Hematology 2021

AMERICAN SOCIETY OF HEMATOLOGY EDUCATION PROGRAM

63rd ASH® Annual Meeting and Exposition

Editors
Adam Cuker, MD, MS
Executive Editor, Hematology 2021
Associate Professor of Medicine
Perelman School of Medicine
University of Pennsylvania
Philadelphia, PA

Mario Cazzola, MD
Deputy Editor, Hematology 2021
Research Group Leader on Myeloid Neoplasms
Fondazione IRCCS Policlinico San Matteo & University of Pavia
Pavia, Italy

Ann LaCasce, MD, MSc

Deputy Editor, Hematology 2021
Associate Professor of Medicine
Dana-Farber Cancer Institute
Boston, MA

Stella Chou, MD
Deputy Editor, Hematology 2021
Associate Professor of Pediatrics
The Children’s Hospital of Philadelphia
University of Pennsylvania
Philadelphia, PA

David Garcia, MD
Special Section Editor, Hematology 2021
Professor of Medicine
University of Washington
Seattle, WA

AMERICAN SOCIETY OF HEMATOLOGY • WASHINGTON, DC

Prakash Singh Shekhawat

 ©2021 by the American Society of Hematology
ISSN 1520–4383 (online)

Hematology, the ASH Education Program, is published Committee on Educational Affairs

annually by the American Society of Hematology (ASH) in Marc J. Kahn (‘21) (Chair)
one volume per year.
Anjali Advani, MD (‘21)
All business correspondence and purchase and reprint Jennifer R. Brown, MD, PhD (‘21)
requests should be addressed to the American Society of
Eldad Dann, MD (‘21)
Hematology, 2021 L Street, NW, Suite 900, Washington, DC
20036, USA; phone: 202–776–0544; fax: 202–776–0545; Faith Davies, MD (‘24)
e-mail: ASH@hematology.org. James Foran, MD (‘22)
Nobuko Hijiya, MD (‘24)
Hematology, the ASH Education Program, is available on the
Internet at https://ashpublications.org/hematology. Margaret Kasner, MD (‘22)
Kah Poh (Melissa) Loh, MD (‘23)
©2021 by The American Society of Hematology. All rights
reserved. Copyright is not claimed in any works of the United Martha P. Mims, MD (‘23)
States Government. Except as expressly permitted in this Casey L. O’Connell, MD (‘23)
statement, no part of this publication may be used (as here Heather A. O’Leary, PhD (‘23)
in after defined) in any form or by any means now or hereafter Adriana Seber, MD, MS (‘21)
known, electronic or mechanical, without permission in
Andrew D. Zelenetz, MD, PhD (‘21)
writing from the Publisher, The American Society of
Hematology. For purposes of this notice, the term “use”
Ex Offcio Members
includes but is not limited to reproduction, photocopying,
storage in a retrieval system, translation, and transmittal. The Martin S. Tallman, MD (‘21) – President
Publisher hereby consents to the use of this publication or Jane N. Winter, MD (‘22) – President-Elect
any part hereof for any noncommercial educational purpose Robert A. Brodsky, MD (‘23) – Vice President
within the health field such as classroom instruction and
clinical and residency training. This publication or any part Liaisons
thereof may be used for educational purposes at confer-
Jessica Altman, MD (‘21) – Senior Executive Editor, ASH-SAP8;
ences, continuing education courses, and other educational
Junior Editor, ASH-SAP7
activity, provided no fee or other compensation is charged,
therefore. All materials so used must acknowledge the Adam Cuker, MD (‘22) – Executive Editor, Hematology; Senior
Publisher’s copyright therein as “©2021 by The American Executive Editor, ASH-SAP7
Society of Hematology.” When requesting the Publisher’s Laura M. De Castro, MD (‘23) – Member, Committee on
permission to use this publication or any part thereof, please Practice
contact the Copyright Clearance Center at 222 Rosewood Cynthia E. Dunbar, MD (‘24) – Secretary
Drive, Danvers, MA 01923, USA; phone: 978–750–8400; fax: Alison Loren, MD (‘21) – 2021 Education Program Co-Chair
978–750–4470; or The American Society of Hematology
Anita Rajasekhar, MD (‘21) – 2021 Education Program Co-Chair
Publishing at 2021 L Street, NW, Suite 900, Washington, DC
20036, USA; phone: 202–776–0544; fax: 202–776–0545. Cover: Computer illustration of eosinophilia showing a view
inside a blood vessel with numerous eosinophils. Source:
Article Citations
Kateryna Kon/Science Photo Library.
Cite articles in this volume by listing Author(s), Title,
Hematology Am Soc Hematol Educ Program. 2021;2021: This activity is supported by educational grants from
beginning page number-ending page number. AstraZeneca; Jazz Pharmaceuticals, Inc.; Merck & Co., Inc.;
Novartis Pharmaceuticals; Pfizer Inc.; and Vertex Pharma-
Hematology Rights and Permission ceuticals Inc.
To request permission to reprint or reuse Hematology
articles, figures, or tables, please visit the Hematology page
at the Copyright Clearance Center (https://marketplace.
copyright.com/rs-ui-web/mp/details/journal/123161909).

The publisher disclaims responsibility for opinions expressed

by the authors.

ii
Contents

ix Editors’ Message
x Reviewers
xi Continuing Medical Education Information

ALL: New Directions for Adult Patients

1 Relapsed ALL: CAR T vs transplant vs novel therapies
NOELLE V. FREY

7 Acute lymphoblastic leukemia in older adults: curtain call for conventional chemotherapy?
MARLISE R. LUSKIN

AML: So Many Options, So Little Time

16 What to use to treat AML: the role of emerging therapies
FELICITAS THOL

24 Whom should we treat with novel agents? Specifc indications for specifc and challenging
populations
LINDSAY WILDE AND MARGARET KASNER

Challenges in Multiple Myeloma Treatment

30 High-risk multiple myeloma: how to treat at diagnosis and relapse?
MARÍA-VICTORIA MATEOS, BORJA PUERTAS MARTÍNEZ, AND VERÓNICA GONZÁLEZ-CALLE

37 Minimal residual disease in multiple myeloma—why, when, where

ANDREW J. YEE AND NOOPUR RAJE

46 Treatment of older adult or frail patients with multiple myeloma

SHAKIRA J. GRANT, CIARA L. FREEMAN, AND ASHLEY E. ROSKO

CLL: Extending Survival

55 Upfront therapy: the case for continuous treatment
CONSTANTINE S. TAM

59 Frontline treatment in CLL: the case for time-limited treatment

VINCENT LÉVY, ALAIN DELMER, AND FLORENCE CYMBALISTA

68 Is there a role for anti-CD20 antibodies in CLL?

HARSH R. SHAH AND DEBORAH M. STEPHENS

Clotting and Bleeding Conundrums

76 Unexplained arterial thrombosis: approach to diagnosis and treatment
JORI E. MAY AND STEPHAN MOLL

85 Cryptogenic oozers and bruisers

KRISTI J. SMOCK AND KAREN A. MOSER

Prakash Singh Shekhawat

Contents | iii
 92 Clots in unusual places: lots of stress, lim­ited data, crit­i­cal deci­sions
CAROL MATHEW AND MARC ZUMBERG

100 EVIDENCE-BASED MINIREVIEW

Should war­fa­rin or a direct oral anti­co­ag­u­lant be used in patients presenting with throm­bo­sis in the
splanch­nic or cere­bral veins?
CAROL MATHEW AND MARC ZUMBERG

CML: Success Breeds More Success

106 TKI dis­con­tin­u­at­ ion in CML: how do we make more patients eli­gi­ble? How do we increase the
chances of a suc­cess­ful treat­ment-free remis­sion?
ANDREAS HOCHHAUS AND THOMAS ERNST

113 Lifelong TKI ther­apy: how to man­age car­dio­vas­cu­lar and other risks
MICHAEL J. MAURO

122 Blast and accel­er­ated phase CML: room for improve­ment

JOAN HOW, VINAYAK VENKATARAMAN, AND GABRIELA SORIANO HOBBS

Coagulation Laboratory Potpourri

129 Direct oral anti­co­ag­u­lant (DOAC) inter­fer­ence in hemo­sta­sis assays
KAREN A. MOSER AND KRISTI J. SMOCK

What Do Aplastic Anemia, PNH, and “Hypoplastic” Leukemia Have in Common?

134 The clin­i­cal and lab­o­ra­tory eval­u­a­tion of patients with suspected hypocellular mar­row fail­ure
SIOBÁN KEEL AND AMY GEDDIS

143 When does a PNH clone have clin­i­cal sig­nif­i­cance?

DARIA V. BABUSHOK

153 When to worry about inherited bone mar­row fail­ure and mye­loid malig­nancy pre­dis­po­si­tion
syn­dromes in the set­ting of a hypocellular mar­row
ANUPAMA NARLA

Defeating Diffuse, Double-Hit, and Dogged Non-Hodgkin Lymphoma

157 Double-hit lym­phoma: opti­miz­ing ther­apy
KIERON DUNLEAVY

164 Relapsed dis­ease: off-the-shelf immunotherapies vs cus­tom­ized engineered prod­ucts

REEM KARMALI

Emerging Therapies and Considerations for Sickle Cell Disease

174 Gene ther­apy for sickle cell dis­ease: where we are now?
JULIE KANTER AND COREY FALCON

181 Hematopoietic cell trans­plan­ta­tion for sickle cell dis­ease: updates and future direc­tions
LAKSHMANAN KRISHNAMURTI

190 EVIDENCE-BASED MINIREVIEW

In young chil­dren with severe sickle cell dis­ease, do the ben­e­fits of HLA-iden­ti­cal sib­ling donor HCT
out­weigh the risks?
NIKETA SHAH AND LAKSHMANAN KRISHNAMURTI

196 Clinical trial con­sid­er­ations in sickle cell dis­ease: patient-reported out­comes, data ele­ments,
and the stake­holder engage­ment frame­work
SHERIF M. BADAWY

iv  |  Hematology 2021  |  ASH Education Program

Hemophilia Update: Our Cup Runneth Over
206 How do we opti­mally uti­lize fac­tor con­cen­trates in per­sons with hemo­philia?
MING Y. LIM

215 EVIDENCE-BASED MINIREVIEW

For over­weight or obese per­sons with hemo­philia A, should fac­tor VIII dos­ing be based on ideal or
actual body weight?
NICOLETTA MACHIN AND MING Y. LIM

219 Factor-mimetic and rebalancing ther­a­pies in hemo­philia A and B: the end of fac­tor con­cen­trates?
PATRICK ELLSWORTH AND ALICE MA

226 Hemophilia gene ther­apy: ush­er­ing in a new treat­ment par­a­digm?

LINDSEY A. GEORGE

Hodgkin Lymphoma: Cure and Optimal Survivorship

234 Controversies in the man­age­ment of early-stage Hodgkin lym­phoma
KRISTIE A. BLUM

240 How to choose first sal­vage ther­apy in Hodgkin lym­phoma: tra­di­tional che­mo­ther­apy vs novel agents
JULIA DRIESSEN, SANNE H. TONINO, ALISON J. MOSKOWITZ, AND MARIE JOSÉ KERSTEN

247 Double-refrac­tory Hodgkin lym­phoma: tack­ling relapse after brentuximab vedotin and check­point
inhib­i­tors
NARENDRANATH EPPERLA AND MEHDI HAMADANI

How Can We Ensure That Everyone Who Needs a Transplant Can Get One?
254 Allogeneic hema­to­poi­etic cell trans­plan­ta­tion for older patients
RICHARD J. LIN AND ANDREW S. ARTZ

264 Increasing access to allo­ge­neic hema­to­poi­etic cell trans­plant: an inter­na­tional per­spec­tive

VANDERSON ROCHA, GIANCARLO FATOBENE, AND DIETGER NIEDERWIESER, FOR THE BRAZILIAN SOCIETY OF BONE
MARROW TRANSPLANTATION AND THE WORLDWIDE NETWORK FOR BLOOD AND MARROW TRANSPLANTATION

275 Increasing access to allotransplants in the United States: the impact of race, geog­ra­phy, and
socio­eco­nom­ics
SANGHEE HONG AND NAVNEET S. MAJHAIL

Immunology 101: What the Practicing Hematologist Needs to Know

281 Immunology 101: fun­da­men­tal immu­nol­ogy for the prac­tic­ing hema­tol­o­gist
SHANNON A. CARTY

287 How immu­no­de­fi­ciency can lead to malig­nancy

SUNG-YUN PAI, KATHRYN LURAIN, AND ROBERT YARCHOAN

296 Modified T cells as ther­a­peu­tic agents

NATHAN SINGH

Indolent Lymphomas
303 What fac­tors guide treat­ment selec­tion in myco­sis fungoides and Sezary syn­drome?
YOUN H. KIM

313 Follicular lym­phoma: is there an opti­mal way to define risk?

CARLA CASULO

320 Does MRD have a role in the man­age­ment of iNHL?

ILARIA DEL GIUDICE, IRENE DELLA STARZA, AND ROBIN FOÀ

Inherited Red Cell Disorders Beyond Hemoglobinopathies

331 Diagnosis and clin­i­cal man­age­ment of red cell mem­brane dis­or­ders
THEODOSIA A. KALFA

Prakash Singh Shekhawat

Contents | v
 341 Diagnosis and clin­i­cal man­age­ment of enzymopathies
LUCIO LUZZATTO

353 Diamond-Blackfan ane­mia

LYDIE M. DA COSTA, ISABELLE MARIE, AND THIERRY M. LEBLANC

It Takes a Village: Maximizing Supportive Care and Minimizing Toxicity During Childhood Leukemia
Therapy
361 Fungal diag­nos­tic test­ing and ther­apy: nav­i­gat­ing the neutropenic period in chil­dren with high-
risk leu­ke­mia
BRIAN T. FISHER

368 Minimizing car­diac tox­ic­ity in chil­dren with acute mye­loid leu­ke­mia

HARI K. NARAYAN, KELLY D. GETZ, AND KASEY J. LEGER

376 Managing ther­apy-asso­ci­ated neu­ro­tox­ic­ity in chil­dren with ALL

DEEPA BHOJWANI, RAVI BANSAL, AND ALAN S. WAYNE

Let the CHIP(s) Fall Where They May? Burdens and Benefits of Diagnosing Clonal Hematopoiesis
384 CHIP: is clonal hema­to­poi­e­sis a sur­ro­gate for aging and other dis­ease?
LUKASZ P. GONDEK

390 Mechanisms of somatic trans­for­ma­tion in inherited bone mar­row fail­ure syn­dromes

HARUNA BATZORIG CHOIJILSUREN, YEJI PARK, AND MOONJUNG JUNG

399 When are idi­o­pathic and clonal cytopenias of unknown sig­nif­i­cance (ICUS or CCUS)?
AFAF E. W. G. OSMAN

Management Strategies for Sickle Cell Disease

405 Optimizing man­age­ment of sickle cell dis­ease in patients under­go­ing sur­gery
CHARITY I. OYEDEJI AND IAN J. WELSBY

411 Treatment dilem­mas: strat­eg ­ ies for priapism, chronic leg ulcer dis­ease, and pul­mo­nary
hyper­ten­sion in sickle cell dis­ease
ROBERTA C.G. AZBELL AND PAYAL CHANDARANA DESAI

MDS: Beyond a One-Size-Fits-All Approach

418 Have we reached a molec­ul­ar era in myelodysplastic syn­dromes?
MARIA TERESA VOSO AND CARMELO GURNARI

428 Lower risk but high risk

AMY E. DeZERN

435 EVIDENCE-BASED MINIREVIEW

Molecular pre­ci­sion and clin­i­cal uncer­tainty: should molec­u­lar pro­fil­ing be rou­tinely used to guide
risk strat­i­fi­ca­tion in MDS?
DAN­IEL R. RICHARDSON AND AMY E. DeZERN

439 Oral hypomethylating agents: beyond con­ve­nience in MDS

ELIZABETH A. GRIFFITHS

MPN: New Directions

710 MPN and throm­bo­sis was hard enough . . . ​now there’s COVID-19 throm­bo­sis too
ANNA FALANGA

448 EVIDENCE-BASED MINIREVIEW

Are DOACs an alter­na­tive to vita­min K antag­on
­ ists for treat­ment of venous throm­bo­em­bo­lism in
patients with MPN?
FRANCESCA SCHIEPPATI AND ANNA FALANGA

453 Novel ther­a­pies vs hema­to­poi­etic cell trans­plan­ta­tion in mye­lo­fi­bro­sis: who, when, how?
JAMES ENGLAND AND VIKAS GUPTA

463 Running inter­feron inter­fer­ence in treating PV/ET: meet­ing unmet needs

ELIZABETH A. TRAXLER AND ELIZABETH O. HEXNER

vi  |  Hematology 2021  |  ASH Education Program

Multidisciplinary Hematologic Disorders: What Is the Hematologist’s Role?
469 Hereditary hem­or­rhagic tel­an­gi­ec­ta­sia (HHT): a prac­ti­cal guide to man­age­ment
ADRIENNE M. HAMMILL, KATIE WUSIK, AND RAJ S. KASTHURI

478 Chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion: anticoagulation and beyond

KARLYN A. MARTIN AND MICHAEL J. CUTTICA

485 Rebalanced hemo­sta­sis in liver dis­ease: a mis­un­der­stood coagulopathy

LARA N. ROBERTS

Neutropenia: Doing More with Less

492 Diagnosis and ther­a­peu­tic deci­sion-mak­ing for the neutropenic patient
JAMES A. CONNELLY AND KELLY WALKOVICH

504 Understanding neutropenia sec­ond­ary to intrin­sic or iat­ro­genic immune dysregulation

KELLY WALKOVICH AND JAMES A. CONNELLY

514 Impaired myelopoiesis in con­gen­i­tal neutropenia: insights into clonal and malig­nant hema­to­poi­e­sis
JULIA T. WARREN AND DAN­IEL C. LINK

Perioperative Consultative Hematology: Can You Clear My Patient for Surgery?

521 Perioperative con­sul­ta­tive hema­tol­ogy: can you clear my patient for a pro­ce­dure?
ALLISON ELAINE BURNETT, BISHOY RAGHEB, AND SCOTT KAATZ

529 How to man­age bleed­ing dis­or­ders in aging patients need­ing sur­gery

MOUHAMED YAZAN ABOU-ISMAIL AND NATHAN T. CONNELL

536 Heparin-induced throm­bo­cy­to­pe­nia and car­dio­vas­cu­lar sur­gery

ALLYSON M. PISHKO AND ADAM CUKER

Pregnancy in Special Populations: Challenges and Solutions

545 How to eval­u­ate and treat the spec­trum of TMA syn­dromes in preg­nancy
MARIE SCULLY

552 Pregnancy in spe­cial pop­u­la­tions: chal­lenges and solu­tions prac­ti­cal aspects of man­ag­ing von
Willebrand dis­ease in preg­nancy
OZLEM TURAN AND REZAN ABDUL KADIR

559 Prevention and man­age­ment of venous throm­bo­em­bo­lism in preg­nancy: cut­ting through the
prac­tice var­i­a­tion
LESLIE SKEITH

Survivorship After Allogeneic Hematopoietic Cell Transplant

570 Psychosocial and finan­cial issues after hema­to­poi­etic cell trans­plan­ta­tion
DAVID BUCHBINDER AND NANDITA KHERA

578 Noninfectious com­pli­ca­tions of hema­to­poi­etic cell trans­plan­ta­tion

KIRSTEN M. WILLIAMS

587 Infectious com­pli­ca­tions and vac­cines

PER LJUNGMAN

The Changing Landscape of α- and β-Thalassemia Diagnosis and Treatment

592 Advances in the man­age­ment of α-thal­as­se­mia major: rea­sons to be opti­mis­tic
PAULINA HORVEI, TIPPI MACKENZIE, AND SANDHYA KHARBANDA

600 β-Thalassemia: evolv­ing treat­ment options beyond trans­fu­sion and iron che­la­tion
ARIELLE L. LANGER AND ERICA B. ESRICK

607 Optimal strat­e­gies for car­rier screen­ing and pre­na­tal diag­no­sis of α- and β-thal­as­se­mia
CHERYL MENSAH AND SUJIT SHETH
Prakash Singh Shekhawat
Contents | vii
 The COVID Crash: Lessons Learned from a World on Pause
614 How to rec­og­nize and man­age COVID-19-asso­ci­ated coagulopathy
GLO­RIA F. GERBER AND SHRUTI CHATURVEDI

621 COVID-19 and throm­bo­sis: searching for evi­dence

BRIGHT THILAGAR, MOHAMMAD BEIDOUN, RUBEN RHOADES, AND SCOTT KAATZ

628 Passive immune ther­a­pies: another tool against COVID-19

LISE J. ESTCOURT

Update in Graft-versus-Host Disease

642 Acute GVHD: think before you treat
LAURA F. NEWELL AND SHERNAN G. HOLTAN

648 Updates in chronic graft-ver­sus-host dis­ease

BETTY K. HAMILTON

655 What else do I need to worry about when treating graft-ver­sus-host dis­ease?
AREEJ EL-JAWAHRI

What’s New in Plasma Cell Disorders?

662 Advances in MGUS diag­no­sis, risk strat­i­fi­ca­tion, and man­age­ment: intro­duc­ing myeloma-defining
genomic events
OLA LANDGREN

673 Smoldering mul­ti­ple mye­loma: evolv­ing diag­nos­tic cri­te­ria and treat­ment strat­e­gies
ALISSA VISRAM, JOSELLE COOK, AND RAHMA WARSAME

682 AL amy­loid­osis: untangling new ther­a­pies

SUSAN BAL AND HEATHER LANDAU

What’s New in the Transfusion Management of Sickle Cell Disease?

689 How to avoid the prob­lem of eryth­ro­cyte alloimmunization in sickle cell dis­ease
FRANCE PIRENNE, A­LINE FLOCH, AND ANOOSHA HABIBI

696 Indications for transfusion in the management of sickle cell disease

HYOJEONG HAN, LISA HENSCH, AND VENÉE N. TUBMAN

704 Management of hemo­lytic trans­fu­sion reac­tions

JEANNE E. HENDRICKSON AND ROSS M. FASANO

viii  |  Hematology 2021  |  ASH Education Program

Editors’ Message

Welcome to the 63rd annual meeting of the American Society of Hematology. Two years
after the first cases of COVID-19 were reported, we are learning to co-exist with the virus.
Many of us are thrilled to be able to attend a scientific congress in person for the first time
since the pandemic began. Others are grateful for the option of virtual participation, which
greatly expands access to the meeting and its cutting-edge content for hematologists
around the world.
This marks our third and final year as Editors of Hematology. During our term, we have
made significant changes to the book. For starters, Hematology is now electronic-only;
gone are the days of schlepping a heavy tome home from the ASH meeting. We have also
sought to make the book more readable and visually appealing. All chapters are based on
clinical cases and include visual abstracts. We have increased the number of tables and
figures per chapter while reducing the amount of text.
Nonetheless, at its core, Hematology remains the same great resource we inherited
three years ago. As always, you can count on high quality content spanning the breadth of
malignant and nonmalignant hematology, based on a state-of-the-art Education Program
developed by Education Program Co-Chairs, Dr. Alison Loren and Dr. Anita Rajasekhar.
This book is the culmination of the efforts of many individuals and their dedication to
ASH, including the authors, reviewers, and ASH staff (Ken April, Glenn Landis, Michelle Lee,
and Joanna Robertson). We hope that you enjoy reading Hematology as much as we have
enjoyed editing it these last three years.

Adam Cuker, MD Ann S. LaCasce, MD Mario Cazzola, MD Stella T. Chou, MD David Garcia, MD

Prakash Singh Shekhawat

ix
 Reviewers

The editors would like to thank the session chairs and the reviewers who worked hard to ensure the reliability of the material
presented in Hematology 2021:

Jeremy S. Abramson Erica B. Esrick Ola Landgren Rachel E. Rau

Ranjana Advani Andrew M. Evens Sophie Lanzkron Farhad Ravandi
Walter Ageno Anna Falanga Stephanie J. Lee Ulrike M. Reiss
Hanny Al-Samkari Emmanuel J. Favaloro Georg Lenz Lawrence Rice
Jennifer Andrews Courtney D. Fitzhugh Jeffrey M. Lipton Heesun J. Rogers
Joseph H. Antin Angela G. Fleischman Mark R. Litzow Cristhiam M. Rojas
Andrew Artz Mary E. Flowers Marlise R. Luskin Hernandez
Daria V. Babushok Gianluca Gaidano Samuel E. Lux David M. Ross
Andrea Bacigalupo Guillermo Garcia-Manero Alice Ma Bruce S. Sachais
Tiziano Barbui Rebecca A. Gardner Ivan Maillard Marie Scully
Renato Bassan Alex Gatt Navneet S. Majhail John F. Seymour
Shannon Bates James N. George Michael Makris Nirmish Shah
Lisa Baumann Kreuziger Lukasz P. Gondek Luca Malcovati Urvi A. Shah
Emily Baumrin Rachael F. Grace Karlyn Martin Bronwen E. Shaw
Neel S. Bhatt John G. Gribben Paul J. Martin Shalini Shenoy
Deepa Bhojwani Elizabeth A. Griffiths Maria-Victoria Mateos Sujit Sheth
Kristie A. Blum Anna B. Halpern Patrick T. McGann Akiko Shimamura
Niccolo Bolli Betty K. Hamilton Neha Mehta-Shah Michelle Sholzberg
Robert A. Brodsky Catherine P.M. Hayward Reid W. Merryman B. Douglas Smith
Andrew Campbell Jeanne Elise Hendrickson Ruben A. Mesa Kristi J. Smock
Shannon Ann Carty Alex F. Herrera Brian Miller Gerald A. Soff
Carla Casulo Jens Hillengass Alison R. Moliterno Jean Soulier
Mario Cazzola Jo Howard Philippe Moreau Brady L. Stein
Shruti Chaturvedi Scott F. Huntington Alison J. Moskowitz Sean R. Stowell
Alan R. Cohen E. Leila Jerome Clay Michael F. Murphy Kendra L. Sweet
Raymond L. Comenzo Theodosia A. Kalfa Peter E. Newburger Felicitas Thol
Nathan T. Connell Manali Kamdar Diane Nugent Judith Trotman
Catherine Cordonnier Julie Kanter Susan M. O’Brien Saad Z. Usmani
Jennifer L. Crombie Margaret Kasner Kristen M. O’Dwyer Adrianna Vlachos
Stacy E Croteau Sioban B. Keel Vivian G. Oehler Oksana Volod
Geoffrey D.E. Cuvelier Nandita Khera Rebecca L. Olin Kelly Walkovich
Anita D’Souza Miriam Kimpton Ashok Pai Mark C. Walters
Michael R. DeBaun Melanie Ann Kirby-Allen Sung-Yun Pai Theodore E. Warkentin
Matteo G. Della Porta Frederikus A. Klok Christopher Jordan Patriquin William George Wierda
Amy E. DeZern Barbara A. Konkle Claire S. Philipp Jennifer J. Wilkes
Courtney D. DiNardo Peter Kouides Joseph Pidala Marcin W. Wlodarski
James Douketis Austin Kulesekararaj Uwe Platzbecker Jasmine Zain
Kieron Dunleavy Shaji K. Kumar Daniel A. Pollyea Alberto Zanella
Christopher C. Dvorak Peter Kurre Steven Pugliese
Matthew Ehrhardt Janet L. Kwiatkowski Margaret V. Ragni

x
Continuing Medical Education Information
Hematology: the ASH Education Program, is an annual publica-
tion that provides practicing hematologists with invaluable infor-
mation on the most important areas of clinical progress.
Hematology 2021 is a peer-reviewed collection of articles
written by the 2021 ASH Education Program speakers and the
Ham-Wasserman Lecturer. The papers showcase groundbreak-
ing advances and new concepts in 32 different fields. Every year,
the periodical provides an updated and comprehensive review
of each of the topics covered in the annual meeting education
sessions.
Educational objectives
1. Employ the knowledge gained regarding the diagnosis and
treatment of malignant and nonmalignant hematologic dis-
orders to improve patient care.
2. Discuss the state-of-the-art therapeutics in hematology.
3. Analyze the potential contribution of novel, not-yet-approved
modalities of therapy to current evidence-based manage-
ment of malignant and nonmalignant hematologic disorders.
Date of release
December 2021
Date of expiration
On December 31, 2022, the ability to earn Continuing Medical
Education credit and American Board of Internal Medicine Main-
tenance of Certification Medical Knowledge points for this prod-
uct expires. This is the last date for users to claim credit for this
product. For questions about credit, please contact the ASH
Education Department at cme@hematology.org or call toll-free
866-828-1231 within United States only; 1-202-776-0544 interna-
tionally.
Accreditation and Credit Designation
The American Society of Hematology is accred-
ited by the Accreditation Council for Continuing
Medical Education to provide continuing medical
education for physicians.
The American Society of Hematology designates this endur-
ing material for a maximum of 40 AMA PRA Category 1 Credits . TM • How
Physicians should claim only the credit commensurate with the
extent of their participation in the activity.
Physicians who participate in this CME activity but are not
licensed in the United States are also eligible for AMA Category
1 PRA CreditTM. To earn these credits, readers must pass two on-
Prakash
line tests (malignant and nonmalignant) Singh
based on Shekhawat
chapters from
the book.
ABIM Maintenance of Certifcation
Successful completion of this CME activi-
ty enables the participant to earn up to 40
Knowledge Points points in the American Board of Internal Med-
icine’s (ABIM) Maintenance of Certification (MOC) program. Par-
ticipants will earn MOC points equivalent to the amount of CME
credits claimed for the activity. It is the CME activity provider’s
responsibility to submit participant completion information to
ACCME for the purpose of granting ABIM MOC credit.
Claiming CME Credits and ABIM points
The estimated time to complete this educational activity is 40
hours. To claim CME credit, users must complete an evaluation
of the product and a test of medical knowledge, both of which
are accessed through ASH Academy On Demand (academy.hema
tology.org). There is a one-time processing fee for claiming CME/
MOC credit. Users with scores of 80% or better on these self-
assessment modules are eligible to claim credit for the activity.
To facilitate claiming of credit, the test for this product is
divided into two subtests, one of which focuses on malignant
hematology content with the other focusing on nonmalignant
content. Successful completion of each test earns the user 20
AMA Category 1 PRA CreditsTM. Users claim CME and/or ABIM
MOC credit for each test individually. You can take one or both
tests, depending on your CME and MOC needs.
Each test consists of 25 questions. The 25 questions can be
answered in one sitting, or a user can save their progress and
return to complete the test at a later time.
The Malignant Hematology Test covers information present-
ed in the following sections:
•
ALL: New Directions for Adult Patients
•
AML: So Many Options, So Little Time
•
Challenges in Multiple Myeloma Treatment
•
CLL: Extending Survival
•
CML: Success Breeds More Success
•
Defeating Diffuse, Double-Hit and Dogged Non-Hodgkin
Lymphoma
• Hodgkin Lymphoma: Cure and Optimal Survivorship
Can We Ensure That Everyone Who Needs a Transplant
Can Get One?
• Immunology 101: What the Practicing Hematologist Needs to
Know
• Indolent Lymphomas
• It Takes a Village: Maximizing Supportive Care and Minimizing
Toxicity During Childhood Leukemia Therapy
• MDS: Beyond a One-Size-Fits-All Approach
xi
• MPN: New Directions
• Normal and Leukemic Stem Cells
• Survivorship After Allogeneic Hematopoietic Cell Transplant
• Update in Graft vs Host Disease
• What’s New in Plasma Cell Disorders?
The Nonmalignant Hematology Test covers information presented
in the following sections:
• Clotting and Bleeding Conundrums
• Coagulation Laboratory Potpourri
• Curative Therapies and Considerations for Sickle Cell Disease
• Dameshek’s Riddle in Practice: Is this Aplastic Anemia, Parox-
ysmal Nocturnal Hemoglobinuria, or Hypoplastic Leukemia?
• Hemophilia Update: Our Cup Runneth Over
• Inherited Red Cell Disorders Beyond Hemoglobinopathies
xii 
• Let the CHIP(s) Fall Where They May? Burdens and Benefits
of Diagnosing Clonal Hematopoiesis
• Management Strategies for Sickle Cell Disease
• Multi-Disciplinary Hematologic Disorders: What Is the Hema-
tologist’s Role?
• Neutropenia: Doing More with Less
• Perioperative Consultative Hematology: Can You Clear My
Patient for Surgery?
• Pregnancy in Special Populations: Challenges and Solutions
• The Changing Landscape of Alpha and Beta-Thalassemia
Diagnosis and Treatment
• The COVID Crash: Lessons Learned from a World on Pause
• What’s New in the Transfusion Management of Sickle Cell
­Disease?
 ALL: NEW DIRECTIONS FOR ADULT PATIENTS

Relapsed ALL: CAR T vs transplant vs novel

therapies

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/1/1851774/1frey.pdf by guest on 13 December 2021

Noelle V. Frey
Cell Therapy and Transplant Program, Abramson Cancer Center, Department of Medicine, Hospital of the University of Pennsylvania, Philadelphia, PA

Chimeric antigen receptor T-cell therapy targeting CD19 (CART19) has expanded the treatment options for patients
with relapsed/refractory (r/r) B-cell acute lymphoblastic leukemia (ALL). The approval of tisagenlecleucel for pediatric
and young adult patients with r/r ALL has allowed broader access for some patients, but the treatment of older adults
is available (at the time of this writing) only within a clinical trial. High remission rates have been consistently observed
with varied CART19 products and treatment platforms, but durability of remissions and thus the potential role of a con-
solidative allogeneic stem cell transplant (SCT) is more uncertain and likely to vary by product and population treated.
The immunologic characteristics of CARTs that confer high response rates also account for the life-threatening toxicities
of cytokine release syndrome and immune effector cell–associated neurotoxicity syndrome, the severity of which also
varies by patient and disease characteristics and product. Further considerations informing a decision to treat include
feasibility of leukapheresis and timeline of manufacture, alternative treatment options available, and the appropriate-
ness of a potential consolidative allogeneic SCT. Advances in the field are under way to improve rate and duration of
responses and to mitigate toxicity.

LEARNING OBJECTIVES
• Understand the efficacy and toxicity outcomes of CART19 in r/r ALL and how they vary by product, patient, and
disease-related factors
• Understand factors that inform a decision to consolidate a recipient of CART19 in CR with allogeneic SCT

Efficacy outcomes of CART19 in relapsed
and refractory ALL
Response
Autologous T cells engineered to express a chimeric anti-
gen receptor T-cell therapy targeted to CD19 (CART19) have
consistently shown high complete remission (CR) rates
(62%-95%) in adult and pediatric patients with relapsed or
refractory (r/r) acute lymphoblastic leukemia (ALL).1–12 The
chimeric antigen receptor (CAR) T cells used in the studies
summarized in Table 1 are made from a patient’s T cells col-
lected through leukapheresis that are then transduced with
a CAR targeting CD19 using a replication-incompetent ret-
rovirus or lentivirus. The CARs include a costimulatory mol-
ecule, which is either CD28 or 41BB depending on product.
Patients typically receive lymphodepleting chemotherapy
(commonly cyclophosphamide and fludarabine) prior to
CART19 infusion. Importantly, despite differences in patient
populations, clinical trial procedures, CAR molecules, and
manufacturing differences, high initial response rates are
maintained. Across studies, remissions are achieved quickly
Prakash Singh Shekhawat
(by 1-month postinfusion), are often minimal residual dis-
ease (MRD) negative, and are not discriminated by muta-
tional status or number and type of prior therapies. The
trial populations represent heavily pretreated patients, with
over a third of patients in some studies having relapsed
after prior allogeneic stem cell transplant (SCT).1–3,6,8,9 Impor-
tantly, several studies have shown CART19 cells tracking
into the central nervous system (CNS) with responses seen
in patients with CNS disease.2,5,10,11 Although the outcomes
discussed here are focused on recipients of CART19, CARTs
to other targets (specifically CD22) have been shown to
be effective.13–16 In a large series of pediatric and adoles-
cent young adult (AYA) patients (N=58), many of whom had
relapsed after CART19 therapy (N=51), anti-CD22 CAR ther-
apy induced a CR in 70% of patients.14
Survival and durability of response
Median overall survival (OS) in most studies using CART19
is beyond 1 year and importantly is noted within some
reports to vary by dose level or other changes to study
CAR T for ALL | 1
Table 1. Outcomes of CART19 in patients with relapsed and refrac­tory acute lym­pho­blas­tic leu­ke­mia

Prior
Reference CART domain No. treated Median age, y blinatumomab, % Prior SCT, % CR, % CRS ICANS
Adult patients
  Frey et al1 41BB 35 34 (21-70) 31 37 69 94% total 40% total
9% grades 4-5 6% grade 3
  Hay et al3 41BB 53 39 (20-76) 20 43 85 75% total 23% total
19% grades 3-4
  Park et al9 CD28 53 44 (23-64) 25 36 83 85% total 42% grades 3-4
26% severe
1 grade 5
  Shah et al12 CD28 55 40 (28-52) 45 42 71 89% total 60% total
25% grade 3 or 23% grades 3-4

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/1/1851774/1frey.pdf by guest on 13 December 2021

higher 1 grade 5
Combined pedi­at­ric and adult patients
  Jiang et al4 41BB 58 28 (10-65) NA 5 88 38% grades 3-5 16% grades 2-3
 Ortíz-Maldonado 41BB 38 24 (3-67) 26 87 85 13.2% grades 3-5 2.6% grade3 or
et al8 higher
  Wang et al16 41BB 23 42 (10-67) NA 0 83 100% total 13% total
18% grade 3
  Maude et al5 41BB 30 14 (5-60) 10 60 90 100% total 43% total
27% severe
Pediatric and AYA patients
  Gardner et al2 41BB 45 12.2 (1.3-25.3) 14 62 93 93% total 49% total
21% severe 21% severe
  Maude et al6 41BB 75 11 (3-23) 0 61 81 77% total 13% grade 3
25% grade 4
  Shah et al11 CD28 50 13.5 (4.3-30.4) 10 40 62 70% total 20% total
22% grades 3-4 8% severe
NA, not available.

design through­ out the clin­i­

cal trial. Multicenter stud­ ies using
con­sis­tent prod­ucts reflect more gen­er­al­iz­able out­comes. In the
piv­otal ELIANA trial lead­ing to approval of tisagenlecleucel for
pedi­at­ric and young adult patients, 75 patients were treated,
with a CR rate of 81%. Median OS was not reached, with event-
free sur­vival (EFS) and OS at 12 months of 50% and 76%, respec­
tively. For respond­ers, the median dura­tion of remis­sion was not
reached.6 In the mul­ti­cen­ter ZUMA-3 study treating 55 adults
with r/r ALL with KTE-X19, CR rate was 71% and median OS was
18.2 months. The median dura­tion of remis­sion for respond­ers
was 12.8 months.12 Another mul­ti­cen­ter study treated 38 adult
and pedi­at­ric patients with CART19 with a CR rate of 71%, with
OS and pro­gres­sion-free sur­vival of 67% and 47%, respec­tively, at
1 year. The median dura­tion of response was 14.8 months.8
Impact of prior blinatumomab on effi­cacy
Blinatumomab is a bispecific T cell engag­ing a sin­gle-chain anti­
body con­struct linking CD3+ T cells with CD19+ B cells. There
is a log­i­cal con­cern that due to its sim­i­lar mech­a­nism of action
and tar­get, prior treat­ment with blinatumomab could adversely
affect out­comes from CART19. For this rea­son, early CART19 stud­
ies, such as the ELIANA study, prohibited prior treat­ment with
blinatumomab or other CD19-targeted ther­ a­
pies.6 Since these
ini­
tial stud­ ies, the use of blinatumomab has increased sig­ nif­i­
cantly, espe­cially in adults, which is reflected in a pro­gres­sively
higher pro­ por­
tion of patients (20%-45%) on CART19 stud­ ies
hav­ing received blinatumomab (see Table 1).1,3,9,12 In the recently
published ZUMA-3 mul­ti­cen­ter trial, 45% of the 55 adult patients
treated with KTE-X19 had received blinatumomab. Although
prior expo­sure to CD19-targeted agents may affect risk of sub­se­
quent CD19– relapse, ini­tial response rates do not seem greatly
affected. The Children’s Hospital of Philadelphia reported out­
comes from 166 patients treated with CART19 at their insti­tu­tion.
The CR rate was 93%, and 67 patients ulti­mately relapsed, 39 with
CD19– dis­ease due to anti­gen escaper. Prior ther­apy with blina-
tumomab was asso­ci­ated with a higher risk for CD19– relapse.17
Importance of per­sis­tence for dura­ble remis­sions
Several ALL tri­ als using CART19 prod­ ucts containing a 4-1BB
costimulatory domain have shown a strong cor­re­la­tion between
CART19 per­ sis­
tence (often represented by B-cell aplasia, a
bio­log­i­cal sur­ro­gate for func­tional per­sis­tence) and CD19+
relapses.1-3,6,7,18 In an anal­
y­
sis of out­
comes with tisagenlecleu-
cel in patients with ALL on the ELIANA trial, the patients with
CD19+ relapses had a more rapid loss of CART per­ sis­
tence
com­ pared with those with dura­ ble remis­ sions. Of inter­
est,
patients who devel­oped a CD19– relapse had per­sis­tence sim­
i­lar to those who had sustained responses with mech­a­nism of
relapse due to anti­gen escape.18 In another pedi­at­ric study using
a dif­fer­ent 4-1BBCART19 prod­uct at Seattle Children’s, a lon­ger

2  |  Hematology 2021  |  ASH Education Program

 dura­tion of B-cell aplasia cor­re­lated sig­nif­i­cantly with the dura­
bil­ity of remis­sion.2 Another study found the median per­ sis­
tence of 41BBCART19 was shorter for those with a CD19+ relapse
(2.5 months) as opposed to a CD19– relapse (6 months).4
Several stud­ies using CD28-CART19 prod­ucts have shown
lim­ited per­sis­tence, but a cor­re­la­tion of out­comes with per­sis­
tence is less clear.9,11,12 In the ZUMA-3 trial using CD28-CART19
(KTE-X19), CARTs were no lon­ger detect­able in 79% of patients
with evaluable sam­ples by 6 months, and B-cell recov­ery had
occurred in all­evaluable ongo­ing respond­ers at 12 months.
The expe­ ri­
ence at Memorial Sloane Kettering Cancer Cen-
ter using their CD28-CART19 showed median per­sis­tence of
14 days (range, 7-138 days), and dura­tion of per­sis­tence did not
cor­re­late with sur­vival.9 In 1 study, how­ever, from the National
Cancer Institute treating 50 pedi­at­ric and AYA patients with a
CD28-CART19 prod­uct, remis­sions were dura­ble only with con-
solidative SCT, which the authors hypoth­e­size may be due to
lim­ited per­sis­tence.11
Given the cor­re­la­tion between 4-1BBCART19 per­sis­tence and
CD19+ dura­bil­ity of remis­sions, it is vital to under­stand cor­re­
lates with per­sis­tence using these CART19 prod­ucts. The group
at the Fred Hutchinson Cancer Research Center found that the
addi­tion of fludarabine to cyclo­phos­pha­mide lymphodepletion
improved per­sis­tence and dis­ease-free sur­vival.19 The impact of
anti­gen load on per­sis­tence has var­ied across stud­ies. The group
from Seattle Children’s found a pos­i­tive cor­re­la­tion, with >15%
bone mar­row blasts cor­re­lat­ing with prolonged per­sis­tence.2
Conversely, the group from did not find an asso­ ci­
a­tion with
low dis­ease bur­den and CD19+ relapses.17 Evaluation of T cells
in apher­e­sis and manufactured 4-1BBCART19 prod­ucts iden­ti­fied
phe­no­typ­i­cal and func­tional attri­butes of CAR CD8 T cells that
cor­re­lated with per­sis­tence.20
Most of the stud­ies in Table 1 use CARs containing a murine
domain that may be a tar­ get for immune-medi­ ated rejec­ tion.
The devel­op­ment of a CART19 prod­uct using a CAR containing a
­human­ized anti-CD19 scFv domain may bypass this ­rejec­tion,
resulting in improved per­ sis­tence and relapsed free sur­ vival
(RFS).7,10 A recent report of out­comes in pedi­at­ric and AYA patients
with r/r ALL treated with a human­ized CART19 based on the back­
bone of CTL019 (tisagenlecleucel) has shown excel­lent responses.
In 41 CART19-​naive patients, CR rate was 98%, and RFS at 1 year
was 84%. Similar to prior stud­ies, ear­lier B-cell recov­ery as a time-
depen­­dent covariate cor­re­lated with worse RFS. In an explor­atory
anal­y­sis, time to B-cell recov­ery was com­pared with a his­tor­i­cal
cohort of CTL019 recip­i­ents, and there was a trend toward a lower
cumu­la­tive inci­dence of B-cell recov­ery by 6 months (15% vs 29%),
but it did not reach sta­tis­ti­cal sig­nif­i­cance.7
Risk of CD19– relapse
CD19– relapses hap­pen when CD19 anti­gen loss occurs through
muta­tion or epi­ge­netic alter­ations, likely in preexisting leu­ke­mia
subclones.21-23 An attrac­tive approach to limit the inci­dence of
CD19– relapse is to infuse CARTs that tar­get more than 1 anti­gen
such as CD19 with either CD20 or CD22. One approach is to gen­
er­ate a sin­gu­lar CART prod­uct that can tar­get more than 1 anti­
gen vs coadministering 2 dis­tinct prod­ucts with dif­fer­ent tar­gets
either con­cur­rently or sequen­tially.13,16,24-26 In 1 study, 20 chil­dren
with r/r ALL who achieved an MRD-CR with CART19 were infused
with CART22 at a median of 1.65 months after CART19 infu­sion.13
No patients received consolidative SCT, and at 1 year, 85% were
Prakash Singh Shekhawat
in CR, an improve­ment when com­pared ret­ro­spec­tively with
out­comes from CART19 alone.13
Role of consolidative SCT
For a patient with­out an ante­ced­ent SCT with an MRD-CR after
CART19, a crit­i­cal ques­tion is whether to con­sol­i­date that remis­
sion with trans­plant if the patient is med­i­cally fit to con­sider this
approach. As always when con­sid­er­ing SCT, treat­ment-related
mor­bid­ity and mor­tal­ity needs to be bal­anced against risk of
relapse. Another con­sid­er­ation is that the CARTs with func­tional
per­sis­tence would be destroyed by SCT, los­ing their ben­e­fit of
ongo­ing tumor sur­veil­lance. There are no stud­ies to date that
ran­dom­ ize patients after CART ther­ apy to allo­ ge­ neic SCT or
obser­va­tion. Furthermore, the deci­sion is not likely gen­er­al­iz­able
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/1/1851774/1frey.pdf by guest on 13 December 2021
across dif­fer­ent CART19 prod­ucts. For exam­ple, patients tak­ing
4-1BBCART19 prod­ucts have bet­ter per­sis­tence (with poten­tially
lon­ger dis­ease-free inter­vals, although more data are needed)
than recip­i­ents of CD28-CART19 prod­ucts. There is con­sid­er­able
var­i­a­tion among clin­i­cal tri­als and ret­ro­spec­tive ana­ly­ses regard­
ing the poten­tial ben­e­fit of a consolidative SCT, and again, no
ran­dom­ized tri­als have for­mally addressed this ques­tion.
The role of consolidative SCT in CD28-CARTs
Recently, the National Cancer Institute reported long-term fol­
low-up of 50 chil­dren and young adults (median age, 13.5 years;
range, 4.3-30.4 years) treated with their CD28-CART19 prod­
uct who clearly benefited from a consolidative SCT. Of the
28 patients who achieved an MRD-CR, 21 (75%) proceeded to
SCT with a median OS of 70.2 months. The cumu­la­tive inci­dence
of relapse after SCT was 9.5% at 24 months. All patients who did
not pro­ceed to SCT relapsed at a median of 152 (range, 94-394)
days.11 However, data from Memorial Sloane Kettering Cancer
Center pro­vided long-term out­comes from 51 adults (median
dura­tion, 18 months) treated with a dif­ fer­
ent CD28-CART19.
Of the 32 patients in MRD-CR, no dif­fer­ence in OS was seen in
patients who did or did not receive SCT after CARTs.9 Recently,
results from the ZUMA-3 mul­ti­cen­ter trial using KTE-X19 (a CART19
with CD28) have been reported. In this study, 55 patients were
treated and 39 patients (71%) achieved CR. Ten patients (18%)
proceeded to SCT, and with sen­si­tiv­ity ana­ly­ses, the median
dura­tion of remis­sion was unchanged by SCT con­sol­i­da­tion.12
The role of consolidative SCT in 4-1BB CARTs
It is clear that a sub­set of patients with ini­tial response to some
41BBCART19 prod­ucts has ongo­ing dura­ble remis­sions with­out
consolidative SCT, which cor­re­lates with per­sis­tence.2,5-8 Even
when a par­tic­u­lar CART19 prod­uct has been shown to have good
func­tional per­sis­tence, how­ever, that per­sis­tence is not observed
across all­patients treated, and CD19+ relapses remain a sig­nif­i­
cant chal­lenge that may be miti­gated by SCT in some patients.1
In addi­tion, even if one antic­i­pates and observes ongo­ing per­
sis­tence, CD19– relapses from anti­ gen escape occur despite
CART19 per­sis­tence. Only a minor­ity of the young adult and pedi­
at­ric patients with relapsed ALL treated with tisagenlecleucel
on the ELIANA study received a consolidative SCT. The EFS of
50% at 12 months is there­fore rep­re­sen­ta­tive of remis­sion dura­
bil­ity with this 4-1BBCART19 prod­uct.6 When CTL019 (the pre­
cur­sor to tisagenlecleucel) was used to treat older adults, those
who proceeded to SCT in MRD-CR had an improved EFS com­
pared with those who were not, although dura­ ble remis­sions
CAR T for ALL  |  3
were seen in each group.1 Similarly, in another study using a dif­
fer­ent 4-1BBCART19, there was a trend toward an improved EFS
in patients bridged to SCT (P = .088). In another study using a
41BBCART19 for r/r ALL, 21 of the 47 patients who achieved MRD-
CR were bridged to SCT. Although there was no dif­fer­ence in OS
between the 2 groups, EFS and RFS were sig­nif­i­cantly prolonged
in the SCT group.4
Toxicity
Due to their mech­a­nism of action, CART19 is asso­ci­ated with
cyto­kine release syn­drome (CRS) and immune effec­tor cell–
asso­ci­ated neu­ro­tox­ic­ity syn­drome (ICANS), which are both
poten­tially life-threat­en­ing.27 The inci­dence of these com­pli­ca­
tions across selected tri­als for adult ALL is sum­ma­rized in Table 1.
ited. Current man­age­ment strat­e­gies are based on inter­ven­tion
with cor­ti­co­ste­roids. Risk fac­tors for severe ICANS are less clear
but cor­re­late with prod­uct used, higher inter­leu­kin 6 lev­els, high
dis­ease bur­den, and more severe sys­temic CRS.9,12,19 CART19 cells
read­ily cross the blood-brain bar­rier (BBB, a poten­tial ben­e­fit for
dis­ease targeting as discussed above) and are detect­able in the
cere­bral spi­nal fluid (CSF) in most treated patients. Their pres­
ence in the CSF, how­ever, does not pre­dict for tox­ic­ity.5 Further
stud­ies are needed to deter­mine the mech­a­nism of action, risk
fac­tors, and opti­mal man­age­ment strat­e­gies of neu­ro­logic tox­ic­
ity after CAR T-cell ther­apy.
Role of CART19 vs other ther­a­pies
Clinicians may have sev­eral sal­vage options avail­­able to treat a

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/1/1851774/1frey.pdf by guest on 13 December 2021

It should be noted that dif­fer­ent grad­ing scales were used patient with r/r ALL. Treatment deci­sions need to be indi­vid­u­al­
across tri­als, chal­leng­ing any com­par­i­son of tox­ic­ity. ized based on prior ther­apy, goals of ther­apy, role of poten­tial
consolidative SCT, and fea­tures of the CART19 prod­uct avail­­able
CRS (see Table 2). Comparing out­comes of CART19 stud­ies with out­
Fever, hemo­dy­namic insta­bil­ity, and hyp­oxia are the core clin­i­ comes from inotuzumab (INO-VATE study) and blinatumomab
cal fea­tures of CRS and used in the American Society for Trans- (TOWER study) should be done with cau­tion given dif­fer­ences
plantation and Cellular Therapy con­sen­sus grad­ing scale.28 The in trial design, patient pop­u­la­tions, dis­ease bur­den, and role
inci­dence of CRS is high across all­CART19 stud­ies com­pared of consolidative SCT in the dif­fer­ent stud­ies. In addi­tion, most
with other B-cell malig­nan­cies. The syn­drome can be self-lim­ited effi­cacy out­comes reported from CART19 stud­ies are for those
(requir­ing only sup­port­ive care with anti­py­ret­ics and intra­ve­nous who made it to infu­sion, discounting fail­ures of treat­ment from
flu­ids) or more seri­ous, requir­ing inter­ven­tion with tocilizumab manufactur­ing or inabil­ity to tol­er­ate the treat­ment delay that
(an anti­body to the inter­leu­kin 6 recep­tor) or cor­ti­co­ste­roids. The is inher­ent in autol­og­ ous CART19 treat­ment. Randomized con­
biol­ogy and risk fac­tors for CRS are well under­stood and con­tinue trolled tri­als are lacking and needed to appro­pri­ately com­pare
to inform man­age­ment strat­e­gies, and fur­ther approaches to out­comes. Acknowledging these lim­i­ta­tions, out­comes from
mit­i­gate CRS are being explored in clin­i­cal tri­als. In ALL, dis­ease mature mul­ti­cen­ter CART19 tri­als (CR = 71%-81%; median OS of
bur­den is a strong pre­dic­tor for the sever­ity of CRS.2,5,9,12,19 Some 11 months not reached; median dura­tion of CR = 12.8 months not
cen­ters have adopted a risk-adapted approach in which a lower reached) com­pare favor­ably with out­comes with blinatumomab
dose of cells is given for patients with higher dis­ease bur­dens.9,19 (CR = 35.1%; median OS = 7.7 months, median dura­tion of CR = 7.3
Others have used more inten­ sive cyto-reduc­ tion for patients months) and inotuzumab (CR = 80.7%; median OS = 7.7 months;
with higher dis­ease bur­dens.10 Several cen­ters are eval­u­at­ing the median dura­tion of CR = 4.7 months).6,8,12,30,31
ben­efi ­ t of inter­ven­ing with tocilizumab for ear­lier grades of CRS
com­pared with the approach in early clin­ic ­ al tri­als that treated at
more severe CRS due to con­cerns that ear­lier inter­ven­tion may Table 2. Challenges of CART19 and solu­tions under inves­ti­ga­tion
mit­i­gate response. In 1 study from the Children’s Hospital of Phil-
adelphia, patients were assigned to high and low (<40%) tumor Challenge Solutions under inves­ti­ga­tion
bur­den cohorts. Those with high dis­ease bur­den received tocili- Relapse
zumab for per­sis­tent fevers. Of 70 patients treated, 15 received
  CD19– relapse due to anti­gen Dual targeted approach (CD19 and
early tocilizumab, which, when com­pared ret­ro­spec­tively with a escape CD22)
sim­i­lar group from an ear­lier study, showed a reduc­tion in grade   CD19+ relapse due to loss of 41BB costimulatory domain
4 CRS from 50% to 27%.29 Seattle Children’s reported their expe­ per­sis­tence Humanized CARTs
ri­ence with ear­lier tocilizumab inter­ven­tion resulting in a lower    Immune medi­ ated rejec­tion Manufacturing changes to select
   Exhaustive phe­ no­type for nonexhaustive phe­no­type
inci­dence of severe CRS with­out an impact on response.2 We
and oth­ers have used a frac­tion­ated dos­ing scheme in which the Logistics
total planned CART dose is infused over 3 days. This approach,   Window for leukapheresis Collect and store cells early in
in which sub­se­quent doses are held for early signs of CRS, allows   Disease con­trol dur­ing dis­ease course
for real-time dose mod­i­fi­ca­tion in response to tox­ic­ity with­out manufactur­ing Develop more rapid manufactur­ing
time
an impact on effi­cacy.1,8 “Off-the-shelf” or allo­ge­neic CARTs
Toxicity
Neurologic tox­ic­ity
The ASTCT con­sen­sus grad­ing for ICANS requires assess­ment   CRS and ICANS Earlier inter­ven­tion with
of a patient’s immune effec­tor cell–asso­ci­ated enceph­a­lop­a­thy tocilizumab or cor­ti­co­ste­roids
Fractionated dos­ing scheme: dose
score, level of con­scious­ness, sei­zure activ­ity, focal motor weak­ mod­i­fi­ca­tion in response to early
ness, and cere­bral edema.28 Similar to CRS, neu­ro­logic events tox­ic­ity
occur within the first few weeks of ther­ apy and have been Dose mod­i­fi­ca­tions by dis­ease
reported in all­the stud­ies sum­ma­rized in Table 1. Our under­stand­ bur­den
Novel anticytokine approaches
ing and abil­ity to mit­i­gate neu­ro­logic tox­ic­ity after CART19 is lim­

4  |  Hematology 2021  |  ASH Education Program

 CART19, inotuzumab, blinatumomab, and che­mo­ther­apy in
patients who are not refrac­tory all­have the poten­tial to suc­cess­
fully induce an MRD– remis­sion that can serve as a suc­cess­ful
bridge to SCT. Although suc­cess­ful, there is no evi­dence that an
MRD– remis­sion from a CART prod­uct vs che­mo­ther­apy or other
targeted approach improves RFS after SCT. One report com­
pared out­comes from patients with r/r ALL who achieved MRD–
CR with CART19 or che­mo­ther­apy and were bridged to SCT. No
dif­fer­ence in cumu­la­tive inci­dence of relapse (11.1% vs 12.8%) or
nonrelapse mor­tal­ity was iden­ti­fied.32
The tox­ic­ity of a spe­cific CART19 prod­uct needs to be con­
sid­ered on an indi­vid­u­al­ized basis and used to inform treat­ment
deci­sions with CART19 vs another avail­­able prod­uct. Patients
with high dis­ease bur­den, advanced age, and comorbid car­
dio­vas­cu­lar dis­ease, for exam­ple, may not tol­er­ate antic­i­pated
severe CRS or ICANS, and another approach may be more
appro­pri­ate. On the other hand, a patient with min­i­mal dis­ease
bur­den would be antic­i­pated to have lower risk of these toxic-
ities from CART19. Other spe­cific fac­tors may influ­ence a cli­ni­
cian’s deci­sion to treat with CART19. For a patient with a low
like­li­hood or desire to pro­ceed to an SCT if a CR is obtained,
CART19 prod­ucts that have been shown to have dura­ble remis­
sions with­out consolidative SCT may be favored over other
prod­ucts. A patient with CNS dis­ease his­tory may ben­e­fit from
CART19, which has been shown to cross the BBB and treat dis­
ease in the CSF.
The logis­tics of autol­o­gous CAR T-cell ther­apy, includ­ing
iden­ti­fy­ing an appro­pri­ate win­dow to per­form leukapheresis
and the need to con­trol dis­ease while cells are being manufac-
tured, con­fer a sig­nif­i­cant dis­ad­van­tage com­pared with other
approaches to care, such as inotuzumab or blinatumomab.
Several inves­ti­ga­tions are under way to explore the poten­
tial ben­e­fit of using allo­ge­neic CARTs from healthy donors in
whom the poten­tial for graft-ver­sus-host dis­ease is abro­gated
by T cell receptor knock­out with gene editing tech­niques. In
early stud­ies with this approach, responses have required high
immu­no­sup­pres­sive ther­apy to min­i­mize immune-medi­ated
rejec­tion.33
Table 3. Clinical sce­nar­ios that favor CART19 vs other ther­a­pies
Clinical sce­nario
Recent CNS dis­ease
Allogeneic SCT not pos­si­ble or desir­able
  Relapse occurs after prior SCT
  High risk for com­pli­ca­tions from SCT
  No donor option
  Patient defers SCT
Rapidly pro­gres­sive dis­ease
Frail or older patient with comorbid dis­ease and high dis­ease bur­den
Frail or older patient with comorbid dis­ease and min­i­mal dis­ease bur­den
Prakash Singh Shekhawat
Real-world expe­ri­ence with CART19 for ALL
There is only 1 CART19 prod­uct approved in the United States
and Europe for r/r B ALL, although more prod­ucts are likely to be
approved for this indi­ca­tion over the next few years. Tisagenle-
cleucel is approved to treat patients up to 25 years of age with
B-cell ALL that is refrac­tory or in sec­ond or greater relapse. This
approval is based on out­comes from the mul­ti­cen­ter phase 2
ELIANA study discussed above.6
Real-world reg­is­try data col­lected after the com­mer­cial­i­za­tion
of tisagenlecleucel have sub­stan­ti­ated these out­comes and rep­
re­sent a larger num­ber and more diverse group of patients than
those treated in the piv­otal trial. Information from 255 pedi­at­ric
and young adult patients with relapsed ALL obtained from the
Center for International Bone Marrow Transplant Registry found a
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/1/1851774/1frey.pdf by guest on 13 December 2021
CR rate of 86% with an EFS and OS of 68.6% and 88.5%, respec­
tively, at 6 months.34 This com­pares favor­ably to the ELIANA trial,
which had a CR rate of 81% and an EFS and OS of 72% and 87%,
respec­tively, at 6 months. Importantly, treat­ment in the real world
has been shown to be safe, with fewer patients hav­ing grade 3
or higher CRS (16% vs 48%) and sim­i­lar num­bers hav­ing grade
3 or higher neu­ro­tox­ic­ity (9% vs 12.7%).34 The lower inci­dence in
CRS may reflect prog­ress in man­age­ment or less advanced dis­
ease at time of treat­ment. In addi­tion to val­i­dat­ing effi­cacy and
safety out­comes, real-world data have pro­vided infor­ma­tion on
pop­u­la­tions not eli­gi­ble for the reg­is­tra­tion trial. For exam­ple, 6%
of patients from the Center for International Bone Marrow Trans-
plant Registry were youn­ger than 3 years, and 15% of patients had
prior blinatumomab; both sub­groups would have been excluded
from ELIANA.
Conclusion
At the time of this writ­ing, there is only 1 CAR T-cell prod­uct,
tisagenlecleucel, approved to treat r/r ALL for patients 25 years
and youn­ger, but more approv­als over the next few years are
antic­
i­
pated. Treatment with CART19 yields high and dura­ ble
response rates for adult and pedi­ at­
ric patients with r/r ALL.
The deci­sion to treat a patient with CART19 depends on sev­eral
patient-, dis­
ease-, and prod­ uct-related fac­tors in con­
junc­
tion
CARTs vs alter­na­tive approach
Consider CART19 due to abil­ity to cross BBB and treat the CNS com­
part­ment. Patients treated with che­mo­ther­apy, blinatumomab, or
inotuzumab are at risk for extramedullary relapse or pro­gres­sion.
CART19 should be con­sid­ered if prod­uct with dura­ble remis­sion with­out
SCT is avail­­able.
CART19 may not be fea­si­ble due to need to iden­tify a win­dow for
leukapheresis and con­trol dis­ease while awaiting man­u­fac­ture. Other
approaches should be con­sid­ered.
Patient may not tol­er­ate antic­i­pated severe CRS or ICANS with CART19.
Consider alter­na­tive approaches.
Patient is likely to do well with blinatumomab, inotuzumab, and CART19.
With CART19, side effects of lymphodepletion need to be con­sid­ered,
although low CART-related tox­ic­ity is antic­i­pated. Decision may be
influ­enced by dura­bil­ity of remis­sions observed with CART19 prod­uct
avail­­able.
CAR T for ALL  |  5
with con­sid­er­ation of other ther­a­pies avail­­able and the poten­
tial role of a consolidative SCT. Approaches are being explored
to min­i­mize CD19+ and CD19– relapses and pre­vent or mit­i­gate
tox­ic­ity, which will con­tinue to affect the treat­ment par­a­digm for
ALL over time (see Table 3).
Conflict-of-inter­est dis­clo­sure
Noelle V. Frey: Novartis Research Funding. Kite Pharmaceuticals
and Syndax Consultancy.
Off-label drug use
Noelle V. Frey: nothing to disclose.
Correspondence
Noelle V. Frey, Hospital of the University of Pennsylvania, Abramson
Cancer Center, Perelman Center for Advanced Medicine, 3400
Civic Center Boulevard, Philadelphia, PA 19104; e-mail: noelle​­.frey
@pennmedicine​­.upenn​­.edu.
References
1. Frey NV, Shaw PA, Hexner EO, et al. Optimizing chi­me­ric anti­gen recep­tor
T-cell ther­apy for adults with acute lym­pho­blas­tic leu­ke­mia. J Clin Oncol.
2020;38(5):415-422.
2. Gardner RA, Finney O, Annesley C, et al. Intent-to-treat leu­ke­mia remis­sion
by CD19 CAR T cells of defined for­mu­la­tion and dose in chil­dren and young
adults. Blood. 2017;129(25):3322-3331.
3. Hay KA, Gauthier J, Hirayama AV, et al. Factors asso­ci­ated with dura­ble EFS
in adult B-cell ALL patients achiev­ing MRD-neg­a­tive CR after CD19 CAR
T-cell ther­apy. Blood. 2019;133(15):1652-1663.
4. Jiang H, Li C, Yin P, et  al. Anti-CD19 chi­me­ric anti­gen recep­tor-mod­i­fied
T-cell ther­apy bridg­ing to allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­
tion for relapsed/refrac­tory B-cell acute lym­pho­blas­tic leu­ke­mia: an open-
label prag­matic clin­i­cal trial. Am J Hematol. 2019;94(10):1113-1122.
5. Maude SL, Frey N, Shaw PA, et  al. Chimeric anti­gen recep­tor T cells for
sustained remis­sions in leu­ke­mia. N Engl J Med. 2014;371(16):1507-1517.
6. Maude SL, Laetsch TW, Buechner J, et  al. Tisagenlecleucel in chil­ dren
and young adults with B-cell lym­pho­blas­tic leu­ke­mia. N Engl J Med.
2018;378(5):439-448.
7. Myers RM, Li Y, Barz Leahy A, et  al. Humanized CD19-targeted chi­me­ric
anti­gen recep­tor (CAR) T cells in CAR-naive and CAR-exposed chil­dren and
young adults with relapsed or refrac­tory acute lym­pho­blas­tic leu­ke­mia.
J Clin Oncol. 2021;39(27):3044-3055.
8. Ortíz-Maldonado V, Rives S, Castellà M, et al. CART19-BE-01: a mul­ti­cen­ter
trial of ARI-0001 cell ther­apy in patients with CD19+ relapsed/refrac­tory
malig­nan­cies. Mol Ther. 2021;29(2):636-644.
9. Park JH, Rivière I, Gonen M, et al. Long-term fol­low-up of CD19 CAR ther­apy
in acute lym­pho­blas­tic leu­ke­mia. N Engl J Med. 2018;378(5):449-459.
10. Wang J, Mou N, Yang Z, et al. Efficacy and safety of human­ized anti-CD19-
CAR-T ther­apy fol­low­ing inten­sive lymphodepleting che­mo­ther­apy for
refrac­tory/relapsed B acute lym­pho­blas­tic leu­kae­mia. Br J Haematol.
2020;191(2):212-222.
11. Shah NN, Lee DW, Yates B, et al. Long-term fol­low-up of CD19-CAR T-cell
ther­apy in chil­dren and young adults with B-ALL. J Clin Oncol. 2021;39(15):
­1650-1659.
12. Shah BD, Ghobadi A, Oluwole OO, et al. KTE-X19 for relapsed or refrac­tory
adult B-cell acute lym­pho­blas­tic leu­kae­mia: phase 2 results of the sin­gle-
arm, open-label, multicentre ZUMA-3 study. Lancet. 2021;398(10299):491-
502.
13. Pan J, Zuo S, Deng B, et al. Sequential CD19-22 CAR T ther­apy induces sus-
tained remis­sion in chil­dren with r/r B-ALL. Blood. 2020;135(5):387-391.
14. Shah NN, Highfill SL, Shalabi H, et al. CD4/CD8 T-cell selec­tion affects chi­
me­ric anti­gen recep­tor (CAR) T-cell potency and tox­ic­ity: updated results
from a phase I anti-CD22 CAR T-cell trial. J Clin Oncol. 2020;38(17):1938-
1950.
6  |  Hematology 2021  |  ASH Education Program
15. Singh N, Frey NV, Engels B, et al. Antigen-inde­pen­dent acti­va­tion enhances
the effi­cacy of 4-1BB-costimulated CD22 CAR T cells. Nat Med. 2021;27(5):
­842-850.
16. Wang N, Hu X, Cao W, et al. Efficacy and safety of CAR19/22 T-cell cock­
tail ther­apy in patients with refrac­tory/relapsed B-cell malig­nan­cies. Blood.
2020;135(1):17-27.
17. Pillai V, Muralidharan K, Meng W, et al. CAR T-cell ther­apy is effec­tive for
CD19-dim B-lym­pho­blas­tic leu­ke­mia but is impacted by prior blinatum-
omab ther­apy. Blood Adv. 2019;3(22):3539-3549.
18. Mueller KT, Waldron E, Grupp SA, et al. Clinical phar­ma­col­ogy of tisagen-
lecleucel in B-cell acute lym­pho­blas­tic leu­ke­mia. Clin Cancer Res. 2018;24
­(24):6175-6184.
19. Turtle CJ, Hanafi LA, Berger C, et al. CD19 CAR-T cells of defined CD4+:CD8+
com­po­si­tion in adult B cell ALL patients. J Clin Invest. 2016;126:2123-2138.
20. Finney OC, Brakke HM, Rawlings-Rhea S, et  al. CD19 CAR T cell prod­uct
and dis­ease attri­butes pre­dict leu­ke­mia remis­sion dura­bil­ity. J Clin Invest.
2019;129(5):2123-2132.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/1/1851774/1frey.pdf by guest on 13 December 2021
21. Jacoby E, Nguyen SM, Fountaine TJ, et  al. CD19 CAR immune pres­ sure
induces B-pre­cur­sor acute lym­pho­blas­tic leu­kae­mia lin­e­age switch expos­
ing inher­ent leukaemic plas­tic­ity. Nat Commun. 2016;7:12320.
22. Rabilloud T, Potier D, Pankaew S, et al. Single-cell pro­fil­ing identifies pre-
existing CD19-neg­a­tive subclones in a B-ALL patient with CD19-neg­a­tive
relapse after CAR-T ther­apy. Nat Commun. 2021;12:865.
23. Singh N, Lee YG, Shestova O, et al. Impaired death recep­tor sig­nal­ing in
leu­ke­mia causes anti­gen-inde­pen­dent resis­tance by induc­ing CAR T-cell
dys­func­tion. Cancer Discov. 2020;10(4):552-567.
24. Qin H, Ramakrishna S, Nguyen S, et al. Preclinical devel­op­ment of biva­lent
chi­me­ric anti­gen recep­tors targeting both CD19 and CD22. Mol Ther Onco-
lytics. 2018;11:127-137.
25. Schneider D, Xiong Y, Wu D, et  al. A tan­dem CD19/CD20 CAR lentiviral
vec­tor drives on-tar­get and off-tar­get anti­gen mod­u­la­tion in leu­ke­mia cell
lines. J Immunother Cancer. 2017;5:42.
26. Shah NN, Johnson BD, Schneider D, et al. Bispecific anti-CD20, anti-CD19
CAR T cells for relapsed B cell malig­nan­cies: a phase 1 dose esca­la­tion and
expan­sion trial. Nat Med. 2020;26(10):1569-1575.
27. Frey NV. Chimeric anti­gen recep­tor T cells for acute lym­pho­blas­tic leu­ke­
mia. Am J Hematol. 2019;94(suppl 1):S24-S27.
28. Lee DW, Santomasso BD, Locke FL, et  al. ASTCT con­sen­sus grad­ing for
cyto­kine release syn­drome and neu­ro­logic tox­ic­ity asso­ci­ated with
immune effec­tor cells. Biol Blood Marrow Transplant. 2019;25(4):625-638.
29. Kadauke S, Myers RM, Li Y, et al. Risk-adapted pre­emp­tive tocilizumab to
pre­vent severe cyto­kine release syn­drome after CTL019 for pedi­at­ric B-cell
acute lym­pho­blas­tic leu­ke­mia: a pro­spec­tive clin­i­cal trial. J Clin Oncol.
2021;39(8):920-930.
30. Kantarjian H, Stein A, Gokbuget N, et al. Blinatumomab ver­sus che­mo­ther­
apy for advanced acute lym­pho­blas­tic leu­ke­mia. N Engl J Med. 2017;376:
­836-847.
31. Kantarjian HM, DeAngelo DJ, Stelljes M, et  al. Inotuzumab ozogamicin
ver­sus stan­dard ther­apy for acute lym­pho­blas­tic leu­ke­mia. N Engl J Med.
2016;375(8):740-753.
32. Zhao YL, Liu DY, Sun RJ, et al. Integrating CAR T-cell ther­apy and trans­plan­
ta­tion: com­par­i­sons of safety and long-term effi­cacy of allo­ge­neic hema­to­
poi­etic stem cell trans­plan­ta­tion after CAR T-cell or che­mo­ther­apy-based
com­plete remis­sion in B-cell acute lym­pho­blas­tic leu­ke­mia. Front Immunol.
2021;12:605766.
33. Ben­ja­min R, Graham C, Yallop D, et al. Genome-edited, donor-derived allo­
ge­neic anti-CD19 chi­me­ric anti­gen recep­tor T cells in pae­di­at­ric and adult
B-cell acute lym­pho­blas­tic leu­kae­mia: results of two phase 1 stud­ies. Lan-
cet. 2020;396:1885-1894.
34. Pasquini MC, Hu ZH, Curran K, et al. Real-world evi­dence of tisagenlec-
leucel for pedi­at­ric acute lym­pho­blas­tic leu­ke­mia and non-Hodgkin lym­
phoma. Blood Adv. 2020;4(21):5414-5424.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000225
 ALL: NEW DIRECTIONS FOR ADULT PATIENTS

Acute lymphoblastic leukemia in older adults:

curtain call for conventional chemotherapy?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/7/1852080/7luskin.pdf by guest on 13 December 2021

Marlise R. Luskin
Dana-Farber Cancer Institute, Department of Medical Oncology, Division of Leukemia, Boston, MA

Unlike younger adults with acute lymphoblastic leukemia (ALL), older adults are rarely cured due to a combination of
intrinsic disease resistance and treatment-related toxicities. Novel therapeutics such as inotuzumab ozogamicin, blinatu-
momab, venetoclax, and ABL kinase inhibitors have high activity in ALL and are well tolerated by older adults. Frontline
treatment regimens for older adults using novel therapeutics with reduction or omission of conventional chemotherapy
are being developed with early results demonstrating high remission rates and lower toxicity, but long-term efficacy and
toxicity data are lacking. Collaboration between academic and pharmaceutical stakeholders is needed to develop clini-
cal trials to define the optimal treatment regimens for older adults with ALL.

LEARNING OBJECTIVES
• Understand the role of novel therapeutics vs CC in initial treatment of older adults with Ph-negative ALL
• Understand the role of tyrosine kinase inhibitor treatment with or without CC or novel therapeutics in
initial treatment of adults with Ph-positive ALL

Outcomes of older adults in the era of conventional
chemotherapy
For most of history, acute lymphoblastic leukemia (ALL)
has been an uncompromising, deadly illness regardless of
age, comorbidities, or social circumstance. Then, in 1948,
Sidney Farber announced 5 temporary remissions among
16 children with acute leukemia treated with the folic acid
antagonist, aminopterin.1 Now, after decades of effort by
numerous clinicians, scientists, and cooperative groups,
more than 90% of children in resourced settings are cured
of ALL with chemotherapy.2 Although celebrated, these
pediatric ALL chemotherapy regimens are notable for
length, complexity, and toxicity.
Traditionally thought of as a pediatric disease, approxi-
mately half of ALL diagnoses occur in younger (18–49 years)
and older (≥50 years) adults in roughly equal porportions.3
Unfortunately, the outstanding outcomes in children
have not been replicated in older cohorts. The standard
approach to treating ALL in adults has been conventional
chemotherapy (CC) programs adapted from pediatric
schedules but with more myelosuppressive agents and a
deemphasis on the noncytotoxic agents that feature prom-
inently in pediatric regimens (corticosteroids, vincristine,
and asparaginase). Results in young adults have improved
due to the use of pediatric-like therapy, with more than
70% now achieving long-term survival, but older adults
aged 55 to 60 or more years have consistently fared more
poorly, with less than 20% cured (Table 1).4–9 Older patients
experience more toxicity, leading to dose reductions,
treatment delays, and high rates of early death and treat-
ment-related mortality in remission.4–7,11,12 In addition, high-
risk genetic features are more common, leading to fewer
remissions and frequent relapse.10,16 Age-based dose mod-
ifications and prospective studies of CC designed specifi-
cally for older adults have not improved outcomes.4,5,12–14,17
There remains no accepted standard-of-care CC regimen
for older adults with ALL.
Clinicians caring for older adults with ALL perceive lim-
ited benefit of CC. A 2019 US Medicare analysis revealed
that only 53.5% of patients with ALL aged 66 or more years
received any treatment within 90 days of diagnosis.18 A
2017 US Surveillance, Epidemiology, and End Results analy-
sis reported a median overall survival (OS) of just 4 months
among adults 60 years or older diagnosed with ALL since
1980, with minimal progress over 3 decades (3-year OS
increasing from 10% to 16%).19 In summary, the chemother-
apy regimens that cure children are not a good match for
older adults.10,16

Prakash Singh Shekhawat

ALL curtain call chemo | 7
Table 1.  Outcomes of older adults treated with CC, select tri­als (see also Gökbuget10)

Trial Age, y N CR, % Early ­mor­tal­ity,% OS, % (95% CI) Death in CR, among those achiev­ing CR, %
Adult tri­als, out­comes by age
  CALGB 91115 All 185 85 8 3 years: 43 (36-50) 8
≥60 35 77 17 17 (9-31) —
30-59 42 84 — 40 (30-51) —
<30 39 90 — 57 (46-68) —
  ECOG 2993/UKALL XII7,11 55-65 100 73 18 5 years: 21 (12-20) 23
<55 1814 93 4 41 (39-43) —
 Hyper-CVAD 12
≥60 122 84 10 5 years: 20 34
<60 409 92 2 48 7

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/7/1852080/7luskin.pdf by guest on 13 December 2021

Prospectively designed for older adults
  DFCI reg­i­men13 51-75 30 67 13 2 years: 52 (33-68) —
60-75 13 76 — 70 (42-86)
51-59 17 58 — 31 (10-55)
 GMALL14 55-85 268 76 14 5 years: 23 6
55-65 — — 7 — —
65-75 — — 14 — —
≥75 — — 37 — —
  PETHEMA ALLOLD0715 >55 56 74 13 32 (at last fol­low-up, 3
median 11.4 months)
DFCI, Dana-Farber Cancer Institute.

Novel approaches to Philadelphia chro­mo­some–neg­a­tive of older adults will allow toxic CC to be de-esca­lated or omit­ted,
ALL in older adults resulting in both improved response rates and less tox­ic­ity.

CASE 1
A 78-year-old man with cor­o­nary artery dis­ease, hyper­ten­sion,
and hyper­ lip­
id­
emia devel­ oped sev­ eral weeks of fatigue and
weight loss. Prior to ill­ness, he vis­ited the gym reg­u­larly engag­
ing in vig­or­ous exer­cise. Laboratory work revealed pan­cy­to­pe­nia
(white blood cell count, 2.9 K/µL; hemo­glo­bin, 11.2 g/dL; plate­
lets, 86 K/µL) with 7% cir­cu­lat­ing blasts. He was diag­nosed with
B-cell acute lymphoblastic leukemia (B-ALL). Immunophenotype
was CD45+, CD34+, HLA-DR+, TdT+, CD19+ (var­ i­
able), CD22+,
CD10−, and CD20−. Karyotype was com­plex with no evi­dence
of the Philadelphia chro­mo­some. Next-gen­er­a­tion sequenc­ing
revealed a path­o­genic muta­tion in TP53. In 2019, he was referred
to our aca­demic cen­ter.
The devel­op­ment of novel che­mo­ther­apy agents—blinatu-
momab and inotuzumab ozogamicin (IO)—for relapsed B-ALL
has led to a reimagining of the treat­ment for older adults with
ALL. IO is a CD22 mono­clo­nal anti­body cova­lently linked to cali-
cheamicin, a cyto­toxic agent (anti­body–drug con­ju­gate). Blina-
tumomab is a bispecific T-cell engager targeting CD19 and CD3.
Each is now approved by the US Food and Drug Administration
(FDA) for relapsed and refrac­tory (R/R) ALL based on phase 3
ran­dom­ized trial data show­ing supe­ri­or­ity of the novel agent
to sal­vage cyto­toxic che­mo­ther­apy in the R/R set­ting.20,21 It is
hoped that using IO and blinatumomab in the first-line treat­ment
IO plus CC
Investigators at MD Anderson Cancer Center (MDACC) have been
test­ing IO plus mini-hyper-CVD (cyclophosphamide, vincristine,
dexamethasone), a lower-inten­sity CC reg­i­men, in older adults
(≥60 years) with untreated Philadelphia chro­mo­some–neg­a­tive
(Ph–) CD22+ B-ALL in a sin­gle-cen­ter phase 2 trial (NCT01371630,
Table 2). In the orig­i­nal design, IO was admin­is­tered on day 3 of
the first 4 of 8 planned CC cycles followed by 36 months of POMP
(6-mercaptopurine, vincristine, methotrexate, prednisone) main­
te­nance. In 2018, MDACC inves­ti­ga­tors reported high response
rates (98% overall response rate [ORR], 96% minimal residual
disease-negativity [MRD-negativity]) and no early mor­tal­ity in
the first 52 patients treated.22 An update of the trial in 2020 (with
70 patients enrolled) was nota­ble for con­tin­ued high response
rates (98% ORR, 96% MRD negativity, with no early deaths) and
an encour­ag­ing 3-year com­plete remis­sion (CR) dura­tion and OS
of 79% and 56%, respec­tively.23
Notable toxicities asso­ ci­
ated with the IO plus mini-hyper-
CVD reg­i­men have included fre­quent prolonged throm­bo­cy­to­
pe­nia (81%), hep­a­to­tox­ic­ity (17% grade ≥3 hyperbilirubinemia,
9% veno-occlu­sive dis­ease), and infec­tions (41% dur­ing induc­
tion, 70% dur­ing con­sol­i­da­tion).23 Death in remis­ sion due to
treat­ment-related tox­ic­ity has been the major chal­lenge. At the
most recent update, 34% (24/70) of responding patients had
died in CR due to sep­sis, veno-occlu­sive dis­ease, or sec­ond­ary
mye­loid malig­nan­cies. Mortality was more com­mon in patients
70 years or older who com­prised ~40% of enrolled patients. In
response to an ini­tial expe­ri­ence with tox­ic­ity, pro­to­col mod­i­

8  |  Hematology 2021  |  ASH Education Program

Table 2.  Novel approaches to Ph– ALL in older adults: published and ongo­ing tri­als*

Regimen-related
Reference Regimen Phase N Line Age, y deaths Response Survival
Kantarjian et al IO + mini-hyper- 2 70 First ≥60 0% early mor­tal­ity 98% ORR Continuous CR
(2018)22 CVD (Blina 34% mor­tal­ity in 88% CR (96% MRD 3 years: 79%
Short et al (2020)23 con­sol­i­da­tion) remis­sion neg­a­tive) Median CCR NR
NCT01371630 OS
3 years: 56%
Median OS 62
months
EWALL-INO IO + mild-inten­sity 2 — First ≥55 — — —
NCT03249870 che­mo­ther­apy
CC con­sol­i­da­tion
Stelljes et al IO induc­tion 2 36 First ≥56 0% early mor­tal­ity 100% CR/CRi EFS

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/7/1852080/7luskin.pdf by guest on 13 December 2021

(2020)24 CC con­sol­i­da­tion (78% MRD 1 year: 87% (95%
GMALL (INITIAL-1) neg­a­tive) CI, 70-100)
NCT03460522 OS
1 year: 87% (95%
CI, 70-100)
Jain et al (2019)25 Venetoclax + 1b/2 19 R/R (8) First (11) All ages ≥60 0% early mor­tal­ity 91% CR (100% —
DFCI/MDACC IST mini-hyper-CVD MRD neg­a­tive)
NCT03319901
Advani et al Blina induc­tion 2 29 First ≥65 0% 66% CR/CRi (92% DFS
(2018)26 Blina con­sol­i­da­tion MRD neg­a­tive) 1 year: 56% (95%
SWOG 1318 CI, 58-90)
NCT02143414 OS
1 year: 65% (95%
CI, 43-80)
Alliance 041703 IO induc­tion 2 — R/R First All ages ≥60 Results pend­ing Results pend­ing Results pend­ing
NCT03739814 Blina con­sol­i­da­tion
*Results at most recent pub­li­ca­tion.
Blina, blinatumomab; CCR, con­tin­u­ous com­plete remis­sion; CRi, complete remission with incomplete count recovery; NR, not reached.

fi­
ca­tions have included IO dose frac­ tion­
ation and reduc­ tion
(to address liver tox­ic­ity) and a dras­tic decrease in the num­ber
of recommended CC cycles (orig­i­nally 8, decreased to 4 for
patients aged 60-69 years and to 2 for patients aged ≥70 years).
The Euro­pean phase 2 EWALL (European Working Group for
Adult ALL)-INO study (NCT03249870) is also explor­ing IO in com­
bi­na­tion with low-inten­sity che­mo­ther­apy for untreated patients
55 years or older with CD22+ ALL. Patients receive IO plus low-
inten­sity che­mo­ther­apy induc­tion for 2 cycles followed by IO-
free che­mo­ther­apy con­sol­i­da­tion and main­te­nance.
In con­trast to the com­bi­na­tion approach taken by MDACC
and EWALL, the German Multicenter Study Group for Adult
Acute Lymphoblastic Leukemia (GMALL) study group is study­
ing a sequen­ tial approach. In the phase 2 INITIAL-1 study
(NCT03460522), untreated older adults (aged >55 years) receive
3 cycles of IO monotherapy followed by CC con­sol­i­da­tion (Table
2). A 100% CR was reported among the first 31 patients treated
(78% MRD neg­a­tive), with the major­ity (29/31) ­able to com­plete
all 3 IO induc­tion cycles with no early deaths.24 Long-term effi­
cacy and tox­ic­ity of the reg­i­men are not yet known.
Venetoclax plus CC
The BCL2 inhib­i­tor venetoclax, an agent with dem­on­strated pre­
clin­i­cal and clin­i­cal activ­ity in relapsed ALL, is also being stud­ied in trial (Alliance 041703 NCT03739814) is inves­ti­gat­ing IO induc­tion
the front­line in com­bi­na­tion with mini-hyper-CVD (NCT03319901,
Table 2).27,28 A phase 1b study of venetoclax plus mini-hyper-CVD
for relapsed/refractory (R/R) (n = 8) and newly diag­ nosed pa-
tients 60 years or older (n = 11) with ALL Prakash Singhsafety
dem­on­strated Shekhawat
(no agent, whereas blinatumomab is more effec­tive and less toxic in
dose-limiting toxicities (DLTs) or early mor­tal­ity), with venetoclax
600 mg daily declared the recommended phase 2 dose.25 Nota-
bly, there was no evi­dence of prolonged cytopenias or liver tox­ic­
ity. The reg­i­men was par­tic­u­larly effec­tive in newly diag­nosed pa-
tients, with 91% (10/11) achiev­ing an MRD-neg­a­tive CR, includ­ing
6 patients with a TP53 muta­tion. Because 9 of 10 respond­ers were
con­sol­i­dated with allo­ge­neic hematopoietic stem cell transplant
(HSCT), there are lim­ited data on late tox­ic­ity. The trial has now
moved to phase 2 (with venetoclax 400  mg) and is cur­ rently
enroll­ing adults 60 years or older with newly diag­nosed Ph– ALL.
A dis­tinct advan­tage of this reg­i­men is that it can be offered to
patients with both B- and T-cell ALL, in con­trast to approaches
that rely on B-lin­e­age restricted antibodies and anti­body–drug
con­ju­gates. Venetoclax is not approved by the FDA for ALL.
Closing the cur­tain on che­mo­ther­apy for Ph– ALL?
Given the effi­cacy of novel agents, a che­mo­ther­apy-free approach
may be pos­si­ble. The US National Clinical Trial Network (NCTN)
trial Southwest Oncology Group (SWOG) 1318 phase 2 study of
blinatumomab induc­tion followed by blinatumomab con­sol­i­da­
tion for newly diag­nosed Ph– B-ALL (aged >65 years) reported a
66% CR rate (92% MRD neg­a­tive) among the first 29 patients with
no early deaths (NCT02143414, Table 2).26 Another phase 2 NCTN
followed by blinatumomab con­sol­i­da­tion (Table 2). This approach
seeks to use each novel agent to max­i­mum advan­tage. IO is ac-
tive regard­less of dis­ease bur­den, mak­ing it an ideal induc­tion

ALL cur­tain call chemo | 9

set­tings of lower dis­ease bur­den and thus a good agent for con­ mer­ase chain reac­tion shows p210 tran­script more than 50%
sol­i­da­tion ther­apy.15,29 international scale (IS).

Encore for che­mo­ther­apy

Blinatumomab and IO achieve high response rates in the R/R set­ The Philadelphia chro­mo­some is pres­ent in approx­i­ma­tely
ting but with short dura­bil­ity.20,21 Although the dura­bil­ity of sin­gle- half of ALLs in older adults.31 Historically asso­ ci­
ated with an
agent novel ther­apy in the upfront set­ting is not known, relapses adverse prog­no­sis, the devel­op­ment of ABL kinase inhib­i­tors has
driven by selec­tive pres­sure would be expected. It is hypoth­e­ resulted in fre­quent cures in youn­ger adults and in older adults
sized that low-inten­sity CC inte­grated into induc­tion and/or con­ has enabled prolonged remis­sions, negat­ing the adverse prog­
sol­id
­ a­tion (as being pur­sued by EWALL-INO and GMALL INITIAL-1 no­sis.32-34 Older adults with ALL are ideal can­di­dates for low-
stud­ies) may con­tinue to ben­e­fit older patients receiv­ing novel inten­sity CC and che­mo­ther­apy-free tyro­sine kinase inhib­i­tor
agents by apply­ ing broader anti­ leu­ke­
mic pres­ sure and dimin- (TKI)–based pro­to­cols.
ishing oppor­ tu­ni­
ties for clonal escape. Novel com­ bi­
na­tion ap-
proaches (such as being stud­ied in A041703) may offer the best TKI plus CC

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/7/1852080/7luskin.pdf by guest on 13 December 2021

of both worlds via a mul­ti­pronged attack with­out che­mo­ther­apy
toxicities. MDACC has added blinatumomab con­ sol­
i­
da­tion to
their IO plus mini-hyper-CVD pro­to­col. Venetoclax plus IO will also
soon be tested in R/R B-ALL due to non­over­lap­ping toxicities and
pre­clin­i­cal evi­dence for syn­ergy.30 If shown to be safe, it may be
use­ful as a front­line reg­i­men. For T-cell ALL, a che­mo­ther­apy-free
option does not yet exist. More research to iden­tify ther­a­peu­tic
vulnerabilities and novel ther­ap ­ eu­tics for this sub­type is needed.
CASE 1 (con­tin­ued)
The patient was treated with venetoclax plus mini-hyper-CVD
on an inves­ti­ga­tional pro­to­col and achieved an MRD-neg­a­tive
CR after 1 cycle. He com­pleted 4 cycles of venetoclax plus
mini-hyper-CVD, after which he was bridged to venetoclax
plus POMP main­te­nance due to tox­ic­ity from CC. He con­tin­
ued POMP main­te­nance for 2 years and remains in an ongo­ing
remis­sion. (See Figure 1 for approach to Ph– ALL in older adults.)
Novel approaches to Ph+ ALL in older adults
CASE 2
A 73-year-old woman with a his­tory of hyper­ten­sion and rheu­
ma­toid arthri­tis was asymp­tom­atic but found to have an abnor­
mal com­plete blood count on rou­tine test­ing: white blood cells
(19 K/µL) with 39% cir­cu­lat­ing blasts that are CD45+, CD34+,
HLA-DR+, TdT+, CD19+, CD10+, and CD20−. Karyotype shows
t(9;22) with addi­tional abnor­mal­i­ties. Reverse tran­scrip­tion poly­
The EWALL has stud­ied TKI plus low-inten­sity CC in adults 55
years or older (NCT028889777, Table 3). Second-gen­er­a­tion TKIs
were com­bined with low-inten­sity induc­tion and con­sol­i­da­tion
CC in the con­sec­u­tive EWALL-PH01 (dasatinib) and EWALL-PH02
(nilotinib) pro­to­cols. Both pro­to­cols resulted in high CR rates
(>95%) with­out induc­tion mor­tal­ity. Long-term sur­vival was par­
tic­u­larly encour­ag­ing in the EWALL-PH02, in which allo­ge­neic
HSCT was pur­sued in 30% (24/79) of patients (4-year OS, 47%; 61%
and 39% in transplanted vs nontransplanted, respec­tively).35,36
Importantly, the Group for Research on Adult Acute Lymphoblas-
tic Leukemia (GRAAPH)-2005 trial, which enrolled youn­ger adults
(18-59 years), dem­on­strated that reduced-inten­sity CC induc­tion
was asso­ci­ated with fewer induc­tion deaths and equiv­a­lent effi­
cacy com­pared with stan­dard-inten­sity CC induc­tion, question-
ing the ben­e­fit of inten­sive CC induc­tion for any patient (regard­
less of age) with Ph+ ALL in the TKI era.37
Curtain call on che­mo­ther­apy for Ph± ALL?
The Ital­ian Gruppo Italiano Malattie EMatologiche dell’Adulto
(GIMEMA) group has pioneered a che­mo­ther­apy-free induc­tion
approach for Ph+ ALL (Table 3). Imatinib plus cor­ti­co­ste­roids
induced uni­ver­sal hema­to­logic remis­sions, although with lim­
ited depth (only 1 of 29 achieved a com­plete molec­u­lar remis­
sion) and dura­bil­ity (median dura­tion of hema­to­logic response
of 8 months).38 Subsequently, the more potent TKI dasatinib
was com­bined with cor­ti­co­ste­roids, with more patients achiev­
ing deep molec­u­lar remis­sions (52.1% achieved BCR-ABL level
<10−3 by day 85).39,40 The orig­i­nal GIMEMA stud­ies were induc­
tion-only stud­ies, with sub­se­quent con­sol­i­da­tion ther­apy for
each patient dic­tated by the treating phy­si­cian. The US NCTN
Alliance 10701 phase 2 study (NCT01256398, Table 3) tested a
che­mo­ther­apy-free dasatinib plus ste­roid induc­tion followed

• Careful assessment of comorbidities and performance status.

• Formal geriatric assessment if able.
• Enroll in clinical trial of novel agent therapy, wherever possible.
• If treating with conventional chemotherapy, apply dose reductions as appropriate
(particularly for asparaginase).
• Ensure CNS prophylaxis is administered per regimen.
• Monitor response per protocol and include MRD assessment (flow cytometry or
molecular).
• Consider referring for allogeneic HSCT if fit and high-risk genetics or persistent
MRD.

Figure 1.  How I treat Ph– ALL in older adults.

10  |  Hematology 2021  |  ASH Education Program

Table 3.  Novel approaches to Ph+ ALL in older adults: published and ongo­ing tri­als*

Regimen-
Reference Regimen Phase N Line Age, y related deaths Response Survival
Rousselot et al Dasatinib 140 mg 2 71 First ≥55 4% (3/71) 96% CR 27% (95% CI, 17-37)
(2016)35 daily + low- induc­tion 65% 3-log ­ 5-year EFS
EWALL PH-01 inten­sity CC 12% (6/71) reduc­tion in 36% (95% CI, 25-47)
NCT028889777 treat­ment- BCR-ABL 5-year OS
related ­ 7 received HSCT
mor­tal­ity 75% of relapses T315I
in CR
Ottmann et al Nilotinib 400 mg 2 79 First ≥55 1% (1/79) 94.4% CR 42% 4-year EFS
(2018)36 twice daily + induc­tion 47% 4-year OS (61%
EWALL PH-02 low-inten­sity 11 died in CR (6 transplanted, 39%
NCT028889777 CC after HSCT) nontransplanted)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/7/1852080/7luskin.pdf by guest on 13 December 2021

24 received alloHSCT, 3
received autoHSCT
Vignetti et al Imatinib 800 mg 2 29 First >60 0% induc­tion 100% CR 48% (95% CI, 28-69)
(2007)38 daily + 2 died in CR 1/27 CMR 1-year DFS
GIMEMA LAL0101-B pred­ni­sone 74% (95% CI, 54-94)
(induc­tion 1-year OS
pro­to­col only)
Foà et al (2011)39 Dasatinib 70 mg 2 53 First ≥18 0% induc­tion 100% CR 69% (95% 61-79)
GIMEMA LAL1205 twice daily + (22%> 60) 52.1% 3-log 20-month OS
pred­ni­sone reduc­tion in 51% (95% CI, 44-59)
(induc­tion BCR-ABL by 20-month DFS
pro­to­col only) day 85
Martinelli et al Ponatinib 45 mg 2 42 First >60 or unfit 0% induc­tion 95.5% CR 87.5% (95% CI, 76.5%-
(2017)40 daily + 1 death in 45% CMR at 24 99.9%) 1-year OS
GIMEMA LAL1811 pred­ni­sone fol­low-up weeks
NCT01641107 (induc­tion pro­to­ ponatinib
col only)
Luskin et al (2019)42 Dasatinib 140 + 1 8 First >50 0% induc­tion 100% CR Not reported
DFCI IST asciminib (dose
NCT03595917 esca­la­tion)
(induc­tion
pro­to­col only)
Foà et al (2020)43 Dasatinib 140 + 2 63 First ≥18 1.6% induc­tion 95% CR With median fol­low-up
GIMEMA LAL2116 pred­ni­sone (median 54) (1 patient) 18 months
NCT02744768 Blinatumomab No reg­i­men 95% (95% CI, 90-100) OS
con­sol­i­da­tion related 88% (95% CI, 80-97) DFS
deaths in CR
Wieduwilt et al Dasatinib 140 + 2 64 First ≥18 (median 0% induc­tion 97% CR 43% 3-year DFS
(2018)41 dexa­meth­a­ 60) deaths (55% allo, 43% auto, 46%
NCTN/Alliance 10701 sone induc­tion chemo)
NCT01256398 CC con­sol­i­da­tion 55% 3-year OS (63% ≤ 60
Allo- vs autoHSCT years, 49%> 60 years)
(75% allo, 71% auto, 55%
chemo)
Relapse
25% allo, 43% auto, 37%
chemo
NCTN/EA9181 TKI + pred­ni­sone 3 — First ≥18-75 Results Results Results pend­ing
NCT04530565 induc­tion pend­ing pend­ing
TKI + hyper-
CVAD vs TKI +
blinatumomab
con­sol­i­da­tion
SWOG 1318 Dasatinib + 2 — First ≥65 Results Results Results pend­ing
NCT02143414 pred­ni­sone pend­ing pend­ing
induc­tion
Dasatinib +
blinatumomab
con­sol­i­da­tion
*Results at most recent pub­li­ca­tion.
CMR, complete molecular response; DFS, dis­ease-free sur­vival; EFS, event-free sur­vival.
Prakash Singh Shekhawat
ALL cur­tain call chemo | 11
by dasatinib plus CC con­sol­i­da­tion and then trans­plant (allo­ge­
neic vs autol­o­gous HSCT or che­mo­ther­apy based on avail­abil­ity
of an human leukocyte antigen (HLA)-matched donor and fit­
ness).41 Deep remis­sions were con­sis­tently obtained (97% CR,
75% ulti­mately achiev­ing a com­plete molec­u­lar response) with
3-year OS of 55% (49% in patients aged >60 years).
Although the GRAAPH-2005 study showed that inten­sive CC
induc­tion was not advan­ta­geous com­pared with lower-inten­sity
CC induc­tion in the con­text of TKIs, the ben­e­fit of adding CC to
TKI-based con­sol­i­da­tion, par­tic­u­larly in patients bridg­ing to allo­
ge­neic HSCT, is not known.
Approaches to pre­vent relapse in Ph± ALL
Despite improve­ments with TKI-based ther­apy, relapse remains
com­mon, with T315I-driven relapses a par­tic­u­lar men­ace.35,39 Ap-
proaches to address this issue include use of ponatinib, which
is active against T315I; dual ABL kinase inhi­bi­tion with dasatinib
and asciminib; and addi­tion of novel agent con­sol­i­da­tion with
blinatumomab.
The GIMEMA group has extended its che­mo­ther­apy-free induc­
tion approach to ponatinib plus ste­roids in older adults (aged ≥60
years), with encour­ ag­ing early results (NCT01641107, Table 3).40
Although tox­ic­ity appears rea­son­able with dose reduc­tions (only
15/42 remained at 45 mg daily at week 24), the risk of car­dio­vas­cu­
lar tox­ic­ity with ponatinib remains a con­cern in older patients. The
planned EWALL-03 study (NCT04688983) is ran­dom­iz­ing patients
to imatinib vs ponatinib in com­bi­na­tion with low-inten­sity CC and
may pro­vide insight regard­ing risk/ben­e­fit ratio of later-gen­er­a­tion
TKIs. A “risk-adapted” approach using early response mile­stones to
select which patients require esca­la­tion of ther­apy to more potent
(and more toxic) later-gen­er­at­ ion TKIs would be attrac­tive.
Combination ABL kinase inhi­bi­tion is being explored in a phase
1 study treating newly diag­nosed adults 50 years or older with
Ph+ ALL with dasatinib and asciminib, a novel myristoyl bind­ing
pocket inhib­i­tor (NCT03595917, Table 3). The hypoth­e­sis is that
dual ABL inhi­bi­tion may induce deeper remis­sions and pre­vent
T315I relapses. Thus far, the reg­i­men appears well tol­er­ated, with
asymp­ tom­ atic amy­
lase and lipase ele­ va­tions being the most
nota­ble adverse event.42 This study is ongo­ing to estab­lish tol­er­
a­bil­ity and recommended dose.
The GIMEMA D-ALBA trial added dasatinib-blinatumomab
con­sol­i­da­tion to dasatinib-ste­roid induc­tion with prom­is­ing
 Careful assessment of comorbidities and performance status.
 Formal geriatric assessment if able.
 Enroll in clinical trial of novel agent therapy, wherever possible.
 I prefer a chemotherapy-free TKI-based induction using imatinib or dasatinib, or
follow published regimen (Table 3).
 Ensure CNS prophylaxis is administered per regimen.
 Monitor response per protocol and include MRD assessment (BCR-ABL
transcript).
 Escalate treatment (later-generation TKI or additional novel agents) if
inadequate response.
 Refer for consideration of allogeneic HSCT if fit.
Figure 2.  How I treat Ph+ ALL in older adults.
12  |  Hematology 2021  |  ASH Education Program
early results (NCT02744768, Table 3).43 The US NCTN SWOG
1318 Cohort 2 (NCT02143414) is study­ing a sim­i­lar approach of
dasatinib-ste­roid induc­tion followed by blinatumomab con­sol­
i­da­tion. A study of ponatinib-ste­roid induc­tion followed by bli-
natumomab con­ sol­
i­
da­
tion is also planned (NCT04722848). An
open ques­tion is whether blinatumomab con­sol­i­da­tion should
be applied to all­patients or used in a risk-adapted man­ner. Fre-
quency of cen­tral ner­vous sys­tem (CNS) relapse will also need to
be mon­i­tored over time in patients treated entirely with TKIs and
novel agents (with only intrathecal (IT) che­mo­ther­apy admin­is­
tered to pre­vent CNS relapse).
Other active agents in ALL, includ­ing IO and venetoclax, also
require eval­u­a­tion as part of treat­ment of Ph+ ALL. The Alliance
041703 is plan­ning to add an arm for Ph+ patients test­ing an IO
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/7/1852080/7luskin.pdf by guest on 13 December 2021
plus ponatinib induc­tion.
The US NCTN is cur­rently accru­ing to EA9181 (NCT04530565),
a phase 3 study of patients up to age 70 years who receive
che­mo­ther­apy-​free induc­tion of TKI plus ste­roids and then are
ran­ dom­ ized to TKI (dasatinib or ponatinib) plus hyper-​
cyclo­ phosphamide, vincristine, doxorubicin, dexamethasone
(CVAD) vs TKI plus blinatumomab con­sol­i­da­tion to com­pare CC
vs novel agent con­sol­i­da­tion. Eligible patients undergo trans­
plan­ta­tion when MRD neg­a­tive, and all­patients receive high-
dose meth­o­trex­ate for CNS pro­phy­laxis.
CASE 2 (con­tin­ued)
The patient was treated with dasatinib and asciminib on a clin­
i­cal trial. She tol­er­ated ther­apy well ini­tially but had an inad­e­
quate response after 3 cycles of ther­apy (CR but MRD+ by flow
cytom­et­ry, BCR/ABL 6.3% IS). She required mul­ti­ple dasatinib
dose reduc­tions due to recur­rent pleu­ral effu­sions. She was
transitioned to ponatinib 30  mg daily and con­tin­ues in remis­
sion (BCR-ABL p210 tran­script unde­tect­able) now 2 years from
diag­no­sis. (See Figure 2 for approach to Ph– ALL in older adults.)
Additional issues in advanc­ing care for older adults
with ALL
The out­look for older adults diag­nosed with ALL is improv­ing
rap­idly. Additional issues fac­ing the field are sum­ma­rized in
Figure 3.
 • Assess and study MRD and other surrogate end points.
• Nonrestrictive inclusion criteria.
• Include quality-of-life and geriatric assessment measures.
• Design strategies to increase recruitment of participants from underrepresented
groups including nonwhite and rural patients.
• Academic and pharmaceutical industry collaborations.
Figure 3.  Important components for ALL clinical trials in older adults.
Curability of older adults
Although sur­vival is poor among older adults treated with CC,
a few fit older patients may tol­er­ate and be cured by CC reg­
i­mens.5,11,12,44,45 In con­trast, there is no evi­dence yet that novel
agent approaches are cura­tive, although they may induce dura­
ble MRD-neg­ a­tive remis­sions. Patients responding to novel
approaches may become eli­gi­ble for cura­tive-poten­tial allo­ge­
neic HSCT, but the role of HSCT is not clear and has not been
con­firmed to be supe­rior to CC for older adults with Ph– ALL
in first CR in the CC era.46 The Alliance/CALGB 10701 study has
attempted to address this ques­tion for Ph+ ALL by assigning
adult patients (aged 18-69 years) to reduced intensity condition-
ing (RIC) allo­ge­neic HSCT vs autol­o­gous HSCT based on avail­
abil­ity of a fully matched donor (Table 3).41 Although 3-year OS
was sim­i­lar between those receiv­ing allo­ge­neic HSCT and other
con­sol­i­da­tion approaches, patients assigned to the allo­ge­neic
approach were less likely to relapse.
Need for ran­dom­ized com­par­i­son
Results of novel agent–based reg­i­mens tested in nonrandom-
ized stud­ies are extremely encour­ag­ing, with strong evi­dence
for supe­ri­or­ity com­pared with his­tor­i­cal CC reg­i­mens based
on high response rates and low tox­ic­ity. The next step will be
to for­mally com­pare a novel and a “stan­dard” CC approach in
Ph– ALL, and thus the NCTN is design­ing a ran­dom­ized phase
2 study of IO plus mini-hyper-CVD (the novel reg­i­men with the
most expe­ri­ence to date) vs age-adjusted hyper-CVAD (Alliance
Negative
Figure 4. Prakash Singh Shekhawat
Treating Ph-negative B-cell ALL in older adults: future trial landscape.
042001). The objec­tive of this trial will be to estab­lish a novel
agent–based bench­mark reg­i­men to which other novel agent
reg­i­mens can be sub­se­quently com­pared (Figure 4).
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/7/1852080/7luskin.pdf by guest on 13 December 2021
Surrogate end points and risk-adapted approaches
Early response end points (such as MRD neg­a­tiv­ity) should be
stud­ied to con­firm validity as sur­ro­gates for sur­vival in older
adults. This could facil­i­tate rapid com­ple­tion of tri­als as well as
devel­op­ment of risk-adapted approaches.
Equity
Novel agents are expen­sive and likely to increase disparities be-
tween resourced and underresourced set­tings. Advocacy for im-
proved access is needed.
Representative enroll­ment and geri­at­ric assess­ment
Almost every older adult has a med­i­cal prob­lem other than ALL.
Trial eli­gi­bil­ity should be designed to allow rep­re­sen­ta­tive enroll­
ment. Unnecessarily restric­ tive inclu­sion cri­te­
ria that exclude
patients based on advanced age, mildly com­ pro­
mised organ
func­tion, or “clin­i­cally insig­nif­i­cant” con­cur­rent malig­nan­cies
should be avoided. Geriatric assess­ment (GA) should be rou­tinely
incor­po­rated into ALL tri­als of older adults (with finan­cial sup­port
to ensure com­ple­tion). GA has been shown to be fea­si­ble in coop­
er­a­tive group can­cer tri­als, includ­ing in an Alliance trial for acute
mye­loid leu­ke­mia.47,48 The phase 2 NCTN trial in devel­ op­ment
plans to incor­po­rate GA into its study design.
winner
ALL cur­tain call chemo | 13
Quality-of-life assess­ment
Quality-of-life end points should be given impor­tance. Incremen-
tal improve­ments in effi­cacy may not be “worth it” to an older
adult if it is asso­ci­ated with sig­nif­i­cant tox­ic­ity and treat­ment
com­plex­ity. Similarly, new reg­i­mens with equiv­a­lent responses
but bet­ter tol­er­a­bil­ity may war­rant fur­ther devel­op­ment.
Cooperation to clob­ber the con­ven­tional approach
By far the big­gest chal­lenge in adult ALL will be to effi­ciently
design and con­duct log­ic ­ al, sequen­tial stud­ies to answer key
ques­tions and make sure patient care “keeps up” with sci­en­
tific advances. In a rare dis­ease, this is no small task. Trials will
need to be devel­oped col­lab­o­ra­tively, opened widely with eq-
uity con­sid­ered, and conducted effi­ciently so that ques­tions
can be iter­a­tively asked and answered for the bet­ter­ment of
patients. In the United States, a NCTN work­ing group involv­ing
rep­re­sen­ta­tives from Alliance, ECOG, and SWOG has formed to
coor­di­nate efforts.
Conclusion
An older adult diag­nosed with ALL in 2021 has more prom­is­ing
treat­ment options than an adult diag­nosed even 5 years ago. The
first task is to expe­di­tiously replace the unac­cept­able tra­di­tional
CC approaches with novel approaches via prac­tice chang­ing
clin­i­cal tri­als. Then we must work to fur­ther define the opti­mal
treat­ment for all­ patients con­sid­er­ing indi­vid­ual (age, comorbid-
ities, and frailty) and dis­ease (immunophenotype, genet­ics, re-
sponse mile­stones) fea­tures and per­sonal pref­er­ences.
Conflict-of-inter­est dis­clo­sure
Marlise R. Luskin: dis­closes no rel­e­vant con­flict of inter­est.
Off-label drug use
Marlise R. Luskin: only clinical trials discussed.
Correspondence
Marlise R. Luskin, Dana-Farber Cancer Institute, Department of
Medical Oncology, Division of Leukemia, 450 Brookline Ave, Dana
2056, Bos­ton, MA 02215; e-mail: marlise_luskin@dfci​­.harvard​­.edu.
References
1. Farber S, Diamond LK. Temporary remis­sions in acute leu­ke­mia in chil­dren
pro­duced by folic acid antag­o­nist, 4-aminopteroyl-glutamic acid. N Engl J
Med. 1948;238(23):787-793.
2. Pui C-H, Yang J-J, Hunger S-P, et al. Childhood acute lym­pho­blas­tic leu­ke­
mia: prog­ress through col­lab­o­ra­tion. J Clin Oncol. 2015;33(27):2938-2948.
3. Surveillance, Epidemiology, and End Results (SEER) Program. Cancer stat
facts: leu­ke­mia—acute lym­pho­cytic leu­ke­mia (ALL). 2020. Accessed 2 July
2021. https:​­/​­/seer​­.cancer​­.gov​­/statfacts​­/html​­/alyl​­.html.
4. Larson RA, Dodge RK, Burns CP, et  al. A five-drug remis­sion induc­tion
reg­i­men with inten­sive con­sol­i­da­tion for adults with acute lym­pho­blas­tic
leu­ke­mia: can­cer and leu­ke­mia group B study 8811. Blood. 1995;85(8):2025-
2037.
5. Larson RA, Dodge RK, Linker CA, et  al. A ran­dom­ized con­trolled trial of
filgrastim dur­ing remis­sion induc­tion and con­sol­i­da­tion che­mo­ther­apy
for adults with acute lym­pho­blas­tic leu­ke­mia: CALGB study 9111. Blood.
1998;92(5):1556-1564.
6. Kantarjian H, Thomas D, O’Brien S, et  al. Long-term fol­low-up results of
hyperfractionated cyclo­phos­pha­mide, vin­cris­tine, doxo­ru­bi­cin, and dexa­
meth­as­ one (hyper-CVAD), a dose-inten­sive reg­i­men, in adult acute lym­
pho­cytic leu­ke­mia. Cancer. 2004;101(12):2788-2801.
7. Rowe JM, Buck G, Burnett AK, et al; ECOG; MRC/NCRI Adult Leukemia
Working Party. Induction ther­apy for adults with acute lym­pho­blas­tic leu­
ke­mia: results of more than 1500 patients from the inter­na­tional ALL trial:
MRC UKALL XII/ECOG E2993. Blood. 2005;106(12):3760-3767.
14  |  Hematology 2021  |  ASH Education Program
8. DeAngelo DJ, Stevenson KE, Dahlberg SE, et al. Long-term out­come of a
pedi­at­ric-inspired reg­i­men used for adults aged 18-50 years with newly
diag­nosed acute lym­pho­blas­tic leu­ke­mia. Leukemia. 2015;29(3):526-534.
9. Stock W, Luger SM, Advani AS, et  al. A pedi­at­ric reg­i­men for older ado­
les­cents and young adults with acute lym­pho­blas­tic leu­ke­mia: results of
CALGB 10403. Blood. 2019;133(14):1548-1559.
10. Gökbuget N. Treatment of older patients with acute lym­pho­blas­tic leu­ke­
mia. Hematology Am Soc Hematol Educ Program. 2016;2016(1):573-579.
11. Sive JI, Buck G, Fielding A, et al. Outcomes in older adults with acute lym­
pho­ blas­
tic leu­
kae­mia (ALL): results from the inter­ na­
tional MRC UKALL
XII/ECOG2993 trial. Br J Haematol. 2012;157(4):463-471.
12. O’Brien S, Thomas DA, Ravandi F, Faderl S, Pierce S, Kantarjian H. Results
of the hyperfractionated cyclo­phos­pha­mide, vin­cris­tine, doxo­ru­bi­cin, and
dexa­meth­a­sone reg­i­men in elderly patients with acute lym­pho­cytic leu­ke­
mia. Cancer. 2008;113(8):2097-2101.
13. Fathi AT, DeAngelo DJ, Stevenson KE, et  al. Phase 2 study of inten­si­fied
che­mo­ther­apy and allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion for
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/7/1852080/7luskin.pdf by guest on 13 December 2021
older patients with acute lym­pho­blas­tic leu­ke­mia. Cancer. 2016;122(15):​
2379-2388.
14. Gokbuget N, Beck J, Bruggemann M. Moderate inten­sive che­mo­ther­apy
includ­ing CNS-pro­phy­laxis with lipo­so­mal cytarabine is fea­si­ble and effec­
tive in older patients with Ph-neg­a­tive acute lym­pho­blas­tic leu­ke­mia (ALL):
results of a pro­spec­tive trial from the Ger­man Multicenter Study Group for
Adult ALL (GMALL). Blood. 2012;120(21):1493.
15. DeAngelo DJ, Advani AS, Marks DI, et  al. Inotuzumab ozogamicin for
relapsed/refrac­tory acute lym­pho­blas­tic leu­ke­mia: out­comes by dis­ease
bur­den. Blood Cancer J. 2020;10(8):81.
16. Gökbuget N. How I treat older patients with ALL. Blood. 2013;122(8):1366-
1375.
17. Ribera J-M, García O, Oriol A, et al; PETHEMA Group, Span­ish Society of
Hematology. Feasibility and results of sub­type-ori­ented pro­to­cols in older
adults and fit elderly patients with acute lym­pho­blas­tic leu­ke­mia: results
of three pro­spec­tive par­al­lel tri­als from the PETHEMA group. Leuk Res.
2016;41:12-20.
18. Kim C, Molony JT, Chia VM, Kota VK, Katz AJ, Li S. Patient char­ac­ter­is­tics,
treat­ment pat­terns, and mor­tal­ity in elderly patients newly diag­nosed with
ALL. Leuk Lymphoma. 2019;60(6):1462-1468.
19. Geyer MB, Hsu M, Devlin SM, Tallman MS, Douer D, Park JH. Overall sur­vival
among older US adults with ALL remains low despite mod­est improve­ment
since 1980: SEER anal­y­sis. Blood. 2017;129(13):1878-1881.
20. Kantarjian HM, DeAngelo DJ, Stelljes M, et  al. Inotuzumab ozogamicin
ver­sus stan­dard ther­apy for acute lym­pho­blas­tic leu­ke­mia. N Engl J Med.
2016;375(8):740-753.
21. Kantarjian H, Stein A, Gökbuget N, et al. Blinatumomab ver­sus che­mo­ther­
apy for advanced acute lym­pho­blas­tic leu­ke­mia. N Engl J Med. 2017;376(9):​
836-847.
22. Kantarjian H, Ravandi F, Short NJ, et al. Inotuzumab ozogamicin in com­bi­
na­tion with low-inten­sity che­mo­ther­apy for older patients with Philadel-
phia chro­mo­some-neg­a­tive acute lym­pho­blas­tic leu­kae­mia: a sin­gle-arm,
phase 2 study. Lancet Oncol. 2018;19(2):240-248.
23. Short NJ, Kantarjian HM, Ravandi F, et  al. Reduced-inten­sity che­mo­ther­
apy with mini-hyper-CVD plus inotuzumab ozogamicin, with or with­out
blinatumomab, in older adults with newly diag­nosed Philadelphia chro­mo­
some-neg­a­tive acute lym­pho­blas­tic leu­ke­mia: results from a phase II study.
Blood. 2020;136(suppl 1):1014.
24. Stelljes M, Raffel S, Wasch R, et al. First results of an open label phase II
study to eval­u­ate the effi­cacy and safety of inotuzumab ozogamicin for
induc­tion ther­apy followed by con­ven­tional che­mo­ther­apy based con­sol­
i­da­tion and main­te­nance ther­apy in patients aged 56 years and older with
acute lym­pho­blas­tic leu­ke­mia (INITIAL-1 trial). Blood. 2020;136(suppl 1):267.
25. Jain N, Stevenson K, Winer ES, et al. A mul­ti­cen­ter phase I study com­bin­
ing venetoclax with mini-hyper-CVD in older adults with untreated and
relapsed/refrac­tory acute lym­pho­blas­tic leu­ke­mia. Blood. 2019;134(suppl
1):3867.
26. Advani AS, Moseley A, O’Dwyer KM, et al. Results of SWOG 1318: a phase
2 trial of blinatumomab followed by POMP (pred­ni­sone, vin­cris­tine, meth­
o­trex­ate, 6-mer­cap­to­pu­rine) main­te­nance in elderly patients with newly
diag­nosed Philadelphia chro­mo­some neg­a­tive B-cell acute lym­pho­blas­tic
leu­ke­mia. Blood. 2018;132(suppl 1):33.
27. Chonghaile TN, Roderick JE, Glenfield C, et al. Maturation stage of T-cell
acute lym­pho­blas­tic leu­ke­mia deter­mines BCL-2 ver­sus BCL-XL depen­
dence and sen­si­tiv­ity to ABT-199. Cancer Discov. 2014;4(9):1074-1087.
28. Lacayo NJ, Pullarkat VA, Stock W, et al. Safety and effi­cacy of venetoclax
in com­bi­na­tion with navitoclax in adult and pedi­at­ric relapsed/refrac­tory
 acute lym­pho­blas­tic leu­ke­mia and lym­pho­blas­tic lym­phoma. Blood.
2019;134(suppl 1):285.
29. Gökbuget N, Dombret H, Bonifacio M, et  al. Blinatumomab for min­i­mal
resid­ual dis­ease in adults with B-cell pre­cur­sor acute lym­pho­blas­tic leu­ke­
mia. Blood. 2018;131(14):1522-1531.
30. Kirchhoff H, Karsli U, Schoenherr C, et al. Venetoclax and dexa­meth­a­sone
synergize with inotuzumab ozogamicin-induced DNA dam­age sig­nal­ing in
B-lin­e­age ALL. Blood. 2021;137(19):2657-2661.
31. Chiaretti S, Vitale A, Cazzaniga G, et al. Clinico-bio­log­i­cal fea­tures of 5202
patients with acute lym­pho­blas­tic leu­ke­mia enrolled in the Ital­ian AIEOP
and GIMEMA pro­ to­cols and strat­ i­
fied in age cohorts. Haematologica.
2013;98(11):1702-1710.
32. Fielding AK. Curing Ph+ ALL: assessing the rel­a­tive con­tri­bu­tions of che­
mo­ther­apy, TKIs, and allo­ge­neic stem cell trans­plant. Hematology Am Soc
Hematol Educ Program. 2019;2019(1):24-29.
33. Ravandi F. How I treat Philadelphia chro­mo­some-pos­i­tive acute lym­pho­
blas­tic leu­ke­mia. Blood. 2019;133(2):130-136.
34. Ribera J-M, García O, Fernández-Abellán P, et al; PETHEMA Group. Lack of
neg­a­tive impact of Philadelphia chro­mo­some in older patients with acute
lym­pho­blas­tic leu­kae­mia in the thyrosine kinase inhib­i­tor era: com­par­i­son
of two pro­spec­tive par­al­lel pro­to­cols. Br J Haematol. 2012;159(4):485-488.
35. Rousselot P, Coudé MM, Gokbuget N, et al; Euro­ pean Working Group
on Adult ALL (EWALL) group. Dasatinib and low-inten­sity che­mo­ther­apy
in elderly patients with Philadelphia chro­mo­some-pos­i­tive ALL. Blood.
2016;128(6):774-782.
36. Ottmann OG, Pfeifer H, Cayuela JM. Nilotinib (Tasigna®) and low inten­sity
che­mo­ther­apy for first-line treat­ment of elderly patients with BCR-ABL1-
pos­i­tive acute lym­pho­blas­tic leu­ke­mia: final results of a pro­spec­tive mul­ti­
cen­ter trial (EWALL-PH02). Blood. 2018;132(suppl 1):31.
37. Chalandon Y, Thomas X, Hayette S, et al; Group for Research on Adult
Acute Lymphoblastic Leukemia (GRAALL). Randomized study of reduced-
inten­sity che­mo­ther­apy com­bined with imatinib in adults with Ph-pos­i­tive
acute lym­pho­blas­tic leu­ke­mia. Blood. 2015;125(24):3711-3719.
38. Vignetti M, Fazi P, Cimino G, et al. Imatinib plus ste­roids induces com­plete
remis­sions and prolonged sur­vival in elderly Philadelphia chro­mo­some-
pos­i­tive patients with acute lym­pho­blas­tic leu­ke­mia with­out addi­tional
che­mo­ ther­apy: results of the Gruppo Italiano Malattie Ematologiche
dell’Adulto (GIMEMA) LAL0201-B pro­to­col. Blood. 2007;109(9):3676-3678.
39. Foà R, Vitale A, Vignetti M, et al; GIMEMAAcute Leukemia Working Party. Dasat-
inib as first-line treat­ment for adult patients with Philadelphia chro­mo­some-
pos­i­tive acute lym­pho­blas­tic leu­ke­mia. Blood. 2011;118(25):6521-6528.
Prakash Singh Shekhawat
40. Martinelli G, Piciocchi A, Papayannidis C, et al. First report of the Gimema
LAL1811 phase II pro­spec­tive study of the com­bi­na­tion of ste­roids with
ponatinib as front­ line ther­apy of elderly or unfit patients with Phila-
delphia chro­mo­some-pos­i­tive acute lym­pho­blas­tic leu­ke­mia. Blood.
2017;130(suppl 1):99.
41. Wieduwilt M, Yin J, Wetzler M, et al. A phase II study of dasatinib and dexa­
meth­a­sone as pri­mary ther­apy followed by trans­plan­ta­tion for adults with
newly diag­nosed Ph/BCR-ABL-pos­i­tive acute lym­pho­blas­tic leu­ke­mia
(Ph+ ALL): final results Alliance/CALGB study 10701. Blood. 2018;132(suppl
1):309.
42. Luskin M, Murakami MA, Stevenson KE, et al. A phase I study of asciminib
(ABL001) in com­bi­na­tion with dasatinib and pred­ni­sone for untreated BCR-
ABL1-pos­i­tive ALL in older adults. Blood. 2019;134(suppl 1):3879.
43. Foà R, Bassan R, Vitale A, et al; GIMEMA Investigators. Dasatinib-blinatum-
omab for Ph-pos­i­tive acute lym­pho­blas­tic leu­ke­mia in adults. N Engl J Med.
2020;383(17):1613-1623.
44. Derman BA, Streck M, Wynne J, et  al. Efficacy and tox­ic­ity of reduced
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/7/1852080/7luskin.pdf by guest on 13 December 2021
vs. stan­dard dose pegylated asparaginase in adults with Philadelphia
chro­mo­some-neg­a­tive acute lym­pho­blas­tic leu­ke­mia. Leuk Lymphoma.
2020;61(3):614-622.
45. Martell MP, Atenafu EG, Minden MD, et  al. Treatment of elderly patients
with acute lym­pho­blas­tic leu­kae­mia using a pae­di­at­ric-based pro­to­col. Br
J Haematol. 2013;163(4):458-464.
46. Wolach O, Stevenson KE, Wadleigh M, et al. Allogeneic trans­plan­ta­tion is
not supe­rior to che­mo­ther­apy in most patients over 40 years of age with
Philadelphia-neg­a­tive acute lym­pho­blas­tic leu­ke­mia in first remis­sion. Am J
Hematol. 2016;91(8):793-799.
47. Hurria A, Cirrincione CT, Muss HB, et al. Implementing a geri­at­ric assess­
ment in coop­er­a­tive group clin­i­cal can­cer tri­als: CALGB 360401. J Clin
Oncol. 2011;29(10):1290-1296.
48. Klepin HD, Ritchie E, Major-Elechi B, et  al. Geriatric assess­ment among
older adults receiv­ing inten­sive ther­apy for acute mye­loid leu­ke­mia: report
of CALGB 361006 (Alliance). J Geriatr Oncol. 2020;11(1):107-113.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000226
ALL cur­tain call chemo | 15
 AML: SO MANY OPTIONS, SO LITTLE TIME

What to use to treat AML: the role

of emerging therapies

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/16/1851427/16thol.pdf by guest on 13 December 2021

Felicitas Thol
Department of Hematology, Hemostasis, Oncology, and Stem Cell Transplantation, Hannover Medical School, Hannover, Germany

The development and approval of novel substances have resulted in substantial improvements in the treatment of acute
myeloid leukemia (AML). In the current era of novel treatment options, genetic and molecular testing at the time of diag-
nosis and relapse becomes increasingly relevant. Midostaurin in combination with intensive chemotherapy is the standard
of care as upfront therapy in younger AML patients with mutated fms-related tyrosine kinase 3 (FLT3). Gilteritinib, a second-
generation FLT3 inhibitor, represents a key drug for relapsed/refractory (R/R) FLT3-mutated AML patients. Targeted therapy
has also been developed for patients with mutated isocitrate dehydrogenase 1 (IDH1) and IDH2. The US Food and Drug Admin-
istration (FDA) approved ivosidenib as a monotherapy for newly diagnosed older adult IDH1-mutated patients and enasid-
enib for R/R IDH2-mutated AML patients. CPX-351, a liposomal formulation of daunorubicin and cytarabine, has become an
important upfront treatment strategy for fit patients with therapy-related AML or AML with myelodysplasia-related changes
that are generally challenging to treat. The antibody drug conjugate gemtuzumab ozogamicin was approved in combination
with intensive therapy for patients with newly diagnosed (FDA/European Medicines Agency [EMA]) as well as R/R CD33+
AML. The combination of venetoclax, an oral selective B-cell leukemia/lymphoma-2 inhibitor, with hypomethylating agents
or low-dose AraC (LDAC) has changed the treatment landscape and prognosis for older adult patients very favorably. The
addition of glasdegib, a small-molecule hedgehog inhibitor, to LDAC is another example of novel options in older patients.
Further substances have shown promising results in early clinical trials.

LEARNING OBJECTIVES
• Learn about the indications and efficacy of newly approved drugs in AML
• Become familiar with a treatment algorithm for newly diagnosed and relapsed AML patients

CLINICAL CASE
A 66-year-old man with no significant underlying health con-
ditions developed leukocytosis, anemia, and thrombocyto-
penia. Bone marrow biopsy revealed an infiltration of 60%
myeloblasts. Immunophenotypically, the blasts were CD34+,
CD13+, and CD33+. He was diagnosed with acute myeloid
leukemia (AML). Mutational screening showed an FLT3-ITD
(allelic ratio 0.6), isocitrate dehydrogenase 2 (IDH2), and
TP53 mutation. The cytogenetic analysis revealed a normal
karyotype.
What are the treatment options for this patient in the
era of novel therapies?
Intensive therapy with novel agents
In recent years, treatment strategies for AML have evolved
beyond “7+3” for younger patients and past monotherapy
with hypomethylating agents (HMAs) for older adult AML
patients (Figure 1). Novel mechanisms of action underlie
these recently approved drugs. Some of the new treat-
ment strategies were designed for or have shown the most
benefit in a distinct genetic group of patients. Therefore,
molecular and cytogenetic analysis becomes increasingly
relevant for the application of novel drugs. In addition, ge-
netic risk stratification at the time of diagnosis is an essen-
tial element for guiding the decision regarding whether
allogeneic hematopoietic stem cell transplantation (HSCT)
is recommended in the first complete remission (CR).1 In
2017 the European LeukemiaNet (ELN) recommended muta-
tional screening for FLT3-ITD (including allelic ratio), NPM1,
CEBPA (biallelic status), TP53, ASXL1, and RUNX1 in addi-
tion to cytogenetic analysis for the allocation to one of the
three prognostic categories.1 Based on the risk-benefit ratio
for allogeneic HSCT, patients in the adverse and possibly
in the intermediate group are candidates for allogeneic

16 | Hematology 2021 | ASH Education Program


Figure 1. Frontline treatment algorithm in AML patients considering recently approved substances. mut, mutated.
HSCT in first CR. In our patient three impor­tant muta­tions were
iden­ti­fied (FLT3-ITD, IDH2, TP53). The detec­tion of FLT3-ITD in
this patient allows the appli­ ca­tion of front­ line-targeted ther­
apy with midostaurin. Midostaurin is an oral first-gen­ er­a­
tion
tyro­sine kinase inhib­i­tor (TKI) and is now the stan­dard of care
in FLT3-mutated AML patients who can receive inten­sive ther­
apy. Its approval by the US Food and Drug Administration (FDA)
and the Euro­pean Medicines Agency (EMA) was based on the
RATIFY trial, a large inter­na­tional phase 3 trial that ran­dom­ized
FLT3-mutated patients to treat­ment with midostaurin vs pla­cebo
in com­bi­na­tion with inten­sive ther­apy.2 Both over­all sur­vival (OS)
and event-free sur­vival (EFS) were sig­nif­i­cantly lon­ger in the mi-
dostaurin com­pared to the pla­cebo arm with­out dif­fer­ences in
severe adverse events. The 4-year OS was 51.4% in the midostau-
rin arm vs 44.3% in the pla­cebo arm. Of note, the sur­vival ben­e­fit
in the midostaurin arm was observed when looking at the whole
cohort. In a sep­a­rate anal­y­sis of molec­u­lar sub­groups (FLT3-ITD
high or low, FLT3-TKD), the sur­vival advan­tage in the midostau-
rin arm did not reach sta­tis­ti­cal sig­nif­i­cance, which is likely due
to the sub­groups being under­pow­ered.2 The sur­vival advan­tage
observed in the midostaurin arm for the whole cohort was due
to a sig­nif­i­cantly lower cumu­la­tive inci­dence of relapse (CIR) if
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/16/1851427/16thol.pdf by guest on 13 December 2021
trans­plan­ta­tion was not taken into account. After the com­ple­
tion of con­sol­i­da­tion ther­apy, midostaurin or pla­cebo was given
for up to 12 cycles as main­te­nance ther­apy. A land­mark anal­y­sis
at the end of con­sol­i­da­tion ther­apy with high-dose cytarabine
showed no sig­nif­i­cant dif­fer­ences in OS and EFS between both
treat­ment arms. Due to these incon­clu­sive results with respect
to the quan­ti­ta­tive effect of main­te­nance ther­apy, the effi­cacy
of the con­tin­ued treat­ment after con­sol­i­da­tion ther­apy is cur­
rently unclear.3 The favor­able safety pro­file of midostaurin was
con­firmed in the RADIUS-X expanded access pro­gram.4
Another novel treat­ment option for newly diag­nosed AML
patients with CD33 expres­ sion on leu­ke­mic blasts is gemtu-
zumab ozogamicin (GO) in com­bi­na­tion with 7 + 3.5 GO com­bines
a CD33-directed anti­body with the che­mo­ther­apy agent cali-
cheamicin. CD33 is a suit­able tar­get in AML because it is highly
expressed in most AML cells and much less on nor­mal hema­to­
poi­etic cells. While prior data have been conflicting, espe­cially
with regard to tox­ic­ity, the French ALFA-0701 trial was designed
to look directly at the effect of GO on the sur­ vival of older
adults under­go­ing inten­sive ther­apy.6 In this ran­dom­ized, open-
label phase 3 study, 280 older patients (aged 50-70 years) were
either treated with GO (n = 140) on day 1, 4, and 7 or with­out
AML ther­apy with newly approved sub­stances  | 17
GO (n = 140) dur­ing the 7 + 3 induc­tion ther­apy.6 Patients achiev­
ing CR were treated with 2 cycles of con­sol­i­da­tion ther­apy with
or with­out GO according to their ini­tial ran­dom­i­za­tion. The EFS
at 2 years was 17.1% in the con­trol arm vs 40.8% in the GO arm.
Two-year OS was also sig­nif­i­cantly bet­ter in the GO (53.2%) vs
stan­dard arm (41.9%; P = 0.037).6 A meta-anal­y­sis by Hills et  al.
looked at the effi­cacy of GO in 5 ran­dom­ized tri­als involv­ing
3325 patients.7 Interestingly, the abso­lute sur­vival ben­e­fit was
most appar­ent in the patient group with favor­able cyto­ge­net­
ics (20.7%), followed by patients with inter­me­di­ate cyto­ge­net­ics
(5.7%).7 GO had no ben­e­fit in patients with adverse cyto­ge­net­
ics.7 In the AMLSG 09-09 trial, NPM1-mutated AML patients were
ran­dom­ized to receive inten­sive che­mo­ther­apy with or with­out
GO. In this trial the EFS was not sig­nif­i­cantly dif­fer­ent between
treat­ment arms.8 However, the early death rate dur­ing induc­tions
was higher, while the CIR was lower in the GO arm. This is in line
with the find­ing that the addi­tion of GO leads to a bet­ter reduc­
tion in NPM1 mutant tran­script lev­els across all­treat­ment cycles.9
Subgroup anal­y­sis revealed that the addi­tion of GO showed a
sig­nif­i­cant ben­e­fit for youn­ger, FLT3-ITD-neg­a­tive, and female
patients.8 The FDA and the EMA approved GO in com­bi­na­tion
with inten­sive che­mo­ther­apy for patients with newly diag­nosed
CD33+ AML. Considering the data outlined above, GO might be
best suited for patients with favor­able cyto­ge­net­ics and with­out
an increased risk for early mor­tal­ity due to under­ly­ing infec­tion.
For patients with newly diag­nosed sec­ond­ary AML (s-AML),
ther­apy-related AML (tAML), or AML with myelodysplasia-related
changes (AML-MRC), treat­ment with CPX-351, a lipo­so­mal for­mu­
la­tion of dau­no­ru­bi­cin and cytarabine in a fixed com­bi­na­tion, is
now avail­­able.10 In the ini­tial phase 2 trial com­par­ing CPX-351 with
stan­dard inten­sive treat­ment in older AML patients, only the sub­
group of patients with sAML and adverse cyto­ge­net­ics showed
a sig­nif­i­cantly higher response rate (57.6% in the CPX-351 arm
vs 31.6% in the stan­dard arm).11 Based on these results, a large
phase 3 trial was ini­ti­ated in older (aged 60-75 years) patients
with high-risk AML or sAML.10 Here the median OS was 9.56 in the
CPX-351 arm and 5.95 months in the 7 + 3 arm. The over­all remis­
sion rate was also sig­nif­i­cantly bet­ter with CPX-351 (47.7%) vs the
stan­dard arm (33.3%).10 Of note, 34% of patients in the CPX-351
arm under­went allo­ge­neic HSCT, com­pared to 25% in the 7 + 3
arm. In an explor­atory land­mark sur­vival anal­y­sis from the time
of trans­ plant, out­ comes were more favor­ able in the CPX-351
arm. This is in line with results from an Ital­ian com­pas­sion­ate-use
pro­gram for CPX-351. Here, CIR was reduced in 71 older adult
AML patients when allo­ge­neic HSCT was performed in the first
CR, with a very favor­able over­all out­come fol­low­ing trans­plan­
ta­tion.12 The encour­ag­ing results of the phase 3 trial resulted in
FDA and EMA approval of CPX-351 for this dis­tinct sub­group of
AML patients. In order to iden­tify these patients with AML with
MRC, cyto­ge­netic anal­y­sis is indis­pens­able.13 Real-life expe­ri­ence
from a French ret­ro­spec­tive study looked at 103 sAML, tAML, or
AML-MRC patients treated with CPX-351.14 The over­all response
rate (ORR) was 59% after induc­tion with an OS of 16.1 months
at a median fol­low-up time of 8.6 months. Importantly, patients
with ASXL1 and RUNX1 muta­tions, who are in the ELN unfa­vor­able
prog­nos­tic group, showed sim­i­lar response rates as wild-type
patients in con­trast to patients with mutated TP53.14 The safety
pro­file was very favor­able for CPX-351, with a low rate of alo­pe­cia
(11%) and gas­tro­in­tes­ti­nal tox­ic­ity (50%). Interestingly, a ret­ro­
spec­tive study involv­ing 25 AML patients who received CPX-351
18  |  Hematology 2021  |  ASH Education Program
as out­pa­tients sug­gests that the admin­is­tra­tion of CPX-351 with­
out planned admis­sion is safe and might lead to a decreased
uti­li­za­tion of health care resources.15
Now we return to our patient. Based on the data above,
he was treated with 7 + 3 and midostaurin. He achieved a CR
after induc­tion ther­apy. Importantly, as he was allo­cated to the
adverse prog­nos­tic group (FLT3-ITD high with­out NPM1 muta­
tion as well as the pres­ence of TP53 muta­tion), it was recom-
mended that he undergo allo­ge­neic HSCT in first CR. However,
for those patients in CR who can­not com­plete inten­sive ther­apy
after induc­tion ther­apy, main­te­nance ther­apy with CC-486 has
become another novel option. CC-486 is an oral HMA that is not
bioequivalent to inject­able azacitidine and has shown effi­cacy in
patients who have devel­oped resis­tance to inject­able HMAs in
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/16/1851427/16thol.pdf by guest on 13 December 2021
prior stud­ies.16,17 In the phase 3 QUAZAR AML-001 trial, 472 AML
patients (aged 55 to 86 years) were ran­dom­ized to receive CC-
486 (238 patients) or pla­cebo (234 patients).18 Median OS as well
as EFS was sig­nif­i­cantly lon­ger in the CC-486 arm (24.7 months,
10.2 months, respec­tively) com­pared to the pla­cebo arm (14.8
months, 4.8 months). Hematological tox­ic­ity with neutropenia
was more com­mon in the CC-486 arm (41% in the CC-486 arm vs
24% in the pla­cebo arm). Based on the QUAZAR AML-001 results,
the FDA and EMA granted approval for CC-486 as a main­te­nance
ther­apy for adult AML patients in CR/CR with incom­plete hema­
to­logic recov­ery (CRi) after inten­sive induc­tion ther­apy who are
unable to com­plete inten­sive cura­tive ther­apy (eg, allo­ge­neic
HSCT). CC-486 should not be substituted for inject­able azacit-
idine because there are dif­fer­ences in phar­ma­co­ki­netic prop­er­
ties between the oral and the inject­able ver­sion.
We still lack suf­fi­cient data about the effi­cacy of com­bin­ing
the above novel agents with each other.
Nonintensive ther­apy with novel agents
Treatment has remained chal­ leng­ ing for older adult patients
who can­not receive inten­sive ther­apy.19 Hypomethylating agents
such as azacitidine and decitabine have become impor­tant treat­
ment strat­eg ­ ies for older patients and have shown sur­vival ben­
e­fits when com­pared to low-dose AraC (LDAC).20 However, the
median OS for patients treated with azacitidine in a phase 3 trial
was 10.4 months and only 7.1 months in a large pop­u­la­tion-based
study.21 In light of the fact that HMA monotherapy does not result
in long-term remis­sions, clin­i­cal tri­als com­bin­ing HMAs (and LDAC)
with venetoclax have dem­ on­strated impres­ sive sur­
vival data.
Venetoclax is an oral, highly selec­tive small-mol­e­cule B-cell leu­
ke­mia/lym­phoma-2 inhib­it­or with poor effi­cacy as a sin­gle agent
in AML.22 However, in vitro and in vivo stud­ies have shown syn­er­
gism between che­mo­ther­apy and venetoclax, supporting the use
of venetoclax com­bi­na­tions in AML. One mech­a­nism of syn­er­gism
between azacitidine and venetoclax is through tran­ scrip­tional
induc­tion of the proapoptotic BH3-only pro­tein NOXA by azaciti-
dine.23,24 The VIALE-C study, an inter­na­tional phase 3 ran­dom­ized
dou­ble-blind trial, com­pared venetoclax vs pla­cebo in com­bi­na­
tion with LDAC in older (≥75 years) or youn­ger patients unfit for
inten­sive ther­apy.25 At a median fol­low-up of 17.5 months, patients
treated with venetoclax and LDAC (VEN-LDAC) achieved a sig­nif­i­
cantly lon­ger median OS com­pared to patients receiv­ing pla­cebo
and LDAC (PBO-LDAC) (8.4 vs 4.1 months). Similarly, the CR rates
as well as EFS were bet­ter in the VEN-LDAC arm com­pared to the
PBO-LDAC arm (EFS, 4.9 months vs 2.1 months). The com­bi­na­
tion of azacitidine and venetoclax (VEN-AZA) yielded even more
encour­ag­ing results (VIALE-A trial).26 Here, older, pre­vi­ously un-
treated AML patients who were inel­i­gi­ble for stan­dard induc­tion
received VEN-AZA or pla­cebo and azacitidine (PBO-AZA).26 The
median OS was 14.7 months in the VEN-AZA arm vs 9.6 months in
the PBO-AZA arm at a median fol­low-up of 20.5 months. Thus, the
median OS in the VEN-AZA arm was lon­ger than reported in any
other prior trial for front­line, older adult AML patients. In addi­tion,
the CR rate, at 36.7% vs 17.9%, was also more favor­able with the
venetoclax com­bi­na­tion. Grade ≥3 throm­bo­cy­to­pe­nia and neu-
tropenia were more com­mon in patients treated with venetoclax.
Likewise, febrile neutropenia occurred in 42% of patients in the
VEN-AZA arm vs 19% of patients in the PBO-AZA arm. Thus, it is
essen­tial to observe the patient closely for hema­to­log­i­cal tox­ic­ity
in the form of cytopenias and fol­low rec­om­men­da­tions for dose
adjust­ments when treating patients with venetoclax com­bi­na­
tions.27 Dose adjust­ments also become nec­es­sary when comed-
ications have strong CYP3A4 inhib­i­tory activ­ity (eg, anti­fun­gals,
fluoroquinolones).27 Recent data sug­gest that IDH1- and IDH2-mu-
tated AML patients achieve impres­sive response rates, suggest-
Figure 2. Treatment of relapsed/refractory AML patients considering recently novel substances. dDLI, donor lymphocyte infusion;
BSC, best supportive care; HU, hydroxyurea; mut, mutated.
Prakash Singh Shekhawat
ing that this molec­u­lar sub­group, espe­cially, ben­e­fits from adding
venetoclax to azacitidine treat­ment.28 The FDA has approved vene-
toclax in com­bi­na­tion with HMAs or LDAC for newly diag­nosed
AML patients ≥75 years of age or patients with comorbidities pre­
clud­ing inten­sive induc­tion ther­apy (the EMA has approved vene-
toclax only in combination with HMAs). As HMA/VEN has evolved
as a stan­dard of care (in countries with approval), this com­bi­na­
tion also rep­re­sents the back­bone of tri­als with novel sub­stances.
Unfortunately, a ret­ ro­
spec­ tive anal­
y­sis sug­gests that once pa-
tients become unre­spon­sive to HMA/VEN the prog­no­sis is very
poor.29 This might be related to the acqui­si­tion of high-risk cyto­
ge­netic and molec­u­lar fea­tures (eg, muta­tions in TP53, N/KRAS,
and/or KIT).29 For patients who are older but lack comorbidities,
the ques­tion arises as to whether inten­sive ther­apy or HMA/VEN
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/16/1851427/16thol.pdf by guest on 13 December 2021
treat­ments are the wiser treat­ment choice. To date, there is no
ran­dom­ized trial com­par­ing inten­sive che­mo­ther­apy with HMA-
VEN head-to-head. However, a ret­ ro­spec­
tive study com­ pared
out­comes of patients with inten­sive che­mo­ther­apy vs decit-
abine over 10 days in com­bi­na­tion with venetoclax (DEC10-VEN).
AML ther­apy with newly approved sub­stances  | 19
Table 1. Recently approved drugs for AML

Significant find­ings in clin­i­cal Issues need­ing spe­cial

Drug Indication Route of admin­is­tra­tion
tri­als atten­tion
Midostaurin Newly diag­nosed FLT3 Oral Phase 3 trial (RATIFY): Careful admin­is­tra­tion of
Rydapt® mut AML in com­bi­ inten­sive che­mo­ther­apy + comedications with strong
na­tion with inten­sive midostaurin vs pla­cebo: CYP3A4 inhib­i­tory activ­ity
ther­apy (FDA, EMA) 4-year OS: 51.4% midostaurin
vs 44.3% pla­cebo
CPX-351 Newly diag­nosed tAML, IV Phase 3 trial: Side effects sim­il­ar to stan­
Vyxeos® AML-MRC (FDA, EMA) median OS 9.6 (CPX-351) vs 6.0 dard che­mo­ther­apy but less
months (stan­dard che­mo­ alo­pe­cia
ther­apy)
Gemtuzumab Newly diag­nosed CD33+ IV Phase 3 trial (ALFA-0701): Hepatotoxicity (includ­ing
ozogamicin AML (FDA, EMA) in inten­sive che­mo­ther­apy +/− veno-occlu­sive dis­ease),

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/16/1851427/16thol.pdf by guest on 13 December 2021

Mylotarg® com­bi­na­tion with GO: infu­sion-related reac­tions,
inten­sive ther­apy, R/R 2-year OS: 53.2% GO vs 41.9% infec­tion
CD33+ AML in com­bi­ stan­dard
na­tion with inten­sive
ther­apy (FDA)
Venetoclax Newly diag­nosed AML Oral VIALE-A trial: Hematological tox­ic­ity (fol­
Venclexta® in patients ≥75 years VEN-AZA vs pla­cebo: low recommended dose
or with comorbidities median OS 14.7 (venetoclax) adjust­ments), neutropenic
pre­clud­ing inten­sive vs 9.6 months (pla­cebo) fever, dose adjust­ments if
ther­apy; in com­bi­ VIALE-C trial: comedications with strong
na­tion with LDAC or LDAC-VEN vs pla­cebo: CYP3A4 inhib­i­tory activ­ity,
HMA (FDA) median OS 8.4 (venetoclax) tumor lysis syn­drome (but
vs 4.1 months (pla­cebo) much rarer than in CLL)
Glasdegib Newly diag­nosed AML Oral Phase 2 BRIGHT AML 1003 trial: Musculoskeletal pain
Daurismo® in patients ≥75 years LDAC +/− glasdegib:
or with comorbidities median OS was 8.8 (LDAC-
pre­clud­ing inten­sive glasdegib) vs 4.9 months
ther­apy; in com­bi­na­ (LDAC alone)
tion with LDAC (FDA,
EMA)
CC-486 Maintenance ther­apy Oral Phase 3 QUAZAR AML-001: GI tox­ic­ity, neutropenia
Onureg® AML patients in CC-486 vs pla­cebo:
CR/CRi after inten­sive median OS 24.7 (CC-486) vs
induc­tion who are 14.8 months (pla­cebo)
unable to com­plete
inten­sive cura­tive
ther­apy
Enasidenib R/R IDH2 mut AML as Oral Phase 1/2 trial: Check IDH2 muta­tional sta­tus
Idhifa® monotherapy (FDA) ORin 40.3% of patients, OS at time of R/R dis­ease,
9.3 months IDH dif­fer­en­ti­a­tion syn­drome,
indi­rect hyperbilirubinemia
Ivosidenib Newly diag­nosed or Oral Phase 1b: Check IDH1 muta­tional sta­tus
Tibsovo® R/R IDH1 mut AML as monotherapy CR in 22% and at time of diag­no­sis and
monotherapy(FDA) OR in 42% of patients R/R dis­ease,
IDH dif­fer­en­ti­a­tion syn­drome,
indi­rect hyperbilirubinemia
Gilteritinib R/R FLT3 mut AML as Oral Phase 3 (ADMIRAL) trial: Check FLT3 muta­tional sta­tus
Xospata® monotherapy (FDA, gilteritinib vs sal­vage che­mo­ at time of R/R dis­ease
EMA) ther­apy:
CR/CRi 34% (gilteritinib) vs
15.4% (sal­vage ther­apy);
OS 9.3 (gilteritinib) vs 5.6
months (sal­vage ther­apy)
CLL, chronic lym­pho­cytic leu­ke­mia; CRi, CR with incom­plete hema­to­logic recov­ery; IV, intra­ve­nous; mut, mutated; OR, over­all response.

20  |  Hematology 2021  |  ASH Education Program

The CR/CRi rates and relapse rates were more favor­able in the
DEC10-VEN arm. In addi­tion, OS was sig­nif­i­cantly lon­ger in DEC10-
VEN-treated patients.30 In sum­mary, the intro­duc­tion of venetoclax
has mark­edly changed the sur­vival out­look for older AML patients.
Glasdegib, a small-mol­e­cule hedge­hog inhib­i­tor, is another
drug that has received approval by the FDA and the EMA for
older, newly diag­nosed AML patients (≥75 years) or patients with
comorbidities that pre­clude inten­sive ther­apy. The approval was
based on a phase 2, ran­dom­ized, open-label, mul­ti­cen­ter study
that com­pared LDAC vs LDAC/glasdegib. Median OS was 8.8
months with the com­bi­na­tion com­pared to 4.9 months with LDAC
alone.31 Of note, a head-to-head com­par­i­son between VEN/LDAC
or VEN/HMA and LDAC/glasdegib is cur­rently miss­ing.
For one molec­u­lar sub­group of patients, targeted ther­apy is
now avail­­able: older AML patients (≥75 years) or AML patients
with sig­ nif­i­
cant comorbidities who carry a muta­ tion in IDH1.
Ivosidenib is an oral, targeted agent that inhib­its mutant IDH1
and con­se­quently blocks the pro­duc­tion of the oncometabolite
2-hydroxyglutarate. Ivosidenib has been stud­ied as monother-
apy in IDH1-mutated AML patients in a phase 1b trial.32 Of the 258
patients in the trial, the major­ity had R/R AML (179 patients). In
the pri­mary effi­cacy cohort, 21.6% of patients achieved CR and
41.6% an over­all response, with a median dura­tion of response
of 9.3 and 6.5 months, respec­tively. The IDH dif­fer­en­ti­a­tion syn­
drome is a dis­tinct side effect of IDH inhib­i­tors and requires a
phy­si­cian’s aware­ness. It was observed in 3.9% of patients in this
trial. The FDA (but not the EMA) approved ivosidenib for older
(≥75 years) patients or patients with sig­nif­i­cant comorbidities
as a monotherapy for newly diag­ nosed, IDH1-mutant as well
as R/R AML. In a phase 1b trial, the com­bi­na­tion of azacitidine
with ivosidenib was well tol­er­ated, and patients showed dura­ble
remis­sions.33 Thus, we eagerly await the results of the pla­cebo-
con­trolled phase 3 trial with this com­bi­na­tion.
Treatment of AML in relapse
Due to clonal evo­lu­tion, it is advis­able to repeat muta­tional pro­
fil­ing at the time of relapse.34 While some muta­tions are sta­ble
dur­ing dis­ease pro­gres­sion, oth­ers are lost or gained at the time
of relapse.35 Interestingly, the fol­low-up data of the RATIFY trial
showed that 46% of patients became FLT3-ITD-neg­a­tive at the
time of dis­ease resis­tance or pro­gres­sion.36 For patients who show
an FLT3 muta­tion in relapse, gilteritinib, an oral, sec­ond-gen­er­a­
tion TKI, is now avail­­able. Gilteritinib was com­pared to sal­vage
ther­apy in a large ran­dom­ized phase 3 trial (ADMIRAL).37,38 Here,
patients were ran­dom­ized to receive gilteritinib vs sal­vage che­
mo­ther­apy according to local inves­ti­ga­tors’ choice. Gilteritinib
was asso­ci­ated with higher CR/CRi rates (34% vs 15.4%). This pos­
i­tive effect trans­lated into a lon­ger OS in the gilteritinib arm (9.3
vs 5.6 months).38 Importantly, severe adverse events were less fre­
quent in the gilteritinib arm. While only 5.7% of patients in the
ADMIRAL trial received midostaurin dur­ing front­line treat­ment, a
ret­ro­spec­tive study of 13 med­i­cal cen­ters ana­lyzed the effi­cacy
of gilteritinib in R/R AML patients pre­vi­ously treated with a TKI.
In this patient pop­u­la­tion, the CR rates were 58% with an OS of
7.8 months, suggesting that gilteritinib is no less effec­tive after
pre­vi­ous TKI treat­ment. This is rel­e­vant, as in the RATIFY trial 11%
of patients devel­oped ITD clones resis­tant to midostaurin dur­ing
dis­ease pro­gres­sion.36
Prakash Singh Shekhawat
For IDH2-mutated AML patients, the FDA (not the EMA)
approved enasidenib, an oral inhib­i­tor of mutated IDH2, as a
monotherapy for R/R AML patients.39,40 The approval was based
on a phase 1/2 trial.29 Here, enasidenib monotherapy achieved
an ORR of 40.3%, with a median dura­tion of response of 5.8
months and a median OS of 9.3 months in R/R IDH2-mutated
AML patients.40
GO also received approval by the FDA (not the EMA) for R/R
patients with CD33+ AML and can be added to inten­sive ther­apy.
So if our patient relapses after HSCT with the same molec­u­lar
pro­file (FLT3-ITD pos, IDH2 mut), targeted ther­apy is avail­­able
with gilteritinib or enasidinib (Figure 2).
Further direc­tions
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/16/1851427/16thol.pdf by guest on 13 December 2021
Various agents cur­rently undergo eval­u­a­tion in early clin­i­cal AML
tri­
als. Magrolimab is a mono­ clo­ nal anti-CD47 anti­body. CD47
is known to be a mac­ro­phage immune check­point that func­
tions as a “don’t eat me” sig­nal to can­cer so that magrolimab
can pro­mote phago­cy­to­sis of leu­ke­mic cells. Magrolimab was
com­bined with azacitidine in a phase 1b trial enroll­ing 52 AML
patients.41 In this trial, 65% of patients achieved an objec­tive
response, with 44% of patients achiev­ing a CR. Magrolimab re-
ceived orphan drug des­ig­na­tion by the FDA for myelodysplastic
syndrome (MDS) and AML and by the EMA for AML in 2020. APR-
246 is a prom­is­ing agent devel­oped for patients with mutated
TP53, as it restores its func­tion as a tumor sup­pres­sor gene.42
Early tri­als have com­bined APR-246 with azacitidine in MDS and
AML. In a phase 1/2 trial, the ORR was 64% with a 36% CR rate
in TP53-mutated AML patients.43 These encour­ag­ing results have
been con­firmed in a sec­ond study show­ing a CR rate of 56%
at 6 cycles. APR-246 has received orphan drug and fast-track
des­ig­na­tions from the FDA for MDS and orphan drug des­ig­na­
tion from the EMA for AML and MDS. Menin is a prom­is­ing tar­
get for AML patients with mixed-lineage leukemia trans­lo­ca­tions,
and menin inhib­i­tors are cur­rently being stud­ied in clin­i­cal tri­
als.44 Many more sub­stances are in early devel­op­ment for AML
patients. Importantly, clin­i­cal tri­als study­ing novel agents with
new com­bi­na­tions are also ongo­ing (eg, IDH inhib­i­tors with che­
mo­ther­apy, FLT3 inhib­i­tors with HMAs). Finally, yet impor­tantly,
advance­ment in mea­sur­able resid­ual dis­ease mon­i­tor­ing is also
likely to influ­ence our treat­ment deci­sions and have a pos­i­tive
impact on out­come.45 In sum­mary, the treat­ment land­scape has
evolved remark­ably over the last 5 years with the approval of a
num­ber of new drugs (Table 1). This has direct impli­ca­tions for
our 66-year-old patient.
Conflict-of-inter­est dis­clo­sure
Felicitas Thol: Advisory Board: Celgene, Novartis, Jazz, Abbvie,
Astellas, Pfizer.
Off-label drug use
Felicitas Thol: nothing to disclose.
Correspondence
Felicitas Thol, Department of Hematology, Hemostasis, Oncol-
ogy, and Stem Cell Transplantation, Hannover Medical School,
Carl-Neuberg Str 1, 30625 Hannover, Germany; e-mail: thol​
­.felicitas@mh-hannover​­.de.
AML ther­apy with newly approved sub­stances  | 21
References
1. Döhner H, Estey E, Grimwade D, et al. Diagnosis and man­age­ment of AML
in adults: 2017 ELN rec­om­men­da­tions from an inter­na­tional expert panel.
Blood. 2017;129(4):424-447.
2. Stone RM, Mandrekar SJ, Sanford BL, et  al. Midostaurin plus che­mo­ther­
apy for acute mye­ loid leu­ ke­mia with a FLT3 muta­ tion. N Engl J Med.
2017;377(5):454-464.
3. Larson RA, Mandrekar SJ, Huebner LJ, et al. Midostaurin reduces relapse in
FLT3-mutant acute mye­loid leu­ke­mia: the Alliance CALGB 10603/RATIFY trial
[published online 2 March 2021]. Leukemia.
4. Roboz GJ, Strickland SA, Litzow MR, et al. Updated safety of midostaurin
plus che­mo­ther­apy in newly diag­nosed FLT3 muta­tion-pos­i­tive acute mye­
loid leu­ke­mia: the RADIUS-X expanded access pro­gram. Leuk Lymphoma.
2020;61(13):3146-3153.
5. Thol F, Schlenk RF. Gemtuzumab ozogamicin in acute mye­loid leu­ke­mia
revisited. Expert Opin Biol Ther. 2014;14(8):1185-1195.
6. Castaigne S, Pautas C, Terré C, et al; Acute Leukemia French Association.
Effect of gemtuzumab ozogamicin on sur­vival of adult patients with de-
novo acute mye­ loid leu­kae­ mia (ALFA-0701): a randomised, open-label,
phase 3 study. Lancet. 2012;379(9825):1508-1516.
7. Hills RK, Castaigne S, Appelbaum FR, et al. Addition of gemtuzumab ozo-
gamicin to induc­tion che­mo­ther­apy in adult patients with acute mye­loid
leu­kae­mia: a meta-anal­y­sis of indi­vid­ual patient data from randomised con­
trolled tri­als. Lancet Oncol. 2014;15(9):986-996.
8. Schlenk RF, Paschka P, Krzykalla J, et  al. Gemtuzumab ozogamicin in
NPM1-mutated acute mye­loid leu­ke­mia: early results from the pro­spec­tive
ran­dom­ized AMLSG 09-09 phase III study. J Clin Oncol. 2020;38(6):623-
632.
9. Kapp-Schwoerer S, Weber D, Corbacioglu A, et al. Impact of gemtuzumab
ozogamicin on MRD and relapse risk in patients with NPM1-mutated AML:
results from the AMLSG 09-09 trial. Blood. 2020;136(26):3041-3050.
10. Lancet JE, Uy GL, Cortes JE, et al. CPX-351 (cytarabine and dau­no­ru­bi­cin)
lipo­some for injec­tion ver­sus con­ven­tional cytarabine plus dau­no­ru­bi­cin in
older patients with newly diag­nosed sec­ond­ary acute mye­loid leu­ke­mia. J
Clin Oncol. 2018;36(26):2684-2692.
11. Lancet JE, Cortes JE, Hogge DE, et al. Phase 2 trial of CPX-351, a fixed 5:1
molar ratio of cytarabine/dau­no­ru­bi­cin, vs cytarabine/dau­no­ru­bi­cin in
older adults with untreated AML. Blood. 2014;123(21):3239-3246.
12. Guolo F, Fianchi L, Minetto P, et al. CPX-351 treat­ment in sec­ond­ary acute
mye­lo­blas­tic leu­ke­mia is effec­tive and improves the fea­si­bil­ity of allo­ge­neic
stem cell trans­plan­ta­tion: results of the Ital­ian com­pas­sion­ate use pro­gram.
Blood Cancer J. 2020;10(10):96.
13. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revi­sion to the World Health
Organization clas­si­fi­ca­tion of mye­loid neo­plasms and acute leu­ke­mia.
Blood. 2016;127(20):2391-2405.
14. Chiche E, Rahmé R, Bertoli S, et al. Real-life expe­ri­ence with CPX-351 and
impact on the out­come of high-risk AML patients: a multicentric French
cohort. Blood Adv. 2021;5(1):176-184.
15. Keiffer G, Hughes K, Zhan T, Wilde L, Palmisiano N, Kasner M. Outpatient
Vyxeos induc­ tion with­ out planned admis­ sion for select patients with
sec­ond­ary acute mye­loid leu­ke­mia (sAML) is safe and yields healthcare
resource sav­ings. Paper presented at: ASH Annual Meeting; (Virtual); 5-8
December 2020.
16. Garcia-Manero G, Gore SD, Kambhampati S, et  al. Efficacy and safety of
extended dos­ing sched­ules of CC-486 (oral azacitidine) in patients with
lower-risk myelodysplastic syn­dromes. Leukemia. 2016;30(4):889-896.
17. Garcia-Manero G, Almeida A, Giagounidis A, et al. Design and ratio­nale of
the QUAZAR Lower-Risk MDS (AZA-MDS-003) trial: a ran­dom­ized phase
3 study of CC-486 (oral azacitidine) plus best sup­port­ive care vs pla­cebo
plus best sup­port­ive care in patients with IPSS lower-risk myelodysplas-
tic syn­dromes and poor prog­no­sis due to red blood cell trans­fu­sion-
depen­dent ane­mia and throm­bo­cy­to­pe­nia. BMC Hematol. 2016;16:12.
18. Wei AH, Döhner H, Pocock C, et al; QUAZAR AML-001 Trial Investigators.
Oral azacitidine main­te­nance ther­apy for acute mye­loid leu­ke­mia in first
remis­sion. N Engl J Med. 2020;383(26):2526-2537.
19. Arellano M, Carlisle JW. How I treat older patients with acute mye­loid leu­
ke­mia. Cancer. 2018;124(12):2472-2483.
20. Dombret H, Seymour JF, Butrym A, et  al. International phase 3 study of
azacitidine vs con­ven­tional care reg­i­mens in older patients with newly
diag­nosed AML with >30% blasts. Blood. 2015;126(3):291-299.
22  |  Hematology 2021  |  ASH Education Program
21. Zeidan AM, Wang R, Wang X, et al. Clinical out­comes of older patients with
AML receiv­ing hypomethylating agents: a large pop­u­la­tion-based study in
the United States. Blood Adv. 2020;4(10):2192-2201.
22. Konopleva M, Pollyea DA, Potluri J, et  al. Efficacy and bio­log­i­cal cor­
re­lates of response in a phase II study of venetoclax monotherapy in
patients with acute mye­log­e­nous leu­ke­mia. Cancer Discov. 2016;6(10):
1106-1117.
23. Jin S, Cojocari D, Purkal JJ, et al. 5-azacitidine induces NOXA to prime AML
cells for venetoclax-medi­ated apo­pto­sis. Clin Cancer Res. 2020;26(13):3371-
3383.
24. Cojocari D, Smith BN, Purkal JJ, et al. Pevonedistat and azacitidine upreg-
ulate NOXA (PMAIP1) to increase sen­si­tiv­ity to venetoclax in pre­clin­i­cal
mod­els of acute mye­loid leu­ke­mia [published online 15 April 2021]. Hae-
matologica.
25. Wei AH, Strickland SA Jr, Hou JZ, et al. Venetoclax com­bined with low-dose
cytarabine for pre­vi­ously untreated patients with acute mye­loid leu­ke­mia:
results from a phase Ib/II study. J Clin Oncol. 2019;37(15):1277-1284.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/16/1851427/16thol.pdf by guest on 13 December 2021
26. DiNardo CD, Jonas BA, Pullarkat V, et al. Azacitidine and venetoclax in pre­
vi­ously untreated acute mye­loid leu­ke­mia. N Engl J Med. 2020;383(7):617-
629.
27. DiNardo CD, Wei AH. How I treat acute mye­loid leu­ke­mia in the era of new
drugs. Blood. 2020;135(2):85-96.
28. Pollyea DA, Dinardo CD, Arellano ML, et al. Results of venetoclax and azac-
itidine com­bi­na­tion in che­mo­ther­apy inel­i­gi­ble untreated patients with
acute mye­loid leu­ke­mia with IDH 1/2 muta­tions. Paper presented at: ASH
Annual Meeting; (Virtual); 5-8 December 2020.
29. Maiti A, Rausch CR, Cortes JE, et  al. Outcomes of relapsed or refrac­tory
acute mye­loid leu­ke­mia after front­line hypomethylating agent and veneto-
clax reg­i­mens. Haematologica. 2021;106(3):894-898.
30. Maiti A, Qiao W, Sasaki K, et  al. Venetoclax with decitabine vs inten­sive
che­mo­ther­apy in acute mye­loid leu­ke­mia: a pro­pen­sity score matched
anal­y­sis strat­i­fied by risk of treat­ment-related mor­tal­ity. Am J Hematol.
2021;96(3):282-291.
31. Cortes JE, Heidel FH, Hellmann A, et  al. Randomized com­par­i­son of low
dose cytarabine with or with­out glasdegib in patients with newly diag­
nosed acute mye­ loid leu­ke­
mia or high-risk myelodysplastic syn­ drome.
Leukemia. 2019;33(2):379-389.
32. DiNardo CD, Stein EM, de Botton S, et  al. Durable remis­sions with ivo-
sidenib in IDH1-mutated relapsed or refrac­ tory AML. N Engl J Med.
2018;378(25):2386-2398.
33. DiNardo CD, Stein AS, Stein EM, et al. Mutant isocitrate dehy­dro­ge­nase 1 inhib­
i­tor ivosidenib in com­bi­na­tion with azacitidine for newly diag­nosed acute
mye­loid leu­ke­mia. J Clin Oncol. 2021;39(1):57-65.
34. Thol F, Schlenk RF, Heuser M, Ganser A. How I treat refrac­tory and early
relapsed acute mye­loid leu­ke­mia. Blood. 2015;126(3):319-327.
35. Cocciardi S, Dolnik A, Kapp-Schwoerer S, et  al. Clonal evo­lu­tion pat­
terns in acute mye­ loid leu­
ke­
mia with NPM1 muta­ tion. Nat Commun.
2019;10(1):2031.
36. Schmalbrock LK, Dolnik A, Cocciardi S, et al. Clonal evo­lu­tion of acute mye­
loid leu­ke­mia with FLT3-ITD muta­tion under treat­ment with midostaurin.
Blood. 2021;137(22):3093-3104.
37. Perl A, Martinelli G, Cortes JE, et al. Gilteritinib sig­nif­i­cantly pro­longs over­
all sur­vival in patients with FLT3-mutated (FLT3mut+) relapsed/refrac­tory
(R/R) acute mye­loid leu­ke­mia (AML): results from the Phase III ADMIRAL
trial. Abstract presented at: AACR Annual Meeting; 29 March-April 3 2019.
Abstract CT184.
38. Perl AE, Martinelli G, Cortes JE, et  al. Gilteritinib or che­ mo­ther­
apy for
relapsed or refrac­tory FLT3-mutated AML. N Engl J Med. 2019;381(18):1728-
1740.
39. Stein EM, DiNardo C, Altman JK, et  al. Safety and effi­cacy of AG-221, a
potent inhib­i­tor of mutant IDH2 that pro­motes dif­fer­en­ti­a­tion of mye­loid
cells in patients with advanced hema­to­logic malig­nan­cies: results of a
phase 1/2 trial. Abstract presented at: ASH 57th Annual Meeting and Expo-
sition; 5-8 Decem­ber 2015; Orlando, FL.
40. Stein EM, DiNardo CD, Pollyea DA, et al. Enasidenib in mutant IDH2 relapsed
or refrac­tory acute mye­loid leu­ke­mia. Blood. 2017;130(6):722-731.
41. Sallman DA, Asch AS, Al Malki MM, et al. The first-in-class anti-CD47 anti­
body magrolimab (5F9) in com­bi­na­tion with azacitidine is effec­tive in
MDS and AML patients: ongo­ing phase 1b results. Blood. 2019;134(suppl 1):
569.
42. Zhang Q , Bykov VJN, Wiman KG, Zawacka-Pankau J. APR-246 reacti- 45. Schuurhuis GJ, Heuser M, Freeman S, et al. Minimal/mea­sur­able resid­ual dis­
vates mutant p53 by targeting cys­ te­
ines 124 and 277. Cell Death Dis. ease in AML: a con­sen­sus doc­u­ment from the Euro­pean LeukemiaNet MRD
2018;9(5):439. Working Party. Blood. 2018;131(12):1275-1291.
43. Sallman DA, DeZern AE, Garcia-Manero G, et al. Eprenetapopt (APR-246)
and azacitidine in TP53-mutant myelodysplastic syn­dromes. J Clin Oncol.
2021;39(14):1584-1594.
44. Krivtsov AV, Evans K, Gadrey JY, et al. A menin-MLL inhib­i­tor induces spe­cific
chro­ma­tin changes and erad­i­cates dis­ease in mod­els of MLL-rearranged leu­ © 2021 by The Amer­i­can Society of Hematology
ke­mia. Cancer Cell. 2019;36(6):660-673.e11673e11. DOI 10.1182/hema­tol­ogy.2021000309

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/16/1851427/16thol.pdf by guest on 13 December 2021

Prakash Singh Shekhawat

AML ther­apy with newly approved sub­stances  | 23
 AML: SO MANY OPTIONS, SO LITTLE TIME

Whom should we treat with novel agents?

Specific indications for specific and challenging
populations

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/24/1851553/24wilde.pdf by guest on 13 December 2021

Lindsay Wilde and Margaret Kasner
Medical Oncology, Thomas Jefferson University, Philadelphia, PA

A relative wealth of new therapies for acute myeloid leukemia (AML) have led to a rapid shift in treatment paradigms for
this disease. Understanding whom, when, and how to treat is more complex than ever before. Here we explore whom
to treat with these available new therapies, focusing on special patient populations that include older adults, those with
relapsed disease, and those with TP53-mutated AML. These high-risk subgroups are some of the most challenging to care
for, but novel treatments are providing them with new hope.

LEARNING OBJECTIVES
• Learn the indications for newly approved and emerging drugs in AML in challenging populations
• Become familiar with a treatment algorithm for older, relapsed, and TP53-mutated AML patients

Introduction becomes more difficult to treat and cure. Five-year overall

The recent approval of a variety of novel therapies for
acute myeloid leukemia (AML) has provided more treat-
ment options for patients than ever before. However,
identifying whom to treat, and how, can be challenging.
Key elements in making these decisions include assessing
patient characteristics and obtaining comprehensive AML
profiling, including cytogenetic and molecular data. Even
with this information, the optimal therapeutic path is not
always clear. Here we highlight the complexities of select-
ing whom to treat with novel AML therapies.
CLINICAL CASE
An 86-year-old man with a past medical history of hyper-
tension and gastroesophageal reflux disease developed
leukocytosis, anemia, and thrombocytopenia. A bone mar-
row biopsy revealed 60% myeloblasts. Immunophenotyp-
ically, the blasts were CD34+, CD13+, and CD33+. He was
diagnosed with AML. Next-generation sequencing showed
mutations in ASXL1 and IDH1. Findings from the karyotype
were normal.
With a median age of 68 years at the time of diagno-
sis, AML is primarily a disease of older adults.1 However,
the incidence of AML not only increases with age but also
survival (OS) for patients diagnosed between the ages of
60 and 64 is 22%, while that of patients diagnosed over the
age of 80 is 1.3%.2 The advent of newer, less toxic treat-
ments may help to improve this disparity.
Outcomes for older adults with AML are poor because of
many complex reasons (Figure 1) First, patient factors such
as comorbidities and decreased baseline performance
status (PS) may prevent patients from receiving optimal
therapy or put them at higher risk of complications. The
number of comorbidities increases with age and corre-
lates with an increase in AML-related and all-cause mortal-
ity.3 Older patients are also more likely to present with an
ECOG PS of 2 or 3, and decreased PS is associated with an
increase in early death.4 Older patients with decreased PS
are more likely to experience early mortality than younger
patients with a similar PS, indicating the complex interplay
between these 2 factors.5,6
Second, disease-related factors also contribute to
poorer outcomes in older patients. Such patients are more
likely to have disease with high-risk chromosomal abnor-
malities (−5 or 5q−, −7 or 7q−, or 17p deletion), and the
incidence of having high-risk mutations in genes such as
ASXL1, RUNX1, and TP53 also increases with age.7 By Euro-
pean LeukemiaNet criteria, older patients more often
have high-risk disease, and survival for these patients is

24 | Hematology 2020 | ASH Education Program


Figure 1.  Factors leading to a poor prognosis in older AML patients.
decreased com­pared to youn­ger patients with sim­i­larly high-risk
dis­ease.8,9 Older patients also more fre­quently have dis­ease that
is inher­ently che­mo­ther­apy-resis­tant due to overexpression of
the MDR1 mul­ti­drug-resis­tance gene.10
Finally, sev­ eral fac­tors influ­
ence the type and inten­ sity of
treat­ments offered to older patients, lead­ing to decreased sur­
vival. Older patients are less likely to be offered up-front treat­
ment for AML and are less likely to undergo allo­ge­neic stem cell
trans­plant.11,12 This is despite the fact that stud­ies have dem­on­
strated that treat­ment of any inten­sity improves sur­vival in this
patient pop­u­la­tion.11 The rea­sons for these disparities include
phy­si­cian bias regard­ing the treat­ment of older patients, care­
giver issues, and other socio­eco­nomic inequities.
Given the urgent need to address disparities in out­comes
for older indi­vid­u­als with AML, in 2020 the Amer­i­can Society of
Hematology published guide­lines for treating these patients.13
The guide­lines are based on a sys­tem­atic review of the lit­er­
a­ture and con­sider many of the pre­vi­ously discussed fac­tors.
Included is a rec­om­men­da­tion to offer anti­leu­ke­mic ther­apy
over best sup­port­ive care and to attempt inten­sive ther­apy
when pos­si­ble. As newer ther­a­pies con­tinue to emerge, these
guide­lines must be adapted and mod­i­fied, par­tic­u­larly with
regard to inten­sity of treat­ment. Ultimately, opti­mal ther­apy for
older adults with AML is that which pro­vi­des the best depth of
response, imparts the least amount of tox­ic­ity, and meets the
goals of the indi­vid­ual patient.
Recent efforts to develop safer and more effec­tive treat­
ments for older adults with AML are begin­ning to pay off. Lead-
ing the way has been the US Food and Drug Administration
(FDA)-approved front­line com­bi­na­tion of azacitidine and vene-
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/24/1851553/24wilde.pdf by guest on 13 December 2021
toclax (AZA-VEN). This lower-inten­sity treat­ment was shown in
the phase 3, mul­ti­cen­ter, ran­dom­ized, dou­ble-blind, pla­cebo-
con­trolled VIALE-A trial to improve OS when com­pared to azac-
itidine plus pla­cebo (14.7 months vs 9.6 months; haz­ard ratio
[HR], 0.66; 95% CI, 0.52-0.85; P < .001) in patients over 75 or
in youn­ger patients with sig­nif­i­cant comorbidities.14 Moreover,
the com­ pos­ite com­ plete remis­sion (CR) rate was 66.4% for
patients in the AZA-VEN group com­pared to 28.3% in the con­
trol group, and rates of red blood cell and plate­let trans­fu­sion
inde­pen­dence were 59.8% and 68.5%, respec­tively. Regarding
safety, rates of grade ≥3 neutropenia and neutropenic fever
were higher in the AZA-VEN group; how­ever, 30-day mor­tal­ity
was sim­i­lar between the 2 groups (7% with AZA-VEN and 6%
with AZA-pla­cebo).
Molecular sub­group anal­y­sis of the VIALE-A study sug­gests
that muta­tions in IDH1 or IDH2 may con­fer a par­tic­u­larly favor­
able response to AZA-VEN. A pooled anal­y­sis of the phase 1b and
phase 3 stud­ies showed that CR/complete remission with par-
tial hematologic recovery (CRh) for patients with an IDH1 muta­
tion was 59% with AZA-VEN vs 9% with AZA-pla­cebo. CR/CRh
for patients with an IDH2 muta­tion was 80% with AZA-VEN vs 6%
with AZA-pla­cebo.15 This improve­ment in remis­sion rates trans­
lated into an improve­ment in dura­tion of response and OS. This
dif­
fered from other molec­ u­
lar sub­sets, such as FLT3-mutated
or TP53-mutated dis­ ease, where responses were bet­ ter with
AZA-VEN, but OS was not affected.
The oral IDH1 inhib­i­tor ivosidenib is another FDA-approved
front­line ther­apy for older patients with AML. This treat­ ment
was ini­tially stud­ied in a phase 1, mul­ti­cen­ter, open-label, dose-
esca­la­tion, dose-expan­sion study in patients with AML with an
Therapies for dif­fi­cult AML pop­u­la­tions | 25
 Newly diagnosed
AML in an older
pa
ent

Clinical trial*

No targetable
Targetable muta on TP53 muta on
muta on

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/24/1851553/24wilde.pdf by guest on 13 December 2021

HMA + venetoclax
HMA + venetoclax
FLT3 IDH1 IDH2 magrolimab +/- HMA
LDAC + glasdegib
APR-246 + aza

HMA + venetoclax
HMA + venetoclax HMA + venetoclax
FLT3 inhibitor +/-
Ivosidenib +/- HMA Enasidenib +/- HMA
HMA

Figure 2.  Algorithm for the treatment of newly diagnosed AML in the older patient.

IDH1 muta­tion.16 The median age in the study was 76.5 years, and
56% of patients were over 75. The CR/CRh rate was 42%, with
a median time to response of 2.8 months. The median fol­low-up
was 23.5 months, with a median OS of 12.6 months. The util­ity
of ivosidenib as a sin­gle agent is lim­ited, but it may be a good
option for patients unable to receive venetoclax or tol­er­ate com­
bi­na­tion treat­ment. A study of azacitidine, venetoclax, and ivos-
idenib is under­way (NCT03471260), and the results, if pos­i­tive,
will increase the value of this agent.
The phase 2 BRIGHT AML 1003 study assessed the com­bi­na­
tion of low-dose cytarabine and glasdegib, a hedge­hog path­
way inhib­i­tor, vs low-dose cytarabine alone.17 The median age
in the glasdegib arm was 77 and in the low-dose cytarabine
arm, 75. The addi­tion of glasdegib improved median OS (8.8 vs
4.6 months; HR, 0.51; 80% CI, 0.39-0.67; P = .0004). This led to
the FDA approval of glasdegib plus low-dose cytarabine for the
treat­ment of older patients with AML. However, given the other
FDA-approved treat­ment options, which appear to be more
effi­ca­cious and equally as safe, the clin­ic
­ al util­ity of this treat­
ment is ques­tion­able. Subsequent sub­set anal­y­sis indi­cated
slightly improved sur­vival in patients with sec­ond­ary AML as
opposed to de novo dis­ease.18 Though no head-to-head com­
par­i­son exists, this sur­vival was shorter than that of patients
treated with AZA-VEN in VIALE-A.14,18
Numerous tri­als eval­u­at­ing targeted ther­a­pies, com­bi­na­tion
ther­a­pies, and novel agents for the front­line treat­ment of older
patients with AML are under­way. In addi­tion, geri­at­ric assess­
ment tools are being devel­oped and eval­ua ­ ted with the hope of
improv­ing treat­ment selec­tion, side effects, and patient qual­ity
of life. These advances will con­tinue to have a pos­i­tive impact on
the lives of older patients with AML. (See Figure 2 for a treat­ment
algo­rithm for older patients.)
CLINICAL CASE
An 81-year-old woman was treated with AZA-VEN for newly diag­
nosed AML. After 2 cycles, her dis­ease went into CR. She remained
on this treat­ment for 7 cycles before her dis­ease relapsed. Repeat
molec­u­lar test­ing showed a new FLT3-ITD muta­tion and a new
muta­tion in IDH2.
In spite of the advances made in the first-line treat­ment of
AML, approx­i­ma­tely 50% of patients expe­ri­ence a relapse.19,20
The treat­ment options for relapsed or refrac­tory (R/R) dis­ease
remain lim­ited, and 5-year OS for these patients is approx­i­ma­tely
10%.21 Outcomes after first-line treat­ment with a hypomethylat-
ing agent and venetoclax are par­tic­ul­arly poor, with a median
sur­vival of 2.4 months.22 At the time of relapse, molec­u­lar test­
ing should be repeated to assess for the pres­ence of tar­get­able
muta­tions. Due to clonal evo­lu­tion, pre­vi­ously iden­ti­fi­able tar­
gets may be lost, or the emer­gence of subclones may lead to the
pres­ence of new tar­gets.23
Several targeted ther­a­pies have been approved for the treat­
ment of R/R AML. For patients with FLT3-mutated dis­ease, gilter-
itinib is a potent oral FLT3 inhib­i­tor that can be used as a sin­gle
agent. In the ran­dom­ized phase 3 ADMIRAL study, gilteritinib
improved OS when com­pared to inves­ti­ga­tor-choice che­mo­
ther­apy (9.3 months vs 5.6 months; two-sided P < .001).24 About
34% of patients receiv­ing gilteritinib had a CR/compelete remis-
sion with incomplete count recovery com­pared to 15.3% in the
che­mo­ther­apy group (risk dif­fer­ence, 18.6 per­cent­age points;
95% CI, 9.8-27.4). Given its mod­est effi­cacy as a sin­gle agent,
gilteritinib is being stud­ied in com­bi­na­tion with other ther­a­pies
in the relapsed and front­line set­ting. Early data on the com­bi­na­
tion of gilteritinib and venetoclax for relapsed FLT3-mutated AML
look prom­is­ing.25

26  |  Hematology 2021  |  ASH Education Program

 Relapsed AML in
an older paent
Clinical trial*
No targetable
mutaon
Gemtuzumab
ozogamicin FLT3
HMA + venetoclax
Gilterinib +/-
HMA
Figure 3.  Algorithm for the treatment of relapsed older AML patients.
The pre­vi­ously men­tioned IDH1 inhib­i­tor, ivosidenib, is also
approved for the treat­ment of R/R AML with an IDH1 muta­tion.
In a large, sin­gle-arm phase 1 study, ivosidenib was shown to be
safe and well tol­er­ated and to induce a CR or CRh in 30.4% of
patients.26 Similarly, the IDH2 inhib­i­tor enasidenib is approved for
the treat­ment of R/R AML with a muta­tion in IDH2. This approval
was based on the phase 1/2 study that showed a CR rate of
19.3% in this pop­u­la­tion.27 However, the phase 3 study for this
drug failed to meet the pri­mary OS end point.28 Combination
stud­ies with enasidenib are under­way.
Finally, the some­what con­tro­ver­sial CD33-targeted anti­body-
drug con­ju­gate gemtuzumab ozogamicin (GO) is also approved
for patients with AML that is R/R. First approved in 2000, GO
was vol­un­tar­ily with­drawn from the US mar­ket in 2010 due to
safety con­cerns. Reapproval was granted in 2017 after fur­ther
review of the data. In the ALFA 9801 trial, GO, which was given as
a sin­gle agent or in com­bi­na­tion with che­mo­ther­apy to patients
with R/R dis­ease, dem­on­strated a 39% CR rate.29 Pooled anal­y­sis
of the data from 3 other phase 2 stud­ies showed a slightly lower
over­all remis­sion rate of 26%.30 There is inter­est in com­bi­na­tion
ther­apy with GO and stud­ies in this area are ongo­ing.
Although these agents rep­re­sent prog­ress for the treat­ment
of patients with R/R AML, response rates and long-term sur­vival
are still lim­ited, and clin­i­cal tri­als remain the pre­ferred option
at first relapse or after the use of targeted ther­ a­
pies. Many
novel agents are under inves­ti­ga­tion, includ­ing small mol­e­cule
inhib­i­tors, immunotherapies, and com­bi­na­tion ther­a­pies, and
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/24/1851553/24wilde.pdf by guest on 13 December 2021
Targetable
mutaon
IDH1 IDH2
Ivosidenib +/- Enasidenib +/-
HMA HMA
new can­di­dates for targeted treat­ments con­tinue to be iden­
ti­fied. (See Figure 3 for a treat­ment algo­rithm for patients with
relapsed dis­ease.)
CLINICAL CASE
A 78-year-old woman with a past med­i­cal his­tory of hyper­ten­
sion devel­oped leu­ko­cy­to­sis, ane­mia, and throm­bo­cy­to­pe­nia.
A bone mar­row biopsy revealed 60% mye­lo­blasts. Immunophe-
notypically, the blasts were CD34+, CD13+, and CD33+. She was
diag­nosed with AML. Next-gen­er­a­tion sequenc­ing showed a
muta­tion in TP53.
TP53, an essen­tial tumor sup­pres­sor gene, is mutated in up
to 10% of AML cases.31 It is found more fre­quently in ther­apy-
related and com­plex-kar­yo­type AML. This muta­tional sta­tus is
an impor­tant prog­nos­tic fac­tor that con­sis­tently cor­re­sponds to
extremely poor out­comes, includ­ing low response rates to tra­
di­tional cyto­toxic che­mo­ther­apy and rare instances of long-term
sur­vival after allo­ge­neic trans­plan­ta­tion.32,33
Unfortunately, the novel agents approved for the treat­ment
of AML, includ­ing ther­apy related and com­plex kar­yo­type, have
not led to sig­nif­i­cant improve­ment in the sur­vival of patients with
TP53-mutated AML. For exam­ple, a 2018 study eval­u­ated this sub­
group of patients in the reg­is­tra­tion trial of CPX-351 vs 7 + 3 in
older adults with newly diag­nosed high-risk or sec­ond­ary AML.34
Patients with the muta­tion had a median sur­vival of 5.7 months
Therapies for dif­fi­cult AML pop­u­la­tions | 27
in the CPX-351 arm and 5.1 months in the 7 + 3 arm, dem­on­strat­
ing no ben­e­fit. A sub­se­quent small study by Kim et al,35 which
looked at the real-world use of CPX-351 in 53 patients, 30% of
whom had TP53-mutated AML, showed more encour­ag­ing results
in this sub­set, with an overall response rate (ORR) of 57% (8/14)
and a 63% measurable residual disease neg­a­tiv­ity rate by next
generation sequencing. However, these ret­ro­spec­tive data should
be interpreted cau­tiously. Skepticism about the effi­cacy of CPX-351
for TP53-­mutated AML remains.
The pre­vi­ously discussed VIALE-A trial showed sig­nif­i­cantly
higher rates of com­ pos­
ite remis­ sion for patients with TP53-
mutated AML who received azacytadine and venetoclax than
those in the con­trol group (55.3% vs 0%, respec­tively; 95% CI,
38.3-71.4; P < .001).14 However, this did not trans­late to an increase
in OS. Disappointingly, com­bin­ing venetoclax with 10 days of
decitabine improved nei­ther response rates nor OS.36 The addi­
tion of venetoclax to more aggres­sive ther­apy (FLAG-IDA) also
failed to show improve­ment in the TP53-mutated AML sub­group.37
Hope for this poor-risk pop­u­la­tion likely lies in the use of
novel agents. For exam­ple, magrolimab is a first-in-class mac­
ro­phage immune check­ point inhib­ i­
tor targeting CD47. CD47
is a “do not eat me” sig­nal overexpressed in mul­ti­ple can­cers,
includ­ing AML.38 Its acti­ va­ tion leads to mac­ ro­
phage immune
eva­ sion. Magrolimab elim­ i­
na­tes tumor cells through mac­ ro­
phage phago­cy­to­sis. Data were presented at ASH 2020 from
a trial using magrolimab plus azacitidine in patients with newly
diag­nosed AML who were inel­i­gi­ble for inten­sive che­mo­ther­apy.
The study pri­mar­ily enrolled patients with TP53-mutated AML
(65%, as they were pref­er­en­tially enrolled after an early pro­to­col
amend­ment).39 The com­bi­na­tion induced a 71% ORR and 48%
CR rate in this high-risk sub­group. Moreover, the median OS for
these patients was 12.9 months. Based upon these data and the
high unmet need in this pop­u­la­tion, a ran­dom­ized phase 3 trial
of magrolimab plus azacitidine vs AZA-VEN or inten­sive induc­tion
as a front­line ther­apy for TP53-mutant AML is pre­par­ing to begin
recruit­ment (NCT04778397).
Another novel agent that tar­gets TP53 more spe­cif­i­cally is
APR-246, a meth­yl­ated PRIMA-1 ana­logue. It is a first-in-class small
mol­e­cule that selec­tively induces apo­pto­sis in TP53 mutant can­
cer cells.40 The agent is spon­ta­ne­ously converted into the active
spe­ cies meth­ y­
lene quinuclidinone, which cova­ lently binds to
cys­te­ine res­i­dues in mutant p53, thereby pro­duc­ing ther­mo­
dy­namic sta­bi­li­za­tion of the pro­tein and shifting equi­lib­rium
toward a func­tional con­for­ma­tion. Two early phase stud­ies were
recently published assessing the com­bi­na­tion of APR-246 and
azacitidine. A phase 1b/2 study in patients with TP53-mutated
myelo dysplasic syndrome (MDS) or low-blast per­cent­age AML
showed the treat­ment to be safe and well tol­er­ated.41 The ORR
was 88%, with a CR rate of 61% for the MDS cohort and 50% for
the AML; median OS was 10.8 months (95% CI, 8.1-13.4). A sim­i­
lar phase 2 study of APR-246 plus azacitidine included patients
with MDS, low-blast AML, and AML with blasts >30%.42 Response
rates and OS for the MDS and low-blast AML cohorts were sim­i­lar
to the pre­vi­ously discussed trial. However, response rates and
sur­vival in patients with AML with >30% blasts were lower (ORR
50%, CR 0%, OS 3 months). These stud­ies are encour­ag­ing for
patients with MDS or low-blast AML but sug­gest that trip­let ther­
apy or other com­bi­na­tions with APR-246 should be explored in
AML with higher blast counts. One such study looking at APR-246
plus azacitidine and venetoclax is ongo­ing (NCT04214860).
28  |  Hematology 2021  |  ASH Education Program
Finally, the ongo­ing trial GLAD-AML (NCT03798678) addresses
this poor-risk pop­ul­a­tion in a ran­dom­ized phase 2 study com­bin­
ing glasdegib with DEC5 or DEC10 in patients with newly diag­
nosed poor-risk AML.
Conclusion
Understanding who will ben­e­fit from cur­rent and future novel
AML ther­ap ­ ies is an evolv­ing area. Patient age and fit­ness, dis­
ease biol­ogy, and treat­ment mech­a­nism of action must all­be
taken into con­sid­er­ation when choos­ing whom to treat and how.
In gen­eral, even low-inten­sity treat­ment pro­vi­des a sur­vival ben­
e­fit over best sup­port­ive care to patients of all­ages and molec­
u­lar and cyto­ge­netic sub­groups. Trials looking to build on recent
improve­ments in out­comes are ongo­ing, includ­ing some with
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/24/1851553/24wilde.pdf by guest on 13 December 2021
dou­blet and trip­let ther­apy as well as oth­ers looking at drug
sequenc­ing, MRD erad­i­ca­tion, and main­te­nance.
Conflict-of-inter­est dis­clo­sure
Lindsay Wilde: no con­flicts to dis­close.
Margaret Kasner: research funding: Astellas, Diachi, Gilead, Jazz,
Telios, Ono, Otsuka, Pfizer; hon­or­ aria: Kite, Jazz, Ono.
Off-label drug use
Lindsay Wilde: no discussion of off-label drug use outside of the
setting of clinical trials.
Margaret Kasner: no discussion of off-label drug use outside of
the setting of clinical trials.
Correspondence
Margaret Kasner, Thomas Jefferson University, Medical Oncology,
834 Chestnut St, Ste 315, Philadelphia, PA 19107; e-mail: margaret​
­.kasner@jefferson​­.edu.
References
1. National Cancer Institute. SEER pre­lim­i­nary can­cer inci­dence rate esti­ma­
tes for 2017, and diag­no­sis years 2000 to 2017. Accessed 2019. https:​­/​­/seer​
­.cancer​­.gov​­/statistics​­/preliminary​­-estimates​­/.
2. Shallis RM, Wang R, Davidoff A, Ma X, Zeidan AM. Epidemiology of acute
mye­loid leu­ke­mia: recent prog­ress and endur­ing chal­lenges. Blood Rev.
2019;36:70-87.
3. Mohammadi M, Cao Y, Glimelius I, Bottai M, Eloranta S, Smedby KE. The
impact of comorbid dis­ease his­tory on all­-cause and can­cer-spe­cific mor­
tal­ity in mye­loid leu­ke­mia and mye­loma—a Swed­ish pop­u­la­tion-based
study. BMC Cancer. 2015;15:850.
4. Appelbaum FR, Gundacker H, Head DR, et al. Age and acute mye­loid leu­
ke­mia. Blood. 2006;107(9):3481-3485.
5. Klepin HD, Estey E, Kadia T. More ver­sus less ther­apy for older adults with
acute mye­loid leu­ke­mia: new per­spec­tives on an old debate. Am Soc Clin
Oncol Educ Book. 2019;39:421-432.
6. Bhatt VR, Chen B, Gyawali B, Lee SJ. Socioeconomic and health sys­tem fac­
tors asso­ci­ated with lower uti­li­za­tion of hema­to­poi­etic cell trans­plan­ta­tion
in older patients with acute mye­loid leu­ke­mia. Bone Marrow Transplant.
2018;53(10):1288-1294.
7. Fröhling S, Schlenk RF, Kayser S, et al; Ger­man-Austrian AML Study Group.
Cytogenetics and age are major deter­mi­nants of out­come in inten­sively
treated acute mye­loid leu­ke­mia patients older than 60 years: results from
AMLSG trial AML HD98-B. Blood. 2006;108(10):3280-3288.
8. Prassek VV, Rothenberg-Thurley M, Sauerland MC, et al. Genetics of acute
mye­loid leu­ke­mia in the elderly: muta­tion spec­trum and clin­i­cal impact
in inten­sively treated patients aged 75 years or older. Haematologica.
2018;103(11):1853-1861.
9. Herold T, Rothenberg-Thurley M, Grunwald VV, et al. Validation and refine­
ment of the revised 2017 Euro­pean LeukemiaNet genetic risk strat­i­fic ­ a­tion
of acute mye­loid leu­ke­mia. Leukemia. 2020;34(12):3161-3172.
10. Leith CP, Kopecky KJ, Godwin J, et  al. Acute mye­ loid leu­ke­
mia in the
elderly: assess­ment of mul­ti­drug resis­tance (MDR1) and cyto­ge­net­ics distin-
 guishes bio­logic sub­groups with remark­ably dis­tinct responses to stan­dard
che­ mo­ ther­apy. A Southwest Oncology Group Study. Blood. 1997;89(9):
3323-3329.
11. Medeiros BC, Satram-Hoang S, Hurst D, Hoang KQ , Momin F, Reyes C.
Big data anal­ y­sis of treat­ ment pat­ terns and out­ comes among elderly
acute mye­ loid leu­ke­mia patients in the United States. Ann Hematol.
2015;94(7):1127-1138.
12. Muffly L, Pasquini MC, Martens M, et al. Increasing use of allo­ge­neic hema­
to­poi­etic cell trans­plan­ta­tion in patients aged 70 years and older in the
United States. Blood. 2017;130(9):1156-1164.
13. Sekeres MA, Guyatt G, Abel G, et al. Amer­i­can Society of Hematology 2020
guide­lines for treating newly diag­nosed acute mye­loid leu­ke­mia in older
adults. Blood Adv. 2020;4(15):3528-3549.
14. DiNardo CD, Jonas BA, Pullarkat V, et al. Azacitidine and venetoclax in pre­vi­
ously untreated acute mye­loid leu­ke­mia. N Engl J Med. 2020;383(7):617-629.
15. Pollyea DA, Dinardo CD, Arellano ML, et al. Results of venetoclax and azacit-
idine com­bi­na­tion in che­mo­ther­apy inel­i­gi­ble untreated patients with acute
mye­loid leu­ke­mia with IDH 1/2 muta­tions. Blood. 2020;136(suppl 1):5-7.
16. Roboz GJ, DiNardo CD, Stein EM, et al. Ivosidenib induces deep dura­ble
remis­sions in patients with newly diag­nosed IDH1-mutant acute mye­loid
leu­ke­mia. Blood. 2020;135(7):463-471.
17. Cortes JE, Heidel FH, Hellmann A, et  al. Randomized com­par­i­son of low
dose cytarabine with or with­out glasdegib in patients with newly diag­
nosed acute mye­ loid leu­ ke­
mia or high-risk myelodysplastic syn­ drome.
Leukemia. 2019;33(2):379-389.
18. Hueser M, Robak T, Montesinos P, et al. Glasdegib (GLAS) plus low-dose
cytarabine (LDAC) in AML or MDS: BRIGHT AML 1003 final report and 4-year
over­all sur­vival (OS) fol­low-up. J Clin Oncol. 2020;38(suppl 15):7509.
19. Zhu C-Y, Chen G-F, Zhou W, et al. Outcome and prog­nos­tic fac­tors of high-
risk acute mye­loid leu­ke­mia after allo­ge­neic hema­to­poi­etic stem cell trans­
plan­ta­tion. Ann Transplant. 2019;24:328-340.
20. Oliva EN, Franek J, Patel D, Zaidi O, Nehme SA, Almeida AM. The real-world
inci­dence of relapse in acute mye­loid leu­ke­mia (AML): a sys­tem­atic lit­er­a­
ture review (SLR). Blood. 2018;132(suppl 1):5188.
21. Ganzel C, Sun Z, Cripe LD, et al. Very poor long-term sur­vival in past and
more recent stud­ies for relapsed AML patients: the ECOG-ACRIN expe­ri­
ence. Am J Hematol. 2018;93(8):1074-1081.
22. Abhishek M, Caitlin RR, Jorge EC, et al. Outcomes of relapsed or refrac­tory
acute mye­loid leu­ke­mia after front­line hypomethylating agent and veneto-
clax reg­i­mens. Haematologica. 2020;106(3):894-898.
23. Vosberg S, Greif PA. Clonal evo­lu­tion of acute mye­loid leu­ke­mia from diag­
no­sis to relapse. Genes Chromosomes Cancer. 2019;58(12):839-849.
24. Perl AE, Martinelli G, Cortes JE, et al. Gilteritinib or che­mo­ther­apy for relapsed
or refrac­tory FLT3-mutated AML. N Engl J Med. 2019;381(18):1728-1740.
25. Perl AE, Daver NG, Pratz KW, et al. Venetoclax in com­bi­na­tion with gilteri-
tinib in patients with relapsed/refrac­tory acute mye­loid leu­ke­mia: a phase
1b study. Blood. 2019;134(suppl 1):3910.
26. DiNardo CD, Stein EM, de Botton S, et al. Durable remis­sions with ivosidenib
in IDH1-mutated relapsed or refrac­tory AML. N Engl J Med. 2018;378(25):
2386-2398.
27. Stein EM, DiNardo CD, Pollyea DA, et al. Enasidenib in mutant IDH2 relapsed
or refrac­tory acute mye­loid leu­ke­mia. Blood. 2017;130(6):722-731.
28. Bristol Meyers Squibb. Bristol Meyers Squibb pro­vi­des update on phase 3
IDHENTIFY trial in patients with relapsed or refrac­tory acute mye­loid leu­ke­
mia. News release. 24 August 2020.
Prakash Singh Shekhawat
29. Merabet F, Boissel N, Reman O, et al. Outcome of relapsed AML patients
aged between 50 and 70: the ALFA group expe­ri­ence (ALFA 9801 study).
Blood. 2007;110(11):1857.
30. Larson RA, Sievers EL, Stadtmauer EA, et al. Final report of the effi­cacy and
safety of gemtuzumab ozogamicin (Mylotarg) in patients with CD33-pos­i­
tive acute mye­loid leu­ke­mia in first recur­rence. Cancer. 2005;104(7):1442-
1452.
31. Papaemmanuil E, Gerstung M, Bullinger L, et al. Genomic clas­si­fi­ca­tion and
prog­no­sis in acute mye­loid leu­ke­mia. N Engl J Med. 2016;374(23):2209-
2221.
32. Kadia TM, Jain P, Ravandi F, et al. TP53 muta­tions in newly diag­nosed acute
mye­loid leu­ke­mia: clinicomolecular char­ac­ter­is­tics, response to ther­apy,
and out­comes. Cancer. 2016;122(22):3484-3491.
33. Najima Y, Sadato D, Harada Y, et al. Prognostic impact of TP53 muta­tion,
monosomal kar­yo­type, and prior mye­loid dis­or­der in nonremission acute
mye­loid leu­ke­mia at allo-HSCT. Bone Marrow Transplant. 2021;56(2):334-
346.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/24/1851553/24wilde.pdf by guest on 13 December 2021
34. Goldberg AD, Talati C, Desai P, et  al. TP53 muta­ tions pre­
dict poorer
responses to CPX-351 in acute mye­loid leu­ke­mia. Blood. 2018;132(suppl
1):1433.
35. Kim GY, Koprivnikar JL, Testi R, et al. Treatment with CPX-351 induces deep
responses and TP53 muta­tion clear­ance in patients with t-AML and AML
MRC, includ­ing youn­ger patients and those with pre-existing MPNs: a real-
world expe­ri­ence. Blood. 2020;136(suppl 1):49-50.
36. Kim K, Maiti A, Kadia TM, et al. Outcomes of TP53-mutant acute mye­loid
leu­ke­mia with venetoclax and decitabine. Blood. 2020;136(suppl 1):33-36.
37. Lachowiez C, Konopleva M, Kadia TM, et al. Interim anal­y­sis of the phase
1b/2 study of the BCL-2 inhib­i­tor venetoclax in com­bi­na­tion with stan­dard
inten­sive AML induc­tion/con­sol­i­da­tion ther­apy with FLAG-IDA in patients
with newly diag­nosed or relapsed/refrac­tory AML. Blood. 2020;136(suppl
1):18-20.
38. Chao MP, Takimoto CH, Feng DD, et al. Therapeutic targeting of the mac­
ro­phage immune check­point CD47 in mye­loid malig­nan­cies. Front Oncol.
2019;9:1380.
39. Sallman DA, Asch A, Kambhampati S, et al. The first-in-class anti-CD47 anti­
body magrolimab com­bined with azacitidine is well-tol­er­ated and effec­
tive in AML patients: phase 1b results. Blood. 2019;134(suppl 1):569.
40. Sallman DA. To tar­get the untargetable: elu­ci­da­tion of syn­ergy of APR-
246 and azacitidine in TP53 mutant myelodysplastic syn­dromes and acute
mye­loid leu­ke­mia. Haematologica. 2020;105(6):1470-1472.
41. Sallman DA, DeZern AE, Garcia-Manero G, et al. Eprenetapopt (APR-246)
and azacitidine in TP53-mutant myelodysplastic syn­dromes. J Clin Oncol.
2021;39(14):1584-1594.
42. Cluzeau T, Sebert M, Rahmé R, et  al. Eprenetapopt plus azacitidine in
TP53-mutated myelodysplastic syn­dromes and acute mye­loid leu­ke­mia: a
phase II study by the groupe fran­co­phone des myélodysplasies (GFM). J
Clin Oncol. 2021;39(14):1575-1583.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000228
Therapies for dif­fi­cult AML pop­u­la­tions | 29
 CHALLENGES IN MULTIPLE MYELOMA TREATMENT

High-risk multiple myeloma: how to treat

at diagnosis and relapse?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/30/1851766/30mateos.pdf by guest on 13 December 2021

María-Victoria Mateos, Borja Puertas Martínez, and Verónica González-Calle
Hospital Universitario de Salamanca, Instituto de Investigación Biomédica de Salamanca (IBSAL), Centro de Investigación del Cancer (IBMCC-USAL,
CSIC), Salamanca, Spain

Patients with multiple myeloma have experienced a great improvement in survival over the past century because of the
introduction of novel therapeutic strategies. However, a subgroup of patients with poorer outcomes than expected is
considered high risk and identified by the presence of patient- and disease-based factors such as frailty, extramedullary
disease, cytogenetic abnormalities, or even relapses occurring earlier than expected according to the baseline factors.
Although the management of patients with high-risk features is not well established because of the lack of specific trials
in this subgroup of patients and because of their underrepresentation in the clinical trials, treatment should be planned
on 2 pillars: (1) poor prognosis with the presence of high-risk features can be at least improved or even abrogated by
achieving a deep and sustained response over time, and (2) this can most likely be obtained through using the best ther-
apeutic options and in a response-adapted way. Some clinical trials that have been planned or are ongoing include only
patients with high-risk features, using the most effective therapies (proteasome inhibitors, immunomodulatory drugs,
and anti-CD38 monoclonal antibodies) as well as chimeric antigen receptor T cells and T-cell engagers that will unravel
what the best therapeutic approach will be to overcome the poor prognosis of the presence of high-risk features.

LEARNING OBJECTIVES
• Identify high-risk myeloma based on patient-, disease-, or outcome-based factors
• Be able to define the key objectives to overcome poor prognosis with the presence of high-risk features
• Define the best therapeutic strategy for patients with high-risk features

CLINICAL CASE maintenance therapy, relapse occurred with reappearance

A 48-year-old man with newly diagnosed (ND), Revised
International Staging System (R-ISS) III (ISS III plus del(17p)),
Bence-Jones κ multiple myeloma (MM) sought treat-
ment and consultation for MM that had been diagnosed
in another institution. The patient had an active lifestyle,
and the workup showed mild anemia and small lytic lesions
in the pelvis and femora as myeloma-defining events. His
Eastern Cooperative Oncology Group performance status
was 1. He was treated with 6 induction cycles of lenalid-
omide, bortezomib, and dexamethasone (RVd), achieving
a very good partial response, followed by high-dose mel-
phalan and autologous stem cell transplantation (HDM-
ASCT), achieving stringent complete remission (sCR) with
minimal residual disease (MRD) positivity. He rejected
a second ASCT and proceeded to consolidation with 2
cycles of lenalidomide, carfilzomib, and dexamethasone,
achieving sCR and MRD negativity. Maintenance with lena-
lidomide was prescribed. Twelve months after starting
of the M-component in urine. He was included in a clini-
cal trial and treated with B-cell maturation antigen (BCMA)
chimeric antigen receptor T (CAR-T) cells, and a new sCR
and MRD negativity were achieved. The patient continues
in follow-up.
How do we define high-risk patients with MM?
Table 1 summarizes the most relevant patient- and disease-
based factors to define high-risk patients.
Patient-based factors
Frailty
For a long time, chronological age influenced treatment
decisions, and the outcome was poor for the elderly. The
International Myeloma Working Group (IMWG), through
a pooled analysis including 869 ND elderly patients en-
rolled in clinical trials, built a simplified geriatric score

30 | Hematology 2021 | ASH Education Program

Table 1. Patient- and dis­ease-based fac­tors for the iden­ti­fi­ca­tion
of high-risk MM
High-risk fea­tures
Patient-based fac­tors
  Frailty sta­tus
Disease-based fac­tors
  Aggressiveness in the
clin­i­cal pre­sen­ta­tion
bor­ing +1q had a shorter median pro­gres­sion-free sur­vival (PFS;
41.9 months) and OS (not reached) com­pared with those with­out
+1q (PFS of 65.1 months, P  =  .002 and OS not reached, P  =  .003).
Definition The neg­a­tive impact on sur­vival with +1q may be more pro­found
if there is ampli­fi­ca­tion of 1q, defined by the pres­ence of 4 or
more cop­ies of chro­mo­some 1q (median PFS of 25.1 months) or in
IMWG frailty score
asso­ci­a­tion with other high-risk CAs (median PFS of 34.6 months).
Modified IMWG frailty score
R-MCI The poor prog­no­sis of patients who have +1q with a coexisting
GAH dele­tion of chro­mo­some 1p (del(1p)) has also been described.6
Although it is well accepted that del(17p) is the CA with more
prog­nos­tic impact in MM, some ques­tions are under debate: (1)
Extramedullary dis­ease (no bone- What is the opti­mal thresh­old to pre­dict poor prog­no­sis? (2)
related plasmacytomas)
Plasma cell leu­ke­mia
Is TP53 an opti­mal molec­u­lar tar­get? (3) What about mono- or
biallelic dele­tion and/or inac­ti­va­tion of TP53 through muta­

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/30/1851766/30mateos.pdf by guest on 13 December 2021

LDH ele­vated
  Cytogenetic ­ del(17p), t(4;14), t(14;16), amp1q, del(1p)
abnor­mal­i­ties
 Mutations TP53
  Biochemical ­ LDH ele­vated
abnor­mal­i­ties β2-microglobulin ≥5.5 mg/L
Albumin lev­els ≤3.5 mg/L
Prognostic scores
 R-ISS R-ISS III: beta2-microglobulin ≥5.5 mg/L
plus either LDH ele­vated or high-risk
CA (del(17p), t(4;14), or t(14;16))
GAH, geri­at­ric assess­ment in hema­tol­ogy; R-MCI, Revised Myeloma
Comorbidity Index.
based on age, comorbidities, and cog­ni­tive and phys­i­cal con­
di­tions to dis­tin­guish among fit (score   =   0), inter­me­di­ate fit­ness
(score  =  1), and frailty (score ≥ 2). Frail patients showed a sig­nif­
i­cantly shorter over­all sur­vival (OS; 57% at 3 years) than unfit
(76% at 3 years) and fit patients (84% at 3 years).1 Many clin­i­cal
tri­als are using an IMWG-mod­i­fied frailty model to iden­tify frail
patients because of the impor­tance of their iden­ti­fi­ca­tion for
the treat­ment deci­sion-mak­ing pro­cess.2
Disease-based fac­tors
Features asso­ci­ated with dis­ease aggres­sive­ness
The pres­ence of extramedullary dis­ease (EMD) or plasma cell leu­
ke­mia (PCL), pri­mary or sec­ond­ary, is infre­quent, but these are
con­sid­ered high-risk fea­tures not only because the plasma cells
escape from the bone mar­row envi­ron­ment but also because
patients with EMD or PCL are dif­fi­cult to treat, with poor out­
comes with the cur­rent ther­a­pies (3-year sur­vival rate is 35% for
EMD3 and median OS is 12 months for PCL4).
Cytogenetic abnor­mal­i­ties
The IMWG cur­ rently rec­ om­ mends the detec­ tion of t(4;14), Functional high-risk patients
t(14;16) and del (17/17p) in selected plasma cells by inter­phase
fluo­res­cent in situ hybrid­iza­tion for the iden­ti­fi­ca­tion of high-risk those patients with no appar­ent high risk at diag­no­sis but who
patients.5 In newly diag­nosed MM (NDMM), the pres­ence of at
least 1 high-risk cyto­ge­netic abnor­mal­ity (CA) is asso­ci­ated with
a median OS of 24.5 months, sig­nif­i­cantly shorter than the 50.5
months observed when there are not any CAs (P  <  .001). This is
far from com­plete and requires being updated.
The gain of the long arm of chro­mo­some 1 (+1q) is a fre­quent
CA observed in approx­i­ma­tely 30% of NDMMs and asso­ci­ated
with poor out­ come. A ret­ ro­ spec­
tive study conducted in 201
patients with NDMM treated with RVd Prakash
reported that Singh Shekhawat
patients har­ its man­age­ment is not easy either. So far, only a few clin­i­cal
tions? Patients with a “dou­ble-hit” biallelic inac­ti­va­tion of TP53
are at high risk, espe­cially if this abnor­mal­ity coex­ists with ISS
3 (1.5-year PFS of 33%) in ND patients.7 The Intergroupe Franco-
phone du Myélome (IFM) has recently reported in a large series
of patients with NDMM that the pres­ence of iso­lated del(17p) was
also asso­ci­ated with a poor out­come, although the poorest out­
come was reported for the dou­ble-hit patients.8
In sum­mary, the opti­mal iden­ti­fi­ca­tion of high-risk MM based
on CA is under con­struc­tion, but in clin­i­cal prac­tice, the CA rec-
ommended by the IMWG, together with abnor­mal­i­ties of chro­
mo­some 1 and muta­tional sta­tus of TP53, if pos­si­ble, would be
nec­es­sary to define the risk at base­line. At the moment of the
relapse, it would be opti­mal to repeat these eval­u­a­tions because
of the clonal evo­lu­tion and the poten­tial acqui­si­tion of new CAs
not detect­able at base­line.
R-ISS
The ISS risk model, includ­ ing albu­min and β2-microglobulin
lev­els, was improved with the incor­po­ra­tion of 2 well-known
dis­ease-related prog­nos­tic bio­mark­ers, CA and serum lac­tate
dehy­dro­ge­nase (LDH) lev­els, which are asso­ci­ated with a higher
pro­lif­er­a­tive activ­ity, resulting in the R-ISS model.
The R-ISS emerged from a large series of patients with NDMM,
and 3 sub­groups were defined: R-ISS I (n  =  871), includ­ing ISS
stage I, no high-risk CA, and nor­mal LDH level with a 5-year OS
rate of 82%; R-ISS III (n  =  295), includ­ing ISS stage III and high-
risk CA or high LDH level with a 5-year OS of 40%; and R-ISS II
(n  =  1894), includ­ing all­ the other pos­si­ble com­bi­na­tions and with
a 5-year OS of 62%.9 This stag­ing sys­tem is still valid, although
there are some lim­i­ta­tions: (1) most patients were assigned to
the R-ISS II, includ­ing those with high-risk CA but not ISS III, (2)
some other high-risk CAs such as +1q were not included, and (3)
other, more com­plex geno­mic abnor­mal­i­ties such as muta­tions
or inac­ti­va­tion of TP53 were not con­sid­ered.
In addi­tion to the above high-risk fea­tures, how do we rec­og­nize
prog­ress within the first 12 to 18 months after an opti­mal first line
of ther­apy? These patients are func­tional high risk with poor prog­
no­sis, and fur­ther inves­ti­ga­tions are required to unravel if there is
a clonal selec­tion or just an inad­e­quate eval­u­a­tion at diag­no­sis.
What is the opti­mal man­age­ment for patients
with high-risk fea­tures?
If the iden­ti­fi­ca­tion of high-risk patients with MM is chal­leng­ing,

Management of high-risk mye­loma  | 31

tri­als have been spe­cif­i­cally conducted in this pop­u­la­tion be-
cause the def­i­ ni­tion is het­ ero­
ge­
neous. In addi­ tion, high-risk
sub­groups in clin­i­cal tri­als are quite small to gen­er­ate solid rec­
om­men­da­tions.
In 2021, we know that poor prog­no­sis with the pres­ence of
high-risk fea­tures can be at least improved or even abro­gated by
the achiev­ing a deep and sustained response over time, which
can most likely be obtained through the use of novel ther­ap ­ eu­tic
options.10
At least 3 large meta-ana­ly­ses sup­port the use of MRD for
mon­i­tor­ing the response in MM because of its prog­nos­tic value.
The most recent one included pub­ li­
ca­
tions up to June 2019,
show­ing that the achieve­ment of unde­tect­able MRD improved
PFS (haz­ard ratio [HR], 0.33) and OS (HR, 0.45) in com­par­i­son
with the pres­ence of MRD. Moreover, its prog­nos­tic impact was
sustained across the dif­fer­ent sub­groups, includ­ing those with
some high-risk fea­ tures such as elderly patients with NDMM,
relapsed/refrac­tory patients, or even the pres­ence of high-risk
CA.11
Of note, the higher the sen­si­tiv­ity thresh­old for the MRD eval­
u­ a­
tion and the lon­ ger the sustained unde­ tect­able MRD over
time, the higher the prog­nos­tic value.
In addi­tion, the achieve­ment of unde­tect­able MRD can con­
vert risk assess­ment in MM into some­thing dynamic, and the high
risk at base­line can be over­come when unde­tect­able MRD is
achieved. In the PETHEMA/GEM2012MENOS65 trial, 458 patients
with NDMM had lon­gi­tu­di­nal assess­ment of MRD after 6 induc­
tion cycles with RVd, autol­o­gous trans­plan­ta­tion, and 2 con­sol­
i­da­tion courses with RVd. The 3-year PFS rate for patients with
R-ISS I, II, or III was com­pa­ra­ble (95%, 94%, and 88%) if MRD was
unde­tect­able after treat­ment. By con­trast, out­comes were pro­
gres­sively poor for patients with R-ISS I, II, and III when MRD was
detect­able, with a 3-year PFS of 62%, 53%, and 28%, respec­
tively, and anal­o­gous results were observed when OS was con­
sid­ered. Similarly, out­come of patients with high-risk CA was
abro­gated when unde­tect­able MRD was achieved.12
One addi­tional aspect needs to be incor­po­rated in the MRD
assess­ment: the MRD eval­u­a­tion out­side of the bone mar­row
through the use of func­tional imag­ing tools such as pos­i­tron
emis­sion tomog­ra­phy/com­put­er­ized tomog­ra­phy.13 Deauville
scores to focal lesions less than 4 and bone mar­ row uptake
show­ing the liver back­ground (Deauville score <4) have been
iden­ti­fied as the best cut­off to define pos­i­tron emis­sion tomog­
ra­phy/com­put­er­ized tomog­ra­phy neg­a­tiv­ity after ther­apy and
com­plete met­ab ­ olic response, as they have been described in
at least 2 clin­i­cal tri­als, the FORTE and CASSIOPETT substudy of
CASSIOPEIA trial.
Management of frail patients
The gen­eral approach described above is fea­si­ble for frail pa-
tients, but tol­er­ab
­ il­ity and qual­ity of life are cru­cial to deliver
treat­ments in order to reach depth responses. At least 1 clin­i­cal
trial has been conducted in unfit and frail patients with NDMM
according to the IMWG frailty index, using ixazomib and dara-
tumumab plus very low dose of dexa­meth­a­sone.14 Preliminary
results are encour­ag­ing, with 1-year OS rates of 96% and 74% for
unfit and frail patients, respec­tively. Subgroup anal­y­sis recently
conducted in the phase 3 tri­als ALCYONE and MAIA have also
shown how the addi­tion of daratumumab to either bortezomib,
mel­pha­lan, and pred­ni­sone (VMP) or Rd (lenalidomida and dexa-
32  |  Hematology 2021  |  ASH Education Program
methasone) has been a ­ ble to sig­nif­i­cantly improve the out­come
of frail patients com­pared with VMP or Rd alone, according to a
mod­i­fied IMWG frailty index (Table 2). In the relapsed-refrac­tory
set­ting, other subanalyses of phase 3 clin­i­cal tri­als have reported
how carfilzomib at dif­fer­ent doses and sched­ules or the com­
bi­na­tion of pomalidomide-dexa­meth­a­sone plus isatuximab is
fea­si­ble and ­able to improve the out­come of frail patients.15,16 Al-
though this infor­ma­tion is obtained from clin­i­cal tri­als, the good
tox­ic­ity pro­file of all­novel agents makes it pos­si­ble to main­tain
ther­apy in the frail pop­u­la­tion.
Management of patients with high-risk CA or R-ISS III
with approved drugs
Proteasome inhib­i­tors (PIs), immu­no­mod­u­la­tory drugs, and cor­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/30/1851766/30mateos.pdf by guest on 13 December 2021
ti­co­ste­roids are the key treat­ment ele­ments of patients with MM
patients with high-risk CA. For trans­plant-eli­gi­ble patients with
NDMM, the ques­ tion about bortezomib or carfilzomib as the
opti­mal PI for high-risk patients remains under debate because
the only phase 3 ran­dom­ized trial com­par­ing bortezomib with
carfilzomib did not show any dif­fer­ence, but it did only include
patients with t(4;14).17
The use of moAbs (monoclonal antibodies) targeting SLAMF7
and CD38 also has been eval­ua ­ ted in this set­ting. Although the
addi­ tion of elotuzumab showed no sig­ nif­i­
cant ben­ e­
fit when
com­bined with RVd in the phase 2 SWOG-1211 study,18 the addi­
tion of daratumumab has been shown to improve the out­come
of patients with NDMM and RRMM (relapsed refractory multiple
myeloma) with high-risk CA in a recent meta-anal­ y­sis.19 HDM-
ASCT con­tin­ues to be the stan­dard of care in high-risk NDMM
because of its capac­ity to achieve a higher unde­tect­able MRD
rate, and tan­dem trans­plant is even con­sid­ered for this pop­u­la­
tion based on the pos­i­tive data from the EMN02 trial, con­firmed
in the STAMINA trial at least in terms of PFS.20,21 However, tan­dem
trans­plant may not be nec­es­sary with the intro­duc­tion of moABs,
espe­cially with the intro­duc­tion of cell ther­apy. Maintenance with
lenalidomide is the stan­dard of care to improve the out­come of
high-risk patients com­pared with obser­va­tion, but it should be
improved through the addi­tion of PIs or moABs, with pre­lim­i­nary
pos­i­tive data.22,23
In the nontransplant-eli­gi­ble, high-risk patients with NDMM,
daratumumab, lenalidomide, and dexa­meth­a­sone would be the
first choice based on the results reported in the MAIA trial, with
a median PFS of 45.3 months com­pared with 29.6 months in the
Rd arm (HR, 0.57).24
In the RR (relapse or refractory) set­ting, the same approach
is valid, and the com­bi­na­tion of choice for patients with high-
risk CA would be those with the higher likely prob­ab ­ il­ity of
achiev­ing unde­tect­able MRD (Table 2).
Of note, the novel drug melflufen flufenamide has shown
prom­is­ing effi­cacy in 45 patients with EMD (extramedullary dis-
ease) included in the HORIZON trial, with an over­all response
of 24% vs 30% in patients with­out EMD.25 Selinexor also may
have a role in treating patients with del(17p) based on its
mech­an ­ ism of action and avail­­able evi­dence in some clin­i­cal
tri­als.26 Belantamb mafodotin, a BCMA-con­ju­gated moAb, pro­
duced a response rate of 33%, which is sim­i­lar to that reported
in patients with high-risk cyto­ge­net­ics.27 Further stud­ies are
required to con­firm this effi­cacy. Table 2 shows the effi­cacy
reported in patients with high-risk CA in the most rel­e­vant clin­
i­cal tri­als in patients with NDMM and RRMM.
Table 2. Clinical stud­ies ongo­ing in patients with high-risk fea­tures
Registration num­ber Study design
ND high-risk MM
NCT03104842 Isatuximab-KRd as induc­tion, con­sol­i­da­tion,
and main­te­nance
NCT03756896 Carfilzomib, pomalidomide, and
dexa­meth­a­sone as main­te­nance after
HDM-ASCT
NCT04025450 Chidamide (HDAC inhib­i­tor)–lenalidomide,
bortezomib, and dexa­meth­a­sone as
induc­tion
NCT02128230 Induction with mel­pha­lan 20–KTD-PACE
followed by mel­pha­lan 80–KTd-PACE plus
ASCT and KTD-PACE con­sol­i­da­tion and
KRd main­te­nance (1 year) and Kd (1 year)
NCT03549442 BCMA CAR-T + huCART19 in dif­fer­ent
sched­ules
NCT04196491 BCMA CAR-T bb2121 (ide-cel) (150-800  ×  106)
followed by lenalidomide as main­te­nance
NCT04436029 Autologous CD8+ T cells expressing an anti-
BCMA chi­me­ric anti­gen recep­tor
NCT04133636 BCMA CAR-T JNJ-68284528 (cilta-cel)
followed by lenalidomide main­te­nance
NCT04133636 BCMA CAR-T JNJ-68284528 (cilta-cel)
followed by lenalidomide and
daratumumab main­te­nance
Relapsed-refrac­tory high-risk MM
NCT03601078 BCMA CAR-T bb2121 (150-450  ×  106)
NCT04133636 BCMA CAR-T JNJ-68284528 (cilta-cel)
NCT03104270 Elotuzumab in com­bi­na­tion with
pomalidomide, carfilzomib, and
dexa­meth­a­sone
cilta-cel, ciltacabtagene autoleucel; CrCL, cre­at­i­nine clear­ance; GEP70, gene expression profiling-70; HDAC, histone deacetylase; ide-cel,
idecabtagene vicleucel; IMiD, immu­no­mod­u­la­tory drug; Kd, carfilzomib and dexa­meth­a­sone; KRd, carfilzomib, lenalidomide and dexa­meth­a­sone;
KTD-PACE, carfilzomib, tha­lid­o­mide, dexa­meth­a­sone, cis­platin, doxo­ru­bi­cin, cyclo­phos­pha­mide, and etoposide; PD, pro­gres­sion dis­ease; VGPR,
very good par­tial response.
Management of func­tional high-risk patients
Although no spe­cific tri­als were performed until very recent-
ly, new tri­als with BCMA-targeted CAR-T cells have focused
on this sub­group of patients (Table 3). Some sub­group anal­
y­sis in phase 3 tri­als focused on early relapses, defined as
those occur­ring within the first 12 to 18 months after the pre­
Prakash Singh Shekhawat
Population
Transplant eli­gi­ble or inel­i­gi­ble
del(17p) in ≥10% of puri­fied cells and/or t(4;14) and/or >3 cop­ies +1q21
ISS II or III
Transplant eli­gi­ble achiev­ing at least par­tial response
Presence of del(17p), t(4;14), t(14;16), t(14;20)
PCL at diag­no­sis
Transplant eli­gi­ble and inel­i­gi­ble
Presence of del(17p), t(4;14), t(14;16), t(14;20)
R-ISS III
IgD/IgE
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/30/1851766/30mateos.pdf by guest on 13 December 2021
Extramedullary plasmacytomas
Peripheral plasma cells by flow cytom­e­try ≥0.165%
Transplant eli­gi­ble
GEP70 risk score of ≥0.66
ISS III or R-ISS III or
Metaphase kar­yo­type with >3 abnor­mal­i­ties except hyperdiploidy
Failure to achieve par­tial response or bet­ter to ini­tial ther­apy based on
PI and IMiD
R-ISS III
High-risk patients who com­pleted pretransplant induc­tion
antimyeloma treat­ment
Less than com­plete response after first-line treat­ment and trans­plant
followed or not by con­sol­i­da­tion
Noneligible for trans­plant patients with ISS III
R-ISS III and
PD <18 months after the first-line treat­ment includ­ing induc­tion,
trans­plant, and lenalidomide main­te­nance
PD <18 months since date of start ini­tial ther­apy, which must con­tain PI,
IMiD, and dexa­meth­a­sone
Less thanVGPR after induc­tion, includ­ing PI, IMiD, and dexa­meth­a­sone
and trans­plant (between 70 and 110 days after trans­plant)
One prior line includ­ing PI, IMiD, and PD within the first 12 months after
trans­plant or the first-line treat­ment for nontransplant eli­gi­ble
More than 2 prior lines, includ­ing PI and IMiD and
del(17p), t(14;16), t(14:20)
PCL
Extramedullary dis­ease
Doubling in lev­els of MM mark­ers in the past 3 months
Refractoriness to their most recent lenalidomide-containing reg­i­men
and PI-based reg­i­men
Renal fail­ure with CrCl between 15 and 30  mL  ×  min­ute
vi­ous ther­apy, show­ing that the addi­tion of carfilzomib to Rd
or daratumumab to Rd vs Rd in the ASPIRE and POLLUX tri­
als resulted in a sig­nif­i­cant ben­e­fit for the tri­ple com­bi­na­tion
com­pared with Rd. A sim­i­lar effect has been recently reported
with the addi­tion of daratumumab to bortezomib or carfilzo-
mib28-31 (Table 2).
Management of high-risk mye­loma  | 33
 Table 3.  Efficacy of cur­rent treat­ment approaches for patients with NDMM with high-risk fea­tures (defined as del(17p), t(4;14), or t(14;16))

Transplant-eli­gi­ble NDMM
Clinical trial SWOG-1211 Cas­si­o­peia Forte Griffin
Population High risk* ITT High risk ITT High risk ITT High risk
Treatment EloVRd vs VRd DVTd vs VTD DVTd vs VTD KRd-T/KRd12 KRd-T/KRd12 DRVd vs RVd DRVd vs RVd
PFS (m)/HR 31 vs 34/0.96 NR/0.47 NR/0.67 NR/0.64 NR/0.51 NR/NA NR/NA
Transplant-inel­i­gi­ble NDMM
Clinical trial SWOG ALCYONE MAIA
Population ITT High risk ITT High risk Frail patients ITT High risk Frail patients
Treatment VRd- > Rd vs Rd VRd- > Rd vs Rd DVMP vs VMP DVMP vs VMP DVMP vs VMP DRd vs Rd DRd vs Rd DRd vs Rd
PFS (m)/HR 43 vs 30/0.74 38 vs 16/NA 36 vs 18/0.50 18 vs 18/0.78 33 vs 19/0.51 NR vs 34/0.54 45 vs 29/0.57 NR vs 30/0.62
Relapsed-refrac­tory MM
Clinical trial POLLUX ASPIRE ELOQUENT-2 TOURMALINE-MM1

34  |  Hematology 2021  |  ASH Education Program

Population ITT High risk Early relapse ITT High risk Early relapse ITT High risk ITT High risk
Treatment DRd vs Rd DRd vs Rd KRd vs Rd KRd vs Rd EloRd vs Rd EloRd vs Rd IRd vs Rd IRd vs Rd
PFS (m)/HR 44.5 vs 17.5/0.44 26.8 vs 8.3/0.37 0.38 26.3 vs 17.3/0.69 23 vs 13.9/0.7 21.4 vs 10.7/0.7 19.4 vs 14.9/0.70 NA/0.72(del17p) 20.6 vs 14.7/0.74 21.4 vs 9.7/0.54
NA/0.56 (t(4;14)
Clinical trial CASTOR ENDEAVOR CANDOR IKEMA
Population ITT High risk ITT High risk ITT High risk Early relapse ITT High risk
Treatment DVd vs Vd DVd vs Vd Kd vs Vd Kd vs Vd DKd vs Kd DKd vs Kd IsaKd vs Kd IsaKd vs Kd
PFS (m)/HR 16.7 vs 7.1/0.31 12.6 vs 6.2/0.41 18.7 vs 9.4/0.53 8.8 vs 6.0/0.7 28.6 vs 15.9/0.59 15.6 vs 5.6/0.49 CRrate 28 vs 3% NR vs 19.1/0.53 NA/0.72
Clinical trial OPTIMISMM BOSTON ICARIA ELOQUENT-3
Population ITT High risk ITT High risk ITT High risk ITT High risk
Treatment PVd vs Vd PVd vs Vd SVd vs Vd SVd vs Vd IsaPd vs Pd IsaPd vs Pd EloPd vs Pd EloPd vs Pd
PFS (m)/HR 11.2 vs 7.1/0.61 NA/0.56 11.2 vs 5.8/0.61 NA/0.67 11.5 vs 6.4/0.59 NA/0.66 10.3 vs 4.7/0.54 0.52
Clinical trial STORM HORIZON DREAMM-2 KARMMA-1
Population ITT High risk ITT High risk ITT High risk ITT High risk
Treatment Sd Sd Melflufen-dex Melflufen-dex Belamaf Belamaf Ide-cel Ide-cel
PFS (m)/HR 3.7 3.3* and 4.6* 4.2 3.0 3.9 2.1 8.8 10.4
*High-risk def­i­ni­tion: gene expres­sion pro­fil­ing high risk, t(14;16), t(14;20), del(17p), amp1q21, plasma cell leu­ke­mia, ele­vated serum LDH (2  ×  upper limit of nor­mal).
Belamaf, belantamaf mafodotin; dex, dexa­meth­a­sone; DKd, daratumumab, carfilzomib, and dexa­meth­a­sone; DRd, daratumumab, lenalidomide, and dexa­meth­a­sone; DRVd, daratumumab,
lenalidomide, bortezomib, and dexa­meth­a­sone; DVd, daratumumab, bortezomib, and dexa­meth­a­sone; DVMP, daratumumab, bortezomib, mel­pha­lan, and pred­ni­sone; DVTd, daratumumab,
bortezomib, tha­lid­o­mide, and dexa­meth­a­sone; Elo, elotuzumab; EloPd, elotuzumab, pomalidomide, and dexa­meth­a­sone; EloRd, elotuzumab, lenalidomide, and dexa­meth­a­sone; Ide-cel,
idecabtagene vicleucel; IRd, ixazomib, lenalidomide, and dexa­meth­a­sone; IsaKd, isatuximab, carfilzomib, and dexa­meth­a­sone; IsaPd, isatuximab, pomalidomide, and dexa­meth­a­sone; ITT,
inten­tion-to-treat pop­u­la­tion; Kd, carfilzomib and dexa­meth­a­sone; KRd, carfilzomib, lenalidomide, and dexa­meth­a­sone; KRd-T, carfilzomib, lenalidomide, and dexa­meth­a­sone followed by
trans­plant; KRd 12, KRd for 12 cycles; m, months; NA, not avail­­able; NR, not reached; PVd, pomalidomide, bortezomib, and dexa­meth­a­sone; Rd, lenalidomide and dexa­meth­a­sone; RVd,
lenalidomide, bortezomib, and dexa­meth­a­sone; Sd, selinexor and dexa­meth­a­sone; SVd, selinexor, bortezomib, and dexa­meth­a­sone; VTd, bortezomib, tha­lid­om ­ ide, and dexa­meth­a­sone.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/30/1851766/30mateos.pdf by guest on 13 December 2021

Management of high-risk patients with CAR-T cell ther­apy
BCMA is an attrac­tive and exten­sively stud­ied tar­get for immu­no­
ther­apy in MM. BCMA-targeting CAR-T cells have dem­on­strated
fast, high, and deep responses in patients with RRMM. The sam­
ple sizes across the dif­fer­ent tri­als are so far rather small but have
included a great pro­por­tion of patients with high-risk fea­tures,
such as high-risk CA, R-ISS III, EMD, or high tumor bur­den.32,33 Ide-
cel (idecabtagene vicleucel), indeed, already has been approved
by US Food and Drug Administration for RRMM after at least 4
rounds of ther­apy, includ­ing PIs, immu­no­mod­u­la­tory drugs, and
anti-CD38, and has been eval­u­ated in 128 patients with RRMM
after a median of 6 prior lines (84% tri­ple refrac­tory). The ORR
(overall response rate) was 73%, includ­ing a com­plete remis­
sion rate of 33% and a median PFS of 8.8 months. These effi­cacy
data were sustained in patients with high-risk fea­tures, such as
EMD (n  =  50), high-risk CA (n  =  45), or high tumor bur­den (n  =  65)34
(Table 2). Many other BCMA-targeted CAR-T cells are under
inves­ti­ga­tion, and some clin­i­cal tri­als are focused in patients with
high-risk fea­tures (Table 3). If results are pos­i­tive, CAR-T cell ther­
apy will rap­idly move as the first choice in patients with high-risk
fea­tures.
Beyond BCMA-targeted CAR-T cells, there are other ther­
a­peu­tic options, such as bispecific moABs targeting not only
BCMA but also GPRC5D, FcRH5, and oth­ers, under eval­u­a­tion in
patients with RRMM, and their effi­cacy will be also eval­u­ated in
patients with high-risk fea­tures.
In sum­mary, in 2021, the iden­ti­fi­ca­tion of high-risk patients
with MM con­tin­ues being an unmet med­i­cal need, and the def­
i­ni­tion should be revisited. Genomics will help us to improve
the iden­ti­fi­ca­tion of these patients, as well as the ther­a­peu­tic
advances, to find the best option for them. Considering the
achieve­ ment of unde­ tect­
able and sustained MRD can abro­
gate the poor prog­no­sis of high-risk fea­tures, the man­age­ment
of these patients should be response adapted and deter­mined
from expo­si­tion to sequen­tial treat­ments based on drugs with a
new and dif­fer­ent mech­a­nism of action. International effort and
research in this regard are required.
Conflict-of-inter­est dis­clo­sure
María-Victoria Mateos: has received hon­o­raria derived from lec­
tures and advi­sory boards from Janssen, BMS-Celgene, Amgen,
Takeda, Abbvie, Sanofi, Oncopeptides, Adaptive, Roche, Pfizer,
Regeneron, GSK, Bluebird Bio, and Sea-Gen.
Borja Puertas Martínez: no com­pet­ing finan­cial inter­ests to
declare.
Verónica González-Calle: has received hon­o­raria from Janssen
and Celgene, received research funding from Janssen, and pro­
vided con­sul­ting or an advi­sory role for Prothena and Janssen.
Off-label drug use
María-Victoria Mateos: nothing to disclose.
Borja Puertas Martínez: nothing to disclose.
Verónica González-Calle: nothing to disclose.
Correspondence
María-Victoria Mateos, Hospital Universitario de Salamanca, Insti-
tuto de Investigación Biomédica de Salamanca (IBSAL), Centro
de Investigación del Cancer (IBMCC-USAL, CSIC), CIBER-ONC
num­ber CB16/12/00233, Paseo San Vicente 58-182, 37007 Sala-
manca, Spain; e-mail: mvmateos@usal​­.es.
Prakash Singh Shekhawat
References
1. Palumbo A, Bringhen S, Mateos M-V, et al. Geriatric assess­ment pre­dicts sur­
vival and toxicities in elderly mye­loma patients: an International Myeloma
Working Group report. Blood. 2015;125(13):2068-2074.
2. Facon T, Dimopoulos MA, Meuleman N, et  al. A sim­ pli­
fied frailty scale
pre­dicts out­comes in trans­plant-inel­i­gi­ble patients with newly diag­
nosed mul­ti­ple mye­loma treated in the FIRST (MM-020) trial. Leukemia.
2020;34(1):224-233.
3. Usmani SZ, Heuck C, Mitchell A, et  al. Extramedullary dis­ease por­tends
poor prog­no­sis in mul­ti­ple mye­loma and is over-represented in high-risk
dis­ease even in the era of novel agents. Haematologica. 2012;97(11):1761–
1767.
4. Gonsalves WI, Rajkumar SV, Go RS, et  al. Trends in sur­vival of patients
with pri­mary plasma cell leu­ke­mia: a pop­u­la­tion-based anal­y­sis. Blood.
2014;124(6):907-912.
5. Sonneveld P, Avet-Loiseau H, Lonial S, et al. Treatment of mul­ti­ple mye­loma
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/30/1851766/30mateos.pdf by guest on 13 December 2021
with high-risk cyto­ ge­net­
ics: a con­sen­sus of the International Myeloma
Working Group. Blood. 2016;127(24):2955-2962.
6. Schmidt TM, Barwick BG, Joseph N, et  al. Gain of chro­ mo­
some 1q is
asso­ci­ated with early pro­gres­sion in mul­ti­ple mye­loma patients treated
with lenalidomide, bortezomib, and dexa­ meth­a­
sone. Blood Cancer J.
2019;9(12):94.
7. Corre J, Munshi NC, Avet-Loiseau H. Risk fac­tors in mul­ti­ple mye­loma: is it
time for a revi­sion? Blood. 2021;137(1):16-19.
8. Corre J, Perrot A, Caillot D, et al. del(17p) with­out TP53 muta­tion con­fers a
poor prog­no­sis in inten­sively treated newly diag­nosed patients with mul­
ti­ple mye­loma. Blood. 2021;137(9):1192-1195.
9. Palumbo A, Avet-Loiseau H, Oliva S, et al. Revised inter­na­tional stag­ing sys­
tem for mul­ti­ple mye­loma: a report from International Myeloma Working
Group. J Clin Oncol. 2015;33(26):2863-2869.
10. Burgos L, Puig N, Cedena M-T, et al. Measurable resid­ual dis­ease in mul­ti­ple
mye­loma: ready for clin­i­cal prac­tice? J Hematol Oncol. 2020;13(1):82.
11. Munshi NC, Avet-Loiseau H, Anderson KC, et  al. A large meta-anal­ y­sis
establishes the role of MRD neg­a­tiv­ity in long-term sur­vival out­comes in
patients with mul­ti­ple mye­loma. Blood Adv. 2020;4(23):5988-5999.
12. Paiva B, Puig N, Cedena M-T, et al; GEM (Grupo Español de Mieloma)/
PETHEMA (Programa Para el Estudio de la Terapéutica en Hemopatías
Malignas) Cooperative Study Group. Measurable resid­ual dis­ease by next-
gen­er­a­tion flow cytom­e­try in mul­ti­ple mye­loma. J Clin Oncol. 2020;38(8):
­784-792.
13. Zamagni E, Nanni C, Dozza L, et  al. Standardization of 18F-FDG-PET/CT
according to Deauville cri­te­ria for met­a­bolic com­plete response def­i­ni­tion
in newly diag­nosed mul­ti­ple mye­loma. J Clin Oncol. 2021;39(2):116-125.
14. Stege CAM, Nasserinejad K, van der Spek E, et al. Efficacy and tol­er­a­bil­ity
of ixazomib, daratumumab and low dose dexa­meth­a­sone (Ixa Dara dex) in
unfit and frail newly diag­nosed mul­ti­ple mye­loma (NDMM) patients; results
of the interim effi­cacy anal­y­sis of the phase II HOVON 143 study. Blood.
2019;134(suppl 1):695-695.
15. Facon T, Niesvizky R, Mateos M-V, et  al. Efficacy and safety of carfilzo-
mib-based reg­ i­
mens in frail patients with relapsed and/or refrac­ tory
­mul­ti­ple mye­loma. Blood Adv. 2020;4(21):5449-5459.
16. Schjesvold FH, Richardson PG, Attal M, et  al. Efficacy of isatuximab with
pomalidomide and dexa­meth­a­sone in elderly patients with relapsed/refrac­
tory mul­ti­ple mye­loma: ICARIA-MM sub­group anal­y­sis. Blood. 2019;134
(suppl 1):1893-1893.
17. Kumar SK, Jaco­bus SJ, Cohen AD, et al. Carfilzomib or bortezomib in com­
bi­na­tion with lenalidomide and dexa­meth­a­sone for patients with newly
diag­nosed mul­ti­ple mye­loma with­out inten­tion for imme­di­ate autol­o­gous
stem-cell trans­plan­ta­tion (ENDURANCE): a multicentre, open-label, phase
3, randomised, con­trolled trial. Lancet Oncol. 2020;21(10):1317-1330.
18. Usmani SZ, Hoering A, Ailawadhi S, et al; SWOG1211 Trial Investigators. Bor-
tezomib, lenalidomide, and dexa­meth­a­sone with or with­out elotuzumab in
patients with untreated, high-risk mul­ti­ple mye­loma (SWOG-1211): pri­mary
anal­y­sis of a randomised, phase 2 trial. Lancet Haematol. 2021;8(1):e45–
e54.
19. Giri S, Grimshaw A, Bal S, et al. Evaluation of daratumumab for the treat­
ment of mul­ti­ple mye­loma in patients with high-risk cyto­ge­netic fac­tors: a
sys­tem­atic review and meta-anal­y­sis. JAMA Oncol. 2020;6(11):1759-1765.
20. Cavo M, Gay F, Beksac M, et  al. Autologous haematopoietic stem-cell
trans­plan­ta­tion ver­sus bortezomib-mel­pha­lan-pred­ni­sone, with or with­
out bortezomib-lenalidomide-dexa­meth­a­sone con­sol­i­da­tion ther­apy, and
lenalidomide main­te­nance for newly diag­nosed mul­ti­ple mye­loma
Management of high-risk mye­loma  | 35
 (EMN02/​HO95): a multicentre, randomised, open-label, phase 3 study.
Lancet Haematol. 2020;7(6):e456-e468.
21. Hari P, Pasquini MC, Stadtmauer EA, et  al. Long-term fol­low-up of BMT
CTN 0702 (STaMINA) of postautologous hema­to­poi­etic cell trans­plan­ta­tion
(autoHCT) strat­eg ­ ies in the upfront treat­ment of mul­ti­ple mye­loma (MM).
J Clin Oncol. 2020;38(15, suppl):8506.
22. Voorhees PM, Kaufman JL, Laubach J, et al. Daratumumab, lenalidomide,
bortezomib, and dexa­meth­a­sone for trans­plant-eli­gi­ble newly diag­nosed
mul­ti­ple mye­loma: the GRIFFIN trial. Blood. 2020;136(8):936-945.
23. Gay F, Musto P, Rota Scalabrini D, et al. Survival anal­y­sis of newly diag­nosed
trans­plant-eli­gi­ble mul­ti­ple mye­loma patients in the ran­dom­ized Forte trial.
Blood. 2020;136(suppl 1):35-37.
24. Facon T, Kumar S, Plesner T, et al. MAIA Trial Investigators. Daratumumab
plus lenalidomide and dexa­ meth­ as­one for untreated mye­ loma. N Engl
J Med. 2019;380(22):2104-2115.
25. Richardson PG, Oriol A, Larocca A, et al. HORIZON (OP-106) Investigators.
Melflufen and dexa­meth­a­sone in heavily pretreated relapsed and refrac­
tory mul­ti­ple mye­loma. J Clin Oncol. 2021;39(7):757-767.
26. Chari A, Vogl DT, Gavriatopoulou M, et al. Oral selinexor-dexa­meth­a­sone
for tri­ple-class refrac­tory mul­ti­ple mye­loma. N Engl J Med. 2019;381(8):727–
738.
27. Cohen AD, Trudel S, Lonial S, et al. DREAMM-2: sin­gle-agent belantamab
mafodotin (GSK2857916) in patients with relapsed/refrac­ tory mul­ti­
ple
mye­loma (RRMM) and high-risk (HR) cyto­ge­net­ics. J Clin Oncol. 2020;38(15,
suppl):8541.
28. Mateos M-V, Goldschmidt H, San-Miguel J, et al. Carfilzomib in relapsed or
refrac­tory mul­ti­ple mye­loma patients with early or late relapse fol­low­ing
prior ther­apy: a sub­group anal­y­sis of the ran­dom­ized phase 3 ASPIRE and
ENDEAVOR tri­als. Hematol Oncol. 2018;36(2):463-470.
36  |  Hematology 2021  |  ASH Education Program
29. Dimopoulos MA, San-Miguel J, Belch A, et  al. Daratumumab plus lenalid-
omide and dexa­meth­a­sone ver­sus lenalidomide and dexa­meth­a­sone in
relapsed or refrac­ tory mul­
ti­
ple mye­ loma: updated anal­y­sis of POLLUX.
Haematologica. 2018;103(12):2088-2096.
30. Spencer A, Lentzsch S, Weisel K, et al. Daratumumab plus bortezomib and
dexa­meth­a­sone ver­sus bortezomib and dexa­meth­a­sone in relapsed or
refrac­tory mul­ti­ple mye­loma: updated anal­y­sis of CASTOR. Haematolog-
ica. 2018;103(12):2079-2087.
31. Weisel K, Geils GF, Karlin L, et al. Carfilzomib, dexa­meth­a­sone, and dara-
tumumab ver­sus carfilzomib and dexa­meth­a­sone in relapsed or refrac­
tory mul­ti­ple mye­loma: sub­group anal­y­sis of the phase 3 can­dor study in
patients with early or late relapse. Blood. 2020;136(suppl 1):37-38.
32. Munshi NC, Anderson LD Jr, Shah N, et  al. Idecabtagene vicleucel in
relapsed and refrac­tory mul­ti­ple mye­loma. N Engl J Med. 2021;384(8):705–
716.
33. Madduri D, Berdeja JG, Usmani SZ, et al. CARTITUDE-1: phase 1b/2 study of
ciltacabtagene autoleucel, a B-cell mat­u­ra­tion anti­gen-directed chi­me­ric
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/30/1851766/30mateos.pdf by guest on 13 December 2021
anti­gen recep­tor T cell ther­apy, in relapsed/refrac­tory mul­ti­ple mye­loma.
Blood. 2020;136(suppl 1):22-25.
34. Raje NS, Siegel DS, Jagannath S, et  al. Idecabtagene vicleucel (ide-cel,
bb2121) in relapsed and refrac­tory mul­ti­ple mye­loma: ana­ly­ses of high-risk
sub­groups in the KarMMa study. Blood. 2020;136(suppl 1):37-38.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000229
 CHALLENGES IN MULTIPLE MYELOMA TREATMENT

Minimal residual disease in multiple

myeloma: why, when, where

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/37/1851756/37yee.pdf by guest on 13 December 2021

Andrew J. Yee1,2 and Noopur Raje1,2
Center for Multiple Myeloma, Massachusetts General Hospital Cancer Center, Boston, MA; and 2Harvard Medical School, Boston, MA
1

Improvements in multiple myeloma therapy have led to deeper responses that are beyond the limit of detection by his-
torical immunohistochemistry and conventional flow cytometry in bone marrow samples. In parallel, more sensitive tech-
niques for assessing minimal residual disease (MRD) through next-generation flow cytometry and sequencing have been
developed and are now routinely available. Deep responses when measured by these assays correspond with improved
outcomes and survival. We review the data supporting MRD testing as well as its limitations and how it may fit in with
current and future clinical practice.

LEARNING OBJECTIVES
• Understand the role of MRD status in prognosis in multiple myeloma
• Explore how to apply MRD testing in clinical practice and its limitations

(daratumumab and isatuximab). With the increasing use of

CLINICAL CASE
A 57-year-old woman with multiple myeloma (MM) has
completed initial therapy with lenalidomide, bortezomib,
and dexamethasone, followed by high-dose melphalan and
autologous stem cell transplant (auto-SCT). She initially pre-
sented with anemia with hemoglobin of 6.1g/dL, and the
disease was staged as Revised International Staging System
stage II. Fluorescence in situ hybridization (FISH) showed
gain of 1q. She is on maintenance therapy with lenalidomide
and ixazomib. Laboratory studies showed no monoclonal
protein, and the serum free light chain ratio was normal. She
mentions that she went to a patient education session and
heard about “minimal residual disease (MRD) testing.” She is
interested in having MRD testing performed.
these agents, especially in 3- and 4-drug combinations,
responses have substantially deepened in newly diagnosed
patients in whom complete responses (CRs) are routinely
achieved, for example, from historically 10% with thalido-
mide and dexamethasone1 to 95% in a recently reported
combination of daratumumab, carfilzomib, lenalidomide,
and dexamethasone (dara-KRd) without the use of high-
dose melphalan.2 Similar trends are also seen in relapsed
disease, especially with the advent of anti-B-cell maturation
antigen (BCMA) directed therapies. To better assess these
improving responses, MRD testing using more sensitive
tools has emerged. This review provides an overview of
MRD assessment in MM and highlights the practical aspects
of MRD testing.

Importance of depth of response

Introduction
Response assessment in MM has traditionally relied on
measuring the monoclonal protein in the serum and urine
and plasma cell involvement in the bone marrow and,
more recently, serum free light chains. The past 2 decades
have seen tremendous progress in the treatment of MM
with the approval and adoption of effective agents such
as proteasome inhibitors (bortezomib, carfilzomib, ixazo-
mib), immunomodulatory drugs (lenalidomide, pomalido-
mide), and, more recently, anti-CD38 monoclonal antibodies
Prakash Singh Shekhawat
Intuitively, the depth of response with myeloma therapy cor-
relates with long-term outcomes. The relationship between
CR and progression-free survival (PFS) and overall survival
(OS) has been consistently demonstrated in a meta-analysis
of trials using intensive therapy combined with older3 or
contemporary therapies.4 Moreover, an analysis of 344
patients treated by the Grupo Español de Mieloma (GEM)
and Programa Para el Estudio de la Terapéutica en Hemo-
patías Malignas (PETHEMA) groups noted differences in sur-
vival between CR, near CR, and very good partial response.5
Minimal residual disease in multiple myeloma | 37
Table 1. Comparison of MRD test­ing with next-gen­er­a­tion flow cytom­e­try and next-gen­er­a­tion sequenc­ing
Characteristic Next-gen­er­a­tion flow cytom­e­try
Requires base­line sam­ple No
Sensitivity <10 −5
Sample require­ments >5 mil­lion cells (more cells to increase sen­si­tiv­ity to 10−6)
Sample processing Requires fresh sam­ple
Assessment for hemo­di­lu­tion Yes
Turnaround time 2-4 days
Commercial avail­abil­ity Reference or expe­ri­enced in-house lab­o­ra­tory
Table adapted from Kumar et al. 9
Similar obser­va­tions with CR hold true in older patients who are
not eli­gi­ble for high-dose ther­apy.6 This raises the ques­tion of
whether fur­ther gains may be seen with even deeper responses
such as MRD-neg­a­tive dis­ease. It should be noted that this rela­
tion­ship between response and out­comes gen­er­ally holds true,
pro­vided that the method of achiev­ing this depth of response
is tol­er­ated well. However, the Eastern Cooperative Oncology
Group study of high-dose dexa­meth­a­sone with lenalidomide7 and
the BELLINI trial of venetoclax with bortezomib and dexa­meth­a­
sone8 are instruc­tive for illus­trat­ing that deeper responses are not
always asso­ci­ated with improve­ments in sur­vival.
The term min­i­mal resid­ual dis­ease con­ven­tion­ally refers to dis­
ease in the bone mar­row space. Measurement of min­i­mal dis­ease in
the bone mar­row is rel­e­vant as it com­monly serves as the res­er­voir
of dis­ease relapse. Methods cur­rently used to detect MRD include
mul­ti­pa­ram­e­ter, next-gen­er­a­tion flow cytom­e­try (NGF) and next-
gen­er­a­tion sequenc­ing (NGS) (Table 1). More than 10 years ago,
flow cytom­e­try was the first tech­nique to eval­u­ate MRD, in which
sen­si­tiv­ity was 10−4 ini­tially.10,11 The sen­ si­
tiv­
ity has improved to
2 × 10−6, and the EuroFlow con­sor­tium has stan­dard­ized the meth­
od­ol­ogy.12 NGS has also emerged in par­al­lel for mea­sur­ing MRD, in
which immu­no­glob­u­lin gene seg­ments are ampli­fied using con­
sen­sus prim­ers and sequenced.13 Currently, the sen­si­tiv­ity of the
Adaptive Clonoseq plat­form (pre­vi­ously known as LymphoSIGHT)
is 6.77 × 10−7 with 20 µg DNA from 1 mL of bone mar­row aspi­rate.14,15
The con­cor­dance between NGF and NGS is high. It exceeded
80% when exam­ined in the FORTE16 and CASSIOPEIA17 tri­als in
newly diag­nosed patients. There was sim­i­larly high con­cor­dance,
85.8%18 and 92.9%,19 when com­par­ing NGF with NGS from a dif­
fer­ent plat­form, LymphoTrack. The choice of assay used for MRD
is based on avail­abil­ity and insti­tu­tional pref­er­ence. NGS by the
Clonoseq assay is com­mer­cially avail­­able through Adaptive, and
in Jan­ua ­ ry 2019, Medi­care announced cov­er­age of this test. NGF
is also com­mer­cially avail­­able, for exam­ple, through Mayo Clinic
ref­er­ence lab­o­ra­tory. A con­sid­er­ation with NGS is that it requires
a base­line sam­ple to pro­vide a trackable sequence; NGF does not
require a base­line sam­ple. In 1 series, a trackable sequence for
NGS could not be iden­ti­fied in 7.8% of sam­ples.20 NGF also has the
advan­tage of assessing for hemo­di­lu­tion by looking for mast cell,
eryth­ro­blast, and B-cell pre­cur­sor pop­u­la­tions.12 Finally, from a
research per­spec­tive, NGF may be a ­ ble to eval­u­ate the bone mar­
row micro­en­vi­ron­ment, which may have prog­nos­tic rel­e­vance.
Several meta-ana­ly­ses have con­sis­tently shown that depth
of response beyond CR cor­re­lates with improve­ment in OS.21-23
Although the ini­tial meta-ana­ly­ses focused on trans­plant-eli­gi­ble
38  |  Hematology 2021  |  ASH Education Program
Next-gen­er­a­tion sequenc­ing
Yes, in order to iden­tify clone to track
<10−6
<2 mil­lion cells
For Clonoseq track­ing: fresh or fro­zen sam­ple
Not pos­si­ble
7 days
Clonoseq (Adaptive); LymphoTrack for in-house use
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/37/1851756/37yee.pdf by guest on 13 December 2021
patients man­aged with inten­sive ther­apy and where MRD was
mostly assessed by older, less sen­si­tive, flow cytom­et­ry (10−4),
the recent meta-anal­ y­
sis extends on prior obser­ va­tions to
include older, trans­ plant-inel­
i­
gi­
ble patients and patients with
relapsed dis­ease.23 Compared with MRD-pos­i­tive dis­ease, MRD-
neg­a­tive sta­tus showed improved PFS (haz­ard ratio, 0.33; 95% CI,
0.29-0.37) and OS (haz­ard ratio, 0.45; 95% CI, 0.39-0.51) across
mul­ti­ple patient pop­u­la­tions, includ­ing in relapsed dis­ease and
high-risk dis­ease.23 Importantly, MRD sta­tus can strat­ify patients
in CR, where OS was 112 vs 82 months for MRD-negative vs
MRD-positive patients, respec­tively.22
Given these find­ings, MRD sta­tus is increas­ingly used as an
end point when com­par­ing dif­fer­ent reg­i­mens, espe­cially now
that reg­im ­ ens are increas­ingly achiev­ing deeper responses. The
International Myeloma Working Group9 and Bone Marrow Trans-
plant Clinical Trials Network (BMT CTN)24 have pro­vided guid­ance
around def­i­ni­tions of MRD and per­for­mance (Table 2), with the
IMWG recommending a sen­si­tiv­ity of 10−5. The use of MRD as a
sur­ro­gate end point for reg­u­la­tory pur­poses is an area of active
dis­cus­sion25 and is being addressed by a con­sor­tium of aca­demic
groups and phar­ma­ceu­ti­cal part­ners, the International Indepen-
dent Team for Endpoint Approval of Myeloma MRD.26-28
The depth of MRD neg­a­tive sta­tus is also impor­tant. This was
ini­tially shown with flow cytom­et­ ry with sen­si­tiv­ity down to 10−4
and where each log reduc­tion in MRD trans­lated into improve­
ment in median OS.29 In the Francophone du Myélome 2009
study of upfront vs deferred auto-SCT after ini­tial ther­apy with
lenalidomide, bortezomib, and dexa­meth­a­sone, MRD sta­tus was
assessed by flow cytom­et­ ry in all­patients, and a sub­set of these
patients was eval­ua ­ ted by more sen­si­tive NGS.30,31 Patients who
were a ­ ble to achieve MRD neg­a­tive sta­tus at 10−6 by NGS, which
is deeper than the recommended IMWG thresh­old of 10−5, had
supe­rior out­comes in PFS and OS com­pared with MRD-pos­i­tive
sta­tus (Figure 1).31 Moreover, the study showed dif­fer­ences in
out­comes between 10−6, 10−5, and 10−4. Prior to starting main­
te­nance ther­apy, patients who were MRD neg­a­tive had sim­i­lar
PFS whether they received trans­plant upfront or not, although
patients in the trans­plant arm were more likely to be MRD neg­a­
tive (29.8% vs 20.5%). Of note is that in the IFM 2009 study, MRD
assess­ments were after com­ple­tion of ini­tial ther­apy, prior to
main­te­nance ther­apy; the effect of high-dose ther­apy in patients
who were already MRD neg­at­ ive prior to high-dose ther­apy was
not addressed. Nevertheless, as long as a deep, MRD-neg­at­ive
response is achieved, the method of achiev­ing the response may
not be as impor­tant. For exam­ple, the CASSIOPEIA study eval­u­ated
Table 2. IMWG MRD cri­te­ria

Response
Sustained MRD neg­a­tive
Flow MRD neg­a­tive
Sequencing MRD neg­a­tive
Imaging plus MRD neg­a­tive
Criteria
MRD neg­a­tiv­ity in the mar­row (NGF or NGS, or both) and by imag­ing (PET CT), con­firmed min­i­mum of 1 year apart.
Subsequent eval­u­a­tions can be used to fur­ther spec­ify the dura­tion of neg­a­tiv­ity (eg, MRD neg­a­tive at 5 years).
Absence of phe­no­typ­i­cally aber­rant clonal plasma cells by NGF on bone mar­row aspi­rates using the EuroFlow stan­
dard oper­a­tion pro­ce­dure for MRD detec­tion in mul­ti­ple mye­loma (or val­i­dated equiv­a­lent method) with a min­i­mum
sen­si­tiv­ity of 1 in 105 nucle­ated cells or higher.
Absence of clonal plasma cells by NGS on bone mar­row aspi­rate in which the pres­ence of a clone is defined as fewer
than 2 iden­ti­cal sequenc­ing reads obtained after DNA sequenc­ing of bone mar­row aspi­rates using the LymphoSIGHT
plat­form (or val­i­dated equiv­a­lent method) with a min­i­mum sen­si­tiv­ity of 1 in 105 nucle­ated cells or higher.
MRD neg­a­tiv­ity as defined by NGF or NGS plus dis­ap­pear­ance of every area of increased tracer uptake found at base­
line or a pre­ced­ing PET CT or decrease to less than medi­as­ti­nal blood pool SUV or decrease to less than that of
surrounding nor­mal tis­sue.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/37/1851756/37yee.pdf by guest on 13 December 2021

Table adapted from Kumar et al.9
SUV, standard uptake value.

daratumumab, bortezomib, tha­lid­o­mide, and dexa­meth­a­sone
(dara-VTd) vs VTd in newly diag­nosed patients under­go­ing high-
dose mel­pha­lan and auto-SCT. Patients who achieved both CR
and MRD-neg­a­tive sta­tus had sim­i­lar PFS, irrespective of treat­
ment arm (although higher-qual­ity responses were more com­
mon in the dara-VTd arm).32 Similar find­ings were observed in the
FORTE study, in which out­comes of patients with MRD-neg­a­tive
dis­ease sustained for 1 year were sim­i­lar, irrespective of the ini­tial
treat­ment KRd vs 12 cycles of KRd with­out auto-SCT vs carfilzo-
mib, cyclo­phos­pha­mide, and dexa­meth­a­sone with auto-SCT.33
In patients with high-risk dis­ease, achiev­ing an MRD-neg­a­tive
response may be even more impor­tant. An anal­y­sis of the PETHEMA/
GEM2012MENOS65 trial showed that MRD-neg­ a­
tive responses
were ­able to over­come poor prog­nos­tic fea­tures at diag­no­sis,
includ­ing Revised International Staging System stage III.34,35 Similar
obser­va­tions were seen for patients with high-risk cyto­ge­net­ics
in IFM 2009 and ear­lier PETHEMA/GEM tri­als.31,36 The find­ings with
MRD extend on pre­vi­ous obser­va­tions where achiev­ing CR was
espe­cially impor­tant in high-risk dis­ease defined by gene expres­
sion pro­fil­ing.37
It is well established with tra­di­tional response cri­te­ria that
dura­bil­ity of response is a pow­er­ful prog­nos­tic fac­tor38,39 and that
loss of CR is asso­ci­ated with infe­rior sur­vival.40 Durability of MRD-
neg­a­tive sta­tus is sim­i­larly impor­tant. This was dem­on­strated
recently in the POLLUX and CASTOR stud­ies, which eval­u­ated
daratumumab with lenalidomide and dexa­meth­a­sone (dara-Rd)
or daratumumab with bortezomib and dexa­meth­a­sone, respec­
tively, using NGS at 10−5 sen­si­tiv­ity.41 Patients with sustained MRD
neg­a­tiv­ity over 12 months had the best out­comes, irrespective of
the treat­ment arm, although this was more likely to be achieved
in the daratumumab-containing com­bi­na­tion. Similar find­ings of
improved out­comes were seen with sustained MRD neg­a­tiv­ity
over 6 or 12 months in newly diag­nosed, trans­plant-inel­i­gi­ble
patients in the ALCYONE (daratumumab, bortezomib, mel­pha­
lan, and pred­ni­sone [dara-VMP] vs VMP) and MAIA (dara-Rd vs
Rd) tri­als.42 Reflecting these obser­va­tions, the IMWG defi­nes

Figure 1. Progression-free survival according to MRD level at the start of maintenance in IFM 2009. Progression-free survival
improves with each log reduction in MRD in IFM 2009 in patients who achieved at least a very good partial response. Figure adapted
from Perrot et al.31 Prakash Singh Shekhawat
Minimal resid­ual dis­ease in mul­ti­ple mye­loma  |  39
a sep­a­rate response cat­e­gory of “sustained MRD neg­a­tive,” in
which assess­ments by mar­row and by imag­ing are con­firmed at
least 1 year apart.9 If 1 year is bet­ter, 2 years may be even bet­ter:
this was dem­on­strated in patients with sustained MRD neg­a­tiv­ity
(by NGF at 10−5) for 2 years in a trial of patients on lenalidomide
main­te­nance.43 Moreover, in this study, loss of MRD neg­at­iv­ity
was actu­ally worse than sustained MRD pos­i­tiv­ity.
There have been sev­eral stud­ies exam­in­ing the pat­terns of
loss of MRD neg­a­tiv­ity and its clin­i­cal rel­e­vance.44-46 For exam­
ple, MRD pro­gres­sion by flow cytom­e­try with sen­si­tiv­ity at 10−4
or by allele-spe­cific oli­go­nu­cle­o­tide poly­mer­ase chain reac­tion
with sen­si­tiv­ity of 10−5 in a series of patients on lenalidomide
main­te­nance antic­i­pated bio­chem­i­cal relapse by 4 months and
clin­i­cal relapse by 9 to 10 months.45 Similarly, in a ret­ro­spec­tive
study using NGS (10−6), molec­u­lar relapse by MRD eval­u­a­tion was
­able to pre­dict clin­i­cal relapse.46 Serial MRD test­ing was a ­ ble to
pre­dict clin­i­cal relapse in 9 of 10 cases, and relapse by IMWG
cri­te­ria occurred at a median of 13 months (range, 1-28 months)
fol­low­ing molec­u­lar relapse. These find­ings raise the ques­tion
of whether ini­ti­at­ing treat­ment at the time of molec­u­lar relapse
rather than waiting for bio­chem­i­cal or clin­ic ­ al relapse could alter
the nat­u­ral his­tory of the dis­ease (see Relapse from MRD Nega-
tivity as Indication for Treatment study below).
Limitations of MRD assess­ment
An inher­ent lim­i­ta­tion in MRD assess­ment is its reli­ance on mea­
sur­ing dis­ease in the bone mar­row. This assess­ment focuses on
plasma cells and does not take into account the bone mar­row
micro­en­vi­ron­ment,47 which may play a role in shap­ing prog­no­
sis. From a prac­ti­cal per­spec­tive, bone mar­row involve­ment may
not be uni­form, such as in the case of macrofocal dis­ease,48 and
per­haps most impor­tant, extramedullary dis­ease may also be
pres­ent. For exam­ple, in the IMAgerie du JEune Myélome study
of the Intergroupe Francophone du Myélome 2009 trial, 26% of
patients with MRD-neg­a­tive dis­ease by flow cytom­e­try (sen­si­tiv­
ity 10−4) had pos­i­tive pos­i­tron emis­sion tomog­ra­phy (PET) com­
puted tomog­ra­phy (CT) find­ings.49 Similar find­ings were seen
in the CASSIOPET substudy of CASSIOPEIA, in which 10.5% of
patients who were neg­a­tive by NGF at 10−5 had pos­it­ ive PET CT.50
This dis­crep­ancy is rel­e­vant, as patients who were MRD neg­a­
tive but PET CT pos­i­tive had sim­i­lar out­comes to patients who
were MRD pos­i­tive. The dis­crep­ancy between MRD neg­a­tiv­ity
and imag­ing is higher in patients with relapsed dis­ease, 50%
vs 12% in newly diag­nosed patients in 1 series.51 Overall, as was
seen in CASSIOPET, patients who are “dou­ble neg­a­tive” on MRD
and imag­ing tended to have the best out­comes, suggesting that
these 2 modal­i­ties com­ple­ment each other. The Deauville scale
used in lym­phoma has been applied to MM to stan­dard­ize “met­
a­bolic response” cri­te­ria by PET and was an inde­pen­dent pre­dic­
tor for improved PFS and OS out­comes.52
Moreover, inher­ent to this dis­cus­sion is the het­ero­ge­ne­ity of
MM, in which the depth of response may not be as impor­tant in
all­ patients. This was pre­vi­ously rec­og­nized with CR, in which
patients with a his­tory of monoclonal gammopathy of undeter-
mined significance53 or with a MGUS-like gene expres­sion pro­file54
had lower CR rates with a tan­dem trans­plant reg­i­men in Total
Therapy 2 but supe­rior out­comes. The cyclin D2 molec­u­lar sub­
type, which char­ac­ter­is­ti­cally includes patients with t(11;14), has
the low­est and slowest cumu­la­tive inci­dence of response, yet has
com­pa­ra­ble out­comes with MRD-pos­i­tive dis­ease com­pared with
40  |  Hematology 2021  |  ASH Education Program
other patients with MRD-neg­a­tive dis­ease.55,56 The chal­lenge at
this time is how to pro­spec­tively iden­tify these patients in whom
an MRD-neg­a­tive response is not as crit­i­cal.
Perhaps the most obvi­ous lim­i­ta­tion for MRD assess­ment is
the require­ment for a bone mar­row aspi­ra­tion pro­ce­dure. This
has moti­vated inves­ti­gat­ing “liq­uid biop­sies,” using the same
tools on the periph­ eral blood. Indeed, anal­ y­
sis of periph­eral
blood pro­vi­des a sys­temic assess­ment and avoids the pit­falls
of het­ero­ge­ne­ity in bone mar­row sam­pling. Methods involv­ing
the periph­eral blood may allow for detecting and mon­i­tor­ing
extramedullary dis­ease that is missed by focus­ing on the bone
mar­row. Using the same NGF meth­od­ol­ogy opti­mized in bone
mar­row on periph­eral blood, the sen­si­tiv­ity is less.57 Forty per­
cent of patients with bone mar­row that was MRD pos­i­tive were
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/37/1851756/37yee.pdf by guest on 13 December 2021
neg­a­tive in the periph­eral blood; all­ patients with cir­cu­lat­ing
plasma cells were MRD pos­i­tive in the bone mar­row. Similarly,
NGS has been explored on periph­eral blood.58 Of patients with
pos­i­tive bone mar­row MRD tests, the test was neg­a­tive in plasma
69% of the time. This may reflect the lower cir­cu­lat­ing DNA bur­
den in periph­eral blood. Other approaches under devel­op­ment
include anal­y­sis of cir­cu­lat­ing free tumor DNA using targeted
muta­tion detec­tion59 or whole-genome low-pass sequenc­ing.60
Mass spec­trom­e­try is now being used to mea­sure mono­clo­
nal gammopathy in the periph­eral blood. There are 2 forms of
mass spec­trom­e­try: matrix-assisted laser desorp­tion ion­i­za­tion
time-of-flight mass spec­ trom­ et­ry (MALDI-TOF-MS) and liq­ uid
chro­ma­tog­ra­phy quad­ru­pole time-of-flight mass spec­trom­e­try
(LC-MS).61 MALDI-TOF-MS has a sen­si­tiv­ity of less than 0.01 g/dL62
and has replaced con­ven­tional serum pro­tein elec­tro­pho­re­
sis at some insti­tu­tions. LC-MS has even more sen­si­tiv­ity than
MALDI-TOF-MS, down to 0.005 g/dL, but has lower through­put.63
Moreover, mass spec­trom­et­ry can dis­tin­guish between “false-
pos­i­tive” bands on pro­tein elec­tro­pho­re­sis from ther­a­peu­tic
mono­clo­nal antibodies such as daratumumab vs the under­ly­ing
dis­ease.64 Given the increased sen­si­tiv­ity of mass spec­trom­e­try
of the periph­eral blood, stud­ies are com­par­ing the per­for­mance
of mass spec­trom­e­try with MRD performed on bone mar­row by
NGS or NGF, for exam­ple, in the Stem Cell Transplant in Myeloma
Incorporating Novel Agents65 and in GEM2012MENOS65 tri­als.66
In one study, LC-MS was esti­mated to be even more sen­si­tive
than NGS at 10−5 and could be used as a screen for MRD.67
Applying MRD to clin­i­cal prac­tice
The data establishing depth of response by MRD test­ ing and
out­comes are robust, and clin­i­cal tri­als now rou­tinely incor­po­
rate MRD test­ing to bench­mark per­for­mance. MRD test­ing is also
being used to strat­ify patients in clin­i­cal tri­als. For exam­ple, the
ECOG Effective Quadruplet Utilization after Treatment Evaluation
trial (NCT04566328) ran­dom­izes patients after ini­tial ther­apy with
dara-Rd to either con­sol­id ­ a­tion with addi­tional dara-Rd or add-
ing bortezomib to dara-Rd, and the study stratifies by MRD sta­
tus. However, apply­ing MRD test­ing to patient care is evolv­ing.
Indeed, the ques­tions raised when this topic was ini­tially cov­ered
in this edu­ca­tion pro­gram 4 years ago con­tinue to be rel­e­vant
now.68 There are ongo­ing tri­als to help answer this ques­tion (Fig-
ure 2). We acknowl­edge that there is sig­nif­i­cant var­i­abil­ity in MRD
use in clin­ic
­ al prac­tice. At this time, there are no pro­spec­tive,
ran­dom­ized data in which the infor­ma­tion from MRD test­ing can
guide treat­ment deci­sions. Nevertheless, if a bone mar­row biopsy
is being performed to con­firm a CR, send­ing the aspi­rate for MRD
 MRDA
Consolidation MRDB
+
Initial therapy + Next line
− Maintenance
or observation
− Observation

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/37/1851756/37yee.pdf by guest on 13 December 2021

Primary Sponsor MRD
Identifier MRD Treatment Study population
outcome modality

A. MRD testing to guide treatment after initial therapy

Randomization to arms
including R maint, RVd
consolidation + R maint,
+ Standard risk
R-isa maint, or isa-RVd
RADAR consolidation + R isa patients after auto PFS Univ. of Leeds NGF
maint SCT (10−5)

− Isa maint

Consolidation with dara-

+
RVd Following induction Univ. of
NCT04140162 MRD NGS
with dara-R Michigan
− Maint dara-R

+ Dara-R
NCT03901963 NGS
After auto-SCT MRD Janssen
AURIGA (10−5)
− R

+ Additional dara-KRd
NCT03224507 After dara-KRd and Univ. of NGS
MRD −5
MASTER auto SCT Alabama (10 )
Observation (after MRD
−
negative × 2)

B. MRD testing to guide treatment on maintenance therapy

+ Dara-Kd MRD negative
NCT04513639 patients
Relapse from randomized to arm
where patients are Oslo Univ. NGF
MRD Negativity PFS
− Continue observation followed by MRD Hospital (10−5)
as Indication for
every 4 months. (In
Treatment other arm, patients
are treated with
Figure 2. Examples of trials evaluating MRD to guide treatment. Treatment decisions with MRD may be broadly divided into
2 categories: (A) following initial therapy to guide treatment intensification or consolidation in patients with MRD positive disease
or (B) in patients on maintenance therapy, to guide discontinuation of treatment. Some trials are also examining “early” initiation of
therapy with the appearance of MRD-positive disease. D, dexamethasone; isa, isatuximab; K, carfilzomib; maint, maintenance; PD,

Prakash Singh Shekhawat

Minimal resid­ual dis­ease in mul­ti­ple mye­loma  |  41
 dara-Kd at PD
by conventional
criteria.)
+ Not applicable MRD NGF
Patients on 3 years Memorial
NCT04221178 negative −5
(10 )
− Discontinue maint of continuous maint Sloan Kettering
at 1 year
Continue maint (off NGS
+ −6
protocol) (10 and
NCT04108624 Patients on maint Univ. of
MRD exploring
MRD2STOP therapy and CR Chicago −7
− Discontinue maint 10 ) and
PET CT
Continue previously After 2 years of
+
assigned maint therapy maintenance
NCT04071457
− Randomization to (previously NGS
DRAMMATIC OS SWOG

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/37/1851756/37yee.pdf by guest on 13 December 2021

continue vs stopping randomized to
SWOG S1803
previously assigned dara-R vs R
maint therapy following auto-SCT)
+ Continue dara-R
− If MRD negative After 24 months of European
NCT03710603 NGS
sustained for 12 maint therapy with PFS Myeloma −5
PERSEUS (10 )
months: discontinue dara-R Network
dara; continue R
+ Dara Patients after 1-2
Polish NGF
NCT03697655 − Observation prior lines of Event-free
Myeloma −5
(10 )
PREDATOR therapy who are survival
Consortium
MRD negative

Figure 2. (continued)
progressive disease; PREDATOR, Pre-emptive Daratumumab Therapy of Minimal Residual Disease Reappearance or Biochemical
Relapse in Multiple Myeloma; R, lenalidomide; RADAR, Risk-Adapted Therapy Directed According to Response; REMNANT, Relapse
from MRD Negativity as Indication for Treatment; V, bortezomib.

test­ing is appro­pri­ate, as it may pro­vide prog­nos­tic infor­ma­tion as
well as estab­lish a ref­er­ence point for sub­se­quent MRD test­ing that
may con­firm sus­tain­abil­ity of response. To increase the sen­si­tiv­ity,
the oper­at­or should pri­or­i­tize the first pull for MRD test­ing, given
hemo­di­lu­tion with sub­se­quent pulls.69 Finally, if MRD assess­ment is
being performed, for com­plete­ness, it may be impor­tant to also
assess for extramedullary dis­ease with imag­ing such as PET CT.
Timing of high-dose mel­pha­lan and SCT
This has been a core ques­tion over the years for trans­plant-
eli­gi­ble patients and con­tin­ues to be an ongo­ing area of debate.
Despite the IFM 2009 trial show­ing sig­nif­i­cant improve­ment in
PFS, the lack of improve­ment in OS30 spurs this ongo­ing debate,
includ­ing with the FORTE trial.33 If out­comes of patients with
MRD-neg­a­tive and espe­cially sustained MRD-neg­a­tive dis­ease
are com­pa­ra­ble, does it mat­ter if this is achieved with­out high-
dose mel­pha­lan? Although the IFM and FORTE stud­ies incor­po­
rated MRD test­ing, this was after com­ple­tion of ini­tial ther­apy.
These stud­ies did not eval­u­ate MRD find­ings before high-dose
mel­pha­lan to inform deci­sion mak­ing.
Should ther­apy change to deepen response?
This is another open ques­tion in the mye­loma field. Attempts at
answer­ing this ques­tion, before the avail­abil­ity of MRD assess­
ments, include the Myeloma XI study of risk-adapted inten­si­fi­
ca­tion,70 which showed that the addi­tion of cyclo­phos­pha­mide,
bortezomib, and dexa­meth­a­sone in patients with sub­op­ti­mal
responses improved PFS. However, cur­ rent prac­ tice does not
inten­sify ther­apy above what was pre­vi­ously planned in patients
who have not achieved an opti­mal response. Because mye­loma
ther­apy is con­tin­u­ous, responses may improve over time. In a ret­
ro­spec­tive study of a real-world prac­tice of patients on lenalid-
omide main­te­nance, 34.3% of patients who were MRD pos­i­tive
(with MRD assess­ment according to local prac­tice) after induc­tion
treat­ment achieved MRD-neg­a­tive sta­tus dur­ing main­te­nance
ther­apy,71 suggesting that a change in ther­apy may not be oblig­a­
tory. The AURIGA study (NCT NCT03901963) is exam­in­ing the role
of adding daratumumab to lenalidomide main­te­nance to eval­u­ate
the ben­e­fit of adding addi­tional ther­apy to deepen a response.
Can we de-esca­late treat­ment?
Current prac­tice is to treat until pro­gres­sion with a com­bi­na­tion
of induc­tion and main­te­nance ther­apy. But can patients step off
this “tread­mill” of con­tin­uo
­ us ther­apy to avoid the adverse events
and bur­den of chronic ther­apy? There are sev­eral tri­als exam­in­
ing de-esca­la­tion of ther­apy. There is an ongo­ing phase 2 study
in newly diag­nosed, trans­plant eli­gi­ble patients, the Monoclonal
Antibody-Based Sequential Therapy for Deep Remission in Multi-
ple Myeloma study (NCT03224507).72 Patients undergo induc­tion
ther­apy with dara-KRd, followed by auto-SCT. Patients who are
MRD neg­a­tive by NGS (10−5) after auto-SCT discontinue treat­ment,
whereas patients who are MRD pos­i­tive con­tinue to undergo con­
sol­id
­ a­tion with dara-KRd for up to 2 cycles until MRD neg­a­tive.
In the PERSEUS trial (NCT03710603), patients on main­ te­
nance
with daratumumab and lenalidomide who are MRD neg­a­tive can
discontinue daratumumab and con­tinue on lenalidomide. In the

42  |  Hematology 2021  |  ASH Education Program

DRAMMATIC trial (SWOG 1803), patients who are MRD neg­a­tive
after ini­
tial ther­
apy are ran­dom­ized to con­ tinue the assigned
main­te­nance vs stop­ping assigned main­te­nance ther­apy.
Should ther­apy change if a patient becomes MRD pos­i­tive?
An ongo­ing ques­tion is the opti­mal tim­ing of treating relapsed
dis­ease. Patients who are treated at the time of bio­ chem­ i­
cal
rather than clin­i­cal relapse have bet­ter out­comes, as seen in a
sub­group anal­y­sis of the ENDEAVOR trial (with the caveat that this
study was not designed to answer this spe­cific ques­tion).73 Could
the out­comes of patients be bet­ter when treated at relapse, with
an even lower bur­den of dis­ease, by MRD? As noted pre­vi­ously,
the appear­ance of MRD-pos­i­tive dis­ease may her­ald bio­chem­i­
cal or clin­i­cal relapse sev­eral months later. The REMNANT study
(NCT04513639) will help answer this ques­tion.74 Patients who are
MRD neg­a­tive after induc­tion ther­apy are ran­dom­ized to start
treat­ment at the time of MRD relapse vs at the time of pro­gres­sive
dis­ease according to IMWG cri­te­ria. However, a lim­i­ta­tion in treat-
ing patients for relapse by MRD cri­te­ria is that cur­rent clin­i­cal tri­als
gen­er­ally require mea­sur­able dis­ease, and MRD pos­i­tiv­ity is not
con­sid­ered mea­sur­able to be eli­gi­ble for the trial. Consequently,
this may limit treat­ment to stan­dard-of-care options instead of
the poten­tially more inno­va­tive ther­a­pies under inves­ti­ga­tion.
CLINICAL CASE (Con­t in­u ed)
Our patient under­went MRD test­ing using Adaptive NGS and was
MRD neg­a­tive. She had MRD test­ing seri­ally for 2 years and was
neg­a­tive on both occa­sions. However, even with the sustained
MRD neg­a­tiv­ity, she pre­ferred to con­tinue with lenalidomide and
ixazomib main­te­nance. Testing for MRD at 5 years after auto-SCT
resulted in a pos­i­tive test, albeit at a low level of 0 to 1 × 10−6. With
this new find­ing, she had fur­ther workup includ­ing whole-body
low-dose CT, which did not show any new find­ings. She has opted
to con­tinue the cur­rent reg­i­men, with a ten­ta­tive plan of repeat­
ing MRD test­ing in 6 months. This case illus­trates some of the chal­
lenges with MRD test­ing, as nei­ther the repeated neg­a­tive results
nor the new low pos­i­tive result prompted a change in treat­ment.
Conclusion
Overall, the field is for­tu­nate that newer treat­ments are lead­ing to
unprec­e­dented depths of response that require newer meth­ods
such as NGS or NGF to mea­sure their effect. What was once a test
restricted to spe­cial­ized research set­tings or clin­i­cal tri­als is now
read­ily avail­­able for any patient. Although there are matur­ing data
on how it adds new prog­nos­tic power, there is a lag in the data
for how to effec­tively use the test to make treat­ment deci­sions.
Ongoing tri­als will pro­vide data and guid­ance on how to incor­po­
rate MRD test­ing into clin­i­cal prac­tice to tai­lor ther­apy. Moreover,
the incor­po­ra­tion of func­tional imag­ing and liq­uid biop­sies will
pro­vide less inva­sive ways of eval­u­at­ing dis­ease bur­den.
Conflict-of-inter­est dis­clo­sure
Andrew J. Yee has consulted for Adaptive, Amgen, BMS, GSK,
Janssen, Karyopharm, Oncopeptides, Sanofi, and Takeda and has
received clin­i­cal trial sup­port from Adaptive, Amgen, BMS, Jans-
sen, and Takeda.
Prakash Singh Shekhawat
Noopur Raje has consulted for Amgen, BMS, Bluebird, GSK, Jans-
sen, and Karyopharm; served on sci­en­tific advi­sory board for Car-
ibou and Immuneel; and received research funding from Bluebird.
Off-label drug use
Andrew J. Yee: no off-label drug use discussed.
Noopur Raje: no off-label drug use discussed.
Correspondence
Noopur Raje, Mas­sa­chu­setts General Hospital, 55 Fruit Street,
Bos­ton, MA 02114; e-mail: nraje@mgh.harvard.edu.
References
1. Cavo M, Zamagni E, Tosi P, et al; Bologna 2002 study. Superiority of tha­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/37/1851756/37yee.pdf by guest on 13 December 2021
lid­o­mide and dexa­meth­a­sone over vin­cris­tine-doxorubicindexamethasone
(VAD) as pri­mary ther­apy in prep­a­ra­tion for autol­o­gous trans­plan­ta­tion for
mul­ti­ple mye­loma. Blood. 2005;106(1):35-39.
2. Landgren O, Hultcrantz M, Diamond B, et  al. Safety and effec­tive­ness of
weekly carfilzomib, lenalidomide, dexa­ meth­a­
sone, and daratumumab
com­bi­na­tion ther­apy for patients with newly diag­nosed mul­ti­ple mye­loma:
the MANHATTAN nonrandomized clin­i­cal trial. JAMA Oncol. 2021;7(6):862-
868.
3. van de Velde HJ, Liu X, Chen G, Cakana A, Deraedt W, Bayssas M. Com-
plete response cor­ re­
lates with long-term sur­ vival and pro­ gres­
sion-free
sur­vival in high-dose ther­ apy in mul­ ti­
ple mye­ loma. Haematologica.
2007;92(10):1399-1406.
4. van de Velde H, Londhe A, Ataman O, et al. Association between com­plete
response and out­comes in trans­plant-eli­gi­ble mye­loma patients in the era
of novel agents. Eur J Haematol. 2017;98(3):269-279.
5. Martinez-Lopez J, Blade J, Mateos M-V, et al; Grupo Español de MM;
Programa para el Estudio de la Terapé utica en Hemopatía Maligna. Long-
term prog­nos­tic sig­nif­i­cance of response in mul­ti­ple mye­loma after stem
cell trans­plan­ta­tion. Blood. 2011;118(3):529-534.
6. Gay F, Larocca A, Wijermans P, et al. Complete response cor­re­lates with
long-term pro­gres­sion-free and over­all sur­vival in elderly mye­loma treated
with novel agents: anal­y­sis of 1175 patients. Blood. 2011;117(11):3025-
3031.
7. Rajkumar SV, Jaco­bus S, Callander NS, et al; Eastern Cooperative Oncology
Group. Lenalidomide plus high-dose dexa­meth­a­sone ver­sus lenalidomide
plus low-dose dexa­meth­a­sone as ini­tial ther­apy for newly diag­nosed mul­
ti­ple mye­loma: an open-label randomised con­trolled trial. Lancet Oncol.
2010;11(1):29-37.
8. Kumar SK, Harrison SJ, Cavo M, et al. Venetoclax or pla­cebo in com­bi­na­tion
with bortezomib and dexa­meth­a­sone in patients with relapsed or refrac­
tory mul­ti­ple mye­loma (BELLINI): a randomised, dou­ble-blind, multicentre,
phase 3 trial. Lancet Oncol. 2020;21(12):1630-1642.
9. Kumar S, Paiva B, Anderson KC, et al. International Myeloma Working Group
con­sen­sus cri­te­ria for response and min­i­mal resid­ual dis­ease assess­ment in
mul­ti­ple mye­loma. Lancet Oncol. 2016;17(8):e328-e346.
10. Rawstron AC, Orfao A, Beksac M, et al; Euro­pean Myeloma Network. Report
of the Euro­pean Myeloma Network on multiparametric flow cytom­e­try in
mul­ti­ple mye­loma and related dis­or­ders. Haematologica. 2008;93(3):431-
438.
11. Paiva B, Vidriales M-B, Cerveró J, et al; GEM (Grupo Español de MM)/PETH-
EMA (Programa para el Estudio de la Terapéutica en Hemopatías Malignas)
Cooperative Study Groups. Multiparameter flow cytometric remis­sion is
the most rel­e­vant prog­nos­tic fac­tor for mul­ti­ple mye­loma patients who
undergo autol­o­gous stem cell trans­plan­ta­tion. Blood. 2008;112(10):4017-
4023.
12. Flores-Montero J, Sanoja-Flores L, Paiva B, et  al. Next gen­er­a­tion flow for
highly sen­si­tive and stan­dard­ized detec­tion of min­i­mal resid­ual dis­ease in
mul­ti­ple mye­loma. Leukemia. 2017;31(10):2094-2103.
13. Martinez-Lopez J, Lahuerta JJ, Pepin F, et  al. Prognostic value of deep
sequenc­ing method for min­i­mal resid­ual dis­ease detec­tion in mul­ti­ple
mye­loma. Blood. 2014;123(20):3073-3079.
14. Ching T, Duncan ME, Newman-Eerkes T, et al. Analytical eval­u­a­tion of the
clonoSEQ assay for establishing mea­sur­able (min­i­mal) resid­ual dis­ease in
acute lym­pho­blas­tic leu­ke­mia, chronic lym­pho­cytic leu­ke­mia, and mul­ti­ple
mye­loma. BMC Cancer. 2020;20(1):612.
15. Clonoseq. Clonoseq assay [tech­ni­cal infor­ma­tion]. Clonoseq; 2021.
Minimal resid­ual dis­ease in mul­ti­ple mye­loma  |  43
16. Oliva S, Genuardi E, Belotti A, et al. Minimal resid­ual dis­ease eval­ua ­ ­tion by
mul­ti­pa­ram­e­ter flow cytom­e­try and next gen­er­a­tion sequenc­ing in the Forte
trial for newly diag­nosed mul­ti­ple mye­loma patients. Blood. 2019;134(suppl 1):
4322.
17. Avet-Loiseau H, Bene MC, Wuilleme S, et  al. Concordance of post-
con­sol­i­da­tion min­i­mal resid­ual dis­ease rates by multiparametric flow
cytom­e­try and next-gen­er­a­tion sequenc­ing in CASSIOPEIA. Clin Lym-
phoma Myeloma Leuk. 2019;19(10):e3-e4.
18. Medina A, Puig N, Flores-Montero J, et al. Comparison of next-gen­er­a­tion
sequenc­ing (NGS) and next-gen­er­a­tion flow (NGF) for min­i­mal resid­ual
dis­ease (MRD) assess­ment in mul­ti­ple mye­loma. Blood Cancer J. 2020;10(10):
108.
19. Ho C, Syed M, Roshal M, et al. Routine eval­u­at­ ion of min­i­mal resid­ual dis­
ease in mye­ loma using next-gen­ er­
a­
tion sequenc­ ing clonality test­ ing:
fea­si­bil­ity, chal­lenges, and direct com­par­i­son with high-sen­si­tiv­ity flow
cytom­e­try. J Mol Diagn. 2021;23(2):181-199.
20. Hussaini MO, Srivastava J, Lee LW, et al. Moffitt can­cer cen­ter 2-year sin­gle-
insti­tu­tion expe­ri­ence with next-gen­er­a­tion sequenc­ing min­i­mal resid­ual
dis­ease detec­tion: clin­i­cal util­ity, appli­ca­tion, and cor­re­la­tion with out­
comes in plasma cell and lym­phoid malig­nan­cies. Blood. 2019;134(suppl
1):4654.
21. Landgren O, Devlin S, Boulad M, Mailankody S. Role of MRD sta­tus in rela­
tion to clin­i­cal out­comes in newly diag­nosed mul­ti­ple mye­loma patients: a
meta-anal­y­sis. Bone Marrow Transplant. 2016;51(12):1565-1568.
22. Munshi NC, Avet-Loiseau H, Rawstron AC, et  al. Association of min­i­mal
resid­ual dis­ease with supe­rior sur­vival out­comes in patients with mul­ti­ple
mye­loma: a meta-anal­y­sis. JAMA Oncol. 2017;3(1):28-35.
23. Munshi NC, Avet-Loiseau H, Anderson KC, et  al. A large meta-anal­ y­sis
establishes the role of MRD neg­a­tiv­ity in long-term sur­vival out­comes in
patients with mul­ti­ple mye­loma. Blood Adv. 2020;4(23):5988-5999.
24. Costa LJ, Derman BA, Bal S, et al. International har­mo­ni­za­tion in performing
and reporting min­i­mal resid­ual dis­ease assess­ment in mul­ti­ple mye­loma
tri­als. Leukemia. 2021;35(1):18-30.
25. Anderson KC, Auclair D, Adam SJ, et al. Minimal resid­ual dis­ease in mye­
loma: appli­ca­tion for clin­i­cal care and new drug reg­is­tra­tion [published
online 28 July 2021]. Clin Cancer Res.
26. Gormley NJ, Farrell AT, Pazdur R. Minimal resid­ual dis­ease as a poten­tial
sur­ro­gate end point-lin­ger­ing ques­tions. JAMA Oncol. 2017;3(1):18-20.
27. Anderson KC, Auclair D, Kelloff GJ, et al. The role of min­i­mal resid­ual dis­ease
test­ing in mye­loma treat­ment selec­tion and drug devel­op­ment: cur­rent
value and future appli­ca­tions. Clin Cancer Res. 2017;23(15):3980-3993.
28. Hol­stein SA, Al-Kadhimi Z, Costa LJ, et  al. Summary of the third annual
blood and mar­row trans­plant clin­i­cal tri­als net­work mye­loma inter­group
work­shop on min­i­mal resid­ual dis­ease and immune pro­fil­ing. Biol Blood
Marrow Transplant. 2020;26(1):e7-e15.
29. Rawstron AC, Gregory WM, de Tute RM, et al. Minimal resid­ual dis­ease in
mye­loma by flow cytom­e­try: inde­pen­dent pre­dic­tion of sur­vival ben­e­fit
per log reduc­tion. Blood. 2015;125(12):1932-1935.
30. Attal M, Lauwers-Cances V, Hulin C, et al; IFM 2009 Study. Lenalidomide,
bortezomib, and dexa­meth­a­sone with trans­plan­ta­tion for mye­loma. N Engl
J Med. 2017;376(14):1311-1320.
31. Perrot A, Lauwers-Cances V, Corre J, et al. Minimal resid­ual dis­ease neg­a­
tiv­ity using deep sequenc­ing is a major prog­nos­tic fac­tor in mul­ti­ple mye­
loma. Blood. 2018;132(23):2456-2464.
32. Avet-Loiseau H, Moreau P, Attal M, et al. Efficacy of daratumumab (DARA)+
bortezomib/tha­lid­o­mide/dexa­meth­a­sone (D-VTd) in trans­plant-eli­gi­ble
newly diag­nosed mul­ti­ple mye­loma (TE NDMM) based on min­i­mal resid­
ual dis­ease (MRD) sta­tus: anal­y­sis of the CASSIOPEIA trial. J Clin Oncol.
2019;37(suppl_15):8017-8017.
33. Oliva S, Genuardi E, Petrucci MT, et  al. Impact of min­i­mal resid­ual dis­
ease (MRD) by mul­ti­pa­ram­e­ter flow cytom­e­try (MFC) and next-gen­er­a­tion
sequenc­ing (NGS) on out­come: results of newly diag­nosed trans­plant-
eli­gi­ble mul­ti­ple mye­loma (MM) patients enrolled in the forte trial. Blood.
2020;136(suppl 1):44-45.
34. Paiva B, Puig N, Cedena M-T, et al; GEM (Grupo Español de Mieloma)/
PETHEMA (Programa Para el Estudio de la Terapéutica en Hemopatías
Malignas) Cooperative Study Group. Measurable resid­ual dis­ease by next-
gen­er­a­tion flow cytom­e­try in mul­ti­ple mye­loma. J Clin Oncol. 2020;38(8):
784-792.
35. Goicoechea I, Puig N, Cedena M-T, et al. Deep MRD pro­fil­ing defi­nes out­
come and unveils dif­fer­ent modes of treat­ment resis­tance in stan­dard- and
high-risk mye­loma. Blood. 2021;137(1):49-60.
44  |  Hematology 2021  |  ASH Education Program
36. Lahuerta J-J, Paiva B, Vidriales M-B, et al; GEM (Grupo Español de Mieloma)/
PETHEMA (Programa para el Estudio de la Terapéutica en Hemopatías
Malignas) Cooperative Study Group. Depth of response in mul­ti­ple mye­
loma: a pooled anal­y­sis of three PETHEMA/GEM clin­ic ­ al tri­als. J Clin Oncol.
2017;35(25):2900-2910.
37. Haessler J, Shaughnessy JD Jr, Zhan F, et al. Benefit of com­plete response in
mul­ti­ple mye­loma lim­ited to high-risk sub­group iden­ti­fied by gene expres­
sion pro­fil­ing. Clin Cancer Res. 2007;13(23):7073-7079.
38. Barlogie B, Anaissie E, Haessler J, et  al. Complete remis­ sion sustained
3 years from treat­ment ini­ti­a­tion is a pow­er­ful sur­ro­gate for extended sur­
vival in mul­ti­ple mye­loma. Cancer. 2008;113(2):355-359.
39. Sidana S, Tandon N, Dispenzieri A, et al. Relapse after com­plete response in
newly diag­nosed mul­ti­ple mye­loma: impli­ca­tions of dura­tion of response
and pat­terns of relapse. Leukemia. 2019;33(3):730-738.
40. Hoering A, Crowley J, Shaughnessy JD Jr, et al. Complete remis­sion in mul­ti­
ple mye­loma exam­ined as time-depen­dent var­ia ­ ble in terms of both onset
and dura­tion in total ther­apy pro­to­cols. Blood. 2009;114(7):1299-1305.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/37/1851756/37yee.pdf by guest on 13 December 2021
41. Avet-Loiseau H, San-Miguel J, Casneuf T, et al. Evaluation of sustained min­
i­mal resid­ual dis­ease neg­a­tiv­ity with daratumumab-com­bi­na­tion reg­i­mens
in relapsed and/or refrac­tory mul­ti­ple mye­loma: anal­y­sis of POLLUX and
CASTOR. J Clin Oncol. 2021;39(10):1139-1149.
42. San-Miguel JF, Avet-Loiseau H, Paiva B, et al. Sustained min­i­mal resid­ual dis­
ease neg­a­tiv­ity with daratumumab in newly diag­nosed mul­ti­ple mye­loma:
MAIA and ALCYONE [published online 16 July 2021]. Blood.
43. Diamond B, Korde N, Lesokhin AM, et al. Dynamics of min­i­mal resid­ual dis­
ease in patients with mul­ti­ple mye­loma on con­tin­uo ­ us lenalidomide main­te­
nance: a sin­gle-arm, sin­gle-cen­tre, phase 2 trial. Lancet Haematol. 2021;8(6):
e422-e432.
44. Ferrero S, Ladetto M, Drandi D, et  al. Long-term results of the GIMEMA
VEL-03-096 trial in MM patients receiv­ing VTD con­sol­i­da­tion after ASCT:
MRD kinet­ics’ impact on sur­vival. Leukemia. 2015;29(3):689-695.
45. Oliva S, Gambella M, Gilestro M, et al. Minimal resid­ual dis­ease after trans­
plan­ta­tion or lenalidomide-based con­sol­i­da­tion in mye­loma patients: a
pro­spec­tive anal­y­sis. Oncotarget. 2017;8(4):5924-5935.
46. Martinez-Lopez J, Wong SW, Shah N, et  al. Clinical value of mea­sur­able
resid­ual dis­ease test­ing for assessing depth, dura­tion, and direc­tion of
response in mul­ti­ple mye­loma. Blood Adv. 2020;4(14):3295-3301.
47. Görgün GT, Whitehill G, Anderson JL, et  al. Tumor-pro­mot­ing immune-
sup­pres­sive mye­loid-derived sup­pres­sor cells in the mul­ti­ple mye­loma
micro­en­vi­ron­ment in humans. Blood. 2013;121(15):2975-2987.
48. Rasche L, Buros A, Weinhold N, et al. The clin­i­cal impact of macrofocal dis­
ease in mul­ti­ple mye­loma dif­fers between pre­sen­ta­tion and relapse. Blood.
2016;128(22):4431.
49. Moreau P, Attal M, Caillot D, et al. Prospective eval­u­a­tion of mag­netic res­o­nance
imag­ing and [(18)F]fluorodeoxyglucose pos­i­tron emis­sion tomog­ra­phy-
com­puted tomog­ra­phy at diag­no­sis and before main­te­nance ther­apy in
symp­tom­atic patients with mul­ti­ple mye­loma included in the IFM/DFCI
2009 trial: results of the IMAJEM study. J Clin Oncol. 2017;35(25):2911-2918.
50. Moreau P, Zweegman S, Perrot A, et al. Evaluation of the prog­nos­tic value
of pos­i­tron emis­sion tomog­ra­phy-com­puted tomog­ra­phy (PET-CT) at
diag­no­sis and fol­low-up in trans­plant-eli­gi­ble newly diag­nosed mul­ti­ple
mye­loma (TE NDMM) patients treated in the phase 3 Cas­si­op ­ eia Study:
results of the Cassiopet Companion Study. Blood. 2019;134(suppl 1):692.
51. Rasche L, Alapat D, Kumar M, et  al. Combination of flow cytom­e­try and
func­tional imag­ing for mon­i­tor­ing of resid­ual dis­ease in mye­loma. Leuke-
mia. 2019;33(7):1713-1722.
52. Zamagni E, Nanni C, Dozza L, et  al. Standardization of 18F-FDG-PET/CT
according to Deauville cri­te­ria for met­a­bolic com­plete response def­i­ni­tion
in newly diag­nosed mul­ti­ple mye­loma. J Clin Oncol. 2021;39(2):116-125.
53. Pineda-Roman M, Bolejack V, Arzoumanian V, et al. Complete response in
mye­loma extends sur­vival with­out, but not with his­tory of prior mono­clo­
nal gammopathy of unde­ter­mined sig­nif­i­cance or smoul­der­ing dis­ease. Br
J Haematol. 2007;136(3):393-399.
54. Zhan F, Barlogie B, Arzoumanian V, et  al. Gene-expres­sion sig­na­ture of
benign mono­clo­nal gammopathy evi­dent in mul­ti­ple mye­loma is linked to
good prog­no­sis. Blood. 2007;109(4):1692-1700.
55. Weinhold N, Heuck CJ, Rosenthal A, et  al. Clinical value of molec­ u­
lar
subtyping mul­ti­ple mye­loma using gene expres­sion pro­fil­ing. Leukemia.
2016;30(2):423-430.
56. Schinke C, Hoering A, Wang H, et al. The prog­nos­tic value of the depth
of response in mul­ti­ple mye­loma depends on the time of assess­ment, risk
sta­tus and molec­u­lar sub­type. Haematologica. 2017;102(8):e313-e316.
57. Sanoja-Flores L, Flores-Montero J, Puig N, et  al. Blood mon­i­tor­ing of cir­
cu­lat­ing tumor plasma cells by next gen­er­a­tion flow in mul­ti­ple mye­loma
after ther­apy. Blood. 2019;134(24):2218-2222.
58. Mazzotti C, Buisson L, Maheo S, et al. Myeloma MRD by deep sequenc­ing
from cir­cu­lat­ing tumor DNA does not cor­re­late with results obtained in the
bone mar­row. Blood Adv. 2018;2(21):2811-2813.
59. Mithraprabhu S, Morley R, Khong T, et al. Monitoring tumour bur­den and
ther­a­peu­tic response through anal­y­sis of cir­cu­lat­ing tumour DNA and extra­
cel­lu­lar RNA in mul­ti­ple mye­loma patients. Leukemia. 2019;33(8):2022-
2033.
60. Guo G, Raje NS, Seifer C, et al. Genomic dis­cov­ery and clonal track­ing in
mul­ti­ple mye­loma by cell-free DNA sequenc­ing. Leukemia. 2018;32(8):1838-
1841.
61. Murray DL, Puig N, Kristinsson S, et al. Mass spec­trom­e­try for the eval­u­
a­tion of mono­clo­nal pro­teins in mul­ti­ple mye­loma and related dis­or­ders:
an International Myeloma Working Group Mass Spectrometry Committee
Report. Blood Cancer J. 2021;11(2):24.
62. Mills JR, Kohlhagen MC, Dasari S, et  al. Comprehensive assess­ ment of
M-pro­teins using nanobody enrich­ment cou­pled to MALDI-TOF mass spec­
trom­e­try. Clin Chem. 2016;62(10):1334-1344.
63. Barnidge DR, Dasari S, Botz CM, et al. Using mass spec­trom­e­try to mon­i­tor
mono­clo­nal immu­no­glob­u­lins in patients with a mono­clo­nal gammopathy.
J Proteome Res. 2014;13(3):1419-1427.
64. Mills JR, Kohlhagen MC, Willrich MAV, Kourelis T, Dispenzieri A, Murray DL.
A uni­ver­sal solu­tion for elim­i­nat­ing false pos­i­tives in mye­loma due to
ther­a­peu­tic mono­clo­nal anti­body inter­fer­ence. Blood. 2018;132(6):670-
672.
65. Dispenzieri A, Krishnan AY, Arendt B, et al. MASS-FIX ver­sus stan­dard meth­
ods to pre­dict for PFS and OS among mul­ti­ple mye­loma patients par­tici­
pat­ing on the STAMINA trial. J Clin Oncol. 2021;39(15, suppl):8009.
66. Puig N, Paiva B, Contreras T, et al. Analysis of min­i­mal resid­ual dis­ease in bone
mar­row by NGF and in periph­eral blood by mass spec­trom­e­try in newly
diag­nosed mul­ti­ple mye­loma patients enrolled in the GEM2012MENOS65
clin­i­cal trial. J Clin Oncol. 2021;39(15, suppl):8010.
Prakash Singh Shekhawat
67. Derman BA, Stefka AT, Jiang K, et al. Measurable resid­ual dis­ease assessed
by mass spec­trom­e­try in periph­eral blood in mul­ti­ple mye­loma in a phase
II trial of carfilzomib, lenalidomide, dexa­meth­a­sone and autol­o­gous stem
cell trans­plan­ta­tion. Blood Cancer J. 2021;11(2):19.
68. Davies FE. Is molec­u­lar remis­sion the goal of mul­ti­ple mye­loma ther­apy?
Hematology Am Soc Hematol Educ Program. 2017;2017(1):205-211.
69. Manasanch EE, Salem DA, Yuan CM, et al. Flow cytometric sen­si­tiv­ity and
char­ac­ter­is­tics of plasma cells in patients with mul­ti­ple mye­loma or its pre­
cur­sor dis­ease: influ­ence of biopsy site and anticoagulation method. Leuk
Lymphoma. 2015;56(5):1416-1424.
70. Jackson GH, Davies FE, Pawlyn C, et al; UK NCRI Haematological Oncol-
ogy Clinical Studies Group. Response-adapted inten­si­fi­ca­tion with cyclo­
phos­pha­mide, bortezomib, and dexa­meth­a­sone ver­sus no inten­si­fi­ca­tion
in patients with newly diag­ nosed mul­ ti­
ple mye­ loma (Myeloma XI): a
multicentre, open-label, randomised, phase 3 trial. Lancet Haematol.
2019;6(12):e616-e629.
71. Alonso R, Cedena M-T, Wong S, et al. Prolonged lenalidomide main­te­nance
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/37/1851756/37yee.pdf by guest on 13 December 2021
ther­apy improves the depth of response in mul­ti­ple mye­loma. Blood Adv.
2020;4(10):2163-2171.
72. Costa LJ, Chhabra S, Godby KN, et al. Daratumumab, carfilzomib, lenalid-
omide and dexa­meth­a­sone (dara-KRd) induc­tion, autol­o­gous trans­plan­ta­
tion and post-trans­plant, response-adapted, mea­sur­able resid­ual dis­ease
(MRD)-based dara-Krd con­ sol­ i­
da­
tion in patients with newly diag­ nosed
mul­ti­ple mye­loma (NDMM). Blood. 2019;134(suppl 1):860.
73. Moreau P, Siegel DS, Goldschmidt H, et al. Subgroup anal­y­sis of patients
with bio­chem­i­cal or symp­tom­atic relapse at the time of enroll­ment in the
endeavor study. Blood. 2018;132(suppl 1):3243.
74. Rasmussen A-M, Askeland FB, Schjesvold F. The next step for MRD in mye­
loma? treating MRD relapse after first line treat­ment in the REMNANT study.
Hemato. 2020;1(2):36-48.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000230
Minimal resid­ual dis­ease in mul­ti­ple mye­loma  |  45
 CHALLENGES IN MULTIPLE MYELOMA TREATMENT

Treatment of older adult or frail patients

with multiple myeloma

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/46/1852113/46grant.pdf by guest on 13 December 2021

Shakira J. Grant,1,* Ciara L. Freeman,2,* and Ashley E. Rosko3
1
Department of Medicine, Division of Hematology, The University of North Carolina at Chapel Hill, Chapel Hill, NC; Lineberger Comprehensive
Cancer Center, The University of North Carolina at Chapel Hill, Chapel Hill, NC; 2Division of Medical Oncology, University of British Columbia,
Vancouver, BC, Canada; and 3Division of Hematology, The Ohio State University, Columbus, OH
*Joint frst authorship.

Older adults with multiple myeloma (MM) are a growing population, and personalizing treatment based on disease and
health status is imperative. Similar to MM staging systems that provide disease-related prognostic information, myeloma-
specific frailty tools can better identify subgroups at greatest risk for treatment-related toxicity and early treatment dis-
continuation, as well as predict overall survival. Several myeloma-specific validated tools are well studied. Although these
fitness/frailty scores have shaped our understanding of the heterogeneity among older adults with myeloma, the applica-
tion of such scores in treatment decision making (ie, transplant considerations, relapse) is an unmet need. Here we outline
how to incorporate frailty assessments in the evaluation of older adults with MM in the clinical setting with consideration
of other factors such as patient preferences, treatment risks/benefits, life expectancy, and disease biology.

LEARNING OBJECTIVES
• Frailty can be objectively characterized in older adults with MM using validated scoring systems.
• Disease and patient factors will influence treatment decisions (health status, biology, risks/benefts, and shared
decision making).

CLINICAL CASE
A 70-year-old woman with newly diagnosed standard-risk,
Revised International Staging System (R-ISS) III, immuno-
globulin G κ multiple myeloma (MM) presents for consul-
tation. Until 4 months ago, she lived independently, had
an active lifestyle, exercised 3 times weekly, and served
as the lead volunteer coordinator for a nonproft organi-
zation. She was also independent in all activities of daily
living (ADLs) and instrumental activities of daily living
(IADLS). She now reports severe lower back pain that
limits her mobility, fatigue, and shortness of breath after
walking 1 block that has limited her function. She has well-
controlled diabetes mellitus (glipizide) and hypertension
(amlodipine, lisinopril) and reports taking several over-the-
counter vitamins to maintain her vitality. She is widowed
and lives alone. Her Eastern Cooperative Oncology Group
(ECOG) performance status is 2. Positron emission tomo-
graphy/computed tomography confrms diffuse lytic lesions
in the humeri, femora, pelvis, and spine along with multiple
compression fractures involving T6 to T8. Her estimated
glomerular fltration rate is 55. She emphasizes that main-
taining her independence and quality of life are her top
priorities. She wants to know if there are tools to gauge her
health and would like the best therapy for her personally.
Myeloma fitness/frailty scores can inform
treatment tolerance
Predicting treatment-related toxicity is especially critical
for older adults with myeloma due to the heterogeneity
that exists in aging. Differences in how individuals age
arise from age-associated losses in physical and cognitive
function and the additive impact of medical comorbidities,
which becomes more prevalent with advancing age.1,2 Pre-
vious studies have shown that performance status (Karnof-
sky, ECOG) is incapable of fully capturing the ftness level
of an older adult.3 Furthermore, although staging systems
such as the ISS and R-ISS can provide valuable disease-
related prognostic information, these staging systems are
incapable of fully discerning the subgroup of patients at
greatest risk for treatment-related toxicity and early treat-
ment discontinuation.4,5

46 | Hematology 2021 | ASH Education Program

Table 1. Myeloma fit­ness/frailty risk scores

Frailty score Geriatric domains Biologic marker Cytogenetics included? Score range Interpretation
IMWG6 ADLs None No 0-2 0 (fit)
IADLs 1 (inter­me­di­ate-fit)
CCI 2 (frail)
R-MCI7 Age None Yes 0-9 0-3 (fit)
Fried frailty 4-6 (inter­me­di­ate- fit)
Lung func­tion 7-9 (frail)
Renal func­tion
Karnofsky per­for­mance sta­tus
Facon frailty scale8 Age None No 0-1 0-1 (nonfrail)
CCI ≥2 ≥2 (frail)
ECOG per­for­mance sta­tus9

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/46/1852113/46grant.pdf by guest on 13 December 2021

Myeloma Research Age CRP No <−0.256 <−0.256 (low risk)
Alliance risk pro­file10 WHO per­for­mance sta­tus9 ISS −0.256-−0.0283 −0.256 to −0.0283
>−0.0283 (medium risk)
>−0.0283 (high risk)
Mayo frailty index11 Age NT- proBNP No 0-3 0 (stage 1)
WHO per­for­mance sta­tus9 1 (stage 2)
2 (stage 3)
3 (stage 4)
Ancona vul­ner­a­bil­ity CCI None No 0-2 0 (low)
score12 WHO per­for­mance sta­tus9 1 (inter­me­di­ate)
2 (high)
CRP, C-reac­tive pro­tein; NT-proBNP, N-terminal pro B-type Natriuretic Peptide; WHO, World Health Organization.

To account for some of these aging-related def­i­cits, the Inter-
national Myeloma Working Group (IMWG) pro­ posed a scor­ ing
sys­tem for mye­loma patient frailty that pre­dicts treat­ment-related
tox­ic­ity and sur­vival, using age, the Katz ADL, the Lawton IADL,
and the Charlson Comorbidity Index (CCI)6 (Table 1). In the clin­i­
cal vignette, the patient has an IMWG score of 1 = inter­me­di­ate-fit.
Among the inter­me­di­ate-fit sub­group, the 3-year over­all sur­vival
(OS) was 76% (haz­ard ratio [HR], 1.61; 95% CI, 1.02-2.56; P = .042),
the cumu­la­tive inci­dence of grade 3 or higher nonhematologic
adverse events at 12 months was 16.7% (HR, 1.23; 95% CI, 0.89-1.71;
P = .217), and the cumu­la­tive inci­dence of treat­ment dis­con­tin­u­a­
tion at 12 months was 20.8% (HR, 1.41; 95% CI, 1.00-2.01; P = .052).6
After discussing her frailty score, the patient inquires
whether addi­tional frailty scales exist and whether
these apply to newly diag­nosed MM
Additional frailty scores have been devel­oped incor­po­rat­ing age,
comorbidities, phys­i­cal per­for­mance, and bio­mark­ers (Table 1).
Engelhardt et al7 found that age, mye­ loma cyto­ge­
net­ics, frailty
(Fried phe­no­type), per­for­mance sta­tus, and pul­mo­nary and re-
nal func­tion (Revised Myeloma Comorbidity Index [R-MCI]) were
prog­ nos­ tic for OS in a pro­ spec­ tive cohort of 801 patients
with newly diag­nosed MM. Of the 801 evaluable patients, 30.8%
were con­sid­ered fit, 55.7% inter­me­di­ate-fit, and 13.5% frail.7 In
the der­i­va­tion cohort, 552 patients clas­si­fied as frail had a near
10-fold greater risk of dying than those con­sid­ered fit (HR, 9.57;
95% CI, 6.52-14.03).7 In our clin­i­cal vignette, we are not pro­vided
with data to esti­mate the patient’s Fried frailty phe­no­type (re-
quires gait speed, grip strength, weight, self-reported exhaus­
tion, and phys­i­cal activ­ity level).13 Neither are we pro­vided with
pul­mo­nary func­tion test data, and as such, we can­not cal­cu­late
this patient’s R-MCI score. A ret­ro­spec­tive anal­y­sis of 1618 trial
par­tic­i­pants enrolled in the FIRST trial14,15 reported on a sim­pli­fied
Prakash Singh Shekhawat
frailty score devel­oped from age, CCI, and ECOG.8 Using this
pre­dic­tive score, patients con­sid­ered frail (score ≥2) had infe­rior
out­comes, espe­cially OS. This sim­pli­fied score was sub­se­quently
exter­nally val­i­dated by Stege et al16 using data from par­tic­i­pants
enrolled in HOVON87/NMSG18. For many cen­ters, lung func­tion
test­ing and for­mal frailty scor­ing (ie, Fried frailty phe­no­type) may
be out of scope of prac­tice. Notwithstanding, frailty scor­ing is
typ­i­cally col­lected by patients as self-report, and func­tional test­
ing such as gait speed requires no cost; these instru­ments are
quick, valid, and cost-effec­tive and inform sur­vival.
Rationale for con­sid­er­ing nontransplant strat­e­gies
in older patients with mye­loma
At this time, the opti­mal tool for assessing frailty for treat­ment
deci­sions (ie, trans­plant and nontransplant) has yet to be deter­
mined. Despite the wide­spread use of the IMWG frailty score
in clin­i­cal and research set­tings, it has its lim­i­ta­tions. Namely,
the use of chro­no­logic age can detract from the con­cept of
bio­logic/func­tional age, and CCI is likely to exclude mye­loma-
spe­cific comorbidities. Recommendations for lon­gi­tu­di­nal assess­
ments of fit­ness and frailty have been based on con­sen­sus expert
opin­ion.17-19 Although these fit­ ness/frailty scores have shaped
our under­stand­ing of the het­ero­ge­ne­ity among older adults with
mye­loma, the appli­ca­tion of such scores in the clin­i­cal set­ting
requires con­sid­er­ation of other fac­tors such as patient pref­er­
ences, life expec­tancy, and dis­ease biol­ogy.
For patients in whom can­di­dacy of trans­plant is a con­cern
due to frailty, there are a num­ber of options. The results from
a sin­gle-cen­ter, sin­gle-arm phase 2 study enroll­ing trans­plant-
inel­i­gi­ble adults with newly diag­nosed MM found the trip­let reg­
i­men of reduced-inten­sity lenalidomide, sub­cu­ta­ne­ous bortezo-
mib, and dexa­meth­a­sone to be tol­er­a­ble and effec­tive.20 Despite
62% of par­tic­i­pants expe­ri­enc­ing low-grade sen­sory neu­rop­a­thy,
Frailty in mul­ti­ple mye­loma  |  47

Figure 1. PFS and OS in matched population aged 64 to 70 years treated with or without ASCT, from myeloma XI subanalysis (with
permission).36
only 1 patient expe­ri­enced grade 3 or higher neu­rop­a­thy symp­
toms. After a 60-month fol­low-up, the median pro­gres­sion-free
sur­vival (PFS) was 41.9 months, and the median OS has still not
been reached. Based on this study’s find­ings, reduced-inten­sity
lenalidomide, sub­cu­ta­ne­ous bortezomib, and dexa­meth­a­
sone could be con­sid­ered for select trans­plant-inel­i­gi­ble older
adults who may be at higher risk for treat­ment-related neu­rop­
a­thy. The excel­lent out­comes achieved with the upfront use of
daratumumab, lenalidomide, and dexa­meth­a­sone (DRd) in this
patient pop­u­la­tion rival those achieved by induc­tion, trans­plant,
and main­te­nance. This rec­om­men­da­tion is supported by find­ings
from the MAIA trial,21 a phase 3 mul­ti­cen­ter study in which 44% of
enrolled par­tic­i­pants were 75 years or older and dem­on­strated
a lon­ger median PFS favor­ing the daratumumab-containing arm.
Among those treated with lenalidomide and dexa­ meth­ a­
sone
alone, the median PFS was 34.4 months vs not reached for those
who received daratumumab, lenalidomide, and dexa­meth­a­sone
48  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/46/1852113/46grant.pdf by guest on 13 December 2021
(HR, 0.53; P < .0001). The lat­est update of the MAIA study reported
an esti­mated 48-month PFS of 60%.22 There are numer­ous non-
transplant reg­i­mens19,23,24; other alter­na­tives could include daratu-
mumab, bortezomib, melphalan, prednisone (D-VMP); borezomib,
lenalidomide, dexamethasone (VRD); cyclophosphadmide, bor-
tezomib, dexamethasone (CyborD). The most com­mon induc­tion
strat­egy for those aged less than 65 to 74 years, in the United
States, is VRD front­line ther­apy.25 Studies are cur­rently under way
eval­u­at­ing whether fit­ness-based or risk-adapted ther­apy assign­
ments in mye­ loma improve dis­ ease- and patient-related out­
comes26-30 and to under­stand whether mye­loma ther­a­pies result
in lon­gi­tu­di­nal changes in an indi­vid­ual’s fit­ness/frailty sta­tus.
Rationale for con­sid­er­ing trans­plant in the older patients
with mye­loma
Autologous stem cell trans­plant (ASCT), even in the era of novel
agents, con­tin­ues to be a part of the ther­a­peu­tic par­a­digm with
clear improve­ments in PFS when incor­po­rated in the upfront set­
ting.31 Historically, the bulk of evi­dence to sup­port the ongo­ing use
of high-dose mel­pha­lan was gen­er­ated from tri­als enroll­ing youn­
ger patients, typ­i­cally 65 years or youn­ger, which excludes almost
two-thirds of patients with newly diag­nosed mye­loma.32 Although
Euro­pean guide­lines con­tinue to rec­om­mend lim­it­ing ASCT to pa-
tients youn­ger than 70 years, there is increas­ing evi­dence to sup­
port the safe use of ASCT in older patients, with empha­sis placed
on bio­log­i­cal age rather than chro­no­log­i­cal age.33,34
A meta-anal­y­sis attempting to quan­tify the ben­e­fit of upfront
ASCT in older patients showed a sig­nif­i­cant improve­ment in OS
(HR, 0.44; 95% CI, 0.34-0.58; P < .001), but authors commented
on the pau­city of evi­dence directly com­par­ing ASCT with non-
ASCT approaches in this age group.35 Another subanalysis of the

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/46/1852113/46grant.pdf by guest on 13 December 2021

large phase 3 mye­loma XI trial attempted to use an overlapping
cohort of age-matched patients (64-70 years) to com­pare out­
comes and tox­ic­ity for those enrolled in the ASCT and non-ASCT
arms.36 Importantly, the deci­sion to enroll in the trans­plant arm
was not ran­dom­ized but based on inves­ti­ga­tor dis­cre­tion and is
a cen­tral lim­i­ta­tion of this study. The study yielded a sig­nif­i­cant
advan­tage in favor of upfront ASCT with improve­ments in PFS
(HR, 0.41; P < .0001) and OS (HR, 0.51; P < .0001), after adjusting
for poten­tial base­line covariates (frailty, response to induc­tion)
(Figure 1). Interestingly, no dif­fer­ences in tox­ic­ity were observed
com­ pared with the cohort youn­ ger than 65 years who also
under­went ASCT. These results echo those reported by a large
anal­y­sis of 2092 patients 70 years or older from the Center for
International Blood and Marrow Transplant Research data­base,
show­ing iden­ti­cal out­comes for patients who were ­able to tol­er­
ate a 200-mg/m2 dose com­pared with youn­ger patients.34 Again,
no dif­fer­ences were seen in nonrelapse mor­tal­ity, but patients
who received dose reduc­ tions (140 
mg/m2) did have infe­ rior
sur­vival esti­ma­tes, thought sec­ond­ary to comorbidities such as
renal fail­ure or frailty. Although the evi­dence is not ideal, an open
dis­cus­sion about the pros and cons of ASCT is warranted, tak­ing
into account the patient’s own pref­er­ences and goals for treat­
ment. Patients with mye­loma in gen­eral do tend to pri­or­i­tize PFS
but also place high value on qual­ity of life, the side effects of
treat­ment, and poten­tial impact that treat­ments have on their
care­giv­ers.37,38

Establishing the risk/ben­e­fit ratio for an older patient with
mye­loma
For the treating cli­ ni­
cian, the chal­ lenge is how best to iden­
tify older patients who will derive the greatest ben­e­fit from up-
front ASCT. As discussed, avail­­able frailty scores can assist with
deci­sion mak­ing, pro­vid­ing an objec­tive assess­ment of the ini­
tial and chang­ing “fit­ness” of patients with mye­loma under­go­
ing ther­apy.17 Although con­sid­ered a stan­dard, it is impor­tant to
remem­ber that the IMWG Frailty Score was derived from a co-
hort of patients deemed to be trans­ plant inel­i­
gi­
ble.6 Similarly,
the recently published sim­pli­fied frailty score8 and UK Myeloma
Research Alliance Risk Profile were also derived and val­i­dated in
a cohort of trans­plant-inel­i­gi­ble patients.10 The R-MCI was derived
from a cohort includ­ing 343 patients 65 years or older and 383 who
under­went ASCT, but the exact over­lap between these 2 groups is
unclear.7 Nonetheless, given that those des­ig­nated as “frail” with
these mea­sures are more likely to expe­ri­ence sig­nif­i­cant toxicities
to induc­tion ther­apy, the addi­tional tox­ic­ity of an ASCT in those
iden­ti­fied as such would likely out­weigh the poten­tial ben­e­fit.6
Prakash Singh Shekhawat
Figure 2. Considerations when evaluating older patients for
consideration of ASCT. PS, performance status.
The hema­to­poi­etic cell trans­plan­ta­tion spe­cific comorbid-
ity index (HCT-CI) was orig­ i­
nally devel­
oped in the set­ ting of
allotransplant but sub­se­quently eval­u­ated in a cohort of 1156
patients with mye­loma and found to be pre­dic­tive of out­come.39
Patients with a score of 1 to 2 and more than 2 (com­pared with
HCT-CI of 0) or a Karnofsky performance status (KPS) less than
90 had infe­rior OS. It is note­wor­thy that treat­ment-related mor­
tal­ity was low at 2% and equiv­a­lent for patients with an HCT-CI
score of 0 or more than 2. The main cause of treat­ment-related
mor­tal­ity was dis­ease relapse and pro­gres­sion. Again, only 23%
of this cohort was 65 years or older, and even fewer were 70
years or older (n = 72, 6%). Regardless, in the absence of a bet­
ter alter­na­tive, it remains a recommended method of eval­u­at­ing
patients pretransplant, with sug­ges­tion to con­sider alter­na­tive
approaches in those with an HCT-CI more than 2 or KPS less
Frailty in mul­ti­ple mye­loma  |  49
Table 2. Relapsed mul­ti­ple mye­loma: early and late relapse stud­ies

Sel/Dex Melflufen/Dex Belantamab-mafadotin Idecabtagene

VPd (OPTIMISMM) Anti-CD38 Anti-SLAMF7 (STORM) (HORIZON OP-106) (DREAMM-2) vicleucel
KRd (ASPIRE) DVd (CASTOR) ERd (ELOQUENT-2) Sel/Pom/Dex Melflufen/Bort/Dex Belamaf/Pom/Dex Ciltacabtagene
(STOMP arm1) (ANCHOR OP-104) (ALGONQUIN) autoleucel
IRd (TOURMALINE-MM1) DRd (POLLUX) EPd (ELOQUENT-3) Sel/Carf/Dex Melflufen/Dara/Dex Belamaf/Bort/Dex Orvacabtagene
(STOMP arm6) (ANCHOR OP-104) (DREAMM-6armB) autoleucel
SVd (BOSTON) DKd (CANDOR) Sel/Dara/Dex P-BCMA-101
(STOMP arm5)
DPd (APOLLO)
Isa-Pd (ICARIA)
Isa-Kd (IKEMA)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/46/1852113/46grant.pdf by guest on 13 December 2021

Table mod­i­fied from Tables 2 to 5 in Nathwani et al.43
Carf,  carfilzomib; Dara, daratumumab; Dex, dexamethasone; DKd, daratumumab, carfilzomib, dexamethasone; DPd, daratumumab, pomalidomide,
dexamethasone; DVd, daratumumab, bortezomib, dexamethasone; EPd, elotuzumab, pomalidomide, dexamethasone; ERd, elotuzumab,
lenalidomide, dexamethasone; IRd, ixazomib, lenalidomide, dexamethasone; Isa, isatuximab; Pom, pomalidomide; Sel, selinexor; SVd, selinexor,
bortezomib, dexamethasone; VPd, bortezomib, pomalidomide, dexamethasone.

than 90.40,41 Other fac­tors that will also influ­ence the dis­cus­sion
around incor­po­ra­tion of ASCT include the biol­ogy of the dis­ease,
prog­nos­tic risk, response to induc­tion, and the wishes of the
patient and their social sit­ua
­ ­tion (Figure 2).
CLINICAL CASE (induc­t ion strat­e­g ies)
This patient had a nor­mal echo and pulmonary function test,
achieved a very good partial response with 3 cycles of induc­tion
with VRD, and had no high-risk cyto­ge­net­ics. Her ECOG was 1 and
KPS 80% at her pretransplant appoint­ment. Although her renal
func­tion improved, she still scored as inter­me­di­ate-fit according
to the R-MCI based on a glomerular filtration rate less than 60,
but HCT-CI was low risk based on the higher thresh­old for renal
func­tion (score 1). She had a sup­port­ive care­giver in good health
who had a solid under­stand­ing of the risks and ben­e­fits. She was
offered upfront ASCT given the low poten­tial risks of tox­ic­ity and
poten­tial ben­e­fits as high­lighted above. Nonetheless, after a long
dis­cus­sion about the rel­a­tive pros and cons, on the basis of per­
sonal pref­er­ence, she opted for a non-ASCT strat­egy.
For fit older patients who decline upfront ASCT, there are a
num­ber of options. The excel­lent out­comes achieved with the
upfront use of DRd in this patient pop­u­la­tion rival those achieved
by induc­tion, trans­plant, and main­te­nance. The lat­est update of
the MAIA study reported an esti­mated 48-month PFS of 60%.22
Other alter­na­tives could include D-VMP, VRD, or CyborD. The
most com­mon induc­tion strat­egy for those aged less than 65 to
74 years, in the United States, is VRD front­line ther­apy and was
sub­se­quently recommended for this patient.25
CLINICAL CASE (relapsed MM)
The patient tol­er­ated VRD induc­tion ther­apy and was in a very good
partial response and con­ tin­
ued on lenalidomide main­ te­
nance.

Table 3. Carfilzomib-based stud­ies: com­par­i­son of PFS and OS by age and fit­ness49

ASPIRE50 ENDEAVOR51 ARROW52

PFS <70 KRd 28.6 Fit KRd 31.4 <65 Kd NE* Fit Kd NE* <65 1 × Kd 12.2 Fit 1 × Kd 15.7
(median,
Rd 17.6 Rd 18.9 Vd 9.5 Vd 12.1 2 × Kd 5.6 2 × Kd 5.7
months)
≥70 KRd 23.8 Frail KRd 24.1 65-74 Kd 15.6 Frail Kd 18.7 65-74 1 × Kd 9.2 Frail 1 × Kd 10.3
Rd 15.9 Vd 9.5 Vd 6.6 2 × Kd 8.4 2 × Kd 6.6
Rd 16.0 ≥75 Kd 18.7 ≥75 1 × Kd 12.2
Vd 8.9 2 × Kd 9.5
OS <70 Not given Fit KRd 55.6 <65 Not given Fit Kd NE* <65 Not given Fit Not given
(median,
Rd 43.3 Vd 42.2
months)
≥70 Frail KRd 36.4 65-74 Frail Kd 33.6 65-74 Frail
Rd 26.2 ≥75 Vd 21.8 ≥75
*not estimable.
Frailty assess­ment based on the IMWG frailty index.6,53
Rd, lenalidomide and dexa­meth­a­sone; Vd, bortezomib and dexa­meth­a­sone.

50  |  Hematology 2021  |  ASH Education Program

 Table 4. Carfilzomib-based stud­ies: com­par­i­son of grade >3 treatment emergent adverse event (TEAE) and car­diac events by age and fit­ness49

ASPIRE50 ENDEAVOR51 ARROW52

Grade ≥3 TEAE, n (%) <70 KRd 236 (81.7) Fit KRd 102 (89) <65 Kd 152 (68.2) Fit Kd 91 (83) <65 1 × Kd 61 (59.2) Fit 1 × Kd 33 (55)
Rd 215 (77.6) Rd 96 (84) Vd 131 (63) Vd 76 (64) 2 × Kd 58 (56.3) 2 × Kd 41 (62)
≥70 KRd 92 (89.3) Frail KRd 86 (93) 65-74 Kd 124 (76.1) Frail Kd 142 (85) 65-74 1 × Kd 62 (68.9) Frail 1 × Kd 64 (81)
Rd 99 (88.4) Rd 94 (94) Vd 128 (69.9) Vd 125 (79) 2 × Kd 64 (63.4) 2 × Kd 42 (70)
≥75 Kd 63 (81.8) ≥75 1 × Kd 38 (84.4)
Vd 46 (70.8) 2 × Kd 23 (74.2)
Grade ≥3 car­diac fail­ure, <70 KRd 6 (2.1) Fit KRd 5 (4) <65 Kd 6 (2.7) Fit Kd 4 (4) <65 1 × Kd 1 (1) Fit 1 × Kd 1 (2)
n (%)
Rd 5 (1.8) Rd 2 (2) Vd 3 (1.4) Vd 2 (2) 2 × Kd 6 (5.8) 2 × Kd 1 (2)
≥70 KRd 9 (8.7) Frail KRd 9 (10) 65-74 Kd 8 (4.9) Frail Kd 15 (9) 65-74 1 × Kd 5 (5.6) Frail 1 × Kd 3 (4)
Rd 2 (1.8) Rd 1 (1) Vd 3 (1.6) Vd 7 (4) 2 × Kd 2 (2) 2 × Kd 5 (8)

Prakash Singh Shekhawat

≥75 Kd 8 (10.4) ≥75 1 × Kd 1 (2.2)
Vd 2 (3.1) 2 × Kd 2 (6.5)
Grade ≥3 ische­mic heart dis­ <70 KRd 8 (2.8) Fit KRd 4 (3) <65 Not given Fit Kd 2 (2) <65 1 × Kd 1 (1) Fit 1 × Kd 1 (1)
ease, n (%)
Rd 7 (2.5) Rd 3 (3) Vd 2 (<1) 2 × Kd 0 2 × Kd 0
≥70 KRd 5 (4.9) Frail KRd 7 (8) 65-74 Frail Kd 8 (5) 65-74 1 × Kd 0 Frail 1 × Kd 0
Rd 1 (0.9) Rd 1 (1) Vd 6 (4) 2 × Kd 1 (1) 2 × Kd 1 (2)
≥75 ≥75 1 × Kd 1 (2.2)
2 × Kd 1 (3.2)
6,54
Frailty assess­ment based on the IMWG frailty index.

Frailty in mul­ti­ple mye­loma  |  51

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/46/1852113/46grant.pdf by guest on 13 December 2021
She returned to volunteering and remained inde­pen­dent. After 3
years on main­te­nance ther­apy, at age 73 years, the patient devel­
oped acom­mu­nity-acquired pneu­mo­nia and was hos­pi­tal­ized for
1 week. She then reported progressing back pain and was iden­
ti­fied to have a new ver­te­bral com­pres­sion defor­mity. Serologic
test­ing revealed a rise in her κ light chains and immu­no­glob­u­lin G
κ immu­no­glob­u­lin pro­tein.
Relapsed MM
When MM relapses, this is a chal­leng­ing time for any patient and
is fur­ther con­founded if the patient’s health sta­tus is frail, clin­i­cal
health is ten­u­ous due to ill­ness, or there is impair­ment in end-
organ func­tion. Importantly, frailty is a dynamic state, and a reas-
sessment can fur­ther define next treat­ment deci­sions. Most stud­
ies sug­gest that health-related qual­ity of life tracks with dis­ease
con­ trol; there­ fore, addressing next steps for relapsed dis­ ease
would also improve health-related qual­ity of life. For relapsed MM,
most phase 3 ran­dom­ized con­trolled tri­als have com­pared the
effec­tive­ness and safety of 3-drug reg­i­mens to 2-drug reg­i­mens,
gen­er­ally depicting PFS ben­e­fits of 3-drug com­bi­na­tions over 2-
drug com­bi­na­tions. For fit patients, a 3-drug reg­i­men has become
the stan­dard of care for relapsed dis­ease. For patients who are frail
according to stan­dard­ized cri­te­ria (eg, IMWG), data are quite lim­
ited on tol­er­ance of relapsed ther­a­pies. Age-strat­i­fied sub­group
anal­y­sis is not an indi­ca­tor of health sta­tus and should not be used
as a sur­ro­gate of tol­er­ance. In this patient sce­nario, a thought­ful
explo­ra­tion of options to per­son­al­ize ther­apy is indi­cated.
Treatment options for relapsed dis­ease can be tai­lored based
on dis­ease biol­ogy, tempo of dis­ease relapse, tol­er­ance of prior
ther­a­pies, health sta­tus, and shared deci­sion mak­ing.42 Drug clas­
ses include immunomodulatory drugs (lenalidomide, pomalido-
mide), proteasome inhibitors (bortezomib, carfilizomib, ixazomib),
mono­clo­nal antibodies (daratumumab, isatuximab, elotuzumab),
B cell maturation antigen tar­gets (balantamab-mafodotin), novel
ther­apy (histone deacetylase inhib­i­tors, selec­tive inhib­i­tors of
nuclear export), or alkylator ther­ apy (cyclo­ phos­pha­mide, mel­
pha­lan, mel­pha­lan flufenamide [pep­tide con­ju­gate]) with ste­roids
(dexa­meth­a­sone, pred­ni­sone) in var­i­ous com­bi­na­tions (Table 2).
Class switching from first-line treat­ment to sec­ond-line ther­apy
is recommended using a dif­fer­ent mech­an ­ ism of action not pre­vi­
ously used. Among frail patients, if 2-drug ther­apy was offered at
induc­tion, such as lenalidomide and dexa­meth­a­sone or bortezo-
mib and dexa­meth­a­sone, if frailty remains or intol­er­ance devel­ops,
the patients could read­ily switch to the alter­nate 2-drug class.
Consistently across stud­ies, dexa­meth­a­sone is dose atten­u­ated
by advanc­ing age and is recommended for bet­ter tol­er­ance. Ste-
roids are a back­bone of mye­loma ther­apy but are asso­ci­ated with
many multiorgan side effects that can affect qual­ity of life and
adher­ence, par­tic­u­larly for older adults.44 Steroid dose atten­u­a­
tions are well char­ac­ter­ized in MM, and low-dose dexa­meth­a­sone
is recommended due to infe­rior sur­vival with higher-dose dexa­
meth­a­sone.45 More recent lit­er­a­ture exam­ined lower dos­ages of
lenalidomide with early dis­con­tin­u­a­tion of dexa­meth­a­sone that
was proven safe and resulted in sim­i­lar out­comes to patients on
con­tin­u­ous lenalidomide-dexa­meth­a­sone.46 Unlike many front­line
stud­ies with ste­roid atten­u­a­tions, lim­ited pro­spec­tive data are
avail­­able for dose atten­u­a­tions of stan­dard relapsed ther­a­pies for
frail patients.
52  |  Hematology 2021  |  ASH Education Program
Most patients ben­e­fit from a mono­clo­nal anti­body, in com­
bi­na­tion, at first relapse. Daratumaumb, an anti-CD38 mono­clo­
nal anti­body, can be offered in com­bi­na­tion with a PI or IMiD at
first relapse or as monotherapy among dis­ease that is dou­ble-
refrac­tory or fol­low­ing third line of treat­ment. Triple-drug ther­apy
for frail patients at relapse may require dose mod­i­fi­ca­tions. As 1
exam­ple, in the ICARIA study,47 eval­u­at­ing isatuximab, pomalido-
mide, and dexa­meth­a­sone vs pomalidomide and dexa­meth­a­sone,
43% of patients required dose reduc­tions of pomalidomide and
33% of patients required dose reduc­tions in dexa­meth­a­sone with
tri­ple ther­apy. However, patients ran­dom­ized to the isatuximab,
pomalidomide, and dexa­meth­a­sone arm had a sub­stan­tially lon­
ger dura­tion of ther­apy com­pared with pomalidomide and dexa­
meth­a­sone (41 vs 24 weeks). In gen­eral, no age- or frailty-based
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/46/1852113/46grant.pdf by guest on 13 December 2021
dose atten­u­a­tions are recommended for mono­clo­nal anti­body
ther­apy.48 In terms of under­ly­ing health sta­tus, par­tic­u­larly rel­e­
vant are periph­eral neu­rop­a­thy, car­dio­vas­cu­lar health, throm­bo­
sis, recur­rent infec­tions, bone mar­row reserve, and pul­mo­nary
health. Cardiovascular events are of sig­nif­i­cant inter­est in aging
adults, par­ tic­u­larly as this applies to carfilizomib-based strat­e­
gies. In a post hoc anal­y­sis,49 eval­u­at­ing 3 carfilizomib-based tri­als
(ASPIRE,50 ENDEAVOR,51 ARROW52), 25% to 35% of patients across
stud­ies were cat­e­go­rized as frail using IMWG cri­te­ria. OS was
reduced, and treat­ ment-emer­gent adverse events were more
com­mon among frail patients (Table 3), includ­ing increased car­
diac fail­ure events (Table 4).49 In a pro­spec­tive study eval­u­at­ing
95 patients receiv­ing a PI (bortezomib or carfilizomib) in relapsed
dis­ease, more than half of patients expe­ri­enced grade 3 car­diac
events, and most events occurred within the first 3 months of
ther­apy.54 Moreover, patients who expe­ri­enced a car­dio­vas­cu­lar
event had infe­rior PFS and worse OS.54 A global assess­ment of
health and reassessment of frailty can fur­ther define next treat­
ment options.
CLINICAL CASE (con­clu­sion)
The patient recov­ered from her hos­pi­tal admis­sion. Recalcula-
tion of the IMWG score remained inter­me­di­ate-fit. The patient
was started on a daratumumab, bortezomib, and dose-reduced
dexa­meth­a­sone for relapsed MM. Special atten­tion to neu­ro-
p­a­thy was tracked closely. The patient benefited from phys­i­cal
ther­apy to assist with recov­ery from back pain and tol­er­ated
ther­apy with­out dif­fi­culty.
The mul­ti­ply­ing ther­a­pies of mye­loma have resulted in chal­
leng­ing clin­i­cal deci­sions. Moreover, the future of MM treat­ment
deci­sions will soon include not only trans­plant deci­sion mak­ing
but also cel­lu­lar ther­apy can­di­dacy. How to best pre­scribe these
ther­a­pies for aging adults has yet to be deter­mined. To rec­on­
cile dif­fer­ences in health with aging, par­tic­u­larly for com­plex MM
deci­sions, a com­pre­hen­sive eval­u­a­tion of health is required. The
MM com­ mu­nity has embraced sev­ eral frailty scor­ ing sys­
tems,
each with lim­i­ta­tions, yet impor­tantly, these impor­tant first steps
of embed­ding assess­ments in clin­ic ­ al tri­als are a start. In sum­
mary, mye­loma treat­ments can be mod­i­fied or informed by frailty
scor­ing sys­tems, and these tools dem­on­strate how geri­at­ric prin­
ci­ples can be incor­po­rated into risk strat­i­fi­ca­tion to improve out­
comes.
Conflict-of-inter­est dis­clo­sure
Shakira J. Grant: no conflicts to disclose.
Ciara L. Freeman received research funding from Lundbeck.
Ashley E. Rosko: no conflicts to disclose.
Off-label drug use
Shakira J. Grant: nothing to disclose.
Ciara L. Freeman: nothing to disclose.
Ashley E. Rosko: nothing to disclose.
Correspondence
Ashley E. Rosko, Division of Hematology, The Ohio State Univer-
sity, 1150C Lin­coln Tower, 1800 Cannon Dr, Colum­bus, OH 43210;
e-mail: ashley​­.rosko@osumc​­.edu.
References
1. Wildes TM, Tuchman SA, Klepin HD, et  al. Geriatric assess­ment in older
adults with mul­ti­ple mye­loma. J Am Geriatr Soc. 2019;67(5):987-991.
2. Hshieh TT, Driver JA, Abel GA. Cognitive impair­ment among older patients
with hema­to­logic can­cers—reply. JAMA Oncol. 2018;4(12):1784.
3. Hamaker ME, Prins MC, Stauder R. The rel­e­vance of a geri­at­ric assess­
ment for elderly patients with a haematological malig­nancy—a sys­tem­atic
review. Leuk Res. 2014;38(3):275-283.
4. Kuroda J, Shimura Y, Ohta K, et al; Kansai Myeloma Forum Investigators.
Limited value of the inter­na­tional stag­ing sys­tem for predicting long-term
out­come of trans­plant-inel­i­gi­ble, newly diag­nosed, symp­tom­atic mul­ti­ple
mye­loma in the era of novel agents. Int J Hematol. 2014;99(4):441-449.
5. Pawlyn C, Cairns D, Kaiser M, et  al. The rel­a­tive impor­tance of fac­tors pre-
dicting out­come for mye­loma patients at dif­fer­ent ages: results from 3894
patients in the Myeloma XI trial. Leukemia. 2020;34(2):604-612.
6. Palumbo A, Bringhen S, Mateos M-V, et  al. Geriatric assess­ ment pre­
dicts sur­vival and toxicities in elderly mye­loma patients: an International
Myeloma Working Group report. Blood. 2015;125:2068-2074.
7. Engelhardt M, Domm A-S, Dold S-M, et  al. A con­cise revised Myeloma
Comorbidity Index as a valid prog­nos­tic instru­ment in a large cohort of 801
mul­ti­ple mye­loma patients. Haematologica. 2017;102(5):910-921.
8. Facon T, Dimopoulos MA, Meuleman N, et  al. A sim­ pli­fied frailty scale
pre­dicts out­comes in trans­plant-inel­i­gi­ble patients with newly diag­
nosed mul­ti­ple mye­loma treated in the FIRST (MM-020) trial. Leukemia.
2020;34(1):224-233.
9. Oken MM, Creech RH, Tormey DC, et al. Toxicity and response cri­te­ria of
the Eastern Cooperative Oncology Group. Am J Clin Oncol. 1982;5(6):649-
655.
10. Cook G, Royle K-L, Pawlyn C, et  al. A clin­i­cal pre­dic­tion model for out­
come and ther­apy deliv­ery in trans­plant-inel­i­gi­ble patients with mye­loma
(UK mye­loma research alli­ance risk pro­file): a devel­op­ment and val­i­da­tion
study. Lancet Haematol. 2019;6(3):e154-e166.
11. Milani P, Vincent Rajkumar S, Merlini G, et  al. N-ter­mi­nal frag­ment of the
type-B natri­uretic pep­tide (NT-proBNP) con­trib­utes to a sim­ple new frailty
score in patients with newly diag­nosed mul­ti­ple mye­loma. Am J Hematol.
2016;91(11):1129-1134.
12. Offidani M, Corvatta L, Polloni C, et  al. Assessment of vul­ner­a­bil­ity mea­
sures and their effect on sur­vival in a real-life pop­u­la­tion of mul­ti­ple mye­
loma patients reg­is­tered at Marche Region Multiple Myeloma Registry. Clin
Lymphoma Myeloma Leuk. 2012;12(6):423-432.
13. Fried LP, Tangen CM, Walston J, et al; Cardiovascular Health Study Collab-
orative Research Group. Frailty in older adults: evi­dence for a phe­no­type.
J Gerontol A Biol Sci Med Sci. 2001;56(3):M146-M156.
14. Benboubker L, Dimopoulos MA, Dispenzieri A, et al; FIRST Trial Team. Lena-
lidomide and dexa­meth­a­sone in trans­plant-inel­i­gi­ble patients with mye­
loma. N Engl J Med. 2014;371(10):906-917.
15. Facon T, Dimopoulos MA, Dispenzieri A, et al. Final anal­y­sis of sur­vival out­
comes in the phase 3 FIRST trial of up-front treat­ment for mul­ti­ple mye­
loma. Blood. 2018;131(3):301-310.
16. Stege CAM, Van Der Holt B, Dinmohamed AG, et al. Validation of the FIRST
sim­pli­fied frailty scale using the ECOG per­ for­
mance sta­ tus instead of
patient-reported activ­i­ties. Leukemia. 2020;34(7):1964-1966.
17. Engelhardt M, Ihorst G, Duque-Afonso J, et  al. Structured assess­ ment
of frailty in mul­ ti­
ple mye­loma as a par­ a­
digm of indi­ vid­ u­
al­
ized treat­
Prakash Singh Shekhawat
ment algo­ rithms in can­ cer patients at advanced age. Haematologica.
2020;105(5):1183-1188.
18. Larocca A, Dold SM, Zweegman S, et  al. Patient-cen­ tered prac­tice in
elderly mye­loma patients: an over­view and con­sen­sus from the Euro­pean
Myeloma Network (EMN). Leukemia. 2018;32(8):1697-1712.
19. Grant SJ, Mian HS, Giri S, et al. Transplant-inel­i­gi­ble newly diag­nosed mul­
ti­ple mye­loma: cur­rent and future approaches to clin­i­cal care: a Young
International Society of Geriatric Oncology review paper. J Geriatr Oncol.
2021;12(4):499-507.
20. O’Donnell EK, Laubach JP, Yee AJ, et al. A phase 2 study of mod­i­fied lena-
lidomide, bortezomib and dexa­meth­a­sone in trans­plant-inel­i­gi­ble mul­ti­ple
mye­loma. Br J Haematol. 2018;182(2):222-230.
21. Facon T, Kumar S, Plesner T, et al; MAIA Trial Investigators. Daratumumab
plus lenalidomide and dexa­ meth­ a­
sone for untreated mye­ loma. N Engl
J Med. 2019;380(22):2104-2115.
22. Kumar SK, Facon T, Usmani SZ, et al. Updated anal­y­sis of daratumumab plus
lenalidomide and dexa­meth­a­sone (D-Rd) ver­sus lenalidomide and dexa­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/46/1852113/46grant.pdf by guest on 13 December 2021
meth­a­sone (Rd) in patients with trans­plant-inel­i­gi­ble newly diag­nosed
mul­ti­ple mye­loma (NDMM): the phase 3 MAIA study. Blood. 2020;136:24-
26.
23. Moreau P, Touzeau C, Vij R, Goldsmith SR, Rosko AE. Newly diag­nosed
mye­loma in 2020. Am Soc Clin Oncol Educ Book. 2020;40:1-15.
24. Berbari HE, Kumar SK. Initial ther­a­peu­tic approaches to patients with mul­ti­ple
mye­loma. Adv Ther. 2021;38(7):3694-3711.
25. Lee HC, Ailawadhi S, Gasparetto C, et al. Treatment pat­terns and out­comes
in elderly patients with newly diag­nosed mul­ti­ple mye­loma: results from
the Connect® MM Registry. Blood Cancer J. 2021;11(7):134.
26. Cairns D, Pawlyn C, Royle K-L, et al. Frailty-adjusted ther­apy in trans­plant
non-eli­gi­ble patients with newly diag­nosed mul­ti­ple mye­loma (FiTNEss): a
UK mye­loma research alli­ance study, mye­loma XIV. Blood. 2019;134(suppl
1):3153.
27. PETHEMA Foundation. Treatment for elderly fit newly diag­nosed mul­ti­
ple mye­loma patients aged between 65 and 80 years. Accessed 1 Apr
2021. https:​­//­ClinicalTrials​­.gov​­/show​­/NCT03742297
28. Onlus FEI. Study to deter­mine the effi­cacy and safety of stan­dard sched­
ule ver­sus a new algo­rithm of dose reduc­tions in elderly and unfit newly
diag­nosed mul­ti­ple mye­loma patients receiv­ing lenalidomide plus ste­
roids. Accessed 1 Apr 2021. https:​­/​­/ClinicalTrials​­.gov​­/show​­/NCT02215980
29. Leeds Uo, UK CR, Takeda, et al. Myeloma XIV: frailty-adjusted ther­apy in
trans­plant non-eli­gi­ble patients with newly diag­nosed mul­ti­ple mye­loma.
Accessed 1 Apr 2021. https:​­/​­/ClinicalTrials​­.gov​­/show​­/NCT03720041
30. Stege CAM, Nasserinejad K, van der Spek E, et al. Efficacy and tol­er­a­bil­ity
of ixazomib, daratumumab and low dose dexa­meth­a­sone (Ixa Dara dex) in
unfit and frail newly diag­nosed mul­ti­ple mye­loma (NDMM) patients; results
of the interim effi­cacy anal­y­sis of the phase II HOVON 143 study. Blood.
2019;134(suppl 1):695.
31. Al Hamed R, Bazarbachi AH, Malard F, Harousseau J-L, Mohty M. Current
sta­tus of autol­o­gous stem cell trans­plan­ta­tion for mul­ti­ple mye­loma. Blood
Cancer J. 2019;9(4):44.
32. Mina R, Bringhen S, Wildes TM, Zweegman S, Rosko AE. Approach to
the older adult with mul­ti­ple mye­loma. Am Soc Clin Oncol Educ Book.
2019;93(39):500-518.
33. Dimopoulos MA, Moreau P, Terpos E, et al; EHA Guidelines Committee. Mul-
tiple mye­loma: EHA-ESMO Clinical Practice Guidelines for diag­no­sis, treat­
ment and fol­low-up. Ann Oncol. 2021;32(3):309-322.
34. Munshi PN, Vesole D, Jurczyszyn A, et  al. Age no bar: a CIBMTR anal­y­sis
of elderly patients under­go­ing autol­o­gous hema­to­poi­etic cell trans­plan­ta­
tion for mul­ti­ple mye­loma. Cancer. 2020;126(23):5077-5087.
35. Mian H, Mian OS, Rochwerg B, Foley R, Wildes TM. Autologous stem
cell trans­plant in older patients (age ≥ 65) with newly diag­nosed mul­ti­
ple mye­ loma: a sys­ tem­atic review and meta-anal­ y­ sis. J Geriatr Oncol.
2020;11(1):93-99.
36. Pawlyn C, Cairns D, Menzies T, et al. Autologous stem cell trans­plan­ta­tion is
safe and effec­tive for fit older mye­loma patients: explor­atory results from
the Myeloma XI trial [published online 3 Decem­ber 2020]. Haematologica.
37. Terpos E, Mikhael J, Hajek R, et  al. Management of patients with mul­ti­
ple mye­loma beyond the clin­i­cal-trial set­ting: under­stand­ing the bal­ance
between effi­cacy, safety and tol­er­a­bil­ity, and qual­ity of life. Blood Cancer
J. 2021;11(2):40.
38. Fifer S, Galinsky J, Richard S. Myeloma patient value map­ping: a dis­crete
choice exper­i­ment on mye­loma treat­ment pref­er­ences in the UK. Patient
Prefer Adherence. 2020;14:1283-1293.
Frailty in mul­ti­ple mye­loma  |  53
39. Saad A, Mahindra A, Zhang M-J, et al. Hematopoietic cell trans­plant comor-
bidity index is pre­dic­tive of sur­vival after autol­o­gous hema­to­poi­etic cell
trans­plan­ta­tion in mul­ti­ple mye­loma. Biol Blood Marrow Transplant.
2014;20(3):402-408.e1408e1.
40. Shah N, Callander N, Ganguly S, et al; Amer­i­can Society for Blood and Mar-
row Transplantation. Hematopoietic stem cell trans­plan­ta­tion for mul­ti­ple
mye­loma: guide­lines from the Amer­i­can Society for Blood and Marrow
Transplantation. Biol Blood Marrow Transplant. 2015;21(7):1155-1166.
41. Rosko A, Giralt S, Mateos M-V, Dispenzieri A. Myeloma in elderly patients:
when less is more and more is more. Am Soc Clin Oncol Educ Book.
2017;37:575-585.
42. Bazarbachi AH, Al Hamed R, Malard F, Harousseau J-L, Mohty M. Relapsed
refrac­tory mul­ti­ple mye­loma: a com­pre­hen­sive over­view. Leukemia.
2019;33(10):2343-2357.
43. Nathwani N, Bertamini L, Banerjee R, Gay F, Shah N, Krishnan A. When and
how to treat relapsed mul­ti­ple mye­loma. Am Soc Clin Oncol Educ Book.
2021;41:358-375.
44. Faiman B, Bilotti E, Mangan PA, Rogers K; IMF Nurse Leadership Board.
Steroid-asso­ci­ated side effects in patients with mul­ti­ple mye­loma: con­
sen­sus state­ment of the IMF Nurse Leadership Board. Clin J Oncol Nurs.
2008;12(3, suppl):53-63.
45. Rajkumar SV, Jaco­bus S, Callander NS, et al; Eastern Cooperative Oncology
Group. Lenalidomide plus high-dose dexa­meth­a­sone ver­sus lenalidomide
plus low-dose dexa­meth­a­sone as ini­tial ther­apy for newly diag­nosed mul­
ti­ple mye­loma: an open-label randomised con­trolled trial. Lancet Oncol.
2010;11(1):29-37.
46. Larocca A, Bonello F, Gaidano G, et al. Dose/sched­ule-adjusted Rd-R vs con­
tin­uo
­ us Rd for elderly, inter­me­di­ate-fit patients with newly diag­nosed mul­ti­
ple mye­loma. Blood. 2021;137(22):3027-3036.
47. Attal M, Richardson PG, Rajkumar SV, et al; ICARIA-MM study group. Isat-
uximab plus pomalidomide and low-dose dexa­meth­a­sone ver­sus poma-
lidomide and low-dose dexa­ meth­ as­one in patients with relapsed and
refrac­ tory mul­ ti­
ple mye­ loma (ICARIA-MM): a randomised, multicentre,
open-label, phase 3 study. Lancet. 2019;394(10214):2096-2107.
54  |  Hematology 2021  |  ASH Education Program
48. Wildes TM, Anderson KC. Approach to the treat­ment of the older, unfit
patient with mye­loma from diag­no­sis to relapse: per­spec­tives of a US
hema­tol­o­gist and a geri­at­ric hema­tol­o­gist. Hematology Am Soc Hematol
Educ Program. 2018;2018(1):88-96.
49. Facon T, Niesvizky R, Mateos M-V, et  al. Efficacy and safety of carfilzo-
mib-based reg­i­mens in frail patients with relapsed and/or refrac­tory mul­ti­
ple mye­loma. Blood Adv. 2020;4(21):5449-5459.
50. Dimopoulos MA, Stewart AK, Masszi T, et al. Carfilzomib, lenalidomide, and
dexa­meth­a­sone in patients with relapsed mul­ti­ple mye­loma categorised
by age: sec­ond­ary anal­y­sis from the phase 3 ASPIRE study. Br J Haematol.
2017;177(3):404-413.
51. Dimopoulos MA, Goldschmidt H, Niesvizky R, et al. Carfilzomib or borte-
zomib in relapsed or refrac­tory mul­ti­ple mye­loma (ENDEAVOR): an interim
over­all sur­vival anal­y­sis of an open-label, randomised, phase 3 trial. Lancet
Oncol. 2017;18(10):1327-1337.
52. Mateos MV, Moreau P, Berenson JR, et  al. Once-weekly vs twice-weekly
carfilzomib (K) dos­ing plus dexa­meth­a­sone (d) in patients with relapsed
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/46/1852113/46grant.pdf by guest on 13 December 2021
and refrac­tory mul­ti­ple mye­loma (RRMM): results of the ran­dom­ized phase 3
study ARROW. J Clin Oncol. 2018;19(7):953-964.
53. Moreau P, Stewart KA, Dimopoulos M, et al. Once-weekly (70 mg/m2) vs
twice-weekly (56 mg/m2) dos­ing of carfilzomib in patients with relapsed
or refrac­tory mul­ti­ple mye­loma: a post hoc anal­y­sis of the ENDEAVOR,
ARROW, and CHAMPION-1 tri­als. Cancer Med. 2020;9(9):2989-2996.
54. Cornell RF, Ky B, Weiss BM, et al. Prospective study of car­diac events dur­
ing proteasome inhib­i­tor ther­apy for relapsed mul­ti­ple mye­loma. J Clin
Oncol. 2019;37(22):1946-1955.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000231
 CLL: EXTENDING SURVIVAL

Upfront therapy: the case for continuous

treatment

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/55/1851744/55tam.pdf by guest on 13 December 2021

Constantine S. Tam
Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia; University of Melbourne, Parkville, Victoria, Australia; and The Royal Melbourne
Hospital, Parkville, Victoria, Australia

Both BTKi and BCL2i are regarded as standards of care for frontline treatment of CLL. In this paper, I present the argu-
ments for favoring BTKi as initial therapy. Venetoclax-based regimens have the advantage of being fixed in duration, but
patients with select high-risk features may experience inferior PFS relative to those without high-risk features.

LEARNING OBJECTIVES
• Understand that both BTK inhibitors (BTKi) and the BCL2 antagonist venetoclax are effective treatments in
frontline CLL
• Understand that BTKis are active across the breadth of prognostic subgroups but are associated with long­term
toxicities and the need for indefnite therapy

median progression­free survival (PFS) of 52 months and

CLINICAL CASE
A 71­year­old man was observed for chronic lymphocytic
leukemia (CLL) for 2 years and progressed with doubling
time of 6 months and bulky lymphadenopathy. He has
stable hypertension, atrial fbrillation, and osteoarthri­
tis. Prognostic workup showed del(17p) by fluorescence
in situ hybridization, unmutated immunoglobulin heavy
chain (IgHV), and TP53 mutation with complex karyotype.
He lives in a rural location and travels 1.5 hours by car to
see his treating oncologist. Should he receive upfront con­
tinuous BTK inhibitor (BTKi) therapy or limited­duration
venetoclax­obinutuzumab?
BTKis and the paradigm for continuous therapy
Therapy for CLL in the chemotherapy era is traditionally
limited in duration due to cumulative myelosuppression
and other side effects.1 This paradigm is now challenged by
the advent of targeted therapies with improved tolerance
profles. In this article, we discuss the advantages of contin­
uous BTKi therapy relative to limited­duration alternatives.
Eight years ago, the phase 1 report of the frst­in­class
BTKi ibrutinib was published in the New England Journal
of Medicine.2 Ibrutinib was groundbreaking for patients
with relapsed/refractory (R/R) CLL, achieving prolonged
7­year overall survival (OS) of 55%,3 compared with historic
median OS expectations of less than 1 year for similar pop­
ulations.4,5 Since then, new generations of BTKis, includ­
ing more specifc covalent inhibitors with the potential for
reduced side effects (eg, acalabrutinib,6 zanubrutinib7) and
even newer “reversible” agents (eg, pirtobrutinib8), have
emerged. Multiple phase 3 studies have confrmed the
superiority of BTKi­based therapies over established com­
parators in CLL, in both the frontline (1L) (RESONATE­2,9
ELEVATE­TN,10 ILLUMINATE,11 E1912,12 ALLIANCE13) and R/R
settings (RESONATE,14 ASCEND15).
Although broadly effective and well tolerated as a class,
the paradigm for BTKi therapy is indefnite therapy due to
incomplete eradication of the CLL population, as manifest
by low complete remission rates of approximately 25% to
35% and approximately 10% in the 1L and R/R settings,3,11,16,17
respectively, and largely absent chances of attaining unde­
tectable minimal residual disease status (uMRD; defned as
<1 CLL cell per 10,000 leukocytes). The addition of the anti­
CD20 rituximab to ibrutinib failed to improve PFS in 2 ran­
domized studies in CLL,13,18 although a slight PFS advantage
to the combination of acalabrutinib and obinutuzumab to
acalabrutinib alone was seen in the ELEVATE­TN study.10,19
Thus, the nature of BTKi­based therapy remains continuous
and indefnite.

Prakash Singh Shekhawat

Upfront therapy: the case for continuous treatment | 55
The alter­na­tive: BCL2 inhib­i­tors and lim­ited-dura­tion
ther­apy
Over the sim­i­lar period, the BCL2 inhib­i­tor venetoclax was devel­
oped.20 Similar to ibrutinib, venetoclax showed remark­able sin­
gle-agent activ­ity in patients with heavily pretreated CLL, with
median PFS of 30 months and 3-year OS of 71%.21 Different from
BTKi, venetoclax is capa­ble of attaining uMRD as monotherapy in
R/R CLL (27%–30% in blood, 16% in mar­row21,22).
The obser­va­tion that some patients in uMRD remis­sions were
­able to main­tain dura­ble responses off ther­apy23,24 led to the design
of time-lim­ited reg­i­mens such as the 24-month venetoclax-rituximab
(Ven-R) reg­ i­
men in the MURANO study of R/R CLL25,26 and the
12-month venetoclax-obinutuzumab (Ven-O) reg­im ­ en in the CLL14
study of 1L CLL.27 These reg­i­mens have the appeal of pro­vid­ing a
rel­a­tively short expo­sure to ther­a­peu­tic agents and their atten­dant
side effects.
However, emerg­ ing data sug­ gest that lim­ ited-dura­ tion
venetoclax reg­i­mens may not be suit­able in all­types of CLL, with
par­tic­ul­ar con­cern for patients with geno­mic high-risk dis­ease. In
the MURANO study, com­pared with the whole-study uMRD rate
of 84% fol­low­ing Ven-R, uMRD rates were 42% for patients with
del(17p), 55% for com­plex kar­yo­type (CK), 48% for TP53 muta­
tion, and 40% to 43% for muta­tions in NOTCH1, XPO1, or BRAF.26
Among patients who responded to Ven-R, 25 patients have
been rechallenged after ini­tial remis­sions of 12 or more months.28
Genomic ana­ly­ses com­par­ing serial sam­ples taken at base­line
(before Ven-R) and at retreatment showed selec­tion for del(17p)
and CK, and patients car­ry­ing these clones were highly unlikely
to reach uMRD from venetoclax reexposure.28
Similarly, in updated data sets of CLL14, patients with IgHV-
unmutated sta­tus, del(17p), or TP53 muta­tions expe­ri­enced shorter
PFS fol­low­ing Ven-O (haz­ard ratio [HR], 2.14, 3.19, and 2.42, respec­
tively), and del(17p) was addi­tion­ally asso­ci­ated with reduced OS
fol­low­ing Ven-O (HR, 3.52).29 Interestingly, clonal regrowth rates
were faster in patients with IgHV-unmutated CLL,30 suggesting
that con­tin­u­ous “sup­pres­sive” BTKi ther­apy may be pref­er­a­ble
in these patients.
BTKis are broadly effec­tive in high geno­mic risk CLL
The ear­li­est expe­ri­ence with ibrutinib in the PCYC-1102 phase
1b/2 trial suggested that del(17p) and/or CK may asso­ci­ated
with infe­rior out­comes: com­pared with the whole-study R/R
CLL median PFS (mPFS) of 52 months, mPFS was 26 months for
del(17p) and 31 months for CK.3 However, the adverse risk in the
CK sub­group was entirely attrib­ut­­able to over­lap with del(17p),
as patients with CK alone (with­out del(17p)) had a favor­able mPFS
of 88 months. In com­par­ing the results of this study with those
reported in sub­se­quent BTKi stud­ies, it is impor­tant to note that
the patients in PCYC-1102 were very heavily pretreated (59% had
≥4 prior lines of ther­apy3) and may be less rep­re­sen­ta­tive of pa­
tients seen in clinic today. Indeed, a clear rela­tion­ship between
increas­ing prior lines of ther­apy and infe­rior PFS has been clearly
established across mul­ti­ple ibrutinib stud­ies.3,31,32
The phase 3 RESONATE study com­pared ibrutinib and ofatu­
mumab in a less heavily pretreated R/R CLL pop­u­la­tion.14 Del(17p),
TP53 muta­tion, and CK were pres­ent in 32%, 51%, and 25% of
patients on the ibrutinib arm, respec­tively.31 Patients with del(17p)
and/or TP53 muta­tion con­tin­ued to show a slight dis­ad­van­tage
(mPFS of 41 months) com­pared with other sub­groups, but no PFS
dif­fer­ences were observed in patients with and with­out CK.31
56  |  Hematology 2021  |  ASH Education Program
Results of ibrutinib in 1L CLL with adverse geno­mics were even
more favor­able. In the long-term fol­low-up of the National Insti­
tutes of Health phase 2 study of 34 patients with TP53 muta­tions
receiv­ing 1L ibrutinib ther­apy, 6-year PFS was 61%.33 When pooled
with the results of 3 other 1L stud­ies (RESONATE-2, ECOG 1912,
and ILLUMINATE), 89 patients in total with TP53 aber­rant CLL were
treated 1L on ibrutinib-based reg­i­mens; in aggre­gate, a 4-year
PFS of 79% was reported.34 Balancing these favor­able reports, the
ALLIANCE 1L study of BR vs ibrutinib vs ibrutinib-rituximab found
a less favor­able PFS for patients with del(17p), with a 2-year PFS
of approx­i­ma­tely 75% for ibrutinib reg­i­mens; the cor­re­spond­ing
2-year PFS for patients with CK was approx­i­ma­tely 90%.13
Next-gen­ er­
a­
tion cova­ lent BTKis such as acalabrutinib and
zanubrutinib show improved tol­er­ance com­pared with ibrutinib in
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/55/1851744/55tam.pdf by guest on 13 December 2021
head-to-head stud­ies.35-37 In one of the larg­est pro­spec­tive series of
patients with del(17p) reported to date (SEQUOIA Arm C, n = 109),
1L treat­ment with zanubrutinib resulted in and over­all response
rate of 95% and an 18-month PFS of 91%.38 Importantly, for those
patients with the highest risk cat­e­gory of del(17p) and con­com­i­
tant CK, a favor­able 18-month PFS of 94% was maintained.38 Simi­
larly favor­able 1L results for TP53 aber­rant CLL were reported in the
ELEVATE-TN study of acalabrutinib vs acalabrutinib-obinutuzumab
vs chlorambucil-obinutuzumab, with an 18-month PFS of more
than 80% for acalabrutinib arms in patients with del(17p) and/or
TP53 muta­tion.10
Finally, all­BTKi stud­ies to date showed no dis­ad­van­tage for
patients with IgHV-unmutated status, with con­sis­tently overlap­
ping PFS curves irrespective of set­ting (1L or R/R), TP53 sta­tus,
or agent stud­ied.10-13,15,17,31,33,38
In aggre­gate, avail­­able data across the cova­lent BTKi expe­ri­
ence suggested that the impact of adverse geno­mics (par­tic­u­
larly TP53 aber­ra­tions) is lim­ited mainly to patients with heavily
pretreated CLL, with largely favor­able out­comes observed in
phase 2 to 3 stud­ies of patients treated in the 1L.10,13,33,34,38 Con­
ceptually, one can argue the­o­ret­i­cally that the patients with the
highest geno­mic risks (ie, del(17p) with CK) may be best served
by indef­i­nite sup­pres­sion of the CLL clone at a time early in the
treat­ment sequence rather than being chal­lenged by inter­mit­
tent ther­a­pies that may give rise to subclonal com­plex­ity later in
the dis­ease course.39
Other advan­tages of con­tin­u­ous BTKi ther­apy
In gen­eral, BTKi ther­apy is straight­for­ward to com­mence and
broadly appli­ca­ble across a range of patients, includ­ing those
with mild to mod­er­ate organ impair­ments, with a low risk of
tumor lysis syn­drome.10-13,15,17,31,33,38 Therefore, BTKis may be par­tic­
u­larly suit­able in patients with high tumor bur­den dis­ease, those
with renal impair­ment, or those sit­ua ­ ­tions (eg, patients resid­ing
in rural com­mu­ni­ties) where real-time mon­i­tor­ing for tumor lysis
syndrome for venetoclax step-up may be logis­ti­cally dif­fi­cult.
Their rel­a­tively broader “cov­er­age” against geno­mic high-risk
CLL also makes them eas­ier agents to admin­is­ter in com­mu­nity-
based clin­ics where advanced test­ing for CLL geno­mics may be
less avail­­able. Indeed, emerg­ing data from real-world reg­is­tries
show low rates of test­ing for TP53 and other high-risk fea­tures in
patients with CLL man­aged in the United States.40
Drawbacks of con­tin­u­ous BTKi ther­apy
Continuous BTKi ther­a­pies do have their price in terms of poten­
tial for emer­gence of resis­tance, side effect bur­den on patient
qual­ity of life, and eco­nomic cost. Addressing the poten­tial of
resis­tance emer­gence, it is the­o­ret­i­cally pos­si­ble that inter­mit­
tent expo­sure to a drug may have a lower risk of induc­ing treat­
ment resis­tance com­pared with long-term expo­sure. However,
as we have seen in the early MURANO retreatment expe­ri­ence,28
even fixed-dura­tion venetoclax expo­sure appears to induce the
emer­gence of geno­mic high-risk, treat­ment-resis­tant clones. To
date, there are no data on whether lim­ited-dura­tion BTKi expo­
sure (eg, in tri­als com­bin­ing ibrutinib and venetoclax)41 may have
less impact on the sub­ se­
quent emer­ gence of BTKi-resis­ tant
muta­tions.
Many patients on con­tin­u­ous BTKi expe­ri­ence side effects that
may be tol­er­a­ble in the short term but cumu­la­tively rep­re­sent a
sub­stan­tial impact on their qual­ity of life in the long term.42 Real-
world stud­ies suggested that up to 41% of patients discontinue
ibrutinib after a median of 17 months of fol­low-up, largely due
to side effects.43 Balanced against these chronic tox­ic­ity con­sid­
er­ations, it is note­wor­thy that all 3 recently reported head-to-
head com­par­i­sons of ibrutinib vs acalabrutinib or zanubrutinib
were con­cor­dant in reporting reduced car­dio­vas­cu­lar and mus­
cu­lo­skel­e­tal com­pli­ca­tions in favor of the sec­ond-gen­er­a­tion
agent.35-37 Therefore, the tol­er­ance pro­file of indef­i­nite BTK inhi­bi­
tion in the sec­ond-gen­er­a­tion BTKi era is likely to be sub­stan­tially
improved com­pared with that of the ibrutinib era.
CLINICAL CASE (Con­t in­u ed)
After dis­cus­sions between the phy­si­cian and patient, our 71-year-
old patient was started on front­line BTKi due to his rural loca­tion
and poor-risk geno­mic fea­tures. Acalabrutinib was cho­sen over
ibrutinib due to a lower risk of car­dio­vas­cu­lar toxicities.
Conflict-of-inter­est dis­clo­sure
Constantine S. Tam: Research funding from Janssen, AbbVie, and
Beigene; hon­o­raria from Janssen, Pharmacyclics, Beigene, and
AbbVie.
Off-label drug use
Constantine S. Tam: none discussed.
Correspondence
Constantine S. Tam, Department of Haematology, Peter MacCal­
lum Cancer Centre, 305 Grattan Street, Melbourne 3050, Victo­
ria, Australia; e-mail: constantine​­.tam@petermac​­.org.
References
1. Tam CS, O’Brien S, Wierda W, et al. Long-term results of the fludarabine,
cyclo­phos­pha­mide, and rituximab reg­i­men as ini­tial ther­apy of chronic
lym­pho­cytic leu­ke­mia. Blood. 2008;112(4):975-980.
2. Byrd JC, O’Brien S, James DF. Ibrutinib in relapsed chronic lym­pho­cytic leu­
ke­mia. N Engl J Med. 2013;369(13):1278-1279.
3. Byrd JC, Furman RR, Coutre SE, et al. Ibrutinib treat­ment for first-line and
relapsed/refrac­tory chronic lym­pho­cytic leu­ke­mia: final anal­y­sis of the piv­
otal phase Ib/II PCYC-1102 study. Clin Cancer Res. 2020;26(15):3918-3927.
4. Keating MJ, O’Brien S, Kontoyiannis D, et al. Results of first sal­vage ther­apy
for patients refrac­tory to a fludarabine reg­i­men in chronic lym­pho­cytic leu­
ke­mia. Leuk Lymphoma. 2002;43(9):1755-1762.
5. Tam CS, O’Brien S, Lerner S, et  al. The nat­ u­ral his­
tory of fludarabine-
refrac­tory chronic lym­pho­cytic leu­ke­mia patients who fail alemtuzumab or
Prakash Singh Shekhawat
have bulky lymph­ade­nop­a­thy. Leuk Lymphoma. 2007;48(10):1931-1939.
6. Byrd JC, Harrington B, O’Brien S, et al. Acalabrutinib (ACP-196) in relapsed
chronic lym­pho­cytic leu­ke­mia. N Engl J Med. 2016;374(4):323-332.
7. Tam CS, Trotman J, Opat S, et al. Phase 1 study of the selec­tive BTK inhib­
i­tor zanubrutinib in B-cell malig­nan­cies and safety and effi­cacy eval­u­a­tion
in CLL. Blood. 2019;134(11):851-859.
8. Mato AR, Shah NN, Jurczak W, et al. Pirtobrutinib in relapsed or refrac­tory
B-cell malig­nan­cies (BRUIN): a phase ½ study. Lancet. 2021;397(10277):892-
901.
9. Burger JA, Tedeschi A, Barr PM, et al; RESONATE-2 Investigators. Ibrutinib
as ini­tial ther­apy for patients with chronic lym­pho­cytic leu­ke­mia. N Engl
J Med. 2015;373(25):2425-2437.
10. Sharman JP, Egyed M, Jurczak W, et al. Acalabrutinib with or with­out obinu­
tuzumab ver­sus chlorambucil and obinutuzmab for treat­ment-naive chronic
lym­pho­cytic leu­kae­mia (ELEVATE TN): a randomised, con­trolled, phase 3
trial. Lancet. 2020;395(10232):1278-1291.
11. Moreno C, Greil R, Demirkan F, et al. Ibrutinib plus obinutuzumab ver­sus
chlorambucil plus obinutuzumab in first-line treat­ ment of chronic lym­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/55/1851744/55tam.pdf by guest on 13 December 2021
pho­cytic leu­kae­mia (iLLUMINATE): a multicentre, randomised, open-label,
phase 3 trial. Lancet Oncol. 2019;20(1):43-56.
12. Shanafelt TD, Wang XV, Kay NE, et al. Ibrutinib-rituximab or chemoimmuno­
therapy for chronic lym­pho­cytic leu­ke­mia. N Engl J Med. 2019;381(5):432-
443.
13. Woyach JA, Ruppert AS, Heerema NA, et  al. Ibrutinib reg­i­mens ver­sus
chemoimmunotherapy in older patients with untreated CLL. N Engl J Med.
2018;379(26):2517-2528.
14. Byrd JC, Brown JR, O’Brien S, et al. Ibrutinib ver­sus ofatumumab in pre­vi­
ously treated chronic lym­phoid leu­ke­mia. N Engl J Med. 2014;371(3):213-
223.
15. Ghia P, Pluta A, Wach M, et al. ASCEND: phase III, ran­dom­ized trial of aca­
labrutinib ver­sus idelalisib plus rituximab or bendamustine plus rituximab
in relapsed or refrac­tory chronic lym­pho­cytic leu­ke­mia. J Clin Oncol. 2020;
38(25):2849-2861.
16. O’Brien SM, Jaglowski S, Byrd JC, et  al. Prognostic fac­tors for com­plete
response to ibrutinib in patients with chronic lym­pho­cytic leu­ke­mia: a
pooled anal­y­sis of 2 clin­i­cal tri­als. JAMA Oncol. 2018;4(5):712-716.
17. Burger JA, Barr PM, Robak T, et al. Long-term effi­cacy and safety of first-line
ibrutinib treat­ment for patients with CLL/SLL: 5 years of fol­low-up from the
phase 3 RESONATE-2 study. Leukemia. 2020;34(3):787-798.
18. Burger JA, Sivina M, Jain N, et  al. Randomized trial of ibrutinib vs ibruti­
nib plus rituximab in patients with chronic lym­pho­cytic leu­ke­mia. Blood.
2019;133(10):1011-1019.
19. Sharman JP, Egyed M, Jurczak W, et al. Acalabrutinib±obinutuzumab ver­sus
obinutuzumab  +  chlorambucil in treat­ment-naïve chronic lym­pho­cytic leu­
ke­mia: ele­vate-TN four-year fol­low up. J Clin Oncol. 2021;39(15, suppl):7509.
20. Roberts AW, Davids MS, Pagel JM, et al. Targeting BCL2 with venetoclax
in relapsed chronic lym­pho­cytic leu­ke­mia. N Engl J Med. 2016;374(4):311-
322.
21. Roberts AW, Ma S, Kipps TJ, et al. Efficacy of venetoclax in relapsed chronic
lym­pho­cytic leu­ke­mia is influ­enced by dis­ease and response var­i­ables.
Blood. 2019;134(2):111-122.
22. Stilgenbauer S, Eichhorst B, Schetelig J, et al. Venetoclax for patients with
chronic lym­pho­cytic leu­ke­mia with 17p dele­tion: results from the full pop­
u­la­tion of a phase II piv­otal trial. J Clin Oncol. 2018;36(19):1973-1980.
23. Seymour JF, Ma S, Brander DM, et al. Venetoclax plus rituximab in relapsed
or refrac­tory chronic lym­pho­cytic leu­kae­mia: a phase 1b study. Lancet
Oncol. 2017;18(2):230-240.
24. Ma S, Seymour JF, Brander DM, et al. Efficacy of venetoclax plus rituximab
for relapsed CLL: 5-year fol­low-up of con­tin­u­ous or lim­ited-dura­tion ther­
apy. Blood. 2021;138(10):836-846.
25. Seymour JF, Kipps TJ, Eichhorst B, et al. Venetoclax-rituximab in relapsed or
refrac­tory chronic lym­pho­cytic leu­ke­mia. N Engl J Med. 2018;378(12):1107-
1120.
26. Kater AP, Wu JQ , Kipps T, et  al. Venetoclax plus rituximab in relapsed
chronic lym­ pho­cytic leu­ ke­
mia: 4-year results and eval­ u­a­
tion of impact
of geno­mic com­plex­ity and gene muta­tions from the MURANO phase III
study. J Clin Oncol. 2020;38(34):4042-4054.
27. Fischer K, Al-Sawaf O, Bahlo J, et al. Venetoclax and obinutuzumab in patients
with CLL and coexisting con­di­tions. N Engl J Med. 2019;380(23):2225-2236.
28. Wu JQ , Seymour JF, Eichhorst B, et al. Impact of unfa­vor­able genet­ics on
min­i­mal resid­ual dis­ease (MRD) response to venetoclax+rituximab retreat­
ment in relapsed or refrac­tory CLL: phase 3 MURANO substudy. Presented
at: EHA2021 Virtual Congress; June 9-12, 2021. Poster EP599.
29. Tausch E, Schneider C, Yosifov D, Robrecht S, Zhang C. Genetic mark­ers
and out­come with front line obinutuzumab plus either chlorambucil or
Upfront ther­apy: the case for con­tin­u­ous treat­ment  |  57
 venetoclax—updated anal­y­sis of the CLL14 trial. Presented at: EHA2021
Virtual Congress; June 9-12, 2021. Poster S144.
30. Al-Sawaf O, Zhang C, Robrecht S, et  al. Clonal dynam­ics after veneto­
clax-obinutuzumab ther­apy: novel insights from the ran­dom­ized, phase 3
CLL14 trial. Blood. 2020;136(suppl 1):22-23.
31. Munir T, Brown JR, O’Brien S, et al. Final anal­y­sis from RESONATE: up to six
years of fol­low-up on ibrutinib in patients with pre­vi­ously treated chronic
lym­pho­cytic leu­ke­mia or small lym­pho­cytic lym­phoma. Am J Hematol.
2019;94(12):1353-1363.
32. Barr PM, Tedeschi A, Munir T, et al. Using ibrutinib in ear­lier lines of treat­ment
results in bet­ter out­comes for patients with chronic lym­pho­cytic leu­ke­mia/
small lym­pho­cytic lym­phoma. Blood. 2019;134(suppl 1):3054.
33. Ahn IE, Tian X, Wiestner A. Ibrutinib for chronic lym­pho­cytic leu­ke­mia with
TP53 alter­ations. N Engl J Med. 2020;383(5):498-500.
34. Allan JN, Shanafelt T, Wiestner A, et al. long-term effi­cacy of first-line ibru­
tinib treat­ment for chronic lym­pho­cytic leu­ke­mia (CLL) with 4 years of
fol­low-up in patients with TP53 aber­ra­tions (del(17p) or TP53 muta­tion): a
pooled anal­y­sis from 4 clin­i­cal tri­als. Blood. 2020;136(suppl 1):23-24.
35. Tam CS, Opat S, D’Sa S, et al. A ran­dom­ized phase 3 trial of zanubrutinib
vs ibrutinib in symp­tom­atic Waldenström mac­ro­glob­u­li­ne­mia: the ASPEN
study. Blood. 2020;136(18):2038-2050.
36. Byrd JC, Hillmen P, Ghia P, et al. First results of a head-to-head trial of aca­
labrutinib vs ibrutinib in pre­vi­ously treated CLL. J Clin Oncol. 2021;39(15,
suppl):7500.
37. Hillmen P, Eichhort B, Brown JR, Lamanna N, O’Brien S. First interim anal­y­
sis of ALPINE study: results of a phase 3 ran­dom­ized study of zanubrutinib
58  |  Hematology 2021  |  ASH Education Program
vs ibrutinib in patients with relapsed/refrac­tory CLL/SLL. Presented at:
EHA2021 Virtual Congress; June 9-12, 2021. Poster LB1900.
38. Tam CS, Robak T, Ghia P, et al. Zanubrutinib monotherapy for patients with
treat­ment naïve chronic lym­pho­cytic leu­ke­mia and 17p dele­tion. Haema-
tologica. 2020;106(9):2354-2363.
39. Landau DA, Tausch E, Taylor-Weiner AN, et al. Mutations driv­ing CLL and
their evo­ lu­
tion in pro­
gres­ sion and relapse. Nature. 2015;526(7574):525-
530.
40. Mato A, Nabhan C, Kay NE, et al. prog­nos­tic test­ing pat­terns and out­comes
of chronic lym­pho­cytic leu­ke­mia patients strat­i­fied by fluo­res­cence in situ
hybrid­iza­tion/cyto­ge­net­ics: a real-world clin­i­cal expe­ri­ence in the con­nect
CLL reg­is­try. Clin Lymphoma Myeloma Leuk. 2018;18(2):114-124.e2124e2.
41. Wierda WG, Tam CS, Allan JN, et  al. Ibrutinib (Ibr) plus venetoclax (Ven)
for first-line treat­ment of chronic lym­pho­cytic leu­ke­mia (CLL)/small lym­
pho­cytic lym­phoma (SLL): 1-year dis­ease-free sur­vival (DFS) results from
the MRD cohort of the phase 2 CAPTIVATE study. Blood. 2020;136(suppl 1):
16-17.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/55/1851744/55tam.pdf by guest on 13 December 2021
42. Lasica M, Tam CS. Management of ibrutinib toxicities: a prac­ti­cal guide.
Curr Hematol Malig Rep. 2020;15(3):177-186.
43. Mato AR, Nabhan C, Thompson MC, et al. Toxicities and out­comes of 621
ibrutinib-treated chronic lym­ pho­cytic leu­ke­ mia patients in the United
States: a real-world anal­y­sis. Haematologica. 2018;103(5):874-879.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000232
 CLL: EXTENDING SURVIVAL

Frontline treatment in CLL: the case

for time-limited treatment

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/59/1851402/59levy.pdf by guest on 13 December 2021

Vincent Lévy,1 Alain Delmer,2 and Florence Cymbalista3
Département de Recherche Clinique, Hôpital Avicenne AP­HP, Université Sorbonne Paris Nord, Bobigny, France, and INSERM CRESS­UMR 1153,
1

Hôpital Saint Louis, Paris, France; 2Hematology Department, Reims University Hospital and Reims Champagne Ardenne University, Reims, France;
and 3Hematology Biology, Hôpital Avicenne, AP­HP, Université Sorbonne Paris Nord, Bobigny, France, and INSERM UMR 978, Bobigny, France

Over the last decade, the advent of Bruton tyrosine kinase inhibitors (BTKi) has profoundly modified the therapeutic
strategy in chronic lymphocytic leukemia (CLL), introducing the concept of treatment until progression. Initially, the
bcl-2 inhibitor venetoclax (VEN) was used as a single agent and then was rapidly combined in VEN-based regimens asso-
ciated with either anti-CD20 or with BTKi. These regimens yielded a high rate of complete remission, leading to their use
as a fixed duration treatment. The decision between continuous treatment with BTKi and VEN-based combinations relies
mostly on comorbidities, comedications, and patient/physician preferences. Notably, with BTKi, cardiovascular comor-
bidities, hypertension, and potential pharmacological interactions should be carefully evaluated. On the other hand, the
risk of tumor lysis syndrome with VEN should be monitored at treatment initiation. TP53 alteration and IGHV mutational
status should also be assessed, as they remain important for therapeutic decisions. Fit patients with a TP53 wild type
and IGHV-mutated CLL may still benefit from fludarabine-cyclophosphamide-rituximab chemoimmunotherapy (CIT), as
it may result in a very long remission duration. VEN-based treatments are well tolerated, and no additional toxicity has
been observed when combined with anti-CD20 or BTKi. The 1-year fixed-duration association of VEN plus obinutuzumab
was evaluated in frontline for older adult patients. Nonetheless, considering the favorable outcome, an extension of
indication for fit younger patients is expected. The association of VEN and BTKi is promising, even if the follow-up is still
short. It is currently being tested against CIT, BTKi continuous treatment, and VEN plus anti-CD20.

LEARNING OBJECTIVES
• Recognize the important biological pretherapeutic parameters for allocating CLL patients to optimal frontline
treatment
• Review the indications of time­limited treatment for the various categories of treatment­naive CLL patients
• Review the various fixed­duration combination therapies and the major trials testing the time­limited strategies
• Be aware of future developments that will allow a more personalized therapeutic approach

Introduction plete response and undetectable minimal residual disease

The concept of time­limited treatment is obviously not
new in chronic lymphocytic leukemia (CLL) since up until
the arrival of BCR inhibitors, treatment relied mainly on
chemotherapy and was therefore of limited duration. In
recent years, Bruton tyrosine kinase inhibitors (BTKi) have
been shown to be superior overall to chemoimmunother­
apy (CIT) in the first­line treatment of CLL, but they must
be given continuously until progression as they have only
a suspensive effect on the disease.1,2 This has led to new
concerns about compliance, quality of life, and cost. The
BCL2 inhibitor venetoclax (VEN), either alone or combined
with other agents, is likely to yield high rates of both com­
Prakash Singh Shekhawat
(uMRD), and it can be discontinued safely after optimal
response is achieved.3
CLINICAL CASE 1
Mr. V., who is 58, is fit and works as an engineer. He has a history
of coronary thrombosis dating from 5 years ago, is on acetyl­
salicylic acid, and takes amlodipine for high blood pressure.
At diagnosis of CLL, his blood count showed 23×109/L lym­
phocytes and no cytopenia. An immunophenotype revealed
Up­front therapy: time­limited treatment | 59
Table 1. Recommended mark­ers for the diag­no­sis of CLL (Euro­pean Research Initiative on CLL rec­om­men­da­tions)

Minimally recommended Other impor­tant mark­ers

CD19 Intracellular Ig light chains kappa and lambda (if absence
CD5 of mem­brane staining)
Ig light chains kappa and lambda (mem­brane staining) CD81
CD23 CD43
CD79b and/or CD22 ROR1
CD20 CD38
FMC7 CD10
CD200
Ig, immu­no­glob­u­lin.

clonal B cells with low CD20 expres­sion har­bor­ing pos­i­tive CD5 55 mL/min, and the direct anti­body test was neg­a­tive. He was

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/59/1851402/59levy.pdf by guest on 13 December 2021

expression, CD23 and CD200hi, low kappa light chain, and CD79b.
CD38 was not expressed on CLL cells. Clinical exam­i­na­tion found
centimetric cer­vi­cal and axil­lary lymph nodes and no spleno­meg­
aly. Five years later in 2010, Mr. V. was still fit, but his CLL had pro­
gressed with lymph­ade­nop­a­thy from 3 to 5 cm (cer­vi­cal, axil­lary,
and ingui­nal) and a spleno­meg­aly 6 cm below the cos­tal mar­gin.
He had lost 3 kg and expe­ri­enced night sweats. His blood counts
were as fol­ lows: lym­ pho­ cytes 120 × 
109/L; hemo­glo­bin 135 g/L;
plate­lets 176 × 10 /L. Biological param­e­ters were as fol­lows: esti­
9 Indications for treat­ment ini­ti­a­tion according to the Interna­
mated glo­mer­u­lar fil­tra­tion rate (eGFR), 70 mL/min; del13q dele­tion
and no 17p dele­tion; mutated IGHV sta­tus with VH 3-30 with 95%
homol­ogy; and hypogammaglobulinemia at 5 g/L. In 2010 he was
started on 6 courses of an oral fludarabine-cyclo­phos­pha­mide-
rituximab (FCR) reg­i­men.
started in 2010 on 6 courses of bendamustine-rituximab (BR).
What if we had to do it again today?
Indication for treat­ment
It is impor­tant to keep in mind that the ther­a­peu­tic options dis­
cussed here are intended for bona fide CLL and that ade­quate
phenotyping (Table 1) should be performed at diag­no­sis.
tional Workshop on Chronic Lymphocytic Leukemia have not
dra­mat­i­cally changed over the years. As exem­pli­fied in these 2
cases, spe­cific treat­ment is indi­cated when patients show signs
of CLL pro­gres­sion. The pro­gres­sion cri­te­ria have been very
clearly defined by the International Workshop on Chronic Lym­
phocytic Leukemia and are sum­ma­rized in Table 2.4

Pretreatment assess­ment and pre­ven­tive mea­sures

CLINICAL CASE 2
Mr. F. is 72 years old, retired, and a pre­vi­ous bank employee. He
is a heavy smoker with type 2 dia­be­tes and dyslipidemia and
takes biguanide and atorvastatin. He was diag­nosed with CLL
in 2009 with a char­ac­ter­is­tic phe­no­type. He rap­idly progressed
into active dis­ease, with mul­ti­ple lymph­ade­nop­a­thy, spleno­
meg­aly, and night sweats. His lym­pho­cyte dou­bling time was
6 months, and blood counts showed hemo­glo­bin 110 × 109/L;
lym­pho­cytes: 130  ×  109/L; and plate­lets: 90 × 109/L. He had VH1-
69 IGHV with 100% homol­ ogy and a nor­ mal kar­yo­ type and
fluo­res­cence in situ hybrid­iza­tion (FISH) anal­y­sis. His eGFR was
International and national soci­e­ties have published com­mon rec­
om­men­da­tions for pre­treat­ment assess­ments, includ­ing his­tory,
phys­i­cal exam­i­na­tion, and stan­dard bio­log­i­cal tests (Table 3).4-6
The cur­rent pretherapeutic workup includes the pre­vi­ous rec­
om­men­da­tions of the CIT era, but some new assess­ments have
been added.
Whenever targeted agents are con­ sid­ered, nota­ bly BTKi,
car­dio­vas­cu­lar comorbidities, hyper­ten­sion, and comedications
should be care­fully assessed.7 The pre­ven­tion of tumor lysis syn­
drome is well established and should be applied to any VEN-
based strat­egy.8
Patients with CLL are at risk of infec­tion with targeted agents as
well, and all­pos­si­ble pre­ven­tive mea­sures should be under­taken

Table 2. Criteria for treat­ment ini­ti­a­tion

•  Progressive BM fail­ure with the devel­op­ment or aggra­va­tion of ane­mia and/or throm­bo­cy­to­pe­nia.

•  Massive or pro­gres­sive or symp­tom­atic spleno­meg­aly.
•  Significantly enlarged or symp­tom­atic or pro­gres­sive lymph­ade­nop­a­thies.
• Progressive lym­pho­cy­to­sis, with an increase of more than 50% over a period of 2 months or an LDT of less than 6 months. In patients with an ini­tial
lym­pho­cyte count <30 × 109/L, LDT alone should not be used as a sin­gle param­e­ter to decide treat­ment ini­ti­a­tion and should be interpreted in the
over­all clin­i­cal con­text.
•  Autoimmune ane­mia and/or throm­bo­cy­to­pe­nia that is poorly respon­sive to cor­ti­co­ste­roids or other stan­dard ther­a­pies.
•  The pres­ence of con­sti­tu­tive symp­toms as defined by 1 or more of the fol­low­ing signs or symp­toms related to the dis­ease:
  Unintentional weight loss of 10% or more in the pre­vi­ous 6 months,
sig­nif­i­cant fatigue (ECOG PS 2 or worse; inabil­ity to per­form usual activ­i­ties),
fever over 38.0 °C for 2 weeks or more with­out signs of infec­tion, and
night sweats last­ing more than a month with no sign of infec­tion.
At least one of the cri­te­ria must be fulfilled.

60  |  Hematology 2021  |  ASH Education Program

Table 3. Pretreatment eval­u­a­tion

Routine prac­tice Clinical tri­als spe­cific

Clinical eval­u­a­tion
  ECOG PS Always
  Constitutive symp­toms Always
  Clinical exam­i­na­tion Always
Cardiovascular comorbidities Always
Comedications Always
Biological stan­dard tests
 Complete blood count and retic­u­lo­cyte Always
count

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/59/1851402/59levy.pdf by guest on 13 December 2021

  LDH, hap­to­glo­bin and direct anti­glob­u­lin test Always
 Serum cre­at­i­nine level and glo­mer­u­lar fil­tra­ Always
tion rate
  Transaminases, bil­i­ru­bin, GGT Always
  LDH and beta-2 microglobulin Always
  Serum pro­tein elec­tro­pho­re­sis Always
  HIV, B and C hep­a­ti­tis serol­ogy Always
 Cryopreservation of blood cells (tumor Desirable Always
library)
Genetics
 Karyotype Desirable Always
 FISH
  17p dele­
tion Always
  11q dele­
tion Desirable Always
  13q dele­
tion Desirable Always
   Trisomy 12 Desirable Always
IGHV muta­tional sta­tus Always
TP53 muta­tions by NGS Always
NGS of recur­rent muta­tions NGI Desirable
Imaging
CT scan of chest, abdo­men, and pel­vis Desirable Recommended
Preventive mea­sures
  Vaccination against pneu­mo­coc­cus Always
  Vaccination against influ­enza (yearly) Always
  Vaccination against SARS-CoV-2 Always
CT, com­puted tomog­ra­phy; GGT, gamma-glutamyl trans­fer­ase; LDH, lac­tate dehy­dro­ge­nase; NGI, not generally indicated.

ated with resis­tance to CIT and remain an unfa­vor­able prog­nos­

before starting ther­apy, nota­bly vac­ci­na­tion (pneu­mo­coc­cus and
annu­ally for influ­enza).9 SARS-CoV-2 vac­ci­na­tion should be per­tic fac­tor with targeted agents.2,3,11,12 It is there­fore man­da­tory to
formed, pref­er­a­bly before starting ther­apy, par­tic­u­larly if the patient
test for both these alter­ations.
is to receive anti-CD20 or BTKi.10 In the 2 cases illus­trated here, in 2010 only 17p FISH was avail­­
In gen­eral, a body com­puted tomog­ra­phy scan may be help­ able. It should be stressed that FISH alone detects only half of
ful to assess the tumor load and risk of tumor lysis syn­drome, TP53 alter­ations, as in approx­i­ma­tely 50% of cases only a TP53
mainly before treat­ment with the BCL2 inhib­i­tor VEN.5 muta­tion is pres­ent and there­fore not iden­ti­fied by FISH anal­y­sis.13
The fol­low­ ing tests have become impor­ tant for deci­Nowadays, sequenc­ing anal­y­sis of TP53 is required before the
sion-
mak­ing. ini­ti­a­tion of ther­apy.4,14 Next-gen­er­a­tion sequenc­ing (NGS) is rap­
idly replacing Sanger sequenc­ing, as it allows the detec­tion of
TP53  Alteration of TP53 func­tion can be medi­ated by either 17p small subclones. In our obser­va­tions, both patients were tested
Prakash
dele­tion, TP53 muta­tion, or both. These Singh
alter­ations are Shekhawat
asso­ci­ ret­ro­spec­tively for TP53 muta­tions and were neg­a­tive.

Up-front ther­apy: time-lim­ited treat­ment  |  61

IGHV  The var­i­able region of immu­no­glob­u­lin heavy chain gene
(IGHV) muta­tional sta­tus, which until recently was only indi­cated
in clin­ i­
cal tri­als, has become an impor­ tant deci­ sion-mak­ing
param­ e­ ter and must be performed before front­ line ther­
apy.4
IGHV muta­tional sta­tus has been con­sid­ered in the last 20 years
to be the most pow­er­ful pre­dic­tive indi­ca­tor. Patients with un­
mutated IGHV genes (UM-IGHV CLL, ie, ≥98% homol­ogy with
germ line sequence) have infe­rior out­comes after any CIT reg­
i­men.11,15,16 Targeted agents have dem­on­strated a clear ben­e­fit
over CIT for these patients.17
Even for FCR inel­i­gi­ble patients, deter­mi­na­tion of IGHV muta­
tional sta­tus is of inter­est. The genetic com­plex­ity appears dif­
fer­ent: UM-IGHV patients accu­mu­late sig­nif­i­cantly more genetic
alter­ations than M-IGHV CLL.18 The neg­ a­
tive impact of this
ATM,21 and some of them have been asso­ci­ated with expo­nen­tial
growth.22 This expanding knowl­edge about growth dynam­ics in
indi­vid­ual patients will be help­ful for predicting the course of the
dis­ease and informing ther­apy in the future but is not yet incor­
po­rated into a treat­ment deci­sion algo­rithm.
Treatment options today
Outside of clin­i­cal tri­als, as of today, 2 dif­fer­ent strat­e­gies may
be con­sid­ered: either con­tin­u­ous ther­apy until pro­gres­sion with
BTKi alone or com­bined with obinutuzumab (O), or time-lim­ited
treat­ment with either CIT or VEN-based reg­i­mens. In this sec­tion,
we will con­sider only the time-lim­ited ther­a­pies.
CIT is the pro­to­type of time-lim­ited treat­ment and may still
be of ben­e­fit for some patients. The ther­a­peu­tic reg­i­mens have

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/59/1851402/59levy.pdf by guest on 13 December 2021

increased muta­tion bur­den has been reduced in UM-IGHV CLL
by targeted agents, but there are dif­fer­ent dom­i­nant muta­genic
mech­a­nisms in M-IGHV and UM-IGHV, under­ly­ing a very likely dif­
fer­ent clonal evo­lu­tion at relapse.19 Therefore, deter­mi­na­tion of
IGHV muta­tional sta­tus has been recommended for all­patients.4,5
Whether a kar­yo­type should be performed in rou­tine prac­
tice remains open. Complex karyotype is a known det­ri­men­tal
fac­tor in clin­i­cal tri­als, both for CIT and VEN.20 It is of inter­est
when it can be eas­ily obtained, but the results still do not clearly
influ­ence treat­ment deci­sion. Similarly, sev­eral recur­rent gene
muta­tions are observed in CLL, such as NOTCH1, SF3B1, and
been mod­u­lated based on fit­ness and age, from chlorambucil
(CLB) and anti-CD20 to BR/FCR.11,15,16 Targeted-agent reg­i­mens
have been com­pared to CIT in front­line (Table 4; Figure 1).
Ibrutinib-rituximab (I-R) until pro­gres­sion has been com­pared
to FCR in a large ran­dom­ized (2:1) phase 3 trial includ­ing 529
patients youn­ger than 70 with no 17p dele­tion.17 I-R is clearly
supe­rior in terms of pro­gres­sion-free sur­vival (PFS) and even OS.
This is related to the large ben­e­fit of I-R over FCR for UM-IGHV
patients. It is note­wor­thy that in M-IGHV cases, PFS after I-R and
FCR remains sim­i­lar at a fol­low-up of 33.6 months despite the
fact that FCR is a 6-month treat­ment. The inci­dence of grade >3

Table 4. Summary of time-lim­ited front­line tri­als

Reference Trial Patients Primary end point Results

Randomized phase 3 tri­als
Shanafelt et al17, FCR vs I-R Age <70 PFS I - R > FCR (89.4% vs 72.9% at 3 y;
ECOG1912 no del 17p FU: 33.6 mo HR: 0.35; 95% CI, 0.22-0.56; P < .001)
N = 529 OS: I - R > FCR (98.8% vs 91.5% at 3 y)
HR: 0.17; 95% CI, 0.05-0.54; P < .001
Al-Sawaf et al3, CLB-O vs VEN-O CIRS >6 and/or Cl. creat PFS VEN-O > CLB-O (NE vs 35.5 mo)
CLL14 30-69 mL/min FU: 39.6 mo HR: 0.31, 95% CI, 0.22–0.44; P < .0001
N = 432
Woyach et al1, ALLIANCE BR/I-R/I Age >65 PFS I-R > BR. HR: 0.38; 95% CI, 0.25-0.59;
N = 447 FU: 38 mo P < .001
I > BR. HR: 0.39; 95% CI, 0.26-0.58;
P < .001
Kater et al23, VEN-I/CLB-O Age >65 or CIRS >6 or Cl. PFS VEN-I > CLB-O
GLOW creat <70 mL/min FU: 27.7 mo HR: 0.216 (95% CI, 0.131-0.357);
N = 211 p < .0001)

Phase 2 tri­als
Michallet et al24, ICLL07 I-O+/−4 courses FCO-I Fit patients, no del 17p or p53 CR + BM MRD <0.01% CR + MRD <0.01%: 62% (95% CI 55%
N = 135 FU: 26.3 to 69%)
Davids et al25 FCR 6 courses-I (stop 2 y if Age <65 CR + BM uMRD 2 mo CR + BM uMRD: 33%, (95% CI 23-
BM uMRD) N = 85 post FCR 44%)
FU: 16.5 mo
Jain et al27 VEN-I Del17P and/or TP53 mut CR + CRi CR + CRi: 88%, CR + uMRD: 61%
and/or del11q and/or FU: 14.8 mo
U-IGHV and/or age >65
N = 80
Wierda et al28, CAPTIVATE VEN-I Age <70 1 y DFS I vs pla­cebo: NS (95% vs 100%)
uMRD : I vs placebo N = 164 FU: NA 30-mo PFS ∼95% in all­arms
MRD+ : I vs V-I
A, acalabrutinib; B, bendamustine; C, Cyclophosphamide; Cl. creat, clear­ing of cre­a­tine; CR, complete remission; DFS, dis­ease-free sur­vival;
F, Fludarabine; FU, median fol­low-up; I, ibrutinib; NA, not appli­ca­ble; NE, not evaluable; NS, not significant; O, obinutuzumab; OS, overall survival;
PFS, progression-free survival; R, rituximab; VEN, venetoclax.

62  |  Hematology 2021  |  ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/59/1851402/59levy.pdf by guest on 13 December 2021

Figure 1. Outline of phase 3 and phase 2 time-lim­ited front­line tri­als.

Prakash Singh Shekhawat

Up-front ther­apy: time-lim­ited treat­ment  |  63
adverse events (AEs) is sim­i­lar in both arms, with fewer grade 3
or more infec­tious com­pli­ca­tions in the I-R arm than in the FCR
(10.5% vs 20.3%).
The ALLIANCE trial com­pared 183 patients receiv­ing benda­
mustine plus rituximab BR to 182 patients receiv­ing ibrutinib plus
rituximab I-R and 182 receiv­ing ibrutinib (I) alone in a phase 3
ran­dom­ized (1:1:1) trial includ­ing older adult patients (≥65 years
of age).1 The median fol­low-up was 38 months. PFS was supe­
rior with the ibrutinib reg­i­men vs the addi­tion of B + R, with no
dif­fer­ence between I and I-R. There was no dif­fer­ence in over­
all sur­vival (OS). Patients receiv­ing BR had a higher num­ber of
grade ≥3 hema­to­logic AEs com­pared to ibrutinib-based treat­
ments (61% vs 41% and 29%) and a lower num­ber of grade ≥3
nonhematologic AEs (63% vs 74% each).
The CLL14 phase 3 trial com­par­ing the venetoclax-obinutu­
zumab (VEN-O) com­bi­na­tion with chlorambucil-obinutuzumab
(CLB-O) ran­dom­ized (1:1) 432 frail patients as defined by a cumu­
lative illnesses rating scale (CIRS) >6 and/or eGFR between
30 and 69 mL/min.3 With a fol­low-up of 39.6 months, PFS and time
to next treat­ment were sig­nif­i­cantly lon­ger with VEN-O in both
M- and UM-IGHV CLL, but no sur­vival ben­efi ­ t was observed. Seri­
ous AEs occurred in 54% and 44% of the patients, respec­tively.
The most com­mon grade ≥3 was neutropenia in both groups
(48% and 54%, respec­tively). Neutropenia occurred mainly dur­
ing the course of obinutuzumab treat­ment.
The GLOW trial is a ran­dom­ized (1:1) phase 3 trial strat­i­fied
on IGHV muta­tional sta­tus and del(11q) com­par­ing ibrutinib and
VEN to CLB-O in patients older than 65 or youn­ger with a CIRS
>6 or with a cre­at­i­nine clear­ance <70 mL/min.23 Patients with a
del(17p) or TP53 muta­tion were not eli­gi­ble. Two hun­dred and
eleven patients were ran­dom­ized (median age, 71 years). After a
median fol­low-up of 27.7 months, VEN-I was sig­nif­i­cantly supe­rior
to CLB + O in terms of PFS (haz­ard ratio [HR], 0.216), com­plete
remis­sion (CR) rate, uMRD in bone mar­row (BM), and time to
next treat­ment. The most fre­quent grade >3 events for patients
treated by I + V were neutropenia (34.9%), diar­rhea (10.4%), and
hyper­ten­sion (7.5%).
In addi­tion to these phase 3 tri­als, sev­eral phase 2 tri­als have
explored the com­bi­na­tion of ibrutinib with either CIT or VEN in
front­line (Table 4; Figure 1).
In the ICLL07FILO trial,24 135 fit TP53 wild-type unmutated IGHV
CLL patients received the com­ bi­
na­
tion of obinutuzumab and
15-month fixed-dura­tion ibrutinib. At month 9, the 120 patients
(92%) who were not in BM CR with MRD <0.01% received
4 courses of FC-O. At the final eval­u­a­tion at month 15, a CR with
BM uMRD rate of 62% was obtained with an accept­able tox­ic­ity.
At a median fol­low-up at 36.7 months, OS and PFS at 3 years
were 95.7% and 98%, respec­tively.
In a phase 2 trial,25 85 patients aged 65 or youn­ger received
I + FCR (6 cycles). Responders con­tin­ued on ibrutinib main­
te­nance for up to 2 years, and patients with BM uMRD were
­able to ­discontinue treat­ment. Thirty-three per­cent of patients
achieved a CR with BM uMRD 2 months after the last course
of FCR. The most com­mon all­-grade toxic effects were hema­
to­log­i­cal. At a median fol­low-up of 16.5 months, 1 death was
observed but no pro­gres­sion. The M. D. Anderson group tested
3 cycles of the I-FC-O com­bi­na­tion in 45 M-IGHV fit patients,
with sim­i­lar excel­lent results.26
The venetoclax-ibrutinib com­bi­na­tion (VEN-I) was tested in
80 “high-risk” patients defined by at least one of the fol­low­ing:
64  |  Hematology 2021  |  ASH Education Program
17p or 11q dele­tion, TP53 muta­tion, UM-IGHV, or an age of 65
years or older.27 In this het­ero­ge­neous pop­ul­a­tion, the CR + CRi
(com­plete remis­sion with incom­plete recov­ery) rate was 88%,
and CR with uMRD was 61%. There was no addi­tional tox­ic­ity in
the com­bi­na­tion of the 2 agents. Considering the short fol­low-
up (14.8 months), no sur­vival data are avail­­able.
Finally, in the CAPTIVATE MRD trial,28 VEN-I was also tested
in 164 “high-risk” patients defined by at least one of the fol­low­
ing: del(17p), del(11q) TP53 muta­tion, com­plex kar­yo­type, or UM-
IGHV. There was an MRD-guided ran­dom­i­za­tion after 12 cycles
of com­ bined treat­ment, and uMRD patients were ran­ domly
assigned to pla­cebo or ibrutinib, the oth­ers being ran­dom­ized to
receive ibrutinib or con­tin­ued VEN-I. PFS rates were >95% across
all­ran­dom­ized arms at 30 months.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/59/1851402/59levy.pdf by guest on 13 December 2021
In sum­mary, all­phase 3 tri­als have shown the supe­ri­or­ity of
targeted agents over CIT when con­sid­er­ing the whole cohort.
This dif­fer­ence is related to the large ben­e­fit of targeted agents
in UM-IGHV cases. When con­sid­er­ing M-IGHV cases, only CLB-O
was infe­rior to a targeted-agent reg­i­men whether com­pared to
ibrutinib or I-O (ILLUMINATE) or VEN-O (CLL14).3,12 Conversely, in
the ECOG trial and ALLIANCE trial no ben­e­fit of BTKi over CIT
treat­ment (FCR or BR) was observed for M-IGHV cases whether
alone or in com­bi­na­tion with rituximab.1,17
Phase 2 tri­als have explored the addi­tion of ibrutinib to an FC-
anti-CD20 back­bone with impres­sive results, but unfor­tu­nately
this strat­egy was not com­pared to chemo-free ther­a­pies. Recent
tri­als have tested time-lim­ited targeted-agent com­bi­na­tions with
very good results. The VEN-I com­bi­na­tion MRD data are very
encour­ag­ing, but fol­low-up is still short for sur­vival anal­y­sis.27,28
It is note­wor­thy that targeted agents show a sig­nif­i­cant activ­
ity in TP53-altered cases. Any chemo-free reg­i­men was supe­rior
to what was pre­vi­ously observed with CIT, as TP53-altered cases
are resis­tant to CIT. Available data are the result of unplanned
sub­group anal­y­sis in tri­als and tend to favor the use of BTKi or
VEN-I over VEN +/− anti-CD20, despite the absence of a cur­rent
head-to-head com­par­i­son.
In the absence of comorbidities pre­vent­ing the use a cer­tain
targeted agent, usu­ally BTKi, the deci­sion of which targeted
agent is pref­er­a­ble is always a mat­ter of debate among phy­
si­cians and of dis­cus­sion with the patient. Besides the case of
TP53 alter­ations, some clin­i­cal sit­u­a­tions may tip the bal­ance in
favor of one or the other targeted agent. For exam­ple, in the
case of a pro­gres­sive CLL with a very short dou­bling time but
no bulky lymph nodes we would tend to favor the use of VEN +
anti-CD20, whereas in the case of a bulky dis­ease, BTKi might
be pre­ferred.
Several tri­als are also ongo­ing that are test­ing mainly time-
lim­ited strat­e­gies (Table 6). They explore com­bi­na­tions with
other BTKi and/or the tri­ple com­bi­na­tion of VEN with BTKi and
anti-CD20. The results will be avail­­able in the near future.
CLINICAL CASE 1 (Con­tin­ued)
How would you treat the first patient today? In 2021, 11 years
after receiv­ing FCR, this patient is still in clin­i­cal CR. These very
long remis­sions were high­lighted by the M. D. Anderson group,
show­ing that a sig­nif­i­cant per­cent­age of M-IGHV patients expe­
ri­ence very prolonged remis­sions (>10 years) after FCR.29
Table 6. Major ongo­ing tri­als eval­u­at­ing fixed-dura­tion targeted ther­a­pies against CIT or con­tin­u­ous targeted ther­a­pies

Trial N pts Trial pop­u­la­tion Compared arms Primary end point

CLL13- GAIA 926 ≥18 y CIT: FCR/BR × 6 cycles MRD at mo 15 (CIT vs GVe)
GCLLSG fit/TP53 abn excluded vs RVe: VEN-R (12 mo) PFS (CIT vs GIVe)
NCT02950051 vs GVe: VEN-O (12 mo)
vs GIVe: VEN-O + I (up to 36 mo;
VEN 12 mo)
ACE-CL-311 780 ≥18 y CIT: FCR/BR × 6 cycles PFS (CIT vs AV)
AstraZeneca fit/TP53 abn excluded vs AV: VEN-O + acalabrutinib
NCT03836261 (15 mo; VEN 12 mo)
vs AVG: VEN-O + acalabrutinib
(15 mo; VEN 12 mo)
CRISTALLO 165 ≥18 y CIT: FCR or BR (if ≥65 y) × 6 BM MRD at mo 15
Hoffmann La Roche fit/TP53 abn excluded cycles

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/59/1851402/59levy.pdf by guest on 13 December 2021

NCT04285567 vs VG: VEN-O (12 mo)
FILO ERADIC 120 ≥18 y CIT: FCR × 6 cycles BM MRD at mo 27
NCT04010668 fit/TP53 abn excluded vs VI: VEN-I (15 or 27 mo
according to MRD)
NCRI UK CLL10 FLAIR 1516 ≤75 y CIT: FCR × 6 cycles PFS
CRUK/12/037 fit/eGFR >30 mL/min vs IR → I*: I + ritux → I (until
del(17p) <20% pro­gres­sion)
vs VI: VEN-I (flex­i­ble dura­tion
according to MRD)
* arm IR replaced by
I monotherapy in 2018
ALLIANCE A041702 454 ≥70 y IO: I + O (until pro­gres­sion) PFS
NCT03737981 vs IVO: VEN-O + I (15 mo; VEN
12 mo)
ECOG-ACRIN EA9161 720 18-69 y IO: I + O (until pro­gres­sion) PFS
NCT03701282 del(17p) excluded vs IVO: VEN-O + I (19 mo; VEN
12 mo)
GCLLSG CLL17 897 ≥18 y I: I (until pro­gres­sion) PFS
NCT04608318 fit and unfit vs VG: VEN-O (12 mo)
w/or w/o TP53 abn vs VI: VEN-I (15 mo; VEN 12 mo)
abn, abnor­mal­i­ties; A, acalabrutinib; I, ibrutinib; O, obinutuzumab; R, rituximab; VEN, venetoclax.

Ten years later, this patient would still be eli­gi­ble for treat­
ment with FCR since he has M-IGHV CLL and no TP53 alter­ation.
In 2021, con­tin­u­ous BTKi can also be con­sid­ered but con­tin­ued
until pro­gres­sion. His comorbidities (high blood pres­ sure, a
his­tory of cor­o­nary throm­bo­sis on ace­tyl­sal­i­cylic acid) do not
restrain FCR but on the con­trary, ibrutinib should be used with
cau­tion. Concerning the 1-year VEN-O com­bi­na­tion, the results
from CLL14 might be extrap­o­lated to youn­ger patients, in which
the data show the supe­ri­or­ity of VEN-O over CLB-O even in
M-IGHV patients. As of yet, how­ever, no data are avail­­able esti­
mat­ing the mag­ni­tude of the effect—if any—com­pared to FCR.
CLINICAL CASE 2 (Con­t in­u ed)
How would you treat the sec­ond patient today? The out­come
was, unfor­tu­nately, poorer in this patient, who progressed
30 months after BR with altered performance status and rap­idly
grow­ing super­fi­cial polyadenopathy and spleno­meg­aly. Despite
ade­quate sup­port­ive care with infec­tion pro­phy­laxis, he died of
sep­sis dur­ing the reintroduction of BR.
All tri­als show the supe­ri­or­ity of targeted agents in UM-IGHV
CLL. Time-lim­ited treat­ments are VEN-based com­bi­na­tions, either
Prakash Singh Shekhawat
with an anti-CD20 or BTKi. Data from the CLL14 trial show the
clear supe­ri­or­ity of a 1-year ­treat­ment of VEN-O over CLB-O.
Moreover, in the CLL14 trial it was recently reported that the
VEN-O com­bi­na­tion could reduce the growth dynam­ics, as a
lower growth rate was observed at r­ eprogression after VEN-O.30
Approval of the targeted agents was obtained through
randomiza­tion against a ref­er­enced ­ther­a­peu­tic ­reg­i­men, ie,
age-adapted CIT. Data from a head-to-head com­par­i­son of chemo-
free reg­i­mens are not avail­­able yet, and for patients eli­gi­ble for
either con­tin­u­ous BTKi or time-lim­ited VEN-based com­bi­na­tions,
a treat­ment deci­sion relies mostly on comorbidities, comedica­
tions, and patient pref­er­ences. It is most likely that the life expec­
tancy of this patient would have been increased with first-line
targeted agents. This patient today would receive a targeted
agent, at least in relapse.
What would be help­ful for adjusting the dura­tion of time-
lim­ited treat­ment?
Most of the data avail­­able to date in front­line have been gen­er­
ated by treat­ment strat­e­gies of an arbi­trarily fixed dura­tion, such
as the CLL14 trial with a 1-year fixed dura­tion of VEN-O.
The value of MRD sta­tus as a sur­ro­gate of PFS has been widely
proven in the con­text of CIT.31 The dem­on­stra­tion of the MRD
Up-front ther­apy: time-lim­ited treat­ment  |  65
value through large tri­als led to approval by the Euro­pean Medi­
cines Agency of the use of uMRD as an inter­me­di­ate end point in
ran­dom­ized clin­i­cal tri­als for drug approval.32 It appeared to be a
wor­thy pre­dic­tor in VEN-based treat­ment.3
Through the anal­ y­sis of a large num­ ber of patients who
received FCR in 3 clin­i­cal tri­als, it was dem­on­strated that high-
sen­si­tiv­ity (0.0007%) MRD assess­ment in blood yielded addi­
tional prog­nos­tic infor­ma­tion beyond the cur­rent stan­dard
sen­si­tiv­ity (0.01%).33 Currently, the dura­tion of time-lim­ited treat­
ment is not guided by MRD sta­tus at the end of treat­ment. In
the future, the use of high-sen­si­tiv­ity MRD assess­ment could be
use­ful for deter­min­ing response at a deep level in poten­tially
cura­tive time-lim­ited reg­im ­ ens. Moreover, MRD kinet­ics will be
needed to bet­ter under­stand the impact of the var­i­ous novel
agents and to be ­able to decide when to discontinue ther­apy in
patients who do or do not achieve unde­tect­able dis­ease.34
Conclusion
Numerous ongo­ing tri­als dem­on­strate both phy­si­cian and
patient aspi­ ra­
tions for time-lim­ ited ther­
a­pies. Combining tar­
geted agents is very prom­is­ing, but it is impor­tant to keep in
mind that fol­low-up is still short and that infor­ma­tion on the effi­
cacy of sub­se­quent lines of ther­apy is lacking. Sensitive MRD
should help tai­lor the opti­mal dura­tion of the var­i­ous strat­eg
­ ies.
Today, a deci­sion between con­tin­u­ous treat­ment with BCR
inhib­i­tors and VEN-based com­bi­na­tions relies mostly on comor­
bidities, comedications, and patient/phy­si­cian pref­er­ences. The
rap­idly increas­ ing under­ stand­ ing of genetic back­ ground and
clonal evo­lu­tion is not yet inte­grated into ther­a­peu­tic deci­sions,
but in the com­ing years it could sup­plant the “one-size-fits-all­”
ten­dency and allow more per­son­al­ized approaches.
Conflict-of-inter­est dis­clo­sure
Vincent Lévy: hon­ o­raria: Abbvie, Astra-Zeneca, CSL Behring,
Janssen.
Alain Delmer: hon­o­raria: Janssen, Abbvie, AstraZeneca, Roche.
Florence Cymbalista: hon­o­raria: Janssen, Abbvie, AstraZeneca,
Roche.
Off-label drug use
Vincent Lévy: nothing to disclose.
Alain Delmer: nothing to disclose.
Florence Cymbalista: nothing to disclose.
Correspondence
Florence Cymbalista, Hematology Biology, CHU Paris Seine-
Saint-Denis - Hôpital Avicenne, 125 rue de Stalingrad, Bobigny,
France 93009; e-mail: florence.cymbalista@aphp.fr.
References
1. Woyach JA, Ruppert AS, Heerema NA, et  al. Ibrutinib reg­i­mens ver­sus
chemoimmunotherapy in older patients with untreated CLL. N Engl J Med.
2018;379(26):2517-2528.
2. Burger JA, Barr PM, Robak T, et al. Long-term effi­cacy and safety of first-line
ibrutinib treat­ment for patients with CLL/SLL: 5 years of fol­low-up from the
phase 3 RESONATE-2 study. Leukemia. 2020;34(3):787-798.
3. Al-Sawaf O, Zhang C, Tandon M, et al. Venetoclax plus obinutuzumab ver­sus
chlorambucil plus obinutuzumab for pre­vi­ously untreated chronic lym­pho­
cytic leu­kae­mia (CLL14): fol­low-up results from a multicentre, open-label,
randomised, phase 3 trial. Lancet Oncol. 2020;21(9):1188-1200.
4. Hallek M, Cheson BD, Catovsky D, et  al. iwCLL guide­lines for diag­no­sis,
indi­ca­tions for treat­ment, response assess­ment, and sup­port­ive man­age­
ment of CLL. Blood. 2018;131(25):2745-2760.
66  |  Hematology 2021  |  ASH Education Program
5. Eichhorst B, Robak T, Montserrat E, et al; Euro­pean Society for Medical
Oncology Guidelines Committee. Chronic lym­pho­cytic leu­kae­mia: ESMO
Clinical Practice Guidelines for diag­no­sis, treat­ment and fol­low-up. Ann
Oncol. 2021;32(1):23-33.
6. Quinquenel A, Aurran-Schleinitz T, Clavert A, et al. Diagnosis and treat­ment
of chronic lym­pho­cytic leu­ke­mia: rec­om­men­da­tions of the French CLL
Study Group (FILO). Hemasphere. 2020;4(5):e473.
7. Gribben JG, Bosch F, Cymbalista F, et al. Optimising out­comes for patients
with chronic lym­pho­cytic leu­kae­mia on ibrutinib ther­apy: Euro­pean rec­
om­men­da­tions for clin­i­cal prac­tice. Br J Haematol. 2018;180(5):666-679.
8. Gribben JG. Practical man­age­ment of tumour lysis syn­drome in veneto­
clax-treated patients with chronic lym­pho­cytic leu­kae­mia. Br J Haematol.
2020;188(6):844-851.
9. Mauro FR, Giannarelli D, Galluzzo CM, et  al. Response to the con­ju­gate
pneu­mo­coc­cal vac­cine (PCV13) in patients with chronic lym­pho­cytic leu­
ke­mia (CLL). Leukemia. 2021;35(3):737-746.
10. Herishanu Y, Avivi I, Aharon A, et  al. Efficacy of the BNT162b2 mRNA
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/59/1851402/59levy.pdf by guest on 13 December 2021
COVID-19 vac­cine in patients with chronic lym­pho­cytic leu­ke­mia. Blood.
2021;137(23):3165-3173.
11. Fischer K, Bahlo J, Fink AM, et al. Long-term remis­sions after FCR chemoim­
munotherapy in pre­vi­ously untreated patients with CLL: updated results of
the CLL8 trial. Blood. 2016;127(2):208-215.
12. Moreno C, Greil R, Demirkan F, et al. Ibrutinib plus obinutuzumab ver­sus
chlorambucil plus obinutuzumab in first-line treat­ ment of chronic lym­
pho­cytic leu­kae­mia (iLLUMINATE): a multicentre, randomised, open-label,
phase 3 trial. Lancet Oncol. 2019;20(1):43-56.
13. Zenz T, Eichhorst B, Busch R, et al. TP53 muta­tion and sur­vival in chronic
lym­pho­cytic leu­ke­mia. J Clin Oncol. 2010;28(29):4473-4479.
14. Campo E, Cymbalista F, Ghia P, et al. TP53 aber­ra­tions in chronic lym­pho­
cytic leu­ke­mia: an over­view of the clin­ic
­ al impli­ca­tions of improved diag­nos­
tics. Haematologica. 2018;103(12):1956-1968.
15. Eichhorst B, Fink AM, Bahlo J, et al; inter­na­tional group of inves­ti­ga­tors;
Ger­man CLL Study Group. First-line chemoimmunotherapy with benda­
mustine and rituximab ver­sus fludarabine, cyclo­phos­pha­mide, and ritux­
imab in patients with advanced chronic lym­pho­cytic leu­kae­mia (CLL10): an
inter­na­tional, open-label, randomised, phase 3, non-infe­ri­or­ity trial. Lancet
Oncol. 2016;17(7):928-942.
16. Goede V, Fischer K, Busch R, et  al. Obinutuzumab plus chlorambucil in
patients with CLL and coexisting con­di­tions. N Engl J Med. 2014;370(12):1101-
1110.
17. Shanafelt TD, Wang XV, Kay NE, et al. Ibrutinib-rituximab or chemoimmuno­
therapy for chronic lym­pho­cytic leu­ke­mia. N Engl J Med. 2019;381(5):432-
443.
18. Landau DA, Tausch E, Taylor-Weiner AN, et al. Mutations driv­ing CLL and
their evo­lu­tion in pro­gres­sion and relapse. Nature. 2015;526(7574):525-530.
19. Burns A, Alsolami R, Becq J, et al. Whole-genome sequenc­ing of chronic
lym­pho­cytic leu­kae­mia reveals dis­tinct dif­fer­ences in the muta­tional land­
scape between IgHVmut and IgHVunmut sub­groups. Leukemia. 2018;32(2):
332-342.
20. Baliakas P, Jeromin S, Iskas M, et al; Euro­pean Research Initiative on CLL.
Cytogenetic com­plex­ity in chronic lym­pho­cytic leu­ke­mia: def­i­ni­tions,
asso­ci­a­tions, and clin­i­cal impact. Blood. 2019;133(11):1205-1216.
21. Lazarian G, Guièze R, Wu CJ. Clinical impli­ca­tions of novel geno­mic dis­cov­
er­ies in chronic lym­pho­cytic leu­ke­mia. J Clin Oncol. 2017;35(9):984-993.
22. Gruber M, Bozic I, Leshchiner I, et  al. Growth dynam­ics in nat­u­rally pro­
gressing chronic lym­pho­cytic leu­kae­mia. Nature. 2019;570(7762):474-479.
23. Arnon Kater A, Carolyn Owen C, Carol Moreno C, et al. Fixed-dura­tion ibru­
tinib and venetoclax (I + V) ver­sus chlorambucil plus obinutuzumab (CLB+O)
for first line (1L) chronic lym­pho­cytic leu­ke­mia (CLL): pri­mary anal­y­sis of the
phase 3 GLOW study. EHA Library. 6 June 2021. Abstract LB1902.
24. Michallet AS, Dilhuydy MS, Subtil F, et  al. Obinutuzumab and ibrutinib
induc­tion ther­apy followed by a min­im ­ al resid­ual dis­ease-driven strat­egy
in patients with chronic lym­pho­cytic leu­kae­mia (ICLL07 FILO): a sin­gle-arm,
multicentre, phase 2 trial. Lancet Haematol. 2019;6(9):e470-e479.
25. Davids MS, Brander DM, Kim HT, et al; Blood Cancer Research Partnership
of the Leukemia and Lymphoma Society. Ibrutinib plus fludarabine, cyclo­
phos­pha­mide, and rituximab as ini­tial treat­ment for youn­ger patients with
chronic lym­pho­cytic leu­kae­mia: a sin­gle-arm, multicentre, phase 2 trial.
Lancet Haematol. 2019;6(8):e419-e428.
26. Jain N, Thompson PA, Burger JA, et  al. Ibrutinib, fludarabine, cyclo­phos­
pha­mide, and obinutuzumab (iFCG) reg­i­men for chronic lym­pho­cytic leu­
ke­mia (CLL) with mutated IGHV and with­out TP53 aber­ra­tions. Leukemia.
Published online 18 May 2021.
27. Jain N, Keating M, Thompson P, et al. Ibrutinib and venetoclax for first-line
treat­ment of CLL. N Engl J Med. 2019;380(22):2095-2103.
28. Wierda WG, Tam CS, Allan JN, et  al. Ibrutinib (Ibr) plus venetoclax (Ven)
for first-line treat­ment of chronic lym­pho­cytic leu­ke­mia (CLL)/small lym­
pho­cytic lym­phoma (SLL): 1-year dis­ease-free sur­vival (DFS) results from
the MRD cohort of the phase 2 CAPTIVATE study. Blood. 2020;136(suppl 1):
16-17, abstract 123.
29. Thompson PA, Tam CS, O’Brien SM, et  al. Fludarabine, cyclo­ phos­pha­
mide, and rituximab treat­ment achieves long-term dis­ease-free sur­vival in
IGHV-mutated chronic lym­pho­cytic leu­ke­mia. Blood. 2016;127(3):303-309.
30. Al-Sawaf O, Zhang C, Robrecht S, et  al. Clonal dynam­ics after veneto­
clax-obinutuzumab ther­apy: novel insights from the ran­dom­ized, phase 3
CLL14 trial. Blood. 2020;136(suppl 1):22-23.
31. Kovacs G, Robrecht S, Fink AM, et al. Minimal resid­ual dis­ease assess­ment
improves pre­dic­tion of out­come in patients with chronic lym­pho­cytic leu­
ke­mia (CLL) who achieve par­tial response: com­pre­hen­sive anal­y­sis of two
phase III stud­ies of the Ger­man CLL study group. J Clin Oncol. 2016;34(31):
3758-3765.
Prakash Singh Shekhawat
32. Euro­pean Medicines Agency Committee for Medicinal Products for Human
Use. Appendix 4 to the guide­line on the eval­u­a­tion of anti­can­cer medic­i­nal
prod­ucts in man [updated 15 Feb­ru­ary 2016]. Accessed 10 Jan­u­ary 2018.
http:​­/​­/www​­.ema​­.europa​­.eu​­/docs​­/en_GB​­/document_library​­/Scientific​
­_guideline​­/2016​­/02​­/WC500201945​­.pdf.
33. Letestu R, Dahmani A, Boubaya M, et al; French Innovative Leukemia Orga­
nization. Prognostic value of high-sen­si­tiv­ity mea­sur­able resid­ual dis­ease
assess­ment after front-line chemoimmunotherapy in chronic lym­pho­cytic
leu­ke­mia. Leukemia. 2021;35(6):1597-1609.
34. Fürstenau M, De Silva N, Eichhorst B, Hallek M. Minimal resid­ual dis­ease
assess­ment in CLL: ready for use in clin­i­cal rou­tine? Hemasphere. 2019;3(5):
e287.
© 2021 by The Amer­i­can Society of Hematology
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/59/1851402/59levy.pdf by guest on 13 December 2021
DOI 10.1182/hema­tol­ogy.2021000233
Up-front ther­apy: time-lim­ited treat­ment  |  67
 CLL: EXTENDING SURVIVAL

Is there a role for anti-CD20 antibodies in CLL?

Harsh R. Shah and Deborah M. Stephens

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/68/1851825/68shah.pdf by guest on 13 December 2021

University of Utah, Salt Lake City, UT

Anti-CD20 monoclonal antibodies (mAbs) have revolutionized the treatment of chronic lymphocytic leukemia (CLL) by
improving survival of patients with CLL in conjunction with chemotherapy. However, the novel targeted agents such
as Bruton tyrosine kinase inhibitors (BTKis) and venetoclax have now mostly replaced chemotherapy in frontline treat-
ment of CLL. Several clinical trials have been conducted to examine the role of anti-CD20 mAbs in combination with
BTK inhibitors and venetoclax. Addition of rituximab to ibrutinib does not improve progression-free survival (PFS) of
treatment-naive patients with CLL, possibly related to ibrutinib’s antagonistic effect on anti-CD20 antibodies. Alterna-
tively, addition of a glycoengineered anti-CD20 mAb obinutuzumab to a more selective BTKi acalabrutinib may improve
PFS but does not improve overall survival of patients with CLL in the frontline setting, pending long-term follow-up. Thus,
we suggest that the addition of an anti-CD20 mAb to a BTKi is of most benefit to patients with autoimmune cytopenia
or rapidly progressive disease. In contrast to BTKis, combination of fixed-duration venetoclax and anti-CD20 mAb can
induce deep remission with high rates of undetectable minimal residual disease, correlating with improved survival of
patients with CLL in both frontline and relapsed/refractory settings. In this review, we discuss clinical trials of BTKis and
venetoclax that have investigated the role of anti-CD20 mAbs in frontline and relapsed settings of CLL treatment. We
also provide an algorithm suggesting how anti-CD20 mAbs may be incorporated in the treatment of patients with CLL,
including specific scenarios.

LEARNING OBJECTIVES
• Anti-CD20 monoclonal antibody plus BTK inhibitor may provide clinical benefit in patients with active autoimmune
cytopenia or rapidly progressive disease.
• Adding anti-CD20 antibody to venetoclax induces deep remissions, allowing for fixed-duration treatment.

I antibody that eradicates CLL cells primarily by means of

CLINICAL CASE
Mr. Smith, a 67-year-old man, was diagnosed with chronic
lymphocytic leukemia (CLL) 3 years ago. He was asymp-
tomatic, without cytopenias, and was observed. Three
months ago, he had purpura and thrombocytopenia.
Evaluation revealed immune thrombocytopenic purpura
(ITP). Over the course of 2 months, his ITP was refractory
to treatment with steroids and intravenous immunoglob-
ulin. Today, Mr. Smith reports drenching night sweats,
fatigue, and early satiety. Laboratory tests show anemia
and thrombocytopenia. He has refractory ITP and meets
criteria for CLL treatment.
Introduction
Anti-CD20 monoclonal antibodies (mAbs) have been inte-
grated in treatment of CLL in combination with chemother-
apy for longer than a decade. Rituximab is a chimeric type
complement-dependent cytotoxicity and antibody-depen-
dent cellular cytotoxicity (ADCC) after binding to CD20
(Figure 1).1 Addition of rituximab to the standard chemo-
therapy agents fludarabine and cyclophosphamide (FCR)
improved overall survival (OS) and progression-free survival
(PFS) compared with fludarabine and cyclophophamide in
the frontline treatment of patients with CLL. This combina-
tion became a standard option for fit patients.2 Although
bendamustine (another standard chemotherapy agent)
with or without rituximab was not directly compared in
a head-to-head clinical trial, bendamustine plus rituximab
(BR) showed favorable overall response rates (ORRs) and
PFS compared with bendamustine alone in treatment-
naive patients with CLL in an imprecise cross-trial com-
parison.3,4 The direct comparison of FCR to BR for treat-
ment-naive fit patients with CLL concluded, though, that
BR was better tolerated in patients older than 65 years and
is considered an option for this population.5 Obinutuzumab
is a potent humanized, glycoengineered type II antibody

68 | Hematology 2021 | ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/68/1851825/68shah.pdf by guest on 13 December 2021
Figure 1. Mechanism of action of select anti-CD20 monoclonal antibodies used to treat CLL. More “+” sign indicates stronger
mechanism of action. ADCP, antibody-dependent cellular phagocytosis; CDC, complement dependent cytotoxicity; MAC, membrane
attack complex.

that also tar­gets CD20 on the cell sur­face but with enhanced
ADCC, anti­body-depen­dent phago­cy­to­sis, and direct cell death
effects com­pared with rituximab (Figure 1).6,7 In treat­ment-naive
patients with CLL with preexisting con­ di­
tions, rituximab or
obinutuzumab in com­bi­na­tion with chlorambucil was com­pared
head-to-head.8 The com­bi­na­tion of chlorambucil and obinutu-
zumab (Chl-O) resulted in higher com­plete remis­sion (CR) rates,
PFS, and OS com­pared with the rituximab com­bi­na­tion, pro­vid­
ing evi­dence to sup­port the supe­ri­or­ity of obinutuzumab com­
pared with rituximab.8,9 Based on these piv­otal stud­ies, the use
Prakash Singh Shekhawat
of anti-CD20 mAbs has a firmly established role in the front­line
treat­ment of CLL when used in com­bi­na­tion with stan­dard che­
mo­ther­apy agents. Over the past few years, the stan­dard of
care has shifted from stan­dard che­mo­ther­apy agents to using
targeted kinase inhib­i­tors, such as Bruton tyro­sine kinase inhib­i­
tors (BTKis) or B-cell lym­phoma 2 (BCL2) inhib­i­tors. It is not clear
whether anti-CD20 mAbs are required for effi­cacy when used in
com­bi­na­tion with these targeted kinase inhib­i­tors. In this study,
we review the avail­­able data surrounding anti-CD20 mAb com­
bi­na­tions in CLL treat­ment.
Anti-CD20 antibodies in CLL | 69
BTK Inhibitors
BTK is a kinase within the TEC fam­ily that leads to down­stream
acti­va­tion of AKT, extra­cel­lu­lar sig­nal–reg­u­lated kinase, and nu-
clear fac­tor κ light-chain enhancer of acti­vated B cells, path­ways
impor­tant for malig­nant B cells’ growth and sur­vival.10-12 Ibrutinib
is an irre­vers­ible BTKi that cova­lently binds to the cys­te­ine 481
amino acid of the BTK enzyme. It has shown sig­nif­i­cant activ­ity in
patients with CLL in both front­line and relapsed/refrac­tory (R/R)
set­tings.13,14 Because of the dura­bil­ity of responses with BTKis as
monotherapy, there is a ques­tion of whether there is a need for
an anti-CD20 mAb in con­junc­tion to improve clin­i­cal effi­cacy.
Ibrutinib ± rituximab or obinutuzumab
The E1912 trial com­pared FCR to ibrutinib plus rituximab (IR) in
treat­ment-naive patients youn­ger than 70 years (Table 1).15 This
study dem­on­strated prolonged PFS and OS of patients receiv­ing
IR com­pared with FCR. Results of this trial led to approval of the
IR com­bi­na­tion in front­line treat­ment of CLL, but this study did
not answer whether addi­tion of rituximab to ibrutinib is nec­es­
sary to obtain sim­i­lar clin­i­cal ben­e­fit.
The A041202 study eval­ua ­ ted front­line ther­apy with ibrutinib
vs IR vs BR in patients with CLL 65 years or older.16 This study
dem­on­strated prolonged PFS with ibrutinib-based treat­ments
com­pared with BR. However, there was no dif­fer­ence in PFS
and OS ben­e­fits between ibrutinib alone and the IR com­bi­na­
tion. Rates of CR (12% vs 7%) and unde­tect­able min­i­mal resid­
ual dis­ease (uMRD) (4% vs 1%) were only mar­gin­ally higher with
the IR com­bi­na­tion. The safety pro­file of both ibrutinib alone and
the IR com­bi­na­tion was sim­il­ar except for slightly higher grade
3 or more neutropenia with IR. Another study (NCT02007044)
com­par­ing ibrutinib vs IR in treat­ment-naive (13% of total) and
R/R (87% of total) set­tings also failed to detect improve­ment
in PFS with IR but con­cluded that time to CR was faster (11.5 vs
22.2 months), and patients had lower lev­els of resid­ual dis­ease
as assessed by flow cytom­e­try with the IR com­bi­na­tion.17 Con-
sidering the existing data and con­tin­u­ous appli­ca­tion of BTKis,
the sig­nif­i­cance of uMRD and its rela­tion to long-term dis­ease
erad­i­ca­tion has not been well established. Based on these 2
stud­ies, addi­tion of rituximab to ibrutinib does not seem to pro­
vide any clin­i­cally mean­ing­ful sur­vival ben­e­fit to most patients
with treat­ment-naive CLL. In a pre­clin­i­cal study, ibrutinib’s off-
tar­get inhi­bi­tion of inter­leu­kin 2 induc­ible tyro­sine kinase led to
downregulation of nat­u­ral killer (NK) cell-medi­ated ADCC and
impaired anti­body-depen­dent phago­cy­to­sis, both of which are
known anti-CD20 mAb mech­a­nisms of action (Figure 1).20,21 This
find­ing has been hypoth­e­sized as one of the expla­na­tions for the
lack of sig­nif­i­cant sur­vival ben­e­fit with the addi­tion of rituximab
to ibrutinib.
The iLLUMINATE study com­ pared front­ line CLL ther­apy of
ibrutinib plus obinutuzumab vs Chl-O.18 This study dem­on­strated
prolonged PFS with the ibrutinib-based reg­i­men in treat­ment-
naive patients with CLL. Adding obinutuzumab to ibrutinib deep­
ened responses (35% with uMRD), but seri­ous adverse events
were more com­mon in this arm. This study led to Food and Drug
Administration (FDA) approval for mar­ket­ing of the ibrutinib and
obinutuzumab com­bi­na­tion in the front­line set­ting. However, no
pro­spec­tive head-to-head com­par­i­sons of ibrutinib vs ibrutinib
plus obinutuzumab have been done, leav­ ing the ques­ tion of
the clin­i­cal sig­nif­i­cance of obinutuzumab’s addi­tion to ibrutinib
unanswered.
70  |  Hematology 2021  |  ASH Education Program
Acalabrutinib ± obinutuzumab
Acalabrutinib is a highly selec­tive BTKi with less off-tar­get inhi­
bi­tion, includ­ing induc­ible tyro­sine kinase, com­pared with
ibrutinib.22,23 In pre­clin­i­
cal stud­ ies, acalabrutinib did not inhibit
anti-CD20 mAb-depen­dent NK-cell medi­ated cyto­tox­ic­ity.22 The
phase 3 ELEVATE treatment-naive (TN) study was designed to an-
swer whether acalabrutinib is supe­rior to chemoimmunotherapy
(Chl-O) in terms of PFS in the front­line treat­ment of CLL. It also
eval­u­ated whether there is a ben­e­fit of adding a more potent an-
ti-CD20 mAb obinutuzumab to acalabrutinib (AO), although the
study was not powered to detect this dif­fer­ence.19 As expected,
patients who received acalabrutinib ±​­ obinutuzumab had improved
PFS com­pared with Chl-O. After 28 months of fol­low-up, the 2-year
PFS rates were not clin­ic ­ ally or sta­tis­ti­cally dif­fer­ent between the
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/68/1851825/68shah.pdf by guest on 13 December 2021
AO and acalabrutinib (93% vs 87%) groups. More patients with
AO had CR (13% vs 1%) and uMRD (56% vs 7% of those eval­u­ated)
com­pared with acalabrutinib alone. AO resulted in higher rates of
grade 3 or higher neutropenia, infec­tions, and infu­sion reac­tions
com­pared with acalabrutinib alone. Findings from this study ulti­
mately led to FDA approval for mar­ket­ing of acalabrutinib ± obinu-
tuzumab for front­line treat­ment of patients with CLL.
As we await the long-term results of the ELEVATE TN study,
the mature data from the orig­i­nal phase 2 study of sin­gle-agent
acalabrutinib in treat­ment-naive CLL are encour­ag­ing.24 After a
median fol­low-up of 53 months, median event-free sur­vival and
dura­tion of response are not reached with a very high ORR of 97%
(7% CR, 90% partial response). In addi­tion, 86% of patients are
still receiv­ing treat­ment, and only 6% have discontinued due to
adverse events. The first data describ­ing a head-to-head com­par­
i­son of acalabrutinib and ibrutinib in patients with R/R CLL were
presented at the Amer­ic ­ an Society of Clinical Oncology meet­ing
in 2021 (ELEVATE Relapsed/Refractory (RR) Study).25 This study
found that acalabrutinib was noninferior to ibrutinib in terms of
PFS and had a bet­ter tox­ic­ity pro­file—impor­tantly, a lower inci­
dence of atrial fibril­ la­tion and hyper­ ten­sion.25 This improved
tox­ic­ity pro­file makes acalabrutinib a very appeal­ing treat­ment
option for treat­ment-naive patients with CLL, espe­cially those
who are elderly or have mul­ti­ple med­ic ­ al comorbidities.
Summary
Of the described tri­als that com­pared a BTKi plus anti-CD20 mAb
vs BTKi alone, none dem­on­strated a clin­i­cally mean­ing­ful supe­
ri­or­ity of the anti-CD20 mAb com­bi­na­tion in terms of improved
PFS or OS. The depth of response (CR and uMRD rates) and the
speed of response (time to CR) are mar­gin­ally improved with the
addi­tion of an anti-CD20 mAb. As such, there are 2 sce­nar­ios in
which addi­tion of an anti-CD20 mAb to a BTKi may be clin­i­cally
mean­ing­ful. The first sce­nario is in the pres­ence of an active and
ste­roid-refrac­tory auto­im­mune cytopenia such as ITP or auto­
im­mune hemo­lytic ane­mia. Both rituximab and BTKi have been
effec­tive as monotherapy in con­trol or res­ol­u­tion of auto­im­mune
cytopenia related to CLL.26-31 Thus, a com­bi­na­tion might lead to
deeper and more rapid con­trol of auto­im­mune cytopenia, espe­
cially in the ste­roid-refrac­tory patients. The sec­ond sce­nario, in
which the upfront addi­tion of a mAb to BTKi could be help­ful, is
in patients need­ing rapid con­trol of dis­ease (eg, lymph­ade­nop­
a­thy endan­ger­ing a crit­i­cal organ). From a prac­ti­cal stand­point,
BTKis can take time to be autho­rized by insur­ance and are not
often included on inpa­tient hos­pi­tal for­mu­lar­ies sec­ond­ary to the
high cost of these ther­ap ­ ies. Anti-CD20 mAbs are often read­ily
 Table 1.  Frontline CLL stud­ies with BTKi ± anti-CD20 mono­clo­nal anti­body

Median
Targeted N (targeted fol­low-up,
Study reg­i­men Comparator reg­i­men) ORR, % CRR, % uMRD, % PFS OS Notable AEs mo
E191215 IR FCR IR = 354 95 17 8 (*n = 276) NR (3-y 89%) NR (3-y 98%) Grade ≥3 infec­tions: FCR = 20.3% 33.6
vs IR = 10.5%; grade ≥3 HTN:
FCR = 8.2% vs IR = 18.8%
A04120216 I ± R BR I = 182 I = 93 I = 7 I = 1 I = NR (2-y 87%) I = NR (2-y 90%) Grade ≥3 neutropenia: 38
IR = 182 IR = 94 IR = 12 IR = 4 IR = NR (2-y 88%) IR = NR (2-y 94%) IR = 21% vs I = 15%
NCT0200704417 IR Ibrutinib I = 104 I = 92.3 I = 20.2 NA I = NR (3-y 86%) I = NR (3-y 89%) No sig­nif­i­cant dif­fer­ence in 36
IR = 104 IR = 92.3 IR = 26 IR = NR (3-y 86.9%) IR = NR (3-y 92%) tox­ic­ity pro­file
TN = 27
RR = 181

Prakash Singh Shekhawat

iLLUMINATE18 IO Chl-O IO = 113 88 19 35 (*n = 113) NR (30-mo 79%) NR (30-mo 86%) SAEs in 58% of IO and 35% of 31.3
Chl-O; obinutuzumab-related
SAEs (febrile neutropenia,
throm­bo­cy­to­pe­nia) in 15% of
the IO arm
ELEVATE TN19 A ± O Chl-O A = 179 A = 86 A = 1 A = 7% of those in A = NR (2-y 87%) A = NR (2-y 95%) Grade ≥3 neutropenia: AO = 30% 28.3
AO = 179 AO = 94 AO = 13 CR/CRi (*n = 14) AO = NR (2-y 93%) AO = NR (2-y 95%) vs A = 10%; grade ≥3 infec­tions:
AO = 56% of those in AO = 21% vs A = 14%
CR/CRi (*n = 43)
*Of those evaluable for uMRD in the bone mar­row/periph­eral blood.
A, acalabrutinib; AE, adverse event; CRi, com­plete response with incom­plete count recov­ery; CRR, com­plete response rate; HTN, hypertension; I, ibrutinib; IO, ibrutinib and obinutuzumab;
NA, not avail­­able; NR, not reached; RR, relapsed/refractory; SAE, seri­ous adverse event; TN, treatment-naive.

Anti-CD20 antibodies in CLL | 71

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/68/1851825/68shah.pdf by guest on 13 December 2021
avail­­able and can expe­dite treat­ment for patients who need a
rapid response. For all­other patients, we rec­om­mend the use
of ibrutinib or acalabrutinib alone in front­line treat­ment of CLL.
CLINICAL CASE (Con­t in­u ed)
Mr. Smith received 6 cycles of obinutuzumab and con­tin­u­ous
oral acalabrutinib and achieved clin­ic
­ al remis­sion. Three years
later, he sought treat­ment from the clinic with a pro­gres­sive
return of drenching night sweats and decline of hemo­glo­bin.
No evi­dence of hemo­ly­sis was detected. A next-gen­er­a­tion
sequenc­ing panel showed a muta­tion of cys­te­ine to ser­ine at
the 481 moi­ety of BTK (C481S).
Venetoclax + rituximab
Venetoclax is an oral, highly selec­tive BCL2 homol­ogy domain
3-mimetic that binds to BCL2, resulting in ini­ti­a­tion of apo­pto­sis
medi­ated by BCL2-asso­ci­ated pro­tein X and BCL2-antag­o­nist/
killer in primed cells.32 Single-agent venetoclax showed prom­is­ing
activ­ity in patients with R/R CLL who had pre­vi­ously progressed
on ibrutinib or idelalisib or had 17p dele­tion.33-35 However, the
com­bi­na­tion of venetoclax with rituximab was found to be a ­ ble
to over­come micro­en­vi­ron­ment-induced resis­tance to veneto-
clax in pre­clin­i­cal stud­ies, lead­ing to its fur­ther inves­ti­ga­tion in
patients with R/R CLL.36 In the phase 1 study of venetoclax plus
rituximab (V-R) in R/R CLL, no sig­nif­i­cant addi­tional toxicities
were found with addi­tion of rituximab. The com­bi­na­tion resulted
in a 2-year PFS of 82% with a 51% CR rate and 57% of the patients
achiev­ing uMRD.37 The depth and dura­ bil­
ity of the response
observed with V-R pop­u­lar­ized the con­cept of a time-lim­ited
course in the treat­ment of patients with CLL, a treat­ment strat­
egy to avoid costs and tox­ic­ity from con­tin­uo ­ us drug expo­sure.
The phase 3 ran­dom­ized MURANO trial com­pared the time-
lim­ited 2-year course of V-R with BR in patients with R/R CLL
(Table 2).38 Rates of 4-year PFS (57% vs 5%) and OS (85% vs 67%)
were higher with V-R com­pared with BR.39 At the end of com­bi­na­
tion ther­apy, 62.4% of patients with V-R had uMRD com­pared with
13.3% of those with BR, and this end point was asso­ci­ated with
Table 2.  Select piv­otal CLL stud­ies with venetoclax + anti-CD20 mono­clo­nal anti­body
N Median
Targeted (targeted
Study Population reg­i­men Comparator reg­i­men) ORR, % CRR, % uMRD, %
MURANO 37-43
R/R V-R BR 194 92
CLL1439-44 Treatment V-O Chl-O 216 85
naive
*Of those evaluable for uMRD in the bone mar­row/periph­eral blood.
CRR, com­plete response rate; NR, not reached.
72  |  Hematology 2021  |  ASH Education Program
supe­rior PFS in both groups. Based on the results of the MURANO
study, the com­bi­na­tion of fixed-dura­tion V-R has been approved
for treat­ment in patients with R/R CLL. No ran­dom­ized study
com­ par­
ing venetoclax vs venetoclax plus rituximab has been
reported. Due to the time-lim­ited nature of this reg­i­men, the deep
responses are more desir­able, and as such, the V-R com­bi­na­tion
has become stan­dard of care when used in the R/R set­ting.
Venetoclax + obinutuzumab
Alternatively, in the front­line treat­ment of CLL, venetoclax has
been stan­dardly paired with obinutuzumab (V-O). The phase 3
CLL14 study com­pared front­line CLL ther­apy with a fixed 1-year
dura­tion of V-O vs Chl-O.40 Patients who received V-O had sig­nif­
i­cantly lon­ger PFS than patients receiv­ing Chl-O (not reached vs
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/68/1851825/68shah.pdf by guest on 13 December 2021
35.6 months).41 Of the patients who received V-O, 76% achieved
uMRD com­pared with 35% for the Chl-O group 3 months after
com­ple­tion of ther­apy. Based on this study, V-O has been ap-
proved by the FDA for mar­ket­ing as a fixed-dura­tion ther­apy for
treat­ment-naive patients with CLL.
Summary
Although the evi­dence from a direct ran­dom­ized study is lack-
ing, com­bi­na­tion of an anti-CD20 mAb and venetoclax in CLL
allows for time-lim­ited ther­apy by achiev­ing clin­i­cally rel­e­vant
improve­ment in PFS, OS, and uMRD rates. Thus, we rec­om­mend
a fixed course of V-R for 2 years in the R/R set­ting and a fixed
course of V-O in the front­line set­ting.
Umbralisib + ublituximab
Ublituximab is a type I glycoengineered anti-CD20 mAb with en-
hanced ADCC that tar­gets a unique epi­tope on CD20 (Figure 1).
Ublituximab has been recently com­bined with umbralisib, a dual
inhib­i­tor of phosphoinositide 3-kinase δ (PI3Kδ) and casein ki-
nase 1ε. In the phase 3 UNITY CLL trial, this com­bi­na­tion (U2)
yielded improved 2-year PFS of 76.6% com­pared with 52.1% with
Chl-O in treat­ment-naive patients and 2-year PFS of 41.3% with
U2 com­pared with 24.8% with Chl-O in patients with R/R CLL.42
The safety pro­file of umbralisib was favor­able com­pared with
his­tor­i­cal data with other PI3K inhib­i­tors. As such, the U2 com­
bi­na­tion will likely be sub­mit­ted for eval­u­at­ion of FDA approval
within the next few years.
fol­low-up,
PFS OS Notable AEs mo
27 62 (*n = 194) 53.6 mo NR (5-y More grade ≥3 59.2
Peripheral 82.1%) neutropenia with
blood V-R; more grade ≥3
febrile neutropenia
and infec­tions with
BR
50 76 (*n = 216) NR (4-y NR Grade ≥3 neutropenia 52.4
Peripheral 74%) (4-y (53%) and throm­
blood 85%) bo­cy­to­pe­nia (13%)
of V-O
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/68/1851825/68shah.pdf by guest on 13 December 2021
Figure 2.  How to incorporate anti-CD20 monoclonal antibody into frontline treatment for patients with CLL. *All eligible patients
should be considered for participation in clinical trials if available. **Anti-C20 monoclonal antibody. +Pending Food and Drug Adminis-
tration approval for marketing for CLL at time of submission. GERD, gastroesophageal reflux disease; IGHV, immunoglobulin variable
heavy chain; TLS, tumor lysis syndrome.

CLINICAL CASE (Con­t in­u ed) incor­po­ra­tion of anti-CD20 mAbs for both front­line and R/R CLL
Mr. Smith received V-R ther­ apy, which led to another clin­ i­
cal is sum­ma­rized in Figures 2 and 3. Given the mul­ti­ple choices for
remis­sion. There is some ques­tion whether main­te­nance with ther­apy, a clin­i­cal trial is always pref­er­a­ble for patients who qual­
anti-CD20 mAbs can improve clin­ i­
cal out­comes. Ofatumumab, ify. However, we have lim­ited our algo­rithms to cur­rently avail­­able
a type I human­ized anti-CD20 mAb with a bet­ter com­ple­ment- ther­a­pies. If using che­mo­ther­apy for a young patient (aged <65
depen­dent cyto­tox­ic­ity pro­file com­pared with rituximab (Figure 1), years) with good prog­nos­tic fac­tors, FCR can be con­sid­ered. If a
is approved for main­te­nance ther­apy after chemoimmunother- patient has ste­roid-refrac­tory auto­im­mune cytopenia or a need
apy in patients with R/R CLL but has not been widely accepted for an urgent response, a com­bi­na­tion of a BTKi and an anti-CD20
into clin­i­cal prac­tice due to the emer­gence of novel agents; mAb can be con­sid­ered. Due to ibrutinib’s suspected antag­o­nism
we do not rou­tinely rec­om­mend its uti­li­za­tion.43 Our algo­rithm on of an anti-CD20 mAb mech­a­nism, we would favor an acalabrutinib

Figure 3.  How to incorporate anti-CD20 monoclonal antibody into subsequent lines of treatment for patients with CLL. *All eligible
patients should be considered for participation in clinical trials if available. **Anti-C20 monoclonal antibody. +Pending Food and Drug
Administration approval for marketing for CLL at time of submission. GERD, gastroesophageal reflux disease; IGHV, immunoglobulin
variable heavy chain; TLS, tumor lysis syndrome.
Prakash Singh Shekhawat
Anti-CD20 antibodies in CLL | 73
back­bone with an anti-CD20 mAb. A fixed 1-year course of V-O
would also be accept­able in this sce­nario. In all­other patients,
we rec­om­mend monotherapy with a BTKi based on the patient’s
comorbidities (eg, acalabrutinib is pre­ ferred for frail/elderly
patients due to a favor­able tox­ic­ity pro­file and should be avoided
in patients with sig­nif­i­cant gas­tro­esoph­a­geal reflux dis­ease requir­
ing pro­ton pump inhib­i­tors due to drug inter­ac­tion) or V-O com­
bi­na­tion in front­line treat­ment. In the R/R set­ting for patients with
CLL with pre­vi­ous intol­er­ance or pro­gres­sion on a BTKi, we would
favor 2-year fixed treat­ment with a V-R com­bi­na­tion. For those
patients with venetoclax-refrac­tory dis­ease, we would use mono-
therapy with a BTKi. Idelalisib plus rituximab is also approved in
the R/R CLL set­ting, but uti­li­za­tion of this PI3K inhib­i­tor requires
care­ful man­age­ment of toxicities. The com­bi­na­tion of umbralisib
and ublituximab is an emerg­ing option in patients who are not
­able to use BTKis or venetoclax in both front­line and R/R set­tings,
but FDA approval for mar­ket­ing is pend­ing.
Conclusion
Addition of an anti-CD20 mAb to che­mo­ther­apy is con­sid­ered
stan­dard in the treat­ment of CLL. However, with the emer­gence
of potent targeted agents such as BTKis and venetoclax, the ques­
tion of the clin­i­cal impact of mAbs has been raised. Several phase
3 tri­als have shown that addi­tion of an anti-CD20 mAb to ibrutinib
does not improve the PFS or OS of patients with CLL, pos­si­bly
related to ibrutinib’s off-tar­get effects. Long-term fol­low-up of
the acalabrutinib plus obinutuzumab com­bi­na­tion is awaited. On
the con­trary, com­bi­na­tion of venetoclax and an anti-CD20 mAb
leads to deep remis­sions with uMRD, allowing for fixed-dura­tion
ther­apy and ulti­mately cor­re­lat­ing with improved sur­vival. The
role of anti-CD20 mAbs remains impor­tant for patients with CLL
who have active auto­im­mune cytopenia or need urgent response
regard­less of a com­bi­na­tion targeted ther­apy being used. Nov-
el anti-CD20 mAbs such as ublituximab are fur­ther being inves­
ti­gated in com­bi­na­tion with novel agents. As a whole, treating
phy­si­cians have a strong arma­men­tar­ium to com­bat CLL.
Acknowledgment  J Med. 2015;373(25):2425-2437.
D.S. is supported by NIH/NCI Grant K23 CA212271.
Conflict-of-inter­est dis­clo­sure
Harsh R. Shah has received research funding from Epizyme.
Deb­or­ ah M. Stephens has received research funding from Acerta,
Gilead, Karyopharm, Verastem, JUNO, Arqule, and Mingsight.
She has served on advi­sory boards for Innate, Jannsen/Pharma-
cyclics, Epizyme, Karyopharm, Beigene, Adaptive, AstraZeneca,
and TG Therapeutics.
Off-label drug use
Harsh R. Shah: Ublituximab and umbralisib are discussed and nei-
ther drug is yet FDA approved.
Deb­or­ ah M. Stephens: Ublituximab and umbralisib are discussed
and neither drug is yet FDA approved.
Correspondence
Deb­or­ah M. Stephens, Huntsman Cancer Institute, University of
Utah, 2000 Circle of Hope, Research South 5509, Salt Lake City,
UT 84112; e-mail: Deb­o­rah​­.stephens@hci​­.utah​­.edu
74  |  Hematology 2021  |  ASH Education Program
References
1. Maloney DG. Mechanism of action of rituximab. Anticancer Drugs. 2001;​
12(suppl 2):​S1-S4.
2. Fischer K, Bahlo J, Fink AM, et al. Long-term remis­sions after FCR chemoim-
munotherapy in pre­vi­ously untreated patients with CLL: updated results of
the CLL8 trial. Blood. 2016;127(2):208-215.
3. Knauf WU, Lissichkov T, Aldaoud A, et al. Phase III ran­dom­ized study of ben-
damustine com­pared with chlorambucil in pre­vi­ously untreated patients
with chronic lym­pho­cytic leu­ke­mia. J Clin Oncol. 2009;27(26):4378-4384.
4. Fischer K, Cramer P, Busch R, et  al. Bendamustine in com­bi­na­tion with
rituximab for pre­vi­ously untreated patients with chronic lym­pho­cytic leu­
ke­mia: a mul­ti­cen­ter phase II trial of the Ger­man Chronic Lymphocytic Leu-
kemia Study Group. J Clin Oncol. 2012;30(26):3209-3216.
5. Eichhorst B, Fink A-M, Bahlo J, et al; International Group of Investigators;
Ger­ man CLL Study Group (GCLLSG). First-line chemoimmunotherapy
with bendamustine and rituximab ver­sus fludarabine, cyclo­phos­pha­mide,
and rituximab in patients with advanced chronic lym­pho­cytic leu­kae­mia
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/68/1851825/68shah.pdf by guest on 13 December 2021
(CLL10): an inter­na­tional, open-label, randomised, phase 3, non-infe­ri­or­ity
trial. Lancet Oncol. 2016;17(7):928-942.
6. Freeman CL, Sehn LH. A tale of two antibodies: obinutuzumab ver­sus ritux-
imab. Br J Haematol. 2018;182(1):29-45.
7. Tobinai K, Klein C, Oya N, Fingerle-Rowson G. A review of obinutuzumab
(GA101), a novel type II anti-CD20 mono­clo­nal anti­body, for the treat­ment
of patients with B-cell malig­nan­cies. Adv Ther. 2017;34(2):324-356.
8. Goede V, Fischer K, Busch R, et  al. Obinutuzumab plus chlorambucil in
patients with CLL and coexisting con­di­tions. N Engl J Med. 2014;370(12):1101-
1110.
9. Goede V, Fischer K, Bosch F, et al. Updated sur­vival anal­y­sis from the CLL11
study: obinutuzumab ver­sus rituximab in chemoimmunotherapy-treated
patients with chronic lym­pho­cytic leu­ke­mia. Blood. 2015;126(23):1733-
1733.
10. Craxton A, Jiang A, Kurosaki T, Clark EA. Syk and Bruton’s tyro­sine kinase
are required for B cell anti­gen recep­tor-medi­ated acti­va­tion of the kinase
Akt. J Biol Chem. 1999;274(43):30644-30650.
11. Petro JB, Rahman SM, Ballard DW, Khan WN. Bruton’s tyro­sine kinase is
required for acti­va­tion of IkappaB kinase and nuclear fac­tor kappaB in
response to B cell recep­tor engage­ment. J Exp Med. 2000;191(10):1745-
1754.
12. Tomlinson MG, Woods DB, McMahon M, et al. A con­di­tional form of Bru-
ton’s tyro­sine kinase is suf­fi­cient to acti­vate mul­ti­ple down­stream sig­nal­ing
path­ways via PLC Gamma 2 in B cells. BMC Immunol. 2001;2:4.
13. Byrd JC, Furman RR, Coutre SE, et al. Ibrutinib treat­ment for first-line and
relapsed/refrac­tory chronic lym­pho­cytic leu­ke­mia: final anal­y­sis of the
piv­otal phase Ib/II PCYC-1102 study. Clin Cancer Res. 2020;26(15):3918-
3927.
14. Burger JA, Tedeschi A, Barr PM, et al; RESONATE-2 Investigators. Ibrutinib
as ini­tial ther­apy for patients with chronic lym­pho­cytic leu­ke­mia. N Engl
15. Shanafelt TD, Wang XV, Kay NE, et al. Ibrutinib-rituximab or chemoimmuno-
therapy for chronic lym­pho­cytic leu­ke­mia. N Engl J Med. 2019;381(5):432-
443.
16. Woyach JA, Ruppert AS, Heerema NA, et  al. Ibrutinib reg­i­mens ver­sus
chemoimmunotherapy in older patients with untreated CLL. N Engl J Med.
2018;379(26):2517-2528.
17. Burger JA, Sivina M, Jain N, et  al. Randomized trial of ibrutinib vs ibruti-
nib plus rituximab in patients with chronic lym­pho­cytic leu­ke­mia. Blood.
2019;133(10):1011-1019.
18. Moreno C, Greil R, Demirkan F, et al. Ibrutinib plus obinutuzumab ver­sus
chlorambucil plus obinutuzumab in first-line treat­ ment of chronic lym­
pho­cytic leu­kae­mia (iLLUMINATE): a multicentre, randomised, open-label,
phase 3 trial. Lancet Oncol. 2019;20(1):43-56.
19. Sharman JP, Egyed M, Jurczak W, et al. Acalabrutinib with or with­out obinu-
tuzumab ver­sus chlorambucil and obinutuzmab for treat­ment-naive chronic
lym­pho­cytic leu­kae­mia (ELEVATE TN): a randomised, con­trolled, phase 3 trial.
Lancet. 2020;395(10232):1278-1291.
20. Kohrt HE, Sagiv-Barfi I, Rafiq S, et  al. Ibrutinib antag­ o­nizes rituximab-
depen­dent NK cell-medi­ated cyto­tox­ic­ity. Blood. 2014;123(12):1957-1960.
21. Mercedes B, María Belén A, Enrique P, et al. Ibrutinib impairs the phago­cy­
to­sis of rituximab-coated leu­ke­mic cells from chronic lym­pho­cytic leu­ke­mia
patients by human mac­ro­phages. Haematologica. 2015;100(4):e140-e142.
22. Rajasekaran N, Sadaram M, Hebb J, et  al. Three BTK-spe­cific inhib­i­tors,
in con­trast to ibrutinib, do not antag­o­nize rituximab-depen­dent NK-cell
medi­ated cyto­tox­ic­ity. Blood. 2014;124(21):3118-3118.
23. Barf T, Covey T, Izumi R, et  al. Acalabrutinib (ACP-196): a cova­lent Bruton
tyro­sine kinase inhib­i­tor with a dif­fer­en­ti­ated selec­tiv­ity and in vivo potency
pro­file. J Pharmacol Exp Ther. 2017;363(2):240-252.
24. Byrd JC, Woyach JA, Furman RR, et  al. Acalabrutinib in treat­ment-naive
chronic lym­pho­cytic leu­ke­mia. Blood. 2021;137(24):3327-3338.
25. Byrd JC, Hillmen P, Ghia P, et al. First results of a head-to-head trial of acal-
abrutinib ver­sus ibrutinib in pre­vi­ously treated chronic lym­pho­cytic leu­ke­mia.
J Clin Oncol. 2021;39(15, suppl):7500.
26. Garcia-Chavez J, Majluf-Cruz A, Montiel-Cervantes L, Esparza M-G, Vela-
Ojeda J; Mex­i­can Hematology Study Group. Rituximab ther­apy for chronic
and refrac­tory immune throm­bo­cy­to­pe­nic pur­pura: a long-term fol­low-up
anal­y­sis. Ann Hematol. 2007;86(12):871-877.
27. Hegde UP, Wilson WH, White T, Cheson BD. Rituximab treat­ment of refrac­
tory fludarabine-asso­ci­ated immune throm­bo­cy­to­pe­nia in chronic lym­pho­
cytic leu­ke­mia. Blood. 2002;100(6):2260-2262.
28. Visco C, Barcellini W, Maura F, Neri A, Cortelezzi A, Rodeghiero F. Autoim-
mune cytopenias in chronic lym­pho­cytic leu­ke­mia. Am J Hematol. 2014;​
89(11):1055-1062.
29. Gupta N, Kavuru S, Patel D, et  al. Rituximab-based che­ mo­ ther­
apy for
ste­roid-refrac­tory auto­im­mune hemo­lytic ane­mia of chronic lym­pho­cytic
leu­ke­mia. Leukemia. 2002;16(10):2092-2095.
30. Vitale C, Salvetti C, Griggio V, et al. Pre-existing and treat­ment-emer­gent
auto­im­mune cytopenias in patients with CLL treated with targeted drugs.
Blood. 2021;137(25):3507-3517.
31. Manda S, Dunbar N, Marx-Wood CR, Danilov AV. Ibrutinib is an effec­tive
treat­ment of auto­im­mune haemolytic anae­mia in chronic lym­pho­cytic leu­
kae­mia. Br J Haematol. 2015;170(5):734-736.
32. Souers AJ, Leverson JD, Boghaert ER, et al. ABT-199, a potent and selec­tive
BCL-2 inhib­i­tor, achieves anti­tu­mor activ­ity while spar­ing plate­lets. Nat
Med. 2013;19(2):202-208.
33. Coutre S, Choi M, Furman RR, et al. Venetoclax for patients with chronic
lym­pho­cytic leu­ke­mia who progressed dur­ing or after idelalisib ther­apy.
Blood. 2018;131(15):1704-1711.
34. Stilgenbauer S, Eichhorst B, Schetelig J, et al. Venetoclax for patients with
chronic lym­pho­cytic leu­ke­mia with 17p dele­tion: results from the full pop­
u­la­tion of a phase II piv­otal trial. J Clin Oncol. 2018;36(19):1973-1980.
Prakash Singh Shekhawat
35. Jones JA, Mato AR, Wierda WG, et al. Venetoclax for chronic lym­pho­cytic
leu­kae­mia progressing after ibrutinib: an interim anal­y­sis of a multicentre,
open-label, phase 2 trial. Lancet Oncol. 2018;19(1):65-75.
36. Rachel T, Erik S, Katinka W, et  al. Resistance to ABT-199 induced by
micro­en­vi­ron­men­tal sig­nals in chronic lym­pho­cytic leu­ke­mia can be
counteracted by CD20 antibodies or kinase inhib­i­tors. Haematologica.
2015;100(8):e302-e306.
37. Seymour JF, Ma S, Brander DM, et al. Venetoclax plus rituximab in relapsed
or refrac­tory chronic lym­pho­cytic leu­kae­mia: a phase 1b study. Lancet
Oncol. 2017;18(2):230-240.
38. Seymour JF, Kipps TJ, Eichhorst B, et al. Venetoclax-rituximab in relapsed or
refrac­tory chronic lym­pho­cytic leu­ke­mia. N Engl J Med. 2018;378(12):1107-
1120.
39. Kater AP, Wu JQ , Kipps T, et  al. Venetoclax plus rituximab in relapsed
chronic lym­ pho­cytic leu­ke­
mia: 4-year results and eval­ u­
a­tion of impact
of geno­mic com­plex­ity and gene muta­tions from the MURANO phase III
study. J Clin Oncol. 2020;38(34):4042-4054.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/68/1851825/68shah.pdf by guest on 13 December 2021
40. Fischer K, Al-Sawaf O, Bahlo J, et al. Venetoclax and obinutuzumab in patients
with CLL and coexisting con­di­tions. N Engl J Med. 2019;380(23):2225-2236.
41. Al-Sawaf O, Zhang C, Robrecht S, et  al. Clonal dynam­ics after veneto-
clax-obinutuzumab ther­apy: novel insights from the ran­dom­ized, phase 3
CLL14 trial. Blood. 2020;136(suppl 1):22-23.
42. Gribben JG, Jurczak W, Jacobs RW, et al. Umbralisib plus ublituximab (U2)
is supe­rior to obinutuzumab plus chlorambucil (O+Chl) in patients with
treat­ment naïve (TN) and relapsed/refrac­tory (R/R) chronic lym­pho­cytic
leu­ke­mia (CLL): results from the phase 3 unity-CLL study. 2020;136(suppl
1):37-39.
43. van Oers M, Smolej L, Petrini M, et al. Ofatumumab main­te­nance pro­longs
pro­gres­sion-free sur­vival in relapsed chronic lym­pho­cytic leu­ke­mia: final
anal­y­sis of the PROLONG study. Blood Cancer J. 2019;9(12):98.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000234
Anti-CD20 antibodies in CLL | 75
 CLOTTING AND BLEEDING CONUNDRUMS

Unexplained arterial thrombosis:

approach to diagnosis and treatment

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/76/1851903/76may.pdf by guest on 13 December 2021

Jori E. May1 and Stephan Moll2
Department of Medicine, Division of Hematology/Oncology, University of Alabama at Birmingham, Birmingham, AL; and 2Department of Medicine,
1

Division of Hematology, University of North Carolina School of Medicine, Chapel Hill, NC

Arterial thrombotic events in younger patients without a readily apparent etiology present significant diagnostic and
management challenges. We present a structured approach to diagnosis with consideration of common causes, includ-
ing atherosclerosis and embolism, as well as uncommon causes, including medications and substances, vascular and
anatomic abnormalities, systemic disorders, and thrombophilias. We highlight areas of management that have evolved
within the past 5 years, including the use of dual-pathway inhibition in atherosclerotic disease, antithrombotic therapy
selection in embolic stroke of undetermined source and left ventricular thrombus, the role of closure of patent foramen
ovale for secondary stroke prevention, and the thrombotic potential of coronavirus disease 2019 infection and vaccina-
tion. We conclude with a representative case to illustrate the application of the diagnostic framework and discuss the
importance of consideration of bleeding risk and patient preference in determining the appropriate management plan.

LEARNING OBJECTIVES
• Perform a structured evaluation of patients with unexplained arterial thrombosis, including consideration of 6
categories of causes
• Select appropriate antithrombotic therapy and other necessary interventions to prevent recurrent arterial
thrombotic events

CLINICAL CASE ularly in the younger patient, hematologists may be called

A 36-year-old woman with no medical history sought
treatment for acute-onset left facial droop and arm weak-
ness. Computed tomography angiography (CTA) revealed
occlusion of the right middle cerebral artery at the M2
segment. She underwent endovascular thrombectomy,
resulting in complete reperfusion. Given the patient’s
young age and no apparent risk factors for stroke, hema-
tology was consulted to assist in the workup of stroke
etiology and to determine a management plan for sec-
ondary stroke prevention.
Introduction
Arterial thrombosis is a leading cause of morbidity and
mortality worldwide.1 The most common forms of arterial
thrombosis, ischemic heart disease (including acute myo-
cardial infarction), and ischemic stroke are managed by
cardiology and neurology, respectively. However, when
arterial events in these sites or in uncommon anatomic
locations occur without a readily apparent etiology, partic-
upon to assist in evaluation and management.
Herein, we present a systematic approach to the diagnos-
tic evaluation and management of patients with unexplained
arterial thrombosis while also highlighting areas where the
literature guiding diagnosis and/or treatment of specific
sites of arterial thrombosis has evolved in the past 5 years.
Defining the clot
An overview of our approach to patients with unexplained
arterial clot is presented in Figure 1. The first step when trying
to determine etiology and subsequent treatment of a throm-
botic event is to confirm the anatomic vascular location and
understand the resultant end-organ damage. Although arte-
rial thrombosis may be clearly evident in some vascular beds,
others, such as those supplying visceral organs or central ret-
inal vessels, may be mischaracterized as arterial events when
they are, in fact, venous. A dedicated review of the patient’s
presenting symptoms, as well as the imaging along with a
discussion with an expert radiologist or ophthalmologist, can
be helpful.

76 | Hematology 2021 | ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/76/1851903/76may.pdf by guest on 13 December 2021
Figure 1.  A structured approach to the diagnosis and management of unexplained arterial thrombosis.2-5 *Includes factor V Leiden,
prothrombin G20210A mutation, protein C deficiency, protein S deficiency, and antithrombin deficiency. aCL, anticardiolipin; aβ2GPI,
anti-β2 glycoprotein 1; CBC, complete blood count; CRP, C-reactive protein; DM, diabetes; ESR, erythrocyte sedimentation rate;
HbA1c, hemoglobin A1c; HLD, hyperlipidemia; HTN, hypertension; LA, lupus anticoagulant; PCR, polymerase chain reaction; PNH,
paroxysmal nocturnal hemoglobinuria; TEE, transesophageal echocardiogram; TTE, transthoracic echocardiogram; UA, urinalysis.

Once an arte­rial throm­botic event is con­firmed, a lit­er­a­ture
search for up-to-date com­ pre­
hen­sive reviews, as well as new
devel­op­ments, rel­e­vant to the site of throm­bo­sis may be a good
next step, as the approach to arte­rial events varies sig­nif­i­cantly by
ana­tomic loca­tion and may rap­idly evolve. Helpful recent reviews
for spe­cific sites of arte­rial throm­bo­sis are presented in Figure 2.6-14
Diagnosis and man­age­ment
Next, the approach to eval­u­a­tion and man­age­ment can be guid-
ed by con­sid­er­ation of 6 categories that can pre­cip­i­tate arte­rial
throm­bo­sis: (1) ath­ero­scle­ro­sis, (2) embolism, (3) sys­temic dis­or­
ders, (4) med­i­ca­tions/sub­stances, (5) vas­cu­lar/ana­tomic abnor­
mal­i­ties, and (6) thrombophilias (Figure 1). Most arte­rial events
occur due to ath­ero­scle­ro­sis and embolism, so focused eval­u­a­
tion for these causes is recommended prior to pro­ceed­ing with
eval­u­a­tion for less com­mon causes.
Importantly, in all­patients in whom antithrombotic ther­apy is
being con­sid­ered, risk of recur­rent throm­bo­sis must be bal­anced
against risk of bleed­ing, so we rec­om­mend a directed his­tory for
bleed­ing risk assess­ment. Risk cal­cu­la­tors15 may be used to guide
questioning and incor­po­rated into the risk con­ver­sa­tion with rec­
og­ni­tion that they are not val­i­dated in this rare pop­u­la­tion.
Atherosclerosis
Diagnosis: The diag­no­sis of ath­ero­scle­rotic dis­ease is often based
on a com­bi­na­tion of imag­ing find­ings and the pres­ence of risk
Prakash Singh Shekhawat
fac­tors, as well as pathol­ogy spec­i­mens if avail­­able. In imag­ing
eval­u­a­tions, iden­ti­fi­ca­tion of lumi­nal narrowing is often used to
deter­mine ath­ero­scle­rotic eti­­ol­ogy of a throm­botic event despite
the fact that vis­i­ble narrowing of an artery is a poor pre­dic­tor of
plaque vul­ner­a­bil­ity.16 This is par­tic­u­larly rel­e­vant in the eval­u­a­tion
of ische­mic stroke—tra­di­tional cri­te­ria require >50% ste­no­sis to
des­ig­nate large-artery ath­ero­scle­ro­sis, but stud­ies sup­port that
nonstenotic plaques may also rup­ture and cause stroke.17 There-
fore, eval­u­a­tion of images for the pres­ence of nonstenotic plaque
and cal­ci­fi­ca­tion may also be use­ful.
Given the lim­i­ta­tions of imag­ing, thor­ough eval­u­a­tion of ath­
ero­scle­rotic risk fac­tors should be pur­sued and, if iden­ti­fied, man­
aged aggres­sively. Lipoprotein(a) is an emerg­ing bio­marker of
ath­ero­scle­rotic car­dio­vas­cu­lar dis­ease (ASCVD) risk, and guide­
lines now rec­om­mend con­sid­er­ation of test­ing in patients with
pre­ma­ture ASCVD or ische­mic stroke (<55 years of age), fam­ily
his­tory of ASCVD, or recur­rent/pro­gres­sive ASCVD despite opti­
mal lipid-low­er­ing ther­apy.2 New treat­ments, includ­ing propro-
tein convertase sub­til­i­sin/kexin type 9 inhib­i­tors, show prom­ise
in lipo­pro­tein(a) low­er­ing and decreas­ing the risk of recur­rent
ische­mic events.
Management: Antiplatelet ther­apy has long been the foun­
da­tion of treat­ment and pre­ven­tion of ath­ero­scle­rotic events.
However, recent stud­ies in sta­ble cor­o­nary artery dis­ease18
and periph­eral artery dis­ease after revas­cu­lar­i­za­tion19 have
dem­on­strated the effi­cacy of dual-path­way inhi­bi­tion (DPI;
Unexplained arte­rial throm­bo­sis  | 77
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/76/1851903/76may.pdf by guest on 13 December 2021
Figure 2. Anatomic sites of unexplained acute arterial thrombosis with citations for relevant comprehensive reviews.6-14 *Organ
infarctions can occur due to arterial thrombosis but also venous thrombosis, low flow states, and other causes. “Defining the clot”
(ie, determining if the end-organ infarction is caused by arterial thrombosis) is a first step in determining further workup and best
management.

not to be con­ fused with dual antiplatelet ther­ apy), incor­
po­rat­ing very low-dose anticoagulation (rivaroxaban 2.5 mg
twice daily) with antiplatelet ther­apy.20 Because of sim­i­lar­
ity in nomen­cla­ture between DPI and dual antiplatelet ther­
apy, as well as sim­i­lar­ity between very low-dose rivaroxaban
2.5 mg and low-dose apixaban 2.5 mg, care­ful use of ter­mi­nol­
ogy and care­ful selec­tion of direct oral anti­co­ag­u­lant dose
are essen­tial to pre­vent med­i­ca­tion error (Table 1). The role of
DPI beyond cur­rently approved indi­ca­tions (sta­ble cor­o­nary
artery dis­ease, periph­eral artery dis­ease after revas­cu­lar­i­za­
tion) has not been stud­ied, but we expect it may be increas­
ingly con­sid­ered for events in uncom­mon ana­tomic loca­tions
where lim­ited evi­dence exists to guide antithrombotic ther­
apy selec­tion.
Embolism
Diagnosis: Embolic events can arise from the heart, where
throm­bus forms due to struc­tural or func­tional abnor­mal­i­ties,
but emboli can orig­in ­ ate any­where in the vas­cu­la­ture and be
com­posed of many mate­ri­als (eg, thromboemboli, atheroem-
boli, fat emboli, sep­tic emboli). Certain imag­ing char­ac­ter­is­tics
(eg, dif­fer­ent ages of ische­mic changes in the same vas­cu­lar
ter­ri­tory) can sug­gest embolic eti­­ol­ogy. In the cere­bral vas­cu­
la­ture, the find­ing of mul­ti­ple infarcts in dif­fer­ent vas­cu­lar ter­
ri­to­ries has clas­si­cally been con­sid­ered an indi­ca­tor of embo-
lism, but a recent inves­ti­ga­tion revealed that the pres­ence of
this pat­tern is unable to con­sis­tently pre­dict, and its absence
is unable to con­sis­tently exclude, an embolic mech­a­nism.21 In
con­trast, a lacu­nar infarct (defined as a sin­gle, small [<15 mm in
diam­e­ter], sub­cor­ti­cal infarct in the dis­tri­bu­tion of a per­fo­rat­ing
artery)22 may sug­gest small-ves­sel dis­ease and can be help­ful to
exclude embolic causes. Also, infarcts in cer­tain ana­tomic sites
are more com­monly caused by embolic phe­nom­e­non (eg, renal
and splenic infarc­tions occur due to embolism in 68% and 43%,
respec­tively13). Again, a dis­cus­sion with an expert radi­ol­o­gist
and a lit­er­a­ture review may be help­ful.
Cardiac imag­ing with trans­tho­racic echo­car­dio­gram is often
the first step in the eval­u­a­tion for an embolic event, to iden­tify
increased atrial size, val­vu­lar dis­ease, and intra­car­diac mate­rial.
Initial eval­u­a­tion for pat­ent fora­men ovale (PFO), through which
venous mate­rial can enter arte­rial cir­cu­la­tion, is performed with
the use of agi­tated saline while the patient is coughing and/or
performing a Valsalva maneu­ver (a “bub­ble study”). Transesopha-
geal echo­car­di­og­ra­phy is the gold stan­dard for PFO mor­pho­logic
char­ac­ter­iza­tion and valve eval­u­a­tion and should be con­sid­ered
if ini­tial stud­ies are unreavealing.23 Imaging of surrounding arteries

78  |  Hematology 2021  |  ASH Education Program

Table 1.  Terminology for anticoagulation and antiplatelet ther­apy in arte­rial throm­bo­sis
Term
Combination ther­apy Dual antiplatelet ther­apy
Dual antithrombotic ther­apy
Dual path­way inhi­bi­tion
Triple antithrombotic ther­apy
Triple ther­apy
DOAC dos­ing Standard-dose DOAC
Reduced-dose DOAC*
Low dose DOAC
Prophylactic dose DOAC
Very low-dose DOAC
*Dosing reg­i­mens for patients with renal impair­ment, low body weight, and/or advanced age. See apixaban and rivaroxaban‡ details.
†
Dosing for patients with AF who meet 2 of the fol­low­ing 3 cri­te­ria: cre­at­i­nine ≥1.5 mg/dL, age ≥80 years, and weight ≤60 kg.
‡
Dosing used in patients with AF with cre­at­i­nine clear­ance ≤50 mL/min.
AF, atrial fibril­la­tion; CAD, cor­o­nary artery dis­ease; DAPT, dual antiplatelet ther­apy; BID, twice daily.
(ie, carotid arteries, aorta) is indi­cated if there is con­cern for an
atheroembolic eti­­ol­ogy. Ambulatory rhythm mon­i­tor­ing is also
recommended, although opti­mal dura­tion and pre­ferred devices
are debated. One group of experts favors at least 30 days in
ische­mic stroke, with con­sid­er­ation of implant­able loop record­
ers in high-risk patients.24 However, existing data have not yet
dem­on­strated that increased arrhyth­mia detec­tion via extended
mon­ i­
tor­
ing reduces risk of recur­ rent stroke, and the opti­ mal
treat­ment for device-detected brief runs of atrial fibril­la­tion is the
sub­ject of ongo­ing clin­i­cal tri­als (NCT01938248, NCT02618577).
Management: When car­diac throm­bo­em­bo­lism asso­ci­ated
with atrial fibril­la­tion is iden­ti­fied, man­age­ment is straight­for­
ward, and evi­ dence-based guide­ lines exist. The man­ age­ment
of patients with pre­sumed cardioembolism with­out iden­ti­fi­able
arrhyth­mia and with­out a documented intra­car­diac throm­bo­em­
bolic source has been explored in ische­mic stroke with the cre­
a­tion of the cat­e­gory of embolic stroke of unde­ter­mined source
(ESUS).25 It was hypoth­e­sized that ESUS was com­monly caused by
covert atrial fibril­la­tion, and there­fore, anticoagulation would be
supe­rior to antiplatelet ther­apy for sec­ond­ary pre­ven­tion. How-
ever, 2 large ran­dom­ized tri­als in patients with ESUS revealed that
the use of anticoagulation (rivaroxaban 15 mg daily, dabigatran
150 mg or 100 mg twice daily) com­pared with aspi­rin 100 mg daily
did not decrease recur­rence yet was asso­ci­ated with increased
bleed­ing. Therefore, antiplatelet agents remain the pre­ferred anti-
thrombotic agent in patients with ESUS.26 Consequently, out­side
of clinic tri­als, there is no cur­rent clin­i­cal use­ful­ness of defin­ing a
stroke as ESUS. The appli­ca­bil­ity of these find­ings to other sites of
unex­plained arte­rial throm­bo­sis is unknown, and the deci­sion to
use antiplatelet ther­apy vs anticoagulation is often empiric.
Management of left ven­tric­u­lar throm­bus is an area of con­
tro­versy and care evo­lu­tion, where despite existing guide­lines
Prakash Singh Shekhawat
Abbreviation Meaning
DAPT Aspirin 81 mg + P2Y12 inhib­i­tor
DAT Anticoagulation + one
antiplatelet agent (com­monly
in patients with AF and CAD)
DPI Very low-dose anticoagulation +
antiplatelet agent(s)
TAT Anticoagulation + 2 antiplatelet
TT agents
— Apixaban 5 mg BID
Dabigatran 150 mg BID
Dabigatran 220 mg daily
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/76/1851903/76may.pdf by guest on 13 December 2021
Edoxaban 60 mg daily
Rivaroxaban 20 mg daily
— Apixaban 2.5 mg BID†
Rivaroxaban 15 mg daily‡
— Apixaban 2.5 mg BID
Rivaroxaban 10 mg daily
— Rivaroxaban 2.5 mg BID
†
recommending war­fa­rin,27-29 off-label use of direct oral anti­co­ag­
u­lants (DOACs) is com­mon. A recent ret­ro­spec­tive cohort study
(RED VELVT, Retrospective Evaluation of DOACs and Vascular
Endpoints of Left Ventricular Thrombi) intro­duced fur­ther uncer­
tainty, as patients tak­ing DOACs had an increased risk of stroke
or sys­temic embolism com­pared with war­fa­rin.30 There are many
lim­i­ta­tions to these data, includ­ing ret­ro­spec­tive and obser­va­
tional design, unknown DOAC dos­ing reg­i­mens, and fre­quent
cross­over, and a sub­se­quent small, unblinded ran­dom­ized trial
(Comparative Study of Oral Anticoagulation in Left Ventricular
Thrombi, No-LVT) dem­on­strated rivaroxaban as being noninferior
to war­fa­rin.31 Pending addi­tional data, we favor war­fa­rin for left
ven­tric­u­lar throm­bus but con­sider patient-spe­cific fac­tors and
have a detailed dis­cus­sion of data lim­i­ta­tions with all­patients.
The role of PFO clo­sure in the pre­ven­tion of recur­rent embolic
events is also debated and has been exten­sively reviewed else­
where.23 PFO is pres­ent in 25% of the adult pop­u­la­tion,32 so the
chal­lenge for cli­ni­cians is to deter­mine when a PFO is coin­ci­
den­tal, there­fore not warranting clo­sure, vs caus­a­tive, there­
fore warranting clo­sure. Updated guide­lines from the Amer­i­can
Academy of Neurology out­line fac­tors to con­sider when deter­
min­ing the value of PFO clo­sure.22 An algo­rithm based on these
guide­lines is presented in Figure 3. Importantly, no sin­gle fac­tor
is suf­fi­cient to jus­tify inter­ven­tion, so a com­pre­hen­sive, patient-
spe­cific eval­u­a­tion in close col­lab­o­ra­tion with neu­rol­ogy and
car­di­­ol­ogy is essen­tial, with sig­nif­i­cant con­sid­er­ation of patient
pref­er­ence.
Medications and sub­stances
After exten­sive eval­u­a­tion for ath­ero­scle­ro­sis and embolism, we
then performed a struc­tured inves­ti­ga­tion of less com­mon causes
(Figure 1). Certain med­i­ca­tions are known to increase throm­botic
Unexplained arte­rial throm­bo­sis  | 79
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/76/1851903/76may.pdf by guest on 13 December 2021
Figure 3.  Algorithm for consideration of PFO closure in secondary stroke prevention based on the 2020 Practice Advisory Update
Summary from the American Academy of Neurology.22 *Shunt descriptor refers to degree of right-to-left shunting, not anatomic
size. ^Includes hypertension, diabetes, hyperlipidemia, or smoking. °Single, small (<15 mm), subcortical, in the distribution of single
penetrating artery.

risk, although asso­ci­a­tions with venous events are more c ­ om­mon
than with arte­rial ones. Furthermore, risk is more con­sis­tently
dem­on­strated with car­dio­vas­cu­lar events and ische­mic stroke,
as throm­bo­ses in other sites are rare, and it is there­fore dif­fi­cult to
deter­mine cau­sal­ity. Example med­i­ca­tions and sub­stances, along
with their pro­posed path­o­phys­i­o­logic mech­a­nism, are presented
in Table 2.
Systemic dis­eases
Patients with malig­nancy are at increased risk for arte­rial throm­
bo­sis in the 5 months prior to diag­no­sis, with highest risk 1 month
prior.42 A thor­ough his­tory and phys­i­cal exam­i­na­tion for signs of
malig­nancy should be performed, with can­cer screen­ing based
on age (eg, colonoscopy, mam­mo­gram) and risk fac­tors (eg, low-
dose com­puted tomog­ra­phy of the chest in patients with tobacco
use). The ben­e­fit of more exten­sive imag­ing in patients with arte­
rial throm­bo­em­bo­lism has not been inves­ti­gated. However, given
the rar­ity of arte­rial events due to can­cer, supported by the lack
of ben­e­fit even in patients with venous throm­bo­em­bo­lism,43 it is
unlikely to be ben­e­fi­cial, and we there­fore do not rec­om­mend it.
Additional categories of sys­temic dis­or­ders to con­sider include
auto­im­mune dis­or­ders (eg, rheu­ma­toid arthri­tis, lupus, vas­cu­li­tis)44
and hema­to­logic dis­or­ders (eg, mye­lo­pro­lif­er­a­tive neo­plasms,
par­ox­ys­mal noc­tur­nal hemoglobulinuria [PNH], sickle cell dis­ease,
plasma cell dis­or­ders, cryoglobulinemia), with pro­posed eval­u­a­
tion outlined in Figure 1.

80  |  Hematology 2021  |  ASH Education Program

Table 2.  Medications asso­ci­ated with increased risk of arte­rial throm­bo­sis33-41

Medication Sites Proposed mech­a­nism(s) Notes

Estrogens33 Stroke, MI Increased plate­let acti­va­tion and
procoagulant fac­tors, decreased
anti­co­ag­u­lant and fibri­no­ly­sis
fac­tors
Androgenic ana­bolic ste­roids34 Stroke, MI Accelerated ath­ero­scle­ro­sis,
coag­u­la­tion abnor­mal­i­ties,
ele­vated hemat­o­crit
Heparin35 PA > stroke, MI Autoimmune response to hep­a­rin-
PF4 com­plex caus­ing hep­a­rin-
induced throm­bo­cy­to­pe­nia
Variable risk reported, likely low
abso­lute risk with com­monly
used lower estro­gen doses.
Higher risk in patients with HTN,
smok­ing
General term for per­for­mance-
enhanc­ing drugs with
promyogenic and andro­genic
effects
Well-documented arte­rial
throm­botic risk, but venous
throm­bo­sis more com­mon

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/76/1851903/76may.pdf by guest on 13 December 2021

Intravenous immu­no­glob­u­lin36 Stroke, MI Many pro­posed: hyper­vis­cos­ity, Risk fac­tors: older age, hyper­ten­sion,
plate­let acti­va­tion, vaso­spasm, hyper­cho­les­ter­­ol­emia. Consider
fac­tor XI con­tent lower dose, slower infu­sion rate
Cocaine37 Stroke, MI, PA Platelet acti­va­tion, Contributes to acute throm­bo­sis
vaso­con­stric­tion, accel­er­ated risk and chronic arte­rial
ath­ero­scle­ro­sis remodeling
Tobacco, can­na­bis38 PA Thromboangiitis obliterans Inflammatory vas­cu­lar dis­ease of
(Buerger dis­ease) the small- and medium-sized
ves­sels of the extrem­i­ties
Erythropoiesis-stim­u­lat­ing Stroke, MI Platelet acti­va­tion, hyper­vis­cos­ity, Risk most clearly documented
agents39 hyper­ten­sion, increased in patients with renal dis­ease
procoagulant fac­tors treated to higher hemo­glo­bin
tar­gets
Anticancer ther­a­pies40
  VEGF inhib­i­tors* Stroke, MI Endothelial dys­func­tion, increased Also asso­ci­ated with hem­or­rhage
procoagulant fac­tors,
accel­er­ated ath­ero­scle­ro­sis,
HTN
 Tamoxifen Stroke Unknown Initial tri­als suggested increased
stroke risk but decreased car­diac
mor­tal­ity, but var­i­able in later
tri­als
 Fluorouracil MI Vasospasm, endo­the­lial dam­age, Cardiotoxicity inci­dence 4% to 19%,
increased procoagulant fac­tors but role of throm­bo­sis vs other
mech­a­nisms unclear
  Immune check­point Stroke, MI, PA Increased procoagulant fac­tors, Venous throm­bo­sis more com­mon
inhib­i­tors41 plate­lets acti­va­tion, impaired than arte­rial (reported 13% vs 2%)
fibri­no­ly­sis, accel­er­ated
ath­ero­scle­ro­sis
*Includes bevacizumab, tyro­sine kinase inhib­i­tors (eg, sorafenib), aflibercept.
HTN, hyper­ten­sion; MI, myo­car­dial infarc­tion; PA, periph­eral artery; PF4, plate­let fac­tor 4; VEGF, vas­cu­lar endo­the­lial growth fac­tor.

Emerging data indi­ cate infec­ tion with severe acute respi­
ra­tory syn­drome coronavirus 2 is asso­ci­ated with an increased
throm­botic risk, with reported occur­rences in mul­ti­ple arte­rial
sites.45 The COVID-19 vac­ cines manufactured by AstraZeneca
and Johnson & Johnson have also been shown to cause a pro-
thrombotic syn­drome that mim­ics hep­a­rin-induced throm­bo­cy­
to­pe­nia, named throm­bo­sis with throm­bo­cy­to­pe­nia syn­drome
(TTS) or vac­cine-induced immune throm­botic throm­bo­cy­to­pe­nia
(VITT). TTS/VITT pres­ents with throm­bo­cy­to­pe­nia and pri­mar­ily
cen­tral venous sinus throm­bo­sis but has also been reported to
cause arte­rial throm­bo­sis.46 A care­ful his­tory of poten­tial COVID-
19 expo­sures, vac­ci­na­tion tim­ing and man­u­fac­turer, COVID-19
test­ing, and plate­let count (if con­cern for TTS/VITT) should be
con­sid­ered in all­patients dur­ing the pan­demic.
Prakash Singh
Vascular and ana­tomic dis­or­ders
Many vas­cu­lar abnor­mal­i­ties with poten­tial to pre­cip­i­tate throm­
bus for­ma­tion have been reviewed in detail.47 Such dis­or­ders may
affect the artery wall (eg, dis­sec­tion, aneu­rysm, fibromuscular
dys­pla­sia, vas­cu­li­tis) and/or cal­i­ber (eg, exter­nal com­pres­sion,
vaso­spasm). Therefore, existing imag­ ing should be reviewed
closely for ves­sel abnor­mal­i­ties. Additional imag­ing with CTA,
mag­netic res­o­nance angi­og­ra­phy, or cath­e­ter-directed angio­
gram may be required to bet­ter visu­al­ize areas of con­cern, with
unique con­sid­er­ations for each vas­cu­lar bed.48
Thrombophilias
The asso­ci­a­tion between inherited throm­botic dis­or­ders and
arte­rial throm­bo­sis is not well char­ac­ter­ized. We have pre­vi­ously
Shekhawat
Unexplained arte­rial throm­bo­sis  | 81
­reviewed the existing lit­er­a­ture,49 which is lim­ited to case series
and ret­ro­spec­tive ana­ly­ses. Given the rar­ity of the stron­ger inherit-
ed thrombophilias and their ques­tion­able role in arte­rial throm­bo­
sis for­ma­tion, we only con­sider test­ing after exten­sive eval­u­a­tion
for other causes in youn­ger patients (youn­ger than 50-55 years),
par­tic­u­larly if there is a fam­ily his­tory of venous or arte­rial throm­
bo­sis. If test­ing is pur­sued, we con­sider eval­u­a­tion for pro­tein C
defi­ciency, pro­tein S defi­ciency, anti­throm­bin defi­ciency, fac­tor V
Leiden muta­tions, and pro­throm­bin 20210 muta­tions. Testing for
fac­tor V Leiden and pro­throm­bin 20210 is performed to iden­tify
homo­zy­gous or com­pound het­ero­zy­gous patients only, as iso­
lated het­ero­zy­gos­ity does not have a clin­i­cally sig­nif­i­cant asso­
ci­
a­tion with arte­ rial throm­botic risk. Importantly, some assays
are inac­cu­rate dur­ing an acute throm­botic epi­sode and while on
anticoagulation, so it is essen­tial to under­stand assay lim­i­ta­tions
to ensure appro­pri­ate tim­ing of col­lec­tion and cor­rect inter­pre­ta­
tion. Antithrombotic agent selec­tion in patients with an iden­ti­fied
inherited thrombophilia is again with­out data for guid­ance. We
con­sider the use of anticoagulation, with or with­out aspi­rin 81 mg
daily, in patients with low bleed­ing risk but with reg­u­lar eval­ua ­­
tion of risk/ben­efi ­ t and con­sid­er­ation of newly published data.
Antiphospholipid syn­ drome (APS) is an acquired thrombo-
philia with a well-documented risk of arte­rial throm­bo­sis, so we
con­sider test­ing based on the updated Sapporo (also known as
Sydney) cri­te­ria5 in all­ patients with­out an iden­ti­fi­able eti­­ol­ogy.
Patients with arte­rial throm­bo­sis and antiphospholipid antibodies
(APLAs) who do not ful­fill Sydney cri­te­ria for APS (eg, low titer
APLA, pos­i­tive assay on only 1 occa­sion) should be man­aged in
the same man­ner as APLA-neg­at­ ive patients with a sim­i­lar throm­
botic event.50 The opti­mal antithrombotic strat­egy for patients
with con­firmed APS is debated; anticoagulation, antiplatelet ther­
apy, or a com­bi­na­tion of both can be con­sid­ered.50 We favor com­
bi­na­tion ther­apy in patients with a low bleed­ing risk based on a
small ran­dom­ized trial dem­on­strat­ing lower inci­dence of stroke
recur­rence and sim­i­lar bleed­ing inci­dence with war­fa­rin (tar­get
inter­na­tional nor­mal­ized ratio [INR] 2-3) plus aspi­rin 100 mg daily
vs aspi­rin alone.51 When anticoagulation is used, war­fa­rin is pre­
ferred, given 2 recent tri­als dem­on­strat­ing an increased risk of
arte­rial throm­bo­sis with rivaroxaban vs war­fa­rin.52,53 However,
debate remains if war­fa­rin is required for all­patients with APS
vs only those with high throm­bo­sis risk fea­tures (tri­ple pos­i­tive),
given the 2 afore­men­tioned tri­als pri­mar­ily included tri­ple-pos­i­tive
patients and had impor­tant methodologic lim­i­ta­tions.54 Although
we gen­er­ally favor war­fa­rin for anticoagulation in patients with
APS, we con­sider DOACs in patients who are sin­gle or dou­ble
pos­i­tive and who (1) have tol­er­ated a DOAC for more than a year,
(2) strug­gle with the lim­i­ta­tions of war­fa­rin (eg, fluc­tu­a­tion in INR,
inabil­ity to com­ply with reg­u­lar mon­i­tor­ing), or (3) have an invalid
INR due to the pres­ence of a lupus anti­co­ag­ul­ant. However, given
the lim­i­ta­tions of existing data, an informed dis­cus­sion with the
patient and the incor­po­ra­tion of patient pref­er­ence are impor­tant
com­po­nents of anticoagulation selec­tion.
CLINICAL CASE (Con­t in­u ed)
Additional eval­u­a­tion was performed based on con­sid­er­ation of
the 6 poten­tial causes of unex­plained arte­rial throm­bo­sis out-
lined in Figure 1.
82  |  Hematology 2021  |  ASH Education Program
1.  Atherosclerosis: The patient con­ firmed no diag­ no­sis of
hyper­ten­sion or dia­be­tes, and she did not use tobacco prod­
ucts. She was not obese and had no known fam­ily his­tory of
pre­ma­ture stroke or car­dio­vas­cu­lar dis­ease. Intracranial vas­cu­lar
imag­ing was reviewed with radi­ol­ogy and did not dem­on­strate
find­ings concerning for ath­ero­scle­ro­sis. Bilateral carotid ultra­so­
nog­ra­phy was performed and there was no sig­nif­i­cant ste­no­sis.
Lipid panel, hemo­glo­bin A1c, and lipo­pro­tein(a) were sent and
with­out abnor­mal­ity.
2.  Embolism: Telemetry dur­ing hos­pi­tal­i­za­tion did not dem­
on­strate arrhyth­mia, so arrange­ments were made at dis­charge
for 30-day ambu­la­tory car­diac mon­i­tor­ing. Trans­tho­racic echo­
car­dio­gram with nor­mal saline bub­ble study was with­out val­vu­
lar abnor­mal­ity, intra­car­diac mate­rial, and increased atrial size
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/76/1851903/76may.pdf by guest on 13 December 2021
but did dem­ on­
strate an intra­ car­diac shunt. Transesophageal
echo­car­di­og­ra­phy revealed a PFO with a mild bidi­rec­tional
shunt. Upper and lower extrem­ity ultra­sounds were neg­a­tive for
deep vein throm­bo­sis. Per the algo­rithm in Figure 3, con­tin­ued
eval­u­a­tion for alter­na­tive eti­­ol­o­gies was pur­sued before deter­
min­ing the role of PFO clo­sure.
3.  Medications and sub­stances: The patient was tak­ing no
med­i­ca­tions. She had a pro­ges­ter­one intra­uter­ine device for
con­ tra­cep­tion. She denied use of nonprescribed sub­ stances,
and a urine drug screen was neg­at­ ive.
4.  Vascular and ana­tomic: CTA of the head and neck was
obtained and did not reveal evi­dence of ste­no­sis or com­pres­
sion, dis­sec­tion, or ves­sel wall thick­en­ing.
5.  Systemic dis­or­ders: The patient denied recent weight loss,
fever, rash, joint pain, or sys­temic symp­toms. Physical exam­i­na­tion
was with­out adenopathy, organomegaly, or breast masses. She
had a nor­mal Pap smear 1 year prior and denied any fam­ily his­tory
of malig­nancy. A com­plete blood count was with­out abnor­mal­ity,
and there was no evi­dence of hemo­ly­sis. JAK2V617F muta­tion test­
ing and flow for PNH were con­sid­ered but not sent pend­ing addi­
tional eval­u­a­tion given low clin­i­cal sus­pi­cion. In the absence of
sys­temic symp­toms concerning for auto­im­mune dis­or­ders, sero­
logic test­ing was also deferred. She had received the COVID-19
vac­cine manufactured by Pfizer-BioNTech 2 months prior.
6.  Thrombophilias: The patient denied a fam­ ily his­
tory of
venous and arte­rial throm­bo­sis in first- or sec­ond-degree rel­a­
tives. She had had 2 preg­nan­cies with­out com­pli­ca­tions and no
mis­car­riages. In the absence of a fam­ily or per­sonal his­tory, the
deci­sion was first made to test for APS, an acquired dis­or­der, prior
to con­sid­er­ation of test­ing for inherited dis­or­ders. Lupus anti­co­ag­
u­lant returned clearly pos­it­ ive, IgG anticardiolipin was ele­vated at
85 microgram of IgG antibody (GPL) (nor­mal, 0-14 GPL), and IgG
anti-β2 gly­co­pro­tein I was ele­vated at 95 GPL (nor­mal, 0-20 GPL).
In sum­mary, the diag­nos­tic workup iden­ti­fied a PFO and clearly
pos­i­tive APLA (“tri­ple pos­i­tive”), likely representing APS pend­ing
repeat test­ing at 12 weeks. After a mul­ti­dis­ci­plin­ary dis­cus­sion
between hema­tol­ogy, car­di­­ol­ogy, and neu­rol­ogy and dis­cus­sion
with the patient, the deci­sion was made to ini­ti­ate anticoagula-
tion for sec­ond­ary stroke pre­ven­tion for likely APS. Warfarin was
recommended given con­cern for increased arte­rial throm­botic
risk with rivaroxaban. The patient had no sig­nif­i­cant bleed­ing risk
fac­tors and was very concerned about the poten­tial for repeat
stroke, so the joint deci­sion was made to add aspi­rin 81 mg daily
with close fol­low-up for reg­u­lar reassessment of bleed­ing risk.
PFO clo­sure was not recommended given the small degree of
right-to-left shunting observed and the fact that the patient would
be con­tin­ued on indef­i­nite anticoagulation for APS regard­less of
plan for PFO clo­sure.
Conclusion
The diag­no­sis and man­age­ment of unex­plained arte­rial throm­
bo­sis are a com­plex, mul­ti­dis­ci­plin­ary effort in which hema­tol­ogy
plays an inte­gral role. A sys­tem­atic approach to eval­u­a­tion is war-
ranted and includes con­sid­er­ation of com­mon eti­­ol­o­gies (ath­ero­
scle­ro­sis and throm­bo­em­bo­lism) as well as uncom­mon eti­­ol­o­gies
(med­i­ca­tions and sub­stances, vas­cu­lar and ana­tomic abnor­mal­
i­ties, sys­temic dis­or­ders, and thrombophilias). Management re-
quires review of avail­­able lit­er­a­ture for spe­cific vas­cu­lar ter­ri­to­ries
or con­di­tions and relies heavily on patient-spe­cific throm­botic
and bleed­ing risk and patient pref­er­ence. Continued close fol­
low-up with reg­u­lar reassessment of the man­age­ment plan is also
essen­tial, as new data or new symp­toms may pres­ent over time,
and new bleed­ing and throm­botic risk fac­tors may develop.
Acknowledgment 
Figures were cre­ated with BioRender​­.com.
Conflict-of-inter­est dis­clo­sure
Jori E. May: no com­pet­ing finan­cial inter­ests to declare.
Stephan Moll: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Jori E. May: off-label use of DOACs for left ventricular thrombus
is discussed.
Stephan Moll: off-label use of DOACs for left ventricular thrombus
is discussed.
Correspondence
Jori E. May, Department of Medicine, Division of Hematology/On-
cology, University of Alabama at Birmingham, 1720 2nd Ave South,
NP 2503, Birmingham, AL 35294; e-mail: jemay@uabmc​­.edu.
References
1. Wendelboe AM, Raskob GE. Global bur­den of throm­bo­sis: epi­de­mi­o­logic
aspects. Circ Res. 2016;118(9):1340-1347.
2. Wilson DP, Jacobson TA, Jones PH, et  al. Use of lipo­pro­tein(a) in clin­i­cal
prac­tice: a bio­marker whose time has come. A sci­en­tific state­ment from
the National Lipid Association. J Clin Lipidol. 2019;13(3):374-392.
3. Roane DW, Griger DR. An approach to diag­no­sis and ini­tial man­age­ment of
sys­temic vas­cu­li­tis. Am Fam Physician. 1999;60(5):1421-1430. PMID: 10524486.
4. Stinton LM, Fritzler MJ. A clin­i­cal approach to auto­an­ti­body test­ing in sys­
temic auto­im­mune rheu­matic dis­or­ders. Autoimmun Rev. 2007;7(1):77-84.
5. Miyakis S, Lockshin MD, Atsumi T, et al. International con­sen­sus state­ment
on an update of the clas­ si­
fi­
ca­
tion cri­
te­
ria for def­i­
nite antiphospholipid
syn­drome (APS). J Thromb Haemost. 2006;4(2):295-306.
6. Saver JL. Cryptogenic stroke. N Engl J Med. 2016;374(21):2065-2074.
7. Gulati R, Behfar A, Narula J, et  al. Acute myo­car­dial infarc­tion in young
indi­vid­u­als. Mayo Clin Proc. 2020;95(1):136-156.
8. Weinberg I, Jaff MR. Nonatherosclerotic arte­rial dis­or­ders of the lower
extrem­i­ties. Circulation. 2012;126(2):213-222.
9. Mart D, Shatzel J, DeLoughery T. Cryptogenic acute limb ische­mia: a ret­
ro­spec­tive cohort study defin­ing a pre­vi­ously undescribed clin­i­cal entity.
J Thromb Thrombolysis. 2018;45(3):397-402.
10. Hayreh SS. Acute ret­i­nal arte­rial occlu­sive dis­or­ders. Prog Retin Eye Res.
2011;30(5):359-394.
11. Massussi M, Scotti A, Lip GYH, Proietti R. Left ven­tric­u­lar throm­bo­sis: new
per­spec­tives on an old prob­lem. Eur Heart J Cardiovasc Pharmacother.
2021;7(2):158-167.
Prakash Singh Shekhawat
12. Brett AS, Azizzadeh N, Miller EM, Collins RJ, Seegars MB, Marcus MA.
Assessment of clin­i­cal con­di­tions asso­ci­ated with splenic infarc­tion in
adult patients. JAMA Intern Med. 2020;180(8):1125-1128.
13. Weber E, Grangeon F, Reynaud Q , et  al. Acute renal and splenic infarc­
tions: a review. QJM. 2020;113(3):186-193.
14. Brandt LJ, Boley SJ. AGA tech­ni­cal review on intes­ti­nal ische­mia. Gastroen-
terology. 2000;118(5):954-968.
15. Klok FA, Huisman MV. How I assess and man­age the risk of bleed­ing in
patients treated for venous throm­bo­em­bo­lism. Blood. 2020;135(10):724–734.
16. Ambrose JA, Tannenbaum MA, Alexopoulos D, et al. Angiographic pro­gres­
sion of cor­o­nary artery dis­ease and the devel­op­ment of myo­car­dial infarc­
tion. J Am Coll Cardiol. 1988;12(1):56-62.
17. Coutinho JM, Derkatch S, Potvin AR, et  al. Nonstenotic carotid plaque
on CT angi­ og­ ra­
phy in patients with cryp­ to­
genic stroke. Neurology.
2016;87(7):665-672.
18. Eikelboom JW, Connolly SJ, Bosch J, et al; COMPASS Investigators. Rivarox-
aban with or with­out aspi­rin in sta­ble car­dio­vas­cu­lar dis­ease. N Engl J Med.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/76/1851903/76may.pdf by guest on 13 December 2021
2017;377(14):1319-1330.
19. Bonaca MP, Bauersachs RM, Anand SS, et al. Rivaroxaban in periph­eral artery
dis­ease after revas­cu­lar­i­za­tion. N Engl J Med. 2020;382(21):1994–2004​.
20. Capodanno D, Bhatt DL, Eikelboom JW, et al. Dual-path­way inhi­bi­tion for
sec­ond­ary and ter­tiary antithrombotic pre­ven­tion in car­dio­vas­cu­lar dis­
ease. Nat Rev Cardiol. 2020;17(4):242-257.
21. Powers WJ, Kam CH, Ritter VS, Fine JP. Diagnostic accu­ racy of acute
infarcts in mul­ti­ple cere­bral cir­cu­la­tions for cardioembolic stroke: lit­er­a­ture
review and meta-anal­y­sis. J Stroke Cerebrovasc Dis. 2020;29(7):104849.
22. Messé SR, Gronseth GS, Kent DM, et  al. Practice advi­sory update sum­
mary: pat­ent fora­men ovale and sec­ond­ary stroke pre­ven­tion: report of
the Guideline Subcommittee of the Amer­i­can Academy of Neurology. Neu-
rology. 2020;94(20):876-885.
23. Collado FMS, Poulin M-F, Mur­phy J-J, Jneid H, Kavinsky C-J. Patent fora­men
ovale clo­sure for stroke pre­ven­tion and other dis­or­ders. J Am Heart Assoc.
2018;7(12).
24. Albers GW, Bernstein RA, Brachmann J, et  al. Heart rhythm mon­i­tor­ing
strat­e­gies for cryp­to­genic stroke: 2015 diag­nos­tics and mon­i­tor­ing stroke
focus group report. J Am Heart Assoc. 2016;5(3):e002944.
25. Hart RG, Diener H-C, Coutts S-B, et al; Cryptogenic stroke/ESUS Interna-
tional Working Group. Embolic strokes of unde­ter­mined source: the case
for a new clin­i­cal con­struct. Lancet Neurol. 2014;13(4):429-438.
26. Ntaios G. Embolic stroke of unde­ter­mined source: JACC review topic of the
week. J Am Coll Cardiol. 2020;75(3):333-340.
27. Guyatt GH, Akl EA, Crowther M, Gutterman DD, Schuünemann HJ. Anti-
thrombotic ther­apy and pre­ven­tion of throm­bo­sis, 9th ed: Amer­i­can
College of Chest Physicians Evidence-Based Clinical Practice Guidelines.
Chest. 2012;141(2):7S–47S.
28. Kernan WN, Ovbiagele B, Black HR, et  al. Guidelines for the pre­ven­tion
of stroke in patients with stroke and tran­sient ische­mic attack: a guide­­­
line for healthcare pro­fes­sion­als from the Amer­i­can Heart Association/
Amer­i­can Stroke Association. Stroke. 2014;45(7):2160-2236.
29. O’Gara PT, Kushner FG, Ascheim DD, et al. 2013 ACCF/AHA guide­line for the
man­age­ment of ST-ele­va­tion myo­car­dial infarc­tion: a report of the Amer­
i­can College of Cardiology Foundation/Amer­i­can Heart Association task
force on prac­tice guide­lines. Circulation. 2013;127(4):e362–e425.
30. Robinson AA, Trankle CR, Eubanks G, et al. Off-label use of direct oral anti­
co­ag­u­lants com­pared with war­fa­rin for left ven­tric­u­lar thrombi. JAMA Car-
diol. 2020;5(6):685.
31. Abdelnabi M, Saleh Y, Fareed A, et  al. Comparative study of oral anti-
coagulation in left ven­tric­u­lar thrombi (no-LVT trial). J Am Coll Cardiol.
2021;77(12):1590-1592.
32. Hagen P, Scholz D, Edwards W. Incidence and size of pat­ent fora­men ovale
dur­ing the first 10 decades of life: an autopsy study of 965 nor­mal hearts.
Mayo Clin Proc. 1984;59(1):17-20.
33. Petitti DB. Hormonal con­tra­cep­tives and arte­rial throm­bo­sis—not risk-free
but safe enough. N Engl J Med. 2012;366(24):2316-2318.
34. Pope HG, Wood RI, Rogol A, Nyberg F, Bowers L, Bhasin S. Adverse health
con­se­quences of per­for­mance-enhanc­ing drugs: an Endocrine Society sci­
en­tific state­ment. Endocr Rev. 2014;35(3):341-375.
35. Warkentin TE. Heparin-induced throm­bo­cy­to­pe­nia: path­o­gen­e­sis and man­
age­ment. Br J Haematol. 2003;121(4):535-555.
36. Marie I, Maurey G, Hervé F, Hellot M, Levesque H. Intravenous immu­no­
glob­u­lin-asso­ci­ated arte­rial and venous throm­bo­sis; report of a series and
review of the lit­er­a­ture. Br J Dermatol. 2006;155(4):714-721.
Unexplained arte­rial throm­bo­sis  | 83
37. Bachi K, Mani V, Jeyachandran D, Fayad ZA, Goldstein RZ, Alia-Klein N. Vas-
cular dis­ease in cocaine addic­tion. Atherosclerosis. 2017;262:154-162.
38. Olin JW. Thromboangiitis obliterans: 110 years old and lit­tle prog­ress made.
J Am Heart Assoc. 2018;7(23):e011214.
39. Lippi G, Franchini M, Favaloro E. Thrombotic com­pli­ca­tions of eryth­ro­poi­e­sis-
stim­u­lat­ing agents. Semin Thromb Hemost. 2010;36(5):537-549.
40. Jiang D, Lee AI. Thrombotic risk from che­mo­ther­apy and other can­cer ther­
a­pies. Cancer Treat Res. 2019;179:87-101.
41. Moik F, Chan W-SE, Wiedemann S, et al. Incidence, risk fac­tors, and out­
comes of venous and arte­rial throm­bo­em­bo­lism in immune check­point
inhib­i­tor ther­apy. Blood. 2021;137(12):1669-1678.
42. Navi BB, Reiner AS, Kamel H, et al. Arterial throm­bo­em­bolic events pre­ced­
ing the diag­no­sis of can­cer in older per­sons. Blood. 2019;133(8):781-789.
43. Carrier M, Lazo-Langner A, Shivakumar S. Screening for occult can­cer in
unpro­voked venous throm­bo­em­bo­lism. J Vasc Surg. 2015;62(6):1679.
44. Scalera A, Emmi G, Squatrito D, et  al. Thrombosis in auto­im­mune dis­
eases: a role for immu­no­sup­pres­sive treat­ments? Semin Thromb Hemost.
2016;42(6):650-661.
45. Cheruiyot I, Kipkorir V, Ngure B, Misiani M, Munguti J, Ogeng’O J. Arterial
throm­bo­sis in coronavirus dis­ease 2019 patients: a rapid sys­tem­atic review.
Ann Vasc Surg. 2021;70:273-281.
46. Cines DB, Bussel JB. SARS-CoV-2 vac­ cine-induced immune throm­ botic
throm­bo­cy­to­pe­nia. N Engl J Med. 2021;384(23):2254-2256.
47. Moll S. Nonarteriosclerotic dis­or­ders of the arte­rial sys­tem. In: Kitchens C,
Konkle B, Kessler C, eds. Consultative Hemostasis and Thrombosis. 4th ed.
Elsevier; 2019:415-429.
84  |  Hematology 2021  |  ASH Education Program
48. Mur­phy DJ, Aghayev A, Steigner ML. Vascular CT and MRI: a prac­ti­cal guide
to imag­ing pro­to­cols. Insights Imaging. 2018;9(2):215-236.
49. May JE, Moll S. How I treat unex­ plained arte­ rial throm­
bo­sis. Blood.
2020;136(13):1487-1498.
50. Ruiz-Irastorza G, Cuadrado MJ, Ruiz-Arruza I, et al. Evidence-based rec­om­
men­da­tions for the pre­ven­tion and long-term man­age­ment of throm­bo­
sis in antiphospholipid anti­body-pos­i­tive patients: report of a task force
at the 13th International Congress on Antiphospholipid Antibodies. Lupus.
2011;20(2):206-218.
51. Okuma H, Kitagawa Y, Yasuda T, Tokuoka K, Takagi S. Comparison between
sin­gle antiplatelet ther­apy and com­bi­na­tion of antiplatelet and anticoag-
ulation ther­apy for sec­ond­ary pre­ven­tion in ische­mic stroke patients with
antiphospholipid syn­drome. Int J Med Sci. 2010;7(1):15-18.
52. Pengo V, Denas G, Zoppellaro G, et al. Rivaroxaban vs war­fa­rin in high-risk
patients with antiphospholipid syn­drome. Blood. 2018;132(13):1365-1371.
53. Ordi-Ros J, Sáez-Comet L, Pérez-Conesa M, et al. Rivaroxaban ver­sus vita­min
K antag­o­nist in antiphospholipid syn­drome. Ann Int Med. 2019;171(10):685–
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/76/1851903/76may.pdf by guest on 13 December 2021
694.
54. Moll S. Rivaroxaban ver­sus vita­min K antag­o­nists in antiphospholipid syn­
drome. Ann Int Med. 2020;172(7):510.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000235
 CLOTTING AND BLEEDING CONUNDRUMS

Cryptogenic oozers and bruisers

Kristi J. Smock1,2 and Karen A. Moser1,2

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/85/1851788/85smock.pdf by guest on 13 December 2021

Department of Pathology, University of Utah, Salt Lake City, UT, and 2ARUP Laboratories, Salt Lake City, UT
1

Bleeding disorders with normal, borderline, or nondiagnostic coagulation tests represent a diagnostic challenge. Disor-
ders of primary hemostasis can be further evaluated by additional platelet function testing modalities, platelet electron
microscopy, repeat von Willebrand disease testing, and specialized von Willebrand factor testing beyond the usual
initial panel. Secondary hemostasis is further evaluated by coagulation factor assays, and factor XIII assays are used to
diagnose disorders of fibrin clot stabilization. Fibrinolytic disorders are particularly difficult to diagnose with current
testing options. A significant number of patients remain unclassified after thorough testing; most unclassified patients
have a clinically mild bleeding phenotype, and many may have undiagnosed platelet function disorders. High-through-
put genetic testing using large gene panels for bleeding disorders may allow diagnosis of a larger number of these
patients in the future, but more study is needed. A logical laboratory workup in the context of the clinical setting and
with a high level of expertise regarding test interpretation and limitations facilitates a diagnosis for as many patients
as possible.

LEARNING OBJECTIVES
• Diagnose bleeding disorders that present with normal conventional coagulation tests
• Identify disorders of primary hemostasis that present diagnostic challenges and recommend specialized testing that
may be diagnostically informative
• Explain a multistep approach for evaluating bleeding disorders of secondary hemostasis, clot stabilization, and fibri-
nolysis

Introduction stabilized by factor XIII (FXIII). Fibrinolysis is the normal

Laboratory coagulation testing plays an important role
in the diagnosis and management of bleeding disorders.
However, patients with a suspected bleeding disorder who
have normal or nondiagnostic initial coagulation tests pres-
ent a significant challenge. Both inherited and acquired
abnormalities should be carefully considered, with close
attention to bleeding assessment tool results, medical his-
tory, family history, medications, liver and renal function,
and physical examination findings.1 In this review, we dis-
cuss how to best approach the laboratory workup beyond
the common tests, including limitations and interpretation
of laboratory testing. The article is organized by hemostasis
component: primary hemostasis, secondary hemostasis,
and fibrinolysis. Primary hemostasis refers to platelet plug
formation that results from platelet adhesion to damaged
vasculature via von Willebrand factor (VWF) and direct
platelet binding, secretion of granules, and platelet aggre-
gation, whereas secondary hemostasis refers to fibrin for-
mation via the coagulation cascade that is subsequently
breakdown of fibrin clot that occurs during healing.2
CASE 1
A 15-year-old girl was evaluated for easy bruising, frequent
nosebleeds, and heavy menstrual bleeding. There was a
vague family history of an uncharacterized mucocutane-
ous bleeding disorder. Her platelet count, blood smear
morphology, basic clotting times, and fibrinogen were
normal. Prior von Willebrand testing had been performed
and showed similar values for VWF antigen (VWF:Ag) and
ristocetin cofactor activity (VWF:RCo) that were either
normal or slightly below the reference interval, not defin-
itive for von Willebrand disease (VWD) and considered
possible “low VWF.” Prior VWF multimer analysis was nor-
mal. She was referred to our laboratory for a repeat von
Willebrand panel and platelet aggregation studies.

Prakash Singh Shekhawat

Workup of bleeding disorders | 85

CASE 2
A 12-year-old boy sought eval­u­a­tion from the hema­tol­ogy clinic
after hav­ing mul­ti­ple epi­sodes of intra­mus­cu­lar bleed­ing that
began after he joined his junior high school foot­ball team. He
reported a prior epi­sode of prolonged bleed­ing after a tooth
extrac­tion. Initial test­ing, includ­ing plate­let count and periph­
eral smear review, basic clot­ting times, and a von Willebrand
panel (VWF:Ag, VWF:RCo, and fac­tor VIII [FVIII] activ­ity), was
nor­mal. Due to the high clin­i­cal sus­pi­cion for a bleed­ing dis­
or­der, there were ques­tions about what addi­tional test­ing to
per­form.
Initial test­ing for a bleed­ing dis­or­der
Coagulation test­ing is most use­ful in patients with an abnor­mal
bleed­ing his­tory, and thus a higher like­li­hood of a bleed­ing dis­
or­der, and is less use­ful for screen­ing unse­lected patients (such
as in the pre­op­er­a­tive set­ting).3,4 Initial test­ing typ­i­cally includes
plate­let count and mor­phol­ogy review, pro­throm­bin time (PT),
acti­vated par­tial throm­bo­plas­tin time (aPTT), and fibrin­o­gen
activ­ity.3,5-7 The most com­mon fibrin­o­gen assay avail­­able in hos­
pi­tal lab­o­ra­to­ries is the Clauss fibrin­o­gen activ­ity assay, which
is a calibrated test based on the throm­bin time. The test will be
abnor­mal in hypofibrinogenemia and most cases of dysfibrino-
genemia. Additional workup of dysfibrinogenemia may include
fibrin­o­gen anti­gen, throm­bin time (TT), reptilase time, and ge-
netic test­ing. Testing for VWD and plate­let func­tion dis­or­ders is
some­times included in the ini­tial eval­u­a­tion and, if not, should
be the next step due to the fre­quency of these dis­or­ders and
inabil­ity of rou­tine tests to detect them.4,5 The aPTT is prolonged
in VWD if the FVIII is suf­fi­ciently low, but the aPTT is nor­mal in
most cases, as is the plate­let count.8 Platelet func­tion dis­or­ders
fre­quently have nor­mal plate­let count and mor­phol­ogy.
In a study designed to eval­ua ­ te the util­ity of com­mon coag­
u­la­tion screen­ing tests, results from 800 patients referred for
bleed­ing dis­or­der eval­u­a­tion were ana­lyzed, and only 11% had 1
or more abnor­mal­i­ties of PT, aPTT, fibrin­o­gen, or TT.3 The sen­si­
tiv­ity of this group of tests to a clin­i­cally sig­nif­i­cant abnor­mal­ity
was only 3.7%, but this increased to 8.5% when a basic VWD
panel was added and to 30% when both a VWD panel and plate­
let func­tion test­ing (by light trans­mis­sion aggregometry) were
added, suggesting that these should be included in the ini­
tial test­ing.3 Early plate­let aggregometry is recommended in a
recent guide­line from the International Society on Thrombosis
and Haemostasis.9
Challenges in diag­no­sis of dis­or­ders of pri­mary
hemo­sta­sis
Patients with dis­or­ders of pri­mary hemo­sta­sis dem­on­strate a
muco­cu­ta­ne­ous pat­tern of imme­di­ate bleed­ing (prolonged time
to achieve ini­tial hemo­sta­sis).2,10 Although the skin bleed­ing time
is an out­dated method for assessing pri­mary hemo­sta­sis, it has
been supplanted in many cen­ters by devices such as the Platelet
Function Analyzer (PFA)–100 (or the PFA-200 in some loca­tions
out­side of the United States) as an ini­tial screen of pri­mary hemo­
sta­sis, pri­mar­ily due to acces­si­bil­ity and ease of use.11 Although
the PFA-100 con­cep­tu­ally rep­re­sents a global test for eval­u­at­ing
pri­mary hemo­sta­sis, the pres­ence of high shear (which results in
a VWF con­for­ma­tional change that exposes the plate­let bind­
86  |  Hematology 2021  |  ASH Education Program
ing domain) and redun­dant plate­let ago­nists (includ­ing col­la­gen
that binds plate­let col­la­gen recep­tors and shear-acti­vated VWF)
tends to mask mild abnor­mal­i­ties, and a nor­mal clo­sure time re-
sult only excludes more severe VWD (such as types 2 and 3) or
severe plate­let dys­func­tion (such as Glanzmann thrombasthenia
and Bernard-Soulier syn­drome).12,13 The PFA-100 and other micro-
fluidics-based assays are ini­ti­ated by the adhe­sion func­tions of
VWF and plate­lets and are thus more sen­si­tive to dis­or­ders of
adhe­sion. PFA-100 sen­si­tiv­ity to VWD is esti­mated at approx­i­
ma­tely 60% to 70% over­all, with milder type 1 phe­no­types and
those with “low VWF” as a bleed­ing risk fac­tor more com­monly
missed.5,12 Although stor­age pool deficiencies and secre­tion de-
fects rep­re­sent the most com­mon her­i­ta­ble plate­let func­tion
abnor­mal­i­ties, stud­ies sug­gest that up to half of these cases are
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/85/1851788/85smock.pdf by guest on 13 December 2021
missed by PFA-100 test­ing.5,12 If there is strong sus­pi­cion for a
bleed­ing dis­or­der of pri­mary hemo­sta­sis, spe­cific VWD test­ing
pan­els and plate­let aggregometry should be used despite the
increased tech­ni­cal and inter­pre­tive com­plex­ity of aggregom-
etry.1,12 Collagen dis­ or­ders, such as Ehlers-Danlos syn­ drome,
should also be con­sid­ered and are eval­u­ated by joint mobil­ity
stud­ies and genetic test­ing in some cases. Because a sig­nif­i­cant
num­ber of patients with Ehlers-Danlos syn­drome who have an
ele­vated bleed­ing score have been shown to also have plate­let
dys­func­tion, plate­let func­tion test­ing is con­sid­ered use­ful in this
set­ting.14
Comprehensive inves­ti­ga­tion for VWD is beyond the scope of
this review, but test­ing should include VWF anti­gen and activ­ity
(such as VWF:RCo or newer assays), FVIII, and addi­tional spe­cial­
ized test­ing as needed to con­firm the diag­no­sis and sub­type.15
Moderate/severe VWD cases are detected most read­ily due to
their more pro­nounced bleed­ing and lab­o­ra­tory abnor­mal­i­ties.
Mild type 1 cases can be chal­leng­ing to diag­nose, some­times
presenting with a nor­mal panel due to the acute phase nature
of VWF.8,15,16 If there is high clin­i­cal sus­pi­cion for VWD (his­tory of
imme­di­ate muco­cu­ta­ne­ous bleed­ing), repeated mea­sure­ments
may be nec­es­sary to uncover the true base­line val­ues.8,15,17 Due to
estro­gen’s effects on VWF lev­els, women with heavy men­strual
bleed­ing are often tested at the onset of men­ses when estro­
gen is low­est, and VWD screen­ing should be avoided in women
on estro­gen ther­apy or dur­ing or imme­di­ately after preg­nancy.
A recent study of ini­tial and fol­low-up VWF test­ing in women
seen in an emer­gency depart­ment for heavy men­strual bleed­ing
showed higher median VWF and FVIII lev­els dur­ing acute bleed­
ing as opposed to fol­low-up, high­light­ing the lim­ited diag­nos­tic
util­ity dur­ing acute bleed­ing.15 The authors also noted that if the
ini­tial FVIII value was near the nor­mal range, the asso­ci­ated VWF
lev­els were closer to base­line, suggesting that because FVIII also
has acute phases, close atten­tion to the FVIII value may help
deter­mine if an acute phase response is pres­ent, but find­ings
should be interpreted with cau­tion.18 Another recent ret­ro­spec­
tive study of 811 pedi­at­ric patients with suspected VWD indi­
cated that 70% of these patients had von Willebrand (VW) panel
results diag­nos­tic of VWD in the first test­ing epi­sode. Patients
with VW panel val­ues between 30 and 50 IU/dL are likely most
at risk for repeat test­ing prior to diag­no­sis; higher VW panel val­
ues are more likely to exclude VWD (neg­at­ive pre­dic­tive value
for VWF anti­gen and activ­ity >75 IU/dL was 94.1% and 97.4%,
respec­tively).15,19
Other set­tings in which an ini­tial VWD panel may be unre-
vealing include rare sub­types of type 2M VWD due to col­la­gen
bind­ing defects because rou­tine pan­els do not typ­i­cally assess
this func­tion.20 A col­la­gen bind­ing assay could be performed
(VWF:CB) and would dem­on­strate decreased col­la­gen bind­ing
out of pro­por­tion to the VWF pro­tein level.5,21 A sub­set of type
2B cases can also appear nor­mal in basic VWD pan­els and have
nor­mal plate­let count and VWF multimers as opposed to the
clas­sic type 2B pro­file of throm­bo­cy­to­pe­nia, miss­ing large VWF
multimers, and decreased VWF activ­ity with a low activ­ity/anti­
gen ratio.22 Low-dose ristocetin-induced plate­let aggre­ga­tion
can uncover the diag­no­sis by dem­on­strat­ing the abnor­mally
high affin­ity of mutant VWF for plate­lets. Cases with nor­mal
multimers have been shown to have dif­fer­ent type 2B muta­
tions and milder bleed­ing as opposed to type 2B patients with
multimeric defects.3,22 Genetic test­ing for VWD is also avail­­able
and has proven most use­ful for type 2 sub­types with chal­leng­
ing lab­o­ra­tory phe­no­types or where phe­no­typic test­ing is not
avail­­able.20
Acquired von Willebrand syn­drome (AVWS), which is a het­
ero­ge­neous dis­or­der due to anti­body-medi­ated clear­ance,
adsorp­tion to plate­lets or neo­plas­tic cells, or pro­te­­ol­y­sis due to
increased shear stress, can be diag­nos­ti­cally chal­leng­ing. Spe-
cific clin­i­cal cir­cum­stances fre­quently asso­ci­ated with AVWS
include patients treated with extra­cor­po­real mem­brane oxy­gen­
a­tion, heart pump (such as Abiomed’s Impella), and left ven­tric­
u­lar assist devices.23,24 VWF val­ues are often in the nor­mal range
due to ade­quate VWF pro­duc­tion and acute phas­ing. Even if the
abso­lute val­ues are nor­mal, a decreased activ­ity/anti­gen ratio
(such as VWF:RCo/Ag or VWF:CB/Ag) is help­ful and may indi­
cate clear­ance of large multimers, which occurs in many, but not
all­, cases of AVWS.25 Subtle loss of large multimers by VWF elec­
tro­pho­re­sis may be the only abnor­mal­ity observed, and upfront
multimer anal­y­sis is rea­son­able if AVWS is suspected.25,26
Although plate­ let aggre­ga­tion is the pre­ ferred ini­ tial test
for assessing plate­let func­tion, it is not sen­si­tive to all­dis­or­
ders and suf­fers in prac­tice from nonstandardization despite
lit­er­a­ture supporting stan­dard­i­za­tion and har­mo­ni­za­tion.4,9,10,27,28
Platelet dense gran­ule defi­ciency is thought to be as com­mon
as VWD, may dem­on­strate nor­mal aggre­ga­tion results, and is
best assessed by dense gran­ ule counts using whole-mount
plate­let elec­tron micros­copy.4,5,13,29 Although elec­tron micros­
copy test­ing has lim­ited avail­abil­ity, it dem­on­strates supe­rior
per­for­mance char­ac­ter­is­tics com­pared with the more widely
avail­­able lumiaggregometry meth­ods that mea­sure aden­o­
sine tri­phos­phate release from plate­let dense gran­ules.4,13,29-31
However, aden­o­sine tri­phos­phate secre­tion anal­y­sis may iden­
tify secre­tion defects that are not due to gran­ule defi­ciency.32
Scott syn­drome (plate­let fac­tor 3 defi­ciency) is a rare reces­sive
bleed­ing dis­or­der caused by abnor­mal­i­ties in plate­let procoag-
ulant activ­ity due to lack of phosphatidylserine (PS) expo­sure
after plate­let acti­va­tion. Calcium-depen­dent coag­u­la­tion fac­
tors require bind­ing to neg­a­tively charged phos­pho­lip­ids, such
as PS, pro­mot­ing for­ma­tion of the tenase and prothrombinase
complexes of coag­u­la­tion and resulting in nor­mal throm­bin and
fibrin gen­er­a­tion. Scott syn­drome is not iden­ti­fied by rou­tine
coag­u­la­tion test­ing or plate­let aggre­ga­tion and requires either
flow cytometric test­ing of plate­let PS expo­sure or genetic test­
ing for ANO6 muta­tions.33 More than 30 inherited plate­let dis­
or­ders, due to abnor­mal­i­ties in more than 50 genes, have been
described, although not all­affect plate­let func­tion.7 Next-gen­
er­a­tion sequenc­ing has been used to inter­ro­gate patients with
Prakash Singh Shekhawat
plate­let func­tion dis­or­ders, high­light­ing genetic com­plex­ity, but
is not com­monly performed.32
Laboratory con­sid­er­ations for patients suspected to have a
bleed­ing dis­or­der of pri­mary hemo­sta­sis are shown in Table 1.
Challenges in diag­no­sis of dis­or­ders of sec­ond­ary
hemo­sta­sis
Patients with dis­or­ders of sec­ond­ary hemo­sta­sis dem­on­strate var­
i­able bleed­ing pat­terns that can include mus­cle and soft tis­sue
bleed­ing and delayed bleed­ing after achiev­ing ini­tial hemo­sta­
sis.4,6 Conventional clot­ting times such as PT and aPTT are pro-
longed by mod­er­ate or severe fac­tor deficiencies but fre­quently
lack sen­si­tiv­ity to milder deficiencies that may be clin­i­cally sig­nif­i­
cant with bleed­ing chal­lenges.4,34,35 Coagulation fac­tor assays can
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/85/1851788/85smock.pdf by guest on 13 December 2021
be ordered in a step­wise approach focus­ing on more com­mon
dis­or­ders first with assess­ment of fac­tors VIII, IX, and XI via clot-
based activ­ity assays. Because some cases of nonsevere hemo­
philia A (HA) dem­on­strate discrepancies between clot-based and
chro­mo­genic FVIII assays based on how spe­cific muta­tions affect
the FVIII pro­tein, it is use­ful to per­form both types of FVIII assays.36
It is esti­mated that 40% of mild HA cases have sig­nif­i­cant assay dis-
crepancies and that 5% to 10% of mild HA cases have nor­mal FVIII
activ­ity by either assay, which can mask the diag­no­sis.5,36 Clinical
phe­no­types have cor­re­lated most closely with the low­est results
observed. A sim­i­lar phe­nom­e­non has been seen in patients with
nonsevere hemophilia B (HB) when clot-based and chro­mo­genic
fac­tor IX activ­ity assays are com­pared, which sup­ports a sim­i­lar
approach.37 Chromogenic fac­tor IX assays are less widely avail­­able
than chro­mo­genic FVIII assays. Because fac­tor activ­i­ties may be
bor­der­line in mild hemo­philia, genetic test­ing (gene sequenc­ing)
to con­firm an HA or HB path­o­genic muta­tion is often help­ful to
diag­nose mild cases or female car­ri­ers.38,39
Additional assays for uncom­mon mild deficiencies of fac­tors
II, V, VII, and X could be con­sid­ered if the ini­tial fac­tor assays are
nor­mal. Alternatively, a large bat­tery of fac­tor assays under­taken
at once can save time and min­i­mize blood draws. Some lab­o­
ra­to­ries may offer reflex­ive bleed­ing dis­or­der pan­els to assist
with timely eval­u­a­tion. Acquired fac­tor deficiencies are usu­ally
mul­ti­ple (due to vita­min K defi­ciency, liver dis­ease, con­sump­tive
coagulopathies, etc.), which has a more pro­nounced effect on
the PT and aPTT, mak­ing them more likely to be iden­ti­fied in the
ini­tial test­ing. It should also be remem­bered that clot­ting time
pro­lon­ga­tions due to a lupus anti­co­ag­u­lant may mask or con­
found iden­ti­fi­ca­tion of a fac­tor defi­ciency or inhib­i­tor. Current
lab­o­ra­tory test­ing guide­lines for lupus anti­co­ag­u­lant iden­ti­fi­ca­
tion indi­cate that fac­tor assays should be performed if there is
sus­pi­cion for a fac­tor defi­ciency (such as a pos­i­tive bleed­ing his­
tory or clin­i­cal con­di­tion asso­ci­ated with fac­tor deficiencies).40
Table 2 dem­on­strates an exam­ple of aPTT reagent sen­si­tiv­
ity to pro­gres­sive sin­gle FVIII defi­ciency or pan-defi­ciency of
all­coag­u­la­tion fac­tors. The upper limit of nor­mal for this aPTT
reagent is 35 sec­onds, and the aPTT is more sen­si­tive to mul­ti­ple
deficiencies as opposed to a sin­gle defi­ciency. Mild HA with FVIII
activ­ity in the 20% to 40% range may dem­on­strate a nor­mal or
min­i­mally prolonged aPTT. Different aPTT reagents may dif­fer in
these pat­terns.
Challenges in diag­no­sis of FXIII defi­ciency
FXIII defi­ciency is a rare reces­sive bleed­ing dis­or­der char­ac­ter­
ized by severe bleed­ing with a delayed bleed­ing pat­tern and
Workup of bleed­ing dis­or­ders  |  87
Table 1. Laboratory con­ sid­er­
ations for patients with a suspected bleed­ ing dis­
or­
der of pri­
mary hemo­ sta­sis and nor­ mal
or nondiagnostic ini­tial tests (nor­mal plate­let count/mor­phol­ogy, rou­tine clot­ting times, VWD panel, and plate­let func­tion test­ing)

Problem Laboratory con­sid­er­ations

VWD
Normal vs low VWF vs VWD masked PFA-100 lacks sen­si­tiv­ity and spec­i­fic­ity for VWD; ini­tial panel should include VWF anti­gen, activ­ity, and
by acute phase response FVIII activ­ity.
Avoid test­ing dur­ing acute bleed­ing epi­sodes, ill­ness, and stress.
Use care to avoid trau­matic phle­bot­omy.
Repeated mea­sure­ments may be nec­es­sary due to nor­mal bio­logic var­i­abil­ity of VWF, par­tic­u­larly with
VWF anti­gen and activ­ity val­ues between 30 IU/dL and 50 IU/dL.
Consider val­ues of other acute phase reac­tants drawn con­cur­rently (FVIII, fibrin­o­gen).
Less com­mon VWD sub­types/ Consider type 2M due to a col­la­gen bind­ing defect. First-line activ­ity tests assess plate­let bind­ing
phe­no­types func­tion rather than col­la­gen bind­ing func­tion. Order VWF:CB to assess for decreased VWF:CB/anti­

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/85/1851788/85smock.pdf by guest on 13 December 2021

Acquired VWS
Platelet func­tion
Platelet func­tion tests are
nonstandardized and variably sen­si­
tive to the most com­mon dis­or­ders
Abbreviations: EM, elec­tron micros­copy; LD-RIPA, low-dose ristocetin-induced plate­let acti­va­tion; PFD, platelet function disorder.
gen ratio.
Consider type 2B due to a muta­tion that results in a nonclassic lab­o­ra­tory phe­no­type (rel­a­tively nor­
mal activ­ity and anti­gen with­out decreased activ­ity/anti­gen ratio, nor­mal multimers and plate­let
count). Perform LD-RIPA to assess for abnor­mally high affin­ity for plate­lets.
LD-RIPA mixing stud­ies (if avail­­able) can be used to dif­fer­en­ti­ate type 2B from plate­let-type VWD.
VWD genetic test­ing can be performed and is most use­ful for type 2 cases with chal­leng­ing lab­o­ra­tory
phe­no­types or when phe­no­typic test­ing is not avail­­able (LD-RIPA not pos­si­ble if local lab­o­ra­to­ries do
not per­form aggre­ga­tion test­ing and LD-RIPA mixing stud­ies are not widely performed). Also use­ful
to dif­fer­en­ti­ate type 2B from plate­let-type VWD and type 2N from HA.
Consideration of and appro­pri­ate test­ing for asso­ci­ated dis­or­ders such as auto­im­mune,
lymphoproliferative or mye­lo­pro­lif­er­a­tive neo­plasms, plasma cell dys­cra­sia, and so on.
Absolute VWF activ­ity and anti­gen val­ues often in the nor­mal range; look for decreased activ­ity/anti­
gen ratio.
VWF multimer anal­y­sis recommended as a first-line test since sub­tle defects of large multimers may be
the only abnor­mal­ity observed.
PFA-100 lacks sen­si­tiv­ity and spec­i­fic­ity for the most com­mon plate­let func­tion dis­or­ders. Platelet
aggregometry is con­sid­ered the gold-stan­dard test but has more lim­ited avail­abil­ity. Some lab­or­ a­to­
ries offer aggregometry with an extended ago­nist panel.
Aggregometry is often nor­mal in com­mon dis­or­ders such as dense gran­ule defi­ciency and secre­tion
defects. Consider plate­let EM and lumiaggregometry.
Mild PFDs remain dif­fi­cult to diag­nose; genetic test­ing has some clin­i­cal util­ity but is not in wide­spread
clin­i­cal use. Consider treating undi­ag­nosed dis­or­ders with tranexamic acid and desmopressin.

nor­mal stan­dard coag­u­la­tion tests. FXIII cross-links fibrin and
incor­po­rates the fibri­no­lytic inhib­i­tor α2-antiplasmin (A2AP)
to pre­vent pre­ma­ture deg­ra­da­tion.41 Qualitative clot sol­u­bil­ity
tests are fre­quently used for screen­ing but have sig­nif­i­cant lim­i­
ta­tions. Solubility tests are nonstandardized between lab­o­ra­to­
ries and sen­si­tive to only the most severe deficiencies (abnor­
mal with <1% activ­ity in our lab­o­ra­tory).41 They also dem­on­strate
nonspecificity because abnor­mal results occur with hypo- or
dysfibrinogenemia, although these con­di­tions would likely be
iden­ti­fied prior to FXIII test­ing by fibrin­o­gen test­ing.41 Activity
assays are pre­ferred because they detect both quan­ti­ta­tive and
qual­it­a­tive abnor­mal­it­ies and are more spe­cific, but they may
have sub­op­ti­mal lower lim­its of quan­ti­ta­tion (our assay can reli­
ably quan­tify down to 5% activ­ity) or use meth­ods that may
over­es­ti­mate FXIII activ­ity.41,42 Some lab­o­ra­to­ries use FXIII anti­
gen tests for ini­tial eval­u­at­ion. FXIII defi­ciency should be sub-
classified as either A or B sub­unit sub­types due to treat­ment
impli­ca­tions (a com­monly used recom­bi­nant ther­apy con­tains
only the A sub­units). Subtyping can be performed using anti­gen
assays or genetic test­ing. Ultimately, accu­rate diag­no­sis and
clas­si­fi­ca­tion of FXIII defi­ciency typ­i­cally requires a com­bi­na­tion
of tests to be used.42
Challenges in diag­no­sis of fibri­no­lytic dis­or­ders
Rare patients have inherited reces­sive defects in the fibri­no­lytic
inhib­i­tors A2AP and plas­min­o­gen acti­va­tor inhib­i­tor 1 (PAI-1),
resulting in a bleed­ing dis­or­der with delayed bleed­ing due to
increased lysis of fibrin clots.1,43 Routine clot­ting times are nor­
mal.1,44 Testing for A2AP and PAI-1 should be reserved until abnor­
mal­i­ties that are more com­mon have been excluded. Testing for
fibri­no­lytic dis­or­ders can be dif­fic
­ ult to inter­pret due to com­plex
inter­ac­tions between fibri­no­lytic acti­va­tors and inhib­i­tors and ef-
fects of preanalytical var­i­ables related to blood draw qual­ity and
tim­ing. Clinically avail­­able PAI-1 assays have tech­ni­cal lim­i­ta­tions
in the low range and may not be ­able to defin­i­tively iden­tify defi-
ciencies.44 PAI-1 is an acute phase reac­tant, which may mask mild
defi­ciency. PAI-1–defi­cient patients have reduced tis­sue plas­min­
o­gen acti­va­tor anti­gen due to faster clear­ance, and tis­sue plas­
min­o­gen acti­va­tor anti­gen should be tested con­cur­rently.
Dominantly inherited profibrinolytic plate­lets due to increased
uro­ki­nase-type plas­min­o­gen acti­va­tor in plate­let alpha gran­ules
result in a bleed­ing dis­or­der known as Que­bec plate­let dis­or­der.
Bleeding is due to intraplatelet plas­min gen­er­a­tion and pro­te­­
ol­y­sis of plate­let pro­teins and coag­u­la­tion fac­tors in the alpha
gran­ules.44 Coagulation and plate­let aggre­ga­tion test­ing is nor­

88  |  Hematology 2021  |  ASH Education Program

Table 2.  aPTT reagent sen­si­tiv­ity to pro­gres­sive sin­gle FVIII defi­ciency or pan-defi­ciency of all­fac­tors

FVIII activ­ity, % aPTT, s* All fac­tor activ­ity, % aPTT, s*

100 28.6 100 29.4
90 28.6 90 30.1
80 29.6 80 31.9
70 30.7 70 33.9**
60 31.5 60 37.4**
50 32.5 50 43.0
40 34.3** 40 51.9
30 36.1** 30 69.9

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/85/1851788/85smock.pdf by guest on 13 December 2021

20 39.1 20 108.0
15 40.5 15 No clot
10 44.5 10 No clot
5 49.6 5 No clot
1 61.1 1 No clot
0.001 79.6 0.001 No clot
*Reference inter­val for this aPTT reagent (24-35 sec­onds).
**The bold font indicates when the aPTT seconds begin to exceed the reference interval.

mal in this con­di­tion (with the excep­tion of absent epi­neph­rine
response), and genetic test­ing for PLAU gene dupli­ca­tion should
be performed for diag­no­sis.44
Because vis­co­elas­tic test­ing meth­ods have become more
com­mon­place, it is use­ful under­stand the expected pat­terns in
patients with fibri­no­lytic dis­or­ders. These assays are often insen­
si­tive to lysis caused by small changes in plas­min­o­gen acti­va­tor
lev­els but can show evi­dence of lysis when there are acti­va­tor
lev­els sig­nif­i­cant enough to over­come PAI-1, includ­ing severe
trauma, car­dio­pul­mo­nary bypass, and the anhepatic phase of
liver trans­plant. Viscoelastic test­ing meth­ods have been used to
opti­mize trans­fu­sion reg­i­mens or other ther­a­pies in these set­
tings, but there is also con­tro­versy around the degree of ben­e­fit
and around sen­si­tiv­ity and spec­i­fic­ity in other types of patients.45
Laboratory con­sid­er­ations for patients suspected to have a
bleed­ing dis­or­der of sec­ond­ary hemo­sta­sis, clot sta­bi­li­za­tion
(FXIII), or fibri­no­ly­sis are shown in Table 3.
Commentary on other test­ing modal­i­ties
Optimized lab­o­ra­tory test­ing approaches can max­i­mize the like­li­
hood of a defin­i­tive diag­no­sis in patients with a bleed­ing dis­or­der.
Although some patients are diag­nosed after thor­ough lab­o­ra­
tory inves­ti­ga­tion, oth­ers remain undi­ag­nosed after an exten­sive
search and fall into the cat­e­gory of unclas­si­fied bleed­ing dis­or­
ders (UBDs), also known as bleed­ing of unknown cause (BUC).16
Mild bleed­ing dis­or­ders are over­rep­re­sented in this group, and
mild plate­let func­tion dis­or­ders are underdiagnosed and remain
diag­nos­ti­cally chal­leng­ing.43 Testing such as vis­co­elas­tic stud­ies, ristocetin-induced plate­let aggre­ga­tion showed an abnor­mally
throm­bin gen­er­a­tion, or other cur­rently research-based test­ing
do not yet have a clear role in this set­ting due to var­i­able per­for­
mance in the lit­er­a­ture or lack of clin­i­cal avail­abil­ity.1,5
Genetic test­ing using large gene pan­els has the poten­tial to
iden­tify rare abnor­mal­i­ties not ade­quately assessed by clin­i­cally
avail­­able tests and may be use­ful for cer­tain patients, although
var­i­ant inter­pre­ta­tion may be dif­fi­cult.7,43,46,47 Genetic abnor­mal­ ing after mild to mod­er­ate trauma, addi­tional test­ing included
i­ties may cor­re­late poorly with thePrakashclin­i­cal bleed­ ing ten­
d ency
Singh Shekhawat
due to mul­ti­ple gene inter­ac­tions or inter­ac­tion with acquired
fac­tors.32 These pan­els may increase diag­no­sis and under­stand­
ing of rare dis­ or­
ders, such as fibri­
no­
lytic dis­
or­
ders, in which
cur­rently avail­­able test­ing is not widely avail­­able, com­plex to
per­form or inter­pret, or lim­ited by poor assay per­for­mance char­
ac­ter­is­tics.16,42,44 When large pan­els were stud­ied in the set­ting
of UBD/BUC, the over­all diag­nos­tic yield was low (3.2% for the
ThromboGenomics panel in a recent study of 2396 patients), and
more study of high-through­put genetic modal­i­ties is needed.43,46
Currently, the Amer­i­can Thrombosis and Hemostasis Network, a
net­work of 140 hemo­sta­sis cen­ters across all­50 states, is offer­
ing genetic test­ing (whole-gene sequenc­ing) for 30 rare clot­ting
and plate­let dis­or­der genes, includ­ing FXIII defi­ciency. This rep­
re­sents a true net­work of diag­nos­tic and ther­a­peu­tic options for
patients and fam­i­lies with rare and ultra-rare bleed­ing dis­or­ders,
and the test­ing is cur­rently pro­vided free of charge. Without a
spe­cific diag­no­sis to tai­lor treat­ment, UBC/BUC is often treated
with tranexamic acid and desmopressin with reported suc­cess.1
Case res­o­lu­tion and con­clu­sion
Case 1
In this female patient with a muco­cu­ta­ne­ous bleed­ing his­tory,
the repeat VWD panel results again showed bor­der­line low activ­
ity and anti­gen results with­out clear dis­cor­dance and a nor­mal
multimer pat­tern (VWF:Ag, 55%; VWF:RCo, 48%; activ­ity/anti­
gen ratio, 0.87). Her plate­lets responded normally to most plate­
let ago­nists in the aggre­ga­tion study, although the low-dose
increased response to low-dose ristocetin (0.5  mg/mL) with
com­plete plate­let aggre­ga­tion. Type 2B VWD was con­firmed by
iden­ti­fi­ca­tion of a muta­tion in VWF exon 28 (p.R1379C).22
Case 2
In this male patient with recent onset of pro­nounced bleed­
plate­let aggre­ga­tion stud­ies (nor­mal), a com­pre­hen­sive pan-

Workup of bleed­ing dis­or­ders  |  89

Table 3. Laboratory con­sid­er­ations for patients with a suspected bleed­ing dis­or­der of sec­ond­ary hemo­sta­sis, clot sta­bi­li­za­tion, or
fibri­no­ly­sis and nor­mal or nondiagnostic ini­tial tests (rou­tine clot­ting times, fibrin­og
­ en)

Problem Laboratory con­sid­er­ations

Factor deficiencies
Mild fac­tor deficiencies may not
pro­long clot­ting times
Prolongation of clot­ting times by a
- Perform fac­tor activ­ity assays with a step­wise approach focus­ing on more com­mon deficiencies first
(fac­tors VIII, IX, and XI), followed by less com­mon deficiencies (fac­tors II, V, VII, and X).
- Consider chro­mo­genic fac­tor VIII and IX assays due to poten­tial for clin­i­cally sig­nif­i­cant dis­crep­ant
results in nonsevere hemo­philia; chro­mo­genic fac­tor IX assays are less widely avail­­able.
- In dis­crep­ant nonsevere HA, a VWD panel may pro­vide a clue in the form of a decreased FVIII
activ­ity/VWF:Ag ratio, even if the abso­lute FVIII result is nor­mal; this is also seen in type 2N VWD.
- Factor activ­i­ties may be bor­der­line, confirm mild cases and female car­ri­ers of HA and HB
with genetic test­ing.
- Perform fac­tor assays if there is a his­tory of bleed­ing or a clin­i­cal con­di­tion strongly asso­ci­ated with

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/85/1851788/85smock.pdf by guest on 13 December 2021

lupus anti­co­ag­u­lant may mask a
fac­tor defi­ciency or inhib­i­tor
  FXIII defi­ciency does not pro­long
clot­ting times
Fibrinolytic dis­or­ders are poorly
defined and do not pro­long
clot­ting times
fac­tor deficiencies. Lupus anticoagulants often interfere with quantification of factor activity in clot-
based factor assays; exercise caution in interpretation.
- Clot sol­u­bil­ity tests are most widely avail­­able world­wide but lack sen­si­tiv­ity (pick up only the most
severe deficiencies) and spec­i­fic­ity (abnor­mal with fibrin­o­gen and other abnor­mal­i­ties).
- Activity assays are the pre­ferred first-line tests but may not accu­rately quan­tify the most severe
deficiencies or may over­es­ti­mate FXIII activ­ity.
- Consider FXIII anti­gen test­ing (sen­si­tive to severe defi­ciency at the cost of miss­ing qual­i­ta­tive abnor­
mal­i­ties) and use of test­ing com­bi­na­tions for diag­no­sis.
- FXIII defi­ciency subtyping should be performed by anti­gen test­ing or genetic test­ing due to impor­tant
treat­ment impli­ca­tions.
- Conditions are rare, and test­ing has lim­ited avail­abil­ity.
- Results are sig­nif­i­cantly affected by blood draw tim­ing and qual­ity.
- Results are dif­fi­cult to inter­pret due to acute phase responses, com­plex inter­ac­tions between fibri­no­
lytic sys­tem com­po­nents, and tech­ni­cal lim­i­ta­tions of assays (deficiencies are dif­fi­cult to iden­tify).
- Test PAI-1 (activ­ity and/or anti­gen) and tis­sue plas­min­o­gen acti­va­tor anti­gen con­cur­rently to look for
expected pat­terns.
- Platelet aggre­ga­tion is mostly nor­mal in Que­bec plate­let syn­drome; per­form genetic test­ing for PLAU
dupli­ca­tion.
- Viscoelastic stud­ies are sen­si­tive to severe defects but may lack sen­si­tiv­ity for mild or mod­er­ate
abnor­mal­i­ties.
- Fibrinolytic dis­or­ders remain dif­fi­cult to diag­nose; genetic test­ing has some clin­i­cal util­ity but is
not in wide­spread clin­i­cal use. Consider treating undi­ag­nosed dis­or­ders with tranexamic acid and
desmopressin

el of clot-based fac­tor assays (nor­mal), and FXIII activ­ity test­ Correspondence

ing (nor­mal). A chro­mo­genic FVIII assay showed a decreased
FVIII activ­ity of 30% (com­pared with the nor­mal clot-based
[1-stage] result of 65%). Mild HA was con­firmed by iden­ti­fi­ca­
tion of an F8 gene muta­tion (p.R546W).48 On rereview, his FVIII
activ­ity/VWF:Ag ratio was decreased on the ini­tial VWD panel
(65%/115%, ratio 0.57), which may have pro­vided a clue to the
diag­no­sis. A dis­cor­dant FVIII activ­ity/VWF:Ag ratio is also seen
in type 2N VWD.49
In con­clu­sion, in-depth knowl­edge of the var­i­abil­ity of bleed­
ing dis­or­der phe­no­types and lab­o­ra­tory test­ing per­for­mance
char­ac­ter­is­tics and lim­i­ta­tions resulted in a con­clu­sive diag­no­sis
for these patients.
Conflict-of-inter­est dis­clo­sure
Kristi J. Smock: no com­pet­ing finan­cial inter­ests to declare.
Karen A. Moser: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Kristi J. Smock: Off-label use of tranexamic acid and DDAVP for
bleeding of unknown cause is discussed.
Karen A. Moser: Off-label use of tranexamic acid and DDAVP for
bleeding of unknown cause is discussed.
Kristi J. Smock, Department of Pathology, University of Utah,
ARUP Laboratories, 500 Chipeta Way, Mail Stop 115-G04, Salt
Lake City, UT 84108; e-mail: kristi​­.smock@aruplab​­.com.
References
1. Thomas W, Downes K, Desborough MJR. Bleeding of unknown cause and
unclas­si­fied bleed­ing dis­or­ders: diag­no­sis, path­o­phys­i­­ol­ogy and man­age­
ment. Haemophilia. 2020;26(6):946-957.
2. Lippi G, Favaloro EJ. Laboratory hemo­sta­sis: from biol­ogy to the bench.
Clin Chem Lab Med. 2018;56(7):1035-1045.
3. Hayward CP, Moffat KA, Liu Y. Laboratory inves­ti­ga­tions for bleed­ing dis­or­
ders. Semin Thromb Hemost. 2012;38(7):742-752.
4. Hayward CPM. How I inves­ti­gate for bleed­ing dis­or­ders. Int J Lab Hematol.
2018;40(suppl 1):6-14.
5. Boender J, Kruip MJ, Leebeek FW. A diag­nos­tic approach to mild bleed­
ing dis­or­ders. J Thromb Haemost. 2016;14(8):1507-1516.
6. Lippi G, Franchini M, Favaloro EJ. Diagnostics of inherited bleed­ ing
dis­or­ders of sec­ond­ary hemo­sta­sis: an easy guide for rou­tine clin­i­cal lab­o­
ra­to­ries. Semin Thromb Hemost. 2016;42(5):471-477.
7. Zaninetti C, Greinacher A. Diagnosis of inherited plate­let dis­or­ders on a
blood smear. J Clin Med. 2020;9(2):539.
8. Bowman ML, James PD. Controversies in the diag­no­sis of type 1 von Wille-
brand dis­ease. Int J Lab Hematol. 2017;39(suppl 1):61-68.
9. Gresele P; Subcommittee on Platelet Physiology of the International Soci-
ety on Thrombosis and Hemostasis. Diagnosis of inherited plate­let func­
tion dis­or­ders: guid­ance from the SSC of the ISTH. J Thromb Haemost.
2015;13(2):314-322.

90  |  Hematology 2021  |  ASH Education Program

10. Mezzano D, Quiroga T, Pereira J. The level of lab­o­ra­tory test­ing required for
diag­no­sis or exclu­sion of a plate­let func­tion dis­or­der using plate­let aggre­
ga­tion and secre­tion assays. Semin Thromb Hemost. 2009;35(2):242-254.
11. Favaloro EJ, Bonar R. An update on qual­ity con­trol for the PFA-100/PFA-200.
Platelets. 2018;29(6):622-627.
12. Favaloro EJ. Clinical util­ity of clo­sure times using the plate­let func­tion ana­
lyzer-100/200. Am J Hematol. 2017;92(4):398-404.
13. Rand ML, Reddy EC, Isra­els SJ. Laboratory diag­no­sis of inherited plate­let
func­tion dis­or­ders. Transfus Apher Sci. 2018;57(4):485-493.
14. Artoni A, Bassotti A, Abbattista M, et al. Hemostatic abnor­mal­i­ties in patients
with Ehlers-Danlos syn­drome. J Thromb Haemost. 2018;16(12):2425-2431.
15. James PD, Connell NT, Ameer B, et al. ASH ISTH NHF WFH 2021 guide­lines
on the diag­no­sis of von Willebrand dis­ease. Blood Adv. 2021;5(1):280-300.
16. Mezzano D, Quiroga T. Diagnostic chal­lenges of inherited mild bleed­ing
dis­or­ders: a bait for poorly explored clin­i­cal and basic research. J Thromb
Haemost. 2019;17(2):257-270.
17. Baronciani L, Peyvandi F. How we make an accu­rate diag­no­sis of von Will-
ebrand dis­ease. Thromb Res. 2020;196:579-589.
18. Brown MC, White MH, Friedberg R, et al. Elevated von Willebrand fac­tor
lev­els dur­ing heavy men­strual bleed­ing epi­sodes limit the diag­nos­tic util­ity
for von Willebrand dis­ease. Res Pract Thromb Haemost. 2021;5(4):e12513.
19. Doshi BS, Rogers RS, Whitworth HB, et al. Utility of repeat test­ing in the
eval­u­a­tion for von Willebrand dis­ease in pedi­at­ric patients. J Thromb Hae-
most. 2019;17(11):1838-1847.
20. Sharma R, Flood VH. Advances in the diag­no­sis and treat­ment of von Will-
ebrand dis­ease. Blood. 2017;130(22):2386-2391.
21. Keesler DA, Flood VH. Current issues in diag­no­sis and treat­ment of von
Willebrand dis­ease. Res Pract Thromb Haemost. 2018;2(1):34-41.
22. Casonato A, Daidone V, Galletta E, Bertomoro A. Type 2B von Willebrand
dis­ease with or with­out large multimers: a dis­tinc­tion of the two sides of
the dis­or­der is long over­due. PLoS One. 2017;12(6):e0179566.
23. Goldfarb M, Czer LS, Lam LD, Moriguchi J, Arabia FA, Volod O. High molec­
u­lar weight von Willebrand fac­tor multimer loss and bleed­ing in patients
with short-term mechan­i­cal cir­cu­la­tory sup­port devices: a case series. J
Extra Corpor Technol. 2018;50(2):77-82.
24. Geisen U, Heilmann C, Beyersdorf F, et al. Non-sur­gi­cal bleed­ing in patients
with ven­tric­u­lar assist devices could be explained by acquired von Wille-
brand dis­ease. Eur J Cardiothorac Surg. 2008;33(4):679-684.
25. Tiede A, Rand JH, Budde U, Ganser A, Federici AB. How I treat the acquired
von Willebrand syn­drome. Blood. 2011;117(25):6777-6785.
26. Blackshear JL, McRee CW, Safford RE, et al. von Willebrand fac­tor abnor­
mal­i­ties and Heyde syn­drome in dys­func­tional heart valve pros­the­ses.
JAMA Cardiol. 2016;1(2):198-204.
27. Munnix ICA, Van Oerle R, Verhezen P, et al. Harmonizing light trans­mis­
sion aggregometry in the Netherlands by implementation of the SSC-
ISTH guide­line. Platelets. 2021;32(4):516-523.
28. Cattaneo M, Cerletti C, Harrison P, et al. Recommendations for the stan­
dard­ i­
za­
tion of light trans­ mis­sion aggregometry: a con­ sen­sus of the
Working Party from the Platelet Physiology Subcommittee of SSC/ISTH
[published online 10 April 2013]. J Thromb Haemost.
29. Brunet JG, Iyer JK, Badin MS, et  al. Electron micros­copy exam­i­na­tion of
plate­let whole mount prep­a­ra­tions to quan­ti­tate plate­let dense gran­ule
num­bers: impli­ca­tions for diag­nos­ing suspected plate­let func­tion dis­or­
ders due to dense gran­ule defi­ciency. Int J Lab Hematol. 2018;40(4):400-
407.
30. Badin MS, Graf L, Iyer JK, Moffat KA, Seecharan JL, Hayward CP. Variability
in plate­let dense gran­ule aden­o­sine tri­phos­phate release find­ings amongst
patients tested mul­ti­ple times as part of an assess­ment for a bleed­ing dis­
or­der. Int J Lab Hematol. 2016;38(6):648-657.
Prakash Singh Shekhawat
31. Dupuis A, Bordet J-C, Eckly A, Gachet C. Platelet δ-stor­age pool dis­ease: an
update. J Clin Med. 2020;9(8):2508.
32. Leo VC, Mor­gan NV, Bem D, et al; UK GAPP Study Group. Use of next-gen­
er­a­tion sequenc­ing and can­di­date gene anal­y­sis to iden­tify under­ly­ing
defects in patients with inherited plate­let func­tion dis­or­ders. J Thromb
Haemost. 2015;13(4):643-650.
33. Millington-Burgess SL, Harper MT. Gene of the issue: ANO6 and Scott syn­
drome. Platelets. 2020;31(7):964-967.
34. Toulon P, Eloit Y, Smahi M, et al. In vitro sen­si­tiv­ity of dif­fer­ent acti­vated
par­tial throm­bo­plas­tin time reagents to mild clot­ting fac­tor deficiencies.
Int J Lab Hematol. 2016;38(4):389-396.
35. Favaloro EJ, Kershaw G, Mohammed S, Lippi G. How to opti­mize acti­vated
par­tial throm­bo­plas­tin time (APTT) test­ing: solu­tions to establishing and
ver­i­fy­ing nor­mal ref­er­ence inter­vals and assessing APTT reagents for sen­si­
tiv­ity to hep­a­rin, lupus anti­co­ag­u­lant, and clot­ting fac­tors. Semin Thromb
Hemost. 2019;45(1):22-35.
36. Al-Huniti A, Sharathkumar A, Krantz M, Watkinson K, Bhagavathi S. Dis-
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/85/1851788/85smock.pdf by guest on 13 December 2021
crepant hemo­philia A: an underdiagnosed dis­ease entity. Am J Clin Pathol.
2020;154(1):78-87.
37. Kihlberg K, Strandberg K, Rosén S, Ljung R, Astermark J. Discrepancies
between the one-stage clot­ting assay and the chro­mo­genic assay in hae-
mophilia B. Haemophilia. 2017;23(4):620-627.
38. Konkle BA, Huston H, Nakaya Fletcher S. In: Adam MP, Ardinger HH, Pagon
RA, et al, eds. Hemophilia A. GeneReviews((R)). University of Washington;
2017.
39. Konkle BA, Huston H, Nakaya Fletcher S. In: Adam MP, Ardinger HH, Pagon
RA, et al, eds. Hemophilia B. GeneReviews((R)). University of Washington;
2017.
40. Moffat KA, Ledford-Kraemer MR, Plumhoff EA, et al. Are lab­o­ra­to­ries fol­low­
ing published rec­om­men­da­tions for lupus anti­co­ag­u­lant test­ing? An inter­
na­tional eval­u­a­tion of prac­tices. Thromb Haemost. 2009;101(1):178-184.
41. Karimi M, Peyvandi F, Naderi M, Shapiro A. Factor XIII defi­ciency diag­no­sis:
chal­lenges and tools. Int J Lab Hematol. 2018;40(1):3-11.
42. Durda MA, Wolberg AS, Kerlin BA. State of the art in fac­tor XIII lab­o­ra­tory
assess­ment. Transfus Apher Sci. 2018;57(6):700-704.
43. Sivapalaratnam S, Collins J, Gomez K. Diagnosis of inherited bleed­ing dis­
or­ders in the geno­mic era. Br J Haematol. 2017;179(3):363-376.
44. Jain S, Acharya SS. Inherited dis­or­ders of the fibri­no­lytic path­way. Transfus
Apher Sci. 2019;58(5):572-577.
45. Longstaff C. Measuring fibri­no­ly­sis: from research to rou­tine diag­nos­tic
assays. J Thromb Haemost. 2018;16(4):652-662.
46. Downes K, Megy K, Duarte D, et al; NIHR BioResource. Diagnostic high-
through­put sequenc­ing of 2396 patients with bleed­ing, throm­botic, and
plate­let dis­or­ders. Blood. 2019;134(23):2082-2091.
47. Simeoni I, Stephens JC, Hu F, et al. A high-through­put sequenc­ing test for
diag­nos­ing inherited bleed­ing, throm­botic, and plate­let dis­or­ders. Blood.
2016;127(23):2791-2803.
48. Johnsen JM, Fletcher SN, Huston H, et al. Novel approach to genetic anal­y­sis
and results in 3000 hemo­philia patients enrolled in the My Life, Our Future
ini­tia­tive. Blood Adv. 2017;1(13):824-834.
49. Seidizadeh O, Peyvandi F, Mannucci PM. Von Willebrand dis­ease type 2N:
an update. J Thromb Haemost. 2021;19(4):909-916.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000236
Workup of bleed­ing dis­or­ders  |  91
 CLOTTING AND BLEEDING CONUNDRUMS

Clots in unusual places: lots of stress, limited

data, critical decisions

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/92/1851781/92mathew.pdf by guest on 13 December 2021

Carol Mathew and Marc Zumberg
Internal Medicine, Division of Hematology/Oncology, University of Florida, Gainesville, FL

Although much less common than deep vein thrombosis of the lower extremities or lungs, clots in unusual locations,
including the splanchnic, cerebral, retinal, upper-extremity, and renal locations, present with significant morbidity and
mortality. In the last 2 decades, treatment of clots in these unusual locations is primarily managed medically, with inter-
ventional and surgical approaches reserved for more severe or refractory cases. The hematologist is well positioned to
provide consultation to organ-specific specialties (ie, neurosurgery, hepatology, ophthalmology), especially because
acquired and congenital hypercoagulability plays a major role, and anticoagulation is often the primary treatment. His-
torically, treatment has been based on expert opinion, but systematic reviews and meta-analyses have recently been
published. Various societies have produced guidelines for the treatment of clots in unusual locations; however, random-
ized clinical trial data remain scarce. In the last few years, increasing data have emerged concerning the efficacy of the
direct oral anticoagulants in treating clots in unusual locations. Cases have recently been described highlighting atypical
thrombosis associated with COVID-19 infection as well as with the ChAdOx1 nCoV-19 (AstraZeneca) vaccine and John-
son and Johnson’s Janssen Ad26.COV2.S vaccine. This article reviews clots in unusual locations with an emphasis on the
splanchnic (mesenteric, portal, splenic, hepatic) and cerebral circulation. Through a case-based approach, key questions
are posed, and data are presented to help guide diagnosis and treatment.

LEARNING OBJECTIVES
• Diagnose and treat clots in the splanchnic and cerebral veins despite (1) heterogeneous clinical presentations,
(2) limited high­level evidence, and (3) concomitant bleeding and clotting risks
• Work within a team consisting of organ­specialized physicians as well as hematologists

Introduction (eg, myeloproliferative diseases in splanchnic vein throm­

Although different definitions exist, clots in unusual sites typ­
ically refer to any location outside of the lower extremities
and pulmonary arteries.1,2 Unlike the treatment of deep vein
thrombosis (DVT) and pulmonary embolism, thrombosis at
unusual sites is much less common, and treatment is typi­
cally guided by expert opinion based on observational stud­
ies or small randomized trials.1,3 Despite the rarity of these
conditions, they are detected with increasing frequency due
to modern radiographic imaging, leading to earlier diagno­
sis and allowing the earlier institution of effective therapy.1
Often, these unique thromboses are provoked by a
pathologic condition in the organ supplied by the spe­
cific venous segments (eg, splenic vein thrombosis in a
patient with pancreatitis; portal vein thrombosis (PVT)
in a patient with cirrhosis).2 Frequently, these events are
associated with acquired and/or congenital hypercoagu­
lable disorders.2,4 Different thrombotic manifestations can
result from the various acquired and inherited conditions
boses [SVT] and estrogens in cerebral vein thrombosis
[CVT]) possibly due to complex interactions with the local
vessel wall, altering hemostatic balance and predisposing
to thrombosis.4 Familiarity with patients with hypercoagu­
lable disorders allows the hematologist to maintain a high
index of suspicion of these rare conditions and remain an
integral part of diagnosis and treatment.
During the prior year, clots in unusual locations such as
the splanchnic, cerebral, and retinal circulation have been
described in patients infected with COVID­19 (Table 1).5,6 In
addition, a subset of vaccine­related splanchnic vein and
cerebral vein clots (termed vaccine­induced thrombocy­
topenia with thrombosis or thrombotic thrombocytopenic
syndrome) that resemble autoimmune heparin­induced
thrombocytopenia with thrombosis has been described by
Greinacher.7 In this case series, 9 patients developed CVT,
and 3 patients developed SVT after vaccination with the
ChAdOx1 nCoV­19 (AstraZeneca) vaccine.7 Thrombosis in

92 | Hematology 2021 | ASH Education Program

Table 1.  COVID-19 vac­ci­na­tion and clots in unusual loca­tions:
key points
• The major­ity of throm­botic events asso­ci­ated with COVID-19 are DVT
and pul­mo­nary embolism
•  A
 n increas­ing num­ber of reports of splanch­nic and CVT have been
reported in asso­ci­a­tion with COVID-19 infec­tion
•  A
 lthough the inci­dence of post­vac­ci­na­tion atyp­i­cal throm­bo­sis
remains quite low, 9 cases of CVT and 3 cases of SVT have been
reported after vac­ci­na­tion with the ChAdOx1 nCoV-19 (AstraZeneca)
vac­cine as well as 6 cases of CVT and 2 of con­com­i­tant SVT after
Johnson & Johnson’s Janssen Ad26.COV2.S vac­cine
    T
○  he lit­er­a­ture surrounding the inci­dence, risk, and path­o­phys­i­­ol­ogy
con­tin­ues to evolve even as this arti­cle was writ­ten
    T
○  hese cases have been asso­ci­ated with throm­bo­cy­to­pe­nia resem­
bling auto­im­mune hep­a­rin-induced throm­bo­cy­to­pe­nia
    T
○  he ISTH has devel­oped treat­ment guide­lines that include
nonheparin anti­co­ag­u­lants as well as IVIG in con­firmed cases
•  M
 ost patients had no obvi­ous under­ly­ing risk fac­tors or known
thrombophilia
•  Most patients were treated with anti­co­ag­u­lant ther­apy
•  Mortality was high
•  C
 OVID-19 test­ing should be con­sid­ered as well as ascer­tain­ment of
vac­cine sta­tus, prod­uct, and tim­ing in patients who pres­ent with
unex­plained splanch­nic or CVT
IVIG, intravenous immunoglobulin.
the cere­bral and splanch­nic cir­cu­la­tion has also been described
after Johnson and Johnson’s Janssen Ad26.COV2.S vac­cine.8 At
the time of this writ­ing, it is recommended that hep­a­rin anti­
coagulation in these patients be avoided until an antiheparin
plate­let fac­tor 4 anti­body or func­tional assay returns neg­a­tive.7
In addi­tion, the International Society on Thrombosis and Haemo­
stasis (ISTH) recently published guide­lines suggesting the use of
intravenous immunoglobulin in addi­tion to nonheparin anticoag­
ulation in con­firmed cases.9
The treat­ment of clots in unusual loca­tions typ­i­cally con­sists
of anticoagulation and less com­monly fibri­no­lytic agents, with
interventional and sur­gi­cal tech­niques playing a sec­ond­ary role
and often reserved for pro­gres­sive organ dys­func­tion or infarc­
tion (eg, mes­en­teric vein throm­bo­sis).1,10,11 Data on the effi­cacy of
the direct oral anti­co­ag­u­lants (DOACs) com­pared to war­fa­rin will
be cov­ered in greater detail in the evi­dence-based minireview
Should Warfarin or a DOAC Be Used in Patients Presenting With
Thrombosis in the Splanchnic or Cerebral Veins?
In the remain­der of this arti­cle, we focus on a case-based
review of throm­bo­sis in the splanch­nic and cere­bral veins. Further
com­pre­hen­sive review of these top­ics as well as clots in other
unusual loca­tions can be found in sev­eral excel­lent reviews.1-4
SVT
SVT refers to venous throm­bo­em­bo­lism of the por­tal vein (PV),
mes­en­teric, splenic, or hepatic veins. Hepatic vein throm­bo­sis
(HVT) is typ­i­cally referred to as Budd-Chiari syn­drome (BCS) and
includes venous out­flow obstruc­tion any­where from the level of
the hepatic venules prox­i­mally to the junc­tion of the inferior vena
cava and right atrium.12 Although the inci­dence of SVT is at least
25 times less than lower-extrem­ity DVT, the pre­sen­ta­tion is often
much more dra­matic.1-4
Prakash Singh Shekhawat
Thrombosis in the splanch­nic cir­cu­la­tion is often due to a local
risk fac­tor such as liver cir­rho­sis, solid tumor, or pan­cre­a­ti­tis, typ­
i­cally in com­bi­na­tion with tran­sient risk fac­tors such as sur­gery or
local inflam­ma­tion.13 Myeloproliferative neo­plasms are the most
com­mon sys­temic risk fac­tor, occur­ring in 40% of HVT and 31%
of non­ma­lig­nant, noncirrhotic PVT.14 Next-gen­er­a­tion sequenc­ing
has also been recently used and has iden­ti­fied a novel JAK-2-exon
12 muta­tion in one-third of tri­ple-neg­a­tive patients with idi­o­
pathic or exclu­sively noncirrhotic SVT.15 Inherited thrombophilic
dis­or­ders are rec­og­nized risk fac­tors reported in mul­ti­ple sys­
temic reviews, but the inci­dence and pen­e­trance dif­fer between
the var­i­ous muta­tions and the var­i­ous sites in the splanch­nic cir­
cu­la­tion.3 SVT is con­sid­ered unpro­voked in 15% to 27% of cases
and diag­nosed inci­den­tally in up to one-third of patients.3
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/92/1851781/92mathew.pdf by guest on 13 December 2021
CLINICAL CASE 1
A 45-year-old man with alco­holic cir­rho­sis noted wors­en­ing
abdom­i­nal dis­ten­tion. Ultrasound revealed a cir­rhotic liver with
new throm­bo­sis of the PV as well as new asci­tes. The plate­let
count was sta­ble at 49 000/µL, and pro­throm­bin time (PT) was
19 sec­onds.
Question
What is the role of anticoagulation in PVT?
Data
When decid­ing on proper treat­ment for a patient with PVT, it is
impor­tant to con­sider whether the throm­bo­sis is acute or chronic
and whether the liver is cir­rhotic or not.16 The goals of anticoagu­
lation in PVT are to increase the chance of recan­a­li­za­tion in order
to improve hepatic blood flow and decrease variceal for­ma­tion
and sub­se­quent bleed­ing.13,17 A recent meta-anal­y­sis by Valeriani
and col­leagues consisted of 7969 SVT patients in 97 stud­ies (54%
of whom had PVT) and documented a recan­a­li­za­tion rate of 58%
vs 22%, a pro­gres­sion rate of 11% vs 15%, and major bleed­ing in
9% vs 16% in those receiv­ing vs not receiv­ing anticoagulation
(Figure 1).13 Compared to no treat­ment, anticoagulation was asso­
ci­ated with higher recan­a­li­za­tion (rel­a­tive risk [RR], 2.29; 95% CI,
1.66-3.44), lower pro­gres­sion (RR, 0.24; 95% CI, 0.13-0.42), major
bleed­ing (RR, 0.73; 95% CI, 0.58-0.92), and over­all mor­tal­ity (RR,
0.45; 95% CI, 0.33-0.60). The con­clu­sion was that anticoagulation
for SVT throm­bo­sis reduced the risk of throm­bus pro­gres­sion and
improved the rate of recan­a­li­za­tion with­out increas­ing bleed­ing
risk.13 Two sys­temic reviews and meta-ana­ly­ses showed the ben­e­
fit of anticoagulation in cir­rhotic patients with SVT.17,18 The first by
Valeriani reported decreased RRs of bleed­ing (RR, 0.52; 95% CI,
1.42-7.17) and mor­tal­ity (RR, 0.42; 95% CI, 0.24-0.73) and higher
recan­a­li­za­tion rates (RR, 3.19; 95% CI, 1.42-7.17) in the anticoagu­
lated group.17 The sec­ond by Ghazaleh also showed that antico­
agulation increased the rate of PV recan­a­li­za­tion and decreased
the rate of variceal bleed­ing in cir­rhotic patients with non­ma­lig­
nant PV throm­bo­sis.18 Prophylactic banding should be insti­tuted
in cir­rhotic patients to decrease variceal bleed­ing.3
An ISTH sub­com­mit­tee intro­duced recent guide­lines recom­
mending early anticoagulation in cir­rhotic and noncirrhotic patients
with acute SVT with no active bleed­ing or con­tra­in­di­ca­tions for at
Clots in unusual places | 93
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/92/1851781/92mathew.pdf by guest on 13 December 2021
Figure 1.  Outcomes of anticoagulant therapy for splanchnic vein thrombosis, as listed on the top.13

least 3 to 6 months.11 In patients with chronic throm­bo­sis, an indi­
vid­u­al­ized care plan is recommended, while treat­ment sim­i­lar to
symp­tom­atic acute throm­bo­sis is recommended for inci­den­tally
detected SVT.11 The 2017 guide­ lines by the Euro­ pean Associa­
tion for the Study of the Liver rec­om­mend anticoagulation for all­
patients with non­ma­lig­nant, noncirrhotic PVT for at least 6 months,
while the 2012 American College of Chest Physicians guide­lines
rec­om­mend only treating symp­tom­atic SVT and not anticoagulat­
ing those inci­den­tally detected unless the SVT is exten­sive or asso­
ci­ated with malig­nancy.2,19,20 However, more recent obser­va­tional
stud­ies and guide­lines have chal­lenged the American College of
Chest Physicians rec­om­men­da­tions.2,19 While still con­tro­ver­sial
and based pri­mar­ily on expert opin­ion, some experts rec­om­mend
decreas­ ing the inten­ sity of anticoagulation to half-ther­ a­ peu­
tic
doses in cir­rhotic patients if the plate­let counts are between 50 000
and 100 000/µL to pro­phy­lac­tic doses between 30 000 and
50 000/µL and with­hold­ing anticoagulation if the plate­let count is
below 30 000/µL.2 In patients with under­ly­ing coagulopathy, low-
molec­u­lar-weight hep­a­rin (LMWH) may be pre­ferred to war­fa­rin,
given the inabil­ity to ade­quately fol­low the effects of war­fa­rin with
PT mon­i­tor­ing. A recent sys­tem­atic review suggested that throm­
bolysis is effec­tive and safe for PVT in noncirrhotic patients who
have failed anticoagulation.21 Figure 2 depicts gen­eral treat­ment
rec­om­men­da­tions for SVT.2

Figure 2.  Recommendations for treatment of SVT.2

94  |  Hematology 2021  |  ASH Education Program


CLINICAL CASE 1 (Con­t in­u ed)
Consistent with the ISTH guide­lines ref­er­enced above, the deci­
sion was made to start anticoagulation given the new, symp­tom­
atic PV clot. We discussed the options of LWMH, war­fa­rin, and
DOACs with the patient. He opted for daily enoxaparin (Lovenox)
injec­tions. Given the plate­let count of 49 000/µL and a PT of 19
sec­onds, the deci­sion was made in con­junc­tion with the patient
and hepatology team to decrease the enoxaparin dose to 1 mg/
kg/d. He under­went suc­cess­ful liver trans­plan­ta­tion 4 months later,
and anticoagulation was stopped after his post­op­er­a­tive recov­ery.
CLINICAL CASE 2
A 43-year-old woman who has no prior med­i­cal his­tory pre­
sented with 2 weeks of pro­gres­sive jaun­dice, abdom­i­nal pain,
and a distended abdo­men. She took no med­i­ca­tions and did
not smoke or con­sume alco­hol. Her fam­ily his­tory was unknown.
Physical exam­i­na­tion was sig­nif­i­cant for scleral icterus and jaun­
dice. The abdo­men was distended, the liver edge was pal­pated
3 cm below the cos­tal mar­gin, and a fluid wave was detected.
There was mild ankle edema, but the rest of the exam­i­na­tion
was unre­mark­able. Hemoglobin was 16.7 g/dL with an MCV of
76 fL and a plate­let count of 608 000/µL. Imaging con­firmed
occlu­sive throm­bo­sis of the main hepatic vein.
Questions
How com­mon is the JAK2 V617F muta­tion in BCS? Is anti­co­ag­u­
lant ther­apy, local inter­ven­tion, or both pre­ferred in newly diag­
nosed BCS?
Data
BCS occurs due to obstruc­tion of the hepatic venous out­flow
any­where between the liver and the heart.12 An under­ly­ing
Figure 3.  Stepwise treatment algorithm for BCS.
Prakash Singh Shekhawat
prothrombotic state is iden­ti­fied in 88% of BCS patients.13 A
mye­lo­pro­lif­er­a­tive neo­plasm (MPN) has been reported in up to
62% of cases of idi­o­pathic BCS, with poly­cy­the­mia vera being
the most com­mon sub­type (18%-43%). A JAK2 V617F muta­tion
is found in 26% to 52% of patients.2 The diag­no­sis of BCS is
the presenting symp­tom of an MPN in 74% of cases, with many
patients hav­ing nor­mal blood counts at the time.2 Calreticulin
muta­tions have been reported in 2.9% of BCS patients, in the
absence of the JAK2 V617F muta­tion.12 The MPL muta­tion has
been reported in only a few cases.12 Any patient who pres­ents
with idi­o­pathic BCS should be eval­u­ated with an MPN genetic
panel consisting of at least the above muta­tions.12 Antiphos­
pholipid anti­body syn­drome and par­ox­ys­mal noc­tur­nal hemo­
glo­bin­uria are also over­rep­re­sented in patients with BCS and
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/92/1851781/92mathew.pdf by guest on 13 December 2021
should be con­sid­ered if an MPN panel is neg­a­tive.12
The man­age­ment of BCS, com­pared to other SVTs, is unique
as treat­ment gen­er­ally con­sists of both short- and long-term
anticoagulation as well as decom­pres­sion of the hepatic venous
out­flow track (Figure 3). Lifelong anticoagulation is typ­i­cally rec­
ommended for BCS, although based mostly on expert opin­ion
with no ran­dom­ized con­trolled tri­als hav­ing been reported.12,19
LMWH or unfractionated hep­a­rin is typ­i­cally begun, and endos­
copy is performed to eval­u­ate for and treat varices.12 Historically,
long-term anticoagulation has been with war­fa­rin.
Endovascular inter­ven­tion con­sists of angio­plasty, stenting,
and thrombolysis aimed at open­ing up the hepatic out­flow
tract and pre­ serv­ing hepatic func­ tion. Transvenous angio­
plasty is cur­rently performed in a major­ity of BCS patients.12,22
Stent place­ment should be pur­sued if there is inad­e­quate pres­
sure reduc­tion with angio­plasty, based on supe­rior results in
a recent ran­dom­ized clin­i­cal trial.12,23 If this is not effec­tive or
fea­si­ble, transjugular intrahepatic portosystemic shunt is usu­
ally performed to reduce por­tal pres­sure.3 Surgical shunting,
while the his­tor­i­cal treat­ment of choice, has mostly given way
to endovascular inter­ven­tion and anticoagulation.1,3 Using the
step­wise approach, 1-year and 5-year sur­vival have improved
Clots in unusual places | 95
to 96% and 89%, respec­tively.11 Liver trans­plan­ta­tion has been lev­els increased over the next 12 hours. The deci­sion was made
performed in patients with acute or pro­gres­sive liver fail­ure.12 to pro­ceed with explor­atory lap­a­rot­omy due to con­cern for
peri­to­ni­tis and necrotic bowel.

CLINICAL CASE 2 (Con­t in­u ed)

The patient was imme­di­ately placed on ther­a­peu­tic unfraction­
ated hep­a­rin, and angio­plasty with stenting of the main hepatic
vein was performed within 48 hours with excel­lent results. JAK-2
V617F test­ing sub­se­quently returned pos­i­tive, and she started
ther­a­peu­tic phle­bot­omy as well as hydroxy­urea. After a dis­cus­
sion about oral anti­co­ag­u­lant options, the patient decided to
be treated with apixaban 5 mg twice a day. Six months later
her hemo­glo­bin and plate­let count remained nor­mal, and there
Endovascular ther­apy (EVT) should be lim­ited to cen­ters with
exper­tise in the pro­ce­dure.24 In the afore­men­tioned sys­tem­atic
review, EVT was only used at 2 cen­ters, and the bowel resec­tion
rate remained high at 65% and 78%, respec­tively.24-26 However,
in an addi­tional study 5 out of 8 patients who under­went EVT
avoided sur­gi­cal resec­tion.27
Surgical explo­ra­tion is often required when bowel ische­mia
is suspected. However, the dis­tinc­tion between revers­ible and
irre­vers­ible bowel ische­mia is dif­fi­cult.24 Often a pri­mary lap­a­

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/92/1851781/92mathew.pdf by guest on 13 December 2021

was no recur­rent throm­bo­sis on a reg­i­men of peri­odic phle­bot­ rot­omy is performed, but only frankly necrotic bowel should be
omy, hydroxy­urea, and apixaban. resected.24 The abdo­men is typ­i­cally left open, and a sec­ond-
look oper­a­tion is performed at a later date.24

CLINICAL CASE 3
A 50-year-old woman with a prior his­tory of a left pop­li­teal DVT
devel­oped severe midabdominal pain, cramping, and fever.
The pain progressed over the next few days, and she presented
to her local emer­gency room. A com­plete blood count showed
a mild leu­ko­cy­to­sis with a left shift. Lactic acid was mod­er­ately
ele­
vated. A CT venogram (CTV) showed throm­ bo­sis of the
supe­rior mes­en­teric vein with bowel wall edema through­out
the jeju­num and prox­i­mal ileum.
Question
Should MVT be man­aged med­i­cally, sur­gi­cally, or both?
MVT is a rare cause of mes­en­teric ische­mia, representing
only 5% to 20% of cases.1,4,24 With improved imag­ing tech­
niques, includ­ ing contrasted CTV or mag­ netic res­
o­
nance
imag­ ing (MRI), the diag­ no­ sis can typ­ i­
cally be made radio­
graph­i­cally instead of sur­gi­cally.1 Guidelines published by the
Euro­pean Association for the Study of the Liver rec­om­mend
treat­ment with LMWH or, pref­er­a­bly, intra­ve­nous hep­a­rin, as
early lap­ ar­ot­
omy may be required.3,4,24 A recent sys­tem­atic
review of 11 published stud­ies recommended early and imme­
di­ate anticoagulation.24
CLINICAL CASE 3 (Con­t in­u ed)
Despite ther­a­peu­tic anticoagulation, intra­ve­nous flu­ids, bowel
rest, and anti­bi­ot­ics, she became hypo­ten­sive and lac­tic acid
CLINICAL CASE 3 (Con­tin­ued)
During the ini­ tial sur­
gi­
cal explo­
ra­
tion, 5 cm of necrotic jeju­
num was resected. Upon tran­sec­tion of the bowel wall, mul­
ti­
ple “worm­ like” clots were extruded. Anticoagulation was
restarted and she clin­i­cally improved and did not require addi­
tional resec­tion. A thrombophilia workup includ­ing JAK-2 V617F
test­ing was under­taken and returned sig­nif­i­cant for high-titer
anticardiolipin and beta-2-gly­co­pro­tein antibodies. Long-term
anticoagulation with war­fa­rin was ini­ti­ated.
CVT
CVT refers to throm­bo­ses of the cere­bral veins or dural venous
sinuses. CVT is rare, with an approx­i­mate inci­dence of 13.2 to 15.7
per mil­lion per­sons per year. CVT is more com­mon in women
than men (3:1) and in youn­ger age groups (mean age of 40).3
Clinical man­i­fes­ta­tions of CVT include neu­ro­log­i­cal symp­
toms rang­ing from head­aches, papilledema, focal def­i­cits, sei­
zures, enceph­a­lop­a­thy, and coma.28 As with other atyp­i­cal sites
of throm­bo­sis, risk fac­tors for CVT can be divided into local,
mechan­i­cal, and sys­temic fac­tors (Table 2).
CVT is con­firmed by neuroimaging. Most patients are ini­
tially eval­u­ated with a noncontrast CT to quickly eval­u­ate for
intra­cra­nial hem­or­rhage as well as other pathol­o­gies. How­
ever, unenhanced CT scans lack the sen­si­tiv­ity to detect the
major­ity of CVT. A con­trast-enhanced CT increases sen­si­tiv­ity

Table 2.  Risk fac­tors for CVT

Local risk fac­tors (8%-12%)
Infections involv­ing the ears,
sinuses, mouth, face, neck, CNS
Adapted from Riva et al.4
CNS, cen­tral ner­vous sys­tem.
Mechanical causes (2%-5%)
Trauma, lum­bar punc­ture,
jug­u­lar vein cath­e­ter­i­za­tion,
neu­ro­sur­gi­cal inter­ven­tions
CNS dis­or­ders (2%)
Arteriovenous malformation, dural
fis­tu­lae, CNS malig­nan­cies
Systemic risk fac­tors
Pregnancy/puer­pe­rium (10%-17% women),
hor­monal treat­ment (50%-53% women),
thrombophilia (approx. 33%), mye­lo­pro­lif­
er­a­tive neo­plasms (3%-4%),
COVID-19 infec­tion, COVID-19 vac­cines
(AstraZeneca, J&J)

96  |  Hematology 2021  |  ASH Education Program

to 88% to 99%, and MRI remains the gold stan­dard for diag­no­
sis.29 CTVs may be use­ful in patients who have a con­tra­in­di­ca­
tion to MRI.3
CLINICAL CASE 4
A 23-year-old woman presented to the emer­gency room with
wors­en­ing head­ache over the past week. She had no sig­nif­i­cant
past med­i­cal his­tory. Her only med­i­ca­tion was a com­bined oral
con­tra­cep­tive pill. A com­puted tomog­ra­phy (CT) scan of the
head with­out con­trast showed a “dense tri­an­gle sign.” While in
the emer­gency room, she suf­fered a tonic-clonic sei­zure. MRI
of the brain showed throm­bo­sis involv­ing the supe­rior sag­it­tal
agulated group (12.7% vs 18.3%; haz­ard ratio, 0.73; CI, 0.44-1.21)
was seen.32 A sub­se­quent sys­tem­atic review and meta-anal­y­sis
showed a trend toward decreased mor­tal­ity and neu­ro­log­i­cal
out­comes with LMWH com­pared to UFH.33 Bleeding events were
sim­i­lar in both groups.33 Based on these stud­ies, the most recent
guide­lines from the Euro­pean Stroke Organization (2017), the
Amer­i­can Heart Association/Amer­i­can Stroke Association (2014),
and the Amer­i­can Academy of Chest Physicians (2012) rec­om­
mend either UFH or LMWH for treat­ment of acute CVT even in
cases with asso­ci­ated cere­bral hem­or­rhage.27,34,35

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/92/1851781/92mathew.pdf by guest on 13 December 2021

sinus with an asso­ci­ated small area of intra­cra­nial hem­or­rhage.
Question
What is the ini­tial treat­ment of acute-phase CVT?
Many cases of CVT are accom­pa­nied by intra­ce­re­bral hem­
or­rhage, mak­ing treat­ment com­pli­cated given the con­cern for
wors­en­ing hem­or­rhage. Two ran­dom­ized tri­als exam­ined the
ben­e­fit of par­en­teral anticoagulation com­pared to pla­cebo in
the treat­ment of acute-phase CVT. In the first trial, published in
1991, patients were ran­dom­ized to intra­ve­nous hep­a­rin treat­ment
(n = 10) vs pla­cebo (n = 10). The hep­a­rin group had improved recov­
ery and bet­ter long-term out­comes.30 The sec­ond trial consisted
of 60 patients ran­dom­ized to nadroparin vs pla­cebo. Better out­
comes in the nadroparin arm were noted, although they did not
reach sta­tis­ti­cal sig­nif­i­cance.31 Although sam­ple sizes were small,
these tri­als dem­on­strated that anticoagulation with hep­a­rin or
LMWH is bet­ter than no treat­ment in acute CVT. Approximately
43% of these patients had intra­cra­nial hem­or­rhage at diag­no­sis,
and none of the patients ran­dom­ized to anticoagulation devel­
oped a new intra­cra­nial hem­or­rhage, whereas 3 patients in the
pla­cebo group expe­ri­enced a new hem­or­rhage.30,31 In the Inter­
national Study on Cerebral Vein and Dural Sinus Thrombosis, a
trend toward lower rates of death or depen­dency in the antico­
CLINICAL CASE 4 (Con­tin­ued)
The patient was started on an anti­ep­i­lep­tic and intra­ve­nous
hep­a­rin; how­ever, she devel­oped wors­en­ing men­tal sta­tus
requir­ing mechan­i­cal intu­ba­tion. Repeat imag­ing did not show
new intra­cra­nial hem­or­rhage. The neurointerventional radi­ol­
ogy team was consulted to eval­u­ate for thrombectomy.
Question
What is the role of endovascular inter­ven­tions in the treat­ment
of CVT?
The evi­dence thus far sug­gests no added ben­e­fit from endo­
vascular thrombolysis or thrombectomy. Data pri­mar­ily come
from sys­tem­atic reviews that show higher fatal­ity rates, more
bleed­ing (mostly intra­cra­nial), and poor neu­ro­log­i­cal out­comes
with thrombolysis (Table 3). A ran­dom­ized con­trolled trial (TO-
ACT) exam­ined the ben­e­fit of thrombolysis in addi­tion to antico­
agulation and was stopped pre­ma­turely for futil­ity.36 Given the
lack of high-qual­ity evi­dence suggesting ben­e­fit, the con­sen­sus
is to avoid up-front EVT except in select cases with high pre­
treat­ment risk of poor out­comes.
CVT can lead to other com­ pli­
ca­
tions, includ­ ing sei­
zures,
ele­vated intra­cra­nial pres­sure, brain her­ni­a­tion, and infec­tions,
requir­ing addi­tional sup­port­ive mea­sures (Figure 4).

Table 3.  Endovascular ther­apy in acute CVT

No of
Trial Study design patients Intervention Results Limitations
Coutinho et al 36
Randomized, 67 Endovascular treat­ment in patients Stopped pre­ma­turely for Small sam­ple size
(TO-ACT trial) con­trolled Inclusion cri­te­ria: ≥1 Risk fac­tor for futil­ity
clin­i­cal dete­ri­o­ra­tion (coma, men­
tal sta­tus changes, CVT in deep
venous sys­tem, intra­ce­re­bral
hem­or­rhage)
Dentali et al37 Systematic review of 156 Mechanical Death rate 9%
15 stud­ies thrombectomy/thrombolysis Major bleed­ing 10% with local
thrombolysis
8% intra­cra­nial hem­or­rhage
58% fatal
Siddiqui et al40 Systematic review of 185 Mechanical 84% good out­come Concern for bias as
42 stud­ies thrombectomy/thrombolysis (71%) 12% mor­tal­ity the stud­ies were
60% pre­treat­ment intra­cra­nial 10% new/wors­ened intra­ce­re­ not blinded
hem­or­rhage bral hem­or­rhage
47% stu­por, coma 95% recan­a­li­za­tion rate

Prakash Singh Shekhawat

Clots in unusual places | 97

Figure 4.  Recommendations for treatment of CVT.2
CLINICAL CASE 4 (Con­t in­u ed)
In the fol­low­ing days, she slowly recov­ered and was extubated,
with a plan to dis­charge her to a reha­bil­it­ a­tion facil­ity.
Question
What is the best long-term treat­ment for CVT?
The Cerebral Vein Thrombosis International Study reported
on 706 patients with CVT. Of these, 85% were treated with hep­a­
rin in the acute phase followed by 84% with war­fa­rin for a median
dura­tion of 12 months. At a median fol­low-up of 40 months, 89.1%
of patients were found to have com­pletely recov­ered.37 In the
VENOST study, 67% of 1144 CVT patients were treated with war­
fa­rin after ini­tial treat­ment with hep­ar­ in. At 1-year fol­low-up, 93.1%
of 691 patients for whom data were avail­­able showed a com­plete
recov­ery.38 Most soci­ety guide­lines rec­om­mend the use of war­fa­
rin, although emerg­ing data show that DOACs may have similar
efficacy and safety as compared to war­fa­rin. This topic will be
cov­ered in greater detail in the accom­pa­ny­ing evi­dence-based
minireview Should Warfarin or a DOAC Be Used in Patients Pre­
senting With Thrombosis in the Splanchnic or Cerebral Veins?
There is no defin­i­tive evi­dence regard­ing the opti­mal dura­tion
of anticoagulation in the treat­ment of CVT. Extrapolating from
evi­dence from lower-extrem­ity DVT, most experts rec­om­mend 3
to 12 months of anticoagulation. If CVT occurred in the set­ting of
a tran­sient risk fac­tor, 3 to 6 months of anticoagulation may be
ade­quate. If there were no pro­vok­ing fac­tors, we treat for at least
98  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/92/1851781/92mathew.pdf by guest on 13 December 2021
6 to 12 months. In cases in which per­sis­tent hyper­co­ag­u­la­bil­ity
and a risk for throm­bo­sis have been iden­ti­fied, long-term anti­
coagulation should be con­sid­ered. Similarly, a lon­ger dura­tion of
anticoagulation is suggested for the treat­ment of recur­rent CVT.
The EXCOA-CVT study is an ongo­ing pro­spec­tive study that aims
to com­pare the out­comes of short-dura­tion (3-6 months) vs long-
term (12 months) anticoagulation treat­ment in CVT patients.39
Conflict-of-inter­est dis­clo­sure
Carol Mathew: no com­pet­ing finan­cial inter­ests to declare.
Marc Zumberg: mem­ber: American Board of Internal Medicine
Hematology Board; Amer­i­can Society of Hematology speaker
reim­burse­ment, law case review.
Off-label drug use
Carol Mathew: direct oral anticoagulants for splanchnic and
cerebral vein thrombosis are discussed.
Marc Zumberg: direct oral anticoagulants for splanchnic and
cerebral vein thrombosis are discussed.
Correspondence
Marc Zumberg, University of Florida Health, 1600 SW Archer Rd,
Gainesville, FL 32610; e-mail: zumbems@med­i­cine​­.ufl​­.edu.
References
1. Rajasekhar A, Zumberg M. Venous throm­bo­ses at unusual sites. In: Kitchens
CS, Alving BM, Kessler CM, eds. Consultative Hemostasis and Thrombosis.
3rd ed. W. B. Saunders Co.; 2013:262-290.
2. Abbattista M, Capecchi M, Martinelli I. Treatment of unusual throm­botic
man­i­fes­ta­tions. Blood. 2020;135(5):326-334.
3. Riva N, Ageno W. Cerebral and splanch­nic vein throm­bo­sis: advances,
chal­lenges, and unanswered ques­tions. J Clin Med. 2020;9(3):743.
4. Shatzel JJ, O’Donnell M, Olson SR, et al. Venous throm­bo­sis in unusual sites:
a prac­ti­cal review for the hema­tol­o­gist. Eur J Haematol. 2019;102(1):53-62.
5. Mowla A, Shakibajahromi B, Shahjouei S, et al. Cerebral venous sinus throm­
bo­sis asso­ci­ated with SARS-CoV-2: a mul­ti­na­tional case series. J Neurol Sci.
2020;419(15 Decem­ber):117183.
6. Singh B, Kaur P, Maroules M. Splanchnic vein throm­bo­sis in COVID-19: a
review of lit­er­a­ture. Dig Liver Dis. 2020;52(12):1407-1409.
7. Greinacher A, Thiele T, Warkentin TE, Weisser K, Kyrle PA, Eichinger S.
Thrombotic throm­bo­cy­to­pe­nia after ChAdOx1 nCov-19 vac­ci­na­tion. N Engl
J Med. 2021;384(22):2092-2101.
8. Schultz NH, Sørvoll IH, Michelsen AE, et al. Thrombosis and throm­bo­cy­to­
pe­nia after ChAdOx1 nCoV-19 vac­ci­na­tion. N Engl J Med. 2021;384(22):2124-
2130.
9. Nazy I, Sachs UJ, Arnold DM, et al. Recommendations for the clin­i­cal and
lab­o­ra­tory diag­no­sis of VITT against COVID-19: com­mu­ni­ca­tion from the
ISTH SSC sub­com­mit­tee on plate­let immu­nol­ogy. J Thromb Haemost.
2021;19(6):1585-1588.
10. Riva N, Carrier M, Gatt A, Ageno W. Anticoagulation in splanch­nic and cere­
bral vein throm­bo­sis: an inter­na­tional vignette-based sur­vey. Res Pract
Thromb Haemost. 2020;4(7):1192-1202.
11. Di Nisio M, Valeriani E, Riva N, Schulman S, Beyer-Westendorf J, Ageno W.
Anticoagulant ther­apy for splanch­nic vein throm­bo­sis: ISTH SSC sub­com­
mit­tee con­ trol of anticoagulation. J Thromb Haemost. 2020;18(7):1562-
1568.
12. Sharma A, Keshava SN, Eapen A, Elias E, Eapen CE. An update on the man­
age­ment of Budd-Chiari syn­drome. Dig Dis Sci. 2021;66(6):1780-1790.
13. Valeriani E, Di Nisio M, Riva N, et  al. Anticoagulant ther­apy for splanch­
nic vein throm­bo­sis: a sys­tem­atic review and meta-anal­y­sis. Blood.
2021;137(9):1233-1240.
14. De Stefano V, Qi X, Betti S, Rossi E. Splanchnic vein throm­bo­sis and mye­
lo­pro­lif­er­a­tive neo­plasms: molec­u­lar-driven diag­no­sis and long-term treat­
ment. Thromb Haemost. 2016;115(2):240-249.
15. Magaz M, Alvarez-Larrán A, Colomer D, et al. Next-gen­er­a­tion sequenc­ing
in the diag­no­sis of non-cir­rhotic splanch­nic vein throm­bo­sis. J Hepatol.
2021;74(1):89-95.
16. Intagliata NM, Caldwell SH, Tripodi A. Diagnosis, devel­op­ment, and treat­
ment of por­tal vein throm­bo­sis in patients with and with­out cir­rho­sis. Gas-
troenterology. 2019;156(6):1582-1599.e11599e1.
17. Valeriani E, Di Nisio M, Riva N, et al. Anticoagulant treat­ment for splanch­nic
vein throm­bo­sis in liver cir­rho­sis: a sys­tem­atic review and meta-anal­y­sis.
Thromb Haemost. 2021;121(7):867-876.
18. Ghazaleh S, Beran A, Aburayyan K, et al. Efficacy and safety of anticoagula­
tion in non-malig­nant por­tal vein throm­bo­sis in patients with liver cir­rho­sis:
a sys­tem­atic review and meta-anal­y­sis. Ann Gastroenterol. 2021;34(1):104-
110.
19. Euro­pean Association for the Study of the Liver. EASL Clinical Practice
Guidelines: vas­cu­lar dis­eases of the liver. J Hepatol. 2016;64(1):179-202.
20. Kearon C, Akl EA, Comerota AJ, et  al. Antithrombotic ther­apy for VTE
dis­ease: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed:
Amer­i­can College of Chest Physicians Evidence-Based Clinical Practice
Guidelines. Chest. 2012;141(suppl 2):e419S-e496S.
21. Cheng Q , Tree K. Systematic review of thrombolysis ther­apy in the man­
age­ment of non-cir­rho­sis-related por­tal vein throm­bo­sis. J Gastrointest
Surg. 2021;25(6):1579-1590.
22. Tripathi D, Sunderraj L, Vemala V, et al. Long-term out­comes fol­low­ing per­
cu­ta­ne­ous hepatic vein recan­a­li­za­tion for Budd-Chiari syn­drome. Liver Int.
2017;37(1):111-120.
23. Wang Q , Li K, He C, et al. Angioplasty with ver­sus with­out rou­tine stent
place­ment for Budd-Chiari syn­drome: a randomised con­trolled trial. Lan-
cet Gastroenterol Hepatol. 2019;4(9):686-697.
Prakash Singh Shekhawat
24. Acosta S, Salim S. Management of acute mes­en­teric venous throm­bo­sis: a
sys­tem­atic review of con­tem­po­rary stud­ies. Scand J Surg. 2020;110(2):123-
129.
25. Liu K, Liu S, Li L, et al. Evaluation of endovascular ther­apy com­bined with
bowel resec­ tion treat­ment on patients with acute mes­ en­teric venous
throm­bo­sis. Ann Vasc Surg. 2020;65(May):72-81.
26. Yang S, Zhang L, Liu K, et al. Postoperative cath­e­ter-directed thromboly­
sis ver­sus sys­temic anticoagulation for acute supe­rior mes­en­teric venous
throm­bo­sis. Ann Vasc Surg. 2016;35(August):88-97.
27. Salim S, Zarrouk M, Elf J, Gottsäter A, Ekberg O, Acosta S. Improved prog­
no­sis and low fail­ure rate with anticoagulation as first-line ther­apy in mes­
en­teric venous throm­bo­sis. World J Surg. 2018;42(11):3803-3811.
28. Ferro JM, Bousser MG, Canhão P, et al; Euro­ pean Stroke Organization.
Euro­pean Stroke Organization guide­line for the diag­no­sis and treat­ment
of cere­bral venous throm­bo­sis—endorsed by the Euro­pean Academy of
Neurology. Eur J Neurol. 2017;24(10):1203-1213.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/92/1851781/92mathew.pdf by guest on 13 December 2021
29. Capecchi M, Abbattista M, Martinelli I. Cerebral venous sinus throm­bo­sis.
J Thromb Haemost. 2018;16(10):1918-1931.
30. Einhäupl KM, Villringer A, Meister W, et  al. Heparin treat­ ment in sinus
venous throm­bo­sis. Lancet. 1991;338(8767):597-600.
31. de Bruijn SF, Stam J. Randomized, pla­cebo-con­trolled trial of anti­co­ag­u­lant
treat­ment with low-molec­u­lar-weight hep­a­rin for cere­bral sinus throm­bo­
sis. Stroke. 1999;30(3):484-488.
32. Ferro JM, Canhão P, Stam J, Bousser MG, Barinagarrementeria F; ISCVT
Investigators. Prognosis of cere­bral vein and dural sinus throm­bo­sis: results
of the International Study on Cerebral Vein and Dural Sinus Thrombosis
(ISCVT). Stroke. 2004;35(3):664-670.
33. Qureshi A, Perera A. Low molec­u­lar weight hep­a­rin ver­sus unfractionated
hep­a­rin in the man­age­ment of cere­bral venous throm­bo­sis: a sys­tem­atic
review and meta-anal­y­sis. Ann Med Surg (Lond). 2017;17(May):22-26.
34. Kernan WN, Ovbiagele B, Black HR, et al; Amer­ i­
can Heart Association
Stroke Council, Council on Cardiovascular and Stroke Nursing, Council on
Clinical Cardiology, and Council on Peripheral Vascular Disease. Guidelines
for the pre­ven­tion of stroke in patients with stroke and tran­sient ische­mic
attack: a guide­line for healthcare pro­fes­sion­als from the Amer­i­can Heart
Association/Amer­i­can Stroke Association. Stroke. 2014;45(7):2160-2236.
35. Lansberg MG, O’Donnell MJ, Khatri P, et al. Antithrombotic and throm­bo­
lytic ther­apy for ische­mic stroke: Antithrombotic Therapy and Prevention of
Thrombosis, 9th ed: Amer­i­can College of Chest Physicians Evidence-Based
Clinical Practice Guidelines. Chest. 2012;141(suppl 2):e601S-e636S.
36. Coutinho JM, Ferro JM, Zuurbier SM, et al. Thrombolysis or anticoagulation
for cere­bral venous throm­bo­sis: ratio­nale and design of the TO-ACT trial.
Int J Stroke. 2013;8(2):135-140.
37. Dentali F, Poli D, Scoditti U, et al; Cerebral Vein Thrombosis International
Study Investigators. Long-term out­comes of patients with cere­bral vein
throm­bo­sis: a mul­ti­cen­ter study. J Thromb Haemost. 2012;10(7):1297-1302.
38. Duman T, Uluduz D, Midi I, et al; VENOST Study Group. A mul­ti­cen­ter study
of 1144 patients with cere­ bral venous throm­bo­ sis: the VENOST study.
J Stroke Cerebrovasc Dis. 2017;26(8):1848-1857.
39. Miranda B, Aaron S, Arauz A, et al. The ben­e­fit of EXtending oral antiCO­
Agulation treat­ment (EXCOA) after acute cere­bral vein throm­bo­sis (CVT):
EXCOA-CVT clus­ter ran­dom­ized trial pro­to­col. Int J Stroke. 2018;13(7):771-
774.
40. Siddiqui FM, Dandapat S, Banerjee C, et  al. Mechanical thrombectomy
in cere­bral venous throm­bo­sis: sys­tem­atic review of 185 cases. Stroke.
2015;46(5):1263-1268.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000237
Clots in unusual places | 99
 CLOTTING AND BLEEDING CONUNDRUMS

EVIDENCE-BASED MINIREVIEW

Should warfarin or a direct oral anticoagulant

be used in patients presenting with thrombosis

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/100/1851729/100mathew.pdf by guest on 13 December 2021

in the splanchnic or cerebral veins?
Carol Mathew and Marc Zumberg
University of Florida, Internal Medicine, Division of Hematology/Oncology, Gainesville, FL

Case 1: A 23-year-old female third-year medical student who has no medical history seeks treatment for abdominal dis-
tention. She takes an estrogen-containing birth control pill and does not smoke or consume alcohol. Family history is
unremarkable. Physical examination is significant for abdominal distention, and an abdominal fluid wave is detected.
Complete blood count is normal. Imaging confirms occlusive thrombosis of the main portal vein. On endoscopy, grade 1
to 2 esophageal varices are noted and banded. Unfractionated heparin is begun. Subsequent workup reveals a homozy-
gous factor V Leiden mutation. Long-term anticoagulation is planned, and she asks if warfarin can be avoided given her
hectic ward rotations, erratic diet, and need for monitoring. Case 2: A 35-year-old woman who has no medical history
seeks treatment for progressively worsening posterior headaches for 1 week. Magnetic resonance imaging of the brain
shows dural sinus thrombosis with associated small areas of petechial cerebral hemorrhage. She is started on a contin-
uous unfractionated heparin infusion and admitted to the hospital for further observation. Her grandmother is on warfa-
rin for atrial fibrillation, and the patient would prefer to avoid warfarin because she does not think she can comply with
the frequent monitoring that will be required. She inquires about other oral anticoagulant options for her condition.

LEARNING OBJECTIVES
• Discuss the relative risks, benefits, and unique characteristics of the DOACs vs warfarin in patients with thrombosis
in the splanchnic or cerebral veins
• Choose an oral anticoagulant in a patient with thrombosis in the splanchnic or cerebral veins

Introduction
When not otherwise contraindicated, most experts and
societal guidelines recommend early anticoagulation in pa-
tients with splanchnic or cerebral vein thrombosis (CVT).1–6
However, many of these recommendations are based pri-
marily on small case series, as only a few randomized trials
exist.4–8 Whether anticoagulation is recommended for a lim-
ited time period or indefinitely depends on whether the clot
was provoked and if the provocation was transient or will be
longstanding.1–6 Most of the studies referenced in this arti-
cle do not clearly identify whether outcomes differed be-
tween provoked and unprovoked splanchnic or CVT. Most
experts recommend 3 to 12 months of anticoagulation for
a provoked splanchnic or CVT. Long-term anticoagulation
is often recommended for persistent (severe thrombophilia,
cancer, myeloproliferative disorder, etc) risk factors, recur-
rent thrombosis, or hepatic vein thrombosis.9 Historically,
warfarin has been the only oral anticoagulant available for
use in patients with chronic splanchnic or CVT.
Since 2010, 4 direct oral anticoagulants (DOACs) have
been approved by the US Food and Drug Administration for
the treatment of venous thromboembolism (VTE) Based on
large, randomized trials, both the CHEST and ASH guidelines
recommend DOACs as the treatment of choice for oral anti-
coagulation in patients with deep vein thrombosis or pul-
monary embolism.9–11 Patients with atypical site thrombosis
were generally excluded from these trials, and therefore
recommendations concerning the use of DOACs in splanch-
nic vein thrombosis (SVT) or CVT remain challenging. The
DOACs have several advantages over warfarin, including
fewer drug-drug interactions, fixed dosing, shorter half-life,
lack of need for monitoring, quick onset and offset, and less
major bleeding in large randomized clinical trials. In the past
several years, small case series and a few randomized trials

100 | Hematology 2021 | ASH Education Program

Table 1. DOAC use in SVT and CVT

•  Evidence is lim­ited but grow­ing. Few ran­dom­ized tri­als have been com­pleted.
•  Most ret­ro­spec­tive and pro­spec­tive data show the DOACs are at least as effec­tive as VKAs, and bleed­ing risk is not higher.
•  Multiple tri­als are cur­rently accru­ing patients.
•  Despite being off label, in a recent sur­vey, DOACs were used in 28% of low bleed­ing risk patients.
•  For patients with SVT, DOACs are contraindicated in Child-Pugh class C liver dis­ease and for rivaroxaban in class B and C liver dis­ease.
• For patients with CVT, con­sider inter­ac­tions with anti­ep­i­lep­tic drugs, and if con­com­i­tant intracranial hemorrhage, con­sider ini­tial use of a
short-act­ing anti­co­ag­u­lant.

have been published reporting on the effi­cacy and bleed­ing risk vein treated with apixaban (n = 20), rivaroxaban (n = 65), and dab-

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/100/1851729/100mathew.pdf by guest on 13 December 2021

of the DOACs in splanch­nic and CVT.7,8,12 Table 1 sum­ma­rizes our
inter­pre­ta­tion of this lit­er­a­ture.
CASE 1
A 23-year-old female third-year med­i­cal stu­dent who has no med­
i­cal his­tory seeks treat­ment for abdom­i­nal dis­ten­tion. She takes
an estro­gen-containing birth con­trol pill and does not smoke or
con­sume alco­hol. Family his­tory is unre­mark­able. Physical exam­
i­na­tion is sig­nif­i­cant for abdom­i­nal dis­ten­tion, and an abdom­i­nal
fluid wave is detected. Complete blood count is nor­mal. Imaging
con­firms occlu­sive throm­bo­sis of the main por­tal vein. On endos­
copy, grade 1 to 2 esoph­a­geal varices are noted and banded.
Unfractionated hep­a­rin is begun. Subsequent workup reveals a
homo­zy­gous fac­tor V Leiden muta­tion. Long-term anticoagula-
tion is planned, and she asks if war­fa­rin can be avoided given
her hec­tic ward rota­tions, erratic diet, and need for mon­i­tor­ing.
Only a sin­gle ran­dom­ized clin­i­cal trial com­par­ing a DOAC to
war­fa­rin has been reported in patients with SVT.12 Hanafy et al12
stud­ied 80 patients with acute non­ma­lig­nant por­tal vein throm­
bo­sis (PVT) with hep­a­ti­tis C–related com­pen­sated liver dis­ease.
After ini­tial treat­ment with enoxaparin at ther­a­peu­tic doses for
3 days, patients at 2 Egyp­tian cen­ters were ran­dom­ized between
war­fa­rin with a tar­get inter­na­tional nor­mal­ized ratio (INR) of 2 to
2.5 and rivaroxaban 10 mg bid. Recanalization of the por­tal vein
was found in 100% of patients in the rivaroxaban group com­
pared with 45% in the war­fa­rin group (P = .001). Major gas­tro­in­
tes­ti­nal (GI) bleed­ing was not seen in the rivaroxaban group but
occurred in 43% of patients in the war­fa­rin group (P = .001).12 No
deaths occurred in the rivaroxaban group com­pared with 36% in
the war­fa­rin group (P = .001).12
Riva and Ageno9 sum­ma­rize most of the remaining evi­dence
reporting on the use DOACs in SVT, which con­sists mostly of
small obser­va­tional tri­als, the major­ity ret­ro­spec­tive in nature.
Serrao et al13 describe a recent pro­spec­tive study of 28 patients
with SVT and ongo­ing risk fac­tors maintained on chronic treat­
ment with war­fa­rin who were switched to DOACs and com­pared
out­comes to 42 patients remaining on war­fa­rin. There was no
dif­fer­ence in throm­botic events between the groups. No major
bleed­ing events occurred in either group, while minor bleed­ing
did not occur in the DOAC group but was documented in 26% of
war­fa­rin-treated patients (P = .09).13
In the larg­est of the ret­ro­spec­tive tri­als, Naymagon et al14 in A 35-year-old woman who has no med­i­cal his­tory seeks treat­
Blood Advances reported on 93 patients with noncirrhotic
Prakash Singh Shekhawat portal ment for pro­gres­sively wors­en­ing pos­te­rior head­aches for 1 week.
igatran (n = 8) com­pared with low molec­u­lar weight hep­a­rin
(LMWH) (n = 70) and war­fa­rin (n = 108).14 Seventy-three per­cent of
patients had predisposing fac­tors for PVT. Complete res­o­lu­tion
of throm­bo­sis was seen in 66% of DOAC-treated patients com­
pared with 57% with LMWH and 31% with war­fa­rin (haz­ard ratio,
2.91; 95% CI, 1.87-4.52 for DOAC vs war­fa­rin).14 There was less
major bleed­ing with the DOACs com­pared with war­fa­rin (haz­ard
ratio for DOACs: war­fa­rin, 0.20; 95% CI, 0.05-0.86; P = .0307).14
A recent sys­tem­atic review and meta-anal­y­sis published in
Blood by Valeriani et al4 reported on 79 patients with SVT in 5
stud­ies who were treated with DOACs. SVT was unpro­voked in
25% of cases. The rates of SVT recan­a­li­za­tion, throm­bo­sis pro­
gres­sion, major bleed­ing, and over­all mor­tal­ity did not sig­nif­i­
cantly dif­fer between type of anti­co­ag­u­lant used.4 A 2021 study
by Kawata et al15 also showed no dif­fer­ence in clot recan­a­li­za­tion
between patients with SVT treated with LMWH/war­fa­rin or with
any of the DOACs. Local risk fac­tors were pres­ent in 118 (76.1%)
patients (sec­ond­ary SVT), whereas 7 (4.5%) patients had no risk
fac­tors and 30 (19.4%) patients had only sys­temic or thrombo-
philia (pri­mary SVT).15
Limited data exist concerning the use of DOACs in Budd-Chiari
syn­drome and mes­en­teric vein throm­bo­sis (MVT). In a ret­ro­spec­
tive anal­y­sis, patients with Budd-Chiari syn­drome treated after
endovascular inter­ ven­ tion with dabigatran (n = 36) were com­
pared with those tak­ing war­fa­rin (n = 62).16 Baseline data on hyper­
co­ag­u­la­ble states were avail­­able in 45 (45.9%) of 98 patients.16
At 12 months, stent patency rates (91% vs 93%), major bleed­ing
(3.5% vs 6.5%), and the com­pos­ite end point of mor­tal­ity and
major bleed­ing (4% vs 8%) did not dif­fer between the dabiga-
tran- and war­fa­rin-treated groups.16 In a ret­ro­spec­tive study of
102 patients with MVT, no dif­fer­ence was seen between patients
treated with a DOAC or war­fa­rin in terms of recan­a­li­za­tion (69%
vs 71%, P = .88) or major bleed­ing (9.1% vs 14.3%, P = .54).16 No
recur­rences of MVT were seen in either group.17
ADAM VTE, com­pared with the HOKUSAI VTE, SELECT-D, and
CARAVAGGIO stud­ies, is the only large, ran­dom­ized treat­ment trial
of DOACs in patients with can­cer to include patients with SVT.18-21
Of the enrolled patients, 8% of those receiv­ing apixaban had a SVT
as the qual­i­fy­ing event, as opposed to 18% in the dalteparin group.
In the group treated with apixaban, there was a sin­gle recur­rent
event (lower extrem­ity throm­bo­sis) and no major bleed­ing.18
CASE 2

Warfarin vs DOAC in SVT or CVT  |  101

Magnetic res­ o­
nance imag­ ing of the brain shows dural sinus
throm­bo­sis with asso­ci­ated small areas of pete­chial cere­bral
hem­or­rhage. She is started on a con­tin­uo­ us unfractionated hep­
a­rin infu­sion and admit­ted to the hos­pi­tal for fur­ther obser­va­
tion. Her grand­mother is tak­ing war­fa­rin for atrial fibril­la­tion, and
the patient would pre­fer to avoid war­fa­rin because she does not
think she can com­ply with the fre­quent mon­i­tor­ing that will be
required. She inquires about other oral anti­co­ag­ul­ant options.
Data supporting the use of DOACs in CVT are emerg­ing. High-
qual­ity evi­dence is still lim­ited, as there has been only 1 ran­dom­
ized con­trolled trial reported.22 The RE-SPECT CVT trial, published
in 2019, is a mul­ti­cen­ter, open-label study that ran­dom­ized 120
patients with CVT to either dabigatran 150 mg twice daily or war­
fa­rin with a goal INR of 2 to 3 for 24 weeks. Approximately 50%
of the patients in each group had iden­ti­fi­able pro­vok­ing fac­tors.
All patients received ini­tial treat­ment with par­en­teral hep­a­rin or
LMWH (5-15 days). The pri­mary end point of the study was new
venous throm­bo­em­bo­lism—new CVT, pul­mo­nary embolism, limb
DVT, SVT, or major bleed­ing.22 Major bleed­ing was defined accord-
ing to the International Society of Thrombosis and Haemostasis
(ISTH) cri­te­ria.23 Mean treat­ment dura­tion was 5.1 months in the
dabigatran arm and 5.3 months in the war­fa­rin group, with a mean
time in ther­a­peu­tic range of 66.1% for war­fa­rin-treated patients.22
There were no recur­rent venous throm­bo­em­bolic events in either
treat­ment group, and the inci­dence of major bleed­ing was low
in this study. Two patients treated with war­ fa­
rin expe­ri­
enced
major intra­cra­nial bleed­ing (3.3%; 95% CI, 0.4%-11.5%), 1 patient
had major GI bleed­ing (1.7%; 95% CI, 0.0%-8.9%), and another
patient expe­ri­enced clin­i­cally rel­e­vant non­ma­jor gen­i­to­uri­nary
bleed­ing.22 The sec­ond­ary end point of cere­bral vein recan­a­li­za­
tion occurred in 33 patients (60.0%; 95% CI, 45.9%-73.0%) in the
dabigatran group and 35 patients (67.3%; 95% CI, 52.9%-79.7%) in
the war­fa­rin group.18 An excel­lent func­tional out­come (mod­i­fied
Rankin scale 0-1 points) was seen in 91.5% and 91.4% in the war­fa­
rin and dabigatran groups, respec­tively.22 Based on the results of
the above trial, dabigatran and war­fa­rin appear to have a sim­i­lar
effi­cacy and safety pro­file in the treat­ment of patients with CVT
after brief, ini­tial ther­apy with unfractionated heparin or LMWH.
A sys­tem­atic review and meta-anal­y­sis of 6 stud­ies (5 obser­
va­tional and 1 ran­dom­ized) exam­in­ing the safety and effi­cacy
of DOACs vs vita­ min K antag­o­
nists (VKAs) was published in
2020.24 There were 412 patients in total, and 151 were treated
with DOACs and 261 with VKAs.20 Provoking fac­tors that were
reported in the stud­ies included oral con­tra­cep­tive use (24.2%,
n = 100/412) and preg­ nancy (6.3%, n  = 
26/412). DOAC effi­ cacy
was com­pa­ra­ble to VKA for cere­bral vein recan­a­li­za­tion (rel­a­tive
risk [RR], 1.02; 95% CI, 0.89-1.16) and excel­lent func­tional out­
comes as mea­sured by a mod­i­fied Rankin scale less than 2 (RR,
1.02; 95% CI, 0.93-1.13).24 There was a trend toward less major
bleed­ing in the DOAC treat­ment group, but this did not reach
sta­tis­ti­cal sig­nif­i­cance (RR, 0.44; 95% CI, 0.12-1.59).24 Another
sys­tem­atic review that included 33 stud­ies (1 ran­dom­ized con­
trolled trial, 5 obser­va­tional, 27 case series/stud­ies) reported
on patients with CVT who were treated with DOACs (n = 279) vs
VKAs (n = 315).25 The most com­mon DOAC was rivaroxaban (47%),
followed by dabigatran (41%), apixaban (10%), and edoxaban
(2%).25 Results showed a sim­i­lar risk of death in the DOAC and
VKA arms (RR, 2.12; 95% CI, 0.29-15.59).25 Rates of new intra­cra­
102  |  Hematology 2021  |  ASH Education Program
nial hem­or­rhage (0.7%) and recur­rent CVT (1.5%) were very low
in patients treated with DOACs.25 Excellent func­tional out­comes
were observed in 94% of patients in the DOAC arm with an RR of
1.13 (95% CI, 1.02-1.25).25
The larg­est pro­spec­tive obser­va­tional study was performed
by Wasay et al26 and published in the Journal of Stroke. This study
reported on patients from Saudi Arabia, Pakistan, Egypt, and
United Arab­ Emirates.26 They included 111 patients—45 on DOACs
and 66 on war­fa­rin. The most com­mon DOAC used was rivarox-
aban (n = 36), followed by dabigatran (n = 9). The war­fa­rin dose was
adjusted to main­tain an INR range of 2 to 3.26 Initiation of war­fa­
rin or a DOAC hap­pened at a median time of 1 week from the
time of CVT diag­no­sis, and the median dura­tion of fol­low-up was
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/100/1851729/100mathew.pdf by guest on 13 December 2021
8 months (range, 6-13 months).26 Most patients had pro­vok­ing
fac­tors, includ­ing preg­nancy/puer­pe­rium (23.1%), sys­temic infec­
tions (32.4%), drugs/oral con­tra­cep­tive pills (9.9%), and other
med­i­cal con­di­tions (6.3%).26 No recur­rent VTE events were seen
in either group. One patient in each group had major bleed­ing
as clas­si­fied by ISTH cri­te­ria. Four patients (6.1%) and 2 patients
(4.4%) in the war­fa­rin and DOAC group, respec­tively, had any
bleed­ing.26 Rusin et al27 pro­spec­tively recruited 36 patients with
CVT who were treated with DOACs, includ­ing dabigatran (n = 18),
rivaroxaban (n = 10), and apixaban (n = 8), for a median dura­tion of
8.5 months. Of the patients, 94.4% (n = 34) were noted to have
com­plete or par­tial recan­a­li­za­tion of the cere­bral veins.23 CVT
recur­rence was observed in 5.6% (n = 2) after anticoagulation was
discontinued.27 Major bleed­ing occurred in 8.3% (n = 3; 2 rivarox-
aban [men­or­rha­gia], 1 dabigatran [GI bleed­ing]), which was rel­a­
tively high com­pared with other stud­ies.27 DOACs as the upfront
treat­ment of acute phase CVT with­out hep­a­rin were exam­ined in
a pro­spec­tive study by Shankar Iyer et al.28 The study was small
(n = 20), and crit­ i­
cally ill patients were excluded.28 All patients
were treated with rivaroxaban only, with­out hep­a­rin or LMWH
ther­apy.28 Cerebral vein recan­a­li­za­tion was seen in all­patients
(60% com­plete and 40% par­tial), and excel­lent func­tional out­
come was noted in 95% (n = 19) of the patients.28 Patients were
treated for a median of 6 months with no major bleed­ing noted.28
At least 5 ret­ro­spec­tive obser­va­tional stud­ies were published
between 2014 and 2019.29-33 Sample sizes were small (range n = 6-
15 patients). Dabigatran was the most com­ monly pre­ scribed
DOAC, followed by rivaroxaban and apixaban, respec­ tively.
Treatment dura­tion ranged from 6 to 12 months. Recurrent VTE
was not observed in any of the patients reported in these ret­ro­
spec­tive stud­ies. Cerebral vein recan­a­li­za­tion rates ranged from
55.6% to 100%.29-33 Bleeding was min­i­mal, with no major bleed­
ing observed. Minor bleed­ing ranged from 0% to 28.6%.29-33
The afore­men­tioned ADAM VTE trial of apixaban in patients
with can­cer included only a sin­gle patient with CVT.18 In a ret­ro­
spec­tive MD Anderson study of 45 patients with can­cer and CVT,
33 were treated with anti­co­ag­u­lants (LMWH, 23 cases; war­fa­rin,
10 cases).34 The spe­cific recur­rent rates of not anticoagulated,
LMWH-treated, and war­fa­rin-treated patients were 25.0%, 13.0%,
and 10.0%, respec­tively, which did not reach sta­tis­ti­cal sig­nif­i­
cance. The inci­dence of intracranial hemorrhage after CVT diag­
no­sis was higher in the group receiv­ing anticoagulation (30.3%;
95% CI, 16.8%-47.1%) vs no anticoagulation (25.0%; 95% CI, 7.6%-
52.9%), with no sig­nif­i­cant sta­tis­ti­cal dif­fer­ence (P = 1.000). The
choice of anti­co­ag­u­lant was asso­ci­ated with bleed­ing events
(LMWH, 34.8%; war­fa­rin, 80.0%; odds ratio, 7.50; 95% CI, 1.05-54.3;
P = .048). No patients were treated with DOACs in this can­cer trial.
Table 2. Ongoing tri­als eval­u­at­ing DOACs in the treat­ment of throm­bo­sis in unusual sites

Site of Primary No. of

Title Identifier Study design Intervention throm­bo­sis out­come patients Sponsor
The effi­cacy and NCT03217448 Phase 3 Dabigatran vs CVT Recanalization 80 Capital Medical
safety of dabigatran (interventional), war­fa­rin for after 6 University,
etexilate com­pared ran­dom­ized, 6 months months Beijing, China
with war­fa­rin for par­al­lel assign­
the anticoagulation ment, open label,
treat­ment of cere­bral sin­gle (out­comes
venous throm­bo­sis: a asses­sor)
pilot study
Multicenter, pro­spec­ NCT03178864 Interventional, Rivaroxaban vs CVT Safety of 100 University of
tive ran­dom­ized open ran­dom­ized, par­ stan­dard of rivaroxaban Brit­ish Colum­

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/100/1851729/100mathew.pdf by guest on 13 December 2021

label, blinded end­point al­lel assign­ment, care bia, Canada
(PROBE) con­trolled trial sin­gle (out­comes
of early anticoagulation asses­sor)
with rivaroxaban ver­sus
stan­dard of care in
deter­min­ing safety at
365 days in symp­tom­
atic cere­bral venous
throm­bo­sis (SECRET)
Comparing treat­ment NCT03747081 Interventional, ran­ Rivaroxaban vs CVT Efficacy of 50 Isfahan
out­comes in CVT dom­ized, par­al­lel war­fa­rin rivaroxaban University
patients who were assign­ment vs war­fa­rin of Medical
treated with war­fa­ Sciences, Iran
rin and rivaroxaban in
Isfahan, Iran
Rivaroxaban com­pared to NCT04569279 Interventional, par­al­ Rivaroxaban vs CVT Change of 71 Damascus
war­fa­rin for treat­ment lel assign­ment, war­fa­rin 6 sinus venous University,
of cere­bral venous open label months throm­bo­sis Syria
throm­bo­sis: a ran­dom­ sever­ity scale
ized con­trolled trial
Comparison of the effi­ NCT03191305 Interventional, Rivaroxaban vs CVT Efficacy of 50 Foundation
cacy of rivaroxaban to nonrandomized, coumadin rivaroxaban University
coumadin (war­fa­rin) in par­al­lel assign­ and com­ Islamabad,
cere­bral venous throm­ ment par­i­son to Pakistan
bo­sis war­fa­rin
Treatment of por­tal, mes­ NCT02627053 Interventional, Rivaroxaban PVT, MVT, Safety and 100 Università
en­teric, and splenic sin­gle group SVT effi­cacy of degli Studi
vein throm­bo­sis with assign­ment, pro­ rivaroxaban dell’Insubria,
rivaroxaban. A pilot, spec­tive, cohort at 3 months Italy
pro­spec­tive cohort study
study
Direct oral anti­co­ag­u­lants NCT04660747 Observational, com­ 3:2 ratio CVT Composite 500 Academisch
for the treat­ment of par­a­tive cohort DOAC/VKA of major Medisch
cere­bral venous throm­ study, pro­spec­ bleed­ing and Centrum—
bo­sis: an inter­na­tional tive symp­tom­atic Universiteit
phase IV study recur­rent van
venous Amsterdam
throm­bo­ (AMC-
sis after 6 UvA), The
months Netherlands
International reg­is­try on NCT03778502 Observational, Any DOAC All sites Evaluate 100 University of
the use of the direct patient reg­is­ the use of Malta, Malta
oral anti­co­ag­u­lants try, pro­spec­tive, DOACs in
for the treat­ment of cohort study unusual-site
unusual site venous VTE and to
throm­bo­em­bo­lism assess safety
and effec­
tive­ness

Prakash Singh Shekhawat

Warfarin vs DOAC in SVT or CVT  |  103
The authors con­cluded that a nar­row ther­a­peu­tic index of antico-
agulation may exist in can­cer-asso­ci­ated CVT.34
Despite being off label, the DOACs are being pre­ scribed
in an increas­ing num­ber of patients with SVT and CVT. A 2020
cross-sec­tional sur­vey of mem­bers of 3 dif­fer­ent hemo­sta­sis and
throm­bo­sis soci­e­ties explored choices of anti­co­ag­u­lant ther­apy
in 4 dif­fer­ent clin­i­cal vignettes.35 Across the 4 vignettes, great
var­i­
abil­
ity existed, but VKAs were the most com­ mon choice
(44%-63%), followed by the DOACs (23%-27%) in low bleed­ing
risk patients.35 In high bleed­ing risk sce­nar­ios, par­en­teral agents
were the sec­ond most com­mon choice.35
Most of the stud­ies presented in this arti­cle have many lim­i­
ta­tions, mak­ing defin­i­tive rec­om­men­da­tions dif­fi­cult; few were
ran­dom­ized, sites of throm­bo­sis dif­fered, type of DOAC use dif­
fered, and length of ther­apy was het­ero­ge­neous. Future stud­ies
should be ran­dom­ized, include well-defined sites of throm­bo­sis
(por­tal, mes­en­teric, hepatic, or cere­bral), choose a sin­gle DOAC
to com­pare with war­fa­rin, and treat for a con­sis­tent length of
time. Multiple tri­als test­ing DOACs in clots in unusual places are
ongo­ing, as sum­ma­rized by Riva et al.36
Current soci­ety guide­lines/expert opin­ion
Riva and Ageno,9 in a state-of-the-art ISTH 2020 report, con­
clude that the cur­rent evi­dence sug­gests that DOACs can be
used in select patients with unusual-site throm­bo­sis given com­
pa­ra­ble effi­cacy and a trend toward lower bleed­ing com­pared
with VKAs. The authors stress that cau­tion should be taken in
high-risk pop­u­la­tions such as those with cir­rho­sis, varices, cen­
tral ner­vous sys­tem infec­tion, or trauma. In addi­tion, they note
that indi­vid­u­al­ized deci­sions need to be con­sid­ered in patients
with Child-Pugh class B to C liver dis­ease, mod­er­ate to severe
renal dys­func­tion, lumi­nal GI malig­nancy (SVT), and those tak­ing
anti­sei­zure med­i­ca­tion (CVT).9
In 2016, rec­om­men­da­tions by the European Association for
the Study of the Liver included LMWH followed by VKAs for non­
ma­lig­nant, noncirrhotic PVT.2
The Euro­ pean Society for Vascular Surgery recommended
LMWH or unfractionated heparin followed by either VKA or DOAC
in patients with MVT.36,37
The most recent guide­lines for CVT treat­ment from the Euro­
pean Stroke Organization and endorsed by the Euro­pean Acad-
emy of Neurology do not rec­om­mend DOAC use in the acute
phase of CVT treat­ment but sug­gest a pref­er­ence for LMWH.38
These guide­lines were published before the results of the RE-
SPECT CVT trial and most of the pro­spec­tive trial data men­tioned
above. With the new infor­ma­tion, upcom­ing guide­lines would
likely include use of DOACs in CVT as sec­ond­ary pro­phy­laxis.
Ongoing tri­als should allow fur­ther rec­om­men­da­tions in the
near future (Table 2).
Author rec­om­men­da­tions
SVT: Based on the avail­­able data, we rec­om­mend use of either
war­fa­rin or a DOAC in patients with SVT. In patients with Child-
Pugh B liver dis­ ease, rivaroxaban should be avoided, and in
Child-Pugh C liver dis­ease, all­of the DOACs are contraindicated.
We rec­om­mend treat­ment for inci­den­tally detected SVT, unless
it is con­firmed to be chronic com­pared with prior scans. In pa-
tients with advanced liver dis­ease in whom the DOACs are con-
traindicated and war­fa­rin would be dif­fi­cult, given the ele­vated
104  |  Hematology 2021  |  ASH Education Program
base­line pro­throm­bin time, we rec­om­mend use of LMWH and
con­sid­er­ation of dose reduc­tion based on the degree of throm­
bo­cy­to­pe­nia and coagulopathy. In SVT, we rec­om­mend treat­
ment of at least 3 to 6 months. Extended-dura­tion anticoagula-
tion should be con­sid­ered in patients with ongo­ing risk fac­tors
for throm­bo­sis.
CVT: Based on the avail­­ able data, we rec­ om­mend use of
either war­fa­rin or dabigatran for treat­ment of an ini­tial epi­sode
of CVT for a dura­tion of at least 3 to 6 months. Extended-dura­tion
anticoagulation should be con­sid­ered in patients with ongo­ing
risk fac­tors for throm­bo­sis. In patients who can­not receive war­
fa­rin or dabigatran, use of rivaroxaban and apixaban can be con­
sid­ered based on pro­spec­tive and ret­ro­spec­tive stud­ies.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/100/1851729/100mathew.pdf by guest on 13 December 2021
Conflict-of-inter­est dis­clo­sure
Carol Mathew: no rel­e­vant con­flicts to dis­close.
Marc Zumberg: mem­ber, ABIM Hematology Board, ASH speaker
reim­burse­ment, law case review.
Off-label drug use
Carol Mathew: DOAC use in splanchnic and cerebral vein throm-
bosis is discussed.
Marc Zumberg: DOAC use in splanchnic and cerebral vein throm-
bosis is discussed.
Correspondence
Marc Zumberg, University of Florida Health, 1600 S.W. Archer Rd,
Gainesville, FL 32610; e-mail: zumbems@med­i­cine​­.ufl​­.edu.
References
1. Di Nisio M, Valeriani E, Riva N, Schulman S, Beyer-Westendorf J, Ageno W.
Anticoagulant ther­apy for splanch­nic vein throm­bo­sis: ISTH SSC sub­com­
mit­tee con­ trol of anticoagulation. J Thromb Haemost. 2020;18(7):1562-
1568.
2. Euro­pean Association for the Study of the Liver. EASL clin­ i­cal prac­tice
guide­lines: vas­cu­lar dis­eases of the liver. J Hepatol. 2016;64(1):179-202.
3. Kearon C, Akl EA, Comerota AJ, et  al. Antithrombotic ther­apy for VTE
dis­ease: antithrombotic ther­apy and pre­ven­tion of throm­bo­sis, 9th ed:
Amer­i­can College of Chest Physicians Evidence-Based Clinical Practice
Guidelines. Chest. 2012;141(2, suppl):e419S-e496S.
4. Valeriani E, Di Nisio M, Riva N, et  al. Anticoagulant ther­apy for splanch­
nic vein throm­bo­sis: a sys­tem­atic review and meta-anal­y­sis. Blood.
2021;137(9):1233-1240.
5. Valeriani E, Di Nisio M, Riva N, et al. Anticoagulant treat­ment for splanch­nic
vein throm­bo­sis in liver cir­rho­sis: a sys­tem­atic review and meta-anal­y­sis.
Thromb Haemost. 2021;121(7):867-876.
6. Ghazaleh S, Beran A, Aburayyan K, et al. Efficacy and safety of anticoagula-
tion in non-malig­nant por­tal vein throm­bo­sis in patients with liver cir­rho­sis:
a sys­tem­atic review and meta-anal­y­sis. Ann Gastroenterol. 2021;34(1):104-
110.
7. de Bruijn SF, Stam J. Randomized, pla­cebo-con­trolled trial of anti­co­ag­ul­ant
treat­ment with low-molec­u­lar-weight hep­a­rin for cere­bral sinus throm­bo­
sis. Stroke. 1999;30(3):484-488.
8. Ferro JM, Canhão P, Stam J, Bousser M-G, Barinagarrementeria F; ISCVT
Investigators. Prognosis of cere­bral vein and dural sinus throm­bo­sis: results
of the International Study on Cerebral Vein and Dural Sinus Thrombosis
(ISCVT). Stroke. 2004;35(3):664-670.
9. Riva N, Ageno W. Direct oral anti­co­ag­ul­ants for unusual-site venous throm­
bo­em­bo­lism. Res Pract Thromb Haemost. 2021;5(2):265-277.
10. Kearon C, Akl EA, Ornelas J. et al. Antithrombotic ther­apy for VTE dis­ease:
CHEST guide­line and expert panel report. Chest. 2016;149(2):315-352.
11. Ortel TL, Neumann I, Ageno W, et  al. Amer­i­can Society of Hematology
2020 guide­lines for man­age­ment of venous throm­bo­em­bo­lism: treat­
ment of deep vein throm­ bo­sis and pul­ mo­
nary embolism. Blood Adv.
2020;4(19):4693-4738.
12. Hanafy AS, Abd-Elsalam S, Dawoud MM. Randomized con­trolled trial of
rivaroxaban ver­sus war­fa­rin in the man­age­ment of acute non-neo­plas­tic
por­tal vein throm­bo­sis. Vascul Pharmacol. 2019;113:86-91.
13. Serrao A, Merli M, Lucani B, et  al. Outcomes of long-term anti­co­ag­u­lant
treat­ment for the sec­ond­ary pro­phy­laxis of splanch­nic venous throm­bo­sis.
Eur J Clin Invest. 2021;51(1):e13356.
14. Naymagon L, Tremblay D, Zubizarreta N, et al. The effi­cacy and safety of
direct oral anti­co­ag­u­lants in noncirrhotic por­tal vein throm­bo­sis. Blood
Adv. 2020;4(4):655-666.
15. Kawata E, Siew D-A, Payne J-G, Louzada M, Kovacs M-J, Lazo-Langner A.
Splanchnic vein throm­bo­sis: clin­i­cal man­i­fes­ta­tions, risk fac­tors, man­age­
ment, and out­comes. Thromb Res. 2021;202:90-95.
16. Sharma S, Kumar R, Rout G, Gamanagatti SR, Shalimar. Dabigatran as an
oral anti­co­ag­u­lant in patients with Budd-Chiari syn­drome post-per­cu­ta­ne­
ous endovascular inter­ven­tion. J Gastroenterol Hepatol. 2020;35(4):654-
662.
17. Salim S, Ekberg O, Elf J, Zarrouk M, Gottsäter A, Acosta S. Evaluation of
direct oral anti­co­ag­u­lants and vita­min K antag­o­nists in mes­en­teric venous
throm­bo­sis. Phlebology. 2019;34(3):171-178.
18. McBan RD II, Wysokinski WE, Le-Rademacher JG, et  al. Apixaban and
dalteparin in active malig­nancy-asso­ci­ated venous throm­bo­em­bo­lism: the
ADAM VTE trial. J Thromb Haemost. 2020;18(2):411-421.
19. Raskob GE, van Es N, Verhamme P, et al; Hokusai VTE can­cer inves­ti­ga­tors.
Edoxaban for the treat­ment of can­cer-asso­ci­ated venous throm­bo­em­bo­
lism. N Engl J Med. 2018;378(7):615-624.
20. Young AM, Marshall A, Thirlwall J, et  al. Comparison of an oral fac­tor Xa
inhib­i­tor with low molec­u­lar weight hep­a­rin in patients with can­cer with
venous throm­bo­em­bo­lism: results of a ran­dom­ized trial (SELECT-D). J Clin
Oncol. 2018;36(20):2017-2023.
21. Agnelli G, Becattini C, Meyer G, et al; Caravaggio Investigators. Apixaban
for the treat­ment of venous throm­bo­em­bo­lism asso­ci­ated with can­cer. N
Engl J Med. 2020;382(17):1599-1607.
22. Ferro JM, Coutinho JM, Dentali F, et al; RE-SPECT CVT Study Group. Safety
and effi­cacy of dabigatran etexilate vs dose-adjusted war­fa­rin in patients
with cere­bral venous throm­bo­sis: a ran­dom­ized clin­i­cal trial. JAMA Neurol.
2019;76(12):1457-1465.
23. Schulman S, Kearon C; Subcommittee on Control of Anticoagulation of the
Scientific and Standardization Committee of the International Society on
Thrombosis and Haemostasis. Definition of major bleed­ing in clin­i­cal inves­
ti­ga­tions of antihemostatic medic­i­nal prod­ucts in non-sur­gi­cal patients. J
Thromb Haemost. 2005;3(4):692-694.
24. Lee GKH, Chen VH, Tan C-H, et al. Comparing the effi­cacy and safety of
direct oral anti­co­ag­u­lants with vita­min K antag­o­nist in cere­bral venous
throm­bo­sis. J Thromb Thrombolysis. 2020;50(3):724-731.
25. Bose G, Graveline J, Yogendrakumar V, et al. Direct oral anti­co­ag­u­lants in
treat­ment of cere­bral venous throm­bo­sis: a sys­tem­atic review. BMJ Open.
2021;11(2):e040212.
Prakash Singh Shekhawat
26. Wasay M, Khan M, Raj­put HM, et al. New oral anti­co­ag­u­lants ver­sus war­
fa­rin for cere­bral venous throm­bo­sis: a multi-cen­ter, obser­va­tional study. J
Stroke. 2019;21(2):220-223.
27. Rusin G, Wypasek E, Papuga-Szela E, Żuk J, Undas A. Direct oral anti­co­ag­u­
lants in the treat­ment of cere­bral venous sinus throm­bo­sis: a sin­gle insti­tu­
tion’s expe­ri­ence. Neurol Neurochir Pol. 2019;53(5):384-387.
28. Shankar Iyer R, Tcr R, Akhtar S, Muthukalathi K, Kumar P, Muthukumar K.
Is it safe to treat cere­bral venous throm­bo­sis with oral rivaroxaban with­
out hep­a­rin? A pre­lim­i­nary study from 20 patients. Clin Neurol Neurosurg.
2018;175:108-111.
29. Geisbüsch C, Richter D, Herweh C, Ringleb PA, Nagel S. Novel fac­tor Xa
inhib­i­tor for the treat­ment of cere­bral venous and sinus throm­bo­sis: first
expe­ri­ence in 7 patients. Stroke. 2014;45(8):2469-2471.
30. Mendonça MD, Barbosa R, Cruz-e-Silva V, Calado S, Viana-Baptista M. Oral
direct throm­bin inhib­i­tor as an alter­na­tive in the man­age­ment of cere­bral
venous throm­bo­sis: a series of 15 patients. Int J Stroke. 2015;10(7):1115-1118.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/100/1851729/100mathew.pdf by guest on 13 December 2021
31. Anticoli S, Pezzella FR, Scifoni G, Ferrari C, Pozzessere C. Treatment of cere­
bral venous throm­bo­sis with rivaroxaban. J Biomed Sci. 2016;5(3):3.
32. Herweh C, Griebe M, Geisbüsch C, et  al. Frequency and tem­po­ral pro­
file of recan­a­li­za­tion after cere­bral vein and sinus throm­bo­sis. Eur J Neurol.
2016;23(4):681-687.
33. Covut F, Kewan T, Perez O, Flores M, Haddad A, Daw H. Apixaban and
rivaroxaban in patients with cere­ bral venous throm­ bo­sis. Thromb Res.
2019;173:77-78.
34. Abelhad NI, Qiao W, Garg N, Rojas-Hernandez CM. Thrombosis and bleed­
ing out­comes in the treat­ment of cere­bral venous throm­bo­sis in can­cer.
Thromb J. 2021;19(1):49.
35. Riva N, Carrier M, Gatt A, Ageno W. Anticoagulation in splanch­nic and cere­
bral vein throm­bo­sis: an inter­na­tional vignette-based sur­vey. Res Pract
Thromb Haemost. 2020;4(7):1192-1202.
36. Riva N, Ageno W. Cerebral and splanch­nic vein throm­bo­sis: advances,
chal­lenges, and unanswered ques­tions. J Clin Med. 2020;9(3):743.
37. Björck M, Koelemay M, Acosta S, et al. Editor’s choice—man­age­ment of the
dis­eases of mes­en­teric arteries and veins: clin­i­cal Practice Guidelines of the
Euro­pean Society of Vascular Surgery (ESVS). Eur J Vasc Endovasc Surg.
2017;53(4):460-510.
38. Ferro JM, Bousser M-G, Canhão P, et al; Euro­pean Stroke Organization.
Euro­pean Stroke Organization guide­line for the diag­no­sis and treat­ment
of cere­bral venous throm­bo­sis—endorsed by the Euro­pean Academy of
Neurology. Eur J Neurol. 2017;24(10):1203-1213.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000319
Warfarin vs DOAC in SVT or CVT  |  105
 CML: SUCCESS BREEDS MORE SUCCESS

TKI discontinuation in CML: how do we make

more patients eligible? How do we increase the

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/106/1851817/106hochhaus.pdf by guest on 13 December 2021

chances of a successful treatment-free remission?
Andreas Hochhaus and Thomas Ernst
Klinik für Innere Medizin II, Universitätsklinikum Jena, Jena, Germany

Treatment-free remission (TFR) is a new and significant goal of chronic myeloid leukemia management. TFR should be
considered for patients in stable deep molecular response (DMR) after careful discussion in the shared decision-making
process. Second-generation tyrosine kinase inhibitors (TKIs) improve the speed of response and the incidence of DMR.
Treatment may be changed to a more active TKI to improve the depth of response in selected patients who have not
reached DMR. Stem cell persistence is associated with active immune surveillance and activation of BCR-ABL1-inde-
pendent pathways, eg, STAT3, JAK1/2, and BCL2. Ongoing studies aim to prove the efficacy of maintenance therapies
targeting these pathways after TKI discontinuation.

LEARNING OBJECTIVES
• Review the options for achieving deep molecular remission in CML patients, with the possibility of discontinuing
therapy and staying in long-term treatment-free remission, which is equivalent to an operational cure
• Discuss the biology of persisting leukemic stem cells as the origin of relapse
• Defne the mechanism of action of experimental therapies to eradicate or silence residual leukemia

Introduction harmonize recommendations for treatment discontinuation

A signifcant proportion of chronic myeloid leukemia (CML)
patients achieve a deep molecular response (DMR) with
tyrosine kinase inhibitor (TKI) therapy defned as residual
BCR-ABL1 transcript levels of at least MR4 on the interna-
tional scale (IS). The speed of response is faster, and the
incidence of DMR at specifc time points is higher under
treatment with second-generation inhibitors (2G-TKIs)
than with imatinib.1
An attempt to discontinue therapy can be considered
after the achievement and long-term maintenance of DMR.
The concept was established with the Stop Imatinib 1 (STIM1)
trial. After a median follow-up of 77 months after stopping
TKI therapy, 38% of patients maintained a molecular remis-
sion. In this trial, patients were eligible for treatment dis-
continuation with negative results of a sensitive quantitative
reverse transcriptase polymerase chain reaction (qRT-PCR)
sustained for 2 years before stopping.2,3 After this innova-
tive pilot study, multiple trials were conducted, each with
different entry criteria and defnitions of relapse, prompt-
ing a restart of TKIs. Concurrently, PCR methodology and
defnitions of PCR sensitivity were standardized in order to
and increase the safety of stopping procedures.4,5
As a multicenter academic approach, the European
Stop Kinase Inhibitor (EURO-SKI, NCT01596114) study ana-
lyzed 755 patients with similar overall results. The minimum
requirements for stopping in most studies were a duration
of TKI therapy of at least 3 years and a sustained DMR of at
least 1 year.6
In general, about 50% of patients will relapse, regardless
of the TKI used. Most molecular recurrences occur within
the frst 6 months after TKI discontinuation. However, treat-
ment discontinuation is safe if eligible patients are care-
fully selected, and high-quality standardized qRT-PCR is
employed. The available data and procedures required
should be discussed with the patient. Some patients eli-
gible for treatment-free remission (TFR) may prefer to
remain on their current treatment. A loss of major molecular
response (MMR; BCR-ABL1>0.1% IS) has been established as
the trigger for restarting therapy.5,7 However, considering
that the rate of DMR in a newly diagnosed patient on cur-
rent treatment options is about 40% and the chance for sta-
ble TFR is about 50%, the overall chance of long-term TFR is

106 | Hematology 2021 | ASH Education Program

only 20% for a CML patient newly diag­nosed in 2021. In addi­tion,
the cost of TKI ther­apy is a sig­nif­i­cant prob­lem for patients and
the health sys­tem in gen­eral. Hence, it is cru­cial to increase the
rate of dura­ble DMR and to con­vert DMR into func­tional cures,
defined as nonrecurrence of active CML after stop­ping TKIs.8
CLINICAL CASE
Our patient is a 27-year-old woman, a teacher, diag­nosed with
chronic-phase CML in May 2017. Calculation of the EUTOS long-
term sur­vival score revealed low-risk dis­ease. Cytogenetics dem­
on­strated the involve­ ment of 3 chro­ mo­somes, 46,XX,t(1;9;22)
(q25;q34;q11), but mul­ti­plex PCR revealed a typ­i­cal e14a2 BCR-
ABL1 tran­script. Using next-gen­er­a­tion sequenc­ing, no BCR-ABL1-
inde­pen­dent muta­tions were detected.Spleen size was nor­mal,
and no med­i­ca­tion or con­com­i­tant dis­eases were reported. There
was a strong desire to estab­lish a fam­ily in the com­ing years. We
thor­oughly discussed with the patient the diag­no­sis, treat­ment
options, goals of ther­apy, and risks and ben­e­fits of each treat­
ment option. Shared deci­sion-mak­ing led to the con­clu­sion to
start nilotinib 300 mg twice daily in June 2017. Treatment was well
tol­er­ated; the only side effect was a grade 1 facial exan­thema.
Blood counts nor­mal­ized within 4 weeks. BCR-ABL1 tran­script lev­
els in periph­eral blood were mea­sured in a spe­cial­ized lab stan­
dard­ized for sen­si­tive quan­ti­fi­ca­tion according to the IS:
Sep­tem­ber 2017: 3.2%
Decem­ber 2017: 0.41%
March 2018: 0.23%
June 2018: 0.072%
Decem­ber 2018: 0.015%
June 2019: 0.0030%
Sep­tem­ber 2019: 0.0015%
Decem­ber 2019: 0.0023%
March 2020: 0.0010%
June 2020: 0% with 96 000 ABL1 tran­scripts (MR4.5).
At this stage, 3 years after the start of ther­apy, the patient
requested that ther­apy be discontinued.
Clinical expe­ri­ence
The pro­spec­tive, nonrandomized EURO-SKI was designed to de-
fine opti­mal con­di­tions for treat­ment dis­con­tin­u­a­tion. Sixty-one
sites in 11 Euro­ pean countries par­ tic­i­
pated. Eligibility cri­te­
ria
included chronic-phase CML, TKI treat­ment for at least 3 years
(with­out treat­ment fail­ure according to Euro­pean LeukemiaNet
rec­om­men­da­tions), and con­firmed MR4 or bet­ter for at least 1
year. The pri­mary end point was molec­u­lar relapse-free sur­vival,
defined as sur­vival with­out a loss of MMR. Secondary end points
were the anal­y­sis of prog­nos­tic fac­tors affect­ing MMR main­te­
nance and the eco­nomic impact of stop­ping ther­apy. Molecu-
lar relapse-free sur­vival was 61% and 50% at 6 and 24 months,
respec­tively. In the prog­nos­tic anal­y­sis of patients who received
first-line imatinib ther­apy, lon­ger treat­ment dura­tion and lon­ger
DMR dura­tion were advan­ta­geous for main­te­nance of MMR over
a period of six months. Treatment dis­con­tin­u­a­tion did not result
in seri­ous adverse events. For study par­tic­i­pants, the over­all cost
sav­ings were esti­mated at about €22 mil­lion.6
Prakash Singh Shekhawat
Confirmation of loss of MMR with a sec­ond PCR is not recom-
mended and could delay restarting ther­apy. Almost all­patients
with molec­u­lar recur­rence regain their ini­tial MR level after a TKI
restart. The choice of TKI for the restart depends on the pre­vi­ous
tol­er­a­bil­ity, comorbidities, risk of long-term tox­ic­ity, and expec­ta­
tions toward a sec­ond TFR attempt. So far, very few of the thou­
sands of patients in TFR tri­als have had unfa­vor­able out­comes.
There are occa­sional reports of pro­gres­sion to blast phase, but
it is uncer­tain whether TKI dis­con­tin­u­a­tion con­trib­uted to the
under­ly­ing biol­ogy of pro­gres­sion. Irrespective of higher eli­gi­bil­ity
rates for TFR stud­ies with sec­ond-gen­er­a­tion TKI, the prob­a­bil­ity
of maintaining TFR has been about 50% after nilotinib or dasatinib
ther­a­pies, sim­i­lar to the results after stop­ping imatinib.9-13
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/106/1851817/106hochhaus.pdf by guest on 13 December 2021
The French RE-STIM trial inves­ti­gated the out­come of sec­ond
or sub­se­quent TFR attempts. Seventy patients who regained
DMR after molec­u­lar recur­rence after the first TKI dis­con­tin­u­a­tion
stopped treat­ment for a sec­ond time. The pro­por­tion of patients
with­out molec­u­lar recur­rence was 35% at 36 months after the
sec­ond TKI dis­con­tin­u­a­tion. RE-STIM data actu­ally sug­gest that
patients in TFR los­ing MMR have a lower chance for a suc­cess­
ful sec­ond TFR attempt than patients with molec­u­lar recur­rence
according to the orig­i­nal STIM def­i­ni­tion (pos­i­tive PCR).14
Patients’ per­spec­tive
In gen­eral, rec­om­men­da­tions for safe dis­con­tin­u­a­tion and treat­
ment restart are well established and published. However, the
expe­ri­ence of patients with regard to their deci­sion about TFR
and their needs dur­ing TFR are widely unknown. Using a sur­vey,
inter­na­tional patient advo­cates sought to check the patient per­
spec­tive, to iden­tify areas of unmet needs, and to cre­ate rec­om­
men­da­tions for improve­ments. Fear or anx­i­ety dur­ing treat­ment
dis­con­tin­u­a­tion were reported by 56% of patients, but only 7%
of patients were asked about their needs for psy­cho­log­i­cal sup­
port. After treat­ment restart, 59% of patients felt scared or anx­
ious, and 56% felt depressed. Twenty-six per­cent of reinitiated
patients received psy­cho­log­i­cal and/or emo­tional sup­port. Six-
ty per­cent of patients expe­ri­enced with­drawal symp­toms after
treat­ment dis­con­tin­u­a­tion. However, 40% of patients with with­
drawal symp­toms reported a lack of sup­port from their phy­si­cian
in symp­tom man­age­ment. Hence, the mon­i­tor­ing of psy­cho­log­
i­cal con­se­quences before, dur­ing, and after TFR should be con­
sid­ered.15
Musculoskeletal and/or joint pain starting a few weeks or
months after TKI dis­con­tin­u­a­tion has been observed in about
25% of patients. This TKI with­drawal syn­drome is likely based on
a release of off-tar­get effects of the TKI. In most cases symp­toms
are mild and tran­sient, but some patients may require tem­po­rary
anti-inflam­ma­tory treat­ment.16 Surveillance and symp­ tom­atic
treat­ment of with­drawal symp­toms should be a pri­or­ity dur­ing
treat­ment dis­con­tin­u­a­tion.15
Predictors of suc­cess
Various prog­nos­tic fac­tors for a suc­cess­ful TFR have been re-
ported. Longer dura­tion of TKI ther­apy and DMR and prior treat­
ment with inter­feron alpha (IFNA) were iden­ti­fied as impor­tant
pre­dic­tors for suc­cess­ful TFR in EURO-SKI; of these, dura­tion of
DMR was shown to be the most impor­tant fac­tor, cor­rob­o­rat­ing
data from a study in Japan.6,17 A Cana­dian TFR study revealed
6 years of imatinib ther­apy and 4.5 years MR4 dura­tion as opti­
mal time frames for a suc­cess­ful TFR.18
TKI dis­con­tin­u­a­tion in CML  |  107
Table 1. Predictive data of eli­gi­bil­ity and sta­bil­ity of TFR19,39

Parameter Impact on eli­gi­bil­ity for TFR Impact on sta­bil­ity of TFR Study exam­ples
Prognostic score DMR more fre­quent in low-risk TFR more sta­ble in low-risk STIM, STIM2, ENESTfreedom,
patients patients; risk of relapse high in TWISTER
high-risk patients
TKI type DMR more fre­quent with 2G-TKIs No major dif­fer­ence
vs imatinib
Initial slope of BCR-ABL1 Probability of DMR bet­ter in No major dif­fer­ence
tran­script decline or patients with EMR
“halv­ing time”
MMR at 12 months Probability of DMR bet­ter in No major dif­fer­ence
patients with MMR

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/106/1851817/106hochhaus.pdf by guest on 13 December 2021

Type of BCR-ABL1 tran­script
Mutations out­side BCR-ABL1
Duration of DMR
Level of MR before stop
Duration of TKI ther­apy
Proportion of nat­u­ral killer cells
EMR, early molec­u­lar response.
DMR more likely in patients with
e14a2 vs e13a2 (incon­sis­tent)
MMR and DMR more likely in
patients lacking BCR-ABL1-
inde­pen­dent muta­tions
TFR more sta­ble with e14a2
(incon­sis­tent)
Data still imma­ture
Longer DMR pre­dic­tive EURO-SKI
Probably yes; unde­tect­able EURO-SKI, ENESTfreedom
BCR-ABL1 pre­dic­tive
Inconsistent, overlapping with STIM
dura­tion of DMR
Reduced risk of relapse after
imatinib and dasatinib but not
after nilotinib

Age and sex have been reported to be asso­ci­ated with
TFR in early stud­ies.6 The Sokal score was reported to pre­dict
TFR out­come, eg, in the STIM study.2 However, the EURO-SKI
study failed to dem­on­strate the pre­dic­tive value of age, sex,
and Sokal score for TFR in patients after first-line imatinib. A
selec­tion bias of patients on IFNA ther­apy prior to imatinib
might play a role. Data published recently indi­cate that the
BCR-ABL1 tran­script type might influ­ence TFR out­come, with
a lower inci­dence of relapse in patients and e14a2 com­pared
to e13a2.19
So far, the dura­tion of first-line nilotinib or dasatinib ther­apy
has not been found to affect the prob­a­bil­ity of TFR suc­cess. The
dura­tion of DMR on sec­ond-line nilotinib was asso­ci­ated with
suc­cess­ful TFR in ENESTop (Table 1).9,10,19
Differences in trial designs must be con­sid­ered when results
from var­i­ous stud­ies are com­pared: the pro­por­tion of suc­cess­ful
TFR attempts is higher in stud­ies that define dis­ease relapse as
loss of MMR than in those that spec­ify loss of unde­tect­able BCR-
ABL1 tran­scripts as the cri­te­rion. In addi­tion, a deeper MR prior
to treat­ment dis­con­tin­u­a­tion may be asso­ci­ated with a bet­ter
chance of suc­cess­ful TFR.17
During fol­low-up, qRT-PCR revealed an increase of BCR-ABL1
tran­scripts:
Sep­tem­ber 2020: 0.06% (IS)
Decem­ber 2020: 0.15%.
Due to loss of MMR, treat­ment reintroduction was discussed.
Good response and good tol­er­a­bil­ity of nilotinib was bal­anced
against a sec­ond chance with an alter­na­tive TKI. The deci­sion
was made to com­mence dasatinib 100 mg/d, lead­ing to a rapid
decline of BCR-ABL1 tran­scripts:
March 2021: 0.0020%.
June 2021: 0% with 290 000 ABL1 tran­scripts (MR5).
A bone mar­row aspi­ra­tion was performed to quan­tify per-
sisting stem cells on a geno­mic level and to char­ac­ter­ize their
envi­ron­ment. The opti­mal dura­tion of TKI treat­ment for the sec­
ond attempt of TFR is unknown. The inhi­bi­tion of BCR-ABL1-inde­
pen­dent path­ways respon­si­ble for stem cell per­sis­tence and for
acti­va­tion of immune sur­veil­lance is a lead­ing topic in cur­rent
research.

Analyzing per­sis­tent stem cells

In sev­
eral stud­ ies, highly sen­si­
tive DNA-based BCR-ABL1 PCR
dem­on­strated per­sis­tence of the CML clone in most patients
CLINICAL CASE (Con­t in­u ed) with neg­a­tive qRT-PCR, indi­cat­ing low BCR-ABL1 expres­sion in
In June 2020, 3 years after starting ther­apy and 1 year in MR4.5, persisting CML stem cells. The detec­tion of the patient-spe­cific
nilotinib ther­apy was discontinued. Grade 2 mus­cle and joint geno­mic BCR-ABL1 break point requires a DNA sam­ple col­lected
pain, respon­sive to acet­amin­op
­ hen and last­ing about 4 weeks, at diag­no­sis.20-22 The anal­y­
sis of bone mar­ row vs periph­
eral
were signs of a mild TKI with­drawal syn­drome. blood for this pur­pose is ongo­ing.23

108  |  Hematology 2021  |  ASH Education Program

 In a Euro­pean study, DNA and mRNA BCR-ABL1 lev­els by quan­ Table 2. Clinical tri­als to main­tain TFR in addi­tion to TKI
ti­ta­tive PCR and by dig­i­tal PCR were com­pared. A good cor­re­
la­tion was found at BCR-ABL1 tran­script lev­els above MR4; the TIGER nilotinib ± peg-IFNA2b (phase 3) NCT01657604
cor­re­la­tion was poor for sam­ples taken in DMR. A com­bi­na­tion PETALs nilotinib ± peg-IFNα2a (phase 3) NCT02201459
of geno­mic and RT-PCR was ­able to bet­ter pre­dict molec­u­lar IFNA main­te­nance ther­apy
relapse-free sur­vival after TKI stop.24   TIGER peg-IFNA2b (phase 3) NCT01657604
Leukemic stem cells are CD34+/CD38−, but this phe­no­type   ENDURE pegylated-pro­line-IFNA2b (phase 3) NCT03117816
is not exclu­ sive to CML. Dipeptidyl pep­ ti­
dase-4 (CD26) may Checkpoint inhib­i­tors
rep­ re­sent a spe­ cific marker for CML stem cells.25 Using the   ACTIW pioglitazone (phase 1), avelumab (phase 2) NCT02767063
CD34+CD38−CD26+ phe­no­type, a study iden­ti­fied CML stem cells   Dasatinib nivolumab (phase 1B) NCT02011945
cir­cu­lat­ing in the periph­eral blood in the major­ity of patients in
JAK inhi­bi­tion
TFR after TKI dis­con­tin­u­a­tion,26 which is, how­ever, still conflict-   Preclinical data
ing. Another study iden­ti­fied persisting leu­ke­mic stem cells in   Nilotinib + ruxolitinib phase 1/2 NCT01914484

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/106/1851817/106hochhaus.pdf by guest on 13 December 2021

the bone mar­row of patients in DMR, in con­trast to the periph­   Nilotinib + ruxolitinib phase 1 NCT02253277
eral blood.23 Further, CD93 expres­ sion has been con­ sis­
tently   Nilotinib + ruxolitinib phase 1 NCT01702064
shown in a lin-CD34+CD38−CD90+ pop­u­la­tion with stem cell char­   ABL-TKI + ruxolitinib phase 2 NCT03610971
ac­ter­is­tics.27   ABL-TKI + ruxolitinib phase 2 NCT03654768
Using fluo­res­cence-acti­vated cell sorting and geno­mic PCR, BCL-2 inhi­bi­tion
resid­ual dis­ease in TFR dif­fered according to the cell lin­e­age.   Preclinical data
BCR-ABL1 was iden­ti­fied pre­dom­i­nantly in the lym­phoid com­   Dasatinib ± venetoclax phase 3 NCT02689440
part­ment and never in granulocytes. B cells were more often BTK inhi­bi­tion
BCR-ABL1+ than T cells. Lineage-spe­cific assess­ment of persist-   Preclinical data
ing dis­ease prior to treat­ment dis­con­tin­u­a­tion could improve the
pre­dic­tion of suc­cess­ful TFR.28

Innovative approaches BCL2

JAK, BTK
In TKI-resis­tant CML, STAT3 inhi­ bi­
tion was ­ able to reduce sur­
vival of CML cells. Kuepper et al iden­ti­fied Jak1 but not Jak2 as the
STAT3-acti­vat­ing kinase. The com­bi­na­tion of BCR-ABL1 and JAK1
inhib­i­tors fur­ther reduced col­ony-forming units from murine CML
bone mar­row and human CML mono­nu­clear cells as well as CD34+
CML cells. Combination ther­apy induced apo­pto­sis even in qui­es­
cent leu­ke­mic stem cells, suggesting JAK1 as a poten­tial ther­a­peu­
tic tar­get for cura­tive CML ther­a­pies.29 Ruxolitinib, a dual JAK1/2
inhib­i­tor, com­bined with nilotinib may selec­tively decrease CML
stem cells.25 A few early-phase stud­ies test­ing the com­bi­na­tion of
BCR-ABL1 targeting TKIs with ruxolitinib were pre­ma­turely closed
due to tox­ic­ity; oth­ers are still recruiting (Table 2). Fc gamma
recep­tor IIb (CD32b) was shown to be crit­i­cal in stem cell per­sis­
tence. Hence, targeting FcγRIIb down­stream sig­nal­ing, eg, with a
BTK inhib­i­tor, pro­vi­des a prom­is­ing ther­a­peu­tic approach.30
BCL2-related antiapoptotic pro­teins are expressed at higher lev­els
in CML stem cells com­pared to nor­mal stem cells. Inhibiting BCL2
increases kill­ing of CML stem cells by TKIs in mod­els. Venetoclax
com­bined with a TKI reduces the serial re-engraft­ment capac­ity
of CML stem cells in mice bet­ter than a TKI alone. Thus, BCL2 inhi­
bi­tion may be another option for targeting persisting CML stem
cells unre­spon­sive to BCR-ABL1 TKIs.31,32 A ran­dom­ized first-line
study test­ing the impact of the addi­tion of venetoclax to dasati-
nib in newly diag­nosed CML patients is ongo­ing (NCT02689440).
However, the impact of venetoclax on per­sis­tent stem cells should
be tested in a main­te­nance for­mat after achieve­ment of DMR.
Immunotherapy
Numerous stud­ies have revealed an asso­ci­a­tion between cel­
lu­lar immune param­e­ters and TFR, includ­ing higher num­bers of
­nat­u­ral killer cells and lower num­bers of both T reg­u­la­tory cells

Figure 1. Innovative strategies to improve chances

Prakash of treatment-free
Singh Shekhawat remission.

TKI dis­con­tin­u­a­tion in CML  |  109

Table 3. Recommendations for TFR

National Comprehensive Cancer French Chronic Myeloid Leukemia

Parameter Euro­pean LeukemiaNet 20205
Mandatory
  Age, phase of the dis­ease, Adult patients, CML in first
dis­ease his­tory chronic phase only. No prior
treat­ment fail­ure.
Motivation, com­mu­ni­ca­tion, Motivated patient with struc­tured
infor­ma­tion com­mu­ni­ca­tion.
Patient’s accep­tance of more fre­
quent PCR tests after stop­ping
Network 202138
Age ≥18 years. Chronic-phase CML.
No prior his­tory of accel­er­ated
or blast-phase CML.
Consultation with a CML spe­cial­
ist to review eli­gi­bil­ity for TKI
dis­con­tin­u­a­tion and poten­tial
risks and ben­e­fits of treat­
Study Group 201839
Age ≥18 years. Chronic-phase CML.
No prior his­tory of allo­ge­neic
stem cell trans­plan­ta­tion, pro­
gres­sion, resis­tance, sub­op­ti­mal
response, or warn­ing.
Adherence to mon­i­tor­ing and
retreatment rec­om­men­da­tions.
Written infor­ma­tion and instruc­
tions may be pro­vided.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/106/1851817/106hochhaus.pdf by guest on 13 December 2021

  Molecular-mon­i­tor­ing infra­
struc­ture
  Molecular-mon­i­tor­ing fre­
quency
Minimal (stop allowed)
  Line of treat­ment
  BCR-ABL1 tran­script type
  Duration of TKI ther­apy
  Duration of DMR
Optimal (stop recommended for con­sid­er­ation)
treat­ment.
Access to high-qual­ity qRT-PCR
using the IS with rapid turn­
around of PCR test results
Monthly molec­u­lar mon­i­tor­ing for the first 6 months, bimonthly dur­ing
months 7-12, and quar­terly there­af­ter (indef­i­nitely)
First-line or sec­ond-line ther­apy
if intol­er­ance was the only rea­
son for chang­ing TKI
Typical e13a2 or e14a2 BCR-ABL1
tran­scripts
Duration of TKI ther­apy >5 years
(>4 years for sec­ond-gen­er­a­tion
TKI)
DMR (MR4 or bet­ter) for >2 years
Duration of TKI ther­apy >5 years
Duration of DMR >3 years if MR4
Duration of DMR >2 years if MR4.5
ment dis­con­tin­u­a­tion, includ­
ing TKI with­drawal syn­drome.
Discontinuation of TKI ther­apy
should only be performed in
consenting patients after a thor­
ough dis­cus­sion of the poten­tial
risks and ben­e­fits.
Access to a reli­able quan­ti­ta­tive
PCR test with a sen­si­tiv­ity of
detec­tion of at least MR4.5
(BCR-ABL1 ≤ 0.0032% IS) that
pro­vi­des results within 2 weeks.
Prior evi­dence of a quan­ti­fi­able
BCR-ABL1 tran­script
On approved TKI ther­apy for at
least 3 years
Stable MR (MR4; BCR-ABL1 ≤ 0.01%
IS) for ≥2 years, as documented
on at least 4 tests, performed at
least 3 mo apart.
Results should be avail­­able within
2-3 weeks. Molecular biol­o­gists
should be aware of TKI dis­con­
tin­u­a­tions in indi­vid­ual patients.
Monthly dur­ing the first half year,
every 2 months. In the sec­ond
half year, quar­terly dur­ing the
sec­ond year and then every 3 to
6 months.
BCR-ABL1 tran­scripts e13a2, e14a2,
or e13a2/e14a2
TKI treat­ment dura­tion ≥5 years.
DMR at least MR4.5.
DMR dura­tion ≥2 years.
MR4.5 is con­firmed in ≥4 con­sec­
u­tive tests. At most 1 of the 4
assess­ments show­ing an MR4 is
accepted.

and CD86+ plasmacytoid den­dritic cells. Cumulative results sug­
gest an immu­no­log­i­cal con­trol of persisting CML stem cells in
patients with suc­cess­ful TFR.19,33
According to pre­clin­i­cal and clin­i­cal stud­ies, IFNA was ­able
to induce cyto­ge­netic remis­sions in a minor­ity of CML patients.
Moreover, a small per­cent­age of patients treated with IFNA sus-
tained dura­ble remis­sions after discontinuing ther­apy. The mech­
a­nisms behind this clin­ic­ al obser­va­tion are not well under­stood;
acti­va­tion of leu­ke­mia-spe­cific immu­nity may be a key fac­tor.34
Treatment with IFNA per­mits the dis­con­tin­u­a­tion of imatinib in
most patients after prior imatinib/IFNA com­bi­na­tion ther­apy and
may result in improved MR. Induction of a pro­tein­ase 3-spe­cific
cyto­toxic T-lym­pho­cyte response by IFNA may con­trib­ute to this
effect.35 Studies inves­ti­gat­ing the com­bi­na­tion of IFNA with nilo-
tinib or dasatinib are ongo­ing.
In a phase 1 study, the com­bi­na­tion of imatinib and ropegin-
terferon α2b was shown to be safe and resulted in a DMR in
patients with chronic-phase CML who did not achieve a DMR with
imatinib alone.36 Clinical tri­als explor­ing the impact of check­
point inhi­bi­tion include nivolumab as a programmed cell death

110  |  Hematology 2021  |  ASH Education Program

1 pro­tein inhib­i­tor in com­bi­na­tion with dasatinib (NCT02011945)
and avelumab as a programmed cell death 1 ligand 1 inhib­i­
tor com­bined with TKIs (ACTIW trial, NCT02767063; Table 2,
Figure 1).37
In any attempt to enhance the sta­bil­ity of TFR, the antic­i­pated
effects should be bal­anced against the risk of addi­tional side
effects caused by the immune acti­va­tion of the block of BCR-
ABL1-inde­pen­dent path­ways.
Recommendations
The rec­om­men­da­tions of the Euro­pean LeukemiaNet panel, the
National Comprehensive Cancer Network, and the French CML
Study Group are sum­ma­rized in Table 3.5,38,39
Certain require­ments are cru­cial: the access to a reli­able, sen­si­
tive, stan­dard­ized qRT-PCR with rapid turn­around and a moti­vated
patient who accepts more fre­ quent PCR ana­ ly­
ses. In gen­
eral,
chronic-phase patients in first-line ther­apy or in sec­ond-line after
TKI intol­er­ance should be con­sid­ered. Data for patients after TKI
resis­tance are still lacking. Standardized qRT-PCR requires spec­i­fi­
ca­tion of the BCR-ABL1 tran­script type. For patients with atyp­i­cal
BCR-ABL1 tran­scripts, the term “indi­vid­ual molec­u­lar response”
has been coined. Such patients could start TFR if DMR can be reli­
ably quan­ti­fied for this spe­cific tran­script (Table 3).40
CLINICAL CASE (Con­t in­u ed)
Our patient decided to con­tinue dasatinib as a monotherapy.
The treat­ ment dura­ tion and the dura­ tion of DMR had been
rather short prior to the attempted treat­ment dis­con­tin­u­a­tion.
Current guide­lines rec­om­mend a treat­ment dura­tion of at least
4 years with a 2G-TKI and a DMR dura­tion of at least 2 years in
MR4.5 (Table 3).5 However, as a low-risk patient with a rapid
MR to the first-line treat­ment, she has a good chance of main-
taining DMR and thus discontinuing ther­apy in about 3 addi­
tional treat­ment years. The var­i­ant cyto­ge­netic trans­lo­ca­tion
does not have an impact on prog­no­sis. The pre­dic­tive impact
of muta­tions out­side BCR-ABL1 is the sub­ject of ongo­ing coop­
er­a­tive stud­ies.
In our clin­i­cal prac­tice, we dis­cuss the con­cept of TFR with
chronic-phase CML patients prior to starting treat­ment. The dis­
cus­sion of the goals of CML treat­ment and the selec­tion of first-
line ther­apy should include the infor­ma­tion that an opti­mal MR
min­i­mizes the risk of pro­gres­sion and max­i­mizes the oppor­tu­nity
for discontinuing ther­apy. The pros­pect of TFR can serve as an
addi­tional moti­va­tor to opti­mize adher­ence to TKI ther­apy.19
Conflict-of-inter­est dis­clo­sure
Andreas Hochhaus: con­sul­tancy: Novartis, Pfizer, BMS, Incyte; re-
search funding: Novartis, Pfizer, BMS, Incyte.
Thomas Ernst: hon­o­raria: Novartis, BMS, Incyte; research fund-
ing: Novartis.
Off-label drug use
Andreas Hochhaus: experimental design (study proposals): BCL-2,
BTK, JAK inhibitors are discussed.
Thomas Ernst: experimental design (study proposals): BCL-2,
BTK, JAK inhibitors are discussed.
Prakash Singh Shekhawat
Correspondence
Andreas Hochhaus, Klinik für Innere Medizin II, Universitätsklini-
kum Jena, Am Klinikum 1, 07740 Jena, Germany; e-mail: andreas​
­.hochhaus@med​­.uni-jena​­.de.
References
1. Kantarjian HM, Hughes TP, Larson RA, et al. Long-term out­comes with front­
line nilotinib ver­sus imatinib in newly diag­nosed chronic mye­loid leu­ke­mia
in chronic phase: ENESTnd 10-year anal­y­sis. Leukemia. 2021;35(2):440-453.
2. Mahon FX, Réa D, Guilhot J, et al; Intergroupe Français des Leucémies
Myéloïdes Chroniques. Discontinuation of imatinib in patients with chronic
mye­loid leu­kae­mia who have maintained com­plete molec­u­lar remis­sion for
at least 2 years: the pro­spec­tive, multicentre Stop Imatinib (STIM) trial.
Lancet Oncol. 2010;11(11):1029-1035.
3. Etienne G, Guilhot J, Rea D, et al. Long-term fol­low-up of the French Stop
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/106/1851817/106hochhaus.pdf by guest on 13 December 2021
Imatinib (STIM1) study in patients with chronic mye­loid leu­ke­mia. J Clin
Oncol. 2017;35(3):298-305.
4. Cross NC, White HE, Colomer D, et al. Laboratory rec­om­men­da­tions for
scor­ing deep molec­u­lar responses fol­low­ing treat­ment for chronic mye­
loid leu­ke­mia. Leukemia. 2015;29(5):999-1003.
5. Hochhaus A, Baccarani M, Silver RT, et al. Euro­pean LeukemiaNet 2020 rec­
om­men­da­tions for treating chronic mye­loid leu­ke­mia. Leukemia. 2020;34(4):
966-984.
6. Saussele S, Richter J, Guilhot J, et al; EURO-SKI Investigators. Discontinuation
of tyro­sine kinase inhib­i­tor ther­apy in chronic mye­loid leu­kae­mia (EURO-SKI):
a prespecified interim anal­y­sis of a pro­spec­tive, multicentre, non-randomised,
trial. Lancet Oncol. 2018;19(6):747-757.
7. Rousselot P, Charbonnier A, Cony-Makhoul P, et al. Loss of major molec­
u­lar response as a trig­ ger for restarting tyro­ sine kinase inhib­ i­
tor ther­
apy in patients with chronic-phase chronic mye­log­e­nous leu­ke­mia who
have stopped imatinib after dura­ble unde­tect­able dis­ease. J Clin Oncol.
2014;32(5):424-430.
8. Zhao H, Deininger MW. Declaration of Bcr-Abl1 inde­pen­dence. Leukemia.
2020;34(11):2827-2836.
9. Radich JP, Hochhaus A, Masszi T, et al. Treatment-free remis­sion fol­low­ing
front­line nilotinib in patients with chronic phase chronic mye­loid leu­ke­mia:
5-year update of the ENESTfreedom trial. Leukemia. 2021;35(5):1344-1355.
10. Hughes TP, Clementino NCD, Fominykh M, et al. Long-term treat­ment-free
remis­sion in patients with chronic mye­loid leu­ke­mia after sec­ond-line nilo-
tinib: ENESTop 5-year update. Leukemia. 2021;35(6):1631-1642.
11. Takahashi N, Nishiwaki K, Nakaseko C, et al. Treatment-free remis­sion after
two-year con­sol­i­da­tion ther­apy with nilotinib in patients with chronic mye­
loid leu­ke­mia: STAT2 trial in Japan. Haematologica. 2018;103(11):1835-1842.
12. Yamaguchi H, Takezako N, Ohashi K, et al. Treatment-free remis­sion after
first-line dasatinib treat­ment in patients with chronic mye­loid leu­ke­mia in
the chronic phase: the D-NewS Study of the Kanto CML Study Group. Int J
Hematol. 2020;111(3):401-408.
13. Imagawa J, Tanaka H, Okada M, et al; DADI Trial Group. Discontinuation of
dasatinib in patients with chronic mye­loid leu­kae­mia who have maintained
deep molec­u­lar response for lon­ger than 1 year (DADI trial): a multicentre
phase 2 trial. Lancet Haematol. 2015;2(12):e528-e535.
14. Legros L, Nicolini FE, Etienne G, et al; French Intergroup for Chronic Myeloid
Leukemias. Second tyro­sine kinase inhib­i­tor dis­con­tin­u­a­tion attempt in
patients with chronic mye­loid leu­ke­mia. Cancer. 2017;123(22):4403-4410.
15. Sharf G, Marin C, Bradley JA, et al. Treatment-free remis­sion in chronic mye­
loid leu­ke­mia: the patient per­spec­tive and areas of unmet needs. Leuke-
mia. 2020;34(8):2102-2112.
16. Richter J, Söderlund S, Lübking A, et  al. Musculoskeletal pain in patients
with chronic mye­loid leu­ke­mia after dis­con­tin­u­a­tion of imatinib: a tyro­
sine kinase inhib­i­tor with­drawal syn­drome? J Clin Oncol. 2014;32(25):2821-
2823.
17. Takahashi N, Tauchi T, Kitamura K, et al; Japan Adult Leukemia Study Group.
Deeper molec­u­lar response is a pre­dic­tive fac­tor for treat­ment-free remis­
sion after imatinib dis­con­tin­u­a­tion in patients with chronic phase chronic
mye­loid leu­ke­mia: the JALSGSTIM213 study. Int J Hematol. 2018;107(2):185-
193.
18. Kim DDH, Novitzky-Basso I, Kim TS, et al. Optimal dura­tion of imatinib treat­
ment/deep molec­u­lar response for treat­ment-free remis­sion after imatinib
dis­con­tin­u­a­tion from a Cana­dian tyro­sine kinase inhib­i­tor dis­con­tin­u­a­tion
trial. Br J Haematol. 2021;193(4):779-791.
19. Ross DM, Hughes TP. Treatment-free remis­sion in patients with chronic
mye­loid leu­kae­mia. Nat Rev Clin Oncol. 2020;17(8):493-503.
TKI dis­con­tin­u­a­tion in CML  |  111
20. Ross DM, Branford S, Seymour JF, et al. Safety and effi­cacy of imatinib ces­
sa­tion for CML patients with sta­ble unde­tect­able min­i­mal resid­ual dis­ease:
results from the TWISTER study. Blood. 2013;122(4):515-522.
21. Ross DM, Branford S, Seymour JF, et al. Patients with chronic mye­loid leu­ke­
mia who main­tain a com­plete molec­u­lar response after stop­ping imatinib
treat­ment have evi­dence of per­sis­tent leu­ke­mia by DNA PCR. Leukemia.
2010;24(10):1719-1724.
22. Ross DM, Pagani IS, Shanmuganathan N, et al; Australasian Leukaemia
and Lymphoma Group. Long-term treat­ ment-free remis­ sion of chronic
mye­loid leu­ke­mia with fall­ing lev­els of resid­ual leu­ke­mic cells. Leukemia.
2018;32(12):2572-2579.
23. Ilhan O, Narli Ozdemir Z, Dalva K, et  al. Leukemic stem cells shall be
searched in the bone mar­row before “tyro­sine kinase inhib­i­tor-dis­con­tin­
u­a­tion” in chronic mye­loid leu­ke­mia. Int J Lab Hematol. Published online
9 April 2021.
24. Machova Polakova K, Zizkova H, Zuna J, et al. Analysis of chronic mye­loid
leu­kae­mia dur­ing deep molec­u­lar response by geno­mic PCR: a traf­fic light
strat­i­fi­ca­tion model with impact on treat­ment-free remis­sion. Leukemia.
2020;34(8):2113-2124.
25. Houshmand M, Simonetti G, Circosta P, et  al. Chronic mye­loid leu­ke­mia
stem cells. Leukemia. 2019;33(7):1543-1556.
26. Bocchia M, Sicuranza A, Abruzzese E, et al. Residual periph­eral blood CD26+
leu­ke­mic stem cells in chronic mye­loid leu­ke­mia patients dur­ing TKI ther­apy
and dur­ing treat­ment-free remis­sion. Front Oncol. 2018;8(30 May):194.
27. Kinstrie R, Horne GA, Morrison H, et  al. CD93 is expressed on chronic
mye­loid leu­ke­mia stem cells and identifies a qui­es­cent pop­u­la­tion which
per­sists after tyro­sine kinase inhib­i­tor ther­apy. Leukemia. 2020;34(6):1613-
1625.
28. Pagani IS, Dang P, Saunders VA, et al. Lineage of mea­sur­able resid­ual dis­
ease in patients with chronic mye­loid leu­ke­mia in treat­ment-free remis­
sion. Leukemia. 2020;34(4):1052-1061.
29. Kuepper MK, Bütow M, Herrmann O, et  al. Stem cell per­sis­tence in CML
is medi­ated by extrin­si­cally acti­vated JAK1-STAT3 sig­nal­ing. Leukemia.
2019;33(8):1964-1977.
30. Parting O, Langer S, Kuepper MK, et al. Therapeutic inhi­bi­tion of FcγRIIb
sig­nal­ing tar­gets leu­ke­mic stem cells in chronic mye­loid leu­ke­mia. Leuke-
mia. 2020;34(10):2635-2647.
31. Carter BZ, Mak PY, Mu H, et al. Combined targeting of BCL-2 and BCR-ABL
tyro­sine kinase erad­i­cates chronic mye­loid leu­ke­mia stem cells. Sci Transl
Med. 2016;8(355):355ra117.
112  |  Hematology 2021  |  ASH Education Program
32. Houshmand M, Garello F, Stefania R, et al. Targeting chronic mye­loid leu­
ke­mia stem/pro­gen­i­tor cells using venetoclax-loaded immunoliposome.
Cancers (Basel). 2021;13(6):1311.
33. Schütz C, Inselmann S, Saussele S, et  al. Expression of the CTLA-4 ligand
CD86 on plasmacytoid den­dritic cells (pDC) pre­dicts risk of dis­ease recur­
rence after treat­ment dis­con­tin­u­a­tion in CML. Leukemia. 2017;31(4):829-836.
34. Talpaz M, Hehlmann R, Quintás-Cardama A, Mercer J, Cortes J. Re-emer­gence
of inter­feron-α in the treat­ ment of chronic mye­ loid leu­ke­ mia. Leukemia.
2013;27(4):803-812.
35. Burchert A, Saussele S, Eigendorff E, et al. Interferon alpha 2 main­te­nance
ther­apy may enable high rates of treat­ment dis­con­tin­ua ­ ­tion in chronic
mye­loid leu­ke­mia. Leukemia. 2015;29(6):1331-1335.
36. Heibl S, Buxhofer-Ausch V, Schmidt S, et  al. A phase 1 study to eval­ua ­ te
the fea­si­bil­ity and effi­cacy of the addi­tion of ropeginterferon alpha-2b to
imatinib treat­ment in patients with chronic phase chronic mye­loid leu­ke­
mia (CML) not achiev­ing a deep molec­ul­ar response (molec­u­lar remis­sion
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/106/1851817/106hochhaus.pdf by guest on 13 December 2021
4.5)-AGMT_CML 1. Hematol Oncol. 2020;38(5):792-798.
37. Hsieh YC, Kirschner K, Copland M. Improving out­comes in chronic mye­
loid leu­ke­mia through harnessing the immu­no­log­i­cal land­scape. Leukemia.
2021;35(5):1229-1242.
38. National Comprehensive Cancer Network. Chronic mye­ loid leu­
ke­
mia.
NCCN Clinical Practice Guidelines in Oncology Version 3.2021. J Natl
Compr Canc Netw. 2020;18(10):1385-1415.
39. Rea D, Ame S, Berger M, et al; French Chronic Myeloid Leukemia Study
Group. Discontinuation of tyro­ sine kinase inhib­ it­ors in chronic mye­ loid
leu­ke­mia: rec­om­men­da­tions for clin­i­cal prac­tice from the French Chronic
Myeloid Leukemia Study Group. Cancer. 2018;124(14):2956-2963.
40. Schäfer V, White HE, Gerrard G, Möbius S, Saussele S, Franke GN, et  al.
Assessment of indi­vid­ual molec­u­lar response in chronic mye­loid leu­ke­mia
patients with atyp­i­cal BCR-ABL1 fusion tran­scripts: rec­om­men­da­tions by
the EUTOS coop­er­a­tive net­work. J Cancer Res Clin Oncol. 2021;147(3):3081-
3089.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000238
 CML: SUCCESS BREEDS MORE SUCCESS

Lifelong TKI therapy: how to manage

cardiovascular and other risks

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/113/1851432/113mauro.pdf by guest on 13 December 2021

Michael J. Mauro
Myeloproliferative Neoplasms Program, Leukemia Service, Memorial Sloan Kettering Cancer Center, New York, NY

Beginning with imatinib and now spanning 6 oral, highly active, and mostly safe agents, the development of specific tar-
geted therapy for patients with chronic myeloid leukemia (CML) has created a new world featuring chronic maintenance
chemotherapy for all treated as such, treatment-free remission, and functional cure after prolonged deep remission in a
subset. As a result comes a necessary shift in focus from acute to chronic toxicity, increasing attention to patient comor-
bidities, and critical thinking around specific adverse events such as metabolic, cardiovascular, and cardiopulmonary
effects, which vary from agent to agent. This review aims to pull together the state of the art of managing the “C” in
CML—a chronic myeloproliferative neoplasm treated at present over many years with oral BCR-ABL-targeted agents in a
population whose overall health can be complex and potentially affected by disease and therapy—and determine how
we can better manage a highly treatable and increasingly curable cancer.

LEARNING OBJECTIVES
• Understand the risks of a CML diagnosis and the importance of comorbidity assessment, especially cardiovascular
• Understand the specifc risks of CML tyrosine kinase inhibitors (approved and forthcoming) and how to screen for
and mitigate adverse events

Introduction events are rare to absent.2 The duration of treatment and

As has been stated often, the treatment of chronic mye-
loid leukemia (CML) was and remains revolutionized, with
lower-risk small-molecule-specifc targeted therapy with
oral ABL tyrosine kinase inhibitors (TKIs) as the overwhelm-
ing mainstay approach. Palliative cytoreductive agents
(hydroxyurea, busulfan) are rarely used except for initial
temporization of blood counts; interferon-based therapy
remains active in other myeloproliferative neoplasms but
is only used rarely and in special circumstances in which
TKI toxicity may be prohibitive (including pregnancy).
Through advances in conditioning, graft-versus-host pre-
vention, posttransplant TKI integration, and alternative
donor sources, allogeneic stem cell transplant continues to
be optimized, yet its use greatly diminished.1 Allografting
was previously viewed as the only “cure” for CML. Pres-
ently, patients treated with TKI therapy may avail them-
selves of the possibility of a “functional cure”—suffcient
treatment with TKI therapy into a sustained deep remission
followed by carefully observed TKI cessation—treatment-
free remission (TFR). Long-term results from the earliest
study of TFR demonstrate, importantly, that durable remis-
sion without TKI therapy can be sustained for many years,
retreatment for failed TFR attempts is effective, and late
Prakash Singh Shekhawat
duration of deep remission remain strong predictors of TFR
success,3 while factors such as minimal residual disease
stability,4 immune factors,5 and myeloid mutations beyond
BCR-ABL1 continue to be investigated as to their role.6
Increasingly, the number of patients potentially eligi-
ble or who have engaged in cessation of TKI therapy (TFR)
grows. A number of unmet needs remain for the optimal
treatment of CML, ranging from maximizing initial response
while minimizing toxicity and progression risk, to safer
and more effcient salvage approaches for heavily treated
multi-TKI resistant or multi-TKI resistant/intolerant or even
multi-TKI intolerant scenarios, to advanced phase or trans-
formed disease; added to this list now is improving the
outcomes of TFR—namely, by increasing the fraction of
successful cessation. Many, or perhaps most, view the cur-
rent goal of therapy to be successful treatment cessation
subsequent to sustained deep remission—what has been
termed as a functional cure. While this path is streamlined
and being made possible for a higher fraction of patients,
assuming the potential for lifelong TKI therapy remains a
pragmatic approach and perhaps the best way to balance
the risks and benefts of treatment strategies for those liv-
ing with CML.
Risks with lifelong TKI therapy in CML | 113
Survival in CML
The long-term fol­low-up of the piv­otal trial in CML, the IRIS
(International Randomized Study of Interferon and STI571) trial,
dem­on­strated the vastly improved out­look for patients treated
with TKI ther­apy com­pared to the prior stan­dard, inter­feron-
based ther­apy,7 heralding the “TKI era.” A com­pre­hen­sive anal­
y­sis of out­comes of cur­rent CML patients’ sur­vival rel­a­tive to
age-matched peers, irrespective of decade of age, con­cluded
no sig­nif­i­cant dif­fer­ence—lead­ing to the wel­come con­clu­sion of
a “nor­mal life span” in the TKI era.8 While the inci­dence of CML
remains steady—Euro­pean CML reg­is­try data cite an annual
inci­dence rate of 0.7 to 1.0/100 000, a median age at diag­no­
sis of 57 to 60 years, and a male/female ratio of 1.2 to 1.7—the
prev­a­lence of CML steadily increases world­wide due to mark­
edly improved sur­vival.9 Estimates in the US pro­ject increase
from a pre–TKI era prev­a­lence of roughly 25 000 indi­vid­u­als
liv­ing with CML to 200 000 plus such indi­vid­u­als over the next
3 decades. Such remark­able suc­cess focuses atten­tion on many
fronts, includ­ing global dis­par­ity, addressed by groups such as
the International CML Foundation, regard­ ing edu­ca­
tion and
treat­ment, and the Max Foundation, regard­ing patient advo­
cacy and med­i­ca­tion access; cost of ther­apy, given the mul­
ti­plier effect of patient and patient-years on TKI ther­apy; and
last and per­haps most impor­tantly, increased atten­tion to the
qual­ity of life and long-term out­come of CML “sur­vi­vors.”
New fron­tier: CML sur­vi­vor­ship
A nat­u­ral exten­sion of highly effec­tive ther­apy, the high frac­tion
of patients ush­ered into deep and sta­ble remis­sions, and the
prom­ise of a grow­ing poten­tial frac­tion of patients deemed func­
tion­ally cured begs the ques­tion of “CML sur­vi­vor­ship.” Distinct
from the past in which CML sur­vi­vors would be a mix of those fol-
lowed after allografting in the pro­to­typ­i­cal posttransplant sur­vi­
vor­ship clinic and a rare long-term responder to inter­feron-based
ther­apy, the pres­ent era calls for the orga­nized care of those
suc­cess­fully treated with TKIs, whether in TFR or not. Organized
care of the leu­ke­mia itself, of course, remains high pri­or­ity; for
those remaining on TKI ther­apy, reg­u­lar blood-based mon­i­tor­ing
with highly sen­si­tive quan­ti­ta­tive poly­mer­ase chain reac­tion for
the BCR-ABL1 fusion con­tin­ues indef­i­nitely per cur­rent rec­om­
men­da­tions.10 For those patients who have under­gone suc­cess­ful
TFR, guid­ance regard­ing opti­mal ini­tial mon­i­tor­ing con­tin­ues to
be scru­ti­nized; sub­se­quent ques­tions will include the more dif­fi­
cult ques­tion of taper­ing of long-term mon­i­tor­ing after suc­cess­
ful TFR.11 Patients with CML thus con­tinue to have long-stand­ing
rela­tion­ships with their CML phy­si­cians, and the devel­op­ment of
a tai­lored approach to the “life­long” TKI algo­rithm, to tackle the
“end” of CML treat­ment for those in TFR, and to delin­eate what
comes next from the per­spec­tive of rel­e­vant fol­low-up and risk
man­age­ment is a next step ripe for devel­op­ment.
CLINICAL CASE
A 55-year-old woman is referred in (via telehealth) for a recent
diag­no­sis of CML. Records indi­cate lower-risk dis­ease by the
Sokal and EUTOS Long Term Survival score, as her white blood
cell count is 32 000 with nor­mal hemo­glo­bin and plate­lets, no
cir­cu­lat­ing blasts, and a mod­est left-shifted dif­fer­en­tial with
3% each baso­phils and eosin­o­phils. She has no spleno­meg­aly,
114  |  Hematology 2021  |  ASH Education Program
and bone mar­row stud­ies sup­port chronic-phase dis­ease mor­
pho­log­i­cally, with cyto­ge­net­ics reveal­ing 100% sole Philadel-
phia chro­mo­some trans­lo­ca­tion in 20 meta­phases and base­line
BCR-ABL tran­scripts of the e14a2 var­i­ant of p210 fusion mea­sur­
ing 90% on the International Scale. She wishes to hear from you
on risks of dis­ease and med­i­ca­tion choices and under­stands
that treat­ment “may not need to be for­ever.”
Risks at pre­sen­ta­tion with a diag­no­sis of CML
The diag­no­sis of CML encompasses a num­ber of signs and symp­
toms, includ­ing fatigue, weight loss, loss of energy, decreased
exer­cise tol­er­ance, low-grade fevers, sweats resulting from hyper­
me­tab­o­lism, bone pain, and early sati­ety resulting from encroach­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/113/1851432/113mauro.pdf by guest on 13 December 2021
ment by an enlarged spleen on the stom­ach, as well as left upper
quad­rant full­ness and pain. Infarction of the spleen can be observed,
espe­cially with marked enlarge­ment. Blood count changes include
leu­ko­cy­to­sis and var­i­able plate­let count ele­va­tion or, with advanc­
ing-phase dis­ease, reduc­tion; ane­mia is gen­er­ally pro­por­tional to
dis­ease bur­den. Bleeding, pete­chiae, and ecchy­mo­ses are more
likely with advanced dis­ease and throm­bo­cy­to­pe­nia.
The pre­sen­ta­tion of CML with marked thrombocytosis (plate­
lets >1 mil­lion) may occur, and vas­cu­lar injury and throm­botic
events appear more com­mon in such cases, as well as a female
pre­dom­i­nance and youn­ger age.12 In chil­dren, throm­bo­sis may be
less likely, but as is seen in other mye­lo­pro­lif­er­a­tive neo­plasms,
acquired von Willebrand syn­drome and mild clin­i­cal bleed­ing
may develop due to the bind­ing of von Willebrand fac­tor multi-
mers, espe­cially large multimers, to plate­lets.13 Not sur­pris­ingly,
a grow­ing num­ber of case reports are emerg­ing iden­ti­fy­ing the
coex­is­tence of mye­lo­pro­lif­er­a­tive “driver” muta­tions, such as
JAK2 v617F and calreticulin along with BCR-ABL fusion, as deep
sequenc­ing becomes more uti­lized, suggesting that bet­ter char­
ac­ter­iza­tion of mye­loid muta­tions in CML may be of clin­i­cal use.
Often iso­lated to set­tings where pre­sen­ta­tion and diag­no­sis
may be delayed, com­pli­ca­tions related to leukostasis in tar­get
organs at risk for vas­cu­lar injury include priapism and ret­i­nop­a­
thy. Leukapheresis can be used for clin­i­cal signs of injury due to
leukostasis and remains a ther­a­peu­tic “bridg­ing” strat­egy dur­ing
CML and preg­nancy, when cytoreductive ther­apy and defin­i­tive
ther­apy with TKIs may pose exces­sive fetal risk.14
Risks asso­ci­ated with the first-line TKIs imatinib, nilotinib,
dasatinib, and bosutinib
The ini­ tial treat­
ment of CML increas­ ingly is the direct use of
TKI ther­apy in order to promptly induce hema­to­logic con­trol/
com­plete hema­to­logic response; cytoreduction with hydroxy­
urea prior to TKI start, if used, may aug­ment the myelosuppres-
sion observed with sub­se­quent TKI ther­apy. Myelosuppression
remains a ubiq­ui­tous ini­tial effect com­men­su­rate with TKI start
and is some­what var­i­able among the avail­­able ABL kinase inhib­i­
tors; in com­par­a­tive tri­als of first-line ther­apy with imatinib, nilo-
tinib, dasatinib, and bosutinib, dasatinib was asso­ci­ated with the
greatest rates of grade 3 or 4 neutropenia and throm­bo­cy­to­pe­
nia. Myelosuppression may be best viewed as a mix of dis­ease
and ther­apy-related effects given its (early) kinet­ics and pro­
pen­sity to resolve; a sub­set of patients, often with higher-risk
fea­tures, may face severe and per­sis­tent myelosuppression pre­
clud­ing TKI deliv­ery and response to ther­apy, pos­ing a spe­cific
intol­er­ance and chal­lenge.
 Other adverse events com­ mon to the entire class of ABL
inhib­i­tors used at diag­no­sis are gen­er­ally revers­ible and pose lit­
tle known long-term risk. Examples of such lab­o­ra­tory find­ings
include trans­am­i­nase ele­va­tion, lipase ele­va­tion (more often bio­
chem­i­cal and lower with pan­creas inflam­ma­tion or pan­cre­a­ti­
tis), hyperbilirubinemia (observed more with nilotinib and often
unmasking Gilbert’s syn­drome), and oth­ers.15 Adverse events more
closely linked to imatinib include edema (periorbital and periph­
eral), mus­cle cramps, mus­cu­lo­skel­e­tal pain, and diar­rhea; those
more closely linked to nilotinib include rash, pru­ri­tus, head­ache,
nasopharyngitis, constipation, abdom­i­nal pain, vomiting, pyrexia,
upper uri­ nary tract infec­
tion, back pain, cough, and asthe­ nia;
those more closely related to dasatinib include, as pre­ vi­
ously
men­tioned, myelosuppression as well as fluid reten­tion, includ­ing
pleu­ral and peri­car­dial effu­sions and head­ache; and those more
closely related to bosutinib include diar­rhea and increases in ala­
nine and aspar­tate ami­no­trans­fer­ase. Common, more fre­quent
adverse events across TKIs include fatigue myal­gias and gas­tro­
in­tes­ti­nal symp­toms such as nau­sea and vomiting. Given a broad
inves­ti­ga­tion into the opti­mal ther­a­peu­tic dose, imatinib plasma
trough lev­els were noted to cor­re­late with effi­cacy and the cor­
rec­tion of sub­op­ti­mal response16; how­ever, drug lev­els did not cor­
re­late with adverse events. Longer fol­low-up of imatinib-treated
patients in the land­mark IRIS trial con­cluded that adverse events
were not typ­i­cally observed after the first year of ther­apy.7
Particular aware­ ness and pre­ pared­ ness are warranted for
adverse events of spe­cific inter­est (asso­ci­ated with greater mor­
bid­ity and poten­tial long-term impact) for TKIs used at diag­no­sis.
Imatinib is asso­ci­ated with hypophosphatemia; fur­ther inves­ti­
ga­tion dem­on­strated a poten­tial link to altered bone min­er­al­i­
za­tion.17 Renal injury, manifesting as a decline of the glo­mer­u­lar
fil­tra­tion rate, was noted with imatinib ther­apy, with reduc­tions
steadily over 4 years’ time before pla­teau and iso­lated to those
with no chronic kid­ney dis­ease prior18; in a sep­a­rate study, nearly
iden­ti­cal changes were observed for bosutinib ther­apy.19 Inter-
estingly, TKI dose did not cor­re­late with injury and revers­ibil­ity
was appar­ent. Nilotinib is asso­ci­ated with impaired fasting glu­
cose lev­els,20 whereas imatinib has been suggested to improve
them21; this effect with nilotinib is reported to be revers­ible, to
be linked to increased body mass index, and not to trig­ger type
2 dia­be­tes.22 Lipid pro­files may show sim­i­lar changes, with ima-
tinib improv­ing lev­els and nilotinib, as early as 3 months into
treat­ment, trig­ger­ing increases in cho­les­terol (both low-den­sity
and high-den­sity lipo­pro­teins and in turn total cho­les­terol); any
effect of dasatinib appears more mod­est.23-25
In con­trast to imatinib, vas­cu­lar adverse events have been
asso­ci­ated with newer-gen­er­a­tion TKIs, most prom­i­nently with
the later-line ther­apy agent ponatinib as well as with the sec­ond-
gen­er­a­tion TKIs dasatinib and nilotinib,26,27 while min­i­mal asso­ci­
a­tion has been observed with bosutinib.28 Long-term fol­low-up
of the ENESTnd trial com­pared nilotinib at var­ied doses vs imati-
nib for newly diag­nosed CML. At the 10-year median fol­low-up,
rates were 16.5% for nilotinib at 300 mg twice daily and 23.5% for
400 mg twice daily vs 3.6% for imatinib, includ­ing in Framingham
car­dio­vas­cu­lar-assessed low-risk patients.29 Mechanisms remain
under inves­ti­ga­tion; pos­tu­lated causes include endo­the­lial cell
effects,30 proinflammatory and pro-oxi­ da­tive stress fac­tors,31
and lipid, coag­u­la­tion, and plate­let effects (some par­a­dox­i­cal),32
among oth­ers. Sequencing stud­ies have suggested dif­fer­en­tial
expres­sion pat­terns in tar­get tis­sues (such as endo­the­lial cells),33
Prakash Singh Shekhawat
and the spec­ter of clonal hema­to­poi­e­sis has been rightly raised
as a poten­tial con­trib­ut­ing fac­tor in throm­botic events in CML.34
Of note, the pres­ence and poten­tial impact of other mye­loid
muta­tions aside from BCR-ABL1 fusion on sev­eral fronts, includ­
ing pri­mary response, the abil­ity to achieve suc­cess­ful TFR in oth­
er­wise “eli­gi­ble” cases, and adverse events, espe­cially vas­cu­lar,
are just begin­ning to be unveiled as broader, deeper sequenc­ing
is performed in CML.6
Pulmonary com­pli­ca­tions of TKI ther­apy are also of spe­cific
inter­est, warranting aware­ness, early inter­ven­tion, and con­sid­er­
ation of spe­cific sur­veil­lance. Most nota­bly, pleu­ral and peri­car­dial
fluid accu­mu­la­tion as well as pul­mo­nary arte­rial hyper­ten­sion have
been observed, most often with dasatinib ther­apy. Initial anal­y­sis
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/113/1851432/113mauro.pdf by guest on 13 December 2021
suggested that car­diac dis­ease and sys­temic arte­rial hyper­ten­sion
were predisposing fac­tors to develop pleu­ral effu­sions on dasat-
inib35; a larger meta-anal­y­sis with lon­ger fol­low-up suggested, in
mul­ti­var­i­ate anal­y­sis, that age was an inde­pen­dent pre­dic­tor.36
Distinct from many other effects asso­ci­ated with TKIs, the risk
of devel­op­ing pleu­ral effu­sions may be spread over many years’
time or the entirety of treat­ment; a 7-year fol­low-up of sec­ond-line
dasatinib noted a cumu­la­tive inci­dence of 28% to 35% (depending
on TKI dose) and a 5% inci­dence of new events in year 7.37 The
DASISION study of front­line dasatinib noted a steady 8% per year
inci­dence and a cumu­la­tive inci­dence of 38% at 5 years and also
cited age as a pre­dic­tor.38 Both dose and sched­ule influ­enced the
like­li­hood of pleu­ral effu­sions in sec­ond-line stud­ies, con­trib­ut­ing
to the change to opti­mized once-daily reduced dos­ing of dasatinib
in CML. Continued inquiry into opti­miz­ing the risk/ben­e­fit bal­ance
with dasatinib is evidenced by a com­pel­ling sin­gle-cen­ter, sin­gle-
arm study of 50 mg dos­ing for newly diag­nosed chronic-phase CML
in which pleu­ral effu­sions were observed in only 6% of patients
after a median fol­low-up of 24 months.39
Specific man­age­ment rec­om­men­da­tions in the case of dasat-
inib pleu­ral effu­sions have been devel­oped40; how­ever, the exact
mech­a­nisms of action and the best mit­i­ga­tion strat­egy remain
elu­sive. Given its pro­file beyond ABL kinase inhi­ bi­tion, as an
inhib­i­tor of the SRC fam­ily of kinases and lym­pho­cyte-spe­cific
pro­tein kinase, a related phe­nom­e­non observed with dasati-
nib ther­apy includes periph­eral blood lym­pho­cy­to­sis,41 with a
large gran­u­lar lym­pho­cyte mor­phol­ogy that is directly cor­rel­
a­tive to dasatinib dos­ing.42 While poten­tially asso­ci­ated with
other inflam­ma­tory phe­nom­ena, includ­ing coli­tis and fol­lic­u­lar
hyper­pla­sia/lymph node enlarge­ment,43 periph­eral blood lym­
pho­cy­to­sis is observed in approx­i­ma­tely one-third of treated
patients and is asso­ci­ated with higher response rates, lon­ger
response dura­tions, and increased over­all sur­vival, support-
ing a poten­tial immu­no­mod­u­la­tory role.44 Additionally, pul­mo­
nary arte­rial hyper­ten­sion has been observed, thank­fully rarely
(<1%).45 It is most often asso­ci­ated with dasatinib (and occurs
less often but has been noted with bosutinib and ponatinib) yet
con­founded by how infre­quently right heart cath­e­ter­i­za­tion is
performed to inves­ti­gate symp­toms prop­erly. Carefully vet­ted
cases (diag­nosed with right heart cath­e­ter­i­za­tion) warranted TKI
dis­con­tin­u­a­tion, and while the major­ity were revers­ible, approx­i­
ma­tely one-third of cases dem­on­strated per­sis­tence,46 val­i­dat­ing
the need for closer atten­tion to this TKI risk.
Risks asso­ci­ated with ponatinib
Considered inde­pen­dently given its more dis­tinct risk pro­file,
ponatinib offers an excel­lent abil­ity to sal­vage poor response in
Risks with life­long TKI ther­apy in CML  |  115
chronic-phase CML, shows effi­cacy in the face of select resis­
tance (ABL kinase domain muta­tion thre­o­nine for iso­leu­cine at
posi­ tion 315-T315I), and has a prob­ a­
bly often mis­ un­der­ stood
adverse event pro­file. Ponatinib expe­ri­ence in a phase 1 study
noted prominent adverse effects, includ­ing rash and acnei-
form der­ma­ti­tis, arthral­gias/fatigue, and a clear risk of pan­creas
inflam­ma­tion greater than other TKIs with lipase ele­va­tion, hyper-
triglyceridemia, and pan­cre­a­ti­tis at equally ele­vated rates47; a
sig­nal of vas­cu­lar tox­ic­ity had not yet emerged. Commensurate
with pub­li­ca­tion of the PACE trial in late 2013,48 the US Food and
Drug Administration (FDA) requested and the man­uf­ac­turer (at
that time, Ariad Pharmaceuticals) agreed to sus­pend the mar­
ket­ing of ponatinib. Data from PACE showed “sig­nif­i­cant anti­leu­
ke­mic activ­ity across categories of dis­ease stage and muta­tion
sta­tus” but reported that seri­ ous arte­ rial throm­botic events
(includ­ing cere­bro­vas­cu­lar, car­dio­vas­cu­lar, and periph­eral vas­cu­
lar events) were ris­ing, ini­tially num­ber­ing 8.9% (2.9% treat­ment
related) dur­ing the study and then going up to 11.8%. The inci­
dence of all­arte­rial throm­botic events, seri­ous or not, was 17.1%.
While the rate was not increas­ing over time, the accu­mu­la­tion
with addi­tional fol­low-up (>24 mo) led to the mar­ket­ing sus­pen­
sion and halt­ing of ponatinib front­line stud­ies. As men­tioned rel­
a­tive to the sec­ond-gen­er­a­tion TKIs nilotinib and bosutinib, the
mech­an ­ isms and basis for such events remain unclear, with sev­
eral hypoth­es­es. A focus on ponatinib in par­tic­ul­ar is ongo­ing
and includes such con­cepts as a throm­botic throm­bo­cy­to­pe­nic
pur­pura-like microangiopathic mech­a­nism.49
Based on addi­tional data from the phase 1 study and the afore­
men­tioned emerg­ing data, the label for ponatinib then included
a vas­cu­lar adverse event (arte­rial and venous) rate of 27% and
a fairly nar­row indi­ca­tion (T315I muta­tion-bear­ing patients and
those in whom “no other TKI is indi­cated”), and a risk eval­u­a­tion
and mit­i­ga­tion strat­egy was requested. In con­trast, ponatinib
retained Euro­pean Medicines Agency mar­ket­ing autho­ri­za­tion
as well as its indi­ca­tion for patients car­ry­ing the T315I muta­tion
and those resis­tant or intol­er­ant to dasatinib or nilotinib and for
whom imatinib was not clin­ic ­ ally appro­pri­ate and con­cluded
a some­what lower vas­cu­lar event rate. Data and fur­ther guid­
ance for the opti­mal use of ponatinib have con­tin­ued to evolve;
as such the 2020 updated guide­lines issued by the Euro­pean
LeukemiaNet rec­om­mend ponatinib in patients with resis­tance
to a sec­ond-gen­er­a­tion TKI (dasatinib, nilotinib, and bosutinib),
even with­out spe­cific muta­tions, unless its use is pre­cluded by
the pres­ence of car­dio­vas­cu­lar risk fac­tors.50 Very recent data
from the dose-find­ing OPTIC study,51 enroll­ing patients with CML
resis­tant/intol­er­ant to ≥2 TKIs or car­ry­ing the T315I muta­tion to
receive 45, 30, or 15 mg ini­tially followed by 15 mg main­te­nance
upon achiev­ing ≤1% BCR-ABL lev­els, sug­gest an opti­mal ben­e­fit/
risk pro­file for the 45 mg starting dose.
What lies at the heart of the mat­ ter—no pun intended—
regard­ing the true risk of vas­cu­lar adverse events is per­haps how
they are quan­ti­fied and attrib­uted. A recent study deployed an
inde­pen­dent end point adju­di­ca­tion com­mit­tee (EAC) includ­
ing car­di­­ol­ogy, hema­tol­ogy, and neu­rol­ogy input to reexamine
the vas­cu­lar adverse events from the PACE study of ponatinib.
It noted that while AOEs iden­ti­fied by the Medical Dictionary
for Regulatory Activities pre­ferred terms num­bered 25%, the
blinded EAC, using Amer­i­can College of Cardiology/Amer­i­can
Heart Association def­i­ni­tions, noted 17%.52 Serious AOEs by the
Medical Dictionary for Regulatory Activities terms num­ bered
116  |  Hematology 2021  |  ASH Education Program
20%, while the EAC ver­i­fied 16%. With these data and a sup­
ple­men­tal new drug appli­ca­tion, the FDA approved an updated
indi­ca­tion for ponatinib in the US—namely, patients with resis­
tance or intol­er­ance to at least 2 prior kinase inhib­i­tors.
On the hori­zon: asciminib
Asciminib is a novel ABL kinase inhib­i­tor, a so-called STAMP (spe­
cif­i­cally targeting the ABL myristoyl pocket) inhib­i­tor. Function-
ing as an allo­ste­ric inhib­i­tor of the deregulated BCR-ABL kinase
by targeting the myristoyl pocket in con­ trast to the active
site (aden­o­sine tri­phos­phate-bind­ing) mech­a­nism of all­ other
licensed ABL kinase inhib­i­tors, asciminib holds the prom­ise of
dis­tinct sin­gle-agent activ­ity in Philadelphia-pos­i­tive leu­ke­mias
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/113/1851432/113mauro.pdf by guest on 13 December 2021
as well as the poten­tial for ratio­nal com­bi­na­tion ther­apy (aden­
o­sine tri­phos­phate-bind­ing + myristoyl-bind­ing ABL inhib­i­tor
ther­apy). Initial phase 1 study reports in mainly chronic-phase
patients with multi-TKI resis­tance/intol­er­ance indi­cated con­
vinc­ing effi­cacy and min­i­mal resis­tance via novel muta­tions.53
A sub­se­quent ran­dom­ized study of asciminib in com­par­i­son to
bosutinib for chronic-phase CML patients with 2 prior ther­a­
pies and no resis­tance muta­tions to bosutinib or T315I muta­tion
con­firmed asciminib’s supe­ri­or­ity regard­ing major molec­u­
lar response at 6 months, the pri­mary end point of the study,
as well as other end points, includ­ing com­plete cyto­ge­netic
response.54 Given its nov­elty as a STAMP inhib­i­tor, its potency
and spec­i­fic­ity for the myristoyl pocket of ABL1, and its broad
inac­tiv­ity against other kinases, includ­ing SRC kinase,55 the tox­
ic­ity pro­file of asciminib to date has also been prom­is­ing. The
phase 1 study lim­it­ing tox­ic­ity was lim­ited to lipase ele­va­tion
with a sin­gle case of pan­cre­a­ti­tis, arthral­gia/myal­gia, abdom­
i­nal pain, and bron­cho­spasm and a sin­gle case of acute cor­
o­nary syn­drome; lipase ele­va­tion/pan­cre­a­ti­tis remains a class
effect prominent with nilotinib and ponatinib and now asso­ci­
ated with asciminib. In the phase 3 ASCEMBL study com­par­ing
asciminib to bosutinib, mark­edly lower-gas­tro­in­tes­ti­nal adverse
events were observed with asciminib com­pared to bosutinib, as
one might expect; how­ever, as with all­TKIs, closer atten­tion is
often paid to vas­cu­lar adverse events, and 2 deaths occurred in
the asciminib arm fall­ing into this cat­e­gory (ische­mic stroke and
arte­rial embolism per the report). Dissecting out any asso­ci­a­
tion and risk will take more time, and rec­og­ni­tion of the heavily
pretreated nature of the patient pop­u­la­tion in the ASCEMBL trial
is warranted. At the pres­ent time, asciminib appears to have
a nar­row spec­trum of tox­ic­ity, and no con­clu­sive evi­dence of a
vas­cu­lar/car­dio­vas­cu­lar sig­nal exists as of yet, nota­bly after a
very long-last­ing (>7 years) phase 1 study. It offers great prom­
ise in resis­tant/intol­er­ant patients, will inspire ongo­ing explo­ra­
tion into com­bi­na­tion approaches, and rep­re­sents a sig­nif­i­cant
advance­ment in the field.
CLINICAL CASE (Con­tin­ued)
After thor­ough dis­cus­sion about risk strat­i­fi­ca­tion and the rel­
a­tive mer­its and risks of avail­­able TKI ther­ap
­ ies spe­cific to her
case, you find your­ self not­
ing the safety and likely suc­ cess
of imatinib-based ther­ apy contrasted with the poten­ tial for
improved early and deep response with any of the sec­ond-
gen­er­a­tion TKI options and presenting either option as rea­son­able.
You pro­ceed to take a more detailed back­ground med­i­cal his­
tory and find her to be free of any active car­dio­vas­cu­lar or met­
a­bolic dis­or­ders such as heart dis­ease, hyper­ten­sion, dia­be­tes,
renal dis­ease, hyper­lip­id­emia, or other chronic con­di­tions. Her
weight is near ideal, and she does not smoke; how­ever, she
can­not offer a fam­ily his­tory as she was adopted as an infant.
When asked about recent screen­ing for the above con­di­tions,
she can­not recall her last pri­mary care appoint­ment as she is
“very healthy and has not needed to go to the doc­tor, espe­
cially dur­ing the pan­demic” and has never seen a car­di­­ol­o­gist
or had any car­dio­vas­cu­lar stud­ies.
Assessing comorbidity in the CML patient pop­u­la­tion
Simply stated, to man­age the car­dio­vas­cu­lar and other risks in
patients with CML, one must embrace the “whole patient”—the
full health his­ tory, active and qui­ es­
cent comorbid con­ di­
tions,
vulnerabilities as best one can judge—plac­ing gen­eral med­i­cal
con­sid­er­ations in clear focus in antic­i­pa­tion of leu­ke­mia care. The
aver­age age at CML diag­no­sis varies widely across the globe,56
rang­ing from medi­ans in the 40s in Asia and Africa, to the 60s in
North America, and to >70 in the Oceanic region (Australia, etc).
The local/regional spe­cif­ics of poten­tial clin­i­cal chal­lenges vary
and may not be well char­ac­ter­ized—eg, in a large ter­tiary refer­
ral cen­ter in Soweto, Johannesburg, South Africa, 7% of all­CML
patients carry a diag­no­sis of both HIV and CML.57 Access to ther­apy
is not equi­ta­ble glob­ally; orga­ni­za­tions such as the non­profit MAX
Foundation (www​­.themaxfoundation​­.org) have partnered with key
spon­sors to facil­i­tate low- or no-cost TKI ther­apy for CML and other
dis­eases, account­ing for >10 mil­lion doses of ther­apy over a recent
span of 3 years. Underserved regions may have lim­ited access to
pro­vid­ers and risk assess­ment, adding to the chal­lenge.
In more eco­nom­i­cally favor­able set­tings, data sug­gest that
comorbidities are com­mon in patients with CML and may influ­
ence treat­ment choice and out­come. A large US series of >2000
CML patients and data from the EUTOS (Euro­pean Treatment
and Outcome Study for CML) study with nearly 3000 cases
found that comorbidities were pres­ent in >50% of CML cases,
and car­dio­vas­cu­lar-related comorbidities such as hyper­ten­sion
and dia­be­tes were pres­ent in approx­i­ma­tely 40% of cases.58,59
The large obser­va­tional SIMPLICITY study reported that the
pri­mary basis for TKI choice (US/EU data set) was for “per­
ceived effi­cacy” in >50%, while only approx­i­ma­tely 9% pri­mar­
ily regarded comorbidities.60 Widely used treat­ment guide­lines
such as the Euro­pean LeukemiaNet have begun to incor­po­
rate a broader con­sid­er­ation of comorbidities into the ini­tial
screen­ing (inclu­sion of lipid pro­file, dia­be­tes screen­ing, etc)
and com­ment on patient selec­tion based on comorbid risks
and adverse event pro­files of the var­i­ous TKIs.50 The US National
Comprehensive Cancer Network,10 while clearly stat­ing TKI tox-
icities and man­age­ment, has yet to sig­nif­i­cantly incor­po­rate
comorbidity assess­ment or related risk strat­i­fi­ca­tion.
Several stud­ ies have clearly driven home the cen­ tral­
ity of
comorbidities in CML out­comes. The Charlson Comorbidity Index,
which incor­po­rates age along with key con­di­tions—includ­ing
myo­car­dial infarc­tion, con­ges­tive heart fail­ure, periph­eral vas­cu­
lar dis­ease, cere­bro­vas­cu­lar acci­dent or tran­sient ische­mic attack,
demen­tia, chronic obstruc­tive pul­mo­nary dis­ease, con­nec­tive
tis­sue dis­ease, pep­tic ulcer dis­ease, liver dis­ease, dia­be­tes melli-
tus, hemi­ple­gia, chronic kid­ney dis­ease, can­cer (blood and solid
Prakash Singh Shekhawat
tumor), and AIDS—was exam­ined in the set­ting of imatinib-based
ther­apy in the Ger­man CML IV study and in the broader set­ting
of imatinib, nilotinib, and dasatinib.61,62 Both stud­ies found that
the Charlson Comorbidity Index was con­sis­tently the stron­gest
pre­dic­tor of sur­vival in CML, includ­ing age-inde­pen­dent anal­y­sis,
and in fact did not affect TKI treat­ment suc­cess, lead­ing to the
con­clu­sion that sur­vival of patients with CML is deter­mined more
by comorbidities than by CML itself.
The who, what, and when of risk assess­ment
and mit­i­ga­tion for the patient with CML
One of my most mem­o­ra­ble quotes from med­i­cal school, from
a cherished leu­ke­mia attend­ing, was sim­ple and pow­er­ful: “You
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/113/1851432/113mauro.pdf by guest on 13 December 2021
can’t be a good hema­tol­o­gist/oncol­o­gist unless you are a good
inter­nist first.” A leu­ke­mia diag­no­sis, even one with such hope,
mul­ti­plic­ity of effec­tive options, and guid­ance based on response
mile­stones as CML has, often leaves other health issues and fol­low-
up in a blur as leu­ke­mia care comes into focus. Leukemia care
engulfs the pres­ent, with fre­quent blood work and check­ups; given
the inten­sity and nuances of care, oth­ers feel it best to step aside
and stay at a dis­tance for fear of neg­a­tively affect­ing can­cer treat­
ment suc­cess. Rather than a point of sep­a­ra­tion, in CML, the time
of diag­no­sis and there­af­ter need to be the oppo­site—a point of
alli­ance between leu­ke­mia spe­cial­ists, pri­mary care/inter­nal med­
i­cine, and oth­ers as needed, par­tic­u­larly in car­dio­vas­cu­lar med­i­
cine. Continued fol­low-up with pri­mary care/inter­nal med­i­cine is
essen­tial for patients with CML. With high rates of remis­sion and
data supporting a nor­mal life expec­tancy, every­thing stays on the
table. Age-appro­pri­ate “other” can­cer screen­ings should pro­ceed,
as well as man­age­ment of the fre­quently observed and poten­tially
exac­er­bated comorbidities of hyper­gly­ce­mia/dia­be­tes, hyper­lip­
id­emia, and hyper­ten­sion and first and fore­most, screen­ing for and
man­age­ment of car­dio­vas­cu­lar dis­ease. Open dia­logue, divi­sion
of labor, and free com­mu­ni­ca­tion regard­ing treat­ment plans and
inter­ven­tions must be established early after diag­no­sis to pre­vent
missed oppor­tu­nity. Seemingly minor things, such as pre­scrip­tion
of acid sup­pres­sion med­i­ca­tion such as pro­ton pump inhib­i­tors,
may severely impair TKI expo­sure and response.63
The sub­spe­cialty field of “cardio-oncol­ogy”—car­di­­ol­o­gists
who focus on can­cer dis­ease and ther­apy com­pli­ca­tions—has
been a boon to hema­tol­ogy/oncol­ogy pro­vid­ers man­ag­ing CML
as the tri­fecta of (1) remark­ably lon­ger sur­vival/nor­mal life span
expec­ta­tions with the devel­op­ment of TKIs, (2) the breadth of
TKI choices, includ­ing more potent BCR-ABL1 inhib­i­tors, and,
unfor­tu­nately, (3) the rec­og­ni­tion that vas­cu­lar/car­dio­vas­cu­lar
AEs may be among the most impactful risks hin­der­ing the over­
all suc­cess of CML ther­apy. Having car­di­­ol­ogy pres­ent as part
of the CML edu­ca­tion ses­sion at the 2017 Amer­i­can Society of
Hematology meet­ing dem­on­strated the need for closer alli­ance,
col­lab­o­ra­tion, and coman­age­ment.64 Provisional guide­lines have
been published in reviews over the last sev­eral years focused
on man­ag­ing car­dio­vas­cu­lar adverse events in CML,65-67 focused
spe­ cif­i­
cally on ponatinib and nilotinib,68-70 as well as posi­ tion
papers suggesting CML clin­ i­
cal trial adverse event reporting
should align with car­dio­vas­cu­lar med­i­cine def­i­ni­tions.71
So, who? Practically speak­ing, inven­tory of comorbid con­di­
tions and vet­ting of nec­es­sary inter­ven­tions, and spe­cific inven­
tory of car­dio­vas­cu­lar risk, is pru­dent for all­ patients. How?
Several val­i­dated tools are read­ily avail­­able to assess car­dio­
vas­cu­lar risk and sub­se­quent event (for exam­ple, 10 year) risk,
Risks with life­long TKI ther­apy in CML  |  117
includ­ing the Framingham “hard” (non­di­ab ­ etic, no prior clau­
di­ca­tion/cor­o­nary heart dis­ease) risk score stem­ming from the
long-run­ning Framingham heart study.72,73 The Euro­pean Society
of Cardiology devel­oped and val­i­dated the SCORE (Systematic
Coronary Risk Evaluation) Risk Charts for use as a car­dio­vas­cu­
lar dis­ease risk assess­ment model for patients in Europe. It is
divided into higher- and lower-risk countries and can be cali-
brated for spe­cific countries’ mor­tal­ity data.74 The SCORE charts
were exam­ined in the set­ting of nilotinib ther­apy by sev­eral
CML groups, includ­ing France and Italy,75,76 dem­on­strat­ing that
base­line assess­ment via SCORE could be a valid tool to iden­
tify patients at high risk of ath­ero­scle­rotic events dur­ing nilo-
tinib treat­ment. Long-term data emerg­ing from the ENESTnd
study of front­line nilotinib, using Framingham risk assess­ment,
cau­tioned that in addi­tion to higher cumu­la­tive rates of car­
dio­vas­cu­lar events reported with nilotinib (300 mg twice daily,
16.5%; 400 mg twice daily, 23.5%) vs imatinib (3.6%), said events
were pos­si­ble in Framingham low-risk patients.29 Other region-
spe­cific risk assess­ment tools can and should be incor­po­rated
into prac­tice with CML patients; in the United Kingdom, the
QRISK score also cal­cu­lates the 10-year risk of car­dio­vas­cu­lar
or cere­bro­vas­cu­lar events spe­cific to the UK pop­u­la­tion and
incor­po­rates postal codes in addi­tion to an increas­ing num­ber
of clin­i­cal pre­dic­tors.77
The “when” of car­dio­vas­cu­lar risk assess­ment—fre­quency or
peri­od­ic­ity of key inter­ven­tions at a gen­eral level at least and ide­
ally patient or patient/TKI risk level—and likely the more pre­cise
details of which diag­nos­tic tools, poten­tial bio­mark­ers, etc, are
most infor­ma­tive con­tinue to be inves­ti­gated. A unique pro­spec­
tive study in the US of car­dio­vas­cu­lar and met­a­bolic param­e­ters
at diag­no­sis of CML and sub­se­quently through treat­ment, agnos­
tic to TKI choice, has reported high rates of base­line risk and con­
tin­ues to gather data on changes with time on treat­ment.78 At
pres­ent, the use of pro­vi­sional cardio-oncol­ogy-derived guide­
lines and leu­ke­mia con­sor­tium-based guide­lines,64,79,80,81 driven
by patient comorbidity and TKI choice, adds sim­ple and lower
cost empiric mon­i­tor­ing based on known risks and thee best-
avail­­able diag­nos­tics/pre­dic­tors of preexisting or accel­er­at­ing
vas­cu­lar dis­ease. At the most basic level, broad use of a sim­ple
ABCDE approach—(a)wareness, (a)nkle-bra­ chial index test­ ing,
(a)spirin ther­apy, (b)lood pres­sure con­trol, (c)igarette ces­sa­tion,
and (c)holesterol low­er­ing—will raise aware­ness and at a min­i­
mum facil­i­tate stan­dard of care inter­ven­tion for comorbid find­
ings in CML patients.
CLINICAL CASE (Con­t in­u ed)
While final­iz­ing the last hema­to­logic assess­ments and confirm-
ing her genetic results and pathol­ogy, you offer the patient
a con­sul­ta­tion in your cen­ter’s pri­mary care clinic, which she
eagerly pur­sues as she is wor­ried about her gen­eral health
given the CML diag­no­sis. The result of your TKI choice dis­cus­
sion led you to rec­om­mend a sec­ond-gen­er­a­tion TKI, and in
dis­cus­sion with the pri­mary care team, a screen­ing eval­u­a­tion
with the cen­ter’s car­di­­ol­ogy group is sched­uled. She under-
goes basic met­a­bolic and func­tional imag­ing and is found to
have a low risk of events in the next 10 years. Cardiovascular
mon­i­tor­ing is pre­scribed by the car­di­­ol­o­gist to con­tinue in the
118  |  Hematology 2021  |  ASH Education Program
pri­mary care and hema­tol­ogy clin­ics with basic blood pres­sure
and met­a­bolic dis­ease screen­ing, with a return to car­di­­ol­ogy if
symp­toms or evi­dence of vas­cu­lar dis­ease devel­ops.
Cardiovascular risk man­age­ment and pre­ven­tion:
too much, too lit­tle?
It is inar­ gu­
able that inter­ven­ tion for comorbidities, such as
aspi­rin or lipid-low­er­ing ther­apy as pri­mary or sec­ond­ary pre­
ven­tion in patients with iden­ti­fied indi­ca­tions based on car­dio­
vas­cu­lar guide­lines, is essen­tial. Intervention based on risk of
com­pli­ca­tions dur­ing TKI ther­apy, based solely on the TKI risk,
is much less clear of a ben­e­fit/risk and ide­ally should be stud­ied
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/113/1851432/113mauro.pdf by guest on 13 December 2021
in con­trolled tri­als. Given the size needed for such tri­als done
in the non-CML set­ting, firm evi­dence-based data for pre­emp­
tive inter­ven­tion may be elu­sive. Smaller directed stud­ies have
looked at pre­ven­ta­tive inter­ven­tions; an Ital­ian study exam­in­
ing ponatinib-treated patients (n = 85), strat­i­fied based on the
Euro­pean Society of Cardiology SCORE charts as was done for
nilotinib, included a pri­mary pre­ven­tion strat­egy using 100 mg
of aspi­ rin daily in a sub­
set (n = 19); no sig­ nif­i­
cant dif­
fer­
ences
were noted, but trends toward decreased AOEs and improved
sur­vival were reported.82 Drug-drug inter­ ac­
tions and other
adverse event risks from adding pre­ven­ta­tive ther­apy are not
to be overlooked. While we may feel com­pelled to inter­vene
with pre­ven­tion for all­patients treated with higher-risk TKIs, risk
strat­i­fi­ca­tion may help jus­tify inter­ven­tion in larger num­bers of
cases, but an “only where indi­cated” approach may be more
pru­dent.
Is TKI ther­apy life­long, and could bet­ter ther­apy help
mit­i­gate risk?
We are at a cross­roads in the treat­ment of CML, hav­ing first
rev­o­lu­tion­ized the prin­ci­pal approach with the devel­op­ment of
TKIs and now mov­ing away from a “trans­plant when pos­si­ble”
approach and expec­ta­tions of dimin­ished sur­vival in oth­ers to
a par­a­digm of “trans­plant only when abso­lutely needed” and
near-nor­mal or nor­mal life expec­tancy for most. Targeted ther­
apy in CML has fos­tered a new modus ope­randi of chronic main­
te­nance che­mo­ther­apy—new in can­cer for the most part—thus
rais­ing rea­son­able fears of mor­bid­ity from con­tin­ual, poten­tially
life­
long che­ mo­ther­
apy. Twenty years after imatinib’s record-
break­ing fast approval by the FDA, we take solace in the rel­a­tive
safety of TKIs, led by imatinib, followed by more hes­i­tant con­
fi­dence in the long-term safety of our more potent and sub­se­
quently devel­oped agents. The advent of TFR as an option for
CML has brought the next cross­road: Can we make CML ther­
apy as lim­ited as pos­si­ble, and func­tion­ally cure the can­cer, in
more/many/all­ patients?
TFR stud­ ies have expanded greatly, and data from tri­ als
performed after pri­mary or sec­ond­ary treat­ment with sec­ond-
gen­er­a­tion TKIs are avail­­able. Trials with nilotinib and dasatinib
as sec­ond-line ther­apy after imatinib resis­tance or intol­er­ance,
or as a means to opti­mize deep remis­sion, have proven that TFR
is both fea­si­ble and sim­i­larly suc­cess­ful to a more uncom­pli­
cated TFR after pri­mary imatinib or sec­ond-gen­er­a­tion TKIs.83,84
While over­all rates of TFR suc­cess may not be higher in stud­ies
with sec­ond-gen­er­a­tion TKIs used as front­line ther­apy,84,85 the
rapid­ity of deep remis­sion may afford more rapid eli­gi­bil­ity to
con­sider TFR and thus reduce treat­ment dura­tion/expo­sure and
the poten­tial like­li­hood of toxicities. If TFR were more per­va­sive
and suc­cess­ful and achiev­able on a more global level, this would
change the per­spec­tive on tox­ic­ity for cer­tain as short-term and
life­long risk are strik­ingly dif­fer­ent. The next steps in long-term
tox­ic­ity assess­ment should incor­po­rate crit­i­cal think­ing into
the rel­a­tive ben­e­fits of more potent ther­a­pies and strat­e­gies
cou­pled with the ben­e­fits of lim­it­ing dura­tion of treat­ment for
patients with CML.
CLINICAL CASE (Con­t in­u ed)
Your patient proceeded to ther­apy with a sec­ond-gen­er­a­tion
TKI and achieved rapid and deep response. Major molec­u­lar
response occurred within 6 months, and deep remis­sion (>MR
4, <0.01% IS BCR-ABL) was pres­ent from 9 months through a
total 3 years of treat­ment. She con­tin­ued with her new pri­
mary phy­si­cian and did not develop any major com­pli­ca­tions.
She was delighted with her response and eager to pur­sue TFR
shortly after her 3 years in remis­sion and 2 ¼ years in deep
molec­u­lar remis­sion. She remains off treat­ment in suc­cess­ful
TFR and is eter­nally grate­ful for you man­ag­ing her CML so well
and ensur­ing she had excel­lent pri­mary care and an intro­duc­
tion to car­di­­ol­ogy care.
Key points
• At diag­no­sis/ini­tial pre­sen­ta­tion of CML, appraisal of the pa-
tient’s full comorbidity pro­file is vitally impor­tant and affects
sur­vival more than the CML.
• CML risk strat­i­fi­ca­tion (Sokal, EUTOS Long Term Survival, etc)
affects TKI response and the suc­cess of poten­tial TFR out­
come and may assist with TKI choice.
• Cardiovascular risk assess­ment (Framingham, SCORE, oth­ers)
at diag­no­sis can also assist with TKI choice and can clar­ify the
need and mer­its of any pre­ven­ta­tive inter­ven­tions.
• TKI choice dic­tates gen­eral (class) and drug-spe­cific base­
line and lon­gi­tu­di­nal mon­i­tor­ing for adverse events; TKIs with
more sig­nif­i­cant car­dio­vas­cu­lar risk war­rant reg­u­lar car­dio­
vas­cu­lar mon­i­tor­ing.
• Partnership and coor­di­na­tion with pri­mary care/gen­eral inter­
nal med­i­cine and sub­spe­cialty med­i­cine, espe­cially car­di­­ol­ogy/
cardio-oncol­ogy, serves CML patients best to iden­tify and
man­age adverse events and con­tinue age-appro­pri­ate health
and other can­cer screen­ings.
• TFR as a goal of ther­apy, by lim­it­ing TKI dura­tion, may sig­nif­i­
cantly mod­ify TKI ther­apy risk, but life­long ther­apy remains a
nec­es­sary par­a­digm and mindset in man­ag­ing the increas­ing
num­ber of patients sur­viv­ing CML.
Conflict-of-inter­est dis­clo­sure
Michael J. Mauro: research funding: Novartis, Bristol Myers
Squibb, Sun Pharma/Sparc, Takeda; con­sul­tancy: Novartis, Bris-
tol Myers Squibb, Takeda, Pfizer.
Off-label drug use
Michael J. Mauro: nothing to disclose.
Prakash Singh Shekhawat
Correspondence
Michael J. Mauro, Memorial Sloan Kettering Cancer Center, 1275
York Ave, Box 489, New York, NY 10065; e-mail: maurom@mskcc​
­.org.
References
1. Craddock CF. We do still trans­plant CML, don’t we? Hematology Am Soc
Hematol Educ Program. 2018;2018(1):177-184.
2. Etienne G, Guilhot J, Rea D, et al. Long-term fol­low-up of the French Stop
Imatinib (STIM1) study in patients with chronic mye­loid leu­ke­mia. J Clin
Oncol. 2017;35(3):298-305.
3. Saussele S, Richter J, Guilhot J, et al; EURO-SKI Investigators. Discontin-
uation of tyro­sine kinase inhib­i­tor ther­apy in chronic mye­loid leu­kae­mia
(EURO-SKI): a prespecified interim anal­y­sis of a pro­spec­tive, multicentre,
non-randomised, trial. Lancet Oncol. 2018;19(6):747-757.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/113/1851432/113mauro.pdf by guest on 13 December 2021
4. Rousselot P, Loiseau C, Delord M, Cayuela JM, Spentchian M. Late molec­
u­lar recur­rences in patients with chronic mye­loid leu­ke­mia expe­ri­enc­ing
treat­ment-free remis­sion. Blood Adv. 2020;4(13):3034-3040.
5. Hsieh YC, Kirschner K, Copland M. Improving out­comes in chronic mye­
loid leu­ke­mia through harnessing the immu­no­log­i­cal land­scape. Leukemia.
2021;35(May):1229-1242.
6. Branford S, Kim DDH, Apperley JF, et al; International CML Foundation
Genomics Alliance. Laying the foun­da­tion for genomically-based risk
assess­ment in chronic mye­loid leu­ke­mia. Leukemia. 2019;33(8):1835-1850.
7. Hochhaus A, Larson RA, Guilhot F, et al; IRIS Investigators. Long-term out­
comes of imatinib treat­ment for chronic mye­loid leu­ke­mia. N Engl J Med.
2017;376(10):917-927.
8. Bower H, Björkholm M, Dickman PW, Höglund M, Lambert PC, Anders-
son TML. Life expec­tancy of patients with chronic mye­loid leu­ke­mia
approaches the life expec­tancy of the gen­eral pop­u­la­tion. J Clin Oncol.
2016;34(24):2851-2857.
9. Höglund M, Sandin F, Simonsson B. Epidemiology of chronic mye­loid leu­
kae­mia: an update. Ann Hematol. 2015;94(suppl 2):S241-S247.
10. National Comprehensive Cancer Network. NCCN guide­lines: chronic mye­
loid leu­ke­mia. Accessed 1 June 2021.
11. Shanmuganathan N, Braley JA, Yong ASM, et al. Modeling the safe min­i­mum
fre­quency of molec­u­lar mon­i­tor­ing for CML patients attempting treat­ment-
free remis­sion. Blood. 2019;134(1):85-89.
12. Findakly D, Arslan W. Clinical fea­ tures and out­ comes of patients with
chronic mye­loid leu­ke­mia presenting with iso­lated thrombocytosis: a sys­
tem­atic review and a case from our insti­tu­tion. Cureus. 2020;12(6):e8788.
13. Knöfler R, Lange BS, Paul F, Tiebel O, Suttorp M. Bleeding signs due to
acquired von Willebrand syn­drome at diag­no­sis of chronic mye­loid leu­kae­
mia in chil­dren. Br J Haematol. 2020;188(5):701-706.
14. Abruzzese E, Mauro M, Apperley J, Chelysheva E. Tyrosine kinase inhib­i­tors
and preg­nancy in chronic mye­loid leu­ke­mia: opin­ion, evi­dence, and rec­om­
men­da­tions. Ther Adv Hematol. 2020;11(31 Octo­ber):2040620720966120.
15. Qosa H, Avaritt BR, Hartman NR, Volpe DA. In vitro UGT1A1 inhi­bi­tion by
tyro­sine kinase inhib­i­tors and asso­ci­a­tion with drug-induced hyperbiliru-
binemia. Cancer Chemother Pharmacol. 2018;82(5):795-802.
16. Guilhot F, Hughes TP, Cortes J, et al. Plasma expo­sure of imatinib and its
cor­re­la­tion with clin­i­cal response in the tyro­sine kinase inhib­i­tor opti­mi­za­
tion and selec­tiv­ity trial. Haematologica. 2012;97(5):731-738.
17. Berman E, Nicolaides M, Maki RG, et al. Altered bone and min­eral metab­o­lism
in patients receiv­ing imatinib mesylate. N Engl J Med. 2006;354(19):2006-
2013.
18. Yilmaz M, Lahoti A, O’Brien S, et  al. Estimated glo­mer­u­lar fil­tra­tion rate
changes in patients with chronic mye­loid leu­ke­mia treated with tyro­sine
kinase inhib­i­tors. Cancer. 2015;121(21):3894-3904.
19. Cortes JE, Gambacorti-Passerini C, Kim DW, et al. Effects of bosutinib treat­
ment on renal func­tion in patients with Philadelphia chro­mo­some-pos­i­tive
leu­ke­mias. Clin Lymphoma Myeloma Leuk. 2017;17(10):684-695.e6695e6.
20. Breccia M, Muscaritoli M, Gentilini F, et al. Impaired fasting glu­cose level as
met­a­bolic side effect of nilotinib in non-dia­betic chronic mye­loid leu­ke­mia
patients resis­tant to imatinib. Leuk Res. 2007;31(12):1770-1772.
21. Gómez-Sámano MA, Baquerizo-Burgos JE, Coronel MFC, et  al. Effect of
imatinib on plasma glu­cose con­cen­tra­tion in sub­jects with chronic mye­
loid leu­ke­mia and gas­tro­in­tes­ti­nal stro­mal tumor. BMC Endocr Disord.
2018;18(1):77.
22. Breccia M, Loglisci G, Salaroli A, Serrao A, Alimena G. Nilotinib-medi­ated
increase in fasting glu­cose level is revers­ible, does not con­vert to type 2
Risks with life­long TKI ther­apy in CML  |  119
 dia­be­tes and is likely cor­re­lated with increased body mass index. Leuk Res.
2012;36(4):e66-e67.
23. Iurlo A, Orsi E, Cattaneo D, et  al. Effects of first- and sec­ond-gen­er­a­tion
tyro­sine kinase inhib­ i­
tor ther­apy on glu­ cose and lipid metab­ o­
lism in
chronic mye­loid leu­ke­mia patients: a real clin­i­cal prob­lem? Oncotarget.
2015;6(32):33944-33951.
24. Rea D, Mirault T, Cluzeau T, et al. Early onset hyper­cho­les­ter­­ol­emia induced
by the 2nd-gen­er­a­tion tyro­sine kinase inhib­i­tor nilotinib in patients with
chronic phase-chronic mye­loid leu­ke­mia. Haematologica. 2014;99(7):1197-
1203.
25. Franklin M, Burns L, Perez S, Yerragolam D, Makenbaeva D. Incidence of
type 2 dia­be­tes mellitus and hyper­lip­id­emia in patients pre­scribed dasat-
inib or nilotinib as first- or sec­ond-line ther­apy for chronic mye­log­e­nous
leu­ke­mia in the US. Curr Med Res Opin. 2018;34(2):353-360.
26. Jain P, Kantarjian H, Boddu PC, et al. Analysis of car­dio­vas­cu­lar and arteri-
othrombotic adverse events in chronic-phase CML patients after front­line
TKIs. Blood Adv. 2019;3(6):851-861.
27. Douxfils J, Haguet H, Mullier F, Chatelain C, Graux C, Dogné JM. Association
between BCR-ABL tyro­sine kinase inhib­i­tors for chronic mye­loid leu­ke­mia
and car­dio­vas­cu­lar events, major molec­u­lar response, and over­all sur­vival:
a sys­tem­atic review and meta-anal­y­sis. JAMA Oncol. 2016;2(5):625-632.
28. Cortes JE, Kantarjian HM, Mauro MJ, et  al. Long-term car­diac, vas­cu­lar,
hyper­ten­sion, and effu­sion safety of bosutinib in patients with Philadelphia
chro­mo­some-pos­i­tive leu­ke­mia resis­tant or intol­er­ant to prior ther­apy. Eur
J Haematol. 2021;106(6):808-820.
29. Kantarjian HM, Hughes TP, Larson RA, et al. Long-term out­comes with front­
line nilotinib ver­sus imatinib in newly diag­nosed chronic mye­loid leu­ke­mia
in chronic phase: ENESTnd 10-year anal­y­sis. Leukemia. 2021;35(2):440-453.
30. Hadzijusufovic E, Albrecht-Schgoer K, Huber K, et al. Nilotinib-induced vas-
culopathy: iden­ti­fi­ca­tion of vas­cu­lar endo­the­lial cells as a pri­mary tar­get
site. Leukemia. 2017;31(11):2388-2397.
31. Bocchia M, Galimberti S, Aprile L, et al. Genetic pre­dis­po­si­tion and induced
pro-inflam­ma­tory/pro-oxi­da­tive sta­tus may play a role in increased athero-
thrombotic events in nilotinib treated chronic mye­loid leu­ke­mia patients.
Oncotarget. 2016;7(44):72311-72321.
32. Pouwer MG, Pieterman EJ, Verschuren L, et  al. The BCR-ABL1 inhib­i­tors
imatinib and ponatinib decrease plasma cho­les­terol and ath­ero­scle­ro­sis,
and nilotinib and ponatinib acti­vate coag­u­la­tion in a trans­la­tional mouse
model. Front Cardiovasc Med. 2018;5(12 June):55.
33. Gover-Proaktor A, Granot G, Pasmanik-Chor M, et al. Bosutinib, dasatinib,
imatinib, nilotinib, and ponatinib dif­fer­en­tially affect the vas­cu­lar molec­u­lar
path­ways and func­tion­al­ity of human endo­the­lial cells. Leuk Lymphoma.
2019;60(1):189-199.
34. Sant’Antonio E, Camerini C, Rizzo V, Musolino C, Allegra A. Genetic het­ero­
ge­ne­ity in chronic mye­loid leu­ke­mia: how clonal hema­to­poi­e­sis and clonal
evo­lu­tion may influ­ence prog­no­sis, treat­ment out­come, and risk of car­dio­
vas­cu­lar events. Clin Lymphoma Myeloma Leuk. 2021;21(9):573-579.
35. Quintás-Cardama A, Kantarjian H, O’Brien S, et  al. Pleural effu­ sion in
patients with chronic mye­log­e­nous leu­ke­mia treated with dasatinib after
imatinib fail­ure. J Clin Oncol. 2007;25(25):3908-3914.
36. Hughes TP, Laneuville P, Rousselot P, et  al. Incidence, out­comes, and risk
fac­tors of pleu­ral effu­sion in patients receiv­ing dasatinib ther­apy for Philadel-
phia chro­mo­some-pos­i­tive leu­ke­mia. Haematologica. 2019;104(1):93-101.
37. Shah NP, Rousselot P, Schiffer C, et  al. Dasatinib in imatinib-resis­tant or
-intol­er­ant chronic-phase, chronic mye­loid leu­ke­mia patients: 7-year fol­low-
up of study CA180-034. Am J Hematol. 2016;91(9):869-874.
38. Cortes JE, Saglio G, Kantarjian HM, et al. Final 5-year study results of DASI-
SION: the dasatinib ver­sus imatinib study in treat­ment-naïve chronic mye­
loid leu­ke­mia patients trial. J Clin Oncol. 2016;34(20):2333-2340.
39. Naqvi K, Jabbour E, Skinner J, et al. Long-term fol­low-up of lower dose dasat-
inib (50 mg daily) as front­ line ther­
apy in newly diag­ nosed chronic-phase
chronic mye­loid leu­ke­mia. Cancer. 2020;126(1):67-75.
40. Cortes JE, Jimenez CA, Mauro MJ, Geyer A, Pinilla-Ibarz J, Smith BD. Pleu-
ral effu­sion in dasatinib-treated patients with chronic mye­loid leu­ke­mia in
chronic phase: iden­ti­fi­ca­tion and man­age­ment. Clin Lymphoma Myeloma
Leuk. 2017;17(2):78-82.
41. Mustjoki S, Ekblom M, Arstila TP, et al. Clonal expan­sion of T/NK-cells dur­
ing tyro­sine kinase inhib­i­tor dasatinib ther­apy. Leukemia. 2009;23(8):1398-
1405.
42. Mustjoki S, Auvinen K, Kreutzman A, et al. Rapid mobi­li­za­tion of cyto­toxic
lym­pho­cytes induced by dasatinib ther­apy. Leukemia. 2013;27(4):914-924.
120  |  Hematology 2021  |  ASH Education Program
43. Ozawa MG, Ewalt MD, Gratzinger D. Dasatinib-related fol­lic­ul­ar hyper­pla­
sia: an underrecognized entity with char­ac­ter­is­tic mor­phol­ogy. Am J Surg
Pathol. 2015;39(10):1363-1369.
44. Schiffer CA, Cortes JE, Hochhaus A, et al. Lymphocytosis after treat­ment
with dasatinib in chronic mye­loid leu­ke­mia: effects on response and tox­ic­
ity. Cancer. 2016;122(9):1398-1407.
45. Weatherald J, Bondeelle L, Chaumais MC, et al. Pulmonary com­pli­ca­tions
of Bcr-Abl tyro­sine kinase inhib­i­tors. Eur Respir J. 2020;56(4):2000279.
46. Weatherald J, Chaumais MC, Savale L, et al. Long-term out­comes of dasat-
inib-induced pul­mo­nary arte­rial hyper­ten­sion: a pop­u­la­tion-based study.
Eur Respir J. 2017;50(1):1700217.
47. Cortes JE, Kantarjian H, Shah NP, et al. Ponatinib in refrac­tory Philadelphia
chro­mo­some-pos­i­tive leu­ke­mias. N Engl J Med. 2012;367(22):2075-2088.
48. Cortes JE, Kim DW, Pinilla-Ibarz J, et al; PACE Investigators. A phase 2 trial
of ponatinib in Philadelphia chro­mo­some-pos­i­tive leu­ke­mias. N Engl J Med.
2013;369(19):1783-1796.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/113/1851432/113mauro.pdf by guest on 13 December 2021
49. Latifi Y, Moccetti F, Wu M, et al. Thrombotic microangiopathy as a cause of
car­dio­vas­cu­lar tox­ic­ity from the BCR-ABL1 tyro­sine kinase inhib­i­tor ponati-
nib. Blood. 2019;133(14):1597-1606.
50. Hochhaus A, Baccarani M, Silver RT, et  al. Euro­pean LeukemiaNet 2020
rec­om­men­da­tions for treating chronic mye­loid leu­ke­mia. Leukemia. 2020;
34(4):966-984.
51. Cortes J, Apperley J, Lomaia E, et  al. OPTIC pri­ mary anal­ y­sis: a dose-
opti­mi­za­tion study of 3 starting doses of ponatinib (PON). J Clin Oncol.
2021;39(suppl 15):7000.
52. Januzzi JL, Garasic J, Kasner S, et al. An inde­pen­dent review of arte­rial occlu­
sive events (AOEs) in the ponatinib (PON) phase II PACE trial (NCT01207440)
in patients (pts) with Ph+ leu­ke­mia. J Clin Oncol. 2020;38(suppl 15):7550.
53. Hughes TP, Mauro MJ, Cortes JE, et al. Asciminib in chronic mye­loid leu­ke­
mia after ABL kinase inhib­i­tor fail­ure. N Engl J Med. 2019;381(24):2315-2326.
54. Hochhaus A, Boquimpani B, Réa D, et al. Efficacy and safety results from
ASCEMBL, a mul­ti­cen­ter, open-label, phase 3 study of asciminib, a first-
in-class STAMP inhib­i­tor, vs bosutinib in patients with chronic mye­loid
leu­ke­mia in chronic phase pre­vi­ously treated with ≥2 tyro­sine kinase
inhib­i­tors. Blood. 2020;136(suppl 2):LBA-4.
55. Wylie AA, Schoepfer J, Jahnke W, et al. The allo­ste­ric inhib­i­tor ABL001 enables
dual targeting of BCR-ABL1. Nature. 2017;543(7647):733-737.
56. Mendizabal AM, Younes N, Levine PH. Geographic and income var­i­a­tions in
age at diag­no­sis and inci­dence of chronic mye­loid leu­ke­mia. Int J Hematol.
2016;103(1):70-78.
57. Patel M, Philip V, Fazel F, et al. Human immu­no­de­fi­ciency virus infec­tion and
chronic mye­loid leu­ke­mia. Leuk Res. 2012;36(11):1334-1338.
58. Jabbour E, Makenbaeva D, Lingohr-Smith M, Lin J. Use of real-world claim
data­bases to assess prev­a­lence of comorbid con­di­tions rel­e­vant to the
treat­ment of chronic mye­log­e­nous leu­ke­mia based on National Compre-
hensive Network Treatment Guidelines. Clin Lymphoma Myeloma Leuk.
2015;15(12):797-802.
59. Hoffmann VS, Baccarani M, Hasford J, et al. The EUTOS pop­u­la­tion-based reg­
is­try: inci­dence and clin­i­cal char­ac­ter­is­tics of 2904 CML patients in 20 Euro­
pean countries. Leukemia. 2015;29(6):1336-1343.
60. Goldberg SL, Cortes JE, Gambacorti-Passerini C, et  al. First-line treat­
ment selec­tion and early mon­i­tor­ing pat­terns in chronic phase-chronic
mye­loid leu­ke­mia in rou­tine clin­i­cal prac­tice: SIMPLICITY. Am J Hematol.
2017;92(11):1214-1223.
61. Saussele S, Krauss MP, Hehlmann R, et al; Schweizerische Arbeitsgemein-
schaft für Klinische Krebsforschung; Ger­man CML Study Group. Impact of
comorbidities on over­all sur­vival in patients with chronic mye­loid leu­ke­
mia: results of the ran­dom­ized CML study IV. Blood. 2015;126(1):42-49.
62. Ono T, Takahashi N, Kizaki M, et  al. Prognostic effect of comorbidities in
patients with chronic mye­ loid leu­ ke­ mia treated with a tyro­ sine kinase
inhib­i­tor. Cancer Sci. 2020;111(10):3714-3725.
63. Takahashi N, Miura M, Niioka T, Sawada K. Influence of H2-recep­tor antag­o­
nists and pro­ton pump inhib­i­tors on dasatinib phar­ma­co­ki­net­ics in Jap­a­nese
leu­ke­mia patients. Cancer Chemother Pharmacol. 2012;69(4):999-1004.
64. Barber MC, Mauro MJ, Moslehi J. Cardiovascular care of patients with
chronic mye­loid leu­ke­mia (CML) on tyro­sine kinase inhib­i­tor (TKI) ther­apy.
Hematology Am Soc Hematol Educ Program. 2017;2017(1):110-114.
65. Seguro FS, Silva CMPDC, Moura CMB, et al. Recommendations for the man­
age­ment of car­dio­vas­cu­lar risk in patients with chronic mye­loid leu­ke­mia
on tyro­sine kinase inhib­i­tors: risk assess­ment, strat­i­fi­ca­tion, treat­ment and
mon­i­tor­ing. Hematol Transfus Cell Ther. 2021;43(2):191-200.
66. Kondapalli L, Worth S, Hawi R, et al. Collaborative car­dio­vas­cu­lar man­age­
ment of patients with chronic mye­loid leu­ke­mia on tyro­sine kinase inhib­i­
tors. Vasc Med. 2020;25(3):246-254.
67. Medeiros BC, Possick J, Fradley M. Cardiovascular, pul­mo­nary, and met­a­
bolic toxicities com­pli­cat­ing tyro­sine kinase inhib­i­tor ther­apy in chronic
mye­loid leu­ke­mia: strat­e­gies for mon­i­tor­ing, detecting, and man­ag­ing.
Blood Rev. 2018;32(4):289-299.
68. Casavecchia G, Galderisi M, Novo G, et al. Early diag­no­sis, clin­i­cal man­age­
ment, and fol­low-up of car­dio­vas­cu­lar events with ponatinib. Heart Fail
Rev. 2020;25(3):447-456.
69. Saussele S, Haverkamp W, Lang F, et  al. Ponatinib in the treat­ment of
chronic mye­loid leu­ke­mia and Philadelphia chro­mo­some-pos­i­tive acute
leu­ke­mia: rec­om­men­da­tions of a Ger­man expert con­sen­sus panel with
focus on car­dio­vas­cu­lar man­age­ment. Acta Haematol. 2020;143(3):217-
231.
70. Valent P, Hadzijusufovic E, Hoermann G, et al. Risk fac­tors and mech­a­nisms
con­trib­ut­ing to TKI-induced vas­cu­lar events in patients with CML. Leuk
Res. 2017;59(August):47-54.
71. Aghel N, Delgado DH, Lipton JH. Cardiovascular events in chronic mye­loid
leu­ke­mia clin­i­cal tri­als. Is it time to reassess and report the events accord-
ing to car­di­­ol­ogy guide­lines? Leukemia. 2018;32(10):2095-2104.
72. Framingham Heart Study. Hard cor­o­nary heart dis­ease. Accessed 18 Octo­ber
2021.
73. D’Agostino RB Sr, Vasan RS, Pencina MJ, et al. General car­dio­vas­cu­lar risk
pro­file for use in pri­mary care: the Framingham Heart Study. Circulation.
2008;117(6):743-753.
74. Euro­pean Society of Cardiology. Score risk charts: the Euro­pean car­dio­vas­
cu­lar dis­ease risk assess­ment mod­els. Accessed 18 Octo­ber 2021.
75. Rea D, Mirault T, Raffoux E, et  al. Usefulness of the 2012 Euro­pean CVD
risk assess­ment model to iden­tify patients at high risk of car­dio­vas­cu­lar
events dur­ing nilotinib ther­apy in chronic mye­loid leu­ke­mia. Leukemia.
2015;29(5):1206-1209.
76. Breccia M, Molica M, Zacheo I, Serrao A, Alimena G. Application of sys­tem­
atic cor­o­nary risk eval­u­a­tion chart to iden­tify chronic mye­loid leu­ke­mia
patients at risk of car­dio­vas­cu­lar dis­eases dur­ing nilotinib treat­ment. Ann
Hematol. 2015;94(3):393-397.
77. Hippisley-Cox J, Coupland C, Brindle P. Development and val­i­da­tion of
QRISK3 risk pre­dic­tion algorithms to estimate future risk of cardiovascular
disease: prospective cohort study. BMJ. 2017;357: j2099.
Prakash Singh Shekhawat
78. Mauro MJ, Oehler VG, Thompson JE, et  al. Cardiovascular and met­a­bolic
risk in patients with chronic mye­loid leu­ke­mia in chronic phase receiv­ing
first-line BCR-ABL1 tyro­sine kinase inhib­i­tors in the United States: base­line
and six-month fol­low-up results from a pro­spec­tive real-world obser­va­
tional study. Blood. 2020;136(suppl 1):39-40.
79. Lyon AR, Dent S, Stanway S, et al. Baseline car­dio­vas­cu­lar risk assess­ment in
can­cer patients sched­uled to receive cardiotoxic can­cer ther­a­pies: a posi­
tion state­ment and new risk assess­ment tools from the Cardio-Oncology
Study Group of the Heart Failure Association of the Euro­pean Society of
Cardiology in col­lab­o­ra­tion with the International Cardio-Oncology Soci-
ety. Eur J Heart Fail. 2020;22(11):1945-1960.
80. Manouchehri A, Kanu E, Mauro MJ, Aday AW, Lindner JR, Moslehi J. Tyrosine
kinase inhib­i­tors in leu­ke­mia and car­dio­vas­cu­lar events: from mech­a­nism
to patient care. Arterioscler Thromb Vasc Biol. 2020;40(2):301-308.
81. Steegmann JL, Baccarani M, Breccia M, et al. Euro­pean LeukemiaNet rec­
om­men­da­tions for the man­age­ment and avoid­ance of adverse events of
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/113/1851432/113mauro.pdf by guest on 13 December 2021
treat­ment in chronic mye­loid leu­kae­mia. Leukemia. 2016;30(8):1648-1671.
82. Caocci G, Mulas O, Abruzzese E, et al. Arterial occlu­sive events in chronic
mye­loid leu­ke­mia patients treated with ponatinib in the real-life prac­tice
are predicted by the Systematic Coronary Risk Evaluation (SCORE) chart.
Hematol Oncol. 2019;37(3):296-302.
83. Hughes TP, Clementino NCD, Fominykh M, et al. Long-term treat­ment-free
remis­sion in patients with chronic mye­loid leu­ke­mia after sec­ond-line nilo-
tinib: ENESTop 5-year update. Leukemia. 2021;35(6):1631-1642.
84. Shah NP, García-Gutiérrez V, Jiménez-Velasco A, et  al. Dasatinib dis­con­
tin­u­a­tion in patients with chronic-phase chronic mye­loid leu­ke­mia and
sta­ble deep molec­u­lar response: the DASFREE study. Leuk Lymphoma.
2020;61(3):650-659.
85. Radich JP, Hochhaus A, Masszi T, et al. Treatment-free remis­sion fol­low­ing
front­line nilotinib in patients with chronic phase chronic mye­loid leu­ke­mia:
5-year update of the ENESTfreedom trial. Leukemia. 2021;35(5):1344-1355.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000239
Risks with life­long TKI ther­apy in CML  |  121
 CML: SUCCESS BREEDS MORE SUCCESS

Blast and accelerated phase CML:

room for improvement

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/122/1851605/122how.pdf by guest on 13 December 2021

Joan How,1–3 Vinayak Venkataraman,1 and Gabriela Soriano Hobbs1
1
Department of Medical Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, MA; 2Division of Hematology, Department
of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA; and 3Department of Medical Oncology, Dana­Farber Cancer
Institute, Harvard Medical School, Boston, MA

Tyrosine kinase inhibitors (TKIs) revolutionized the treatment of chronic myeloid leukemia (CML). With TKI therapy, the
percentage of patients who progress to accelerated phase (AP) or blast phase (BP) CML has decreased from more than
20% to 1% to 1.5% per year. Although AP- and BP-CML occur in a minority of patients, outcomes in these patients are
significantly worse compared with chronic phase CML, with decreased response rates and duration of response to TKI.
Despite this, TKIs have improved outcomes in advanced phase CML, particularly in de novo AP patients, but are often
inadequate for lasting remissions. The goal of initial therapy in advanced CML is a return to a chronic phase followed by
consideration for bone marrow transplantation. The addition of induction chemotherapy with TKI is often necessary for
achievement of a second chronic phase. Given the small population of patients with advanced CML, development of
novel treatment strategies and investigational agents is challenging, although clinical trial participation is encouraged
in AP and BP patients, whenever possible. We review the overall management approach to advanced CML, including TKI
selection, combination therapy, consideration of transplant, and novel agents.

LEARNING OBJECTIVES
• Understand the treatment approach to accelerated and blast phase CML in de novo disease and occurring in
patients with underlying CP­CML
• Understand response rates and outcomes in patients with advanced CML treated with TKI, TKI and chemotherapy,
and hematopoietic stem cell transplantation

blasts, but M351T and T3151 mutations and KMT2A par­

CLINICAL CASE
NM is a 54­year­old woman diagnosed with chronic mye­
loid leukemia (CML) in chronic phase (CP), after present­
ing with a white blood cell (WBC) count of 270 × 109/L,
hemoglobin of 7.5 g/dL, and platelets of 100 × 109/L. She
was initiated on imatinib, which she took irregularly and
eventually discontinued. She established care with a new
physician and resumed imatinib when she was found to
have a WBC count of 68 × 109/L, hemoglobin of 10.4 g/dL,
and platelets of 129 × 109/L. After initial response, periph­
eral blood polymerase chain reaction for BCR­ABL rose
to 35% after 6 months. A bone marrow biopsy specimen
demonstrated CP­CML, and she was switched to dasat­
inib 100 mg daily. After 7 months of treatment on dasati­
nib, BCR­ABL rose from less than 1% to 8.013%. A repeat
bone marrow biopsy specimen did not show increased
tial duplication were identified. She was started on pona­
tinib 45 mg daily. Four months after ponatinib initiation,
polymerase chain reaction for BCR­ABL rose to more than
50%, and bone marrow biopsy specimen showed 25%
myeloblasts along with rising peripheral WBC count. She
was transferred to our hospital and received cytarabine
200 mg/m2 intravenously daily for 7 days and idarubicin
12 mg/m2 for 3 days. Ponatinib was resumed at the end
of induction. She went into a second CP and underwent
haploidentical stem cell transplant with posttransplant
cyclophosphamide. She received fludarabine, melphalan,
and total body irradiation for conditioning. She remains
in complete molecular remission 9 months after trans­
plant and has discontinued tyrosine kinase inhibitor (TKI)
therapy.

122 | Hematology 2021 | ASH Education Program

Introduction
Chronic mye­loid leu­ke­mia is a mye­lo­pro­lif­er­a­tive neo­plasm
char­ac­ter­ized by the Philadelphia chro­mo­some (Ph) that affects
1 to 2 per 100 000 new patients per year and com­prises 15% of
leu­ke­mias in adults.1 The dis­ease is driven by a recip­ro­cal trans­
lo­ca­tion of chro­mo­somes 9 and 22, which results in the BCR-
ABL fusion pro­tein and dysregulated tyro­sine kinase activ­ity.
Patients most com­monly pres­ent in CP, but with­out treat­ment,
CP-CML will prog­ress to accel­er­ated phase (AP-CML) and blast
cri­sis (BP-CML) within 3 to 5 years. Tyrosine kinase inhib­i­tors
rev­o­lu­tion­ized the care of CML with the approval of imatinib,
the only first-gen­er­a­tion TKI. Three sec­ond-gen­er­a­tion TKIs, da­
satinib, nilotinib, and bosutinib, are also avail­­able for front­line
use. Progression to advanced CML is due to con­tin­ued BCR-
ABL activ­ity, which results in not only con­tin­ued pro­lif­er­a­tion of
leu­ke­mic cells but fur­ther genetic insta­bil­ity and DNA dam­age.
This invari­ably leads to clonal evo­lu­tion and muta­tions both in­
side and out­side the BCR-ABL kinase domain, as well as addi­
tional chro­mo­somal abnor­mal­i­ties (ACAs). This review focuses
on treat­ment con­sid­er­ations for patients who have or prog­ress
to AP- and BP-CML, includ­ing TKI selec­tion, con­sid­er­ation, and
tim­ing of hema­to­poi­etic stem cell trans­plan­ta­tion (HSCT), and
emerg­ing ther­a­pies.
Definition and epi­de­mi­­ol­ogy of advanced phase CML
The over­all inci­dence of AP- and BP-CML at diag­no­sis is 3.5%
and 2.2%,2 respec­tively, and with the intro­duc­tion of TKIs, the
num­ber has decreased sig­nif­i­cantly. Long-term fol­low-up of the
International Randomized Study of Interferon and STI571 trial
dem­on­strated 6.9% cumu­la­tive pro­gres­sion to AP- or BP-CML
after 10 years.3 This num­ber is lower in recent long-term stud­
ies, likely due to improve­ments in the man­age­ment of patients
with CP-CML with an inad­e­quate response.4 A lower inci­dence of
Table 1.  Major clas­si­fi­ca­tion sys­tems used in chronic mye­loid leu­ke­mia
MDACC IBMTR
Accelerated phase
  PB blasts 15%-29% PB or BM blasts 10%-29%
  PB blasts + promyelocytes ≥30% BP blasts + promyelocytes >20%
  PB baso­phils ≥20% PB baso­phils ≥20%
  Platelets ≤100 × 109/L (unre­lated Platelets ≤100 × 109/L (unre­lated
to ther­apy) to ther­apy) or >1000 × 109/L
  Splenomegaly (unre­spon­sive to (unre­spon­sive to ther­apy)
ther­apy) Anemia Hb <8 g/dL (unre­spon­sive
to ther­apy)
Splenomegaly (unre­spon­sive to
ther­apy)
  Cytogenetic evo­lu­tion on Cytogenetic evo­lu­tion on treat­
treat­ment ment
Blast phase
  PB or BM blasts ≥30%
  Extramedullary blast pro­lif­er­a­tion
Prakash Singh Shekhawat
BM, bone mar­row; Hb, hemo­glo­bin; PB, periph­eral blood.
trans­for­ma­tion is asso­ci­ated with ini­tial treat­ment with sec­ond-
gen­er­a­tion TKIs com­pared with imatinib.4
Complicating the epi­de­mi­­ol­ogy and treat­ment of advanced
phase CML are the dif­fer­ent clas­si­fi­ca­tion sys­tems used to define
AP- and BP-CML (Table 1), which include the International Blood
and Marrow Transplant Registry (IBMTR),5 M. D. Anderson Can­
cer Center (MDACC),6 Euro­pean LeukemiaNet,7 and World Health
Organization (WHO) cri­te­ria.8 One major dif­fer­ence between clas­
si­fi­ca­tion sys­tems is the thresh­old blast per­cent­age used to dis­tin­
guish CP-, AP-, and BP-CML, with the WHO defin­ing BP as a blast
per­cent­age of more than 20% and all­other clas­si­fi­ca­tion sys­tems
using a thresh­old of more than 30%. Of note, the acqui­si­tion of
major-route ACAs on treat­ment is con­sid­ered a hall­mark of AP-CML,
and the WHO also con­sid­ers the pres­ence of major-route ACAs at
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/122/1851605/122how.pdf by guest on 13 December 2021
diag­no­sis as diag­nos­tic for AP-CML.8 The WHO has also recently
included pro­vi­sional cri­te­ria based on ini­tial response to TKIs (Table
1), which will require fur­ther val­i­da­tion in pro­spec­tive tri­als.8
The vary­ing clas­si­fi­ca­tion sys­tems and their def­i­ni­tions of
advanced CML must be kept in mind when interpreting and apply­
ing trial results to indi­vid­ual patients. The MDACC and IBMTR cri­
te­ria are more fre­quently used as eli­gi­bil­ity cri­te­ria in clin­i­cal tri­als,
and National Comprehensive Cancer Network (NCCN) guide­lines
dis­fa­vor use of the WHO clas­si­fi­ca­tion sys­tem for this rea­son.9
Overall, the lack of uni­for­mity seen in these major clas­si­fi­ca­tion
sys­tems empha­sizes a clin­i­cal spec­trum within each phase of the
dis­ease. Recommendations for treat­ment in AP- or BP-CML are
there­fore rarely one-size-fits-all­.
Treatment of AP-CML
The ini­tial goal of ther­apy in advanced phase CML is to revert to a
CP or a remis­sion prior to HSCT. The hema­to­logic and com­plete
cyto­ge­netic responses (CCyRs) to the var­i­ous TKIs in advanced
CML are sum­ma­rized in Table 2.
Euro­pean LeukemiaNet WHO
PB or BM blasts 15%-29% PB or BM blasts 10%-19%
PB blasts + promyelocytes ≥30%
PB baso­phils ≥20% PB baso­phils ≥20%
Platelets ≤100 × 109/L (unre­lated Platelets ≤100 × 109/L (unre­lated to
to ther­apy) ther­apy) or >1000 × 109/L
(unre­spon­sive to ther­apy)
Splenomegaly (unre­spon­sive to
ther­apy)
Cytogenetic evo­lu­tion on ACA/Ph+ major route, com­plex
treat­ment kar­yo­type, or 3q26.2 abnor­mal­i­ties,
at diag­no­sis
Cytogenetic evo­lu­tion on treat­ment
Provisional: fail­ure to achieve CHR to
first TKI; any indi­ca­tion of resis­tance
to 2 sequen­tial TKIs; occur­rence of
>2 muta­tions on BCR-ABL dur­ing TKI
PB or BM blasts ≥20%
Extramedullary blast pro­lif­er­a­tion
Advanced phase CML: room for improve­ment | 123
Table 2.  Summary of hema­to­logic and cyto­ge­netic responses to TKI in advanced phase CML*

CHR CCyR MMR OS

AP BP AP BP AP AP BP
Imatinib 70%- > 90% 11%-35% 16%-60% 0%-10% 19%-63% 50%-60% at 5 years 7-10 months†
Nilotinib (de novo) >90% — 80%-90% — 70% 90% at 3 years
Nilotinib (progressed) 22%-46% 21%-42% ‡
0%-21% 14%-38% §
10% 60%-70% at 2 years 32% at 2 years
Dasatinib (de novo) >90% — 80%-90% — 70% 90% at 3 years
Dasatinib (progressed) 45%-52% 28%-61%‡ 18%-33% 27%-35%§ 10% 60%-70% at 2 years 30% at 2 years
Bosutinib (progressed) 57%‡ 28%‡ 40%§ 50%§ 11% 60% at 4 years 17% at 4 years
Ponatinib (progressed) 55% ‡
32% ‡
24% 18% §
34% 84% at 1 year 29% at 1 year

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/122/1851605/122how.pdf by guest on 13 December 2021

*Adapted from Bonifacio et al.10
†
Median OS.
‡
Hematologic response.
§
Major cyto­ge­netic response.

In patients presenting with de novo AP-CML, responses to
TKIs are robust. Imatinib results in CCyRs of 60% to 80% and
major molec­u­lar responses (MMRs) of 40% to 60%,11,12 and these
responses are fur­ther improved in de novo AP patients whose
only hall­mark of AP is the pres­ence of ACAs. Second-gen­er­a­tion
TKIs have greater potency and BCR-ABL selec­tiv­ity com­pared
with imatinib. CCyRs and MMRs were 90% and 76% with nilotinib
and dasatinib, respec­tively, in newly diag­nosed AP patients.11
Most patients were diag­nosed with AP based on iso­lated clonal
evo­lu­tion or baso­philia, again dem­on­strat­ing how het­ero­ge­ne­
ity in clas­si­fi­ca­tion sys­tems makes inter­pre­ta­tion of responses
with dif­fer­ent inter­ven­tions chal­leng­ing. Consistent with NCCN
guide­lines, we approach ini­tial man­age­ment of patients with de
novo AP-CML sim­i­larly to patients with CP-CML, espe­cially if they
have low-risk Sokal scores or iso­lated ACA/Ph+ abnor­mal­i­ties
as their only AP fea­ture.9 Of note, cer­tain ACA abnor­mal­i­ties are
higher risk than oth­ers, which may affect treat­ment deci­sions
(Table 3). Milestones are not well defined in AP-CML, but mon­
i­
tor­ing of BCR-ABL is recommended at 3-month inter­ vals, as
in CP. Failure to achieve mile­stones as used in CP-CML should
prompt con­sid­er­ation change in ther­apy and HSCT.
Prognosis, how­ever, is worse in AP patients who prog­ress from
CP while on treat­ment. For these patients, CCyRs and com­plete
Table 3.  Prognostic risk fac­tors in advanced phase CML
Characteristic Poor risk fac­tors
Clinical13,14
Blast % (most impor­tant prog­nos­tic indi­ca­tor,
greater impact in AP com­pared with BP)
Older age
Anemia
Thrombocytopenia
Basophil %
Prior TKI
Myeloid immunophenotype
Chromosomal13,15-17 +8, +Ph, i(17q), +17, +19, +21, 3q26.2, 11q23, −7/7q,
com­plex, del17p, hyperdiploidy, chro­mo­some 15
abnor­mal­i­ties
Molecular15,18 TP53
ASXL1
Acquisition of new muta­tions dur­ing TKI treat­ment
(ABL1 kinase muta­tions, TP53, KMT2D, TET2)
hema­to­logic remis­sion (CHR) rates are 10% to 20% and 20% to
30% for nilotinib,19 30% and 50% for dasatinib,20 and 40% and 57%
for bosutinib.21 Ponatinib is a third-gen­er­a­tion BCR-ABL inhib­i­tor
that over­ comes many muta­ tions that con­fer resis­tance to ear­
lier-gen­er­a­tion TKIs, includ­ing the gate­keeper T315I muta­tion. The
Ponatinib Ph+ ALL and CML Evaluation trial enrolled 267 heavily pre­
treated patients with CML, includ­ing 83 patients in AP.22 Ponatinib
resulted in CCyR and CHR rates of 24% and 55% in AP patients,
respec­ tively, with median dura­ tion of response last­ ing 12.9
months.23 However, arte­rial and venous throm­bo­sis were fre­quent
adverse events (AEs), lead­ing to dis­con­tin­u­a­tion for severe AEs in
11% of patients. The over­all cumu­la­tive inci­dence of AEs was 25%
in the 5-year fol­low-up, includ­ing 21% seri­ous AEs.23 The Ponatinib
in Participants with Resistant Chronic Phase Chronic Myeloid Leu­
kemia to Characterize the Efficacy and Safety of a Range of Doses
(OPTIC) trial eval­u­ated lower starting doses of ponatinib and de-
esca­lated doses once BCR-ABL lev­els reached 1%, and lon­ger-term
fol­low-up of this cohort may sup­port alter­nate dos­ing of ponatinib
to improve tol­er­a­bil­ity and effi­cacy.24 Omecataxine is a pro­tein syn­
the­sis inhib­i­tor with approval by the US Food and Drug Administra­
tion (FDA) for AP-CML resis­tant or intol­er­ant to TKIs, includ­ing the
T315I muta­tion. In a phase 2 trial of heavily pretreated patients with
CML, among the 51 AP patients, 29% achieved CHR and 4% CCyR.25
These responses are less than that seen with ponatinib but can be
con­sid­ered in this patient pop­u­la­tion with few remaining options.
Ultimately, selec­tion of a TKI is affected by patient comorbid­
ity, costs, prior treat­ment, and BCR-ABL muta­tional sta­tus. Better
def­i­ni­tions of AP-CML are needed with more refined risk strat­i­fic ­ a­
tion and treat­ment indi­ca­tions, as cer­tain patients with de novo
AP-CML behave sim­i­lar to those with CP-CML. Table 3 sum­ma­rizes
clin­i­cal, chro­mo­somal, and molec­u­lar risk fac­tors that, if pres­ent in
AP patients, may war­rant more aggres­sive approaches. Patients
with AP-CML with excess blasts often need treat­ment such as BP-
CML. Given the higher response rates, later-gen­er­a­tion TKIs are
pre­ferred over imatinib for de novo AP, and a switch in TKI is war­
ranted for patients progressing from CP; if appli­ca­ble, selec­tion
should be made based on ABL1 kinase domain muta­tions. Pona­
tinib is the only FDA-approved TKI with activ­ity against the T315I
muta­tion, although it car­ries a risk of vas­cu­lar occlu­sive events.
Although our patient did not dem­on­strate overt signs of AP-CML,
pro­gres­sion on imatinib and then dasatinib is concerning. Given

124  |  Hematology 2021  |  ASH Education Program

the pres­ence of the T315I muta­tion while tak­ing dasatinib, she
was switched to ponatinib for fur­ther man­age­ment.
Treatment of BP-CML
Chemotherapy in addi­tion to a TKI is gen­er­ally recommended in
patients with BP-CML, with the type of induc­tion che­mo­ther­apy
guided by the mye­loid or lym­phoid lin­e­age of the blasts. Induc­
tion che­mo­ther­apy is recommended in con­junc­tion with TKI as re­
sponse rates asso­ci­ated with TKI treat­ments alone are inad­e­quate.
Response to sin­gle-agent imatinib is sig­nif­i­cantly lower in BP-CML,
with CCyRs occur­ring in approx­i­ma­tely 10% of patients and median
over­all sur­vival (OS) around 7 to 10 months.26 Responses are slightly
improved with sec­ond gen­er­a­tion TKIs.27,28 In the Ponatinib Ph+ ALL
and CML Evaluation trial, ponatinib dem­on­strated an 18% CCyR in
were too small to make a defin­i­tive com­par­i­son, dasatinib led to
improved out­comes com­pared with imatinib. Rates of CCyR and
MMR were also sig­nif­i­cantly higher with TKI and che­mo­ther­apy
com­bi­na­tion.13 However, intense ther­apy may not be fea­si­ble in
all­patients. TKIs with hypomethylating agents have dem­on­strated
some effi­cacy and OS ben­e­fit.32,33
Overall, BP-CML is a rare entity, and defin­ing uni­fied treat­ment
guide­lines is chal­leng­ing. However, patients with high blast bur­
den almost always need treat­ment with stan­dard che­mo­ther­apy
in addi­tion to TKIs. For the rare patient with de novo blast cri­sis,
some experts con­sider use of TKI alone with close mon­i­tor­ing,
although there are lit­tle data to guide this approach.10 Patients
who are unfit for che­mo­ther­apy and with lower blast bur­den may
also be can­di­dates for TKI alone with close mon­i­tor­ing. In the

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/122/1851605/122how.pdf by guest on 13 December 2021

62 patients with BP-CML and a median dura­tion of response of 6
months.22 Of note, nilotinib is not FDA approved for BP-CML.
Multiple che­mo­ther­apy reg­i­mens have been explored in com­bi­
na­tion with TKIs in small, ret­ro­spec­tive stud­ies.13 Cytarabine-based
reg­i­mens, includ­ing 7-day con­tin­u­ous infu­sion of cytarabine
200 mg/m2 and dau­no­ru­bi­cin 60 mg/m2 on days 1 to 3 (7 + 3)29
and fludarabine, cytarabine, granulocyte col­ony-stim­u­lat­ing fac­
tor, and idarubicin, are com­monly used for CML in mye­loid blast
cri­sis.30 Hyperfractionated cyclo­phos­pha­mide, vin­cris­tine, doxo­
ru­bi­cin, and dexa­meth­a­sone in com­bi­na­tion with dasatinib has
dem­on­strated effi­cacy in lym­phoid blast cri­sis.31 Cytogenetic
responses of approx­i­ma­tely 30% are seen with mye­loid induc­
tion reg­i­mens in com­bi­na­tion with imatinib.29 A ret­ro­spec­tive
anal­y­sis of 477 patients with BP-CML treated with che­mo­ther­apy,
TKI and che­mo­ther­apy, or non-TKI-based ther­apy dem­on­strated
that patients treated with TKI in com­bi­na­tion with che­mo­ther­apy
had supe­rior sur­vival com­pared with TKI alone or non-TKI-based
ther­apy (5-year OS, 30% vs 14% vs 9%).13 Although the num­bers
case of our patient, she was treated with stan­dard 7 + 3 induc­tion
che­mo­ther­apy and ponatinib given the T315I muta­tion. We also
use higher doses of TKI for BP-CML com­pared with CP-CML, con­
sis­tent with FDA label­ing indi­ca­tions.
HSCT in advanced CML
HSCT is an impor­tant man­age­ment tool in advanced CML given
poorer responses and inad­e­quate response dura­bil­ity to ini­tial
ther­apy.34 For BP-CML espe­cially, most long-term sur­vi­vors are
trans­plant recip­i­ents. Recent anal­y­sis dem­on­strates 5-year OS
rates of more than 90% for patients under­go­ing trans­plan­ta­tion
in CP.35 This decreases to 60% in patients who undergo trans­
plan­ta­tion in AP/BP.35 Patients who have trans­plan­ta­tion after
achieve­ment of a sec­ond CP do sig­nif­i­cantly worse com­pared
with patients who undergo trans­plan­ta­tion in their first CP. The
choice of myeloablative vs reduced-inten­sity con­di­tion­ing is un­
clear, although the lat­ter may be a rea­son­able alter­na­tive as it
offers sim­i­lar sur­vival rates at the cost of ear­lier posttransplant

Table 4.  Novel ther­a­pies in advanced CML under inves­ti­ga­tion

Drug class Clinical tri­als N Response

Asciminib (BCR-ABL TKI) Hughes et al (phase 1—asciminib in CP/AP CML after TKI fail­ure)
41
N = 9 (AP) CHR 8/9 (AP)
NCT02081378 (phase 1—asciminib ± TKI in CP/AP/BP CML after TKI MMR 1/9 (AP)
fail­ure)
NCT03595917 (phase 1—asciminib + dasatinib + pred­ni­sone in Ph+
ALL/CML lym­phoid BP)
HQP1351(BCR-ABL TKI) Jiang et al43 (phase 2—HQP1351 in T315I-mutated CP/AP CML) N = 23 (AP) MHR 78% (AP)
MCyR 52% (AP)
K0706 (BCR-ABL TKI) NCT02629692 (phase 1/2—K0706 in AP/CP/BP CML after TKI fail­ure)
PF-114 (BCR-ABL TKI) Turkina et al44 (phase 1—PF-114 CP/AP CML after TKI fail­ure or T315I) N = 51 (CP/AP) MHR 42% (CP/AP)
MCyR 29% (CP/AP)
PHA-739358 (aurora kinase inhib­i­tor) Borthakur et al45 (phase 1—PHA-739358 in AP/BP CML) N = 29 (AP/BP) HR 14% (AP/BP)
SCH 6636 (farnesyl trans­fer­ase inhib­i­tor) Cortes et al (phase 1—SCH 6636 CP/AP/BP CML after imatinib
46
N = 15 (AP/BP) CHR 14% (AP/BP)
fail­ure)
Venetoclax (BCL2 inhib­i­tor) Maiti et al42 (ret­ro­spec­tive—venetoclax + TKI) N = 9 (BP) ORR 75% (BP)
BP1001 (lipo­so­mal Grb-2 anti­sense Ohanian et al47 (phase 1/1b—BP1001 ± low-dose cytarabine in N = 5 (BP) ORR 1/5 (BP)
oli­go­nu­cle­o­tide) advanced mye­loid malig­nan­cies)
Nivolumab (anti–PD-1) NCT02011945 (phase 1b—nivolumab + dasatinib in CP/AP CML)
Inotuzumab (anti-CD22) Jain et al48 (phase 1/2—inotuzumab+bosutinib in Ph+ ALL/CML lym­ N = 2 (BP) ORR 1/2 (BP)
phoid BP)
BCL2, B-cell lym­phoma 2; HR, hema­to­logic response; MCyR, major cyto­ge­netic response; MHR, major hema­to­logic response; ORR, over­all
response rate.
Prakash Singh Shekhawat
Advanced phase CML: room for improve­ment | 125

Figure 1.  Algorithm for treatment of (A) de novo or (B) progressed advanced phase CML.
relapse.36 Overall 5-year OS rates were 53% in both myeloabla­
tive and reduced-inten­sity con­di­tion­ing groups.
For AP patients, how­ever, tim­ing of trans­plant can be dif­fi­
cult. Certain de novo AP patients have out­comes sim­i­lar to CP-
CML, with 1 study suggesting that trans­plant can be deferred
in AP patients who lack high-risk char­ac­ter­is­tics, includ­ing total
CML dis­ease dura­tion more than 12 months, hemo­glo­bin less
than 10 g/dL, and periph­eral blood blasts more than 5%.37 For
most de novo AP patients, treat­ment with TKI alone is likely suf­
fi­cient, with sub­se­quent trans­plan­ta­tion deci­sions based on the
patients’ response to ther­apy. However, HSCT is recommended
in all­eli­gi­ble patients who have progressed to AP from CP and all­
BP patients, includ­ing those presenting with de novo dis­ease.9
Once a patient has under­gone HSCT, the use and dura­tion of
TKI main­te­nance after trans­plan­ta­tion are unknown. In Ph+ acute
lym­pho­blas­tic leu­ke­mia, main­te­nance TKI improves leu­ke­mia-
free sur­vival, relapse, and OS.38 Smaller pro­spec­tive stud­ies have
dem­ on­ strated safety of main­ te­nance TKI with lower rates of
126  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/122/1851605/122how.pdf by guest on 13 December 2021
relapse in CML.39 However, a recent IBMTR anal­y­sis of main­te­
nance TKI after trans­plant for CML showed no ben­e­fit at 100 days
in terms of OS, leu­ke­mia-free sur­vival, relapse rates, trans­plant-
related mor­tal­ity, and chronic graft-ver­sus-host dis­ease.40 NCCN
guide­lines rec­om­mend 12 months of main­te­nance TKI, although
data guid­ing this rec­om­men­da­tion are unclear.9 Minimal resid­ual
dis­ease mon­i­tor­ing is an essen­tial tool to deter­mine dura­tion and
need for TKI main­te­nance, although firm guide­lines in this area
are needed. Our patient was offered trans­plant after reverting to
CP with ponatinib and induc­tion che­mo­ther­apy. She discontin­
ued ponatinib at 6 months given its side effect pro­file and lack of
resid­ual dis­ease on fol­low-up.
Novel ther­a­peu­tics for advanced CML
Despite dra­matic improve­ments in out­comes for patients with
CP-CML, patients with AP- and BP-CML have sig­nif­i­cantly worse
prog­no­sis, and thus new ther­a­peu­tic approaches are needed.
Table 4 sum­ma­rizes novel ther­a­peu­tics cur­rently under inves­ti­ga­
tion. Asciminib, an allo­ste­ric inhib­i­tor that binds to the myristoyl
site of the BCR-ABL tyro­sine kinase, has dem­on­strated effi­cacy
in heavily pretreated patients with CML.41 In 9 patients with AP-
CML, 8 had a CHR, and 1 of 9 had an MMR. As asciminib tar­gets a
dis­tinct bind­ing site com­pared with other TKIs, its devel­op­ment
and recent approval open the pos­si­bil­ity for future TKI-TKI drug
com­bi­na­tions, which are being eval­u­ated in phase 1 and 2 tri­als
(NCT03595917; NCT03578367). Other novel small-mol­e­cule inhib­
i­tors, includ­ing newer-gen­er­a­tion TKI inhib­i­tors, are also under
eval­u­a­tion (Table 4). Venetoclax, a selec­tive B-cell lym­phoma 2
inhib­i­tor, has shown syn­er­gism with TKIs in elim­i­nat­ing leu­ke­mic
stem cells in advanced CML, with encour­ag­ing response rates
for venetoclax and TKI com­bi­na­tions in ret­ro­spec­tive stud­ies for
CML in mye­loid BP.42 Given dys­func­tion in immune sur­veil­lance
within CML, immune ther­a­pies could poten­tially be lev­er­aged in
advanced phase CML, par­tic­u­larly in TKI-resis­tant patients. Over­
all, treat­ment of advanced phase CML remains a chal­lenge, and
pre­ven­tion of dis­ease pro­gres­sion is the most par­a­mount strat­
egy. Clinical trial par­tic­i­pa­tion is encour­aged for all­patients who
develop or have advanced phase CML.
Conclusions
TKI ther­apy rev­o­lu­tion­ized the care of patients with CML. Al­
though the inci­dence of AP- and BP-CML has declined, prog­no­sis
is sig­nif­i­cantly worse, and TKI ther­apy is not as effec­tive. In addi­
tion to small num­bers and lim­ited evi­dence to guide treat­ment,
it is impor­tant to rec­og­nize that there is het­ero­ge­ne­ity in the
def­i­ni­tions of AP and BP, com­pli­cat­ing treat­ment deci­sions. Our
gen­eral approach to patients with advanced CML is sum­ma­rized
in Figure 1. Given lim­ited options in advanced CML, pre­ven­tion of
pro­gres­sion with strict adher­ence to evi­dence-based treat­ment
guide­lines for CP-CML remains the best strat­egy; how­ever, novel
ther­a­peu­tics and treat­ment strat­e­gies are being explored for the
small group of patients who do trans­form.
Conflict-of-inter­est dis­clo­sure
Joan How: no conflicts of interest.
Vinayak Venkataraman: no conflicts of interest.
Gabriela Soriano Hobbs: Scientific Advisory Board (SAB) for No­
vartis, Celgene/BMS, Constellation, AbbVie, Blueprint Medicines,
and Incyte. Research sup­port from Constellation and Incyte.
Off-label drug use
Joan How: nothing to disclose.
Vinayak Venkataraman: nothing to disclose.
Gabriela Soriano Hobbs: nothing to disclose.
Correspondence
Gabriela Soriano Hobbs, Zero Emerson Office 138, Bos­ton, MA
02114; e-mail: ghobbs@part­ners​­.org.
References
1. Shallis RM, Wang R, Davidoff A, Ma X, Podoltsev NA, Zeidan AM. Epidemi­
ology of the clas­si­cal mye­lo­pro­lif­er­a­tive neo­plasms: the four cor­ners of an
expan­sive and com­plex map. Blood Rev. 2020;42:100706.
2. Hoffmann VS, Baccarani M, Hasford J, et al. The EUTOS pop­u­la­tion-based reg­is­
try: inci­dence and clin­i­cal char­ac­ter­is­tics of 2904 CML patients in 20 Euro­pean
Countries. Leukemia. 2015;29(6):1336-1343.
3. Hochhaus A, Larson RA, Guilhot F, et al; IRIS Investigators. Long-term out­
comes of imatinib treat­ment for chronic mye­loid leu­ke­mia. N Engl J Med.
2017;376(10):917-927.
Prakash Singh Shekhawat
4. Cortes JE, Gambacorti-Passerini C, Deininger MW, et al. Bosutinib ver­sus
imatinib for newly diag­nosed chronic mye­loid leu­ke­mia: results from the
ran­dom­ized BFORE trial. J Clin Oncol. 2018;36(3):231-237.
5. Speck B, Bortin MM, Champlin R, et al. Allogeneic bone-mar­row trans­plan­
ta­tion for chronic mye­log­e­nous leu­kae­mia. Lancet. 1984;1(8378):665-668.
6. Kantarjian HM, Dixon D, Keating MJ, et al. Characteristics of accel­er­ated dis­
ease in chronic mye­log­e­nous leu­ke­mia. Cancer. 1988;61(7):1441-1446.
7. Baccarani M, Deininger MW, Rosti G, et al. Euro­pean LeukemiaNet rec­om­
men­da­tions for the man­age­ment of chronic mye­loid leu­ke­mia: 2013. Blood.
2013;122(6):872-884.
8. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revi­sion to the World Health
Organization clas­si­fi­ca­tion of mye­loid neo­plasms and acute leu­ke­mia.
Blood. 2016;127(20):2391-2405.
9. Network NCC. Myeloproliferative neo­plasms (ver­sion 3.2021). Accessed
31 Jan­u­ary 2021. https:​­/​­/www​­.nccn​­.org​­/professionals​­/physician_gls​­/pdf​­/
mpn_blocks​­.pdf.
10. Bonifacio M, Stagno F, Scaffidi L, Krampera M, Di Raimondo F. Management
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/122/1851605/122how.pdf by guest on 13 December 2021
of chronic mye­loid leu­ke­mia in advanced phase. Front Oncol. 2019;9:1132.
11. Ohanian M, Kantarjian HM, Quintas-Cardama A, et al. Tyrosine kinase inhib­
i­tors as ini­tial ther­apy for patients with chronic mye­loid leu­ke­mia in accel­
er­ated phase. Clin Lymphoma Myeloma Leuk. 2014;14(2):155-162.e1162e1.
12. Rea D, Etienne G, Nicolini F, et al. First-line imatinib mesylate in patients
with newly diag­nosed accel­er­ated phase-chronic mye­loid leu­ke­mia. Leu-
kemia. 2012;26(10):2254-2259.
13. Jain P, Kantarjian HM, Ghorab A, et al. Prognostic fac­tors and sur­vival out­
comes in patients with chronic mye­loid leu­ke­mia in blast phase in the tyro­
sine kinase inhib­i­tor era: cohort study of 477 patients. Cancer. 2017;123(22):​
4391-4402.
14. Lauseker M, Bachl K, Turkina A, et  al. Prognosis of patients with chronic
mye­loid leu­ke­mia presenting in advanced phase is defined mainly by blast
count, but also by age, chro­mo­somal aber­ra­tions and hemo­glo­bin. Am J
Hematol. 2019;94(11):1236-1243.
15. Ochi Y, Yoshida K, Huang Y-J, et al. Clonal evo­lu­tion and clin­i­cal impli­ca­
tions of genetic abnor­mal­i­ties in blastic trans­for­ma­tion of chronic mye­loid
leu­kae­mia. Nat Commun. 2021;12(1):2833.
16. Wang W, Cortes JE, Lin P, et  al. Clinical and prog­nos­tic sig­nif­i­cance of
3q26.2 and other chro­mo­some 3 abnor­mal­i­ties in CML in the era of tyro­
sine kinase inhib­i­tors. Blood. 2015;126(14):1699-1706.
17. Hehlmann R, Voskanyan A, Lauseker M, et al; SAKK and the Ger­man CML
Study Group. High-risk addi­tional chro­mo­somal abnor­mal­i­ties at low blast
counts her­ald death by CML. Leukemia. 2020;34(8):2074-2086.
18. Kim T, Tyndel MS, Kim HJ, et  al. Spectrum of somatic muta­tion dynam­
ics in chronic mye­loid leu­ke­mia fol­low­ing tyro­sine kinase inhib­i­tor ther­apy.
Blood. 2017;129(1):38-47.
19. Nicolini FE, Masszi T, Shen Z, et al. Expanding Nilotinib Access in Clinical
Trials (ENACT), an open-label mul­ti­cen­ter study of oral nilotinib in adult
patients with imatinib-resis­tant or -intol­er­ant chronic mye­loid leu­ke­mia in
accel­er­ated phase or blast cri­sis. Leuk Lymphoma. 2012;53(5):907-914.
20. Kantarjian H, Cortes J, Kim DW, et  al. Phase 3 study of dasatinib 140mg
once daily ver­sus 70mg twice daily in patients with chronic mye­loid leu­
ke­mia in accel­er­ated phase resis­tant or intol­er­ant to imatinib: 15-month
median fol­low-up. Blood. 2009;113(25):6322-6329.
21. Gambacorti-Passerini C, Kantarjian HM, Kim D-W, et  al. Long-term effi­
cacy and safety of bosutinib in patients with advanced leu­ke­mia fol­low­ing
resis­tance/intol­er­ance to imatinib and other tyro­sine kinase inhib­i­tors. Am
J Hematol. 2015;90(9):755-768.
22. Cortes JE, Kim D-W, Pinilla-Ibarz J, et al; PACE Investigators. A phase 2 trial
of ponatinib in Philadelphia chro­mo­some-pos­i­tive leu­ke­mias. N Engl J Med.
2013;369(19):1783-1796.
23. Cortes JE, Kim D-W, Pinilla-Ibarz J, et  al. Ponatinib effi­cacy and safety in
Philadelphia chro­ mo­some-pos­ i­
tive leu­ ke­
mia: final 5-year results of the
phase 2 PACE trial. Blood. 2018;132(4):393-404.
24. Kantarjian H, Deininger M, Abruzzese E, et al. Efficacy and safety of pona­
tinib (PON) in patients with chronic-phase chronic mye­loid leu­ke­mia (CP-
CML) who failed one or more sec­ ond-gen­ er­
a­tion (2G) tyro­ sine kinase
inhib­i­tors (TKIs): ana­ly­ses based on PACE and optic. Blood. 2020;136:43-44.
25. Khoury HJ, Cortes J, Baccarani M, et  al. Omacetaxine mepesuccinate in
patients with advanced chronic mye­loid leu­ke­mia with resis­tance or intol­
er­ance to tyro­sine kinase inhib­i­tors. Leuk Lymphoma. 2015;56(1):120-127.
26. Sawyers CL, Hochhaus A, Feldman E, et al. Imatinib induces hema­to­logic and
cyto­ge­netic responses in patients with chronic mye­log­e­nous leu­ke­mia in
mye­loid blast cri­sis: results of a phase II study. Blood. 2002;99(10):3530-3539.
Advanced phase CML: room for improve­ment | 127
27. Nicolini FE, Basak GW, Kim D-W, et  al. Overall sur­ vival with ponatinib
ver­sus allo­ge­neic stem cell trans­plan­ta­tion in Philadelphia chro­mo­some-
pos­i­tive leu­ke­mias with the T315I muta­tion. Cancer. 2017;123(15):2875-
2880.
28. Saglio G, Hochhaus A, Goh YT, et  al. Dasatinib in imatinib-resis­ tant or
imatinib-intol­er­ant chronic mye­loid leu­ke­mia in blast phase after 2 years
of fol­low-up in a phase 3 study: effi­cacy and tol­er­a­bil­ity of 140 mil­li­grams
once daily and 70 mil­li­grams twice daily. Cancer. 2010;116(16):3852-3861.
29. Deau B, Nicolini FE, Guilhot J, et al. The addi­tion of dau­no­ru­bi­cin to imatinib
mesylate in com­bi­na­tion with cytarabine improves the response rate and
the sur­vival of patients with mye­loid blast cri­sis chronic mye­log­e­nous leu­
ke­mia (AFR01 study). Leuk Res. 2011;35(6):777-782.
30. Milojkovic D, Ibrahim A, Reid A, Foroni L, Apperley J, Marin D. Efficacy of
com­bin­ing dasatinib and FLAG-IDA for patients with chronic mye­loid leu­
ke­mia in blastic trans­for­ma­tion. Haematologica. 2012;97(3):473-474.
31. Strati P, Kantarjian H, Thomas D, et al. HCVAD plus imatinib or dasatinib in
lym­phoid blastic phase chronic mye­loid leu­ke­mia. Cancer. 2014;120(3):373-
380.
32. Ghez D, Micol J-B, Pasquier F, et al. Clinical effi­cacy of sec­ond gen­er­a­tion
tyro­sine kinase inhib­i­tor and 5-azacytidine com­bi­na­tion in chronic mye­
log­e­nous leu­kae­mia in mye­loid blast cri­sis. Eur J Cancer. 2013;49(17):3666-
3670.
33. Ruggiu M, Oberkampf F, Ghez D, et al. Azacytidine in com­bi­na­tion with tyro­
sine kinase inhib­i­tors induced dura­ble responses in patients with advanced
phase chronic mye­log­e­nous leu­ke­mia. Leuk Lymphoma. 2018;59(7):1659-
1665.
34. Hehlmann R, Lauseker M, Jung-Munkwitz S, et  al. Tolerability-adapted
imatinib 800 mg/d ver­sus 400 mg/d ver­sus 400 mg/d plus inter­feron-α in
newly diag­nosed chronic mye­loid leu­ke­mia. J Clin Oncol. 2011;29(12):1634-
1642.
35. Saussele S, Lauseker M, Gratwohl A, et al; Ger­man CML Study Group. Allo­
geneic hema­to­poi­etic stem cell trans­plan­ta­tion (allo SCT) for chronic mye­
loid leu­ke­mia in the imatinib era: eval­u­a­tion of its impact within a sub­group
of the ran­dom­ized Ger­man CML Study IV. Blood. 2010;115(10):1880-1885.
36. Chhabra S, Ahn KW, Hu Z-H, et al. Myeloablative vs reduced-inten­sity con­
di­tion­ing allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion for chronic mye­
loid leu­ke­mia. Blood Adv. 2018;2(21):2922-2936.
37. Jiang Q , Xu L-P, Liu D-H, et al. Imatinib mesylate ver­sus allo­ge­neic hema­
to­poi­etic stem cell trans­plan­ta­tion for patients with chronic mye­log­e­nous
leu­ke­mia in the accel­er­ated phase. Blood. 2011;117(11):3032-3040.
38. Brissot E, Labopin M, Beckers MM, et al. Tyrosine kinase inhib­i­tors improve
long-term out­come of allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion
for adult patients with Philadelphia chro­mo­some pos­i­tive acute lym­pho­
blas­tic leu­ke­mia. Haematologica. 2015;100(3):392-399.
128  |  Hematology 2021  |  ASH Education Program
39. Olavarria E, Siddique S, Griffiths MJ, et al. Posttransplantation imatinib as
a strat­egy to post­pone the require­ment for immu­no­ther­apy in patients
under­go­ing reduced-inten­sity allo­grafts for chronic mye­loid leu­ke­mia.
Blood. 2007;110(13):4614-4617.
40. DeFilipp Z, Ancheta R, Liu Y, et al. Maintenance tyro­sine kinase inhib­it­ors
fol­low­ing allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion for chronic
mye­log­e­nous leu­ke­mia: a cen­ter for inter­na­tional blood and mar­row trans­
plant research study. Biol Blood Marrow Transplant. 2020;26(3):472-479.
41. Hughes TP, Mauro MJ, Cortes JE, et al. Asciminib in chronic mye­loid leu­ke­
mia after ABL kinase inhib­i­tor fail­ure. N Engl J Med. 2019;381(24):2315-2326.
42. Maiti A, Rausch CR, Cortes JE, et al. Outcomes of relapsed or refrac­tory acute
mye­loid leu­ke­mia after front­line hypomethylating agent and venetoclax reg­
i­mens. Haematologica. 2021;106(3):894-898.
43. Jiang Q , Huang X, Chen Z, et al. Novel BCR-ABL1 tyro­sine kinase inhib­i­tor
(TKI) HQP1351 (olverembatinib) is effi­ca­cious and well tol­er­ated in patients
with T315I-mutated chronic mye­ loid leu­ke­
mia (CML): results of piv­ otal
(phase II) tri­als. Blood. 2020;136(suppl 1):50-51.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/122/1851605/122how.pdf by guest on 13 December 2021
44. Turkina AG, Vinogradova O, Lomaia E, et al. Phase-1 study of PF-114 mesylate
in CML fail­ing prior tyro­sine kinase-inhib­i­tor ther­apy. Blood. 2018;132(suppl
1):790.
45. Borthakur G, Dombret H, Schafhausen P, et al. A phase I study of danusertib
(PHA-739358) in adult patients with accel­er­ated or blastic phase chronic
mye­loid leu­ke­mia and Philadelphia chro­mo­some-pos­i­tive acute lym­pho­
blas­tic leu­ke­mia resis­tant or intol­er­ant to imatinib and/or other sec­ond
gen­er­a­tion c-ABL ther­apy. Haematologica. 2015;100(7):898-904.
46. Cortes J, Jabbour E, Daley GQ , et al. Phase 1 study of lonafarnib (SCH 66336)
and imatinib mesylate in patients with chronic mye­loid leu­ke­mia who have
failed prior sin­gle-agent ther­apy with imatinib. Cancer. 2007;110(6):1295-
1302.
47. Ohanian M, Tari Ashizawa A, Garcia-Manero G, et  al. Liposomal Grb2
anti­ sense oligodeoxynucleotide (BP1001) in patients with refrac­ tory or
relapsed haematological malig­nan­cies: a sin­gle-cen­tre, open-label, dose-
esca­la­tion, phase 1/1b trial. Lancet Haematol. 2018;5(4):e136-e146.
48. Jain N, Maiti A, Ravandi F, et al. Inotuzumab ozogamicin with bosutinib for
relapsed or refrac­tory Philadelphia chro­mo­some pos­i­tive acute lym­pho­
blas­tic leu­ke­mia or lym­phoid blast phase of chronic mye­loid leu­ke­mia. Am
J Hematol. 2021;96(8):1000-1007.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000240
 COAGULATION LABORATORY POTPOURRI

Direct oral anticoagulant (DOAC) interference

in hemostasis assays

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/129/1851777/129moser.pdf by guest on 13 December 2021

Karen A. Moser1 and Kristi J. Smock2
Department of Pathology, University of Utah School of Medicine, Salt Lake City, UT; and 2ARUP Institute for Clinical and Experimental Pathology,
1

Salt Lake City, UT

Direct oral anticoagulants (DOACs) are a group of direct coagulation factor inhibitors including both direct thrombin
inhibitors and direct factor Xa inhibitors. These medications may cause hemostasis assay interference by falsely increas-
ing or decreasing measured values, depending on the analyte. Considering the potential for DOAC interference in a vari-
ety of hemostasis assays is essential to avoid erroneous interpretation of results. Preanalytic strategies to avoid DOAC
interference include selecting alternatives to clot-based hemostasis assays in patients taking DOACs when possible and
sample collection timed when the patient is off anticoagulant therapy or at the expected drug trough. Clinical labora-
tories may also provide educational materials that clearly describe possible interferences from DOAC, develop testing
algorithms to aid in detection of DOAC in submitted samples, use DOAC-neutralizing agents to remove DOACs before
continuing with testing, and write interpretive comments that explain the effects of DOAC interference in hemostasis
tests. Using a combination of the described strategies will aid physicians and laboratorians in correctly interpreting
hemostasis and thrombosis laboratory tests in the presence of DOACs.

LEARNING OBJECTIVES
• Describe the patterns of interference in hemostasis assays due to direct thrombin inhibitors and direct Xa inhibi­
tors
• List potential strategies physicians and clinical laboratories can use to decrease DOAC interference in hemostasis
assays

Since the introduction of the direct oral anticoagulants
(DOACs) in the 2010s, clinical laboratories have struggled
to mitigate the effects of these drugs in clot­based hemo­
stasis and thrombosis assays. DOACs are a group of di­
rect coagulation factor inhibitors that include both direct
thrombin inhibitors (dabigatran) and direct Xa inhibitors
(rivaroxaban, apixaban, edoxaban).1 The DOACs may cause
assay interference by falsely increasing or decreasing mea­
sured values, depending on the analyte. Data from both
individual laboratory studies and external quality assess­
ment programs describe expected patterns with a variety
of reagents (Table 1).1-8
embolism, prompting testing for lupus anticoagulant (LA)
and treatment with rivaroxaban. Laboratory results for
both her initial LA profile as well as follow­up testing 12
weeks later are included in Table 2.
Case 2: A 37­year­old man sought treatment for an
unprovoked pulmonary embolism and was evaluated for
thrombophilia risk factors, including LA, while taking apix­
aban therapy. Laboratory results for his initial LA panel are
included in Table 2. An anti­Xa activity assay calibrated for
unfractionated heparin (UFH) showed measurable anti­Xa
activity in the plasma sample submitted for the LA profile.

Patterns of DOAC interference in hemostasis assays

CLINICAL CASES
Case 1: A 54­year­old woman was diagnosed with pneu­
monia and was bedridden due to the severity of her ill­
ness. A few weeks later, she developed a pulmonary
Prakash Singh Shekhawat
For basic hemostasis assays such as prothrombin time (PT)
and activated partial thromboplastin time (aPTT), response
to DOACs varies considerably by drug, drug concentration,
and reagent. In general, the aPTT may be prolonged with
dabigatran but does not tend to be prolonged by direct
Xa inhibitors.1,2,9 A normal aPTT is insufficient to exclude the
DOAC interference | 129
Table 1.  Patterns of DOAC inter­fer­ence in hemo­sta­sis/throm­bo­sis assays
Expected change Assays
Clotting time pro­lon­ga­tion • aPTT (dabigatran > direct Xa inhib­i­tors)
• PT (rivaroxaban > edoxaban > apixaban)
• Thrombin time (dabigatran)
False increase • Clot-based pro­tein C activ­ity
• Clot-based pro­tein S activ­ity
• Antithrombin activ­ity (in fac­tor IIa–based assays with
dabigatran, in fac­tor Xa–based assays with direct Xa
inhib­i­tors)
• Activated pro­tein C resis­tance ratio
False decrease • aPTT-based fac­tor assays (VIII, IX, XI, XII)
• PT-based fac­tor assays (II, V, VII, X)
False pos­i­tive (or poten­tially • LA assays
false neg­a­tive)
No change • Clauss fibrin­o­gen activ­ity (for most reagents, rare
meth­ods show false decrease in pres­ence of high
con­cen­tra­tions of dabigatran)
• D-dimer
• Chromogenic pro­tein C activ­ity
• Free and total pro­tein S anti­gen
• Anticardiolipin, anti-β2GP1 ELISAs
• von Willebrand activ­ity and anti­gen assays
• DNA-based assays (eg, fac­tor V Leiden muta­tion, pro­
throm­bin G20210A muta­tion)
pres­ence of dabigatran.1 The PT may be prolonged with rivar­
oxaban and edoxaban but tends to be rel­a­tively insen­si­tive to
the pres­ence of apixaban in all­reagents eval­u­ated to date.2,9 The
dif­fer­ence between the observed PT pro­lon­ga­tion in the clin­i­cal
cases high­lights the dif­fer­en­tial effects of rivaroxaban and apix­
aban on the PT. Dabigatran does not tend to pro­long the PT.2,9
In mixing tests for both aPTT and PT, DOACs will appear as non­
spe­cific inhib­i­tors, with the inhib­it­ or effect most pro­nounced at
higher drug con­cen­tra­tions.1 Fibrinogen activ­ity tends to be unaf­
fected by DOACs, par­tic­u­larly the com­monly used Clauss fibrin­
o­gen assay, which uses plasma dilu­tion as well as a high con­
cen­tra­tion of throm­bin to con­vert patient fibrin­o­gen to fibrin.2-6
Factitiously decreased fibrin­ og­en activ­ ity was reported by
some lab­o­ra­to­ries at higher dabigatran con­cen­tra­tions (385 and
744 ng/mL) in an exter­nal qual­ity assess­ment pro­gram.3 D-dimer
assays are typ­i­cally immu­no­as­says, such as enzyme-linked immu­
no­sor­bent assay (ELISAs) or latex immu­no­as­says; these meth­ods
are not affected by the pres­ence of DOACs.1
Among spe­cial­ized hemo­sta­sis assays, those eval­u­at­ing for
thrombophilia risk fac­tors are at par­tic­u­lar risk for inter­fer­ence,
given that many of these assays are clot based and may be mea­
sured in patients who have throm­bo­sis and are treated with
anti­co­ag­u­lant ther­apy.10-12 The clin­i­cal cases illus­trate two typ­
i­cal exam­ples of how LA assays are affected by DOACs. Dilute
Russell viper venom time (DRVVT) and aPTT-based LA assays
(eg, plate­let neu­tral­i­za­tion pro­ce­dure, hex­ag­o­nal phos­pho­lipid
neu­tral­i­za­tion) may show false-pos­i­tive results in the pres­ence
of DOACs.10,12,13 False-pos­it­ive DRVVT-based test­ing appears to
be a par­tic­u­lar risk with rivaroxaban, but there also appears to
be risk of DRVVT false neg­a­tives for LA in sam­ples containing
apixaban.14 Just as with screen­ing aPTT and PT, the mixing test
steps in DRVVT and aPTT-based LA assays may show a non­spe­
cific inhib­i­tor pat­tern.13 Assays for clot-based pro­tein C, pro­tein
130  |  Hematology 2021  |  ASH Education Program
Notes
Effects on clot­ting times are reagent depen­dent.
aPTT and PT mixing tests are expected to show incom­
plete cor­rec­tion in the pres­ence of DOACs.
False increase in pro­tein C, pro­tein S, and anti­throm­bin
activ­i­ties may result in mis­di­ag­no­sis of a patient with
true defi­ciency as nor­mal.
Falsely ele­vated acti­vated pro­tein C resis­tance ratio may
result in mis­di­ag­no­sis of a patient with fac­tor V Leiden
muta­tion as nor­mal.
Dilutions in fac­tor assays may show non­spe­cific inhib­i­tor
effect.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/129/1851777/129moser.pdf by guest on 13 December 2021
Includes aPTT- and DRVVT-based assays, among other
clot­ting time-based LA assays; effects are drug and
reagent depen­dent.
S, and anti­throm­bin may be falsely increased in the pres­ence
of DOACs, which may result in a false-neg­a­tive result.2-6,11 That
is, a patient with a true defi­ciency may have a nor­mal result if
plasma is tested for pro­tein C, pro­tein S, or anti­throm­bin activ­ity
using clot-based assays in the pres­ence of DOACs. DOACs do
not affect chro­mo­genic pro­tein C activ­ity assays or total pro­tein
C anti­gen assays.1,11 Protein S activ­ity assays are clot based and
expe­ri­ence inter­fer­ence by DOACs, whereas free and total pro­
tein S anti­gen assays are immu­no­as­says and do not show DOAC
inter­fer­ence.3,4,15 Antithrombin activ­ity assays may be either fac­
tor IIa based or fac­tor Xa based; the design deter­mines which
DOACs will inter­fere. Factor IIa–based assays show false ele­va­
tion with direct throm­bin inhib­i­tors, whereas fac­tor Xa–based
assays show false ele­va­tion with direct Xa inhib­i­tors.3-6 Activated
pro­tein C resis­tance assays also have the poten­tial to give false-
neg­a­tive results in the pres­ence of DOACs.11,16
Other clot-based hemo­sta­sis assays, such as fac­tor assays,
may show decreased activ­ ity in the pres­ ence of DOACs.1,3,4,11
Hemostasis assays based on ELISA or other immu­no­as­say meth­
ods (eg, von Willebrand fac­tor anti­gen, solid-phase antiphospho­
lipid antibodies) will not show inter­fer­ence by DOACs; the results
of these assays will not be affected if performed in a sam­ple from
a patient receiv­ing DOAC ther­apy.1 Likewise, DNA-based assays
(eg, fac­tor V Leiden muta­tion, pro­throm­bin G20210A muta­tion)
will be unaf­fected by the pres­ence of DOACs.
One par­tic­u­larly chal­leng­ing area where direct Xa inhib­i­tor
inter­fer­ence can limit ther­a­peu­tic mon­i­tor­ing of another drug is
the spe­cial case of patients switching from a direct Xa inhib­i­tor
to ther­apy with UFH or low molec­u­lar weight hep­a­rin (LMWH).17
Early reports of effi­cacy of DOAC-Stop (D-S; Haematex Research)
for remov­ing the mea­sur­able anti-Xa activ­ity effect of rivaroxaban
and apixaban but not hep­a­rins raise the pos­si­bil­ity that plas­mas
in patients transitioning between a direct Xa inhib­i­tor and UFH
Table 2.  LA panel results for clin­i­cal cases 1 and 2 anti­throm­bin anti­gen) do not detect rare deficiencies due to pro­
tein dys­func­tion and would not rep­re­sent appro­pri­ate sub­sti­tutes
Case 1: for func­tional assays. If a mea­sure­ment must be made, test­ing at
Case 1: LA panel Case 2: expected drug trough con­cen­tra­tion is recommended: either 12
ini­tial 12 weeks ini­tial Reference hours postdose for DOACs dosed twice daily or 24 hours post­
Test LA panel later LA panel inter­val
dose for DOACs admin­is­tered once daily.9,14 LA test sys­tems (aPTT
PT(s) 23.0 12.8 14.3 12.0-15.5 and DRVVT based) have dem­on­strated inter­fer­ence from DOACs
DRVVT screen(s) 57 35 91 33-44 even at low drug con­cen­tra­tions; there­fore, test­ing a sam­ple col­
DRVVT 1:1 mix(s) 52 NA 80 33-44 lected at an expected drug trough con­cen­tra­tion may not com­
pletely elim­i­nate the pos­si­bil­ity of inter­fer­ence.21
DRVVT con­firm(s) Negative NA Positive Negative
aPTT screen(s) 61 38 119 32-48 What can clin­i­cal lab­o­ra­to­ries do to avoid DOAC
TT(s) 15.7 NA 15.1 14.7-19.5 inter­fer­ence in hemo­sta­sis assays?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/129/1851777/129moser.pdf by guest on 13 December 2021

aPTT 1:1 mix(s) 49 NA 85 32-48
PNP Negative NA Positive Negative
Hex phos Positive NA NA Negative
neu­tral­i­za­tion
DRVVT con­firm, PNP, and hex phos neutralization are LA con­fir­ma­tory
reagents containing high phos­pho­lipid con­cen­tra­tions.
hex phos neu­tral­i­za­tion, hex­ag­o­nal phase phos­pho­lipid neu­tral­i­za­tion
test; NA, not applicable; PNP, plate­let neu­tral­i­za­tion pro­ce­dure;
TT, throm­bin time.
or LMWH could be treated with D-S prior to mea­sure­ment with a
UFH- or LMWH-calibrated anti-Xa activ­ity assay or aPTT for ther­
a­peu­tic mon­i­tor­ing.17-19 It is impor­tant to note that the degree to
which acti­vated car­bon adsorbs LMWH is incom­pletely under­
stood, and fur­ther study in this area is needed before rou­tinely
using DOAC-neu­tral­iz­ing com­pounds in sam­ples also containing
LMWH.19
What can phy­si­cians do to avoid DOAC inter­fer­ence
in hemo­sta­sis assays?
Physicians can choose from a few dif­fer­ent prac­ti­cal strat­e­gies to
min­i­mize DOAC inter­fer­ence in hemo­sta­sis assays. The sim­plest
and best way to avoid DOAC inter­fer­ence is to avoid order­ing
and performing hemo­ sta­
sis assays in patients tak­ ing DOACs.
Another option would be to stop DOAC ther­apy for 2 to 3 days
prior to collecting a sam­ple for hemo­sta­sis test­ing; how­ever, giv­
en the risk of throm­bo­sis with inter­rup­tion or change of treat­
ment, this option may not be clin­i­cally fea­si­ble in most cases.11,12,20
Briefly transitioning to an anti­co­ag­u­lant with less assay inter­fer­
ence, such as LMWH, could also be an option but also may not
be prac­ti­cal or fea­si­ble in many cases. In some clin­i­cal sit­u­a­tions,
test­ing in patients tak­ing DOACs may be desired (eg, in patients
requir­ing indef­i­nite anticoagulation or detecting an LA in the set­
ting of unpro­voked throm­bo­sis to gain infor­ma­tion about recur­
rence risk).12 If test­ing is under­taken, phy­si­cians can con­sider both
choice of assay and tim­ing of sam­ple col­lec­tion to decrease the
pos­si­bil­ity of DOAC inter­fer­ence. For analytes in which there is a
choice between a func­tional assay and an assay such as a chro­
mo­genic, an anti­genic, or a DNA-based assay where inter­fer­ence
is not expected, phy­si­cians can choose the assay with­out expect­
ed inter­fer­ence. For exam­ple, using this prin­ci­ple would favor
muta­tion anal­y­sis for fac­tor V Leiden over func­tional test­ing with
acti­vated pro­tein C resis­tance assays in a patient tak­ing DOACs
in whom thrombophilia test­ing is desired. Simply substitut­ing an­
other test does not work in all­cases in which alter­nate assays are
avail­­able. Some tests that lack inter­fer­ence (such as pro­tein C or
Prakash Singh Shekhawat
Clinical lab­o­ra­to­ries like­wise have mul­ti­ple options for miti­gat­
ing the effects of DOAC inter­fer­ence in hemo­sta­sis and throm­
bo­sis assays.14 A lab­o­ra­tory can pro­vide edu­ca­tional mate­rial to
phy­si­cians and other health care pro­vid­ers using the lab­o­ra­tory’s
ser­vices that include clear descrip­tions of expected pat­terns of
DOAC inter­fer­ence, sim­i­lar to Table 1.22 Laboratories may also de­
sign test­ing algo­rithms that allow detec­tion of inter­fer­ing anti­
co­ag­u­lants in plasma sam­ples in cases where a patient’s his­tory
of anti­co­ag­u­lant ther­apy is not avail­­able.11,13 In clin­i­cal case 1, the
prolonged PT offered a clue to the pres­ence of an anti­co­ag­u­lant
med­i­ca­tion, either war­fa­rin or a direct Xa inhib­i­tor. Detecting
anti-Xa activ­ity in the sam­ple is another, more sen­si­tive way to
iden­tify a direct Xa inhib­i­tor in a plasma sam­ple; throm­bin time
can be used to iden­tify the pres­ence of direct throm­bin inhib­i­
tors and hep­a­rins.11,13 Laboratories should also pro­vide inter­pre­
tive result com­ments that explain the poten­tial for DOAC inter­
fer­ence in the assays performed, par­tic­u­larly for LA pro­files as
recommended in cur­rent International Society on Thrombosis
and Haemostasis (ISTH) and Clinical and Laboratory Standards
Institute (CLSI) guide­lines.13,23 Some lab­o­ra­to­ries may elect not
to com­plete test­ing for a LA in the event DOAC inter­fer­ence is
detected and may sim­ply issue a com­ment stat­ing that DOAC
inter­fer­ence ren­ders results of LA test­ing uninterpretable.23
Recently, adsorbing agents that can neu­tral­ize DOAC effects
in plasma sam­ples in the hemo­sta­sis lab­o­ra­tory have been devel­
oped, although these agents are not yet FDA approved. Clinical
lab­o­ra­to­ries have exten­sive expe­ri­ence with remov­ing hep­a­rin
effects from plasma sam­ ples using heparinase or polybrene.24
Using DOAC-neu­tral­iz­ing agents is an attrac­tive pos­si­bil­ity that
could be incor­ po­
rated into lab­ o­ra­
tory workflows that already
use hep­a­rin neu­tral­iz­ers in a sim­i­lar fash­ion. Currently avail­­able
DOAC-neu­tral­iz­ing agents in tab­let form include DOAC-Stop (D-S;
Haematex Research) and DOAC-Remove (D-R; 5-Diagnostics
­
AG).11 These agents are com­posed of acti­vated car­bon-containing
adsor­bent com­pounds. One D-S or D-R tab­let must be added to
1 mL patient plasma, with time allowed for adsorp­tion of DOACs
and sub­se­quent cen­tri­fu­ga­tion and removal of DOAC-free plasma
for test­ing (Figure 1).10,18,25 DOAC-spiked plas­mas that show false-
pos­i­tive LA results gen­er­ally con­vert to neg­a­tive LA results with
DRVVT- and aPTT-based test­ ing fol­low­
ing D-S treat­ ment, with
results com­pa­ra­ble to neu­tral­i­za­tion with idarucizumab or andex­
anet alfa.10,19,26,27 One tab­ let of D-S in 1  mL plasma can report­
edly neu­tral­ize up to 708 ng/mL apixaban, 1060 ng/mL edoxaban,
1020 ng/mL rivaroxaban, and 360 ng/mL dabigatran.10 D-S or D-R
has also been reported to decrease DOAC inter­fer­ence in chro­
mo­ genic and clot-based fac­ tor VIII activ­ ity,18,26 1-stage fac­ tor
IX activ­ity,26 throm­bin gen­er­a­tion assays,28 acti­ vated pro­ tein C
DOAC inter­fer­ence  | 131
 A. DOAC-neutralizing tablets with B. DOAC-neutralizing filter extracting
activated carbon (eg, DOAC-Stop, DOACs through noncovalent binding in
DOAC-Remove) solid phase (eg, DOAC Filter)

Filter
Plasma
Connector
1 tablet

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/129/1851777/129moser.pdf by guest on 13 December 2021

added to
1 mL
patient Filter DOAC-free
Sample Filter placed on apparatus
citrated plasma in
mixed for 5 connecter attached centrifuged
plasma microtainer
minutes on to microtainer with 15 minutes, and ready
rocking DOAC-free 600 µL platelet-poor 300 g, room for testing
mixer and plasma plasma loaded temperature
centrifuged supernatant
5 minutes removed
at 2000 g and ready
for testing

Figure 1.  DOAC-neutralizing processes with tablet-based adsorbing agents (A) and filter systems (B).

resis­tance,25,29 and anti­throm­bin activ­ity.29,30 D-S has been reported
to cause decreased fac­tor activ­i­ties in treated plas­mas, rais­ing
the pos­si­bil­ity of adsorp­tion of coag­u­la­tion fac­tors in addi­tion to
DOACs; how­ever, D-S does not pro­duce aPTT pro­lon­ga­tion with
all­reagents tested.10,31 Caution in result inter­pre­ta­tion is recom­
mended when using DOAC-neu­tral­iz­ing agents in the clin­i­cal lab­o­
ra­tory as com­plete neu­tral­i­za­tion was not observed in all­cases.10,18
One even more recently described option, DOAC Filter (Diagnos­
tica Stago), is a car­tridge containing a solid phase designed to
extract DOACs through noncovalent bind­ing when 600 µL citrated
plasma is added and centri­fuged at 300 × g for 15 min­utes.32 DOAC
Filter treat­ ment resulted in unde­ tect­able lev­ els of DOACs as
assessed by quan­ti­ta­tive assays and no change in selected hemo­
sta­sis assays (PT, aPTT, fibrin­o­gen, anti­throm­bin activ­ity, pro­tein C
activ­ity, acti­vated pro­tein C resis­tance, LA tests) in postfiltration
sam­ples.32 Further study is needed to bet­ter describe the effi­cacy
of DOAC-neu­tral­iz­ing reagents and the risk for inad­ver­tent adsorp­
tion of other coag­u­la­tion fac­tors from plasma in the clin­i­cal lab­o­
ra­tory. It is also impor­tant to remem­ber that performing addi­tional
assays to detect DOACs and use of neu­tral­iz­ing agents adds cost
and time to test­ing and requires addi­tional patient plasma. Labo­
ratories will need to care­fully con­sider how these meth­ods would
fit into existing test workflows.
CLINICAL CASE (Con­t in­u ed)
Case 1: This patient had an ini­tial pos­it­ ive LA pro­file with a fol­
low-up pro­file 12 weeks later in which an LA was not detected;
both pro­files were performed while the patient was receiv­ing
rivaroxaban. Differential diag­nos­tic con­sid­er­ations for this pat­
tern include a tran­sient LA, dif­fer­ences in tim­ing of draw (eg,
first pro­file drawn at drug peak and sec­ond pro­file drawn at
drug trough), or nonadherence to the rivaroxaban reg­i­men at
the time of the sec­ond pro­file. The prolonged PT in the first LA
pro­file sug­gests one of the lat­ter 2 pos­si­bil­i­ties.
Case 2: This case dem­on­strates the rel­a­tive insen­si­tiv­ity of
the PT to apixaban. The mea­sur­able direct Xa activ­ity sug­gests
that the pos­i­tive LA results obtained were caused by apixaban
inter­fer­ence. A repeat sam­ple col­lected when the patient is not
tak­ing apixaban, repeat sam­ ple col­ lected at expected drug
trough, or treating the cur­rent sam­ple with a DOAC-neu­tral­iz­ing
agent and repeat­ ing the LA pro­ file are pos­ si­ble options to
decrease apixaban inter­fer­ence.
Conclusion
Considering the poten­tial for DOAC inter­fer­ence in a vari­ety of
hemo­sta­sis assays is essen­tial to avoid erro­ne­ous inter­pre­ta­tion of
results. Preanalytic strat­e­gies to avoid DOAC inter­fer­ence include
avoiding clot-based hemo­sta­sis assays in patients tak­ing DOACs
and sam­ple col­lec­tion timed when the patient is not tak­ing anti­
co­ag­u­lant ther­apy or at the expected drug trough (as opposed
to peak or ran­dom sam­ples). Strategies clin­i­cal lab­o­ra­to­ries may
employ to avoid DOAC inter­fer­ence include pro­vid­ing edu­ca­tional
mate­ri­als that clearly describe pos­si­ble inter­fer­ences from DOACs,
devel­op­ing test­ing algo­rithms to aid in the detec­tion of DOACs
in sub­mit­ted sam­ples, using DOAC-neu­tral­iz­ing agents to remove

132  |  Hematology 2021  |  ASH Education Program

DOACs before con­tinu­ing with test­ing, and writ­ing inter­pre­tive
com­ments that explain the effects of DOAC inter­fer­ence in hemo­
sta­sis tests. Using a com­bi­na­tion of the described strat­e­gies will
aid phy­si­cians and laboratorians in cor­rectly interpreting hemo­
sta­sis and throm­bo­sis lab­o­ra­tory tests in the pres­ence of DOACs.
Conflict-of-inter­est dis­clo­sures
Karen A. Moser: dis­close no rel­e­vant con­flicts of inter­est.
Kristi J. Smock: dis­close no rel­e­vant con­flicts of inter­est.
Off-label drug use
Karen A. Moser: nothing to disclose.
Kristi J. Smock: nothing to disclose.
update of the guide­lines for lupus anti­co­ag­u­lant detec­tion and inter­pre­ta­
tion. J Thromb Haemost. 2020;18(11):2828-2839.
14. Favaloro EJ, Mohammed S, Curnow J, Pasalic L. Laboratory test­ing for lupus
anti­co­ag­u­lant (LA) in patients tak­ing direct oral anti­co­ag­u­lants (DOACs):
poten­tial for false pos­i­tives and false neg­a­tives. Pathology. 2019;51(3):292-
300.
15. Smock KJ, Plumhoff EA, Meijer P, et al. Protein S test­ing in patients with
pro­tein S defi­ciency, fac­tor V Leiden, and rivaroxaban by North Amer­i­can
Specialized Coagulation Laboratories. Thromb Haemost. 2016;116(1):50-57.
16. Moore GW, Van Cott EM, Cutler JA, Mitchell MJ, Adcock DM; Subcommittee
on Plasma Coagulation Inhibitors. Recommendations for clin­i­cal lab­o­ra­
tory test­ing of acti­vated pro­tein C resis­tance; com­mu­ni­ca­tion from the
SSC of the ISTH. J Thromb Haemost. 2019;17(9):1555-1561.
17. Douxfils J, Adcock DM, Bates SM, et  al. 2021 Update of the International
Council for Standardization in Haematology rec­ om­men­ da­ tions for lab­
o­ra­tory mea­sure­ment of direct oral anti­co­ag­u­lants. Thromb Haemost.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/129/1851777/129moser.pdf by guest on 13 December 2021

2021;121(8):1008-1020.
Correspondence 18. Platton S, Hunt C. Influence of DOAC Stop on coag­ u­
la­tion assays in
sam­ples from patients on rivaroxaban or apixaban. Int J Lab Hematol.
Karen A. Moser, ARUP Laboratories, 500 Chipeta Way, Mail Stop
2019;41(2):227-233.
115-G04, Salt Lake City, UT 84108; e-mail: karen​­.moser@hsc​­.utah​ 19. Frans G, Meeus P, Bailleul E. Resolving DOAC inter­fer­ence on aPTT, PT, and
­.edu. lupus anti­co­ag­u­lant test­ing by the use of acti­vated car­bon. J Thromb Hae-
most. 2019;17(8):1354-1362.
20. Godier A, Dincq A-S, Martin A-C, et al. Predictors of pre-pro­ce­dural con­cen­
References tra­tions of direct oral anti­co­ag­u­lants: a pro­spec­tive multicentre study. Eur
1. Gosselin RC, Adcock DM, Bates SM, et al. International Council for Standard­
Heart J. 2017;38(31):2431-2439.
ization in Haematology (ICSH) rec­om­men­da­tions for lab­o­ra­tory mea­sure­
21. Ratzinger F, Lang M, Belik S, et  al. Lupus-anti­co­ag­u­lant test­ing at NOAC
ment of direct oral anti­co­ag­u­lants. Thromb Haemost. 2018;118(3):437-450.
trough lev­els. Thromb Haemost. 2016;116(2):235-240.
2. Gosselin R, Grant RP, Adcock DM. Comparison of the effect of the anti-Xa
22. College of Amer­i­can Pathologists. Hematology and Coagulation Check-
direct oral anti­co­ag­u­lants apixaban, edoxaban, and rivaroxaban on coag­u­
list. College of Amer­i­can Pathologists; 2021.
la­tion assays. Int J Lab Hematol. 2016;38(5):505-513.
23. CLSI. Laboratory Testing for the Lupus Anticoagulant; Approved Guide-
3. Bonar R, Favaloro EJ, Mohammed S, Pasalic L, Sioufi J, Marsden K. The effect
line. CLSI doc­u­ment H60-A. Clinical and Laboratory Standards Institute;
of dabigatran on haemostasis tests: a com­pre­hen­sive assess­ment using in
2014.
vitro and ex vivo sam­ples. Pathology. 2015;47(4):355-364.
24. Harenberg J, Reichel T, Malsch R, Hirsh J, Rustagi P. Multicentric eval­u­a­
4. Bonar R, Favaloro EJ, Mohammed S, et al. The effect of the direct fac­tor Xa
tion of heparinase on aPTT, throm­bin clot­ting time and a new PT reagent
inhib­i­tors apixaban and rivaroxaban on haemostasis tests: a com­pre­hen­sive
based on recom­ bi­
nant human tis­ sue fac­tor. Blood Coagul Fibrinolysis.
assess­ment using in vitro and ex vivo sam­ples. Pathology. 2016;48(1):60-71.
1996;7(4):453-458.
5. Van Blerk M, Bailleul E, Chatelain B, et al. Influence of apixaban on com­
25. Jilma-Stohlawetz P, Lysy K, Sunder-Plassmann R, et al. Limitations of a cali­
monly used coag­u­la­tion assays: results from the Bel­gian national exter­nal
brated, quan­ti­ta­tive APC-R assay under rou­tine con­di­tions. Int J Lab Hema-
qual­ity assess­ment scheme. Int J Lab Hematol. 2017;39(4):402-408.
tol. 2021;43(2):318-323.
6. Van Blerk M, Bailleul E, Chatelain B, et  al. Influence of dabigatran and
26. Favaloro EJ, Gilmore G, Bonar R, et al. Reducing the effect of DOAC inter­
rivaroxaban on rou­tine coag­u­la­tion assays: a nation­wide Bel­gian sur­vey.
fer­ence in lab­o­ra­tory test­ing for fac­tor VIII and fac­tor IX: a com­par­a­tive
Thromb Haemost. 2015;113(1):154-164.
study using DOAC Stop and andexanet alfa to neu­ tral­
ize rivaroxaban
7. Hillarp A, Gustafsson KM, Faxälv L, et al. Effects of the oral, direct fac­tor
effects. Haemophilia. 2020;26(2):354-362.
Xa inhib­i­tor apixaban on rou­tine coag­u­la­tion assays and anti-FXa assays.
27. Baker SA, Jin J, Pfaffroth C, Vu T, Zehnder JL. DOAC–Stop in lupus anti­co­
J Thromb Haemost. 2014;12(9):1545-1553.
ag­u­lant test­ing: direct oral anti­co­ag­u­lant inter­fer­ence removed in most
8. Hillarp A, Baghaei F, Fagerberg Blixter I, et  al. Effects of the oral, direct
sam­ples. Res Pr Thromb Haemost. 2021;5(2):314-325.
fac­ tor Xa inhib­i­
tor rivaroxaban on com­ monly used coag­ u­la­
tion assays.
28. Kopatz WF, Brinkman HJM, Meijers JCM. Use of DOAC Stop for elim­i­na­
J Thromb Haemost. 2011;9(1):133-139.
tion of anti­co­ag­u­lants in the throm­bin gen­er­a­tion assay. Thromb Res.
9. Douxfils J, Ageno W, Samama C-M, et  al. Laboratory test­ing in patients
2018;170:97-101.
treated with direct oral anti­co­ag­u­lants: a prac­ti­cal guide for cli­ni­cians.
29. Favresse J, Lardinois B, Sabor L, et al. Evaluation of the DOAC-Stop® pro­ce­
J Thromb Haemost. 2018;16(2):209-219.
dure to over­come the effect of DOACs on sev­eral thrombophilia screen­ing
10. De Kesel PM, Devreese KMJ. Direct oral anti­co­ag­u­lant adsorp­tion: impact
tests. TH Open. 2018;2(2):e202-e209.
on lupus anti­co­ag­u­lant test­ing—review of the lit­er­a­ture and eval­u­a­tion on
30. Ząbczyk M, Natorska J, Kopytek M, Malinowski KP, Undas A. The effect of
spiked and patient sam­ples. J Thromb Haemost. 2020;18(8):2003-2017.
direct oral anti­co­ag­u­lants on anti­throm­bin activ­ity test­ing is abolished by
11. Siriez R, Dogné J-M, Gosselin R, Laloy J, Mullier F, Douxfils J. Comprehen­
DOAC-stop in venous throm­bo­em­bo­lism patients. Arch Pathol Lab Med.
sive review of the impact of direct oral anti­co­ag­u­lants on thrombophilia
2021;145(1):99-104.
diag­nos­tic tests: prac­ti­cal rec­om­men­da­tions for the lab­o­ra­tory. Int J Lab
31. Riva N, Vella K, Hickey K, et al. The effect of DOAC-Stop(®) on sev­eral oral
Hematol. 2021;43(1):7-20.
and par­en­teral anti­co­ag­u­lants. Int J Lab Hematol. 2021;43(4):171-175.
12. Tripodi A, Cohen H, Devreese KMJ. Lupus anti­co­ag­u­lant detec­tion in anti­
32. Sevenet PO, Cucini V, Hervé T, et al. Evaluation of DOAC Filter, a new device
coagulated patients: guid­ ance from the Scientific and Standardization
to remove direct oral anti­co­ag­u­lants from plasma sam­ples. Int J Lab Hema-
Committee for lupus anti­ co­
ag­u­
lant/antiphospholipid antibodies of the
tol. 2020;42(5):636-642.
International Society on Thrombosis and Haemostasis. J Thromb Haemost.
2020;18(7):1569-1575.
13. Devreese KMJ, de Groot PG, de Laat B, et al. Guidance from the Scientific
and Standardization Committee for lupus anti­co­ag­u­lant/antiphospholipid © 2021 by The Amer­i­can Society of Hematology
antibodies of the International Society on Thrombosis and Haemostasis: DOI 10.1182/hema­tol­ogy.2021000241

Prakash Singh Shekhawat

DOAC inter­fer­ence  | 133
 WHAT DO APLASTIC ANEMIA, PNH, AND “ HYPOPLASTIC ” LEUKEMIA HAVE IN COMMON ?

The clinical and laboratory evaluation of patients

with suspected hypocellular marrow failure

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/134/1852140/134geddis.pdf by guest on 13 December 2021

Siobán Keel1 and Amy Geddis2
University of Washington, Seattle, WA; and 2Seattle Children’s Hospital, Seattle, WA
1

The overlap in clinical presentation and bone marrow features of acquired and inherited causes of hypocellular marrow
failure poses a significant diagnostic challenge in real case scenarios, particularly in nonsevere disease. The distinction
between acquired aplastic anemia (aAA), hypocellular myelodysplastic syndrome (MDS), and inherited bone marrow fail-
ure syndromes presenting with marrow hypocellularity is critical to inform appropriate care. Here, we review the workup
of hypocellular marrow failure in adolescents through adults. Given the limitations of relying on clinical stigmata or fam-
ily history to identify patients with inherited etiologies, we outline a diagnostic approach incorporating comprehensive
genetic testing in patients with hypocellular marrow failure that does not require immediate therapy and thus allows time
to complete the evaluation. We also review the clinical utility of marrow array to detect acquired 6p copy number-neutral
loss of heterozygosity to support a diagnosis of aAA, the complexities of telomere length testing in patients with aAA,
short telomere syndromes, and other inherited bone marrow failure syndromes, as well as the limitations of somatic muta-
tion testing for mutations in myeloid malignancy genes for discriminating between the various diagnostic possibilities.

LEARNING OBJECTIVES
• Frame a comprehensive clinical and laboratory evaluation of hypocellular marrow failure based on disease severity
• Recognize that acquired 6p CN-LOH in the context of hypocellular marrow failure favors a diagnosis of aAA
• Recognize that germline mutations in SAMD9, SAMD9L, and GATA2 are enriched among young patients with MDS
with chromosome 7 abnormalities
• Recognize the clinical utility and limitations of telomere length testing by flow-FISH in discriminating between
causes of hypocellular marrow failure

CLINICAL CASE
A 28-year-old woman seeks treatment for fatigue and easy
bruising. Complete blood count reveals a hemoglobin of
10g/dL, absolute neutrophil count of 1.5×109/L, platelet
count of 45×109/L, and an absolute reticulocyte count of
70 000/µL. Marrow aspirate is hypocellular for age with
no frank morphologic dysplasia. Trephine biopsy specimen
shows an estimated marrow cellularity of 30% and no retic-
ulin fibrosis. Flow cytometry shows normal myeloid matura-
tion and no abnormal or expanded myeloid blast population.
Defining hypocellular marrow failure, the differential
diagnosis, and rationale for workup
Hypocellular marrow failure is characterized by peripheral
cytopenia(s) and bone marrow cellularity deemed low for
age and results from acquired and inherited causes that can
overlap in clinical presentation and bone marrow features.
For the purpose of this manuscript and to avoid ambiguity,
we restrict the term aplastic anemia to the immunolog-
ically mediated disease entity, acquired aplastic anemia
(aAA). Our workup centers on classifying patients as hav-
ing hypocellular marrow failure requiring urgent therapy,
including disease defined as severe by modified Camitta’s
criteria1 (Table 1) or nonsevere hypocellular marrow failure,
which we define as cytopenia(s) and a hypocellular mar-
row for age and not meeting criteria for severe disease and
who do not require urgent therapy.
Causes of hypocellular marrow failure are shown in
Table 2. aAA is bimodal in its age at presentation, is most
commonly idiopathic, and accounts for most patients with
severe disease.3 Rarely and enriched among pediatric pre-
sentations, aAA is due to an underlying inborn error in

134 | Hematology 2021 | ASH Education Program

Table 1.  Severe hypocellular mar­row fail­ure clas­si­fi­ca­tion
cri­te­ria
Marrow cel­lu­lar­ity
<25% or
25%-50% with <30% resid­ual hema­to­poi­etic cells
AND cytopenias (at least 2 of 3)
Absolute neu­tro­phil count <500 × 109/L
Platelets <20 × 109/L
Absolute retic­u­lo­cyte count <60 × 109/L
The immu­no­log­i­cally medi­ated dis­ease, severe aAA, accounts for the
major­ity of severe hypocellular mar­row fail­ure.
immu­nity such as X-linked lymphoproliferative dis­or­der.4 Accu-
rate diag­no­sis of patients with nonsevere hypocellular mar­row
fail­ure is clin­i­cally more chal­leng­ing and requires con­sid­er­ation
of hypocellular myelodysplastic syn­drome (MDS) with or with­
out an under­ly­ing clas­si­cal inherited bone mar­row fail­ure syn­
drome or inherited mye­loid malig­nancy pre­dis­po­si­tion syn­drome
(IBMFS/MMP). Among the most com­mon clas­si­cal IBMFSs, Fan-
coni ane­mia (FA), short telo­mere syn­dromes, and Shwachman-Di-
amond syn­drome (SDS) may pres­ent with hypocellular mar­row
fail­ure. While rare and in con­trast to MDS in the elderly, child­hood
MDS is most com­monly hypocellular,5 and also, approx­i­ma­tely 15%
of adults with MDS have a hypocellular mar­row.6 Further com­pli­
cat­ing the diag­no­sis, patients with IBMFS/MMP (eg, muta­tions in
GATA27) may have cytopenias and mega­kar­yo­cytic atypia with­
out overt malig­nant clonal dis­ease that can be mis­taken for MDS.
Important diag­nos­tic pos­si­bil­i­ties to con­sider in young per­sons
with hypocellular MDS char­ac­ter­ized by chro­mo­some 7 abnor­
mal­i­ties are GATA2 defi­ciency8,9 and SAMD9/9L dis­or­ders.10 While
IBMFS/MMP may be suspected based on per­ti­nent his­tory and
phys­i­cal exam find­ings (Table 3), cryp­tic pre­sen­ta­tions of these
dis­or­ders are increas­ingly rec­og­nized and thus war­rant con­sid­
er­ation even in the absence of these clues.3 Last, in all­patients,
the poten­tial of a revers­ible cause—namely, med­i­ca­tion-induced
hypocellular mar­row fail­ure—should be excluded.
The cor­rect diag­no­sis of hypocellular mar­row fail­ure is clin­
i­
cally par­ a­mount. Patients with severe aAA are treated with
either hema­to­poi­etic stem cell trans­plant (HSCT) or immu­no­sup­
pres­sive ther­apy (IST) with equine antithymocyte glob­u­lin and
cyclo­spor­ine with or with­out the addi­tion of a thrombopoietin
recep­tor ago­nist.11 Marrow fail­ure due to IBMFS/MMP or MDS is
unlikely to respond to IST, and HSCT may be indi­cated. Recog-
nition of a her­i­ta­ble dis­or­der has impor­tant impli­ca­tions to the
selec­tion of an appro­pri­ate con­di­tion­ing reg­i­men, eval­u­a­tion of
poten­tial related stem cell donors, and sur­veil­lance for hema­to­
poi­etic and nonhematopoietic com­pli­ca­tions, as well as coun­sel­
ing for fam­ily mem­bers.
Clinical and lab­o­ra­tory eval­u­a­tion of hypocellular
mar­row fail­ure
A diag­nos­tic approach to hypocellular mar­row fail­ure is shown
in Figure 1. Notably, since mar­row cel­lu­lar­ity can be patchy, it
is crit­i­cal that the biopsy spec­i­men be of ade­quate size (aim
for 2 cm) and qual­ity to ensure it is rep­re­sen­ta­tive. Numerous
guide­ lines detailing the recommended and suggested ini­ tial
lab­o­ra­tory stud­ies of hypocellular mar­row fail­ure exist.12–14 Our
recommended and suggested lab­ o­
ra­tory stud­ ies are listed
Prakash Singh Shekhawat
Table 2.  Differential diag­no­sis of hypocellular mar­row fail­ure
Acquired Inherited
aAA Classical IBMFS/MMP
Fanconi anemia
Anorexia nervosa
Short telo­mere syn­dromes
Shwachman-Diamond Syndrome
GATA2 defi­ciency
SAMD9/SAMD9L dis­or­ders
Hypocellular myelodysplastic Inborn errors of immu­nity (eg, X-linked
syndrome lymphoproliferative dis­or­der)
Medications/tox­ins
This table is not exhaus­tive and is intended to cap­ture the more com­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/134/1852140/134geddis.pdf by guest on 13 December 2021
monly encoun­ter enti­ties that can pres­ent as aplastic ane­mia. In our
clin­i­cal expe­ri­ence, the IBMFSs/MMPs listed here are not always
char­ac­ter­ized by hypocellular mar­rows. Gelatinous deg­ra­da­tion of the
bone mar­row also may be seen in anorexia nervosa.2
in Table 3. A tai­lored his­tory and phys­i­cal exam­i­na­tion can be
help­ful in iden­ti­fy­ing the under­ly­ing cause (Table 4), par­tic­u­larly
doc­u­men­ta­tion of prior com­plete blood counts as longstanding
cytopenias or mac­ro­cy­to­sis can indi­cate a her­i­ta­ble cause. As
IBMFS/MMP can occur with­out clin­i­cal stig­mata or fam­ily his­tory,
an unre­mark­able his­tory and exam­i­na­tion do not reli­ably exclude
these dis­or­ders; the pres­ence of select dis­ease fea­tures should
raise clin­i­cal sus­pi­cion (Table 5). If an MDS-defin­ing cyto­ge­netic
abnor­mal­ity is absent, the dis­tinc­tion between aAA and hypo-
cellular MDS is based on mor­pho­logic fea­tures (dys­pla­sia and in-
creased blasts).15 Notably, cyto­ge­netic abnor­mal­i­ties can occur
in aAA. Trisomy 8 and del(13q) are seen in aAA, where both carry
a favor­able prog­no­sis and are asso­ci­ated with response to IST.16,17
Hypocellular MDS, par­ tic­
u­larly among young patients, should
prompt con­sid­er­ation of an under­ly­ing IBMFS/MMP.
Diagnostic test­ing in hypocellular mar­row fail­ure requir­ing
ther­apy (Figure 1A) is done simul­ta­neously, rather than sequen­
tially, to allow defin­i­tive treat­ment to start expe­di­tiously to limit
trans­fu­sions and the risk of seri­ous infec­tion. We aim to ini­ti­ate
treat­ment within approx­i­ma­tely 3 weeks of pre­sen­ta­tion, which
allows time to com­plete diag­nos­tic stud­ies and donor eval­u­a­
tions. Evaluation of hypocellular mar­ row fail­
ure not requir­ ing
urgent ther­apy can typ­i­cally pro­ceed in a more step­wise fash­ion
(Figure 1B). There are clin­i­cal set­tings where genetic test­ing for
IBMFS/MMP may not be per­ti­nent (eg, elderly patient with mul­
ti­ple comorbidities and no poten­tially affected fam­ily mem­bers
in whom test­ing would not inform the patient’s or fam­ily’s care).
CLINICAL CASE (Con­tin­ued)
Old records are requested and doc­ u­
ment sev­ eral nor­
mal
past com­ plete blood counts. Medical his­ tory and exam­ i­
na­
tion are unre­mark­able. Chromosomal break­age stud­ies return
unre­mark­able. Telomere length test­ing falls in the 30th age-
adjusted per­cen­tile.
Chromosomal break­age test
Given the impact a diag­no­sis of FA has on treat­ment and sur­veil­
lance strat­e­gies, a chro­mo­somal break­age test is indi­cated in
Evaluation of hypocellular mar­row fail­ure | 135
Table 3.  Suggested ini­tial lab­o­ra­tory eval­u­a­tion of hypocellular mar­row fail­ure

Implications for
Study Source Diagnosis treat­ment
Peripheral blood
Flow cytometry for PNH PB aAA IST
Chromosomal break­age test­ing PB FA HSCT
Modified con­di­tion­ing
Telomere lengths by flow-FISH PB STS HSCT
Modified con­di­tion­ing
Immunoglobulins, lym­pho­cyte sub­sets, natural killer PB Inborn error of immu­nity, GATA2 defi­ciency ­ HSCT
cell func­tion syn­drome, XLP Modified con­di­tion­ing
Infection pro­phy­laxis

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/134/1852140/134geddis.pdf by guest on 13 December 2021

Pancreatic isoamylase PB SDS
HLA typ­ing on patient and full sib­lings PB Severe aAA or MDS HSCT
Donor test­ing
Bone mar­row
Morphologic review of mar­row BM MDS, GATA2 defi­ciency—mega­kar­yo­cytic atypia
Routine kar­yo­type BM MDS, IBMFS/MMP HSCT
FISH BM MDS, IBMFS/MMP HSCT
includ­ing −7/del, −5/del, +8, del(q20)
Chromosome geno­mic array BM 6p CN-LOH in aAA; germline CNAs in
IBMFS/MMP; risk strat­i­fi­ca­tion of NK MDS, 7q
LOH in SAMD9/9L dis­or­ders
Somatic multigene genetic test­ing for muta­tions in BM See text for complexities
mye­loid malig­nancy genes
Cultured skin fibro­blasts
Germline multigene genetic test­ing Fibroblasts* IBMFS/MMP HSCT
Modified con­di­tion­ing
Donor test­ing
Studies recommended in all­patients are bolded.
*When clin­i­cally pos­si­ble, cul­tured skin fibro­blasts are the recommended DNA source for germline test­ing.
BM, bone mar­row aspi­rate; NK, nor­mal kar­yo­type; PB, periph­eral blood; PNH, par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria.

all­patients with hypocellular mar­row fail­ure. Abnormal test re-
sults should be followed with genetic test­ing of a germline tis­sue
source to try to iden­tify the caus­a­tive muta­tion as this pro­vi­des
addi­tional prog­nos­tic infor­ma­tion regard­ing can­cer and other
risks as well as facil­i­tates car­rier test­ing for fam­ily mem­bers.18
Breakage stud­ies allow diag­no­sis of patients whose muta­tions
are missed by stan­dard genetic test­ing and inform the func­tional
con­se­quence of genetic var­i­ants that might oth­er­wise be of un-
clear clin­i­cal sig­nif­i­cance. This test­ing is performed by cul­tur­ing
either periph­eral blood lym­pho­cytes or skin fibro­blasts in the
pres­ence of DNA cross-linking agents and com­par­ing the num­
ber of chro­mo­somal breaks, includ­ing rearrangements, gaps, en-
doreduplications, and exchanges, to con­trols under base­line and
stim­u­lated con­di­tions. Results are reported as both the num­ber
of cells with abnor­mal breaks and the num­ber of breaks per cell.
Although periph­eral blood lym­pho­cytes are the most acces­si­
ble source of tis­sue for break­age and genetic stud­ies, lym­pho­
cytes are sus­cep­ti­ble to the phe­nom­e­non of somatic mosa­i­cism
resulting from expan­sion of hema­to­poi­etic stem cells that have
under­gone molec­u­lar rever­sion that cor­rects the FA phe­no­type.
This is esti­mated to occur in approx­i­ma­tely 10% to 25% of pa-
tients with FA.19 Somatic rever­sion may be suspected if fra­gil­ity
test­ing performed on lym­pho­cytes shows 2 pop­ul­a­tions of cells,
some appearing nor­mal and oth­ers with mul­ti­ple breaks per cell.
Negative results should be con­firmed in cul­tured fibro­blasts if FA
is strongly suspected. Breakage stud­ies on fibro­blasts can also
be con­sid­ered when stud­ies in lym­pho­cytes fail due to severe
leu­ko­pe­nia or poor growth in cul­ture.
Complexities of telo­mere length test­ing
The short telo­mere syn­dromes are a group of genetic dis­or­ders
that are caused by muta­tions in com­po­nents of the telomerase
enzyme and other telo­mere main­te­nance genes. Telomere length
test­ing by flow cytom­et­ ry and fluo­res­cence in situ hybrid­iza­tion
(flow-FISH) is the gold stan­dard because of its high repro­duc­ibil­
ity. It mea­sures sin­gle-cell telo­mere length using a fluorescently
labeled probe that hybrid­izes to telo­mere DNA. Clinical test­ing
is performed on a periph­eral blood sam­ple, and avail­­able assays
report results in total lym­pho­cytes and granulocytes (so called
2-panel test­ing) or granulocytes and total lym­pho­cytes and lym­
pho­cyte sub­sets (naive T cells, mem­ory T cells, B cells, nat­u­ral
killer cells), so-called 6-panel test­ing. Two-panel test­ing appears
suf­fi­cient for the ini­tial diag­nos­tic workup of a patient with hypo-
cellular mar­row fail­ure.20

136  |  Hematology 2021  |  ASH Education Program

 Initial work-up to include:
Pancytopenia & - History and physical examination (see Table 4)
hypocellular marrow CBC, retic, and marrow cellularity c/w - Cytogenetics (routine karyotype, FISH, and array) on BM
severe aplastic anemia - PNH screen on PB
- Chromosomal breakage test on PB
Laboratory work-up and HLA - Telomere length by flow-FISH
typing underway - Pancreatic isoamylase
- HLA typing on patient and full siblings
1 wk
Normal morphology and cytogenetics? Consider Consider underlying IBMFS/MMP
No hypocellular MDS
Yes

2 wk Negative or Initial work-up

PNH or 6p CN-LOH positive positive

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/134/1852140/134geddis.pdf by guest on 13 December 2021

Severe aAA IBMFS/MMP
-PNH or 6p CN-LOH
further supports diagnosis

3 wk
Rapid panel-based NGS
Initiate therapy testing for IBMFS/MMP
(IST or HSCT) of patient & related
donor on PB

If severe lymphopenia, MDS, or suspicion for IBMF/AL-MDS obtain a skin biopsy to culture
skin fibroblasts in case needed for chromosomal fragility testing or germline genetic testing.
If clinical suspicion of Shwachman-Diamond Syndrome.
Is an MDS-defining finding present?

Pancytopenia & CBC, retic, and marrow cellularity consistent Initial work-up to include:
hypocellular marrow with non-severe hypocellular marrow failure - History and physical examination Table 4)
(not requiring urgent therapy) - Cytogenetics (routine karyotype, FISH, and array) on BM
- PNH screen on PB
- Chromosomal breakage study on PB
Laboratory work-up underway
- Telomere length by flow-FISH
1 wk

Initial work-up
2 wk Negative and Positive or
PNH or 6p CN-LOH positive PNH/6p CN-LOH negative

• Consider targeted capture or WES of myeloid

malignancy genes (include BCOR/BCORL) on BM

3 wk Skin biopsy to culture

Nonsevere aAA
fibroblasts

Panel-based NGS testing for IBMFS/MMP • There are clinical settings where genetic testing
on fibroblast DNA IBMFS/MMP may not be pertinent
6 wk & syndrome specific testing
Monitor

IBMFS/MMP and/or hypocellular MDS or aAA

Does not reliably distinguish among causes of hypocellular marrow failure; may provide some prognostic information.
Serum pancreatic isoamylase for Shwachman-Diamond Syndrome
If MDS-defining finding present.

Figure 1. Diagnostic approach to hypocellular marrow failure based on severity. (A) Diagnostic approach to severe hypocellu-
lar marrow failure requiring therapy—time-limited. (B) Diagnostic approach to nonsevere hypocellular marrow failure not requiring
therapy—not time-limited. Time-limited indicates the urgency to get the workup completed before starting therapy. BM, bone mar-
row aspirate; PB, peripheral blood; PNH, paroxysmal nocturnal hemoglobinuria.

Prakash Singh Shekhawat

Evaluation of hypocellular mar­row fail­ure | 137
Table 4.  Focused med­i­cal his­tory and phys­i­cal exam­i­na­tion

Questions Supportive infor­ma­tion Implications

Presentation Duration of symp­toms Prior CBC/diff Inherited vs acquired
Bleeding Platelet count, coag­u­la­tion stud­ies Transfusion require­ments
Fatigue Hemoglobin, pulse oximetry, CXR Transfusion require­ments
S/Sx of infec­tion Fever, respi­ra­tory symp­toms Identify and treat infec­tions
Medical his­tory Adrenal hypo­pla­sia SAMD9
Ataxia Brain MRI SAMD9L
Atypical infec­tions Immunologic test­ing, vac­ci­na­tion his­tory SDS, GATA2, STS, SAMD9/9L
Birth his­tory Birth weight, ges­ta­tional age, IUGR, ane­mia DBA, FA, SAMD9/9L, SDS, STS

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/134/1852140/134geddis.pdf by guest on 13 December 2021

Cancer or leu­ke­mia Type of che­mo­ther­apy and toxicities IBMFS, sec­ond­ary MDS
Congenital malformations Echo, abdom­i­nal US, brain MRI, x-rays FA
GI (diar­rhea, dif­fi­culty swallowing) LFTs, abdom­i­nal US, endos­copy aAA, SDS, STS, PNH
Growth Growth chart FA
Hearing loss Audiogram FA, GATA2
Nutrition (vegan, weight loss, sup­ple­ments) B12/MMA/HC, ceru­lo­plas­min/cop­per B12, folate, cop­per defi­ciency
Pulmonary dis­ease (TE fis­tula, fibro­sis, CXR, CT, PFTs, bron­chos­copy FA, GATA2, STS, SAMD9/9L
pneu­mo­nia, alve­o­lar proteinosis)
Renal dis­ease (con­gen­i­tal anom­aly, dark urine) Urinalysis, renal US FA, PNH
Skeletal dys­pla­sia, skel­e­tal abnor­mal­i­ties X-rays SDS, FA, DBA
Family his­tory Ancestry Founder muta­tions in cer­tain
pop­u­la­tions
Consanguinity Autosomal reces­sive dis­or­ders
Full sib­lings Potential mar­row donors
Marrow fail­ure, MDS, leu­ke­mia, can­cer IBMFS
Examination find­ings Syndromic facies IBMFS
Microcephaly STS, SAMD9/9L
Leukoplakia STS
Hepatomegaly SDS
Abnormal thumbs, radial abnor­mal­i­ties FA
Nail dys­pla­sia STS
Lymphedema GATA2
Café-au-lait spots FA
Reticulated pig­men­ta­tion on neck STS
Warts GATA2
CBC/diff, com­plete blood count with dif­fer­en­tial; CT, com­put­er­ized tomog­ra­phy; CXR, chest x-ray; GI, gas­tro­in­tes­ti­nal; IUGR, intra­uter­ine growth
retar­da­tion; GATA2, GATA2 defi­ciency; HC, homocysteine; LFT, liver function testing; MMA, methylmalonic acid; mag­netic res­o­nance imag­ing;
PFT, pul­mo­nary func­tion test; S/Sx, signs/symp­toms; SAMD9/9L, SAMD9/SAMD9L dis­or­ders; TE, tracheoesophageal fis­tula; US, ultrasound.

Telomere length (TL) assess­ment is recommended in the ini­tial
diag­nos­tic workup of hypocellular mar­row fail­ure. TL assess­ment
is use­ful in rul­ing out a diag­no­sis of a con­sti­tu­tional short telo­mere
syn­drome upfront in patients with mar­ row fail­ ure. Lymphocyte
telo­mere length above the age-adjusted 50th per­ cen­tile has a
100% neg­a­tive pre­dic­tive value in iden­ti­fy­ing a clin­i­cally mean­ing­
ful muta­tion in a short telomere syndrome (STS) gene in pedi­at­
ric and adult patients.21 Among patients youn­ger than 40 years, a
lym­pho­cyte telo­mere length below the age-adjusted first per­cen­
tile strongly supports a diagnosis of a STS but lack specificity.21,22
Telomere lengths below the age-adjusted first per­ cen­tile have
been reported in patients with other IBMFSs, includ­ ing FA and
SDS,21,23 and so con­sid­er­ation of the clin­i­cal pre­sen­ta­tion and addi­
tional genetic test­ing results are nec­es­sary to estab­lish the cor­rect
diag­no­sis. In a pre­dom­i­nantly pedi­at­ric and youn­ger adult cohort
of patients youn­ger than 40 years (0-31 years; median, 14 years)
presenting with idi­o­pathic mar­row fail­ure, all­of the patients with
immu­no­ther­apy-respon­sive mar­row fail­ure had lym­pho­cyte telo­
mere lengths above the age-adjusted first per­cen­tile.21 Nearly half of
the patients with genet­i­cally con­firmed short telo­mere syn­dromes
older than 40 years have telo­mere length mea­sure­ments above
the age-adjusted first per­cen­tile, and so telo­mere lengths above

138  |  Hematology 2021  |  ASH Education Program

Table 5.  Features of select inherited bone mar­row fail­ure and inherited MDS/leu­ke­mia pre­dis­po­si­tion syn­dromes that
may pres­ent as aplastic ane­mia

Syndrome Congenital find­ings Malignancy risk Screening test Genetics Other hints
Fanconi anemia Ear abnor­mal­i­ties, heart defects, short AML, MDS, GYN CA, Increased chro­mo­ FANCA, C, G account Sensitivity to
stat­ure, skin pig­men­ta­tion (café- head and neck some break­age for 95% of cases che­mo­ther­apy
au-lait spots or hypopigmentation), CA, and oth­ers
skel­e­tal anom­a­lies (thumbs, arms),
TE fis­tula, tri­an­gu­lar facies, uro­gen­
i­tal defects
GATA2 defi­ciency Lymphedema, immu­no­de­fi­ciency AML, MDS GATA2 Megakaryocyte
with atyp­i­cal atypia, pedi­at­ric
myco­bac­te­rial infec­tions MDS with mono­
somy 7, del7q, tri­

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/134/1852140/134geddis.pdf by guest on 13 December 2021

somy 8, der(1;7)
SAMD9/SAMD9L MIRAGE (SAMD9): MDS, infec­tion, AML, MDS SAMD9, SAMD9L Pediatric MDS with
dis­or­ders restric­tion of growth, adre­nal hypo­ mono­somy 7,
pla­sia, gen­i­tal phe­no­types, and del7q, or CN-LOH
enter­op­a­thy 7q
Ataxia-pan­cy­to­pe­nia syn­drome
(SAMD9L): cer­e­bel­lar atro­phy and
white mat­ter hyperintensities, gait
dis­tur­bance, nys­tag­mus
Short telo­mere Young adults: infec­tions, nail dys­tro­ AML, MDS Short telo­mere lengths DKC1, RTEL1, TERT,
syn­dromes phy, oral leukoplakia, skin hyper­ Rectal ade­no­car­ (cor­re­lates with TERC, TINF2,
pig­men­ta­tion ci­noma, SCC phe­no­type) RTEL1 account for
Adults: emphy­sema, early hair anus/oral cav­ major­ity of cases
graying, immune defi­ciency, liver ity/tongue
fibro­sis/cir­rho­sis, mac­ro­cy­to­sis,
pul­mo­nary AVMs and HPS
Shwachman- Pancreatic insuf­fi­ciency (can improve AML, MDS Low pan­cre­atic SBDS, SRP54, ELF1, Isolated neutropenia,
Diamond with age), skel­e­tal abnor­mal­i­ties isoamylase (>3 years DNAJC21 somatic muta­tions
Syndrome old) and low fecal in EIF6 and TP53
elas­tase (pediatric
and adult patients)
AML, acute myeloid leukemia; AVM, arteriovenous malformations; CA, cancer; GYN CA, gynecological cancers; HPS, hepatopulmonary syndrome;
SCC, squamous cell carcinoma; TE, tracheoesophageal fis­tula.

this thresh­old do not exclude the diag­no­sis of a short telo­mere
syn­drome in older patients and requires diagnosis here addi­tional
clin­i­cal and genetic infor­ma­tion.21,22 Telomere length assess­ment in
var­i­ous causes of hypocellular mar­row fail­ure are shown in Figure 2.
CLINICAL CASE (Con­t in­u ed)
Marrow kar­ yo­
type returns 46,XX. Microarray on the mar­ row
aspi­rate sam­ple iden­ti­fied copy number-neutral loss of het­ero­
zy­gos­ity of 6p in 10% of the cells. Flow cytometry on periph­eral
blood did not iden­tify any glycosylphosphatidylinositol (GPI)-
defi­cient pop­u­la­tions.
Diagnostic util­ity of select genetic abnor­mal­i­ties
in hypocellular mar­row fail­ure
Acquired copy number-neutral loss of het­ ero­zy­
gos­
ity of
chro­mo­some arm 6p and chro­mo­somal microarray test­ing on
plat­forms includ­ing sin­gle-nucle­o­tide poly­mor­phism probes
More than 70% of patients (and over 60% of pedi­at­ric patients) with
aAA have clonal hema­to­poi­e­sis with somatic muta­tions detected
by whole-exome sequenc­ing or sin­gle-nucle­o­tide poly­mor­phism
Prakash Singh Shekhawat
array early in the course of their dis­ease.24,25 Clonal hema­to­poi­e­sis
does not equate to malig­nancy and reflects an acquired genet-
ic change in a hema­to­poi­etic stem cell, allowing for com­pet­i­tive
out­growth of the mutated hema­to­poi­etic stem cell’s prog­eny. The
2 most com­mon somatic muta­tions in aAA are muta­tional inac­ti­
va­tion of PIGA and the loss of HLA class I alleles. The clin­i­cal util­ity
of detecting PIGA loss or the func­tional con­se­quence of a detect­
able par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria clone in hypocellular
mar­row fail­ure is cov­ered in Education Session entitled “When
does a PNH clone have clinical significance?”26 PIGA loss is most
com­monly diag­nosed by flow-cytometric anal­y­sis to detect cells
lacking GPI-linked pro­teins. This study is more sen­si­tive when per-
formed on a periph­eral blood sam­ple com­pared with a bone mar­
row aspi­rate sam­ple as the GPI-linked pro­teins typ­i­cally assessed
are most highly expressed on periph­eral blood cells. Thus, test­ing
on periph­eral blood as opposed to a mar­row sam­ple for a par­ox­
ys­mal noc­tur­nal hemo­glo­bin­uria clone is recommended.27
Loss of HLA class I alleles occurs either through a loss-of-func­tion
muta­tion in 1 of the HLA class I genes or, more com­monly, through
the loss of 1 paren­tal HLA hap­lo­type through the acqui­si­tion of
copy number-neutral loss of het­ero­zy­gos­ity of chro­mo­some arm
6p (6p CN-LOH) involv­ing the human leu­ko­cyte anti­gen locus. A
prevailing hypoth­e­sis is that the loss of HLA class I alleles dis­rupts
Evaluation of hypocellular mar­row fail­ure | 139
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/134/1852140/134geddis.pdf by guest on 13 December 2021
Figure 2.  Utility of telomere length testing in the diagnosis of hypocellular marrow failure. The red circles denote patients with
bone marrow failure secondary to a genetically confirmed diagnosis of a short telomere syndrome. The blue circles denote patients
with presumed idiopathic acquired aplastic anemia based on response to treatment with IST. The green circles denote patients with
bone marrow failure due to an underlying genetically confirmed inherited bone marrow failure syndrome apart from a short telomere
syndrome (LIG4, RUNX1, GATA2). The black circles denote patients diagnosed with a short telomere syndrome at >40 years of age.
BMF, bone marrow failure. Based on data from Alder et al.21

anti­gen pre­sen­ta­tion, allowing non­dom­i­nant clones to evade the
immune sys­tem. An anal­o­gous path­o­phys­i­­ol­ogy may con­trib­ute
to posttransplant relapse in haploidentical HSCT recip­ i­
ents, in
whom the relapsed dis­ease has acquired loss of the HLA hap­lo­
type that dif­fered from the donor’s hap­lo­type, thus damp­en­ing
the graft-ver­sus-leu­ke­mia effect.26,28,29 In aAA, the HLA alleles spe­
cif­i­cally lost through 6p CN-LOH vary across age and eth­nic ori­gin
of patients. That said, stud­ies report a bias of the miss­ing HLA
allele in 6p CN-LOH to par­tic­u­lar HLA types. This, together with
the find­ing that the HLA allele is com­monly involved across the
6p CN-LOH reported in aAA, sug­gests the find­ing is linked to the
path­o­gen­e­sis of aAA and not merely a sec­ond­ary event. Although
addi­tional stud­ies are needed to clar­ify the impact of acquired 6p
CN-LOH on response to IST, the pres­ence of 6p CN-LOH appears
to be diag­nos­ti­cally infor­ma­tive in distinguishing patients with
immune-medi­ated mar­row fail­ure from those with inherited bone
mar­row fail­ure con­di­tions.30-33 Acquired 6p CN-LOH appears to be
rel­a­tively spe­cific for aAA, occur­ring in less than 1% of patients
with MDS or in nor­mal aging.30,34,35 The prev­a­lence of clonal 6p
CN-LOH among pedi­at­ric and adult patients with aAA is approx­
i­ma­tely 10% to 13%.24,33,36 Constitutional 6p homo­zy­gos­ity is not
enriched among patients with aAA.
Clinical test­ing for 6p CN-LOH requires chro­mo­somal microar-
ray anal­y­sis (CMA). Outside of detecting 6p CN-LOH, addi­tional
ratio­nale supporting inclu­sion of CMA in the workup of hypocel-
lular mar­row fail­ure includes that this test­ing can be performed
on DNA extracted from resid­ ual cell pel­ lets, aspi­
rate smears,
touch preps, or for­ma­lin-fixed, par­af­fin embed­ded tis­sues. Cells
obtained from patients with hypocellular mar­ row fail­ure may
not grow well in cul­ture for kar­yo­typic anal­y­sis. In addi­tion to
detecting CN-LOH, CMA can iden­tify microdeletions and micro-
duplications beyond the res­o­lu­tion of con­ven­tional kar­yo­type
and FISH,37 poten­tially pro­vid­ing a ratio­nale for closer clin­i­cal sur­
veil­lance for malig­nant clonal evo­lu­tion. In addi­tion, CMA may
detect copy num­ber alter­ations (CNAs) missed by some targeted
next-gen­er­a­tion sequenc­ing (NGS)–based test­ing. Causative

140  |  Hematology 2021  |  ASH Education Program

 The evaluation of hypocellular marrow failure

Inial worku
workup:
History and phy
physical
morphology and cytogene
cs
Marrow morph
- 6p CN-LOH
PNH
PNH Chromosomal b breakage test
TTelomere
Te lomere length
AA HLA typing of ppa
ent and siblings

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/134/1852140/134geddis.pdf by guest on 13 December 2021

6p CN-LOH MDS
Severe disease
- ANC <500 x 109/L
- Plt <20 x 109/L Acquired Inherited
- Abs re
c <60 x 109/L

Nonsevere disease Likely acquired Germline

- Not requiring urgent genec tesng
therapy Consider acquired or inherited

Figure 3. The evaluation of hypocellular marrow failure.

CNAs have been reported in a num­ber of IBMFs/MMPs, includ­ing
FA and oth­ers.38,39 In 1 cohort study of patients with suspected
inherited bone mar­row fail­ure, CMA iden­ti­fied germline path­o­
genic CNAs in 16.4% (11/67) of patients tested.40
Somatic muta­tion test­ing for muta­tions in mye­loid
malig­nancy genes
The pro­ file of somatic muta­ tions iden­ ti­
fied by NGS strat­ e­
gies
in aAA, pedi­at­ric MDS, hypocellular MDS in adults, and IBMFS/
MMP dif­fers from one another. An under­stand­ing of this con­text-
depen­dent clonal land­scape and its dynam­ics con­tin­ues to pro­
vide impor­tant insights into malig­nant and non­ma­lig­nant clonal
evo­lu­tion and, in select set­tings, offers prog­nos­tic insights. This
has been recently reviewed.40-42 The cur­rent diag­nos­tic util­ity of
NGS pan­els and whole-exome sequenc­ing for somatic muta­tions
in mye­loid malig­nancy genes to reli­ably dis­tin­guish among these
hypocellular mar­row fail­ure syn­dromes in the clinic remains unclear.
A spe­cific somatic muta­tional land­scape that can reli­ably iden­tify
bona fide MDS (in the absence of defin­i­tive mor­pho­logic or cyto­
ge­netic find­ings) for clin­i­cal prac­tice in this set­ting has not been
established. Approximately 25% to 30% of patients with aAA have
clonal muta­tions in genes that are asso­ci­ated with both mye­loid
malig­nancy and with aging, includ­ing ASXL1, DNMT3A, JAK2, and
TP53. These muta­tions are pres­ent at low var­i­ant allele fre­quency
sim­i­lar to what is seen in age-related clonal hema­to­poi­e­sis, and
these genes can also be mutated in adult hypocellular MDS.24,43,44
In addi­tion, although muta­tions in the epi­ge­netic mod­i­fier genes,
BCOR and BCORL1, appear sub­tly enriched in aAA com­pared with
hypocellular MDS, both dis­eases can carry these muta­tions, and
so this find­ing is not suf­fi­cient to dis­tin­guish between these dis­or­
ders.44 In con­trast to typ­i­cal adult MDS, muta­tions in the spliceo-
some com­plex (ie, SF3B1, SRSF2, ZRSR2, U2AF1) are rare across
Prakash Singh Shekhawat
refrac­tory cytopenia of child­ hood,9,11 adult hypo­plas­tic MDS,44,45
and aAA, 24,44
so this find­ing is also not infor­ma­tive diag­nos­ti­cally.
When fea­si­ble, this test­ing can be con­sid­ered for its poten­tial
prog­nos­tic value.46 As exam­ples, the acqui­si­tion of biallelic TP53
muta­tions is asso­ci­ated with devel­op­ment of leu­ke­mia in SDS,47
and in aAA, BCOR and BCORL1 muta­tions are asso­ci­ated with re-
sponse to immu­no­sup­pres­sion and a favor­able prog­no­sis.24
CLINICAL CASE (Con­tin­ued)
Based on her workup, the patient was diag­nosed with nonse-
vere aAA and is being followed by serial blood counts.
Conclusion
The eval­u­a­tion of hypocellular mar­row fail­ure (Figure 3) is based
on dis­ease sever­ity and aims to dis­tin­guish among the inherited
and acquired causes to inform opti­mal patient care.
Acknowledgments 
We thank patients and fam­i­lies for par­tici­pat­ing in mar­row fail­
ure research and Mary Armanios at Johns Hopkins University for
pro­vid­ing Figure 2.
Conflict-of-inter­est dis­clo­sures
Amy Geddis: no rel­e­vant con­flicts of inter­est.
Siobán Keel: no rel­e­vant con­flicts of inter­est.
Off-label drug use
Amy Geddis: nothing to disclose.
Siobán Keel: nothing to disclose.
Evaluation of hypocellular mar­row fail­ure | 141
Correspondence
Siobán Keel, University of Washington, Division of Hematology,
Seattle, WA 98105; e-mail: sioban@uw​­.edu.
References
1. Camitta BM, Rappeport JM, Parkman R, Nathan DG. Selection of patients
for bone mar­ row trans­ plan­
ta­
tion in severe aplastic ane­ mia. Blood.
1975;45(3):355-363.
2. Abella E, Feliu E, Granada I, et al. Bone mar­row changes in anorexia nervosa
are cor­re­lated with the amount of weight loss and not with other clin­i­cal
find­ings. Am J Clin Pathol. 2002;118(4):582-588.
3. Keel SB, Scott A, Sanchez-Bonilla M, et  al. Genetic fea­ tures of myelo-
dysplastic syn­drome and aplastic ane­mia in pedi­at­ric and young adult
patients. Haematologica. 2016;101(11):1343-1350.
4. Pachlopnik Schmid J, Canioni D, Moshous D, et al. Clinical sim­i­lar­i­ties and
dif­fer­ences of patients with X-linked lymphoproliferative syn­drome type
1 (XLP-1/SAP defi­ciency) ver­sus type 2 (XLP-2/XIAP defi­ciency). Blood.
2011;117(5):1522-1529.
5. Niemeyer CM, Baumann I. Classification of child­hood aplastic ane­mia and
myelodysplastic syn­drome. Hematol Am Soc Hematol Educ Program.
2011;2011:84-89.
6. Nazha A, Seastone D, Radivoyevitch T, et al. Genomic pat­terns asso­ci­ated
with hypo­plas­tic com­pared to hyper­plas­tic myelodysplastic syn­dromes.
Haematologica. 2015;100(11):e434-e437.
7. McReynolds LJ, Calvo KR, Holland SM. Germline GATA2 muta­tion and bone
mar­row fail­ure. Hematol Oncol Clin North Am. 2018;32(4):713-728.
8. Pastor V, Hirabayashi S, Karow A, et  al. Mutational land­scape in chil­dren
with myelodysplastic syn­dromes is dis­tinct from adults: spe­cific somatic
driv­ers and novel germline var­i­ants. Leukemia. 2017;31(3):759-762.
9. Wlodarski MW, Hirabayashi S, Pastor V, et al; EWOG-MDS. Prevalence, clin­
i­cal char­ac­ter­is­tics, and prog­no­sis of GATA2-related myelodysplastic syn­
dromes in chil­dren and ado­les­cents. Blood. 2016;127(11):1387-1397, quiz 1518.
10. Schwartz JR, Ma J, Lamprecht T, et al. The geno­mic land­scape of pedi­at­ric
myelodysplastic syn­dromes. Nat Commun. 2017;8(1):1557.
11. Townsley DM, Scheinberg P, Winkler T, et  al. Eltrombopag added
to stan­dard immu­no­sup­pres­sion for aplastic ane­mia. N Engl J Med.
2017;376(16):1540-1550.
12. Killick SB, Bown N, Cavenagh J, et al; Brit­ish Society for Standards in Hae-
matology. Guidelines for the diag­no­sis and man­age­ment of adult aplastic
anae­mia. Br J Haematol. 2016;172(2):187-207.
13. Peslak SA, Olson T, Babushok DV. Diagnosis and treat­ment of aplastic ane­
mia. Curr Treat Options Oncol. 2017;18(12):70.
14. Shimano KA, et al. Am J Hematol. 2021. Online ahead of print.
15. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revi­sion to the World Health
Organization clas­si­fi­ca­tion of mye­loid neo­plasms and acute leu­ke­mia.
Blood. 2016;127(20):2391-2405.
16. Hosokawa K, Sugimori N, Katagiri T, et  al. Increased glycosylphosphati-
dylinositol-anchored pro­tein-defi­cient granulocytes define a benign sub­
set of bone mar­row fail­ures in patients with tri­somy 8. Eur J Haematol.
2015;95(3):230-238.
17. Ishiyama K, Karasawa M, Miyawaki S, et al. Aplastic anae­mia with 13q-: a
benign sub­set of bone mar­row fail­ure respon­sive to immu­no­sup­pres­sive
ther­apy. Br J Haematol. 2002;117(3):747-750.
18. Chandrasekharappa SC, Lach FP, Kimble DC, et al; NISC Comparative
Sequencing Program. Massively par­al­lel sequenc­ing, aCGH, and RNA-seq
tech­nol­o­gies pro­vide a com­pre­hen­sive molec­u­lar diag­no­sis of Fanconi
ane­mia. Blood. 2013;121(22):e138-e148.
19. Nicoletti E, Rao G, Bueren JA, et al. Mosaicism in Fanconi ane­mia: con­cise
review and eval­u­a­tion of published cases with focus on clin­i­cal course of
blood count nor­mal­i­za­tion. Ann Hematol. 2020;99(5):913-924.
20. Alter BP, Rosenberg PS, Giri N, Baerlocher GM, Lansdorp PM, Savage SA.
Telomere length is asso­ci­ated with dis­ease sever­ity and declines with age
in dyskeratosis congenita. Haematologica. 2012;97(3):353-359.
21. Alder JK, Hanumanthu VS, Strong MA, et  al. Diagnostic util­ ity of telo­
mere length test­ing in a hos­pi­tal-based set­ting. Proc Natl Acad Sci U S A.
2018;115(10):E2358-E2365.
22. Schratz KE, Haley L, Danoff SK, et al. Cancer spec­trum and out­comes in the
Men­de­lian short telo­mere syn­dromes. Blood. 2020;135(22):1946-1956.
23. Alter BP, Baerlocher GM, Savage SA, et al. Very short telo­mere length by
flow fluo­res­cence in situ hybrid­iza­tion identifies patients with dyskeratosis
congenita. Blood. 2007;110(5):1439-1447.
24. Yoshizato T, Dumitriu B, Hosokawa K, et al. Somatic muta­tions and clonal
hema­to­poi­e­sis in aplastic ane­mia. N Engl J Med. 2015;373(1):35-47.
142  |  Hematology 2021  |  ASH Education Program
25. Babushok DV, Perdigones N, Perin JC, et al. Emergence of clonal hema­to­
poi­es­ is in the major­ity of patients with acquired aplastic ane­mia. Cancer
Genet. 2015;208(4):115-128.
26. Shaw YB. et al. Blood Advances 2021: The predictive value of PNH clones,
6p CN-LOH, and clonal TCR gene rearrangement for aplastic anemia. 5(16).
27. Borowitz MJ, Craig FE, Digiuseppe JA, et al; Clinical Cytometry Society.
Guidelines for the diag­no­sis and mon­i­tor­ing of par­ox­ys­mal noc­tur­nal
hemo­glo­bin­uria and related dis­or­ders by flow cytom­e­try. Cytometry B
Clin Cytom. 2010;78(4):211-230.
28. Villalobos IB, Takahashi Y, Akatsuka Y, et al. Relapse of leu­ke­mia with loss
of mismatched HLA resulting from uni­pa­ren­tal disomy after haploiden-
tical hema­to­poi­etic stem cell trans­plan­ta­tion. Blood. 2010;115(15):3158-
3161.
29. Vago L, Perna SK, Zanussi M, et  al. Loss of mismatched HLA in leu­ke­mia
after stem-cell trans­plan­ta­tion. N Engl J Med. 2009;361(5):478-488.
30. Afable MG II, Wlodarski M, Makishima H, et al. SNP array-based karyotyp-
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/134/1852140/134geddis.pdf by guest on 13 December 2021
ing: dif­fer­ences and sim­i­lar­i­ties between aplastic ane­mia and hypocellular
myelodysplastic syn­dromes. Blood. 2011;117(25):6876-6884.
31. Babushok DV, Olson TS, Bessler M. Somatic muta­tions and clonal hema­to­
poi­e­sis in aplastic ane­mia. N Engl J Med. 2015;373(17):1673.
32. Babushok DV, Xie HM, Roth JJ, et al. Single nucle­o­tide poly­mor­phism array
anal­y­sis of bone mar­row fail­ure patients reveals char­ac­ter­is­tic pat­terns of
genetic changes. Br J Haematol. 2014;164(1):73-82.
33. Katagiri T, Sato-Otsubo A, Kashiwase K, et al; Japan Marrow Donor Pro-
gram. Frequent loss of HLA alleles asso­ci­ated with copy num­ber-neu­tral
6pLOH in acquired aplastic ane­mia. Blood. 2011;118(25):6601-6609.
34. Mohamedali AM, Gäken J, Ahmed M, et al. High con­cor­dance of geno­mic
and cyto­ge­netic aber­ra­tions between periph­eral blood and bone mar­row
in myelodysplastic syn­drome (MDS). Leukemia. 2015;29(9):1928-1938.
35. Score J, Chase A, Forsberg LA, et  al. Detection of leu­ke­mia-asso­ci­ated
muta­tions in periph­eral blood DNA of hematologically nor­mal elderly indi­
vid­u­als. Leukemia. 2015;29(7):1600-1602.
36. Betensky M, Babushok D, Roth JJ, et al. Clonal evo­lu­tion and clin­i­cal sig­nif­
i­cance of copy num­ber neu­tral loss of het­ero­zy­gos­ity of chro­mo­some arm
6p in acquired aplastic ane­mia. Cancer Genet. 2016;209(1-2):1-10.
37. Kanagal-Shamanna R, Hodge JC, Tucker T, et  al. Assessing copy num­ber
aber­ra­tions and copy neu­tral loss of het­ero­zy­gos­ity across the genome
as best prac­tice: an evi­dence based review of clin­i­cal util­ity from the
Cancer Genomics Consortium (CGC) Working Group for Myelodysplas-
tic Syndrome, Myelodysplastic/Myeloproliferative and Myeloproliferative
Neoplasms. Cancer Genet. 2018;228-229:197-217.
38. Kuramitsu M, Sato-Otsubo A, Morio T, et al. Extensive gene dele­tions in Jap­
a­nese patients with Diamond-Blackfan ane­mia. Blood. 2012;119(10):2376-
2384.
39. Waespe N, Dhanraj S, Wahala M, et al. The clin­i­cal impact of copy num­
ber var­i­ants in inherited bone mar­row fail­ure syn­dromes. NPJ Genom Med.
2017;2:18.
40. Sperling AS, Gib­son CJ, Ebert BL. The genet­ics of myelodysplastic syn­
drome: from clonal haematopoiesis to sec­ond­ary leu­kae­mia. Nat Rev Can-
cer. 2017;17(1):5-19.
41. Sun L, Babushok DV. Secondary myelodysplastic syn­drome and leu­ke­mia
in acquired aplastic ane­mia and par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria.
Blood. 2020;136(1):36-49.
42. Tsai FD, Lindsley RC. Clonal hema­to­poi­e­sis in the inherited bone mar­row
fail­ure syn­dromes. Blood. 2020;136(14):1615-1622.
43. Lane AA, Odejide O, Kopp N, et al. Low fre­quency clonal muta­tions recov­
er­able by deep sequenc­ing in patients with aplastic ane­mia. Leukemia.
2013;27(4):968-971.
44. Negoro E, Nagata Y, Clemente MJ, et al. Origins of myelodysplastic syn­
dromes after aplastic ane­mia. Blood. 2017;130(17):1953-1957.
45. Bono E, McLornan D, Travaglino E, et  al. Clinical, his­to­path­o­log­ic ­ al and
molec­u­lar char­ac­ter­iza­tion of hypo­plas­tic myelodysplastic syn­drome. Leu-
kemia. 2019;33(10):2495-2505.
46. Kulasekararaj AG, Jiang J, Smith AE, et al. Somatic muta­tions iden­tify a sub­
group of aplastic ane­mia patients who prog­ress to myelodysplastic syn­
drome. Blood. 2014;124(17):2698-2704.
47. Kennedy AL, Myers KC, Bowman J, et al. Distinct genetic path­ways define
pre-malig­nant ver­sus com­pen­sa­tory clonal hema­to­poi­e­sis in Shwach-
man-Diamond syn­drome. Nat Commun. 2021;12(1):1334.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000244
 WHAT DO APLASTIC ANEMIA, PNH, AND “ HYPOPLASTIC ” LEUKEMIA HAVE IN COMMON ?

When does a PNH clone have clinical signifcance?

Daria V. Babushok

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/143/1852074/143babushok.pdf by guest on 13 December 2021

Division of Hematology-Oncology, Department of Medicine, Hospital of the University of Pennsylvania, Philadelphia, PA; and Comprehensive Bone
Marrow Failure Center, Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, PA

Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired blood disease caused by somatic mutations in the phosphati-
dylinositol glycan class A (PIGA) gene required to produce glycophosphatidyl inositol (GPI) anchors. Although PNH cells
are readily identified by flow cytometry due to their deficiency of GPI-anchored proteins, the assessment of the clinical
significance of a PNH clone is more nuanced. The interpretation of results requires an understanding of PNH pathogene-
sis and its relationship to immune-mediated bone marrow failure. Only about one-third of patients with PNH clones have
classical PNH disease with overt hemolysis, its associated symptoms, and the highly prothrombotic state characteristic
of PNH. Patients with classical PNH benefit the most from complement inhibitors. In contrast, two-thirds of PNH clones
occur in patients whose clinical presentation is that of bone marrow failure with few, if any, PNH-related symptoms.
The clinical presentations are closely associated with PNH clone size. Although exceptions occur, bone marrow failure
patients usually have smaller, subclinical PNH clones. This review addresses the common scenarios that arise in evalu-
ating the clinical significance of PNH clones and provides practical guidelines for approaching a patient with a positive
PNH result.

LEARNING OBJECTIVES
• Interpret the validity and significance of PNH flow cytometry results
• Recognize the significance of PNH clones in the diagnostic assessment of patients with bone marrow failure
• Recognize indications for starting complement inhibitor therapy in PNH

Introduction prevalent in patients with immune-mediated bone mar-

Paroxysmal nocturnal hemoglobinuria (PNH) is an acquired
blood disease caused by somatic mutations in the phos-
phatidylinositol glycan class A (PIGA) gene.1 PIGA encodes
an enzyme required for the first step of glycophosphatidyl
inositol (GPI) anchor biosynthesis. PIGA-mutant (PNH) cells
lack all GPI-anchored proteins, including 2 essential com-
plement regulatory proteins, CD55 (complement decay-
accelerating factor) and CD59 (inhibitor of the complement
membrane attack complex). The deficiency of CD55 and
CD59 makes PNH red blood cells (PNH RBCs) susceptible to
complement-mediated hemolysis.
Although very small polyclonal populations of PNH cells
of approximately 0.001% to 0.005% (approximately 10 to
50 per million) can be detected in most healthy individ-
uals,2,3 PNH hematopoietic cells have no intrinsic growth
advantage and do not clonally expand under normal con-
ditions.4-6 Slightly larger PNH clones of 0.02% to 0.03% can
occur in approximately 1% of healthy individuals; however,
more significant clonal expansions of PNH cells do not occur
in the absence of pathology.2,3,7 In contrast, PNH clones are
Prakash Singh Shekhawat
row failure.8-10 Nearly half of patients with acquired aplastic
anemia (AA), a T-cell-mediated autoimmune bone marrow
failure disorder, develop PNH clones.8-10 In AA, PNH cells
are thought to evade the hematopoietic stem and progen-
itor cell (HSPC)-directed autoimmune attack, and conse-
quently, PNH cells outgrow wild-type cells forming PNH
clones.1 PNH clones can also emerge in patients with myel-
odysplastic syndrome (MDS) and, more rarely, in patients
with myeloproliferative neoplasms.11,12 Although immune-
mediated bone marrow failure is the primary risk factor for
developing PNH clones, patients can be diagnosed with
classical PNH without a known diagnosis of bone marrow
failure or a history of cytopenias. In such cases, prior sub-
clinical marrow failure is frequently presumed; alternatively,
secondary proliferative mutations within a PNH HSPC can
promote PNH clone expansion.1,13
Studies of patients with positive PNH flow cytometry
results have found that PNH clones have a bimodal size
distribution closely related to the patients’ clinical pre-
sentations12,14 (Figure 1, Table 1). Approximately one-third
Evaluation and significance of PNH clones | 143

Figure 1. The bimodal distribution of PNH clone sizes closely
correlates with the patients’ clinical presentations. The styl-
ized schematic depicts the bimodal distribution of granulocyte
PNH clone sizes (black line) based on published PNH clone size
distributions.12,14 Red and blue area plots illustrate the different
clinical presentations that correlate with the extremes of PNH
clone sizes. Patients with bone marrow failure who have an asso-
ciated PNH clone are shown in red. Classical PNH presentations
are shown in blue.
of patients have “clas­si­cal PNH,” with overt hemo­ly­sis, mul­ti­ple
PNH symp­toms, and an increased risk of throm­bo­ses. Patients
with clas­si­cal PNH have a mean PNH clone size of more than
70%.11,12 In con­trast, two-thirds of PNH clones occur in patients
who have cytopenias and bone mar­row fail­ure with­out clin­i­cal
hemo­ly­sis. Most bone mar­row fail­ure patients are largely asymp­
tom­atic from PNH and have small PNH clones (mean clone size of
approx­i­ma­tely 11%) (Figure 1, Table 1).11,12
Interpreting the sig­nif­i­cance of a PNH flow
cytom­e­try result
CLINICAL CASE 1: FALSE-POSITIVE PNH TEST
A 55-year-old man with decades-long throm­bo­cy­to­pe­nia was
referred for an eval­u­a­tion of bone mar­row fail­ure. The patient had
mac­ro­cytic ane­mia with a hemo­glo­bin of 10.6  g/dL and a mean
Table 1.  Clinical categories of patients with PNH clones
PNH granulocyte
PNH cat­e­gory clone Clinical hemo­ly­sis
Bone mar­row fail­ure with an asso­ci­ated PNH clone
  Subclinical Small (usu­ally Absent
(major­ity of <30%)
patients)
 Symptomatic Moderate May be pres­ent
Classical PNH Large (usu­ally Present
>50%)
144  |  Hematology 2021  |  ASH Education Program
cor­pus­cu­lar vol­ume of 121.4 fL, an abso­lute retic­u­lo­cyte count
of 45.8 × 103 cells/µL, and plate­lets of 29 × 103 cells/µL. The white
blood cells (WBCs) were low at 2.9 × 103 cells/µL with 58.9%
granulocytes, 6.2% mono­cytes, and 34.9% lym­pho­cytes. Bone
mar­row was hypocellular with­out dys­plas­tic changes. PNH flow
cytom­et­ ry revealed an iso­lated PNH clone of 2.9% in RBCs only,
with no PNH granulocytes. Lactate dehy­dro­ge­nase (LDH) was
nor­mal at 191 U/L. Haptoglobin was bor­der­line low at 32 mg/dL.
Given the absence of granulocyte PNH clone and the lack of
hemo­ly­sis, a false-pos­i­tive PNH clone was suspected. PNH flow
cytom­et­ ry test­ing was repeated, with no evi­dence of PNH cells
on sev­eral sub­se­quent tests. Additional workup showed short
lym­pho­cyte telo­mere lengths for age and a path­o­genic var­i­ant
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/143/1852074/143babushok.pdf by guest on 13 December 2021
in TERC, confirming the diag­no­sis of short telo­mere syn­drome.
The diverse clin­i­cal pre­sen­ta­tions and var­ied PNH cell type
involve­ment require sys­tem­atic con­sid­er­ation of sev­eral clin­i­cal
and lab­or­a­tory fac­tors to deter­mine the sig­nif­i­cance of a “pos­
i­tive” (or a “neg­at­ive”) PNH flow cytom­et­ry result for a given
patient. At a min­ i­
mum, an eval­ u­at­ion of a PNH result should
include (1) PNH flow cytom­e­try, which should be performed on
periph­eral blood with quan­ti­fi­ca­tion of PNH neu­tro­phils and PNH
RBCs; (2) a bone mar­row aspi­rate and biopsy, includ­ing cyto­ge­
net­ics (test­ing for somatic muta­tions asso­ci­ated with hema­to­
logic malig­nan­cies can also be help­ful for prog­nos­tic assess­ment
of patients with hypocellular mar­row fail­ure13 and in patients in
whom PNH arose in asso­ci­a­tion with MDS or mye­lo­pro­lif­er­a­tive
neo­plasms); (3) lab­o­ra­tory stud­ies to eval­u­ate for hemo­ly­sis; and
(4) his­tory assessing for prior throm­botic events and cytopenias
(Figure 2). Here, I dis­cuss sev­eral com­mon clin­i­cal sce­nar­ios that
arise when eval­ua ­ t­ing patients with a pos­i­tive PNH test to illus­
trate the sig­nif­i­cance of PNH clones in clas­si­cal PNH and bone
mar­row fail­ure.
PIGA muta­tions occur early in hema­to­poi­e­sis in hema­to­poi­
etic stem cells or multipotent pro­gen­i­tors.7,12,14,15 In most patients,
PNH clones can be detected in all­or most mature hema­to­poi­
etic lin­e­ages, includ­ing granulocytes, mono­cytes, plate­lets,
RBCs, and the B, T, and nat­u­ral killer cells.7,12,15 More lim­ited lin­e­
age involve­ment can occur but is more char­ac­ter­is­tic of MDS, in
which PNH clones are also more likely to be tran­sient.7
Next-gen­er­a­tion sequenc­ing of the PIGA gene in patients with
PNH iden­ti­fied 2 or more inde­pen­dent PIGA muta­tions in 85% of
Laboratory mark­ers Risk of Utility of com­ple­ment
of hemo­ly­sis PNH symp­toms throm­bo­sis inhib­i­tor ther­apy
May be pres­ent, Absent Low Do not ben­e­fit from
depending on com­ple­ment inhib­i­tor
clone size
May be pres­ent May be pres­ent Intermediate Some patients may
ben­e­fit from
com­ple­ment inhib­i­tor
Present Present High Complement inhib­i­tor
improves out­comes

Figure 2. The recommended evaluation algorithm for patients with a newly identified PNH clone. CBC, complete blood count;
NGS, next-generation sequencing.
patients with PNH and at least 3 inde­pen­dent PIGA muta­tions
in 57% of patients.16 Because mul­ti­ple inde­pen­dent muta­tions
together com­prise PNH clones detected by flow cytom­e­try, flow
cytom­e­try is much more sen­si­tive than next-gen­er­a­tion sequenc­
ing for PNH detec­tion. In con­trast to genetic ana­ly­ses, which are
more sen­si­tive when performed on bone mar­row than periph­
eral blood by cap­tur­ing acquired genetic alter­ations in HSPCs,
the standard clin­ i­
cal PNH flow cytom­ e­
try eval­ u­
ates only the
mature cell pop­u­la­tions (granulocytes, mono­cytes, and RBCs) that
are well represented in periph­eral blood. PNH clones can thus be
sen­si­tively detected in periph­eral blood with­out a require­ment for
a bone mar­row biopsy. Sometimes, sub­mit­ting a hypocellular bone
mar­row spec­i­men for PNH flow cytom­e­try may par­a­dox­i­cally result
in lower sen­si­tiv­ity for PNH detec­tion if the sub­mit­ted aspi­rate is of
an inad­e­quate vol­ume containing insuf­fi­cient granulocytes to detect
very small PNH clones. While PNH flow cytometry of other cell pop-
ulations in the bone marrow can be performed, these have not
been standardized for high sensitivity PNH cell analysis and can be
affected by variation in cell surface marker expression across various
hematopoietic subsets.
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/143/1852074/143babushok.pdf by guest on 13 December 2021
Clinically, most flow cytom­e­try lab­o­ra­to­ries now use high-
sen­si­tiv­ity PNH assay, allowing the detec­tion of PNH clones as
small as 0.01% in granulocyte, mono­cyte, and RBC lin­e­ages. PNH
granulocyte and mono­cyte pro­por­tions are closely cor­re­lated
in most patients.11,12 Unlike PNH RBCs, which are sus­cep­ti­ble to
com­ple­ment-medi­ated hemo­ly­sis, PNH granulocytes have a nor­
mal life span in vivo.17 For this rea­son, the PNH clone size is com­
monly esti­mated using the granulocyte lin­e­age. At a min­i­mum,
PNH flow cytom­e­try should pro­vide a quan­ti­fi­ca­tion of PNH pro­
por­tions in granulocytes and RBCs.
To ensure that the iden­ti­fied GPI-neg­a­tive cell pop­u­la­tions rep­
re­sent PNH clones and are not cells with aber­rant expres­sion or an
inherited defi­ciency of a sin­gle GPI-anchored pro­tein,18 at least 2
GPI-anchored mark­ers should be used to iden­tify PNH cells. The
incor­po­ra­tion of the fluo­ro­chrome-con­ju­gated nonlysing form
of the proaerolysin toxin, which binds to the gly­can moi­ety of
GPI-anchored pro­teins, has fur­ther improved the sen­si­tiv­ity and
accu­racy of PNH leu­ko­cyte detec­tion.19 Granulocyte and mono­
cyte PNH clones are quan­ti­fied based on com­plete GPI-anchored
pro­tein defi­ciency. In con­trast, both the com­plete (type III) and
Evaluation and sig­nif­i­cance of PNH clones | 145
the par­tial (type II) GPI-anchored pro­tein defi­ciency are mea­sured
for RBCs. Type II cells have a small frac­tion of nor­mal GPI-anchored
pro­tein lev­els and arise from either hypo­mor­phic (mis­sense) PIGA
muta­tions or the pas­sive trans­fer of GPI-anchored pro­teins to PNH
cells.16,20 Rarely, type II cells may sig­nif­i­cantly out­num­ber type III
cells, in which case the under­ly­ing PIGA muta­tion is com­monly a
mis­sense var­i­ant in PIGA, lead­ing to a hypofunctioning allele and a
par­tial GPI-anchored pro­tein defi­ciency.16
In patients not receiv­ing com­ple­ment inhib­i­tors, PNH RBCs are
sen­si­tive to com­ple­ment-medi­ated lysis and have a half-life of only
4 to 8 days.21 The short­ened life span of PNH RBCs causes a sig­nif­
i­cantly smaller appar­ent RBC PNH clone size com­pared with PNH
granulocytes. A cor­­ol­lary to this is that the lack of the expected
clone size dif­fer­ence between PNH granulocytes and RBCs in an
untreated patient with PNH implies a nor­mal PNH RBC life span
and a lack of clin­i­cally sig­nif­i­cant hemo­ly­sis; in prac­tice, this can
arise due to a pre­dom­i­nant type II clone that is less sus­cep­ti­ble
to hemo­ly­sis. Because RBC trans­fu­sions will dilute the PNH RBC
frac­tion, a trans­fu­sion his­tory should be taken into account when
interpreting PNH RBC clone size. When clin­i­cally fea­si­ble, PNH
flow cytom­e­try should be performed before RBC trans­fu­sions.
Most PNH clones per­sist long term, of which some (approx­
i­ma­tely 6% to 18%) will expand, while 17% to 24% of clones will
become smaller over time.9,10,12,14,22 Thus, serial eval­u­a­tion (eg,
annu­ally and with changes in hemo­lytic param­e­ters) is recom-
mended for all­patients with PNH clones to mon­i­tor for PNH
clone expan­ sion. Periodic mon­ i­
tor­
ing can also be help­ ful in
patients with clas­si­cal PNH, up to 15% of whom can undergo
spon­ta­ne­ous remis­sion over time and may no lon­ger require
ther­apy.13 Because most PNH clones arise in patients with AA,
annual PNH screen­ing is advised for patients with AA to iden­tify
newly emer­gent PNH clones.23
Reflecting on case 1, the pat­tern of an iso­lated PNH RBC clone
with­out the involve­ment of granulocytes is unex­pected. This pat­
tern raises a pos­si­bil­ity of a false-pos­i­tive PNH clone or, per­haps,
a tran­sient sin­gle-lin­e­age PNH clone isolated to the ery­throid lin­
e­age. Clinically, the patient has no lab­o­ra­tory or clin­i­cal hemo­
ly­sis, which is not nec­es­sar­ily unex­pected given the rel­a­tively
small RBC clone size. Because PNH cells are iden­ti­fied by their
lack of staining for sur­face mark­ers, spe­cial­ized flow cytom­e­try
pro­to­cols are required to pre­vent falsely misidentifying debris
and other nonstaining cells as PNH clones.24 Historically, fail­ure
to use opti­mal meth­od­ol­ogy has been asso­ci­ated with false-
­pos­i­tive results in 16% to 25% of clin­i­cal PNH tests in mul­ti­cen­ter
com­par­i­sons.11,25 The implementation of the International Clinical
Cytometry Society/Euro­pean Society for Clinical Cell Analyses
con­sen­sus guide­lines for detec­tion of GPI-defi­cient cells in PNH
has dra­mat­i­cally improved the accu­racy and sen­si­tiv­ity of PNH
test­ing. Compared with rou­tine PNH anal­y­sis suit­able for iden­
ti­fy­ing PNH clones more than 1%, which misses nearly 40% of
smaller PNH clones,19 the high sen­ si­tiv­
ity PNH assay enables
accu­ rate iden­ti­
fi­
ca­
tion of clones down to 0.01%.19,24 Because
rare healthy indi­vid­u­als can have tran­sient expan­sions of PNH
cells reaching the level of detec­tion of the high-sen­si­tiv­ity PNH
assay,2,3,7 repeat­ing PNH flow cytom­e­try can help to ascer­tain
per­sis­tence of very rare pop­u­la­tions of PNH cells.
In sum, the tech­ ni­
cal char­ ac­ter­
is­tics of a given PNH flow
cytom­e­try assay can sig­nif­i­cantly affect the accu­racy of results.
A care­ful review of the PNH test­ing meth­od­ol­ogy, with close
atten­tion to the assay sen­si­tiv­ity and spec­i­fic­ity param­e­ters, the
pres­ence of PNH cells in mul­ti­ple lin­e­ages, and the expected dif­
146  |  Hematology 2021  |  ASH Education Program
fer­ence between the granulocyte and RBC PNH clone sizes, can
help inter­pret the accu­racy of both the pos­i­tive and neg­a­tive
PNH results in a given patient (Figure 2). Follow-up test­ing with
high-sen­si­tiv­ity PNH assay, performed according to the Inter-
national Clinical Cytometry Society/Euro­pean Society for Clin-
ical Cell Analyses guide­lines, can clar­ify unex­pected results and
resolve inconsistencies.
PNH clones in immune-medi­ated bone mar­row fail­ure
CLINICAL CASE 2: DIAGNOSTIC SIGNIFICANCE OF PNH
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/143/1852074/143babushok.pdf by guest on 13 December 2021
CLONES IN AA
A 26-year-old pre­vi­ously healthy woman sought treat­ment from
the emer­gency room after expe­ri­enc­ing 1 month of fatigue and
dyspnea on exer­tion. Complete blood count revealed pan­cy­
to­pe­nia, with WBCs of 1.07  ×  103 cells/µL, with 24.3% neu­tro­
phils and 75.7% lym­pho­cytes, a hemo­glo­bin of 4.3 g/dL with
low abso­lute retic­u­lo­cytes of 23  ×  103 cells/µL, and plate­lets of
12 × 103 cells/µL. Physical exam­i­na­tion was nota­ble for pal­lor and
scattered pete­chiae. There was no fam­ily his­tory of blood count
abnor­mal­i­ties or con­di­tions sus­pi­cious for inherited mar­row fail­
ure syn­dromes. Additional lab­o­ra­tory stud­ies showed a mildly
ele­vated LDH of 239  U/L (1.2 × the upper limit of nor­mal [ULN] in
the test­ing lab­o­ra­tory), a low hap­to­glo­bin less than 30  mg/dL,
and nor­mal indi­rect bil­i­ru­bin of 0.9  mg/dL. The mar­row was
mark­edly hypocellular for age with­out dys­plas­tic changes. Cyto-
genetics were nor­mal. Flow cytometry revealed a PNH clone of
19.11% in neu­tro­phils, 17.83% in mono­cytes, and 0.32% in RBCs.
A diag­ no­sis of severe immune-medi­ ated AA with an asso­ ci­
ated PNH clone was made. The patient had no matched sib­ling
donors and was treated with immu­no­sup­pres­sive ther­apy with
horse antithymocyte glob­u­lin, cyclo­spor­ine, and eltrombopag,
achiev­ing com­plete remis­sion. One year after ini­tial diag­no­sis,
a fol­low-up eval­u­a­tion revealed a hemo­glo­bin of 11.9  g/dL and
48 × 103 retic­u­lo­cytes/µL with no clin­i­cal or lab­o­ra­tory evi­dence
of hemo­ly­sis that included a nor­mal LDH of 186  U/L. Follow-up
PNH flow cytom­e­try showed a smaller PNH clone of 8.26% in
neu­tro­phils, 7.6% in mono­cytes, and 1.59% in RBCs.
Because the clonal expan­sion of PNH cells is closely linked to
the HSPC-directed auto­im­mune attack in AA,1 a PNH clone can
pro­vide an impor­tant diag­nos­tic clue about the immune-medi­
ated path­o­gen­e­sis of mar­row aplasia.8,26 The presenting fea­tures
of AA—pan­cy­to­pe­nia with hypocellular bone mar­row—can be
brought on by dif­fer­ent eti­­ol­o­gies and are not spe­cific for the
diag­no­sis of AA. For this rea­son, the diag­no­sis of AA requires
sys­tem­atic exclu­sion of all­ alter­na­tive causes of mar­row fail­ure,
includ­ing nutri­tional deficiencies, infec­tions, or tox­ins. Of par­tic­
u­lar impor­tance for chil­dren and young and mid­dle-aged adults
is the exclu­sion of inherited bone mar­row fail­ure syn­dromes,
a costly and time-con­sum­ing pro­cess that can leave lin­ger­ing
diag­nos­tic uncertainties. Consequently, clin­i­cal test­ing with a
high pos­i­tive pre­dic­tive value for immune-medi­ated AA can be
valu­able by streamlining the diag­nos­tic eval­u­at­ ion and allowing
for prompt ini­ti­a­tion of AA-directed ther­a­pies.
The pre­ dic­
tive value of PNH clones for the diag­ no­sis of
immune-medi­ated AA and the exclu­sion of inherited bone mar­row
fail­ure was eval­ua ­ ted in 2 insti­tu­tional cohorts of bone mar­row
fail­ure patients8,26 (Table 2). PNH clones were iden­ti­fied in 46% of
patients with AA but in none of the eval­u­ated inherited bone mar­
row fail­ure patients. A com­bined anal­y­sis of both cohorts showed
that PNH granulocyte clones, regard­less of clone size, have an
approx­i­ma­tely 100% pos­i­tive pre­dic­tive value for AA and have
25-fold higher odds of being found in AA com­pared with inher-
ited bone mar­row fail­ure.8 Two addi­tional stud­ies found no PNH
clones or PIGA muta­tions in more than 130 unse­lected patients
with an inherited mar­row fail­ure dis­or­der, Shwachman-Diamond
syn­drome (SDS),27,28 pro­vid­ing addi­tional sup­port to the spec­i­fic­
ity of PNH clones for immune-medi­ated AA.
In clin­i­
cal prac­tice, a Bayes the­ ory nomo­ gram tool, which
incor­po­rates the pre­test prob­a­bil­ity of AA, can be used to esti­
anom­a­lies and mod­er­ate cytopenias from birth who devel­oped
wors­en­ing cytopenias and new trans­fu­sion require­ments fol­low­
ing inter­feron ther­apy for hep­a­ti­tis C. Finding a PNH clone, in this
or sim­i­lar sit­u­a­tions of poorly explained cytopenias, can bring AA
higher on the dif­fer­en­tial diag­no­sis. In a hypo­thet­i­cal case with a
low pre­test prob­a­bil­ity of 2%, a pos­i­tive PNH clone would raise
the post­test prob­a­bil­ity of AA into a mod­er­ate range of 65%.8
Because PNH clones can emerge later in the dis­ease course,
repeat­ing PNH flow cytom­e­try can help to iden­tify newly emer­
gent clones in patients who may have tested neg­a­tive at diag­
no­sis.23 PNH clones found later in the dis­ease course may be
par­tic­u­larly help­ful by affirming the immune-medi­ated eti­­ol­ogy
of AA in patients with refrac­tory AA, in whom the diag­no­sis of

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/143/1852074/143babushok.pdf by guest on 13 December 2021

mate the post­test prob­a­bil­ity of AA after a pos­i­tive PNH result
(Figure 3).8 In Case 2, a pre­vi­ously healthy patient devel­oped
pan­cy­to­pe­nia as a young adult, with­out a fam­ily his­tory or phys­
i­cal exam­i­na­tion find­ings sug­ges­tive of a con­gen­i­tal mar­row fail­
ure dis­or­der. Based on this pre­sen­ta­tion, the clin­i­cal sus­pi­cion of
immune-medi­ated acquired AA is mod­er­ate to high. Using a con­
ser­va­tive esti­mate of a mod­er­ate pre­test prob­a­bil­ity of AA (eg,
60%), the pres­ence of a PNH clone would raise the post­test prob­
a­bil­ity of AA to more than 99%, allowing to con­fi­dently estab­lish
the diag­no­sis of immune-medi­ated AA and ini­ti­ate appro­pri­ate
ther­ apy. Notably, a neg­ a­
tive PNH test does not sig­ nif­i­
cantly
change the pre­test prob­a­bil­ity of AA, and the absence of a PNH
clone should not be used to exclude AA. PNH test­ing can sim­i­larly
help in sit­u­a­tions where a pri­ori clin­i­cal sus­pi­cion of AA is low.
For exam­ple, a diag­no­sis of AA may be discounted in a young
adult with a known syndromic dis­or­der with mul­ti­ple con­gen­i­tal
immune-medi­ated AA may be in ques­tion.
In most stud­ies of AA out­comes, PNH clones, regard­less of clone
size, were asso­ci­ated with improved response to immu­no­sup­pres­
sive ther­apy and bet­ter over­all prog­no­sis.9,13,29-34 The rea­sons for
improved responses to immu­no­sup­pres­sion in patients with PNH
clones are incom­pletely under­stood but may be par­tially due to a
more accu­rate diag­no­sis of immune-medi­ated AA. PNH clones are
also asso­ci­ated with a lower rate of MDS pro­gres­sion.9,31,34
When does a PNH clone require therapy?
CLINICAL CASE 3 (CLASSICAL PNH)
A 20-year-old male col­lege ath­lete sought treat­ment from the
emer­gency room 3 times over the past 6 months with acute

Table 2.  Frequency of PNH in AA com­pared with inherited bone mar­row fail­ure dis­or­ders

Disease: num­ PNH test­ing Frequency of PNH

Study Year Population Study design ber of patients method clones Comparison
DeZern et al 26
2014 JHU cohort Retrospective AA: n = 132 Flow cytometry on AA: 61 of 132 (46%) • PPV for AA: 100%
IBMF: n = 20 periph­eral blood IBMF: 0 of 20 (0%) • NPV for AA: 54%
(sen­si­tiv­ity range
>0.01%-0.1%)
Shah et al8 2021 Penn/CHOP Retrospective AA: n = 126 Flow cytometry on AA: 58 of 126 (46%) • AA vs IBMF, OR
cohort Inherited: n = 9 periph­eral blood Inherited and other: 11.10, P < .05
Other: n = 13 (sen­si­tiv­ity range 0 of 22 (0%) • AA vs all­
>0.01%-1%) inherited dis­or­
ders includ­ing
IBMF, OR 16.23,
P < .05
• AA vs all­non-AA,
non-PNH, OR
38.43, P < .05
• PPV for AA vs
IBMF: 100%
• NPV for AA vs
IBMF: 48.5%
Keller et al27 2002 Camp Sunshine Retrospective n = 28 Flow cytometry on 0 of 28 patients (0%) NA
SDS cohort periph­eral blood
(PNH clones >1%)
Kennedy et al28 2014 SDS cohort Retrospective SDS: n = 99 Targeted, error- 0 of 110 patients NA
SDS-like: n = 11 corrected with PIGA muta­
sequenc­ing of tions (0%)
PIGA in bone
mar­row (>0.1%
VAF)
IBMF, inherited bone mar­row fail­ure; Inherited, inherited hema­to­logic dis­or­ders; JHU, Johns Hopkins University; NA, not applicable; NPV, neg­a­tive pre­dic­tive
Prakash Singh Shekhawat
value; OR, odds ratio; Penn/CHOP, University of Pennsylvania/Children’s Hospital of Philadelphia; PPV, pos­i­tive pre­dic­tive value; VAF, var­i­ant allele frac­tion.

Evaluation and sig­nif­i­cance of PNH clones | 147


Figure 3. The Bayesian (Fagan’s) nomogram demonstrating
the pretest and posttest probability of having a diagnosis of
­immune-mediated AA based on PNH flow cytometry results.
PNH clones have a near 100% positive predictive value for the
diagnosis of immune-mediated AA, with a positive LR of approx-
imately 92 for immune-mediated AA.8 Shown is an example cor-
responding to Case 2 in the text. To apply the nomogram tool,
one first has to estimate the pretest probability of AA based on
the clinical presentation. This previously healthy patient devel-
oped pancytopenia in young adulthood without any physical
examination findings or family history suggestive of a congenital
marrow failure disorder. Based on this presentation, her pretest
probability of immune-mediated acquired AA is estimated as
moderate to high. The green line shows the results for a con-
servative estimate of a moderate (approximately 60%) pretest
probability of AA, crossing at the positive LR of 92. Detection of
a PNH clone increases the posttest probability of AA to 99.3%,
allowing to confidently establish the diagnosis of immune-medi-
ated AA and initiate appropriate therapy. Importantly, a negative
PNH test would not significantly change the pretest probability.
Starting with a pretest probability of 60%, a negative PNH test
(shown by the red line crossing at the negative LR of 0.54) will
result in a posttest probability of 44.9%. The calculations for the
Bayesian nomogram plot are based on Shah et al.8
onset of nau­ sea, vomiting, and abdom­ i­
nal pain, each time
resolv­
ing with hydra­ tion and sup­ port­ive care. Contrast-en-
hanced abdom­i­nal com­puted tomog­ra­phy revealed an edem­
a­tous mes­en­tery with prominent mes­en­teric lymph nodes
with­out other pathol­ogy. Liver func­tion stud­ies showed indi­
rect hyperbilirubinemia of 2.7  mg/dL and an ele­vated aspar­tate
ami­no­trans­fer­ase of 81  U/mL. A com­plete blood count showed
WBCs of 12 × 103 cells/µL, hemo­glo­bin of 12.0  g/dL, and plate­
148  |  Hematology 2021  |  ASH Education Program
lets of 109 × 103 cells/µL, which downtrended to WBCs of 4 × 103
cells/µL, hemo­ glo­bin of 10.3  g/dL, and plate­ lets of 95 × 103
cells/µL after intra­ ve­
nous hydra­ tion. LDH was ele­ vated at
596 IU/L (3.10 × ULN), hap­to­glo­bin was unde­tect­able, and abso­
lute retic­u­lo­cytes were 112  ×  103 cells/µL. The Coombs test
was neg­a­tive, and the D-dimer was ele­vated to 3.87  µg/mL.
Upper extrem­ity super­fi­cial throm­bo­phle­bi­tis was noted at an
old intra­ve­nous line place­ment site. Flow cytometry revealed
a PNH clone with 78.6% PNH granulocytes, 87.3% PNH mono­
cytes, and 0.8% type II and 11.2% type III PNH RBCs. The patient
recalled mul­ti­ple epi­sodes of darker-col­ored urine. A bone mar­
row biopsy, performed to assess for under­ly­ing bone mar­row
fail­ure, showed a 40% cel­lu­lar mar­row with­out dys­plas­tic fea­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/143/1852074/143babushok.pdf by guest on 13 December 2021
tures, a nor­mal kar­yo­type, and a ­dis­ease-­asso­ci­ated somatic
BCOR muta­tion. The patient was diag­nosed with clas­si­cal
PNH, which likely arose from sub­clin­i­cal nonsevere AA. He was
started on folic acid and ini­ti­ated low-dose aspi­rin for man­age­
ment of throm­bo­phle­bi­tis and gen­eral thromboprophylaxis.
Following vac­ci­na­tion for Neisseria men­in­git­i­des, he started
eculizumab ther­apy with full symp­tom res­o­lu­tion. Once rav-
ulizumab became avail­­able, he was switched to ravulizumab
for patient con­ve­nience. Three years after the ini­tial diag­no­sis,
PNH flow cytom­e­try showed an expan­sion of his PNH clone to
92.8% in granulocytes, 94.1% in mono­cytes, and 36.5% in RBCs
(3.4% type II and 33.1% type III). Laboratory stud­ies performed
before the main­te­nance ravulizumab infu­sion showed WBCs
of 3.9 × 103 cells/µL, hemo­glo­bin of 13.0  g/dL, and plate­lets of
162 × 103 cells/µL, with retic­u­lo­cytes of 217 × 103 cells/µL and LDH
of 282  U/L (1.47 × ULN).
The deci­sion to ini­ti­ate com­ple­ment inhib­i­tor ther­apy in a pa-
tient with a PNH clone depends on mul­ti­ple fac­tors, includ­ing
clone size, clin­i­cal hemo­ly­sis, and symp­tom bur­den (Table 1).
Patients whose pri­mary clin­i­cal pic­ture is that of bone mar­row
fail­ure on aver­age have smaller PNH clones, have lit­tle in the way
of PNH symp­toms, and rarely ben­efi ­ t from PNH-directed ther­
apy. The ane­mia, fre­quently seen in patients with AA, most com­
monly stems from the under­ly­ing bone mar­row fail­ure and is not
caused by hemo­ly­sis. Thus, a care­ful eval­u­a­tion of the eti­­ol­ogy
of ane­mia with par­tic­u­lar atten­tion to ade­quate retic­u­lo­cyte re-
sponse is essen­tial when decid­ing on the util­ity of adding com­
ple­ment inhi­bi­tion to stan­dard AA ther­apy.
In con­trast, com­ple­ment inhib­i­tors improve out­comes in
patients with clas­si­cal PNH,35 who have overt hemo­ly­sis, are fre­
quently highly symp­tom­atic, and invari­ably have large (gen­er­ally
>50%12,14) PNH clones. Nearly one-third of patients with clas­si­
cal PNH have abdom­i­nal pain.36 Other clas­si­cal PNH symp­toms
include fatigue, dyspnea, esoph­a­geal spasm, erec­tile dys­func­
tion, neu­ro­logic dys­func­tion, and gall­stone dis­ease. Before the
avail­abil­ity of com­ple­ment inhib­i­tors, 37% to 39% of patients
with clas­si­cal PNH had throm­botic events, fre­quently involv­ing
unusual vas­cu­lar beds (eg, mes­en­teric and cere­bro­vas­cu­lar),
which were the lead­ing cause of mor­tal­ity.36-38 Less rec­og­nized
seri­ous com­pli­ca­tions of PNH include sub­clin­i­cal pul­mo­nary
emboli and pul­mo­nary hyper­ten­sion,39 cere­bro­vas­cu­lar ische­
mia,40 and renal dys­func­tion.41
Risk fac­ tors for throm­ bo­sis in patients with PNH clones
include granulocyte clone size more than 50%, LDH 1.5 or more
times the ULN, high PNH symp­tom bur­den, and pre­vi­ous throm­
 Table 3.  Risk fac­tors for throm­botic com­pli­ca­tions in PNH

PNH test­ing Effect of PNH clone

Study Year Population Study design No. of patients method Risk fac­tor for throm­bo­sis size
de Latour 2008 Patients treated in 58 Retrospective 460 Ham test or flow • Age >55 years: HR 1.8 >50% clones asso­ci­
et al36 hema­to­logic cen­ cytom­e­try (PNH • Transfusions: HR 1.7 ated with higher
ters in France clones >5%) • Thrombosis at diag­no­sis: HR 3.7 risk of TE: HR 3.2
• Warfarin as pri­mary pro­phy­laxis: HR 5.2
Lee et al44 2013 South Korean Retrospective 301 Ham test, sucrose- • LDH ≥1.5 × ULN: OR 7.0 No sig­nif­i­cant effect
National PNH lysis test, and Higher pre­dic­tive value in com­bi­na­tion with of clone size
Registry flow cytom­e­ symp­toms: PNH clone <20%:
try (no min­i­mal • LDH ≥1.5 × ULN plus abdom­i­nal pain: OR 17.8 16% TE
cut­off) • LDH ≥1.5 × ULN plus chest pain: OR 19.0 PNH clone 20%-
• LDH ≥1.5 × ULN plus dyspnea: OR 10.3 50%: 19% TE
• LDH ≥1.5 × ULN plus hemo­glo­bin­uria: OR 10.3 PNH clone >50%:
20% TE
Schrezenmeier 2014 International PNH Retrospective 900 Clinical diag­no­ • LDH ≥1.5 × ULN: 15.6% (vs 8.4% <1.5 × ULN) PNH clone <10%:
et al45 Registry sis and/or flow • PNH clone ≥50% 5.3% TE
cytom­e­try (no PNH clone 10%-
min­i­mal cut­off) 49%: 7.7% TE
PNH clone ≥50%:
15.4% TE
Long et al46 2017 Peking Union Medical Retrospective 104 Flow cytometry • PNH clone ≥50%: OR 9.78 PNH clone ≥50%:
College Hospital (PNH clones >1%) • ABO gene rs495828 GT+TT: OR 5.63 OR 9.78
• ABO gene rs2519093 TC+TT: 5.95
Griffin et al. 2019 Leeds UK PNH Retrospective study 25 of 429 patients Flow cytometry • Group 1: PNH white cells >30%, PNH red cells In patients with
201947 Database of throm­bo­sis in had PNH clone (PNH clones <10%, LDH <2 × ULN: 6 of 11 (54%) had TE TEs, PNH clone

Prakash Singh Shekhawat

patients with PNH >10% and LDH >10%) • Group 2: PNH white cells >30%, PNH red ranged from 49%
with PNH clone <2 × ULN cells >10% with higher pro­por­tion of type to 100%; median
>10% and LDH II red cells than type III red cells, LDH clone sizes were
<2 × ULN) <2 × ULN: 2 of 11 (18%) had TE 93% (group 1) and
• Group 3: PNH white cells 10%-30%, PNH 89% (group 2)
red cells <10%, LDH <2 × ULN (0 of 3 had
throm­bo­sis)
Huang et al.48 2019 Peking Union Medical Retrospective 99 Flow cytometry • PNH clone ≥80%: OR 1.056 Only eval­u­ated cut­
College Hospital (PNH clones >1%) • Hemoglobin ≤7.5 g/dL: OR 4.202 off of PNH clone
• Platelets >100 × 109/L: OR 6.547 ≥80%: OR 1.056
• ABO gene rs495828 = G: OR 5.243
Hoechsmann 2020 International PNH Retrospective case 57 TE cases, 189 non- NA • Recent HDA: OR 2.65 PNH clone ≥50%
et al43 [ab­ Registry con­trol TE con­trols • LDH ≥1.5 × ULN plus 2-3 HDA symp­toms:
stract] OR 8.61
• LDH ≥1.5 × ULN plus ≥4 HDA symp­toms: OR
14.5
• History of TE: OR 3.6
• History of MAVE: OR 2.17
• Recent pro­phy­lac­tic anticoagulation: OR
4.35
Füreder et al49 2020 Austrian PNH net­ Retrospective 59 Flow cytometry 14 of 59 patients had TE; 5 of 14 patients had Larger clone size in
work clone size recorded, which ranged from patients with TE
80% to 100%

Evaluation and sig­nif­i­cance of PNH clones | 149

HDA, recent high dis­ease activ­ity, defined as occur­ring within 6 months of TE event, LDH ≥1.5 × ULN, and hemo­glo­bin <10 g/dL or at least 1 of the fol­low­ing symp­toms: abdom­i­nal pain, dyspnea,
dys­pha­gia, fatigue, hemo­glo­bin­uria, and male erec­tile dys­func­tion; HR, haz­ard ratio; MAVE, major adverse vas­cu­lar event; OR, odds ratio; TE, throm­bo­em­bolic event.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/143/1852074/143babushok.pdf by guest on 13 December 2021
botic events (Table 3).36-38,42-44 When eval­ua ­ t­ing LDH, it is impor­
tant to com­pare a patient’s LDH result with the ref­er­ence range
from the test­ing lab­o­ra­tory, as nor­mal LDH val­ues vary between
test­ing sites due to var­i­a­tions in test­ing meth­od­ol­o­gies and plat­
forms. RBC clones more than 3% to 5% and granulocyte clones
over approx­i­ma­tely 23% are pre­dic­tive of ele­vated LDH, with
each 10% rise in granulocyte clone size asso­ci­ated with approx­i­
ma­tely 1.6 higher odds of throm­bo­sis.37 Although PNH clone size
cor­re­lates closely with hemo­ly­sis and LDH,45 the rela­tion­ship
between clone size, hemo­ly­sis, and throm­botic risk is not always
lin­ear. Rare patients with more than 50% PNH granulocytes have
devel­oped throm­botic events despite a near-nor­mal or nor­mal
LDH.47 Conversely, patients with PNH clones smaller than 50%
can be at increased risk of throm­bo­sis.44,45 Although hypo­mor­
phic PIGA muta­tions cause only par­tial GPI-anchored pro­tein
defi­ ciency (type II cells), which could be less sus­ cep­ ti­
ble to
hemo­ly­sis, throm­bo­ses in patients with type II ­PNH-pre­dom­i­nant
clones have been reported.47
In Case 3, the patient has near-nor­mal hemo­glo­bin and nota­
bly does not require ther­apy due to ane­mia. However, he is at
high risk of throm­botic com­pli­ca­tions. His risk fac­tors for throm­
bo­sis include a large PNH clone more than 50%, overt hemo­ly­sis
with LDH more than 3 times the ULN, high symp­tom bur­den, and
pre­vi­ous super­fi­cial throm­bo­phle­bi­tis. As evidenced by mes­en­
teric edema and ele­vated D-dimer, his abdom­i­nal pain is likely
caused by micro­vas­cu­lar ische­mia in the mes­en­teric vas­cu­la­ture.
He requires prompt ini­ti­a­tion of a com­ple­ment inhib­i­tor to pre­
vent life-threat­en­ing throm­botic com­pli­ca­tions.
The prothrombotic state in PNH is man­aged most effec­tively
by com­ple­ment inhi­bi­tion, which reduces throm­botic events from
approx­i­ma­tely 7.4% to 1% per patient-year.50 The mech­a­nism of
throm­bo­sis in PNH involves abnor­mal plate­let acti­va­tion by both
com­ple­ment and hemo­lytic pro­cesses as well as abnor­mal clot
com­po­si­tion.37 Untreated patients with PNH have higher fibrin­
o­gen and throm­bin gen­er­a­tion lev­els, lead­ing to faster-forming
fibrin clots that are harder to break down; these abnor­mal­i­ties
are improved by eculizumab.51 Primary anticoagulation pro­phy­
laxis has been pro­posed for patients with more than 50% PNH
granulocytes who have no con­tra­in­di­ca­tions to anticoagulation.38
However, pri­ mary anticoagulation pro­ phy­
laxis in PNH remains
con­tro­ver­sial. Patients with PNH who have con­com­i­tant throm­
bo­cy­to­pe­nia have increased risks of bleed­ing with anticoagula-
tion, which, before the advent of com­ple­ment inhib­i­tors, have
con­trib­uted to PNH mor­tal­ity. In addi­tion, unlike com­ple­ment
inhib­i­tors, ther­a­peu­tic anticoagulation alone does not entirely
pre­vent throm­bo­sis in patients with PNH.37,38,50 In sit­u­a­tions where
patients with clas­si­cal PNH do not have access to com­ple­ment
inhib­i­tors, pri­mary anticoagulation pro­phy­laxis should be con­sid­
ered. After com­ple­ment inhib­i­tor ther­apy has been ini­ti­ated, pri­
mary anticoagulation pro­phy­laxis can be discontinued.
As of 2021, the US Food and Drug Administration has now
approved 3 com­ple­ment inhib­it­ ors for the treat­ment of PNH. Ecu-
lizumab and the lon­ger-act­ing ravulizumab both tar­get ter­mi­nal
com­ple­ment C5 and effec­tively abro­gate intra­vas­cu­lar hemo­ly­sis,
reduce throm­botic events, and improve qual­ity of life of patients
with PNH.52-55 However, approx­ im
­ a­tely 4% to 27% of patients
treated with C5 inhib­i­tors con­tinue to expe­ri­ence ane­mia due
to extra­vas­cu­lar hemo­ly­sis from the accu­mu­la­tion of unop­posed
C3b on PNH RBCs.56 Although C5 inhi­bi­tion does not address
extra­vas­cu­lar hemo­ly­sis, the recently approved, first-in-class
C3 inhib­i­tor pegcetacoplan, which acts more prox­i­mally in the
150  |  Hematology 2021  |  ASH Education Program
com­ple­ment cas­cade, was found to be supe­rior to eculizumab
in improv­ing hemo­glo­bin and nor­mal­iz­ing hemo­lytic mark­ers.57
Approved for both front­line or sec­ond-line ther­apy of PNH, the
twice-weekly, sub­cu­ta­ne­ously admin­is­tered agent pegcetaco-
plan will be par­tic­u­larly use­ful in patients who remain mod­er­ately
or severely ane­mic despite C5 inhi­bi­tion because of extra­vas­cu­lar
hemo­ly­sis. Several other prox­i­mal com­ple­ment inhib­i­tors, some
of which are orally admin­is­tered, are in clin­ic­ al tri­als.58
Summary
PNH remains a very rare, orphan dis­ease with a widely avail­­able
and highly sen­si­tive screen­ing test, the results of which can have
nuanced impli­ca­tions for diag­no­sis, fol­low-up, and com­ple­ment
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/143/1852074/143babushok.pdf by guest on 13 December 2021
inhib­i­tor ther­apy. PNH clones are patho­gno­monic for immune-
medi­ated bone mar­row fail­ure. Patients with AA, who com­prise
most indi­vid­u­als with PNH clones, have smaller PNH clones and
are rarely symp­tom­atic from PNH. In the absence of hemo­ly­sis
or PNH symp­toms, patients with AA with sub­clin­i­cal PNH clones
do not ben­e­fit from com­ple­ment inhib­i­tors but should be mon­
i­tored pro­spec­tively as most clas­si­cal PNH arises out of AA. In
con­trast, com­ple­ment inhib­i­tors improve out­comes of patients
with clas­si­cal PNH, who invari­ably have large PNH clones, clin­i­
cal hemo­ly­sis, and a high bur­den of PNH-asso­ci­ated symp­toms.
An indi­vid­u­al­ized approach is required when eval­u­at­ing patients
between these extremes, and refer­ral to a ter­tiary care cen­ter
with exper­tise in man­ag­ing PNH and AA is recommended.
Acknowledgments
The author would like to thank the cur­rent and for­mer mem­
bers of the Penn/CHOP Bone Marrow Failure Center for help­ful
dis­cus­sions as well as patients and their fam­i­lies for par­tici­pat­
ing in research stud­ies of bone mar­row fail­ure. This work was
­supported by NHLBI grant K08 HL132101.
Conflict-of-inter­est dis­clo­sure
Daria V. Babushok: declares no com­pet­ing finan­cial inter­ests.
Off-label drug use
Daria V. Babushok: no off-label drug use discussed.
Correspondence
Daria V. Babushok, Room 808 Biomedical Research Build-
ing II/III, 421 Curie Blvd, Philadelphia, PA 19104; e-mail: daria​
­.babushok@pennmedicineupenn​­.edu.
References
1. Hill A, DeZern AE, Kinoshita T, Brodsky RA. Paroxysmal noc­tur­nal haemo-
globinuria. Nat Rev Dis Primers. 2017;3:17028.
2. Hu R, Mukhina GL, Piantadosi S, Barber JP, Jones RJ, Brodsky RA. PIG-A
muta­tions in nor­mal hema­to­poi­e­sis. Blood. 2005;105(10):3848-3854.
3. Araten DJ, Nafa K, Pakdeesuwan K, Luzzatto L. Clonal pop­u­la­tions of hema­
to­poi­etic cells with par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria geno­type and
phe­no­type are pres­ent in nor­mal indi­vid­u­als. Proc Natl Acad Sci U S A.
1999;96(9):5209-5214.
4. Shin T-H, Baek E-J, Corat M-A-F, et al. CRISPR/Cas9 PIG -A gene editing in
non­hu­man pri­mate model dem­on­strates no intrin­sic clonal expan­sion of
PNH HSPCs. Blood. 2019;133(23):2542-2545.
5. Murakami Y, Kinoshita T, Maeda Y, Nakano T, Kosaka H, Takeda J. Differ-
ent roles of glycosylphosphatidylinositol in var­i­ous hema­to­poi­etic cells
as revealed by a mouse model of par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria.
Blood. 1999;94(9):2963-2970.
6. Keller P, Tremml G, Rosti V, Bessler M. X inac­ti­va­tion and somatic cell selec­
tion res­cue female mice car­ry­ing a Piga-null muta­tion. Proc Natl Acad Sci
U S A. 1999;96(13):7479-7483.
7. Katagiri T, Kawamoto H, Nakakuki T, et al. Individual hema­to­poi­etic stem
cells in human bone mar­row of patients with aplastic ane­mia or myelo-
dysplastic syn­drome sta­bly give rise to lim­ited cell lin­e­ages. Stem Cells.
2013;31(3):536-546.
8. Shah YB, Priore SF, Li Y, et al. The pre­dic­tive value of PNH clones, 6p CN-
LOH, and clonal TCR gene rearrangement for aplastic ane­mia diag­no­sis.
Blood Adv. 2021;5(16):3216-3226.
9. Fattizzo B, Ireland R, Dun­lop A, et al. Clinical and prog­nos­tic sig­nif­i­cance
of small par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria clones in myelodysplastic
syn­drome and aplastic ane­mia [published online 4 March 2021]. Leukemia.
10. Sugimori C, Mochizuki K, Qi Z, et al. Origin and fate of blood cells defi­cient
in glycosylphosphatidylinositol-anchored pro­ tein among patients with
bone mar­row fail­ure. Br J Haematol. 2009;147(1):102-112.
11. Richards SJ, Whitby L, Cullen MJ, et  al. Development and eval­u­a­tion of
a sta­bi­lized whole-blood prep­a­ra­tion as a pro­cess con­trol mate­rial for
screen­ing of par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria by flow cytom­e­try.
Cytometry B Clin Cytom. 2009;76(1):47-55.
12. Cannizzo E, Raia M, De Propris MS, et al. Features, rea­son for test­ing, and
changes with time of 583 par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria clones from
529 patients: a mul­ti­cen­ter Ital­ian study. Ann Hematol. 2019;98(5):1083-
1093.
13. Sun L, Babushok DV. Secondary myelodysplastic syn­drome and leu­ke­mia
in acquired aplastic ane­mia and par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria.
Blood. 2020;136(1):36-49.
14. Richards SJ, Dickinson AJ, Cullen MJ, et al. Presentation clin­i­cal, haemato-
logical and immunophenotypic fea­tures of 1081 patients with GPI-defi­cient
(par­ox­ys­mal noc­tur­nal haemoglobinuria) cells detected by flow cytom­e­try.
Br J Haematol. 2020;189(5):954-966.
15. Yoroidaka T, Hosokawa K, Imi T, et al. Hematopoietic stem pro­gen­i­tor cells
lacking HLA dif­fer from those lacking GPI-anchored pro­teins in the hier­
ar­chi­cal stage and sen­si­tiv­ity to immune attack in patients with acquired
aplastic ane­mia [published online 6 April 2021]. Leukemia.
16. Gurnari C, Pagliuca S, Patel BJ, et al. Implication of PIGA geno­type on eryth­
ro­cytes phe­no­type in par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria. Leukemia.
2021;35(8):2431-2434.
17. Brubaker LH, Essig LJ, Mengel CE. Neutrophil life span in par­ox­ys­mal noc­
tur­nal hemo­glo­bin­uria. Blood. 1977;50(4):657-662.
18. Reid ME. Cromer-related blood group anti­gens and the gly­co­syl phospha-
tidylinositol-linked pro­tein, decay-accel­er­at­ing fac­tor DAF (CD55). Immu-
nohematology. 1990;6(2):27-29.
19. Borowitz MJ, Craig FE, Digiuseppe JA, et al; Clinical Cytometry Society.
Guidelines for the diag­no­sis and mon­i­tor­ing of par­ox­ys­mal noc­tur­nal
hemo­glo­bin­uria and related dis­or­ders by flow cytom­e­try. Cytometry B
Clin Cytom. 2010;78(4):211-230.
20. Sloand EM, Mainwaring L, Keyvanfar K, et al. Transfer of glycosylphosphati-
dylinositol-anchored pro­teins to defi­cient cells after eryth­ro­cyte trans­fu­sion
in par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria. Blood. 2004;104(12):3782-3788.
21. Rosse WF. The life-span of com­ple­ment-sen­si­tive and -insen­si­tive red cells
in par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria. Blood. 1971;37(5):556-562.
22. Le Garff-Tavernier M, Debliquis A, Boyer T, et al. Persistence of PNH clones
over time: insights from the mid-term anal­y­sis of the French nation­wide
mul­ti­cen­ter pro­spec­tive obser­va­tional study. Amer­i­can Society of Hema-
tology. Blood. 2019;134(suppl 1):1218.
23. Killick SB, Bown N, Cavenagh J, et al; Brit­ish Society for Standards in Hae-
matology. Guidelines for the diag­no­sis and man­age­ment of adult aplastic
anae­mia. Br J Haematol. 2016;172(2):187-207.
24. Illingworth AJ, Marinov I, Sutherland DR. Sensitive and accu­rate iden­ti­fi­ca­
tion of PNH clones based on ICCS/ESCCA PNH Consensus Guidelines—a
sum­mary. Int J Lab Hematol. 2019;41 (suppl 1):73-81.
25. Debliquis A, Wagner-Ballon O, Le Garff-Tavernier M, et al; HPN-AFC Group.
Evaluation of par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria screen­ing by flow
cytom­e­try through multicentric interlaboratory com­par­i­son in four countries.
Am J Clin Pathol. 2015;144(6):858-868.
26. DeZern AE, Symons HJ, Resar LS, Borowitz MJ, Armanios MY, Brodsky RA.
Detection of par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria clones to exclude inher-
ited bone mar­row fail­ure syn­dromes. Eur J Haematol. 2014;92(6):467-470.
27. Keller P, Debaun MR, Rothbaum RJ, Bessler M. Bone mar­row fail­ure in Shwa-
chman-Diamond syn­drome does not select for clonal haematopoiesis of
the par­ox­ys­mal noc­tur­nal haemoglobinuria phe­no­type. Br J Haematol.
2002;119(3):830-832.
28. Kennedy AL, Myers KC, Bowman J, et al. Distinct genetic path­ways define
pre-malig­nant ver­sus com­pen­sa­tory clonal hema­to­poi­e­sis in Shwachman-
Diamond syn­drome. Nat Commun. 2021;12(1):1334.
Prakash Singh Shekhawat
29. Zhao X, Zhang L, Jing L, et al. The role of par­ox­ys­mal noc­tur­nal hemo­glo­
bin­uria clones in response to immu­no­sup­pres­sive ther­apy of patients with
severe aplastic ane­mia. Ann Hematol. 2015;94(7):1105-1110.
30. Sugimori C, Chuhjo T, Feng X, et  al. Minor pop­u­la­tion of CD55–CD59–
blood cells pre­dicts response to immu­no­sup­pres­sive ther­apy and prog­no­
sis in patients with aplastic ane­mia. Blood. 2006;107(4):1308-1314.
31. Wang H, Chuhjo T, Yasue S, Omine M, Nakao S. Clinical sig­nif­i­cance of a
minor pop­u­la­tion of par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria-type cells in
bone mar­row fail­ure syn­drome. Blood. 2002;100(12):3897-3902.
32. Ishiyama K, Chuhjo T, Wang H, Yachie A, Omine M, Nakao S. Polyclonal
hema­to­poi­e­sis maintained in patients with bone mar­row fail­ure har­bor­
ing a minor pop­u­la­tion of par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria-type cells.
Blood. 2003;102(4):1211-1216.
33. Wang B, He B, Zhu Y-D, Wu W. The pre­dic­tive value of pre-treat­ment
par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria clone on response to immu­no­sup­
pres­sive ther­apy in patients with aplastic ane­mia: a meta-anal­y­sis. Hema-
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/143/1852074/143babushok.pdf by guest on 13 December 2021
tology. 2020;25(1):464-472.
34. Yoshizato T, Dumitriu B, Hosokawa K, et al. Somatic muta­tions and clonal
hema­to­poi­e­sis in aplastic ane­mia. N Engl J Med. 2015;373(1):35-47.
35. Kelly RJ, Hill A, Arnold LM, et al. Long-term treat­ment with eculizumab in
par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria: sustained effi­cacy and improved
sur­vival. Blood. 2011;117(25):6786-6792.
36. de Latour RP, Mary JY, Salanoubat C, et al; French Society of Hematol-
ogy; French Association of Young Hematologists. Paroxysmal noc­ tur­
nal hemo­glo­bin­uria: nat­u­ral his­tory of dis­ease subcategories. Blood.
2008;112(8):3099-3106.
37. Hill A, Kelly RJ, Hillmen P. Thrombosis in par­ox­ys­mal noc­tur­nal hemo­glo­bin­
uria. Blood. 2013;121(25):4985-4996, quiz 5105.
38. Hall C, Richards S, Hillmen P. Primary pro­ phy­ laxis with war­ fa­
rin pre­
vents throm­bo­sis in par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria (PNH). Blood.
2003;102(10):3587-3591.
39. Hill A, Sapsford RJ, Scally A, et  al. Under-rec­ og­ nized com­ pli­
ca­tions in
patients with par­ox­ys­mal noc­tur­nal haemoglobinuria: raised pul­mo­
nary pres­sure and reduced right ven­tric­u­lar func­tion. Br J Haematol.
2012;158(3):409-414.
40. Barcellini W, Scola E, Lanfranconi S, et al. Paroxysmal noc­tur­nal hemo­glo­
bin­uria (PNH): brain MRI ische­mic lesions in neu­ro­log­i­cally asymp­tom­atic
patients. Sci Rep. 2018;8(1):476.
41. Kokoris SI, Gavriilaki E, Miari A, et al. Renal involve­ment in par­ox­ys­mal noc­
tur­nal hemo­glo­bin­uria: an update on clin­i­cal fea­tures, path­o­phys­i­­ol­ogy
and treat­ment. Hematology. 2018;23(8):558-566.
42. Ge M, Li X, Shi J, Shao Y, Zheng Y. Clinical fea­tures and prog­nos­tic fac­tors
of Asian patients with par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria: results from a
sin­gle cen­ter in China. Ann Hematol. 2012;91(7):1121-1128.
43. Hoechsmann B, Peffault De Latour R, Hill A, et al. Risk fac­tors for throm­
botic events in patients with PNH: a nested case-con­trol study in the inter­
na­tional PNH reg­is­try. Blood. 2020;136(suppl 1):6-8.
44. Lee JW, Jang JH, Kim JS, et al. Clinical signs and symp­toms asso­ci­ated with
increased risk for throm­bo­sis in patients with par­ox­ys­mal noc­tur­nal hemo­
glo­bin­uria from a Korean Registry. Int J Hematol. 2013;97(6):749-757.
45. Schrezenmeier H, Muus P, Socié G, et al. Baseline char­ac­ter­is­tics and dis­
ease bur­den in patients in the International Paroxysmal Nocturnal Hemo-
globinuria Registry. Haematologica. 2014;99(5):922-929.
46. Long Z, Du Y, Li H, Han B. Polymorphism of the ABO gene asso­ci­ate with
throm­bo­sis risk in patients with par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria.
Oncotarget. 2017;8(54):92411-92419.
47. Griffin M, Hillmen P, Munir T, et  al. Significant hemo­ly­sis is not required
for throm­bo­sis in par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria. Haematologica.
2019;104(3):e94-e96.
48. Huang Y, Liu X, Chen F, et al. Prediction of throm­bo­sis risk in patients with
par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria. Ann Hematol. 2019;98(10):2283-
2291.
49. Füreder W, Sperr WR, Heibl S, et  al. Prognostic fac­tors and fol­low-up
param­e­ters in patients with par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria (PNH):
expe­ri­ence of the Austrian PNH net­work. Ann Hematol. 2020;99(10):2303-
2313.
50. Hillmen P, Muus P, Dührsen U, et  al. Effect of the com­ple­ment inhib­i­tor
eculizumab on throm­bo­em­bo­lism in patients with par­ox­ys­mal noc­tur­nal
hemo­glo­bin­uria. Blood. 2007;110(12):4123-4128.
51. Macrae FL, Peacock-Young B, Bowman P, et al. Patients with par­ox­ys­mal
noc­tur­nal hemo­glo­bin­uria dem­on­strate a prothrombotic clot­ting phe­no­
type which is improved by com­ple­ment inhi­bi­tion with eculizumab. Am J
Hematol. 2020;95(8):944-952.
Evaluation and sig­nif­i­cance of PNH clones | 151
52. Hillmen P, Young NS, Schubert J, et al. The com­ple­ment inhib­i­tor eculizumab
in par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria. N Engl J Med. 2006;355(12):1233-
1243.
53. Brodsky RA, Young NS, Antonioli E, et al. Multicenter phase 3 study of the
com­ple­ment inhib­i­tor eculizumab for the treat­ment of patients with par­ox­
ys­mal noc­tur­nal hemo­glo­bin­uria. Blood. 2008;111(4):1840-1847.
54. Lee JW, Sicre de Fontbrune F, Wong Lee Lee L, et  al. Ravulizumab
(ALXN1210) vs eculizumab in adult patients with PNH naive to com­ple­ment
inhib­i­tors: the 301 study. Blood. 2019;133(6):530-539.
55. Kulasekararaj AG, Hill A, Rottinghaus ST, et al. Ravulizumab (ALXN1210) vs
eculizumab in C5-inhib­i­tor-expe­ri­enced adult patients with PNH: the 302
study. Blood. 2019;133(6):540-549.
56. Brodsky RA, Peffault de Latour R, Rottinghaus ST, et al. Characterization of
break­through hemo­ly­sis events observed in the phase 3 ran­dom­ized stud­
152  |  Hematology 2021  |  ASH Education Program
ies of ravulizumab ver­sus eculizumab in adults with par­ox­ys­mal noc­tur­nal
hemo­glo­bin­uria. Haematologica. 2021;106(1):230-237.
57. Hillmen P, Szer J, Weitz I, et al. Pegcetacoplan ver­sus eculizumab in par­ox­
ys­mal noc­tur­nal hemo­glo­bin­uria. N Engl J Med. 2021;384(11):1028-1037.
58. Risitano AM, Marotta S. Toward com­ple­ment inhi­bi­tion 2.0: next gen­er­a­
tion anticomplement agents for par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria. Am
J Hematol. 2018;93(4):564-577.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000245
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/143/1852074/143babushok.pdf by guest on 13 December 2021
 WHAT DO APLASTIC ANEMIA, PNH, AND “ HYPOPLASTIC ” LEUKEMIA HAVE IN COMMON ?

When to worry about inherited bone marrow failure

and myeloid malignancy predisposition syndromes
in the setting of a hypocellular marrow

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/153/1851462/153narla.pdf by guest on 13 December 2021

Anupama Narla
Stanford University School of Medicine, Stanford, CA

With our increasing understanding of inherited marrow failure and myeloid malignancy predisposition syndromes, it has
become clear that there is a wide phenotypic spectrum and that these diseases must be considered in the differential
diagnosis of both children and adults with unexplained defects in hematopoiesis. Moreover, these conditions are not as
rare as previously believed and may present as aplastic anemia, myelodysplastic syndrome, or malignancy over a range
of ages. Establishing the correct diagnosis is essential because it has implications for treatment, medical management,
cancer screening, and family planning. Our goal is to highlight insights into the pathophysiology of these diseases,
review cryptic presentations of these syndromes, and provide useful references for the practicing hematologist.

LEARNING OBJECTIVES
• Review the basic clinical features of the inherited BM failure and myeloid malignancy predisposition syndromes
that can lead to aplastic anemia, myelodysplasia, and malignancy
• Emphasize the role of a thorough physical exam and detailed family history to identify these disorders and deter­
mine the need for genetic testing for the patient and family
• Discuss the initial diagnostic workup of these disorders and potential treatment modifications if a specific condi­
tion is identified

Introduction: Inherited bone marrow failure syndromes
Inherited bone marrow failure syndromes (iBMFs) encom­
pass a diverse collection of diseases. While they are rare
causes of hematologic disorders, it is essential for the
practicing hematologist to be aware of iBMFs. These dis­
orders have a wide phenotypic spectrum and may pres­
ent cryptically in adult patients with cytopenias in one or
more lineages. Furthermore, our evolving understanding
of the pathophysiology of these disorders is critical to
our general knowledge of the hematopoietic system. An
overview of these syndromes is summarized in Table 1. We
have highlighted the disorders that have a predisposition
to aplastic anemia, myelodysplastic syndrome (MDS), and
malignancy. A brief overview of each of these disorders
follows.
Congenital amegakaryocytic thrombocytopenia
Congenital amegakaryocytic thrombocytopenia (CAMT) is
a rare autosomal recessive disease characterized by an iso­
lated, severe hypomegakaryocytic thrombocytopenia that
Prakash Singh Shekhawat
presents in infancy and progresses to pancytopenia and
bone marrow (BM) failure in later childhood. Clinical man­
ifestations can include petechiae at birth and intracranial
hemorrhage, but unlike other iBMFs, congenital anomalies
are rare. CAMT is most commonly due to mutations in MPL,
which encodes for the thrombopoietin receptor. In contrast
to the activating MPL mutations seen in myeloproliferative
neoplasms, CAMT mutations always lead to diminished or
absent signaling of the thrombopoietin receptor. In gen­
eral, loss­of­function mutations are associated with more
severe thrombocytopenia and early­onset pancytopenia
(CAMT type 1). In cases with MPL gene missense mutations,
patients have a transient increase in platelet counts dur­
ing the first year of life (CAMT type 2). Heterozygous muta­
tions in MECOM (MDS1 and EVI1 complex locus) have been
reported to be causative of a rare association of CAMT and
radioulnar synostosis. Mutations in MECOM can present
with a wide range of phenotypes ranging from thrombo­
cytopenia to full­blown hypocellular marrow failure/aplastic
anemia early in life, again demonstrating the importance of
Bone marrow failure syndromes and aplastic anemia | 153
Table 1. Features of inherited bone nar­row and mye­loid malig­nancy pre­dis­po­si­tion syn­dromes that may pres­ent as aplastic
ane­mia/hypocellular mar­row

Disease Presentation Useful clin­i­cal resources

Amegakaryocytic thromobocytopenia
Diamond-Blackfan ane­mia
Dyskeratosis congenita
Fanconi ane­mia
Nonsyndromic throm­bo­cy­to­pe­nia, pan­cy­to­pe­nia,
BM fail­ure
Macrocytic ane­mia with reticulocytopenia, con­gen­i­tal
anom­a­lies/short stat­ure
Triad of abnor­mal skin pig­men­ta­tion, oral leukoplakia,
and nail dys­tro­phy; aplastic ane­mia, pul­mo­nary fibro­sis
Cytopenias, con­gen­i­tal malformations, can­cer at a young
age, sen­si­tiv­ity to toxic effects of che­mo­ther­apy or
radi­a­tion, MDS/AML
https:​­/​­/www​­.pdsa​­.org​­/inherited​
­-thrombocytopenia​­.html
https:​­/​­/dbafoundation​­.org​­/
https:​­/​­/www​­.dbar​­.org​­/
https:​­/​­/teamtelomere​­.org​­/
https:​­/​­/www​­.fanconi​­.org​­/

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/153/1851462/153narla.pdf by guest on 13 December 2021

RUNX1 Thrombocytopenia, MDS/AML https:​­/​­/www​­.runx1​­-fpd​­.org​­/
GATA-2 defi­ciency Recurrent infec­tions, lymphedema, warts, pul­mo­nary https:​­/​­/rarediseases​­.org​­/gard​­-rare​­-disease​­
alve­o­lar proteinosis, MDS/AML /gata2​­-deficiency​­/
SAMD9/9L dis­or­ders MIRAGE (myelodysplasia, infec­tion, restric­tion of growth, https:​­/​­/www​­.stjude​­.org​­/samd9
adre­nal hypo­pla­sia, gen­i­tal phe­no­types, enter­op­a­thy),
MDS/AML
Severe con­gen­i­tal neutropenia Neutropenia, recur­rent bac­te­rial infec­tions https:​­/​­/neutropenianet​­.org​­/
Shwachman-Diamond syn­drome Neutropenia, exo­crine pan­cre­atic insuf­fi­ciency, MDS/AML http:​­/​­/sdsregistry​­.org​­/

screen­ing for these rare dis­eases and the need for ongo­ing sci­en­
tific dis­cov­ery.
Diamond-Blackfan ane­mia
Diamond-Blackfan ane­ mia (DBA) was orig­ i­
nally described by
Josephs in 1936 and fur­ ther char­ ac­ter­
ized by Diamond and
Blackfan in 1938 as a con­gen­i­tal hypo­plas­tic ane­mia.1 In addi­tion
to hypo­plas­tic ane­mia, the dis­or­der is char­ac­ter­ized by mac­ro­
cy­to­sis, reticulocytopenia, and ele­vated lev­els of eryth­ro­cyte
aden­ o­
sine deam­ i­
nase. Additionally, half of patients with DBA
also pres­ent with phys­i­cal abnor­mal­i­ties, includ­ing short stat­ure,
thumb abnor­ mal­i­
ties (clas­
si­
cally, a triphalangeal thumb), cra­
nio­fa­cial defects, and cleft lip/pal­ate. DBA was the first dis­ease
to be linked to impaired ribo­some func­tion and is the founding
mem­ber of a group of dis­or­ders now known as ribosomopa­
thies.2 Several other iBMFs as well as the 5q− MDS and can­cers
have sub­se­quently been linked to muta­tions in genes encoding
for ribo­somal pro­teins or pro­teins required for nor­mal ribo­some
func­tion.3,4 While the pre­dis­po­si­tion to MDS is lower than other
iBMFs, there is a risk of other tumors, includ­ing colo­rec­tal can­
cer. In addi­tion, DBA is known to have incom­plete pen­e­trance,
as dem­on­strated both by the obser­va­tion of sib­lings who carry
the same muta­tion dis­cor­dant for the pres­ence of ane­mia and by
the occur­rence of spon­ta­ne­ous remis­sions in affected patients,
which is why cli­ni­cians need to have a high index of sus­pi­cion
in patients with unex­plained ane­mia and mar­row abnor­mal­i­ties.
Dyskeratosis congenita
In 1910 a case report was the first to define a syn­drome, ulti­mately
named dyskeratosis congenita (DC), char­ac­ter­ized by a triad of
abnor­mal skin pig­men­ta­tion, oral leukoplakia, and nail dys­tro­
phy.5 The spec­trum of DC has expanded con­sid­er­ably since then
to include effects on every organ sys­tem, par­tic­u­larly the BM.
Other clin­i­cal find­ings may include early graying of hair or hair
loss, short stat­ure, devel­op­men­tal delay, blepharitis, peri­odon­tal
dis­ease, pul­mo­nary fibro­sis, esoph­a­geal ste­no­sis, ure­thral ste­no­
sis, liver dis­ease, and avas­cu­lar necro­sis of the hips or shoul­ders.
Almost 90% of patients with DC will even­tu­ally develop a cyto­
penia of one or more periph­eral blood lin­e­ages, and BM fail­ure is
the major cause of death. Often BM fail­ure devel­ops in the sec­
ond or third decade of life, but it can occur at birth or as late as
the sev­enth decade of life. The BM abnor­mal­i­ties can evolve into
MDS or acute mye­loid leu­ke­mia (AML).6
The clin­i­cal diag­no­sis of DC can be chal­leng­ing given its phe­
no­typic het­ero­ge­ne­ity, dif­fer­ent modes of inher­i­tance (X-linked,
auto­so­mal reces­sive, and auto­so­mal dom­i­nant), and var­i­able age
of onset. However, despite the wide spec­trum of the dis­ease,
rang­ing from clas­sic DC to aplastic ane­mia, it is clear that the
under­ly­ing pathol­ogy is due to defec­tive telo­mere main­te­nance.
The term “DC” is now used inter­change­ably with telo­mere biol­
ogy dis­ or­
ders and short telo­ mere syn­ dromes. The func­ tional
assess­ment of telo­mere length with the find­ing of a telo­mere
length less than the first per­cen­tile for age in leu­ko­cyte sub­sets
is highly sen­si­tive but not spe­cific to the diag­no­sis of these dis­
or­ders and has been seen in other inherited dis­or­ders.7,8 Regard­
less, the eval­ua ­ ­tion of telo­mere length is crit­i­cal for any patient
with pan­cy­to­pe­nia and abnor­mal mar­row find­ings given the
impli­ca­tions for screen­ing and mod­i­fi­ca­tions in stem cell trans­
plant con­di­tion­ing.
Fanconi ane­mia
Fanconi ane­mia (FA) was first described by Guido Fanconi in 1927 in
3 broth­ers with pan­cy­to­pe­nia and con­gen­i­tal abnor­mal­i­ties.9 The
dis­ease typ­i­cally progresses through sev­eral clin­i­cal stages, orga­
nized by age. In the first stage, in infancy and early child­hood, con­
gen­i­tal anom­a­lies may be pres­ent, although they are not required
for the diag­no­sis of FA and range from mild to severe. The most
com­ mon malformations include short stat­ ure, hypopigmented

154  |  Hematology 2021  |  ASH Education Program

or café au lait spots, thumb or radial ray abnor­mal­i­ties, micro- or
hydro­ceph­aly, struc­tural renal anom­a­lies, hypogonadism, and
devel­op­men­tal delay.
Within the first decade, patients may pres­ent with throm­bo­cy­
to­pe­nia and mac­ro­cy­to­sis before progressing to BM fail­ure, when
the diag­no­sis is often first made. During ado­les­cence and adult­
hood, the risk of AML and MDS becomes very high. In older adults,
a range of solid tumors, par­tic­u­larly squa­mous cell car­ci­no­mas of
the head/neck and gen­i­to­uri­nary track, can be seen.10 Throughout
life, the hema­to­poi­etic phe­no­type can change because of genetic
rever­sion or clonal evo­lu­tion, so a high index of sus­pi­cion and a
care­ful his­tory, includ­ing a fam­ily his­tory of can­cer pre­dis­po­si­tion,
is essen­tial. These patients are highly sen­si­tive to che­mo­ther­apy
and radi­a­tion, which is why it is crit­i­cal to rule out this diag­no­sis
prior to trans­plant for any patient with unex­plained pan­cy­to­pe­nia.
Severe con­gen­i­tal neutropenia
Severe con­gen­i­tal neutropenia (SCN) is char­ac­ter­ized by abso­
lute neu­tro­phil counts con­sis­tently less than 200/uL with recur­
rent severe infec­tions, which often develop within the first few
months of life. Evaluation of the BM dem­on­strates a mye­loid mat­
u­ra­tion arrest. The dis­ease can be auto­so­mal reces­sive (Kostmann
syn­drome), X-linked, or auto­so­mal dom­i­nant. In Kostmann syn­
drome, muta­tions have been iden­ti­fied in the HAX1, G6PC3, and
GFI1 genes. In the X-linked dis­ease, muta­tions have been found in
the WAS gene. Finally, muta­tions in the ELANE gene are the most
com­mon cause of SCN and cause auto­so­mal dom­i­nant dis­ease. As
with Shwachman-Diamond syn­drome, below), addi­tional somatic
muta­tions are often acquired over time, which serve as impor­tant
mark­ers for dis­ease risk and may affect treat­ment deci­sions.
Shwachman-Diamond syn­drome
Shwachman-Diamond syn­drome (SDS, also known as Shwachman-
Diamond-Oski syn­drome) was ini­tially described in the early 1960s
as a dis­or­der of exo­crine pan­cre­atic dys­func­tion and BM fail­ure.11
Patients gen­er­ally pres­ent with ste­at­or­rhea, growth fail­ure, and
recur­rent infec­tions. The hema­to­logic abnor­mal­i­ties are pre­
dom­i­nantly neutropenia, although ane­mia, throm­bo­cy­to­pe­nia,
and pan­cy­to­pe­nia can occur. Skeletal abnor­mal­i­ties (short stat­
ure, delayed appear­ ance of normally shaped epiph­ y­
ses, and
pro­gres­sive metaphyseal thick­en­ing/dys­pla­sia) as well as poor
growth are com­mon in SDS patients. Patients with­out severe
pan­cre­atic dis­ease may pres­ent later in child­hood or as adults,
and some patients may pres­ent with aplastic ane­mia or AML as
the ini­tial man­i­fes­ta­tion of the dis­ease.12
The diag­no­sis of SDS relies on clin­i­cal find­ings. Exocrine pan­
cre­atic dys­func­tion, which can be sub­clin­i­cal, can be established
by low serum tryp­sin­o­gen in patients youn­ger than 3 years or low
pan­cre­atic isoamylase in patients older than 3 years. In addi­tion,
patients may have low fecal elas­tase or a fatty pan­creas that can
be dem­on­strated by imag­ing. Genetic test­ing can be help­ful to
con­firm the diag­no­sis, but a neg­a­tive test does not exclude the
diag­no­sis. In 90% of affected patients, this auto­so­mal reces­sive
dis­or­der is due to muta­tions in the Shwachman-Bodian-Diamond
syn­drome (SBDS) gene.13 It is essen­tial to dis­tin­guish SDS from
cys­tic fibro­sis (the most com­mon cause of exo­crine pan­cre­atic
insuf­fi­ciency in chil­dren), con­gen­i­tal neutropenia, and Pearson
syn­drome.
Prakash Singh Shekhawat
Myeloid malig­nancy pre­dis­po­si­tion syn­dromes
GATA-2 defi­ciency
The con­stel­la­tion of dis­sem­i­nated nontuberculous myco­bac­te­
rial infec­tions; sus­cep­ti­bil­ity to viral infec­tions, espe­cially human
pap­il­lo­ma­vi­rus and Epstein-Barr virus; and fun­gal infec­tions, cou­
pled with pro­found monocytopenia and B-cell and nat­u­ral killer
cell lymphopenias, was syndromically described as MonoMAC
and den­dritic cell, mono­cyte, B, and nat­u­ral killer lym­phoid defi­
ciency. Mutations in GATA-2 were iden­ti­fied in 2011 as the cause
of this group of dis­or­ders, which are also asso­ci­ated with mar­
row fail­ure, MDS, and AML.14 The dis­ease fol­lows an auto­so­mal
dom­i­nant mode of inher­i­tance as well as a large num­ber of spo­
radic cases. Presentations vary widely among affected mem­bers
even within the same fam­ily, but the BM is often ini­tially hypocel­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/153/1851462/153narla.pdf by guest on 13 December 2021
lular and may have strik­ing multilineage dys­pla­sia.
RUNX1
RUNX1 is a mem­ber of the core-bind­ing fac­tor fam­ily of tran­scrip­
tion fac­tors and is indis­pens­able for defin­i­tive hema­to­poi­e­sis;
it is one of the most fre­quently mutated genes in a vari­ety of
hema­to­log­i­cal malig­nan­cies. Germ line muta­tions in RUNX1 have
been shown to cause a famil­ial plate­let dis­or­der with asso­ci­ated
malig­nan­cies, includ­ing pedi­at­ric MDS. Patients may pres­ent
with iso­lated throm­bo­cy­to­pe­nia and mega­kar­yo­cytic dysmor­
phia or atypia on base­line BM eval­u­a­tion before progressing to
pan­cy­to­pe­nia, multilineage dys­pla­sia, increased blasts, and the
acqui­si­tion of addi­tional somatic muta­tions. A sub­set of patients
may pres­ent with MDS/AML at a young age. Early rec­og­ni­tion of
germ line muta­tion and pre­dis­po­si­tion to mye­loid malig­nancy
per­mits appro­pri­ate treat­ment, ade­quate mon­i­tor­ing for dis­ease
pro­gres­sion, and proper donor selec­tion for hema­to­poi­etic stem
cell trans­plan­ta­tion, as well as genetic coun­sel­ing of affected
patients and their fam­ily mem­bers.
SAMD9/9L dis­or­ders
Germ line SAMD9 and SAMD9L muta­tions have now been iden­ti­
fied as the cause of a spec­trum of mul­ti­sys­tem dis­or­ders that carry
a mark­edly increased risk of devel­op­ing mye­loid malig­nan­cies with
chro­mo­some 7 abnor­mal­i­ties, includ­ing mono­somy 7, del7q, and
CN-loh of 7q. Affected indi­vid­u­als dis­play a highly var­i­able clin­i­cal
course that ranges from mild and tran­sient dyspoietic changes in
the BM to a rapid pro­gres­sion of MDS or AML with mono­somy 7.
These data have impli­ca­tions for under­stand­ing how SAMD9 and
SAMD9L muta­tions con­trib­ute to mye­loid trans­for­ma­tion and for
rec­og­niz­ing, coun­sel­ing, and treating affected fam­i­lies.
The role of genetic test­ing
Many of the dis­or­ders discussed have spe­cific, iden­ti­fied gene
muta­tions, but there is increas­ing inter­est in under­stand­ing what
other genes may be involved within these dis­eases as well as
in dis­or­ders in which no genes have been iden­ti­fied. Increasing
num­bers of pan­els are being conducted both in com­mer­cial labs
and uni­ver­si­ties that cover BMF only vs whole exome sequenc­
ing. The deci­sion about spe­cific tests to be performed should be
done in con­sul­ta­tion with a genet­i­cist, which will also help guide
deci­sions about fam­ily test­ing. This infor­ma­tion is crit­i­cal for sur­
veil­lance and may help steer treat­ment options regard­ing stem
cell donor trans­plan­ta­tion.
Bone mar­row fail­ure syn­dromes and aplastic ane­mia  |  155
Acquired BM fail­ure syn­dromes that can pres­ent
with signs of mar­row aplasia
Although beyond the scope of this chap­ter, it is worth not­ing
that sev­eral genetic causes of acquired BM fail­ure should be con­
sid­ered in the eval­u­a­tion of patients with cytopenias and hypo­
plas­tic mar­row. Again, a detailed phys­i­cal exam and a care­ful
fam­ily his­tory will help to deter­mine the appro­pri­ate workup and
genetic test­ing. For exam­ple, VEXAS syn­drome is a mono­genic
dis­ease of adult­hood caused by somatic muta­tions in UBA1 in
hema­to­poi­etic pro­gen­i­tor cells that can pres­ent with a range of
immu­no­logic and hema­to­logic symp­toms.15
Conclusions
Maintaining a high sus­pi­cion for rare con­gen­i­tal and acquired
causes of BM fail­ure is crit­ic ­ al when eval­ua
­ t­ing patients of all­
ages with unusual cytopenias. A detailed phys­i­cal exam­i­na­tion
and fam­ily his­tory are also crit­i­cal. These guide the sub­se­quent
lab­o­ra­tory workup, sur­veil­lance sched­ule for malig­nancy, and
poten­tial ther­a­peu­tic options.
Conflict-of-inter­est dis­clo­sure
Anupama Narla: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Anupama Narla: nothing to disclose.
Correspondence
Anupama Narla, Stanford University School of Medicine, CCSR
South 1215b, 269 Campus Dr, Stanford, CA 94305-5162; e-mail:
anunarla@stanford​­.edu.
156  |  Hematology 2021  |  ASH Education Program
References
1. Diamond LK, Blackfan KD. Hypoplastic ane­mia. Am J Dis Child. 1938;15:307.
2. Narla A, Ebert BL. Ribosomopathies: human dis­or­ders of ribo­some dys­
func­tion. Blood. 2010;115(16):3196-3205.
3. Ebert BL, Pretz J, Bosco J, et al. Identification of RPS14 as a 5q- syn­drome
gene by RNA inter­fer­ence screen. Nature. 2008;451(7176):335-339.
4. Pelletier J, Thomas G, Volarević S. Ribosome bio­gen­e­sis in can­cer: new
play­ers and ther­a­peu­tic ave­nues. Nat Rev Cancer. 2018;18(1):51-63.
5. Zinsser F. Atrophia cutis reticularis cum pig­men­ta­tions, dystrophia unguium
at leukoplakis oris (poikioodermia atrophicans vascularis Jacobi). Ikono­
graphia Dermatologica. 1910;5:219-233.
6. Savage SA, Alter BP. Dyskeratosis congenita. Hematol Oncol Clin North
Am. 2009;23(2):215-231.
7. Alder JK, Hanamanthu VS, Strong MA, et al. Diagnostic util­ity of telo­mere
length test­ing in a hos­pi­tal-based set­ting. PNAS. 2018;115(10):E2358-E2365.
8. Schratz KE, Haley L, Danoff SK, et al. Cancer spec­trum and out­comes in the
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/153/1851462/153narla.pdf by guest on 13 December 2021
Men­de­lian short telo­mere syn­dromes. Blood. 2020;135(22):1946-1956.
9. Lobitz S, Velleuer E. Guido Fanconi (1892-1979): a jack of all­trades. Nat Rev
Cancer. 2006;6(11):893-898.
10. Bagby GC, Alter BP. Fanconi ane­mia. Semin Hematol. 2006;43(3):147-156.
11. Shwachman H, Diamond LK, Oski FA, Khaw KT. The syn­drome of pan­cre­atic
insuf­fi­ciency and bone mar­row dys­func­tion. J Pediatr. 1964;65(5):645-663.
12. Myers KC, Davies SM, Shimamura A. Clinical and molec­ul­ar path­o­phys­i­­ol­
ogy of Shwachman-Diamond syn­drome: an update. Hematol Oncol Clin
North Am. 2013;27(1):117-128, ix.
13. Boocock GR, Morrison JA, Popovic M, et al. Mutations in SBDS are asso­ci­
ated with Shwachman-Diamond syn­drome. Nat Genet. 2003;33(1):97-101.
14. Hsu AP, Sampaio EP, Khan J, et al. Mutations in GATA2 are asso­ci­ated with
the auto­so­mal dom­i­nant and spo­radic monocytopenia and myco­bac­te­rial
infec­tion (MonoMAC) syn­drome. Blood. 2011;118(10):2653-2655.
15. Grayson PC, Patel BA, Young NS. VEXAS syn­drome. Blood. 2021;137(26):3591-
3594.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000246
 DEFEATING DIFFUSE, DOUBLE - HIT, AND DOGGED NON - HODGKIN LYMPHOMA

Double-hit lymphoma: optimizing therapy

Kieron Dunleavy

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/157/1852000/157dunleavy.pdf by guest on 13 December 2021

Division of Hematology and Oncology, Georgetown Lombardi Comprehensive Cancer Center, Georgetown University Hospital, Washington, DC

Aggressive B-cell lymphoma is a heterogeneous entity with disparate outcomes based on clinical and pathological char-
acteristics. While most tumors in this category are diffuse large B-cell lymphoma (DLBCL), the recognition that some
cases have high-grade morphology and frequently harbor MYC and BCL2 and/or BCL6 translocations has led to their
separate categorization. These cases are now considered distinct from DLBCL and are named “high-grade B-cell lym-
phoma” (HGBL). Most are characterized by distinct rearrangements, but others have high-grade morphological features
without these and are called HGBL-not otherwise specified. Studies have demonstrated that this group of diseases leads
to poor outcomes following standard rituximab, cyclophosphamide, hydroxydaunorubicin, vincristine, and prednisone
therapy; retrospective and recent single-arm, multicenter studies suggest they should be approached with dose-intense
treatment platforms. As yet, this has not been validated in randomized trial settings due to the rarity of these diseases. In
the relapsed and refractory setting, novel approaches such as anti-CD19 chimeric antigen receptor T cells and antibodies
against CD19 have demonstrated high efficacy in this subgroup. Recently, genomic studies have made much progress in
investigating some of the molecular underpinnings that drive their lymphomagenesis and have paved the way for testing
additional novel approaches.

LEARNING OBJECTIVES
• Understand recent developments in the elucidation of MYC and BCL2 aberrations that are helpful in the categori-
zation of these lymphomas and the implications for therapeutic approaches
• Review recent clinical outcomes using various strategies for HGBL, both in the up-front and relapsed/refractory
setting, and understand which novel therapies may be useful in managing these diseases

B-cell lymphoma (HGBL) with MYC and BCL2 transloca-

CLINICAL CASE
A 46-year-old previously well man presented with a 4-week
history of bilateral neck swelling, fevers, night sweats,
and moderate weight loss. Fluorodeoxyglucose-positron
emission tomographic (FDG-PET) imaging demonstrated
hypermetabolic diffuse lymphadenopathy (node diameters
up to 4 cm: maximum standardized uptake values >25) on
both sides of the diaphragm in addition to bone marrow
involvement. His lactate dehydrogenase was elevated; his
Eastern Cooperative Oncology Group performance status
was 1. HIV testing was negative. Following an inconclusive
fine-needle aspiration, he underwent an excisional biopsy
of a right 3-cm cervical lymph node. This demonstrated
aggres sive CD20+ B -cell lymphoma; tumor cells were
CD10+, BCL2+ (>60%), and MYC+ (>90%), and Ki67 was
>90%. Fluorescence in situ hybridization (FISH) studies
demonstrated a rearrangement of MYC and BCL2 with no
BCL6 rearrangement. Hence, his final diagnosis was stage
IVB (International Prognostic Index [IPI] 3), high-grade
Prakash Singh Shekhawat
tions. He received treatment with 6 cycles of rituximab,
cyclophosphamide, hydroxydaunorubicin, vincristine, and
prednisone (R-CHOP) that included intrathecal prophy-
laxis with methotrexate and achieved a complete remis-
sion by end-of-treatment PET scan. Eight weeks following
the PET scan, he developed recurrent symptoms and neck
lymphadenopathy. Reimaging and subsequent biopsy
confirmed recurrent disease.
Introduction
Diffuse large B-cell lymphoma (DLBCL) is now recognized
as and continues to evolve as a clinically and molecu-
larly heterogeneous disease.1-5 While the addition of rit-
uximab to anthracycline-based treatment has resulted in
an overall survival (OS) benefit, standard R-CHOP ther-
apy remains noncurative for a substantial proportion of
people. Among the challenges faced in managing newly
diagnosed patients are accurately predicting who will not
New developments in aggressive B-cell lymphoma | 157
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/157/1852000/157dunleavy.pdf by guest on 13 December 2021
Figure 1. Category of aggressive B-cell lymphomas “HGBLs with MYC and BCL2 and/or BCL6 rearrangements” described in the
2016 revision to the World Health Organization classification of tumors of hematopoietic and lymphoid tumors. Most cases with
MYC and BCL2 rearrangements are of GCB origin, whereas most cases with BCL6 rearrangements are of ABC origin. This category
includes DH lymphomas, which involve MYC and BCL2 or MYC and BCL6, as well as THLs that involve MYC, BCL2, and BCL6. When
translocated, MYC may have an IG or non-IG partner gene, with the former associated with an inferior outcome. In a large study, 7.9%
of tumors with DLBCL morphology were assigned to HGBL-DHL/THL, composing 13.3% of GCB and 1.7% of ABC DLBCL.30

be cured with R-CHOP and decid­ing if alter­na­tive approaches
should be used in cer­tain patient sub­sets. In sev­eral stud­ies over
a long time span, the IPI has been a reli­able pre­dic­tor of out­
come; the prog­nos­tic value of base­line tumor bio­log­i­cal fac­tors
is much less well established and remains very con­tro­ver­sial.6
Different groups have reported out­comes and asso­ci­a­tions with
tumor bio­log­i­cal char­ac­ter­is­tics that vary con­sid­er­ably, par­tic­u­
larly when com­par­ing pro­spec­tive and ret­ro­spec­tive data. It is
well established that most DLBCL cases are of ger­mi­nal cen­ter
B-cell (GCB) or acti­vated B-cell (ABC) ori­gin, and many stud­ies
have dem­on­strated an infe­rior out­come for the lat­ter group. In
attempts to improve out­come in this sub­set, efforts have been
under­way to add agents to R-CHOP that have selec­tive activ­
ity in ABC-DLBCL. However, 2 recently reported large ran­dom­
ized stud­ies did not show a ben­e­fit to adding lenalidomide or
ibrutinib in ABC- or non-GCB-DLBCL.7-9 Since the con­ cep­ tion
of these tri­als, the cat­e­go­ri­za­tion of DLBCL has under­gone fur­
ther refine­ment and is focused on more accu­rately iden­ti­fy­ing
prognostically mean­ing­ful genetic driv­ers of tumor­i­gen­e­sis,
which paves the way for novel and more pre­cise therapeutic
approaches.1-3 As well as cell-of-ori­gin-focused stud­ies, the rec­
og­ni­tion that aber­rant MYC and BCL2 expres­sion is asso­ci­ated
with infe­rior out­comes has led to inves­ti­ga­tions aimed at uncov-
ering the mech­an ­ is­tic basis for MYC and BCL2 overexpression.
In par­al­lel with this, fol­low­ing improved out­comes with dose-
intense approaches in ret­ro­spec­tive com­par­i­sons, alter­na­tives
to R-CHOP are being inves­ti­gated in these sub­sets. Coming back
to our patient, based on his high IPI score and “dou­ble-hit” sta­
tus, his predicted cur­abil­ity rate with R-CHOP is low. While this is
clear from ret­ro­spec­tive data, the negative prognostic impact of
double-hit is not as well established in prospective experiences,
and the opti­mal man­age­ment of cases such as these remains
con­tro­ver­sial. The goal of this review is to explore this ques­tion
in the con­text of emerg­ing bio­log­i­cal insights and novel clin­i­cal
data for this sub­set of patients.
HGBL cases are now con­sid­ered a dis­tinct entity in the 2016
World Health Organization Lymphoid Tumor Classification.10 They
encom­pass a sub­set of aggres­sive cases that, despite overlap-
ping clin­i­cal and path­o­log­i­cal char­ac­ter­is­tics, are dif­fer­ent from
the par­ ent enti­ ties of DLBCL and BL (Figure 1). The cat­ e­gory
includes lym­pho­mas that were pre­vi­ously named “Burkitt-like”
and “high grade” as well as “dou­ble-hit” or “tri­ple-hit.” In addi­
tion to the major cat­e­gory “HGBL with MYC and BCL2 and/or
BCL6 trans­lo­ca­tions,” there is a sub­set called “HGBL-not oth­er­
wise spec­ifi­ ed” (NOS). Most cases with a sin­gle MYC rearrange-
ment fall under DLBCL, as do most cases with high MYC and BCL2
expres­sion but with­out rearrangements. The updated cat­e­go­ri­
za­tion is help­ful in the clinic, as HGBL behaves more aggres­sively
than DLBCL and likely requires dis­tinct ther­a­peu­tic approaches.11
Though dou­ble-hit/tri­ple-hit lym­phoma (DHL/THL) and double
protein expresser (DPE) cases have MYC and BCL2 aber­ra­tions
in com­mon, they are dis­tinct in terms of their lymphomagenesis
because DLH/THL is mostly GCB derived, while DPE cases are
prin­ci­pally of ABC ori­gin (Figure 2).
Recent advances in under­stand­ing HGBL cases
Recent stud­ies have set out to bet­ter under­stand the bio­log­i­
cal under­pin­nings of HGBL cases and elu­ci­date the spe­cific
char­ac­ter­is­tics asso­ci­ated with infe­rior out­comes. While FISH test­
ing is cur­rently the gold stan­dard to iden­tify HGBL with MYC and

158  |  Hematology 2021  |  ASH Education Program


Figure 2. Categories of double-expresser lymphomas. These are typically cases that have a high-protein expression of MYC and
BCL2. Most DPE cases that are associated with rearrangements of MYC and BCL2 are of GCB origin, whereas most cases that do not
harbor these rearrangements are of ABC origin. The proportion of DLBCL cases that are double expressers has been calculated at
between 21% and 44% across various studies.31
BCL2 and/or BCL6 rearrangements, emerg­ing data from more
com­plex geno­mic stud­ies sug­gest that FISH has sen­si­tiv­ity lim­i­
ta­tions com­pared to tech­niques such as whole-exome sequenc­
ing.12 Therefore, an impor­tant clin­i­cally appli­ca­ble ques­tion arises:
Does FISH ade­quately iden­tify—within GCB-DLBCL and HGBL—
cases with aber­ra­tions of MYC/BCL2 and/or BCL6 that por­
tend an infe­rior out­come? Probably not, and many cases with
crit­i­cal DHL/THL aber­ra­tions are not iden­ti­fied by FISH alone.
A dou­ble-hit gene expres­sion sig­na­ture (DHIT-sig) was recently
pro­posed based on the anal­y­sis of RNA sequenc­ing data from
157 cases of GCB-DLBCL (includ­ing HGBL) treated with R-CHOP.4
This was a 104 gene sig­na­ture that represented/dis­tin­guished
27% of GCB-DLBCLs—and while the major­ity of GCB-DLBCLs
with high-grade mor­ phol­
ogy had this sig­ na­ture, only half of
the DHIT-sig+ cases har­bored con­cur­rent MYC and BCL2 rear-
rangements.4 DHIT-sig+ cases had a sig­nif­i­cantly infe­rior time to
dis­ease pro­gres­sion com­pared to DHIT− cases. Another recent
study dem­on­strated that DHIT-sig+ cases with con­com­i­tant
TP53 abnor­mal­i­ties had a par­tic­u­larly poor out­come.13 It is inter­
est­ing to con­sider HGBL cases in the con­text of newer molec­
u­lar clas­si­fi­ca­tions of DLBCL and pro­pose where they may lie
within newly defined, more poten­tially tar­get­able, sub­groups.1-3
In 1 recent study that com­pre­hen­sively genet­i­cally ana­lyzed 304
DLBCL cases and defined 5 dis­tinct sub­sets (clus­ters 1-5), tumors
with co-occur­ring BCL2 and MYC struc­tural var­i­ants were sig­nif­i­
cantly more fre­quent in clus­ter 3, a GCB-defined sig­na­ture.1
The influ­ence of a rearrangement of MYC is likely affected
by the MYC part­ner gene, whether it is an immu­no­glob­u­lin (IG)
or a non-IG gene.14 In Bur­kitt lym­phoma, the part­ner of MYC is
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/157/1852000/157dunleavy.pdf by guest on 13 December 2021
an IG gene in almost all­cases. In con­trast, approx­i­ma­tely 50%
of HGBL-DHL/THLs have a non-IG part­ner. A recent study dem­
on­strated infe­rior out­comes for MYC DHL/THL cases with MYC
translocated to an IG part­ ner vs a non-IG part­ner fol­low­ing
treat­ment with R-CHOP.14 In interpreting the results of stud­ies
looking at the prog­nos­tic impact of DHL/THL and sin­gle-hit lym­
phoma (SHL) MYC aber­ra­tions, it is impor­tant to con­sider recent
geno­mic stud­ies that have elu­ci­dated novel genetic sub­types of
DLBCL.1-4 Within GCB tumors, newly char­ac­ter­ized sub­sets with
dis­tinct clin­i­cal out­comes exhibit the co-occur­rence of spe­cific
genetic alter­ations. There is likely sig­nif­i­cant over­lap between
aber­ra­tions in these sub­sets and DHIT-sig+ cases that require
future inves­ti­ga­tion.
Approach to HGBL with MYC and BCL2 and/or BCL6
rearrangements
Currently, no widely accepted stan­dard approach exists to the
ini­tial man­age­ment of these cases. Clinical pre­sen­ta­tion dif­fer­
ences between DLBCL and HGBL have not been well defined,
but early stud­ies dem­on­strated fre­quent extranodal involve­ment
and a higher rate of cen­tral ner­vous sys­tem (CNS) dis­ease in the
lat­ter entity. Multiple ret­ro­spec­tive and obser­va­tional stud­ies
have dem­on­strated that fol­low­ing R-CHOP treat­ment, sur­vival
is sig­nif­i­cantly infe­rior com­pared to patients who do not har­bor
these aber­ra­tions. Early ret­ro­spec­tive stud­ies dem­on­strated a
par­tic­u­larly adverse out­come for this sub­set that is less strik­ing
in some recent stud­ies. This may be partly explained by a his­
tor­i­cal selec­tion bias in apply­ing FISH test­ing to select patients
with more aggres­sive clin­i­cal pre­sen­ta­tions vs the more recent
New devel­op­ments in aggres­sive B-cell lym­phoma  |  159
stan­dard of apply­ing it to the major­ity of new cases. A recent
large-scale ret­ro­spec­tive anal­y­sis of DLBCL patient out­comes fol­
low­ing R-CHOP—from pro­spec­tive tri­als and patient reg­is­tries—
reported inter­est­ing find­ings.14 While the anal­y­sis showed that
patients with a MYC rearrangement had a sig­nif­i­cantly shorter
pro­gres­sion-free sur­vival (PFS) and OS, the adverse impact of the
rearrangement was con­fined to those har­bor­ing a con­cur­rent
BCL2 and/or BCL6 rearrangement and an IG vs non-IG part­ner
with MYC. It is impor­tant to note that this study only included
cases with DLBCL mor­phol­ogy, and this may have con­trib­uted
to bet­ter out­comes com­pared to his­tor­i­cal expe­ri­ences in which
blastoid and Burkitt morphologies were included.
Should approaches beyond R-CHOP be con­ sid­ered for
mul­ti­cen­ter study of 53 patients with MYC rearrangement
(MYC-R) aggres­ sive B-cell lym­phoma.15 More cases were DHL
than SHL-MYC-R. The major­ity of patients had advanced-stage
dis­ease (81%), and 48-month event-free sur­vival (EFS) and OS
were 71% and 77%, respec­tively. A phase 2 HOVON trial added
lenalidomide to R-CHOP in 82 patients, hypoth­e­siz­ing that
because lenalidomide downregulates MYC and its tar­get genes
it may be ther­a­peu­ti­cally advan­ta­geous in MYC-R lym­phoma.15
Sixty-five per­cent of cases had a DHL or THL, and 2-year EFS and
OS were 63% and 73%, respec­tively. A recent Brit­ish study eval­
u­ated patients with high-risk DLBCL using cyclo­phos­pha­mide,
vin­cris­tine, doxo­ru­bi­cin, high-dose meth­o­trex­ate, ifosfamide,
etoposide, and high-dose cytarabine (R-CODOX-M/R-IVAC).16 All

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/157/1852000/157dunleavy.pdf by guest on 13 December 2021

HGBCL? Two recent pro­spec­tive stud­ies of DLBCL patients
har­bor­ing a MYC rearrangement (with a high pro­por­tion hav­
ing DHL) were reported. Based on ret­ro­spec­tive com­par­i­sons
show­ing that DHL cases did bet­ter with more dose-inten­sive
approaches com­pared to R-CHOP, dose-adjusted etoposide,
pred­ni­sone, vin­cris­tine, cyclo­phos­pha­mide, hydroxydaunoru-
bicin, and rituximab (DA-EPOCH-R) was tested in a pro­spec­tive,
patients had an IPI score of 3 or higher. Most were GCB, and
over 10% had DHL, suggesting that a high pro­por­tion may have
had HGBL. Two-year PFS was 67.9%, and 2-year OS was 76%, and
while not directly com­pared to R-CHOP, the results were supe­rior
to his­tor­i­cal expe­ri­ences using R-CHOP in a sim­i­lar pop­u­la­tion.
Leppa and col­leagues inves­ti­gated dose-dense immunochemo-
therapy in 139 patients with high-risk DLBCL and dem­on­strated a

Figure 3. Outline for the workup and management of aggressive B-cell lymphomas. Typically, morphological, immunohistochemis-
try, and FISH analysis are performed to differentiate DLBCL from HGBL. HGBL cases are divided into those that are DHL/THL or NOS.
For DLBCL cases that have a high-protein expression of MYC and BCL2, which are usually of ABC origin, dose-intensive therapy or
enrollment in a clinical trial should be considered, particularly for patients with a high IPI score.

160  |  Hematology 2021  |  ASH Education Program

5-year OS rate of 83%.17 The out­come of patients with BCL2/MYC
DHL was sim­i­lar to patients with­out rearrangements, suggesting
a ben­e­fit from dose-intense ther­apy in the DHL group. Also, a
large, recently published French ret­ro­spec­tive study eval­u­ated
160 patients with HGBL (with MYC and BCL2 and/or BCL6) and
dem­on­strated a sig­nif­i­cantly lon­ger PFS for inten­sive ther­apy vs
R-CHOP.18 HGBL-NOS cases are much rarer, and hence, there are
a pau­city of data to inform on opti­mal treat­ment.
In selecting ther­apy for patients with HGBL, the stage of dis­
ease and IPI char­ac­ter­is­tics may be impor­tant (Figure 3). Retro-
spective expe­ri­ences have dem­on­strated good out­comes for
patients with lim­ited-stage aggres­sive B-cell lym­phoma despite
high-risk cyto­ge­net­ics.19 It may be rea­son­able to approach
stem cell trans­plan­ta­tion. Unfortunately, 6 weeks fol­low­ing
the trans­ plant, he had fur­ther pro­ gres­
sive dis­ease. He then
went on to receive anti-CD19 chi­me­ric anti­gen recep­tor (CAR)
T-cell ther­apy (axicabtagene ciloleucel) and had a com­plete
response. He was still in remis­sion 6 months fol­low­ing the com­
ple­tion of CAR T cells.
Approach to relapsed/refrac­tory HGBL
Given the rar­ity of these tumors, it is unknown if relapsed/
refrac­tory HGBL should be approached dif­fer­ently to DLBCL with­
out high-risk cyto­ge­net­ics. Some ret­ro­spec­tive stud­ies have
shown that patients with these dis­ eases (com­pared to other

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/157/1852000/157dunleavy.pdf by guest on 13 December 2021

these early-stage patients with stan­dard R-CHOP. For higher- aggres­sive B-cell lym­pho­mas) fare more poorly fol­low­ing autol­o­
stage and high-IPI patients, more inten­sive immunochemother- gous stem cell trans­plan­ta­tion.22 This may be partly explained by
apy approaches are rea­son­able to con­sider, under­stand­ing that the fact that HGBL cases are more likely to receive more inten­
there are a pau­city of robust com­par­i­son data (Table 1). HGBL sive immunochemotherapy than R-CHOP in the up-front set­ting.
tumors are not uncom­monly encoun­tered in the set­ting of HIV Recently published and ongo­ing stud­ies looking at approaches
infec­tion, and based on equiv­ a­lent out­
comes to HIV− cases such as targeting CD19 in the relapsed/refrac­tory set­ting have,
in recent pro­ spec­tive stud­ ies (where up to 25% of patients inter­est­ingly, shown that HGBL biol­ogy is not asso­ci­ated with a
accrued were HIV+), they should not be approached dif­fer­ently. worse out­come.23,24 Considering that this has not been the expe­
ri­ence with autol­o­gous stem cell trans­plan­ta­tion, it sug­gests a
Approach to DPE lym­pho­mas poten­tial ben­e­fit to mov­ing these new treat­ment modal­i­ties to the
MYC, BCL2, and BCL6 are overexpressed by sev­ eral mech­
a­ front­line set­ting.
nisms other than gene rearrangements, and a high pro­por­tion of
aggres­sive B-cell lym­phoma cases have high-pro­tein expres­sion
Promising new approaches
of MYC and BCL2 and/or BCL6. These DPE cases are asso­ci­ated
Many prom­is­ing approaches are in devel­op­ment for this group
with an infe­rior out­come fol­low­ing stan­dard ther­apy—when lack-
of dis­eases. First, strat­e­gies that incor­po­rate targeting BCL2 and
ing rearrangements, they are typ­i­cally of ABC ori­gin. How best to
MYC are under inves­ti­ga­tion in tri­als.25,26 Undoubtedly, the con­
approach them ther­a­peu­ti­cally is unclear. There is no evi­dence
cur­rent high expres­sion of MYC and BCL2, irrespective of path­o­
from ret­ro­spec­tive stud­ies that intensive therapy approaches
gen­e­sis, is asso­ci­ated with a higher risk of treat­ment fail­ure, but
are superior. These patients should there­fore be con­sid­ered for
the indi­vid­ual con­tri­bu­tions of these path­ways, in terms of con­
enroll­ment in clin­i­cal tri­als, where agents and approaches that
fer­ring resis­tance, are not well under­stood. Venetoclax, a highly
tar­get the key path­o­ge­netic mech­a­nisms underpinning MYC and
selec­tive inhib­i­tor of BCL2, has activ­ity in sev­eral lym­pho­mas and
BCL2 acti­va­tion are being investigated.
was recently com­bined with R-CHOP che­mo­ther­apy in a front­line
phase 2 study for patients with DLBCL (phase 2 CAVALI study).27
Role of CNS pro­phy­laxis in HGBL
While this com­ bi­
na­tion dem­ on­strated good activ­ ity and the
As is the case in DLBCL, the role of CNS pro­phy­laxis in HGBL is
poten­tial to improve out­come in a BCL2 immunohistochemistry
con­tro­ver­sial, and unfor­tu­nately, pro­spec­tive stud­ies that could
sub­group, it was asso­ci­ated with increased myelotoxicity com­
poten­tially clar­ify this ques­tion are lacking.20,21 From ret­ro­spec­
pared to R-CHOP alone. Currently, a ran­ dom­ ized pro­ spec­tive
tive data, the inci­dence of CNS relapse in early-stage or low-IPI
study of immunochemotherapy with or with­ out venetoclax is
cases is very low, suggesting that CNS pro­phy­laxis may have very
ongo­ing (NCT03984448). Inhibitors of MYC are also in devel­op­
min­i­mal poten­tial ben­e­fit.19 However, sev­eral series have now
ment; small-mol­e­cule inhib­i­tors of the bromodomain and extra-
dem­on­strated high CNS relapse rates in advanced-stage and
terminal domain pro­teins are also inter­est­ing with respect to MYC.
high-IPI cases, and some ret­ro­spec­tive sin­gle-arm com­par­i­son
Epignentic inhibitors such as those targeting histone deacetylase
expe­ri­ences sug­gest that pro­phy­laxis may dimin­ish CNS relapse
(HDAC)3 and EZH2 inhibitors are under evaluation in pre-clinical
inci­dence.17 It is very chal­leng­ing to con­duct a large-scale pro­
studies. Recently, up-front stud­ies have started to incor­ po­rate
spec­tive trial to prop­erly address this ques­tion and reli­ably guide
novel strat­e­gies such as anti-CD19 CAR T cells in treat­ments for
clin­i­cal prac­tice. Until that is done, how to approach CNS pro­
high-risk DLBCL and HGBCL patients who do not have early com­
phy­laxis will remain con­tro­ver­sial, with a lack of con­sen­sus. Given
plete responses.28
our patient’s stage IV dis­ease with bone mar­row infil­tra­tion and
DHL sta­tus and the high poten­tial for CNS spread, we decided to
insti­tute intra­the­cal CNS pro­phy­laxis. Conclusions
HGBLs are a huge ther­a­peu­tic chal­lenge, and their opti­mal man­
age­ment remains unde­fined at this time. It is crit­i­cal to con­
tinue to make inroads in under­stand­ing their biol­ogy and how
that inter­acts and over­laps with other key genetic and func­
CLINICAL CASE (Con­t in­u ed) tional driv­ers of lymphomagenesis. In that regard, recent work
Following con­f ir­m a­t ion of refrac­tory/relapsed dis­ease, the such as the iden­ti­fi­ca­tion of new prog­nos­tic sig­na­tures such as
patient went on to receive rituximab, ifosfamide, carboplatin, the DHIT-sig is wel­comed; addi­tion­ally, the def­i­ni­tion of novel,
and etoposide (R-ICE) che­mo­ther­aPrakash
py followedSingh Shekhawat
by autol­
o­gous poten­tially more action­able, DLBCL sub­groups is help­ful in the

New devel­op­ments in aggres­sive B-cell lym­phoma  |  161

Table 1. Select recent stud­ies in aggres­sive B-cell lym­phoma looking at (dif­fer­en­tial) out­comes of patients with HGBL
and DHL/THL

Study N Patient pop­u­la­tion/study DHL/THL % Treatment Outcome

Rosenwald 2383: (MYC-R DLBCL and HGBL/ 5.8% R-CHOP MYC-R was asso­ci­ated with
et al14 in 11%) ret­ro­spec­tive anal­y­sis of shorted PFS and OS; neg.
pro­spec­tive and patient prog­nos­tic impact of MYC-R
reg­is­try stud­ies only with BCL2 and/or
BCL6 and an IG part­ner.
Dunleavy 53 MYC-R and aggres­sive B-cell Approx 44%* had MYC-R DA-EPOCH-R 4-year EFS and OS were 71%
et al15 lym­phoma/pro­spec­tive, (SH); 56% had DHL/THL and 77%. No dif­fer­ence for
sin­gle-arm, mul­ti­cen­ter SH vs DHL/THL.
trial

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/157/1852000/157dunleavy.pdf by guest on 13 December 2021

Chamuleau 82 MYC-R DLBCL/pro­spec­tive, Approx 27%* had MYC-R R-CHOP + 2-year EFS and OS were 63%
et al29 sin­gle-arm mul­ti­cen­ter trial (SH); 73% had DHL/THL lenalidomide and 73%.
Leppä 139 DLBCL and high-IPI 12% had DHL Dose-dense chemo 5-year FFS and OS were 74%
et al17 score/high-risk cohort/ (MTX/R-CHOEP-14, and 83%. No sig­nif­i­cant
pro­spec­tive, sin­gle-arm, ARA C worse out­come for DHL
mul­ti­cen­ter trial group.
McMillan 111 DLBCL and IPI 3-5; 12% had 12% had DHL; FISH R-CODOX-M/R-IVAC 2-year PFS and OS were 68%
et al16 HGBL/pro­spec­tive study. performed in approx. 50% and 76%. No worse out­
come for DHL.
Laude 160 All patients had HGBL/ret­ro­ 81% had DHL; 19% had THL R-CHOP vs inten­sive At 32 months, 2 and 4-year
et al18 spec­tive study che­mo­ther­apy PFS were 40% and 28% for
R-CHOP; 57% and 52% for
inten­sive ther­apy.
Of cases tested.
ARA C, cytarabine; CHOEP, CHOP with etoposide; FFS, fail­ure-free sur­vival; MTX, meth­o­trex­ate.

quest to bet­ ter under­ stand the molec­ u­lar basis of MYC and
BCL2 dysregulation. It is clear that stan­dard R-CHOP treat­ment
remains inad­e­quate for an unac­cept­ably high pro­por­tion of
cases; a sub­set of these cases may ben­e­fit from dose-inten­sive
approaches, but this needs bet­ter inves­ti­ga­tion in, ide­ally, ran­
dom­ized pro­spec­tive tri­als. The chal­lenge is that these dis­eases
are rare, and until these trial results are avail­­able, stan­dards can
only be based on ret­ro­spec­tive/obser­va­tional expe­ri­ences and
sin­gle-arm pro­spec­tive stud­ies. In the relapsed and refrac­tory
set­ting, HGBL cases do not fare worse than DLBCL cases fol­low­
ing novel strat­e­gies, as discussed ear­lier. This sug­gests a role
for these approaches in ear­lier lines of ther­apy for HGBL. Addi-
tionally, many prom­is­ing novel agents are under inves­ti­ga­tion for
these dis­eases. Until we aug­ment cure rates sig­nif­i­cantly for this
patient pop­u­la­tion with a widely avail­­able strat­egy, it is a pri­or­ity
to con­sider refer­ral of HGBL patients to prom­is­ing clin­i­cal tri­als.
Conflict-of-inter­est dis­clo­sure
Kieron Dunleavy: advi­sory board mem­ber: Genmab, Morphosys,
Beigene, Daiichi Sankyo, Abbvie, ADC Therapeutics, Incyte, Astra
Zeneca, Genetech, Janssen.
Off-label drug use
Kieron Dunleavy: nothing to disclose.
Correspondence
Kieron Dunleavy, Division of Hematology and Oncology, George-
town Lombardi Comprehensive Cancer Center, Georgetown
University Hospital, 3800 Reservoir Rd NW, Washington, DC
20057; e-mail: kmd322@georgetown.edu.
References
1. Chapuy B, Stewart C, Dunford AJ, et al. Molecular sub­types of dif­fuse large
B cell lym­phoma are asso­ci­ated with dis­tinct path­o­genic mech­a­nisms and
out­comes. Nat Med. 2018;24(5):679-690.
2. Reddy A, Zhang J, Davis NS, et al. Genetic and func­tional driv­ers of dif­fuse
large B cell lym­phoma. Cell. 2017;171(2):481-494.e15494e15.
3. Schmitz R, Wright GW, Huang DW, et al. Genetics and path­o­gen­e­sis of dif­
fuse large B-cell lym­phoma. N Engl J Med. 2018;378(15):1396-1407.
4. Ennishi D, Jiang A, Boyle M, et  al. Double-hit gene expres­sion sig­na­ture
defi­nes a dis­ tinct sub­group of ger­ mi­
nal cen­
ter B-cell-like dif­
fuse large
B-cell lym­phoma.J Clin Oncol. 2019;37(3):190-201.
5. Lap CJ, Nassereddine S, Dunleavy K. Novel bio­log­i­cal insights and new
devel­op­ments in man­age­ment of Bur­kitt lym­phoma and high-grade B-cell
lym­phoma. Curr Treat Options Oncol. 2021;22(7):60.
6. Frontzek F, Ziepert M, Nickelsen M, et  al. Rituximab plus high-dose che­
mo­ther­apy (MegaCHOEP) or con­ven­tional che­mo­ther­apy (CHOEP-14) in
young, high-risk patients with aggres­sive B-cell lym­phoma: 10-year fol­low-
up of a randomised, open-label, phase 3 trial. Lancet Haematol. 2021;8(4):
e267-e277.
7. Younes A, Sehn LH, Johnson P, et al; PHOENIX Investigators. Randomized
phase III trial of ibrutinib and rituximab plus cyclo­phos­pha­mide, doxo­ru­bi­
cin, vin­cris­tine, and pred­ni­sone in non-ger­mi­nal cen­ter B-cell dif­fuse large
B-cell lym­phoma. J Clin Oncol. 2019;37(15):1285-1295.
8. Nowakowski GS, Chiappella A, Gascoyne RD, et  al. ROBUST: a phase III
study of lenalidomide plus R-CHOP ver­sus pla­cebo plus R-CHOP in pre­vi­
ously untreated patients with ABC-type dif­fuse large B-cell lym­phoma. J Clin
Oncol. 2021;39(12):1317-1328.
9. Nowakowski GS, Hong F, Scott DW, et  al. Addition of lenalidomide to
R-CHOP improves out­comes in newly diag­nosed dif­fuse large B-cell lym­
phoma in a ran­dom­ized phase II US inter­group study ECOG-ACRIN E1412.
J Clin Oncol. 2021;39(12):1329-1338.
10. Ott G. Aggressive B-cell lym­pho­mas in the update of the 4th edi­tion of the
World Health Organization clas­si­fi­ca­tion of haematopoietic and lym­phatic
tis­sues: refine­ments of the clas­si­fi­ca­tion, new enti­ties and genetic find­ings.
Br J Haematol. 2017;178(6):871-887.
11. Alsharif R, Dunleavy K. Bur­kitt lym­phoma and other high-grade B-cell lym­
pho­mas with or with­out MYC, BCL2, and/or BCL6 rearrangements. Hema-
tol Oncol Clin North Am. 2019;33(4):587-596.

162  |  Hematology 2021  |  ASH Education Program

12. Hilton LK, Tang J, Ben-Neriah S, et al. The dou­ble-hit sig­na­ture identifies
dou­ble-hit dif­fuse large B-cell lym­phoma with genetic events cryp­tic to
FISH. Blood. 2019;134(18):1528-1532.
13. Song JY, Perry AM, Herrera AF, et al. Double-hit sig­na­ture with TP53 abnor­
mal­i­ties pre­dicts poor sur­vival in patients with ger­mi­nal cen­ter type dif­
fuse large B-cell lym­phoma treated with R-CHOP. Clin Cancer Res. 2021;
27(6):1671-1680.
14. Rosenwald A, Bens S, Advani R, et al. Prognostic sig­nif­i­cance of MYC rear-
rangement and trans­lo­ca­tion part­ner in dif­fuse large B-cell lym­phoma: a study
by the Lunenburg Lymphoma Biomarker Consortium. J Clin Oncol. 2019;37
(35):3359-3368.
15. Dunleavy K, Fanale MA, Abramson JS, et  al. Dose-adjusted EPOCH-R
(etoposide, pred­ni­sone, vin­cris­tine, cyclo­phos­pha­mide, doxo­ru­bi­cin, and
rituximab)in untreated aggres­sive dif­fuse large B-cell lym­phoma with MYC
rearrangement: a pro­spec­tive, multicentre, sin­gle-arm phase 2 study.
Lancet Haematol. 2018;5(12):e609-e617.
16. McMillan AK, Phil­lips EH, Kirkwood AA, et al. Favourable out­comes for high-
risk dif­fuse large B-cell lym­phoma (IPI 3-5) treated with front-line R-CODOX-
M/R-IVAC che­mo­ther­apy: results of a phase 2 UK NCRI trial. Ann Oncol.
2020;31(9):1251-1259.
17. Leppä S, Jørgensen J, Tierens A, et al. Patients with high-risk DLBCL ben­e­
fit from dose-dense immunochemotherapy com­bined with early sys­temic
CNS pro­phy­laxis. Blood Adv. 2020;4(9):1906-1915.
18. Laude MC, Lebras L, Sesques P, et al. First-line treat­ment of dou­ble-hit and
tri­ple-hit lym­pho­mas: sur­vival and tol­er­ance data from a ret­ro­spec­tive
mul­ti­cen­ter French study. Am J Hematol. 2021;96(3):302-311.
19. Torka P, Kothari SK, Sundaram S, et al. Outcomes of patients with lim­ited-
stage aggres­sive large B-cell lym­phoma with high-risk cyto­ge­net­ics. Blood
Adv. 2020;4(2):253-262.
20. Wilson MR, Eyre TA, Martinez-Calle N, et  al. Timing of high-dose meth­o­
trex­ate CNS pro­phy­laxis in DLBCL: an anal­y­sis of tox­ic­ity and impact on
R-CHOP deliv­ery. Blood Adv. 2020;4(15):3586-3593.
21. Chin CK, Cheah CY. How I treat patients with aggres­sive lym­phoma at high
risk of CNS relapse. Blood. 2017;130(7):867-874.
22. Herrera AF, Mei M, Low L, et al. Relapsed or refrac­tory dou­ble-expressor
and dou­ble-hit lym­pho­mas have infe­rior pro­gres­sion-free sur­vival after
autol­o­gous stem-cell trans­plan­ta­tion. J Clin Oncol. 2017;35(1):24-31.
Prakash Singh Shekhawat
23. Abramson JS, Palomba ML, Gordon LI, et al. Lisocabtagene maraleucel for
patients with relapsed or refrac­tory large B-cell lym­pho­mas (TRANSCEND
NHL 001): a multicentre seam­less design study. Lancet. 2020;396(10254):
839-852.
24. Caimi PF, Ai W, Alderuccio JP, et  al. Loncastuximab tesirine in relapsed or
refrac­tory dif­fuse large B-cell lym­phoma (LOTIS-2): a multicentre, open-label,
sin­gle-arm, phase 2 trial. Lancet Oncol. 2021;22(6):790-800.
25. Deng M, Xu-Monette ZY, Pham LV, et al. Aggressive B-cell lym­phoma with
MYC/TP53 dual alter­ations dis­plays dis­tinct clinicopathobiological fea­tures
and response to novel targeted agents. Mol Cancer Res. 2021;19(2):249-
260.
26. Fiskus W, Mill CP, Perera D, et al. BET pro­te­­ol­y­sis targeted chi­mera-based
ther­apy of novel mod­els of Richter trans­for­ma­tion-dif­fuse large B-cell lym­
phoma. Leukemia. 2021;35(9):2621-2634.
27. Morschhauser F, Feugier P, Flinn IW, et al. A phase 2 study of venetoclax
plus R-CHOP as first-line treat­ment for patients with dif­fuse large B-cell
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/157/1852000/157dunleavy.pdf by guest on 13 December 2021
lym­phoma. Blood. 2021;137(5):600-609.
28. Neelapu S, Dickinson M, Ulrickson M, et  al. Interim anal­y­sis of ZUMA-12:
a phase 2 study of axicabtagene ciloleucel (Axi-Cel) as first-line ther­apy
in patients with high risk large B-cell lym­phoma. Abstract presented at:
62nd Amer­i­can Society of Hematology Annual Meeting and Exposition; 5-8
Decem­ber 2020; Atlanta, GA. Abstract 405.
29. Chamuleau MED, Burggraaff CN, Nijland M, et al. Treatment of patients with
MYC rearrangement pos­ i­
tive large B-cell lym­phoma with R-CHOP plus
lenalidomide: results of a mul­ti­cen­ter HOVON phase II trial. Haematolog-
ica. 2020;105(12):2805-2812.
30. Scott DW, King RL, Staiger AM, et  al. High-grade B-cell lym­phoma with
MYC and BCL2 and/or BCL6 rearrangements with dif­fuse large B-cell lym­
phoma mor­phol­ogy. Blood. 2018;131(18):2060-2064.
31. Dunleavy K. Double-hit lym­pho­mas: cur­rent par­a­digms and novel treat­ment
approaches. Hematology Am Soc Hematol Educ Program. 2014;2014(1):
107-112.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000247
New devel­op­ments in aggres­sive B-cell lym­phoma  |  163
 DEFEATING DIFFUSE, DOUBLE - HIT, AND DOGGED NON - HODGKIN LYMPHOMA

Relapsed disease: off-the-shelf immunotherapies

vs customized engineered products

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/164/1851658/164karmali.pdf by guest on 13 December 2021

Reem Karmali
Northwestern University Feinberg School of Medicine, Chicago, IL

Innovations in immuno-oncology for lymphomas have outpaced therapeutic developments in any other cancer histology.
In the 1990s, rituximab, a CD20 monoclonal antibody, drastically changed treatment paradigms for B-cell non-Hodgkin
lymphomas (B-NHLs). In parallel, the concept that T cells could be genetically reprogrammed and regulated to address
tumor cell evasion was developed. Twenty years later, this concept has materialized—3 customized engineered CD19 chi-
meric antigen receptor T-cell (CART) constructs have been embraced as third-line therapies and beyond for aggressive
B-NHL. Responses with CARTs are durable in 30% to 40% of patients, with consistent results in older patients, primary
refractory disease, high-grade B-cell lymphoma, and patients with concurrent secondary central nervous system dis-
ease, all features historically associated with poorer outcomes. Challenges associated with the administration of CARTs
include cumbersome and time-consuming manufacturing processes, toxicities, and cost, not to mention a substantial
risk of relapse. Fortunately, as our understanding of how to manipulate the immune system to achieve full antitumor
potential has grown, so has the rapid development of off-the-shelf immunotherapies, with CD20/CD3 bispecific antibod-
ies standing out above all others. These agents have shown promising activity in aggressive B-NHL and have the potential
to circumvent some of the challenges encountered with customized engineered products. However, toxicities remain
substantial, dosing schedules intensive, and experience limited with these agents. Novel customized and off-the-shelf
therapeutics as well as rational combinations of these agents are underway. Ultimately, growing experience with both
customized engineered and off-the-shelf immunotherapies will provide guidance on optimal methods of delivery and
sequencing.

LEARNING OBJECTIVES
• Understand the strengths and limitations of CD19 CARTs as a customized engineered product vs off-the-shelf
immunotherapies such as bispecifc CD20/CD3 antibodies for the treatment of aggressive B-NHL
• Review clinical effcacy and safety data for CD19 CARTs and bispecifc CD20/CD3 antibodies
• Optimize a therapeutic algorithm for relapsed/refractory aggressive B-cell lymphoma with the inclusion of CARTs
and off-the-shelf immunotherapies

CLINICAL CASE: PART 1
A 65-year-old male presented with lower back and flank
pain, fevers, and weight loss. Magnetic resonance imag-
ing of the lumbar spine showed a paraspinal mass. Posi-
tron emission tomography-computed tomography scans
showed diffuse lymphadenopathy with bone marrow
involvement and highest uptake in bulky retroperitoneal
lymph nodes and the paraspinal mass. A core biopsy of
the paraspinal mass confrmed high-grade B-cell lym-
phoma with dual rearrangements of MYC and BCL2 (also
known as double-hit lymphoma). The patient was treated
with six cycles of DA-EPOCH-R and achieved a complete
metabolic response at the completion of therapy. Twelve
months later, the patient relapsed. He was treated with
two cycles of R-ICE with complete response (CR) and
consolidated with an autologous stem cell transplanta-
tion (ASCT). Unfortunately, scans 3 months post-ASCT
demonstrated disease recurrence. He was referred to our
institution to discuss treatment options for his second
relapse.
Introduction
To date, salvage high-dose chemotherapy with ASCT re-
mains the standard second-line treatment for relapsed or
refractory (R/R) diffuse large B-cell lymphoma (DLBCL)

164 | Hematology 2021 | ASH Education Program

regard­less of under­ly­ing high-risk bio­logic fea­tures.1 Howev-
er, few patients are cured with this inten­sive approach, and
appli­ca­bil­ity is lim­ited by comorbidities, advanced age, and/or
che­mo­ther­apy-insen­si­tive dis­ease.2,3 In the era predating the
use of immunotherapies, patients with refrac­tory dis­ease or
relapse within 12 months of ASCT had dis­mal out­comes. In the
SCHOLAR-1 mul­ti­cen­ter ret­ro­spec­tive study, the objec­tive re-
sponse rate (ORR) to the next line of ther­apy was 26% (CR, 7%),
with a median over­all sur­vival (OS) rate of 6.3 months in such
patients.2
Fortunately, the treat­ment land­scape has rap­idly evolved for
R/R DLBCL, with cus­tom­ized engineered immunotherapies—more
spe­cif­i­cally, CD19 chi­me­ric anti­gen recep­tor T cells (CARTs)—and
syn­drome (ICANS). Restaging scans 30 days and 90 days post-
CART dem­on­strated a com­plete met­a­bolic response.
Limitations of CARTs and the emer­gence of off-the-shelf
immunotherapies
Although CARTs have changed the treat­ment par­a­digm for R/R
aggres­sive B-cell lym­pho­mas, the ther­a­peu­tic has its lim­i­ta­tions.
First, CARTs have to be engineered for each indi­vid­ual patient,
with a poten­tial for logis­ti­cal delays from the time of patient
iden­ti­fi­ca­tion to CART infu­sion as well as a risk of manufactur­ing
fail­ure. Second, sig­nif­i­cant toxicities are asso­ci­ated with CART
ther­apy that include CRS and ICANS. Such toxicities may pre­

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/164/1851658/164karmali.pdf by guest on 13 December 2021

off-the-shelf immunotherapies tak­ing cen­ter stage.
clude patients with cer­tain comorbid con­di­tions. Most impor­
CARTs ther­apy tantly, although CARTs offer dura­ble responses in some, 60% to
CARTs are autol­o­gous T cells that have been genet­i­cally reengi- 70% of patients will still relapse.4,7,8
neered using viral trans­duc­tion to express a CAR that tar­gets a Since the approval of CARTs for aggres­sive B-cell ther­apy, the
spe­cific tumor anti­gen. For B-cell lym­pho­mas, the CAR includes FDA has approved a wave of immunotherapies (with or with­out
an extra­cel­lu­lar moi­ety derived from an anti-CD19 sin­gle-chain var­ che­mo­ther­apy) and targeted approaches for R/R DLBCL. These
i­able frag­ment for anti­gen rec­og­ni­tion and intra­cel­lu­lar domains include the com­bi­na­tions of anti-CD19 mono­clo­nal anti­body
includ­ing a costimulatory domain, CD28 or 41BB, in tan­dem with (mAb) tafasitamab and lenalidomide,9 anti-CD79b anti­body-drug
a CD3ζ-acti­vat­ing domain. The US Food and Drug Administration con­ju­gate (ADC) polatuzumab vedotin with bendamustine and
(FDA) has approved three con­structs for the treat­ment of R/R rituximab,10 and monotherapy with anti-CD19 ADC loncastuximab
aggres­sive B-cell lym­pho­mas, includ­ing DLBCL, high-grade B-cell tesirine,11 or selinexor, an orally avail­­able selec­tive inhib­i­tor of
lym­phoma, transformed fol­lic­u­lar lym­phoma, and pri­mary medi­as­ nuclear export.12 Each of these options should be con­sid­ered
ti­nal B-cell lym­phoma, after 2 prior lines of sys­temic ther­apy, and for patients who are either poor can­di­dates for CART or who
they show high response rates with dura­ble remis­sions. The first relapse after CART, although data supporting these appli­ca­tions
con­struct, approved in 2017, for this pop­u­la­tion was axicabtagene are lim­ited.
ciloleucel (axi-cel), containing a CD28 costimulatory domain. The In addi­tion to the mAbs and ADCs listed above, a num­ber
mul­ti­cen­ter phase 1/2 ZUMA-1 trial eval­u­ated axi-cel and had the of other off-the-shelf immunotherapies have been eval­u­ated in
lon­gest fol­low-up of all­CART tri­als of greater than 4 years (n = 101); aggres­sive lym­pho­mas (Figure 1).13-20 Bispecific T-cell-engag­ing
responses were dura­ble, with a median OS of 25.8 months and antibodies (BsAbs) have emerged as a novel class of off-the-shelf
a 4-year OS rate of 44%.4,5 The phase 2 JULIET study of tisagen- immunotherapies with clear effi­ cacy in R/R aggres­ sive B-cell
lecleucel dem­on­strated that the effi­cacy is com­pa­ra­ble for this lym­ pho­ mas, includ­ing for those patients relaps­ ing after CART
41BB-containing CART, with a more favor­able tox­ic­ity pro­file.6,7 The ther­apy. BsAbs are designed to simul­ta­neously bind to CD3 epsi­
TRANSCEND study, the larg­est CART trial, eval­u­ated the 41BB con­ lon, a com­po­nent of the T-cell recep­tor com­plex, and CD20 on
struct lisocabtagene maraleucel (liso-cel), manufactured uniquely the cell sur­face of malig­nant B cells, cre­at­ing an “immune syn­
through the sep­a­rate trans­duc­tion, expan­sion, and admin­is­tra­tion apse” that redi­rects T-cell cyto­toxic activ­ity against malig­nant B
of equal tar­get doses of CD4+ and CD8+ CARTs. This trial estab- cells. Four agents eval­u­ated in R/R aggres­sive B-cell lym­pho­mas
lished the appli­ca­bil­ity of CARTs to a broader pop­u­la­tion, includ­ include mosunetuzumab, epcoritamab, glofitamab, and odronex-
ing patients with prior allo­ge­neic stem cell trans­plan­ta­tion and tamab.15-19 Unlike their pre­de­ces­sor blinatumomab, a CD19/CD3
those with sec­ond­ary cen­tral ner­vous sys­tem involve­ment.8 bispecific T-cell engager, CD20/CD3 BsAbs have a lon­ger half-life,
allowing for greater ease of admin­is­tra­tion, and appear to offer
higher response rates in R/R aggres­sive B-cell lym­pho­mas.20 Fur-
thermore, CD20/CD3 BsAbs have the poten­tial to cir­cum­vent the
short­com­ings of CARTs while pro­vid­ing high rates of response.20
CLINICAL CASE: PART 2 Herein, the focus is on com­par­ing and contrasting fea­tures
The patient was enrolled in the TRANSCEND trial with liso-cel. The of CARTs and BsAbs as pro­to­types of cus­tom engineered vs off-
patient under­went leukapheresis for T cells followed by bridg­ing the-shelf immunotherapies, respec­tively, with the strengths and
ther­apy with rituximab and high-dose ste­roids for rap­idly pro­ lim­i­ta­tions of each modal­ity outlined (Table 1).
gres­sive dis­ease. Four weeks after leukapheresis, manufactur­ing
was com­plete. Lymphodepleting che­mo­ther­apy with fludara- Off-the-shelf immunotherapies vs CARTs: ease
bine at 30 mg/m2/d intra­ve­nously (IV) and cyclo­phos­pha­mide at of admin­is­tra­tion
300 mg/m2/d IV for 3 days was admin­is­tered, followed by infu-­ CARTs are cus­tom­ized prod­ucts engineered for each indi­vid­ual
sion of liso-cel at a dose of 100 × 106 cells. He expe­ri­enced grade patient. The manufactur­ing pro­cess has been refined to ensure
2 cyto­kine release syn­drome (CRS) with fever, mild hypo­ten­sion that the end prod­ucts meet spec­i­fi­ca­tions for via­bil­ity and com­
requir­ing intra­ve­nous flu­ids, and mild hyp­oxia 6 days after the po­si­tion.21 However, despite every pre­cau­tion taken, suc­cess­ful
admin­is­tra­tion of CART and was man­aged effec­tively with the IL- man­u­fac­ture is not guaranteed—the final prod­uct may not meet
6 recep­tor antag­o­nist tocilizumab. This resolved within 5 days. spec­i­fi­ca­tions or may entirely fail to gen­er­ate. Furthermore, the
He had no signs of immune effec­torPrakash Singh Shekhawatcess can be lengthy.
cell-asso­ci­ated neu­
ro­
logic pro­

Off-the-shelf vs cus­tom engineered immu­no­ther­apy | 165


Figure 1.  Evolving landscape for customized engineered and off-the-shelf immunotherapies in aggressive B-NHL. ADCC: antibody-
dependent cell cytotoxicity; ADCP, antibody-dependent cellular phagocytosis; BiTE, bispecific T-cell engager; PD-1, programmed cell
death 1; PD-L1, programmed cell death ligand 1.
For both the ZUMA-1 and JULIET tri­als, a min­i­mum abso­lute
lym­pho­cyte count was required, which may be pro­hib­i­tive
in heavily pretreated patients. Rates of manufactur­ ing fail­
ure
ranged from 1% to 7%. Conversely, in the TRANSCEND trial,
despite lax­ity in eli­gi­bil­ity with no require­ment for min­i­mum
abso­lute lym­pho­cyte count, manufactur­ing fail­ure occurred in
only 2 patients.4,7,8
Factoring in the time from patient iden­ti­fi­ca­tion, leukaphere-
sis, and manufactur­ing to even­tual admin­is­tra­tion, with expected
logis­ti­cal delays along the way, the turn­around time for CARTs is
unpre­dict­able and typ­i­cally greater than 3 to 4 weeks. Accord-
ingly, CARTs may not be fea­si­ble for patients with rap­idly pro­
gres­sive dis­ease. Bridging ther­apy is often needed dur­ing the
manufactur­ing period, with lim­ited guid­ance as to which ther­a­
pies are most effec­tive for this pur­pose. Allogeneic off-the- shelf
CARTs would solve this par­tic­u­lar issue, lead­ing to timely acces­
si­bil­ity, but their devel­op­ment remains in infancy.22
Bispecific antibodies as an off-the-shelf option can be used
with­out ex vivo T-cell prep­a­ra­tion, allowing imme­di­ate treat­
ment. However, unlike CARTs, treat­ment dura­tion with BsAbs is
prolonged, cre­at­ing issues for acces­si­bil­ity and ease of admin­is­
tra­tion. These agents are admin­is­tered every 1 to 4 weeks either
IV or sub­cu­ta­ne­ously, with shorter inter­vals early in the treat­ment
course, and may be pur­sued for 12 cycles and beyond, depend-
ing on the agent used and the dura­bil­ity of response.15-17,19
Off-the-shelf immunotherapies vs CARTs: toxicities
and appli­ca­tion in vul­ner­a­ble pop­u­la­tions
Side effects asso­ci­ated with CARTs include CRS and ICANS, pro-
longed cytopenia, and impair­ment of humoral immu­nity with
increased risk of infec­tion (Figure 2). CRS is the most com­mon
and has been described in 42% to 93% of patients, with grade
≥3 events occur­ring in 2% to 22% of patients.4,7,8 The path­o­phys­i­­
ol­ogy of CRS has been attrib­uted to an upsurge in cyto­kines and
chemokines upon acti­va­tion of CARTs after engage­ment with
166  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/164/1851658/164karmali.pdf by guest on 13 December 2021
malig­nant cells with IL-6 as a pri­mary driver.4,23 Rates of ICANS
range from 21% to 64%, with grade ≥3 events described in 10%
to 28%.4,7,8 The mech­a­nism of ICANS is elu­sive but has been asso­
ci­ated with high cyto­kine lev­els as well.4
Variability in the pre­sen­ta­tion, prev­al­ence, and inten­sity of
CRS and ICANS across CART con­structs has been attrib­uted to
patient-related, dis­ease-related, and prod­uct-spe­cific fac­tors.
For prod­uct-spe­cific fac­tors, dif­fer­ences in T-cell expan­sion and
pro­lif­er­a­tion kinet­ics con­ferred by the CD28 vs 41BB costimula-
tion domains may explain the higher CRS and ICANS rates asso­
ci­ated with axi-cel.4,7,8
Although the prev­ a­
lence of CRS and ICANS is high with
CARTs, these toxicities are effec­ tively man­ aged. Early mit­ i­
ga­tion strat­e­gies with anti-IL-6 ther­apy and/or ste­roids have
improved the safety pro­file of CARTs with­out hav­ing an impact
on CART func­tion, effi­cacy, or per­sis­tence.24,25 Real-world data
with CARTs and the move­ment toward out­pa­tient CART admin­
is­tra­tion are tes­ta­ments to suc­cess­ful tox­ic­ity mit­i­ga­tion.26,27 For
example, data cap­tured from real-world expe­ri­ence with axi-cel
showed that 43% of patients would not have met eli­gi­bil­ity cri­
te­ria for the registrational ZUMA-1 trial because of comorbidi-
ties. Despite the inclu­sion of older and sicker patients, toxicities
and clin­i­cal out­comes were sim­i­lar for these patients com­pared
to out­comes in the piv­otal trial.26 Additionally, the fea­si­bil­ity of
CART admin­is­tra­tion has been dem­on­strated more for­mally in
the older adult and unfit pop­u­la­tion. The phase 2 PILOT trial
was the first to assess the safety and effi­cacy of liso-cel as a
sec­ond-line ther­apy for trans­plant-inel­i­gi­ble patients with R/R
aggres­sive lym­phoma. This included patients ≥70 years of age or
with impaired organ func­tion includ­ing mod­er­ate car­dio­my­op­a­
thy (left ventricular ejection fraction ≥40%-50%) and/or  pul­mo­
nary  impair­ment  (DLCO ≤60% but blood-oxygen saturation ≥ 92%).
Rates of CRS, ICANS, and response were com­pa­ra­ble to those
of the TRANSCEND study with liso-cel in third line ther­apy and
beyond.28
 Table 1.  Summary of clin­i­cal tri­als for CD19 CARTs and bispecific antibodies

Axi-cel4,34 Tisa-cel6,7 Liso-cel8 Mosunetuzumab15 Epcoritamab16 Glofitamab17,18 Odronextamab19

Structure CD19/CD3z/CD28 CD19/CD3z/41BB CD19/CD3z/41BB Full-length human­ CD20/CD3 BsAb CD20/CD3 Hinge-sta­bi­lized,
ized, IgG1 BsAb with fully human IgG4
CD20/CD3 BsAb 2:1 molec­u­lar CD20/CD3 BsAb
con­fig­u­ra­tion
with biva­lent
bind­ing to
CD20 and
mono­va­lent
bind­ing to
CD3
Route of IV IV IV IV or SC SC IV IV
admin­is­tra­tion
Trial/NCT ZUMA-1 JULIET TRANSCEND NCT02500407 NCT03625037 NCT03075696 NCT02290951
N = treated 101 111 269 141 45 127 71
(aggres­sive
his­tol­ogy)
Median lines 3 (2-4) Median not 3 (2-4) 3 (1-14); 23 pts with 3 (1-6); 6 pts with 3 (1-13) 3 (1-11); 29 pts with
prior ther­apy reported; prior CART prior CART prior CART
(range) range 1-6
Dosing 2 × 106/kg × 1 dose 0.6-6.0 × 108 × 1 0.5-1.0 × 108 × 1 dose Step-up dos­ing on Flat dose in 28-day Step-up dos­ing Step-up dose
dose days 1, 8, and 15 of cycles (q1w: on cycle (C) consisting of ini­tial
cycle 1, followed cycles 1–2; q2w: 1, day (D) 1 dose at week (W)
by fixed doses on cycles 3–6; q4w and 8 and 1, an inter­me­di­ate

Prakash Singh Shekhawat

day 1 of each 21- there­af­ter) until then at the dose at W2, and
day cycle for up to dis­ease pro­gres­ tar­get dose there­af­ter a fixed
17 cycles sion or unac­cept­ from C2D1, q weekly dose until
able tox­ic­ity 3 weeks for up W12 followed by
to 12 cycles main­te­nance q2w
Median time to Leukapheresis → Enrollment → Leukapheresis → infu­
man­u­fac­ture infu­sion: 17 days infu­sion: 21- sion: 24-day tar­get
day tar­get Immediate off-the-shelf immu­no­ther­apy
Median 4 years 40.3 months 18.8 months Not reported 8.3 months 13.5 months 3.9 months
fol­low-up
ORR (%) 82 52 73 35 67 71 • Without prior CART
(n = 11): ORR 55%
• Post-CART (n = 24):
ORR 33%
CR (%) 54 40 53 19 33 64 • Without prior CART
(n = 11): CR 55%
• Post-CART (n = 24):
CR 21%

Off-the-shelf vs cus­tom engineered immu­no­ther­apy | 167

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/164/1851658/164karmali.pdf by guest on 13 December 2021
 Table 1.  Summary of clin­i­cal tri­als for CD19 CARTs and bispecific antibodies (Continued)

Axi-cel4,34 Tisa-cel6,7 Liso-cel8 Mosunetuzumab15 Epcoritamab16 Glofitamab17,18 Odronextamab19

Median DOR 8.1 Not reached Not reached Median time from Not yet mature 5.5 • Without prior CART:
(mo) first CR: 8.8 10.3
• Post-CART: 2.8
Median PFS (mo) 5.8 <6 months; not 6.8 Not yet mature Not yet mature 2.9 Not yet mature

168  |  Hematology 2021  |  ASH Education Program

reached in
CRs
Median OS (mo) 25.8 11.1 21.1 Not yet mature Not yet mature Not yet mature Not yet mature
CRS (%) All grades: 93 All grades: 58 All grades: 42 All grades: 28 All grades: 58 All grades: 50 All grades: 62
≥G3: 13 ≥G3: 22 ≥G3: 2 ≥G3: 1.4 ≥G3: 0 ≥G3: 3.5 ≥G3: 7
ICANS (%) All grades: 64 All grades: 21 All grades: 30 All grades: 1.4 All grades: 6 All grades: 5.3 All grades: not
≥G3: 28 ≥G3: 12 ≥G3:10 ≥G3: 0 ≥G3: 3 ≥G3: 0 reported
≥G3: 2.3
Population appli­ DLBCL/PMBCL/TFL DLBCL/TFL DLBCL/PMBCL/trans­ Aggressive B-cell Aggressive B-cell Aggressive B-cell Aggressive B-cell lym­
ca­bil­ity/con­ for­ma­tion from lym­phoma prior to lym­phoma prior to lym­phoma phoma prior to and
sid­er­ations indo­lent lym­phoma; and post-CART and post-CART post-CART
FL grade 3B; pre­
ferred for sys­temic
+ con­cur­rent sec­
ond­ary CNS dis­ease
or post-alloSCT
alloSCT, allo­ge­neic stem cell trans­plan­ta­tion; DOR, dura­tion of response; FL: follicular lymphoma; IgG: immu­no­glob­u­lin G; PMBCL, pri­mary medi­as­ti­nal B-cell lym­phoma; TFL, transformed
fol­lic­u­lar lym­phoma; tisa-cel, tisagenlecleucel.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/164/1851658/164karmali.pdf by guest on 13 December 2021

 Figure 2.  CART or BsAb treatment-related adverse effects of interest with an incidence of ≥10% and ≥5%, respectively.
Taken col­lec­tively, such toxicities should not pre­clude the use
of CARTs in older patients. Based on clin­i­cal trial and real-world
expe­ri­ence, both CD28 and 41BB con­structs appear to be rea­
son­able options for older adult and/or frail patients or those who
have comorbid con­di­tions, rec­og­niz­ing the dif­fer­ences in tox­ic­
ity pro­files for these con­structs. Practically, how­ever, one might
favor the use of 41BB con­structs in patients with under­ly­ing neu­
ro­logic comorbidities, given the lower rates of ICANS asso­ci­
ated with these con­structs. Additionally, the grow­ing prac­tice
of out­pa­tient admin­is­tra­tion of CARTs with pref­er­ence for 41BB
con­structs in this con­text is likely to lead to wider access and
uti­li­za­tion of this ther­a­peu­tic.
Bispecific antibodies can also pro­ duce CRS and neu­ ro­
logic
toxicities, seem­ingly at lower fre­quen­cies and sever­ity, although
tox­ic­ity data are emerg­ing and not yet mature. Additional toxici-
ties described for BsAbs with an inci­dence of ≥10% include pyrexia,
reac­tion at the injec­tion site, head­aches, and cytopenia (Figure 2).
Rates of CRS for BsAbs range from 28% to 62%, with grade ≥3
events in 0% to 7% of patients, and tend to dis­si­pate after 1 to
2 cycles of admin­is­tra­tion.15-17,19 CRS appears to be driven by IL-6
with BsAbs as well and is man­aged effec­tively with tocilizumab
if needed.15-17,19 Rates of ICANS for BsAbs are not clearly reported;
rates of grade ≥3 events range from 0% to 3%.15-17,19 Clinical and bio­
logic pre­dic­tors of CRS with BsAbs remain unclear. For mosunetu-
zumab, aggres­ sive dis­ ease his­
tol­
ogy and a base­ line ele­vated
C-reac­tive pro­tein appear to pre­dict greater neu­ro­logic tox­ic­ity.29
Like CARTs, BsAbs have dem­ on­
strated fea­si­
bil­ity in older
patients and patients with comorbid con­ di­
tions. As a sin­ gle
agent, mosunetuzumab was eval­u­ated as front­line ther­apy in 19
patients aged ≥80 years or 60 to 79 years with func­tional impair­
ments or comorbid con­di­tions pre­clud­ing the use of full-dose
chemo-immu­no­ther­apy and dem­on­strated effi­cacy with remark­
able tol­er­a­bil­ity.30
Furthermore, a num­ber of strat­e­gies are being employed to
opti­mize the dos­ing and tol­er­a­bil­ity of BsAbs. For exam­ple, step-
up dos­ing for BsAbs is rou­tine. The sub­cu­ta­ne­ous for­mu­la­tion of
mosunetuzumab was shown to reduce the sever­ity of CRS; CRS
events were mild, tran­sient, and delayed in onset and required
min­i­mal inter­ven­tion with no grade ≥3 events reported.31 With
glofitamab, the use of a cytoreductive anti-CD20 mAb and step-
up dos­ing have been shown to mit­i­gate CRS.17,18 Although it is
unclear which strat­egy is most effec­tive in decreas­ing tox­ic­ity, col­
lec­tively, these strat­e­gies may allow for higher-dose drug admin­
is­tra­tion and foreseeably improved response rates with BsAbs.
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/164/1851658/164karmali.pdf by guest on 13 December 2021
Off-the-shelf immunotherapies vs CARTs: effi­cacy
and sequenc­ing
For all 3 FDA-approved CD19 CARTs, the pat­terns and dura­bil­ity of
response are sim­i­lar (Table 1). Response rates range from 52% to
82%, with CR rates of 40% to 54%.4,7,8 Long-term fol­low-up data
for CARTs sug­gest that these responses are dura­ble, par­tic­u­larly
for patients in CR. For the ZUMA-1 study, the median fol­low-up
is now greater than 4 years, the median OS rate is 25.8 months,
and the 4-year OS rate is 44% (n = 101).5 In the JULIET study, the
median fol­low-up is 40.3 months (n = 115). Although the median
OS was 11.1 months, the pro­gres­sion-free sur­vival (PFS) rates at
24 and 36 months were 33% and 31%, suggesting that few pa-
tients who achieve a CR will relapse beyond 24 months.6 For the
TRANSCEND study, with a shorter fol­low-up, median PFS and OS
were 6.8 and 21.1 months, respectively.8
For BsAbs, response rates in aggres­sive lym­pho­mas range
from 33% to 71%, with CRs of 19% to 64%, and may depend on
prior CART expo­sure. However, expe­ri­ence with BsAbs is still
lim­ited; fol­low-up is short and data on dura­bil­ity of response
are lacking. Similarly, the impact of these agents on sur­vival
com­pared to CART is not clear. What is clear is that these
agents do main­tain their effects in patients with relapse after
CART (Table 1).
For instance, results for mosunetuzumab in 30 patients who
had received prior CART ther­apy were high­lighted, and 18 patients
were eval­u­ated for response. In this sub­group, mosunetuzumab
led to CART expan­sion and gen­er­ated an ORR of 39% and a CR
rate of 22% with long-last­ing responses and tol­er­a­ble safety.15
Similarly, odronextamab was eval­u­ated in patients post-CARTs
(n = 24) and dem­on­strated encour­ag­ing activ­ity with an ORR of
33% and a CR rate of 21%.19
With clin­i­cal expe­ri­ence of sequenc­ing strat­e­gies in R/R
DLBCL essen­tially lim­ited to CARTs as a third-line ther­apy fol-
lowed by BsAbs, this sequence remains favored (Figure 3).
One could con­sider CD20/CD3 BsAbs as a bridge to CARTs in
patients with rap­idly pro­gres­sive dis­ease or even as a bridge
to allo­ge­neic stem cell trans­plan­ta­tion. Given the poten­tial for
T-cell exhaus­tion with programmed cell death ligand 1 upreg-
ulation in tar­get cells seen with BsAbs, whether uti­liz­ing BsAbs
prior to leukapheresis could have an impact on the qual­ ity
of harvested T cells for CART man­u­fac­ture is ques­tion­able.32
As both CARTs and BsAbs make their way to ear­lier lines of
ther­apy, how best to sequence these agents will con­tinue to
evolve.30,33
Off-the-shelf vs cus­tom engineered immu­no­ther­apy | 169

Figure 3.  Algorithm for preferred and alternative treatment options for R/R DLBCL that includes customized engineered and off-
the-shelf immunotherapies.
Off-the-shelf immunotherapies vs CARTs: effi­cacy
in high-risk pop­u­la­tions
Ahead of off-the-shelf immunotherapies, CARTs are being eval­
u­ated in sev­eral patient sub­sets with poor prog­no­ses and high
unmet needs. First, patients with dou­ble-expressor or high-grade
B-cell lym­pho­mas with rearrangements of MYC and BCL2 and/or
BCL6, also known as dou­ble- and tri­ple-hit lym­pho­mas, were in-
cluded in piv­otal tri­als for all 3 FDA-approved CARTs. Response
rates in these sub­sets were sim­i­lar to those seen for all­patients.7,8,34
Trials are also under­way for CARTs in patients with intrin­sic
che­mo­ther­apy resis­tance. Both the ZUMA-7 (NCT03391466) and
TRANSFORM (NCT03575351) tri­als have com­pared axi-cel or liso-
cel, respec­tively, vs ASCT as sec­ond-line ther­apy for patients with
pri­mary refrac­tory dis­ease or relapse within 12 months of front­line
ther­apy, with mature results eagerly awaited. The ZUMA-12 study
is eval­u­at­ing axi-cel in patients with large B-cell lym­phoma who
had either high-grade lym­phoma or an international prognostic
index score ≥3 and a pos­i­tive interim positron emission tomog-
raphy after 2 cycles of R-CHOP/R-CHOP-like ther­apy.35 Thus far,
of 12 response-evaluable patients the ORR is 92%, with a CR rate
of 75%. Longer-term fol­low-up of these tri­als will pro­vide greater
insight into the ben­e­fit of CARTs in pri­mary refrac­tory patients.
Experience with off-the-shelf immunotherapies is lim­ited in
these high-risk pop­u­la­tions. In fact, for the L-MIND study, which
eval­u­ated the CD19 human­ized anti­body tafasitamab with lena-
lidomide, patients with pri­mary refrac­tory dis­ease and/or high-
grade B-cell lym­pho­mas with rearrangements of MYC and BCL2
and/or BCL6 were excluded.9 Data for BsAbs remain imma­ture,
with inad­e­quate ana­ly­ses of sub­set pop­u­la­tions with high-grade
B-cell lym­phoma and/or refrac­tory dis­ease. It is antic­i­pated that
this infor­ma­tion will become more read­ily avail­­able with ongo­ing
fol­low-up.
CLINICAL CASE: PART 3
Surveillance scans in our patient were conducted 180 days
post-CART with con­cern for relapse in the retroperitoneum. A
170  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/164/1851658/164karmali.pdf by guest on 13 December 2021
biopsy of a ret­ro­per­i­to­neal node iden­ti­fied CD19− relapse of
dis­ease. The patient was sub­se­quently offered a clin­i­cal trial
with a novel CD20/CD3 bispecific anti­body.
Off-the-shelf immunotherapies vs CARTs: mech­a­nisms
of resis­tance and future direc­tions
Predictors of response and relapse with regard to both engi-
neered prod­ucts and off-the-shelf ther­ap ­ ies remain elu­sive. For
CARTs, it is clear that relapses can occur despite the per­sis­
tence of reengineered T cells. CART exhaus­tion stem­ming from
an immu­no­sup­pres­sive tumor micro­en­vi­ron­ment (TME) and host
sys­temic inflam­ma­tion along with intrin­sic T-cell dys­func­tion
may explain this phe­nom­e­non.6,36 These find­ings sug­gest oppor­
tu­ni­ties for com­bi­na­tions with immuno-onco­log­i­cal agents such
as check­point inhib­i­tors, tyro­sine kinase inhib­i­tors, and immu­
no­mod­u­la­tory agents that may reinvigorate per­sis­tent CARTs,
although this runs the risk of increased tox­ic­ity.37
Given that BsAbs also rely on the patient’s own T cells, one
expects T-cell exhaus­tion and dys­func­tion to be rel­e­vant mech­a­
nisms of resis­tance to said ther­a­peu­tics as well.
Allogeneic CARTs afford the oppor­tu­nity to min­i­mize the con­
tri­bu­tion of T-cell dys­func­tion to relapse risk but may not be a ­ ble
to over­come the immu­no­sup­pres­sive effects of the TME (Table 2).
Limitations asso­ci­ated with this modal­ity also include a risk of
increased immune toxicities, graft-ver­sus-host dis­ease, and pos­
si­ble rejec­tion.21 Allogeneic nat­u­ral killer (NK) CARs rep­re­sent
another immunocellular plat­form with sev­eral advan­tages over
allo­ge­neic CARTs—they can be selected from non-HLA related
healthy donors, will not cause graft-ver­ sus-host dis­
ease, and
are less prone to the inhib­i­tory effects of the TME (Table 2).38
Similarly, bispecific dual-affinity retargeting (DART) pro­ teins
designed to tar­get LAG3 and programmed cell death 1 may bet­
ter over­come the neg­a­tive effects of the TME and have dem­
on­strated responses in CART-treated and naive patients.39 As an
added ben­e­fit, all­ afore­men­tioned prod­ucts rep­re­sent off-the-
shelf options.
Table 2.  Novel dual-targeted autol­o­gous CARTs, allo­ge­neic CARTs, and NK CARs cur­rently in clin­i­cal trial

Clinicaltrials​­.gov iden­ti­fier Sponsor Phase Target Class

Dual-targeted autol­o­gous CARTs
NCT04186520 Medical College of Wisconsin 1/2 CD19/CD20 Autologous dual-tar­get CART
NCT03287817 Autolus Limited 1/2 CD19/CD22 Autologous dual-tar­get CART
followed by lim­ited dura­
tion of anti-PD-1 anti­body
pembrolizumab
Allogeneic CARTs
NCT03939026 Allogene Therapeutics 1/2 CD19 Single-tar­get allo­ge­neic CART
NCT03398967 Chi­nese PLA General Hospital 1/2 CD19/CD20 or CD19/CD22 Dual-tar­get allo­ge­neic CART

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/164/1851658/164karmali.pdf by guest on 13 December 2021

NK CARs
NCT03056339 M. D. Anderson Cancer Center 1/2 CD19 Cord blood CD19 NK CAR
NCT03774654 Baylor College of Medicine 1 CD19 NK CAR

In rare cir­cum­stances, relapses after CD19 CARTs have been
attrib­uted to CD19 anti­gen escape. Mechanisms of anti­gen escape
include altered CD19 mem­brane traf­fick­ing and/or inter­nal­i­za­tion,
expres­sion of CD19 splice var­i­ants that lack the tar­get epi­tope,
or muta­tions in the CD19 gene that lead to disrupted mem­brane
anchor­age.40-43 Of note, concern for antigen escape/loss with use
of CD19-mAb tafasitamab or CD19 ADC loncastuximab tesirine
drives apprehension to utilize these agents prior to CARTs.
CD20 anti­gen loss has also been described in approx­i­ma­
tely 25% of patients treated with anti-CD 20 mAbs.44 Possible
expla­na­tions include loss through clonal selec­tion, epi­ge­netic
downregulation, inter­nal­i­za­tion of CD20, or arti­fact due to ritux-
imab-bound CD20.44,45 It has yet to be deter­mined whether this
is a rel­e­vant mech­a­nism of resis­tance to CD20/CD3 BsAbs.
Several dual-targeting CARTs that con­cur­rently tar­get 2 anti­gens
and would effec­tively address the chal­lenge of anti­gen escape are
now in devel­op­ment (Table 2). This includes a CD19/CD20 CART
that has dem­on­strated a high response rate of 82% (CR, 64%) with­
out added tox­ic­ity.46 Rational com­bi­na­tions of immunotherapies
directed at mul­ti­ple anti­gens are also a con­sid­er­ation.
these toxicities are milder and may not require such close atten­
tion. However, the fre­quency and dura­tion of admin­is­tra­tion can
be cum­ber­some. Ultimately, a clearer under­stand­ing of the cost-
effec­tive­ness of off-the-shelf and cus­tom­ized engineered immu-
notherapies is needed.
Conclusions
CARTs have changed the treat­ment land­scape for R/R aggres­
sive B-cell lym­pho­mas, pro­vid­ing dura­ble responses in patients
with his­tor­i­cally poor out­comes. CD20/CD3 BsAbs rep­re­sent a
prom­is­ing new class of off-the-shelf immu­no­ther­apy that is high-
ly active and offers the oppor­tu­nity to cir­cum­vent some of the
chal­ lenges faced with the admin­ is­
tra­
tion of CARTs. Although
expe­ri­ence favors the use of CARTs over other immunotherapies
at pres­ent, fur­ther stud­ies and lon­ger-term fol­low-up are needed
to elu­ci­date opti­mal sequenc­ing. Along with ratio­nal com­bi­na­
tions, a num­ber of other off-the-shelf immunotherapies, includ­
ing novel CARs, are being explored to opti­mize ease of admin­
is­tra­tion, safety, and effi­cacy, and they will undoubt­edly lead to
mea­sur­able impacts on patient out­comes.

Off-the-shelf immunotherapies vs CARTs: at what cost?

Both CARTs and BsAbs have been asso­ci­ated with a high finan­cial
bur­den. The cost of FDA-approved CARTs ranges from approx­i­
ma­tely $373,000 to $410,000 and is even higher when fac­tor­ing
in the price asso­ci­ated with the logis­tics of CART admin­is­tra­tion
and the man­age­ment of toxicities. In the TRANSCEND study, rel­
e­vant trial-observed health care resource uti­li­za­tion and costs
were sig­nif­i­cantly greater among patients with grade ≥3 CRS
and/or ICANS (22.8%).47 These data favor the use of 41BB CARTs,
which are asso­ci­ated with a low inci­dence of severe CRS/ICANS
and sup­port the devel­op­ment of safer CART options.
With BsAbs, some of these costs are circumvented, but many
over­lap, given the tox­ic­ity pro­file. The price tag for CD20/CD3
BsAbs has yet to be established. However, if one is to learn any­
thing from the blinatumomab story, these may not be a cheaper
alter­na­tive.
One also needs to con­sider the social bur­den asso­ci­ated with
the admin­is­tra­tion of both CARTs and BsAbs. With CARTs, CRS
and ICANs can have a delayed onset and require care­giver sup­
port for the first 1 to 2 months after admin­is­tra­tion.4,7,8 For BsAbs,
Prakash Singh Shekhawat
Conflict-of-inter­est dis­clo­sure
Reem Karmali: speak­ers’ bureau: AstraZeneca, BeiGene, Kite/Gil-
ead, Morphosys; con­ sul­
tant: Kite/Gilead, BMS/Celgene/Juno,
Karyopharm, Janssen/Pharmacyclics, Morphosys, Epizyme; re-
search funding: Kite/Gilead, BMS/Celgene/Juno, Takeda.
Off-label drug use
Reem Karmali: mosunetuzumab, epcoritamab, glofitamab, and
odronextamab are discussed.
Correspondence
Reem Karmali, Northwestern University Feinberg School of Med-
icine, 676 N St Clair St, Ste 850, Chicago, IL 60611; e-mail: reem​
­.karmali@north­west­ern​­.edu.
References
1. Zelenetz AD, Gordon LI, Abramson JS, et  al. NCCN guide­lines insights:
B-cell lym­pho­mas, ver­sion 3.2019. J Natl Compr Canc Netw. 2019;17(6):
650-661.
Off-the-shelf vs cus­tom engineered immu­no­ther­apy | 171
2. Crump M, Neelapu SS, Farooq U, et al. Outcomes in refrac­tory dif­fuse large
B-cell lym­phoma: results from the inter­na­tional SCHOLAR-1 study. Blood.
2017;130(16):1800-1808.
3. Gisselbrecht C, Glass B, Mounier N, et al. Salvage reg­i­mens with autol­o­
gous trans­plan­ta­tion for relapsed large B-cell lym­phoma in the rituximab
era. J Clin Oncol. 2010;28(27):4184-4190.
4. Neelapu SS, Locke FL, Bartlett NL, et  al. Axicabtagene ciloleucel CAR
T-cell ther­ apy in refrac­ tory large B-cell lym­ phoma. N Engl J Med.
2017;377(26):2531-2544.
5. Jacobson C, Locke FL, Ghobadi A, et  al. Long-term sur­vival and grad­ual
recov­ery of B cells in patients with refrac­tory large B cell lym­phoma treated
with axicabtagene ciloleucel (Axi-Cel). Blood. 2020;136(suppl 1):40-42.
6. Jaeger U, Bishop MR, Salles G, et al. Myc expres­sion and tumor-infil­trat­ing
T cells are asso­ci­ated with response in patients (Pts) with relapsed/refrac­
tory dif­fuse large B-cell lym­phoma (r/r DLBCL) treated with tisagenlecleu-
cel in the JULIET trial. Blood. 2020;136(suppl 1):48-49.
7. Schuster SJ, Bishop MR, Tam CS, et al; JULIET Investigators. Tisagenlecleu-
cel in adult relapsed or refrac­tory dif­fuse large B-cell lym­phoma. N Engl J
Med. 2019;380(1):45-56.
8. Abramson JS, Palomba ML, Gordon LI, et  al. Lisocabtagene maraleu-
cel for patients with relapsed or refrac­ tory large B-cell lym­ pho­mas
(TRANSCEND NHL 001): a multicentre seam­ less design study. Lancet.
2020;396(10254):839-852.
9. Salles G, Duell J, González Barca E, et al. Tafasitamab plus lenalidomide in
relapsed or refrac­tory dif­fuse large B-cell lym­phoma (L-MIND): a multicentre,
pro­spec­tive, sin­gle-arm, phase 2 study. Lancet Oncol. 2020;21(7):978-988.
10. Sehn LH, Herrera AF, Flowers CR, et al. Polatuzumab vedotin in relapsed or
refrac­tory dif­fuse large B-cell lym­phoma. J Clin Oncol. 2020;38(2):155-165.
11. Hamadani M, Radford J, Carlo-Stella C, et al. Final results of a phase 1 study
of loncastuximab tesirine in relapsed/refrac­tory B-cell non-Hodgkin lym­
phoma. Blood. 2020;137(19):2634-2645.
12. Kalakonda N, Maerevoet M, Cavallo F, et  al. Selinexor in patients with
relapsed or refrac­tory dif­fuse large B-cell lym­phoma (SADAL): a sin­gle-arm,
mul­ ti­
na­tional, multicentre, open-label, phase 2 trial. Lancet Haematol.
2020;7(7):e511-e522.
13. Ansell SM, Minnema MC, Johnson P, et al. Nivolumab for relapsed/refrac­
tory dif­fuse large B-cell lym­phoma in patients inel­i­gi­ble for or hav­ing failed
autol­o­gous trans­plan­ta­tion: a sin­gle-arm, phase II study. J Clin Oncol.
2019;37(6):481-489.
14. Advani R, Flinn I, Popplewell L, et al. CD47 block­ade by Hu5F9-G4 and rit-
uximab in non-Hodgkin’s lym­phoma. N Engl J Med. 2018;379(18):1711-1721.
15. Schuster SJ, Bartlett NL, Assouline S, et al. Mosunetuzumab induces com­
plete remis­ sions in poor prog­ no­sis non-Hodgkin lym­ phoma patients,
includ­ing those who are resis­tant to or relaps­ing after chi­me­ric anti­gen
recep­tor T-cell (CAR-T) ther­a­pies, and is active in treat­ment through mul­ti­
ple lines. Blood. 2019;134(suppl 1):6.
16. Hutchings M, Mous R, Clausen MR, et  al. Subcutaneous epcoritamab
induces com­plete responses with an encour­ag­ing safety pro­file across
relapsed/refrac­ tory B-cell non-Hodgkin lym­ phoma sub­ types, includ­ ing
patients with prior CAR-T ther­apy: updated dose esca­la­tion data. Blood.
2020;136(suppl 1):45-46.
17. Hutchings M, Carlo-Stella C, Bachy E, et  al. Glofitamab step-up dos­ing
induces high response rates in patients with hard-to-treat refrac­tory or
relapsed non-Hodgkin lym­phoma. Blood. 2020;136(suppl 1):46-48.
18. Hutchings M, Morschhauser F, Iacoboni G, et al. Glofitamab, a novel, biva­
lent CD20-targeting T-cell-engag­ing bispecific anti­body, induces dura­ble
com­plete remis­sions in relapsed or refrac­tory B-cell lym­phoma: a phase I
trial. J Clin Oncol. 2021;39(18):JCO.20.03175.
19. Bannerji R, Allan JN, Arnason JE, et al. Odronextamab (REGN1979), a human
CD20 × CD3 bispecific anti­body, induces dura­ble, com­plete responses in
patients with highly refrac­tory B-cell non-Hodgkin lym­phoma, includ­ing
patients refrac­tory to CAR T ther­apy. Blood. 2020;136(suppl 1):42-43.
20. Coyle L, Morley NJ, Rambaldi A, et  al. Open-label, phase 2 study of bli-
natumomab as sec­ond sal­vage ther­apy in adults with relapsed/refrac­
tory aggres­ sive B-cell non-Hodgkin lym­ phoma. Leuk Lymphoma.
2020;61(9):2103-2112.
21. Subklewe M. BiTEs bet­ter than CAR T cells. Blood Adv. 2021;5(2):607-612.
22. Depil S, Duchateau P, Grupp SA, Mufti G, Poirot L. “Off-the-shelf” allo­
ge­neic CAR T cells: devel­op­ment and chal­lenges. Nat Rev Drug Discov.
2020;19(3):185-199.
23. Teachey DT, Lacey SF, Shaw PA, et al. Identification of pre­dic­tive bio­mark­
ers for cyto­kine release syn­drome after chi­me­ric anti­gen recep­tor T-cell
172  |  Hematology 2021  |  ASH Education Program
ther­apy for acute lym­pho­blas­tic leu­ke­mia. Cancer Discov. 2016;6(6):664-
679.
24. Molina JC, Shah NN. CAR T cells bet­ter than BiTEs. Blood Adv. 2021;5(2):602-
606.
25. Topp M, Van Meerten T, Houot R, et al. Earlier ste­roid use with axicabta-
gene ciloleucel (Axi-Cel) in patients with relapsed/refrac­tory large B cell
lym­phoma. Blood. 2019;134(suppl 1):243.
26. Nastoupil LJ, Jain MD, Feng L, et al. Standard-of-care axicabtagene ciloleu-
cel for relapsed or refrac­tory large B-cell lym­phoma: results from the US
lym­phoma CAR T con­sor­tium. J Clin Oncol. 2020;38(27):3119-3128.
27. Pasquini MC, Hu ZH, Curran K, et  al. Real-world evi­dence of tisagenlec-
leucel for pedi­at­ric acute lym­pho­blas­tic leu­ke­mia and non-Hodgkin lym­
phoma. Blood Adv. 2020;4(21):5414-5424.
28. Sehgal AR, Hildebrandt G, Ghosh N, et  al. Lisocabtagene maraleucel
(liso-cel) for treat­ ment of sec­ ond-line (2L) trans­
plant noneligible (TNE)
relapsed/refrac­tory (R/R) aggres­sive large B-cell non-Hodgkin lym­phoma
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/164/1851658/164karmali.pdf by guest on 13 December 2021
(NHL): updated results from the PILOT study. J Clin Oncol. 2020;38(suppl
15):8040.
29. Diefenbach C, Assouline S, Bosch F, et al. An indi­vid­u­al­ized risk mit­i­ga­tion
approach for safety: expe­ ri­
ence from the mosunetuzumab (CD20/CD3
bispecific anti­body) devel­op­ment pro­gram in rela­tion to neu­ro­tox­ic­ity risk.
Blood. 2019;134(suppl 1):4728.
30. Olszewski AJ, Avigdor A, Babu S, et al. Single-agent mosunetuzumab is a
prom­is­ing safe and effi­ca­cious che­mo­ther­apy-free reg­i­men for elderly/un-
fit patients with pre­vi­ously untreated dif­fuse large B-cell lym­phoma. Blood.
2020;136(suppl 1):43-45.
31. Matasar MJ, Cheah CY, Yoon DH, et al. Subcutaneous mosunetuzumab in
relapsed or refrac­tory B-cell lym­phoma: prom­is­ing safety and encour­ag­ing
effi­cacy in dose esca­la­tion cohorts. Blood. 2020;136(suppl 1):45-46.
32. Feucht J, Kayser S, Gorodezki D, et al. T-cell responses against CD19+ pedi­
at­ric acute lym­pho­blas­tic leu­ke­mia medi­ated by bispecific T-cell engager
(BiTE) are reg­ u­ lated con­
trarily by PD-L1 and CD80/CD86 on leu­ ke­
mic
blasts. Oncotarget. 2016;7(47):76902-76919.
33. Phil­lips TJ, Olszewski AJ, Munoz J, et al. Mosunetuzumab, a novel CD20/CD3
bispecific anti­body, in com­bi­na­tion with CHOP con­fers high response rates
in patients with dif­fuse large B-cell lym­phoma. Blood. 2020;136(suppl 1):37-
38.
34. Locke FL, Ghobadi A, Jacobson CA, et al. Long-term safety and activ­ity of
axicabtagene ciloleucel in refrac­tory large B-cell lym­phoma (ZUMA-1): a
sin­gle-arm, multicentre, phase 1-2 trial. Lancet Oncol. 2019;20(1):31-42.
35. Neelapu SS, Dickinson M, Ulrickson ML, et  al. Interim anal­y­sis of ZUMA-
12: a phase 2 study of axicabtagene ciloleucel (Axi-Cel) as first-line ther­
apy in patients (Pts) with high-risk large B cell lym­phoma (LBCL). Blood.
2020;136(suppl 1):49.
36. Locke FL, Rossi JM, Neelapu SS, et  al. Tumor bur­den, inflam­ma­tion, and
prod­uct attri­butes deter­mine out­comes of axicabtagene ciloleucel in large
B-cell lym­phoma. Blood Adv. 2020;4(19):4898-4911.
37. Siddiqi T, Abramson JS, Palomba ML, et al. Correlation of patient char­ac­ter­
is­tics and bio­mark­ers with clin­i­cal out­comes of JCAR017 in R/R aggres­sive
B-NHL (TRANSCEND NHL 001 study). J Clin Oncol. 2018;36(suppl 5):122.
38. Basar R, Daher M, Rezvani K. Next-gen­er­a­tion cell ther­a­pies: the emerg­ing
role of CAR-NK cells. Blood Adv. 2020;4(22):5868-5876.
39. Wang J, Asch AS, Hamad N, et al. A phase 1, open-label study of MGD013, a
bispecific DART® mol­e­cule bind­ing PD-1 and LAG-3 in patients with relapsed
or refrac­tory dif­fuse large B-cell lym­phoma. Blood. 2020;136(suppl 1):21-22.
40. Abramson JS, Lunning M, Palomba ML. Chimeric anti­gen recep­tor T-cell
ther­a­pies for aggres­sive B-cell lym­pho­mas: cur­rent and future state of the
art. Am Soc Clin Oncol Educ Book. 2019;39:446-453.
41. Sotillo E, Barrett DM, Black KL, et al. Convergence of acquired muta­tions and
alter­na­tive splic­ing of CD19 enables resis­tance to CART-19 immu­no­ther­apy.
Cancer Discov. 2015;5(12):1282-1295.
42. Braig F, Brandt A, Goebeler M, et al. Resistance to anti-CD19/CD3 BiTE in
acute lym­pho­blas­tic leu­ke­mia may be medi­ated by disrupted CD19 mem­
brane traf­fick­ing. Blood. 2017;129(1):100-104.
43. Orlando EJ, Han X, Tribouley C, et al. Genetic mech­a­nisms of tar­get anti­
gen loss in CAR19 ther­apy of acute lym­pho­blas­tic leu­ke­mia. Nat Med.
2018;24(10):1504-1506.
44. Hiraga J, Tomita A, Sugimoto T, et al. Down-reg­u­la­tion of CD20 expres­sion
in B-cell lym­phoma cells after treat­ment with rituximab-containing com­
bi­na­tion chemotherapies: its prev­a­lence and clin­i­cal sig­nif­i­cance. Blood.
2009;113(20):4885-4893.
45. Schmitz K, Brugger W, Weiss B, Kaiserling E, Kanz L. Clonal selec­tion of
CD20-neg­a­tive non-Hodgkin’s lym­phoma cells after treat­ment with anti-
CD20 anti­body rituximab. Br J Haematol. 1999;106(2):571-572.
46. Shah NN, Johnson BD, Schneider D, et al. Bispecific anti-CD20, anti-CD19
CAR T cells for relapsed B cell malig­nan­cies: a phase 1 dose esca­la­tion and
expan­sion trial. Nat Med. 2020;26(10):1569-1575.
Prakash Singh Shekhawat
47. Abramson JS, Siddiqi T, Garcia J, et al. Cytokine release syn­drome and neu­
ro­log­i­cal event costs in lisocabtagene maraleucel-treated patients in the
TRANSCEND NHL 001 trial. Blood Adv. 2021;5(6):1695-1705.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000249
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/164/1851658/164karmali.pdf by guest on 13 December 2021
Off-the-shelf vs cus­tom engineered immu­no­ther­apy | 173
 EMERGING THERAPIES AND CONSIDERATIONS FOR SICKLE CELL DISEASE

Gene therapy for sickle cell disease:

where we are now?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/174/1851456/174kanter.pdf by guest on 13 December 2021

Julie Kanter1 and Corey Falcon2
1
Department of Medicine, University of Alabama at Birmingham, Birmingham, AL; and 2Department of Pediatrics, Ochsner Hospital for Children,
Jefferson, LA

The landscape of sickle cell disease (SCD) treatment continues to evolve rapidly, with new disease-modifying therapies in
development and potentially curative options on the horizon. Until recently, allogeneic stem cell transplant has been the
only proven cure for SCD. Gene therapy is rising to the forefront of the discussion as a potentially curative or highly disease-
modifying option for abating the complications of the disease. Understanding the different types of gene therapy in use,
the differences in their end points, and their potential risks and benefits will be key to optimizing the long-term use of
this therapy.

LEARNING OBJECTIVES
• Have an improved understanding of the different types of gene therapy
• Be able to have a more thoughtful conversation with patients regarding gene therapy

mostly individuals living in low­resource settings or minor­

CLINICAL CASE
A 22­year­old woman with sickle cell anemia (HbSS disease)
presents to you, a sickle cell specialist (as a referral from her
community hematologist­oncologist), to discuss the possi­
bilities of curative treatment. She has a history of frequent
vaso­occlusive (VOC) crises and goes to the emergency
department about 5 times per year, where she is admitted
most of those times. She has never had a stroke (that she
is aware of). She has been on hydroxyurea (HU), 1000mg
per day, since she was 9, pretty reliably. She takes 30mg
of morphine sulfate extended release tablets (MS Contin,
Purdue) taken by mouth twice a day, 10mg of oxycodone
every 4 hours as needed, and folic acid (in addition to HU).
She is on no other medications. She has no full siblings and
does not know her father. She lives with her mother and
2 half siblings approximately 2 hours from your clinic.
On exam, she weighs 60 kg, with a normal body mass
index, a heart rate of 75, a respiration rate of 20, a blood oxy­
gen level at 93%, and a blood pressure reading of 118/70.
Sickle cell disease (SCD) has been well characterized for
over 100 years, with the first clinical report published in 1910
describing it as the “first molecular disease.”1 Despite this
long scientific history, progress toward identifying a cure
has been slow, likely due in part to the fact that SCD affects
ity populations living in high­resource nations. Further, the
development of disease­modifying therapies has also been
sluggish, with only 1 medication available until 2017.
SCD results from the inheritance of at least 1 copy of
sickle hemoglobin (HbS) and a second copy of a gene
encoding HbS or another abnormal hemoglobin. HbS is
produced when the β­globin gene (HBB) contains a single
E6V missense mutation resulting in the replacement of β6
glutamic acid by valine.2 On deoxygenation, HbS polymer­
izes, leading to abnormally shaped red cells and multiple
downstream clinical sequelae, including hemolysis, vaso­
occlusion and subsequent progressive and irreversible
organ damage, decreased quality of life, and early death.3
Although HbS polymerization, vaso­occlusion, and hemo­
lytic anemia are central to the pathophysiology of SCD, the
resultant pathological events are more consistent with sec­
ondary vascular­endothelial dysfunction and widespread
inflammation such that the complications of SCD are better
understood as inflammatory vascular disorders.
The source of SCD pathology remains a single­point
mutation resulting in the production of abnormal pro­
tein (HbS) and subsequent hemoglobin polymerization
within the red blood cell. The rate of HbS polymerization
is highly variable and depends on multiple factors, includ­
ing the amount of HbS per erythrocyte, the amount of other
nonsickling hemoglobin per erythrocyte, and the overall

174 | Hematology 2021 | ASH Education Program

eryth­ro­cyte hemo­glo­bin level. Thus, suf­fi­cient healthy, nor­mal
adult hemo­glo­bin (HbA) in each cell, as in indi­vid­u­als who carry
the sickle cell trait (HbAS), will pre­vent poly­mer­i­za­tion and reduce
the symp­toms and sequelae of the dis­ease except under rare
instances of extreme phys­i­o­log­i­cal stress.4,5 Similarly, fetal hemo­
glo­bin (HbF) has an antisickling effect and can reduce or pos­si­
bly elim­i­nate hemo­glo­bin poly­mer­i­za­tion if in suf­fi­cient quan­tity
within the ­eryth­ro­cyte.6
Four dif­fer­ent SCD-spe­cific med­i­ca­tions have now been
approved to treat the effects of SCD. HU, the first approved, has
been proven to decrease the fre­quency of VOC, reduce stroke
risk in some affected patients, and improve the qual­ity and length
of life in many affected indi­vid­u­als.7 However, HU has not been
proven as a uni­ver­sally accept­able dis­ease-mod­i­fy­ing ther­apy
because many indi­vid­u­als con­tinue to develop sig­nif­i­cant dis­ease-
related com­pli­ca­tions or are unwill­ing to main­tain drug adher­
ence due to con­cerns with side effects, fer­til­ity, or tol­er­ance.
L-glu­ta­mine, approved in 2017 to treat SCD, has anti­ox­i­dant prop­
er­ties that have been shown to ame­lio­rate SCD and decrease the
fre­quency of VOC.8 Long-term dura­bil­ity and util­ity have yet to be
dem­on­strated. The newest med­i­ca­tions, crizanlizumab and vox­
elotor, have proven effi­ca­cious in their own right by decreas­ing
VOCs and decreas­ing hemo­ly­sis, respec­tively, but nei­ther med­
i­ca­tion is likely to have a demon­stra­ble cura­tive effect, and both
require ongo­ing, life­long treat­ment.9,10 Further, these med­i­ca­tions
are not yet proven to reduce, delay, or sta­bi­lize organ-spe­cific
com­pli­ca­tions such as renal dis­ease, avas­cu­lar necro­sis, or pul­mo­
nary hyper­ten­sion.
CLINICAL CASE (Con­t in­u ed)
You review her cur­rent dis­ease sta­tus, com­pli­ca­tions, and con­
cerns and spe­ cif­i­
cally dis­
cuss the impor­ tance of doing a full
screen­ing assess­ment of all­organ sys­tems to eval­u­ate the SCD-
related com­pli­ca­tions. You per­form a full set of labs, an eval­u­a­tion
for iron over­load (fer­ri­tin, liver mag­netic res­o­nance imag­ing), an
echo­car­dio­gram (due to low oxy­gen at base­line), a workup for
alloimmunization, and a kid­ney screen (albu­min/cre­at­i­nine ratio).
Her lab results are con­sis­tent with HU adher­ence, and the echo­
car­dio­gram shows triv­ial tri­cus­pid regur­gi­ta­tion and mild left
ven­tric­u­lar hyper­tro­phy but is oth­er­wise nor­mal. Labs show that
fer­ri­tin is 1200 µg/L, liver iron con­cen­tra­tion is 5.1 mg/g, and there
is no con­cern for sig­nif­i­cant alloimmunization. Unfortunately, her
renal assess­ment shows marked pro­tein­uria.
You dis­ cuss treat­ ment options with the patient, includ­ ing
poten­ tially increas­ ing HU to the max­ i­
mum tol­ er­
ated dose or
adding addi­tional med­i­cal ther­a­pies. In frus­tra­tion, how­ever, she
says, “I want to be cured of this wretched dis­ease!”
One method by which to cor­rect the path­o­phys­i­o­log­i­cal
abnor­mal­ity in SCD is to replace the abnor­mal HbS with more
func­tional HbA. The clin­i­cal ben­e­fit of this molec­u­lar replace­
ment has been clearly dem­on­strated by the suc­cess of hema­
to­poi­etic stem cell trans­plant (HSCT) for SCD. In this sce­nario,
new hema­to­poi­etic stem cells (HSCs) are used as the vehi­cle
to deliver healthy hemo­glo­bin and there­fore elim­i­nate eryth­ro­
cyte sick­ling and the sec­ond­ary effects of this path­o­log­i­cal pro­
Prakash Singh Shekhawat
cess.11,12 Multiple stud­ies have dem­on­strated the effi­cacy of HSCT
for SCD, includ­ing donor HSC with HbAS.13
While highly effi­ca­cious, HSCT is not an option for all­indi­vid­
u­als liv­ing with SCD. The best out­comes reported are in youn­ger
indi­vid­u­als with SCD who have a matched sib­ling donor; unfor­
tu­nately, rel­a­tively few indi­vid­u­als with SCD have such a donor.14
Improvements in haploidentical HSCT for SCD are rap­idly evolv­ing
and will sig­nif­i­cantly increase the avail­­able donor pool. However,
there are still poten­tial draw­backs that can include graft fail­ure,
delayed immune recon­sti­tu­tion, infer­til­ity, sec­ond­ary malig­nancy,
and graft-ver­sus-host dis­ease (GVHD).15 Further, it is unlikely any
allo­ge­neic trans­plant could com­pletely remove the risk of GVHD
dis­ease or the need for long-term immune sup­pres­sion ther­apy,
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/174/1851456/174kanter.pdf by guest on 13 December 2021
both of which carry their own risks of sub­se­quent com­pli­ca­tions.
Thus, there remains a dem­on­strated need for other means of pro­
vid­ing gene trans­fer into HSCs with­out the same immu­no­logic
risks. In this sce­nario, trans­plan­ta­tion of genet­i­cally mod­i­fied
autol­o­gous HSCs pro­vi­des a poten­tial alter­na­tive ther­apy.
CLINICAL CASE (Con­tin­ued)
You dis­cuss poten­tially cura­tive options with your patient that
include haploidentical HSCT and gene ther­ apy, as you have
already deter­mined that she does not have any full sib­lings. Her
mother works full time and has a his­tory of lupus that is on active
treat­ment (and so can­not donate HSCs). She is inter­ested in con­
sid­er­ing haploidentical trans­plant from a sib­ling but is wor­ried
about the dis­tance from home (weekly appoint­ments to mod­u­
late her immune sup­pres­sion, the need to stay in close prox­im­ity
after trans­plant, a con­cern for GVHD and sub­se­quent rehospital­
ization). Using shared deci­sion-mak­ing, it seems she may be best
suited for gene ther­apy, as there is no risk of GVHD and no need
to mon­i­tor immune sup­pres­sion med­i­ca­tion. She asks about the
dif­fer­ent types of gene ther­a­pies avail­­able.
Broadly speak­ing, 4 main types of gene ther­apy are avail­­able for
the treat­ment of SCD. These include gene addi­tion ther­apy, gene
editing, gene silenc­ing, and gene cor­rec­tion ther­apy (Figure 1).
Each type of ther­apy dif­fers in the means by which it induces the
replace­ment of HbS with nonsickling hemo­glo­bin. The nonsick­
ling hemo­glo­bin is the tar­get pro­tein in each ther­apy and will
need to be eval­u­ated using mul­ti­ple novel meth­ods and terms.
1. Gene addi­tion ther­apy is the addi­tion of a new gene using a
viral vec­tor (usu­ally) to deliver a nonsickling glo­bin gene to
the stem cells. In this pro­ce­dure the native HbS gene is not
altered, resulting in the pro­duc­tion of both the new hemo­
glo­bin and the native HbS. Several exam­ples of gene addi­tion
ther­apy are ongo­ing that use a lentiviral vec­tor (LVV) to house
and deliver a new gene.16
2. Gene editing is most often used to describe a pro­cess of gene
dis­rup­tion in the con­text of SCD and can be used to tar­get sup­
pres­sors of HbF as a way to both increase HbF and decrease
HbS. Elements of DNA within a gene can be targeted using a
guide that can iden­tify and tightly bind to the tar­get with high
spec­i­fic­ity cou­pled with an enzyme to cut the DNA, induc­ing
dou­ble-stranded breaks. The spe­cific DNA cut allows one to
Gene ther­apy for sickle cell dis­ease  |  175
 change the sequence with high pre­ci­sion, usu­ally resulting in an
inser­tion and dele­tion. This type of gene ther­apy most often tar­
gets a dif­fer­ent part of DNA (sep­a­rate from the HbS muta­tion)
to pro­duce an increase in HbF pro­duc­tion while recip­ro­cally
suppressing HbS pro­duc­tion.16 Specifically, many of the cur­rent
ther­a­pies tar­get the BCL11A gene, a neg­a­tive reg­u­la­tor of HbF.
In this exam­ple, the gene editing is used to turn off the reg­u­la­
tion of HbF in order to increase HbF pro­duc­tion.
3. Gene silenc­ing uses the reg­u­la­tion of gene expres­sion in a
cell to pre­vent the expres­sion and resul­tant pro­duc­tion of
cer­tain pro­teins. Similar to gene editing, this type of ther­apy
is being used to sup­press the BC11A gene, resulting in an in­
crease in HbF while recip­ro­cally suppressing HbS pro­duc­tion.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/174/1851456/174kanter.pdf by guest on 13 December 2021

In con­trast to gene editing, this type of gene ther­apy has thus
far relied upon viral vec­tor deliv­ery (sim­i­lar to gene addi­tion)
to deliver an anti­sense to mes­sen­ger RNA to sup­press the
gene prod­uct instead of cut­ting the gene.17
4. Gene cor­rec­tion can be performed in sev­eral dif­fer­ent ways.
In most cases, how­ever, a guide RNA is used to iden­tify the
tar­get muta­tion for cut­ting, and then editing occurs with the
simul­ ta­
neous deliv­ ery of tem­ plate DNA of the cor­ rect se­
quence, directing homol­ogy-directed repair (HDR; Figure 2).
Though this is cur­rently the least effi­cient method, efforts are
under­way to improve gene cor­rec­tion, includ­ing through the
inser­tion of DNA, direct base editing, and prime editing. This
is the only type of gene ther­apy that cur­rently aims to elim­
i­nate HbS pro­duc­tion and intro­duce a nonsickling hemo­glo­
bin simul­ta­neously.18
At pres­ent, all­types of gene ther­apy use the same over­all
pro­ce­dure. For each study listed, patients first undergo inten­sive
screen­ing. It is impor­tant to note that the cri­te­ria for each study
dif­fer slightly, but all­require the indi­vid­ual to have had sig­nif­i­cant
SCD-related com­pli­ca­tions (as a rea­son for under­go­ing a study
pro­ce­dure) and to have suf­fic ­ ient organ func­tion to undergo
the che­mo­ther­apy prep­a­ra­tion required. Once fully screened,
patients need to undergo stem cell col­lec­tion using plerixafor
mobi­li­za­tion and apher­e­sis. This pro­cess may be under­taken
more than once to ensure suf­fi­cient stem cells are col­lected
for manip­u­la­tion as well as for backup. Once the stem cells are
appro­pri­ately altered, all­types of gene ther­apy uti­lize che­mo­
ther­apy (to make room for altered/manip­u­lated stem cells). In all­
but 1 trial below (MOMENTUM), these stud­ies use busul­fan che­
mo­ther­apy to pro­vide myeloablation to ensure opti­mal stem cell
engraft­ment. The MOMENTUM trial uses a “reduced-inten­sity”
che­mo­ther­apy reg­i­men using mel­pha­lan.
Current and upcom­ing clin­ic ­ al tri­als using gene ther­apy are
detailed in Table 1. Note that some of the tri­als listed tar­get
the BCL11A gene (its ery­throid enhancer of mes­sen­ger RNA), a
repres­sor of γ-glo­bin expres­sion, in order to induce HbF pro­duc­
tion in adult eryth­ro­cytes, while oth­ers use ran­dom viral vec­tor
inser­tion to result in HbA pro­duc­tion, and 1 study, nota­bly, tar­
gets the sickle cell gene muta­tion itself.17-20
Current clin­i­cal tri­als of lentiviral gene ther­apy based on the
addi­tion of a mod­i­fied β-glo­bin gene (HbAT87Q ) have accu­
mu­lated the most data so far and have dem­on­strated a ben­
e­fit in the reduc­tion of sig­nif­i­cant VOCs in SCD. However, the
data remain early, and results regard­ing improve­ments in long-
term dura­bil­ity and organ func­tion are forth­com­ing. CRISPR
Figure 1. Cartoon rendering of different gene therapies. (A) Gene
editing, (B) gene correction, (C) gene silencing, and (D) gene
addition via viral vector.
Therapeutics and Vertex Pharmaceuticals and inves­ ti­
ga­
tors
from Bos­ton C ­ hildren’s Hospital have also presented data on
their CLIMB and short hair­ pin RNA (shRNA) stud­ ies, respec­
tively, show­ing that the potent antisickling prop­er­ties of HbF
com­bined with the lower lev­els of HbS result in the ­res­o­lu­tion
of VOCs as well. These gene editing and silenc­ing stud­ies, how­
ever, are even ear­lier in their reporting and long-term out­comes.
Gene cor­rec­tion ther­apy will be the next type of gene ther­
apy to enter the clin­i­cal space. Gene cor­rec­tion ther­apy includes
a com­bi­na­tion of gene editing and gene addi­tion. The CEDAR
study, which received US Food and Drug Administration clear­
ance to move for­ward to a phase 1 clin­i­cal trial in early 2021, will
con­cur­rently use a high-fidelity clus­tered reg­u­larly interspaced
short pal­in­dromic repeats (CRISPR)-Cas9 ribo­nu­cleo­pro­tein (RNP)
to induce a dou­ble-strand DNA break­age and HDR with a nonin­
tegrating adeno-asso­ci­ated virus-6 donor DNA repair tem­plate
to pro­duce a new gene prod­uct. Unlike the gene-editing ther­
a­pies that rely on non­ho­mol­o­gous end join­ing and sub­se­quent
inser­tion and dele­tion for­ma­tion in the edited space, gene cor­
rec­tion ther­a­pies rely on the more com­pli­cated and his­tor­i­cally
less effi­cient HDR.21
CLINICAL CASE (Con­tin­ued)
You and your res­i­dent leave the room to allow the patient and
her fam­ily time to dis­cuss and digest all­of this infor­ma­tion. Your
res­id
­ ent astutely asks how it will be pos­si­ble to com­pare out­
comes and mea­sure effi­cacy in tri­als that use dif­fer­ent meth­
ods of hemo­glo­bin induc­tion and dif­fer­ent types of hemo­glo­bin
pro­duc­tion.

176  |  Hematology 2021  |  ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/174/1851456/174kanter.pdf by guest on 13 December 2021
Figure 2. CRISPR-Cas-9-induced double-stranded break and its sequential repair pathways. Left: nonhomologous end joining. Right:
HDR, which requires the insertion of a homologous DNA strand used as a template for a high-fidelity double-stranded DNA break.
PAM, protospacer adjacent motif; sgRNA, single-guide RNA.

Table 1. Current and upcom­ing stud­ies of gene ther­apy in SCD

Genetic silenc­ing
Study name LentiGlobin DREPAGLOBE CLIMB PRECIZN-1 of BCL11A MOMENTUM CEDAR
Type of gene ther­apy Gene addi­tion Gene addi­tion Gene editing Gene editing Gene silenc­ing Gene addi­tion Gene cor­rec­tion
Editing tool NA NA CRISPR-Cas9 RNP Zinc fin­ger ShRNA NA HiFi CRISPR-
Cas9 RNP
Type of stem cell Transduction Transduction Electroporation Transfection Transduction Transduction Electroporation
manip­u­la­tion with zinc fin­
ger nucle­ase
mRNA
Vector (y/n) BB305 LVV DROBE 1 LVV None None BCH-BB694 LVV γG16D LVV Nonintegrating
that encodes AAV6 donor
a microRNA- DNA repair
adapted shRNA tem­plate
Genetic tar­get (y/n) NA NA Erythroid lin­e­age- 11A (BCL11A) BCL11A mRNA N/a Sickle muta­tion
spe­cific locus (aden­o­sine—
enhancer of the (ery­throid > thy­mine
BCL11A gene enhancer) [A— > T]
Drug prod­uct LentiGlobin DREPAGLOBE CTX001 BIVV003 BCH-BB694 ARU-180126 GPH101
BB305
Protein prod­uct HbAT87Q βAS3, an antisickling HbF HbF HbF HbFG16D HbA
β-glo­bin pro­tein
(AS3) containing
3 amino acid
sub­sti­tu­tions in
the wild-type HBB
Prakash Singh Shekhawat
Gene ther­apy for sickle cell dis­ease  |  177

Figure 3. Gene therapy process and evaluation measures. DP, drug product; VCN, vector copy number.
There are sev­eral ways to eval­ua ­ te effi­cacy in gene ther­apy
for SCD. At the end of the pro­cess, the most impor­tant ques­
tion is how much of the new hemo­glo­bin (pro­tein prod­uct of the
gene ther­apy) is being pro­duced over time. However, through­
out the pro­cess are many steps in which interim eval­u­a­tions of
effi­cacy are use­ful. These steps and their eval­u­a­tion tech­niques
are detailed in Figure 3.
As noted, it is most impor­tant to deter­mine how much non­
sickling hemo­glo­bin is in each red blood cell and how much of
it is a prod­uct of the gene ther­apy vs the myeloablation (which
can result in stress eryth­ ro­
poi­e­
sis that causes HbF pro­ duc­
tion). This can be eval­u­ated by the trans­duc­tion effi­ciency, or
the per­cent­age of blood stem cells hav­ing incor­po­rated the
desired genetic mate­rial. Longitudinal stud­ies are needed to
deter­mine the dura­bil­ity of gene ther­apy. Last, it will be impor­
tant to learn over time which symp­toms and/or com­pli­ca­tions
of SCD improve with gene ther­apy and if the out­comes dif­fer
depending on the type and amount of hemo­glo­bin prod­uct.
At this time it is not clear what per­cent­age of stem cells must
receive genetic cor­rec­tions to result in suf­fic ­ ient nonsickling
hemo­glo­bin. Additional fol­low-up assess­ments need to include
both lab­ o­ra­tory and rheologic mea­ sures of hemo­ ly­
sis and
adhe­sion in addi­tion to in-depth patient-reported out­comes.
Finally, it is most impor­tant to assess whether gene ther­apy
can pre­vent vaso-occlu­sion (which types and to what extent)
and can either sta­bi­lize or resolve organ com­pli­ca­tions due to
SCD. These find­ings remain unclear at this time but are highly
nec­es­sary in advanc­ing out­comes in SCD. For exam­ple, up to
10% of per­sons with sickle cell ane­mia may develop end-stage
renal dis­ease. At this time it is unclear whether even allo­ge­neic
trans­ plant can pre­ vent the devel­ op­ ment of end-stage renal
dis­ease once some­one has devel­oped chronic kid­ney dis­ease;
gene ther­apy results are fur­ther behind. While data regard­ing
out­comes for VOC appear clearer, the long-term organ-spe­cific
response to gene ther­apy will truly mea­sure its effi­cacy.
CLINICAL CASE (Con­t in­u ed)
You and the res­i­dent return to the patient and her fam­ily. She
does not have chil­dren but does want to have a fam­ily later in
life. She is wor­ried about the poten­tial for infer­til­ity. Further, she
178  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/174/1851456/174kanter.pdf by guest on 13 December 2021
has read about a patient who got leu­ke­mia after early-stage
gene ther­apy and wants to know the risks of gene ther­apy.
There are real and poten­tial risks involved with gene ther­apy
of all­types. The che­mo­ther­apy used in myeloablation car­ries a
high risk of infer­til­ity (nearly 100%) and also results in mucositis,
nau­sea, loss of appe­tite, alo­pe­cia, and other usu­ally revers­ible
com­pli­ca­tions. Infertility is a major source of con­cern that must
be addressed. While fer­til­ity pres­er­va­tion is pos­si­ble for some
indi­vid­u­als, it is nei­ther uni­ver­sally avail­­able nor effi­ca­cious. It
is impor­tant to have patients meet with fer­til­ity spe­cial­ists to
review the avail­­able options and their asso­ci­ated risks and ben­
e­fits. Secondary malig­nancy is another major risk of gene ther­
apy. Chemotherapy such as busul­ fan car­ ries an inde­ pen­dent
long-term risk of sec­ ond­ ary malig­ nancy in patients under­ go­
ing both allo­ge­neic and autol­o­gous trans­plant. Another poten­
tial cause of a sec­ond­ary malig­nancy is the trans­plan­ta­tion of
poten­tially dam­aged HSCs. Individuals with SCD are affected by
chronic inflam­ma­tion and endo­the­lial dam­age as well as hyp­
oxic bone infarc­tion and con­stant eryth­ro­poi­etic stress.22 These
man­i­fes­ta­tions of SCD likely dam­age the HSCs and may result
in a pre­dis­po­si­tion to malig­nant trans­for­ma­tion. At this point it
is unclear how high this risk is or if it can be suit­ably miti­gated
with changes already in use in gene ther­apy pro­to­cols or future
changes to come. Two patients in the ini­tial LentiGlobin HGB-
206 trial devel­oped acute mye­log­e­nous leu­ke­mia at 3 and
5 years post autol­o­gous gene ther­apy.23 Currently, the workup
sug­gests that the vec­tor is not asso­ci­ated with the malig­nancy
and that per­haps there is an inher­ent increased risk in those with
SCD wors­ened by low cell dose, low vec­tor copy num­ber, and
a return to the SCD phe­no­type of high eryth­ro­poi­etic stress.
These risks may have been miti­gated by the use of a plerixafor-
medi­ated stem cell har­vest (in place of bone mar­row col­lec­tion)
and precollection trans­fu­sion ther­apy; how­ever, the degree to
which this will reduce the long-term risk is unknown.
Regarding gene addi­tion, the major con­cern is the poten­tial
risk of an inser­tion at a pro­moter site that causes unwanted cel­lu­
lar pro­lif­er­a­tion or malig­nant trans­for­ma­tion. This issue occurred
recently when a patient with cere­bral adre­nal leu­ko­dys­tro­phy
devel­oped myelodysplastic syn­drome after receiv­ing gene ther­
apy. The viral vec­tor, Lenti-D (Bluebird Bio), dif­fers from those
used in SCD-related gene ther­apy but is sim­i­larly designed to add
func­tional cop­ies of a gene into a stem cell. However, based on
the loca­tion of the LVV inser­tion, there is con­cern that it caused
the myelodysplastic syn­ drome (press report, Bluebird Bio
Aug 13, 2021). This has not yet been observed (in any LVV-based
gene ther­apy for SCD) but remains a poten­tial risk. Gene editing
can also result in unin­tended mod­i­fi­ca­tions at other points along
the genome out­side of the targeted DNA sequence (off-tar­get
effects). Current tech­nol­ogy allows us to iden­tify off-tar­get muta­
tions that occur at a high fre­quency, but it is pos­si­ble that the
more rare ones could (the­o­ret­i­cally) escape detec­tion and give
cells growth or sur­vival advan­tages that pro­mote can­cer. Further,
it is clear that when electroporation is used there is decreased
stem cell sur­vival24; it is unclear if this will result in any addi­tional
neg­a­tive out­comes.
CLINICAL CASE (Con­t in­u ed)
Your patient returns the fol­low­ing week with her fam­ily. After
sev­eral con­ver­sa­tions with you (her phy­si­cian), her fam­ily, and
her pas­tor, she decides she under­stands the risks and ben­e­fits
of gene ther­apy and wants to move for­ward. Your res­i­dent asks
why it took 3 weeks for her to reach this con­clu­sion.
A sickle cell expert and team famil­iar with all­of the ther­a­
peu­tic options for the treat­ment of SCD should be required to
help an indi­vid­ual with SCD make a sig­nif­i­cant treat­ment deci­
sion such as gene ther­apy. The patient should be made aware
of the real­is­tic expec­ta­tions and poten­tial risks of gene ther­
apy, with an empha­sis on the impor­tance of shared deci­sion-
mak­ing between the patient, fam­ily, and phy­si­cian.25 Ample
time detailing the known and unknown aspects of gene ther­
apy with the patient and their fam­ily will facil­i­tate mean­ing­
ful dis­cus­sion and dia­logue. It is also impor­tant to real­ize that
both ben­e­fits and risks are inher­ent in new pro­ce­dures, but the
poten­tial for cure is also avail­­able. Finally, gene ther­apy still
requires myeloablation, which may limit uptake and par­tic­i­pa­
tion based on both the risks of infer­til­ity as well as any base­line
organ dys­func­tion that would make a poten­tial patient inel­i­gi­
ble. Thus, thor­ough delib­er­a­tion should be under­taken before
screen­ing is ini­ti­ated.
Recent data from gene ther­a­pies in prog­ress dem­on­strate
the need for the long-term fol­low-up and con­sis­tent col­lec­tion of
data using com­mon data ele­ments to ensure that out­comes can
be com­pared across tri­als as well as to the nat­u­ral his­tory of SCD.
Additional data are needed regard­ing the use of gene ther­apy
for the spe­cific reme­di­a­tion of organ-related com­pli­ca­tions in
SCD as well as in patients whose pri­mary presenting symp­tom is
severe, chronic pain. Finally, enhanced SCD sur­veil­lance and lon­
gi­tu­di­nal stud­ies are needed to bet­ter under­stand, quan­ti­tate,
and com­pare the poten­tial for malig­nancy in this pop­u­la­tion.
Conflict-of-inter­est dis­clo­sure
Julie Kanter: hon­o­raria: Novartis, Graphite Bio, Forma Therapeu­
tics, Agios, Beam Therapeutics, Guidepoint Global, GLG; research
Prakash Singh Shekhawat
funding: National Institutes of Health, National Heart, Lung, and
Blood Institute, Health Resources and Services Administration,
Centers for Disease Control and Prevention.
Corey Falcon: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Julie Kanter: nothing to disclose.
Corey Falcon: nothing to disclose.
Correspondence
Julie Kanter, University of Alabama at Birmingham, 1720 2nd St
South, Birmingham, AL 35294; e-mail: jkanter@uabmc​­.edu.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/174/1851456/174kanter.pdf by guest on 13 December 2021
References
1. Pauling L. On the sta­bil­ity of the S(8) mol­e­cule and the struc­ture of fibrous
sul­fur. Proc Natl Acad Sci USA. 1949;35(9):495-499.
2. Noguchi CT, Torchia DA, Schechter AN. Determination of deoxyhemoglo­
bin S poly­mer in sickle eryth­ro­cytes upon deox­y­gen­ation. Proc Natl Acad
Sci USA. 1980;77(9):5487-5491.
3. Bunn HF. Pathogenesis and treat­ment of sickle cell dis­ease. N Engl J Med.
1997;337(11):762-769.
4. Heller P, Best WR, Nelson RB, Becktel J. Clinical impli­ca­tions of sickle-cell
trait and glu­cose-6-phos­phate dehy­dro­ge­nase defi­ciency in hos­pi­tal­ized
black male patients. N Engl J Med. 1979;300(18):1001-1005.
5. Platt OS, Thorington BD, Brambilla DJ, et al. Pain in sickle cell dis­ease: rates
and risk fac­tors. N Engl J Med. 1991;325(1):11-16.
6. Ware RE, Davis BR, Schultz WH, et al. Hydroxycarbamide ver­sus chronic
trans­fu­sion for main­te­nance of transcranial doppler flow veloc­i­ties in chil­dren
with sickle cell anae­mia-TCD with trans­fu­sions chang­ing to hydroxy­urea
(TWiTCH): a multicentre, open-label, phase 3, non-infe­ri­or­ity trial. Lancet.
2016;387(10019):661-670.
7. Niihara Y, Miller ST, Kanter J, et al; Investigators of the Phase 3 Trial of
l-Glutamine in Sickle Cell Disease. A phase 3 trial of l-glu­ta­mine in sickle cell
dis­ease. N Engl J Med. 2018;379(3):226-235.
8. Ataga KI, Kutlar A, Kanter J. Crizanlizumab in sickle cell dis­ease. N Engl
J Med. 2017;376(18):1796.
9. Vichinsky E, Hoppe CC, Ataga KI, et al; HOPE Trial Investigators. A phase
3 ran­dom­ized trial of voxelotor in sickle cell dis­ease. N Engl J Med. 2019;​
381(6):​509-519.
10. Bernaudin F, Souillet G, Vannier JP, et al. Report of the French expe­ri­ence
concerning 26 chil­dren transplanted for severe sickle cell dis­ease. Bone
Marrow Transplant. 1997;19(suppl 2):112-115.
11. Walters MC, Patience M, Leisenring W, et al. Bone mar­row trans­plan­ta­tion
for sickle cell dis­ease. N Engl J Med. 1996;335(6):369-376.
12. Al-Khabori M, Al-Ghafri F, Al-Kindi S, et al. Safety of stem cell mobi­li­za­tion in
donors with sickle cell trait. Bone Marrow Transplant. 2015;50(2):310-311.
13. Eapen M, Brazauskas R, Walters MC, et al. Effect of donor type and con­
di­tion­ing reg­i­men inten­sity on allo­ge­neic trans­plan­ta­tion out­comes in
patients with sickle cell dis­ease: a ret­ro­spec­tive multicentre, cohort study.
Lancet Haematol. 2019;6(11):e585-e596.
14. Tanhehco YC, Bhatia M. Hematopoietic stem cell trans­plan­ta­tion and cel­
lu­lar ther­apy in sickle cell dis­ease: where are we now? Curr Opin Hematol.
2019;26(6):448-452.
15. Magrin E, Miccio A, Cavazzana M. Lentiviral and genome-editing strat­e­gies
for the treat­ment of β-hemo­glo­bin­op­a­thies. Blood. 2019;134(15):1203-1213.
16. Esrick EB, Lehmann LE, Biffi A, et al. Post-tran­scrip­tional genetic silenc­ing
of BCL11A to treat sickle cell dis­ease. N Engl J Med. 2021;384(3):205-215.
17. Wilkinson AC, Dever DP, Baik R, et al. Cas9-AAV6 gene cor­rec­tion of beta-
glo­bin in autol­o­gous HSCs improves sickle cell dis­ease eryth­ro­poi­e­sis in
mice. Nat Commun. 2021;12(1):686.
18. Frangoul H, Altshuler D, Cappellini MD, et al. CRISPR-Cas9 gene editing for
sickle cell dis­ease and β-thal­as­se­mia. N Engl J Med. 2021;384(3):252-260.
19. Kanter J, Thompson AA, Mapara MY, et al. Updated results from the Hgb-
206 study in patients with severe sickle cell dis­ ease treated under a
revised pro­to­col with lentiglobin gene ther­apy using plerixafor-mobilised
haematopoietic stem cells. HemaSphere. 2019;3(suppl 1):754-755.
20. Li Y, Maule J, Neff JL, et al. Myeloid neo­plasms in the set­ting of sickle cell
dis­ease: an intrin­sic asso­ci­a­tion with the under­ly­ing con­di­tion rather than
Gene ther­apy for sickle cell dis­ease  |  179
 a coin­ci­dence: report of 4 cases and review of the lit­er­a­ture. Mod Pathol. 25. Grimley M, Asnani M, Shrestha A et  al. Early results from a phase 1/2
2019;32(12):1712-1726. study of Aru-1801 gene ther­apy for sickle cell dis­ease (SCD): manufactur­
21. Leonard A, Tisdale JF. A pause in gene ther­apy: reflecting on the unique ing pro­cess enhance­ments improve effi­cacy of a mod­i­fied gamma glo­bin
chal­lenges of sickle cell dis­ease. Mol Ther. 2021;29(4):1355-1356. len­ti­vi­rus vec­tor and reduced inten­sity con­di­tion­ing trans­plant. Blood.
22. Métais JY, Doerfler PA, Mayuranathan T, et al. Genome editing of HBG1 and 2020;136(suppl 1):20-21.
HBG2 to induce fetal hemo­glo­bin. Blood Adv. 2019;3(21):3379-3392.
23. Sullivan KM, Horwitz M, Osunkwo I, Shah N, Strouse JJ. Shared deci­sion-
mak­ing in hema­to­poi­etic stem cell trans­plan­ta­tion for sickle cell dis­ease.
Biol Blood Marrow Transplant. 2018;24(5):883-884.
24. Naik RP, Smith-Whitley K, Hassell KL, et al. Clinical out­comes asso­ci­ated
with sickle cell trait: a sys­tem­atic review. Ann Intern Med. 2018;169(9): © 2021 by The Amer­i­can Society of Hematology
619-627. DOI 10.1182/hema­tol­ogy.2021000250

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/174/1851456/174kanter.pdf by guest on 13 December 2021

180  |  Hematology 2021  |  ASH Education Program

 EMERGING THERAPIES AND CONSIDERATIONS FOR SICKLE CELL DISEASE

Hematopoietic cell transplantation for sickle cell

disease: updates and future directions

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/181/1851576/181krishnamurti.pdf by guest on 13 December 2021

Lakshmanan Krishnamurti
Aflac Cancer and Blood Disorders Center, Children’s Healthcare of Atlanta, Emory University, Atlanta, GA

Excellent outcomes in hematopoietic cell transplantation (HCT) from HLA-identical siblings, improvements in condition-
ing regimens, novel graft-versus-host disease prophylaxis, and the availability of alternative donors have all contributed
to the increased applicability and acceptability of HCT for sickle cell disease (SCD). In young children with symptomatic
SCD with an available HLA-identical related donor, HCT should be carefully considered. HCT from alternative donors is
typically undertaken only in patients with severe symptoms, causing or likely to cause organ damage, and in the context
of clinical trials. Patients undergoing HCT for SCD require careful counseling and preparation. They require careful moni-
toring of unique organ toxicities and complications during HCT. Patients must be prospectively followed for a prolonged
time to determine the long-term outcomes and late effects of HCT for SCD. Thus, there is a need for a universal, longi-
tudinal clinical registry to follow patients after HCT for SCD in conjunction with individuals who do not receive HCT to
compare outcomes. Antibody-based conditioning and ex-vivo umbilical cord blood expansion are likely to improve the
availability and acceptability of HCT. In addition, new disease-modifying drugs and the emerging option of the autol-
ogous transplantation of gene-modified hematopoietic progenitor cells are likely to expand the available therapeutic
options and make decision-making by patients, physicians, and caregivers even more complicated. Future efforts must
also focus on determining the impact of socioeconomic status on access to and outcomes of HCT and the long-term
impact of HCT on patients, families, and society.

LEARNING OBJECTIVES
• Discuss indications, conditioning, donor options, timing, outcomes, and decision­making in HCT for SCD
• Understand the impact of recipient ages and the availability of HLA­identical donors on outcomes of HCT for SCD
• Review emerging options in alternate­donor HCT for SCD.

L­glutamine has been shown to reduce the rate of VOE

CLINICAL CASE
A 12­year­old girl with HbSS­type sickle cell disease (SCD)
has been having recurrent episodes of vaso­occlusive pain
(VOE). She has been on hydroxyurea (HU) since the age of
9 months. Her clinical course remains severe despite an ade­
quate trial of L­glutamine 3 years ago and, more recently,
of voxelotor. She has an 8­year­old HLA­identical sibling.
The patient’s pediatric hematologist inquires whether she
should talk to this family about considering hematopoietic
cell transplantation (HCT).
Introduction
Comprehensive care and disease modification of SCD with
HU can decrease morbidity and organ dysfunction and
improve health­related quality of life (QoL) and survival.1
Prakash Singh Shekhawat
and related hospitalizations.2 Voxelotor has been shown
to increase the mean hemoglobin (Hb) level from baseline
compared with placebo and may be particularly useful in
individuals who have continued anemia and hemolysis.3
Crizanlizumab has been demonstrated to reduce the fre­
quency of VOE.4 HCT, however, remains the only treatment
with curative intent. When performed in young patients
from HLA­identical related donors, HCT results in excel­
lent overall survival (OS) and event­free survival (EFS).5­11
However, the lack of an available HLA­identical family
donor remains a significant limitation in the applicability of
matched sibling donor HCT. The optimization of support­
ive care, the development of novel conditioning regimens,
and the availability of the options of HCT from alternative
donors have enhanced the applicability of HCT for SCD.
However, it is a sobering fact that, despite its increase in
HCT for SCD | 181
Table 1. Risk score based on age and type of donor.

3-year prob­a­bil­ity/
inci­dence % (95% CI)
Death
Age, with­out
years Age score Type of donor Donor score Total score EFS GF Graft fail­ure ≤
≤12 0 HLA-matched sib­ling 0 0 92 (89-94)     2 (0-4) 6       (4-9)
0 HLA-matched rel­a­tive 2 2 62 (43-76)     8 (2-19) 30        (15-47)
0 Matched unre­lated donor 1 1 83 (69-91)     8 (2-18) 8 (2-18)
0 Mismatched unre­lated donor 2 2 68 (55-79)     5 (1-13) 27 (16-38)
≥13 1 HLA-matched sib­ling 0 1 87  (81-92)     7 (4-11) 5 (2-10)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/181/1851576/181krishnamurti.pdf by guest on 13 December 2021

1 HLA mismatched rel­a­tive 2 3 52 (38-65)    10   (4-18) 38 (24-51)
1 Matched unre­lated donor 1 2 50    (34-64) 29 (17-43)  21   (10-33)
1 Mismatched unre­lated donor 2 3 49 (31-66) 23 (9-40) 28 (31-44)
GF, graft fail­ure. Reproduced with per­mis­sion from Brazauskas et al.12

the last decade,11 HCT has been applied to only a tiny frac­tion of
patients with SCD, includ­ing those with severe dis­ease man­i­fes­
ta­tions. This arti­cle aims to review the cur­rent sta­tus of HCT for
SCD and address future direc­tions in the field.
Considerations in deci­sion-mak­ing about HCT for SCD
Recipient age and donor HLA match
The diag­no­sis of SCD is made at birth, and chil­dren are often well
at a young age. However, the clin­i­cal course waxes and wanes and
progresses unpre­dict­ably with age. Families can usu­ally ascer­tain
very early if their child with SCD has a poten­tial HLA-iden­ti­cal sib­
ling since sib­lings are often close in age. Thus, the cru­cial ques­tion
is, at what age and when in the clin­i­cal course should HCT be
con­sid­ered? Gluckman et al9 reported in a study com­bin­ing the
Euro­pean Society for Blood and Marrow Transplantation and Cen­
ter for International Blood and Marrow Transplant Research reg­
is­tries that out­comes of HCT from HLA-iden­ti­cal related donors
are excel­lent, but EFS decreases with increas­ing age at HCT (haz­
ard ratio [HR], 1.09; P < .001). Cappelli et al8 reported 100% OS
and 93% EFS in chil­dren under 5 years of age. Brazauskas et al12
performed an anal­y­sis of 1425 patients with SCD who under­went
HCT between 2008 and 2017. They found that patients aged 12
or youn­ger with an HLA-matched sib­ling donor had the best out­
come, with a 3-year EFS of 92%.12 Age at HCT and type of donor
were pre­dic­tive of EFS. Patients ≤12 years under­go­ing HCT from
an HLA-iden­ti­cal sib­ling are the best risk group. Patients ≤12 years
receiv­ing HCT from an unre­lated donor and patients ≥13 years
from an HLA-iden­ti­cal donor are at inter­me­di­ate risk. All other
patients are in the high-risk group (Table 1).
Indications for HCT
HCT for SCD has been performed in patients with severe SCD-
related com­pli­ca­tions. Common rea­sons to pro­ceed with HCT
include a his­tory of stroke, the need for chronic blood trans­
fu­sions with the atten­dant risks of trans­fu­sional iron over­load
and fea­tures of dis­ease sever­ity, and pre­dic­tors of pre­ma­ture
mor­tal­ity, such as recur­rent VOE and recur­rent acute chest
syn­drome.13-16 In patients enrolled in early clin­i­cal tri­als of HCT
for SCD, the most com­mon indi­ca­tions were stroke in 57% of
patients and fre­quent VOE in 23% of patients.17 More recently,
the most com­mon indi­ca­tion for HCT has been recur­rent VOE in
over 70% of cases.9,18-20
The rea­sons for the shift in indi­ca­tions for HCT are unknown. It
is pos­si­ble that with the decline in the inci­dence of stroke,21 there
are fewer patients with stroke pres­ent and con­sid­er­ing HCT. It is
also pos­si­ble that a shift has occurred in the per­cep­tion of patients,
care­giv­ers, and phy­si­cians regard­ing recur­rent VOE being an appro­
pri­ate indi­ca­tion for HCT.22-24 The num­ber of hos­pi­tal­i­za­tions or
emer­gency room vis­its for SCD-asso­ci­ated pain has long been
con­sid­ered a sur­ro­gate mea­sure of the total bur­den of pain. How­
ever, the fre­quency of health care uti­li­za­tion may be an inad­e­quate
mea­sure of the daily bur­den of pain since many patients man­age
most of their pain at home.25 A hos­pi­tal or emer­gency room visit
rep­re­sents a small frac­tion of the pain expe­ri­ence.26 Thus, some
patients with a severe bur­den of pain may not meet eli­gi­bil­ity cri­te­
ria because they do not have fre­quent health care uti­li­za­tion. Acute
inter­mit­tent vaso-occlu­sive pain is the hall­mark of SCD, but more
than half of the adults with SCD tran­si­tion from acute inter­mit­tent
pain to chronic per­sis­tent pain. Chronic pain, defined by dura­tion
as the pres­ence of pain on most days of the pre­vi­ous 6 months, is a
sig­nif­i­cant cause of mor­bid­ity and impaired QoL in SCD.27 However,
chronic pain may not affect all­indi­vid­u­als with the same degree of
dis­abil­ity and inter­fer­ence with activ­i­ties.
A def­i­ni­tion of chronic pain based on dura­tion alone does not
con­sider the mul­ti­ple dimen­sions of the con­di­tion or cap­ture the
extent of asso­ci­ated dis­abil­ity. The US National Pain Strategy has
pro­posed high-impact chronic pain (HICP) as an extreme phe­
no­type of chronic pain asso­ci­ated with severe dis­abil­ity.28 HICP,
deter­mined by screen­ing patients for fre­quent daily pain and the
pres­ence of dis­abil­ity,29 is begin­ning to be used as an eli­gi­bil­ity
cri­te­rion for HCT for SCD even in the absence of fre­quent health
care uti­li­za­tion.28,29 However, more research is required on how
to inte­grate screen­ing for HICP into clin­i­cal care and deter­mine
from elec­tronic health records if an indi­vid­ual has HICP. Post-HCT
patients with SCD show improve­ment in pain inter­fer­ence, opi­oid
use, hos­pi­tal­i­za­tion, and QoL.18,30-33 A sub­group of patients with
the pre-HCT fea­tures of sig­nif­i­cantly higher pain bur­dens, anx­i­ety,
and the use of long-act­ing opi­oids before HCT have per­sis­tent

182  |  Hematology 2021  |  ASH Education Program

chronic pain beyond 1 year post HCT.33 Thus, the iden­ti­fi­ca­tion of
HICP pre-HCT, care­ful prep­a­ra­tion of patients, includ­ing behav­
ioral health con­sul­ta­tion, and long-term mul­ti­modal reha­bil­i­ta­tion
post-HCT are cru­cial to opti­mize out­comes in these patients.
Donor con­sid­er­ations
Type of donor
The major­ity of cases of HCT for SCD reported to inter­na­tional
reg­is­tries rep­re­sent HLA-iden­ti­cal sib­ling donor HCT.11 Outcomes
of HCT from HLA-iden­ti­cal donors are supe­rior to those from alter­
nate donors,11,34 but the avail­abil­ity of HLA-iden­ti­cal donor HCT is
severely lim­ited by the lack of suit­able fam­ily donors.35 The use of
alter­nate donors to expand the donor pool for patients with SCD
HCT has been the sub­ject of inten­sive inves­ti­ga­tion. HCT from
unre­lated donors results in sta­ble donor-derived hema­to­poi­e­sis
but is asso­ci­ated with sig­nif­i­cant HCT-related mor­bid­ity and mor­
tal­ity.36 The addi­tion of costimulatory block­ade using abatacept for
graft-ver­sus-host dis­ease (GVHD) pro­phy­laxis has shown prom­ise
in miti­gat­ing the risk of severe GVHD in HCT for SCD and is the
sub­ject of an ongo­ing mul­ti­cen­ter clin­i­cal trial (NCT 03924401).37
HCT from HLA-haploidentical famil­ial donors using in-vivo or ex-
vivo T-cell deple­tion has been reported to be safe and effec­tive in
early-phase clin­i­cal tri­als and is the sub­ject of ongo­ing mul­ti­cen­ter
clin­i­cal tri­als (NCT 03263559, NCT04201210).20,38-41
Stem cell source
No study to date has com­pared the out­comes from HCT using
dif­fer­ent stem cell sources. A higher rate of nonengraftment led
to the pre­ma­ture clo­sure of the unre­lated donor (URD) umbil­
i­
cal cord blood (UCB) arm of the BMT CTN 0601 study.42 A
recent case series sug­gests that the addi­tion of thio­tepa to the
reduced-inten­sity con­di­tion­ing (RIC) reg­i­men may enhance the
engraft­ment of UCB.43 Peripheral blood as a stem cell source has
been asso­ci­ated with a higher risk of chronic GVHD (CGVHD).11
Donor char­ac­ter­is­tics
The impact of donor age in HCT for SCD has not been reported.
However, donor age has been reported as a risk fac­tor for CGVHD
fol­low­ing HCT for leu­ke­mia.44 Thus, donor age is likely sig­nif­i­cant
since GVHD does not con­vey any ben­e­fit in HCT for SCD. Increas­
ingly, patients with SCD are receiv­ing HCT from famil­ial haploiden­
tical donors, such as par­ents or older sib­lings; hence, it is cru­cial
to deter­mine the con­tri­bu­tion of donor age to out­comes. In addi­
tion, ABO major or minor incom­pat­i­bil­ity may add to the risk of
graft stem cell loss,42 delayed red cell engraft­ment decreased OS
[44], and pure red cell aplasia.45 Therefore, donor size is an impor­
tant con­sid­er­ation to ensure an ade­quate cell dose with­out com­
pro­mis­ing donor safety. For pedi­at­ric donors for HCT for SCD, the
usual prac­tice is to accept a min­i­mum donor size ≥ 10 kg and age
≥ 1 year. Donor-recip­i­ent cyto­meg­a­lo­vi­rus serostatus is matched
(D−/R− or D+/R+) when­ever pos­si­ble to min­i­mize cyto­meg­a­lo­vi­
rus dis­ease risks. High-titer donor-directed HLA antibodies may
pre­dict an increased risk for graft rejec­tion. Therefore, recip­i­ents
should be screened for the pres­ence of donor-directed HLA anti­
bodies. If high-titer antibodies are found to be pres­ent, desen­si­ti­
za­tion must be con­sid­ered before pro­ceed­ing to HCT.45,46
Conditioning reg­i­men for HCT
Most often, matched related donor (MRD) HCT has been per­
formed fol­low­ing a myeloablative con­di­tion­ing reg­i­men.9 While
Prakash Singh Shekhawat
early case series reported the use of a myeloablative com­bi­na­tion
of busul­fan and cyclo­phos­pha­mide,17 more recent series describe
the use of RIC.9,36,42,43,47,48 The most com­ mon RIC approach is
substitut­ing cyclo­phos­pha­mide with fludarabine in com­bi­na­tion
with another agent,9 usu­ally an alkylator such as busul­fan or mel­
pha­lan.48 Several reduced-tox­ic­ity con­di­tion­ing reg­i­mens have
been described, includ­ing the use of reduced doses of busul­fan
com­bined with fludarabine with or with­out cyclo­phos­pha­mide
or the sub­sti­tu­tion of busul­fan with treosulfan or with the addi­
tion of thio­tepa.18,49-52 Nonmyeloablative con­di­tion­ing with 200
cGy of total body irra­di­a­tion and fludarabine resulted in poor
long-term engraft­ment.53,54 A com­bi­na­tion of total body irra­di­
a­tion (300 cGy) with alemtuzumab resulted in high OS and EFS
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/181/1851576/181krishnamurti.pdf by guest on 13 December 2021
in adults and chil­dren.19,31,55,56 Pretransplant immu­no­sup­pres­sion
with 2 courses of fludarabine and dexa­meth­a­sone to pre­vent
graft fail­ure has been piloted in haploidentical- and URD HCT
for SCD.45,57
Prophylaxis for GVHD
The deple­tion of T cells in vivo using either antithymocyte glob­
u­lin (ATG; 70.6%) or alemtuzumab (11.5%) has been used exten­
sively in patients under­go­ing MRD HCT and may be impor­tant for
the pre­ven­tion of graft fail­ure.9 Bernaudin et al reported that the
use of ATG may decrease the rate of graft fail­ure from 22.6% to
3%.58 ATG has not been fre­quently used in MRD UCB trans­plan­ta­
tion (UCBT).59 The deple­tion of T cells in vivo with alemtuzumab
has been used in MRD HCT and URD UCBT.19,42,43,48 The deple­tion
of T cells in vivo with posttransplant cyclo­phos­pha­mide with ATG
or alemtuzumab has also been used for haploidentical HCT for
SCD.45,47,60-62 The deple­tion of T cells ex vivo with CD34+ selec­
tion,63 CD3/CD19 deple­tion,64 or α/β T-cell recep­tor and CD19
deple­tion has also been used in haploidentical HCT for SCD.65 Ex
vivo α/β t-cell recep­tor and CD19 deple­tion may reduce the risk
of GVHD but may be asso­ci­ated with delayed immune recon­
sti­tu­tion and an increased risk of infec­tion and graft fail­ure.38,65
The most com­monly used GVHD pro­phy­laxis con­sists of calci­
neurin inhib­i­tors (CNIs), which are often com­bined with meth­o­
trex­ate (MTX) or mycophenolate mofetil.9 Locatelli et al reported
that the addi­tion of MTX fol­low­ing MRD UCBT reduced EFS.59
Therefore, mycophenolate mofetil is substituted for MTX in MRD
UCBT.42,43 The addi­tion of selec­tive inhi­bi­tion of T-cell costimula­
tion with abatacept can decrease GVHD and thus improve the
safety pro­file and appli­ca­bil­ity of HCT to SCD.37
Cell dose con­sid­er­ations
For MRD UCB, a total nucle­ated cell (TNC) dose for UCB >3 × 107
TNC/kg recip­i­ent weight is pref­er­a­ble.66 If the UCB cell dose is
low, a com­bi­na­tion of UCB and bone mar­row may be used.67 In
URD UCB, TNC >5 × 107/kg increased engraft­ment and dis­ease-
free sur­vival.68 A tar­get cell dose of 4 × 108 to 5 × 108 TNC/kg for
bone mar­row and 4 × 107 to 5 × 107 TNC/kg for UCB (prethaw) is
recommended.
Engraftment fol­low­ing HCT
Risk fac­tors for graft fail­ure include HLA mis­match, high titers
of donor-directed HLA antibodies, the inten­sity of con­di­tion­ing,
or the pres­ence of active infec­tion at the time of engraft­ment.69
A com­bi­na­tion of donor chi­me­rism in the lym­phoid lin­e­age
(CD3) and mye­loid lin­e­age (CD15 or CD33), total Hb level, and
sickle cell hemo­glo­bin (HbS) per­cent­age are used to eval­u­ate the
HCT for SCD  |  183
robust­ness of engraft­ment and donor-derived hema­to­poi­e­sis.
Whole-blood donor chi­me­rism of 11% to 74% may be asso­ci­ated
with sta­ble donor-derived eryth­ro­poi­e­sis.70-72 Myeloid and lym­
phoid lin­e­age-spe­cific chi­me­rism may pro­vide addi­tional infor­ma­
tion, but ery­throid cell chi­me­rism assays are still being eval­u­ated
in research stud­ies.73,74 Stable mye­loid chi­me­rism of at least 20%
to 25% is asso­ci­ated with sta­ble donor-derived eryth­ro­poi­e­sis.18,71,72
HbS >50% sug­ gests the imma­ nence of autol­ o­
gous recov­ ery. If
there is mixed or declin­ing donor chi­me­rism or an increas­ing HbS
per­cent­age, closer mon­i­tor­ing with more fre­quent assess­ments of
donor chi­me­rism may be nec­es­sary. Donor lym­pho­cyte infu­sions
are asso­ci­ated with a sig­nif­i­cant risk of GVHD, and their role in
improv­ing donor chi­me­rism is unknown.
Most patients rejecting the allo­ graft recon­sti­
tute autol­ o­
gous hema­to­poi­e­sis,58,71 even in the case of alter­na­tive-donor
HCT.18,36,43,60,71 Marrow aplasia or prolonged cytopenia are indi­ca­
tions for an urgent sal­vage HCT.18 Consideration of a sec­ond HCT
should be deferred for at least 6 months fol­low­ing a first HCT
that has resulted in graft fail­ure with autol­o­gous recon­sti­tu­tion.
Organ func­tion con­sid­er­ations spe­cific to HCT for SCD
Neurological
In the ini­tial case series of patients who under­went MRD HCT, sei­
zures and hem­or­rhagic stroke were observed in 30% of patients.75
The risk fac­tors for neu­ro­log­i­cal com­pli­ca­tions included (1) a his­
tory of prior stroke, (2) hyper­ten­sion due to neu­ro­log­i­cal and renal
dys­func­tion exac­er­bated by CNIs and cor­ti­co­ste­roid use, (3) hem­
or­rhagic stroke in patients with preexisting cere­bral vasculopathy
with post-HCT throm­bo­cy­to­pe­nia,58 and (4) pos­te­rior revers­ible
enceph­a­lop­a­thy syn­drome (PRES) (22%-34%).36,76 There is an
increased risk of PRES in SCD patients that is exac­er­bated dur­ing
HCT and may result in decreased OS and EFS.76,77 As com­pared to
indi­vid­u­als matched for age, sex, and race, indi­vid­u­als with SCD
have lower blood pres­sure (BP).78 In patients with SCD, BP above
the 50th per­cen­tile for age may be asso­ci­ated with an increased
risk of stroke.78,79 Careful mon­i­tor­ing of BP and aggres­sive man­age­
ment of hyper­ten­sion are par­tic­u­larly impor­tant when patients are
receiv­ing both cor­ti­co­ste­roids and CNIs since both drugs are likely
to con­trib­ute to the devel­op­ment of hyper­ten­sion. In BMT CTN
0601, a mul­ti­cen­ter trial of URD blood and mar­row trans­plan­ta­tion
(BMT) for SCD, there was a high inci­dence of PRES.36 The GVHD
pro­phy­laxis in this study included the use of pred­ni­sone through
day +28. Essential pre­cau­tions to pre­vent neu­ro­log­i­cal com­pli­
ca­tions such as sei­zures or PRES include (1) starting anti­ep­i­lep­tic
sei­zure pro­phy­laxis before con­di­tion­ing, espe­cially if busul­fan is
used, and con­tinu­ing sei­zure pro­phy­laxis for the dura­tion of CNI
admin­is­tra­tion, (2) care­fully mon­i­tor­ing and strictly con­trol­ling
BP,17,75 with a tar­get BP within 10% of the median for age and sex
for SCD patients as described by Pegelow et al,78 (3) maintaining
nor­mal mag­ne­sium lev­els by mag­ne­sium sup­ple­men­ta­tion,80,81
and (4) admin­is­ter­ing pro­phy­lac­tic plate­let trans­fu­sions to main­
tain a plate­let count >50 000/µL and red blood cell trans­fu­sion as
needed to main­tain an Hb level of 9 to 11 g/dL. Post HCT, the rec­
om­men­da­tion is to con­sider obtaining brain mag­netic res­o­nance
angi­og­ra­phy/imag­ing (MRA/MRI) at 1 and 2 years after HCT and
then every 2 years as clin­i­cally indi­cated in patients with a his­tory
of stroke, moyamoya pretransplant, PRES, or another neu­ro­tox­ic­
ity dur­ing HCT.82 An age-appro­pri­ate neurocognitive eval­u­a­tion
should be obtained at 1 year. It should be repeated every 2 years if
there is a his­tory of a neu­ro­toxic com­pli­ca­tion dur­ing HCT.82
184  |  Hematology 2021  |  ASH Education Program
Cardiopulmonary
In adult SCD patients, the com­bi­na­tion of echo­car­dio­graphic
tri­cus­pid regurgitant jet veloc­ity >3.0 m/s and brain natri­uretic
pep­tide >160 pg/mL is a strong pre­dic­tor of pre­ma­ture mor­tal­
ity.83,84 Following suc­cess­ful HCT there may be an improve­ment
of tri­cus­pid regurgitant jet veloc­ity.18 Post HCT, obtaining an
echo­car­dio­gram annu­ally for 5 years is recommended. In addi­
tion, an eval­u­a­tion of car­diac iron by T2 MRI at 1 year after HCT
should be con­sid­ered in patients with mod­er­ate or severe iron
over­load at HCT.82
Unique infec­tion risks
SCD patients undergo autoinfarction of the spleen variably with
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/181/1851576/181krishnamurti.pdf by guest on 13 December 2021
age and have impair­ment of splenic func­tion and an increased
risk of pneu­mo­coc­cal sep­sis,85-87 which most often includes non­
vaccine sero­types.88 Overall, the risk of infec­tion in SCD patients
is increased due to splenic infarc­tion, defec­tive opsonization of
encap­su­lated organ­isms, impair­ment of T- and B-cell immune
func­tion, and the pres­ence of infarcted devitalized bone.89 SCD
patients may recover splenic func­tion on HU fol­low­ing the insti­
tu­tion of chronic trans­fu­sion or fol­low­ing suc­cess­ful HCT.90 How­
ever, patients who are older at the time of HCT and those with
exten­sive CGVHD are at higher risk of poor post-HCT splenic
recov­ery.91 While pneu­mo­coc­cal infec­tions are rare fol­low­ing
HCT for SCD,91 deaths due to over­whelm­ing pneu­mo­coc­cal sep­
sis have been reported. Therefore, pneu­mo­coc­cal pro­phy­laxis,
care­ful mon­i­tor­ing of splenic func­tion recov­ery post HCT, and
timely reimmunization starting with con­ju­gated pneu­mo­coc­cal
vac­cines 6 months post trans­plant are impor­tant for the pre­ven­
tion of seri­ous pneu­mo­coc­cal infec­tions.92
Management of trans­fu­sional iron over­load
Patients may have trans­fu­sional iron over­load prior to HCT and
may have also received sev­eral trans­fu­sions of packed red blood
cells with the HCT. Patients who have received a life­time trans­
fu­sion bur­den ≥10 trans­fu­sions and serum fer­ri­tin ≥1500 ng/mL
are eval­u­ated pre-HCT for liver iron over­load and liver fibro­sis.
MR of the abdo­men is performed for quan­ti­fic ­ a­tion of iron. In
addi­tion, ultra­sound or MR elastography may be performed to
assess the degree of fibro­sis. In patients with evi­dence of severe
iron over­load or liver fibro­sis, a hepatology con­sul­ta­tion and liver
biopsy may be con­sid­ered for deter­min­ing the safety of pro­
ceed­ing with HCT. Patients with cir­rho­sis, bridg­ing fibro­sis, or
active hep­at­i­tis are too high risk and may not tol­er­ate an HCT
con­di­tion­ing reg­i­men.93 In patients with base­line iron over­load,
the removal of excess body iron stores must be insti­tuted with
oral iron che­la­tion, by monthly phle­bot­omy, or by a com­bi­na­tion
thereof after mea­sure­ment of resid­ual iron over­load by serum fer­
ri­tin and MRI.94-97 Iron che­la­tion may be ini­ti­ated post HCT when
patients are off immu­no­sup­pres­sion, have no evi­dence of GVHD
or drug-induced liver dam­age, and are trans­fu­sion inde­pen­dent.
Renal
Serum blood urea nitro­gen/cre­at­i­nine and glo­mer­u­lar fil­tra­
tion rate or 24-hour cre­at­i­nine clear­ance and SCD-asso­ci­ated
pro­tein­uria are crit­i­cal con­sid­er­ations pre-HCT and fol­low-ups
for recov­ ery post HCT.82 A prolonged course of CNIs places
patients at a high risk of renal dys­func­tion. Patients with SCD
with prior com­ ple­
ment-medi­ ated vas­ cu­lar injury due to their
under­ly­ing pri­mary hemo­lytic dis­ease and addi­tional stress­ors
dur­ing the HCT pro­cess may develop pro­gres­sive endo­the­lial
injury and end-organ dys­func­tion.98 Case reports and small case
series of trans­plant-asso­ci­ated throm­botic microangiopathy in
patients with SCD and fol­low­ing HCT sug­gest the need to mon­
i­tor trans­plant-asso­ci­ated throm­botic microangiopathy, con­
trol hyper­ten­sion, con­sider alter­na­tives to CNIs, and treat with
eculizumab when appro­pri­ate.99-103
Gonadal dam­age and impaired fer­til­ity
HCT recip­i­ents for SCD are at high risk of gonadal dam­age and
infer­til­ity.104 However, data are lacking on the patient and care­
giver per­spec­tive on the impor­tance of fer­til­ity in mak­ing deci­
sions about HCT, the costs of and access to fer­til­ity pres­er­va­tion,
and the emo­tional and psy­cho­log­i­cal impact of gonadal dam­age
on the long-term sur­vi­vors of HCT.105 Sperm bank­ing in males and
oocyte cryo­pres­er­va­tion in females are stan­dard pro­ce­dures
that can mit­i­gate the risk of infer­til­ity.106 Ovarian tis­sue bank­ing
is an accept­able fer­til­ity-pres­er­va­tion tech­nique that is no lon­
ger con­sid­ered exper­i­men­tal and is the only method to pre­serve
fer­til­ity for pre­pu­ber­tal girls.106 Testicular tis­sue cryo­pres­er­va­tion
in pre­pu­ber­tal boys is under inves­ti­ga­tion.107,108 Fertility pres­
er­va­tion pro­ce­dures pre-HCT must also con­sider that HU may
cause gonadal dam­age and have an adverse impact on male and
female fer­til­ity in SCD patients.109,110
Med­ic­aid cov­er­age of fer­til­ity pres­er­va­tion is extremely lim­
ited both in the scope of ben­e­fits and the num­ber of states that
require such a ben­e­fit.111 For exam­ple, only 15 states require fer­til­
ity pres­er­va­tion cov­er­age in pri­vate insur­ance plans, and 5 states
extend this ben­e­fit only to females. In addi­tion, state stat­utes
pro­vide var­i­able cov­er­age based on mar­i­tal sta­tus, diag­no­sis,
length of fer­til­ity prob­lems, and the mon­e­tary limit of the ben­e­fit.
Long-term and late effects of HCT
HCT can sta­bi­lize organ func­tion and ame­lio­rate man­i­fes­ta­tions
of SCD; patients need sys­tem­atic fol­low-up of mon­i­tor­ing SCD
com­pli­ca­tions and the long-term and late effects of HCT accord­
ing to established con­sen­sus guide­lines.82,112 Further, most of the
published lit­er­a­ture on the late effects of HCT for SCD is based
on HCT for malig­nan­cies. There is, there­fore, a need for a uni­ver­
sal, lon­gi­tu­di­nal clin­i­cal reg­is­try to fol­low out­comes after HCT
for SCD.
Values and pref­er­ences of patients, fam­i­lies,
and care­giv­ers
The deci­sion to undergo HCT for SCD involves mak­ing com­plex
trade-offs between the prom­ise of relief of dis­ease man­i­fes­ta­
tions, sta­bi­li­za­tion of organ func­tion, halt­ing of dis­ease pro­gres­
sion, and improve­ment of lon­gev­ity and QoL, on the one hand,
with sub­stan­tial treat­ment bur­den and mor­bid­ity in the short
term, and the risk of new long-term sequelae such as CGVHD,
infer­til­ity, and sub­se­quent malig­nancy, on the other hand.24 Many
fac­tors are involved in patient and care­giver deci­sion-mak­ing for
SCD.24,113,114 These include the unpre­dict­able onset and pro­gres­
sion of dis­ease com­pli­ca­tions, the repeated need to seek health
care, a poor QoL, a wors­en­ing of dis­ease-related com­pli­ca­tions,
the need to make major ther­a­peu­tic deci­sions, and con­cerns
about the long-term con­se­quences of SCD. Families that are
aware of and have access to HCT, have strong fam­ily sup­port,
and have an avail­­able HLA-iden­ti­cal sib­ling donor are more likely
to con­sider HCT.113 Overall, fam­i­lies face deci­sional con­flict at
Prakash Singh Shekhawat
lev­els that result in deci­sional delays or uncertainties.114 The per­
cep­tion that the cur­rent dis­ease bur­den has become unac­cept­
able is a com­mon con­sid­er­ation for HCT for SCD among patients
and their phy­si­cians.22-24,113 The avail­abil­ity of an HLA-iden­ti­cal sib­
ling donor may be viewed as prov­i­den­tial by par­ents and is a
sig­nif­i­cant fac­tor when con­sid­er­ing HCT. Some patients or care­
giv­ers will not con­sider HCT at any level of risk.115,116 Almost three-
quar­ters of patients report being will­ing to take a mod­est risk of
≥ 5% mor­tal­ity, whereas 57% are will­ing to take the ≥ 10% risk of
GVHD. Following a suc­cess­ful HCT, SCD patients and care­giv­ers
do not report deci­sional regret, even with active CGVHD.113
The phy­si­cian per­spec­tive on deci­sion-mak­ing
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/181/1851576/181krishnamurti.pdf by guest on 13 December 2021
Their past expe­ri­ences and the out­comes of pre­vi­ous patients
may influ­ence how phy­si­cians respond in mak­ing rec­om­men­da­
tions to patients about seek­ing HCT con­sul­ta­tion. For exam­ple,
Bakshi et al, in a qual­i­ta­tive study of phy­si­cian deci­sion-mak­ing
in dis­ease-mod­i­fy­ing ther­apy for SCD, found that the phy­si­cian’s
per­cep­tion of the sever­ity of dis­ease and the abil­ity of a given
patient to adhere to the med­i­ca­tion and treat­ment reg­i­men is also
a sig­nif­i­cant con­sid­er­ation in their recommending HCT for SCD.23
Socioeconomic sta­tus and HCT for SCD
Racial and socio­ eco­ nomic disparities may adversely affect
access to and out­comes of HCT in minor­ity pop­u­la­tions.117,118
There is also a sub­ stan­
tial regional dis­ par­
ity in donor search
cov­er­age, trans­plant pro­ce­dures, hos­pi­tal­i­za­tions, med­i­ca­tions,
transportation, and lodg­ing pro­vided by dif­fer­ent states’ Med­
ic­aid pro­grams for HCT.119 Mupfudze et al have described the
bur­den of nav­i­gat­ing eli­gi­bil­ity for Med­ic­aid, the peer review
pro­cess for var­i­ous com­po­nents and pro­ce­dures for HCT, the dif­
fer­ence between fee-for-ser­vice and com­pre­hen­sively man­aged
plans, and the bur­den that HCT poses to fam­i­lies seek­ing HCT for
SCD.120 Patients enrolled in Med­ic­aid had a lower EFS (HR, 2.36;
95% CI, 1.44-3.85; P = .0006) and a higher cumu­la­tive inci­dence of
graft fail­ure fol­low­ing HCT (HR, 2.57; 95% CI, 1.43-4.60; P = .0015)
com­pared to pri­vately insured patients with SCD.121 Registry data
pro­vide no clue to the fac­tors con­trib­ut­ing to worse out­comes
among Med­ic­aid-insured patients. Afri­can Amer­i­cans, His­pan­ics,
and indi­vid­u­als with Med­ic­aid cov­er­age have been dem­on­strated
to have worse out­comes fol­low­ing treat­ment for hema­to­log­i­cal
malig­nan­cies and fol­low­ing HCT.117,118,122-124 These find­ings pro­vide
the ratio­nale for fur­ther study of the com­plex inter­ac­tion of psy­
cho­so­cial func­tion­ing, health behav­iors, racial and eth­nic dispar­
ities, poor socio­eco­nomic sta­tus, and health care disparities on
out­comes of HCT in patients with SCD.117,118,122-124
Future direc­tions
Despite prog­ress in the con­di­tion­ing reg­i­mens, donor options,
GVHD pro­phy­laxis, and out­comes of allo­ge­neic HCT over the last
25 years, HCT has been applied to a tiny pro­por­tion of patients
with SCD. Advances in sev­eral fields, includ­ing ame­lio­ra­tion of the
risk of mor­bid­ity, mor­tal­ity, and long-term sequelae, are nec­es­
sary to increase the appli­ca­bil­ity and accept­abil­ity of HCT for SCD.
Antibody-based myeloablation approaches to trans­plant con­di­
tion­ing through the use of anti-c-kit antibodies, anti-c-kit anti­body
con­ju­gated to the drug saporin, a ribo­some-inactivating pro­tein
with potent cell-cycle-inde­pen­dent cyto­toxic activ­ity, and c-kit
targeted chi­me­ric anti­gen recep­tor T cells can reduce or elim­i­nate
the need for che­mo­ther­apy-based con­di­tion­ing reg­i­mens.125-131
HCT for SCD  |  185
A nonchemotherapy con­di­tion­ing reg­i­men can fur­ther reduce
acute mor­bid­ity related to mucositis and mar­row aplasia, gonadal
tox­ic­ity, and sub­se­quent malig­nancy related to alkylators. The
refine­ment of GVHD pro­phy­laxis using posttransplant cyclo­phos­
pha­mide or costimulatory block­ade with abatacept can reduce
acute GVHD and CGVHD and their sequelae. UCB is lim­ited by
both the avail­abil­ity of HLA-matched donors and the lim­ited cell
dose of avail­­able UCB units. Ex-vivo expan­sion of the cell dose of
UCB units is likely to make more and pos­si­bly bet­ter HLA-matched
units avail­­able. Advances in stem cell expan­sion and stem cell
engi­neer­ing, includ­ing iden­ti­fy­ing crit­i­cal cyto­kines, pro­teins, and
small drug ago­nists, have led to clin­i­cal pro­to­cols for expanding
hema­to­poi­etic stem cells and improved engraft­ment. URD UCBT
with omidubicel, a nic­o­tin­amide-based, ex vivo-expanded UCB
prod­uct, is asso­ci­ated with rapid engraft­ment but a high rate of
GVHD in patients with SCD.132 Ongoing clin­i­cal tri­als can refine
the ex-vivo expan­sion of hema­to­poi­etic stem cells, with future
appli­ca­tions to SCD.133 Pharmacokinetics-directed pre­ci­sion dos­
ing can improve the safety and effi­cacy of in-vivo T-cell deple­tion
with alemtuzumab and myeloablation with mel­pha­lan.134,135
Impaired fer­til­ity is an unin­tended con­se­quence of the con­di­
tion­ing reg­i­men. Improving the accept­abil­ity of HCT for SCD will
require advo­cacy with state gov­ern­ments to expand access to a
fer­til­ity pres­er­va­tion ben­e­fit, remove the spou­sal require­ment,
expand the diag­nos­tic cri­te­ria, and include the ben­e­fit in Med­ic­aid
plans.111 Further study of improv­ing the aware­ness of, access to, and
out­comes of HCT in under­served pop­u­la­tions can help us bet­ter
under­stand how to make HCT promptly avail­­able to a greater pro­
por­tion of SCD patients who could ben­e­fit from the pro­ce­dure.
The advances in HCT are occur­ring con­tem­po­ra­ne­ously with
explo­sive growth in clin­i­cal tri­als of novel dis­ease-mod­i­fy­ing
ther­a­pies, on the one hand, and gene ther­apy strat­e­gies, on the
other hand. How the inter­ac­tion of these devel­op­ments is likely
to affect the future sta­tus of HCT in the man­age­ment of SCD is
unknown. Still, it is safe to say that we have embarked on a jour­
ney of increased and improved ther­a­peu­tic options for patients
with SCD.
CLINICAL CASE (Con­t in­u ed)
This 12-year-old girl with HbSS with fea­tures of severe dis­ease
phe­no­type is at risk for the long-term com­pli­ca­tions of SCD.
Her clin­i­cal course has remained severe despite a trial of HU,
L-glu­ta­mine, and voxelotor. She has excel­lent risk fac­tors for
HCT from her HLA-iden­ti­cal sib­ling. I would rec­om­mend that
the hema­tol­o­gist have a series of con­ver­sa­tions with the fam­ily
to edu­cate them about HCT and then refer them to HCT con­
sul­ta­tion and explore research pro­ce­dures for ovar­ian tis­sue
pres­er­va­tion. If the fam­ily pro­vi­des informed con­sent, we will
pro­ceed to HCT with a pretransplant trans­fu­sion to lower HbS.
During HCT we will main­tain tar­get Hb and plate­let counts by
trans­fu­sions, care­fully man­age BP, and main­tain nor­mal serum
mag­ne­sium lev­els to pre­vent PRES.
Conflict-of-inter­est dis­clo­sure
Lakshmanan Krishnamurti: no com­pet­ing finan­cial inter­ests to
declare.
186  |  Hematology 2021  |  ASH Education Program
Off-label drug use
Lakshmanan Krishnamurti: nothing to disclose.
Correspondence
Lakshmanan Krishnamurti, Children’s Healthcare of Atlanta-Egleston,
1405 Clifton Road NE, Atlanta, GA 30322; e-mail: lakshmanan.
krishnamurti@emory​­.edu
References
1. McGann PT, Ware RE. Hydroxyurea ther­apy for sickle cell ane­mia. Expert
Opin Drug Saf. 2015;14(11):1749-1758.
2. Cieri-Hutcherson NE, Hutcherson TC, Conway-Habes EE, Burns BN, White
NA. Systematic review of l-glu­ta­mine for pre­ven­tion of vaso-occlu­sive pain
cri­sis in patients with sickle cell dis­ease. Pharmacotherapy. 2019;39(11):
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/181/1851576/181krishnamurti.pdf by guest on 13 December 2021
1095-1104.
3. Blair HA. Voxelotor: first approval. Drugs. 2020;80(2):209-215.
4. Blair HA. Crizanlizumab: first approval. Drugs. 2020;80(1):79-84.
5. Sahdev I, Brochstein J, Werther N, Stiles J. Timing of alemtuzumab with
respect to day of bone mar­row infu­sion and its effects upon engraft­ment
and graft-ver­sus-host dis­ease in patients with sickle cell dis­ease: a sin­gle-
insti­tu­tionalstudy. J Pediatr Hematol Oncol. 2020;42(8):e718-e722.
6. Bernaudin F, Dalle JH, Bories D, et al; Société Française de Greffe de Moelle
et de Thérapie Cellulaire. Long-term event-free sur­ vival, chi­
me­ rism and
fer­til­ity out­comes in 234 patients with sickle-cell ane­mia youn­ger than 30
years after myeloablative con­di­tion­ing and matched-sib­ling trans­plan­ta­tion
in France. Haematologica. 2020;105(1):91-101.
7. Saraf SL, Rondelli D. Allogeneic hema­to­poi­etic stem cell trans­plan­ta­tion for
adults with sickle cell dis­ease. J Clin Med. 2019;8(10):1565.
8. Cappelli B, Volt F, Tozatto-Maio K, et al; for Eurocord, the Cellular Therapy
and Immunobiology Working Party, and the Paediatric Diseases Working Party
of the Euro­pean Society for Blood and Marrow Transplantation. Risk fac­tors
and out­comes according to age at trans­plan­ta­tion with an HLA-iden­ti­cal
sib­ling for sickle cell dis­ease. Haematologica. 2019;104(12):e543-e546.
9. Gluckman E, Cappelli B, Bernaudin F, et al; for Eurocord, the Pediatric Work­
ing Party of the Euro­pean Society for Blood and Marrow Transplantation,
and the Center for International Blood and Marrow Transplant Research.
Sickle cell dis­ease: an inter­na­tional sur­vey of results of HLA-iden­ti­cal sib­ling
hema­to­poi­etic stem cell trans­plan­ta­tion. Blood. 2017;129(11):1548-1556.
10. Walters MC, De Castro LM, Sullivan KM, et al. Indications and results of HLA-
iden­ti­cal sib­ling hema­to­poi­etic cell trans­plan­ta­tion for sickle cell dis­ease.
Biol Blood Marrow Transplant. 2016;22(2):207-211.
11. Eapen M, Brazauskas R, Walters MC, et al. Effect of donor type and con­
di­tion­ing reg­i­men inten­sity on allo­ge­neic trans­plan­ta­tion out­comes in
patients with sickle cell dis­ease: a ret­ro­spec­tive multicentre, cohort study.
Lancet Haematol. 2019;6(11):e585-e596.
12. Brazauskas R, Scigliuolo GM, Wang HL, et al. Risk score to pre­dict event-
free sur­ vival after hema­ to­
poi­etic cell trans­ plant for sickle cell dis­ ease.
Blood. 2020;136(5):623-626.
13. Brandow AM, Panepinto JA. Hydroxyurea use in sickle cell dis­ease: the bat­
tle with low pre­scrip­tion rates, poor patient com­pli­ance and fears of toxic­
ities. Expert Rev Hematol. 2010;3(3):255-260.
14. Darbari DS, Wang Z, Kwak M, et al. Severe pain­ful vaso-occlu­sive cri­ses and
mor­tal­ity in a con­tem­po­rary adult sickle cell ane­mia cohort study. PLoS
One. 2013;8(11):e79923.
15. Quinn CT, Lee NJ, Shull EP, Ahmad N, Rogers ZR, Buchanan GR. Prediction
of adverse out­comes in chil­dren with sickle cell ane­mia: a study of the
Dallas Newborn Cohort. Blood. 2008;111(2):544-548.
16. Prasad R, Hasan S, Castro O, Perlin E, Kim K. Long-term out­ comes in
patients with sickle cell dis­ease and fre­quent vaso-occlu­sive cri­ses. Am J
Med Sci. 2003;325(3):107-109.
17. Walters MC, Patience M, Leisenring W, et al. Bone mar­row trans­plan­ta­tion
for sickle cell dis­ease. N Engl J Med. 1996;335(6):369-376.
18. Krishnamurti L, Neuberg DS, Sullivan KM, et al. Bone mar­row trans­plan­ta­
tion for ado­les­cents and young adults with sickle cell dis­ease: results of a
pro­spec­tive mul­ti­cen­ter pilot study. Am J Hematol. 2019;94(4):446-454.
19. Hsieh MM, Fitzhugh CD, Weitzel RP, et al. Nonmyeloablative HLA-matched
sib­ling allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion for severe sickle
cell phe­no­type. JAMA. 2014;312(1):48-56.
20. Bolaños-Meade J, Cooke KR, Gamper CJ, et al. Effect of increased dose of
total body irra­di­a­tion on graft fail­ure asso­ci­ated with HLA-haploidentical
trans­plan­ta­tion in patients with severe haemoglobinopathies: a ­pro­spec­tive
clin­i­cal trial. Lancet Haematol. 2019;6(4):e183-e193.
21. Fullerton HJ, Adams RJ, Zhao S, Johnston SC. Declining stroke rates in Cali­
fornian chil­dren with sickle cell dis­ease. Blood. 2004;104(2):336-339.
22. Bakshi N, Katoch D, Sinha CB, et al. Assessment of patient and care­giver
atti­tudes and approaches to deci­ sion-mak­ ing regard­ ing bone mar­ row
trans­plant for sickle cell dis­ease: a qual­i­ta­tive study. JAMA Netw Open.
2020;3(5):e206742.
23. Bakshi N, Sinha CB, Ross D, Khemani K, Loewenstein G, Krishnamurti L. Pro­
ponent or col­lab­o­ra­tive: phy­si­cian per­spec­tives and approaches to dis­ease
mod­i­fy­ing ther­a­pies in sickle cell dis­ease. PLoS One. 2017;12(7):e0178413.
24. Sinha CB, Bakshi N, Ross D, Loewenstein G, Krishnamurti L. Primary care­
giver deci­sion-mak­ing in hema­to­poi­etic cell trans­plan­ta­tion and gene ther­
apy for sickle cell dis­ease. Pediatr Blood Cancer. 2021;68(1):e28749.
25. Smith WR, Penberthy LT, Bovbjerg VE, et  al. Daily assess­ment of pain in
adults with sickle cell dis­ease. Ann Intern Med. 2008;148(2):94-101.
26. Smith WR, Bovbjerg VE, Penberthy LT, et  al. Understanding pain and
improv­ing man­age­ment of sickle cell dis­ease: the PiSCES study. J Natl Med
Assoc. 2005;97(2):183-193.
27. Field JJ, Ballas SK, Campbell CM, et al. AAAPT diag­nos­tic cri­te­ria for acute
sickle cell dis­ease pain. J Pain. 2019;20(7):746-759.
28. Von Korff M, Scher AI, Helmick C, et al. United States national pain strat­
egy for pop­u­la­tion research: con­cepts, def­i­ni­tions, and pilot data. J Pain.
2016;17(10):1068-1080.
29. Pitcher MH, Von Korff M, Bushnell MC, Porter L. Prevalence and pro­file of
high-impact chronic pain in the United States. J Pain. 2019;20(2):146-160.
30. Han J, Holden CC, Ahluwalia AY, et al. Chronic opi­oid use can be reduced
or discontinued after haematopoietic stem cell trans­plan­ta­tion for sickle
cell dis­ease. Br J Haematol. 2020;191(3):e70-e72.
31. Saraf SL, Oh AL, Patel PR, et al. Nonmyeloablative stem cell trans­plan­ta­
tion with alemtuzumab/low-dose irra­di­a­tion to cure and improve the qual­
ity of life of adults with sickle cell dis­ease. Biol Blood Marrow Transplant.
2016;22(3):441-448.
32. Badawy SM, Beg U, Liem RI, Chaudhury S, Thompson AA. A sys­tem­atic
review of qual­ity of life in sickle cell dis­ease and thal­as­se­mia after stem cell
trans­plant or gene ther­apy. Blood Adv. 2021;5(2):570-583.
33. Darbari DS, Liljencrantz J, Ikechi A, et al. Pain and opi­oid use after rever­sal
of sickle cell dis­ease fol­low­ing HLA-matched sib­ling haematopoietic stem
cell trans­plant. Br J Haematol. 2019;184(4):690-693.
34. Gluckman E, Cappelli B, Scigliuolo GM, De la Fuente J, Corbacioglu S. Alter­
native donor hema­to­poi­etic stem cell trans­plan­ta­tion for sickle cell dis­ease
in Europe. Hematol Oncol Stem Cell Ther. 2020;13(4):181-188.
35. Mentzer WC, Heller S, Pearle PR, Hackney E, Vichinsky E. Availability of
related donors for bone mar­row trans­plan­ta­tion in sickle cell ane­mia. Am J
Pediatr Hematol Oncol. 1994;16(1):27-29.
36. Shenoy S, Eapen M, Panepinto JA, et  al. A trial of unre­lated donor mar­
row trans­plan­ta­tion for chil­dren with severe sickle cell dis­ease. Blood.
2016;128(21):2561-2567.
37. Ngwube A, Shah N, Godder K, Jacobsohn D, Hulbert ML, Shenoy S. Abata­
cept is effec­tive as GVHD pro­phy­laxis in unre­lated donor stem cell trans­
plan­ta­tion for chil­dren with severe sickle cell dis­ease. Blood Adv. 2020;4(16):
3894-3899.
38. Foell J, Kleinschmidt K, Jakob M, Troeger A, Corbacioglu S. Alternative donor:
αß/CD19 T-cell-depleted haploidentical hema­to­poi­etic stem cell trans­plan­
ta­tion for sickle cell dis­ease. Hematol Oncol Stem Cell Ther. 2020;13(2):
98-105.
39. Foell J, Schulte JH, Pfirstinger B, et  al. Haploidentical CD3 or α/β T-cell
depleted HSCT in advanced stage sickle cell dis­ease. Bone Marrow Trans-
plant. 2019;54(11):1859-1867.
40. Patel DA, Akinsete AM, Connelly JA, Kassim AA. T-cell deplete ver­sus T-cell
replete haploidentical hema­to­poi­etic stem cell trans­plan­ta­tion for sickle
cell dis­ease: where are we? Expert Rev Hematol. 2019;12(9):733-752.
41. Patel DA, Dhedin N, Chen H, et  al. Early viral reactivation despite excel­
lent immune recon­sti­tu­tion fol­low­ing haploidentical bone mar­row trans­
plant with post-trans­plant cytoxan for sickle cell dis­ease. Transpl Infect
Dis. 2020;22(1):e13222.
42. Kamani NR, Walters MC, Carter S, et al. Unrelated donor cord blood trans­
plan­ta­tion for chil­dren with severe sickle cell dis­ease: results of one cohort
from the phase II study from the Blood and Marrow Transplant Clinical
Trials Network (BMT CTN). Biol Blood Marrow Transplant. 2012;18(8):1265-
1272.
43. Abra­ham A, Cluster A, Jacobsohn D, et al. Unrelated umbil­i­cal cord blood
trans­plan­ta­tion for sickle cell dis­ease fol­low­ing reduced-inten­sity con­di­
tion­ing: results of a phase I trial. Biol Blood Marrow Transplant. 2017;23(9):
1587-1592.
Prakash Singh Shekhawat
44. Watkins BK, Horan J, Storer B, Martin PJ, Carpenter PA, Flowers ME. Recip­
ient and donor age impact the risk of devel­op­ing chronic GvHD in chil­
dren after allo­ge­neic hema­to­poi­etic trans­plant. Bone Marrow Transplant.
2017;52(4):625-626.
45. Pawlowska AB, Cheng JC, Karras NA, et  al. HLA haploidentical stem cell
trans­plant with pretransplant immu­no­sup­pres­sion for patients with sickle
cell dis­ease. Biol Blood Marrow Transplant. 2018;24(1):185-189.
46. Ciurea SO, Cao K, Fernandez-Vina M, et al. The Euro­pean Society for Blood
and Marrow Transplantation (EBMT) Consensus Guidelines for the detec­
tion and treat­ment of donor-spe­cific anti-HLA antibodies (DSA) in hap­
loidentical hema­to­poi­etic cell trans­plan­ta­tion. Bone Marrow Transplant.
2018;53(5):521-534.
47. Wiebking V, Hütker S, Schmid I, Immler S, Feuchtinger T, Albert MH. Reduced
tox­ic­ity, myeloablative HLA-haploidentical hema­to­poi­etic stem cell trans­
plan­ta­tion with post-trans­plan­ta­tion cyclo­phos­pha­mide for sickle cell dis­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/181/1851576/181krishnamurti.pdf by guest on 13 December 2021
ease. Ann Hematol. 2017;96(8):1373-1377.
48. King AA, Kamani N, Bunin N, et al. Successful matched sib­ling donor mar­
row trans­plan­ta­tion fol­low­ing reduced inten­sity con­di­tion­ing in chil­dren
with hemo­glo­bin­op­a­thies. Am J Hematol. 2015;90(12):1093-1098.
49. Gyurkocza B, Sandmaier BM. Conditioning reg­i­mens for hema­to­poi­etic cell
trans­plan­ta­tion: one size does not fit all. Blood. 2014;124(3):344-353.
50. Bacigalupo A, Ballen K, Rizzo D, et al. Defining the inten­sity of con­di­tion­
ing reg­i­mens: work­ing def­i­ni­tions. Biol Blood Marrow Transplant. 2009;15
(12):1628-1633.
51. Horan JT, Haight A, Dioguardi JL, et al. Using fludarabine to reduce expo­sure
to alkylating agents in chil­dren with sickle cell dis­ease receiv­ing busul­fan,
cyclo­phos­pha­mide, and antithymocyte glob­u­lin trans­plant con­di­tion­ing:
results of a dose de-esca­la­tion trial. Biol Blood Marrow Transplant.2015;21(5):
900-905.
52. Strocchio L, Zecca M, Comoli P, et al. Treosulfan-based con­di­tion­ing reg­i­
men for allo­ge­neic haematopoietic stem cell trans­plan­ta­tion in chil­dren with
sickle cell dis­ease. Br J Haematol. 2015;169(5):726-736.
53. Iannone R, Casella JF, Fuchs EJ, et  al. Results of min­i­mally toxic nonmy­
eloablative trans­plan­ta­tion in patients with sickle cell ane­mia and beta-
thal­as­se­mia. Biol Blood Marrow Transplant. 2003;9(8):519-528.
54. Horan JT, Liesveld JL, Fenton P, Blumberg N, Walters MC. Hematopoietic
stem cell trans­plan­ta­tion for mul­ti­ply trans­fused patients with sickle cell
dis­ease and thal­as­se­mia after low-dose total body irra­di­a­tion, fludarabine,
and rab­bit anti-thy­mo­cyte glob­u­lin. Bone Marrow Transplant. 2005;35(2):
171-177.
55. Guilcher GM, Monagel D, Leaker M, et al. Alemtuzumab/low dose TBI con­
di­tion­ing facil­i­tates sta­ble long-term donor hema­to­poi­etic cell engraft­
ment from sib­ling donors in chil­dren with sickle cell dis­ease. Blood. 2017;
130(suppl 1):4592.
56. Guilcher GMT, Monagel DA, Nettel-Aguirre A, et  al. Nonmyeloablative
matched sib­ling donor hema­to­poi­etic cell trans­plan­ta­tion in chil­dren and
ado­les­cents with sickle cell dis­ease. Biol Blood Marrow Transplant. 2019;
25(6):1179-1186.
57. Kharya G, Bakane AN, Rauthan AM. Pretransplant mye­loid and immune
sup­pres­sion, reduced tox­ic­ity con­di­tion­ing with posttransplant cyclo­phos­
pha­mide: ini­tial out­comes of novel approach for matched unre­lated donor
hema­to­poi­etic stem cell trans­plant for hemo­glo­bin­op­a­thies. Pediatr Blood
Cancer. 2021;68(4):e28909.
58. Bernaudin F, Socie G, Kuentz M, et al; for the Société Française de Greffe
de Moelle et de Thérapie Cellulair. Long-term results of related myeloab­
lative stem-cell trans­plan­ta­tion to cure sickle cell dis­ease. Blood. 2007;
110(7):2749-2756.
59. Locatelli F, Kabbara N, Ruggeri A, et al; Eurocord; Euro­pean Society for Blood
and Marrow Transplantation. Outcome of patients with hemo­glo­bin­op­a­thies
given either cord blood or bone mar­row trans­plan­ta­tion from an HLA-iden­ti­cal
sib­ling. Blood. 2013;122(6):1072-1078.
60. Bolaños-Meade J, Fuchs EJ, Luznik L, et al. HLA-haploidentical bone mar­
row trans­plan­ta­tion with posttransplant cyclo­phos­pha­mide expands the
donor pool for patients with sickle cell dis­ease. Blood. 2012;120(22):4285-
4291.
61. Fitzhugh CD, Hsieh MM, Taylor T, et  al. Cyclophosphamide improves
engraft­ment in patients with SCD and severe organ dam­age who undergo
haploidentical PBSCT. Blood Adv. 2017;1(11):652-661.
62. de la Fuente J, Dhedin N, Koyama T, et  al. Haploidentical bone mar­row
trans­plan­ta­tion with post-trans­plan­ta­tion cyclo­phos­pha­mide plus thio­tepa
improves donor engraft­ment in patients with sickle cell ane­mia: results
of an inter­na­tional learn­ing col­lab­o­ra­tive. Biol Blood Marrow Transplant.
2019;25(6):1197-1209.
HCT for SCD  |  187
63. Gilman AL, Eckrich MJ, Epstein S, et  al. Alternative donor hema­to­poi­etic
stem cell trans­plan­ta­tion for sickle cell dis­ease. Blood Adv. 2017;1(16):1215-
1223.
64. Foell J, Pfirstinger B, Rehe K, Wolff D, Holler E, Corbacioglu S. Haploidenti­
cal stem cell trans­plan­ta­tion with CD3+−/CD19+− depleted periph­eral stem
cells for patients with advanced stage sickle cell dis­ease and no alter­na­tive
donor: results of a pilot study. Bone Marrow Transplant. 2017;52(6):938-
940.
65. Gaziev J, Isgrò A, Sodani P, et al. Haploidentical HSCT for hemo­glo­bin­op­
a­thies: improved out­comes with TCRαβ+/CD19+-depleted grafts. Blood
Adv. 2018;2(3):263-270.
66. Locatelli F, Pagliara D. Allogeneic hema­to­poi­etic stem cell trans­plan­ta­tion
in chil­dren with sickle cell dis­ease. Pediatr Blood Cancer. 2012;59(2):372-
376.
67. Soni S, Boulad F, Cowan MJ, et  al. Combined umbil­i­cal cord blood and
bone mar­row from HLA-iden­ti­cal sib­ling donors for hema­to­poi­etic stem
cell trans­plan­ta­tion in chil­dren with hemo­glo­bin­op­a­thies. Pediatr Blood
Cancer. 2014;61(9):1690-1694.
68. Ruggeri A, Eapen M, Scaravadou A, et al; Eurocord Registry; Center for
International Blood and Marrow Transplant Research; New York Blood Cen­
ter. Umbilical cord blood trans­plan­ta­tion for chil­dren with thal­as­se­mia and
sickle cell dis­ease. Biol Blood Marrow Transplant. 2011;17(9):1375-1382.
69. Olsson R, Remberger M, Schaffer M, et al. Graft fail­ure in the mod­ern era
of allo­ge­neic hema­to­poi­etic SCT. Bone Marrow Transplant. 2013;48(4):
537-543.
70. Walters MC, Patience M, Leisenring W, et al; Multicenter Investigation of
Bone Marrow Transplantation for Sickle Cell Disease. Stable mixed hema­to­
poi­etic chi­me­rism after bone mar­row trans­plan­ta­tion for sickle cell ane­mia.
Biol Blood Marrow Transplant. 2001;7(12):665-673.
71. Abra­ham A, Hsieh M, Eapen M, et al; National Institutes of Health; Cen­
ter for International Blood and Marrow Transplant Research. Relationship
between mixed donor-recip­i­ent chi­me­rism and dis­ease recur­rence after
hema­to­poi­etic cell trans­plan­ta­tion for sickle cell dis­ease. Biol Blood Mar-
row Transplant. 2017;23(12):2178-2183.
72. Fitzhugh CD, Cordes S, Taylor T, et al. At least 20% donor mye­loid chi­me­
rism is nec­es­sary to reverse the sickle phe­no­type after allo­ge­neic HSCT.
Blood. 2017;130(17):1946-1948.
73. Wu CJ, Gladwin M, Tisdale J, et  al. Mixed haematopoietic chi­me­rism for
sickle cell dis­ease pre­vents intra­vas­cu­lar haemolysis. Br J Haematol.
2007;139(3):504-507.
74. Krishnamurti L, Kharbanda S, Biernacki MA, et  al. Stable long-term donor
engraft­ment fol­low­ing reduced-inten­sity hema­to­poi­etic cell trans­plan­ta­tion
for sickle cell dis­ease. Biol Blood Marrow Transplant. 2008;14(11):1270-1278.
75. Walters MC, Sullivan KM, Bernaudin F, et al. Neurologic com­pli­ca­tions after
allo­ge­neic mar­row trans­plan­ta­tion for sickle cell ane­mia. Blood. 1995;85(4):
879-884.
76. Gaziev J, Marziali S, Paciaroni K, et  al. Posterior revers­ible enceph­a­lop­a­
thy syn­drome after hema­to­poi­etic cell trans­plan­ta­tion in chil­dren with
hemo­glo­bin­op­a­thies. Biol Blood Marrow Transplant. 2017;23(9):1531-1540.
77. Solh Z, Taccone MS, Marin S, Athale U, Breakey VR. Neurological PRESen­
tations in sickle cell patients are not always stroke: a review of pos­te­rior
revers­ible enceph­a­lop­a­thy syn­drome in sickle cell dis­ease. Pediatr Blood
Cancer. 2016;63(6):983-989.
78. Pegelow CH, Colangelo L, Steinberg M, et al. Natural his­tory of blood pres­
sure in sickle cell dis­ease: risks for stroke and death asso­ci­ated with rel­a­
tive hyper­ten­sion in sickle cell ane­mia. Am J Med. 1997;102(2):171-177.
79. Kassim AA, Galadanci NA, Pruthi S, DeBaun MR. How I treat and man­age
strokes in sickle cell dis­ease. Blood. 2015;125(22):3401-3410.
80. Thompson CB, June CH, Sullivan KM, Thomas ED. Association between cyc­
losporin neu­ro­tox­ic­ity and hypomagnesaemia. Lancet. 1984;2(8412):1116-
1120.
81. June CH, Thompson CB, Kennedy MS, Loughran TP Jr, Deeg HJ. Correlation
of hypo­mag­ne­se­mia with the onset of cyclo­spor­ine-asso­ci­ated hyper­ten­
sion in mar­row trans­plant patients. Transplantation. 1986;41(1):47-51.
82. Shenoy S, Gaziev J, Angelucci E, et  al. Late effects screen­ing guide­lines
after hema­to­poi­etic cell trans­plan­ta­tion (HCT) for hemo­glo­bin­op­a­thy: con­
sen­sus state­ment from the Second Pediatric Blood and Marrow Transplant
Consortium inter­na­tional con­fer­ence on late effects after pedi­at­ric HCT.
Biol Blood Marrow Transplant. 2018;24(7):1313-1321.
83. Gladwin MT, Sachdev V, Jison ML, et  al. Pulmonary hyper­ ten­sion as a
risk fac­tor for death in patients with sickle cell dis­ease. N Engl J Med.
2004;350(9):886-895.
188  |  Hematology 2021  |  ASH Education Program
84. Gladwin MT, Barst RJ, Gibbs JS, et al; walk-PHaSST Investigators and
Patients. Risk fac­tors for death in 632 patients with sickle cell dis­ease in
the United States and United Kingdom. PLoS One. 2014;9(7):e99489.
85. Rogers ZR, Wang WC, Luo Z, et al; BABY HUG Investigators. Biomarkers of
splenic func­tion in infants with sickle cell ane­mia: base­line data from the
BABY HUG trial. Blood. 2011;117(9):2614-2617.
86. Brown AK, Sleeper LA, Miller ST, Pegelow CH, Gill FM, Waclawiw MA; for the
Cooperative Study of Sickle Cell Disease. Reference val­ues and hema­to­logic
changes from birth to 5 years in patients with sickle cell dis­ease. Arch
Pediatr Adolesc Med. 1994;148(8):796-804.
87. Gale HI, Bobbitt CA, Setty BN, et al. Expected sonographic appear­ance of
the spleen in chil­dren and young adults with sickle cell dis­ease: an update.
J Ultrasound Med. 2016;35(8):1735-1745.
88. Martin OO, Moquist KL, Hennessy JM, Nelson SC. Invasive pneu­mo­coc­
cal dis­ease in chil­dren with sickle cell dis­ease in the pneu­mo­coc­cal con
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/181/1851576/181krishnamurti.pdf by guest on 13 December 2021
­ju­gate vac­cine era. Pediatr Blood Cancer. 2018;65(1):e26713.
89. Ochocinski D, Dalal M, Black LV, et al. Life-threat­en­ing infec­tious com­pli­
ca­tions in sickle cell dis­ease: a con­cise nar­ra­tive review. Front Pediatr.
2020;8(20 Feb­ru­ary):38.
90. Khatib R, Rabah R, Sarnaik SA. The spleen in the sick­ling dis­or­ders: an
update. Pediatr Radiol. 2009;39(1):17-22.
91. Nickel RS, Seashore E, Lane PA, et al. Improved splenic func­tion after hema­
to­poi­etic stem cell trans­plant for sickle cell dis­ease. Pediatr Blood Cancer.
2016;63(5):908-913.
92. Tomblyn M, Chiller T, Einsele H, et al; Center for International Blood and
Marrow Research; National Marrow Donor pro­gram; Euro­pean Society for
Blood and Marrow Transplantion; Amer­i­can Society of Blood and Marrow
Transplantation; Cana­dian Blood and Marrow Transplant Group; Infectious
Diseases Society of America; Society for Healthcare Epidemiology of Amer­
ica; Association of Medical Microbiology and Infectious Disease Canada;
Centers for Disease Control and Prevention. Guidelines for pre­vent­ing infec­
tious com­pli­ca­tions among hema­to­poi­etic cell trans­plan­ta­tion recip­i­ents:
a global per­spec­tive. Biol Blood Marrow Transplant. 2009;15(10):1143-1238.
93. Ishak KG. Pathologic fea­ tures of chronic hep­ a­
ti­
tis. Am J Clin Pathol.
2000;113(1):40-55.
94. Angelucci E, Pilo F. Management of iron over­load before, dur­ing, and after
hema­to­poi­etic stem cell trans­plan­ta­tion for thal­as­se­mia major: trans­plan­
ta­tion and iron over­load. Ann NY Acad Sci. 2016;1368(1):115-121.
95. Angelucci E, Muretto P, Lucarelli G, et al; for the Ital­ian Cooperative
Group for Phlebotomy Treatment of Transplanted Thalassemia by Bone
Marrow Transplantation. Phlebotomy to reduce iron over­load in patients
cured of thal­as­se­mia by bone mar­row trans­plan­ta­tion. Blood. 1997;90(3):
994-998.
96. Giardini C, Galimberti M, Lucarelli G, et al. Desferrioxamine ther­apy accel­
er­ates clear­ance of iron depos­its after bone mar­row trans­plan­ta­tion for
thal­as­sae­mia. Br J Haematol. 1995;89(4):868-873.
97. Schettini F, De Mattia D, Sabato V, Del Vecchio GC. Compliance with iron-
che­la­tion treat­ment after bone mar­row trans­plan­ta­tion. Lancet. 1994;343
(8897):604.
98. Jodele S. Double trou­ble: com­ple­ment-medi­ated throm­botic microangi­
opathy in patients with hemo­glo­bin­op­a­thies after stem cell trans­plan­ta­
tion. Pediatr Blood Cancer. 2017;64(9).
99. Abusin GA, Abu-Arja R, Bajwa RPS, Horwitz EM, Auletta JJ, Rangarajan HG.
Severe trans­ plant-asso­ci­
ated throm­ botic microangiopathy in patients
with hemo­glo­bin­op­a­thies. Pediatr Blood Cancer. 2017;64(9):e26503.
100. Bayer G, von Tokarski F, Thoreau B, et al. Etiology and out­comes of throm­
botic microangiopathies. Clin J Am Soc Nephrol. 2019;14(4):557-566.
101.  Bhunia N, Abu-Arja R, Bajwa RPS, Auletta JJ, Rangarajan HG. Successful
treat­ment with eculizumab for pos­te­rior revers­ible enceph­a­lop­a­thy syn­
drome due to under­ly­ing trans­plant-asso­ci­ated throm­botic microangiop­
athy in patients transplanted for sickle cell dis­ease. Pediatr Blood Cancer.
2019;66(10):e27912.
102. Davies JOJ, Hart AC, De La Fuente J, Bain BJ. Macrophage acti­va­tion syn­
drome and post-trans­plant microangiopathy fol­low­ing haploidentical
bone mar­row trans­plan­ta­tion for sickle cell ane­mia. Am J Hematol. 2018;​
93(4):588-589.
103. Maigne G, Ferlicot S, Galacteros F, et  al. Glomerular lesions in patients
with sickle cell dis­ease. Medicine (Baltimore). 2010;89(1):18-27.
104. Walters MC, Hardy K, Edwards S, et al; Multicenter Study of Bone Marrow
Transplantation for Sickle Cell Disease. Pulmonary, gonadal, and cen­tral
ner­vous sys­tem sta­tus after bone mar­row trans­plan­ta­tion for sickle cell
dis­ease. Biol Blood Marrow Transplant. 2010;16(2):263-272.
 105. 
Mishkin AD, Mapara MY, Barhaghi M, Reshef R. Fertility con­ cerns and
access to care for stem cell trans­plan­ta­tion can­di­dates with sickle cell
dis­ease. Biol Blood Marrow Transplant. 2020;26(8):e192-e197.
106. Practice Committee of the Amer­i­can Society for Reproductive Medicine.
Fertility pres­ er­
va­tion in patients under­ go­ing gonadotoxic ther­ apy or
gonad­ec­tomy: a com­mit­tee opin­ion. Fertil Steril. 2019;112(6):1022-1033.
107.  Brodigan K, Kapadia M, Frazier AL, et al. Safety of sur­gi­cal fer­til­ity pres­er­
va­tion pro­ce­dures in chil­dren prior to hema­to­poi­etic stem cell trans­plant.
Transplant Cell Ther. 2021;27(8):696.e1-696696.e4.
108. Lautz TB, Harris CJ, Laronda MM, Erickson LL, Rowell EE. A fer­til­ity pres­
er­va­tion toolkit for pedi­at­ric sur­geons car­ing for chil­dren with can­cer.
Semin Pediatr Surg. 2019;28(6):150861.
109. Joseph L, Jean C, Manceau S, et al. Effect of hydroxy­urea expo­sure before
puberty on sperm param­e­ters in males with sickle cell dis­ease. Blood.
2021;137(6):826-829.
110.  Pecker LH, Hussain S, Christianson MS, Lanzkron S. Hydroxycarbamide
expo­sure and ovar­ian reserve in women with sickle cell dis­ease in the
Multicenter Study of Hydroxycarbamide. Br J Haematol. 2020;191(5):880-
887.
111.  Benoit M, Chiles K, Hsieh M. The land­scape of cov­er­age for fer­til­ity pres­
er­va­tion in male pedi­at­ric patients. Urol Pract. 2018;5(3):198-204.
112. Shenoy S, Angelucci E, Arnold SD, et al. Current results and future research
pri­or­i­ties in late effects after hema­to­poi­etic stem cell trans­plan­ta­tion for
chil­dren with sickle cell dis­ease and thal­as­se­mia: a con­sen­sus state­ment
from the Second Pediatric Blood and Marrow Transplant Consortium
inter­na­tional con­fer­ence on late effects after pedi­at­ric hema­to­poi­etic
stem cell trans­plan­ta­tion. Biol Blood Marrow Transplant. 2017;23(4):552-
561.
113. Khemani K, Ross D, Sinha C, Haight A, Bakshi N, Krishnamurti L. Experi­
ences and deci­sion mak­ing in hema­to­poi­etic stem cell trans­plant in sickle
cell dis­ease: patients’ and care­giv­ers’ per­spec­tives. Biol Blood Marrow
Transplant. 2018;24(5):1041-1048.
114. Schulz GL, Kelly KP, Holtmann M, Armer JM. Navigating deci­sional con­flict
as a fam­ily when fac­ing the deci­sion of stem cell trans­plant for a child or
ado­les­cent with sickle cell dis­ease. Patient Educ Couns. 2021;104(5):1086-
1093.
115. van Besien K, Koshy M, Anderson-Shaw L, et al. Allogeneic stem cell trans­
plan­ta­tion for sickle cell dis­ease. A study of patients’ deci­sions. Bone Mar-
row Transplant. 2001;28(6):545-549.
116. Kodish E, Lantos J, Stocking C, Singer PA, Siegler M, Johnson FL. Bone mar­
row trans­plan­ta­tion for sickle cell dis­ease: a study of par­ents’ deci­sions.
N Engl J Med. 1991;325(19):1349-1353.
117. Majhail NS, Nayyar S, Santibañez ME, Mur­phy EA, Denzen EM. Racial dis­
parities in hema­to­poi­etic cell trans­plan­ta­tion in the United States. Bone
Marrow Transplant. 2012;47(11):1385-1390.
118. Majhail NS, Omondi NA, Denzen E, Mur­phy EA, Rizzo JD. Access to hema­
to­poi­etic cell trans­plan­ta­tion in the United States. Biol Blood Marrow
Transplant. 2010;16(8):1070-1075.
119. Preussler JM, Farnia SH, Denzen EM, Majhail NS. Variation in Med­ ic­
aid cov­er­age for hema­to­poi­etic cell trans­plan­ta­tion. J Oncol Pract.
2014;10(4):e196-e200.
120.  Mupfudze TG, Preussler JM, Sees JA, SanCartier M, Arnold SD, Devine S.
A qual­i­ta­tive anal­y­sis of state Med­ic­aid cov­er­age ben­e­fits for allo­ge­neic
hema­to­poi­etic cell trans­plan­ta­tion (alloHCT) for patients with sickle cell
dis­ease (SCD). Transplant Cell Ther. 2021;27(4):345-351.
Prakash Singh Shekhawat
121. Abu Al Hamayel N, Waldfogel JM, Hannum SM, et al. Pain expe­ri­ences of
adults with sickle cell dis­ease and hema­to­poi­etic stem cell trans­plan­ta­
tion: a qual­i­ta­tive study. Pain Med. 2021;22(8):1753-1759.
122. Isaac EI, Meisman AR, Drucker K, et al. The rela­tion­ship between health
disparities, psy­cho­so­cial func­tion­ing and health out­comes in pedi­at­ric
hema­tol­ogy-oncol­ogy and stem cell trans­plant pop­u­la­tions: rec­om­men­
da­tions for clin­i­cal care. Int J Environ Res Public Health. 2020;17(7):2218.
123. Patel MI, Ma Y, Mitchell B, Rhoads KF. How do dif­fer­ences in treat­ment
impact racial and eth­nic disparities in acute mye­loid leu­ke­mia? Cancer
Epidemiol Biomarkers Prev. 2015;24(2):344-349.
124. Kirtane K, Lee SJ. Racial and eth­nic disparities in hema­to­logic malig­nan­
cies. Blood. 2017;130(15):1699-1705.
125. Li Z, Czechowicz A, Scheck A, Rossi DJ, Mur­phy PM. Hematopoietic chi­me­
rism and donor-spe­cific skin allo­graft tol­er­ance after non-genotoxic CD117
anti­body-drug-con­ju­gate con­di­tion­ing in MHC-mismatched allotrans­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/181/1851576/181krishnamurti.pdf by guest on 13 December 2021
plantation. Nat Commun. 2019;10(1):616.
126. Czechowicz A, Palchaudhuri R, Scheck A, et al. Selective hema­to­poi­etic
stem cell abla­ tion using CD117-anti­ body-drug-con­ ju­
gates enables safe
and effec­tive trans­plan­ta­tion with immu­nity pres­er­va­tion. Nat Commun.
2019;10(1):617.
127.   Witt RG, Wang B, Nguyen QH, Eikani C, Mattis AN, MacKenzie TC. Deple­
tion of murine fetal hema­to­poi­etic stem cells with c-kit recep­tor and CD47
block­ade improves neo­na­tal engraft­ment. Blood Adv. 2018;2(24):3602-
3607.
128. Arai Y, Choi U, Corsino CI, et al. Myeloid con­di­tion­ing with c-kit-targeted
CAR-T cells enables donor stem cell engraft­ment. Mol Ther. 2018;26(5):
1181-1197.
129.  Yokoi T, Yokoi K, Akiyama K, et  al. Non-myeloablative preconditioning
with ACK2 (anti-c-kit anti­body) is effi­cient in bone mar­row trans­plan­ta­tion
for murine mod­els of mucopolysaccharidosis type II. Mol Genet Metab.
2016;​119(3):232-238.
130. Xue X, Pech NK, Shelley WC, Srour EF, Yoder MC, Dinauer MC. Antibody
targeting KIT as pretransplantation con­di­tion­ing in immu­no­com­pe­tent
mice. Blood. 2010;116(24):5419-5422.
131. Czechowicz A, Kraft D, Weissman IL, Bhattacharya D. Efficient trans­plan­
ta­tion via anti­body-based clear­ance of hema­to­poi­etic stem cell niches.
Science. 2007;318(5854):1296-1299.
132. Parikh S, Brochstein JA, Galamidi E, Schwarzbach A, Kurtzberg J. ­Allogeneic
stem cell trans­plan­ta­tion with omidubicel in sickle cell dis­ease. Blood
Adv. 2021;5(3):843-852.
133. Zimran E, Papa L, Hoffman R. Ex vivo expan­sion of hema­to­poi­etic stem
cells: finally transitioning from the lab to the clinic. Blood Rev. 2021;​50(2):​
100853.
134. Chandra S, Mizuno K, Zhao J, et  al. Test-dose phar­ma­co­ki­net­ics guided
mel­pha­lan dose adjust­ment in reduced inten­sity con­di­tion­ing allo­ge­neic
trans­plant for non-malig­nant dis­or­ders. Published online 1 June 2021. Br J
Clin Pharmacol.
135. Dong M, Emoto C, Fukuda T, et al. Model-informed pre­ci­sion dos­ing for
alemtuzumab in pae­di­at­ric and young adult patients under­go­ing allo­ge­
neic haematopoietic cell trans­plan­ta­tion. Published online 28 June 2021.
Br J Clin Pharmacol. 2021.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000251
HCT for SCD  |  189
 EMERGING THERAPIES AND CONSIDERATIONS FOR SICKLE CELL DISEASE

EVIDENCE-BASED MINIREVIEW

In young children with severe sickle cell disease,

do the benefits of HLA-identical sibling donor HCT

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/190/1851641/190shah.pdf by guest on 13 December 2021

outweigh the risks?
Niketa Shah1 and Lakshmanan Krishnamurti2
Section of Pediatric Hematology/Oncology/BMT, Yale School of Medicine, New Haven, CT; and 2Aflac Cancer and Blood Disorders Center,
1

Children’s Healthcare of Atlanta, Emory University, Atlanta, GA

In case 1, a 14-month-old male child with sickle cell disease (SCD) was referred for evaluation for an allogeneic hematopoi-
etic stem cell transplant (HCT). The patient had a history of dactylitis 3 times in his first year of life and febrile episodes twice
at the consult. His 4-year-old sister was found to be human leukocyte antigen (HLA) identical. The patient was started on
hydroxyurea (HU) at 2.5 years of age. His parents again sought consultation when he was 5 years old because of concerns
about his medical condition. At the time, the patient had experienced 2 vaso-occlusive pain episodes (VOEs) requiring
hospitalization during the previous 2 years. He had also experienced intermittent pain crises requiring rest at home for
2 to 3 days. The child has not attended school in person due to the COVID-19 pandemic. The family is considering HCT
but is ambivalent about it because of potential toxicity. In case 2, an 8-year-old female child is 3 years out from HCT for
SCD from her HLA-identical sibling. Before HCT, despite receiving HU, she had experienced >5 VOEs requiring hospital-
ization and 2 episodes of acute chest syndromes in the previous 3 years. She had also been missing almost 50 days of
school days each year. After HCT, she is now attending school regularly and participating in all normal age-appropriate
activities. The parents believe that HCT has been transformative in their child’s life.

LEARNING OBJECTIVES
• Understand therapeutic options for a young child with SCD
• Describe the current outcomes of MSD HCT
• Critically appraise the role of MSD HCT for SCD in young children

Introduction Outcomes with comprehensive care and

Sickle cell disease (SCD) is an autosomal recessive inher-
ited hemoglobinopathy. Over 300 000 affected babies are
born each year with SCD, which is expected to increase to
over 400 000 a year by 2050.1,2 Although the overwhelm-
ing majority of patients live in Africa, India, or the Middle
East, growing numbers of individuals with SCD are encoun-
tered in areas that were historically not endemic for malaria
because of population migrations. In the United States, it
is estimated that 100 000 individuals are affected with
SCD.2-4 SCD is associated with a significant risk of morbidity
and premature mortality.5 Thus, SCD is a significant public
health problem worldwide that disproportionately affects
populations residing in resource-poor settings. Hemato-
poietic cell transplantation (HCT) can ameliorate long-term
SCD-related morbidity and its consequences.
disease-modifying therapy for SCD
Advances such as newborn screening, penicillin prophy-
laxis, pneumococcal immunization, and comprehensive care
have improved survival in children with SCD. Hydroxyurea
(HU) has been proven to effectively decrease the frequency
of vaso-occlusive pain episodes (VOEs), acute chest syn-
drome, frequency of transfusions, and death in multiple
randomized controlled trials.6 In addition, HU is an effective
substitute for chronic transfusions to prevent primary stroke
among high-risk pediatric patients with SCD who have
abnormal transcranial Doppler (TCD) flow velocity.5 Obser-
vational studies have demonstrated a relationship between
HU use and decreased rates of hospitalization and blood
transfusions.6 Thus, this relatively inexpensive orally admin-
istered medication can be used globally to ameliorate SCD

190 | Hematology 2021 | ASH Education Program

com­pli­ca­tions and improve sur­vival. Nouraie et al7 reported the
results of the pulmonary hypertension and the hypoxic response
in sickle cell disease study that esti­mated that the sur­vival of a
cohort of chil­dren, ado­les­cents, and young adults with SCD was
99% to 18 years, and this declined to 94% at 25 years. Stroke was the
most com­mon cause of death. Elevated sys­tolic pul­mo­nary artery
pres­sure as reflected in triscupid regurgitant jet velocity (TRV)
≥2.7 m/s, high serum fer­ri­tin con­cen­tra­tion, low forced expiratory
velocity (FEV1)/forced vital capacity, and high neu­tro­phil count
were bio­mark­ers of increased risk of death, pre­dom­i­nantly after
the tran­si­tion to adult­hood. Recently, newer dis­ease-mod­i­fy­ing
agents approved by the US Food and Drug Administration (FDA)
offer added poten­tial to improve out­comes for patients with SCD.
The new dis­ease-mod­i­fy­ing drugs include ­voxelotor, which binds
to the N-ter­mi­nus of sickle hemoglobin to sta­bi­lize oxy­gen­ated
sick­ling and pre­vent sick­ling8; L-glu­ta­mine, which reduces oxi­da­
tive stress and could result in fewer epi­sodes of VOE9; and crizan-
lizumab, which acts on cytoadherence of sickle red blood cells,
leu­ko­cytes, and plate­lets to the endo­the­lium.10 These drugs offer
the poten­tial for fur­ther ame­lio­rat­ing com­pli­ca­tions of SCD and
pro­vide an alter­na­tive for HCT for dis­ease mod­i­fi­ca­tion. However,
in the absence of long-term data, we do not know how the addi­
tional FDA-approved SCD ther­a­pies will mod­ify the clin­i­cal his­tory
of patients with SCD.
The improve­ment in sur­vival of chil­dren with SCD between
1979 and 2005 seemed to mir­ror a sim­i­lar decrease in adult sur­
vival dur­ing the same period.11 Although indi­vid­ual cen­ters in
high-resource set­tings report a median sur­vival of 58 to 67 years
over­all, there has been lit­tle improve­ment in sur­vival for adults
with SCD.12 In adults, there is a pro­gres­sive loss of pul­mo­nary
FEV1 at 49 cc/y.13 More than 61% of patients with SCD aged
18 to 30 years develop either micro- or macroalbuminuria.14 There
is a wors­en­ing change in tri­cus­pid regurgitant jet veloc­ity over
time.15 There is neurocognitive decline and worse per­for­mance
IQ , processing speed, cog­ni­tive func­tion, and exec­u­tive func­
tion with age.16 Acute-on-chronic mor­bid­ity leads to reduced
health-related qual­ity of life (HRQOL), depres­sion and anx­i­ety,
dimin­ished edu­ca­tional and occu­pa­tional ful­fill­ment, and pre­
ma­ture mor­tal­ity, con­trib­ut­ing to health care costs..2,17,18 Disease-
mod­i­fy­ing ther­a­pies and sup­port­ive care can improve out­comes
but place a sub­stan­tial bur­den of care, require­ment of med­i­ca­
tion adher­ence, and inter­rup­tion of daily life.17
Thus, a child receiv­ing com­pre­hen­sive care in high-resource
set­tings has an esti­mated 99% sur­vival into adult­hood. However,
there may be pro­gres­sive organ dam­age, impaired qual­ity of life,
con­sid­er­able mor­bid­ity in child­hood, and risk of pre­ma­ture mor­
tal­ity in adult­hood. However, it is cru­cial to rec­og­nize that it is
not yet known if the addi­tional FDA-approved SCD ther­a­pies will
mod­ify the clin­i­cal his­tory of patients with SCD in adult­hood.
Outcomes after matched sib­ling donor bone mar­row
trans­plant
HCT for SCD was first performed in 1984 from a human leu­ko­
cyte anti­gen (HLA) matching sib­ling donor (MSD) to treat acute
mye­loid leu­ke­mia in a patient who also had SCD and paved the
way for a cura­tive option for SCD. The early trans­plant series
using myeloablative con­di­tion­ing, such as a trans­plant for malig­
nant dis­eases, showed over­all sur­vival (OS) and event-free sur­
vival (EFS) in the range of 77% to 95% and 93% to 95% fol­low­ing
HCT from an HLA MSD, respec­tively.18,19 In the past decade, there
Prakash Singh Shekhawat
has been fur­ther improve­ment in the out­comes of HCT for SCD,
with OS and EFS in the range of 94% to 100% and 91% to 100%,
respec­ tively, with a min­ i­
mum risk of mor­ tal­
ity with reduced-
inten­sity/tox­ic­ity reg­i­mens.20-24 An inter­na­tional, ret­ro­spec­tive,
reg­is­try-based anal­y­sis reported an OS of 92.9% and EFS of 91.4%
in 1000 chil­dren who under­went HCT between 1986 and 2013.21
On mul­ti­var­i­ate anal­y­sis, sur­vival was bet­ter with bone mar­row
than periph­eral blood stem cells graft and those who under­went
trans­plan­ta­tion after 2006.21 Cappelli et al,22 in a study of recip­
i­ent age as a pre­dic­tor of the out­come of MSD HCT, reported
an OS and EFS of 100% and 93%, respec­tively, in chil­dren 0 to
5 years of age. Bernaudin et al25 reported out­comes of MSD HCT
for SCD from 1988 to 2012 and described the improve­ment in
out­come since 2005. They noted that for MSD since 2005, OS is
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/190/1851641/190shah.pdf by guest on 13 December 2021
98.6% and EFS is 97.4%. The use of reduced-inten­sity or nonmy-
eloablative con­di­tion­ing strat­e­gies such as com­bin­ing alemtu-
zumab with fludarabine and mel­pha­lan,26 as well as low-dose
total body irra­di­a­tion (300 cGy),20,27 has been asso­ci­ated with
high OS and EFS in both adults and chil­dren.
The risk-ben­e­fit trade-off in HCT for SCD
The deci­sion to pro­ceed with cura­tive options for SCD rep­re­sents
a com­plex risk-ben­e­fit trade-off. On one hand, sur­vival to adult­
hood is almost uni­ver­sal in chil­dren receiv­ing the best com­pre­
hen­sive care. The dis­ease, how­ever, progresses in the major­ity to
cumu­la­tive organ dam­age, mor­bid­ity, and mor­tal­ity in adult­hood.
On the other hand, fol­low­ing HLA MSD HCT in young chil­dren,
there is a high prob­a­bil­ity of sur­vival, with a poten­tial for alle­vi­at­
ing symp­toms related to SCD, sta­bi­liz­ing organ func­tion, and the
pos­si­bil­ity of the patient being ­able to live rel­a­tively every­day life.
But HCT itself is asso­ci­ated with sub­stan­tial com­pli­ca­tions in the
short term, a sig­nif­i­cant bur­den of care that can be dis­rup­tive to
the patient and care­giv­ers, and a small risk of mor­tal­ity.
Furthermore, the long-term sequelae of HCT, such as graft-
ver­sus-host dis­ease, infer­til­ity, and sub­se­quent malig­nancy, could
be unin­tended con­se­quences of this ther­apy (Figure 1). This deci­
sional dilemma is also fur­ther compounded by the patient’s age,
the sever­ity of the cur­rent clin­i­cal course, the type of donor
avail­­able, and the avail­abil­ity of other treat­ment options. Thus,
par­ents con­sid­er­ing HCT for their young child face mul­ti­ple com­
plex trade-offs: the pri­mary trade-off for a fam­ily of a 5-year-old
is sur­vival of the child in the decade when they age from 5 to 15
years. Although no study has com­pared HCT with stan­dard of
care in chil­dren with SCD, published stud­ies sug­gest that sur­vival
Figure 1. Risk-benefit paradigm for curative therapy for SCD.
HLA iden­ti­cal donor HCT in young chil­dren  |  191
to the next decade for a 5-year-old child with SCD fol­low­ing
stan­dard of care or MSD HCT may be com­pa­ra­ble. However, a
par­ent may also con­sider other trade-offs such as the out­comes
of HCT in child­hood vs the poten­tial for mor­bid­ity and mor­tal­
ity in adult­hood, HRQOL with or with­out HCT, and cura­tive HCT
vs chronic GVHD, infer­til­ity, car­dio­vas­cu­lar risk, and sub­se­quent
malig­nancy.28 How fam­i­lies resolve these deci­sional trade-offs is
likely to be highly indi­vid­u­al­ized and depen­dent on sev­eral fac­
tors unique to the cir­cum­stances of the patient and fam­ily. To
assist the hema­tol­o­gist in the con­ver­sa­tion with the fam­ily of a
young child with SCD who has an HLA-iden­ti­cal sib­ling, we sum­
ma­rize the avail­­able evi­dence as answers to crit­i­cal ques­tions
that may arise in the minds of patients and their care­giv­ers as
they con­tem­plate HCT.
Who has been offered HCT for SCD?
HCT for SCD has been offered to patients with severe SCD-
related com­pli­ca­tions. Indications for HCT include stroke, the
need for chronic blood trans­ fu­sions to pre­
vent SCD-related
com­pli­ca­tions, and severe dis­ease com­pli­ca­tions such as recur­
rent VOE, recur­rent acute chest syn­drome, sickle nephrop­a­thy,
or ret­i­nop­a­thy that at least affects vision in 1 eye or osteonecro-
sis.18,29 In early clin­i­cal tri­als of HCT for SCD, the most com­mon
indi­ca­tions were stroke in 57% of patients and fre­quent VOEs in
23% of patients. The most com­mon indi­ca­tion for HCT recently
has been recur­rent VOE in more than 70% of the cases.21 The
change in the most fre­quent indi­ca­tion for HCT may reflect the
decrease in the inci­dence of stroke because of bet­ter com­pre­
hen­sive care, screen­ing for stroke risk, and pri­mary pre­ven­tion
of stroke by the insti­tu­tion of chronic blood trans­fu­sion. It is
also pos­si­ble that phy­si­cians and patients are more accepting
of HCT for indi­vid­ua ­ ls with recur­rent VOE. Utilization of health
care for recur­rent VOE is being used as a marker of dis­ease
sever­ity and a pre­dic­tor of long-term sur­vival in adults. Yet,
health care uti­li­za­tion for VOE is not an accu­rate mea­sure of
pain bur­ den because many VOE epi­ sodes are man­ aged at
home. Thus, some patients may have chronic dis­abling pain
but may not access health care for var­i­ous rea­sons.30 There is,
there­fore, a need to bet­ter esti­mate pain bur­den to inform risk-
ben­e­fit strat­i­fi­ca­tion in con­sid­er­ation of HCT for SCD. An ongo­
ing clin­i­cal trial is study­ing HCT from HLA matched sib­ling HCT
for chil­dren with the less severe clin­i­cal phe­no­type (ClinicalTri
als​­.gov NCT04018937).
Is it pos­si­ble to offer MSD HCT to every child with SCD?
The appli­ca­bil­ity of HCT is lim­ited because only 14% to 18% of
patients with SCD have an avail­­able HLA-iden­ti­cal fam­ily donor.31
Most sib­ling donors can have sickle cell trait. However, HCT from
a donor with sickle cell trait is safe and effec­tive.31,32 Clinical tri­
als are eval­u­at­ing alter­nate donor HCT. An esti­mated 19% of the
patients may find a suit­able HLA matched unre­lated donor,32 and
most are likely to find a haploidentical fam­ily donor. If the safety
and effi­cacy of alter­nate donor HCT are proven, a more sig­nif­
i­cant pro­por­tion of patients may be pro­vided HCT. Thus, only
a small minor­ity of patients can be offered HCT from an HLA-
iden­ti­cal fam­ily donor.
What is the best time to per­form HCT?
Age at HCT is the most crit­ ic ­al deter­
mi­
nant of out­ come in
patients with SCD with an avail­­able HLA MSD. Patients receiv­ing
192  |  Hematology 2021  |  ASH Education Program
HCT who are youn­ger than 5 years have an EFS of 93% and an
OS of 100% com­pared with 89% and 95% for those aged 5 to 10
years and 81% and 88% for those older than 15 years, respec­
tively.22 Gluckman et al21 report that for every 1-year incre­ment
in age, there was a 9% increase in the haz­ard ratio for treat­ment
fail­ure—that is, graft fail­ure or death. Similarly, for every 1-year
incre­ment in age, there was a 10% increase in the haz­ard ratio
for death.22 Risk of graft fail­ure, grade II to IV acute graft-ver­sus-
host dis­ease (GVHD), and chronic GVHD are low­est in chil­dren
youn­ger than 5 years at HCT. Gluckman et al21 reported from the
Euro­pean Society for Blood and Marrow Transplantation reg­is­try
that many cen­ters have diverse expe­ri­ence in performing HCT
for SCD. This is mir­rored by the report by Bernaudin et al23 in a
sin­gle-cen­ter study with a 5-year EFS of 97.9% (95% con­fi­dence
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/190/1851641/190shah.pdf by guest on 13 December 2021
inter­val, 95.5%-100%). Brazauskas et al24 devel­oped a risk scor­ing
sys­tem that included age at HCT and donor type to pre­dict EFS
after HCT for SCD. They con­cluded that patients 12 years and
youn­ger with an HLA-iden­ti­cal fam­ily donor have the low­est risk.
Another pri­mary con­sid­er­ation in the tim­ing of HCT for SCD
is the pres­ence or immi­nence of organ dam­age. Therefore, HCT
should pref­er­a­bly be offered to patients who have been pre-
dicted to have a poor out­come but have not expe­ri­enced severe
organ dam­age to avoid seri­ous side effects asso­ci­ated with HCT
pro­ce­dures. However, SCD organ dam­age is insid­i­ous, with pro­
gres­sive loss of FEV1 at 49 cc/y,13 61% of patients with SCD aged
18 to 30 years devel­op­ing either micro- or macroalbuminuria,14
wors­ en­
ing changes in tri­ cus­
pid regurgitant jet veloc­ ity over
time,15 and neurocognitive decline with poor per­for­mance IQ ,
processing speed, cog­ni­tive func­tion, and exec­u­tive func­tion.16
Because over­all out­comes of MSD HCT for SCD are best when
HCT is performed at age 5 years or youn­ger, the con­sen­sus opin­
ion of an Euro­pean Society for Blood and Marrow Transplantation
expert panel is to con­sider MSD at a young age.33
What about long-term side effects and qual­ity of life after
HCT com­ pared with patients with SCD who have not
received cura­tive ther­a­pies?
Improvement in HRQOL out­comes in patients with SCD fol­low­ing
HCT has been reported in mul­ti­ple stud­ies.34-39 This includes sig­
nif­i­cant improve­ment in phys­i­cal, emo­tional, psy­cho­so­cial, gen­
eral health, bodily pain, pain inter­fer­ence, and vital­ity domains
of HRQOL. It has also been reported that car­diac, neu­ro­logic,
and neuropsychologic func­tions sta­bi­lize after HCT.19 In the most
recent report, for chil­dren with SCD requir­ing chronic trans­fu­sion
because of per­sis­tently ele­vated TCD veloc­i­ties, MSD HSCT was
sig­nif­i­cantly asso­ci­ated with lower TCD veloc­i­ties at 1 year com­
pared with stan­dard care.25 One of the sig­nif­i­cant risks fol­low­ing
HCT is gonadal dam­age and infer­til­ity.19 Fertility pres­er­va­tion by
sperm bank­ing or cryo­pres­er­va­tion of oocytes can poten­tially
mit­i­gate the risk of infer­til­ity. Ovarian tis­sue cryo­pres­er­va­tion in
pre­pu­ber­tal females is now con­sid­ered the stan­dard of clin­i­cal
care. Experimental pro­ ce­
dures such as tes­ tic­
u­lar tis­
sue cryo­
pres­er­va­tion are still under inves­ti­ga­tion.40,41 On the other hand,
HU also adversely affects male and female fer­til­ity in patients
with SCD.42,43 The risk of sub­se­quent malig­nancy fol­low­ing HCT
remains small.41,44,45 Late effects of HCT include car­dio­vas­cu­lar,
pul­mo­nary, and endo­crine com­pli­ca­tions; dys­func­tion of the thy­
roid gland, gonads, liver, and kid­neys; infer­til­ity; iron over­load;
bone dis­eases; infec­tion; malig­nancy; and neuropsychological
effects.46 Most of these late effects have been stud­ied in patients
under­go­ing HCT for can­cers or bone mar­row fail­ure syn­dromes.
There is a need to gen­er­ate data on the long-term out­come of
HCT for SCD to address this ele­ment of uncer­tainty in the deci­
sion-mak­ing regard­ing HCT for SCD.47,48
Is HCT cost-effec­tive?
Although the cost-effec­tive­ness of HCT for SCD is still being eval­u­
ated, early data sug­gest HCT is asso­ci­ated with decreased hos­pi­
tal­i­za­tion, health care uti­li­za­tion, and opi­oid uti­li­za­tion.49 In adults
with SCD, Saraf et al50 reported lower health care costs by the sec­
ond year post-HCT (median of $16 281 vs $64 634 pre-HSCT [P = .01]
vs $54 082 in the stan­dard-of-care group [P = .05]). They reported
a median reduc­tion of –$20 833/patient/y (interquartile range,
–$67 078 to +$4442/patient/y) in health care costs com­pared
to pre-HCT in the sec­ond year post-HSCT (P = .05). In chil­dren,
Arnold et al,51 while reviewing 2 years pre- and post-HCT data,
reported sig­nif­i­cant reduc­tions in admis­sions (P < .001), length of
stay (P < .001), and cost (P = .008) for HCT patients who received
their HCT before the age of 10 years.
Should we wait for gene ther­apy clin­i­cal tri­als or for gene
ther­apy to be approved instead of con­sid­er­ing allo­ge­neic
HCT?
Early phase clin­i­cal tri­als of autol­o­gous trans­plan­ta­tion of hema­
to­poi­etic pro­gen­i­tors mod­i­fied by gene addi­tion or gene edit-
ing hold the prom­ise of long-term dis­ease ame­lio­ra­tion with­out
the need for an allo­ge­neic donor and asso­ci­ated risk of GVHD.
Gene ther­apy stud­ies have so far enrolled patients 12 years or
older. Future stud­ies may offer enroll­ment to youn­ger chil­dren.
The stud­ies typ­i­cally exclude chil­dren who have an avail­­able HLA
MSD. The cur­rent clin­i­cal hold of the Hgb-206 gene ther­apy trial
(ClinicalTrials.gov: NCT02140554) after 2 par­tic­i­pants devel­oped
acute mye­loid leu­ke­mia under­scores the fact that there remain
many unknowns in this exper­i­men­tal ther­apy.52-58 Thus, until a via­
ble, safe strat­egy has been implemented for the gene ther­apy
approach, HCT is still con­sid­ered the best cura­tive option for
young chil­dren with SCD if they have an avail­­able HLA MSD, par­
tic­u­larly with reduced-inten­sity/tox­ic­ity con­di­tion­ing reg­i­mens.
There have been no com­par­a­tive tri­als of allo­ge­neic HCT and
autol­o­gous gene ther­apy for any other dis­eases. Thus, fur­ther
stud­ies are required to deter­mine the rel­a­tive safety and effi­cacy
of the 2 treat­ment modal­i­ties.
Are there indi­vid­ual dif­fer­ences in how patients and fam­i­lies
con­sider the infor­ma­tion and make deci­sions about HCT?
The deci­ sion for HCT for SCD involves com­ plex deci­ sional
trade-offs. Parents decid­ing to pro­ceed with HCT face many
dilem­mas such as the unpre­dict­able onset and pro­gres­sion of
SCD com­pli­ca­tions, repeated need to seek health care with
SCD, pos­si­ble poor qual­ity of life with SCD, the immi­nence of
mak­ing a major ther­a­peu­tic deci­sion, and con­cerns about the
long-term con­se­quences of the dis­ease. The feel­ing that the
dis­ease bur­den has become unac­cept­able is fre­quently cited
as a trig­ger for the paren­tal deci­sion.28,59,60 The avail­abil­ity of an
HLA MSD may be viewed as prov­i­den­tial by some par­ents. Par-
ents vary greatly in their will­ing­ness to accept risks asso­ci­ated
with HCT to achieve poten­tial ben­e­fits. HCT-asso­ci­ated mor­
bid­ity, includ­ing mucositis, GVHD, risk of death, and infer­til­ity, is
a sig­nif­i­cant rea­son that gives fam­i­lies pause in accepting HCT
options. Overall, fam­i­lies face sub­stan­tial deci­sional con­flict.60
Prakash Singh Shekhawat
Some patients or care­giv­ers would not con­sider HCT at any
level of risk.61,62 Almost three-fourths of patients report being
will­ing to take a mod­est risk of ≥5% mor­tal­ity risk, whereas 57%
are ready to take ≥10% risk of GVHD. Some par­ents and care­
giv­ers may also have dif­fi­culty mak­ing deci­sions about HCT for
and on behalf of a child who is cur­rently doing well and may
have to deal with long-term sequelae of HCT.28 However, fol­
low­ing suc­cess­ful HCT, par­ents do not report deci­sional regret
even if their child has had severe HCT-related com­pli­ca­tions.59
Thus, the deci­ sion to undergo HCT is highly indi­ vid­
u­al­
ized
and is driven by the per­cep­tion of dis­ease sever­ity; the val­ues
and pref­er­ences of patients, care­giv­ers, and their phy­si­cians
regard­ing dis­ease-mod­i­fy­ing options; and acute and long-term
sequelae of HCT.28,63
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/190/1851641/190shah.pdf by guest on 13 December 2021
Summary and rec­om­men­da­tions
Young chil­dren with SCD have an excel­lent out­look for sur­vival to
adult­hood, and new drugs may fur­ther improve out­comes. How-
ever, many chil­ dren may have severe dis­ ease man­ i­
fes­
ta­
tions,
organ dam­age, and impaired qual­ity of life. Disease pro­gres­sion
may be unpre­dict­able in adult­hood, with severe con­se­quences
for the patient. HCT from HLA MSDs in young chil­dren is asso­ci­
ated with excel­lent OS and EFS, sta­bi­li­za­tion of organ func­tion,
decreased health care uti­li­za­tion, and improved qual­ity of life.
These ben­e­fits come at the cost of acute mor­bid­ity, the sub­stan­
tial short-term bur­den of care, a small risk of HCT-related mor­tal­
ity, late sequelae such as infer­til­ity, and a slight increase in the
risk of sub­se­quent malig­nancy.
Furthermore, with increas­ing age, there is a decline in the
out­comes of HCT. Parents and care­giv­ers approach these com­
plex trade-offs in ways unique to their life sit­u­a­tions and expe­
ri­ences and their per­cep­tion of dis­ease bur­den. We, there­fore,
rec­om­mend rou­tine HLA typ­ing of full sib­lings and con­sul­ta­tion
with a phy­si­cian knowl­edge­able about HCT as 1 com­po­nent of
com­pre­hen­sive and sup­port­ive care, which includes the use of
cur­rent dis­ease-mod­i­fy­ing ther­a­pies. The com­plex­ity of deci­
sion-mak­ing regard­ing HCT neces­si­tates a bal­anced dis­cus­sion
of the deci­sion with the fam­ily in shared deci­sion-mak­ing about
HCT. Communication about the patient between the hema­tol­
ogy and HCT teams is also likely to be ben­e­fi­cial. Families may
also ben­e­fit from a con­sul­ta­tion with sup­port groups and peers
who may have nav­i­gated a sim­i­lar deci­sional dilemma.
The 8-year-old patient described in the sec­ ond vignette
illus­trates such a pro­cess. The patient had severe dis­ease man­
i­fes­ta­tions and impaired qual­ity of life. Routine HLA typ­ing had
revealed the avail­abil­ity of an HLA MSD. The fam­ily engaged in
the con­sul­ta­tive and deci­sional pro­cess and decided to pro­ceed
with HCT. Following HCT from the HLA MSD, the child is cur­rently
lead­ing a nor­mal life.
The tod­dler in the first vignette is begin­ning to man­i­fest
com­pli­ca­tions of SCD. The par­ents, how­ever, are ambiv­a­lent
about HCT. We would rec­om­mend con­tinu­ing com­pre­hen­sive
med­i­cal care, includ­ing the addi­tion of dis­ease-mod­i­fy­ing ther­
apy and sup­port for treat­ment adher­ence and psy­cho­so­cial
func­tion­ing. If the fam­ily were to elect to pro­ceed with HCT in
the future, we rec­om­mend that HCT be offered in the con­text
of a mul­ti­cen­ter clin­i­cal trial. Active con­sid­er­ation must also be
given to pro­vid­ing fer­til­ity pres­er­va­tion pro­ce­dures pre-HCT
and long-term fol­low-up post-HCT to mon­i­tor for late effects
of treat­ment.
HLA iden­ti­cal donor HCT in young chil­dren  |  193
Conflict-of-inter­est dis­clo­sure
Niketa Shah: no com­pet­ing finan­cial inter­ests to declare.
Lakshmanan Krishnamurti: no com­pet­ing finan­cial inter­ests to
declare.
Off-label drug use
Niketa Shah: no off-label drug use has been discussed.
Lakshmanan Krishnamurti: no off-label drug use has been
discussed.
Correspondence
Lakshmanan Krishnamurti, Children’s Healthcare of Atlanta–
Egleston, 1405 Clifton Road NE, Atlanta, GA 30322; e-mail:
lakshmanan​­.krishnamurti@choa​­.org; lakshmanan.krishnamurti@
emory.edu.
References
1. Stenger EO, Shenoy S, Krishnamurti L. How I treat sickle cell dis­ease with
hema­to­poi­etic cell trans­plan­ta­tion. Blood. 2019;134(25):2249-2260.
2. Hassell KL. Population esti­ma­tes of sickle cell dis­ease in the U.S. Am J Prev
Med. 2010;38(4, suppl):S512-S521.
3. Hassell KL. Sickle cell dis­ease: a con­tin­ued call to action. Am J Prev Med.
2016;51(1, suppl 1):S1-S2.
4. Salinas Cisneros G, Thein SL. Recent advances in the treat­ment of sickle
cell dis­ease. Front Physiol. 2020;11:435.
5. Ware RE, Davis BR, Schultz WH, et al. Hydroxycarbamide ver­sus chronic
trans­fu­sion for main­te­nance of transcranial Dopp­ler flow veloc­i­ties in chil­
dren with sickle cell anae­mia—TCD With Transfusions Changing to Hydroxy-
urea (TWiTCH): a multicentre, open-label, phase 3, non-infe­ri­or­ity trial.
Lancet.2016;387(10019):661-670.
6. Ware RE, McGann PT, Quinn CT. Hydroxyurea for chil­dren with sickle cell
ane­mia: pre­scribe it early and often. Pediatr Blood Cancer. 2019;66(8):
e27778.
7. Nouraie M, Darbari DS, Rana S, et al. Tricuspid regur­gi­ta­tion veloc­ity and
other bio­mark­ers of mor­tal­ity in chil­dren, ado­les­cents and young adults
with sickle cell dis­ease in the United States: the PUSH study. Am J Hematol.
2020;95(7):766-774.
8. Blair HA. Voxelotor: first approval. Drugs. 2020;80(2):209-215.
9. Wilmore DW. Food and Drug Administration approval of glu­ta­mine for
sickle cell dis­ease: suc­cess and pre­cau­tions in glu­ta­mine research. JPEN J
Parenter Enteral Nutr. 2017;41(6):912-917.
10. Carden MA, Little J. Emerging dis­ease-mod­i­fy­ing ther­a­pies for sickle cell
dis­ease. Haematologica. 2019;104(9):1710-1719.
11. Lanzkron S, Carroll CP, Haywood C Jr. Mortality rates and age at death from
sickle cell dis­ease: U.S., 1979-2005. Public Health Rep. 2013;128(2):110-116.
12. DeBaun MR, Ghafuri DL, Rodeghier M, et  al. Decreased median sur­vival
of adults with sickle cell dis­ease after adjusting for left trun­ca­tion bias: a
pooled anal­y­sis. Blood. 2019;133(6):615-617.
13. Field JJ, DeBaun MR, Yan Y, Strunk RC. Growth of lung func­tion in chil­dren
with sickle cell ane­mia. Pediatr Pulmonol. 2008;43(11):1061-1066.
14. Guasch A, Navarrete J, Nass K, Zayas CF. Glomerular involve­ment in adults
with sickle cell hemo­glo­bin­op­a­thies: prev­a­lence and clin­i­cal cor­re­lates of
pro­gres­sive renal fail­ure. J Am Soc Nephrol. 2006;17(8):2228-2235.
15. Gladwin MT, Sachdev V, Jison ML, et  al. Pulmonary hyper­ ten­
sion as a
risk fac­tor for death in patients with sickle cell dis­ease. N Engl J Med.
2004;350(9):886-895.
16. Vichinsky EP, Neumayr LD, Gold JI, et al; Neuropsychological Dysfunction
and Neuroimaging Adult Sickle Cell Anemia Study Group. Neuropsycho-
logical dys­func­tion and neuroimaging abnor­mal­i­ties in neu­ro­log­i­cally
intact adults with sickle cell ane­mia. JAMA. 2010;303(18):1823-1831.
17. Shah N, Bhor M, Xie L, et  al. Treatment pat­terns and eco­nomic bur­den
of sickle-cell dis­ ease patients pre­ scribed hydroxy­urea: a ret­ ro­spec­ tive
claims-based study. Health Qual Life Outcomes. 2019;17(1):155.
18. Walters MC, Patience M, Leisenring W, et al. Bone mar­row trans­plan­ta­tion
for sickle cell dis­ease. N Engl J Med. 1996;335(6):369-376.
19. Walters MC, Hardy K, Edwards S, et al; Multicenter Study of Bone Marrow
Transplantation for Sickle Cell Disease. Pulmonary, gonadal, and cen­tral
194  |  Hematology 2021  |  ASH Education Program
ner­vous sys­tem sta­tus after bone mar­row trans­plan­ta­tion for sickle cell dis­
ease. Biol Blood Marrow Transplant. 2010;16(2):263-272.
20. Hsieh MM, Kang EM, Fitzhugh CD, et al. Allogeneic hema­to­poi­etic stem-cell
trans­plan­ta­tion for sickle cell dis­ease. N Engl J Med. 2009;361(24):2309-
2317.
21. Gluckman E, Cappelli B, Bernaudin F, et al; Eurocord, the Pediatric Work-
ing Party of the Euro­pean Society for Blood and Marrow Transplantation,
and the Center for International Blood and Marrow Transplant Research.
Sickle cell dis­ease: an inter­na­tional sur­vey of results of HLA-iden­ti­cal sib­
ling hema­to­poi­etic stem cell trans­plan­ta­tion. Blood. 2017;129(11):1548-
1556.
22. Cappelli B, Volt F, Tozatto-Maio K, et al; Eurocord, the Cellular Therapy and
Immunobiology Working Party (CTIWP) and the Paediatric Diseases Work-
ing Party (PDWP) of the EBMT. Risk fac­tors and out­comes according to
age at trans­plan­ta­tion with an HLA-iden­ti­cal sib­ling for sickle cell dis­ease.
Haematologica. 2019;104(12):e543-e546.
23. Bernaudin F, Dalle J-H, Bories D, et al; Société Française de Greffe de Moelle
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/190/1851641/190shah.pdf by guest on 13 December 2021
et de Thérapie Cellulaire. Long-term event-free sur­vival, chi­me­rism and
fer­til­ity out­comes in 234 patients with sickle-cell ane­mia youn­ger than
30 years after myeloablative con­di­tion­ing and matched-sib­ling trans­plan­
ta­tion in France. Haematologica. 2020;105(1):91-101.
24. Brazauskas R, Scigliuolo GM, Wang H-L, et al. Risk score to pre­dict event-
free sur­ vival after hema­to­poi­
etic cell trans­ plant for sickle cell dis­ ease.
Blood. 2020;136(5):623-626.
25. Bernaudin F, Verlhac S, Peffault de Latour R, et al; DREPAGREFFE Trial Inves-
tigators. Association of matched sib­ling donor hema­to­poi­etic stem cell
trans­plan­ta­tion with transcranial doppler veloc­i­ties in chil­dren with sickle
cell ane­mia. JAMA. 2019;321(3):266-276.
26. King AA, Kamani N, Bunin N, et al. Successful matched sib­ling donor mar­
row trans­plan­ta­tion fol­low­ing reduced inten­sity con­di­tion­ing in chil­dren
with hemo­glo­bin­op­a­thies. Am J Hematol. 2015;90(12):1093-1098.
27. Hsieh MM, Fitzhugh CD, Weitzel RP, et al. Nonmyeloablative HLA-matched
sib­ling allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion for severe sickle
cell phe­no­type. JAMA. 2014;312(1):48-56.
28. Sinha CB, Bakshi N, Ross D, Loewenstein G, Krishnamurti L. Primary care­
giver deci­sion-mak­ing in hema­to­poi­etic cell trans­plan­ta­tion and gene ther­
apy for sickle cell dis­ease. Pediatr Blood Cancer. 2021;68(1):e28749.
29. Walters MC, Patience M, Leisenring W, et  al. Collaborative mul­ti­cen­ter
inves­ti­ga­tion of mar­row trans­plan­ta­tion for sickle cell dis­ease: cur­rent
results and future direc­tions. Biol Blood Marrow Transplant. 1997;3(6):310-
315.
30. Smith WR, Bovbjerg VE, Penberthy LT, et  al. Understanding pain and
improv­ing man­age­ment of sickle cell dis­ease: the PiSCES study. J Natl Med
Assoc. 2005;97(2):183-193.
31. Mentzer WC, Heller S, Pearle PR, Hackney E, Vichinsky E. Availability of
related donors for bone mar­row trans­plan­ta­tion in sickle cell ane­mia. Am J
Pediatr Hematol Oncol. 1994;16(1):27-29.
32. Gragert L, Eapen M, Williams E, et al. HLA match like­li­hoods for hema­to­poi­
etic stem-cell grafts in the U.S. reg­is­try. N Engl J Med. 2014;371(4):339-348.
33. Angelucci E, Matthes-Martin S, Baronciani D, et al; EBMT Inborn Error and
EBMT Paediatric Working Parties. Hematopoietic stem cell trans­plan­ta­tion
in thal­as­se­mia major and sickle cell dis­ease: indi­ca­tions and man­age­ment
rec­om­men­da­tions from an inter­na­tional expert panel. Haematologica.
2014;99(5):811-820.
34. Badawy SM, Beg U, Liem RI, Chaudhury S, Thompson AA. A sys­tem­atic
review of qual­ity of life in sickle cell dis­ease and thal­as­se­mia after stem cell
trans­plant or gene ther­apy. Blood Adv. 2021;5(2):570-583.
35. Gallo AM, Patil C, Adeniyi T, Hsu LL, Rondelli D, Saraf S. Health-related
qual­ity of life and per­sonal life goals of adults with sickle cell dis­ease after
hema­to­poi­etic stem cell trans­plan­ta­tion. West J Nurs Res. 2019;41(4):555-
575.
36. Saraf SL, Oh AL, Patel PR, et al. Nonmyeloablative stem cell trans­plan­ta­
tion with alemtuzumab/low-dose irra­di­a­tion to cure and improve the qual­
ity of life of adults with sickle cell dis­ease. Biol Blood Marrow Transplant.
2016;22(3):441-448.
37. Krishnamurti L, Neuberg DS, Sullivan KM, et al. Bone mar­row trans­plan­
ta­tion for ado­les­cents and young adults with sickle cell dis­ease: results
of a pro­spec­tive mul­ti­cen­ter pilot study. Am J Hematol. 2019;94(4):446-
454.
38. Arnold SD, Bhatia M, Horan J, Krishnamurti L. Haematopoietic stem cell trans­
plan­ta­tion for sickle cell dis­ease—cur­rent prac­tice and new approaches.
Br J Haematol. 2016;174(4):515-525.
39. Bhatia M, Kolva E, Cimini L, et al. Health-related qual­ity of life after allo­ge­
neic hema­to­poi­etic stem cell trans­plan­ta­tion for sickle cell dis­ease. Biol
Blood Marrow Transplant. 2015;21(4):666-672.
40. Brodigan K, Kapadia M, Frazier AL, et al. Safety of sur­gi­cal fer­til­ity pres­er­
va­tion pro­ce­dures in chil­dren prior to hema­to­poi­etic stem cell trans­plant.
Transplant Cell Ther. 2021;27(8):696.e1-696696.e4.
41. Lautz TB, Harris CJ, Laronda MM, Erickson LL, Rowell EE. A fer­til­ity pres­er­
va­tion toolkit for pedi­at­ric sur­geons car­ing for chil­dren with can­cer. Semin
Pediatr Surg. 2019;28(6):150861.
42. Joseph L, Jean C, Manceau S, et al. Effect of hydroxy­urea expo­sure before
puberty on sperm param­e­ters in males with sickle cell dis­ease. Blood.
2021;137(6):826-829.
43. Pecker LH, Hussain S, Christianson MS, Lanzkron S. Hydroxycarbamide
expo­sure and ovar­ian reserve in women with sickle cell dis­ease in the Mul-
ticenter Study of Hydroxycarbamide. Br J Haematol. 2020;191(5):880-887.
44. Kahn JM, Brazauskas R, Tecca HR, et  al. Subsequent neo­plasms and
late mor­tal­ity in chil­dren under­go­ing allo­ge­neic trans­plan­ta­tion for
non­ma­lig­nant dis­eases. Blood Adv. 2020;4(9):2084-2094.
45. Ghannam JY, Xu X, Maric I, et al. Baseline TP53 muta­tions in adults with SCD
devel­op­ing mye­loid malig­nancy fol­low­ing hema­to­poi­etic cell trans­plan­ta­
tion. Blood. 2020;135(14):1185-1188.
46. Inamoto Y, Lee SJ. Late effects of blood and mar­row trans­plan­ta­tion. Hae-
matologica. 2017;102(4):614-625.
47. Shenoy S, Angelucci E, Arnold SD, et al. Current results and future research
pri­or­i­ties in late effects after hema­to­poi­etic stem cell trans­plan­ta­tion for
chil­dren with sickle cell dis­ease and thal­as­se­mia: a con­sen­sus state­ment
from the sec­ond pedi­at­ric blood and mar­row trans­plant con­sor­tium inter­
na­tional con­fer­ence on late effects after pedi­at­ric hema­to­poi­etic stem cell
trans­plan­ta­tion. Biol Blood Marrow Transplant. 2017;23(4):552-561.
48. Shenoy S, Gaziev J, Angelucci E, et  al. Late effects screen­ing guide­lines
after hema­to­poi­etic cell trans­plan­ta­tion (HCT) for hemo­glo­bin­op­a­thy: con­
sen­sus state­ment from the sec­ond pedi­at­ric blood and mar­row trans­plant
con­sor­tium inter­na­tional con­fer­ence on late effects after pedi­at­ric HCT.
Biol Blood Marrow Transplant. 2018;24(7):1313-1321.
49. Hsieh MM, Fitzhugh CD, Tisdale JF. Allogeneic hema­to­poi­etic stem cell trans­
plan­ta­tion for sickle cell dis­ease: the time is now. Blood. 2011;118(5):1197-
1207.
50. Saraf SL, Ghimire K, Patel P, et al. Improved health care uti­li­za­tion and costs
in transplanted ver­sus non-transplanted adults with sickle cell dis­ease.
PLoS One. 2020;15(2):e0229710.
51. Arnold SD, Brazauskas R, He N, et al. Clinical risks and healthcare uti­li­za­
tion of hema­to­poi­etic cell trans­plan­ta­tion for sickle cell dis­ease in the USA
using merged data­bases. Haematologica. 2017;102(11):1823-1832.
Prakash Singh Shekhawat
52. Brendel C, Williams DA. Current and future gene ther­a­pies for hemo­glo­
bin­op­a­thies. Curr Opin Hematol. 2020;27(3):149-154.
53. Curtis SA, Shah NC. Gene ther­apy in sickle cell dis­ease: pos­si­ble util­ity and
impact. Cleve Clin J Med. 2020;87(1):28-29.
54. Demirci S, Uchida N, Tisdale JF. Gene ther­apy for sickle cell dis­ease: an
update. Cytotherapy. 2018;20(7):899-910.
55. Frangoul H, Altshuler D, Cappellini MD, et al. CRISPR-Cas9 gene editing for
sickle cell dis­ease and β-thal­as­se­mia. N Engl J Med. 2021;384(3):252-260.
56. Ikawa Y, Miccio A, Magrin E, Kwiatkowski JL, Rivella S, Cavazzana M. Gene
ther­apy of hemo­glo­bin­op­a­thies: prog­ress and future chal­lenges. Hum Mol
Genet. 2019;28(R1):R24-R30.
57. Jones RJ, DeBaun MR. Leukemia after gene ther­apy for sickle cell dis­ease:
inser­tional muta­gen­e­sis, busul­fan, both, or nei­ther. Blood. 2021;138(11):942-
947.
58. Leonard A, Tisdale J, Abra­ham A. Curative options for sickle cell dis­ease:
haploidentical stem cell trans­plan­ta­tion or gene ther­apy? Br J Haematol.
2020;189(3):408-423.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/190/1851641/190shah.pdf by guest on 13 December 2021
59. Khemani K, Ross D, Sinha C, Haight A, Bakshi N, Krishnamurti L. Experi-
ences and deci­sion mak­ing in hema­to­poi­etic stem cell trans­plant in sickle
cell dis­ease: patients’ and care­giv­ers’ per­spec­tives. Biol Blood Marrow
Transplant. 2018;24(5):1041-1048.
60. Schulz GL, Kelly KP, Holtmann M, Armer JM. Navigating deci­sional con­flict
as a fam­ily when fac­ing the deci­sion of stem cell trans­plant for a child or
ado­les­cent with sickle cell dis­ease. Patient Educ Couns. 2021;104(5):1086-
1093.
61. van Besien K, Koshy M, Anderson-Shaw L, et al. Allogeneic stem cell trans­
plan­ta­tion for sickle cell dis­ease: a study of patients’ deci­sions. Bone Mar-
row Transplant. 2001;28(6):545-549.
62. Kodish E, Lantos J, Stocking C, Singer PA, Siegler M, Johnson FL. Bone mar­
row trans­plan­ta­tion for sickle cell dis­ease: a study of par­ents’ deci­sions.
N Engl J Med. 1991;325(19):1349-1353.
63. Bakshi N, Katoch D, Sinha CB, et al. Assessment of patient and care­giver atti­
tudes and approaches to deci­sion-mak­ing regard­ing bone mar­row trans­
plant for sickle cell dis­ease: a qual­i­ta­tive study. JAMA Netw Open. 2020;3(5):
e206742.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000322
HLA iden­ti­cal donor HCT in young chil­dren  |  195
 EMERGING THERAPIES AND CONSIDERATIONS FOR SICKLE CELL DISEASE

Clinical trial considerations in sickle cell disease:

patient-reported outcomes, data elements,
and the stakeholder engagement framework

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/196/1851725/196badawy.pdf by guest on 13 December 2021

Sherif M. Badawy
Department of Pediatrics, Northwestern University Feinberg School of Medicine, Chicago, IL; and Division of Hematology, Oncology, and Stem Cell
Transplant, Ann and Robert H. Lurie Children’s Hospital of Chicago, Chicago, IL

Patients with sickle cell disease (SCD) have significant impairment in their quality of life across the life span as a con-
sequence of serious disease burden with several SCD-related complications. A number of disease-modifying therapies
are currently available, yet long-term clinical benefits in real-world settings remain unclear. Over the past few years,
a number of important initiatives have been launched to optimize clinical trials in SCD in different ways, including:
(1) established panels through a partnership between the American Society of Hematology (ASH) and the US Food and
Drug Administration; (2) the ASH Research Collaborative SCD Clinical Trials Network; (3) the PhenX Toolkit (consensus
measures for Phenotypes and eXposures) in SCD; and (4) the Cure Sickle Cell Initiative, led by the National Heart, Lung,
and Blood Institute. Electronic patient-reported outcomes assessment is highly recommended, and patient-reported
outcomes (PROs) should be evaluated in all SCD trials and reported using Standard Protocol Items Recommendations
for Interventional Trials guidelines. Patient-centered outcomes research (PCOR) approaches and meaningful stakeholder
engagement throughout the process have the potential to optimize the execution and success of clinical trials in SCD
with considerable financial value. This article reviews several clinical trial considerations in SCD related to study design
and outcomes assessment as informed by recent initiatives as well as patient-centered research approaches and stake-
holder engagement. A proposed hematology stakeholder-engagement framework for clinical trials is also discussed.

LEARNING OBJECTIVES
• Review key considerations for SCD clinical trials related to PROs, medication adherence, developmental issues in
children, and the COVID­19 pandemic
• Review efforts to optimize clinical trials and outcomes assessment in SCD, such as ASH­FDA panels, the ASH Re­
search Collaborative Clinical Trials Network, the PhenX Toolkit, and the CureSCi
• Review the evidence for patient­centered research (PCR) and stakeholder engagement and their potential role in
successful trials and cost savings

CLINICAL CASE
A 19­year­old man with hemoglobin SS disease presents
for his regular clinic visit. He has had no hospitalizations
over the past 5 years since he started taking daily hydroxy­
urea with good adherence. He believes that hydroxyurea
helped him a great deal with his quality of life, but he also
understands that not every sickle cell patient feels the
same. He has learned about other approved and emerg­
ing therapies, and he is fascinated by the science behind
them. Given his interest in pursuing a career in medicine,
he inquires about the possibility of getting involved in
sickle cell research to help other sickle cell patients bene­
ft from these therapies.
Introduction
Sickle cell disease (SCD) is an inherited hemoglobin dis­
order affecting about 100 000 individuals in the United
States and more than 20 million people worldwide, mainly
of African descent.1,2 SCD is a chronic, debilitating medical
condition that affects patients across their life span and is
associated with signifcant morbidity and early mortality.2,3

196 | Hematology 2021 | ASH Education Program

SCD patients suf­fer from a num­ber of acute and chronic com­
pli­ca­tions, includ­ing pain epi­sodes, acute chest syn­drome, car­
dio­pul­mo­nary dis­ease, kid­ney dam­age, liver impair­ment, splenic
seques­ tra­
tion, avas­
cu­lar necro­ sis, stroke, priapism, and other
end organ dam­age.2,3 These com­pli­ca­tions lead to sig­nif­i­cant
impair­ment in patient-reported out­comes (PROs) among chil­
dren and adults with SCD, espe­cially in the phys­i­cal and psy­cho­
so­cial domains.4-6 Patients with SCD have increased health care
uti­li­za­tion with fre­quent hos­pi­tal­i­za­tions and emer­gency depart­
ment vis­its.7-9 Treatment approaches include pre­ven­tive strat­e­
gies (eg, pen­i­cil­lin pro­phy­laxis, transcranial Dopp­ler screen­ing),
acute man­age­ment (eg, opi­oids), dis­ease-mod­i­fy­ing ther­a­pies
(ie, hydroxy­urea, L-glu­ta­mine, voxelotor, and crizanlizumab), and
cura­tive options (eg, hema­to­poi­etic stem cell trans­plan­ta­tion,
gene ther­apy, or gene editing).3
Some ear­lier clin­i­cal tri­als in SCD faced a num­ber of logis­ti­
cal chal­lenges related to recruit­ment and reten­tion.10-12 Barriers
to suc­cess­ful trial com­ple­tion included con­cerns from par­ents
as to the neces­sity of the research, patient belief that research
was only needed for those with more severe dis­ease, or anx­i­ety
related to pre­vi­ous research expe­ri­ence.11-13 A num­ber of facil­i­ta­
tors were also iden­ti­fied from these ear­lier stud­ies, includ­ing edu­
cat­ing peers, explaining trial ratio­nales more clearly, ­improv­ing
the read­abil­ity of con­sent/assent forms, explaining study pro­to­
cols using vid­eos and other inno­va­tive illus­tra­tions, and leverag­
ing patient-cen­ tered research (PCR) approaches that involve
patients, par­ents/care­giv­ers and other stake­hold­ers (here­to­fore
referred to as “stake­hold­ers”) across all­stages of the research
pro­cess, from plan­ning a study to the dis­sem­i­na­tion of find­ings.11-13
Despite some ini­tial dif­fi­cul­ties in research, sev­eral tri­als in SCD
to date have been suc­cess­fully com­pleted, lead­ing to sig­nif­i­cant
prog­ress in the field of SCD with new ther­a­pies now approved
by the US Food and Drug Administration (FDA), pro­vid­ing clin­i­
cal ben­e­fits to many SCD patients. Currently, there are sev­eral
active, ongo­ing clin­i­cal tri­als with novel dis­ease-mod­i­fy­ing ther­
a­pies and cura­tive approaches, such as gene ther­apy, gene edit­
ing, and hema­to­poi­etic stem cell trans­plan­ta­tion with var­i­ous
reg­i­mens and donors.14,15 Nevertheless, efforts to improve clin­i­cal
tri­als, includ­ing involv­ing stake­hold­ers in tri­als, are still ongo­ing.
Over the past few years, a num­ber of impor­tant ini­tia­tives
have been launched to opti­mize clin­i­cal tri­als in SCD in dif­fer­
ent ways. First, con­sen­sus rec­om­men­da­tions for evi­dence-based
SCD end points have been devel­oped as a result of a col­lab­o­
ra­tive effort from 7 pan­els of patients, cli­ni­cians, and research­
ers established through a part­ner­ship between the Amer­i­can
Society of Hematology (ASH) and the FDA.16,17 Second, the ASH
Research Collaborative SCD Clinical Trials Network has focused
on build­ing part­ner­ships with the SCD com­mu­nity and stake­
hold­ers, establishing col­lab­o­ra­tion across SCD cen­ters, stream­
lining clin­i­cal trial oper­a­tions with a sin­gle insti­tu­tional review
board approval, and facil­i­tat­ing data shar­ing with a cen­tral­ized
data repos­i­tory through the ASH Research Collaborative Data
Hub.18 Third, the PhenX (Phenotypes and eXposures) pro­ject has
been funded by dif­fer­ent sources, includ­ing the National Human
Genome Research Institute, the National Institute on Drug
Abuse, the Office of Behavioral and Social Sciences Research,
the National Institute of Mental Health, the National Heart, Lung,
and Blood Institute (NHLBI), the National Institute on Minority
Health and Health Disparities, the National Cancer Institute,
Prakash Singh Shekhawat
and the Tobacco Regulatory Science Program of the National
Institutes of Health (NIH). In SCD, PhenX efforts have focused
on selecting high-qual­ity SCD-related out­come mea­sures to be
included in the Toolkit (con­sen­sus mea­sures; www​­.phenxtoolkit​
­.org), guided by the Sickle Cell Disease Research and Scientific
Panel.19 Finally, the Cure Sickle Cell Initiative (CureSCi), led by
the NHLBI, has cen­tered on inno­vat­ing genetic ther­a­pies, nur­
tur­ing a col­lab­o­ra­tive, PCR envi­ron­ment, and establishing data
stan­dards for SCD clin­i­cal tri­als.20 In par­tic­u­lar, inter­est in cura­tive
ther­a­pies in the SCD com­mu­nity has been grow­ing as evi­dence
has emerged and con­tin­ues to mate­ri­al­ize from ongo­ing clin­i­cal
tri­als.
This arti­cle aims to review sev­eral clin­i­cal trial con­sid­er­ations
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/196/1851725/196badawy.pdf by guest on 13 December 2021
in SCD related to study design and out­ come assess­ ment as
informed by recent ini­tia­tives, in par­tic­u­lar PROs as well as PCR
approaches and stake­holder engage­ment.
Clinical trial con­sid­er­ations
PROs and e-PROs
PROs have been defined as “out­ comes reported directly by
patients them­ selves and not interpreted by an observer.”21
Health-related qual­ity of life (HRQOL) is a PRO that is defined as
“a mul­ti­di­men­sional con­cept that usu­ally includes self-report of
the way in which phys­i­cal, emo­tional, social, or other domains
of well-being are affected by a dis­ease or its treat­ment.”21 A
proxy- or par­ent/care­giver-reported out­come is also com­monly
used to eval­u­ate PROs and/or HRQOL among pedi­at­ric pop­u­
la­tions. Major reg­u­la­tory author­i­ties have rec­og­nized the value
of includ­ing PROs eval­u­a­tion in clin­i­cal tri­als to inform clin­i­cal
deci­sion-mak­ing, phar­ma­ceu­ti­cal-label­ing claims, and prod­uct
reim­burse­ment.22 The inclu­sion of PROs in clin­i­cal trial pro­to­cols
should be well planned in advance and reported according to
the Standard Protocol Items Recommendations for Interven­
tional Trials guide­lines.23 Evaluating PROs in clin­i­cal tri­als that
involve SCD patients pro­vi­des an oppor­tu­nity to mea­sure the
impact of a given treat­ment on their indi­vid­ual func­tion­ing and
well-being.5
A num­ber of fac­tors should be con­sid­ered when assessing
PROs in SCD clin­i­cal tri­als, such as eli­gi­bil­ity cri­te­ria (eg, age), rel­e­
vant domains of inter­est, psy­cho­met­ric prop­er­ties (eg, respon­sive­
ness, validity, reli­abil­ity, and floor/ceil­ing effects), min­i­mal clin­i­cally
impor­tant dif­fer­ences, generic vs dis­ease-spe­cific approaches,
par­tic­i­pants’ bur­den (ie, sur­vey length), and mode of admin­is­tra­
tion.5 Generic mea­sures pro­vide insight into the bur­den of SCD
com­pared to healthy indi­vid­u­als and those with other chronic
med­i­cal con­di­tions.5 On the other hand, dis­ease-spe­cific mea­
sures bet­ter exam­ine the dif­fer­ences and effects of treat­ments or
inter­ven­tions across dif­fer­ent patient groups and within indi­vid­
ual SCD patients.5 Thus, a com­bined approach using generic and
dis­ease-spe­cific instru­ments to eval­u­ate SCD patients’ PROs is
highly recommended, with a pref­er­ence for patient self-report­
ing over proxy reporting when pos­si­ble.4,5 Further, the ASH-FDA
panel for PROs has pro­vided detailed guid­ance and rec­om­men­
da­tions on PROs selec­tion in SCD clin­i­cal tri­als (Table 1).16
Finally, the mode of PROs eval­u­a­tion is another key con­sid­
er­ation. The use of elec­tronic approaches or e-PROs has been
recommended by the e-PRO con­sor­tiums of the International
Society for Quality of Life Research, the Professional Society for
Health Economics and Outcomes Research, and the reg­u­la­tors
Clinical trial con­sid­er­ations in SCD  |  197
Table 1. PROs rec­om­men­da­tions in CureSCi CDEs (ver­sion 1.0)
Subdomain Population
Pain inten­sity Adults and chil­dren ≥8 years old
Pain impact/inter­fer­ence Adults, SCD-spe­cific
Children 5-18 years, SCD-spe­cific
Children/adults, not SCD-spe­cific
Pain: mixed Children 5-18 years, SCD-spe­cific
Painful cri­ses Adults, SCD-spe­cific
Emotional impact of SCD Adults, SCD-spe­cific
Children, SCD-spe­cific
Negative affect: mixed Children, not SCD-spe­cific
Low mood Children/adults, not SCD-spe­cific
Anxiety Children/adults, not SCD-spe­cific
Fatigue Children/adults, not SCD-spe­cific
Children, SCD-spe­cific
Sleep dis­tur­bance Children/adults, not SCD-spe­cific
Adults, SCD-spe­cific
Adults, not SCD-spe­cific
Children, not SCD-spe­cific
General func­tion Adults, not SCD-spe­cific
Social func­tion Adults, SCD-spe­cific
Physical func­tion Adults, SCD-spe­cific
Adults, not SCD-spe­cific
Children, not SCD-spe­cific
Global health/QOL Adults, not SCD-spe­cific
Children, not SCD-spe­cific
Global cog­ni­tion Children, 0-3.5 years old
Children, 2.5-7 years old
Children, 6-16 years old
Adults
Children and adults†
Executive func­tion­ing Children 3-7, 8-11, and ≥12 years
Children ≥9 years old
Children and adults, 8-89 years
Children and adults, 7-89 years
Processing speed Children ≥7 years old
Adults
Working mem­ory Children ≥7 years old
*Pediatric PROMIS mea­sures are avail­­able for chil­dren self-report ≥8 years old and proxy report.
†Adult PROMIS mea­sures are avail­­able.
ASCQ-Me, Adult Sickle Cell Quality of Life Measurement Information System; CHAD, children and adolescents; NRS, Numeric Rating Scale; NT, NIH
Toolbox; PROMIS, Patient-Reported Outcomes Measurement Information System; VAS, Visual Analog Scale; WISC, Wechsler Intelligence Scale for
Children; WPPSI-IV, Wechsler Preschool and Primary Scale of Intelligence.
Publicly available at https:​­/​­/curesickle​­.org​­/sites​­/scdc​­/files​­/Doc​­/SC​­/Patient_Reported_Outcomes_Recommendations_Summary​­.pdf.
198  |  Hematology 2021  |  ASH Education Program
Measure Classification
NRS Core
VAS Supplemental
ASCQ-Me Pain Impact Core
PedsQL Pain Impact SCD mod­ule Core
PROMIS Pain Interferencea,b Core
PedsQL Pain and Hurt, SCD mod­ules Core
ASCQ-Me Pain Episodes Core
ASCQ-Me Emotional Impact Supplemental, highly recommended
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/196/1851725/196badawy.pdf by guest on 13 December 2021
PedsQL, SCD Module Emotions Supplemental, highly recommended
PedsQL, SCD Module Worrying Supplemental
PROMIS Physical Stress Experience1 Supplemental, highly recommended
PROMIS Emotional Distress: Depression 1,2
Supplemental, highly recommended
PROMIS Emotional Distress: Anxiety1,2 Supplemental, highly recommended
Pediatric/Adult PROMIS Fatigue1,2 Core
PedsQL Multidimensional Fatigue Scale Core
PROMIS Sleep Disturbance1,2 Supplemental, highly recommended
ASCQ-Me Sleep Impact Supplemental, highly recommended
Pittsburgh Sleep Quality Index Supplemental
Epworth Sleepiness Scale Supplemental
Epworth Sleepiness Scale (CHAD) Supplemental
Cana­dian Occupational Performance Supplemental
ASCQ-Me Social Functioning Impact Supplemental
ASCQ-Me Stiffness Impact Supplemental, highly recommended
PROMIS - Physical Function (PF) 12a2 Supplemental
Pediatric PROMIS - PF Mobility1 Supplemental
Pediatric PROMIS - PF Upper Extremity1 Supplemental
PROMIS 10 Global Health 2
Core
PROMIS 7 + 2 Global Health* Core
Bayley-III Supplemental
WPPSI-IV (4th edi­tion) Supplemental
WISC-V (5th edi­tion) Supplemental
Wechsler Adult Intelligence Scale Supplemental
NIH Toolbox Supplemental, highly recommended
Flanker Inhibitory Control/Attention (NT) Supplemental, highly recommended
Dimensional Change Card Sort Test (NT) Supplemental, highly recommended
Trail Making Test, parts A and B Supplemental
Delis-Kaplan Executive Function System Supplemental
Wisconsin Card Sort Test Supplemental
Pattern Comparison Processing Speed Test Supplemental, highly recommended
Processing Speed Index Supplemental
List Sorting Working Memory Test (NT) Supplemental, highly recommended
(eg, the FDA).23-25 e-PROs have the fol­low­ing advan­tages: (1) more
pre­cise, com­plete, timely, and high-qual­ity data, (2) bet­ter adher­
ence to study pro­to­col, (3) pos­si­ble PRO remind­ers and real-
time mon­i­tor­ing, (4) less recall bias, (5) fewer data entry errors,
(6) lev­er­aged com­put­er­ized adap­tive test­ing when needed,
(7) inte­ grated skip pat­ terns for rel­e­vant ques­tions, (8) ligh­
ter
work­loads for staff, (9) pos­si­ble cost sav­ings and envi­ron­men­tal
friend­li­ness with less paper print­ing, and (10) high accept­abil­ity
rat­ings from patients.5,24,26 Given the ubiq­ui­tous access to smart­
phones and tab­lets as well as the grow­ing evi­dence and accept­
abil­ity of mobile health inter­ven­tions among SCD patients,27,28
e-PROs should be strongly con­ sid­
ered in SCD clin­ i­
cal tri­
als,
whether pro­vid­ing patients with a ded­i­cated device for e-PROs
or allowing patients to down­ load an app and use their own
phone—an approach we call “Bring your own device,” or BYOD.
Medication adher­ence
Adherence to any new med­i­ca­tion is a crit­i­cal com­po­nent of the
suc­cess of any clin­i­cal trial in SCD, yet often lit­tle atten­tion is
given to ways to mon­i­tor and opti­mize adher­ence dur­ing the
course of a study. A num­ber of objec­tive and sub­jec­tive adher­
ence mea­sures can be con­sid­ered in the set­ting of a clin­i­cal
trial, which might vary based on the study design (eg, ran­dom­
ized con­trolled trial vs real-world com­par­a­tive effec­tive­ness
trial). Objective mea­ sures of med­ i­
ca­
tion adher­ ence include
bio­chem­i­cal mea­sures (eg, drug lev­els, bio­mark­ers), elec­tronic
mon­i­tor­ing (eg, elec­tronic pill bot­tles or smartphone app logs),
directly observed ther­apy (ie, in-per­son or mobile), dig­i­tal pills
(eg, Pro­teus), pill counts, and phar­macy records (eg, pre­scrip­
tion refills).29 These mea­sures pro­vide a more accu­rate view of
a patient’s adher­ ence behav­ ior but require some addi­ tional
resources. In con­trast, sub­jec­tive mea­sures of med­i­ca­tion adher­
ence include a patient’s self-reported adher­ence, using sur­veys
(eg, paper and pen­cil or elec­tronic) or inter­views, and phy­si­cian
assess­ments.29 These mea­sures are sim­ple, short, and inex­pen­
sive and can pro­vide insight into poten­tial adher­ence bar­ri­ers;
nev­ er­the­ less, social desir­ abil­
ity and recall bias are impor­ tant
con­sid­er­ations. It is worth not­ing that recent col­lab­o­ra­tive, mul­
ti­dis­ci­plin­ary efforts led to the devel­op­ment of the PROMIS Med­
ication Adherence Scale (PMAS), and its psy­cho­met­ric eval­u­a­tion
is under­way in sev­eral ongo­ing tri­als.30 PMAS is listed in C
­ ureSCi’s
com­mon data ele­ments (CDEs). Given that both objec­tive and
sub­jec­tive mea­sures of adher­ence have the poten­tial to cap­ture
var­i­ous aspects of med­i­ca­tion-tak­ing behav­ior, a mul­ti­modal
strat­ egy is highly recommended.29,31 Further, in ran­ dom­ ized
con­trolled tri­als eval­u­at­ing the effi­cacy of a new med­i­ca­tion in
SCD, it is likely advan­ta­geous to use tools that can mon­i­tor and
enhance adher­ence behav­ior to opti­mize the clin­i­cal ben­e­fits of
a given ther­apy for study par­tic­i­pants. In addi­tion, this can pro­
vide a more pre­cise assess­ment of the dif­fer­ences in study out­
comes based on expo­sure or adher­ence to either exper­i­men­tal
drug vs pla­cebo or active com­par­a­tor.
CDEs (CureSCi and PhenX Toolkit for SCD)
One of the goals of the NHLBI-led CureSCi is to stan­dard­ize data
col­lec­tion forms for all­clin­i­cal research stud­ies in SCD, includ­ing
those with prom­is­ing genetic approaches.20 In 2021 the first set
of CDEs were assem­bled and final­ized. These CDEs serve as a
crit­i­cal resource for SCD clin­i­cal tri­als in the effort to improve the
effi­ciency of clin­i­cal stud­ies, enhance data qual­ity, enable data
Prakash Singh Shekhawat
shar­ing, and edu­cate young inves­ti­ga­tors on var­i­ous aspects of
clin­i­cal research meth­od­ol­ogy.20 Table 2 includes an over­view of
the pro­posed CDEs in CureSCi. In addi­tion, all­core data ele­ments
that are essen­tial for the ini­ti­a­tion of any clin­i­cal research study
in SCD are included in a Start-Up Resource Listing doc­u­ment.20
The PhenX Toolkit for SCD is another key NHLBI-funded ini­tia­tive.
The goal of the PhenX Measures for SCD Research pro­ject is to
help research­ers bet­ter under­stand the path­o­phys­i­­ol­ogy, nat­u­ral
his­tory, and treat­ment approaches for SCD. The PhenX Toolkit in
SCD is a frame­work for out­comes assess­ment and data shar­ing
across var­i­ous SCD research pro­jects that allows for poten­tial
com­par­i­sons across stud­ies.19
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/196/1851725/196badawy.pdf by guest on 13 December 2021
Developmental issues in pedi­at­ric tri­als
Some issues should be con­ sid­
ered when plan­ ning out­ comes
assess­ment in an SCD clin­i­cal trial that involves chil­dren and
ado­les­cents, such as age-appro­pri­ate psy­cho­met­ric prop­er­ties
for PROs, patient or proxy reports or both, and devel­op­men­tal
level.12 Children and ado­les­cents expe­ri­ence many cog­ni­tive,
psy­cho­log­i­cal, and phys­i­cal changes over time and show wide
var­i­abil­ity in their emo­tional, social, atten­tional, and intel­lec­tual
lev­els of devel­op­ment.12 These dif­fer­ences have impor­tant impli­
ca­tions for pedi­at­ric clin­i­cal trial design and implementation,
espe­cially in behav­ioral and interventional areas, where a one-
size-fits-all­approach is far from ideal.
Patient-cen­tered research and stake­holder-engage­ment
frame­work in hema­tol­ogy
Stakeholder involve­ment in dif­fer­ent stages of sickle cell research
has been lim­ited, includ­ing in devel­op­ment, design, implemen­
tation, and dis­sem­i­na­tion. Most clin­i­cal tri­als in SCD have his­tor­
i­cally focused on sur­ro­gate end points, such as hos­pi­tal­i­za­tions
and emer­gency room vis­its, as mark­ers of dis­ease activ­ity, with
less empha­sis on PROs or patient-cen­tered out­comes research
(PCOR). The prob­lem with this approach is the pos­si­bil­ity of
miss­ing what patients and other stake­hold­ers care about the
most, mak­ing clin­i­cal trial find­ings less rel­e­vant to many of them,
at least in their view. The Patient-Centered Outcomes Research
Institute (PCORI) was established by the US Congress in 2010
to address this issue. Since then, PCORI has funded sev­eral SCD
pro­jects at dif­fer­ent stages and with a wide range of bud­gets
and scopes (Table 3). Other gov­ern­ment agencies also sup­port
PCOR pro­jects with var­i­ous lev­els of expected patient and stake­
holder involve­ ment (Table 4). The FDA Patient-Focused Drug
Development ini­tia­tive is another key effort to include patients’
per­spec­tives on their med­i­cal con­di­tions, the symp­toms that
have the most impact on their daily lives, and the avail­­able ther­
a­pies and to bet­ter under­stand the fac­tors that drive treat­ment
deci­sions and a will­ing­ness to par­tic­i­pate in clin­i­cal tri­als.25 More­
over, PCOR high­ lighted impor­ tant out­comes that were often
overlooked by inves­ti­ga­tors, such as patient- or proxy-reported
PROs, treat­ment sat­is­fac­tion, care­giver/par­ent bur­den, work
time off, transportation costs, and out-of-pocket costs.
PCOR pro­ jects in SCD most often focus on inves­ ti­
ga­
tor-
ini­ti­ated com­par­a­tive effec­tive­ness tri­als eval­u­at­ing dif­fer­ent
established treat­ment approaches. Involving patients and stake­
hold­ers in clin­i­cal trial deci­sions, using mea­sures such as PROs
and other out­comes, is crit­i­cal to ensure the rel­e­vance of these
assess­ments to the larger SCD com­mu­nity. Moreover, in 2017
the Clinical Trials Transformation Initiative (CTTI) outlined, by
Clinical trial con­sid­er­ations in SCD  |  199
Table 2. NIH-NHLBI CureSCi CDEs (ver­sion 1.0)

Domain Subdomain Class Recommendations

Participant char­ac­ter­is­tics
Demographics C Demographics
Baseline abnor­mal hema­to­poi­e­sis Behavioral his­tory short form
C
Transfusion his­tory Medical his­tory Surgical his­tory
General health his­tory
Behavioral his­tory Medical history supplemental elements
S
Hospitalization form Sleep assess­ment (ped form)
C Social sta­tus
Education school ques­tion­naire Social deter­mi­nants screen
Social his­tory
S ACEs screen chil­dren (1-17 years) ACEs screen adults (  ≥18 years)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/196/1851725/196badawy.pdf by guest on 13 December 2021

Acute ane­mia Chronic ane­mia
Disease and treat­ment-
Asthma out­comes instru­ment rec­om­men­da­tions (highly recommended)
related events Asthma S
Asthma out­comes Over-read spi­rom­e­try report form
Fertility/bone C Endocrine, infer­til­ity, and bone health
6-min­ute walk test Pulmonary func­tion test
C
Lung dis­ease assess­ments guide­lines
Lung
PROMIS dyspnea func­tional lim­i­ta­tions
S
Pulmonary hyper­ten­sion PROMIS dyspnea sever­ity
Pain C Acute chest syn­drome SCD-related acute pain­ful epi­sodes
C Priapism core
Priapism
S Priapism Impact Profile (PIP) Priapism ques­tion­naire
Renal C Renal func­tion assess­ments
C Acute spleen Chronic spleen
Spleen
S Spleen assess­ment from the pedi­at­ric HU phase 3 clin­i­cal trial (BABY HUG)
Leg ulcers Retinopathy Avascular necro­sis
Other S
Chronic mal­nu­tri­tion Guidelines mal­nu­tri­tion iden­ti­fi­ca­tion
Assessments and
C Cardiac MRI Echocardiogram Brain MRI
exam­i­na­tions Imaging diag­nos­tics
S Functional MRI Brain MRA Imaging TCD
C Genetic diag­nos­tic test­ing Hemoglobin var­i­ant anal­y­sis
Laboratory tests
S Immune func­tion form Lab assess­ments-genet­ics/assays
Nonimaging S Electrocardiogram
Physical exam S Physical exam NIH Stroke Scale
Vital signs C Vital signs and blood gases
Treatment and
C Prior and con­com­i­tant med­i­ca­tions
inter­ven­tions Drugs
S PROMIS Medical Adherence Scale (PMAS) Asthma med­i­ca­tions list
Drug prod­uct Hematopoietic cel­lu­lar trans­plant infu­sion
C
Genetics and assays sum­mary of rec­om­men­da­tions
Therapies S Adhesion and vis­cos­ity Apheresis Conditioning reg­i­men

E Adhesion mol­e­cules assay

Cytopenia Genotoxicity Iron over­load

C
Adverse events and Infusion-related tox­ic­ity Infection form
toxicities
Adverse events New malig­nancy Toxicity form
S
Cellular ther­apy essen­tial data fol­low-up form

200  |  Hematology 2021  |  ASH Education Program

Table 2. (continued)

Domain Subdomain Class Recommendations

PROs C See details in Table 1
Outcomes and end points
Mortality C Death form
ACEs: adverse child­hood expe­ri­ences; Class: clas­si­fi­ca­tion; C: core; E: explor­atory; HU: hydroxy­urea; MRA: mag­netic res­o­nance angi­og­ra­phy; MRI:
mag­netic res­o­nance imag­ing; Ped: pedi­at­ric; S: sup­ple­men­tal; TCD: transcranial Dopp­ler.
Publicly available at https:​­/​­/curesickle​­.org​­/system​­/files​­/Sickle_Cell_Disease_CDE_Highlight_Summary​­.pdf.

phases of research, dif­fer­ent approaches to incor­po­rate patient Involving stake­hold­ers with diverse back­grounds pro­vi­des

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/196/1851725/196badawy.pdf by guest on 13 December 2021

and stake­holder input across the con­tin­uum of a clin­i­cal trial.
The CTTI also reported some exam­ples of poten­tial ben­e­fits for
PCOR, such as enhanc­ing the rel­e­vance of research ques­tions to
patients and stake­hold­ers, choos­ing the most appro­pri­ate pri­
mary and sec­ond­ary study out­comes, improv­ing strat­e­gies for
engage­ment, recruit­ment, and reten­tion, addressing bar­ri­ers to
par­tic­i­pa­tion, keep­ing study bur­den to a min­i­mum, and opti­miz­
ing over­all clin­i­cal trial expe­ri­ence.32
the needed insight into per­sonal expe­ri­ences man­ag­ing SCD,
cul­tural con­sid­er­ations, adher­ence bar­ri­ers, and poten­tial strat­e­
gies to opti­mize the uptake of approved ther­a­pies as well as the
research approach and accept­abil­ity of study assess­ments.33-35
An impor­tant part of stake­hold­ers’ engage­ment is clar­i­fy­ing the
sci­en­tific ratio­nale for the choice of study design and exam­in­ing
the fea­si­bil­ity of includ­ing spe­cific out­comes for a given trial with
clear expec­ta­tions of time lines and lev­els of involve­ment.33-35

Table 3. Examples of SCD pro­jects funded by the PCORI

Project title Project type Budget Time line

We’ll Take the Village: Engaging the Community to Develop Better Health - Tier I Pipeline to pro­posal $15 000 2015
We’ll Take the Village: Engaging the Community to Better Health - Tier II Pipeline to pro­posal $25 000 2016-2017
OMPASS: COMmunity Participation to Advance the Sickle Cell Story Pipeline to pro­posal $50 000 2017-2018
National Sickle Cell Advocate Network (NSCAN) Engagement award $249 855 2016-2018
Tennessee Sickle Cell Disease Network Project Engagement award $249 963 2014-2017
Disseminating Results: Missed SCD Clinic Appointments and the Health Belief Model Engagement award $417 106 2019-2021
Automating Quality and Safety Benchmarking for Children: Meeting the Needs of Health PCORnet dem­on­stra­tion $1 264 641 2016-2021
Systems and Patients
Engaging Parents of Children With SCA and Providers in Shared-Decision Making for HU Research pro­ject $1 962 454 2017-2023
Comparative Effectiveness of a Decision Aid for Therapeutic Options in Sickle Cell Research pro­ject $2 143 228 2013-2018
Disease
PATient Navigator to rEduce Readmissions—The PArTNER Study Research pro­ject $2 054 803 2013-2018
Patient-Centered Comprehensive Medication Adherence Management System to Research pro­ject $2 148 331 2013-2018
Improve Effectiveness of Disease Modifying Therapy With HU in Patients With SCD
Comparing Two Ways to Help Patients With SCD Manage Pain (CaRISMA) Research pro­ject $4 343 821 2019-2024
Comparing Patient Centered Outcomes in the Management of Pain Between Research pro­ject $4 358 545 2014-2020
Emergency Departments and Dedicated Acute Care Facilities for Adults With SCD
National Pediatric Learning Health System (PEDSnet) - phase 1 PCORnet: CDRN (phase I) $6 459 893 2013-2015
Mid-South Clinical Data Research Network - phase 1 PCORnet: CDRN (phase 1) $6 672 017 2013-2015
Community Health Workers and Mobile Health for Emerging Adults Transitioning Research pro­ject $8 456 632 2017-2024
SCD Care (COMETS Trial)
Research Action for Health Network (REACHnet) PCORnet: CDRN (phase 2) $8 641 395 2015-2019
Comparative Effectiveness of Peer Mentoring Versus Structured Education-Based Research pro­ject $9 753 462 2017-2024
Transition Programming for the Management of Care Transitions in Emerging
Adults With SCD
Mid-South Clinical Data Research Network PCORnet: CDRN (phase 2) $10 064 128 2015-2019
Projects are orga­nized by level of funding sup­port from low to high.
CDRN, clin­i­cal data research net­work; HU, hydroxy­urea; PCORnet, National Patient-Centered Clinical Research Network.
Prakash Singh Shekhawat
Clinical trial con­sid­er­ations in SCD  |  201
Table 4. Various funding agencies, lev­els of patients, care­giver and stake­holder engage­ment, and poten­tial ben­efi
­ ts

Funding agency Level of engage­ment Benefits

Agency for Healthcare Research and Quality (AHRQ ) Desirable Possibly ben­e­fi­cial
Center for Medi­care and Med­ic­aid Services (CMS) Innovation Center Required Beneficial
National Institutes of Health (NIH) Potentially advan­ta­geous Possibly ben­e­fi­cial
Patient-Centered Outcomes Research Institute (PCORI) Expected Beneficial
Pharmaceutical com­pa­nies (indus­try) Potentially advan­ta­geous Beneficial
Professional soci­e­ties and orga­ni­za­tions Desirable Possibly ben­e­fi­cial
Note: Level of engage­ment is defined as the depth and the extent to which patients, care­giv­ers, and stake­hold­ers are engaged in dif­fer­ent stages of
a given research pro­ject that is pro­posed for funding by any of the listed agencies.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/196/1851725/196badawy.pdf by guest on 13 December 2021

Stakeholders may par­tic­i­pate in reg­u­lar study calls (eg, steering
com­mit­tee) and engage in detailed research dis­cus­sions in which
they can offer poten­tial solu­tions to unex­pected chal­lenges and
hur­dles hin­der­ing par­tic­i­pants’ recruit­ment, reten­tion, and fol­low-
up.33-35 Stakeholders may be given the oppor­tu­nity to con­trib­ute
to schol­arly prod­ucts from the research pro­ject and par­tic­i­
pate in edu­ca­tional ini­tia­tives for the dis­sem­i­na­tion of research
find­ings.
Figure 1  rep­re­sents  a pro­posed stake­holder engage­ment  frame-
work for clin­i­cal tri­als and research stud­ies in hema­tol­ogy, includ­ing
SCD. The first layer (base of the pyr­am ­ id) includes the broader SCD
com­mu­nity, with online engage­ment ­strat­e­gies such as polls, sur­
veys, blogs, social media, and dis­cus­sion boards. The sec­ond layer
(mid­dle) involves more planned stake­holder engage­ment for in-
depth insight into dif­fer­ent aspects of the research, with activ­i­ties
such as com­mu­nity stu­dios, focus groups, and/or inter­views. This

Research
Partners

Planned Engagement
(community studios,
focus groups, interviews)

Broad Online Engagement (polls,

surveys, blogs, social media,
and discussion boards)

Paents Caregivers Researchers Clinicians Community

Professional Policy Payers

Industry
Sociees Maker

Figure 1. Hematology stakeholder-engagement framework in clinical trials and research studies.

202  |  Hematology 2021  |  ASH Education Program

engage­ment approach is essen­tial to pro­vide a safe envi­ron­ment
for stake­hold­ers to give unbi­ased and crit­i­cal feed­back based on
their val­ues, expe­ri­ences, and back­grounds—espe­cially those who
have no internet access, are not active on social media, or have lim­
ited health lit­er­acy. Finally, the third layer (top) rep­re­sents research
part­ners who are driv­ing the clin­i­cal trial or the research study,
includ­ing inves­ti­ga­tors, indus­try part­ners, and selected, actively
engaged patients and stake­hold­ers. This hema­tol­ogy stake­holder-
engage­ment frame­work (Figure 1) cap­tures var­i­ous poten­tial stake­
hold­ers who either should or could be involved in clin­i­cal tri­als or
research stud­ies in hema­tol­ogy, with increas­ing lev­els of involve­
ment as we move toward the top of the pyr­a­mid. Engagement of
all­these part­ners across dif­fer­ent lev­els of the pyr­a­mid, includ­ing
plan might be help­ful to out­line the involve­ment of stake­hold­ers
across dif­fer­ent stages of tri­als or research pro­jects.
Historically, indus­ try-spon­
sored clin­ i­
cal tri­
als did not often
include sig­nif­i­cant spon­sors of stake­holder engage­ment beyond
small pre­lim­i­nary stud­ies, and this hes­i­tancy may have been due
to a lack of famil­iar­ity with PCOR meth­od­ol­ogy and/or the unclear
return on invest­ment of study ben­e­fits. This has changed over
the last decade, and cur­rently, a num­ber of ongo­ing clin­i­cal tri­
als in SCD have established advi­sory boards with dif­fer­ent stake­
hold­ ers, includ­ ing patients, care­ giv­ers, and advo­ cacy groups.
Furthermore, CTTI recently pro­posed an approach or a con­cep­
tual finan­cial model to eval­u­ate the value of stake­holder engage­
ment in clin­i­cal tri­als.38 CTTI’s model is based on an esti­mated

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/196/1851725/196badawy.pdf by guest on 13 December 2021

stake­hold­ers and com­mu­nity orga­ni­za­tions, is essen­tial to ensure
that clin­i­cal tri­als focus on mean­ing­ful out­comes for patients. This
frame­work also facil­i­tates col­lab­o­ra­tion and part­ner­ship between
research­ers and stake­hold­ers while pri­or­i­tiz­ing out­comes of high
value to patients.
Standardized train­ing may be needed to ensure that dif­fer­ent
stake­hold­ers are equipped with the skill set and ade­quate prep­
a­ra­tion needed to be actively involved in the trial or the pro­
ject as research part­ners.33-35 A num­ber of PCOR com­pe­ten­cies
and engage­ment prin­ci­ples have been reported in the lit­er­a­ture
(Table 5).36,37 Furthermore, establishing a detailed engage­ment
expected net pres­ent value (ENPV) incor­po­rat­ing cost, time, rev­
e­nue, and risk as cru­cial busi­ness driv­ers.38 In an exam­ple using an
oncol­ogy devel­op­ment pro­gram, the authors reported a poten­
tial mean­ing­ful impact of stake­holder engage­ment by avoiding
pro­to­col amend­ments and enhanc­ing enroll­ment, reten­tion, and
com­ple­tion of study assess­ments. This pos­i­tive impact in pre-
phase 2 and pre-phase 3 tri­als was asso­ci­ated with an increase
in NPV ($62 mil­lion and $65 mil­lion, respec­tively) and ENPV ($35
mil­lion and $75 mil­ lion), adding sub­ stan­tial finan­
cial value to
these tri­als. With a hypo­thet­i­cal ini­tial invest­ment of $100 000
ded­i­cated to opti­miz­ing stake­holder engage­ment strat­e­gies in

Table 5. PCR com­pe­ten­cies and prin­ci­ples

A. Competencies
I. Knowledge II. Skills III. Attitudes
Cultural con­text Communication Community val­ues
Knowledge about dis­ease Conflict man­age­ment Emotional intel­li­gence
Logistical con­sid­er­ations Critical think­ing General atti­tudes toward PCR
Participatory approaches Group par­tic­i­pa­tion Openness and trust
Agenda set­ting Leadership Personal attri­butes
Research meth­od­ol­ogy Project man­age­ment Personal growth
Understanding of data Teamwork Professional growth
Understanding PCR Prioritization Self-reflec­tion
B. Principles
I. Shared learn­ing expe­ri­ence Involvement of patients, care­giv­ers, and other stake­hold­ers in all­aspects of the research
Researchers and team mem­bers learn­ing about PCR meth­od­ol­ogy
Training for patients, care­giv­ers, and other stake­hold­ers on research prin­ci­ples
II. Collaborations Cultural sen­si­tiv­ity and mutual respect
Fair com­pen­sa­tion for effort and time
Inclusion and diver­sity for all­pro­ject-related activ­i­ties and part­ner­ships
Planning ahead for meet­ings, tasks, and mile­stones with real­is­tic time lines
III. Bidirectional rela­tion­ships Patient, care­giv­ers, and other stake­hold­ers are involved as research part­ners
Well-defined roles and strat­e­gies informed by col­lab­o­ra­tive dis­cus­sions
IV. Trustworthiness Clear and trans­par­ent com­mu­ni­ca­tions
Shared deci­sion-mak­ing pro­cess
Sharing infor­ma­tion and data openly

Prakash Singh Shekhawat

Clinical trial con­sid­er­ations in SCD  |  203
a clin­i­cal trial, there may be a return on invest­ment, in both NPV
and ENPV, that exceeds the invest­ment 500-fold.38
COVID-19 pan­demic and clin­i­cal tri­als
More recently, the COVID-19 pan­demic has led to dis­rup­tions in
our daily rou­tines, per­son­ally and pro­fes­sion­ally, in dif­fer­ent ways,
includ­ing interrupting the exe­cu­tion of clin­i­cal tri­als.39,40 Most insti­
tu­tions stopped new enroll­ments and allowed the con­tin­ua ­ ­tion
of interventional tri­als when there were poten­tial clin­i­cal ben­e­fits
for the par­tic­i­pants; how­ever, many reported chal­lenges related
to delayed and rescheduled study vis­its, pro­ce­dures, and assess­
ments and over­all dif­fi­culty reaching patients.40 Many of these
vul­ner­a­ble patients were at risk from expo­sure to COVID-19, and
some were inten­tion­ally avoiding health care facil­i­ties or obey­ing
stay-at-home-orders.39 This sit­u­a­tion high­lights the need for con­
sid­er­able adapt­abil­ity using a hybrid strat­egy in design­ing future
clin­i­cal tri­als to com­plete all­ planned study pro­ce­dures.40 Some
pro­posed strat­e­gies include (1) pri­or­i­tiz­ing pri­mary out­comes
over explor­ atory ones; (2) alter­ nat­
ing strat­e­
gies for out­ comes
assess­ment; (3) collecting remote data using phone inter­views or
online tools; (4) obtaining phone num­bers and e-mail addresses
for patients and 3 fam­ily mem­bers or friends to ensure maintained
con­tact; (5) using dif­fer­ent meth­ods to con­tact par­tic­i­pants,
includ­ing text mes­sag­ing, phone calls, e-mail, or social media;
(6) employing telemedicine; (7) arranging home vis­its by health
care work­ers wear­ing per­sonal pro­tec­tive equip­ment; (8) allow­
ing study med­ic ­ a­tions to be taken at home; (9) mak­ing use of
con­cierge ser­vices; (10) using local instead of cen­tral lab facil­
i­ties; and (11) esca­lat­ing incen­tives.12,40,41 Other approaches
should be con­sid­ered to achieve the highest level of reten­tion
and adher­ence to study inter­ven­tions, and the sta­tis­ti­cal anal­y­
sis plan for pri­mary and sec­ond­ary out­comes should be revised
to reflect any expected mean­ing­ful effects or influ­ence relate
to the pan­demic.41 For behav­ioral clin­ic ­ al tri­als, efforts should
be directed to lever­ age widely avail­­able and user-friendly
online sur­veys, data­bases, and web-based appli­ca­tions to opti­
mize e-consenting and remote enroll­ment, with com­ple­tion
of all­study assess­ments and deliv­ery of study inter­ven­tions
­conducted vir­tu­ally.
CLINICAL CASE (Con­t in­u ed)
The patient has been actively involved as an advi­sory board
mem­ber in a few inves­ti­ga­tor-ini­ti­ated tri­als at our insti­tu­tion. He
had sig­nif­i­cant input in these tri­als, such as feed­back on con­sent
and assent forms and the selec­tion of HRQOL domains that are
more rel­e­vant to SCD patients. In addi­tion, he was ­able to pro­
vide crit­i­cal insight to inform our efforts to develop and refine
a mobile app for ado­les­cents and young adults as a behav­ioral
inter­ven­tion to mon­i­tor and improve med­i­ca­tion adher­ence and
HRQOL. His pas­sion for med­i­cine was reinforced, and he is deter­
mined to become a hema­tol­o­gist car­ing for adult SCD patients.
Conclusions
Several clin­i­cal trial con­sid­er­ations in SCD are key to suc­cess.
PROs are sig­nif­i­cantly impaired among SCD patients and should
be included in all­clin­i­cal tri­als. A mul­ti­modal strat­egy is highly
204  |  Hematology 2021  |  ASH Education Program
recommended to assess adher­ence out­comes. Developmental
dif­fer­ences among chil­dren and ado­les­cents with SCD should
inform the study approach. Engaging patients and stake­hold­
ers in SCD clin­i­cal tri­als in mean­ing­ful ways is crit­i­cal to ensure
that their voices are heard and that study designs and out­
comes are rel­e­vant to them, which is essen­tial for future suc­
cess­ful dis­sem­i­na­tion and implementation. This engage­ment is a
dynamic, bidi­rec­tional com­mit­ment that is mutu­ally ben­e­fi­cial to
all­involved part­ners, and it has the poten­tial to improve health
out­comes in the larger pop­u­la­tion of pedi­at­ric and adult SCD
patients. Stakeholders can play a major role in clos­ing the gap
between data-heavy research find­ings from clin­i­cal tri­als and
their impli­ca­tions in clin­i­cal prac­tice. It is crit­i­cal to keep stake­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/196/1851725/196badawy.pdf by guest on 13 December 2021
hold­ers engaged and inter­ested through­out the research pro­
cess, and the sus­tain­abil­ity of this part­ner­ship is key. Evaluating
the ­finan­cial value of stake­holder engage­ment is impor­tant to
esti­mate the poten­tial cost sav­ings for SCD clin­i­cal tri­als, which
might be of ­con­sid­er­able finan­cial value. Timely adap­ta­tions to
address unusual cir­cum­stances, such as the COVID-19 pan­demic,
are often cru­cial.
Acknowledgments
This pro­ject was supported by a grant (K23HL150232, PI: Bad­
awy) from the National Heart, Lung, and Blood Institute of the
National Institutes of Health. The con­tent is solely the respon­si­
bil­ity of the author and does not nec­es­sar­ily rep­re­sent the views
of the National Institutes of Health.
I would like to thank Drs. Robert Liem, Jane Holl, David Cella,
Tonya Palermo, and Alexis Thompson for their advice while writ­
ing this arti­cle.
Conflict-of-inter­est dis­clo­sure
Sherif M. Badawy: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Sherif M. Badawy: nothing to disclose.
Correspondence
Sherif M. Badawy, Ann and Robert H. Lurie Children’s Hospital of
Chicago, 225 E Chicago Ave, Box #30, Chicago, IL 60611; e-mail:
sbadawy@luriechildrens​­.org.
References
1. Hassell KL. Population esti­ma­tes of sickle cell dis­ease in the U.S. Am J Prev
Med. 2010;38(4):S512-S521.
2. Piel FB, Steinberg MH, Rees DC. Sickle cell dis­ ease. N Engl J Med.
2017;376(16):1561-1573.
3. Kato GJ, Piel FB, Reid CD, et al. Sickle cell dis­ease. Nat Rev Dis Primers.
2018;4(March):18010.
4. Panepinto JA. Health-related qual­ity of life in patients with hemo­glo­bin­
op­a­thies. Hematology Am Soc Hematol Educ Program. 2012;2012(Jan­u­
ary):284-289.
5. King AA, Badawy SM, Panepinto J, Anie K, Jonassaint C, Treadwell M. Psy­
chosocial bur­den in sickle cell dis­ease. In: Gladwin M, Novelli E, Kato G,
eds. Sickle Cell Disease. McGraw Hill; 2021 558-578.
6. Badawy SM, Thompson AA, Lai JS, Penedo FJ, Rychlik K, Liem RI. Health-
related qual­ity of life and adher­ence to hydroxy­urea in ado­les­cents and
young adults with sickle cell dis­ease. Pediatr Blood Cancer. 2017;64(6):
e26369.
7. Raphael JL, Mei M, Mueller BU, Giordano T. High resource hos­pi­tal­i­za­tions
among chil­dren with vaso-occlu­sive cri­ses in sickle cell dis­ease. Pediatr
Blood Cancer. 2012;58(4):584-590.
8. Brousseau DC, Owens PL, Mosso AL, Panepinto JA, Steiner CA. Acute
care uti­li­za­tion and rehospitalizations for sickle cell dis­ease. JAMA. 2010;​
303(13):1288-1294.
9. Badawy SM, Thompson AA, Holl JL, Penedo FJ, Liem RI. Healthcare uti­li­
za­tion and hydroxy­urea adher­ence in youth with sickle cell dis­ease. Pediatr
Hematol Oncol. 2018;35(5-6):297-308.
10. Lebensburger JD, Hilliard LM, Pair LE, Oster R, Howard TH, Cutter GR. Sys­
tematic review of interventional sickle cell tri­als reg­is­tered in ClinicalTrials​
­.gov. Clin Trials. 2015;12(6):575-583.
11. Lebensburger JD, Sidonio RF, Debaun MR, Safford MM, Howard TH, Scarinci
IC. Exploring bar­ ri­
ers and facil­i­
ta­
tors to clin­
i­
cal trial enroll­
ment in the
con­ text of sickle cell ane­ mia and hydroxy­ urea. Pediatr Blood Cancer.
2013;60(8):1333-1337.
12. Palermo TM, Kashikar-Zuck S, Friedrichsdorf SJ, Powers SW. Special con­sid­
er­ations in conducting clin­i­cal tri­als of chronic pain man­age­ment inter­ven­
tions in chil­dren and ado­les­cents and their fam­i­lies. Pain Rep. 2019;4(3):e649.
13. Liem RI, Cole AH, Pelligra SA, Mason M, Thompson AA. Parental atti­tudes
toward research par­tic­i­pa­tion in pedi­at­ric sickle cell dis­ease. Pediatr Blood
Cancer. 2010;55(1):129-133.
14. Carden MA, Little J. Emerging dis­ease-mod­i­fy­ing ther­a­pies for sickle cell
dis­ease. Haematologica. 2019;104(9):1710-1719.
15. Leonard A, Tisdale J, Abra­ham A. Curative options for sickle cell dis­ease:
haploidentical stem cell trans­plan­ta­tion or gene ther­apy? Br J Haematol.
2020;189(3):408-423.
16. Farrell AT, Panepinto J, Carroll CP, et al. End points for sickle cell dis­ease
clin­i­cal tri­als: patient-reported out­comes, pain, and the brain. Blood Adv.
2019;3(23):3982-4001.
17. Farrell AT, Panepinto J, Desai AA, et  al. End points for sickle cell dis­ease
clin­i­cal tri­als: renal and car­dio­pul­mo­nary, cure, and low-resource set­tings.
Blood Adv. 2019;3(23):4002-4020.
18. Amer­i­can Society of Hematology Research Collaborative. Clinical tri­als net­
work. Accessed 3 Octo­ber 2020. https:​­/​­/www​­.ashre​searchcollaborative​
­.org​­/s​­/​clinical​­-trials​­-network.
19. Eckman JR, Hassell KL, Huggins W, et al. Standard mea­sures for sickle cell
dis­ease research: the PhenX Toolkit sickle cell dis­ease col­lec­tions. Blood
Adv. 2017;1(27):2703-2711.
20. National Heart, Lung, and Blood Institute. Cure sickle cell ini­ tia­
tive.
Accessed 3 Octo­ber 2021. https:​­/​­/www​­.nhlbi​­.nih​­.gov​­/science​­/cure​­-sickle​
­-cell​­-initiative.
21. Calvert M, Blazeby J, Altman DG, et al; Consolidated Standards of Report­
ing Trials Patient-Reported Outcomes Group. Reporting of patient-
reported out­comes in ran­dom­ized tri­als: the CONSORT PRO exten­sion.
JAMA. 2013;309(8):814-822.
22. Tunis SR, Stryer DB, Clancy CM. Practical clin­i­cal tri­als: increas­ing the value
of clin­i­cal research for deci­sion mak­ing in clin­i­cal and health pol­icy. JAMA.
2003;290(12):1624-1632.
23. Calvert M, Kyte D, Mercieca-Bebber R, et al; Standard Protocol Items: Rec­
ommendations for Clinical Trials Patient-Reported Outcome Group. Guide­
lines for inclu­sion of patient-reported out­comes in clin­i­cal trial pro­to­cols:
the SPIRIT-PRO exten­sion. JAMA. 2018;319(5):483-494.
24. Coons SJ, Gwaltney CJ, Hays RD, et al; International Society for Pharma­
coeconomics and Outcomes Research ePRO Task Force. Recommenda­
tions on evi­dence needed to sup­port mea­sure­ment equiv­a­lence between
elec­tronic and paper-based patient-reported out­come (PRO) mea­sures:
ISPOR ePRO Good Research Practices Task Force report. Value Health.
2009;12(4):419-429.
25. US Food and Drug Administration. The voice of the patient: a series of
reports from the U.S. Food and Drug Administration’s (FDA’s) patient-
focused drug devel­op­ment ini­tia­tive. https://www.fda.gov/media/89898/
Prakash Singh Shekhawat
download; https://www.fda.gov/industry/prescription-drug-user-fee-​
ame​nd​ments/​fda-led-patient-focused-drug-development-pfdd-public-
meetings#sicklecell.
26. Coons SJ, Eremenco S, Lundy JJ, O’Donohoe P, O’Gorman H, Malizia W.
Capturing patient-reported out­come (PRO) data elec­tron­i­cally: the past,
pres­ ent, and prom­ ise of ePRO mea­ sure­ment in clin­i­
cal tri­
als. Patient.
2015;8(4):301-309.
27. Badawy SM, Thompson AA, Liem RI. Technology access and smartphone
app pref­er­ences for med­i­ca­tion adher­ence in ado­les­cents and young
adults with sickle cell dis­ease. Pediatr Blood Cancer. 2016;63(5):848-852.
28. Badawy SM, Cronin RM, Hankins J, et al. Patient-cen­tered eHealth inter­ven­
tions for chil­dren, ado­les­cents, and adults with sickle cell dis­ease: sys­tem­
atic review. J Med Inter­net Res. 2018;20(7):e10940.
29. Lehmann A, Aslani P, Ahmed R, et  al. Assessing med­i­ca­tion adher­ence:
options to con­sider. Int J Clin Pharm. 2014;36(1):55-69.
30. Peipert JD, Badawy SM, Baik SH, et  al. Development of the NIH patient-
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/196/1851725/196badawy.pdf by guest on 13 December 2021
reported out­comes mea­sure­ment infor­ma­tion sys­tem (PROMIS) ­med­i­ca­tion
adher­ence scale (PMAS). Patient Prefer Adherence. 2020;14(9 June):
971-983.
31. Hansen RA, Kim MM, Song L, Tu W, Wu J, Murray MD. Comparison of meth­
ods to assess med­i­ca­tion adher­ence and clas­sify nonadherence. Ann Phar-
macother. 2009;43(3):413-422.
32. Bloom D, Beetsch J, Harker M, et al. The rules of engage­ment: CTTI rec­om­
men­da­tions for suc­cess­ful col­lab­o­ra­tions between spon­sors and patient
groups around clin­i­cal tri­als. Ther Innov Regul Sci. 2018;52(2):206-213.
33. Concannon TW, Grant S, Welch V, et al; Multi Stakeholder ­Engagement
Consortium. Practical guid­
­ ance for involv­ing stake­ hold­
ers in health
research. J Gen Intern Med. 2019;34(3):458-463.
34. Forsythe L, Heckert A, Margolis MK, Schrandt S, Frank L. Methods and
impact of engage­ ment in research, from the­ ory to prac­ tice and back
again: early find­ings from the Patient-Centered Outcomes Research Insti­
tute. Qual Life Res. 2018;27(1):17-31.
35. Forsythe LP, Ellis LE, Edmundson L, et al. Patient and stake­holder engage­
ment in the PCORI pilot pro­jects: descrip­tion and les­sons learned. J Gen
Intern Med. 2016;31(1):13-21.
36. Frisch N, Atherton P, Doyle-Waters MM, et  al. Patient-ori­ented research
com­pe­ten­cies in health (PORCH) for research­ers, patients, healthcare pro­
vid­ers, and deci­sion-mak­ers: results of a scop­ing review. Res Involv Enga-
gem. 2020;6(10 Feb­ru­ary):4.
37. Sheridan S, Schrandt S, Forsythe L, Hilliard TS, Paez KA; Advisory Panel
on Patient Engagement (2013 Inaugural Panel). The PCORI engage­ment
rubric: prom­ is­
ing prac­ tices for partnering in research. Ann Fam Med.
2017;15(2):165-170.
38. Levitan B, Getz K, Eisenstein EL, et al. Assessing the finan­cial value of patient
engage­ ment: a quan­ ti­
ta­
tive approach from CTTI’s patient groups and
clin­i­cal tri­als pro­ject. Ther Innov Regul Sci. 2018;52(2):220-229.
39. van Dorn A. COVID-19 and readjusting clin­ i­
cal tri­
als. Lancet. 2020;
396(10250):523-524.
40. McDermott MM, Newman AB. Preserving clin­i­cal trial integ­rity dur­ing the
coronavirus pan­demic. JAMA. 2020;323(21):2135-2136.
41. Flem­ing TR, Labriola D, Wittes J. Conducting clin­i­cal research dur­ing the
COVID-19 pan­demic: protecting sci­en­tific integ­rity. JAMA. 2020;324(1):
33-34.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000252
Clinical trial con­sid­er­ations in SCD  |  205
 HEMOPHILIA UPDATE: OUR CUP RUNNETH OVER

How do we optimally utilize factor concentrates

in persons with hemophilia?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/206/1851572/206lim.pdf by guest on 13 December 2021

Ming Y. Lim
Division of Hematology and Hematologic Malignancies, Department of Internal Medicine, University of Utah, Salt Lake City, UT

The current mainstay of therapy for hemophilia is to replace the deficient clotting factor with the intravenous administra-
tion of exogenous clotting factor concentrates. Prophylaxis factor replacement therapy is now considered the standard
of care in both pediatric and adult patients with hemophilia with a severe phenotype to protect musculoskeletal health
and improve quality of life. Heterogeneity in bleeding presentation among patients with hemophilia due to genetic,
environmental, and treatment-related factors has been well described. Accordingly, the World Federation of Hemophilia
recommends an individualized prophylaxis regimen that considers the factors mentioned above to meet the clinical
needs of the patient, which can vary over time. This review focuses on the practical points of choosing the type of factor
concentrate, dose, and interval while evaluating appropriate target trough factor levels and bleeding triggers such as
level of physical activity and joint status. We also discuss the use of a pharmacokinetics assessment and its incorporation
in the clinic for a tailored approach toward individualized management. Overall, adopting an individualized prophylaxis
regimen leads to an optimal utilization of factor concentrates with maximum efficacy and minimum waste.

LEARNING OBJECTIVES
• Describe the different environmental and treatment-related variables involved in determining an appropriate pro-
phylaxis regimen
• Understand the importance of an individualized prophylaxis regimen that meets the clinical needs of hemophilia
patients

CLINICAL CASE Introduction

A 15-year-old adolescent boy with moderate hemophilia A
(HA; baseline factor VIII 2%) presented to a hematology clinic
to discuss whether he should start prophylaxis with clotting
factor concentrates (CFCs). He was diagnosed at the age of
3 by his astute pediatrician when he developed severe right
knee hemarthrosis after tripping while running. Throughout
his childhood and adolescent years, he required infrequent
on-demand administration of CFCs predominantly due to
trauma-induced acute knee hemarthrosis from playing soft-
ball and basketball. He recalled 3 episodes of spontaneous
knee hemarthrosis occurring about once per year in the last
3 years. He was now in high school and wanted to play bas-
ketball competitively, with 3 practice games per week and
more during tournament season. He was concerned about
his increased risk of bleeding given his history of spontane-
ous joint bleeds and the high likelihood of musculoskeletal
injuries, such as sprained ankles and knee ligament tears,
that can occur while playing basketball.
HA and hemophilia B (HB) are X-linked bleeding disorders
that result from decreased or deficient plasma clotting fac-
tors VIII (FVIII) and IX (FIX), respectively. The severity of
hemophilia has traditionally been defined based on the de-
gree of the clotting factor deficiency, with levels of >5% to
40% classified as mild hemophilia, 1% to 5% as moderate,
and <1% as severe.1 The severity and frequency of bleeding
in hemophilia typically correlate with the degree of the
clotting factor deficiency (Table 1).
The mainstay of therapy for hemophilia is to replace
the deficient clotting factor, usually by intravenous admin-
istration of exogenous CFC. This can be episodic to treat
bleeding events or on a regular basis (prophylaxis) to pre-
vent bleeding episodes.1 Heterogeneity in bleeding pre-
sentation between patients with similar severity of disease
or between patients with HA and HB is well recognized.2-4
Similarly, patients with moderate hemophilia can present
phenotypically as those with severe disease.5 The basis
for this phenotypic variation is multifactorial and includes

206 | Hematology 2021 | ASH Education Program

Table 1.  Association of bleed­ing man­i­fes­ta­tions with sever­ity
of dis­ease based on plasma clot­ting fac­tor lev­els
Severity Plasma clot­ting fac­tor lev­els Bleeding man­i­fes­ta­tions
Mild 5%-<40% of nor­mal Bleeding with major
trauma or sur­gery; rare
spon­ta­ne­ous bleed­ing
Moderate 1%-5% of nor­mal Bleeding with minor trauma
or sur­gery; occa­sional
spon­ta­ne­ous bleed­ing
Severe <1% of nor­mal Spontaneous bleed­ing ­
pre­dom­i­nantly in the joints
or mus­cles but can occur
any­where, includ­ing
life-threat­en­ing intra­cra­nial
bleed
Adapted with permission from Srivastava, A, Santagostino, E, Dougall, A,
et al. WFH Guidelines for the Management of Hemophilia, 3rd edition.9
© 2020 World Federation of Hemophilia. Haemophilia © 2020 John
Wiley & Sons Ltd.
(1) genetic fac­tors such as the F8/F9 geno­type (null vs nonnull
var­i­ants),6 con­com­i­tant thrombophilic gene var­i­ants or other
bleed­ing dis­or­ders,7 and intrin­sic pro- and anti­co­ag­u­lant activ­
ity,6 and (2) envi­ron­men­tal fac­tors that can vary for an indi­vid­
ual over time, such as func­tional sta­tus, lev­els and pat­terns of
phys­i­cal activ­ity, and joint sta­tus.8 Rather than a one-size-fits-all­
approach, the opti­mal uti­li­za­tion of CFC takes into account both
genetic and envi­ron­men­tal fac­tors as well as treat­ment-related
fac­tors such as fac­tor type and dose, indi­vid­ual and prod­uct
phar­ma­co­ki­net­ics (PK) of fac­tor, and indi­vid­ual pref­er­ence and
adher­ence (Figure 1).
In this review, we focus on pro­phy­laxis fac­tor replace­ment
ther­apy and prac­ti­cal points around choos­ing the type of con­
cen­trate, dose, and inter­val while con­sid­er­ing trough fac­tor lev­
els and bleed­ing trig­gers. Finally, we briefly dis­cuss the use of
PK assess­ment for indi­vid­u­al­ized man­age­ment and its incor­po­
ra­tion in the clinic. The use of CFC in the man­age­ment of hemo­
philia patients with inhib­i­tors and the opti­mal uti­li­za­tion of fac­tor
replace­ment ther­apy dur­ing sur­gery are beyond the scope of
this paper. Readers are directed to the World Federation of
Hemophilia (WFH) guide­lines on the dos­age and dura­tion of CFC
cov­er­age for major and minor sur­ger­ies (Srivastava et al; Table
7-2).9 During the perioperative period, both stan­dard half-life
(SHL) and extended half-life (EHL) CFC can be given via con­tin­u­
ous infu­sion or inter­mit­tent bolus.10-12
Prophylaxis as the stan­dard of care
Since com­ ple­
tion of the Joint Outcome Study,13 and ESPRIT14
study for children with severe HA, and the SPINART study15 for
adolescents and adults with severe HA, the use of pro­phy­laxis
fac­tor replace­ment ther­apy is now con­sid­ered the stan­dard of
care. The recently updated WFH guide­lines rec­om­mend that
pedi­at­ric and adult patients with hemo­philia with a severe phe­
no­type (to include those with mod­er­ate hemo­philia with a se-
vere phe­no­type) be on pro­phy­laxis to pre­vent spon­ta­ne­ous and
break­through bleed­ing at all­times.9 Specifically for chil­dren, the
WFH guide­lines rec­om­mend the early ini­ti­a­tion of pro­phy­laxis
ide­ally before age 3, as this has been shown to have bet­ter long-
term joint out­comes com­pared to starting after age 6.9,16,17
Prakash Singh Shekhawat
Types of fac­tor con­cen­trates
For a rare dis­ or­der, the arma­ men­ tar­
ium of CFCs avail­­able is
note­wor­thy, with a wide range of prod­ucts in use around the
world. The 2 main types of SHL CFC are the virally inactivated
plasma-derived lyoph­i­lized fac­tor con­cen­trates and the recom­
bi­nant fac­tor con­cen­trates manufactured from genet­i­cally engi-
neered cells; both have high clin­i­cal effi­cacy and safety.18 Table 2
lists the CFCs cur­rently licensed for use in the United States. An
updated list of all­cur­rently avail­­able prod­ucts glob­ally and their
manufactur­ing details are avail­­able in the WFH Online Registry of
Clotting Factor Concentrates.19
For HA patients, there is a clin­i­cal con­cern that recom­bi­nant
FVIII con­cen­trates may con­trib­ute to a higher rate of inhib­i­tor
devel­op­ment in pre­vi­ously untreated patients (PUPs) com­pared
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/206/1851572/206lim.pdf by guest on 13 December 2021
to plasma-derived FVIII con­ cen­trates, although this remains
highly con­tro­ver­sial.20,21 Conversely, the risk of inhib­i­tor devel­
op­ment in HB patients is not thought to be related to the type
of CFC.9 Inhibitors are immu­no­glob­u­lin G (IgG) alloantibodies to
exog­e­nous FVIII or FIX that neu­tral­ize the func­tion of infused CFC
and mostly occur within the first 50 expo­sures.1,22 An expo­sure is
defined as any infu­sion of a FVIII/FIX-containing prod­uct in a 24-
hour period.1 Meta-ana­ly­ses of obser­va­tional stud­ies, includ­ing
patient-level data, found no dif­fer­ences in inhib­i­tor rates when
com­par­ing all­ plasma-derived FVIII con­cen­trates to all­ recom­bi­
nant FVIII con­cen­trates.20,23,24 In con­trast, the much-awaited Sur-
vey of Inhibitors in Plasma-Product Exposed Toddlers (SIPPET)
study, the only pro­spec­tive, ran­dom­ized con­trolled trial, found
that PUPs treated with recom­bi­nant FVIII prod­ucts had a higher
cumu­ la­
tive inci­
dence of inhib­ i­
tors (44.5%; 95% CI, 34.7-54.3)
com­pared with those treated with plasma-derived FVIII prod­ucts
(26.8%; 95% CI, 18.4-35.2).21 The gen­er­al­iz­abil­ity and appli­ca­bil­
ity of SIPPET have been widely debated in numer­ous con­fer­ences
and review arti­cles and is beyond the scope of this review. The
inter­ested reader is directed to the National Hemophilia Foun-
dation’s Medical and Scientific Advisory Council doc­u­ment 243
for a detailed expla­na­tion of the dif­fer­ences between SIPPET
and prior obser­va­tional stud­ies and their rec­om­men­da­tions.25
Additionally, the Euro­ pean Medicines Agency Pharmacovigi-
lance Risk Assessment Committee con­cluded in 2017 that no
clear evi­dence exists of inhib­i­tor risk dif­fer­ences between plas-
ma-derived and recom­bi­nant fac­tor VIII con­cen­trates.26 Ongo-
ing research through the Amer­i­can Thrombosis and Hemostasis
Network 8 PUPs Matter study, a lon­gi­tu­di­nal obser­va­tional study
collecting treat­ment and inhib­i­tor data on chil­dren with mod­
er­ate and severe HA or HB born on or after 1 Jan­u­ary 2010, may
help pro­vide addi­tional clar­ity on this issue.27 At this time, the
WFH does not express a pref­er­ence for prod­uct type and rec­
om­mends prod­uct selec­tion based on local avail­abil­ity, cost, and
pro­vider/patient pref­er­ence.9
For pre­vi­ously treated patients, clas­si­cally defined as hav­
ing received >150 expo­sure (although oth­ers have con­sid­ered
it to be >50 expo­sure),28 the risk of inhib­i­tor devel­op­ment is
low.22 Additionally, stud­ies indi­cate no increased risk of inhib­
i­tor devel­op­ment when switching to another type or brand
of fac­tor.29,30 Whether to use plasma-derived vs recom­bi­nant
fac­tor con­cen­trates is a mat­ter of dealer’s choice. Practically,
regional avail­abil­ity and pay­ers (gov­ern­ment or insur­ance
com­pa­nies) pref­er­ence due to nego­ti­ated pro­cure­ment dis­
counts (as well as out-of-pocket cost for US patients) play a
Optimal pro­phy­laxis fac­tor replace­ment ther­apy | 207

Figure 1.  Factors that lead to optimal utilization of CFC. Created with BioRender​.com.
larger role in deter­min­ing the spe­cific prod­uct of CFC to pre­
scribe within each type.
Dose and inter­val
The dose and inter­val of a pro­phy­laxis reg­i­men depend on the
half-life of the CFC—the time for fac­tor con­cen­tra­tion to reach
half of its orig­i­nal peak con­cen­tra­tion. In adults, the half-life can
range between 8 and 12 hours for FVIII con­cen­trates and 18 and
30 hours for FIX con­cen­trates, depending on the brand of fac­
tor; in chil­dren the half-life is shorter.31,32 Although most man­u­
fac­tur­ers pro­vide recommended starting doses for pro­phy­laxis
(Table 2), mul­ti­ple reg­im­ ens exist and vary widely. These reg­i­
mens range from a high-dose approach that involves the admin­
is­tra­tion of 25 to 40 IU/kg per dose every other day (for SHL FVIII
con­cen­trates) or twice per week (for SHL FIX con­cen­trates) to a
low-dose approach that involves once-weekly or twice-weekly
infu­sion and/or using lower doses to a reg­i­men in between these
2 approaches. The low-dose approach is pre­dom­i­nantly used in
resource-constrained countries where access to CFCs is lim­ited.
The ben­e­fits of such an approach over epi­sodic treat­ment have
been dem­on­strated with­out incur­ring a huge finan­cial bur­den.33
Thus, depending on local resources and eco­nomic con­straints,
208  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/206/1851572/206lim.pdf by guest on 13 December 2021
a starting reg­i­men of one of the afore­men­tioned approaches
is selected that is mutu­ ally accept­
able to both cli­
ni­
cian and
patient/care­giver and then tai­lored (esca­lated or de-esca­lated)
to the patient’s clin­i­cal needs.
Target trough lev­els
Historically, the goal of pro­ phy­laxis replace­
ment ther­ apy has
been to tar­get a trough fac­tor level of at least 1% (ie, convert-
ing a patient with severe hemo­philia to the typ­i­cal bleed­ing
­phe­no­type of mod­er­ate hemo­philia) to pre­vent spon­ta­ne­ous
joint bleed and to reduce dam­age to mus­cu­lo­skel­et­ al health.9,31,34
However, given the afore­men­tioned phe­no­typic var­i­a­tions ob-
served, the tra­di­tional tar­get trough level of 1% does not pre­vent
bleed­ing in all­hemo­philia patients.35 Also, for patients with mod­
er­ate hemo­philia with fre­quent spon­ta­ne­ous bleed­ing (a severe
phe­no­type), a tar­get trough level of 1% is clin­i­cally unhelp­ful.36
So what should the tar­get trough level be to achieve zero
bleeds for all­and is this attain­­able? Studies in mild and mod­er­
ate hemo­ philia patients observed that the annual num­ ber of
joint bleeds decreases as base­line fac­tor activ­ity increases and
approaches zero with a base­line fac­tor activ­ity of >15% to 30%.4,37
Predictive mod­el­ing stud­ies in severe HA patients on pro­phy­laxis
Table 2.  CFCs licensed in the US to treat HA and HB

Brand (non­pro­pri­etary name)
Factor VIII prod­ucts
  Kogenate FS® (octocog alfa)
 Advate® (octocog alfa)
 Kovaltry (octocog alfa)
®
 NovoEight® (turoctocog alfa)
  Xyntha®/ReFacto AF®
Technology
Second gen­er­a­tion: full-length recom­bi­nant
Third gen­er­a­tion: full-length recom­bi­nant
Third gen­er­a­tion: full-length recom­bi­nant
Third gen­er­a­tion: B-domain trun­cated
Third gen­er­a­tion: B-domain deleted
Manufacturer-recommended starting dose for pro­phy­laxis use*
Adults: 25 IU/kg 3 times weekly; chil­dren: 25 IU/kg every other day
20-40 IU/kg every other day
Adults/ado­les­cents: 20-40 IU/kg 2-3 times weekly; chil­dren ≤12
years: 25-50 IU/kg 2 times weekly, 3 times weekly, or every
other day
Adults/ado­les­cents: 20-50 IU/kg 3 times weekly or 20-40 IU/kg
every other day; chil­dren <12 years: 25-60 IU/kg 3 times weekly
or 25-50 IU/kg every other day
Adults/ado­les­cents: 30 IU/kg 3 times weekly; chil­dren <12 years:

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/206/1851572/206lim.pdf by guest on 13 December 2021

(moroctocog alfa) 25 IU/kg every other day
 Nuwiq® (simoctocog alfa) Fourth gen­er­a­tion: B-domain deleted Adults/ado­les­cents: 30-40 IU/kg every other day; chil­dren <12
years: 30-50 IU/kg every other day or 3 times weekly
  Hemofil M® Plasma-derived immunoaffinity-puri­fied FVIII Not pro­vided
 Koate ®
Plasma-derived containing FVIII and von Not pro­vided
Willebrand fac­tor
 Humate-P® Plasma-derived containing FVIII and von Not pro­vided
Willebrand fac­tor
 Alphanate® Plasma-derived containing FVIII and von Not pro­vided
Willebrand fac­tor
Factor IX prod­ucts
 BeneFIX® (nonacog alfa)† Recombinant FIX Age ≥16 years: 100 IU/kg weekly
 Ixinity® (trenonacog alfa)‡ Recombinant FIX Adults: 40-70 IU/kg twice weekly
 Rixubis® (nonacog gamma) Recombinant FIX Age ≥12 years: 40-60 IU/kg twice weekly; chil­dren <12 years:
60-80 IU/kg twice weekly
 Alphanine Plasma derived Not pro­vided
*Recommended doses based on US pack­age insert.
†In the US, BeneFIX® is approved for pro­phy­laxis use in those aged ≥16 years only, whereas in the EU it is approved for pro­phy­laxis use at all­ages.
‡In the US, Ixinity® is approved for pro­phy­laxis use in adults only. It is not approved in the EU.

fac­tor replace­ment ther­apy projected that every 1% rise in trough
FVIII lev­els resulted in a 2% increase in the num­ber of patients who
might achieve zero bleeds in a year.38 Accordingly, most cli­ni­cians
favor targeting a higher trough level (>3%-5% or higher) but are
prohibited from doing so due to the con­straints of SHL CFCs that
require almost daily infu­sions to achieve this level. In addi­tion to
the bur­den of daily dos­ing, this strat­egy is also cost-pro­hib­i­tive.
Bleeding trig­gers
The level and pat­tern of phys­i­cal activ­ity as well as under­ly­ing
mus­cu­lo­skel­e­tal health can act as bleed­ing trig­gers, con­trib­ut­
ing to the bleed­ing phe­no­type. School-aged chil­dren and col­
lege stu­dents often engage in sports and should be encour­aged
to do so, even if some restric­tions or mod­i­fi­ca­tions are nec­es­
sary.39 Several ret­ro­spec­tive stud­ies con­cluded that par­tic­i­pa­tion
in orga­nized sports (includ­ing high-risk sports such as bas­ket­ball
and foot­ball) by chil­dren and ado­les­cents with severe hemo­philia
on pro­phy­laxis fac­tor replace­ment ther­apy was not asso­ci­ated
with increased bleed­ing com­pli­ca­tions.40,41 However, depending
on the activ­ity, a change in the pro­phy­laxis dose and/or sched­ule
or an addi­tional preactivity fac­tor infu­sion may be indi­cated. Even
though it is widely agreed that higher fac­tor lev­els are required
with higher-risk phys­i­cal activ­i­ties, clin­i­cal data on what con­sti­
tutes a pro­tec­tive fac­tor level to safely par­tic­i­pate in high-risk
Prakash Singh Shekhawat
activ­i­ties are lacking. A Delphi con­sen­sus state­ment suggested a
fac­tor level of 3% to 5% for mild phys­i­cal activ­ity and 5% to 15%
for higher-risk phys­i­cal activ­ity,42 whereas a more recent expert
solic­i­ta­tion exer­cise suggested much higher lev­els of up to 40%
to 50% (Table 3).43 Regardless of fac­tor lev­els, addi­tional strat­e­
gies includ­ing proper coaching and super­vi­sion, appro­pri­ate use
of safety equip­ment, and suit­able foot­wear are equally impor­tant
to max­i­mize safety for these patients.44 To ensure safe par­tic­i­pa­
tion, a well-informed dis­cus­sion among the health care pro­vider,
patient, par­ents, and coaches should take place. The National
Hemophilia Foundation’s Playing It Safe guide may be help­ful in
these dis­cus­sions, as it cat­e­go­rizes the level of risk and pro­vi­des
infor­ma­tion on safety mea­sures for >80 sport­ing activ­i­ties.39
Hemophilia patients who have under­ ly­ing joint dam­ age
(tar­get joints) from prior bleeds at a young age usu­ally require
higher fac­ tor trough lev­ els and hence higher fac­ tor use to
pre­vent future bleeds. The age at first joint bleed is an early
indi­ca­tor of the patient’s bleed­ing phe­no­type.2 Patients who
expe­ri­ence their first joint bleed ear­lier in life tend to have
higher annual CFC uti­li­za­tion and to develop more arthrop­a­thy
in later years than patients who have their first joint bleed at a
later age.2 This high­lights the impor­tance of early pro­phy­laxis,
as even a few bleeds prior to starting pro­phy­laxis can con­trib­
ute to long-term joint dam­age.13
Optimal pro­phy­laxis fac­tor replace­ment ther­apy | 209
Table 3.  Estimates of min­i­mum fac­tor lev­els cor­re­spond­ing to National Hemophilia Foundation categories of risks
for phys­i­cal activ­i­ties

Minimum fac­tor lev­els (Mean ± SD)†

Risk level Examples of phys­i­cal activ­i­ties* Without joint mor­bid­ity With joint mor­bid­ity‡
1 – low Aquatics, archery, ellip­ti­cal machine, frisbee, golf, hik­ing, tai chi, 4  ±  2 7  ±  2
snor­kel­ing, sta­tion­ary bike, step­per, swim­ming, walk­ing
1.5 – low-mod­er­ate Baseball, bas­ket­ball, bicy­cling, body-sculpting class, canoe­ing, 7  ±  2 11  ±  2
cheerleading, cir­cuit train­ing, dance, fish­ing, frisbee, hik­ing,
horse­back rid­ing, indoor-cycling class, kay­ak­ing, Pilates, indoor
rock climbing, row­ing, row­ing machine, non­mo­tor­ized scooter,
skate­board­ing, ice skat­ing, inline skat­ing, ski machine, soft­ball,
step­per, strength train­ing, T-ball, tread­mill, yoga, Zumba class
2 – mod­er­ate Baseball, bas­ket­ball, bicy­cling, boot camp work­out class, 11  ±  2 12  ±  2

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/206/1851572/206lim.pdf by guest on 13 December 2021

bowl­ing, canoe­ing, cardio kick­box­ing class, cheerleading,
dance, div­ing, fish­ing, touch foot­ball, gym­nas­tics, CrossFit,
horse­back rid­ing, indoor-cycling class, Jet Ski, jumping rope,
kay­ak­ing, mar­tial arts, Pilates, river rafting, rock climbing,
run­ning/­­jog­ging, motor­ized and non­mo­tor­ized scoot­ers,
scuba div­ing, skate­board­ing, ice skat­ing, cross-coun­try ski­ing,
water ski­ing, soc­cer, soft­ball, surf­ing, ten­nis, track and field,
vol­ley­ball, yoga, Zumba class
2.5 – mod­er­ate-high Baseball, bas­ket­ball, bicy­cling, bounce houses, canoe­ing, 19  ±  2 21  ±  2
cheerleading, dance, div­ing, gym­nas­tics, CrossFit, hockey,
horse­back rid­ing, Jet Ski, kay­ak­ing, mar­tial arts, moun­tain bik­
ing, rac­quet­ball, out­door rock climbing, motor­ized and
non­mo­tor­ized scooter, scuba div­ing, skate­board­ing, ice
skat­ing, down­hill ski­ing, water ski­ing, snow­board­ing, soc­cer,
soft­ball, surf­i ng, track and field, tram­po­line, vol­ley­ball,
water polo
3 – high Bicycling, BMX rac­ing, bounce houses, box­ing, dance, div­ing, 38  ±  2 47  ±  2
tackle foot­ball, gym­nas­tics, CrossFit, hockey, Jet Ski, lacrosse,
mar­tial arts, dirt bikes, powerlifting, out­door rock climbing,
rodeo, rugby, snow­mo­bil­ing, soc­cer, tram­po­line, wres­tling
* Examples of phys­i­cal activ­i­ties based on National Hemophilia Foundation categories of risk. Depending on the inten­sity of the phys­i­cal activ­ity,
some activ­i­ties are in mul­ti­ple risk-level categories.39
†Minimum fac­tor lev­els for par­tic­i­pa­tion in phys­i­cal activ­i­ties based on an expert elic­i­ta­tion exer­cise.43
‡Joint mor­bid­ity was defined as hav­ing one or more “dam­aged joints”—ie, any joint per­ma­nently dam­aged as a result of the patient’s bleed­ing dis­or­
der or asso­ci­ated with chronic joint pain and/or lim­ited range of move­ment due to com­pro­mised joint integ­rity.43

CLINICAL CASE (Con­t in­u ed)
The deci­sion was made to start the patient on pro­phy­laxis fac­
tor replace­ ment ther­apy, given his his­ tory of 3 spon­ ta­
ne­ous
joint bleeds. He had received recom­bi­nant SHL FVIII CFC in the
past for treat­ment of his acute joint bleeds and opted to con­
tinue with the same prod­uct. The prod­uct was on his insur­ance
com­pany’s pre­ferred drug list and was on for­mu­lary at the l­ocal
hos­pi­tal. He started at a dose of 25 IU/kg 3 times per week,
to be infused prior to his bas­ket­ball prac­tices on Tues­day and
Thurs­day even­ings and Sat­ur­day morn­ing, to reduce the risk of
bleed­ing from bas­ket­ball injury and to reduce spon­ta­ne­ous joint
bleeds at all­times. Occasionally, when he had addi­tional bas­ket­
ball matches on Sat­ur­day even­ing, he would infuse 40 IU/kg on
Sat­ur­day morn­ing instead. In addi­tion, strat­e­gies to min­i­mize his
injury risk by using eye pro­tec­tion, elbow pads and kneepads,
mouth guards, ath­letic sup­port­ers, and proper foot­wear were
discussed with the patient. He learned how to self-infuse, and his
bas­ket­ball coach also received infu­sion train­ing.
Over the next 3 years, he con­tin­ued on the same reg­i­men
and reported no injury-related bleeds from playing bas­ket­ball.
He noted break­through joint bleeds involv­ing the right knee
occur­ring about twice a year, which he treated using his pro­
phy­laxis 25 IU/kg dose each time. He then started col­lege and
returned to playing bas­ket­ball casu­ally. Because of this and a
part-time job to sup­port his tuition, his adher­ence to the pro­
phy­laxis regimen dropped. He missed a dose about once every
2 weeks due to com­pet­ing demands on his time. He began to
expe­ri­ence more break­through joint bleeds, par­tic­u­larly in his
right knee, about once every month. He returned to the clinic
to dis­cuss his options.
EHL CFCs
The high treat­ment bur­den asso­ci­ated with pro­phy­laxis fac­tor
replace­ment ther­apy, given the fre­quency of infu­sions using
SHL CFC, often leads to a less than opti­mal degree of adher­
ence. In the last decade, EHL prod­ucts were engineered to re-
duce infu­sion fre­quency and to main­tain a higher fac­tor trough
­level for bet­ter pro­tec­tion against spon­ta­ne­ous bleed­ing. This
was achieved using tech­ niques includ­ ing fusion with either
albu­min or the mono­meric Fc frag­ment of immu­no­glob­u­lin G1
(IgG1) or con­ju­ga­tion with poly­eth­yl­ene gly­col (PEG). In the first

210  |  Hematology 2021  |  ASH Education Program

Table 4.  Characteristics of EHL CFCs

Brand (non­pro­pri­etary name)
Factor VIII prod­ucts
  Eloctate®/Elocta®
(efmoroctocog alfa)
  Adynovate®/Adynovi®
(rurioctocog alfa pegol)
  Jivi® (damoctocog alfa pegol)
Manufacturer-recommended
Licensed for starting dose for Recommended assay for
Technology pro­phy­laxis use pro­phy­laxis use* mon­i­tor­ing
B-domain deleted FVIII fused US & EU: all­ages Adults: 50 IU/kg every 4 days; One stage or chro­mo­genic
to Fc por­tion of IgG1 chil­dren <6 years: 50 IU/kg
twice weekly
Full-length recom­bi­nant FVIII US: all­; EU: ≥12 years Adolescents/adults: 40-50 One stage or chro­mo­genic
with 20 kDa PEG IU/kg 2 times weekly;
chil­dren <12 years: 55 IU/kg
2 times weekly
B-domain deleted FVIII with US & EU: ≥12 years 30–40 IU/kg twice weekly Chromogenic or one stage

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/206/1851572/206lim.pdf by guest on 13 December 2021

  Esperoct® (turoctocog alfa
pegol)
  Afstyla® (lonoctocog alfa)
Factor IX prod­ucts
  Alprolix® (efrenonacog alfa)
  Idelvion® (albutrepenonacog
alfa)
  Rebinyn®/Refixia® (nonacog
beta pegol)
*Recommended doses based on US pack­age insert except for Refixia®, which is licensed for pro­phy­laxis use in the EU only.
†Assay for mon­i­tor­ing based on World Federation of Hemophilia rec­om­men­da­tions.9 Further details on val­i­dated reagents are avail­­able at Gray et al
and Kitchen et al.51,52
60 kDa PEG with val­i­dated reagents†
B-domain trun­cated FVIII US: all­; EU: ≥12 years Adolescents/adults: 50 Chromogenic or one stage
with 40 kDa PEG IU/kg every 4 days; with val­i­dated reagents†
chil­dren <12 years:
65 IU/kg twice weekly
Single-chain recom­bi­nant US & EU: all­ages Adults: 20-50 IU/kg 2-3 Chromogenic assay
FVIII times weekly; chil­dren <12
years: 30-50 IU/kg
2-3 times weekly
Recombinant FIX fused to Fc US & EU: all­ages Adolescents/adults: 50 IU/kg Chromogenic or one stage
por­tion of IgG1 weekly or 100 IU/kg every with val­i­dated reagents†
10 days; chil­dren <12 years:
60 IU/kg weekly
Recombinant FIX fused to US & EU: all­ages Adolescents/adults: 25-40 One stage with val­i­dated
albu­min IU/kg weekly; chil­dren <12 reagents†
years: 40-55 IU/kg weekly
Recombinant FIX with 40 kDa EU: ≥12 years 40  IU/kg weekly Chromogenic or one stage
PEG with val­i­dated reagents†

tech­nique, the neo­na­tal Fc recep­tor inhib­its lyso­somal deg­ra­
da­tion of fusion pro­teins and recy­cles them back into the cir­cu­
la­tion,45 whereas PEGylation delays the deg­ra­da­tion and renal
elim­i­na­tion of its attached clot­ting fac­tor.46 A novel strat­egy to
increase the sta­bil­ity and affin­ity of FVIII to von Willebrand fac­
tor (VWF) to reduce the risk of inhib­i­tor devel­op­ment led to
the devel­op­ment of sin­gle-chain forms of FVIII.47 This strat­egy
resulted in the added ben­e­fit of a favor­able half-life, which al-
lows for dos­ing twice per week, sim­i­lar to other EHL FVIII prod­
ucts.48 Although this strat­egy was not intended to extend the
half-life, it is included as an EHL prod­uct for the sake of com­
plete­ness. All cur­rently avail­­able EHL CFCs have been shown to
be effi­ca­cious in the pre­ven­tion and treat­ment of bleeds with
no evi­dence of any clin­i­cal safety issues,9,49 albeit the­o­ret­i­cal
con­ cerns remain regard­ ing life­
long use of PEGylated CFC.50
This has led to vary­ing reg­u­la­tory approval for some PEGylated
prod­ucts for pro­phy­laxis use in the pedi­at­ric pop­u­la­tion (Ta-
ble 4). When selecting an EHL CFC, the same prac­ti­cal mea­
sures apply—local avail­abil­ity, payer pref­er­ence, and cost. For
EHL prod­ucts, an addi­tional deci­sion point for cli­ni­cians is the
abil­ity or avail­abil­ity of local assays to mon­i­tor FVIII/FIX lev­els
accu­rately. Certain EHL CFCs require chro­mo­genic fac­tor as-
says or one-stage FVIII/FIX assays that have been val­i­dated for
Prakash Singh Shekhawat
use (Table 4).51,52 Hence, the lack of appro­pri­ate FVIII/FIX assays
for mon­i­tor­ing may also influ­ence prod­uct selec­tion.
The avail­abil­ity of EHL CFCs has decreased the treat­ment bur­
den, espe­cially for patients with HB, lead­ing to high adher­ence
rates for pro­phy­laxis.53 The prolonged half-life of EHL CFCs trans­
lates to dos­ing twice per week or every 4 days for FVIII CFC and
once every 7 to 14 days for FIX CFC. The dis­crep­ancy in dos­ing
inter­vals is due to the depen­dence of FVIII on the half-life of its
chap­er­one VWF; thus, the prolonged half-life is mod­est at 1.5 to
1.7 times the half-life of SHL FVIII CFC. This lim­i­ta­tion has been
over­come by a new class of FVIII that builds on the Fc fusion
tech­nol­ogy by adding a region of VWF and XTEN® poly­pep­tides,
phys­i­cally decoupling it from endog­e­nous VWF.54 Early-phase
clin­i­cal stud­ies dem­on­strated a 3- to 4-fold increase in half-life
observed, pos­si­bly reduc­ing the dos­ing fre­quency of FVIII for
pro­phy­laxis to once a week.55 Phase 3 clin­i­cal tri­als are ongo­ing,
with results expected in 2022 (NCT04161495).
In selected HA patients who require higher trough lev­ els,
a switch to EHL CFCs admin­is­tered at the reg­u­lar dose inter­
val (ie, every other day) used for SHL CFCs may be appro­pri­ate.
The pre­sumed ben­e­fit predicted from targeting higher trough
lev­els was con­firmed in the recently published PROPEL study
that ran­dom­ized 115 severe HA patients on pro­phy­laxis fac­tor
Optimal pro­phy­laxis fac­tor replace­ment ther­apy | 211
Table 5.  A prac­ti­cal guide for adopting PK assess­ment in the clinic

Question
How should I sched­ule the clinic visit?
What are the ideal mea­sure­ment time points?*
What pop­u­la­tion PK soft­ware is avail­­able?
What infor­ma­tion is required?*
What do I do with the results?
Answer
Time/day of infu­sion and clinic visit should be coor­di­nated to ensure fac­tor lev­els can be drawn at
required time points to min­i­mize mul­ti­ple patient vis­its and incon­ve­nience
For SHL FVIII – predose, 2-4 hours, 24 ± 4 hours, 48 ± 6 hours; for SHL FIX – predose, any time on day
2 and 3; for EHL – as above and add a time point at 60-84 hours for FVIII and at 2-14 days for FIX
A glob­ally acces­si­ble online tool, such as WAPPS-Hemo, or prod­uct-spe­cific tools by the prod­uct
man­u­fac­tur­ers
Age, base­line fac­tor level, prod­uct name, total dose admin­is­tered, body weight, height, infu­sion
day, infu­sion time, postinfusion fac­tor lev­els, and tim­ing of sam­ples
Share and explain results with patients/care­giv­ers; tai­lor treat­ment reg­i­men based on results,
if appro­pri­ate; inte­grate results into patient’s med­i­cal records

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/206/1851572/206lim.pdf by guest on 13 December 2021

*Ideal time points and infor­ma­tion required are based on WAPPS-Hemo (www​­.wapps​­-hemo​­.org). Please see prod­uct man­uf­ac­tur­ers’ man­ual if using
prod­uct-spe­cific tools. Adapted with permission from Hermans C, Dolan G. Therapeutic Advances in Hematology (Volume 11).66 © The Author(s),
2020; Reprinted by Permission of SAGE Publications.

replace­ment ther­apy using rurioctocog alfa pegol (an EHL prod­
uct with a half-life of 14 to 16 hours) to 2 FVIII trough lev­els, 1%
to 3% vs 8% to 12%.56 As expected, dur­ing the 6-month study
period 85% of patients in the higher-trough arm achieved zero
spon­ta­ne­ous joint bleeds vs 65% in the lower-trough arm. How-
ever, achiev­ing the higher tar­get trough lev­els required a higher
bur­den of ther­apy, with 72.4% need­ing infu­sion every 24 to 48
hours vs 19.3% in the lower-trough arm. Notably, 60% of patients
in the ­lower-trough arm expe­ri­enced zero spon­ta­ne­ous bleeds,
whereas 24% of those in the higher-trough arm con­tin­ued to
have spon­ta­ne­ous bleeds. This sup­ports 2 key points: (1) bleed­
ing events decrease as tar­get trough fac­tor level increases, and
(2) there is a need for per­son­al­ized treat­ment. The lat­ter indi­
rectly answers the ques­tion that the appro­pri­ate tar­get trough
level is that at which the patient expe­ri­ences zero bleeds while
pre­serv­ing an active (or sed­en­tary) life­style.
Individual PK assess­ment
For patients with HA, there is a wide var­i­a­tion of interpatient PK
han­dling of infused fac­tor depen­dent on age, body mass, blood
group, and VWF lev­els.57-60 PK stud­ies on interpatient han­dling
of FIX CFC have been fewer, but it is thought to have sim­i­lar
interpatient var­i­abil­ity.61,62 Hence, to opti­mize pro­phy­laxis and
fac­tor uti­li­za­tion, the WFH cur­rently rec­om­mends indi­vid­u­al­ized
PK assess­ment.9
Traditionally, obtaining PK eval­u­a­tion was a huge incon­ve­
nience due to the need for a wash­out period (abstaining for 72
hours from FVIII CFC use and 96-120 hours for FIX CFC use, thus
intro­duc­ing eth­i­cal con­cerns for increased bleed­ing risk) and
fre­quent fac­tor mea­sure­ments for 48 to 72 hours after infu­sion
of a pro­phy­laxis dose. Nowadays, the avail­abil­ity of ­pop­u­la­tion-​
based PK mod­els, such as WAPPS-Hemo (www​­.wapps​­-hemo​
­.org), enables a Bayes­ian esti­ma­tion of indi­vid­ual PK from only 2
or 3 fac­tor mea­sure­ments with­out a wash­out period and allows
for the increased use of PK mon­i­tor­ing in rou­tine clin­i­cal prac­
tice.63,64 A prac­ti­cal guide for adopting PK assess­ ment in the
clinic is shown in Table 5.
A typ­i­cal PK pro­file con­tains many PK param­e­ters, includ­ing
area under the curve, in vivo recov­ery, half-life, and clear­ance.
Detailed descrip­tions of these PK param­e­ters are described in
recently published review arti­cles by Delavenne and Dargaud,
Hermans and Dolan, and Iorio.65-67 Practically, the most impor­
tant esti­ma­tes in the PK pro­file are the time to a targeted (or
trough) fac­tor level or the fac­tor level at a spe­cific time after the
infu­sion.64,65,67 One can then sim­u­late the effect of using dif­fer­ent
doses and/or infu­sion fre­quen­cies to achieve the desired fac­tor
level or to deter­mine the appro­pri­ate­ness of performing a phys­i­
cal activ­ity given the fac­tor level at a spe­cific time. In my expe­ri­
ence, the abil­ity to deter­mine the lat­ter has pro­vided sig­nif­i­cant
peace of mind to hemo­philia patients with active life­styles.
CLINICAL CASE (Con­tin­ued)
We discussed switching to the EHL ver­sion of the patient’s
cur­rent SHL CFC. He declined to undergo a PK assess­ment on
his cur­rent SHL reg­i­men but was agree­­able to doing so after
switching. He was started on an EHL CFC at 50 IU/kg once
every 4 days. After 2 months on the new reg­im ­ en, he came to
the clinic for a PK eval­u­a­tion that showed a half-life of 16.25
hours and an esti­mated trough level of 3.3% prior to the next
dose. One year later, he reports zero spon­ta­ne­ous joint bleeds.
Monitoring real-world hemo­static effec­tive­ness
Even though all­avail­­able CFCs have been shown to be effi­ca­
cious in clin­i­cal tri­als, con­tin­u­ous mon­i­tor­ing of hemo­static
out­comes in clin­i­cal prac­tice is recommended. Recently, it has
been reported that a small sub­set of HB patients on EHL FIX
CFCs for pro­phy­laxis expe­ri­enced unex­pected spon­ta­ne­ous
break­through bleed­ing despite ade­quate FIX lev­els.68,69 These
patients had to switch back to SHL CFCs, switch to a dif­fer­ent
EHL CFC, or reduce the dos­ing inter­val from 14 days to 7 to 10
days to main­tain ade­quate hemo­static con­trol. The cause of this
­unex­pected hemo­static out­come remains unclear, and fur­ther
research is warranted.68,69 Yet this high­lights the need for con­tin­
u­ous assess­ment of real-world hemo­static out­comes out­side of
clin­i­cal tri­als when newly approved treat­ment options for hemo­
philia are intro­duced in rou­tine clin­i­cal prac­tice.
Conclusion
In the era of pre­ci­sion med­i­cine, it is clear that pro­phy­laxis fac­tor
replace­ment ther­apy should be indi­vid­u­al­ized for opti­mal fac­
tor uti­li­za­tion and that “one reg­i­men does not fit all­.” Different

212  |  Hematology 2021  |  ASH Education Program

aspects of the treat­ ment, includ­ing prod­ uct type, dose, and
inter­val, as well as targeted trough lev­els and bleed­ing trig­
gers, are involved in tai­lor­ing the reg­i­men to the clin­i­cal needs
of the patient. One aspect not explic­itly discussed is that effec­
tive pro­phy­laxis is an ongo­ing col­lab­o­ra­tive effort that relies on
shared deci­sion-mak­ing between the patient and the cli­ni­cian.
All of the afore­men­tioned con­sid­er­ations are irrel­e­vant if the pa-
tient’s pref­er­ences and val­ues are not taken into account. As the
com­plex­ity of treat­ment options increases with the avail­abil­ity
of non­fac­tor ther­a­pies and gene ther­apy, it is crit­i­cal that both
patients and cli­ni­cians are actively involved in this col­lab­o­ra­tive
pro­cess to opti­mize treat­ment and over­all patient out­comes.
Acknowledgment
Ming Y. Lim received a 2018 Mentored Research Award from the
Hemostasis and Thrombosis Research Society, which was sup-
ported by an edu­ca­tional grant from Bioverativ, a Sanofi com­pany.
Conflict-of-inter­est dis­clo­sure
Ming Y. Lim: hon­o­rar­ium for par­tic­i­pa­tion in advi­sory boards: Sa-
nofi Genzyme, Argenx, Dova Pharmaceuticals, Hema Biologics;
hon­o­rar­ium and travel expenses for edu­ca­tional par­tic­i­pa­tion:
Hemostasis and Thrombosis Research Society Trainee Workshop
supported by Novo Nordisk.
Off-label drug use
Ming Y. Lim: nothing to disclose.
Correspondence
Ming Y. Lim, Division of Hematology and Hematologic Malignan-
cies, Department of Internal Medicine, University of Utah, 2000
Circle of Hope, Rm 4126, Salt Lake City, UT 84112; e-mail: ming​
­.lim@hsc​­.utah​­.edu.
References
1. Blanchette VS, Key NS, Ljung LR, et al; Subcommittee on Factor VIII, Factor
IX and Rare Coagulation Disorders of the Scientific and Standardization
Committee of the International Society on Thrombosis and Hemostasis.
Definitions in hemo­ philia: com­mu­ni­
ca­
tion from the SSC of the ISTH. J
Thromb Haemost. 2014;12(11):1935-1939.
2. van Dijk K, Fischer K, van der Bom JG, Grobbee DE, van den Berg HM. Vari-
ability in clin­i­cal phe­no­type of severe haemophilia: the role of the first joint
bleed. Haemophilia. 2005;11(5):438-443.
3. Castaman G, Matino D. Hemophilia A and B: molec­u­lar and clin­i­cal sim­i­lar­i­
ties and dif­fer­ences. Haematologica. 2019;104(9):1702-1709.
4. Soucie JM, Monahan PE, Kulkarni R, Konkle BA, Mazepa MA; US Hemophilia
Treatment Center Network. The fre­quency of joint hem­or­rhages and pro­
ce­dures in nonsevere hemo­philia A vs B. Blood Adv. 2018;2(16):2136-2144.
5. Scott MJ, Xiang H, Hart DP, et  al. Treatment reg­i­mens and out­comes in
severe and mod­er­ate haemophilia A in the UK: the THUNDER study. Hae-
mophilia. 2019;25(2):205-212.
6. Santagostino E, Mancuso ME, Tripodi A, et al. Severe hemo­philia with mild
bleed­ing phe­no­type: molec­u­lar char­ac­ter­iza­tion and global coag­u­la­tion
pro­file. J Thromb Haemost. 2010;8(4):737-743.
7. Bos MH, Meijerman DW, van der Zwaan C, Mertens K. Does acti­vated pro­
tein C-resis­tant fac­tor V con­trib­ute to throm­bin gen­er­a­tion in hemo­philic
plasma? J Thromb Haemost. 2005;3(3):522-530.
8. Zhou JY, Barnes RFW, Foster G, Iorio A, Cramer TJ, von Drygalski A. Joint
bleed­ing ten­den­cies in adult patients with hemo­philia: it’s not all­phar­ma­
co­ki­net­ics. Clin Appl Thromb Hemost. 2019;25­(Jan­u­ary-Decem­ber):107602
9619862052.
9. Srivastava A, Santagostino E, Dougall A, et al; WFH Guidelines for the Man-
agement of Hemophilia pan­el­ists and coau­thors. WFH Guidelines for the
Management of Hemophilia, 3rd edi­tion. Haemophilia. 2020;26(suppl 6):1–
158.
Prakash Singh Shekhawat
10. Pabinger I, Mamonov V, Windyga J, et al. Results of a ran­dom­ized phase
III/IV trial com­par­ing inter­mit­tent bolus ver­sus con­tin­u­ous infu­sion of anti-
haemophilic fac­ tor (recom­ bi­
nant) in adults with severe or mod­ er­
ately
severe haemophilia A under­go­ing major ortho­pae­dic sur­gery. Haemo-
philia. 2021;27(3):e331-e339.
11. Pan-Petesch B, Nagao A, Karim FA, et al. Efficacy and safety of rIX-FP in
sur­gery: an update from a phase 3b exten­sion study. Thromb Res. 2020;​
193(Sep­tem­ber):139-141.
12. Santagostino E, Lalezari S, Reding MT, et al. Safety and effi­cacy of BAY 94-
9027, an extended-half-life fac­tor VIII, dur­ing minor sur­gi­cal pro­ce­dures in
patients with severe haemophilia A. Haemophilia. 2021;27(4):e559-e562.
13. Manco-Johnson MJ, Abshire TC, Shapiro AD, et al. Prophylaxis ver­sus epi­
sodic treat­ment to pre­vent joint dis­ease in boys with severe hemo­philia. N
Engl J Med. 2007;357(6):535-544.
14. Gringeri A, Lundin B, von Mackensen S, Mantovani L, Mannucci PM; ESPRIT
Study Group. A ran­dom­ized clin­i­cal trial of pro­phy­laxis in chil­dren with
hemo­philia A (the ESPRIT Study). J Thromb Haemost. 2011;9(4):700-710.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/206/1851572/206lim.pdf by guest on 13 December 2021
15. Manco-Johnson MJ, Lundin B, Funk S, et  al. Effect of late pro­ phy­laxis
in hemo­ philia on joint sta­ tus: a ran­ dom­ ized trial. J Thromb Haemost.
2017;15(11):2115-2124.
16. Warren BB, Thornhill D, Stein J, et al. Young adult out­comes of child­hood
pro­phy­laxis for severe hemo­philia A: results of the Joint Outcome Continu-
ation Study. Blood Adv. 2020;4(11):2451-2459.
17. Manco-Johnson MJ, Soucie JM, Gill JC; Joint Outcomes Committee of the
Universal Data Collection, US Hemophilia Treatment Center Network. Pro-
phylaxis usage, bleed­ing rates, and joint out­comes of hemo­philia, 1999 to
2010: a sur­veil­lance pro­ject. Blood. 2017;129(17):2368-2374.
18. Farrugia A. Guide for the Assessment of Clotting Factor Concentrates. 3rd
ed. World Federation of Hemophilia; 2017. Accessed 10 May 2021. https:​­/​­/
www1​­.wfh​­.org​­/publication​­/files​­/pdf​­-1271​­.pdf.
19. World Federation of Hemophilia. World Federation of Hemophilia online
reg­is­try of clot­ting fac­tor con­cen­trates. Accessed 10 May 2021. https:​­/​­/
www1​­.wfh​­.org​­/custom​­/CFC​­/index​­.html.
20. Gouw SC, van der Bom JG, Ljung R, et al; PedNet and RODIN Study Group.
Factor VIII prod­ucts and inhib­i­tor devel­op­ment in severe hemo­philia A. N
Engl J Med. 2013;368(3):231-239.
21. Peyvandi F, Mannucci PM, Garagiola I, et al. A ran­dom­ized trial of fac­tor VIII
and neu­tral­iz­ing antibodies in hemo­philia A. N Engl J Med. 2016;374(21):​
2054-2064.
22. Wight J, Paisley S. The epi­de­mi­­ol­ogy of inhib­i­tors in haemophilia A: a sys­
tem­atic review. Haemophilia. 2003;9(4):418-435.
23. Franchini M, Coppola A, Rocino A, et al; Ital­ian Association of Hemophilia
Centers Working Group. Systematic review of the role of FVIII con­cen­trates
in inhib­ i­
tor devel­ op­ment in pre­ vi­
ously untreated patients with severe
hemo­philia a: a 2013 update. Semin Thromb Hemost. 2013;39(7):752-766.
24. Marcucci M, Mancuso ME, Santagostino E, et al. Type and inten­sity of FVIII
expo­sure on inhib­i­tor devel­op­ment in PUPs with haemophilia A: a patient-
level meta-anal­y­sis. Thromb Haemost. 2015;113(5):958-967.
25. National Hemophilia Foundation. Recommendation on SIPPET (sur­vey of
inhib­i­tors in plasma-prod­uct-exposed tod­dlers): results and rec­om­men­da­
tions for treat­ment prod­ucts for pre­vi­ously untreated patients with hemo­
philia A. Accessed 10 May 2021. https:​­/​­/www​­.hemophilia​­.org​­/healthcare​
­-professionals​­/guidelines​­-on​­-care​­/masac​­-documents​­/masac​­-document-​
­243​­-recommendation​­-on​­-sippet​­-survey​­-of​­-inhibitors​­-in​­-plasma​­-product​
­-exposed​­-toddlers​­-results​­-and​­-recommendations​­-for​­-treatment​­-products​
­-for​­-previously​­-untreated​­-patients​­-with.
26. Euro­pean Medicines Agency. PRAC con­firms its pre­vi­ous con­clu­sion
on risk of inhib­i­tor devel­op­ment with fac­tor VIII med­i­cines. Accessed 31
August 2021. https:​­/​­/www​­.ema​­.europa​­.eu​­/en​­/documents​­/referral​­/factor​
­-viii​­-article​­-31​­-referral​­-prac​­-confirms​­-its​­-previous​­-conclusion​­-risk​­-inhibitor​
­-development_en​­.pdf.
27. Thornburg C, Friedman K, Guerrera M, et al. U.S. cohort study of pre­vi­ously
untreated patients with con­gen­i­tal hemo­philia (ATHN 8: PUPs study): asso­
ci­a­tion between fam­ily his­tory and age of diag­no­sis. Blood. 2020;136(suppl
1):39-40.
28. White GC, DiMichele D, Mertens K, et  al. Utilization of pre­vi­ously treated
patients (PTPs), non­ in­ fected patients (NIPs), and pre­ vi­
ously untreated
patients (PUPs) in the eval­u­a­tion of new fac­tor VIII and fac­tor IX con­cen­trates:
rec­om­men­da­tion of the Scientific Subcommittee on Factor VIII and Factor IX
of the Scientific and Standardization Committee of the International Society
on Thrombosis and Haemostasis. Thromb Haemost. 1999;81(3):462.
Optimal pro­phy­laxis fac­tor replace­ment ther­apy | 213
29. Dubé E, Bonnefoy A, Merlen C, et al. A pro­spec­tive sur­veil­lance study of
inhib­i­tor devel­op­ment in haemophilia A patients fol­low­ing a pop­u­la­tion
switch to a third-gen­er­a­tion B-domain-deleted recom­bi­nant fac­tor VIII.
Haemophilia. 2018;24(2):236-244.
30. Sidonio RF, Cheng D, Guelcher C, et al. Lack of inhib­i­tor devel­op­ment in
the Amer­i­can throm­bo­sis and hemo­sta­sis net­work (ATHN)-2 fac­tor switch-
ing study: pre­lim­i­nary report of pri­mary out­come. Blood. 2019;134(suppl
1):1114.
31. Collins PW, Björkman S, Fischer K, et  al. Factor VIII require­ ment to
main­tain a tar­get plasma level in the pro­phy­lac­tic treat­ment of severe
hemo­philia A: influ­ences of var­i­ance in phar­ma­co­ki­net­ics and treat­ment
reg­i­mens. J Thromb Haemost. 2010;8(2):269-275.
32. Collins PW, Fischer K, Morfini M, Blanchette VS, Björkman S; International
Prophylaxis Study Group Pharmacokinetics Expert Working Group. Impli-
cations of coag­u­la­tion fac­tor VIII and IX phar­ma­co­ki­net­ics in the pro­phy­lac­
tic treat­ment of haemophilia. Haemophilia. 2011;17(1):2-10.
33. Gouider E, Jouini L, Achour M, et al. Low dose pro­phy­laxis in Tunisian chil­­
dren with haemophilia. Haemophilia. 2017;23(1):77-81.
34. Collins PW, Blanchette VS, Fischer K, et al; rAHF-PFM Study Group. Break-
through bleed­ ing in rela­ tion to predicted fac­ tor VIII lev­
els in patients
receiv­ing pro­phy­lac­tic treat­ment for severe hemo­philia A. J Thromb Hae-
most. 2009;7(3):413-420.
35. Ahnström J, Berntorp E, Lindvall K, Björkman S. A 6-year fol­low-up of dos­
ing, coag­u­la­tion fac­tor lev­els and bleed­ings in rela­tion to joint sta­tus in the
pro­phy­lac­tic treat­ment of haemophilia. Haemophilia. 2004;10(6):689-697.
36. den Uijl I, Biesma D, Grobbee D, Fischer K. Outcome in mod­er­ate haemo-
philia. Blood Transfus. 2014;12(suppl 1):s330-s336.
37. den Uijl IE, Fischer K, Van Der Bom JG, Grobbee DE, Rosendaal FR, Plug I.
Analysis of low fre­quency bleed­ing data: the asso­ci­a­tion of joint bleeds
according to base­line FVIII activ­ity lev­els. Haemophilia. 2011;17(1):41-
44.
38. Chowdary P, Fischer K, Collins PW, et  al. Modeling to pre­dict fac­tor VIII
lev­els asso­ci­ated with zero bleeds in patients with severe hemo­philia A
ini­ti­ated on ter­tiary pro­phy­laxis. Thromb Haemost. 2020;120(5):728-736.
39. Anderson A, Forsyth A; National Hemophilia Foundation. Playing It Safe:
Bleeding Disorders, Sports and Exercise. National Hemophilia Founda-
tion; 2017. Accessed 10 May 2021. https:​­/​­/www​­.changinghemophilia​­.com​
­/content​­/dam​­/biopharm​­/changinghemophilia​­/managing​­-hemophilia​­/
lifestyle​­-tips​­/sports​­-and​­-exercise​­/Playing​­-It​­-Safe_0​­.pdf.
40. Ross C, Goldenberg NA, Hund D, Manco-Johnson MJ. Athletic par­tic­i­pa­
tion in severe hemo­philia: bleed­ing and joint out­comes in chil­dren on pro­
phy­­laxis. Pediatrics. 2009;124(5):1267-1272.
41. Versloot O, Timmer MA, de Kleijn P, et al. Sports par­tic­i­pa­tion and sports
injuries in Dutch boys with haemophilia. Scand J Med Sci Sports. 2020;30(7):​
1256-1264.
42. Iorio A, Iserman E, Blanchette V, et al. Target plasma fac­tor lev­els for per­
son­al­ized treat­ment in haemophilia: a Delphi con­sen­sus state­ment. Hae-
mophilia. 2017;23(3):e170-e179.
43. Martin AP, Burke T, Asghar S, Noone D, Pedra G, O’Hara J. Understanding
min­i­mum and ideal fac­tor lev­els for par­tic­i­pa­tion in phys­i­cal activ­i­ties by
peo­ple with haemophilia: an expert elic­i­ta­tion exer­cise. Haemophilia.
2020;26(4):711-717.
44. McGee S, Raffini L, Witmer C. Organized sports par­tic­i­pa­tion and the asso­
ci­a­tion with injury in pae­di­at­ric patients with haemophilia. Haemophilia.
2015;21(4):538-542.
45. Roopenian DC, Akilesh S. FcRn: the neo­na­tal Fc recep­tor comes of age. Nat
Rev Immunol. 2007;7(9):715-725.
46. Ivens IA, Baumann A, McDonald TA, Humphries TJ, Michaels LA, Mathew
P. PEGylated ther­a­peu­tic pro­teins for haemophilia treat­ment: a review for
haemophilia care­giv­ers. Haemophilia. 2013;19(1):11-20.
47. Mahlangu J, Kuliczkowski K, Karim FA, et al; AFFINITY Investigators. Efficacy
and safety of rVIII-sin­gle chain: results of a phase 1/3 mul­ti­cen­ter clin­i­cal
trial in severe hemo­philia A. Blood. 2016;128(5):630-637.
48. Zhang Y, Roberts J, Tortorici M, et  al. Population phar­ma­co­ki­net­ics of
recom­bi­nant coag­u­la­tion fac­tor VIII-sin­gle chain in patients with severe
hemo­philia A. J Thromb Haemost. 2017;15(6):1106-1114.
49. Mahlangu J, Young G, Hermans C, Blanchette V, Berntorp E, Santagostino
E. Defining extended half-life rFVIII—a crit­i­cal review of the evi­dence. Hae-
mophilia. 2018;24(3):348-358.
50. Lubich C, Allacher P, de la Rosa M, et al. The mys­tery of antibodies against
poly­eth­yl­ene gly­col (PEG)—what do we know? Pharm Res. 2016;33(9):​
2239-2249.
214  |  Hematology 2021  |  ASH Education Program
51. Gray E, Kitchen S, Bowyer A, et  al. Laboratory mea­ sure­ment of fac­
tor replace­ ment ther­ a­
pies in the treat­ ment of con­ gen­ i­
tal haemophilia:
a United Kingdom Haemophilia Centre Doctors’ Organisation guide­line.
Haemophilia. 2020;26(1):6-16.
52. Kitchen S, Tiefenbacher S, Gosselin R. Factor activ­ity assays for mon­it­ or­ing
extended half-life FVIII and fac­tor IX replace­ment ther­a­pies. Semin Thromb
Hemost. 2017;43(3):331-337.
53. Mancuso ME, Oldenburg J, Boggio L, et al. High adher­ence to pro­phy­laxis
reg­i­mens in haemophilia B patients receiv­ing rIX-FP: evi­dence from clin­ic ­ al
tri­als and real-world prac­tice. Haemophilia. 2020;26(4):637-642.
54. Seth Chhabra E, Liu T, Kulman J, et al. BIVV001, a new class of fac­tor VIII
replace­ment for hemo­philia A that is inde­pen­dent of von Willebrand fac­tor
in pri­ma­tes and mice. Blood. 2020;135(17):1484-1496.
55. Konkle BA, Shapiro AD, Quon DV, et al. BIVV001 fusion pro­tein as fac­tor VIII
replace­ment ther­apy for hemo­philia A. N Engl J Med. 2020;383(11):1018–
1027.
56. Klamroth R, Windyga J, Radulescu V, et  al. Rurioctocog alfa pegol
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/206/1851572/206lim.pdf by guest on 13 December 2021
PK-guided pro­ phy­laxis in hemo­ philia A: results from the phase 3
PROPEL study. Blood. 2021;137(13):1818-1827.
57. Björkman S, Folkesson A, Jönsson S. Pharmacokinetics and dose require­
ments of fac­tor VIII over the age range 3-74 years: a pop­ul­a­tion anal­y­sis
based on 50 patients with long-term pro­phy­lac­tic treat­ment for haemo-
philia A. Eur J Clin Pharmacol. 2009;65(10):989-998.
58. Fischer K, Pendu R, van Schooten CJ, et al. Models for pre­dic­tion of fac­tor
VIII half-life in severe haemophiliacs: dis­tinct approaches for blood group
O and non-O patients. PLoS One. 2009;4(8):e6745.
59. Henrard S, Speybroeck N, Hermans C. Impact of being under­weight or
over­weight on fac­tor VIII dos­ing in hemo­philia A patients. Haematologica.
2013;98(9):1481-1486.
60. Tang L, Leong L, Sim D, et al. von Willebrand fac­tor con­trib­utes to lon­ger
half-life of PEGylated fac­tor VIII in vivo. Haemophilia. 2013;19(4):539-545.
61. Björkman S, Ahlén V. Population phar­ma­co­ki­net­ics of plasma-derived fac­
tor IX in adult patients with haemophilia B: impli­ca­tions for dos­ing in pro­
phy­laxis. Eur J Clin Pharmacol. 2012;68(6):969-977.
62. Iorio A, Fischer K, Blanchette V, et al; Pharmacokinetic Expert Work-
ing Group of the International Prophylaxis Study Group. Tailoring treat­
ment of haemophilia B: account­ ing for the dis­ tri­
bu­tion and clear­ ance
of stan­dard and extended half-life FIX con­cen­trates. Thromb Haemost.
2017;117(6):1023-1030.
63. WAPPS-Hemo Research Network. Web-acces­si­ble pop­u­la­tion phar­ma­co­
ki­netic ser­vice—hemo­philia. Accessed 14 May 2021. https:​­/​­/www​­.wapps​
­-hemo​­.org.
64. Iorio A, Blanchette V, Blatny J, Collins P, Fischer K, Neufeld E. Estimating
and interpreting the phar­ma­co­ki­netic pro­files of indi­vid­ual patients with
hemo­philia A or B using a pop­u­la­tion phar­ma­co­ki­netic approach: com­mu­
ni­ca­tion from the SSC of the ISTH. J Thromb Haemost. 2017;15(12):2461–
2465.
65. Delavenne X, Dargaud Y. Pharmacokinetics for haemophilia treat­ers: mean­
ing of PK param­e­ters, inter­pre­ta­tion pit­falls, and use in the clinic. Thromb
Res. 2020;192(May):52-60.
66. Hermans C, Dolan G. Pharmacokinetics in rou­ tine haemophilia clin­ i­
cal
prac­tice: ratio­nale and modal­i­ties—a prac­ti­cal review. Ther Adv Hematol.
2020;11(Octo­ber):2040620720966888.
67. Iorio A. Using phar­ma­co­ki­net­ics to indi­vid­u­al­ize hemo­philia ther­apy.
Hematology Am Soc Hematol Educ Program. 2017;2017(1):595-604.
68. Kleiboer B, Nielsen B, Ma AD, Abajas Y, Monroe DM, Key NS. Excessive
break­ through bleed­ ing in haemophilia B patients on fac­ tor IX-albu­
min
fusion pro­tein pro­phy­lac­tic ther­apy: a sin­gle cen­tre case series. Haemo-
philia. 2020;26(1):e23-e25.
69. Malec LM, Croteau SE, Callaghan MU, Sidonio RF Jr. Spontaneous bleed­
ing and poor bleed­ing response with extended half-life fac­tor IX prod­
ucts: a sur­vey of select US haemophilia treat­ment cen­tres. Haemophilia.
2020;26(3):e128-e129.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000310
 HEMOPHILIA UPDATE: OUR CUP RUNNETH OVER

EVIDENCE-BASED MINIREVIEW

For overweight or obese persons with

hemophilia A, should factor VIII dosing be based

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/215/1851813/215machin.pdf by guest on 13 December 2021

on ideal or actual body weight?
Nicoletta Machin1,2 and Ming Y. Lim3
Department of Medicine, Division of Hematology/Oncology, University of Pittsburgh, Pittsburgh, PA; 2Hemophilia Center of Western Pennsylvania,
1

Pittsburgh, PA; and 3Department of Internal Medicine, Division of Hematology and Hematologic Malignancies, University of Utah, Salt Lake City, UT

LEARNING OBJECTIVES
• Review evidence of the impact of body weight on FVIII in vivo recovery
• Review evidence of using ideal body weight for FVIII dosing in overweight and obese persons with hemophilia A

CLINICAL CASE
A 35­year­old man with severe hemophilia A and a body mass
index (BMI) of 38 (height, 180 cm; weight, 123 kg) presents
for his annual comprehensive care visit. He is on prophylaxis
using factor replacement therapy with a recombinant stan­
dard half­life coagulation factor VIII (FVIII) product that is
dosed according to actual body weight (ABW). He reports
zero spontaneous joint bleeds over the past 12 months. He
has an upcoming elective laparoscopic cholecystectomy
for symptomatic cholelithiasis. His hematologist wonders if
his perioperative dosing as well as prophylaxis dosing of
FVIII should be adjusted considering his increased BW.
Introduction
For persons with hemophilia A, factor replacement therapy
with FVIII concentrates is the cornerstone of treatment for
managing and preventing bleeding episodes. FVIII concen­
trate is typically dosed on a per­kilogram basis. The number
of FVIII units needed to achieve a desired circulating FVIII
level is empirically calculated using the following formula:
the person’s physical build does not markedly differ from
the average. Despite this caveat, the IVR of 2 has been
adopted widely for FVIII dose calculations irrespective of
body composition. With over 31% of the US hemophilia
population classified as obese, the appropriateness of an
IVR of 2 as a “one­size­fits­all” approach in this population
is questionable.2
Impact of body composition on FVIII IVR
In persons who are overweight or obese, the rise in BW is
primarily the result of increased adipose tissue, which con­
tains less vascular space than lean mass. The plasma volume
per kilogram of BW decreases as BW increases, resulting in
a lower plasma volume per kilogram of BW.3 Since FVIII is
primarily restricted to the vascular compartment, the same
amount of FVIII concentrate (in units per kilogram) infused
in an obese person likely results in a higher circulating plas­
ma FVIII level compared to a nonobese person.4
Several clinical studies have demonstrated that FVIII IVR
increases with increasing BMI and BW (Table 1).5-8 The IVR
is calculated by

FVIII dose =
(
BW in kg  ×  desired FVIII increase  IU
dL ) IVR =
(
BW in kg  ×   observed FVIII recovery in IU
dL
−preinfusion FVIII in IU
dL  )
2 FVIII dose in IU

This formula is based on a FVIII in vivo recovery (IVR) of
2, which assumes that each unit of FVIII concentrate infused
per kilogram of BW increases the circulating FVIII level by
2 IU/dL.1 The IVR of 2 was first described in 1981 by Ingram
based on data from 19 persons with hemophilia A with an
ABW between 27 and 91 kg.1 He concluded that the dose
calculation using a plasma volume Prakash
of 0.5  dL/kgSingh
appliesShekhawat
if
In addition, a population pharmacokinetic (PK) model­
ing study found that changes in ideal body weight (IBW)
account for the most significant amount of interindivid­
ual FVIII PK variability.9 Given these findings, this minire­
view seeks to compare the evidence for IBW vs ABW for
the dosing of FVIII in overweight and obese persons with
hemophilia A.

IBW vs ABW dosing for FVIII | 215

Table 1. Summary of clin­i­cal stud­ies eval­u­at­ing FVIII IVR strat­i­fied by BMI

Study design Intervention (median IU, Strongest pre­dic­tor

Reference (age, years) n range) BMI groups (kg/m2) (n) IVR (IU dL−1/IU kg−1) of IVR
Henrard et al5 Prospective obser­va­tional 46 A dose of rFVIII (2000, 18.5-24.9 (26) 1.88 BW
(mean, 40.4 ± 12.3) 980-4200) 25.0-29.9 (14) 2.30
≥30.0 (6) 2.70
Henrard et al6 Retrospective pooled 201 A dose of rFVIII (3745, <18.5 (9) 1.72 BMI
anal­y­sis of 8 PK tri­als 1953-8794) 18.5-24.9 (105) 2.03
(median, 26; IQR, 25.0-29.9 (52) 2.18
21-38) >30.0 (35) 2.68
Henrard et al7 Retrospective pool 66 A dose of rFVIII (2778, Normal (43) 1.93 BMI-for-age
anal­y­sis of 6 PK tri­als 1675-5420) Overweight (7) 2.12
(median, 14.5; IQR, Obese (16)b 2.65

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/215/1851813/215machin.pdf by guest on 13 December 2021

12.8-15.6)a
Tiede et al8 Prospective obser­va­tional 35 rFVIII 50 IU/kg by ABW <18.5 (5) 2.2c BMI
(mean, 37.4; range, 18.5-24.9 (7) 2.9c
23.0–57.0) 25.0-29.9 (9) 2.9c
30.0-34.9 (7) 3.2c
≥35 (7) 3.5c
a
Only trial in chil­dren.
b
Based on the BMI-for-age per­cen­tiles (nor­mal, 5th-84th; over­weight, 85th-94th; obese, >95th).
c
End point reported as IVR at 30 min­utes using a geo­met­ric mean.
IQR, interquartile range.

Methods
We conducted a sys­tem­atic search of MEDLINE, Embase, and the
Cochrane Database of Systematic Reviews from 1 Jan­u­ary 1981
to 10 May 2021. Our search included MeSH and the key words
“hemo­philia A,” “phar­ma­co­ki­netic,” and “dose” and yielded 830
abstracts (Figure 1). References of the resulting stud­ies and nar­
ra­tive reviews were screened. There were 594 unique arti­cles
after 236 dupli­cates were removed. Of the unique arti­cles, 181
were review papers, 117 were eval­ua ­ t­ing new fac­tor con­cen­
trates, 87 were case reports and alter­na­tive trial designs, 68 in­
volved non­fac­tor ther­apy, 49 involved inhib­i­tor patients, 38 were
not on hemo­philia A, and 27 had no weight infor­ma­tion. Full-text
reviews by 2 inde­pen­dent review­ers were conducted on the 27
remaining arti­cles, and 3 stud­ies were ulti­mately included: 1 ret­
ro­spec­tive obser­va­tional study and 2 cross­over tri­als.10-12
Results
All three stud­ies used dif­fer­ent out­come mea­sures: peak FVIII
lev­els after a 50 IU/kg dose and clin­i­cal effec­tive­ness10; per­cent­
age of actual FVIII recov­ery lev­els over expected lev­els after a
50 IU/kg dose11; and FVIII IVR (Table 2).12 Graham et al performed
a ret­ro­spec­tive study reporting peak FVIII lev­els for 6 adults
with BMIs ≥30 treated with recom­bi­nant FVIII 50 IU/kg dosed
by IBW.10 These patients had been previously receiving recom­
binant FVIII at 50 IU/kg dosed by ABW for prophylaxis or were
undergoing surgery. With IBW dos­ ing, 5 of the 6 patients
achieved expected mean peak FVIII lev­els of 100%. The patient
with the low­est peak level (78%) had the highest weight (151   kg,
BMI 45.8) and was retreated with IBW plus 25%, resulting in a
peak FVIII level of 107%. During the 3-month pro­phy­laxis and/or
sur­gi­cal period for each patient, there was no increase in spon­ta­
ne­ous or trau­matic bleeds and no unex­pected bleed­ing or trans­
fu­sion require­ment with sur­gery.
Blair et al conducted a cross­over trial of both ABW and IBW
dos­ing in 16 adults and chil­dren (median BMI 33.1; range, 25.6-
41.8) and com­pared actual FVIII recov­ery lev­els with expected-
recov­ery lev­els after a FVIII 50 IU/kg dose.11 Participants used
their per­sonal brand of fac­tor, and both stan­dard and extended
half-life prod­ ucts were included. With ABW dos­ ing, 15 of 16
patients achieved above-expected recov­ery rang­ing from 0.1 to
0.96 above expected (mean, 0.44). In the IBW dos­ing arm, 8 of 16
patients achieved above-expected recov­ery (range, 0.01-0.37).
Of the 8 patients who did not achieve above-expected recov­ery,
6 were within 10% from expected-recov­ery.
The sec­ond cross­over trial, a 3 × 3 × 3 design of ABW vs IBW vs
lean body mass (LBM), was conducted in 19 adults (mean BMI,
29.5) to deter­mine which descrip­tors of BW achieved the tar­
geted IVR of 2 with bet­ter pre­ci­sion.12 The mean IVR postrecom­
binant fac­tor VIII (rFVIII) infu­sion for ABW-, LBM-, and IBW-based
dos­ing was 2.46, 2.22, and 2.29, respec­tively. The pro­por­tion of
par­tic­i­pants with an IVR of 2.00% ± 10% was 31.1%, 44.2%, and
49.0% for ABW-, LBM-, and IBW-based dos­ing, respec­tively.
Discussion
Our sys­tem­atic minireview iden­ti­fied 2 pro­spec­tive cross­over
stud­ies that directly com­pared IBW to ABW dos­ing of FVIII con­
cen­trates and 1 ret­ro­spec­tive study that com­pared IBW dos­ing
to his­tor­i­cal ABW dos­ing in over­weight and obese per­sons with
hemo­philia A.10-12 A con­cern with using IBW is the risk of under­
dosing, resulting in sub­ther­a­peu­tic FVIII lev­els and increased risk
of bleed­ ing. Overall, these 3 stud­ ies dem­ on­strated favor­
able
out­comes using dif­fer­ent PK mea­sures when dosed by IBW and
sup­port the use of IBW to per­form indi­vid­u­al­ized PK anal­y­sis in
over­weight and obese per­sons with hemo­philia A.
Despite favor­able PK mea­sures, a clin­i­cal con­cern is hemo­static
effec­tive­ness when using IBW for dos­ing FVIII con­cen­trates. Of
the stud­ies iden­ti­fied, only the ret­ro­spec­tive study by Graham et
al pro­vided 3 months of data on hemostastic effec­tive­ness for the
IBW dos­ing strat­egy.10 Even though the hemo­static effec­tive­ness
of IBW dos­ing was favor­able in both the perioperative and pro­

216  |  Hematology 2021  |  ASH Education Program

 830 records idenfied through MEDLINE (156),
EMBASE (459) and Cochrane database (215)

594 records screened by abstract a‚er 236

duplicates removed 567 arcles excluded

• Review papers (181)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/215/1851813/215machin.pdf by guest on 13 December 2021

• New factor product study (117)
• Other trial designs/ case reports (87)
• Nonfactor therapy (68)
• Inhibitor paents (49)
• Wrong disease (38)
27 full-text reviewed for eligibility • No weight informaon (27)

24 arcles excluded

• No weight comparison (18)

• Review papers (2)
• Other trial designs (3)
• Alternave dosing schedule (1)

3 arcles included

Figure 1. Study flow diagram.

Table 2. Findings of 3 stud­ies eval­u­at­ing out­comes based on IBW

Mean (range)
Peak FVIII level or Achieved peak FVIII
Study design expected recov­ery level or expected
Reference (age, years) n Intervention ABW (kg) BMI IBW or IVR recov­ery or IVR
Grahamet al 10
Retrospective 6 rFVIII 50 IU/kg 116 (91-151) 36.0 (30.0-45.8) 74.4 (67.0– Peak FVIII level 100% 83% (5 of 6)a
(median, by IBW 84.5) (78-123)
33; range,
25-47)
Blair et al11 Prospective 16 FVIII 50 IU/kg NR Median, 33.1 (25.6-41.8) NR ABW, 140% expected ABW, 94% (15 of 16)b;
cross­over (±20%) by recov­ery (83-196); IBW, 87.5% (14 of 16)b
(median, both ABW IBW, 100% expected
21.5; range, and IBW recov­ery (56-137)
12-53)
Seaman et al12 Prospective 19 A dose of rFVIII 93.0 ± 13.6 29.2 ± 3.5 (Max, 38.3) 73.6 ± 6.3 ABW: IVR, 2.46; ABW, 31.1%c;
cross­over based on IBW: IVR, 2.29; IBW, 49.0%c;
(mean, ABW, IBW, LBM: IVR, 2.22 LBM, 44.2%c
34.6 ± 11.3) and LBM and
an IVR of 2.0
a
Above expected peak FVIII level of 100% or within 10% below expected peak FVIII level.
b
Above expected recov­ery or within 10% below expected recov­ery.
c
Within 10% of targeted IVR of 2.00.
LBW, lean body weight.
Prakash Singh Shekhawat
IBW vs ABW dos­ing for FVIII  |  217
phy­laxis set­ting, this study was lim­ited by the small sam­ple size of
6 adults. Particularly in the perioperative set­ting, the appli­ca­tion
of IBW dos­ing seems appeal­ing to avoid over­treat­ment with the
unnec­es­sary high FVIII peaks asso­ci­ated with ABW dos­ing, which
could lead to com­pli­ca­tions such as throm­bo­sis and inhib­i­tor
devel­op­ment, espe­cially in per­sons with nonsevere hemo­philia
A who have life­long inhib­i­tor risk.13 A case-con­trol study found
that inten­sive FVIII treat­ment was a risk fac­tor for inhib­it­ or devel­
op­ment in per­sons with nonsevere hemo­philia A even after more
than 50 expo­sure days.14 Larger cohort stud­ies with long-term
fol­low-up are crit­i­cal to fully eval­u­ate the effec­tive­ness of dos­ing
FVIII by IBW in both the perioperative and pro­phy­laxis set­ting.
Additionally, given the cost of FVIII con­cen­trates, dos­ing based
on IBW is a cost-sav­ings approach if hemo­static effec­tive­ness is
com­pa­ra­ble to ABW dos­ing. Graham et al cal­cu­lated that IBW
dos­ing resulted in a 48.9% reduc­tion in FVIII usage dur­ing the
3-month period, with an annu­al­ized mean sav­ings of $133,000 per
patient.10 However, with­out robust effec­tive­ness data, it is dif­fi­
cult to judge the cost-effec­tive­ness of IBW dos­ing at this time.
While body com­po­si­tion has a dem­on­strated impact on FVIII
PKs, it should be remem­bered that other fac­tors may play a role
and should also be con­sid­ered.15 Age has been iden­ti­fied as an
inde­pen­dent mod­i­fier of FVIII clear­ance, with increased clear­
ance seen in early child­hood.16 von Willebrand fac­tor lev­els and
the ABO blood group are noted to alter FVIII clear­ance but were
not found to be sig­nif­i­cant deter­mi­na­tes of IVR in the stud­ies
reviewed.5,6,8 Individual PKs also vary between fac­tor concen­
trate products (recom­ bi­
nant, plasma-derived, and extended
half-life), and the stud­ies included in this review only eval­u­ated
recom­ bi­
nant prod­ ucts. Overall, even though numer­ ous indi­
vid­ual fac­tors can account for inter­in­di­vid­ual PK pro­files, body
com­po­si­tions appear to be the stron­gest deter­mi­nate.5,6,8
In sum­mary, although con­clu­sive rec­om­men­da­tions can­not
be made based on the small num­ber of stud­ies and the lack of
stan­dard­i­za­tion of out­comes, we sug­gest the fol­low­ing:
1. For overweight and obese persons with hemophilia A, we
suggest that an individualized PK analysis be performed using
IBW to determine adequate FVIII in vivo recovery (Grade 2B).
2. If there is adequate FVIII in vivo recovery, we suggest using an
IBW dosing strategy in the perioperative setting (Grade 2C).
3. If there is adequate FVIII in vivo recovery, we suggest using
an IBW dosing strategy in the prophylaxis setting (Grade 2C).
We empha­size the need for larger and long-term clin­i­cal
stud­ies to deter­mine the hemo­static effec­tive­ness of using
an IBW dos­ing strat­egy in the perioperative and pro­phy­laxis
set­ting.
Acknowledgment  tology Am Soc Hematol Educ Program. 2017;2017(1):595-604.
Ming Y. Lim received a 2018 Mentored Research Award from the
Hemostasis and Thrombosis Research Society, which was sup­
ported by an edu­ca­tional grant from Bioverativ, a Sanofi com­pany.
Conflict-of-inter­est dis­clo­sure
Ming Y. Lim: advi­sory board: Sanofi Genzyme, Argenx, Dova Phar­
maceuticals, Hema Biologics; hon­o­raria: Hemostasis and Thrombo­
218  |  Hematology 2021  |  ASH Education Program
sis Research Society Trainee Workshop supported by Novo Nordisk.
Nicoletta Machin: research funding: Takeda Pharmaceuticals.
Off-label drug use
Ming Y. Lim: nothing to disclose.
Nicoletta Machin: nothing to disclose.
Correspondence
Ming Y. Lim, Division of Hematology and Hematologic Malignan­
cies, Department of Internal Medicine, University of Utah, 2000
Circle of Hope, Rm 4126, Salt Lake City, UT 84112; e-mail: ming​
­.lim@hsc​­.utah​­.edu.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/215/1851813/215machin.pdf by guest on 13 December 2021
References
1. Ingram GI. Calculating the dose of fac­tor VIII in the man­age­ment of haemo­
philia. Br J Haematol. 1981;48(2):351-354.
2. Wilding J, Zourikian N, Di Minno M, et al. Obesity in the global haemophilia
pop­u­la­tion: prev­a­lence, impli­ca­tions and expert opin­ions for weight man­
age­ment. Obes Rev. 2018;19(11):1569-1584.
3. Feldschuh J, Enson Y. Prediction of the nor­mal blood vol­ume: rela­tion of
blood vol­ume to body hab­i­tus. Circulation. 1977;56(suppl 4, pt 1):605-612.
4. Björkman S, Berntorp E. Pharmacokinetics of coag­u­la­tion fac­tors: clin­i­
cal rel­e­vance for patients with haemophilia. Clin Pharmacokinet. 2001;​
40(11):815-832.
5. Henrard S, Speybroeck N, Hermans C. Body weight and fat mass index as
strong pre­dic­tors of fac­tor VIII in vivo recov­ery in adults with hemo­philia A.
J Thromb Haemost. 2011;9(9):1784-1790.
6. Henrard S, Speybroeck N, Hermans C. Impact of being under­weight or
over­weight on fac­tor VIII dos­ing in hemo­philia A patients. Haematologica.
2013;98(9):1481-1486.
7. Henrard S, Hermans C. Impact of being over­weight on fac­tor VIII dos­ing in
chil­dren with haemophilia A. Haemophilia. 2016;22(3):361-367.
8. Tiede A, Cid AR, Goldmann G, et al. Body mass index best pre­dicts recov­
ery of recom­bi­nant fac­tor VIII in under­weight to obese patients with severe
haemophilia A. Thromb Haemost. 2020;120(2):277-288.
9. van Moort I, Preijers T, Hazendonk HCAM, et al; OPTI-CLOT Study Group.
Dosing of fac­tor VIII con­cen­trate by ideal body weight is more accu­rate
in over­weight and obese haemophilia A patients. Br J Clin Pharmacol.
2021;87(6):2602-2613.
10. Graham A, Jaworski K. Pharmacokinetic anal­y­sis of anti-hemo­philic fac­tor in
the obese patient. Haemophilia. 2014;20(2):226-229.
11. Blair A, Felgenhauer J, Recht M, Kruse-Jarres R. Comparison of ideal ver­sus
actual body weight fac­tor dos­ing in hemo­philia A [ab­stract]. Haemophilia.
25(S2): 3-77.
12. Seaman CD, Yabes JG, Lalama CM, Ragni MV. Factor VIII con­cen­trate dos­ing
with lean body mass, ideal body weight and total body weight in over­
weight and obe­sity: a ran­dom­ized, con­trolled, open-label, 3 × 3 cross­over
trial. Haemophilia. 2021;27(3):351-357.
13. Eckhardt CL, van Velzen AS, Peters M, et al; INSIGHT Study Group. Fac­
tor VIII gene (F8) muta­tion and risk of inhib­i­tor devel­op­ment in nonsevere
hemo­philia A. Blood. 2013;122(11):1954-1962.
14. Abdi A, Eckhardt CL, van Velzen AS, et al; INSIGHT Study Group. Treat­
ment-related risk fac­tors for inhib­i­tor devel­op­ment in non-severe hemo­
philia A after 50 cumu­la­tive expo­sure days: a case-con­trol study. J Thromb
Haemost. 2021;19(9):2171-2181.
15. Iorio A. Using phar­ma­co­ki­net­ics to indi­vid­u­al­ize hemo­philia ther­apy. Hema-
16. Björkman S. Comparative phar­ma­co­ki­net­ics of fac­tor VIII and recom­bi­nant
fac­tor IX: for which coag­u­la­tion fac­tors should half-life change with age?
Haemophilia. 2013;19(6):882-886.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000317
 HEMOPHILIA UPDATE: OUR CUP RUNNETH OVER

Factor-mimetic and rebalancing therapies

in hemophilia A and B: the end of factor
concentrates?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/219/1851769/219ellsworth.pdf by guest on 13 December 2021

Patrick Ellsworth and Alice Ma
Department of Medicine, Division of Hematology, University of North Carolina, Chapel Hill, NC

Hemophilia A (HA) and B are inherited bleeding disorders caused by a deficiency of factor VIII or factor IX, respectively.
The current standard of care is the administration of recombinant or purified factor. However, this treatment strategy
still results in a high economic and personal burden to patients, which is further exacerbated by the development of
inhibitors—alloantibodies to factor. The treatment landscape is changing, with nonfactor therapeutics playing an increas-
ing role in what we consider to be the standard of care. Emicizumab, a bispecific antibody that mimics the function of
factor VIIIa, is the first such nonfactor therapy to gain US Food and Drug Administration approval and is rapidly changing
the paradigm for HA treatment. Other therapies on the horizon seek to target anticoagulant proteins in the coagulation
cascade, thus “rebalancing” a hemorrhagic tendency by introducing a thrombotic tendency. This intricate hemostatic
balancing act promises great things for patients in need of more treatment options, but are these other therapies going
to replace factor therapy? In light of the many challenges facing these therapies, should they be viewed as a replacement
of our current standard of care? This review discusses the background, rationale, and potential of nonfactor therapies as
well as the anticipated pitfalls and limitations. This is done in the context of a review of our current understanding of the
many aspects of the coagulation system.

LEARNING OBJECTIVES
• Describe the basis of emicizumab action
• Describe potential targets used in rebalancing therapies for hemophilia treatment

Introduction in hemophilia in subsequent years.3 However, even with

Congenital hemophilia A (HA) and B (HB) are inherited
bleeding disorders caused by a deficiency of factor VIII
(FVIII) and factor IX (FIX), respectively. Many barriers have
been overcome leading up to our current standard of care
in the form of modern factor therapies.1
The administration of factor concentrates (derived
from human plasma) to replace the missing factor is the
most straightforward treatment approach and started in
the 1970s. However, many hemophilia patients fell victim
to the HIV pandemic, when contaminated factor products
infected many patients. With the sequencing and cloning
of the FVIII and FIX genes in the 1980s and technological
advances to inactivate and purify concentrates, modern
recombinant and plasma­derived factor products ushered
in a new era of treatment.1,2 Recombinant and plasma­
derived factor products have ascended as the major hemo­
philia treatment and have remained the standard of care
Prakash Singh Shekhawat
infectious concerns all but eradicated in new patients,
treatment with factor products poses a risk of developing
inhibitors, which are neutralizing alloantibodies to exog­
enous FVIII or FIX proteins recognized as foreign by the
body’s immune system.
Inhibitor development remains a significant complica­
tion in the treatment of patients with HA and HB and leads
to bleeding despite factor therapy. The phenomenon is
more common in severe HA than in nonsevere HA, with an
incidence of 25% to 30%.4 In HB, inhibitor incidence is 3% to
5% in general but is higher in populations enriched for null
mutations.5 Even with attempted immune tolerance induc­
tion and immunosuppression, inhibitors recur in up to 30%
of HA and 20% of HB patients.5-7 First­line bleed treatment in
inhibitor patients is generally with bypassing agents (BPA)
such as recombinant activated factor VII (rFVIIa) or activated
prothrombin complex concentrates (aPCC), which have
Treating hemophilia without factor? | 219
sim­i­lar effi­cacy and side effect pro­files in ret­ro­spec­tive ana­ly­ses.
Although gen­er­ally effec­tive, BPA can­not be mon­i­tored by stan­
dard lab assays and have a reported fail­ure rate rang­ing from 7%
to 11.6% and throm­bo­sis rates between 4% and 6.5%.8 As such,
non­fac­tor alter­na­tives to cir­cum­vent these com­pli­ca­tions have
been of inter­est. These ther­a­pies are part of 2 broad categories:
fac­tor mimet­ics and rebalancing ther­a­pies. This review discusses
the indi­vid­ual drugs either cur­rently in use, in trial, or in pre­clin­
i­cal inves­ti­ga­tion. A dis­cus­sion of gene ther­apy, desmopressin,
and antifibrinolytics is omit­ted in favor of a focus on the basic
sci­ence, ratio­nale, and con­sid­er­ations of these emerg­ing non­
fac­tor ther­a­pies.
CLINICAL CASES
Case 1: A 10-year-old boy with severe HB devel­oped an inhib­i­tor
at age 4, suc­cess­fully com­plet­ing immune tol­er­ance induc­tion
at that time. He expe­ri­enced bleed­ing while on pro­phy­lac­tic
recom­bi­nant FIX infu­sions years later and was found to have a
recur­rence of his inhib­i­tor. He had a cen­tral line inserted for BPA
use while restarting immu­no­sup­pres­sion. He still has evi­dence of
an inhib­i­tor, and his par­ents are ask­ing about alter­na­tive ther­a­
pies, espe­cially those that could allow him to live with­out cen­tral
venous access and such fre­quent infu­sion cen­ter vis­its.
Case 2: A 26-year-old man has severe HA and has not been on fac­
tor pro­phy­laxis since age 12. He is afraid of bleed­ing and avoids
var­i­ous activ­i­ties out of cau­tion but still expe­ri­ences spon­ta­ne­ous
bleed­ing, pre­dom­i­nantly in his ankles. He never learned to infuse
fac­tor and states that he is afraid of needles. Additionally, he has
no insur­ance cov­er­age. He is ask­ing if there are now any options
other than infus­ing intra­ve­nous fac­tor to treat his hemo­philia.
Hemostatic tar­gets of non­fac­tor ther­a­pies
A review of the clot­ting cas­cade and its role in the hemo­philia
phe­no­type is nec­es­sary to a dis­cus­sion of mimetic and rebalanc­
ing ther­a­pies.
Hemostasis is ini­ti­ated by the extrin­sic path­way of the coag­
u­la­tion cas­cade and then ampli­fied by the con­tact, or intrin­sic,
path­way. These inter­ac­tions gen­er­ate throm­bin, which then
cleaves to fibrin­o­gen to form a sta­ble fibrin clot (Figure 1).
Given the pri­macy of the extrin­sic, or ini­ti­a­tion, path­way in
phys­i­o­logic hemo­sta­sis, why should a defi­ciency of FVIII or FIX
lead to such severe bleed­ing? Because once fibrin depo­si­tion
begins and throm­bin is formed, throm­bin also acts “upstream”
to acti­vate addi­tional FVIII, FXI, and FV, lead­ing to more throm­bin
gen­er­a­tion via the intrin­sic path­way (also called the ampli­fi­ca­
tion path­way) in what is referred to as the throm­bin burst. With­
out FVIII or FIX, this phe­nom­e­non does not occur, and throm­bin
gen­er­at­ ion remains too mea­ger to form a sta­ble fibrin clot. The
objec­tive of either a fac­tor mimetic or a rebalancing ther­apy is
to restore throm­bin gen­er­a­tion in patients with hemo­philia, thus
achiev­ing clin­i­cal hemo­sta­sis with­out throm­botic com­pli­ca­tions.
Factor mimetic ther­apy
Emicizumab (Genentech, San Francisco, CA) is a human­ ized,
bispecific, mono­clo­nal immu­no­glob­u­lin G4 anti­body (mAb) that
binds to acti­vated FIX (FIXa) and FX, thereby performing the
220  |  Hematology 2021  |  ASH Education Program
func­tion of FVIII by bring­ing FIXa and FX into close enough prox­
im­ity to facil­i­tate FX acti­va­tion.
First approved for bleed pro­phy­laxis in HA with inhib­it­ors in
the US by the US Food and Drug Administration in 2018, emici­
zumab has been appro­pri­ately rec­og­nized as sig­nal­ing a new
era of HA treat­ment. In all­phase 3 tri­als, emicizumab pro­phy­laxis
has led to dras­tic reduc­tions in annu­al­ized bleed­ing rates (ABRs),
with median ABRs of 2.6 for all­dos­ ing reg­i­
mens in patients
through the HAVEN 1 through 4 tri­als and with over 80% of par­
tic­i­pants expe­ri­enc­ing no bleeds after week 24 of ther­apy.9,10
Considering data show­ing diminishing FVIII expres­sion months
to years after adeno-asso­ci­ated virus gene ther­apy,11 emicizumab
or a sim­i­lar mimetic ther­apy may well be the dom­i­nant par­a­digm
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/219/1851769/219ellsworth.pdf by guest on 13 December 2021
in HA treat­ment for some time. In clin­i­cal prac­tice it is being suc­
cess­fully used in infants and other pre­vi­ously untreated patients,
who now grow up with­out bleed­ing or fac­tor expo­sure. Given
the low inci­dence of anti­drug antibodies, the ease of admin­is­
tra­tion with min­i­mal instruc­tion, and the extremely long half-life,
patients may expe­ri­ence a de facto cure with reg­u­lar admin­is­
tra­tion.
However, much remains unanswered in emicizumab use, and
opti­mal man­age­ment requires an expe­ri­enced hema­tol­o­gist.
Standard coag­u­la­tion tests and thus one-stage, clot-based FVIII
activ­ity assays (FVIII OSCA) are ren­dered unre­li­able due to the
inde­pen­dence of emicizumab from throm­bin gen­er­a­tion in ini­ti­
at­ing its activ­ity.
Thromboses and throm­botic microangiopathy were observed
in early tri­als, and all­were asso­ci­ated with the con­com­i­tant use
of aPCC.10 The mech­a­nism of this poten­tially dev­as­tat­ing adverse
effect is still unknown, though elu­ci­dat­ing this may yield insights
help­ful for cli­ni­cians treating with emicizumab in gen­eral.
Other anti­body fac­tor VIIIa mimet­ics with the same mech­
a­nism of action, bind­ing fac­tors IXa and X in close prox­im­ity,
lead­ing to throm­bin-inde­pen­dent FXa acti­va­tion, are in devel­
op­ment. Mim8 (Novo Nordisk, Bagsvaerd, Denmark) is a next-
gen­ er­
at­ion bispecific anti­ body to FIXa and X that has been
shown in vivo and in vitro to potently pro­mote throm­bin gen­er­a­
tion in the absence of FVIII.12 It is cur­rently being inves­ti­gated as
a next-gen­er­a­tion FVIII mimetic in a phase 1/2 trial (clinicaltrials​
­.gov). Another next-gen­er­at­ ion bispecific FVIII mimetic in devel­
op­ment is BS-027125 (Bioverativ, Waltham, MA), cur­rently in pre­
clin­i­cal eval­u­a­tion.13,14
FVIII is a cofac­tor rather than a ser­ine pro­te­ase, mak­ing it
ame­na­ble to ther­apy replacing it with a non­fac­tor com­pound,
leav­ing HB treat­ment want­ing a sim­i­lar par­a­digm change. Fortu­
nately, inge­nious manip­u­la­tion of the clot­ting cas­cade prom­ises
a sim­i­lar new era for HB. Approaches with roots in obser­va­tional
sci­ence are driv­ ing emerg­ ing hemo­ philia treat­ments known
as rebalancing ther­a­pies. These ther­a­pies seek to “rebalance”
coag­ u­la­
tion to a more nor­ mal state by alter­ ing the inher­
ent
phys­i­o­logic mod­u­la­tion of coag­u­la­tion (Table 1).
Rebalancing ther­a­pies as a treat­ment for hemo­philia
siRNA ther­a­peu­tics
With the advent of genome sequenc­ing, sci­en­tists iden­ti­fied
patients with severe hemo­philia who coinherited var­i­ous pro­
thrombotic gene muta­tions and displayed a milder phe­no­type.15
A pre­pon­der­ance of evi­dence exists for coinheritance of FVIII
and pro­throm­bin muta­tions, so it is per­haps fit­ting that target­
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/219/1851769/219ellsworth.pdf by guest on 13 December 2021
Figure 1.  Coagulation cascade. A simplified representation of the “coagulation cascade”. Note the role that the tenase complexes
play in thrombin generation. The intrinsic tenase complex consists of factors VIIIa, IXa, and X on a phospholipid surface with phos­
phatidylserine exposure, usually a platelet, and facilitates the activation of factor X. The extrinsic tenase complex consists of factor
VIIa, tissue factor, and factor X, likewise leading to the activation of factor X. Note the many feedback mechanisms of activation that
thrombin performs. Although the generation and exposure of TF at the site of vascular endothelial is the primary initiator of coagu­
lation via the extrinsic pathway, the intrinsic tenase pathway is important because active TF has only limited availability in vivo and
TFPI’s constitutive activity inhibits the extrinsic tenase complex from generating adequate thrombin for a stable clot (see reference 19
for a more detailed treatment of this topic). In a PTT test, a test on which clot-based factor assays are built, phospholipid and calcium
are added to a sample anticoagulated with sodium citrate (a calcium chelator that inhibits the Ca++ dependent steps as noted in the
figure). Thrombin is added to the assay and further generated by the thrombin burst (see text). The activation of factor VIII or IX is
the rate-limiting step in the assay. Factors are labeled by their traditional roman numeral. TF, tissue factor; TFPI, tissue factor pathway
inhibitor; EPCR, endothelial protein C receptor; APC, activated protein C.

ing the func­tion of throm­bin would be an early con­tender for a lead­ing to increased total throm­bin gen­er­ated with a hemo­static
rebalancing ther­apy. chal­lenge.16,17
Antithrombin (AT) is an endog­ e­
nous pro­
tein that nat­ u­rally In early-phase clin­i­cal tri­als to date, doses of fitusiran were
reg­u­
lates the func­ tion of active throm­
bin. Fitusiran (Alnylam, targeted to lower AT lev­els to 20%, which nor­mal­izes throm­bin
Cambridge, MA) is a small-mol­e­cule RNA inter­fer­ence ther­a­peu­ gen­er­a­tion and reduces bleed­ing. However, tri­als were briefly
tic that acts by bind­ing and degrading the mRNA encoding AT, put on hold in 2017 after a patient died after devel­op­ing a dural
Prakash Singh Shekhawat
Treating hemo­philia with­out fac­tor? | 221
Table 1. Summary of non-factor therapies

Name of Drug Mechanism ofAc on CurrentStatus Route ofAdministra on DosingSchedule

Factor VIIIMimecs

Emicizumab1 Humanized monoclonal IgG4 anbody with specificity to FIXa and FX FDA approved in US, EMA approved in Europe Subcutaneous injecon Generally weekly or every other week, though approved for intervals
up to every 4 weeks

Mim82 Monoclonal anbody specific to FIXa and FX Phase 1/2 trial Subcutaneous injecon Weekly and monthly injecons are being evaluated in the phase 2
poron of the study.3

BS-0271254 Monoclonal anbody specific to FIXa and FX Preclinical evaluaon TBD TBD

siRNA Therapies

Fitusiran3,5 siRNA against AT3 mRNA, leading to decreased AT protein translaon Phase 2 and 3 trials ongoing Subcutaneous injecon Monthly

siRNA against Protein S mRNA, leading to decreased Protein S translaon. Preclinical/animal models TBD TBD

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/219/1851769/219ellsworth.pdf by guest on 13 December 2021

Protein S siRNA (unnamed)7

TFPI Inhibitors

Concizumab8 Humanized monoclonal IgG4 against TFPI, Kunitz domain 2 Phase 2 and 3 trials ongoing. Phase 3 trials in paents without Subcutaneous injecon Daily
inhibitors are sll recruing.

Befovacimab(BAY1093884)8 Humanized monoclonal IgG4 against TFPI, Kunitz domains 1 and 2 Phase 2 trial terminated due to 2 paents with arterial thrombi Subcutaneous injecon Weekly

Marstacimab(PF-06741086)9 Monoclonal Ab against TFPI Phase 1 – 3 trials are accruing Subcutaneous injecon Weekly

MG11133,8 Monoclonal Ab against TFPI Phase 1 trials in healthy volunteers. Preclinical, non-human Subcutaneous injecon Weekly
primate models showed decreased bleeding.

BAX49910 Aptamer disrupng TFPI binding to extrinsic tenase complex. Phase 1/2 studies terminated due to bleeding in subjects. Intravenous N/A

Other Serine Protease Targets

Serpin PC3,11 Alpha-1-antrypsin-like serine protease which inhibits acvated protein C Phase 1/2 trial evaluang in hemophilia A and B paents Subcutaneous or intravenous TBD

HAPC157312 Monoclonal anbody to acvated protein C Preclinical, non-human primate models have shown decreased TBD TBD
bleeding.

Protein Z related protease
inhibitor (unnamed)13
Protease Nexin-1 inhibitor
(unnamed)14
Inacve, mutant protein Z, prevenng protein Z (a cofactor) binding to Preclinical, in vitro thrombin generaon is enhanced. TBD TBD
protein Z-dependent protease inhibitor, prevenng FXa inhibion by the
endogenous complex.
Endogenous glycoprotein that is secreted by acvated platelets and inhibits Preclinical murine knockout model, no specific drug in TBD TBD
FXIa and thrombin, among other coagulaon factors. development.

This table includes currently approved therapeucs (emicizumab), those in advanced phase trials (fitusiran, concizumab, marstacimab), and others which are either in preclinical invesgaon or suspended invesgaon as far as the authors are currently aware. References for informaon
provided in text. Paent convenience and acceptability are important consideraons and noted in this table. An-TFPI mAb drugs are administered by daily subcutaneous injecons, making them less convenient than siRNA promises to be. Ulmately though, physician opinion, paent
symptoms and personal preferences and values will decide appropriate therapy.
FVIII - Factor VIII, FIXa - acvated factor IX, FX - Factor X, siRNA - small interfering RNA, mRNA - messenger RNA, TFPI - ssue factor pathway inhibitor

1. Callaghan MU, Negrier C, Paz-Priel I, et al. Long-term outcomes with emicizumab prophylaxis for hemophilia A with or without FVIII inhibitors from the HAVEN 1-4 studies. Blood. 2021;137(16):2231-2242. doi:10.1182/blood.2020009217
2. Kjellev SL, Østergaard H, Greisen PJ, et al. Mim8 - a Next-Generaon FVIII Mimec Bi-Specific Anbody - Potently Restores the Hemostac Capacity in Hemophilia a SeŸngs in Vitro and In Vivo. Blood. 2019;134(Supplement_1):96-96. doi:10.1182/blood-2019-122817
3. clinicaltrials.gov
4. Aleman MM, Jindal S, Leksa N, Peters R, Salas J. Phospholipid-Independent Acvity of Fviiia Mimec Bispecific Anbodies in Plasma. Blood. 2018;132(Supplement 1):2461-2461. doi:10.1182/blood-2018-99-119226
5. Machin N, Ragni MV. An invesgaonal RNAi therapeuc targeng anthrombin for the treatment of hemophilia A and B. J Blood Med. 2018;9:135-140. doi:10.2147/JBM.S159297
6. Pasi KJ, Lissitchkov T, Mamonov V, et al. Targeng of anthrombin in hemophilia A or B with invesgaonal siRNA therapeuc fitusiran - results of the phase 1 inhibitor cohort. J Thromb Haemost. February 2021. doi:10.1111/jth.15270
7. Prince R, Bologna L, ManeŸ M, et al. Targeng ancoagulant protein S to improve hemostasis in hemophilia. Blood. 2018;131(12):1360-1371. doi:10.1182/blood-2017-09-800326
8. Mahlangu JN. Progress in the Development of An-ssue Factor Pathway Inhibitors for Haemophilia Management. Front Med (Lausanne). 2021;8:670526. doi:10.3389/fmed.2021.670526
9. Patel-He¥ S, Marn EJ, Mohammed BM, et al. Marstacimab, a ssue factor pathway inhibitor neutralizing anbody, improves coagulaon parameters of ex vivo dosed haemophilic blood and plasmas. Haemophilia. 2019;25(5):797-806. doi:10.1111/hae.13820
10. Spadarella G, Di Minno A, Milan G, et al. Paradigm shi¦ for the treatment of hereditary haemophilia: Towards precision medicine. Blood Rev. 2020;39:100618. doi:10.1016/j.blre.2019.100618
11. Weyand AC, Pipe SW. New therapies for hemophilia. Blood. 2019;133(5):389-398. doi:10.1182/blood-2018-08-872291
12. Zhao X-Y, Wilmen A, Wang D, et al. Targeted inhibion of acvated protein C by a non-acve-site inhibitory anbody to treat hemophilia. Nat Commun. 2020;11(1):2992. doi:10.1038/s41467-020-16720-9
13. Aymonnier K, Kawecki C, Arocas V, Boula¦ali Y, Bouton MC. Serpins, new therapeuc targets for hemophilia. Thromb Haemost. 2021;121(3):261-269. doi:10.1055/s-0040-1716751

sinus throm­bo­sis fol­low­ing high-dose rFVIII admin­is­tra­tion to Another pre­clin­i­cal siRNA ther­a­peu­tic shares fitusiran’s mech­
treat a bleed dur­ing the open-label exten­sion of the trial.17 The a­nism but silences pro­tein S expres­sion rather than AT. It is in
hold was lifted after pro­to­col amend­ments were made. The drug pre­clin­i­cal inves­ti­ga­tion for even­tual use in hemo­philia.19
had no throm­botic com­pli­ca­tions in a phase 1 cohort of patients
with inhib­i­tors,18 and results of the phase 2 trial have not been Tissue fac­tor path­way inhib­i­tors
published. Subsequently, fol­low­ing fur­ther non­fa­tal throm­botic Another way to exploit inher­ent mech­a­nisms to pro­mote hemo­
events, clin­i­cal tri­als were again held, and doses were reduced in sta­sis is by blocking tis­sue fac­tor path­way inhib­i­tor (TFPI), an
Octo­ber 2020. Revised dos­ing and AT tar­gets have been adopted endog­e­nous ser­ine pro­te­ase inhib­i­tor that pre­vents the acti­va­
for ongo­ing phase 3 par­tic­i­pants (clinicaltrials​­.gov). tion of FX by the TF-FVIIa com­plex, thus lim­it­ing the degree of
In tri­als, fitusiran is now admin­is­tered as a sub­cu­ta­ne­ous injec­ throm­bin gen­er­a­tion via the con­tact path­way. By disrupting TFPI
tion every other month. Since the decrease in AT affects the com­ bind­ing to this extrin­sic tenase com­plex, Xa and throm­bin gen­er­
mon path­way, this can treat either HA or HB, poten­tially with or a­tion are increased. This approach has been suc­cess­ful via mono­
with­out inhib­i­tors. clo­nal antibodies to var­i­ous domains of the TFPI mol­ec ­ ule.20

222  |  Hematology 2021  |  ASH Education Program

 Concizumab (Novo Nordisk) is a human­ized immu­no­glob­u­lin
G4 anti-TFPI anti­body to the sec­ond Kunitz (K2) domain of TFPI.
Building on obser­va­tional data in humans, ani­mal mod­els dem­
on­strated restored throm­bin gen­er­a­tion and decreased bleed­
ing despite deficiencies in FVIII or FIX.21 Phase 1 and 2 tri­als in
hemo­philia patients with­out inhib­i­tors dem­on­strated reduced
ABRs and no throm­bo­em­bolic events. Phase 3 tri­als in HA and HB
were tem­po­rar­ily suspended due to non­fa­tal throm­botic events
in early 2020 but have since resumed. In addi­tion, anti­drug anti­
bodies have been observed in tri­als for concizumab.22,23
Befovacimab (BAY1093884, Bayer) is spe­cific to both K2 and
K1 domains, blocking TFPI bind­ing to both FVIIa and FX in the
extrin­sic tenase com­plex and enhanc­ing throm­bin gen­er­a­tion
despite deficiencies in the con­tact path­way. In an early study, it
decreased bleed­ing epi­sodes, but the trial was ter­mi­nated when
2 patients devel­oped cere­bral arte­rial thrombi, and one devel­
oped a cere­bral venous throm­bus.22,23
PF-06741086 (Marstacimab, Pfizer) has been shown to nor­
mal­ize coag­u­la­tion in hemo­philia patient plas­mas ex vivo and is
being eval­u­ated in HA and HB patients (clinicaltrials​­.gov). There
have been no throm­botic com­pli­ca­tions to date, with reduc­
tion of ABRs to zero in most par­ tic­
i­
pants.22,24 MG1113 (Green
Cross Corporation) is another TFPI mAb like­wise being tested
in healthy vol­un­teers (clinicaltrials​­.gov), with non­hu­man pri­mate
mod­els hav­ing dem­on­strated in vivo reduc­tion of blood loss.22
An aptamer derived from recom­bi­nant human TFPI (BAX499,
Takeda) was also devel­oped and found to effi­ciently inhibit TFPI
in vitro and in vivo, with dose-depen­dent increases in throm­
bin gen­er­a­tion and decreased bleed­ing in ani­mal mod­els and
in early-phase human tri­als.15,25 Development is on hold due to
bleed­ing in sub­jects.15
Other ser­ine pro­te­ase tar­gets
SerpinPC is an alpha-1-antitrypsin-like ser­ine pro­te­ase mod­i­fied
to inhibit activated protein C (APC).26 By pre­vent­ing the action
of APC, which inhib­its FV acti­va­tion, throm­bin gen­er­a­tion is like­
wise restored despite fac­tor deficiencies in the con­tact sys­tem
via enhanced com­mon path­way activ­ity (Figure 1).27 This is cur­
rently in a phase 1/2 trial, with approx­i­ma­tely 50 par­tic­i­pants
hav­ing both HA and HB (clinicaltrials​­.gov). Recently, a mono­clo­
nal anti­body to acti­vated pro­tein C (HAPC1573) has been shown
in non­hu­man pri­ma­tes to pre­vent bleed­ing.28
Protein Z (PZ)-related pro­te­ase inhib­i­tor is an inac­tive mutant
PZ, cofac­tor to PZ-depen­dent pro­te­ase inhib­i­tor. Endogenously,
the PZ-depen­dent pro­te­ase inhib­i­tor com­plex inhib­its FXa. By
pre­vent­ing this com­plex for­ma­tion through var­i­ous meth­ods,
pre­clin­i­cal data have shown increased throm­bin gen­er­a­tion in
vitro.29
Finally, pro­ te­
ase nexin-1 has been pro­ posed as a tar­ get.
Endogenously, this gly­co­pro­tein is expressed by acti­vated plate­
lets and inhib­its FXIa, throm­bin, and other fac­tors. Hemophilic
mice showed decreased bleed­ing in pro­te­ase nexin-1 knock­out
mod­els, mak­ing this another prom­is­ing tar­get.29
The eco­nom­ics of non­fac­tor ther­apy
Although poten­tially par­a­digm chang­ing for patients, non­fac­tor
ther­a­pies may come at a par­a­digm-chang­ing price. Cost ana­ly­ses
of new mono­clo­nal antibodies for var­i­ous indi­ca­tions are com­mon
in the lit­er­a­ture. Although ana­ly­ses sug­gest cost-effec­tive­ness for
Prakash Singh Shekhawat
some mAb according to many met­rics for rare dis­eases, equity in
access to treat­ment with these expen­sive med­i­ca­tions must be
ensured for these ben­e­fits to be real­ized.30
Some authors have found that mAb costs for malig­nant and
non­ma­lig­nant hema­to­logic dis­or­ders skew higher than those
marketed for use in other dis­or­ders.31 However, emicizumab has
actu­ally been shown to be highly cost-effec­tive in sev­eral real-
world ana­ly­ses, owing to its superb effi­cacy and the already high
cost of hemo­philia care with fac­tor infu­sions and inhib­i­tor devel­
op­ment.32,33 Current data describe an existing high bur­den in
treating those patients who fail fac­tor ther­apy or can­not secure
access to effec­tive fac­tor ther­apy.34,35 In this con­text, costs of
new ther­a­pies may be off­set by a reduced bur­den of the com­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/219/1851769/219ellsworth.pdf by guest on 13 December 2021
pli­ca­tions of stan­dard ther­apy. However, con­cerns remain about
the long-term effects of these ther­a­pies, which may carry hid­den
eco­nomic costs and dis­ad­van­tages.
Concerns and pit­falls
For fac­tor mimetic and rebalancing ther­a­pies that have reached
a suf­fi­ciently advanced stage in clin­i­cal tri­als, a reduc­tion in ABRs
has indeed been noted. Note that this effi­cacy has been shown
thus far in pro­phy­laxis only, includ­ing in emicizumab. This is to
be expected con­sid­er­ing the mech­a­nism of action of these ther­
a­peu­tics but is impor­tant in eval­u­at­ing the lim­i­ta­tions and poten­
tial appli­ca­tion of these new and emerg­ing drugs.
As noted in the dis­cus­sion of these agents, every non­fac­tor
ther­apy that has reached clin­i­cal trial has had throm­botic side
effects with the excep­tion of marstacizumab.23,29,36 With emici­
zumab, all­throm­ bo­ ses and throm­ botic microangiopathy epi­
sodes were asso­ci­ated with aPCC use, and safety oth­er­wise is
becom­ing well established.10 But safety in the set­ting of sur­ger­
ies and trauma remains an unre­solved prob­lem. Although the
cur­rent aim of non­fac­tor ther­apy is pro­phy­laxis, the con­com­i­tant
man­age­ment of perioperative and trauma-related bleed­ing is of
grow­ing inter­est,35-37 with con­sen­sus that fac­tor con­cen­trate use
is still nec­es­sary in these cases. Though stan­dard­ized, opti­mized
approaches do not yet exist, cur­rent con­sen­sus guide­lines pro­
vide detailed guid­ance based on real-world use thus far.35,38,39
Additionally, expe­ri­ence in com­pet­i­tive ath­letes is lacking,
though expert opin­ion gen­er­ally con­sid­ers monotherapy insuf­fi­
cient if break­through bleed­ing is observed.23,37
Because mimetic and rebalancing ther­a­pies alter hemo­sta­sis
either inde­pen­dently or by directly mod­i­fy­ing reg­u­la­tory mech­a­
nisms of coag­u­la­tion, patient selec­tion will also need to take into
account throm­botic risk fac­tors such as obe­sity, existing car­dio­
vas­cu­lar dis­ease, dia­be­tes, and other con­di­tions.
Aside from effi­cacy and safety in gen­eral, mean­ing­ful trial end
points are essen­tial for emerg­ing ther­a­pies. For exam­ple, some
extended half-life fac­tor prod­ucts, despite show­ing ade­quate
trough lev­els in clin­i­cal trial, have been found in real-world use
to have infe­rior bleed­ing response.38 The approval and wide clin­
i­cal adop­tion of emicizumab reflect its proven effi­cacy in reduc­
ing ABRs in patients with and with­out inhib­i­tors.39 As mon­i­tor­ing
will not be as straight­for­ward as with fac­tor replace­ment, ABR
reduc­tion must remain the end point of choice in non­fac­tor ther­
apy tri­als.15,23,40
The mon­i­tor­ing of non­fac­tor ther­a­pies is com­pli­cated and
impos­si­ble with stan­dard screen­ing tests and assays, a promi­
nent sub­ject of con­cern.40 This is because these drugs exert their
Treating hemo­philia with­out fac­tor? | 223
effect inde­pen­dently of acti­va­tion by throm­bin itself, the reac­
tion noted above being the rate-lim­it­ing step in such clot-based
assays,39 or by directly mod­i­fy­ing throm­bin gen­er­a­tion. Though
OSCA and chro­mo­genic fac­tor assays are widely avail­­able, val­i­
dated mon­i­tor­ing of non­fac­tor agents is forth­com­ing.
Patients started on emicizumab while still hav­ ing active
inhib­i­tors are another pop­u­la­tion who will ben­e­fit from fur­ther
research to opti­mize inhib­i­tor erad­i­ca­tion, with approaches
likely to dif­fer between pedi­at­ric and adult patients.41,42 Alloan­
tibodies to the drugs them­selves and inhib­it­ or recur­rence while
on mimetic ther­apy are emerg­ing issues as well.29,43,44
Other con­cerns with the advent of fac­tor-mimetic ther­ap ­ ies,
espe­cially in long-term use, are that they do not per­form some
impor­tant func­tions of the fac­tor mol­e­cules in maintaining vas­
cu­lar integ­rity and interacting with the cel­lu­lar com­po­nents of
blood.39 Additionally, TFPI knock­ out mice develop pre­ ma­ture
ath­ero­scle­ro­sis, suggesting that mod­u­lat­ing the hemo­static sys­
tem could have unin­tended con­se­quences out­side of the acute
hemo­static effect.45 Indeed, the act of rebalancing a sys­tem that
exists in such a com­plex bal­ance is bound to be a dif­fi­cult prop­
o­si­tion, and long-term fol­low-up data will be crit­i­cal in all­emerg­
ing ther­a­pies.
Factor mimetic and rebalancing ther­ a­
pies will con­ tinue to
coevolve with fac­tor ther­apy. These ther­ap ­ ies and the accom­
pa­ny­ing com­plex­ity involved in their use will con­tinue to require
com­pre­hen­sive, expert care. The myr­iad offer­ings will match the
com­plex needs of our patients and even­tu­ally lead to supe­rior
care over­all. While expert opin­ion still regards fac­tor replace­ment
as the cur­rent stan­dard of care for hemo­philia,3,37 this “stan­dard” is
con­tinu­ing to change. Future ther­apy, pro­tean as it may be, has a
com­mon goal: a stan­dard of result that our patients do not bleed.
Final thoughts
There will be a need for fac­tor replace­ment ther­apy in the treat­
ment of hemo­philia for the fore­see­able future, but non­fac­tor ther­
a­pies ful­fill a great need and rep­re­sent a new phase of hemo­philia
care. Some of those needs not cur­rently ame­na­ble to non­fac­tor
ther­ap­ ies include the man­age­ment of sur­ger­ies and trauma.
Are non­fac­tor ther­a­pies the end of fac­tor ther­apy? Perhaps
the more impor­tant ques­tion is whether they should be the end
of fac­tor ther­apy. Preliminary data seem to sug­gest that alter­ing
the hemo­static bal­ance is a fraught prop­o­si­tion, and ulti­mately,
the best option is that which yields the best result for a given
patient. Some patients will not want new options to replace cur­
rent treat­ments. Maybe they will be reassured by the num­ber of
choices. Clinicians should find reas­sur­ance in this as well.
Returning to our intro­duc­tory cases, are they can­di­dates for
non­fac­tor ther­a­pies? Certainly, our young man in Case 1 would
clearly ben­e­fit from ther­apy to free him from fre­quent BPA use in
a cen­tral line and the risks of ongo­ing immu­no­sup­pres­sion. Our
gen­tle­man from Case 2 may have more to think about. Someone
who has never had fac­tor ther­apy opti­mized lacks a com­pel­ling
rea­son to con­sider newer ther­apy, though the deci­sion would
ulti­mately rest on his val­ues and per­cep­tion of his qual­ity of life. As
we have seen with emicizumab, even patients with­out inhib­i­tors
and other dif­fi­cul­ties have benefited from non­fac­tor treat­ment.
Also, the sim­plic­ity of sub­cu­ta­ne­ous dos­ing in a patient who has
not learned self-infu­sion is likely to improve bleed pro­phy­laxis.
Even if these non­fac­tor tools are not the end of fac­tor ther­apy,
they do seem poised to inau­gu­rate a new chap­ter in the his­tory
224  |  Hematology 2021  |  ASH Education Program
of hemo­philia—a chap­ter of pre­ci­sion treat­ment, in which many
roads lead to a life unen­cum­bered by the threat of bleed­ing.
Conflict-of-inter­est dis­clo­sure
Patrick Ellsworth: Is an NHF-Takeda Clinical Research Fellowship
award recipient.
Alice Ma: research funding and hon­or­ aria: Takeda.
Off-label drug use
Patrick Ellsworth: no off-label drug use has been discussed.
Alice Ma: no off-label drug use has been discussed.
Correspondence
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/219/1851769/219ellsworth.pdf by guest on 13 December 2021
Alice Ma, University of North Carolina at Chapel Hill, Houpt Bldg,
CB# 7305, 170 Manning Dr, 3rd Fl, Chapel Hill, NC 27599-7305;
e-mail: alice_ma@med​­.unc​­.edu.
References
1. Franchini M, Mannucci PM. The his­ tory of hemo­ philia. Semin Thromb
Hemost. 2014;40(5):571-576.
2. Mannucci PM. Hemophilia ther­apy: the future has begun. Haematologica.
2020;105(3):545-553.
3. Aledort L, Mannucci PM, Schramm W, Tarantino M. Factor VIII replace­ment
is still the stan­dard of care in haemophilia A. Blood Transfus. 2019;17(6):479-
486.
4. Witmer C, Young G. Factor VIII inhib­i­tors in hemo­philia A: ratio­nale and lat­est
evi­dence. Ther Adv Hematol. 2013;4(1):59-72.
5. Dolan G, Benson G, Duffy A, et al. Haemophilia B: where are we now and
what does the future hold? Blood Rev. 2018;32(1):52-60.
6. Santoro C, Quintavalle G, Castaman G, et  al. Inhibitors in hemo­philia B.
Semin Thromb Hemost. 2018;44(6):578-589.
7. Antun A, Monahan PE, Manco-Johnson MJ, et al. Inhibitor recur­rence after
immune tol­er­ance induc­tion: a mul­ti­cen­ter ret­ro­spec­tive cohort study.
J Thromb Haemost. 2015;13(11):1980-1988.
8. Ma AD, Kessler CM, Al-Mondhiry HA, Gut RZ, Cooper DL. US expe­ri­ence
with recom­ bi­
nant fac­tor VIIa for sur­ gery and other inva­ sive pro­
ce­
dures
in acquired haemophilia: anal­ y­
sis from the Hemostasis and Thrombosis
Research Society Registry. Haemophilia. 2016;22(1):e18-e24.
9. Yada K, Nogami K. Spotlight on emicizumab in the man­ age­ment of
hemo­philia A: patient selec­tion and spe­cial con­sid­er­ations. J Blood Med.
2019;10(2 July):171-181.
10. Callaghan MU, Negrier C, Paz-Priel I, et al. Long-term out­comes with emi­
cizumab pro­phy­laxis for hemo­philia A with or with­out FVIII inhib­i­tors from
the HAVEN 1-4 stud­ies. Blood. 2021;137(16):2231-2242.
11. Pasi KJ, Rangarajan S, Mitchell N, et al. Multiyear fol­low-up of AAV5-hFVIII-
SQ gene ther­apy for hemo­philia A. N Engl J Med. 2020;382(1):29-40.
12. Kjellev SL, Østergaard H, Greisen PJ, et al. Mim8—a next-gen­er­a­tion FVIII
mimetic bi-spe­cific anti­body—potently restores the hemo­static capac­ity
in hemo­philia a set­tings in vitro and in vivo. Blood. 2019;134(suppl 1):96.
13. Leksa NC, Aleman MM, Goodman AG, Rabinovich D, Peters R, Salas J.
Intrinsic dif­ fer­
ences between FVIIIa mimetic bispecific antibodies and
FVIII pre­vent assign­ment of FVIII-equiv­a­lence. J Thromb Haemost. 2019;​
17(7):1044-1052.
14. Aleman MM, Jindal S, Leksa N, Peters R, Salas J. Phospholipid-inde­pen­dent
activ­ity of Fviiia mimetic bispecific antibodies in plasma. Blood. 2018;132
(­suppl 1):2461.
15. Spadarella G, Di Minno A, Milan G, et  al. Paradigm shift for the treat­
ment of hered­i­tary haemophilia: towards pre­ci­sion med­i­cine. Blood Rev.
2020;39(Jan­u­ary):100618.
16. Pasi KJ, Rangarajan S, Georgiev P, et al. Targeting of anti­throm­bin in hemo­
philia A or B with RNAi ther­apy. N Engl J Med. 2017;377(9):819-828.
17. Machin N, Ragni MV. An inves­ti­ga­tional RNAi ther­a­peu­tic targeting anti­
throm­bin for the treat­ment of hemo­philia A and B. J Blood Med. 2018;9(22
August):135-140.
18. Pasi KJ, Lissitchkov T, Mamonov V, et al. Targeting of anti­throm­bin in hemo­
philia A or B with inves­ti­ga­tional siRNA ther­a­peu­tic fitusiran—results of the
phase 1 inhib­i­tor cohort. J Thromb Haemost. 2021;19(6):1436-1446.
19. Prince R, Bologna L, Manetti M, et al. Targeting anti­co­ag­u­lant pro­tein S to
improve hemo­sta­sis in hemo­philia. Blood. 2018;131(12):1360-1371.
20. Chowdary P. Inhibition of tis­ sue fac­tor path­
way inhib­i­
tor (TFPI) as a
treat­ment for haemophilia: ratio­nale with focus on concizumab. Drugs.
2018;78(9):881-890.
21. Morfini M, Marchesini E. The avail­abil­ity of new drugs for hemo­philia treat­
ment. Expert Rev Clin Pharmacol. 2020;13(7):721-738.
22. Mahlangu JN. Progress in the devel­op­ment of anti-tis­sue fac­tor path­way
inhib­i­tors for haemophilia man­age­ment. Front Med (Lausanne). 2021;8(5
May):670526.
23. Jiménez-Yuste V, Auerswald G, Benson G, et  al. Practical con­sid­er­ations
for non­fac­tor-replace­ment ther­a­pies in the treat­ment of haemophilia with
inhib­i­tors. Haemophilia. 2021;27(3):340-350.
24. Patel-Hett S, Martin EJ, Mohammed BM, et  al. Marstacimab, a tis­sue fac­
tor path­way inhib­i­tor neu­tral­iz­ing anti­body, improves coag­u­la­tion param­
e­ters of ex vivo dosed haemophilic blood and plas­ mas. Haemophilia.
2019;25(5):797-806.
25. Miesbach W, Eladly F. Current and future options of haemophilia A treat­
ments. Expert Opin Biol Ther. 2021:1 April:1-8.
26. Pelland-Marcotte MC, Carcao MD. Hemophilia in a chang­ ing treat­ ment
land­scape. Hematol Oncol Clin North Am. 2019;33(3):409-423.
27. Weyand AC, Pipe SW. New ther­a­pies for hemo­philia. Blood. 2019;133(5):389-
398.
28. Zhao XY, Wilmen A, Wang D, et al. Targeted inhi­bi­tion of acti­vated pro­tein
C by a non-active-site inhib­i­tory anti­body to treat hemo­philia. Nat Com-
mun. 2020;11(1):2992.
29. Aymonnier K, Kawecki C, Arocas V, Boulaftali Y, Bouton MC. Serpins, new
ther­a­peu­tic tar­gets for hemo­philia. Thromb Haemost. 2021;121(3):261-269.
30. Park T, Griggs SK, Suh DC. Cost effec­tive­ness of mono­clo­nal anti­body ther­
apy for rare dis­eases: a sys­tem­atic review. BioDrugs. 2015;29(4):259-274.
31. Hernandez I, Bott SW, Patel AS, et al. Pricing of mono­clo­nal anti­body ther­
a­pies: higher if used for can­cer? Am J Manag Care. 2018;24(2):109-112.
32. Cortesi PA, Castaman G, Trifirò G, et  al. Cost-effec­tive­ness and bud­get
impact of emicizumab pro­phy­laxis in haemophilia A patients with inhib­i­
tors. Thromb Haemost. 2020;120(2):216-228.
33. Polack B, Trossaërt M, Cousin M, Baffert S, Pruvot A, Godard C. Cost-
effec­tive­ness of emicizumab vs bypassing agents in the pre­ ven­
tion of
bleed­ing epi­sodes in haemophilia A patients with anti-FVIII inhib­i­tors in
France. Haemophilia. 2021;27(1):e1-e11.
34. Buckner TW, Bocharova I, Hagan K, et al. Health care resource uti­li­za­tion
and cost bur­den of hemo­philia B in the United States. Blood Adv. 2021;5(7):​
1954-1962.
Prakash Singh Shekhawat
35. D’Angiolella LS, Cortesi PA, Rocino A, et  al. The socio­eco­nomic bur­den
of patients affected by hemo­philia with inhib­i­tors. Eur J Haematol. 2018;​
101(4):​435-456.
36. Lillicrap D, Fijnvandraat K, Young G, Mancuso ME. Patients with hemo­philia
A and inhib­i­tors: pre­ven­tion and evolv­ing treat­ment par­a­digms. Expert Rev
Hematol. 2020;13(4):313-321.
37. Srivastava A, Santagostino E, Dougall A, et al; World Federation of Hemo­
philia Guidelines for the Management of Hemophilia pan­ el­ists and co-
authors. WFH Guidelines for the Management of Hemophilia, 3rd edi­tion.
Haemophilia. 2020;26 (suppl 6):1-158.
38. Malec LM, Croteau SE, Callaghan MU, Sidonio RF Jr. Spontaneous bleed­
ing and poor bleed­ing response with extended half-life fac­tor IX prod­
ucts: a sur­vey of select US haemophilia treat­ment cen­tres. Haemophilia.
2020;26(3):e128-e129.
39. Lenting PJ, Denis CV, Christophe OD. Emicizumab, a bispecific anti­body
rec­og­niz­ing coag­u­la­tion fac­tors IX and X: how does it actu­ally com­pare to
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/219/1851769/219ellsworth.pdf by guest on 13 December 2021
fac­tor VIII? Blood. 2017;130(23):2463-2468.
40. Lenting PJ. Laboratory mon­i­tor­ing of hemo­philia A treat­ments: new chal­
lenges. Blood Adv. 2020;4(9):2111-2118.
41. Escuriola-Ettingshausen C, Auerswald G, Königs C, et  al. Optimizing the
man­age­ment of patients with haemophilia A and inhib­i­tors in the era of
emicizumab: rec­ om­ men­ da­tions from a Ger­ man expert panel. Haemo-
philia. 2021;27(3):e305-e313.
42. Linari S, Castaman G. Concomitant use of rFVIIa and emicizumab in peo­ple
with hemo­philia A with inhib­i­tors: cur­rent per­spec­tives and emerg­ing clin­
i­cal evi­dence. Ther Clin Risk Manag. 2020;16(May 22):461-469.
43. Harroche A, Sefiane T, Desvages M, et al. Non-inhib­i­tory antibodies induc­ing
increased emicizumab clear­ance in a severe haemophilia A inhib­i­tor patient.
Haematologica. 2021;106(8):2287-2290.
44. Capdevila L, Borgel D, Lasne D, et al. Reappearance of inhib­i­tor in a toler­
ized patient with severe haemophilia A dur­ing FVIII-free emicizumab ther­
apy. Haemophilia. 2021;27(4):e581-e584.
45. Xiao J, Jin K, Wang J, et al. Conditional knock­out of TFPI-1 in VSMCs of mice
accel­er­ates ath­ero­scle­ro­sis by enhanc­ing AMOT/YAP path­way. Int J Car-
diol. 2017;228(Feb­ru­ary):605-614.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000253
Treating hemo­philia with­out fac­tor? | 225
 HEMOPHILIA UPDATE: OUR CUP RUNNETH OVER

Hemophilia gene therapy: ushering in a new

treatment paradigm?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/226/1851420/226george.pdf by guest on 13 December 2021

Lindsey A. George
Department of Pediatrics, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA; and Division of Hematology and
Raymond G. Perelman Center for Cellular and Molecular Therapeutics, Children’s Hospital of Philadelphia, Philadelphia, PA

After 3 decades of clinical trials, repeated proof-of-concept success has now been demonstrated in hemophilia A and
B gene therapy. Current clinical hemophilia gene therapy efforts are largely focused on the use of systemically admin-
istered recombinant adeno-associated viral (rAAV) vectors for F8 or F9 gene addition. With multiple ongoing trials,
including licensing studies in hemophilia A and B, many are cautiously optimistic that the first AAV vectors will obtain reg-
ulatory approval within approximately 1 year. While supported optimism suggests that the goal of gene therapy to alter
the paradigm of hemophilia care may soon be realized, a number of outstanding questions have emerged from clinical
trial that are in need of answers to harness the full potential of gene therapy for hemophilia patients. This article reviews
the use of AAV vector gene addition approaches for hemophilia A and B, focusing specifically on information to review
in the process of obtaining informed consent for hemophilia patients prior to clinical trial enrollment or administering a
licensed AAV vector.

LEARNING OBJECTIVES
• Understand information important to review with hemophilia patients prior to consenting for a clinical trial
or approved AAV vector
• Differentiate information that is well understand from information that remains incompletely understood

occurring AAV, which is nonpathogenic in humans and rep-

CLINICAL CASE
A 62-year-old man with severe hemophilia A (HA) with a his-
tory of an inhibitor, eradicated by immune tolerance induc-
tion, on emicizumab prophylaxis is interested in enrolling
in a HA gene therapy trial or receiving a licensed adeno-
associated viral (AAV) vector that will be available soon.
His comorbidities include 3 target joints, a prior 2-decade
history of the hepatitis C virus (HCV) infection with stage 2
liver fibrosis, and HIV well controlled on antiretroviral med-
ications. He wants to know what is known and unknown
about hemophilia gene therapy.
Introduction: What are the current approaches
for hemophilia gene therapy?
Clinical approaches in hemophilia gene therapy nearly exclu-
sively use systemically delivered recombinant AAV (rAAV)
vectors to target hepatocyte expression of an episom-
ally maintained transgene expressed under a hepatocyte
promoter (Table 1). The vector is engineered from naturally
lication defective. rAAV vectors are produced in cell culture
using either HEK293 mammalian cells or baclovirus/Sl9 insect
cell culture. The design and manufacturing of rAAV vectors
was recently reviewed by Li and Samulski.1
While hepatocytes are the natural site of factor IX (FIX)
synthesis, liver sinusoidal endothelial cells are the pre-
dominate site of factor VIII (FVIII) synthesis.2 Most HA rAAV
vectors contain a B-domain deleted FVIII, FVIII-SQ trans-
gene that retains full procoagulant function and accom-
modates AAV vector packaging constraints (4.7 kB).3 This
exact amino acid sequence has been used for recombinant
protein HA therapy for 2 decades (eg, Xyntha/ReFacto®,
Pfizer) without evidence of increased risk of inhibitor for-
mation. A single trial is using an alternative B-domain
deleted FVIII, FVIII-V3 that consists of the SQ linker with 6
additional N-linked glycosylation sites to improve expres-
sion.4,5 All enrolling hemophilia B (HB) trials have adapted
the use of the FIX-Padua (FIX-R338L) mutation, which is a
naturally occurring missense mutation with approximately
8-fold greater specific activity relative to wild-type FIX.6-10

226 | Hematology 2021 | ASH Education Program

Table 1. Ongoing hemo­philia A and B gene ther­apy tri­als

Manufacturing Capsid
Sponsor Clinicaltrials​­.gov plat­form Transgene sero­type Dose (vg/kg) Phase Ref
HB
  UCL/St. Jude NCT00979238 Mammalian scFIX AAV8 2 × 1011 to 2 × 1012 1/2 16,45

 Pfizer/Spark NCT03681273 Mammalian ssFIX-R338L SPK100 5 × 1011 1/2, 3 9

NCT03307980
 uniQure NCT02396342 Sl9 (insect) ssFIX-R338L AAV5 2 × 1013 3 8

 Freeline NCT03369444 Mammalian scFIX-R338L AAVS3 7.5 × 10 11 10

9.5 × 1011
HA
 BioMarin NCT02576795 Sl9 (insect) ssFVIII-SQ AAV5 6 × 1013 1/2, 3 11,15

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/226/1851420/226george.pdf by guest on 13 December 2021

NCT03392974
NCT03370913
 Sangamo/Pfizer NCT03061201 Sl9 (insect) ssFVIII-SQ AAV6 3 × 1013 1/2, 3 18

 Spark NCT03003533 Mammalian ssFVIII-SQ LK03 5 × 10 to 2 × 10

11 12
1/2 14

NCT03432520
  UCL/St. Jude NCT03001830 Mammalian ssFVIII-V3 AAV8 6 × 1011 to 6 × 1012 1/2 4,5

 Bayer/Ultragenyx NCT03588299 Mammalian ssFVIII-SQ AAVhu37 5 × 10 to 2 × 10

12 13
1/2 58

 Takeda NCT03370172 Mammalian ssFVIII-SQ AAV8 1/2

sc, self-com­ple­men­tary; ss, sin­gle stranded; UCL, University College London.

What are the short- and long-term rAAV safety
con­sid­er­ations?
In the 300-fold range of rAAV vec­ tor doses (2 × 1011 to 6 × 1013
vg/kg) thus far eval­u­ated in hemo­philia clin­i­cal tri­als, there have
been no major safety con­cerns.
In the short term, a small num­ber of sub­jects across 3 tri­als
at rAAV vec­tor doses rang­ing from 2 × 1012 to 6 × 1013 vg/kg have
expe­ri­enced acute vec­tor infu­sion reac­tions, includ­ing a sin­gle
inci­dence of ana­phy­laxis and a cou­ple of inci­dences of either
myal­gias, fever, and/or hypo­ten­sion of unclear eti­­ol­ogy that may
be con­sis­tent with an innate immune response to rAAV11-14; these
events have not clearly cor­re­lated with clin­i­cal out­comes post
vec­tor.
Because AAV effi­ciently tar­gets the liver, the bulk of safety con­
sid­er­ations of sys­temic rAAV deliv­ery have focused on hep­a­to­tox­
ic­ity. Within hemo­philia tri­als, a sin­gle study outlined multimonth
trans­am­i­nase ele­va­tions post vec­tor of unclear eti­­ol­ogy that were
not asso­ci­ated with liver dys­func­tion and resolved.11,15 Beyond
this, most trans­am­i­nase ele­va­tions observed in hemo­philia tri­als
occurred in the set­ting of an immune response to the rAAV cap­
sid.9,14,16-18 This AAV cap­sid immune response is hypoth­e­sized to
occur when CD8+ T cells rec­og­nize cap­sid pep­tides on the sur­
face of trans­duced hepa­to­cytes, trig­ger­ing a cel­lu­lar immune
response that results in clear­ance of the trans­duced cells.19 While
the AAV cap­sid immune response has not posed safety con­cerns
in hemo­philia, it has dimin­ished or prevented effi­cacy in some
tri­als.9,16-18 Close mon­i­tor­ing for a cap­sid immune response within
the first 3 months or so post vec­tor clas­si­cally includes fac­tor
assays, liver func­ tion stud­ ies, and periph­ eral blood mono­ nu­
clear cell inter­feron-γ anticapsid enzyme-linked immu­no­sor­bent
spot assays that, when interpreted as con­sis­tent with a cap­sid
immune response, trig­ger inter­ven­tion with immune-mod­u­lat­ing
ther­a­pies to main­tain transgene expres­sion; glu­co­cor­ti­coids have
been used pre­dom­i­nantly,9,16-18 but other immune-mod­u­lat­ing
Prakash Singh Shekhawat
agents have been used con­cur­rently with glu­co­cor­ti­coids or in
iso­la­tion to main­tain transgene expres­sion.10,13,14 Additional study
is needed to under­stand which reg­i­men best main­tains transgene
expres­sion. Further still, because some stud­ies have employed
glu­co­cor­ti­coids while simul­ta­neously reporting that par­tic­i­pants
had no cel­lu­lar immune response, addi­tional study is needed to
dis­tin­guish a cel­lu­lar immune response vs other poten­tial eti­­ol­o­
gies for ele­vated trans­am­i­nases.12,13,15
While safety in hemo­philia has been excel­lent, rAAV vec­tors
devel­oped for other mono­genic dis­or­ders sug­gest there may
be dose-lim­it­ing hepatic toxicities with sys­temic rAAV vec­tor
doses >1 × 1014 vg/kg. Specifically, Zolgensma (Novartis), an
rAAV9 vec­tor and the first licensed sys­temic rAAV vec­tor, deliv­
ered at a dose of 1.1 × 1014 vg/kg dem­on­strated not only remark­
able effi­cacy for spi­nal mus­cu­lar atro­phy but also evi­dence of
hepatic tox­ic­ity in clin­i­cal trial such that it was approved with a
boxed safety warn­ing for acute liver injury. Subsequently, a case
report outlined 2 chil­dren presenting 6 to 8 weeks post Zol-
gensma infu­sion with acute liver fail­ure that resolved with glu­
co­cor­ti­coid inter­ven­tion.20 While the exact eti­­ol­ogy is unclear,
the tim­ing post vec­tor, liver biopsy spec­i­mens dem­on­strat­ing
CD8+ T-cell infil­tra­tion, and respon­sive­ness to ste­roids sug­gest
an immune-medi­ated pro­cess that is pos­si­bly con­sis­tent with
an rAAV cap­sid immune response. These obser­va­tions sup­port
the pos­si­bil­ity that the rAAV cap­sid immune response has safety
impli­ca­tions and sup­port using the low­est pos­si­ble rAAV vec­tor
dose to min­i­mize safety con­cerns and max­i­mize effi­cacy. Addi-
tionally, a study using an rAAV8 vec­tor for X-linked myotubular
myop­a­thy reported that 3 of 17 boys in its 3.5 × 1014 vg/kg dose
cohort devel­oped liver fail­ure approx­i­ma­tely 6 weeks post vec­
tor and ulti­mately succumbed to com­pli­ca­tions.21 The gene ther­
apy com­mu­nity eagerly awaits a com­pos­ite anal­y­sis of these
obser­va­tions. Nonetheless, these tragic events sup­port pos­si­ble
dose-lim­it­ing toxicities of sys­temic rAAV vec­tors at doses >1 × 1014
Update on clin­i­cal hemo­philia gene ther­apy  |  227
vg/kg, which are >1.6- to 500-fold higher than the rAAV doses
used in hemo­philia.
Finally, dor­sal root gan­glia (DRG) tox­ic­ity has recently emerged
as a poten­tial short- or long-term safety con­cern. Overwhelm-
ingly, these obser­va­tions are his­to­log­i­cal find­ings in large-ani­mal
mod­els. Specifically, a meta-anal­y­sis of NHP data (n = 33 vec­tors,
n = 256 NHPs) by a sin­gle lab­o­ra­tory fol­low­ing het­ero­ge­neous
routes of admin­is­tra­tion (intra­the­cal, intracisternal magna, and
sys­temic), vec­tor doses, AAV cap­sids, and transgenes reported
no or mild his­to­log­i­cal evi­dence of DRG tox­ic­ity in most ani­mals
that appeared to be dose depen­dent and more com­mon with
direct cen­tral ner­vous sys­tem admin­is­tra­tion.22 Importantly, clin­
i­cal symp­tom­atol­ogy in only 2 ani­mals was attrib­uted to DRG
tox­ic­ity. Subsequent NHP stud­ies by the same group suggested
DRG tox­ic­ity may be related to an unfolded pro­tein response
abro­gated when inhibiting transgene expres­sion in the DRG with,
as yet, unclear clin­i­cal trans­la­tion.23 Thus far, in the full cadre of
clin­ic­ al AAV tri­als span­ning 3 decades, var­io ­ us routes of admin­
is­tra­tion for a vari­ety of dis­or­ders, and an approx­i­mate 1000-fold
range of rAAV vec­tor doses,24 no DRG tox­ic­ity clin­i­cal symp­toms
have been observed out­side of a sin­gle recently reported case.
A trial par­tic­i­pant who received an rAAVrh10 vec­tor (a cap­sid not
cur­rently used for hemo­philia) by intra­the­cal injec­tion to deliver
microRNA for amyotrophic lat­eral scle­ro­sis presented with symp­
toms and imag­ing con­sis­tent with DRG tox­ic­ity; symp­toms
dimin­ished with glu­co­cor­ti­coid and sup­port­ive care inter­ven­tion
but had not been com­pletely ame­lio­rated at the time of pub­
li­ca­tion.25 Whether DRG tox­ic­ity will be route of admin­is­tra­tion,
cap­sid, or rAAV vec­tor dose depen­dent and/or a poten­tial rAAV
vec­tor plat­form tox­ic­ity is unclear.
The major iden­ti­fied long-term safety con­cerns of sys­temic
rAAV vec­tors are the risks of hep­a­to­tox­ic­ity and genotoxicity.
Direct sequenc­ing data in ani­mals and humans dem­on­strate that
AAV inte­gra­tion can occur and has a pro­pen­sity for sites of active
tran­scrip­tion.26-30 Data in neo­na­tal mice dem­on­strated AAV inte­
gra­tion that disrupted the murine Rian locus (that does not have
a human ortholog) and resulted in near 100% pen­e­trance of hepa­
to­cel­lu­lar car­ci­noma (HCC).29 A sub­se­quent murine study dem­
on­strated that the risk of HCC cor­re­lated directly with the rAAV
vec­tor dose and degree of cel­lu­lar divi­sion and that hepa­to­cyte-
spe­ cific enhancer and pro­ moter ele­
ments protected against
onco­gen­e­sis, while non-hepa­to­cyte-spe­cific ele­ments increased
HCC risk.27 This would thus sup­port using the low­est pos­si­ble
effec­tive vec­tor dose with hepa­to­cyte-spe­cific pro­moter and
enhancer ele­ments to min­i­mize genotoxicity risk. Recently, stud­
ies in HA dogs followed for 10 years post rAAV vec­tor dem­on­
strated evi­dence of clonal pro­lif­er­a­tion at sites of AAV inte­gra­tion
in liver tis­sue with­out tumor­i­gen­e­sis but were the first large-ani­
mal data to sup­port the the­o­ret­i­cal risk of genotoxicity.30
Nearly all­severe hemo­philia patients >40 years of age con-
tracted iat­ ro­genic hep­ a­ti­
tis B virus and/or HCV, which are
known risk fac­tors for HCC such that the hemo­philia pop­ul­a­tion
has an increased risk of HCC rel­a­tive to the gen­eral pop­ul­a­
tion.31 Published 15-year fol­low-up data of the first par­tic­i­pants
to receive a sys­temic rAAV vec­tor (HB sub­jects) dem­on­strated
no evi­dence of long-term hepatoxicity, but the report is lim­ited
in inter­pre­ta­tion due to small cohort size and the absence of
detect­able transgene expres­sion or direct sequenc­ing data to
con­firm the per­sis­tence of trans­duced hepa­to­cytes.32 A sin­gle
clin­ic­ al trial par­tic­ip
­ ant in an HB trial with prior long-term HCV
228  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/226/1851420/226george.pdf by guest on 13 December 2021
Figure 1. Biochemical characterization of FIX-Padua (FIX-R338L).
Data support that FIX-R338L requires factor VIIIa (FVIIIa) cofactor
function for enzymatic activity to generate factor Xa (FXa) from
FX and is similarly activated for factor XIa (FIXa) and inactivated
by antithrombin (AT). Modified with permission from Samelson-
Jones et al.7
infec­tion and other risk fac­tors for HCC devel­oped HCC approx­
i­ma­tely 1 year fol­low­ing vec­tor infu­sion. Recently presented
tumor sequenc­ing data dem­on­strated the expected ran­dom,
low-fre­quency AAV inte­gra­tions with­out evi­dence of clonality
as well as well-described HCC-asso­ ci­
ated genetic changes,
suggesting that rAAV vec­tor admin­is­tra­tion did not con­trib­ute
to HCC devel­op­ment.33 There is thus far no clin­i­cal evi­dence to
sug­gest that sys­temic rAAV vec­tor deliv­ery is a risk fac­tor for
the devel­op­ment of HCC. Ultimately, cohorts out­side the adult
hemo­philia pop­ul­a­tion may more closely mimic the risk fac­tors
iden­ti­fied in mice for the devel­op­ment of HCC post sys­temic
rAAV vec­ tor infu­
sion. Specifically, infants and tod­dlers (who
have inher­ent greater degrees of cel­lu­lar divi­sion than adults)
who receive higher sys­temic rAAV vec­tor doses that con­tain
ubiq­ui­tous pro­mot­ers will be impor­tant pop­u­la­tions to fol­low
to eval­ua ­ te long-term AAV genotoxicity.
What are the transgene safety con­sid­er­ations?
The gen­ er­ ally iden­ti­
fied safety con­
cerns related to transgene
expres­sion include an immu­no­log­i­cal response to the expressed
pro­tein and direct transgene tox­ic­ity. Taking the lat­ter first, epi­
de­mi­o­log­i­cal stud­ies have outlined that supraphysiologic FIX:C
has a mod­er­ate, inde­pen­dent risk of venous throm­bo­sis (odds
ratio, 1.8-4.0) rel­a­tive to supraphysiologic FVIII:C (OR, 8.8-21.3).34
Biochemical data sup­port that FIX-R338L func­tion, like wild-type
FIX, requires FVIIIa cofac­tor func­tion for enzy­matic activ­ity and
is sim­i­
larly acti­ vated by FXIa and inactivated by anti­ throm­ bin
(Figure 1), supporting that FIX-R338L is reg­u­lated the same as wild-
type FIX7 and is not inher­ently not prothrombotic. Consistent with
this, observed throm­botic events in mice expressing w ­ t-FIX or
FIX-R338L cor­re­lated with the degree of supratherapeutic FIX:C,
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/226/1851420/226george.pdf by guest on 13 December 2021
Figure 2. Aggregated reported annualized bleeding rate data for hemophilia A and B clinical trials before (blue) and after (maroon)
receiving the described recombinant AAV vectors. The dotted line denotes an annualized bleeding rate of 1.AAV8-wt-FIX, also known
as scAAV2/8-LP1-hFIXco. Data from Nathwani et al.45 AMT-060 data from Miesbach et al.12 AMT-061, now etranocogene dezaparvovec,
data from Pipe et al.8 SPK-9001, now fidancogene elaparvovec, data from George et al.9 BMN270, now valoctocogene roxaparvovec,
data from Pasi et al.11 SPK-8011 data from George et al.14

inde­ pen­ dent of the expressed transgene.35 Concurrently, HB
gene ther­apy tri­als using the FIX-R338L transgene achieved FIX:C
in the range of mild or nor­mal HB with­out safety con­cerns. A sin­
gle FIX-R338L trial reported throm­bo­sis in a sub­ject who achieved
supraphysiologic FIX:C with mul­ti­ple prothrombotic comorbid-
ities (obe­sity, kid­ney fail­ure with an arte­rio­ve­nous fis­tula).10 This
obser­va­tion when paired with epi­de­mi­o­log­i­cal data of supraphys-
iologic fac­tor activ­ity and venous throm­bo­sis risk under­scores
the impor­tance of maintaining expres­sion within a ther­a­peu­tic
­win­dow. In addi­tion to throm­bo­sis risk, the expres­sion of FVIII or
FIX-R388L could impart direct cel­lu­lar tox­ic­ity, which is one pos­si­
ble cause of the unex­pected declin­ing FVIII expres­sion observed
in the first HA gene ther­apy trial detailed below.
While mul­ti­ple immu­no­log­i­cal responses to the expressed
transgene are pos­si­ble, the most concerning is the devel­op­ment
of inhib­i­tory antibodies, which have not been observed in clin­i­
cal trial. However, cur­rent trial enroll­ment includes patients who
are the least likely to develop an inhib­i­tor (eg, par­tic­i­pants with
prior or cur­rent inhib­i­tors are excluded and must have >50-150
Prakash Singh Shekhawat
fac­tor expo­sures). Nonetheless, inhib­i­tor risk post gene ther­apy
is gen­er­ally thought to be unlikely because of small- and large-
ani­mal data that dem­on­strate the abil­ity of hepa­to­cyte-directed
gene ther­apy to induce tol­er­ance to FVIII, FIX, or FIX-R338L.35-37
These ani­mal stud­ies show that future gene trans­fer may ulti­
mately be an effec­tive means for FVIII tol­er­ance induc­tion in
HA patients with inhib­i­tors; ini­ti­at­ing clin­i­cal tri­als for tol­er­ance
induc­tion was recently supported by con­sen­sus rec­om­men­da­
tions fol­low­ing a National Heart, Lung, and Blood Institute State
of the Science of FVIII inhib­i­tors.38
What are the pre­lim­i­nary effi­cacy data?
The approx­i­mate homo­ge­neous and remark­able phe­no­typic
ame­lio­ra­tion that has been observed with het­ero­ge­neous trans-
gene-derived fac­ tor activ­
ity supporting fac­ tor activ­
ity itself
should not be the sole distinguishing fea­ ture of clin­
i­
cal pro­
grams (Figure 2). Specifically, cur­rently avail­­able data are con­sis­
tent with HA nat­u­ral his­tory stud­ies that sug­gest fac­tor activ­ity
>10% ame­lio­rates spon­ta­ne­ous bleed­ing; how­ever, higher fac­tor
Update on clin­i­cal hemo­philia gene ther­apy  |  229

Figure 3. FVIII activity 1 and 2 years post vector infusion in trial participants who received a therapeutic vector dose (6 × 1013 vg/kg
of AAV5-hFVIII vs 5 × 1011 to 2 × 1012 vg/kg of SPK 8011) that maintained expression and were followed >2 years post vector. FVIII ac-
tivity is reported by 1-stage assay for SPK-8011 participants and CSA in participants who received AAV5-hFVIII. Data for AAV5-hFVIII
are from Pasi et al11 and data for SPK-8011 are from George et al.14
activ­ity is likely required among patients with marked preexist-
ing joint dis­ease. Most tri­als report an unpre­dict­able range of
expres­sion of 5- to 10-fold var­ia ­ bil­ity (some tri­als up to 50- to
100-fold) such that it is not cur­rently pos­si­ble to tar­get expres­
sion based on indi­vid­ual patient needs.9,10,11,13,14,39 Correspondingly,
an up to 7-fold var­i­abil­ity in trans­duc­tion was dem­on­strated in a
chi­me­ric murine model of human hepa­to­cytes.40 This degree of
var­i­able expres­sion in a clonal mouse pop­u­la­tion is con­sis­tent
with the mul­ ti­
ple var­
i­ables that exist from vec­ tor infu­
sion to
transgene expres­sion1 that may neces­si­tate tol­er­ance for a range
of transgene expres­sion, albeit hope­fully to a lesser degree than
has thus far been reported. This under­scores the need for an
accu­rate mea­sure­ment of in vivo activ­ity of transgene hemo­
static func­tion to define the range of tol­er­a­ble safe and effi­ca­
cious expres­sion.
In HA and HB efforts, there is a dis­crep­ancy in transgene-
derived fac­tor activ­ity by one-state assay (OSA) and chro­mo­genic
assay (CSA). Transgene-derived FVIII:C by OSA mea­sures approx­
i­ma­tely 1.6-fold higher than CSA in humans15,39-42 and mice,43 which
dif­fers from the expe­ri­ence with recom­bi­nant pro­tein prod­ucts
of the same amino acid sequence. Which assay cor­re­lates with
in vivo hemo­static ben­e­fit is unclear. Current tri­als tar­get exog­e­
nous FVIII expres­sion, and whether this mildly alters bio­chem­i­cal
prop­er­ties of the pro­tein is hypoth­e­sized (eg, changes in von Wil-
lebrand fac­tor affin­ity or FVIII cleav­age by throm­bin or fac­tor Xa)
but not deter­mined. Similarly, FIX-R338L deter­mined that FIX:C
mea­sures higher by OSA vs CSA FIX:C; how­ever, unlike HA, this
obser­va­tion is also maintained with recom­bi­nant FIX-R338L and
con­sis­tent with the avail­­able bio­chem­i­cal under­stand­ing of FIX-
R338L.44 Additionally, unlike HA tri­als, OSA deter­mined that FIX:C
from FIX-R338L expres­sion dif­fers by ini­ti­at­ing reagents.44
How long will the ther­apy last?
Thus far, clin­i­cal trial efforts in HB gene trans­fer mir­ror hemo­
philia canine mod­ els of rAAV-medi­ ated gene trans­ fer that
dem­on­strated no decline in FVIII or FIX expres­sion for up to
230  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/226/1851420/226george.pdf by guest on 13 December 2021
8 years post gene trans­fer.16,30,42,45,46 In con­trast, the first phase
1/2 HA gene ther­apy trial of valoctocogene roxaparvovec
at a dose of 6 × 1013 vg/kg dem­on­strated an aver­age loss of
approx­i­ma­tely 40% of transgene expres­sion from year 1 to
year 2 post vec­tor (n = 7)11,15; sim­i­lar obser­va­tions were observed
among the 17 phase 3 par­tic­i­pants administered the same vec­
tor dose who were followed for >2 years.13 Available phase
1/2 data of par­tic­i­pants 5 years post valoctocogene roxapar-
vovec dem­on­strated a con­tin­ued, but lesser, decline in FVIII
expres­sion in years 2 to 5.47 In con­trast, a phase 1/2 study of
SPK-8011 for HA at vec­tor doses 5 × 1011 to 2 × 1012 vg/kg dem­
on­strated dif­fer­ent phar­ma­co­ki­net­ics of expres­sion such that
the 12 par­tic­i­pants with sustained expres­sion out­side the cel­
lu­lar immune response and followed for >2 years dem­ on­
strated no appar­ent decline in expres­sion from year 1 to year 2
(Figure 3).14 Some of these sub­jects have been followed for up to
4 years and dem­on­strate gen­er­ally sta­ble FVIII expres­sion after
1 year post vec­tor; these pre­lim­i­nary SPK-8011 data gen­er­ally
sup­port the cur­rent strat­egy of targeting hepa­to­cytes for mul­
ti­year sta­ble FVIII expres­sion. Given that the transgene is main-
tained epi­som­ally, some degree of declin­ing expres­sion over
mul­ti­year fol­low-up may be antic­i­pated, although it is unclear
how rap­idly, but the mech­a­nism of loss or the more marked
decline in FVIII expres­sion from year 1 to year 2 observed with
valoctocogene roxaparvovec is unknown. Possible hypoth­ e­
ses include (1) an unfolded pro­tein response, which has been
described in mam­ma­lian expres­sion sys­tems of recom­bi­nant
FVIII pro­duc­tion and in mouse mod­els of supratherapeutic but
not low lev­els of FVIII expres­sion post rAAV48-50; (2) pro­moter
silenc­ing; (3) an ongo­ing unde­tected/unmit­i­gated immune
response to AAV or the transgene; or (4) fail­ure to form sta­ble
concatermized epi­somal DNA due to prop­er­ties of the vec­tor.
What do AAV neu­tral­iz­ing antibodies mean?
AAV neu­tral­iz­ing antibodies (NAb) are incor­po­rated into most
eli­gi­bil­ity cri­te­ria in hemo­philia and other dis­ease mod­els

Figure 4. Summary of known and unknown variables for recombinant AAV gene edition clinical efforts for in hemophilia A and B.
using sys­temic AAV vec­tors. While results of the assay have
an impact on eli­gi­bil­ity, the assays are not stan­dard­ized, can
yield highly var­i­able results depending on the assay meth­
ods (eg, trans­duc­tion vs enzyme-linked immu­no­sor­bent assay
based, employed multiplicity of infection, etc), and do not
always have a clear cor­re­la­tion with clin­i­cal out­come. None-
theless, most avail­­able data sup­port the fact that high-titer
AAV NAb pre­clude tar­get tis­sue trans­duc­tion of a sys­tem­i­
cally admin­is­tered rAAV vec­tor and thus limit effi­cacy. A sin­gle
AAV5 trial using a dose of 2 × 10 vg/kg is enroll­ing par­tic­i­pants
inde­pen­dent of AAV NAb sta­tus and has dem­on­strated trans-
gene expres­sion in all­ par­tic­i­pants except the par­tic­i­pant with
the highest mea­sured AAV NAb (>1:3000).8,13 These data chal­
lenge existing notions that would oth­er­wise pre­dict preex-
isting AAV NAb to pre­clude effi­cacy. Details of the AAV NAb,
thus far unre­ported, are nec­es­sary to inter­pret the data and
whether these data are spe­cific to rAAV5 and/or the vec­tor
dose used or may be par­tially explained by meth­ods used to
per­form the AAV NAb.
Following rAAV vec­tor admin­is­tra­tion, patients uni­ver­sally
develop per­ sis­
tent, multiserotype cross-reac­ tive AAV NAb
that avail­­able data sug­gest would pre­clude repeat rAAV vec­
tor effi­cacy.32,51 This under­scores that patients inter­ ested in
gene ther­apy must rec­og­nize that the cur­rent state of clin­i­cal
devel­op­ment likely only per­mits 1 sys­temic rAAV vec­tor infu­
sion. Thus, outlining the cur­rent known and unknown infor­ma­
tion regard­ing rAAV gene ther­apy for hemo­philia is essen­tial
to pro­vide true informed con­sent for both clin­i­cal tri­als and
antic­i­pated soon-to-be licensed vec­tors (Figure 4); this will be
nec­es­sary to avoid “buyer’s remorse” should one receive an
rAAV vec­tor that is ulti­mately found to be infe­rior to another
rAAV vec­tor.
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/226/1851420/226george.pdf by guest on 13 December 2021
What other approaches are in devel­op­ment?
Safely achiev­ing sustained, sta­ble, and pre­dict­able FVIII or FIX
expres­ sion, even in the pres­ ence of an immune response, is
an unre­al­ized goal of hemo­philia gene ther­apy. Beyond gene
addi­tion approaches, gene editing using lipid nanoparticles to
deliver mRNA encoding Cas9 and gRNA and a donor FIX cDNA
tem­plate via an rAAV vec­tor to knockin F9 into the albu­men
locus have dem­on­strated nor­mal lev­els of FIX expres­sion in an
NHP model.52 In addi­tion, lentiviral vec­tors for either sys­temic
infu­sion,53 ex vivo trans­duc­tion of hema­to­poi­etic stem cells,54-56
or induced plu­rip­o­tent stem cells are being pur­sued in pre­clin­
i­cal inves­ti­ga­tion and early-phase clin­i­cal trial (clinicaltrials.gov;
NCT03818763).57
Will hemo­philia gene ther­apy ever be main­stream?
Building on the remark­able successes the field has seen over the
past approx­i­ma­tely 5 years will be pred­i­cated on the thought­ful
inves­ti­ga­tion of ques­tions that have emerged from clin­i­cal tri­als.
Optimism in the field will con­tinue to prog­ress, buoyed by 3 gen­
eral strengths: (1) a strong under­stand­ing of the molec­u­lar and
bio­chem­i­cal basis of HA and HB, (2) an orga­nized patient advo­
cacy and phy­si­cian pro­vider net­work that fos­ters basic research
and clin­i­cal tri­als, and (3) a rap­idly expanding knowl­edge base in
gene ther­apy with sig­nif­i­cant finan­cial invest­ment. Collectively,
these strengths ensure that gene ther­apy will ful­fill its prom­ise to
alter the par­a­digm of hemo­philia care.
Acknowledgments 
This work was supported by NIH/NHLBI K08 HL 146991 (L. A.
George).
Dr. George thanks Ben­ja­min Samelson-Jones, MD/PhD, for his
thought­ful review of the arti­cle.
Update on clin­i­cal hemo­philia gene ther­apy  |  231
Conflict-of-inter­est dis­clo­sure
Lindsey A. George: sci­en­tific advi­sory board mem­ber: STRM.Bio;
data safety mon­i­tor­ing com­mit­tee mem­ber: Avrobio; con­sul­tancy:
Intellia, Biomarin, Pfizer, Bayer.
Off-label drug use
Lindsey A. George: glucocorticoids were discussed.
Correspondence
Lindsey A. George, University of Pennsylvania School of Med-
icine, Colket Translational Research Bldg, Rm 5016, 3501 Civic
Center Blvd, Philadelphia, PA 19104; e-mail: georgel@email​­.chop​
­.edu.
References
1. Li C, Samulski RJ. Engineering adeno-asso­ci­ated virus vec­tors for gene
ther­apy. Nat Rev Genet. 2020;21(4):255-272.
2. Shahani T, Covens K, Lavend’homme R, et al. Human liver sinu­soi­dal endo­
the­lial cells but not hepa­to­cytes con­tain fac­tor VIII. J Thromb Haemost.
2014;12(1):36-42.
3. Pittman DD, Alderman EM, Tomkinson KN, Wang JH, Giles AR, Kaufman
RJ. Biochemical, immu­no­log­i­cal, and in vivo func­tional char­ac­ter­iza­tion of
B-domain-deleted fac­tor VIII. Blood. 1993;81(11):2925-2935.
4. McIn­tosh J, Lenting PJ, Rosales C, et al. Therapeutic lev­els of FVIII fol­low­ing
a sin­gle periph­eral vein admin­is­tra­tion of rAAV vec­tor encoding a novel
human fac­tor VIII var­i­ant. Blood. 2013;121(17):3335-3344.
5. Nathwani AC, Tuddenham E, Chowdary P, McIn­tosh J, et al. GO-8: pre­lim­
i­nary results of a phase I/II dose esca­la­tion trial of gene ther­apy for hae-
mophilia A using a novel human fac­tor VIII var­i­ant. Blood. 2018;132(suppl 1):
489.
6. Simioni P, Tormene D, Tognin G, et al. X-linked thrombophilia with a mutant
fac­tor IX (fac­tor IX Padua). N Engl J Med. 2009;361(17):1671-1675.
7. Samelson-Jones BJ, Finn JD, George LA, Camire RM, Arruda VR. Hyperactiv-
ity of fac­tor IX Padua (R338L) depends on fac­tor VIIIa cofac­tor activ­ity. JCI
Insight. 2019;4(14):e128683.
8. Pipe SW, Recht M, Key NS, et al. HOPE-B inves­ti­ga­tors 52 week effi­cacy and
safety of etranacogene dezaparvovec in adults with severe or mod­er­ate-
severe hemo­philia B: data from the phase 3 HOPE-B gene ther­apy trial
[ab­stract]. Res Pract Thromb Haemost. 2021;5(suppl 2).
9. George LA, Sullivan SK, Giermasz A, et al. Hemophilia B gene ther­apy with a
high-spe­cific-activ­ity fac­tor IX var­i­ant. N Engl J Med. 2017;377(23):2215-2227.
10. Chowdary P, Makris M, Evans G, et  al. A novel adeno asso­ci­ated virus
(AAV) gene ther­apy (FLT180a) achieves nor­mal FIX activ­ity lev­els in severe
hemo­philia B (HB) patients (B-AMAZE study). Res Pract Thromb Haemost.
2020;4(suppl 2):17.
11. Pasi KJ, Rangarajan S, Mitchell N, et al. Multiyear fol­low-up of AAV5-hFVIII-
SQ gene ther­apy for hemo­philia A. N Engl J Med. 2020;382(1):29-40.
12. Miesbach W, Meijer K, Coppens M, et  al. Gene ther­apy with adeno-
asso­ci­ated virus vec­tor 5-human fac­tor IX in adults with hemo­philia B.
Blood. 2018;131(9):1022-1031.
13. Ozelo M, Mahlangu J, Pasi K, Giermasz A, et al; GENEr8-1 Trial Group. Efficacy
and safety of valoctocogene roxaparvovec adeno-asso­ci­ated virus gene
trans­fer for severe hemo­philia A: results from the phase 3 GENEr8-1 trial
[ab­stract]. Res Pract Thromb Haemost. 2021;5(suppl 1). Abstract OC 26.1.
14. George LA, Monahan PE, Eyster ME, Sullivan SK, et al. Phase I/II trial of SPK-
8011: sta­ble and dura­ble FVIII expres­sion after AAV gene trans­fer for hemo­
philia A [ab­stract]. Res Pract Thromb Haemost. 2021;5(suppl 1). Abstract
OC 67.2.
15. Rangarajan S, Walsh L, Lester W, et  al. AAV5-fac­tor VIII gene trans­fer in
severe hemo­philia A. N Engl J Med. 2017;377(26):2519-2530.
16. Nathwani AC, Tuddenham EG, Rangarajan S, et al. Adenovirus-asso­ci­ated
virus vec­tor-medi­ated gene trans­fer in hemo­philia B. N Engl J Med. 2011;
365(25):2357-2365.
17. Manno CS, Pierce GF, Arruda VR, et al. Successful trans­duc­tion of liver in
hemo­philia by AAV-fac­tor IX and lim­i­ta­tions imposed by the host immune
response. Nat Med. 2006;12(3):342-347.
18. Konkle BA, Walsh CE, Escobar MA, et al. BAX 335 hemo­philia B gene ther­
apy clin­ i­
cal trial results: poten­ tial impact of CpG sequences on gene
expres­sion. Blood. 2021;137(6):763-774.
232  |  Hematology 2021  |  ASH Education Program
19. Mingozzi F, Maus MV, Hui DJ, et  al. CD8(+) T-cell responses to adeno-
asso­ci­ated virus cap­sid in humans. Nat Med. 2007;13(4):419-422.
20. Feldman AG, Parsons JA, Dutmer CM, et al. Subacute liver fail­ure fol­low­ing
gene replace­ment ther­apy for spi­nal mus­cu­lar atro­phy type 1. J Pediatr.
2020;225(Octo­ber):252-258.e1258e1.
21. Shieh PB, Bönnemann CG, Müller-Felber W, et  al. Re: “Moving Forward
After Two Deaths in a Gene Therapy Trial of Myotubular Myopathy” by Wil-
son and Flotte. Hum Gene Ther. 2020;31(15-16):787.
22. Hordeaux J, Buza EL, Dyer C, et al. Adeno-asso­ci­ated virus-induced dor­sal
root gan­glion pathol­ogy. Hum Gene Ther. 2020;31(15-16):808-818.
23. Hordeaux J, Buza EL, Jeffrey B, et al. MicroRNA-medi­ated inhi­bi­tion of trans-
gene expres­sion reduces dor­sal root gan­glion tox­ic­ity by AAV vec­tors in
pri­ma­tes. Sci Transl Med. 2020;12(569):eaba9188.
24. Wang D, Tai PWL, Gao G. Adeno-asso­ci­ated virus vec­tor as a plat­form for
gene ther­apy deliv­ery. Nat Rev Drug Discov. 2019;18(5):358-378.
25. Mueller C, Berry JD, McKenna-Yasek DM, et  al. SOD1 sup­ pres­sion with
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/226/1851420/226george.pdf by guest on 13 December 2021
adeno-asso­ ci­ated virus and microRNA in famil­ ial ALS. N Engl J Med.
2020;383(2):151-158.
26. Nault JC, Datta S, Imbeaud S, et al. Recurrent AAV2-related inser­tional muta­
gen­e­sis in human hepa­to­cel­lu­lar car­ci­no­mas. Nat Genet. 2015;47(10):1187-
1193.
27. Chandler RJ, LaFave MC, Varshney GK, et  al. Vector design influ­ences
hepatic genotoxicity after adeno-asso­ ci­ated virus gene ther­ apy. J Clin
Invest. 2015;125(2):870-880.
28. Nakai H, Montini E, Fuess S, Storm TA, Grompe M, Kay MA. AAV sero­type
2 vec­tors pref­er­en­tially inte­grate into active genes in mice. Nat Genet.
2003;34(3):297-302.
29. Donsante A, Miller DG, Li Y, et  al. AAV vec­tor inte­gra­tion sites in mouse
hepa­to­cel­lu­lar car­ci­noma. Science. 2007;317(5837):477.
30. Nguyen GN, Everett JK, Kafle S, et al. A long-term study of AAV gene ther­
apy in dogs with hemo­philia A identifies clonal expan­sions of trans­duced
liver cells. Nat Biotechnol. 2021;39(1):47-55.
31. Mazepa MA, Monahan PE, Baker JR, Riske BK, Soucie JM; US Hemophilia Treat-
ment Center Network. Men with severe hemo­philia in the United States:
birth cohort anal­ y­sis of a large national data­ base. Blood. 2016;127(24):
3073-3081.
32. George LA, Ragni MV, Rasko JEJ, et  al. Long-term fol­low-up of the first
in human intra­vas­cu­lar deliv­ery of AAV for gene trans­fer: AAV2-hFIX16 for
severe hemo­philia B. Mol Ther. 2020;28(9):2073-2082.
33. Schmidt M, Foster GM, Coppens M, Thomsen H, et  al. Liver safety case
report from the phase 3 HOPE-B gene ther­apy trial in adults with hemo­philia
B [ab­stract]. Res Pract Thromb Haemost. 2021;5(suppl 1). Abstract OC 67.4.
34. Rietveld IM, Lijfering WM, le Cessie S, et al. High lev­els of coag­u­la­tion fac­
tors and venous throm­bo­sis risk: stron­gest asso­ci­a­tion for fac­tor VIII and
von Willebrand fac­tor. J Thromb Haemost. 2019;17(1):99-109.
35. Crudele JM, Finn JD, Siner JI, et al. AAV liver expres­sion of FIX-Padua pre­
vents and erad­i­cates FIX inhib­i­tor with­out increas­ing thrombogenicity in
hemo­philia B dogs and mice. Blood. 2015;125(10):1553-1561.
36. Finn JD, Ozelo MC, Sabatino DE, et al. Eradication of neu­tral­iz­ing antibod-
ies to fac­tor VIII in canine hemo­philia A after liver gene ther­apy. Blood.
2010;116(26):5842-5848.
37. Mingozzi F, Hasbrouck NC, Basner-Tschakarjan E, et al. Modulation of tol­er­
ance to the transgene prod­uct in a non­hu­man pri­mate model of AAV-medi­
ated gene trans­fer to liver. Blood. 2007;110(7):2334-2341.
38. Ragni MV, George LA; for Members of Working Group 1, the National Heart,
Lung, and Blood Institute State of the Science Workshop on Factor VIII Inhib-
itors: Generating a National Blueprint for Future Research. The national blue­
print for future fac­tor VIII inhib­i­tor clin­i­cal tri­als: NHLBI State of the Science
(SOS) Workshop on fac­tor VIII inhib­i­tors. Haemophilia. 2019;25(4):581-589.
39. Konkle BA, Stine K, Visweshwar N, Harrington TJ, et al. Updated fol­low-up
of the Alta study, a phase 1/2, open-label, adap­tive, dose-rang­ing study to
assess the safety and tol­er­ab ­ il­ity of SB-525 gene ther­apy. Blood. 2019;
134(suppl 1):2060.
40. Zou C, Vercauteren KOA, Michailidis E, et  al. Experimental var­i­ables that
affect human hepa­to­cyte AAV trans­duc­tion in liver chi­me­ric mice. Mol Ther
Methods Clin Dev. 2020;18(Sep­tem­ber):189-198.
41. Rosen S, Tiefenbacher S, Robinson M, et al. Activity of transgene-pro­duced
B-domain-deleted fac­tor VIII in human plasma fol­low­ing AAV5 gene ther­
apy. Blood. 2020;136(22):2524-2534.
42. Nathwani AC, Tuddenham E, Chowdary P, et al. Adeno-asso­ci­ated medi­
ated gene trans­fer for hemo­philia B: 8 year fol­low up and impact of remov­
ing” empty viral par­ti­cles” on safety and effi­cacy of gene trans­fer. Blood.
2019;132(suppl 1):491.
43. Sternberg AR, Davidson RJ, Samelson-Jones BJ, George L. In vitro and in
vivo mod­els to under­stand one-stage and chro­mo­genic fac­tor VIII activ­ity
assay dis­crep­ancy of hepa­to­cyte-derived fac­tor VIII [ab­stract]. Res Pract
Thromb Haemost. 2021;5(suppl 1). Abstract OC 75.3.
44. Robinson MM, George LA, Carr ME, et al. Factor IX assay discrepancies in
the set­ting of liver gene ther­apy using a hyper­func­tional var­i­ant fac­tor IX-
Padua. J Thromb Haemost. 2021;19(5):1212-1218.
45. Nathwani AC, Reiss UM, Tuddenham EG, et al. Long-term safety and effi­cacy
of fac­tor IX gene ther­apy in hemo­philia B. N Engl J Med. 2014;371(21):1994-
2004.
46. Niemeyer GP, Herzog RW, Mount J, et al. Long-term cor­rec­tion of inhib­i­
tor-prone hemo­philia B dogs treated with liver-directed AAV2-medi­ated
fac­tor IX gene ther­apy. Blood. 2009;113(4):797-806.
47. Pasi KR, Rangarajan S, TM Robinson, Lester W, et al. Hemostatic response
is maintained for up to 5 years fol­low­ing treat­ment with valoctocogene
roxaparvovec, an AAV5-hFVIII-SQ gene ther­apy for severe hemo­philia A
[ab­stract]. Res Pract Thromb Haemost. 2021;5(suppl 1). Abstract OC 67.1.
48. Lange AM, Altynova ES, Nguyen GN, Sabatino DE. Overexpression of fac­
tor VIII after AAV deliv­ery is tran­siently asso­ci­ated with cel­lu­lar stress
in hemo­philia A mice. Mol Ther Methods Clin Dev. 2016;3(28 Sep­tem­
ber):16064.
49. Zolotukhin I, Markusic DM, Palaschak B, Hoffman BE, Srikanthan MA,
Herzog RW. Potential for cel­lu­lar stress response to hepatic fac­tor VIII
expres­sion from AAV vec­tor. Mol Ther Methods Clin Dev. 2016;3(28 Sep­
tem­ber):16063.
50. Poothong J, Pottekat A, Siirin M, et al. Factor VIII exhib­its chap­er­one-depen­
dent and glu­cose-reg­u­lated revers­ible amy­loid for­ma­tion in the endo­plas­
mic retic­u­lum. Blood. 2020;135(21):1899-1911.
Prakash Singh Shekhawat
51. Long BR, Veron P, Kuranda K, et al. Early phase clin­i­cal immu­no­ge­nic­ity of
valoctocogene roxaparvovec, an AAV5-medi­ated gene ther­apy for hemo­
philia A. Mol Ther. 2021;29(2):597-610.
52. Huang H MEC, Bialek P, Wang C, et  al. CRISPR/Cas9-medi­ated targeted
inser­tion of human F9 achieves ther­a­peu­tic cir­cu­lat­ing pro­tein lev­els in mice
and non-human pri­ma­tes [ab­stract]. Mol Ther. 2019;27(April):7. Abstract 11.
53. Cantore A, Ranzani M, Bartholomae CC, et al. Liver-directed lentiviral gene
ther­apy in a dog model of hemo­philia B. Sci Transl Med. 2015;7(277):277ra28.
54. Du LM, Nurden P, Nurden AT, et  al. Platelet-targeted gene ther­apy with
human fac­tor VIII establishes haemostasis in dogs with haemophilia A. Nat
Commun. 2013;4(Novem­ber):2773.
55. Shi Q , Wilcox DA, Fahs SA, et al. Factor VIII ectop­i­cally targeted to plate­
lets is ther­a­peu­tic in hemo­philia A with high-titer inhib­i­tory antibodies. J
Clin Invest. 2006;116(7):1974-1982.
56. Greene TK, Wang C, Hirsch JD, et al. In vivo effi­cacy of plate­let-deliv­ered,
high spe­cific activ­ity fac­tor VIII var­i­ants. Blood. 2010;116(26):6114-6122.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/226/1851420/226george.pdf by guest on 13 December 2021
57. Follenzi A, Benten D, Novikoff P, Faulkner L, Raut S, Gupta S. Transplanted
endo­the­lial cells repopulate the liver endo­the­lium and cor­rect the phe­no­
type of hemo­philia A mice. J Clin Invest. 2008;118(3):935-945.
58. Pipe S, Becka M, Detering E, Vanevski K, Lissitchkov T. First-in-human gene
ther­apy study of AAVhu37 cap­sid vec­tor tech­nol­ogy in severe hemo­philia
A. Blood. 2019;134(suppl 1):4630.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000254
Update on clin­i­cal hemo­philia gene ther­apy  |  233
 HODGKIN LYMPHOMA: CURE AND OPTIMAL SURVIVORSHIP

Controversies in the management of early-stage

Hodgkin lymphoma

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/234/1851889/234blum.pdf by guest on 13 December 2021

Kristie A. Blum
Department of Hematology and Medical Oncology, Emory Winship Cancer Institute, Atlanta, GA

Positron emission tomography (PET)–adapted chemotherapy and radiotherapy approaches are currently used for the ini-
tial treatment of early-stage Hodgkin lymphoma (HL) with progression-free survival and overall survival exceeding 85%
and 95%, respectively. However, despite general agreement on the prognostic value of interim PET in HL, frontline treat-
ment approaches vary among institutions with respect to how pretreatment clinical risk factors determine treatment
selection, the definition of PET negativity, which chemotherapy regimen to initiate and how many cycles to administer,
and when to incorporate radiation. Furthermore, as recent trials have confirmed improved efficacy and manageable
toxicity when brentuximab and checkpoint inhibitors are combined with frontline regimens such as doxorubicin, vinblas-
tine, and dacarbazine in advanced-stage HL, these agents are now under evaluation as frontline therapy in early-stage
HL. A number of issues will affect the use of these agents in early-stage HL, including the costs, early and late toxicities
with these agents, patient population (favorable or unfavorable risk groups), how to incorporate them (concurrently or
sequentially), and whether they can ultimately replace cytotoxic therapy with similar efficacy and fewer late effects.
Future treatment paradigms for early-stage HL may change significantly once randomized studies are completed incor-
porating these agents into frontline therapy. Ideally, frontline use of brentuximab and checkpoint inhibitors in early-stage
HL will result in improved outcomes compared with current PET-adapted approaches with decreased risks of late toxici-
ties that continue to afflict long-term survivors of HL.

LEARNING OBJECTIVES
• Describe current frontline treatment approaches in patients with early-stage HL
• Describe recent clinical trials and controversies surrounding the incorporation of brentuximab and checkpoint inhib-
itors into frontline therapy in early-stage HL

white blood cell count of 12,400/microliter, hemoglobin of

CLINICAL CASE
A 20-year-old woman has bilateral cervical lymph node
enlargement but no fevers, night sweats, or weight loss.
She was initially treated with antibiotics and steroids by her
primary care physician without improvement. Core needle
biopsy specimen of a right cervical lymph node was suspi-
cious but not diagnostic for classic Hodgkin lymphoma (HL).
Excisional lymph node biopsy confirmed classic HL, and
positron emission tomography (PET) scan demonstrated
numerous cervical, supraclavicular, mediastinal, and axillary
nodes, including a left posterior cervical node of 1.2×1cm
with standardized uptake value (SUV) 7.3, a right supracla-
vicular node measuring 2.7×3.5cm with SUV 10.6, a medias-
tinal node measuring 2.3×1.3cm with SUV 5.5, and right and
left axillary lymphadenopathy. No intra-abdominal or splenic
uptake was noted. Her laboratory testing demonstrated a
11.9g/dL, platelet count of 471 000, and an elevated eryth-
rocyte sedimentation rate (ESR) of 77mm/h. Therefore, she
has nonbulky stage IIA classic HL, with adverse features
including involvement of 5 nodal sites and an elevated ESR.
Introduction
A number of controversies surround treatment decisions
in patients diagnosed with early-stage HL, and institutions
vary in their approaches to these patients. Clinical trials that
guide treatment decisions in early-stage HL differ in the
designation of pretreatment risk factors, eligibility criteria,
definition of response based on interim PET, and how novel
agents are incorporated. In addition, long-term follow-up
is lacking from many of these trials, and recent studies con-
tinue to show that patients with classic HL remain at risk

234 | Hematology 2021 | ASH Education Program

for late com­pli­ca­tions from ther­apy despite improve­ments and
reduc­tions in ther­apy. Therefore, cur­rent treat­ment approaches
need to bal­ance the com­pet­ing risks of achiev­ing high clin­i­cal
effi­cacy while min­i­miz­ing late tox­ic­ity. This review sum­ma­rizes
recent data regard­ing late toxicities in patients with early-stage
HL, PET-adapted ther­a­peu­tic approaches for these patients, the
treat­ment of bulky HL, and the incor­po­ra­tion of brentuximab
and check­point inhib­i­tors into the treat­ment of early-stage HL.
In addi­tion, the review high­lights ongo­ing ques­tions in the field,
includ­ing which che­mo­ther­apy back­bone to ini­ti­ate, the num­ber
of treat­ment cycles to admin­is­ter, def­i­ni­tions of PET neg­a­tiv­ity,
when to incor­po­rate radio­ther­apy, how to treat high-risk pa-
tients who are interim or end-of-ther­apy PET pos­i­tive, treat­ment
of bulky HL, and how to include brentuximab and check­point
inhib­i­tors in front­line ther­apy.
Late toxicities of treat­ment of clas­sic HL
in the mod­ern era
Despite advances in front­line che­mo­ther­apy and radio­ther­apy
for patients with early-stage HL, these patients still expe­ri­ence
a life­long risk of sec­ond­ary malig­nan­cies and car­dio­vas­cu­lar
dis­ease. In an anal­y­sis of Surveillance, Epidemiology, and End
Results data on 20,007 sur­vi­vors of HL aged 20 to 74 years diag­
nosed from 2000 to 2015, Dores et al1 ascertained that the risk of
death due to noncancer causes and sec­ond neo­plasms remains
sig­nif­i­cantly ele­vated over the gen­eral US pop­u­la­tion, even in
the mod­ern treat­ment era. In this study, 60% of patients had
stage I/II dis­ease, and median fol­low-up was 8 years. Noncancer
causes of mor­tal­ity in early-stage patients included inter­sti­tial
lung dis­ease, infec­tion, benign hema­to­logic dis­ease (cytopenias
and clot­ ting), heart dis­ease, and dia­be­tes.1 Other neo­ plasms
accounted for 25% of all­nonlymphoma deaths regard­ less of
stage, with 145 cases of sec­ond­ary neo­plasms in early-stage and
144 cases in advanced-stage patients. Stage-spe­cific mor­tal­ity
trends are declin­ing in the mod­ern treat­ment era with all­-cause,
noncancer, and other neo­plasm stan­dard­ized mor­tal­ity ratios of
7.0, 2.9, and 2.9 for early-stage patients treated from 1983 to 1991
vs 2.9, 1.5, and 1.6 treated from 2001 to 2009, respec­tively. How-
ever, fol­low-up is short for those patients treated from 2001 to
2009, and these mor­tal­ity rates still exceed rates in the gen­eral
pop­u­la­tion.
Focusing on sec­ond­ary neo­plasms, the Children’s Oncology
Group published a 10-year fol­low-up on AHOD0031, a trial of 1711
patients aged up to 21 years treated with response-adapted
ther­apy.2 All patients ini­tially received doxo­ru­bi­cin, bleomycin,
vin­cris­tine, etoposide, pred­ni­sone, and cyclo­phos­pha­mide, and
rapid early respond­ers (deter­mined by PET) were ran­domly allo­
cated to 21 Gy involved field radio­ther­apy (IFRT) or obser­va­tion.
Slow respond­ers were ran­dom­ized to IFRT or dexa­meth­a­sone,
etoposide, cis­platin, and cytarabine plus IFRT. Ten-year event-
free sur­vival and over­all sur­vival (OS) were not sig­nif­i­cantly dif­
fer­ent in the rapid early respond­ers treated with obser­va­tion
vs IFRT and also not dif­fer­ent in slow respond­ers treated with
dexa­meth­a­sone, etoposide, cis­platin, and cytarabine plus IFRT
vs IFRT alone. The cumu­la­tive inci­dence of sec­ond neo­plasms
was 1.3% with 17 sec­ond malig­nan­cies (3 cases of acute mye­
loid leu­ke­mia, 11 solid tumors, and 3 cases of NHL). Sixteen of
these malig­nan­cies occurred in patients treated with com­bined
modal­ity ther­apy (CMT) with doxo­ru­bi­cin, bleomycin, vin­cris­tine,
etoposide, pred­ni­sone, and cyclo­phos­pha­mide plus 21 Gy IFRT,
Prakash Singh Shekhawat
and 9 of 11 solid tumors occurred within the radi­a­tion field. Most
solid tumors were pap­il­lary thy­roid car­ci­noma, but there was
1 breast can­cer that did not develop until 13 years after study
enroll­ment. Therefore, it is likely with lon­ger fol­low-up that addi­
tional sec­ond­ary neo­plasms, includ­ing breast and lung can­cers,
will be observed despite a PET-adapted approach that lim­its
both che­mo­ther­apy and radi­a­tion doses.
In 3905 Dutch patients aged 15 to 50 years who were treated
between 1965 and 2000, with 60.5% receiv­ing CMT, the cumu­
la­tive inci­dence of sub­se­quent solid neo­plasms did not dif­fer
sig­nif­i­cantly (P = .71) among treat­ment eras: 1965 to 1976, 1977 to
1988, and 1989 to 2000.3 In addi­tion, the risk for any sec­ond can­
cer remained high for up to 40 years after treat­ment for HL, with
a 48.5% cumu­la­tive inci­dence. Bright et al4 also showed that the
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/234/1851889/234blum.pdf by guest on 13 December 2021
risk of solid tumors con­tin­ues to increase annu­ally, even up to
35 years after treat­ment for HL. In their study of 16 971 sur­vi­vors
of HL aged 15 to 39 years who were treated from 1971 to 2006,
the cumu­la­tive inci­dence of sub­se­quent pri­mary neo­plasms was
0.9% and 26.6% in females and 0.6% and 16.5% in males at 10 and
35 years from diag­no­sis. Taken together, these stud­ies all­dem­
on­strate that the risk of sec­ond neo­plasms con­tin­ues to rise with
long-term fol­low-up and that most of these sec­ond solid tumors
are not observed until after at least 10 years from diag­no­sis. In
addi­tion, the risk of sec­ond malig­nant neo­plasms con­tin­ues to
per­sist in the cur­rent treat­ment era despite improved che­mo­
ther­apy and reduced radi­a­tion expo­sure.
PET-adapted front­line treat­ment
As a result of the persisting evi­dence of late com­pli­ca­tions even
with mod­ern che­mo­ther­apy and radi­a­tion tech­niques and with
the proven prog­nos­tic sig­nif­i­cance of interim PET in early-stage
HL,5 many ado­les­cent and young adult (AYA) and adult patients
with early-stage HL cur­rently receive PET-adapted ther­apy in
an effort to fur­ther reduce che­mo­ther­apy and radi­a­tion expo­
sure. The stud­ies supporting PET-adapted ther­apy dif­fer slightly
in eli­gi­bil­ity cri­te­ria, in the treat­ment reg­i­men, and in defin­ing
PET neg­a­tiv­ity, mak­ing com­par­i­sons across stud­ies chal­leng­ing.
Table 1 sum­ma­rizes these stud­ies that cur­rently guide treat­ment
in AYA and adult HL.
In the UK RAPID study,7 602 nonbulky stage I to IIA patients
(32.3% unfa­vor­able by German Hodgkin Study Group cri­te­ria)
received 3 cycles of doxorubicin, bleomycin, vinblastine, and
dacarbazine followed by a PET scan. Patients with a Deauville
score of 1 to 2 were con­ sid­ered PET neg­ a­
tive and ran­
dom­
ized to either 30 Gy IFRT or no fur­ther treat­ment, and patients
with a Deauville score of 3 to 5 under­went a fourth cycle of
ABVD and IFRT. Seventy-four per­cent of patients were PET
­neg­a­tive, and 3-year pro­gres­sion-free sur­vival (PFS) was 94.6%
with IFRT and 90.8% with obser­va­tion (P = .02). No OS ben­e­fit
was observed, with a 3-year OS of 97.1% com­pared with 99%
with and with­out radi­a­tion (P = .27). For the PET-pos­i­tive patients,
3-year PFS was 83%. In an anal­y­sis of the RAPID study by Deau-
ville score,7 infe­rior out­comes were seen only in patients with
scores of 5, with 5-year PFS of 91.5%, 91.1%, 95.3%, 87.5%, and
61.9% in patients with scores of 1, 2, 3, 4, and 5, respec­tively.
Cancer and Leukemia Group B 50604 exam­ ined a PET-
adapted approach in patients aged 18 to 60 years with nonbulky
stage I and II HL with and with­out B-symp­toms.8 All patients
received 2 cycles of ABVD, interim PET, and, if PET neg­a­ tive
(Deauville scores 1-3), 2 addi­tional ABVD cycles with­out IFRT.
Management of early-stage HL  | 235
 Table 1. PET-adapted ther­a­peu­tic tri­als in early-stage HL
Study (median fol­low-up)
6
UK RAPID (5.0 years)
CALGB 506048 (3.8 years)
EORTC9 (4.5-5.1 years)
236  |  Hematology 2021  |  ASH Education Program
GHSG HD1610 (45 months)
GHSG HD1711 (46.2 months)
eBEACOPP, esca­lated bleomycin, etoposide, doxo­ru­bi­cin, cyclo­phos­pha­mide, vin­cris­tine, pro­car­ba­zine and pred­ni­sone; EN, XXX; F, favor­able; FFTF, free­dom from treat­ment fail­ure; neg, neg­a­
tive; pos, pos­i­tive; U, unfa­vor­able.
Patient pop­u­la­tion (N) Risk fac­tors at enroll­ment PET neg­a­tive Treatment arms PFS or FFTF OS
Stage IA or IIA (n = 602) Nonbulky Deauville 1-2 PET neg: ABVD × 3 3-year 90.8% 3-year 99.0%
PET neg: ABVD × 3 + 30 Gy IFRT 3-year 94.6% 3-year 97.1%
PET pos: ABVD × 4 + 30 Gy IFRT 3-year 83%
Stage I or II (n = 164) Nonbulky Deauville 1-3 PET neg: ABVD × 4 3-year PFS 91%
PET pos: ABVD × 2, eBEACOPP × 2, 3-year PFS 66%
30 Gy IFRT
Stage I or II (n = 1950) Favorable (no risk fac­tors) Deauville 1-2 (F) PET neg: ABVD × 3 + 30 Gy INRT 5-year PFS 99% 5-year OS 100%
Unfavorable (any of the fol­low­ing) (F) PET neg: ABVD × 4 5-year PFS 87.1% 5-year OS 99.6%
1. Age ≥50
(U) PET neg: ABVD × 4 + 30 Gy INRT 5-year PFS 92.1% 5-year OS 96.7%
2. Bulk
(U) PET neg: ABVD × 6 5-year 89.6% 5-year OS 98.3%
3. >3 nodal sites
4. ESR ≥50 (or 30 if B-symp­toms) (F/U) PET pos: ABVD × 4 + 30 Gy INRT 5-year PFS 77.4% 5 year OS 89.3%
(F/U) PET pos: ABVD × 2, eBEACOPP × 2, 5-year PFS 90.6% 5 year OS 96.0%
30 Gy INRT
Stage I or II (N = 1150) Favorable (none of the fol­low­ing Deauville 1-2 PET neg: ABVD × 2 + 20 Gy IFRT 5-year PFS 93.4% 5-year OS 98.1%
risk fac­tors): PET neg: ABVD × 2 5-year PFS 86.1% 5-year OS 98.4%
1. Bulk PET pos: ABVD × 2 + 20 Gy IFRT 5-year PFS 88.4% 5-year OS 97.9%
2. Extranodal sites
3. >2 nodal areas
4. ESR ≥50 (or 30 if B-symp­toms)
Stage I or II (N = 1100) Unfavorable (with one of the Deauville 1-2 PET neg­a­tive: eBEACOPP/ABVD × 4 +  5-year PFS 97.7% 5-year OS 98.7%
fol­low­ing risk fac­tors): 30 Gy IFRT 5-year PFS 95.9% 5-year OS 98.8%
1. Bulk PET neg: eBEACOPP/ABVD × 4 5-year PFS 94.2% 5-year OS 99.2%
2. Extranodal sites PET pos: eBEACOPP/ABVD × 4 + 30 Gy
3. >2 nodal areas IFRT
4. ESR ≥50 (or 30 if B-symp­toms)
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/234/1851889/234blum.pdf by guest on 13 December 2021
PET-pos­i­tive patients switched to esca­lated bleomycin, etopo-
side, doxo­ru­bi­cin, cyclo­phos­pha­mide, vin­cris­tine, pro­car­ba­zine, the asso­ci­a­tion of PFS by Deauville score from the RAPID,11 HD16,9
and pred­ni­sone (BEACOPP) with IFRT. In total, 164 patients were
enrolled, 26% with B-symp­toms and 58% unfa­vor­able by GHSG
cri­te­ria. With the expan­sion of PET neg­a­tiv­ity to Deauville scores
of 1 to 3, 91% of patients were PET neg­a­tive com­pared with 76%
if Deauville scores of 1 to 2 were used. Three-year PFS was 91%
for PET-neg­a­tive patients and 66% for PET-pos­i­tive patients. For
patients with Deauville scores of 1 to 2 (n = 113), 3 (n = 22), and 4 to
5 (n = 14), 3-year PFS was 94%, 77%, and 67%, respec­tively.
The larg­est PET-adapted study, European Organization for
Research and Treatment of Cancer H10, enrolled 1950 stage I to II
patients, with 754 favor­able and 1196 unfa­vor­able HL.9 In the con­
trol arms, treat­ment consisted of 3 (favor­able) or 4 (unfa­vor­able)
ABVD cycles and involved nodal radio­ther­apy (INRT), regard­
less of PET results. In the exper­i­men­tal arms, patients received
2 cycles of ABVD, interim PET, and, if PET neg­a­tive (Deauville
scores 1-2), 2 (favor­able) or 4 (unfa­vor­able) addi­tional cycles of
ABVD. PET-pos­ i­
tive patients switched to esca­ lated BEACOPP
and INRT. Eighty-seven per­cent of favor­able and 77.6% of unfa­
vor­able patients were PET neg­a­tive. Five-year PFS rates in the
favor­able PET-neg­a­tive patients were 99% with 3 cycles of ABVD
and INRT vs 87.1% with 4 cycles of ABVD alone. In unfa­vor­able
PET-neg­a­tive patients, 5-year PFS was 92.1% with 4 cycles of
ABVD and INRT vs 89.6% with 6 cycles of ABVD. In both favor­
able and unfa­vor­able cohorts, noninferiority for che­mo­ther­apy
alone could not be dem­on­strated. Five-year OS was not sig­nif­
i­cantly dif­fer­ent at 99.6% and 98.3% with ABVD only and 100%
and 96.7% with ABVD and INRT in favor­able and unfa­vor­able
patients, respec­tively. For PET-pos­i­tive patients, 5-year PFS with
esca­lated BEACOPP was 90.6% com­pared with 77.4% with 3 to 4
cycles of ABVD and INRT (P = .002), with no OS ben­e­fit (P = .062).
The GHSG conducted the HD16 trial in 1150 patients with favor­
able stage I to II HL exam­in­ing 2 cycles of ABVD and 20 Gy IFRT
com­pared with PET-guided treat­ment with 2 cycles of ABVD, PET,
and no radio­ther­apy if PET neg­a­tive (Deauville scores 1-2) and 20
Gy IFRT if PET pos­i­tive.10 In PET-neg­a­tive patients, 5-year PFS was
93.4% with CMT and 86.1% with ABVD alone (P = .04). No OS ben­
e­fit was noted with 5-year OS of 98.1% with CMT and 98.4% with
ABVD (P = .12). Notably, PFS was sig­nif­i­cantly worse in patients
with Deauville scores of 4 to 5, with 5-year PFS of 80.9%, 93.1%,
and 93.2% for scores of 4 to 5, 1 to 3, and 1 to 2, respec­tively.
The GHSG HD17 trial exam­ined 2 cycles of esca­lated BEACOPP
plus 2 cycles of ABVD followed by 30 Gy IFRT in unfa­vor­able ear-
ly-stage HL com­pared with an exper­i­men­tal arm omit­ting IFRT in
PET-neg­a­tive (Deauville scores 1-2) patients after 4 cycles of che­
mo­ther­apy. In the PET-neg­a­tive patients, 5-year PFS was 97.7%
11
with CMT, not sta­tis­ti­cally dif­fer­ent com­pared with 95.9% with
che­mo­ther­apy alone. As in the HD16 trial, a Deauville score of 4
to 5 was a sig­nif­i­cant risk fac­tor for poor PFS, whereas scores of
1 to 3 were not asso­ci­ated with PFS in the mul­ti­var­i­able model.
Although the RAPID, EORTC H10, and GHSG HD16 tri­als all­ dis­ease (n = 152).
failed to show noninferiority in PFS with PET-adapted omis­sion
of radio­ther­apy com­pared with CMT, par­tic­u­larly in favor­able
patients, there was no OS ben­e­fit with inclu­sion of radi­a­tion.
Therefore, the treating phy­si­cian needs to weigh the risks of
poten­ tial late toxicities with radio­ ther­apy against the risk of
relapse in their patients when deter­min­ing if radio­ther­apy should
be included. Based on data show­ing con­tin­ued risk of late sec­
ond malig­nan­cies with mod­ern radio­ther­apy tech­niques and the
lack of OS ben­e­fit with IFRT, my prac­tPrakash
ice is to omitSingh Shekhawat
radio­ther­
apy in leads to dura­ble remis­sions in bulky HL.
PET-neg­a­tive patients. On the basis of the sec­ond­ary ana­ly­ses of
and HD1710 stud­ies, which dem­on­strated sig­nif­i­cantly worse PFS
only in patients with Deauville scores of 4 to 5, I define PET neg­a­
tive as Deauville scores of 1 to 3 and often omit radio­ther­apy even
in patients with a Deauville score of 3. With this approach, more
than 90% of patients with early-stage HL can avoid radio­ther­apy.
However, as all­patients in the RAPID, HD16, and HD17 tri­als with
a Deauville score of 3 received CMT and CALGB 50604 dem­on­
strated an infe­rior PFS of 77% in 22 patients with a Deauville score
of 3 treated with che­mo­ther­apy alone, the use of radio­ther­apy
is cer­tainly jus­ti­fied in patients with a Deauville score of 3 and
should be discussed with the patient and a mul­ ti­
dis­
ci­
plin­
ary
treat­ment team and weighed against poten­tial late effects.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/234/1851889/234blum.pdf by guest on 13 December 2021
In addi­ tion to the ongo­ ing con­ tro­
versy over PET-directed
radio­ther­apy, these tri­als also high­light other debates in early-​
stage HL, includ­ing num­ber of che­mo­ther­apy cycles, which reg­i­
men to start (ABVD or esca­lated BEACOPP), and treat­ment of
PET-pos­i­tive patients. Although the low-risk patient could receive
as few as 2 to 4 ABVD cycles, unfa­vor­able patients in the EORTC
and HD17 tri­als seem to have improved out­comes with 6 cycles of
ABVD or esca­ lated BEACOPP. In my own prac­ tice, I tend to
admin­is­ter 3 to 4 cycles of ABVD alone in nonbulky, unfa­vor­able
patients who are interim PET neg­a­tive based on the UK Rapid and
CALGB stud­ies but rec­og­nize that 6 cycles can also be con­sid­
ered for these patients, par­tic­u­larly if omit­ting radio­ther­apy.
For those patients with interim PET scores of 4 to 5, it remains
unclear if these patients should switch ther­apy from ABVD to
esca­lated BEACOPP. In EORTC H10, esca­lated BEACOPP and INRT
in interim PET-pos­i­tive patients improved PFS to 90.6% vs 77.4%
with ABVD and INFRT. However, in the RAPID and HD16 tri­als,
PET-pos­i­tive patients received ABVD and IFRT with­out che­mo­
ther­apy inten­si­fi­ca­tion, and PFS was 87.6% and 88.4%, respec­
tively. In my own prac­tice, I avoid the use of esca­lated BEACOPP
even in interim PET-pos­i­tive patients due to the poten­tial risks of
infer­til­ity and sec­ond­ary malig­nan­cies with the reg­i­men. I typ­i­
cally rec­om­mend IFRT for a Deauville score of 4 and biopsy fol-
lowed by sal­vage che­mo­ther­apy and autol­o­gous trans­plant for
a Deauville score of 5.
Treatment of bulky dis­ease
In bulky HL, 4 stud­ies sug­gest that radi­a­tion can be elim­i­nated
in PET-neg­a­tive patients with­out com­pro­mis­ing out­comes. The
Brit­ish Colombia Cancer Agency omit­ted radio­ther­apy in patients
with stage I to II bulky, IIB, and III to IV HL who were PET neg­
a­tive (Deauville scores 1-3) after 6 cycles of ABVD.12 Eighty-four
per­cent were PET neg­a­tive and did not receive radi­a­tion. Five-
year free­dom from treat­ment fail­ure was 89% for PET-neg­a­tive
com­pared with 56% for PET-pos­i­tive patients. In PET-neg­a­tive
patients with bulk (n = 112), 5-year free­dom from treat­ment fail­
ure was 89% com­pared with 88.5% for PET-neg­a­tive nonbulky
In CALGB 50801, 101 patients with bulky stage I to II HL were
treated with 2 cycles of ABVD followed by interim PET.13 PET-neg­
a­tive (Deauville scores 1-3) patients received 4 addi­tional cycles
of ABVD and no radi­a­tion, and PET-pos­i­tive patients received 4
cycles of esca­lated BEACOPP and 30 Gy involved site radio­ther­
apy (ISRT). Seventy-eight per­cent were PET neg­a­tive, and 3-year
PFS was 93.1% com­pared with 89.7% for PET-pos­i­tive patients,
confirming that a PET-adapted approach elim­i­nat­ing radi­a­tion
Management of early-stage HL  | 237
 In the Gruppo Italiano Terapie Innovative nei Linfomi/Fonda-
zione Italiana Linfomi HD 0607 study,14 patients with stage IIB to
IVB HL who were PET neg­a­tive after 6 cycles of ABVD and who
had a large nodal mass 5 cm or larger under­went ran­dom­i­za­tion
to 30 Gy IFRT or no fur­ther treat­ment. In the 296 PET-neg­a­tive
patients with a large nodal mass, there was no sig­nif­i­cant dif­
fer­ence in 3-year PFS with a PFS of 93% with­out radio­ther­apy
com­pared with 97% with IFRT (P = .29). Even when lim­it­ing the
anal­y­sis to patients with bulk more than 10 cm, 3-year PFS was
94% with IFRT and 86% for obser­va­tion (P = .34). Last, the UK
RATHL study included 500 patients with bulky or high-risk
stage II HL and dem­on­strated a 90.9% PFS in these patients if
PET neg­a­tive after 6 cycles of ABVD.15 Therefore, based on these
stud­ies, I omit IFRT in patients with bulky early-stage HL who
achieve a neg­a­tive PET.
Incorporation of novel agents
Unlike advanced-stage HL in which large ran­dom­ized stud­ies
have exam­ined brentuximab vedotin (BV) and check­point inhib­
i­tors com­bined with doxo­ru­bi­cin, vin­blas­tine, and dacarbazine
(AVD) or BEACOPP,16-18 only small stud­ies have been com­pleted
incor­po­rat­ing these agents into front­line ther­apy for early-stage
HL. Abramson et al19 eval­u­ated com­bined AVD/BV with­out radio­
ther­apy in 34 patients with nonbulky early-stage HL. Patients re-
ceived 1 cycle of BV alone on days 1 and 15 followed by 4 cycles
of com­ bined AVD/BV. The com­ plete response (CR) rate was
52% after the lead-in cycle of BV and 97% after 2 AVD/BV cycles.
Three-year PFS was 94%. Grade 3 to 4 events included sen­sory
neu­rop­at­hy (23%), neutropenia (62%), and febrile neutropenia
(35%), although neutropenic fever declined once growth fac­tor
sup­port was man­dated. Park and col­leagues exam­ined con­sol­i­
da­tion of patients with nonbulky (defined as ≤7.5 cm) early-stage
HL with 6 cycles of BV after 2 to 6 cycles of ABVD.20 Of the 41
patients enrolled, 36 com­pleted the planned 6 BV doses. With
a con­sol­i­da­tion approach, 95% achieved a CR, and 3-year PFS
was 92%.
Kumar et al21 conducted a pilot study of CMT using AVD/BV
and 30 Gy ISRT in 30 patients with early unfa­vor­able HL. For-
ty-seven per­cent of patients had bulky dis­ease. Twenty-seven
patients achieved CR, and 1-year PFS was 93.3%. This trial
recently enrolled 3 addi­tional cohorts test­ing the fea­si­bil­ity of
reduc­ing the ISRT dose to 20 Gy, reduc­ing the radi­a­tion field
with con­sol­i­da­tion vol­ume radi­a­tion, and elim­i­nat­ing ISRT.22 To
date, 27% of the 117 patients enrolled across all 4 cohorts had
bulky dis­ease, and CR rates were 93% to 97% in the CMT cohorts
and 97% with AVD/BV alone. With a median fol­low-up of 3.8
years, 2-year PFS was 94% in all­patients, with a 2-year PFS of
96.6% in the che­mo­ther­apy alone arm, although fol­low-up in this
no radio­ther­apy cohort is still short at 2.2 years.
Two stud­ies have explored check­point inhib­it­ ors in early-stage
HL. The GHSG conducted a trial in 109 patients with unfa­vor­able
or bulky HL exam­in­ing con­cur­rent nivolumab/AVD for 4 cycles or
sequen­tial ther­apy with 4 nivolumab cycles, 2 nivolumab/AVD
cycles, and 2 AVD cycles.23 All patients received 30 Gy ISRT.
Eighty-seven per­cent and 26% of patients achieved a CR after 2
nivolumab/AVD cycles or 4 nivolumab cycles, respec­tively. At the
com­ple­tion of treat­ment, the CR rates were 83% and 84%, and
2-year PFS was 100% and 98% in the com­bi­na­tion and sequen­tial
arms, respec­tively. Grade 3 to 4 toxicities were sim­il­ar with both
approaches, and hypo­ thy­
roid­
ism was the most fre­ quent late
effect, persisting in 17% of patients. Allen and col­leagues com­
238  |  Hematology 2021  |  ASH Education Program
pleted a phase 2 study of 3 doses of pembrolizumab followed by
4 to 6 cycles of AVD with­out radio­ther­apy in 12 early-stage and 18
advanced-stage patients, includ­ing 10 patients with bulky tumors
more than 10 cm.24 After 3 cycles of pembrolizumab, the CR rate
was 37% and improved to 100% after 2 cycles of AVD. Two-year
PFS was 100%. Grade 3 to 4 events included neutropenia in 3
patients and transaminitis, lymphopenia, diar­rhea, and Bell palsy
in 1 patient each.
Therefore, BV and check­ point inhib­ i­
tors appear safe and
effec­tive in early-stage and bulky HL with PFS exceed­ing 90%;
how­ever; large ran­dom­ized tri­als are nec­es­sary to estab­lish the
safety, cost-effec­tive­ness, and long-term effi­cacy in com­par­i­
son to cur­rent PET-adapted ABVD ther­apy. Although sequen­tial
admin­is­tra­tion may min­i­mize tox­ic­ity, these prolonged treat­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/234/1851889/234blum.pdf by guest on 13 December 2021
ment approaches may lead to sig­nif­i­cant edu­ca­tion, childcare,
finan­cial, and employ­ment con­straints in an AYA pop­u­la­tion.
Alternatively, as dem­on­strated by Abramson et al19 and Allen et
al,24 finan­cial costs and treat­ment risks may be min­i­mized with
1 to 3 lead-in cycles of BV or check­point inhib­i­tor prior to AVD,
and these lim­ited approaches should also be eval­u­ated. In con­
clu­sion, although it is likely that the par­a­digm of treat­ment of
early-stage HL will likely shift to incor­po­rate BV and check­point
inhib­i­tors front­line, which agents to use, how to incor­po­rate
them (lead-in, sequen­tial, or con­cur­rent admin­is­tra­tion), which
reg­i­men (AVD or BEACOPP) to com­bine with, and whether con­
ven­tional cyto­toxic ther­apy can ulti­mately be elim­in ­ ated are
still under inves­ti­ga­tion.
CLINICAL CASE (Con­tin­ued)
The 20-year-old woman with nonbulky stage IIA clas­sic HL with
2 adverse risk fac­tors (5 nodal sites and ele­vated ESR) started
ABVD che­mo­ther­apy, and after 2 cycles, PET was performed,
dem­on­strat­ing a right supraclavicular node of 2.1 × 1.5 cm with
SUV 5.9 com­pared with 2.9 × 3 cm with SUV 7.3 pre­vi­ously, new
right axil­lary lymph node of 1.5 × 1.1 cm with SUV 7.5, and wors­
en­ing ante­rior medi­as­ti­nal nodal mass with SUV 10.2 com­pared
with 6.3 pre­ vi­
ously (Deauville score 5). Core needle biopsy
spec­ i­
men of the supraclavicular node con­ firmed per­ sis­
tent
clas­sic HL. She stopped ABVD ther­apy and began 4 cycles of
com­bined BV/nivolumab sal­vage ther­apy. After 4 cycles, she
achieved a CR with a Deauville score of 3. She proceeded to
autol­o­gous trans­plant followed by 30 Gy pro­ton ther­apy to
bilat­eral neck, medi­as­ti­num, and axil­lary nodal sites. She ini­ti­
ated BV main­te­nance after radi­a­tion and remains on treat­ment
for a planned max­i­mum of 16 cycles.
Conflict-of-inter­est dis­clo­sure
Kristie A. Blum: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Kristie A. Blum: Brentuximab, nivolumab, and pembrolizumab are
not FDA approved as frontline therapy in early-stage HL.
Correspondence
Kristie A. Blum, Department of Hematology and Medical Oncolo-
gy, Emory Winship Cancer Institute, 1365 Clifton Road NE, B4013,
Atlanta, GA 30322; e-mail: kablum@emory​­.edu.
References
1. Dores GM, Curtis RE, Dalal NH, et al. Cause-spe­cific mor­tal­ity fol­low­ing ini­
tial che­mo­ther­apy in a pop­u­la­tion-based cohort of patients with clas­si­cal
Hodgkin lym­phoma, 2000-2016. J Clin Oncol. 2020;38(35):4149-4162.
2. Giulino-Roth L, Pei Q , Buxton A, et al. Subsequent malig­nant neo­plasms
among chil­dren with Hodgkin lym­phoma: a report from the Children’s
Oncology Group. Blood. 2021;137(11):1449-1456.
3. Schaapveld M, Aleman BM, van Eggermond AM, et al. Second can­cer risk
up to 40 years after treat­ment for Hodgkin’s lym­phoma. N Engl J Med.
2015;373(26):2499-2511.
4. Bright CJ, Reulen RC, Winter DL, et  al. Risk of sub­se­quent pri­mary neo­
plasms in sur­vi­vors of ado­les­cent and young adult can­cer (Teenage and
Young Adult Cancer Survivor Study): a pop­u­la­tion-based, cohort study.
Lancet Oncol. 2019;20(4):531-545.
5. Zinzani PL, Rigacci L, Stefoni V, et  al. Early interim 18F-FDG PET in Hod-
gkin’s lym­phoma: eval­u­a­tion on 304 patients. Eur J Nucl Med Mol Imaging.
2012;39(1):4-12.
6. Radford J, Illidge T, Counsell N, et al. Results of a trial of PET-directed ther­
apy for early-stage Hodgkin’s lym­phoma. N Engl J Med. 2015;372(17):1598-
1607.
7. Barrington SF, Phil­lips EH, Counsell N, et al. Positron emis­sion tomog­ra­phy
score has greater prog­nos­tic sig­nif­i­cance than pre­treat­ment risk strat­i­fi­ca­
tion in early-stage Hodgkin lym­phoma in the UK RAPID Study. J Clin Oncol.
2019;37(20):1732-1741.
8. Straus DJ, Jung S-H, Pitcher B, et al. CALGB 50604: risk-adapted treat­ment
of nonbulky early-stage Hodgkin lym­phoma based on interim PET. Blood.
2019;132(10):1013-1021.
9. André MPE, Girinsky T, Federico M, et al. Early pos­i­tron emis­sion tomog­
ra­phy response-adapted treat­ment in stage I and II Hodgkin lym­phoma:
final results of the ran­dom­ized EORTC/LYSA/FIL H10 trial. J Clin Oncol.
2017;35(16):1786-1794.
10. Fuchs M, Goergen H, Kobe C, et al. Positron emis­sion tomog­ra­phy-guided
treat­ment in early-stage favor­able Hodgkin lym­phoma: final results of the
inter­na­
tional, ran­
dom­ ized phase III HD16 trial by the Ger­ man Hodgkin
Study Group. J Clin Oncol. 2019;37(31):2835-2845.
11. Borchmann P, Plütschow A, Kobe C, et al. PET-guided omis­sion of radio­ther­
apy in early-stage unfavourable Hodgkin lym­phoma (GHSG HD17): a multi-
centre, open-label, randomised, phase 3 trial. Lancet Oncol. 2021;22(2):
223-234.
12. Savage KJ, Connors JM, Villa DR, et al. Advanced stage clas­si­cal Hodgkin
lym­phoma patients with a neg­a­tive PET-scan fol­low­ing treat­ment with
ABVD have excel­lent out­comes with­out the need for consolidative radio­
ther­apy regard­less of dis­ease bulk at pre­sen­ta­tion. Blood. 2015;126(123):579.
Prakash Singh Shekhawat
13. LaCasce AS, Dockter T, Ruppert AS, et  al. CALGB 50801 (Alliance): PET
adapted ther­apy in bulky stage I/II clas­sic Hodgkin lym­phoma (cHL). J Clin
Oncol. 2021;39(15, suppl):7507.
14. Gallamini A, Tarella C, Viviani S, et al. Early che­mo­ther­apy inten­si­fi­ca­tion
with esca­lated BEACOPP in patients with advanced-stage Hodgkin lym­
phoma with a pos­i­tive interim pos­i­tron emis­sion tomog­ra­phy/com­puted
tomog­ra­phy scan after two ABVD cycles: long-term results of the GITIL/
FIL HD 0607 trial. J Clin Oncol. 2018;36(5):454-462.
15. Johnson P, Federico M, Kirkwood A, et al. Adapted treat­ment guided by
interim PET-CT scan in advanced Hodgkin’s lym­ phoma. N Engl J Med.
2016;374(25):2419-2429.
16. Eichenauer DA, Plütschow A, Kreissl S, et al. Incorporation of brentuximab
vedotin into first-line treat­ ment of advanced clas­ si­
cal Hodgkin’s lym­
phoma: final anal­y­sis of a phase 2 randomised trial by the Ger­man Hodgkin
Study Group. Lancet Oncol. 2017;18(12):1680-1687.
17. Straus DJ, Długosz-Danecka M, Alekseev S, et al. Brentuximab vedotin with
che­mo­ther­apy for stage III/IV clas­si­cal Hodgkin lym­phoma: 3-year update
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/234/1851889/234blum.pdf by guest on 13 December 2021
of the ECHELON-1 study. Blood. 2020;135(10):735-742.
18. Ramchandren R, Domingo-Domènech E, Rueda A, et  al. Nivolumab for
newly diag­nosed advanced-stage clas­sic Hodgkin lym­phoma: safety and
effi­cacy in the phase II checkMate 205 Study. J Clin Oncol. 2019;37(23):1997-
2007.
19. Abramson JS, Arnason JE, LaCasce AS, et  al. Brentuximab vedotin, doxo­
ru­bi­cin, vin­blas­tine, and dacarbazine for nonbulky lim­ited-stage clas­si­cal
Hodgkin lym­phoma. Blood. 2019;134(7):606-613.
20. Park SI, Shea TC, Olajide O, et  al. ABVD followed by BV con­sol­i­da­tion in
risk-strat­i­fied patients with lim­ited-stage Hodgkin lym­phoma. Blood Adv.
2020;4(11):2548-2555.
21. Kumar A, Casulo C, Yahalom J, et al. Brentuximab vedotin and AVD followed
by involved-site radio­ther­apy in early stage, unfa­vor­able risk Hodgkin lym­
phoma. Blood. 2016;128(11):1458-1464.
22. Kumar A, Casulo C, Advani RH, et al. Brentuximab vedotin com­bined with
che­mo­ther­apy in patients with newly diag­nosed early-stage, unfa­vor­able-
risk Hodgkin lym­phoma. J Clin Oncol. 2021;39(20):2257-2265.
23. Bröckelmann PJ, Goergen H, Keller U, et al. Efficacy of nivolumab and AVD in
early-stage unfa­vor­able clas­sic Hodgkin lym­phoma: the ran­dom­ized phase
2 Ger­man Hodgkin Study Group NIVAHL trial. JAMA Oncol. 2020;6(6):872.
24. Allen PB, Savas H, Evens AM, et  al. Pembrolizumab followed by AVD in
untreated early unfa­vor­able and advanced-stage clas­si­cal Hodgkin lym­
phoma. Blood. 2021;137(10):1318-1326.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000255
Management of early-stage HL  | 239
 HODGKIN LYMPHOMA: CURE AND OPTIMAL SURVIVORSHIP

How to choose first salvage therapy

in Hodgkin lymphoma: traditional
chemotherapy vs novel agents

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/240/1851809/240driessen.pdf by guest on 13 December 2021

Julia Driessen,1 Sanne H. Tonino,1 Alison J. Moskowitz,2 and Marie José Kersten1
Department of Hematology, Amsterdam UMC, University of Amsterdam, LYMMCARE, Cancer Center Amsterdam, Amsterdam, The Netherlands;
1

and 2Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, NY

Approximately 10% to 30% of patients with classical Hodgkin lymphoma (cHL) develop relapsed or refractory (R/R)
disease. Of those patients, 50% to 60% show long-term progression-free survival after standard salvage chemotherapy
followed by high-dose chemotherapy (HDCT) and autologous stem cell transplant (ASCT). In the past decade, novel
therapies have been developed, such as the CD30-directed antibody–drug conjugate brentuximab vedotin and immune
checkpoint inhibitors, which have greatly extended the treatment possibilities for patients with R/R cHL. Several phase
1/2 clinical trials have shown promising results of these new drugs as monotherapy or in combination with chemother-
apy, but unfortunately, very few randomized phase 3 trials have been performed in this setting, making it difficult to give
evidence-based recommendations for optimal treatment sequencing. Two important goals for the improvement in the
treatment of R/R cHL can be identified: (1) increasing long-term progression-free and overall survival by optimizing risk-
adapted treatment and (2) decreasing toxicity in patients with a low risk of relapse of disease by evaluating the need for
HDCT/ASCT in these patients. In this review, we discuss treatment options for patients with R/R cHL in different settings:
patients with a first relapse, primary refractory disease, and in patients who are ineligible or unfit for ASCT. Results of
clinical trials investigating novel therapies or strategies published over the past 5 years are summarized.

LEARNING OBJECTIVES
• Describe current and emerging therapies for patients with R/R Hodgkin lymphoma
• Understand the importance for patients with R/R Hodgkin lymphoma to achieve a CMR before HDCT/ASCT

phamide, vincristine, procarbazine, and prednisone. After 2

CLINICAL CASE
A 30-year-old woman presented with a persistent painless
lump in the neck without B-symptoms. A biopsy of a right
supraclavicular node was performed, which showed a clas-
sical Hodgkin lymphoma (cHL). An 18F-fluorodeoxyglucose
positron emission tomography (PET)–computed tomogra-
phy (CT) scan revealed lymphadenopathy bilaterally in the
supraclavicular and infraclavicular region, retrosternally,
and in the mediastinum; hence, cHL stage IIA unfavorable
was diagnosed.
After oocyte preservation, treatment with adriamycin,
bleomycin, vinblastine, and dacarbazine was initiated with
the intention to administer a total of 6 cycles in case of a com-
plete metabolic response (CMR) after 2 cycles. However, the
interim PET-scan showed only a partial metabolic response
(PR) (Deauville score 4), and the treatment was intensifed
to escalated bleomycin, etoposide, adriamycin, cyclophos-
cycles, a CMR was reached and the patient received consol-
idative involved node radiotherapy (30 Gy).
Unfortunately, 1 year later, the patient presented with
night sweats and severe itching. Imaging revealed exten-
sive lymphadenopathy above and below the diaphragm,
and a biopsy confrmed the relapse. Salvage chemother-
apy with dexamethasone, high-dose cytarabine, and cis-
platin (DHAP) was initiated, which resulted in a CMR after
2 cycles, and stem cells were mobilized and collected
after a third cycle of DHAP with the intention to proceed
to high-dose chemotherapy (HDCT) followed by autolo-
gous stem cell transplant (ASCT) rescue.
Introduction
Approximately 10% to 30% of patients with cHL will relapse or
are primary refractory (R/R) to frst-line treatment. Standard

240 | Hematology 2021 | ASH Education Program

Table 1. Overview of first-sal­vage che­mo­ther­apy reg­i­mens since 2010

Refractory,
Study N Intervention n (%) CR pre-ASCT ORR pre-ASCT PFS OS
Josting et al 279 DHAP 0 (0) CT: 24% CT: 71% 3 years: 69% (no 3 years: 85% (no
(2010)13 (RCT) sig­nif­i­cant dif­fer­ence sig­nif­i­cant dif­fer­ence
between arms) between arms)
Moskowitz et al 105 ICE 48 (46) PET/gal­lium: 61% CT: 59% 4 years: 56% 4 years: 72%
(2010)14 CT: 33%
Moskowitz et al 97 ICE + GVD 41 (42) PET: 60% after ICE — 51 months: 70% 51 months: 80%
(2012)11 (PET-adapted 78% after GVD
sequen­tial)
Labrador et al 82 ESHAP 41 (50) PET/gal­lium: 50% PET/ gal­lium: 67% Median PFS: 56 months 5 years: 73%

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/240/1851809/240driessen.pdf by guest on 13 December 2021

(2014)15 (ret­
ro­spec­tive)
Santoro et al 58 BeGEV 27 (46) PET: 73% PET: 83% 5 years: 59% 5 years: 78%
(2016)16
BeGEV, bendamustine, gemcitabine, and vinorelbine.

sal­vage che­mo­ther­apy and ­con­sol­i­da­tion with HDCT/ASCT leads
to long-term pro­gres­sion-free sur­vival (PFS) in about 50% to 60%
of patients. Until recently, patients who relapsed after ASCT or
were inel­i­gi­ble for ASCT had lim­ited treat­ment options, and 50% of
those patients even­tu­ally died of the dis­ease.1 In the past decades,
sev­eral novel ther­a­peu­tic options for patients with R/R cHL have
become avail­­able, includ­ing brentuximab vedotin (BV) and immune
check­point inhib­i­tors (CPIs), lead­ing to high CMR rates pre-ASCT,
espe­cially when com­bined with che­mo­ther­apy.2 Achieving a CMR
prior to ASCT appears to be the most impor­tant prog­nos­tic fac­tor
for PFS.3-10 Therefore, a risk- and PET-adapted treat­ment approach
could prob­a­bly lead to higher cure rates.11 On the other hand, the
bur­den of late toxicities related to HDCT, such as sec­ond­ary malig­
nan­cies and infer­til­ity, is con­sid­er­able, espe­cially as the dis­ease
typ­i­cally affects patients early in life. For this rea­son, decreas­ing
tox­ic­ity is one of the main goals in the treat­ment of R/R cHL.12
In this edu­ca­tional ses­sion, we dis­cuss the results of stud­ies
that incor­po­rated novel ther­a­pies and response-adapted treat­
ment and how this could be implemented in stan­dard prac­tice
to improve out­comes for patients with R/R cHL.
Treatment for patients with a first relapse or pri­mary
refrac­tory dis­ease after first-line treat­ment
Conventional sal­vage che­mo­ther­apy results in pre-ASCT com­
plete response (CR) rates of about 20% to 25% and over­ all
response rates (ORRs) of 60% to 70%, based on eval­u­a­tion by
CT scan.13 More recent stud­ies reporting response rates based
on func­tional imag­ing using PET or gal­lium showed CMR rates of
50% to 60% after ifosfamide, carboplatin, and etoposide (ICE)
or etoposide, meth­yl­pred­nis­o­lone, high-dose cytarabine, and
cis­platin (ESHAP). Even higher CMR rates were reported for the
bendamustine, gemcitabine, and vinorelbine reg­i­men (73%) and
a sequen­tial ICE–gemcitabine, vinorelbine, and lipo­so­mal doxo­
ru­bi­cin (GVD) approach (78%) in which patients with no CR on
ICE received addi­ tional che­mo­ther­
apy with GVD before pro­
ceed­ing to ASCT.11,14-16 PFS ranges between 50% and 60% with an
over­all sur­vival (OS) of 70% to 80% at 5 years.11,13-16 Overall, there
seem to be no sig­nif­i­cant dif­fer­ences with regard to out­come
between the most com­monly used reg­i­mens (ie, ICE/DHAP/
Prakash Singh Shekhawat
Salvage treatment in relapsed/refractory Hodgkin lymphoma  |  241
ESHAP) (Table 1). However, ran­dom­ized con­trolled tri­als (RCTs)
com­par­ing dif­fer­ent sal­vage che­mo­ther­apy reg­i­mens or a PET-
adapted approach are lacking.
BV and check­point inhib­i­tors
cHL is char­ ac­
ter­
ized by the pres­ ence of a minor­ity of bi- or
mul­ ti­
nu­cle­ated Hodgkin and Reed-Sternberg (HRS) cells that
uni­ver­sally express CD30 in an inflam­ma­tory tumor micro­en­vi­
ron­ment. BV is an anti-CD30 mono­clo­nal anti­body con­ju­gated
to the micro­tu­bule-disrupting agent monomethyl auristatin-E.17
PD-L1 and PD-L2 are upregulated by HRS cells in about 90%
of patients and induce T-cell exhaus­tion, which con­trib­utes to
immune escape of HRS cells.18 CPIs are mono­clo­nal antibodies
that block the inter­ac­tion between inhib­i­tory ligands such as PD-
L1 and PD-L2 on the tumor cells and PD-1 recep­tors on immune
effec­tor cells.
Several stud­ies have inves­ti­gated the use of BV in com­bi­na­
tion with che­mo­ther­apy as first sal­vage reg­i­men and showed
high CMR rates prior to ASCT of up to 83%, with 2-year PFS
rates rang­ ing from 63% to 81% (Table 2).3-10 In 5 stud­ ies, BV
was com­bined with che­mo­ther­apy in 2 to 6 cycles followed by
ASCT in patients with PR or CMR, whereas in 2 stud­ies, patients
were treated ini­tially with 4 to 6 admin­is­tra­tions of BV mono-
therapy; patients with a CMR could pro­ceed directly to ASCT,
whereas patients with a PR received addi­tional sal­vage che­mo­
ther­apy with­out BV.4,5 This PET-adapted approach is inter­est­ing
because approx­i­ma­tely 30% to 50% of patients could pro­ceed
to ASCT after BV monotherapy only, thereby avoiding tox­ic­ity
from sal­vage che­mo­ther­apy in these patients. Moreover, a trial
inves­ti­gat­ing the com­bi­na­tion of BV and the CPI nivolumab as
pre-ASCT sal­vage reg­i­men showed that 67 of 91 patients could
pro­ceed directly to ASCT after BV-nivolumab, with­out sal­vage
che­mo­ther­apy. The study revealed low tox­ic­ity of this reg­i­men
com­pared with sal­vage che­mo­ther­apy.10
Thus far, in stud­ies that incor­po­rated a PET-adapted strat­egy,
PFS seems to be sim­i­lar for patients who proceeded to ASCT
directly after hav­ing a CMR on BV, BV-nivolumab, or ICE alone as
for those patients who needed addi­tional sal­vage che­mo­ther­
apy to achieve a CMR.4,5,10 This con­firms that the most i­mpor­tant
Table 2. Overview of first-sal­vage reg­i­mens containing BV or CPI

Refractory, CMR pre-ASCT,

Study N Intervention Schedule n (%) n (%) 2-year PFS 2-year OS
Moskowitz et al 65 BV + sequen­tial ICE BV 1.2 mg/kg d1, 8, 15 of 34 (52) 54 (83) 82% 97%
(2017)4 28-d cycles, 2 cycles.
ICE sal­vage in case of
Deauville >3.
Herrera et al 57 BV + sequen­tial ICE/GVD BV 1.8 mg/kg every 21 d, 4 35 (61) 37 (65) 67% 93%
(2018)5 cycles. Last 2 cycles BV
esca­la­tion to 2.4 mg/kg
in n = 8 patients with
PR/SD. Salvage che­mo­
ther­apy at dis­cre­tion of

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/240/1851809/240driessen.pdf by guest on 13 December 2021

treating phy­si­cian.
Cole et al 45 BV + gemcitabine BV 1.8 mg/kg on d1 and d8 29 (64) 28 (67) — 1 year: 95%
(2018)6 every 21 days, 4 cycles.
In com­bi­na­tion with
gemcitabine.
LaCasce et al 55 BV + bendamustine BV 1.8 mg/kg every 21 d, 28 (51) 39 (74) 63% 94%
(2018)9 2-6 cycles. In com­bi­na­
tion with bendamustine.
Post-ASCT BV
monotherapy main­te­
nance up to 16 cycles.
Garcia-Sanz 66 BV + ESHAP BV 1.8 mg/kg every 21 40 (61) 46 (70) 71% 90%
et al (2019)8 days, 4 cycles. In com­
bi­na­tion with 3 cycles of
ESHAP.
Broccoli et al 40 BV + bendamustine BV 1.8 mg/kg every 21 20 (50) 30 (79) 68% 97%
(2019)7 days, 4-6 cycles. In
com­bi­na­tion with
bendamustine.
Abuelgasim 28 BV + IGEV BV 1.8 mg/kg every 12 (43) includ­ing 70% 73.5% (100% for 87.1% (100% for
et al (2019)19 21 days, 2-4 cycles. In n = 14 with >1 patients with patients with
com­bi­na­tion with IGEV. line of ther­apy. first relapse) first relapse)
64% received BV con­sol­
i­da­tion after ASCT.
Kersten et al 67 BV + DHAP BV 1.8 mg/kg every 21 30 (45) 53 (82) 78% 96%
(2021)3 days, 3 cycles. In com­bi­
na­tion with DHAP.
Advani et al 91 BV + nivolumab BV 1.8 mg/kg and 38 (42) 61 (67) 78% 93%
(2021)10 nivolumab 3.0 mg/kg
every 21 days, 4 cycles.
Moskowitz et al 39 Pembrolizumab + GVD Pembrolizumab 200 mg 16 (41) 36 (95) 1 year: 100% 1 year: 100%
(2021)20 every 21 days, 4 cycles.
In com­bi­na­tion with
GVD.
d, day; IGEV, ifosfamide, gemcitabine, vinorelbine, and pred­nis­o­lone; SD, sta­ble dis­ease.

goal is to achieve a CMR before ASCT and that patients who
do not respond ini­tially can poten­tially be res­cued with addi­
tional sal­vage che­mo­ther­apy and pro­ceed to ASCT if they sub­
se­quently reach a CMR.
Figure 1 pro­poses a flow­chart of PET-adapted treat­ment in
patients with R/R cHL. Because BV and CPI are not yet avail­­able
or reim­bursed as first sal­vage treat­ment in many countries, sal­
vage che­mo­ther­apy is often still the stan­dard approach.
High-dose che­mo­ther­apy and ASCT
With increas­ing CMR rates pre-ASCT, one might ques­tion the need
for con­sol­i­da­tion with HDCT/ASCT in all­patients. HDCT/ASCT
is asso­ci­ated with imme­di­ate tox­ic­ity such as cytopenias and
mucositis and long-term tox­ic­ity such as infer­til­ity and sec­ond­ary
malig­nan­cies.12 Superiority for HDCT/ASCT over mini-carmustine,
etoposide, cytarabine, and melphalan (BEAM) (ie, reduced-dose
BEAM with­out ASCT) in R/R cHL was shown in 2 ran­dom­ized
con­trolled tri­als (RCTs) in 1993 and 2002.21,22 However, in these
tri­als, patients did not receive any sal­vage che­mo­ther­apy before
BEAM or mini-BEAM. In addi­tion, with the advent of effec­tive
drugs such as BV and CPIs, a sub­set of patients may be cured
with sal­vage treat­ment alone.17
A recently published study using pembrolizumab and GVD
che­ mo­ ther­
apy followed by HDCT/ASCT showed a very high

242  |  Hematology 2021  |  ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/240/1851809/240driessen.pdf by guest on 13 December 2021

Figure 1. Flowchart of treatment for R/R cHL. IGEV, ifosfamide, gemcitabine, vinorelbine, and prednisolone; SCT, stem cell transplant.

­ re-ASCT CMR rate of 95%, and with a median fol­low-up of 1 year,

p There is a high unmet need for RCTs in the pre-ASCT R/R
no pro­gres­sions have occurred.20 This study has now started a set­ting. Future stud­ies should focus on the opti­mal sequence of
sec­ond part in which patients with a CMR after 2 cycles of pem- using CPIs and BV in sal­vage treat­ment and con­sol­i­da­tion strat­e­
brolizumab-GVD con­tinue with 2 addi­tional cycles of pembroli- gies to induce high CMR rates while at the same time min­i­miz­ing
zumab-GVD followed by pembrolizumab con­sol­i­da­tion instead early and late tox­ic­ity. Eventually, an RCT should be performed
of HDCT/ASCT. Results of this rev­ o­
lu­
tion­
ary approach could to estab­lish the role of risk- and PET-adapted treat­ment in R/R
change the treat­ment of patients with R/R cHL sig­nif­i­cantly. cHL, includ­ing the role of HDCT/ASCT.
Prakash Singh Shekhawat
Salvage treatment in relapsed/refractory Hodgkin lymphoma  |  243
Patients with pri­mary refrac­tory dis­ease
Primary refrac­tory dis­ease and a short inter­val between first-line
treat­ment and relapse are impor­tant poor prog­nos­tic fac­tors
for response to sal­vage ther­apy and PFS.3,11,14 CMR rates to sal­
vage ther­apy for refrac­tory patients are usu­ally lower com­pared
with relapsed patients: 73% vs 86% after DHAP, 64% vs 84% after
BV-bendamustine, and 53% vs 77% after BV-nivolumab, respec­
tively.3,9,10 However, in patients who do achieve a CMR to sal­vage
ther­apy and pro­ceed to ASCT, post-ASCT PFS for pri­mary refrac­
tory and relapsed patients is sim­i­lar.3,11 A ret­ro­spec­tive anal­y­sis in
78 patients who progressed on one or more sal­vage reg­i­mens
and who were treated with reg­i­mens containing CPI showed
favor­able results, with 59% of patients achiev­ing a CMR and a
post-ASCT PFS of 81% at 18 months.23 This sug­gests that treat­
ment with CPI may improve chemosensitivity of pre­vi­ously che-
morefractory dis­ease. Interestingly, pre-ASCT CMR sta­tus was
not sig­nif­i­cantly prog­nos­tic for post-ASCT PFS in this cohort,
suggesting a PR might be suf­fi­cient for pro­ceed­ing to ASCT after
CPI in this patient pop­u­la­tion.
Patients with a late relapse
Although most relapses after first-line treat­ment occur within
the first 2 years, a minor­ity of patients has a late relapse more
than 5 years after first-line treat­ment.24 A large ret­ro­spec­tive
anal­y­sis showed that late relapses occur more fre­quently in
patients with early-stage favor­ able dis­
ease com­ pared with
patients with unfa­vor­able or advanced-stage dis­ease. Whether
these relapses rep­re­sent true relapses or a new sec­ond cHL
man­i­fes­ta­tion, pos­si­bly due to genetic or envi­ron­men­tal risk
fac­tors in these patients, remains largely unknown. About half
of these patients were treated with ASCT, which was asso­ci­
ated with favor­able PFS and OS com­pared with other sal­vage
ther­ap­ ies; how­ever, non-ASCT approaches, such as those using
com­bi­na­tion che­mo­ther­apy or occa­sion­ally radi­a­tion alone,
could be con­sid­ered depending on the patient’s ini­tial treat­
ment and under­ly­ing comorbidities.24
Maintenance treat­ment after ASCT
For patients with a high risk of relapse after ASCT, main­te­nance
treat­ment with BV can be con­sid­ered. In a study investigating
BV maintenance, 329 patients with unfavorable risk R/R cHL
(defined as primary refractory disease, relapse <1 year, or extran-
odal disease) received either up to 16 cycles of BV maintenance
or placebo after ASCT.25 The study showed improve­ment in PFS
in patients receiv­ing BV main­te­nance, with a 5-year PFS of 59%
vs 41% for pla­cebo. However, there was no dif­fer­ence in OS,
prob­a­bly because 87% of patients who relapsed in the pla­cebo
arm received BV at the sub­se­quent relapse. Therefore, the use
of BV main­te­nance after ASCT could poten­tially be restricted to
patients with at least 2 risk fac­tors, or alter­na­tively, its use could
be delayed until progression. With the increas­ing use of BV in
the first-line set­ting, it is also impor­tant to inves­ti­gate whether
patients who relapse after BV in com­bi­na­tion with che­mo­ther­apy
will still show advan­tage of BV maintenance after ASCT.26 Alter-
natively, CPI could be used as main­te­nance treat­ment; a small
phase 2 trial in 30 high-risk patients showed high post-ASCT
PFS.27 The com­bi­na­tion of CPI and BV main­te­nance in 59 high-
risk patients has also shown prom­is­ing results, with 5 patients
with PR converting to CR dur­ing main­te­nance.28 Further stud­ies
should inves­ti­gate the role of post-ASCT main­te­nance in high-
244  |  Hematology 2021  |  ASH Education Program
risk patients with CPI and/or BV vs reserv­ing these treat­ments
for a sub­se­quent relapse.
BV and CPI treat­ment for patients who relapse after ASCT
or are inel­i­gi­ble for ASCT
Patients who relapse after ASCT or are inel­i­gi­ble for ASCT due to
che­mo­ther­apy-resis­tant dis­ease gen­er­ally have a poor prog­no­
sis.1 In Table 3, we sum­ma­rize the most impor­tant recent stud­ies
in patients with R/R cHL who have pro­gres­sion after at least 1
line of sal­vage treat­ment. The first break­through in the post-ASCT
set­ting was the appli­ca­tion of monotherapy with BV in heavily
pretreated R/R patients, which showed an ORR of 75% and a
CMR rate of 34% with a median PFS of 20.5 months in those with
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/240/1851809/240driessen.pdf by guest on 13 December 2021
a CMR. The PFS rate at 5 years, how­ever, was only 22% with an
OS of 41%, high­light­ing the need for addi­tional treat­ment options
(Table 3).17
A study that investigated pembrolizumab monotherapy showed
an ORR of 69% and a CMR rate of 22%, with a 2-year PFS of 31%
and OS of 91%.2 Several dif­fer­ent CPIs and also com­bi­na­tions of
CPIs have been inves­ti­gated in R/R cHL.30,31,33,34 In a phase 1 trial, 64
patients were ran­dom­ized between ipilimumab-BV, nivolumab-BV,
and tri­ple ther­apy with ipilimumab-nivolumab-BV. The trial showed
dif­fer­ences in tox­ic­ity pro­file and effi­cacy between the 3 reg­i­mens,
with the highest per­cent­age of grade 3/4 adverse events in the
trip­let and ipilimumab-BV group, whereas the highest ORR and
CMR rates were found in the trip­let and nivolumab-BV group.34
In a recently published head-to-head com­par­i­son of mono-
therapy with pembrolizumab vs monotherapy with BV, pembroli-
zumab showed a sig­nif­i­cantly higher median PFS of 13.2 months
vs 8.3 months for BV.35 The inci­dence of adverse events was com­
pa­ra­ble between the 2 groups, with immune-medi­ated adverse
events in the pembrolizumab arm and neu­rop­a­thy in the BV arm.
Importantly, in patients who received ear­lier treat­ment with
BV or CPI, response rates seem to be sim­i­lar to patients who
have not received BV or CPI before. A ret­ro­spec­tive study eval­
u­ated 18 patients with R/R cHL and 10 patients with R/R ana­
plas­tic large cell lym­phoma who received treat­ment with BV
in 2 lines and showed CMR and ORR that are com­pa­ra­ble with
patients who received BV for the first time.37 Retreatment with
CPI in 78 patients with R/R cHL who relapsed after nivolumab
also showed com­pa­ra­ble effi­cacy.38
Role of radio­ther­apy in the man­age­ment of R/R cHL
The role of radio­ther­apy in the R/R set­ting has not been revis-
ited well in this era of novel treat­ment options. Radiotherapy
can be used pre-ASCT or post-ASCT on resid­ual lesions or in
patients with extranodal or bulky dis­ease and as part of the
con­di­tion­ing reg­i­men using total lym­phoid irra­di­a­tion, but
com­par­a­tive data about effi­cacy of radio­ther­apy in these set­
tings are scarce and out­dated.39 Earlier stud­ies have shown
that patients who receive radio­ther­apy have a decreased risk
of local recur­rence, and thus for patients with lim­ited-stage
dis­ease at relapse, radio­ther­apy may be an effec­tive option.39
Using radio­ther­apy in patients who have a PR pre-ASCT would
be an inter­est­ing strat­egy to increase the CMR rate, and stud­ies
inves­ti­gat­ing this approach are warranted. In addi­tion, the syn­
er­gis­tic effects of radi­a­tion with immu­no­ther­apy, as described
in a few case reports, should be inves­ti­gated more exten­sively
and could be an option for patients who relapse after ASCT or
are inel­i­gi­ble for ASCT.
Table 3. Overview of recently reported trial results incor­po­rat­ing BV or CPIs for patients after ≥1 line of sal­vage treat­ment

Study N Intervention CMR ORR PFS OS

Chen et al (2016)17 102 BV 34% 75% 5 years: 22% 5 years: 41%
O’Connor et al (2018) 29
65 BV + bendamustine 32% 71% 2 years: 50%* 2 years: 65%*
Armand et al (2018)18 243 Nivolumab 16% 69% 1 year: 50%* 1 year: 92%
Chen et al (2019)2 210 Pembrolizumab 27.6% 71.9% 2 years: 31.3% 2 years: 90.9%
Shi et al (2019)30 92 Sintilimab 34% 80.4% 6 month: 77.6% 6 month: 100%
Song et al (2019) 31
75 Camrelizumab 28% 76% 9 month: 76.6% 9 month: 96%*
Armand et al (2020)32 31 Pembrolizumab 19% 58% 2 years: 30% 2 years: 87%
Song et al (2020) 33
70 Tislelizumab 63.9% 87.1% 9 month: 74.5% 9 month: 98.6%

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/240/1851809/240driessen.pdf by guest on 13 December 2021

Diefenbach et al (2020)34 64 Ipilimumab + BV vs 60%/65%/84% 80%/94%/95% 1 year: 59%/77%/87% 1 year:
(includ­ing n = 26 first nivolumab + BV 90%*/80%*/95%*/
relapse) vs ipilimumab +
nivolumab + BV
Kuruvilla et al (2021)35 304 Pembrolizumab (n = 151) 25% vs 24% 66% vs 54% 2 years: 35% vs 25% —
vs BV (n = 152)
Liu et al (2021)36 61 Camrelizumab 79% vs 32% for 95% vs 89% 2 years: 67% vs 42% 2 years: 100%
monotherapy vs camrelizumab +
camrelizumab + decitabine vs
decitabine camrelizumab
*Data have been extracted from avail­­able Kaplan-Meier curves using WebPlot Digitizer.

General con­sid­er­ations Conflict-of-inter­est dis­clo­sure

In con­clu­sion, the pri­mary goal of treat­ment for patients with
R/R cHL is to achieve a CMR before HDCT/ASCT as this sig­nif­
i­cantly cor­re­lates with a favor­able out­come after ASCT. Using
a sequen­tial approach, treat­ment inten­sity and tox­ic­ity can be
reduced in a sub­set of fast-responding patients. Patients who are
inel­i­gi­ble for ASCT or relapse after ASCT can be treated with BV
or CPIs. There is a need to develop novel ther­a­pies to increase
response rates with­out increas­ing tox­ic­ity. One of the next goals
for clin­i­cal tri­als is to inves­ti­gate which patients can pos­si­bly
be cured with­out HDCT/ASCT. Risk-strat­i­fied and PET-adapted
pro­spec­tive stud­ies could help achieve this goal. Optimized risk
strat­i­fi­ca­tion and response eval­u­a­tion will guide future treat­ment
deci­sions and will help to find the right treat­ment for the right
patient.
Julia Driessen: travel sup­port from Takeda.
Sanne H. Tonino: no rel­e­vant dis­clo­sures.
Alison J. Moskowitz: hon­o­raria: Seattle Genetics; con­sul­ting or advi­
sory role: Seattle Genetics, Takeda, Imbrium Therapeutics, Merck,
Janpix, Kyowa Kirin International, miRagen, ADC Therapeutics, Bris-
tol Myers Squibb; research funding: Incyte, Seattle Genetics, Merck,
Bristol Myers Squibb, miRagen, ADC Therapeutics, BeiGene.
Marie José Kersten: hon­o­raria for con­sul­ting or advi­sory boards:
Kite/Gilead, Novartis, BMS/Celgene, Miltenyi Biotech, Research
funding Takeda, Roche, Celgene, Kite/Gilead.
Off-label drug use
Julia Driessen: nothing to disclose.
Sanne H. Tonino: nothing to disclose.
Alison J. Moskowitz: nothing to disclose.
Marie José Kersten: nothing to disclose.

CLINICAL CASE (Con­t in­u ed) Correspondence

Despite the ini­tial CMR after 2 cycles of DHAP, soon after the
third and last cycle of DHAP, B-symp­toms and itching returned
and a relapse was again con­firmed by PET-CT and a biopsy.
Given the poor prog­no­sis in this patient with chemorefrac-
tory dis­ease, com­bi­na­tion treat­ment with BV and nivolumab was
started based on the encour­ag­ing results of a phase 1/2 trial.10
Already after 1 cycle of BV-nivolumab, her B-symp­toms disap-
peared, and after 4 cycles, she reached a CMR. We decided to
pro­ceed to haploidentical ASCT, because the patient ini­ tially
progressed dur­ing sal­vage che­mo­ther­apy. As men­tioned above,
though, emerg­ing data on the role of ASCT after CPI for patients
refrac­tory to mul­ti­ple prior lines of che­mo­ther­apy sug­gest that
ASCT could be con­sid­ered in this set­ting as well.23
Prakash Singh Shekhawat
Salvage treatment in relapsed/refractory Hodgkin lymphoma  |  245
Marie José Kersten, Amsterdam UMC, University of Amsterdam,
Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands; e-mail:
m​­. j​­.kersten@amsterdamumc​­.nl.
References
1. von Tresckow B, Müller H, Eichenauer DA, et al. Outcome and risk fac­
tors of patients with Hodgkin lym­phoma who relapse or prog­ress after
autol­o­gous stem cell trans­plant. Leuk Lymphoma. 2014;55(8):1922-
1924.
2. Chen R, Zinzani PL, Lee HJ, et  al. Pembrolizumab in relapsed or refrac­
tory Hodgkin lym­phoma: 2-year fol­low-up of KEYNOTE-087. Blood. 2019;
134(14):1144-1153.
3. Kersten MJ, Driessen J, Zijlstra JM, et al. Combining brentuximab vedotin with
dexa­meth­a­sone, high-dose cytarabine and cis­platin as sal­vage treat­ment
in relapsed or refrac­tory Hodgkin lym­phoma: the phase II ­HOVON/LLPC
trans­plant BRaVE study. Haematologica. 2021;106(4):1129-1137.
4. Moskowitz AJ, Schöder H, Gavane S, et al. Prognostic sig­nif­i­cance of base­
line met­ ab
­ olic tumor vol­ume in relapsed and refrac­ tory Hodgkin lym­
phoma. Blood. 2017;130(20):2196-2203.
5. Herrera AF, Palmer J, Martin P, et al. Autologous stem-cell trans­plan­ta­tion
after sec­ond-line brentuximab vedotin in relapsed or refrac­tory Hodgkin
lym­phoma. Ann Oncol. 2018;29(3):724-730.
6. Cole PD, McCarten KM, Pei Q , et  al. Brentuximab vedotin with gemcit-
abine for pae­di­at­ric and young adult patients with relapsed or refrac­tory
Hodgkin’s lym­phoma (AHOD1221): a Children’s Oncology Group, multi-
centre sin­gle-arm, phase 1-2 trial. Lancet Oncol. 2018;19(9):1229-1238.
7. Broccoli A, Argnani L, Botto B, et al; Fondazione Italiana Linfomi ONLUS.
First sal­vage treat­ment with bendamustine and brentuximab vedotin in
Hodgkin lym­phoma: a phase 2 study of the Fondazione Italiana Linfomi.
Blood Cancer J. 2019;9(12):100.
8. Garcia-Sanz R, Sureda A, de la Cruz F, et  al. Brentuximab vedotin and
ESHAP is highly effec­tive as sec­ond-line ther­apy for Hodgkin lym­phoma
patients (long-term results of a trial by the Span­ish GELTAMO Group). Ann
Oncol. 2019;30(4):612-620.
9. LaCasce AS, Bociek RG, Sawas A, et al. Brentuximab vedotin plus benda-
mustine: a highly active first sal­vage reg­i­men for relapsed or refrac­tory
Hodgkin lym­phoma. Blood. 2018;132(1):40-48.
10. Advani R, Moskowitz AJ, Bartlett NL, et al. Brentuximab vedotin in com­bi­
na­tion with nivolumab in relapsed or refrac­tory Hodgkin lym­phoma: 3-year
study results. Blood. 2021;138(6):427-438.
11. Moskowitz CH, Matasar MJ, Zelenetz AD, et al. Normalization of pre-ASCT,
FDG-PET imag­ing with sec­ond-line, non-cross-resis­tant, che­mo­ther­apy
pro­ grams improves event-free sur­ vival in patients with Hodgkin lym­
phoma. Blood. 2012;119(7):1665-1670.
12. Myers RM, Hill BT, Shaw BE, et al. Long-term out­comes among 2-year sur­
vi­vors of autol­o­gous hema­to­poi­etic cell trans­plan­ta­tion for Hodgkin and
dif­fuse large b-cell lym­phoma. Cancer. 2018;124(4):816-825.
13. Josting A, Müller H, Borchmann P, et  al. Dose inten­sity of che­mo­ther­apy
in patients with relapsed Hodgkin’s lym­phoma. J Clin Oncol. 2010;28(34):
5074-5080.
14. Moskowitz CH, Yahalom J, Zelenetz AD, et al. High-dose chemo-radio­ther­
apy for relapsed or refrac­tory Hodgkin lym­phoma and the sig­nif­i­cance of
pre-trans­plant func­tional imag­ing. Br J Haematol. 2010;148(6):890-897.
15. Labrador J, Cabrero-Calvo M, Pérez-López E, et  al. ESHAP as sal­ vage
ther­apy for relapsed or refrac­tory Hodgkin’s lym­phoma. Ann Hematol.
2014;93(10):1745-1753.
16. Santoro A, Mazza R, Pulsoni A, et al. Five-year results of the BEGEV sal­vage
reg­i­men in relapsed/refrac­tory clas­si­cal Hodgkin lym­phoma. Blood Adv.
2020;4(1):136-140.
17. Chen R, Gopal AK, Smith SE, et al. Five-year sur­vival and dura­bil­ity results
of brentuximab vedotin in patients with relapsed or refrac­tory Hodgkin
lym­phoma. Blood. 2016;128(12):1562-1566.
18. Armand P, Engert A, Younes A, et  al. Nivolumab for relapsed/refrac­tory
clas­sic Hodgkin lym­phoma after fail­ure of autol­o­gous hema­to­poi­etic cell
trans­plan­ta­tion: extended fol­low-up of the multicohort sin­gle-arm phase II
checkmate 205 trial. J Clin Oncol. 2018;36(14):1428-1439.
19. Abuelgasim KA, Alzahrani M, Alsharhan Y, et al. Chemoimmunotherapy with
brentuximab vedotin com­bined with ifosfamide, gemcitabine, and vinorel-
bine is highly active in relapsed or refrac­tory clas­si­cal Hodgkin lym­phoma.
Bone Marrow Transplant. 2019;54(7):1168-1172.
20. Moskowitz AJ, Shah G, Schöder H, et  al. Phase II trial of pembrolizumab
plus gemcitabine, vinorelbine, and lipo­so­mal doxo­ru­bi­cin as sec­ond-line
ther­apy for relapsed or refrac­tory clas­si­cal Hodgkin lym­phoma [published
online 25 June 2021]. J Clin Oncol.
21. Schmitz N, Pfistner B, Sextro M, et al; Ger­man Hodgkin’s Lymphoma Study
Group; Lymphoma Working Party of the Euro­pean Group for Blood and
Marrow Transplantation. Aggressive con­ven­tional che­mo­ther­apy com­
pared with high-dose che­mo­ther­apy with autol­o­gous haemopoietic stem-
cell trans­plan­ta­tion for relapsed chemosensitive Hodgkin’s dis­ease: a
randomised trial. Lancet. 2002;359(9323):2065-2071.
22. Linch DC, Winfield D, Goldstone AH, et al. Dose inten­si­fi­ca­tion with autol­
o­gous bone-mar­row trans­plan­ta­tion in relapsed and resis­tant Hodgkin’s
246  |  Hematology 2021  |  ASH Education Program
dis­ ease: results of a BNLI randomised trial. Lancet. 1993;341(8852):
1051-1054.
23. Merryman RW, Redd RA, Nishihori T, et al. Autologous stem cell trans­plan­
ta­tion after anti-PD-1 ther­apy for mul­ti­ply relapsed or refrac­tory Hodgkin
lym­phoma. Blood Adv. 2021;5(6):1648-1659.
24. Bröckelmann PJ, Goergen H, Kohnhorst C, et al. Late relapse of clas­si­cal
Hodgkin lym­phoma: an anal­y­sis of the Ger­man Hodgkin Study Group HD7
to HD12 tri­als. J Clin Oncol. 2017;35(13):1444-1450.
25. Moskowitz CH, Nademanee A, Masszi T, et al; AETHERA Study Group.
Brentuximab vedotin as con­sol­i­da­tion ther­apy after autol­o­gous stem-cell
trans­plan­ta­tion in patients with Hodgkin’s lym­phoma at risk of relapse or
pro­gres­sion (AETHERA): a randomised, dou­ble-blind, pla­cebo-con­trolled,
phase 3 trial. Lancet. 2015;385(9980):1853-1862.
26. Connors JM, Jurczak W, Straus DJ, et al; ECHELON-1 Study Group. Brentux-
imab vedotin with che­mo­ther­apy for stage III or IV Hodgkin’s lym­phoma.
N Engl J Med. 2018;378(4):331-344.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/240/1851809/240driessen.pdf by guest on 13 December 2021
27. Armand P, Chen Y-B, Redd R-A, et al. PD-1 block­ade with pembrolizumab
for clas­si­cal Hodgkin lym­phoma after autol­o­gous stem cell trans­plan­ta­tion.
Blood. 2019;134(1):22-29.
28. Herrera AF, Chen L, Nieto Y, et al. Consolidation with nivolumab and bren-
tuximab vedotin after autol­o­gous hema­to­poi­etic cell trans­plan­ta­tion in
patients with high-risk Hodgkin lym­phoma. Blood. 2020;136(suppl 1):19-20.
29. O’Connor OA, Lue JK, Sawas A, et  al. Brentuximab vedotin plus benda-
mustine in relapsed or refrac­tory Hodgkin’s lym­phoma: an inter­na­tional,
multicentre, sin­gle-arm, phase 1-2 trial. Lancet Oncol. 2018;19(2):257-266.
30. Shi Y, Su H, Song Y, et al. Safety and activ­ity of sintilimab in patients with
relapsed or refrac­tory clas­si­cal Hodgkin lym­phoma (ORIENT-1): a multi-
centre, sin­gle-arm, phase 2 trial. Lancet Haematol. 2019;6(1):e12-e19.
31. Song Y, Wu J, Chen X, et  al. A sin­gle-arm, mul­ti­cen­ter, phase II study
of camrelizumab in relapsed or refrac­tory clas­si­cal Hodgkin lym­phoma.
Clin Cancer Res. 2019;25(24):7363-7369.
32. Armand P, Kuruvilla J, Michot J-M, et al. KEYNOTE-013 4-year fol­low-up of
pembrolizumab in clas­si­cal Hodgkin lym­phoma after brentuximab vedotin
fail­ure. Blood Adv. 2020;4(12):2617-2622.
33. Song Y, Gao Q , Zhang H, et al. Treatment of relapsed or refrac­tory clas­si­cal
Hodgkin lym­phoma with the anti-PD-1, tislelizumab: results of a phase 2,
sin­gle-arm, mul­ti­cen­ter study. Leukemia. 2020;34(2):533-542.
34. Diefenbach CS, Hong F, Ambinder RF, et  al. Ipilimumab, nivolumab, and
brentuximab vedotin com­bi­na­tion ther­a­pies in patients with relapsed or
refrac­tory Hodgkin lym­ phoma: phase 1 results of an open-label, multi-
centre, phase ½ trial. Lancet Haematol. 2020;7(9):e660-e670.
35. Kuruvilla J, Ramchandren R, Santoro A, et al; KEYNOTE-204 inves­ ti­
ga­
tors. Pembrolizumab ver­sus brentuximab vedotin in relapsed or refrac­
tory clas­si­cal Hodgkin lym­phoma (KEYNOTE-204): an interim anal­y­sis of a
multicentre, randomised, open-label, phase 3 study. Lancet Oncol. 2021;
22(4):512-524.
36. Liu Y, Wang C, Li X, et al. Improved clin­ic
­ al out­come in a ran­dom­ized phase II
study of anti-PD-1 camrelizumab plus decitabine in relapsed/refrac­tory Hod-
gkin lym­phoma. J Immunother Cancer. 2021;9(4):e002347.
37. Fukuhara N, Yamamoto G, Tsujimura H, et al. Retreatment with brentuximab
vedotin in patients with relapsed/refrac­tory clas­si­cal Hodgkin lym­phoma
or sys­temic ana­plas­tic large-cell lym­phoma: a mul­ti­cen­ter ret­ro­spec­tive
study. Leuk Lymphoma. 2020;61(1):176-180.
38. Manson G, Brice P, Herbaux C, et  al. Efficacy of anti-PD1 re-treat­ment in
patients with Hodgkin lym­phoma who relapsed after anti-PD1 dis­con­tin­
u­a­tion. Haematologica. 2020;105(11):2664-2666.
39. Constine LS, Yahalom J, Ng AK, et  al. The role of radi­ a­tion ther­
apy in
patients with relapsed or refrac­tory Hodgkin lym­phoma: guide­lines from
the inter­na­tional lym­phoma radi­a­tion oncol­ogy group. Int J Radiat Oncol
Biol Phys. 2018;100(5):1100-1118.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000311
HODGKIN LYMPHOMA: CURE AND OPTIMAL SURVIVORSHIP

Double-refractory Hodgkin lymphoma:

tackling relapse after brentuximab vedotin
and checkpoint inhibitors

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/247/1851712/247epperla.pdf by guest on 13 December 2021

Narendranath Epperla1 and Mehdi Hamadani2
Division of Hematology, Department of Medicine, The James Cancer Hospital and Solove Research Institute, Ohio State University, Columbus, OH;
1

and 2Blood and Marrow Transplant Program and Cellular Therapy Program, Department of Medicine, Medical College of Wisconsin, Milwaukee, WI

The approval of brentuximab vedotin (BV) and checkpoint inhibitors (CPI) has revolutionized the management of relapsed/
refractory classical Hodgkin lymphoma (cHL) patients. In recent years these agents have rapidly moved to earlier lines
of therapy, including post-autologous hematopoietic cell transplant (auto-HCT) consolidation, pre-HCT salvage, and the
frontline treatment setting. This shift in practice means that double-refractory (refractory to both BV and CPI) cHL is
becoming an increasingly common clinical problem. In patients who are not eligible for clinical trials, conventional cyto-
toxic and targeted therapies (off label) may be a potential option. In patients who are transplant eligible, early referral to
allogeneic HCT should be considered given the significant improvement in transplant outcomes in the contemporary era.
Cellular therapy options including CD30.chimeric antigen receptor T cells, Epstein-Barr virus-directed cytotoxic T cells,
and CD16A/30 bispecific natural killer cell engagers appear promising and are currently in clinical trials.

LEARNING OBJECTIVES
• Identify promising therapeutic agents for cHL refractory to both BV and checkpoint inhibitors
• Highlight contemporary outcomes of allogeneic transplantation in patients with heavily pretreated cHL
• Recognize emerging cellular therapies and summarize the results of clinical trials evaluating these therapies

CLINICAL CASE Introduction

A 32-year-old Caucasian man was diagnosed with stage IV-B
classical Hodgkin lymphoma (cHL). He achieved a complete
metabolic remission following six cycles of frst-line treat-
ment with adriamycin, vinblastine, bleomycin, and dacar-
bazine. The patient experienced a biopsy-proven relapse
9 months after fnishing this chemotherapy. At the time of
relapse, the patient had extranodal involvement in the axial
skeleton and liver. His disease remained refractory to second-
line treatment with ifosfamide, carboplatin, and etoposide.
The patient achieved a second complete metabolic remis-
sion with a brentuximab vedotin (BV) and nivolumab com-
bination and subsequently went on to receive autologous
hematopoietic cell transplantation (auto-HCT) followedby
BV consolidation. While on consolidation treatment, the
patient relapsed again and was initiated on pembrolizumab,
which resulted in a partial remission lasting approximately
18 months. At the current progression, the patient is ft, with
excellent organ function, and his donor search has identifed
a haploidentical sibling.
Prakash Singh
Although most patients with cHL are cured with initial che-
motherapy, up to 20% to 25% of patients will relapse or
have primary refractory (R/R) disease.1 While high-dose
therapy and an auto-HCT consolidation can cure cHL
patients responding to subsequent treatments, only about
half of these patients will attain durable disease control.2,3
The advent of novel agents, especially BV and checkpoint
inhibitors (CPI), in the past decade has not only revolution-
ized the management of R/R cHL patients; in recent years
these agents have rapidly moved to earlier lines of therapy,
including post-auto-HCT consolidation (Figure 1 depicts
the increased utilization of BV consolidation post-auto-HCT
in the United States based on reporting to the Center for
International Blood and Marrow Transplant Research [CIB-
MTR] registry),3 pre-HCT salvage,4 and front-line treatment
setting.5 This shift in practice means that the scenario de-
scribed in the clinical case (above) of a cHL patient with
double-refractory (refractory to both BV and CPI) disease
is becoming a common clinical problem. Unlike patients
Shekhawat
with relapsed non-Hodgkin lymphoma (NHL; median age in

Double-refractory Hodgkin lymphoma | 247

 No maintenance Brentuximab maintenance
2020 256 125
2019 390
Year of Transplant
2018 353
2017 309
2016 388
2015 596
2014 665
2013 677
2012 706
0 100 200 300
Frequency
Figure 1. Increased use of BV as post-autologous transplant consolidation therapy in the US. Red arrow: year of AETHERA trial
publication; 2020 data incomplete. Based on CIBMTR registry data.
the late 60s), the median age of relapsed cHL patients is typ­i­cally
in the 30s.6,7 The young age of these R/R patients means that in
addi­tion to selecting the next sal­vage option (typ­i­cally expected
to pro­vide short-term dis­ease con­trol), the treating team must
also con­sider ther­a­peu­tic modal­i­ties that can pro­vide dura­ble
dis­ease con­trol and hope­fully cure. The lat­ter may not be fea­si­
ble for the sub­set of dou­ble-refrac­tory patients with advanced
age, com­pro­mised organ func­tion, and/or poor per­for­mance
sta­tus.
In this arti­cle we dis­cuss the treat­ment options for dou­ble-
refrac­tory cHL, begin­ning with poten­tial sal­vage options and
followed by opti­mal con­sol­i­da­tion approaches. At the out­set we
acknowl­edge that there are no uni­ver­sally accepted def­i­ni­tions
of BV or CPI refrac­tory dis­ease. In this review, dou­ble-refrac­tory
dis­ease is loosely defined as cHL that does not achieve at least
a par­tial remis­sion or progresses while on BV and CPI (-based
treat­ments). It is impor­tant to remem­ber that cHL patients pre­
vi­ously achiev­ing remis­sion with either BV or CPI (or reg­i­mens
containing these drugs) and sub­se­quently relaps­ing after a treat­
ment-free inter­val may be can­di­dates for rechallenge with these
agents.
Salvage options in dou­ble-refrac­tory cHL
Conventional cyto­toxic ther­apy options
Clinical trial enroll­ment, if avail­­able, should be a pri­or­ity in dou­ble-
refrac­tory cHL. Off trial, in a fit patient con­ven­tional multiagent
che­mo­ther­apy as a bridge to allo­ge­neic (allo) HCT is a rea­son­
able option. In patients who are not can­di­dates for allo-HCT,
sin­
gle-agent cyto­ toxic ther­ apy with a pal­ li­
a­tive intent is rea­
son­able. The choice of che­mo­ther­apy (reg­i­men) is often empir-
ic and depends on the patient’s prior treat­ments, mar­row re-
serve, and organ func­tion. For instance, while both plat­i­num- or
gemcitabine-based ther­a­pies are active, plat­i­num-containing
reg­i­mens may be less desir­able in patients with chronic kid­ney
dis­ease, and a gemcitabine-based option could be chal­leng­
ing in patients with a poor mar­row reserve. A detailed dis­cus­
248  |  Hematology 2021  |  ASH Education Program
Other maintenance
33
235 95
316 74
305 102
233 64
106 38
20 35
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/247/1851712/247epperla.pdf by guest on 13 December 2021
5 34
713
400 500 600 700 800
sion of sal­vage reg­i­mens is beyond the scope of this review, but
Table 1 sum­ma­rizes the com­monly con­sid­ered sal­vage options in
dou­ble-refrac­tory cHL patients.8-17
Radiation ther­apy
In relapsed patients with lim­ited treat­ment options, sal­vage radi­
a­tion should be con­sid­ered. Limited ret­ro­spec­tive data show
effec­tive short-term dis­ease con­trol with radi­a­tion ther­apy,
par­tic­u­larly in patients with local­ized nodal dis­ease.18 Although
pre­clin­i­cal mod­els and case reports dem­on­strate syn­er­gism be-
tween radi­at­ion and CPIs,19-21 it is unclear if this is a fea­si­ble ap-
proach in dou­ble-refrac­tory cHL patients.
Investigational agents for dou­ble-refrac­tory cHL
Camidanlumab tesirine
Camidanlumab tesirine (cami) is an anti­ body-drug con­ ju­
gate
(pyrrolobenzodiazepine) directed against CD25. After impres­
sive anti­tu­mor activ­ity was noted in a phase 1, dose-esca­la­tion,
and dose-expan­sion study of cami in R/R cHL patients (Table
2),22 a phase 2 study was conducted using 45 µg/kg q3 week
dos­ing for two cycles followed by 30 µg/kg Q3 weeks (in cHL
with prior BV and CPI treat­ments). Cami was active in R/R cHL
patients based on the pre­lim­i­nary results (n  =  51), with an over­all
response rate (ORR) of 83% and a com­plete response (CR) rate
of 38%.23 Grade 3 or higher adverse events were reported in 63%
(n  =  32) of cHL patients.23 Of note, there were 3 (6.4%) patients
with Guillain-Barré syn­drome (GBS)/polyradiculopathy (n  =  1 with
grade 4 GBS, n  =  1 with grade 2 GBS, and n  =  1 with grade 2 radic-
ulopathy) that led to an enroll­ment pause. Following a review of
safety and effi­cacy data (by inde­pen­dent review), the enroll­ment
pause was lifted. The study recently com­pleted accrual, and final
results are not yet avail­­able, but the pre­lim­i­nary results with cami
are encour­ag­ing and may pro­vide an addi­tional ther­a­peu­tic op-
tion for patients with R/R cHL, par­tic­u­larly those with dual-refrac­
tory dis­ease. Strategies to mit­i­gate the auto­im­mune neu­ro­log­i­cal
tox­ic­ity of this agent will enhance its risk/ben­e­fit pro­file.
Table 1.  Conventional cyto­toxic ther­apy options in R/R cHL

Study Median age Median

Agent design (range), years Total, N Prior AHCT (n) ORR% (CR%) PFS/EFS (mo) Reference
Gemcitabine based
Gemcitabine II 35 (19-58) 23 0 39 (9) 6.7 Santoro et al8
GDP II 36 (19–57) 23 0 69 (17) NR Baetz et al9
GVD I/II 33 (19–83) 91 40 70 (19) 8.5* Bartlett et al10
IGEV II 30 (17-59) 91 NA 81 (54) 3-y FFP 53% Santoro et al11
Platinum based
ICE II 27 (12-59) 65 NA 88 (26) 58%† Moskowitz et al12

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/247/1851712/247epperla.pdf by guest on 13 December 2021

DHAP II 34 (21-64) 102 NA 89 (21) NR Josting et al13
ESHAP II 34 (18-66) 22 2 73 (40) 3-y DFS 27% Aparicio et al14
Vinca alka­loids
Vinblastine II 31 (23-48) 17 17 59 (12) 8.3 Little et al15
Vinorelbine II NR 24 4 50 (14) 6 Devizzi et al16
Alkylators
Bendamustine II 34 (21–75) 35 27 53 (33) 5.2 Moskowitz et al17
*Median EFS was not reached in the trans­plant-naive group, while it was 8.5 months in patients with prior trans­plant.
†At a median fol­low-up of 43 months, as ana­lyzed by intent to treat, the EFS was 58%.
ASCT, autol­o­gous hema­to­poi­etic cell trans­plan­ta­tion; DFS, dis­ease-free sur­vival; DHAP, dexa­meth­a­sone, high-dose cytarabine, cis­platin; EFS,
event-free sur­vival; ESHAP, etoposide, meth­yl­pred­nis­o­lone, high-dose cytarabine, cis­platin; FFP, free­dom from pro­gres­sion; GDP, gemcitabine,
dexa­meth­a­sone, cis­platin; GVD, gemcitabine, vinorelbine, lipo­so­mal doxo­ru­bi­cin; ICE, ifosfamide, carboplatin, etoposide; IGEV, ifosfamide,
gemcitabine, etoposide, vinorelbine; NA, not appli­ca­ble; NR, not reported; PFS, pro­gres­sion-free sur­vival.

Mammalian tar­get of rapamycin inhib­i­tors
Preclinical stud­ies have shown that the PI3K-Akt-mam­ma­lian tar­
get of rapamycin (mTOR) path­way plays an impor­tant role in the
growth and sur­vival of Reed-Sternberg cells.24 Everolimus is an
oral inhib­i­tor of mTOR that is active in R/R cHL (Table 2).25 Recent-
ly, com­bi­na­tion ther­a­pies with everolimus have been explored in
R/R cHL with encour­ag­ing pre­lim­i­nary activ­ity. In a phase 1/2
study of heavily pretreated cHL patients (n  =  15), the com­bi­na­
tion of everolimus and itacitinib (an oral Jak inhib­i­tor) pro­duced
an ORR and CR rate of 79% and 14%, respec­tively. Of note, 94%
(n  =  14) were dou­ble refrac­tory in the study. The nota­ble grade 3
or higher toxicities included throm­bo­cy­to­pe­nia (43%), neutrope-
nia (21%), infec­tion (7%), and hyper­ten­sion (7%).26
Immunomodulatory agents
Lenalidomide mod­u­lates the immune micro­en­vi­ron­ment in lym­
phoid malig­nan­cies by interacting with ubiquitin E3 ligase cere-
blon and degrading Ikaros tran­ scrip­
tion fac­
tors.27 Single-agent
lenalidomide has pro­duced an ORR rang­ing from 13% to 30%
across stud­ies in R/R cHL (Table 2).28,29 The major tox­ic­ity is hema­
to­log­i­cal adverse events, which are slightly higher with con­tin­u­
ous dos­ing.29 Combination approaches have also been stud­ied,
mainly with his­tone deacetylase (HDAC) inhib­i­tors asso­ci­ated with
sig­nif­i­cant tox­ic­ity with­out any improve­ment in response rates.30
HDAC inhib­i­tors
HDAC inhib­i­tors are epi­ge­netic mod­u­la­tors that induce cell death
in cHL cell lines by inhibiting STAT6-medi­ated Th-2 cyto­kine and
TARC (thy­mus and acti­va­tion-reg­u­lated chemokine) pro­duc­tion.31
While sin­gle-agent HDAC inhib­i­tors have lim­ited activ­ity in R/R
cHL (Table 2),32,33 the com­bi­na­to­rial approach with mTOR inhib­i­
tors seems to be prom­is­ing. In a study that com­bined vorinostat
with mTOR inhib­i­tors (everolimus and sirolimus) in heavily pre-
treated cHL (n  =  40), the ORR and CR rates were 55% and 33%,
respec­tively.34 While the study included patients with prior BV
fail­ure, none of the patients in the study received CPI.
Transplant and cel­lu­lar ther­apy options
Allogeneic HCT
Adoptive immu­no­ther­apy in the form of an allo-HCT offers a poten­
tially cura­tive option for heavily pretreated R/R cHL patients,
includ­ing those relaps­ing after a prior auto-HCT, BV, and CPI.6,35,36
Historically, allo-HCT in R/R cHL has been asso­ci­ated with high
rates of nonrelapse mor­tal­ity (NRM; 35%-45% at 1 year) and dis­ease
relapse (up to 50%).37 These rates (with an obvi­ous neg­a­tive im-
pact on refer­ral pat­terns) and the avail­abil­ity of novel agents in cHL
have led to a decline in the use of allo-HCT in cHL in the US in the
last decade (Figure 2A, CIBMTR reg­is­try data). However, in dual-
refrac­tory cHL, the role of allo-HCT war­rants reappraisal, espe­cially
when focus­ing on mod­ern out­come data from the CIBMTR show­
ing remark­able improve­ment in sur­vival out­comes (Figure 2B, CIB-
MTR reg­is­try data) for these cHL patients fol­low­ing allografting.
Unlike the his­tor­i­cal data in which NRM rates were almost 50% and
3-year over­all sur­vival (OS) 25% to 30%,37 in the con­tem­po­rary se-
ries the NRM rates have declined to approx­i­ma­tely 10%, and 3-year
OS is approx­i­ma­tely 60%.7 These improve­ments could be due to
bet­ter con­di­tion­ing approaches,7,38 advances in graft-ver­sus-host
dis­ease (GVHD) pre­ven­tion, and over­all advances in sup­port­ive

Prakash Singh Shekhawat

Double-refrac­tory Hodgkin lym­phoma | 249
Table 2.  Investigational ther­a­pies in R/R cHL

Study Median age Median

Agent design (range), years Total (N) Prior AHCT (n) ORR% (CR%) PFS/EFS (mo) Reference
Cami I* 38 (31-53) 57 NR 75 (44) NR Hamadani et al22
Cami II 36 (20-74) 51 31 83 (38) NR Herrera et al23
Everolimus II 32 (19-77) 57 38 46 (9) 8 Johnston et al25
Lenalidomide II 37 (18–74) 15 10 13 NR Kuruvilla et al28
Lenalidomide II 38 (20–83) 42 31 30 8.2 Fehniger et al29
Vorinostat II 42 (20-71) 25 11 4 7.2 Kirschbaum
et al32
Panobinostat II 32 (18-75) 129 129 27 (4) 6.9 Younes et al33

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/247/1851712/247epperla.pdf by guest on 13 December 2021

Idelalisib II 42 (21-80) 25 18 20 (4) 2.3 Gopal et al51
Ibrutinib† Retrosp 35 (26-72) 7 5 57 (43) NA Badar et al52
*Data shown for 30 µg/kg and 45 µg/kg cohorts.
†A clin­i­cal trial is ongo­ing inves­ti­gat­ing the role of ibrutinib in R/R cHL (NCT02824029).
NA, not appli­ca­ble; NR, not reported; Retrosp, ret­ro­spec­tive.

care and trans­plant prac­tices. In addi­tion, donor avail­abil­ity is no
lon­ger a bar­rier. Several stud­ies looked at the out­comes of alter­na­
tive donor allo-HCT for R/R cHL and reported excel­lent out­comes
and low NRM (1 year, 10%-15%) among patients who received
posttransplant cyclo­phos­pha­mide (PTCy)-based haploidentical
(haplo) allo-HCT.35,39
In cur­rent prac­tice, most of the cHL patients under­go­ing allo-
HCT have CPI expo­sure in the imme­di­ate pre-HCT period. Allo-
HCT in this set­ting (post CPI) may be asso­ci­ated with an increased
risk of early posttransplant com­pli­ca­tions, espe­cially severe acute
GVHD.40 It has been pos­tu­lated that this phe­nom­e­non is likely
due to the higher fre­ quency of interferon-γ-pro­ duc­ ing effec­
tor T cells and a more pro­nounced T helper 1 dif­fer­en­ti­a­tion in
patients exposed to CPI before allo-HCT.41 PTCy has been shown
to not only abro­ gate the CPI-induced immune acti­ va­tion but
also pro­mote the vig­or­ous recov­ery of reg­u­la­tory T cells, lead­
ing to immune tol­er­ance and thereby suppressing GVHD to mild
lev­els.41,42 In a recently published “real-world” anal­ y­
sis of cHL
patients who under­ went allo-HCT after CPI treat­ ments, those
who received haplo/PTCy had a lower cumulative incidence of
relapse (2 year  =  7%) and excel­lent OS (2 year  =  85%). The time
from CPI to allo-HCT was an impor­tant pre­dic­tor of GVHD risk,
wherein patients with a lon­ger time from CPI expo­sure to allo-
HCT (>80 days) had a decreased risk of severe acute GVHD.43
Given these improve­ments in recent years, early refer­ral to trans­
plant pro­grams must be con­sid­ered for fit cHL patients with dual-
refrac­tory dis­ease that responds to avail­­able sal­vage treat­ments.
Cellular ther­a­pies
Given the remark­able activ­ity of chi­me­ric anti­gen recep­tor mod­i­
fied T (CAR-T) cell ther­apy in R/R B-cell NHL, this approach is un-
der inves­ti­ga­tion in cHL. In a phase 1/2 study of heavily pretreated
cHL (median prior lines  =  7) includ­ing BV, CPI, auto-HCT, and/or
allo-HCT, CD30.CAR-T cell ther­apy was safe and well tol­er­ated.
The ORR was 72% in those receiv­ing fludarabine-based lympho-
depletion (n  =  31) with a CR rate of 59%,44 but 1-year pro­gres­sion-
free sur­vival was only 36%, underscoring the need to improve the
plat­form. Currently, a larger phase 2 study is ongo­ing to eval­u­ate
the effi­cacy of CD30.CAR-T cells in R/R cHL (NCT04268706).
Epstein-Barr virus (EBV)-directed cyto­toxic T cells (CTLs)
directed against LMP 1 and LMP 2 have activ­ity in EBV+ cHL.
EBNA1, LMP 1, and LMP 2 are attrac­tive tar­gets and have been
clin­i­cally tested.45,46 In patients with EBV-asso­ ci­ ated cHL and
NHL, CTLs with enhanced activ­ity against LMP 1/2 were tested
with prom­is­ing results, wherein a response to ther­apy was noted
in 13 of 21 patients with active dis­ease, includ­ing 11 patients
achiev­ing CR.46 Another small study used LMP-spe­ cific CTLs
expressing dom­i­nant-neg­a­tive transforming growth fac­tor β
recep­ tor to coun­ ter­act transforming growth fac­ tor β-asso­ ci­
ated immune sup­pres­sion in the tumor micro­en­vi­ron­ment with
good clin­i­cal results.47 Currently, a phase 1 study using EBV CTLs
expressing CD30 chi­ me­
ric recep­tor for CD30+ lym­ pho­ mas is
ongo­ing (NCT01192464).
AFM13 is a first-in-class innate cell engager that acti­vates
the immune sys­tem and tar­gets CD30+ hema­to­logic can­cers.
AFM13 induces the spe­ cific and selec­ tive kill­
ing of CD30+
cells by engag­ing CD16 on nat­u­ral killer (NK) cells and CD30
on the sur­face of the cHL cells (bispecific NK cell engagers,
or BIKEs). Given the min­ i­
mal sin­
gle-agent activ­ ity of AFM13
in R/R cHL (11.5%),48 it has been com­bined with CPI (AFM13+
pembrolizumab in CPI-naive R/R cHL) and with cyto­ kine-
acti­vated cord blood-derived NK cells.49 The pre­lim­i­nary safety
and effi­cacy data on first-in-human cord blood-derived NK
BIKEs (CD16A/CD30) in heavily pretreated cHL (n   =  4) are
pro­voc­a­tive. The ORR/CR rate was 100% and 50%, respec­
tively, among the 4 patients treated with this approach, with
no evi­dence of cyto­kine release syn­drome, neu­ro­tox­ic­ity,
or GVHD noted.50 Currently, the study is enroll­ing patients
(NCT04074746).
CLINICAL CASE (Con­tin­ued)
The patient in the clin­i­cal case achieved a CR with cami on a
clin­i­cal trial and went on to receive a reduced-inten­sity con­di­
tion­ing haplo-HCT from his sib­ling donor. At the last fol­low-up,
the patient was in remis­sion with mild active chronic GVHD.

250  |  Hematology 2021  |  ASH Education Program

 A 250
196 199 201
Number of Transplants 200 184 186
168 168
157
147
150 136
127
114 106 109
105
100

50

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/247/1851712/247epperla.pdf by guest on 13 December 2021

Year of Transplant

B 100 p<0.001

80
Probability, %

60

40

20
1990-1995 1996-2000 2001-2005
2006-2010 2011-2015 2016-2020
0
0 1 2 3 4 5
Years
Figure 2.  (A) Utilization of allo-HCT for cHL in the US. (B) Overall survival of patients undergoing allo-HCT in the US from 1990 to 2020.
Based on CIBMTR registry data.

Conclusions ceutical Company; Consultancy: Incyte Corporation; ADC Ther-

Although the advent of BV and CPI has changed the ther­a­peu­
tic land­scape of cHL man­age­ment, these drugs are not cura­tive
(in an R/R set­ting). As these agents are moved ear­lier into the
treat­ment lines, many more patients will become dou­ble refrac­
tory. Conventional cyto­ toxic and targeted ther­ a­
pies may be
con­sid­ered in these patients, who are oth­er­wise not eli­gi­ble for
clin­i­cal tri­als or allo-HCT. In patients who are trans­plant eli­gi­ble,
allo-HCT should be strongly con­ sid­ered given the sig­ nif­i­
cant
improve­ment in out­comes with allo-HCT in the con­tem­po­rary
era. Preliminary results with the cel­lu­lar ther­apy options that in-
clude CD30.CAR-Ts, EBV CTLs, and CD30 BIKEs appear prom­is­
ing and are cur­rently in clin­i­cal tri­als. As we con­tinue on our path
toward a greater under­stand­ing of the biol­ogy of cHL, we will
be ­able to iden­tify addi­tional ther­a­peu­tic tar­gets with the goal
of pro­vid­ing dura­ble responses while min­i­miz­ing tox­ic­ity in R/R
cHL patients.
Conflict-of-inter­est dis­clo­sure
Narendranath Epperla: reports Speaker’s Bureau: Verastem and
Beigene; Advisory Board: Karyopharm; Honorarium: Genzyme.
Mehdi Hamadani: research sup­ Prakash Takeda
port/funding: Singh Pharma-
Shekhawat
apeutics; Pharmacyclics, Omeros, Kite. Speaker’s Bureau: Sanofi
Genzyme, AstraZeneca, BeiGene.
Off-label drug use
Narendranath Epperla: camidanlumab tesirine, AFM13, lenalido-
mide, everolimus, vorinostat, idelalisib, ibrutinib, panabinostat.
Mehdi Hamadani: camidanlumab tesirine, AFM13, lenalidomide,
everolimus, vorinostat, idelalisib, ibrutinib, panabinostat.
Correspondence
Mehdi Hamadani, Blood and Marrow Transplant Program and
Cellular Therapy Program, Department of Medicine, Medical Col-
lege of Wisconsin, 9200 W Wisconsin Ave, Ste C5500, Milwau-
kee, WI 53226; e-mail: mhamadani@mcw​­.edu.
References
1. Johnson P, Federico M, Kirkwood A, et al. Adapted treat­ment guided by
interim PET-CT scan in advanced Hodgkin’s lym­ phoma. N Engl J Med.
2016;374(25):2419-2429.
2. Schmitz N, Pfistner B, Sextro M, et al; Ger­man Hodgkin’s Lymphoma Study
Group; Lymphoma Working Party of the Euro­pean Group for Blood and
­Marrow Transplantation. Aggressive con­ven­tional che­mo­ther­apy ­com­pared

Double-refrac­tory Hodgkin lym­phoma | 251

 with high-dose che­mo­ther­apy with autol­o­gous haemopoietic stem-​cell
trans­
­ plan­ta­
tion for relapsed chemosensitive Hodgkin’s dis­ ease: a ran-
domised trial. Lancet. 2002;359(9323):2065-2071.
3. Moskowitz CH, Nademanee A, Masszi T, et al; AETHERA Study Group.
Brentuximab vedotin as con­sol­i­da­tion ther­apy after autol­o­gous stem-cell
trans­plan­ta­tion in patients with Hodgkin’s lym­phoma at risk of relapse or
pro­gres­sion (AETHERA): a randomised, dou­ble-blind, pla­cebo-con­trolled,
phase 3 trial. Lancet. 2015;385(9980):1853-1862.
4. Advani R, Moskowitz AJ, Bartlett NL, et al. Brentuximab vedotin in com­bi­
na­tion with nivolumab in relapsed or refrac­tory Hodgkin lym­phoma: 3-year
study results. Blood. 2021;138(6):427-438.
5. Connors JM, Jurczak W, Straus DJ, et al; ECHELON-1 Study Group. Brentux-
imab vedotin with che­mo­ther­apy for stage III or IV Hodgkin’s lym­phoma. N
Engl J Med. 2018;378(4):331-344.
6. Badar T, Epperla N, Szabo A, et al. Trends in postrelapse sur­vival in clas­sic
Hodgkin lym­phoma patients after expe­ri­enc­ing ther­apy fail­ure fol­low­ing
auto-HCT. Blood Adv. 2020;4(1):47-54.
7. Ahmed S, Ghosh N, Ahn KW, et  al. Impact of type of reduced-inten­
sity con­di­tion­ing reg­i­men on the out­comes of allo­ge­neic haematopoi-
etic cell trans­plan­ta­tion in clas­si­cal Hodgkin lym­phoma. Br J Haematol.
2020;190(4):573-582.
8. Santoro A, Bredenfeld H, Devizzi L, et al. Gemcitabine in the treat­ment of
refrac­tory Hodgkin’s dis­ease: results of a mul­ti­cen­ter phase II study. J Clin
Oncol. 2000;18(13):2615-2619.
9. Baetz T, Belch A, Couban S, et al. Gemcitabine, dexa­meth­a­sone and cis­
platin is an active and non-toxic che­mo­ther­apy reg­i­men in relapsed or
refrac­tory Hodgkin’s dis­ease: a phase II study by the National Cancer Insti-
tute of Canada Clinical Trials Group. Ann Oncol. 2003;14(12):1762-1767.
10. Bartlett NL, Niedzwiecki D, Johnson JL, et al; Cancer Leukemia Group B.
Gemcitabine, vinorelbine, and pegylated lipo­so­mal doxo­ru­bi­cin (GVD),
a sal­vage reg­im­ en in relapsed Hodgkin’s lym­phoma: CALGB 59804. Ann
Oncol. 2007;18(6):1071-1079.
11. Santoro A, Magagnoli M, Spina M, et  al. Ifosfamide, gemcitabine, and
vinorelbine: a new induc­tion reg­i­men for refrac­tory and relapsed Hod-
gkin’s lym­phoma. Haematologica. 2007;92(1):35-41.
12. Moskowitz CH, Nimer SD, Zelenetz AD, et al. A 2-step com­pre­hen­sive high-
dose chemoradiotherapy sec­ond-line pro­gram for relapsed and refrac­tory
Hodgkin dis­ease: anal­y­sis by intent to treat and devel­op­ment of a prog­
nos­tic model. Blood. 2001;97(3):616-623.
13. Josting A, Rudolph C, Reiser M, et al; Participating Centers. Time-inten­si­fied
dexa­meth­a­sone/cis­platin/cytarabine: an effec­tive sal­vage ther­apy with
low tox­ic­ity in patients with relapsed and refrac­tory Hodgkin’s dis­ease.
Ann Oncol. 2002;13(10):1628-1635.
14. Aparicio J, Segura A, Garcerá S, et al. ESHAP is an active reg­i­men for relaps­
ing Hodgkin’s dis­ease. Ann Oncol. 1999;10(5):593-595.
15. Little R, Wittes RE, Longo DL, Wilson WH. Vinblastine for recur­rent Hod-
gkin’s dis­ease fol­low­ing autol­o­gous bone mar­row trans­plant. J Clin Oncol.
1998;16(2):584-588.
16. Devizzi L, Santoro A, Bonfante V, et al. Vinorelbine: an active drug for the
man­age­ment of patients with heavily pretreated Hodgkin’s dis­ease. Ann
Oncol. 1994;5(9):817-820.
17. Moskowitz AJ, Hamlin PA Jr, Perales MA, et  al. Phase II study of benda-
mustine in relapsed and refrac­ tory Hodgkin lym­ phoma. J Clin Oncol.
2013;31(4):456-460.
18. Goda JS, Massey C, Kuruvilla J, et al. Role of sal­vage radi­at­ion ther­apy for
patients with relapsed or refrac­tory Hodgkin lym­phoma who failed autol­o­
gous stem cell trans­plant. Int J Radiat Oncol Biol Phys. 2012;84(3):e329-e335.
19. Deng L, Liang H, Burnette B, et  al. Irradiation and anti-PD-L1 treat­
ment syn­er­gis­ti­cally pro­mote anti­tu­mor immu­nity in mice. J Clin Invest.
2014;124(2):687-695.
20. Weichselbaum RR, Liang H, Deng L, Fu YX. Radiotherapy and immu­no­ther­
apy: a ben­e­fi­cial liai­son? Nat Rev Clin Oncol. 2017;14(6):365-379.
21. Michot JM, Mazeron R, Dercle L, et al. Abscopal effect in a Hodgkin lym­
phoma patient treated by an anti-programmed death 1 anti­body. Eur J
Cancer. 2016;66(Octo­ber):91-94.
22. Hamadani M, Collins G, Caimi P, Samaniego F, Spira A, et  al. Camidan-
lumab tesirine in relapsed/refrac­tory lym­phoma: a phase 1, mul­ti­cen­ter,
open-label, dose-esca­la­tion, dose-expan­sion study. Lancet Haematol.
2021;8(6):e433-e445.
23. Herrera AF, Carlo-Stella C, Collins GP, et  al. Preliminary results of a phase
2 study of camidanlumab tesirine (cami), a novel pyrrolobenzodiazepine-​
based anti­body-drug con­ju­gate, in patients with relapsed or refrac­tory Hod-
gkin lym­phoma. Blood. 2020;136(suppl 1):21-23.
252  |  Hematology 2021  |  ASH Education Program
24. Dutton A, Reynolds GM, Dawson CW, Young LS, Murray PG. Constitutive
acti­va­tion of phos­pha­ti­dyl-inositide 3 kinase con­trib­utes to the sur­vival of
Hodgkin’s lym­phoma cells through a mech­a­nism involv­ing Akt kinase and
mTOR. J Pathol. 2005;205(4):498-506.
25. Johnston PB, Pinter-Brown LC, Warsi G, White K, Ramchandren R. Phase 2
study of everolimus for relapsed or refrac­tory clas­si­cal Hodgkin lym­phoma.
Exp Hematol Oncol. 2018;7(May):12.
26. Svoboda J, Barta SK, Landsburg DJ, et  al. Everolimus plus itacitinib in
relapsed/refrac­tory clas­si­cal Hodgkin lym­phoma: results of a phase I/II
inves­ti­ga­tor ini­ti­ated trial (EVITA study). Blood. 2020;136(suppl 1):20-21.
27. Kritharis A, Coyle M, Sharma J, Evens AM. Lenalidomide in non-Hodgkin
lym­phoma: bio­log­i­cal per­spec­tives and ther­a­peu­tic oppor­tu­ni­ties. Blood.
2015;125(16):2471-2476.
28. Kuruvilla J, Taylor D, Wang L, Blattler C, Keating A, Crump M. Phase II trial
of lenalidomide in patients with relapsed or refrac­tory Hodgkin lym­phoma.
Blood. 2008;112(11):3052-3052.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/247/1851712/247epperla.pdf by guest on 13 December 2021
29. Fehniger TA, Larson S, Trinkaus K, et al. A phase 2 mul­ti­cen­ter study of
con­tin­u­ous dose lenalidomide in relapsed or refrac­tory clas­si­cal Hodgkin
lym­phoma. Blood. 2012;120(21):1623.
30. Maly JJ, Chris­tian BA, Zhu X, et al. A phase I/II trial of panobinostat in com­
bi­na­tion with lenalidomide in patients with relapsed or refrac­tory Hodgkin
lym­phoma. Clin Lymphoma Myeloma Leuk. 2017;17(6):347-353.
31. Buglio D, Georgakis GV, Hanabuchi S, et al. Vorinostat inhib­its STAT6-medi­
ated TH2 cyto­kine and TARC pro­duc­tion and induces cell death in Hodgkin
lym­phoma cell lines. Blood. 2008;112(4):1424-1433.
32. Kirschbaum MH, Goldman BH, Zain JM, et al. A phase 2 study of vorinos-
tat for treat­ment of relapsed or refrac­tory Hodgkin lym­phoma: Southwest
Oncology Group Study S0517. Leuk Lymphoma. 2012;53(2):259-262.
33. Younes A, Sureda A, Ben-Yehuda D, et  al. Panobinostat in patients with
relapsed/refrac­tory Hodgkin’s lym­phoma after autol­o­gous stem-cell trans­
plan­ta­tion: results of a phase II study. J Clin Oncol. 2012;30(18):2197-2203.
34. Janku F, Park H, Call SG, et al. Safety and effi­cacy of vorinostat plus siroli-
mus or everolimus in patients with relapsed refrac­tory Hodgkin lym­phoma.
Clin Cancer Res. 2020;26(21):5579-5587.
35. Ahmed S, Kanakry JA, Ahn KW, et al. Lower graft-ver­sus-host dis­ease and
relapse risk in post-trans­plant cyclo­phos­pha­mide-based haploidentical ver­
sus matched sib­ling donor reduced-inten­sity con­di­tion­ing trans­plant for
Hodgkin lym­phoma. Biol Blood Marrow Transplant. 2019;25(9):1859-1868.
36. Shah NN, Hamadani M. Is there still a role for allo­ge­neic trans­plan­ta­tion in
the man­age­ment of lym­phoma? J Clin Oncol. 2021;39(5):487-498.
37. Sureda A, Robinson S, Canals C, et al. Reduced-inten­sity con­di­tion­ing com­
pared with con­ven­tional allo­ge­neic stem-cell trans­plan­ta­tion in relapsed
or refrac­tory Hodgkin’s lym­phoma: an anal­y­sis from the Lymphoma Work-
ing Party of the Euro­pean Group for Blood and Marrow Transplantation.
J Clin Oncol. 2008;26(3):455-462.
38. Ghosh N, Ahmed S, Ahn KW, et al. Association of reduced-inten­sity con­di­
tion­ing reg­i­mens with over­all sur­vival among patients with non-Hodgkin
lym­phoma under­go­ing allo­ge­neic trans­plant. JAMA Oncol. 2020;6(7):1011-
1018.
39. Martínez C, Gayoso J, Canals C, et al; Lymphoma Working Party of the Euro­
pean Group for Blood and Marrow Transplantation. Post-trans­plan­ta­tion
cyclo­phos­pha­mide-based haploidentical trans­plan­ta­tion as alter­na­tive to
matched sib­ling or unre­lated donor trans­plan­ta­tion for Hodgkin lym­phoma:
a Registry Study of the Lymphoma Working Party of the Euro­pean Society
for Blood and Marrow Transplantation. J Clin Oncol. 2017;35(30):3425-
3432.
40. Merryman RW, Kim HT, Zinzani PL, et al. Safety and effi­cacy of allo­ge­neic
hema­to­poi­etic stem cell trans­plant after PD-1 block­ade in relapsed/refrac­
tory lym­phoma. Blood. 2017;129(10):1380-1388.
41. Nieto JC, Roldán E, Jiménez I, et al. Posttransplant cyclo­phos­pha­mide after
allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion mit­i­gates the immune acti­va­
tion induced by pre­vi­ous nivolumab ther­apy. Leukemia. 2020;34(12):3420-
3425.
42. Ikegawa S, Meguri Y, Mizuhara K, et  al. Pretransplant nivolumab fur­
ther enhanced Treg expan­ sion after posttransplant cyclo­ phos­
pha­
mide:
another aspect for immune tol­er­ance by PTCy after nivolumab. Leukemia.
2021;35(3):929-931.
43. Merryman RW, Castagna L, Giordano L, et  al. Allogeneic trans­plan­ta­tion
after PD-1 block­ade for clas­sic Hodgkin lym­phoma. Leukemia. 2021;35(9):
2672-2683.
44. Ramos CA, Grover NS, Beaven AW, et al. Anti-CD30 CAR-T cell ther­apy in
relapsed and refrac­tory Hodgkin lym­phoma. J Clin Oncol. 2020;38(32):3794-
3804.
45. Icheva V, Kayser S, Wolff D, et  al. Adoptive trans­fer of Epstein-Barr virus
(EBV) nuclear anti­gen 1-spe­cific T cells as treat­ment for EBV reactivation
and lymphoproliferative dis­or­ders after allo­ge­neic stem-cell trans­plan­ta­
tion. J Clin Oncol. 2013;31(1):39-48.
46. Bollard CM, Gottschalk S, Torrano V, et al. Sustained com­plete responses
in patients with lym­phoma receiv­ing autol­o­gous cyto­toxic T lym­pho­
cytes targeting Epstein-Barr virus latent mem­brane pro­teins. J Clin Oncol.
2014;32(8):798-808.
47. Bollard CM, Tripic T, Cruz CR, et al. Tumor-spe­cific T-cells engineered to
over­come tumor immune eva­sion induce clin­i­cal responses in patients
with relapsed Hodgkin lym­phoma. J Clin Oncol. 2018;36(11):1128-1139.
48. Rothe A, Sasse S, Topp MS, et  al. A phase 1 study of the bispecific anti-
CD30/CD16A anti­body con­struct AFM13 in patients with relapsed or refrac­
tory Hodgkin lym­phoma. Blood. 2015;125(26):4024-4031.
49. Bartlett NL, Herrera AF, Domingo-Domenech E, et al. A phase 1b study of
AFM13 in com­bi­na­tion with pembrolizumab in patients with relapsed or
refrac­tory Hodgkin lym­phoma. Blood. 2020;136(21):2401-2409.
Prakash Singh Shekhawat
50. Rezvani K. CAR NK cells: a drive to the future of cell ther­apy. Paper pre-
sented at: AACR Annual Meeting 2021; April 10-15, 2021; vir­ tual online
meet­ing.
51. Gopal AK, Fanale MA, Moskowitz CH, et  al. Phase II study of idelalisib, a
selec­tive inhib­i­tor of PI3Kδ, for relapsed/refrac­tory clas­si­cal Hodgkin lym­
phoma. Ann Oncol. 2017;28(5):1057-1063.
52. Badar T, Astle J, Kakar IK, Zellner K, Hari PN, Hamadani M. Clinical activ­
ity of ibrutinib in clas­si­cal Hodgkin lym­phoma relaps­ing after allo­ge­neic
stem cell trans­plan­ta­tion is inde­pen­dent of tumor BTK expres­sion. Br J
Haematol. 2020;190(2):e98-e101.
© 2021 by The Amer­i­can Society of Hematology
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/247/1851712/247epperla.pdf by guest on 13 December 2021
DOI 10.1182/hema­tol­ogy.2021000256
Double-refrac­tory Hodgkin lym­phoma | 253
 HOW CAN WE ENSURE THAT EVERYONE WHO NEEDS A TRANSPLANT CAN GET ONE ?

Allogeneic hematopoietic cell transplantation

for older patients

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/254/1851535/254lin.pdf by guest on 13 December 2021

Richard J. Lin1 and Andrew S. Artz2
Memorial Sloan Kettering Cancer Center, New York, NY; and 2City of Hope National Medical Center, Duarte, CA
1

Hematologic malignances are more common and often higher risk in older patients. Allogeneic hematopoietic cell trans-
plantation (alloHCT) best enables long-term disease control for patients with poor risk or relapsed/refractory hema-
tologic malignancies such as acute myeloid leukemia, myelodysplastic syndromes, or myelofibrosis. Rates of alloHCT
among older patients, while still relatively low compared with younger patients, have risen sharply over the past decade.
Accumulating evidence supports alloHCT for patients ≥60 years of age relative to non-HCT therapies based on improved
overall and disease-free survival. However, a significant proportion of older adults have limitations characterized by geri-
atric assessment. A systematic process to evaluate and optimize older patients may improve decision making, transplant
outcomes, and alloHCT access. We present case-based studies to illustrate a stepwise and rational approach to proper
older patient evaluation, pretransplant optimization, and posttransplant care with attention to important geriatric issues
and quality of life.

LEARNING OBJECTIVES
• Describe access barriers to allogeneic hematopoietic cell transplantation for older adults
• Understand the role of GA, management, and optimization strategies for an older adult throughout the alloHCT
process

Introduction
Allogeneic hematopoietic cell transplantation (alloHCT)
remains the best­established curative option for many
patients with advanced hematologic malignancies, par­
ticularly myeloid neoplasms.1 In recent years, we have
witnessed significant advances in reducing transplant­
related mortality, manipulating graft­versus­leukemia
effect to prevent/treat relapse, and developing alloHCT
as a platform for novel cellular therapies.2,3 Older age
may have been the most formidable and important bar­
rier, representing the next frontier.4 The demographics
of blood cancer, especially myeloid malignancies, with
a median age of onset in the late 60s to early 70s and
frequently higher risk underscore the need.5 The era of
alloHCT is upon us; the Center for International Blood
and Marrow Transplantation Research (CIBMTR) reports
that patients aged ≥60 years comprised more than 40%
of adult alloHCT volume in the United States (Figure 1).6
In this review, we discuss unique challenges facing old­
er patients in alloHCT and strategies to improve their
outcomes.
CASE 1
Mr. RM is a 73­year­old man with coronary artery dis­
ease, hypertension, diabetes, and moderate obesity who
resides in a rural town with his wife and children in an
active lifestyle. One year ago, he initiated hypomethylat­
ing agent therapy through his local oncologist for newly
diagnosed high­risk, transfusion­requiring myelodysplas­
tic syndrome (MDS) with excess blasts. The MDS evolved
to acute myeloid leukemia (AML) 1 year later, prompting
induction with liposomal daunorubicin and cytarabine.
His treatment course was complicated by neutropenic
fever and bacteremia. A follow­up bone marrow biopsy
demonstrated complete remission. Should Mr. RM be
referred for consolidation alloHCT?
AlloHCT vs chemotherapy in older patients
Older patients, especially those in their 70s, face the unique
challenge of finite life expectancy that may be further con­
strained by medical comorbidities.7 AlloHCT for older pa­
tients with AML poses the dual dangers of complications

254 | Hematology 2021 | ASH Education Program


Figure 1.  Trends in alloHCT in the United States by increasing recipient age (N = total number of alloHCTs during each calendar year;
Transplant, % reflects the percentage of alloHCT in each age group by calendar year). Data generously provided by the Center for
International Blood and Marrow Transplant Research.
includ­ing death after alloHCT with­out relapse (nonrelapse mor­
tal­ity) and dis­ease relapse. As such, it is imper­a­tive that phy­si­cians
and patients weigh the ben­e­fits and risks of alloHCT vs nontrans­
plant approaches, ide­ally early in the treat­ment course. Sever­
al pop­u­la­tion-based stud­ies have shown that invari­ably, older
patients with inter­me­di­ate- or poor-risk AML (which com­prise
most newly diag­nosed AMLs in older patients) rarely sur­vive for
more than 5 years with­out an alloHCT.8,9 In a study com­par­ing
patients with AML aged ≥60 years treated with con­sol­i­da­tion
che­ mo­ ther­
apy alone in first com­ plete remis­sion in sev­ eral
national coop­er­a­tive tri­als vs sim­i­larly aged patients under­go­ing
alloHCT in first com­plete remis­sion from the con­tem­po­rary CIB­
MTR trans­plant reg­is­try,10 sur­vival was worse for alloHCT in the
first 9 months posttransplant rel­a­tive to con­sol­i­da­tion on tri­als.
However, after 5 years, alloHCT sig­nif­i­cantly benefited patient
over­all sur­vival (OS) at 28.6% vs 13.8% in the chemo-con­sol­i­da­
tion cohort (haz­ard ratio, 0.53; P < .0001). Table 1 high­lights sim­i­
lar find­ings from sev­eral reg­is­try stud­ies com­par­ing alloHCT with
nontransplant chemo-con­sol­i­da­tion tri­als for AML.11-13 In addi­tion,
3 pro­spec­tive, donor vs no-donor stud­ies for patients with AML
were published in abstract form, which also sup­ports alloHCT in
this pop­u­la­tion (Table 1). The most recently reported Blood and
Marrow Transplantation Clinical Trial Network (BMT CTN) 1102
pro­spec­tively stud­ied bio­log­i­cally assigned, newly diag­nosed
high-risk patients with MDS aged 60 to 75 years to alloHCT with
a matched donor vs hypomethylating ther­apy with­out alloHCT
in the absence of a matched donor; the pres­ence of a matched
donor con­ferred a 3-year OS advan­tage of 47.9% vs 26.6%.14
AlloHCT out­comes in older patients
Associated with many advances in trans­plan­ta­tion, the num­ber
and pro­por­tion of total alloHCT con­tinue to rise in patients aged
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/254/1851535/254lin.pdf by guest on 13 December 2021
≥60 years with hema­to­logic malig­nan­cies (Figures 1 and 2), fur­
ther stim­u­lated by wider donor avail­abil­ity, includ­ing haploiden­
tical, for most patients. Rashidi et al15 performed a meta-anal­y­sis
of 13 stud­ies of patients with AML 60 years and older who under­
went alloHCT. The 2-year relapse-free sur­vival and OS were 44%
and 45%, respec­tively, suggesting that alloHCT is a via­ble op­
tion for these patients. Similar find­ings were dem­on­strated for
patients with a vari­ety of hema­to­logic malig­nan­cies.16-18 Even
among patients older than 70 years, a recent CIBMTR anal­y­sis
showed accept­able if not prom­is­ing 2-year pro­gres­sion-free sur­
vival and OS of 32% and 39%, respec­tively, in het­ero­ge­neous
dis­eases, donor sources, and reg­i­mens.19 These data rein­force
that chro­no­logic age alone, at least up to 75 years, should not
exclude an older patient from alloHCT can­di­dacy. Rather, we
pro­pose the patient’s “phys­i­o­logic” age should be eval­u­ated,
along with a com­pre­hen­sive assess­ment of the patient’s goals
of care, qual­ity of life (QOL), and the eco­sys­tem, includ­ing care­
giv­ers, social sup­port sys­tem, finan­cial resources, and liv­ing sit­
u­a­tion (Figure 3).4,20 Although beyond the scope of this review,
even among reduced-inten­sity reg­i­mens, a range of trans­plant
inten­si­ties exist that must be indi­vid­u­al­ized based on patient
health and dis­ease risk.19-22 Furthermore, graft-ver­sus-host dis­
ease (GVHD) remains a major cause of mor­bid­ity and func­tional
impair­ment in this pop­u­la­tion, prompting con­sid­er­ation of lower
GVHD plat­forms (Figure 3).23,24
Transplant access bar­ri­ers for older patients
Referral bias and other bar­ri­ers limit access among older pa­
tients to alloHCT. A recent sys­temic review of 26 stud­ies showed
that chro­no­logic older age is the sin­gle most impor­tant bar­rier
to refer patients for alloHCT con­sid­er­ation.25 Specifically, opin­
ions dif­fer mark­edly among hema­tol­o­gists/oncol­o­gists, trans­
Allotransplant in older adults  | 255
 Table 1.  Multicenter stud­ies com­par­ing alloHCT to non-HCT con­sol­i­da­tion strat­eg
­ ies in older patients with AML or MDS

AlloHCT donor
Study N Age range Study design dis­ease Comparison groups Survival out­comes Comments
source/con­di­tion­ing
Farag et al. AlloHCT: 94 60-70 Retrospective, mul­ti­cen­ter AlloHCT: CIBMTR Registry Matched: 77% 3-year OS: OS ben­e­fit in all­ cyto­­
(2011)11 Non-HCT: 96 anal­y­sis Non-HCT: CT in CALGB Alternative: 23% AlloHCT: 37% ge­netic risk groups.
AML in CR1 RIC/NMA: 100% Non-HCT: 25%
3-year LFS:
AlloHCT: 32%
Non-HCT: 15%
Kurosawa et al. AlloHCT: 152 50-70 Retrospective, mul­ti­cen­ter AlloHCT: Jap­a­nese Registry Matched: 76% 3-year OS: 183 patients in the CT
(2011)12 Non-HCT: 884 anal­y­sis Non-HCT: Jap­a­nese Registry Alternative: 24% (15% AlloHCT: 62% group even­tu­ally
AML in CR1 cord) Non-HCT: 51% received HCT.
MA: 38% 3-year RFS: OS ben­e­fits in inter­me­
RIC/NMA: 62% AlloHCT: 56% di­ate-risk group.
Non-HCT: 29%
Versluis et al. AlloHCT: 97 ≥60 Retrospective anal­y­sis of AlloHCT: HOVON-SAKK tri­als Matched: 92% 5-year OS: Benefits are seen in
(2015)13 Non-HCT: 177 mul­ti­cen­ter tri­als Non-HCT: HOVON-SAKK tri­als Alternative: 8% AlloHCT: 35% both inter­me­di­ate-
None: 366 AML in CR1 RIC/NMA: 100% Non-HCT: 26% risk and adverse risk
5-year RFS: groups.

256  |  Hematology 2021  |  ASH Education Program

AlloHCT: 32%
Non-HCT: 20%
Ustun et al. AlloHCT: 431 60-77 Retrospective, mul­ti­cen­ter AlloHCT: CIBMTR Registry Matched: 65% 5-year OS: OS/DFS is worse in
(2019)10 Non-HCT: 211 anal­y­sis Non-HCT: CT in coop­er­a­ Alternative: 35% AlloHCT: 28.6% the first 9 months
AML in CR1 tive group tri­als (CALGB, (24% cord) Non-HCT: 13.8% for alloHCT. OS/DFS
SWOG, ECOG) MA: 29% 5-year DFS: ben­e­fits more for
RIC/NMA: 71% AlloHCT: 23.7% poor-risk patients.
Non-HCT: 11.1%
*Niederwieser AlloHCT: 150 (donor) 50-75 Prospective, intent-to-treat, AlloHCT: Donor group Matched: 79% 9-year LFS: Benefits seen across
et al. (2016) Non-HCT: 205 (no donor vs no-donor trial Non-HCT: No donor con­sol­i­ Mismatched: 21% Donor: 25% both inter­me­di­
ASCO donor) AML in CR1 da­tion group RIC/NMA: 100% No donor: 14% ate- and high-risk
Abstract 9-year RI: groups.
e18501 Donor: 42%
No donor: 78%
*Brune et al. AlloHCT: 77 (donor) 50-70 Prospective, intent-to-treat, AlloHCT: Donor group Matched: 100% 3-year OS: 7 patients in con­trol
(2018) ASH Non-HCT: 68 (no donor vs no-donor trial Non-HCT: No donor con­sol­i­ RIC/NMA: 100% Donor: 45% group crossed over.
Abstract 205 donor) AML in CR1 da­tion group No donor: 48% Extremely high relapse
3-year RFS: rate in both arms
Donor: 40% (49% vs 60%)
No donor: 35%
*Foran et al. AlloHCT: 135 (donor) ≥60 Prospective, intent-to-treat, AlloHCT: Donor group Matched: 100% Median OS: 44 patients in the
(2018) EHA Non-HCT: 225 (no donor vs no-donor trial Non-HCT: No donor con­sol­i­ RIC/NMA: 100% Donor: 22.1 months donor group and
Abstract donor) AML in CR1 da­tion group No donor: 13.4 months 25 patients in the
S857 (ECOG (P = .013) no-donor group
E2906) crossed over.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/254/1851535/254lin.pdf by guest on 13 December 2021

 Table 1.  Multicenter stud­ies com­par­ing alloHCT to non-HCT con­sol­i­da­tion strat­e­gies in older patients with AML or MDS (Continued)

AlloHCT donor
Study N Age range Study design dis­ease Comparison groups Survival out­comes Comments
source/con­di­tion­ing
Nakamura et al. AlloHCT: 260 (donor) 50-75 Prospective, intent-to-treat, AlloHCT: Donor group Matched: 100% 3-year OS: Similar ben­e­fits in 3-
(2021) (BMT Non-HCT: 124 (no donor vs no-donor trial Non-HCT: No donor con­sol­i­ RIC/NMA: 100% Donor: 47.9% year LFS.
CTN 1102)14 donor) MDS IPSS Intermediate- da­tion group No donor: 26.6% No decrease in QOL.
2/high risk 3-year OS (as treated
anal­y­sis):

Prakash Singh Shekhawat

AlloHCT: 47.4%
Non-HCT: 16%
*Meeting abstract only. ASCO, American Society of Clinical Oncology; ASH, American Society of Hematology; CALGB, Cancer and Leukemia Group B; CR1, first complete remission; CT,
chemotherapy consolidation; DFS, disease-free survival; ECOG, Eastern Cooperative Oncology Group; EHA, European Hematology Association; HOVON-SAKK, Dutch-Belgian Hemato-Oncology
Cooperative Group and the Swiss Group for Clinical Cancer Research; IPSS, International Prognostic Scoring System; LFS, leukemia-free survival; MA, myeloablative conditioning; RFS, relapse
free survival; RI, relapse incidence; RIC/NMA, reduced-intensity/nonmyeloablative conditioning; SWOG, Southwest Oncology Group.

Allotransplant in older adults  | 257

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/254/1851535/254lin.pdf by guest on 13 December 2021

Figure 2. Trends in alloHCT in the United States for patients 76 years or older (N = total number of transplants). Data generously
provided by the Center for International Blood and Marrow Transplant Research.
plant phy­si­cians, and trans­plant cen­ters regard­ing the upper age
limit for alloHCT, likely as a result of indi­vid­ual expe­ri­ence and
exper­tise.26,27 Routine aging assess­ment could neu­tral­ize het­ero­
ge­ne­ity in opin­ion; how­ever, the lack of stan­dard­ized geri­at­ric
assess­ment (GA) tools and resources to accom­plish them chal­
lenges phys­i­o­logic aging eval­u­a­tion.27 Other noted fac­tors hin­
der­ing access included non­white eth­nic ori­gin, insur­ance sta­tus,
higher comorbidities, and lower socio­eco­nomic sta­tus. Given
recent advances in trans­plan­ta­tion using alter­na­tive donors such
as haploidentical and mismatched donors, lack of a matched do­
nor should not be exclu­sion­ary even among older patients.28,29
There are sev­eral poten­tial mit­i­ga­tion strat­e­gies to reduce
access bar­ ri­
ers. First and fore­
most, dis­ ease indi­ ca­
tions for
alloHCT should be clearly defined for older patients to sup­ple­
ment stan­dard alloHCT guide­lines,30 account­ing for worse out­
comes for AML, MDS, and acute lym­pho­blas­tic leu­ke­mia even in
the same dis­ease risk group. Rather than a dichot­o­mous sin­gle
deci­sion point of “fit” or “not fit” for trans­plant, we rec­om­mend
expe­dited refer­ral for alloHCT eval­u­a­tion in the appro­pri­ate dis­
ease indi­ca­tions for patients 60 years or older in the pres­ence of
ade­quate base­line func­tional sta­tus and with­out severe organ
comorbidities (Figure 4). We must strive to enroll patients aged
>75 years on alloHCT stud­ies; until then, the deci­sion must be
indi­vid­u­al­ized in this age group. Figure 2 quantifies the lim­ited
appli­ca­tion of alloHCT in this cohort but also the sub­ stan­tial
258  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/254/1851535/254lin.pdf by guest on 13 December 2021
increase in uti­ li­
za­tion. Second, we should explore inno­ va­ tive
approaches to incor­po­rate phys­i­o­logic aging eval­u­a­tion by
GA in the rou­tine care of older patients, such as an embed­ded
geri­at­ric hema­tol­ogy clinic and telemedicine plat­form.31,32 Last,
we must invest in greater edu­ca­tional and out­reach efforts to
raise aware­ness of the emerg­ing, prom­is­ing alloHCT out­come
data, the role of GA, and clin­i­cal trial oppor­tu­ni­ties spe­cif­i­cally
designed for older patients.26
CASE 1 (Continued)
Mr. RM had sev­eral telemedicine vis­its with the trans­plant phy­si­
cian, a clin­i­cal nurse coor­di­na­tor, and a social worker, all­located
at an aca­demic med­i­cal cen­ter 200 miles away. Cognizant of his
comorbid con­di­tions, nec­es­sary eval­u­a­tion, and poten­tial early
loss of QOL from alloHCT, Mr. RM and his fam­ily expressed a desire
to pro­ceed. He also com­pleted a remote, video-assisted GA,
which dem­on­strated pre­served self-reported func­tional sta­tus,
mobil­ity, and cog­ni­tion. In par­al­lel, the unre­lated donor search
proceeded, iden­ ti­
fy­
ing a young matched unre­ lated donor.
During che­mo­ther­apy con­sol­i­da­tion locally, he under­went pre­
transplant test­ing also through his local oncol­o­gist. Four weeks
later, he began a reduced-inten­sity trans­plant reg­i­men inclu­sive
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/254/1851535/254lin.pdf by guest on 13 December 2021
Figure 3.  Solving the puzzle of alloHCT for older patients with hematologic malignancies. MRD, measurable residual disease; PPx,
prophylaxis.

Figure 4.  How we perform alloHCT for an older patient with a hematologic malignancy. HLA, human leukocyte antigen. *Severe
comorbidities: New York Heart Association class 4 heart failure, severe renal dysfunction or end-stage renal disease on dialysis, Child
class C liver cirrhosis, Gold stage 4 chronic obstructive lung disease, metastatic solid tumor, dementia, or any comorbidity signifi­
cantly limiting life expectancy.
Prakash Singh Shekhawat
Allotransplant in older adults  | 259
of posttransplant cyclo­phos­pha­mide for GVHD pro­phy­laxis on
the BMT CTN 1301 Progress 3 trial (NCT02345850) from a young
well-matched unre­lated donor.
GA in alloHCT
The shift from fit­ness alone to assessing resilience to dis­ease-
related and trans­plant-related stress­ors broad­ens interventional
oppor­tu­ni­ties that may widen access (Figure 3). The term resil­
iency encompasses both the intrin­sic, “phys­i­o­logic” aging pro­
cess and the extrin­sic “eco­sys­tem,” includ­ing care­giver, social
sup­port, finance, and resources; GA com­bined with stan­dard
trans­plant psy­cho­so­cial eval­u­a­tion achieves this goal.4 GA is a
mul­ti­dis­ci­plin­ary diag­nos­tic pro­cess that identifies med­i­cal,
func­tional, and psy­cho­so­cial lim­i­ta­tions of an older per­son and
place him or her on a con­tin­u­ous spec­trum of fit­ness, vul­ner­a­bil­
ity, and frailty and fur­ther informs a mul­ti­dis­ci­plin­ary care plan
to max­i­mize healthy aging, as illus­trated in the fol­low­ing case.33
CASE 2
Mrs. LK is a 70-year-old woman who had stage II early breast
can­cer 2 years ago that was treated with sur­gery, radi­a­tion,
and adju­vant che­mo­ther­apy with no evi­dence of dis­ease, mod­
er­ate chronic obstruc­tive pul­mo­nary dis­ease, oste­o­ar­thri­tis,
and atrial fibril­la­tion. She sought treat­ment from her pri­mary
care phy­si­cian for fatigue and was found to have pan­cy­to­pe­
nia with periph­eral blasts. A bone mar­row biopsy spec­i­men
established the diag­ no­sis of AML har­ bor­ing a mono­ somy 7.
Due to her comorbidities, low-inten­sity induc­tion com­menced
with azacitidine and venetoclax, which was com­pli­cated only
by ongo­ing cytopenia. Repeat bone mar­row eval­u­at­ion after
1 cycle dem­on­strated com­plete remis­sion but with pos­i­tive
mea­sur­able resid­ual dis­ease by mul­ti­color flow cytom­e­try and
cyto­ge­net­ics. She was referred for trans­plant con­sul­ta­tion. The
GA revealed depen­dence in sev­eral instru­men­tal activ­i­ties of
daily liv­ing, recently depressed mood, and a screen­ing test
pos­i­tive for mild cog­ni­tive impair­ment. She would like to pur­
sue cura­tive-intent alloHCT con­sol­id ­ a­tion if pos­si­ble. She has a
highly sup­port­ive fam­ily and care­giver who con­cur and under­
stand that trans­plant tox­ic­ity may be pro­hib­i­tive, espe­cially
con­sid­er­ing the GA-defined def­i­cits and comorbidities. What is
the appro­pri­ate next step?
GA domains affect alloHCT out­comes
Physiologic aging established through GA, cou­pled with antic­i­
pated stress­ors of the dis­ease and treat­ments, begins to paint a
pic­ture of phys­i­cal resilience. Serial GA may enrich under­stand­ing
of resilience or “bounce back” after treat­ment. In the con­text of
alloHCT, the GA should address the extrin­sic eco­sys­tem, includ­
ing psy­cho­so­cial sup­port, care­giver sup­port, and resources for
alloHCT (Figure 3). Artz and col­leagues conducted the ini­tial pi­
lot study of GA in alloHCT and found sig­nif­i­cant asso­ci­a­tions of
pretransplant geri­at­ric impair­ments in func­tion and mobil­ity with
adverse sur­vival out­comes fol­low­ing alloHCT.34,35 Subsequently,
sev­eral groups inde­pen­dently val­i­dated these find­ings and found
260  |  Hematology 2021  |  ASH Education Program
addi­tional, prognostically impor­tant domains such as cog­ni­tion,
med­ic ­ a­tion, and frailty scales. These stud­ies are sum­ma­rized in
Table 2.36-44 The ongo­ing BMT CTN 1704 trial (Composite Health
Assessment Risk Model [CHARM]) is a large national study pro­
spec­tively using a stan­dard GA and other mea­sures prior to al­
loHCT among patients ≥60 years old, aiming to con­firm these
find­ings and/or iden­tify addi­tional risk fac­tors (NCT03992352).
CASE 2 (Continued)
The trans­ plant team recommended short-term defer­ ral to
address GA-defined def­i­cits while con­tinu­ing che­mo­ther­apy
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/254/1851535/254lin.pdf by guest on 13 December 2021
to deepen dis­ease response. Mrs. LK under­went reha­bil­it­ a­tive
ther­apy with phys­i­cal and occu­pa­tional ther­apy supplemented
by home walk­ing and strength­en­ing super­vised by her fam­
ily. The res­o­lu­tion of trans­fu­sion-depen­dent ane­mia fur­ther
boosted phys­i­cal recov­ery. The geri­at­rics team man­aged poly­
pharmacy by actively deprescribing non­es­sen­tial med­i­ca­tions
thought to con­trib­ute to the mild cog­ni­tive def­i­cits. Repeated
GA 2 months later dem­on­strated improved func­tional sta­tus
and cog­ni­tion (no lon­ger in the impaired range). Depressive
symp­toms resolved with more social engage­ment and phys­ic ­ al
inde­pen­dence. Based on these results and another informed
dis­cus­sion, the patient and trans­plant team elected to pro­
ceed. She sub­se­quently under­went reduced-inten­sity con­di­
tion­ing alloHCT using her 36-year-old haploidentical son with
posttransplant cyclo­phos­pha­mide for pre­ven­tion of GVHD.
The patient had a care­giver starting the day before trans­plant
infu­sion and con­tinu­ing through­out. The trans­plant admis­sion
was com­pli­cated by an epi­sode of delir­ium ini­tially rec­og­nized
by the care­giver. After exclud­ing organic causes, she received
hal­o­per­i­dol as needed, and occu­pa­tional ther­apy pre­scribed
inten­ sive cog­ ni­
tive exer­ cises. She was discharged on post­
transplant day +37 to home with a walker and a home exer­cise
reg­i­men, avoiding a sub­acute reha­bil­i­ta­tion facil­ity. She con­tin­
ued “vir­tual” clin­ics vis­its and phys­i­cal face-to-face encoun­ters
and ongo­ing reha­bil­i­ta­tion.
Geriatric man­age­ment and opti­mi­za­tion
While GA may uncover vulnerabilities in older patients con­sid­
er­ing alloHCT, how best to opti­mize patients prior to alloHCT
remains a work in prog­ress. Challenges include short time avail­­
able before alloHCT due to the pace of dis­ease and delayed
refer­ral, nonmodifiable def­i­cits such as comorbidities, and lim­
ited insti­tu­tional resources. Low-inten­sity inter­ven­tions would
be ideal; how­ ever, the BMT CTN conducted a mul­ ti­
cen­
ter,
ran­dom­ized study of struc­tured home exer­cise and a stress man­
age­ment pro­gram prior to trans­plan­ta­tion, find­ing no improve­
ment in phys­i­cal and men­tal func­tion­ing posttransplant.45 While
not lim­ited to older patients, this accen­tu­ated the need for tar­
geted and/or more inten­sive pretransplant opti­mi­za­tion. Recent­
ly, Derman, Artz and col­leagues46 conducted the first pilot study
apply­ing GA-guided inter­ven­tions in a mul­ti­dis­ci­plin­ary team
clinic (MDC) to opti­mize patients prior to trans­plant. They found
that, com­pared to his­tor­i­cal cohorts with sim­i­lar dis­ease and
trans­plant char­ac­ter­is­tics, the MDC cohort expe­ri­enced fewer
 Table 2.  Studies illus­trat­ing prognostically impor­tant geri­at­ric def­i­cits in older patients under­go­ing alloHCT

Impairment in GA domains with impact on out­comes

Age, median
Study N (range), y Study design dis­ease Comorbidity Function Mobility Cognition Medication Frailty scale
35
Muffly et al. (2014) 203 58 (54-63) Prospective, sin­gle-cen­ter pilot study NRM ↑ NRM ↑ OS ↓
All dis­eases OS ↓ OS ↓
Deschler et al. (2018)36 106 66 (60-78) Prospective, sin­gle-cen­ter trial NRM ↑ OS ↓ OS ↓
Myeloid neo­plasms PFS ↓ PFS ↓
Huang et al. (2020)37 148 62 (50-76) Prospective, sin­gle-cen­ter pilot study OS ↓
All dis­eases PFS ↓
NRM ↑
Toxicities ↑
Pamukcuoglu et al. 52 59 (40-73) Prospective, sin­gle-cen­ter pilot study OS ↓
(2019)40 All dis­eases Toxicities ↑

Prakash Singh Shekhawat

Salas et al. (2021)41 168 58 (19-77) Prospective, sin­gle-cen­ter pilot study NRM ↑ NRM ↑
All dis­eases OS ↓
Olin et al. (2020)38 330 63 (50-77) Multicenter, CIBMTR reg­is­try NRM ↑ NRM ↑
All dis­eases OS ↓
Polverelli et al. (2020)42 228 64 (60-76) Two-cen­ter ret­ro­spec­tive study NRM ↑
OS ↓
Lin et al. (2020)39 457 66 (60-79) Single-cen­ter, ret­ro­spec­tive study NRM ↑ NRM ↑
OS ↓
PFS ↓
Bhargava et al. (2020)43 114 68 (65-75) Single-cen­ter, ret­ro­spec­tive study Toxicities ↑
OS ↓
Sugidono et al. (2021)44 148 62 (50-76) Single-cen­ter, ret­ro­spec­tive study OS ↓
NRM, nonrelapse mor­tal­ity; PFS, pro­gres­sion-free sur­vival.

Allotransplant in older adults  | 261

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/254/1851535/254lin.pdf by guest on 13 December 2021
inpa­tient deaths, shorter length of stay, fewer dis­charges to a
skilled nurs­ing facil­ity, and improved sur­vival. The crit­i­cal com­
po­nents of the MDC approach likely involve more care­ful patient
selec­tion, targeted opti­mi­za­tion, and mul­ti­dis­ci­plin­ary col­lab­o­
ra­tion.46,47 In addi­tion, early rec­og­ni­tion, espe­cially of eco­sys­tem
bar­ri­ers, through rou­tine eval­u­a­tion best affords oppor­tu­ni­ties to
opti­mize. Telehealth and a shared care model, for exam­ple, may
alle­vi­ate dis­tance bar­ri­ers for rou­tine pre- and posttransplant vis­
its, at least when patients can safely reside at home.47 GA-guided
man­age­ment and inte­gra­tion of geri­at­ric prin­ci­ples of care
should not be lim­ited to pretransplant care. The devel­op­ment of
geri­at­ric syn­dromes of func­tional decline, fall, delir­ium, and cog­
ni­tive impair­ment posttransplant is not uncom­mon, and these
syn­dromes are asso­ci­ated with impaired sur­vival and QOL.24,48,49
In addi­tion, dis­charge to a reha­bil­i­ta­tion facil­ity posttransplant
has been shown to be a marker of poor sur­vival.50 These issues
require fur­ther pro­spec­tive val­i­da­tion with patient-cen­tric out­
comes of func­tion and QOL.
How we per­form alloHCT in an older patient
We sum­ma­rize our approach to alloHCT in an older patient with
hema­to­logic malig­nancy in Figure 4, work­ing toward suc­cess­ful
com­ple­tion of a final check­list. We rec­om­mend that hema­tol­o­
gists, patients, and insti­tu­tions first con­sider the “ABCDE” to tri­
age (early) refer­ral. We believe resil­iency mea­sure­ment, through
GA or equiv­a­lent, is essen­tial in older can­di­dates to sup­ple­ment
stan­dard pre-alloHCT test­ing and the sub­jec­tive “fit­ness” cri­te­
ria. We advo­cate a col­lab­or­ a­tive model partnering the trans­plant
team and the dis­ease man­age­ment team (when sep­a­rate) to har­
mo­nize dis­ease ther­apy with antic­i­pated trans­plant tim­ing, often
dic­tated by donor avail­abil­ity. Disease treat­ment may occur dis­
tant from the trans­plant cen­ter, par­tic­u­larly as a range of lower-
inten­sity treat­ments exist for com­mon alloHCT indi­ca­tions. This
shared care model ensures more uni­form mes­sag­ing to patients
related to alloHCT plans, risks, and ben­e­fits from all­phy­si­cians.
Shared care pro­motes the par­al­lel pro­cess of max­i­miz­ing resil­
ience through GA-targeted inter­ven­tions and prep­a­ra­tion of the
supporting eco­sys­tem dur­ing dis­ease treat­ment. Alignment of
these pro­ cesses facil­i­
tates meet­ ing a “final check­list” before
alloHCT (Figure 4). We acknowl­edge that not all­older patients
who embark on this pro­cess will ulti­mately pur­sue alloHCT be­
cause of dis­ease relapse, inad­e­quate resilience, and/or changes
in goals of care, underscoring the value of mul­ti­ple touch points
to dis­cuss patient goals and recalibrate patient expec­ta­tions
about the like­li­hood of meet­ing the final check­list.
Conclusion and future direc­tions
We rec­om­mend alloHCT as a stan­dard of care option for older
patients with high-risk hema­ to­
logic malig­ nan­cies best estab­
lished for AML and MDS. Not only has uti­li­za­tion in older patients
risen mark­edly, but out­comes in older patients also con­tinue
to improve due to incor­po­ra­tion of novel trans­plant plat­forms
with reduced toxicities, an increased donor pool, and the bet­ter
selec­tion and care of older trans­plant patients. Moreover, we are
begin­ning to appre­ci­ate the impact of aging biol­ogy on trans­
plant out­comes and to explore mech­a­nism-based, ther­a­peu­tic
inter­ven­tions to tar­get aging path­ways.51 The con­ver­gence of
suc­cess in dis­ease-based ther­a­pies, edu­ca­tion to address age
262  |  Hematology 2021  |  ASH Education Program
mis­con­cep­tions, novel inter­ven­tions to bol­ster patient resilience,
and trans­plant reg­i­mens prom­ises more wide­spread and more
suc­cess­ful appli­ca­tion of alloHCT for older adults with high-risk
hema­to­logic malig­nan­cies.
Acknowledgments
R.J.L. acknowl­edges the sup­port from the ASH Clinical Research
Training Institute and the ASH Scholar Award. We acknowl­edge
edi­to­rial sup­port from Sally Mokhtari. We thank Dr. Sergio Gi­
ralt for com­ments on the con­cept. This work was also support­
ed in part by the NIH/NCI Cancer Center Support Grant P30
CA008748 to Memorial Sloan Kettering Cancer Center and P30
CA033572 to City of Hope National Medical Center. The con­tent
is solely the respon­si­bil­ity of the authors and does not nec­es­sar­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/254/1851535/254lin.pdf by guest on 13 December 2021
ily rep­re­sent the offi­cial views of the National Institutes of Health.
Likewise, the views expressed in this arti­cle are those of the au­
thors and do not reflect the posi­tion of the CIBMTR.
Conflict-of-inter­est dis­clo­sure
Richard J. Lin: no com­pet­ing finan­cial inter­ests to declare.
Andrew S. Artz: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Richard J. Lin: nothing to disclose.
Andrew S. Artz: nothing to disclose.
Correspondence
Andrew S. Artz, Department of Hematology and HCT, Center for
Cancer and Aging, City of Hope National Medical Center, 1500 E.
Duarte Road, Duarte, CA 91010; e-mail: aartz@coh​­.org.
References
1. Copelan EA. Hematopoietic stem-cell trans­plan­ta­tion. N Engl J Med. 2006;
354(17):1813-1826.
2. Granot N, Storb R. History of hema­ to­poi­etic cell trans­
plan­ta­tion: chal­
lenges and prog­ress. Haematologica. 2020;105(12):2716-2729.
3. Chabannon C, Kuball J, Bondanza A, et al. Hematopoietic stem cell trans­
plan­ta­tion in its 60s: a plat­ form for cel­lu­lar ther­
a­pies. Sci Transl Med.
2018;10(436).
4. Artz AS. Biologic vs phys­i­o­logic age in the trans­plant can­di­date. Hematol­
ogy Am Soc Hematol Educ Program. 2016;2016(1):99-105.
5. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer sta­tis­tics, 2021. CA Cancer
J Clin. 2021;71(1):7-33.
6. D’Souza A, Fretham C, Lee SJ, et al. Current use of and trends in hema­to­
poi­etic cell trans­plan­ta­tion in the United States. Biol Blood Marrow Trans­
plant. 2020;26(8):e177-e182.
7. Yourman LC, Lee SJ, Schonberg MA, Widera EW, Smith AK. Prognostic indi­
ces for older adults: a sys­tem­atic review. JAMA. 2012;307(2):182-192.
8. Pulte D, Jansen L, Castro FA, Brenner H. Changes in the sur­vival of older
patients with hema­to­logic malig­nan­cies in the early 21st cen­tury. Cancer.
2016;122(13):2031-2040.
9. Vasu S, Kohlschmidt J, Mrózek K, et al. Ten-year out­come of patients with
acute mye­loid leu­ke­mia not treated with allo­ge­neic trans­plan­ta­tion in first
com­plete remis­sion. Blood Adv. 2018;2(13):1645-1650.
10. Ustun C, Le-Rademacher J, Wang H-L, et al. Allogeneic hema­to­poi­etic cell
trans­plan­ta­tion com­pared to che­mo­ther­apy con­sol­i­da­tion in older acute
mye­loid leu­ke­mia (AML) patients 60-75 years in first com­plete remis­sion
(CR1): an alli­ance (A151509), SWOG, ECOG-ACRIN, and CIBMTR study. Leu­
kemia. 2019;33(11):2599-2609.
11. Farag SS, Maharry K, Zhang M-J, et al; Acute Leukemia Committee of the
Center for International Blood and Marrow Transplant Research and Cancer
and Leukemia Group B. Comparison of reduced-inten­sity hema­to­poi­etic
cell trans­plan­ta­tion with che­mo­ther­apy in patients age 60-70 years with
 acute mye­log­e­nous leu­ke­mia in first remis­sion. Biol Blood Marrow Trans­
plant. 2011;17(12):1796-1803.
12. Kurosawa S, Yamaguchi T, Uchida N, et al. Comparison of allo­ge­neic hema­
to­poi­etic cell trans­plan­ta­tion and che­mo­ther­apy in elderly patients with
non-M3 acute mye­log­e­nous leu­ke­mia in first com­plete remis­sion. Biol
Blood Marrow Transplant. 2011;17(3):401-411.
13. Versluis J, Hazenberg CL, Passweg JR, et al; HOVON and SAKK Leukemia
Groups. Post-remis­sion treat­ment with allo­ge­neic stem cell trans­plan­ta­
tion in patients aged 60 years and older with acute mye­loid leu­kae­mia: a
time-depen­dent anal­y­sis. Lancet Haematol. 2015;2(10):e427-e436.
14. Nakamura R, Saber W, Martens MJ, et  al. Biologic assign­ ment trial of
reduced-inten­sity hema­to­poi­etic cell trans­plan­ta­tion based on donor
avail­abil­ity in patients 50-75 years of age with advanced myelodysplastic
syn­drome [published online 9 June 2021]. J Clin Oncol.
15. Rashidi A, Ebadi M, Colditz GA, DiPersio JF. Outcomes of allo­ge­neic stem
cell trans­ plan­ta­
tion in elderly patients with acute mye­ loid leu­
ke­
mia: a
sys­tem­atic review and meta-anal­y­sis. Biol Blood Marrow Transplant.
2016;22(4):651-657.
16. Sorror ML, Sandmaier BM, Storer BE, et  al. Long-term out­comes among
older patients fol­low­ing nonmyeloablative con­di­tion­ing and allo­ge­neic
hema­to­poi­etic cell trans­plan­ta­tion for advanced hema­to­logic malig­nan­
cies. JAMA. 2011;306(17):1874-1883.
17. Ringdén O, Boumendil A, Labopin M, et al. Outcome of allo­ge­neic hema­
to­poi­etic stem cell trans­plan­ta­tion in patients age >69 years with acute
mye­log­e­nous leu­ke­mia: on behalf of the Acute Leukemia Working Party
of the Euro­pean Society for Blood and Marrow Transplantation. Biol Blood
Marrow Transplant. 2019;25(10):1975-1983.
18. Atallah E, Logan B, Chen M, et  al. Comparison of patient age groups in
trans­plan­ta­tion for myelodysplastic syn­drome: the Medi­care cov­er­age
with evi­dence devel­op­ment study. JAMA Oncol. 2020;6(4):486-493.
19. Muffly L, Pasquini MC, Martens M, et al. Increasing use of allo­ge­neic hema­
to­poi­etic cell trans­plan­ta­tion in patients aged 70 years and older in the
United States. Blood. 2017;130(9):1156-1164.
20. Olin RL. Delivering inten­sive ther­a­pies to older adults with hema­to­logic
malig­nan­cies: strat­e­gies to per­son­al­ize care. Blood. 2019;134(23):2013-2021.
21. Hourigan CS, Dillon LW, Gui G, et  al. Impact of con­di­tion­ing inten­sity of
allo­ge­neic trans­plan­ta­tion for acute mye­loid leu­ke­mia with geno­mic evi­
dence of resid­ual dis­ease. J Clin Oncol. 2020;38(12):1273-1283.
22. Dillon LW, Gui G, Logan BR, et  al. Impact of con­di­tion­ing inten­sity and
geno­ mics on relapse after allo­ ge­neic trans­ plan­
ta­
tion for patients with
myelodysplastic syn­drome. JCO Precis Oncol. 2021;5:265-274.
23. El-Jawahri A, Pidala J, Inamoto Y, et al. Impact of age on qual­ity of life, func­
tional sta­tus, and sur­vival in patients with chronic graft-ver­sus-host dis­
ease. Biol Blood Marrow Transplant. 2014;20(9):1341-1348.
24. Lin RJ, Baser RE, Elko TA, et al. Geriatric syn­dromes in 2-year, pro­gres­sion-
free sur­vi­vors among older recip­i­ents of allo­ge­neic hema­to­poi­etic cell
trans­plan­ta­tion. Bone Marrow Transplant. 2021;56(1):289-292.
25. Flannelly C, Tan B-E, Tan J-L, et al. Barriers to hema­to­poi­etic cell trans­plan­
ta­tion for adults in the United States: a sys­tem­atic review with a focus on
age. Biol Blood Marrow Transplant. 2020;26(12):2335-2345.
26. Meyer C, Mau L-W, Mur­phy E-A, et al. Addressing knowl­edge gaps in acute
mye­loid leu­ke­mia to improve refer­ral for hema­to­poi­etic cell trans­plan­ta­
tion con­sul­ta­tion. J Natl Compr Canc Netw. 2019;17(12):1473-1481.
27. Mishra A, Preussler JM, Bhatt VR, et al. Breaking the age bar­rier: phy­si­cians’
per­cep­tions of can­di­dacy for allo­ge­neic hema­to­poi­etic cell trans­plan­ta­
tion in older adults. Transplant Cell Ther. 2021;27(7):617.e1-617.e7.
28. Kasamon YL, Bolaños-Meade J, Prince GT, et  al. Outcomes of nonmye­
loablative HLA-haploidentical blood or mar­row trans­plan­ta­tion with high-
dose post-trans­plan­ta­tion cyclo­phos­pha­mide in older adults. J Clin Oncol.
2015;33(28):3152-3161.
29. Shaw BE, Jimenez-Jimenez AM, Burns LJ, et al. National mar­row donor pro­
gram-spon­sored mul­ti­cen­ter, phase II trial of HLA-mismatched unre­lated
donor bone mar­row trans­plan­ta­tion using post-trans­plant cyclo­phos­pha­
mide. J Clin Oncol. 2021;39(18):1971-1982.
30. Kanate AS, Perales M-A, Hamadani M. Eligibility cri­te­ria for patients under­
go­ing allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion. J Natl Compr Canc
Netw. 2020;18(5):635-643.
31. Goede V, Stauder R. Multidisciplinary care in the hema­tol­ogy clinic: imple­
mentation of geri­at­ric oncol­ogy. J Geriatr Oncol. 2019;10(3):497-503.
Prakash Singh Shekhawat
32. Wall SA, Knauss B, Compston A, et al. Multidisciplinary telemedicine and
the impor­tance of being seen. J Geriatr Oncol. 2020;11(8):1349-1351.
33. Wildiers H, Heeren P, Puts M, et al. International Society of Geriatric Oncol­
ogy con­sen­sus on geri­at­ric assess­ment in older patients with can­cer. J Clin
Oncol. 2014;32(24):2595-2603.
34. Muffly LS, Boulukos M, Swanson K, et al. Pilot study of com­pre­hen­sive geri­
at­ric assess­ment (CGA) in allo­ge­neic trans­plant: CGA cap­tures a high prev­
a­lence of vulnerabilities in older trans­plant recip­i­ents. Biol Blood Marrow
Transplant. 2013;19(3):429-434.
35. Muffly LS, Kocherginsky M, Stock W, et al. Geriatric assess­ment to pre­dict
sur­vival in older allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion recip­i­ents.
Haematologica. 2014;99(8):1373-1379.
36. Deschler B, Ihorst G, Schnitzler S, Bertz H, Finke J. Geriatric assess­ment and
qual­ity of life in older patients con­sid­ered for allo­ge­neic hema­to­poi­etic
cell trans­plan­ta­tion: a pro­spec­tive risk fac­tor and serial assess­ment anal­y­
sis. Bone Marrow Transplant. 2018;53(5):565-575.
37. Huang L-W, Sheng Y, Andreadis C, et al. Functional sta­tus as mea­sured by
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/254/1851535/254lin.pdf by guest on 13 December 2021
geri­at­ric assess­ment pre­dicts infe­rior sur­vival in older allo­ge­neic hema­
to­poi­etic cell trans­plan­ta­tion recip­i­ents. Biol Blood Marrow Transplant.
2020;26(1):189-196.
38. Olin RL, Fretham C, Pasquini MC, et  al. Geriatric assess­ ment in older
alloHCT recip­i­ents: asso­ci­a­tion of func­tional and cog­ni­tive impair­ment
with out­comes. Blood Adv. 2020;4(12):2810-2820.
39. Lin RJ, Elko TA, Devlin SM, et al. Impact of geri­at­ric vulnerabilities on allo­
ge­neic hema­to­poi­etic cell trans­plan­ta­tion out­comes in older patients with
hema­to­logic malig­nan­cies. Bone Marrow Transplant. 2020;55(1):157-164.
40. Pamukcuoglu M, Bhatia S, Weisdorf DJ, et al. Hematopoietic cell trans­plant-
related toxicities and mor­tal­ity in frail recip­i­ents. Biol Blood Marrow Trans­
plant. 2019;25(12):2454-2460.
41. Salas MQ , Atenafu EG, Bascom O, et al. Pilot pro­spec­tive study of frailty
and func­tion­al­ity in rou­tine clin­i­cal assess­ment in allo­ge­neic hema­to­poi­
etic cell trans­plan­ta­tion. Bone Marrow Transplant. 2021;56(1):60-69.
42. Polverelli N, Tura P, Battipaglia G, et al. Multidimensional geri­at­ric assess­
ment for elderly hema­to­log­i­cal patients (≥60 years) sub­mit­ted to allo­ge­
neic stem cell trans­plan­ta­tion: a French-Ital­ian 10-year expe­ri­ence on 228
patients. Bone Marrow Transplant. 2020;55(12):2224-2233.
43. Bhargava D, Arora M, DeFor TE, et al. Use of poten­tially inap­pro­pri­ate med­
i­ca­tions in older allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion recip­i­ents.
Biol Blood Marrow Transplant. 2020;26(12):2329-2334.
44. Sugidono M, Lo M, Young R, et al. Impact of polypharmacy prior to allo­ge­
neic hema­to­poi­etic stem cell trans­plan­ta­tion in older adults. Transplant
Cell Ther. 2021;27(4):344.e1-344.e5.
45. Jacobsen PB, Le-Rademacher J, Jim H, et al. Exercise and stress man­age­
ment train­ing prior to hema­to­poi­etic cell trans­plan­ta­tion: Blood and Mar­
row Transplant Clinical Trials Network (BMT CTN) 0902. Biol Blood Marrow
Transplant. 2014;20(10):1530-1536.
46. Derman BA, Kordas K, Ridgeway J, et al. Results from a mul­ti­dis­ci­plin­ary
clinic guided by geri­at­ric assess­ment before stem cell trans­plan­ta­tion in
older adults. Blood Adv. 2019;3(22):3488-3498.
47. Wildes TM, Artz AS. Characterize, opti­mize, and har­mo­nize: car­ing for older
adults with hema­to­logic malig­nan­cies. Am Soc Clin Oncol Educ Book.
2021;41:e266-e274.
48. Fann JR, Alfano CM, Roth-Roemer S, Katon WJ, Syrjala KL. Impact of delir­
ium on cog­ni­tion, dis­tress, and health-related qual­ity of life after hema­to­
poi­etic stem-cell trans­plan­ta­tion. J Clin Oncol. 2007;25(10):1223-1231.
49. Lin RJ, Hilden PD, Elko TA, et al. Burden and impact of mul­ti­fac­to­rial geri­
at­ric syn­dromes in allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion for older
adults. Blood Adv. 2019;3(1):12-20.
50. Wall SA, Zhao Q , Vasu S, Rosko A. Discharge dis­po­si­tion fol­low­ing hema­
to­poi­etic cell trans­plan­ta­tion: predicting the need for reha­bil­i­ta­tion and
asso­ci­a­tion with sur­vival. Transplant Cell Ther. 2021;27(4):337.e1-337.e7.
51. Lin RJ, Elias HK, van den Brink MRM. Immune recon­sti­tu­tion in the aging
host: oppor­tu­ni­ties for mech­a­nism-based ther­apy in allo­ge­neic hema­to­
poi­etic cell trans­plan­ta­tion. Front Immunol. 2021;12:674093.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000257
Allotransplant in older adults  | 263
 HOW CAN WE ENSURE THAT EVERYONE WHO NEEDS A TRANSPLANT CAN GET ONE ?

Increasing access to allogeneic hematopoietic

cell transplant: an international perspective

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021

Vanderson Rocha,1,2,3 Giancarlo Fatobene,1,2,3 and Dietger Niederwieser,4 for the Brazilian Society of Bone
Marrow Transplantation and the Worldwide Network for Blood and Marrow Transplantation
Laboratorio de Investigação Médica (LIM) 31, Serviço de Hematologia e Terapia Celular, Hospital das Clínicas, Faculdade de Medicina, Universidade
1

de São Paulo, São Paulo, Brazil; 2Eurocord, Paris, France; 3Hospital Vila Nova Star - Rede D’Or, São Paulo, Brazil; and 4University of Leipzig, Leipzig,
Germany

Allogeneic hematopoietic cell transplantation (allo-HCT) is a highly complex, costly procedure for patients with onco-
logic, hematologic, genetic, and immunologic diseases. Demographics and socioeconomic status as well as donor avail-
ability and type of health care system are important factors that influence access to and outcomes following allo-HCT.
The last decade has seen an increase in the numbers of allo-HCTs and teams all over the world, with no signs of satura-
tion. More than 80 000 procedures are being performed annually, with 1 million allo-HCTs estimated to take place by the
end of 2024. Many factors have contributed to this, including increased numbers of eligible patients (older adults with or
without comorbidities) and available donors (unrelated and haploidentical), improved supportive care, and decreased
early and late post-HCT mortalities. This increase is also directly linked to macro- and microeconomic indicators that affect
health care both regionally and globally. Despite this global increase in the number of allo-HCTs and transplant centers,
there is an enormous need for increased access to and improved outcomes following allo-HCT in resource-constrained
countries. The reduction of poverty, global economic changes, greater access to information, exchange of technologies,
and use of artificial intelligence, mobile health, and telehealth are certainly creating unprecedented opportunities to
establish collaborations and share experiences and thus increase patient access to allo-HCT. A specific research agenda
to address issues of allo-HCT in resource-constrained settings is urgently warranted.

LEARNING OBJECTIVES
• Understand the panorama of allogeneic hematopoietic cell transplantation (allo-HCT) activity worldwide
• Review factors affecting the access to allo-HCT across the globe
• Discuss possible strategies for expanding allo-HCT activity in resource-constrained areas

Introduction (4) improvements in early and late outcomes due to better

Allogeneic hematopoietic cell transplantation (allo-HCT) is a
highly specialized, technologically sophisticated, resource-
intense, and expensive procedure for patients with oth-
erwise incurable hematologic diseases. Numerous factors
influence whether a patient eligible for allo-HCT actually
goes on to receive it: donor availability, social issues, eco-
nomic status, and health care system.1 The degree to which
these factors, particularly socioeconomic factors, might
influence access to allo-HCT is a question of debate.1
In recent decades a worldwide increase in the use of
allo-HCT has been seen due to (1) better donor availabil-
ity, (2) optimization of indications and rapid evolution of
molecular diagnostic/prognostic techniques, (3) use of
novel reduced-intensity conditioning regimens for older
adult patients and/or patients with comorbidities, and
supportive care. However, allo-HCT is still associated with
significant morbidity and mortality, requiring advanced
care consisting of significant infrastructure and a network
of specialists from all fields of medicine.
The World Health Organization (WHO) has recognized
the broad idea of the provision of medical products of
human origin as an important global medical task and a
governmental responsibility at the national level.2,3 Data
collection and data analysis are also recognized as inte-
gral parts of the treatment to achieve an efficient and cost-
effective use of resources. The Worldwide Network for
Blood and Marrow Transplantation (WBMT), as a nongov-
ernmental organization in official relations with the WHO,
has taken up the challenge of collecting global HCT activity
data. Information on HCT trends over time provides a sound

264 | Hematology 2021 | ASH Education Program

basis for sci­en­tific soci­e­ties, pol­i­ti­cians, and health care agencies
to develop or opti­mize national HCT pro­grams.
This arti­cle pres­ents a global over­view of the num­bers and
types of allo-HCT in dif­fer­ent world regions col­lected by mem­
bers of the WBMT (the Euro­pean Group for Blood and Mar-
row Trans­plantation, the Center for International Blood and
Marrow Transplant Research, and the Asia-Pacific Blood and Mar-
row Transplantation Group) or directly from trans­plant cen­ters
(where no regional reg­is­try is in place) using a uni­form reporting
sheet.4,5 As many patient, dis­ease, donor, and trans­plant-related
fac­tors may improve out­comes, includ­ing those affect­ing access
to allo-HCT, this arti­cle also describes these aspects.
Worldwide allo-HCT activ­ity and over­all trends
Worldwide, 1 298 897 HCTs (42.9% allo-HCT) pro­ce­dures have
been recorded from the first HCT in 1957 to 2016.5 The annual
activ­ity increased con­tin­u­ously from 46 563 in 2006 (the foun­
da­tion of WBMT and the first global report) to 82 718 in 2016
(the lat­est com­plete global sur­vey, Table 1),4-6 amounting to a
global increase of 77.6% since 2006, which was some­what higher
in allo-HCT (89.0%) than in auto-HCT (68.9%).5
Allo-HCT team activ­ity in 2016
In 2016, 38 425 first allo-HCTs, in com­par­i­son to 20 333 in 2006,
were performed world­wide (Figure 1). The increase took place
in all­regions of the world, and in con­trast to 2006, more related
(53.6%) than unre­lated HCTs were reported.5 Increases of 20%
and 50% were noted in the num­ber of allo-HCT teams and allo-
HCTs, respec­tively, while the allo-HCT/team ratio var­ied from
15.4 in 2006 to 23.3 in 2016 (Figure 2A). Team den­sity (TD, the
num­ber of teams/10 mil­lion per­sons) was highest in North Amer-
ica (NA; 23.72), followed by Europe (EUR; 18.53), the South-East
Asian/Western Pacific Region (SEAR/WPR; 3.91), Latin America
Figure 1. Allo-HCT activity according to donor type (related and unrelated), irrespective of source and matching (PBSC, BM, or CB;
matched or mismatched), and region Prakash
comparingSingh
2006 toShekhawat
2016. BM, bone marrow; CB, cord blood; PBSC, peripheral blood stem cell.
(LA; 3.27), and the Eastern Med­i­ter­ra­nean/Afri­can Region (EMR/
AFR; 1.98; Figure 2B). Accordingly, allo-HCT/team ratios were
higher in EMR/AFR and NA, dem­on­strat­ing a par­tic­u­larly high
con­cen­tra­tion of pro­ce­dures in few trans­plant cen­ters in EMR/AFR
(Figure 2B).5
Trends in indi­ca­tions, donor type, and stem cell source
The num­ber of trans­plants for all­indi­ca­tions using all­type of
donors (except for plasma cell dis­or­ders and lym­pho­mas in
related allo-HCT; data not shown) increased.5 Leukemia was
the most fre­quent indi­ca­tion, of which acute mye­loid leu­ke­mia
(AML) amounted 14 334 HCTs. HCT for myelodysplastic syn­
dromes (MDS) and, to a lower extent, mye­lo­pro­lif­er­a­tive dis­or­
ders increased steadily over the obser­va­tion period. Increased
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021
allo-HCT activ­i­ties were observed for almost all­indi­ca­tions in
all­regions except for solid tumors in almost all­regions but
SEAR/WPR and lymphoproliferative dis­or­ders in NA (Figure 3).
The highest incre­ments were observed for severe aplastic ane­
mia and hemo­glo­bin­op­a­thies (Δ2007-2016 > 1000%) and for leu­
ke­mia (Δ2007-2016 > 550%; data not shown).5
Allo-HCTs (range, 0-7850) were reported from 76 countries,
includ­ing grafts from unre­lated donors (UDs) and from cord blood
(CB) in 55 and in 41 countries, respec­tively. Absolute UD-HCT num­
bers ranged from 0 to 4311, and those of CB ranged from 0 to 1233
in indi­vid­ual countries. Overall, related-HCT has become more
fre­quent than UD-HCT, starting in 2014, due to the increased use
of related haploidentical (haplo) donors (39.5% of related HCT).
Haplo-HCTs (n = 8131) were evenly dis­trib­uted in 62 countries, with
abso­lute num­bers rang­ing from 0 to 2554. It is not sur­pris­ing that
more haplo-HCTs were performed in regions with­out or with few
UD reg­is­tries (LA, EMR, AFR, and SEAR/WPR).
Of the 38 425 allo-HCTs, 20% were derived from bone mar­
row (BM), 73% from periph­eral blood (PB), and 7% from CB.5
Factors affect­ing global access to allotransplant  |  265
 A
Number of HCTs

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021

allo-HCT
B
World Regions

Number of HCTs
HCT/team
Figure 2. (A) Number of allo-HCTs, number of allo-HCT teams, and allo-HCTs per team. (B) Number of allo-HCT teams, number of
allo-HCTs per team, and TD according to region in 2016.

Factors affect­ing access to allo-HCT (Figure 4)
Donor avail­abil­ity
UD reg­is­tries: impact of eth­nic­ity and eco­nomic issues.  In the
absence of an HLA-iden­ti­cal related donor, hema­to­poi­etic cells
from HLA-matched or mismatched UD, haplo donors, or CB cells
can be used. Countries with UD reg­is­tries increased from 2 in
1987 to 57 in 2012 and 119 in 2021. Accordingly, the donor pool has
increased in the last 40 years to >40 mil­lion UD, includ­ing CB (May
2021; https://wmda.info/). Rates of UD per 10 000 inhab­i­tants
vary per con­ti­nent (2.98 in Africa, 16.5 in Asia, 61.64 in Oceania,
179.45 in South America, 198.28 in NA, and 247.33 in Europe; Joris
M., per­sonal com­mu­ni­ca­tion, May 2021). The num­bers of unre­
lated grafts (BM, PB, and CB cells) expanded >7-fold from 1997
to 2020, with increased use of PB (except for the COVID-19 pan­
demic period in 2020) and decreased use of BM and CB cells
(Figure 5A). Figure 5B shows the global exchange of prod­ucts
(BM + PB) in 2020. In a National Marrow Donor Program study,
the prob­a­bil­ity of find­ing a UD for black Amer­i­cans is <17%.7 In
a recent study of the Brazilian UD reg­is­try (REDOME), this prob­
a­bil­ity was <10% for sickle cell dis­ease patients.8 In addi­tion,

266  |  Hematology 2021  |  ASH Education Program

Delta (%)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021

Figure 3. Change in % of allo-HCT activity from 2006 and 2016 activity according to diagnosis and region. Adapted with permission
of Haematologica.5

access to searching for and using UD grafts is lim­ited by finan­
cial issues even in dif­ fer­ent areas of high-income countries
(HICs). In the US, Paulson et al. found that patients with leu­ke­
mia from poorer areas were less likely to undergo UD allo-HCT
com­pared to patients from wealth­ier countries.9 Similar find­ings
were also reported in Italy.10 Despite many ret­ro­spec­tive stud­ies
and some pro­spec­tive tri­als reporting sim­i­lar out­comes after UD,
CB, or haplo-HCT, the use of UD over haplo or CB depends on
patient/donor eth­nic­ity, avail­abil­ity of a UD reg­is­try, and finan­
cial issues—graft-related costs being an impor­tant fac­tor in low-
and mid­dle-income countries (LICs, LMICs).
Impact of haplo donors on access to allo-HCT.  The use of haplo
donors is increas­ing world­wide, but inter­est­ingly, this is more evi­
dent in LICs and LMICs, prob­a­bly due to the unavail­abil­ity of HLA-
matched UD, HCT-asso­ ci­
ated costs, infra­ struc­ture of cen­ ters,
lack of access to drugs, and other fac­tors. According to the last
WBMT report, the haplo-HCT trans­plant rate (TR; HCT/10 mil­lion

Figure 4. Factors associated

Prakash Singhwith access to allo-HCT: an international perspective.
Shekhawat
Factors affect­ing global access to allotransplant  |  267

Figure 5. (A) Unrelated BM, mobilized PB, and CB grafts shipped from 1997 to 2020. (B) Global exchange of hematopoietic cell
products (BM + PB) in 2020.
pop­u­la­tion) represented 32% and 26% of all­allo-HCTs in SEA/W-
PR and LA, respec­ tively, but only 14% in Europe and EMR/
AFR and 4.9% in NA.5
Patient demo­graph­ics: age, gen­der, and eth­nic­ity
Studies inves­ti­gat­ing age as a pre­dic­tor of access to HCT indi­cate
that youn­ger patients are more likely to receive a trans­plant com­
pared to older patients, even in HICs.11,12 Only 5.5% of older adult
patients with AML (n = 17 555) under­went HCT in the US between
2003 and 2012. Other fac­tors asso­ci­ated with a lower like­li­hood
of receiv­ing HCT included care at a non­ac­a­demic hos­pi­tal, race
other than White, Charlson comorbidity score of ≥1, unin­sured
sta­tus, and lower edu­ca­tional sta­tus.12 No sig­nif­i­cant gen­der
effect has been reported.13 The role of race in turn must be inter-
preted with cau­tion because race is a com­plex social, cul­tural,
and polit­i­cal con­struct and not only a bio­log­ic ­ al con­cept. The
avail­abil­ity of a donor for non-Cau­ca­sians and eth­nic minor­i­ties
has improved over the last decades with the use of haplo donors
and CB units.14 However, many stud­ ies have linked eco­ nomic
issues with race and eth­nic­ity to address access to allo-HCT.9,15
Social fac­tors
Geographic dis­tance. The dis­tance between a patient and an
HCT facil­ity appears to be an impor­tant fac­tor to HCT access.16
To ensure proper fol­low-up care, allo-HCT patients with lim­ited
geo­graphic access must decide whether to make numer­ ous
long trips or relo­cate near the HCT facil­ity, either of which can be
a sig­nif­i­cant finan­cial bur­den. There are conflicting results from
published data on the role of dis­tance from the trans­plant cen­ter
or the impact of urban/rural areas on out­comes after allo-HCT.9,17
However, in a recent study, poor access to care, defined as low
socio­eco­nomic sta­tus and a far dis­tance to the trans­plant cen­ter,
was asso­ci­ated with increased 1-year mor­tal­ity.18
Patient/fam­ily atti­tudes and avail­abil­ity of care­giv­ers. Family
and patient psy­cho­log­i­cal fac­tors may be impor­tant for HCT
access and out­come. To our knowl­edge no study has eval­u­ated
how psy­cho­log­ic ­ al fac­tors affect access, yet diag­noses of depres­
sion, anx­i­ety, or post­trau­matic stress dis­or­der fol­low­ing HCT were
asso­ci­ated with sub­op­ti­mal health out­comes and increased mor­
tal­ity.19-21 Caregiver avail­abil­ity may also be a bar­rier to patients
con­sid­er­ing and pro­ceed­ing to HCT.22 Parents/guard­ians are often
the nat­u­ral care­giv­ers for their chil­dren, but for adult patients, the
268  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021
quan­tity and qual­ity of rela­tion­ships and the care­giver’s income
loss may affect the avail­abil­ity of such a per­son.23,24
Economics
Macro- and micro­eco­nomic fac­tors: impact on access to trans­
plan­ta­tion.  From a global per­spec­tive, access to and rates of allo-
HCT are closely related to the socio­eco­nomic sta­tus of countries
and regions of the globe. International find­ings show strik­ing dif­
fer­ences in abso­lute trans­plant num­bers, TD, and the spread of
allo-HCT, which are affected mainly by a coun­try’s or region’s
macro- or micro­eco­nomic fac­tors related to resources and infra­
struc­ture.4 For instance, a recent study showed that adults with
AML treated in São Paulo (Brazil) had infe­rior 5-year over­all sur­
vival (OS) and higher early mor­tal­ity pri­mar­ily due to multiresis-
tant gram-neg­a­tive bac­te­rial and fun­gal infec­tions com­pared to
patients from Oxford (UK). Importantly, Brazilian patients were
less likely to undergo allo-HCT (28% vs 75%; P < .001) and waited
lon­ger for HCT (median, 23.8 vs 7.2 months; P < .001).25
The fol­low­ing eco­nomic indi­ca­tors have been asso­ci­ated
with TR (HCTs/10 mil­lion), other trans­plant indi­ca­tors,4 and out­
comes26-28: (1) human devel­op­ment index (HDI), a com­pos­ite var­
i­able containing infor­ma­tion about life expec­tancy, edu­ca­tion,
and gross national income (GNI); (2) health care expen­di­ture
(HCE) per cap­ita; (3) GNI per cap­ita; (4) World Bank categories:
HICs, HMICs, LMICs, and LICs.
Gratwohl et al4 and Niederwieser et al5 dem­on­strated that
global num­bers of allo-HCT are still increas­ing. By the end of
2024, almost 1 mil­lion allo-HCT will be performed in 76 of the
194 WHO mem­ber states, yet no HCT was performed in coun-
tries with a pop­u­la­tion of <300 000 inhab­i­tants, sur­face area
<700 km2, and GNI < US$1260/per­son. TR was higher in countries
with greater gross domes­tic prod­uct, GNI, and HCE per per­
son, higher HDI, more donors, and larger CB banks.4 The asso­
ci­a­tion between trans­plant and mac­ro­eco­nomic fac­tors var­ied
sub­stan­tially between donor types, WHO regions, and World
Bank categories. Macroeconomic resources gen­er­ally showed
higher asso­ci­at­ions with auto-HCT than allo-HCT, with no sub­
stan­tial dif­fer­ences for related- or UD-HCT. Ease of access to
HCT as deter­mined by TD gen­er­ally showed a higher asso­ci­a­
tion for auto- than allo-HCT. Generally, asso­ ci­
a­tions between
mac­ ro­
eco­
nomic fac­ tors and TR were higher in the Americas
and SEAR/WPR Regions. Considering the World Bank catego-
ries, asso­ci­a­tions were gen­er­ally low, with a stron­ger impact of
 Table 1. Increase in HCT num­bers from 2006 to 2016 according to trans­plant type (allo­ge­neic and autol­o­gous), donor type, and source4-6

Year 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Allogeneic HCT 20,333 21,904 23,989 25,539 27,189 29,815 31,926 33,572 35,333 35,897 38,425
Related 11,002 11,611 12,362 12,641 12,945 14,303 15,493 16,305 17,705 18,515 20,594
 Identical n.a. 10,107 10,599 10,624 10,912 11,751 12,322 12,475 12,579 12,511 12,463
 Nonidentical n.a. 1,504 1,763 2,017 2,033 2,552 3,171 3,830 5,126 6,004 8,131
 CB 0 102 92 131 149 143 251 239 298 358 117

Prakash Singh Shekhawat

Unrelated 9,331 10,293 11,627 12,898 14,244 15,512 16,433 17,267 17,628 17,382 17,831
 CB 1,722 2,031 2,266 2,508 2,804 2,927 2,859 2,787 2,568 2,474 2,477
Autologous HCT 26,230 26,805 27,547 30,081 31,730 33,756 36,220 37,464 38,188 41,698 44,293
Total 46,563 48,709 51,536 55,620 58,919 63,571 68,146 71,036 73,521 77,595 82,718
n.a.

Factors affect­ing global access to allotransplant  |  269

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021
finan­cial resources on the LMIC cat­e­gory and TD for auto-HCT in
the HMIC cat­e­gory.
Little data are avail­­able on whether eco­nomic fac­tors may
also explain the TR dif­fer­ences found among regions of a par­
tic­u­lar coun­try. This microlevel sce­nario gen­er­ally involves
the health care sys­tem and eco­nom­ics, while clin­i­cal dif­fer­
ences should not be expected to cause dif­fer­ent TR among
regions within a coun­try with a pub­lic health care sys­tem.
An inter­est­ing Span­ish study showed that auto-HCT TR was
asso­ci­ated with only TD, whereas mac­ro­eco­nomic deter­mi­
nants exerted a strong influ­ence in the case of allo-TR.29 In
par­tic­u­lar, an increase of €1000 in gross domes­tic prod­uct per
cap­ita would cause an aver­age increase of 1.4 trans­plants per
1 ­mil­lion inhab­i­tants.29 Therefore, there may be inequal­ity in
access to HCT by region due to per cap­ita income, TD, and
pub­lic hos­pi­tal expen­di­ture. These find­ings may be use­ful to
health author­i­ties in mak­ing deci­sions both at the mac­ro­eco­
nomic and local lev­els.
Macro- and micro­eco­nomic fac­tors: impact on out­comes of
allo-HCT.  Studies ana­lyz­ing socio­eco­nomic fac­tors and out­
comes after allo-HCT are restricted to patients with suf­fic ­ ient
resources to pro­ceed to HCT and might not be rep­re­sen­ta­tive
of all­trans­plant can­di­dates. Hence, although research on out­
comes fol­low­ing allo-HCT in dis­ad­van­taged groups is impor­tant,
it is equally rel­e­vant to ensure that patients in these groups actu­
ally have access to trans­plan­ta­tion.
In 2010 we eval­u­ated the asso­ci­a­tion of HDI with rates and
out­comes of allo-HCT for acute leu­ke­mia in Euro­pean countries.
Allo-HCTs performed in countries belong­ing to the upper HDI
cat­e­gory were asso­ci­ated with higher leu­ke­mia-free sur­vival
com­ pared to other categories.27 In another Euro­ pean study,
lower cur­rent HCE and HDI were asso­ci­ated with a decreased
prob­a­bil­ity of OS in allo-HCT recip­i­ents with acute lym­pho­blas­tic
leu­ke­mia.26 An impor­tant Center for International Blood and Mar-
row Transplant Research study ana­lyz­ing 11 261 allo-HCT recip­
i­ents with acute lym­pho­blas­tic leu­ke­mia across 38 countries
showed that trans­plants performed in countries within the low­
est HDI quar­tile were asso­ci­ated with an OS infe­rior to countries
in the highest HDI quar­tile.30
The asso­ci­a­tion between coun­try-level mac­ro­eco­nomic indi­
ca­tors and out­comes after allo-HCT needs fur­ther eval­u­at­ion.
HCT in lower-income set­tings may be asso­ci­ated with decreased
avail­­able resources, espe­cially early post HCT.31 In fact, prior work
has documented exten­sive var­i­abil­ity in the pro­vi­sion of sup­port­
ive prac­tices across countries.30 It is pos­si­ble that care deliv­ery-
related fac­tors may have con­trib­uted to the high pro­por­tion of
deaths from infec­tions seen in HMIC/LMIC/LIC set­tings (eg, lack
of ade­quate inten­sive care sup­port).25,30,32 Alternatively, worse
mac­ro­eco­nomic indi­ca­tors may reflect deficiencies in train­ing,
expe­ri­ence, or staffing mod­els in cen­ters in resource-constrained
set­tings. On a patient level, lower socio­eco­nomic sta­tus may be
asso­ci­ated with adverse out­comes owing to chronic stress, low
edu­ca­tional level, unaffordability of med­i­ca­tions, and dif­fi­cul­ties
in reaching the trans­plant cen­ter.
Health care sys­tems
Health care is a sec­tor where gov­ern­ments have a lead­ing role.
Access to and out­comes fol­low­ing allo-HCT depend on how
much a gov­ern­ment invests in gen­eral health and ter­tiary care.
To our knowl­edge, there are no data com­par­ing the type of
270  |  Hematology 2021  |  ASH Education Program
health sys­tem (pub­lic vs pri­vate or uni­ver­sal vs non­uni­ver­sal) and
access to allo-HCT.
HCT is costly and may drain con­sid­er­able resources from pri­
mary care, mater­nal and child­hood care, or coun­try­wide vac­ci­na­
tion pro­grams. On the other hand, car­ing for patients with endemic
dis­eases (eg, hemo­glo­bin­op­a­thies in most emerg­ing countries)
may be costly over time, and HCT may be more cost-effec­tive in
the long run.33 Thus, when it comes to stra­te­gic pri­or­it­ies, coun-
tries should tai­lor the implementation of ter­tiary care facil­i­ties
according to socio­eco­nomic con­di­tions and coun­try-spe­cific dis­
eases. It is also crit­i­cal to eval­u­ate the clin­ic
­ al/eco­nomic effec­
tive­ness and sus­tain­abil­ity of such an endeavor.34 Unfortunately,
there are scarce data on what type of donor or graft-ver­ sus-
host dis­ease (GVHD) pro­phy­laxis could be most cost-effec­tive in
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021
devel­op­ing econ­o­mies, an issue that urgently war­rants fur­ther
research.
Referral of patients to trans­plant and posttransplant care
The com­ plex inter­play of patients, refer­ ring phy­ si­
cians, pay­
ors, and trans­plant cen­ter-related fac­tors have been pre­vi­ously
described.35 These fac­tors need to be com­pre­hen­sively stud­ied
to deliver opti­mal care. Both pay­ors and accred­i­ta­tion agencies
should also attempt to ele­vate the stan­dards of care affect­ing
trans­plant out­comes after dis­charge from the trans­plant cen­ter.35
National and inter­na­tional reg­u­la­tions
In many health care set­tings, bench­mark­ing for com­plex pro­
ce­dures has become a man­da­tory require­ment by stake­hold­
ers to assure clin­i­cal per­for­mance, cost-effec­tive­ness, and
patient safety.36 Benchmarking has also been inte­grated into
the Foundation for the Accreditation of Cellular Therapy-Joint
Accreditation Committee of the International Society for Cel-
lular Therapy and the Euro­pean Society for Blood and Mar-
row Transplantation qual­ity man­age­ment stan­dards and has
been established mainly for devel­oped countries but not for
resource-constrained nations.36 Developing mean­ing­ful bench­
mark­ing sys­tems for countries with sim­i­lar mac­ro­eco­nomic
fac­tors may help address and cor­rect underperformance.
Establishing new HCT pro­grams
Table 2 sum­ ma­rizes impor­ tant steps in establishing new or
opti­miz­ing established HCT pro­grams. Close coop­er­a­tion with
health author­i­ties, pol­i­ti­cians, and phy­si­cians is needed to
ensure equal­ity of treat­ment around the world. It is essen­tial to
allo­cate funds for this treat­ment and have sup­port from regional
and global sci­en­tific soci­e­ties in asso­ci­a­tion with the WHO. Fur-
thermore, HCT-spe­ cific min­ i­
mum require­ ments are needed.
The Transplant Center and Recipient Issues Standing Commit-
tee of the WBMT orga­nized a struc­tured review and ana­lyzed,
described, and scored (by inde­pen­dent trans­plant phy­si­cians)
min­ i­
mum require­ ments to estab­ lish an HCT pro­ gram.37 This
struc­tured set of rec­om­men­da­tions guides the pri­or­i­ti­za­tion to
estab­lish a trans­plant pro­gram and set the path for expan­sion
and fur­ther devel­op­ment.37
How to improve access to allo-HCT
Access to allo-HCT is depen­dent on the mul­ti­ple fac­tors reported
above and sum­ ma­ rized in Figure 6. Improvements in access
can be obtained by eval­u­at­ing the need for HCT in indi­vid­ual
countries and the ways in which to improve. The eas­i­est way
would be to improve the TR in existing cen­ters and estab­lish
Table 2. How to improve access to HCT

Target Topic Actions

Benchmarking activ­i­ties among
countries and regions
Starting new pro­grams
Global HCT activ­ity reports
Alerting health author­i­ties and pol­i­ti­cians about the
need for pro­grams in countries with low HCT activ­ity
Essential med­i­ca­tion
Training of phy­si­cians, nurses, tech­ni­cians, and data
man­ag­ers
Infrastructure
Site visit from expe­ri­enced phy­si­cians
Biannual sur­vey since 20064,6,47-50
Organization of WBMT work­shops in coop­er­a­tion with
the WHO51
Published pre­vi­ously52
Scientific soci­e­ties; accredited trans­plant
cen­ters
Define essen­tial infra­struc­ture37,53
Role of sci­en­tific soci­e­ties

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021

Optimizing existing pro­grams
FACT, Foundation for the Accreditation of Cellular Therapy; JACIE, Joint Accreditation Committee of the International Society for Cellular Therapy.
Financial aspects Optimize treat­ment
Twinning and telemedicine Supervisory telemedicine42
Outcome reg­is­tries Establish out­come reg­is­tries
Analysis of dif­fer­ent tech­niques54
Accreditation Liaise with JACIE/FACT
Utilization of HCT world­wide Analyzing inci­dence (tumor reg­is­tries) and HCT activ­i­ties
for each dis­ease in regions and countries55
Establishing alter­nate donor reg­is­tries Describe chal­lenges in devel­op­ing countries56
Establishing clin­i­cal stud­ies Structures for local reg­is­tries, noninterventional,
interventional stud­ies

new trans­plant cen­ters in regions with low TD. This eval­u­a­tion appli­ca­tions rep­re­sent a unique oppor­tu­nity to nar­row the gap
pro­vi­des com­par­i­sons for bench­mark­ing and some hints on how of access to HCT between devel­oped and devel­op­ing nations.
to cre­ate or opti­mize trans­plant pro­grams. New tech­nol­o­gies Telehealth has been asso­ci­ated with decreas­ing costs, gains
may help in this endeavor. in pro­duc­ tiv­
ity, patient adher­ence, and improved access to
health care.38 Since the 2000s there have been a few reports
Role of new tech­nol­o­gies on the use of telehealth in HCT recip­i­ents,39,40 but the COVID-19
Even in LICs and LMICs, wide­ spread adop­ tion of the mobile pan­demic really caused the field to gain trac­tion world­wide,
phone, invest­ments in elec­tronic med­i­cal records, devel­op­ments includ­ing in the devel­op­ing world. At our pub­lic insti­tu­tion in
in cloud com­ put­
ing, and emer­ gent mobile health (mHealth) Brazil (V.R., G.F.), a recent sur­vey among 232 HCT recip­i­ents

Prakash
Figure 6. Increasing access and Singh Shekhawat
possibly outcomes to allo-HCT using new technologies: an international perspective.

Factors affect­ing global access to allotransplant  |  271

dur­ing the COVID-19 pan­demic showed that one-third of them
spent >120 min­utes to travel to and from the out­pa­tient clinic,
and 42% had a cost >US $10.00 (equiv­a­lent to 5% of the min­i­
mum wage/month). Approximately half of them had engaged
in at least 1 pre­vi­ous inter­ac­tion via telehealth with our cen­ter,
and 91% of these patients con­sid­ered it a good expe­ri­ence.41
Although not fit for all­clin­i­cal sce­nar­ios, telehealth may be effi­
cient and com­ple­men­tary to in-per­son inter­ac­tions with HCT
patients, may allow a cost reduc­tion for patients, and may be
espe­cially use­ful in the long-term fol­low-up of HCT sur­vi­vors,40
which often requires insti­tu­tions expe­ri­enced in the man­age­
ment of chronic GVHD and late com­pli­ca­tions. Telementoring is
also a prom­is­ing approach to implementing allo-HCT pro­grams
in that con­text.42
Wearable devices (eg, the Apple Watch) rep­re­sent another
mHealth modal­ ity increas­ ingly assim­ i­
lated in every­ day life.
These devices can be con­tin­u­ous or inter­mit­tently cap­ture
bio­log­ i­
cal data, and recent expe­ ri­ences have been tested
in can­cer and HCT sur­vi­vors.43 Wearable devices along with
telehealth approaches reduced hos­pi­tal readmissions fol­low­
ing liver trans­plan­ta­tion.44 Coupling the mas­ sive data gen­
er­ated by these tech­nol­o­gies, elec­tronic med­i­cal records,
and bio­mark­ers with arti­fi­cial intel­li­gence may poten­tially
pre­dict trans­ plant com­ pli­ca­tions such as febrile neutrope-
nia and GVHD hours or days in advance.45 In fact, arti­fi­cial
intel­li­gence has been used to improve the diag­no­sis of birth
asphyxia and dia­betic ret­i­nop­a­thy and pro­vide treat­ment rec­
om­men­da­tions in can­cer in dif­fer­ent low-income countries.46
With decreas­ing costs, these tech­nol­o­gies could expand out­
pa­tient allo-HCT and allow bet­ter uti­li­za­tion of hos­pi­tal HCT
beds in regions where trans­plant waiting lists are a real­ity.
Prioritization of trans­plant can­di­dates on waiting lists may also
be assisted by intel­li­gent algo­rithms that take into account
per­sonal and dis­ease char­ac­ter­is­tics, pretransplant workup
sched­ul­ing, local trans­plant cen­ter per­for­mance, and evi­
dence-based data.
Although tech­no­log­i­cal advances offer unprec­e­dented oppor­
tu­ni­ties to enable access to HCT, the val­i­da­tion of effec­tive solu­
tions demands high-qual­ ity data sets, internet con­ nec­tiv­
ity,
infor­ma­tion tech­nol­ogy infra­struc­ture, and clin­i­cal tri­als, which
are time- and cost-con­sum­ing. Hence, stake­hold­ers must ensure
that actions are taken to guide sus­tain­able, equi­ta­ble tech­nol­
ogy implementation in the HCT realm to ful­fill its prom­ises.
Research agenda for resource-constrained countries
Most of the cur­rent evi­dence in HCT has been gen­er­ated in HICs
and may not be appli­ ca­ble to resource-constrained set­ tings.
Many clin­i­cal chal­lenges faced by devel­op­ing countries (ie, the
major­ity of the world’s pop­u­la­tion) are under­rep­re­sented in the
HCT lit­er­a­ture (eg, evidenced-based man­age­ment of waiting
lists, peritransplant mul­ti­drug-resis­tant bac­te­ria, trop­i­cal dis­
eases). A research agenda, with a par­tic­u­lar focus on clin­i­cal tri­
als, directed to and conducted in resource-constrained countries
is essen­tial to effec­tively tackle these issues. Initiatives fos­ter­ing
inter­na­tional coop­er­a­tion, funding, and local research capac­ity,
such as the Amer­i­can Society of Hematology Global Research
Award and Clinical Research Training Institutes, should be mul­
ti­plied to advance access to HCT and other cel­lu­lar ther­a­pies
world­wide.
272  |  Hematology 2021  |  ASH Education Program
Conflict-of-inter­est dis­clo­sure
Vanderson Rocha: no com­pet­ing finan­cial inter­ests to declare.
Giancarlo Fatobene: no com­pet­ing finan­cial inter­ests to declare.
Dietger Niederwieser: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Vanderson Rocha: no off-label drug use is discussed.
Giancarlo Fatobene: no off-label drug use is discussed.
Dietger Niederwieser: no off-label drug use is discussed.
Correspondence
Vanderson Rocha, University of São Paulo, Rua Dr Ovidio Pires de
Campos, São Paulo, Brazil 05403-010; e-mail: vanderson​­.rocha​
­@hc​­.fm​­.usp​­.br.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021
References
1. Majhail NS, Omondi NA, Denzen E, Mur­phy EA, Rizzo JD. Access to hema­to­
poi­etic cell trans­plan­ta­tion in the United States. Biol Blood Marrow Trans-
plant. 2010;16(8):1070-1075.
2. Warwick RM, Chapman J, Pruett TL, Wang H. Globally con­sis­tent cod­ing
sys­tems for med­i­cal prod­ucts of human ori­gin. Bull World Health Organ.
2013;91(5):314-314A.
3. World Health Organization. WHO guid­ing prin­ci­ples on human cell, tis­sue
and organ trans­plan­ta­tion. Accessed 19 Octo­ber 2021. https:​­/​­/www​­.who​
­.int​­/transplantation​­/Guiding_PrinciplesTransplantation_WHA63​­.22en​­.pdf.
4. Gratwohl A, Pasquini MC, Aljurf M, et al; Worldwide Network for Blood and
Marrow Transplantation. One mil­lion haemopoietic stem-cell trans­plants: a
ret­ro­spec­tive obser­va­tional study. Lancet Haematol. 2015;2(3):e91-100.
5. Niederwieser D, Baldomero H, Bazuaye N, et  al. One and a half mil­lion
hema­to­poi­etic stem cell trans­plants: con­tin­u­ous and dif­fer­en­tial improve­
ment in world­wide access with the use of non-iden­ti­cal fam­ily donors. Pub-
lished online 12 August 2021. Haematologica.
6. Niederwieser D, Baldomero H, Szer J, et al. Hematopoietic stem cell trans­
plan­ta­tion activ­ity world­wide in 2012 and a SWOT anal­y­sis of the World-
wide Network for Blood and Marrow Transplantation Group includ­ing the
global sur­vey. Bone Marrow Transplant. 2016;51(6):778-785.
7. Gragert L, Eapen M, Williams E, et al. HLA match like­li­hoods for hema­to­poi­
etic stem-cell grafts in the U.S. reg­is­try. N Engl J Med. 2014;371(4):339-348.
8. Nunes K, Aguiar VRC, Silva M, et al. How ances­try influ­ences the chances
of find­ing unre­lated donors: an inves­ti­ga­tion in admixed Brazilians. Front
Immunol. 2020;11(6 Novem­ber):584950.
9. Paulson K, Brazauskas R, Khera N, et al. Inferior access to allo­ge­neic trans­plant
in dis­ad­van­taged pop­u­la­tions: a Center for International Blood and Marrow
Transplant Research anal­y­sis. Biol Blood Marrow Transplant. 2019;25(10):
2086-2090.
10. Milone G, Sacchi N, Gallina A, et al. Access to alter­na­tive donor hema­to­poi­
etic search and trans­plan­ta­tion for acute leu­ke­mia in dif­fer­ent macro-regions
of Italy: a GITMO/IBMDR study. Bone Marrow Transplant. 2018;53(3):291-299.
11. Hershenfeld SA, Matelski J, Ling V, Paterson M, Cheung M, Cram P. Utili-
sation and out­comes of allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion in
Ontario, Canada, and New York State, USA: a pop­u­la­tion-based ret­ro­spec­
tive cohort study. BMJ Open. 2020;10(10):e039293.
12. Bhatt VR, Chen B, Gyawali B, Lee SJ. Socioeconomic and health sys­tem fac­
tors asso­ci­ated with lower uti­li­za­tion of hema­to­poi­etic cell trans­plan­ta­tion
in older patients with acute mye­loid leu­ke­mia. Bone Marrow Transplant.
2018;53(10):1288-1294.
13. Joshua TV, Rizzo JD, Zhang MJ, et  al. Access to hema­to­poi­etic stem cell
trans­plan­ta­tion: effect of race and sex. Cancer. 2010;116(14):3469-3476.
14. Barker JN, Boughan K, Dahi PB, et al. Racial disparities in access to HLA-
matched unre­lated donor trans­plants: a pro­spec­tive 1312-patient anal­y­sis.
Blood Adv. 2019;3(7):939-944.
15. Bona K, Brazauskas R, He N, et  al. Neighborhood pov­erty and pedi­at­ric
allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion out­comes: a CIBMTR anal­y­
sis. Blood. 2021;137(4):556-568.
16. Delamater PL, Uberti JP. Geographic access to hema­to­poi­etic cell trans­plan­
ta­tion ser­vices in the United States. Bone Marrow Transplant. 2016;51(2):
241-248.
17. Khera N, Gooley T, Flowers MED, et al. Association of dis­tance from trans­
plan­ta­tion cen­ter and place of res­i­dence on out­comes after allo­ge­neic
 hema­to­poi­etic cell trans­plan­ta­tion. Biol Blood Marrow Transplant. 2016;
22(7):1319-1323.
18. Jamy O, Chen A, Battles K, et al. Impact of access to care on 1-year mor­
tal­ity fol­low­ing allo­ge­neic blood or mar­row trans­plan­ta­tion. Bone Marrow
Transplant. 2021;56(6):1364-1372.
19. Amonoo HL, Barclay ME, El-Jawahri A, Traeger LN, Lee SJ, Huffman JC.
Positive psy­cho­log­i­cal con­structs and health out­comes in hema­to­poi­etic
stem cell trans­plan­ta­tion patients: a sys­tem­atic review. Biol Blood Marrow
Transplant. 2019;25(1):e5-e16.
20. El-Jawahri A, Traeger L, Greer JA, et al. Effect of inpa­tient pal­li­a­tive care dur­
ing hema­to­poi­etic stem-cell trans­plant on psy­cho­log­i­cal dis­tress 6 months
after trans­plant: results of a ran­dom­ized clin­i­cal trial. J Clin Oncol. 2017;
35(32):3714-3721.
21. Ehrlich KB, Miller GE, Scheide T, et al. Pre-trans­plant emo­tional sup­port is
asso­ci­ated with lon­ger sur­vival after allo­ge­neic hema­to­poi­etic stem cell
trans­plan­ta­tion. Bone Marrow Transplant. 2016;51(12):1594-1598.
22. Preussler JM, Mau LW, Majhail NS, et al. Caregiver avail­abil­ity and patient
access to hema­to­poi­etic cell trans­plan­ta­tion: social worker per­spec­tives
inform prac­tice. Support Care Cancer. 2019;27(11):4253-4264.
23. Frey P, Stinson T, Siston A, et al. Lack of care­giv­ers lim­its use of out­pa­tient
hema­to­poi­etic stem cell trans­plant pro­gram. Bone Marrow Transplant.
2002;30(11):741-748.
24. Sundaramurthi T, Wehrlen L, Friedman E, Thomas S, Bevans M. Hemato-
poietic stem cell trans­plan­ta­tion recip­i­ent and care­giver fac­tors affect­ing
length of stay and readmission. Oncol Nurs Forum. 2017;44(5):571-579.
25. Silveira DRA, Coelho-Silva JL, Silva WF, et  al. A mul­ti­cen­ter com­par­a­tive
acute mye­loid leu­ke­mia study: can we explain the dif­fer­ences in the out­
comes in resource-constrained set­tings? Leuk Lymphoma. 2021;62(1):147-
157.
26. Giebel S, Labopin M, Ibatici A, et al. Association of mac­ro­eco­nomic fac­tors
with nonrelapse mor­tal­ity after allo­ge­neic hema­to­poi­etic cell trans­plan­
ta­tion for adults with acute lym­pho­blas­tic leu­ke­mia: an anal­y­sis from the
Acute Leukemia Working Party of the EBMT. Oncologist. 2016;21(3):377-
383.
27. Giebel S, Labopin M, Ehninger G, et al; Acute Leukemia Working Party of
the Euro­pean Group for Blood and Marrow Transplantation. Association
of Human Development Index with rates and out­comes of hema­to­poi­
etic stem cell trans­ plan­ta­
tion for patients with acute leu­ ke­mia. Blood.
2010;116(1):122-128.
28. Frankiewicz A, Peczynski C, Giebel S, et al. Association of coun­try-spe­cific
socio­eco­nomic fac­tors with sur­vival of patients who expe­ri­ence severe
clas­sic acute graft-vs.-host dis­ease after allo­ge­neic hema­to­poi­etic cell
trans­plan­ta­tion: an anal­y­sis from the Transplant Complications Working
Party of the EBMT. Front Immunol. 2020;11(July 23):1537.
29. Espigado I, Ortega-Ortega M, Montero-Granados R, Rodriguez-Torres N,
Márquez-Malaver FJ. Differences in stem cell trans­plan­ta­tion activ­ity among
regions in Spain: an eco­ nomic expla­ na­
tion. Bone Marrow Transplant.
2016;51(11):1537-1539.
30. Wood WA, Brazauskas R, Hu ZH, et al. Country-level mac­ro­eco­nomic indi­
ca­tors pre­dict early post-allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion
sur­vival in acute lym­pho­blas­tic leu­ke­mia: a CIBMTR anal­y­sis. Biol Blood
Marrow Transplant. 2018;24(9):1928-1935.
31. Hong S, Brazauskas R, Hebert KM, et al. Community health sta­tus and out­
comes after allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion in the United
States. Cancer. 2021;127(4):609-618.
32. da Silva WF, da Rosa LI, Seguro FS, et al. Salvage treat­ment for refrac­tory
or relapsed acute mye­loid leu­ke­mia: a 10-year sin­gle-cen­ter expe­ri­ence.
Clinics. 2020;75:1-8.
33. Arnold SD, Jin Z, Sands S, Bhatia M, Kung AL, Satwani P. Allogeneic hema­to­
poi­etic cell trans­plan­ta­tion for chil­dren with sickle cell dis­ease is ben­e­fi­cial
and cost-effec­tive: a sin­gle-cen­ter anal­y­sis. Biol Blood Marrow Transplant.
2015;21(7):1258-1265.
34. Chaudhri NA, Aljurf M, Almohareb FI, et al. Establishing an autol­o­gous
ver­sus allo­ge­neic hema­to­poi­etic cell trans­plant pro­gram in nations with
emerg­ing econ­o­mies. Hematol Oncol Stem Cell Ther. 2017;10(4):173-
177.
35. Majhail NS, Jagasia M. Referral to trans­plant cen­ter for hema­to­poi­etic cell
trans­plan­ta­tion. Hematol Oncol Clin North Am. 2014;28(6):1201-1213.
36. Snowden JA, Saccardi R, Orchard K, et al. Benchmarking of sur­vival out­
comes fol­low­ing haematopoietic stem cell trans­plan­ta­tion: a review of
existing pro­cesses and the intro­duc­tion of an inter­na­tional sys­tem from
Prakash Singh Shekhawat
the Euro­pean Society for Blood and Marrow Transplantation (EBMT) and
the Joint Accreditation Committee of ISCT and EBMT (JACIE). Bone Marrow
Transplant. 2020;55(4):681-694.
37. Pasquini MC, Srivastava A, Ahmed SO, et  al. Worldwide Network for
Blood and Marrow Transplantation (WBMT) rec­om­men­da­tions for estab-
lishing a hema­to­poi­etic cell trans­plan­ta­tion pro­gram (part I): min­i­mum
require­ments and beyond. Hematol Oncol Stem Cell Ther. 2020;13(3):131-
142.
38. Snoswell CL, Taylor ML, Comans TA, Smith AC, Gray LC, Caffery LJ. Deter-
mining if telehealth can reduce health sys­tem costs: scop­ing review. J Med
Inter­net Res. 2020;22(10):e17298.
39. DuHamel KN, Mosher CE, Winkel G, et  al. Randomized clin­ i­
cal trial of
tele­phone-admin­is­tered cog­ni­tive-behav­ioral ther­apy to reduce post-
trau­matic stress dis­or­der and dis­tress symp­toms after hema­to­poi­etic
stem-cell trans­plan­ta­tion. J Clin Oncol. 2010;28(23):3754-3761.
40. Syrjala KL, Yi JC, Artherholt SB, et al. An online ran­dom­ized con­trolled trial,
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021
with or with­out prob­lem-solv­ing treat­ment, for long-term can­cer sur­vi­vors
after hema­to­poi­etic cell trans­plan­ta­tion. J Cancer Surviv. 2018;12(4):560-
570.
41. Cordeiro A, Fatobene G, Mariano LCB, et al. Telehealth in hema­to­poi­etic
cell trans­plan­ta­tion: per­spec­tive from patients at a pub­lic hos­pi­tal in Brazil.
Blood. 2020;136(suppl 1):26.
42. Frutos C, Enciso ME, von Glasenapp A, Quiroz A, Batista J, Niederwieser D.
Bridging the gap using telemedicine: opti­miz­ing an existing autol­o­gous
hema­to­poi­etic SCT unit into an allo­ge­neic hema­to­poi­etic SCT unit in Para-
guay with the help of the WBMT. Blood Adv. 2019;3(suppl 1):45-47.
43. Mussetti A, Salas MQ , Condom M, et al. Use of telehealth for domi­cil­i­ary
fol­low-up after hema­to­poi­etic cell trans­plan­ta­tion dur­ing the COVID-19
pan­demic: pro­spec­tive pilot study. JMIR Form Res. 2021;5(3):e26121.
44. Lee TC, Kaiser TE, Alloway R, Woodle ES, Edwards MJ, Shah SA. Telemedi-
cine based remote home mon­i­tor­ing after liver trans­plan­ta­tion: results of a
ran­dom­ized pro­spec­tive trial. Ann Surg. 2019;270(3):564-572.
45. Abbasi J. Wearable dig­i­tal ther­mom­e­ter improves fever detec­tion. JAMA.
2017;318(6):510.
46. Owoyemi A, Owoyemi J, Osiyemi A, Boyd A. Artificial intel­ li­
gence for
healthcare in Africa. Front Digit Heal. 2020;2(July):6.
47. Gratwohl A, Baldomero H, Gratwohl M, et al; Worldwide Network of Blood
and Marrow Transplantation. Quantitative and qual­i­ta­tive dif­fer­ences in use
and trends of hema­to­poi­etic stem cell trans­plan­ta­tion: a global obser­va­
tional study. Haematologica. 2013;98(8):1282-1290.
48. Jaimovich G, Martinez Rolon J, Baldomero H, et al. Latin America: the next
region for haematopoietic trans­plant prog­ress. Bone Marrow Transplant.
2017;52(5):671-677.
49. Baldomero H, Aljurf M, Zaidi SZA, et al; Eastern Med­i­ter­ra­nean Blood and
Marrow Transplantation Group; Afri­ can Blood and Marrow Transplanta-
tion Group; Worldwide Network for Blood and Marrow Transplantation.
Narrowing the gap for hema­to­poi­etic stem cell trans­plan­ta­tion in the East-
Med­i­ter­ra­nean/Afri­can region: com­par­i­son with global HSCT indi­ca­tions
and trends. Bone Marrow Transplant. 2019;54(3):402-417.
50. Gratwohl A, Baldomero H, Aljurf M, et al; Worldwide Network of Blood and
Marrow Transplantation. Hematopoietic stem cell trans­plan­ta­tion: a global
per­spec­tive. JAMA. 2010;303(16):1617-1624.
51. Harif M, Weisdorf D, Novitzky N, et al. Special report: sum­mary of the first
meet­ing of Afri­can Blood and Marrow Transplantation (AfBMT) group, Casa-
blanca, Morocco, April 19–21, 2018 held under the aus­pices of the World-
wide Network for Blood and Marrow Transplantation (WBMT). Hematol
Oncol Stem Cell Ther. 2020;13(4):202-207.
52. El Fakih R, Greinix H, Koh M, et  al. Worldwide Network for Blood and
Marrow Transplantation (WBMT) rec­om­men­da­tions regard­ing essen­
tial med­ i­
ca­tions required to estab­ lish an early stage hema­ to­poi­etic
cell trans­plan­ta­tion pro­gram. Transplant Cell Ther. 2021;27(3):267.e1-
267.e5.
53. Aljurf M, Weisdorf D, Hashmi SK, et al. Worldwide Network for Blood and
Marrow Transplantation (WBMT) rec­ om­men­
da­tions for establishing a
hema­to­poi­etic stem cell trans­plan­ta­tion pro­gram in countries with lim­ited
resources (part II): clin­i­cal, tech­ni­cal and socio-eco­nomic con­sid­er­ations.
Hematol Oncol Stem Cell Ther. 2020;13(1):7-16.
54. Yoshimi A, Baldomero H, Horowitz M, et al; Worldwide Network of Blood
and Marrow Transplantation. Global use of periph­eral blood vs bone mar­row
as source of stem cells for allo­ge­neic trans­plan­ta­tion in patients with bone
mar­row fail­ure. JAMA. 2016;315(2):198-200.
Factors affect­ing global access to allotransplant  |  273
55. Cowan AJ, Baldomero H, Atsuta Y, et al. The global state of hema­to­poi­etic ing alter­nate donor reg­is­tries. Bone Marrow Transplant. 2019;54(8):1179-
cell trans­plan­ta­tion for mul­ti­ple mye­loma: an anal­y­sis of the Worldwide 1188.
Network of Blood and Marrow Transplantation data­base and the global
bur­den of dis­ease study. Biol Blood Marrow Transplant. 2020;26(12):2372-
2377.
56. Aljurf M, Weisdorf D, Alfraih F, et  al. Worldwide Network for Blood and © 2021 by The Amer­i­can Society of Hematology
Marrow Transplantation (WBMT) spe­cial arti­cle, chal­lenges fac­ing emerg­ DOI 10.1182/hema­tol­ogy.2021000258

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/264/1851868/264rocha.pdf by guest on 13 December 2021

274  |  Hematology 2021  |  ASH Education Program

 HOW CAN WE ENSURE THAT EVERYONE WHO NEEDS A TRANSPLANT CAN GET ONE ?

Increasing access to allotransplants in the

United States: the impact of race, geography,
and socioeconomics

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/275/1852104/275hong.pdf by guest on 13 December 2021

Sanghee Hong1 and Navneet S. Majhail2
Department of Hematology and Oncology, University Hospitals, Case Western Reserve University, Cleveland, OH; and 2Blood and Marrow
1

Transplant Program, Taussig Cancer Center, Cleveland Clinic, Cleveland, OH

Allogeneic hematopoietic cell transplantation (HCT) is particularly susceptible to racial, socioeconomic, and geographic
disparities in access and outcomes given its specialized nature and its availability in select centers in the United States.
Nearly all patients who need HCT have a potential donor in the current era, but racial minority populations are less likely
to have an optimal donor and often rely on alternative donor sources. Furthermore, prevalent health care disparity fac-
tors are further accentuated and can be barriers to access and referral to a transplant center. Research has primarily
focused on defining and quantifying a variety of social determinants of health and their association with access to allo-
geneic HCT, with a focus on race/ethnicity and socioeconomic status. However, research on interventions is lacking and
is an urgent unmet need. We discuss the role of racial, socioeconomic, and geographic disparities in access to allogeneic
HCT, along with policy changes to address and mitigate them and opportunities for future research.

LEARNING OBJECTIVES
• Understand the association of race, geography, and socioeconomic status with access to allogeneic transplanta-
tion in the United States
• Highlight opportunities to evaluate, mitigate, and address social access-related barriers to allogeneic transplantation

be referred to a transplant center in Cleveland, which is

CLINICAL CASE
A 40-year-old African American woman from rural Ohio
has relapsed acute myeloid leukemia (AML). She is hos-
pitalized at a regional hospital close to home to receive
salvage chemotherapy. She is a single mother with a
10-year-old son and lives in an area that has one of the
highest rates of poverty in Ohio. She does not have a car
and uses public transportation. She used to earn an hourly
wage as a waitress but has been unemployed with no
health care benefits for the past year since being laid off
at the onset of the COVID epidemic. She had been feel-
ing very fatigued and had noticed spontaneous bruising
for 4 weeks before the diagnosis of relapse; she was con-
cerned about leukemia recurrence but did not want to
see her oncologist given the lack of health insurance and
concern about paying medical bills. She does not have
any immediate family in the vicinity. Her oncologist has
discussed an allogeneic hematopoietic cell transplanta-
tion (HCT) for her AML and the fact that she will need to
Prakash Singh Shekhawat
90 miles away from where she lives.
Introduction
Although allogeneic HCT is potentially curative for many
patients with high-risk hematologic malignancies and oth-
er diseases, it is a highly specialized and complex proce-
dure that requires comprehensive clinical infrastructure to
facilitate referral, donor search, transplant hospitalization
and supportive care, and posttransplant follow-up. The
number of patients receiving allogeneic HCT continues to
increase in the US every year with improvements in tech-
nology and supportive care, use of less intense condition-
ing regimens that allow transplantation in older and frail
patients, and greater availability of suitable donors.1 How-
ever, it is also recognized that many patients who might
otherwise benefit do not receive allogeneic HCT.2-7 Several
patient-specific barriers to accessing HCT have been iden-
tified. Historically, a lack of suitably HLA-matched donors
Access to allogeneic transplantation in the US | 275
Table 1. Sociodemographic fac­tors asso­ci­ated with access to allo­ge­neic HCT

Referencea Population Access var­i­able(s) Key find­ings

Jabo et al3 Age ≥15 years; patients with
ALL/AML in California Cancer
Registry; 2003-2012
Dehn et al10 All ages; donor searches through
Be the Match reg­is­try; 2016
Barker et al12 Age ≤70 years; sin­gle-cen­ter study
of patients under­go­ing unre­
Age, race/eth­nic­ity, geog­ra­phy, Higher rate of HCT in patients aged ≤40 and in
SES mar­ried patients; women more likely to receive
HCT for ALL; lower rates of HCT in His­panic
and non-His­panic Black patients; no asso­ci­a­
tion between dis­tance and HCT uti­li­za­tion; low
neigh­bor­hood quin­tile SES index asso­ci­ated
with less uti­li­za­tion of HCT
Race/eth­nic­ity White patients more likely to receive HCT com­
pared to Black patients
Race/eth­nic­ity Patients of Euro­pean ances­try more likely to
receive 8/8 HLA-MUD HCT trans­plant than

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/275/1852104/275hong.pdf by guest on 13 December 2021

lated donor search; 2005-2017
Bhatt et al2 Age 61-75 years; National Cancer
Database, patients with AML;
2003-2012
Paulson et al5 Age <66 years; patients with
AML/ALL/MDS reported to
CIBMTR and SEER; 2000-2010
Delamater and Uberti18 All ages; sev­eral pub­lic data­bases
Getta et al27 Age ≤70 years; sin­gle-cen­ter study
of MDS patients; 2008-2015
Table shows rep­re­sen­ta­tive stud­ies in US pop­u­la­tions published since 2015.
a
non-Euro­pean ances­try and less likely to have
no MUD or cord-blood grafts
Age, race/eth­nic­ity, geog­ra­phy, Lower like­li­hood of receiv­ing HCT in patients
SES, insur­ance cov­er­age who were older, non-White, of lower edu­ca­
tional sta­tus, unin­sured, on Med­ic­aid/
Medi­care, or received care at non­ac­a­demic
facil­ity; no dif­fer­ence in HCT rates among urban
vs rural facil­ity; higher like­li­hood of receiv­ing
HCT in patients who lived ≥37 miles from facil­ity;
no asso­ci­a­tion with median house­hold income
SES, geog­ra­phy Higher county lev­els of pov­erty asso­ci­ated with
lower trans­plant rates; rural vs urban sta­tus
was not asso­ci­ated with HCT uti­li­za­tion
Geography Overall, 66% of US pop­ul­a­tion lives within 60 min­
utes’ travel time and 94% within 3 hours’ travel
time of HCT facil­ity; geo­graphic access to HCT
facil­ity varies by state
Age Patients ≥65 years were less likely to be referred
for HCT eval­u­a­tion; mar­i­tal sta­tus and insur­
ance type were not asso­ci­ated with trans­plant
refer­ral

MDS, myelodysplastic syn­drome.

used to be a bar­rier to HCT, but alter­na­tive donors (eg, hap-
loidentical, mismatched unre­lated, and umbil­i­cal cord blood) are
now used rou­tinely, and nearly all­patients have a suit­able donor
for trans­plan­ta­tion. However, age-related, racial, eco­nomic, and
other social disparities con­tinue to limit access to allo­ge­neic
HCT in the US, and many patients who would oth­er­wise ben­efi ­t
are not referred for and do not receive trans­plan­ta­tion.
As high­lighted by the case above, access to HCT is mod­er­
ated by a com­plex inter­play of sev­eral socio­cul­tural, eco­nomic,
dis­ease, treating pro­vider, hos­pi­tal-related, and health-sys­tem-
related fac­tors. At a patient level, disparities in access can more
fac­tors that often tend to be closely related (eg, race/eth­nic­ity,
insur­ance sta­tus, edu­ca­tion level, pov­erty, employ­ment sta­tus;
Table 1). Studies have established asso­ci­a­tions between age,
sex, race/eth­nic­ity, insur­ance cov­er­age, and socio­eco­nomic
sta­ tus (SES) and the uti­ li­
za­
tion of HCT, and less evi­ dence is
avail­­able for other fac­tors such as mar­i­tal sta­tus, lan­guage bar­
ri­ers, dis­tance from trans­plant cen­ter, and care­giver avail­abil­ity.7
Although the focus of this review is bar­ri­ers to access, the same
disparities also influ­ence short-term and long-term out­comes
fol­low­ing HCT, and the con­tem­po­rary lit­er­a­ture in this area is
sum­ma­rized in Table 2.
In a dis­cus­sion of disparities in access to trans­plan­ta­tion, it is
impor­tant to acknowl­edge the lim­i­ta­tions of the existing lit­er­a­
ture. Data on patients who receive HCT are robust and cap­tured
well by insti­tu­tional and national reg­is­tries (eg, the Center for
International Blood and Marrow Transplant Research [CIBMTR]).
However, data on patients who are can­ di­
dates for and may
poten­tially ben­e­fit from HCT are not read­ily avail­­able. National
reg­is­tries and sec­ond­ary data­bases (eg, the Surveillance, Epide-
miology, and End Results Program [SEER] and sin­gle- or multi-
payer data­bases) often do not include the details required to
deter­mine whether HCT was indi­cated for a given patient (eg,
dis­ease risk, remis­sion sta­tus, donor avail­abil­ity). Furthermore,
sociodemographic bar­ri­ers are a com­plex con­struct, are chal­
leng­ing to define, and are not cap­tured reli­ably at an indi­vid­ual
level; hence, most stud­ies focus on pop­ul­a­tion-level indi­ca­tors
to define disparities (eg, median house­hold income based on
zip code of res­i­dence). Studies in HCT recip­i­ents have eval­u­ated
com­pos­ite mea­sures that com­bine sev­eral health dis­par­ity fac­
tors, but these instru­ments need fur­ther val­i­da­tion.8,9 Qualitative
stud­ies are also needed to con­tex­tu­al­ize the quan­ti­ta­tive lit­er­
a­ture, to deepen our under­stand­ing of access bar­ri­ers, and to
iden­tify impactful and timely inter­ven­tions to address them.

276  |  Hematology 2021  |  ASH Education Program

Table 2. Sociodemographic and cen­ter fac­tors asso­ci­ated with out­comes of allo­ge­neic HCT

Referencea Population Variable(s) Key find­ings

Bona et al28 Age ≤18 years; all­diag­noses;
first allo­ge­neic HCT recip­i­ents
reported to CIBMTR; 2006-2015
Hong et al8 Age ≥18 years; all­diag­noses;
first allo­ge­neic HCT recip­i­ents
reported to CIBMTR; 2014-2016
SES
Several (county-level indi­ca­tors
of com­mu­nity health)
In chil­dren with malig­nant dis­ease, high neigh­bor­hood
pov­erty level asso­ci­ated with higher NRM and Med­ic­aid
insur­ance sta­tus asso­ci­ated with higher NRM and infe­
rior OS (vs pri­vate insur­ance); no asso­ci­a­tion between
neigh­bor­hood pov­erty and HCT out­comes for non­ma­
lig­nant dis­ease
Patients resid­ing in counties with worse com­mu­nity
health sta­tus had infe­rior OS; among patients with
hema­to­logic malig­nancy, worse com­mu­nity health
sta­tus was asso­ci­ated with infe­rior OS and higher risks
of NRM

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/275/1852104/275hong.pdf by guest on 13 December 2021

Madbouly et al29 All ages; all­ diag­noses; allo­ge­neic Race/eth­nic­ity (ances­try) Higher recip­i­ent-donor Afri­can genetic admix­ture asso­ci­
HCT using 10/10 allele matched ated with lower OS and DFS and higher NRM
MUD reported to CIBMTR; 1995–
2001
Majhail et al16 Adult HCT cen­ters; all­diag­noses; Center vol­ume Higher 100-day and 1-year OS in high-vol­ume (>40 allo­
allo­ge­neic HCT reported to ge­neic HCT/year) vs low-vol­ume cen­ters; pres­ence of
CIBMTR; 2008-2010 and 2012– sur­vi­vor­ship pro­gram asso­ci­ated with higher 1-year OS
2014
Khera et al30 Adult; all­diag­noses; first allo­ge­ Geography (dis­tance from HCT No asso­ci­a­tion of dis­tance and OS, NRM, or relapse;
neic HCT recip­i­ents at sin­gle cen­ter) trend toward higher NRM with increased dis­tance in
cen­ter; 2000-2010 nonmyeloablative HCT recip­i­ents
Bhatt et al31 All ages; hema­to­logic malig­nancy; Time to insur­ance approval Time to insur­ance approval for HCT var­ied between pri­
sin­gle-cen­ter study of first auto vate and pub­lic pay­ers but was not asso­ci­ated with OS
and allo­ge­neic HCT recip­i­ents;
2007-2011
Table shows rep­re­sen­ta­tive stud­ies in US pop­u­la­tions published since 2015.
a

DFS, dis­ease-free sur­vival; NRM, nonrelapse mor­tal­ity; OS, over­all sur­vival.

Racial bar­ri­ers to HCT
Race and eth­nic­ity are par­tic­u­larly rel­e­vant when con­sid­er­ing
access to allo­ge­neic HCT. First, racial disparities that are rou­
tinely prev­a­lent in health care apply to HCT and in fact may be
accen­tu­ated given the com­plex­ity and expense of the pro­ce­
dure along with its restricted avail­abil­ity in select cen­ters in the
US. Second, there is an ele­ment of donor avail­abil­ity asso­ci­ated
with race and very spe­cific to HCT. Unrelated donor reg­is­tries
are over­rep­re­sented by donors of Euro­pean ances­try, and White
patients have a higher chance of find­ing an HLA-matched unre­
lated donor (MUD).10-12 Using data from the National Marrow Do-
nor Program reg­is­try, Gragert et al showed that the like­li­hood of
find­ing a high-res­o­lu­tion HLA 8/8 allele MUD was 75% for White
peo­ple of Euro­pean descent and only 16% for Black peo­ple of
South or Central Amer­i­can ances­try.11 Given that the major­ity of
patients do not have an HLA-iden­ti­cal sib­ling donor, this dis­par­
ity in MUD avail­abil­ity by race/eth­nic­ity has sig­nif­i­cant impli­ca­
tions on trans­plant uti­li­za­tion and, ulti­mately, sur­vival and oth-
er out­comes after HCT. There is a pos­si­bil­ity that this dis­par­ity
may worsen in the future. In another anal­y­sis that mod­eled the
like­li­hood of find­ing HLA-iden­ti­cal sib­lings and unre­lated do-
nors, Besse et al showed that the aver­age num­ber of sib­lings
and sib­ling match prob­a­bil­ity vary by patient age and race, and
young minor­ity patients are at greatest risk for not find­ing an
HLA-matched donor.13
In addi­tion to race-related donor issues, minor­ity pop­u­la­tions
often have social and eco­nomic bar­ri­ers to refer­ral and donor
search. In gen­eral, His­panic and Black pop­u­la­tions in the US have
Prakash Singh Shekhawat
lower median house­ hold income com­ pared to Whites.14 This
dis­par­ity often trans­lates to expo­sure to adverse social deter­
mi­nants of health in the for­mer, includ­ing a greater like­li­hood of
resid­ing in areas with high pov­erty lev­els, inad­e­quate health care
cov­er­age, and lower lev­els of health lit­er­acy. Racial dis­par­ity in
access to HCT has been well documented, includ­ing a recent
study that showed adult His­panic and Black patients with AML
and acute lym­pho­blas­tic leu­ke­mia hav­ing a lower prob­a­bil­ity of
pro­ceed­ing with HCT.3 The mech­a­nism by which these social
deter­mi­nants of health have an impact on the ulti­mate receipt
of allo­ge­neic HCT is com­plex, as dem­on­strated in a study by
Clay et  al.15 They showed that the rea­sons for not receiv­ing a
trans­plant dif­fered by race. Patient deci­sion/treat­ment reluc­
tance and sta­ble dis­ease sta­tus not severe enough to war­rant
trans­plant were the most impor­tant rea­sons for not pro­ceed­
ing in Euro­pean Amer­i­can patients, whereas comorbidities and
phy­si­cian deci­sion were the main rea­sons for not pro­ceed­ing in
Afri­can Amer­i­can patients. Psychosocial or com­pli­ance con­cerns
were iden­ti­fied more often in Afri­can Amer­i­can patients as a rea­
son for not pro­ceed­ing with HCT.
Socioeconomic bar­ri­ers to HCT
In addi­tion to race/eth­nic­ity, the asso­ci­a­tion of SES with access
to HCT has been well documented, with most stud­ies using
US Census tract data to define SES. This is a lim­i­ta­tion of the
existing lit­er­a­ture since SES is most accu­rate when it is patient
self-reported. Regardless, using zip codes to esti­mate median
house­hold income is a well-val­i­dated method for defin­ing SES in
Access to allo­ge­neic trans­plan­ta­tion in the US  |  277

Figure 1. Framework for investigating interventions to address racial, socioeconomic, and geographic disparities in access to
allogeneic HCT.
health care research. In the con­tem­po­rary lit­er­a­ture, Jabo et al
have reported an inverse asso­ci­a­tion of neigh­bor­hood SES with
HCT uti­li­za­tion; com­pared to the highest-quin­tile SES, patients
resid­ing in the low­est quin­tile had a lower like­li­hood of under­go­
ing HCT (adjusted rel­a­tive risk, 0.63; 95% CI, 0.47-0.84 for acute
lym­pho­blas­tic leu­ke­mia and 0.52; 0.43-0.64, for AML).3 Similar-
ly, Paulson et al, using data from the CIBMTR, showed that res­
i­dence in counties with high lev­els of pov­erty was asso­ci­ated
with a lower prob­a­bil­ity of receiv­ing HCT (in mul­ti­var­i­able anal­
y­sis, esti­mated rate ratio was 0.86 per 10% increase in county
pop­u­la­tion below the pov­erty line; P  <  .01).5
Studies that have been ­able to inves­ti­gate the role of social
and eco­ nomic deter­ mi­
nants in greater detail sug­ gest some
mech­a­nisms by which SES influ­ences access to HCT. In a study
using the National Cancer Database, Bhatt et al found that the
pri­mary payer for HCT cov­er­age was sig­nif­i­cantly asso­ci­ated
with HCT uti­li­za­tion.2 Compared to pri­vate insur­ance, patients
were less likely to receive HCT if they had Med­ic­aid (odds ratio
[OR], 0.3; 95% CI, 0.3-0.5; P < .0001), Medi­care (OR, 0.7; 0.6-0.8;
P < .0001), unin­sured (OR, 0.2; 0.1-0.5; P  = .0003), and unknown
insur­ance sta­tus (OR, 0.1; 0.1-0.3; P  <  .0001).
Geographic bar­ri­ers to HCT
Given its spe­cial­ized and highly reg­u­lated nature, need for expe­
ri­enced per­son­nel, and infrastructural require­ments, HCT is avail­­
able through approx­i­ma­tely 200 trans­plant pro­grams in the US.
There is a ratio­nale for restricting HCT to select cen­ters since
a vol­ume-out­come rela­tion­ship has been dem­on­strated for this
pro­ce­dure.16,17 However, this does cause a bar­rier to some pa-
tients who need to travel long dis­tances to access a trans­plant
cen­ter. Overall, 48% and 79% of the US adult pop­ul­a­tion and
43% and 72% of the pedi­at­ric pop­u­la­tion have access to an HCT
facil­ity within 30 and 90 min­utes’ travel time from their homes,
278  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/275/1852104/275hong.pdf by guest on 13 December 2021
respec­tively.18 There is sig­nif­i­cant var­i­a­tion by state; eg, >70%
of adult res­i­dents in Arizona, Maryland, New Jersey, New York,
Rhode Island, and Washington, DC, live within 30 min­utes of an
HCT, whereas 6% of the US pop­u­la­tion must travel >3 hours to
access a trans­plant facil­ity.18
Since the major­ity of the US pop­u­la­tion live in rea­son­ably
close prox­im­ity to an HCT cen­ter, stud­ies have shown no defin­i­
tive asso­ci­at­ ion between dis­tance from the trans­plant pro­gram
or rural/urban res­i­dence sta­tus and receipt of allo­ge­neic HCT.2,3
Interestingly, a recent study has shown that patients a ­ ble to
travel lon­ger dis­tances (≥37 miles) for care were more likely to
receive trans­plant, pos­si­bly indi­cat­ing bet­ter patient sta­tus or
bet­ter receipt of care in ter­tiary refer­ral hos­pi­tals.2 As illus­trated
in the case at the begin­ning of this arti­cle, geo­graphic disparities
are likely accen­tu­ated in patients who are socio­eco­nom­i­cally
under­served to begin with.
Although not spe­cif­i­cally a focus of this review, other social
deter­mi­nants of health can be related to racial, socio­eco­nomic,
and geo­graphic disparities and ulti­mately affect access to allo­ge­
neic HCT. Some exam­ples of such bar­ri­ers include age, sex, patient
pref­er­ence, edu­ca­tional sta­tus, health lit­er­acy level, psy­chi­at­ric
dis­abil­ity, sub­stance abuse, mar­i­tal sta­tus, lan­guage bar­ri­ers, and
lack of com­pli­ance with med­i­cal care.7
Opportunities to address disparities in access to HCT
Research to date in the field of HCT has largely focused on
under­stand­ing and defin­ing health care disparities in access
to and out­comes of allo­ge­neic HCT, and acknowl­edg­ing and
quan­ ti­
fy­
ing them is an impor­ tant first step. Less work has
been done around inves­ti­gat­ing inter­ven­tions to resolve or
mit­i­gate these disparities. Some rea­sons for this are related to
the long-stand­ing sys­temic inequities in health care that need
to be addressed at the soci­e­tal level, the lack of val­i­dated
health care inter­ven­tions to address these disparities, the dif­
fi­culty in gen­er­al­iz­ing stud­ies given that social deter­mi­nants
may vary at the local and indi­vid­ual lev­els, and the fact that
such disparities are often out­side the con­trol of what a trans­
plant cen­ter can real­is­ti­cally influ­ence. Furthermore, the addi­
tional resources and effort required to bring in patients from
dis­ad­van­taged pop­u­la­tions for HCT may not be pri­or­i­tized at
the refer­ring pro­vider and trans­plant cen­ter level. Neverthe-
less, oppor­tu­ni­ties exist for inter­ven­tions to ensure that pa-
tients with racial, socio­eco­nomic, and geo­graphic chal­lenges
receive appro­pri­ate trans­plant-related care (Figure 1).
A major time point for inter­ven­tion to improve access is refer­
ral from a patient’s oncol­o­gist to the trans­plant cen­ter so that
indi­ca­tion and can­di­dacy for trans­plan­ta­tion can be deter­mined,
and a donor search can be ini­ti­ated in a timely man­ner.19 Early
refer­ral in the dis­ease course also allows for the iden­ti­fi­ca­tion of
psy­cho­so­cial and SES fac­tors that may later hin­der pro­ceed­ing
with trans­plan­ta­tion so that they can be addressed early. Social
work­ers and care coor­di­na­tors are an inte­gral part of the trans­
plant team and play an impor­tant role in this early eval­u­a­tion
and in gath­er­ing appro­pri­ate resources for patients. Some exam­
ples of such inter­ven­tions include refer­rals for grants to off­set
out-of-pocket costs, assessing care­giver sup­port, and help­ing
with local hous­ing for patients who must tem­po­rar­ily relo­cate
to be close to the trans­plant cen­ter.20,21 Particularly, pro­grams
often require a ded­i­cated care­giver for HCT recip­i­ents, and inter­
ven­tions to iden­tify and sup­port care­giv­ers are needed.21 Addi-
tionally, the devel­op­ment and implementation of tools to assess
social chal­lenges can facil­i­tate indi­vid­u­al­ized inter­ven­tions for
patients. Some instru­ments that have been eval­u­ated in HCT
recip­i­ents include the Psychosocial Assessment of Candidates
for Transplantation scale, Transplant Evaluation Rating Scale, and
Stanford Integrated Psychosocial Assessment for Transplant.22-25
Data are needed for spe­cial pop­u­la­tions (eg, les­bian, gay, bisex­
ual, trans­gen­der, queer patients; health-illit­er­ate indi­vid­u­als; or
non­na­tive English speak­ers) so that appro­pri­ate inter­ven­tions
can be designed to help them access HCT.
In this same con­text, opti­miz­ing health care deliv­ery for allo­ge­
neic HCT recip­i­ents by improv­ing care coor­di­na­tion among pri­
mary care phy­si­cians, refer­ring hema­tol­o­gists/oncol­o­gists, and
trans­plant cen­ters can help mit­i­gate the social bar­ri­ers and chal­
lenges that patients and fam­i­lies face. Khera et al have described
a patient-cen­tered care frame­work to coor­di­nate allo­ge­neic HCT
deliv­ery.26 They lay out var­ i­ous phases in the HCT con­ tin­
uum
and the roles of var­i­ous health care pro­vid­ers. Their frame­work
can be expanded to address social dis­par­ity-related fac­tors that
play a role through­out the trans­plant jour­ney. It can also inform
research on inter­ven­tions since all­stake­hold­ers, includ­ing health
care pro­vid­ers from out­side the trans­plant cen­ter net­work, need
to be engaged to address these disparities. A com­mon thread for
patients as they move through dif­fer­ent phases and sites of care
is their payer, and we need to explore oppor­tu­ni­ties for pro­vider-
payer col­lab­o­ra­tions to address bar­ri­ers to trans­plant.
There is an urgent need for more research and funding to
sup­port the inves­ti­ga­tion of inno­va­tive inter­ven­tions to address
socio­eco­nomic and geo­graphic disparities in access to allo­ge­
neic HCT. There is also a soci­e­tal respon­si­bil­ity to address health
care sys­tem fac­tors that operate at a sys­temic level and ulti­mately
have an impact on access to HCT. Although cel­lu­lar ther­apy was
not the focus of this review, sim­i­lar health care dis­par­ity fac­tors
Prakash Singh Shekhawat
also apply and may in fact be worse given the costs of newer chi­
me­ric anti­gen recep­tor T-cell ther­a­pies. Ultimately, the ben­e­fit of
inno­va­tions in HCT and cel­lu­lar ther­apy can be fully real­ized when
all­patients who may ben­e­fit actu­ally receive these pro­ce­dures.
CLINICAL CASE (Con­tin­ued)
The treating hema­tol­o­gist-oncol­o­gist contacted the trans­plant
cen­ter early in the patient’s treat­ment course. Local and trans­
plant cen­ter social work­ers were a ­ ble to enroll the patient in
a state Med­ic­aid pro­gram and refer her for grants and other
ser­vices (eg, transportation assis­tance). An ini­tial HCT con­sult
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/275/1852104/275hong.pdf by guest on 13 December 2021
was conducted through telemedicine, and a donor search was
ini­ti­ated. A close friend stepped in to serve as a ded­i­cated care­
giver dur­ing trans­plant. A MUD was iden­ti­fied, and the patient
was ­able to suc­cess­fully pro­ceed with an allo­ge­neic trans­plant.
Conflict-of-inter­est dis­clo­sure
Sanghee Hong: no com­pet­ing finan­cial inter­ests to declare.
Navneet S. Majhail: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Sanghee Hong: none discussed.
Navneet S. Majhail: none discussed.
Correspondence
Navneet S. Majhail, Blood and Marrow Transplant Program,
Taussig Cancer Center, Cleveland Clinic, 9500 Euclid Ave, CA60,
Cleveland, OH 44195; e-mail: majhain@ccf​­.org.
References
1. D’Souza A, Fretham C, Lee SJ, et al. Current use of and trends in hema­to­
poi­etic cell trans­plan­ta­tion in the United States. Biol Blood Marrow Trans-
plant. 2020;26(8):e177-e182.
2. Bhatt VR, Chen B, Gyawali B, Lee SJ. Socioeconomic and health sys­tem fac­
tors asso­ci­ated with lower uti­li­za­tion of hema­to­poi­etic cell trans­plan­ta­tion
in older patients with acute mye­loid leu­ke­mia. Bone Marrow Transplant.
2018;53(10):1288-1294.
3. Jabo B, Mor­gan JW, Martinez ME, Ghamsary M, Wieduwilt MJ. Sociodemo-
graphic disparities in che­mo­ther­apy and hema­to­poi­etic cell trans­plan­ta­tion
uti­li­za­tion among adult acute lym­pho­blas­tic and acute mye­loid leu­ke­mia
patients. PLoS One. 2017;12(4):e0174760.
4. Majhail NS, Omondi NA, Denzen E, Mur­phy EA, Rizzo JD. Access to hema­to­
poi­etic cell trans­plan­ta­tion in the United States. Biol Blood Marrow Trans-
plant. 2010;16(8):1070-1075.
5. Paulson K, Brazauskas R, Khera N, et al. Inferior access to allo­ge­neic trans­plant
in dis­ad­van­taged pop­u­la­tions: a cen­ter for inter­na­tional blood and mar­row
trans­plant research anal­y­sis. Biol Blood Marrow Transplant. 2019;25(10):2086-
2090.
6. Khera N, Deeg HJ, Kodish E, Rondelli D, Majhail N. Allogeneic hema­
to­poi­etic cell trans­plan­ta­tion and other expen­sive cel­lu­lar ther­a­pies: a
mir­a­cle for the few but off lim­its to many? J Clin Oncol. 2020;38(12):1268-
1272.
7. Flannelly C, Tan BE, Tan JL, et al. Barriers to hema­to­poi­etic cell trans­plan­
ta­tion for adults in the United States: a sys­tem­atic review with a focus on
age. Biol Blood Marrow Transplant. 2020;26(12):2335-2345.
8. Hong S, Brazauskas R, Hebert KM, et al. Community health sta­tus and out­
comes after allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion in the United
States. Cancer. 2021;127(4):609-618.
9. Hong S, Rybicki LA, Corrigan D, Schold JD, Majhail NS. Community risk
score for eval­u­at­ing health care disparities in hema­to­poi­etic cell trans­plan­
ta­tion. Biol Blood Marrow Transplant. 2018;24(4):877-879.
10. Dehn J, Chitphakdithai P, Shaw BE, et al. Likelihood of pro­ceed­ing to allo­
ge­neic hema­to­poi­etic cell trans­plan­ta­tion in the United States after search
Access to allo­ge­neic trans­plan­ta­tion in the US  |  279
 acti­va­tion in the national reg­is­try: impact of patient age, dis­ease, and
search prog­no­sis. Transplant Cell Ther. 2021;27(2):184.e181-184.e113.
11. Gragert L, Eapen M, Williams E, et al. HLA match like­li­hoods for hema­to­
poi­etic stem-cell grafts in the U.S. reg­is­try. N Engl J Med. 2014;371(4):339-
348.
12. Barker JN, Boughan K, Dahi PB, et al. Racial disparities in access to HLA-
matched unre­lated donor trans­plants: a pro­spec­tive 1312-patient anal­y­sis.
Blood Adv. 2019;3(7):939-944.
13. Besse K, Maiers M, Confer D, Albrecht M. On mod­el­ing human leu­ko­cyte
anti­gen-iden­ti­cal sib­ling match prob­a­bil­ity for allo­ge­neic hema­to­poi­etic
cell trans­plan­ta­tion: esti­mat­ing the need for an unre­lated donor source.
Biol Blood Marrow Transplant. 2016;22(3):410-417.
14. Guzman GG. Household income by race and His­panic ori­gin: 2005–2009 and
2015–2019. US Census Bureau Amer­i­can Community Survey Brief ACSBR-7.
Decem­ber 2020. Accessed 16 Sep­tem­ber 2021. https:​­/​­/www​­.census​­.gov​­
/content​­/dam​­/Census​­/library​­/publications​­/2020​­/acs​­/acsbr19​­-07​­.pdf.
15. Clay A, Peoples B, Zhang Y, et  al. Population-based anal­y­sis of hema­to­
logic malig­nancy refer­rals to a com­pre­hen­sive can­cer cen­ter, refer­rals for
blood and mar­row trans­plan­ta­tion, and par­tic­i­pa­tion in clin­i­cal trial, sur­
vey, and biospecimen research by race. Biol Blood Marrow Transplant.
2015;21(8):1488-1494.
16. Majhail NS, Mau LW, Chitphakdithai P, et  al. Transplant cen­ter char­ac­ter­is­
tics and sur­vival after allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion in adults.
Bone Marrow Transplant. 2020;55(5):906-917.
17. Majhail NS, Mau LW, Chitphakdithai P, et  al. National sur­vey of hema­to­
poi­etic cell trans­plan­ta­tion cen­ter per­son­nel, infra­struc­ture, and mod­els of
care deliv­ery. Biol Blood Marrow Transplant. 2015;21(7):1308-1314.
18. Delamater PL, Uberti JP. Geographic access to hema­to­poi­etic cell trans­plan­
ta­tion ser­vices in the United States. Bone Marrow Transplant. 2016;51(2):
241-248.
19. Majhail NS, Jagasia M. Referral to trans­plant cen­ter for hema­to­poi­etic cell
trans­plan­ta­tion. Hematol Oncol Clin North Am. 2014;28(6):1201-1213.
20. Preussler JM, Mau LW, Majhail NS, et al. Patient hous­ing bar­ri­ers to hema­to­
poi­etic cell trans­plan­ta­tion: results from a mixed-meth­ods study of trans­
plant cen­ter social work­ers. Support Care Cancer. 2016;24(3):1167-1174.
21. Preussler JM, Mau LW, Majhail NS, et al. Caregiver avail­abil­ity and patient
access to hema­to­poi­etic cell trans­plan­ta­tion: social worker per­spec­tives
inform prac­tice. Support Care Cancer. 2019;27(11):4253-4264.
280  |  Hematology 2021  |  ASH Education Program
22. Hong S, Rybicki L, Corrigan D, et  al. Psychosocial Assessment of Candi-
dates for Transplant (PACT) as a tool for psy­cho­log­i­cal and social eval­u­
a­tion of allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion recip­i­ents. Bone
Marrow Transplant. 2019;54(9):1443-1452.
23. Harashima S, Yoneda R, Horie T, et al. Psychosocial Assessment of Candidates
for Transplantation scale (PACT) and sur­vival after allo­ge­neic hema­to­poi­etic
stem cell trans­plan­ta­tion. Bone Marrow Transplant. 2019;54(7):1013-1021.
24. Solh MM, Speckhart D, Sol­o­mon SR, et al. The Transplant Evaluation Rat-
ing Scale pre­dicts over­all sur­vival after allo­ge­neic hema­to­poi­etic stem cell
trans­plan­ta­tion. Blood Adv. 2020;4(19):4812-4821.
25. Hoodin F, Kalbfleisch KR. Factor anal­y­sis and validity of the Transplant Eval-
uation Rating Scale in a large bone mar­row trans­plant sam­ple. J Psycho-
som Res. 2003;54(5):465-473.
26. Khera N, Martin P, Edsall K, et al. Patient-cen­tered care coor­di­na­tion in hema­
to­poi­etic cell trans­plan­ta­tion. Blood Adv. 2017;1(19):1617-1627.
27. Getta BM, Kishtagari A, Hilden P, et al. Allogeneic hema­to­poi­etic stem cell
trans­plan­ta­tion is underutilized in older patients with myelodysplastic syn­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/275/1852104/275hong.pdf by guest on 13 December 2021
dromes. Biol Blood Marrow Transplant. 2017;23(7):1078-1086.
28. Bona K, Brazauskas R, He N, et  al. Neighborhood pov­erty and pedi­at­ric
allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion out­comes: a CIBMTR anal­y­
sis. Blood. 2021;137(4):556-568.
29. Madbouly A, Wang T, Haagenson M, et al. Investigating the asso­ci­a­tion of
genetic admix­ture and donor/recip­i­ent genetic dis­par­ity with trans­plant
out­comes. Biol Blood Marrow Transplant. 2017;23(6):1029-1037.
30. Khera N, Gooley T, Flowers MED, et  al. Association of dis­ tance from
trans­plan­ta­tion cen­ter and place of res­i­dence on out­comes after allo­ge­
neic hema­to­poi­etic cell trans­plan­ta­tion. Biol Blood Marrow Transplant.
2016;22(7):1319-1323.
31. Bhatt VR, Loberiza FR Jr, Schmit-Pokorny K, Lee SJ. Time to insur­ ance
approval in pri­ vate and pub­ lic pay­
ers does not influ­ ence sur­ vival in
patients who undergo hema­to­poi­etic cell trans­plan­ta­tion. Biol Blood Mar-
row Transplant. 2016;22(6):1117-1124.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000259
 IMMUNOLOGY 101: WHAT THE PRACTICING HEMATOLOGIST NEEDS TO KNOW

Immunology 101: fundamental immunology

for the practicing hematologist

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/281/1851987/281carty.pdf by guest on 13 December 2021

Shannon A. Carty
Division of Hematology and Oncology, Department of Internal Medicine, University of Michigan, Ann Arbor, MI

From an evolutionary perspective, the immune system developed primarily to protect the host from pathogens. In the
continuous balance between killing pathogens and protecting host tissues, selective pressures have shaped the discrim-
inatory functions of the immune system. In addition to protection against microbial pathogens, the immune system also
plays a critical role in antitumor immunity. Immune dysfunction, either under- or overactivity, is found in a wide range of
hematologic disorders. Here we review the fundamental features of the immune system and the key concepts critical to
understanding the impact of immune dysfunction on hematologic disorders.

LEARNING OBJECTIVES
• Describe the cardinal features of the innate vs adaptive arms of the immune system
• Define the mechanisms that result in an inadequate immune response
• Understand the immunotherapeutic approaches to overcome immune dysfunction

age. This response is mediated by germ line­encoded pattern

CLINICAL CASE
A 70­year­old man in previously good health presents with 6
weeks of fatigue and drenching night sweats. The physical
exam is notable for cervical, axillary, and inguinal lymphade­
nopathy. Laboratory studies are significant for mild anemia
and an elevated lactate dehydrogenase level. Lymph node
biopsy reveals CD20+CD10+Bcl6+ large B cells consistent
with a diffuse large B­cell lymphoma, germinal center sub­
type. He is treated with rituximab, cyclophosphamide, doxo­
rubicin hydrochloride, vincristine sulfate, and prednisone,
also known as R­CHOP, immunochemotherapy for 6 cycles
and obtains a complete remission. However, his lymphoma
relapses within 6 months, and he has progressive disease
despite salvage immunochemotherapy. He then discusses
chimeric antigen receptor T cells (CAR­T) therapy vs a bispe­
cific T­cell engager (BiTE) clinical trial with his hematologist.
Introduction
Innate immune system
The immune system is classically divided into 2 primary
arms—the innate and the adaptive (Table 1). The cardinal fea­
ture of the innate immune system is the rapid but nonspecific
response to a broad repertoire of pathogens and tissue dam­
Prakash Singh Shekhawat
recognition receptors (PRRs) that can recognize conserved
features of pathogens or damaged cells, termed pathogen­
associated molecular patterns or damage­associated molec­
ular patterns. Cells of the innate immune system include
granulocytes,monocytes/macrophages,dendriticcells(DCs),
and natural killer (NK) cells. Through PRRs, innate immune
cells recognize broadly conserved structures present in a
large group of micro­organisms, such as bacterial lipopoly­
saccharide, a component of gram­negative bacteria, or
pathogen­associated nucleic acids such as double­stranded
RNA. There are 4 main classes of PRRs: toll­like receptors,
C­type lectin receptors, NOD­like receptors, and RIG­I­like
receptors.
Innate immune cells exert their effector functions through
the phagocytosis of infected cells and/or extracellular organ­
isms, the production of inflammatory cytokines, and the
release of soluble mediators, such as cytokines and chemo­
kines. Different pathogens trigger distinctive innate immune
responses as distinct effector functions are needed for effec­
tive clearance. For instance, neutrophils are required to pro­
tect against certain fungal pathogens, such as Aspergillus
fumigatus and Candida albicans,1 the putative underlying
mechanism for the increased risk of invasive fungal infections
in patients with prolonged neutropenia following high­dose
chemotherapy or allogeneic hematopoietic stem cell (HSC)
Fundamentals of immunology | 281
Table 1. Key char­ac­ter­is­tics of innate vs adap­tive immune cells
Innate Adaptive
Timing of response Minutes to hours Days
Self vs non­self No Yes
Diversity Limited Extensive
Receptors Broad, germ line encoded Highly spe­cific,
somat­i­cally
rearranged
Cell types Granulocytes, mac­ro­phages, B cells, T cells
DCs, NK cells
trans­plant.2 Furthermore, some innate immune cells such as DCs
and mac­ ro­
phages func­
tion as anti­gen presenting cells (APCs).
Antigen pre­sen­ta­tion is a pro­cess through which small pep­tides
are incor­po­rated into the bind­ing groove of the major his­to­com­
pat­i­bil­ity com­plex (MHC; also known as human leu­ko­cyte anti­gen)
pro­teins. During an infec­tion, small pep­tides derived from path­o­
gen pro­teins are presented in the con­text of self-MHC on the sur­
face of APCs to help acti­vate T lym­pho­cytes and thus the adap­tive
immune response.
More recently, immune cells termed “innate lym­phoid cells”
(ILCs), which have fea­tures of both the innate and adap­tive immune
cells, have been rec­og­nized. ILCs respond quickly to infec­tion and
do not express anti­gen recep­tors like innate immune cells but pro­
duce sim­i­lar inflam­ma­tory cyto­kines as T lym­pho­cytes.3
Adaptive immune sys­tem
In con­trast to the innate immune response, the car­di­nal fea­
tures of the adap­tive immune response include anti­gen spec­i­
fic­ity and the for­ma­tion of immu­no­log­i­cal mem­ory, which is the
abil­ity of lym­pho­cytes to respond to a pre­vi­ously encoun­tered
anti­ gen rap­ idly and more effi­ ciently upon reexposure. B and
T lym­pho­cytes are the pri­mary cells of the adap­tive immune sys­
tem. Both lin­ea ­ ges express anti­gen recep­tors, the B-cell recep­
tor (BCR) and T-cell recep­tor (TCR), respec­tively, which undergo
somatic rearrangement to allow B and T cells to rec­og­nize and
respond to spe­cific anti­gens. B cells are the pri­mary medi­at­ors
of the humoral (ie, anti­body-medi­ated) immune response, which
func­tions to pro­tect the host from extra­cel­lu­lar micro-organ­isms
and pre­vent the spread of intra­cel­lu­lar path­o­gens. T cells medi­ate
the cel­lu­lar immune response, which both sup­ports the humoral
immu­nity and acts as the pri­mary effec­tor cell to kill virally infected
or transformed host cells.
Lymphocytes of the adap­tive immune sys­tem share com­mon
fea­tures of anti­gen spec­i­fic­ity and the for­ma­tion of long-lived
immune mem­ory. Specificity is achieved by each lym­pho­cyte
bear­ing an anti­gen recep­tor that rec­og­nizes a sin­gle epi­tope,
which is attained by the somatic rearrangement of sev­eral sets
of gene seg­ments encoding the BCR, which is also a mem­brane-
bound form of antibodies or immu­no­glob­u­lins (Ig), and TCRs to
gen­er­ate recep­tor diver­sity. During B- and T-cell devel­op­ment,
the var­i­able (V), diver­sity (D), and join­ing (J) gene seg­ments are
somat­i­cally rearranged through DNA recom­bi­na­tion medi­ated
by the recom­bi­na­tion-acti­vat­ing genes (RAG1/2) in a rel­a­tively
anal­og ­ ous pro­cess occur­ring at the light and heavy chains of
the immunoglobin loci or the α and β chains of the TCR locus.4-7
The com­bi­na­to­rial diver­sity of V(D)J recom­bi­na­tion is fur­ther
282  |  Hematology 2021  |  ASH Education Program
enhanced with an addi­tional layer of junc­tional diver­sity when
nucle­ot­ides are added or subtracted between dif­fer­ent gene
seg­ments dur­ing the recom­bi­na­tion pro­cess. Each resul­tant
B or T cell expresses an anti­gen recep­tor with unique spec­i­fic­ity.
Estimates sug­gest that such diver­sity could lead to 1015 to 1018
unique BCR com­bi­na­tions and 1015 to 1020 dis­tinct TCRs; how­ever,
unique func­tional pop­u­la­tions are sig­nif­i­cantly smaller due to
devel­op­men­tal selec­tion events. Exploiting this fact, poly­mer­ase
chain reac­tions to detect somat­i­cally rearranged immu­no­glob­u­
lin or TCR loci are a com­monly used molec­u­lar hematopathology
tech­nique to eval­u­ate for clonality in pathol­ogy spec­i­mens.
B cells: B cells develop in the bone mar­row from HSCs, where
they undergo sev­eral pro­gres­sive stages of lin­e­age spec­i­fi­ca­
tion and com­mit­ment. B-cell com­mit­ment and devel­op­ment are
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/281/1851987/281carty.pdf by guest on 13 December 2021
reg­u­lated by a coor­di­nated net­work of tran­scrip­tion fac­tors as
well as accom­pa­ny­ing epi­ge­netic changes. In the bone mar­row,
devel­op­ing B cells rearrange the heavy-chain and then the light-
chain immu­no­glob­u­lin loci. Following the gen­er­a­tion of an anti­
gen recep­tor, devel­op­ing B cells must pass sev­eral check­points
to pre­vent the pro­duc­tion of self-reac­tive lym­pho­cytes.8 Once
each chain is suc­cess­fully rearranged, a pro­cess called alle­lic
exclu­sion occurs to pre­vent rearrangement of the other allele,
which allows each indi­vid­ual cell to pro­duce only a recep­tor of
sin­gle-anti­gen spec­i­fic­ity. Light-chain loci also undergo isotypic
exclu­sion, in which only 1 type of light chain (κ or λ) is expressed
by each indi­vid­ual B cell. Once a func­tional BCR forms (as a sur­
face IgM), the anti­gen recep­tor is tested for self-reac­tiv­ity in a
pro­cess known as cen­tral tol­er­ance to elim­i­nate autoreactive
cells. Immature B cells then migrate out of the bone mar­row
to the spleen, where they undergo fur­ther mat­ur­a­tion. In the
periph­ery, B cells that encoun­ter and rec­og­nize self-anti­gen in
the absence of infec­tion will be deleted or become anergic in
a pro­cess known as periph­eral tol­er­ance. Failure of these tol­
er­ance mech­a­nisms can result in B-cell-medi­ated auto­im­mu­nity.
Naive (ie, anti­gen-inex­pe­ri­enced) B cells are acti­vated by sol­u­
ble anti­gen through their BCR, which then ini­ti­ates an intra­cel­lu­lar
sig­nal­ing cas­cade. The BCR can also deliver the anti­gen intra­cel­
lu­larly for anti­gen processing to allow B cells to pres­ent anti­genic
pep­tides to T cells. Some anti­gens can acti­vate B cells with­out
T cell help (T inde­pen­dent) or require T cell help (T depen­dent).
In T-depen­dent responses, B cells receive addi­tional acti­va­tion
sig­nals from CD4+ T fol­lic­u­lar helper cells that rec­og­nizes the anti­
genic pep­tide. Activated B cells can dif­fer­en­ti­ate into sev­eral dif­
fer­ent mature sub­sets,9 includ­ing plasmablasts and lon­ger-lived
plasma cells, as well as migrate to the ger­mi­nal cen­ter to undergo
fur­ther mat­u­ra­tion, includ­ing somatic hypermutation of the
V regions of the immu­no­glob­u­lin genes to gen­er­ate higher-affin­ity
antibodies (affin­ity mat­u­ra­tion) and class switching to allow
B cells to pro­duce antibodies with dif­fer­ent effec­tor func­tions
(IgG, IgE, IgA). Molecular sub­types of dif­fer­ent lym­pho­mas have
sim­il­ar gene expres­sion sig­na­tures as nor­mal B-cell sub­sets, such as
the ger­mi­nal cen­ter B-cell sub­type and the acti­vated B-cell sub­type
(or non-ger­mi­nal cen­ter B cell) of dif­fuse large B-cell lym­phoma,10
which can pre­dict response to cer­tain targeted ther­ap ­ ies.11
T cells: T lym­pho­cytes are one of the few cells of the immune
sys­tem that develop out­side the bone mar­row. T cell pre­cur­
sors migrate from the bone mar­row to the thy­mus, where they
mature into func­tional T cells. Collectively, mature T cells must
pos­sess a diverse TCR rep­er­toire capa­ble of rec­og­niz­ing the
immense num­ber of for­eign anti­gens that will be encoun­tered
over the host’s life­time. In con­trast to B cells that respond to sol­
u­ble anti­gen, T cells are stim­u­lated by small anti­genic pep­tides
presented on the sur­face of other cells by MHC mol­e­cules. Since
the TCR binds anti­genic pep­tides plus some amino acid res­i­dues
of self-MHC pro­tein, it is crit­i­cal that only T cells with a TCR ­able
to rec­og­nize self-MHC, albeit with lim­ited affin­ity, be exported to
the periph­ery. In addi­tion, it is essen­tial that T cells pos­sess TCRs
that only respond to for­eign anti­gen and that those that pos­
sess TCRs rec­og­niz­ing self-pep­tide in the con­text of self-MHC
are elim­i­nated dur­ing devel­op­ment to pre­vent auto­im­mu­nity.
Similar to B-cell devel­op­ment, devel­op­ing T cells undergo sev­
eral key check­points to ensure the devel­op­ment of func­tional,
non-self-reac­tive T cells.12 In the thy­mus, devel­op­ing αβT cells
undergo 2 fur­ther check­points, known as pos­i­tive and neg­a­tive
selec­tion, to deter­mine TCR fit­ness. Positive selec­tion occurs
when thy­mo­cytes inter­act with self-anti­gens presented in the
con­text of class I or class II MHC on cor­ti­cal thy­mic epi­the­lial
cells. The thy­mo­cytes that engage anti­gen/MHC with inter­me­
di­ate affin­ity are protected from apo­pto­sis, whereas those cells
that can­not inter­act with self-anti­gen/MHC die by apo­pto­sis.
The remaining cells then undergo neg­ a­
tive selec­ tion, dur­ing
which cells that react too strongly to self-anti­gen/MHC undergo
apo­pto­sis. This pro­cess helps pro­tect the host from T cells that
pos­ sess TCRs with high reac­ tiv­
ity against self-pep­ tides pre­
sented by self-MHC, which would lead to auto­im­mu­nity. Follow­
ing pos­i­tive and neg­a­tive selec­tion, only remaining thy­mo­cytes
with mod­er­ate affin­ity for pep­tide/MHC undergo lin­e­age com­
mit­ment to become CD4+ or CD8+ T cells, which then migrate to
the periph­ery. Within this con­tin­uum of per­mit­ted reac­tiv­ity, the
devel­op­ing CD4+ thy­mo­cytes that pos­sess the highest affin­ity for
self-pep­tide/MHC develop into reg­u­la­tory T cells (Tregs), char­
ac­ter­ized by the expres­sion of the tran­scrip­tion fac­tor FoxP3 and
the sur­face expres­sion of CD25. Tregs are a spe­cial­ized sub­set of
CD4+ T cells that sup­press acti­vated CD4+ and CD8+ T cells as well
as other immune cells.
Once in the periph­ery, naive T cells cir­cu­late until the TCR
rec­og­ni­tion of pep­tide-MHC complexes induces an intra­cel­lu­
lar bio­chem­i­cal sig­nal­ing cas­cade that acti­vates a pro­gram of
clonal pro­lif­er­a­tion and dif­fer­en­ti­a­tion (with addi­tional input
from costimulatory mol­e­cules and cyto­kines), thus transform­
ing the naive T cell into an effec­tor T cell. CD4+ and CD8+ T cells
undergo anal­o­gous dif­fer­en­ti­a­tion pro­cesses, over the span of
sev­ eral days, to acquire func­ tional matu­ rity but play dis­ tinct
func­tional roles in the adap­tive immune response to path­o­gens.
Naive cells of both lin­e­ages are acti­vated through their TCRs,
and their dif­fer­en­ti­a­tion is influ­enced by a com­bi­na­tion of sig­
nals, includ­ing TCR sig­nal strength, costimulatory ligands, and
the local cyto­kine milieu. The inte­gra­tion of these sig­nals drives
the expres­sion of key tran­scrip­tion fac­tors and effec­tor mol­e­
cules, which endow the acti­vated T cell with its indi­vid­u­al­ized
func­tion. Activated CD8+ T cells func­tion to induce death in host
cells fol­low­ing acti­va­tion, whereas CD4+ T cells func­tion pri­mar­
ily through cyto­kine pro­duc­tion or via direct cell-to-cell con­tact
to acti­vate other immune cells. Depending on the sig­nals dur­ing
dif­fer­en­ti­a­tion, naive CD4+ T cells have the poten­tial to develop
into 1 of sev­eral prin­ci­pal effec­tor (or “helper”) sub­sets, includ­ing
TH1, TH2, TH17, and T fol­lic­u­lar helper (TFH) cells.13 Different sub­sets
of periph­eral T-cell lym­pho­mas have been described that likely
arise from these sub­sets.14,15 In addi­tion to the thymically derived
Prakash Singh Shekhawat
Tregs, naive CD4+ T cells can also be induced to become Tregs
and exert immu­no­sup­pres­sive func­tions in the appro­pri­ate con­
text. CD8+ T cells become ter­mi­nally dif­fer­en­ti­ated effec­tor cells
or long-lived mem­ory cells fol­low­ing anti­gen stim­u­la­tion.16
What con­sti­tutes an effec­tive immune response?
A suc­cess­ful immune response results in the clear­ance of the path­
o­gen and the for­ma­tion of immu­no­logic mem­ory while ­pre­vent­ing
host dam­age. It requires the care­ful and ­inte­grated orches­tra­tion
of a mul­ti­step pro­cess involv­ing diverse cell types, cyto­kines, and
ana­tomic local­i­za­tion. A ­pro­to­typ­i­cal immune response can be
bro­ken down into 3 main stages: (1) acute inflam­ma­tory response
to the invad­ing path­o­gen by the innate immune sys­tem, (2) anti­
gen pre­sen­ta­tion, and (3) ­adap­tive immune response involv­ing
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/281/1851987/281carty.pdf by guest on 13 December 2021
CD4+ and CD8+ T cells and B ­lym­pho­cytes.
Innate acute inflam­ma­tory response: The first line of defense
is typ­i­cally bar­rier sur­faces that seek to pre­vent path­o­gen entry
into the host, includ­ ing the epi­ the­ lial sur­
faces and muco­ sal
­sur­faces of the respi­ra­tory, gas­tro­in­tes­ti­nal, or uro­gen­i­tal tracts.
Following entry through these bar­ri­ers, most path­o­gens cause a
local infec­tion in the tis­sues. Innate immune cells rec­og­nize path­
o­gen-asso­ci­ated molec­u­lar pat­terns and dam­age-asso­ci­ated
molec­u­lar pat­terns through their germ line-encoded PRRs,
which ini­ti­ates an intra­cel­lu­lar sig­nal­ing cas­cade to acti­vate their
unique effec­tor func­tions. Cytokine and chemokine pro­duc­tion
pro­motes local inflam­ma­tion, attracting cir­cu­lat­ing innate effec­
tor cells such as neu­tro­phils and mono­cytes. If the ini­tial innate
immune response fails to con­ trol the infec­ tion, the adap­ tive
immune response is trig­gered in local lym­phoid tis­sues by the
pre­sen­ta­tion of path­o­gen-derived anti­gens by APCs.
Antigen pre­sen­ta­tion: Following path­o­gen expo­sure, APCs
pro­cess for­eign pep­tides and pres­ent them on self-MHC. MHC
class I and class II mol­e­cules func­tion sim­i­larly to deliver and
pres­ent short pep­tides on the cell sur­face to allow these pep­
tides to be rec­og­nized by CD8+ or CD4+ T cells, respec­tively.
The pep­tides presented by the dif­fer­ent clas­ses orig­i­nate from
either intra­cel­lu­lar pro­teins for MHC class I or engulfed extra­cel­
lu­lar pro­teins for MHC class II. All cells of the body are at risk of
malig­nant trans­for­ma­tion or infec­tion by intra­cel­lu­lar microbes,
which requires a CD8+ T cell response, and thus all­nucle­ated
cells express MHC class I. Meanwhile, only pro­fes­sional APCs,
such as DCs, mac­ro­phages, and B cells, express MHC class II.
Additionally, a sub­set of APCs can pres­ent inter­nal­ized anti­gens
in the con­text of MHC class I mol­e­cules via a mech­a­nism called
cross-pre­sen­ta­tion, thus allowing CD8+ T-cell acti­va­tion against
path­o­gens or tumors that do not directly infect DCs. Inflamma­
tory cyto­ kines help acti­ vate APCs and pro­ mote upregulation
costimulatory mol­e­cules, such as B7.1 (CD80) and B7.2 (CD86)
on the APC cell sur­face. Activated APCs then travel from the site
of infec­tion to draining lymph nodes, where they pres­ent anti­
gen to naive T cells. T cells that rec­og­nize anti­gen:self-MHC on
an acti­vated APC pro­ceed to sig­nal through the multimeric TCR
com­plex. Costimulation by the bind­ing of CD28 on the T-cell sur­
face to a B7 mol­e­cule on the APC is required for T cell acti­va­tion.
A fail­ure to receive both the TCR (sig­nal 1) and costimulation (sig­
nal 2) typ­i­cally leads to T-cell anergy and apo­pto­sis. T cell acti­
va­tion can be fur­ther mod­u­lated by addi­tional costimulatory or
coinhibitory sig­nals.17 Activating sig­nals include those pro­vided
via 4-1BB (CD37), OX40 (CD134), and CD27.
Fundamentals of immu­nol­ogy  |  283
 Adaptive immune response: Following acti­va­tion by APCs,
anti­gen-spe­cific T cells and B cells undergo clonal expan­sion
and dif­fer­en­ti­a­tion over sev­eral days. B cells pro­duce path­o­gen-
spe­cific antibodies that can acti­vate com­ple­ments to directly kill
path­o­gens, opsonize path­o­gens to pro­mote phago­cy­to­sis, and
pro­mote anti­body-depen­dent cell-medi­ated cyto­tox­ic­ity. Acti­
vated CD4+ T cells release cyto­kines to enhance other immune
cell func­tion, such as mac­ro­phage-medi­ated phago­cy­to­sis.
Effector CD8+ T cells release proinflammatory cyto­kines and can
directly kill infected cells via cytol­y­sis. Following path­o­gen clear­
ance, the major­ity of anti­gen-spe­cific lym­pho­cytes undergo cell
death, resulting in clonal con­trac­tion. A small sub­set remains
as long-lived mem­ory lym­pho­cytes that have the capac­ity to
respond to the same path­o­gen more rap­idly upon rechallenge,
lead­ing to less severe or sub­clin­i­cal infec­tion.
Importantly, the immune sys­tem has sev­eral mech­a­nisms to
pre­serve a care­ful bal­ance between allowing appro­pri­ate acti­
va­tion of the immune sys­tem and pre­vent­ing immune overacti­
vation and sub­se­quent tis­sue dam­age of the host. For exam­ple,
in con­trast to the costimulatory sig­nals outlined above, coinhib­
itory sig­nals, such as those medi­ated by ligands to programmed
cell death 1 pro­tein (PD-1), cyto­toxic T-lym­pho­cyte anti­gen
4 (CTLA-4), and lym­pho­cyte-acti­va­tion gene 3 (LAG3), serve to
sup­press immune acti­va­tion and act as a bio­log­ic ­ al rheo­stat to
pre­vent tis­sue dam­age from a hyper­ac­tive immune response.
In addi­tion, Tregs medi­ate their immu­no­sup­pres­sive func­tion in
trans on other immune cells, includ­ing CD4+ and CD8+ T cells,
B cells, DCs, NK cells, and mac­ro­phages, through CTLA-4 expres­
sion; pro­duc­tion of the sup­pres­sive cyto­kines IL-10, IL-35, and
transforming growth fac­ tor β; the con­ ver­sion of extra­ cel­lu­
lar
aden­o­sine tri­phos­phate to aden­o­sine; and the con­sump­tion of
local IL-2.
Mechanisms that result in an inad­e­quate immune
response
The increased risk of malig­nan­cies that develop in patients with
pri­mary (more recently termed inborn errors of immu­nity) or
sec­ond­ary (acquired) immunodeficiencies high­lights the impor­
tance of immune sur­veil­lance to pre­vent can­cer.18 The malig­
nant trans­for­ma­tion of host cells pres­ents unique chal­lenges
to the immune response since the cells are self in ori­gin, and
self-tol­er­ance may pre­vent the devel­op­ment of a full immune
response. Multiple mech­a­nisms in the tumor micro­en­vi­ron­ment
dampen effec­tive anti­tu­mor responses (Table 2). The selec­tive
Table 2. Mechanisms that dampen immune response in the
tumor micro­en­vi­ron­ment and cor­re­spond­ing immu­no­ther­a­
peu­tic approaches
Mechanisms to evade immune
response Immunotherapeutic approaches
Loss of anti­ge­nic­ity of tumor cells CAR-T, BiTE
(ex: loss of MHC, downregulation
of costimulatory mol­e­cules)
Increased expres­sion of inhib­i­tory Immune check­point block­ade
ligands on tumor cells or in
micro­en­vi­ron­ment
Suppressive immune pop­u­la­tions in CD47 block­ade (exper­i­men­tal)
tumor micro­en­vi­ron­ment
284  |  Hematology 2021  |  ASH Education Program
pres­ sures of an immune response can pro­ mote the loss of
anti­ge­nic­ity of the tumor itself through sev­eral poten­tial mech­
a­nisms. Lymphomas, and less com­monly leu­ke­mias, can lose
MHC class I or class II expres­sion.19-23 Additionally, tumor cells
can downregulate costimulatory mol­e­cules to pre­vent T cell
acti­va­tion, dysregulation of anti­gen-processing machin­ery can
lead to the loss of anti­gen pre­sen­ta­tion, and selec­tive pres­
sure could lead to the out­growth of a tumor clone that no lon­
ger expresses rec­ og­
nized anti­ gens. Furthermore, the tumor
micro­en­vi­ron­ment in many hema­to­logic malig­nan­cies is immu­
no­sup­pres­sive, with the upregulation of inhib­i­tory ligands on
the malig­nant cells or other cells. For instance, cer­tain lym­
phoma cells may overexpress PD-L1/2, which may be driven
by viral infec­tion or genetic alter­ations.24-26 Immunosuppressive
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/281/1851987/281carty.pdf by guest on 13 December 2021
cells in the tumor micro­en­vi­ron­ment, such as Tregs or mac­ro­
phages, can also dampen the immune response to leu­ke­mias
and lym­pho­mas.27
Modulating the immune sys­tem to over­come immune
dys­func­tion
Immunotherapeutic approaches to over­come or bypass immune
dys­func­tion have had clin­i­cal suc­cess in many hema­to­logic dis­
or­ders. The most long-stand­ing exam­ple of immu­no­ther­apy to
treat hema­to­logic malig­nan­cies is the allo­ge­neic HSC trans­plant,
first used in humans in 1957.28 Discoveries aris­ing from decades
of research have found that the reconstituted donor-derived
immune sys­tem, par­tic­u­larly donor lym­pho­cytes, medi­ates a
graft-ver­sus-leu­ke­mia/lym­phoma effect but also leads to graft-
ver­sus-host dis­ease, dur­ing which acti­vated donor T cells rec­og­
nize the host tis­sue as for­eign, lead­ing to tis­sue dam­age.
More recent immu­no­ther­a­peu­tic advances have been built
on the rec­og­ni­tion that tumor-spe­cific T cells undergo a pro­
cess termed exhaus­tion, dur­ing which chronic anti­gen stim­u­
la­tion results in T cells becom­ing increas­ingly less respon­sive,
lead­ing to decreased cyto­kine pro­duc­tion and the inabil­ity to
pro­mote cytol­y­sis.29 Concomitantly, exhausted cells upregu­
late inhib­i­tory cell sur­face recep­tors. The best stud­ied of these
inhib­i­tory recep­tors is PD-1, which binds its ligands, PD-L1
and PD-L2, expressed on acti­ vated mac­ ro­phages and other
APCs. The engage­ment of PD-1 damp­ens the T-cell response,
which normally occurs after ini­tial TCR acti­va­tion, likely to pre­
vent exces­sive responses and is sub­se­quently downregulated.
Exhausted T cells, how­ever, con­tinue to express this inhib­i­tory
recep­tor.30 Therapeutic targeting of the PD-1 axis has shown effi­
cacy in solid tumors and some lym­pho­mas, includ­ing clas­si­cal
Hodgkin lym­phoma and pri­mary medi­as­ti­nal B-cell lym­phoma.31-34
However, only a sub­ set of clin­ i­
cal responses is long-last­
ing,
pos­ si­
bly because PD-1/PD-L1 block­ ade alters sig­
nal­
ing path­
ways within the responding T cells but not the epi­ge­netic land­
scape.35,36 Current clin­i­cal tri­als are ongo­ing to exam­ine whether
the com­bi­na­tion of epi­ge­netic ther­apy—ie, with hypomethyl­
ating agents or his­ tone deacetylase inhib­ i­
tors—and immune
check­point block­ade improves response rates or dura­bil­ity.
Another recent advance in mod­u­lat­ing the immune sys­tem
to fight hema­to­logic malig­nan­cies or viral infec­tion is the use
of redirected autol­ o­gous T cells. The most explored of these
approaches is the use of CAR-T to treat relapsed/refrac­tory
CD19-expressing B-cell malig­nan­cies.37 This approach modifies
autol­o­gous T cells with the intro­duc­tion of an engineered anti­
gen recep­tor to pro­mote an anti­tu­mor response. The engineered
­ nti­gen recep­tor com­bines an extra­cel­lu­lar domain of a sin­gle-
a 10. Alizadeh AA, Eisen MB, Davis RE, et al. Distinct types of dif­fuse large B-cell
chain var­i­able frag­ment, which is a fusion of the var­i­able regions of
the heavy and light chains of an immu­no­glob­u­lin rec­og­niz­ing the
CD19 anti­gen, with intra­cel­lu­lar domains of the CD3ζ chain and
the sig­nal­ing domain of a costimulatory mol­e­cule, such as CD28
or 4-1BB. The addi­tion of the costimulatory domain improves the
per­sis­tence and effi­cacy of engineered T cells,38 which mir­rors the
known require­ment of costimulation for nor­mal T cell func­tion.
Another method of redirecting autol­o­gous T cells includes the
use of BiTEs, which are fusion pro­teins containing 2 anti­body rec­
og­ni­tion domains.39 One domain rec­og­nizes the CD3 com­plex on
autol­o­gous T cells, and a sec­ond rec­og­nizes a sur­face recep­tor
pres­ent on the tumor cells, such as CD19 or CD20 on B-cell malig­
nan­cies. BiTEs bring the T cells into the prox­im­ity of the tumor
cells, and the engage­ment of both domains pro­motes the acti­va­
tion of T cells, cyto­kine pro­duc­tion, and cyto­tox­ic­ity inde­pen­dent
of the TCR spec­i­fic­ity of the T cell. Currently, only blinatumomab,
a CD3xCD19 BiTE, is approved for B-cell pre­cur­sor acute lym­pho­
blas­tic leu­ke­mia, but many oth­ers are cur­rently under­go­ing clin­i­
cal tri­als for B-cell leu­ke­mias and lym­pho­mas.
A prom­is­ing, yet underexplored, approach is that of targeting
the immu­no­sup­pres­sive tumor micro­en­vi­ron­ment in hema­to­logic
malig­nan­cies to pro­mote a more effec­tive T-cell response. One
mech­a­nism to achieve this out­come would be to induce mac­ro­
phage phago­cy­to­sis of transformed cells and thereby enhance
tumor anti­gen pre­sen­ta­tion to host T lym­pho­cytes. This could be
achieved by block­ade of CD47 (also known as the “don’t eat me
sig­nal”) on tumor cells, which enhances tumor anti­gen pre­sen­ta­
tion to T cells.40 Clinically, CD47 block­ade in lym­phoma patients
has shown some poten­tially prom­is­ing early-phase clin­i­cal trial
results.41,42
Conflict-of-inter­est dis­clo­sure
Shannon A. Carty: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Shannon A. Carty: nothing to disclose.
Correspondence
Shannon A. Carty, University of Michigan, 109 Zina Pitcher Pl, Rm
1526, Ann Arbor, MI 48109; e-mail: scarty@med​­.umich​­.edu.
References
1. Romani L, Mencacci A, Cenci E, Del Sero G, Bistoni F, Puccetti P. An immu­
no­reg­u­la­tory role for neu­tro­phils in CD4+ T helper sub­set selec­tion in mice
with can­di­di­a­sis. J Immunol. 1997;158(5):2356-2362.
2. Lionakis MS, Levitz SM. Host con­trol of fun­gal infec­tions: les­sons from basic
stud­ies and human cohorts. Annu Rev Immunol. 2018;36(April):157-191.
3. Vivier E, Artis D, Colonna M, et al. Innate lym­phoid cells: 10 years on. Cell.
2018;174(5):1054-1066.
4. Brack C, Hirama M, Lenhard-Schuller R, Tonegawa S. A com­plete immu­no­
glob­u­lin gene is cre­ated by somatic recom­bi­na­tion. Cell. 1978;15(1):1-14.
5. Seidman JG, Edgell MH, Leder P. Immunoglobulin light-chain struc­tural gene
sequences cloned in a bac­te­rial plas­mid. Nature. 1978;271(5645):582-585.
6. Yanagi Y, Yoshikai Y, Leggett K, Clark SP, Aleksander I, Mak TW. A human T
cell-spe­cific cDNA clone encodes a pro­tein hav­ing exten­sive homol­ogy to
immu­no­glob­u­lin chains. Nature. 1984;308(5955):145-149.
7. Hedrick SM, Cohen DI, Nielsen EA, Davis MM. Isolation of cDNA clones
encoding T cell-spe­cific mem­brane-asso­ci­ated pro­teins. Nature. 1984;308
(5955): 149-153.
8. Melchers F. Checkpoints that con­trol B cell devel­op­ment. J Clin Invest.
2015;125(6):2203-2210.
9. Cyster JG, Allen CDC. B cell responses: cell inter­ac­tion dynam­ics and deci­
sions. Cell. 2019;177(3):524-540.
Prakash Singh Shekhawat
lym­phoma iden­ti­fied by gene expres­sion pro­fil­ing. Nature. 2000;403(6769):
503-511.
11. Wilson WH, Young RM, Schmitz R, et al. Targeting B cell recep­tor sig­nal­
ing with ibrutinib in dif­fuse large B cell lym­phoma. Nat Med. 2015;21(8):
922-926.
12. Shah DK, Zúñiga-Pflücker JC. An over­view of the intrathymic intri­ca­cies of
T cell devel­op­ment. J Immunol. 2014;192(9):4017-4023.
13. Zhu J. T helper cell dif­fer­en­ti­a­tion, het­ero­ge­ne­ity, and plas­tic­ity. Cold Spring
Harb Perspect Biol. 2018;10(10):a030338.
14. Wang T, Feldman AL, Wada DA, et al. GATA-3 expres­sion identifies a high-
risk sub­set of PTCL, NOS with dis­ tinct molec­u­lar and clin­ i­cal fea­tures.
Blood. 2014;123(19):3007-3015.
15. de Leval L, Rickman DS, Thielen C, et  al. The gene expres­sion pro­file of
nodal periph­eral T-cell lym­phoma dem­on­strates a molec­u­lar link between
angioimmunoblastic T-cell lym­phoma (AITL) and fol­lic­u­lar helper T (TFH)
cells. Blood. 2007;109(11):4952-4963.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/281/1851987/281carty.pdf by guest on 13 December 2021
16. Jameson SC, Masopust D. Understanding sub­set diver­sity in T cell mem­ory.
Immunity. 2018;48(2):214-226.
17. Chen L, Flies DB. Molecular mech­a­nisms of T cell co-stim­u­la­tion and co-
inhi­bi­tion. Nat Rev Immunol. 2013;13(4):227-242.
18. Kebudi R, Kiykim A, Sahin MK. Primary immu­no­de­fi­ciency and can­cer in
chil­dren: a review of the lit­er­a­ture. Curr Pediatr Rev. 2019;15(4):245-250.
19. Roemer MG, Advani RH, Redd RA, et al. Classical Hodgkin lym­phoma with
reduced β2M/MHC class I expres­sion is asso­ci­ated with infe­rior out­come
inde­pen­dent of 9p24.1 sta­tus. Cancer Immunol Res. 2016;4(11):910-916.
20. Roemer MGM, Redd RA, Cader FZ, et al. Major his­to­com­pat­i­bil­ity com­plex
class II and programmed death ligand 1 expres­sion pre­dict out­come after
programmed death 1 block­ade in clas­sic Hodgkin lym­phoma. J Clin Oncol.
2018;36(10):942-950.
21. Rimsza LM, Roberts RA, Miller TP, et al. Loss of MHC class II gene and pro­
tein expres­sion in dif­fuse large B-cell lym­phoma is related to decreased
tumor immunosurveillance and poor patient sur­vival regard­less of other
prog­nos­tic fac­tors: a fol­low-up study from the Leukemia and Lymphoma
Molecular Profiling Project. Blood. 2004;103(11):4251-4258.
22. Roberts RA, Wright G, Rosenwald AR, et al. Loss of major his­to­com­pat­i­bil­
ity class II gene and pro­tein expres­sion in pri­mary medi­as­ti­nal large B-cell
lym­ phoma is highly coor­ di­nated and related to poor patient sur­ vival.
Blood. 2006;108(1):311-318.
23. Brouwer RE, van der Heiden P, Schreuder GM, et al. Loss or downregula­
tion of HLA class I expres­sion at the alle­lic level in acute leu­ke­mia is infre­
quent but func­tion­ally rel­e­vant, and can be restored by inter­feron. Hum
Immunol. 2002;63(3):200-210.
24. Green MR, Rodig S, Juszczynski P, et al. Constitutive AP-1 activ­ity and EBV
infec­tion induce PD-L1 in Hodgkin lym­pho­mas and posttransplant lymph­
oproliferative dis­or­ders: impli­ca­tions for targeted ther­apy. Clin Cancer
Res. 2012;18(6):1611-1618.
25. Chen BJ, Chapuy B, Ouyang J, et al. PD-L1 expres­sion is char­ac­ter­is­tic of a
sub­set of aggres­sive B-cell lym­pho­mas and virus-asso­ci­ated malig­nan­cies.
Clin Cancer Res. 2013;19(13):3462-3473.
26. Roemer MG, Advani RH, Ligon AH, et  al. PD-L1 and PD-L2 genetic alter­
ations define clas­si­cal Hodgkin lym­phoma and pre­dict out­come. J Clin
Oncol. 2016;34(23):2690-2697.
27. Höpken UE, Rehm A. Targeting the tumor micro­en­vi­ron­ment of leu­ke­mia
and lym­phoma. Trends Cancer. 2019;5(6):351-364.
28. Cieri N, Maurer K, Wu CJ. 60 years young: the evolv­ing role of allo­ge­neic
hema­to­poi­etic stem cell trans­plan­ta­tion in can­cer immu­no­ther­apy. Cancer
Res. 2021;81(17):4373-4384.
29. McLane LM, Abdel-Hakeem MS, Wherry EJ. CD8 T cell exhaus­tion dur­ing
chronic viral infec­tion and can­cer. Annu Rev Immunol. 2019;37(April):457-495.
30. Day CL, Kaufmann DE, Kiepiela P, et  al. PD-1 expres­sion on HIV-spe­cific
T  cells is asso­ci­ated with T-cell exhaus­tion and dis­ease pro­gres­sion.
Nature. 2006;443(7109):350-354.
31. Ansell SM, Lesokhin AM, Borrello I, et al. PD-1 block­ade with nivolumab in
relapsed or refrac­tory Hodgkin’s lym­phoma. N Engl J Med. 2015;372(4):
311-319.
32. Younes A, Santoro A, Shipp M, et  al. Nivolumab for clas­si­cal Hodgkin’s
lym­phoma after fail­ure of both autol­o­gous stem-cell trans­plan­ta­tion and
brentuximab vedotin: a multicentre, multicohort, sin­gle-arm phase 2 trial.
Lancet Oncol. 2016;17(9):1283-1294.
33. Zinzani PL, Ribrag V, Moskowitz CH, et al. Safety and tol­er­a­bil­ity of pem­
brolizumab in patients with relapsed/refrac­tory pri­mary medi­as­ti­nal large
B-cell lym­phoma. Blood. 2017;130(3):267-270.
Fundamentals of immu­nol­ogy  |  285
34. Armand P, Rodig S, Melnichenko V, et al. Pembrolizumab in relapsed or refrac­
tory pri­mary medi­as­ti­nal large B-cell lym­phoma. J Clin Oncol. 2019;37(34):
3291-3299.
35. Pauken KE, Sammons MA, Odorizzi PM, et al. Epigenetic sta­bil­ity of exhausted
T cells lim­its dura­bil­ity of reinvigoration by PD-1 block­ade. Science. 2016;
354(6316):1160-1165.
36. Ghoneim HE, Fan Y, Moustaki A, et  al. De novo epi­ge­netic pro­grams
inhibit PD-1 block­ade-medi­ated T cell reju­ve­na­tion. Cell. 2017;170(1):142-157.
e19157e19.
37. June CH, O’Connor RS, Kawalekar OU, Ghassemi S, Milone MC. CAR T cell
immu­no­ther­apy for human can­cer. Science. 2018;359(6382):1361-1365.
38. Milone MC, Fish JD, Carpenito C, et  al. Chimeric recep­ tors containing
CD137 sig­nal trans­duc­tion domains medi­ate enhanced sur­vival of T cells
and increased anti­leu­ke­mic effi­cacy in vivo. Mol Ther. 2009;17(8):1453-1464.
286  |  Hematology 2021  |  ASH Education Program
39. Goebeler ME, Bargou RC. T cell-engag­ing ther­a­pies—BiTEs and beyond.
Nat Rev Clin Oncol. 2020;17(7):418-434.
40. Liu X, Pu Y, Cron K, et al. CD47 block­ade trig­gers T cell-medi­ated destruc­
tion of immu­no­genic tumors. Nat Med. 2015;21(10):1209-1215.
41. Advani R, Flinn I, Popplewell L, et  al. CD47 block­ade by Hu5F9-G4 and
rituximab in non-Hodgkin’s lym­phoma. N Engl J Med. 2018;379(18):1711-1721.
42. Ansell SM, Maris MB, Lesokhin AM, et al. Phase I study of the CD47 blocker
TTI-621 in patients with relapsed or refrac­tory hema­to­logic malig­nan­cies.
Clin Cancer Res. 2021;27(8):2190-2199.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000260
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/281/1851987/281carty.pdf by guest on 13 December 2021
 IMMUNOLOGY 101: WHAT THE PRACTICING HEMATOLOGIST NEEDS TO KNOW

How immunodeficiency can lead to malignancy

Sung-Yun Pai,1 Kathryn Lurain,2 and Robert Yarchoan2

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/287/1852069/287pai.pdf by guest on 13 December 2021

Immune Deficiency Cellular Therapy Program, Center for Cancer Research, National Cancer Institute, Bethesda, MD; and 2HIV and AIDS Malignancy
1

Branch, Center for Cancer Research, National Cancer Institute, Bethesda, MD

Immunodeficiency, whether acquired in the case of human immunodeficiency virus (HIV) infection or congenital due to
inborn errors of immunity (IEIs), presents clinically with not only infection and immune dysregulation but also increased
risk of malignancy. The range of malignancies seen is relatively limited and attributable to the particular cellular and
molecular defects in each disease. CD4+ T-cell lymphopenia in people living with HIV infection (PLWH) and certain IEIs
drive the predisposition to aggressive B-cell non-Hodgkin lymphomas, including certain rare subtypes rarely seen in
immunocompetent individuals. PLWH and IEI that lead to profound T-cell lymphopenia or dysfunction also are at risk
of cancers related to oncogenic viruses such as Kaposi sarcoma herpesvirus, Epstein-Barr virus, human papillomavirus
(HPV), and Merkel cell polyomavirus. IEIs that affect natural killer cell development and/or function heavily predispose
to HPV-associated epithelial cancers. Defects in DNA repair pathways compromise T- and B-lymphocyte development
during immune receptor rearrangement in addition to affecting hematopoietic and epithelial DNA damage responses,
resulting in both hematologic and nonhematologic cancers. Treatment of cancers in immunodeficient individuals should
be curative in intent and pursued in close consultation with disease experts in immunology and infectious disease.

LEARNING OBJECTIVES
• Understand the immunologic defects in PLWH infection and their relationship to the development of cancer
• Recognize that IEIs affecting lymphocytes and/or DNA repair predispose to cancer

and positive for CD138, CD38, and CD45, as well as both

CLINICAL CASE 1, PART 1
A 28­year­old cisgender man who immigrated to the
United States from Uganda 10 years ago presents to
the emergency room with shortness of breath. He also
reports 6 months of weight loss, fatigue, and fevers. On
examination, the patient is in respiratory distress and has
raised, violaceous plaques on his trunk and extremities.
Chest x­ray reveals a large left­sided pleural effusion.
The patient undergoes testing for human immunodefi­
ciency virus (HIV), which is positive. His HIV viral load is
150 000 copies/mL, with a CD4+ T­cell count of 110 cells/µL.
After consultation with an HIV specialist and a pharma­
cist, the patient is started on combination antiretroviral
therapy (ART) including bictegravir, emtricitabine, and
tenofovir alafenamide along with sulfamethoxazole/
trimethoprim and valacyclovir for opportunistic infection
prophylaxis. Biopsy specimen of one of the skin lesions is
consistent with Kaposi sarcoma (KS). Preliminary results
from a therapeutic and diagnostic thoracentesis indicate
a B­cell lymphoma. Further pathologic workup indicates
the malignant B cells are negative for CD20 and CD19
Prakash Singh Shekhawat
Kaposi sarcoma herpesvirus (KSHV) and Epstein­Barr virus
(EBV), indicating a diagnosis of primary effusion lym­
phoma (PEL).
Cancers arising in people with HIV/AIDS
Cancer and HIV/AIDS have been linked since the first cluster
of cases of KS and pneumocystis pneumonia were reported
in 1981. It was soon recognized that people living with HIV
(PLWH) were susceptible to other cancers, particularly
aggressive B­cell non­Hodgkin lymphomas (NHLs) (Table 1).1
With the discovery of KSHV (also called human herpesvi­
rus 8) in 1994 and its subsequent finding as the causative
agent of KS, it became apparent that many, but not all, HIV­
associated cancers are caused by oncogenic viruses.2
HIV infects CD4+ T cells and certain other cell types, and
the primary immune defect is CD4+ T­cell lymphopenia. HIV
infection also causes other disruptions of immune function,
including depletion of other lymphocyte populations, defects
in lymphocyte function, increased inflammatory cytokines,
and chronic inflammation. These immunologic defects impair
How immunodeficiency can lead to malignancy | 287
Table 1. Etiology and inci­dence of can­cers among PLWH

Role of SIR in the

Etiologic virus Cancer type immu­no­sup­pres­sion Other risk fac­tors United States3,4 Comment
KSHV Kaposi sar­coma +++ Men who have sex with men 498 Very com­mon in parts of
sub-Saharan Africa
EBV Non-Hodgkin lym­phoma* ++ to ++++ HBV, HCV 12
Hodgkin lym­phoma ++ Male sex 7.7 Mixed cel­lu­lar­ity is most
com­mon sub­type
HPV Anal can­cer + Tobacco, anal warts 19 Incidence increas­ing
Cervical can­cer + Tobacco 3.2 Very com­mon in resource-
lim­ited regions
Oropharyngeal can­cer + Tobacco, alco­hol 1.6-2.2† Incidence increas­ing

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/287/1852069/287pai.pdf by guest on 13 December 2021

Vulvar can­cer + Tobacco 9.4
Penile can­cer + Tobacco, lack of cir­cum­ci­sion 5.3
HBV, HCV Hepatocellular can­cer + Alcohol, cir­rho­sis 3.2
MCPyV Merkel cell car­ci­noma + Sun expo­sure 2.6 At least 15% of cases not
asso­ci­ated with MCPyV
None or Lung can­cer + Tobacco, pul­mo­nary 2 Most com­mon can­cer-related
unknown infec­tions cause of death
Nonmelanoma skin can­cer + Sun expo­sure, fam­ily his­tory 2.1
Conjunctival car­ci­noma + Sun expo­sure 5.5 Rare in the United States;
occurs most often in
sub-Saharan Africa; viral
eti­­ol­ogy suspected
*See Figure 2.
†
SIR is 1.6 for HPV-related can­cers and 2.2 for HPV-unre­lated can­cers.
SIR, stan­dard­ized inci­dence ratio; +, low; ++, moderate; +++, significant; ++++, very significant role for immunosuppression in pathogenesis.

con­trol of onco­genic viruses, such as KSHV, EBV, human pap­il­lo­ma­
vi­rus (HPV), and Merkel cell polyomavirus (MCPyV), lead­ing to the
devel­op­ment of can­cer (Table 1, Figure 1). KSHV is an essen­tial cause
of KS, and pop­u­la­tions with a higher prev­al­ence of KSHV infec­tion
expe­ri­ence higher rates of KS, such as men who have sex with men
and those liv­ing in cer­tain areas of sub-Saharan Africa. PLWH also
have a higher expo­sure to high-risk types of HPV, increas­ing the
risk for cer­vi­cal and anal can­cer. Increased rates of tobacco use and
infec­tion with hep­a­ti­tis B virus (HBV) and hep­a­ti­tis C virus (HCV)
also con­trib­ute to higher can­cer rates in PLWH (Table 1).
The devel­op­ment of an effec­tive 3-drug ART in the mid-1990s
allowed for mean­ing­ful res­to­ra­tion of CD4+ T-cell counts, and
AIDS was converted from a death sen­tence to a chronic dis­ease.
ART sub­stan­tially reduced the inci­dence of the can­cers asso­ci­
ated with pro­found immu­no­de­fi­ciency such as KS and cer­tain
B-cell lym­pho­mas. However, as PLWH live lon­ger, they have an
increased cumu­la­tive like­li­hood of devel­op­ing can­cers asso­ci­
ated with mod­er­ate decreases in CD4+ T-cell counts (Table 1).
PLWH also remain at risk of can­cers not asso­ci­ated with immu­
no­de­fi­ciency. In fact, can­cer is now one of the most com­mon or
the most com­mon cause of death in PLWH in regions where ART
is read­ily avail­­able. Interestingly, PLWH do not have an increased
inci­dence of cer­tain com­mon can­cers, such as breast, colon, and
pros­tate can­cer; how­ever, they may pres­ent at more advanced
stages or with more aggres­sive sub­types.5,6
NHL is the most com­mon can­cer among PLWH in the United
States.3 NHL is asso­ci­ated with mod­er­ate to severe immu­no­sup­
pres­sion depending on the sub­type (Figure 2). Hodgkin lym­
phoma is also increased but gen­er­ally occurs at pre­served CD4+
T-cell counts and is asso­ci­ated with use of ART.7 Diffuse large
B-cell (DLBCL) is the most com­mon NHL sub­type in PLWH, fol­
lowed by Bur­kitt lym­phoma and rare sub­types occur­ring almost
exclu­sively in PLWH, par­ tic­
u­
larly PEL and plasmablastic lym­
phoma. The inci­dence of pri­mary cen­tral ner­vous sys­tem lym­
phoma, which is asso­ci­ated with pro­found immu­no­sup­pres­sion,
dras­ti­cally declined with the advent of ART and remains rare.
Lymphoma often pres­ents at advanced stages in PLWH with
more rapid pro­ gres­sion, fre­quent B symp­ toms, and fre­ quent
involve­ ment of extranodal sites and the cen­ tral ner­
vous sys­
tem than the gen­eral pop­u­la­tion.8 The depth of the CD4+ nadir,
uncon­trolled HIV vire­mia, and inter­rup­tions in ART are all­asso­ci­
ated with increased risk of NHL.9-11 A rel­a­tively large per­cent­age
of HIV-asso­ci­ated lym­pho­mas is asso­ci­ated with EBV, includ­ing
almost 100% of pri­mary cen­tral ner­vous sys­tem lym­pho­mas.12
KSHV is the caus­a­tive agent of PEL, but most are also coinfected
with EBV. Chronic B-cell acti­va­tion in PLWH con­trib­utes to lym­
phomagenesis, and coinfection with HBV and/or HCV may also
increase the risk of NHL through this mech­a­nism.13
As exem­pli­fied by case 1, patients with malig­nan­cies that are
more fre­quent in HIV/AIDS should be screened for HIV infec­tion,
espe­cially if they are in a high-risk group. Before the devel­op­ment
of effec­tive ART, treat­ment with cura­tive-intent che­mo­ther­apy for
HIV-asso­ci­ated malig­nan­cies was nearly impos­si­ble given the pro­
found immu­no­sup­pres­sion and risk of infec­tious ­com­pli­ca­tions.

288  |  Hematology 2021  |  ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/287/1852069/287pai.pdf by guest on 13 December 2021

Figure 1. Key relationships between immune defects and oncogenesis. General relationships are shown in A, and key relationships
specific to the indicated diseases are shown in B-H. Yellow boxes denote key immunologic defects observed in HIV/AIDS and/or IEI;
green boxes denote secondary defects; orange boxes denote infectious agents that contribute to malignancies in ­immunodeficiency;
purple ovals show the principal malignancies associated with the immune defects. DC, dendritic cell; HHV-8, human herpesvirus 8.

Prakash Singh Shekhawat

How immu­no­de­fi­ciency can lead to malig­nancy  |  289

Figure 2. Pathologic, immune, and viral characteristics of HIV-associated NHL. ABC, activated B cell; GCB, germinal center B cell;
IRF4, interferon regulatory factor 4; MUM1, multiple myeloma 1; PBL, plasmablastic lymphoma. *Indicates MYC gene rearrangements.
This figure was created on biorender​.com.
With ART, it has been shown that most can­cers aris­ing in PLWH
are best treated with the same reg­ i­
mens used to treat HIV-
neg­a­tive patients.14,15 The National Comprehensive Cancer Net­
work has published prac­tice guide­lines for can­cer in peo­ple with
HIV, HIV-related NHL, and KS. PLWH should con­tinue ART while
under­go­ing can­cer ther­apy, and oncol­o­gists should work closely
with HIV spe­cial­ists to ensure opti­mal HIV man­age­ment and infec­
tion pro­phy­laxis. Oncologists should con­sult with phar­ma­cists to
avoid drug-drug inter­ac­tions, espe­cially if patients are tak­ing HIV
pro­te­ase inhib­i­tors, which are metab­o­lized by the cyto­chrome
P450 enzymes. One of the most excit­ing devel­op­ments in can­
cer ther­apy is the suc­cess of immune check­point inhib­it­ors tar­
geting PD-1, PD-L1, or CTLA-4. Despite ini­tial tox­ic­ity and effi­cacy
con­cerns, recent stud­ies have shown them to be safe and effec­
tive in expected can­cer types in PLWH with more than 100 CD4+
T cells/µL.16 It should be stressed that HIV infec­tion is a chronic,
man­ age­
able con­ di­
tion, and US Food and Drug Administration
guid­ance endorses inclu­sion of PLWH with 350 or more CD4+
cells/µL for all­can­cer clin­i­cal tri­als as well as their inclu­sion at
lower CD4+ T cell counts if patients have poten­tially cur­able malig­
nan­cies or if the study agents have been shown to have activ­ity
in a given can­cer.17 PLWH treated for can­cer are at increased risk
for sec­ond­ary pri­mary can­cers, fur­ther underscoring the need for
vig­i­lance and adher­ence to can­cer screen­ing guide­lines.18
290  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/287/1852069/287pai.pdf by guest on 13 December 2021
CLINICAL CASE 1, PART 2
For the patient’s PEL, an oncol­ o­
gist pre­scribes cura­ tive-intent
treat­ment with dose-adjusted etoposide, pred­ni­sone, vin­cris­tine,
cyclo­phos­pha­mide, and doxo­ru­bi­cin (EPOCH) in addi­tion to con­
tinu­ing ART. After 6 cycles of EPOCH, the patient’s lym­phoma is in
com­plete remis­sion, his CD4+ T cell count is 90 cells/µL, and his
HIV viral load is unde­tect­able. Two months later, the patient devel­
ops increas­ing num­bers of nod­u­lar KS lesions on his lower extrem­
i­ties. His oncol­o­gist pre­scribes lipo­so­mal doxo­ru­bi­cin to treat the
KS while the patient’s immune sys­ tem recon­ sti­
tutes fol­low­ing
EPOCH. After 4 cycles, there is sig­nif­i­cant flat­ten­ing and light­en­ing
of the KS lesions, and lipo­so­mal doxo­ru­bi­cin is discontinued. One
year fol­low­ing EPOCH treat­ment, the patient’s PEL is in remis­sion,
the cuta­ne­ous KS lesions are barely vis­i­ble, CD4+ T-cell count is
350 cells/µL, and HIV viral load remains unde­tect­able.
CLINICAL CASE 2
A 25-year-old man with a his­tory of recur­rent respi­ra­tory tract
infec­tions was referred to an immu­nol­o­gist for eval­u­a­tion. He
was found to have dysgammaglobulinemia, poor anti­ gen-
spe­cific responses, hyper-IgM, and IgA defi­ciency, con­sis­tent
with a diag­no­sis of com­mon var­i­able immu­no­de­fi­ciency. Sus­
pecting an inborn error of immu­ nity (IEI), the immu­ nol­o­gist
ordered whole-exome sequenc­ing. While awaiting fur­ther test­
ing, the patient presented to a local emer­gency room with
cer­vi­cal lymph­ade­nop­a­thy and a medi­as­ti­nal mass and was
treated with emer­gent cor­ti­co­ste­roids. The biopsy spec­i­men
was ini­tially read as Hodgkin lym­phoma, but his immu­nol­o­gist
questioned the diag­ no­sis and requested EBV test­ ing of the
tumor. Rereview of pathol­ogy found that what were thought
to be Reed-Sternberg cells were large atyp­i­cal cells that had
com­pletely effaced the nodal archi­tec­ture and were pos­i­tive
for EBV, CD20, PAX5, and CD30, con­sis­tent with EBV+ DLBCL
­lym­phoma. Whole-exome sequenc­ing showed a het­ero­zy­gous
E1021K var­i­ant in the PIK3CD gene. After fur­ ther lym­ phoma
treat­ment, he under­went suc­cess­ful allo­ge­neic trans­plan­ta­tion
with a matched unre­lated donor.
Cancers aris­ing in peo­ple with IEIs
Advances in genet­ics have led to a ver­i­ta­ble explo­sion of defined
pri­mary immu­no­de­fi­ciency dis­eases, now termed IEIs. The Inter­
national Union of Immunological Societies (IUIS) cur­rently lists
more than 450 dis­eases, cat­e­go­rized broadly into 10 phe­no­typic
groups (Table 2).19-21 The speed with which these dis­or­ders are
being described is evi­dent by the num­ber of dis­or­ders in each
bien­nial pub­li­ca­tion (191 in 2011, 430 in 2019) and the fact that the
2019 meet­ing report, published in Jan­u­ary 2020, was suf­fi­ciently
out­dated that a sup­ple­ment adding 26 addi­tional mono­genic
gene defects was published in Jan­u­ary 2021.
At least 10% of IEIs are stated within the IUIS reports to be
asso­ci­ated with malig­nan­cies. The dis­tri­bu­tion of IEIs asso­ci­ated
with malig­nancy is not even with respect to cat­e­gory; very few
fall into the categories of defects in intrin­sic and innate immu­
nity, com­ple­ment defects, and autoinflammatory dis­or­ders.22,23
As reviewed below, dis­tinct and overlapping mech­a­nisms are
respon­si­ble for a pre­dis­po­si­tion to malig­nancy among patients
with IEIs (Figure 1).
General path­o­genic mech­a­nisms
As in HIV/AIDS, T-cell lymphopenia and dys­func­tion gen­er­ally
pre­dis­pose to both poor tumor sur­veil­lance and oppor­tu­nis­tic
infec­tion with onco­genic viruses. T-cell dys­func­tion may lead to
dysregulated pro­lif­er­a­tion of B cells, which may be fur­ther exac­
er­bated by con­di­tions that lead to B-cell lymphopenia or intrin­
si­cally aber­rant B-cell sig­nal­ing.
Natural killer (NK) cells are crit­ i­
cal for con­
trol of cer­tain
viruses, par­tic­u­larly DNA viruses such as HPV and EBV. Control
of viral infec­tions in skin or tis­sue is often depen­dent on anti­
gen presenting cells such as mono­ cytes and den­ dritic cells,
which must undergo acti­va­tion, changes in shape, and migra­tion
through tis­sues to inter­act with T cells and traf­fic to sec­ond­ary
lym­phoid organs.
Mutations in a host of genes involved in DNA repair pre­dis­
pose to lym­phoid malig­nancy and epi­the­lial malig­nancy, due
to the gen­eral need for faith­ful DNA repair in all­tis­sues. Repair
of dou­ ble-stranded breaks gen­ er­
ated in the T-cell recep­ tor
and immu­no­glob­u­lin loci by RAG1/RAG2 depends on the non­­
ho­mol­o­gous end-join­ing path­way. Thus, defects in genes such
as LIG4, DCLRE1C, XLF1, and NBS1 tend to result in T- and B-cell
Prakash Singh Shekhawat
lymphopenia and fur­ther­more increase the chance of trans­lo­
ca­tions asso­ci­ated with lym­phoma. Outside of lym­pho­cytes,
muta­tions in these genes and oth­ers involved in homol­o­gous
recom­bi­na­tion and the response to DNA dam­age (eg, BLM, the
gene mutated in Bloom syn­drome) ren­der the patients radi­a­tion
sen­si­tive. Oncogenesis is com­mon par­tic­u­larly in rap­idly divid­ing
tis­sues, includ­ing skin, gas­tro­in­tes­ti­nal tract, breast, gen­i­to­uri­
nary tract, soft tis­sues, and bone mar­row.
Defects of hema­to­poi­etic stem cell (HSC) fit­ness, dif­fer­en­ti­
a­tion, and/or sur­vival con­trib­ute to bone mar­row fail­ure, in the
set­ting of rep­li­ca­tive stress and clonal hema­to­poi­e­sis of a lim­
ited pool of HSCs. Accumulation of addi­tional muta­tions pro­
motes the devel­op­ment of myelodysplasia and ulti­mately acute
mye­loid leu­ke­mia. In these dis­or­ders (eg, Shwachman-Diamond
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/287/1852069/287pai.pdf by guest on 13 December 2021
syn­drome, dyskeratosis congenita, or severe con­gen­i­tal neutro­
penia), the immu­no­de­fi­ciency is typ­i­cally related to neutropenia
rather than defects of lym­pho­cytes or adap­tive immu­nity.
Individual dis­or­ders with more than 1 mech­a­nism involved
Wiskott-Aldrich syn­drome and ded­i­ca­tor of cyto­ki­ne­sis 8
defi­ciency.  Wiskott-Aldrich syn­drome (WAS) is an X-linked dis­or­
der caused by muta­tions in the WAS gene, resulting in a syn­drome
of microthrombocytopenia, eczema, infec­tions, auto­im­mu­nity,
and pre­dis­po­si­tion to lym­phoma. WAS pro­tein is crit­i­cal for actin
cyto­skel­e­tal rearrangement, which in turn is required for for­ma­
tion of the immune syn­apse in response to T, B, and NK cell-cell
sig­nal­ing, for reg­u­la­tory T-cell func­tion, mono­cyte cyto­kine pro­
duc­tion, and lym­pho­cyte and den­dritic cell shape and mobil­ity.24
Progressive CD8+ T-cell lymphopenia, impaired T cell receptor
sig­nal­ing and acti­va­tion, poor NK cell func­tion, over­ac­tive B cell
receptor sig­nal­ing, and dysregulated mono­cyte acti­va­tion all­
con­ spire to pro­ mote lymphomagenesis with and with­ out EBV
infec­tion (Figure 1).
Dedicator of cyto­ki­ne­sis 8 (DOCK8) defi­ciency is an auto­so­
mal reces­sive dis­or­der that shares fea­tures with WAS, includ­ing
eczema, infec­tions, and auto­im­mu­nity, but also is char­ac­ter­ized
by her­petic viral skin infec­tions and HPV-driven malig­nancy.25
DOCK8 is also crit­i­cal for actin cyto­skel­e­tal func­tion and func­
tions in a com­ plex with WAS pro­ tein. Immune syn­apses are
poorly formed in DOCK8-defi­cient cells, sim­i­lar to WAS. Several
dis­tinct cel­lu­lar defects in DOCK8-defi­cient patients that may
under­lie HPV-driven can­cers include severe impair­ment of den­
dritic cell motil­ity in skin, decreased num­bers of plasmacytoid
den­dritic cells that fail to pro­duce inter­feron α, and increased
death of immune cells, includ­ing skin-res­i­dent CD8+ T cells, due
to frag­men­ta­tion while attempting to tra­verse through tis­sues
(Figure 1).26
Common var­i­able immu­no­de­fi­ciency.  Patients with com­mon
var­i­able immu­no­de­fi­ciency (CVID) have defi­ciency of anti­body
pro­duc­tion and humoral immu­nity, with vary­ing degrees of T-
cell dys­func­tion and immune dysregulation (Figure 1). Because
most patients with CVID are not genet­i­cally char­ac­ter­ized, there
is broad het­ero­ge­ne­ity in dis­ease man­i­fes­ta­tions. Lymphomas
and lymphoproliferation are com­mon, occur­ring in 17.2% of 1091
patients with CVID reg­is­tered in the US Immunodeficiency Net­
work.27 B-cell lymphopenia is com­mon, and the addi­tional rep­
li­ca­tive stress on B cells in the con­text of chronic inflam­ma­tion
is a likely con­trib­u­tor. In con­trast to the dis­tri­bu­tion of B-cell
lym­pho­mas in oth­er­wise healthy adults, there are few fol­lic­u­lar
How immu­no­de­fi­ciency can lead to malig­nancy  |  291
Table 2. IEIs asso­ci­ated with malig­nan­cies

Associated malig­nan­cies
Myelodysplastic Gastric ade­
B- or T-cell lym­pho­mas HPV-driven syn­drome and/or no­car­ci­noma
Number IUIS cat­e­gory Examples (ref­er­ence) can­cers mye­loid leu­ke­mia or lym­phoma Other
1 Immunodeficiencies SCID SCID (ADA, RAG1, LIG4, — — — OX40 defi­ciency
affect­ing DCLRE1C, coronin 1a) (KS)
cel­lu­lar and ZAP70 CD40L
humoral immu­nity (cholangio
carcinoma)
2 Combined WAS WAS — — Ataxia-tel­an­gi­ Ataxia-tel­an­gi­ec­
immunodeficiencies Ataxia- DOCK8 defi­ciency ec­ta­sia ta­sia (breast,
with asso­ci­ated or tel­an­gi­ec­ CARD11 auto­so­mal ­ thy­roid, liver,
syndromic fea­tures ta­sia dom­i­nant loss of brain, acute

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/287/1852069/287pai.pdf by guest on 13 December 2021

func­tion lym­pho­blas­tic
Ataxia-tel­an­gi­ec­ta­sia leu­ke­mia)
Nijmegen break­age
syn­drome
Bloom syn­drome
Immunodeficiency
with cen­tro­meric
insta­bil­ity and facial
anom­a­lies
Schimke immuno-
osse­ous dys­pla­sia
Cartilage hair
hypo­pla­sia
DiGeorge syn­drome
CHARGE syn­drome
PNP defi­ciency
3 Predominantly CVID APDS — — CVID —
anti­body CVID Selective IgA
deficiencies Selective IgA defi­ciency
defi­ciency
4 Diseases of immune ALPS ALPS — — — —
dysregulation IPEX X-linked
HLH lymphoproliferative
dis­ease type 1
XMEN
ITK defi­ciency
CD27/CD70 defi­ciency
CTPS1 defi­ciency
IL10R defi­ciency
5 Congenital defects Chronic gran­ — GATA2 GATA2 — —
of phago­cyte u­lo­ma­tous defi­ciency defi­ciency
num­ber or dis­ease Severe
func­tion Severe con­gen­i­tal
con­gen­i­tal neutropenia
neutropenia (ELANE, HAX1,
Leukocyte other)
adhe­sion Shwachman-
defi­ciency Diamond
syn­drome
X-linked
neutropenia
due to gain of
func­tion muta­
tions in WAS
SMARCAD2
6 Defects in intrin­sic IL-12 and WHIM syn­drome Epidermo­­ — — —
and innate inter­feron dysplasia
immu­nity γ path­way verruciformis
defects WHIM syn­drome

292  |  Hematology 2021  |  ASH Education Program

Table 2. (continued)

Associated malig­nan­cies
Myelodysplastic Gastric ade­
B- or T-cell lym­pho­mas HPV-driven syn­drome and/or no­car­ci­noma
Number IUIS cat­e­gory Examples (ref­er­ence) can­cers mye­loid leu­ke­mia or lym­phoma Other
7 Autoinflammatory Familial — — — — —
dis­or­ders Med­i­ter­ra­
nean fever
Aicardi-
Goutieres
syn­drome
8 Complement Hereditary — — — — —
deficiencies angioedema

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/287/1852069/287pai.pdf by guest on 13 December 2021

9 Bone mar­row fail­ure Fanconi Dyskeratosis congenita — Fanconi ane­mia — Fanconi ane­mia
ane­mia Dyskeratosis (breast, skin,
Dyskeratosis congenita head and
congenita Coats plus neck, other)
syn­drome
MIRAGE (SAMD9)
Ataxia
pan­cy­to­pe­
nia syn­drome
(SAMD9L)
10 Phenocopies of IEI Somatic — — — — —
muta­tions
lead­ing to
ALPS
RAS-
asso­ci­ated
auto­im­mune
leukopro-
liferative
dis­ease
CHARGE, coloboma, heart anomaly, choanal atresia, retardation, genital and ear anomalies; HLH, hemophagocytic lymphohistiocytosis; IPEX,
immune dysregulation, polyendocrinopathy, enteropathy, X-linked; MIRAGE, myelodysplasia, infection, restriction of growth, adrenal hypoplasia,
genital phenotypes, and enteropathy; PNP, purine nucleoside phosphorylase; RAS, rat sarcoma; SCID, severe combined immunodeficiency; WHIM,
warts, hypogammaglobulinemia, infections, and myelokathexis.

lym­ pho­ mas reported, skewing instead to extranodal mar­ ginal
zone lym­ phoma and mucosa-asso­ ci­
ated lym­ phoid tis­ sue lym­
phoma.28 Gastric can­cers occur in patients with CVID more fre­
quently than the gen­eral pop­u­la­tion and appear to be related to
chronic inflam­ma­tion and cell pro­lif­er­a­tion, pos­si­bly in con­junc­
tion with Helicobacter pylori infec­tion. In an Ital­ian reg­is­try of 455
patients with CVID, 25 patients had gas­tric can­cer, with a stan­
dard­ized inci­dence ratio of 6.4, and this was the highest cause
of death. The gas­tric can­cers in the Ital­ian series arose out of a
back­ground of atro­phic gas­tri­tis, intes­ti­nal meta­pla­sia, or dys­pla­
sia.29 Lymphoid and/or gran­u­lo­ma­tous infil­tra­tion of the gas­tro­
in­tes­ti­nal tract or lungs may fur­ther trig­ger excess immune cell
pro­lif­er­a­tion. Although the inci­dence of malig­nant or pro­lif­er­a­tive
dis­or­ders is clearly increased, rou­tine sur­veil­lance is unlikely to be
effec­tive for early detec­tion. Meticulous sup­port­ive care, main­
taining a high degree of sus­pi­cion for malig­nancy, and aggres­sive
workup of symp­toms are key to man­ag­ing patients with CVID.
Activated phosphoinositide 3-kinase-δ syn­drome. A sub­set of
patients pre­vi­ously diag­nosed with CVID have been shown to
har­
bor gain-of-func­ tion muta­ tions in the genes encoding the
cat­a­lytic sub­unit (p110δ) or biallelic loss-of-func­tion muta­tions in
the reg­u­la­tory sub­unit (p85α) of the phosphoinositide 3-kinase-δ
Prakash Singh Shekhawat
(PI3K-δ) pro­tein. These muta­tions result in increased PI3K sig­nal­
ing in lym­pho­cytes, where PI3K-δ is expressed.30 The fact that
muta­ tions in p110δ respon­ si­
ble for acti­vated PI3K-δ syn­drome
(APDS) are pres­ent as somatic muta­tions in DLBCL, man­tle cell
lym­phoma, and rarely Bur­kitt lym­phoma sup­ports a B-cell intrin­
sic role of over­ac­tive PI3K sig­nal­ing in lymphomagenesis due to
uncon­trolled B-cell pro­lif­er­a­tion in the con­text of class switch
recom­bi­na­tion in ger­mi­nal cen­ters. These patients may have lym­
phoma, as in case 2, even before IEI is fully rec­og­nized. In addi­
tion, T cells in patients with APDS have impaired func­tion and
cyto­tox­ic­ity, have exhausted CD8+ CD57+ T cells, and com­monly
develop T-cell lymphopenia and low naive T-cell counts over time
(Figure 1).31
Ataxia-tel­an­gi­ec­ta­sia.  Biallelic muta­tions in the auto­so­mal gene
ATM result in the syn­drome of ataxia-tel­an­gi­ec­ta­sia, char­ac­ter­ized
by immu­no­de­fi­ciency, pro­gres­sive cer­e­bel­lar ataxia, oculocuta­
neous tel­an­gi­ec­ta­sia, and can­cer pre­dis­po­si­tion.32 ATM is a kinase
that has at least 2 crit­i­cal roles in DNA repair. ATM is acti­vated dur­
ing dou­ble-stranded breaks gen­er­a­tion (includ­ing dur­ing variable-
diversity-joining region recom­bi­na­tion) and recruits the MRE11/
RAD51/NBS1 com­plex in a feed-for­ward cycle by phos­phor­
y­lat­ing his­tone H2AX. ATM-depen­dent phos­phor­y­la­tion of other
How immu­no­de­fi­ciency can lead to malig­nancy  |  293
tar­gets medi­ates cell cycle arrest until DNA dam­age can be re­
solved. The broad func­ tion of ATM in DNA repair, within and
out­side of lym­pho­cytes, results in mul­ti­ple dif­fer­ent can­cers,
includ­ing acute leu­ke­mia (both lym­phoid and mye­loid), ­Hodgkin
lym­phoma, NHL, and car­ci­no­mas (Figure 1).33 The degree of B-cell
lymphopenia and anti­body defi­ciency cor­re­lates with can­cer risk.
GATA2 defi­ciency.  Germline het­ero­zy­gous muta­tions in the
GATA2 gene, a tran­scrip­tion fac­tor crit­ic ­ al for HSC devel­op­ment
and homeo­sta­sis, result in a wide vari­ety of clin­i­cal man­i­fes­ta­
tions, includ­ing atyp­i­cal myco­bac­te­rial infec­tion, DNA viral infec­
tions (herpes simplex virus, HPV, EBV, cyto­meg­al­o­vi­rus), lymph­
edema, bone mar­ row fail­ure, myelodysplastic syn­ drome, and
acute mye­loid leu­ke­mia.34 GATA defi­ciency results in mul­ti­ple cel­
lu­lar defects, includ­ing numer­ic ­ al defi­ciency of mono­cytes, den­
dritic cells, NK cells, and B cells; pro­gres­sive stem cell loss; and
accu­mu­la­tion of addi­tional cyto­ge­netic abnor­mal­i­ties such as
mono­somy 7 and tri­somy 8. Whether the loss of mature immune
cells is a man­i­fes­ta­tion of HSC loss vs cell intrin­sic defects in sur­
vival or homeo­sta­sis is not clear. Here the immune defect per se,
par­tic­u­larly the lack of NK cells, appears crit­i­cal for the devel­op­
ment of HPV-driven can­cer.35 EBV+ lym­phoma is far less com­mon
in GATA2 defi­ciency than EBV vire­mia and is related to immune
defects includ­ing excess pro­lif­er­a­tion of B cells, poor den­dritic
cell num­ber and func­tion, and thereby poor T-cell response to
EBV.36 The pre­dis­po­si­tion to mye­loid leu­ke­mia in con­trast is likely
due to the par­al­lel defect in hema­to­poi­e­sis (Figure 1).
Principles of treat­ment of can­cer in patients with IEIs.  In gen­
eral, patients with IEIs and can­ cer should undergo stan­ dard
treat­ment, with the caveat that radi­a­tion-sen­si­tive patients (eg,
ataxia-tel­an­gi­ec­ta­sia) require dose mod­i­fi­ca­tion or omis­sion alto­
gether of alkylating agents and radi­a­tion. Immunosuppressive
agents such as cor­ti­co­ste­roids, mono­clo­nal antibodies that tar­
get B cells, fludarabine, and targeted inhib­i­tors of JAK/STAT or
BTK path­ways fur­ther pre­dis­pose to oppor­tu­nis­tic infec­tions.
Additional anti­bi­otic pro­phy­laxis and sched­uled replace­ment of
immu­no­glob­u­lins to main­tain trough lev­els are crit­i­cal com­po­
nents of sup­port­ive care.23
Correction of the under­ly­ing predisposing con­di­tion with allo­
ge­neic HSC trans­plan­ta­tion con­sol­i­dates can­cer treat­ment and
should be con­sid­ered when fea­si­ble; it is the only ave­nue for long-
term pre­ven­tion of recur­rence and defin­i­tive cure. The rapid rise in
dis­cov­ery of the genes respon­si­ble for IEIs has been paralleled by
the devel­op­ment of gene ther­apy approaches, using inte­grat­ing
viral vec­tors to trans­duce autol­o­gous HSCs that are transplanted
after appro­pri­ate con­di­tion­ing.37 Gene ther­apy is an excit­ing alter­
na­tive to allo­ge­neic HSC trans­plant that obvi­ates the need for a
matched donor and avoids immu­no­logic com­pli­ca­tions such as
graft rejec­tion and graft-ver­sus-host dis­ease alto­gether. Newer
tech­nol­o­gies based on gene editing also have prom­ise.
Acknowledgments  1869.
This work was supported by funding from the Intramural Re­
search Program, National Institutes of Health, National Cancer
Institute, Center for Cancer Research. The authors acknowl­edge
the con­tri­bu­tion of Dr Ramya Ramaswami, who cre­ated Figure 2,
and Dr Luigi Notarangelo, who pro­vided case 2.
294  |  Hematology 2021  |  ASH Education Program
Conflict-of-inter­est dis­clo­sure
Sung-Yun Pai has no con­flicts of inter­est to dis­close and is sup­
ported by Center for Cancer Research, National Cancer Institute.
Sung-Yun Pai’s spouse is coin­ven­tor of a pat­ent on com­bi­na­tion
ther­ap
­ ies, includ­ing abiraterone to treat pros­tate can­cer.
Kathryn Lurain receives research funding from Bristol Myers
Squibb-Celgene Corporation, Merck, EMD Serono, Janssen Re­
search, Lentigen Corp., CTI Biopharma and the National Cancer
Institute.
Robert Yarchoan’s research is funded in part by Cooperative
Research and Development Agreements (CRADAs) bet­ ween
Celgene Corporation (now Bristol Myers Squibb, Co.) and the
National Cancer Institute. He has also used drugs for his clin­i­cal
and/or lab­o­ra­tory research pro­vided to the NCI by Genentech
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/287/1852069/287pai.pdf by guest on 13 December 2021
Corp., Merck and Co., EMD Serono, Janssen Research, Lentigen
Corp., and CTI Biopharma. Robert Yarchoan is a coin­ven­tor on
pat­ents on the treat­ment of KSHV-induced lym­phoma using
immu­no­mod­u­la­tory com­pounds and uses of bio­mark­ers, on
pep­tide vac­cines for HIV and other viral dis­eases, and on inter­
leu­kin (IL)–12 as a treat­ment for Kaposi sar­coma. His spouse,
who is also a US gov­ern­ment employee, has pat­ents on KSHV
viral IL-6, ana­lyz­ing sin­gle-cell epigenomes, induc­ing inter­nal­i­
za­tion of sur­face recep­tors, and vasostatin. It is his under­stand­
ing that for­eign pat­ents have also been or will be filed for these
inven­tions. These inven­tions were all­made as full-time employ­
ees of the US gov­ern­ment under 45 Code of Federal Regula­
tions Part 7. All rights, title, and inter­est to these pat­ents have
been or should by law be assigned to the US Department of
Health and Human Services. The gov­ern­ment con­veys a por­tion
of the roy­al­ties it receives to its employee-inven­tors under the
Federal Technology Transfer Act of 1986 (P.L. 99-502).
Off-label drug use
Sung-Yun Pai: Because most IEI present in childhood and very
few drugs are formally approved for use in children, off-label
drug use is discussed.
Kathryn Lurain: Off-label use of EPOCH for HIV-associated lym­
phomas and primary effusion lymphoma was discussed.
Robert Yarchoan: We broadly discuss the use of checkpoint in­
hibitors in PLWH but do not specify tumors for which they are
FDA approved.
Correspondence
Sung-Yun Pai, National Institutes of Health, Building 10, Room
1-5142, 10 Center Dr, Bethesda, MD 20892; e-mail: sung-yun​
­.pai@nih​­.gov.
References
1. Yarchoan R, Uldrick TS. HIV-asso­ ci­
ated can­ cers and related dis­ eases.
N Engl J Med. 2018;378(22):2145.
2. Chang Y, Cesarman E, Pessin MS, et al. Identification of her­pes­vi­rus-like DNA
sequences in AIDS-asso­ci­ated Kaposi’s sar­coma. Science. 1994;266(5192):1865-
3. Hernández-Ramírez RU, Shiels MS, Dubrow R, Engels EA. Cancer risk in
HIV-infected peo­ple in the USA from 1996 to 2012: a pop­u­la­tion-based,
­reg­is­try-link­age study. Lancet HIV. 2017;4(11):e495-e504.
4. Silverberg MJ, Leyden W, Warton EM, Quesenberry CP Jr, Engels EA, Asgari
MM. HIV infec­tion sta­tus, immu­no­de­fic ­ iency, and the inci­dence of non-­
mel­a­noma skin can­cer. J Natl Cancer Inst. 2013;105(5):350-360.
5. Coghill AE, Engels EA, Schymura MJ, Mahale P, Shiels MS. Risk of breast,
pros­tate, and colo­rec­tal can­cer diag­noses among HIV-infected indi­vid­u­als
in the United States. J Natl Cancer Inst. 2018;110(9):959-966.
6. Brandão M, Bruzzone M, Franzoi M-A, et  al. Impact of HIV infec­tion on
base­line char­ac­ter­is­tics and sur­vival of women with breast can­cer. AIDS.
2021;35(4):605-618.
7. Biggar RJ, Jaffe ES, Goedert JJ, Chaturvedi A, Pfeiffer R, Engels EA. Hod­
gkin lym­phoma and immu­no­de­fi­ciency in per­sons with HIV/AIDS. Blood.
2006;108(12):3786-3791.
8. Noy A. Optimizing treat­ment of HIV-asso­ci­ated lym­phoma. Blood. 2019;
1­34(17):1385-1394.
9. Bohlius J, Schmidlin K, Costagliola D, et al; Collaboration of Observational
HIV Epidemiological Research Europe (COHERE) Study Group. Incidence
and risk fac­tors of HIV-related non-Hodgkin’s lym­phoma in the era of com­
bi­na­tion antiretroviral ther­apy: a Euro­pean multicohort study. Antivir Ther.
2009;14(8):1065-1074.
10. Hernández-Ramírez RU, Qin L, Lin H, et al; North Amer­i­can AIDS Cohort
Collaboration on Research and Design of the International Epidemiologic
Databases to Evaluate AIDS. Association of immu­no­sup­pres­sion and HIV
viraemia with non-Hodgkin lym­phoma risk over­all and by sub­type in peo­
ple liv­ing with HIV in Canada and the USA: a multicentre cohort study.
Lancet HIV. 2019;6(4):e240-e249.
11. Silverberg MJ, Neuhaus J, Bower M, et al. Risk of can­cers dur­ing interrupted
antiretroviral ther­apy in the SMART study. AIDS. 2007;21(14):1957-1963.
12. Shindiapina P, Ahmed EH, Mozhenkova A, Abebe T, Baiocchi RA. Immunol­
ogy of EBV-related lymphoproliferative dis­ease in HIV-pos­i­tive indi­vid­u­als.
Front Oncol. 2020;10:1723.
13. Wang Q , De Luca A, Smith C, et al; Hepatitis Coinfection and Non Hodgkin
Lymphoma pro­ject team for the Collaboration of Observational HIV Epide­
miological Research Europe (COHERE) in EuroCoord. Chronic hep­a­ti­tis B
and C virus infec­tion and risk for non-Hodgkin lym­phoma in HIV-infected
patients: a cohort study. Ann Intern Med. 2017;166(1):9-17.
14. Montoto S, Shaw K, Okosun J, et al. HIV sta­tus does not influ­ence out­come
in patients with clas­si­cal Hodgkin lym­phoma treated with che­mo­ther­apy
using doxo­ru­bi­cin, bleomycin, vin­blas­tine, and dacarbazine in the highly
active antiretroviral ther­apy era. J Clin Oncol. 2012;30(33):4111-4116.
15. Rengan R, Mitra N, Liao K, Armstrong K, Vachani A. Effect of HIV on sur­
vival in patients with non-small-cell lung can­cer in the era of highly active
antiretroviral ther­apy: a pop­u­la­tion-based study. Lancet Oncol. 2012;​
13(12):​1203-1209.
16. Uldrick TS, Gonçalves PH, Abdul-Hay M, et al; Cancer Immunotherapy Tri­
als Network (CITN)-12 Study Team. Assessment of the safety of pembroli­
zumab in patients with HIV and advanced can­cer—a phase 1 study. JAMA
Oncol. 2019;5(9):1332-1339.
17. Food and Drug Administration (FDA). Cancer Clinical Trial Eligibility Cri-
teria: Patients With HIV, Hepatitis B Virus, or Hepatitis C Virus Infections:
Food and Drug Administration Guidance for Industry. Washington, DC:
FDA; 2020.
18. Hessol NA, Whittemore H, Vittinghoff E, et al. Incidence of first and sec­ond
pri­mary can­cers diag­nosed among peo­ple with HIV, 1985-2013: a pop­u­la­
tion-based, reg­is­try link­age study. Lancet HIV. 2018;5(11):e647-e655.
19. Picard C, Bobby Gaspar H, Al-Herz W, et al. International Union of Immu­
nological Societies: 2017 pri­mary immu­no­de­fi­ciency dis­eases com­mit­tee
report on inborn errors of immu­nity. J Clin Immunol. 2018;38(1):96-128.
20. Tangye SG, Al-Herz W, Bousfiha A, et al. Human inborn errors of immu­nity:
2019 update on the clas­si­fi­ca­tion from the International Union of Immuno­
logical Societies expert com­mit­tee. J Clin Immunol. 2020;40(1):24-64.
Prakash Singh Shekhawat
21. Tangye SG, Al-Herz W, Bousfiha A, et al. The ever-increas­ing array of novel
inborn errors of immu­nity: an interim update by the IUIS com­mit­tee. J Clin
Immunol. 2021;41(3):666-679.
22. Derpoorter C, Bordon V, Laureys G, Haerynck F, Lammens T. Genes at the
cross­ road of pri­ mary immunodeficiencies and can­ cer. Front Immunol.
2018;9:2544.
23. Riaz IB, Faridi W, Patnaik MM, Abra­ham RS. A sys­tem­atic review on pre­
dis­po­si­tion to lym­phoid (B and T cell) neo­pla­sias in patients with pri­mary
immunodeficiencies and immune dysregulatory dis­or­ders (inborn errors of
immu­nity). Front Immunol. 2019;10:777.
24. Mastio J, Saeed MB, Wurzer H, Krecke M, Westerberg LS, Thomas C. Higher
inci­dence of B cell malig­nan­cies in pri­mary immunodeficiencies: a com­bi­
na­tion of intrin­sic geno­mic insta­bil­ity and exo­cy­to­sis defects at the immu­
no­log­i­cal syn­apse. Front Immunol. 2020;11:581119.
25. Albert MH, Freeman AF. Wiskott-Aldrich Syndrome (WAS) and ded­i­ca­tor of
cyto­ki­ne­sis 8- (DOCK8) defi­ciency. Front Pediatr. 2019;7:451.
26. Biggs CM, Keles S, Chatila TA. DOCK8 defi­ciency: insights into path­o­phys­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/287/1852069/287pai.pdf by guest on 13 December 2021
i­­ol­ogy, clin­i­cal fea­tures and man­age­ment. Clin Immunol. 2017;181:75-82.
27. Yakaboski E, Fuleihan RL, Sullivan KE, Cunningham-Rundles C, Feuille E.
Lymphoproliferative dis­ease in CVID: a report of types and fre­quen­cies
from a US patient reg­is­try. J Clin Immunol. 2020;40(3):524-530.
28. Wehr C, Houet L, Unger S, et al. Altered spec­trum of lym­phoid neo­plasms
in a sin­gle-cen­ter cohort of com­mon var­i­able immu­no­de­fi­ciency with
immune dysregulation. J Clin Immunol. 2021;41(6):1250-1265.
29. Pulvirenti F, Pecoraro A, Cinetto F, et al. Gastric can­cer is the lead­ing cause
of death in Ital­ian adult patients with com­mon var­i­able immu­no­de­fi­ciency.
Front Immunol. 2018;9:2546.
30. Preite S, Gomez-Rodriguez J, Cannons JL, Schwartzberg PL. T and B-cell
sig­nal­ing in acti­vated PI3K delta syn­drome: from immu­no­de­fi­ciency to
auto­im­mu­nity. Immunol Rev. 2019;291(1):154-173.
31. Durandy A, Kracker S. Increased acti­va­tion of PI3 kinase-δ pre­dis­poses to
B-cell lym­phoma. Blood. 2020;135(9):638-643.
32. Zaki-Dizaji M, Akrami SM, Abolhassani H, Rezaei N, Aghamohammadi A.
Ataxia tel­an­gi­ec­ta­sia syn­drome: moon­light­ing ATM. Expert Rev Clin Immu-
nol. 2017;13(12):1155-1172.
33. Suarez F, Mahlaoui N, Canioni D, et  al. Incidence, pre­ sen­
ta­tion, and
prog­no­sis of malig­nan­cies in ataxia-tel­an­gi­ec­ta­sia: a report from the
French national reg­ is­
try of pri­
mary immune deficiencies. J Clin Oncol.
2015;33(2):202-208.
34. Hsu AP, McReynolds LJ, Holland SM. GATA2 defi­ciency. Curr Opin Allergy
Clin Immunol. 2015;15(1):104-109.
35. Moon WY, Powis SJ. Does nat­u­ral killer cell defi­ciency (NKD) increase the
risk of can­cer? NKD may increase the risk of some virus induced can­cer.
Front Immunol. 2019;10:1703.
36. Cohen JI. GATA2 Deficiency and Epstein-Barr virus dis­ease. Front Immunol.
2017;8:1869.
37. Kohn LA, Kohn DB. Gene ther­a­pies for pri­mary immune deficiencies. Front
Immunol. 2021;12:648951.
DOI 10.1182/hema­tol­ogy.2021000261
How immu­no­de­fi­ciency can lead to malig­nancy  |  295
 IMMUNOLOGY 101: WHAT THE PRACTICING HEMATOLOGIST NEEDS TO KNOW

Modified T cells as therapeutic agents

Nathan Singh

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/296/1851965/296singh.pdf by guest on 13 December 2021

Washington University School of Medicine, St Louis, MO

Immunotherapy is now a well-established modality in the treatment of cancer. Although several platforms to redirect the
immune response exist, the use of genetically modified T cells has garnered particular attention in recent years. This is
due, in large part, to their success in the treatment of B-cell malignancies. Adoptively transferred T cells have also dem-
onstrated efficacy in the treatment of systemic viral infections that occur following hematopoietic cell transplantation
prior to immune reconstitution. Here we discuss the techniques that enable redirection of T lymphocytes to treat cancer
or infection and the current indications for these therapies.

LEARNING OBJECTIVES
• Describe the methods that enable production of redirected T cells
• Review the current indications for approved T-cell products

CLINICAL CASE the multicomponent TCR genes are edited to produce

A 56-year-old man seeks treatment from his primary care phy-
sician for 2 months of progressive fatigue and night sweats.
Physical examination reveals axillary lymphadenopathy. Labo-
ratory studies reveal only an elevated lactate dehydrogenase
but are otherwise unremarkable. Positron emission tomogra-
phy (PET) reveals several sites of nodal fluorodeoxyglucose-
uptake. Lymph node biopsy specimen ultimately reveals
CD19+ CD10+ BCL6+ mature B cells with histopathologic and
cytogenetic features consistent with germinal center-type
large B-cell lymphoma. His Eastern Cooperative Oncology
Group (ECOG) performance status is assessed as 1, and he
initiates chemoimmunotherapy with combination rituximab,
cyclophosphamide, doxorubicin, vincristine, and prednisone
(R-CHOP). After 6 cycles, he has evidence of disease progres-
sion and undergoes salvage chemotherapy with rituximab,
ifosfamide, carboplatin, and etoposide (R-ICE). After 2 cycles,
PET reveals persistent progressive disease, demonstrating
chemotherapy resistance.
T cells as therapeutic agents
Several cellular attributes make T cells extremely effective
at their native function of controlling infection. First, T cells
are endowed with the ability to differentiate “self” from
“non-self” antigens, reflecting exquisite specificity. This
determination is made by the surface membrane antigen
receptor called the T-cell receptor (TCR). In a manner sim-
ilar to antibody production, the large genes encoding the
TCR undergo somatic recombination, a process in which
smaller genes. This rearrangement results in the production
of two recombined genes from more than 1018 potential
rearrangements, endowing each individual T-cell clone
with an essentially unique TCR. In addition to this antigen
specificity, T cells are potent killers of target cells, ensur-
ing efficient viral eradication. Upon antigen recognition,
T cells undergo rapid proliferation to generate a large
cytotoxic immune response. Finally, antigen-specific
T cells are able to differentiate into long-lived memory
cells, capable of mounting memory responses decades
after the initial infectious insult. Together, these features
not only make T cells highly evolved and efficient con-
trollers of infection but also make them very attractive as
anticancer agents.
Although T cells have great potential to serve as antileu-
kemic agents by recognition of host leukemia as “non-self”
(this is the primary therapeutic effect of allogeneic hema-
topoietic cell transplantation [HCT]), they also recognize
healthy host tissues as foreign, resulting in the syndrome of
graft-versus-host disease. Extrication of these 2 processes,
graft-versus-leukemia and graft-versus-host, has given rise
to the field of T-cell engineering for the purpose of redi-
recting this potent immune response toward cancer or
uncontrolled viral infection.
Redirecting the T-cell response
The processes of engineering T cells for use as anticancer
or antiviral therapeutics are fundamentally distinct and will
be addressed separately.

296 | Hematology 2021 | ASH Education Program


Figure 1. Process of manufacturing CAR T cells.
Cancer
Redirection of the T-cell response is depen­dent on the type of
anti­gen they are intended to tar­get. Antigens fall into 2 broad
categories—intra­cel­lu­lar and cell sur­face. For a T cell to rec­og­nize
pro­teins that are intra­cel­lu­lar, these pro­teins must be processed
and presented by major his­to­com­pat­i­bil­ity com­plex (MHC) pro­
teins for rec­og­ni­tion by TCRs. Introduction of a trans­genic TCR
that has been designed, or more often iden­ti­fied among nat­
u­rally occur­ ring TCRs, to be spe­ cific for a can­cer-asso­ ci­
ated
anti­gen allows for rec­og­ni­tion of intra­cel­lu­lar can­cer anti­gens.
This pro­cess, which recon­structs all­com­po­nents of the native
TCR com­plex to acti­vate the T-cell response, has thus far been
met with lim­ited suc­cess. Identification of immu­no­genic anti­
gens and TCR cog­na­tes is dif­fi­cult. Furthermore, iden­ti­fi­ca­tion of
can­cer-spe­cific anti­gens is unusual, and can­cer cells often down-
regulate sur­face MHC. Although sev­eral stud­ies have eval­u­ated
the effi­cacy of these agents,1 dis­ease con­trol has been mod­est.
For these rea­sons, an approach using syn­thetic anti­gen recep­
tors that are not depen­dent on MHC has been devel­oped. The
most well devel­oped of these are chi­me­ric anti­gen recep­tors
(CARs). These hybrid recep­tors con­tain an anti­gen rec­og­ni­tion
domain derived from a mono­clo­nal anti­body. This strat­egy al-
lows for MHC-inde­pen­dent anti­gen rec­og­ni­tion while pre­serv­
ing much of the native TCR-driven acti­va­tion machin­ery. The
down­side, of course, is loss of the abil­ity to tar­get intra­cel­lu­lar
anti­gens. This extra­cel­lu­lar anti­gen-bind­ing domain is linked to
intra­cel­lu­lar T-cell acti­vat­ing domains. Currently, only 2 sig­nal­
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/296/1851965/296singh.pdf by guest on 13 December 2021
ing domain struc­tures are used in the sev­eral CAR T-cell prod­
ucts approved for use by the US Food and Drug Administration
(FDA). Both con­tain the CD3ζ sig­nal­ing domain, which reca­pit­u­
lates the clas­si­cal “sig­nal 1” of T-cell stim­u­la­tion through the TCR.
These 2 struc­tures dif­fer in the com­po­si­tion of their costimulato-
ry domains, one employing CD28 and the other 41BB. Correla-
tive data from clin­i­cal tri­als dem­on­strate that CD28-based CARs
tend to undergo greater and more rapid expan­sion after deliv­ery
but do not per­sist for long peri­ods. Alternatively, 41BB-based
CARs do not expand as quickly or to the same degree but can
per­sist for years. Both CAR- and TCR-based approaches rely
on genetic engi­neer­ing to deliver either of these trans­genic
recep­tors to T cells. Significant advances in gene ther­apy us-
ing viral vec­tors have enabled effi­cient deliv­ery of genes en-
coding these recep­tors to T cells.
The pro­cess of manufactur­ing autol­o­gous T cells with syn­
thetic anti­ gen recep­ tors (Figure 1) begins with an ini­ tial leu-
kapheresis to iso­ late periph­ eral blood mono­ nu­
clear cells
(PBMCs). This PBMC prod­uct undergoes processing to purify
T cells, which are then cul­tured with a stim­u­la­tory agent, often
mag­netic beads containing stim­u­la­tory antibodies directed at
CD3 and CD28. This stim­u­la­tion induces T-cell acti­va­tion, DNA
rep­li­ca­tion, and pro­lif­er­a­tion. Following stim­u­la­tion, gene ther­
apy vec­tors (most often len­ti­vi­rus or ret­ro­vi­rus) are added to the
T-cell cul­tures, which medi­ate trans­fer of the tar­get transgene.
These engineered cells are then grown in cul­ture for sev­eral days
and finally are cryopreserved for even­tual infu­sion. All of these
Modified T cells as ther­a­peu­tic agents  |  297

Figure 2. Process of manufacturing virus-specific cells.
steps, whether at an aca­demic cen­ter for treat­ment on clin­i­cal
tri­als or at a com­mer­cial pro­duc­tion facil­ity, are performed under
Good Manufacturing Practice guide­lines. At sev­eral time points
dur­ing this manufactur­ing pro­cess, qual­ity con­trol mea­sures are
performed to ensure Good Manufacturing Practice stan­dards.
Prior to deliv­ery of the T-cell prod­uct, patients receive lympho-
depleting che­mo­ther­apy. Similar to the con­cept in allo­ge­neic
HCT, this lymphodepletion facil­i­tates suc­cess­ful engraft­ment of
the engineered prod­uct. This step can also fur­ther aid the ther­a­
peu­tic response by reduc­ing dis­ease bur­den, as CAR T cells are
still most often used for lym­phoid malig­nan­cies.
Infection
The pri­mary indi­ca­tion for adop­tive T-cell immu­no­ther­apy in the
treat­ment of infec­tious dis­eases is clin­i­cally sig­nif­i­cant viral infec­
tion fol­low­ing HCT. Over the past 2 decades, strat­eg ­ ies to treat
viruses that are respon­si­ble for sig­nif­i­cant mor­bid­ity and mor­tal­
ity, such as cyto­meg­a­lo­vi­rus,2 Epstein-Barr virus (EBV),3 ade­no­vi­
rus,4,5 human her­pes virus 6,6 BK virus,7 var­i­cella zoster virus,8 or
broad-spec­trum anti­vi­ral treat­ments,9 have been devel­oped. Most
stud­ies have been performed using autol­o­gous cells, derived from
either the patient them­selves in the set­ting of autol­o­gous trans­
plan­ta­tion or the donor in the set­ting of allo­ge­neic trans­plan­ta­
tion. The fun­da­men­tal prin­ci­ple under­ly­ing the gen­er­a­tion of virus-
spe­cific T cells (VSTs) in both set­tings is to expand a preexisting
pop­u­la­tion of anti­gen-spe­cific mem­ory T cells. This requires that
the patient or donor has pre­vi­ously been exposed to that virus and
an appro­pri­ate method to pres­ent stim­u­la­tory anti­gens to VSTs.
298  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/296/1851965/296singh.pdf by guest on 13 December 2021
Several strat­e­gies exist to iso­late and expand VSTs from a
bulk T-cell pop­u­la­tion ex vivo and dif­fer based on the type of
anti­gen and method of anti­gen pre­sen­ta­tion (Figure 2). The sim­
plest of these is to deliver a whole viral lysate to anti­gen present-
ing cells (APCs), such as den­dritic cells, and cocul­ture these with
T cells. This approach has lim­ited clin­i­cal util­ity, because there
is an under­ly­ing risk of trans­fer of a live infec­tious par­ti­cle. An
alter­na­tive is deliv­er­ing puri­fied or recom­bi­nant viral pro­teins or
plas­mid DNA encoding viral pro­teins to APCs. The most widely
used approach employs the use of pools of pep­tide deliv­ered
to APCs.10 After selec­tion of the stim­ul­a­tion strat­egy, VSTs are
most often expanded in repeat stim­u­la­tion cul­tures. This pro­cess
results in the pro­duc­tion of a large num­ber of poly­clonal T cells,
but manufactur­ing time is often sev­eral weeks to even months
long. An alter­na­tive approach is direct iso­la­tion of existing VSTs
from periph­eral blood. Antigen-spe­cific T cells will bind pep­tide-
HLA multimers, and these multimers can be iso­lated along with
bound T cells. This approach can fea­si­bly only be applied in set­
tings of high VST fre­quency and as such has lim­ited util­ity.
Clinical prod­ucts for the treat­ment of can­cer
The major­ity of clin­i­cal expe­ri­ence using engi­neer­ing T cells is
in the treat­ment of B-cell malig­nan­cies. Currently, 5 CAR T-cell
prod­ucts are approved for com­mer­cial use by the FDA; 4 of
these tar­get the B-cell anti­gen CD19, and 1 tar­gets the plasma
cell anti­gen B-cell mat­u­ra­tion anti­gen (BCMA). Although these
prod­ucts dif­fer in sev­eral ways, the pri­mary var­i­ance is derived
from the struc­ture of the recep­tor’s costimulatory domains. In
Table 1. Pivotal trials of CAR T cells for B cell malignancies

Costimulatory
Characteristic Target domain Patient group No. (evaluable) CR rate (%) FDA approved? Reference
ALL
  Penn/CHOP phase CD19 41BB Children 60 93 Yes 12
1/2a
  ELIANA phase 2 CD19 41BB Children 75 81 Yes 11
(tisagenlecleucel)
  MSKCC phase 1 CD19 CD28 Children 25 75 No 35
  Seattle phase 1/2 CD19 41BB Children 43 93 No 36
  NCI phase 1 CD19 CD28 Children 51 61 No 37

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/296/1851965/296singh.pdf by guest on 13 December 2021

  MSKCC phase 1 CD19 CD28 Adults 53 83 No 38
  Penn phase 1 CD19 41BB Adults 20 90 No 39
  NCI phase 1 CD22 41BB Children 15 73 No 40
  Penn/CHOP phase 1 CD22 41BB Adults and 8 50 No 41
chil­dren
Non-Hodgkin lym­phoma
 Tisagenlecleucel CD19 41BB Adults 93 40 Yes 13
  Axicabtagene CD19 CD28 Adults 101 54 Yes 14, 15
ciloleucel
  Lisocabtagene CD19 41BB Adults 256 53 Yes 23
maraleucel
  Brexucabtagene CD19 CD28 Adults 60 62 Yes 17
autoleucel
Multiple mye­loma
  Idecabtagene BCMA 41BB Adults 128 33 Yes 18
vicleucel
Chronic lym­pho­cytic leu­ke­mia
  Penn phase 2 CD19 41BB Adults 24 25 No 42
Trials iden­ti­fied by loca­tion and/or even­tual com­mer­cial prod­uct name.
CHOP, Children’s Hospital of Philadelphia; MSKCC, Memorial Sloan Kettering Cancer Center; NCI, National Cancer Institute; Penn, University of
Pennsylvania; Seattle, Seattle Children’s Hospital.

addi­tion to these FDA-approved prod­ucts, sev­eral oth­ers have
been eval­u­ated in clin­i­cal tri­als. Key find­ings from these stud­ies
are sum­ma­rized in Table 1.
Acute lym­pho­blas­tic leu­ke­mia
Currently, the only approved prod­ uct for the man­ age­
ment
of acute lym­ pho­blas­
tic leu­
ke­
mia (ALL) is tisagenlecleucel, a
CD19-targeted CAR containing the 41BB costimulatory do-
main.11,12 It is approved for patients up to 25 years of age who are
refrac­tory or in sec­ond or later relapse.
Non-Hodgkin lym­phoma
Four prod­ucts are approved for the treat­ment of non-Hodgkin
lym­phoma in adults. Tisagenlecleucel is indi­cated for patients
with dif­fuse large B-cell lym­phoma (DLBCL), high-grade B-cell
lym­phoma, or DLBCL aris­ing from fol­lic­u­lar lym­phoma that is
relapsed or refrac­tory after 2 or more lines of ther­apy.13 Axicabta-
gene ciloleucel is also directed at CD19 but con­tains a costimula-
tory domain derived from CD28.14,15 It is approved for adults with
DLBCL, pri­mary medi­as­ti­nal large B-cell lym­phoma, high-grade
B-cell lym­phoma, and DLBCL aris­ing from fol­lic­u­lar lym­phoma
that is relapsed or refrac­tory after 2 or more lines of ther­apy.
Prakash Singh Shekhawat
Lisocabtagene maraleucel also con­tains the 41BB costimulatory
domain but is deliv­ered as 2 indi­vid­ual prod­ucts: one com­posed
of CD4+ and the other of CD8+ T cells.16 It is indi­cated for all­of
the lym­phoma sub­types men­tioned above, as well as grade 3B
fol­lic­u­lar lym­phoma and DLBCL aris­ing from nonfollicular indo­lent
his­tol­o­gies. Finally, brexucabtagene autoleucel is struc­tur­ally
iden­ti­cal to axicabtagene ciloleucel but undergoes an addi­tional
B-cell deple­tion step dur­ing prod­uct manufactur­ing and is ap-
proved for relapsed or refrac­tory man­tle cell lym­phoma.17
Multiple mye­loma
Idecabtagene vicleucel was recently approved for adult patients
with relapsed or refrac­tory mul­ti­ple mye­loma after 4 or more
lines of ther­apy, includ­ing an immu­no­mod­u­la­tory agent, a prote-
asome inhib­i­tor, and an anti-CD38 mono­clo­nal anti­body.18 Unlike
the other approved ther­a­pies, it tar­gets the BCMA on plasma
cells. Like tisagenlecleucel and listocabtagene maraleucel, it
con­tains the 41BB costimulatory domain.
Patient selec­tion
Given the avail­abil­ity of CAR T-cell prod­ucts, par­tic­u­larly in the
man­age­ment of relapsed or refrac­tory non-Hodgkin lym­phoma,
Modified T cells as ther­a­peu­tic agents  |  299

Figure 3. Currently enrolling CAR T cell trials across the globe.
a key clin­i­cal con­sid­er­ation is when to use this ther­apy and which
ther­apy to use. At many aca­demic cen­ters, CAR T-cell ther­apy is
com­monly con­sid­ered for patients with pro­gres­sive lym­phoma
after sec­ond-line che­mo­ther­apy or in the set­ting of relapse after
autol­o­gous trans­plan­ta­tion. Decision mak­ing about using CAR T
cells vs autol­o­gous trans­plan­ta­tion, spe­cif­i­cally in the set­ting of
failed first-line ther­apy, is an active topic of dis­cus­sion and will be
assisted by the results of cur­rent ongo­ing ran­dom­ized clin­ic ­ al tri­
als (NCT03391466, NCT03570892, NCT03575351).
Incidence and man­age­ment of toxicities is also a rel­e­vant
con­sid­er­ation. The most com­mon tox­ic­ity occur­ring fol­low­
ing CAR T-cell ther­apy is cyto­kine release syn­drome (CRS), a
sys­temic hyperinflammatory syn­drome driven by supraphysi-
ologic pro­duc­tion of inter­leu­kin 6 (IL-6) bystander mye­loid-lin­
e­age cells in response to robust T-cell acti­va­tion.19,20 The use
of IL-6 and IL-6 recep­tor inhib­i­tors, such as tocilizumab, has
sig­nif­i­cantly improved out­comes for patients. Another highly
mor­bid tox­ic­ity is immune effec­tor cell-asso­ci­ated neu­ro­tox­ic­
ity syn­drome (ICANS),21 a poorly under­stood neu­ro­psy­chi­at­ric
syn­drome that is com­monly man­aged with cor­ti­co­ste­roids and
anti­ep­i­lep­tics.22 In rare cases, ICANS can prog­ress and lead to
cere­bral edema, a much more seri­ous con­di­tion that is pri­mar­ily
man­aged with high-dose ste­roids. Intriguingly, both toxicities
appear to be more severe with CD28-based CAR T cells, and for
this rea­son, many cli­ni­cians will elect to use 41BB-based ther­
apy, if avail­­able, for patients who are more frail. Clinical tri­als of
CAR T prod­ucts only enrolled patients with high-per­for­mance
sta­tuses (ECOG 0-1). However, real-world data sug­gest that
patients who are not as fit can be suc­cess­fully treated with CAR
ther­apy.23 Indeed, given the man­age­ment options now avail­­
able for CRS and ICANS, most patients are con­sid­ered eli­gi­ble
for this ther­apy. Other toxicities include prolonged pan­cy­to­pe­
nia and a dura­ble loss of B cells with asso­ci­ated hypogamma-
globulinemia.
300  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/296/1851965/296singh.pdf by guest on 13 December 2021
Finally, given the autol­o­gous nature of this ther­apy, prod­
uct manufactur­ ing is also worth con­ sid­er­
ing. Several stud­ ies
have dem­on­strated that both pre­vi­ous expo­sure to che­mo­ther­
apy and the immu­no­sup­pres­sive envi­ron­ment encum­bered by
active malig­nancy can sig­nif­i­cantly impair the effec­tive expan­
sion of col­lected T cells.24,25 Several cen­ters have explored “early
bank­ing” pro­ce­dures, in which high-risk patients undergo apher­
e­sis and stor­age of T cells prior to ini­ti­at­ing first-line ther­apy to
ensure avail­abil­ity of “fit” T cells should they be needed. This
approach remains inves­ti­ga­tional and should only be con­sid­ered
in the con­text of a clin­i­cal trial.
The avail­abil­ity of CAR T-cell prod­ucts has increased sig­nif­i­
cantly in the past 5 years. For chil­dren with ALL, ther­apy is almost
always deliv­ered in an aca­demic set­ting, which also facil­i­tates
avail­abil­ity of tisagenlecleucel. At the time of writ­ing, there were
324 actively recruiting clin­i­cal tri­als of CAR T-cell prod­ucts. Most
of these are in the United States (148) and China (150), reflect-
ing the dis­par­ity in access to this plat­form that exists across the
globe (Figure 3).
Clinical prod­ucts for the treat­ment of infec­tion
Currently, there are no FDA-approved prod­ucts avail­­able for the
treat­ment of viral infec­tion using T cells. Three phase 3 stud­ies are
cur­rently ongo­ing (NCT04832607, NCT04390113, NCT03394365)
eval­u­at­ing the effi­cacy of VSTs. One (NCT03394365) is focused on
the effi­cacy of a VST prod­uct targeting EBV anti­gens in the man­
age­ ment of EBV-asso­ ci­
ated posttransplant lymphoproliferative
dis­or­der,26 as well chronic active EBV and EBV-driven hemophago-
cytic lymphohistiocytosis (HLH). Greater than 100 clin­i­cal tri­als
at all­phases are cur­rently enroll­ing across the globe. An excit­ing
pros­pect is the devel­op­ment of allo­ge­neic VSTs that can serve as
an “off-the-shelf” treat­ment, elim­i­nat­ing the time delay between
dis­ease onset and treat­ment.27

CLINICAL CASE (Con­t in­u ed)
After discussing the dis­ tinc­
tions between autol­ o­
gous trans­
plan­ta­tion and CAR T cells, the patient elects to pro­ceed with
CAR T cells. He undergoes apher­e­sis for planned ther­apy with
tisagenlecleucel, with suc­cess­ful prod­uct manufactur­ing. He
receives con­di­tion­ing che­mo­ther­apy with fludarabine and
cyclo­phos­pha­mide, followed by CAR T-cell infu­sion. He expe­
ri­ences chills and rig­ors on day +3, clas­si­fied as grade 1 CRS.
These resolve spon­ta­ne­ously on day +5. After count recov­ery
on day +11, he is discharged from the hos­pi­tal with­out incur­ring
any fur­ther tox­ic­ity. Thirty days after treat­ment, PET/com­puted
tomog­ra­phy reveals a par­tial response. PET/com­puted tomog­
ra­phy at 3 months reveals com­plete remis­sion.
Challenges and next-gen­er­a­tion T cells for can­cer
Although great prog­ ress has been made in the use of engi-
neered T cells, par­tic­u­larly with regard to the treat­ment of hema­
to­logic can­cers, the real­ity is that most patients treated with
CD19-directed CAR T cells will not achieve dura­ble remis­sions.
The 2 fea­tures that have been iden­ti­fied as crit­i­cal bar­ri­ers to the
broader suc­cess of this ther­apy are (1) ini­tial fail­ure that leads to
non­re­sponse and (2) acquired fail­ure that leads to dis­ease re-
lapse. The for­mer is more com­mon in lym­phoma, whereas the
lat­ter is more com­mon in ALL. Several stud­ies have iden­ti­fied
T cell–intrin­sic fea­tures that are asso­ci­ated with ini­tial fail­ure,
such as expres­sion of dys­func­tion-asso­ci­ated genes, ter­mi­nal
dif­fer­en­ti­a­tion, and impaired T-cell per­sis­tence.25,28-30 New strat­
e­gies to enhance T-cell fit­ness using inno­va­tive manufactur­ing
pro­cesses have gen­er­ated great excite­ment and open ave­nues
to the manufactur­ing of T cells with enhanced func­tion.31,32
Several mech­a­nisms have been described that lead to dis­ease
relapse, most thor­oughly the phe­nom­e­non of anti­gen escape. In
approx­i­ma­tely 40% of patients with ALL who relapse, the recur­
rent leu­ke­mia lacks detect­able sur­face CD19, enabling dis­ease
out­growth.33,34 To over­ come anti­ gen escape, sev­ eral groups
have employed dual-anti­ gen targeting to more thor­ oughly
cap­ture malig­nant clones and reduce the like­li­hood of anti­gen
loss as a mech­a­nism of relapse (NCT03241940, NCT03330691,
NCT03448393).
Conflict-of-inter­est dis­clo­sure
Nathan Singh holds sev­eral pat­ents related to CAR and other
engineered cell ther­a­pies.
Off-label drug use
Nathan Singh: nothing to disclose.
Correspondence
Nathan Singh, Washington University School of Medicine, 660 S.
Euclid Avenue, St Louis, Missouri 63110; e-mail: singh​­.nathan@gmail​
­.com.
References
1. Stadtmauer EA, Fraietta JA, Davis MM, et al. CRISPR-engineered T cells in
patients with refrac­tory can­cer. Science. 2020;367(6481):eaba7365.
2. Berger C, Turtle CJ, Jensen MC, Riddell SR. Adoptive trans­fer of virus-spe­
cific and tumor-spe­cific T cell immu­nity. Curr Opin Immunol. 2009;21(2):
224-232. Prakash Singh Shekhawat
3. Heslop HE, Slobod KS, Pule MA, et  al. Long-term out­ come of EBV-
spe­cific T-cell infu­sions to pre­vent or treat EBV-related lymphoproliferative
dis­ease in trans­plant recip­i­ents. Blood. 2010;115(5):925-935.
4. Leen AM, Christin A, Myers GD, et al. Cytotoxic T lym­pho­cyte ther­apy with
donor T cells pre­vents and treats ade­no­vi­rus and Epstein-Barr virus infec­
tions after haploidentical and matched unre­lated stem cell trans­plan­ta­tion.
Blood. 2009;114(19):4283-4292.
5. Zandvliet ML, Falkenburg JH, van Liempt E, et  al. Combined CD8+ and
CD4+ ade­no­vi­rus hexon-spe­cific T cells asso­ci­ated with viral clear­ance
after stem cell trans­plan­ta­tion as treat­ment for ade­no­vi­rus infec­tion. Hae-
matologica. 2010;95(11):1943-1951.
6. Gerdemann U, Keukens L, Keirnan JM, et al. Immunotherapeutic strat­e­gies
to pre­vent and treat human her­pes­vi­rus 6 reactivation after allo­ge­neic stem
cell trans­plan­ta­tion. Blood. 2013;121(1):207-218.
7. Blyth E, Clancy L, Simms R, et al. BK virus-spe­cific T cells for use in cel­lu­lar
ther­apy show spec­i­fic­ity to mul­ti­ple anti­gens and polyfunctional cyto­kine
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/296/1851965/296singh.pdf by guest on 13 December 2021
responses. Transplantation. 2011;92(10):1077-1084.
8. Blyth E, Gaundar SS, Clancy L, et al. Clinical-grade var­i­cella zoster virus-spe­
cific T cells pro­duced for adop­tive immu­no­ther­apy in hemo­poi­etic stem cell
trans­plant recip­i­ents. Cytotherapy. 2012;14(6):724-732.
9. Papadopoulou A, Gerdemann U, Katari UL, et  al. Activity of broad-spec­
trum T cells as treat­ment for AdV, EBV, CMV, BKV, and HHV6 infec­tions after
HSCT. Sci Transl Med. 2014;6(242):242ra83.
10. Koehne G, Hasan A, Doubrovina E, et al. Immunotherapy with donor T cells
sen­si­tized with overlapping pentadecapeptides for treat­ment of per­sis­
tent cyto­meg­a­lo­vi­rus infec­tion or vire­mia. Biol Blood Marrow Transplant.
2015;21(9):1663-1678.
11. Maude SL, Laetsch TW, Buechner J, et al. Tisagenlecleucel in chil­dren and young
adults with B-cell lym­pho­blas­tic leu­ke­mia. N Engl J Med. 2018;378­(5):439-448­.
12. Maude SL, Frey N, Shaw PA, et  al. Chimeric anti­gen recep­tor T cells for
sustained remis­sions in leu­ke­mia. N Engl J Med. 2014;371(16):1507-1517.
13. Schuster SJ, Bishop MR, Tam CS, et al; JULIET Investigators. Tisagenlecleu-
cel in adult relapsed or refrac­tory dif­fuse large B-cell lym­phoma. N Engl J
Med. 2019;380(1):45-56.
14. Locke FL, Ghobadi A, Jacobson CA, et al. Long-term safety and activ­ity of
axicabtagene ciloleucel in refrac­tory large B-cell lym­phoma (ZUMA-1): a
sin­gle-arm, multicentre, phase 1-2 trial. Lancet Oncol. 2019;20(1):31-42.
15. Neelapu SS, Locke FL, Bartlett NL, et  al. Axicabtagene ciloleucel CAR
T-cell ther­ apy in refrac­ tory large B-cell lym­ phoma. N Engl J Med.
2017;377(26):2531-2544.
16. Abramson JS, Palomba ML, Gordon LI, et  al. Lisocabtagene maraleu-
cel for patients with relapsed or refrac­ tory large B-cell lym­ pho­ mas
(TRANSCEND NHL 001): a multicentre seam­ less design study. Lancet.
2020;396(10254):839-852.
17. Wang M, Munoz J, Goy A, et al. KTE-X19 CAR T-cell ther­apy in relapsed or
refrac­tory man­tle-cell lym­phoma. N Engl J Med. 2020;382(14):1331-1342.
18. Munshi NC, Anderson LD Jr, Shah N, et  al. Idecabtagene vicleucel in
relapsed and refrac­tory mul­ti­ple mye­loma. N Engl J Med. 2021;384(8):
705-716.
19. Singh N, Hofmann TJ, Gershenson Z, et al. Monocyte lin­e­age-derived IL-6
does not affect chi­me­ric anti­gen recep­tor T-cell func­tion. Cytotherapy.
2017;19(7):867-880.
20. Giavridis T, van der Stegen SJC, Eyquem J, Hamieh M, Piersigilli A, Sadelain M.
CAR T cell-induced cyto­kine release syn­drome is medi­ated by mac­ro­phages
and abated by IL-1 block­ade. Nat Med. 2018;24(6):731-738.
21. Gust J, Ponce R, Liles WC, Garden GA, Turtle CJ. Cytokines in CAR T cell-
asso­ci­ated neu­ro­tox­ic­ity. Front Immunol. 2020;11:577027.
22. Maus MV, Alexander S, Bishop MR, et  al. Society for immu­no­ther­apy of
can­cer (SITC) clin­i­cal prac­tice guide­line on immune effec­tor cell-related
adverse events. J Immunother Cancer. 2020;8(2):e001511.
23. Nastoupil LJ, Jain MD, Feng L, et al. Standard-of-care axicabtagene ciloleu-
cel for relapsed or refrac­tory large B-cell lym­phoma: results from the US
lym­phoma CAR T con­sor­tium. J Clin Oncol. 2020;38(27):3119-3128.
24. Das RK, Vernau L, Grupp SA, Barrett DM. Naïve T-cell def­i­cits at diag­no­sis
and after che­mo­ther­apy impair cell ther­apy poten­tial in pedi­at­ric can­cers.
Cancer Discov. 2019;9(4):492-499.
25. Singh N, Perazzelli J, Grupp SA, Barrett DM. Early mem­ory phe­no­types
drive T cell pro­lif­er­a­tion in patients with pedi­at­ric malig­nan­cies. Sci Transl
Med. 2016;8(320):320ra3.
26. Prockop S, Doubrovina E, Suser S, et  al. Off-the-shelf EBV-spe­cific T cell
immu­no­ther­apy for rituximab-refrac­tory EBV-asso­ci­ated lym­phoma fol­low­
ing trans­plan­ta­tion. J Clin Invest. 2020;130(2):733-747.
27. Tzannou I, Papadopoulou A, Naik S, et al. Off-the-shelf virus-spe­cific T cells to
treat BK virus, human her­pes­vi­rus 6, cyto­meg­a­lo­vi­rus, Epstein-Barr virus, and
Modified T cells as ther­a­peu­tic agents  |  301
 ade­no­vi­rus infec­tions after allo­ge­neic hema­to­poi­etic stem-cell trans­plan­ta­
tion. J Clin Oncol. 2017;35(31):3547-3557.
28. Deng Q , Han G, Puebla-Osorio N, et al. Characteristics of anti-CD19 CAR T
cell infu­sion prod­ucts asso­ci­ated with effi­cacy and tox­ic­ity in patients with
large B cell lym­pho­mas. Nat Med. 2020;26(12):1878-1887.
29. Fraietta JA, Lacey SF, Orlando EJ, et al. Determinants of response and resis­
tance to CD19 chi­me­ric anti­gen recep­tor (CAR) T cell ther­apy of chronic
lym­pho­cytic leu­ke­mia. Nat Med. 2018;24(5):563-571.
30. Singh N, Lee YG, Shestova O, et al. Impaired death recep­tor sig­nal­ing in
leu­ke­mia causes anti­gen-inde­pen­dent resis­tance by induc­ing CAR T-cell
dys­func­tion. Cancer Discov. 2020;10(4):552-567.
31. Lynn RC, Weber EW, Sotillo E, et  al. c-Jun overexpression in CAR T cells
induces exhaus­tion resis­tance. Nature. 2019;576(7786):293-300.
32. Weber EW, Parker KR, Sotillo E, et  al. Transient rest restores func­tion­al­
ity in exhausted CAR-T cells through epi­ ge­ netic remodeling. Science.
2021;372(6537):eaba1786.
33. Orlando EJ, Han X, Tribouley C, et al. Genetic mech­a­nisms of tar­get anti­
gen loss in CAR19 ther­apy of acute lym­pho­blas­tic leu­ke­mia. Nat Med.
2018;24(10):1504-1506.
34. Sotillo E, Barrett DM, Black KL, et al. Convergence of acquired muta­tions
and alter­na­tive splic­ing of CD19 enables resis­tance to CART-19 immu­no­
ther­apy. Cancer Discov. 2015;5(12):1282-1295.
35. Curran KJ, et al. Toxicity and response after CD19-specific CAR T-cell therapy
in pediatric/young adult relapsed/refractory B-ALL. Blood. 2019;134:2361–
2368.
302  |  Hematology 2021  |  ASH Education Program
36. Gardner RA, et al. Intent-to-treat leukemia remission by CD19 CAR T cells
of defined formulation and dose in children and young adults. Blood.
2017;129:3322-3331.
37. Lee DW, et al. T cells expressing CD19 chimeric antigen receptors for acute lym-
phoblastic leukaemia in children and young adults: a phase 1 dose-escalation
trial. Lancet. 2015;385:517-528.
38. Park JH, et al. Long-term follow-up of CD19 CAR therapy in acute lympho-
blastic leukemia. N Engl J Med. 2018;378:449-459.
39. Frey NV, et al. Optimizing chimeric antigen receptor T-cell therapy for
adults with acute lymphoblastic leukemia. J Clin Oncol. 2020;38:415-422.
40. Fry TJ, et al. CD22-targeted CAR T cells induce remission in B-ALL that
is naive or resistant to CD19-targeted CAR immunotherapy. Nat Med.
2018;24:20-28.
41. Singh N, et al. Antigen-independent activation enhances the efficacy of
4-1BB-costimulated CD22 CAR T cells. Nat Med. 2021.
42. Frey NV, et al. Long-term outcomes from a randomized dose optimization
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/296/1851965/296singh.pdf by guest on 13 December 2021
study of chimeric antigen receptor modified T cells in relapsed chronic
lymphocytic leukemia. J Clin Oncol. 2020;38:2862-2871.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000262
 INDOLENT LYMPHOMAS

What factors guide treatment selection in

mycosis fungoides and Sezary syndrome?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/303/1852064/303kim.pdf by guest on 13 December 2021

Youn H. Kim
Department of Dermatology, Stanford University School of Medicine, Palo Alto, CA

Cutaneous T-cell lymphoma (CTCL) comprises a spectrum of T-cell lymphomas with primary skin involvement. Mycosis
fungoides (MF) and Sezary syndrome (SS) are the common subtypes of CTCL in which patients present with widely
diverse profiles of skin involvement and varying extents of extracutaneous disease. Patients with early-stage disease have
an excellent prognosis and are managed primarily with skin-directed therapies; however, those with advanced-stage MF
or SS often require multiple lines and recurrent courses of systemic therapies. Many options are available when consid-
ering systemic agents, and it is often challenging to know how to prioritize therapies to address a patient’s objective
disease and quality of life issues. Appreciating the disease heterogeneity and understanding the patient’s overall disease
profile (eg, skin, lymph nodes, blood, large cell transformation) serve as a useful framework in aligning therapies that
can optimally treat active sites of disease. Tissue or blood biomarkers can be integrated into our process of prioritizing
therapies and personalizing management in MF or SS. Multidisciplinary management and optimizing supportive care are
additional key elements for a favorable outcome. Appropriate patients with high-risk disease should be considered for
allogeneic hematopoietic stem cell transplant.

LEARNING OBJECTIVES
• Understand the clinical heterogeneity of patients with MF and SS
• Appreciate the unique quality of life issues associated with cutaneous disease
• Identify the clinical, pathological, and other biomarkers that help prioritize therapies and personalize clinical man­
agement in MF and SS

Introduction skin electron beam therapy [TSEBT], extracorporeal pho­

Mycosis fungoides (MF) and Sezary syndrome (SS) are the
most common subtypes of cutaneous T­cell lymphoma
(CTCL), and although the skin is the primary site of involve­
ment, the malignant T cells may expand in the lymph node
(LN), visceral, and blood compartments.1 In the skin lesions
of early­stage MF, significant inflammatory infiltrate of
immune stimulated profile is observed. However, as the
disease progresses the malignant T cells acquire a Th2
phenotype accompanied by a microenvironment skewed
toward a more immune­suppressive Th2 cytokine profile.2,3
A subset of patients with advanced MF or SS undergo a
transformation into large cell disease (LCT) that is often
associated with more aggressive biological behavior.
Patients with MF and SS subgroups are often distinct clini­
cally and biologically but also share similarities, and clinical
features can shift from MF to SS and vice versa in the course
of a patient’s disease.
The clinical management of MF/SS involves therapies
developed specifically for patients with CTCL (eg, total
Prakash Singh Shekhawat
topheresis, bexarotene, vorinostat, romidepsin) and those
originally developed to treat other lymphomas but co­
opted for CTCL (eg, brentuximab vedotin [BV], pralatrex­
ate). More recently, new systemic therapies have emerged,
built upon rationally selected targets in MF/SS (eg, mog­
amulizumab [MOGA], anti­KIR3DL2 antibody; Table 1).
Although these newer agents represent a major advance­
ment, they have not altered our fundamental treatment
strategy. Only a slim minority of patients will ever be cured of
their disease, and most patients will receive numerous lines
of therapy during their lives. Each failed treatment takes a
physical and psychological toll on patients. We have long
known that a one­size­fits­all approach cannot work for this
heterogeneous disease. As our therapeutic armamentarium
continues to expand, we are now challenged with how to
best match patients to the treatments most likely to ben­
efit them. Furthermore, it is unclear how best to sequence
or combine therapies, and given the lack of curative thera­
pies, the ones with fewer cumulative toxicities and durable
Personalizing treatment selection in MF/SS | 303
 Table 1. Clinical activ­ity of sys­temic ther­a­pies in CTCL* by dis­ease sub­type (clin­i­cal stage, MF vs SS), com­part­ments (skin, LN, blood), ±LCT; tol­er­a­bil­ity

Global/com­pos­ite clin­i­cal activ­ity of all­

Tolerability
com­part­ments*
fac­tors in chronic
Treatment N; sub­type ORR/CR % (n/N) DOR, median PFS, median Skin activ­ity LN activ­ity Blood activ­ity LCT included ther­apy
Standard ther­a­pies
 Bexarotene†,22 56; MF/SS 45% (25/56)/1 CR 42.7 weeks 13.9 weeks Skin RR = pri­mary Limited LN data: Of 17 SS, 8 had N/a Hyperlipidemia
(300 mg/m2 = opti­ IIB-IV; end point: Of 25 skin OR, par­tial Sezary
mal dose) SS (17/56) IIB, 57% (13/23); 7 w/clin­i­cal data:
III, 32% (6/19); LAD - 4/8 PR
IVA, 44% (4/9); 3/7 PR or CR 3/8 SD
IVB, 40% (2/5); 4/7 PD 1/8 PD
SS, 24% (4/17, all­
doses)
 Extracorporeal 51; MF/SS 63% (32/51)/8 CR 22.4 months N/a Compartment-spe­cific data not reported but 16% CR N/a None
photopheresis‡,24 III-IV; SS (39/51) w/clear­ing of blood Sezary dis­ease
 Vorinostat63 74; MF/SS IB-IV; 30% (22/74)/1 CR >185 days TTP median Skin RR = pri­mary LN RR 42% 14/27 SS w/SC Yes Fatigue,
SS (30/74) after 281 days 148 days end point: (10/24) reduc­tion dysgeusia,
IB/IIA, 31% (4/13); >25% ↓plate­lets

304  |  Hematology 2021  |  ASH Education Program

IIB-IV, 30% (18/61);
skin tumors, 23%
(5/22);
SS, 33% (10/30)
 Romidepsin14 96; MF/SS 34% (33/96)/6 CR; 15 months TTP median 8 Skin RR 40% (38/96) LN response Blood RR 38% Yes Fatigue,
IB-IV; SS SS (B1-2), 32% months 35% (13/37) (14/37, 2 dysgeusia,
(37/96) (12/37) by RECIST CR); B2, 46% ↓plate­lets
(6/13, 2 CR)
 Pralatrexate§,37 29; CTCL 45% (13/29)/1 CR Not reached; Not reached; Compartment-spe­cific data not reported, response by Yes Mucositis
(opti­mal CTCL IB-IV; SS (8/29) 73% cont OR 388 days stage:
dose 15 mg/m2 at 6 months >15 mg/m2 MF by stage
weekly 3/4) IB 60% (3/5);
IIB 67% (4/6);
IVA 60% (3/5);
SS 25% (2/8)
 Brentuximab11 48; MF IA-IV; 65% (31/48)/ 15.1 months; 15.9 months; Skin RR 77% (37/48); 2 w/LN+, stage Excluded B2/SS Yes (17/48) PN
CD30 > 10% 5 CR; ORR4 LCT MF/ALCL LCT 15.5 median DOR 20.6 IVA: ORR
(no SS) 65% (11/17) months months, MF/ALCL 100% (2/2,
1 CR)
 Mogamulizumab¶,13 186; MF/SS 28% (52/186)/6 14.1 months; MF 7.7 months Skin RR 42% LN RR 17% Blood RR 68% No (LCT Rash
IB-IV; SS CR; MF 21% 13.1 months; (78/186); median (21/124); (83/122, excluded)
(81/186) (22/105); SS SS 17.3 TTR 3.0 months; median TTR 54 CR);
37% (30/81) months median DOR 20.6 3.3 months; median TTR
months median DOR 1.1 months;
15.5 months median DOR
25.5 months
 Pembrolizumab41 24; MF/SS IIB- 38% (9/24)/2 CR; Not reached; Not reached; Skin RR 38% (9/24); N/a Baseline w/B2, Yes (4/24) irAE (coli­tis, pneu­
IV (23/24); SS 27% (4/15); median PFS at 1 year, 6/24, >90% skin n = 6, 17% (1/6) mo­ni­tis)
SS (15/24) LCT 25% (1/4) fol­low-up 65% clear­ing
58 weeks

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/303/1852064/303kim.pdf by guest on 13 December 2021

 Table 1. (continued)

Global/com­pos­ite clin­i­cal activ­ity of all­

Tolerability
com­part­ments*
fac­tors in chronic
Treatment N; sub­type ORR/CR % (n/N) DOR, median PFS, median Skin activ­ity LN activ­ity Blood activ­ity LCT included ther­apy
  Liposomal doxo­ru­bi­ 49; 41% (20/49)/3 CR 6 months 6.2 months; Skin RR 53% (26/49); LN ± vis­ceral Excluded SS Yes
cin§,39 (20 mg/m2, MF IIB-IV (no TTP median best PR/CR dis­ease had
day 1, day 15, SS) 7.4 months lower RR than
28-day cycle) skin-only (22%
vs 52%)
 Gemcitabine§,40 31; MF/SS; SS 65% (20/31)/3 CR; 4.1 months N/a Compartment-spe­cific data not reported but case exam­ Yes (13/31)
(1000 mg/m2, day (11/33) LCT 54% (7/13); ples with activ­ity in LN/blood dis­ease
1, 8, 15) SS 73% (8/11)
  Investigative
ther­a­pies
(w/peer-reviewed
papers)
 Anti-KIR3DL2 anti­ 44; SS (35/44); SS, 43% (15/35); 13.8 months; 8.2 months; SS Of SS: skin RR 51% Of SS: LN RR 11% Of SS: blood RR Yes (6/35 SS, Well tol­er­ated
body (lacutamab), MF IB-IV MF, 13% (1/8); SS 13.8 11.7 months, (18/35, 3 CR) (2/18, 1 CR) 56% (19/34, 5/8 MF)
phase 1/2 study (8/44) LCT no OR months; MF MF 3.9 9 CR)
data (ongo­ing piv­ 6.9 months months
otal trial in SS)25
 Duvelisib64 19; MF/SS; MF 32% no CR; MF n/a; 4.5 months Compartment-spe­cific data not reported Yes (4/19) ↑LFTs
(13/19); SS 38% (5/13); LCT DOR range 0.7-
(5/19) 25% (1/4); SS 10.1 months;
20% (1/5) TTR 2.4

Prakash Singh Shekhawat

months
 Lenalidomide65 32; MF/SS 28% (9/32) no CR 10 months; 8 months Skin RR 53% (10/19) Reports activ­ity, Blood RR 38% N/a Fatigue, skin
IB-IV; TTR 25 mg, MF IB-IV no details (5/13) flares, tran­sient
SS (11/32) 2 months; ↑blood SCs
10 mg,
4 months
*Response assess­ment tools and/or cri­te­ria for clin­i­cal end points may vary across CTCL tri­als.
†
Time to response is lon­ger than other listed treatments, 300 mg/m2, TTR = 15.7 weeks (180 days), >300 mg/m2, TTR = 8.4 weeks (59 days); poor details on extracutaneous dis­ease sites (4/9
entered as stage IVB had LN+ only); “rate of PD” = 39% (22/56) and 32% (12/38) for 300 mg/m2 and >300 mg/m2; median TTR for other treatment ~2 cycles.
‡
Median time to response 8 months.
§
Caution when used with RT, espe­cially TSEBT.
¶
Caution when used to bridge to allo­ge­neic HSCT (pos­si­ble risk of severe GVHD).
ALCL, ana­plas­tic large cell lym­phoma; DOR, duration of response; irAE, immune-related adverse event; LFT, liver function test; n/a, not appli­ca­ble; OR, odds ratio; PFS, pro­gres­sion-free sur­vival;
PN, peripheral neuropathy; RECIST, response eval­u­a­tion cri­te­ria in solid tumors; RT, radiation therapy; SD, stable disease; TTR, time to response; TTP, time to progression.

Personalizing treat­ment selec­tion in MF/SS  |  305

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/303/1852064/303kim.pdf by guest on 13 December 2021
responses are pri­or­i­tized. In selected high-risk advanced-stage
MF or SS, allo­ge­neic hema­to­poi­etic stem cell trans­plant (HSCT)
should be con­sid­ered as a poten­tial cura­tive ther­apy.
CLINICAL CASE
A 64-year-old male presented with a 2-year his­tory of pro­gres­
sive skin rash that cur­rently affected >80% of the body sur­face
area with gen­er­al­ized erythroderma asso­ci­ated with severe,
debil­i­tat­ing itching. Topical and sys­temic ste­roids were min­
i­
mally help­ ful. Skin biop­sies showed an atyp­ i­
cal T-cell infil­
trate. Blood flow dem­on­strated an abnor­mal T-cell pop­u­la­tion
expressing CD3 and CD4 but lacking CD7 and CD26, a phe­
no­type con­sis­tent with Sezary cells (SCs), which represented
68% of lym­pho­cytes with an abso­lute count of 1250/mm3; T-cell
recep­tor next-gen­er­a­tion sequenc­ing (NGS) stud­ies supported
a clonal T-cell pro­cess in the skin and blood. Whole-body pos­
i­tron emis­sion tomog­ra­phy/com­puted tomog­ra­phy (PET/CT)
imag­ing showed an approx­im ­ a­tely 2 cm axil­lary and ingui­nal
lymph­ade­nop­a­thy (LAD) with a max­i­mum stan­dard­ized uptake
value of 3 to 4, con­sis­tent with reac­tive or dermatopathic LNs.
A diag­no­sis of SS was established, presenting with low-bur­den
SCs and clin­i­cally reac­tive LAD and thus clin­i­cal stage IVA1. The
patient was treated with oral bexarotene with over­all global
par­ tial response (75% improve­ ment of his skin dis­ ease and
>50% reduc­tion of his SCs), and sup­port­ive care was pro­vided
with top­ic ­ al ste­roids, emol­lients, and gabapentin for itching.
Six months later while con­tinu­ing main­te­nance bexarotene, he
Thinner skin disease, erythroderma,
blood disease
Mogamulizumab
Lacutamab
(investigational)
Sézary cells Romidepsin
Pembrolizumab
Figure 1. Examples of newer systemic agents and their clinical activity in different patient profiles.
306  |  Hematology 2021  |  ASH Education Program
noted wors­en­ing of his skin dis­ease with recur­rent erythrod­
erma. A skin biopsy con­firmed CTCL with­out evi­dence of LCT.
His blood Sezary flow now showed >10,000 SCs/mm3. Updated
PET/CT find­ings again supported reac­tive or dermatopathic
LAD with no hepatosplenomegaly.
Recognizing dis­ease het­ero­ge­ne­ity within the skin
and across extracutaneous com­part­ments
MF and SS are unique in their strik­ing het­ero­ge­ne­ity of clin­i­cal,
bio­log­i­cal, and molec­u­lar fea­tures and the need to indi­vid­u­al­
ize man­age­ment for an opti­mal out­come. This het­ero­ge­ne­ity is
fur­ther com­pli­cated by var­i­able clin­i­cal behav­ior across dis­ease
com­part­ments (eg, skin, LN, blood). A mean­ing­ful clin­i­cal out­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/303/1852064/303kim.pdf by guest on 13 December 2021
come of treat­ment in MF/SS con­sists of a dura­ble global (com­
pos­ite of all­com­part­ments) objec­tive response and improve­ment
of qual­ity of life (QoL) mea­sures,4-7 and achiev­ing these goals
requires indi­vid­u­al­i­za­tion of ther­apy. Traditionally, the clin­i­cal
stage has been a pri­mary con­sid­er­ation in treat­ment selec­tion;
how­ever, addi­tional guid­ance is often needed given the het­ero­
ge­ne­ity within stage group­ings. The extent/bur­den of dis­ease
or sta­tus of other established prog­nos­tic fac­tors (eg, LCT) in the
skin or other dis­ease com­part­ments fur­ther con­trib­utes to treat­
ment selec­tion.8
In early-stage MF with patch and/or plaque skin dis­ ease,
skin-directed treat­ment is pre­ferred, and improve­ment of QoL
is usu­ally the pri­mary focus of treat­ment. Primary skin-directed
treat­ments include top­i­cal prep­a­ra­tions of ste­roids, mechlore­
thamine, ret­i­noids, phototherapy (eg, nar­row­band ultra­vi­o­let
B), and TSEBT.8 TSEBT is very effec­tive at low doses (eg, 12 Gy)
Tumor-type, thicker skin disease,
large-cell transformation, LN disease
Brentuximab vedotin
Pralatrexate
and most appro­pri­ate in patients who have wide­spread thicker
or folliculotropic (hair fol­li­cle involve­ment) skin dis­ease.9 If skin-
directed treat­ment options fail to man­age a patient’s dis­ease,
then esca­lat­ing to sys­temic agents is appro­pri­ate.
Patients with advanced-stage MF and SS have vary­ing clin­i­
cal pre­sen­ta­tions rang­ing from tumor-type skin dis­ease to gen­
er­al­ized erythroderma in which the type and extent of their skin
and extracutaneous dis­ease (LN and/or blood dis­ease) drive the
ther­ a­
pies that best address a patient’s over­ all dis­
ease. Many
stan­dard treat­ment options exist for advanced-stage patients,
includ­ing sys­temic bio­log­i­cal and targeted ther­a­pies or tra­di­
tional cyto­toxic che­mo­ther­apy.8 The ulti­mate selec­tion of the pri­
mary ther­apy will be based on the assess­ment of the skin and the
sever­ity of any extracutaneous dis­ease and on aligning ther­a­pies
that address the most symp­tom­atic and threat­en­ing dis­ease.
Systemic treat­ments often exhibit dif­fer­en­tial effi­cacy across
skin, blood, and nodal com­part­ments (Table 1 and Figure 1). The
response inter­pre­ta­tion can be fur­ther com­pli­cated by the vary­
ing response cri­te­ria and clin­i­cal end points used in ear­lier clin­
i­cal tri­als in CTCL; how­ever, more recent tri­als have uti­lized the
inter­na­tional con­sen­sus cri­te­ria that have helped com­pare results
across stud­ies.4 There are fre­quent mixed responses in the skin or
across dif­fer­ent com­part­ments with dis­ease, and ther­a­pies that
reduce skin dis­ease may not be effec­tive in clear­ing the blood
or LN dis­ease, and vice versa. For exam­ple, in a patient with MF
and wide­spread ulcerating skin tumors (±LCT), we would select
treat­ments with known reli­able activ­ity in the skin com­part­ment
(eg, TSEBT, BV).9-12 However, in a patient with SS and high-bur­
den Sezary dis­ease, we would opt for ther­a­pies that are effi­cient
at reduc­ing the blood dis­ease (eg, MOGA, romidepsin).13,14 Fur­
thermore, some ther­a­pies do not have established, mean­ing­ful
activ­ity in LCT or bulky dis­ease, whereas some ther­a­pies have
shown activ­ity across all­com­part­ments, includ­ing LCT dis­ease.
Traditional com­bi­na­tion che­mo­ther­apy reg­i­mens in sys­temic
TCLs are often reserved for patients with refrac­tory, advanced
dis­ease or when allo­ge­neic HSCT is planned for con­sol­i­da­tion.8
Lastly, the poten­tial toxicities and tol­er­a­bil­ity of chronic ther­apy
must be con­sid­ered in pri­or­i­tiz­ing treat­ments and per­son­al­iz­ing
man­age­ment.
Importance of mul­ti­dis­ci­plin­ary care and man­age­ment
of QoL issues: essen­tial role of sup­port­ive ther­apy
Ideally, the man­age­ment of patients with CTCL involves a mul­ti­
dis­ci­plin­ary approach. It is help­ful to estab­lish col­lab­o­ra­tive col­
leagues in hema­tol­ogy/med­i­cal oncol­ogy, radi­a­tion oncol­ogy,
pathol­ogy, and der­ma­tol­ogy who are inter­ested in cuta­ne­ous
lym­phoma. To address the exten­sive skin dis­ease, spe­cial­ized
radi­a­tion meth­ods such as TSEBT serve as an impor­tant ther­a­
peu­tic alter­na­tive; how­ever, access to cen­ters with TSEBT can
be an issue. Low-dose TSEBT can be com­bined with treat­ments
that have great effi­cacy in the blood and/or LN com­part­ments
to improve the over­all com­pos­ite response and response dura­
tion.15,16 Patients may have exten­sive skin wounds or atyp­i­cal skin
infec­tions or develop skin lesions that clin­i­cally and/or path­o­
log­i­cally mimic CTCL, includ­ing an inflam­ma­tory skin reac­tion to
lym­phoma ther­apy (eg, MOGA, pembrolizumab). Thus, a close
col­lab­o­ra­tion across spe­cial­ties is essen­tial in pro­vid­ing appro­
pri­ate clin­i­cal care and opti­mal out­come in CTCL.
Intolerable itching is a com­mon QoL in CTCL, espe­cially in SS.
The effec­tive man­age­ment of itching also reduces the risk of sec­
Prakash Singh Shekhawat
ond­ary skin infec­tions as patients often have com­pro­mised skin
bar­rier func­tion and are com­monly col­o­nized with Staphylococ-
cus aureus.17 Management with an anti-staph sup­port­ive reg­i­men
has been shown to improve their itching, clear their skin, and
lower infec­tion-asso­ci­ated com­pli­ca­tions.18 The vis­i­ble nature of
CTCL adds a unique neg­a­tive life impact because it con­stantly
reminds patients of their dis­ease and often pro­hib­its patients
from social inter­ac­tion, inten­si­fy­ing the emo­tional bur­den of
their strug­gles.19,20 Thus, in con­trast to other indo­lent lym­pho­mas
such as low-grade fol­lic­u­lar lym­phoma, in which watch­ful wait­
ing can be accept­able, patients with CTCL are often on con­stant
treat­ment, and QoL ele­ments may lead to a deci­sion to change
ther­apy with­out objec­tive dis­ease pro­gres­sion. Optimizing sup­
port­ive care is essen­tial, and clin­i­cal end points such as time to
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/303/1852064/303kim.pdf by guest on 13 December 2021
next sig­nif­i­cant ther­apy may be use­ful mea­sures of over­all clin­i­
cal ben­e­fit.21
Prioritizing treat­ment options in SS
In patients with SS, the leu­ke­mic sub­type of CTCL, clin­i­cal man­
age­ment is often strat­i­fied by the bur­den of the blood Sezary
dis­ease. Primary sys­temic ther­apy for those with a lower Sezary
bur­ den (<5,000 SCs/mm3) includes oral bexarotene,22 meth­
o­trex­ate,23 and extra­cor­po­real photopheresis.24 In those with
a higher Sezary bur­ den, his­ tone deacetylase inhib­
i­
tors (eg,
romidepsin) or newer ther­a­pies, such as MOGA, with more reli­
able activ­ity in the blood com­part­ment would be appro­pri­ate
front­line options (Table 1).14 With MOGA, an anti-CCR4 mono­clo­
nal anti­body in which CCR4 is highly and con­sis­tently expressed
by SCs, objec­tive blood response was observed in 68% (83/122)
of patients with 44% (54/122) achiev­ ing com­plete remis­sion
(CR) in the piv­otal trial, MAVORIC.13 The median time to response
in the blood was 1 month, and the median dura­tion of blood
response was 26 months. The skin response rate was 42% with a
lower nodal response rate of 17% and a global (com­pos­ite of all­
com­part­ments) over­all response rate of 28%. However, patients
with active LCT were excluded in the MOGA piv­otal trial. Given
the impres­sive clin­i­cal activ­ity in the blood com­bined with a
favor­able tol­er­a­bil­ity pro­file, MOGA may be a pre­ferred option
in high-bur­den SS with­out skin LCT or LN dis­ease. Similarly,
the anti-KIR3DL2 anti­body (lacutamab) cur­rently under clin­
i­cal devel­op­ment has been shown to be more effec­tive in
reduc­ing blood dis­ease com­pared to nodal dis­ease (blood
response rate [RR] 56% vs nodal RR 11%).25 Conversely, BV,
which dem­on­strated reli­able effi­cacy in the skin (includ­ing
LCT) and the LN dis­ease,11 does not have established activ­ity
in Sezary dis­ease and SS and was excluded in the piv­otal trial
in CTCL. In SS with exten­sive LN dis­ease (±LCT), romidepsin
may be pri­or­i­tized over MOGA or other options, as romidepsin
has dem­on­strated con­sis­tent clin­i­cal activ­ity across com­part­
ments (Table 1).
Length of ther­apy and main­te­nance strat­e­gies
To address the chronic course of MF/SS, newer sys­temic ther­a­
pies were devel­oped in which clin­i­cal stud­ies allowed patients
to con­tinue treat­ment with­out a fixed dura­tion until intol­er­ance
or dis­ease pro­gres­sion occurred. However, for ther­a­pies that have
dose-cumu­la­tive toxicities such as the periph­eral neu­rop­a­thy
asso­ci­ated with BV treat­ment, we should con­sider a fixed treat­
ment course (<6-8 cycles) and re-treat with BV later when
needed or use reduced doses of BV.10,26.27 MOGA can be con­tin­ued
Personalizing treat­ment selec­tion in MF/SS  |  307
long-term, but in one-third of patients, it may lead to a rash
(median time to rash, 105 days),13,28-30 and the sever­ ity of the
rash may affect the length of treat­ment with MOGA. Despite the
occur­rence of MOGA, patients can be re-treated with MOGA. For
chronic ther­apy in CTCL, treat­ment options or strat­e­gies that
min­im
­ ize immune sup­pres­sion will be pre­ferred given the high-
risk of recur­rent skin infec­tions.
After ini­tial dis­ease reduc­tion, a main­te­nance strat­egy of a sus­
tain­able treat­ment reg­i­men can be con­sid­ered. This may involve
increas­ing the treat­ment inter­val of intra­ve­nous ther­a­pies or transi­
tioning to an oral agent.31 Utilizing a less intense main­te­nance dose
reg­im
­ en of ther­a­pies may lower treat­ment-related toxicities and
pro­vide a lon­ger over­all clin­i­cal ben­e­fit. Integrating skin-directed
ther­apy is not only essen­tial in com­bi­na­tion with sys­temic agents
but nec­es­sary when the sys­temic dis­ease is cleared, and patients
then have pri­mar­ily skin-lim­ited dis­ease.
CLINICAL CASE (CON­T IN­UED)
Given the higher bur­den of blood Sezary dis­ease and lack of
LCT in the skin, the patient was treated with MOGA and after
2 cycles expe­ri­enced >90% skin improve­ment and reduc­tion of
itching. His blood Sezary dis­ease cleared promptly with­out mea­
sur­able abnor­mal cir­cu­lat­ing T cells (B0) by flow cytom­e­try. After
4 months of con­tin­ued great response with con­tin­ued blood CR
(B0), MOGA was discontinued, and the patient con­tin­ued top­i­cal
ther­apy as needed for lim­ited resid­ual skin dis­ease. The patient’s
blood Sezary remained clear and his skin near CR off MOGA, but
after sev­eral months, he expe­ri­enced dis­ease pro­gres­sion in the
skin with rap­idly wors­en­ing, wide­spread thick plaques and nod­
u­lar, tumor-type dis­ease. Skin biopsy showed LCT with 10–40%
of neo­plas­tic T cells expressing CD30 by immu­no­his­to­chem­is­try.
PET/CT now showed more nota­ble LAD, and a core needle sam­
pling showed involve­ment with T-cell lym­phoma (CD30 expres­
sion of 30%). Given the sig­nif­i­cant CD30 expres­sion in the skin
and LN com­part­ments, BV was selected as the next ther­apy.
Management of LCT in MF and SS
LCT in MF or SS is rec­og­nized as one of the stron­gest inde­pen­dent
adverse prog­nos­tic indi­ca­tors,32,33 and thus knowl­edge of LCT is
a key fac­tor in treat­ment selec­tion (Table 1). However, nota­ble
inter-rater (pathol­o­gist to pathol­o­gist) var­i­abil­ity is observed
in the inter­pre­ta­tion of LCT as the cri­te­ria is an arbi­trary cut­off
value (>25% of malig­nant T cells are large cells) by his­to­path­o­
logic eval­u­a­tion.34 Accordingly, a sub­set of patients des­ig­nated
as LCT may have a more indo­lent course; there­fore, appro­pri­ate
clin­ic
­ al judg­ment is essen­tial for opti­mal man­age­ment. In con­
trast to sys­temic lym­pho­mas in which anthracycline-based ther­
apy and autol­o­gous trans­plants are con­sid­ered in the set­ting of
transformed dis­ease, LCT in MF/SS is man­aged dif­fer­ently, with
greater indi­vid­u­al­ized tai­lor­ing for the extent and biol­ogy of the
transformed dis­ease.
CD30 expres­sion is not a require­ment for LCT cri­te­ria, but
the median CD30 expres­sion is higher in the lesional tis­sue of
patients with LCT com­pared to those with­out LCT,35 and thus
BV is often a highly effec­tive treat­ment in MF with LCT.10,12,36 In
308  |  Hematology 2021  |  ASH Education Program
the phase 3 trial of BV in CD30+ CTCL, the pri­mary end point
of an objec­tive response rate last­ing >4 months (ORR4) was
higher in patients with LCT than with­out LCT (65% vs 39%).11,12
Pralatrexate and romidepsin, each as a sin­ gle agent, have
also dem­on­strated clin­i­cal activ­ity in MF or SS with LCT,14,37,38
though the data in these sin­gle agents are not as robust as
in the piv­otal trial of BV. MOGA would not be pri­or­i­tized in
patients with LCT given the lack of established data.13 Older
agents such as lipo­so­mal doxo­ru­bi­cin or gemcitabine have
greater tox­ic­ity pro­files but can be used in MF/SS with LCT in
refrac­tory set­tings or when newer agents are not avail­­able.39,40
A recent clin­ i­
cal trial of pembrolizumab in patients with
relapsed/refrac­tory MF or SS showed prom­is­ing clin­i­cal activ­
ity in patients with LCT and can be con­sid­ered in relapsed or
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/303/1852064/303kim.pdf by guest on 13 December 2021
refrac­tory set­tings.41
Biomarker-guided selec­tion and inte­gra­tion of ther­a­pies
in MF/SS-CTCL
Currently, numer­ous avail­­able and inves­ti­ga­tional ther­a­pies in
MF/SS-CTCL can tar­get the cell sur­face mol­e­cules of malig­nant
T cells, disrupted cellular pathways, and/or the tumor micro­en­vi­
ron­ment (TME), includ­ing the immune milieu (Table 1). Reliable
bio­mark­ers that enrich the patient sub­sets that can ben­e­fit with
spe­ cific ther­ a­pies would be ideal, and such bio­ marker infor­
ma­tion may help select and pri­or­i­tize ther­a­pies among a list of
options.
Of the tar­gets on the malig­nant cell sur­face in CTCL, CD30 has
the most robust data, with a CD30 targeting agent, BV. With the
use of more sen­si­tive tools for CD30 detec­tion (eg, mul­ti­spec­
tral imag­ing), sam­ples with nondetectable CD30 expres­sion by
rou­tine immu­no­his­to­chem­is­try (IHC) and light micros­copy dem­
on­strated cell-sur­face CD30 mol­e­cules.10 In MF/SS, the inter- and
intrapatient (interlesional) CD30 expres­ sion lev­els (by rou­
tine
IHC) can be highly var­i­able in the skin, and BV has dem­on­strated
sig­nif­i­cant clin­i­cal activ­ity in patients with a spec­trum of CD30
expres­sion in 2 inde­pen­dent stud­ies in MF/SS.10,36,42 Patients with
>5% CD30max (max­i­mum CD30 level from mul­ti­ple skin sam­ples)
had more reli­able over­all clin­i­cal response com­pared with those
with CD30max <5%; how­ever, mean­ing­ful activ­ity was observed
across all­CD30 lev­els. Further, in the piv­otal study of BV in CD30+
CTCL (ALCANZA), patients with mul­ti­ple skin biop­sies showed a
nota­ble var­i­abil­ity of CD30 lev­els, rang­ing from nondetectable
to >50% expres­sion.11,12 Thus, BV is a treat­ment option even in
those with low/neg­li­gi­ble lev­els of CD30 expres­sion, espe­cially
in refrac­tory set­tings.
Therapies such as MOGA can tar­get the malig­nant cells as
well as TME, as the tar­get mol­e­cule CCR4 is expressed not only
by the malig­nant T cells but also by the reg­u­la­tory T cells. CCR4 is
a chemokine recep­tor asso­ci­ated with skin traf­fick­ing, but unlike
CD30, the mol­ e­
cule is fre­ quently and con­ sis­
tently expressed
in the lesional skin (median CCR4, 80%; range of 1%-100% by
IHC) and may not be use­ful as a bio­marker in CTCL.13 However,
some patients with MF/SS have neg­li­gi­ble CCR4 expres­sion, and
fur­ther stud­ies will be needed to estab­lish the role of MOGA in
these patients. Further, assess­ment for CCR4 expres­sion is not
rou­tinely avail­­able, so a stan­dard­ized pro­to­col for CCR4 assess­
ment and a bet­ter under­stand­ing of its cor­re­la­tion with clin­i­cal
out­come needs to be established.43 Although both CCR4 and
KIR3DL2 are both com­monly expressed in skin dis­ease with LCT,
nei­ther MOGA nor anti-KIR3DL2 anti­body (lacutamab) has estab­
lished activ­ity data in patients with LCT.25,44
Newer bio­marker plat­forms (eg, action­able NGS pan­els, TME
pro­fil­ing) may help guide treat­ment selec­tion and pri­or­i­tize ther­
a­pies in selected set­tings.45,46 The knowl­edge of Th2 skewing
with dis­ease pro­gres­sion has led to the selec­tion of immune
ther­a­pies that shift the immune pro­file from Th2 to Th1, includ­
ing inter­ fer­
ons, IL-12, and toll-like recep­tor ago­ nists. Agents
known to deplete reg­u­la­tory T cells in the TME, such as MOGA or
CD25-targeting agents,47 and to acti­vate mac­ro­phages (increase
the “eat-me” sig­nal), such as agents targeting the CD47-SIRPα
axis (eg, SIRPαFc, magrolimab),48,49 may be con­sid­ered as sin­
gle agents or part of com­bi­na­tion reg­i­mens. Patients with geno­
mic data that dem­on­strate path­o­genic alter­ation in the T-cell
An NGS study by Ungewickell et al reported that LCT is asso­
ci­
ated with a high muta­ tion bur­den and recur­ rent PLCG1
alter­ations.52 Furthermore, recent stud­ies explor­ing the pro­
grammed cell death 1 (PD-1) and programmed cell death
1 ligand 1 (PD-L1) axis show a pos­si­ble link between PD-L1
struc­tural var­i­ants and LCT,45 supporting the hypoth­e­sis that
patients with LCT may have more TME immune alter­ations that
allow tumor growth. These patients with PD-L1 alter­ation have
expe­ri­enced clin­i­cal ben­e­fit with anti-PD-1 anti­body such as
pembrolizumab.41,53

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/303/1852064/303kim.pdf by guest on 13 December 2021

immune reg­u­la­tory axis may ben­e­fit from immune ther­a­pies that
include check­point inhib­i­tors.41
In patients with refrac­tory or high-risk dis­ease, using NGS
plat­forms to eval­u­ate for addi­tional prog­nos­tic clues or action­
able path­o­genic genetic var­i­ants can help pri­or­i­tize ther­a­pies.
For exam­ple, JAK inhib­i­tors may be con­sid­ered in patients with
relapsed/refrac­tory dis­ease with a path­o­genic muta­tion of
JAK1/3.50 Moreover, we can extrap­o­late geno­mic bio­marker data
from other T-cell lym­pho­mas. For exam­ple, CCR4 gain of func­
tion muta­tions have been asso­ci­ated with supe­rior clin­i­cal out­
comes in adult T-cell leu­ke­mia/lym­phoma patients treated with
MOGA.51 Thus, if such poten­tial path­o­genic var­i­ants of CCR4 are
noted by NGS data, per­haps MOGA can be pri­or­i­tized over other
ther­a­pies in high-bur­den Sezary patients or in refrac­tory dis­ease.
As pre­vi­ously discussed, LCT is a very impor­tant clin­i­cal
and path­o­logic fac­tor to con­sider when selecting ther­apy.
CLINICAL CASE (Con­tin­ued)
The patient was treated with BV, with great clin­i­cal response
in the skin and flat­ten­ing of his thicker, tumor-type skin dis­
ease that showed LCT by pathol­ogy. PET/CT obtained after
3 cycles of BV showed clin­i­cal CR of LN dis­ease. The blood
com­part­ment remained clear of mea­sur­able dis­ease by flow
cytom­e­try. The patient had clin­i­cal fea­tures con­sis­tent with a
high-risk prog­nos­tic pro­file, includ­ing a his­tory of high blood
SC bur­den, LN dis­ease, and LCT.33 At the ini­ti­a­tion of BV, allo­
ge­neic HSCT was discussed as a poten­tial cura­tive ther­apy in
this high-risk set­ting. The patient’s sib­ling was iden­ti­fied as a
10/10 matched donor. The patient was 67 years old at this time
point, and he agreed to move forth with allo­ge­neic HSCT using
TSEBT and a total lym­phoid irra­di­a­tion/antithymocyte glob­u­lin
nonmyeloablative reg­i­men.

Table 2. Allogeneic HSCT in MF/SS-CTCL

Reference N Intervention Efficacy Transplant-related toxicities

Hosing et al 55
47 6% abla­tive ORR  =  n/a, 2-year PFS 40%, 2-year OS 67% 1-year NRM 10%
94% RIC/NMA 4-year PFS 26%, 4-year OS 51% 2-year NRM 17%
Gr 2–4 aGVHD 40%;
cGVHD 28%
Duarte et al56 60 33% abla­tive ORR  =  n/a, 2-year PFS 34%, 2-year OS 54% 1-year NRM 20%
67% RIC/NMA 5-year PFS 32%, 5-year OS 46% 7-year NRM 22%
7-year PFS 30%, 7-year OS 44%
Paralkar et al57 12 17% abla­tive ORR  =  67%, 2-year PFS 23%, 2-year OS 1-year NRM 25%
83% RIC 56%
De Masson et al58 37 32% abla­tive ORR  =  n/a, 2-year PFS 31%, 2-year OS 1-year NRM 18%
68% RIC/NMA 57% 2-year NRM 18%
Gr 2-4 aGVHD 76%;
cGVHD 44%
Lechowicz 129 36% abla­tive ORR  =  n/a, 1-year PFS 31%, 1-year OS 54% 1-year NRM 19%
et al59 64% RIC/NMA 5-year PFS 17%, 5-year OS 32% 5-year NRM 22%
Gr 2-4 aGVHD 41%; cGVHD 43%
Isufi et al60 16 (n  =  23; 16 MF/SS; RIC except 2 haploidentical CR rate  =  56% 100-day NRM 12%
7 G/D TCL) OS 75% (12 of 16 patients) w/median fol­ Gr 2-4 aGVHD 50%;
low-up 5.5 years cGVHD 56%
Weng et al61 35 100% NMA ORR (CR)  =  80% (57%), 2-year PFS 60%, 1-year NRM 3%
2-year OS 68% 2-year NRM 14%
5-year PFS 41%, 3-year OS 62%, Gr 2-4 aGVHD 16%;
5-year OS 56% cGVHD 32%
aGVHD, acute GVHD; cGVHD, chronic GVHD; Gr, grade; NMA, non-myeloablative; RIC, reduced-inten­sity con­di­tion­ing.
Prakash Singh Shekhawat
Personalizing treat­ment selec­tion in MF/SS  |  309
Allogeneic HSCT as a path to estab­lish dura­ble remis­sion
in high-risk MF/SS
Patients with advanced-stage MF and SS have a median sur­
vival of <5 years, are often treated with mul­ti­ple sequen­tial sys­
temic ther­a­pies, and even­tu­ally become refrac­tory to avail­­able
agents.33 High-dose che­mo­ther­apy followed by autol­o­gous
HSCT has not shown a dura­ble ben­e­fit,54 but allo­ge­neic HSCT
that relies on a graft vs lym­phoma effect has led to long-term
remis­sions (Table 2).55-61
There is some con­tro­versy regard­ing which sub­set of patients
should move forth for allo­ge­neic trans­plant in MF/SS and when
in their dis­ease course it is appro­pri­ate. Allogeneic HSCT should
be con­sid­ered for appro­pri­ate patients with stage IIB through IV
dis­ease who are refrac­tory to mul­ti­ple lines of sys­temic ther­
apy, dem­on­strate declin­ing dura­tion of clin­i­cal ben­e­fit, or pre­
s­ent with a very high-risk pro­file. A large inter­na­tional study of
prog­ nos­ tic fac­
tors in advanced-stage MF and SS (retro-CLIPI)
iden­ti­fied 3 risk groups with dis­tinct prog­nos­tic out­comes based
on the num­ber of key adverse prog­nos­tic param­e­ters, includ­ing
extracutaneous dis­ease and LCT.33 Patients in the high-risk group
with a 5-year over­all sur­vival (OS) of 28% were clearly appro­pri­
ate for allo­ge­neic HSCT. A sig­nif­i­cant pro­por­tion of patients in
the inter­me­di­ate-risk (5-year OS, 44%) group may be con­sid­ered
for allo­ge­neic HSCT. The 5-year OS with allo­ge­neic HSCT in the
published stud­ies ranged from 32% to 56%,55-62 likely supporting
an improved sur­vival ben­e­fit pro­vided that most patients who
received allo­ge­neic HSCT included those with inter­me­di­ate- to
high-risk retro-CLIPI pro­files. The opti­mal tim­ing for allo­ge­neic
HSCT is when the dis­ease is well con­trolled (CR or near CR of all­
dis­ease com­part­ments) and before the dis­ease has progressed
to a state in which the chance of response or sur­vival with allo­
ge­neic HSCT is low. The skin com­part­ment is a fre­quent site of
posttransplant dis­ease relapse, and thus TSEBT is often an impor­
tant and nec­es­sary com­po­nent of the pre­pa­ra­tory strat­egy in
MF/SS. The median age of patients with advanced MF or SS at
diag­no­sis is 60 to 65 years, so an allo­ge­neic reg­i­men with fewer
trans­plant-related toxicities has also been explored.61 For each
patient, a deci­sion to trans­plant requires care­ful con­sid­er­ation of
the long-term ben­e­fits and sig­nif­i­cant risks, includ­ing the poten­
tial for graft-ver­sus-host dis­ease (GVHD) and dis­ease relapse after
trans­plant.
In patients with aggres­sive MF or SS, there remains an unmet
need for addi­tional safe and effec­tive ther­a­pies that can bridge
patients to allo­ge­neic HSCT. And espe­cially for those who are
not eli­gi­ble for allo­ge­neic HSCT or have no opti­mal donors, new
and novel immune ther­a­pies and approaches are needed. As
these inno­va­tive ther­a­pies are made avail­­able, it is essen­tial to
inte­grate them to man­age patients suc­cess­fully and to con­tinue
explor­ing bio­mark­ers that can help us guide our selec­tion or pri­
or­i­ti­za­tion of treat­ment options.
Conclusions and future direc­tions
The clin­i­cal, path­o­log­i­cal, and bio­log­i­cal fea­tures in MF and SS
are highly het­ero­ge­neous, with sig­nif­i­cant inter- and intra-­patient
var­i­abil­ity that requires a per­son­al­ized approach to man­age­
ment. Available and inves­ti­ga­tive ther­a­pies often have dif­fer­en­
tial activ­ity across dis­ease com­part­ments, and thus in addi­tion
to clin­i­cal stage and the pres­ence of LCT, com­part­men­tal activ­
ity is an impor­tant ele­ment that guides treat­ment selec­tion. In
appro­pri­ate high-risk set­tings, patients should be con­sid­ered for
310  |  Hematology 2021  |  ASH Education Program
allo­ge­neic HSCT. The unique QoL issues bring unique chal­lenges
in man­ag­ing patients with CTCL.
Ultimately, in this highly het­ero­ge­neous dis­ease group we hope
to estab­lish bio­mark­ers and strat­e­gies that enrich (bet­ter align sub­
sets) for dura­ble clin­i­cal response and address resis­tance/escape
mech­a­nisms while min­i­miz­ing toxicities. More per­son­al­ized and
tol­er­ab
­ le immunotherapies may be a path­way to achieve a more
dura­ ble or sustained clin­ i­
cal response. Advances in bio­ marker-
guided inves­ti­ga­tions in CTCL will enable us to apply evi­dence-
based approaches in pri­ or­
i­
tiz­
ing ther­
a­
pies and help develop
com­bi­na­tion strat­e­gies that may lead to the clin­i­cal syn­ergy of
com­bined agents.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/303/1852064/303kim.pdf by guest on 13 December 2021
Conflict-of-inter­est dis­clo­sure
Youn H. Kim: Advisory board, Research funding; Innate: Research
funding; Corvus: Research funding; Galderma: Advisory board,
Research funding; CRISPR therapeutics: Research funding; Secura
Bio: Advisory board; Trillium: Research funding.
Off-label drug use
Youn H. Kim: none.
Correspondence
Youn H. Kim, Stanford University School of Medicine, 780 Welch
Rd, CJ220D, C. J. Huang Bldg, Palo Alto, CA 94304; e-mail:
younkim@stanford​­.edu.
References
1. Willemze R, Cerroni L, Kempf W, et al. The 2018 update of the WHO-EORTC
clas­si­fi­ca­tion for pri­mary cuta­ne­ous lym­pho­mas. Blood. 2019;133(16):1703-
1714.
2. Kim EJ, Hess S, Richardson SK, et al. Immunopathogenesis and ther­apy of
cuta­ne­ous T cell lym­phoma. J Clin Invest. 2005;115(4):798-812.
3. Bobrowicz M, Fassnacht C, Ignatova D, Chang YT, Dimitriou F, Guenova
E. Pathogenesis and ther­apy of pri­mary cuta­ne­ous T-cell lym­phoma: Col­
legium Internationale Allergologicum (CIA) update 2020. Int Arch Allergy
Immunol. 2020;181(10):733-745.
4. Olsen EA, Whittaker S, Kim YH, et al. Clinical end points and response cri­
te­ria in myco­sis fungoides and Sezary syn­drome: a con­sen­sus state­ment
of the International Society for Cutaneous Lymphomas, the United States
Cutaneous Lymphoma Consortium, and the Cutaneous Lymphoma Task
Force of the Euro­pean Organisation for Research and Treatment of Cancer.
J Clin Oncol. 2011;29(18):2598-2607.
5. Kim YH, Demierre MF, Kim EJ, et al. Clinically mean­ing­ful reduc­tion in pru­
ri­tus in patients with cuta­ne­ous T-cell lym­phoma treated with romidepsin.
Leuk Lymphoma. 2013;54(2):284-289.
6. Dummer R, Prince HM, Whittaker S, et al. Patient-reported qual­ity of life in
patients with relapsed/refrac­tory cuta­ne­ous T-cell lym­phoma: results from
the randomised phase III ALCANZA study. Eur J Cancer. 2020;133(July):120-
130.
7. Porcu P, Hudgens S, Horwitz S, et al. Quality of life effect of the anti-CCR4
mono­clo­nal anti­body mogamulizumab ver­sus vorinostat in patients with
cuta­ne­ous T-cell lym­phoma. Clin Lymphoma Myeloma Leuk. 2021;21(2):97-
105.
8. Mehta-Shah N, Horwitz SM, Ansell S, et al. NCCN guide­lines insights: pri­
mary cuta­ne­ous lym­pho­mas, ver­sion 2.2020. J Natl Compr Canc Netw.
2020;18(5):522-536.
9. Hoppe RT, Harrison C, Tavallaee M, et al. Low-dose total skin elec­tron beam
ther­apy as an effec­tive modal­ity to reduce dis­ease bur­den in patients with
myco­sis fungoides: results of a pooled anal­y­sis from 3 phase-II clin­ic ­ al tri­
als. J Am Acad Dermatol. 2015;72(2):286-292.
10. Kim YH, Tavallaee M, Sundram U, et al. Phase II inves­ti­ga­tor-ini­ti­ated study
of brentuximab vedotin in myco­sis fungoides and Sézary syn­drome with
var­i­able CD30 expres­sion level: a multi-insti­tu­tion col­lab­o­ra­tive pro­ject.
J Clin Oncol. 2015;33(32):3750-3758.
11. Prince HM, Kim YH, Horwitz SM, et al; ALCANZA Study Group. Brentuximab
vedotin or phy­si­cian’s choice in CD30-pos­i­tive cuta­ne­ous T-cell lym­phoma
(ALCANZA): an inter­na­tional, open-label, randomised, phase 3, multicentre
trial. Lancet. 2017;390(10094):555-566.
12. Kim YH, Prince HM, Whittaker S, et al. Response to brentuximab vedotin
ver­sus phy­si­cian’s choice by CD30 expres­sion and large cell trans­for­ma­tion
sta­tus in patients with myco­sis fungoides: an ALCANZA sub-anal­y­sis. Eur J
Cancer. 2021;148(May):411-421.
13. Kim YH, Bagot M, Pinter-Brown L, et al; MAVORIC Investigators. Mogamuli­
zumab ver­sus vorinostat in pre­vi­ously treated cuta­ne­ous T-cell lym­phoma
(MAVORIC): an inter­na­tional, open-label, randomised, con­trolled phase 3
trial. Lancet Oncol. 2018;19(9):1192-1204.
14. Whittaker SJ, Demierre MF, Kim EJ, et al. Final results from a mul­ti­cen­ter,
inter­na­tional, piv­otal study of romidepsin in refrac­tory cuta­ne­ous T-cell
lym­phoma. J Clin Oncol. 2010;28(29):4485-4491.
15. Jothishankar B, Almazan T, Kim Y, et al. Romidepsin and total skin elec­tron
beam ther­apy in advanced stage myco­sis fungoides and Sézary syn­drome.
Br J Haematol. 2019;186(2):377-379.
16. Fong S, Hong EK, Khodadoust MS, et al. Low-dose total skin elec­tron beam
ther­apy com­bined with mogamulizumab for refrac­tory myco­sis fungoides
and Sézary syn­drome. Adv Radiat Oncol. 2021;6(3):100629.
17. Ahern K, Gilmore ES, Poligone B. Pruritus in cuta­ne­ous T-cell lym­phoma: a
review. J Am Acad Dermatol. 2012;67(4):760-768.
18. Lindahl LM, Willerslev-Olsen A, Gjerdrum LMR, et  al. Antibiotics inhibit
tumor and dis­ ease activ­ ity in cuta­ne­ ous T-cell lym­phoma. Blood.
2019;134(13):1072-1083.
19. Demierre MF, Kim YH, Zackheim HS. Prognosis, clin­i­cal out­comes and qual­
ity of life issues in cuta­ne­ous T-cell lym­phoma. Hematol Oncol Clin North
Am. 2003;17(6):1485-1507.
20. Jonak C, Porkert S, Oerlemans S, et  al. Health-related qual­ity of life in
cuta­ne­ous lym­pho­mas: past, pres­ent and future. Acta Derm Venereol.
2019;99(7):640-646.
21. Campbell BA, Scarisbrick JJ, Kim YH, Wilcox RA, McCormack C, Prince HM.
Time to next treat­ment as a mean­ing­ful end­point for tri­als of pri­mary cuta­
ne­ous lym­phoma. Cancers (Basel). 2020;12(8):2311.
22. Duvic M, Hymes K, Heald P, et al; Bexarotene Worldwide Study Group. Bex­
arotene is effec­tive and safe for treat­ment of refrac­tory advanced-stage
cuta­ne­ous T-cell lym­phoma: mul­ti­na­tional phase II-III trial results. J Clin
Oncol. 2001;19(9):2456-2471.
23. Zackheim HS, Epstein EH Jr. Low-dose meth­o­trex­ate for the Sézary syn­
drome. J Am Acad Dermatol. 1989;21(suppl 4, pt 1):757-762.
24. Quaglino P, Knobler R, Fierro MT, et al. Extracorporeal photopheresis for
the treat­ment of erythrodermic cuta­ne­ous T-cell lym­phoma: a sin­gle cen­
ter clin­i­cal expe­ri­ence with long-term fol­low-up data and a brief over­view
of the lit­er­a­ture. Int J Dermatol. 2013;52(11):1308-1318.
25. Bagot M, Porcu P, Marie-Cardine A, et  al. IPH4102, a first-in-class anti-
KIR3DL2 mono­clo­nal anti­body, in patients with relapsed or refrac­tory cuta­
ne­ous T-cell lym­phoma: an inter­na­tional, first-in-human, open-label, phase
1 trial. Lancet Oncol. 2019;20(8):1160-1170.
26. Corbin ZA, Nguyen-Lin A, Li S, et al. Characterization of the periph­eral neu­
rop­a­thy asso­ci­ated with brentuximab vedotin treat­ment of myco­sis fun­
goides and Sézary syn­drome. J Neurooncol. 2017;132(3):439-446.
27. Khan N, Noor S, Geller S, et al. A phase II trial of reduced dose brentuximab
vedotin for cuta­ne­ous T-cell lym­phoma. Paper presented at: International
Conference on Malignant Lymphoma; June 2021; Lugano, Switzerland.
28. Bagot M, Dalle S, Sokol L, et  al. Long-term clin­i­cal ben­e­fit to anti-CCR4
mogamulizumab: results from the phase 3 MAVORIC study in pre­vi­ously
treated cuta­ne­ous T-cell lym­phoma. Blood. 2018;132(suppl 1):2901.
29. Musiek A, Whittaker S, Horwitz SM, et al. Characterization and out­comes in
patients with mogamulizumab-asso­ci­ated skin reac­tions in the MAVORIC
trial. Blood. 2020;136(suppl 1):23-24.
30. Hirotsu KE, Neal TM, Khodadoust MS, et  al. Clinical char­ac­ter­iza­tion of
mogamulizumab-asso­ci­ated rash dur­ing treat­ment of myco­sis fungoides
or Sézary syn­drome. JAMA Dermatol. 2021;157(6):700-707.
31. Martinez-Escala ME, Kuzel TM, Kaplan JB, et  al. Durable responses with
main­te­nance dose-spar­ing reg­i­mens of romidepsin in cuta­ne­ous T-cell
lym­phoma. JAMA Oncol. 2016;2(6):790-793.
32. Arulogun SO, Prince HM, Ng J, et al. Long-term out­comes of patients with
advanced-stage cuta­ne­ous T-cell lym­phoma and large cell trans­for­ma­tion.
Blood. 2008;112(8):3082-3087.
33. Scarisbrick JJ, Prince HM, Vermeer MH, et al. Cutaneous Lymphoma Interna­
tional Consortium study of out­come in advanced stages of myco­sis fungoi­
des and Sézary syn­drome: effect of spe­cific prog­nos­tic mark­ers on sur­vival
Prakash Singh Shekhawat
and devel­op­ment of a prog­nos­tic model. J Clin Oncol. 2015;33(32):3766-
3773.
34. Gru AA, Kim J, Pulitzer M, et al. The use of cen­tral pathol­ogy review with
dig­i­tal slide scan­ning in advanced-stage myco­sis fungoides and Sézary
syn­drome: a multi-insti­tu­tional and inter­na­tional pathol­ogy study. Am J
Surg Pathol. 2018;42(6):726-734.
35. Edinger JT, Clark BZ, Pucevich BE, Geskin LJ, Swerdlow SH. CD30 expres­
sion and pro­lif­er­a­tive frac­tion in nontransformed myco­sis fungoides. Am J
Surg Pathol. 2009;33(12):1860-1868.
36. Duvic M, Tetzlaff MT, Gangar P, Clos AL, Sui D, Talpur R. Results of a phase
II trial of brentuximab vedotin for CD30+ cuta­ne­ous T-cell lym­phoma and
lymphomatoid papulosis. J Clin Oncol. 2015;33(32):3759-3765.
37. Horwitz SM, Kim YH, Foss F, et al. Identification of an active, well-tol­er­ated
dose of pralatrexate in patients with relapsed or refrac­tory cuta­ne­ous
T-cell lym­phoma. Blood. 2012;119(18):4115-4122.
38. O’Connor OA, Pro B, Pinter-Brown L, et  al. Pralatrexate in patients with
relapsed or refrac­tory periph­eral T-cell lym­phoma: results from the piv­otal
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/303/1852064/303kim.pdf by guest on 13 December 2021
PROPEL study. J Clin Oncol. 2011;29(9):1182-1189.
39. Dummer R, Quaglino P, Becker JC, et al. Prospective inter­na­tional mul­ti­cen­
ter phase II trial of intra­ve­nous pegylated lipo­so­mal doxo­ru­bi­cin monoche­
motherapy in patients with stage IIB, IVA, or IVB advanced myco­ sis
fungoides: final results from EORTC 21012. J Clin Oncol. 2012;30(33):4091-
4097.
40. Duvic M, Talpur R, Wen S, Kurzrock R, David CL, Apisarnthanarax N. Phase
II eval­u­a­tion of gemcitabine monotherapy for cuta­ne­ous T-cell lym­phoma.
Clin Lymphoma Myeloma. 2006;7(1):51-58.
41. Khodadoust MS, Rook AH, Porcu P, et al. Pembrolizumab in relapsed and
refrac­tory myco­sis fungoides and Sézary syn­drome: a mul­ti­cen­ter phase II
study. J Clin Oncol. 2020;38(1):20-28.
42. Rahbar Z, Li S, Tavallaee M, Novoa RA, Kim J, Kim YH. Variability in the
expres­sion of immu­no­his­to­chem­i­cal mark­ers: impli­ca­tions for bio­marker
inter­pre­ta­tion in cuta­ne­ous T-cell lym­phoma. J Invest Dermatol. 2018;138(5):
1204-1206.
43. Fujii K, Sakamoto Y, Masaki A, et al. Immunohistochemistry for CCR4 C-ter­mi­nus
pre­dicts CCR4 muta­tions and mogamulizumab effi­cacy in adult T-cell leu­ke­mia/
lym­phoma. J Pathol Clin Res. 2021;7(1):52-60.
44. Duvic M, Pinter-Brown LC, Foss FM, et  al. Phase ½ study of mogamuli­
zumab, a defucosylated anti-CCR4 anti­body, in pre­vi­ously treated patients
with cuta­ne­ous T-cell lym­phoma. Blood. 2015;125(12):1883-1889.
45. Beygi S, Fernandez-Pol S, Duran G, et al. Pembrolizumab in myco­sis fungoi­
des with PD-L1 struc­tural var­i­ants. Blood Adv. 2021;5(3):771-774.
46. Houlahan KE, Curtis C. A tumor “per­son­al­ity” test to guide ther­a­peu­tic deci­
sion mak­ing. Cancer Cell. 2021;39(6):747-749.
47. Kawai H, Ando K, Maruyama D, et al. Phase II study of E7777 in Jap­a­nese
patients with relapsed/refrac­ tory periph­eral and cuta­ne­ous T-cell lym­
phoma. Cancer Sci. 2021;112(6):2426-2435.
48. Folkes AS, Feng M, Zain JM, Abdulla F, Rosen ST, Querfeld C. Targeting CD47
as a can­cer ther­a­peu­tic strat­egy: the cuta­ne­ous T-cell lym­phoma expe­ri­
ence. Curr Opin Oncol. 2018;30(5):332-337.
49. Johnson LDS, Banerjee S, Kruglov O, et al. Targeting CD47 in Sezary syn­
drome with SIRPalphaFc. Blood Adv. 2019;3(7):1145-1153.
50. Gomez-Arteaga A, Margolskee E, Wei MT, van Besien K, Inghirami G, Hor­
witz S. Combined use of tofacitinib (pan-JAK inhib­i­tor) and ruxolitinib (a
JAK1/2 inhib­ i­
tor) for refrac­ tory T-cell prolymphocytic leu­ ke­mia (T-PLL)
with a JAK3 muta­tion. Leuk Lymphoma. 2019;60(7):1626-1631.
51. Sakamoto Y, Ishida T, Masaki A, et  al. CCR4 muta­tions asso­ci­ated with
supe­rior out­come of adult T-cell leu­ke­mia/lym­phoma under mogamuli­
zumab treat­ment. Blood. 2018;132(7):758-761.
52. Ungewickell A, Bhaduri A, Rios E, et al. Genomic anal­y­sis of myco­sis fun­
goides and Sézary syn­drome identifies recur­rent alter­ations in TNFR2. Nat
Genet. 2015;47(9):1056-1060.
53. Lesokhin AM, Ansell SM, Armand P, et al. Nivolumab in patients with relapsed
or refrac­tory hema­to­logic malig­nancy: pre­lim­i­nary results of a phase
Ib study. J Clin Oncol. 2016;34(23):2698-2704.
54. Wu PA, Kim YH, Lavori PW, Hoppe RT, Stockerl-Goldstein KE. A meta-
anal­y­sis of patients receiv­ing allo­ge­neic or autol­o­gous hema­to­poi­etic
stem cell trans­plant in myco­sis fungoides and Sézary syn­drome. Biol Blood
Marrow Transplant. 2009;15(8):982-990.
55. Hosing C, Bassett R, Dabaja B, et al. Allogeneic stem-cell trans­plan­ta­tion in
patients with cuta­ne­ous lym­phoma: updated results from a sin­gle insti­tu­
tion. Ann Oncol. 2015;26(12):2490-2495.
56. Duarte RF, Boumendil A, Onida F, et al. Long-term out­come of allo­ge­neic
hema­to­poi­etic cell trans­plan­ta­tion for patients with myco­sis fungoides
Personalizing treat­ment selec­tion in MF/SS  |  311
 and Sézary syn­drome: a Euro­pean Society for Blood and Marrow Trans­
plantation Lymphoma Working Party extended anal­ y­
sis. J Clin Oncol.
2014;32(29):3347-3348.
57. Paralkar VR, Nasta SD, Morrissey K, et  al. Allogeneic hema­to­poi­etic SCT
for pri­mary cuta­ne­ous T cell lym­pho­mas. Bone Marrow Transplant.
2012;47(7):940-945.
58. de Masson A, Beylot-Barry M, Bouaziz JD, et al; French Study Group on
Cutaneous Lymphomas and Société Française de Greffe de Moëlle et
Thérapie Cellulaire. Allogeneic stem cell trans­ plan­ta­
tion for advanced
cuta­ne­ous T-cell lym­pho­mas: a study from the French Society of Bone Mar­
row Transplantation and French Study Group on Cutaneous Lymphomas.
Haematologica. 2014;99(3):527-534.
59. Lechowicz MJ, Laz­a­rus HM, Carreras J, et al. Allogeneic hema­to­poi­etic cell
trans­plan­ta­tion for myco­sis fungoides and Sezary syn­drome. Bone Marrow
Transplant. 2014;49(11):1360-1365.
60. Isufi I, Seropian S, Gowda L, et al. Outcomes for allo­ge­neic stem cell trans­
plan­ta­tion in refrac­tory myco­sis fungoides and pri­mary cuta­ne­ous gamma
delta T cell lym­pho­mas. Leuk Lymphoma. 2020;61(12):2955-2961.
61. Weng WK, Arai S, Rezvani A, et al. Nonmyeloablative allo­ge­neic trans­plan­ta­
tion achieves clin­i­cal and molec­u­lar remis­sion in cuta­ne­ous T-cell lym­phoma.
Blood Adv. 2020;4(18):4474-4482.
312  |  Hematology 2021  |  ASH Education Program
62. Mehta-Shah N, Kommalapati A, Teja S, et  al. Successful treat­ment of
mature T-cell lym­phoma with allo­ge­neic stem cell trans­plan­ta­tion: the
larg­est mul­ti­cen­ter ret­ro­spec­tive anal­y­sis. Paper presented at: 62nd
Amer­i­can Society of Hematology Annual Meeting; 5-8 Decem­ber 2020;
vir­tual meet­ing.
63. Olsen EA, Kim YH, Kuzel TM, et al. Phase IIb mul­ti­cen­ter trial of vorinostat
in patients with per­sis­tent, pro­gres­sive, or treat­ment refrac­tory cuta­ne­ous
T-cell lym­phoma. J Clin Oncol. 2007;25(21):3109-3115.
64. Horwitz SM, Koch R, Porcu P, et al. Activity of the PI3K-δ,γ inhib­i­tor duv­
elisib in a phase 1 trial and pre­clin­i­cal mod­els of T-cell lym­phoma. Blood.
2018;131(8):888-898.
65. Querfeld C, Rosen ST, Guitart J, et al. Results of an open-label mul­ti­cen­ter
phase 2 trial of lenalidomide monotherapy in refrac­tory myco­sis fungoides
and Sezary syn­drome. Blood. 2014;123(8):1159-1166.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/303/1852064/303kim.pdf by guest on 13 December 2021
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000263
 INDOLENT LYMPHOMAS

Follicular lymphoma: is there an optimal

way to define risk?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/313/1851762/313casulo.pdf by guest on 13 December 2021

Carla Casulo
University of Rochester, Rochester, NY

Follicular lymphoma (FL) has a long natural history and typically indolent behavior. In the present era, there are a plethora
of prognostic factors combining clinical, biological, and genetic data to determine patient prognosis and help develop
treatment strategies over the course of a patient’s lifetime. The rapid pace of tumor-specific and clinical advances in FL
has created a challenge in the prioritization and implementation of these factors into clinical practice. Developing a com-
prehensive understanding of existing prognostic markers in FL will help select optimal ways of utilization in the clinical
setting and investigate opportunities to define and intervene upon risk at FL diagnosis and disease recurrence.

LEARNING OBJECTIVES
• Describe patient- and tumor-specific factors associated with risk of disease progression, histologic transforma-
tion, and premature death from FL
• Review established prognostic models in FL and their strengths and weaknesses when applied in clinical practice
• Discuss approaches to incorporate the clinical and mutational data into novel prognostic tools at different points
in the patient’s lifetime

Introduction
In American and European adults, follicular lymphoma (FL)
is the second most common non-Hodgkin lymphoma.1
The long natural history and typically indolent behavior
of FL provide opportunities to identify prognostic infor-
mation combining clinical, biological, and genetic data to
determine patient prognosis and help define treatment
strategies patients will need over their lifetime. However,
the rapid pace of scientific discoveries in FL encompass-
ing both tumor-specific and clinical advances has made
it challenging to prioritize the most clinically relevant
and essential prognostic parameters. Importantly, some
of the prognostic information currently available often
yields redundant or conflicting results, thereby limiting
implementation into practice (Figure 1). A comprehensive
overview of all prognostic markers to assess risk in FL is
beyond the scope of this review, as it would be inclusive
of genetic, genomic, clinical, and diagnostic imaging-
based parameters.2-7 However, through a discussion of a
patient case, this review examines some of the prognostic
factors along the FL continuum, assesses optimal ways of
utilization in clinical practice, and explores opportunities
to define risk in the relapsed setting.
CLINICAL CASE
A 52-year-old man presented with painless cervical lymph-
adenopathy noticed while shaving. He has a history of
hypertension and hyperlipidemia. He recently noted being
significantly fatigued but denied weight loss or night
sweats. Laboratory tests revealed hemoglobin of 10.2g/dL
(reference range, male: 13.7-17.5g/dL), lactate dehydroge-
nase (LDH) of 270U/dL (upper limit of normal, 225U/dL),
and an increased β2 microglobulin level (4.2mg/L; upper
limit of normal, 3.0mg/L). No other abnormalities were
noted. An excisional lymph node biopsy specimen showed
grade 1 to 2 FL comprising small mature centrocytes in a
well-preserved follicular pattern that stained positive for
CD10, CD20, and BCL2. Flow cytometry detected a clonal
population of B cells. Fluorescent in situ hybridization
detected t(14;18). Bone marrow biopsy specimen revealed
small paratrabecular lymphoid aggregates encompassing
10% of the bone marrow space. Positron emission tomog-
raphy (PET) showed hypermetabolic lymphadenopathy
in the bilateral cervical chain, right axillae, retroperito-
neum, and right inguinal area with standard uptake value
of 7 to 11. Lymphadenopathy ranged from 3.2 to 4.7 cm.

Prakash Singh Shekhawat

Defining risk in FL | 313
 FLIPI FLIPI2

Variables Variables
Nodal sites Age >60
Elevated LDH Bone marrow involvement
Age >60 Hemoglobin <12 g/dL
Stage III/IV Greatest LN diameter >6 cm
Hemoglobin ≤12 g/dL Serum β2 microglobulin > upper limit of normal

Score Score

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/313/1851762/313casulo.pdf by guest on 13 December 2021

0-1 Low risk 0-1 Low risk
2 Intermediate risk 2 Intermediate risk
3-5 High risk 3-5 High risk

Survival Survival
0-1 5-year OS 91% 0-1 5-year PFS 80%
2 5-year OS 78% 2 5-year PFS 52%
3-5 5-year OS 53% 3-5 5-year PFS 19%

FLEX
PRIMA-PI

Variables Variables
β2 microglobulin >3 mg/L Male sex, SPD in highest quartile,
Bone marrow involvement grade 3A, >2 extranodal sites, ECOG
PS >1, hemoglobin <12 g/dL, β2
microglobulin > normal, peripheral
Score blood natural killer cell count <100/µL,
Low β2 microglobulin <3 mg/L, increased LDH
negative bone marrow
Intermediate β2 microglobulin ≤3 mg/L,
+ bone marrow
High β2 microglobulin >3 mg/L Score
0-2 Low risk
3-9 High risk
Survival
Low 5-year PFS 69%
Survival
Intermediate 5-year PFS 55%
Low 3-year PFS 86%
High 5-year PFS 37%
High 3-year PFS 68%

Figure 1. Clinical prognostic tools and risk scores. ECOG, Eastern Cooperative Oncology Group; LN, lymph node; PS, performance
status.

314  |  Hematology 2021  |  ASH Education Program

The patient ini­ti­ated ther­apy with bendamustine and rituximab
and received 6 cycles.
At pres­ent, there is no agreed-on, uni­form def­i­ni­tion of risk
in FL. However, the sur­vival of FL span­ning over 2 decades and
mul­ti­ple avail­­able novel ther­a­peu­tics with high response rates
and long peri­ods of remis­sion pro­vide a rea­son­able bench­mark
of expected patient out­comes at diag­no­sis. The con­cept of high-
risk FL has emerged from data dem­on­strat­ing increased prob­a­bil­
ity of pre­ma­ture death for patients refrac­tory to alkylator-based
ther­apy or anti-CD20 mono­clo­nal antibodies in the first-line set­
ting, as well as patients with early dis­ease recur­rence fol­low­ing
An over­ view of the cur­ rently avail­­
able clin­
i­
cal and tumor-
based mark­ers asso­ci­ated with high risk in FL reveals a mul­ti­tude
of var­i­ables (Table 1). Based on this patient’s male sex, pres­ence
of >4 nodal sites, high tumor bur­den, ele­vated LDH, advanced
stage, and low hemo­glo­bin, the patient has sev­eral fac­tors asso­
ci­ated with higher prob­a­bil­ity of poor out­come. Accordingly,
the patient’s Follicular Lymphoma International Prognosis Index
(FLIPI) score is 4, high risk, and based on this assess­ment, the
patient’s over­all sur­vival (OS) is esti­mated to be 70% at 10 years
for patients in the rituximab era.9,15 The FLIPI incor­po­rates 5 clin­i­
cal var­i­ables (nodal sites, ele­vated LDH, age >60 years, advanced
stage, and hemo­glo­bin <12 g/dL) into a mul­ti­var­i­ate model to
pre­dict OS. It was constructed to improve sen­si­tiv­ity com­pared

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/313/1851762/313casulo.pdf by guest on 13 December 2021

front­line ther­apy and early his­to­logic trans­for­ma­tion.8-10 For the with the International Prognostic Index for indo­lent lym­phoma
pur­poses of this review, we con­sider patients at increased risk of and based on ret­ro­spec­tive data from patients diag­nosed in the
these adverse out­comes as high risk. 1980s and 1990s treated het­ero­ge­neously at var­i­ous cen­ters.

Table 1. Variables asso­ci­ated with “high-risk FL” (increased risk of death, early recur­rence, trans­for­ma­tion)

Variable Outcome Reference

Clinical/patient based
  Grade 3b vs grades 1-3a Inferior OS Wahlin et al,11 Mercadal et al12
  Increased β2 microglobulin from nor­mal Inferior PFS and OS Press et al,13 Bachy et al14
  Large nodal mass (>7 cm) Inferior PFS Solal-Céligny et al,15 Nooka et al,16 Buske et al17
  Greater than 4 nodal sites Inferior PFS and OS Solal-Céligny et al15
  Bone mar­row involve­ment Inferior PFS and OS Bachy et al14
  Low hemo­glo­bin (under 12 g/dL) Inferior PFS and OS Solal-Céligny et al,15 Nooka et al16
  Advanced stage Inferior PFS and OS Brice et al,18 Buske et al17
  High FL tumor bur­den based on GELF cri­te­ria Inferior PFS and OS Brice et al,18 Buske et al17
  Low lym­pho­cyte to mono­cyte ratio Increased risk of trans­for­ma­tion Gao et al,19 Mohsen et al,20 Mozas et al21
Imaging based
  High base­line SUV max of PET scan Inferior OS and increased risk of early Mir et al,22 Barrington and Meignan23
dis­ease pro­gres­sion (mixed data)
  End of ther­apy PET pos­i­tive (Deauville 4-5) Inferior PFS and OS Trotman et al3
  High total met­a­bolic tumor vol­ume Inferior PFS and OS (mixed data) Barrington and Meignan,23 Skander 2019,50 Meignan et al4
Tumor based
  23-gene GEP sig­na­ture score Inferior PFS Huet et al5
  Genomic alter­ations (gain of chro­mo­some 2p; Inferior PFS Qu et al24
dele­tion 17p, subclonal TP53 muta­tions)
  Low intratumoral immune infil­tra­tion Increased risk of early pro­gres­sion Tobin et al25
  High expres­sion of genes from tumor asso­ci­ Inferior PFS and OS Dave et al,26 Kridel et al27
ated mac­ro­phages
  Increased ctDNA at diag­no­sis and at Inferior PFS and OS Delfau-Larue et al,6 Zohren et al,28 Sarkozy et al29
com­ple­tion of ther­apy
  Select muta­tions in sev­eral genes, includ­ing Enriched in tumors of early progressed Kridel et al30
TP53, SOCS1, B2M, and MYD88 patients
Treatment based
  Disease recur­rence within 24 months of diag­ Inferior OS Casulo et al,31 Jurinovic et al,32 Maurer et al33
no­sis or ther­apy
  Histologic trans­for­ma­tion 18-24 months after Inferior OS Link et al,8 Wagner-Johnson et al,34 Sarkozy et al35
diag­no­sis or ther­apy
  Multiply relapsed dis­ease Inferior PFS Batlevi et al,9 Link et al,36 Rivas-Delgado et al10
  Refractory dis­ease Inferior PFS Batlevi et al,9 Link et al,36 Rivas-Delgado et al10
Prakash Singh Shekhawat
GELF, Groupe d’Etude des Lymphomas Folliculaires; SUV, standard uptake value.

Defining risk in FL  |  315

The FLIPI yields 3 risk clas­si­fi­ca­tions: low (0-1 fac­tor), inter­me­di­
ate (2-3 fac­tors), and high (4-5 fac­tors) with a dis­tri­bu­tion of low,
inter­me­di­ate, and high risk in 36%, 37%, and 27% of patients,
respec­tively. Despite its devel­op­ment in the prerituximab era, the
FLIPI’s prog­nos­tic legit­i­macy has been val­i­dated numer­ous times.16,17
This sur­vival esti­mate may seem dis­pro­por­tion­ately low when
con­sid­er­ing recent stud­ies eval­u­at­ing prog­no­sis and cause of
death in FL. An anal­y­sis by Sarkozy et al51 noted that 10-year
OS was approx­i­ma­tely 75% in patients with newly diag­nosed
FL treated in the rituximab era, but a high-risk FLIPI score did
increase the cumu­ la­tive inci­dence of lym­ phoma-related mor­
tal­ity (inci­dence of lym­phoma-related mor­tal­ity at 10 years was
4.0% for low-risk FLIPI, 10.0% for inter­me­di­ate-risk FLIPI, and 27%
cyclophosphamide, doxorubicin, vincristine, and prednisone or
rituximab, cyclophosphamide, vincristine, and prednisone. More-
over, for patients with FL treated in the GALLLIUM study, bone
mar­row biopsy was not pre­dic­tive of out­come.38 As with the FLIPI
and FLIPI2, the PRIMA-PI model does not inform selec­tion of ther­
apy for this patient or mod­i­fi­ca­tion of treat­ment based on his
risk score.
Application of these mod­els in the
bendamustine/obinutuzumab era
The FLIPI, FLIPI-2, and PRIMA-PI were devel­ oped in patients
treated with alkylators but did not include the more fre­quently
used bendamustine. Although these indi­ces have been val­i­dated

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/313/1851762/313casulo.pdf by guest on 13 December 2021

for high-risk FLIPI, P < .001). The patient’s high-risk FLIPI score and
high tumor bur­den there­fore inform us of a like­li­hood of adverse
out­come and the advan­tages of treat­ment ini­ti­a­tion.
What are other clin­i­cal prog­nos­tic mod­els avail­­able
in the rituximab era?
As rituximab has dras­ ti­
cally improved patient sur­ vival in the
mod­ern era, it is expected and observed that prog­nos­tic indi­ces
have less sen­si­tiv­ity and spec­i­fic­ity for out­comes. The FLIPI2 was
devel­oped using 1093 patients with newly diag­nosed FL in need
of treat­ment, included nearly 70% of patients treated in the ritux-
imab era, and was val­i­dated in an inde­pen­dent cohort of Ital­ian
patients.37 The FLIPI2 included 5 covariates based on their rel­a­
tive fre­quency and included ele­vated β2 microglobulin, lymph
node diam­ e­
ter lon­ ger than 6  cm, bone mar­ row involve­ ment,
low hemo­glo­bin, and age greater than 60 years. This model
was designed to pre­dict pro­gres­sion-free sur­vival (PFS) and, if
applied to this patient, would sim­i­larly yield a high-risk score of
3 and pro­vide a 3-year PFS esti­mate of approx­i­ma­tely 39% in the
rituximab era. The validity of the FLIPI 2 was sim­i­larly con­firmed
exter­nally in a cohort of 1135 patients from the PRIMA study who
received che­mo­ther­apy with or with­out main­te­nance rituximab.
The Primary RItuximab and MAintenance (PRIMA) prog­ nos­
tic index (PRIMA-PI) esti­mated disparities in sur­vival based on
the num­ber of fac­tors pres­ent in patients treated in the PRIMA
study.14 Using only 2 param­e­ters (bone mar­row involve­ment and
increased β2 microglobulin), the pri­mary end point was to pre­
dict PFS. In this case, the patient’s bone mar­row involve­ment
and increased β2 microglobulin would also result in a high-risk
score but pre­dict a dif­fer­ent 5-year PFS of 69% and a 38% risk of
expe­ri­enc­ing early ther­apy fail­ure (Table 2). Of impor­tance, the
PRIMA-PI was only pre­dic­tive in patients treated with rituximab,
ret­ro­
spec­tively in small series, only the Follicular Lymphoma
Evaluation Index (FLEX) model was pre­dic­tive of out­comes in
patients receiv­ing bendamustine and the novel anti-CD20 anti­
body obinutuzumb.39 This model includes 9 var­i­ables: male sex,
sum of prod­ucts (SPD) in highest quar­tile, grade 3A, >2 extran-
odal sites, Eastern Cooperative Oncology Group per­for­mance
sta­tus >1, hemo­glo­bin <12 g/dL, β2 microglobulin higher than
nor­mal, periph­eral blood nat­u­ral killer cell count <100/µL, and
increased LDH. A high score (3-9 fac­tors) was pre­dic­tive of poor
PFS (3-year PFS of 86% for low risk vs 68% for high risk), and sec­
ond­ary end points also showed poor OS and increased risk of
early pro­gres­sion. Given the lack of read­ily avail­­able var­i­ables
tested in this model, as part of a sen­si­tiv­ity anal­y­sis, the authors
also eval­u­ated miss­ing data. They found that the sur­vival pro­file
of patients with miss­ing data appeared sim­i­lar to that of com­
plete cases, with a PFS haz­ard ratio for miss­ing vs com­plete FLEX
data of 0.89 (95% CI, 0.68-1.17; P = .40). In this patient’s case, his
score would be 4 (high risk), based on sex, low hemo­glo­bin, and
increased LDH and β2 microglobulin. Other param­e­ters included
in the FLEX were not avail­­able for eval­ua ­ ­tion. This pre­dicts a
3-year PFS of 86%. The clear dis­ad­van­tage of this prog­nos­tic tool
is the rou­tine lack of sev­eral of the param­e­ters cal­cu­lated. In the
FLEX man­u­script, SPD was used as a sur­ro­gate for tumor bur­den.
During model devel­op­ment, SPD had among the highest haz­ard
ratios for PFS. As SPD requires radio­logic cal­cu­la­tion, bulky dis­
ease may be an alter­na­tive.
Although a Cox pro­por­tional haz­ards anal­y­sis of PFS showed
that FLIPI and PRIMA-PI had a lower prog­ nos­tic value across
treat­ment reg­i­mens than the FLEX, it is not imme­di­ately appar­
ent how this score will become rou­tinely used in clin­i­cal prac­
tice, and fur­ther study is desir­able in real-world set­tings and in
the con­text of non-chemoimmunotherapeutic strat­e­gies such as
lenalidomide and rituximab or rituximab monotherapy.

Table 2. Case patient’s risk assess­ment based on avail­­able
clin­i­cal cal­cu­la­tors/data
Clinical Risk Calculator/
Prognostic Data
FLIPI score high risk
FLIPI2 score high risk
PRIMA-PI high risk
FLEX score high risk
Early relapse
Tumor-based, bio­logic, and genetic var­i­ables
FL cells reside within a rich tumor micro­en­vi­ron­ment (TME) envel-
oped by both non­ma­lig­nant cells and crit­i­cal immune com­po­
nents that pro­vide sym­bi­otic influ­ences on the FL tumor cell itself
Survival Outcomes
and back onto the TME. This con­tri­bu­tion to patient out­come
10-year OS of 70% was first established by Dave et al,26 who dem­on­strated that high
3-year PFS of 39% expres­ sion of genes from FL tumor-asso­ ci­
ated mac­ ro­
phages
was indic­a­tive of poor out­comes. Yet, more con­tem­po­rary stud­ies
5-year PFS of 37%, 38% risk of
hav­ing treat­ment fail­ure within show conflicting results when rituximab is incor­po­rated into treat­
24 months ment reg­i­mens, as dem­on­strated by Kridel et al,27 who showed
3-year PFS of 68%
that high num­bers of CD163+ mac­ro­phages were inde­pen­dent
pre­dic­tors of lon­ger sur­vival in patients receiv­ing anthracycline-
5-year OS 50%
based reg­i­mens. In other work, the mag­ni­tude of intratumoral

316  |  Hematology 2021  |  ASH Education Program

immune infil­tra­tion was found to be an impor­tantcom­po­nent of
poor out­comes, in par­tic­u­lar early treat­ment fail­ure. A study from
the Princess Alexandra Hospital eval­ u­
ated whether there was
an immune infil­tra­tion pro­file in FL asso­ci­ated with early dis­ease
pro­gres­sion.25
Tobin et al25 performed targeted gene sequenc­ ing using
NanoString tech­nol­ogy from par­af­fin-embed­ded tis­sue and mul­
ti­spec­tral immu­no­flu­o­res­cence on a tis­sue microarray that was
applied to 2 groups: a dis­cov­ery cohort of 132 patients from the
Princess Alexandra Hospital with early and advanced stage FL
who received either che­mo­ther­apy or obser­va­tion and 2 inde­
pen­dent val­i­da­tion cohorts of 198 patients with advanced stage
dis­ease treated with R-CHOP and R-CVP from the Ger­man Low
Grade Lymphoma Study Group and the Brit­ish Colum­bia Can-
cer Agency. They also performed T-cell rep­er­toire anal­y­sis, flow
cytom­e­try, immu­no­flu­o­res­cence, and next-gen­er­a­tion sequenc­
ing. They showed that immune mol­e­cules from T cells, mac­ro­
phages, or immune check­points were clus­tered into either high
expres­ sion or low expres­ sion. Low lev­ els of immune mark­ ers
iden­ti­fied patients enriched for early pro­gres­sion, and PDL2 was
the marker with highest accu­racy to dis­tin­guish groups of low or
high immune infil­tra­tion. This was val­i­dated in group of uni­formly
treated patients from the Brit­ish Colum­bia Cancer Agency and the
Ger­man Low Grade Lymphoma Study Group, which showed that
tumors with a low immune infil­tra­tion had higher early dis­ease-
related (progression of disease within 24 months of diagnosis/
treatment of FL) events. Immune infil­tra­tion by intratumoral T cells
was quan­ti­fied in con­junc­tion with fluorodeoxyglucose-PET in a
study by Nath et al.40 Lymph nodes from patients with high total
met­a­bolic tumor vol­ume had an inverse asso­ci­a­tion with num­bers
of intratumoral T cells but increased malig­nant B-cell infil­tra­tion.
This sug­gests dis­crep­ant pre­dic­tive util­ity of FDG-PET in patients
with FL depending on the extent of T-cell deple­tion caused by the
type of chemoimmunotherapy that is admin­is­tered.
Gene expres­sion pro­fil­ing (GEP) stud­ies by Huet et al5 defined
a gene sig­ na­
ture cor­ re­lat­
ing with adverse PFS in FL. They
defined a model based on the expres­sion of 23 genes char­ac­
ter­is­tic of B-cell centroblasts that was prog­nos­tic inde­pen­dent
of the FLIPI score. In a mul­ti­var­i­ate anal­y­sis, those iden­ti­fied as
high risk by the 23-gene pre­dic­tor had a 5-year PFS of 26% com­
pared with 73% in patients with a low-risk sig­na­ture. This was
con­firmed in 3 inde­pen­dent val­i­da­tion cohorts. The per­for­mance
of the 23-gene sig­na­ture was eval­u­ated in GALLIUM, in which
no prog­nos­tic effect was observed for any of the gene sig­na­
tures based on anti­body treat­ment arm. However, che­mo­ther­
apy choice did have a sig­nif­i­cant inter­ac­tion between PFS and
the 23-gene sig­na­ture, in which patients with a high-risk score
receiv­ing CHOP/CVP had unfa­vor­able out­comes com­pared with
those treated with bendamustine, suggesting che­ mo­ther­ apy
depen­dence of the high-risk sig­na­ture score.41
At pres­ent, the assess­ment of these immune mark­ers and the
capa­bil­ity to per­form GEP stud­ies are not stan­dard­ized at diag­
no­sis, and some tech­niques are not read­ily avail­­able and hence
remain part of a research approach.
Combining clin­i­cal and genetic mod­els
The com­bi­na­tion of key epi­ge­netic muta­tions with a high-risk
FLIPI score and patient per­for­mance sta­tus was accom­plished by
Pastore et al42 to yield the m7-FLIPI. Although prior stud­ies empha­
sized the rel­e­vance of sin­gle gene alter­ations, m7-FLIPI included a
Prakash Singh Shekhawat
com­pre­hen­sive col­lec­tion of recur­rent gene muta­tions as well as
clin­i­cal risk fac­tors in patients with FL. In the mul­ti­var­i­ate model,
high-risk FLIPI score and Eastern Cooperative Oncology Group
>1 car­ried sig­nif­i­cant weight, but EZH2 and ARID1A muta­tions
iden­ti­fied a favor­able risk pop­u­la­tion. The m7-FLIPI improved risk
strat­i­fi­ca­tion by reclassifying patients pre­vi­ously clas­si­fied as
high-risk FLIPI into the low-risk m7-FLIPI group, which was a bet­
ter reflec­tion of their actual FL risk. However in a val­i­da­tion study
of patients treated with bendamustine or obinutuzumab in the
GALLIUM study, although the m7-FLIPI was prog­nos­tic in patients
receiv­ing CHOP/CVP and still outperformed the FLIPI, it lost its
prog­ nos­ tic sig­
nif­i­
cance in patients receiv­ ing bendamustine or
obinutuzumab.43 It is hypoth­e­sized that this inter­ac­tion may be
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/313/1851762/313casulo.pdf by guest on 13 December 2021
driven by EZH2 muta­tions, but this remains to be fur­ther val­i­
dated.41 In nonchemotherapy-containing reg­i­mens, m7-FLIPI was
also not prognostic of out­comes, rais­ing the notion that che­mo­
ther­apy depen­dence influ­ences the high-risk sig­na­ture score.44
To eval­u­ate the spe­cific end point of POD24, Jurinovic et al32
reexamined the m7-FLIPI. Using the same patient cohorts and
muta­tional data, a POD24-spe­cific prog­nos­tic model was devel­
oped, described as the POD24-PI. It included FLIPI but was
refined to include muta­ tional sta­tus of only 3 genes (EP300,
FOXO1, and EZH2). Compared with the m7-FLIPI, the new
POD24-PI model had greater sen­si­tiv­ity to pre­dict POD24 but
was not supe­rior to other approaches due to lower spec­i­fic­ity.
The Bio-clinical FLIPI (Bio-FLIPI) sim­i­larly inte­grated the FLIPI with
genes impli­cated in the TME and iden­ti­fied sev­eral asso­ci­ated
with early treat­ment fail­ure. Nonetheless, only lack of intrafollic-
ular CD4 expres­sion was pre­dic­tive of treat­ment fail­ure.45 As in
other exam­ples, assess­ment of muta­tional sta­tus is not stan­dard­
ized and remains part of a research approach.
Dynamic risk assess­ment through­out the patient’s his­tory
CLINICAL CASE (Con­tin­ued)
The patient com­pleted 6 cycles of treat­ment with­ out com­
pli­ca­tions. His PET response at the end of ther­apy was com­
pat­i­ble with a com­plete met­a­bolic response, an achieve­ment
asso­ci­ated with favor­able out­comes.3,46,47 As part of an ongo­ing
clin­i­cal trial, eval­u­a­tion of min­i­mal resid­ual dis­ease (MRD) was
performed at diag­no­sis, end of ther­apy, and every 6 months for
2 years to assess cor­re­la­tion with dis­ease pro­gres­sion.
Several stud­ies have reported the asso­ci­a­tion of MRD assess­
ment as prog­nos­tic of out­come in FL. This can be achieved by
assess­ment of cir­cu­lat­ing pre­treat­ment cell-free DNA frag­ments
from apo­pto­tic tumor cells (cir­cu­lat­ing tumor DNA [ctDNA]) and by
assess­ment of BCL2 immunoglobulin heavy chain lev­els quan­ti­fied
by poly­mer­ase chain reac­tion (PCR). Identification of the t(14;18)
trans­lo­ca­tion can be qual­i­ta­tively and ­quan­ti­ta­tively ­mea­sured in
the bone mar­row and periph­eral blood but is l­im­ited to patients
har­bor­ing the t(14;18) and can also be found in the blood of healthy
patients. Other tech­nol­ogy using next-gen­er­a­tion sequenc­ing
meth­ods can detect a large spec­trum of genetic alter­ations, such
as the Cancer Personalized Profiling by Deep Sequencing method
and quan­ti­ta­tion of ctDNA encoding the clonalyl rearranged V(D)
J Ig recep­tor gene sequence of FL cells.29,48
Defining risk in FL  |  317
 A pro­spec­tive Ger­
man study ana­ lyzed BCL2 IGH lev­ els
by PCR in pre- and posttreatment periph­ eral blood in 173
patients with FL, find­ing an adverse impact of high pre­treat­
ment lev­els on sub­se­quent PFS.28 As expected, high lev­els of
MRD at the con­clu­sion of ther­apy also had poor out­comes.
The achieve­ment of neg­a­tive or low ctDNA or MRD sim­i­larly
enhances prog­nos­tic infor­ma­tion on patient out­come. Studies
in other lym­pho­mas have dem­on­strated that dynamic ctDNA
lev­els dur­ing and after ther­apy not only can pre­dict clin­i­cal
out­comes but may also synergize with other prog­nos­tic fac­
tors such as total met­a­bolic tumor vol­ume.6 The best use of
these molec­u­lar assess­ments of response will be dur­ing or
fol­low­ing treat­ment to facil­i­tate early inter­ven­tion for recur­
rence, assess changes in muta­tional pro­file, and be used to
guide next ther­apy.
CLINICAL CASE (Con­t in­u ed)
Approximately 1 year fol­low­ing ther­apy, the patient expe­ri­enced
dis­ease relapse with high tumor bur­den, low-grade FL. The
adverse prog­nos­tic impact of early dis­ease recur­rence (within
24 months of diag­no­sis) on sur­vival has been well established.31,49
Few prog­nos­tic indi­ces have been tested for use to pre­dict out­
comes at relapse out­side of FLIPI, and this is an unmet need. He
joined a ran­dom­ized phase 2 clin­ic ­ al study ran­dom­iz­ing patients
to obinutuzumab with umbralisib, lenalidomide, or CHOP and
was ran­dom­ized to the obinutuzumab with umbralisib arm
(NCT03269669). He expe­ ri­
enced a par­ tial met­
a­bolic response
followed by rapid pro­gres­sion of dis­ease. A repeat biopsy con­
firmed low-grade FL. He began ther­apy with axicabtagene cilo-
leucel on a clin­i­cal study (NCT03105336), now approved by the US
Food and Drug Administration for relapsed FL in the United States
(https:​­/​­/www​­.fda​­.gov​­/drugs​­/resources​­-information​­-approved​
­-drugs​­/fda​­-grants​­-accelerated​­-approval​­-axicabtagene​­-ciloleucel​
­-relapsed​­-or​­-refractory​­-follicular​­-lymphoma).
Conclusions
A mul­ti­tude of prog­nos­tic var­i­ables are at our dis­posal to under­
stand and pre­dict sur­vival for FL and to iden­tify patients at high
risk of poor out­come. However, sev­eral of these tools remain
inac­ces­ si­
ble in daily prac­tice and have not been ade­ quately
tested to select ther­apy. Unfortunately, none of the risk mod­els
described here have been val­i­dated as a tool to select or adapt
ther­apy for patients with FL. Although under­stand­ing prog­no­sis
is impor­tant, ulti­mately the goal of any of these tests should be
to guide ther­apy. Studies of risk-adapted ther­apy based on MRD
are ongo­ing in dif­fuse large B-cell lym­phoma and begin­ning in
FL (Luminari et al52), but this approach remains inves­ti­ga­tional.
Importantly, assess­ment of risk for the refrac­tory patient or at
relapse is an area of ongo­ing inves­ti­ga­tion and may con­trib­ute
to a more pre­cise abil­ity to select the opti­mal treat­ment par­a­
digm. The ideal way to merge existing clin­i­cal and bio­logic data
in a sophis­ti­cated and accu­rate model for high-risk FL remains a
work in prog­ress.
Conflict-of-inter­est dis­clo­sure
Carla Casulo: Research funding from Gilead, Verastem, Genen-
tech, and BMS.
318  |  Hematology 2021  |  ASH Education Program
Off-label drug use
Carla Casulo: nothing to disclose.
Correspondence
Carla Casulo, University of Rochester, 601 Elmwood Ave, Roches-
ter, NY 14642; e-mail: carla_casulo@urmc​­.rochester​­.edu.
References
1. Teras LR, DeSantis CE, Cerhan JR, Morton LM, Jemal A, Flowers CR. 2016 US
lym­phoid malig­nancy sta­tis­tics by World Health Organization sub­types.
CA Cancer J Clin. 2016;66(6):443-459.
2. Cottereau AS, Versari A, Luminari S, et al. Prognostic model for high-tumor-
bur­den fol­lic­u­lar lym­phoma inte­grat­ing base­line and end-induc­tion PET: a
LYSA/FIL study. Blood. 2018;131(22):2449-2453.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/313/1851762/313casulo.pdf by guest on 13 December 2021
3. Trotman J, Luminari S, Boussetta S, et al. Prognostic value of PET-CT after
first-line ther­apy in patients with fol­lic­u­lar lym­phoma: a pooled anal­y­sis
of cen­ tral scan review in three multicentre stud­ ies. Lancet Haematol.
2014;1(1):e17-e27.
4. Meignan M, Cottereau AS, Versari A, et al. Baseline met­a­bolic tumor vol­ume
pre­dicts out­come in high-tumor-bur­den fol­lic­u­lar lym­phoma: a pooled
anal­y­sis of three mul­ti­cen­ter stud­ies. J Clin Oncol. 2016;34(30):3618-3626.
5. Huet S, Tesson B, Jais J-P, et al. A gene-expres­sion pro­fil­ing score for pre­
dic­tion of out­come in patients with fol­lic­u­lar lym­phoma: a ret­ro­spec­tive
train­ing and val­i­da­tion anal­y­sis in three inter­na­tional cohorts. Lancet
Oncol. 2018;19(4):549-561.
6. Delfau-Larue M-H, van der Gucht A, Dupuis J, et al. Total met­a­bolic tumor
vol­ume, cir­cu­lat­ing tumor cells, cell-free DNA: dis­tinct prog­nos­tic value in
fol­lic­u­lar lym­phoma. Blood Adv. 2018;2(7):807-816.
7. Barrington S TJ, Sahin D, et al. Baseline PET-derived met­a­bolic tumor vol­
ume met­rics did not pre­dict out­comes in fol­lic­u­lar lym­phoma patients
treated with first-line immuno che­mo­ther­apy and anti­body main­te­nance
in the phase III GALLIUM study. Blood. 2018;132(suppl 1):2882.
8. Link BK, Maurer MJ, Nowakowski GS, et al. Rates and out­comes of fol­lic­ul­ar
lym­phoma trans­for­ma­tion in the immunochemotherapy era: a report from
the University of Iowa/MayoClinic Specialized Program of Research Excel-
lence Molecular Epidemiology Resource. J Clin Oncol. 2013;31(26):3272-
3278.
9. Batlevi CL, Sha F, Alperovich A, et al. Follicular lym­phoma in the mod­ern
era: sur­ vival, treat­
ment out­ comes, and iden­ ti­
fi­
ca­tion of high-risk sub­
groups. Blood Cancer J. 2020;10(7):74.
10. Rivas-Delgado A, Magnano L, Moreno-Velázquez M, et al. Response dura­
tion and sur­vival shorten after each relapse in patients with fol­lic­u­lar lym­
phoma treated in the rituximab era. Br J Haematol. 2019;184(5):753-759.
11. Wahlin BE, Yri OE, Kimby E, et al. Clinical sig­nif­i­cance of the WHO grades of
fol­lic­u­lar lym­phoma in a pop­ul­a­tion-based cohort of 505 patients with long
fol­low-up times. Br J Haematol. 2012;156(2):225-233.
12. Mercadal S, Sancho JM, Climent F, et  al. Long-term out­come com­par­ing
his­to­log­i­cal grades of fol­lic­u­lar lym­phoma patients treated with immu-
nochemotherapy as first-line ther­apy: a ret­ro­spec­tive anal­y­sis from two
insti­tu­tions. Eur J Haematol. 2020;104(3):198-206.
13. Press OW, Unger JM, Rimsza LM, et  al. A com­par­a­tive anal­y­sis of prog­
nos­tic fac­tor mod­els for fol­lic­ul­ar lym­phoma based on a phase III trial of
CHOP-rituximab ver­sus CHOP +131iodine—tositumomab. Clin Cancer Res.
2013;19(23):6624-6632.
14. Bachy E, Maurer MJ, Habermann TM, et al. A sim­pli­fied scor­ing sys­tem in
de novo fol­lic­ul­ar lym­phoma treated ini­tially with immunochemotherapy.
Blood. 2018;132(1):49-58.
15. Solal-Céligny P, Roy P, Colombat P, et al. Follicular lym­phoma inter­na­tional
prog­nos­tic index. Blood. 2004;104(5):1258-1265.
16. Nooka AK, Nabhan C, Zhou X, et al. Examination of the fol­lic­u­lar lym­phoma
inter­na­tional prog­nos­tic index (FLIPI) in the National LymphoCare study
(NLCS): a pro­spec­tive US patient cohort treated pre­dom­i­nantly in com­mu­
nity prac­tices. Ann Oncol. 2013;24(2):441-448.
17. Buske C, Hoster E, Dreyling M, Hasford J, Unterhalt M, Hiddemann W.
The Follicular Lymphoma International Prognostic Index (FLIPI) sep­a­rates
high-risk from inter­me­di­ate- or low-risk patients with advanced-stage fol­
lic­ul­ar lym­phoma treated front-line with rituximab and the com­bi­na­tion
of cyclo­phos­pha­mide, doxo­ru­bi­cin, vin­cris­tine, and pred­ni­sone (R-CHOP)
with respect to treat­ment out­come. Blood. 2006;108(5):1504-1508.
18. Brice P, Bastion Y, Lepage E, et al. Comparison in low-tumor-bur­den fol­
lic­u­lar lym­pho­mas between an ini­tial no-treat­ment pol­icy, prednimustine,
 or inter­feron alfa: a ran­dom­ized study from the Groupe d’Etude des Lym-
phomes Folliculaires. Groupe d’Etude des Lymphomes de l’Adulte. Clinical
trial ran­dom­ized con­trolled trial. J Clin Oncol. 1997;15(3):1110-1117.
19. Gao F, Hu J, Zhang J, Xu Y. Prognostic value of periph­eral blood lym­pho­
cyte/mono­cyte ratio in lym­phoma. J Cancer. 2021;12(12):3407-3417.
20. Mohsen A, Taalab M, Abousamra N, Mabed M. Prognostic sig­nif­i­cance of
abso­lute lym­pho­cyte count, abso­lute mono­cyte count, and abso­lute lym­
pho­cyte count to abso­lute mono­cyte count ratio in fol­lic­u­lar non-Hodgkin
lym­phoma. Clin Lymphoma Myeloma Leuk. 2020;20(9):e606-e615.
21. Mozas P, Rivero A, Rivas-Delgado A, et  al. A low lym­pho­cyte-to-mono­
cyte ratio is an inde­pen­dent pre­dic­tor of poorer sur­vival and higher risk
of his­to­log­i­cal trans­for­ma­tion in fol­lic­u­lar lym­phoma. Leuk Lymphoma.
2021;62(1):104-111.
22. Mir F, Barrington SF, Brown H, et al. Baseline SUVmax did not pre­dict his­
to­log­i­cal trans­for­ma­tion in fol­lic­u­lar lym­phoma in the phase 3 GALLIUM
study. Blood. 2020;135(15):1214-1218.
23. Barrington SF, Meignan M. Time to pre­pare for risk adap­ta­tion in lym­phoma
by stan­dard­iz­ing mea­sure­ment of met­a­bolic tumor bur­den. J Nucl Med.
2019;60(8):1096-1102.
24. Qu X, Li H, Braziel RM, et al. Genomic alter­ations impor­tant for the prog­
no­sis in patients with fol­lic­u­lar lym­phoma treated in SWOG study S0016.
Blood. 2019;133(1):81-93.
25. Tobin JWD, Keane C, Gunawardana J, et al. Progression of dis­ease within
24 months in fol­lic­u­lar lym­phoma is asso­ci­ated with reduced intratumoral
immune infil­tra­tion. J Clin Oncol. 2019;37(34):3300-3309.
26. Dave SS, Wright G, Tan B, et al. Prediction of sur­vival in fol­lic­u­lar lym­phoma
based on molec­u­lar fea­tures of tumor-infil­trat­ing immune cells. N Engl J
Med. 2004;351(21):2159-2169.
27. Kridel R, Xerri L, Gelas-Dore B, et  al. The prog­nos­tic impact of CD163-
pos­i­tive mac­ro­phages in fol­lic­u­lar lym­phoma: a study from the BC can­
cer agency and the lym­ phoma study asso­ ci­
a­
tion. Clin Cancer Res.
2015;21(15):3428-3435.
28. Zohren F, Bruns I, Pechtel S, et  al. Prognostic value of cir­cu­lat­ing Bcl-2/
IgH lev­els in patients with fol­lic­u­lar lym­phoma receiv­ing first-line immu-
nochemotherapy. Blood. 2015;126(12):1407-1414.
29. Sarkozy C, Huet S, Carlton VE, et al. The prog­nos­tic value of clonal het­ero­
ge­ne­ity and quan­ti­ta­tive assess­ment of plasma cir­cu­lat­ing clonal IG-VDJ
sequences at diag­no­sis in patients with fol­lic­u­lar lym­phoma. Oncotarget.
2017;8(5):8765-8774.
30. Kridel R, Chan FC, Mottok A, et al. Histological trans­for­ma­tion and pro­gres­
sion in fol­lic­u­lar lym­phoma: a clonal evo­lu­tion study. PLoS Med. 2016;13(12):
e1002197.
31. Casulo C, Byrtek M, Dawson KL, et al. Early relapse of fol­lic­u­lar lym­phoma
after rituximab plus cyclo­phos­pha­mide, doxo­ru­bi­cin, vin­cris­tine, and pred­
ni­sone defi­nes patients at high risk for death: an anal­y­sis from the national
lymphocare study. J Clin Oncol. 2015;33(23):2516-2522.
32. Jurinovic V, Kridel R, Staiger AM, et al. Clinicogenetic risk mod­els pre­dict
early pro­gres­sion of fol­lic­u­lar lym­phoma after first-line immunochemother-
apy. Blood. 2016;128(8):1112-1120.
33. Maurer MJ, Ghesquieres H, Jais JP, et al. Event-free sur­vival at 24 months is
a robust end point for dis­ease-related out­come in dif­fuse large B-cell lym­
phoma treated with immunochemotherapy. J Clin Oncol. 2014;32(10):1066-
1073.
34. Wagner-Johnston ND, Link BK, Taylor M, Bartlett NL. Risk fac­tors for early
trans­for­ma­tion of fol­lic­u­lar lym­phoma (FL): report from the National Lym-
phoCare Study (NLCS). Blood. 2009;114(22):2698.
35. Sarkozy C, Trneny M, Xerri L, et al. Risk fac­tors and out­comes for patients with
fol­lic­u­lar lym­phoma who had his­to­logic trans­for­ma­tion after response to
first-line immunochemotherapy in the PRIMA trial. J Clin Oncol. 2016;34(22):
2575-2582.
36. Link BK, Day BM, Zhou X, et al. Second-line and sub­se­quent ther­apy and
out­ comes for fol­ lic­
u­lar lym­
phoma in the United States: data from the
obser­va­tional National LymphoCare Study. Br J Haematol. 2019;184(4):660-
663.
37. Federico M, Bellei M, Marcheselli L, et  al. Follicular Lymphoma Interna-
tional Prognostic Index 2: a new prog­nos­tic index for fol­lic­u­lar lym­phoma
Prakash Singh Shekhawat
devel­oped by the inter­na­tional fol­lic­u­lar lym­phoma prog­nos­tic fac­tor pro­ject.
J Clin Oncol. 2009;27(27):4555-4562.
38. Rutherford SC, Herold M, Hiddemann W, et  al. Impact of bone mar­row
biopsy on response assess­ ment in immunochemotherapy-treated lym­
phoma patients in GALLIUM and GOYA. Blood Adv. 2020;4(8):1589-1593.
39. Mir F, Mattiello F, Grigg A, et  al. Follicular Lymphoma Evaluation Index
(FLEX): a new clin­i­cal prog­nos­tic model that is supe­rior to existing risk
scores for predicting pro­gres­sion-free sur­vival and early treat­ment fail­ure
after front­ line immunochemotherapy. Am J Hematol. 2020;95(12):1503-
1510.
40. Nath K, Law S-C, Sabdia M-B, et  al. Intratumoral T cells have a dif­fer­en­
tial impact on FDG-PET param­ e­
ters in fol­ lic­
u­lar lym­
phoma. Blood Adv.
2021;5(12):2644-2649.
41. Bolen CR, Mattiello F, Herold M, et  al. Treatment depen­dence of prog­
nos­tic gene expres­sion sig­na­tures in de novo fol­lic­u­lar lym­phoma. Blood.
2021;137(19):2704-2707.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/313/1851762/313casulo.pdf by guest on 13 December 2021
42. Pastore A, Jurinovic V, Kridel R, et al. Integration of gene muta­tions in risk
prog­nos­ti­ca­tion for patients receiv­ing first-line immunochemotherapy for
fol­lic­u­lar lym­phoma: a ret­ro­spec­tive anal­y­sis of a pro­spec­tive clin­i­cal trial
and val­i­da­tion in a pop­u­la­tion-based reg­is­try. Lancet Oncol. 2015;16(9):1111-
1122.
43. Jurinovi V, Passerini V, Oestergaard M, et al. Evaluation of the m7-FLIPI in
patients with fol­lic­u­lar lym­phoma treated within the gal­lium trial: EZH2
muta­tion sta­tus may be a pre­dic­tive marker for dif­fer­en­tial effi­cacy of che­
mo­ther­apy. Blood. 2019;134(suppl 1):122.
44. Lockmer S, Ren W, Brodtkorb M, et al. M7-FLIPI is not prog­nos­tic in fol­lic­
u­lar lym­phoma patients with first-line rituximab chemo-free ther­apy. Br J
Haematol. 2020;188(2):259-267.
45. Mondello P, Fama A, Larson MC, et al. Intrafollicular CD4+ T-cells as an inde­
pen­dent pre­dic­tor of early clin­i­cal fail­ure in newly diag­nosed fol­lic­u­lar lym­
phoma. Blood. 2019;134(suppl 1):121.
46. Trotman J, Barrington SF, Belada D, et al; PET inves­ti­ga­tors from the GAL-
LIUM study. Prognostic value of end-of-induc­tion PET response after first-
line immunochemotherapy for fol­lic­u­lar lym­phoma (GALLIUM): sec­ond­ary
anal­y­sis of a randomised, phase 3 trial. Lancet Oncol. 2018;19(11):1530-1542.
47. Pott C, Hoster E, Kehden B, et al. Minimal resid­ual dis­ease response at end
of induc­tion and dur­ing main­te­nance cor­re­lates with updated out­come in
the phase III GALLIUM study of obinutuzumab- or rituximab-based immu-
nochemotherapy in pre­vi­ously untreated fol­lic­u­lar lym­phoma patients.
Blood. 2018;132(suppl 1):396.
48. Scherer F, Kurtz DM, Newman AM, et al. Distinct bio­log­i­cal sub­types and pat­
terns of genome evo­lu­tion in lym­phoma revealed by cir­cu­lat­ing tumor DNA.
Sci Transl Med. 2016;8(364):364ra155.
49. Maurer MJ, Ghesquières H, Link BK, et  al. Diagnosis-to-treat­ment inter­
val is an impor­tant clin­i­cal fac­tor in newly diag­nosed dif­fuse large B-cell
lym­phoma and has impli­ca­tion for bias in clin­i­cal tri­als. J Clin Oncol. 2018;
36(16):1603-1610.
50. Skander J, Fredrickson J, Coimbra A, Carano RAD, Christoffer T, El-Galaly C,
Knapp A, Nielsen TG, Sahin D, Bengtsson T, de Crespigny, A. A fully auto-
mated measurement of total metabolic tumor burden in diffuse large B-cell
lymphoma and follicular lymphoma. Blood. 2019;34 (suppl 1):4666.
51. Sarkozy C, Maurer MJ, Link BK, Ghesquieres H, Nicolas E, Thompson CA,
Traverse-Glehen A, Feldman AL, Allmer C, Slager SL, Ansell SM, Habermann
TM, Bachy E, Cerhan JR, Salles G. Cause of death in follicular lymphoma in
the first decade of the rituximab era: a pooled analysis of French and US
cohorts. J Clin Oncol. 2019 Jan 10;37(2):144-152.
52. Luminar S, Manni M, Galimberti S, et al. Response-adapted postinduction
strategy in patients with advanced-stage follicular lymphoma: The FOLL12
Study. Journal of Clinical Oncology. 2021 (Oct. 28).
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000264
Defining risk in FL  |  319
 INDOLENT LYMPHOMAS

Does MRD have a role in the management

of iNHL?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021

Ilaria Del Giudice, Irene Della Starza, and Robin Foà
Hematology, Department of Translational and Precision Medicine, Sapienza University, Rome, Italy

Among indolent non-Hodgkin lymphomas (iNHLs), the analysis of measurable/minimal residual disease (MRD) has been
extensively applied to follicular lymphoma (FL). Treatment combinations have deeply changed over the years, as well as
the techniques to measure MRD, which is currently evaluated only in the setting of clinical trials. Here, we discuss the evi-
dence on the role of molecular monitoring in the management of FL. Mature data support the quantification of molecular
tumor burden at diagnosis as a tool to stratify patients in risk categories and of MRD evaluation at the end of treatment
to predict progression-free survival and overall survival. Moreover, MRD deserves further studies as a tool to refine the
clinical/metabolic response and to modulate treatment intensity/duration. Patients with a higher relapse probability can
be identified, but the relevance of continuous molecular follow-up should be clarified by kinetic models of MRD analysis.
Being the BCL2/heavy chain immunoglobulin gene hybrid rearrangement detectable in about 50% to 60% of advanced FL
and in 30% of positron emission tomography/computed tomography–staged localized FL, technical advancements such
as next-generation sequencing/target locus amplification may allow broadening the FL population carrying a molecular
marker. Droplet digital polymerase chain reaction can better quantify MRD at low levels, and novel sources of DNA, such
as cell-free DNA, may represent a noninvasive tool to monitor MRD. Finally, MRD in other iNHLs, such as lymphoplasmacytic
lymphoma/Waldenström macroglobulinemia and marginal zone lymphoma, is beginning to be explored.

LEARNING OBJECTIVES
• Describe the accumulating data on MRD monitoring in patients with advanced and localized FL in the immuno-
chemotherapy and chemo-free era
• Understand how MRD might be used in the clinical management of patients with FL
• Discuss how new technologies could overcome the current limitations in order to widely apply MRD to patients
with FL and to other indolent NHLs

Introduction
Indolent non-Hodgkin lymphomas (iNHLs) often spread to
the bone marrow (BM) and/or peripheral blood (PB) and
can be monitored through the identification at diagno-
sis of a disease marker to be followed during treatment.1
Minimal/measurable residual disease (MRD) analysis has
been extensively applied to follicular lymphoma (FL), which
represents the main focus of this review.2,3 MRD in other
iNHLs is starting to be explored, and we briefly comment
on this at the end of the article.
Follicular lymphoma
FL is the second most frequent non-Hodgkin lymphoma.
Despite the improvement in outcome, mainly due to the
combination of anti-CD20 monoclonal antibodies with che-
motherapy, most patients relapse, and the disease remains
uncurable.4 The clinical course of FL is heterogeneous, with
some patients not requiring treatment for years, others expe-
riencing long-lasting remissions after first-line treatment,
and ~20% rapidly relapsing within 24 months from treatment
initiation.5 Transformation into an aggressive lymphoma,
whose frequency has decreased with the use of rituximab
(R), strongly impairs patients’ survival.6 The current prog-
nostic scores (follicular lymphoma international prognostic
index [FLIPI], FLIPI-2, PRIMA-prognostic index [PRIMA-PI],
follicular lymphoma evaluation index [FLEX]), based on clin-
ical parameters, fail to predict the clinical course of individ-
ual patients.7,8 Knowledge of FL’s biological heterogeneity,
which relies on the complex interactions between nonma-
lignant immune/stromal components and tumor cells, has
not identified clinically applicable predictive biomarkers.9,10
The molecular hallmark of FL, the t(14;18)(q32;q21) trans-
location, resulting in the hybrid BCL2/heavy chain immu-
noglobulin gene (IGH) rearrangement, is the first necessary

320 | Hematology 2021 | ASH Education Program

but not suf­fi­cient step in lymphomagenesis.11 The BCL2/IGH gene
can be used to sup­port FL diag­no­sis and to eval­u­ate treat­ment
response in terms of MRD. The first evi­dence on the value of MRD
in FL goes back to 1991.12 Thereafter, treat­ment com­bi­na­tions
have deeply changed, as well as the tech­niques to mea­sure MRD,
which is cur­rently eval­u­ated only in the set­ting of clin­i­cal tri­als.7
Here, we dis­cuss the cur­rent evi­dence on the role of MRD in
the man­age­ment of FL, starting from 2 clin­i­cal cases.
CLINICAL CASES
Patient 1. A 49-year-old man with FL grade 2, Ann Arbor stage IV,
high-risk FLIPI, high-tumor bur­den and major breakpoint region
(MBR)+ in PB/BM was treated in a clin­i­cal trial from July 2017 with
R-bendamustine, followed by 3 courses of 4-weekly doses of R
main­te­nance, com­pleted in August 2019. At the end of induc­tion
(EOI), a pos­i­tron emis­sion tomog­ra­phy (PET)−/MRD+ response
was documented. MRD remained pos­i­tive through­out main­te­
nance and at the end of treat­ment. In March 2021, 19 months after
main­te­nance com­ple­tion, he presented with an abdom­i­nal bulky
dis­ease recur­rence. The new biopsy spec­i­men showed FL grade
3A; the BM biopsy spec­i­men showed no lym­phoma infil­tra­tion.
Patient 2. Based on the biopsy spec­i­men of a 3-cm ingui­nal
lymph node, in May 2016, an asymp­tom­atic 63-year-old woman
was diag­nosed with FL grade 1 to 2, Ann Arbor stage I, MBR+ in
PB/BM. She received involved-site radi­a­tion (24 Gy), followed
by ofatumumab con­sol­i­da­tion for MRD+ response, as per clin­
i­cal trial. She became MRD−, which persisted for 1.5 years at a
6-month MRD mon­i­tor­ing. In August 2018, she returned MRD+ in
the BM, shortly followed by a clin­i­cal relapse.
Is molec­u­lar tumor bur­den at diag­no­sis use­ful
to strat­ify ­patients in risk categories?
The tumor bur­den molec­u­lar quan­ti­fi­ca­tion at diag­no­sis pre­dicts
pro­gres­sion-free sur­vival (PFS) in patients with FL, in both ad-
vanced and local­ized stages (Table 1). Rambaldi et al13 select-
ed patients with FL for the pres­ence of BCL2/IGH pos­i­tiv­ity in
the BM and treated them with cyclo­phos­pha­mide, doxo­ru­bi­cin,
vin­cris­tine, and pred­ni­sone (CHOP), followed by 4-weekly R in
MRD+ cases, and divided them into low (43%), inter­ me­ di­
ate
(34%), and high risk (23%) according to BM BCL2/IGH real-time
quan­ti­ta­tive (RQ )–poly­mer­ase chain reac­tion (PCR) lev­els at
diag­no­sis. High BCL2/IGH+ lev­els were sig­nif­i­cantly asso­ci­ated
with a reduced event-free sur­vival (EFS) and were an inde­pen­
dent pre­dic­tor of poor clin­i­cal and molec­u­lar response in mul­ti­
var­i­ate anal­y­sis (MVA). In con­trast, PB BCL2/IGH lev­els at diag­no­
sis were not pre­dic­tive of EFS.
In large first-line tri­ als employing R-based chemoimmuno-
therapy for advanced FL, BCL2/IGH was found in 51% to 66% of
enrolled patients.17-19,22 Conversely, in the unpub­lished MRD results
from the Gallium trial, in which obinutuzumab (G)–based che­mo­
ther­apy plus G-main­te­nance resulted supe­rior to R-che­mo­ther­apy
plus R-main­te­nance, a clonal marker was detected in 88% of 1101
patients, higher than in other stud­ies, because both BCL2/IGH
and IGH rearrangements were screened by con­sen­sus PCR.20,21
RQ-PCR strat­i­fied patients with low, inter­me­di­ate, and high
BM tumor bur­den at diag­no­sis with a sig­nif­i­cantly dif­fer­ent PFS,
both in elderly patients treated within the ML17638 trial (short
Prakash Singh Shekhawat
chemoimmunotherapy followed by a con­sol­i­da­tion with R plus
a ran­dom­ized main­te­nance) and in the FOLL05 trial, ran­domly
assigning patients to R-CHOP, R-fludarabine, and mitoxantrone
or R-cyclo­phos­pha­mide, vin­cris­tine, and pred­ni­sone (CVP).17,18 In
the for­mer, the molec­u­lar tumor bur­den was cor­re­lated to BM
inva­sion but not with FLIPI.17 In the lat­ter, high BCL2/IGH lev­
els at enroll­ment, documented in patients with high FLIPI and
FLIPI2 scores, were sig­nif­i­cantly asso­ci­ated with a lower over­all
response rate and a higher relapse rate and retained a neg­a­tive
impact on 3-year PFS besides high FLIPI and lack of com­plete
response (CR), inde­pen­dently from the ran­dom­i­za­tion arm.18
Likewise, in the NHL1-2003 trial, com­par­ing R-CHOP to R-benda-
mustine, high pre­treat­ment PB BLC2/IGH lev­els were asso­ci­ated
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021
with BM involve­ment, stage IV, high tumor bur­den, and increased
β2-microglobulin, being an inde­pen­dent prog­nos­tic marker for
PFS and fur­ther strat­i­fy­ing patients with inter­me­di­ate/high FLIPI.19
In the Relevance trial, com­ par­
ing the che­ mo­ ther­
apy-free
reg­i­men lenalidomide  +  R (R2) vs R  +  che­mo­ther­apy followed by
main­te­nance, molec­u­lar tumor bur­den at diag­no­sis quan­ti­fied
by drop­let dig­i­tal PCR (ddPCR) was sig­nif­i­cantly asso­ci­ated with
the MRD sta­tus at week 24 (EOI).22
In early stage FL, despite a neg­a­tive BM biopsy, BCL2/IGH+
cells spread­ing from the orig­i­nal lymph node could be detected
in the PB/BM at diag­no­sis in more than 50% of patients.24,26 The
84-month PFS was 90.9% in patients with unde­tect­able/low lev­
els (<1   ×   10−5) of cir­cu­lat­ing BCL2/IGH+ cells quan­ti­fied by ddPCR
at diag­no­sis vs 38% in those with higher lev­els (P = .015).24
Final com­ment. Both in advanced and local­ized FL, quan­ti­
fi­ca­tion of molec­u­lar tumor bur­den at base­line by RQ-PCR or
ddPCR pre­dicts PFS, being an inde­pen­dent prog­nos­tic marker.
Circulating BCL2/IGH lev­els, which may sim­ply reflect not only
tumor bur­ den but also the enhanced lym­ phoma cell migra­
tion and inva­sive­ness, could help to refine our capac­ity to risk-
strat­ify patients.
Is MRD use­ful to refine the clin­i­cal response?
In the pre-R era, sev­eral tri­als showed that autol­o­gous stem cell
trans­plant induced higher rates of molec­u­lar remis­sion (60%-70%)
than stan­dard anthracycline-based che­mo­ther­apy (30%-50%).2,3
In the R era, the addi­tion of R induced a con­ver­sion to a MRD−
sta­tus in 70% of patients MRD+ after CHOP and in 70% to 84% af-
ter R-fludarabine, mitoxantrone, and dexa­ meth­a­
sone (FND).13,14,17
The com­bi­na­tion of che­mo­ther­apy with R ­front­line showed an
­increased MRD− response rate at EOI com­pared with che­mo­ther­
apy alone, with a pref­er­en­tial MRD clear­ance in the PB (Figure 1,
Table 1).13-15,17-22,24,26-28 In the Gallium trial, the BM MRD− rates at EOI
in the R-chemo arm were higher than ­pre­vi­ously reported and in-
creased fur­ther in the G-arm, e ­ spe­cially after CHOP/CVP. Indeed,
G abro­gated the com­part­ment- and ­chemo-related ­effects ­obser-
ved in the R-arm. MRD response rates in the PB were sim­i­lar across
all­chemo reg­i­mens (96% G-bendamustine [B], 96% G-CHOP, 94%
G-CVP), as well as in the BM (93% after G-B, 93% after G-CHOP).20,21
The impor­tance of MRD was also shown in the Relevance
trial, where MRD was quan­ti­fied for the first time by ddPCR. At
EOI, 98% and 78% of patients achieved a com­plete molec­u­lar
response in the PB and BM. A com­plete molec­u­lar response was
reached more fre­quently with R2 (90%) than with R-chemo (77%)
(P    =   .022) (Figure 1, Table 1).22
In advanced stage FL, since MRD− sta­tus after treat­ment can be
detected in patients in CR or partial response (PR), the molec­u­lar
Minimal resid­ual dis­ease in indo­lent NHL  |  321
 Table 1. Studies reporting on MRD detec­tion and mon­i­tor­ing in FL

Advanced
Study dis­ease Patients Therapy Tissue Method Marker Tumor bur­den MRD− % Clinical impact Follow-up
Rambaldi BCL2/IGH  + 128 (79 CHOP   ×   6 BM    ±    PB Nested Enrolled PCR  + — End of CHOP: 3-year FFR 17 months
et al13 FL, received R) 4 weeks R PCR patients BM MRD− 36% (43/118) BM MRD− after CHOP (8-39)
untreated in PCR  + (BCL2/IGH PB MRD− 35% (43/121) 52%
patients 100%) After R: MRD− 59  % at BM MRD− after R at
week 12, 74  % at week 28, week 44, 3-year FFR
and 63% at week 44 57% vs 20% MRD+
Rambaldi BCL2/IGH  + 86 (same BM RQ- Enrolled PCR    + At base­line: See above FFR: 64  % MRD− vs 32  % 56 months
et al14 FL, series) PCR patients Low (1 cell in 103–105): MRD  +  (P    <    .006) (40-75)
untreated (BCL2/IGH 43  %;
100%) inter­me­di­ate (1 cell in
102–103): 34  %;
high (>1 cell in 102): 23  %
CR: 71  % in low vs 26  %
in high
EFS: 59  % in

322  |  Hematology 2021  |  ASH Education Program

low/inter­me­di­ate
vs 32% in high
Ladetto High-risk FL 134 Phase 3 BM Nested 73/104 (70  %) — Overall, 65  % (39/60) MRD− PFS MRD− vs MRD  + 51 months
et al15 untreated R-HDS vs PCR BCL2/IGH After CHOP-R: MRD− 44% (P    <    .001), regard­less
<60 years CHOP   ×   6 + (65)  +  IGH (8) After R-HDS: MRD− 80% of treat­ment arm
R   ×   4 (P  <  .002)
(GITMO/IIL)
Bruna High-risk FL 134 Update of the BM Nested — Overall, 65% (39/60) MRD− 13-year OS 13 years
et al16 untreated phase 3 PCR R-HDS 64.5  % vs 68.5  %
<60 years R-HDS vs CHOP-R
CHOP   ×   6  + 13-year OS
R   ×   4 BM MRD− 82.1  % vs
(GITMO/IIL) BM MRD  +   51.9  %
(P    =.030)
20/24 (83  %) patients
with 4-year long-
term MRD− in first CR
Ladetto FL >60 years, 227    ML17638 BM RQ- 116/227 PFS: 80  % in low vs 75  % At EOI: 3-year PFS: at end 42 months
et al17 untreated R-FND ×   4   +   4R PCR (BCL2/IGH in inter­me­di­ate vs MRD− 70  % of con­sol­i­da­tion:
±  4R every 51  %) 66  % in high End of con­sol­i­da­tion: 72  % MRD− vs 39  %
2 months PFS: 61% at month  +  42 MRD− 84  % MRD    +    (16  %) (P   <.007)
in both PCR  +  and Median PFS: 12 months
PCR  − at base­line MRD− (RQ-PCR);
75  % of MRD− free
of PD at 36 months
(P  < .001)
PFS main­te­nance arm:
83  % PCR− vs 60  %
PCR  + (P    = .007)
PFS obser­va­tion arm:
71  % PCR− vs 50  %  
PCR   + (P     <  .001)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021

 Table 1. (Continued)

Advanced
Study dis­ease Patients Therapy Tissue Method Marker Tumor bur­den MRD− % Clinical impact Follow-up
Galimberti FL untreated 415 Phase 3 BM Nested 220/415 At diag­no­sis: EOI Relapse rate: 33% NA
et al18 FOLL05 trial PCR; (BCL2/IGH 61.9% PCR+ non-CR vs MRD− (qPCR): 109/154 (70.8%) for MRD− vs 41%
R-CHOP, R- RQ- 53%) 38.1% PCR− (P  = .027) (no dif­fer­ence between arms) for MRD+ at EOI
CVP, R-FN PCR RQ-PCR in 105 BM biopsy+ 32% non-CR R-CHOP 39% (P  = .363)
cases vs 21.8% in BM− cases R-FN 36% 3-year PFS 64.3% for
(P  = .021) R-CVP 25% MRD− vs 53.1 for
3-year PFS: 74% in RQ decrease >3 log: MRD+ at EOI (P = .08)
PCR−/BM− at diag­ R-CHOP 42.1% 3-year PFS 66% for
no­sis vs 55% in R-FN 36.8% R-CVP 21.1% 63 MRD− vs 41%
PCR+/BM+ (P  = .04) (P  = .07) for 24 MRD+ at
ORR 38.9% in high RQ vs 12 months from EOI
76.6% in low (P  = .006) (P  = .015)
Relapses: 22% in 3-year PFS 84% for
<1 × 10–4 cop­ies vs 46 MRD− vs 50%
78% in >1 × 10–4 cop­ies for 19 MRD+ at
(P  = .033) at base­line 24 months (P  = .014)
3-year PFS: 80% in low vs
59% in high (P  = .015)
Zohren FL, untreated 114 Phase 3 PB RQ- 114/173 Median PFS: 22 months Overall MRD− 78/92 (85%) 3-year PFS 63% over­all 41 months
et al19 NHL1-2003 trial PCR (BCL2/IGH in high BCL2/IGH After R-B (0-69)
R-CHOP vs R-B 66%) (ratio >1, n  =  28) vs NR MRD− 43/48 (89.6%)
in inter­me­di­ate (ratio After R-CHOP
0.1-1, n  =  24) vs NR in MRD− 35/44 (79.5%)
low (<0.1, n  =  55)
High vs inter­me­di­ate:

Prakash Singh Shekhawat

HR, 4.28; 95% CI,
1.70-10.77 (P  = .002)
High vs low: HR, 3.02;
95% CI, 1.55-5.86
(P  = .001)
Pott FL, untreated 1101 Phase 3 PB   ±   BM RQ- 968/1101 Clonal marker vs no MI PB MRD−: G-vs R-based treat­ 4-year PFS: 80% early 57 months
et al20,21 Gallium trial R PCR (BCL2/IGH marker at base­line: ment (94% vs 89%, P   =.013) respond­ers (PB MI
vs G (CVP, B, Nested +  IGH 88%) stage IV 61% vs 34%; EOI PB   ±   BM MRD−: G- vs R-based MRD−/EOI MRD−)
CHOP)  +  main­ PCR 815 (74%) BM biopsy infil­tra­ treat­ment (92% vs 85%, P   =  .0041) vs 30% PB MI
te­nance suit­able for tion 58.2% vs 14.3%; EOI BM MRD−: G- vs R-based treat­ MRD   +/EOI MRD   +
RQ-PCR high-risk FLIPI 44.5% ment (93% vs 83%, P  =.0014), NS (P   < .0001) vs 60%
vs 30.8%; extranodal in the PB late respond­ers, PB
involve­ment 70% vs EOI MRD− PB  ±  BM: R-CVP 76% vs MI MRD   +/EOI MRD−
40% G-CVP 91.4%; R-CHOP 77.8% vs (P   = .0133)
G-CHOP 91.3%; R-B 89.6% vs EOI MRD− sig­nif­i­
G-B 92.5% cantly affected PFS
EOI PB MRD−: R-CVP 79% vs G-CVP (HR, 0.38; 95% CI,
94%; R-CHOP 93% vs G-CHOP 0.26-0.56; P < .0001)
96%; R-B 96% vs G-B 96% and OS (HR, 0.35;
EOI BM MRD−: R-CHOP 74% vs 95% CI, 0.17-0.72;
G-CHOP 93%; R-B 87% vs G-B P   = .0027)
93%
EOI MRD+: 22/24 (92%) patients
in the G-chemo arm and 36/46
(78%) patients in the R-chemo

Minimal resid­ual dis­ease in indo­lent NHL  |  323

arm achieved MRD neg­a­tiv­ity
early dur­ing main­te­nance
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021
 Table 1. (Continued)

Advanced
Study dis­ease Patients Therapy Tissue Method Marker Tumor bur­den MRD− % Clinical impact Follow-up
Delfau- FL untreated 440 Phase 3 PB   ±   BM ddPCR 222/440 Tumor bur­den 10 times At EOI (week 24): 3-year PFS: 84% for NA
Larue Relevance trial (BCL2/IGH higher in PB and BM of PB MRD− 98%, BM MRD− 78% MRD− (in PB and/or
et al22 R2 (18 cycles of 50.45%) patients with week 24 R2 arm: MRD− 105/117; 90% BM) vs 55% for
lenalidomide BCL2/IGH+ MRD+, com­pared to vs R-chemo MRD− 70/90; MRD+ (P   = .015)
plus R, patients: patients with week 24 77% (P  =  .022) 3-year PFS: 85% for BM
followed by stage III-IV MRD− (P  = .03 and .02) MRD− vs 54% for BM
R main­te­ dis­ease MRD+ (P   = .011)
nance ther­ (96% vs 90%, MRD+ at EOI: R-chemo
apy every 8 P  =  .004), arm (HR, 3.3; 95%
weeks for FLIPI score >1 CI, 1.2-9.2; P   = .02)
12 cycles) (91% vs 83%, vs R2 arm (HR, 2;
vs R-chemo P  =  .002), 95% CI, 0.6-6.8;
(CHOP) BM involve­ P    = .27)
followed by ment (62% vs
R-main­te­ 48%, P  =  .003)
nance every
8 weeks for

324  |  Hematology 2021  |  ASH Education Program

12 cycles
Pott et al23 R refrac­tory FL Phase 3 PB    ±    BM RQ- 71% (228/319) At MI: PB MRD− 79% (41/52) in G-B PFS (HR, 0.33; P < .0001) 31.8
GADOLIN PCR (63% vs 47% (17/36) in B (P   = .0029) and OS (HR, 0.39; months
trial (and BCL2/IGH+ At EOI, PB or BM MRD− 86% P   = .008) for MRD−
G-B  +  G sub­se­ [MBR, mcr, (54/63) in G-B vs 55% (30/55) in PFS in MRD +3.3
main­te­nance quent 3′MBR] and B (P   = .0002) months in both arms
vs B nested) 72% IGH) in MRD− 8.54 B vs
52% (166/319) 35.71 G-B
suit­able for
RQ-PCR ≤10-4
Localized dis­
Study ease Patients Therapy Tissue Method Marker Tumor bur­den MRD− % Clinical impact Follow-up
Pulsoni Stage I (78%) 67 IFRT (24-30 Gy) PB   ±   BM Nested 72% 84-month PFS: 75% for After RT: MRD −50% Different PFS in MRD+ 82
et al24 Stage II (22%) (+4 weeks R in PCR BCL2/IGH- vs 59% for After R: 16/19 patients post-IFRT: months
untreated MRD+ from RQ- BCL2/IGH+ at base­ (84%) MRD− between untreated (17-196)
FL 2005 in 19 PCR line (P  = .26) patients vs patients
patients) ddPCR 84-month PFS: 90.9% in treated with R after
11patients with MRD 2005 (P   = .049)
<10−5 vs 38% in Different PFS of MRD+
19 patients with MRD patients dur­ing the
≥10−5 by ddPCR at fol­low-up vs per­sis­
base­line (P  = .015) tently MRD− patients
(P   = .038)
Herfarth Stage I (56%) 85 (60 long MIR trial 83 (64 Overall 24/64 57% below 10−4 MRD−: 20/21 (95%) at week 18 15/19 patients MRD− 29.6
et al25 Stage II (44%) term) R  ×  8 cycles  + PB, 13 (38%; PB with no pro­gres­sion; months
untreated IFRT (30-40 BM) 35%, BM 3 patients relapsed,
FL Gy) 46%) in 2 asso­ci­ated or
Stage II predicted by MRD+
> stage I sam­ple
(P  = .0038)
Only tri­als includ­ing anti-CD20 mono­clo­nal antibodies have been included.
FN, fludarabine and mitoxantrone; FND, fludarabine, mitoxantrone, and dexa­meth­a­sone; IFRT, involved field radio­ther­apy; MIR, Mabthera and Involved field Radiotherapy (NCT00509184); NA,
not available; NS, not significant; ORR, over­all response rate; qPCR, quan­ti­ta­tive PCR.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021


Figure 1. Rates of MRD negativity after anti-CD20 based treatments in advanced FL. *Data are expressed as proportion of MRD−
cases at EOI. gDNA, genomic DNA.
and ­clin­i­cal/radio­logic response can be com­bined to refine prog­
no­sis. Ladetto et al17 showed that at month +8, 69% of patients
were CR/nested PCR−, 15% PR/PCR−, and 11% CR/PCR+, with
a 3-year PFS of 77%, 59%, and 45%, respec­tively; the 3 PR/PCR+
cases relapsed within 23 months. Similar results were shown in
the FOLL05, where PFS was sig­nif­i­cantly lon­ger in CR-PR/PCR−
than in CR-PR/PCR+ patients.18 As in the Relevance study,22 the
pre­dic­tive value of MRD was supe­rior to clin­i­cal response in MVA.18
Luminari et al,29 in a small sub­group of patients (n = 41) from
the FOLL05 trial, suggested the com­ple­men­tary role of MRD
and PET at EOI in defin­ing the response to treat­ment, with a
con­cor­dance of 76%. The 3-year PFS was 78%, 50%, and 27%
in 28 PET−/MRD−, 8 PET  −    /MRD  +  , and 5 PET+ cases, respec­
tively (P   =   .015). The con­com­i­tant PET/MRD neg­a­tiv­ity was asso­
ci­ated with a bet­ter out­come: 5-year PFS 75% PET−/MRD− vs
35% PET+ or MRD+ (P = .012).29 The com­ple­men­tary role of PET
and MRD was fur­ther supported in 298 patients from the Gallium
trial: PET−/MRD+ or PET  +  /MRD− patients had a 2.1 higher risk of
pro­gres­sion or death than those with both PET−/MRD−. Never-
theless, 15% of PET−/MRD− patients progressed within 2.5 years
from EOI.30
Final com­ment. The intro­duc­tion of chemoimmunotherapy with
R and G has allowed an increase in the rates of MRD neg­a­tiv­ity at
EOI up to 70% to 80% and 90%, respec­tively. MRD anal­y­sis is a sen­
si­tive tool to refine clin­i­cal response assess­ment in FL. In addi­tion,
the com­bi­na­tion of molec­u­lar and met­a­bolic response assess­
ment is a prom­is­ing and valu­able tool to be fur­ther explored.
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021
Is MRD use­ful to pre­dict PFS?
The prog­nos­tic impact of MRD in advanced FL has been dem­on­
strated across dif­fer­ent treat­ment strat­e­gies. All chemoimmuno-
therapy tri­als, with or with­out main­te­nance, are asso­ci­ated with
a sig­nif­i­cant improve­ment of PFS in molec­u­lar respond­ers, inde­
pen­dent from other prog­nos­tic fac­tors (Table 1). Rambaldi et al13
showed that the achieve­ment of a BM PCR− sta­tus was asso­ci­ated
with a higher free­dom-from-recur­rence (FFR) (64% vs 32% for PCR+
patients). Interestingly, this study antic­i­pated later obser­va­tions17,18,21
(ie, most PCR− patients after 6 cycles were already neg­a­tive after 3
cycles, and a delayed MRD clear­ance induced by R was pos­si­ble). In
the phase 3 Gruppo Italiano Trapianto Midollo Osseo/Intergruppo
Italiano Linfomi (GITMO/IIL) trial, com­par­ing high-dose sequen­tial
che­mo­ther­apy with R and auto­graft (R–high dose sequen­tial che­
mo­ther­apy with autografting [HDS]) to R-CHOP, molec­u­lar remis­
sion—documented in 44% of R-CHOP–treated patients and 80% of
R-HDS–treated patients—was the stron­gest pre­dic­tor of PFS, EFS,
and FFR.15 The out­come of patients achiev­ing an MRD− response
was sim­i­lar regard­less of the treat­ment received, as was the out­
come of patients remaining MRD+.15 In the phase 3 ML17638 trial,17
MRD neg­a­tiv­ity (by PCR and RQ-PCR) at the end of con­sol­i­da­tion
(month +8) was asso­ci­ated with a lon­ger 3-year PFS, representing
an inde­pen­dent pre­dic­tor of out­come, besides clin­i­cal response
and FLIPI.17 Galimberti et al18 showed that the BM MRD sta­tus at 12
and 24 months from EOI, but not at the EOI, predicted the 3-year
PFS. In a MVA includ­ing FLIPI, BM involve­ment, qual­ity of response,
and arm of ther­apy, MRD per­sis­tence at month +12 (besides BM
Minimal resid­ual dis­ease in indo­lent NHL  |  325
involve­ment) and at month +24 (alone) retained a poor prog­nos­tic
role. Zohren et al19 showed that a per­sis­tent PB BCL2/IGH pos­i­tiv­
ity (n = 14, 15%) after R-che­mo­ther­apy was asso­ci­ated with a shorter
PFS (8.7 months vs not reached, P = .002), despite the 2-log reduc­
tion in BCL2/IGH lev­els and in both treat­ment arms. By MVA, both
pre­treat­ment and posttreatment BCL2/IGH lev­els were sig­nif­i­cant
prog­nos­tic fac­tors, the for­mer being the stron­gest.
Recently, 3 stud­ies pro­vided the first evi­dence that a MRD−
achieve­ment fol­low­ing treat­ment is asso­ci­ated also with a pro-
longed OS. The long-term fol­low-up of a Brit­ish trial showed the
lon­ger sur­vival of MRD− respond­ers after 2 dif­fer­ent che­mo­ther­
apy reg­i­mens in the pre-R era.31 The updated results of the phase
3 GITMO/IIL trial showed a 13-year OS of 82.1% and 51.9% for BM
MRD− vs MRD+ patients.16 Moreover, in a sub­group of molec­u­
larly mon­i­tored patients for a median of 4 years since treat­ment
com­ple­tion, 20 of 24 (83%) patients alive in CR remained in first
MRD− response, rais­ing the hypoth­e­sis of a func­tional cure also
for FL. In the Gallium trial, at an updated median fol­low-up of
57 months, MRD− patients at EOI (n = 564) had a lon­ger PFS than
MRD+ (n = 70) (P < .0001) and a bet­ter OS (P = .0027).21
The Gallium trial also opened the way to novel con­cepts.21 (i)
The role of main­te­nance either maintained or increased MRD−
response rates. Indeed, two-thirds of MRD− responses were pre­
served through­out main­te­nance (G, 67.0%; R, 63.2%) with no
dif­fer­ence between the 2 arms in the rate of con­ver­sion to MRD+
(6.3% vs 6.1%, respec­tively). Furthermore, within MRD+ patients
at EOI, 92% of 24 patients in the G-chemo and 78% of 46 in the
R-chemo arm became MRD− in the first months of main­te­nance;
the few cases who were never MRD− rap­idly progressed. (ii)
For the first time, MRD was assessed not only at EOI but also
at the mid­dle of induc­tion (MI). PFS was sig­nif­i­cantly lon­ger for
PB MRD− vs MRD+ at MI (P < .0001), with early respond­ers (MI
MRD−/EOI MRD−) hav­ing the best prog­no­sis.
In the Relevance trial, MRD− or MRD+ patients in the PB and/or
BM at EOI had a 3-year PFS of 84% and 55%, respec­tively, being sig­
nif­i­cant only in the BM and in the R-chemo but not in the R2 arm.22
Even in the relapse/refrac­tory set­ting, the GADOLIN trial
showed that MI MRD in the PB and/or BM was sig­nif­i­cantly dif­fer­
ent between the G-B and the B arm. In MVA, MRD sta­tus at EOI
(85.7% G-B vs 55% B-arm), treat­ment arm and FLIPI sta­tus were
strongly pre­dic­tive of PFS and, to a lesser extent, of OS.23
In local­ized FL, a monocentric expe­ ri­
ence dem­ on­ strated
that after treat­ment with local 24 to 30 Gy radiotherapy (RT),
dis­ap­pear­ance of cir­cu­lat­ing BCL2/IGH+ cells in the PB and/or
BM occurred in 50% (n  =  20/40) of patients.24 Additional treat­
ment with R in MRD+ patients after RT or who had molec­u­larly
relapsed dur­ing the fol­low-up achieved a molec­u­lar CR in 84% of
cases: the 82-month PFS of MRD+ R-treated patients was sig­nif­i­
cantly bet­ter than his­tor­i­cal con­trols, and the relapse prob­a­bil­ity
was sig­nif­i­cantly lower in MRD− patients after RT or after R than
that of patients remaining MRD+.24
Final com­ment. MRD neg­a­tiv­ity is pre­dic­tive of a bet­ter PFS
in all­clin­i­cal tri­als conducted in the past 2 decades, even in
relapsed patients, and pos­si­bly of a lon­ger sur­vival in stud­ies
with a prolonged fol­low-up. Assessment of MRD at ear­lier time
points with respect to EOI can also be infor­ma­tive.
Is the prolonged molec­u­lar fol­low-up of clin­i­cal rel­e­vance?
The prog­nos­tic value of sequen­tial BCL2/IGH assess­ment to pre­
dict clin­ic
­ al relapse is supported by lim­ited evi­dences and often
326  |  Hematology 2021  |  ASH Education Program
ham­pered by the lack of adher­ence to long-term MRD eval­u­a­
tions.18,19,21,27 In FOLL05, molec­ul­ar recur­rence pre­ceded clin­i­cal re-
lapse by a median of 5 months in 9 of 10 evaluable cases.18 Zohren
et al19 over a 41-month fol­low-up showed that 49 patients remained
BCL2/IGH– and 43 converted to BCL2/IGH+. Qualitative molec­u­lar
relapse was not indic­a­tive of an infe­rior PFS, unless asso­ci­ated with
high BCL2/IGH lev­els. In 20 patients who were rig­or­ously sam­pled,
the inter­val from BCL2/IGH redetection to clin­i­cal relapse was 9.5
months. In the Gallium trial, patients fail­ing to become MRD− had a
high like­li­hood of expe­ri­enc­ing early pro­gres­sion or death.21
More rel­e­vant than the MRD punc­tual eval­u­a­tion is the kinetic
anal­y­sis of molec­u­lar results over time. So far, few data are avail­­
able in FL. The first evi­dence came from Ladetto et al,17 who
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021
showed that the accu­mu­la­tion of MRD− results over time reduced
the like­li­hood of relapse. Kinetic mod­els of MRD anal­y­sis in FL are
awaited.
Final com­ment. Patients with FL with a higher relapse prob­
a­bil­ity can be iden­ti­fied, but the tim­ing of clin­i­cal relapse is not
accu­rately predicted by the cur­rent MRD ana­ly­ses. Accumulating
evi­dence sug­gests that a kinetic model of MRD anal­ys­is, more
than a sin­gle result, could help in antic­ip ­ at­ing dis­ease recur­rence.
Can we use MRD to mod­u­late treat­ment?
The first attempts to use MRD as a tool to guide ther­a­peu­tic choic-
es were pro­vided in advanced FL, where R could con­vert MRD+
after CHOP to MRD−, and in local­ized FL after RT.13,24 Two recent
large pro­spec­tive Ital­ian Fondazione Italiana Linfomi (FIL) tri­als ex-
plored the use of MRD to mod­u­late treat­ment in FL.32,33 FOLL12
was a phase 3 trial designed to prove the fea­si­bil­ity of a postin-
duction mod­u­la­tion of the stan­dard 2-year R main­te­nance accord-
ing to met­a­bolic response (Deauville score 1-3) and MRD response
at EOI after chemoimmunotherapy in advanced FL. In the exper­i­
men­tal arm, PET−/MRD− patients (29%) under­went obser­va­tion,
PET−/MRD+ patients (4%) received repeated weekly R, and PET+
patients had 1 dose of ibritumomab tiuxetan followed by stan­
dard R main­te­nance. The exper­i­men­tal arm showed a sig­nif­i­cantly
infe­rior PFS com­pared with the stan­dard 2-year R main­te­nance,
par­tic­u­larly for PET−/MRD− patients.32 It appears that a sin­gle
molec­u­lar response eval­u­a­tion at EOI is prob­a­bly not suf­fi­cient to
indi­cate dis­ease erad­i­ca­tion and to deintensify treat­ment. Results
of MRD mon­i­tor­ing will be shown at this meet­ing by Ladetto et al
(Abstract submission #146773).
The MIRO’ (Molecularly Immuno-Radio-therapy Oriented,
EUDRACT 2012-001676-11) trial for local­ ized FL used MRD to
guide post-RT treat­ment.33 Thirty per­cent of patients had cir­
cu­lat­ing BCL2/IGH+ cells,25,33 18 MRD+ patients after RT (60%)
and 8 MRD+ dur­ing fol­low-up received ofatumumab. With this
strat­egy, 91.7% of patients achieved a molec­ul­ar CR, with a sug-
gested clin­i­cal ben­e­fit after 38 months of fol­low-up.
The main lim­i­ta­tion in the intro­duc­tion of MRD mon­i­tor­ing in
the man­age­ment of patients with FL, espe­cially if con­sid­er­ing
only the BCL2/IGH rearrangement, is represented by the lack of
a molec­u­lar tar­get in a nota­ble pro­por­tion of cases. It is hoped
that tech­ni­cal advance­ments cur­rently under inves­ti­ga­tion will
over­come these lim­i­ta­tions (Figure 2, Tables 2-3).34,35-43
Final com­ment. R main­te­nance holds and increases the rates
of MRD neg­a­tiv­ity. Two recent tri­als (FOLL12 and MIRO’) explored
MRD-driven mod­u­la­tion of the postinduction ther­apy in FL, with
treat­ment inten­si­fi­ca­tion or deintensification in MRD+ and MRD−
patients at the EOI, respec­tively. Final results are awaited.
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021
Figure 2. Standard and new techniques to monitor MRD in FL.

Table 2. Principles of stan­dard and next-gen­er­a­tion tech­nol­o­gies in MRD mon­i­tor­ing

Technology References Potential advan­tages and open issues

RQ-PCR 1,35 The most val­i­dated and stan­dard­ized quan­ti­ta­tive method for MRD detec­tion in FL. Despite remark­able sen­si­tiv­ity and
spec­i­fic­ity, RQ-PCR has nota­ble lim­i­ta­tions due to the lack of BCL2/IGH tar­get in about 40% of advanced FL and 65%
to 70% of local­ized FL, cur­rently not eli­gi­ble for MRD assess­ment, and to the chal­lenge of very low MRD lev­els, where
it is dif­fi­cult to dis­sect if the sig­nal observed by PCR (not quan­ti­fi­able) is due to few resid­ual leu­ke­mic cells or to a
non­spe­cific ampli­fi­ca­tion of nor­mal DNA.
NGS 34,36 Potential deeper sen­si­tiv­ity com­pared with the clas­sic meth­ods.
A sen­si­tiv­ity of 10−6 is achiev­able only when high amounts of DNA are used.
Broad spec­trum of iden­ti­fi­able molec­u­lar tar­gets: a “cap­ture-based” pro­to­col cov­er­ing the cod­ing V, D, J genes of the
IGH loci was capa­ble of detecting clonal rearrangements in 87% (21/24) of lymphoproliferative dis­or­ders.
TLA 37,38 Alternative NGS tool capa­ble of sequenc­ing struc­tural var­i­ants, usu­ally not detected by con­ven­tional PCR approaches,
thanks to the selec­tive ampli­fi­ca­tion and sequenc­ing of entire genes on the basis of the cross-linking of phys­i­cally
prox­i­mal DNA loci.
ddPCR 39,40 A third-gen­er­a­tion quan­ti­ta­tive method based on the par­ti­tion of the tem­plate DNA into water-in-oil drop­lets in which
PCR ampli­fi­ca­tion occurs, allowing the quan­ti­fi­ca­tion of nucleic acid tar­gets with­out the need of the cal­i­bra­tion
curve.
Sensitivity, accu­racy, and repro­duc­ibil­ity at least com­pa­ra­ble to that of RQ-PCR.
A good per­for­mance in the MRD quan­ti­fi­ca­tion in at least 20% to 30% of sam­ples resulting pos­i­tive but not quan­ti­fi­able
by RQ-PCR.
cfDNA 41-43 Plasma is a poten­tially impor­tant source of DNA (ie, cfDNA), poten­tially use­ful to iden­tify dis­tinct bio­log­i­cal sub­types
of lym­pho­mas and to pro­vide insights into the pat­terns of geno­mic evo­lu­tion/resis­tance through­out treat­ment in all­
com­part­ments, not only PB and BM.
A non­in­va­sive tool to mon­i­tor MRD in non-Hodgkin lym­phoma through patient-spe­cific IGH rearrangements, to iden­tify
indi­vid­u­als at an increased risk of relapse and to detect relapse before clin­i­cal evi­dence of dis­ease.
Most cfDNA orig­i­na­tes from leu­ko­cytes, and only a small frac­tion (<10%) is tumor derived, known as ctDNA. ctDNA
con­cen­tra­tion varies among patients and dif­fers according to the type, loca­tion, and stage of can­cer, with some
pro­duc­ing extremely low con­cen­tra­tions. Thus, it is unlikely that cfDNA will reach the sen­si­tiv­ity of MRD anal­y­sis on
cir­cu­lat­ing cells in FL. Before broad clin­i­cal implementation, issues on preanalytical fac­tors must be addressed in
order to achieve con­sis­tent and repro­duc­ible results.
A major stan­dard­i­za­tion effort is under way within the EuroClonality (https:​­/​­/www​­.euroclonalityngs​­.org​­/usr​­/pub​­/pub​­.php) and EuroMRD Consortium
(www​­.euromrd​­.org), to estab­lish guide­lines for NGS and ddPCR MRD anal­y­sis and their future appli­ca­tion in stan­dard clin­i­cal prac­tice.
cfDNA, cell-free DNA; NGS, next-gen­er­a­tion sequenc­ing; TLA, tar­get locus ampli­fi­ca­tion.
Prakash Singh Shekhawat
Minimal resid­ual dis­ease in indo­lent NHL  |  327
Table 3. First appli­ca­tions of next-gen­er­a­tion tech­nol­o­gies to molec­u­lar mon­i­tor­ing in FL

Study Patients Marker/tis­sue/tim­ing Method Potential impact

Genuardi et al38 20 FL with no MBR BCL2/TLA TLA BCL2/TLA in 8 (40%) of “marker-neg­a­tive” cases
or mcr BM The new BCL2/TLA mark­ers were suit­able for RQ-PCR MRD
At diag­no­sis and for MRD anal­y­sis in 4 of 5 cases.
MRD by BCL2-TLA reached good sen­si­tiv­ity lev­els.
Cavalli et al40 67 early stage FL BCL2/IGH ddPCR Concordance between ddPCR and RQ-PCR: 82%
PB, BM ddPCR iden­ti­fied a MBR marker in 8 of 18 (44%) sam­ples that
At diag­no­sis and for MRD resulted in MBR−/mcr− by qual­it­ a­tive nested PCR.
Molecular tumor bur­den at diag­no­sis ≥10−5 sig­nif­i­cantly
predicted PFS only when quan­ti­fied by ddPCR but not by
RQ-PCR.
Higher sen­si­tiv­ity of ddPCR in RQ-PCR PNQ sam­ples.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021

Sarkozy et al42 34 FL from PRIMA trial IGH NGS 29 (85%) had 1 or more tumor clonotypes in the tumor
Tumor biopsy biopsy spec­i­men.
Plasma 25 (74%) had 1 or more tumor clonotypes in plasma.
At diag­no­sis 18 of 24 (75%) patients with an IGH clonotype had sev­eral
detect­able subclones in the tumor or in the plasma.
13 of 24 (54%) showed a subclone detected in both the
plasma and the tumor.
>50% of cases showed a dif­fer­ent dis­tri­bu­tion of subclones
between tumor and plasma.
High ctDNA lev­els at diag­no­sis predicted short PFS in MVA.
Delfau-Larue et al43 FL PET TMTV (n = 133) ddPCR 23 of 68 cfDNA were BCL2/IGH+ (ctDNA ≤10% cfDNA).
BCL2/IGH High cor­re­la­tion between CTCs and TMTV and between
PB CTC (n = 68) cfDNA and TMTV
PB cfDNA (n = 61) CTCs pre­dic­tive of out­come in uni­var­i­ate anal­y­sis but not in
At diag­no­sis MVA
Total cfDNA lev­els and TMTM are inde­pen­dent pre­dic­tors of
out­come.
CTC, cir­cu­lat­ing tumor cell; PNQ , positive not quantifiable; TMTV, total met­a­bolic tumor vol­ume.

CLINICAL CASES (Con­t in­u ed)
Patient 1. The per­sis­tence of MRD pos­it­iv­ity dur­ing the 2-year
treat­ment despite a PET neg­ a­
tiv­
ity sug­ gests a resis­
tance
to R-bendamustine treat­ ment. Indeed, the patient quickly
relapsed. In such patients, a pos­si­ble inten­si­fi­ca­tion/switch of
treat­ment could be explored in clin­ic ­ al tri­als.
Patient 2. Although in this local­ized FL, clin­i­cal relapse was
antic­i­pated by a molec­u­lar con­ver­sion to MRD pos­i­tiv­ity, this is
not always the case with the cur­rent MRD mon­i­tor­ing tim­ing and
modal­i­ties. The clin­i­cal ben­e­fit of adding an anti-CD20 mono­clo­
nal anti­body to RT in local­ized FL or to treat molec­u­lar relapses
needs to be addressed in well-designed clin­i­cal tri­als.
Other iNHLs: lymphoplasmacytic lym­phoma/
Waldenström mac­ro­glob­u­li­ne­mia and mar­ginal
zone lym­phoma
Although the role of MRD mon­i­tor­ing in FL is pro­gres­sively
increas­ing, MRD in other iNHLs, such as lymphoplasmacytic lym­
phoma/Waldenström mac­ro­glob­u­li­ne­mia (WM) and mar­ginal
zone lym­ phoma, is only recently starting to be explored.44,45
In WM, MYD88L265P is a diag­nos­tic and pre­dic­tive bio­marker of
ther­apy response. Besides allele-spe­cific RQ-PCR, ddPCR has re-
cently proved to be a suit­able and sen­si­tive tool for MYD88L265P
screen­ing and MRD mon­i­tor­ing.44 Both unsorted BM and PB sam­
ples can be reli­ably tested, as well as cir­cu­lat­ing tumor DNA
(ctDNA), which rep­re­sents an attrac­tive and less inva­sive alter­
na­tive to BM for MYD88L265P detec­tion.44 MYD88L265P detec­tion
in the cere­bral spi­nal fluid by ddPCR is also use­ful to diag­nose
Bing-Neel syn­drome.46
Promising results have been pre­lim­i­nar­ily shown in splenic mar­
ginal zone lym­phoma, where MRD has been assessed in BM and
PB by ddPCR using IGH allele-spe­cific oli­go­nu­cle­o­tide prim­ers in
the phase 2 BRISMA/IELSG36 (Bendamustine-rituximab as first-
line treatment of splenic marginal zone lymphoma/International
Extranodal Lymphoma Study Group) trial.45
Concluding remarks
Despite sev­eral decades of research, MRD anal­y­sis in FL has not
yet entered the day-to-day clin­ i­
cal prac­tice. Indeed, out­
side
of clin­i­cal tri­als, MRD results should not be used to take clin­i­cal
deci­sions in the real-life man­age­ment of patients with FL. The is-
sue of BCL2/IGH+ non­ma­lig­nant B cells that can be found in the
PB of healthy indi­vid­u­als47 is mar­ginal in treated patients with FL,
because the emer­gence of a BCL2/IGH clone unre­lated to the
dis­ease is very unlikely and rarely requires the need of prov­ing
the sequence iden­tity of the rearrangement.
To move MRD in FL from clin­i­cal tri­als to daily prac­tice, fur­
ther stud­ies are needed to over­come the limit represented by
the lack of a molec­u­lar marker for all­patients; to estab­lish an
inte­grated risk strat­i­fi­ca­tion; to dem­on­strate whether MRD anal­
y­sis can be inte­grated with PET for a refined def­i­ni­tion of “poor
respond­ers,” can­di­dates to exper­i­men­tal approaches; to explore
other MRD-adapted treat­ment modal­i­ties, pos­si­bly with kinetic
mod­els of MRD anal­y­sis; and to eval­ua ­ te if chemo-free com­bi­

328  |  Hematology 2021  |  ASH Education Program

na­tions have an advan­tage in terms of MRD achieve­ment com­
pared with chemoimmunotherapy. Results from new-gen­er­a­tion
clin­i­cal tri­als (eg, FIL FOLL12 and Gallium) are eagerly awaited.
Data are mature to design MRD-driven clin­i­cal tri­als that incor­
po­rate the molec­u­lar mon­i­tor­ing in the man­age­ment of patients
with FL.
Acknowledgments
Work partly supported by Associazione Italiana per la Ricerca sul
Cancro (AIRC), Metastases 5   x   1000 Special Program, N° 21198,
Milan, Italy (Robin Foà). The authors thank Dr. Marco Ladetto and
the FIL MRD net­work.
Conflict-of-inter­est dis­clo­sure
Ilaria Del Giudice: AstraZeneca, Tolero (advi­sory board).
Irene Della Starza: no conflicts to disclose.
Robin Foà: hon­ o­raria from Janssen, Amgen, Novartis, Incyte,
­Servier, and Sanofi.
Off-label drug use
Ilaria Del Giudice: nothing to disclose.
Irene Della Starza: nothing to disclose.
Robin Foà: nothing to disclose.
Correspondence
Robin Foà, Hematology, Department of Translational and Pre-
cision Medicine, Sapienza University, Via Benevento 6, 00161
Rome, Italy; e-mail: rfoa@bce​­.uniroma1​­.it
References
1. Pott C, Brüggemann M, Ritgen M, van der Velden VHJ, van Dongen JJM,
Kneba M. MRD detec­tion in B-cell non-Hodgkin lym­pho­mas using Ig gene
rearrangements and chro­mo­somal trans­lo­ca­tions as tar­gets for real-time
quan­ti­ta­tive PCR. Methods Mol Biol. 2019;1956:199-228.
2. Galimberti S, Genuardi E, Mazziotta F, et al. The min­i­mal resid­ual dis­ease
in non-Hodgkin’s lym­pho­mas: from the lab­o­ra­tory to the clin­i­cal prac­tice.
Front Oncol. 2019;9:528.
3. Grimaldi D, Genuardi E, Ferrante M, Ferrero S, Ladetto M. Minimal resid­ual
dis­ease in indo­lent lym­pho­mas: a crit­i­cal assess­ment. Curr Treat Options
Oncol. 2018;19(12):71.
4. Sarkozy C, Maurer MJ, Link BK, et al. Cause of death in fol­lic­u­lar lym­phoma
in the first decade of the rituximab era: a pooled anal­y­sis of French and US
cohorts. J Clin Oncol. 2019;37(2):144-152.
5. Casulo C, Byrtek M, Dawson KL, et al. Early relapse of fol­lic­u­lar lym­phoma
after rituximab plus cyclo­phos­pha­mide, doxo­ru­bi­cin, vin­cris­tine, and pred­
ni­sone defi­nes patients at high risk for death: an anal­y­sis from the National
LymphoCare Study. J Clin Oncol. 2015;33(23):2516-2522.
6. Federico M, Caballero Barrigón MD, Marcheselli L, et al. Aristotle Consor-
tium. Rituximab and the risk of trans­for­ma­tion of fol­lic­u­lar lym­phoma: a
ret­ro­spec­tive pooled anal­y­sis. Lancet Haematol. 2018;5(8):e359-e367.
7. Dreyling M, Ghielmini M, Rule S, et al; ESMO Guidelines Committee. Newly
diag­nosed and relapsed fol­lic­u­lar lym­phoma: ESMO Clinical Practice Guide-
lines for diag­no­sis, treat­ment and fol­low-up. Ann Oncol. 2021;32(3):298-
308.
8. Mir F, Mattiello F, Grigg A, et  al. Follicular Lymphoma Evaluation Index
(FLEX): a new clin­i­cal prog­nos­tic model that is supe­rior to existing risk
scores for predicting pro­gres­sion-free sur­vival and early treat­ment fail­ure
after front­ line immunochemotherapy. Am J Hematol. 2020;95(12):1503-
1510.
9. Pastore A, Jurinovic V, Kridel R, et al. Integration of gene muta­tions in risk
prog­nos­ti­ca­tion for patients receiv­ing first-line immunochemotherapy for
fol­lic­u­lar lym­phoma: a ret­ro­spec­tive anal­y­sis of a pro­spec­tive clin­i­cal trial
and val­i­da­tion in a pop­u­la­tion-based reg­is­try. Lancet Oncol. 2015;16(9):1111-
1122.
10. Huet S, Tesson B, Jais J-P, et al. A gene-expres­sion pro­fil­ing score for pre­
dic­tion of out­come in patients with fol­lic­u­lar lym­phoma: a ret­ro­spec­tive
Prakash Singh Shekhawat
train­ing and val­i­da­tion anal­y­sis in three inter­na­tional cohorts. Lancet
Oncol. 2018;19(4):549-561.
11. Huet S, Sujobert P, Salles G. From genet­ics to the clinic: a trans­la­tional per­
spec­tive on fol­lic­u­lar lym­phoma. Nat Rev Cancer. 2018;18(4):224-239.
12. Gribben JG, Freedman AS, Neuberg D, et al. Immunologic purg­ing of mar­
row assessed by PCR before autol­o­gous bone mar­row trans­plan­ta­tion for
B-cell lym­phoma. N Engl J Med. 1991;325(22):1525-1533.
13. Rambaldi A, Lazzari M, Manzoni C, et  al. Monitoring of min­i­mal ­resid­ual
­dis­ease after CHOP and rituximab in pre­vi­ously untreated patients with
fol­lic­u­lar lym­phoma. Blood. 2002;99(3):856-862.
14. Rambaldi A, Carlotti E, Oldani E, et al. Quantitative PCR of bone mar­row
BCL2/IgH+ cells at diag­no­sis pre­dicts treat­ment response and long-term
out­come in fol­lic­u­lar non-Hodgkin lym­phoma. Blood. 2005;105(9):3428-
3433.
15. Ladetto M, De Marco F, Benedetti F, et al; Gruppo Italiano Trapianto di
Midollo Osseo (GITMO); Intergruppo Italiano Linfomi (IIL). Prospective, mul­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021
ti­cen­ter ran­dom­ized GITMO/IIL trial com­par­ing inten­sive (R-HDS) ver­sus
con­ven­tional (CHOP-R) chemoimmunotherapy in high-risk fol­lic­u­lar lym­
phoma at diag­no­sis: the supe­rior dis­ease con­trol of R-HDS does not trans­
late into an over­all sur­vival advan­tage. Blood. 2008;111(8):4004-4013.
16. Bruna R, Benedetti F, Boccomini C, et al. Prolonged sur­vival in the absence
of dis­ease-recur­rence in advanced-stage fol­lic­u­lar lym­phoma fol­low­ing
chemo-immu­no­ther­apy: 13-year update of the pro­spec­tive, mul­ti­cen­ter
ran­dom­ized GITMO-IIL trial. Haematologica. 2019;104(11):2241-2248.
17. Ladetto M, Lobetti-Bodoni C, Mantoan B, et al; Fondazione Italiana Linfomi.
Persistence of min­i­mal resid­ual dis­ease in bone mar­row pre­dicts out­come
in fol­lic­u­lar lym­pho­mas treated with a rituximab-inten­sive pro­gram. Blood.
2013;122(23):3759-3766.
18. Galimberti S, Luminari S, Ciabatti E, et  al. Minimal resid­ual dis­ease after
con­ven­tional treat­ment sig­nif­i­cantly impacts on pro­gres­sion-free sur­vival
of patients with fol­lic­u­lar lym­phoma: the FIL FOLL05 trial. Clin Cancer Res.
2014;20(24):6398-6405.
19. Zohren F, Bruns I, Pechtel S, et  al. Prognostic value of cir­cu­lat­ing Bcl-2/
IgH lev­els in patients with fol­lic­u­lar lym­phoma receiv­ing first-line immu-
nochemotherapy. Blood. 2015;126(12):1407-1414.
20. Pott C, Hoster E, Kehden B, et al. Minimal resid­ual dis­ease in patients with
fol­lic­u­lar lym­phoma treated with obinutuzumab or rituximab as first-line
induc­tion immunochemotherapy and main­te­nance in the phase 3 GALLIUM
study. Blood. 2016;128(22):613.
21. Pott C, Hoster E, Kehden B, et al. Minimal resid­ual dis­ease response at end
of induc­tion and dur­ing main­te­nance cor­re­lates with updated out­come in
the phase III GALLIUM study of obinutuzumab- or rituximab-based immu-
nochemotherapy in pre­vi­ously untreated fol­lic­u­lar lym­phoma patients.
Blood. 2018; 132(suppl 1):396.
22. Delfau-Larue M-H, Boulland M-L, Beldi-Ferchiou A, et  al. Lenalidomide/
rituximab induces high molec­u­lar response in untreated fol­lic­u­lar lym­phoma:
LYSA ancil­lary RELEVANCE study. Blood Adv. 2020;4(14):3217-3223.
23. Pott C, Sehn LH, Belada D, et  al. MRD response in relapsed/refrac­tory
FL after obinutuzumab plus bendamustine or bendamustine alone in the
­GADOLIN trial. Leukemia. 2020;34(2):522-532.
24. Pulsoni A, Della Starza I, Cappelli LV, et al. Minimal resid­ual dis­ease monitor-
ing in early stage fol­lic­u­lar lym­phoma can pre­dict prog­no­sis and drive treat­
ment with rituximab after radio­ther­apy. Br J Haematol. 2020;188(2):249-258.
25. Herfarth K, Borchmann P, Schnaidt S, et al. Rituximab with involved field
irra­di­a­tion for early-stage nodal fol­lic­u­lar lym­phoma: results of the MIR
study. HemaSphere. 2018;2(6):e160.
26. Pulsoni A, Starza ID, Frattarelli N, et  al. Stage I/II fol­ lic­
u­lar lym­ phoma:
spread of BCL2/IGH+ cells in blood and bone mar­row from pri­mary site of
dis­ease and pos­si­bil­ity of clear­ance after involved field radio­ther­apy. Br J
Haematol. 2007;137(3):216-220.
27. Hirt C, Schüler F, Kiefer T, et  al. Rapid and sustained clear­ance of cir­cu­
lat­ing lym­phoma cells after che­mo­ther­apy plus rituximab: clin­i­cal sig­nif­i­
cance of quan­ti­ta­tive t(14;18) PCR mon­i­tor­ing in advanced stage fol­lic­u­lar
lym­phoma patients. Br J Haematol. 2008;141(5):631-640.
28. Boccomini C, Ladetto M, Rigacci L, et al. A brief rituximab, bendamustine,
mitoxantrone (R-BM) induc­ tion followed by rituximab con­ sol­i­
da­
tion in
elderly patients with advanced fol­lic­u­lar lym­phoma: a phase II study by
the Fondazione Italiana Linfomi (FIL). Br J Haematol. 2021;193(2):280-289.
29. Luminari S, Galimberti S, Versari A, et  al. Positron emis­sion tomog­ra­phy
response and min­i­mal resid­ual dis­ease impact on pro­gres­sion-free sur­vival
in patients with fol­lic­u­lar lym­phoma: a sub­set anal­y­sis from the FOLL05 trial
of the Fondazione Italiana Linfomi. Haematologica. 2016;101(2):e66-e68.
Minimal resid­ual dis­ease in indo­lent NHL  |  329
30. Pott C, Davies A, Hiddemann W, et al. Metabolic (PET) and MRD response
con­fer reduced risk of pro­gres­sion or death in patients treated within the
phase III Gallium study. HemaSphere. 2018;2(suppl 1):171-172.
31. Bishton MJ, Rule S, Wilson W, et  al. The UK NCRI study of chlorambucil,
mitoxantrone and dexa­meth­as­ one (CMD) ver­sus fludarabine, mitoxantrone
and dexa­meth­a­sone (FMD) for untreated advanced stage fol­lic­u­lar lym­
phoma: molec­u­lar response strongly pre­dicts prolonged over­all sur­vival.
Br J Haematol. 2020;190(4):545-554.
32. Luminari S, Galimberti, S, Versari A, et al. Response adapted post induc­tion
ther­apy in fol­lic­u­lar lym­phoma: updated results of the FOLL12 trial by the
Fondazione Italiana Linfomi (FIL). Hematol Oncol. 2021;39(suppl 2):133.
33. Pulsoni A, Tosti ME, Ferrero S, et  al. Early-stage fol­lic­u­lar lym­phoma: first
results of the FIL “Miro” study, a mul­ti­cen­ter phase II trial com­bin­ing local
radio­ther­apy and MRD-driven immu­no­ther­apy. Blood. 2019;134(suppl 1):124.
34. Ladetto M, Brüggemann M, Monitillo L, et al. Next-gen­er­a­tion sequenc­ing
and real-time quan­ti­ta­tive PCR for min­i­mal resid­ual dis­ease detec­tion in
B-cell dis­or­ders. Leukemia. 2014;28(6):1299-1307.
35. Della Starza I, Cavalli M, De Novi LA, et al. Minimal resid­ual dis­ease (MRD) in
non-Hodgkin lym­pho­mas: Interlaboratory repro­duc­ibil­ity on mar­row sam­
ples with very low lev­els of dis­ease within the FIL (Fondazione Italiana Lin-
fomi) MRD Network. Hematol Oncol. 2019;37(4):368-374.
36. Wren D, Walker BA, Brüggemann M, et  al. Comprehensive trans­ lo­
ca­
tion and clonality detec­ tion in lymphoproliferative dis­ or­
ders by next-
gen­er­a­tion sequenc­ing. Haematologica. 2017;102:e57-e60.
37. de Vree PJ, de Wit E, Yilmaz M, et al. Targeted sequenc­ing by prox­im­ity
liga­tion for com­pre­hen­sive var­i­ant detec­tion and local haplotyping. Nat
Biotechnol. 2014;32(10):1019-1025.
38. Genuardi E, Klous P, Mantoan B, et  al. Targeted locus ampli­fic ­ a­tion to
detect molec­u­lar mark­ers in man­tle cell and fol­lic­u­lar lym­phoma. Hematol
Oncol. 2021;39(3):293-303.
39. Drandi D, Kubiczkova-Besse L, Ferrero S, et  al. Minimal resid­ual dis­ease
detec­tion by drop­let dig­i­tal PCR in mul­ti­ple mye­loma, man­tle cell lym­
phoma, and fol­lic­u­lar lym­phoma: a com­par­i­son with real-time PCR. J Mol
Diagn. 2015;17(6):652-660.
330  |  Hematology 2021  |  ASH Education Program
40. Cavalli M, De Novi LA, Della Starza I, et al. Comparative anal­y­sis between
RQ-PCR and dig­i­tal drop­let PCR of BCL2/IGH gene rearrangement in the
periph­eral blood and bone mar­row of early stage fol­lic­ul­ar lym­phoma. Br J
Haematol. 2017;177(4):588-596.
41. Lakhotia R, Roschewski M. Circulating tumour DNA in B-cell lym­pho­mas:
cur­rent state and future pros­pects. Br J Haematol. 2021;193(5):867-881.
42. Sarkozy C, Huet S, Carlton VEH, et  al. The prog­ nos­tic value of clonal
­het­ero­ge­ne­ity and quan­ti­ta­tive assess­ment of plasma cir­cu­lat­ing clonal
IG-VDJ sequences at diag­no­sis in patients with fol­lic­u­lar lym­phoma. Onco­
target. 2017;8(5):8765-8774.
43. Delfau-Larue MH, van der Gucht A, Dupuis J, et al. Total met­a­bolic tumor
vol­ume, cir­cu­lat­ing tumor cells, cell-free DNA: dis­tinct prog­nos­tic value in
fol­lic­u­lar lym­phoma. Blood Adv. 2018;2(7):807-816.
44. Drandi D, Genuardi E, Dogliotti I, et al. Highly sen­si­tive MYD88L265P muta­
tion detec­tion by drop­let dig­i­tal poly­mer­ase chain reac­tion in Walden-
ström mac­ro­glob­u­li­ne­mia. Haematologica. 2018;103(6):1029-1037.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/320/1852135/320giudice.pdf by guest on 13 December 2021
45. Ferrero S, Ladetto M, Beldjord K, et al. First appli­ca­tion of min­i­mal resid­
ual dis­ease anal­y­sis in splenic mar­ginal zone lym­phoma tri­als: results from
BRISMA/IELSG36 phase II study. Hematol Oncol. 2019;37(52):224-225.
46. Hiemcke-Jiwa LS, Minnema MC, Radersma-van Loon JH, et  al. The use
of drop­ let dig­i­
tal PCR in liq­uid biop­sies: a highly sen­ si­
tive tech­nique
for MYD88 p.(L265P) detec­ tion in cere­bro­spi­
nal fluid. Hematol Oncol.
2018;36(2):429-435.
47. Roulland S, Lebailly P, Lecluse Y, et al. Long-term clonal per­sis­tence and
evo­ lu­tion of t(14;18)-bear­ ing B cells in healthy indi­ vid­
u­als. Leukemia.
2006;20(1):158-162.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000312
 INHERITED RED CELL DISORDERS BEYOND HEMOGLOBINOPATHIES

Diagnosis and clinical management of red cell

membrane disorders

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/331/1851848/331kalfa.pdf by guest on 13 December 2021

Theodosia A. Kalfa
Cancer and Blood Diseases Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH; and Department of Pediatrics, University of
Cincinnati College of Medicine, Cincinnati, OH

Heterogeneous red blood cell (RBC) membrane disorders and hydration defects often present with the common clinical
findings of hemolytic anemia, but they may require substantially different management, based on their pathophysiology.
An accurate and timely diagnosis is essential to avoid inappropriate interventions and prevent complications. Advances in
genetic testing availability within the last decade, combined with extensive foundational knowledge on RBC membrane
structure and function, now facilitate the correct diagnosis in patients with a variety of hereditary hemolytic anemias
(HHAs). Studies in patient cohorts with well-defined genetic diagnoses have revealed complications such as iron overload
in hereditary xerocytosis, which is amenable to monitoring, prevention, and treatment, and demonstrated that splenec-
tomy is not always an effective or safe treatment for any patient with HHA. However, a multitude of variants of unknown
clinical significance have been discovered by genetic evaluation, requiring interpretation by thorough phenotypic assess-
ment in clinical and/or research laboratories. Here we discuss genotype-phenotype correlations and corresponding clini-
cal management in patients with RBC membranopathies and propose an algorithm for the laboratory workup of patients
presenting with symptoms and signs of hemolytic anemia, with a clinical case that exemplifies such a workup.

LEARNING OBJECTIVES
• Discuss comprehensive evaluation for RBC membrane disorders, including phenotypic and genetic testing
• Review clinical management considerations tailored to the genetic diagnosis and pathophysiology of RBC
membrane disorders
• Elaborate on patient cases of RBC membrane skeleton disorders and hydration defects

the amount of normal α-spectrin produced from the affected

CLINICAL CASE
A European American boy presented with nonimmune
hemolyticanemia and hyperbilirubinemia as a neonate,
requiring red blood cell (RBC) transfusion and phototherapy.
He continued to require frequent transfusions every 4 to
8 weeks. There was no family history of hemolytic anemia.
His mother had some spherocytes on her blood smear, but
she was asymptomatic with no evidence of hemolysis. Since
the patient had been regularly transfused since birth, an RBC
phenotypic evaluation was not feasible to provide a diag-
nosis. Therefore, genetic evaluation on a next-generation
sequencing (NGS) panel of genes associated with hereditary
hemolytic anemia (HHA)was performed that showed a non-
sense SPTA1 mutation (c.4295del, p.Leu1432*)and the low-
expression SPTA1 polymorphism c.4339-99C>T (also known
as αLEPRA or low expression PRAgue). αLEPRA, a deep intronic
variant that affects splicing, causes a significant decrease in
Prakash Singh Shekhawat
allele. Compound heterozygosity of αLEPRA with a null SPTA1
mutation in trans is the most common cause of autoso-
mal recessive (AR) hereditary spherocytosis (HS) due to α-
spectrin deficiency.1-3 In addition, a variant of unknown clini-
cal significance (VUCS) was identified in PIEZO1: c.6205G>A
(p.Val2069Met), which, if pathogenic, could cause a compo-
nent of xerocytosis in this patient. Splenectomy in hereditary
xerocytosis (HX) is associated with life-threatening thrombo-
philia; therefore, it was important to explore the contribu-
tion of this PIEZO1 VUCS to the patient’s disease.
Parental studies were performed and confirmed that the
SPTA1 mutations were positioned in trans and were there-
fore pathogenic, causing HS. The father carried αLEPRA and
the PIEZO1 p.Val2069Met variant while the mother carried
the nonsense SPTA1 mutation p.Leu1432*. The patient had
osmotic gradient ektacytometry performed 8 weeks after
a transfusion that showed decreased EImax (maximum
Red cell membrane disorders | 331
elon­ga­tion index; ie, the max­im
­ um deformability of the RBCs),
com­pat­i­ble with HS but not typ­i­cal since Omin (ie, the hypo­tonic
osmo­lal­ity in which the elon­ga­tion index is min­i­mal) and Ohyp
(ie, the osmo­lal­ity value at which the cells’ aver­age max­i­mum
diam­e­ter is half of EImax) were nor­mal because nor­mal donor
RBCs were still pres­ent in his blood (Figure 1A).4 Both par­ents
had a nor­mal hemo­glo­bin (Hgb) and retic­ul­o­cyte count as well
as nor­mal ektacytometry. Each nor­ mal SPTA1 gene pro­ duces
α-spectrin well in excess; there­fore, it was not sur­pris­ing that
nei­ther par­ent had HS. In addi­tion, the ektacytometry curve of
the father’s blood sam­ple showed no evi­dence of HX despite
car­ry­ing the PIEZO1 VUCS. RBC cat­ion con­tent was eval­u­ated for
both par­ents and was within nor­mal range, offer­ing addi­tional
reas­sur­ance that the PIEZO1 p.Val2069Met var­i­ant was benign.
The child con­tin­ued with chronic trans­fu­sions and che­la­tion,
but his iron over­load appeared to be poorly con­trolled, and there­
fore a par­tial sple­nec­tomy was performed at 4 years of age, at least
2 weeks after pneu­mo­coc­cal and menin­go­coc­cal vac­ci­na­tions
were admin­is­tered.5 He remained trans­fu­sion-free for 18 months
with an Hgb of 8.7 to 11 g/dL, but his ane­mia grad­u­ally wors­ened,
requir­ing fre­quent trans­fu­sions again. A total sple­nec­tomy was
performed at 7 years of age since the remaining splenic tis­sue had

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/331/1851848/331kalfa.pdf by guest on 13 December 2021

Figure 1. HS. (A) Ektacytometry of blood samples from the patient before and after splenectomy and from his parents. The patient had
AR HS due to compound heterozygosity of SPTA1 c.4295del (p.L1432*), shared with the mom, and SPTA1 c.4339-99C>T, shared with
the dad. The dad and the proband were also found to carry the PIEZO1 VUCS c.6205G>A (p.Val2069Met). The parents had a normal
ektacytometry with no evidence of xerocytosis for the father, offering reassurance that the PIEZO1 VUCS is benign, while the patient
after splenectomy had a typical HS curve. Before splenectomy, the patient was chronically transfused. Ektacytometry at that time
was performed at a nadir before the next transfusion and 2 months after a previous one. Transfused RBCs have obviously modified the
curve, giving a falsely normal Omin (ie, the hypotonic osmolality where the elongation index is minimal). The value of Omin provides
information on the initial surface-to-volume ratio of the cell sample. A shift to the right reflects a decrease in the surface area/volume
ratio and corresponds to increased osmotic fragility. (B) Blood smear of the patient about 18 months post partial splenectomy before
starting to require transfusions again, demonstrating significant anisopoikilocytosis and polychromasia as well as many spherocytes,
typical for AR SPTA1-associated HS. (C) Blood smear of the patient 5 years after total splenectomy, stable and without need of transfu-
sions since the time of that surgery. Moderate anisopoikilocytosis and several spherocytes are still noted. Scale bar = 14 mm. Details on
the ektacytometry assay and parameters can be found at https:​/​/www​.cincinnatichildrens​.org​/service​/c​/cancer​-blood​/hcp​/clinical​
-laboratories​/erythrocyte​-diagnostic​-lab​/ektacytometry.

332  |  Hematology 2021  |  ASH Education Program

impres­sively regrown to 435 g, based on the pathol­ogy report.
The patient has had no fur­ther trans­fu­sion require­ment for the past
6 years, with a Hgb now in the 12.7 to 16.7 g/dL range, an abso­lute
retic­u­lo­cyte count (ARC) of 80 to 150 × 103/µL, and a resolved iron
over­load. Ektacytometry revealed the typ­i­cal HS curve (Figure 1A);
a blood smear at 5.5 years of age (before the total sple­nec­tomy)
dem­ on­ strated many spherocytes and sig­ nif­i­
cant poikilocytosis
and polychromasia, as expected with α-spectrin defi­ciency (Fig-
ure 1B).6 A recent blood smear, 5 years after the total sple­nec­tomy,
showed an improved RBC mor­phol­ogy, although still with many
spherocytes and mod­er­ate poikilocytosis (Figure 1C).
Hereditary spherocytosis (HS)
Genotype/phe­no­type cor­re­la­tions
HS, the most com­mon of the RBC mem­brane dis­or­ders, is caused
by muta­tions in the genes SPTA1, SPTB, ANK1, EPB42, or SLC4A1,
lead­ing to an RBC mem­brane skel­e­ton defi­cient in α- or β-spectrin,
ankyrin, pro­tein 4.2, or band 3, respec­tively.6-9 These pro­teins
build the scaf­fold and the ver­ti­cal con­nec­tions of the RBC mem­
brane skel­e­ton with the lipid mem­brane; their defi­ciency allows
mem­ brane loss and, con­ se­
quently, the for­ ma­tion of sphero-
cytes.6,9 Patients with auto­ so­mal dom­ i­
nant (AD) HS due to
SLC4A1, SPTB, or ANK1 muta­tions or AR EPB42-HS pres­ent with a
range of well-com­pen­sated hemo­ly­sis up to a mod­er­ately severe
ane­mia (Hb <8 g/dL) with brisk reticulocytosis (>10%).6,10 Severe
AR HS typ­i­cally pres­ents as a trans­fu­sion-depen­dent dis­ease from
early infancy and is most fre­quently due to a null SPTA1 var­i­ant in
trans to an SPTA1 allele with the αLEPRA poly­mor­phism, allowing for
a small resid­ual expres­sion of nor­mal α-spectrin, as in the case
above.2,3 Rarely, a sim­i­lar phe­no­type of AR HS may be due to a null
ANK1 muta­tion in trans to a pro­moter or mis­sense var­i­ant of ANK1
decreas­ing but not oblit­er­at­ing the incor­po­ra­tion of ankyrin into
the mem­brane skel­e­ton.11 With the advent of intra­uter­ine trans­
fu­sions for severe pre­na­tal ane­mia, we now see the most severe
HS cases, which used to be embry­o­nal-lethal dur­ing the third tri­
mes­ter of preg­nancy: these are cases due to biallelic SPTA1 null
var­i­ants or homo­zy­gous SLC4A1 var­i­ants,3,12 which would remain
trans­fu­sion-depen­dent even after sple­nec­tomy.
Clinical man­age­ment
Neonatal hyperbilirubinemia, presenting as early as within the
first 24 hours of life, is fre­quently the presenting sign of HS of any
sever­ity. Since fast-ris­ing uncon­ju­gated bil­i­ru­bin poses a risk of
kernicterus, appro­pri­ate mon­i­tor­ing and treat­ment with photo-
therapy or, rarely, with exchange trans­fu­sion is needed. A trans­
fu­sion require­ment in mild or mod­er­ate spherocytosis may not
develop for 2 to 3 weeks after birth, when the nor­mal neo­na­tal
hyposplenism resolves. It is dif­fi­cult to pre­dict the phe­no­type of
HS dur­ing the first year of life: trans­fu­sion depen­dency may con­
tinue dur­ing the first 9 months of life due to delayed eryth­ro­poi­e­
tin response.13 During that time, eval­u­a­tion for the genetic cause
of the dis­ease is attrac­tive for phy­si­cians and fam­i­lies, to inform
prog­no­sis. Infants with an AD form of HS may ben­e­fit from eryth­
ro­poi­e­tin admin­is­tra­tion to wean off trans­fu­sions faster.14
It is not rare for patients with mild HS who have com­pen­
sated hemo­ly­sis to pres­ent later in life, after a viral infec­tion, such
as with Epstein-Barr virus or par­vo­vi­rus B19; with a hemo­lytic or
aplastic cri­sis; or with gall­stones or spleno­meg­aly on phys­i­cal
exam. Occasionally, tod­dlers are diag­nosed with HS after refer­ral
Prakash Singh Shekhawat
to a hema­tol­o­gist because their Hgb does not improve with iron
sup­ple­men­ta­tion—one of the bright exam­ples of why a com­
plete blood count (CBC) for eval­u­a­tion of ane­mia should always
be asso­ci­ated with a retic­u­lo­cyte count.
Spherocytes are pre­ma­turely destroyed in the spleen as they
are impeded in pass­ ing through the interendothelial sinu­ soi­
dal slits and even­tu­ally phago­cy­tosed by the red pulp mac­ro­
phages. Therefore, sple­nec­tomy has been used for decades as
the stan­dard of care for HS, with the aim of decreas­ing hemo­ly­
sis and improv­ing ane­mia. However, increased aware­ness of the
postsplenectomy increased risks of sep­sis; par­a­sitic infec­tions
of eryth­ro­cytes; and, pos­si­bly, increased thrombophilia and ath­
ero­scle­ro­sis due to the rise in cho­les­terol,15,16 at least par­tially
due to ame­lio­ra­tion of the increased eryth­ro­poi­e­sis, along with
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/331/1851848/331kalfa.pdf by guest on 13 December 2021
accu­mu­la­tion of fol­low-up data on the effi­cacy of par­tial sple­nec­
tomy,17,18 has led to a more indi­vid­u­al­ized approach rec­og­niz­ing
the options of total vs par­tial vs no sple­nec­tomy. Therefore, it
is cur­rently rec­og­nized that sple­nec­tomy is not needed for all­
patients with HS. A form of sple­nec­tomy is indi­cated for patients
with HS who have severe ane­ mia requir­
ing fre­ quent trans­ fu­
sions or hav­ing com­pro­mised qual­ity of life; par­tial sple­nec­tomy
should be strongly con­sid­ered as an option when the patient is
youn­ger than 5 years of age and for AD HS.5
In the clin­i­cal case above, in which the genetic diag­no­sis con­
firmed an SPTA1 null muta­tion in trans to αLEPRA, known to cause
severe trans­fu­sion-depen­dent spherocytosis, sple­nec­tomy was
the expected course of action. However, the addi­tional find­
ing of the PIEZO1 VUCS com­pli­cated the deci­sion: if this var­i­
ant was path­o­genic, an addi­tional com­po­nent of HX could be
caus­ing intra­vas­cu­lar hemo­ly­sis, lead­ing to an increased risk
of postsplenectomy thrombophilia.19 This is not an uncom­mon
sce­nario since PIEZO1 has numer­ous VUCS. Parental stud­ies to
sep­a­rate the var­i­ants and exam­ine the asso­ci­ated phe­no­type by
ektacytometry and RBC cat­ion con­tent are help­ful to guide clin­
i­cal deci­sions, as in our case. Given the con­cern regard­ing the
patient’s increas­ing iron over­load despite che­la­tion, the fam­ily
and the pri­mary hema­tol­o­gist decided to pur­sue sple­nec­tomy
by the time he was 4. Thus, par­tial sple­nec­tomy was performed
to mit­i­gate hemo­ly­sis while pre­serv­ing splenic immune func­
tion.5,7,17,18,20 Although par­ tial sple­
nec­tomy (leav­ ing either the
supe­rior or infe­rior pole of the spleen) in HS is fre­quently a per­
ma­nent solu­tion,5 fol­low-up sur­gery with total sple­nec­tomy may
be needed, espe­cially in cases of SPTA1-asso­ci­ated AR HS.
Notable excep­ tions of sple­ nec­tomy effi­
cacy are the rare
cases due to a com­plete defi­ciency of α-spectrin or band 3, who
require either a life­long pro­gram of trans­fu­sions and iron che­la­
tion or stem cell transplantion.3,12
Not all­spherocytes are due to HS. The his­tory, blood smear,
and ektacytometry in non-trans­fu­sion-depen­dent con­gen­i­tal
dyserythropoietic ane­mia type II (CDA-II) may closely resem­
ble HS (Figure 2).4,21 Suboptimal reticulocytosis and high or
high-nor­mal fer­ri­tin val­ues should trig­ger a genetic workup to
explore this dif­fer­en­tial diag­no­sis. In addi­tion, warm auto­im­
mune hemo­lytic ane­mia fre­quently causes enough RBC mem­
brane loss to resem­ ble spherocytosis, both in eryth­ ro­
cyte
mor­phol­ogy and ektacytometry. Therefore, the first tests rec-
ommended in the workup of a newly presenting hemo­ly­sis are
direct and indi­rect anti­glob­u­lin tests (DAT and IAT) to eval­u­ate
for the pos­si­bil­ity of an immune-medi­ated cause of hemo­lytic
ane­mia since this rad­i­cally alters man­age­ment (Figure 3).
Red cell mem­brane dis­or­ders  |  333

Figure 2. CDA-II is in the differential diagnosis for HS. A 4-year-
old African American girl presented with DAT-negative, mild
hemolytic anemia (Hgb, 10 g/dL) with a reticulocyte count of
2.4%, a normal ARC of 86 × 103/µL, and mild jaundice. She had a
history of prolonged neonatal jaundice treated with photother-
apy and no transfusion requirement. Ektacytometry showed a
curve that resembled HS, and a blood smear showed sphero-
cytes, marked poikilocytosis, and no polychromasia. Inade-
quate reticulocytosis and a ferritin of 80 ng/mL, at a generous
level for her age with no concurrent inflammation, triggered
further evaluation with sequencing of an HHA gene panel that
revealed 2 SEC23B mutations, c.40C>T (p.R14W) and c.367-
3A>G. Follow-up targeted sequencing of her parents con-
firmed that these 2 variants were in trans, causing CDA-II. Scale
bar = 14 mm.
Hereditary elliptocytosis and hered­i­tary
pyropoikilocytosis (HE/HPP)
Genotype/phe­no­type cor­re­la­tions
Hereditary elliptocytosis (HE) is caused by het­ero­zy­gous muta­
tions in SPTA1, SPTB, and EPB41, the genes encoding α-spectrin,
β-spectrin, and pro­tein 4.1R, respec­tively. HE-caus­ing SPTA1 and
SPTB var­i­ants are typ­i­cally located in the spectrin tetrameriza-
tion domain or, less fre­quently, along the chain or close to the
junc­tional com­plex and pro­duce a qual­i­ta­tive rather than quan­ti­
ta­tive defect in the spectrin tet­ra­mer; the altered spectrin chains
334  |  Hematology 2021  |  ASH Education Program
incor­po­rate into the mem­brane skel­e­ton and weaken its hor­
i­zon­tal scaf­fold.9,22 These var­i­ants have increased prev­a­lence in
malaria-endemic regions (up to 3% in West Africa), supporting
the hypoth­e­sis that elliptocytosis con­fers a sur­vival advan­tage
to malaria. EPB41 muta­tions are respon­si­ble for fewer than 15%
of HE cases; they lead to altered or defi­cient pro­tein 4.1R, com­
pro­mis­ing the spectrin-4.1R-actin asso­ci­a­tion at the junc­tions.7,23
Very rare cases of mild HE have been described due to homo­
zy­gous non­sense muta­tions in GYPC, caus­ing a com­plete defi­
ciency of glycophorin C and, con­se­quently, a par­tial defi­ciency
of pro­tein 4.1R in the junc­tional com­plex.
HE is char­ac­ter­ized by ellip­ti­cally shaped RBCs and is fre­
quently asymp­tom­atic with min­i­mal or no hemo­ly­sis, nor­mal
Hgb, and a nor­mal or bor­der­line high retic­u­lo­cyte count. Nev-
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/331/1851848/331kalfa.pdf by guest on 13 December 2021
ertheless, patients may pres­ent with exac­er­bated neo­na­tal jaun­
dice or develop epi­sodic hemo­ly­sis with illnesses that cause
hyper­pla­sia of the retic­u­lo­en­do­the­lial sys­tem, eg, viral hep­a­ti­tis,
infec­tious mono­nu­cle­o­sis, bac­te­rial infec­tions, and malaria.7,23
Hereditary pyropoikilocytosis (HPP) is char­ ac­ter­
ized by
in­
creased poikilocytosis, RBC frag­ men­ ta­
tion, and elliptocytes
on the periph­eral smear (Figure 4). The most com­mon form of
HPP is diag­nosed in infants of Afri­can ances­try, who pres­ent with
neo­na­tal jaun­dice and fre­quently a trans­fu­sion require­ment, and
is due to an SPTA1 HE-caus­ing muta­tion in trans to the SPTA1
splic­ing var­ i­
ant c.6531-12C>T, known as αLELY (low expres­ sion
LYon).23,24 αLELY causes a 50% decrease in α-spectrin expres­sion
but is clin­i­cally silent in nor­mal indi­vid­u­als even in the homo­zy­
gous state since α-spectrin is pro­duced well in excess. However,
in trans to an HE-caus­ing SPTA1 muta­tion, αLELY leads to an HPP
phe­no­type because it allows for increased rel­a­tive incor­po­ra­tion
of the abnor­mal spectrin chain into the mem­brane skel­e­ton. Of
note, αLELY has a minor allele fre­quency of up to 25.5% (gnomad​
­.broadinstitute​­.org); there­fore, it is fairly com­mon for an indi­vid­
ual with non­he­mo­lytic HE to have a child with HPP.
HPP can also be caused by homo­zy­gous or com­pound het­ero­
zy­gous HE-caus­ing var­i­ants of SPTA1, SPTB, or EPB41 or a com­bi­
na­tion of SPTA1 and SPTB alleles with HE muta­tions.23,24 Hemolysis
and poikilocytosis per­sist in those patients (true HPP), while the
sever­ity of the dis­ease may vary from com­pen­sated hemo­ly­sis
(eg, a case with com­bined SPTA1 p.L155dup and SPTB p.W2024R)
up to trans­fu­sion depen­dency unre­spon­sive to sple­nec­tomy (eg,
a child with com­pound het­ero­zy­gos­ity for spectrin St. Claude
SPTA1 c.2806-13T>G,24 pro­duc­ing a trun­cated α-spectrin pro­tein
and a novel non­sense SPTA1 muta­tion p.R118* in trans).8 An exam­
ple of true HPP due to homo­zy­gous SPTB p.F2014V muta­tion with
trans­fu­sion depen­dency that responded well to sple­nec­tomy
is detailed in Figure 5, to be contrasted with the exam­ple case
of infan­tile HPP in Figure 4. My lim­ited per­sonal expe­ri­ence with
these rare cases sug­gests that reticulocytopenia while on trans­
fu­sions in severe HHA, such as the reces­sively inherited SPTA1-
asso­ci­ated HS or severe true HPP, pre­dicts a poor response to
sple­nec­tomy. A bet­ter-informed ratio­nal approach to treat­ment
is being devel­oped, as we are gaining more insights into the spe­
cific genetic muta­tions caus­ing these dis­eases.
Clinical man­age­ment
HE is fre­quently diag­nosed inci­den­tally on a blood smear review
or dur­ing inves­ti­ga­tion for gall­stones or spleno­meg­aly or in a
par­
ent of an infant with neo­ na­
tal jaun­
dice diag­
nosed with
HPP. Infants with the com­mon form of HPP (SPTA1 HE-var­i­ant in
 Hemolysis positive Consider physical agents
DAT and Work-up for autoimmune
(with or without anemia) including
reticulocytes, indirect Coombs or alloimmune causes
MAHA/HUS/aHUS/TTP
unconjugated bilirubin,
haptoglobin
negative
Consider PNH
Globin gene and Consider
Recently or yes
Family h istory HHA/CDA panel or subpanels Wilson disease
chronically
and depending on family history and
transfused patient
blood smear r eview smear review
no
of patient and parents
to assist with diagnosis abnormal
Hemoglobin Globin gene sequencing and
electrophoresis deletion/duplication analysis

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/331/1851848/331kalfa.pdf by guest on 13 December 2021

Suboptimal normal
reticulocytosis,
Ektacytometry abnormal
iron overload,
or o smotic fragility RBC membrane disorder
skeletal
and/or EMA flow cytometry
abnormalities
normal
or

RBC membrane disorders gene panel,

abnormal RBC enzymopathies
CDA gene RBC enzyme activity especially if
panel gene panel splenectomy is contemplated
normal

binuclear or multinuclear erythroblasts Untypable CDA

Bone marrow studies (consider CDA Registries)
normal except erythroid hyperplasia
Figure 3. Proposed algorithm for laboratory workup of a patient presenting with hemolysis with or without anemia. In many cases of
mild HS and most cases of PIEZO1-associated HX, anemia may be well compensated by reticulocytosis. Although our focus here is on
RBC membrane disorders, the differential includes other causes of hemolytic disorders that are mentioned in this algorithm. Evaluation
for autoimmune or, especially in an infant, alloimmune hemolytic anemia with DAT and IAT is the first testing recommended since such
a diagnosis is frequently acute and evolving, requiring immediate action. Of note, warm autoimmune hemolytic anemia in children and
occasionally adults with underlying immune dysregulation may be atypical and conventionally DAT-negative.52-54 Consideration should
also be given to the possibility of MAHA, PNH, and Wilson disease. The cases described in this review, especially that presented in
Figure 6, demonstrate utilization of this algorithm. A blood smear review of the patient and parents and attention to the RBC indices
­including MCV, MCHC, and RDW, along with hemolytic markers (unconjugated bilirubin, lactate dehydrogenase, h ­ aptoglobin—of
note, haptoglobin is reliable after 6 months of life since earlier it may be low due to decreased production by the ­infant’s liver rather
than ­increased consumption) and ferritin and transferrin saturation to consider iron-loading inefficient ­erythropoiesis, can ­provide hints
as to the differential diagnosis. In a non-chronically transfused patient, we suggest phenotypic evaluation considering the ­differential
of globin disorders, followed by RBC membranopathies and enzymopathies. Rare causes of HHA such as unstable Hgb disorders and
CDAs also need to be considered. When suboptimal reticulocytosis or iron overload or skeletal abnormalities are noted in a patient
with hemolytic anemia, the possibility of CDA should be considered and pursued. The combination of blood smear review and osmotic
gradient ektacytometry frequently helps to narrow the differential while alerting clinicians of rare possibilities. Osmotic gradient ekta-
cytometry evaluates the deformability of RBCs as they are subjected to constant shear stress in a medium of increasing osmolality in a
laser diffraction viscometer and is the reference technique for differential diagnosis of erythrocyte membrane and hydration disorders
when a recent transfusion does not interfere with phenotypic evaluation of the patient.4 Flow cytometry with eosin-5′-maleimide bind-
ing of band 3 and Rh-related proteins is a rapid screening test for RBC membrane disorders characterized by membrane loss.55 Osmot-
ic fragility is increased in HS and expected to be decreased in HX (however, it is reported as normal in patients with KCNN4 Arg352His
mutation31). When the patient is recently or chronically transfused, as is typically the case for infants with HHA, genetic evaluation
with clinically available NGS panels or research-based whole-exome sequencing or whole-genome sequencing may provide an accu-
rate diagnosis necessary for appropriate management decisions. Laboratories that offer sequencing on genes or panels associated
with HHAs and red cell membrane disorders can be found by searching the ­Genetic Testing R ­ egistry: https:​/​/www​.ncbi​.nlm​.nih​.gov​/
gtr​/. e.g. https:​/​/www​.ncbi​.nlm​.nih​.gov​/gtr​/all​/tests​/​?term=RBC%20membrane%20disorders. aHUS, atypical ­hemolytic uremic syn-
drome; HUS, hemolytic uremic syndrome; MAHA, microangiopathic hemolytic anemia; PNH, paroxysmal nocturnal hemoglobinuria;
RDW, RBC distribution width; TTP, thrombotic thrombocytopenic purpura.

Prakash Singh Shekhawat

Red cell mem­brane dis­or­ders  |  335
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/331/1851848/331kalfa.pdf by guest on 13 December 2021
Figure 4. Transient infantile HPP. An African American girl presented within the first day of life after delivery at term with nonimmune
hemolytic anemia and neonatal jaundice. She was treated with triple phototherapy but did not require RBC transfusion. (A) The blood
smear of the patient with HPP showing marked anisocytosis and poikilocytosis with bizarre microcytes and fragmented cells along
with elliptocytes (scale bar = 14 µm). (B) The blood smear of the mother, who had a normal Hgb and reticulocyte count, was notable
for elliptocytes indicating HE. (C) Ektacytometry showed the trapezoid curve characteristic for HE and HPP with decreased deform-
ability for both mother and patient. The patient’s sample was sequenced by NGS on an RBC membrane gene panel, which revealed
the SPTA1 HE-causing variant c.460_463insTTG (p.L155dup) as well as the SPTA1 c.6531-12C>T polymorphism known as αLELY. Targeted
sequencing for these variants revealed that the mother was heterozygous for SPTA1 p.L155dup. The mother did not carry αLELY, which
was presumably inherited by the father. The compound heterozygosity of an HE-causing SPTA1 mutation in trans to αLELY confirmed
the diagnosis of infantile HPP. The child continued to have a mild anemia with reticulocytosis (Hgb, 10.2 g/dL with 3.8% reticulocytes)
and poikilocytosis, with RDW up to 18.7% at her 2-year-old follow-up, but a CBC with reticulocyte count and blood smear at 5 years
of age indicated transition to a nonhemolytic HE phenotype. RDW, RBC distribution width.

trans to αLELY) may have mod­er­ate to severe hemo­lytic ane­mia
requir­ing fre­quent trans­fu­sions early in life, but the phe­no­type
improves after the first 1 to 2 years, evolv­ing to typ­i­cal HE.7 HPP
due to biallelic HE-caus­ing muta­tions may pres­ent with mild to
severe chronic ane­mia with long-term trans­fu­sion depen­dence.
Most but not all­of the cases with severe hemo­ly­sis and a chronic
trans­fu­sion require­ment respond to sple­nec­tomy.8,24
Hereditary xerocytosis (HX)
Genotype/phe­no­type cor­re­la­tions
HX, also called dehydrated hered­i­tary stomatocytosis, is an AD
dis­ease caused by muta­tions in PIEZO1 or KCNN4,25-32 cod­ing
respec­ tively for the mechanosensitive cat­ ion chan­ nel PIEZO1
and the cal­ cium ion-acti­ vated potas­ sium (K+) chan­nel known
as the Gardos chan­nel. HX RBCs have an abnor­mal K+ leak not
com­pen­sated by a pro­por­tional intra­cel­lu­lar sodium (Na+) gain,
lead­ing to cel­ lu­
lar dehy­ dra­ tion. Pseudohyperkalemia may be
noted if anal­y­sis of elec­tro­lytes in blood spec­i­mens is delayed.
HX is char­ac­ter­ized by intra­vas­cu­lar hemo­ly­sis and intrahepatic
RBC destruc­tion that per­sists after sple­nec­tomy (there­fore,
splenec­tomy is not only dan­ger­ous but also inef­fec­tive as an
HX treat­ment) as well as iron over­load dis­pro­por­tion­ate to the
trans­fu­sion his­tory.19,33 Significant phe­no­typic var­i­abil­ity exists,
prob­a­bly due to the large num­ber of caus­at­ive var­i­ants, coin-
heritance of var­i­ants in other genes affect­ing RBC phe­no­type,34
and mul­ti­ple mech­a­nisms con­trib­ut­ing to the path­o­gen­e­sis.35
HX patients may pres­ent with mild to mod­er­ate hemo­lytic ane­
mia, usu­ally due to an aminoterminal PIEZO1 muta­tion or KCNN4
var­i­ants, or, with fully com­pen­sated hemo­ly­sis, due to the most
com­mon carboxyterminal PIEZO1 var­i­ants (Figure 6).19 Despite
the RBC dehy­dra­tion, mean cel­lu­lar vol­ume (MCV) is high nor­mal
to nor­mal in HX, likely due to the effects of the mutated chan­
nels in eryth­ro­poi­e­sis.36 Cases with poly­cy­the­mia despite hemo­
ly­sis have also been reported,37 pointing to the hypoth­e­sis that
patients with HX may have increased eryth­ro­poi­etic drive for
their level of Hgb because of rel­at­ively low 2,3-DPG lev­els and

336  |  Hematology 2021  |  ASH Education Program


Figure 5. Persistent (true) HPP. A boy of Iranian descent present-
ed with severe anemia since infancy. He remained transfusion
dependent; therefore, his erythrocyte phenotype was not eval-
uable. (A) A blood smear of the father (practically identical with
the blood smear of the mother) and (B) ektacytometry show-
ing a trapezoid shape typical for HE for both parental samples
indicated that they both had elliptocytosis. NGS on a panel of
RBC membrane disorder genes revealed that the patient was
homozygous for a novel missense mutation in the SPTB gene
(c.6040T>G, p.F2014V), affecting the spectrin self-association
site. Both parents were heterozygous for the same mutation.24
con­se­quently high Hb-oxy­gen affin­ity and exac­er­bated tis­sue
and renal hyp­oxia.38,39 Hepatic and/or myo­car­dial iron over­load
may be the presenting sign for some patients when not diag­
nosed before adult­hood. Evaluating for hemo­chro­ma­to­sis gene
muta­tions is recommended since this may alter clin­i­cal prog­no­
sis and care. Iron over­load may also be accel­er­ated in cases of
KCNN4-HX,19 as well as in cases of aminoterminal PIEZO1 path­o­
log­i­cal var­i­ants such as the p.V598M (per­sonal com­mu­ni­ca­tion
on 2 patients in their early 20s with pref­er­en­tial and severe myo­
car­dial iron over­load).
Prakash Singh Shekhawat
Clinical man­age­ment
Genetic coun­sel­ing and close mon­i­tor­ing of preg­nan­cies should
be pro­vided when either par­ent has HX because of the risk of
peri­na­tal edema and non­im­mune hydrops fetalis.19
Splenectomy is contraindicated in HX due to PIEZO1 muta­
tions because it pre­dis­poses patients to life-threat­en­ing venous
and arte­rial throm­bo­em­bolic com­pli­ca­tions, reported to hap­pen
within a month up to 28 years post sple­nec­tomy.19,40,41 Although
no throm­botic com­pli­ca­tions were observed after sple­nec­tomy
in 12 patients with KCNN4-asso­ci­ated HX after fol­low-up for 2 to
44 years,19,31 sple­nec­tomy should be avoided any­way since it pro­vi­
des no ther­a­peu­tic ben­e­fit in either PIEZO1- or KCNN4-asso­ci­ated
HX because intra­vas­cu­lar and intrahepatic RBC destruc­tion per­
sists after sple­nec­tomy.33
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/331/1851848/331kalfa.pdf by guest on 13 December 2021
Management is cur­rently focused on sup­port­ive care for ane­
mia and hemo­ly­sis com­pli­ca­tions and on pre­ven­tive mon­i­tor­ing
to quan­ti­tate and treat iron over­load as needed. Screening with
fer­ri­tin and trans­fer­rin sat­u­ra­tion should be followed (even with a
mod­er­ate increase in fer­ri­tin) by T2* mag­netic res­o­nance imag­
ing of the liver and heart. Iron over­load is much more eas­ily and
effec­tively treated with che­la­tion when diag­nosed early. Phle-
botomy has also been used to address HX-asso­ci­ated iron over­
load. However, based on the fact that hemo­chro­ma­to­sis in HX
is mul­ti­fac­to­rial and includes exces­sive eryth­ro­poi­e­sis caus­ing
increased erythroferrone that suppresses hepcidin as well as the
direct sup­pres­sion of hepcidin,35,42 the effi­cacy of phle­bot­omy to
treat iron over­load in HX needs fur­ther eval­u­a­tion.43
Specific med­i­ca­tions for the treat­ment of HX are not clin­i­cally
avail­­able. A prom­is­ing option may be a spe­cific inhib­i­tor for the
Gardos chan­nel, senicapoc, since KCNN4 is the final com­mon
medi­a­tor of RBC dehy­dra­tion in HX.8,44 Senicapoc was shown to
be well tol­er­ated in a phase 3 trial in sickle cell dis­ease, where
it reduced RBC dehy­dra­tion and hemo­ly­sis and increased Hgb
lev­els.45 A phase 1/2 “proof-of-con­cept study of senicapoc in
patients with famil­ ial dehydrated stomatocytosis caused by
the V282M muta­tion in the Gardos (KCNN4) chan­nel” has been
posted on clinicaltrials​­.gov and is cur­rently recruiting.
Overhydration syn­dromes
Genotype/phe­no­type cor­re­la­tions
Genetic eti­­ol­ogy in overhydration syn­dromes is var­i­able, along
with a phe­no­type of mild to severe hemo­lytic ane­mia.46,47 Blood
smear is remark­able for abun­dant stomatocytes; RBC phys­i­­ol­ogy
is char­ac­ter­ized by a large net increase in intra­cel­lu­lar Na+ and
water not ade­quately com­pen­sated by a decrease in K+. MCV is
typ­i­cally increased, and mean cor­pus­cu­lar hemo­glo­bin con­cen­
tra­tion (MCHC) is decreased. Osmotic fra­gil­ity is increased, and
ektacytometry is right-shifted.
The most severe form of overhydrated hered­i­tary stomatocyto-
sis is caused by het­ero­zy­gous mis­sense muta­tions in RHAG, the
gene cod­ing for the red cell ammo­nium trans­porter rhesus-
asso­ci­ated gly­co­pro­tein; both muta­tions described so far
(p.Phe65Ser and p.Ile61Arg) alter amino acids lin­ing the puta­tive
trans­port chan­nel, mak­ing it per­me­able to Na+ ions.48-50
Cryohydrocytosis, Southeast Asian ovalocytosis, and band 3
Ceinge are caused by cer­tain het­ero­zy­gous muta­tions in SLC4A1,
which alter the band 3 chan­nel to medi­ate cat­ion leak in cold tem­
per­a­tures.46,47 Patients with stomatin-defi­cient ­cryohydrocytosis,
caused by muta­tions in SLC2A1 cod­ing for glu­cose trans­porter
type 1, pres­ent with mod­er­ate hemo­lytic ane­mia, cat­a­racts due
Red cell mem­brane dis­or­ders  |  337
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/331/1851848/331kalfa.pdf by guest on 13 December 2021
Figure 6. HX due to heterozygous PIEZO1 mutation. A European American girl, born at term, developed neonatal hyperbilirubinemia
that prevented discharge from the nursery for 4 days while receiving treatment with phototherapy. Scleral icterus was again noted
at 10 months of age, and a CBC indicated reticulocytosis but a normal Hgb. She continued to have chronic hemolysis and jaundice
and had an extensive workup including negative DAT, normal Hgb electrophoresis, RBC enzyme activity testing showing normal or
increased activity, osmotic fragility testing that was (probably mistakenly) reported as normal rather than as decreased, alpha and
beta globin genetic testing that revealed no mutation or copy number variation, a negative PNH screen by flow cytometry, and bone
marrow studies showing erythroid hyperplasia, no dyserythropoiesis, and markedly increased iron stores. (A) At 10 years of age, she
had osmotic gradient ektacytometry performed that demonstrated a typical HX curve, with left shift due to decreased Omin and
Ohyp (Figure 2A). The CBC at the time indicated an Hgb of 14.2 g/dL, an MCV of 96 fl, an MCH of 36.5 pg, an MCHC of 37.9 g/dL, a
reticulocyte count 16.9%, and an ARC of 643 × 103/µL. (B) A blood smear was significant for polychromasia, macrocytosis, and occa-
sional stomatocytes, target cells, and dense fragmented cells. Of note, sometimes the blood smear may be deceptively normal, with
only a few target cells and no stomatocytes. A determination of the RBC intracellular cation confirmed a reduced K+ content without
a corresponding increase in Na+ content. NGS for an RBC membrane disorder panel revealed 1 of the most common PIEZO1 variants
(c.7367G>A, p.R2456H) causing HX. PNH, paroxysmal nocturnal hemoglobinuria.

to the altered cat­ion per­me­abil­ity of the lens, and neu­ro­log­i­cal ulation should be con­sid­ered in patients who have inad­ver­tently
dis­or­ders caused by disrupted glu­cose trans­port. been splenectomized.

Clinical man­age­ment Conclusions

Management is sup­port­ive for ane­mia and hemo­ly­sis com­pli­ Decades of research on the molec­ u­
lar path­
o­gen­e­
sis of RBC
ca­tions. Splenectomy is usu­ally par­tially effec­tive in reduc­ing mem­brane dis­or­ders have cul­mi­nated,9,51 upon the advent of
hemo­ly­sis, but it car­ries a high risk of throm­bo­em­bolic com­pli­ NGS over the last 10 years, in a high suc­cess rate of genetic diag­
ca­tions and pul­mo­nary hyper­ten­sion.40,41 Prophylactic anticoag- no­sis for patients with such dis­eases. Genetic diag­no­sis is highly

338  |  Hematology 2021  |  ASH Education Program

ben­e­fi­cial for trans­fu­sion-depen­dent patients and likely pru­dent
to con­sider before pro­ceed­ing to the irre­vers­ible treat­ment of
sple­nec­tomy, which is not always effec­tive or safe. Increased
infor­ma­tion on geno­type-phe­no­type cor­re­la­tion along with the
cur­rent wider avail­abil­ity of par­tial sple­nec­tomy calls for a new
con­sen­sus state­ment of experts in the field on updated man­age­
ment guide­lines.
Acknowledgments  14. Tchernia G, Delhommeau F, Perrotta S, et al; Euro­pean Society for Paediat-
This work was supported by National Heart, Lung, and Blood In-
stitute grant R01 HL152099.
We are grate­ful to the patients, fam­i­lies, and refer­ring phy­
si­cians par­tici­pat­ing in our diag­nos­tic stud­ies of hered­i­tary
hemo­lytic ane­mia and to the teams of the Cincinnati Children’s
Erythrocyte Diagnostic Laboratory and Molecular Genetics Lab-
oratory for their con­tri­bu­tion to the diag­nos­tic stud­ies of cases
presented in the text and fig­ures.
Clinical out­comes reported in some of these cases are
based upon per­sonal com­mu­ni­ca­tion with Drs. Marya Alman-
soori, Jeanne Boudreaux, Mor­gan McLemore, Charles Quinn, and
Audrey Woerner.
Conflict-of-inter­est dis­clo­sure
Theodosia A. Kalfa: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Theodosia A. Kalfa: nothing to disclose.
Correspondence
Theodosia A. Kalfa, Cancer and Blood Diseases Institute, Cincin-
nati Children’s Hospital Medical Center, 3333 Burnet Ave, MLC
7015, Cincinnati, OH 45229-3039; e-mail: theodosia​­.kalfa@cchmc​
­.org.
References
1. Wichterle H, Hanspal M, Palek J, Jarolim P. Combination of two mutant
alpha spectrin alleles under­lies a severe spherocytic hemo­lytic ane­mia. J
Clin Invest. 1996;98(10):2300-2307.
2. Gallagher PG, Maksimova Y, Lezon-Geyda K, et al. Aberrant splic­ing con­
trib­utes to severe α-spectrin-linked con­gen­i­tal hemo­lytic ane­mia. J Clin
Invest. 2019;129(7):2878-2887.
3. Chonat S, Risinger M, Sakthivel H, et  al. Corrigendum: the spec­trum of
SPTA1-asso­ci­ated hered­i­tary spherocytosis. Front Physiol. 2019(18 Octo­
ber);10:1331.
4. Da Costa L, Suner L, Galimand J, et al; Society of Hematology and Pediatric
Immunology Group; French Society of Hematology. Diagnostic tool for red
blood cell mem­brane dis­or­ders: assess­ment of a new gen­er­a­tion ektacy-
tometer. Blood Cells Mol Dis. 2016;56(1):9-22.
5. Rothman JA, Stevens JL, Gray FL, Kalfa TA. How I approach hered­i­tary hemo­
lytic ane­mia and sple­nec­tomy. Pediatr Blood Cancer. 2020;67(11):e28337.
6. Eber S, Lux SE. Hereditary spherocytosis—defects in pro­teins that con­nect
the mem­brane skel­e­ton to the lipid bila­yer. Semin Hematol. 2004;41(2):118-
141.
7. Mohandas N. Inherited hemo­lytic ane­mia: a pos­ses­sive begin­ner’s guide.
Hematology Am Soc Hematol Educ Program. 2018;2018(1):377-381.
8. Risinger M, Kalfa TA. Red cell mem­brane dis­or­ders: struc­ture meets func­
tion. Blood. 2020;136(11):1250-1261.
9. Lux SE. Disorders of the red cell mem­brane. In: Orkin SH, Fisher DE, Gins-
burg D, et al, eds. Nathan and Oski’s Hematology and Oncology of
Infancy and Childhood. Saunders Elsevier; 2015:515-579.
10. Kalfa TA, Connor JA, Begtrup AH. EPB42-related hered­i­tary spherocytosis. In:
Adam MP, Ardinger HH, Pagon RA, et al, eds. GeneReviews((R)). University of
Washington; 2016. https://www.ncbi.nlm.nih.gov/books/NBK190102/
Prakash Singh Shekhawat
11. Eber SW, Gonzalez JM, Lux ML, et  al. Ankyrin-1 muta­ tions are a major
cause of dom­i­nant and reces­sive hered­i­tary spherocytosis. Nat Genet.
1996;13(2):214-218.
12. Kager L, Bruce LJ, Zeitlhofer P, et al. Band 3 nullVIENNA, a novel homo­zy­
gous SLC4A1 p.Ser477X var­i­ant caus­ing severe hemo­lytic ane­mia, dyseryth-
ropoiesis and com­plete dis­tal renal tubu­lar aci­do­sis. Pediatr Blood Cancer.
2017;64(3):e26227.
13. Delhommeau F, Cynober T, Schischmanoff PO, et  al. Natural his­tory of
hered­i­tary spherocytosis dur­ing the first year of life. Blood. 2000;95(2):393-
397.
ric Haematology and Immunology Working Group on Hemolytic Anemias.
Recombinant eryth­ro­poi­e­tin ther­apy as an alter­na­tive to blood trans­fu­
sions in infants with hered­i­tary spherocytosis. Hematol J. 2000;1(3):146-
152.
15. Crary SE, Buchanan GR. Vascular com­ pli­
ca­
tions after sple­ nec­tomy for
hema­to­logic dis­or­ders. Blood. 2009;114(14):2861-2868.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/331/1851848/331kalfa.pdf by guest on 13 December 2021
16. Schilling RF, Gangnon RE, Traver MI. Delayed adverse vas­ cu­
lar events
after sple­nec­tomy in hered­i­tary spherocytosis. J Thromb Haemost. 2008;
6(8):1289-1295.
17. Englum BR, Rothman J, Leonard S, et al; Splenectomy in Congenital Hemo-
lytic Anemia Consortium. Hematologic out­comes after total sple­nec­tomy
and par­tial sple­nec­tomy for con­gen­i­tal hemo­lytic ane­mia. J Pediatr Surg.
2016;51(1):122-127.
18. Pincez T, Guitton C, Gauthier F, et  al. Long-term fol­low-up of sub­to­tal
sple­nec­tomy for hered­i­tary spherocytosis: a sin­gle-cen­ter study. Blood.
2016;127(12):1616-1618.
19. Picard V, Guitton C, Thuret I, et al. Clinical and bio­log­i­cal fea­tures in PIEZO1-
hered­i­tary xerocytosis and Gardos channelopathy: a ret­ro­spec­tive series
of 126 patients. Haematologica. 2019;104(8):1554-1564.
20. Rice HE, Englum BR, Rothman J, et al; Splenectomy in Congenital Hemo-
lytic Anemia Consortium. Clinical out­comes of sple­nec­tomy in chil­dren:
report of the sple­nec­tomy in con­gen­i­tal hemo­lytic ane­mia reg­is­try. Am J
Hematol. 2015;90(3):187-192.
21. Danise P, Amendola G, Nobili B, et al. Flow-cytometric anal­y­sis of eryth­
ro­cytes and retic­u­lo­cytes in con­gen­i­tal dyserythropoietic anae­mia type II
(CDA II): value in dif­fer­en­tial diag­no­sis with hered­i­tary spherocytosis. Clin
Lab Haematol. 2001;23(1):7-13.
22. Harper SL, Sriswasdi S, Tang HY, Gaetani M, Gallagher PG, Speicher DW.
The com­mon hered­i­tary elliptocytosis-asso­ci­ated α-spectrin L260P muta­
tion perturbs eryth­ro­cyte mem­branes by sta­bi­liz­ing spectrin in the closed
dimer con­for­ma­tion. Blood. 2013;122(17):3045-3053.
23. Gallagher PG. Hereditary elliptocytosis: spectrin and pro­tein 4.1R. Semin
Hematol. 2004;41(2):142-164.
24. Niss O, Chonat S, Dagaonkar N, et  al. Genotype-phe­no­type cor­re­la­tions
in hered­i­tary elliptocytosis and hered­i­tary pyropoikilocytosis. Blood Cells
Mol Dis. 2016;61(Octo­ber):4-9.
25. Zarychanski R, Schulz VP, Houston BL, et  al. Mutations in the mechano-
transduction pro­tein PIEZO1 are asso­ci­ated with hered­i­tary xerocytosis.
Blood. 2012;120(9):1908-1915.
26. Andolfo I, Alper SL, De Franceschi L, et al. Multiple clin­i­cal forms of dehy-
drated hered­i­tary stomatocytosis arise from muta­tions in PIEZO1. Blood.
2013;121(19):3925-3935, S1-S12.
27. Albuisson J, Murthy SE, Bandell M, et al. Dehydrated hered­i­tary stomatocy-
tosis linked to gain-of-func­tion muta­tions in mechanically acti­vated PIEZO1
ion chan­nels. Nat Commun. 2013;4(2013):1884.
28. Glogowska E, Lezon-Geyda K, Maksimova Y, Schulz VP, Gallagher PG. Muta-
tions in the Gardos chan­nel (KCNN4) are asso­ci­ated with hered­i­tary xero-
cytosis. Blood. 2015;126(11):1281-1284.
29. Rapetti-Mauss R, Lacoste C, Picard V, et al. A muta­tion in the Gardos chan­
nel is asso­ci­ated with hered­i­tary xerocytosis. Blood. 2015;126(11):1273-1280.
30. Andolfo I, Russo R, Manna F, et al. Novel Gardos chan­nel muta­tions linked
to dehydrated hered­ i­
tary stomatocytosis (xerocytosis). Am J Hematol.
2015;90(10):921-926.
31. Fermo E, Bogdanova A, Petkova-Kirova P, et al. “Gardos channelopathy”:
a var­i­ant of hered­i­tary stomatocytosis with com­plex molec­u­lar reg­u­la­tion.
Sci Rep. 2017;7(1):1744.
32. Rivera A, Vandorpe DH, Shmukler BE, et  al. Erythrocyte ion con­tent and
dehy­dra­tion mod­u­late max­i­mal Gardos chan­nel activ­ity in KCNN4 V282M/+
hered­i­tary xerocytosis red cells. Am J Physiol Cell Physiol. 2019;317(2): C287-
C302.
Red cell mem­brane dis­or­ders  |  339
33. Platt OS, Lux SE, Nathan DG. Exercise-induced hemo­ly­sis in xerocytosis.
Erythrocyte dehy­dra­tion and shear sen­si­tiv­ity. J Clin Invest. 1981;68(3):631-
638.
34. Risinger M, Glogowska E, Chonat S, et al. Hereditary xerocytosis: diag­nos­
tic con­sid­er­ations. Am J Hematol. 2018;93(3):E67-E69.
35. Jankovsky N, Caulier A, Demagny J, et al. Recent advances in the path­o­
phys­i­­ol­ogy of PIEZO1-related hered­i­tary xerocytosis. Am J Hematol. 2021;
96(8):1017-1026.
36. Caulier A, Jankovsky N, Demont Y, et al. PIEZO1 acti­va­tion delays ery­throid
dif­fer­en­ti­a­tion of nor­mal and hered­i­tary xerocytosis-derived human pro­
gen­i­tor cells. Haematologica. 2020;105(3):610-622.
37. Filser M, Giansily-Blaizot M, Grenier M, et al. Increased inci­dence of germ-
line PIEZO1 muta­tions in indi­vid­u­als with idi­o­pathic erythrocytosis. Blood.
2021;137(13):1828-1832.
38. Albala MM, Fortier NL, Glader BE. Physiologic fea­tures of hemo­ly­sis asso­
ci­ated with altered cat­ion and 2,3-diphosphoglycerate con­tent. Blood.
1978;52(1):135-141.
39. Kiger L, Oliveira L, Guitton C, et  al. Piezo1-xerocytosis red cell metabo-
lome shows impaired gly­col­y­sis and increased hemo­glo­bin oxy­gen affin­ity.
Blood Adv. 2021;5(1):84-88.
40. Stewart GW, Amess JA, Eber SW, et  al. Thrombo-embolic dis­ease after
sple­nec­tomy for hered­i­tary stomatocytosis. Br J Haematol. 1996;93(2):303-
310.
41. Iolascon A, Andolfo I, Barcellini W, et al; Working Study Group on Red Cells
and Iron of the Euro­ pean Hematology Association. Recommendations
regard­ing sple­nec­tomy in hered­i­tary hemo­lytic ane­mias. Haematologica.
2017;102(8):1304-1313.
42. Archer NM, Shmukler BE, Andolfo I, et al. Hereditary xerocytosis revisited.
Am J Hematol. 2014;89(12):1142-1146.
43. Kim A, Nemeth E. New insights into iron reg­u­la­tion and eryth­ro­poi­es­is.
Curr Opin Hematol. 2015;22(3):199-205.
44. Rapetti-Mauss R, Picard V, Guitton C, et al. Red blood cell Gardos chan­nel
(KCNN4): the essen­tial deter­mi­nant of eryth­ro­cyte dehy­dra­tion in hered­i­
tary xerocytosis. Haematologica. 2017;102(10):e415-e418.
45. Ataga KI, Reid M, Ballas SK, et al; ICA-17043-10 Study Investigators.
Improvements in haemolysis and indi­ ca­
tors of eryth­
ro­
cyte sur­vival do
not cor­re­late with acute vaso-occlu­sive cri­ses in patients with sickle cell
340  |  Hematology 2021  |  ASH Education Program
dis­ease: a phase III ran­dom­ized, pla­cebo-con­trolled, dou­ble-blind study
of the Gardos chan­ nel blocker senicapoc (ICA-17043). Br J Haematol.
2011;153(1):92-104.
46. Bruce LJ. Hereditary stomatocytosis and cat­ion-leaky red cells—recent
devel­op­ments. Blood Cells Mol Dis. 2009;42(3):216-222.
47. Gallagher PG. Disorders of eryth­ro­cyte hydra­tion. Blood. 2017;130(25):2699-
2708.
48. Stewart AK, Shmukler BE, Vandorpe DH, et al. Loss-of-func­tion and gain-of-
func­tion phe­no­types of stomatocytosis mutant RhAG F65S. Am J Physiol
Cell Physiol. 2011;301(6):C1325-C1343.
49. Genetet S, Ripoche P, Picot J, et  al. Human RhAG ammo­nia chan­nel is
impaired by the Phe65Ser muta­ tion in overhydrated stomatocytic red
cells. Am J Physiol Cell Physiol. 2012;302(2):C419-C428.
50. Bruce LJ, Guizouarn H, Burton NM, et  al. The mono­va­lent cat­ion leak in
overhydrated stomatocytic red blood cells results from amino acid sub­sti­
tu­tions in the Rh-asso­ci­ated gly­co­pro­tein. Blood. 2009;113(6):1350-1357.
51. Lux SE IV. Anatomy of the red cell mem­brane skel­e­ton: unanswered ques­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/331/1851848/331kalfa.pdf by guest on 13 December 2021
tions. Blood. 2016;127(2):187-199.
52. Kalfa TA. Warm anti­body auto­im­mune hemo­lytic ane­mia. Hematology Am
Soc Hematol Educ Program. 2016;2016(1):690-697.
53. Voulgaridou A, Kalfa TA. Autoimmune hemo­lytic ane­mia in the pedi­at­ric
set­ting. J Clin Med. 2021;10(2):216.
54. Kamesaki T, Toyotsuji T, Kajii E. Characterization of direct anti­glob­ul­in test-
neg­a­tive auto­im­mune hemo­lytic ane­mia: a study of 154 cases. Am J Hema-
tol. 2013;88(2):93-96.
55. King MJ, Smythe JS, Mushens R. Eosin-5-maleimide bind­ ing to band 3
and Rh-related pro­teins forms the basis of a screen­ing test for hered­i­tary
spherocytosis. Br J Haematol. 2004;124(1):106-113.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000265
 INHERITED RED CELL DISORDERS BEYOND HEMOGLOBINOPATHIES

Diagnosis and clinical management

of enzymopathies

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021

Lucio Luzzatto
Department of Hematology and Blood Transfusion, Muhimbili University of Health and Allied Sciences, Dar es Salaam, Tanzania; and University of
Florence, Firenze, Italy

At least 16 genetically determined conditions qualify as red blood cell enzymopathies. They range in frequency from ultra-
rare to rare, with the exception of glucose-6-phosphate dehydrogenase deficiency, which is very common. Nearly all these
enzymopathies manifest as chronic hemolytic anemias, with an onset often in the neonatal period. The diagnosis can be
quite easy, such as when a child presents with dark urine after eating fava beans, or it can be quite difficult, such as when
an adult presents with mild anemia and gallstones. In general, 4 steps are recommended: (1) recognizing chronic hemolytic
anemia; (2) excluding acquired causes; (3) excluding hemoglobinopathies and membranopathies; (4) pinpointing which
red blood cell enzyme is deficient. Step 4 requires 1 or many enzyme assays; alternatively, DNA testing against an appro-
priate gene panel can combine steps 3 and 4. Most patients with a red blood cell enzymopathy can be managed by good
supportive care, including blood transfusion, iron chelation when necessary, and splenectomy in selected cases; however,
some patients have serious extraerythrocytic manifestations that are difficult to manage. In the absence of these, red blood
cell enzymopathies are in principle amenable to hematopoietic stem cell transplantation and gene therapy/gene editing.

LEARNING OBJECTIVES
• Update clinical, hematologic, biochemical, and molecular approaches to the diagnosis of red blood cell enzymop-
athies
• Pinpoint the role of supportive treatment, targeted treatment, and advanced forms of treatment in the manage-
ment of red blood cell enzymopathies
• Update knowledge about the molecular basis of red blood cell enzymopathies

Introduction essential for the detoxification of hydrogen peroxide (H2O2)

Red blood cell enzymopathies are genetic disorders affecting
the intraerythrocytic metabolism.1,2 Red blood cells, having
sacrificed their nucleus and other organelles to the unself-
ish task of providing oxygen to all other cells, are limited in
their capacity to withstand metabolic impairment and also
limited in how they will be affected: when one important en-
zyme is deficient, the red blood cell life span is nearly always
compromised, with consequent hemolysis. Most red blood
cell enzymes are ubiquitously expressed “housekeeping” en-
zymes (Table 1); therefore, there may be pathology in other
tissues as well. Because red blood cells do not have protein
synthesis, however, they are in general more vulnerable.
The main metabolic fuel of the red blood cell is plasma
glucose, and its breakdown within the red blood cell provides
2 key compounds: adenosine triphosphate (ATP), required
mainly for the operation of the cation pump, and nicotinamide
adenine dinucleotide phosphate, reduced (NADPH), required
mainly for keeping up the supply of reduced glutathione (GSH)
Prakash Singh Shekhawat
and of other reactive oxygen species (ROS). ATP is produced
by reactions of the “classic” glycolytic pathway; NADPH is
produced by the pentose phosphate pathway (Figure 1).3
Therefore, it is natural to classify red blood cell enzymopa-
thies according to these 2 groups. A third group comprises
enzymes involved in nucleotide metabolism (Table 1).
Features of hemolytic anemias due to enzymopathies
can be best illustrated through clinical summaries of indi-
vidual patients.
CLINICAL CASE
A 1-year-old boy born in 1963 from parents not known to
be consanguineous, although both were from Morcone
(a small town not far from Naples, Italy), was found by his
local pediatrician to have anemia with reticulocytosis. Since
birth he had had moderate jaundice that did not respond
Red blood cell enzymopathies | 341
Table 1.  List of red blood cell enzymopathies under­ly­ing CNSHA*

Extraerythrocytic
Chromosomal
Enzyme (Acronym) Gene Clinical Manifesta- Notes
location
tions†
Glycolytic Hexokinase (HK) HK1 10q22 May ben­e­fit from sple­nec­tomy;
path­way36 bone marrow transplantation
(BMT) has been done.
Glucose-6-phos­phate GPI 19q31.1 Rarely myop­a­thy, Ranks sec­ond in fre­quency within
isom­er­ase (GPI) central nervous the gly­co­lytic path­way group
system (CNS) (after PK). May ben­e­fit from
sple­nec­tomy.
Phosphofructokinase (PFK) PFKM 12q13.3 Myopathy, Primarily a mus­cle dis­ease, with
myoglobinuria glycogenosis in mus­cle. CNSHA

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021

mild.7
Aldolase ALDOA 16q22-24 Myopathy, CNS Fever may trig­ger poten­tially fatal
mas­sive rhabdomyolysis.37
Triose phos­phate isom­er­ TPI1 12p13.31 CNS (severe), car­ Disease very rare, but same muta­tion
ase (TPI) dio­my­op­a­thy (p. Glu104Asp) found in sev­eral
cases, suggesting founder effect.
Glyceraldehyde-3- GAPDH 12p13.31- Myopathy Hemolytic ane­mia reported in some
phos­phate dehy­dro­ge­ cases but link with enzyme defi­
nase (GAPD) ciency not clearly established.
Bisphosphoglycerate BPGM 7q33 Nohemo­ly­sis; erythrocytosis can be
mutase (DPGM) attrib­uted to low 2,3-DPG, left-
shifted Hb-O2 dis­so­ci­a­tion curve,
rel­a­tive hyp­oxia.
Phosphoglycerate kinase PGK1 Xq21.1 CNS myop­a­thy May ben­e­fit from sple­nec­tomy;
(PGK) BMT has been done. Early-onset
par­kin­son­ism reported.38
Pyruvate kinase (PK) PKLR 1q22 Iron over­load Ranks first in fre­quency in the gly­co­
lytic enzyme group. May ben­e­fit
from sple­nec­tomy; BMT has been
done.
Redox16,39 Glucose-6-phos­phate dehy­ G6PD Xq28 Very rarely Inalmost all­cases, only AHA from
dro­ge­nase (G6PD) granulocytes exog­e­nous trig­ger‡; CNSHA only
with class I var­i­ants.
Glutathione synthase GSS 20q11.22 CNS May be asso­ci­ated with high
5-oxoproline and met­a­bolic
aci­do­sis.40
Glutathione reduc­tase GSR 8p12 Cataracts AHA from exog­e­nous trig­ger
(fav­ism).
γ-glutamylcysteine synthase GCLC 6p12.1 CNS Mutations affect cat­a­lytic sub­unit.41
Cytochrome b5 reduc­tase CYB5R3 22q13.2 CNS Methemoglobinemia rather than
(CBR) hemo­ly­sis.
Nucleotide Adenylate kinase (AK) AK1 9q34.11 CNS May ben­e­fit from sple­nec­tomy.
metab­o­lism
Pyrimidine 5′-nucle­o­tid­ase NTSC3A 7p14.3 Ranks third in fre­quency among all­
(P5N) red blood cell enzymopathies
(leav­ing aside G6PD). May ben­e­fit
from sple­nec­tomy.
*The molec­u­lar basis of an enzymopathy is in the muta­tion(s) in the respec­tive gene. Most enzymopathies are auto­so­mal reces­sive dis­or­ders.
Therefore, the patient has a muta­tion in each of the 2 alle­lic genes encoding the respec­tive enzyme: the muta­tion may be the same in both alleles
(homo­zy­gos­ity), or there may be 2 dif­fer­ent muta­tions (com­pound het­ero­zy­gos­ity or biallelic muta­tions). In most cases the muta­tion causes insta­
bil­ity of the pro­tein: since mature red blood cells can­not make pro­teins, the enzyme activ­ity with which a retic­u­lo­cyte is endowed will decay as
the red blood cell ages in cir­cu­la­tion.
†
Since we are deal­ing with ubiq­ui­tously expressed genes, the pres­ence of such man­i­fes­ta­tions is not sur­pris­ing. Whether they are pres­ent or not,
and to what extent, depends mainly on 3 fac­tors: (1) nonerythroid cells may express alter­na­tive forms of a par­tic­u­lar enzyme, from the same gene
or from other genes; (2) if the main mech­a­nism of defi­ciency is enzyme insta­bil­ity, cells other than eryth­ro­cytes may com­pen­sate by increased
bio­syn­the­sis; (3) if the main mech­a­nism of defi­ciency is a qual­i­ta­tive change (eg, in the active site), all­cells expressing the gene will be affected.
The last 2 fac­tors may explain why extraerythrocytic man­i­fes­ta­tions may vary even within the same enzymopathy.
‡
G6PD defi­ciency is wide­spread in malaria-endemic areas, con­se­quent on malaria hav­ing been the selec­tive agent for this poly­mor­phism.42 In areas
endemic for Plasmodium vivax malaria, a test for G6PD defi­ciency should be car­ried out before giv­ing primaquine or tafenoquine, the only drugs
that pre­vent relapse. This has been a strong stim­u­lus for the devel­op­ment of quan­ti­ta­tive point-of-care tests for G6PD defi­ciency.43

342  |  Hematology 2021  |  ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021

Figure 1. Red blood cell metabolism. Glycolysis (the Embden-Meyerhof pathway) generates ATP required for cation transport and for
membrane maintenance, while NADH maintains hemoglobin iron in a reduced state. The hexose monophosphate shunt generates
the NADPH that is used to reduce GSH, which protects the red blood cell against oxidant stress (Figure 7); 6-phosphogluconate, after
decarboxylation, can be recycled via pentose sugars to glycolysis. At the side of glycolysis is the Rapoport-Luebering cycle, which
provides 2,3-isphosphoglycerate (2,3-DPG): its level is a critical determinant of the oxygen affinity of hemoglobin. G6PD deficiency is
highly prevalent in many parts of the world (Figure 6); all other enzymopathies are rare to ultrarare. Among them, the order of preva-
lence is PK, followed by P5N, followed by GPI.

Prakash Singh Shekhawat

Red blood cell enzymopathies  | 343
to barbiturates. Over several years his hemoglobin (Hb) level
fluctuated in the range from 7 to 11g/dL, and he received sev-
eral blood transfusions. At the age of 18, when he was fully reas-
sessed (B. Rotoli, MD, in Naples), physical examination showed
pallor, mild icterus, and splenomegaly. There were no neurolog-
ical signs or muscle weakness. Hb was 10.4 G/dL; mean corpus-
cular volume, 104 fL, reticulocytes, 480×109/L; and unconjugated
bilirubin, 5mg/dL; erythrocyte morphology showed anisocytosis
and basophilic stippling in 4% of cells. There were no abnormal
hemoglobins, osmotic fragility was normal, and the autohemolysis
test showed increased lysis not corrected by glucose. A 51Cr study
showed reduction of the red blood cell half-life by about 40%,
with increased radioactive uptake in both the spleen and liver.
A panel of red blood cell enzyme assays (A. Zanella, MD, Milan)
revealed a marked deficiency of glucose-6-phosphate isomer-
ase (GPI).4 The patient, who had been placed on daily folic acid
(5mg/day), continued to require an average of 2.5 blood units
per year until, at the age of 25, he developed obstructive jaun-
dice due to gallstones. He underwent surgery that included cho-
lecystectomy, choledochoduodenostomy, and splenectomy. The
spleen weighed 500g; microscopic examination showed marked
congestion, erythrophagocytosis, and small foci of erythropoie-
sis. A liver biopsy showed evidence of cholestasis. The Hb level
increased and remained for years in the range of 10 to 11g/dL;
therefore, regular blood transfusions were no longer needed.
Reticulocytes remained high (in the range of 250-350×109/L). At
the age of 27, the patient completed his doctorate in philosophy.
When the patient was 31 years old, sequencing of the GPI gene
revealed that the patient was homozygous for a p.Q343R muta-
tion identical to that independently observed at about the same
time in a patient from Japan.5,6
At the age of 44, the patient developed severe jaundice that
persisted for weeks (the bilirubin peaked at 80mg/dL) and was
admitted to a unit that specialized in liver transplantation (Dr G.
Ramadori, Göttingen). Imaging studies revealed an accessory
spleen. A liver biopsy (reviewed by Dr Tania Roskam, Leuven)
showed massive bilirubin overload, cholestasis, ballooned hepato-
cytes, ceroidosis of the Kupffer cells, and moderate centrolobular
perisinusoidal fibrosis but no evidence of cirrhosis. On supportive
treatment (including transfusion of 10 units of blood), the patient
gradually improved. Quantitative magnetic resonance-imaging
studies (R. Galanello, MD, Cagliari) revealed significant iron over-
load in the liver; deferoxamine was recommended as the iron-
chelating agent least likely to be hepatotoxic.
At the age of 52, after 3 further episodes of severe jaundice,
imaging studies showed stenosis of the choledochoduodenos-
tomy with proximal dilatation of the common duct, numerous
stones, possible cholangitis, and evidence of iron in the acces-
sory spleen and in the liver. The patient underwent surgery
(Dr A.D. Pinna, Bologna) consisting of removing innumerable
stones, including a large one, reconstructing the biliary-intes-
tinal anastomosis, and removing the accessory spleen. Immu-
nization against capsulated bacteria was carried out. Since that
time the patient has been stable clinically and hematologically,
with an Hb of about 11 g/dL, reticulocytes about 4 times the
upper normal limit, and mildly elevated bilirubin. The patient,
now 58, is head librarian in an academic institution in Rome; he
has recently published a book on the liberalism of the Italian phi-
losopher Benedetto Croce. The last time I had contact with him
it was to recommend that he receive a COVID-19 vaccination.
344 | Hematology 2021 | ASH Education Program
Glycolytic enzymopathies
Patient 1 recapitulates the clinical and hematologic features of
patients not just with GPI deficiency but with chronic nonsphero-
cytic hemolytic anemia (CNSHA) due to a deficiency of any of
the glycolytic enzymes: chronic hemolysis, splenomegaly, gall-
stones. What varies greatly is the severity of these features, not
only from one enzymopathy to another but even from patient to
patient within the same enzymopathy. This explains why the age
of presentation may range from death in utero to adulthood. The
anemia is normocytic and normochromic: if it appears to be mac-
rocytic, it is usually on account of marked reticulocytosis. A short-
age of folic acid might also contribute to a high mean corpuscular
volume, and since the increased rate of erythropoiesis increases
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021
the requirement of this vitamin, folic acid supplementation is rec-
ommended. Although there is considerable individual variation in
tolerating anemia, in all chronic anemias the body tends to adapt.
The best-known red blood cell intrinsic mechanism of adaptation
is an increase in bisphosphoglycerate (2,3-DPG) that shifts the
Hb-O2 dissociation curve to the right: this can be expected if the
flow of glycolysis is impaired downstream, but not upstream, of the
Rapoport-Luebering cycle (Figure 1). Indeed, with phosphofructoki-
nase deficiency 2,3-DPG decreases,7 but with pyruvate kinase (PK)
deficiency, it increases, and the increased O2 delivery to the tissues
is of benefit to the patient.8
The pathophysiology of hemolysis in this group of enzy-
mopathies consists mainly of the removal of red blood cells by
macrophages, ie, extravascular hemolysis. That is why splenec-
tomy is so often beneficial (Figure 2). It cannot be claimed to
Figure 2. Splenectomy does not cure but does ameliorate CN-
SHA of PK deficiency. In each of these 10 patients with PK de-
ficiency, a significant increase in the steady-state hemoglobin
level was seen after splenectomy (left panel). Whereas in most
types of CNSHA an improvement of anemia, resulting from de-
creased hemolysis, is usually associated with a decrease in the
reticulocyte count, in PK deficiency reticulocytes often increase
after splenectomy. This paradoxical phenomenon suggests that
the spleen selectively removes PK-deficient reticulocytes; the
mechanism is not yet well understood.48 From Zanella et al.8
 PK deficiency G6PD deficiency
(autosomal) (X-linked)

Homozygous
normal

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021

Heterozygous

Homozygous
deficient

Figure 3. Different phenotypes of heterozygotes for red blood cell enzymopathies. In a heterozygote for PK deficiency, encoded by
an autosomal gene (Table 1), the level of enzyme is about one-half of normal in all red blood cells. Since this level of enzyme is suffi-
cient, there are no clinical consequences—ie, PK deficiency is recessive. In a heterozygote for deficiency of G6PD, encoded by an X-
linked gene, the situation is quite different: X chromosome inactivation generates red blood cell mosaicism, whereby some red blood
cells are entirely normal, and others are G6PD deficient. Therefore, G6PD deficiency is expressed in heterozygotes: it is not recessive.

be a rad­i­cal treat­ment, but if it raises the Hb level suf­fi­ciently
to avoid reg­u­lar blood trans­fu­sion, it will also avoid, in most
cases, the need for iron che­la­tion ther­apy, lead­ing to a sub­stan­
tial improve­ment in qual­ity of life. The acces­sory spleen was
an uncalled-for extra hitch in patient 1: over the years, being a
site of hemo­ly­sis, it may have become larger than it had been
orig­i­nally, and per­haps that is why it was missed dur­ing the first
oper­a­tion.
Except for phos­pho­glyc­er­ate kinase defi­ciency, which is
X-linked, all­other gly­co­lytic enzymopathies are auto­so­mal reces­
sive: ie, het­ero­zy­gotes are not affected. This must mean that a
50% reduc­tion in enzyme activ­ity (Figure 3) does not com­pro­mise
red blood cell sur­vival. Enzymopathy patients are either homo­zy­
gous for a loss of func­tion muta­tion, or they have 2 dif­fer­ent muta­
tions in the 2 alleles at the same locus (com­pound het­ero­zy­gotes,
also referred to as biallelic). Whenever a patient is homo­zy­gous
for a very rare muta­tion, it is likely that the par­ents, even if not
known to be con­san­guin­e­ous, may be ances­trally related. In other
genetic dis­or­ders, there are now algo­rithms aiming to esti­mate
the prob­a­bil­ity that a par­tic­u­lar pre­vi­ously unknown muta­tion is or
is not path­o­genic. In the case of enzymopathies, this is not gen­er­
ally nec­es­sary because a loss of enzyme activ­ity is in itself a good
read­out of path­o­ge­nic­ity.
Prakash Singh Shekhawat
The autohemolysis test used in patient 1 was devel­oped
by J.V Dacie.9 It was based on the obser­va­tion that when red
blood cells from patients with a wide range of hered­i­tary ane­
mias are incu­bated in vitro at 37 °C for 24 to 48 hours, some
frac­tion of them undergo hemo­ly­sis. If the abnor­mal­ity is in the
mem­brane, sup­ply­ing extra glu­cose usu­ally helps to reduce
hemo­ly­sis; if the abnor­mal­ity is instead in gly­col­y­sis, glu­cose
does not help. It was a clever test that gave a strong hint—cor­
rect in our case—that the patient had an enzymopathy. The
test is so eco­nom­i­cal that in prac­tice it is no lon­ger used. The
sus­pi­cion of enzymopathy must be raised in any patient with
chronic hemo­lytic ane­mia presenting early in life, or even later,
who does not have a hemo­glo­bin­op­a­thy and the mor­phol­ogy
char­ac­ter­is­tic of red blood cell membranopathies; there­fore,
sound clin­i­cal hema­tol­ogy is still par­a­mount. Once the sus­pi­
cion is for­mu­lated, the choice is between a bat­tery of quan­
ti­ta­tive enzyme assays and a geno­mic approach. Today the
lat­ter is prob­a­bly pref­er­a­ble: it con­sists in obtaining from the
patient’s DNA the sequence of genes whose muta­tions may
cause a red blood cell enzymopathy, or con­gen­i­tal hemo­lytic
ane­mia in gen­eral.10 This approach has been used suc­cess­fully
with a panel of 71 genes,11 at a price (to the Ital­ian National
Health Service) of about $950 per patient; sev­eral com­mer­cial
Red blood cell enzymopathies  | 345
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021
Figure 4. Red blood morphology in select enzymopathies. Left panel: in a patient with CNSHA due to PK deficiency, the blood smear
is not diagnostic; however, the presence of “prickle red cells” should raise suspicion. From Mahendra et al.49 Middle panel: “bite cells,”
hemighosts and microspherocytes, characteristic of oxidative hemolysis, are seen in this smear from a child who received a dap-
sone-chlorproguanyl combination for the treatment of acute P. falciparum malaria. From Pamba et al.50 Right panel: red blood cells
with fine basophilic stippling in a patient with P5N deficiency. This enzymopathy is the most common inherited cause of basophilic
stippling, which is also seen with lead poisoning (lead inhibits many enzymes, including P5N, and it can produce a phenocopy of P5N
deficiency). From Rees, Duley, and Marinaki.51

pan­els are now avail­­able. The geno­mic approach is also advan­
ta­geous when the patient is heavily trans­fused, mak­ing a red
blood cell enzyme assay unre­li­able.
Among the gly­co­lytic enzymopathies, the one that has been
best char­ac­ter­ized in terms of man­age­ment is PK defi­ciency (red
blood cell mor­phol­ogy in Figure 4),12 mainly because it is the
most com­mon (although, with an esti­mated fre­quency in the
United States of 5 per 100 000, it is still a rare dis­ease); there­fore,
the range of the clin­i­cal bur­den is bet­ter known. From the clin­i­
cal point of view, it is note­wor­thy that iron over­load is fre­quent,
even in patients who are not trans­fu­sion depen­dent.13 This may
be due to hepcidin sup­pres­sion, con­sis­tent with a com­po­nent
of inef­fec­tive eryth­ro­poi­e­sis in the ane­mia of PK defi­ciency.14
In this respect it is inter­est­ing that, upon diag­nos­tic test­ing of
inherited ane­mias by sequenc­ing a gene panel, sev­eral patients
who had been diag­ nosed with con­ gen­i­
tal dyserythropoietic
ane­mia had in fact PKLR muta­tions, ie, PK defi­ciency.11
Mitapivat, an allo­ste­ric acti­va­tor of PK (Figure 5), pro­duced a
sig­nif­i­cant increase in hemo­glo­bin level in one-half of 52 patients
with PK defi­ciency in a phase 3 multicentric clin­i­cal trial; this was
asso­ci­ated with a decrease in bil­i­ru­bin and in retic­u­lo­cytes.15
Mitapivat may be on track to become the first approved drug for
a red blood cell enzymopathy.
In the man­age­ment of patient 1, at least 7 senior col­leagues
from 4 dif­fer­ent countries with spe­cial­ized exper­tise in dif­fer­ent
areas have been involved: I found their help invalu­able. At the
same time, it must be noted that a per­son with a poten­tially
dis­abling dis­ease can still hold on and have a qual­ity of life that
enables him to con­trib­ute to the life and cul­ture of oth­ers.
CLINICAL CASE
A 3-year-old girl from Egypt was seen in London for an assess­ment
of her clin­i­cal state. Her par­ents also requested genetic coun­sel­
ing. The patient was the cou­ple’s first­born. Her height and weight
were in the lower 10th per­cen­tile, and she was thin but did not
appear mal­nour­ished. She had bilat­eral choreoathetosis, invol­un­
tary mus­cle move­ments and spasms, and upward eye move­ments;
her speech was lim­ited and dysarthric. Her his­tory was highly sig­
nif­i­cant in that she had been born at 35 weeks’ ges­ta­tion and at
birth had had severe jaun­dice, but no hos­pi­tal records were avail­­
able. On one occasion after eating fava beans, she had become
lethargic and had passed dark urine. At the time of her visit, her
blood count, includ­ing retic­u­lo­cytes, was nor­mal: there was no
evi­dence of CNSHA. The 3 mem­bers of the fam­ily were all­Rh+,
and there was no ABO incom­pat­i­bil­ity setup. Red blood cell glu­
cose-6-phos­phate dehy­dro­ge­nase (G6PD) assays yielded val­ues
of 1.5 in the patient, 7.2 in her mother, and 0.6 in her father (nor­mal
val­ues, 7-10 IU/g Hb). The most likely diag­no­sis was severe neu­
ro­log­i­cal dam­age from kernicterus. We inferred from the G6PD
results that the father was hemi­zy­gous G6PD defi­cient: sequence
anal­y­sis of his G6PD gene iden­ti­fied a pre­vi­ously unknown muta­
tion, p.N135T, that was hence named G6PD Cairo; the daugh­ter

346  |  Hematology 2021  |  ASH Education Program


Figure 5. Mitapivat is an allosteric activator of PK. Top panel: the sigmoid-shaped dependence of PK activity on the concentration of
AG-348 (mitapivat) suggests cooperative interaction among the PK subunits. Bottom panel: diagram of the three-dimensional struc-
ture of the PK tetramer with bound mitapivat (gray). Blue: the substrate phosphoenolpyruvate PEP; brown: the physiological activator
fructose-1,6-bisphosphate (FBP). Modified from Kung et al.52
was het­ero­zy­gous and the mother, nor­mal. The mode of inher­
i­tance of G6PD defi­ciency and its impli­ca­tions were explained
to the fam­ily. Father and daugh­ter were advised not to eat fava
beans. We informed them that the prob­lem of neo­na­tal jaun­dice
(NNJ) was unlikely to recur if a sub­se­quent baby were male, but
unfor­tu­nately, it might recur in a female, although it might be less
severe. We empha­sized the impor­tance of hav­ing the baby in a
loca­tion where NNJ could be appro­pri­ately man­aged.
Redox enzymopathies
The most prominent enzymopathy in this group is indeed G6PD
defi­ciency.16 From an epi­de­mi­o­log­i­cal point of view, whereas the
gly­co­lytic enzymopathies are all­rare to ultra­rare, here we are
at the other end of the spec­trum: G6PD defi­ciency is pres­ent in
an esti­mated 500 mil­lion peo­ple world­wide (Figure 6). From a
clin­i­cal point of view, we are deal­ing with a very dif­fer­ent sit­u­a­
tion. Whereas the gly­co­lytic enzymopathies cause life­long CN-
SHA, G6PD defi­ciency in itself does not cause ane­mia and is not
even a dis­ease; rather, it is a genetic abnor­mal­ity that in most
cases remains asymp­tom­atic for a life­time (except for the very
rare variants that cause CNSHA, clinically similar to that from
glycolytic enzymopathies16). However, it can man­i­fest at birth
through NNJ (as in patient 2) or at any age,17 like a thun­der­bolt
out of a blue sky, in the form of acute hemo­lytic ane­mia (AHA),
when the per­son is chal­lenged by an agent that causes oxi­da­
tive dam­age to the red blood cells (Figure 4).18 The agent may
be the inges­tion of fava beans, or infec­tion, or a drug (prob­a­bly
in that order of fre­quency: for the mech­a­nism, see Figure 7).19
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021
In G6PD-defi­cient per­sons, AHA is a pro­to­type exam­ple of a
gene-envi­ron­ment inter­ac­tion caus­ing a dis­ease that can be
life-threat­en­ing. Favism in chil­dren can be fatal if not promptly
treated, and rasburicase, poten­tially a life­saver in a patient with
an incum­bent tumor lysis syn­drome, can become a seri­ous haz­
ard if the patient hap­pens to be G6PD defi­cient (Figure 7B)—a
typ­i­cal exam­ple of a pharmacogenetic haz­ard.20
Patient 2 illus­trates that NNJ, a clin­i­cal man­i­fes­ta­tion of G6PD
defi­ciency known for decades,21,22 is no less impor­tant than
AHA. In most G6PD defi­cient babies, the hyperbilirubinemia is
mild: it over­laps with “phys­i­o­log­i­cal jaun­dice” of the new­born.
However, in some cases, like this one, it was severe enough to
cause kernicterus. In many countries, G6PD defi­ciency is the
most com­mon cause of severe NNJ: in 1 study in Nigeria among
babies admit­ted to the hos­pi­tal after show­ing clin­i­cal signs of
kernicterus, the fre­quency of G6PD defi­ciency was 78%—com­
pared to 22% in the gen­eral pop­u­la­tion.23 The under­ly­ing G6PD
var­i­ant was G6PD A−, the same that accounts for many cases
of NNJ, includ­ing severe ones, in the United States and one of
the rea­sons why neo­na­tal screen­ing for G6PD defi­ciency has
been advo­cated.17,24 The rea­sons for sever­ity may be envi­ron­
men­ tal (eg, infec­ tion, expo­sure to naph­ tha­ lene) or genetic
(coex­is­tence of 1 or 2 UGT1A1 mutant alleles),25 and they are
in part unknown. The man­age­ment of G6PD-related NNJ is no
dif­fer­ent from that due to other causes: this child should have
had exchange blood trans­fu­sion, but unfor­tu­nately, she did not.
G6PD Cairo, the G6PD var­i­ant first iden­ti­fied in patient 2, was
sub­se­quently found to be quite com­mon in Egypt and else­
where, par­tic­u­larly in patients presenting with acute fav­ism.26
Red blood cell enzymopathies  | 347
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021
Figure 6. Epidemiology of G6PD deficiency throughout the world. Each country on the map is shaded in a color based on the best
estimate of the mean frequency of G6PD deficiency allele(s) in that country (this is the same as the frequency of G6PD-deficient
males). The larger panel gives a color-coded list of 10 common G6PD variants associated with G6PD deficiency: asterisk-shaped sym-
bols in the corresponding colors are shown in the countries where these variants have been observed (for graphic reasons symbols
could not be inserted in all countries where the respective variants are present). Modified from Luzzatto, Ally, and Notaro.16

The impor­tance of G6PD being X-linked can­not be overem-
phasized. First, since males have only 1 X chro­mo­some, they
are either hemi­zy­gous G6PD nor­mal or hemi­zy­gous G6PD defi­
cient; females can be homo­zy­gous nor­mal or G6PD defi­cient
(biallelic, whether homo­zy­gous or com­pound het­ero­zy­gous)
or het­ero­zy­gous for G6PD defi­ciency (ie, with 1 nor­mal allele).
Second, in any pop­u­la­tion, according to the Hardy-Weinberg
rule, het­ero­zy­gous females are more com­mon than hemi­zy­gous
G6PD-defi­cient males (Table 2). Third, since G6PD is sub­ject to
X chro­mo­some inac­ti­va­tion, in het­ero­zy­gous females there is
red blood cell mosa­i­cism, whereby a pro­por­tion of red blood
cells are just as defi­cient in G6PD activ­ity as in a hemi­zy­gous
male (Figure 3). In het­ero­zy­gotes the expres­sion of G6PD defi­
ciency is highly var­i­able, and since X inac­ti­va­tion can be ran­
domly “skewed,” their enzyme lev­ els range from nor­ mal to
mark­edly defi­cient so that in an indi­vid­ual female the pro­por­
tion of G6PD defi­cient red blood cells may be high (even up to
90%). This last point is clin­i­cally rel­e­vant, as it was unfor­tu­nately
in our patient 2.
Given the high fre­quency of G6PD defi­ciency in many parts
of the world (Figure 6), blood donors may be G6PD defi­cient.
A recent care­ful inves­ti­ga­tion has found that after blood bank
stor­age for 6 weeks the major­ity of G6PD-defi­cient blood units
still meet, 24 hours after trans­fu­sion, the in vivo red blood cell
sur­vival tar­get of >75%.27
Some ultra­rare enzymopathies in the redox cat­eg ­ ory involve
either the bio­syn­the­sis or the func­tion of GSH (Table 1). GSH
reduc­tase defi­ciency can man­i­fest as fav­ism,28 confirming the
key role of GSH in the defense of red blood cells against oxi­da­
tive stress.

348  |  Hematology 2021  |  ASH Education Program

A Divicine
Primaquine
(Fava Beans)

ROS
from Neutrophils

O2- NADP
Glucose 6-Phosphate

Glucose 6-Phosphate
Superoxide Dismutase Dehydrogenase
Rasburicase
GSH 6-Phosphoglucono-
Uric Acid H2O2
δ-Lactone
Glutathione
Catalase Reductase NADPH
Prx2-SH-

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021

6-Phosphogluconolactonase
H2O GSSG
Glutathione 6-Phosphogluconate
Peroxidase
6-Phosphogluconate
Prx2-S-S-
Dehydrogenase

Ribulose 5-Phosphate
B Divicine Primaquine
(Fava Beans)

Oxidative
ROS
Damage
O2-
from Neutrophils
Glucose 6-Phosphate
NADP Glucose 6-Phosphate
Superoxide Dismutase
Dehydrogenase
Rasburicase
GSH 6-Phosphoglucono-δ-Lactone
Uric Acid H2O2
Glutathione 6-Phosphoglucono
Catalase Reductase Lactonase
Prx2-SH-
H2O GSSG NADPH 6-Phosphogluconate
Glutathione
Peroxidase 6-Phosphogluconate
Prx2-S-S- Dehydrogenase

Ribulose 5-phosphate
Figure 7. The role of G6PD in protecting red blood cells from oxidative damage. (A) In G6PD-normal red blood cells, G6PD and
6-phosphogluconate dehydrogenase—2 of the enzymes of the PPP—provide an ample supply of NADPH, which in turn regenerates
GSH when this is oxidized by ROS (eg, superoxide, O2− and H2O2). Thus, when O2− (meant here to represent itself and other ROS) is
produced by pro-oxidant compounds such as primaquine, or the glucosides in fava beans (divicine), or the oxidative burst of neutro-
phils, these ROS are rapidly neutralized; similarly, when rasburicase administered to degrade uric acid produces an equimolar amount
of H2O2, this is rapidly degraded by the combined action of GSH peroxidase, catalase, and Prx2 (peroxiredoxin-2: all 3 mechanisms
are NADPH dependent). (B) In G6PD-deficient red blood cells, where the enzyme activity is reduced, NADPH production is limited,
and it may not be sufficient to cope with the excess ROS generated by pro-oxidant compounds and the consequent excess H2O2.
This diagram also explains why a defect in GSH reductase has very similar consequences to G6PD deficiency. Modified from Luzzatto,
Nannelli, and Notaro.53 PPP, pentose phosphate pathway.

Cytochrome b5 reduc­tase (CBR) defi­ciency is an enzymop-

athy that does not cause hemo­lytic ane­mia: it instead causes
reces­sive con­gen­i­tal met­he­mo­glo­bi­ne­mia,29 often presenting as
cya­no­sis in infants, that may mis­lead one to sus­pect con­gen­i­tal
heart dis­ease. This NADH-depen­dent enzyme (for­merly known
as diaph­o­rase), capa­ble of reduc­ing met­he­mo­glo­bin (Fe+++) back
to hemo­glo­bin (Fe++), is encoded by CYB5R3.* Most mis­sense
muta­tions in this gene cause only met­he­mo­glo­bi­ne­mia (type I
dis­ease), whereas non­sense muta­tions, dele­tions, and some spe­
Prakash Singh Shekhawat
*In red blood cells there is also an NADPH-depen­dent enzyme that can
reduce met­he­mo­glo­bin back to hemo­glo­bin: it was for­merly called fla­vin
reduc­tase and is now known to be iden­ti­cal to bil­i­ver­din-IX reduc­tase.30
Once again it is from genetic evi­dence that we know that CBR is the
enzyme respon­si­ble for the phys­i­o­logic main­te­nance of hemo­glo­bin in
the Fe++ state since muta­tions of CYB5R3 cause met­he­mo­glo­bi­ne­mia,
whereas muta­tions of BLVRB do not.
Red blood cell enzymopathies  | 349
Table 2. G6PD defi­ciency in sev­eral sam­ple pop­u­la­tions

Frequency of G6PD defi­ciency

Frequency of G6PD defi­ciency in in females, %
Examples
(hemi­zy­gous) males, %
Heterozygous Homozygous
US 2.5 4.9 0.06
Nigeria 22 34 4.8
Sardinia 12 21 1.4
Thailand 7.3 13.5 0.5
Frequency in US from Chinevere et al44; in Nigeria, from Luzzatto and Allan45; in Sardinia, from Cocco et al46; in Thailand, from Bancone et al.47 The
fre­quency of G6PD defi­ciency in hemi­zy­gous males is iden­ti­cal to the fre­quency of the G6PD mutant allele (or the sum total of the fre­quen­cies if
more than 1 mutant allele is involved). From the male fre­quency, it is easy to cal­cu­late the fre­quen­cies of the female geno­types, on the assump­tion

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021

that the pop­u­la­tion is in Hardy-Weinberg equi­lib­rium. The data in this table show that, as all­text­books say, homo­zy­gous females are few com­
pared to hemi­zy­gous males; on the other hand, het­ero­zy­gous females are almost twice as com­mon as hemi­zy­gous males (which most text­books
fail to say). Because of epi­ge­netic mosa­ic
­ ism con­se­quent on X chro­mo­some inac­ti­va­tion, all­het­ero­zy­gotes have some G6PD defi­cient red blood
cells: on aver­age 50%, but with a wide range of var­ia ­ ­tion. This is why in het­ero­zy­gotes clin­i­cal man­i­fes­ta­tions are, on aver­age, less severe, but not
always so. In every series of patients with fav­ism, there are always het­ero­zy­gous girls.19

cific mis­sense muta­tions cause, in addi­tion, a dev­as­tat­ing neu­
ro­log­i­cal syn­drome (type 2 dis­ease, prob­a­bly medi­ated by the
role of the enzyme in the bio­syn­the­sis of mye­lin phos­pho­lip­ids).
A point of prac­ti­cal impor­tance is that in patients with met­he­mo­
glo­bi­ne­mia (whether from CRB defi­ciency or from hemo­glo­bin
M) pulse oximetry mea­sured with bed­side devices is not reli­able.
CLINICAL CASE
A 12-year-old boy from Izmir, Turkey, the first­born from first-cousin
par­ents with nor­mal blood counts, was referred to an aca­demic
hema­tol­ogy depart­ment because of hemo­lytic ane­mia diag­nosed
at the age of 2. The ane­mia was mod­er­ate most of the time but with
exac­er­ba­tions, con­com­i­tant to infec­tion, that some­times required
blood trans­fu­sion. Physical exam­i­na­tion revealed pal­lor, jaun­dice,
hepa­to­meg­aly (2 cm below the cos­tal bor­der) and spleno­meg­aly
(4cm below the cos­tal bor­der). He had ane­mia, with a hemo­glo­
bin of 7.5 g/dL and a reticulocytosis of 160 × 109/L. White blood cell
count, plate­lets, and liver enzymes were within the nor­mal range. A
direct Coombs test was neg­a­tive. The patient showed an increased
uncon­ju­gated bil­i­ru­bin of 2.1 mg/dL (upper normal limit 0.8 mg/dL)
and lac­tate dehy­dro­ge­nase 678 IU (upper normal limit 300). Hapto-
globin was unde­tect­able. A periph­eral blood smear revealed aniso-
poikilocytosis, mac­ro­cytes, poly­chro­matic cells, rare elliptocytes
and spherocytes, and baso­philic stip­pling (Figure 4). Hemoglobin
elec­tro­pho­re­sis was nor­mal, and a test for unsta­ble hemo­glo­bin
was neg­a­tive. Folic acid and vita­min B12 lev­els were nor­mal. An
osmotic fra­gil­ity test was nor­mal. A bone mar­row aspi­rate showed
ery­throid hyper­pla­sia; an iron stain did not reveal sideroblasts. An
enzyme assay panel yielded nor­mal val­ues of gly­co­lytic enzymes
and of G6PD. The ultra­vi­o­let spec­trum of an extract of red blood
cell metab­o­lites showed an abnor­mal purine/pyrim­id ­ ine ratio, sug­
ges­tive of pyrim­i­dine 5′-nucle­o­tid­ase (P5N) defi­ciency. Sequencing
of the NT5C3 gene in the patient’s DNA revealed homo­zy­gos­ity for
a frame­shift muta­tion (c.393-394del TA, K132Rfs7*). Both par­ents
were het­ero­zy­gous for the same muta­tion.31
Courtesy of Dr Paola Bianchi.
Nucleotide metab­o­lism enzymopathies
P5N defi­ciency, although rare, ranks third in fre­quency among
enzymopathies (after G6PD defi­ ciency and PK defi­ ciency).32
P5N, encoded by the gene NT5C3A, is par­tic­u­larly impor­tant
dur­ing the last stage of ery­throid mat­u­ra­tion. Reticulocytes,
hav­ ing no RNA syn­ the­
sis, are unable to recy­ cle pyrim­ i­
dine
nucle­o­tides aris­ing from RNA deg­ra­da­tion (as would be the
case in other cells), and these must be hydro­lyzed to nucle­
o­sides before they can exit through the mem­brane. With P5N
defi­cient, encum­brance by pyrim­i­dine nucle­o­tides seems to
hin­der the abil­ity of retic­u­lo­cytes to get rid of their last ribo­
somes or of their rem­nants. These become vis­i­ble under the
shape of baso­philic stip­pling,33 a mor­pho­log­i­cal abnor­mal­ity
char­ac­ter­is­tic of P5N defi­ciency (Figure 4) but not unique to
it (see patient 1). The hema­to­logic con­se­quences of this en-
zymopathy are illus­trated by patient 3, who has a mod­er­ately
severe CNSHA.
The other (ultra­rare) enzymopathy in this group is adenylate
kinase (AK) defi­ciency. AK, by cat­a­lyz­ing the inter­con­ver­sion
of ADP into ATP and adenosine mono-phosphate (AMP), is gen­
er­ally impor­tant in con­trol­ling the level of these com­pounds.3
In red blood cells ATP is cru­cial for the cat­ion pump, and AMP,
a pre­cur­sor of the sig­nal­ing mol­e­cule cyclic AMP,3 may be
impor­tant in maintaining eryth­ro­cyte deformability.34 Indeed,
in AK defi­ciency there is evi­dence of intra­vas­cu­lar hemo­ly­sis.35
For both P5N and AK, it is through their respec­tive genetic
defects that we know they have an essen­tial func­tion in red
blood cells.
Conclusion
Red blood cell enzymopathies are now under­stood at the bio­
chem­i­cal and molec­u­lar lev­els, and their clin­i­cal and hema­to­
logic fea­tures are rea­son­ably well char­ac­ter­ized. G6PD defi­
ciency stands out in terms of epi­de­mi­­ol­ogy because it is a
wide­spread, mostly asymp­tom­atic abnor­mal­ity: it bears wit­
ness to the role of malaria in shap­ing human evo­lu­tion, and its
clin­i­cal man­i­fes­ta­tions are largely pre­vent­able and man­age­
able. Although rec­og­niz­ing chronic hemo­lytic ane­mia is not dif­
fi­cult, the other enzymopathies are prob­a­bly still underdiag-
nosed; DNA test­ing by gene pan­els should grad­u­ally over­come
this diag­nos­tic gap. In con­trast, the targeted treat­ment of red

350  |  Hematology 2021  |  ASH Education Program

blood cell enzymopathies is nearly uncharted ter­ri­tory. Since
most of them qual­ify as orphan dis­eases, there are incen­tives
to devel­op­ing new ther­a­pies.
Acknowledgments  ciency. Blood. 2020;136(11):1225-1240.
I have been for­tu­nate to work on red blood cell enzymopathies,
par­tic­u­larly G6PD defi­ciency, in 5 countries. Since I can­not list
all­those to whom I am grate­ful, I will only men­tion Olugbemiro
Sodeinde (Ibadan), the late Graziella Persico (Naples), Tom Vul-
liamy (London), Letizia Longo (New York), Rosario Notaro (Flor-
ence), and Stella Malangahe (Dar es Salaam). Research sup­port
was received from the World Health Organization, National Re-
search Council (Italy), Medical Research Council (UK), and Na-
tional Institutes of Health (in Nigeria and in the US). I am espe­
cially grate­ful to all­patients with enzymopathies for what I have
learned from them.
Conflict-of-inter­est dis­clo­sure
Lucio Luzzatto: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Lucio Luzzatto: nothing to disclose.
Correspondence
Lucio Luzzatto, Department of Hematology and Blood Transfusion,
Muhimbili University of Health and Allied Sciences, 65001 Dar es Sa-
laam, Tanzania; e-mail: lluzzatto@blood​­.ac​­.tz.
References
1. Luzzatto L. Hemolytic ane­mias. In: Jameson JL, Fauci AS, Kasper DL, Hauser
SL, Longo DL, Loscalzo J, eds. Harrison’s Principles of Internal Medicine.
Vol. 1. McGraw-Hill Education; 2018:708-723.
2. Grace RF, Glader B. Red blood cell enzyme dis­or­ders. Pediatr Clin North
Am. 2018;65(3):579-595.
3. Berg JM, Tymoczko JL, Stryer L. Biochemistry. 5th ed. W. H. Freeman and
Co.; 2002.
4. Alfinito F, Ferraro F, Rocco S, et  al. Glucose phos­phate isom­er­ase (GPI)
“Morcone”: a new var­i­ant from Italy. Eur J Haematol. 1994;52(5):263-266.
5. Baronciani L, Zanella A, Bianchi P, et  al. Study of the molec­u­lar defects
in glu­cose phos­phate isom­er­ase-defi­cient patients affected by chronic
hemo­lytic ane­mia. Blood. 1996;88(6):2306-2310.
6. Kanno H, Fujii H, Hirono A, et al. Molecular anal­y­sis of glu­cose phos­phate
isom­er­ase defi­ciency asso­ci­ated with hered­i­tary hemo­lytic ane­mia. Blood.
1996;88(6):2321-2325.
7. Vora S, Davidson M, Seaman C, et al. Heterogeneity of the molec­u­lar lesions
in inherited phos­pho­fruc­to­ki­nase defi­ciency. J Clin Invest. 1983;72(6):1995-
2006.
8. Zanella A, Fermo E, Bianchi P, Chiarelli LR, Valentini G. Pyruvate kinase defi­
ciency: the geno­type-phe­no­type asso­ci­a­tion. Blood Rev. 2007;21(4):217-
231.
9. Dacie JV. Part 1—The con­gen­i­tal anae­mias. In: The Haemolytic Anaemias.
Vol. 1. Churchill; 1960:1-339.
10. Roy NB, Wilson EA, Henderson S, et  al. A novel 33-gene targeted rese-
quencing panel pro­vi­des accu­rate, clin­i­cal-grade diag­no­sis and improves
patient man­ age­ ment for rare inherited anae­ mias. Br J Haematol.
2016;175(2):318-330.
11. Russo R, Andolfo I, Manna F, et al. Multi-gene panel test­ing improves diag­
no­sis and man­age­ment of patients with hered­i­tary ane­mias. Am J Hematol.
2018;93(5):672-682.
12. Grace RF, Barcellini W. Management of pyru­vate kinase defi­ciency in chil­
dren and adults. Blood. 2020;136(11):1241-1249.
13. van Beers EJ, van Straaten S, Morton DH, et al. Prevalence and man­age­ment
of iron over­load in pyru­vate kinase defi­ciency: report from the Pyruvate
Kinase Deficiency Natural History Study. Haematologica. 2019;104(2):e51-
e53.
14. Mojzikova R, Koralkova P, Holub D, et al. Iron sta­tus in patients with pyru­
vate kinase defi­ ciency: neo­ na­
tal hyperferritinaemia asso­ ci­
ated with a
Prakash Singh Shekhawat
novel frame­shift dele­tion in the PKLR gene (p.Arg518fs), and low hepcidin
to fer­ri­tin ratios. Br J Haematol. 2014;165(4):556-563.
15. Grace RF, Rose C, Layton DM, et al. Safety and effi­cacy of mitapivat in pyru­
vate kinase defi­ciency. N Engl J Med. 2019;381(10):933-944.
16. Luzzatto L, Ally M, Notaro R. Glucose-6-phos­phate dehy­dro­ge­nase defi­
17. Kaplan M, Hammerman C. The need for neo­na­tal glu­cose-6-phos­phate
dehy­dro­ge­nase screen­ing: a global per­spec­tive. J Perinatol. 2009;29(suppl
1):S46-S52.
18. Luzzatto L, Poggi VE. Glucose-6-phos­phate dehy­dro­ge­nase defi­ciency. In:
O’rkin SH, Fisher DE, Ginsburg D, Look TA, Lux SE, Nathan DG, eds. Nathan
and Oski’s Hematology and Oncology of Infancy and Childhood. Vol. 1.
Elsevier Saunders; 2015:609-629.
19. Luzzatto L, Arese P. Favism and glu­cose-6-phos­phate dehy­dro­ge­nase defi­
ciency. N Engl J Med. 2018;378(1):60-71.
20. Zaramella P, De Salvia A, Zaninotto M, et  al. Lethal effect of a sin­gle
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021
dose of rasburicase in a pre­term new­born infant. Pediatrics. 2013;131(1):
e309-e312.
21. Doxiadis SA, Fessas P, Valaes T. Glucose-6-phos­ phate dehy­ dro­ge­
nase
defi­ciency. A new aetiological fac­tor of severe neo­na­tal jaun­dice. Lancet.
1961;1(7172):297-301.
22. Capps FP, Gilles HM, Jolly H, Worlledge SM. Glucose-6-phos­phate dehy­
dro­ge­nase defi­ciency and neo­na­tal jaun­dice in Nigeria: their rela­tion to the
use of pro­phy­lac­tic vita­min K. Lancet. 1963;2(7304):379-383.
23. Bienzle U, Effiong C, Luzzatto L. Erythrocyte glu­cose 6-phos­phate dehy­
dro­ge­nase defi­ciency (G6PD type A−) and neo­na­tal jaun­dice. Acta Paedi-
atr Scand. 1976;65(6):701-703.
24. Kaplan M, Hammerman C. Glucose-6-phos­ phate dehy­ dro­ge­ nase defi­
ciency: a hid­den risk for kernicterus. Semin Perinatol. 2004;28(5):356-364.
25. Huang MJ, Kua KE, Teng HC, Tang KS, Weng HW, Huang CS. Risk fac­tors for
severe hyperbilirubinemia in neo­na­tes. Pediatr Res. 2004;56(5):682-689.
26. Reading NS, Sirdah MM, Shubair ME, et  al. Favism, the commonest form
of severe hemo­lytic ane­mia in Palestinian chil­dren, varies in sever­ity with
three dif­fer­ent var­i­ants of G6PD defi­ciency within the same com­mu­nity.
Blood Cells Mol Dis. 2016;60:58-64.
27. Francis RO, D’Alessandro A, Eisenberger A, et al. Donor glu­cose-6-phos­phate
dehy­dro­ge­nase defi­ciency decreases blood qual­ity for trans­fu­sion. J Clin
Invest. 2020;130(5):2270-2285.
28. Kamerbeek NM, van Zwieten R, de Boer M, et al. Molecular basis of glu­ta­thi­
one reduc­tase defi­ciency in human blood cells. Blood. 2007;109(8):3560-
3566.
29. Percy MJ, Lappin TR. Recessive con­gen­i­tal methaemoglobinaemia: cyto­
chrome b(5) reduc­tase defi­ciency. Br J Haematol. 2008;141(3):298-308.
30. Komuro A, Tobe T, Hashimoto K, et al. Molecular clon­ing and expres­sion of
human liver bil­i­ver­din-IX beta reduc­tase. Biol Pharm Bull. 1996;19(6):796-
804.
31. Köker SA, Oymak Y, Bianchi P, et  al. A new homo­ zy­gous muta­ tion
(c.393-394del TA/c.393-394del TA) in the NT5C3 gene asso­ ci­
ated with
pyrim­i­dine-5′-nucle­o­tid­ase defi­ciency: a case report. J Pediatr Hematol
Oncol. 2019;41(8):e484-e486.
32. Zanella A, Bianchi P, Fermo E, Valentini G. Hereditary pyrim­i­dine 5′-nucle­o­
tid­ase defi­ciency: from genet­ics to clin­i­cal man­i­fes­ta­tions. Br J Haematol.
2006;133(2):113-123.
33. Cheson BD, Rom WN, Webber RC. Basophilic stip­pling of red blood cells: a
non­spe­cific find­ing of mul­ti­ple eti­­ol­ogy. Am J Ind Med. 1984;5(4):327-334.
34. Semenov AN, Shirshin EA, Muravyov AV, Priezzhev AV. The effects of dif­fer­
ent sig­nal­ing path­ways in adenylyl cyclase stim­u­la­tion on red blood cells
deformability. Front Physiol. 2019;10:923.
35. Dongerdiye R, Kamat P, Jain P, et al. Red cell adenylate kinase defi­ciency in
India: iden­ti­fi­ca­tion of two novel mis­sense muta­tions (c.71A>G and c.413G>A).
J Clin Pathol. 2019;72(6):393-398.
36. van Wijk R, van Solinge WW. The energy-less red blood cell is lost: eryth­
ro­cyte enzyme abnor­mal­i­ties of gly­col­y­sis. Blood. 2005;106(13):4034-
4042.
37. Mamoune A, Bahuau M, Hamel Y, et al. A ther­mo­la­bile aldol­ase A mutant
causes fever-induced recur­rent rhabdomyolysis with­out hemo­lytic ane­
mia. PLoS Genet. 2014;10(11):e1004711.
38. Sakaue S, Kasai T, Mizuta I, et  al. Early-onset par­kin­son­ism in a ped­i­gree
with phos­pho­glyc­er­ate kinase defi­ciency and a het­ero­zy­gous car­rier: do
PGK-1 muta­tions con­trib­ute to vul­ner­a­bil­ity to par­kin­son­ism? NPJ Parkin-
sons Dis. 2017;3:13.
39. van Zwieten R, Verhoeven AJ, Roos D. Inborn defects in the anti­ox­i­dant
sys­tems of human red blood cells. Free Radic Biol Med. 2014;67:377-386.
Red blood cell enzymopathies  | 351
40. Soylu Ustkoyuncu P, Mutlu FT, Kiraz A, Tag Balkis Z, Yel S. A rare cause of
neo­na­tal hemo­lytic ane­mia: glu­ta­thi­one syn­the­tase defi­ciency. J Pediatr
Hematol Oncol. 2018;40(1):e45-e49.
41. Mañú Pereira M, Gelbart T, Ristoff E, et al. Chronic non-spherocytic hemo­
lytic ane­mia asso­ci­ated with severe neu­ro­log­i­cal dis­ease due to gam-
ma-glutamylcysteine syn­ the­tase defi­ ciency in a patient of Moroccan
ori­gin. Haematologica. 2007;92(11):e102-e105.
42. Luzzatto L. G6PD defi­ciency: a poly­mor­phism bal­anced by het­ero­zy­gote
advan­tage against malaria. Lancet Haematol. 2015;2(10):e400-e401.
43. Pal S, Bansil P, Bancone G, et  al. Evaluation of a novel quan­ti­ta­tive test
for glu­cose-6-phos­phate dehy­dro­ge­nase defi­ciency: bring­ing quan­ti­ta­tive
test­ing for glu­cose-6-phos­phate dehy­dro­ge­nase defi­ciency closer to the
patient. Am J Trop Med Hyg. 2019;100(1):213-221.
44. Chinevere TD, Murray CK, Grant E Jr, Johnson GA, Duelm F, Hospenthal
DR. Prevalence of glu­cose-6-phos­phate dehy­dro­ge­nase defi­ciency in U.S.
Army per­son­nel. Mil Med. 2006;171(9):905-907.
45. Luzzatto L, Allan NC. Relationship between the genes for glu­ cose-6-
phos­phate dehy­dro­ge­nase and for haemoglobin in a Nigerian pop­u­la­tion.
Nature. 1968;219(5158):1041-1042.
46. Cocco P, Todde P, Fornera S, Manca MB, Manca P, Sias AR. Mortality in a
cohort of men expressing the glu­cose-6-phos­phate dehy­dro­ge­nase defi­
ciency. Blood. 1998;91(2):706-709.
352  |  Hematology 2021  |  ASH Education Program
47. Bancone G, Menard D, Khim N, et al. Molecular char­ac­ter­iza­tion and map­
ping of glu­ cose-6-phos­phate dehy­ dro­ ge­
nase (G6PD) muta­ tions in the
Greater Mekong Subregion. Malar J. 2019;18(1):20.
48. van Vuren AJ, van Beers EJ, van Wijk R. A pro­posed con­cept for defec­tive
mitophagy lead­ing to late stage inef­fec­tive eryth­ro­poi­e­sis in pyru­vate
kinase defi­ciency. Front Physiol. 2020;11:609103.
49. Mahendra P, Dollery CT, Luzzatto L, Bloom SR. Pyruvate kinase defi­ciency.
Association with G6PD defi­ciency. BMJ. 1992;305(6856):760-762.
50. Pamba A, Richardson ND, Carter N, et al. Clinical spec­trum and sever­ity of
hemo­lytic ane­mia in glu­cose 6-phos­phate dehy­dro­ge­nase-defi­cient chil­
dren receiv­ing dap­sone. Blood. 2012;120(20):4123-4133.
51. Rees DC, Duley JA, Marinaki AM. Pyrimidine 5′ nucle­o­tid­ase defi­ciency. Br J
Haematol. 2003;120(3):375-383.
52. Kung C, Hixon J, Kosinski PA, et al. AG-348 enhances pyru­vate kinase activ­
ity in red blood cells from patients with pyru­vate kinase defi­ciency. Blood.
2017;130(11):1347-1356.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/341/1851446/341luzzatto.pdf by guest on 13 December 2021
53. Luzzatto L, Nannelli C, Notaro R. Glucose-6-phos­ phate dehy­ dro­
ge­nase
defi­ciency. Hematol Oncol Clin North Am. 2016;30(2):373-393.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000266
 INHERITED RED CELL DISORDERS BEYOND HEMOGLOBINOPATHIES

Diamond-Blackfan anemia
Lydie M. Da Costa,1–4 Isabelle Marie,1,5 and Thierry M. Leblanc5

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/353/1851415/353costa.pdf by guest on 13 December 2021

Service d’Hématologie Biologique (Hematology Diagnostic Lab), AP­HP, Hôpital Robert Debré, Paris, France; 2University of Paris, Paris, France;
1

HEMATIM EA4666, Amiens, France; 4Laboratory of Excellence for Red Cells, LABEX GR­Ex, Paris, France; and 5Immuno­Hematology Department,
3

AP­HP, Hôpital Robert­Debré, Paris, France

Diamond-Blackfan anemia (DBA) is an inherited bone marrow failure syndrome, characterized as a rare congenital bone
marrow erythroid hypoplasia (OMIM#105650). Erythroid defect in DBA results in erythroblastopenia in bone marrow as
a consequence of maturation blockade between the burst forming unit–erythroid and colony forming unit–erythroid
developmental stages, leading to moderate to severe usually macrocytic aregenerative (<20×109/L of reticulocytes)
anemia. Congenital malformations localized mostly in the cephalic area and in the extremities (thumbs), as well as short
stature and cardiac and urogenital tract abnormalities, are a feature of 50% of the DBA-affected patients. A significant
increased risk for malignancy has been reported. DBA is due to a defect in the ribosomal RNA (rRNA) maturation as a
consequence of a heterozygous mutation in 1 of the 20 ribosomal protein genes. Besides classical DBA, some DBA-like
diseases have been identified. The relation between the defect in rRNA maturation and the erythroid defect in DBA has
yet to be fully defined. However, recent studies have identified a role for GATA1 either due to a specific defect in its
translation or due to its defective regulation by its chaperone HSP70. In addition, excess free heme-induced reactive oxy-
gen species and apoptosis have been implicated in the DBA erythroid phenotype. Current treatment options are either
regular transfusions with appropriate iron chelation or treatment with corticosteroids starting at 1 year of age. The only
curative treatment for the anemia of DBA to date is bone marrow transplantation. Use of gene therapy as a therapeutic
strategy is currently being explored.

LEARNING OBJECTIVES
• Recognize a case of DBA
• Manage anemia and potential clinical complications of DBA

CLINICAL CASE
The family described below is an excellent illustration of
the various issues that are encountered in the diagnosis
and treatment of Diamond­Blackfan anemia (DBA) due to
incomplete penetrance.
The family is the largest of the 417 DBA families regis­
tered in the French DBA registry (Observatoire Français
de l’Anémie de Blackfan­Diamond) and includes 3 gen­
erations of affected patients (Figure 1). In this family, the
mode of inheritance is familial, as is the case for 45% of the
DBA­affected patients registered in the different registries
around the world. In 55% of individuals, DBA is the result of
sporadic or de novo inheritance. Interestingly, in the illus­
trated family, all 4 children, 2 girls and 2 boys, developed
DBA with different phenotypes (Figure 1). Of note, in the
1980s, no DBA gene had been identifed, and steroids were
started following DBA diagnosis in all instances, which is
not the recommended treatment currently: steroids should
be initiated only after the frst year of life.1
1. The child, UPN#1447, at age 2.3 years with normal white
and platelet cell counts, was initially diagnosed with
transient erythroblastopenia of childhood (TEC) on the
basis of an isolated severe normochromic, macrocytic
mean corpuscular volume (MCV=91.4 fL) aregenerative
(7.5×109/L reticulocytes) anemia (hemoglobin (Hb) level
at 79g/L) with erythroblastopenia documented on the
bone marrow smear (1% of total erythroblasts [Eb, nor­
mal 5%­30%]). Bone marrow cellularity was normal with
no signs of dysplasia. Parvovirus B19 serology was nega­
tive for both immunoglobulin M and immunoglobulin G.

Prakash Singh Shekhawat

Diamond­Blackfan anemia, from a clinical case | 353

Figure 1. Family tree of the described DBA family.
The patient recov­ered with no treat­ment except for a short
course of ste­roid treat­ment (for a few weeks) with no red cell
blood trans­fu­sions. At the same time, increased expres­sion
of hemo­glo­bin F (HbF) at 9% (nor­mal <2% at 6-24 months of
age) along with a sig­nif­i­cantly ele­vated eryth­ro­cyte aden­o­
sine deam­in ­ ase (eADA) activ­ity at 5.95 nmol/min/mg Hb (or
U/g Hb) (nor­mal: 1.50 ± 0.2) has been reported. UPN#1447, ini­
tially diag­nosed with TEC, has been finally diag­nosed with
DBA after the iden­ti­fi­ca­tion of a RPS19 gene muta­tion. In ret­
ro­spect, the asso­ci­a­tion of pure red blood cell hypo­plas­tic
ane­mia, in con­junc­tion with an increased per­cent­age of HbF
and eADA activ­ity, and response to ste­roid was in accor­dance
with this diag­no­sis.1,2 However, it is to be noted some cli­ni­
cians con­sider TEC to be one of the DBA phe­no­types with
a low pen­e­trance.3 TEC and DBA should be dif­fer­en­ti­ated,
but it is some­times very dif­fi­cult to dis­crim­i­nate between
the 2 phe­no­types as only the time course of the evo­lu­tion
of ane­mia can clearly dis­tin­guish between them (see Table 1
for dif­fer­en­tial diag­no­sis between TEC and DBA). In order to
be cau­tious, we rec­om­mend a molec­u­lar screen­ing and the
care­ful fol­low-up of any child with TEC or inform the par­ents
to bring back the child for fur­ther eval­u­a­tion in case of recur­
rence of the ane­mia (Table 2).
. UPN#1821 was the sec­ond child of the fam­ily, a girl born 2 years
2 DBA-affected patient. He was diag­nosed at 1 month and 3
later. In con­trast to UPN#1447, UPN#1821 exhibited a 2-month-
old, severe ane­mia with a nadir of 20 g/L Hb. The ane­mia was
normochromic and mac­ro­cytic (high MCV of 114.8 fL), which is
clas­si­cal for DBA, and with a mod­est degree of regen­er­a­tion
354  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/353/1851415/353costa.pdf by guest on 13 December 2021
(60 × 109/L retic­u­lo­cytes). The char­ac­ter­is­tic erythroblastope­
nia was noted on the bone mar­row smear with only 1% of Eb.
UPN#1821 exhibited fea­tures of fetal eryth­ro­poi­e­sis with HbF
at 18% and a large increase in eADA activ­ity at 5 nmol/min/mg
Hb. A val­gus foot defor­mity was noted as the only con­gen­i­tal
malformation. UPN#1821 received a reg­u­lar dose of 2 mg/kg/d
of ste­roid for a month, which allowed for trans­fu­sion inde­pen­
dency, and was sub­se­quently maintained on low-dose ste­roid
ther­apy (0.1  mg/kg/d). However, she became later ste­ roid
resis­tant, and she is cur­rently being reg­u­larly trans­fused every
3 weeks (Table 2).
3. UPN#1822 was diag­nosed with DBA at 1 month of age. Surpris­
ingly at this time, the bone mar­row smear did not exhibit the
char­ac­ter­is­tic DBA erythroblastopenia with 31% of ery­throid
pre­cur­sors. However, a large increase in eADA activ­ ity of
4 nmol/min/mg Hb was noted. Posterior hypo­spa­dias was no­
ticed. He received an ini­tial dose of ste­roid at 2 mg/kg/d, and
the dose was grad­u­ally decreased to 0.2 mg/kg/d, which en­
abled the sta­bi­li­za­tion of the Hb level. The ste­roid treat­ment
lasted for 18 years and 10 months, at which point the patient
became ste­roid inde­pen­dent and main­tains a rea­son­able Hb
level with no need for ste­roids (Table 2).
4. Finally, UPN#1213, the fourth child of this fam­ily, was also a
weeks of age with erythroblastopenia with 4% of Eb in the bone
mar­row. Like his sib­lings, his eADA activ­ity was increased to
4.38 nmol/min/mg Hb. Hypospadias was once again noticed.
He was treated with an ini­tial dose of 2 mg/kg/d of ste­roid,
Table 1. Differential diag­no­sis between DBA and tran­sient TEC

Characteristic DBA TEC

Median age at diag­no­sis 2 months >1 year
Inheritance Sporadic (55%) or dom­i­nant (45%) Not inherited
Congenital anom­a­lies In 50% None
Pure red blood cell aplasia (bone mar­row Yes Yes
biopsy) or erythroblastopenia (bone
mar­row aspi­ra­tion)
Hb level Low Low
Reticulocyte count <20 × 109/L <20 × 109/L
MCV Usually high Normal

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/353/1851415/353costa.pdf by guest on 13 December 2021

eADA activ­ity Normal to high Normal
HbF Normal to high Normal
Allelic var­i­a­tion in a RP gene or another 70% to 80% of the patients with No muta­tion found
gene involved in DBA-like cases DBA

which nor­mal­ized the Hb level. He is now 32 years old and
still under­go­ing ste­roid ther­apy at 10 mg per day. At the last
fol­low-up, he exhibited a mod­er­ate mac­ro­cytic (MCV at 105.9
fL) ane­mia with an Hb level at 108 g/L with 38.6 × 109/L retic­u­lo­
cytes, in con­junc­tion with nor­mal white blood cell and plate­let
counts (Table 2).
Clinical and bio­log­i­cal pre­sen­ta­tion of DBA: take-home
mes­sage
DBA is usu­ally char­ac­ter­ized by a mod­er­ate to severe, mac­ro­cytic
aregenerative ane­mia. The other cell lin­e­ages are usu­ally nor­mal,
but on occa­ sion, neutropenia, throm­ bo­cy­ to­
pe­
nia, and, in some
instances, even thrombocytosis may be noted at the time of diag­no­
sis. The erythroblastopenia (<5% of ery­throid pre­cur­sors) in an oth­
er­wise nor­mal bone mar­row (no dys­pla­sia and nor­mal cel­lu­lar­ity) on
the bone mar­row smear or pure hypo­plas­tic ane­mia on bone mar­
row biopsy can con­firm the diag­no­sis. Bone mar­row exam­i­na­tion is
man­da­tory to avoid mis­di­ag­no­sis. Erythroblastopenia is the con­se­
quence of block­ade in ery­throid dif­fer­en­ti­a­tion between the burst
forming unit–ery­throid and col­ony forming unit–ery­throid pro­gen­
i­tor stages.4 The other bio­log­i­cal fea­tures include increased eADA
activ­ity, which is ele­vated in 90% of the nontransfused patients with
DBA and the per­sis­tence of fea­tures of fetal eryth­ro­poi­e­sis (high HbF
per­cent­age). They should both be mea­sured at diag­no­sis before
trans­fu­sion or at least 3 months fol­low­ing trans­fu­sion (Table 3). In
50% of the patients with DBA, var­i­ous malformations are reported
mostly in the cephalic area and the extrem­i­ties, with the clas­si­cal
but rare triphalangeal thumbs (Table 4).1,5
Going back to the herein reported cases, the molec­ u­
lar
screen­ing of the pro­band (UPN#1447) and her sis­ter (UPN#1821)
and 2 broth­ers (UPN#1822 and UPN#1213) iden­ti­fied a het­ero­zy­
gous 4–base pair dele­tion near the donor splice site of exon 2 in
the RPS19 gene (NM_001022.3: c.71 + 3_71 + 6del; p.?). This var­i­ant
has indeed been found in the puta­tive TEC case (UPN#1447), and
it should have been enough to rule out TEC. This alle­lic var­i­a­tion
was found in the nonanemic (Hb 121 g/L; 78.4 × 109/L retic­u­lo­cytes)
mother, who is con­sid­ered a so-called silent phe­no­type, another
dis­tinct fea­ture of DBA. In addi­tion, the mother exhibited a nor­mal
MCV (96 fL) and an ele­vated eADA level (4.21 nmol/min/mg Hb).
Silent DBA phe­no­type indi­vid­u­als could be either the ­par­ent or

Table 2. Phenotype of the 4 patients with DBA

Characteristic UPN#1447 UPN#1821 UPN#1822 UPN#1213

Age, mo* 27.6 2 1 1.8
Hb, g/L* 79 20 NA NA
MCV, fL* 91.4 114.8 NA NA
Reticulocyte count, × 10 /L*
9
7.5 60 NA NA
eADA, nmol/min/mg Hb* 5.95 5 4 4.38
HbF, %* 9 18 NA NA
Bone mar­row 1% total Eb 1% total Eb 31 4
Congenital abnor­mal­i­ties None Valgus foot Posterior hypo­spa­dias Hypospadias
Hbat the last fol­low-up, g/L 110 90 NA 108
Treatment at the last fol­ Treatment inde­pen­dence Transfusion Treatment inde­pen­dence Steroid
low-up (deceased)
*At pre­sen­ta­tion.
NA, not avail­­able.
Prakash Singh Shekhawat
Diamond-Blackfan ane­mia, from a clin­i­cal case  |  355
Table 3. Diagnosis cri­te­ria of DBA (from the inter­na­tional sys­tolic heart mur­mur due to ane­mia at 120 g/L (nor­mal value
con­sen­sus con­fer­ence1) >140 g/L), which decreased to 75 g/L at day 18 after birth. He
was trans­fused with 70 mL of red blood cells. He has since been
Diagnosis cri­te­ria Age less than 1 year reg­u­larly trans­fused every 3 to 4 weeks. He did not pres­ent any
Macrocytic ane­mia with no other sig­nif­i­cant con­gen­i­tal abnor­mal­i­ties. After 1 year as recommended, ste­
cytopenias roid ther­apy was started, but there was no response. Chelation
ther­apy with deferasirox was started but had to be interrupted
Reticulocytopenia
due to liver tox­ic­ity. Deferoxamine treat­ment (1000 mg/d) by
Normal mar­row cel­lu­lar­ity with a pau­city of nightly infu­sion over 10 hours, 5 out of 7 days, was started to
ery­throid pre­cur­sors
Supporting
cri­te­ria
Table 4. Malformations iden­ti­fied in DBA from Willig et al5
 Major Gene muta­tion described in “clas­si­cal” DBA

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/353/1851415/353costa.pdf by guest on 13 December 2021

Positive fam­ily his­tory Craniofacial Hypertelorism
 Minor Elevated eADA activ­ity Broad, flat nasal bridge
Congenital anom­a­lies described in “clas­si­cal” DBA Cleft pal­ate
Elevated HbF High arched pal­ate
Noevi­dence of another inherited bone mar­row Microcephaly
fail­ure syn­drome
Micrognathia
Microtia
a sib­ling of a DBA pro­band: they are not ane­mic but may exhibit Low-set ears
a mac­ro­cy­to­sis, an ele­vated eADA, and/or a muta­tion in a ribo­
Low hair­line
somal pro­tein (RP) gene. Patients with the “silent phe­no­type”
have a risk of DBA com­pli­ca­tions such as malig­nan­cies and may Epicanthus
trans­mit the dis­ease as dem­on­strated in this fam­ily: the mother Ptosis
has trans­mit­ted the var­i­ant to all 4 of her off­spring, reinforcing the Ophthalmologic Congenital glau­coma
dom­i­nant inher­i­tance in this fam­ily. DBA has been diag­nosed in 2
Strabismus
mem­bers of the third gen­er­at­ion in this fam­ily. UPN#1822 expe­
ri­enced a fetal loss (UPN#1660) in utero at 23 weeks +3 days of Congenital cat­a­ract
a female nondysmorphic fetus exhibiting a severe intra­uter­ine Neck Short neck
growth retar­da­tion (weight: 436.5 g, which is between the 5th
Webbed neck
and the 10th per­cen­tiles; head cir­cum­fer­ence: 18 cm, 5th per­cen­
tile; height [ver­tex/coc­cyx]: 20 cm, 50th per­cen­tile for 23--week Sprengel defor­mity
amen­or­rhea) with oligohydramnios but with­out hydrops fetalis. Klippel-Feil defor­mity
Interestingly, the pla­centa exhibited defects in the ves­sels with Thumbs Triphalangeal
ste­no­ sis, and some villi were hypovascularized. However, no
Duplex or bifid
erythroblastopenia was noted. The same alle­lic var­i­a­tion has been
iden­ti­fied in this fetus. We assume that the major growth retar­ Hypoplastic
da­tion in rela­tion with DBA may be the ori­gin of the fetal death Flat thenar emi­nence
in utero (Table 5). Hydrops fetalis is a fea­ture of DBA, and its fre­
Absent radial artery
quency is likely underestimated. Since our descrip­tion of the first
case of spo­radic fetal loss due a muta­tion in the RPS19 gene,6 we Urogenital Absent kid­ney
sub­se­quently iden­ti­fied addi­tional fetal cases mostly in asso­ci­a­ Horseshoe kid­ney
tion with the RPS19 gene and also in asso­ci­at­ion with the RPL15 Hypospadias
gene.7 These find­ings rein­force, as we stated ear­lier,8 that major
Cardiac Ventricular sep­tal
com­pli­ca­tions can occur dur­ing preg­nancy and the impor­tance of defect
the fol­low-up of the mother and her off­spring dur­ing preg­nancy
Atrial sep­tal defect
with puta­tive aspi­rin treat­ment in order to pre­vent the pla­centa
vas­cu­lar com­pli­ca­tions as noted in our case.8 Coarctation of the
In addi­tion to UPN#1822, UPN#1821 also had 5 off­spring, aorta
among them twins (UPN#779, UNP#1114, UPN#1264, UPN#1573, Complex car­diac
UNP#1574) (Figure 1). Prenatal diag­no­sis was declined, and out anom­a­lies
of the 5 chil­dren, only 1 boy was affected by DBA and car­ried Other mus­cu­lo­skel­e­tal Growth retar­da­tion
the same famil­ial alle­lic var­i­a­tion in the RPS19 gene. UNP#1264 Syndactyly
was born at a nor­mal ges­ta­tional time of 40 weeks with nor­mal
Neuromotor Learning dif­fi­cul­ties
men­su­ra­tion but a con­junc­ti­val and skin pal­lor, tachy­car­dia, and

356  |  Hematology 2021  |  ASH Education Program

Table 5. Mode of DBA rev­e­la­tion take-home mes­sage
Median age at 2 months; neo­na­tes (16% of the cases); hydrops fetalis
Aregenerative usu­ally mac­ro­cytic ane­mia
Erythroblastopenia in an oth­er­wise normocellular bone mar­row
Malformations in par­tic­u­lar cleft pal­ate, thumbs anom­a­lies, heart and uro­gen­i­tal tract anom­a­lies
Short stat­ure, includ­ing intra­uter­ine growth retar­da­tion
eADA ele­va­tion
Fetal eryth­ro­poi­e­sis fea­tures (ele­vated HbF per­cent­age after 6 months)
Complications of preg­nancy
Malignancies
Very rarely aplastic ane­mia
man­age iron over­load. Stem cell trans­plan­ta­tion was con­tem­
plated, and the sib­lings were tested for the known famil­ial alle­
lic var­i­a­tion, which was not found in any of them. Unfortunately,
none of them was HLA iden­ti­cal. He is cur­rently 6 years old
and has been recently under­gone trans­plan­ta­tion with a fully
matched unre­lated donor.
DBA treat­ment: take-home mes­sage
DBA treat­ment in 2021 still relies on either chronic blood trans­fu­
sions (Hb con­cen­tra­tion to be maintained >90 g/L) with opti­mal
iron che­la­tion ther­apy started when the fer­ri­tin level is >500 µg/L
or treat­ment with cor­ti­co­ste­roids.9-11 The cor­ti­co­ste­roids (pred­
ni­sone or pred­nis­o­lone) are to be intro­duced only after the first
year of life in order to enable the max­i­mal growth because short
stat­ure is part of the malformative syn­drome. The cor­ti­co­ste­
roids are intro­duced at a dose of 2 mg/kg/d, and their effi­cacy
is usu­ally seen starting at 2 weeks fol­low­ing ini­ti­a­tion of ther­
apy, reflected by increased retic­u­lo­cyte count and Hb level. If
there is no response to ste­roid ther­apy after 1 month, there is no
rea­son to con­tinue this ther­apy. In case of effi­cacy, the cor­ti­co­
ste­roid should be grad­u­ally decreased to a min­i­mum dose that
can main­tain the Hb level above 90 g/L. The max­i­mal con­tin­u­
ous cor­ti­co­ste­roid dose must be <0.3 mg/kg/d (<0.5 mg/kg/d
in the countries where access to trans­fu­sions is dif­fi­cult or dan­
ger­ous). In case of corticoresistance or corticodependence (>0.3
[or >0.5 mg/kg/d]), trans­fu­sions with iron che­la­tion are cur­rently
the only alter­na­tive option of treat­ment,9-11 and hema­to­poi­etic
stem cell trans­plan­ta­tion (HSCT) is indi­cated fol­low­ing the avail­
abil­ity of either an HLA-iden­ti­cal sib­ling in whom DBA or the
silent phe­no­type has been excluded12-15 or with a fully matched
(10/10) unre­lated donor.14,16 HSCT should be performed ide­ally
before the age of 5 years and cer­tainly <10 years to avoid HSCT
com­ pli­
ca­
tions. HSCT is also indi­ cated in patients with clonal
evo­lu­tion.12-15 An alter­na­tive source of hema­to­poi­etic stem cells
should be con­sid­ered as exper­i­men­tal approaches but may be
indi­cated for patients with clonal evo­lu­tion. HSCT cures ane­mia
and pre­vents the risk of myelodysplastic syn­drome (MDS) and
leu­ke­mia, but patients with DBA still need to be care­fully mon­i­
tored for the risk of posttransplant solid tumors.17
Last, leu­cine has been found to be effec­tive in small num­ber
of patients with DBA.18 New ther­a­peu­tic options such as gene
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/353/1851415/353costa.pdf by guest on 13 December 2021
ther­apy are prob­a­bly the most prom­is­ing option,19,20 along with
other targeted ther­a­pies.21 Recently, some inno­va­tive ther­a­pies
have been pro­posed such as cal­mod­u­lin inhib­i­tors16 and met­
formin,22 and it is likely that other new ther­a­peu­tic options will
emerge from trans­la­tional research (trial NCT03966053 with tri­
fluo­per­a­zine).23-26
Genetics in DBA: take-home mes­sage
An alle­lic var­i­a­tion, always in a het­ero­zy­gous state in a RP, is found
in approx­i­ma­tely 70% to 80% of the DBA-affected cases.27-30
Mutations in 20 RP genes have been iden­ti­fied (Figure 2). The
most fre­quently mutated gene and the one iden­ti­fied in our
clin­i­cal case is the RPS19 gene, which accounts for 25% of DBA
cases glob­ally.31 Large dele­tions in these RP genes have been
found in approx­i­ma­tely 20% of the DBA cases, mostly in RPS17,
RPL35a, and RPS19 genes.32 Thus, at least 70% of the patients
with DBA carry an alle­lic var­i­a­tion, includ­ing a large dele­tion in
only 8 RP genes, namely, RPS19, RPL5, RPS26, RPL11, RPL35a,
RPS10, RPS24, and RPS17 (Figure 2). Multiple path­o­genic RP
muta­tions in a patient with DBA have not been reported to
date. Phenotypic/geno­typic cor­re­la­tion is not read­ily evi­
dent, although it is well established that patients with DBA
car­ry­ing a RPS19 gene muta­tion exhibit less malformation
com­pared with oth­ers but appear to exhibit a more severe
hema­ to­logic phe­ no­ type and are fre­ quently man­ aged by a
trans­fu­sion pro­gram.33 Neutropenia is more fre­quently asso­
ci­ated with RPL35a,34 car­diac anom­a­lies with RPS24,35 cleft
pal­ate with RPL5, and thumbs anom­a­lies with RPL1136 gene
muta­tions. Other RP genes have been found to be mutated
in small sub­sets of DBA-affected patients, each affect­ing <1%
of DBA cohorts (Figure 2). A muta­tion in a RP gene involved in
DBA is asso­ci­ated with defec­tive ribo­somal RNA (rRNA) mat­
u­ra­tion, which is con­sid­ered the sig­na­ture fea­ture of the DBA
dis­ease, and its doc­u­men­ta­tion con­firms the path­o­ge­nic­ity of
alle­lic var­i­a­tion of unknown sig­nif­i­cance. DBA-like dis­ease is
respon­si­ble for ane­mia with erythroblastopenia and a cer­tain
degree of ste­roid response but with the absence of a defect in
rRNA mat­u­ra­tion. These DBA-like dis­eases are related to EPO37
and GATA-138,39 gene muta­tions. The DBA and DBA-like dis­eases
con­sti­tute the DBA syn­drome. Despite erythroblastopenia and
a cer­tain degree of ste­roid response, ADA2 defi­ciency related
Diamond-Blackfan ane­mia, from a clin­i­cal case  |  357

Figure 2. Genes mutated in DBA and DBA-like cases.
to CECR1 or ADA2 gene muta­tions is not con­sid­ered to belong
to DBA syn­drome (Figure 2).30,40,41
The DBA fam­ily we described has been affected by malig­nan­
cies. The mother, who rep­re­sents a DBA silent phe­no­type, was
diag­nosed with breast can­cer that was rap­idly fatal. The oldest
off­spring, UNP#1447, car­ry­ing the famil­ial RPS19 alle­lic var­i­a­tion but
never need­ing any treat­ment, was diag­nosed at age 36 years with
a rec­tal ade­no­car­ci­noma and passed away fol­low­ing 16 months
of treat­ment. These obser­va­tions point to the fact that iden­ti­fy­
ing patients with DBA (and not misdiagnosing them with TEC) is
impor­tant and that the so-called silent DBA phe­no­types should
be con­sid­ered in patients with DBA in their fol­low-up, par­tic­u­larly
for the risk of malig­nan­cies. A recent study established a risk of
5% for malig­nan­cies in nontransplanted patients with DBA in the
National Cancer Institute cohort17 and in the Amer­i­can DBA reg­is­
try42,43 with an observed/expected ratio of 4.8 for any malig­nancy,
44.7 for colon car­ci­noma, 9.4 for lung can­cer, 42.4 for oste­o­genic
sar­coma, 352 for MDS, and 28.8 for acute mye­loid leu­ke­mia.42,43
However, com­pared with the other inherited bone mar­row fail­
ure syn­dromes, the observed/expected ratio is lower for DBA.17
To date, there are no guide­lines for MDS/acute mye­loid leu­ke­mia
and solid tumor screen­ing, except recent pre­lim­i­nary rec­om­men­
da­tions on colo­rec­tal car­ci­noma,44 but this may be warranted and
is cur­rently being discussed among DBA coop­er­a­tive groups.
In con­clu­sion, DBA is a fas­ci­nat­ing and com­plex ery­throid
dis­or­der, as illus­
trated from the study of the DBA fam­ ily we
358  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/353/1851415/353costa.pdf by guest on 13 December 2021
described. Progress is being made in our under­stand­ing of the
molec­u­lar basis for DBA, path­o­phys­i­­ol­ogy of the dis­ease, and
devel­op­ing and pur­su­ing new ther­a­peu­tic options. We antic­i­
pate that these advances will enable bet­ter clin­i­cal man­age­ment
of the patients with DBA in the com­ing years.
Acknowledgments 
Our spe­cial thanks go to the affected indi­vid­ua ­ ls, their fam­i­lies,
and the DBA French patient asso­ci­a­tion (Association Française
de la Maladie de Blackfan-Diamond [AFMBD], pres­ i­
dent Mar­
cel Hibert). We sin­cerely thank our 2 men­tors, Gil Tchernia and
Mohandas Narla, for their sup­port, friend­ship, and many fruit­
ful sci­en­tific dis­cus­sions. We also thank all­ our col­lab­o­ra­tors in
France and around the world involved in the muta­tional screen­
ing anal­y­sis in France (Cécile Masson, Christine Bole, and the
team from the Genomics Core Facility, Institut Imagine-Structure
Fédérative de Recherche Necker, and Anaëlle Jaouen, Ludivine
David, Julie Galimand, and Hélène Bourdeau, tech­ni­cians from
the hema­tol­ogy lab in R. Debré hos­pi­tal, Paris). L.M.D.C. is sup­
ported by #ANR EJPRD/ANR-19-RAR4-0016 (Euro­ pean Union’s
Horizon 2020 research and inno­va­tion pro­gram under the EJP
RD COFUND-EJP No 825575), the Laboratory of Excellence for
Red Cells ((LABEX GR-Ex)-ANR Avenir-11-LABX-0005-02), and the
French National PHRC OFABD (DBA reg­is­try). T.M.L. and L.M.D.C.
are supported by the ANR grant EJPRD DBAGenCure (coor­di­na­
tors Juan Ruben and Susana Navarro Ordonez).
Conflict-of-inter­est dis­clo­sure
Lydie M. Da Costa: no com­pet­ing finan­cial inter­ests to declare.
Isabelle Marie: no com­pet­ing finan­cial inter­ests to declare.
Thierry M. Leblanc: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Lydie M. Da Costa: nothing to disclose.
Isabelle Marie: nothing to disclose.
Thierry M. Leblanc: nothing to disclose.
Correspondence
Lydie M. Da Costa, Service d’Hématologie Biologique (Hematol­
ogy Diagnostic Lab), AP-HP, Hôpital Robert Debré, 48 bou­le­vard
Serurier, Paris 75019, France; e-mail: lydie​­.dacosta@rdb​­.aphp​­.fr.
References
1. Vlachos A, Ball S, Dahl N, et al; Participants of Sixth Annual Daniella Maria
Arturi International Consensus Conference. Diagnosing and treating Dia­
mond Blackfan anae­ mia: results of an inter­ na­tional clin­
i­
cal con­sen­sus
con­fer­ence. Br J Haematol. 2008;142(6):859-876.
2. Willig TN, Pérignon JL, Gustavsson P, et  al. High aden­o­sine deam­i­nase
level among healthy pro­bands of Diamond Blackfan ane­mia (DBA) coseg­
regates with the DBA gene region on chro­ mo­ some 19q13. The DBA
Working Group of Société d’Immunologie Pédiatrique (SHIP). Blood.
1998;92(11):4422-4427.
3. Gustavsson P, Klar J, Matsson H, et al. Familial tran­sient erythroblastopenia
of child­hood is asso­ci­ated with the chro­mo­some 19q13.2 region but not
caused by muta­tions in cod­ing sequences of the ribo­somal pro­tein S19
(RPS19) gene. Br J Haematol. 2002;119(1):261-264.
4. Ohene-Abuakwa Y, Orfali KA, Marius C, Ball SE. Two-phase cul­ture in Dia­
mond Blackfan ane­mia: local­i­za­tion of ery­throid defect. Blood. 2005;105
(2):838-846.
5. Willig T-N, Niemeyer CM, Leblanc T, et al. Identification of new prog­no­sis
fac­tors from the clin­i­cal and epi­de­mi­o­logic anal­y­sis of a reg­is­try of 229
Diamond-Blackfan ane­mia patients. Pediatr Res. 1999;46(5):553.
6. Da Costa L, Chanoz-Poulard G, Simansour M, et al. First de novo muta­tion
in RPS19 gene as the cause of hydrops fetalis in Diamond-Blackfan ane­mia.
Am J Hematol. 2013;88(4):340-341.
7. Wlodarski MW, Da Costa L, O’Donohue M-F, et  al. Recurring muta­tions
in RPL15 are linked to hydrops fetalis and treat­ment inde­pen­dence in
­Diamond-Blackfan ane­mia. Haematologica. 2018;103(6):949-958.
8. Faivre L, Meerpohl J, Da Costa L, et  al. High-risk preg­ nan­cies in Dia­
mond-Blackfan ane­mia: a sur­vey of 64 preg­nan­cies from the French and
Ger­man reg­is­tries. Haematologica. 2006;91(4):530-533.
9. Berdoukas V, Nord A, Carson S, et al. Tissue iron eval­u­a­tion in chron­i­cally
trans­fused chil­dren shows sig­nif­i­cant lev­els of iron load­ing at a very young
age. Am J Hematol. 2013;88(11):E283-E285.
10. Roggero S, Quarello P, Vinciguerra T, Longo F, Piga A, Ramenghi U. Severe
iron over­ load in Blackfan-Diamond ane­ mia: a case-con­ trol study. Am J
Hematol. 2009;84(11):729-732.
11. Vlachos A, Muir E. How I treat Diamond-Blackfan ane­mia. Blood. 2010;116
(19):3715-3723.
12. Strahm B, Loewecke F, Niemeyer CM, et al. Favorable out­comes of hema­
to­poi­etic stem cell trans­plan­ta­tion in chil­dren and ado­les­cents with
­Diamond-Blackfan ane­mia. Blood Adv. 2020;4(8):1760-1769.
13. Fagioli F, Quarello P, Zecca M, et al. Haematopoietic stem cell trans­plan­
ta­tion for Diamond Blackfan anae­mia: a report from the Ital­ian Associa­
tion of Paediatric Haematology and Oncology Registry. Br J Haematol.
2014;165(5):673-681.
14. Peffault de Latour R, Peters C, Gib­son B, et al; Pediatric Working Party of
the Euro­pean Group for Blood and Marrow Transplantation; Severe Aplas­
tic Anemia Working Party of the Euro­pean Group for Blood and Marrow
Transplantation. Recommendations on hema­to­poi­etic stem cell trans­plan­
ta­tion for inherited bone mar­row fail­ure syn­dromes. Bone Marrow Trans-
plant. 2015;50(9):1168-1172.
15. Vlachos A, Federman N, Reyes-Haley C, Abramson J, Lipton JM. Hemato­
poietic stem cell trans­plan­ta­tion for Diamond Blackfan ane­mia: a report
Prakash Singh Shekhawat
from the Diamond Blackfan Anemia Registry. Bone Marrow Transplant.
2001;27(4):381-386.
16. Taylor AM, Macari ER, Chan IT, et al. Calmodulin inhib­i­tors improve eryth­ro­
poi­e­sis in Diamond-Blackfan ane­mia. Sci Transl Med. 2020;12(566):eabb5831.
17. Alter BP, Giri N, Savage SA, Rosenberg PS. Cancer in the National Cancer
Institute inherited bone mar­row fail­ure syn­drome cohort after fif­teen years
of fol­low-up. Haematologica. 2018;103(1):30-39.
18. Vlachos A, Atsidaftos E, Lababidi ML, et al. L-leu­cine improves ane­mia and
growth in patients with trans­fu­sion-depen­dent Diamond-Blackfan ane­mia:
results from a mul­ti­cen­ter pilot phase I/II study from the Diamond-Blackfan
Anemia Registry. Pediatr Blood Cancer. 2020;67(12):e28748.
19. Jaako P, Debnath S, Olsson K, et al. Gene ther­apy cures the ane­mia and
lethal bone mar­row fail­ure in a mouse model of RPS19-defi­cient Diamond-​
Blackfan ane­mia. Haematologica. 2014;99(12):1792-1798.
20. Liu Y, Dahl M, Debnath S, et al. Successful gene ther­apy of Diamond-Black­
fan ane­mia in a mouse model and human CD34+ cord blood hema­to­poi­etic
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/353/1851415/353costa.pdf by guest on 13 December 2021
stem cells using a clin­i­cally appli­ca­ble lentiviral vec­tor [published online 14
Jan­u­ary 2021]. Haematologica.
21. Siva K, Ek F, Chen J, et al. A phe­no­typic screen­ing assay identifies mod­u­la­
tors of Diamond Blackfan ane­mia. SLAS Discov. 2019;24(3):304-313.
22. Wilkes MC, Siva K, Varetti G, et  al. Metformin-induced sup­ pres­
sion of
Nemo-like kinase improves eryth­ro­poi­e­sis in pre­clin­i­cal mod­els of Diamond-
Blackfan ane­mia through induc­tion of miR-26a. Exp Hematol. 2020;91:65-77.
23. Kristiansson A, Gram M, Flygare J, Hansson SR, Åkerström B, Storry JR. The
role of α1-microglobulin (A1M) in eryth­ro­poi­e­sis and eryth­ro­cyte homeo­
sta­sis-ther­a­peu­tic oppor­tu­ni­ties in hemo­lytic con­di­tions. Int J Mol Sci.
2020;21(19):7234.
24. Sjögren SE, Flygare J. Progress towards mech­a­nism-based treat­ment for
Diamond-Blackfan ane­mia. ScientificWorldJournal. 2012;2012:184362.
25. Wilkes MC, Siva K, Chen J, et al. Diamond Blackfan ane­mia is medi­ated by
hyper­ac­tive Nemo-like kinase. Nat Commun. 2020;11(1):3344.
26. Doulatov S, Vo LT, Macari ER, et al. Drug dis­cov­ery for Diamond-Blackfan
ane­mia using reprogrammed hema­to­poi­etic pro­gen­i­tors. Sci Transl Med.
2017;9(376):eaah5645.
27. Da Costa L, Leblanc T, Mohandas N. Diamond-Blackfan ane­mia. Blood.
2020;136(11):1262-1273.
28. Da Costa L, Narla A, Mohandas N. An update on the path­o­gen­e­sis and diag­
no­sis of Diamond-Blackfan ane­mia. F1000Res. 2018;7:F1000 Faculty Rev-
1350.
29. Da Costa L, O’Donohue M-F, van Dooijeweert B, et al. Molecular approaches
to diag­nose Diamond-Blackfan ane­mia: the EuroDBA expe­ri­ence. Eur J Med
Genet. 2018;61(11):664-673.
30. Ulirsch JC, Verboon JM, Kazerounian S, et al. The genetic land­scape of Dia­
mond-Blackfan ane­mia. Am J Hum Genet. 2018;103(6):930-947.
31. Draptchinskaia N, Gustavsson P, Andersson B, et  al. The gene encoding
ribo­somal pro­tein S19 is mutated in Diamond-Blackfan anae­mia. Nat Genet.
1999;21(2):169-175.
32. Quarello P, Garelli E, Brusco A, et al. High fre­quency of ribo­somal pro­tein
gene dele­tions in Ital­ian Diamond-Blackfan ane­mia patients detected by
mul­ti­plex liga­tion-depen­dent probe ampli­fi­ca­tion assay. Haematologica.
2012;97(12):1813-1817.
33. Arbiv OA, Cuvelier G, Klaassen RJ, et al. Molecular anal­y­sis and geno­type-
phe­no­type cor­re­la­tion of Diamond-Blackfan ane­mia. Clin Genet. 2018;​
93(2):​320-328.
34. Gianferante MD, Wlodarski MW, Atsidaftos E, et al. Genotype-phe­no­type
asso­ci­a­tion and var­i­ant char­ac­ter­iza­tion in Diamond-Blackfan ane­mia
caused by path­o­genic var­i­ants in RPL35A. Haematologica. 2021;106(5):​
1303-1310.
35. Vlachos A, Osorio DS, Atsidaftos E, et  al. Increased prev­a­lence of con­
gen­i­tal heart dis­ease in chil­dren with Diamond Blackfan ane­mia sug­gests
unrec­og­nized Diamond Blackfan ane­mia as a cause of con­gen­i­tal heart dis­
ease in the gen­eral pop­u­la­tion: a report of the Diamond Blackfan ane­mia
reg­is­try. Circ Genom Precis Med. 2018;11(5):e002044.
36. Gazda HT, Sheen MR, Vlachos A, et  al. Ribosomal pro­ tein L5 and L11
muta­tions are asso­ci­ated with cleft pal­ate and abnor­mal thumbs in
­Diamond-Blackfan ane­mia patients. Am J Hum Genet. 2008;83(6):769-780.
37. Kim AR, Ulirsch JC, Wilmes S, et al. Functional selec­tiv­ity in cyto­kine sig­nal­
ing revealed through a path­o­genic EPO muta­tion. Cell. 2017;168(6):1053-
1064.e151064e15.
38. Ludwig LS, Gazda HT, Eng JC, et  al. Altered trans­la­tion of GATA1 in Dia­
mond-Blackfan ane­mia. Nat Med. 2014;20(7):748-753.
Diamond-Blackfan ane­mia, from a clin­i­cal case  |  359
39. Sankaran VG, Ghazvinian R, Do R, et al. Exome sequenc­ing identifies GATA1
muta­tions resulting in Diamond-Blackfan ane­mia. J Clin Invest. 2012;122
­(7):2439-2443.
40. Lee PY. Vasculopathy, immu­no­de­fi­ciency, and bone mar­row fail­ure: the
intrigu­ing syn­drome caused by defi­ ciency of aden­o­sine deam­ i­
nase 2.
Front Pediatr. 2018;6:282.
41. Ulirsch JC, Verboon JM, Kazerounian S, et al. The genetic land­scape of
­Diamond-Blackfan ane­mia. Am J Hum Genet. 2019;104(2):356.
42. Lipton JM, Federman N, Khabbaze Y, et al; Diamond-Black Anemia Registry.
Osteogenic sar­coma asso­ci­ated with Diamond-Blackfan ane­mia: a report
from the Diamond-Blackfan Anemia Registry. J Pediatr Hematol Oncol.
2001;23(1):39-44.
360  |  Hematology 2021  |  ASH Education Program
43. Vlachos A, Rosenberg PS, Atsidaftos E, et al. Increased risk of colon can­
cer and oste­o­genic sar­coma in Diamond-Blackfan ane­mia. Blood. 2018;132
­(20):2205-2208.
44. Lipton JM, Molmenti CLS, Hussain M, et al. Colorectal can­cer screen­ing and
sur­veil­lance strat­egy for patients with Diamond Blackfan ane­mia: pre­lim­i­
nary rec­om­men­da­tions from the Diamond Blackfan Anemia Registry. Pedi-
atr Blood Cancer. 2021;68(8):e28984.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000314
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/353/1851415/353costa.pdf by guest on 13 December 2021
Contents

ix Editors’ Message
x Reviewers
xi Continuing Medical Education Information

ALL: New Directions for Adult Patients

1 Relapsed ALL: CAR T vs transplant vs novel therapies
NOELLE V. FREY

7 Acute lymphoblastic leukemia in older adults: curtain call for conventional chemotherapy?
MARLISE R. LUSKIN

AML: So Many Options, So Little Time

16 What to use to treat AML: the role of emerging therapies
FELICITAS THOL

24 Whom should we treat with novel agents? Specifc indications for specifc and challenging
populations
LINDSAY WILDE AND MARGARET KASNER

Challenges in Multiple Myeloma Treatment

30 High-risk multiple myeloma: how to treat at diagnosis and relapse?
MARÍA-VICTORIA MATEOS, BORJA PUERTAS MARTÍNEZ, AND VERÓNICA GONZÁLEZ-CALLE

37 Minimal residual disease in multiple myeloma—why, when, where

ANDREW J. YEE AND NOOPUR RAJE

46 Treatment of older adult or frail patients with multiple myeloma

SHAKIRA J. GRANT, CIARA L. FREEMAN, AND ASHLEY E. ROSKO

CLL: Extending Survival

55 Upfront therapy: the case for continuous treatment
CONSTANTINE S. TAM

59 Frontline treatment in CLL: the case for time-limited treatment

VINCENT LÉVY, ALAIN DELMER, AND FLORENCE CYMBALISTA

68 Is there a role for anti-CD20 antibodies in CLL?

HARSH R. SHAH AND DEBORAH M. STEPHENS

Clotting and Bleeding Conundrums

76 Unexplained arterial thrombosis: approach to diagnosis and treatment
JORI E. MAY AND STEPHAN MOLL

85 Cryptogenic oozers and bruisers

KRISTI J. SMOCK AND KAREN A. MOSER

Prakash Singh Shekhawat

Contents | iii
 92 Clots in unusual places: lots of stress, lim­ited data, crit­i­cal deci­sions
CAROL MATHEW AND MARC ZUMBERG

100 EVIDENCE-BASED MINIREVIEW

Should war­fa­rin or a direct oral anti­co­ag­u­lant be used in patients presenting with throm­bo­sis in the
splanch­nic or cere­bral veins?
CAROL MATHEW AND MARC ZUMBERG

CML: Success Breeds More Success

106 TKI dis­con­tin­u­at­ ion in CML: how do we make more patients eli­gi­ble? How do we increase the
chances of a suc­cess­ful treat­ment-free remis­sion?
ANDREAS HOCHHAUS AND THOMAS ERNST

113 Lifelong TKI ther­apy: how to man­age car­dio­vas­cu­lar and other risks
MICHAEL J. MAURO

122 Blast and accel­er­ated phase CML: room for improve­ment

JOAN HOW, VINAYAK VENKATARAMAN, AND GABRIELA SORIANO HOBBS

Coagulation Laboratory Potpourri

129 Direct oral anti­co­ag­u­lant (DOAC) inter­fer­ence in hemo­sta­sis assays
KAREN A. MOSER AND KRISTI J. SMOCK

What Do Aplastic Anemia, PNH, and “Hypoplastic” Leukemia Have in Common?

134 The clin­i­cal and lab­o­ra­tory eval­u­a­tion of patients with suspected hypocellular mar­row fail­ure
SIOBÁN KEEL AND AMY GEDDIS

143 When does a PNH clone have clin­i­cal sig­nif­i­cance?

DARIA V. BABUSHOK

153 When to worry about inherited bone mar­row fail­ure and mye­loid malig­nancy pre­dis­po­si­tion
syn­dromes in the set­ting of a hypocellular mar­row
ANUPAMA NARLA

Defeating Diffuse, Double-Hit, and Dogged Non-Hodgkin Lymphoma

157 Double-hit lym­phoma: opti­miz­ing ther­apy
KIERON DUNLEAVY

164 Relapsed dis­ease: off-the-shelf immunotherapies vs cus­tom­ized engineered prod­ucts

REEM KARMALI

Emerging Therapies and Considerations for Sickle Cell Disease

174 Gene ther­apy for sickle cell dis­ease: where we are now?
JULIE KANTER AND COREY FALCON

181 Hematopoietic cell trans­plan­ta­tion for sickle cell dis­ease: updates and future direc­tions
LAKSHMANAN KRISHNAMURTI

190 EVIDENCE-BASED MINIREVIEW

In young chil­dren with severe sickle cell dis­ease, do the ben­e­fits of HLA-iden­ti­cal sib­ling donor HCT
out­weigh the risks?
NIKETA SHAH AND LAKSHMANAN KRISHNAMURTI

196 Clinical trial con­sid­er­ations in sickle cell dis­ease: patient-reported out­comes, data ele­ments,
and the stake­holder engage­ment frame­work
SHERIF M. BADAWY

iv  |  Hematology 2021  |  ASH Education Program

Hemophilia Update: Our Cup Runneth Over
206 How do we opti­mally uti­lize fac­tor con­cen­trates in per­sons with hemo­philia?
MING Y. LIM

215 EVIDENCE-BASED MINIREVIEW

For over­weight or obese per­sons with hemo­philia A, should fac­tor VIII dos­ing be based on ideal or
actual body weight?
NICOLETTA MACHIN AND MING Y. LIM

219 Factor-mimetic and rebalancing ther­a­pies in hemo­philia A and B: the end of fac­tor con­cen­trates?
PATRICK ELLSWORTH AND ALICE MA

226 Hemophilia gene ther­apy: ush­er­ing in a new treat­ment par­a­digm?

LINDSEY A. GEORGE

Hodgkin Lymphoma: Cure and Optimal Survivorship

234 Controversies in the man­age­ment of early-stage Hodgkin lym­phoma
KRISTIE A. BLUM

240 How to choose first sal­vage ther­apy in Hodgkin lym­phoma: tra­di­tional che­mo­ther­apy vs novel agents
JULIA DRIESSEN, SANNE H. TONINO, ALISON J. MOSKOWITZ, AND MARIE JOSÉ KERSTEN

247 Double-refrac­tory Hodgkin lym­phoma: tack­ling relapse after brentuximab vedotin and check­point
inhib­i­tors
NARENDRANATH EPPERLA AND MEHDI HAMADANI

How Can We Ensure That Everyone Who Needs a Transplant Can Get One?
254 Allogeneic hema­to­poi­etic cell trans­plan­ta­tion for older patients
RICHARD J. LIN AND ANDREW S. ARTZ

264 Increasing access to allo­ge­neic hema­to­poi­etic cell trans­plant: an inter­na­tional per­spec­tive

VANDERSON ROCHA, GIANCARLO FATOBENE, AND DIETGER NIEDERWIESER, FOR THE BRAZILIAN SOCIETY OF BONE
MARROW TRANSPLANTATION AND THE WORLDWIDE NETWORK FOR BLOOD AND MARROW TRANSPLANTATION

275 Increasing access to allotransplants in the United States: the impact of race, geog­ra­phy, and
socio­eco­nom­ics
SANGHEE HONG AND NAVNEET S. MAJHAIL

Immunology 101: What the Practicing Hematologist Needs to Know

281 Immunology 101: fun­da­men­tal immu­nol­ogy for the prac­tic­ing hema­tol­o­gist
SHANNON A. CARTY

287 How immu­no­de­fi­ciency can lead to malig­nancy

SUNG-YUN PAI, KATHRYN LURAIN, AND ROBERT YARCHOAN

296 Modified T cells as ther­a­peu­tic agents

NATHAN SINGH

Indolent Lymphomas
303 What fac­tors guide treat­ment selec­tion in myco­sis fungoides and Sezary syn­drome?
YOUN H. KIM

313 Follicular lym­phoma: is there an opti­mal way to define risk?

CARLA CASULO

320 Does MRD have a role in the man­age­ment of iNHL?

ILARIA DEL GIUDICE, IRENE DELLA STARZA, AND ROBIN FOÀ

Inherited Red Cell Disorders Beyond Hemoglobinopathies

331 Diagnosis and clin­i­cal man­age­ment of red cell mem­brane dis­or­ders
THEODOSIA A. KALFA

Prakash Singh Shekhawat

Contents | v
 341 Diagnosis and clin­i­cal man­age­ment of enzymopathies
LUCIO LUZZATTO

353 Diamond-Blackfan ane­mia

LYDIE M. DA COSTA, ISABELLE MARIE, AND THIERRY M. LEBLANC

It Takes a Village: Maximizing Supportive Care and Minimizing Toxicity During Childhood Leukemia
Therapy
361 Fungal diag­nos­tic test­ing and ther­apy: nav­i­gat­ing the neutropenic period in chil­dren with high-
risk leu­ke­mia
BRIAN T. FISHER

368 Minimizing car­diac tox­ic­ity in chil­dren with acute mye­loid leu­ke­mia

HARI K. NARAYAN, KELLY D. GETZ, AND KASEY J. LEGER

376 Managing ther­apy-asso­ci­ated neu­ro­tox­ic­ity in chil­dren with ALL

DEEPA BHOJWANI, RAVI BANSAL, AND ALAN S. WAYNE

Let the CHIP(s) Fall Where They May? Burdens and Benefits of Diagnosing Clonal Hematopoiesis
384 CHIP: is clonal hema­to­poi­e­sis a sur­ro­gate for aging and other dis­ease?
LUKASZ P. GONDEK

390 Mechanisms of somatic trans­for­ma­tion in inherited bone mar­row fail­ure syn­dromes

HARUNA BATZORIG CHOIJILSUREN, YEJI PARK, AND MOONJUNG JUNG

399 When are idi­o­pathic and clonal cytopenias of unknown sig­nif­i­cance (ICUS or CCUS)?
AFAF E. W. G. OSMAN

Management Strategies for Sickle Cell Disease

405 Optimizing man­age­ment of sickle cell dis­ease in patients under­go­ing sur­gery
CHARITY I. OYEDEJI AND IAN J. WELSBY

411 Treatment dilem­mas: strat­eg ­ ies for priapism, chronic leg ulcer dis­ease, and pul­mo­nary
hyper­ten­sion in sickle cell dis­ease
ROBERTA C.G. AZBELL AND PAYAL CHANDARANA DESAI

MDS: Beyond a One-Size-Fits-All Approach

418 Have we reached a molec­ul­ar era in myelodysplastic syn­dromes?
MARIA TERESA VOSO AND CARMELO GURNARI

428 Lower risk but high risk

AMY E. DeZERN

435 EVIDENCE-BASED MINIREVIEW

Molecular pre­ci­sion and clin­i­cal uncer­tainty: should molec­u­lar pro­fil­ing be rou­tinely used to guide
risk strat­i­fi­ca­tion in MDS?
DAN­IEL R. RICHARDSON AND AMY E. DeZERN

439 Oral hypomethylating agents: beyond con­ve­nience in MDS

ELIZABETH A. GRIFFITHS

MPN: New Directions

710 MPN and throm­bo­sis was hard enough . . . ​now there’s COVID-19 throm­bo­sis too
ANNA FALANGA

448 EVIDENCE-BASED MINIREVIEW

Are DOACs an alter­na­tive to vita­min K antag­on
­ ists for treat­ment of venous throm­bo­em­bo­lism in
patients with MPN?
FRANCESCA SCHIEPPATI AND ANNA FALANGA

453 Novel ther­a­pies vs hema­to­poi­etic cell trans­plan­ta­tion in mye­lo­fi­bro­sis: who, when, how?
JAMES ENGLAND AND VIKAS GUPTA

463 Running inter­feron inter­fer­ence in treating PV/ET: meet­ing unmet needs

ELIZABETH A. TRAXLER AND ELIZABETH O. HEXNER

vi  |  Hematology 2021  |  ASH Education Program

Multidisciplinary Hematologic Disorders: What Is the Hematologist’s Role?
469 Hereditary hem­or­rhagic tel­an­gi­ec­ta­sia (HHT): a prac­ti­cal guide to man­age­ment
ADRIENNE M. HAMMILL, KATIE WUSIK, AND RAJ S. KASTHURI

478 Chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion: anticoagulation and beyond

KARLYN A. MARTIN AND MICHAEL J. CUTTICA

485 Rebalanced hemo­sta­sis in liver dis­ease: a mis­un­der­stood coagulopathy

LARA N. ROBERTS

Neutropenia: Doing More with Less

492 Diagnosis and ther­a­peu­tic deci­sion-mak­ing for the neutropenic patient
JAMES A. CONNELLY AND KELLY WALKOVICH

504 Understanding neutropenia sec­ond­ary to intrin­sic or iat­ro­genic immune dysregulation

KELLY WALKOVICH AND JAMES A. CONNELLY

514 Impaired myelopoiesis in con­gen­i­tal neutropenia: insights into clonal and malig­nant hema­to­poi­e­sis
JULIA T. WARREN AND DAN­IEL C. LINK

Perioperative Consultative Hematology: Can You Clear My Patient for Surgery?

521 Perioperative con­sul­ta­tive hema­tol­ogy: can you clear my patient for a pro­ce­dure?
ALLISON ELAINE BURNETT, BISHOY RAGHEB, AND SCOTT KAATZ

529 How to man­age bleed­ing dis­or­ders in aging patients need­ing sur­gery

MOUHAMED YAZAN ABOU-ISMAIL AND NATHAN T. CONNELL

536 Heparin-induced throm­bo­cy­to­pe­nia and car­dio­vas­cu­lar sur­gery

ALLYSON M. PISHKO AND ADAM CUKER

Pregnancy in Special Populations: Challenges and Solutions

545 How to eval­u­ate and treat the spec­trum of TMA syn­dromes in preg­nancy
MARIE SCULLY

552 Pregnancy in spe­cial pop­u­la­tions: chal­lenges and solu­tions prac­ti­cal aspects of man­ag­ing von
Willebrand dis­ease in preg­nancy
OZLEM TURAN AND REZAN ABDUL KADIR

559 Prevention and man­age­ment of venous throm­bo­em­bo­lism in preg­nancy: cut­ting through the
prac­tice var­i­a­tion
LESLIE SKEITH

Survivorship After Allogeneic Hematopoietic Cell Transplant

570 Psychosocial and finan­cial issues after hema­to­poi­etic cell trans­plan­ta­tion
DAVID BUCHBINDER AND NANDITA KHERA

578 Noninfectious com­pli­ca­tions of hema­to­poi­etic cell trans­plan­ta­tion

KIRSTEN M. WILLIAMS

587 Infectious com­pli­ca­tions and vac­cines

PER LJUNGMAN

The Changing Landscape of α- and β-Thalassemia Diagnosis and Treatment

592 Advances in the man­age­ment of α-thal­as­se­mia major: rea­sons to be opti­mis­tic
PAULINA HORVEI, TIPPI MACKENZIE, AND SANDHYA KHARBANDA

600 β-Thalassemia: evolv­ing treat­ment options beyond trans­fu­sion and iron che­la­tion
ARIELLE L. LANGER AND ERICA B. ESRICK

607 Optimal strat­e­gies for car­rier screen­ing and pre­na­tal diag­no­sis of α- and β-thal­as­se­mia
CHERYL MENSAH AND SUJIT SHETH
Prakash Singh Shekhawat
Contents | vii
 The COVID Crash: Lessons Learned from a World on Pause
614 How to rec­og­nize and man­age COVID-19-asso­ci­ated coagulopathy
GLO­RIA F. GERBER AND SHRUTI CHATURVEDI

621 COVID-19 and throm­bo­sis: searching for evi­dence

BRIGHT THILAGAR, MOHAMMAD BEIDOUN, RUBEN RHOADES, AND SCOTT KAATZ

628 Passive immune ther­a­pies: another tool against COVID-19

LISE J. ESTCOURT

Update in Graft-versus-Host Disease

642 Acute GVHD: think before you treat
LAURA F. NEWELL AND SHERNAN G. HOLTAN

648 Updates in chronic graft-ver­sus-host dis­ease

BETTY K. HAMILTON

655 What else do I need to worry about when treating graft-ver­sus-host dis­ease?
AREEJ EL-JAWAHRI

What’s New in Plasma Cell Disorders?

662 Advances in MGUS diag­no­sis, risk strat­i­fi­ca­tion, and man­age­ment: intro­duc­ing myeloma-defining
genomic events
OLA LANDGREN

673 Smoldering mul­ti­ple mye­loma: evolv­ing diag­nos­tic cri­te­ria and treat­ment strat­e­gies
ALISSA VISRAM, JOSELLE COOK, AND RAHMA WARSAME

682 AL amy­loid­osis: untangling new ther­a­pies

SUSAN BAL AND HEATHER LANDAU

What’s New in the Transfusion Management of Sickle Cell Disease?

689 How to avoid the prob­lem of eryth­ro­cyte alloimmunization in sickle cell dis­ease
FRANCE PIRENNE, A­LINE FLOCH, AND ANOOSHA HABIBI

696 Indications for transfusion in the management of sickle cell disease

HYOJEONG HAN, LISA HENSCH, AND VENÉE N. TUBMAN

704 Management of hemo­lytic trans­fu­sion reac­tions

JEANNE E. HENDRICKSON AND ROSS M. FASANO

viii  |  Hematology 2021  |  ASH Education Program

IT TAKES A VILLAGE: MAXIMIZING SUPPORTIVE CARE AND MINIMIZING TOXICITY DURING CHILDHOOD LEUKEMIA THERAPY

Fungal diagnostic testing and therapy:

navigating the neutropenic period
in children with high-risk leukemia

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/361/1851411/361fisher.pdf by guest on 13 December 2021

Brian T. Fisher
Department of Pediatrics, Children’s Hospital of Philadelphia, Philadelphia, PA; and Department of Biostatistics, Epidemiology and Informatics,
Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA

Children, adolescents, and young adults receiving intensive chemotherapy for acute myeloid leukemia or high-risk or
relapsed acute lymphoblastic leukemia sustain prolonged periods of neutropenia that predispose them to invasive fun-
gal disease (IFD). For many decades the standard of care for these patients was to initiate empirical antifungal therapy
after a period of prolonged fever and neutropenia. Recent publications have yielded important evidence on the utility of
different diagnostic and therapeutic approaches aimed at reducing the impact of IFD among these patients during these
vulnerable periods. This case-based review highlights and interprets the published data to provide context for the IFD
diagnostic and therapeutic recommendations proposed in multiple published guidelines. Personalized approaches are
offered at points where evidence is lacking. Time points where specific knowledge gaps exist are identified along the
clinical trajectory of the prolonged neutropenic period to illustrate areas for future investigation.

LEARNING OBJECTIVES
• Learn the evidence supporting prophylaxis, preemptive, and empirical antifungal therapy during periods of pro-
longed neutropenia secondary to intensive chemotherapy for leukemia
• Understand the utility and limitations of existing fungal biomarkers at specifc clinical time points during pro-
longed periods of neutropenia

CLINICAL CASE
An 11-year-old girl with recently diagnosed acute myeloid
leukemia (AML) is admitted to the hospital for induction
2 chemotherapy. She receives cyarabine, daunorubicin,
etoposide, and gemtuzumab ozogamicin in induction 1 and
is scheduled to receive cytarabine, daunorubicin, and etopo-
side for induction 2. What prophylactic antifungal regimen
and fungal surveillance testing should be employed once she
develops neutropenia? Approximately 2 weeks into her neu-
tropenic period (absolute neutrophil count <200 cells/µL),
she develops fevers without any localizing signs or symp-
toms. Blood cultures are drawn, and cefepime is initiated for
empirical antibiotic therapy. The blood cultures are negative
and she remains stable, but fevers persist for more than 96
hours. What adjustments should be made to her antifungal
regimen, if any, and what additional diagnostic testing, if any,
should be performed?
Prakash Singh Shekhawat
Introduction
Invasive fungal diseases (IFDs) have long been identifed as
important opportunistic infections in children, adolescents,
and young adults receiving chemotherapy for AML and for
high-risk or relapsed acute lymphoblastic leukemia (ALL).
The epidemiology of IFDs in these patient populations is
displayed in Table 1. The interpretation of IFD incidence
across each study needs to consider the IFD defnition used,
whether antifungal prophylaxis was administered, and the
study design. Early reports from pediatric AML chemother-
apy clinical trials estimated an IFD incidence of 14% to 23%
per cycle of chemotherapy.1 More recent investigations
have found a lower but still substantial incidence of IFD,
ranging from 3% to 7%.2,3 The decline in IFD rates among
patients with AML is likely multifactorial, including the use of
published research criteria for defning probable or proven
IFDs that are more restrictive,3 the decreased use of ste-
roids in this population, and the optimization of supportive
care measures. The rates of IFD in ALL are less well defned,
Fungal management in leukemia | 361
Table 1. Epidemiology of IFDs from select pedi­at­ric leu­ke­mia cohorts

Author, year Study design
AML
Sung et al1 Summary of adverse
event data from
che­mo­ther­apy RCT
Fisher et al2 RCT of 2 anti­fun­gal
pro­phy­laxis reg­i­mens
ALL
Rosen et al4 Retrospective sin­gle-
cen­ter obser­va­tional
cohort
Patient type Antifungal pro­phy­laxis IFD def­i­ni­tion IFD inci­dence
Newly diag­nosed AML Not rou­tinely used Not spec­i­fied 18% to 23% inci­dence per
che­mo­ther­apy course
De novo, sec­ond­ary, Fluconazole or 2008 EORTC/MSG cri­te­ria33 3.1% in caspofungin
or relapsed AML caspofungin sub­jects*;
7.2% in fluconazole sub­jects*
Any type of newly Not rou­tinely used Not spec­i­fied 10%†
diag­nosed ALL

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/361/1851411/361fisher.pdf by guest on 13 December 2021

Wang et al7 Retrospective mul­ti­cen­ter Reported by ALL type Varied by health care 2008 EORTC/MSG cri­te­ria33 6.2% over­all†;
obser­va­tional cohort and risk strata cen­ter 5.8% SR ALL†;
8.7% HR ALL†;
3.8% nonrelapsed ALL†;
12.9% relapsed refrac­tory ALL†
Castagnola Retrospective sin­gle- Any type of newly Not rou­tinely used 2008 EORTC/MSG cri­te­ria33 5.4% over­all†;
et al6 cen­ter obser­va­tional diag­nosed ALL 15.2% HR ALL†
cohort
*5-month cumu­la­tive inci­dence.
†Cumulative inci­dence for dura­tion of che­mo­ther­apy.
CTCAE, com­mon ter­mi­nol­ogy cri­te­ria for adverse events; EORTC/MSG, Euro­pean Organization for Research and Treatment of Cancer/Mycoses Study
Group; IFD, invasive fungal disease; RCT, ran­dom­ized con­trolled trial.

but avail­­able data sug­gest the sub­stan­tial pres­ence of IFD. Initial
obser­va­tional stud­ies across all­types of pedi­at­ric ALL reported
an IFD rate of 10%, with IFD rates nearing 20% in relapsed ALL
patients.4,5 However, these stud­ies did not employ the published
proven or prob­a­ble IFD cri­te­ria and thus may have overestimated
the true rate of IFD. Contemporary stud­ies sug­gest that proven
or prob­a­ble IFD rates in stan­dard-risk ALL patients have declined
to 5.8% but remain high at 12.9% for relapsed ALL patients and at
8.7% to 15.2% for high-risk ALL patients.6,7
Although rates of IFD have var­ied by time period and by leu­
ke­mia type, their per­sis­tence as oppor­tu­nis­tic infec­tions in these
patient pop­u­la­tions remains a con­cern. IFDs con­tinue to be a
feared com­pli­ca­tion of leu­ke­mia treat­ment because of the poten­
tial for sig­nif­i­cant mor­bid­ity and mor­tal­ity asso­ci­ated with these
path­ o­gens. Invasive can­ di­di­
as­is is the most com­ mon form of
proven or prob­a­ble IFD and has been asso­ci­ated with a 10% attrib­
ut­­able mor­tal­ity in chil­dren.8 Proven or prob­ab ­ le inva­sive mold
dis­eases, such as asper­gil­lo­sis and mucormycosis, are less com­
mon but have esti­mated case fatal­ity rates in excess of 30%.9,10
Multiple fac­tors pre­dis­pose chil­dren with leu­ke­mia to IFD. They
include neutropenia, expo­sure to high-dose ste­roids, hyper­gly­
ce­mia, and prolonged broad-spec­trum anti­bi­ot­ics.11 Prolonged
neutropenia is by far the most documented of these fac­tors.
The dura­tion of neutropenia at which the risk for IFD increases
sig­nif­i­cantly is not clearly defined and likely varies by patient.
However, in gen­eral the risk for IFD is esti­mated to increase sub­
stan­tially in patients with neutropenia last­ing for more than 10
days. The dura­tion of neutropenia will vary depending on the
che­mo­ther­apy reg­i­men admin­is­tered, but most con­tem­po­rary
che­mo­ther­apy reg­i­mens for AML and high-risk or relapsed ALL
will result in mul­ti­ple epi­sodes of sustained abso­lute neu­tro­phil
counts below 200 cells/µL that last for 2 to 3 weeks.11,12 While
neutropenia is the dom­i­nant risk fac­tor for IFD, the impor­tance
of recent ste­roid expo­sure in esca­lat­ing the risk of IFD in leu­ke­
mia patients should not be overlooked.13,14 The declin­ing reli­ance
on ste­roids in con­tem­po­rary AML che­mo­ther­apy reg­i­mens and
increas­ing reli­ance on dexa­meth­a­sone in cer­tain ALL reg­i­mens
might explain some of the reported decline in IFD inci­dence in
the for­mer and increase in IFD inci­dence in the lat­ter.15,16
Results from a small ran­ dom­ized trial performed by Pizzo
et al in the early 1980s supported the util­ity of empir­i­cal anti­
fun­gal ther­apy in patients with per­sis­tent fever and neutrope-
nia despite broad-spec­trum empir­i­cal anti­bi­otic ther­apy.17 This
approach quickly became the stan­dard of care and has remained
as such for the bet­ter part of 4 decades. In recent years there has
been increas­ing focus on advanc­ing best prac­tices for reduc­ing
IFD mor­bid­ity and mor­tal­ity in pedi­at­ric AML and ALL patients.
Several impor­tant inves­ti­ga­tions and inter­na­tional guide­lines
have been published that pro­ vide guid­ ance on the util­ ity of
pro­phy­lac­tic and pre­emp­tive anti­fun­gal ther­apy approaches as
well as fun­gal diag­nos­tic tools for these patient pop­u­la­tions. In
this review we use the pro­vided clin­i­cal case to illus­trate how
published evi­ dence can sup­ port impor­ tant IFD man­ age­ment
deci­sions at var­i­ous time points dur­ing a period of prolonged
neutropenia after leu­ke­mia che­mo­ther­apy. Knowledge gaps in
need of future inves­ti­ga­tion are noted.
Should anti­fun­gal pro­phy­laxis be admin­is­tered dur­ing
prolonged neutropenia peri­ods?
In the first decade of the cur­rent cen­tury, the approach to anti­
fun­gal pro­phy­laxis for neutropenia after receipt of inten­sive che­
mo­ther­apy for leu­ke­mia var­ied sig­nif­i­cantly. An inter­na­tional
sur­vey performed by Sung et al found that 77% and 91% of Chil-
dren’s Oncology Group Centers and Berlin-Frankfurt-Muen­ster
Group Centers, respec­ tively, were using 5 dif­ fer­ ent anti­
fun­
gal
agents to admin­is­ter anti­fun­gal pro­phy­laxis to chil­dren with AML.18

362  |  Hematology 2021  |  ASH Education Program

This var­i­a­tion in prac­tice was not par­tic­u­larly sur­pris­ing because
of the lim­ited data from pedi­at­ric obser­va­tional or ran­dom­ized
stud­ies com­par­ing the effec­tive­ness of dif­fer­ent anti­fun­gal pro­
phy­laxis approaches. Adult neutropenia man­age­ment guide­lines
did exist at that time and recommended anti­fun­gal pro­phy­laxis
dur­ing peri­ods of neutropenia resulting from inten­sive che­mo­
ther­apy. This included a spe­cific rec­om­men­da­tion of posacon-
azole pro­ phy­laxis for patients older than 13 years receiv­ ing
che­mo­ther­apy for AML, which was based on ran­dom­ized con­
trolled trial data reveal­ing the effi­cacy of posaconazole com­
pared to fluconazole or itraconazole in an adult AML pop­u­la­tion.19
Fortunately, larger pedi­at­ric com­par­a­tive effec­tive­ness stud­ies
soon followed. A ret­ro­spec­tive obser­va­tional cohort study lev­er­
aged inpa­tient hos­pi­tal admin­is­tra­tive data and var­i­a­tion in anti­
fun­gal pro­phy­laxis uti­li­za­tion to reveal that anti­fun­gal pro­phy­laxis
in chil­dren, ado­les­cents, and young adults with AML did reduce
mor­tal­ity and resource uti­li­za­tion com­pared to no pro­phy­laxis.20
This study con­firmed the pre­sumed ben­e­fit of anti­fun­gal pro­
phy­laxis in pedi­at­ric patients but could not deter­mine whether
the pro­phy­lac­tic agent needed to include antimold activ­ity. A
sub­se­quent ran­dom­ized con­trolled trial performed within the
Children’s Oncology Group (ACCL0933) helped to answer this
ques­tion by com­par­ing the effi­cacy of caspofungin pro­phy­laxis
with fluconazole pro­phy­laxis.2 In this trial caspofungin pro­phy­
laxis sig­nif­i­cantly lowered the risk of proven or prob­a­ble IFD,
spe­cif­i­cally reduc­ing the fre­quency of inva­sive asper­gil­lo­sis.
As these and other pedi­at­ric stud­ies began to pop­u­late the lit­
er­a­ture, an inter­na­tional sup­port­ive care com­mit­tee endeav­ored
to col­late and inter­pret the data in a pedi­at­ric-spe­cific, sys­temic
anti­fun­gal pro­phy­laxis guide­line.21 This guide­line included a sys­
tem­atic review of stud­ies com­par­ing dif­fer­ent anti­fun­gal pro­phy­
laxis reg­i­mens. These data informed a strong rec­om­men­da­tion
for antimold pro­phy­laxis in pedi­at­ric AML patients and a weak
rec­om­men­da­tion for antimold pro­phy­laxis after inten­sive che­
mo­ther­apy for high-risk or relapsed ALL. The lat­ter was a weak
rec­om­men­da­tion because of lim­ited IFD epi­de­mi­­ol­ogy and com­
par­a­tive effec­tive­ness data spe­cific to this pop­u­la­tion. This weak
rec­om­men­da­tion cou­pled with chal­lenges in admin­is­ter­ing anti­
fun­gal pro­phy­laxis to chil­dren with ALL (eg, greater poten­tial for
drug-drug inter­ac­tions and man­age­ment mostly in the out­pa­tient
set­ting) has likely resulted in less rou­tine use of anti­fun­gal pro­phy­
laxis in patients with ALL com­pared to those with AML. This could,
in part, explain the pre­vi­ously cited increas­ing rates of IFD in ALL
patients. Randomized con­trolled tri­als assessing the ideal pro­phy­
laxis approach in ALL-spe­cific patient pop­u­la­tions are warranted.
Notably, the guide­line remained flex­i­ble on which antimold
agent to use for pro­phy­laxis in either AML or ALL patients, sug-
gesting an echinocandin or an azole with antimold activ­ity. The
choice of anti­fun­gal agent depends on fac­tors such as the age
of the child, inpa­tient vs out­pa­tient set­ting of neutropenia man­
age­ment, and con­cern for drug-drug inter­ac­tions. The guide­line
did rec­om­mend against using a sys­temic amphotericin for­mu­la­
tion for pro­phy­laxis because of a documented increased risk of
adverse effects.
Before adapting guide­line rec­om­men­da­tions to their patients,
cli­ni­cians need to assess the epi­de­mi­­ol­ogy of IFD across their
local leu­ke­mia pop­u­la­tion. A cen­ter could have dif­fer­ing rates
of proven or prob­a­ble IFD and/or a dif­fer­ent fun­gal path­o­gen
dis­tri­bu­tion that under­mines the gen­er­al­iz­abil­ity of results from
published stud­ ies to their patient pop­ u­
la­
tion. In this set­ ting,
Prakash Singh Shekhawat
a mul­ti­dis­ci­plin­ary qual­ity improve­ment ini­tia­tive inclu­sive of
oncol­ogy, infec­tious dis­eases, and anti­mi­cro­bial stew­ard­ship
experts could be help­ful to guide deci­sions on local fun­gal pre­
ven­tion efforts.
My cur­ rent per­ sonal pref­
er­ ence is to use an echinocandin
as the anti­fun­gal pro­phy­laxis agent. This is because echinocan-
dins have a rel­a­tively lim­ited side effect pro­file with less con­cern
about alter­ing the phar­ma­co­ki­net­ics of other admin­is­tered med­
i­ca­tions. Additionally, our patients with antic­i­pated prolonged
neutropenia after inten­sive che­mo­ther­apy are often mon­i­tored
in the hos­pi­tal for their peri­ods of neutropenia, so a once-a-day
intra­ve­nous med­i­ca­tion is fea­si­ble. Finally, chal­lenges in achiev­ing
the appro­pri­ate antimold azole dose for each child is a lim­it­ing
com­po­nent of this class of anti­fun­gal agents. However, as out­pa­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/361/1851411/361fisher.pdf by guest on 13 December 2021
tient obser­va­tion of chil­dren dur­ing peri­ods of prolonged neutro-
penia is increas­ingly uti­lized, the ben­e­fits of enteral pro­phy­laxis
will increase, which may favor antimold azoles. Knowledge gaps
still exist that if filled would improve the abil­ity to uti­lize azoles in
this set­ting. These gaps include ideal dos­ing for chil­dren, spe­cif­i­
cally for posaconazole, and tar­get troughs for pro­phy­laxis.
Is sur­veil­lance test­ing with fun­gal bio­mark­ers warranted
dur­ing prolonged neutropenia?
An increas­ing num­ber of com­mer­cially avail­­able non-cul­ture-
based fun­gal bio­mark­ers offer the poten­tial for sur­veil­lance test­
ing to detect the pres­ence of an IFD at early onset. The most
well stud­ied of these bio­mark­ers are the Aspergillus galactoman-
nan (GM) enzyme immu­no­as­say (EIA; Platelia™) and the beta-D-
glu­can (BDG) assay (Fungitell®). The GM EIA is designed to detect
GM, which is a cell wall com­po­nent of Aspergillus spe­cies, while
the BDG assay aims to detect BDGs found in the cell walls of var­
i­ous path­o­genic fungi, includ­ing Aspergillus and Candida spp.
Results from stud­ies assessing the util­ity of these 2 bio­mark­ers as
a sur­veil­lance tool for early IFD dur­ing peri­ods of prolonged neu-
tropenia in adults were received with much opti­mism. Maertens
et al assessed the GM EIA’s abil­ity to detect inva­sive asper­gil­lo­sis
dur­ing 362 prolonged neutropenic peri­ods in adults receiv­ing
che­mo­ther­apy for hema­to­logic malig­nancy or con­di­tion­ing for a
hema­to­poi­etic cell trans­plan­ta­tion (HCT).22 They found the oper­
at­ing char­ac­ter­is­tics of the GM EIA for sur­veil­lance test­ing to be
rea­son­able (sen­si­tiv­ity, spec­i­fic­ity, and pos­i­tive pre­dic­tive val­ues
(PPV) and neg­a­tive pre­dic­tive val­ues (NPV) were 72.9%, 99.1%,
93.1%, and 70.8%, respec­tively). In a sim­i­larly designed study,
Odabasi et al assessed the BDG assay as a sur­veil­lance tool for
proven or prob­a­ble IFD in adults with neutropenia after AML che­
mo­ther­apy. They found the BDG assay oper­at­ing char­ac­ter­is­tics
to also be rea­son­able (sen­si­tiv­ity, spec­i­fic­ity, PPV, and NPV were
100%, 90%, 43%, and 100%, respec­tively).23 These 2 pro­spec­
tive obser­va­tional cohorts led to US Food and Drug Adminis-
tration approval for these bio­mark­ers, and many pedi­at­ric and
adult cen­ters adopted an IFD sur­veil­lance test­ing approach (ie,
weekly or twice-weekly GM EIA and BDG assays) dur­ing peri­ods
of prolonged neutropenia but prior to the onset of any signs or
symp­toms of IFD.
The trans­port­abil­ity of these adult data to chil­dren was appro­
pri­ately questioned. First, an assess­ ment of the BDG assay in
oth­er­wise healthy chil­dren suggested higher base­line lev­els of
BDG in chil­dren and ado­les­cents, which could lead to fre­quent
false-pos­i­tive results.24 Subsequently, a series of smaller stud­
ies were published assessing the GM EIA and the BDG assay in
Fungal man­age­ment in leu­ke­mia  |  363
Table 2. Operating char­ac­ter­is­tics of the Aspergillus GM EIA and BDG assay for sur­veil­lance test­ing in pedi­at­ric patients with
can­cer or under­go­ing HCT

Study cohort Study’s IFD

Author, year size inci­dence rate Sensitivity Specificity PPV NPV
Aspergillus GM EIA
Hovi et al34 98 8.0% 93% 50% 14% 99%
Steinbach et al35 64 0% 87% 0% 0% 98%
Hayden et al36 56 30.4% 87% 65% 69% 85%
Badiee et al37 62 16.1% 92% 90% 82% 96%
Koltze et al38
34 17.6% 100% 83% 100% 97%
Fisher et al26 425 0.7% 0% 98.8% 0% 99.9%

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/361/1851411/361fisher.pdf by guest on 13 December 2021

BDG assay
Badiee et al37 62 16.1% 46% 50% 26% 71%
Koltze et al38
34 17.6% 29% 83% 20% 89%
Fisher et al26 425 1.4% 0% 94.7% 0% 99.8%
Data in this table adapted from Lehrnbecher et al, Table 3.25

pedi­at­ric patients with prolonged neutropenia after che­mo­ther­apy
or HCT con­di­tion­ing reg­i­mens. These stud­ies were sys­tem­at­i­cally
reviewed by Lehrnbecher et al,25 and the results are sum­ma­rized
in Table 2. The oper­at­ing char­ac­ter­is­tics var­ied widely for both
assays across stud­ies included in the review. The authors con­
cluded that lower rates (ie, lower pre­test prob­a­bil­ity) of proven or
prob­ab­ le IFD in the more recent cohorts increased the like­li­hood
of false-pos­i­tive results, which lim­ited the util­ity of sur­veil­lance
test­ing. A more recent large obser­va­tional cohort of pedi­at­ric AML
patients con­firmed the lim­ited util­ity of the GM EIA and the BDG
assays for sur­veil­lance test­ing (Table 2).26 Prior to these pub­li­ca­
tions, our cen­ter had been employing weekly sur­veil­lance test­ing
with the GM EIA. After reviewing the data from these pedi­at­ric
stud­ies, we have elected to stop sur­veil­lance test­ing.
Other nonculture diag­ nos­
tic plat­forms besides the GM EIA
and BDG assays may be appro­ pri­ate for sur­ veil­
lance test­ing.
These would include mold-spe­ cific poly­ mer­ase chain reac­ tion
(PCR) assays, spe­cif­i­cally for Aspergillus spp. and path­o­gens of
the Mucorales order, and plasma micro­bial cell-free DNA (cfDNA)
sequenc­ ing plat­forms. Assessments of mold-spe­ cific PCRs in
pedi­at­ric cohorts have focused on sce­nar­ios in which the patient
is displaying clin­i­cal symp­toms concerning for IFD (ie, prolonged
fever and neutropenia) but they have not been assessed under a
sur­veil­lance pro­to­col.25 A recent inves­ti­ga­tion reported the expe­
ri­ence of monthly cfDNA sequenc­ing for sur­veil­lance of IFD in 40
at-risk pedi­at­ric patients with can­cer or under­go­ing HCT.27 While
a sub­set of patients had cfDNA sequenc­ing that revealed the pres­
ence of a fun­gal path­o­gen, it is not clear that sur­veil­lance cfDNA
test­ing improved the clin­i­cal man­age­ment of these patients. Addi-
tionally, on some of the cfDNA sequenc­ing results mul­ti­ple path­o­
gens of unclear sig­nif­i­cance were iden­ti­fied. Future inves­ti­ga­tions
in larger pedi­at­ric cohorts are nec­es­sary before con­sid­er­ing the
com­mit­ment of health care resources toward sur­veil­lance test­ing
with either PCR or cfDNA sequenc­ing diag­nos­tic plat­forms.
Empirical anti­fun­gal ther­apy
As noted above, Pizzo et al published their land­mark ran­dom­
ized trial in 1982. A total of 34 chil­dren, ado­les­cent, and young
adult patients with per­sis­tent fever and neutropenia, despite
broad-spec­trum anti­bi­otic ther­apy, were ran­dom­ized to receive
or not receive sys­temic con­ven­tional ampho­ter­i­cin B ther­apy.17
Those sub­jects ran­dom­ized to anti­fun­gal ther­apy were observed
to have a faster res­ol­u­tion of fever. This approach was labeled
“empir­i­cal anti­fun­gal ther­apy for prolonged fever and neutrope-
nia.” Multiple stud­ies assessing the util­ity of empir­i­cal anti­fun­gal
ther­apy were sub­se­quently performed, and col­lec­tively, these
stud­ies con­firmed the ben­e­fits of empir­i­cal anti­fun­gal ther­apy.28
Based on these data, the 2017 update of the inter­na­tional pedi­at­
ric fever and neutropenia guide­lines recommended that patients
at high risk for IFD and with fever and neutropenia persisting
lon­ger than 96 hours receive empir­i­cal anti­fun­gal ther­apy with
either an echinocandin or lipo­so­mal ampho­ter­i­cin B.29
However, this rec­om­men­da­tion existed prior to the recently
published rec­om­men­da­tion to admin­is­ter antimold pro­phy­laxis
to patients with antic­i­pated prolonged neutropenia after inten­
sive leu­ke­mia che­mo­ther­apy. This rec­om­men­da­tion to admin­is­ter
antimold pro­phy­laxis at the start of neutropenia raises the ques­
tion of whether empir­ic ­ al anti­fun­gal ther­apy is still needed. As the
exam­ple case illus­trates, some pedi­at­ric patients will develop pro-
longed fever and neutropenia despite receiv­ing both pro­phy­lac­tic
antimold ther­apy and empir­i­cal broad-spec­trum anti­bi­ot­ics at the
onset of fever. In the ran­dom­ized trial of caspofungin vs fluconazole
pro­phy­laxis in pedi­at­ric AML, many patients devel­oped prolonged
fever and neutropenia—even those ran­dom­ized to the caspofungin
treat­ment arm.2 Unfortunately, there are lim­ ited data to guide
whether a patient like the one in our clin­i­cal case should con­tinue
the antimold pro­phy­lac­tic agent started at the begin­ning of neu-
tropenia or if the prolonged fever and neutropenia should prompt
a tran­si­tion to a dif­fer­ent antimold agent to serve as empir­i­cal anti­
fun­gal ther­apy for the remain­der of the neutropenic period. This is
an impor­tant knowl­edge gap deserv­ing fur­ther inves­ti­ga­tion.
Until data are avail­­able, cli­ni­cians will need to dis­cuss locally
how they want to approach such patients. Notably, the break­
through proven or prob­a­ble IFD rate for sub­jects receiv­ing caspo-
fungin pro­phy­laxis in the ran­dom­ized trial was 3.1%,2 which may
be too high for some cli­ni­cians not to change to a dif­fer­ent empir­
i­cal antimold ther­a­peu­tic. However, a true expan­sion in anti­fun­gal
cov­er­age from pro­phy­lac­tic caspofungin would likely neces­si­

364  |  Hematology 2021  |  ASH Education Program

 An 11-year-old female with AML
develops neutropenia aer
inducon II chemotherapy

Surveillance fungal biomarker tesng

Start an-mold prophylaxis therapy
• Do NOT perform surveillance biomarker
• Suggest an echinocandin as first-line therapy
tesng if on an-mold prophylaxis
• If unable to administer echinocandin
• Posive biomarker test result more likely to
consider an-mold azole
be a false posive

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/361/1851411/361fisher.pdf by guest on 13 December 2021

On echinocandin prophylaxis she
develops new fever. Despite empirical
anbiocs her fever and neutropenia
persist for >96 hours

What are next steps?

Comprehensive Clinical Exam

If no focal or systemic signs besides fever:
• Connue an-mold agent started as
prophylaxis and monitor closely
• Alternavely, could pursue a pre-empve
diagnosc approach (see text)
If any focal or systemic signs concerning for IFD:
• Broaden an-mold anfungal therapy
• Consult with immunocompromised infecous
diseases specialist to guide addional
diagnosc tesng

Figure 1. Conceptual model for antifungal diagnostic and therapeutic decisions during neutropenia after chemotherapy for AML.

tate a tran­si­tion to a lipid ampho­ter­i­cin B for­mu­la­tion, which can
have a sig­nif­i­cant tox­ic­ity risk. Therefore, in this sit­u­a­tion I elect
to review the his­tory and cur­rent clin­i­cal state of each patient
before decid­ing whether to broaden beyond the pro­phy­lac­tic
anti­fun­gal agent. If a patient has no symp­toms beyond fever and
neutropenia and the patient’s med­i­cal his­tory and expo­sure his­
tory do not increase my con­cern for IFD, I will rec­om­mend main-
taining ther­apy with the same antimold pro­phy­laxis agent.
Preemptive anti­fun­gal ther­apy for prolonged fever
and neutropenia
Preemptive anti­ fun­
gal ther­apy is a con­ cept gar­ner­
ing increas­
ing inter­est. Evidence from stud­ies assessing this approach may
even­tu­ally solve the clin­i­cal conun­drum of whether a patient with
prolonged fever and neutropenia should remain on anti­ fun­gal
pro­phy­laxis vs transitioning to empir­i­cal anti­fun­gal ther­apy. A pre­
emp­tive ther­apy approach relies on the uti­li­za­tion of results from
a clin­i­cal exam­i­na­tion and from diag­nos­tic stud­ies that inform the
deci­sion to alter ther­apy for a given patient. In 2009 Cordonnier
et al com­pared a pre­emp­tive anti­fun­gal approach to an empir­i­cal
one in 293 pedi­at­ric and adult patients with prolonged or recur­
rent fever and neutropenia. The pre­emp­tive approach included
clin­i­cal exam­i­na­tions, GM EIA test­ing, and radio­graphic imag­ing.
Any signs or symp­toms of IFD from those stud­ies prompted the
ini­ti­a­tion or broad­en­ing of anti­fun­gal ther­apy. Preemptive ther­apy
Prakash Singh Shekhawat
was noninferior for sur­vival and resulted in a sig­nif­i­cant reduc­tion
in anti­fun­gal expo­sure.30 However, sig­nif­i­cantly more sub­jects in
the pre­emp­tive arm sustained a prob­a­ble or proven IFD. Santolaya
et al performed a sim­i­lar ran­dom­ized trial in 149 pedi­at­ric patients
with per­sis­tent fever and neutropenia.31 Patients in the pre­emp­
tive arm had a sig­nif­i­cant reduc­tion in anti­fun­gal expo­sure and
mor­tal­ity rates sim­i­lar to those started on empir­i­cal anti­fun­gal
ther­apy. The 8th Euro­pean Conference on Infections in Leukaemia
ref­er­enced the lat­ter study in stat­ing their grade B rec­om­men­
da­tion that pre­emp­tive ther­apy may be con­sid­ered as an alter­
na­tive approach to empir­i­cal ther­apy in pedi­at­ric patients.32 The
chal­lenge with a pre­emp­tive approach in pedi­at­ric patients is
that it often relies on fun­gal bio­mark­ers to guide final deci­sions.
Pediatric-spe­cific data on the util­ity of these bio­mark­ers to either
detect or exclude IFD at the time point of prolonged fever and
neutropenia are not read­ily avail­­able. As such it is dif­fi­cult to rely
on these bio­mark­ers for this pre­emp­tive vs empir­i­cal clin­i­cal deci­
sion. Hopefully, future inves­ti­ga­tions will define the util­ity of exist-
ing and novel bio­mark­ers for this clin­i­cal sce­nario.
Summary
The diag­nos­tic and ther­a­peu­tic approaches to IFD in pedi­at­
ric patients with AML and high-risk or relapsed ALL con­tinue to
evolve. In the past decade, increas­ingly avail­­able pedi­at­ric-spe­
cific evi­dence has informed pedi­at­ric-spe­cific guide­lines. Figure 1
Fungal man­age­ment in leu­ke­mia  |  365
pro­vi­des a con­cep­tual model of my inter­pre­ta­tion and appli­ca­
tion of some of the existing data and guide­lines detailed above
using the pro­posed clin­i­cal case. Antimold ther­apy for chil­dren
with antic­i­pated prolonged neutropenia after AML or ALL che­
mo­ther­apy should be con­sid­ered the stan­dard of care. While GM
EIA or BDG assays should be dis­cour­aged for IFD sur­veil­lance,
more data are needed on using these and other novel bio­mark­
ers at other clin­i­cal time points. Finally, for patients who develop
prolonged fever and neutropenia while on antimold pro­phy­laxis,
it is not always nec­es­sary to tran­si­tion to a dif­fer­ent empir­i­cal
anti­fun­gal agent. More data are needed to deter­mine the ideal
approach to this lat­ter clin­i­cal sce­nario.
Conflict-of-inter­est dis­clo­sure
Brian T. Fisher: research funding: Pfizer, Merck; chair: Safety Mon-
itoring Board, Astellas inves­ti­ga­tion of isavuconazole.
Off-label drug use
Brian T. Fisher: nothing to disclose.
Correspondence
Brian T. Fisher, Division of Infectious Diseases, Children’s Hospital
of Philadelphia, Roberts Pediatric Research Center, 2716 South
Str, Rm 10-362, Philadelphia, PA 19146; e-mail: fisherbria@chop​
­.edu.
References
1. Sung L, Lange BJ, Gerbing RB, Alonzo TA, Feusner J. Microbiologically
documented infec­ tions and infec­ tion-related mor­tal­
ity in chil­dren with
acute mye­loid leu­ke­mia. Blood. 2007;110(10):3532-3539.
2. Fisher BT, Zaoutis T, Dvorak CC, et  al. Effect of caspofungin vs fluco-
nazole pro­phy­laxis on inva­sive fun­gal dis­ease among chil­dren and young
adults with acute mye­loid leu­ke­mia: a ran­dom­ized clin­i­cal trial. JAMA.
2019;322(17):1673-1681.
3. Donnelly JP, Chen SC, Kauffman CA, et  al. Revision and update of the
con­sen­sus def­i­ni­tions of inva­sive fun­gal dis­ease from the Euro­pean Orga-
nization for Research and Treatment of Cancer and the Mycoses Study
Group Education and Research Consortium. Clin Infect Dis. 2020;71(6):
1367-1376.
4. Rosen GP, Nielsen K, Glenn S, Abelson J, Deville J, Moore TB. Invasive fun­
gal infec­tions in pedi­at­ric oncol­ogy patients: 11-year expe­ri­ence at a sin­gle
insti­tu­tion. J Pediatr Hematol Oncol. 2005;27(3):135-140.
5. Leahey AM, Bunin NJ, Belasco JB, Meek R, Scher C, Lange BJ. Novel multi-
agent che­mo­ther­apy for bone mar­row relapse of pedi­at­ric acute lym­pho­
blas­tic leu­ke­mia. Med Pediatr Oncol. 2000;34(5):313-318.
6. Castagnola E, Palmisani E, Mesini A, Saffioti C, Micalizzi C, Dufour C. Com-
ment on: inva­sive fun­gal infec­tions in chil­dren with acute lym­pho­blas­tic
leu­ke­mia. Pediatr Blood Cancer. 2020;67(1):e28035.
7. Wang SS, Kotecha RS, Bernard A, et al. Invasive fun­gal infec­tions in chil­dren
with acute lym­pho­blas­tic leu­kae­mia: results from four Aus­tra­lian cen­tres,
2003-2013. Pediatr Blood Cancer. 2019;66(10):e27915.
8. Zaoutis TE, Argon J, Chu J, Berlin JA, Walsh TJ, Feudtner C. The epi­de­mi­­
ol­ogy and attrib­ut­­able out­comes of candidemia in adults and chil­dren
hos­pi­tal­ized in the United States: a pro­pen­sity anal­y­sis. Clin Infect Dis.
2005;41(9):1232-1239.
9. Burgos A, Zaoutis TE, Dvorak CC, et al. Pediatric inva­sive asper­gil­lo­sis: a
mul­ti­cen­ter ret­ro­spec­tive anal­y­sis of 139 con­tem­po­rary cases. Pediatrics.
2008;121(5):e1286-e1294.
10. Wattier RL, Dvorak CC, Hoffman JA, et  al. A pro­ spec­ tive, inter­
na­
tional
cohort study of inva­sive mold infec­tions in chil­dren. J Pediatric Infect Dis
Soc. 2015;4(4):313-322.
11. Fisher BT, Robinson PD, Lehrnbecher T, et al. Risk fac­tors for inva­sive fun­gal
dis­ease in pedi­at­ric can­cer and hema­to­poi­etic stem cell trans­plan­ta­tion: a
sys­tem­atic review. J Pediatric Infect Dis Soc. 2018;7(3):191-198.
12. Alexander S, Fisher BT, Gaur AH, et al; Children’s Oncology Group. Effect of
levofloxacin pro­phy­laxis on bac­ter­emia in chil­dren with acute leu­ke­mia or
under­go­ing hema­to­poi­etic stem cell trans­plan­ta­tion: a ran­dom­ized clin­i­cal
trial. JAMA. 2018;320(10):995-1004.
366  |  Hematology 2021  |  ASH Education Program
13. Johnston DL, Lewis V, Yanofsky R, et al. Invasive fun­gal infec­tions in pae­di­
at­ric acute mye­loid leu­kae­mia. Mycoses. 2013;56(4):482-487.
14. Lai HP, Chen YC, Chang LY, et al. Invasive fun­gal infec­tion in chil­dren with
per­sis­tent febrile neutropenia. J Formos Med Assoc. 2005;104(3):174-179.
15. Kavcic M, Fisher BT, Li Y, et al. Induction mor­tal­ity and resource uti­li­za­tion
in chil­dren treated for acute mye­loid leu­ke­mia at free-stand­ing pedi­at­ric
hos­pi­tals in the United States. Cancer. 2013;119(10):1916-1923.
16. Teachey DT, O’Connor D. How I treat newly diag­nosed T-cell acute lym­
pho­blas­tic leu­ke­mia and T-cell lym­pho­blas­tic lym­phoma in chil­dren. Blood.
2020;135(3):159-166.
17. Pizzo PA, Robichaud KJ, Gill FA, Witebsky FG. Empiric anti­bi­otic and anti­
fun­gal ther­apy for can­cer patients with prolonged fever and granulocyto-
penia. Am J Med. 1982;72(1):101-111.
18. Lehrnbecher T, Ethier MC, Zaoutis T, et al. International var­i­a­tions in infec­
tion sup­port­ive care prac­tices for pae­di­at­ric patients with acute mye­loid
leu­kae­mia. Br J Haematol. 2009;147(1):125-128.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/361/1851411/361fisher.pdf by guest on 13 December 2021
19. Cornely OA, Maertens J, Winston DJ, et  al. Posaconazole vs. fluconazole
or itraconazole pro­phy­laxis in patients with neutropenia. N Engl J Med.
2007;356(4):348-359.
20. Fisher BT, Kavcic M, Li Y, et  al. Antifungal pro­ phy­
laxis asso­ ci­
ated with
decreased induc­tion mor­tal­ity rates and resources uti­lized in chil­dren with
new-onset acute mye­loid leu­ke­mia. Clin Infect Dis. 2014;58(4):502-508.
21. Lehrnbecher T, Fisher BT, Phil­ lips B, et  al. Clinical prac­ tice guide­
line
for sys­temic anti­fun­gal pro­phy­laxis in pedi­at­ric patients with can­cer
and hema­to­poi­etic stem-cell trans­plan­ta­tion recip­i­ents. J Clin Oncol.
2020;38(27):3205-3216.
22. Maertens J, Verhaegen J, Lagrou K, Van Eldere J, Boogaerts M. Screening
for cir­cu­lat­ing galactomannan as a non­in­va­sive diag­nos­tic tool for inva­sive
asper­gil­lo­sis in prolonged neutropenic patients and stem cell trans­plan­ta­
tion recip­i­ents: a pro­spec­tive val­i­da­tion. Blood. 2001;97(6):1604-1610.
23. Odabasi Z, Mattiuzzi G, Estey E, et al. Beta-D-glu­can as a diag­nos­tic adjunct
for inva­sive fun­gal infec­tions: val­i­da­tion, cut­off devel­op­ment, and per­for­
mance in patients with acute mye­log­e­nous leu­ke­mia and myelodysplastic
syn­drome. Clin Infect Dis. 2004;39(2):199-205.
24. Smith PB, Ben­ja­min DK Jr, Alexander BD, Johnson MD, Finkelman MA, Stein-
bach WJ. Quantification of 1,3-beta-D-glu­can lev­els in chil­dren: pre­lim­i­nary
data for diag­nos­tic use of the beta-glu­can assay in a pedi­at­ric set­ting. Clin
Vaccine Immunol. 2007;14(7):924-925.
25. Lehrnbecher T, Robinson PD, Fisher BT, et al. Galactomannan, β-D-glu­can,
and poly­mer­ase chain reac­tion-based assays for the diag­no­sis of inva­
sive fun­ gal dis­ ease in pedi­ at­
ric can­ cer and hema­ to­
poi­ etic stem cell
trans­plan­ta­tion: a sys­tem­atic review and meta-anal­y­sis. Clin Infect Dis.
2016;63(10):1340-1348.
26. Fisher BT, Westling T, Boge CLK, et al. Prospective eval­ua ­ ­tion of galacto-
mannan and (1 → 3) β-d-glu­can assays as diag­nos­tic tools for inva­sive fun­
gal dis­ease in chil­dren, ado­les­cents, and young adults with acute mye­loid
leu­ke­mia receiv­ing fun­gal pro­phy­laxis. J Pediatric Infect Dis Soc. 2021;
10(8):864-871.
27. Armstrong AE, Rossoff J, Hollemon D, Hong DK, Muller WJ, Chaudhury
S. Cell-free DNA next-gen­er­a­tion sequenc­ing suc­cess­fully detects infec­
tious path­o­gens in pedi­at­ric oncol­ogy and hema­to­poi­etic stem cell trans­
plant patients at risk for inva­sive fun­gal dis­ease. Pediatr Blood Cancer.
2019;66(7):e27734.
28. Goldberg E, Gafter-Gvili A, Robenshtok E, Leibovici L, Paul M. Empirical
anti­fun­gal ther­apy for patients with neutropenia and per­sis­tent fever: sys­
tem­atic review and meta-anal­y­sis. Eur J Cancer. 2008;44(15):2192-2203.
29. Lehrnbecher T, Robinson P, Fisher B, et al. Guideline for the man­age­ment of
fever and neutropenia in chil­dren with can­cer and hema­to­poi­etic stem-cell
trans­plan­ta­tion recip­i­ents: 2017 update. J Clin Oncol. 2017;35(18):2082-2094.
30. Cordonnier C, Pautas C, Maury S, et al. Empirical ver­sus pre­emp­tive anti­
fun­gal ther­apy for high-risk, febrile, neutropenic patients: a ran­dom­ized,
con­trolled trial. Clin Infect Dis. 2009;48(8):1042-1051.
31. Santolaya ME, Alvarez AM, Acuña M, et al. Efficacy of pre-emptive ver­sus
empir­i­cal anti­fun­gal ther­apy in chil­dren with can­cer and high-risk febrile
neutropenia: a ran­dom­ized clin­i­cal trial. J Antimicrob Chemother. 2018;
73(10):2860-2866.
32. Groll AH, Pana D, Lanternier F, et al; 8th Euro­pean Conference on Infections
in Leukaemia. 8th Euro­pean Conference on Infections in Leukaemia: 2020
guide­lines for the diag­no­sis, pre­ven­tion, and treat­ment of inva­sive fun­gal
dis­eases in pae­di­at­ric patients with can­cer or post-haematopoietic cell
trans­plan­ta­tion. Lancet Oncol. 2021;22(6):e254-e269.
33. De Pauw B, Walsh TJ, Donnelly JP, et al; Euro­pean Organization for Research
and Treatment of Cancer/Invasive Fungal Infections Cooperative Group;
 National Institute of Allergy and Infectious Diseases Mycoses Study Group
Consensus Group. Revised def­i­ni­tions of inva­sive fun­gal dis­ease from the
Euro­pean Organization for Research and Treatment of Cancer/Invasive
Fungal Infections Cooperative Group and the National Institute of Allergy
and Infectious Diseases Mycoses Study Group (EORTC/MSG) Consensus
Group. Clin Infect Dis. 2008;46(12):1813-1821.
34. Hovi L, Saxen H, Saarinen-Pihkala UM, Vettenranta K, Meri T, Richardson
M. Prevention and mon­ i­
tor­
ing of inva­sive fun­
gal infec­
tions in pedi­ at­
ric
patients with can­cer and hema­to­logic dis­or­ders. Pediatr Blood Cancer.
2007;48(1):28-34.
35. Steinbach WJ, Addison RM, McLaughlin L, et  al. Prospective Aspergillus
galactomannan anti­gen test­ing in pedi­at­ric hema­to­poi­etic stem cell trans­
plant recip­i­ents. Pediatr Infect Dis J. 2007;26(7):558-564.
Prakash Singh Shekhawat
36. Hayden R, Pounds S, Knapp K, et al. Galactomannan antigenemia in pedi­
at­ric oncol­ogy patients with inva­sive asper­gil­lo­sis. Pediatr Infect Dis J.
2008;27(9):815-819.
37. Badiee P, Alborzi A, Karimi M, et  al. Diagnostic poten­tial of nested PCR,
galactomannan EIA, and beta-D-glu­can for inva­sive asper­gil­lo­sis in pedi­at­
ric patients. J Infect Dev Ctries. 2012;6(4):352-357.
38. Koltze A, Rath P, Schöning S, et al. β-D-glu­can screen­ing for detec­tion of
inva­sive fun­gal dis­ease in chil­dren under­go­ing allo­ge­neic hema­to­poi­etic
stem cell trans­plan­ta­tion. J Clin Microbiol. 2015;53(8):2605-2610.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000267
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/361/1851411/361fisher.pdf by guest on 13 December 2021
Fungal man­age­ment in leu­ke­mia  |  367
 IT TAKES A VILLAGE: MAXIMIZING SUPPORTIVE CARE AND MINIMIZING TOXICITY DURING CHILDHOOD LEUKEMIA THERAPY

Minimizing cardiac toxicity in children

with acute myeloid leukemia

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/368/1851685/368narayan.pdf by guest on 13 December 2021

Hari K. Narayan,1 Kelly D. Getz,2 and Kasey J. Leger3
1
Department of Pediatrics, University of California San Diego, La Jolla, CA; 2Departments of Biostatistics, Epidemiology & Informatics and Pediatrics,
Perelman School of Medicine, University of Pennsylvania; Division of Oncology, The Children’s Hospital of Philadelphia, Philadelphia, PA; and
3
Department of Pediatrics, University of Washington, Seattle Children’s Hospital, Seattle, WA

Anthracycline chemotherapy remains an integral component of modern pediatric acute myeloid leukemia (AML) regimens
and is often delivered at high doses to maximize cancer survival. Unfortunately, high-dose anthracyclines are associated with
a significant risk of cardiotoxicity, which may result in early and/or long-term left ventricular systolic dysfunction and heart
failure. Moreover, the development of cardiotoxicity during pediatric AML therapy is associated with lower event-free and
overall survival, which may be partially attributable to incomplete anthracycline delivery. A combined strategy of primary
cardioprotection and close cardiac monitoring can maximize chemotherapy delivery while reducing the toxicity of inten-
sive AML therapy. Primary cardioprotection using dexrazoxane reduces short-term cardiotoxicity without compromising
cancer survival. Liposomal anthracycline formulations, which are under active investigation, have the potential to mitigate
cardiotoxicity while also improving antitumor efficacy. Primary cardioprotective strategies may reduce but not eliminate
the risk of cardiotoxicity; therefore, close cardiac monitoring is also needed. Standard cardiac monitoring consists of serial
echocardiographic assessments for left ventricular ejection fraction decline. Global longitudinal strain has prognostic utility
in cancer therapy-related cardiotoxicity and may be used as an adjunct assessment. Additional cardioprotective measures
should be considered in response to significant cardiotoxicity; these include cardiac remodeling medications to support
cardiac recovery and anthracycline dose interruption and/or regimen modifications. However, the withholding of anthracy-
clines should be limited to avoid compromising cancer survival. A careful approach to cardioprotection during AML therapy
is critical to maximize the efficacy of leukemia treatment while minimizing the short- and long-term risks of cardiotoxicity.

LEARNING OBJECTIVES
• Recognize the impact of anthracycline-associated cardiotoxicity on cardiac and leukemia outcomes in the treat-
ment of pediatric acute myeloid leukemia
• Understand the role of primary cardioprotection during anthracycline-containing chemotherapy for pediatric
acute myeloid leukemia
• Understand the strengths and limitations of cardiac imaging modalities and the role of echocardiography during
pediatric acute myeloid leukemia therapy

cycles including 300 mg/m2 daunorubicin and 48 mg/m2

CLINICAL CASE
A 15-year-old girl was diagnosed with acute myeloid leu-
kemia (AML) after presenting with fatigue, menorrhagia,
and fever. Cytogenetic analyses demonstrated a t(9;11)
KMT2A-MLLT3 fusion. The patient commenced induc-
tion therapy with cytarabine, daunorubicin with dexra-
zoxane, and gemtuzumab ozogamicin. After induction,
there was no evidence of residual leukemia, and she was
stratified to low-risk AML treatment per the standard arm
of the Children’s Oncology Group (COG) trial AAML1831
(NCT04293562) with planned receipt of 5 chemotherapy
mitoxantrone. On day 14 of intensification 1 (following
300 mg/m2 daunorubicin), she was admitted to the inten-
sive care unit with klebsiella septic shock. After pressor
initiation for fluid-refractory hypotension, echocardiogra-
phy demonstrated a significant decline in left ventricular
ejection fraction (LVEF) from 60% at baseline to 35%.
Introduction
While clinical outcomes of pediatric AML have improved
over time with intensification of therapy, the risk of leukemia

368 | Hematology 2021 | ASH Education Program

recur­rence and ther­apy-related mor­bid­ity and mor­tal­ity remains
high.1 Anthracyclines are crit­i­cal for opti­mal sur­vival,2 and deliv­
ered doses in AML sub­sets treated with­out consolidative trans­
plant may exceed 600 mg/m2 of doxo­ru­bi­cin equiv­a­lents after
account­ing for the recently rec­og­nized 10:1 cardiotoxic dose
equiv­a­lence of mitoxantrone in com­par­i­son with doxo­ru­bi­cin.3
As a result, anthracycline-asso­ci­ated cardiotoxicity is com­mon,
with late car­dio­my­op­a­thy rates of up to 27%.4 Early LV sys­tolic
dys­func­tion (LVSD) dur­ing AML ther­apy is also com­mon and has
impor­tant impli­ca­tions for car­dio­vas­cu­lar out­comes and sur­vival.
The inci­dence and impact of early LVSD were eval­u­ated in 2 re-
cent ran­dom­ized phase 3 COG tri­als in de novo pedi­at­ric AML.5,6
In AAML0531, 12% of chil­dren with non-FLT3-mutant AML devel­

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/368/1851685/368narayan.pdf by guest on 13 December 2021

oped grade 2 + LVSD dur­ing or fol­low­ing ther­apy (n =  1022; 5-
year fol­low-up).6 In AAML1031, with more com­pre­hen­sive ascer­
tain­ment due to man­dated site reporting of car­diac func­tional
param­e­ters, the rate of grade 2 + LVSD was 21% (n =  1014; median
3.5-year fol­low-up).5 In fur­ther anal­y­sis of AAML0531, those who
devel­oped LVSD on ther­apy were much more likely to have per­
sis­tent or recur­rent LVSD dur­ing off-pro­to­col fol­low-up (haz­ard
ratio [HR], 12.1; 95% CI, 4.2-34.8). Moreover, on-ther­apy grade
2 + LVSD was asso­ci­ated with a >15% reduc­tion in 5-year event-
free and over­all sur­vival (Figure 1).6 Given that this pro­to­col re-
quired dis­con­tin­u­a­tion of anthracyclines after the devel­op­ment
of LVSD, infe­ rior sur­ vival was likely driven in part by incom­
plete anthracycline deliv­ery com­pro­mis­ing che­mo­ther­a­peu­tic
effi­cacy. Thus, devel­op­ing an effec­tive strat­egy to reduce the
cardiotoxicity of AML ther­apy while maintaining leu­ke­mia-free
sur­vival is crit­i­cal. A com­bined approach of pri­mary cardiopro-
tection and close car­diac mon­i­tor­ing to guide tox­ic­ity-respon­
sive cardioprotective mea­sures dur­ing che­mo­ther­apy can help
bal­ance the some­what com­pet­ing goals of max­i­mal ther­apy
Figure 1. Event-free (A) and Overall (B) survival according to
deliv­ery and cardiotoxicity reduc­tion to opti­mize both leu­ke­mia
the occurrence of Grade 2 + LVSD, in the presence or absence
and car­dio­vas­cu­lar out­comes.
of associated bloodstream infection, on COG trial AAML0531.
Grade 2 + LVSD was defined as LVEF <50% or LVFS <24% based
Primary cardioprotection
on Common Terminology Criteria for Adverse Events V3.0.
Primary cardioprotection involves the con­cur­rent use of cardi-
Adapted with permission from Getz et al.6
oprotective med­ i­
ca­
tions dur­
ing anthracycline ther­ apy or the
use of less cardiotoxic agents to reduce cardiotoxicity risk. Two
impor­tant pri­mary cardioprotective strat­e­gies per­ti­nent to pedi­
at­ric AML ther­apy are dexrazoxane and lipo­so­mal anthracyclines. equiv­a­lence ratio to doxo­ru­bi­cin, wherein dexrazoxane is dosed
at 5-10:1 with dau­no­ru­bi­cin, 40:1 with mitoxantrone, and 50:1
Dexrazoxane with idarubicin.10 However, given evolv­ing dose equiv­a­lence ra-
Dexrazoxane is the only US Food and Drug Administration-ap- tios as new data sets emerge,3 dexrazoxane dos­ing should also
proved drug for pre­vent­ing anthracycline-induced cardiotoxici- take into con­sid­er­ation dem­on­strated phar­ma­co­ki­net­ics, safety,
ty. Its approval is restricted to met­a­static breast can­cer patients and effi­cacy. Studies of dexrazoxane in com­bi­na­tion with agents
receiv­ing >300 mg/m2 of doxo­ru­bi­cin based on data across mul­ com­monly used in AML (ie, dau­no­ru­bi­cin, mitoxantrone, idaru-
ti­ple ran­dom­ized phase 3 tri­als dem­on­strat­ing lower rates of car­ bicin) have shown safety and/or effi­cacy in dose ranges/ratios
diac events and/or heart fail­ure with con­cur­rent dexrazoxane.7 sim­i­lar to those pro­posed above.7,10
In 2014 dexrazoxane was des­ig­nated as an orphan drug for the Dexrazoxane has been shown to sig­nif­i­cantly reduce short-
“pre­ven­tion of car­dio­my­op­a­thy for chil­dren and ado­les­cents 0 term LVSD in anthracycline-treated chil­dren with acute lym­pho­
through 16 years of age treated with anthracyclines,” allowing blas­tic leu­ke­mia, non-Hodgkin lym­phoma, and Ewing sar­coma.11-13
for its use in pedi­at­ric AML. Dexrazoxane mit­i­gates anthracy- Evaluations of dexrazoxane use in pedi­at­ric AML, though prom­is­
cline-induced cardiomyocyte injury by stim­u­lat­ing the selec­tive ing, have his­tor­i­cally been lim­ited to small sin­gle-insti­tu­tion stud­
deg­ra­da­tion of topoisomerase IIβ, a molec­u­lar tar­get of anthra- ies.14,15 More recently, the short-term cardioprotective effects of
cyclines, thereby reduc­ing DNA dam­age and oxi­da­tive stress.8 dexrazoxane were ana­lyzed in the COG trial for de novo AML,
Dexrazoxane is admin­is­tered as a bolus infu­sion over 15 min­utes AAML1031.5 Dexrazoxane use was nonrandomized and deter­
imme­di­ately prior to anthracycline admin­is­tra­tion at a dose ratio mined by insti­ tu­tional prac­tice; 96 (9.5%) patients received
of 10:1 with doxo­ru­bi­cin.9 Dosing with alter­na­tive anthracycline dexrazoxane and 918 (90.5%) did not. The 4-year risk for grade
agents has his­tor­i­cally been based on their cardiotoxicity dose 2 + LVSD was 45% lower with dexrazoxane than with­out (26.5%
Prakash Singh Shekhawat
Minimizing cardiotoxicity in pedi­at­ric AML  |  369

Figure 2. Temporal trends in LVEF by dexrazoxane exposure.
Standard errors of the mean LVEF measurements are presented
as vertical bars. The (*) denotes treatment courses that include
an anthracycline. Adapted with permission from Getz et al.5
vs 42.2%; HR =  0.55; 95% CI, 0.36-0.86; P =  .009), with greater
reduc­tions in risk for higher-grade cardiotoxicity. While dexra-
zoxane did not pro­vide com­plete pro­tec­tion against LVSD, its
use was asso­ci­ated with smaller declines in LVEF (Figure 2), less
wors­en­ing over time, and a trend toward greater like­li­hood of
LVSD res­o­lu­tion. Although label­ing warns that dexrazoxane may
increase the myelosuppressive effects of che­mo­ther­apy, dura­
tions of neutropenia, inten­sive care unit require­ments, and rates
of mucositis and blood­stream infec­tion did not dif­fer with dexra-
zoxane expo­sure. Additionally, 5-year event-free (49.0% vs 45.1%;
P =  .534) and over­all sur­vival esti­ma­tes (65.0% vs 61.9%; P  =  .613)
were com­pa­ra­ble between those who did and did not receive
dexrazoxane. However, sig­ nif­i­
cantly lower treat­ ment-related
mor­tal­ity with dexrazoxane (4.6% vs 12.6%; P =  .024) sug­gests
that it may pre­vent acute car­diac events that con­trib­ute to early
patient deaths.
Despite com­pel­ling evi­dence of an early cardioprotective ben­
e­fit, <10% of AML patients receive dexrazoxane, likely due to early
con­cerns for increased sec­ond­ary malig­nancy risk.5 This asso­ci­a­tion
was ini­tially suggested in an anal­y­sis of two Hodgkin lym­phoma
tri­als that included the ran­dom­ized admin­is­tra­tion of dexrazoxane
in the con­text of mod­er­ate-dose anthracycline, etoposide, and
alkylator ther­apy, P9425 (n =  216) and P9426 (n =  255).16 However, in
sub­se­quent ana­ly­ses of these tri­als and the COG T-cell acute lym­
pho­blas­tic leu­ke­mia/lym­phoma trial P9404 (n =  537) with lon­ger
fol­low-up (median, 12.6 years),17 dexrazoxane did not sig­nif­i­cantly
increase the 10-year relapse risk (16.1% with dexrazoxane vs 19.1%
with­out; HR =  0.81; 95% CI, 0.60-1.08), over­all mor­tal­ity (12.8% vs
12.2%; HR =  1.03; 95% CI, 0.73-1.45), or deaths due to sec­ond­ary
can­cers (2.0% vs 1.6%; HR =  1.24; 95% CI, 0.49-3.15; Figure 3). More-
over, these find­ings remained sim­il­ar in long-term ana­ly­ses of the
indi­vid­ual Hodgkin lym­phoma tri­als. Likewise, sec­ond­ary malig­
nan­cies were rare over the fol­low-up on AAML1031 (n =  4), with only
one occur­ring after dexrazoxane ther­apy.5
Overall, suf­fi­
cient data exist to sup­ port the use of dexra-
zoxane in chil­dren with AML to mit­i­gate short-term LVSD and
improve the poten­tial for tar­get anthracycline dose deliv­ery. It
370  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/368/1851685/368narayan.pdf by guest on 13 December 2021
Figure 3. Cumulative incidence of relapse and overall mortali-
ty in the combined COG randomized trials of DRZ. Cumulative
incidences at 10 years were not significantly different by DRZ
status for either outcome. DRZ, dexrazoxane; DRZ+, exposed to
DRZ; DRZ−, not exposed to DRZ. Adapted with permission from
Chow et al.17
should be noted that dexrazoxane reduces but does not elim­
i­nate short-term cardiotoxicity. In addi­tion, fur­ther research is
needed to assess non­car­diac toxicities such as typhlitis rates,
which were qual­i­ta­tively higher in Hodgkin lym­phoma patients
receiv­ing dexrazoxane in P9425 (8.4% vs 2.8%) and to deter­mine
the long-term effects of dexrazoxane on car­dio­vas­cu­lar mor­bid­
ity and mor­tal­ity.18
Liposomal anthracyclines
Liposomal deliv­ery sys­tems for anthracyclines are a prom­is­ing
strat­egy to mit­i­gate cardiotoxicity while enhanc­ing anti­tu­mor
effi­cacy.19-21 Liposomal drug encap­ su­
la­
tion reduces its pen­ e­
trance into the tight cap­il­lary junc­tions of the heart while allow-
ing sustained sys­temic cir­cu­la­tion and pref­er­en­tial accu­mu­la­tion
in the tumor.22 Meta-ana­ly­ses of adult tri­als com­par­ing lipo­so­mal
doxo­ru­bi­cin to free drug dem­on­strate a >50% reduc­tion in clin­
i­cal and sub­clin­i­cal car­diac dys­func­tion with lipo­so­mal for­mu­la­
tions.23,24 Intensification of che­mo­ther­apy with lipo­so­mal encap­
su­lated dau­no­ru­bi­cin, DaunoXome® (DNX), has proven effec­tive
and tol­er­a­ble in the con­text of pedi­at­ric AML stud­ies conducted
by the International Berlin-Frankfurt-Münster Study Group. In
the relapse set­ting, DNX plus fludarabine and cytarabine (FLAG)
dem­on­strated enhanced leu­ke­mic effi­cacy and com­pa­ra­ble car-
diotoxicity com­pared to FLAG alone.25 Children with de novo
AML expe­ri­enced sim­i­lar rates of cardiotoxicity and less over­
all treat­ment-related tox­ic­ity with DNX, given at a higher than
equiv­a­lent dose vs idarubicin, dur­ing induc­tion.26 DNX is no lon­
ger being manufactured and thus is not avail­­able for use in AML.
CPX-351 is a lipo­so­mal for­mu­la­tion of dau­no­ru­bi­cin plus cytar-
abine recently approved by the US Food and Drug Administration
in adults and chil­dren with ther­apy-related AML or AML/myelo-
dysplastic syn­drome, given its favor­able effi­cacy and tol­er­a­ble
safety pro­file.27,28 The cardioprotective effects of CPX-351 are not
established; how­ever, clin­i­cal cardiotoxicity has been described
in early tri­als. A ran­dom­ized study of CPX-351 vs dau­no­ru­bi­cin
and cytarabine in 309 older adults with sec­ond­ary AML dem­on­
strated sim­i­lar rates of cardiotoxicity despite nearly dou­ble the
rate of postremission consolidative anthracycline deliv­ery on the
CPX-351 arm.28 In 38 heavily anthracycline pretreated chil­dren
with relapsed AML who received CPX-351 followed by FLAG on
the COG phase 1/2 study, grade 2 + LVSD occurred in 7 (18%).27
There are a pau­city of data to inform how this com­pares with
tra­di­tional anthracycline or nonanthracycline sal­vage reg­i­mens.
CPX-351 is cur­rently being stud­ied in the up-front set­ting by COG
in a phase 3 ran­dom­ized study in de novo AML (NCT04293562).
If found to mit­i­gate cardiotoxicity while enhanc­ing or even main-
taining leu­ke­mia-free sur­vival, lipo­so­mal anthracyclines could
sub­stan­tially improve ther­a­peu­tic options for AML.
Cardiac mon­i­tor­ing and tox­ic­ity-respon­sive
cardioprotection
While pri­mary cardioprotection may reduce the risk for cardio-
toxicity, the above data dem­ on­strate that these approaches
have not elim­i­nated car­diac risk. Thus, close monitoring of car­
diac func­tion dur­ing and after AML ther­apy remains crit­i­cal to
clin­i­cal man­age­ment and is used to guide tox­ic­ity-respon­sive
cardioprotective inter­ven­tions.
Cardiac mon­i­tor­ing
Cardiac mon­i­tor­ing dur­ing pedi­at­ric AML ther­apy typ­i­cally con­
sists of serial precycle echo­ car­di­
og­ra­
phy to assess for LVSD,
manifested as declines in LV frac­tional short­en­ing (LVFS), LVEF,
and/or global lon­gi­tu­di­nal strain (GLS). Each of these meth­ods
has strengths and lim­i­ta­tions that are rel­e­vant to clin­i­cal man­
age­ment (Table 1).
LVFS is the frac­tional cav­ity diam­e­ter change between dias­
tole and sys­tole. It is assessed with two-dimen­sional (2D) or
M-mode echo­car­di­og­ra­phy via a lin­ear inter­ro­ga­tion through the
cen­ter of the LV. LVFS has a long his­tor­i­cal pre­ce­dent and is easy
to acquire and mea­sure.29 However, it is sub­op­ti­mal in screen­
ing given its sig­nif­i­cant angle depen­dence and poor reli­abil­ity
between acqui­si­tions.30
LVEF is the frac­tional cav­ity vol­ume change between dias­
tole and sys­tole. Cardiac mag­netic res­o­nance imag­ing (CMRI)
is the opti­mal modal­ity for LVEF der­i­va­tion given its high-res­
o­lu­tion, direct LV mea­sure­ment in 3 dimen­sions with­out geo­
met­ric assump­tions.31 However, CMRI is cur­rently reserved for
sec­ond­ary eval­u­a­tion, pri­mar­ily related to its lesser avail­abil­ity.32
Echocardiography, the stan­ dard pri­ mary screen­ ing modal­ ity,
can be used to derive LVEF using mul­ti­ple meth­ods. The Teich-
holz for­mula can be used to cal­cu­late LVEF from lin­ear dimen­
sions; how­ ever, this method is not recommended due to its
sig­nif­i­cant geo­met­ric assump­tions and resulting inaccuracy.33,34
Alternate 2D LVEF der­i­va­tions include the five/sixths area-length
(ie, “bul­let”) method and the biplane Simpson’s method; pref­er­
ence among these is gen­er­ally lab­o­ra­tory depen­dent. Although
supe­rior to Teichholz, these meth­ods also have lim­i­ta­tions due
to chal­lenges with angle depen­dence, mea­sure­ment repro­duc­
ibil­ity, and geo­met­ric assump­tions such that their sen­si­tiv­ity to
detect abnor­mal LVEF com­pared to CMRI in child­hood can­cer
sur­vi­vor pop­u­la­tions is only approx­i­ma­tely 50%.33,35
Three-dimen­sional (3D) echo­car­di­og­ra­phy allows com­plete
LV visu­al­i­za­tion, yield­ing a more repro­duc­ible and accu­rate LVEF
mea­sure­ment in com­par­i­son with 2D meth­ods; the sen­si­tiv­ity to
detect LVSD in com­par­i­son with MRI is 53% to 68% in child­hood
Prakash Singh Shekhawat
can­cer sur­vi­vor pop­u­la­tions.33,35 However, lim­i­ta­tions include the
need for spe­cial­ized equip­ment, soft­ware, and train­ing and a
pau­city of data supporting its use in youn­ger chil­dren with can­
cer. When avail­­able, 3D LVEF is the pre­ferred screen­ing method
in ado­les­cents and young adults.32
Although valid and impor­tant heart fail­ure screen­ing mea­
sures, LVEF and LVFS declines may be late find­ings in the pro­
gres­sion of anthracycline-related car­dio­my­op­a­thy.36 Strain,
spe­cif­i­cally GLS, has been pro­posed as an alter­na­tive mea­sure to
facil­i­tate ear­lier, more sen­si­tive cardiotoxicity detec­tion. Strain
is the frac­tional change in myo­car­dial fiber length, or defor­ma­
tion, dur­ing the car­diac cycle. GLS rep­re­sents strain along the LV
long axis, aver­aged across the LV. GLS declines pre­cede and are
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/368/1851685/368narayan.pdf by guest on 13 December 2021
prog­nos­tic of LVEF declines in adults dur­ing and after anthracy-
cline ther­apy.37 GLS abnor­mal­i­ties are also prev­a­lent in child­hood
can­cer sur­vi­vors in the pres­ence of nor­mal LVEF and are asso­
ci­ated with established car­dio­vas­cu­lar risk fac­tors, suggesting
that they are clin­i­cally rel­e­vant.38 In recent years GLS has been
incor­po­rated in adult cardio-oncol­ogy imag­ing guide­lines, and
its use is becom­ing more wide­spread.32 However, 1-year results
from the only ran­dom­ized trial com­par­ing the effi­cacy of a GLS-
vs LVEF-guided approach to cardioprotective ther­apy ini­ti­a­tion
in adults receiv­ing anthracyclines failed to dem­on­strate a ben­e­fit
of GLS.39 In addi­tion, pedi­at­ric data attesting to the util­ity of GLS
dur­ing anthracycline ther­apy are lim­ited. Based on the cur­rent
evi­dence, GLS may serve as an adjunct mea­sure, but AML treat­
ment deci­sions should not be altered based on GLS alone.40
Toxicity-respon­sive cardioprotection dur­ing che­mo­ther­apy
Cardiac mon­i­tor­ing pro­vi­des the oppor­tu­nity for addi­tional cardi-
oprotective inter­ven­tions in response to cardiotoxicity detec­tion,
with the goal of reduc­ing fur­ther car­diac injury and supporting
car­diac recov­ery. The two main strat­e­gies of cardioprotection in
response to LVSD dur­ing che­mo­ther­apy include with­hold­ing an-
thracyclines (ide­ally tem­po­rar­ily) and using car­diac remodeling
med­i­ca­tions. There are lim­ited data to guide cli­ni­cians regard­ing
when to insti­tute these inter­ven­tions; acknowl­edg­ing this pau­
city, our pro­posed man­age­ment strat­egy is depicted in Figure 4.
Dose inter­rup­tions or reduc­tions in anthracycline deliv­ery should
be min­i­mized given the poten­tial to reduce che­mo­ther­a­peu­tic
effi­cacy. Brief delays may be con­sid­ered in the set­ting of LVSD,
with the goal of func­tional recov­ery and resump­tion of anthracy-
cline deliv­ery. Withholding anthracycline should be con­sid­ered
in cases of sig­nif­i­cant dys­func­tion; in these set­tings we sug­gest
replacing it with an inten­sive non-anthracycline-containing che­
mo­ther­apy block such as high-dose cytarabine and asparaginase
(Capizzi II). The pres­ence of per­sis­tent and/or sig­nif­i­cant LVSD
should prompt car­di­­ol­ogy con­sul­ta­tion, fur­ther car­diac test­ing,
and con­sid­er­ation of car­diac remodeling med­i­ca­tions such as
angio­ten­sin converting enzyme inhib­i­tor or beta-blocker ther­
apy; fur­ther pedi­at­ric data are needed to deter­mine the ben­
e­fit of these med­i­ca­tions in this set­ting. Recovery of func­tion
after these tox­ic­ity-respon­sive inter­ven­tions may allow for the
resump­tion or con­tin­u­a­tion of anthracycline deliv­ery.
Summary
The goals of care dur­ ing pedi­at­
ric AML ther­ apy are to max­ i­
mize cure while min­i­miz­ing short- and long-term cardiotoxici-
ty. To pro­mote the safety and effi­cacy of che­mo­ther­apy, treat­
ment reg­i­mens should include pri­mary cardioprotection in the
Minimizing cardiotoxicity in pedi­at­ric AML  |  371
Table 1. Indices used to assess for LVSD

Threshold for sig­

Measure Strengths Limitations nif­i­cant LVSD Recommendation for use

LVFS

• Long his­tor­i­cal pre­ce­dent •  Highly angle depen­dent <24%a Adjunct mea­sure only except
• Easy to acquire and mea­sure • Poor reli­abil­ity (~7% error in the set­ting of inad­e­quate
when in the nor­mal range)30 win­dows for alter­nate mea­
sures

LVEF

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/368/1851685/368narayan.pdf by guest on 13 December 2021

Teichholz for­mula • Easy to acquire and mea­sure • Significant geo­met­ric <50%a Not recommended
(from lin­ear dimen­ assump­tions
sions) • Inaccurate

2D echo • Standard across most lab­o­ •  Geometric assump­tions May be used as pri­mary
(5/6 area length or ra­to­ries •  Measurement error screen­ing
biplane Simpson’s • Established util­ity in heart • Only ~50% sen­si­tiv­ity to
method) fail­ure detect LVSD in com­par­i­son
with CMRI (in sur­vi­vor data)

3D echo •  Better repro­duc­ibil­ity • Requires spe­cial­ized train­ing Preferred for pri­mary screen­
• 53%–68% sen­si­tiv­ity to and equip­ment ing in ado­les­cents and young
detect LVSD in com­par­i­son • Limited data in youn­ger adults, when avail­­able
with CMRI (in sur­vi­vor data) chil­dren

CMRI • Optimal LVEF repro­duc­ibil­ity
and accu­racy
• Can mea­sure other poten­
tially impor­tant car­diac
fea­tures (eg, pre­cise mea­
sure­ment of mass; edema,
fibro­sis)
• Requires spe­cial­ized train­ing Secondary eval­u­a­tion as indi­
and equip­ment catedb
•  Less avail­­ableb
• Compliance chal­lenges in
youn­ger chil­drenb
• An improve­ment in out­
comes has not been dem­on­
strated in adult cardio-oncol­
ogy pop­u­la­tions
• Limited pedi­at­ric cardio-
oncol­ogy data

GLS

• More sen­si­tive mea­sure of • An improve­ment in out­ >−16% (approx­i­ Adjunct mea­sure only (fur­ther
car­diac func­tion comes has not been dem­on­ mate)42 or ≥15% study needed)
• Prognostic of sub­se­quent strated in adult cardio-oncol­ wors­en­ing from
LVEF declines37 ogy pop­u­la­tions39 base­line valuec
• Limited pedi­at­ric cardio-
oncol­ogy data

The table depicts the der­iv ­ a­tions of imag­ing-based mea­sures of left ven­tric­ul­ar sys­tolic func­tion. LVFS is the frac­tional change in cav­ity diam­e­ter dur­ing the car­diac cycle. LVEF
is the frac­tional change in cav­ity vol­ume dur­ing the car­diac cycle. Strain is the frac­tional change in myo­car­dial fiber length, or defor­ma­tion, dur­ing the car­diac cycle, typ­i­cally
derived from speckle-track­ing anal­y­sis of 2D echo­car­dio­graphic images. GLS is the aver­age defor­ma­tion in the lon­gi­tu­di­nal dimen­sion. 2D or 3D echo­car­dio­graphic der­i­va­tions
of LVEF should be the pri­mary screen­ing mea­sure used to assess for LVSD dur­ing pedi­at­ric AML ther­apy. Strengths, lim­it­ a­tions, and rec­om­men­da­tions for use are based on gen­
eral car­dio­vas­cu­lar-imag­ing quan­ti­fi­ca­tion guide­lines and adult cardio-oncol­ogy guide­lines, with addi­tional sup­port­ive ref­er­ences listed within the table and foot­notes.29,32,34,40
a
LVFS and LVEF cut­offs are based on grade 2 sever­ity events for the terms “left ven­tric­u­lar sys­tolic dys­func­tion” and “ejec­tion frac­tion decreased” in the National Cancer
Institute Common Terminology Criteria for Adverse Events (Versions 3.0 and 5.0, respec­tively). Precise cut­offs may vary among treat­ment pro­to­cols, and cli­ni­cians should
con­sult the spe­cific pro­to­col for fur­ther guid­ance.
b
Some indi­ca­tions for CMRI in sec­ond­ary eval­u­a­tion are poor echo­car­dio­graphic win­dows resulting in inad­e­quate assess­ment of func­tion, bor­der­line/inde­ter­mi­nate echo­
car­dio­graphic results in which a more pre­cise LVEF esti­mate would change clin­i­cal man­age­ment, or base­line or sig­nif­i­cant LVSD to eval­u­ate for under­ly­ing car­dio­my­op­a­
thy.31,32 CMRI does not employ ion­iz­ing radi­a­tion, and assess­ment of LV vol­umes, mass, and LVEF does not require intra­ve­nous con­trast. However, youn­ger chil­dren require
seda­tion/anes­the­sia to com­plete CMRI, which is an impor­tant con­sid­er­ation.43 Increasing avail­abil­ity cou­pled with ongo­ing tech­no­log­i­cal improve­ments that reduce motion
arti­facts/patient com­pli­ance demands, scan­ning time, and postprocessing time may result in increased CMRI use in pedi­at­ric can­cer pop­u­la­tions in the future.
c
GLS val­ues are typ­i­cally expressed as a neg­a­tive per­cent­age; less neg­a­tive val­ues reflect worse sys­tolic func­tion (eg, −16% is worse than −18%). The limit of nor­mal has not
been pre­cisely established in pedi­at­rics; the reported limit of −16% is an impre­cise esti­mate based on prior pedi­at­ric stud­ies.42 The wors­en­ing of 15% from the base­line value
is based on adult cardio-oncol­ogy guide­lines and refers to the per­cent wors­en­ing from the base­line value,32 not the abso­lute wors­en­ing in GLS (eg, a change from −20% to
−17% is a 15% wors­en­ing). LVEDD, left ven­tric­u­lar end-dia­stolic diam­e­ter; LVEDV, left ven­tric­u­lar end-dia­stolic vol­ume; LVESD, left ven­tric­u­lar end-sys­tolic diam­e­ter; LVESV,
left ven­tric­u­lar end-sys­tolic vol­ume.

372  |  Hematology 2021  |  ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/368/1851685/368narayan.pdf by guest on 13 December 2021
Figure 4. Proposed management of cardiotoxicity during pediatric AML therapy. This figure depicts our proposed algorithm to
manage cardiotoxicity during AML therapy, with the goal of balancing cardiac safety with effective delivery of AML-directed che-
motherapy. LVEF cutoffs are based on National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) V3.0 for
LVSD: grade 2 LVSD is defined as asymptomatic LVEF decline of 40% to 49%. Grade 3 LVSD is defined as LVEF decline of 20% to 39%,
or symptomatic heart failure. These LVEF thresholds are consistent with grade 2/3 “ejection fraction decreased” in CTCAE V5.0.
*Additional testing may include biomarkers (eg, N-terminal-pro B-type natriuretic peptide), CMRI, or alternate testing as indicated.
Biomarkers currently have an established role in secondary/further evaluation but are under investigation as a primary screening
modality. ^Cardiac remodeling medications may include angiotensin converting enzyme inhibitors (eg, enalapril) and beta-blockers
(eg, carvedilol). #When omission of anthracyclines is necessary due to persistent LVSD, efforts should be made to maintain the inten-
sity of chemotherapy, if appropriate. Thus, we suggest replacing with an intensive non-anthracycline-containing chemotherapy block,
such as high-dose cytarabine and asparaginase (Capizzi II).

form of con­cur­rent dexrazoxane and car­diac mon­i­tor­ing with Acknowledgments

addi­tional tox­ic­ity-respon­sive cardioprotective inter­ven­tions as
indi­cated. Given the del­i­cate bal­ance needed to man­age com­
pet­ing onco­logic and car­dio­vas­cu­lar risks, a close col­lab­o­ra­tion
between oncol­o­gists and car­di­­ol­o­gists is crit­i­cal to opti­miz­ing
patient care. This col­lab­o­ra­tion should begin dur­ing che­mo­ther­
apy and con­tinue in sur­vi­vor­ship, with long-term mon­i­tor­ing and
the poten­ tial implementation of sec­ ond­ary and ter­ tiary heart
fail­ure pre­ven­tion mea­sures.41 As onco­logic ther­a­pies con­tinue
to rap­idly evolve, the devel­op­ment of ded­i­cated, inter­dis­ci­plin­
ary pedi­at­ric cardio-oncol­ogy pro­grams may serve to meet this
impor­tant need.
Hari K. Narayan is supported by Padres Pedal the Cause/RADY
no. #PTC2020 and UC San Diego Moore Cancer Center, Special-
ized Cancer Control Support Grant NIH/NCI P30CA023100. Kelly
Getz is supported by an National Heart, Lung, and Blood Insti-
tute Career Development Award (1K01HL143153-01).
Conflict-of-inter­est dis­clo­sure
Hari K. Narayan: nothing to disclose.
Kelly Getz: nothing to disclose.
Kasey J. Leger: research funding: Abbott Diagnostics; advi­sory
board: Jazz Pharmaceuticals.

Off-label drug use

CLINICAL CASE (Con­t in­u ed)
The patient was diag­nosed with grade 3 LVSD in the set­ting
of sep­sis. With inten­sive sup­port­ive care, the patient recov­
ered and was discharged on cycle day 34. LVEF improved to
48% on echo­car­di­og­ra­phy at dis­charge, with fur­ther recov­ery
to 55% 1.5 weeks later. The patient was admit­ted for inten­si­
fi­ca­tion 2, includ­ing full-dose mitoxantrone with dexrazoxane,
and her che­mo­ther­apy course was com­pleted with­out fur­ther
com­pli­ca­tions. The patient is cur­rently 1 year fol­low­ing ther­apy
and is being mon­i­tored with annual echo­car­dio­grams due to
high-dose anthracycline expo­sure and a his­tory of on-ther­apy
LVSD.
Prakash Singh Shekhawat
Hari K. Narayan: Off-label use of several drugs is discussed:
angiotensin converting enzyme inhibitors (ex. enalapril); beta
blockers (ex. Carvedilol); asparaginase dexrazoxane (given under
FDA orphan drug designation for pediatric cancers, but not tech-
nically a labeled indication).
Kelly Getz: Off-label use of several drugs is discussed: angioten-
sin converting enzyme inhibitors (ex. enalapril); beta blockers
(ex. Carvedilol); asparaginase dexrazoxane (given under FDA or-
phan drug designation for pediatric cancers, but not technically
a labeled indication).
Kasey J. Leger: Off-label use of several drugs is discussed: angio-
tensin converting enzyme inhibitors (ex. enalapril); beta blockers
(ex. Carvedilol); asparaginase dexrazoxane (given under FDA or-
phan drug designation for pediatric cancers, but not technically
a labeled indication).
Minimizing cardiotoxicity in pedi­at­ric AML  |  373
Correspondence
Kasey J. Leger, Department of Hematology/Oncology, Cancer
and Blood Disorders Center, Seattle Children’s Hospital, 4800
Sand Point Way NE, M/S MB8.501, Seattle, WA 98105; e-mail:
kasey​­.leger@seattlechildrens​­.org.
References
1. Aplenc R, Meshinchi S, Sung L, et al. Bortezomib with stan­dard che­mo­ther­
apy for chil­dren with acute mye­loid leu­ke­mia does not improve treat­ment
out­comes: a report from the Children’s Oncology Group. Haematologica.
2020;105(7):1879-1886.
2. Fernandez HF, Sun Z, Yao X, et  al. Anthracycline dose inten­si­fi­ca­tion in
acute mye­loid leu­ke­mia. N Engl J Med. 2009;361(13):1249-1259.
3. Feijen EAM, Leisenring WM, Stratton KL, et al. Derivation of anthracycline
and anthra­qui­none equiv­al­ence ratios to doxo­ru­bi­cin for late-onset cardio-
toxicity. JAMA Oncol. 2019;5(6):864-871.
4. Barlogis V, Auquier P, Bertrand Y, et al. Late car­dio­my­op­a­thy in child­hood
acute mye­loid leu­ke­mia sur­vi­vors: a study from the L.E.A. pro­gram. Hae-
matologica. 2015;100(5):e186-e189.
5. Getz KD, Sung L, Alonzo TA, et al. Effect of dexrazoxane on left ven­tric­u­
lar sys­tolic func­tion and treat­ment out­comes in patients with acute mye­
loid leu­ke­mia: a report from the Children’s Oncology Group. J Clin Oncol.
2020;38(21):2398-2406.
6. Getz KD, Sung L, Ky B, et al. Occurrence of treat­ment-related cardiotoxic-
ity and its impact on out­comes among chil­dren treated in the AAML0531
clin­i­cal trial: a report from the Children’s Oncology Group. J Clin Oncol.
2019;37(1):12-21.
7. Cvetković RS, Scott LJ. Dexrazoxane: a review of its use for cardioprotec-
tion dur­ing anthracycline che­mo­ther­apy. Drugs. 2005;65(7):1005-1024.
8. Lyu YL, Kerrigan JE, Lin CP, et al. Topoisomerase IIβ-medi­ated DNA dou­ble-
strand breaks: impli­ca­tions in doxo­ru­bi­cin cardiotoxicity and pre­ven­tion by
dexrazoxane. Cancer Res. 2007;67(18):8839-8846.
9. Schuchter LM, Hensley ML, Meropol NJ, Winer EP; Amer­i­can Society of Clin-
ical Oncology Chemotherapy and Radiotherapy Expert Panel. 2002 update
of rec­om­men­da­tions for the use of che­mo­ther­apy and radio­ther­apy pro­tec­
tants: clin­i­cal prac­tice guide­lines of the Amer­i­can Society of Clinical Oncol-
ogy. J Clin Oncol. 2002;20(12):2895-2903.
10. Vachhani P, Shin S, Baron J, Thompson JE, Wetzler M, Griffiths EA, Ontiveros
EP, Spangenthal EJ, Wang ES. Dexrazoxane for cardioprotection in older
adults with acute myeloid leukemia. Leuk Res Rep. 2017 Apr 14;7:36-39.
11. Harake D, Franco VI, Henkel JM, Miller TL, Lipshultz SE. Cardiotoxicity in
child­hood can­cer sur­vi­vors: strat­e­gies for pre­ven­tion and man­age­ment.
Future Cardiol. 2012;8(4):647-670.
12. Lipshultz SE, Sambatakos P, Maguire M, et al. Cardiotoxicity and cardiopro-
tection in child­hood can­cer. Acta Haematol. 2014;132(3-4):391-399.
13. Wexler LH. Ameliorating anthracycline cardiotoxicity in children with cancer:
clinical trials with dexrazoxane. Semin Oncol. 1998 Aug;25(4 Suppl 10):86-92.
14. Sánchez-Medina J, Gonzalez-Ramella O, Gallegos-Castorena S. The effect
of dexrazoxane for clin­i­cal and sub­clin­i­cal cardiotoxicity in chil­dren with
acute mye­loid leu­ke­mia. J Pediatr Hematol Oncol. 2010;32(4):294-297.
15. Schloemer NJ, Brickler M, Hoffmann R, et al. Administration of dexrazoxane
improves car­diac indi­ces in chil­dren and young adults with acute mye­loid
leu­ke­mia (AML) while maintaining sur­vival out­comes. J Pediatr Hematol
Oncol. 2017;39(5):e254–e258.
16. Tebbi CK, London WB, Friedman D, et al. Dexrazoxane-asso­ci­ated risk for
acute mye­loid leu­ke­mia/myelodysplastic syn­drome and other sec­ond­ary
malig­nan­cies in pedi­at­ric Hodgkin’s dis­ease. J Clin Oncol. 2007;25(5):493–
500.
17. Chow EJ, Asselin BL, Schwartz CL, et al. Late mor­tal­ity after dexrazoxane
treat­ ment: a report from the chil­ dren’s oncol­
ogy group. J Clin Oncol.
2015;33(24):2639-2645.
18. Schwartz CL, Constine LS, Villaluna D, et  al. A risk-adapted, response-
based approach using ABVE-PC for chil­dren and ado­les­cents with inter­
me­di­ate- and high-risk Hodgkin lym­phoma: the results of P9425. Blood.
2009;114(10):2051-2059.
19. Forssen EA, Coulter DM, Proffitt RT. Selective in vivo local­i­za­tion of dau­no­ru­
bi­cin small unilamellar ves­ic­ les in solid tumors. Cancer Res. 1992;52(12):3255–
3261.
20. Lowis S, Lewis I, Elsworth A, et al; United Kingdom Children’s Cancer
Study Group New Agents; Société Française d’Oncologie Pédiatrique
Pharmacology Group. A phase I study of intra­ve­nous lipo­so­mal dau­no­ru­
374  |  Hematology 2021  |  ASH Education Program
bi­cin (DaunoXome) in pae­di­at­ric patients with relapsed or resis­tant solid
tumours. Br J Cancer. 2006;95(5):571-580.
21. Bellott R, Auvrignon A, Leblanc T, et  al. Pharmacokinetics of lipo­so­mal
dau­ no­ru­bi­
cin (DaunoXome) dur­ ing a phase I-II study in chil­ dren with
relapsed acute lym­pho­blas­tic leu­kae­mia. Cancer Chemother Pharmacol.
2001;47(1):15-21.
22. Waterhouse DN, Tardi PG, Mayer LD, Bally MB. A com­par­i­son of lipo­so­mal
for­mu­la­tions of doxo­ru­bi­cin with drug admin­is­tered in free form: chang­ing
tox­ic­ity pro­files. Drug Saf. 2001;24(12):903-920.
23. Rafiyath SM, Rasul M, Lee B, Wei G, Lamba G, Liu D. Comparison of safety
and tox­ic­ity of lipo­so­mal doxo­ru­bi­cin vs. con­ven­tional anthracyclines: a
meta-anal­y­sis. Exp Hematol Oncol. 2012;1(1):10.
24. van Dalen EC, Michiels EM, Caron HN, Kremer LC. Different anthracycline
der­i­vates for reduc­ing cardiotoxicity in can­cer patients. Cochrane Data-
base Syst Rev. 2010;2010(5):CD005006.
25. Kaspers GJ, Zimmermann M, Reinhardt D, et al. Improved out­come in pedi­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/368/1851685/368narayan.pdf by guest on 13 December 2021
at­ric relapsed acute mye­loid leu­ke­mia: results of a ran­dom­ized trial on lipo­
so­mal dau­no­ru­bi­cin by the International BFM Study Group. J Clin Oncol.
2013;31(5):599-607.
26. Creutzig U, Zimmermann M, Bourquin JP, et al. Randomized trial com­par­
ing lipo­so­mal dau­no­ru­bi­cin with idarubicin as induc­tion for pedi­at­ric acute
mye­loid leu­ke­mia: results from Study AML-BFM 2004. Blood. 2013;122(1):
37-43.
27. Lancet JE, Uy GL, Cortes JE, et al. CPX-351 (cytarabine and dau­no­ru­bi­cin)
lipo­some for injec­tion ver­sus con­ven­tional cytarabine plus dau­no­ru­bi­cin in
older patients with newly diag­nosed sec­ond­ary acute mye­loid leu­ke­mia.
J Clin Oncol. 2018;36(26):2684-2692.
28. Cooper TM, Absalon MJ, Alonzo TA, et al. Phase I/II study of CPX-351 fol-
lowed by fludarabine, cytarabine, and granulocyte-col­ony stim­u­lat­ing fac­
tor for chil­dren with relapsed acute mye­loid leu­ke­mia: a report from the
chil­dren’s oncol­ogy group. J Clin Oncol. 2020;38(19):2170-2177.
29. Lopez L, Colan SD, Frommelt PC, et al. Recommendations for quan­ti­fi­ca­tion
meth­ods dur­ing the per­for­mance of a pedi­at­ric echo­car­dio­gram: a report
from the Pediatric Measurements Writing Group of the Amer­i­can Society
of Echocardiography Pediatric and Congenital Heart Disease Council.
J Am Soc Echocardiogr. 2010;23(5):465-495, quiz, 576-577.
30. Selamet Tierney ES, Hollenbeck-Pringle D, Lee CK, et al; Pediatric Heart
Network Investigators. Reproducibility of left ven­tric­u­lar dimen­sion ver­sus
area ver­sus vol­ume mea­sure­ments in pedi­at­ric patients with dilated car­
dio­my­op­a­thy. Circ Cardiovasc Imaging. 2017;10(11):e006007.
31. Leiner T, Bogaert J, Friedrich MG, et al. SCMR Position Paper (2020) on clin­
i­cal indi­ca­tions for car­dio­vas­cu­lar mag­netic res­o­nance. J Cardiovasc Magn
Reson. 2020;22(1):76.
32. Plana JC, Galderisi M, Barac A, et  al. Expert con­sen­sus for multimodality
imag­ing eval­ua ­ ­tion of adult patients dur­ing and after can­cer ther­apy: a
report from the Amer­ i­
can Society of Echocardiography and the Euro­
pean Association of Cardiovascular Imaging. J Am Soc Echocardiogr.
2014;27(9):911-939.
33. Armstrong GT, Plana JC, Zhang N, et al. Screening adult sur­vi­vors of child­
hood can­cer for car­dio­my­op­a­thy: com­par­i­son of echo­car­di­og­ra­phy and
car­diac mag­netic res­o­nance imag­ing. J Clin Oncol. 2012;30(23):2876-2884.
34. Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for car­diac cham­
ber quan­ti­fi­ca­tion by echo­car­di­og­ra­phy in adults: an update from the
Amer­i­can Society of Echocardiography and the Euro­pean Association of
Cardiovascular Imaging. J Am Soc Echocardiogr. 2015;28(1):1-39.e14.
35. Toro-Salazar OH, Ferranti J, Lorenzoni R, et  al. Feasibility of echo­car­dio­
graphic tech­niques to detect sub­clin­i­cal can­cer ther­a­peu­tics-related car­
diac dys­func­tion among high-dose patients when com­pared with car­diac
mag­netic res­o­nance imag­ing. J Am Soc Echocardiogr. 2016;29(2):119-131.
36. Lipshultz SE, Lipsitz SR, Sallan SE, et al. Long-term enalapril ther­apy for left
ven­tric­u­lar dys­func­tion in doxo­ru­bi­cin-treated sur­vi­vors of child­hood can­
cer. J Clin Oncol. 2002;20(23):4517-4522.
37. Oikonomou EK, Kokkinidis DG, Kampaktsis PN, et al. Assessment of prog­
nos­tic value of left ven­tric­u­lar global lon­gi­tu­di­nal strain for early pre­dic­tion
of che­mo­ther­apy-induced cardiotoxicity: a sys­tem­atic review and meta-
anal­y­sis. JAMA Cardiol. 2019;4(10):1007-1018.
38. Armstrong GT, Joshi VM, Ness KK, et al. Comprehensive echo­car­dio­graphic
detec­tion of treat­ment-related car­diac dys­func­tion in adult sur­vi­vors of
child­hood can­cer: results from the St. Jude Lifetime Cohort Study. J Am
Coll Cardiol. 2015;65(23):2511-2522.
39. Thavendiranathan P, Negishi T, Somerset E, et al; SUCCOUR Investigators.
Strain-guided man­age­ment of poten­tially cardiotoxic can­cer ther­apy. J Am
Coll Cardiol. 2021;77(4):392-401.
40. Zamorano JL, Lancellotti P, Rodriguez Muñoz D, et  al. 2016 ESC Position 43. Fratz S, Chung T, Greil GF, et al. Guidelines and pro­to­cols for car­dio­vas­cu­lar
Paper on can­cer treat­ments and car­dio­vas­cu­lar tox­ic­ity devel­oped under mag­netic res­o­nance in chil­dren and adults with con­gen­i­tal heart dis­ease:
the aus­pices of the ESC Committee for Practice Guidelines: the Task Force SCMR expert con­sen­sus group on con­gen­i­tal heart dis­ease. J Cardiovasc
for can­cer treat­ments and car­dio­vas­cu­lar tox­ic­ity of the Euro­pean Society Magn Reson. 2013;15(1):51.
of Cardiology (ESC). Eur Heart J. 2016;37(36):2768-2801.
41. Arme­ nian SH, Gelehrter SK, Chow EJ. Strategies to pre­ vent anthracy-
cline-related con­ges­tive heart fail­ure in sur­vi­vors of child­hood can­cer. Car-
diol Res Pract. 2012;2012(1):713294.
42. Levy PT, Machefsky A, Sanchez AA, et al. Reference ranges of left ven­tric­u­lar
strain mea­sures by two-dimen­sional speckle-track­ing echo­car­di­og­ra­phy in
chil­dren: a sys­tem­atic review and meta-anal­y­sis. J Am Soc Echocardiogr. © 2021 by The Amer­i­can Society of Hematology
2016;29(3):209-225.e6. DOI 10.1182/hema­tol­ogy.2021000268

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/368/1851685/368narayan.pdf by guest on 13 December 2021

Prakash Singh Shekhawat

Minimizing cardiotoxicity in pedi­at­ric AML  |  375
 IT TAKES A VILLAGE: MAXIMIZING SUPPORTIVE CARE AND MINIMIZING TOXICITY DURING CHILDHOOD LEUKEMIA THERAPY

Managing therapy-associated neurotoxicity

in children with ALL

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/376/1851679/376bhojwani.pdf by guest on 13 December 2021

Deepa Bhojwani,1 Ravi Bansal,2 and Alan S. Wayne1
Division of Hematology-Oncology, Children’s Hospital Los Angeles, Norris Comprehensive Cancer Center, Keck School of Medicine,
1

University of Southern California, Los Angeles, CA; and 2Division of Research on Children, Youth and Families, Children’s Hospital Los Angeles, Keck
School of Medicine, University of Southern California, Los Angeles, CA

Several chemotherapeutic agents and novel immunotherapies provide excellent control of systemic and central nervous
system (CNS) leukemia but can be highly neurotoxic. The manifestations of subacute methotrexate neurotoxicity are
diverse and require vigilant management; nonetheless, symptoms are transient in almost all patients. As methotrexate is
a crucial drug to prevent CNS relapse, it is important to aim to resume it after full neurologic recovery. Most children tol-
erate methotrexate rechallenge without significant delays or prophylactic medications. Neurotoxicity is more frequent
with newer immunotherapies such as CD19– chimeric antigen receptor T (CAR T) cells and blinatumomab. A uniform
grading system for immune effector cell–associated neurotoxicity syndrome (ICANS) and algorithms for management
based on severity have been developed. Low-grade ICANS usually resolves within a few days with supportive measures,
but severe ICANS requires multispecialty care in the intensive care unit for life-threatening seizures and cerebral edema.
Pharmacologic interventions include anticonvulsants for seizure control and glucocorticoids to reduce neuroinflamma-
tion. Anticytokine therapies targeted to the pathophysiology of ICANS are in development. By using illustrative patient
cases, we discuss the management of neurotoxicity from methotrexate, CAR T cells, and blinatumomab in this review.

LEARNING OBJECTIVES
• Discuss management of methotrexate neurotoxicity and rechallenge strategy
• Review assessment and management of ICANS associated with CAR T-cell therapy and blinatumomab

Introduction predictors and mediators of immune effector cell-associated

Optimization of chemotherapeutic regimens has enabled
substitution of cranial irradiation with systemic and intrathe-
cal agents for prophylaxis of central nervous system (CNS)
disease in childhood acute lymphoblastic leukemia (ALL).1
However, even radiation-free contemporary regimens can
cause severe neurotoxicity, leading to morbidity and mor-
tality, therapy delays, and fear in patients and families. The
incidence of symptomatic neurotoxicity during frontline ALL
therapy is approximately 10% to 12%, and its manifestations
are varied, as are the inciting agents.2,3 Methotrexate-related
stroke-like syndrome (SLS), posterior reversible encepha-
lopathy syndrome, cerebral sinus venous thrombosis, and
seizures are the common CNS events that result from stan-
dard ALL chemotherapeutic agents.4 With increasing use
of immunotherapies for patients with ALL, neurotoxicity is
being reported more frequently, with incidence of up to
50% in patients treated with CD19- chimeric antigen recep-
tor T (CAR T)-cell therapy.5 Multiple studies are investigating
neurotoxicity syndrome (ICANS), with the goal of identify-
ing those at high risk and developing interventions to miti-
gate its severe manifestations. In this review, we discuss the
management of neurotoxicity associated with methotrexate,
CAR T-cell therapy, and blinatumomab in children with ALL.
CLINICAL CASE 1
A 14-year-old Hispanic girl with high-risk B-ALL without CNS
involvement was receiving standard-of-care chemother-
apy according to the Children’s Oncology Group proto-
col AALL0232. She had tolerated chemotherapy relatively
well, including 6 doses of intrathecal methotrexate in the
induction and consolidation phases. The first interim main-
tenance phase was initiated with 15mg of intrathecal meth-
otrexate followed by 5g/m2 of intravenous methotrexate
over 24 hours. Methotrexate cleared appropriately, and she

376 | Hematology 2021 | ASH Education Program


Figure 1. Diffusion-weighted MRI of patient 1 four hours after
onset of symptoms. DWI demonstrated abnormal diffusion in bi-
lateral centrum semiovale (red arrows), greater on the left side.
These findings indicate increased diffusion of water in the affect-
ed regions suggestive of cytotoxic edema.
received 3 doses of leucovorin 15 mg/m2 at hours 42, 48, and 54.
Nine days later, she presented to the emer­gency depart­ment
with acute-onset right-sided leg weak­ness, which progressed to
the right arm and face over a 2-hour period. In addi­tion to right
paraparesis and facial droop, she was noted to have apha­sia and
dys­ar­thria. Magnetic res­o­nance imag­ing (MRI) performed 4 hours
after the onset of symp­toms showed restricted dif­fu­sion bilat­er­
ally in the cen­trum semiovale, greater on the left side (Figure 1).
She was admit­ted to the inten­sive care unit for close obser­va­
tion. During the next 24 hours, her speech wors­ened, and she
became more con­fused. There was mod­er­ate improve­ment in
right-sided weak­ness, but she devel­oped weak­ness in the left arm
and leg. With con­tin­ued sup­port­ive care, begin­ning day 3 of hos­
pi­tal­i­za­tion, her symp­toms grad­u­ally improved, facial asymmetry
resolved, her speech became clearer, and she was a ­ ble to swal­
low. She required phys­i­cal ther­apy for 2 more days to regain her
base­line strength and ambu­late inde­pen­dently.
Subacute meth­o­trex­ate neu­ro­tox­ic­ity
Approximately 3% to 7% of chil­dren treated with con­tem­po­rary
front­line ALL reg­i­mens develop sub­acute meth­o­trex­ate neu­ro­
tox­ic­ity.6-8 Seizures, enceph­a­lop­a­thy, apha­sia, and SLS are com­
mon pre­sen­ta­tions.7-9 Symptoms typ­i­cally begin 3 to 11 days after
intra­the­cal or intra­ve­nous meth­o­trex­ate, can wax and wane over
sev­eral hours or days, and are char­ac­ter­is­ti­cally tran­sient and
resolve in 1 to 4 days. Most patients make a full recov­ery, although
recov­ery may require sev­eral months of inten­sive reha­bil­i­ta­tion in
a small sub­set of patients with severe SLS.9 The ­defin­ing char­
ac­ter­is­tics of meth­o­trex­ate-SLS are listed in Table 1.10 In brief,
­meth­o­trex­ate-SLS is defined as neu­ro­tox­ic­ity that occurs within
Prakash Singh Shekhawat
21 days of intra­ve­nous or intra­the­cal meth­o­trex­ate with spec­i­fied
clin­i­cal and radio­graphic fea­tures, while exclud­ing other iden­ti­fi­
able causes. Risk fac­tors for meth­o­trex­ate neu­ro­tox­ic­ity ­iden­ti­fied
in mul­ti­var­i­able ana­ly­ses are age >10 years, His­panic eth­nic­ity,
and ele­vated serum aspar­tate trans­am­i­nase dur­ing induc­tion/
con­sol­i­da­tion.8,11 Coadministration of intra­the­cal meth­o­trex­ate in
sys­temic cytarabine- and cyclo­phos­pha­mide-containing treat­
ment blocks has been noted as a period of increased risk.8,9
Leukoencephalopathy or sub­cor­ti­cal white mat­ter changes
are the char­ac­ter­is­tic MRI find­ings of meth­o­trex­ate neu­ro­tox­
ic­
ity. Diffusion-weighted imag­ ing (DWI) is the most sen­ si­ tive
modal­ity dur­ing the acute phase and shows abnor­mally high
dif­fu­sion of water in brain areas that cor­re­spond with clin­i­cally
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/376/1851679/376bhojwani.pdf by guest on 13 December 2021
appar­ent neu­ro­logic def­i­cits.12 Because changes noted on DWI
can be tran­sient, a neg­a­tive MRI does not exclude a diag­no­sis
of meth­o­trex­ate neu­ro­tox­ic­ity. On the other hand, white mat­ter
hyperintensities in T2-weighted and fluid-attenuated inversion
recovery MRIs can per­sist and evolve over time; there­fore, await-
ing com­plete nor­mal­i­za­tion of MRI find­ings prior to rechallenge
with meth­o­trex­ate is not prac­ti­cal.
The path­o­phys­i­­ol­ogy of sub­acute meth­o­trex­ate neu­ro­tox­ic­
ity is not com­pletely under­stood. Accumulation of homocyste-
ine and its metab­o­lites in plasma and cere­bro­spi­nal fluid (CSF) is
toxic to vas­cu­lar endo­the­lium. Release of aden­o­sine and neu­ro­
nal injury are also pro­posed mech­a­nisms.13
Acute man­age­ment of meth­o­trex­ate neu­ro­tox­ic­ity
Supportive care and close mon­i­tor­ing are the main­stays of man­
age­ment dur­ing the acute epi­sode. A sin­gle, brief sei­zure will
not require phar­ma­co­logic inter­ven­tion, but anti­con­vul­sants are
admin­is­tered for prolonged or repeated sei­zures. Symptoms can
wax and wane; there­fore, fre­quent neu­ro­logic exam­i­na­tions
guide sub­se­quent man­age­ment. Oral feeds should be held until
swallowing is deemed safe. Because apha­sia and dys­ar­thria are
distressing to patients and fam­i­lies, psy­cho­so­cial sup­port, reas­
sur­ance, and pro­vid­ing alter­nate meth­ods of com­mu­ni­ca­tion may
lessen anx­i­ety. Computerized tomog­ra­phy (CT) is not gen­er­ally
help­ful but may be needed if symp­toms are sug­ges­tive of CNS
bleed. CSF exam­i­na­tion is only indi­cated if the patient is febrile
and there are con­cerns for CNS infec­tion. As detailed above, MRI
with DWI can sup­port the diag­no­sis of meth­o­trex­ate neu­ro­tox­
ic­ity, but its util­ity in cases of mild, tran­sient, and clas­si­cal symp­
toms is lim­ited. Input from a neu­rol­o­gist is valu­able in most cases,
par­tic­u­larly with prolonged symp­toms. Severe SLS is best man­
aged in a crit­i­cal care set­ting, as patients may require intu­ba­tion
with mechan­i­cal ven­ti­la­tion for air­way pro­tec­tion.
Based on the pre­sumed mech­a­nisms of meth­o­trex­ate neu­ro­
tox­ic­ity and evi­dence of homocysteine caus­ing excit­atory effects
on N-methyl-D-aspar­tate recep­tors, the use of the N-methyl-D-
aspar­ tate antag­ o­nist dextromethorphan has been reported in
the acute set­ting and for sec­ond­ary pro­phy­laxis.14 To com­bat
vaso­di­la­ta­tion from aden­o­sine, ami­noph­yl­line use has also been
described, as it dis­places aden­o­sine from its recep­tor on endo­
the­lial cells.15 However, supporting such approaches is dif­fi­cult
due to a lack of con­trols, par­tic­u­larly in the set­ting of the tran­sient
nat­u­ral course of meth­o­trex­ate neu­ro­tox­ic­ity and rapid res­o­lu­
tion of symp­toms with­out phar­ma­ceu­ti­cal inter­ven­tion in most
patients. In addi­ tion, dextromethorphan itself can cause CNS
effects such as diz­zi­ness and rest­less­ness, whereas ami­noph­yl­line
Neurotoxicity in chil­dren with ALL  |  377
Table 1. Defining char­ac­ter­is­tics of meth­o­trex­ate-SLS (reproduced with per­mis­sion10)

Neurotoxicity occur­ring within 21 days of intra­ve­nous or intra­the­cal meth­o­trex­ate with 3 char­ac­ter­is­tics that all­need to be fulfilled:
  1. New onset of one or more of pare­sis or paral­y­sis; move­ment dis­or­der or bilat­eral weak­ness; apha­sia or dys­ar­thria; altered men­tal sta­tus includ­ing
con­scious­ness (e.g., som­no­lence, con­fu­sion, dis­ori­en­ta­tion, and emo­tional labil­ity); and/or sei­zures with at least one of the other symp­toms.
  2. Either char­ac­ter­is­tic, but often tran­sient, white mat­ter changes indi­cat­ing leukoencephalopathy on MRI or a char­ac­ter­is­tic clin­i­cal course with
wax­ing and wan­ing symp­toms usu­ally lead­ing to com­plete (some­times par­tial) res­o­lu­tion within a week.
  3.  No other iden­ti­fi­able cause.
Characteristic oval-shaped lesions of the sub­cor­ti­cal white mat­ter (mostly fron­tal or pari­e­tal) on MRI are best seen on dif­fu­sion-weighted (hyper­in­
tense) or appar­ent dif­fu­sion coef­fi­cient (hypointense) images. Can be graded 1-5 according to CTCAEv4.03 for enceph­a­lop­a­thy.
This con­sen­sus def­i­ni­tion was devel­oped by the Ponte de Legno inter­na­tional child­hood ALL group to assist cli­ni­cians in dif­fer­en­ti­at­ing meth­o­trex­ate
-SLS from sim­i­lar neurotoxicities (eg, pos­te­rior revers­ible enceph­a­lop­a­thy syn­drome) and to enable reli­able com­par­i­sons of inci­dence of meth­o­trex­ate
-SLS between dif­fer­ent tri­als.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/376/1851679/376bhojwani.pdf by guest on 13 December 2021

can cause tachy­car­dia, anx­i­ety, head­ache, and sei­zures.16 Current
gaps in knowl­edge regard­ing the util­ity of dextromethorphan
and ami­noph­yl­line for the treat­ment and pro­phy­laxis of meth­o­
trex­ate neu­ro­tox­ic­ity pre­clude spe­cific rec­om­men­da­tions.
Rechallenge with meth­o­trex­ate after the first epi­sode
of neu­ro­tox­ic­ity
Multiple stud­ies have val­i­dated the safety of rechallenge with
meth­o­trex­ate after the first epi­sode of neu­ro­tox­ic­ity, as 82% to
92% of patients did not develop recur­rent symp­toms.6-9 Impor-
tantly, omis­sion of meth­o­trex­ate for CNS-directed ther­apy after a
neu­ro­toxic epi­sode increased the risk of CNS relapse, as dem­on­
strated in a large Aus­tra­lian cohort of 1251 pedi­at­ric patients with
ALL, includ­ing 95 who expe­ri­enced meth­o­trex­ate neu­ro­tox­ic­ity.8
The 5-year CNS relapse-free sur­vival was 89% when intra­the­cal
meth­o­trex­ate was per­ma­nently discontinued and replaced with
alter­nate intra­the­cal ther­apy after a symp­tom­atic neu­ro­tox­ic­ity
epi­sode, com­pared with 95% if patients con­tin­ued to receive
pro­to­col-pre­scribed doses of intra­the­cal meth­o­trex­ate (P = .047).
These data high­light the impor­tance for the con­tin­u­a­tion of opti­
mal ther­apy to max­i­mize cure.
Rechallenge with meth­o­trex­ate should be delayed until full
neu­ro­logic recov­ery, which is typ­i­cally less than 7 days. For some
patients, this may result in the omis­sion or delay of meth­o­trex­ate
ther­apy by 1 to 2 weeks. Successful resump­tion of stan­dard ther­
apy is pos­si­ble in most patients, with­out the need for exces­sive
delays, sub­sti­tu­tions, or pro­phy­lac­tic med­i­ca­tions. Substitution of
intra­the­cal meth­o­trex­ate by cytarabine with or with­out hydro­cor­
ti­sone can be con­sid­ered tem­po­rar­ily for 1 or 2 doses, with the
resump­tion of intra­the­cal meth­o­trex­ate as soon as pos­si­ble. Leu-
covorin res­cue of 2 sys­temic doses (5 mg/m2/dose) at hours 24
and 30 after intra­the­cal meth­o­trex­ate rechallenge is a rea­son­able
strat­egy to attempt to min­i­mize recur­rent epi­sodes. For sys­temic
meth­ot­rex­ate rechallenge, early leucovorin res­cue at hour 30 or
36 after the start of the 24-hour infu­sion of high-dose meth­o­trex­
ate could be con­sid­ered. Although there are no ran­dom­ized stud­
ies for this sec­ond­ary pro­phy­laxis, leucovorin is the only drug with
strong clin­ic­ al evi­dence for reduc­ing the inci­dence of meth­o­trex­
ate neu­ro­tox­ic­ity.17-19 The the­o­ret­i­cal con­cern of “res­cue” of intra­
the­cal meth­o­trex­ate by leucovorin and con­se­quent reduc­tion of
its anti­leu­ke­mic effect is off­set by the abil­ity to con­tinue admin­is­
tra­tion of this impor­tant drug to reduce risk of relapse. Additional
leucovorin res­cue may not be needed indef­i­nitely, espe­cially if
the patient tol­er­ates the sub­se­quent rechallenge well.
CLINICAL CASE 1 (Con­tin­ued)
This patient made a full recov­ery 5 days after the onset of symp­
toms. Her sec­ond course of intra­ve­nous high-dose meth­o­trex­
ate was delayed by 2 weeks, dur­ing which she con­tin­ued to
receive mer­cap­to­pu­rine. Leucovorin res­cue was ini­ti­ated early
at hour 30 and given every 6 hours for 5 doses dur­ing courses
2 and 3. As she did not develop recur­rent neu­ro­tox­ic­ity, the last
course of high-dose meth­o­trex­ate was admin­is­tered with the
stan­dard leucovorin res­cue. She com­pleted all­planned ther­apy
for ALL with­out fur­ther com­pli­ca­tions.
Recurrence of meth­o­trex­ate neu­ro­tox­ic­ity
Due to the rar­ity of recur­rences, there are no avail­­able guide­
lines for a sec­ond rechallenge. Patients with recur­rent brief sei­
zures may tol­er­ate sub­se­quent meth­o­trex­ate under the cover of
appro­pri­ate anti­con­vul­sants for the remain­der of ther­apy. Very
lim­ited data are avail­­able to assess out­comes of sec­ond rechal-
lenge after recur­rent SLS. In the UKALL 2003 report, 4 patients
with a sec­ond epi­sode of SLS con­tin­ued to receive meth­o­trex­ate
after com­plete neu­ro­logic recov­ery.9 Of these, 3 patients tol­er­
ated meth­o­trex­ate well, and 1 devel­oped per­sis­tent neu­ro­logic
def­i­cits, suggesting that a sub­set of patients may con­tinue to
receive meth­o­trex­ate safely even after 2 epi­sodes of neu­ro­tox­
ic­ity. However, cau­tion is warranted as there are case reports of
patients with poor neu­ro­logic out­comes and rare fatal­i­ties after
repeated epi­sodes of meth­o­trex­ate neu­ro­tox­ic­ity.9,20 In these
uncom­mon sce­nar­ios, par­tic­u­larly when patients do not make
a full recov­ery in the expected time frame or have recur­rent
severe meth­o­trex­ate neu­ro­tox­ic­ity, com­ple­tion of CNS-directed
ther­apy with alter­nate agents would require an indi­vid­u­al­ized
approach based on the patient’s risk of CNS relapse.
CLINICAL CASE 2
The patient is a 16-year-old Cau­ca­sian male with B-ALL in sec­ond
relapse with high bone mar­row leu­ke­mia bur­den. He received
levetiracetam for sei­zure pro­phy­laxis and fludarabine and cyclo­
phos­pha­mide for lymphodepletion prior to infu­sion of autol­o­
gous anti-CD19 CAR T cells. On postinfusion day 3, he devel­oped
grade 2 cyto­kine release syn­drome (CRS) manifested by fevers,

378  |  Hematology 2021  |  ASH Education Program

hyp­oxia, and hypo­ten­sion.21 Initial man­ age­ment included oxy­
gen sup­ple­men­ta­tion, intra­ve­nous fluid bolus, tocilizumab, and
meth­yl­pred­nis­o­lone. After tran­sient improve­ment, grade 2 CRS
recurred with wors­en­ing hypo­ten­sion on day 5, requir­ing dopa­
mine infu­sion. A sec­ond dose of tocilizumab was admin­is­tered for
CRS. On day 6, he was lethar­gic, with expres­sive dys­pha­sia and
an immune effec­tor cell–asso­ci­ated enceph­a­lop­a­thy (ICE) score
of 5, meet­ing cri­te­ria for grade 2 ICANS.21 Dexamethasone was ini­
ti­ated for ICANS, and CT of the head was performed, which was
nor­mal. However, his neu­ro­logic sta­tus rap­idly declined through­
out the day, and he devel­oped a gen­er­al­ized tonic-clonic sei­zure
that abated after 2 doses of lorazepam. He was given a load­ing
dose of levetiracetam and trans­ferred to the inten­sive care unit.
He was stu­por­ous on exam­i­na­tion and required intu­ba­tion for
apneic epi­sodes. ICANS had progressed to grade 4. Aggressive
neuroprotective mea­sures were insti­tuted with ele­va­tion of head
of bed and sup­port­ive inter­ven­tions to main­tain normotension
and nor­mo­ther­mia. Anakinra and high-dose meth­yl­pred­nis­o­lone
were ini­ti­ated.
ICANS
Accurate assess­ments of the inci­dence and sever­ity of neu­ro­
tox­ic­ity were lim­ited by dif­fer­ing def­i­ni­tions and grad­ing cri­te­ria
used in ear­lier CAR T stud­ies. Therefore, the Amer­i­can Society
of Transplantation and Cellular Therapy (ASTCT) published con­
sen­ sus def­i­
ni­
tions and grad­ ing in 2019 for uni­ for­
mity across
Figure 2. ASTCT consensus grading for ICANS. Adapted from Lee et al21 and Traube et al44 with permission. ICANS grading (blue box)
has 5 elements. The first element assesses encephalopathy and uses a scoring tool: ICE (green box) for patients ≥12 years and CAPD
(orange box) for patients <12 years of age. CAPD, Cornell Assessment of Pediatric Delirium; ICE, immune effector cell–associated
encephalopathy; ICP, increased intracranial pressure.
Prakash Singh Shekhawat
stud­ies (Figure 2).21 ICANS is defined as “a dis­or­der char­ac­ter­
ized by a path­o­logic pro­cess involv­ing the CNS fol­low­ing any
immune ther­apy that results in the acti­va­tion or engage­ment
of endog­e­nous or infused T cells and/or other immune effec­tor
cells.”21 With CD19 CAR T ther­apy, neu­ro­tox­ic­ity of all­grades
has been reported in 28% to 64% of patients and grades ≥3
in 13% to 50%.5 Fatal ICANS was reported in approx­i­ma­tely 3%
of adult patients.22 ICANS begins in the first week postinfusion
(median 4-6 days) in 90% of patients, soon after the onset of
CRS or as CRS symp­toms sub­side. Cases of ICANS with­out CRS
are uncom­mon and have less severe man­i­fes­ta­tions. Typically,
neu­ro­logic symp­toms are tran­sient and resolve in 5 to 11 days
but can be noted for up to 8 weeks.5,23 Delayed-onset ICANS
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/376/1851679/376bhojwani.pdf by guest on 13 December 2021
at 1 month after infu­sion has also been described.24 Early signs
of ICANS include expres­sive dys­pha­sia, tremor, and dysgraphia
and can prog­ress to sei­zures, depressed level of con­scious­
ness, enceph­a­lop­a­thy, and coma. Cerebral edema is a dreaded
con­se­quence of ICANS. Baseline risk fac­tors for ICANS include
high bone mar­row dis­ease bur­den and preexisting neu­ro­logic
def­i­cits, whereas postinfusion risk fac­tors include higher-grade
CRS, inflam­ma­tory mark­ers such as fever and increased fer­ri­
tin, and high serum lev­els of spe­cific cyto­kines (eg, inter­leu­
kin [IL]–2, IL-6, IL-10, IL-15).22,23,25 Although mech­a­nisms lead­ing
to ICANS are unclear, exper­i­men­tal mod­els sug­gest cyto­kine-
induced neuroinflammation and endo­the­lial acti­va­tion, with
sub­se­ quent dis­ rup­ tion of the blood-brain bar­ rier and direct
neu­ro­nal injury.22,25
Neurotoxicity in chil­dren with ALL  |  379
Acute man­age­ment of ICANS
Commercially avail­­able CAR T-cell prod­ucts in the United States
are governed under the Food and Drug Administration Risk Eval-
uation and Mitigation Strategy drug safety pro­gram to ensure
that health care pro­vid­ers and rel­e­vant staff are trained in the
rec­og­ni­tion and man­age­ment of toxicities asso­ci­ated with CAR T
ther­apy, includ­ing ICANS. With broader expe­ri­ence gained in the
man­age­ment of ICANS, sev­eral clin­i­cal man­age­ment guide­lines
have been published in the past 4 years.26-28 General rec­om­men­
da­tions are sum­ma­rized in Figure 3.
Although pro­phy­lac­tic anti­con­vul­sants are not used uni­
formly due to lim­ited evi­dence for pre­ven­tion of ICANS-related
sei­zures, they are recommended in patients with a his­tory of
sei­zures or other neu­ro­logic comorbidities. Close clin­i­cal mon­
i­
tor­ing and atten­ tive sup­port­
ive care are often suf­fi­
cient for
grade 1 ICANS. Tocilizumab (anti–IL-6 recep­tor anti­body) is only
indi­cated for con­cur­rent CRS, not for treat­ment of ICANS. Due
to its large size (molec­u­lar weight 149 kDA), tocilizumab does
not cross the blood-brain bar­rier and may also facil­i­tate a par­a­
dox­ic­ al shift of IL-6 from the blood to the brain.22 Neuroimaging
pref­er­a­bly with MRI, elec­tro­en­ceph­a­lo­gram (EEG), and lum­bar
punc­ture with open­ing pres­sure can sup­port the diag­no­sis of
ICANS and rule out other causes of neu­ro­logic symp­toms such
as infec­tion, CNS leu­ke­mia. or hem­or­rhage. MRI changes of focal
or dif­fuse cere­bral edema guide ICANS grad­ing and man­age­
ment. Leptomeningeal enhance­ment and mul­ti­fo­cal microhem-
Figure 3. General guidelines for management of ICANS. Frequent neurologic assessments, multidisciplinary supportive care, and glu-
cocorticoids form the backbone of ICANS management. In addition, several novel strategies are being studied for refractory ICANS,
as described in the text. ATG, anti–thymocyte globulin; CRP, C-reactive protein; EVD, external ventricular drain; ICU, intensive care
unit; LP, lumbar puncture; NPO, nothing by mouth.
380  |  Hematology 2021  |  ASH Education Program
orrhages have also been observed with severe ICANS.22 In an
study of adult patients, EEG showed dif­fuse slowing in 76% of
patients dur­ing acute ICANS and was also help­ful in detecting
sub­clin­i­cal sei­zures.22 Characteristic CSF find­ings for ICANS are
mod­er­ate leu­ko­cy­to­sis with lym­pho­cyte pre­dom­i­nance and
mark­edly ele­vated pro­tein.22
For grade 2 ICANS, glu­co­cor­ti­coids (dexa­meth­a­sone 0.2 mg/
kg/dose, max­i­mum 10 mg every 6 hours, or meth­yl­pred­nis­o­lone
equiv­al­ents) should be strongly con­sid­ered as brief courses do
not neg­at­ively affect the effi­cacy of the CAR T cells and may
shorten the dura­ tion of symp­ toms.29 For higher-grade ICANS
3 or 4, along with con­tin­ued sup­port­ive care in col­lab­o­ra­tion
with intensivists, neu­rol­o­gists, and neu­ro­sur­geons, the dose of
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/376/1851679/376bhojwani.pdf by guest on 13 December 2021
glu­co­cor­ti­coids should be increased (meth­yl­pred­nis­o­lone 10 or
20 mg/kg/dose, max­i­mum 500 or 1000 mg every 12 hours) until
improve­ment, followed by rapid taper. In addi­tion, mul­ti­ple anti­
con­vul­sants for opti­mal sei­zure con­trol, air­way pro­tec­tion by
intu­ba­tion and mechan­i­cal ven­ti­la­tion, and decreas­ing intra­cra­
nial pres­sure with hyperosmolar ther­apy or neu­ro­sur­gi­cal inter­
ven­tion may be warranted.
There are very lim­ited clin­i­cal data on other phar­ma­co­logic
inter­ven­tions targeted to the path­o­phys­i­­ol­ogy of ICANS. Siltux-
imab binds to IL-6 (unlike tocilizumab that binds to the IL-6 recep­
tor) and facil­i­tates removal of IL-6 from the cir­cu­la­tion, leav­ing
less IL-6 to enter the brain.27 Anakinra, the IL-1 recep­tor antag­o­
nist, may also be of ben­e­fit as mono­cyte-derived IL-1 can pre­cede
and induce IL-6 secre­tion.30,31 The molec­u­lar weight of anakinra
is 17.3 kDA, and ther­a­peu­tic con­cen­tra­tions can be achieved in
the CSF.32 Another potential approach to reduce neuroinflamma-
tory cyto­kine pro­duc­tion by mono­cytes is the use of lenzilumab,
the anti–granulocyte-mac­ro­phage col­ony-stim­u­lat­ing fac­tor
mono­clo­nal anti­body. Lenzilumab reduced neuroinflammation
after CAR T-cells in pre­clin­i­cal mod­els.33 Clinical tri­als of these
agents as pro­ phy­lac­
tic strat­e­
gies for ICANS are ongo­ ing
(NCT04359784, NCT04150913, NCT04432506, NCT04148430,
NCT04205838, NCT04314843). Dasatinib, a tyro­ sine kinase
inhib­i­tor, reduces CAR T-cell sig­nal­ing revers­ibly by inter­fer­ing
with lym­pho­cyte-spe­cific pro­tein tyro­sine kinase.34 In a murine
model, the dose of dasatinib was titrated to achieve par­tial or
com­plete inhi­bi­tion of CAR T-cell func­tion.34 This strat­egy of mod­
u­la­tion of CAR T-cell func­tion with dasatinib is being stud­ied as
an inter­ven­tion for CRS and ICANS in an ongo­ing clin­i­cal trial
(NCT04603872). The use of intra­the­cal hydro­cor­ti­sone, meth­
o­trex­ate, and cytarabine in ste­roid-refrac­tory ICANS has been
described in case reports.35,36 Additional stud­ies are needed to
inves­ti­gate if such attempts to reduce neuroinflammation by
intra­the­cal ther­apy and elim­i­nate immune effec­tor cells in the
CSF with cyto­toxic agents will be ben­e­fi­cial in the man­age­ment
of ICANS with­out neg­a­tively affect­ing CNS leu­ke­mia con­trol by
CAR T cells. However, in severe life-threat­ en­
ing ICANS, elim­ i­
na­tion of the CAR T cells is delib­er­ate, for exam­ple, with anti–
thy­mo­cyte glob­u­lin36 or cyclo­phos­pha­mide or by acti­vat­ing CAR
T-cell sui­cide approaches designed into spe­cific con­structs.37
CLINICAL CASE 2 (Con­t in­u ed)
With con­tin­ued aggres­sive neurocritical care and mul­ti­modal
ther­apy, his clin­i­cal con­di­tion began to improve on day 9, evi-
denced by spon­ta­ne­ous move­ments and with­drawal to pain.
No fur­ther sei­zure activ­ity was noted on EEG. Methylpredniso-
lone was tapered day 10 onward, and he was extubated on day
11. He con­tin­ued to have resid­ual apha­sia that grad­u­ally recov­
ered to base­line by day 15.
Table 2. Summary of man­i­fes­ta­tions and man­age­ment of neu­ro­tox­ic­ity from meth­o­trex­ate, CAR T cells, and blinatumomab
Characteristic Methotrexate
Frequency of neu­ro­tox­ic­ity 3%-7%
Risk fac­tors Older age, His­panic eth­nic­ity,
his­tory of ele­vated trans­am­i­nases
Time of onset from ther­apy 3-11 days
ini­ti­a­tion
Common signs/symp­toms Seizures, enceph­a­lop­a­thy, SLS
Average time to res­o­lu­tion 1-4 days
MRI find­ings Leukoencephalopathy
Management Supportive care
Prakash Singh Shekhawat
CLINICAL CASE 3
A 9-year old His­panic girl with B-ALL in first relapse had min­i­mal
resid­ual dis­ease 0.2% after reinduction che­mo­ther­apy. Blinatu-
momab con­tin­u­ous infu­sion 15 µg/kg/d was ini­ti­ated, and she
was mon­i­tored inpa­tient. On day 4, she complained of head­ache
that improved with acet­amin­o­phen. Three hours later, she devel­
oped acute-onset con­fu­sion and delir­ium. She was agi­tated and
moaning, she could not fol­low com­mands, and her speech was
inco­her­ent. She was assessed to have grade 2 enceph­a­lop­a­thy.
Blinatumomab infu­sion was imme­di­ately stopped, and dexa­
meth­a­sone was admin­is­tered. Within an hour of these inter­ven­
tions, her symp­toms began to improve, and 6 hours later, her
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/376/1851679/376bhojwani.pdf by guest on 13 December 2021
neu­ro­logic sta­tus was back to nor­mal. Because symp­toms did
not recur, blinatumomab infu­sion was restarted after 3 days at
a lower dose of 5 µg/kg/d, along with dexa­meth­a­sone premed-
ication. The dose of blinatumomab was esca­lated to 15 µg/kg/d
a week later, and she tol­er­ated the rest of the course well.
Neurotoxicity after blinatumomab
Similar prin­ci­ples of grad­ing and man­age­ment of ICANS from
CAR T apply to neu­ro­tox­ic­ity from blinatumomab. All clin­i­cal tri­
als to date have reported grades of indi­vid­ual neu­ro­toxic events
based on the Common Terminology Criteria for Adverse Events
(CTCAE), but future stud­ies will likely fol­low the ASTCT ICANS
guide­lines. The inci­dence of blinatumomab-asso­ci­ated neu­ro­
tox­ic­ity was 54% in adult patients with relapsed/refrac­tory ALL38
and 24% in a sim­i­lar pedi­at­ric cohort.39 Of all­neu­ro­logic events,
10% to 17% were sei­zures. In adult patients, symp­toms began at
a median of 9 days and also included tremor, som­no­lence, diz­zi­
ness, con­fu­sion, and enceph­a­lop­a­thy.38 Median time to onset of
neu­ro­tox­ic­ity may be ear­lier in pedi­at­ric patients.39
Blinatumomab has a very short elim­i­na­tion half-life (1.25
hours), and most toxicities resolve after interrupting the infu­sion
and ini­ti­at­ing dexa­meth­a­sone.38,39 Blinatumomab infu­sion should
be stopped for any sei­zure and for grade 3 or 4 nonseizure neu­ro­
tox­ic­ity.40 Prompt ini­ti­a­tion of dexa­meth­a­sone (0.2-0.4 mg/kg/d,
max­i­mum 24 mg, in 3 divided doses for 1-3 days) is recommended
CAR T cells Blinatumomab
28%-64% 24%-54%
High dis­ease bur­den, high-grade Older age, non­white race, prior
CRS, preexisting neu­ro­logic neu­ro­logic events, >2 prior
def­i­cits sal­vage ther­a­pies
4-6 days Median 9 days in adult patients,
may be ear­lier in pedi­at­ric
patients
Seizures, dys­pha­sia, enceph­a­lop­a­thy, Seizures, tremor, som­no­lence,
cere­bral edema enceph­a­lop­a­thy
5-11 days 1-5 days
Cerebral edema, leptomeningeal Insufficient data avail­­able
enhance­ment, mul­ti­fo­cal
microhemorrhages
Neurocritical care, ste­roids, anti­con­ Steroids, inter­rup­tion of
vul­sants, anticytokine ther­a­pies blinatumomab infu­sion
Neurotoxicity in chil­dren with ALL  |  381
for any severe neu­ro­tox­ic­ity. Blinatumomab is per­ma­nently dis-
continued if the patient devel­ops more than 1 sei­zure, any grade 4
neu­ro­tox­ic­ity, or if neu­ro­logic recov­ery is delayed > 7 days. For all­
other patients, blinatumomab can be restarted at one-third of the
dose (ie, 5 µg/kg, max­i­mum 9 µg) with dexa­meth­a­sone premedi-
cation (5 mg/m2, max­i­mum 20 mg) after improve­ment to grade ≤ 1
and a min­im ­ um of 3 days. If neu­ro­tox­ic­ity does not recur, blinatu-
momab can be esca­lated to full dose after 7 days. Secondary sei­
zure pro­phy­laxis with anti­con­vul­sants may be con­sid­ered for the
remain­der of the cycle and for sub­se­quent cycles. Rechallenge of
blinatumomab is not recommended after recur­rence of the ini­tial
neu­ro­toxic event.
Conclusions
Table 2 sum­ma­rizes the man­i­fes­ta­tions and man­age­ment of neu­
ro­tox­ic­ity from meth­o­trex­ate, CAR T cells, and blinatumomab. A
detailed under­stand­ing of the mech­an ­ isms of these neurotoxici-
ties will lead to the devel­op­ment of targeted approaches for pre­
ven­tion and treat­ment. Until then, close mon­i­tor­ing and atten­tive,
mul­ti­modal sup­port­ive care form the back­bone of man­ag­ing clin­
i­cal man­i­fes­ta­tions. In addi­tion, glu­co­cor­ti­coids and anticytokine
ther­a­pies dampen the neuroinflammatory cas­cade in ICANS. Neu-
rologic symp­toms are usu­ally tran­sient, and most patients appear
to make a full recov­ery. However, screen­ing MRIs detect leuko-
encephalopathy in 20% of chil­dren receiv­ing meth­o­trex­ate dur­
ing front­line ALL ther­apy,7 and even asymp­tom­atic patients with
leukoencephalopathy are at high risk for long-term adverse neu-
robehavioral prob­lems.41 Survivors of child­hood ALL treated with
che­mo­ther­apy alone con­tinue to have infe­rior neurocognitive
out­comes, par­tic­u­larly in domains of exec­u­tive func­tion, atten­
tion, and mem­ory.42 Similar pedi­at­ric data are not avail­­able after
immu­no­ther­apy, but a recent study noted clin­i­cally sig­nif­i­cant
neu­ro­psy­chi­at­ric prob­lems in 50% of adult patients 1 to 5 years
after CD19 CAR T-cell ther­apy.43 Therefore, in addi­tion to devel­
op­ing strat­e­gies to decrease the fre­quency and mor­bid­ity of
acute neu­ro­tox­ic­ity from ALL ther­apy, it is also imper­a­tive to
devise approaches to reduce its con­se­quences on long-term
out­comes.
Conflict-of-inter­est dis­clo­sure
Deepa Bhojwani: no com­pet­ing finan­cial inter­ests to declare.
Ravi Bansal: no com­pet­ing finan­cial inter­ests to declare.
Alan S. Wayne: receives research funding from Kite Pharma (a Gilead
com­pany) and Institut de Recherches Internationales Servier.
Off-label drug use
Deepa Bhojwani: multiple off-label drugs are discussed in the
article: dextromethorphan, aminophylline, siltuximab, anakinra,
lenzilumab, and dasatinib.
Ravi Bansal: multiple off-label drugs are discussed in the article:
dextromethorphan, aminophylline, siltuximab, anakinra, lenzi-
lumab, and dasatinib.
Alan S. Wayne: multiple off-label drugs are discussed in the article:
dextromethorphan, aminophylline, siltuximab, anakinra, lenzi-
lumab, and dasatinib.
Correspondence
Deepa Bhojwani, MD, Division of Hematology-Oncology,
­Children’s Hospital Los Angeles, 4650 Sunset Blvd., MS 54, Los
Angeles, CA 90027; email: dbhojwani@chla.usc.edu
382  |  Hematology 2021  |  ASH Education Program
References
1. Vora A, Andreano A, Pui C-H, et  al. Influence of cra­nial radio­ther­apy on
out­come in chil­dren with acute lym­pho­blas­tic leu­ke­mia treated with con­
tem­po­rary ther­apy. J Clin Oncol. 2016;34(9):919-926.
2. Parasole R, Petruzziello F, Menna G, et al. Central ner­vous sys­tem com­pli­
ca­tions dur­ing treat­ment of acute lym­pho­blas­tic leu­ke­mia in a sin­gle pedi­
at­ric insti­tu­tion. Leuk Lymphoma. 2010;51(6):1063-1071.
3. Banerjee J, Niinimäki R, Lähteenmäki P, et al. The spec­trum of acute cen­tral
ner­vous sys­tem symp­toms dur­ing the treat­ment of child­hood acute lym­
pho­blas­tic leu­kae­mia. Pediatr Blood Cancer. 2020;67(2):e27999.
4. Orgel E, Bhojwani D. Medical sup­port­ive care for treat­ment-related tox­ic­ity
in child­hood ALL. In: Vora A, ed. Childhood Acute Lymphoblastic Leuke-
mia. Springer International Publishing; 2017:299-321.
5. Gust J, Ceppi F, Turtle C. Neurotoxicities after CART T-cell immu­no­ther­
apy. In: Lee D, Shah N, eds. Chimeric Antigen Receptor T-Cell Therapies for
Cancer. Elsevier; 2020:83-105.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/376/1851679/376bhojwani.pdf by guest on 13 December 2021
6. Badke C, Flem­ing A, Iqbal A, et al. Rechallenging with intra­the­cal meth­o­trex­
ate after devel­op­ing sub­acute neu­ro­tox­ic­ity in chil­dren with hema­to­logic
malig­nan­cies. Pediatr Blood Cancer. 2016;63(4):723-726.
7. Bhojwani D, Sabin ND, Pei D, et al. Methotrexate-induced neu­ro­tox­ic­ity and
leukoencephalopathy in child­hood acute lym­pho­blas­tic leu­ke­mia. J Clin
Oncol. 2014;32(9):949-959.
8. Mateos MK, Marshall GM, Barbaro PM, et al. Methotrexate-related cen­tral
neu­ro­tox­ic­ity: clin­i­cal char­ac­ter­is­tics, risk fac­tors and genome-wide asso­
ci­a­tion study in chil­dren treated for acute lym­pho­blas­tic leu­ke­mia [pub-
lished online 11 Feb­ru­ary 2021]. Haematologica.
9. Bond J, Hough R, Moppett J, Vora A, Mitchell C, Goulden N. Stroke-like syn­
drome caused by intra­the­cal meth­o­trex­ate in patients treated dur­ing the
UKALL 2003 trial. Leukemia. 2013;27(4):954-956.
10. Schmiegelow K, Attarbaschi A, Barzilai S, et al; Ponte di Legno tox­ic­ity
work­ing group. Consensus def­i­ni­tions of 14 severe acute toxic effects for
child­hood lym­pho­blas­tic leu­kae­mia treat­ment: a Delphi con­sen­sus. Lancet
Oncol. 2016;17(6):e231-e239.
11. Taylor OA, Brown AL, Brackett J, et al. Disparities in neu­ro­tox­ic­ity risk and
out­comes among pedi­at­ric acute lym­pho­blas­tic leu­ke­mia patients. Clin
Cancer Res. 2018;24(20):5012-5017.
12. Inaba H, Khan RB, Laningham FH, Crews KR, Pui C-H, Daw N-C. Clinical
and radio­log­i­cal char­ac­ter­is­tics of meth­o­trex­ate-induced acute enceph­
a­lop­a­thy in pedi­at­ric patients with can­cer. Ann Oncol. 2008;19(1):178-
184.
13. Schmiegelow K. Advances in indi­vid­ual pre­dic­tion of meth­o­trex­ate tox­ic­
ity: a review. Br J Haematol. 2009;146(5):489-503.
14. Afshar M, Birnbaum D, Golden C. Review of dextromethorphan admin­is­
tra­tion in 18 patients with sub­acute meth­o­trex­ate cen­tral ner­vous sys­tem
tox­ic­ity. Pediatr Neurol. 2014;50(6):625-629.
15. Razi W, Haque AU, Sadiq H, Ullah R, Jabbar N, Mirza S. Safety and effi­cacy of
ami­noph­yl­line in intra­the­cal meth­o­trex­ate-related neu­ro­log­i­cal tox­ic­ity in
large pedi­at­ric oncol­ogy cen­tre. J Coll Physicians Surg Pak. 2021;30(4):481-
484.
16. Lexicomp. Riverwoods, IL: Wolters Kluwer Health, Inc. Accessed 15 June
2021. http:​­/​­/online​­.lexi​­.com
17. Mahoney DH Jr, Shuster JJ, Nitschke R, et al. Acute neu­ro­tox­ic­ity in chil­dren
with B-pre­cur­sor acute lym­phoid leu­ke­mia: an asso­ci­a­tion with inter­me­di­ate-
dose intra­ve­nous meth­o­trex­ate and intra­the­cal tri­ple ther­apy—a Pediatric
Oncology Group study. J Clin Oncol. 1998;16(5):1712-1722.
18. Winick NJ, Bowman WP, Kamen BA, et al. Unexpected acute neu­ro­logic
tox­ic­ity in the treat­ment of chil­dren with acute lym­pho­blas­tic leu­ke­mia. J
Natl Cancer Inst. 1992;84(4):252-256.
19. Cohen IJ. Neurotoxicity after high-dose meth­o­trex­ate (MTX) is ade­quately
explained by insuf­fi­cient folinic acid res­cue. Cancer Chemother Pharma-
col. 2017;79(6):1057-1065.
20. Dabagh S, David H, Young S, Doan A, Bhojwani D. Severe, fatal meth­o­trex­ate-
related neu­ro­tox­ic­ity in 2 ado­les­cent patients with ALL. J Pediatr Hematol
Oncol. 2020;42(8):e839-e844.
21. Lee DW, Santomasso BD, Locke FL, et  al. ASTCT con­ sen­
sus grad­ing
for cyto­kine release syn­drome and neu­ro­logic tox­ic­ity asso­ci­ated with
immune effec­tor cells. Biol Blood Marrow Transplant. 2019;25(4):625-
638.
22. Gust J, Hay KA, Hanafi L-A, et al. Endothelial acti­va­tion and blood-brain bar­
rier dis­rup­tion in neu­ro­tox­ic­ity after adop­tive immu­no­ther­apy with CD19
CAR-T cells. Cancer Discov. 2017;7(12):1404-1419.
23. Gofshteyn JS, Shaw PA, Teachey DT, et al. Neurotoxicity after CTL019 in a
pedi­at­ric and young adult cohort. Ann Neurol. 2018;84(4):537-546.
24. Tallantyre EC, Evans NA, Parry-Jones J, Mor­gan MPG, Jones CH, Ingram W.
Neurological updates: neu­ro­log­i­cal com­pli­ca­tions of CAR-T ther­apy. J Neu-
rol. 2021;268(4):1544-1554.
25. Morris EC, Neelapu SS, Giavridis T, Sadelain M. Cytokine release syn­drome
and asso­ci­ated neu­ro­tox­ic­ity in can­cer immu­no­ther­apy [published online
17 May 2021]. Nat Rev Immunol. 2021.
26. Neelapu SS, Tummala S, Kebriaei P, et al. Chimeric anti­gen recep­tor T-cell
ther­apy—assess­ment and man­age­ment of toxicities. Nat Rev Clin Oncol.
2018;15(1):47-62.
27. Maus MV, Alexander S, Bishop MR, et  al. Society for Immunotherapy of
Cancer (SITC) clin­i­cal prac­tice guide­line on immune effec­tor cell-related
adverse events. J Immunother Cancer. 2020;8(2):e001511.
28. Cornillon J, Hadhoum N, Roth-Guepin G, et  al. Management of CAR-T
cell-related enceph­a­lop­a­thy syn­drome in adult and pedi­at­ric patients: rec­
om­men­da­tions of the French Society of Bone Marrow trans­plan­ta­tion and
cel­lu­lar ther­apy (SFGM-TC) [in French]. Bull Cancer. 2020;107(1S):S12-S17.
29. Karschnia P, Jordan JT, Forst DA, et al. Clinical pre­sen­ta­tion, man­age­ment,
and bio­mark­ers of neu­ro­tox­ic­ity after adop­tive immu­no­ther­apy with CAR T
cells. Blood. 2019;133(20):2212-2221.
30. Strati P, Ahmed S, Kebriaei P, et al. Clinical effi­cacy of anakinra to mit­i­gate
CAR T-cell ther­apy-asso­ci­ated tox­ic­ity in large B-cell lym­phoma. Blood
Adv. 2020;4(13):3123-3127.
31. Norelli M, Camisa B, Barbiera G, et al. Monocyte-derived IL-1 and IL-6 are
dif­fer­en­tially required for cyto­kine-release syn­drome and neu­ro­tox­ic­ity
due to CAR T cells. Nat Med. 2018;24(6):739-748.
32. Galea J, Ogungbenro K, Hulme S, et al. Intravenous anakinra can achieve
exper­i­men­tally effec­tive con­cen­tra­tions in the cen­tral ner­vous sys­tem
within a ther­a­peu­tic time win­dow: results of a dose-rang­ing study. J Cereb
Blood Flow Metab. 2011;31(2):439-447.
33. Sterner RM, Sakemura R, Cox MJ, et al. GM-CSF inhi­bi­tion reduces cyto­kine
release syn­drome and neuroinflammation but enhances CAR-T cell func­
tion in xeno­grafts. Blood. 2019;133(7):697-709.
34. Mestermann K, Giavridis T, Weber J, et  al. The tyro­sine kinase inhib­i­tor
dasatinib acts as a phar­ma­co­logic on/off switch for CAR T cells. Sci Transl
Med. 2019;11(499):eaau5907.
35. Shah NN, Johnson BD, Fenske TS, Raj RV, Hari P. Intrathecal che­mo­ther­apy
for man­age­ment of ste­roid-refrac­tory CAR T-cell-asso­ci­ated neu­ro­tox­ic­ity
syn­drome. Blood Adv. 2020;4(10):2119-2122.
Prakash Singh Shekhawat
36. Wang M, Jain P, Chi TL, et al. Management of a patient with man­tle cell lym­
phoma who devel­oped severe neu­ro­tox­ic­ity after chi­me­ric anti­gen recep­
tor T-cell ther­apy in ZUMA-2. J Immunother Cancer. 2020;8(2):e001114.
37. Caulier B, Enserink JM, Wälchli S. Pharmacologic con­trol of CAR T cells. Int
J Mol Sci. 2021;22(9):4320.
38. Stein AS, Schiller G, Ben­ja­min R, et al. Neurologic adverse events in patients
with relapsed/refrac­tory acute lym­pho­blas­tic leu­ke­mia treated with blina-
tumomab: man­age­ment and miti­gat­ing fac­tors. Ann Hematol. 2019;98(1):
159-167.
39. von Stackelberg A, Locatelli F, Zugmaier G, et al. Phase I/phase II study of
blinatumomab in pedi­at­ric patients with relapsed/refrac­tory acute lym­
pho­blas­tic leu­ke­mia. J Clin Oncol. 2016;34(36):4381-4389.
40. BLINCYTO (blinatumomab) for injec­tion. Vol. 2021. Accessed 15 June 2021.
https:​­/​­/www​­.accessdata​­.fda​­.gov​­/drugsatfda_docs​­/label​­/2017​­/125557​
s008lbl​­.pdf
41. Cheung YT, Sabin ND, Reddick WE, et al. Leukoencephalopathy and long-
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/376/1851679/376bhojwani.pdf by guest on 13 December 2021
term neurobehavioural, neurocognitive, and brain imag­ing out­comes in
sur­vi­vors of child­hood acute lym­pho­blas­tic leu­kae­mia treated with che­mo­
ther­apy: a lon­gi­tu­di­nal anal­y­sis. Lancet Haematol. 2016;3(10):e456-e466.
42. Cheung YT, Krull KR. Neurocognitive out­ comes in long-term sur­ vi­vors of
child­hood acute lym­pho­blas­tic leu­ke­mia treated on con­tem­po­rary treat­ment
pro­to­cols: a sys­tem­atic review. Neurosci Biobehav Rev. 2015;53:108-120.
43. Ruark J, Mullane E, Cleary N, et al. Patient-reported neu­ro­psy­chi­at­ric out­
comes of long-term sur­vi­vors after chi­me­ric anti­gen recep­tor T cell ther­
apy. Biol Blood Marrow Transplant. 2020;26(1):34-43.
44. Traube C, Silver G, Kearney J, et al. Cornell assess­ment of pedi­at­ric delir­
ium: a valid, rapid, obser­va­tional tool for screen­ing delir­ium in the PICU.
Crit Care Med. 2014;42(3):656-663.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000269
Neurotoxicity in chil­dren with ALL  |  383
 LET THE CHIP(S) FALL WHERE THEY MAY ? BURDENS AND BENEFITS OF DIAGNOSING CLONAL HEMATOPOIESIS

CHIP: is clonal hematopoiesis a surrogate for aging

and other disease?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/384/1851752/384gondek.pdf by guest on 13 December 2021

Lukasz P. Gondek
The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD

Somatic mutations are an unavoidable consequence of aging tissues. Even though most mutations are functionally silent,
some may affect genes critical to proper tissue self-renewal and differentiation, resulting in the outgrowth of affected
cells, also known as clonal expansion. In hematopoietic tissue such clonal dominance is known as clonal hematopoiesis
(CH). Sporadic CH is frequent in aging and affects over 10% of individuals beyond the fifth decade of life. It has been
associated with an increased risk of hematologic malignancies and cardiovascular disease. In addition to aging, CH has
been observed in other hematologic conditions and confers an adaptation of hematopoietic stem cells (HSCs) to vari-
ous environmental stressors and cell-intrinsic defects. In the presence of extrinsic stressors such as genotoxic therapies,
T-cell-mediated immune attack, or inflammation, somatic mutations may result in augmentation of HSC fitness. Such
attuned HSCs can evade the environmental insults and outcompete their unadapted counterparts. Similarly, in inherited
bone marrow failures, somatic mutations in HSCs frequently lead to the reversion of inherited defects. This may occur
via the direct correction of germline mutations or indirect compensatory mechanisms. Occasionally, such adaptation
may involve oncogenes or tumor suppressors, resulting in malignant transformation. In this brief article, we focus on the
mechanisms of clonal dominance in various clinical and biological contexts.

LEARNING OBJECTIVES
• Review the etiology and nomenclature of CH
• Discuss the mechanism of clonal expansion and stem cell ftness in various clinical and biological contexts

Hemogram (normal range) Additional studies (normal range)

CLINICAL CASE
The patient is a 69-year-old man who is a computer program-
mer and has a medical history of hypertension, hyperlipid-
emia, coronary artery disease, and rheumatoid arthritis. He
was recently found to have mild normocytic normochromic
anemia during his annual physical exam. His initial workup
is shown in the table below (abnormal values are in bold).
Absolute reticulocyte count
25 (24.1-87.7 K/µL)
RDW: 15 (11%-14%)
Hgb, hemoglobin; MCV, mean corpuscular volume; PLT, platelet;
RBC, red blood cell; RDW, red cell distribution width; TIBC, total
iron-binding capacity; WBC, white blood cell.

A bone marrow biopsy revealed normocellular marrow

Hemogram (normal range)
WBC: 6.2 (4.5-11 K/µL)
RBC: 4.0 (4.5-5.9M/µL)
PLT: 189 (150-350 K/µL)
Hgb: 12.2 (13-16.3g/dL)
MCV: 90 (80-100 fL)
Additional studies (normal range)
Serum iron: 50 (60-150µg/dL)
Ferritin: 520 (30-400ng/mL)
Transferrin 102 (200-400mg/dL)
TIBC: 250 (300-360µg/dL)
B12 and folate within normal limits
for age and trilineage hematopoiesis with no dysplasia
and no increased blasts. Normal iron stores and no ringed
sideroblasts were noted on iron stain. Cytogenetics stud-
ies showed a normal male karyotype. A next-generation
sequencing (NGS) targeted myeloid panel revealed a
DNMT3A R882C mutation at a variant allele frequency
(VAF) of 2.5%.

384 | Hematology 2021 | ASH Education Program

Introduction
The term “clonal hema­to­poi­e­sis” (CH) describes the dom­i­nance of
a hema­to­poi­etic clone aris­ing from a sin­gle hema­to­poi­etic stem
cell (HSC) rel­a­tive to the rest of the hema­to­poi­etic cells. Since
the intro­duc­tion of NGS tech­nol­ogy in hema­to­poi­e­sis research,
an asso­ci­a­tion between CH and var­i­ous hema­to­logic and nonhe-
matologic con­di­tions has been reported. Thus, a num­ber of new
terms and acro­nyms comprising CH have been pro­posed.
For the pur­pose of this review, CH of inde­ter­mi­nate poten­tial
(CHIP) and age-related clonal hema­to­poi­e­sis (ARCH) will be con­
sid­ered syn­on­y­mous and will denote CH char­ac­ter­ized by the
pres­ence of 1 or more somatic muta­tions in genes asso­ci­ated
with hema­to­logic malig­nan­cies in oth­er­wise healthy indi­vid­u­als
with­out obvi­ous hema­to­logic con­di­tions. Moreover, the muta­
tions must be pres­ent at a VAF ≥2%, although con­sen­sus on the
bio­log­i­cally and clin­i­cally mean­ing­ful VAF cut­off is unknown and
has been a sub­ject of ongo­ing research.1-3
Clonal cytopenia of unde­ ter­
mined sig­ nif­i­
cance (CCUS)
describes the pres­ence of CH in indi­vid­u­als with per­sis­tent uni-
or multilineage cytopenias that can­not be explained by other
con­di­tions.3 Since cytopenia in CCUS is thought to be due to
the under­ly­ing clonal pro­cess, this con­di­tion often rep­re­sents an
early mye­loid neo­plasm.2 A detailed descrip­tion and the clin­i­cal
con­se­quences of CCUS are pro­vided in an accom­pa­ny­ing review
by Dr. Osman.
Idiopathic cytopenia of unde­ter­mined significance (ICUS)
denotes per­sis­tent uni- or multilineage cytopenia unex­plained
by other con­di­tions and hav­ing no evi­dence of CH.4
We pro­pose that the term CH be used for any clonal pro­cess
that does not ful­fill the cri­te­ria for the above-men­tioned con­di­
tions (Table 1).
Table 1. Definition of clin­i­cal enti­ties asso­ci­ated with CH
Clonal hema­to­poi­e­sis of inde­ter­mi­nate poten­tial (CHIP)
  •  Presence of somatic muta­tion(s) in gene(s) asso­ci­ated with hema­to­
logic malig­nan­cies
  •  Normal hemogram
  •  No clin­i­cal of path­o­log­i­cal evi­dence of hema­to­logic malig­nancy
Age-related clonal hema­to­poi­e­sis (ARCH)
  •  Any CH asso­ci­ated with aging
Clonal cytopenia of unde­ter­mined sig­nif­i­cance (CCUS)
  •  Presence of somatic muta­tion(s) in gene(s) asso­ci­ated with hema­to­
logic malig­nan­cies
  •  Presence of clin­i­cally rel­e­vant and per­sis­tent cytopenia in 1 or more
periph­eral blood cell lin­e­ages
  •  Diagnostic cri­te­ria of mye­loid neo­plasm not fulfilled
  •  No other causes of cytopenia and molec­u­lar aber­ra­tion
Idiopathic cytopenia of unde­ter­mined sig­nif­i­cance (ICUS)
  •  Presence of clin­i­cally rel­e­vant and per­sis­tent cytopenia in 1 or more
periph­eral blood cell lin­e­ages
  •  Not explained by any other dis­ease
  •  Diagnostic cri­te­ria of mye­loid neo­plasm not fulfilled
Prakash Singh Shekhawat
The eti­­ol­ogy of clonal dom­i­nance
Aging
Despite the human body’s very effec­tive DNA repair machin­ery,
somatic muta­tions are inev­i­ta­ble, and their num­ber increases
with time. In fact, every tis­ sue accu­mu­ lates somatic muta­
tions dur­ing devel­op­ment and aging, resulting in the gen­er­a­
tion of cells with dif­fer­ent geno­types, a phe­nom­e­non known
as somatic mosa­i­cism.5,6 This is par­tic­u­larly prev­a­lent in highly
pro­lif­er­a­tive tis­sues such as the hema­to­poi­etic sys­tem, and
somatic muta­tions are pres­ent in the blood of nearly all­older
adults.7,8 Fortunately, because of their pre­dom­i­nantly sto­chas­
tic nature, most muta­tions affect noncoding regions and thus
remain func­tion­ally silent. Assuming that approx­i­ma­tely 1000
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/384/1851752/384gondek.pdf by guest on 13 December 2021
HSCs are active and con­ trib­ute equally to human hema­ to­
poi­e­sis, somatic mosa­i­cism would be expected at a level of
0.001 in the absence of clonal advan­tage of an indi­vid­ual HSC.
In fact, anal­y­sis of the periph­eral blood of patients in the fifth
decade of life using ultradeep sequenc­ing showed the pres­
ence of somatic mosa­i­cism with a median VAF of 0.002 in 95%
of indi­vid­u­als. Interestingly, most clones remained sta­ble over
time.8 Thus, somatic mosa­ i­
cism is fre­
quent with aging but
rarely results in clonal advan­tage and sub­se­quent CH. The rel­
a­tive expan­sion of a sin­gle clone can arise either due to the
increased fit­ ness and growth advan­ tage of an HSC (large
numer­a­tor) or as a result of the HSC’s pool con­trac­tion (small
denom­i­na­tor). Pathogenic muta­tions, often in can­cer-asso­
ci­ated genes, fre­quently result in the expan­sion of HSCs and
clonal dom­i­nance. Nonfunctional var­i­ants, on the other hand,
serv­ing only as “clonal mark­ers,” may become detect­able in the
case of HSC pool attri­tion and when ultradeep or wide breadth
of cov­er­age NGS is applied. Consistent with this notion, CH was
found to be sig­nif­i­cantly more prev­a­lent when whole-genome
sequenc­ing meth­ods were applied.7 What then is the under­
ly­ing mech­a­nism of hema­to­poi­etic clonal expan­sion in aging
hema­ to­poi­ e­
sis? Undoubtedly, the HSC pool decreases with
age; how­ever, sto­chas­tic HSC attri­tion does not appear to be
the major means behind clin­i­cally con­se­quen­tial CH (Figure 1)
CH seems to be pre­dom­i­nantly driven by a cell auton­o­mous
increase in HSC self-renewal and pos­i­tive selec­tion rather than
genetic drift.9 This is mostly due to qual­i­ta­tive and/or quan­ti­
ta­tive changes in pro­teins encoded by so-called driver genes.
DNMT3A and TET2 are among the most com­monly mutated
genes in ARCH and belong to the class of epi­ge­netic mod­i­fi­ers.
Since both are essen­tial in self-renewal and dif­fer­en­ti­a­tion, they
impose their effects in a cell auton­o­mous man­ner.
The DNMT3A gene encodes a methyltransferase respon­si­ble
for de novo DNA meth­yl­a­tion that is crit­i­cal for tran­si­tion from
one cell state to another and thus is required for proper tis­sue
devel­op­ment and stem cell main­te­nance. Animal stud­ies have
shown that the loss of DNMT3A aug­ments HSC self-renewal with
only a mod­est impact on dif­fer­en­ti­a­tion, lead­ing to the grad­ual
expan­sion of mutated HSC and clonal dom­i­nance.10
TET2 is the sec­ond most fre­quently mutated gene in CHIP.
The prod­uct of this gene belongs to the fam­ily of dioxygen-
ases. TET2 medi­ates DNA demethylation by cat­a­lyz­ing the oxi­
da­ tion of 5-methylcytosine to 5-hydroxymethylcytosine that
remains unmethylated dur­ing sub­se­quent cell divi­sion.11 Simi-
larly to DNMT3A loss, ani­mal mod­els of TET2 knock­out exhibit
Mechanism of clonal hema­to­poi­e­sis  |  385

Figure 1. Context-dependent mechanism of clonal expansion.
increased HSC self-renewal and skewed hema­to­poi­etic dif­fer­
en­ti­a­tion favor­ing myelomonocytic lin­e­age.12 Similar gain of
self-renewal has been observed with ASXL1 muta­tions.13 The
impact of less com­mon CH muta­tions on clonal growth advan­
tage, par­tic­ul­arly in genes involved in mRNA splic­ing, is not well
under­stood.
Cell-extrin­sic fac­tors
A very mod­est increase in the fit­ness of an indi­vid­ual HSC often
requires years or decades to gain dom­i­nance over unmutated
coun­ter­parts. This mostly indo­lent pro­cess may be fur­ther accel­
er­ated by a pleth­ora of extra­cel­lu­lar stress­ors, such as increased
pro­lif­er­a­tive pres­sure, inflam­ma­tion, and genotoxic agents. The
most pro­found stress that can be imposed on the hema­to­poi­etic
sys­tem is related to hema­to­poi­etic recon­sti­tu­tion after allo­ge­
neic bone mar­row trans­plan­ta­tion. Recent stud­ies have dem­on­
strated that CH clones pres­ent in allo­ge­neic stem cell donors
undergo accel­er­ated expan­sion in the recip­i­ents while remain-
ing rel­at­ ively dor­mant in the donors.14-16 Inflammation is another
pre­sump­tive mod­i­fier of CH and has been fre­quently asso­ci­ated
with tis­sue aging. Mutated HSC may gain a selec­tive growth
advan­tage when exposed to proinflammatory sig­nals such as IL-6
or IFN-gamma.17,18
386  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/384/1851752/384gondek.pdf by guest on 13 December 2021
Immune attack.  The immune destruc­tion of hema­to­poi­etic cells
is a hall­mark of acquired aplastic ane­mia (AA). CH is pres­ent in up
to 50% of AA patients.19 Perhaps the best exam­ple of clonal im-
mune escape is the clonal advan­tage of the par­ox­ys­mal noc­tur­
nal hemo­glo­bin­uria (PNH) clone aris­ing from phosphatidylinosi-
tol gly­can class A-defi­cient HSC. The PNH clone can be found at
low lev­els in most healthy indi­vid­u­als, but the cells lack a com­
pet­i­tive advan­tage under homeo­static con­di­tions. Under im-
mune pres­sure, gly­co­syl-phosphatidylinositol-defi­cient cells are
less sus­cep­ti­ble to T-cell-medi­ated kill­ing and out­com­pete their
unmutated coun­ter­parts.20 This obser­va­tion raises the pos­si­bil­ity
that the gly­co­syl-phosphatidylinositol-anchored pro­tein might
be a crit­i­cal tar­get rec­og­nized by T cells. Inactivating muta­tions
or loss-of-het­ero­zy­gos­ity of HLA class 1 alleles, allowing for resis­
tance against autoreactive T cells, is another exam­ple of HSC
adap­ta­tion to immune-medi­ated stress.21 The role of other muta­
tions in increased HSC fit­ness is less clear. DNMT3A and ASXL1
are seen in up to 15% of patients with AA.19 Also, muta­tions in
BCOR and BCORL1 (tran­scrip­tional core­pres­sors) are fre­quently
asso­ci­ated with AA. Unlike clones with DNMT3A and ASXL1 muta­
tions that may expand over time and lead to malig­nant trans­
for­ma­tion, clones car­ry­ing BCOR and BCORL1 muta­ tions re-
main sta­ble and sel­dom result in pro­gres­sion to acute mye­loid
leu­ke­mia/myelodysplastic syn­drome.19 The exact mech­a­nism of
the afore­men­tioned muta­tions to aug­ment HSC fit­ness in the
pres­ence of T-cell-medi­ated attack is unclear.
Genotoxic ther­a­pies. Unlike already primed HSCs with muta­
tions resulting in improved fit­ness, some hema­to­poi­etic clones
lack the clonal advan­tage under homeo­static con­di­tions but are
capa­ble of evad­ing extra­cel­lu­lar insults or intra­cel­lu­lar defects
(Figure 1). Patients treated with genotoxic che­ mo­ ther­apy or
radio­ther­apy are at risk for a ther­apy-related mye­loid neo­plasm.
It is now well rec­og­nized that min­is­cule hema­to­poi­etic clones
car­ry­ing TP53 or PPM1D muta­tions fre­quently not detect­able by
stan­dard NGS tech­niques are pres­ent long before treat­ment with
Cell-intrin­sic defects
In addi­tion to spo­radic CH asso­ci­ated with aging or shaped by
extra­cel­lu­lar stress­ors, CH can also rep­re­sent an adap­ta­tion to
intrin­
sic HSC defects under­ ly­
ing inherited bone mar­ row fail­
ure syn­dromes (iBMFs). iBMFs occur as a result of germline HSC
muta­tions and include short telo­mere syn­dromes (STS), impaired
ribo­some bio­gen­e­sis (Shwachman-Diamond syn­drome; SDS),
an increase in DNA dam­age (Fanconi ane­mia; FA), and SAMD9/
SAMD9L muta­tions (MIRAGE syn­drome). Several com­pen­sa­
tory mech­a­nisms may lead to an increase in HSC self-renewal
in iBMFs. Randomly occur­ring muta­tions in HSCs may result in
somatic rever­sion of the genetic defect, allowing the corrected
clone to “res­cue” hema­to­poi­etic out­put. This may be achieved

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/384/1851752/384gondek.pdf by guest on 13 December 2021

genotoxic agents.22 It is likely that func­tion­ally incon­se­quen­tial
somatic muta­tions in TP53 and PPM1D, under homeo­static con­di­
tions, are a­ ble to resist genotoxic stress. Given that these muta­
tions likely con­fer an inad­e­quate response to DNA dam­age, it is
not sur­pris­ing that while exposed to genotoxic ther­apy they may
cir­cum­vent apo­pto­sis and acquire addi­tional DNA defects. The
com­mon exam­ple is the loss of the wild-type TP53 allele result-
ing in a com­plete loss of func­tion and sub­se­quent accu­mu­la­tion
of addi­tional genetic defects and leu­ke­mic trans­for­ma­tion.23,24
via the direct cor­rec­tion of germline muta­tion or an indi­rect com­
pen­sa­tory mech­a­nism (Table 1).25 Somatic rever­sion of DKC1, a
gene fre­quently mutated in STS, or gain-of-func­tion muta­tion in
the TERT pro­moter lead­ing to increased mRNA expres­sion have
been observed in STS. Direct cor­rec­tion of dam­ag­ing muta­tions
in FA patients and the full or par­tial func­tion res­to­ra­tion of sev­eral
genes in the FANC com­plex have also been observed. Since
bone mar­row fail­ure in FA is mainly medi­ated by p53-induced
apo­pto­sis, the emer­gence of CH with TP53 loss-of-func­tion

Table 2. CH char­ac­ter­is­tics in var­i­ous clin­i­cal con­texts and selec­tive pres­sures

Selective pres­sure Clinical con­text CH char­ac­ter­is­tics Frequency (%)

Extrinsic/intrin­sic Aging DNMT3A 8
TET2 2
ASXL1 2
Extrinsic AA DNMT3A 9
ASXL1 13
BCOR 6
PNH TET2 8
SUZ12 3
U2AF1 3
ASXL1 3
JAK2 5
Chemotherapy TP53 3
PPM1D 15
Intrinsic DC Backmutation 11
TERT pro­moter 5
FA Gene con­ver­sion 9
FANC point muta­tion Case reports
FANC frame­shift muta­tion Case reports
SDS Isochromosome 7q 44
Monosomy 7 13
Deletion 20q 22
TP53 48
SCN CSF3R 12
MIRAGE syn­drome Deletion 7q 18
UPD7q 20
DC, dyskeratosis congenita; SCN, severe con­gen­i­tal neutropenia.

Prakash Singh Shekhawat

Mechanism of clonal hema­to­poi­e­sis  |  387
may pro­vide bet­ter sur­vival and improve­ment in hema­to­poi­
etic out­put. The lat­ter is an exam­ple of the indi­rect cor­rec­
tion of a germline defect. Similarly, TP53 muta­tions may res­cue
bone mar­row fail­ure in SDS patients in whom abnor­mal ribo­
some bio­gen­e­sis trig­gers p53-medi­ated apo­pto­sis. The cleav­
age of EIF6 by SBDS pro­tein is essen­tial for proper ribo­some
assem­bly. Thus, either muta­tions in EIF6 or chro­mo­some 20q
dele­tion resulting in EIF6 haploinsufficiency con­fers the indi­
rect cor­rec­tion of SBDS muta­tions.26 Finally, the dupli­ca­tion of
the hypo­mor­phic SBDS allele, through the emer­gence of CH
with iso­ chro­mo­ some 7q, may par­ tially reverse the dis­ ease
phe­no­type.27 In con­trast, in SAMD9/SAMD9L-medi­ated bone
mar­row fail­ure, somatic rever­sion may be accom­plished by the
loss of the mutant allele located on chro­mo­some 7q (Table 2).28
The role of CH in iBMFs is discussed in detailed by Dr. Jung in
an accom­pa­ny­ing arti­cle.
CH is com­mon in BMFs and fre­quently rep­re­sents an adap­ta­
tion of HSCs to extra­cel­lu­lar or cell-intrin­sic stress­ors, improv­ing
stem cell fit­ness and periph­eral blood count. However, such cor­
rec­tion may result in a higher inci­dence of hema­to­logic malig­
nan­cies, espe­cially when involv­ing puta­tive onco­genes or tumor
sup­pres­sors.
CLINICAL CASE (Con­t in­u ed)
The patient had a sin­gle hot spot muta­tion in DNMT3A with a
VAF of 2.5%. This likely rep­re­sents spo­radic CH asso­ci­ated with
aging. The patient’s normocytic normochromic and hypopro-
liferative ane­mia is likely related to chronic inflam­ma­tion given
the patient’s his­tory of rheu­ma­toid arthri­tis and the ele­va­tion of
inflam­ma­tory mark­ers (fer­ri­tin). The hemogram and iron stud­ies
are highly sug­ges­tive of ane­mia of inflam­ma­tion (AI), par­tic­u­
larly in the absence of nutri­tional deficiencies or other appar­
ent abnor­mal­i­ties, includ­ing an unre­mark­able bone mar­row
exam­i­na­tion. It appears that the ane­mia is not caused solely by
an under­ly­ing CH. Additionally, the pres­ence of a minor clone
account­ing for only 5% of the total hema­to­poi­etic out­put (het­
ero­zy­gous muta­tion with a VAF of 2.5%) would not be suf­fi­cient
to have a sig­nif­i­cant impact on the hema­to­poi­etic out­put from
95% nonclonal coun­ter­parts in a cell-auton­o­mous man­ner. How-
ever, the exact VAF cut­off asso­ci­ated with cytopenias is not yet
established, and the val­ues are likely muta­tion-spe­cific. Of note,
DNMT3A muta­tions seem to be more fre­quent in older patients
with AI, but the pres­ence of small clones (<5% VAF) has no
impact on clin­i­cal out­come.29 Even though the patient may be
at risk for pro­gres­sion to myelodysplastic syn­drome or acute
mye­loid leu­ke­mia, the prob­ab ­ il­ity appears to be small given the
lack of high-risk CH fea­tures such as mul­ti­ple muta­tions, spli-
ceosome, IDH1/2 or TP53 muta­tions, and high VAF.30-32 Since
the pres­ence of CHIP is an impor­tant risk fac­tor for a major car­
dio­vas­cu­lar event, the patient should be coun­seled on life­style
mod­i­fi­ca­tion to address mod­i­fi­able car­dio­vas­cu­lar risk fac­tors.
Conflict-of-inter­est dis­clo­sure
Lukasz P. Gondek: no com­pet­ing finan­cial inter­ests to declare.
388  |  Hematology 2021  |  ASH Education Program
Off-label drug use
Lukasz P. Gondek: nothing to disclose.
Correspondence
Lukasz P. Gondek, The Sidney Kimmel Comprehensive Cancer
Center, Johns Hopkins University School of Medicine, 1650 Or-
leans St, CRB1-290, Baltimore, MD 21287; e-mail: lgondek1@jhmi​
­.edu.
References
1. Steensma DP, Bejar R, Jaiswal S, et al. Clonal hema­to­poi­e­sis of inde­ter­mi­
nate poten­tial and its dis­tinc­tion from myelodysplastic syn­dromes. Blood.
2015;126(1):9-16.
2. Gondek LP, DeZern AE. Assessing clonal haematopoiesis: clin­i­cal bur­dens
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/384/1851752/384gondek.pdf by guest on 13 December 2021
and ben­e­fits of diag­nos­ing myelodysplastic syn­drome pre­cur­sor states.
Lancet Haematol. 2020;7(1):e73-e81.
3. DeZern AE, Malcovati L, Ebert BL. CHIP, CCUS, and other acro­nyms: def­i­ni­
tion, impli­ca­tions, and impact on prac­tice. Am Soc Clin Oncol Educ Book.
2019;39(Jan­u­ary):400-410.
4. Valent P, Bain BJ, Bennett JM, et  al. Idiopathic cytopenia of unde­ter­
mined sig­nif­i­cance (ICUS) and idi­o­pathic dys­pla­sia of uncer­tain sig­nif­
i­cance (IDUS), and their dis­tinc­tion from low risk MDS. Leuk Res. 2012;
36(1):1-5.
5. De S. Somatic mosa­i­cism in healthy human tis­sues. Trends Genet. 2011;27(6):
217-223.
6. Risques RA, Kennedy SR. Aging and the rise of somatic can­cer-asso­ci­ated
muta­tions in nor­mal tis­sues. PLoS Genet. 2018;14(1):e1007108.
7. Zink F, Stacey SN, Norddahl GL, et  al. Clonal hema­to­poi­e­sis, with and
with­ out can­ di­date driver muta­ tions, is com­mon in the elderly. Blood.
2017;130(6):742-752.
8. Young AL, Challen GA, Birmann BM, Druley TE. Clonal haematopoiesis
harbouring AML-asso­ci­ated muta­tions is ubiq­ui­tous in healthy adults. Nat
Commun. 2016;7(August):12484.
9. Watson CJ, Papula AL, Poon GYP, et  al. The evo­lu­tion­ary dynam­ics and
fit­ness land­scape of clonal hema­to­poi­e­sis. Science. 2020;367(6485):1449-
1454.
10. Challen GA, Goodell MA. Clonal hema­to­poi­e­sis: mech­a­nisms driv­ing dom­
i­nance of stem cell clones. Blood. 2020;136(14):1590-1598.
11. Ko M, Huang Y, Jankowska AM, et al. Impaired hydrox­yl­ation of 5-methylcy-
tosine in mye­loid can­cers with mutant TET2. Nature. 2010;468(7325):839-
843.
12. Li Z, Cai X, Cai CL, et  al. Deletion of Tet2 in mice leads to dysregulated
hema­to­poi­etic stem cells and sub­se­quent devel­op­ment of mye­loid malig­
nan­cies. Blood. 2011;118(17):4509-4518.
13. Abdel-Wahab O, Gao J, Adli M, et al. Deletion of Asxl1 results in myelodyspla-
sia and severe devel­op­men­tal defects in vivo. J Exp Med. 2013;210(12):2641-
2659.
14. Gondek LP, Zheng G, Ghiaur G, et al. Donor cell leu­ke­mia aris­ing from clonal
hema­to­poi­e­sis after bone mar­row trans­plan­ta­tion. Leukemia. 2016;30(9):
1916-1920.
15. Wong WH, Bhatt S, Trinkaus K, et al. Engraftment of rare, path­o­genic donor
hema­to­poi­etic muta­tions in unre­lated hema­to­poi­etic stem cell trans­plan­
ta­tion. Sci Transl Med. 2020;12(526).
16. Frick M, Chan W, Arends CM, et  al. Role of donor clonal hema­to­poi­e­
sis in allo­ge­neic hema­to­poi­etic stem-cell trans­plan­ta­tion. J Clin Oncol.
2019;37(5):375-385.
17. Meisel M, Hinterleitner R, Pacis A, et al. Microbial sig­nals drive pre-leukae-
mic myeloproliferation in a Tet2-defi­cient host. Nature. 2018;557(7706):580-
584.
18. Hormaechea-Agulla D, Matatall KA, Le DT, et  al. Chronic infec­tion drives
Dnmt3a-loss-of-func­tion clonal hema­to­poi­e­sis via IFNγ sig­nal­ing. Cell Stem
Cell. 2021;28(8):1428-1442.e61442e6.
19. Yoshizato T, Dumitriu B, Hosokawa K, et al. Somatic muta­tions and clonal
hema­to­poi­e­sis in aplastic ane­mia. N Engl J Med. 2015;373(1):35-47.
20. Cooper JN, Young NS. Clonality in con­text: hema­to­poi­etic clones in their
mar­row envi­ron­ment. Blood. 2017;130(22):2363-2372.
21. Babushok DV, Perdigones N, Perin JC, et al. Emergence of clonal hema­to­
poi­e­sis in the major­ity of patients with acquired aplastic ane­mia. Cancer
Genet. 2015;208(4):115-128.
22. Gib­son CJ, Lindsley RC, Tchekmedyian V, et al. Clonal hema­to­poi­e­sis asso­
ci­ated with adverse out­comes after autol­o­gous stem-cell trans­plan­ta­tion
for lym­phoma. J Clin Oncol. 2017;35(14):1598-1605.
23. Bernard E, Nannya Y, Hasserjian RP, et al. Implications of TP53 alle­lic state
for genome sta­bil­ity, clin­i­cal pre­sen­ta­tion and out­comes in myelodysplas-
tic syn­dromes. Nat Med. 2020;26(10):1549-1556.
24. Jasek M, Gondek LP, Bejanyan N, et al. TP53 muta­tions in mye­loid malig­nan­
cies are either homo­zy­gous or hemi­zy­gous due to copy num­ber-neu­tral
loss of het­ero­zy­gos­ity or dele­tion of 17p. Leukemia. 2010;24(1):216-219.
25. Pasca S, Gondek LP. Clonal hema­to­poi­e­sis and bone mar­row fail­ure syn­
dromes. Best Pract Res Clin Haematol. 2021;34(2):101273.
26. Kennedy AL, Myers KC, Bowman J, et al. Distinct genetic path­ways define
pre-malig­nant ver­sus com­pen­sa­tory clonal hema­to­poi­e­sis in Shwachman-
Diamond syn­drome. Nat Commun. 2021;12(1):1334.
27. Minelli A, Maserati E, Nicolis E, et al. The iso­chro­mo­some i(7)(q10) car­ry­ing
c.258 + 2t>c muta­tion of the SBDS gene does not pro­mote devel­op­ment of
mye­loid malig­nan­cies in patients with Shwachman syn­drome. Leukemia.
2009;23(4):708-711.
Prakash Singh Shekhawat
28. Pastor VB, Sahoo SS, Boklan J, et  al. Constitutional SAMD9L muta­tions
cause famil­ial myelodysplastic syn­drome and tran­sient mono­somy 7. Hae-
matologica. 2018;103(3):427-437.
29. van Zeventer IA, de Graaf AO, Wouters HJCM, et al. Mutational spec­trum
and dynam­ics of clonal hema­to­poi­e­sis in ane­mia of older indi­vid­u­als.
Blood. 2020;135(14):1161-1170.
30. Abelson S, Collord G, Ng SWK, et al. Prediction of acute mye­loid leu­kae­mia
risk in healthy indi­vid­u­als. Nature. 2018;559(7714):400-404.
31. Desai P, Mencia-Trinchant N, Savenkov O, et al. Somatic muta­tions pre­cede
acute mye­loid leu­ke­mia years before diag­no­sis. Nat Med. 2018;24(7):1015-
1023.
32. Rossi M, Meggendorfer M, Zampini M, et  al. Clinical rel­e­vance of clonal
hema­to­poi­e­sis in the oldest-old pop­u­la­tion. Published online 14 June 2021.
Blood.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/384/1851752/384gondek.pdf by guest on 13 December 2021
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000270
Mechanism of clonal hema­to­poi­e­sis  |  389
 LET THE CHIP(S) FALL WHERE THEY MAY ? BURDENS AND BENEFITS OF DIAGNOSING CLONAL HEMATOPOIESIS

Mechanisms of somatic transformation in

inherited bone marrow failure syndromes

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/390/1852164/390choijilsuren.pdf by guest on 13 December 2021

Haruna Batzorig Choijilsuren,1,2,* Yeji Park,1,* and Moonjung Jung1
1
Division of Hematology, Department of Medicine, School of Medicine, Johns Hopkins University, Baltimore, MD; and 2Department of Molecular and
Cellular Biology, Krieger School of Arts and Sciences, Johns Hopkins University, Baltimore, MD
*These authors contributed equally.

Inherited bone marrow failure syndromes (IBMFS) cause hematopoietic stem progenitor cell (HSPC) failure due to ger-
mline mutations. Germline mutations influence the number and fitness of HSPC by various mechanisms, for example,
abnormal ribosome biogenesis in Shwachman-Diamond syndrome and Diamond-Blackfan anemia, unresolved DNA cross-
links in Fanconi anemia, neutrophil maturation arrest in severe congenital neutropenia, and telomere shortening in short
telomere syndrome. To compensate for HSPC attrition, HSPCs are under increased replication stress to meet the need for
mature blood cells. Somatic alterations that provide full or partial recovery of functional deficit implicated in IBMFS can
confer a growth advantage. This review discusses results of recent genomic studies and illustrates our new understand-
ing of mechanisms of clonal evolution in IBMFS.

LEARNING OBJECTIVES
• Learn recent advances in our understanding of clonal hematopoiesis in IBMFS
• Recognize early signs of malignant transformation in IBMFS

Introduction ideal timing for transplant. Although there are no consen­

Inherited bone marrow failure syndromes (IBMFS) manifest as
ineffective and stressed hematopoiesis due to germline muta­
tions that cause hematopoietic stem progenitor cell (HSPC)
failure. The causes of HSPC failure are different depending on
disease mechanisms: unresolved DNA cross­links in Fanconi
anemia (FA), telomere shortening in dyskeratosis congenita
(DC) (also known as short telomere syndromes [STSs]), and
ribosomopathy in Shwachman­Diamond syndrome (SDS)
and Diamond­Blackfan anemia (DBA).1 However, clinical con­
sequences are similar in that patients experience ineffec­
tive hematopoiesis, leading to replicative stress and HSPC
exhaustion, and frequently develop malignant transformation
(Table 1).
Successful hematopoietic stem cell transplant (HSCT)
can cure bone marrow failure (BMF) and prevent leuke­
mic transformation; however, transplant outcome is poor
in patients who have already developed malignant clones
before HSCT.2,3 Besides, HSCT comes with a price of poten­
tial end­organ damage and accelerated solid tumorigene­
sis. Therefore, the major goal of surveillance is to detect
early signs of malignant transformation and determine the
sus surveillance guidelines with a high level of evidence
due to a lack of randomized controlled trials, experts gen­
erally agree with serial monitoring of blood counts and
bone marrow (BM). Risk­based surveillance recommenda­
tions are summarized in the Figure 1.
Clonal hematopoiesis (CH) is frequently seen in older
individuals. Ten percent to 20% of individuals older than 70
years carry somatic mutations in blood, most frequently in
genes involved in epigenetic regulation (DNMT3A, TET2,
and ASXL1), which is associated with about 1% per year risk
for progression to myeloid malignancies.4 Clonal hema­
topoiesis in IBMFS has a couple of important differences
compared with CH observed in older, otherwise healthy
individuals; it occurs more frequently and early in life. The
mutational spectrum is also quite different. This is likely
because preexisting molecular and cellular defects due
to germline mutations cause stem cell exhaustion at base­
line, and somatic mutations that can compensate under­
lying defects confer a growth advantage.5 It is currently
not established whether the detection of clones with par­
ticular mutations or progressive expansion of such clones

390 | Hematology 2021 | ASH Education Program

Table 1. Inherited BMF syn­dromes

Causative genes Frequent chro­mo­ Frequent somatic

Syndrome ­(inher­i­tance) Etiology Clinical man­i­fes­ta­tions somal abnor­mal­i­ties muta­tions
SDS SBDS (AR)—m/c Ribosomopathy Hematopoietic/onco­logic: 20q– (inter­sti­tial EIF6—improve ribo­
DNAJC21 (AR) neutropenia, ane­mia, throm­ dele­tion q11.21- some bio­gen­e­sis
SRP54 (AD) bo­cy­to­pe­nia, aplastic ane­ q13.32)—loss of and pro­tein trans­la­
ELF1 (AR) mia, MDS, AML EIF6 tion effi­ciency
Nonhematologic: exo­crine pan­ i(7)(q10)—dupli­ca­tion
cre­atic insuf­fi­ciency, fail­ure of hypo­mor­phic High risk:
to thrive, mal­ab­sorp­tion, SBDS allele Biallelic TP53—
short stat­ure, skel­e­tal abnor­ impaired cell cycle
mal­i­ties (chondrodysplasia or High risk: check­point and
con­gen­i­tal tho­racic dys­ –7/7q– apo­pto­sis
Complex

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/390/1852164/390choijilsuren.pdf by guest on 13 December 2021

tro­phy), endo­crine abnor­
mal­i­ties, liver dys­func­tion,
cog­ni­tive impair­ment
FA FANCA (AR)—m/c Inability to repair Hematologic/onco­logic: 1q+ Somatic rever­sion
FANCB (XR) DNA inter­strand throm­bo­cy­to­pe­nia, ane­mia, of FANC genes
FANCC (AR) cross-links neutropenia, aplastic ane­mia, High risk: by back muta­
FANCD1/BRCA2 (AR) MDS, leu­ke­mia, squa­mous 3q+ tion, intra­genic
FANCD2 (AR) cell car­ci­noma (head and –7/7q– recom­bi­na­tion,
FANCE (AR) neck, anogenital, esoph­ Complex sec­ond site muta­
FANCF (AR) a­geal), embry­o­nal tumors Cryptic RUNX1 trans­ tions suppressing
FANCG (AR) (medul­lo­blas­toma, neu­ro­ lo­ca­tion the effect of the
FANCI (AR) blas­toma, Wilms tumor in orig­i­nal muta­tion,
FANCJ/BRIP1/BACH1 patients with biallelic BRCA2 frame-restor­ing
(AR) muta­tions) sec­ond site inser­
FANCL (AR) Nonhematopoietic: short stat­ tions and dele­tions
FANCM (AR) ure, low birth weight, fail­ure
FANCN/PALB2 (AR) to thrive, micro­ceph­aly, High risk:
FANCO/RAD51C (AR) microphthalmia, hear­ing loss, RUNX1
FANCP/SLX4 (AR) tri­an­gu­lar face, micrognathia, RAS path­way (KRAS,
FANCQ/ERCC4 (AR) car­diac abnor­mal­i­ties (pat­ent PTPN11)—(small
FANCR/RAD51 (AD) duc­tus arte­ri­o­sus, atrial sep­ num­ber of cases)
FANCS/BRCA1 (AR) tal defect, ven­tric­u­lar sep­tal
FANCT/UBE2T (AR) defect), tracheoesophageal
FANCU/XRCC2 (AR) fis­tula, esoph­a­geal atre­sia,
FANCV/REV7 (AR) horse­shoe or ectopic kid­neys,
FANCW/RFWD3 (AR) thumb/radius abnor­mal­i­ties,
hypo­plas­tic the­nar emi­nence,
clinodactyly, café-au-lait
spots, hypo/hyper­pig­men­ta­
tion of the skin
SCN ELANE (AD)—m/c Neutrophil mat­u­ Hematologic/onco­logic: severe High risk: CSF3R—enhanced
HAX1 (AR) ra­tion arrest and neutropenia at birth, recur­ –7/7q– and prolonged
GFI1 (AD) apo­pto­sis rent infec­tions, MDS, AML Complex response to G-CSF
G6PC3 (AR)
JAGN1 (AR) High risk:
TCIRG1 (AD) RUNX1
WAS (XR)
CSF3R (AD, AR)
Short telo­mere TERT (AD, AR)—m/c Telomere short­en­ing Hematologic/onco­logic: High risk: Somatic rever­sion—
syn­drome TERC (AD) throm­bo­cy­to­pe­nia, ane­mia, –7/7q– back muta­tion
DKC1 (XR) neutropenia, pan­cy­to­pe­ Complex of DKC1, TERT
NOLA3/NOP10 (AR) nia, MDS, AML, immu­no­ pro­moter GOF
NOLA2/NHP2 (AR) de­fi­ciency, squa­mous cell muta­tion, mitotic
TINF2 (AD) car­ci­noma recom­bi­na­tion
WRAP53/TCAB1 (AR) Nonhematopoietic: the muco­
CTC1 (AR) cu­ta­ne­ous triad (retic­u­lated High risk:
RTEL1 (AD, AR) skin ­pig­men­ta­tion, nail dys­ TP53a
ACD/TPP1 (AD, AR) tro­phy, oral leukoplakia), short
PARN (AD, AR) stat­ure, low birth weight,
NAF1 (AD) fail­ure to thrive, hear­ing loss,
STN1 (AD) ret­i­nop­a­thy, muco­sal stric­
NPM1 (AD) tures, pul­mo­nary fibro­sis, liver
ZCCHC8 (AD) fibro­sis

Prakash Singh Shekhawat

Clonal hema­to­poi­e­sis in IBMFS  |  391
Table 1. Inherited BMF syndromes (Continued)

Causative genes Frequent chro­mo­ Frequent somatic

Syndrome ­(inher­i­tance) Etiology Clinical man­i­fes­ta­tions somal abnor­mal­i­ties muta­tions
DBA RPS19 (AD)—m/c Ribosomopathy; red Hematologic/onco­logic: Abnormal cyto­ge­net­ TP53,a PPM1D, ASXL1
RPL3 (AD) blood cell aplasia Anemia, reticulocytopenia, ics infre­quent (small num­ber of
RPL5 (AD) MDS, AML, solid tumors cases)b
RPL7 (AD) (oste­o­genic sar­coma, lung,
RPL11 (AD) colon and cer­vix)
RPL14 (AD) Nonhematologic: low birth
RPL15 (AD) weight, growth retar­da­tion,
RPL18 (AD) devel­op­men­tal delay, short
RPL19 (AD) stat­ure, micro­ceph­aly, other
RPL23 (AD) cra­nio­fa­cial malformation,
RPL26 (AD) con­gen­i­tal glau­coma or cat­

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/390/1852164/390choijilsuren.pdf by guest on 13 December 2021

RPL27 (AD) a­ract, stra­bis­mus, short neck,
RPL31 (AD) thumb abnor­mal­i­ties, horse­
RPL35a (AD) shoe kid­ney, hypo­spa­dias,
RPL36 (AD) car­diac abnor­mal­i­ties
PRS7 (AD)
RPS8 (AD)
RPS10 (AD)
RPS15 (AD)
RPS17 (AD)
RPS24 (AD)
RPS26 (AD)
RPS27 (AD)
RPS27A (AD)
RPS28 (AD)
RPS29 (AD)
GATA1 (XR)
TSR2 (XR)
SAMD9/SAMD9L SAMD9 (AD) Growth inhi­bi­tion MIRAGE syn­drome –7/7q– Somatic rever­sion
dis­or­ders SAMD9L (AD) Hematologic/onco­logic: 5q– by LOF SAMD9 or
Thrombocytopenia, ane­mia, SAMD9L muta­tions
MDS with mono­somy 7, ETV6, RUNX1, SETBP1
recur­rent infec­tions, (small num­ber of
bleed­ing cases)
Nonhematologic: intra­uter­ine
growth restric­tion, devel­
op­men­tal delay, adre­nal
hypo­pla­sia, chronic diar­rhea,
gen­i­tal anom­a­lies
Ataxia-pan­cy­to­pe­nia syn­drome
Hematologic/onco­logic:
Pancytopenia, ane­mia, MDS
with mono­somy 7/del(7q),
AML, recur­rent infec­tions
Nonhematologic: ataxia, cer­
e­bel­lar atro­phy, ret­i­nal dys­
func­tion, behav­ioral abnor­
mal­i­ties, alve­o­lar proteinosis
The dif­fer­en­tial effects of biallelic vs het­ero­zy­gous TP53 muta­tions to leu­ke­mic trans­for­ma­tion have not been stud­ied in IBMFS other than SDS.
a

Whether somatic muta­tions con­trib­ute to malig­nant trans­for­ma­tion in DBA has not been fully inves­ti­gated.
b

m/c, most com­mon; AD, auto­so­mal dom­i­nant; AR, auto­so­mal reces­sive; GOF, gain of func­tion; XR, X-linked reces­sive.

war­rants empir­ic
­ al treat­ments before frank cytopenias or myel­
odysplastic syn­ dromes (MDS)/acute mye­ loid leu­ke­
mia (AML)
arise. For exam­ple, CSF3R muta­tions in severe con­gen­i­tal neutro­
penia (SCN) or het­ero­zy­gous TP53 muta­tions in SDS can per­sist
for years with­out pro­gres­sion to malig­nancy, even in the set­ting
of mul­ti­ple muta­tions in a given patient. Determination of ideal
tim­ing for HSCT in the set­ting of CH in IBMFS requires fur­ther
stud­ies.
Somatic rever­ sion is the acquired alter­ ation of inherited
path­o­genic muta­tion, resulting in the par­tial or full func­tional
recov­ery of the affected pro­tein. It has been reported in many
inherited genetic dis­ or­
ders, and poten­ tial mech­ an­isms have
been reported as back muta­tion, intra­genic recom­bi­na­tion, sec­
ond site muta­tions suppressing the effect of the orig­in ­ al muta­
tion, frame-restor­ing sec­ond site inser­tions, and dele­tions.6 It
can con­trib­ute to CH by con­fer­ring sur­vival ben­e­fit to reverted
clones, but the expan­sion of hema­to­poi­etic clones is not always
asso­ci­ated with higher pro­pen­sity for leu­ke­mic trans­for­ma­
tion. This may be due to the over­all improved fit­ness of HSPC
by correcting under­ly­ing func­tional defects while keep­ing the

392  |  Hematology 2021  |  ASH Education Program

 At diagnosis
- CBC
- BM with cytogenetics/FISH, NGS myeloid panel
- germline genetic testing
- chromosome breakage test for FA
- telomere length (flow-FISH) for STS
- red cell adenosine deaminase for DBA
- pancreatic isoamlyase and/or trypsinogen for SDS
- HLA typing

Risk Stratification for Surveillance

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/390/1852164/390choijilsuren.pdf by guest on 13 December 2021

Low-risk patients
- stable and mild cytopenias
- transfusion independent
- no BM dysplasia
- no cytogenetic abnormalities
Intermediate-risk patients
- mild or moderate cytopenias, but stable
- transfusion independent
- no BM dysplasia
- non-high-risk cytogenetic abnormalities
High-risk patients
- falling or increasing counts
- severe cytopenias
- transfusion dependent
- BM dysplasia or increased blasts
- high-risk cytogenetics/FISH (del7q/-7, complex;
RUNX1 translocation or +3q in FA)
- biallelic TP53 # or RUNX1 mutations

CBC every 3-6 months CBC every 1-3 months

BM every 1 year BM every 3-6 months
NGS panel every 1-3 years* NGS panel every 1 year*
Refer to HSCT for advanced planning CBC every 1-2 months
Immediate BM and repeat every 3-6 months
HSCT with the best available donor

Figure 1. Surveillance recommendations for patients with IBMFS based on the risk of malignant transformation. Detailed diagnostic
workup for BMF is described by DeZern and Churpek.7 Once diagnosed, patients are followed by CBC with differential, BM aspiration
and biopsy with cytogenetics, FISH, and NGS myeloid panel at regular intervals. The frequency of follow-up depends on the disease
and mutation types, baseline counts and BM findings, and the presence of interval changes on subsequent visits. Although the NGS
myeloid panel is not a part of standard practice for the management of IBMFS, it is becoming more widely used with advancement of
our understanding in CH in both IBMFS and the general population. NGS myeloid panel results are particularly useful to select patients
who are at high risk of malignant transformation without overt clinical signs of disease progression. We can increase the frequency
of surveillance and initiate discussion and planning for early intervention for the best possible outcome in these high-risk patients.
*The optimal frequency and the best genomic source (PB vs BM) of NGS myeloid panel testing have not been fully established.
#
Although biallelic TP53 mutations are known high-risk features, conventional NGS techniques do not readily provide allelic status of a
given mutation. The severity of BMF is defined as the following: mild—ANC <1500/µL, hemoglobin ≥8 g/dL, or platelets 50-150 K/µL;
moderate—ANC <1000/µL, hemoglobin <8 g/dL, or platelets <50 K/µL; severe—ANC <500/µL, hemoglobin <8 g/dL, or platelets
<30 K/µL. ANC, absolute neutrophil count; CBC, complete blood count; FISH, fluorescence in situ hybridization; PB, peripheral blood.

cell cycle check­point intact, most impor­tantly p53, which can
actively sup­press tumor­i­gen­e­sis.
In this review, the term CH will be used to describe a phe­nom­
e­non where any HSPC clone with somatic geno­mic alter­ations—
includ­ing sin­gle-nucle­o­tide var­i­ants, indels, large dele­tions, or
gross chro­mo­somal abnor­mal­i­ties—con­trib­utes to a larger pool
of mature blood cells than what is expected for any sin­gle HSPC
clone under homeo­sta­sis.
in the mye­ loid lin­
e­age, but no frank dys­ pla­
sia is seen and
blasts are not increased. Cytogenetics showed del(20q) in 2 of
20 cells. Peripheral blood mye­loid neo­plasm panel sequenc­ing
result reveals 2 muta­tions in the EIF6 gene: D112N (var­i­ant alle­lic
fre­quency [VAF] 5%) and N106S (VAF 2%) and 2 muta­tions in the
TP53 gene: R248W (VAF 1%) and R110L (VAF 0.4%). The hema­
tol­o­gist decides to mon­i­tor his counts every 3 months, as well
as BM every 3 to 6 months, and repeat blood mye­loid neo­plasm
panel sequenc­ing in 1 year to mon­i­tor the TP53 mutant clones.

CLINICAL CASE Shwachman-Diamond syn­drome

A 21-year-old man with SDS presented to clinic for a 6-month SDS is char­ac­ter­ized by con­gen­i­tal anom­a­lies; BMF, with neutro­
fol­low-up. The patient has mild neutropenia and mild throm­ penia being the most prominent fea­ture; and exo­crine pan­cre­
bo­cy­to­pe­nia, and blood counts have been sta­ble in the past atic insuf­fi­ciency.8 Biallelic muta­tions in the SBDS gene are most
6 months. The BM is hypocellular with mild dysmorphologies com­mon (>90%). Less fre­quently, muta­tions in DNAJC21, SRP54,

Prakash Singh Shekhawat

Clonal hema­to­poi­e­sis in IBMFS  |  393
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/390/1852164/390choijilsuren.pdf by guest on 13 December 2021
Figure 2. Clonal hematopoiesis in IBMFS. Underlying genetic mutations lead to ineffective and stressed hematopoiesis, and u ­ ltimately
HSPC attrition. Genetic variations occurred during lifetime of HSPC may confer a growth advantage if it corrected underlying genetic
and/or cellular defects by somatic reversion. Alternatively, bypassing cell cycle check point by acquiring high risk genetic features,
such as loss of p53, may confer a growth advantage. Clones that reverted underlying defects may have competitive advantage and
higher fitness, resulting in improved counts and stable hematologic course. Whereas clones with high risk somatic alterations (eg,
TP53 mutations or monosomy 7) may result in clonal expansion without functional recovery of underlying defects. These clones are
more prone to acquire additional secondary mutations and/or progress to MDS/AML.

and EFL1 pres­ent with clin­i­cal fea­tures of SDS. About 20% to 30%
of patients develop MDS/AML, which is fre­quently accom­pa­nied
by com­plex kar­yo­type.2,9-11
A recent sequenc­ing study shed light on our under­stand­ing
of the mech­a­nisms of clonal evo­lu­tion in SDS. Clonal hema­to­poi­
e­sis was com­mon in patients with SDS (72% among those with­
out frank MDS/AML).12 Interestingly, mutated genes in SDS were
dif­fer­ent from those of the gen­eral pop­u­la­tion. Mutations in EIF6
were the most com­mon (59%), followed by TP53 (39.8%) and
CSNK1A1 (7.2%), whereas muta­tions in DNMT3A, TET2, or ASXL1
were less fre­quent. The most com­monly observed chro­mo­somal
abnor­mal­ity was also del(20q) (47%), which is the loca­tion for
the EIF6 gene. In another study, CH was observed in 59% of
patients, and TP53 muta­tions were the most fre­quent (48%) (of
note, EIF6 was not included in the gene panel in this study).13 The
prev­a­lence of TP53 muta­tions increased with age; about 80%
of patients older than 10 years had at least one TP53 muta­tion,
and fre­quently two or more TP53 mutations were dis­cov­ered in
a given patient. The pres­ence of TP53 muta­tions among patients
with SDS is not an impending sign of leu­ke­mic trans­for­ma­tion,
even though TP53 muta­ tions are fre­
quently encoun­ tered in
MDS/AML and por­tend a poor prog­no­sis.

394  |  Hematology 2021  |  ASH Education Program

 The SBDS pro­tein coop­er­ates with the GTPase EFL1 to release
EIF6 from the nascent 60S ribo­somal sub­unit in the cyto­sol, so
that the 60S sub­unit is free to asso­ci­ate with the 40S sub­unit and
form the mature 80S ribo­some.14 Therefore, the loss-of-func­tion
(LOF) EIF6 var­i­ants would result in increased avail­abil­ity of the 60S
sub­unit for mat­u­ra­tion and com­pen­sate for the defec­tive SBDS.
In fact, EIF6 var­i­ants were fre­quently found in patients with SDS,
which was asso­ci­ated with clonal expan­sion due to improved
trans­ la­
tional effi­
ciency and stem cell fit­ ness. While func­ tional
com­pen­sa­tion for pri­mary ribo­somal defects improved sur­vival of
HSPC in SDS, it did not nec­es­sar­ily increase the risk of malig­nant
trans­for­ma­tion. On the other hand, biallelic TP53 muta­tions were
almost always observed in patients with leu­ke­mic trans­for­ma­
tion. P53 is a tran­scrip­tion fac­tor that medi­ates cell cycle arrest
and apo­pto­sis in response to cel­lu­lar stress sig­nals. Biallelic LOF
muta­tions in TP53 allow HSPCs to prog­ress through cell cycle
and escape apo­pto­sis even in the pres­ence of crit­i­cal cel­lu­lar
stresses posed by ribosomopathy, thereby con­fer­ring a growth
advan­tage. Single-cell DNA sequenc­ing on serial sam­ples from
patients who devel­oped AML revealed that patients can have
mul­ti­ple inde­pen­dent TP53 mutant clones for a long time, and
the par­tic­u­lar clone car­ry­ing biallelic muta­tions can give rise to
leu­ke­mia over sev­eral years.12 The pres­ence of biallelic muta­tions
usu­ally can­not be assessed in the con­ven­tional bulk sequenc­ing,
which only informs VAF but not alle­lic sta­tus of a given muta­tion
in cells. This raises a ques­tion whether performing sin­gle-cell DNA
sequenc­ing reg­u­larly on patients with known TP53 mutant clones
can help select patients with biallelic TP53 muta­tions who are at
risk of leu­ke­mic trans­for­ma­tion. Of course, this expen­sive mon­i­
tor­ing approach will need to be care­fully assessed in clin­i­cal tri­als
to con­firm effi­cacy and cost-effec­tive­ness.
DBA is another inherited ribosomopathy due to het­ero­zy­gous
muta­tions in the com­po­nents of either the small 40S or large 60S
ribo­somal sub­units.8 It is char­ac­ter­ized by red blood cell aplasia
and con­gen­i­tal anom­aly. The risk of MDS/AML is rel­a­tively low
com­pared with other IBMFS discussed here, with a cumu­la­tive
inci­dence of AML of 5% by the age of 46 years.15 The muta­tional
spec­trum lead­ing to MDS/AML has not been sys­tem­at­i­cally or
func­ tion­ally stud­ied in DBA. Somatic muta­ tions in TP53 and
PPM1D in MDS have been reported in spo­radic case reports,16-18
which raises a pos­ si­
bil­
ity that sim­ i­
lar mech­ a­
nisms may drive
malig­nant trans­for­ma­tion in DBA and SDS.
Fanconi ane­mia
FA is the most com­mon IBMFS. At least 22 genes have been iden­
ti­fied (FANCA-FANCW).19,20 The under­ly­ing cel­lu­lar defect is the
inabil­ity to repair DNA cross-links, lead­ing to apo­pto­sis via the
p53 path­way, or trans­for­ma­tion if p53 is inactivated.21 Patients
have con­gen­i­tal anom­a­lies, BMF, and can­cer pre­dis­po­si­tion,
most nota­bly squa­mous cell car­ci­noma of the head and neck.20
Patients have a 30% to 40% cumu­la­tive risk of mye­loid malig­
nan­cies by age 40 years.9,22 Myeloid malig­nan­cies often arise
from clones with abnor­mal kar­yo­types, char­ac­ter­is­ti­cally unbal­
anced trans­lo­ca­tions lead­ing to gains or losses of chro­mo­somes.
Gain of chro­mo­some 1q is the most com­mon kar­yo­typic abnor­
mal­ity in FA, which can be seen with­out MDS/AML and does not
indi­cate impending trans­for­ma­tion.23 Conversely, mono­ somy
7/del(7q) are signs of impending trans­ for­
ma­tion, if not trans­
formed already, which requires urgent HSCT. Gain of chro­mo­
some 3q, which is asso­ci­ated with increased expres­sion of EVI1,
Prakash Singh Shekhawat
is also highly char­ac­ter­is­tic in FA and fre­quently pre­cedes mono­
somy 7/del(7q).24-26 RUNX1 abnor­mal­i­ties, includ­ing cryp­tic trans­
lo­ca­tion, also indi­cate high-risk of trans­for­ma­tion.23
Single-nucle­o­tide poly­mor­phism array anal­y­sis is one of the
non­in­va­sive mon­i­tor­ing meth­ods that can iden­tify chro­mo­somal
abnor­mal­i­ties from blood sam­ples. From 130 patients with FA
(mean age, 14.5 years; range, 0-50 years), sin­gle-nucle­o­tide poly­
mor­phism array detected chro­mo­somal mosaic events (CMEs) in
16 patients (12.3%).27 Nine of 16 patients car­ried mul­ti­ple CMEs.
The pres­ence of CMEs was asso­ci­ated with higher inci­dence of
hema­to­logic and solid can­cers, as well as increased risk of death.
Large-scale next gen­er­a­tion sequenc­ing (NGS) stud­ies have
not been published in FA; there­fore, it is dif­fi­cult to spec­u­late the
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/390/1852164/390choijilsuren.pdf by guest on 13 December 2021
con­tri­bu­tion of somatic muta­tions to clonal evo­lu­tion. Prior stud­
ies using Sanger sequenc­ ing of selected onco­ genes showed
largely neg­a­tive results, but that may be due to the small num­
ber of genes included in the panel and Sanger sequenc­ing not
being sen­si­tive enough.23
More recently, targeted mye­loid panel sequenc­ing performed
on 16 patients with FA with MDS/AML revealed 10 patients had
detect­able somatic muta­tions, most fre­quently involv­ing RUNX1,
and the RAS path­ way genes (KRAS and PTPN11).28 Of these
16 patients, 12 patients had either chro­mo­some 3 or 7 abnor­
mal­i­ties and/or RUNX1 muta­tions, confirming the prior reports.
Another NGS study of 7 patients with FA (includ­ing 1 with AML)
also con­firmed fre­quent CH from a young age, vir­tu­ally observed
in all­patients tested, regard­less of the pres­ence of periph­eral
blood cytopenia or leu­ke­mia.29 One AML sam­ple in that study
car­ried mul­ti­ple MDS/AML muta­tions, includ­ing TP53. Although
a clear pat­tern of clonal evo­lu­tion can­not be deter­mined from
these stud­ ies due to small sam­ ple sizes, a rel­ a­
tive lack of
DNMT3A, TET2, and SF3B1 was observed in both stud­ies, sug­
gesting the dif­fer­ent somatic muta­tional spec­trum in FA.
Approximately 15% to 18% of patients with FA are reported
to have somatic rever­sion. It fre­quently results in CH, because
recov­ery of the pro­tein func­tion often con­fers a growth advan­
tage. It can result in improved counts and sta­ble hema­to­logic
course, with lon­ger sur­vival, lower risk of MDS/AML, and less like­
li­hood to require HSCT.30-32 However, malig­nant trans­for­ma­tion
in coexisting nonrevertant FA cells may still occur, requir­ing reg­
u­lar monitoring.30
Severe con­gen­i­tal neutropenia
SCN is char­ac­ter­ized by severe neutropenia from birth, com­pli­
cated by recur­rent infec­tions. Treatment with granulocyte colony-
stimulating factor (G-CSF) can increase the cir­cu­lat­ing neu­tro­phil
counts and is effec­tive for pre­ven­tion of recur­rent infec­tions. The
most com­mon cause is the ELANE gene muta­tions (∼50%), fol­
lowed by HAX1 and G6PC3 (10%-20%).13
About 20% of patients tak­ing G-CSF develop MDS/AML in
their life­time.33,34 Somatic muta­ tions in CSF3R (encoding the
G-CSF recep­tor) and RUNX1 have been fre­quently observed in
those who devel­oped MDS/AML.35 From a targeted sequenc­ing
study of 40 patients, CH was fre­quently observed (62%; mean
age, 16.6 years).13 As expected, the most fre­quently observed
muta­tion was the CSF3R muta­tion (40%).13 All CSF3R var­i­ants
were non­sense muta­tions resulting in trun­ca­tion of cyto­plas­mic
domain of the G-CSF recep­tor, lead­ing to prolonged and hyper­
ac­tive sig­nal­ing in response to G-CSF. Therefore, HSPCs car­ry­ing
CSF3R muta­tions have a growth advan­tage in the pres­ence of
Clonal hema­to­poi­e­sis in IBMFS  |  395
G-CSF. Independent CSF3R-mutant clones can coex­ist, expand
over many years, or some­times dis­ap­pear.
Another very fre­quently observed muta­tion in MDS/AML was
the RUNX1 muta­tion (64.5%), and in fact, approx­i­ma­tely 80% of
them also had the con­cur­rent CSF3R muta­tions.33 Serial eval­u­
a­tion of a patient who devel­oped AML with CSF3R and RUNX1
muta­tions sug­gests that the CSF3R muta­tions were likely ini­ti­
at­ing events con­fer­ring a pro­lif­er­a­tive advan­tage, and an addi­
tional RUNX1 muta­tion later in the dis­ease pro­gres­sion led to a
block in dif­fer­en­ti­a­tion and fueled leu­ke­mo­gen­e­sis.
Short telo­mere syn­drome
Telomere length is reg­ u­lated by coop­ er­
a­
tion of telomerases
and telomeric pro­ tein complexes (shelterin and CST [CTC1-
STN1-TEN1]).36 Germline muta­tions in telo­mere com­po­nents may
result in short telo­mere length, which is asso­ci­ated with BMF,
hepatic and pul­mo­nary fibro­sis, and can­cer pre­dis­po­si­tion.37
The telo­meres are short­ened on each cell divi­sion due to an
“end-rep­li­ca­tion prob­lem,” where 5′ to 3′ DNA poly­mer­ases can
only extend the DNA strand in one direc­tion in the pres­ence of a
tem­plate sequence, leav­ing the extreme 5′ end of the chro­mo­
some as a gap in rep­li­ca­tion.38 Critically short telo­mere lengths,
below the first per­cen­tile for age-matched con­trols, cause cel­
lu­lar senes­cence and pre­ma­ture cell death, as well as a clin­i­cal
syn­drome known as STS (also known as DC or telo­mere biol­ogy
dis­or­der). Patients exhibit a wide spec­trum of dis­ease man­i­fes­ta­
tions, from clas­si­cal DC with a muco­cu­ta­ne­ous triad (oral leuko­
plakia, lacy retic­u­lar skin hyper­pig­men­ta­tion, and nail dys­tro­phy)
and early onset BMF to adult-onset MDS/AML, pul­mo­nary fibro­
sis, or liver dis­ease with­out appar­ent BMF.39
All MDS/AML cases in STS were asso­ci­ated with abnor­mal
kar­yo­types, most com­monly mono­somy 7 (56%).40 The somatic
muta­ tional land­scapes in these patients (n  = 
14; median age
53 years) were sim­i­lar to that of MDS/AML in the gen­eral pop­
u­la­tion. In patients with­out MDS/AML (median age, 59 years),
6 of 20 eval­ua ­ ted patients had CH (30%) with the mean VAF of
8.2% (range, 2.2%-32.9%). Mutations in TP53 (n = 3) were most
com­monly observed, followed by TET2 (n = 2).40 On the other
hand, 2 other stud­ies with slightly youn­ger indi­vid­u­als reported
that somatic muta­tions asso­ci­ated with MDS/AML were rarely
observed, whereas CH was com­mon.41,42 Whether the pres­ence
of somatic muta­tions dis­cov­ered by NGS is asso­ci­ated with an
increased risk of malig­nant trans­for­ma­tion, with­out other coex­
isting high-risk clinical features, is not known at this time.
Somatic rever­sion of the telomerase activ­ity in the periph­eral
blood has been reported. Examples include acquired gain-of-
func­tion muta­tions in the TERT pro­moter region in patients with
the TERT or PARN het­ero­zy­gous muta­tions, lead­ing to overex­
pression of the wild-type allele.43 A back muta­tion in DKC142 and
mitotic recom­bi­na­tion of the TERC gene resulting in isodisomy
of the wild-type allele in patients with germline het­ero­zy­gous
TERC muta­tions have also been reported.44 These somatic rever­
sion events can sta­bi­lize the telo­mere lengths and allow clonal
expan­sion of the reverted clones.
SAMD9/SAMD9L dis­or­ders
Germline muta­tions in SAMD9 and SAMD9L cause MIRAGE (myel­
odysplasia, infec­tion, restric­tion of growth, adre­nal hypo­pla­sia,
gen­i­tal phe­no­types, and enter­op­a­thy) syn­drome and ataxia pan­
cy­to­pe­nia syn­drome, respec­tively.45,46 SAMD9 and SAMD9L are
396  |  Hematology 2021  |  ASH Education Program
paralogs on the chro­mo­some 7q21.2. Both SAMD9 and SAMD9L
pro­teins are endosome fusion facil­i­ta­tors, and they play a role
in deg­ra­da­tion of growth fac­tor recep­tors by inter­nal­i­za­tion.
Gain-of-func­tion muta­tions in SAMD9 or SAMD9L have a growth-
inhib­i­tory effect by reduc­ing Raf/MEK/ERK sig­nal­ing, and loss
of mutated allele con­ fers growth advan­ tage. Mutant allele is
lost fre­quently by acquired chro­mo­some 7 loss (mono­somy 7 or
del(7q)) or con­com­i­tant somatic LOF muta­tions.47,48
Conclusion
Clonal hema­ to­
poi­e­sis is com­
monly observed in patients with
IBMFS at a much youn­ger age than in the gen­eral pop­u­la­tion. The
muta­tional spec­trum is also dif­fer­ent in IBMFS: epi­ge­netic reg­u­la­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/390/1852164/390choijilsuren.pdf by guest on 13 December 2021
tors, such as DNMT3A, TET2, and ASXL1, the most fre­quently mu­
tated genes in the aging pop­u­la­tion, are rel­a­tively rare—except
in the older patients with STS. Instead, somatic geno­mic alter­
ations that can restore the orig­i­nal func­tion of pro­tein (eg, somat­
ic rever­sion of the caus­a­tive muta­tion or com­pen­sa­tory sec­ond
muta­tions) or that allow path­o­logic adap­ta­tion for sur­vival (eg,
by inac­ti­va­tion of p53 or loss of mutant allele) are pref­er­en­tially
selected for clonal expan­sion (Figure 2). Reduced HSPC num­
bers and fit­ness at base­line also cre­ate an envi­ron­ment in which
clones with such genetic alter­ations out­com­pete other HSPCs
and quickly estab­lish clonal dom­i­nance. However, based on cur­
rently avail­­able evi­dence, CH itself, unless accom­pa­nied by high-
risk cyto­ge­netic abnor­mal­i­ties, pro­gres­sive cytopenias, and/or
BM mor­pho­logic changes, does not require imme­di­ate defin­i­tive
treat­ments even in the set­ting of IBMFS (Table 1). Future stud­
ies should address whether pre­emp­tive inter­ven­tions based on
high-risk CH fea­tures improve patient out­come in clin­i­cal tri­als.
CLINICAL CASE (Con­tin­ued)
This patient car­ries the 2 most com­mon somatic muta­tions in
SDS, EIF6 and TP53, and these muta­tions are likely aris­ing from
inde­pen­dent clones. Deletion 20q is also a com­mon cyto­ge­
netic find­ing that results in the loss of EIF6. The pres­ence of TP53
muta­tions is a high-risk fea­ture; how­ever, given the small clone
sizes and sta­ble blood counts and BM mor­phol­ogy, impending
leu­ke­mic trans­for­ma­tion is unlikely. Although there is no avail­­able
con­sen­sus guide­line, more fre­quent fol­low-up with blood counts
and BM exam­i­na­tion along with serial NGS panel sequenc­ing is
desir­able for early detec­tion of dis­ease pro­gres­sion.
Acknowledgments 
This work was supported in part by National Heart Lung and
Blood Institute grant K99 HL150628 (M.J.). Authors thank Dr. Robert
Brodsky and Dr. Amy DeZern for their constructive feedback.
Conflict-of-inter­est dis­clo­sure
Haruna Batzorig Choijilsuren: no con­flicts to dis­close.
Yeji Park: no con­flicts to dis­close.
Moonjung Jung: no con­flicts to dis­close.
Off-label drug use
Haruna Batzorig Choijilsuren: nothing to disclose.
Yeji Park: nothing to disclose.
Moonjung Jung: nothing to disclose.
Correspondence
Moonjung Jung, Division of Hematology, Department of Medi­
cine, School of Medicine, Johns Hopkins University, 720 Rutland
Ave, Ross Research Building Room 1032B, Baltimore, MD 21205;
e-mail: mjung@jhmi​­.edu.
References
1. Shimamura A. Aplastic ane­mia and clonal evo­lu­tion: germ line and somatic
genet­ics. Hematology Am Soc Hematol Educ Program. 2016;2016(1):
74-82.
2. Myers KC, Furutani E, Weller E, et  al. Clinical fea­tures and out­comes of
patients with Shwachman-Diamond syn­drome and myelodysplastic syn­
drome or acute mye­loid leu­kae­mia: a multicentre, ret­ro­spec­tive, cohort
study. Lancet Haematol. 2020;7(3):e238–e246.
3. Giardino S, de Latour RP, Aljurf M, et al; Severe Aplastic Anemia and Chronic
Malignancies Working Parties of Euro­pean Blood and Marrow Transplanta­
tion group. Outcome of patients with Fanconi ane­mia devel­op­ing myel­
odysplasia and acute leu­ke­mia who received allo­ge­neic hema­to­poi­etic
stem cell trans­plan­ta­tion: a ret­ro­spec­tive anal­y­sis on behalf of EBMT group.
Am J Hematol. 2020;95(7):809-816.
4. Jaiswal S, Ebert BL. Clonal hema­to­poi­e­sis in human aging and dis­ease. Sci-
ence. 2019;366(6465):eaan4673.
5. Tsai FD, Lindsley RC. Clonal hema­to­poi­e­sis in the inherited bone mar­row
fail­ure syn­dromes. Blood. 2020;136(14):1615-1622.
6. Hirschhorn R. In vivo rever­sion to nor­mal of inherited muta­tions in humans.
J Med Genet. 2003;40(10):721-728.
7. DeZern AE, Churpek JE. Approach to the diagnosis of aplastic anemia.
Blood Adv. 2021;5(12):2660-2671.
8. Shimamura A, Alter BP. Pathophysiology and man­age­ment of inherited
bone mar­row fail­ure syn­dromes. Blood Rev. 2010;24(3):101-122.
9. Alter BP, Giri N, Savage SA, Rosenberg PS. Cancer in the National Cancer
Institute inherited bone mar­row fail­ure syn­drome cohort after fif­teen years
of fol­low-up. Haematologica. 2018;103(1):30-39.
10. Cada M, Segbefia CI, Klaassen R, et al. The impact of cat­e­gory, cytopathol­
ogy and cyto­ge­net­ics on devel­op­ment and pro­gres­sion of clonal and malig­
nant mye­loid trans­for­ma­tion in inherited bone mar­row fail­ure syn­dromes.
Haematologica. 2015;100(5):633-642.
11. Donadieu J, Fenneteau O, Beaupain B, et al; Associated inves­ti­ga­tors of
the French Severe Chronic Neutropenia Registry. Classification of and
risk fac­ tors for hema­ to­
logic com­ pli­
ca­tions in a French national cohort
of 102 patients with Shwachman-Diamond syn­ drome. Haematologica.
2012;97(9):1312-1319.
12. Kennedy AL, Myers KC, Bowman J, et al. Distinct genetic path­ways define
pre-malig­nant ver­sus com­pen­sa­tory clonal hema­to­poi­e­sis in Shwachman-
Diamond syn­drome. Nat Commun. 2021;12(1):1334.
13. Xia J, Miller CA, Baty J, et al. Somatic muta­tions and clonal hema­to­poi­e­sis
in con­gen­i­tal neutropenia. Blood. 2018;131(4):408-416.
14. Warren AJ. Molecular basis of the human ribosomopathy Shwachman-
Diamond syn­drome. Adv Biol Regul. 2018;67:109-127.
15. Vlachos A, Rosenberg PS, Atsidaftos E, Alter BP, Lipton JM. Incidence of
neo­pla­sia in Diamond Blackfan ane­mia: a report from the Diamond Blackfan
Anemia Registry. Blood. 2012;119(16):3815-3819.
16. Lindsley RC, Saber W, Mar BG, et  al. Prognostic muta­tions in myelodys­
plastic syn­drome after stem-cell trans­plan­ta­tion. N Engl J Med. 2017;376(6):
536-547.
17. Simkins A, Bannon SA, Khoury JD, et al. Diamond-blackfan ane­mia predis­
posing to myelodysplastic syn­drome in early adult­hood. JCO Precis Oncol.
2017;22(1):1-5.
18. Schaefer EJ, Lindsley RC. Significance of clonal muta­tions in bone mar­
row fail­ure and inherited myelodysplastic syn­drome/acute mye­loid leu­ke­
mia pre­dis­po­si­tion syn­dromes. Hematol Oncol Clin North Am. 2018;32(4):
643-655.
19. Knies K, Inano S, Ramírez MJ, et al. Biallelic muta­tions in the ubiquitin ligase
RFWD3 cause Fanconi ane­mia. J Clin Invest. 2017;127(8):3013-3027.
20. Nalepa G, Clapp DW. Fanconi anae­mia and can­cer: an intri­cate rela­tion­
ship. Nat Rev Cancer. 2018;18(3):168-185.
21. Ceccaldi R, Parmar K, Mouly E, et al. Bone mar­row fail­ure in Fanconi ane­mia
is trig­gered by an exac­er­bated p53/p21 DNA dam­age response that impairs
hema­to­poi­etic stem and pro­gen­i­tor cells. Cell Stem Cell. 2012;11(1):36-49.
Prakash Singh Shekhawat
22. Kutler DI, Singh B, Satagopan J, et  al. A 20-year per­spec­tive on the
International Fanconi Anemia Registry (IFAR). Blood. 2003;101(4):1249-
1256.
23. Quentin S, Cuccuini W, Ceccaldi R, et al. Myelodysplasia and leu­ke­mia of
Fanconi ane­mia are asso­ci­ated with a spe­cific pat­tern of geno­mic abnor­
mal­i­ties that includes cryp­tic RUNX1/AML1 lesions. Blood. 2011;117(15):
e161–e170.
24. Tönnies H, Huber S, Kuhl J-S, Gerlach A, Ebell W, Neitzel H. Clonal chro­mo­
somal aber­ra­tions in bone mar­row cells of Fanconi ane­mia patients: gains
of the chro­mo­somal seg­ment 3q26q29 as an adverse risk fac­tor. Blood.
2003;101(10):3872-3874.
25. Cioc AM, Wagner JE, MacMillan ML, DeFor T, Hirsch B. Diagnosis of myel­
odysplastic syn­drome among a cohort of 119 patients with Fanconi ane­
mia: mor­pho­logic and cyto­ge­netic char­ac­ter­is­tics. Am J Clin Pathol.
2010;133(1):92-100.
26. Meyer S, Bristow C, Wappett M, et al. Fanconi ane­mia (FA)-asso­ci­ated 3q
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/390/1852164/390choijilsuren.pdf by guest on 13 December 2021
gains in leu­ke­mic trans­for­ma­tion con­sis­tently tar­get EVI1, but do not affect
low TERC expres­sion in FA. Blood. 2011;117(22):6047-6050.
27. Reina-Castillón J, Pujol R, López-Sánchez M, et al. Detectable clonal mosa­
i­cism in blood as a bio­marker of can­cer risk in Fanconi ane­mia. Blood Adv.
2017;1(5):319-329.
28. Chao MM, Thomay K, Goehring G, et al. Mutational spec­trum of Fanconi
ane­mia asso­ci­ated mye­loid neo­plasms. Klin Padiatr. 2017;229(6):329-
334.
29. Heidemann S, Bursic B, Zandi S, et al. Cellular and molec­u­lar archi­tec­ture
of hema­to­poi­etic stem cells and pro­gen­i­tors in genetic mod­els of bone
mar­row fail­ure. JCI Insight. 2020;5(4):e131018.
30. Gregory JJ Jr, Wagner JE, Verlander PC, et al. Somatic mosa­i­cism in Fanconi
ane­mia: evi­dence of geno­typic rever­sion in lymphohematopoietic stem
cells. Proc Natl Acad Sci U S A. 2001;98(5):2532-2537.
31. Soulier J, Leblanc T, Larghero J, et al. Detection of somatic mosa­i­cism and
clas­si­fi­ca­tion of Fanconi ane­mia patients by anal­y­sis of the FA/BRCA path­
way. Blood. 2005;105(3):1329-1336.
32. Ramírez MJ, Pujol R, Trujillo-Quintero JP, et  al. Natural gene ther­apy by
reverse mosa­ i­
cism leads to improved hema­ tol­
ogy in Fanconi ane­ mia
patients. Am J Hematol. 2021;96(8):989-999.
33. Skokowa J, Steinemann D, Katsman-Kuipers JE, et  al. Cooperativity
of RUNX1 and CSF3R muta­tions in severe con­gen­i­tal neutropenia: a
unique path­way in mye­loid leu­ke­mo­gen­e­sis. Blood. 2014;123(14):2229-
2237.
34. Rosenberg PS, Zeidler C, Bolyard AA, et al. Stable long-term risk of leu­kae­mia
in patients with severe con­gen­i­tal neutropenia maintained on G-CSF ther­
apy. Br J Haematol. 2010;150(2):196-199.
35. Link DC. Mechanisms of leu­ke­mic trans­for­ma­tion in con­gen­i­tal neutrope­
nia. Curr Opin Hematol. 2019;26(1):34-40.
36. Lim CJ, Cech TR. Shaping human telo­meres: from shelterin and CST com­
plexes to telomeric chro­ma­tin orga­ni­za­tion. Nat Rev Mol Cell Biol. 2021;​
22(4):283-298.
37. Calado RT, Young NS. Telomere dis­eases. N Engl J Med. 2009;361(24):2353–
2365.
38. Lingner J, Cooper JP, Cech TR. Telomerase and DNA end rep­li­ca­tion: no
lon­ger a lag­ging strand prob­lem? Science. 1995;269(5230):1533-1534.
39. Schratz KE. Extrahematopoietic man­ i­
fes­
ta­
tions of the short telo­ mere
syn­dromes. Hematology Am Soc Hematol Educ Program. 2020;2020(1):
115-122.
40. Schratz KE, Haley L, Danoff SK, et al. Cancer spec­trum and out­comes in the
Men­de­lian short telo­mere syn­dromes. Blood. 2020;135(22):1946-1956.
41. Kirschner M, Maurer A, Wlodarski MW, et  al. Recurrent somatic muta­
tions are rare in patients with cryp­tic dyskeratosis congenita. Leukemia.
2018;32(8):1762-1767.
42. Perdigones N, Perin JC, Schiano I, et al. Clonal hema­to­poi­e­sis in patients
with dyskeratosis congenita. Am J Hematol. 2016;91(12):1227-1233.
43. Maryoung L, Yue Y, Young A, et al. Somatic muta­tions in telomerase pro­
moter coun­ter­bal­ance germline loss-of-func­tion muta­tions. J Clin Invest.
2017;127(3):982-986.
44. Jongmans MC, Verwiel ET, Heijdra Y, et  al. Revertant somatic mosa­i­cism
by mitotic recom­bi­na­tion in dyskeratosis congenita. Am J Hum Genet.
2012;90(3):426-433.
45. Narumi S, Amano N, Ishii T, et al. SAMD9 muta­tions cause a novel mul­ti­sys­
tem dis­or­der, MIRAGE syn­drome, and are asso­ci­ated with loss of chro­mo­
some 7. Nat Genet. 2016;48(7):792-797.
Clonal hema­to­poi­e­sis in IBMFS  |  397
46. Nagamachi A, Matsui H, Asou H, et al. Haploinsufficiency of SAMD9L, an
endosome fusion facil­i­ta­tor, causes mye­loid malig­nan­cies in mice mim­
ick­ing human dis­eases with mono­somy 7. Cancer Cell. 2013;24(3):305-
317.
47. Wong JC, Bryant V, Lamprecht T, et al. Germline SAMD9 and SAMD9L muta­
tions are asso­ci­ated with exten­sive genetic evo­lu­tion and diverse hema­to­
logic out­comes. JCI Insight. 2018;3(14):e121086.
398  |  Hematology 2021  |  ASH Education Program
48. Nagata Y, Narumi S, Guan Y, et  al. Germline loss-of-func­tion SAMD9 and
SAMD9L alter­ations in adult myelodysplastic syn­ dromes. Blood. 2018;​
132(21):2309-2313.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000271
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/390/1852164/390choijilsuren.pdf by guest on 13 December 2021
 LET THE CHIP(S) FALL WHERE THEY MAY ? BURDENS AND BENEFITS OF DIAGNOSING CLONAL HEMATOPOIESIS

When are idiopathic and clonal cytopenias

of unknown signifcance (ICUS or CCUS)?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/399/1852160/399osman.pdf by guest on 13 December 2021

Afaf E. W. G. Osman
Division of Hematology and Hematologic Malignancies, University of Utah, Salt Lake City, UT

Rapid advances in sequencing technology have led to the identification of somatic mutations that predispose a signifi-
cant subset of the aging population to myeloid malignancies. Recently recognized myeloid precursor conditions include
clonal hematopoiesis of indeterminate potential (CHIP) and clonal cytopenia of unknown significance (CCUS). These
conditions can present diagnostic challenges and produce unwarranted anxiety in some instances. While the risk of pro-
gression to myeloid malignancies is very low in CHIP, true CCUS confers an exponential increase in risk. Idiopathic cyto-
penia of unknown significance (IDUS) lacks the predisposing genetic mutations and has a variable course. In this review
we define the early myeloid precursor conditions and their risk of progression. We present our diagnostic approach to
patients with unexplained cytopenias and discuss the clinical consequences of CHIP and CCUS.

LEARNING OBJECTIVES
• Identify myeloid precursor conditions (CCUS and CHIP) and their clinical consequences
• Understand the diagnostic approach to patients with unexplained cytopenias
• Understand the role of next­generation sequencing in the workup of unexplained cytopenias

Introduction CLINICAL CASE

Cancer precursor states were frst recognized in solid tu­ An 80­year­old man with pancytopenia was referred
mors and led to a better understanding of the multistep to our clinic. He had a past medical history of chronic
processes underlying oncogenesis. In hematology, mono­ obstructive pulmonary disease and recurrent deep vein
clonal gammopathy of undetermined signifcance and thrombosis. His medications included an oral anticoagu­
monoclonal B­cell lymphocytosis are well­characterized lant and antihypertensive therapy. He did not report any
cancer precursor states. These conditions are defned by fatigue, recurrent infections, fever, or night sweats. Initial
the presence of clonally expanded hematopoietic cells, the testing showed the following:
absence of overt malignancy, and increased risk of devel­
oping lymphoid malignancies. In parallel, clonal hema­ Hemogram (normal range) Additional studies (normal range)
topoiesis and clonal cytopenias are emerging as newly
WBC: 3.16 (4.5-11 K/µL) Serum iron: 65 (60-150 mcg/dL)
recognized precursor states that increase the risk of devel­
oping myeloid malignancies, in particular myelodysplastic RBC: 3.14 (4.5-5.9M/µL) Ferritin: 266 (30-400ng/mL)
syndrome (MDS) and acute myeloid leukemia (AML). Rap­ PLT: 86 (150-350 K/µL) RDW: 20.7 (11%-14%)
id advancements in genomic technology have resulted Hgb: 10.2 (13-16.3g/dL) Absolute reticulocyte count: 43.2
in the detection of cancer­associated mutations without (47-152 K/µL)
the morphologic changes characteristic of MDS and AML. MCV: 96 (80-100 fL) TSH, B12, folate, zinc, copper, and lac­
This has led to the identifcation of new precursor states tate dehydrogenase within normal
with increased predisposition to myeloid malignancies. limits
In this brief review, we discuss the defnition, clinical con­ Antinuclear antibody negative
sequences, and management of early myeloid precursor MCV, mean corpuscular volume; RBC, red blood cell; WBC, white
states. blood cell.

Prakash Singh Shekhawat

CHIP, CCUS, and ICUS: defning myeloid precursors | 399
 Morphologic exam­i­na­tion of the bone mar­row showed a nor­
mocellular mar­row with trilineage hema­to­poi­e­sis and no evi­
dence of dys­pla­sia or increased blasts. Flow cytometry did not
show abnor­mal cells. Conventional karyotyping showed 2 abnor­
mal cell lines: 46,XY,del(12)(q24.1),add(22)(q13)[3]/46,XY,add(10)
(q22)[2]/46,XY[18]. Next-gen­er­a­tion sequenc­ing (NGS) from his
bone mar­row revealed the pres­ence of 3 muta­tions:
1 .  TP53 c.743G>A, p.Arg248Gln (NM_000546.5)
  Variant allele fre­quency: 18.6%
2.  U2AF1 c.471G>T, p.Gln157His (NM_006758.2)
  Variant allele fre­quency: 19.7%
3.  DNMT3A c.2173 + 1G>A, p.? (NM_175629.2)
 Variant allele frequency: 19.2% Clonal hematopoiesis of inde­
terminate potential
In 2014, 3 groups of research­ers simul­ta­neously dis­cov­ered the
pres­ence of can­cer-asso­ci­ated muta­tions in the periph­eral blood
of seem­ingly nor­mal older indi­vid­u­als.1-3 The term clonal hema­
to­poi­e­sis of inde­ter­mi­nate poten­tial (CHIP) was coined and re­
fers to the pres­ence of muta­tions asso­ci­ated with hema­to­logic
malig­nan­cies in indi­vid­u­als with nor­mal blood counts and no evi­
dence of hema­to­logic malig­nan­cies (Table 1).4 These muta­tions
rep­re­sent somatic events that lead to the clonal expan­sion of
hema­ to­poi­etic stem cells. ARCH (age-related clonal hema­ to­
poi­es­ is) is another acro­nym used to describe this phe­nom­en ­ on.
The bio­log­i­cal mech­a­nisms under­ly­ing CHIP are addressed in an
accom­pa­ny­ing review by L. Gondek.
CHIP is rel­a­tively com­mon in the aging pop­u­la­tion, with ini­
tial stud­ies esti­mat­ing a prev­a­lence of roughly 10% among indi­
vid­u­als aged 70 to 80. In those stud­ies, CHIP con­fers a 10-fold
increase in the risk of devel­op­ing a hema­to­logic malig­nancy.1-3
This trans­lates to a rate of pro­gres­sion of 0.5% to 1% per year, sim­
i­lar to the rate of pro­gres­sion from mono­clo­nal gammopathy of
unde­ter­mined sig­nif­i­cance and mono­clo­nal B-cell lym­pho­cy­to­sis
to lym­phoid malig­nan­cies.1-3 Accordingly, most older indi­vid­u­als
with CHIP have nor­mal blood counts, no fea­tures of hema­to­logic
malig­nan­cies, and a very low risk of pro­gres­sion. However, the
risk of pro­gres­sion from CHIP to mye­loid malig­nan­cies can vary
and depends on sev­eral fac­tors. The size of the CHIP clone cor­re­
lates with the mag­ni­tude of risk. A var­i­ant allele fre­quency (VAF)
cut­off of 1% to 2% was ini­tially used to define CHIP. This cut­off was
based on the lower limit of detec­tion of whole exome sequenc­
ing but may not be ideal for cur­rent clin­ic ­ al prac­tice. Clones with
a VAF of 10% or more dem­on­strate a 50-fold increase in the risk
Table 1.  The most com­monly mutated genes in CHIP/CCUS
and their respec­tive func­tions
Genes Function/path­way
DNMT3A, TET2, ASXL1, IDH1, IDH2 Epigenetic reg­u­la­tors
SF3B1, SRSF2, U2AF1 RNA splic­ing
TP53, PPM1D DNA repair/cell cycle
JAK2, CBL, GNAS, GNB1 Signaling
BCOR, BCORL1 Transcription reg­u­la­tion
Mutations in DNMT3A, TET2, and ASXL1 are the most fre­quent muta­tions
in CHIP and con­fer a lower risk of pro­gres­sion in sol­i­tary.
400  |  Hematology 2021  |  ASH Education Program
of devel­op­ing a hema­to­logic malig­nancy.1-3 The iden­tity of the
mutated genes affects the risk of pro­gres­sion. While genes that
code for epi­ge­netic mod­i­fi­ers (DNMT3A, TET2, and ASXL1) are the
most com­monly mutated in CHIP, they do not con­fer the highest
risk of pro­gres­sion. Two stud­ies exam­ined CHIP in blood sam­ples
taken years before patients devel­oped AML and showed a higher
prev­a­lence of CHIP in these preleukemic sam­ples.5,6 Mutations in
splic­ing-fac­tor genes (U2AF1, SRSF2, and SF3B1) as well as IDH1,
IDH2, and TP53 con­ferred a higher risk of pro­gres­sion.5,6
The risk of pro­gres­sion is also influ­enced by the num­ber of
muta­tions in an indi­vid­ual with CHIP. Multiple driver muta­tions in
the same indi­vid­ual are thought to reflect clonal evo­lu­tion and
the acqui­si­tion of pas­sen­ger muta­tions in the path­way toward
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/399/1852160/399osman.pdf by guest on 13 December 2021
frank malig­nancy.
Blood counts are nor­mal by def­i­ni­tion in CHIP, but a higher
red cell dis­tri­bu­tion width (RDW) seems to pre­dict a higher risk
of pro­gres­sion.5 A higher RDW is thought to reflect under­ly­ing
inef­fec­tive hema­to­poi­e­sis and often her­alds fall­ing counts and
malig­nant trans­for­ma­tion.
Clinical con­se­quences and man­age­ment of CHIP
Currently, there are no clear clin­i­cal indi­ca­tions to screen for
CHIP. Nevertheless, due to expanding indi­ca­tions for genetic
sequenc­ing, CHIP is increas­ingly diag­nosed in the clin­i­cal set­
ting. Current indi­ca­tions for genetic sequenc­ing include germ­
line sequenc­ing for the eval­u­a­tion of can­cer pre­dis­po­si­tion, for
guid­ance on ther­apy choices in solid tumors, for detecting cir­cu­
lat­ing tumor DNA, and in the workup of unex­plained cytopenias.
In patients with solid tumors and their rel­a­tives, test­ing for
germline pre­ dis­
po­
si­
tion is rou­ tinely done on DNA extracted
from the periph­eral blood. Germline pre­dis­po­si­tion genetic pan­
els can include genes involved in CHIP, such as TP53. Not infre­
quently, these pan­els will return results in which muta­tions are
detected at a VAF below the usual het­ero­zy­gos­ity rates (VAFs
below 40%). These clones could rep­re­sent CHIP, and addi­tional
tis­sue test­ing may be needed to clar­ify this.7,8 Similarly, test­ing
for cir­cu­lat­ing tumor DNA is done using blood sam­ples and could
uncover CHIP clones.9 CHIP clones can also be uncov­ered when
sequenc­ing is conducted using solid-tumor tis­sue sam­ples due
to con­tam­i­na­tion with blood cells.10
With the increase in CHIP diag­noses, sev­eral large cen­ters,
includ­ing ours, have started ded­i­cated CHIP clin­ics to eval­u­ate
and fol­low these patients.11 A diag­no­sis of CHIP in older indi­vid­u­
als could pro­duce unwar­ranted anx­i­ety. The risk of hema­to­logic
malig­nan­cies for car­ri­ers of CHIP who lack high-risk fea­tures is
minor. It is impor­tant to reas­sure these indi­vid­u­als that most of
them will not develop hema­to­logic malig­nan­cies and do not
require clin­i­cal inter­ven­tion.12 On the other hand, clones with
larger VAFs (>10%) and clones with muta­tions in high-risk genes
or with muta­tions in more than 1 driver gene war­rant closer
mon­i­tor­ing with serial blood counts. A bone mar­row exam­i­na­
tion should be pur­sued if these patients develop cytopenias. The
workup for cytopenias is fur­ther detailed below.
Certain CHIP clones can carry sig­nif­i­cant risk to patients with
solid tumors who will undergo treat­ment with cyto­toxic che­
mo­ther­apy and/or radi­a­tion. Cytotoxic che­mo­ther­apy and ion­
iz­ing radi­a­tion are likely to intro­duce a selec­tive pres­sure that
leads to the expan­sion of CHIP clones with muta­tions in the DNA
repair path­ways (TP53, CHEK2, and PPM1D). The pres­ence of
CHIP muta­tions increases the risk of treat­ment-related mye­loid
 Persistent cytopenias:
hemoglobin <13.0 g/dL [males], <12.0 g/dL [females]; absolute neutrophil count
<1·8 × 109/L; platelets <150 × 109/L
Rule out nutritional, autoimmune,
infectious, toxic, and neoplastic
processes
Unexplained cytopenias
Bone marrow aspiration and
biopsy
Cytogenetic evaluation of the
bone marrow
NGS panel including genes
associated with myeloid
malignancies
No evidence of clonality
ICUS
Figure 1. The diagnostic workup for cytopenias starts with ruling out underlying etiologies, including nutritional, autoimmune,
infectious, drug-mediated, and neoplastic processes. NGS plays an important role in the workup of unexplained cytopenias to
differentiate between CCUS and ICUS.
neo­plasms.13 This find­ing has impor­tant clin­i­cal impli­ca­tions for
patients with solid tumors under­ go­
ing adju­vant/neoadjuvant
che­mo­ther­apy and/or radi­a­tion. Currently, no clin­i­cal guide­lines
call for screen­ing for CHIP in this pop­u­la­tion.
Special atten­tion should also be paid to the car­dio­vas­cu­lar risk
asso­ci­ated with CHIP. The increase in all­-cause mor­tal­ity detected
in ear­lier CHIP stud­ies was mostly attrib­ut­­able to an increase in
fatal car­dio­vas­cu­lar events, includ­ing ische­mic stroke, cor­o­nary
heart dis­ease, and early-onset myo­car­dial infarc­tion.1,2,14 Mutations
in the 2 most com­mon CHIP genes, TET2 and DNMT3A, are asso­
ci­ated with an increased risk of car­dio­vas­cu­lar dis­ease through
proinflammatory mech­a­nisms.1,14 In addi­tion, JAK2 V617F-mutated
CHIP is asso­ci­ated with venous throm­bo­sis, includ­ing deep vein
throm­bo­sis and pul­mo­nary embolism.15 Carriers of CHIP were also
found to have worse heart fail­ure out­comes and worse out­comes
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/399/1852160/399osman.pdf by guest on 13 December 2021
Dysplasia or
increased
myeloblasts
MDS/AML
MDS
defining
cytogenetic
abnormalities
SOMATIC
mutation CCUS
VAF>2%
after aor­tic valve implan­ta­tion pro­ce­dures.12-14 In our CHIP clinic,
we assess the pres­ence of clas­sic risk fac­tors of car­dio­vas­cu­lar
dis­ease, such as hyper­lip­id­emia, dia­be­tes mellitus, hyper­ten­sion,
and smok­ing. We refer patients who require inter­ven­tions to our
cardio-oncol­o­gist for fur­ther man­age­ment.
The diag­nos­tic approach to unex­plained cytopenias
Unexplained cytopenias are defined as per­sis­tent cytopenias in 1
or more cell lin­e­ages with­out a clear under­ly­ing eti­­ol­ogy (hemo­
glo­bin <13.0 g/dL [males], <12.0 g/dL [fe­males]; abso­lute neu­tro­
phil count <1.8 × 109/L; plate­lets <150 × 109/L).16 The workup for
cytopenias starts with rul­ing out nutri­tional, auto­im­mune, infec­
tious, toxic, and neo­plas­tic causes.17 The term “unex­plained cy­
topenias” is used when the ori­gin of cytopenias is not explained
by these causes or by con­com­i­tant illnesses.
CHIP, CCUS, and ICUS: defin­ing mye­loid pre­cur­sors | 401
Table 2.  The unique clin­i­cal fea­tures dif­fer­en­ti­at­ing ICUS, CHIP, CCUS, and MDS
ICUS CHIP
Cytopenias + −
Somatic muta­tions − +
Morphologic dys­pla­sia − −
Increased blasts − −
Risk of trans­for­ma­tion to AML Very low Very low
Suggested man­age­ment and - CBC with dif­fer­en­tial, his­ - Routine health main­te­
fol­low-up* tory, and phys­ic­ al annu­ nance.
ally or semi­an­nu­ally - Consider annual CBC with
dif­fer­en­tial
- Assess car­dio­vas­cu­lar risk
+, pres­ent; −, absent.
*Evidence-based guide­lines for man­age­ment of CHIP, CCUS, and idi­o­pathic dys­pla­sia of unknown sig­nif­i­cance are cur­rently lacking. Follow-up should
be tai­lored to each patient’s indi­vid­ual risk and needs.
MDS is one of the most com­mon hema­to­logic malig­nan­cies and
fre­quently pres­ents with unex­plained cytopenias.17 The workup of
unex­plained cytopenias starts with rul­ing out MDS (Figure 1). MDS
is a group of het­ero­ge­neous clonal hema­to­poi­etic malig­nan­cies
char­ac­ter­ized by inef­fec­tive hema­to­poi­e­sis and increased risk
of AML. The diag­no­sis of MDS requires mor­pho­log­i­cal stud­ies of
periph­eral blood and bone mar­row and cyto­ge­net­ics to iden­tify
non­ran­dom chro­mo­somal abnor­mal­i­ties. Diagnostic cri­te­ria for
MDS include dys­pla­sia in at least 10% of cells in any hema­to­poi­
etic lin­e­age, increased mye­lo­blasts, or MDS-defin­ing cyto­ge­netic
abnor­mal­i­ties detected through karyotyping.18 It is impor­tant to
note that mor­pho­logic cri­te­ria for MDS suf­fer from great interpro­
vider var­i­abil­ity in terms of detecting dys­pla­sia.19
NGS plays an impor­tant role in the workup of unex­plained
cytopenias. Although not included in the cur­rent diag­nos­tic cri­
te­ria for MDS, NGS pan­els from the periph­eral blood can aid in
the diag­no­sis of MDS and other hema­to­logic malig­nan­cies.20,21
The absence of can­ cer-asso­ci­
ated somatic muta­ tions in the
periph­ eral blood has a high neg­a­ tive pre­ dic­tive value for an
under­ly­ing mye­loid malig­nancy.22 On the other hand, the pres­
ence of these muta­tions in the periph­eral blood raises the sus­
pi­cion of an under­ly­ing hema­to­logic malig­nancy. The pres­ence
of larger clones with VAFs of ≥10% and clones with 2 or more
muta­tions in the periph­eral blood has a pos­i­tive pre­dic­tive value
of 0.86 and 0.88 for the diag­no­sis of a mye­loid neo­plasm in cyto­
penic patients. Conversely, smaller clones involv­ing only 1 muta­
tion in the com­mon epi­ge­netic reg­u­la­tors (DNM3TA, TET2, and
ASXL1) have a low pre­dic­tive value unless a sec­ond muta­tion is
pres­ent in the same sam­ple. Mutations in spliceosome genes
(SF3B1, SRSF2, and U2AF1), JAK2, and RUNX1 have the highest
pre­dic­tive value for mye­loid neo­plasms irrespective of co-occur­
ring muta­tions.23
In the absence of diag­nos­tic cri­te­ria for MDS, unex­plained
cytopenia could rep­re­sent one of two enti­ties depending on
the pres­ ence of clonal genetic abnor­ mal­ i­
ties: clonal cytope­
nia of unknown sig­nif­i­cance (CCUS) or idi­o­pathic cytopenia of
unknown sig­nif­i­cance (ICUS).24-26 It is impor­tant to dif­fer­en­ti­ate
between CCUS and ICUS because the prog­no­sis and man­age­
402  |  Hematology 2021  |  ASH Education Program
CCUS MDS
+ +
+ +
− +
− ±
Low Low to very high
- CBC with dif­fer­en­tial every 3-6 mo Treatment per stage-
- History and phys­i­cal annu­ally or spe­cific guide­lines
semi­an­nu­ally
- Repeat bone mar­row biopsy if
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/399/1852160/399osman.pdf by guest on 13 December 2021
counts worsen
- Assess car­dio­vas­cu­lar risk
ment are dif­fer­ent (Table 2). ICUS is defined as unex­plained cyto­
penia that does not meet the min­i­mal cri­te­ria for MDS. CCUS is
defined as ICUS with one or more somatic muta­tions typ­i­cally
found in patients with mye­loid malig­nan­cies. Idiopathic dys­pla­
sia of unknown sig­nif­i­cance is a con­di­tion with­out cytopenia that
usu­ally rep­re­sents a reac­tive pro­cess.24
The clin­i­cal sig­nif­i­cance of CCUS
The cur­rent model of pro­gres­sion from CHIP to a mye­loid malig­
nancy includes a step­wise pro­gres­sion through a num­ber of dis­
ease states. CHIP is the ear­li­est iden­ti­fi­able entity in this model;
by def­i­ni­tion, it requires the pres­ence of nor­mal blood counts.
The devel­op­ment of cytopenia rep­re­sents the next man­i­fes­ta­tion
of a dys­func­tional hema­to­poi­etic sys­tem progressing toward a
frank mye­loid malig­nancy. Cytopenia in CCUS is a result of the
clonal hema­to­poi­etic pro­cess lead­ing to inef­fec­tive hema­to­poi­
e­sis. It is impor­tant to rule out other causes of cytopenia before
diag­nos­ing CCUS. For exam­ple, a patient with iso­lated throm­
bo­cy­to­pe­nia and a low-level DNMT3A muta­tion could have idi­
o­pathic throm­bo­cy­to­pe­nic pur­pura (ITP). If the throm­bo­cy­to­pe­
nia responds to treat­ment with ste­roids, the patient is unlikely to
have CCUS. The DNMT3A muta­tion in this sit­u­a­tion is less sig­nif­i­
cant because DNMT3A-mutated CHIP is com­mon in healthy older
indi­vid­u­als.1-3
CCUS con­fers a much higher risk of pro­gres­sion to a mye­
loid malig­nancy than ICUS, with a 10-year cumu­la­tive prob­ab ­ il­
ity of pro­gres­sion of 82% in CCUS vs 9% in ICUS. Furthermore,
CCUS clones with a VAF >20% are asso­ci­ated with a more than
95% risk of pro­gres­sion to a mye­loid malig­nancy in a 10-year
period.23 Consequently, a VAF of 20% is likely a bet­ter cut­off to
define CCUS than the 2% used for defin­ing CHIP. The pres­ence
of muta­tions in spe­cific genes (such as U2AF1, ZRSR2, SRSF2,
JAK2, or RUNX1) and the pres­ence of mul­ti­ple muta­tions also her­
ald a high risk of pro­gres­sion of up to 80% to 90% in 5 years.23
Although the term “CCUS” is now accepted by experts in the
field, I would argue that this is a mis­no­mer because of the high
risk of pro­gres­sion to mye­loid malig­nancy in true CCUS. Larger
clonal events can also be seen in CCUS and are asso­ci­ated with
a risk of pro­gres­sion. Trisomy 8 detected by fluo­res­cence in situ
hybrid­iza­tion sig­naled the pres­ence of a low-level clonal pro­cess
and increased risk of sub­se­quent MDS in a group of patients with
unex­plained cytopenias.27 Focal copy num­ber aber­ra­tions and
copy-neu­tral loss of het­ero­zy­gos­ity detected by sin­gle-nucle­o­
tide poly­mor­phism assays are asso­ci­ated with worse sur­vival in
patients with CCUS.28
On the other hand, the clin­i­cal course for ICUS is more indo­
lent. Patients with ICUS and no evi­dence of clonality can be mon­
i­tored with peri­odic blood counts. Patients with CCUS require
closer mon­i­tor­ing of their blood counts and repeat bone mar­row
exam­i­na­tion if counts worsen (Table 2).
The clone size and the num­ber of muta­tions in CCUS tend
to be much higher than in CHIP. The pat­tern of muta­tions in
CCUS tends to resem­ble muta­tions found in lower-risk MDS.29,30
Furthermore, patients with high-risk CCUS had sim­i­lar sur­vival
to those with mye­ loid neo­plasms.23 High-risk CCUS lies in a
close con­tin­uum to lower-risk MDS, and the line between these
2 enti­ties con­tin­ues to blur. Patients with CCUS can also have
low-level dys­ pla­
sia, and some patients may develop trans­ fu­
sion-depen­dent ane­mia.29 While no large ran­dom­ized tri­als exist
in this space, a sin­gle-cen­ter study exam­in­ing the treat­ment of
trans­fu­sion-depen­dent CCUS showed some suc­cess with MDS-
type ther­a­pies, includ­ing growth fac­tors and hypomethylating
agents.31 Patients with high-risk CCUS who develop trans­fu­sion
depen­dency or signs and symp­toms of bone mar­row fail­ure may
ben­e­fit from treat­ment algo­rithms used in low-risk MDS. Consen­
sus treat­ment and mon­i­tor­ing guide­lines for CCUS are cur­rently
lacking.
Our patient was diag­nosed with high-risk CCUS. Although
his periph­eral counts were highly sus­pi­cious for MDS, he did not
have dys­pla­sia, increased mye­lo­blasts, or MDS-defin­ing cyto­ge­
netic abnor­mal­i­ties. The patient was coun­seled about his con­di­
tion and is being mon­i­tored with monthly hemograms due to the
high-risk nature of his muta­tions. A repeat bone mar­row exam­
i­na­tion will be pur­sued if his counts worsen or if he devel­ops
symp­toms of bone mar­row fail­ure.
Conflict-of-inter­est dis­clo­sure
Afaf E. W. G. Osman: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Afaf E. W. G. Osman: use of hypomethylating agents and growth
factors in CCUS is briefly discussed in this article.
Correspondence
Afaf E. W. G. Osman, Division of Hematology and Hematologic
Malignancies, University of Utah, 2000 Circle of Hope Dr, Salt
Lake City, UT 84103; e-mail: afaf​­.osman@hsc​­.utah​­.edu.
References
1. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hema­to­poi­e­
sis asso­ci­ated with adverse out­comes. N Engl J Med. 2014;371(26):2488-
2498.
2. Genovese G, Kähler AK, Handsaker RE, et  al. Clonal hema­to­poi­e­sis and
blood-can­cer risk inferred from blood DNA sequence. N Engl J Med. 2014;​
371(26):​2477-2487.
3. Xie M, Lu C, Wang J, et al. Age-related muta­tions asso­ci­ated with clonal
hema­to­poi­etic expan­sion and malig­nan­cies. Nat Med. 2014;20(12):1472-
1478.
Prakash Singh Shekhawat
4. Steensma DP, Bejar R, Jaiswal S, et al. Clonal hema­to­poi­e­sis of inde­ter­mi­
nate poten­tial and its dis­tinc­tion from myelodysplastic syn­dromes. Blood.
2015;126(1):9-16.
5. Abelson S, Collord G, Ng SWK, et al. Prediction of acute mye­loid leu­kae­mia
risk in healthy indi­vid­u­als. Nature. 2018;559(7714):400-404.
6. Desai P, Mencia-Trinchant N, Savenkov O, et al. Somatic muta­tions pre­cede
acute mye­loid leu­ke­mia years before diag­no­sis. Nat Med. 2018;24(7):1015-
1023.
7. Slavin TP, Coffee B, Bernhisel R, et  al. Prevalence and char­ac­ter­is­tics of
likely-somatic var­i­ants in can­cer sus­cep­ti­bil­ity genes among indi­vid­u­als
who had hered­i­tary pan-can­cer panel test­ing. Cancer Genet. 2019;235-
236(June):31-38.
8. Weitzel JN, Chao EC, Nehoray B, et  al. Somatic TP53 var­i­ants fre­quently
con­found germ-line test­ing results. Genet Med. 2018;20(8):809-816.
9. Jensen K, Konnick EQ , Schweizer MT, et  al. Association of clonal hema­
to­poi­e­sis in DNA repair genes with pros­tate can­cer plasma cell-free DNA
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/399/1852160/399osman.pdf by guest on 13 December 2021
test­ing inter­fer­ence. JAMA Oncol. 2021;7(1):107-110.
10. Bolton KL, Gillis NK, Coombs CC, et al. Managing clonal hema­to­poi­e­sis in
patients with solid tumors. J Clin Oncol. 2019;37(1):7-11.
11. Bolton KL, Zehir A, Ptashkin RN, et al. The clin­i­cal man­age­ment of clonal
hema­to­poi­e­sis: cre­a­tion of a clonal hema­to­poi­e­sis clinic. Hematol Oncol
Clin North Am. 2020;34(2):357-367.
12. Gondek LP, DeZern AE. Assessing clonal haematopoiesis: clin­i­cal bur­dens
and ben­e­fits of diag­nos­ing myelodysplastic syn­drome pre­cur­sor states.
Lancet Haematol. 2020;7(1):e73-e81.
13. Bolton KL, Ptashkin RN, Gao T, et  al. Cancer ther­apy shapes the fit­ness
land­scape of clonal hema­to­poi­e­sis. Nat Genet. 2020;52(11):1219-1226.
14. Jaiswal S, Natarajan P, Silver AJ, et  al. Clonal hema­to­poi­e­sis and risk of
ath­ero­scle­rotic car­dio­vas­cu­lar dis­ease. N Engl J Med. 2017;377(2):111-121.
15. Edelmann B, Gupta N, Schnoeder TM, et al. JAK2-V617F pro­motes venous
throm­bo­sis through β1/β2 integrin acti­va­tion. J Clin Invest. 2018;128(10):​
4359-4371.
16. Greenberg PL, Tuechler H, Schanz J, et  al. Cytopenia lev­ els for aiding
estab­ lish­ment of the diag­ no­
sis of myelodysplastic syn­ dromes. Blood.
2016;128(16):2096-2097.
17. Malcovati L, Hellström-Lindberg E, Bowen D, et al; Euro­pean Leukemi­
aNet. Diagnosis and treat­ment of pri­mary myelodysplastic syn­dromes in
adults: rec­om­men­da­tions from the Euro­pean LeukemiaNet. Blood. 2013;​
122(17):2943-2964.
18. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revi­sion to the World Health
Organization clas­si­fi­ca­tion of mye­loid neo­plasms and acute leu­ke­mia.
Blood. 2016;127(20):2391-2405.
19. Parmentier S, Schetelig J, Lorenz K, et al. Assessment of dys­plas­tic hema­
to­poi­ e­ sis: les­
sons from healthy bone mar­ row donors. Haematologica.
2012;97(5):723-730.
20. Haferlach T, Nagata Y, Grossmann V, et al. Landscape of genetic lesions in
944 patients with myelodysplastic syn­dromes. Leukemia. 2014;28(2):241-
247.
21. Zheng G, Chen P, Pallavajjalla A, et  al. The diag­nos­tic util­ity of targeted
gene panel sequenc­ing in dis­crim­i­nat­ing eti­­ol­o­gies of cytopenia. Am J
Hematol. 2019;94(10):1141-1148.
22. Steensma DP. How I use molec­u­lar genetic tests to eval­u­ate patients who
have or may have myelodysplastic syn­dromes. Blood. 2018;132(16):1657-
1663.
23. Malcovati L, Gallì A, Travaglino E, et al. Clinical sig­nif­i­cance of somatic
muta­ tion in unex­ plained blood cytopenia. Blood. 2017;129(25):3371-
3378.
24. Valent P, Bain BJ, Bennett JM, et al. Idiopathic cytopenia of unde­ter­mined
sig­nif­i­cance (ICUS) and idi­o­pathic dys­pla­sia of uncer­tain sig­nif­i­cance (IDUS),
and their dis­tinc­tion from low risk MDS. Leuk Res. 2012;36(1):1-5.
25. Valent P, Orazi A, Steensma DP, et al. Proposed min­i­mal diag­nos­tic cri­te­ria
for myelodysplastic syn­dromes (MDS) and poten­tial pre-MDS con­di­tions.
Oncotarget. 2017;8(43):73483-73500.
26. Valent P, Horny HP, Bennett JM, et  al. Definitions and stan­dards in the
diag­no­sis and treat­ment of the myelodysplastic syn­dromes: con­sen­sus
state­ments and report from a work­ing con­fer­ence. Leuk Res. 2007;31(6):
727-736.
27. Petrova-Drus K, Hasserjian R, Pozdnyakova O, et  al. Clinicopathologic
eval­u­a­tion of cytopenic patients with iso­lated tri­somy 8: a detailed
com­par­i­son between idi­o­pathic cytopenia of unknown sig­nif­i­cance and
low-grade myelodysplastic syn­drome. Leuk Lymphoma. 2017;58(3):569-
577.
CHIP, CCUS, and ICUS: defin­ing mye­loid pre­cur­sors | 403
28. Mikkelsen SU, Safavi S, Dimopoulos K, et al. Structural aber­ra­tions are asso­ 31. Xie Z, Nanaa A, Saliba AN, et al. Treatment out­come of clonal cytopenias of
ci­ated with poor sur­vival in patients with clonal cytopenia of unde­ter­mined unde­ter­mined sig­nif­i­cance: a sin­gle-insti­tu­tion ret­ro­spec­tive study. Blood
sig­nif­i­cance. Haematologica. 2021;106(6):1762-1766. Cancer J. 2021;11(3):43.
29. Kwok B, Hall JM, Witte JS, et  al. MDS-asso­ci­ated somatic muta­tions and
clonal hema­to­poi­e­sis are com­mon in idi­o­pathic cytopenias of unde­ter­
mined sig­nif­i­cance. Blood. 2015;126(21):2355-2361.
30. Cargo CA, Rowbotham N, Evans PA, et al. Targeted sequenc­ing identifies
patients with pre­clin­i­cal MDS at high risk of dis­ease pro­gres­sion. Blood. © 2021 by The Amer­i­can Society of Hematology
2015;​126(21):2362-2365. DOI 10.1182/hema­tol­ogy.2021000272

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/399/1852160/399osman.pdf by guest on 13 December 2021

404  |  Hematology 2021  |  ASH Education Program

 MANAGEMENT STRATEGIES FOR SICKLE CELL DISEASE

Optimizing management of sickle cell disease

in patients undergoing surgery

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/405/1851532/405oyedeji.pdf by guest on 13 December 2021

Charity I. Oyedeji1 and Ian J. Welsby1,2
1
Division of Hematology, Department of Medicine, Duke University School of Medicine, Durham, NC; 2Department of Anesthesiology, Duke University
Medical Center, Durham, NC

Individuals with sickle cell disease (SCD) are likely to be referred for surgery at some point in their lifetime due to a
high incidence of musculoskeletal and intrabdominal complications such as avascular necrosis and gallbladder disease.
Preoperative optimization is a multidisciplinary process that involves a hematologist with SCD expertise, an anesthe-
siologist, and the surgical team. The type and risk classification of the surgery, disease severity, medications, baseline
hemoglobin, transfusion history, and history of prior surgical complications are often documented. Clinicians should
consider perioperative risk assessment that includes determining the patient’s functional status and cardiovascular risk
and screening for obstructive sleep apnea. Many patients will require preoperative transfusion to reduce the risk of
postoperative complications such as acute chest syndrome and vaso-occlusive pain crises. The hematologist should
consider the patient’s preoperative transfusion requirements and ensure that the surgical team has an appropriate plan
for postoperative observation and management. This often includes follow-up laboratory studies, a postoperative pain
management plan, and venous thromboembolism prophylaxis. The transfusion plan should be patient-specific and take
into account the SCD genotype, baseline hemoglobin, disease severity, risk classification of the surgery, and history of
prior surgical complications. In the intraoperative and postoperative period, dehydration, hypothermia, hypotension,
hypoxia, and acidosis should be avoided, and incentive spirometry should be utilized to minimize complications such as
acute chest syndrome. In this review we discuss preoperative, intraoperative, and postoperative strategies to optimize
patients with SCD undergoing surgery.

LEARNING OBJECTIVES
• Understand strategies for optimizing patients with sickle cell disease undergoing surgery
• Review indications for preoperative transfusion for patients with sickle cell disease
• Understand options for postoperative pain management and venous thromboembolism prophylaxis in patients
with sickle cell disease

vative therapies including nonsteroidal anti­inflammatory

CLINICAL CASE
A 46­year­old African American woman with hemoglobin
(Hb) sickle­beta plus thalassemia (HbSβ+) and right hip
avascular necrosis (AVN) was referred to a sickle cell clinic
for optimization prior to right total hip replacement.
She reported progressive bilateral hip pain that had
been worse on the right for the past 10 years. She denied
having a precipitating injury. The pain worsened with pro­
longed sitting and standing or walking short distances.
She was most concerned that the hip pain made it diffi­
cult for her to play with her grandchildren. Three years ear­
lier, magnetic resonance imaging (MRI) of the right hip had
shown a small area of AVN of the femoral head. She was
referred to orthopedics and was managed with conser­
Prakash Singh Shekhawat
drugs, a corticosteroid injection, and physical therapy. Her
pain and mobility continued to worsen. A repeat MRI of the
hips 2 years later showed bilateral AVN that was worse in
the right hip than the left. Due to the failure of conservative
management, orthopedics recommended a right total hip
replacement.
Her medical history included sickle cell disease (SCD)
retinopathy, anxiety, depression, hypersomnia, and gas­
troesophageal reflux. She had no history of stroke, venous
thromboembolism (VTE), or chronic cardiopulmonary com­
plications. She had contracted pneumonia 8 years prior,
and her last transfusion had been 30 years earlier. Her med­
ications included methadone, oxycodone, pantoprazole,
aripiprazole, buspirone, folic acid, and dronabinol. She was
Perioperative management of sickle cell disease | 405
not on hydroxy­urea due to an intol­er­ance. She had no prior sur­
ger­ies and no aller­gies to med­i­ca­tions. She lived alone but had 2
sup­port­ive adult chil­dren. She smoked 1 cig­a­rette per day, drank
alco­hol occa­sion­ally, and had no illicit drug use.
On phys­i­cal exam, her blood pres­sure was 117/78 mmHg; heart
rate, 102/min; res­pi­ra­tion rate, 18/min; tem­per­a­ture, 36.7 °C;
pain score, 8/10; and weight, 73 kg. The mus­cu­lo­skel­e­tal exam
showed dif­fi­culty mount­ing the exam table and pain on inter­nal
and exter­nal hip rota­tion, with the right hip worse than the left.
Her Hb level was 8.7 g/dL; hemat­o­crit level, 34.6%; white blood
cell count, 5.7 × 109/L; plate­let count; 198 000; and abso­lute
retic­u­lo­cyte count, 93.3  ×  109/L, with 2.68% retic­u­lo­cytes. Her
Hb elec­tro­pho­re­sis showed a sickle hemo­glo­bin [HbS] level of
67.0%; HbA2, 4.6%; HbA, 17.7%; fetal hemo­glo­bin [HbF], 10.7%.
Her com­plete met­a­bolic pro­file, lac­tate dehy­dro­ge­nase, fer­
ri­tin, and urine microalbumin were all­within nor­mal. Coro­
navirus was not detected on SARS-CoV-2 poly­mer­ase chain
reac­tion.
Introduction
SCD is one of the most com­mon inherited dis­or­ders in the world.1
Patients with SCD expe­ri­ence red blood cell sick­ling, lead­ing to
micro­vas­cu­lar occlu­sion that results in com­pli­ca­tions such as
acute vaso-occlu­sive pain cri­ses, acute chest syn­drome, chronic
organ dam­age, and mus­cu­lo­skel­e­tal com­pli­ca­tions.2 Surgical
com­pli­ca­tions are more com­mon in patients with SCD com­pared
to the gen­eral pop­u­la­tion due to their increased risk of post­op­
er­a­tive acute chest syn­drome, infec­tions, vaso-occlu­sive pain
cri­ses, and 30-day sur­gi­cal mor­tal­ity of 1.1%.3,4 Failure to appro­
pri­ately opti­mize a patient with SCD perioperatively can lead to
com­pli­ca­tions such as acute chest syn­drome, which is asso­ci­ated
with an increased risk of death.5,6 With appro­pri­ate plan­ning,
man­age­ment, and post­op­er­a­tive mon­i­tor­ing, health care pro­vid­
ers can increase the like­li­hood of opti­mal sur­gi­cal out­comes.
Determine if the sur­gery is really nec­es­sary
Given the high risk of both sur­gi­cal and post­op­er­a­tive com­pli­
ca­tions in patients with SCD, it is imper­a­tive to first deter­mine
whether con­ser­va­tive man­age­ment strat­e­gies have truly failed.
Consider the risks asso­ci­ated with sur­gery vs con­tinu­ing with
con­ser­va­tive man­age­ment, includ­ing refer­ral to a spe­cial­ist cen­
ter offer­ing less inva­sive options. Ascertain a patient’s per­sonal
moti­va­tions for and expected out­comes of hav­ing sur­gery to
deter­mine whether sur­gery will allow them to achieve their per­
sonal goals. This pro­cess should be a shared deci­sion between
the patient and health care pro­vid­ers.
Preoperative man­age­ment
Once the team has planned to pro­ceed with sur­gery, the next
step is to eval­u­ate the patient’s risk of perioperative com­pli­ca­
tions. This is typ­i­cally performed by an anes­the­si­­ol­o­gist and a
hema­tol­o­gist with exper­tise in SCD either in sep­a­rate or mul­ti­
dis­ci­plin­ary out­pa­tient vis­its or by vir­tual meet­ing. Consider the
sur­gi­cal indi­ca­tion and the type of sur­gi­cal pro­ce­dure the patient
is under­go­ing and any spe­cial con­sid­er­ations (Table 1). For chil­
dren, ensure they are up to date on their SCD-spe­cific health
screen­ing, which includes transcranial Dopp­ler to assess their
stroke risk. Document the risk of the sur­gery; SCD sever­ity based
on their geno­type; com­pli­ca­tions such as stroke, renal dis­ease,
406  |  Hematology 2021  |  ASH Education Program
or pul­mo­nary hyper­ten­sion; non-SCD comorbidities; recent hos­
pi­tal­i­za­tions; recent infec­tions; base­line Hb; trans­fu­sion his­tory
with a his­tory of trans­fu­sion reac­tions and ease of crossmatch;
and prior sur­gi­cal com­pli­ca­tions.7,8 The hema­tol­o­gist should
con­sider trans­fu­sion require­ments and tar­get sickle Hb per­cent
(%HbS) based on the patient’s geno­type, phe­no­type, and Hb.
Medications must be reviewed to avoid drug-drug inter­ac­tions
and neph­ro­toxic med­i­ca­tions in the set­ting of perioperative fluid
shifts. The perioperative team should con­sider an appro­pri­ate
acu­ity of post­op­er­a­tive mon­i­tor­ing, includ­ing lab­o­ra­tory test­ing,
a pain man­age­ment plan, and VTE pro­phy­laxis.
Preoperative risk assess­ment
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/405/1851532/405oyedeji.pdf by guest on 13 December 2021
The anes­the­si­­ol­o­gist typ­i­cally assesses the patient’s perioperative
risk and fit­ness for sur­gery and plans anes­the­sia accord­ingly. Sur­
geries are strat­i­fied into risk categories based on their poten­tial for
intraoperative blood loss and post­op­er­a­tive com­pli­ca­tions (ie, low,
mod­er­ate, or high risk; Table 2). The patient’s his­tory of strokes,
acute chest syn­drome, obstruc­tive sleep apnea (OSA), adverse
reac­tions to seda­tion, or recur­rent VTE is documented since these
increase the patient’s risk of perioperative com­pli­ca­tions.
Functional capac­ity is often mea­sured by the abil­ity to per­
form met­a­bolic equiv­a­lent tasks.9 Patients unable to per­form ≥4
met­ab ­ olic equiv­al­ent tasks (ie, climbing a flight of stairs) have an
increased risk of car­diac events.9,10 Consider OSA screen­ing since
the con­di­tion increases a patient’s risk of perioperative hyp­ox­
emia.11,12 The risk of car­dio­vas­cu­lar events is often assessed using
val­i­dated mod­els such as the revised car­diac risk index. Patients
at high risk for car­diac events and with a known his­tory of car­
dio­pul­mo­nary dis­ease such as pul­mo­nary hyper­ten­sion, car­diac
symp­toms, or poor func­tional sta­tus should con­sider addi­tional
test­ing with echo­car­dio­gram, stress test­ing, and/or a car­di­­ol­
ogy con­sul­ta­tion.10
Preoperative trans­fu­sion
Preoperative opti­mi­za­tion for patients with SCD often includes
sim­ple trans­fu­sion or red cell exchange (RCE). The pri­mary goal
of pre­op­er­a­tive trans­fu­sion is to reduce the risk of post­op­er­a­tive
com­pli­ca­tions by increas­ing the Hb and reduc­ing the %HbS.13
Patients with SCD are at increased risk of post­op­er­at­ ive mor­bid­
ity pri­mar­ily related to acute chest syn­drome and pain cri­ses and
have a higher risk of mor­tal­ity com­pared to the gen­eral pop­u­la­
tion. The evi­dence on pre­op­er­a­tive trans­fu­sion in both pedi­at­ric
and adult pop­u­la­tions shows var­i­able out­comes in pre­vent­ing
adverse out­comes and mor­tal­ity.13,14
The Transfusion Alternatives Preoperatively in Sickle Cell Dis­
ease (TAPS) study, which was a mul­ti­cen­ter ran­dom­ized con­trolled
trial (RCT) com­par­ing no trans­fu­sion with trans­fu­sion within 10
days of low- or mod­er­ate-risk sur­gery for patients with geno­type
hemoglobin SS disease (HbSS) or sickle-beta zero thalassemia
(HbSβ0), showed that patients who received pre­op­er­a­tive trans­fu­
sion had a lower risk of post­op­er­a­tive acute chest syn­drome and
other clin­i­cally impor­tant com­pli­ca­tions (15% vs 39%; P = .023).15
Despite this, there was no dif­fer­ence in post­op­er­a­tive pain cri­sis,
hos­pi­tal length of stay, or readmission rates between patients
who received pre­op­er­a­tive trans­fu­sion com­pared with no trans­
fu­sion.13,15 A mul­ti­cen­ter RCT com­par­ing pre­op­er­a­tive con­ser­va­tive
trans­fu­sion (increas­ing Hb to 10  g/dL) to aggres­sive trans­fu­sion
(using RCE to decrease %HbS to <30%) found no sig­nif­i­cant dif­
fer­ence in the fre­quency of seri­ous com­pli­ca­tions.5 Ten per­cent of
Table 1. Common indi­ca­tions for sur­gery in sickle cell dis­ease and con­sid­er­ations

Surgical indi­ca­tion Surgery Surgical con­sid­er­ations

Severe avas­cu­lar necro­sis
Tonsillar hyper­tro­phy, pedi­at­ric
OSA, and/or recur­rent ton­sil­li­tis
Cholelithiasis
Total hip replace­ment27,28
ENT sur­gery such as ton­sil­lec­tomy
and adenoidectomy14,29
Cholecystectomy30
- High risk of periprosthetic and wound infec­tion and respi­ra­tory
com­pli­ca­tions
- Monitor tem­per­a­ture, fluid bal­ance, and Hb
- Postoperative pro­phy­lac­tic anticoagulation for a min­i­mum of 10-14 days
and up to 35 days; LMWH is recommended
- Early mobi­li­za­tion
- Chest phys­io­ther­apy with incen­tive spi­rom­e­try
- At increased risk of noc­tur­nal hyp­oxia pre- and post­op­er­a­tively
- Monitor for dehy­dra­tion and post­op­er­a­tive oral fluid intake
- Use least inva­sive tech­nique (lap­a­ro­scopic)
- Monitor for pul­mo­nary com­pli­ca­tions (atel­ec­ta­sis, acute chest syn­

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/405/1851532/405oyedeji.pdf by guest on 13 December 2021

drome) and vaso-occlu­sive cri­sis of the liver
Splenic seques­tra­tion Splenectomy31,32 - Use least inva­sive tech­nique (lap­a­ro­scopic)
- At risk for postsplenectomy infec­tion; immu­nize with pneu­mo­coc­cal,
menin­go­coc­cal, Haemophilus influ­enza at least 10-14 days before sur­
gery or post­op­er­a­tively if emer­gency sur­gery is required
- Consider pro­phy­lac­tic anticoagulation to pre­vent postsplenectomy
por­tal vein sys­tem throm­bo­sis
Priapism Surgical shunt or penile pros­the­sis33 - At risk for post­op­er­a­tive erec­tile dys­func­tion, penile gan­grene, and
per­i­neal abscess
Pregnancy com­pli­ca­tions (eg, Cae­sar­ian sec­tion26,34 - Hydrate and main­tain spO2 > 94%
pre­eclamp­sia) - Consider trans­fu­sion to Hb 10 g/dL
- Monitor fluid bal­ance to avoid vol­ume over­load
- Postpartum pro­phy­lac­tic anticoagulation for both vag­i­nal and Cae­
sar­ean deliv­er­ies for 6 weeks starting 6-12 hours after all­deliv­ery (no
sooner than 4 hours after epi­du­ral cath­e­ter removal). LMWH and war­fa­
rin are not contraindicated with breastfeeding.
Moyamoya dis­ease Cerebral revas­cu­lar­i­za­tion35 - At risk for reduced cere­bral blood flow due to gen­eral anes­the­sia
- Aggressive fluid hydra­tion prior to sur­gery
- Utilize intraoperative EEG
Valvular dis­ease Cardiac sur­gery36 - At risk for hem­or­rhage, stroke, renal fail­ure, and acute chest syn­drome
- For valve replace­ment, at risk of hemo­ly­sis, throm­bo­sis, and infec­tion
- Use incen­tive spi­rom­e­try, anticoagulation, and pro­phy­lac­tic anti­bi­ot­ics
- Maintain nor­mo­ther­mia to min­i­mize vaso­con­stric­tion and sick­ling
EEG, elec­tro­en­ceph­a­lo­gram; ENT, ear, nose, and throat; spO2, oxy­gen sat­u­ra­tion.

patients in each group devel­oped acute chest syn­drome.5 Based
on avail­­able evi­dence, some experts have con­ cluded that the
ben­e­fit of pre­op­er­a­tive trans­fu­sion is mostly related to increas­ing
Hb rather than reduc­ing %HbS.13 Although the cur­rent evi­dence
supporting pre­op­er­a­tive trans­fu­sion is low qual­ity, the ben­e­fits
out­weigh the harms; there­fore, pre­op­er­a­tive trans­fu­sion is still
recommended.13
There is a lack of con­sen­sus on the best prac­tices for pre­op­er­
a­tive and peripartum trans­fu­sions. High-qual­ity data are needed
in the era of hydroxy­ urea and other new dis­ ease-mod­ i­
fy­
ing
ther­ a­
pies. We advise that cli­ ni­
cians fol­
low national and local
guide­ lines. According to the Amer­ i­
can Society of Hematology
trans­fu­sion guide­lines, in some sit­u­a­tions RCE may reduce the risk
of acute chest syn­drome and pain cri­sis in patients with geno­type
HbSS/HbSβ0, Hb <9 g/dL, or a severe phenotype (characterized
as hav­ing a his­tory of stroke, recur­rent acute chest syn­drome,
or prior severe post­op­er­a­tive com­pli­ca­tions) or any per­son with
SCD under­go­ing high-risk sur­gery. Patients with a severe phe­no­
type are most likely to ben­e­fit from achiev­ing a post-RCE goal of
%HbS <30%. For patients with Hb <9 g/dL with­out a severe phe­
no­type under­go­ing mod­er­ate-risk sur­gery, sim­ple trans­fu­sion is
Prakash Singh Shekhawat
recommended with a post­trans­fu­sion goal of a Hb level no greater
than 11 g/dL in order to avoid hyper­vis­cos­ity. For patients with Hb
>9 g/dL under­go­ing mod­er­ate-risk sur­gery, RCE is recommended.
For patients with Hb >9g/dL with­out a severe phe­no­type under­
go­ing low-risk sur­gery, con­sider pro­ceed­ing with­out trans­fu­sion.13
Patients under­go­ing emer­gency sur­gery may receive sim­ple trans­
fu­sion to increase Hb to 10 g/dL, if indi­cated. Sometimes, pre­op­
er­a­tive trans­fu­sion may not be accept­able due to the poten­tial for
delaying a life-sav­ing sur­gery; there­fore, intraoperative or post­op­
er­a­tive trans­fu­sion may be con­sid­ered.13 See the sum­mary of trans­
fu­sion rec­om­men­da­tions in Table 3.
Alloimmunization
The inci­dence of alloimmunization in SCD ranges from 7% to 58%
depending on age, num­ber of pre­vi­ous trans­fu­sions, and use of
red cell phe­no­typic matching.16,17 The TAPS study noted no dif­fer­
ence in alloimmunization between patients receiv­ing pre­op­er­a­
tive trans­fu­sion com­pared to no trans­fu­sion likely due to the use
of extended red cell phe­no­typic matching, and par­tic­i­pants had
pre­vi­ously received fewer trans­fu­sions com­pared to other stud­
ies.15 When com­par­ing aggres­sive trans­fu­sion to con­ser­va­tive
Perioperative man­age­ment of sickle cell dis­ease  |  407
Table 2. Surgical risk clas­si­fi­ca­tion

Low risk Moderate risk High risk

Dental extrac­tions Head and neck sur­gery Cardiac sur­gery (eg, valve replace­ment)
Ophthalmologic pro­ce­dures (ie, laser sur­gery, Orthopedic sur­gery (eg, total hip replace­ment) Intrathoracic sur­gery
cat­a­ract sur­gery)
Hernia repair Urologic sur­gery Brain sur­gery
Dilation and curet­tage Cholecystectomy Transplant sur­gery (heart, lung, liver, kid­ney)
Wound debride­ment Splenectomy Major vas­cu­lar sur­gery (eg, aorta repair)
Endoscopy Hysterectomy Major spine sur­gery
Superficial tis­sue biopsy (eg, breast biopsy) Uterine abla­tion Surgeries requir­ing prolonged gen­eral
anes­the­sia >4 hours (eg, Whipple pro­ce­dure)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/405/1851532/405oyedeji.pdf by guest on 13 December 2021

Cae­sar­ian sec­tion
Appendectomy

trans­fu­sion, an RCT showed an increased risk of alloimmuniza­
tion with aggres­sive trans­fu­sion; there­fore, many cen­ters pre­fer
sim­ple trans­fu­sion over RCE when appro­pri­ate.5 Patients with a
his­tory of mul­ti­ple alloantibodies, delayed hemo­lytic trans­fu­sion
reac­tion (DHTR), and/or hyperhemolysis have an increased risk
of adverse out­comes related to trans­fu­sion; there­fore, care­ful
con­sid­er­ation should be given prior to trans­fu­sion. The risk of
trans­fu­sion-related com­pli­ca­tions should be discussed and doc­
umented, and in some sit­ua ­ ­tions, trans­fu­sion is not pos­si­ble. In
the event an emer­gent trans­fu­sion is required, con­sider hav­ing
crossmatched blood avail­­able on-site even if there is no plan
for trans­fu­sion. The blood should be anti­gen-neg­a­tive for any
pos­i­tive antibodies and pre­vi­ous antibodies even if those anti­
bodies are no lon­ger detect­able in order to min­i­mize the risk of
DHTR. For life-threat­en­ing ane­mia, con­sider immu­no­sup­pres­sive
ther­apy if trans­fu­sion is abso­lutely required.8,13,18 In patients with
reli­gious objec­tions to trans­fu­sion or if trans­fu­sion is not pos­si­
ble due to severe alloimmunization/pre­vi­ous DHTR, it is impor­
tant to opti­mize the Hb in advance with med­i­ca­tions such as
hydroxy­urea, voxelotor, and ery­throid-stim­u­lat­ing agents.
Pain man­age­ment plan
The patient’s cur­rent pain reg­im­ en should be reviewed, and a
post­op­er­a­tive pain man­age­ment plan should be devel­oped, often
sim­i­lar to a patient’s usual inpa­tient acute pain plan and includ­ing
patient-con­trolled anal­ge­sia (PCA).7 The deci­sion to include con­
tin­u­ous basal opi­oid infu­sion vs on-demand dos­ing only should
be patient-spe­cific and deter­mined by the patient’s level of post­
op­er­a­tive pain and opi­oid tol­er­ance.19 Referral to the local acute
pain team is advised to ensure appro­pri­ate opi­ate dos­ing and
nonopiate adjuncts, such as a short course of non­ste­roi­dal anti-
inflam­ma­tory drugs (5-7 days if no renal risk), intra­ve­nous (IV) ket­
amine or lido­caine, regional blocks, and acet­amin­o­phen.19
Intraoperative period and imme­di­ately after sur­gery
The most impor­ tant fac­tor to con­ sider intraoperatively is to
avoid imbal­ances in vol­ume sta­tus, tem­per­a­ture, acid-base bal­
ance, blood pres­ sure, and oxy­ gen­
a­tion since derange­ ments
increase red blood cell sick­ling, which can result in acute organ
injury. It can also man­i­fest as a vaso-occlu­sive pain cri­sis, as acute
chest syn­drome, as an acute kid­ney injury, or even as an ische­mic
stroke. Practical strat­e­gies to main­tain euvolemia include avoid­
ing prolonged fasting prior to sur­gery with­out IV flu­ids, mon­i­
tor­ing fluid intake and out­put, and decreas­ing IV flu­ids as soon
as patients are a ­ ble to main­tain ade­quate oral fluid intake. Avoid
extremely cold or hot ambi­ent tem­per­a­tures pre­op­er­a­tively, in
the oper­at­ing room, and in the recov­ery space while using fluid
and con­vec­tive warming tech­nol­ogy aggres­sively.

Table 3. Transfusion rec­om­men­da­tions based on sickle cell geno­type, hemo­glo­bin, and sur­gi­cal risk clas­si­fi­ca­tion

Hemoglobin
Sickle cell geno­type (g/dL) Surgical risk Transfusion rec­om­men­da­tion
HbSS or HbSβ - thal­as­se­mia
0
<9 Low or mod­er­ate Simple trans­fu­sion, par­tial exchange, or RCE
HbSS or HbSβ0- thal­as­se­mia >9 Low or mod­er­ate Partial exchange or RCE
HbSC, HbSβ -thal­as­se­mia, and HbSS on hydroxy­urea and
+
>9 Low No trans­fu­sion
ele­vated HbF with­out severe phe­no­type*
HbSC or HbSβ+- thal­as­se­mia >9 Moderate Partial exchange or RCE
All geno­types — High RCE
*Severe phe­no­type is defined as hav­ing a his­tory of stroke, recur­rent acute chest syn­drome, or prior severe sur­gi­cal com­pli­ca­tions.
HbSC, hemoglobin SC disease.

408  |  Hematology 2021  |  ASH Education Program

Postoperative period
All patients with SCD should use an incen­tive spi­rom­e­ter post­
op­er­a­tively since this has been shown to reduce the inci­dence
of atel­ec­ta­sis and acute chest syn­drome in hos­pi­tal­ized patients
with SCD.20,21 Monitor for post­op­er­a­tive infec­tion, and if a patient’s
tem­per­a­ture is ≥38 °C, inves­ti­gate for a source of the fever with
blood and urine cul­tures, wound inspec­tion, and chest x-ray, as
indi­cated.8 Minimize the risk of hyp­oxia with the judi­cious use of
anal­ge­sics to avoid respi­ra­tory depres­sion. If oxy­gen sat­u­ra­tion
falls by ≥2% below the patient’s base­line or is ≤94%, pro­vide sup­
ple­men­tal oxy­gen and eval­u­ate the patient with con­sid­er­ation
for post­op­er­a­tive com­pli­ca­tions such as acute chest syn­drome
and pul­mo­nary embolism. The eval­u­a­tion often includes phys­i­
cal exam­i­na­tion, arte­rial blood gas, and chest imag­ing. Consider
checking the met­a­bolic pro­file after sur­gery to assess for acute
kid­ney injury and elec­tro­lyte derange­ments.
Anticoagulation man­age­ment
SCD causes hyper­co­ag­u­la­bil­ity due to mul­ti­ple alter­ations in
hemo­sta­sis, includ­ing increased plate­let acti­va­tion, acti­va­tion
of the clot­ting cas­cade, impaired fibri­no­ly­sis, and intra­vas­cu­lar
hemo­ly­sis.22,23 The inci­dence of VTE in indi­vid­u­als with SCD is
as high as 12% before the age of 40.24 Hospitalized adults with
SCD should receive VTE pro­phy­laxis if there is no con­tra­in­di­ca­
tion since VTE in SCD is asso­ci­ated with increased mor­tal­ity.25
There is insuf­fi­cient evi­dence on post­op­er­a­tive VTE pro­phy­laxis
in chil­dren. For both pedi­at­ric and adult pop­u­la­tions, we sug­gest
pre­scrib­ing VTE pro­phy­laxis based on national or local guide­
lines. Most inpa­tient teams use low-molec­u­lar-weight hep­a­rin
(LMWH) or low-dose unfractionated hep­a­rin. Patients under­go­
ing sur­ger­ies that require prolonged immo­bi­li­za­tion, such as hip
sur­gery, require extended VTE pro­phy­laxis (Table 1). Perioper­
ative teams should con­sider the use of inter­mit­tent pneu­matic
com­pres­sion dur­ing the hos­pi­tal stay and encour­age early mobi­
li­za­tion. LMWH or war­fa­rin is recommended for post­par­tum VTE
pro­phy­laxis in all­indi­vid­u­als with SCD who plan to breastfeed
since DOACs are contraindicated (Table 1).26
CLINICAL CASE (Con­tin­ued)
Our patient’s perioperative risk scores were documented by
the anes­the­si­­ol­o­gist as low risk, with phys­i­cal sta­tus con­sis­tent
with mild sys­temic dis­ease with­out sub­stan­tial lim­i­ta­tions and
a low risk of hav­ing OSA. She received RCE 4 days before sur­
gery. Post-RCE Hb improved to 11.4 g/dL and HbS was 24.8%.
She under­went total right hip replace­ment under spi­nal anes­the­
sia with­out com­pli­ca­tions. Her post­op­er­a­tive Hb was 9.6 g/dL.
An anes­the­sia acute pain ser­vice man­aged her post­op­er­a­tive
pain. A con­tin­u­ous infu­sion of ropivicane via a cath­e­ter inserted
under ultra­sound guid­ance into the adduc­tor canal was supple­
mented by IV hydromorphone PCA and a low-dose ketamine
infu­sion. Her home meth­a­done was started once she was tol­
er­at­ing a diet, and the PCA and ketamine were discontinued on
post­op­er­a­tive day (POD) 1. Cyclobenzaprine and ketorolac were
started once renal func­tion was deter­mined to be sta­ble.
For post­op­er­a­tive pro­phy­laxis for deep vein throm­bo­sis, she
received 30 mg of LMWH twice a day for 30 days, which started
on POD 1. She was discharged on POD 2 with out­pa­tient reha­bil­
Prakash Singh Shekhawat
i­ta­tion 3 days a week. On fol­low-up in the hema­tol­ogy clinic, she
reported sig­nif­i­cant improve­ments in mobil­ity, range of motion,
and pain. She no lon­ger required the use of an assist device to
ambu­late and was a ­ ble to meet her goal of being a ­ ble to play
with her grandchildren.
Conclusion
Preoperative opti­mi­za­tion is an essen­tial part of care for indi­vid­
u­als with SCD under­go­ing sur­gery. People with SCD are more
likely to have sur­gi­cal com­pli­ca­tions com­pared to the gen­eral
pop­u­la­tion, so it is imper­a­tive that con­sul­ta­tion with an SCD spe­
cial­ist be included in the opti­mi­za­tion plan. Appropriate opti­mi­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/405/1851532/405oyedeji.pdf by guest on 13 December 2021
za­tion and man­age­ment prior to, dur­ing, and after sur­gery can
reduce the risk of adverse out­comes in indi­vid­u­als with SCD.
More research is needed to improve perioperative man­age­ment
guide­lines for SCD.
Acknowledgments
Charity I. Oyedeji rec­og­nizes sup­port from the Amer­i­can Soci­
ety of Hematology Research Training Award for Fellows and the
Duke Center for Research to Advance Healthcare Equity, which
is supported by the National Institute on Minority Health and
Health Disparities under award num­ber U54MD012530.
Conflict-of-inter­est dis­clo­sure
Charity I. Oyedeji: no com­pet­ing finan­cial inter­ests to declare.
Ian J. Welsby: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Charity I. Oyedeji: nothing to disclose.
Ian J. Welsby: nothing to disclose.
Correspondence
Charity I. Oyedeji, Division of Hematology, Department of Med­
icine, Duke University School of Medicine, 315 Trent Dr, Hanes
House, Ste 261, DUMC Box 3939, Dur­ham, NC 27710; e-mail: char
­ity​­.oyedeji@duke​­.edu.
References
1. Modell B, Darlison M. Global epi­de­mi­­ol­ogy of haemoglobin dis­or­ders and
derived ser­vice indi­ca­tors. Bull World Health Organ. 2008;86(6):480-
487.
2. Kato GJ, Piel FB, Reid CD, et al. Sickle cell dis­ease. Nat Rev Dis Primers.
2018;4(March):18010.
3. Buck J, Davies SC. Surgery in sickle cell dis­ease. Hematol Oncol Clin North
Am. 2005;19(5):897-902, vii.
4. Koshy M, Weiner SJ, Miller ST, et al. Surgery and anes­the­sia in sickle cell dis­
ease. Cooperative Study of Sickle Cell Diseases. Blood. 1995;86(10):3676-3684.
Accessed 27 Sep­tem­ber 2021.
5. Vichinsky EP, Haberkern CM, Neumayr L, et al; Preoperative Transfusion in
Sickle Cell Disease Study Group. A com­par­i­son of con­ser­va­tive and aggres­
sive trans­fu­sion reg­i­mens in the perioperative man­age­ment of sickle cell
dis­ease. N Engl J Med. 1995;333(4):206-213.
6. Allareddy V, Roy A, Lee MK, et al. Outcomes of acute chest syn­drome in
adult patients with sickle cell dis­ease: pre­dic­tors of mor­tal­ity. PLoS One.
2014;9(4):e94387.
7. Adjepong KO, Otegbeye F, Adjepong YA. Perioperative man­age­ment of
sickle cell dis­ease. Mediterr J Hematol Infect Dis. 2018;10(1):e2018032.
8. Walker I, Trompeter S, Howard J, et  al. Guideline on the peri-oper­a­tive
man­age­ment of patients with sickle cell dis­ease: guide­line from the Asso­
ciation of Anaesthetists. Anaesthesia. 2021;76(6):805-817.
Perioperative man­age­ment of sickle cell dis­ease  |  409
9. Weinstein AS, Sigurdsson MI, Bader AM. Comparison of pre­ op­er­
a­tive
assess­ ment of patient’s met­ a­
bolic equiv­a­lents (METs) esti­ mated from
his­tory ver­sus mea­sured by exer­cise car­diac stress test­ing. Anesthesiol Res
Pract. 2018;2018(Sep­tem­ber):5912726.
10. Fleisher LA, Fleischmann KE, Auerbach AD, et al. 2014 ACC/AHA guide­line
on perioperative car­dio­vas­cu­lar eval­u­a­tion and man­age­ment of patients
under­ go­ing non­ car­
diac sur­ gery: a report of the Amer­ i­
can College of
Cardiology/Amer­i­can Heart Association Task Force on Practice Guidelines.
Circulation. 2014;130(24):e278-e333.
11. Katz T, Schatz J, Roberts CW. Comorbid obstruc­ tive sleep apnea and
increased risk for sickle cell dis­ease mor­bid­ity. Sleep Breath. 2018;22(3):797-
804.
12. Chung F, Abdullah HR, Liao P. STOP-Bang Questionnaire: a prac­ ti­
cal
approach to screen for obstruc­tive sleep apnea. Chest. 2016;149(3):631-
638.
13. Chou ST, Alsawas M, Fasano RM, et  al. Amer­i­can Society of Hematology
2020 guide­lines for sickle cell dis­ease: trans­fu­sion sup­port. Blood Adv.
2020;4(2):327-355.
14. Waldron P, Pegelow C, Neumayr L, et al. Tonsillectomy, adenoidectomy,
and myringotomy in sickle cell dis­ease: perioperative mor­bid­ity. Preop­
erative Transfusion in Sickle Cell Disease Study Group. J Pediatr Hematol
Oncol. 1999;21(2):129-135.
15. Howard J, Malfroy M, Llewelyn C, et  al. The Transfusion Alternatives Pre-
operatively in Sickle Cell Disease (TAPS) study: a randomised, con­trolled,
multicentre clin­i­cal trial. Lancet. 2013;381(9870):930-938.
16. Chou ST, Jackson T, Vege S, Smith-Whitley K, Friedman DF, Westhoff
CM. High prev­ a­
lence of red blood cell alloimmunization in sickle cell
dis­ease despite trans­ fu­sion from Rh-matched minor­ ity donors. Blood.
2013;122(6):1062-1071.
17. Lasalle-Williams M, Nuss R, Le T, et  al. Extended red blood cell anti­gen
matching for trans­fu­sions in sickle cell dis­ease: a review of a 14-year expe­
ri­ence from a sin­gle cen­ter (CME). Transfusion. 2011;51(8):1732-1739.
18. Noizat-Pirenne F, Habibi A, Mekontso-Dessap A, et al. The use of rituximab
to pre­vent severe delayed haemolytic trans­fu­sion reac­tion in immu­nized
patients with sickle cell dis­ease. Vox Sang. 2015;108(3):262-267.
19. Brandow AM, Carroll CP, Creary S, et  al. Amer­i­can Society of Hematology
2020 guide­lines for sickle cell dis­ease: man­age­ment of acute and chronic
pain. Blood Adv. 2020;4(12):2656-2701.
20. Niss O, Cole-Jenkins C, Davis B, et al. Prevention of acute chest syn­drome
by implementing a stan­dard­ized pro­cess to improve incen­tive spi­rom­e­try
use in hos­pi­tal­ized patients with sickle cell dis­ease. Blood. 2017;130(suppl
1):132.
21. Bellet PS, Kalinyak KA, Shukla R, Gelfand MJ, Rucknagel DL. Incentive spi­
rom­e­try to pre­vent acute pul­mo­nary com­pli­ca­tions in sickle cell dis­eases.
N Engl J Med. 1995;333(11):699-703.
410  |  Hematology 2021  |  ASH Education Program
22. Kelley D, Jones LT, Wu J, Bohm N. Evaluating the safety and effec­tive­ness of
venous throm­bo­em­bo­lism pro­phy­laxis in patients with sickle cell dis­ease.
J Thromb Thrombolysis. 2017;43(4):463-468.
23. Naik RP, Streiff MB, Lanzkron S. Sickle cell dis­ease and venous throm­bo­em­
bo­lism: what the anticoagulation expert needs to know. J Thromb Throm-
bolysis. 2013;35(3):352-358.
24. Brunson A, Lei A, Rosenberg AS, White RH, Keegan T, Wun T. Increased
inci­dence of VTE in sickle cell dis­ease patients: risk fac­tors, recur­rence and
impact on mor­tal­ity. Br J Haematol. 2017;178(2):319-326.
25. Novelli EM, Huynh C, Gladwin MT, Moore CG, Ragni MV. Pulmonary
embolism in sickle cell dis­ease: a case-con­trol study. J Thromb Haemost.
2012;10(5):760-766.
26. Bates SM, Middeldorp S, Rodger M, James AH, Greer I. Guidance for the treat­
ment and pre­ven­tion of obstet­ric-asso­ci­ated venous throm­bo­em­bo­lism.
J Thromb Thrombolysis. 2016;41(1):92-128.
27. Kamath AF, McGraw MH, Isra­el­ite CL. Surgical man­age­ment of osteonecro­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/405/1851532/405oyedeji.pdf by guest on 13 December 2021
sis of the fem­o­ral head in patients with sickle cell dis­ease. World J Orthop.
2015;6(10):776-782.
28. Falck-Ytter Y, Francis CW, Johanson NA, et al. Prevention of VTE in ortho­pe­
dic sur­gery patients: Antithrombotic Therapy and Prevention of Thrombosis,
9th ed: Amer­i­can College of Chest Physicians Evidence-Based Clinical
Practice Guidelines. Chest. 2012;141(suppl 2):e278S-e325S.
29. Warrier R, Chauhan A, Athale U. Tonsillectomy and adenoidectomy for
obstruc­tive sleep apnea in sickle cell ane­mia. Indian J Pediatr. 2010;77(6):669-
672.
30. Martins RA, Soares RS, Vito FB, et al. Cholelithiasis and its com­pli­ca­tions
in sickle cell dis­ease in a uni­ver­sity hos­pi­tal. Rev Bras Hematol Hemoter.
2017;39(1):28-31.
31. Al-Salem AH. Splenic com­pli­ca­tions of sickle cell ane­mia and the role of sple­nec­
tomy. ISRN Hematol. 2011;2011(October 2010):864257.
32. Zhang N, Yao Y, Xue W, Wu S. Early pro­phy­lac­tic anticoagulation for por­tal
vein sys­tem throm­bo­sis after sple­nec­tomy: a sys­tem­atic review and meta-
anal­y­sis. Biomed Rep. 2016;5(4):483-490.
33. Olujohungbe A, Burnett AL. How I man­age priapism due to sickle cell dis­
ease. Br J Haematol. 2013;160(6):754-765.
34. Jain D, Atmapoojya P, Colah R, Lodha P. Sickle cell dis­ease and preg­nancy.
Mediterr J Hematol Infect Dis. 2019;11(1):e2019040.
35. Smith ER, Scott RM. Surgical man­age­ment of moyamoya syn­drome. Skull
Base. 2005;15(1):15-26.
36. Crawford TC, Carter MV, Patel RK, et al. Management of sickle cell dis­ease
in patients under­go­ing car­diac sur­gery. J Card Surg. 2017;32(2):80-84.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000274
 MANAGEMENT STRATEGIES FOR SICKLE CELL DISEASE

Treatment dilemmas: strategies for priapism,

chronic leg ulcer disease, and pulmonary
hypertension in sickle cell disease

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/411/1851512/411azbell.pdf by guest on 13 December 2021

Roberta C.G. Azbell1,2 and Payal Chandarana Desai1,3
The Ohio State University Wexner Medical Center, Department of Internal Medicine, Columbus, OH; 2Division of Hospital Medicine, Columbus, OH;
1

and 3Division of Hematology and Oncology, Columbus, OH

Sickle cell disease is a disorder characterized by chronic hemolytic anemia and multiorgan disease complications.
Although vaso-occlusive episodes, acute chest syndrome, and neurovascular disease frequently result in complication
and have well-documented guidelines for management, the management of chronic hemolytic and vascular-related com-
plications, such as priapism, leg ulcers, and pulmonary hypertension, is not as well recognized despite their increasing
reported prevalence and association with morbidity and mortality. This chapter therefore reviews the current updates
on diagnosis and management of priapism, leg ulcers, and pulmonary hypertension.

LEARNING OBJECTIVES
• Differentiate the types of priapism presenting in patients with SCD and understand current therapeutic and pre­
ventative regimens
• Understand phenotypic presentation of and current therapeutic options for chronic leg ulcerations
• Be familiar with the current screening, diagnostic, and therapeutic recommendations for pulmonary hypertension
in SCD

Background tologist. He followed regularly with his pediatric hematol­

Sickle cell disease (SCD) is a β­hemoglobinopathy that is ogist and is currently taking hydroxyurea, folic acid, and
hallmarked by red blood cell (RBC) sickling and chronic cholecalciferol. His SCD­related complications include
hemolytic anemia. Point mutations in both β­globin genes elevated transcranial Doppler velocity, 3 episodes of ACS,
lead to an unstable hemoglobin molecule (HbS),1 which and once­yearly VOE hospitalizations. At today’s visit, he
polymerizes into chains to form a “sickled” shape. This reports new episodes of painful, intermittent erections that
ultimately causes vaso­occlusion, intravascular hemolysis, occur about 4 nights per week and last about 3 to 4 hours.
and endothelial dysfunction. Although the complications They awaken him from sleep and sometimes resolve with
of SCD are broad, there are recognized overlapping “sub­ rigorous exercise.
phenotypes” of the disease. The high hemoglobin/high
viscosity subtype is associated with increased episodes of
acute chest syndrome (ACS) and vaso­occlusive episodes Priapism
(VOEs),2 whereas the hyperhemolytic subtype is associated Definition, types, prevalence, and pathophysiology
with increased vascular complications such as priapism, Priapism is defined as an erection lasting greater than 4
leg ulceration, and pulmonary hypertension.2­4 This chapterhours. Priapism can occur in patients with SCD of all ages
reviews the definitions, pathophysiology, diagnosis, and but may peak between 20 and 25 years of age.5 In SCD,
management of those vascular complications. increased RBC sickling occurs within the corpus cavernosa,
possibly as a result of abnormal endothelial adherence, rel­
ative acidosis during sleep and during erection, and defi­
cient erection control mechanisms involving an impaired
CLINICAL CASE nitric oxide–cyclic GMP pathway.6 Priapism is characterized
A 24­year­old man with SCD (genotype hemoglobin SS) as low flow (ischemic) or high flow (nonischemic/arterial).
Prakash
presents to a clinic to establish care Singh
with an adult Shekhawat
hema­ Nonischemic priapism is caused by trauma and damage to

Treatment priapism, ulcers, pulmonary hypertension | 411

the cavernosal artery and is not a uro­logic emer­gency as there is
no last­ing dam­age to penile tis­sue. Ischemic priapism, how­ever,
can cause last­ing tis­sue dam­age and requires prompt uro­logic
eval­u­at­ion. When ische­mic priapism lasts less than 4 hours and
recurs, it is referred to as stuttering. Both stuttering and ische­
mic priapism are described as pain­ful, which can help dif­fer­en­ti­
ate them from nonischemic priapism.7
Management
Although increased hydra­ tion and pain man­ age­ ment can be
attempted for priapism last­ing less than 4 hours,8 first-line ther­
apy for acute ische­mic priapism is decom­pres­sion via aspi­ra­tion,
followed by intracavernous injec­tion (ICI) of sym­pa­tho­mi­metic
agents.7,9 Penile blood is then sent for blood gas test­ing to con­
firm eti­­ol­ogy, although in patients with SCD, it is often ische­mic.
Sympathomimetics work via α-medi­ated vaso­con­stric­tion within
the cor­pus cavernosa, and uro­logic data sug­gest that aspi­ra­tion
plus ICI is more effec­tive than aspi­ra­tion alone.7 These data were
supported in a pro­spec­tive study of 15 patients with sickle cell
ane­mia.8
RBC exchange trans­fu­sions have also been shown to be safe
and effec­tive in both acute and pre­ven­ta­tive man­age­ment of pri­
apism.10 There have been pre­vi­ously described adverse effects
char­ac­ter­ized by cere­bral vas­cu­lar events (   Association of SCD,
priapism, exchange transfusion, and neurological events syn­
drome) in this set­ting,11,12 but it is thought to be more related to
post­trans­fu­sion hemo­glo­bin and asso­ci­ated increased vis­cos­ity,
rather than the actual trans­fu­sion itself.
Although no other ther­a­pies have been shown to be effec­tive
in large tri­als for the acute man­age­ment of priapism, case reports
sug­gest some ther­a­pies may be effec­tive in the pre­ven­ta­tive set­
ting. A 2017 Cochrane review ana­lyz­ing the data on these ther­
a­pies, which included ephed­ rine, sildenafil, stilboestrol, and
etilefrine, found no dif­fer­ence in effi­cacy between treat­ment and
pla­cebo in the pre­ven­tion of recur­rent priapism.13 Oral sym­pa­
tho­mi­met­ics such as ephed­rine, pseudoephedrine, and etilefrine
work sys­tem­i­cally via the same mech­a­nism as sym­pa­tho­mi­metic
ICI. Other pro­posed ther­a­pies dis­rupt the abnor­mal nitric oxide–
cGMP path­way and include phosphodiesterase 5 inhib­i­tors such
as sildenafil and tadalfil. Although some case reports sup­port
their use for ­reduc­tion in recur­rence with long-term dos­ing,13,14
the data on the use of sildenafil in pul­mo­nary hyper­ten­sion (PH)
raise con­cern for increased VOEs with its use.15 Hormonal ther­a­
pies sup­press andro­genic effects on penile erec­tion9 and include
GNrH a ­go­ nists, estro­gens, antiandrogens, and ketoconazole.
Despite reported suc­ cess in pre­ vent­
ing recur­ rent priapism
events, sig­nif­i­cant adverse effects include erec­tile dys­func­tion,
mood changes, and hot flashes and poten­tial for car­dio­vas­cu­
lar com­pli­ca­tions.9,14 Baclofen, a selec­tive γ-aminobutyric acid
recep­tor ago­nist, was effec­tive in treat­ment for sleep-related
priapism in a UK cohort study but was not effec­tive in the SCD
cohort.14
Although none of these ther­a­pies have been inves­ti­gated for
detu­mes­cence in acute priapism epi­sodes, they have been anec­
dot­ally used with some suc­cess. The com­bi­na­tion of pseudo­
ephedrine and baclofen has been par­tic­u­larly effec­tive in some
patients within our prac­tice and can be trialed while awaiting
(but should not delay) urol­ogy con­sul­ta­tion (Figure 1). Regarding
the main­stays of SCD ther­a­pies, the direct effect of hydroxy­urea
412  |  Hematology 2021  |  ASH Education Program
on priapism has not been well stud­ied. There are, how­ever, case
reports suggesting hydroxy­urea and trans­fu­sion are ben­efi ­ ­cial in
pre­vent­ing recur­rent epi­sodes of stuttering priapism.16,17
Novel ther­a­pies
A phase 2 sin­gle-arm trial is cur­rently inves­ti­gat­ing the effi­cacy
and safety of crizanlizumab in the man­age­ment of SCD-related
priapism (ClinicalTrials​­
.gov Identifier NCT03938454). Crizanli­
zumab is a mono­clo­nal anti­body targeting the leu­ko­cyte adhe­
sion mod­u­la­tor P-selectin and was approved by the US Food and
Drug Administration in 2019 to reduce VOEs in SCD.2
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/411/1851512/411azbell.pdf by guest on 13 December 2021
CLINICAL CASE (Con­tin­ued)
The patient is diag­nosed with stuttering priapism and is started
on daily baclofen for pre­ven­tion. The fre­quency of epi­sodes
decreases over 2 years, and he remains clin­ic­ ally sta­ble for the
next 10 years. However, he now reports a new, recur­rent wound
on his inner right ankle.
Leg ulcers
Definition, types, prev­a­lence, and path­o­phys­i­­ol­ogy
The prev­a­lence of sickle cell leg ulcers (SCLUs) is unknown, but
many patients report their first ulcer by the sec­ond decade of
life. The pro­posed path­o­phys­i­­ol­ogy in SCD involves predispos­
ing vasculopathy, local skin insult, pain-medi­ated neu­ro­genic
inflam­ma­tion, and low ten­sile strength of scar tis­sue.3 SCLUs are
often located in the lower extrem­i­ties, spe­cif­i­cally where there
is poor cir­cu­la­tion, thin skin, and less sub­cu­ta­ne­ous fat (eg, peri­
malleolar).3,4 Ulcers can be sin­gu­lar, stuttering, or chronic and fre­
quently recur at the site of pre­vi­ous ulcer­a­tions. A distinguishing
fea­ture from other venous ulcers is the pres­ence of pain.3 Prior
trauma is a poten­tial predisposing fac­tor.3,18 Other asso­ci­ated
fac­tors include HbSS geno­type,3,19 male sex, age older than
20 years, and reduced hemo­glo­bin F (HbF) lev­els.19 There has
been prior debate in the lit­er­a­ture as to the role of hydroxy­urea
and pre­dis­po­si­tion to SCLUs. However, sev­eral large reviews
have dis­ puted this. Of the 505 patients being screened for
pul­mo­nary hyper­ten­sion, 39% were tak­ing hydroxy­urea. There
was no sta­tis­ti­cal dif­fer­ence in prev­a­lence of leg ulcers among
patients with and with­out hydroxy­urea.20
Management
There are var­i­ous pro­posed inter­ven­tions for SCLUs, ­includ­ing
pharmaceutical treat­ments (vas­cu­lar, anti­ox­i­dants, growth fac­tors,
HbF syn­the­sis stim­u­la­tors), top­i­cal treat­ments (wound care, anti­
bi­ot­ics, growth fac­tors, ste­roids), and ­sur­gi­cal/nonpharmaceutical
agents. A 2020 Cochrane review found 6 ran­dom­ized con­trol
­tri­als (RCTs) inves­ti­gat­ing these ther­a­pies.4 One study inves­ti­gat­
ing isoxsuprine, a vas­cu­lar agent, had mixed results when eval­
u­at­ing com­plete ulcer clo­sure. Of the 30 patients, only 23.9%
healed (of those, 63.6% treat­ment vs 36.4% pla­cebo), 36.9% had
improved ulcers (47% treat­ment vs 53% pla­cebo), 15.2% had no
change, and 23.9% had dete­ri­o­ra­tion of ulcers.4,21 Arginine buty­
rate, an HbF stim­u­la­tor, dem­on­strated 30% com­plete clo­sure
com­pared with 8% in the stan­dard local care arm.4,22
 3RVVLEOH EHQHILWVRI+8
WUDQVIXVLRQ WKHUDS\ ORQJWHUP
3UHYHQWLRQ
RUDOV\PSDWKRPLPHWLFV
KRUPRQDO WKHUDSLHV

/DVWLQJKRXUV
VWXWWHULQJ

,QFUHDVHG RUDORU,9K\GUDWLRQ
$FXWH 3DLQPDQDJHPHQW
3VHXGRHSKHGULQH EDFORIHQ

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/411/1851512/411azbell.pdf by guest on 13 December 2021

3ULDSLVP

(While awaiting urology can trial

acute stuttering therapies)

8URORJLFFRQVXOWDWLRQ IRU
,VFKHPLF LQWUDFDYHUQRVDO DVSLUDWLRQ 
V\PSDWKRPLPHWLF LQMHFWLRQ
/DVWLQJKRXUV

$UWHULDO 1RLQWHUYHQWLRQ QHHGHG

Figure 1. How I treat priapism. For episodes lasting greater than 4 hours, urgent urologic consultation for prompt aspiration +/- sym­
pathomimetic injection is recommended. For stuttering episodes, it is reasonable to attempt increased hydration and pain manage­
ment. Long-term oral sympathomimetics may be beneficial in the prevention of recurrent episodes, as well as therapy with hydroxy­
urea (HU) or chronic transfusions. Hu, hydroxyurea; IV, intravenous.

Wound care. Local wound care is cru­cial to treating SCLUs
and involves debride­ment, infec­tion con­trol, and main­te­nance
of a moist wound envi­ron­ment.19 Appropriate com­pres­sive ban­
dages and gar­ments (eg, stock­ings) and Unna boots are impor­
tant in maintaining a moist envi­ron­ment for healing (Figure 2).
Topical anti­bi­ot­ics are not rou­tinely recommended due to risk
of con­tact sen­si­ti­za­tion, bac­te­rial resis­tance, and lack of mois­
ture bal­ance.19
Pain man­age­ment. SCLUs cause sig­nif­i­cant pain and mor­bid­
ity, and man­age­ment should include ade­quate anal­ge­sic ther­
apy. Topical anes­thet­ics may be used, but severe pain fre­quently
war­ rants opi­ oid ther­ apy. Neuropathic pain may respond to
selec­tive sero­to­nin recep­tor inhib­i­tors, pregabalin, gabapentin,
or tri­cy­clic anti­de­pres­sants. Given the chro­nic­ity of some ulcers,
they may require chronic, not just acute, reg­i­mens.19
Transfusions.  Chronic trans­fu­sions decrease HbS per­cent­age,
RBC sick­ling, and hemo­ly­sis-related pathol­ogy. Transfusions
have been used in both acute and pre­ven­ta­tive set­tings19 for
SCLUs and as perioperative sup­port for patients who require
sur­gery.3 There are no RCTs to deter­mine goal hemo­glo­bin, HbS
­per­cent­age, or dura­tion of ther­apy. Generally, a 6-month trial is
used to eval­u­ate for response, and if the response is pos­i­tive,
trans­fu­sions are con­tin­ued until com­plete ulcer clo­sure.18
Novel ther­a­pies. A phase 2 tri­ple-blind RCT assessing the tol­
er­a­bil­
ity and effi­ cacy of top­ i­
cal sodium nitrate is under way
(­ClinicalTrials​­.gov Identifier NCT02863068), after the 2014 phase 1
cohort trial results were prom­is­ing and dem­on­strated a significant
increase in periwound blood flow, as well as dose-depen­dent
decreases in SCLU size.23
In addi­tion, a dou­ble-blind RCT is eval­u­at­ing the safety and
effi­cacy of intra­der­mal deferoxamine patches (ClinicalTrials​­.gov
Identifier NCT04058197).
CLINICAL CASE (Con­tin­ued)
Our patient’s ulcer com­pletely resolved with wound care and a
short trans­fu­sion pro­to­col. He required a 6-month dose increase
in his chronic pain reg­ i­
men and did not have any fur­ ther
recur­rences over the past 6 years. At his office visit today, he
notes wors­en­ing dyspnea with reg­u­lar activ­i­ties and increas­
ing fatigue.

Prakash Singh Shekhawat

Treatment priapism, ulcers, pul­mo­nary hyper­ten­sion  |  413
 &RPSUHVVLYH EDQGDJHV
&RPSUHVVLRQJDUPHQWV
/RFDOGHEULGHPHQW
Wound Care 0RLVWZRXQGEHGPDLQWHQDQFH
7UHDWPHQW RIVXSHULPSRVHG
LQIHFWLRQV

7RSLFDODQHVWKHWLFV
16$,'VRSLRLGVGHSHQGLQJRQ
Analgesia VHYHULW\ RISDLQ
$GHTXDWHLQFUHDVHG DQDOJHVLFV

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/411/1851512/411azbell.pdf by guest on 13 December 2021

IRUGUHVVLQJFKDQJHV

$YRLGLOOILWWLQJ VKRHV
$YRLGWUDXPD
Prevention 7UHDWPHQW RIXQGHUO\LQJ
GLVHDVH6&'WKHUDS\
Figure 2. Multimodal approach to SCLUs. The treatment of leg ulcers requires comprehensive wound care, adequate analgesics, and
preventative measures. NSAID, nonsteroidal anti-inflammatory drug.
Pulmonary hyper­ten­sion
Presentation and diag­no­sis
Progressive dyspnea on exer­tion or exer­cise lim­i­ta­tions, espe­
cially with exertional hyp­oxia, should prompt fur­ther inves­ti­ga­
tion for PH.24 On exam­i­na­tion, patients may have signs of right
heart fail­ure, includ­ing ele­vated jug­u­lar venous pul­sa­tion, accen­
tu­ated or fixed S2 split­ting, tri­cus­pid regur­gi­ta­tion (TR) mur­mur,
and/or periph­eral edema.
Although right heart cath­e­ter­i­za­tion (RHC) is the gold stan­
dard of diag­no­sis, its inva­sive­ness has led to the devel­op­ment of
more prac­ti­cal prog­nos­tic tools to pre­dict the pres­ence of PH. TR
max­i­mum veloc­ity on echo­car­dio­gram can esti­mate pul­mo­nary
arte­rial sys­tolic pres­sure. A TR veloc­ity (TRV) more than 2.5 m/s
was dem­on­strated to pre­dict the pres­ence of PH,24 and recent
stud­ies have shown that greater than 50% of patients with a TRV
of 2.9 m/s are diag­nosed with PH on RHC.25 Elevated brain natri­
uretic peptide (BNP) lev­els are also asso­ci­ated with the pres­ence
of PH, and ele­vated N-terminal-pro-BNP lev­els are an inde­pen­
dent risk fac­tor for mor­tal­ity in patients with SCD.24,26,27 A 2011
pro­spec­tive study in France found that in patients with a TRV
from 2.5 to 2.8 m/s, NT-pro-BNP more than 164, and a 6-min­ute
walk dis­tance less than 333 m, the pos­i­tive pre­dic­tive value of
diag­nos­ing PH on RHC was 62% (false-pos­it­ ive rate of 7%).28
Screening
Despite the asso­ci­a­tion of PH with increased mor­tal­ity, there
is no con­sen­sus regard­ing screen­ing guide­lines, par­tially due
414  |  Hematology 2021  |  ASH Education Program
RSWLPL]DWLRQ
to the lack of evi­dence dem­on­strat­ing treat­ment ben­e­fit.29
The Amer­ i­
can Thoracic Society published guide­ lines in 2014
regard­ing risk strat­i­fi­ca­tion, screen­ing, and man­age­ment of
PH in patients with SCD. Based on evi­dence that suggested
13% of patients with a nor­mal echo­car­dio­gram develop an ele­
vated TRV after 3 years of fol­low-up,30 the Amer­i­can Thoracic
Society recommended a screen­ing echo­car­dio­gram every 1 to
3 years in all­patients with SCD.31 However, fol­low-up stud­ies sug­
gest that the rate of pro­gres­sion to PH may actu­ally be slower.32
The most recent National Institutes of Health and Amer­ i­
can
Society of Hematology guide­lines rec­om­mend against rou­tine
screen­ing in asymp­tom­atic patients with SCD33,34 and rec­om­
mend obtaining a diag­nos­tic echo­car­dio­gram only for symp­
tom­atic patients (Figure 3).
Definition, types (groups), prev­a­lence, and path­o­phys­i­­ol­ogy
PH is defined as a rest­ ing mean pul­ mo­ nary arte­rial pres­
sure
(mPAP) more than 25 mm Hg by RHC.24 In SCD, recur­rent hemo­ly­
sis and its dis­rup­tion of the vas­cu­lar endo­the­lium ulti­mately leads
to smooth mus­cle pro­lif­er­a­tion and adven­ti­tial fibro­blast accu­
mu­la­tion.35 This, in turn, leads to increased pul­mo­nary vas­cu­lar
resis­tance, right ven­tric­u­lar (RV) dys­func­tion, and decreased
car­diac out­put.36 The prev­al­ence of PH is thought to be 6% to
10% in SCD36 and can be fur­ther char­ac­ter­ized by the loca­tion of
hyper­ten­sion rel­a­tive to the cap­il­lary sys­tem.
Precapillary or pul­mo­nary arte­rial hyper­ten­sion (PAH) is defined
by a mPAP greater than 25 mm Hg and nor­mal left ven­tric­u­lar end-
 Signs and Symptoms
• '\VSQHD DWUHVWRUZLWKH[HUWLRQWKDWLVRXWRI
SURSRUWLRQWRNQRZQFRQGLWLRQ
• +\SR[HPLDDWUHVWRUZLWKH[HUWLRQ
• &KHVW SDLQDWUHVWRUZLWKH[HUWLRQ
• ,QFUHDVHLQH[HUFLVHOLPLWDWLRQ
• +LVWRU\RIUHFXUUHQWK\SR[HPLDDWUHVWRUZLWKH[HUWLRQ
• (YLGHQFHRIVOHHSGLVRUGHUHGEUHDWKLQJ
K\SR[HPLD
• +LVWRU\RIV\QFRSHRUSUHV\QFRSH
• 6ZLWKSUHGRPLQDQW3RUQHZXQH[SHFWHGPXUPXU 759PV
• 6LJQVRIKHDUWIDLOXUHDQGRUIOXLGRYHUORDG
And one or both of the following:

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/411/1851512/411azbell.pdf by guest on 13 December 2021

0:'P
2EWDLQD %13!SJP/
5HIHUWR
GLDJQRVWLF
3+VSHFLDOLVW
HFKRFDUGLRJUDP

759PV
Comorbid conditions
• &RQQHFWLYHWLVVXH GLVHDVH
• 8QGHUO\LQJSXOPRQDU\GLVHDVH DVWKPD
*Assist with SCD-specific periprocedural planning:
Disease complications associated with PH • &RQVLGHUWUDQVIXVLRQV
• /HJXOFHUV • +ROGDQWLFRDJXODWLRQLIDSSOLFDEOH
• 3ULDSLVP • (QVXUHZDUPEODQNHWDQGSRVWSURFHGXUH,9IOXLGV
• $OEXPLQXULD DUHUHDGLO\DYDLODEOH

Figure 3. How I approach PH screening. Every patient with SCD should be screened for “red-flag” signs and symptoms of PH at
routine visits (when there is no acute illness present). Any red-flag sign/symptom, or the presence of certain comorbidities or dis­
ease-specific complications, should be further evaluated with a diagnostic echocardiogram. A moderately elevated TRV, or a mild
elevation with other prognostic factors, should warrant further referral to a specialist. IV, intravenous; 6MWD, 6-minute walk distance.

dia­stolic pres­sure less than 15 mm Hg and falls under PH group I.
Postcapillary or pul­mo­nary venous hyper­ten­sion is defined by an
mPAP greater than 25 mm Hg with an ele­vated left ven­tric­u­lar
end-dia­stolic pres­sure higher than 15 mm Hg,24 and it can be fur­
ther grouped into PH World Health Organization groups II to V
based on eti­­ol­ogy. Pre- and postcapillary are dis­trib­uted evenly
in patients with SCD, and patients often have fea­tures of both.25,35
Management
PAH-spe­cific ther­a­pies include the fol­low­ing.
Endothelin-recep­tor antag­o­nists. Endothelin 1 is a potent vaso­
con­stric­tor that is ele­vated in SCD. The Randomized, Placebo-
Controlled, Double-Blind, Multicenter, Parallel Group Study to As­
sess the Efficacy, Safety and Tolerability of Bosentan in Patients
With Symptomatic Pulmonary Arterial Hypertension Associated
With Sickle Cell Disease (ASSET)-1 and -2 tri­als in 2009 were de­
signed to inves­ti­gate the safety, effi­cacy, and tol­er­a­bil­ity of bosen­
tan, an endothelin 1 recep­tor antag­o­nist, in patients with PAH and
PH, respec­tively. Unfortunately, both tri­als were ter­mi­nated early
due to slow enroll­ment. However, the lim­ited data obtained dem­
on­strated tol­er­a­bil­ity and increased exer­cise tol­er­ance,37 which
were fur­ther supported by the find­ings of Minniti et al38 in 2009.
In that cohort study, patients with RHC-con­firmed PH receiv­ing
bosentan or ambrisentan had reduced BNP lev­els, reduced TRVs,
and improved 6-minute walk test.
PDE-5 inhib­i­tors. Although sildenafil has been shown to de­
crease TRV and BNP,39 a 2011 mul­ti­cen­ter trial of sildenafil was
stopped pre­ma­turely due to increased hos­pi­tal­i­za­tion for VOEs.
Potential con­tri­bu­tions to these events include lack of SCD ther­
apy opti­mi­za­tion prior to the inter­ven­tion and sildenafil-spe­cific
mech­a­nisms that may cause myal­gias and inflam­ma­tory pain.15
Other PAH ther­a­pies. Prostanoids (epoprostenol, treprostinil,
iloprost, beraprost) have not spe­cif­i­cally been stud­ied in SCD
but have shown ben­e­fit in non-SCD patients with PH.40 In addi­
tion, 1 case study dem­on­strated decreased pul­mo­nary arte­rial
sys­tolic pres­sure in 10 patients treated with 5 days of L-argi­nine.41
PH ther­a­pies
Intensification of SCD-spe­cific ther­a­pies. The direct ben­e­fit of
hydroxy­urea on PH has not been stud­ied, but it reduces epi­
sodes of VOEs and ACS, both of which are asso­ci­ated with an

Prakash Singh Shekhawat

Treatment priapism, ulcers, pul­mo­nary hyper­ten­sion  |  415
increase in PA pres­sures. Hydroxyurea should there­fore indi­rectly
decrease the risk of acute right-sided heart fail­ure and death in
patients with under­ly­ing PH.24 The effect of chronic trans­fu­sions
on PH has also not been for­mally stud­ied; how­ever, because it
reduces com­pli­ca­tions of SCD sim­i­lar to hydroxy­urea, it should
the­o­ret­i­cally improve out­comes in patients with under­ly­ing PH.
Treatment of comorbidities.  Patients who have con­cur­rent right
heart fail­ure or albu­min­uria can be treated with diuret­ics but cau­
tiously, given the pos­si­bil­ity of vol­ume deple­tion and increased
RBC sick­ling.24,31 In addi­tion, patients with under­ly­ing sleep ap­
nea and throm­botic dis­ease should be man­aged accord­ingly.
Novel ther­a­pies
A phase 2 dou­ble-blind RCT is eval­u­at­ing the safety, tol­er­a­bil­
ity, and effi­cacy of riociguat (ClinicalTrials​­.gov NCT02633397) in
patients with SCD. Several of the sec­ond­ary out­come mea­sures
spe­cif­i­cally relate to mark­ers of PH.
In addi­tion, a phase 3 ran­dom­ized par­al­lel-arm trial is eval­u­at­
ing the effect of RBC exchange ther­apy com­pared with stan­dard
of care on car­dio­vas­cu­lar risk (ClinicalTrials​­.gov NCT04084080).
Secondary out­comes spe­cif­i­cally eval­u­ate the effect on mark­ers
of PH.
Conflict-of-inter­est dis­clo­sure
Roberta C.G. Azbell: no conflicts to disclose.
Payal Chandarana Desai: con­sul­tant for GBT for grant review;
advi­sory board for Forma; funding from the National Institutes of
Health, University of Tennessee, and University of Pittsburgh; and
speaker for Novartis.
Off-label drug use
Roberta C.G. Azbell: During the discussion of prevention of recur­
rent priapism, we discuss the off-label use of oral sympathomimet­
ics, baclofen, phosphodiesterase 5 inhibitors, hormonal therapies
(GnRH agonists, estrogens, antiandrogens), ketoconazole, hydroxy­
urea, and exchange transfusion. With regard to management
of SCLUs, we discuss the off-label use of exchange transfusions,
hydroxyurea, topical sodium nitratre, and intradermal deferoxamine
patches. We discuss the off-label use of hydroxyurea in pulmonary
hypertension.
Payal Chandarana Desai: During the discussion of prevention of
recurrent priapism, we discuss the off-label use of oral sympatho­
mimetics, baclofen, phosphodiesterase 5 inhibitors, hormonal ther­
apies (GnRH agonists, estrogens, antiandrogens), ketoconazole,
hydroxyurea, and exchange transfusion. With regard to manage­
ment of SCLUs, we discuss the off-label use of exchange transfusions,
hydroxyurea, topical sodium nitratre, and intradermal deferoxamine
patches. We discuss the off-label use of hydroxyurea in pulmonary
hypertension.
Correspondence
Payal Chandarana Desai, The Ohio State University, 1800 Cannon
Drive, 1100D Lincoln Tower, Columbus, OH 43210. e-mail: payal​
­.desai@osumc​­.edu.
References
1. Rees DC, Williams TN, Gladwin MT. Sickle-cell dis­ ease. Lancet. 2010;
376(9757):2018-2031.
2. Ofori-Acquah SF. Sickle cell dis­
ease as a vas­
cu­
lar dis­
or­
der. Expert Rev
Hematol. 2020;13(6):645-653.
416  |  Hematology 2021  |  ASH Education Program
3. Minniti CP, Kato GJ. How we treat sickle cell patients with leg ulcers clin­i­cal
cases 40th anni­ver­sary issue. Am J Hematol. 2016;91(1):22-30.
4. Martí-Carvajal AJ, Knight-Madden JM, Martinez-Zapata MJ. Interventions for
treating leg ulcers in peo­ple with sickle cell dis­ease. Cochrane Database
Syst Rev. 2014;2014(12):CD008394.
5. Emond AM, Holman R, Hayes RJ, Serjeant GR. Priapism and impo­tence in
homo­zy­gous sickle cell dis­ease. Arch Intern Med. 1980;140(11):1434-1437.
6. Bivalacqua TJ, Burnett AL. Priapism: new con­cepts in the path­o­phys­i­­ol­ogy
and new treat­ment strat­e­gies. Curr Urol Rep. 2006;7(6):497-502.
7. Mon­ta­gue DK, Jarow J, Broderick GA, et al; Members of the Erectile Dys­
function Guideline Update Panel; Amer­i­can Urological Association. Amer­
i­can Urological Association guide­line on the man­age­ment of priapism. J
Urol. 2003;170(4, pt 1):1318-1324.
8. Mantadakis E, Ewalt DH, Cavender JD, Rogers ZR, Buchanan GR. Outpatient
penile aspi­ra­tion and epi­neph­rine irri­ga­tion for young patients with sickle
cell ane­mia and prolonged priapism. Blood. 2000;95(1):78-82.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/411/1851512/411azbell.pdf by guest on 13 December 2021
9. Broderick GA. Priapism and sickle–cell ane­mia: diag­no­sis and non­sur­gi­cal
ther­apy. J Sex Med. 2012;9(1):88-103.
10. Ballas SK, Lyon D. Safety and effi­cacy of blood exchange trans­fu­sion for
priapism com­pli­cat­ing sickle cell dis­ease. J Clin Apher. 2016;31(1):5-10.
11. Siegel JF, Rich MA, Brock WA. Association of sickle cell dis­ease, priapism,
exchange trans­fu­sion and neu­ro­log­i­cal events: ASPEN syn­drome. J Urol.
1993;150(5, pt 1):1480-1482.
12. Rackoff WR, Ohene-Frempong K, Month S, Scott JP, Neahring B, Cohen AR.
Neurologic events after par­tial exchange trans­fu­sion for priapism in sickle
cell dis­ease. J Pediatr. 1992;120(6):882-885.
13. Chinegwundoh FI, Smith S, Anie KA. Treatments for priapism in boys
and men with sickle cell dis­ease. Cochrane Database Syst Rev. 2017;9(9):
CD004198.
14. Johnson MJ, McNeillis V, Chiriaco G, Ralph DJ. Rare dis­or­ders of pain­ful
erec­tion: a cohort study of the inves­ti­ga­tion and man­age­ment of stuttering
priapism and sleep-related pain­ful erec­tion. J Sex Med. 2021;18(2):376-384.
15. Machado RF, Barst RJ, Yovetich NA, et al; walk-PHaSST Investigators
and Patients. Hospitalization for pain in patients with sickle cell dis­ease
treated with sildenafil for ele­vated TRV and low exer­cise capac­ity. Blood.
2011;118(4):855-864.
16. Anele UA, Mack AK, Resar LMS, Burnett AL. Hydroxyurea ther­apy for pria­
pism pre­ven­tion and erec­tile func­tion recov­ery in sickle cell dis­ease: a case
report and review of the lit­er­a­ture. Int Urol Nephrol. 2014;46(9):1733-1736.
17. Hassan A, Jam’a A, Al Dabbous IA. Hydroxyurea in the treat­ment of sickle
cell asso­ci­ated priapism. J Urol. 1998;159(5):1642.
18. Trent JT, Kirsner RS. Leg ulcers in sickle cell dis­ease. Adv Skin Wound Care.
2004;17(8):410-416.
19. Ladizinski B, Bazakas A, Mistry N, Alavi A, Sibbald RG, Salcido R. Sickle cell
dis­ease and leg ulcers. Adv Skin Wound Care. 2012;25(9):420-428.
20. Minniti CP, Taylor JG VI, Hildesheim M, et  al. Laboratory and echo­car­di­
og­ra­phy mark­ers in sickle cell patients with leg ulcers. Am J Hematol.
2011;86(8):705-708.
21. Serjeant GR, Howard C. Isoxsuprine hydro­chlo­ride in the ther­apy of sickle
cell leg ulcer­a­tion. West Indian Med J. 1977;26(3):164-166.
22. McMahon L, Tamary H, Askin M, et al. A ran­dom­ized phase II trial of argi­nine
buty­rate with stan­dard local ther­apy in refrac­tory sickle cell leg ulcers. Br J
Haematol. 2010;151(5):516-524.
23. Minniti CP, Gorbach AM, Xu D, et al. Topical sodium nitrite for chronic leg
ulcers in patients with sickle cell anae­mia: a phase 1 dose-find­ing safety
and tol­er­a­bil­ity trial. Lancet Haematol. 2014;1(3):e95-e103.
24. Ataga KI, Klings ES. Pulmonary hyper­ten­sion in sickle cell dis­ease: diag­
no­sis and man­age­ment. Hematology Am Soc Hematol Educ Program.
2014;2014(1):425-431.
25. Gordeuk VR, Castro OL, Machado RF. Pathophysiology and treat­ment of
pul­mo­nary hyper­ten­sion in sickle cell dis­ease. Blood. 2016;127(7):820-828.
26. Machado RF, Anthi A, Steinberg MH, et al. Risk of death in sickle cell dis­
ease. Vasc Med. 2006;296(3):310-318.
27. Machado RF, Hildesheim M, Mendelsohn L, Remaley AT, Kato GJ, Gladwin
MT. NT-pro brain natri­uretic pep­tide lev­els and the risk of death in the coop­
er­at­ ive study of sickle cell dis­ease. Br J Haematol. 2011;154(4):512-520.
28. Parent F, Bachir D, Inamo J, et  al. A hemo­dy­namic study of pul­mo­nary
hyper­ten­sion in sickle cell dis­ease. N Engl J Med. 2011;365(1):44-53.
29. Yawn BP, Buchanan GR, Afenyi-Annan AN, et al. Management of sickle cell
dis­ease: sum­ mary of the 2014 evi­ dence-based report by expert panel
mem­bers. JAMA. 2014;312(10):1033-1048.
30. Ataga KI, Moore CG, Jones S, et al. Pulmonary hyper­ten­sion in patients with
sickle cell dis­ease: a lon­gi­tu­di­nal study. Br J Haematol. 2006;134(1):109-115.
31. Klings ES, Machado RF, Barst RJ, et al; Amer­i­can Thoracic Society Ad Hoc
Committee on Pulmonary Hypertension of Sickle Cell Disease. An offi­cial
Amer­i­can Thoracic Society clin­i­cal prac­tice guide­line: diag­no­sis, risk strat­
i­fi­ca­tion, and man­age­ment of pul­mo­nary hyper­ten­sion of sickle cell dis­
ease. Am J Respir Crit Care Med. 2014;189(6):727-740.
32. Desai PC, May RC, Jones SK, et al. Longitudinal study of echo­car­di­og­ra­phy-
derived tri­cus­pid regurgitant jet veloc­ity in sickle cell dis­ease. Br J Haema-
tol. 2013;162(6):836-841.
33. Liem RI, Lanzkron S, D Coates T, et  al. Amer­i­can Society of Hematology
2019 guide­lines for sickle cell dis­ease: car­dio­pul­mo­nary and kid­ney dis­
ease. Blood Adv. 2019;3(23):3867-3897.
34. National Heart, Lung, and Blood Institute. Evidence-based man­age­ment
of sickle cell dis­ease: expert panel, 2014. Accessed 26 May 2019. https:­/​/​­
­​­www​­.nhlbi​­.nih​­.gov​­/sites​­/default​­/files​­/media​­/docs​­/sickle​­-cell​­-disease​
­-report 020816_0.pdf
35. Wood KC, Gladwin MT, Straub AC. Sickle cell dis­ease: at the cross­roads of
pul­mo­nary hyper­ten­sion and dia­stolic heart fail­ure. Heart. 2020;106(8):562-
568.
36. Potoka KP, Gladwin MT. Vasculopathy and pul­mo­nary hyper­ten­sion in sickle
cell dis­ease. Am J Physiol Lung Cell Mol Physiol. 2015;308(4):L314-L324.
Prakash Singh Shekhawat
37. Barst RJ, Mubarak KK, Machado RF, et al; ASSET study group. Exercise
capac­ity and haemodynamics in patients with sickle cell dis­ease with pul­
mo­nary hyper­ten­sion treated with bosentan: results of the ASSET stud­ies.
Br J Haematol. 2010;149(3):426-435.
38. Minniti CP, Machado RF, Coles WA, Sachdev V, Gladwin MT, Kato GJ. Endo­
thelin recep­tor antag­o­nists for pul­mo­nary hyper­ten­sion in adult patients
with sickle cell dis­ease. Br J Haematol. 2009;147(5):737-743.
39. Machado RF, Martyr S, Kato GJ, et al. Sildenafil ther­apy in patients with sickle
cell dis­ease and pul­mo­nary hyper­ten­sion. Br J Haematol. 2005;130(3):445-
453.
40. Machado RF, Gladwin MT. Chronic sickle cell lung dis­ease: new insights
into the diag­no­sis, path­o­gen­e­sis and treat­ment of pul­mo­nary hyper­ten­
sion. Br J Haematol. 2005;129(4):449-464.
41. Morris CR, Morris SM Jr, Hagar W, et al. Arginine ther­apy: a new treat­ment
for pul­mo­nary hyper­ten­sion in sickle cell dis­ease? Am J Respir Crit Care
Med. 2003;168(1):63-69.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/411/1851512/411azbell.pdf by guest on 13 December 2021
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000275
Treatment priapism, ulcers, pul­mo­nary hyper­ten­sion  |  417
 MDS: BEYOND A ONE - SIZE - FITS -ALL APPROACH

Have we reached a molecular era

in myelodysplastic syndromes?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/418/1851957/418voso.pdf by guest on 13 December 2021

Maria Teresa Voso1,2 and Carmelo Gurnari1,3
Department of Biomedicine and Prevention, University of Rome Tor Vergata, Rome, Italy; 2Santa Lucia Foundation, IRCCS, Neuro­Oncohematology,
1

Rome, Italy; and 3Translational Hematology and Oncology Research Department, Taussig Cancer Center, Cleveland Clinic, OH

Myelodysplastic syndromes (MDS) are characterized by heterogeneous biological and clinical characteristics, leading
to variable outcomes. The availability of sophisticated platforms of genome sequencing allowed the discovery of recur-
rently mutated genes, which have led to a new era in MDS. This is reflected by the 2016 update of the World Health
Organization classification, in which the criteria to define MDS with ringed sideroblasts include the presence of SF3B1
mutations. Further, the detection of somatic mutations in myeloid genes at high variant allele frequency guides the diag-
nostic algorithm in cases with cytopenias, unclear dysplastic changes, and normal karyotypes, supporting MDS over
alternative diagnoses. SF3B1 mutations have been shown to play a positive prognostic role, while mutations in ASXL1,
EZH2, RUNX1, and TP53 have been associated with a dismal prognosis. This is particularly relevant in lower- and interme-
diate-risk disease, in which a higher number of mutations and/or the presence of “unfavorable” somatic mutations may
support the use of disease-modifying treatments. In the near future, the incorporation of mutation profiles in currently
used prognostication systems, also taking into consideration the classical patient clinical variables (including age and
comorbidities), will support a more precise disease stratification, eg, the assignment to targeted treatment approaches
or to allogeneic stem cell transplantation in younger patients.

LEARNING OBJECTIVES
• Discuss the impact of somatic mutations in MDS and how to integrate molecular data within current diagnostic
and prognostic classifications
• Review the clinical management of patients with MDS according to the disease mutational status and identify
actionable targets

CLINICAL CASE 1 findings, a myelodysplastic syndrome (MDS) with RS and

A 78­year­old woman with an unremarkable medical his­
tory presented to our clinic with isolated macrocytic
anemia (hemoglobin [Hb], 9 g/dL; mean corpuscular vol­
ume, 107 fL). Metabolic as well as nutritional deficien­
cies and other common causes of anemia were ruled out
after an initial diagnostic workup, while the endogenous
erythropoietin (EPO) level was 358 mU/mL. The bone
marrow (BM) evaluation revealed expansion of erythro­
poiesis (70% of marrow cellularity) with marked dysplas­
tic changes in 20% of erythroid cells and the presence of
1% blasts and 12% ring sideroblasts (RS) by iron staining
(Perls’ Prussian blue; Figure 1). The karyotype was nor­
mal, while next­generation sequencing (NGS) mutational
analysis identified a canonical SF3B1K700E mutation at a
variant allelic frequency (VAF) of 21%. Based on these
single­lineage dysplasia was diagnosed.
Introduction
MDS is a highly heterogeneous group of disorders defined
by the presence of ineffective hematopoiesis with periph­
eral blood (PB) cytopenias, dysplastic changes in ≥10% of
cells of one or more myeloid lineages, and a variable risk
of progression to acute myeloid leukemia (AML).1 Clinical
presentations of MDS reflect its biological heterogeneity.
Based on the major risk­prognostication systems, the In­
ternational Prognostic Scoring System (IPSS) and its revi­
sion (IPSS­R),2, 3 two­thirds of patients will present with
lower­risk disease (LR­MDS) and are generally treated with
supportive care. In these cases the goal of treatment is the

418 | Hematology 2021 | ASH Education Program


Figure 1. Perls’ Prussian blue stain of the patient’s BM smear
showing RS. Classical type 3 RS are shown (according to Mufti
et al57), with multiple siderotic granules in a perinuclear position
surrounding the nucleus (lower-right cell) or encompassing at
least one-third of the nuclear circumference (middle cell).
ame­lio­ra­tion of cytopenias, represented by ane­mia in about 70%
of cases, through trans­fu­sion sup­port, eryth­ro­poi­e­tin admin­is­tra­
tion, or lenalidomide treat­ment in MDS with del(5q). Conversely,
higher-risk MDS patients, defined by inter­me­di­ate to very high
IPSS-R,4 require dis­ease-mod­i­fy­ing treat­ments (DMT) such as hy­
pomethylating agents (HMA) and a timely eval­u­a­tion for allo­ge­
neic hema­to­poi­etic stem cell trans­plan­ta­tion (HSCT), the most
effec­tive treat­ment and the only poten­tially cura­tive option.5
Integration of molec­u­lar data into clas­si­fi­ca­tion
and prog­nos­ti­ca­tion sys­tems
Paralleling the var­i­able clin­i­cal pre­sen­ta­tion, MDS biol­ogy is fas­
ci­nat­ingly com­plex. In the last decade, the avail­abil­ity of sophis­
ti­cated genome-sequenc­ing plat­forms led to the iden­ti­fi­ca­tion
of recur­rently mutated genes, shed­ding light on new aspects of
the dis­ease patho­bi­­ol­ogy. Accordingly, del(5q) and SF3B1 muta­
tions have been incor­po­rated in the most recent World Health
Organization (WHO) clas­si­fi­ca­tion and define spe­cific MDS sub­
types.1 However, the com­mon risk-prog­nos­ti­ca­tion tools (ie, the
IPSS and IPSS-R), used world­wide for clin­i­cal deci­sion-mak­ing
and trial eli­gi­bil­ity, consider only clin­i­cal/cyto­ge­netic var­i­ables.
Briefly, the IPSS score groups patients in a 4-tier model (low,
inter­me­di­ate-1, inter­me­di­ate-2, and high),2 while the more recent
IPSS-R includes 5 risk groups (very low, low, inter­me­di­ate, high,
and very high), improv­ing strat­i­fi­ca­tion of kar­yo­type and of the
pro­ por­tion of blasts and tak­ ing into account the sever­ ity of
cytopenias (Table 1).3 The IPSS-R, ini­tially devel­oped in patients
receiv­ing sup­port­ive treat­ment, has since been val­i­dated in
patients treated with DMT and ther­apy-related MDS.6,7
The advent of NGS and the grow­ing body of evi­dence con­
cerning the prog­nos­tic impact of somatic muta­tions in MDS have
posed new chal­lenges, such as the incor­po­ra­tion of this infor­
ma­tion into the established risk-strat­i­fi­ca­tion mod­els. To this
end, sev­eral stud­ies have dem­on­strated the impact of somatic
Prakash Singh Shekhawat
mye­ loid gene muta­ tions on sur­vival in patients with MDS.8-10
The num­ber of muta­tions itself has been shown to play a sig­nif­
i­cant prog­nos­tic role by most stud­ies, with patients car­ri­ers of
over 3 driver muta­tions char­ac­ter­ized by reduced sur­vival and a
higher prob­a­bil­ity of leu­ke­mia pro­gres­sion.11-13 This is par­tic­u­larly
rel­e­vant in lower- or inter­me­di­ate-risk MDS, in which a higher
num­ber of somatic muta­tions, or the pres­ence of “unfa­vor­able”
muta­tions (ie, TP53), may refine the prog­no­sis and set the indi­
ca­tion for DMT or HSCT.14 In a sur­vey of 104 genes, Haferlach et
al showed by uni­var­i­ate anal­y­sis that among the 25 genes found
to affect sur­vival, only 5 (ASXL1, KRAS, PRPF8, RUNX1, and SF3B1)
retained sig­nif­i­cance after correcting for confounding fac­tors.8
These muta­tions were incor­po­rated into a novel hybrid geno­
mic-clin­i­cal model that more accu­rately predicted MDS patients’
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/418/1851957/418voso.pdf by guest on 13 December 2021
prog­no­sis when com­pared to both the IPSS-R and the “gene-
only” model.8 Another study from the Cleveland Clinic iden­ti­fied
one favor­able (SF3B1) and 5 unfa­vor­able gene muta­tions (ASXL1,
EZH2, NPM1, RUNX1, and TP53) in a cohort of 508 MDS patients.10
These data led to the design of a score, includ­ing age, IPSS-R,
and the muta­tional sta­tus of EZH2, SF3B1, and TP53, that out­
performed the clas­ si­cal MDS risk-pre­ dic­tion mod­ els (Table 1)
and could pre­dict lon­gi­tu­di­nal dis­ease evo­lu­tion regard­less of
treat­ment.10 Of note, these scores unveiled discrepancies in risk
assign­ment for a not neg­li­gi­ble frac­tion of patients. For instance,
if tak­ing into con­sid­er­ation the IPSS-R, the hybrid “powered”
model upstaged 58% of patients pre­vi­ously cat­e­go­rized as inter­
me­di­ate risk to high as well as 26% from lower to high risk.9
More recently, by apply­ing Bayes­ian net­works and Dirichlet
pro­cesses in a study enroll­ing 2043 patients anno­tated for the
pres­ence of muta­tions in 47 fre­quently mutated genes, Bersanelli
et al iden­ti­fied 8 clus­ters, each char­ac­ter­ized by dif­fer­ent genetic
lesions, clin­i­cal phe­no­types, and prog­no­ses.15 By using a ran­dom-
effects Cox mul­ti­state model, the authors com­bined 63 clin­i­cal
and geno­mic var­i­ables to pro­file indi­vid­ual patients’ prog­no­ses,
yield­ing a C-index of 0.75 and 0.74 when cross-val­i­dated inter­
nally and in an inde­pen­dent cohort, respec­tively.15 This sem­i­nal
study paves the way for an MDS prog­nos­ti­ca­tion sys­tem not only
account­ing for tra­di­tional clin­i­cal and mor­pho­log­i­cal var­i­ables
but also incor­po­rat­ing recent advances in MDS biol­ogy.
Limitations to the appli­ca­tion of geno­mics
to MDS diag­nos­tics
If the idea of a hybrid geno­mic-clin­i­cal model sounds appeal­ing,
sev­eral fac­tors still play a role in deter­min­ing the actual fea­si­
bil­ity and repro­duc­ibil­ity of such an approach. There is no gen­
eral con­sen­sus yet on sequenc­ing tech­niques and data anal­y­sis
across lab­o­ra­to­ries. The clin­i­cal impact of some var­i­ants on pa­
tients’ out­comes remains unknown, and of utmost impor­tance,
bio­log­i­cal dif­fer­ences in muta­tion char­ac­ter­is­tics such as VAF,
type (mis­sense vs trun­cat­ing), and loca­tion (func­tional domain
vs other posi­tions across the gene) are still not homo­ge­neously
con­sid­ered. For instance, TP53 muta­tions are clas­si­cally con­sid­
ered prognostically unfa­vor­able. However, a recent coop­er­a­tive
study dem­on­strated that their clin­i­cal impact is beyond the mere
binary pres­ence/absence.16 Indeed, Bernard et al con­firmed that
besides VAF, the TP53 alle­lic state is a strong pre­dic­tive fac­tor,
show­ing that patients with monoallelic TP53 muta­tions and VAF
≤22% were char­ac­ter­ized by sur­vival and response to ther­apy
sim­i­lar to wild-type coun­ter­parts.16
Toward a tai­lored molec­u­lar approach in MDS | 419
Table 1.  Comparison of existing prog­nos­ti­ca­tion mod­els in MDS

Model Variables C index Reference

International Prognostic Scoring System (IPSS)2 Karyotype 0.65 2
BM blasts %
Number of cytopenias
Revised International Prognostic Scoring System Karyotype 0.67 3
(IPSS-R)3 BM blasts %
Degree of cytopenias
MD Anderson Cancer Center (MDACC)55 Karyotype 0.65 49
BM blasts %
Degree of cytopenias
Age
Performance sta­tus

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/418/1851957/418voso.pdf by guest on 13 December 2021

Prior trans­fu­sion
World Health Organization-based Prognostic Scoring Karyotype 0.65 50
System (WPSS)56 WHO cat­e­gory
Transfusion require­ment
Genoclinical model according to Haferlach et al8 Gender NA 8
Age
Degree of cytopenias (PLT, Hb)
BM blasts %
Karyotype
ASXL1, CBL, ETV6, EZH2, KRAS, LAMB4, NCOR2,
NF1, NPM1, NRAS, PRPF8, RUNX1, TET2, TP53
Genoclinical model according to Nazha et al10 Karyotype 0.71 10
BM blasts %
Degree of cytopenias
Age
WHO cat­e­gory
Disease ontog­eny (sec­ond­ary vs de novo)
TP53, EZH2, and SF3B1
Genoclinical model according to Bersanelli et al15 63 clin­i­cal and molec­u­lar var­i­ables 0.75 14
PLT, plate­lets; NA, not avail­­able.

In addi­tion, the costs of genome-scan­ning approaches (despite Key clin­i­cal point

drop­ping dra­mat­i­cally in the last years) still rep­re­sent an obsta­cle In Case 1, the results of NGS were deter­ mi­ nant in assigning
for a world­wide hybrid geno­mic-clin­i­cal appli­ca­tion used on a the patient to the cor­rect WHO-defined cat­e­gory (MDS-RS), in
day-by-day basis, espe­cially in devel­op­ing countries. defin­ing the prog­no­sis, and in guid­ing the clin­i­cal man­age­ment
toward targeted treat­ment with luspatercept.

From biol­ogy to clas­si­fi­ca­tion and tai­lored treat­ment:

CLINICAL CASE 1 (Con­t in­u ed)
Our 78-year-old woman had an LR-MDS (IPSS score, 0; IPSS-R,
2). According to the com­ mon guide­ lines, the patient was
started on alpha EPO (40 000 UI sub­cu­ta­ne­ously once weekly,
increased to 80 000 UI weekly after 12 weeks). However, after
6 months the hemo­glo­bin lev­els remained unchanged, and she
even­ tu­
ally became trans­ fu­
sion depen­ dent (1 red blood cell
unit [RBC]/week). At this point, con­sid­er­ing the diag­no­sis of
SF3B1-mutant MDS with RS and sin­gle-lin­e­age dys­pla­sia and
the erythropoiesis-stimulating agent fail­ure, the patient was
enrolled in the MEDALIST trial and started on luspatercept.17
A pro­ gres­sive increase in hemo­ glo­bin lev­els and a reduc­
tion of trans­fu­sion needs were observed after 8 weeks, with
achieve­ment of a minor hema­to­logic improve­ment—ery­throid
response (1 RBC unit/3 weeks), according to the revised Inter­
national Working Group 2018 hema­to­log­i­cal response cri­te­ria—
still persisting after 27 months on treat­ment.18
the case of SF3B1 muta­tions
As show­cased by our patient, SF3B1-mutant MDS is an emblem
of geno­type/phe­no­type asso­ci­a­tion, char­ac­ter­ized by a spe­cific
prog­nos­tic out­come and the avail­abil­ity of a tai­lored treat­ment.
SF3B1 is the most com­monly mutated splic­ing-fac­tor gene and
defi­nes a well-char­ac­ter­ized group of MDS with RS, lead­ing to its
def­i­ni­tion as “the lord of the rings” of MDS.19-21 The asso­ci­a­tion is
so strong that if the muta­tion is pres­ent, the thresh­old of RS pro­
por­tion in the BM to sat­isfy the WHO MDS-RS cri­te­ria low­ers from
15% to 5% (as in our patient).1 In a recent study, the International
Working Group argues for the rec­og­ni­tion of SF3B1-mutant MDS
as a dis­tinct noso­logic entity and pro­poses spe­cific diag­nos­tic
cri­te­ria (Table 2).22 This plea is supported by the strong geno­
type/phe­no­type asso­ci­a­tion char­ac­ter­ized by the prev­a­lence of
female sex, a higher degree of ane­mia and lower BM blast per­
cent­ ages, a favor­ able dis­ease course, the pres­ ence of SF3B1
muta­tions as a founding event in the major­ity of cases, and the
recent avail­abil­ity of a spe­cific treat­ment.17,22

420  |  Hematology 2021  |  ASH Education Program

Table 2.  Proposed diag­nos­tic cri­te­ria for the subentity of MDS with mutated SF3B1
1. Cytopenia defined by stan­dard hema­to­logic ref­er­ence val­ues
2. Presence of a somatic SF3B1 muta­tion
3. Isolated ery­throid or multilineage dys­pla­sia, with or with­out RS
4. BM blasts <5% and PB blasts <1%
5. WHO cri­te­ria not fulfilling any other cat­e­gory
6. Normal kar­yo­type or any cyto­ge­netic abnor­mal­ity other than del(5q), mono­somy 7, inv(3) or abnor­mal 3q26, com­plex (≥3)
7. Presence of any addi­tional somatic gene muta­tion other than RUNX1 and/or EZH2*
*Additional JAK2V617F, CALR, or MPL muta­tions strongly sup­port the diag­no­sis of myelodysplastic syndrome/myeloproliferative neoplasm with ring
sideroblasts and thrombocytosis (MDS/MPN-RS-T). Adapted from Malcovati et al.22
SF3B1, together with del(5q), rep­re­sents an ideal exem­plum
of how progresses in dis­ease patho­bi­­ol­ogy can not only improve
dis­ease clas­si­fi­ca­tion but also open pos­si­bil­i­ties for tai­lored treat­
ments. The haploinsufficiency of CSNK1A1 and RPS14 rep­re­sents
the molec­u­lar lesions under­ly­ing the exqui­site sen­si­tiv­ity of del(5q)
cases to lenalidomide.23,24 The pres­ence of SF3B1 muta­tions gen­er­
ates aber­rant splic­ing events due to misrecognition of 3′ splice site
with deg­ra­da­tion of about 50% of the aber­rant mRNAs via non­
sense-medi­ated mRNA decay.25 Moreover, the ABCB7 and PPOX
genes, involved in mito­ chon­ drial iron metab­ o­lism, have been
found to be sig­nif­i­cantly downregulated in SF3B1-mutant MDS and
are thought to be asso­ci­ated with the pres­ence of RS.22,26 These
molec­u­lar dis­tur­bances gen­er­ate a higher degree of inef­fec­tive
eryth­ro­poi­e­sis and, there­fore, of ane­mia. Luspatercept is a recom­
bi­nant inhib­i­tor of transforming growth fac­tor β a ­ ble to reduce
SMAD2/3 sig­nal­ing, over­com­ing the impair­ment imprinted by the
genetic lesion and enabling late-stage ery­throid dif­fer­en­ti­a­tion. In
the sin­gle-arm phase 2 trial (PACE-MDS) enroll­ing ane­mic LR-MDS
patients, luspatercept induced hema­to­logic improve­ment in 69%
of RS-pos­i­tive cases vs 43% of RS-neg­a­tive cases. When looking at
SF3B1 sta­tus, hema­to­logic improve­ment was enriched in mutated
patients (77% vs 40% in neg­a­tive cases).27 These results were later
con­firmed in patients with MDS-RS by the ran­dom­ized phase 3
MEDALIST trial, which showed that luspatercept (given at a dose
of 1-1.75 mg/kg sub­cu­ta­ne­ously every 21 days) induced trans­fu­
sion inde­pen­dence in 38% of patients at 8 weeks vs 13% in the
pla­cebo arm.17 This trial led to the approval of luspatercept by the
US Food and Drug Administration and the Euro­pean Medicines
Agency for patients with LR-MDS with RS and trans­fu­sion-depen­
dent ane­mia, who have an unsat­is­fac­tory response to or are inel­i­
gi­ble for eryth­ro­poi­e­tin-based ther­apy.
CLINICAL CASE 2
A 55-year-old woman was admit­ted to our hos­pi­tal after notic­
ing the appear­ance of pete­chiae over her arms and legs and
after a find­ing of pan­cy­to­pe­nia (Hb, 9 g/dL; abso­lute neu­tro­
phil count, 0.8 × 109/L; plate­lets, 13 109/L) at a com­plete blood
count eval­u­a­tion. Her BM aspi­ra­tion was hypocellular, and the
kar­yo­type was nor­mal. A core biopsy con­firmed the BM hypo­
cellularity and showed <5% CD34+ cells with the pres­ence of
rare micromegakaryocytes and a nor­mal reticulin-staining pat­
tern. The patient was started on EPO 40 000 UI/week, but no
changes in Hb lev­els were observed. NGS anal­y­sis using a 30-
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/418/1851957/418voso.pdf by guest on 13 December 2021
gene mye­loid panel iden­ti­fied somatic muta­tions in 4 genes
(DNMT3A, RUNX1, SRSF2, and TET2) at a median VAF of 12%
(range, 1.5%-12.6%). Based on the pres­ence of severe multilin­
ear cytopenias and uni­lin­ear dys­pla­sia and the evi­dence of <5%
BM blasts, a diag­no­sis of MDS unclas­si­fi­able (MDS-U) with sin­
gle-lin­e­age dys­pla­sia was made. The patient’s IPSS score was
0.5 (inter­me­di­ate-1), while IPSS-R was 4 (inter­me­di­ate). Taking
into con­sid­er­ation the young age at pre­sen­ta­tion, the sever­ity
of cytopenias lead­ing to trans­fu­sion depen­dency of RBCs and
plate­lets, and the muta­tional pro­file, a deci­sion to pro­ceed to
upfront HSCT was made. She had no major com­pli­ca­tions from
the trans­plan­ta­tion pro­ce­dure and is pres­ently in com­plete
remis­sion (CR) at 24 months from HSCT.
Myeloid gene muta­tion anal­y­sis as a help­ful diag­nos­tic
tool for MDS
As shown by the patient’s pre­sen­ta­tion, the num­ber and type of
muta­tional events iden­ti­fied by NGS were help­ful not only in sup­
porting the diag­no­sis of MDS but also for decid­ing to pro­ceed
to HSCT. As men­tioned before, somatic muta­tions (except SF3B1)
are not part of cur­rent clas­si­fi­ca­tion schemes of MDS, whose hall­
mark remains the pres­ence of BM dys­pla­sia, and there is no sin­gle
muta­tion that can be con­sid­ered 100% patho­gno­monic of the dis­
ease. However, the assess­ment of dys­pla­sia in hypocellular cases
with low blast counts and nor­mal kar­yo­types poses chal­lenges
even well-expe­ri­enced mor­phol­o­gists. In these cases, NGS may
add impor­tant infor­ma­tion since over 90% of patients with mye­
loid dis­eases have been shown to pres­ent somatic muta­tions.11,12
Indeed, when using a well-constructed gene panel, the absence
of any molec­u­lar lesion in a patient with iso­lated cytopenia with­
out BM dys­pla­sia and a nor­mal kar­yo­type has a high neg­a­tive pre­
dic­tive value and should prompt the con­sid­er­ation of alter­na­tive
diag­noses (eg, auto­im­mune cytopenias), espe­cially in youn­ger
patients.28 Conversely, the iden­ ti­
fi­
ca­
tion of somatic muta­
tions
may under­pin a diag­no­sis of MDS, as in our patient, or direct phy­
si­cians toward alter­na­tive diag­noses (ie, large gran­u­lar lym­pho­
cytic leu­ke­mia in the case of STAT3 muta­tions).29 When apply­ing
NGS to rou­tine diag­nos­tics, clonal hema­to­poi­e­sis of indetermined
poten­tial (CHIP) must be taken into account, since it is a very fre­
quent find­ing in older adult indi­vid­u­als (up to 60% over the age
of 80) and is char­ac­ter­ized by muta­tions usu­ally affect­ing one
iso­lated gene (fre­quently, DNMT3A, TET2, and ASXL1), at a 2% to
30% VAF.14,30-32 In the con­text of CHIP, a high num­ber of muta­tions
and increased alle­lic bur­den are indi­ca­tions for fre­quent hema­to­
Toward a tai­lored molec­u­lar approach in MDS | 421
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/418/1851957/418voso.pdf by guest on 13 December 2021
Figure 2.  Model of disease evolution in MDS. Upper panel: risk factors commonly associated with the development of myeloid disor­
ders (germline variants, smoking, aging, and cancer treatment exposure such as chemo/radiotherapy). Middle panel: a hypothetical
model of clonal evolution in which the founding event (or germline predisposition lesion) leads to subsequent acquisition/loss of new
somatic mutations in a linear/branching fashion. Lower panel: the typical acquisition of methylation during MDS progression, which
leads to the silencing of genes such as tumor suppressor and the disruption of many cell pathways (DNA repair, apoptosis, cell cycle,
cell adherence).

log­i­cal fol­low-up since these muta­tions may be the pro­dromes
for the onset of MDS.14,30,31 It is worth­while to under­line here that
the pres­ence of cytopenias and BM dys­pla­sia remains an essen­tial
require­ment for the diag­no­sis of MDS, while CHIP refers to a con­
di­tion in which only somatic muta­tional events are pres­ent.1 Other
four-letters acronyms are used to define the case of patients with
somatic mutations and cytopenias in the absence of BM dysplasia
(CCUS, clonal cytopenia of uncertain significance), BM dysplasia
without cytopenia and clonal events (IDUS, idiopathic dysplasia of
unknown significance), and cytopenia without dysplasia and
mutations (ICUS, idiopathic cytopenias of uncertain significance).14
The increas­ ing num­ber of somatic muta­ tions also mir­ rors
dis­ease pro­gres­sion, which is an intrin­sic char­ac­ter­is­tic of MDS
(Figure 2). This implies not only lin­ear and branching clonal evo­
lu­tion, defined by the acqui­si­tion or loss of somatic muta­tions,
respec­tively, but also increased DNA meth­yl­a­tion lev­els.33,34 For
these rea­sons, reassessment of the muta­tional pro­file at the time
of dis­ease pro­gres­sion is indi­cated to ensure treat­ment opti­mi­
za­tion, espe­cially in youn­ger patients.
Key clin­i­cal points
In Case 2 the deci­sion to pro­ceed to HSCT upfront was mostly
based on clin­i­cal con­sid­er­ations, includ­ing the PB tri­lin­ear cyto­
penia lead­ing to RBC and plate­let trans­fu­sion depen­dency. The
avail­abil­ity of the muta­tion pro­file, which showed the pres­ence
of 4 muta­tions also affect­ing RUNX1 and SRSF2 genes, was fur­
ther sup­port­ive of the diag­no­sis of MDS and of the high prob­a­bil­
ity of rapid dis­ease pro­gres­sion.
Considerations on allo­ge­neic HSCT in patients with MDS
in the molec­u­lar era
Although allo­ge­neic HSCT rep­re­sents the only cura­tive option
in MDS, only 10% to 15% of patients will even­tu­ally undergo this
pro­ce­dure. Indeed, the demo­graph­ics of the dis­ease (ie, most
patients are diag­nosed with MDS at ages >70) con­sti­tute a major
fac­tor lim­it­ing HSCT acces­si­bil­ity. However, the risk of relapse
remains the major out­ come deter­ mi­
nant even after HSCT. In
this con­text, Della Porta et al iden­ti­fied lesions in ASXL1, RUNX1,
and TP53 genes as inde­pen­dent pre­dic­tors of dis­mal out­come.35

422  |  Hematology 2021  |  ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/418/1851957/418voso.pdf by guest on 13 December 2021
Figure 3. Time course of the identification of genes involved in myeloid neoplasms with germline predisposition. Time line de­
picting the discovery of the most important genes involved in the development of myeloid disorders with germline predisposition.
Different colors represent the associated peculiar features (eg, platelet disorders or organ dysfunction). For gene groups (eg, Fanconi
anemia), the year of discovery of the first gene associated with the disease is indicated.

More impor­tantly, the role of these somatic events was inde­pen­
dent of the IPSS-R risk score, empha­siz­ing once more the need to
inte­grate clin­i­cal and molec­u­lar fac­tors. Indeed, the com­bi­na­tion
of IPSS-R plus muta­tional char­ac­ter­is­tics changed the pre­dic­tion
of posttransplant out­comes for 34% of cases. Further, in another
study new somatic events were found to be acquired at relapse,
with the pres­ence of dis­ease-related muta­tions at 30 days from
HSCT asso­ci­ated with a high risk of pro­gres­sion at a median of
67 days from muta­tion detec­tion.36
In this line, Heuser et al iden­ti­fied muta­tions of NRAS, U2AF1,
IDH2, and TP53 and/or the pres­ence of a com­plex kar­yo­type as
adverse mark­ers of sur­vival in MDS patients under­go­ing HSCT.37
Although dif­fer­ent stud­ies found diverse pat­terns of gene muta­
tions influ­enc­ing out­comes, TP53 muta­tions remain among the
major deter­mi­nants of poor prog­no­sis post HSCT, being con­
sis­tently asso­ci­ated across stud­ies with reduced sur­vival and a
high risk of relapse. In a large cohort of 1514 MDS patients under­
Prakash Singh Shekhawat
go­ing allo­ge­neic HSCT, patients older than 40 years har­bor­ing
muta­tions in TP53 and RAS path­way genes had sig­nif­i­cantly
shorter sur­vival and a higher risk of relapse than their wild-type
coun­ter­parts.38 While RAS path­way muta­tions were unfa­vor­able
only for patients receiv­ing reduced-inten­sity con­di­tion­ing, TP53
muta­tions were asso­ci­ated with a dis­mal prog­no­sis irrespec­
tive of the inten­sity of the con­di­tion­ing reg­i­men. Likewise, TP53
muta­tions had an impact on the out­come of HSCT when asso­
ci­ated with com­plex kar­yo­type, while RAS path­way muta­tions
were prognostically unfa­vor­able in the con­text of MDS/mye­lo­
pro­lif­er­a­tive neo­plasms in another study.39
Germline muta­tions
The sit­u­a­tion becomes even haz­ier when con­sid­er­ing genes
mutated in mye­loid neo­plasms with germline pre­dis­po­si­tion,
which have been increas­ingly iden­ti­fied in the last few years
(Figure 3). The pre­cise rec­og­ni­tion of such var­i­ants is impor­tant
Toward a tai­lored molec­u­lar approach in MDS | 423

Figure 4.  Exemplificative list of genes recurrently mutated in MDS, with impact on clinical features and on treatment options.
Shown are exemplary gene mutations, their prognostic impact, the associated recurrent clinical features, and possible therapeutic
interventions.
for a cor­rect patient man­age­ment (espe­cially for genes with
other organ involve­ment, such as Fanconi ane­mia cases and
oth­ers), for genetic coun­sel­ing, and ulti­mately for donor selec­
tion in patients eli­gi­ble for HSCT (unre­lated vs related donors).40
In these cases the var­ia ­ nts must be con­firmed on a nonhemato­
poietic tis­sue (eg, fibro­blasts, hair fol­li­cles, or nails), and genet­
ic coun­sel­ing must be required. Generally, these patients pres­
ent with youn­ger age, a pos­it­ive fam­ily his­tory, and in some
instances with extrahematologic fea­tures (such as in Emberger
syn­drome, or MonoMAC in case of GATA2 defects).41 However,
a low pen­e­trance may result in an absent fam­ily his­tory, and
late age at MDS diag­no­sis (eg, in DDX41-pos­i­tive cases) may
chal­lenge the rec­og­ni­tion.42 As dem­on­strated by a sem­i­nal
study, these issues play a sig­nif­i­cant role in the HSCT con­text,
where donor selection becomes of utmost impor­tance. Indeed,
the top mutated genes in youn­ger patients (<40 years) includ­
ed GATA2, PIGA, and SBDS, which are recur­rently mutated in
mye­loid neo­plasms with germline pre­dis­po­si­tion, while muta­
tions in TET2, DNMT3A, SRSF2, and SF3B1 were sig­nif­i­cantly less
com­mon.38
424  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/418/1851957/418voso.pdf by guest on 13 December 2021
Implications of molec­u­lar data for treat­ment strat­egy
Several emerg­ing targeted ther­a­pies are cur­rently under eval­
u­ a­
tion for patients with MDS and recur­ rent gene muta­ tions
(Figure 4). For instance, vita­min C is ­able to restore the met­a­
bolic impair­ment imprinted by TET2 muta­tions (NCT03682029,
NCT03999723) while prom­is­ing results have been obtained in
vitro with splic­ing inhib­i­tors (eg, H3B-8800),43 but unfor­tu­nately,
they have not been con­firmed by the phase 1, first-in-human clin­
i­cal study (NCT02841540).
APR-246 is a small mol­e­cule ­able to reestablish p53 func­tions
by restor­ing its con­for­ma­tion and has shown encour­ag­ing clin­
i­cal results in vivo in com­bi­na­tion with HMA.44 A recent phase
1b/2 study of its com­bi­na­tion with azacitidine (AZA) in patients
with TP53-mutant MDS or AML with 20% to 30% BM blasts
showed a 73% over­all response rate (ORR), with 50% of patients
achiev­ing CR and 58%, a cyto­ge­netic response.45 The effi­cacy of
this com­bi­na­tion was also underlined by the reduc­tion of muta­
tional bur­den and of p53 expres­sion by immu­no­his­to­chem­is­try.45
However, the phase 3 of this trial (NCT03745716) missed the pri­
mary end point, as the dif­fer­ence in CR between the 2 arms did
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/418/1851957/418voso.pdf by guest on 13 December 2021
Figure 5. Clinicobiological characterization of MDS and tailored treatment. MDS patients represent a heterogeneous multitude
characterized by different clinical, karyotypic, morphologic, and molecular features. In particular, the lower panel demonstrates how
the incorporation of molecular information into currently available prognostication schemes will enable in the near future better
prognostication and tailored treatments.

not meet the predefined thresh­old for sta­tis­ti­cal sig­nif­i­cance
(33.3% vs 22.4% in the exper­i­men­tal vs AZA-monotherapy group,
respec­tively; P = .13).46 These data high­light the time­less need to
con­firm phase 1 and 2 results with larger phase 3 stud­ies even in
patients with molec­u­larly defined dis­eases, which dis­play wide
clin­i­cal het­ero­ge­ne­ity.
Beyond spe­cific TP53 acti­va­tors, magrolimab, a first-in-class
anti-CD47 anti­body, has shown par­tic­u­lar activ­ity in TP53-mu­
tated MDS/AML, as revealed by the results of a phase 1b trial in
com­bi­na­tion with AZA in this set­ting.47 These 2 drugs act syn­er­
gis­ti­cally, induc­ing “eat me” sig­nals on leu­ke­mic stem cells and
restor­ing mac­ro­phage-medi­ated phago­cy­to­sis.
The recent results of the IDH1/2 inhib­ i­
tors ivosidenib and
enasidenib in AML have also opened new sce­nar­ios for patients
with MDS.48-50 In a phase 2 study, enasidenib in com­bi­na­tion with
AZA showed an ORR of 67% in newly diag­nosed IDH2-mutant
high-risk-MDS patients, and a 50% ORR was obtained when enas­
idenib is used as a sin­gle agent in HMA-treated patients. The
pre­lim­i­nary anal­y­sis of another study (NCT02074839) cur­rently
eval­u­at­ing ivosidenib in IDH1-mutant cases dem­on­strated a 42%
ORR in relapsed/refrac­tory MDS.51,52 Furthermore, DiNardo and
col­leagues recently showed that ivosidenib in com­bi­na­tion with
AZA induced responses in 78% of newly diag­nosed patients with
IDH1-mutant AML inel­i­gi­ble for inten­sive ther­apy and IDH1 muta­
tion clear­ance in 71% of those achiev­ing CR.50 Another study
com­bin­ing ivosidenib plus venetoclax with or with­out AZA in the
same patient group (NCT03471260) is cur­rently ongo­ing, and the
pre­lim­i­nary results have dem­on­strated the tol­er­a­bil­ity and effec­
tive­ness of this com­bi­na­tion (ORR, 89%).53
Conclusions
MDS is a puz­zling dis­or­der with a bio­log­i­cal intri­cacy reflected
by the dif­fi­cul­ties and lim­i­ta­tions of existing clas­si­fi­ca­tions and
prog­nos­ti­ca­tion sys­tems. The abun­dance of molec­u­lar infor­ma­
tion opens new sce­nar­ios in the clin­i­cal set­ting, indeed inau­gu­
rat­ing a new era in MDS and delin­eat­ing a more pre­cise, objec­
tive, and per­son­al­ized path (Figure 5).
The bioinformatic inter­pre­ta­tion of the clin­i­cal role of somatic
muta­ tions is still prob­lem­ atic. Nevertheless, machine-learn­ ing
approaches may in the near future rep­re­sent a valu­able tool to
solve the short­com­ings of cur­rent diag­nos­tic schemes, pos­si­
bly unveiling fur­ther prog­nos­tic impli­ca­tions. By com­bin­ing

Prakash Singh Shekhawat

Toward a tai­lored molec­u­lar approach in MDS | 425
mor­pho­logic and molec­u­lar data, these approaches will enable
the iden­ti­fi­ca­tion of non­ran­dom geno­typic/mor­pho­logic rela­tion­
ships, bet­ter defin­ing clin­i­cally rel­e­vant phe­no­types.54 However,
a lim­i­ta­tion of these stud­ies is dem­on­strated by the sta­tis­ti­cal
power of the sam­ple size, which does not account for the clin­i­cal
and prog­nos­tic weight of less fre­quently mutated genes. Larger
cohorts of patients will help unravel the com­plex­ity of MDS biol­
ogy with the intent of gen­er­at­ing molec­u­larly ori­ented clas­si­fi­
ca­tion and prog­nos­ti­ca­tion sys­tems with rel­e­vant clin­i­cal util­ity.
Acknowledgments
We thank the Amer­ i­
can Ital­
ian Cancer Foundation Fellowship
Program (to C. G.), AIRC 5 × 1000 call “Metastatic dis­ease: the key
unmet need in oncol­ogy to MYNERVA” pro­ject, #21267 (MYeloid
NEoplasms Research Venture AIRC), and MIUR grant N. 2017WX­
R7ZT (to M. T. V.).
We would also like to acknowl­edge Stefan Hohaus for help­ful
com­ments on the man­u­script.
Conflict-of-inter­est dis­clo­sure
Maria Teresa Voso: no com­pet­ing finan­cial inter­ests to declare.
Carmelo Gurnari: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Maria Teresa Voso: nothing to disclose.
Carmelo Gurnari: nothing to disclose.
Correspondence
Maria Teresa Voso, Department of Biomedicine and Prevention,
University of Tor Vergata, Via Montpellier 1, 00133 Rome, Italy; e-
mail: voso@med​­.uniroma2​­.it.
References
1. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revi­sion to the World Health
Organization clas­si­fi­ca­tion of mye­loid neo­plasms and acute leu­ke­mia.
Blood. 2016;127(20):2391-2405.
2. Greenberg P, Cox C, LeBeau MM, et al. International scor­ing sys­tem for eval­
u­at­ing prog­no­sis in myelodysplastic syn­dromes. Blood. 1997;89(6):2079-
2088.
3. Greenberg PL, Tuechler H, Schanz J, et al. Revised inter­na­tional prog­nos­tic
scor­ing sys­tem for myelodysplastic syn­dromes. Blood. 2012;120(12):2454-
2465.
4. Pfeilstöcker M, Tuechler H, Sanz G, et al. Time-depen­dent changes in mor­
tal­ity and trans­for­ma­tion risk in MDS. Blood. 2016;128(7):902-910.
5. Platzbecker U. Treatment of MDS. Blood. 2019;133(10):1096-1107.
6. Voso MT, Fenu S, Latagliata R, et  al. Revised International Prognostic
Scoring System (IPSS) pre­dicts sur­vival and leu­ke­mic evo­lu­tion of myelo­
dysplastic syn­dromes sig­nif­i­cantly bet­ter than IPSS and WHO Prognostic
Scoring System: val­i­da­tion by the Gruppo Romano Mielodisplasie Ital­ian
Regional Database. J Clin Oncol. 2013;31(21):2671-2677.
7. Kuendgen A, Nomdedeu M, Tuechler H, et  al. Therapy-related myelo­
dysplastic syn­dromes deserve spe­cific diag­nos­tic sub-clas­si­fi­ca­tion and
risk-strat­i­fi­ca­tion—an approach to clas­si­fi­ca­tion of patients with t-MDS.
Leukemia. 2021;35(3):835-849.
8. Haferlach T, Nagata Y, Grossmann V, et al. Landscape of genetic lesions in
944 patients with myelodysplastic syn­dromes. Leukemia. 2014;28(2):241-
247.
9. Nazha A, Al-Issa K, Hamilton BK, et al. Adding molec­u­lar data to prog­nos­tic
mod­els can improve pre­dic­tive power in treated patients withmyelodys­
plastic syn­dromes. Leukemia. 2017;31(12):2848-2850.
10. Nazha A, Narkhede M, Radivoyevitch T, et  al. Incorporation of molec­ul­ar
data into the Revised International Prognostic Scoring System in treated
patients with myelodysplastic syn­ dromes. Leukemia. 2016;30(11):2214-
2220.
11. Bejar R, Stevenson K, Abdel-Wahab O, et al. Clinical effect of point muta­
tions in myelodysplastic syn­ dromes. N Engl J Med. 2011;364(26):2496-
2506.
426  |  Hematology 2021  |  ASH Education Program
12. Papaemmanuil E, Gerstung M, Malcovati L, et al; Chronic Myeloid Disorders
Working Group of the International Cancer Genome Consortium. Clinical
and bio­log­i­cal impli­ca­tions of driver muta­tions in myelodysplastic syn­
dromes. Blood. 2013;122(22):3616-3627, quiz 3699.
13. Falconi G, Fabiani E, Piciocchi A, et  al. Somatic muta­tions as mark­ers of
out­come after azacitidine and allo­ge­neic stem cell trans­plan­ta­tion in high­
er-risk myelodysplastic syn­dromes. Leukemia. 2019;33(3):785-790.
14. Fenaux P, Haase D, Santini V, et al; Euro­pean Society for Medical Oncol­
ogy Guidelines Committee. Myelodysplastic syn­ dromes: ESMO Clinical
Practice Guidelines for diag­no­sis, treat­ment and fol­low-up†☆. Ann Oncol.
2021;32(2):142-156.
15. Bersanelli M, Travaglino E, Meggendorfer M, et al. Classification and per­
son­al­ized prog­nos­tic assess­ment on the basis of clin­i­cal and geno­mic fea­
tures in myelodysplastic syn­dromes. J Clin Oncol. 2021;39(11):1223-1233.
16. Bernard E, Nannya Y, Hasserjian RP, et al. Implications of TP53 alle­lic state
for genome sta­bil­ity, clin­i­cal pre­sen­ta­tion and out­comes in myelodysplas­
tic syn­dromes. Nat Med. 2020;26(10):1549-1556.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/418/1851957/418voso.pdf by guest on 13 December 2021
17. Fenaux P, Platzbecker U, Mufti GJ, et al. Luspatercept in patients with low­
er-risk myelodysplastic syn­dromes. N Engl J Med. 2020;382(2):140-151.
18. Platzbecker U, Fenaux P, Adès L, et  al. Proposals for revised IWG 2018
hema­to­log­i­cal response cri­te­ria in patients with MDS included in clin­i­cal
tri­als. Blood. 2019;133(10):1020-1030.
19. Papaemmanuil E, Cazzola M, Boultwood J, et al; Chronic Myeloid Disorders
Working Group of the International Cancer Genome Consortium. Somatic
SF3B1 muta­tion in myelodysplasia with ring sideroblasts. N Engl J Med.
2011;365(15):1384-1395.
20. Yoshida K, Sanada M, Shiraishi Y, et al. Frequent path­way muta­tions of splic­ing
machin­ery in myelodysplasia. Nature. 2011;478(7367):64-69.
21. Palomo L, Solé F. SF3B1: the lord of the rings in MDS. Blood. 2020;136(2):149-
151.
22. Malcovati L, Stevenson K, Papaemmanuil E, et al. SF3B1-mutant MDS as a
dis­tinct dis­ease sub­type: a pro­posal from the International Working Group
for the Prognosis of MDS. Blood. 2020;136(2):157-170.
23. Ebert BL, Pretz J, Bosco J, et al. Identification of RPS14 as a 5q− syn­drome
gene by RNA inter­fer­ence screen. Nature. 2008;451(7176):335-339.
24. Schneider RK, Ademà V, Heckl D, et  al. Role of casein kinase 1A1 in the
biol­ogy and targeted ther­apy of del(5q) MDS. Cancer Cell. 2014;26(4):509-
520.
25. Shiozawa Y, Malcovati L, Gallì A, et al. Aberrant splic­ing and defec­tive
mRNA pro­duc­tion induced by somatic spliceosome muta­tions in myel­
odysplasia. Nat Commun. 2018;9(1):3649.
26. Nikpour M, Scharenberg C, Liu A, et al. The trans­porter ABCB7 is a medi­a­
tor of the phe­no­type of acquired refrac­tory ane­mia with ring sideroblasts.
Leukemia. 2013;27(4):889-896.
27. Platzbecker U, Germing U, Götze KS, et al. Luspatercept for the treat­ment
of anae­mia in patients with lower-risk myelodysplastic syn­dromes (PACE-
MDS): a multicentre, open-label phase 2 dose-find­ing study with long-term
exten­sion study. Lancet Oncol. 2017;18(10):1338-1347.
28. Steensma DP. How I use molec­u­lar genetic tests to eval­u­ate patients who
have or may have myelodysplastic syn­dromes. Blood. 2018;132(16):1657-
1663.
29. Koskela HL, Eldfors S, Ellonen P, et  al. Somatic STAT3 muta­tions in large
gran­u­lar lym­pho­cytic leu­ke­mia. N Engl J Med. 2012;366(20):1905-1913.
30. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hema­to­poi­e­sis
asso­ci­ated with adverse out­comes. N Engl J Med. 2014;371(26):2488-2498.
31. Genovese G, Kähler AK, Handsaker RE, et  al. Clonal hema­to­poi­e­sis and
blood-can­ cer risk inferred from blood DNA sequence. N Engl J Med.
2014;371(26):2477-2487.
32. van Zeventer IA, Salzbrunn JB, de Graaf AO, et al. Prevalence, pre­dic­tors,
and out­ comes of clonal hema­ to­poi­e­sis in indi­
vid­
ua­ls aged ≥80 years.
Blood Adv. 2021;5(8):2115-2122.
33. da Silva-Coelho P, Kroeze LI, Yoshida K, et al. Clonal evo­lu­tion in myelodys­
plastic syn­dromes. Nat Commun. 2017;8(21 April):15099.
34. Figueroa ME, Skrabanek L, Li Y, et  al. MDS and sec­ond­ary AML dis­play
unique pat­ terns and abun­ dance of aber­ rant DNA meth­ yl­
a­
tion. Blood.
2009;114(16):3448-3458.
35. Della Porta MG, Gallì A, Bacigalupo A, et al. Clinical effects of driver somatic
muta­tions on the out­comes of patients with myelodysplastic syn­dromes
treated with allo­ge­neic hema­to­poi­etic stem-cell trans­plan­ta­tion. J Clin
Oncol. 2016;34(30):3627-3637.
36. Duncavage EJ, Jacoby MA, Chang GS, et al. Mutation clear­ance after trans­
plan­ta­tion for myelodysplastic syn­drome. N Engl J Med. 2018;379(11):1028-
1041.
37. Heuser M, Gabdoulline R, Löffeld P, et al. Individual out­come pre­dic­tion for
myelodysplastic syn­drome (MDS) and sec­ond­ary acute mye­loid leu­ke­mia
from MDS after allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion. Ann Hematol.
2017;96(8):1361-1372.
38. Lindsley RC, Saber W, Mar BG, et al. Prognostic muta­tions in myelodysplas­
tic syn­drome after stem-cell trans­plan­ta­tion. N Engl J Med. 2017;376(6):536-
547.
39. Yoshizato T, Nannya Y, Atsuta Y, et al. Genetic abnor­mal­i­ties in myelodys­
plasia and sec­ond­ary acute mye­loid leu­ke­mia: impact on out­come of stem
cell trans­plan­ta­tion. Blood. 2017;129(17):2347-2358.
40. University of Chicago Hematopoietic Malignancies Cancer Risk Team. How
I diag­nose and man­age indi­vid­u­als at risk for inherited mye­loid malig­nan­
cies. Blood. 2016;128(14):1800-1813.
41. Wlodarski MW, Hirabayashi S, Pastor V, et al; Euro­pean Working Groups of
Myelodysplastic Syndromes. Prevalence, clin­i­cal char­ac­ter­is­tics, and prog­
no­sis of GATA2-related myelodysplastic syn­dromes in chil­dren and ado­les­
cents. Blood. 2016;127(11):1387-1397, quiz 1518.
42. Polprasert C, Schulze I, Sekeres MA, et al. Inherited and somatic defects in
DDX41 in mye­loid neo­plasms. Cancer Cell. 2015;27(5):658-670.
43. Cimmino L, Dolgalev I, Wang Y, et al. Restoration of TET2 func­tion blocks
aber­rant self-renewal and leu­ke­mia pro­gres­sion. Cell. 2017;170(6):1079-
1095.e201095e20.
44. Maslah N, Salomao N, Drevon L, et  al. Synergistic effects of PRIMA-1Met
(APR-246) and 5-azacitidine in TP53-mutated myelodysplastic syn­dromes
and acute mye­loid leu­ke­mia. Haematologica. 2020;105(6):1539-1551.
45. Sallman DA, DeZern AE, Garcia-Manero G, et al. Eprenetapopt (APR-246)
and azacitidine in TP53-mutant myelodysplastic syn­dromes. J Clin Oncol.
2021;39(14):1584-1594.
46. Aprea Therapeutics. Aprea Therapeutics announces results of pri­ mary
end­point from phase 3 trial of eprenetapopt in TP53 mutant myelodys­
plastic syn­dromes (MDS). Accessed 6 Sep­tem­ber 2021. https:​­/​­/ir​­.aprea​
­.com​­/news​­-releases​­/news​­-release​­-details​­/aprea​­-therapeutics​­-announces​
­-results​­-primary​­-endpoint​­-phase​­-3.
47. Sallman DA, Malki MA, Asch AS, et al. Tolerability and effi­cacy of the first-in-
class anti-CD47 anti­body magrolimab com­bined with azacitidine in MDS and
AML patients: phase Ib results. J Clin Oncol. 2020;38(suppl 15):7507-7507.
48. DiNardo CD, Stein EM, de Botton S, et  al. Durable remis­sions with ivos­
idenib in IDH1-mutated relapsed or refrac­tory AML. N Engl J Med. 2018;​
378(25):​2386-2398.
Prakash Singh Shekhawat
49. Stein EM, DiNardo CD, Pollyea DA, et al. Enasidenib in mutant IDH2 relapsed
or refrac­tory acute mye­loid leu­ke­mia. Blood. 2017;130(6):722-731.
50. DiNardo CD, Stein AS, Stein EM, et al. Mutant isocitrate dehy­dro­ge­nase 1
inhib­i­tor ivosidenib in com­bi­na­tion with azacitidine for newly diag­nosed
acute mye­loid leu­ke­mia. J Clin Oncol. 2021;39(1):57-65.
51. Richard-Carpentier G, DeZern AE, Takahashi K, et  al. Preliminary results
from the phase II study of the IDH2-inhib­ i­
tor enasidenib in patients
with high-risk IDH2-mutated myelodysplastic syn­ dromes (MDS). Blood.
2019;134(suppl 1):678-678.
52. Foran JM, DiNardo CD, Watts JM, et al. Ivosidenib (AG-120) in patients with
IDH1-mutant relapsed/refrac­ tory myelodysplastic syn­ drome: updated
enroll­ ment of a phase 1 dose esca­ la­
tion and expan­ sion study. Blood.
2019;134(suppl 1):4254.
53. Lachowiez CA, Borthakur G, Loghavi S, et  al. Phase Ib/II study of the
IDH1-mutant inhib­i­tor ivosidenib with the BCL2 inhib­i­tor venetoclax +/-
azacitidine in IDH1-mutated hema­to­logic malig­nan­cies. J Clin Oncol.
2020;38(suppl 15):7500.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/418/1851957/418voso.pdf by guest on 13 December 2021
54. Nagata Y, Zhao R, Awada H, et  al. Machine learn­ing dem­on­strates that
somatic muta­tions imprint invari­ant mor­pho­logic fea­tures in myelodys­
plastic syn­dromes. Blood. 2020;136(20):2249-2262.
55. Kantarjian H, O’Brien S, Ravandi F, et al. Proposal for a new risk model in
myelodysplastic syn­drome that accounts for events not con­sid­ered in the
orig­i­nal International Prognostic Scoring System. Cancer. 2008;113(6):1351-
1361.
56. Malcovati L, Germing U, Kuendgen A, et  al. Time-depen­dent prog­nos­tic
scor­ing sys­tem for predicting sur­vival and leu­ke­mic evo­lu­tion in myelodys­
plastic syn­dromes. J Clin Oncol. 2007;25(23):3503-3510.
57. Mufti GJ, Bennett JM, Goasguen J, et al; International Working Group on
Morphology of Myelodysplastic Syndrome. Diagnosis and clas­si­fi­ca­tion of
myelodysplastic syn­drome: International Working Group on Morphology
of myelodysplastic syn­drome (IWGM-MDS) con­sen­sus pro­pos­als for the
def­i­ni­tion and enu­mer­a­tion of mye­lo­blasts and ring sideroblasts. Haema-
tologica. 2008;93(11):1712-1717.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000276
Toward a tai­lored molec­u­lar approach in MDS | 427
 MDS: BEYOND A ONE - SIZE - FITS -ALL APPROACH

Lower risk but high risk

Amy E. DeZern

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/428/1851953/428dezern.pdf by guest on 13 December 2021

Division of Hematologic Malignancies, Sidney Kimmel Cancer Center at Johns Hopkins, Baltimore, MD

Risk stratification is crucial to the appropriate management of most cancers, but in patients with myelodysplastic syn-
dromes (MDS), for whom expected survival can vary from a few months to more than a decade, accurate disease prog-
nostication is especially important. Currently, patients with MDS are often grouped into higher-risk (HR) vs lower-risk (LR)
disease using clinical prognostic scoring systems, but these systems have limitations. Factors such as molecular genetic
information or disease characteristics not captured in the International Prognostic Scoring System–Revised (IPSS-R) can
alter risk stratification and identify a subset of patients with LR-MDS who actually behave more like those with HR-MDS.
This review describes the current identification and management of patients with LR-MDS whose condition is likely to
behave in a less favorable manner than predicted by the IPSS-R.

LEARNING OBJECTIVES
• Describe limitations of currently available prognostic scoring systems for patients with MDS
• Discuss the current status of biologic upstaging, specifcally with respect to the lower­risk patient who may require
treatments usually administered to higher­risk patients
• Highlight the heterogeneity in WHO­defned treatment­related MDS
• Review which lower­risk patients by the IPSS­R may beneft from early consideration of allogeneic transplant

Introduction of 0.82×109/L. Bone marrow biopsy specimen exhibits hyper­

In myelodysplastic syndromes (MDS), an optimized under­
standing of a patient’s risk (even outside of the prognosti­
cation systems) is critical as therapeutic strategies can vary
from observation of mild cytopenias or supportive treat­
ments with transfusions through active chemotherapy and
even early allogeneic hematopoietic stem transplantation
(BMT). Here I consider lower­risk (LR) disease to encompass
very low­, low­, and intermediate­risk categories assigned
by the International Prognostic Scoring System–Revised
(IPSS­R).1 The intermediate­risk group in particular is hetero­
geneous with a substantial proportion behaving like higher
risk (HR) with early disease progression and risk of death.
Treatment­related MDS (tMDS) is another prognostic chal­
lenge in the LR arena.3
CLINICAL CASE 1
A 68­year­old woman was treated 15 years ago for breast
adenocarcinoma including adjuvant chemotherapy. Labora­
tory values now reveal a platelet (PLT) count of 102×109/L,
hemoglobin (Hgb) of 8.2g/dL, and white blood cell (WBC)
count of 2.65×109/L with an absolute neutrophil count (ANC)
cellularity, trilineage dysplasia and 2% blasts, karyotype with
8 of 20 metaphases with chromosome 21 loss, and next­
generation sequencing (NGS) revealing a RUNX1 mutation
with variant allele frequency (VAF) of 26% and a monoallelic
TP53 mutation with a VAF of 11%. IPSS­R score is low risk at
3 with 2 points for intermediate cytogenetics and 1 point for
Hgb. She received a transfusion for anemia at diagnosis.
CLINICAL CASE 2
A 69­year­old woman also was treated for breast can­
cer with radiation therapy at 50 Gy. She had new ane­
mia (Hgb, 9.9 g/dL; PLTs, 359 × 109/L; and ANC, 1.6 × 109/L).
A bone marrow biopsy specimen was 30% cellular with
dysgranulopoiesis and erythroid dysplasia. Blasts were
1%. Karyotype had 17 of 20 cell lines with trisomy 8 and a
DNMT3A mutation with a VAF of 41%. IPSS­R score was low
risk at 2, with 2 points for intermediate cytogenetics only.
Current prognostic strategies are imperfect
The use of prognostic scoring systems is common in many
malignant diseases and important for predicting the sur­

428 | Hematology 2021 | ASH Education Program

Table 1. Currently avail­­able prog­nos­tic scor­ing sys­tems in MDS
Blood Marrow
System
Hgb PLTs ANC Blasts Cyto Age
IPSS + + + + + +
WPSS + +
MDS LR + + + + + +
MDAPSS + + + + + +
FPSS + + + Higher-risk dis­ease and includes azacytidine-
IPSS-R + + + + + +
The prog­nos­tic scor­ing sys­tems are listed in order of pub­li­ca­tion with anno­ta­tions on the var­i­ables included and com­ments on their use in lower
risk dis­ease.
Cyto, cyto­ge­net­ics; FPSS, French Prognostic Scoring System; MDAPSS, MD Anderson Prognostic Scoring System; MDS LR, scor­ing sys­tem for
patients with lower-risk MDS; PS, per­for­mance sta­tus; Txn, trans­fu­sions; WPSS, WHO Classification-Based Prognostic Scoring System.
vival of indi­vid­u­als, but these sys­tems are often based on ret­ro­
spec­tive patient data with het­er­og­e­nous treat­ments and often
incor­po­rate only a few salient clin­i­cal or bio­log­i­cal fea­tures that
were sta­tis­ti­cally sig­nif­i­cant in a given model. Although the IPSS
and its IPSS-R1 (among oth­ers) are use­ful tools for clin­i­cal deci­
sion mak­ing (Table 1), these scor­ing sys­tems have draw­backs
and may fail to cap­ture impor­tant prog­nos­tic infor­ma­tion at the
indi­vid­ual level.
The key lim­i­ta­tion to the prog­nos­ti­ca­tion sys­tems is their
inabil­ity to cap­ture all­ rel­e­vant biol­ogy. For exam­ple, cyto­ge­net­ics
is the only bio­log­i­cal param­e­ter included in the IPSS-R, and unfor­
tu­nately, although an IPSS revi­sion incor­po­rat­ing NGS molec­u­lar
data is in devel­op­ment, this has yet to be reported. At the 2015
meeting of the American Society of Hematology, Bejar et al2 from
the International Working Group for Prognosis in MDS–Molecular
Committee reviewed ana­ly­ses that showed somatic muta­tions
in MDS are asso­ci­ated with clin­i­cal fea­tures and pre­dict prog­no­
sis inde­pen­dent of the IPSS-R. Bersanelli et al4 recently showed
that inte­gra­tion of clin­i­cal data with NGS pro­fil­ing improves the
accu­racy of cur­rently avail­­able prog­nos­tic scores. The authors
pro­vided evi­dence from another large, inter­na­tional data­base
that MDS could be clas­si­fied into 8 dis­tinct sub­types according
to spe­cific geno­mic fea­tures.4 These sub­groups do not cor­re­
late with mor­pho­logic categories defined by the cur­rent World
Health Organization (WHO) clas­si­fi­ca­tion and displayed sig­nif­i­
cantly dif­fer­ent clin­i­cal phe­no­types and out­come.4
The IPSS-R was based only on stan­dard G-banded meta­phase
karyotyping. A recent study exam­ined whole-genome sequenc­
ing (WGS) of patients with MDS or acute mye­loid leu­ke­mia (AML)
for fea­si­bil­ity assess­ment.5 Among 42 patients with MDS in this
study, the con­ven­tional results (trans­lo­ca­tions and copy num­ber
var­i­a­tion) were con­firmed in all­cases by WGS; nearly 25% had
addi­tional chro­mo­somal abnor­mal­i­ties iden­ti­fied by WGS that, if
Prakash Singh Shekhawat
Patient
Comments
Txn PS
Has been revised now (2012) using the same
prog­nos­tic param­e­ters (num­ber and depth or
cytopenias, mar­row blast per­cent­age [more gran­
u­lar] and kar­yo­type), was a
­ ble to reclassify nearly
25% of lower-risk MDS patients into a higher-risk
cat­e­gory
+ Time depen­dent which is use­ful over the course of
patient’s dis­ease
May be applied spe­cif­i­cally to lower-risk dis­ease
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/428/1851953/428dezern.pdf by guest on 13 December 2021
with more var­i­ables
+ Contains more clin­i­cal var­i­ables, espe­cially age and
trans­fu­sion bur­den, which could upstage a patient
treated patients, likely is not use­ful in lower-risk
patients, even for upstaging
Used most com­monly at diag­no­sis and for clin­i­cal
trial eli­gi­bil­ity, lim­i­ta­tions discussed in text
incor­po­rated into IPSS-R, would have altered the risk cat­e­gory.5
Although I do know that a chro­mo­somal alter­ation iden­ti­fied
only by WGS has the same prog­nos­tic mean­ing as that iden­ti­fied
by meta­phase karyotyping, I believe this sug­gests fur­ther need
for effi­cient incor­po­ra­tion of addi­tional cyto­ge­netic and molec­u­
lar data into prog­nos­tic scor­ing sys­tems.
CLINICAL CASES 2 (continued)
Patient 1 went on to receive 14 cycles of azacytidine. Transplant
options were pur­ sued but no avail­­ able donor located. She
remained red blood cell trans­fu­sion depen­dent (RBC-TD) with
Hgb rou­tinely below 8 g. Repeat mar­row assess­ment con­tin­ued
to show LR fea­tures; ther­apy was not altered. She ulti­mately
progressed to AML 19 months after pre­sen­ta­tion. Patient 2 has
been followed with every 3-monthly blood counts and bien­nial
mar­row assess­ments for 6 years with near-com­plete sta­bil­ity.
Patients 1 and 2 high­light fur­ther sig­nif­i­cant clin­i­cal lim­i­ta­
tion in the IPSS-R: how to cat­e­go­rize tMDS. About 15% of MDS
are tMDS and gen­er­ally con­sid­ered at HR. These dis­or­ders are
of clin­i­
cal impor­tance for patients and of grow­ ing aca­ demic
inter­est, espe­cially as patients live lon­ger due to more effec­
tive ther­a­pies. With closer fol­low-up, this cat­e­gory is likely to
be the larg­est increase in patients with “higher-risk” low-risk
MDS. The IPSS-R cohort did not con­tain any patients with tMDS,
and the prog­no­sis of these patients is com­pli­cated by sev­eral
non-MDS fac­tors, includ­ing prior toxicities of ther­apy (trans­fu­
sions, alloantibodies, organ lim­i­ta­tions), as well as other med­i­
cal comorbidities. There is the com­mon per­cep­tion that patients
with tMDS have high-risk dis­ease, regard­less of IPSS-R score, and
there­fore war­rant aggres­sive ther­apy. Patient 2’s course would
Lower risk but high risk | 429
Table 2. Features in lower-risk MDS that sug­gest higher-risk behav­ior

MDS char­ac­ter­is­tic Feature asso­ci­ated with lesser prog­no­sis

Etiology of MDS Treatment related, can behave in a het­er­og­e­nous fash­ion
Fibrosis in core biopsy Grade 2 or higher
Cytopenia Symptomatic neutropenia
Decrease in PLTs >25%
Ongoing RBC trans­fu­sion depen­dence
Anemia or throm­bo­cy­to­pe­nia refrac­tory to trans­fu­sions
Karyotype Clonal emer­gence of unfa­vor­able kar­yo­type
Somatic muta­tions Multiple muta­tions (≥3 somatic muta­tions)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/428/1851953/428dezern.pdf by guest on 13 December 2021

T53, RUNX1, ASXL1 muta­tions
Absence of SF3B1 muta­tion (espe­cially in MDS with ring sideroblasts)
Multiple somatic muta­tions
Inherited pre­dis­po­si­tion Patients with known germline var­i­ant in their dis­ease may be less likely to
respond to tra­di­tional ther­a­pies and require stem cell ther­apy sooner
Treatment response Primary treat­ment fail­ure vs sec­ond­ary fail­ure

argue against this. According to the cur­rent def­i­ni­tion of the
WHO, tMDS is defined by the his­tory of receipt of che­mo­ther­apy
and/or radio­ther­apy for nonmyeloid malig­nancy or med­i­cal con­
di­tion.6 This can­not always allow for dis­ease that may not truly
be attrib­uted to the pre­vi­ous ther­apy (per­haps in patient 2) as
the dis­ease behav­ior is more akin to de novo MDS. Nonetheless,
the IPPS-R has been shown to be prognostically inad­eq ­ uate in
tMDS, as in patient 1.7 The US MDS Clinical research con­sor­tium
reported shorter sur­vival in tMDS than in de novo MDS, includ­ing
in LR patients.8 Zeidan et al8 ana­lyzed out­comes in 370 patients
with tMDS to under­stand the prog­nos­tic util­ity of cur­rent risk
strat­i­fi­ca­tion tools in tMDS. The over­all sur­vival (OS) of patients
with tMDS was sig­nif­i­cantly worse than those with de novo dis­
ease (median OS, 19 vs 46 months).
The genet­ ics of tMDS has been inves­ ti­
gated to explain if
their “risk het­ero­ge­ne­ity” (as displayed by both cases) might
be par­tially related to inad­ver­tent inclu­sion of coin­ci­den­tal sec­
ond dis­ease due to genetic pre­dis­po­si­tion.9 A his­tory of ther­
apy might mean there was clonal selec­tion at some ances­tral
point, but the biol­ogy is het­ero­ge­neous, and some have more
indo­lent dis­ease, as in patient 2. Previous can­cer ther­apy with
radi­at­ ion, plat­i­num, and topoisomerase II inhib­i­tors, as in patient
1, pref­er­en­tially selects for muta­tions in DNA dam­age response
genes (TP53, PPM1D, CHEK2) in a recent large series.10 These lat­
ter patients will likely all­behave in an HR fash­ion, regard­less of
IPSS-R, and should be treated more aggres­sively with seri­ally
reassessment and alter­ations to ther­a­peu­tics. The man­age­ment
of patients with tMDS remains chal­leng­ing, and care­ful sur­veil­
lance and open­ness to change ther­apy as course dic­tates—to
lower inten­sity or higher inten­sity—are crit­i­cal.
WBC count was 2.4 × 109/L, ANC was 1.36 × 109/L, Hgb was
7.9 g/dL, and PLTs were 99 × 109/L. Marrow biopsy spec­i­men was
95% cel­lu­lar with ery­throid pre­dom­i­nance, trilineage dyspoi­
esis, and 3% blasts. There were 25% to 30% ring sideroblasts
with a dif­fuse increase in retic­u­lar fibro­sis. His final diag­no­sis
was MDS with ring sideroblasts and multilineage dys­pla­sia. His
kar­yo­type was 46,XY, and NGS showed SF3B1 R625H with a VAF
of 38% with­out comutations to sug­gest mye­lo­fi­bro­sis. IPSS-R
score was inter­me­di­ate-risk dis­ease at 3.5 with 1 for cyto­ge­net­
ics, 1 for 3% blasts, 1 for Hgb less than 8 g, and 0.5 for PLTs less
than 100 × 109/L.
Clinical lim­i­ta­tions in use of prog­nos­tic scor­ing sys­tems also
exist. Table 2 sum­ma­rizes addi­tional poten­tial poor prog­nos­tic
fac­tors in LR patients that fall out­side the IPSS-R. Grade 2 or more
mar­row fibro­sis may worsen prog­no­sis of patients with MDS.11
RBC trans­fu­sion depen­dency is an inde­pen­dent poor prog­nos­
ti­ca­tor. Recently, Hiwase et al12 exam­ined whether RBC-TD adds
prog­nos­tic value to the IPSS-R. In a mul­ti­var­i­ate anal­y­sis, their
cohort dem­on­strates that devel­op­ment of RBC-TD at any time
dur­ing the course of MDS is asso­ci­ated with poor OS, inde­pen­
dent of IPSS-R.12 The LeukaemiaNet MDS reg­is­try recently dem­on­
strated that a rel­at­ ive PLT drop of more than 25% results in a 22%
OS at 5 years.13 When the PLT drop is com­bined with RBC-TD at
6 months from diag­no­sis, the OS is a stag­ger­ingly low 9%.13 This
is a straight­for­ward and non­in­va­sive way to pre­dict evo­lu­tion
to HR dis­ease for LR MDS. The addi­tion of the inter­me­di­ate-risk
group to the IPSS-R has also been a clin­i­cal chal­lenge as it is
most often con­sid­ered still LR yet is highly het­ero­ge­neous. Ben­
ton et al14 performed an anal­y­sis of 298 inter­me­di­ate-risk patients
to assess HR fea­tures. Age older than 66 years, periph­eral blood
blasts of 2% or more, and RBC trans­fu­sion were sig­nif­i­cantly
asso­ci­ated with infe­rior sur­vival.14 Patient 3 high­lights sev­eral of
these issues.

CLINICAL CASE 3 Adverse molec­ul­ar fea­tures may “upstage” an LR patient

A pre­vi­
ously healthy 39-year-old man sought treat­ment for Most cases of MDS have 1 or sev­eral somatic gene muta­tions.
fatigue and cytopenias. He had no spleno­ meg­aly. His total SF3B1MUT is gen­er­ally con­sid­ered a favor­able prog­nos­tic marker.15

430  |  Hematology 2021  |  ASH Education Program

Despite this, SF3B1MUT MDS is het­ero­ge­neous in its path­o­logic
fea­tures, treat­ment responses, and clin­i­cal out­comes,16 as in
patient 3. Understanding this het­ero­ge­ne­ity is cru­cial for dis­
ease prog­nos­ti­ca­tion and man­age­ment, and it is rel­e­vant to a
recent pro­posal by the MDS International Working Group that
SF3B1MUT MDS should be clas­si­fied as a dis­tinct noso­logic enti­
ty.15 Although his­tor­i­cally, SF3B1MUT gen­er­ally rep­re­sents indo­lent
dis­ease in LR MDS, comutation with RUNX1 and EZH2 has been
asso­ci­ated with a worse prog­no­sis.15
The inde­ pen­ dent poor prog­ nos­
tic value of many somatic
muta­tions, includ­ing ASXL1, SRSF2, RUNX1, and TP53, has been
established ret­ro­spec­tively.17 The inte­gra­tion of these unfa­vor­
able muta­tions in spe­cific scor­ing sys­tems for LR MDS has been

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/428/1851953/428dezern.pdf by guest on 13 December 2021

reported. In addi­tion, an increased num­ber of muta­tions pres­ent
in a sin­gle patient also sug­gests a less favor­able out­come.18
Of par­tic­u­lar impor­tance is the TP53 muta­tion. Specifically,
this is pres­ent in 20% of cases of LR MDS with iso­lated dele­tion
5q minus syn­drome (del5q; but rare in other LR MDS) and asso­ci­
ated with an HR of AML and poorer sur­vival.19 However, a recent
study ana­ lyzed more than 3000 patients with MDS for TP53
muta­tions and alle­lic imbal­ances.19 One-third of TP53-mutated
patients had monoallelic muta­tions, whereas two-thirds had mul­
ti­ple mutations con­sis­tent with biallelic targeting. Established
asso­ci­a­tions with com­plex kar­yo­type, few co-occur­ring muta­
tions, high-risk pre­sen­ta­tion, and poor out­comes were spe­cific to
multihit patients only. Interestingly, monoallelic patients did not
dif­fer from TP53 wild-type patients in out­comes and response to Figure 1. Therapeutic paradigm in LR MDS. MDS-RS, myelodys­
ther­apy in this series.19 Nonetheless, most still use the pres­ence plastic syndrome with ring sideroblasts.
of TP53 muta­tion (even monoallelic) to upstage LR dis­ease given
its long asso­ci­a­tion with adverse out­comes, as in patient 1.
dif­fer­en­ti­a­tion to inherited pre­dis­po­si­tion beyond poten­tial bio­
logic under­stand­ing of increased risk fea­tures to LR dis­ease.20

Outcomes in LR dis­ease with cur­rently avail­­able ther­a­pies

CLINICAL CASE 3 (Con­t in­u ed)
Four months after diag­no­sis, the patient remained well, but blood
counts were lower and treat­ ment warranted for cytopenias,
despite still inter­me­di­ate risk by IPSS-R on repeat mar­row. Con­
cern for his MDS diag­no­sis at age less than 40 years prompted
deeper genetic assess­ment. Ultimately, a germline pre­dis­po­si­
tion gene asso­ci­ated with mye­loid malig­nan­cies was iden­ti­fied
in the patient (and also in his brother and father) with the RTEL1
muta­tion. He then under­went unre­lated donor BMT with nor­mal
counts and full donor chi­me­rism with­out com­pli­ca­tions at 1 year.
Inherited pre­dis­po­si­tion to MDS may alter
risk assess­ment
Inherited var­i­ants are rec­og­nized increas­ingly as predisposing
patients to MDS, and germline sus­cep­ti­bil­ity to mye­loid malig­nan­
cies is included in the lat­est WHO clas­si­fi­ca­tion of hema­to­poi­etic
malig­nan­cies. Moreover, clin­i­cal guide­lines now call for assess­
ment of germline pre­dis­po­si­tion at MDS diag­no­sis.20,21 A recent
series showed that the fre­quency of germline var­i­ants in adults
aged 18 to 40 years diag­nosed with MDS is high (con­sis­tent with
other reports), and the asso­ci­ated germline syn­drome is often not
appar­ent from the patient’s med­i­cal or fam­ily his­tory.22 Patient 3,
with his youn­ger age, illus­trates this need for deeper genet­ics
assess­ment. These patients may behave in a vari­ety of ways with
HR man­i­fes­ta­tions; coexisting somatic muta­tions are not uncom­
mon in these patients.23 There are also many ben­e­fits of real-time
Prakash Singh Shekhawat
Figure 1 depicts many of the stan­dard treat­ments24-27 often em­
ployed in LR dis­ease, regard­less of HR fea­tures, as well as fre­quent
need for reassessment if response to treat­ment is not pos­i­tive. Al­
though hypomethylating agents (HMAs), such as azacitidine and
decitabine, and lenalidomide (LEN) are not approved in many
countries for LR dis­ease, the focus here is on drugs for con­sid­
er­ation in our patients and the avail­­able data in the lit­er­a­ture.
For throm­bo­cy­to­pe­nic LR patients, eltrombopag is mod­estly
clin­i­cally effec­tive in rais­ing PLT counts and reduc­ing bleed­ing
events.28 It is not cur­rently approved for MDS and not yet known
if it will be safe. The assess­ment of long-term safety and effi­cacy
of eltrombopag and its effect on sur­vival (phase 2 part of study) is
still ongo­ing28 and likely will be held to a high stan­dard to ensure
no increased clonal evo­lu­tion. Romiplostim has also been stud­ied
in MDS and reduces PLT trans­fu­sions and bleed­ing events with
good OS with­out increas­ing rates of AML.29 Clinical tri­als of novel
ther­a­pies in LR dis­ease, such as imetelstat,30 may yet be a ­ ble to
mod­ify dis­ease risk and change the nat­u­ral his­tory.
Prebet et  al. have looked at out­comes posttreatment with
LEN in LR MDS with and with­out del5q.31,32 In patients with del5q,
OS fol­low­ing LEN fail­ure was a mere 23 months,31 whereas in
non-del5q, it was 43 months.32 LEN resis­tance in del5q is also
asso­ci­ated with a high risk of AML and a 5-year prob­a­bil­ity of
sur­vival of only 25%.31 For non-del5q, sub­se­quent ther­apy with
HMAs was asso­ci­ated with a prolonged sur­vival com­pared with
best sup­port­ive care (median OS, 51 vs 36 months, P = .01).32 This
Lower risk but high risk | 431
sug­gests an early switch to HMAs in LEN-resis­tant patients is
rea­son­able. The time to resis­tance or fail­ure of ther­apy varies in
the lit­er­at­ure, but a trial prob­a­bly should be no lon­ger than 16
weeks for non-del5q patients as more than 90% of respond­ers
were seen in the phase 3 MDS-005 trial after those 4 cycles.25
In the same study, the authors explored gene muta­tions and
response to LEN.33 The ana­ly­ses revealed that patients with 4 or
more muta­tions were more likely to have a high serum eryth­ro­
poi­et­in (EPO) level (>500 mU/mL); in addi­tion, high serum EPO
level was asso­ci­ated with the pres­ence of muta­tions in any of
5 genes asso­ci­ated with poorer out­comes in the Bejar et al34
anal­y­sis (ASXL1, ETV6, EZH2, RUNX1, and TP53). Taken together
with the observed asso­ci­a­tion between ASXL1 muta­tions and
low response rate to LEN, this pro­vi­des fur­ther insight into pre­
vi­ous data that high base­line EPO level pre­dicts for non­re­sponse
to LEN in these LR patients.25,33 Finally, even in LR patients, non­
re­sponse to HMAs was asso­ci­ated with a median sur­vival of 17
months in a US MDS con­sor­tium study,35 which cor­re­lates with
increas­ing num­ber of muta­tions.36 Thus, response to all­lines of
ther­apy in LR MDS should also be con­sid­ered for tran­si­tions of
ther­apy. Clinical tri­als should also always be con­sid­ered. Increas­
ingly, tri­als for patients with MDS include a clause for eli­gi­bil­ity
that allows for upstaging, such as “patients with inter­me­di­ate
risk by R-IPSS with high-risk molec­ul­ar fea­tures includ­ing TP53,
ASXL1, EZH2, and/or RUNX1 muta­tions are also eli­gi­ble.” I believe
this is a rea­son­able prac­tice to apply broadly.
Consideration of allo­ge­neic trans­plan­ta­tion (BMT) in LR
dis­ease
Historically, BMT has been reserved for patients with HR MDS,
given its poten­tial cura­tive ben­e­fit is off­set by the mor­bid­ity and
mor­tal­ity that can result from the pro­ce­dure.37,38 Nonetheless,
efforts have been made to weigh the toxicities of this “higher
risk, higher reward” pro­ce­dure in LR MDS, and many LR patients
are offered BMT. The National Comprehensive Network Guide­
lines and other con­sen­sus guide­lines rec­om­mend BMT for pa­
tients with MDS early in their dis­ease if they have inter­me­di­ate 2
or high-risk cat­e­gory per the IPSS and later dur­ing their dis­ease
course (prior to AML pro­gres­sion) if they have LR dis­ease.
Although no pro­ spec­ tive series exist cur­ rently, there are
reported out­ comes of BMT in LR patients included in many
ret­ro­spec­tive series.39-43 Only 1 ded­i­cated ret­ro­spec­tive study
reviewed out­ comes in LR patients, and the results were not
encour­ag­ing.42 In a study of patients who under­went trans­plan­ta­
tion between 2000 and 2011 by the European Society for Blood
and Marrow Transplantation (EBMT), 246 IPSS low-risk (21%) or
inter­me­di­ate 1 (79%) patients had a 3-year OS and PFS of 57%
and 54%, respec­tively. There was a high nonrelapse mor­tal­ity
at 30%. The best out­come was seen in patients who received
periph­eral blood stem cells and had a matched related donor,
although alter­na­tive donors were rel­a­tively rare. The con­di­tion­
ing reg­i­men inten­sity and patient age did not appear to influ­ence
out­come in this het­er­og­e­nous series.42 In a series from Asia, LR
patients had a sig­nif­i­cantly bet­ter out­come than HR patients.43
Prognosis was also sig­nif­i­cantly influ­enced by the genetic risk.
LR patients who dis­play HR fea­tures such as a com­plex kar­yo­
type or TP53 muta­tions do poorly, whereas LR 5q minus patients
had an OS of about 40%.43 Additional expla­na­tions of incor­po­
ra­tion of bio­logic data into BMT out­come ana­ly­ses are seen in a
large series from the Center for International Blood and Marrow
432  |  Hematology 2021  |  ASH Education Program
Transplant Research. The authors reported 1514 patients with
MDS who under­went trans­plan­ta­tion, of whom 116 were con­sid­
ered LR. These patients were con­sid­ered favor­able, with an age
at trans­plant from less than 1 year to less than 40 years, no tMDS,
no TP53 or Ras path­way gene muta­tion, no throm­bo­cy­to­pe­nia,
and mar­row blasts less than 15%.40 Although not a clas­si­cal def­i­ni­
tion of LR patients, this cohort of chil­dren and young adults had a
favor­able 3-year OS of 82%.40 The sup­ple­men­tary data review OS
curves by IPSS spe­cif­i­cally and sug­gest that low-risk and inter­me­
di­ate 1 patients with­out TP53 muta­tions have an OS at 3 years of
approx­i­ma­tely 50%.
To aid in deci­sion mak­ing, some series from the EBMT and
Center for International Blood and Marrow Transplant Research
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/428/1851953/428dezern.pdf by guest on 13 December 2021
have devel­oped scor­ing sys­tems or deci­sion mod­els that adjust
to indi­vid­ual patient risk. Although not spe­cific to LR patients, the
molec­u­lar and other clin­i­cal fea­tures asso­ci­ated with less desir­
able out­comes can been seen. Both of these series had an OS in
LR patients of about 70% at 3 years or 4 years.41,44 The Gruppo
Italiano Trapianto Midollo Osseo e Terapia Cellulare reported a
BMT Mar­kov anal­y­sis on 1728 patients with MDS cat­e­go­rized by
the IPSS-R. Similar to older ana­ly­ses, a sur­vival advan­tage was
seen when delaying trans­plant in very low- and low-risk patients,
whereas post­pone­ment was del­e­te­ri­ous in inter­me­di­ate-risk
patients.45 This is in keep­ing with the cur­rent approach to only
con­sider BMT in LR patients with dis­ease evo­lu­tion or lack of
response to pri­mary treat­ments. Patients with increased RBC
needs, lack of response to HMAs, bone mar­ row fibro­ sis, HR
molec­u­lar fea­tures, and tMDS are asso­ci­ated with a poor prog­
no­sis, and early con­sid­er­ation for BMT is often suggested.35,46,47
Genomic fea­tures are rel­e­vant for predicting sur­vival after BMT,
supporting the ratio­nale to include this infor­ma­tion for trans­plan­
ta­tion deci­sion mak­ing in MDS.48,49 Many of these fea­tures would
have been appli­ca­ble in both patients 1 and 3.
Additional data includ­ing pro­spec­tive stud­ies, poten­tially
with a ran­ dom­ ized design com­ pared with the cur­ rent stan­
dard of delaying BMT, and includ­ing a wider donor pool with
haploidentical donors are needed for an improved under­stand­
ing and estab­lish­ment of the BMT approach for LR patients. An
ongo­ ing pro­ spec­tive nonrandomized trial in France is enroll­
ing patients with LR dis­ ease and at least 1 high-risk fea­ ture
(NCT02757989), and results will be use­ful to the field. Nonethe­
less, BMT remains the only poten­tially cura­tive option for MDS,
and no options avail­­able at pres­ent seem to be a ­ ble to replace
this path to poten­tial cure; we must use this for patients when
the risk/ben­e­fit pro­file is ratio­nal.
Conclusions
LR MDS is a het­ero­ge­neous group of dis­or­ders. Currently avail­­
able prog­nos­tic scor­ing sys­tems aid in risk strat­i­fi­ca­tion but do
not cap­ture all­impor­tant var­i­ables, and serial reassessment of
the patient to fully clas­sify indi­vid­ual risk is impor­tant. The goal
is a pro­ac­tive approach in all­LR patients to extend quan­tity of
life with qual­ity through use of all­avail­­able treat­ments avail­­able.
Conflict-of-inter­est dis­clo­sure
Amy E. DeZern has received hon­ o­
raria from Bristol Meyers
Squibb, Novartis, Taiho, and Abbvie as a con­sul­tant.
Off-label drug use
Amy E. DeZern: imetelstat or eltrombopag for MDS.
Correspondence
Amy E. DeZern, Division of Hematologic Malignancies, Sidney
Kimmel Cancer Center at Johns Hopkins, 1650 Orleans St, CRBI
Room 3M87, Baltimore, MD 21287-0013; e-mail: adezern1@jhmi​
­.edu.
References
1. Greenberg PL, Tuechler H, Schanz J, et al. Revised inter­na­tional prog­nos­tic
scor­ing sys­tem for myelodysplastic syn­dromes. Blood. 2012;120(12):2454-
2465.
2. Bejar R et al. Somatic mutations in MDS patients are associated with clin­
ical features and predict prognosis independent of the IPSS-R: Analysis of
combined datasets from the International Working Group for Prognosis in
MDS-Molecular Committee/Blood/American Society of Hematology (ash­
publications.org). Blood. 2015;126(23):907.
3. Kuendgen A, Nomdedeu M, Tuechler H, et  al. Therapy-related myelo­
dysplastic syn­dromes deserve spe­cific diag­nos­tic sub-clas­si­fi­ca­tion and
risk-strat­i­fi­ca­tion-an approach to clas­si­fi­ca­tion of patients with t-MDS. Leu-
kemia. 2021;35(3):835-849.
4. Bersanelli M, Travaglino E, Meggendorfer M, et al. Classification and per­
son­al­ized prog­nos­tic assess­ment on the basis of clin­i­cal and geno­mic fea­
tures in myelodysplastic syn­dromes. J Clin Oncol. 2021;39(11):1223-1233.
5. Duncavage EJ, Schroeder MC, O’Laughlin M, et al. Genome sequenc­ing as
an alter­na­tive to cyto­ge­netic anal­y­sis in mye­loid can­cers. N Engl J Med.
2021;384(10):924-935.
6. Swerdlow S. WHO Classification of Tumours of Haematopoietic and Lym-
phoid Tissues. Lyon International Agency for Research on Cancer; 2017.
7. Nazha A, Seastone DP, Keng M, et al. The Revised International Prognostic
Scoring System (IPSS-R) is not pre­dic­tive of sur­vival in patients with sec­
ond­ary myelodysplastic syn­dromes. Leuk Lymphoma. 2015;56(12):3437-
3439.
8. Zeidan AM, Al Ali N, Barnard J, et al. Comparison of clin­i­cal out­comes and
prog­nos­tic util­ity of risk strat­i­fi­ca­tion tools in patients with ther­apy-related
vs de novo myelodysplastic syn­dromes: a report on behalf of the MDS Clin­
ical Research Consortium. Leukemia. 2017;31(6):1391-1397.
9. Kuzmanovic T, Patel BJ, Sanikommu SR, et al. Genomics of ther­apy-related
mye­loid neo­plasms. Haematologica. 2020;105(3):e98-e101.
10. Bolton KL, Ptashkin RN, Gao T, et  al. Cancer ther­apy shapes the fit­ness
land­scape of clonal hema­to­poi­e­sis. Nat Genet. 2020;52(11):1219-1226.
11. Ramos F, Robledo C, Izquierdo-García FM, et al; Span­ish Group for Myelo­
dysplastic Syndromes (GESMD). Bone mar­row fibro­sis in myelodysplastic
syn­dromes: a pro­spec­tive eval­u­a­tion includ­ing muta­tional anal­y­sis. Onco-
target. 2016;7(21):30492-30503.
12. Hiwase DK, Singhal D, Strupp C, et al. Dynamic assess­ment of RBC-trans­fu­sion
depen­dency improves the prog­nos­tic value of the revised-IPSS in MDS
patients. Am J Hematol. 2017;92(6):508-514.
13. Itzykson R, Crouch S, Travaglino E, et al; Euro­pean MDS Registry mem­bers.
Early plate­let count kinet­ics has prog­nos­tic value in lower-risk myelodys­
plastic syn­dromes. Blood Adv. 2018;2(16):2079-2089.
14. Benton CB, Khan M, Sallman D, et al. Prognosis of patients with inter­me­di­
ate risk IPSS-R myelodysplastic syn­drome indi­cates var­i­able out­comes and
need for mod­els beyond IPSS-R. Am J Hematol. 2018;93(10):1245-1253.
15. Malcovati L, Stevenson K, Papaemmanuil E, et al. SF3B1-mutant MDS as a
dis­tinct dis­ease sub­type: a pro­posal from the International Working Group
for the Prognosis of MDS. Blood. 2020;136(2):157-170.
16. Dalton WB, Helmenstine E, Pieterse L, et al. The K666N muta­tion in SF3B1
is asso­ci­ated with increased pro­gres­sion of MDS and dis­tinct RNA splic­ing.
Blood Adv. 2020;4(7):1192-1196.
17. Bejar R, Stevenson KE, Caughey BA, et al. Validation of a prog­nos­tic model
and the impact of muta­tions in patients with lower-risk myelodysplastic
syn­dromes. J Clin Oncol. 2012;30(27):3376-3382.
18. DeZern AE. Treatments targeting MDS genet­ics: a fool’s errand? Hematol-
ogy Am Soc Hematol Educ Program. 2018;2018(1):277-285.
19. Bernard E, Nannya Y, Hasserjian RP, et al. Implications of TP53 alle­lic state
for genome sta­bil­ity, clin­i­cal pre­sen­ta­tion and out­comes in myelodysplas­
tic syn­dromes. Nat Med. 2020;26(10):1549-1556.
20. Schratz KE, DeZern AE. Genetic pre­dis­po­si­tion to myelodysplastic syn­
drome in clin­i­cal prac­tice. Hematol Oncol Clin North Am. 2020;34(2):333-
356.
21. Greenberg PL, Stone RM, Al-Kali A, et al. Myelodysplastic syn­dromes, ver­
sion 2.2017, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr
Canc Netw. 2017;15(1):60-87.
Prakash Singh Shekhawat
22. Feurstein S, Churpek JE, Walsh T, et al. Germline var­i­ants drive myelodys­
plastic syn­drome in young adults. Leukemia. 2021;35(8):2439-2444.
23. Schratz KE, Haley L, Danoff SK, et al. Cancer spec­trum and out­comes in the
Men­de­lian short telo­mere syn­dromes. Blood. 2020;135(22):1946-1956.
24. Fenaux P, Platzbecker U, Mufti GJ, et al. Luspatercept in patients with low­
er-risk myelodysplastic syn­dromes. N Engl J Med. 2020;382(2):140-151.
25. Santini V, Almeida A, Giagounidis A, et  al. Randomized phase III study
of lenalidomide ver­ sus pla­ cebo in RBC trans­ fu­sion-depen­dent patients
with lower-risk non-del(5q) myelodysplastic syn­ dromes and inel­ i­
gi­
ble for or refrac­tory to eryth­ro­poi­e­sis-stim­u­lat­ing agents. J Clin Oncol.
2016;34(25):2988-2996.
26. Jabbour E, Short NJ, Montalban-Bravo G, et al. Randomized phase 2 study
of low-dose decitabine vs low-dose azacitidine in lower-risk MDS and
MDS/MPN. Blood. 2017;130(13):1514-1522.
27. Stahl M, DeVeaux M, de Witte T, et al. The use of immu­no­sup­pres­sive ther­
apy in MDS: clin­i­cal out­comes and their pre­dic­tors in a large inter­na­tional
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/428/1851953/428dezern.pdf by guest on 13 December 2021
patient cohort. Blood Adv. 2018;2(14):1765-1772.
28. Oliva EN, Alati C, Santini V, et al. Eltrombopag ver­sus pla­cebo for low-risk
myelodysplastic syn­dromes with throm­bo­cy­to­pe­nia (EQoL-MDS): phase 1
results of a sin­gle-blind, randomised, con­trolled, phase 2 supe­ri­or­ity trial.
Lancet Haematol. 2017;4(3):e127-e136.
29. Kantarjian HM, Fenaux P, Sekeres MA, et  al. Long-term fol­low-up for up
to 5 years on the risk of leukaemic pro­ gres­sion in throm­ bo­cy­to­
pe­nic
patients with lower-risk myelodysplastic syn­dromes treated with romi­
plostim or pla­cebo in a randomised dou­ble-blind trial. Lancet Haematol.
2018;5(3):e117-e126.
30. Steensma DP, Fenaux P, Van Eygen K, et al. Imetelstat achieves mean­ing­ful
and dura­ble trans­fu­sion inde­pen­dence in high trans­fu­sion-bur­den patients
with lower-risk myelodysplastic syn­dromes in a phase II study. J Clin Oncol.
2021;39(1):48-56.
31. Prebet T, Cluzeau T, Park S, et al. Outcome of patients treated for myelo­
dysplastic syn­dromes with 5q dele­tion after fail­ure of lenalidomide ther­
apy. Oncotarget. 2017;8(47):81926-81935.
32. Prebet T, Toma A, Cluzeau T, et al. Outcome of patients treated for myel­
odysplastic syn­dromes with­out dele­tion 5q after fail­ure of lenalidomide
ther­apy. Oncotarget. 2017;8(23):37866-37874.
33. Santini V, Fenaux P, Giagounidis A, et  al. Impact of somatic muta­tions on
response to lenalidomide in lower-risk non-del(5q) myelodysplastic syn­
dromes patients. Leukemia. 2021;35(3):897-900.
34. Bejar R, Stevenson K, Abdel-Wahab O, et al. Clinical effect of point muta­
tions in myelodysplastic syndromes. N Engl J Med. 2011;364:2496-506.
35. Jabbour EJ, Garcia-Manero G, Strati P, et al. Outcome of patients with low-
risk and inter­ me­di­ate-1-risk myelodysplastic syn­ drome after hypometh­
ylating agent fail­ ure: a report on behalf of the MDS Clinical Research
Consortium. Cancer. 2015;121(6):876-882.
36. Montalban-Bravo G, Takahashi K, Patel K, et  al. Impact of the num­ber of
muta­tions in sur­vival and response out­comes to hypomethylating agents
in patients with myelodysplastic syn­dromes or myelodysplastic/mye­lo­
pro­lif­er­a­tive neo­plasms. Oncotarget. 2018;9(11):9714-9727.
37. Cutler CS, Lee SJ, Greenberg P, et al. A deci­sion anal­y­sis of allo­ge­neic bone
mar­row trans­plan­ta­tion for the myelodysplastic syn­dromes: delayed trans­
plan­ta­tion for low-risk myelodysplasia is asso­ci­ated with improved out­
come. Blood. 2004;104(2):579-585.
38. Koreth J, Pidala J, Perez WS, et al. Role of reduced-inten­sity con­di­tion­ing
allo­ge­neic hema­to­poi­etic stem-cell trans­plan­ta­tion in older patients with
de novo myelodysplastic syn­dromes: an inter­na­tional col­lab­o­ra­tive deci­
sion anal­y­sis. J Clin Oncol. 2013;31(21):2662-2670.
39. Polona Novak P, Zabelina T, Wolschke C, Ayuk F, Christopeit M, Kröger N.
Allogeneic stem cell trans­plan­ta­tion for patients with lower-risk myelodys­
plastic syn­drome. Biol Blood Marrow Transplant. 2020;26(11):2047-2052.
40. Lindsley RC, Saber W, Mar BG, et al. Prognostic muta­tions in myelodysplas­
tic syn­drome after stem-cell trans­plan­ta­tion. N Engl J Med. 2017;376(6):536-
547.
41. Gagelmann N, Eikema D-J, Stelljes M, et  al. Optimized EBMT trans­plant-
spe­cific risk score in myelodysplastic syn­dromes after allo­ge­neic stem-cell
trans­plan­ta­tion. Haematologica. 2019;104(5):929-936.
42. Robin M, Porcher R, Zinke-Cerwenka W, et al. Allogeneic haematopoietic
stem cell trans­plant in patients with lower risk myelodysplastic syn­drome:
a ret­ro­spec­tive anal­y­sis on behalf of the Chronic Malignancy Working Party
of the EBMT. Bone Marrow Transplant. 2017;52(7):1081.
43. Yoshizato T, Nannya Y, Atsuta Y, et al. Genetic abnor­mal­i­ties in myelodys­
plasia and sec­ond­ary acute mye­loid leu­ke­mia: impact on out­come of stem
cell trans­plan­ta­tion. Blood. 2017;129(17):2347-2358.
Lower risk but high risk | 433
44. Shaffer BC, Ahn KW, Hu Z-H, et al. Scoring sys­tem prog­nos­tic of out­come
in patients under­go­ing allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion for
myelodysplastic syn­drome. J Clin Oncol. 2016;34(16):1864-1871.
45. Della Porta MG, Jackson CH, Alessandrino EP, et  al. Decision anal­y­sis of
allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion for patients with myel­
odysplastic syn­ drome strat­ i­fied according to the revised International
Prognostic Scoring System. Leukemia. 2017;31(11):2449-2457.
46. de Witte T, Bowen D, Robin M, et al. Allogeneic hema­to­poi­etic stem cell
trans­plan­ta­tion for MDS and CMML: rec­om­men­da­tions from an inter­na­
tional expert panel. Blood. 2017;129(13):1753-1762.
47. Prébet T, Gore SD, Esterni B, et al. Outcome of high-risk myelodysplastic
syn­drome after azacitidine treat­ment fail­ure. J Clin Oncol. 2011;29(24):3322-
3327.
434  |  Hematology 2021  |  ASH Education Program
48. Della Porta MG, Alessandrino EP, Bacigalupo A, et al; Gruppo Italiano Trapi­
anto di Midollo Osseo. Predictive fac­tors for the out­come of allo­ge­neic
trans­plan­ta­tion in patients with MDS strat­i­fied according to the revised
IPSS-R. Blood. 2014;123(15):2333-2342.
49. Della Porta MG, Gallì A, Bacigalupo A, et al. Clinical effects of driver somatic
muta­tions on the out­comes of patients with myelodysplastic syn­dromes
treated with allo­ge­neic hema­to­poi­etic stem-cell trans­plan­ta­tion. J Clin
Oncol. 2016;34(30):3627-3637.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000277
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/428/1851953/428dezern.pdf by guest on 13 December 2021
 MDS: BEYOND A ONE - SIZE - FITS -ALL APPROACH

EVIDENCE-BASED MINIREVIEW

Molecular precision and clinical uncertainty:

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/435/1852132/435richardson.pdf by guest on 13 December 2021

should molecular profiling be routinely used
to guide risk stratification in MDS?
Daniel R. Richardson1 and Amy E. DeZern2
Division of Hematology, Department of Medicine, Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC;
1

Department of Oncology, Sidney Kimmel Cancer Center at Johns Hopkins, Baltimore, MD

2

This is a focused clinical vignette and review of the literature in MDS to discuss the application of molecular sequencing
for risk stratification in MDS. The authors utilize an exemplar patient case and explain the advantages and disadvantages,
based on available data, of routine use of this testing for MDS patients.

LEARNING OBJECTIVES
• Review the prognostic significance of mutational profiles in MDS
• Identify appropriate approaches to use molecular profiling for risk stratification in MDS

development of next­generation sequencing (NGS) assays

CLINICAL CASE
A 67­year­old woman sought treatment for progressive
fatigue initially attributed to caring for her 18­month­old
grandson. She was found to have pancytopenia (hemo­
globin of 10.2 g/dL, mean cell volume of 104 fL, platelet
count of 77 × 109/L, and white blood cell count of 2.1 × 109/L
with absolute neutrophil count of 0.7 × 109/L). Bone mar­
row biopsy specimen demonstrated trilineage dysplasia
with 1% blasts. Cytogenetics were 46,XX [20]. Interna­
tional Prognostic Scoring System – Revised (IPSS­R) risk
category was low. Mutational testing demonstrated muta­
tions in ASXL1 (variant allele fraction [VAF] 34%), SRSF2
(VAF 42%), TET2 (VAF 18%), and SETBP1 (VAF 12%). Does
this molecular profile modify risk stratification or manage­
ment decisions in this patient?
Introduction
Myelodysplastic syndromes (MDS) are a group of related
hematologic neoplasms that cause dysplasia and ineffective
hematopoiesis and have a propensity to progress to acute
myeloid leukemia (AML). The diagnosis of MDS requires mor­
phologic dysplasia in 1 or more cell lines and/or character­
istic cytogenetic abnormalities. Over the past decade, the
Prakash Singh Shekhawat
has allowed for the identification of more than 50 different
recurrent driver gene mutations that are associated with
MDS.1,2 Most patients with MDS will have at least 1 identi­
fied somatic mutation, whereas a typical patient will har­
bor a median of 2 or 3 driver mutations.3 NGS assays are
increasingly used in the diagnostic evaluation of patients
with cytopenias.4 Numerous studies have also evaluated the
prognostic significance of such mutations. Best practices
for integrating molecular testing into existing risk prediction
modeling have yet to be established, and substantial clinical
variability exists among practicing hematologists.5 Here we
review the current evidence for use of molecular profiling to
guide risk stratification in clinical practice.
Molecular precision
Several notable studies using NGS and high­throughput
sequencing have demonstrated the independent prog­
nostic significance of recurrent molecular mutations in
patients with MDS. Consistently identified driver mutant
genes include mutations involved in RNA splicing (SF3B1,
SRSF2, U2AF1, and ZRSR2), DNA methylation (TET2, DN-
MT3A, IDH1, and IDH2), chromatin modification (ASXL1 and
EZH2), transcription regulation (RUNX1), DNA repair (TP53),
and signal transduction (CBL, NRAS, and KRAS).6
Routine molecular profiling in MDS for risk | 435
 Papaemmanuil et al1 sequenced 111 genes across 738 patients
with MDS, chronic myelomonocytic leukemia, and MDS/mye­lo­
pro­lif­er­a­tive neo­plasms. Mutations in SF3B1 (24%), TET2 (22%), and
SFSF2 (14%) were most com­mon, followed by ASXL1, DNMT3A,
RUNX1, U2AF1, TP53, and EZH2. Mutations in TP53, U2AF1, RUNX1,
SRSF2, IDH2, CUX1, ASXL1, and BCOR were neg­a­tively asso­ci­ated
with over­all sur­vival (OS), whereas muta­tions in SF3B1 were asso­
ci­ated with improved OS. Leukemia-free sur­vival was neg­a­tively
asso­ci­ated with increased num­ber of driver muta­tions and cyto­
ge­netic abnor­mal­i­ties.
Bejar et al7 in a study using NGS and mass spec­trom­e­try–
based genotyping of 439 patients with MDS dem­on­strated that
muta­tions in 10 genes were asso­ci­ated with OS (TET2, ASXL1,
RUNX1, TP53, EZH2, NRAS, JAK2, ETV6, CBL, and IDH2). Mutations
in 6 genes (TP53, EZH2, ETV6, RUNX1, CBL, and ASXL1) neg­a­tively
affected OS after adjusting for IPSS risk group and age. The pres­
ence of a TP53 muta­tion had the larg­est inde­pen­dent asso­ci­a­tion
with sur­vival among muta­tions (haz­ard ratio, 2.48).
Haferlach et al3 used high-through­put deep sequenc­ing of
104 genes to ana­lyze the geno­mic and clin­i­cal data of more
than 900 patients with MDS. Twenty-five genes were iden­ti­fied
to be asso­ci­ated with sur­vival out­comes on uni­var­i­ate anal­y­sis.
Only muta­tions in SF3B1 were asso­ci­ated with an improved clin­
i­cal out­come. Other recent ana­ly­ses con­firm the impor­tance of
muta­tions iden­ti­fied in these stud­ies, high­light­ing the prog­nos­
tic value of spe­cific muta­tions inde­pen­dent of IPSS-R.5,8
Variant allele fre­quency is the per­cent­age of var­i­ant reads
in the total num­ber of reads (var­i­ant plus ref­er­ence reads) and
is used to approx­ im
­ ate the pro­ por­tion of DNA mol­ e­
cules in
the orig­i­nal spec­i­men that carry the spe­cific var­i­ant. Tradition­
ally, muta­tions with a VAF of 5% or less are not thought to be
clin­i­cally sig­nif­i­cant. More data are emerg­ing that prog­nos­tic
out­comes asso­ci­ated with indi­vid­ual muta­tions may be depen­
dent on the VAF of muta­tions, although this effect may be lim­
ited to cer­tain muta­tions (such as TP53, TET2, and SF3B1).9-12 In
TP53-mutated MDS, for exam­ple, increas­ing VAF was asso­ci­ated
with worse prog­no­sis, with an increase in haz­ard ratio of 1.02%
per 1% VAF increase.13 Patients with a low TP53-muta­tional bur­
den (VAF <10%) and with­out a com­plex kar­yo­type may have out­
comes bet­ter than expected. Those patients with a high VAF are
less likely to respond to hypomethylating agents.13
Molecular pro­ fil­
ing is also essen­ tial in esti­
mat­
ing risk for
patients with clonal hema­to­poi­e­sis of inde­ter­mi­nate poten­
tial (CHIP) and clonal cytopenia of unde­ter­mined sig­nif­i­cance
(CCUS), pre­cur­sor states to MDS. For exam­ple, in patients with
CCUS, muta­ tional sta­ tus is pre­dic­
tive of risk of pro­gres­
sion:
muta­tions in U2AF1, ZRSR2, SRSF2, JAK2, or RUNX1 are asso­ci­
ated with a high risk of pro­gres­sion to MDS (as high as 80%-
90% at 5 years), whereas iso­lated muta­tions in TET2, ASXL1, and
DNMT3A are asso­ci­ated with a lower risk of pro­gres­sion (~50%
at 5 years).14 Similar to find­ings in MDS, emerg­ing data have dem­
on­strated that dif­fer­ent muta­tional pat­terns and VAF val­ues are
likely impor­tant in predicting over­all risk in CHIP and CCUS.15
Although iden­ti­fic ­ a­tion of somatic muta­tions by NGS is an
impor­tant step for­ward in risk strat­i­fi­ca­tion for patients with MDS,
indi­vid­ual muta­tions are only a part of a more com­plex land­scape
of geno­mic abnor­mal­it­ies that drive pathol­ogy and deter­mine
out­comes. Using the same cohort of 944 patients men­tioned
above, Haferlach et al3 pro­posed novel prog­nos­tic mod­els incor­
po­rat­ing genetic data that strat­ify patients into 4 prognostically
dis­tinct sub­groups. These mod­els con­sis­tently outperformed
436  |  Hematology 2021  |  ASH Education Program
IPSS-R, dem­on­strat­ing the fea­si­bil­ity and impor­tance of incor­po­
rat­ing molec­u­lar data into prog­nos­tic mod­els. By apply­ing com­
pu­ta­tional approaches to large inter­na­tional data­bases, Bersanelli
et al16 explored the com­plex­ity of mul­ti­ple coexisting muta­tions
and iden­ti­fied 8 prognostically dis­tinct “geno­mic sub­types” of
MDS such as MDS with TP53 muta­tions and/or com­plex kar­yo­
type, MDS with AML-like muta­tional pat­terns, and SRSF2-related
MDS. Furthermore, by inte­grat­ing 63 clin­i­cal and geno­mic var­i­
ables, the authors were a ­ ble to develop a model to gen­er­ate an
esti­mated per­son­al­ized prob­a­bil­ity of sur­vival, which also sig­nif­i ­
cantly improved upon IPSS-R prog­nos­ti­ca­tion. Such approaches
likely bet­ter rep­re­sent the dis­ease biol­ogy in MDS, but fur­ther
work needs to be done to dem­on­strate the util­ity of such com­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/435/1852132/435richardson.pdf by guest on 13 December 2021
plex risk strat­i­fi­ca­tion sys­tems for clin­i­cal deci­sion mak­ing.
Clinical uncer­tainty
Despite the remark­ able tech­ no­log­
i­
cal prog­ ress that has en­
abled rapid molec­u­lar pro­fil­ing of patients with MDS and pre­
cur­sor states, sub­stan­tial clin­i­cal uncer­tainty remains about
how to apply molec­ul­ar data to indi­vid­ual patients. Numerous
prog­nos­tic mod­els have already been val­i­dated for use in MDS
and are in wide­ spread use, includ­ ing the IPSS, MDS WHO
Classification-based Prognostic Scoring System (WPSS), Lower-Risk
Prognostic Scoring System (LR-PSS), Global MD Anderson Prog­
nostic Scoring System (MDAPSS), French Prognostic Scoring
System (FPSS), and IPSS-R, among oth­ers. Multiple novel mod­els
have been pro­posed that seek to com­bine existing prog­nos­tic
scor­ing sys­tems with muta­tional data from emerg­ing stud­ies.1,3,16
Clinical use of these advanced mod­els incor­po­rat­ing genetic
data is cur­rently lim­ited due to the com­plex­ity of the mod­els
and the lack of fur­ther val­i­da­tion stud­ies. A stan­dard scor­ing sys­
tem incor­po­rat­ing molec­u­lar data and clin­i­co­path­o­logic data
remains to be established.
The lack of an accepted prog­nos­tic scor­ing sys­tem incor­
po­rat­ing molec­u­lar data makes prog­nos­ti­ca­tion for indi­vid­ual
patients par­tic­ul­arly chal­leng­ing. Hematologists already face the
chal­lenge of deter­min­ing which existing risk pre­dic­tion model
best applies to each patient; the addi­tion of molec­u­lar pro­fil­
ing adds yet another layer of com­ plex­ity and clin­i­
cal uncer­
tainty. Multiple ques­ tions remain: Are the adverse effects of
muta­tions on out­comes uni­form across stan­dard­ized risk mod­
els? Are they uni­form across risk categories? Is there a ceil­ing
effect to adverse prog­nos­tic fac­tors? For patients with mul­ti­ple
muta­ tions iden­ ti­fied, how should risk be esti­ mated? Multiple
stud­ies have dem­on­strated the adverse effect of hav­ing mul­ti­
ple coexisting muta­tions.1,3,7 Should this “comutation” effect be
thought of in par­al­lel or in sequence to the effect of indi­vid­ual
muta­tions? These ques­tions, along with many oth­ers, remain to
be addressed to improve risk strat­i­fi­ca­tion for patients with MDS.
Implications for clin­i­cal prac­tice
Although the use of molec­u­lar pro­fil­ing adds com­plex­ity to prog­
nos­ti­ca­tion and obtaining NGS is often hin­dered by the lack of
insur­ance cov­er­age, we sug­gest com­plet­ing molec­u­lar pro­fil­ing
on all­patients suspected or con­firmed to have MDS. The rea­son
for this is straight­for­ward: in some patients, molec­u­lar pro­fil­ing
will sub­stan­tially change man­age­ment. For the patient men­tioned
above, although IPSS-R risk is low, sev­eral fea­tures of the molec­u­lar
pro­file are impor­tant. The ASXL1 muta­tion, assum­ing it is a canon­
i­cal muta­tion resulting in a read­ing frame­shift, has been shown
to be asso­ci­ated with worse out­comes inde­pen­dent of IPSS-R.7
Furthermore, the fact that 4 onco­genic muta­tions were iden­ti­fied
also por­tends worse out­comes. The median leu­ke­mia-free sur­vival
for patients with 4 iden­ti­fied muta­tions was only 18 months in the
study by Papaemmanuil et al.1 So, although by IPSS-R, this patient
is low risk (esti­mated OS of 5.3 years and leu­ke­mia-free sur­vival of
10.8 years), the molec­u­lar pro­file should prompt more fre­quent
clin­i­cal mon­i­tor­ing and make con­sid­er­ation of early ther­apy rea­
son­able, includ­ing dis­cus­sions of an allo­ge­neic stem cell trans­
plant can­di­dacy. Several addi­tional sce­nar­ios require men­tion­ing.
Upstaging lower-risk dis­ease
Patients with lower-risk MDS often expe­ri­ence long-term sur­
vival with­out the need for aggres­sive ther­apy such as allo­ge­
enroll­ment in a clin­i­cal trial for patients with a com­plex kar­yo­
type and a TP53 muta­tion due to recently emerg­ing encour­ag­ing
results of small mol­e­cules that induce apo­pto­sis in TP53-mutated
cells such as APR-246. Response rates of more than 70% with a 50%
com­plete remis­sion rate were shown in 29 patients with MDS
receiv­ing APR-246 on the phase 1b/2 trial.28
Targetable muta­tions
Although there are cur­rently no ther­a­pies approved by the US Food
and Drug Administration that tar­get indi­vid­ual muta­tions in MDS,
the iden­ti­fi­ca­tion of spe­cific muta­tions may make a patient eli­gi­ble
for AML-style treat­ments in which targeted agents are approved or
for a clin­i­cal trial. Current ther­a­peu­tic clin­i­cal tri­als include patients

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/435/1852132/435richardson.pdf by guest on 13 December 2021

neic hema­to­poi­etic cell trans­plant. The iden­ti­fi­ca­tion of patients with iden­ti­fied muta­tions in TP53 (ClinicalTrials​­.gov Identifier:
who are higher risk for dis­ease com­pli­ca­tions is impor­tant to NCT04638309), IDH1 (numer­ous stud­ies includ­ing NCT03173248,
guide ther­a­peu­tic deci­sions. In a large anal­y­sis of patients pre­ NCT03503409, NCT03839771), IDH2 (NCT04522895, NCT03839771,
dicted to have lower-risk MDS by MDS-LR, the iden­ti­fi­ca­tion of NCT03383575), RUNX1 (NCT04874194), TET2 (NCT03397173), FLT3
a muta­tion in EZH2 predicted worse-than-expected out­comes (NCT04493138, NCT04027309), CBL (NCT04493138), and mul­ti­ple
after adjusting for mul­ti­ple rel­e­vant covariates.17 Similarly, the muta­tions (NCT02670525).
find­ing of a muta­tion with an inde­pen­dent poor prog­nos­tic sig­
nif­i­cance, includ­ing ASXL1, SRSF2, ETV6, RUNX1, and TP53, in
patients should prompt con­sid­er­ation of ear­lier ini­ti­a­tion of dis­ Existing chal­lenges
ease-mod­i­fy­ing ther­apy and/or closer clin­i­cal mon­i­tor­ing as in Obtaining qual­ity molec­u­lar pro­fil­ing on some patients remains
the case above. Furthermore, the iden­ti­fi­ca­tion of a muta­tion in a chal­lenge. Many insur­ance com­pa­nies deny cov­er­age for NGS
NPM1 (~4% of cases) has been asso­ci­ated with an aggres­sive dis­ assays or require prior autho­ri­za­tion that may be dif­fi­cult to ob­
ease course and rel­a­tively rapid pro­gres­sion to AML often with­ tain. We encour­age clear cod­ing as MDS to help with bill­ing cov­
in 12 months of diag­no­sis.18 Treating patients with NPM1 muta­ er­age. A let­ter of med­i­cal neces­sity cit­ing the National Compre­
tions sim­i­lar to patients with AML with inten­sive che­mo­ther­apy hensive Cancer Network guide­lines may also be of use.
and allo­ge­neic stem cell trans­plant has been asso­ci­ated with Some ref­er­ence lab­o­ra­to­ries are not rou­tinely reporting VAF,
improved out­comes.19 Similarly, FLT3 muta­tions, which are less which sub­stan­tially lim­its the inter­pre­ta­tion of data, and/or are
fre­quent in MDS com­pared with AML (~2% vs 35%), are asso­ci­ not clearly delin­eat­ing whether indi­vid­ual muta­tions are path­o­
ated with a com­plex kar­yo­type and a more rapid pro­gres­sion to genic. A con­sen­sus panel by the Association for Molecular Pathol­
AML, although the data are mixed.20-22 The iden­ti­fi­ca­tion of a new ogy with liai­son rep­re­sen­ta­tion from the Amer­i­can College of
FLT3-ITD muta­tion at the time of pro­gres­sion from MDS to AML is Medical Genetics and Genomics, Amer­ i­
can Society of Clinical
asso­ci­ated with dis­mal out­comes (1 month median OS).23 Oncology, and College of Amer­i­can Pathologists rec­om­mends
a 4-tiered reporting approach: tier I, var­i­ants with strong clin­i­cal
SF3B1-mutated MDS sig­nif­i­cance; tier II, var­i­ants with poten­tial clin­i­cal sig­nif­i­cance; tier
SF3B1-muated MDS is often con­ sid­
ered a dis­
tinct sub­
type of III, var­i­ants of unknown clin­i­cal sig­nif­i­cance; and tier IV, var­i­ants
MDS with an indo­lent course.24 SF3B1 muta­tions have a pos­i­ deemed benign or likely benign.29 Importantly, according to the
tive prog­nos­tic value on OS and risk of pro­gres­sion. This favor­ Amer­i­can College of Medical Genetics and Genomics guide­lines, a
able effect on prog­no­sis is inde­pen­dent of IPSS-R for very low var­i­ant of uncer­tain sig­nif­i­cance should not influ­ence clin­i­cal deci­
and low IPSS-R categories but is not retained for high- or very sion mak­ing. NGS may also iden­tify a germline var­i­ant (eg, biallelic
high-risk groups.24 Comutations medi­ate this effect. Mutations in muta­tion in CEBPA, RUNX1, or other genes often evidenced by a
epi­ge­netic reg­u­la­tors, includ­ing TET2, DNMT3A, and ASXL1, do VAF of around 50%), which may have imme­di­ate prac­ti­cal impli­
not appear to affect sur­vival of patients with an SF3B1 muta­tion, ca­tions for donor selec­tion if the patient is a trans­plant can­di­date,
whereas RUNX1 and EZH2 muta­tions have a sig­nif­i­cant asso­ci­a­ and influ­ence fam­ily mem­bers.30 Referral to a genetic coun­selor is
tion with increased risk of dis­ease pro­gres­sion and decreased rea­son­able when germline muta­tions are a con­sid­er­ation.
OS inde­pen­dent of IPSS-R.24,25
Conclusion
The pres­ence or absence of a TP53 muta­tion In sum­mary, we rec­om­mend the use of NGS to pro­vide cli­ni­cians
The adverse prog­nos­tic effect of TP53 muta­tions in MDS has with addi­tional infor­ma­tion for risk strat­i­fi­ca­tion for patients with
long been rec­og­nized. In patients with an iso­lated dele­tion in MDS. Conclusive rec­om­men­da­tions of how to inte­grate cur­rent
5q, 20% of patients har­bor a TP53 muta­tion that con­fers a higher risk prog­nos­ti­ca­tion sys­tems and molec­u­lar pro­files can­not be
risk to AML pro­gres­sion, worse OS, and early pro­gres­sion in lena­ made based on the var­i­abil­ity in data qual­ity and het­er­og­e­nous ef­
lidomide-treated patients.26 TP53 muta­tion sta­tus also appears fects of muta­tions on patients. Treatment deci­sions must be made
crit­i­cal to medi­ate the adverse effect of a com­plex kar­yo­type. on a case-by-case basis with ther­apy aligned with patient pref­er­
Patients with a com­plex kar­yo­type with­out a TP53 muta­tion ences. Increasingly, we are see­ing the dem­on­stra­tion of the asso­
have sig­nif­i­cantly bet­ter sur­vival than would be suggested from ci­a­tion between molec­u­lar pro­files and clin­i­cal fea­tures that can
prog­nos­ti­ca­tion based on karyotyping alone.27 In both cases, the pre­dict prog­no­sis inde­pen­dent of clin­i­cal scor­ing sys­tems. The
pres­ence or absence of a TP53 muta­tion could sub­stan­tially al­ devel­op­ment of val­i­dated prog­nos­tic mod­els, induc­ing molec­u­lar
ter clin­i­cal man­age­ment. Specifically, Prakash
we wouldSingh Shekhawat
highly con­
sider data, will bring great value to the field. Hopefully, we are not too

Routine molec­u­lar pro­fil­ing in MDS for risk  |  437

far away from this being a real­ity. The International Working Group
for the Prognosis of MDS and the MDS Foundation are progressing
in the devel­op­ment of an inte­grated clin­i­cal-molec­u­lar data­base
that should sup­port the devel­op­ment of “IPSSR-Molecular.”
Recommendations
The rec­om­men­da­tions are labeled as either “strong” or “con­di­
tional” according to the GRADE approach.
(1) For all­patients suspected of hav­ing or con­firmed to have
MDS, the authors rec­om­mend molec­u­lar pro­fil­ing with NGS to
iden­tify rel­e­vant muta­tions (strong rec­om­men­da­tion based
on mod­er­ate cer­tainty in the evi­dence of effects).
(2) For patientpdfs found to have canon­i­cal muta­tions in genes
known to con­trib­ute to sur­vival out­comes (eg, SF3B1, ASXL1,
SRSF2, ETV6, RUNX1, and TP53), the authors sug­gest cli­ni­
cians incor­po­rate the poten­tial addi­tional risk into existing
prog­nos­tic scor­ing sys­tems (con­di­tional rec­om­men­da­tion
based on low cer­tainty in the evi­dence of effects).
(3) For patients who are eli­gi­ble based on molec­u­lar pro­files, the au­
thors sug­gest cli­ni­cians should offer enroll­ment in a clin­i­cal trial to
patients (con­di­tional rec­om­men­da­tion based on low cer­tainty in
the evi­dence of effects).
(4) For all­patients with MDS, the authors rec­om­mend assess­
ing patient pref­er­ences to align treat­ment rec­om­men­da­
tions with patients’ will­ing­ness to tol­er­ate risks for ben­e­fits
(strong rec­om­men­da­tion based on mod­er­ate cer­tainty in
the evi­dence of effects).
Conflict-of-inter­est dis­clo­sure
Dan­iel R. Richardson: no conflicts to disclose.
Amy E. DeZern has received con­sul­tancy fees from BMS, Abbvie,
Novartis, and Taiho.
Off-label drug use
Dan­iel R. Richardson: nothing to disclose.
Amy E. DeZern: nothing to disclose.
Correspondence
Amy E. DeZern, Department of Oncology, Sidney Kimmel Cancer
Center at Johns Hopkins, 1650 Orleans Street, CRBI Room 3M87
Baltimore, MD 21287-0013; e-mail: adezern1@jhmi​­.edu.
References
1. Papaemmanuil E, Gerstung M, Malcovati L, et al; Chronic Myeloid Disorders
Working Group of the International Cancer Genome Consortium. Clinical
and bio­log­i­cal impli­ca­tions of driver muta­tions in myelodysplastic syn­
dromes. Blood. 2013;122(22):3616-3627, quiz 3699.
2. Ogawa S. Genetics of MDS. Blood. 2019;133(10):1049-1059.
3. Haferlach T, Nagata Y, Grossmann V, et al. Landscape of genetic lesions in
944 patients with myelodysplastic syn­dromes. Leukemia. 2014;28(2):241-
247.
4. Bejar R. Myelodysplastic syn­dromes diag­no­sis: what is the role of molec­u­
lar test­ing? Curr Hematol Malig Rep. 2015;10(3):282-291.
5. Nazha A, Sekeres MA, Gore SD, Zeidan AM. Molecular test­ing in myelodys­
plastic syn­dromes for the prac­tic­ing oncol­o­gist: will the prog­ress ful­fill the
prom­ise? Oncologist. 2015;20(9):1069-1076.
6. Cazzola M, Della Porta MG, Malcovati L. The genetic basis of myelodyspla­
sia and its clin­i­cal rel­e­vance. Blood. 2013;122(25):4021-4034.
7. Bejar R, Stevenson K, Abdel-Wahab O, et al. Clinical effect of point muta­
tions in myelodysplastic syn­ dromes. N Engl J Med. 2011;364(26):2496-
2506.
438  |  Hematology 2021  |  ASH Education Program
8. Tefferi A, Gangat N, Mudireddy M, et al. Mayo alli­ance prog­nos­tic model
for myelodysplastic syn­dromes: inte­gra­tion of genetic and clin­i­cal infor­ma­
tion. Mayo Clin Proc. 2018;93(10):1363-1374.
9. Sallman DA, Komrokji R, Vaupel C, et al. Impact of TP53 muta­tion var­i­ant
allele fre­ quency on phe­ no­type and out­ comes in myelodysplastic syn­
dromes. Leukemia. 2016;30(3):666-673.
10. Hirsch CM, Nazha A, Kneen K, et al. Consequences of mutant TET2 on clon­
ality and subclonal hier­ar­chy. Leukemia. 2018;32(8):1751-1761.
11. Pellagatti A, Roy S, Di Genua C, et al. Targeted resequencing anal­y­sis of
31 genes com­monly mutated in mye­loid dis­or­ders in serial sam­ples from
myelodysplastic syn­drome patients show­ing dis­ease pro­gres­sion. Leuke-
mia. 2016;30(1):247-250.
12. Bernard E, Nannya Y, Hasserjian RP, et al. Implications of TP53 alle­lic state
for genome sta­bil­ity, clin­i­cal pre­sen­ta­tion and out­comes in myelodysplastic
syn­dromes. Nat Med. 2020;26(10):1549-1556.
13. Montalban-Bravo G, Kanagal-Shamanna R, Benton CB, et al. Genomic con­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/435/1852132/435richardson.pdf by guest on 13 December 2021
text and TP53 allele fre­quency define clin­i­cal out­comes in TP53-mutated
myelodysplastic syn­dromes. Blood Adv. 2020;4(3):482-495.
14. Malcovati L, Gallì A, Travaglino E, et  al. Clinical sig­nif­i­cance of somatic
muta­tion in unex­plained blood cytopenia. Blood. 2017;129(25):3371-3378.
15. Galli A, Todisco G, Catamo E, et al. Relationship between clone met­rics and
clin­i­cal out­come in clonal cytopenia [published online 13 July 2021]. Blood.
16. Bersanelli M, Travaglino E, Meggendorfer M, et al. Classification and per­
son­al­ized prog­nos­tic assess­ment on the basis of clin­ic ­ al and geno­mic fea­
tures in myelodysplastic syn­dromes. J Clin Oncol. 2021;39(11):1223-1233.
17. Bejar R, Stevenson KE, Caughey BA, et al. Validation of a prog­nos­tic model
and the impact of muta­tions in patients with lower-risk myelodysplastic
syn­dromes. J Clin Oncol. 2012;30(27):3376-3382.
18. Forghieri F, Paolini A, Morselli M, et al. NPM1 muta­tions may reveal acute mye­
loid leu­ke­mia in cases oth­er­wise mor­pho­log­i­cally diag­nosed as myelodys­
plastic syn­dromes or myelodysplastic/mye­lo­pro­lif­er­a­tive neo­plasms. Leuk
Lymphoma. 2015;56(11):3222-3226.
19. Montalban-Bravo G, Kanagal-Shamanna R, Sasaki K, et al. NPM1 muta­tions
define a spe­cific sub­group of MDS and MDS/MPN patients with favor­able
out­comes with inten­sive che­mo­ther­apy. Blood Adv. 2019;3(6):922-933.
20. Bains A, Luthra R, Medeiros LJ, Zuo Z. FLT3 and NPM1 muta­tions in myelo­
dysplastic syn­dromes: fre­quency and poten­tial value for predicting pro­
gres­sion to acute mye­loid leu­ke­mia. Am J Clin Pathol. 2011;135(1):62-69.
21. Daver N, Strati P, Jabbour E, et  al. FLT3 muta­ tions in myelodysplas­
tic syn­ drome and chronic myelomonocytic leu­ ke­mia. Am J Hematol.
2013;88(1):56-59.
22. Xu F, Han R, Zhang J, et al. The role of FLT3-ITD muta­tion on de novo MDS in
Chi­nese pop­u­la­tion. Clin Lymphoma Myeloma Leuk. 2019;19(2):e107-e115.
23. Badar T, Patel KP, Thompson PA, et al. Detectable FLT3-ITD or RAS muta­tion
at the time of trans­for­ma­tion from MDS to AML pre­dicts for very poor out­
comes. Leuk Res. 2015;39(12):1367-1374.
24. Malcovati L, Stevenson K, Papaemmanuil E, et al. SF3B1-mutant MDS as a
dis­tinct dis­ease sub­type: a pro­posal from the International Working Group
for the Prognosis of MDS. Blood. 2020;136(2):157-170.
25. Malcovati L, Karimi M, Papaemmanuil E, et al. SF3B1 muta­tion identifies a
dis­tinct sub­set of myelodysplastic syn­drome with ring sideroblasts. Blood.
2015;126(2):233-241.
26. Jädersten M, Saft L, Smith A, et al. TP53 muta­tions in low-risk myelodys­
plastic syn­dromes with del(5q) pre­dict dis­ease pro­gres­sion. J Clin Oncol.
2011;29(15):1971-1979.
27. Haase D, Stevenson KE, Neuberg D, et al; International Working Group for
MDS Molecular Prognostic Committee. TP53 muta­tion sta­tus divi­des myel­
odysplastic syn­dromes with com­plex kar­yo­types into dis­tinct prog­nos­tic
sub­groups. Leukemia. 2019;33(7):1747-1758.
28. Sallman DA, DeZern AE, Garcia-Manero G, et al. Eprenetapopt (APR-246)
and azacitidine in TP53-mutant myelodysplastic syn­dromes. J Clin Oncol.
2021;39(14):1584-1594.
29. Li MM, Datto M, Duncavage EJ, et al. Standards and guide­lines for the inter­
pre­ta­tion and reporting of sequence var­i­ants in can­cer: a joint con­sen­sus
rec­om­men­da­tion of the Association for Molecular Pathology, Amer­i­can
Society of Clinical Oncology, and College of Amer­i­can Pathologists. J Mol
Diagn. 2017;19(1):4-23.
30. Kennedy AL, Shimamura A. Genetic pre­dis­po­si­tion to MDS: clin­i­cal fea­tures
and clonal evo­lu­tion. Blood. 2019;133(10):1071-1085.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000320
 MDS: BEYOND A ONE - SIZE - FITS -ALL APPROACH

Oral hypomethylating agents: beyond

convenience in MDS

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/439/1852126/439griffiths.pdf by guest on 13 December 2021

Elizabeth A. Griffiths
Roswell Park Comprehensive Cancer Center, Buffalo, NY

Oral hypomethylating agents (HMAs) represent a substantial potential boon for patients with myelodysplastic syndrome
(MDS) who have previously required between 5 and 7 visits per month to an infusion clinic to receive therapy. For patients
who respond to treatment, ongoing monthly maintenance visits represent a considerable burden to quality of life, and
for those who are early in therapy, these sequential visits may tax transportation and financial resources that would be
optimally distributed over the treatment cycle to facilitate transfusion support. The availability of oral HMAs may support
the optimal application of these agents by contributing to adherence and lessening the burden of therapy, potentially
encouraging patients to stay on longer-term treatment. Distinct pharmacokinetic profiles for the recently approved oral
HMAs (oral azacitidine and decitabine-cedazuridine) result in differential toxicity profiles and have prompted their clinical
trial development in lower- and higher-risk MDS, respectively.

LEARNING OBJECTIVES
• Describe the challenges of effective therapy with HMAs for patients with MDS and the barriers to this therapy
• Understand the PK profiles for HMAs given parenterally and contrast these with the PK profiles following oral
administration of unmodified oral azacitidine and decitabine/cedazuridine
• Recognize how the PK for oral HMAs may differ from an established parenteral regimen and anticipate the chal-
lenges associated with transitioning from one regimen to another

with parenteral azacitidine 75 mg/m2 given subcutane-

CLINICAL CASE 1
A 78-year-old woman presented to our clinic after routine
blood work demonstrated thrombocytopenia and neutro-
penia. The patient was otherwise in reasonable health for
her age with comorbid medically managed hypertension.
She was widowed but lived independently, with minimal
support from her local adult children. A complete blood
count (CBC) at the initial evaluation demonstrated a white
blood cell (WBC) count of 1.4 × 109/µL, with an abso-
lute neutrophil count of 0.5 × 109/µL, hemoglobin (Hg)
of 11 g/µL, and platelet count of 25 k/µL. She was dem-
onstrated to be nutrient replete for B12, folate, iron, and
copper. A bone marrow (BM) aspiration and biopsy were
performed, showing refractory anemia with excess blast-2
(15% blasts); cytogenetics were normal. Next-generation
sequencing (NGS) revealed mutations in IDH2 and SRSF2.
A diagnosis of higher-risk myelodysplastic syndrome
(MDS) was rendered (International Prognostic Scoring
System [IPSS]-Revised [R], 5.5 points [high risk]; IPSS 2.0
points [intermediate-2]), and the patient started therapy
Prakash Singh Shekhawat
ously (SQ ) 7 days of a 28-day schedule. The initial treat-
ment was characterized by the development of red blood
cell (RBC) and platelet transfusion dependence in cycles
1 and 2. She presented prior to cycle 3 of therapy, and the
CBC showed a WBC count of 1.0 × 109/µL with absolute
neutrophil count 0.2 × 109/µL, an Hg of 8 g/µL, and plate-
lets, 105 k/µL. She came to the clinic accompanied by
her children. She stated that she was no longer willing to
continue therapy. She reported intolerable loss of quality
of life and independence due to the daily infusion clinic
visits in the first week of each cycle and the burden of
transfusion dependence. She was adamant that she would
take no further chemotherapy despite an early response
to treatment that suggested she was likely to derive a sur-
vival benefit from continued treatment (early hematologic
improvement [HI] of the platelet lineage). She requested
enrollment in home hospice care and passed away 1 month
after therapy discontinuation.Today, the availability of oral
agents might have convinced this patient to continue
treatment, optimizing her survival and improving her
Oral hypomethylating agents for MDS | 439

Figure 1. Chemical structure of cytidine, azacitidine analogues, and the breakdown products of CDA.
qual­ity of life. As it stood, this patient accrued all­the tox­ic­ity
from the use of this agent, includ­ing ini­tial loss of qual­ity of life,
with­out real­iz­ing the lon­ger-term poten­tial ben­e­fits of trans­fu­
sion inde­pen­dence (TI) and sur­vival, which she might have been
expected to achieve given her early evi­dence of response.
Introduction
Hypomethylating agents (HMAs) have been the main­ stay of
treat­ment for patients with higher-risk MDS since the ini­tial reg­u­
la­tory approval of monthly treat­ments with par­en­teral azacitidine
and decitabine in the early 2000s.1 The shared active metab­ol­ite
of both agents (~15% of the admin­is­tered azacitidine dose, con-
verted by the action of ribo­nu­cle­o­tide reduc­tase) is decitabine
tri­phos­phate, an ana­logue of the nucle­o­side cyti­dine (Figure 1).
Although 85% of par­en­ter­ally admin­is­tered azacitidine is incor­
po­rated into RNA, interrupting pro­tein syn­the­sis and induc­ing
apo­pto­sis, the clin­ic
­ al impact of this incor­po­ra­tion on its activ­
ity remains an area of active inves­ti­ga­tion.1 When decitabine tri­
phos­phate is incor­po­rated into DNA whose par­ent strand bears
a methyl group in the con­text of a cyto­sine-phospho-gua­nine
sequence, DNA methyltransferase enzymes (par­tic­u­larly DNMT-1)
440  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/439/1852126/439griffiths.pdf by guest on 13 December 2021
bind it irre­vers­ibly, cre­at­ing an adduct that results in both DNA
dam­age and the deple­tion of DNMT enzymes.2 With repeated
cycles of DNA rep­li­ca­tion, loss of the meth­yl­at­ion enzymes re-
sults in the pas­sive demethylation of DNA at mul­ti­ple loci. At low
doses that do not cause over­whelm­ing cyto­tox­ic­ity, these drugs
induce tran­sient gene-spe­cific and global hypomethylation that
are hypoth­e­sized to medi­ate their response. Historically, activ­ity
has been pos­tu­lated to result from the reexpression of epi­ge­
net­i­cally silenced tumor sup­pres­sor genes, but more recently,
reexpression of endog­e­nous ret­ro­vi­ral ele­ments with enhanced
immune rec­og­ni­tion of MDS pro­gen­i­tors has been hypoth­e­sized
as another puta­tive mech­a­nism.3 Despite ongo­ing con­tro­versy
over the defin­i­tive expla­na­tion for ther­a­peu­tic effi­cacy, opti­mal
patient out­comes from HMA ther­apy seem to require long-term
treat­ment on a reg­u­lar and unin­ter­rupted sched­ule.4-7
Dose, sched­ule, and treat­ment con­tin­u­a­tion are crit­i­cal
to opti­mal HMA response
When given on a monthly basis for repeated cycles, these agents
improve cytopenias, lower BM blasts, improve qual­ity of life, de-
crease trans­for­ma­tion to acute mye­loid leu­ke­mia (AML), and im-
prove sur­vival for select patients.8-10 Despite years of study, no
effec­tive pre­treat­ment pre­dic­tors of response are clin­i­cally appli­
ca­ble.11 About half of HMA-treated patients will respond, most
with improve­ments in cytopenias (HI of the affected lin­e­age)
and a minor­ity (per­haps 15%) with com­plete response (CR), but
responses can take sev­eral months to develop (National Com-
prehensive Cancer Center guide­lines rec­om­mend a min­i­mum of
6 cycles of treat­ment before response assess­ment) and are gen­
er­ally of lim­ited dura­bil­ity.12,13 Critically, hema­to­log­i­cal improve­
ment (HI) in any lin­e­age is asso­ci­ated with sur­vival ben­e­fit for
patients who con­tinue ther­apy; with sustained treat­ment some
indi­vid­u­als will have improved blood counts across more than
one lin­e­age, even converting to CR.7,13,14 Unfortunately, when pa-
tients discontinue ther­apy, rapid pro­gres­sion and relapse with
poor sur­vival are likely.15-21 Good responses and improved sur­vival
are often accom­pa­nied by recov­ery of the plate­let count in the
first 1 to 3 cycles of treat­ment, an early bio­marker for those who
are likely benefit­ing from treat­ment.22,23
As for the patient in our first case, early cycles of HMA ther­apy
are often char­ac­ter­ized by the wors­en­ing of cytopenias and the
devel­op­ment of trans­fu­sion depen­dence. This has the poten­tial
to cre­ate a sub­stan­tial bur­den for patients and fam­i­lies, espe­cially
since each monthly cycle has his­tor­i­cally required 5 to 7 con­sec­
u­tive daily doses of intra­ve­nous (IV; decitabine, azacitidine) or SQ
(azacitidine) treat­ment at a des­ig­nated infu­sion cen­ter. Given the
early wors­en­ing of cytopenias, opti­mal sup­port­ive care requires
weekly or even twice-weekly vis­its dur­ing the sub­se­quent 3 weeks
to assess trans­fu­sion needs as well as the pro­phy­lac­tic use of anti­
mi­cro­bi­als.24 The devel­op­ment of RBC and plate­let trans­fu­sion
depen­dence and neutropenia early in the course of treat­ment can
trig­ger cli­ni­cians to inap­pro­pri­ately reduce doses, delay, or even
discontinue ther­apy; deci­sions known to com­pro­mise effi­cacy.25,26
Such deci­sions may explain the dif­fer­ence between the sur­vival
ben­e­fit documented in con­trolled pop­u­la­tions of patients treated
with HMAs in the con­text of clin­i­cal tri­als com­pared to out­comes
in unse­lected pop­u­la­tions of patients, where improved sur­vival
rates have been less con­sis­tent.10,21,27,28
A set of recent ret­ro­spec­tive ana­ly­ses using the sur­veil­lance,
epi­de­mi­­ol­ogy, and end results Medi­care-linked data­base in a large
cohort of 644 patients with higher-risk MDS eval­u­ated between
2011 and 2015 sought to inves­ti­gate the way in which patients
treated in a real-world set­ting rou­tinely receive ther­apy. This study
dem­on­strated that almost 30% of patients discontinued HMAs
before com­ plet­ing 4 cycles—a major­ ity after com­ plet­ing only
1 cycle of treat­ment.29 A more detailed anal­y­sis of this same cohort
reveals that an addi­tional 20% of patients had early dose delays of
>90 days between cycles, which likely obvi­ates response.30 Over-
all, among the approx­i­ma­tely 50% of patients with higher-risk
MDS, those who did not receive their HMA on sched­ule for at least
4 cycles had sub­stan­tially higher rates of health care uti­li­za­tion
char­ac­ter­ized by increased emer­gency room vis­its, higher rates
of hos­pi­tal­i­za­tion, and more admis­sions to skilled nurs­ing facil­i­ties
and hos­pice. Suboptimally treated patients in this cohort were
more likely to be older and unpartnered.
Real-world stud­ies fur­ther dem­on­strate the rel­a­tively lim­ited
ini­ti­a­tion of HMA treat­ment for high-risk patients, with only 12%
to 30% of pre­sum­ably eli­gi­ble patients receiv­ing ther­apy at all­,
depending upon the report.20,27,30 These data high­light the fact
that despite the avail­abil­ity of HMAs as pri­mary ther­apy for MDS
for more than 15 years, many patients do not receive any ther­apy
at all­, and among those who do, early dis­con­tin­u­a­tion or inad­e­
Prakash Singh Shekhawat
quate dose inten­sity dur­ing early cycles prob­a­bly com­pro­mises
response.30 Such treat­ment deci­sions are likely to have an adverse
impact on qual­ity of life and com­pro­mise sur­vival for those MDS
patients, who will accrue all­the tox­ic­ity from HMA ther­apy and
none of the ben­e­fits (HI, CR, or sur­vival) of sustained ther­apy.24
The impor­tance of ade­quate sup­port and ther­apy per­sis­tence for
patients with higher-risk MDS can­not, there­fore, be overstated.
Oral agents, which might limit the need for treat­ment-related
infu­sion clinic vis­its and allow patients a sched­ule based solely on
trans­fu­sion needs, have the poten­tial to decrease rates of early
dis­con­tin­u­a­tion due to treat­ment bur­den and thereby result in
improved qual­ity of life, effi­cacy, and sur­vival for patients with
MDS. Furthermore, although most responses to HMAs have been
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/439/1852126/439griffiths.pdf by guest on 13 December 2021
shown to occur within 4 to 6 cycles of treat­ment, con­tin­u­a­tion
of ther­apy beyond the first HI results in fur­ther improve­ment
in response cat­e­gory for about half of those on treat­ment.7,18
Indeed, the best documented response seems to man­i­fest at
least 2 to 3 cycles after the first documented HI response, with
opti­mal responses in 1 study manifesting as late as cycle 12 for a
major­ity of patients.13,18
Given the ongo­ ing bur­
den of treat­ ment, con­ tin­
ued ther­
apy on sched­ule must be empha­sized to patients and should
not be underrecognized. Early on, trans­fu­sion depen­dence may
increase patient will­ing­ness to come to the office for treat­ment,
but in the long term, these vis­its can become oner­ous. Ongo-
ing stud­ies are test­ing whether patients on an established par­
en­teral HMA reg­i­men can be switched to a lower-inten­sity oral
reg­i­men to max­i­mize adher­ence with­out com­pro­mis­ing effi­cacy
(eg, NCT04806906). Such changes to our ther­a­peu­tic approach
may also improve the real-world out­comes asso­ci­ated with HMA
treat­ment when com­pared with those seen in clin­i­cal tri­als. If
pre­ma­ture dis­con­tin­u­a­tion in responding patients is driven by
doc­tors and patients unwill­ing to con­tinue ther­apy, at least in
part due to the bur­den of infu­sion clinic appoint­ments, I believe
oral ther­a­pies may help us max­i­mize HMA ben­e­fit for patients.
CLINICAL CASE 2
A 69-year-old man with a med­ i­
cal his­
tory nota­ble for obe­
sity,
dia­be­tes, and bilat­eral knee oste­o­ar­thri­tis was referred to our
prac­tice after a pre­op­er­a­tive CBC performed for a planned knee
replace­ment sur­gery dem­on­strated throm­bo­cy­to­pe­nia. The CBC
showed a nor­mal WBC count, plate­lets of 40 000/µL, and ane­mia
with an Hb of 10 g; the mean cor­pus­cu­lar vol­ume was 106 fL. The
remain­der of the CBC and a com­pre­hen­sive pro­file were nor­mal,
and nutri­tional workup dem­on­strated a suf­fi­ciency of vita­min B12,
folate, and iron; fer­ri­tin was 600 (min­i­mally ele­vated), and the
serum erythropoietin level was 580 IU (ele­vated). The periph­eral
blood smear was nota­ble for pelgeroid neu­tro­phils, throm­bo­cy­
to­pe­nia, and RBC anisocytosis. A BM aspi­rate and biopsy were
hypercellular and ery­throid pre­dom­i­nant with 6% blasts and trilin-
eage dyspoeisis. Cytogenetics were 47 XY, + 8 in 21 meta­phases.
NGS was nota­ble for muta­tions in TET2 and ASXL1. A diag­no­sis
of inter­me­di­ate-risk MDS (IPSS score, inter­me­di­ate-1; IPSS-R,
5 points [high risk]) was ren­dered. After dis­cus­sion of prog­no­
sis and options, he elected to enroll in a phase 1b phar­ma­co­
ki­
netic (PK) study of oral ­ decitabine-cedazuridine fixed-dose
tab­let 35 mg/100 mg. Initial cycles were char­ac­ter­ized by grade
Oral hypomethylating agents for MDS  |  441
Table 1. PK char­ac­ter­is­tics of the HMAs
Bioavailability of sin­gle oral
Agent dose (% of par­en­teral)
Azacitidine IV34,40 100%
Azacitidine SQ34,40 89%
CC-486 40,52
11%
Decitabine IV35 100%
Decitabine PO 41
3.9%-14.1%
C-DEC43 60% (55-65) D1; 106% (98-114) D5
4 throm­bo­cy­to­pe­nia and neutropenia and plate­let trans­fu­sion
depen­dence. After cycle 5 the patient had recov­ery of the plate­let
count. A BM biopsy after cycle 4 dem­on­strated a blast per­cent­age
of 4% with ongo­ing dys­pla­sia. Referral for trans­plant was made,
and a fully matched sib­ling donor was iden­ti­fied. The patient cur­
rently remains on pro­to­col cycle 12 of oral decitabine-cedazuridine,
given 4 days of a 28-day cycle (the dose was reduced after cycle
8 for neutropenia last­ing more than 2 weeks and BM show­ing
ongo­ing ther­a­peu­tic response). He has declined to pro­ceed with
trans­plant due to con­cerns about trans­plant-related mor­tal­ity. He
enjoys an excel­lent qual­ity of life and remains trans­fu­sion inde­
pen­dent (TI) and nonneutropenic with a plate­ let count above
100 k/µL. These counts allow him to use non­ste­roi­dal agents for
man­age­ment of his knee pain with­out bleed­ing risk.
PKs of par­en­teral vs oral HMAs
Parenteral daily HMA treat­ment with azacitidine and decitabine
results in blood drug lev­els that peak between 15 and 30 min­
utes after admin­ is­tra­
tion and an elim­ i­na­tion half-life between
1 to 2 hours for azacitidine and 35 to 40 min­utes for decitabine
(Table 1).31-35 The short half-life of par­en­ter­ally admin­is­tered HMAs
is medi­ated largely by the ubiq­ui­tous expres­sion of the enzyme
cyti­dine deam­i­nase (CDA), which is highly expressed in a vari­ety
of human tis­sues, includ­ing the liver, the BM, and the gas­tro­in­
tes­ti­nal (GI) tract. CDA rap­idly degrades these agents into the
inac­tive metab­o­lites of uri­dine (Figure 1).32,36
The early devel­op­ment of oral HMAs was char­ac­ter­ized by
the rec­og­ni­tion that when given alone oral admin­is­tra­tion results
in mark­ edly lower peak plasma con­ cen­ tra­
tions with a wide
range of bio­avail­abil­ity when com­pared with IV or SQ dos­ing
strat­e­gies.33,37-41 Table 1 pro­vi­des a com­par­i­son of half-life, sin­gle-
dose bio­avail­abil­ity (based upon area under the con­cen­tra­tion
time curve), time to max­im ­ al blood drug level (Tmax), and max­
i­mal blood con­cen­tra­tion achieved after the admin­is­tered dose
(Cmax) for each drug depending upon the route of admin­is­tra­
tion. Differences in bio­avail­abil­ity are medi­ated by wide inter-
and intraindividual ranges of CDA expres­sion, with higher lev­els
in men, in those with spe­cific sin­gle-nucle­o­tide poly­mor­phisms,
and in the con­text of inflam­ma­tory driv­ers.32,36,42
Oral decitabine/cedazuridine
Based upon the dem­on­stra­tion of phar­ma­co­log­i­cal equiv­a­
lence to IV decitabine, in July 2020 the US reg­u­la­tory author­i­
ties approved the first oral HMA, decitaine-cedazuridine (C-DEC;
442  |  Hematology 2021  |  ASH Education Program
Cmax in ng/mL (%
T1/2 Tmax (range) coef­fi­cient of var­i­a­tion)
4 h 0.5 h Similar to SQ
4 h 0.5 h (0.2-1.1) 750 (54%)
0.5 h 1 h (0.47-2) 145 (64%)
0.5 h 1 h 147 (49%)
0.36-0.93 h 0.5 h —
1.5 h 1 h (0.3-3.0) 145 (55%)
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/439/1852126/439griffiths.pdf by guest on 13 December 2021
35/100 mg), for patients with higher-risk MDS (inter­me­di­ate/high
risk by IPSS).43 In con­trast with another recently approved agent,
unmod­i­fied oral azacitidine (CC-486), which will be discussed
sub­se­quently, C-DEC has been shown in a ran­dom­ized phase 3
study to reli­ably reca­pit­u­late the blood drug lev­els over time
(cumu­la­tive area under the con­cen­tra­tion-time curve; AUC∞; Fig-
ure 2) resulting from daily treat­ment for 5 days with IV decit-
abine, and the 2 can there­fore be treated inter­change­ably for
the pur­poses of clin­i­cal man­age­ment.44,45
As expected, based upon the ubiq­ui­tous expres­sion of CDA
in the liver and GI tract, there is sub­stan­tial first-pass metab­o­
lism of oral decitabine. Single-agent oral decitabine was shown
in a phase 1 study to result in low bio­avail­abil­ity with sub­stan­
tial var­i­a­tion across indi­vid­u­als when com­pared with IV decit-
abine.41 To over­come the var­i­able PKs resulting from dif­fer­en­tial
CDA lev­els within and across indi­vid­u­als, a novel inhib­i­tor of CDA
(CDAi) given the des­ig­na­tion E7727 (sub­se­quently cedazuridine)
was devel­oped and exten­sively tested in ani­mals.46 This agent
dem­on­strated excel­lent bio­avail­abil­ity and a wide safety mar­gin
in non­hu­man pri­ma­tes.46,47 In a series of early-phase stud­ies, the
com­bi­na­tion of cedazuridine with oral decitabine (devel­oped
ini­tially as ASTX 727 and sub­se­quently des­ig­nated decitabine-
cedazuridine) was tested to develop a com­bi­na­tion pill designed
to reca­pit­u­late the PK and phar­ma­co­dy­namic (PD) admin­is­tra­tion
of IV decitabine dosed at 20 mg/m2 for 5 days.47 Serial blood col­
lec­tion was performed, and changes in global meth­yl­at­ ion using
the repet­i­tive DNA sequence long inter­spersed nuclear ele­ment
1 (a sur­ro­gate for global DNA meth­yl­a­tion) was used as a tool for
com­par­ing phar­ma­co­dy­namic (PD) effi­cacy.48 Given the sub­stan­
tial intra- and inter­in­di­vid­ual var­i­abil­ity in CDA lev­els, the devel­
op­ment pro­gram for this agent used each patient as their own
con­trol and com­pared the PK/PD for IV decitabine against the
PK/PD for the oral com­bi­na­tion.47-49 Early stud­ies used PK/PD
read­outs to titrate the oral doses of E7727 and decitabine in
order to cre­ate a near iden­tity with IV decitabine admin­is­tra­tion
(Figure 2a). Ultimately, the 2 were com­bined into a fixed-dose
com­bi­na­tion tab­let containing 35 mg of decitabine and 100 mg
of cedazuridine.43-45
The phase 3 study of C-DEC, upon which approval was based,
ran­dom­ized and treated 133 patients in a cross­over design to
receive either sequence A: C-DEC orally for 5 out of 28 days
in cycle 1 followed by IV decitabine 20 mg/m2 × 5 out of 28 days in
cycle 2 and sub­se­quently C-DEC for cycles 3 onward or sequence
B: IV decitabine 20 mg/m2   ×   5 out of 28 days in cycle 1 followed
by C-DEC orally daily for 5/28 days in cycle 2 and sub­se­quently
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/439/1852126/439griffiths.pdf by guest on 13 December 2021
Figure 2. Concentration time curves for (a) IV decitabine vs oral C-DEC and (b) SQ azacitidine vs CC-486.40,46 Figure 2a was re-
produced from Future Oncol. 2021;17(16):2077-2087 and was modified only by renumbering for the purpose of this article. Figure
2b was reproduced from Leukemia 2016;30(4):889-896 and was also modified only by renumbering. Both works are licensed under
the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. http:​/​/creativecommons​.org​/licenses​
/by​-nc​-nd​/4​.0​/

C-DEC for cycles 3 onward; cycles were to be repeated every
28 days (Figure 3). The pri­mary end point for this phase 3 study was
PK for cumu­la­tive AUC equiv­a­lence between IV and oral dos­ing of
decitabine. Eligible patients had MDS by French-Amer­i­can-­­Brit­ish
clas­si­fi­ca­tion, includ­ing chronic myelomonocytic leu­ke­mia, or
MDS deemed inter­me­di­ate 1, 2, or high risk by the IPSS. The clin­
i­cal response to ther­apy was assessed according to ­International
Working Group 2006 cri­te­ria by an inde­pen­dent review com­mit­
tee. Enrolled patients were of median age 71 years (range, 44-88),
65% were male, 12% had chronic myelomonocytic leu­ke­mia, the
remain­der had MDS, and 42% had >5% mar­row blasts. Transfu-
sion depen­dence for either RBCs or plate­lets was pres­ent in 43%.
After a median fol­low-up of 12.6 months, com­plete remis­sion was
reported in 28 (21%) and HI in 30 (23%) patients for an over­all
response rate of 43%. An addi­tional 23 patients dem­on­strated
mar­row CR with­out HI, although blast clear­ance with­out blood

Figure 3. Summary comparison of the 2 completed phase 3 studies of C-DEC and CC-486 in MDS. Data abstracted from ­Garcia-Manero
et al40,49,55 and Savona et al.45 AZA, azacitidine; DAC, decitabine.
Prakash Singh Shekhawat
Oral hypomethylating agents for MDS  |  443
count improve­ ment is an end point of unclear clin­ ic
­al sig­
nif­i­
cance. Adverse events (AEs) asso­ci­ated with C-DEC were con­
sis­tent with the famil­iar toxicities from early cycles of par­en­teral
HMA ther­apy, nota­bly cytopenias (≥grade 3: neutropenia, 52%;
throm­bo­cy­to­pe­nia, 50%; ane­mia, 40%; leu­ko­pe­nia, 21%) and
infec­tions (febrile neutropenia, 26%; pneu­mo­nia, 12%; sep­sis, 7%);
in con­trast to CC-486, GI toxicities were not prominent.
While phase 3 data have been reported only in abstract form,
they largely reca­pit­u­late the recently published results from a
phase 2 mul­ti­cen­ter ran­dom­ized cross­over study of sim­i­lar design
that enrolled and treated 80 patients. In this phase 2 study,
responses (CR + HI) were observed in 44% of patients, with a
major­ity of first responses seen by cycle 3 and best responses by
cycles 5 or beyond. Among base­line trans­fu­sion-depen­dent (TD)
patients, about 50% became TI. Overall responses and tox­ic­ity
from C-DEC appear com­men­su­rate with those from par­en­ter­ally
admin­is­tered HMAs, and this is not sur­pris­ing given the iden­ti­cal
blood drug lev­els (AUC) achieved with this drug. Many patients,
like our patient in Case 2, will have increased trans­fu­sion needs
dur­ing early cycles of ther­apy. The same aggres­ sive sup­port-
­ive care, includ­ing pro­phy­lac­tic anti­mi­cro­bi­als (a mold-active
anti­fun­ gal, an oral agent that cov­ ers gram neg­ a­
tives, and an
anti-herpes simplex virus agent) for those with prolonged neu-
tropenia and reg­u­lar eval­u­a­tion of blood work for deter­mi­na­tion
of trans­fu­sion needs (man­dated weekly dur­ing cycles 1-3 by the
clin­i­cal trial and gen­er­ally performed at least 1-2 times per week
in my own prac­tice as a stan­dard of care), par­tic­u­larly dur­ing
early ther­apy cycles, is crit­i­cal to opti­mize responses from this
agent, as for all­HMAs.4,24,25
Oral C-DEC is cur­rently the only oral HMA approved for MDS.
This agent was approved only for higher-risk MDS dis­ease cate-
gories, but alter­nate dos­ing strat­e­gies with this agent are cur­
rently under inves­ti­ga­tion to deter­mine the poten­tial ben­e­fit
for those with lower-risk dis­ ease (NCT03502668). Preliminary
reports of effi­cacy in higher-risk MDS from the phase 2 and 3
tri­als dem­on­strate clin­i­cal activ­ity and response that are at least
com­pa­ra­ble with (and pos­si­bly supe­rior to) the response to IV
decitabine. Although cur­rent guide­lines favor SQ azacitidine as
the first-choice HMA for patients with MDS given the sur­vival
ben­e­fit reported for this agent, decitabine and C-DEC are both
con­sid­ered accept­able alter­na­tives.10,50 Despite this caveat, sub­
se­quent reports from clin­ic ­ al tri­als of azacitidine and decitabine
sug­gest that sur­vival dif­fer­ences in the ini­tial clin­i­cal tri­als may
stem from issues of trial design rather than true dif­fer­ences in
effi­cacy, and a major­ity of prac­tic­ing MDS experts would con­
sider these agents largely inter­change­able.25
Unmodified oral azacitidine
In Sep­tem­ber 2020 unmod­i­fied oral azacitidine (sub­se­quently
des­ig­nated CC-486) at a dose of 300 mg daily for 14 out of 28
days (not 21 days, as tested in the MDS study) per month was
approved by the US Food and Drug Administration as main­te­
nance ther­apy for inter­me­di­ate- and poor-cyto­ge­netic-risk AML
patients in com­plete remis­sion fol­low­ing an inten­sive induc­tion
reg­i­men; this agent is not cur­rently approved for patients with
MDS.51,52 Although this agent shares its name with par­en­ter­ally
admin­is­tered azacitidine, the dis­tinct PK pro­files resulting from
oral admin­is­tra­tion of this agent ren­der it a dif­fer­ent drug (Figure
2b). Substitution of CC-486 for IV or SQ azacitidine is not accept­
able for patients with MDS.
444  |  Hematology 2021  |  ASH Education Program
As shown pre­vi­ously for sin­gle-agent decitabine, the exten­
sive expres­sion of CDA results in low bio­avail­abil­ity of unmod­i­
fied oral azacitidine. The pilot study of CC-486 ini­tially dosed it
daily for 7 days in an attempt to reca­pit­u­late the stan­dard par­
en­teral dos­ing sched­ule. In this study the max­i­mum tol­er­ated
dose was iden­ti­fied to be 480 mg, with a mean oral bio­avail­
abil­ity between 6% and 20%. Relatively poor oral bio­avail­abil­ity
could not be over­come due to dose-lim­it­ing grades 3 and 4 GI
tox­ic­ity (diar­rhea [12.2%], nau­sea [7.3%], vomiting [7.3%]) at the
highest doses; other grade 3/4 AEs included febrile neutropenia
(19.5%) and fatigue (9.8%).53 With this reg­i­men, hypomethylation
was low­est at day 15 and recov­ered to base­line by the end of
the cycle.53,54 While some responses (mostly HI) were observed
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/439/1852126/439griffiths.pdf by guest on 13 December 2021
in this early study, decreases in meth­yl­at­ion across a range of
tested loci (the PD bio­marker for activ­ity) were less robust with
CC-486 com­pared with the par­en­teral azacitidine for­mu­la­tion.
Given the prevailing hypoth­e­sis that the opti­mal hypometh-
ylating activ­ity of HMAs requires incor­po­ra­tion dur­ing S phase as
well as the rel­a­tively slow pro­lif­er­a­tive rate of MDS pro­gen­i­tors,
sub­se­quent drug devel­op­ment focused on extending the dos­ing
sched­ule to 14 or 21 days to allow opti­mal drug incor­po­ra­tion.38
Lower peak blood drug lev­els, with the sustained low-level expo­
sure asso­ci­ated with this dos­ing strat­egy, would be expected to
improve response rates. Indeed, exten­sion of the dos­ing sched­ule
in this man­ner in a PK/PD study dem­on­strated effec­tive induc­
tion of hypomethylation using doses of 200 mg twice daily for 14
days or 300 mg daily for 14 or 21 days that were low­est at day 22
of treat­ment; hypomethylation of selected loci persisted at day
28.33 This dose sched­ule was tol­er­a­ble, and although a major­ity of
patients expe­ri­enced GI (84%) and hema­to­log­ic ­ al (81%) toxicities,
these were not dose lim­it­ing.38 The oral admin­is­tra­tion of CC-486
results in rapid absorp­tion from the GI tract, with food con­sump­
tion and gas­tric pH hav­ing lim­ited impacts on blood drug lev­els.37
Subset ana­ly­ses of early CC-486 stud­ies have suggested that low-
er-risk patients, par­tic­u­larly those with throm­bo­cy­to­pe­nia, might
derive par­tic­u­lar ben­e­fit from such a low-dose oral reg­i­men.39
Recently, a phase 3 study of CC-486 in TD lower-risk MDS
patients (by IPSS cri­te­ria) reported results on 216 patients ran­
dom­ized 1:1 between pla­cebo and CC-486 dosed at 300 mg daily
for 21 days of a 28-day sched­ule (Figure 3).55 Eligibility required
an Eastern Cooperative Oncology Group per­for­mance sta­tus of
at least 2, throm­bo­cy­to­pe­nia (plate­let count ≤75   ×   109/L), and
RBC trans­fu­sion depen­dence of at least 2 units per month for at
least 2 months (International Working Group 2006 cri­te­ria).12 The
pri­mary end point for this study was RBC TI; the use of exog­e­
nous growth fac­tors was not per­mit­ted.
Enrolled and treated patients in this study were of median
age 74 years (range, 30-89), with IPSS-defined inter­me­di­ate-1 risk
dis­ease (by IPSS-R, 28% had high or very high-risk MDS). They
were heavily RBC TD (requir­ing about 3 units/month) and base­
line throm­bo­cy­to­pe­nic (median 25   ×   109/L; ~30% plate­let TD; 40%
with base­line plate­let count <20   ×   109/L). The study met its pri­mary
end point of RBC TI in 30% of the CC-486 patients vs 11% of pla­
cebo-treated patients; HI of the plate­let count was also higher
for CC-486-treated patients—24% vs 4.6%. Responses took on
aver­age 2.4 months, and decreased trans­fu­sion needs for both
RBCs and plate­lets were sustained with CC-486 treat­ment (for a
median of 10 months for RBCs and 12 months for plate­lets); bilin-
eage improve­ments (in both Hb and plate­let counts) were seen
in about a quar­ter of the CC-486-treated patients. With a median
fol­low-up of about 13 months, an under­pow­ered over­all sur­vival
esti­mate was the same in both arms, although the rate of AML pro­
gres­sion was lower in the CC-486 arm. Toxicities were as expected
with an HMA-based ther­apy and were worst in the first cycles of
treat­ment. Most patients reported low-grade GI tox­ic­ity asso­ci­
ated with tak­ing CC-486. Grades 3 and 4 cytopenias (neutropenia,
47% vs 12%; throm­bo­cy­to­pe­nia, 29% vs 16%) and infec­tions (febrile
neutropenia, 28% vs 10%) were more com­mon in the CC-486-
treated patients vs those who got pla­cebo. A higher rate of early
death was reported in the CC-486 arm (16 vs 6 patients), pre­dom­
i­nantly asso­ci­ated with bac­te­rial infec­tions as well as neutropenia.
The phase 3 data reviewed above sug­gest sub­stan­tial activ­
ity in TD lower-risk MDS patients, likely supporting an expan­sion
of the labeled indi­ca­tion in the near future.55 High rates of early
infec­tious death in this lower-risk pop­u­la­tion should at least war­
rant the rou­tine use of pro­phy­lac­tic anti-infec­tives in order to
mit­i­gate risk and may give the Food and Drug Administration
pause in the deci­sion to approve this agent.24 Since CC-486 has
a mark­edly dis­tinct PK/PD pro­file from par­en­teral azacitidine,
the two agents are not inter­change­able (Figure 2b); addi­tional
stud­ies are needed to deter­mine whether CC-486 is appro­pri­
ate for indi­vid­u­als with higher-risk MDS. Presently, CC-486 use in
patients with MDS is lim­ited to the con­text of clin­i­cal tri­als.
Other oral HMAs in devel­op­ment
Other oral HMAs are in clin­i­cal devel­op­ment for patients with
MDS. Among these are NTX-301 (NCT04167917), a novel orally bio­
avail­able DNMT1 inhib­i­tor, and ASTX-030, a com­bi­na­tion of azacit-
idine with oral cedazuridine (ASTX-030; NCT04256317). The lat­ter
is being devel­oped using a com­bined phase 1 to 3 clin­i­cal trial
based upon the PK equiv­a­lence model pioneered for C-DEC (Fig-
ure 3). It is likely that in the near future we will have approval of at
least 3 oral HMAs for patients with MDS. Moreover, sev­eral stud­ies
are either planned or under­way to com­bine oral HMAs with other
prom­is­ing MDS drugs, includ­ing pevonidostat, venetoclax, and sa-
batolimab, among oth­ers (NCT04655755, NCT04878432). Histori-
cally, com­bi­na­to­rial stud­ies have failed to dem­on­strate improve­
ments over the sin­gle agent.2,56 The hypoth­e­ses for com­bi­na­tion
fail­ures include pos­tu­lated inad­e­quate dose inten­sity for the HMA
and antag­o­nism between agents due to con­ve­nience-dic­tated
overlapping sched­ules, among other rea­sons.2,56 The sub­sti­tu­tion
of an oral HMA in such com­bi­na­tions might facil­i­tate improved
dose/sched­ule com­bi­na­tions that could the­o­ret­i­cally enhance
effi­cacy and poten­tially decrease treat­ment mod­i­fi­ca­tions.2,56-58
CLINICAL CASE 3
A 78-year-old gen­tle­man with a his­tory of high-risk MDS (base­
line hypercellular dys­plas­tic mar­row with blasts of 15%, nor­mal
kar­yo­type in 20 meta­phases, NGS with 2 dis­tinct TET2 muta­
tions, base­line trilineage cytopenias) has been maintained on SQ
azacitidine 50 mg/m2 daily for 7 days every 28 days. The patient
achieved com­plete remis­sion after 6 cycles of ther­apy dosed
at 75 mg/m2 daily for 7 days every 28 days and has been non-
TD since cycle 5. Due to the devel­op­ment of neutropenia after
cycle 9, a BM biopsy was repeated that showed ongo­ing CR
with a hypocellular mar­row. The azacitidine dose was reduced
to 50 mg/m2 for 7 days of a 28-day cycle, and the patient
Prakash Singh Shekhawat
remained TI and nonneutropenic. Due to con­cern about the risk
from COVID-19, he requested tran­si­tion to an oral reg­i­men, hav­
ing learned from a recent patient edu­ca­tion event about the
approval of C-DEC. His phy­ si­
cian agreed to the switch, and
3 weeks into cycle 1, the patient presented with short­ness of
breath and fatigue. A CBC dem­on­strated an Hb of 5.7 g, an ANC
of 0.6   ×   109/µL, and plate­lets of 150   ×   109/µL. The patient and his
phy­si­cian became concerned about the pos­si­bil­ity of dis­ease
pro­gres­sion, and an urgent BM biopsy was performed.
Switching from par­en­teral to oral ther­apy
Recent short­ages of par­en­teral azacitidine and con­cerns about
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/439/1852126/439griffiths.pdf by guest on 13 December 2021
the risk of COVID-19 infec­tion posed to patients with can­cer by
expo­sure to infu­sion clin­ics have resulted in rapid sub­sti­tu­tion of
C-DEC for par­en­teral HMAs.59,60 It is impor­tant to rec­og­nize that
C-DEC results in iden­ti­cal PKs with IV decitabine. However, and as
empha­sized in the Introduction, the PKs of decitabine and azaciti-
dine are not iden­ti­cal. For patients who have been maintained on
dose-reduced azacitidine, the sub­sti­tu­tion of oral C-DEC rep­re­
sents a sub­stan­tial dose increase in terms of both peak lev­els and
total AUC expo­sure to HMA (Table 1). C-DEC is avail­­able only as a
fixed-dose tab­let; thus, dose reduc­tions require either a decrease
in the num­ber of days of ther­apy or dos­ing on an interrupted
sched­ule. The pack­age insert for C-DEC sug­gests dose reduc­tions
from 5 to 4 days, sub­se­quently to 3 days, and ulti­mately dos­ing
every other day on days 1, 3, and 5. Due to the action of cedazuri-
dine, which accu­mu­lates dur­ing the 5-day dos­ing period, the AUC
for decitabine when C-DEC is dosed daily is sub­stan­tially higher
than when it is dosed every other day. Therefore, dose reduc­tions
to a sched­ule of every other day result in sub­stan­tially lower decit-
abine expo­sures. When transitioning patients from an established
par­en­teral HMA reg­i­men, par­tic­u­larly when dose reduc­tions have
been made, it is impor­tant to rec­og­nize these dif­fer­ences in drug
deliv­ery and to either plan for up-front dose reduc­tion to mit­i­gate
cytopenias or to reinstitute appro­pri­ate mon­i­tor­ing for tox­ic­ity
dur­ing the ini­tial treat­ment period. The patient described in case
3 under­went a BM biopsy to exclude dis­ease pro­gres­sion, which
dem­ on­
strated only hypocellularity. After a 2-week delay from
his sched­uled cycle of treat­ment, blood counts fully recov­ered.
C-DEC ther­apy was resumed, dose reduced to 3 days per cycle
this time, and blood counts were checked weekly dur­ing the next
2 cycles of ther­apy. No fur­ther trans­fu­sions were required, and the
patient remained nonneutropenic. A fail­ure to rec­og­nize the dif­
fer­ence between SQ azacitidine and oral C-DEC in this case could
have resulted in sig­nif­i­cant risk to this patient. Failure to plan for,
or to pre­emp­tively adjust dos­ing in antic­i­pa­tion of these dif­fer­
ences, might prompt both patient and phy­si­cian reluc­tance to
con­tinue with oral ther­apy and thereby con­trib­ute to a loss of dis­
ease response or frank relapse.
Conclusions
The advent of orally bio­avail­able HMAs rep­re­sents a sub­stan­
tial poten­tial ben­e­fit for patients with MDS. While both oral
decitabine (C-DEC) and oral azacitidine (CC-486) have received
reg­u­
la­
tory approval, these two agents are dis­ tinct, and only
C-DEC is cur­ rently approved for patients with MDS. C-DEC
fixed-dose tab­lets given for 5 days pro­vide iden­ti­cal PK expo­
sure to IV decitabine for 5 days, while CC-486 is dosed daily for
14 or 21 days and results in lower peak blood-drug lev­els than SQ
Oral hypomethylating agents for MDS  |  445
or IV azacitidine but a more prolonged expo­sure win­dow. The dif­
fer­ences in PK and PD pro­files for these two agents ren­der them
quite dis­tinct in terms of both poten­tial tox­ic­ity and mech­a­nisms
of action. As a treating cli­ni­cian, I sub­sti­tute C-DEC directly 1:1 for
IV decitabine at a dose of 20 mg/m2 for 5 days. By con­trast, CC-
486 can­not be con­sid­ered equiv­a­lent to IV or SQ azacitidine, and
in the con­text of MDS, its effi­cacy must be tested pro­spec­tively.
Orally bio­avail­able HMAs have the poten­tial to prove more
con­ve­nient for MDS patients and fam­i­lies, spar­ing them repeated
vis­its to infu­sion clin­ics for the sole pur­pose of drug deliv­ery, but
these agents result in sub­stan­tial cytopenias that require rou­tine
assess­ment of trans­fu­sion needs and con­sid­er­ation for pro­phy­
lac­tic anti­mi­cro­bi­als. Most patients treated with these agents in
clin­i­cal trial devel­oped neutropenia, which was espe­cially prom-
inent dur­ing early cycles of ther­apy, and neutropenic fever was
among the most fre­quently reported AEs. As with par­en­teral
HMA ther­apy, opti­mal response to oral HMAs will require sus-
tained on-sched­ule treat­ment cycles and opti­mized sup­port­ive
care. When we con­sider switching to an oral reg­i­men in an effort
to opti­mize qual­ity of life for a patient, par­tic­ul­ar care should
be taken to rec­og­nize the dif­fer­ent PK pro­files for the new oral
reg­i­men and to antic­ip ­ ate the poten­tial toxicities resulting from
this change.
Acknowledgments  posal for mod­i­fi­ca­tion of the International Working Group (IWG) response
Elizabeth A. Griffith received research sup­port from the Rappa-
port Family Foundation and the Charles D and Mary A Bauer Foun-
dation. This work was supported by Roswell Park Comprehensive
Cancer Center and National Cancer Institute grant P30CA016056.
The author would also like to acknowl­ edge David Eifrig,
Michael Nemeth, Valeria Santini, and Sophia Balderman for edi­
to­rial review of the arti­cle.
Conflict-of-inter­est dis­clo­sure
Elizabeth A. Griffiths: research funding: Apellis Pharmaceu-
ticals, Alexion Pharmaceuticals, Astex Pharmaceuticals, Cel-
gene/Bristol Myers-Squibb, Celldex Therapeutics, Genentech;
hon­o­raria: Abbvie, Astex Pharmaceuticals, Bos­ton Biomedical,
Celgene/Bristol Myers-Squibb, Novartis Oncology, Takeda On-
cology, Taiho Oncology.
Off-label drug use
Elizabeth A. Griffiths: This review discusses the use of CC-486/
unmodified oral azacitidine for MDS, an indication that is not FDA
approved.
Correspondence
Elizabeth A. Griffiths, Roswell Park Comprehensive Cancer
Center, Elm and Carlton St, Buffalo, NY 14263; e-mail: elizabeth​
­.griffiths@roswellpark​­.org.
References
1. Stomper J, Rotondo JC, Greve G, Lübbert M. Hypomethylating agents
(HMA) for the treat­ment of acute mye­loid leu­ke­mia and myelodysplastic
syn­dromes: mech­a­nisms of resis­tance and novel HMA-based ther­a­pies.
Leukemia. 2021;35(7):1873-1889.
2. Fandy TE, Herman JG, Kerns P, et  al. Early epi­ge­netic changes and DNA
dam­age do not pre­dict clin­i­cal response in an overlapping sched­ule of
5-azacytidine and entinostat in patients with mye­loid malig­nan­cies. Blood.
2009;114(13):2764-2773.
3. Jones PA, Ohtani H, Chakravarthy A, De Carvalho DD. Epigenetic ther­apy in
immune-oncol­ogy. Nat Rev Cancer. 2019;19(3):151-161.
446  |  Hematology 2021  |  ASH Education Program
4. Santini V, Prebet T, Fenaux P, et  al. Minimizing risk of hypomethylating
agent fail­ure in patients with higher-risk MDS and prac­ti­cal man­age­ment
rec­om­men­da­tions. Leuk Res. 2014;38(12):1381-1391.
5. Voso MT, Breccia M, Lunghi M, et  al. Rapid loss of response after with­
drawal of treat­ment with azacitidine: a case series in patients with higher-
risk myelodysplastic syn­dromes or chronic myelomonocytic leu­ke­mia. Eur
J Haematol. 2013;90(4):345-348.
6. Tendas A, Lissia MF, Piccioni D, et  al. Obstacles to adher­ence to azaciti-
dine admin­is­tra­tion sched­ule in out­pa­tient myelodysplastic syn­drome and
related dis­or­ders. Support Care Cancer. 2015;23(2):303-305.
7. Silverman LR, Fenaux P, Mufti GJ, et  al. Continued azacitidine ther­ apy
beyond time of first response improves qual­ity of response in patients with
higher-risk myelodysplastic syn­dromes. Cancer. 2011;117(12):2697-2702.
8. Steensma DP, Baer MR, Slack JL, et  al. Multicenter study of decitabine
admin­is­tered daily for 5 days every 4 weeks to adults with myelodysplastic
syn­dromes: the Alternative Dosing for Outpatient Treatment (ADOPT) trial.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/439/1852126/439griffiths.pdf by guest on 13 December 2021
J Clin Oncol. 2009;27(23):3842-3848.
9. Kantarjian H, Issa JP, Rosenfeld CS, et al. Decitabine improves patient out­
comes in myelodysplastic syn­dromes: results of a phase III ran­dom­ized
study. Cancer. 2006;106(8):1794-1803.
10. Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al; International Vidaza High-
Risk MDS Survival Study Group. Efficacy of azacitidine com­ pared with
that of con­ven­tional care reg­i­mens in the treat­ment of higher-risk myel-
odysplastic syn­dromes: a randomised, open-label, phase III study. Lancet
Oncol. 2009;10(3):223-232.
11. Santini V, Ossenkoppele GJ. Hypomethylating agents in the treat­ment of
acute mye­loid leu­ke­mia: a guide to opti­mal use. Crit Rev Oncol Hematol.
2019;140(August):1-7.
12. Cheson BD, Greenberg PL, Bennett JM, et al. Clinical appli­ca­tion and pro­
cri­te­ria in myelodysplasia. Blood. 2006;108(2):419-425.
13. Silverman LR, Fenaux P, Mufti GJ, et al. The effects of con­tin­ued azacitidine
treat­ment cycles on response in higher risk patients with myelodysplastic
syn­dromes: an update. Ecancermedicalscience. 2008;2(12 August):118.
14. Gore SD, Fenaux P, Santini V, et al. A mul­ti­var­i­ate anal­y­sis of the rela­tion­
ship between response and sur­vival among patients with higher-risk myel-
odysplastic syn­ dromes treated within azacitidine or con­ ven­tional care
reg­i­mens in the ran­dom­ized AZA-001 trial. Haematologica. 2013;98(7):
1067-1072.
15. Prébet T, Gore SD, Esterni B, et al. Outcome of high-risk myelodysplastic
syn­drome after azacitidine treat­ment fail­ure. J Clin Oncol. 2011;29(24):3322-
3327.
16. Jabbour E, Garcia-Manero G, Batty N, et  al. Outcome of patients with
myelodysplastic syn­ drome after fail­ ure of decitabine ther­ apy. Cancer.
2010;116(16):3830-3834.
17. Jabbour EJ, Garcia-Manero G, Strati P, et al. Outcome of patients with low-
risk and inter­ me­ di­
ate-1-risk myelodysplastic syn­ drome after hypometh-
ylating agent fail­ ure: a report on behalf of the MDS Clinical Research
Consortium. Cancer. 2015;121(6):876-882.
18. Cabrero M, Jabbour E, Ravandi F, et al. Discontinuation of hypomethylating
agent ther­apy in patients with myelodysplastic syn­dromes or acute mye­
log­e­nous leu­ke­mia in com­plete remis­sion or par­tial response: ret­ro­spec­tive
anal­y­sis of sur­vival after long-term fol­low-up. Leuk Res. 2015;39(5):520-
524.
19. Dinmohamed AG, van Norden Y, van de Loosdrecht AA, Jongen-Lavrencic
M. Effectiveness of azacitidine in higher-risk myelodysplastic syn­dromes.
Leukemia. 2016;30(8):1795-1796.
20. Bernal T, Martínez-Camblor P, Sánchez-García J, et al; Span­ish Group on
Myelodysplastic Syndromes; PETHEMA Foundation; Span­ ish Society of
Hematology. Effectiveness of azacitidine in unse­ lected high-risk myel-
odysplastic syn­ dromes: results from the Span­ ish reg­is­
try. Leukemia.
2015;29(9):1875-1881.
21. Zeidan AM, Stahl M, Hu X, et al. Long-term sur­vival of older patients with
MDS treated with HMA ther­apy with­out sub­se­quent stem cell trans­plan­ta­
tion. Blood. 2018;131(7):818-821.
22. Jung HA, Maeng CH, Kim M, Kim S, Jung CW, Jang JH. Platelet response
dur­ing the sec­ond cycle of decitabine treat­ment pre­dicts response and
sur­vival for myelodysplastic syn­drome patients. Oncotarget. 2015;6(18):
­16653-16662.
23. Itzykson R, Thépot S, Quesnel B, et al; Groupe Francophone des Myelodys-
plasies. Prognostic fac­tors for response and over­all sur­vival in 282 patients
with higher-risk myelodysplastic syn­ dromes treated with azacitidine.
Blood. 2011;117(2):403-411.
24. Santini V, Fenaux P, Mufti GJ, et al. Management and sup­port­ive care mea­
sures for adverse events in patients with myelodysplastic syn­ dromes
treated with azacitidine. Eur J Haematol. 2010;85(2):130-138.
25. Zeidan AM, Hu X, Zhu W, et al. Does pro­vider expe­ri­ence treating patients
(pts) with myelodysplastic syn­dromes (MDS) explain dura­tion of hypometh-
ylating agent (HMA) ther­apy and over­all sur­vival (OS)? A large pop­u­la­tion-
based anal­y­sis. Blood. 2018;132(suppl 1):370.
26. Stein EM, Bonifacio G, Latremouille-Viau D, et  al. Treatment pat­ terns
and out­comes in patients with myelodysplastic syn­dromes treated with
hypomethylating agents: a SEER-Medi­ care anal­
y­
sis. Leuk Lymphoma.
2021;62(6):1411-1421.
27. Grinblatt DL, Sekeres MA, Komrokji RS, Swern AS, Sullivan KA, Narang M.
Patients with myelodysplastic syn­dromes treated with azacitidine in clin­i­
cal prac­tice: the AVIDA reg­is­try. Leuk Lymphoma. 2015;56(4):887-895.
28. Davidoff AJ, Hu X, Bewersdorf JP, et al. Hypomethylating agent (HMA) ther­
apy use and sur­vival in older adults with refrac­tory ane­mia with excess
blasts (RAEB) in the United States (USA): a large pro­pen­sity score-matched
pop­u­la­tion-based study. Leuk Lymphoma. 2020;61(5):1178-1187.
29. Zeidan AM, Joshi N, Kale H, et al. Predictors of hypomethylating agent dis­
con­tin­u­a­tion among patients with higher-risk myelodysplastic syn­dromes.
J Clin Oncol. 2021;39(suppl 15):7043.
30. Corman S, Joshi N, Wert T, Kale H, Hill K, Zeidan AM. Under-use of
hypomethylating agents in patients with higher-risk myelodysplastic syn­
drome in the United States: a large pop­u­la­tion-based anal­y­sis. Clin Lym-
phoma Myeloma Leuk. 2021;21(2):e206-e211.
31. Marcucci G, Silverman L, Eller M, Lintz L, Beach CL. Bioavailability of azaciti-
dine sub­cu­ta­ne­ous ver­sus intra­ve­nous in patients with the myelodysplastic
syn­dromes. J Clin Pharmacol. 2005;45(5):597-602.
32. Mahfouz RZ, Jankowska A, Ebrahem Q , et  al. Increased CDA expres­sion/
activ­ity in males con­trib­utes to decreased cyti­dine ana­log half-life and likely
con­trib­utes to worse out­comes with 5-azacytidine or decitabine ther­apy.
Clin Cancer Res. 2013;19(4):938-948.
33. Laille E, Shi T, Garcia-Manero G, et al. Pharmacokinetics and phar­ma­co­
dy­nam­ics with extended dos­ing of CC-486 in patients with hema­to­logic
malig­nan­cies. PLoS One. 2015;10(8):e0135520.
34. Vidaza (azacitidine) for injec­tion. Package insert, https://www.accessdata
.fda.gov/drugsatfda_docs/label/2008/​050794s​011lbl.pdf accessed 10/1/21.
35. Dacogen® (decitabine) for injec­tion. Package insert, https://www.accessdata​
.fda.gov/drugsatfda_docs/label/2018/​021790s021lbl­.pdf accessed 10/­1/21.
36. Watanabe S, Uchida T. Expression of cyti­dine deam­i­nase in human solid
tumors and its reg­u­la­tion by 1 alpha, 25-dihydroxyvitamin D3. Biochim Bio-
phys Acta. 1996;1312(2):99-104.
37. Laille E, Savona MR, Scott BL, Boyd TE, Dong Q , Skikne B. Pharmacokinetics
of dif­fer­ent for­mu­la­tions of oral azacitidine (CC-486) and the effect of food
and mod­i­fied gas­tric pH on phar­ma­co­ki­net­ics in sub­jects with hema­to­logic
malig­nan­cies. J Clin Pharmacol. 2014;54(6):630-639.
38. Savona MR, Kolibaba K, Conkling P, et  al. Extended dos­ing with CC-486
(oral azacitidine) in patients with mye­loid malig­nan­cies. Am J Hematol.
2018;93(10):1199-1206.
39. Garcia-Manero G, Scott BL, Cogle CR, et al. CC-486 (oral azacitidine) in patients
with myelodysplastic syn­dromes with pre­treat­ment throm­bo­cy­to­pe­nia. Leuk
Res. 2018;72(Sep­tem­ber):79-85.
40. Garcia-Manero G, Gore SD, Kambhampati S, et  al. Efficacy and safety of
extended dos­ing sched­ules of CC-486 (oral azacitidine) in patients with
lower-risk myelodysplastic syn­dromes. Leukemia. 2016;30(4):889-896.
41. Mistry B, Jones MM, Kubiak P, et al. A phase 1 study to assess the abso­lute
bio­avail­abil­ity and safety of an oral solu­tion of decitabine in sub­jects with
myelodysplastic syn­dromes (MDS). Blood. 2011;118(21):3801.
42. Frances A, Cordelier P. The emerg­ing role of cyti­dine deam­i­nase in human
dis­eases: a new oppor­tu­nity for ther­apy? Mol Ther. 2020;28(2):357-366.
43. Inqovi® (decitabine and cedazuridine) tab­lets. Package insert, https://www
.accessdata.fda.gov/drugsatfda_docs/label/2020/212576s000lbl.pdf
accessed 10/1/2021.
Prakash Singh Shekhawat
44. Garcia-Manero G, McCloskey J, Griffiths EA, et al. Pharmacokinetic expo­
sure equiv­a­lence and pre­lim­i­nary effi­cacy and safety from a ran­dom­ized
cross over phase 3 study (ASCERTAIN study) of an oral hypomethylating
agent ASTX727 (cedazuridine/decitabine) com­ pared to IV decitabine.
Blood. 2019;134(suppl 1):846.
45. Savona MR, McCloskey JK, Griffiths EA, et  al. Clinical effi­cacy and safety
of oral decitabine/cedazuridine in 133 patients with myelodysplastic syn­
dromes (MDS) and chronic myelomonocytic leu­ ke­ mia (CMML). Blood.
2020;136(suppl 1):37-38.
46. Thota S, Oganesian A, Azab M, Griffiths EA. Role of cedazuridine/decit-
abine in the man­age­ment of myelodysplastic syn­drome and chronic myel-
omonocytic leu­ke­mia. Future Oncol. 2021;17(16):2077-2087.
47. Savona MR, Odenike O, Amrein PC, et  al. Results of first in human (FIH)
phase 1 phar­ma­co­ki­netic (PK) guided dose-esca­la­tion study of ASTX727,
a com­bi­na­tion of the oral cyti­dine deam­i­nase inhib­i­tor (CDAi) E7727 with
oral decitabine in sub­jects with myelodysplastic syn­dromes (MDS). Blood.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/439/1852126/439griffiths.pdf by guest on 13 December 2021
2015;126(23):1683.
48. Savona MR, Odenike O, Amrein PC, et al. An oral fixed-dose com­bi­na­tion
of decitabine and cedazuridine in myelodysplastic syn­dromes: a multi-
centre, open-label, dose-esca­ la­
tion, phase 1 study. Lancet Haematol.
2019;6(4):e194-e203.
49. Garcia-Manero G, Griffiths EA, Steensma DP, et al. Oral cedazuridine/decit-
abine for MDS and CMML: a phase 2 phar­ma­co­ki­netic/phar­ma­co­dy­namic
ran­dom­ized cross­over study. Blood. 2020;136(6):674-683.
50. National Comprehensive Cancer Network. Myelodysplastic syn­ dromes:
Clinical prac­tice guide­lines in oncol­ogy. J Nat Compr Canc Netw. 2003;1(4):
456-471.
51. Wei AH, Döhner H, Pocock C, et al; QUAZAR AML-001 Trial Investigators.
Oral azacitidine main­te­nance ther­apy for acute mye­loid leu­ke­mia in first
remis­sion. N Engl J Med. 2020;383(26):2526-2537.
52. Onureg (azacitidine) tab­lets. Package insert, https://www.accessdata.fda​
.gov/drugsatfda_docs/label/2020/​214120s​000lbl.pdf accessed 10/1/21.
53. Garcia-Manero G, Stoltz ML, Ward MR, Kantarjian H, Sharma S. A pilot phar­
ma­co­ki­netic study of oral azacitidine. Leukemia. 2008;22(9):1680-1684.
54. Garcia-Manero G, Gore SD, Cogle C, et al. Phase I study of oral azacitidine
in myelodysplastic syn­ dromes, chronic myelomonocytic leu­ ke­
mia, and
acute mye­loid leu­ke­mia. J Clin Oncol. 2011;29(18):2521-2527.
55. Garcia-Manero G, Santini V, Almeida A, et  al. Phase III, ran­dom­ized, pla­
cebo-con­trolled trial of CC-486 (oral azacitidine) in patients with lower-risk
myelodysplastic syn­dromes. J Clin Oncol. 2021;39(13):1426-1436.
56. Sekeres MA, Othus M, List AF, et al. Randomized phase II study of azacitidine
alone or in com­bi­na­tion with lenalidomide or with vorinostat in higher-risk
myelodysplastic syn­dromes and chronic myelomonocytic leu­ke­mia: North
Amer­i­can Intergroup Study SWOG S1117. J Clin Oncol. 2017;35(24):2745-
2753.
57. Cameron EE, Bachman KE, Myöhänen S, Herman JG, Baylin SB. Synergy of
demethylation and his­tone deacetylase inhi­bi­tion in the re-expres­sion of
genes silenced in can­cer. Nat Genet. 1999;21(1):103-107.
58. Gore SD, Baylin S, Sugar E, et al. Combined DNA methyltransferase and his­
tone deacetylase inhi­bi­tion in the treat­ment of mye­loid neo­plasms. Cancer
Res. 2006;66(12):6361-6369.
59. Kuderer NM, Choueiri TK, Shah DP, et al; COVID-19 and Cancer Consortium.
Clinical impact of COVID-19 on patients with can­cer (CCC19): a cohort
study. Lancet. 2020;395(10241):1907-1918.
60. Zeidan AM, Boddu PC, Patnaik MM, et  al. Special con­sid­er­ations in the
man­age­ment of adult patients with acute leu­kae­mias and mye­loid neo­
plasms in the COVID-19 era: rec­om­men­da­tions from a panel of inter­na­-
tional experts. Lancet Haematol. 2020;7(8):e601-e612.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000278
Oral hypomethylating agents for MDS  |  447
 MPN: NEW DIRECTIONS
EVIDENCE-BASED MINIREVIEW
Are DOACs an alternative to vitamin K
antagonists for treatment of venous
thromboembolism in patients with MPN?
Francesca Schieppati1 and Anna Falanga1,2
1
Department of Transfusion Medicine and Hematology, Hospital Papa Giovanni XXIII, Bergamo, Italy; and 2Department of Medicine and Surgery,
University of Milan Bicocca, Milan, Italy
LEARNING OBJECTIVES
• Understand the current strategies of VTE treatment and prevention of recurrences in patients with MPN
• Identify the pros and cons of traditional VTE treatment with VKAs vs newer approaches with DOACs in light of the
available evidence
CLINICAL CASE
A 55-year-old man with a history of JAK2-mutated essen-
tial thrombocythemia (ET) presented to the emergency
room for the sudden onset of severe pain and swelling
in his right leg during the night. He denied any recent
trauma, surgery, or infection. He was in good general con-
dition except for a modest shortness of breath. An elec-
trocardiogram showed a sinus tachycardia; the oxygen
saturation was 95% in room air. At physical examination, a
palpable cord at the right thigh was appreciable, associ-
ated with marked edema and erythema of the calf. A com-
pression ultrasonography revealed an absence of color
flow compatible with complete deep vein thrombosis of
the femoral and popliteal veins of the right leg. A chest
computed tomography scan also showed bilateral seg-
mental pulmonary embolism. His past medical history was
otherwise mute except for being a heterozygous carrier
of the factor V Leiden variant and having hemorrhoids in
the past. He was on low-dose aspirin as his sole medica-
tion. What is the best treatment of venous thromboembo-
lism in this ET patient?
Introduction
Classical BCR/ABL-negative myeloproliferative neoplasms
(MPN) include polycythemia vera (PV), essential thrombo-
cythemia (ET), and primary myelofibrosis (PMF). Patients
with MPN are at high risk of thrombotic manifestations,
which considerably affect morbidity and mortality, espe-
cially in younger patients.1,2 Up to 30% of patients pres-
448 | Hematology 2021 | ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/448/1851806/448schieppati.pdf by guest on 13 December 2021
ent with a thrombotic event before or at MPN diagnosis.3
Although arterial thrombotic events (ATEs) are twice as
common as venous events, MPN patients have a 4-fold
increased risk of ATEs and a 10-fold increased risk of venous
thromboembolism (VTE) shortly after diagnosis compared
to the general population, and the thrombotic rate remains
significantly elevated throughout the follow-up.2
MPN-related VTE manifests most often as deep vein
thrombosis (DVT) of the legs and/or pulmonary embo-
lism (PE; accounting for 40%-90% of all cases in different
reports), although thromboses at unusual sites, ie, involv-
ing the splanchnic and cerebral districts, are remarkably
frequent in these patients.4
In PV and ET, treatment is aimed at preventing throm-
botic complications, and risk-stratification for treatment
decisions is based on thrombotic risk factors, including an
age over 60, a history of thrombosis, and, only in ET, the
presence of the JAK2V617F mutation.5,6 In this regard, the
substantial involvement of JAK2V617F and clonal hemato-
poiesis in thrombosis development in MPN has emerged.7
Current strategies of thromboprophylaxis include the
use of low-dose aspirin (75-100 mg) once daily in both high-
and low-risk PV patients (aged <60 years and no thrombo-
sis history) and in low-/intermediate-risk ET patients (JAK2
mutated OR aged >60 years and no thrombosis history),5
based on 2 randomized controlled trials in PV and on 1 ret-
rospective analysis in ET, respectively.8-10 Low-dose aspi-
rin showed a considerable but non-statistically significant
beneficial effect on mortality from thrombotic events and
did not prevent major cardiovascular and venous throm-
botic events, taken individually.8,11 Thromboprophylaxis also
Table 1. Principal stud­ies includ­ing MPN patients on VKA treat­ment for usual site VTE

Median
N included Overall throm­bo­sis fol­low-up
Reference Study pop­u­la­tion in the study N on VKAs recur­rence (A/V) VTE recur­rence Major bleed­ing (years)
De Stefano PV/ET with at least 494 90 33.6% (7.6% pt-y)* 13.1% (3% pt-y)* 5.4% (0.9% pt-y)* 5.3
et al14 1 epi­sode of throm­bo­sis 7.7% (0.9% pt-y)†
(ATE and VTE) (2.8 pt-y)‡
Hernandez- PV/ET receiv­ing VKA for a 150 150 28% (6.0% pt-y)† 24% (2.7% ON vs 11.3% (over­all 1.7% pt-y, 7.7
Boluda first VTE or ATE epi­sode 9.0% OFF, p)† 1.8% ON vs 1.5%
et al16 OFF)†
De Stefano PV/ET/PMF on sys­temic 206 155 21.8% (6.5% pt-y)* 17.4% (5.2% pt-y)* 6.4% (2.4% ON vs 0.7 3
et al15 anticoagulation for a 12.2% (4.7% pt-y)† 9.6% (4.2% pt-y ON OFF)†
first VTE epi­sode vs 9.6% pt-y OFF)†

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/448/1851806/448schieppati.pdf by guest on 13 December 2021

Wille et al17 PV/ET/PMF with a first 78 40 — 20.5% (6.0% pt-y)* 26.9% 2
VTE epi­sode
*Entire cohort.
†Patients on VKA only.
‡Patients on VKA and aspi­rin.
A/V, arterial/venous; N, num­ber of patients; ON, on VKA; OFF, off VKA; Pt-y, patient-years.

includes phle­bot­omy in all­PV patients and myelosuppression
with cytoreductive ther­apy in high-risk PV and ET patients.5,12
Treatment of acute VTE and sec­ond­ary pre­ven­tion of VTE
recur­rence
Despite pro­phy­laxis, the reported inci­dence of VTE is 0.6% to
1.0% in patient-years across the dif­fer­ent MPN sub­types and is
con­sid­er­ably higher than the annual inci­dence of 0.1% to 0.2%
observed in the gen­eral pop­u­la­tion.4 Detected risk fac­tors for
a first VTE epi­sode in MPN patients include an age >60, a pre­
vi­ous his­tory of VTE, a his­tory of major bleed­ing, leu­ko­cy­to­sis,
inherited thrombophilia (in youn­ger patients), and JAK2V617F (in
ET and PMF).4
In light of a lack of pro­spec­tive stud­ies addressing the effi­
cacy and safety of the newer direct oral anti­co­ag­u­lants (DOACs)
in MPN patients, based on expert opin­ion the ini­tial treat­ment
for acute VTE in MPN patients should start with low-molec­u­lar-
weight hep­a­rin (LMWH) or fondaparinux followed by vita­min K
antag­o­nists (VKAs), targeting an inter­na­tional nor­mal­ized ratio
(INR) of 2.5 for at least 3 to 6 months.4,13
The effi­cacy and safety of VKAs in the MPN set­ting has been
eval­u­ated in 4 ret­ro­spec­tive stud­ies, includ­ing a very recent one
(Table 1).14-17 The annual inci­dence rate of VTE recur­rence ranged
between 3% and 6% in patient-years.14-17 VKA treat­ment was
asso­ci­ated with a sig­nif­i­cant reduc­tion in VTE recur­rences in all
4 stud­ies and ATEs in 1 study.16
With regard to the over­ all dura­ tion of anticoagulation in
these patients, while there is con­sen­sus on con­tinu­ing life­long
treat­ment in patients with splanch­nic vein throm­bo­sis,18 the opti­
mal treat­ment dura­tion for VTE at the usual sites is uncer­tain.
Two stud­ies com­par­ing VKA indef­i­nite treat­ment vs dis­con­tin­
u­a­tion after 6 months showed a sig­nif­i­cantly greater inci­dence
of recur­rence in the group that discontinued VKAs (2.7%-4.2%
in patient-years vs 9%-9.6%).15,16 Indeed, VKA sus­pen­sion resulted
in a 2- to 3-fold increased risk of recur­rence.15,16 In addi­tion, MPN
patients showed a higher rate of 5-year recur­rence after anti­
co­ag­u­lant with­drawal com­pared to non-MPN patients (42% vs
Prakash Singh Shekhawat
29%, respec­tively).15 However, it is impor­tant to con­sider that the
cumu­la­tive inci­dence of VTE recur­rence in MPN patients receiv­
ing ade­quate VKA treat­ment still remains greater than that of the
gen­eral pop­u­la­tion (7.8% vs 1.8%-3.5% at 1 year, respec­tively).15
In the stud­ies shown in Table 1, the bleed­ing inci­dence dur­
ing VKA treat­ment ranged between 0.9% and 2.8% patient-years
and sig­nif­i­cantly increased only when admin­is­tered in com­bi­na­
tion with aspi­rin (com­pared to patients off VKAs). Nevertheless,
bleed­ing com­pli­ca­tions with VKAs look higher in MPN com­pared
to non-MPN patients (up to 2.8% vs 1.2%-2.2% in patient-years,
respec­tively),4 with dis­ease-related fac­tors con­trib­ut­ing to the
over­all higher hem­or­rhagic risk in MPN patients com­pared to the
gen­eral pop­u­la­tion.3 This is a very rel­e­vant aspect when choos­
ing the type and dura­tion of anti­co­ag­u­lant ther­apy. Improving
the effi­cacy and safety of anticoagulation in MPN patients with
VTE still rep­re­sents an open issue.
In the last years, more ther­ a­
peu­ tic options for VTE have
become avail­­able with the advent of the DOACs, includ­ing the
fac­tor IIa inhib­i­tor dabigatran and the fac­tor Xa (FXa) inhib­i­tors
rivaroxaban, apixaban, edoxaban, and oth­ers. In the non-MPN
set­ting, DOACs have become the first treat­ment choice for DVT
and PE.19 Moreover, FXa inhib­i­tors have been tested spe­cif­i­cally
in the can­cer pop­u­la­tion by means of RCTs, show­ing a good effi­
cacy and safety pro­file com­pared to the stan­dard ther­apy with
LMWH.20-22 Thus, expert guide­lines have recently included these
drugs in the recommended treat­ment options for can­cer-asso­
ci­ated VTE.23,24 No pro­spec­tive con­trolled stud­ies on the use
of DOACs have been conducted with MPN patients so far. In
any case, some obser­va­tional ret­ro­spec­tive stud­ies have been
published in recent years eval­u­at­ing the effi­cacy and safety of
DOACs in MPN (Table 2).
Three small stud­ies described an over­all throm­botic recur­
rence of 0% to 4% involv­ing only arte­rial dis­tricts.25-27 Major
bleed­ing was reported in 0% to 12% of patients, with 3 cases of
clin­i­cally rel­e­vant non­ma­jor bleed­ing in asso­ci­a­tion with aspi­
rin.26 A recent large ret­ro­spec­tive study includ­ing 442 patients
treated with DOACs for atrial fibril­la­tion (AF) or for VTE pro­vided
DOACs vs VKAs in MPN-asso­ci­ated VTE treat­ment  |  449
Table 2. Studies includ­ing at least 20 MPN patients on DOAC treat­ment for usual site VTE

Median
Study N on N on N on N on N on Overall throm­botic fol­low-up
Reference pop­u­la­tion DOAC rivar apix edox dabig recur­rence VTE recur­rence Major bleed­ing (years)
Ianotto et al25
PV/ET receiv­ing 25* 16 9 — — 4% (1 stroke) 0 12% 2.1
DOAC for AF
or VTE
Curto-Garcia PV/ET/PMF/ 32 17 14 1 0 3% (1 mes­en­teric 0 0% 2.1
et al26 MDS-MPN ische­mia)
receiv­ing
DOAC for VTE
Serrao et al27 PV/ET/PMF 71† 26 21 14 10 0% — 0% 1
receiv­ing

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/448/1851806/448schieppati.pdf by guest on 13 December 2021

DOAC for AF
or VTE
Barbui et al28 PV/ET/PMF 442‡ 187 157 48 50 4.9% (2.1% pt-y) (AF) 1.5% (0.6% pt-y) (AF) 6.9% (3.0 pt-y) (AF) 1.7
receiv­ing 9.2% (4.5% pt-y) (VTE) 7.1% (3.4% pt-y) (VTE) 5.0% (2.3% pt-y) (VTE)
DOAC for AF
or VTE
*13 patients receiv­ing DOAC for AF, 4 for AF and stroke, and 8 for VTE.
†35 patients receiv­ing DOAC for AF; 36 for VTE.
‡203 patients receiv­ing DOAC for AF; 239 for VTE.
Apix, apixaban; dabig, dabigatran; edox, edoxaban; rivar, rivaroxaban.

more exten­sive infor­ma­tion on the rates of thrombohemor-
rhagic com­pli­ca­tions in this set­ting.28 Specifically, the inci­
dence of a first VTE event in patients receiv­ing a DOAC for AF
was 0.6% in patient-years, while the inci­dence of recur­rent
VTE in patients receiv­ing a DOAC for a prior VTE was 3.4% in
patient-years, which was no dif­fer­ent from the recur­rence inci­
dence observed in the VKA stud­ies.14-17 Moreover, annual rates of
major bleed­ing ranged from 2.3% to 3% in patient-years,28 also
sim­i­lar to VKAs. Therefore, on the basis of lim­ited avail­­able evi­
dence, DOACs and VKAs seem to have a com­pa­ra­ble risk/ben­
e­fit pro­file in the treat­ment and sec­ond­ary pre­ven­tion of VTE in
MPN patients. Finally, 2 recent small stud­ies tried to ret­ro­spec­
tively com­pare the out­comes of MPN patients treated with
VKAs or DOACs29,30 (Table 3). In the study by Huenerbein et al,
despite a higher relapse rate seen in the VKA group com­pared
to the DOAC group, throm­bo­sis-free sur­vival was no dif­fer­ent
between the 2 groups.29 In the study by Fedorov et al, the rates
of throm­bo­sis were also com­pa­ra­ble between the 2 anti­co­ag­
u­lant reg­i­mens. In both stud­ies the rate of major bleed­ing was
sim­i­lar for the 2 anti­co­ag­u­lants.29,30 Interestingly, a sig­nif­i­cantly
higher VTE recur­rence rate was observed after the dis­con­tin­u­
a­tion of either drug.30
In this set­ting it is impor­tant to recall that cytoreduction is
recommended in PV/ET who have a his­tory of throm­bo­sis or
are expe­ri­enc­ing a first VTE epi­sode in the fol­low-up.5 However,
recent stud­ ies show that despite the dem­ on­
strated effi­ cacy
of hydroxy­urea (HU) at cytoreduction to pre­vent pri­mary and
recur­rent arte­rial events,14,31 its action in the pre­ven­tion of first or
recur­rent venous throm­bo­sis is more lim­ited.32,33 Other cytore-
ductive or dis­ease-mod­i­fy­ing agents (ie, ruxolitinib, anagrelide,
inter­ feron alpha, and ropeginterferon) have proved valu­ able
alter­ na­tives to HU for dis­ ease con­ trol, although for most of
these drugs there is no con­trolled evi­dence show­ing their supe­
ri­or­ity over HU at pre­vent­ing VTE. Since the inci­dence of ATEs
and VTE remains high in MPN patients despite cytoreduction,
fur­ther ther­a­peu­tic pro­pos­als are needed, pos­si­bly addressing
addi­tional mech­a­nisms besides myelosuppression and targeting
other thrombogenic path­ways.7

Table 3. Retrospective stud­ies com­par­ing throm­bo­sis recur­rence and major bleed­ing in MPN patients receiv­ing VKA or DOAC

Overall throm­botic
Median
recur­rence VTE recur­rence Major bleed­ing
N on N on fol­low-up
Reference Study pop­u­la­tion VKA DOAC VKA DOAC VKA DOAC VKA DOAC (years)
Huenerbein PV/ET/PMF/ 45 26 48.8% 15.3% 24.4% 11.5% 8.88% 7.6% 3.2
et al29 MPN-U on sys­temic anticoagulation for
VTE or ATE
Fedorov PV/ET/PMF/ 31 22 19.4% 22.7% — — 6.4% 4.5% 1.2
et al30 MPN-U on sys­temic anticoagulation for
VTE or ATE
MPN-U, mye­lo­pro­lif­er­a­tive neo­plasm-unclas­si­fi­able: PV, plycythemia vera.

450  |  Hematology 2021  |  ASH Education Program

 When it comes to choos­ing an anti­co­ag­u­lant agent, the pros
and cons of treat­ment with VKAs or DOACs for MPN-asso­ci­ated
VTE can be sum­ma­rized as fol­lows:
VKA pros:
•  Larger stud­ies eval­u­at­ing effi­cacy and safety
•  Longer fol­low-up
•  Reduced VTE recur­rence
•  Reduced ATE recur­rence
•  Long-term pro­tec­tion from VTE recur­rence
VKA cons:
•  Laboratory mon­i­tor­ing of INR
•  Increased bur­den in case of life­long treat­ment
•  Higher bleed­ing risk in MPN than in non-MPN patients
•  Residual risk of recur­rence even with an INR in the ther­a­peu­tic
range
DOACs pros:
•  Easier to admin­is­ter
•  Routine lab­o­ra­tory mon­i­tor­ing is unnec­es­sary
•  Good effi­ cacy and safety pro­ file in stud­ies with can­
cer
patients
•  More appeal­ing than VKAs for indef­i­nite treat­ment
DOACs cons:
•  Limited stud­ies eval­u­at­ing effi­cacy and safety
•  Shorter fol­low-up
•  Higher bleed­ing risk in MPN than in non-MPN patients
•  Residual risk of recur­rence even dur­ing treat­ment
In con­clu­sion, the cur­rent evi­dence, although lim­ited, shows
sim­i­lar pat­terns in terms of the effi­cacy and safety of VKAs and
DOACs for the treat­ment and pre­ven­tion of recur­rent VTE in
MPN. While we wait for a ran­dom­ized com­par­i­son between
the 2 reg­i­mens, treat­ment deci­sions should be guided accord-
ing to indi­vid­ual fac­tors (ie, renal func­tion, bleed­ing risk pro­file,
con­com­i­tant med­i­ca­tions) as well as patient pref­er­ences. The
opti­mal dura­tion of anticoagulation is uncer­tain, but sub­stan­
tial evi­dence indi­cates that recur­rence risk is par­tic­u­larly high in
these patients even years after the index event. To help decide
on the dura­tion, it is wise to per­form a care­ful assess­ment of the
patient’s risk fac­tors for recur­rence (such as unpro­voked VTE,
prox­i­mal DVT, pul­mo­nary embo­li­za­tion, thrombophilia, etc) and
to plan a peri­od­i­cal reassessment of risk fac­tors for throm­bo­sis
and bleed­ing dur­ing the fol­low-up.
CLINICAL CASE (Con­t in­u ed)
We discussed pros and cons with our patient, who was rel­a­
tively young and in good form. We agreed to start HU and an
FXa inhib­i­tor for VTE treat­ment. This deci­sion was based on
the patient’s pref­er­ence for a less oner­ous reg­i­men and on our
expe­ri­ence with DOACs in can­cer patients. We aimed to pur­
sue a long-term treat­ment, con­sid­er­ing his unpro­voked prox­i­
mal DVT and PE and the pres­ence of addi­tional risk fac­tors such
as JAK2 muta­tion and inherited thrombophilia. We suspended
aspi­rin to reduce his bleed­ing risk. Nevertheless, we peri­od­i­
cally assess his throm­bo­sis and bleed­ing risk, will­ing to adjust
Prakash Singh Shekhawat
our treat­ment strat­e­gies in case of changes in the patient’s dis­
ease pat­tern or the avail­abil­ity of new evi­dence on anticoagu-
lation modal­i­ties in MPN.
Conflict-of-inter­est dis­clo­sure
Francesca Schieppati: no com­ pet­ ing finan­ cial inter­
ests to
declare.
Anna Falanga: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Francesca Schieppati: nothing to disclose.
Anna Falanga: nothing to disclose.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/448/1851806/448schieppati.pdf by guest on 13 December 2021
Correspondence
Francesca Schieppati, Department of Transfusion Medicine and
Hematology, Hospital Papa Giovanni XXIII, Piazza OMS n 1, Berga-
mo 24127, Italy; e-mail: fschieppati@gmail​­.com.
References
1. Falanga A, Marchetti M, Schieppati F. Prevention and man­ age­
ment of
throm­bo­sis in BCR/ABL-neg­a­tive mye­lo­pro­lif­er­a­tive neo­plasms. Hamo-
staseologie. 2021;41(1):48-57.
2. Hultcrantz M, Björkholm M, Dickman PW, et al. Risk for arte­rial and venous
throm­bo­sis in patients with mye­lo­pro­lif­er­a­tive neo­plasms: a pop­u­la­tion-
based cohort study. Ann Intern Med. 2018;168(5):317-325.
3. Rungjirajittranon T, Owattanapanich W, Ungprasert P, Siritanaratkul N,
Ruchutrakool T. A sys­tem­atic review and meta-anal­y­sis of the prev­a­lence of
throm­bo­sis and bleed­ing at diag­no­sis of Philadelphia-neg­a­tive mye­lo­pro­
lif­er­a­tive neo­plasms. BMC Cancer. 2019;19(1):184.
4. De Stefano V, Finazzi G, Barbui T. Antithrombotic ther­apy for venous throm­
bo­em­bo­lism in mye­lo­pro­lif­er­a­tive neo­plasms. Blood Cancer J. 2018;8(7):65.
5. Tefferi A, Barbui T. Polycythemia vera and essen­tial thrombocythemia: 2021
update on diag­no­sis, risk-strat­i­fi­ca­tion and man­age­ment. Am J Hematol.
2020;95(12):1599-1613.
6. Barbui T, Finazzi G, Carobbio A, et  al. Development and val­i­da­tion of an
International Prognostic Score of throm­bo­sis in World Health Organization-
essen­tial thrombocythemia (IPSET-throm­bo­sis). Blood. 2012;120(26):5128-
5133, quiz 5252.
7. Moliterno AR, Ginzburg YZ, Hoffman R. Clinical insights into the ori­gins
of throm­bo­sis in mye­lo­pro­lif­er­a­tive neo­plasms. Blood. 2021;137(9):1145-
1153.
8. Landolfi R, Marchioli R, Kutti J, et al; Euro­pean Collaboration on Low-Dose
Aspirin in Polycythemia Vera Investigators. Efficacy and safety of low-dose
aspi­rin in poly­cy­the­mia vera. N Engl J Med. 2004;350(2):114-124.
9. Italiano Studio Policitemia GRUPPO. Low-dose aspi­rin in polycythaemia
vera: a pilot study. Br J Haematol. 1997;97(2):453-456.
10. Alvarez-Larrán A, Cervantes F, Pereira A, et  al. Observation ver­sus anti-
platelet ther­apy as pri­mary pro­phy­laxis for throm­bo­sis in low-risk essen­tial
thrombocythemia. Blood. 2010;116(8):1205-1210, quiz 1387.
11. Squizzato A, Romualdi E, Passamonti F, Middeldorp S. Antiplatelet drugs for
polycythaemia vera and essen­tial thrombocythaemia. Cochrane Database
Syst Rev. 2013;(4):CD006503.
12. Marchioli R, Finazzi G, Specchia G, et al; CYTO-PV Collaborative Group.
Cardiovascular events and inten­sity of treat­ment in poly­cy­the­mia vera. N
Engl J Med. 2013;368(1):22-33.
13. Barbui T, De Stefano V, Falanga A, et al. Addressing and pro­pos­ing solu­tions
for unmet clin­i­cal needs in the man­age­ment of mye­lo­pro­lif­er­a­tive neo­
plasm-asso­ci­ated throm­bo­sis: a con­sen­sus-based posi­tion paper. Blood
Cancer J. 2019;9(8):61.
14. De Stefano V, Za T, Rossi E, et al; GIMEMA CMD-Working Party. Recurrent
throm­bo­sis in patients with poly­cy­the­mia vera and essen­tial thrombo-
cythemia: inci­dence, risk fac­tors, and effect of treat­ments. Haematolog-
ica. 2008;93(3):372-380.
15. De Stefano V, Ruggeri M, Cervantes F, et al. High rate of recur­rent venous
throm­bo­em­bo­lism in patients with mye­lo­pro­lif­er­a­tive neo­plasms and effect
of pro­phy­laxis with vita­min K antag­o­nists. Leukemia. 2016;30(10):2032-
2038.
DOACs vs VKAs in MPN-asso­ci­ated VTE treat­ment  |  451
16. Hernández-Boluda JC, Arellano-Rodrigo E, Cervantes F, et al; Grupo
Español de Enfermedades Mieloproliferativas Filadelfia Negativas. Oral
anticoagulation to pre­vent throm­bo­sis recur­rence in poly­cy­the­mia vera
and essen­tial thrombocythemia. Ann Hematol. 2015;94(6):911-918.
17. Wille K, Sadjadian P, Becker T, et al. High risk of recur­rent venous throm­bo­
em­bo­lism in BCR-ABL-neg­a­tive mye­lo­pro­lif­er­a­tive neo­plasms after ter­mi­
na­tion of anticoagulation. Ann Hematol. 2019;98(1):93-100.
18. Finazzi G, De Stefano V, Barbui T. Splanchnic vein throm­ bo­
sis in mye­
lo­pro­lif­er­a­tive neo­plasms: treat­ment algo­rithm 2018. Blood Cancer J.
2018;8(7):64.
19. Kearon C, Akl EA, Ornelas J, et al. Antithrombotic ther­apy for VTE dis­ease:
CHEST guide­line and expert panel report. Chest. 2016;149(2):315-352.
20. Raskob GE, van Es N, Verhamme P, et al; Hokusai VTE Cancer Investigators.
Edoxaban for the treat­ment of can­cer-asso­ci­ated venous throm­bo­em­bo­
lism. N Engl J Med. 2018;378(7):615-624.
21. Young AM, Marshall A, Thirlwall J, et  al. Comparison of an oral fac­tor Xa
inhib­i­tor with low molec­u­lar weight hep­a­rin in patients with can­cer with
venous throm­bo­em­bo­lism: results of a ran­dom­ized trial (SELECT-D). J Clin
Oncol. 2018;36(20):2017-2023.
22. Agnelli G, Becattini C, Meyer G, et al; Caravaggio Investigators. Apixaban
for the treat­ment of venous throm­bo­em­bo­lism asso­ci­ated with can­cer. N
Engl J Med. 2020;382(17):1599-1607.
23. Lyman GH, Carrier M, Ay C, et al. Amer­i­can Society of Hematology 2021
guide­lines for man­age­ment of venous throm­bo­em­bo­lism: pre­ven­tion and
treat­ment in patients with can­cer. Blood Adv. 2021;5(4):927-974.
24. Key NS, Khorana AA, Kuderer NM, et al. Venous throm­bo­em­bo­lism pro­phy­
laxis and treat­ment in patients with can­cer: ASCO clin­i­cal prac­tice guide­
line update. J Clin Oncol. 2020;38(5):496-520.
25. Ianotto JC, Couturier MA, Galinat H, et al. Administration of direct oral anti­
co­ag­u­lants in patients with mye­lo­pro­lif­er­a­tive neo­plasms. Int J Hematol.
2017;106(4):517-521.
26. Curto-Garcia N, Doyle AJ, Breen KA, et  al. Outcomes of patients receiv­
ing direct oral anti­co­ag­u­lants for mye­lo­pro­lif­er­a­tive neo­plasm-asso­ci­ated
452  |  Hematology 2021  |  ASH Education Program
venous throm­bo­em­bo­lism in a large ter­tiary cen­tre in the UK. Br J Haema-
tol. 2020;189(3):e79-e81.
27. Serrao A, Breccia M, Napolitano M, et al. A mul­ti­cen­ter real-life study on
anti­co­ag­u­lant treat­ment with direct oral anti­co­ag­u­lants in patients with Ph-
neg­a­tive mye­lo­pro­lif­er­a­tive neo­plasms. Am J Hematol. 2020;95(12):E329-
E332.
28. Barbui T, De Stefano V, Carobbio A, et  al. Direct oral anti­co­ag­u­lants for
mye­lo­pro­lif­er­a­tive neo­plasms: results from an inter­na­tional study on 442
patients. Blood. 2020;136(suppl 1):42-43.
29. Huenerbein K, Sadjadian P, Becker T, et  al. Direct oral anti­ co­ag­u­
lants
(DOAC) for pre­ven­tion of recur­rent arte­rial or venous throm­bo­em­bolic
events (ATE/VTE) in mye­lo­pro­lif­er­a­tive neo­plasms. Ann Hematol. 2021;
100(8):2015-2022.
30. Fedorov K, Goel S, Kushnir M, Billett HH. Direct oral anti­co­ag­u­lants for pre­
ven­tion of recur­rent throm­bo­sis in mye­lo­pro­lif­er­a­tive neo­plasms. Blood.
2019;134(suppl 1):4193.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/448/1851806/448schieppati.pdf by guest on 13 December 2021
31. Barbui T, Vannucchi AM, Finazzi G, et  al. A reappraisal of the ben­e­fit-risk
pro­file of hydroxy­urea in poly­cy­the­mia vera: a pro­pen­sity-matched study.
Am J Hematol. 2017;92(11):1131-1136.
32. De Stefano V, Rossi E, Carobbio A, et al. Hydroxyurea pre­vents arte­rial and
late venous throm­botic recur­rences in patients with mye­lo­pro­lif­er­a­tive
neo­plasms but fails in the splanch­nic venous dis­trict: pooled anal­y­sis of
1500 cases. Blood Cancer J. 2018;8(11):112.
33. Barbui T, De Stefano V, Ghirardi A, Masciulli A, Finazzi G, Vannucchi AM. Dif-
ferent effect of hydroxy­urea and phle­bot­omy on pre­ven­tion of arte­rial and
venous throm­bo­sis in poly­cy­the­mia vera. Blood Cancer J. 2018;8(12):124.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000318
 MPN: NEW DIRECTIONS

Novel therapies vs hematopoietic cell

transplantation in myelofibrosis: who, when, how?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/453/1851800/453england.pdf by guest on 13 December 2021

James England and Vikas Gupta
Princess Margaret Cancer Centre, University of Toronto, Toronto, Ontario, Canada

Myelofibrosis is one of the classical Philadelphia chromosome–negative myeloproliferative neoplasms characterized by

progressive marrow failure and chronic inflammation. Discovery of the JAK2 mutation paved the way for development
of small molecular inhibitors and further facilitated the research in understanding of molecular biology of the disease.
Development of novel medications and synergistic combinations with standard JAK inhibitor (JAKi) therapy may have the
potential to improve depth and duration of disease control and symptomatic benefit, whereas advancements in alloge-
neic hematopoietic stem cell transplantation (HCT) have improved tolerability and donor availability, allowing for more
patients to pursue this potentially curative therapy. The increase in options for medical therapy and changing risk profile
of HCT is leading to increased complexity in counseling patients on choice of management strategy. In this case-based
review, we summarize our approach to symptom-directed medical therapy, including the use of novel drugs and combi-
nation therapies currently under study in advanced clinical trials. We outline our recommendations for optimal timing of
HCT, including risk-adapted selection for early HCT as opposed to delayed HCT after upfront JAKi therapy, as well as the
use of pretransplant JAKi and alternative donor sources.

LEARNING OBJECTIVES
• Outline a risk-adapted approach for selection of upfront HCT vs nontransplant therapies for MF in chronic phase
• Overview the progress of symptom-directed JAKi therapy toward disease-modifying novel therapies for MF

Introduction ent but is limited by significant mortality and morbidity.

Myelofibrosis (MF) encompasses a group of clonal stem
cell disorders that are either primary myelofibrosis or arise
from a preceding polycythemia vera (post-PV MF) or essen-
tial thrombocythemia (ET; post-ET MF). Although somatic
mutations and disease biology may differ between primary
and secondary MF, the management approach remains
similar. Somatic mutations in myeloproliferative neoplasm
(MPN) driver (JAK2, CALR, MPL) and other myeloid malig-
nancy genes perpetuate increased cytokine expression,
inflammation, and fibrosis within the niche environment
of the bone marrow (BM).1 The clinical manifestations of
MF include cytokine-mediated constitutional symptoms
(night sweats, weight loss, fever), cytopenias from progres-
sive marrow failure, extramedullary hematopoiesis, and
mechanical symptoms from splenomegaly. MF is a heterog-
enous disease with variable rates of progression to death
and blastic transformation depending on underlying clini-
cal and genetic risk factors.
Allogeneic hematopoietic stem cell transplantation
(HCT) is the only therapy with potential for cure at pres-
Prakash Singh Shekhawat
Accurate estimation of risk to inform patient-centered
decision making is a priority concern in managing MF. The
approval of JAK inhibitor (JAKi) therapy with ruxolitinib
and, more recently, fedratinib provides an alternative
effective approach to manage constitutional and spleno-
megaly-related symptoms in patients with MF who are not
candidates or do not wish to pursue HCT. JAKi therapy is
not curative and has limited duration of clinical benefit
as half of patients experience treatment failure after 2 to
3 years.2,3 A variety of novel agents are in advanced clini-
cal trials for MF, and these emerging therapies may lead
to disease modification or improved depth and duration
of symptom control.4 Conversely, progress in support-
ive care, conditioning, graft-versus-host disease (GVHD)
prophylaxis, and increasing safety of alternative donor
sources may allow for more patients to benefit from HCT.
In this case-based review, we outline our approach and
framework for individualized decision making for patients
with MF on selection of HCT vs nontransplant strategies,
including emerging therapies through clinical trials.
Therapeutic landscape for transplant eligible MF | 453

Figure 1.  Timing options for consideration of HCT in MF.
CLINICAL CASES
Case 1: A 60-year old white woman had symp­toms of slowly
pro­gres­sive ane­mia, abdom­i­nal dis­com­fort, and early sati­ety.
Physical exam­in ­ a­tion showed a spleen pal­pa­ble 12 cm below the
cos­tal mar­gin (BCM). Hemoglobin was 95 g/L, white blood cell
count was 7.0 × 109/L, plate­lets were 150 × 109/L, and periph­eral
blood (PB) blasts were 1.2%. Bone mar­row biopsy (BMBx) spec­i­
men was con­sis­tent with diag­no­sis of overt pri­mary mye­lo­fi­bro­
sis with World Health Organization grade 3 fibro­sis and blasts
less than 5%. Cytogenetics showed nor­mal kar­yo­type. Targeted
next-gen­er­a­tion sequenc­ing dem­on­strated type 1 CALR muta­
tion as the sole abnor­mal­ity (variant allele frequency (VAF) 35%).
Case 2: A 63-year-old man of east Indian ori­gin with a long-
stand­ing diag­no­sis of PV lost the need for phle­bot­omy. He was
asymp­tom­atic, and phys­i­cal exam­i­na­tion showed pal­pa­ble
spleen 7 cm BCM. Complete blood count showed the fol­low­ing:
hemo­glo­bin, 130 g/L; white blood cell count, 15.3 × 109/L; plate­
lets, 47 × 109/L; and PB blasts, 2%. BMBx spec­i­men was con­sis­
tent with post-PV MF with World Health Organization grade 2
fibro­sis and blasts less than 5%. Cytogenetics dem­on­strated a
dele­tion of 11q in 15 of 20 meta­phases. The fol­low­ing path­o­genic
muta­tions were noted in next-gen­er­a­tion sequenc­ing stud­ies:
JAK2 V617F (VAF 80%), ASXL1 (VAF 27%), and IDH1 (VAF 36%).
Optimal tim­ing of HCT in MF?
Optimal tim­ing of HCT is not well defined, but cur­rent expert
opin­ions rec­om­mend HCT to patients with MF who have Dynam-
ic International Prognostic Scoring System (DIPSS) inter­me­di­ate-
454  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/453/1851800/453england.pdf by guest on 13 December 2021
2/high-risk dis­ease and inter­me­di­ate-1 risk with addi­tional risk
fac­tors.5,6 Transplant stud­ies have reported bet­ter out­comes in
patients treated ear­lier in the dis­ease course.7 However, the same
group of patients also has bet­ter and more dura­ble response
with JAKi ther­apy.8,9 To our knowl­edge, there is no pro­spec­tive
study to com­pare the out­comes of HCT vs nontransplant ther­a­
pies in MF, and lim­ited ret­ro­spec­tive stud­ies were either done
in the pre-JAKi era or had a very small pro­por­tion of patients
treated with JAKi.7,10 Lack of com­par­a­tive data in this area has led
to sig­nif­i­cant var­i­a­tions in prac­tice on appli­ca­tion of HCT in MF.
When discussing with a patient regard­ing the trans­plant tim­
ing, var­i­ous time points for trans­plant can be con­sid­ered (Figure 1):
a.  Upfront trans­plan­ta­tion: early HCT usu­ally within a few months
of diag­no­sis
b.  Delayed trans­plan­ta­tion: HCT is delayed as long as one can,
trans­plant offered at early signs of fail­ure of nontransplant
ther­apy
c.  Leukemic pro­gres­sion: HCT at time of trans­for­ma­tion to acute
phase (AP) (PB or BM blasts 10%-19%) or blast phase (BP) (PB
or BM blasts ≥20%)
There is no one-size-fits-all­approach in patients with MF;
how­ever, we strongly rec­om­mend against waiting for dis­ease to
prog­ress to AP/BP as out­comes after the HCT are poor and many
patients never make it to the trans­plant stage.11-13
Our goal in the begin­ning is to under­stand the nat­u­ral his­
tory of the dis­ ease based on best avail­­ able infor­ma­ tion on
clin­i­cal and genetic risk fac­tors. This infor­ma­tion helps in under­
stand­ing the expected sur­vival and the like­li­hood of response
and the dura­bil­ity of nontransplant ther­apy. Patients with an
Table 1.  Identifying high-risk patients with sur­vival less than 5 years in mye­lo­fi­bro­sis

Risk strat­i­fi­ca­tion model Risk group Median OS, y Pros Cons

DIPSS14 Intermediate-2 4.0 Easy appli­ca­bil­ity based on Does not include any infor­ma­tion
clin­i­cal/lab­o­ra­tory var­i­ables on genetic var­i­ables that may
High 1.5
have sig­nif­i­cant impact on the
nat­u­ral his­tory of the dis­ease
DIPSS +15 Intermediate-2 2.9 Includes cytopenias and ­ No muta­tional data
cyto­ge­netic data
High 1.3
MIPSS70*16 High 3.1 Includes impact of MPN driver No cyto­ge­netic data
genes and HMR† genes
MIPSS70 + 2.0*17 High 4.1 Cytogenetic data
Driver/HMR‡ genes

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/453/1851800/453england.pdf by guest on 13 December 2021

Very high 1.8
MPN per­son­al­ized risk§18 TP53/–17p/–5/–5q 2.4 Combination of Study heavily weighted toward
clin­ical/genetic/ PV/ET patients
cyto­ge­netic data Copy num­ber var­i­a­tion not
included in most clin­i­cal NGS
reports
*http:​­/​­/www​­.mipss70score​­.it​­/.
†HMR (ASXL1, IDH1/2, EZH2, SRSF2).
‡Includes ASXL1, IDH1/2, EZH2, SRSF2; adds U2AF1 Q157.
§
https:​­/​­/www​­.sanger​­.ac​­.uk​­/science​­/tools​­/progmod​­/progmod​­/.
NGS, next-gen­er­a­tion sequenc­ing.

expected sur­vival of less than 5 years and/or less like­li­hood
of dura­ble response from nontransplant ther­apy are defined
as hav­ing high-risk MF (Table 1). We rec­om­mend the option of
upfront HCT in these patients as part of shared deci­sion mak­ing
incor­po­rat­ing patient fit­ness, val­ues, and pref­er­ences, not­ing
that many patients will choose to delay or forgo trans­plant fol­
low­ing coun­sel­ing.19,20
Evolution of risk assess­ment in MF
Significant prog­ ress has been made in the past decade on
under­stand­ing the nat­u­ral his­tory of MF (Figure 2). Early prog­
nos­tic mod­els incor­po­rated only basic lab­o­ra­tory var­i­ables.21,22
Using clin­i­cal and lab­o­ra­tory param­e­ters, the International
Working Group for Myelofibrosis Research and Treatment
devel­oped the International Prognostic Scoring System valid at
the time of diag­no­sis of MF and fur­ther val­i­dated in the DIPSS
appli­ca­ble at any time dur­ing the dis­ease course.14,23 Mayo Clinic
inves­ti­ga­tors refined this score by adding trans­fu­sion requir­ing
ane­mia, throm­bo­cy­to­pe­nia, and cyto­ge­net­ics lead­ing to the
DIPSS plus score.15 The impact of MPN driver muta­tions (JAK2,
CALR, and MPL) sub­se­quently showed a more indo­lent course
asso­ci­ated with the type 1 CALR muta­tion, whereas other mye­
loid malig­nancy high molec­u­lar risk (HMR) muta­tions, includ­
ing ASXL1, EZH2, IDH1/2, and SRSF2, were linked with poorer
prog­no­sis.26-28 Mutational data alone have been used as the
basis for the Genetically Inspired Prognostic Scoring System,
whereas inte­ gra­
tion of all­the above fac­ tors resulted in the
Mutation-Enhanced International Prognostic Scoring System
(MIPSS70), which was fur­ ther refined to include cyto­ ge­ netic
risk in the MIPSS70 plus model.16,24 Mutations in the U2AF1 Q157
hotspot were included as an addi­tional HMR in the MIPSS70 plus
v2.0 (http:​­/​­/www​­.mipss70score​­.it).17
Using a large cohort of 2035 patients with MPN (1321 with
ET, 356 with PV, and 309 with pri­mary or sec­ond­ary MF and
49 with other MPNs), Grinfeld et al18 iden­ti­fied 8 geno­mic sub­
groups with dis­tinct clin­i­cal and prog­nos­tic fea­tures. This was
used to develop a per­son­al­ized MPN risk cal­cu­la­tor that incor­po­
rates clin­i­cal, cyto­ge­netic, and muta­tion data to pre­dict sur­vival
and leu­ke­mic trans­for­ma­tion (https:​­/​­/cancer​­.sanger​­.ac​­.uk​­/mpn​
­-multistage). A key obser­va­tion in this study was the poor sur­vival

Figure 2.  Timeline of prognostic risk models for MF and their components. Reproduced with permission from Davidson and Gupta.25
Lille,21 Cervantes,22 IPSS,23 DIPSS,14 DIPSS-plus,15 MIPSS70/MIPSS70 plus,16 GIPSS,24 MIPSS70 + v.2.0,17 and Personalized.18
Prakash Singh Shekhawat
Therapeutic land­scape for trans­plant eli­gi­ble MF | 455

Figure 3.  Management algorithm for transplant-eligible patients with MF in the chronic phase.
in patients with mutated TP53, a rare yet influ­en­tial muta­tion in
chronic phase MF but one whose prog­nos­tic sig­nif­i­cance had
not been pre­vi­ously described.
Mutation anal­y­sis in mye­loid neo­plasms is a rap­idly progress-
ing field, and inter­ac­tions of var­i­ous subclonal pop­u­la­tions and
their down­stream effects on clin­i­cal out­comes are just begin­ning
to be under­stood. The def­i­ni­tion of HMR muta­tions will likely
evolve fur­ther with inclu­sion of other can­di­date gene muta­tions,
such as the RAS path­way (NRAS, KRAS, and CBL) and NFE2,
which have been shown to con­fer poorer prog­no­sis in MF.29,30
Survival fol­ low­ing HCT has been shown to cor­ re­
late with
DIPSS cat­ eg
­ ory, but this does not include trans­ plant-spe­
cific
var­i­ables that may influ­ence clin­i­cal out­comes.7 The clin­i­cal-
molec­u­lar MF trans­plant scor­ing sys­tem, incor­po­rat­ing clin­i­cal
data, donor type, and muta­ tion sta­tus for ASXL1/CALR/MPL,
pre­dicts over­all sur­vival (OS) and nonrelapse mor­tal­ity (NRM) in
pri­mary and sec­ond­ary MF.31
We strongly rec­om­mend that a phy­si­cian should use the max­
i­mal avail­­able infor­ma­tion on clin­i­cal, lab­o­ra­tory, and genetic
param­ et­ers depending on the access to the tests. If muta­
tional test­ing is avail­­able, we rec­om­mend using the MIPSS70 or
MIPSS70 plus v2.0 and oth­er­wise use the DIPSS.
CLINICAL CASES (Con­t in­u ed)
In case 1, although the patient had a DIPSS score 3 (ie, intermediate-
2 risk), the tempo of the dis­ease was indo­lent likely due to the
type 1 CALR muta­tion with­out addi­tional somatic muta­tions or
cyto­ge­netic abnor­mal­i­ties reflected in a low MIPSS70 + v2.0 risk.
Although symp­tom­atic, the patient was at rel­a­tively lower risk of
456  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/453/1851800/453england.pdf by guest on 13 December 2021
sig­nif­i­cant dis­ease pro­gres­sion, and symp­toms were ame­na­ble
to JAKi ther­apy; there­fore, we would delay the HCT for such a
patient until time of JAKi fail­ure.The patient in case 2 had high-risk
genetic abnor­mal­i­ties, includ­ing an adverse cyto­ge­netic abnor­
mal­ity and 2 HMR muta­tions. Although the patient had no symp­
toms and was clin­i­cally well, there was high risk of dis­ease pro­
gres­sion or leu­ke­mic trans­for­ma­tion. Therefore, we would refer
such a patient for con­sid­er­ation of upfront HCT.
Our algo­rith­mic approach to use of HCT in trans­plant-eli­gi­ble
patients is sum­ma­rized in Figure 3.
Nontransplant ther­a­pies in MF
Splenomegaly and con­sti­tu­tional symp­toms
Although rec­om­men­da­tion for HCT is based on risk assess­ment,
cur­rent med­i­cal ther­apy is directed at man­age­ment of symp­
toms. Ruxolitinib has dem­ on­
strated improve­ments in spleno­
meg­aly, MF-related symp­tom bur­den, and health-related qual­ity
of life.32,33 Initial tri­als did not include patients with lower IPSS
risk scores, and sub­se­quent stud­ies have shown symp­tom ben­
e­fit from ruxolitinib in a lower-risk pop­u­la­tion.8 Rates of spleen
response cor­re­late with ruxolitinib dos­ing, and cytopenias are
the main dose-lim­it­ing tox­ic­ity. Other JAKi have been stud­ied
in phase 3 tri­als with fedratinib approved by the US Food and
Drug Administration for use in MF.34 Momelotinib and pacritinib
are other JAKi going through addi­tional phase 3 tri­als (Table 2)
and may have dif­fer­en­ti­at­ing value for patients with ane­mia and
severe throm­bo­cy­to­pe­nia, respec­tively. Factors that pre­dict a
shorter dura­tion of response to front-line JAKi ther­apy include
higher DIPSS score, trans­fu­sion depen­dence, num­ber of muta­
tions, and ASXL1/EZH2 muta­tions spe­cif­i­cally.9,35
 Table 2.  Ongoing phase 3 tri­als in mye­lo­fi­bro­sis

Class of investigational Population/key inclu­sion

Agent agent Trial n Phase 2 clin­i­cal ben­e­fit Primary out­come Key Secondary out­comes Comparator cri­te­ria
JAKi-naive patients
Pelabresib + BET inhib­i­tor MANIFEST-2 310 Spleen reduc­tion SVR ≥35% at 24 TSS at 24 weeks Placebo +
ruxolitinib Less ane­mia weeks ruxolitinib
BM fibro­sis reduc­tion
Navitoclax + BCL2 inhib­i­tor TRANSFORM-1 230 Spleen reduc­tion SVR ≥35% at 24 TSS at 24 weeks Placebo +
ruxolitinib weeks OS ruxolitinib
Reduction in BM fibro­sis
Parsaclisib + PI3Kδ inhib­i­tor LIMBER-313 440 Spleen reduc­tion SVR ≥35% at 24 TSS Placebo +
ruxolitinib weeks OS ruxolitinib
Luspatercept + ActRII ligand trap INDEPENDENCE 309 Transfusion inde­pen­dence/​ RBC trans­fu­sion Anemia improve­ment Placebo + JAKi Transfusion depen­dent
JAKi ane­mia improve­ment inde­pen­dence at Duration of ben­e­fit
24 weeks
Pacritinib JAKi PACIFICA 348 Spleen reduc­tion SVR ≥35% at 24 TSS Randomized 2:1 Platelets <50 000/µL
Better tol­er­a­bil­ity in weeks OS Physician selected Includes patients with prior
throm­bo­cy­to­pe­nic BAT JAKi expo­sure
patients
Jaktinib JAKi — 105 SVR ≥35% at 24 Transfusion depen­dence Hydroxyurea
weeks
Following first-line JAKi expo­sure or fail­ure
Momelotinib JAKi MOMENTUM 180 Anemia improve­ment TSS at 24 weeks Transfusion inde­pen­dence Randomized 2:1 Patients with ane­mia <100 g/L
Spleen response rate Danazol

Prakash Singh Shekhawat

Fedratinib JAKi FREEDOM-2 192 Spleen reduc­tion SVR ≥35% at 24 TSS Randomized 2:1
weeks OS BAT
Navitoclax + BCL2 inhib­i­tor TRANSFORM-2 330 Spleen reduc­tion SVR ≥35% at 24 TSS at 24 weeks BAT
ruxolitinib weeks OS
Reduction in BM fibro­sis
Parsaclisib + PI3Kδ inhib­i­tor LIMBER-304 212 Spleen reduc­tion SVR ≥35% 24 weeks TSS Placebo +
ruxolitinib OS ruxolitinib
Imetelstat Telomerase inhib­i­tor IMpactMF 320 Symptom score reduc­tion OS TSS at 24 weeks BAT
OS PFS
SVR ≥35% at 24 weeks
Reduction in BM fibro­sis
BAT, best avail­­able ther­apy; BCL2, B-cell lymphoma 2; BET, bromodomain and extraterminal; PFS, pro­gres­sion-free sur­vival; SVR, spleen vol­ume response; TSS, total symp­tom score.
Source: https:​­/​­/www​­.clinicaltrials​­.gov accessed on April 30, 2021.

Therapeutic land­scape for trans­plant eli­gi­ble MF | 457

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/453/1851800/453england.pdf by guest on 13 December 2021
 Deepening and prolonging the spleen and symp­ tom After 6 years, her spleen began to increase in size again, up
response is a pri­mary goal of many novel com­bi­na­tion strat­e­ to 15 cm BCM, with recur­rence of abdom­in ­ al pain and early sati­
gies adding addi­tive/syn­er­gis­tic agents to ruxolitinib in symp­ ety. The patient was started on fedratinib and had sig­nif­i­cant
tom­atic treat­ment-naive patients (Table 2). In addi­tion to the improve­ment in spleen size. Subsequently, this patient under­
con­sti­tu­tive acti­va­tion of JAK-STAT path­ways, there is evi­dence went 10/10 matched unre­lated donor (MUD) trans­plant using
that cyto­kine bur­den, dis­ease pro­gres­sion, and JAKi resis­tance reduced-inten­sity con­di­tion­ing and remains well after HCT.
can be affected through aber­rant acti­va­tion of other path­ways
that medi­ate cell pro­lif­er­a­tion, sur­vival, and cyto­kine sig­nal­
ing.1,4 Promising agents used in com­bi­na­tion with ruxolitinib
JAKi fail­ure
cur­ rently in phase 3 eval­ u­
a­
tion are sum­ ma­
rized in Table 2.
Prior stud­ies eval­u­at­ing out­comes fol­low­ing dis­con­tin­u­a­tion of
Pelabresib is a bromodomain and extra-ter­mi­nal domain inhib­
ruxolitinib report a median sur­vival of 11 to 16 months (Table 3),
i­tor that reduces nuclear fac­tor–κB–medi­ated inflam­ma­tion.
with very short sur­vival in those with AP/BP trans­for­ma­tion.39-42
Navitoclax is a BCL-2/BCL-XL inhib­ i­
tor that reduces prosur-
Although infor­ma­tive, we note that in real-world prac­tice, many

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/453/1851800/453england.pdf by guest on 13 December 2021

vival sig­nal­ing and pro­motes apo­pto­sis in the malig­nant clone.
patients con­tinue on ruxolitinib despite clear evi­dence of dis­
Parsaclisib, a phosphatidylinositol 3-kinase (PI3K) δ inhib­it­or,
ease pro­gres­sion due to the lack of alter­na­tive med­i­cal ther­apy.
reduces abnor­mal PI3K/mTOR/AKT acti­va­tion that pro­motes
With the approval of fedratinib and avail­abil­ity of novel com­bi­na­
inflam­ma­tion and JAKi resis­tance. In phase 2 stud­ies, com­bi­na­
tion strat­e­gies as part of clin­i­cal tri­als (Table 2), early rec­og­ni­tion
tions of these agents with ruxolitinib have exhibited good rates
of JAKi ther­apy fail­ure and its pat­tern is of par­am ­ ount impor­tance
of spleen reduc­tion and symp­tom score improve­ment, with
for selecting appro­pri­ate options for these patients. We use the
pelabresib and navitoclax containing reg­i­mens dem­on­strat­ing
cri­te­ria pro­posed by the Cana­dian MPN Group and clas­sify the
evi­dence of reduced BM fibro­sis in some patients.4 Pelabresib,
JAKi fail­ure by categories such as loss of spleen/symp­tom re-
navitoclax, and parsaclisib com­ bi­
na­tions with ruxolitinib are
sponse, severe cytopenias, or pro­gres­sion to AP/BP.43
being com­pared with ruxolitinib monotherapy in treat­ment-
Due to poor prog­no­sis of JAKi fail­ure, we rec­om­mend HCT in
naive patients in the phase 3 MANIFEST-2, TRANSFORM-1, and
any patient with first-line JAKi fail­ure (Figure 3), an approach sup-
LIMBER-313 tri­als, respec­tively.
ported by lim­ited ret­ro­spec­tive data dem­on­strat­ing improved
sur­vival with HCT com­pared with sup­port­ive care.40
Cytopenias
Management of cytopenias is an unmet need in MF, both in
oth­er­wise asymp­ tom­ atic patients and in those whose JAKi Pretransplant JAKi ther­apy
treat­ment is com­pli­cated by ane­mia and throm­bo­cy­to­pe­nia. In patients referred for HCT, mas­ sive spleno­ meg­ aly has been
Supportive care relies on trans­fu­sion of red blood cells and asso­ci­ated with delayed count recov­ery and pos­si­bly worse OS.44
plate­lets, whereas phar­ma­co­logic ther­apy is lim­ited to danazol Limited evi­ dence is avail­­
able to define an appro­ pri­
ate tar­get
and eryth­ro­poi­e­sis-stim­u­lat­ing agents. Single-agent pelabresib spleen size, although sple­nec­tomy is con­sid­ered in patients with
is being inves­ti­gated for improv­ing ane­mia as monotherapy for mas­sive spleno­meg­aly that is refrac­tory to med­i­cal ther­apy.44,45
patients who do not have an indi­ca­tion for JAKi ther­apy.4 En- Use of peritransplant JAKi has sig­nif­i­cant inter­est given the
hancement of late-stage eryth­ro­cyte devel­op­ment through the poten­tial for both dis­ease con­trol and impact on GVHD (Table
inhi­bi­tion of transforming growth fac­tor β/SMAD sig­nal­ing with 4). Preliminary results from the JAK-ALLO trial raised safety
activin recep­tor ligand traps, such as luspatercept, has dem­on­ con­cerns with tumor lysis syn­drome and car­dio­genic shock,
strated effi­cacy in improv­ing ane­mia and reduc­ing trans­fu­sion, poten­tially due to cyto­kine flare fol­low­ing sud­den ces­sa­tion
and it will be eval­ua ­ ted fur­ther in the phase 3 INDEPENDENCE
study (Table 2).
Table 3.  Survival and pre­dic­tive fac­tors fol­low­ing ruxolitinib
Asymptomatic patients fail­ure/dis­con­tin­u­a­tion
There is cur­rently no med­ic ­ al ther­apy that has clear ben­e­fit in
asymp­tom­atic patients; we only con­sider it in the set­ting of a Median OS (95% CI)
fol­low­ing ruxolitinib Factors predicting lower
clin­i­cal trial for an agent with dis­ease-mod­i­fy­ing poten­tial. Im-
Reference fail­ure, mo sur­vival
mune ther­ ap­ ies, includ­
ing inter­feron agents, and antifibrotic
agents such as PRM-151 have shown some prom­ise in early stud­ Newberry 14 (10-18) Platelets <260 at ruxolitinib start
et al. (2017)39 Platelets <100 at dis­con­tin­u­a­tion
ies but require con­fir­ma­tion in con­trolled tri­als.4,36-38 Emergent muta­tions at
dis­con­tin­u­a­tion
Kuykendall 13 Lack of sal­vage ther­apy
et al. (2018)40
CLINICAL CASES (Con­t in­u ed) Mascarenhas 11.1 (8.4-14.5) Age >65 years
In case 1, the patient responded well to upfront ruxolitinib with et al. (2020)41 Charlson Comorbidity Index
res­o­lu­tion of symp­toms and reduc­tion in her spleen size to a na- Palandri et al. 13.2 (8.0-22.7) Blast phase at fail­ure
dir of 6 cm BCM. Such a patient could also be con­sid­ered for an (2020)42 Hb <100 g/L
Peripheral blasts >1%
upfront com­bi­na­tion ther­apy trial.

458  |  Hematology 2021  |  ASH Education Program

 Table 4.  Prospective stud­ies of pre-HCT ruxolitinib ther­apy

Number of Number of Splenectomy SAE dur­ing

patients patients who prior to HCT, Ruxolitinib ruxolitinib
Reference recruited under­went HCT n (%) Taper ­dis­con­tin­u­a­tion taper Conditioning OS at 2 y, % NRM at 2 y, % GF, % GVHD
Gupta et al. (2019)47 21 19* 0 (0) 4 days 1 day prior to start None RIC (Flu-Bu) 63 28% 16 aGVHD grade 3-4, 16%
5 MSD of con­di­tion­ing cGVHD 76% at 2 y
14 UD
Salit et al. (2020)48 34 28† 3 (11) 5 mg every After 2-3 days of None Multiple 86 7% 0 aGVHD grade 3-4, 22%
14 MRD 3 days con­di­tion­ing cGVHD, 41%
11 UD ther­apy
3 UCB
Robin et al. (2021)46 64 59‡ 19 (30) 15 days (n = 17) 1 day prior to start TLS 3§ RIC (Flu-Mel) 55 46% 0 aGVHD grade 3-4, 44%
18 MSD of con­di­tion­ing RWS 3 23% MSD cGVHD 37% at 2 y
32 MUD Heart fail­ure 50% MUD
14 UD 9/10 77% UD 9/10
None (n = 42) 1 day prior to start Heart fail­ure
of con­di­tion­ing
*One patient with dis­ease pro­gres­sion to AML, 1 patient with sud­den death dur­ing ruxolitinib phase.
†Three patients with pro­gres­sion to AML dur­ing ruxolitinib phase, 3 patients not meet­ing cri­te­ria for trans­plant.
‡One donor with­drawal. One patient con­tin­ued tak­ing ruxolitinib, 3 patients died with­out HCT; 2 with heart fail­ure fol­low­ing sple­nec­tomy.
§No TLS fol­low­ing pro­to­col amend­ment to abrupt ruxolitinib dis­con­tin­u­a­tion.
aGVHD, acute GVHD; AML, acute meloid leukemia; cGVHD, chronic GVHD; Flu-Bu, fludarabine/busulfan; Flu-Mel, fludarabine-melphalan; GF, graft fail­ure; RIC, reduced-inten­sity con­di­tion­ing;
RWS, ruxolitinib with­drawal syn­drome; SAE, sig­nif­i­cant adverse event; TLS, tumor lysis syn­drome; UCB, umbilical cord blood.

Prakash Singh Shekhawat

Table 5.  HCT using alter­na­tive donors in mye­lo­fi­bro­sis

Reference Donor type n Conditioning GVHD ­pro­phy­laxis OS (95% CI) NRM (95% CI) Graft fail­ure GVHD
55
Raj et al. (2019) Mismatched related donor 56 MAC (70%)/RIC (30%) PTCy in 79% 56% (41%-70%) at 2 y 38% (24%-51%) at 2 y Primary 9% aGVHD* 28%
Secondary 13% cGVHD 45% at 1 y
Murata et al. (2019)56 Unrelated cord blood 13 — — 48% at 1 y 41% (22%-60%) at 1 y — aGVHD* 31%
27% at 4 y 62% (35%-81%) at 4 y cGVHD 15% at 1 y
Kunte et al. (2020)54 Haploidentical 58 NMA (52%)/RIC (45%) PTCy in 100% 69% (55%-80%) at 2 y 21% (11%-34%) at 2 y Primary 9% aGVHD* 44% at 6 m
cGVHD 31% at 2 y
*Acute GVHD encompasses grades II to IV.
MAC, myeloablative con­di­tion­ing; NMA, non-myeloablative; PTCy, posttransplant cyclo­phos­pha­mide.

Therapeutic land­scape for trans­plant eli­gi­ble MF | 459

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/453/1851800/453england.pdf by guest on 13 December 2021
of ruxolitinib, neces­ si­
tat­
ing a pause on the study.46 How-
ever, sub­se­quent tri­als do not sug­gest sub­stan­tial risk to this
approach.47,48 The JAK-ALLO results dem­ on­strated improved
safety, with no tumor lysis syn­drome or ruxolitinib with­drawal
syn­drome, fol­low­ing pro­to­col amend­ment to an abrupt ruxoli-
tinib dis­con­tin­u­a­tion prior to con­di­tion­ing.46 The impact of pre-
transplant JAKi on posttransplant out­comes, includ­ing GHVD,
is not clear, with ret­ro­spec­tive stud­ies dem­on­strat­ing sim­i­lar
out­comes in those with or with­out prior JAKi ther­apy.49 We rec­
om­mend using JAKi ther­apy prior to HCT in patients with sig­nif­
i­cant MF-related symp­toms (Figure 3), with a taper­ing sched­ule
lead­ing up to, or a few days after, the start of con­di­tion­ing to
avoid any with­drawal symp­toms.46,47,50
CLINICAL CASES (Con­t in­u ed)
In case 2, the patient was referred for trans­plant, but no fully
matched sib­ ling donors (MSDs) or unre­ lated donors (UDs)
could be iden­ti­fied. The patient pre­ferred not to pur­sue hap-
loidentical trans­ plant ini­
tially. However, there was increase
in blast counts in PB to 8% within 6 months, and the patient
opted for haploidentical HCT with his son as a donor at that
time point.
Role of alter­na­tive donor trans­plant in MF
HCT from a matched sib­ling or unre­lated donor is the pre­ferred
source for HCT in MF, with many early stud­ies dem­on­strat­ing
infe­rior out­comes using a mismatched unre­lated donor.51,52 Re-
cent reporting trends from the Center for International Blood
and Marrow Transplant Research show slowly increas­ing adap­
tion in the use of haploidentical donors in MF as advance­ments
in sup­port­ive care and the pre­ven­tion of GVHD, includ­ing the
use of posttransplant cyclo­phos­pha­mide, have allowed for im-
proved safety (Table 5).25 A major lim­i­ta­tion of all­these stud­ies
is small sam­ple size. Bregante et al53 report the increas­ing use
of haploidentical donors from 2011 to 2014 com­pared with 2000
to 2010, with sim­i­lar out­comes com­pared with MSDs among the
alter­na­tive donor group (77% haploidentical) in the later time
period. Initial results from Kunte et al54 also show encour­ag­ing
out­comes in a ret­ro­spec­tive cohort of haploidentical HCT with
posttransplant cyclo­ phos­ pha­
mide. Data fol­ low­ing the use of
related donors with 2 or more anti­gen mis­matches have been
reported from the Euro­ pean Society for Blood and Marrow
Transplantation reg­is­try dem­on­strat­ing rea­son­able 2-year OS
(56%) and GVHD rates but higher rates of graft fail­ure (22%) and
NRM at 2 years (38%).55 Use of unre­lated cord blood also pres­
ents sim­i­lar con­cerns, with a report from the Japan Society for
Hematopoietic Cell Transplantation dem­on­strat­ing lower rates
of engraft­ment, high NRM, and only 27% 4-year OS.56 Limited
data on haploidentical trans­plants sup­port the poten­tial use in
selected patients, although many patients may opt for med­i­
cal ther­apy when coun­seled on the added risks asso­ci­ated with
alter­na­tive donors.15 Ongoing large stud­ies through the Center
for International Blood and Marrow Transplant Research com­
par­ing out­comes of haploidentical trans­plants vs MUDs and mis-
matched donors should fur­ther clar­ify and sup­port the posi­tion­
ing of haploidentical trans­plants in the man­age­ment of MF (Tania
Jain, Johns Hopkins University, per­sonal com­mu­ni­ca­tion, 2021).
460  |  Hematology 2021  |  ASH Education Program
Conclusion
Advancements in both med­i­cal ther­apy and HCT opti­mi­za­tion
have improved and broad­ened ther­a­peu­tic options for patients
with MF. We rec­om­mend a risk-adapted strat­egy for tim­ing of
HCT and symp­tom-directed approach to med­i­cal ther­apy. These
rec­ om­men­ da­tions should be incor­ po­rated as part of shared
deci­sion mak­ing encompassing patient pref­er­ences and val­ues
to decide on an indi­vid­u­al­ized approach in each patient. Where
pos­si­ble, efforts should be made to enroll these patients in pro­
spec­tive tri­als or reg­is­tries for bet­ter under­stand­ing of com­par­
a­tive out­comes.
Acknowledgments
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/453/1851800/453england.pdf by guest on 13 December 2021
This pro­ject was supported by a grant from the Elizabeth and
Tony Comper Foundation and the Princess Margaret Cancer
Centre Foundation (V.G.).
Conflict-of-inter­est dis­clo­sure
James England: No com­pet­ing finan­cial inter­ests to declare.
Vikas Gupta: Consultancy and advi­sory board: Novartis, BMS-
Celgene, Sierra Oncology, Pfizer, Abb Vie, Roche.
Off-label drug use
James England: Off-label use of ruxolitinib in the peri-transplant
period is discussed in the article.
Vikas Gupta: Off-label use of ruxolitinib in the peri-transplant pe-
riod is discussed in the article.
Correspondence
Vikas Gupta, Princess Margaret Cancer Centre, University of
Toronto, 700 University Avenue, 6-326, Toronto, Ontario, M5G
1Z5 Canada; e-mail: vikas​­.gupta@uhn​­.ca.
References
1. Koschmieder S, Chatain N. Role of inflam­ma­tion in the biol­ogy of mye­lo­pro­
lif­er­a­tive neo­plasms. Blood Rev. 2020;42(suppl 1):100711.
2. Harrison CN, Vannucchi AM, Kiladjian J-J, et  al. Long-term find­ings from
COMFORT-II, a phase 3 study of ruxolitinib vs best avail­­able ther­apy for
mye­lo­fi­bro­sis. Leukemia. 2016;30(8):1701-1707.
3. Verstovsek S, Mesa RA, Gotlib J, et al; COMFORT-I inves­ti­ga­tors. Long-
term treat­ ment with ruxolitinib for patients with mye­ lo­
fi­
bro­
sis: 5-year
update from the ran­dom­ized, dou­ble-blind, pla­cebo-con­trolled, phase 3
COMFORT-I trial. J Hematol Oncol. 2017;10(1):55.
4. Bankar A, Gupta V. Investigational non-JAK inhib­i­tors for chronic phase
mye­lo­fi­bro­sis. Expert Opin Investig Drugs. 2020;29(5):461-474.
5. Gupta V, Hari P, Hoffman R. Allogeneic hema­to­poi­etic cell trans­plan­ta­
tion for mye­lo­fi­bro­sis in the era of JAK inhib­i­tors. Blood. 2012;120(7):1367-
1379.
6. Kröger NM, Deeg JH, Olavarria E, et  al. Indication and man­age­ment of
allo­ge­neic stem cell trans­plan­ta­tion in pri­mary mye­lo­fi­bro­sis: a con­sen­
sus pro­ cess by an EBMT/ELN inter­ na­tional work­ ing group. Leukemia.
2015;29(11):2126-2133.
7. Gowin K, Ballen K, Ahn KW, et al. Survival fol­low­ing allo­ge­neic trans­plant in
patients with mye­lo­fi­bro­sis. Blood Adv. 2020;4(9):1965-1973.
8. Gupta V, Griesshammer M, Martino B, et  al. Analysis of pre­ dic­tors of
response to ruxolitinib in patients with mye­ lo­
fib
­ ro­
sis in the phase 3b
expanded-access JUMP study. Leuk Lymphoma. 2021;62(4):918-926.
9. Spiegel JY, McNamara C, Kennedy JA, et al. Impact of geno­mic alter­ations
on out­comes in mye­lo­fi­bro­sis patients under­go­ing JAK1/2 inhib­­i­tor ther­
apy. Blood Adv. 2017;1(20):1729-1738.
10. Kröger N, Giorgino T, Scott BL, et al. Impact of allo­ge­neic stem cell trans­
plan­ta­tion on sur­vival of patients less than 65 years of age with pri­mary
mye­lo­fi­bro­sis. Blood. 2015;125(21):3347-3350, quiz 3364.
11. McNamara C, Spiegel J, Xu W, et  al. Outcomes fol­ low­ ing fail­
ure of
JAK1/2 inhib­i­tor ther­apy in patients with mye­lo­fi­bro­sis is depen­dent
 on the pat­tern of fail­ure. Clin Lymphoma Myeloma Leuk. 2019;19(suppl
1):S351–S352.
12. Gupta V, Kennedy JA, Capo-Chichi J-M, et al. Genetic fac­tors rather than
blast reduc­ tion deter­ mine out­ comes of allo­ ge­
neic HCT in BCR-ABL-­​
neg­a­tive MPN in blast phase. Blood Adv. 2020;4(21):5562-5573.
13. Gupta V, Kim S, Hu Z-H, et  al. Comparison of out­comes of HCT in blast
phase of BCR-ABL1-MPN with de novo AML and with AML fol­low­ing MDS.
Blood Adv. 2020;4(19):4748-4757.
14. Passamonti F, Cervantes F, Vannucchi AM, et  al. A dynamic prog­nos­tic
model to pre­dict sur­vival in pri­mary mye­lo­fi­bro­sis: a study by the IWG-MRT
(International Working Group for Myeloproliferative Neoplasms Research
and Treatment). Blood. 2010;115(9):1703-1708.
15. Gangat N, Caramazza D, Vaidya R, et  al. DIPSS plus: a refined Dynamic
International Prognostic Scoring System for pri­ mary mye­ lo­
fi­
bro­sis that
incor­po­rates prog­nos­tic infor­ma­tion from kar­yo­type, plate­let count, and
trans­fu­sion sta­tus. J Clin Oncol. 2011;29(4):392-397.
16. Guglielmelli P, Lasho TL, Rotunno G, et  al. MIPSS70: muta­tion-enhanced
inter­na­tional prog­nos­tic score sys­tem for trans­plan­ta­tion-age patients
with pri­mary mye­lo­fi­bro­sis. J Clin Oncol. 2018;36(4):310-318.
17. Tefferi A, Guglielmelli P, Lasho TL, et al. MIPSS70 + ver­sion 2.0: muta­tion and
kar­yo­type-enhanced inter­na­tional prog­nos­tic scor­ing sys­tem for pri­mary
mye­lo­fi­bro­sis. J Clin Oncol. 2018;36(17):1769-1770.
18. Grinfeld J, Nangalia J, Baxter EJ, et al. Classification and per­son­al­ized prog­
no­sis in mye­lo­pro­lif­er­a­tive neo­plasms. N Engl J Med. 2018;379(15):1416–
1430.
19. Walker AR. How to approach shared deci­sion mak­ing when deter­min­ing
con­sol­i­da­tion, main­te­nance ther­apy, and trans­plan­ta­tion in acute mye­loid
leu­ke­mia. Hematology Am Soc Hematol Educ Program. 2020;2020(1):51–
56.
20. Smith E, Huang J, Viswabandya A, et  al. Association of fac­tors influ­enc­
ing selec­tion of upfront hema­to­poi­etic cell trans­plan­ta­tion ver­sus non-
trans­plant ther­a­pies in mye­lo­fi­bro­sis: trans­plant deci­sions in MF. Transplant
Cell Ther. 2021;27(7):600.e1-600.e8.
21. Dupriez B, Morel P, Demory JL, et al. Prognostic fac­tors in agnogenic mye­
loid meta­pla­sia: a report on 195 cases with a new scor­ing sys­tem. Blood.
1996;88(3):1013-1018.
22. Cervantes F, Barosi G, Demory JL, et  al. Myelofibrosis with mye­loid
meta­pla­sia in young indi­vid­u­als: dis­ease char­ac­ter­is­tics, prog­nos­tic
fac­tors and iden­ti­fi­ca­tion of risk groups. Br J Haematol. 1998;102(3):
684-690.
23. Cervantes F, Dupriez B, Pereira A, et al. New prog­nos­tic scor­ing sys­tem
for pri­mary mye­lo­fi­bro­sis based on a study of the International Working
Group for Myelofibrosis Research and Treatment. Blood. 2009;113(13):
2895-2901.
24. Tefferi A, Guglielmelli P, Nicolosi M, et al. GIPSS: genet­i­cally inspired prog­
nos­tic scor­ing sys­tem for pri­mary mye­lo­fi­bro­sis. Leukemia. 2018;32(7):1631–
1642.
25. Davidson MB, Gupta V. Application of stem cell ther­apy in mye­lo­fi­bro­sis.
Hematol Oncol Clin North Am. 2021;35(2):391-407.
26. Guglielmelli P, Rotunno G, Fanelli T, et  al. Validation of the dif­fer­en­tial
prog­nos­tic impact of type 1/type 1-like ver­sus type 2/type 2-like CALR
muta­tions in mye­lo­fi­bro­sis. Blood Cancer J. 2015;5(10):e360.
27. Vannucchi AM, Lasho TL, Guglielmelli P, et al. Mutations and prog­no­sis in
pri­mary mye­lo­fi­bro­sis. Leukemia. 2013;27(9):1861-1869.
28. Rumi E, Pietra D, Pascutto C, et al; Associazione Italiana per la Ricerca sul
Cancro Gruppo Italiano Malattie Mieloproliferative Investigators. Clinical
effect of driver muta­tions of JAK2, CALR, or MPL in pri­mary mye­lo­fi­bro­sis.
Blood. 2014;124(7):1062-1069.
29. Coltro G, Rotunno G, Mannelli L, et  al. RAS/CBL muta­tions pre­dict resis­
tance to JAK inhib­i­tors in mye­lo­fi­bro­sis and are asso­ci­ated with poor prog­
nos­tic fea­tures. Blood Adv. 2020;4(15):3677-3687.
30. Marcault C, Zhao LP, Maslah N, et  al. Impact of NFE2 muta­tions on AML
trans­for­ma­tion and over­all sur­vival in patients with mye­lo­pro­lif­er­a­tive neo­
plasms (MPN) [published online 4 May 2021]. Blood.
31. Gagelmann N, Ditschkowski M, Bogdanov R, et al. Comprehensive clin­i­cal-
molec­u­lar trans­plant scor­ing sys­tem for mye­lo­fi­bro­sis under­go­ing stem
cell trans­plan­ta­tion. Blood. 2019;133(20):2233-2242.
32. Harrison C, Kiladjian J-J, Al-Ali H-K, et  al. JAK inhi­bi­tion with ruxolitinib
ver­sus best avail­­able ther­apy for mye­lo­fi­bro­sis. N Engl J Med. 2012;366(9):
787–798.
33. Verstovsek S, Mesa RA, Gotlib J, et al. A dou­ble-blind, pla­cebo-con­trolled
trial of ruxolitinib for mye­lo­fi­bro­sis. N Engl J Med. 2012;366(9):799-807.
Prakash Singh Shekhawat
34. Mullally A, Hood J, Harrison C, Mesa R. Fedratinib in mye­lo­fi­bro­sis. Blood
Adv. 2020;4(8):1792-1800.
35. Patel KP, Newberry KJ, Luthra R, et al. Correlation of muta­tion pro­file and
response in patients with mye­lo­fi­bro­sis treated with ruxolitinib. Blood.
2015;126(6):790-797.
36. Sørensen AL, Mikkelsen SU, Knudsen TA, et al. Ruxolitinib and inter­feron-α2
com­bi­na­tion ther­apy for patients with poly­cy­the­mia vera or mye­lo­fi­bro­sis:
a phase II study. Haematologica. 2020;105(9):2262-2272.
37. Kiladjian J-J, Soret-Dulphy J, Resche-Rigon M, et al. Ruxopeg, a multi-cen­
ter Bayes­ian phase 1/2 adap­tive ran­dom­ized trial of the com­bi­na­tion of
ruxolitinib and pegylated inter­feron alpha 2a in patients with mye­lo­pro­lif­
er­a­tive neo­plasm (MPN)–asso­ci­ated mye­lo­fi­bro­sis. Blood. 2018;132(suppl
1):581.
38. Verstovsek S, Talpaz M, Wadleigh M, et  al. S828: a ran­dom­ized, dou­ble
blind phase 2 study of 3 dif­fer­ent doses of PRM-151 in patients with mye­
lo­fi­bro­sis who were pre­vi­ously treated with or inel­i­gi­ble for ruxolitinib.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/453/1851800/453england.pdf by guest on 13 December 2021
HemaSphere. 2019;3(suppl 1):367.
39. Newberry KJ, Patel K, Masarova L, et al. Clonal evo­lu­tion and out­comes
in mye­lo­fi­bro­sis after ruxolitinib dis­con­tin­u­a­tion. Blood. 2017;130(9):1125-
1131.
40. Kuykendall AT, Shah S, Talati C, et al. Between a rux and a hard place: eval­u­
at­ing sal­vage treat­ment and out­comes in mye­lo­fi­bro­sis after ruxolitinib dis­
con­tin­u­a­tion. Ann Hematol. 2018;97(3):435-441.
41. Mascarenhas J, Mehra M, He J, Potluri R, Loefgren C. Patient char­ac­ter­is­
tics and out­comes after ruxolitinib dis­con­tin­u­a­tion in patients with mye­
lo­fi­bro­sis. J Med Econ. 2020;23(7):721-727.
42. Palandri F, Breccia M, Bonifacio M, et al. Life after ruxolitinib: rea­sons for dis­
con­tin­u­a­tion, impact of dis­ease phase, and out­comes in 218 patients with
mye­lo­fi­bro­sis. Cancer. 2020;126(6):1243-1252.
43. Gupta V, Cerquozzi S, Foltz L, et  al. Patterns of ruxolitinib ther­apy fail­
ure and its man­age­ment in mye­lo­fi­bro­sis: per­spec­tives of the Cana­dian
Myeloproliferative Neoplasm Group. JCO Oncol Pract. 2020;16(7):351-
359.
44. Malato A, Rossi E, Tiribelli M, Mendicino F, Pugliese N. Splenectomy in
mye­lo­fi­bro­sis: indi­ca­tions, effi­cacy, and com­pli­ca­tions. Clin Lymphoma
Myeloma Leuk. 2020;20(9):588-595.
45. Polverelli N, Mauff K, Kröger N, et al. Impact of spleen size and sple­nec­
tomy on out­comes of allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion
for mye­lo­fi­bro­sis: a ret­ro­spec­tive anal­y­sis by the chronic malig­nan­cies
work­
­ ing party on behalf of Euro­ pean soci­ ety for blood and mar­ row
trans­plan­ta­tion (EBMT). Am J Hematol. 2021;96(1):69-79.
46. Robin M, Porcher R, Orvain C, et al. Ruxolitinib before allo­ge­neic hema­to­
poi­etic trans­plan­ta­tion in patients with mye­lo­fi­bro­sis on behalf SFGM-TC
and FIM groups. Bone Marrow Transplant. 2021;56(8):1-12.
47. Gupta V, Kosiorek HE, Mead A, et  al. Ruxolitinib ther­ apy followed by
reduced-inten­sity con­di­tion­ing for hema­to­poi­etic cell trans­plan­ta­tion for
mye­lo­fi­bro­sis: mye­lo­pro­lif­er­a­tive dis­or­ders research Consortium 114 study.
Biol Blood Marrow Transplant. 2019;25(2):256-264.
48. Salit RB, Scott BL, Stevens EA, Baker KK, Gooley TA, Deeg HJ. Pre-
hema­ to­poi­etic cell trans­ plant ruxolitinib in patients with pri­ mary
and sec­ond­ary mye­lo­fi­bro­sis. Bone Marrow Transplant. 2020;55(1):
70-76.
49. Shahnaz Syed Abd Kadir S, Christopeit M, Wulf G, et al. Impact of ruxoli-
tinib pre­treat­ment on out­comes after allo­ge­neic stem cell trans­plan­ta­tion
in patients with mye­lo­fi­bro­sis. Eur J Haematol. 2018;101(3):305-317.
50. Shanavas M, Popat U, Michaelis LC, et al. Outcomes of allo­ge­neic hema­
to­poi­etic cell trans­plan­ta­tion in patients with mye­lo­fi­bro­sis with prior
expo­ sure to Janus kinase ½ inhib­ i­tors. Biol Blood Marrow Transplant.
2016;22(3):432-440.
51. Kröger N, Holler E, Kobbe G, et  al. Allogeneic stem cell trans­plan­ta­tion
after reduced-inten­sity con­di­tion­ing in patients with mye­lo­fi­bro­sis: a
pro­spec­tive, mul­ti­cen­ter study of the Chronic Leukemia Working Party
of the Euro­ pean Group for Blood and Marrow Transplantation. Blood.
2009;114(26):5264-5270.
52. Gupta V, Malone AK, Hari PN, et al. Reduced-inten­sity hema­to­poi­etic cell
trans­plan­ta­tion for patients with pri­mary mye­lo­fi­bro­sis: a cohort anal­y­sis
from the Center for International Blood and Marrow Transplant Research.
Biol Blood Marrow Transplant. 2014;20(1):89-97.
53. Bregante S, Dominietto A, Ghiso A, et  al. Improved out­come of alter­na­
tive donor trans­plan­ta­tions in patients with mye­lo­fi­bro­sis: from unre­
lated to haploidentical fam­ ily donors. Biol Blood Marrow Transplant.
2016;22(2):324-329.
Therapeutic land­scape for trans­plant eli­gi­ble MF | 461
54. Kunte SJ, Rybicki L, Viswabandya A, et  al. Haploidentical allo­ ge­
neic
hema­to­poi­etic cell trans­plan­ta­tion with post-trans­plant cyclo­phos­pha­mide
in patients with mye­lo­fi­bro­sis: a multi-insti­tu­tional expe­ri­ence. Blood.
2020;136(suppl 1):­33-34.
55. Raj K, Eikema D-J, McLornan D-P, et al. Family mismatched allo­ge­neic stem
cell trans­plan­ta­tion for mye­lo­fi­bro­sis: report from the Chronic Malignancies
Working Party of Euro­pean Society for Blood and Marrow Transplantation.
Biol Blood Marrow Transplant. 2019;25(3):522-528.
462  |  Hematology 2021  |  ASH Education Program
56. Murata M, Takenaka K, Uchida N, et al. Comparison of out­comes of allo­
ge­neic trans­plan­ta­tion for pri­mary mye­lo­fi­bro­sis among hema­to­poi­etic
stem cell source groups. Biol Blood Marrow Transplant. 2019;25(8):1536-
1543.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000279
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/453/1851800/453england.pdf by guest on 13 December 2021
 MPN: NEW DIRECTIONS

Running interferon interference in treating

PV/ET: meeting unmet needs

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/463/1852047/463traxler.pdf by guest on 13 December 2021

Elizabeth A. Traxler and Elizabeth O. Hexner
Department of Medicine and Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA

Enthusiasm about interferons for the treatment of myeloproliferative neoplasms has recently arisen. How does a non-
targeted therapy selectively target the malignant clone? Many foundational questions about interferon treatment are
unanswered, including who, when, and for how long do we treat. Using an individual case, this review touches on gaps
in risk assessment in polycythemia vera (PV) and essential thrombocythemia (ET) and the history of treatment with inter-
ferons. How is it that this proinflammatory cytokine effectively treats ET and PV, themselves proinflammatory states? We
summarize existing mechanistic and clinical data, the molecular context as a modifier for treatment response, the estab-
lishment of treatment goals, and the challenges that lie ahead.

LEARNING OBJECTIVES
• Define treatment goals in PV/ET
• Summarize molecular responses using IFN in PV/ET

overall prognosis of her disease and wonders about the

CLINICAL CASE
A 32­year­old woman with at least 2 years of blood count
abnormalities is referred to hematology. Two years prior,
early in a complicated pregnancy, she had an abnormal
complete blood count: white blood cell count, 12.7 thou/µL;
hemoglobin, 16g/dL; mean corpuscular volume, 82 fL; and
platelets, 415 thou/µL. She was admitted with preeclamp­
sia 25 weeks into her pregnancy and was closely observed,
treated with low­dose aspirin, and delivered at 32 weeks.
Her newborn daughter was born small for gestational age.
When first evaluated by hematology 19 months later, she
described chronic itching for 7 to 8 years, worse after a
shower and absent during pregnancy but present again
and mild and manageable. She had no visual symptoms or
headaches and no history of clots. Medications included
an oral contraceptive. A repeat complete blood count
showed persistent abnormalities: WBC, 11.3 tho/µL; hemo­
globin, 17.9g/dL; mean corpuscular volume, 76 fL; platelets,
692 tho/µL. Additional testing, including JAK2V617F muta­
tion analysis, detected a mutation at an allele frequency
of 32.6%. Serum ferritin was 7ng/mL, and serum erythro­
poietin was 2 mIU/mL (range, 3–19 mIU/mL), confirming a
diagnosis of polycythemia vera (PV). While her symptoms
now feel mild and manageable, she is concerned about the
Prakash Singh Shekhawat
safety of future pregnancies, which she desires.
Introduction
Our patient has PV, and her experience is common to many
when diagnosed: long­standing but nonspecific symptoms
and a lag from the time of blood count abnormalities to con­
firmation of diagnosis. More pernicious is her experience in
pregnancy, fraught with complications that are too com­
mon in women with PV, including the overlooked real­time
diagnosis. Nevertheless, by standard vascular risk stratifi­
cation criteria, our patient is considered to have low­risk
disease, a designation of little comfort to many young peo­
ple with myeloproliferative neoplasms (MPN). How can we
best address her immediate concerns related to thriving as
a “thirty­something,” the intermediate goal of growing her
family, and, finally, the long­term and often dominant con­
cern related to progression of disease and early mortality?
Many are hopeful that interferons (IFNs), especially long­
acting formulations, might address such patient concerns.
MPN as inflammatory diseases
Classical MPN without the Philadelphia­negative chromo­
some are clonal diseases classified into 3 subtypes: PV,
IFN in PV/ET | 463
essen­tial thrombocythemia (ET), and pri­mary mye­lo­fi­bro­sis (PMF).
They are clin­i­cally char­ac­ter­ized by pro­lif­er­a­tive bone mar­row,
abnor­mal blood counts, spleno­meg­aly, and con­sti­tu­tional symp­
toms, along with a long-term higher risk of sec­ond­ary bone mar­
row fibro­ sis and acute leu­ ke­mic trans­for­
ma­ tion. In 2005 the 9
onco­genic driver muta­tion was iden­ti­fied and found to be pres­ent
in nearly all­patients with PV and about 50% to 60% of patients
with ET or PMF. Activating muta­tions in thrombopoietin recep­
tor (MPL) and calreticulin (CALR) account for the large major­ity of
remaining cases. These driver muta­tions func­tion­ally con­verge in a
com­mon path­way, resulting in aber­rant JAK2-STAT sig­nal­ing.
The genetic driv­ers of MPN con­verge, with grow­ing evi­dence
that proinflammatory pro­cesses are cen­tral to MPN patho­bi­­ol­ogy.
MPN are acquired dis­eases of stem cells, and the JAK2 path­way is
con­sid­ered an impor­tant inflam­ma­tory driver. Patients with MPN
have increased lev­els of inflam­ma­tory cyto­kines, dysregulation of
immune-related genes, and enhanced oxi­da­tive stress.1,2 As of the
sum­mer of 2021, JAK inhib­i­tors remain the only approved med­i­
ca­tions in the United States for PV and PMF, and cen­tral to their
activ­ity is the abil­ity to mod­u­late cyto­kine pro­files and improve
symp­ toms. Despite off-label use, IFNs predated JAK inhib­i­tors
for the treat­ment of MPN. IFNs have been con­sid­ered a stan­dard
treat­ment for decades, predating even the iden­ti­fi­ca­tion of the
driver muta­tions. But we are now left with a cen­tral par­a­dox of
treating a proinflammatory state with a proinflammatory cyto­kine.
Optimally, the expe­ri­ence of IFN use and mech­a­nis­tic pre­clin­i­cal
and clin­ic­ al data will lead us on a path to res­o­lu­tion of this par­a­
dox and, more impor­tantly, improved out­comes for our patients.
History of IFN use as it relates to the treat­ment
of PV and ET
Let us now con­sider treating our patient. Recombinant IFN-al­
pha has been a treat­ment option for patients with hema­to­logic
malig­nan­cies for decades and was first used to treat MPN over
30 years ago. In the pre–tyro­sine kinase era, IFN was ­able to
induce deep remis­sions in a small sub­set of patients with chronic
mye­loid leu­ke­mia, while con­ven­tional chemotherapies failed.3 In
Philadelphia chro­mo­some-neg­a­tive MPN, an early study reported
on 12 patients with improve­ment in thrombocytosis with 2 to 10
weeks of recom­ bi­
nant IFN.4,5 Despite other early tri­ als show­
ing prom­is­ing responses, the side effects with IFN forced unac­
cept­able dis­con­tin­u­a­tion rates, and con­se­quently, its use was
not widely adopted. MPN patients can expe­ri­ence debil­i­tat­ing
con­ sti­
tu­
tional symp­ toms, felt to be cyto­ kine driven, mak­ ing
expected but overlapping IFN-related side effects poten­ tially
even less well tol­er­ated in this pop­u­la­tion. The devel­op­ment of
pegylated IFNs (pegIFNs), a lon­ger-act­ing for­mu­la­tion that is bet­
ter tol­er­ated and requires less fre­quent dos­ing, renewed inter­
est in this treat­ment approach. More recent stud­ies, sum­ma­rized
here, have shown prom­is­ing responses in hema­to­logic abnor­mal­
i­ties, symp­toms, and anticlonal activ­ity with the use of pegIFN.
The exact action of IFN in MPN remains an area of intense
inves­ti­ga­tion. Supported mech­a­nisms include direct cyto­toxic
effects on the malig­nant clonal cells in addi­tion to enhanc­ing the
host innate and adap­tive immune response. IFN may increase
the expres­sion of neoantigens on malig­nant clonal cells, ren­der­
ing them more sus­cep­ti­ble to immunosurveillance,6 and it can
also induce emer­gence of the hema­to­poi­etic stem cell from qui­
es­cence to allow for apo­pto­sis (exten­sively reviewed in Kiladjian,
Mesa, and Hoffman et al).7 Ideally, cor­rel­a­tive sci­ence from pro­
464  |  Hematology 2021  |  ASH Education Program
spec­tive stud­ies of peg-IFN can help elu­ci­date the mech­a­nisms
most rel­e­vant for opti­mal response.
Goals of ther­apy and risk strat­i­fi­ca­tion
The goals of treating patients with MPN include ame­lio­ra­tion
of symp­toms, reduc­tion in throm­bo­sis risk, con­trol of abnor­mal
blood counts, reduc­tion in long-term con­se­quences, and, hope­
fully, delay of dis­ease pro­gres­sion. In addi­tion to an often dom­i­
nant con­cern for pro­gres­sion to advanced mye­lo­fi­bro­sis and/or
acute leu­ke­mia, addi­tional con­se­quences of MPN ther­a­pies can
accu­mu­late over time; thus, thought­ful treat­ment rec­om­men­da­
tions are cru­cial.
Therapeutic phle­bot­omy (TP) alone was found to be supe­rior
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/463/1852047/463traxler.pdf by guest on 13 December 2021
to alkylating agents in one of the early piv­otal stud­ies of inter­
ven­tions for PV, is the back­bone of treat­ment in newly diag­nosed
PV, and per­sists as an impor­tant Hip­po­cratic reminder that sim­
ple can be bet­ter.8 TP is par­tic­u­larly crit­i­cal in the acute set­ting
at diag­no­sis when signs or symp­toms of hyper­vis­cos­ity are pres­
ent. A more recent foun­da­tional study, CYTO-PV, con­firmed that
hemat­oc ­ rit con­trol below 45%, by means of TP or cytoreduc­
tive med­i­ca­tions, reduced the risk of vas­cu­lar events by approx­
i­ma­tely 4-fold.9 But TP only con­trols red blood cell count, not
plate­lets and white blood cells, so pro­lif­er­at­ive dis­ease is only
par­tially addressed with this sat­is­fy­ingly sim­ple and safe inter­
ven­tion. In addi­tion, iron defi­ciency—some­times pres­ent even
prior to phle­bot­omy ini­ti­at­ion in PV patients such as ours—is
induced or wors­ened by TP, which often drives plate­let counts
higher still and in aggre­gate may cause or exac­er­bate micro­vas­
cu­lar or neurocognitive symp­toms. PegIFN, by con­trast, induces
trilineage blood cell count low­er­ing, and when responses are
seen, it can offer the pos­si­bil­ity of maintaining blood count con­
trol while still correcting iron defi­ciency, thus interrupting the
vicious TP-iron defi­ciency-thrombocytosis cycle (Figure 1).
Comparative effi­cacy data to guide the selec­tion of ini­tial
cytoreductive ther­apy in PV and ET are lim­ited, and ther­apy is
influ­enced by con­ve­nience, cost, tol­er­a­bil­ity, and safety. PegIFN
is not cur­rently approved in the US for treat­ment of ET or PV, yet
con­sen­sus guide­lines rec­om­mend its con­sid­er­ation as a first-line
option for cytoreduction, par­tic­u­larly in youn­ger patients and
those who desire preg­nancy.10
Efficacy of pegIFN
Hematologic response
In tri­als using stan­dard for­mu­la­tions of recom­bi­nant IFN, hema­to­
logic responses were observed in about 80% of patients, who also
reported a sig­nif­i­cant reduc­tion in MPN-asso­ci­ated symp­toms,
includ­ing pruritus. More recent stud­ies with lon­ger-act­ing peg-
IFN have reported a sim­i­lar effect size. Two inde­pen­dent phase
2 stud­ies using peginterferon alfa-2a in PV and ET both dem­on­
strated effec­tive cytoreduction with less dis­con­tin­u­a­tion due to
tox­ic­ity. The French “PV-Nord” group (PVN1) conducted a study of
37 patients with newly diag­nosed PV and boasted count nor­mal­i­
za­tion in 95% of the cohort at 12 months.11 Testifying to the tol­er­a­
bil­ity of this for­mu­la­tion, the vast major­ity of patients remained on
the treat­ment for at least 1 year. Surprisingly, a great major­ity of
these responses were dura­ble at 6-year fol­low-up, even after stop­
ping or switching ther­a­pies.12 In a sep­a­rate study of patients with
both PV and ET, response rates were around 80% for both dis­
ease sub­types.13 Noncomparative stud­ies have defined the effec­
tive­ness of pegIFN in hydroxy­urea (HU) resis­tance/intol­er­ance.14
 Polycythemia vera
with hematocrit
above goal

Cytoreductive therapy

Therapeutic
phlebotomy
Progressive
microvascular
symptoms

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/463/1852047/463traxler.pdf by guest on 13 December 2021

Iron deficiency

Reactive
thrombocytosis

Figure 1. Vicious cycle: iron deficiency, polycythemia vera, and thrombocytosis.

While these have both gar­ nered sup­ port for their effi­
cacy in
nor­mal­iz­ing blood cell counts and reduc­ing phle­bot­omy needs,
ran­
dom­ ized tri­als com­par­
ing pegIFN to HU may also pro­ vide
impor­tant direct com­par­a­tive effi­cacy. MPN-RC-112 is a trial that
ran­dom­izes high-risk PV and ET patients to first-line treat­ment
with either pegIFN or HU. A final anal­y­sis of the trial has yet to
be published, but interim trial updates report an over­all response
rate of 78% with peginterferon alfa-2a, not sig­nif­i­cantly dif­fer­ent
from that seen with HU (Table 1). Symptom ben­e­fit appears to be
most pro­nounced for both agents in patients with a high base­line
symp­tom bur­den, though more patient-reported out­come data
are needed to make con­clu­sions about symp­tom ben­e­fit for both
HU and pegIFN. Comparative rates of throm­bo­ses will be a key
end point for deter­min­ing the best treat­ment across the age span.
Molecular effects: defin­ing dis­ease mod­i­fi­ca­tion
In addi­tion to impres­sive hema­to­logic responses, stud­ies using
pegIFN have reported patients achiev­ing a com­plete molec­u­
lar response. Defined as an unde­tect­able JAK2V617F allele fre­
quency, for the first time these data suggested that exog­e­nous
IFN could poten­tially be directly targeting the MPN clone.11 The
JAK2V617F allele bur­den was pro­gres­sively reduced in about 70%
of PV patients and 40% of ET patients. Moreover, the JAK2V617F
muta­tion became unde­tect­able in 24% of PV patients in the PVN1
study.15 Slightly lower rates of molec­u­lar remis­sion—15% in PV
and 6% in ET patients—were reported in the US study. This is not
sur­pris­ing given a more het­ero­ge­neous study pop­u­la­tion with a
lon­ger dis­ease dura­tion. In both dis­ease groups, the pres­ence
of addi­tional muta­tions such as TET2, ASXL1, IDH2, and TP53
was asso­ci­ated with poorer molec­u­lar responses.16 Along with
improve­ments in both symp­toms and cell counts, these molec­
u­lar responses raise the pos­si­bil­ity and hope that this treat­ment
could selec­tively and dura­bly tar­get the malig­nant clone and
even allow hema­tol­o­gists to con­tem­plate a cura­tive treat­ment.
This aspi­ra­tion is often cap­tured with the phrase “dis­ease mod­i­
fy­ing” when treating patients with PV and ET.

Table 1. Long-acting interferon trials

Study Phase Comparison Patients Follow-up dura­tion Hematologic CR Molecular response

PVN1 2 peginterferon alfa-2a 37 with PV 31.4 mo 94.6% 18.9% unde­tected
72.4% with >50% reduc­tion
MDACC 2 peginterferon alfa-2a 40 with PV 20 mo 70% (PV) 76% (ET) 14% and 6% unde­tected
39 with ET (PV vs ET)
54% and 38% with >50%
reduc­tion (PV vs ET)
PV-PROUD/ 3 Ropeginterferon vs HU 257 with PV 45.5 mo IFN vs HU 44% vs 51% at 12 mo (IFN
CONTINUATION-PV 71% vs 51% at 36 months vs HU)
66% vs 27% at 36 mo (IFN
vs HU)
Prakash Singh Shekhawat
IFN in PV/ET  |  465
Will I be on this for­ever? Treatment dura­tion with pegIFN
Whether dura­ble dis­ease mod­i­fi­ca­tion or cure is achiev­able with
pegIFN has not yet been answered. How these gen­er­ally encour­
ag­ing results trans­late into clin­i­cal prac­tice can be var­i­able, and
the opti­mal treat­ment dura­tion with pegIFN is unknown. If a com­
plete molec­u­lar remis­sion is achieved, is a main­te­nance phase of
treat­ment required? Tolerability and safety are crit­i­cal to these
deci­sions, par­tic­u­larly in a pop­u­la­tion with often near-nor­mal life
expec­tancy. Longer-term fol­low-up stud­ies suggested an over­all
dis­con­tin­u­a­tion rate of 22% for treat­ment-related tox­ic­ity. These
rates decreased over time, though treat­ment-lim­it­ing grade
3 or 4 toxicities were still observed after 60 months on ther­apy.17
If pro­vider and patient elect to pur­sue a treat­ment-free remis­sion
period in patients achiev­ing a com­plete molec­u­lar response,
what are the param­et­ ers for resum­ing treat­ment? Should blood
count mon­i­tor­ing only drive retreatment deci­sions, or is mon­i­
tor­ing for ear­lier mea­sur­able resid­ual dis­ease by JAK2V617F poly­
mer­ase chain reac­tion a bet­ter stan­dard? If the lat­ter, what is the
thresh­old of detect­abil­ity driv­ing deci­sions? At what fre­quency
should patients be mon­i­tored? These and many more related
ques­tions remain unanswered.
Vascular risk
Many of the major life-lim­it­ing con­se­quences of MPN stem from
vas­cu­lar com­pli­ca­tions, with rates of throm­bo­sis reported at
5.5 events per 100 patients per year.18 The published lon­gi­tu­
di­nal data sug­gest that IFN use may mit­i­gate vas­cu­lar risk, but
com­par­a­tive data are lacking to robustly answer these impor­
tant ques­tions. Encouragingly, the PVN1 trial did not report any
throm­bo­sis events in 6 years of fol­low-up, while the phase 2 US
trial reported a throm­bo­sis inci­dence of 1.22 per 100 patients per
year.17 Additional long-term fol­low-up may help define any true
vas­cu­lar ben­efi­ t, but these data sug­gest that pegIFN is asso­ci­
ated with an over­all low throm­bo­sis risk.
Patients with splanch­nic vein throm­bo­sis (SVT) rep­re­sent a
par­tic­u­larly high-risk group. MPN are a com­mon predisposing
risk fac­tor for SVT, involv­ing hepatic, splenic, por­tal, or mes­en­
teric veins, and it has been rec­og­nized that patients with SVT are
at increased risk of devel­op­ing recur­rent throm­bo­ses, bleed­ing
com­pli­ca­tions, and arte­rial throm­bo­ses. Furthermore, HU fails
to pre­vent recur­rent SVT.19 The MPD-RC 111 study set to address
whether IFN could poten­tially fill this clin­i­cal need in a pro­spec­
tive trial eval­u­at­ing peg-IFN in 20 patients with a his­tory of SVT.14
Hematologic response rates were 70% at 12 months, on par
with pre­vi­ous stud­ies of all­-com­ers, and SVT did not recur in any
patients dur­ing 2.2 years of fol­low-up.
Ropeginterferon alfa-2b
Efforts to fur­ther improve the activ­ity and tol­er­ab­ il­ity of pegIFN
treat­ment resulted in the devel­op­ment of ropeginterferon alfa-2b,
a monopegylated IFN dosed every 14 days. The Euro­pean Med­
icines Agency approved ropeginterferon alfa-2b in 2019 for PV
with­out symp­tom­atic spleno­meg­aly; a deci­sion from the US Food
and Drug Association was expected in March 2021 but has yet to
be decided at the time of this writ­ing. If approved, ropeginter­
feron alfa-2b would be the first IFN spe­cif­i­cally approved for MPN
in the US. Euro­pean Medicines Agency approval was based on a
reg­is­tra­tion study known as PROUD-PV and its exten­sion study,
CONTINUATION-PV. These stud­ies eval­u­ated ropeginterferon alfa-
2b com­pared with HU using a noninferiority design. PROUD-PV
466  |  Hematology 2021  |  ASH Education Program
enrolled 257 PV patients with early-stage dis­ease, includ­ing
patients newly requir­ing cytoreductive ther­apy or those who had
been treated with HU for fewer than 3 years.20 Randomization was
strat­i­fied by age, his­tory of throm­bo­em­bo­lism, and prior HU use.
The pri­mary out­come was com­posed of hema­to­logic response
and nor­mal­i­za­tion of spleen size at 12 months, achieved in 21%
of patients in the ropeginterferon alfa-2b group and 28% in the
HU group. Longer treat­ ment and fol­low-up showed that 53%
of patients achieved a com­ plete hema­ to­log­
i­
cal response with
improved dis­ease bur­den in the ropeginterferon alfa-2b group
vs 38% of patients receiv­ing HU (P = .044 at 36 months). Throm­
boembolic events in both arms were sim­i­larly rare. Reductions in
JAK2V617F allele fre­quency appeared to improve over time, pre­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/463/1852047/463traxler.pdf by guest on 13 December 2021
dom­in ­ antly in the ropeginterferon alfa-2b group, supporting the
dis­ease-mod­i­fy­ing claim that appears to be unique to pegIFN as
a class.
Dosing: pegIFN and ropeginterferon
High doses of IFN are poorly tol­er­ated. This was established
in the nonpegylated for­mu­la­tion era and in dose-find­ing stud­
ies of pegIFN using up to 360 µg weekly.13 But deter­mi­na­tion of
the opti­mal starting and main­te­nance dose for both pegIFN and
ropeginterferon is a work in prog­ress. Conventional phar­ma­co­
dy­namic stud­ies are chal­leng­ing when the tar­get(s) and mech­
a­nism of action of these agents remain some­what obscure. In
addi­tion, the rel­a­tively long time to response, with slow hema­
to­logic responses and sub­se­quent deeper molec­u­lar responses,
begs a phil­os­oph­i­cal dilemma between “hit it hard and hit it
early” vs “slow and steady wins the race” since a lin­ear and pre­
dict­able dose-response and dose-tox­ic­ity rela­tion­ship does not
appear to exist, and maintaining patients on a tol­er­a­ble dose
for a long period of time may be of par­tic­ul­ar impor­tance. Our
approach is to start at a low dose, 45 µg weekly, of pegIFN alpha-
2a (Pegasys, the only cur­rently avail­­able US for­mu­la­tion) for a
min­i­mum of 4 weeks. Close mon­i­tor­ing of mood, liver, and thy­
roid func­tion dic­tate fur­ther dose mod­i­fi­ca­tions or inter­rup­tions.
Some patients stay at 45 µg weekly, and oth­ers increase to 90 µg
weekly; doses as high as 135 or 180 µg weekly are less com­mon
but can also be tol­er­ated by some and may be help­ful in the
absence of a hema­to­logic response at lower doses.
Potential toxicities and mon­i­tor­ing for IFN ther­a­pies
Although lon­ger-act­ing IFNs have improved tol­er­a­bil­ity com­
pared to stan­dard for­mu­la­tions, patients still com­monly expe­ri­
ence adverse effects. Fatigue and flu-like symp­toms are prev­a­lent.
These tend to respond to acet­amin­op ­ hen or non­ste­roi­dal anti-
inflam­ma­tory med­i­ca­tions and can abate with time. Efforts to
min­ im
­ ize intolerabilities include low-dose run-in peri­ ods with
dose-esca­la­tion algo­rithms. Toxicity mon­i­tor­ing dur­ing treat­ment
includes blood counts as well as period thy­roid and liver func­tion
test­ing and depres­sion screen­ing. Dose adjust­ments for cytope­
nias, liver dys­func­tion, and mood or neurocognitive effects are
crit­i­cal to safety and opti­mal response.
Are there dif­fer­en­tial responses to IFN depending
on muta­tions?
As men­ tioned, a post hoc anal­ y­
sis of an early pegIFN study
astutely observed that fail­ing to achieve a com­plete molec­u­lar
remis­sion was asso­ci­ated with molec­u­larly more com­plex dis­ease
beyond the driver muta­tion. The pres­ence of muta­tions in TET2,
ASXL1, EZH2, DNMT3A, and IDH1/2 decreased the like­li­hood of
achiev­ing a molec­u­lar remis­sion.16 Much more recently, detailed
ana­ly­ses of serial sam­ples found that leu­ke­mic pro­gen­i­tor cells
with homo­zy­gous JAK2V617F muta­tions were more sen­si­tive to
IFN com­pared to cells with a sin­gle (het­ero­zy­gous) JAK2V617F
muta­tion and CALR-mutated cells. In addi­tion, for the het­ero­zy­
gous JAK2V617F cells, a higher dose of IFN appeared to be impor­
tant for effec­tive clear­ance of the clone.21 Thus, response to IFN
appears to be influ­enced by driver muta­tion type and zygos­ity
and by addi­tional dis­ease-asso­ci­ated muta­tions.
Pregnancy
What will we tell our patient about her desire for addi­tional
preg­nan­cies? While data on the man­age­ment of PV in the set­
ting of preg­nancy are lim­ited, the avail­­able stud­ies indi­cate that
early fetal loss is com­mon, as is intra­uter­ine growth restric­tion.
Preterm deliv­ery is com­mon, and there is an esti­mated fetal
sur­vival of 50%.22 In gen­eral, the con­tin­u­a­tion of aspi­rin is rec­
ommended for vir­tu­ally all­preg­nant women, and IFNs are con­
sid­ered safe and should gen­er­ally be con­tin­ued in patients for
whom cytoreductive treat­ ment was indi­ cated prepregnancy,
either for vas­cu­lar risk reduc­tion or symp­tom­atic con­trol.
CLINICAL CASE (Con­t in­u ed)
Now 2 years old, our patient’s daugh­ter is thriv­ing and meet­
ing all­her devel­op­men­tal mile­stones. Our patient started TP,
con­tin­ues on aspi­rin, transitioned to a non-estro­gen-containing
reli­able form of con­tra­cep­tion, and ini­ti­ated pegIFN alpha-2a,
ini­tially at a low dose (45 µg weekly) and after 6 weeks, with­out
dis­cern­able side effects, increased to 90 µg weekly. Her blood
counts have improved, and her phle­bot­omy require­ments have
decreased. She is fol­low­ing with her obstet­rics group and with
mater­nal fetal med­i­cine, plan­ning for a sec­ond preg­nancy. She
will con­tinue with both aspi­rin and pegIFN through­out preg­
nancy if she con­ tin­ues to tol­
er­ate it as well as pro­
phy­lac­tic
doses of low-molec­u­lar-weight hep­a­rin for 6 weeks post­par­tum.
Conclusions
Patients with PV and ET have a gen­er­ally favor­able prog­no­
sis, usu­ally sur­viv­ing decades and liv­ing full lives. But they live
with an incur­able chronic ill­ness that for some has debil­i­tat­
ing symp­toms, and harder to cap­ture is the exis­ten­tial uncer­
tainty patients feel about the pos­si­bil­ity of dis­ease pro­gres­sion.
There is excite­ ment around pegIFNs, nontargeted ther­ a­ pies
that selec­tively tar­get the malig­nant MPN clone. Many foun­
da­tional ques­tions—who, when, for how long—require more
pre­cise answers, but the prog­ress toward bet­ter out­comes is
unde­ni­able.
Conflict-of-inter­est dis­clo­sure
Elizabeth A. Traxler: no com­pet­ing finan­cial inter­ests to declare.
Elizabeth O. Hexner: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Elizabeth A. Traxler: off-label drug use includes hydroxyurea, pegin­
terferon alfa-2a, and Ropeginterferon alfa-2b.
Prakash Singh Shekhawat
Elizabeth O. Hexner: off-label drug use includes hydroxyurea, pegin­
terferon alfa-2a, and Ropeginterferon alfa-2b.
Correspondence
Elizabeth O. Hexner, Division of Hematology Oncology, Perelman
Center for Advanced Medicine, 3400 Civic Center Blvd, Philadel­
phia, PA 19104; e-mail: hexnere@pennmedicine​­.upenn​­.edu.
References
1. Bjørn ME, Hasselbalch HC. The role of reac­tive oxy­gen spe­cies in mye­lo­
fi­bro­sis and related neo­plasms. Mediators Inflamm. 2015;2015(Octo­ber):
648090.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/463/1852047/463traxler.pdf by guest on 13 December 2021
2. Hasselbalch HC, Thomassen M, Riley CH, et al. Whole blood tran­scrip­tional
pro­fil­ing reveals dereg­u­la­tion of oxi­da­tive and antioxidative defence genes
in mye­lo­fi­bro­sis and related neo­plasms. Potential impli­ca­tions of downreg­
ulation of Nrf2 for geno­mic insta­bil­ity and dis­ease pro­gres­sion. PLoS ONE.
2014;9(11):e112786.
3. Talpaz M, Kantarjian HM, McCredie K, Trujillo JM, Keating MJ, Gutterman JU.
Hematologic remis­sion and cyto­ge­netic improve­ment induced by recom­
bi­nant human inter­feron alpha A in chronic mye­log­e­nous leu­ke­mia. N Engl
J Med. 1986;314(17):1065-1069.
4. Silver RT. Recombinant inter­feron-alpha for treat­ment of polycythaemia
vera. Lancet. 1988;2(8607):403.
5. Tichelli A, Gratwohl A, Berger C, et al. Treatment of thrombocytosis in mye­
lo­pro­lif­er­a­tive dis­or­ders with inter­feron alpha-2a. Blut. 1989;58(1):15-19.
6. Xiong Z, Yan Y, Liu E, et al. Novel tumor anti­gens elicit anti-tumor humoral
immune reac­tions in a sub­set of patients with poly­cy­the­mia vera. Clin
Immunol. 2007;122(3):279-287.
7. Kiladjian JJ, Mesa RA, Hoffman R. The renais­sance of inter­feron ther­apy for
the treat­ment of mye­loid malig­nan­cies. Blood. 2011;117(18):4706-4715.
8. Berk PD, Goldberg JD, Silverstein MN, et al. Increased inci­dence of acute
leu­ke­mia in poly­cy­the­mia vera asso­ci­ated with chlorambucil ther­apy. N
Engl J Med. 1981;304(8):441-447.
9. Marchioli R, Finazzi G, Specchia G, et al; CYTO-PV Collaborative Group.
Cardiovascular events and inten­sity of treat­ment in poly­cy­the­mia vera. N
Engl J Med. 2013;368(1):22-33.
10. Mesa RA, Jamieson C, Bhatia R, et al. NCCN guide­lines insights: mye­lo­pro­lif­
er­a­tive neo­plasms, Version 2.2018. J Natl Compr Canc Netw. 2017;15(10):1193-
1207.
11. Kiladjian JJ, Cassinat B, Turlure P, et  al. High molec­u­lar response rate of
poly­cy­the­mia vera patients treated with pegylated inter­feron alpha-2a.
Blood. 2006;108(6):2037-2040.
12. Kiladjian JJ. Long-term treat­ment with inter­feron alfa for mye­lo­pro­lif­er­a­tive
neo­plasms. Lancet Haematol. 2017;4(4):e150-e151.
13. Quintás-Cardama A, Kantarjian H, Manshouri T, et al. Pegylated inter­feron
alfa-2a yields high rates of hema­to­logic and molec­u­lar response in patients
with advanced essen­tial thrombocythemia and poly­cy­the­mia vera. J Clin
Oncol. 2009;27(32):5418-5424.
14. Mascarenhas J, Kosiorek H, Prchal J, et al. A pro­spec­tive eval­u­a­tion of pegy-
lated inter­feron alfa-2a ther­apy in patients with poly­cy­the­mia vera and
essen­tial thrombocythemia with a prior splanch­nic vein throm­bo­sis. Leu-
kemia. 2019;33(12):2974-2978.
15. Kiladjian JJ, Cassinat B, Chevret S, et al. Pegylated inter­feron-alfa-2a induces
com­plete hema­to­logic and molec­u­lar responses with low tox­ic­ity in poly­cy­
the­mia vera. Blood. 2008;112(8):3065-3072.
16. Quintás-Cardama A, Abdel-Wahab O, Manshouri T, et al. Molecular anal­
y­sis of patients with poly­cy­the­mia vera or essen­tial thrombocythemia
receiv­ing pegylated inter­feron α-2a. Blood. 2013;122(6):893-901.
17. Masarova L, Patel KP, Newberry KJ, et  al. Pegylated inter­feron alfa-2a in
patients with essen­tial thrombocythaemia or polycythaemia vera: a post-
hoc, median 83 month fol­low-up of an open-label, phase 2 trial. Lancet
Haematol. 2017;4(4):e165-e175.
18. Marchioli R, Finazzi G, Landolfi R, et al. Vascular and neo­plas­tic risk in a large
cohort of patients with poly­cy­the­mia vera. J Clin Oncol. 2005;23(10):2224-
2232.
19. De Stefano V, Rossi E, Carobbio A, et al. Hydroxyurea pre­vents arte­rial and
late venous throm­botic recur­rences in patients with mye­lo­pro­lif­er­a­tive
neo­plasms but fails in the splanch­nic venous dis­trict: pooled anal­y­sis of
1500 cases. Blood Cancer J. 2018;8(11):112.
IFN in PV/ET  |  467
20. Gisslinger H, Klade C, Georgiev P, et al; PROUD-PV Study Group. Ropegin­
terferon alfa-2b ver­sus stan­dard ther­apy for polycythaemia vera (PROUD-PV
and CONTINUATION-PV): a randomised, non-infe­ri­or­ity, phase 3 trial and its
exten­sion study. Lancet Haematol. 2020;7(3):e196-e208.
21. Mosca M, Hermange G, Tisserand A, et al. Inferring the dynamic of mutated
hema­to­poi­etic stem and pro­gen­i­tor cells induced by IFNα in mye­lo­pro­lif­
er­a­tive neo­plasms. Published online 18 August 2021. Blood.
468  |  Hematology 2021  |  ASH Education Program
22. Gangat N, Joshi M, Shah S, et al. Pregnancy out­comes in mye­lo­pro­lif­er­a­
tive neo­plasms: a Mayo Clinic report on 102 preg­nan­cies. Am J Hematol.
2020;95(5):E114-E117.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000280
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/463/1852047/463traxler.pdf by guest on 13 December 2021
 MULTIDISCIPLINARY HEMATOLOGIC DISORDERS: WHAT IS THE HEMATOLOGIST ’ S ROLE ?

Hereditary hemorrhagic telangiectasia (HHT):

a practical guide to management

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/469/1852038/469hammill.pdf by guest on 13 December 2021

Adrienne M. Hammill,1 Katie Wusik,2 and Raj S. Kasthuri3
Cancer and Blood Diseases Institute, Division of Hematology, Cincinnati Children’s Hospital Medical Center, and Department of Pediatrics,
1

University of Cincinnati College of Medicine, Cincinnati, OH; 2Division of Human Genetics, Cincinnati Children’s Hospital Medical Center, Cincinnati, OH;
and 3Division of Hematology, Department of Medicine, University of North Carolina School of Medicine, Chapel Hill, NC

Hereditary hemorrhagic telangiectasia (HHT), the second most common inherited bleeding disorder, is associated with
the development of malformed blood vessels. Abnormal blood vessels may be small and cutaneous or mucosal (telangi-
ectasia), with frequent complications of bleeding, or large and visceral (arteriovenous malformations [AVMs]), with addi-
tional risks that can lead to significant morbidity and even mortality. HHT can present in many different ways and can be
difficult to recognize, particularly in younger patients in the absence of a known family history of disease or epistaxis, its
most common manifestation. HHT is commonly diagnosed using the established Curaçao clinical criteria, which include
(1) family history, (2) recurrent epistaxis, (3) telangiectasia, and (4) visceral AVMs. Fulfillment of 3 or more criteria provides
a definite diagnosis of HHT, whereas 2 criteria constitute a possible diagnosis of HHT. However, these criteria are insuf-
ficient in children to rule out disease due to the age-dependent development of some of these criteria. Genetic testing,
when positive, can provide definitive diagnosis of HHT in all age groups. Clinical course is often complicated by signif-
icant epistaxis and/or gastrointestinal bleeding, leading to anemia in half of adult patients with HHT. The management
paradigm has recently shifted from surgical approaches to medical treatments aimed at control of chronic bleeding, such
as antifibrinolytic and antiangiogenic agents, combined with aggressive iron replacement with intravenous iron. Guide-
lines for management of HHT, including screening and treatment, were determined by expert consensus and originally
published in 2009 with updates and new guidelines in 2020.

LEARNING OBJECTIVES
• Recognize signs and symptoms of HHT
• Know when to test/screen patients for additional complications of the disease
• Understand the approach to treatment of various manifestations of HHT

ther questioning revealed that her mother and brother also

CLINICAL CASE
Presentation. A 17-year-old previously healthy girl presented
to neurology with a complaint of intractable headaches and
a history of possible prior concussions during sports. Mag-
netic resonance imaging (MRI) without contrast revealed
some asymmetry of the cerebral vessels, and subsequent
MRI with contrast and magnetic resonance arteriography
revealed an arteriovenous malformation (AVM) involving the
left perisylvian fissure and insular areas. This was not felt to
be the cause of her headaches, and an electroencephalo-
gram ruled out seizure activity. She was referred to neuro-
surgery for further management. The neurosurgeon noted
telangiectasia on her fingertips and elicited a history of
heavy recurrent nosebleeds since the age of 9 years. Fur-
had frequent epistaxis and migraine headaches and that her
mother had undergone a lung lobectomy for AVMs. At that
time, the neurosurgeon informed the family that the patient
likely had hereditary hemorrhagic telangiectasia (HHT).
Diagnosis. The patient was referred to our HHT center
20 months after seeing neurosurgery at the age of 18 years.
She continued to have significant migraine headaches
despite medication. She had nosebleeds approximately 1/wk,
often gushing, lasting up to 15 minutes (Epistaxis Sever-
ity Score [ESS] 3.7). Physical examination confirmed tel-
angiectasia on her fingers, lower lip, and tongue. Family
history was confirmed, including distant cousins with brain
and lung AVMs. The patient met Curaçao clinical criteria for
definite HHT (+telangiectasia, +epistaxis, +AVM) at this visit

Prakash Singh Shekhawat

The ABCs of HHT | 469
even in the absence of a diag­no­sis in a first-degree rel­a­tive. She
under­went genetic test­ing to con­firm the diag­no­sis, and a path­o­
genic var­ia
­ nt in the endoglin gene ENG was iden­ti­fied.
Screening. Additional screen­ing was performed to eval­ua ­ te
for lung AVMs. A delayed-bub­ble echo was pos­i­tive, and sub­se­
quent com­puted tomog­ra­phy angi­og­ra­phy of the chest revealed
a left lower lobe AVM with a nidus mea­sur­ing nearly 2 × 1 × 1 cm,
with a feed­ing artery mea­sur­ing 4 mm.
When ini­ tial lung AVM screen­ing is neg­a­
tive, it should be
repeated every 5 years in chil­dren1,2 and every 5 to 10 years
in adults.
Treatment. She under­went embo­li­za­tion of her pul­mo­nary
AVM with dra­matic improve­ment in her migraine head­aches. At
the age of 22 years, she under­went pre­ci­sion radio­ther­apy to mul­
ti­ple brain AVMs due to con­cern for pro­gres­sive venous ectasia.
Follow-up. She con­tin­ues to be followed in our HHT Center of
Excellence at reg­u­lar inter­vals to mon­i­tor for new or recanalized
pul­mo­nary AVMs (bub­ble echo every 3-5 years) and changes in
epi­staxis (ESS at every visit). She gets annual lab­or­a­tory tests to
screen for com­pli­ca­tions of liver AVMs (hepatic panel includ­ing
γ-glutamyl trans­fer­ase [GGT] to assess liver func­tion and brain
natri­uretic pep­tide [BNP] for heart func­tion) and ane­mia (com­
plete blood count [CBC], fer­ ri­tin, total iron bind­
ing capac­ ity
[TIBC], and retic­u­lo­cyte count) related to nose­bleeds or poten­
tial gas­tro­in­tes­ti­nal (GI) bleed­ing. She takes anti­bi­ot­ics prior
to den­tal pro­ce­dures, includ­ing reg­u­lar cleaning, for infec­tion
pro­phy­laxis, due to her known lung AVMs as per HHT treat­ment
guide­lines. She is also followed in the cere­bro­vas­cu­lar clinic for
her cere­bral AVMs.
Annual lab­o­
ra­
tory tests for adults followed in our cen­ ter
include CBC, fer­ri­tin, retic­u­lo­cyte count, TIBC, hepatic pro­file,
GGT, BNP,2 and vita­min D.3,4
Her fam­ily mem­bers also under­went screen­ing for vis­ceral vas­
cu­lar lesions given their clin­i­cal diag­noses of HHT once our
patient was diag­nosed. Her brother was found to have both
mul­ti­ple brain AVMs, which were treated with radio­ther­apy,
and mul­ti­ple lung AVMs, which were treated with coil embo­li­
za­tion. Her mother was found to have addi­tional lung AVMs that
required embo­ li­
za­
tion. Furthermore, her mother proceeded
to develop severe iron defi­ciency ane­mia sec­ond­ary to both
severe epi­staxis and GI bleed­ing and is tak­ing main­te­nance
ther­apy with sys­temic bevacizumab in addi­tion to inter­mit­tent
iron infu­sions to main­tain fer­ri­tin more than 50 ng/mL.
Introduction
Hereditary hem­or­rhagic tel­an­gi­ec­ta­sia is the sec­ond most com­
mon inherited bleed­ing dis­or­der, occur­ring in 1/5000 to 1/10,000
peo­ple.6 It is gen­er­ally inherited in an auto­so­mal dom­i­nant fash­
ion but can occur de novo, includ­ing in mosaic form in some
pro­bands.7-11 Bleeding is the most com­mon symp­tom, occur­ring
from nasal, cuta­ne­ous, or GI tel­an­gi­ec­ta­ses. Unlike bleed­ing dis­
or­ders such as hemo­philia or von Willebrand dis­ease, bleed­ing in
HHT is sec­ond­ary to devel­op­ment of malformed blood ves­sels,
which result in ectasia and increased fra­gil­ity. This leads to local­
470  |  Hematology 2021  |  ASH Education Program
ized bleed­ing from the abnor­mal ves­sels, rather than the gen­eral
risk of bleed­ing and post­sur­gi­cal bleed­ing asso­ci­ated with defi-
ciencies of plasma coag­u­la­tion fac­tors or plate­let dys­func­tion.
Diagnosis
The Curaçao clin­i­cal diag­nos­tic cri­te­ria, first for­mal­ized in 2000,
are shown in Table 1.12 The first is epi­staxis; to meet this cri­te­rion,
nose­bleeds should be recur­rent and spon­ta­ne­ous. These may be
gushing or drip­ping, as there is no require­ment for sever­ity.
While there is no widely accepted nose­bleed fre­quency to
meet this cri­te­rion, we gen­er­ally use more than 4 per year with
no other cause.5 A his­tory of night­time nose­bleeds is par­tic­u­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/469/1852038/469hammill.pdf by guest on 13 December 2021
larly sug­ges­tive of HHT.
Telangiectases, or small bright red spots, typ­i­cally occur at char­ac­
ter­is­tic sites in HHT, includ­ing lips, oral cav­ity (par­tic­u­larly tongue),
fin­gers, and nose (Figure 1). These tel­an­gi­ec­ta­sia can be tiny and
appear like red freck­les; they gen­er­ally blanch with pres­sure, with
rapid reperfusion upon release. Visceral lesions can also be pres­
ent, includ­ing GI tel­an­gi­ec­ta­ses and larger high-flow lesions, includ­
ing (in order of fre­quency) liver, lung, brain, and spi­nal AVMs. These
can be asymp­tom­atic and may only be detected with screen­ing.
A fam­ily his­tory of sim­il­ar symp­toms in a first-degree rel­a­tive (par­
ent, sib­ling, or child) is usu­ally pres­ent even if they do not always
carry a diag­no­sis of HHT. Underrecognition and delayed diag­no­sis
is a con­sid­er­able prob­lem in HHT,13 and there is a need to increase
aware­ness among health care pro­vid­ers.
There is sim­i­larly no stan­dard required tel­an­gi­ec­ta­sia count; in
our cen­ters, we use more than 4.5
Clinical cri­te­ria, val­i­dated in adults, may not be suf­fic­ ient to rule out
dis­ease in chil­dren youn­ger than 16 years.14 In 1 study by Pahl et al,14
the pos­i­tive pre­dic­tive value of the clin­i­cal cri­te­ria was extremely
high in chil­dren (99% over­all), but the neg­a­tive pre­dic­tive value
was only 54% over­all and even lower in chil­dren youn­ger than
5 years. This is due to the age-depen­dent devel­op­ment of 2 of the
Table 1. Curaçao cri­te­ria for the clin­i­cal diag­no­sis of HHT12
Telangiectases Multiple, at char­ac­ter­is­tic sites
  Lips, oral cav­ity, fin­gers, nose
Epistaxis Recurrent spon­ta­ne­ous nose­bleeds
Visceral involve­ment GI tel­an­gi­ec­ta­sia
Pulmonary AVM
Hepatic AVM
Cerebral VM
Spinal AVM
Family his­tory First-degree rel­a­tive with known HHT
  Parent, sib­ling, or child
Definite if ≥3 cri­te­ria are pres­ent OR if path­o­genic var­i­ant iden­ti­fied in
known HHT gene
Suspected if 2 cri­te­ria are pres­ent
Unlikely if <2 cri­te­ria are pres­ent

Figure 1. Telangiectases in HHT.
cri­te­ria. Nosebleeds typ­i­cally develop in the sec­ond decade of
life and do not com­monly require med­i­cal atten­tion in child­hood,
although many patients do develop recur­rent and/or severe epi­
staxis requir­ing med­i­cal and/or sur­gi­cal inter­ven­tions by mid­dle
age. Development of tel­an­gi­ec­ta­sia is also age depen­dent, typ­i­
cally noted in the sec­ond and third decades. Thus, even when a
child has a par­ent with HHT (1 clin­i­cal cri­te­rion), in the absence of
imag­ing performed to screen for vis­ceral AVMs, affected chil­dren
may not meet the diag­nos­tic thresh­old for dis­ease.
Genetic test­ing has become more avail­­able and afford­able
and can be used to make the HHT diag­no­sis.15 HHT-caus­ing muta­
tions have been iden­ti­fied in sev­eral genes in the transforming
growth fac­tor β path­way, with the major­ity of path­o­genic var­i­
ants found in ENG and ACVRL1. In 1 recent study, ENG and ACVRL1
muta­tions were found to com­prise up to 96% of cases of “clas­
sic HHT” meet­ing strictly applied Curaçao cri­te­ria.16 SMAD4 gene
muta­tions account for approx­i­ma­tely 10% of ENG- and ACVRL1-
neg­a­tive cases of HHT, account­ing for 1% to 2% of cases over­
all. SMAD4 also causes juve­nile polyposis, resulting in GI pol­yps
16,17
and a can­cer pre­dis­po­si­tion syn­drome, 18
and has Singh
Prakash been impli­
cated
Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/469/1852038/469hammill.pdf by guest on 13 December 2021
in aortopathy as well.19 Rarely, HHT cases meet­ing clin­i­cal cri­te­ria
may be caused by GDF2 (BMP9) or RASA1.20 Overlap has been
observed between HHT and cap­il­lary‑ malformation–AVM syn­
drome caused by EPHB4.21 In prac­tice, most test­ing is now per-
formed as part of multigene pan­els consisting of 5 to 6 genes
(ENG, ACVRL1, SMAD4, RASA1, GDF2, and EPHB4).
Genetic test­ing is recommended to assist in establishing the
diag­no­sis in indi­vid­u­als who do not meet Curaçao cri­te­ria or in
those who are asymp­tom­atic or min­i­mally symp­tom­atic, includ­
ing young chil­dren. Genetic test­ing can also be used to iden­tify
the caus­a­tive muta­tion in a fam­ily with clin­i­cally con­firmed HHT
or to estab­lish a diag­no­sis in rel­a­tives of a per­son with a known
caus­a­tive muta­tion.15 However, cur­rent test­ing remains unable
to iden­tify a caus­a­tive genetic change in up to 10% to 15% of
patients with a clin­i­cal diag­no­sis.
Several lab­ o­
ra­
to­ries now offer free fam­ ily test­
ing within a
spec­i­fied time period when a path­o­genic var­i­ant is iden­ti­fied
in 1 patient.
The ABCs of HHT  |  471

Figure 2. Screening recommendations for HHT, assembled from multiple guidelines.1,15
Guidelines have been established regard­ ing the diag­ no­sis,
screen­ing, and treat­ment of peo­ple liv­ing with HHT. The orig­i­nal
guide­lines were published in 2011 and cov­ered diag­no­sis of HHT,
epi­staxis, cere­bral vas­cu­lar malformations, pul­mo­nary AVMs, GI
bleed­ing, and liver vas­cu­lar malformations.15 The Second Inter-
national guide­lines were published in 2020, with updated guide­
lines for epi­staxis, GI bleed­ing, and liver vas­cu­lar malformations,
as well as new guide­lines on ane­mia and anticoagulation, pedi­
at­rics, and preg­nancy and deliv­ery.1
Clinical man­i­fes­ta­tions and man­age­ment
Initial screen­ing
Once the diag­no­sis of HHT is made or suspected, all­patients
should undergo ini­tial screen­ing for poten­tial com­pli­ca­tions.
Although a long-time area of debate in the care of patients
with HHT, recent pedi­at­ric care guide­lines clar­ify that chil­dren
require workup and screen­ing as well as adults, because even
though the out­ward clin­i­cal signs (tel­an­gi­ec­ta­sia) and symp­
toms (epi­staxis) may so far be absent, pedi­at­ric patients can still
have vis­ceral AVMs that can cause mor­bid­ity and even mor­tal­
ity. Screening pro­ce­dures are sum­ma­rized in Figure 2 from the
guide­lines for pul­mo­nary AVMs, liver vas­cu­lar malformations,
brain vas­cu­lar malformations, and pedi­at­ric care.
Epistaxis
Epistaxis, the most prominent symp­tom in most peo­ple with HHT,
can sig­nif­i­cantly inter­fere with daily activ­i­ties, includ­ing work,
school, and qual­ity of life (QoL). The updated guide­lines out­line
a num­ ber of rec­ om­men­ da­
tions for con­
trol of nose­ bleeds in a
step­wise fash­ion.1 The ESS was pre­vi­ously established for use in
HHT to facil­i­tate mea­sure­ment of nose­bleeds based on a num­
472  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/469/1852038/469hammill.pdf by guest on 13 December 2021
ber of fac­tors, includ­ing fre­quency, dura­tion, amount of bleed­
ing and pres­ence of ane­mia, and inter­ven­tions needed to con­trol
the bleed­ing.22 This score should be cal­cu­lated at every visit and
before and after any inter­ven­tions to mea­sure their impact (online
cal­cu­la­tor read­ily avail­­able at https:​­/​­/curehht​­.org​­/resource​­/epi-
staxis​­-severity​­-score​­/). ESS can be used to guide treat­ment of epi­
staxis, as shown in Figure 3. Although his­tor­ic ­ ally approaches were
pri­mar­ily sur­gi­cal, med­i­cal ther­a­pies includ­ing antifibrinolytics
(tranexamic acid, epsi­lon aminocaproic acid)23,24 and sys­temic anti-
angiogenics have become the main­stay of treat­ment strat­e­gies in
recent years, as discussed in more detail below under “Anemia.”
GI bleed­ing
GI bleed­ing is also a com­mon symp­tom in adults with HHT older
than 50 years, reported in 13% to 30% of patients. Most GI tel­
an­gi­ec­ta­sias occur in the stom­ach (46%-75%) and small bowel
(56%-91%).25 GI bleed­ing can be highly mor­bid in adults, with
chronic GI bleed­ing resulting in severe iron defi­ciency ane­mia
requir­ing fre­ quent intra­ ve­nous iron replace­ ment and/or red
blood cell trans­ fu­
sions, with decreased QoL. GI bleed­ ing is
rarely sig­nif­i­cant in chil­dren except in those car­ry­ing the SMAD4
geno­type and then is typ­i­cally related to bleed­ing from pol­yps.
GI bleed­ing should be suspected in the pres­ence of iron defi­
ciency with or with­out ane­mia, par­tic­u­larly when this is out of
pro­por­tion to reported epi­staxis. The first step in eval­u­a­tion is
gen­er­ally refer­ral to a gas­tro­en­ter­ol­o­gist famil­iar with HHT for
upper endos­copy, as stool occult blood tests are not reli­able
in the set­ting of ongo­ing epi­staxis. Recommended treat­ment is
deter­mined based on the sever­ity of GI bleed­ing and coexisting
ane­mia; rec­om­men­da­tions for the workup and treat­ment of GI
bleed­ing can be found in Figure 4.
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/469/1852038/469hammill.pdf by guest on 13 December 2021
Figure 3. Epistaxis Severity Score (ESS)22 and management of epistaxis in HHT.1 *Therapy currently under investigation. ENT, ear,
nose, and throat; NSAID, nonsteroidal anti-inflammatory drug.

Anemia stud­ied in a pro­spec­tive man­ner in HHT, it has become the de

Anemia can result from epi­staxis, GI bleed­ing, or the com­bi­na­ facto stan­dard of care for severe/refrac­tory ane­mia in HHT. The
tion of the two and is pres­ent in more than 50% of adults with effi­cacy of the oral antiangiogenic agents pomalidomide (PATH-
HHT.26 Anemia, and even iron defi­ciency with­out ane­mia, should HHT trial; clinicaltrials​­.gov NCT03910244) and pazopanib (clin­i­cal
be man­aged aggres­sively to cor­rect ane­mia and main­tain ade­ trial open­ing soon) are being eval­u­ated pro­spec­tively in clin­i­
quate iron stores. In addi­tion to well-known signs and symp­toms cal tri­als fol­low­ing prom­is­ing data in pilot stud­ies.31,32 Additional
of ane­mia, includ­ing fatigue, short­ness of breath, lightheadedness, agents, includ­ing doxy­cy­cline33 (NCT03397004, NCT04167085)
pal­lor, and car­diac stress, there have been addi­tional asso­ci­a­tions and tacrolimus (NCT04646356), are also being actively eval­u­ated.
reported in patients with HHT. In patients with liver AVMs, iron
defi­ciency results in sub­stan­tial increased risk of aber­rant man­ Visceral AVMs
ga­nese depo­si­tion in the brain, lead­ing to Parkinson-like neu­ro­ Visceral AVMs, although less com­mon than bleed­ing symp­toms,
logic symp­toms.27 Paradoxically, an increased risk for throm­bo­sis can also cause sig­nif­i­cant mor­bid­ity and mor­tal­ity if unde­tected
has been reported in HHT, which appears to cor­re­late with iron and/or untreated. Therefore, it is impor­tant to be aware of poten­
defi­ciency.28 Such throm­botic events are gen­er­ally treated with tial signs and symp­toms of AVMs and fol­low cur­rent screen­ing
anticoagulation, with asso­ci­ated increased bleed­ing risk. While guide­lines. In doing so, indi­vid­u­als with HHT who undergo rec-
anticoagulation is not abso­lutely contraindicated in HHT, it can be ommended screen­ing and treat­ment at an HHT Center of Excel-
quite dif­fi­cult to bal­ance bleed­ing risks and clot­ting risks in these
lence can expect a nor­mal life span.34,35
patients. Complete rec­om­men­da­tions for man­age­ment of ane­mia Brain vas­cu­lar malformations seen in HHT include AVM, arte­
and anticoagulation in HHT can be found in Figure 5. rio­
ve­ nous fis­ tula (AVF), vein of Galen malformations, cere­ bral
cav­ern­ous malformations, devel­op­men­tal venous anom­a­lies, and
Our goal is to try to achieve and main­tain fer­ri­tin above 50 ng/mL cap­il­lary tel­an­gi­ec­ta­sia. Of these, AVF and AVM are con­sid­ered
in the set­ting of ongo­ing recur­rent bleed­ing. highest risk for com­pli­ca­tions. Symptoms of AVM vary by loca­
tion and can include sei­zure, head­ache, or, in the case of rup­ture,
For ane­mia resulting from epi­staxis or GI bleed­ing refrac­tory to hem­or­rhagic stroke; how­ever, most brain AVMs are asymp­tom­
local ther­a­pies and first-line med­i­cal treat­ments, sys­temic anti- atic and found on rou­tine screen­ing. Management of brain vas­
angiogenic ther­a­pies should be con­sid­ered in com­bi­na­tion with cu­lar malformations, once iden­ti­fied, is pur­sued col­lab­o­ra­tively
aggres­ sive iron replace­ ment. To date, the most widely used with neu­ro­ra­di­ol­ogy, neu­rol­ogy, neu­ro­sur­gery, and/or neuroin-
antiangiogenic agent in HHT is bevacizumab, a mono­clo­nal anti– terventional radi­ol­ogy. Treatment options for high-risk lesions
vas­cu­lar endo­the­lial growth fac­tor anti­body given as an intra­ve­ include embo­li­za­tion by a skilled neurointerventional radi­ol­o­gist,
nous infu­sion.29,30 Although sys­temicPrakash
bevacizumab has not been
Singh Shekhawatresec­tion by a neu­ro­sur­geon, or focal radi­a­tion (gamma knife or

The ABCs of HHT  |  473


Figure 4. Guidelines for the management of GI bleeding in HHT.1 *Mild GI bleeding defined as meeting hemoglobin goals with oral
iron. **Moderate GI bleeding defined as meeting hemoglobin goals with intravenous oral iron. ***Severe GI bleeding defined as not
meeting hemoglobin goals despite adequate iron replacement or requiring blood transfusions. CRC, colorectal cancer; EGD, esoph-
agogastroduodenoscopy.
pro­ton) admin­is­tered by a radi­a­tion oncol­o­gist; choice of ther­
apy will depend on risk of the lesion, size, and loca­tion, includ­ing
con­sid­er­ations of acces­si­bil­ity and elo­quence.
Lung AVMs can be asymp­tom­atic or may cause a num­ber of
signs and symp­toms. These include chronic man­i­fes­ta­tions such
as fatigue, short­ness of breath, hyp­oxia, asthma-like symp­toms
poorly respon­sive to β-ago­nists, migraine head­aches, or dra­matic
and life-threat­en­ing pre­sen­ta­tions such as par­a­dox­i­cal embo­li­
za­tion lead­ing to stroke, tran­sient ische­mic attack, or cere­bral
abscess. AVM rup­ture caus­ing mas­sive hemop­ty­sis and/or spon­
ta­ne­ous hemothorax is a rare pre­sen­ta­tion. Cerebral abscess risk
has been strongly asso­ci­ated with den­tal work, includ­ing rou­tine
cleaning, and anti­bi­otic pro­phy­laxis should be admin­is­tered prior
to den­tal work to all­patients with known lung AVMs and those
who have not yet under­gone lung AVM screen­ing to pre­vent
this com­pli­ca­tion. Many HHT cen­ters also rec­om­mend anti­bi­otic
pro­phy­laxis for other “dirty” pro­ce­dures that could poten­tially
intro­duce bac­te­ria into the blood­stream, includ­ing colonoscopy
and other GI/GU (genitourinary) pro­ce­dures, but this is not yet
stan­dard­ized. Treatment of lung AVMs is gen­er­ally car­ried out by
interventional radi­ol­o­gists using coil embo­li­za­tion. The deci­sion
to treat lung AVMs is based on size of the AVM nidus and diam­e­ter
of the feed­ing ves­sels (typ­i­cally >2-2.5 mm diam­e­ter depending
474  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/469/1852038/469hammill.pdf by guest on 13 December 2021
on the cen­ter) but may also be con­sid­ered for slightly smaller
AVMs based on symp­toms. After treat­ment, con­tin­ued mon­i­tor­
ing is nec­es­sary to rule out recan­a­li­za­tion or for­ma­tion of new
AVMs; inter­vals vary on a case-by-case basis. An update to the
pul­mo­nary AVM guide­lines is planned in the near future.
Liver AVMs are the most com­mon vis­ceral AVM found in HHT,
but they are often asymp­tom­atic and have been his­tor­i­cally hard
to treat. The updated guide­lines rec­om­mend screen­ing, as it may
be pos­si­ble to pick up sequelae of the hepatic AVMs ear­lier this
way, includ­ing high-out­put heart fail­ure, pul­mo­nary hyper­ten­
sion, por­tal hyper­ten­sion, and enceph­a­lop­a­thy. When pres­ent,
liver AVMs should be followed by a hepatologist but, depending
on the man­i­fes­ta­tions, may require car­di­­ol­ogy and/or pulmonol-
ogy or pul­mo­nary hyper­ten­sion exper­tise. Treatment options for
hepatic AVMs have been lim­ited, as embo­li­za­tion often leads to
fur­ther liver com­pli­ca­tions. Current approaches include orthot-
opic liver trans­plant or sys­temic med­i­cal ther­ap­ ies such as beva-
cizumab, reviewed in more detail above in the “Anemia” sec­tion.
Pregnancy and child­birth
Genetic coun­ sel­
ing is recommended pre­ con­cep­tion to dis­
cuss risks of trans­mis­sion and/or pre­na­tally to explore mul­ti­ple
options for genetic test­ing of the child, from pre­im­plan­ta­tion, to
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/469/1852038/469hammill.pdf by guest on 13 December 2021
Figure 5. Guidelines for the management of anemia and anticoagulation in HHT.1 HGB, hemoglobin.

Figure 6. Guidelines for pregnancy and delivery in HHT.1 CT, computed tomography; VM, vascular malformation.
Prakash Singh Shekhawat
The ABCs of HHT  |  475
in utero, to post­na­tal options such as cord blood. Guidelines for
preg­nancy and deliv­ery were also devel­oped, includ­ing screen­
ing rec­om­men­da­tions for AVMs, outlined in Figure 6. Women
who have not been screened recently or who have known lung
AVMs and/or those with high-risk brain vascular malformations
should be followed by a mul­ ti­
dis­
ci­
plin­
ary team, includ­ ing a
mater­nal-fetal med­i­cine spe­cial­ist, at a ter­tiary care cen­ter. Lung
AVMs in need of treat­ment should be embolized in the sec­ond
tri­mes­ter if pos­si­ble and patients coun­seled that hemop­ty­sis is
an emer­gency, based on increas­ing evi­dence of mater­nal mor­
bid­ity and even mor­tal­ity dur­ing preg­nancy and child­birth,36 due
at least in part to increased risk of lung AVM rup­ture.37,38
Conclusion
Patients with HHT can ini­tially pres­ent to any of sev­eral dif­fer­ent
med­i­cal spe­cial­ists, depending on their presenting symp­toms,
and underrecognition of the dis­ease risks diag­nos­tic delays and
increased mor­bid­ity. It is impor­tant that poten­tial pro­vid­ers have
a high index of sus­pi­cion in the pres­ence of the Curaçao clin­i­
cal cri­te­ria for HHT. These patients can pres­ent to hema­tol­o­gists
in many ways: with ane­mia and iron defi­ciency (due to bleed­
ing from nasal or GI tel­an­gi­ec­ta­ses), with poly­cy­the­mia (as a nor­
mal phys­i­o­logic response in the set­ting of pul­mo­nary AVMs), or
for sys­temic ther­a­pies for other com­pli­ca­tions of HHT (severe
ane­mia requir­ ing sig­
nif­i­
cant iron/blood replace­ ment or pul­
mo­nary hyper­ten­sion or high-out­put car­diac fail­ure related to
liver AVMs).39 Where HHT Centers of Excellence exist, cur­rently
at 30 sites in North America, the cen­ter’s direc­tor or codi­rec­tor
will likely pro­vide a med­ic ­ al home for these often com­pli­cated
patients; how­ever, no such cen­ter exists in most of the med­i­cal
cen­ters across the coun­try. Elsewhere, hema­tol­o­gist-oncol­o­gists
may find them­selves in the role of pri­mary pro­vider, par­tic­u­larly
as more targeted med­i­cal ther­a­pies become avail­­able as, to date,
most of these drugs have been used pre­vi­ously in the treat­ment
of can­cers. Given this and the rel­a­tively high fre­quency of HHT as
a hered­i­tary bleed­ing dis­or­der, an under­stand­ing of mod­ern HHT
man­age­ment is cru­cial for the prac­tic­ing hema­tol­o­gist.
Conflict-of-inter­est dis­clo­sure
Adrienne M. Hammill: none related to this paper. But AH has clin-
ical trials (for other diseases) with novartis, cerecor, Venthera,
Merck; had study drug provided by Pfizer; serves as a consultant
for Novartis.
Katie Wusik: none reported.
Raj S. Kasthuri: none reported.
Off-label drug
Adrienne M. Hammill: All drugs listed here are off label. None has
FDA approval. They are not attributable to a single author.
Katie Wusik: All drugs listed here are off label. None has FDA ap-
proval. They are not attributable to a single author.
Raj S. Kasthuri: All drugs listed here are off label. None has FDA
approval. They are not attributable to a single author.
Correspondence
Adrienne M. Hammill, Cancer and Blood Diseases Institute, Divi-
sion of Hematology, Cincinnati Children’s Hospital Medical Center,
and Department of Pediatrics, University of Cincinnati College of
Medicine, 3333 Burnet Ave, Cincinnati, OH 45245; e-mail: adrienne​
­.hammill@cchmc​­.org.
476  |  Hematology 2021  |  ASH Education Program
References
1. Faughnan ME, Mager JJ, Hetts SW, et al. Second International Guidelines for
the Diagnosis and Management of Hereditary Hemorrhagic Telangiectasia.
Ann Intern Med. 2020;173(12):989-1001.
2. Ginon I, Decullier E, Finet G, et al. Hereditary hem­or­rhagic tel­an­gi­ec­ta­sia,
liver vas­cu­lar malformations and car­diac con­se­quences. Eur J Intern Med.
2013;24(3):e35-e39.
3. Weber LM, McDonald J, Whitehead K. Vitamin D lev­els are asso­ci­ated with
epi­staxis sever­ity and bleed­ing dura­tion in hered­i­tary hem­or­rhagic tel­an­
gi­ec­ta­sia. Biomark Med. 2018;12(4):365-371.
4. Ratjen A, Au J, Carpenter S, John P, Ratjen F. Growth of pul­mo­nary arte­rio­ve­-
nous malformations in pedi­at­ric patients with hered­i­tary hem­or­rhagic tel­
an­gi­ec­ta­sia. J Pediatr. 2019;208:279-281.
5. McDonald J, Stevenson DA, Whitehead K. Incidence of epi­staxis and tel­an­
gi­ec­ta­ses in the gen­eral pop­u­la­tion. Angiogenesis. 2017;18:531.
6. Kjeldsen AD, Vase P, Green A. Hereditary haemorrhagic tel­an­gi­ec­ta­sia: a
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/469/1852038/469hammill.pdf by guest on 13 December 2021
pop­u­la­tion-based study of prev­a­lence and mor­tal­ity in Dan­ish patients.
J Intern Med. 1999;245(1):31-39.
7. McDonald J, Wooderchak-Donahue WL, Henderson K, Paul E, Morris A,
Bayrak-Toydemir P. Tissue-spe­cific mosa­i­cism in hered­i­tary hem­or­rhagic
tel­an­gi­ec­ta­sia: impli­ca­tions for genetic test­ing in fam­i­lies. Am J Med Genet
A. 2018;176(7):1618-1621.
8. Tørring PM, Kjeldsen AD, Ousager LB, Brusgaard K. ENG muta­tional mosa­
i­cism in a fam­ily with hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia. Mol Genet
Genomic Med. 2018;6(1):121-125.
9. Best DH, Vaughn C, McDonald J, et  al. Mosaic ACVRL1 and ENG muta­
tions in hered­i­tary haemorrhagic tel­an­gi­ec­ta­sia patients. J Med Genet.
2011;48(5):358-360.
10. Lee NP, Matevski D, Dumitru D, Piovesan B, Rushlow D, Gallie BL. Identifi-
cation of clin­i­cally rel­e­vant mosa­i­cism in type I hered­i­tary haemorrhagic
tel­an­gi­ec­ta­sia. J Med Genet. 2011;48(5):353-357.
11. Gedge F, McDonald J, Phansalkar A, et al. Clinical and ana­lyt­i­cal sensitivities
in hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia test­ing and a report of de novo
muta­tions. J Mol Diagn. 2007;9(2):258-265.
12. Shovlin CL, Guttmacher AE, Buscarini E, et al. Diagnostic cri­te­ria for hered­
i­tary hem­or­rhagic tel­an­gi­ec­ta­sia (Rendu-Osler-Weber syn­drome). Am J
Med Genet. 2000;91(1):66-67.
13. Pierucci P, Lenato GM, Suppressa P, et  al. A long diag­ nos­
tic delay in
patients with hered­i­tary haemorrhagic tel­an­gi­ec­ta­sia: a ques­tion­naire-
based ret­ro­spec­tive study. Orphanet J Rare Dis. 2012;7:33.
14. Pahl KS, Choudhury A, Wusik K, et al. Applicability of the Curaçao cri­te­ria
for the diag­no­sis of hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia in the pedi­at­ric
pop­u­la­tion. J Pediatr. 2018;197:207-213.
15. Faughnan ME, Palda VA, Garcia-Tsao G, et al; HHT Foundation International—
Guidelines Working Group. International guide­lines for the diag­no­sis and
man­age­ment of hered­i­tary haemorrhagic tel­an­gi­ec­ta­sia. J Med Genet.
2011;48(2):73-87.
16. McDonald J, Bayrak-Toydemir P, DeMille D, Wooderchak-Donahue W,
Whitehead K. Curaçao diag­nos­tic cri­te­ria for hered­i­tary hem­or­rhagic tel­
an­gi­ec­ta­sia is highly pre­dic­tive of a path­o­genic var­i­ant in ENG or ACVRL1
(HHT1 and HHT2). Genet Med. 2020;22(7):1201-1205.
17. Gallione CJ, Richards JA, Letteboer TG, et  al. SMAD4 muta­tions found in
unse­lected HHT patients. J Med Genet. 2006;43(10):793-797.
18. Gallione C, Aylsworth AS, Beis J, et al. Overlapping spec­tra of SMAD4 muta­
tions in juve­nile polyposis (JP) and JP-HHT syn­drome. Am J Med Genet A.
2010;152A(2):333-339.
19. Heald B, Rigelsky C, Moran R, et al. Prevalence of tho­racic aortopathy in
patients with juve­nile polyposis syn­drome–hered­i­tary hem­or­rhagic tel­an­
gi­ec­ta­sia due to SMAD4. Am J Med Genet A. 2015;167(8):1758-1762.
20. Hernandez F, Huether R, Carter L, et al. Mutations in RASA1 and GDF2 iden­
ti­fied in patients with clin­i­cal fea­tures of hered­i­tary hem­or­rhagic tel­an­gi­ec­
ta­sia. Hum Genome Var. 2015;2:15040.
21. Wooderchak-Donahue WL, Akay G, Whitehead K, et al. Phenotype of CM-
AVM2 caused by var­i­ants in EPHB4: how much over­lap with hered­i­tary
hem­or­rhagic tel­an­gi­ec­ta­sia (HHT)? Genet Med. 2019;21(9):2007-2014.
22. Hoag JB, Terry P, Mitchell S, Reh D, Merlo CA. An epi­staxis sever­ity score for
hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia. Laryngoscope. 2010;120(4):838-843.
23. Gaillard S, Dupuis-Girod S, Boutitie F, et al; ATERO Study Group. Tranexamic
acid for epi­staxis in hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia patients: a
Euro­pean cross-over con­trolled trial in a rare dis­ease. J Thromb Haemost.
2014;12(9):1494-1502.
24. Geisthoff UW, Seyfert UT, Kübler M, Bieg B, Plinkert PK, König J. Treatment of
epi­staxis in hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia with tranexamic acid—a
 dou­ble-blind pla­cebo-con­trolled cross-over phase IIIB study. Thromb Res.
2014;134(3):565-571.
25. Canzonieri C, Centenara L, Ornati F, et al. Endoscopic eval­u­a­tion of gas­tro­
in­tes­ti­nal tract in patients with hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia and
cor­re­la­tion with their geno­types. Genet Med. 2014;16(1):3-10.
26. Kasthuri RS, Montifar M, Nelson J, et al; Brain Vascular Malformation Consor-
tium HHT Investigator Group. Prevalence and pre­dic­tors of ane­mia in hered­
i­tary hem­or­rhagic tel­an­gi­ec­ta­sia. Am J Hematol. 2017;92(10):e591-e593.
27. Serra MM, Besada CH, Cabana Cal A, et al. Central ner­vous sys­tem man­ga­
nese induced lesions and clin­i­cal con­se­quences in patients with hered­i­tary
hem­or­rhagic tel­an­gi­ec­ta­sia. Orphanet J Rare Dis. 2017;12(1):92.
28. Livesey JA, Manning RA, Meek JH, et  al. Low serum iron lev­els are asso­
ci­ated with ele­vated plasma lev­els of coag­u­la­tion fac­tor VIII and pul­mo­
nary emboli/deep venous throm­bo­ses in rep­li­cate cohorts of patients with
hered­i­tary haemorrhagic tel­an­gi­ec­ta­sia. Thorax. 2012;67(4):328-333.
29. Iyer VN, Apala DR, Pannu BS, et al. Intravenous bevacizumab for refrac­tory
hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia-related epi­staxis and gas­tro­in­tes­ti­
nal bleed­ing. Mayo Clin Proc. 2018;93(2):155-166.
30. Al-Samkari H, Kasthuri RS, Parambil JG, et al. An inter­na­tional, mul­ti­cen­ter
study of intra­ve­nous bevacizumab for bleed­ing in hered­i­tary hem­or­rhagic
tel­an­gi­ec­ta­sia: the InHIBIT-Bleed study. Haematologica. 2020;106(8):2161-
2169.
31. McCrae K, Swaidani S, Samour M, Silver B, Parambil J, Thomas S. Pomalido-
mide in HHT: results of a pilot study. Angiogenesis. 2019;22(4):624.
32. Faughnan ME, Gossage JR, Chakinala MM, et al. Pazopanib may reduce bleed­
ing in hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia. Angiogenesis. 2019;22(1):145-
155.
Prakash Singh Shekhawat
33. Meek M, Womble P, Jordan A, Kanaan A, Meek J. Oral doxy­cy­cline for the
treat­ment of epi­staxis in patients with hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­
sia. Angiogenesis. 2015;18(4):530-530.
34. de Gussem EM, Edwards CP, Hosman AE, et  al. Life expec­tancy of par­
ents with hered­i­tary haemorrhagic tel­an­gi­ec­ta­sia. Orphanet J Rare Dis.
2016;11:46.
35. Iyer VN, Brinjikji W, Pannu BS, et al. Effect of cen­ter vol­ume on out­comes
in hos­pi­tal­ized patients with hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia. Mayo
Clin Proc. 2016;91(12):1753-1760.
36. Dupuis O, Delagrange L, Dupuis-Girod S. Hereditary haemorrhagic tel­an­
gi­ec­ta­sia and preg­nancy: a review of the lit­er­a­ture. Orphanet J Rare Dis.
2020;15(1):5.
37. de Gussem EM, Lausman AY, Beder AJ, et  al. Outcomes of preg­nancy in
women with hered­i­tary hem­or­rhagic tel­an­gi­ec­ta­sia. Obstet Gynecol.
2014;123(3):514-520.
38. Shovlin CL, Sodhi V, McCarthy A, Lasjaunias P, Jackson JE, Sheppard MN.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/469/1852038/469hammill.pdf by guest on 13 December 2021
Estimates of mater­nal risks of preg­nancy for women with hered­i­tary hae-
morrhagic tel­an­gi­ec­ta­sia (Osler-Weber-Rendu syn­drome): suggested
approach for obstet­ric ser­vices. BJOG. 2008;115(9):1108-1115.
39. Al-Samkari H. Hereditary hem­or­rhagic tel­an­gi­ec­ta­sia: sys­temic ther­a­pies,
guide­lines, and an evolv­ing stan­dard of care. Blood. 2021;137(7):888-895.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000281
The ABCs of HHT  |  477
 MULTIDISCIPLINARY HEMATOLOGIC DISORDERS: WHAT IS THE HEMATOLOGIST ’ S ROLE ?

Chronic thromboembolic pulmonary

hypertension: anticoagulation and beyond

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/478/1852108/478martin.pdf by guest on 13 December 2021

Karlyn A. Martin1 and Michael J. Cuttica2,3
Division of Hematology/Oncology and 2Division of Pulmonary and Critical Care Medicine, Department of Medicine, Northwestern University
1

Feinberg School of Medicine, Chicago, IL; and 3Bluhm Cardiovascular Institute at Northwestern Memorial Hospital, Chicago, IL

Chronic thromboembolic pulmonary hypertension (CTEPH) is a rare complication in pulmonary embolism (PE) survivors,
characterized by chronic vascular occlusion and pulmonary hypertension. The identification and diagnosis of CTEPH
requires a stepwise approach, starting with symptom evaluation, functional evaluation, screening imaging, and progress-
ing to interventional hemodynamic assessment. On the backbone of anticoagulation, CTEPH management necessitates
a multidisciplinary approach. Surgical pulmonary thromboendarterectomy (PTE) is the only potentially curative option.
In nonoperable disease or residual disease after PTE, interventional balloon pulmonary angioplasty and/or pulmonary-
vasodilator therapies can be offered, in collaboration with interventional and vascular pulmonary colleagues. As it is a
disease that can cause high morbidity and mortality, CTEPH requires a high index of suspicion to diagnose and treat in
patients following PE.

LEARNING OBJECTIVES
• Understand that CTEPH is a rare complication of acute PE that has high morbidity and mortality, necessitating a
high index of suspicion
• Apply diagnostic algorithms to evaluate patients for CTEPH and understand multidisciplinary treatment strategies,
including anticoagulation, surgical and interventional treatments, and pulmonary hypertension therapy

that is associated with impaired quality of life, dyspnea,

CLINICAL CASE
A 55-year-old man with a history of unprovoked pulmo-
nary embolism (PE) 18 months earlier sought treatment for
shortness of breath. Although he initially improved, he never
returned to baseline functional status. For the past 6 months,
dyspnea on exertion has progressed despite appropriate anti-
coagulation. His oxygen saturation is 92%. Computed tomog-
raphy (CT) angiogram is obtained and reported to show PE
(Figure 1A). Subsequent echocardiogram shows a dilated
right ventricle (RV) with severely reduced right ventricular
function (Figure 1B). He was diagnosed with submassive PE
and underwent catheter directed thrombolysis. Following
thrombolysis, his hypoxemia worsened, and he was trans-
ferred to our institution for consideration of embolectomy.
Introduction
As many as 50% of patients have a chronic functional limita-
tion up to 1 year after PE, termed the “post-PE syndrome,”
and reduced exercise tolerance.1,2 Post-PE syndrome is de-
fined as the presence of functional or cardiac impairment
(without another non-PE explanation), chronic thrombo-
embolic disease (CTED), or chronic thromboembolic pul-
monary hypertension (CTEPH), occurring after at least 3
months of effective anticoagulation for acute PE.3 CTED
and CTEPH share common features of exertional dyspnea
and persistent thromboembolic material in the pulmonary
artery tree. However, CTED lacks pulmonary hypertension
(PH) at rest, whereas in CTEPH, resting PH is present, as
defined by a mean pulmonary artery pressure (mPAP) of
25mm Hg or more, and a pulmonary capillary wedge pres-
sure of 15mm Hg or less.3 Of note, a change to decrease
mPAP to more than 20mm Hg to define PH has been pro-
posed but is not yet incorporated into diagnostic criteria
of CTEPH.4,5
Although a significant number of patients are diag-
nosed with CTEPH without a known prior acute PE (estima-
tes vary from 25% to 67%),6,7 hematologists are more likely
to encounter CTEPH following acute PE. CTEPH affects

478 | Hematology 2021 | ASH Education Program


Figure 1. Clinical Imaging from a representative case. PTE, pulmonary thromboendarterectomy.
approx­ i­
ma­tely 1% to 5% of PE sur­ vi­
vors.8 One meta-anal­y­sis
dem­on­strated the pooled inci­dence of CTEPH fol­low­ing acute
PE to be 0.56% (95% CI, 0.1%-1.0%) for “all­-com­ers,” 3.2% (95%
CI, 2.0%-4.4%) for “sur­vi­vors,” and 2.8% (95% CI, 1.5%-4.1%) for
“sur­vi­vors with­out major comorbidities.”8 However, when the
expected inci­dence of CTEPH based on the num­ber of inci­dent
acute PE cases per year is com­pared with the num­ber of pul­mo­
nary thromboendarterectomy (PTE) sur­ger­ies performed each
year, it is highly likely that CTEPH is underdiagnosed and there­
fore undertreated. A study mod­el­ing epi­de­mi­o­logic data from
Europe, Japan, and the United States, for exam­ple, esti­mated
that only 16% of CTEPH cases would be diag­nosed in 2015, and
70% or more of those diag­nosed would be in the set­ting of
advanced heart fail­ure.9 Furthermore, data from both a sur­vey of
inter­na­tional phy­si­cians and ret­ro­spec­tive claims study showed
that diag­nos­tic tests to eval­u­ate CTEPH are under­used.10,11 As
CTEPH has high mor­bid­ity and mor­tal­ity—caus­ing pre­ma­ture
death in more than 50% of untreated patients within 5 years
of diag­no­sis—but has poten­tially cura­tive inter­ven­tions, a high
index of sus­pi­cion is crit­i­cal to iden­tify the dis­ease.12
When to sus­pect CTEPH
CTEPH and/or CTED should be con­sid­ered in PE sur­vi­vors with
per­sis­tent dyspnea for more than 3 months after a diag­no­sis of
acute PE, despite ade­quate anticoagulation, or in those who ini­
tially improve but sub­se­quently develop wors­en­ing func­tional
lim­i­ta­tions and dyspnea with exer­cise between 3 and 24 months
fol­low­ing acute PE. Onset of CTEPH is rare after 24 months fol­
low­ing ini­tial PE diag­no­sis.13 In addi­tion, CTEPH should be sus-
pected if echo­ car­
dio­gram obtained for suspected acute PE
shows increased RV wall thick­ness or tri­cus­pid valve peak sys­tolic
gra­di­ent more than 60 mm Hg, both of which sug­gest changes
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/478/1852108/478martin.pdf by guest on 13 December 2021
beyond acute RV over­load.14 Symptoms of CTEPH can be insid­
i­ous, such as exer­cise intol­er­ance and dyspnea on exer­tion, or
more severe, such as leg swell­ing, chest pain, and syn­cope that
can occur with right heart (RH) fail­ure. As screen­ing echo­car­dio­
gram in all­PE sur­vi­vors has a high false-pos­i­tive rate,15 eval­u­a­tion
for CTEPH should be reserved for those with symp­toms.16
Numerous risk fac­tors have been his­tor­i­cally asso­ci­ated with
CTEPH, includ­ing sple­nec­tomy, chronic inflam­ma­tory dis­or­
ders, indwell­ing cath­e­ters, ele­vated fac­tor VIII, and unpro­voked
or recur­ rent venous thromboembolism (VTE), among oth­ ers,
reviewed else­where.17 More recently, in a post hoc patient-level
anal­y­sis of 3 large pro­spec­tive cohorts of more than 700 PE sur­
vi­vors, of whom 2.8% devel­oped CTEPH, pre­dic­tors of CTEPH
were unpro­voked PE (odds ratio [OR], 20; 95% CI, 2.7 to >100),
onset of symp­toms more than 14 days prior to diag­no­sis (OR, 7.9;
95% CI, 3.3-19), hypo­thy­roid­ism (OR, 4.3; 95% CI, 1.4-13), and RV
dys­func­tion on pre­sen­ta­tion (OR, 4.1; 95% CI, 1.4-12), whereas
dia­be­tes mellitus and throm­bo­lytic ther­apy had infi­nitely low OR
for devel­op­ing CTEPH.18
How to iden­tify and diag­nose CTEPH
When CTEPH/CTED is suspected, a step­wise eval­u­a­tion aims
to iden­tify pul­mo­nary vas­cu­lar dis­ease related to nonresolving
throm­bus (Figure 2). The diag­nos­tic eval­u­a­tion also con­cur­
rently allows for assess­ment of treat­ment options and sur­gi­cal
can­di­dacy.
Although no uni­ver­sally agreed-on diag­nos­tic algo­rithm
exists to eval­u­ate for CTEPH in a patient post-PE with dyspnea,
we start with an echo­car­dio­gram cou­pled with a basic func­
tional test such as the 6-min­ ute walk test (6MWT).19 Several
guide­lines also pro­pose diag­nos­tic strat­e­gies.4,14 Notably, how­
ever, if PH is strongly suspected, then we elim­i­nate the 6MWT.
CTEPH: anticoagulation and beyond  |  479
 Persistent symptoms despite 3 months of
ancoagulaon following acute PE

Transthoracic echocardiogram Equivocal findings; suspicion

and for CTEPH/CTED remains
6-minute walk test

Persistent or new right heart changes Cardiopulmonary exercise tesng

and/or funconal limitaon on walk test (CPET)

V/Q scan Findings suggesve of

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/478/1852108/478martin.pdf by guest on 13 December 2021

+/- CT angiogram pulmonary vascular limitaon

Refer to CTEPH center

Right heart catheterizaon with pulmonary

angiogram
+/- CT angiogram

Assessment of operability by
muldisciplinary CTEPH team

Technically operable Technically inoperable

PTE Medical therapy BPA

Figure 2. Diagnostic algorithm in evaluation for CTEPH/CTED.

Echocardiogram is used to esti­mate prob­a­bil­ity of PH, with inter­
me­di­ate to high prob­a­bil­ity suggested by esti­mated peak tri­cus­
pid valve regur­gi­ta­tion veloc­ity more than 2.8m/s, or 2.8 m/s
or less associated with at least one of the following: RV end
dia­stolic diam­et­er (EDD) more than 30 mm or RVEDD/left ven-
tricular EDD more than 0.9, hypokinesia of the RV free wall, or
exertional dyspnea.3,14 Importantly, echo­car­dio­gram allows for
screen­ing with­out expo­sure to radi­a­tion or con­trast dye.20 The
6MWT is a sim­ple test that can be performed in any clinic. As
a patient walks for 6 min­utes, dis­tance walked, heart rate, and
oxy­gen sat­ur­a­tion by pulse oximetry are mea­sured. The 6MWT
pro­vi­des a basic assess­ment of car­dio­pul­mo­nary func­tion and
fit­ness, along with infor­ma­tion on func­tional, car­dio­vas­cu­lar,
and gas exchange response to exer­cise. Abnormalities in either
echo­car­dio­gram or 6MWT in a symp­tom­atic patient more than
3 months beyond acute PE should trig­ger fur­ther eval­u­a­tion.
Occasionally, more advanced exer­cise test­ing is needed to
eval­u­ate per­sis­tent symp­toms. Formal car­dio­pul­mo­nary exer­cise
test­ing plays a key role in the eval­u­a­tion of dyspnea in a patient
post-PE, and guide­lines sug­gest using car­dio­pul­mo­nary exer­cise
test­ing in patients with low prob­a­bil­ity of PH on echo­car­di­og­ra­
phy but con­tin­ued sus­pi­cion for CTEPH/CTED based on symp­
toms.16 The pat­tern of increased dead space ven­ti­la­tion with a
wid­en­ing A-a gra­di­ent and flat­tened stroke vol­ume in response
to exer­cise sug­gests a pul­mo­nary vas­cu­lar lim­i­ta­tion and need
for fur­ther inva­sive test­ing.
Once a func­tional lim­i­ta­tion and/or appro­pri­ate echo­car­dio­
graphic abnor­mal­i­ties raise suf­fi­cient con­cern for CTEPH/CTED,
the next step in eval­u­a­tion is documenting unre­solved vas­cu­lar
occlu­sion. Although a full dis­cus­sion of the roles of CT angi­og­ra­phy
(CTA) and ven­ti­la­tion-per­fu­sion (VQ ) scan­ning in the eval­u­a­tion of
CTEPH is beyond the scope of this arti­cle, both imag­ing modal­i­
ties can add impor­tant infor­ma­tion in the assess­ment and deter­
min­ing appro­pri­ate inter­ven­tion (Table 1). While the gap between
sen­si­tiv­ity of V/Q and CTA scan is narrowing, VQ scan remains
the pre­ferred test for screen­ing for CTEPH/CTED,4 as nonocclu-
sive changes to the ves­sels that occur in CTEPH (such as mural
thrombi, webs, and stric­tures) can be missed on rou­tine CTA reads
but are iden­ti­fia­ ble on VQ because of their effect on per­fu­sion to
dis­tal areas of the lung.21 Furthermore, VQ has lower expo­sure

480  |  Hematology 2021  |  ASH Education Program

Table 1. Comparison of diag­nos­tic tests used in the eval­u­a­tion for CTEPH

Diagnostic test Findings in CTEPH Advantages Disadvantages/lim­i­ta­tions

Imaging
 Echocardiogram • Evidence of PH or RH strain: RV
dila­tion, RV sys­tolic dys­func­tion, RA
dila­tion, PASP ele­va­tion, flat­ten­ing of
inter­ven­tric­u­lar sep­tum
 VQ • Mismatched defects in per­fu­sion and
ven­ti­la­tion defects
•  Heterogeneity of per­fu­sion
  CT PA • Pulmonary arteries: PA dila­tion,
webs/bands, eccen­tric fill­ing
•  Non-inva­sive
• No expo­sure to radi­a­tion or con­
trast dye
• Highest sen­si­tiv­ity to rule out
CTEPH
•  Less radi­a­tion expo­sure
•  No con­trast dye expo­sure
• Defining vas­cu­lar anat­omy can
aid in sur­gi­cal assess­ment
•  Not spe­cific or sen­si­tive to CTEPH
•  Misses CTED
• Limited access as performed in nuclear
med­i­cine
•  Unable to pro­vide alter­na­tive diag­no­sis
•  Sensitive but not spe­cific for CTEPH
• CTEPH find­ings can some­times be sub­
tle and require exper­tise and atten­tion

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/478/1852108/478martin.pdf by guest on 13 December 2021

defects, mural thrombi, lumi­nal
narrowing with poststenotic dila­tion,
com­plete occlu­sion, pouch defects
•  Heart: RV dila­tion, sep­tal flat­ten­ing
• Lungs: mosaic atten­u­a­tion, large
bron­chial artery col­lat­er­als
Invasive test­ing
  Pulmonary angio­gram •  Ring lesions/ring-like ste­no­sis
•  Poststenotic dila­tion
•  Total occlu­sion
•  Vascular webs
  RH cath­e­ter­i­za­tion •  mPAP ≥25 mm Hg
•  Wedge ≤15 mm Hg
Functional test­ing
 Cardiopulmonary • Increased dead space ven­ti­la­tion
exer­cise test­ing with wid­en­ing A-a gra­di­ent, flat­tened
stroke vol­ume in response to exer­cise
  6-min­ute walk test • Functional lim­i­ta­tions and decreased
oxy­gen sat­u­ra­tion with exer­cise
IV, intra­ve­nous; PASP, pulmonary artery systolic pressure; RA, right atrium.
• Provides data on screen­ing for
con­com­i­tant lung dis­ease/alter­
nate diag­no­sis that can aid in
sur­gi­cal risk assess­ment
• Provides com­plete hemo­dy­
namic assess­ment and aids in
sur­gi­cal plan­ning
• Provides com­plete hemo­dy­
namic assess­ment and aids in
sur­gi­cal plan­ning
• Can detect lim­i­ta­tions in func­tion
and car­dio­pul­mo­nary response
to exer­cise
• Simple to per­form, low cost,
min­i­mal risk
to diag­nose
•  Exposure to IV con­trast
•  Exposure to radi­a­tion
• Less sen­si­tive than VQ (neg CT does
not exclude CTEPH)
• Invasive
• Invasive
•  Requires spe­cial­ized test­ing unit
• Requires arte­rial blood gas
mea­sure­ment
•  Not spe­cific to CTEPH

to radi­a­tion and no con­trast dye expo­sure.21 Notably, how­ever,
VQ scans are under­used: in 1 study, 43% of patients under­go­ing
workup for PH did not get the recommended VQ scan.22
After asso­ci­at­ing ongo­ing symp­toms, func­tional lim­i­ta­tion, and/
or echo­car­dio­graphic changes to per­fu­sion abnor­mal­i­ties on imag­
ing, patients should be referred to an expe­ri­enced CTEPH cen­ter
for right heart cath­e­ter­i­za­tion (RHC) with pul­mo­nary angio­gram (PA
gram).16 The diag­no­sis of PH of any kind, includ­ing CTEPH, relies on
a com­pre­hen­sive hemo­dy­namic assess­ment com­pleted dur­ing an
RHC. These data are crit­i­cal for confirming the cor­rect PH diag­no­
sis and pro­vid­ing prog­nos­tic infor­ma­tion that informs treat­ment
deci­sions.23 Coupling a diag­nos­tic RHC with PA gram to bet­ter
define chronic throm­bo­em­bolic lesions not only allows for diag­
nos­tic and prog­nos­tic infor­ma­tion but are the final data needed to
deter­mine sur­gi­cal can­di­dacy.
CLINICAL CASE (Con­t in­u ed)
On arrival to our insti­tu­tion, the patient had per­sis­tence of his
chronic dyspnea despite thrombolysis and anticoagulation,
which raised sus­pi­cion for CTEPH. On review of CTA by expe­
ri­enced radi­ol­o­gists, find­ings of eccen­tric mural thrombi and
webs were more con­sis­tent with CTEPH, and VQ scan con­
Prakash Singh Shekhawat
firmed marked ven­ti­la­tion per­fu­sion mis­matches (Figure 1A and
1C). He under­went RHC with PA gram, which con­firmed CTEPH.
Management con­sid­er­ations of CTEPH
The back­bone of treat­ment of CTEPH requires life­long anticoag-
ulation to pre­vent acute VTE recur­rence. There are also 3 other
treat­ment arms (detailed below) used to address the chronic
vas­cu­lar occlu­sions and PH to improve hemo­dy­nam­ics and qual­
ity of life: (1) sur­gi­cal PTE, (2) interventional bal­loon pul­mo­nary
angio­plasty (BPA), and (3) pul­mo­nary vaso­di­la­tor med­i­ca­tions.
Treatment strat­egy is selected based on an indi­vid­ual patient
and hemo­dy­namic char­ac­ter­is­tics. Although the focus of PTE
and BPA is typ­i­cally patients with CTEPH, it is impor­tant to note
that patients with CTED may also be offered sur­gi­cal or pro­ce­
dural inter­ven­tions if symp­toms affect qual­ity of life.16 Although
anticoagulation falls within the hema­ tol­
ogy exper­
tise, defin­
i­
tive sur­gi­cal, interventional, and PH med­i­cal ther­apy for CTEPH
neces­si­tates a mul­ti­dis­ci­plin­ary team involv­ing hema­tol­ogy, car­
dio­tho­racic sur­gery, radi­ol­ogy, car­di­­ol­ogy, and pulmonology.
Anticoagulation
Anticoagulation is recommended as stan­dard of care for CTEPH
(so long as min­ i­
mal bleed­ ing risk). Of note, whether chronic
anticoagulation is indi­cated in CTED is not clear, and no guide­
CTEPH: anticoagulation and beyond  |  481
lines offer for­mal rec­om­men­da­tions; in our prac­tice, we con­tinue
anticoagulation in symp­tom­atic CTED to pre­vent VTE recur­rence
if low bleed­ing risk. Vitamin K antag­o­nists (VKAs) are cur­rently
recommended as the anti­co­ag­u­lant of choice in patients with
CTEPH given his­tor­i­cal expe­ri­ence, as data on direct-acting oral
anticoagulants (DOACs) are lim­ited.16 A ret­ro­spec­tive study com­
pared out­comes fol­low­ing PTE in those treated with war­fa­rin
(n = 700) and DOACs (n = 200) and found sig­nif­i­cantly higher rates
of VTE recur­rence with DOACs (4.62%/per­son-year) com­pared
with VKA (0.76%/per­ son-year) (P = .008), with no dif­ fer­
ence in
over­all sur­vival and sim­i­lar rates of major bleed­ing (0.67%/per­son-
year vs 0.68%/per­son-year for VKA and DOAC, respec­tively).24
Furthermore, VTE recur­rence occurred at a median of 5.8 months
(interquartile range 5.4 months) after PTE, suggesting choice of
anti­co­ag­u­lant in the first 6 months may be par­tic­u­larly impor­tant,
although this requires fur­ther eval­u­a­tion. In the absence of robust
data, in our prac­tice, we treat with VKA for 6 months fol­low­ing
PTE and then allow for switch to DOAC per patient pref­er­ence.
Prospective stud­ies are needed to deter­mine opti­mal anticoagu-
lation strat­egy fol­low­ing PTE. In addi­tion, pulmonologists and/or
car­dio­tho­racic sur­geons may need hema­tol­ogy exper­tise on
opti­mal anti­co­ag­u­lants in dis­tinct clin­i­cal sit­u­a­tions, such as hep­
a­rin-induced throm­bo­cy­to­pe­nia and antiphospholipid syn­drome
(APS). Reported rates of APS in CTEPH cohorts range from 3%
to 7%,25 and although rou­tine thrombophilia test­ing does not af-
fect man­age­ment in patients with CTEPH who require life­long
anticoagulation, the pres­ence of APS would, given evi­dence of
higher rates of recur­rent throm­bo­sis in patients with APS tak­ing
rivaroxaban com­pared with VKA.26,27
Pulmonary thromboendarterectomy
PTE remains the only poten­tially cura­tive option for CTEPH and
thus is con­sid­ered the gold stan­dard where appro­pri­ate. All pa-
tients should be referred to an expe­ri­enced cen­ter to deter­mine
whether obstruc­tive pul­mo­nary artery lesions are ame­na­ble to
sur­gi­cal removal (“tech­ni­cally oper­a­ble”) and whether the patient
is a sur­gi­cal can­di­date. As PTE is tech­ni­cally chal­leng­ing, patients
should be referred to a pro­gram with an expe­ri­enced sur­geon,
con­sid­ered to be more than 30 to 50 PTEs performed annu­ally.16
Between 10% and 50% of patients are deemed nonoperable,6 and
these patients should be referred for a sec­ond opin­ion at a cen­
ter with sig­nif­i­cant expe­ri­ence. Current in-hos­pi­tal mor­tal­ity rates
fol­low­ing PTE are less than 5% (<2% in expe­ri­enced cen­ters),28
and sur­vival exceeds more than 90% at 1 year and more than 70%
at 10 years.29 Hematologists are fre­quently asked about the role
of inferior vena cava (IVC) fil­ters perioperatively. Although IVC fil­
ters have fallen out of favor prior to most sur­ger­ies given a lack
of clear mor­tal­ity ben­e­fit and increased com­pli­ca­tion risks, there
are no ded­i­cated pro­spec­tive or obser­va­tional stud­ies spe­cif­i­
cally in the CTEPH pop­u­la­tion.30 However, given that this patient
pop­u­la­tion is com­mit­ted to life­long anticoagulation, the rou­tine
place­ment of IVC fil­ters is not cur­rently recommended.31
Balloon pul­mo­nary angio­plasty
BPA is avail­­able for (1) inop­er­a­ble CTEPH, because of the pres­
ence of either dis­tal lesions not ame­na­ble to sur­gi­cal inter­ven­
tion (“tech­ni­cally inop­er­a­ble”) or comorbidities that pre­clude
sur­gery, or (2) per­sis­tent or recur­rent CTEPH fol­low­ing PTE. BPA
is performed by cath­e­ter­i­za­tion via fem­o­ral or jug­u­lar access and
does not require gen­eral anes­the­sia or cir­cu­la­tory arrest. Using
482  |  Hematology 2021  |  ASH Education Program
image guid­ance, a bal­loon is inflated at the site of the obstruc­
tive lesion to com­ press the intra­ vas­
cu­
lar fibrotic occlu­ sion
against the wall, in an attempt to restore the vas­cu­lar lumen.
Most patients require mul­ti­ple ses­sions for BPA, with an num­
ber of ses­sions between 4 and 8, typ­i­cally over sev­eral months.32
Studies have shown effi­cacy of BPA in reduc­ing pulmonary vas-
cular resistance (PVR) and increas­ing func­tional activ­ity, with low
rates of com­pli­ca­tions.33,34 Postprocedural com­pli­ca­tions occur
in approx­i­ma­tely 10% of patients, includ­ing pul­mo­nary vas­cu­lar
injury (eg, wire per­fo­ra­tion), reperfusion injury, and pul­mo­nary
hem­or­rhage. Critically, hema­tol­o­gists should be aware of BPA as
a treat­ment option and refer patients with nonoperable CTEPH
to a cen­ter with BPA exper­tise.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/478/1852108/478martin.pdf by guest on 13 December 2021
Medical ther­apy of PH
Medical ther­apy of PH should be addressed by pul­mo­nary vas­
cu­lar spe­cial­ists with exper­tise in PH. Currently, riociguat is the
only approved ther­apy for the treat­ment of inop­er­a­ble or resid­
ual/recur­rent CTEPH fol­low­ing PTE. A sol­u­ble guanylate cyclase
stim­u­la­tor, riociguat works via the nitric oxide path­way to pro­
mote vaso­di­la­ta­tion of the pul­mo­nary arte­rial bed and has been
shown to improve 6-min­ute walk dis­tance and decrease PVR.35
Published case series and clin­i­cal tri­als have explored the role
of other pul­mo­nary vaso­di­la­tors such as syn­thetic pros­ta­cy­
clins and endothelin recep­tor antag­o­nists for the treat­ment of
CTEPH, but riociguat cur­rently remains the only approved ther­
apy.36-38 Although the role of targeted PH ther­a­pies in patients
with oper­ a­ble CTEPH remains unclear, lim­ ited avail­­
able data
do not sup­port a role for “pre­treat­ment” pend­ing PTE sur­gery.
Further research is needed to deter­mine whether high-risk sub­
sets of patients, such as those with very ele­vated PVR, could
ben­e­fit.39,40
How to man­age patients with CTEPH after PTE
Long-term man­age­ment of CTEPH involves anticoagulation and
mon­i­tor­ing for recur­rent symp­toms. Currently, all­ patients who
have a CTEPH diag­no­sis are maintained on indef­i­nite anticoagu-
lation to pre­vent recur­rent VTE. The opti­mal inten­sity is unclear;
cur­rently, most are maintained on ther­a­peu­tic doses if bleed­
ing risk is low. Although PTE is poten­tially cura­tive, up to 35%
of patients have a mPAP of 25 mm Hg or more post-PTE.41-43 In
addi­tion, most stud­ies fol­low patients for only 5 years after PTE,
and recur­rent PH may occur as late as 10 years after sur­gery, sug-
gesting inci­dence of recur­rent PH may be even higher.41 There-
fore, long-term mon­i­tor­ing is essen­tial, with par­tic­u­lar atten­tion
on assessing for symp­toms sug­ges­tive of recur­rent dis­ease. If
recur­rent CTEPH occurs, options include BPA and/or med­i­cal PH
ther­apy. In rare instances of recur­rent severe dis­ease, repeat PTE
may be performed.
Conclusion
CTEPH is a rare but sig­ nif­i­
cant com­pli­ ca­tion in PE sur­ vi­
vors,
with high rates of mor­bid­ity and mor­tal­ity. Therefore, it requires
aware­ness of and atten­tive­ness to symp­toms, which, if pres­
ent, should prompt an appro­pri­ate diag­nos­tic eval­u­a­tion. Once
CTEPH is diag­nosed, man­age­ment requires mul­ti­dis­ci­plin­ary
care. Although PTE remains the only poten­ tially cura­
tive op-
tion, BPA and med­i­cal ther­apy for PH are options for nonoper-
able CTEPH that can improve symp­toms and car­dio­pul­mo­nary
func­tion. Further stud­ies should eval­ u­ ate opti­ mal long-term
anti­co­ag­u­lant agent and inten­sity for both CTEPH and CTED, the
pre­ferred strat­egy (CTEPH vs BPA) for dif­fer­ent patient risk pro­
files, and role of com­bi­na­tion ther­apy.
CLINICAL CASE (Con­t in­u ed)
Upon review of the patient’s case in a mul­ti­dis­ci­plin­ary CTEPH
con­fer­ence, he was deemed a PTE can­di­date based on clot dis­
tri­bu­tion, asso­ci­ated per­fu­sion abnor­mal­i­ties and hemo­dy­namic
changes, age, and lack of comorbidities. He under­went suc­cess­
ful PTE with excel­lent hemo­dy­namic response (Figure 1D and 1E).
Two years after PTE, he con­tin­ues on war­fa­rin. He has no symp­
toms of dyspnea on exer­tion or func­tional lim­i­ta­tions.
Conflict-of-inter­est dis­clo­sure
Karlyn A. Martin reports funding from Janssen Scientific Affairs for
an inves­ti­ga­tor-ini­ti­ated study out­side the sub­mit­ted work.
Michael J. Cuttica reports grants and con­sul­ting fees from Bayer,
United Therapeutics, and Actelion.
Off-label drug use
Karlyn A. Martin: nothing to disclose.
Michael J. Cuttica: nothing to disclose.
Correspondence author
Karlyn A. Martin, Division of Hematology/Oncology, Department
of Medicine, Northwestern University Feinberg School of Medi-
cine, 645 N Michigan Ave, Suite 1020, Chicago, IL 60611; e-mail:
karlyn​­.martin@north­west­ern​­.edu.
References
1. Kahn SR, Hirsch AM, Akaberi A, et  al. Functional and exer­cise lim­i­ta­tions
after a first epi­sode of pul­mo­nary embolism: results of the ELOPE pro­spec­
tive cohort study. Chest. 2017;151(5):1058-1068.
2. Sanchez O, Helley D, Couchon S, et  al. Perfusion defects after pul­mo­
nary embolism: risk fac­tors and clin­i­cal sig­nif­i­cance. J Thromb Haemost.
2010;8(6):1248-1255.
3. Le Gal G, Carrier M, Castellucci LA, et al; ISTH CDE Task Force. Development
and implementation of com­mon data ele­ments for venous throm­bo­em­bo­
lism research: on behalf of SSC Subcommittee on offi­cial Communication
from the SSC of the ISTH. J Thromb Haemost. 2021;19(1):297-303.
4. Delcroix M, Torbicki A, Gopalan D, et al. ERS state­ment on chronic throm­
bo­em­bolic pul­mo­nary hyper­ten­sion. Eur Respir J. 2021;57(6):2002828.
5. Simonneau G, Montani D, Celermajer DS, et al. Haemodynamic def­i­ni­tions
and updated clin­i­cal clas­si­fi­ca­tion of pul­mo­nary hyper­ten­sion. Eur Respir J.
2019;53(1):1801913.
6. Pepke-Zaba J, Delcroix M, Lang I, et al. Chronic throm­bo­em­bolic pul­mo­
nary hyper­ten­sion (CTEPH): results from an inter­na­tional pro­spec­tive reg­
is­try. Circulation. 2011;124(18):1973-1981.
7. Condliffe R, Kiely DG, Gibbs JS, et  al. Prognostic and aetiological fac­
tors in chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion. Eur Respir J.
2009;33(2):332-338.
8. Ende-Verhaar YM, Cannegieter SC, Vonk Noordegraaf A, et al. Incidence of
chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion after acute pul­mo­nary
embolism: a con­tem­po­rary view of the published lit­er­a­ture. Eur Respir J.
2017;49(2):1601792.
9. Gall H, Hoeper MM, Richter MJ, Cacheris W, Hinzmann B, Mayer E. An epi­
de­mi­o­log­i­cal anal­y­sis of the bur­den of chronic throm­bo­em­bolic pul­mo­
nary hyper­ten­sion in the USA, Europe and Japan. Eur Respir Rev. 2017;​
26(143):160121.
10. Gall H, Preston IR, Hinzmann B, et al. An inter­na­tional phy­si­cian sur­vey of
chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion man­age­ment. Pulm
Circ. 2016;​6(4):​472-482.
Prakash Singh Shekhawat
11. Tapson VF, Platt DM, Xia F, et al. Monitoring for pul­mo­nary hyper­ten­sion fol­
low­ing pul­mo­nary embolism: the INFORM study. Am J Med. 2016;129(9):978–
985.e2.
12. Riedel M, Stanek V, Widimsky J, Prerovsky I. Longterm fol­low-up of patients
with pul­mo­nary throm­bo­em­bo­lism: late prog­no­sis and evo­lu­tion of hemo­
dy­namic and respi­ra­tory data. Chest. 1982;81(2):151-158.
13. Pengo V, Lensing AW, Prins MH, et al; Thromboembolic Pulmonary Hyper-
tension Study Group. Incidence of chronic throm­bo­em­bolic pul­mo­nary
hyper­ten­sion after pul­mo­nary embolism. N Engl J Med. 2004;350(22):2257–
2264.
14. Konstantinides SV, Meyer G, Becattini C, et al; The Task Force for the diag­
no­sis and man­age­ment of acute pul­mo­nary embolism of the Euro­pean
Society of Cardiology (ESC). 2019 ESC Guidelines for the diag­no­sis and
man­age­ment of acute pul­mo­nary embolism devel­oped in col­lab­o­ra­tion
with the Euro­pean Respiratory Society (ERS): the Task Force for the diag­
no­sis and man­age­ment of acute pul­mo­nary embolism of the Euro­pean
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/478/1852108/478martin.pdf by guest on 13 December 2021
Society of Cardiology (ESC). Eur Respir J. 2019;54(3):1901647.
15. Klok FA, van Kralingen KW, van Dijk APJ, Heyning FH, Vliegen HW, Huisman
MV. Prospective car­dio­pul­mo­nary screen­ing pro­gram to detect chronic
throm­bo­em­bolic pul­mo­nary hyper­ten­sion in patients after acute pul­mo­
nary embolism. Haematologica. 2010;95(6):970-975.
16. Wilkens H, Konstantinides S, Lang IM, et  al. Chronic throm­ bo­em­ bolic
pul­mo­nary hyper­ten­sion (CTEPH): updated rec­om­men­da­tions from the
Cologne Consensus Conference 2018. Int J Cardiol. 2018;272S:69-78.
17. Lang IM, Madani M. Update on chronic throm­bo­em­bolic pul­mo­nary hyper­
ten­sion. Circulation. 2014;130(6):508-518.
18. Klok FA, Dzikowska-Diduch O, Kostrubiec M, et al. Derivation of a clin­i­cal
pre­dic­tion score for chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion
after acute pul­mo­nary embolism. J Thromb Haemost. 2016;14(1):121-128.
19. ATS Committee on Proficiency Standards for Clinical Pulmonary Function
Laboratories. ATS state­ment: guide­lines for the six-min­ute walk test. Am J
Respirat Crit Care Med. 2002;166(1):111-117.
20. Dzikowska-Diduch O, Kostrubiec M, Kurnicka K, et al. The post-pul­mo­nary
syn­drome—results of echo­car­dio­graphic driven fol­low up after acute pul­
mo­nary embolism. Thromb Res. 2020;186:30-35.
21. Tunariu N, Gibbs SJR, Win Z, et  al. Ventilation-per­fu­sion scin­tig­ra­phy is
more sen­si­tive than multidetector CTPA in detecting chronic throm­bo­em­
bolic pul­mo­nary dis­ease as a treat­able cause of pul­mo­nary hyper­ten­sion. J
Nucl Med. 2007;48(5):680-684.
22. McLaughlin VV, Langer A, Tan M, et al; Pulmonary Arterial Hypertension-​
Quality Enhancement Research Initiative (PAH-QuERI) Investigators. Con-
temporary trends in the diag­no­sis and man­age­ment of pul­mo­nary arte­rial
hyper­ten­sion: an ini­tia­tive to close the care gap. Chest. 2013;143(2):324–
332.
23. Kovacs G, Dumitrescu D, Barner A, et  al. Definition, clin­i­cal clas­si­fi­ca­tion
and ini­tial diag­no­sis of pul­mo­nary hyper­ten­sion: updated rec­om­men­
da­tions from the Cologne Consensus Conference 2018. Int J Cardiol.
2018;272S:11-19.
24. Bunclark K, Newnham M, Chiu Y-D, et  al. A mul­ti­cen­ter study of antico-
agulation in oper­a­ble chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion. J
Thromb Haemost. 2020;18(1):114-122.
25. Bunclark K, Ansaripour A, Sheares K, Toshner M. Response to let­ter to the
edi­tor: direct oral anti­ co­
ag­u­
lants in throm­botic antiphospholipid syn­
drome asso­ci­ated with chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion.
J Thromb Haemost. 2020;18(3):756-757.
26. Pengo V, Denas G, Zoppellaro G, et al. Rivaroxaban vs war­fa­rin in high-risk
patients with antiphospholipid syn­drome. Blood. 2018;132(13):1365-1371.
27. Ordi-Ros J, Sáez-Comet L, Pérez-Conesa M, et al. Rivaroxaban ver­sus vita­
min K antag­o­nist in antiphospholipid syn­drome: a ran­dom­ized noninferior-
ity trial. Ann Intern Med. 2019;171(10):685-694.
28. Madani MM, Auger WR, Pretorius V, et  al. Pulmonary end­ ar­
ter­
ec­tomy:
recent changes in a sin­gle insti­tu­tion’s expe­ri­ence of more than 2,700
patients. Ann Thorac Surg. 2012;94(1):97-103.
29. Jenkins D, Madani M, Fadel E, D’Armini AM, Mayer E. Pulmonary end­ar­ter­ec­
tomy in the man­age­ment of chronic throm­bo­em­bolic pul­mo­nary hyper­ten­
sion. Eur Respir Rev. 2017;26(143):160111.
30. Bikdeli B, Chatterjee S, Desai NR, et al. Inferior vena cava fil­ters to pre­vent
pul­mo­nary embolism: sys­tem­atic review and meta-anal­y­sis. J Am Coll Car-
diol. 2017;70(13):1587-1597.
31. Galiè N, Humbert M, Vachiery J-L, et al; ESC Scientific Document Group.
2015 ESC/ERS Guidelines for the diag­no­sis and treat­ment of pul­mo­nary
hyper­ten­sion: the Joint Task Force for the Diagnosis and Treatment of Pul-
monary Hypertension of the Euro­pean Society of Cardiology (ESC) and
CTEPH: anticoagulation and beyond  |  483
 the Euro­pean Respiratory Society (ERS): endorsed by: Association for Euro­
pean Paediatric and Congenital Cardiology (AEPC), International Society
for Heart and Lung Transplantation (ISHLT). Eur Heart J. 2016;37(1):67-119.
32. Fukui S, Ogo T, Morita Y, et al. Right ven­tric­u­lar reverse remodelling after
bal­loon pul­mo­nary angio­plasty. Eur Respir J. 2014;43(5):1394-1402.
33. Ogo T, Fukuda T, Tsuji A, et  al. Efficacy and safety of bal­loon pul­mo­nary
angio­plasty for chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion guided
by cone-beam com­puted tomog­ra­phy and elec­tro­car­dio­gram-gated area
detec­tor com­puted tomog­ra­phy. Eur J Radiol. 2017;89:270-276.
34. Olsson KM, Wiedenroth CB, Kamp J-C, et al. Balloon pul­mo­nary angio­plasty
for inop­er­a­ble patients with chronic throm­bo­em­bolic pul­mo­nary hyper­
ten­sion: the ini­tial Ger­man expe­ri­ence. Eur Respir J. 2017;49(6):1602409.
35. Ghofrani H-A, D’Armini A-M, Grimminger F, et al; CHEST-1 Study Group. Rio-
ciguat for the treat­ment of chronic throm­bo­em­bolic pul­mo­nary hyper­ten­
sion. N Engl J Med. 2013;369(4):319-329.
36. Sadushi-Kolici R, Jansa P, Kopec G, et al. Subcutaneous treprostinil for the
treat­ment of severe non-oper­a­ble chronic throm­bo­em­bolic pul­mo­nary
hyper­ten­sion (CTREPH): a dou­ble-blind, phase 3, randomised con­trolled
trial. Lancet Respir Med. 2019;7(3):239-248.
37. Ghofrani H-A, Simonneau G, D’Armini A-M, et al; MERIT study inves­ti­ga­
tors. Macitentan for the treat­ment of inop­er­a­ble chronic throm­bo­em­bolic
pul­mo­nary hyper­ten­sion (MERIT-1): results from the multicentre, phase 2,
randomised, dou­ble-blind, pla­cebo-con­trolled study. Lancet Respir Med.
2017;5(10):785-794.
38. Jaïs X, D’Armini AM, Jansa P, et al; Bosentan Effects in iNopErable Forms of
chronIc Thromboembolic pul­mo­nary hyper­ten­sion Study Group. Bosentan
484  |  Hematology 2021  |  ASH Education Program
for treat­ment of inop­er­a­ble chronic throm­bo­em­bolic pul­mo­nary hyper­ten­
sion: BENEFiT (Bosentan Effects in iNopErable Forms of chronIc Thrombo-
embolic pul­mo­nary hyper­ten­sion), a ran­dom­ized, pla­cebo-con­trolled trial.
J Am Coll Cardiol. 2008;52(25):2127-2134.
39. Reesink HJ, Surie S, Kloek JJ, et al. Bosentan as a bridge to pul­mo­nary end­
ar­ter­ec­tomy for chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion. J Tho-
rac Cardiovasc Surg. Jan 2010;139(1):85-91.
40. Jensen KW, Kerr KM, Fedullo PF, et  al. Pulmonary hyper­ten­sive med­ic ­ al
ther­apy in chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion before pul­
mo­nary thromboendarterectomy. Circulation. 2009;120(13):1248-1254.
41. Cannon JE, Su L, Kiely DG, et al. Dynamic risk strat­i­fi­ca­tion of patient long-
term out­come after pul­mo­nary end­ar­ter­ec­tomy: results from the United
Kingdom National Cohort. Circulation. 2016;133(18):1761-1771.
42. Ogino H, Ando M, Matsuda H, et al. Jap­a­nese sin­gle-cen­ter expe­ri­ence of
sur­gery for chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion. Ann Thorac
Surg. 2006;82(2):630-636.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/478/1852108/478martin.pdf by guest on 13 December 2021
43. Condliffe R, Kiely DG, Gibbs JS, et al. Improved out­comes in med­i­cally and
sur­gi­cally treated chronic throm­bo­em­bolic pul­mo­nary hyper­ten­sion. Am J
Respir Crit Care Med. 2008;177(10):1122-1127.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000282
 MULTIDISCIPLINARY HEMATOLOGIC DISORDERS: WHAT IS THE HEMATOLOGIST ’ S ROLE ?

Rebalanced hemostasis in liver disease:

a misunderstood coagulopathy

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/485/1851749/485roberts.pdf by guest on 13 December 2021

Lara N. Roberts
King’s College Hospital NHS Foundation Trust, Denmark Hill, London, United Kingdom

The combination of frequently abnormal hemostatic markers and catastrophic bleeding as seen with variceal hemorrhage
has contributed to the longstanding misperception that chronic liver disease (CLD) constitutes a bleeding diathesis. Lab-
oratory studies of hemostasis in liver disease consistently challenge this with global coagulation assays incorporating
activation of the protein C pathway demonstrating rebalanced hemostasis. It is now recognized that bleeding in CLD is
predominantly secondary to portal hypertension (rather than a coagulopathy) and additionally that these patients are at
increased risk of venous thrombosis, particularly in the portal venous system. This narrative review describes the current
understanding of hemostasis in liver disease, as well as the periprocedural management of hemostasis and anticoagula-
tion for management of venous thromboembolism in patients with CLD.

LEARNING OBJECTIVES
• Describe the changes leading to rebalanced hemostasis in CLD
• Recognize factors influencing periprocedural bleeding risk in patients with CLD
• Evaluate the need for anticoagulation of PVT in patients with liver disease and select the most suitable agent

liver failure (ACLF).1 Most patients admitted to the hospital

CLINICAL CASE
A 46-year-old woman with a new diagnosis of presumed
alcohol-related cirrhosis presents to her local hospital
with increasing abdominal swelling, pain, and jaundice.
She had been admitted 1 year prior with alcohol with-
drawal but did not attend planned hepatology outpa-
tient reviews. She has continued to drink 0.5 L of vodka
daily. Initial laboratory investigation reveals the follow-
ing: hemoglobin, 90 g/L; platelets, 43 × 109/L; bilirubin,
199 µmol/L; albumin, 26 g/L; international normalized
ratio (INR), 2.2; prothrombin time (PT), 33 seconds; and
creatinine, 126 µmol/L. The Child-Turcotte-Pugh class is
C and Model for End Stage Liver Disease score is 25. She
commences spironolactone but there is no improvement
in ascites or weight, and her renal function further deteri-
orates. Therapeutic paracentesis is planned and the med-
ical team requests authorization of fresh-frozen plasma
(FFP) (aiming for an INR of <1.5) to proceed.
The natural history of chronic liver disease (CLD) is charac-
terized by phases of stable cirrhosis progressing to acute
decompensation (AD), from which patients may recover or
progress to multiorgan failure, known as acute-on-chronic
Prakash Singh Shekhawat
are in a state of AD, often following a precipitating event.
Common complications pertain to liver failure or portal
hypertension and comprise jaundice, encephalopathy,
ascites, and variceal bleeding. Although bleeding is com-
mon in liver disease, it is now recognized that most sponta-
neous bleeding events are predominantly gastrointestinal
(GI) and secondary to portal hypertension (refer to Table 1
for incidence).2-4
The liver has a key role in the synthesis of both pro- and
anticoagulant proteins, along with pro- and antifibrinolyt-
ics and thrombopoietin.9 Portal hypertension in CLD leads
to splenomegaly and thrombocytopenia. Progression of
CLD with synthetic failure is accompanied by prolonga-
tion of PT.10 As PT simply measures time to first detection
of clot in plasma following activation with tissue factor and
calcium, it does not reflect in vivo hemostasis in patients
with CLD in which natural anticoagulants (protein C, pro-
tein S, and antithrombin) are reduced in parallel, and it does
not detect increased factor VIII and von Willebrand factor.
Global coagulation assays, such as thrombin generation
incorporating activation of the protein C pathway, demon-
strate that patients with CLD have rebalanced hemostasis
with some evidence of hypercoagulability. With increasing
disease severity (as in ACLF), wide interindividual variation
Rebalanced hemostasis in liver disease | 485
Table 1. Incidence of bleed­ing and por­tal vein throm­bo­sis in CLD reported in pro­spec­tive stud­ies

Child-Pugh
Study set­ting Study design Participants, n class A, n (%) Incidence
Bleeding
  Italy, 2012-20143 Prospective, mul­ti­cen­ter 280 53 Overall: 5.5%/year
Major: 3.6%/year
Minor: 1.9%/year
Portal hyper­ten­sive: ~3.3%/year
Nonvariceal GI: 1.9%/year
Intracranial: 0.17%/year
  United States, Decem­ber Prospective, sin­gle-cen­ter 83 0 Overall: 40%
2009 to Jan­u­ary 20104 hos­pi­tal­ized cohort Variceal: 17%
Nonvariceal GI: 12%

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/485/1851749/485roberts.pdf by guest on 13 December 2021

Portal vein throm­bo­sis
  Italy, 2012-20145 Prospective, mul­ti­cen­ter 753 397 (53) 6.1/100 patient years
  No PVT at enroll­ment 692 4.1
  Prior PVT at enroll­ment 61 18.9
  France/Belgium, 2000-2006 6
Prospective, mul­ti­cen­ter 1243 863 (69) 1 y: 4.6%
Cumulative 5 y: 10.7%
  France, 2014-20177 Prospective, sin­gle cen­ter 108 75 (77) 1 y: 10.8%
  Portugal, 2014-20198 Prospective, sin­gle cen­ter 241 184 (76) 1 y: 3.7%
Cumulative 3 y: 7.6%
Modified from Roberts and Bernal (2020)2 with per­mis­sion from Thieme.

in throm­bin gen­er­a­tion poten­tial is seen with a more pre­car­i­ous
hemostatic bal­ance.10-12 Similarly, throm­bo­cy­to­pe­nia and plate­let
dys­func­tion are coun­tered by increased von Willebrand fac­tor,
with increased high molec­u­lar weight multimers due to reduced
ADAMTS13.13,14 The changes in hemo­sta­sis asso­ci­ated with CLD
are sum­ma­rized in Table 2.
Periprocedural bleed­ing risk
It has been repeat­edly dem­on­strated over the past 40 years that
nei­ther prolonged PT/INR nor throm­bo­cy­to­pe­nia inde­pen­dently
pre­dict periprocedural bleed­ing.9,15-17 These param­e­ters con­tinue
to be mea­sured with attempts at cor­rec­tion com­mon in this set­
ting, despite poten­tial for harm asso­ci­ated with trans­fu­sion.18,19
Procedural bleed­ ing risk in patients with CLD is not well
defined; var­i­able prac­tice in attempting to cor­rect hemo­static
mark­ers and a lack of con­sis­tent def­i­ni­tion for major bleed­ing
are key con­trib­u­tory fac­tors.20 It is pro­posed that pro­ce­dures
with an inci­dence of major bleed­ing of more than 2% or bleed­
ing with poten­tial to cause organ dam­age/death be con­sid­ered
high risk.20 Ascites is a fre­quent com­pli­ca­tion of CLD, devel­op­ing
in approaching two-thirds of patients within 10 years of cir­rho­sis
diag­no­sis.21 Therapeutic paracentesis is recommended as first-
line treat­ment for large-vol­ume asci­tes, and this sce­nario is com­
monly encoun­tered. The risk of bleed­ing fol­low­ing paracentesis
is low, esti­mated at 0.2%.22 Acute kid­ney injury is a rec­og­nized
inde­pen­dent risk fac­ tor for postparacentesis hematoperito-
neum.23
Prophylactic hemo­static inter­ven­tions
Despite evi­
dence that major sur­ gery such as liver trans­plan­
ta­
tion can be safely performed with­ out correcting abnor­ mal
hemo­static mark­ers,24 attempts at cor­rec­tion remain com­mon
in the pro­ce­dural set­ting.18,19 In Spain, a national sur­vey (of pre­
dom­i­nantly hepatologists, 69%) from 2017 reported the major­
ity of respon­dents would attempt to cor­rect PT prior to major
pro­ce­dures, with 17% also attempting to cor­rect prior to low-
risk pro­ce­dures.19 Almost all­respon­dents reported attempting
to cor­rect throm­bo­cy­to­pe­nia prior to major sur­gery, with 35%
also attempting to cor­rect prior to low-risk pro­ce­dures, pre­dom­
i­nantly using a thresh­old for plate­lets of less than 50 × 109/L. How-
ever, attempts to cor­rect throm­bo­cy­to­pe­nia and coagulopathy
with trans­fu­sion are asso­ci­ated with addi­tional risks. Administra-
tion of a “ther­a­peu­tic” vol­ume of FFP pre­dis­poses to both trans­
fu­sion-asso­ci­ated cir­cu­la­tory over­load and trans­fu­sion-related
acute lung injury.25 Furthermore, due to sub­se­quent increases in
por­tal venous pres­sure,26 FFP admin­is­tra­tion may par­a­dox­i­cally
increase the risk of bleed­ing (eg, dur­ing endo­scopic variceal
band liga­tion or transjugular portosystemic shunt place­ment).
In addi­tion, lab­o­ra­tory stud­ies dem­on­strate that patients with
CLD have preexisting nor­mal to hyper­co­ag­u­la­ble pro­files, with
FFP trans­fu­sion hav­ing a min­i­mal impact on improv­ing PT but
increas­ing hyper­co­ag­u­la­bil­ity.27,28 Platelet trans­fu­sion results in
var­i­able plate­let count incre­ment,28,29 and ran­dom­ized con­trolled
tri­als of plate­let trans­fu­sion in other pop­u­la­tions (eg, intra­ce­re­
bral bleed­ing) are asso­ci­ated with worse out­comes.30 Throm-
bopoietin recep­tor ago­nists have been dem­on­strated to bet­ter
cor­rect plate­let count prior to elec­tive pro­ce­dures.31,32 However,
the sem­i­nal stud­ies included both low- and high-risk pro­ce­dures
and lacked clin­i­cally impor­tant pri­mary out­comes. Meta-anal­y­sis
sug­gests their use may reduce periprocedural bleed­ing.33
Given the risks asso­ci­a­tion with plasma trans­fu­sion and lack
of evi­dence to dem­on­strate ben­e­fit, cor­rec­tion of prolonged PT
prior to paracentesis has been advised against since 2003 by
inter­na­tional soci­e­ties. Recommendations for high-risk bleed­ing

486  |  Hematology 2021  |  ASH Education Program

Table 2. Summary of hemostatic changes in CLD and lab­o­ra­tory evi­dence supporting rebalanced hemo­sta­sis

Laboratory evi­dence to sup­port

Characteristic Factors asso­ci­ated with bleed­ing Factors asso­ci­ated with throm­bo­sis rebalanced hemo­sta­sis
Platelets • Thrombocytopenia •  ↑ vWF • Weak
•  Platelet dys­func­tion • ↓ADAMTS13 • Difficult to rep­li­cate plate­let-
• Anemia •  ↑Platelet acti­va­tion endo­the­lial inter­ac­tion in lab­o­ra­tory
•  Endothelial acti­va­tion set­ting
Coagulation •  ↓FII, V, VII, IX, X, XI • ↓Protein C, pro­tein S, anti­throm­bin • Thrombin gen­er­a­tion nor­mal to
•  ↓Fibrinogen (in AD, ACLF) • ↑FVIII increased when mea­sured with pro­
• ↓FXIII •  ↓Fibrin clot per­me­abil­ity tein C path­way acti­va­tion
•  ↓Rates of fibrin poly­mer­i­za­tion • Weak evi­dence for rebalancing of
fibrin clot strength/sta­bil­ity
Fibrinolysis •  ↑tPA (not rebalanced by ↑PAI-1) • ↓Plasminogen • Uncertain; both

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/485/1851749/485roberts.pdf by guest on 13 December 2021

• ↓α2-Antiplasmin hyper/hypofibrinolysis seen in
• ↓TAFI ACLF, poten­tially reflecting the del­i­
cate nature of rebalance
Adapted from Northup et al20 with per­mis­sion from Wiley.
ADAMTS13, a disintegrin and metalloproteinase thrombospondin type 1 motif; F, fac­tor; PAI-1, plas­min­o­gen acti­va­tor inhib­i­tor 1; TAFI, throm­bin
activatable fibri­no­ly­sis inhib­i­tor; tPA, tis­sue plas­min­o­gen acti­va­tor; vWF, von Willebrand fac­tor.

pro­ce­dures are sum­ma­rized in Table 3.20,34-36 Vitamin K defi­ciency
may be rel­e­vant in those with poor diet and/or mal­ab­sorp­tion,
and a sin­gle dose of 10 mg more than 12 hours prior to inter­ven­
tions in such patients is advo­cated by some.35,36
CLINICAL CASE (Con­t in­u ed)
This patient received a sin­gle dose of vita­min K on admis­sion
with no improve­ment in the PT. After reassuring the med­i­cal
team that the pro­ce­dural bleed­ing risk of paracentesis is low
and that PT/INR and plate­let count are not a mea­sure of bleed­
ing risk, ther­ a­
peu­tic paracentesis is performed under ultra­
sound guid­ance with­out com­pli­ca­tion.
She recov­ers from this acute epi­sode of decom­pen­sa­tion and
remains absti­nent from alco­hol fol­low­ing hos­pi­tal dis­charge. On
rou­tine screen­ing ultra­sound for hepa­to­cel­lu­lar car­ci­noma 2
years later, she is diag­nosed with occlu­sive por­tal vein throm­bo­
sis (PVT) extending to the supe­rior mes­en­teric venous junc­tion.
The spleen is enlarged (20 cm), and the patient has the fol­low­
ing lab­o­ra­tory val­ues: hemo­glo­bin, 110  g/L; plate­lets, 57  ×  109/L;
bil­i­ru­bin, 16 µmol/L; albu­min, 40 g/L; PT, 18 sec­onds; INR, 1.3;
sodium, 138 µmol/L; and cre­at­i­nine, 116 µmol/L. She is Child-
Turcotte-Pugh class A, and Model for End Stage Liver Disease
score is 13. An upper GI endos­copy performed dur­ing the pre­
vi­ous admis­sion revealed mild por­tal hyper­ten­sive gastropathy
with no esoph­a­geal varices. The med­i­cal team seek advice on
the role for anticoagulation in the con­text of throm­bo­cy­to­pe­nia.
Venous throm­bo­em­bo­lism in CLD
PVT is the most com­mon throm­botic event affect­ing patients
with CLD, with its prev­a­lence reported in up to 26% in those
listed for liver trans­plan­ta­tion and increas­ing in par­al­lel with dis­
ease sever­ity.38 CLD is also asso­ci­ated with an increased risk of
deep vein throm­bo­sis (DVT) and pul­mo­nary embolism. A meta-
anal­y­sis of pre­dom­i­nantly ret­ro­spec­tive cohort stud­ies reported
Prakash Singh Shekhawat
cir­rho­sis was asso­ci­ated with an increased odds ratio for venous
throm­bo­em­bo­lism (VTE) of 1.7 (95% CI, 1.3-2.2).39 The inci­dence
of PVT in recent pro­spec­tive stud­ies is sum­ma­rized in Table 1.
The Ital­ian obser­va­tional cohort of 753 patients with cir­rho­sis
reported an inci­dence rate of 6 per 100 patient years.5 Of note,
half of the events were asymp­tom­atic and detected on rou­tine
screen­ing for hepa­to­cel­lu­lar car­ci­noma. Those with prior PVT at
study entry had a sig­nif­i­cantly higher rate of PVT com­pared with
those with­out (18.9 vs 4.0 per 100 patient years).5 The only inde­
pen­dent risk fac­tors for PVT iden­ti­fied in this study were p ­ re­vi­ous
PVT and degree of throm­bo­cy­to­pe­nia (with lower counts asso­
ci­ated with increased risk). A fur­ther pro­spec­tive obser­va­tional
cohort of 241 patients with cir­rho­sis reported a cumu­la­tive inci­
dence of 3.7% and 7.6% at 1 and 3 years, respec­tively.8 Throm-
bocytopenia and pre­vi­ous decom­pen­sa­tion were iden­ti­fied as
inde­pen­dent pre­dic­tors of PVT. Of note, in both cohorts, most
patients were Child-Pugh class A (see Table 1).
This sce­nario illus­trates again that con­ven­tional lab­o­ra­tory
tests do not reflect the under­ly­ing hemo­static milieu in patients
with liver dis­ease. The par­a­dox­i­cal increase in PVT risk asso­ci­
ated with throm­bo­cy­to­pe­nia likely reflects increas­ing sever­ity of
both liver dis­ease and por­tal hyper­ten­sion with reduced por­tal
venous flow.8,40 Only a sin­gle, small ret­ro­spec­tive study (n = 53)
has exam­ined the influ­ence of hyper­co­ag­u­la­bil­ity mea­sured
with throm­ bin gen­ er­a­
tion on PVT risk.41 Patients with base­
line “thrombomodulin resis­tance” (defined as an endog­e­nous
throm­bin poten­tial ratio >95th per­cen­tile of nor­mal con­trols) had
an 8-fold increase in risk of PVT over 4 years. Local hyper­co­ag­
u­la­bil­ity medi­ated by bac­te­rial trans­lo­ca­tion, inflam­ma­tion, and
endotoxemia are pro­posed as addi­tional con­trib­u­tors but have
not been con­firmed in pro­spec­tive stud­ies.42
Anticoagulation in CLD
The need for anticoagulation in this sce­nario of inci­den­tal PVT in
the absence of intes­ti­nal ische­mia or planned liver trans­plan­ta­
tion (in which anas­to­mo­sis may be com­pro­mised) is uncer­tain.
A recent meta-anal­y­sis of 33 stud­ies com­pris­ing 1696 patients
with PVT reported spon­ta­ne­ous com­plete recan­a­li­za­tion in 18%
Rebalanced hemo­sta­sis in liver dis­ease  |  487
Table 3. Thresholds for coag­ul­a­tion param­e­ters prior to high-risk pro­ce­dures in patients with CLD
Characteristic AASLD 202120 ACG 202037
PT/INR Do not cor­rect Do not cor­rect
Platelet count Do not cor­rect >50 × 10 /L
9
Fibrinogen Do not cor­rect No spe­cific rec­om­men­da­tion
VHA Do not use rou­tinely May be use­ful
*Give vita­min K if INR >2.5; do not use FFP/prothrombin complex concentrate.
Table adapted from Northup et al20 with per­mis­sion from Wiley.
AASLD, Amer­i­can Association for the Study of Liver Disease; ACG, Amer­i­can College of Gastroenterology; AGA, Amer­i­can Gastroenterology
Association; SIR, Society of Interventional Radiology; VHA, vis­co­elas­tic hemo­static assay.
of patients (n  = 
33/180) not receiv­ ing anticoagulation. How-
ever, anticoagulation was asso­ci­ated with a sig­nif­i­cant 2-fold
improve­ment in over­all and com­plete recan­a­li­za­tion rates com­
pared with no anticoagulation. The risk of throm­bus pro­gres­sion
was also sig­nif­i­cantly reduced (risk ratio [RR], 0.26; 95% CI, 0.14-
0.49). Six stud­ies reported over­all sur­vival, with anticoagulation
asso­ci­ated with improve­ment (RR, 1.11; 95% CI, 1.03-1.21).43 There
are clearly a num­ber of lim­it­a­tions to this anal­y­sis. In our case,
given the throm­bus is occlu­sive and extends to the mes­en­teric
junc­tion, anticoagulation is likely to be ben­e­fi­cial in pre­vent­ing
fur­ther exten­sion to the mes­en­teric vas­cu­la­ture and achiev­ing
recan­a­li­za­tion, with poten­tial reduc­tion in por­tal hyper­ten­sion
(given the lack of portosystemic col­lat­eral cir­cu­la­tion). Guide-
lines rec­om­mend obtaining cross-sec­tional imag­ing (com­puted
tomog­ra­phy/mag­netic res­o­nance imag­ing) to con­firm the diag­
no­sis and extent of throm­bo­sis, as well as to exclude malig­nant
obstruc­tion prior to treat­ment ini­ti­a­tion.20 There is a clear role
for anticoagulation in patients with symp­tom­atic PVT to pre­vent
pro­gres­sion and intes­ti­nal ische­mia.
The prin­ci­pal con­cern in anticoagulating patients with liver
dis­ease is the con­com­i­tant increased risk of bleed­ing. In the
afore­men­tioned meta-anal­y­sis, the pooled over­all rate of bleed­
ing asso­ci­ated with anticoagulation from 19 stud­ies was 2.8%,
with fatal bleed­ing 0.7% (from 25 stud­ies).43 The pooled rate of
esoph­a­geal variceal bleed­ing from 17 stud­ies was 2%. The use
of anticoagulation did not increase the risk of bleed­ ing (RR,
0.78; 95% CI, 0.47-1.3) from 4 stud­ ies reporting out­ comes in
those both on and off anticoagulation, and inter­est­ingly, anti-
coagulation was asso­ci­ated with a reduced risk of both upper
GI and variceal bleed­ing (RR, 0.26; 95% CI, 0.11-0.65).43 These
data sug­gest anticoagulation does not increase the risk of vari-
ceal hem­or­rhage (and may in fact reduce the risk, poten­tially by
reduc­ing por­tal hyper­ten­sion). It is recommended that upper GI
endos­copy be performed to iden­tify and enable treat­ment of
high-risk varices.20 In prac­tice, anticoagulation can be ini­ti­ated
while awaiting fur­ther endo­scopic eval­u­a­tion. Endoscopic vari-
ceal banding is a low bleed­ing risk pro­ce­dure,20 with pro­ce­dural
bleed­ing uncom­mon and the major­ity of bleed­ing sec­ond­ary to
band ulcer­at­ion, presenting some days later.15,44 A sin­gle obser­
va­tional study found no increased bleed­ing risk asso­ci­ated with
use of low molec­u­lar weight hep­a­rin (LMWH) (169 pro­ce­dures
in 80 patients) at the time of variceal band liga­tion.45 However,
proceduralists may tem­po­rar­ily inter­rupt anticoagulation fol­low­
ing variceal banding when the per­ceived bleed­ing risk due to
sec­ond­ary band ulcer­a­tion is thought to out­weigh the risk of
throm­bo­sis exten­sion.
488  |  Hematology 2021  |  ASH Education Program
AGA 201935 SIR 201936
Do not cor­rect INR >2.5*
>50 × 10 /L
9
>30 × 109/L
>1.2 g/L >1 g/L
No spe­cific rec­om­men­da­tion No spe­cific rec­om­men­da­tion
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/485/1851749/485roberts.pdf by guest on 13 December 2021
Thrombocytopenia is a rec­og­nized risk fac­tor for bleed­ing
with anticoagulation; a small obser­va­tional cohort reported a
plate­let count of less than 50 × 109/L as a pre­dic­tor for bleed­
ing.46 In these patients, I con­sider ini­ti­at­ing LMWH at pro­phy­
lac­tic to inter­me­di­ate doses and titrate the dose as tol­er­ated. I
favor inter­me­di­ate doses for symp­tom­atic and/or exten­sive PVT.
For lim­ited extent/asymp­tom­atic events in which a deci­sion for
anticoagulation is made, I favor pro­phy­lac­tic doses at ini­ti­a­tion.
Severe throm­bo­cy­to­pe­nia (<30 × 109/L) is uncom­ mon, and in
such cases, I would look for and treat con­trib­ut­ing fac­tors but
still con­sider pro­phy­lac­tic LMWH on a case-by-case basis. If the
plate­let count improves, ther­ap ­ eu­tic doses can then be used. As
this patient has a plate­let count more than 50 × 109/L, full-dose
anticoagulation should be used.
For venous throm­bo­em­bo­lism in patients with­out CLD, direct
oral anti­co­ag­u­lants (DOACs) are now the agents of choice due
to their pre­dict­able phar­ma­co­ki­net­ics, elim­i­nat­ing the need for
rou­tine drug mon­i­tor­ing and favor­able safety pro­file.47 Charac-
teristics of avail­­able anti­co­ag­u­lants are sum­ma­rized in Table 4.
Of note, patients with CLD were excluded from the ran­dom­ized
con­trolled tri­als establishing their effi­cacy in VTE, and there is
lit­tle evi­dence for use of DOACs in treat­ment of PVT in cir­rho­sis.
The use of DOACs is not recommended in patients with increased
sever­ity of CLD (see Table 4). A recent sys­tem­atic review iden­ti­
fied 5 ret­ro­spec­tive cohort stud­ies com­pris­ing 239 patients with
CLD treated for DVT, splanch­nic vein throm­bo­sis, or atrial fibril­la­
tion.48 The anal­y­sis sug­gests com­pa­ra­ble effi­cacy and safety but
is lim­ited by the small sam­ple size and ret­ro­spec­tive nature of
the stud­ies. A small ran­dom­ized con­trolled trial (n = 80) in Egypt
com­pared rivaroxaban 10 mg daily with war­fa­rin for treat­ment
of acute PVT in patients with com­pen­sated hep­a­ti­tis C cir­rho­
sis (pre­dom­i­nantly after sple­nec­tomy) for a var­i­able treat­ment
dura­tion based on throm­bus extent and recan­a­li­za­tion.49 Patients
were mon­i­tored fortnightly with ultra­sound imag­ing with a pri­
mary out­come of com­plete recan­a­li­za­tion. The pri­mary out­come
was achieved in sig­nif­i­cantly more patients receiv­ing rivarox-
aban (85% vs 45%), with recan­a­li­za­tion occur­ring at a mean of 2.5
months. There was no major bleed­ing reported in the rivaroxaban
arm and no recur­rent events at 12 months. A fur­ther small study
com­par­ing out­comes in those treated with edoxaban (n = 20) to
war­fa­rin (n = 30) fol­low­ing 2 weeks of danaparoid also reported
supe­rior clot regres­sion with edoxaban.50 Although these find­
ings are prom­is­ing, it is impor­tant to note that nei­ther war­fa­rin
arm pro­vided stan­dard of care in that there was no bridg­ing until
ther­a­peu­tic INR was reached, with a sub­ther­a­peu­tic INR tar­get
(1.5-2.0) in 1 study. This is impor­tant as ­recan­a­li­za­tion rates are
 Table 4. Available anti­co­ag­u­lants and con­sid­er­ations for use in man­age­ment of VTE in CLD

Characteristic VKA LMWH Dabigatran Apixaban Edoxaban Rivaroxaban

Dosing Variable od, based on Weight based, once or Twice daily Twice daily Once daily Twice daily for 3 weeks,
INR twice daily then once daily
Initiation Overlap with LMWH until As above ≥5 days LMWH, then 10 mg twice daily for 5 days LMWH, then switch 15 mg twice daily for 3
INR >2. Use alter­nate switch 7 days weeks
agent if base­line INR
increased
Standard dose Variable as above As above 150 mg bd 5 mg twice daily from 60 mg od 20 mg once daily from
day 8 day 22
Dose reduc­tion NA CrCl 15-30 mL, reduce 110 mg bd for age ≥80 Consider dose reduc­tion 30 mg od for CrCl <50 mL/min Consider dose reduc­tion
dose years or con­com­i­tant to 2.5 mg bd from or weight <60 kg to 10 mg from
verap­a­mil and con­sider 6 months 6  months
for other high-risk
groups
Drug mon­i­tor­ing Y N N N N N
Hepatobiliary/intes­ti­nal Predominant Minimal 20% 73% 50% 34%
elim­i­na­tion

Prakash Singh Shekhawat

CLD
CPC A Y Y Y Y Caution, avoid if ALT/AST Y
<2 × ULN and bil­i­ru­bin
CPC B Y Y Limited expe­ri­ence, no Y in absence of N
<1.5 × ULN
change in expo­sure coagulopathy
in n = 12
CPC C Y Y N N N N
CKD
CrCl, 15-30 mL/min Y ↓Dose Do not use Limited expe­ri­ence Limited expe­ri­ence Limited expe­ri­ence
<15 mL/min Y Avoid Do not use Do not use Do not use Do not use
Modified from Northup et al20 with per­mis­sion from Wiley.
ALT, ala­nine ami­no­trans­fer­ase; AST, aspar­tate ami­no­trans­fer­ase; bd, twice daily; CKD, chronic kid­ney dis­ease; CPC, Child-Pugh clas­si­fi­ca­tion; CrCl, cre­at­i­nine clear­ance; N, no; NA, not applicable;
od, once daily; ULN, upper limit of nor­mal; Y, yes.

Rebalanced hemo­sta­sis in liver dis­ease  |  489

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/485/1851749/485roberts.pdf by guest on 13 December 2021
improved with early anticoagulation.46 Further ade­quately pow-
ered stud­ies with the use of LMWH bridg­ing until ther­a­peu­tic
INR is achieved in the com­par­a­tor arm are required. In prac­tice,
LMWH is often ini­ti­ated until com­ple­tion of upper GI endos­copy,
fol­low­ing which a vita­min K antag­o­nist (VKA) can be ini­ti­ated.
DOACs may be a rea­son­able alter­na­tive in cases with dif­fi­culty
com­ply­ing to var­i­able dos­ing of VKA or the mon­i­tor­ing require­
ments (eg, lim­ited venous access and/or unable to use a point-
of-care INR device). For patients with sig­nif­i­cantly prolonged PT
at base­line, I favor extended LMWH pro­vided con­tin­ued par­
en­teral admin­is­tra­tion is accept­able to the patient, given the
opti­mal INR tar­get is unknown in those with sig­nif­i­cantly raised
base­line INR.
For PVT, in which a deci­sion to ini­ti­ate anticoagulation is made,
the min­i­mum recommended dura­tion of anti­co­ag­u­lant ther­apy is
6 months.20,37 Guidelines sug­gest reimaging to eval­ua ­ te recan­a­
li­za­tion to guide the ongo­ing need for anticoagulation.20 There
are, how­ever, no high-qual­ity data to inform whether incom­plete
recan­al­i­za­tion is asso­ci­ated with a higher risk of recur­rence or
other adverse out­comes related to por­tal hyper­ten­sion. Small
obser­va­tional stud­ies sug­gest the risk of early recur­rent PVT is
high (up to 38%) fol­low­ing dis­con­tin­u­a­tion of anticoagulation
in patients with advanced CLD.42,46 In addi­tion, pre­vi­ous PVT is
a strong pre­dic­tor for recur­rence in pro­spec­tive obser­va­tional
cohorts.5 However, given the lack of ran­dom­ized con­trolled tri­
als to defin­i­tively inform the risk of both PVT recur­rence and
bleed­ing, the fea­si­bil­ity of a mul­ti­cen­ter pro­spec­tive study to
deter­ mine the opti­ mal dura­ tion of anticoagulation should be
eval­u­ated. Currently, those listed for liver trans­plan­ta­tion fre­
quently remain on extended anticoagulation to reduce the risk
of recur­rent PVT, par­tic­ul­arly as the pres­ence of occlu­sive PVT
at trans­plan­ta­tion affects both tech­ni­cal suc­cess and posttrans-
plant sur­vival.51 For patients with DVT/pul­mo­nary embolism, the
min­im ­ um dura­tion of anticoagulation is 3 months. All patients
should be reviewed prior to com­plet­ing 3 months of treat­ment
to con­sider the need for extended anticoagulation in the pres­
ence of per­sis­tent risk fac­tors (eg, inflam­ma­tory bowel dis­ease)
and/or for events not asso­ci­ated with a major pro­vok­ing fac­tor
(eg, major sur­gery/hos­pi­tal­i­za­tion).52
CLINICAL CASE (Con­t in­u ed)
Therapeutic anticoagulation with LMWH was ini­ti­ated while
fur­
ther endos­copy was planned (unchanged from pre­ vi­
ous,
with no esoph­a­geal varices seen with mild por­tal hyper­ten­sive
gastropathy). Following this, war­fa­rin was ini­ti­ated with a tar­
get INR of 2.5. LMWH was con­tin­ued until an INR of more than
2 was achieved. She remained on war­fa­rin for 12 months, and
reimaging con­firmed par­tial recan­a­li­za­tion of the por­tal vein.
Apart from minor trau­matic bruis­ing, there were no sig­nif­i­cant
bleed­ing com­pli­ca­tions dur­ing treat­ment.
Conclusion
In sum­ mary, despite fre­ quently deranged hemo­ static mark­
ers in patients with CLD, there is a dual pro­pen­sity to bleed­ing
and throm­bo­sis (with a pre­pon­der­ance of por­tal vein throm­bo­
sis). Variceal bleed­ing sec­ond­ary to por­tal hyper­ten­sion is the
490  |  Hematology 2021  |  ASH Education Program
most fre­quent bleed­ing man­i­fes­ta­tion. There is lit­tle evi­dence
to sup­port pro­phy­lac­tic cor­rec­tion of hemo­static mark­ers in the
periprocedural set­ting, and use of pre­emp­tive trans­fu­sion sup­
port should be avoided. Anticoagulation is likely to improve out­
comes fol­low­ing occlu­sive por­tal vein throm­bo­sis, even in the
pres­ence of hemo­static abnor­mal­i­ties. However, the opti­mal
dura­tion of anticoagulation and role of DOACs remains uncer­tain.
Conflict-of-inter­est dis­clo­sure
Lara N. Roberts: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Lara N. Roberts: nothing to disclose.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/485/1851749/485roberts.pdf by guest on 13 December 2021
Correspondence
Lara N. Roberts, King’s Thrombosis Centre, Department of Haema-
tological Medicine, King’s College Hospital NHS Foundation Trust,
Denmark Hill, London SE5 9RS, UK; e-mail: lara​­.roberts@nhs​­.net.
References
1. D’Amico G, Garcia-Tsao G, Pagliaro L. Natural his­tory and prog­nos­tic indi­
ca­tors of sur­vival in cir­rho­sis: a sys­tem­atic review of 118 stud­ies. J Hepatol.
2006;44(1):217-231.
2. Roberts LN, Bernal W. Incidence of bleed­ing and throm­bo­sis in patients
with liver dis­ease. Semin Thromb Hemost. 2020;46(6):656-664.
3. Basili S, Raparelli V, Napoleone L, et al; PRO-LIVER Collaborators. Platelet
count does not pre­dict bleed­ing in cir­rhotic patients: results from the PRO-
LIVER study. Am J Gastroenterol. 2018;113(3):368-375.
4. Shah NL, Northup PG, Caldwell SH. A clin­i­cal sur­vey of bleed­ing, throm­
bo­sis, and blood prod­uct use in decompensated cir­rho­sis patients. Ann
Hepatol. 2012;11(5):686-690.
5. Violi F, Corazza GR, Caldwell SH, et al; PRO-LIVER Collaborative Group.
Incidence and recur­rence of por­tal vein throm­bo­sis in cir­rhotic patients.
Thromb Haemost. 2019;119(3):496-499.
6. Nery F, Chevret S, Condat B, et al; Groupe d’Etude et de Traitement du
Carcinome Hépatocellulaire. Causes and con­ se­
quences of por­ tal vein
throm­bo­sis in 1,243 patients with cir­rho­sis: results of a lon­gi­tu­di­nal study.
Hepatology. 2015;61(2):660-667.
7. Nery F, Correia S, Macedo C, et al. Nonselective beta-block­ers and the risk
of por­tal vein throm­bo­sis in patients with cir­rho­sis: results of a pro­spec­tive
lon­gi­tu­di­nal study. Aliment Pharmacol Ther. 2019;49(5):582-588.
8. Noronha Ferreira C, Marinho RT, Cortez-Pinto H, et al. Incidence, pre­dic­tive
fac­tors and clin­i­cal sig­nif­i­cance of devel­op­ment of por­tal vein throm­bo­sis
in cir­rho­sis: a pro­spec­tive study. Liver Int. 2019;39(8):1459-1467.
9. Lisman T, Hernandez-Gea V, Magnusson M, et al. The concept of rebal-
anced hemostasis in patients with liver disease: Communication from the
ISTH SSC working group on hemostatic management of patients with liver
disease. Thromb Haemost. 2021;19(4):1116-22.
10. Fisher C, Patel VC, Stoy SH, et al. Balanced haemostasis with both hypo- and
hyper-coag­u­la­ble fea­tures in crit­i­cally ill patients with acute-on-chronic-
liver fail­ure. J Crit Care. 2018;43:54-60.
11. Gatt A, Riddell A, Calvaruso V, Tuddenham EG, Makris M, Burroughs AK.
Enhanced throm­bin gen­er­a­tion in patients with cir­rho­sis-induced coagu-
lopathy. J Thromb Haemost. 2010;8(9):1994-2000.
12. Tripodi A, Primignani M, Chantarangkul V, et  al. Thrombin gen­er­a­tion in
patients with cir­rho­sis: the role of plate­lets. Hepatology. 2006;44(2):440-
445.
13. Lisman T, Bongers TN, Adelmeijer J, et al. Elevated lev­els of von Willebrand
fac­tor in cir­rho­sis sup­port plate­let adhe­sion despite reduced func­tional
capac­ity. Hepatology. 2006;44(1):53-61.
14. Reuken PA, Kussmann A, Kiehntopf M, et al. Imbalance of von Willebrand
fac­tor and its cleav­ing pro­te­ase ADAMTS13 dur­ing sys­temic inflam­ma­tion
superimposed on advanced cir­rho­sis. Liver Int. 2015;35(1):37-45.
15. Ewe K. Bleeding after liver biopsy does not cor­re­late with indi­ces of
periph­eral coag­u­la­tion. Dig Dis Sci. 1981;26(5):388-393.
16. Vieira da Rocha EC, D’Amico EA, Caldwell SH, et al. A pro­spec­tive study of
con­ven­tional and expanded coag­u­la­tion indi­ces in predicting ulcer bleed­ing
after variceal band liga­tion. Clin Gastroenterol Hepatol. 2009;7(9):988-993.
17. Napolitano G, Iacobellis A, Merla A, et  al. Bleeding after inva­ sive pro­
ce­dures is rare and unpredicted by plate­let counts in cir­rhotic patients
with throm­bo­cy­to­pe­nia. Eur J Intern Med. 2017;38:79-82.
18. Desborough MJ, Hockley B, Sekhar M, et al; National Audit Collaborative.
Patterns of blood com­po­nent use in cir­rho­sis: a nation­wide study. Liver Int.
2016;36(4):522-529.
19. Fortea JI, Puente Á, Ezcurra I, et al. Management of haemostatic alter­ations
and asso­ci­ated dis­or­ders in cir­rho­sis in Spain: a national sur­vey. Dig Liver
Dis. 2019;51(1):95-103.
20. Northup PG, Garcia–Pagan JC, Garcia–Tsao G, et  al. Vascular liver dis­or­
ders, por­tal vein throm­bo­sis, and pro­ce­dural bleed­ing in patients with
liver dis­ease: 2020 Practice Guidance by the Amer­i­can Association for the
Study of Liver Diseases. Hepatology. 2021;73(1):366-413.
21. Ginés P, Quintero E, Arroyo V, et al. Compensated cir­rho­sis: nat­u­ral his­tory
and prog­nos­tic fac­tors. Hepatology. 1987;7(1):122-128.
22. Pache I, Bilodeau M. Severe haemorrhage fol­low­ing abdom­i­nal paracen-
tesis for asci­tes in patients with liver dis­ease. Aliment Pharmacol Ther.
2005;21(5):525-529.
23. Hung A, Garcia-Tsao G. Acute kid­ney injury, but not sep­sis, is asso­ci­ated
with higher pro­ce­dure-related bleed­ing in patients with decompensated
cir­rho­sis. Liver Int. 2018;38(8):1437-1441.
24. Massicotte L, Denault AY, Beaulieu D, et al. Transfusion rate for 500 con­sec­
u­tive liver trans­plan­ta­tions: expe­ri­ence of one liver trans­plan­ta­tion cen­ter.
Transplantation. 2012;93(12):1276-1281.
25. Green L, Bolton-Maggs P, Beattie C, et al. Brit­ish Society of Haematology
guide­lines on the spec­trum of fresh fro­zen plasma and cryoprecipitate
prod­ucts: their han­dling and use in var­i­ous patient groups in the absence
of major bleed­ing. Br J Haematol. 2018;181(1):54-67.
26. Zimmon DS, Kessler RE. The por­tal pres­sure-blood vol­ume rela­tion­ship in
cir­rho­sis. Gut. 1974;15(2):99-101.
27. Rassi AB, d’Amico EA, Tripodi A, et al. Fresh fro­zen plasma trans­fu­sion in
patients with cir­rho­sis and coagulopathy: effect on con­ven­tional coag­u­la­
tion tests and thrombomodulin-mod­i­fied throm­bin gen­er­a­tion. J Hepatol.
2020;72(1):85-94.
28. von Meijenfeldt FA, van den Boom BP, Adelmeijer J, Roberts LN, Lisman T,
Bernal W. Prophylactic fresh fro­zen plasma and plate­let trans­fu­sion have
a prothrombotic effect in patients with liver dis­ease. J Thromb Haemost.
2021;19(3):664-676.
29. Tripodi A, Primignani M, Chantarangkul V, et al. Global hemo­sta­sis tests in
patients with cir­rho­sis before and after pro­phy­lac­tic plate­let trans­fu­sion.
Liver Int. 2013;33(3):362-367.
30. Baharoglu MI, Cordonnier C, Al-Shahi Salman R, et al; PATCH Investigators.
Platelet trans­fu­sion ver­sus stan­dard care after acute stroke due to spon­ta­
ne­ous cere­bral haemorrhage asso­ci­ated with antiplatelet ther­apy (PATCH):
a randomised, open-label, phase 3 trial. Lancet. 2016;387(10038):2605-
2613.
31. Peck-Radosavljevic M, Simon K, Iacobellis A, et  al. Lusutrombopag for
the treat­ment of throm­bo­cy­to­pe­nia in patients with chronic liver dis­ease
under­go­ing inva­sive pro­ce­dures (L-PLUS 2). Hepatology. 2019;70(4):1336-
1348.
32. Terrault N, Chen Y-C, Izumi N, et  al. Avatrombopag before pro­ce­dures
reduces need for plate­let trans­fu­sion in patients with chronic liver dis­ease
and throm­bo­cy­to­pe­nia. Gastroenterology. 2018;155(3):705-718.
33. Lindquist I, Olson SR, Li A, et al. The effi­cacy and safety of thrombopoeitin
recep­tor ago­nists in patients with chronic liver dis­ease under­go­ing elec­
tive pro­ce­dures: a sys­tem­atic review and meta-anal­y­sis [published online
18 Jan­u­ary 2021]. Platelets.
34. Moore KP, Wong F, Gines P, et  al. The man­age­ment of asci­tes in cir­rho­
sis: report on the con­sen­sus con­fer­ence of the International Ascites Club.
Hepatology. 2003;38(1):258-266.
35. O’Leary JG, Greenberg CS, Patton HM, Caldwell SH. AGA clin­i­cal prac­tice
update: coag­u­la­tion in cir­rho­sis. Gastroenterology. 2019;157(1):34-43.
36. Patel IJ, Rahim S, Davidson JC, et al. Society of interventional radi­ol­ogy
con­sen­sus guide­lines for the periprocedural man­age­ment of throm­botic
Prakash Singh Shekhawat
and bleed­ing risk in patients under­go­ing per­cu­ta­ne­ous image-guided
inter­ven­tions—part II: rec­om­men­da­tions. J Vasc Interv Radiol. 2019;30
­(8):1168-1184.
37. Simonetto DA, Singal AK, Garcia-Tsao G, Caldwell SH, Ahn J, Kamath PS.
ACG clin­i­cal guide­line: dis­or­ders of the hepatic and mes­en­teric cir­cu­la­tion.
Am J Gastroenterol. 2020;115(1):18-40.
38. Ponziani FR, Zocco MA, Senzolo M, Pompili M, Gasbarrini A, Avolio AW.
Portal vein throm­bo­sis and liver trans­plan­ta­tion: impli­ca­tions for waiting
list period, sur­gi­cal approach, early and late fol­low-up. Transplant Rev
(Orlando). 2014;28(2):92-101.
39. Ambrosino P, Tarantino L, Di Minno G, et al. The risk of venous throm­bo­em­
bo­lism in patients with cir­rho­sis: a sys­tem­atic review and meta-anal­y­sis.
Thromb Haemost. 2017;117(1):139-148.
40. Abdel-Razik A, Mousa N, Elhelaly R, Tawfik A. De-novo por­tal vein throm­
bo­ sis in liver cir­rho­sis: risk fac­tors and cor­ re­la­
tion with the model for
end-stage liver dis­ ease scor­ ing sys­ tem. Eur J Gastroenterol Hepatol.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/485/1851749/485roberts.pdf by guest on 13 December 2021
2015;27(5):585-592.
41. La Mura V, Tripodi A, Tosetti G, et al. Resistance to thrombomodulin is asso­
ci­ated with de novo por­tal vein throm­bo­sis and low sur­vival in patients
with cir­rho­sis. Liver Int. 2016;36(9):1322-1330.
42. Senzolo M, Garcia-Tsao G, García-Pagán JC. Current knowl­edge and man­
age­ment of por­tal vein throm­bo­sis in cir­rho­sis. J Hepatol. 2021;75(2):442-
453.
43. Wang L, Guo X, Xu X, et  al. Anticoagulation favors throm­bus recan­a­li­za­
tion and sur­vival in patients with liver cir­rho­sis and por­tal vein throm­bo­sis:
results of a meta-anal­y­sis. Adv Ther. 2021;38(1):495-520.
44. Giannini EG, Giambruno E, Brunacci M, et al. Low fibrin­o­gen lev­els are asso­
ci­ated with bleed­ing after varices liga­tion in throm­bo­cy­to­pe­nic cir­rhotic
patients. Ann Hepatol. 2018;17(5):830-835.
45. Bianchini M, Cavani G, Bonaccorso A, et al. Low molec­u­lar weight hep­a­rin
does not increase bleed­ing and mor­tal­ity post-endo­scopic variceal band
liga­tion in cir­rhotic patients. Liver Int. 2018;38(7):1253-1262.
46. Delgado MG, Seijo S, Yepes I, et al. Efficacy and safety of anticoagulation
on patients with cir­rho­sis and por­tal vein throm­bo­sis. Clin Gastroenterol
Hepatol. 2012;10(7):776-783.
47. Ortel TL, Neumann I, Ageno W, et  al. Amer­i­can Society of Hematology
2020 guide­lines for man­age­ment of venous throm­bo­em­bo­lism: treat­
ment of deep vein throm­ bo­sis and pul­ mo­ nary embolism. Blood Adv.
2020;4(19):4693-4738.
48. Hoolwerf EW, Kraaijpoel N, Büller HR, van Es N. Direct oral anti­co­ag­u­
lants in patients with liver cir­rho­sis: a sys­tem­atic review. Thromb Res.
2018;170:102-108.
49. Hanafy AS, Abd-Elsalam S, Dawoud MM. Randomized con­trolled trial of
rivaroxaban ver­sus war­fa­rin in the man­age­ment of acute non-neo­plas­tic
por­tal vein throm­bo­sis. Vascul Pharmacol. 2019;113:86-91.
50. Nagaoki Y, Aikata H, Daijyo K, et  al. Efficacy and safety of edoxaban for
treat­ ment of por­ tal vein throm­ bo­sis fol­
low­ing danaparoid sodium in
patients with liver cir­rho­sis. Hepatol Res. 2018;48(1):51-58.
51. Rodríguez-Castro KI, Porte RJ, Nadal E, Germani G, Burra P, Senzolo M.
Management of nonneoplastic por­tal vein throm­bo­sis in the set­ting of liver
trans­plan­ta­tion: a sys­tem­atic review. Transplantation. 2012;94(11):1145-
1153.
52. Kearon C, Ageno W, Cannegieter SC, et al; Subcommittees on Control of
Anticoagulation, and Predictive and Diagnostic Variables in Thrombotic
Disease. Categorization of patients as hav­ing pro­voked or unpro­voked
venous throm­bo­em­bo­lism: guid­ance from the SSC of ISTH. J Thromb Hae-
most. 2016;14(7):1480-1483.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000283
Rebalanced hemo­sta­sis in liver dis­ease  |  491
 NEUTROPENIA: DOING MORE WITH LESS

Diagnosis and therapeutic decision-making

for the neutropenic patient

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021

James A. Connelly1 and Kelly Walkovich2
Department of Pediatrics, Vanderbilt University Medical Center, Nashville, TN; and 2Department of Pediatrics, University of Michigan, Ann Arbor, MI
1

Determining the cause of a low neutrophil count in a pediatric or adult patient is essential for the hematologist’s clini-
cal decision-making. Fundamental to this diagnostic process is establishing the presence or lack of a mature neutrophil
storage pool, as absence places the patient at higher risk for infection and the need for supportive care measures. Many
diagnostic tests, eg, a peripheral blood smear and bone marrow biopsy, remain important tools, but greater understand-
ing of the diversity of neutropenic disorders has added new emphasis on evaluating for immune disorders and genetic
testing. In this article, a structure is provided to assess patients based on the mechanism of neutropenia and to prioritize
testing based on patient age and hypothesized pathophysiology. Common medical quandaries including fever manage-
ment, need for growth factor support, risk of malignant transformation, and curative options in congenital neutropenia
are reviewed to guide medical decision-making in neutropenic patients.

LEARNING OBJECTIVES
• Recognize the different pathophysiologic mechanisms of neutropenia
• Evaluate the cause of neutropenia
• Assess the infectious and malignancy risk in neutropenia patients
• Provide directed therapy and supportive care to neutropenia patients

Introduction and no complaints concerning for an underlying rheuma-

Neutropenia, defined as an absolute neutrophil count (ANC)
<1500/µL after the first year of life, is a common abnormal-
ity confronting the hematologist. Establishing the correct
diagnosis and differentiating benign vs severe pathologic
mechanisms of neutropenia are critical to inform medical
decision-making regarding infectious risk, the utility of sup-
portive care measures, the possibility of malignant trans-
formation, and curative approaches. This article provides
a diagnostic algorithm for evaluation of the pediatric and
adult neutropenic patient followed by evidence-based
approaches, where available, or expert consensus recom-
mendations on the management of medical complications.
Additionally, the complexity inherent to diagnosing and
managing neutropenia is illustrated via a patient case.
CLINICAL CASE
A 16-year-old generally healthy African American adoles-
cent boy is referred for an ANC of 1200/µL incidentally iden-
tified on labs obtained as part of a sports-related physical.
He has no history of recent, recurrent, or unusual infections
tologic condition or malignancy. He takes no medications
and denies recreational drug use. His mother denies any
family history of malignancy, bone marrow failure (BMF)
syndromes, or immunodeficiency from the maternal side
of the family. His height is in the 10th percentile; weight is
in the 10th percentile. The physical exam is unremarkable,
including a detailed oral, skeletal, and skin assessment.
A review of his peripheral smear is consistent with mild neu-
tropenia; neutrophils are normal in morphologic appear-
ance. Hemoglobin level, platelet counts, mean corpuscular
volume, and the remainder of differential are unremarkable.
A repeat complete blood count (CBC) obtained within
1 month of the referral redemonstrates an ANC of 1100/µL.
The patient remains in excellent health and as such is coun-
seled that his ANC result is most likely secondary to the
absence of the Duffy antigen, a phenotype most often seen
in individuals of African, Caribbean, Middle Eastern, and/or
West Indian descent.1 Return instructions are provided.
Author commentary: This adolescent is healthy, with an
incidental finding of neutropenia during a well-child exami-
nation. There are no other CBC abnormalities, and the ANC
is stable on repeat examination. There are no concerning

492 | Hematology 2021 | ASH Education Program

find­ings on his­tory or phys­i­cal exam to sus­pect a con­gen­i­tal
disorder or under­ly­ing malig­nancy that would impair ­neu­tro­phil
pro­duc­tion dur­ ing this ini­
tial visit. Individuals with Duffy-null
(Fy[a−b−]) sta­tus have a lower neu­tro­phil count but are at no
increased risk of infec­tions and should not be labeled as neu-
tropenic. Providing edu­ca­tion for concerning signs and symp­
toms of an evolv­ing mar­row pro­cess with return instruc­tions is
an appro­pri­ate action.
Neutropenia clas­si­fi­ca­tion by path­o­phys­i­­ol­ogy
Classically, the degree of neutropenia is subdivided into mild
(ANC of 1000-1499/µL), mod­ er­
ate (ANC of 500-999/µL), or
severe (ANC <500/µL), with the term “agran­u­lo­cy­to­sis” reserved
for patients with an ANC <200/µL and an absence of neu­tro­phil
pre­cur­sors on bone mar­row exam­i­na­tion. This clas­si­fi­ca­tion sys­
tem is applied to noninfants only because neo­na­tes have an ele­
vated ANC dur­ing the first 2 weeks of life followed by a lower
ANC limit of 1000/µL until 1 year of age. Although this orga­ni­za­
tion is instruc­tive for indi­vid­u­als with a decreased capac­ity for
neu­tro­phil pro­duc­tion, in which the risk for infec­tion increases
as the ANC falls below 1000/µL and the dura­tion of neutropenia
length­ens, it is not as valu­able for predicting infec­tious risk for
other mech­a­nisms of neutropenia. It also fails to rec­og­nize the
eth­nic dif­fer­ences in the ANC sec­ond­ary to Duffy-null sta­tus that
fre­quently result in an ANC <1500/µL.1
An alter­ na­ tive approach is to clas­ sify neutropenia by the
path­o­phys­i­­ol­ogy driv­ing the decreased periph­eral blood ANC.2
The advan­tage of this clas­si­fi­ca­tion sys­tem is that it pro­vi­des a
frame­work to inter­pret the periph­eral blood ANC in the con­text
of diag­nos­tic inves­ti­ga­tions such as a bone mar­row ­exam­i­na­tion
and pro­vi­des insights into man­age­ment. Central to this approach
is appre­ci­at­ing whether the neutropenic patient has a decreased
or nor­mal bone mar­row reserve because cir­cu­lat­ing neu­tro­phils
account for only 3% to 5% of the total body neu­tro­phil count and
there­fore may not reflect the abil­ity to mobi­lize mye­loid cells in
response to exter­nal threats.
Patients may have hin­dered pro­lif­er­a­tion and/or ­mat­u­ra­tion of to assess for recent infec­tion as tran­sient postinfectious neutro-
their neu­tro­phils from intrin­sic defects or extrin­sic fac­tors result- penia is com­mon. The med­i­ca­tion his­tory should be reviewed for
ing in decreased mar­row reserves. Intrinsic defects in myelopoi-
esis stem from germline path­o­genic var­i­ants, eg, ELANE-related
neutropenia, or somatic genetic alter­ations, eg, acquired myel-
odysplastic syn­drome (MDS), that have a neg­a­tive impact on the
effi­ciency or devel­op­ment of mye­loid pro­gen­i­tors into func­tional
poly­mor­pho­nu­clear neu­tro­phils. These patients are at high risk for
infec­tion with­out sup­port­ive and/or defin­i­tive care and may have
a risk of pro­gres­sion to MDS and acute mye­loid leu­ke­mia (AML).
In con­trast, patients with extrin­sic lim­i­ta­tions on myelopoi-
esis have exter­nal pres­sures that inhibit an oth­er­wise nor­mal
bone mar­row from ade­quately pro­duc­ing neu­tro­phils either
through injury to or phys­i­cal obscu­ra­tion of the mar­row niche,
eg, ­neu­ro­blas­toma or fibro­sis or nutri­tional deficiencies. These
patients are also at increased risk of infec­tion; how­ever, many
of the exter­nal pres­sures are treat­able with prompt rec­og­ni­tion.
Patients with a nor­mal mar­row reserve can have neutropenia
through 3 dif­fer­ent mech­a­nisms: inef­fec­tive traf­fick­ing, immune-
medi­ated destruc­tion, or increased con­sump­tion. For patients
with inef­fec­tive traf­fick­ing, neu­tro­phils mature appro­pri­ately but or­ders, eg, anorexia nervosa, or rec­re­a­tional drug use should
are either unable to egress from the mar­row, eg, due to defects
in the CXCL12/CXCR4 axis, as in warts, hypogammaglobulinemia,
infec­tions, myelokathexis (WHIM) syn­ Prakash
drome, or Singh Shekhawat
are stored inef­ aplastic ane­mia sec­ond­ary to ecstasy (3,4-methylenedioxymeth-
fi­ciently, as seen in inap­pro­pri­ate seques­tra­tion of neu­tro­phils
within the mar­gin­ated splenic pool in hypersplenism.3 Immune-
medi­ated destruc­tion is com­mon after infec­tions or drug insults,
can be driven by T- and B-cell auto­im­mu­nity (eg, pri­mary auto­
im­mune neutropenia) as well as innate immune cell i­nflam­ma­tory
cyto­ kine pro­duc­tion (eg, chronic idi­ o­
pathic neutropenia), or
may man­ i­
fest sec­ond­
ary to broader immune syn­ dromes (eg,
sec­ond­ary auto­im­mune neutropenia).4 Increased con­sump­tion is
typ­i­cally driven by height­ened demand dur­ing infec­tions.
Differential diag­no­sis and eval­u­a­tion of neutropenia
Most patients presenting for eval­u­a­tion of neutropenia will have
benign causes. The most likely eti­­ol­o­gies are pri­mar­ily driven by
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021
age (Table 1), high­light­ing the impor­tance of an age-focused his­
tory. For neo­na­tes, it is impor­tant to assess the mater­nal health
with regard to any mater­nal hyper­ten­sion/pre­eclamp­sia asso­
ci­
ated with restricted pla­ cen­ tal blood flow and sub­ se­quent
reduced neo­na­tal bone mar­row pro­duc­tion as well as known
auto­im­mu­nity,5 eg, sys­temic lupus erythematosus or prior preg­
nan­cies com­pli­cated by known antineutrophil immu­no­glob­u­lin
G (IgG) antibodies that may cross the pla­centa and result in allo-
immune or isoimmune neutropenia. Significant com­pli­ca­tions at
deliv­ery, eg, the prolonged rup­ture of mem­branes with chorio-
amnionitis or neo­na­tal sep­sis, can exhaust neu­tro­phil supplies,
resulting in prolonged neutropenia. Additionally, pre­ ma­ture
infants are more likely to have neutropenia as the pro­duc­tion of
neu­tro­phils occurs pri­mar­ily dur­ing the third tri­mes­ter.6
For infants and chil­dren with benign eti­­ol­o­gies, includ­ing pri­
mary auto­im­mune neutropenia due to the devel­op­ment of anti-
neutrophil autoantibodies, symp­toms are usu­ally lacking, and
the dis­cov­ery of neutropenia is inci­den­tal. Additional ques­tions
to screen for more concerning or revers­ible eti­­ol­o­gies of neu-
tropenia include detailing the pres­ence or absence of recur­rent
fevers, oral ulcers, gin­gi­vi­tis, den­tal infec­tions, recur­rent or deep-
seated pyo­genic infec­tions, coli­tis, and nutri­tional deficiencies
sec­ond­ary to restric­tive diets (eg, veg­an­ism or extended total
par­en­teral nutri­tion use) or mal­ab­sorp­tion. Care should be taken
com­monly offending agents used in chil­dren, eg, anti­ep­i­lep­tics
and anti­bi­ot­ics, as with­drawal, close mon­i­tor­ing, or tran­si­tion to
an alter­nate agent may be indi­cated.
Further fea­tures ele­vat­ing con­cern for an under­ly­ing inborn
error of immu­nity,7 such as a per­sonal or fam­ily his­tory of pan­cre­
atic insuf­fi­ciency, malig­nancy (espe­cially mye­loid and lym­phoid),
early edentulism or abnor­mal teeth, early graying, liver or lung
fibro­sis, myco­bac­te­rial infec­tions, warts, lymphedema, non­ma­
lig­nant lymphoproliferation, vas­cu­li­tis/stroke, auto­im­mu­nity, or
child­ hood death from infec­ tion should be solicited because
many con­gen­i­tal neutropenias or syn­dromes inclu­sive of neutro-
penia pres­ent in the first years of life but may have an incom­plete
spec­trum of symp­toms.
For ado­les­cents and young adults, a his­tory sim­i­lar to infants
and chil­dren should be obtained with addi­tional screen­ing ques­
tions related to rheumatologic dis­or­ders since the inci­dence of
sec­ond­ary auto­im­mune neutropenia increases with age. Addi-
tional his­tory ques­tions aimed at assessing risk for eat­ing dis­
be com­pleted. Neutropenia sec­ond­ary to the use of cocaine
adul­ ter­
ated with levamisole is documented,8 as are cases of
Diagnosis and deci­sion-mak­ing for neutropenia  |  493
Table 1. Etiologies of neutropenia by mech­a­nism and age

Neonates Infants/chil­dren Adolescents and young adults Adults

Decreased bone mar­row reserve
Intrinsic defects in - Inborn errors of
myelopoiesis myelopoiesis (eg, ELANE-
related neutropenia)
Extrinsic lim­i­ta­tions on - Maternal hyper­ten­sion
myelopoiesis - Prematurity
- Viral infec­tion
- Infiltration of the mar­row
space (eg, neu­ro­blas­toma,
osteopetrosis)
- Inborn errors of - Inborn errors of myelopoiesis - Acquired MDS
myelopoiesis (eg, SDS, (eg, GATA2 haploinsufficiency, - Inborn errors of
ELANE-related neutropenia, BMF syn­dromes) myelopoiesis (eg, GATA2
GATA2 haploinsufficiency) haploinsufficiency)
- Viral infec­tion - Viral infec­tion - Medications
- Nutritional deficiencies - Medications - Viral infec­tion
- Medications - Idiopathic aplastic ane­mia - Nutritional deficiencies (eg,
- Infiltration of the mar­row - Nutritional deficiencies (eg, vita­min B12 defi­ciency, cop­
space (eg, acute leu­ke­mia) anorexia nervosa) per defi­ciency)
- Drug abuse - Idiopathic aplastic ane­mia

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021

- Alcoholism - Infiltration of the mar­row
space (eg, T-LGL, fibro­sis,
mye­loma)
- Thymoma with pure WBC
aplasia
- Drug abuse
- Alcoholism
Normal bone mar­row reserve
Ineffective traf­fick­ing - Inborn errors of immu­nity - Inborn errors of immu­nity - Inborn errors of immu­nity (eg, - Hypersplenism
(eg, WHIM syn­drome) (eg, WHIM syn­drome) WHIM syn­drome)
- Hypersplenism
Immune-medi­ated - Postinfectious anti­body- - Postinfectious anti­body- - Postinfectious anti­body- - Drug-related anti­body-
destruc­tion medi­ated neutropenia medi­ated neutropenia medi­ated neutropenia medi­ated neutropenia
- Isoimmune neutropenia - Autoimmune neutropenia - Drug-related anti­body- - Postinfectious anti­body-
- Alloimmune neutropenia of infancy medi­ated neutropenia medi­ated neutropenia
- Drug-related anti­body- - Primary auto­im­mune - Autoimmune neutropenia
medi­ated neutropenia neutropenia asso­ci­ated with auto­im­mune
- Autoimmune neutropenia dis­or­ders
asso­ci­ated with inborn errors - Primary auto­im­mune
of immu­nity neutropenia
- Cytokine-induced apo­pto­
sis (eg, chronic idi­o­pathic
neutropenia)
Increased con­sump­tion - Neonatal sep­sis - Bacterial infec­tion - Bacterial infec­tion - Bacterial infec­tion
WBC, white blood cell.

amphetamine) and inhal­ing glue.9,10 Alcoholism can also result
in neutropenia from impaired granulopoiesis.11 Many of the BMF
syn­dromes pres­ent with clin­i­cally sig­nif­i­cant cytopenias in the
sec­ond and third decades of life, so like youn­ger chil­dren, a full
his­tory should be obtained to screen for under­ly­ing eti­­ol­o­gies
that may have been pre­vi­ously qui­es­cent.
For adults, acquired eti­­ol­o­gies, par­tic­u­larly drug expo­sures,
remain the most likely eti­­ol­ogy, and thus a care­ful review of
the patient’s med­i­ca­tion expo­sures is crit­i­cal. Nutritional defi-
ciencies are also com­ mon, eg, cop­ per defi­ciency sec­ond­
ary
to gas­tric bypass.12 Relative to pedi­at­ric patients, both auto­
im­mu­nity and malig­nancy are more likely in adults, so focused
ques­tions regard­ing arthral­gias, arthri­tis, rashes, fatigue, unex­
plained fevers, bone pain, weight loss, and night sweats should
be included in the his­tory.
The phys­i­cal exam eval­ua ­ ­tion for all­patients should focus
on the oral cav­ity to assess for aphthous ulcers, gin­gi­vi­tis, and
den­tal abnor­mal­i­ties sig­nal­ing clin­i­cally sig­nif­i­cant neutropenia
as well as on the assess­ment of height, skel­e­tal abnor­mal­i­ties,
skin changes (eg, albi­nism, malar rash, vit­il­igo, cuta­ne­ous vas­
cu­li­tis, fol­lic­u­li­tis, warts, eczema), nail anom­a­lies, lymph­ade­nop­
a­thy, and hepatosplenomegaly, which are poten­tial clues to an
under­ly­ing eti­­ol­ogy.
Diagnostically, all­patients ben­efi ­ t from a CBC and dif­fer­en­tial,
a mea­sure­ment of liver and renal func­tion, and a direct review
of the periph­eral blood smear to ensure a lack of pseudoneutro-
penia due to in vitro leu­ko­cyte aggre­ga­tion (due to ethylenedi-
aminetetraacetic acid or cold agglu­ti­nins). The remain­der of the
diag­nos­tic eval­u­a­tion should be tai­lored to both the age and pre-
senting clin­ic
­ al fea­tures of the patient (Table 2). An impor­tant fac­
tor in deter­min­ing risk of infec­tion in patients with a low periph­eral
ANC is the pres­ence or absence of a bone mar­row reserve pool of
mature mye­loid cells. Although a bone mar­row biopsy can pro­vide
this infor­ma­tion, it usu­ally is not indi­cated in the ini­tial eval­u­a­tion if
the his­tory and phys­i­cal are nonconcerning for bone mar­row dys­
func­tion. Alternative meth­ods to assess for a bone mar­row reserve
include documenting a rise in the ANC either dur­ing an infec­tious
epi­sode, although fail­ure to increase the ANC may be a false-neg­
a­tive in the set­ting of a viral infec­tion, or through a granulocyte
col­ony-stim­u­lat­ing fac­tor (G-CSF) stim­u­la­tion test in which an ANC
is mea­sured before and 4 to 6 hours after a sin­gle dose.
For neo­na­tes, screen­ing for mater­nal IgG antineutrophil anti-
bodies may assist in secur­ing a diag­no­sis of alloimmune or
isoimmune neutropenia. For infants and chil­dren, antineutro-
phil antibodies can be assessed to lend sup­port to a suspected
diag­no­sis of auto­im­mune neutropenia of infancy; how­ever, the

494  |  Hematology 2021  |  ASH Education Program

Table 2. Laboratory eval­u­a­tion of neutropenia by age
Neonates Infants/chil­dren
Recommended for all­patients
CBC with dif­fer­en­tial CBC with dif­fer­en­tial
Liver and renal func­tion Liver and renal func­tion
Peripheral blood smear Peripheral blood smear
To be con­sid­ered based on patient-spe­cific his­tory
Maternal antineutrophil antibodies Antineutrophil antibodies
Vitamin B12, folate,
cop­per
Amylase iso­en­zymes
Fecal elas­tase
Telomere length
Chromosomal break­age
ADA2 enzyme activity
IgG/A/M/E
T/B/NK sub­sets
T/B/NK, T cell/B cell/nat­u­ral killer cell.
absence of such antibodies does not pre­clude the d ­ iag­no­sis.
If an anti­body is not detected, though, the neutropenia is
often referred to as chronic idi­o­pathic neutropenia of child­
hood.13 Additionally, pos­i­tive antineutrophil antibodies do not
guar­an­tee the absence of other eti­­ol­o­gies, includ­ing severe con­
gen­i­tal neutropenia (SCN).14 Secondary to con­cerns regard­ing the
poor sen­si­tiv­ity and spec­i­fic­ity of avail­­able assays, the mea­sure­
ment of antineutrophil antibodies is not uni­formly prac­ticed by
hema­tol­o­gists. Test per­for­mance can be improved by includ­ing
both the granulocyte immu­no­flu­o­res­cence test and the granulo-
cyte agglu­ti­na­tion test and by repeat­ing the assay.15
Depending on the solicited his­ tory, screen­ ing for pan­cre­
atic insuf­fi­ciency with isoamylase, fecal elas­tase, and a pan­cre­
atic ultra­sound and/or skel­e­tal imag­ing may be use­ful because
many patients with Shwachman-Diamond syn­drome (SDS) have
clin­i­cally sub­tle pan­cre­atic or skel­e­tal defects.16 Other screen­ing
tests for BMF, eg, telo­mere length (short telo­mere syn­dromes),
chro­ mo­somal break­ age (Fanconi ane­ mia), or ADA2 enzyme
activ­ity (a defi­ciency of ADA2), may be appro­pri­ate. Similarly,
screen­ing for under­ly­ing immune defi­ciency with quan­ti­ta­tive
Igs and enu­mer­a­tion of lym­pho­cyte sub­sets and func­tion or for
auto­im­mune dis­ease with an anti­nu­clear anti­body (ANA) may be
of value. Evaluation with folate, vita­min B12, and cop­per is rec-
ommended in all­older adult patients with suspected nutri­tional
deficiencies.
For adults, ANA, rheu­ma­toid fac­tor, eryth­ro­cyte sed­i­men­ta­
tion rate (ESR), and C-reac­tive pro­tein (CRP) tests are valu­able to
screen for auto­im­mune dis­ease. Screening for HIV and for T-cell
large gran­u­lar lym­pho­cyte leu­ke­mia (T-LGL) is recommended
even in patients with­out lym­pho­cy­to­sis or a his­tory of auto­im­
mune dis­ease.17 Depending on the his­tory, addi­tional tests such
as a serum pro­tein elec­tro­pho­re­sis or imag­ing may be warranted.
Regardless of age, if a sat­is­fac­tory eti­­ol­ogy of the neutropenia
is not deter­mined or other key phys­i­cal anom­a­lies or his­tory ele­
ments prompt con­cern for an under­ly­ing dis­or­der, then a bone
mar­row aspi­rate and biopsy to assess myelopoiesis and evi­dence
of clonal evo­lu­tion is warranted. Additionally, given the increased
Prakash Singh Shekhawat
Adolescents and young adults Adults
CBC with dif­fer­en­tial CBC with dif­fer­en­tial
Liver and renal func­tion Liver and renal func­tion
Peripheral blood smear Peripheral blood smear
HIV anti­body screen­ing
T-LGL flow cytom­e­try
Vitamin B12, folate, cop­per Vitamin B12, folate, cop­per
HIV Urine drug screen
Urine drug screen ANA, rheu­ma­toid fac­tor
Amylase iso­en­zymes ESR, CRP
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021
Fecal elas­tase Amylase iso­en­zymes
Telomere length Fecal elas­tase
Chromosomal break­age Telomere length
ADA2 enzyme activity Chromosomal break­age
ANA, ESR, CRP ADA2 enzyme activity
IgG/A/M/E IgG/A/M/E
T/B/NK sub­sets T/B/NK sub­sets
rec­og­ni­tion of neutropenia with inborn errors of hema­to­poi­e­sis
and immu­nity and lim­i­ta­tions in the clin­i­cal clas­si­fi­ca­tion of dis­
ease,18 genetic sequenc­ing is increas­ingly being uti­lized to secure
a molec­u­lar diag­no­sis.19 Confirmation of a genetic diag­no­sis is
valu­able because it facil­i­tates bio­log­i­cally ratio­nal and per­son­
ally tai­lored prog­nos­ti­ca­tion, sup­port­ive care, and ther­apy. For
patients with neutropenia, it can offer spe­cific insights into the
util­ity of G-CSF and the risk of trans­for­ma­tion to malig­nan­cies
and inform deci­sions related to donor selec­tion and con­di­tion­
ing reg­i­mens for hema­to­poi­etic stem cell trans­plant (HSCT). In
some cases, par­tic­u­larly in patients with neutropenia as part of
an inborn error of immu­nity, the genetic diag­no­sis may iden­tify
an exploit­able molec­u­lar path­way that is amenable to targeted
ther­apy. For fur­ther infor­ma­tion on the inter­sec­tion between neu-
tropenia and inborn errors of immu­nity, the read­ers are directed
to the accom­pa­ny­ing arti­cle “Understanding Neutropenia Sec-
ondary to Intrinsic or Iatrogenic Immune Dysregulation” by Dr.
Walkovich. Patients with suspected benign causes of neutrope-
nia should have repeat CBCs every few months to mon­i­tor for
res­o­lu­tion of neutropenia. If the neutropenia does not resolve
in an expected time frame based on the suspected mech­a­nism,
reassessment is warranted. This may include repeat genetic test­
ing if the ini­tial test­ing was conducted more than 2 years ear­lier
to account for newly dis­cov­ered inborn errors of hema­to­poi­e­sis
and immu­nity.
CLINICAL CASE (Con­tin­ued)
Four years later at age 20, the patient returns prior to a planned
sur­gi­cal pro­ce­dure to remove his wis­dom teeth and is con-
cerned his neutropenia may place him at risk of infec­tion. He is
clin­i­cally doing well, but a CBC is nota­ble for an ANC of 1000/µL
and a plate­let count of 90 000/µL. Given the new throm­bo­cy­
to­pe­nia, a bone mar­row biopsy is performed that dem­on­strates
hypocellularity (50%) and mild mye­loid and mega­kar­yo­cyte dys­
pla­sia. Karyotyping reveals a small del(20q) clone but no abnor­
Diagnosis and deci­sion-mak­ing for neutropenia  |  495
mal blast pop­u­la­tion by flow cytom­e­try. The com­pre­hen­sive
mar­row report men­tions the pos­si­bil­ity of SDS, but there is no
his­tory of diar­rhea, and an inherited neutropenia next-gen­er­a­
tion sequenc­ing (NGS) panel is neg­at­ive for path­o­genic var­i­
ants. Given the lack of mor­pho­log­ic ­ al evi­dence of MDS or AML,
it is recommended that the patient be followed closely with
monthly CBCs and a repeat bone mar­row in 3 months.
Author com­men­tary: Unlike his first visit, there are sev­
eral fea­tures dur­ing this eval­u­a­tion concerning for a con­gen­i­
tal neutropenia dis­or­der. His mar­row aspi­ra­tion dem­on­strated
hypocellularity and cel­lu­lar dys­cra­sias, and his cyto­ge­net­ics
iden­ ti­
fied a com­ mon and likely benign clonal hema­ to­
poi­
e­
sis seen in SDS. The phy­si­cian rec­og­nized the asso­ci­at­ion of
del(20q) with SDS but assumed that a lack of ste­at­or­rhea and a
neg­a­tive NGS panel was suf­fic ­ ient to rule out SDS. However, the
clas­sic pre­sen­ta­tion of diar­rhea and neutropenia is only seen
in approx­i­ma­tely 50% of SDS patients.15 It is also impor­tant to
review the genes pres­ent in NGS pan­els to ascer­tain whether
they incor­po­rate suspected genes of inter­est. SBDS may not
be included on some NGS pan­els sec­ond­ary to the pres­ence
of a pseudogene (SBDSP1), requir­ing high-qual­ity map­ping and
infor­mat­ics or other approaches, such as Sanger sequenc­ing,
for dif­fer­en­ti­a­tion. Additional genes (EFL1, DNAJC21, SRP54)
have recently been dis­cov­ered to be asso­ci­ated with SDS and
may not yet have been added to some pan­els. A high sus­pi­cion
of SDS should prompt other screens to detect pan­cre­atic exo­
crine insuf­fi­ciency, includ­ing serum pan­cre­atic isoamylase level
and fecal elas­tase tests, and an abdom­in ­ al ultra­sound to eval­
u­ate for pan­cre­atic atro­phy/fatty replace­ment. The phy­si­cian
should have ordered these tests to help deter­mine a clin­i­cal
diag­no­sis of SDS. The patient’s mar­row find­ings are com­mon
for SDS, but as the diag­no­sis is still unknown in this patient,
close fol­low-up as pre­scribed by the phy­si­cian is appro­pri­ate.
Therapeutic deci­sion-mak­ing for the neutropenic patient
Following iden­ti­fi­ca­tion of the neutropenic patient and eval­ua ­­
tion for an under­ly­ing cause, chal­lenges arise regard­ing treat­
ment deci­sions, mon­i­tor­ing for infec­tion and/or MDS and AML,
and, for some dis­or­ders, con­sid­er­ation of cura­tive options, eg,
HSCT. Unfortunately, for many neutropenic dis­or­ders and com­
pli­ca­tions the evi­dence to sup­port med­i­cal deci­sion-mak­ing is
lacking or reli­ant on expe­ri­ence gained in onco­logic neutrope-
nic patients. The devel­op­ment of neutropenia reg­is­tries such as
the Severe Chronic Neutropenia International Registry (SCNIR)
and the Shwachman-Diamond Syndrome Registry has pro­vided
impactful data to close this infor­ma­tion gap, but much guid­ance
on clin­i­cal deci­sion-mak­ing still depends on expert and con­sen­
sus opin­ion. Knowing this lim­it­a­tion in the field, the fol­low­ing
dis­cus­sion on treating the neutropenic patient is based on evi­
dence where avail­­able but also observes stan­dard treat­ment
prac­ tices in which expert con­ sen­
sus has been achieved and
areas of ongo­ing man­age­ment debate among hema­tol­o­gists.
Management of fever
Historically, fevers have been treated aggres­sively, even in well-
looking patients, over con­cerns that signs of inflam­ma­tion may
be dimin­ished with­out infil­trat­ing neu­tro­phils and that the risk of
not treating an occult bac­ter­emia will lead to poor out­comes.
However, fre­quent hos­pi­tal admis­sions and anti­bi­otic use may
496  |  Hematology 2021  |  ASH Education Program
lead to harm­ful con­se­quences, includ­ing aller­gic reac­tions, an
increase in bac­te­rial resis­tance to anti­bi­ot­ics, hos­pi­tal-acquired
infec­tions, large costs to the health care sys­tem, and a decreased
qual­ity of life for patients with chronic neutropenia. This care­
ful bal­ance must be weighed for each neutropenic patient who
pres­ents with fever, and the approach will dif­fer if the patient
is a known neutropenic patient or pres­ents with newly iden­ti­
fied neutropenia in the con­text of fever. Additional fac­tors such
as the pos­si­bil­ity of neu­tro­phil dys­func­tion, the dura­tion of the
neutropenia, the sta­tus of vac­ci­na­tions, the pres­ence of a cen­
tral venous cath­et­ er, and any con­cur­rent immune defects related
to the under­ly­ing dis­or­der or immu­no­sup­pres­sive med­i­ca­tions
must also be con­sid­ered in deter­min­ing the risk of bac­ter­emia in
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021
the neutropenic patient.
Fever man­
age­
ment in the newly diag­
nosed neutropenic
­patient
The iden­ti­fi­ca­tion of iso­lated neutropenia in a pre­vi­ously healthy
patient presenting with fever is chal­leng­ing, as the abil­ity to
mount neu­tro­phil immu­nity to lower the risk of bac­ter­emia is
likely unknown. Several stud­ies have exam­ined the risk of severe
bac­te­rial infec­tion (SBI) in pre­vi­ously healthy chil­dren who pres­
ent with fever and neutropenia, and the major­ity have dem­on­
strated these patients to be at low risk, with a sim­i­lar prev­a­lence
of bac­te­rial blood and uri­nary tract infec­tions (UTIs) as nonneu-
tropenic patients presenting with fever.20 The larg­ est study,
conducted at the Bos­ ton Children’s Emergency Department,
US, eval­ua ­ ted the out­comes of 1888 pedi­at­ric patients who pre-
sented for eval­u­a­tion of unsus­pected and iso­lated neutropenia.21
For febrile infants <3 months of age, the SBI rate was 6.2%, which
was sim­i­lar to the SBI rate of 7% reported in a prior study of non-
neutropenic infants.22 SBI was very rare in chil­dren over 3 months
of age (2.0%; 1 bac­ter­emia, 15 UTIs, 0 men­in­gi­tis), with the sin­
gle case of bac­ter­emia presenting with obvi­ous clin­i­cal signs of
sep­tic shock in a 9-month-old with Streptococcus pneumoniae.
A study in Barcelona, Spain, eval­u­ated 190 well-appearing febrile
and unsus­ pected neutropenia patients and dem­ on­
strated an
SBI in 4 patients (2.1%; 2 UTIs, 2 pneu­mo­nias), but 101 patients
(53.2%) were admit­ted, and 48 (47.5%) received broad-spec­trum
anti­bi­ot­ics.23 A case con­trol study com­pared pedi­at­ric patients
>3 months of age who presented with fever and unsus­pected
neutropenia with an age-matched con­trol group with­out neutro-
penia. SBIs were iden­ti­fied at a higher rate in the con­trol group
(19 vs 6 patients), which may be explained by the asso­ci­a­tion of
viral infec­tions with fever and neutropenia (Assaf Harofeh Medi-
cal Center, Israel).24 Two addi­tional stud­ies dem­on­strated sim­i­lar
low SBI rates in only 1 of 52 (1.9%) patients aged 1.5 to 36 months
(Children’s Hospital of Wisconsin, US) and in 0 of 51 well-appear-
ing chil­dren but in 5 of 17 (30%) who appeared ill (Dana-Dwek
Children Hospital, Israel).25,26 Higher rates of bac­te­rial infec­tions
have been reported in stud­ies focused on hos­pi­tal­ized neutro-
penic patients (12.7%-14.9%, with 1 study dem­on­strat­ing 0%),27
which may have resulted from the admis­sion of sicker patients
and the infec­tious envi­ron­ment of the pop­u­la­tion, as bru­cel­lo­
sis and rick­ett­sia, bac­te­rial infec­tions known to be asso­ci­ated
with neutropenia, were reported eti­­ol­o­gies in 3 out of 4 of these
­inpa­tient-only stud­ies.28-31
The con­clu­sion from these stud­ies is that the risk of bac­te­rial
infec­tion in the out­pa­tient set­ting for the well-appearing pedi­
at­ric patient with iso­lated and unsus­pected neutropenia is low
and sim­i­lar to patients with nonneutropenic fever. Screening for
bac­te­rial infec­tion, includ­ing a blood cul­ture, a uri­nal­y­sis with
urine cul­ture, and, for symp­tom­atic patients, a chest x-ray, is
suggested, but aggres­sive man­age­ment includ­ing admis­sion
and empiric anti­bi­ot­ics may not be nec­es­sary for all­patients.
Patients who are unwell, have addi­tional cytopenias that raise
con­cern for malig­nancy or BMF, have other con­sti­tu­tional fea­
tures, have a his­tory of infec­tions, or have a fam­ily his­tory of
hema­ to­logic malig­ nan­cies or cytopenias should enact more
aggres­ sive mea­ sures includ­ ing admis­
sion for empiric anti­ bi­
ot­ics that include Pseudomonas spp. and gram-neg­a­tive cov­
er­
age. Similarly, adults who pres­ ent with severe neutropenia
(ANC <500/µL) are more likely to have seri­ous dis­or­ders such as
under­ly­ing malig­nancy and should be admit­ted for eval­u­a­tion for
SBIs.17 Medications asso­ci­ated with the sup­pres­sion of myelopoi-
esis that are not of vital impor­tance should also be discontinued
because drug-induced agran­u­lo­cy­to­sis can lead to fatal­ity with
con­tin­ued expo­sure.32
Fever man­age­ment in the established neutropenic patient
The approach to the patient with known neutropenia is often less
chal­leng­ing, as the eti­­ol­ogy of the neutropenia and the his­tory
of infec­tious com­pli­ca­tions may guide man­age­ment. Disorders
such as pri­mary auto­im­mune neutropenia and chronic idi­o­pathic
neutropenia typ­ i­
cally have ade­ quate bone mar­ row stor­ age
pools of neu­tro­phils to respond to infec­tion and there­fore pres­
ent a low risk for SBIs. In a study of 43 pedi­at­ric patients with
pri­mary auto­im­mune or chronic idi­o­pathic neutropenia, there
were 133 emer­gency room encoun­ters for fever with no patients
iden­ti­fied as hav­ing a true bac­te­rial infec­tion (Texas Children’s
Cancer and Hematology Centers, US).33 In the Ital­ian Neutrope-
nia Registry, over the course of 9.5 years they iden­ti­fied only 1
blood­stream infec­tion in 38 patients with auto­im­mune neutro-
penia and 2 blood­stream infec­tions in 23 idi­o­pathic neutropenia
patients.34 For the pedi­at­ric patient with auto­im­mune or chronic
idi­o­pathic neutropenia, if a rise in the neu­tro­phil count can be
dem­on­strated dur­ing times of infec­tion and there is no his­tory
of severe infec­tions, eval­u­a­tion by the child’s pri­mary care phy­si­
cian should be suf­fi­cient, with emer­gency room vis­its lim­ited to
when the patient appears unwell.35
Adults with pri­mary auto­im­mune and chronic idi­o­pathic neu-
tropenia usu­ally do not develop severe infec­tions, although they
may suf­fer from recur­rent sto­ma­ti­tis and respi­ra­tory infec­tions.35
This was dem­on­strated in a French Severe Chronic Neutropenia
Registry report in which adult patients with pri­mary auto­im­mune
and chronic idi­o­pathic neutropenia had low rates of SBIs at 3.85
per 100 patient-years.36 Secondary auto­im­mune neutropenia in
both chil­dren and adults is often asso­ci­ated with addi­tional cyto-
penias, immune defects, or immu­no­sup­pres­sive med­i­ca­tions
and car­ries a higher risk of infec­tion. The Ital­ian Neutropenia
Registry reported SBIs in 10 of 26 (40%) patients with sec­ond­
ary auto­im­mune neutropenia, which is sig­nif­i­cantly higher than
in the 11.8% of patients with pri­mary auto­im­mune neutropenia,
over the course of 15 years.37 Extra cau­tion should there­fore be
taken in patients with sec­ond­ary auto­im­mune neutropenia who
pres­ent with fever, includ­ing con­sid­er­ation of fun­gal and other
atyp­i­cal infec­tions related to addi­tional T-cell immu­no­de­fi­ciency.
Disorders of impaired myelopoiesis, includ­ing SCN, will lead
to an impaired abil­ity to mount neu­tro­phil immu­nity in the set­
ting of infec­tion. For patients with a poor response to G-CSF, an
Prakash Singh Shekhawat
emer­gent eval­u­a­tion for any fever is nec­es­sary sec­ond­ary to the
risk of bac­te­rial sep­sis. However, patients who have con­sis­tent
ANCs above 1000/µL are at low risk of infec­tion,13 and man­age­
ment of fever may be based on the clin­i­cal appear­ance of the
patient, with unwell patients eval­u­ated in the emer­gency room
and well patients eval­u­ated in out­pa­tient clin­ics.
Indications for G-CSF
The use of G-CSF is mainly employed in patients with defec­tive
myelopoiesis and has proven to be a life­sav­ing inter­ven­tion in
patients with SCN. The SCNIR rec­om­mends a starting dose of
5  µg/kg/d for patients with SCN with esca­la­tion to 10  µg/kg/d
and then an increase by incre­ments of 10  µg/kg/d at 14-day inter­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021
vals while the ANC remains below 1000/µL.38 The main­te­nance
dose of G-CSF for SCN is var­i­able depending on the genetic dis­
or­der as well as the spe­cific path­o­genic var­i­ant within the gene,
with a median dose of 7.3  µg/kg/d needed to achieve an ANC
between 1000 and 1500/µL.39 Higher ANCs are not nec­es­sary in
most patients, and maintaining the smallest effec­tive dose avoids
the con­se­quences of higher dos­ing, includ­ing bone pain, spleno­
meg­aly, and the­o­ret­i­cal higher selec­tive pres­sure for hema­to­poi­
etic clones more respon­sive to G-CSF.40 For cyclic neutropenia,
the dose is much lower, typ­i­cally 1 to 3  µg/kg/d given every 1 to
3 days, which short­ens the dura­tion of the neutropenic period
and effec­tively elim­i­na­tes sep­sis and death from sep­sis based on
out­come data from the SCNIR.13 Granulocyte-mac­ro­phage-CSF
treat­ment is inef­fec­tive in most forms of con­gen­i­tal neutropenia,
as, unlike G-CSF, it is unable to induce nic­o­tin­amide phosphori-
bosyl trans­fer­ase-depen­dent granulopoiesis.41
G-CSF use in acquired causes of neutropenia is typ­i­cally guided
by the patient his­tory and is likely safe, as no defin­i­tive asso­ci­a­tion
with G-CSF use and can­cer devel­op­ment has been established in
this patient pop­u­la­tion.35 G-CSF treat­ment for pri­mary auto­im­mune
and chronic idi­o­pathic neutropenia should not be deter­mined
based on the detec­tion of antineutrophil antibodies but rather is
recommended for patients with severe neutropenia with recur­
rent infec­tions or sto­ma­ti­tis.42 As mye­loid pro­duc­tion is rel­a­tively
pre­served in these con­di­tions, small doses of 0.5 to 3  µg/kg/d
of G-CSF given every 1 to 3 days can be used to increase the
ANC to a tar­get main­te­nance of 1000 to 1500/uL.43 In the French
Severe Chronic Neutropenia Registry, the use of G-CSF pro­vided
a clin­i­cal ben­e­fit in adult patients with pri­mary auto­im­mune or
chronic idi­o­pathic neutropenia in 23 of 24 and 20 of 26 rou­tinely
sched­uled and spo­radic G-CSF cohorts, respec­tively.36 Intermit-
tent dos­ing of G-CSF may be required in patients who pres­ent
with severe infec­tion or in the set­ting of sur­gi­cal pro­ce­dures at
high risk for infec­tion or poor wound healing.44 G-CSF is almost
always given in patients with suspected drug-induced agran­u­
lo­cy­to­sis, although evi­dence supporting this rec­om­men­da­tion is
lacking,35 and may also be used in patients with only mod­er­ate
drug-induced sup­pres­sion of myelopoiesis in whom con­tin­u­a­tion
of the offending agent is nec­es­sary.
Role of immune sup­pres­sion in auto­im­mune neutropenia
Immune sup­pres­sion to treat pri­mary auto­im­mune and chronic
idi­o­pathic neutropenia is not recommended for chil­dren due to
poor response and low infec­tious risk and has lim­ited suc­cess in
adults.44 In the French Severe Congenital Neutropenia Registry,
response to com­monly used immu­no­sup­pres­sant agents (gluco-
corticosteroids, meth­o­trex­ate, cyclo­spor­ine) was observed in half
Diagnosis and deci­sion-mak­ing for neutropenia  |  497
Table 3. Congenital neutropenia dis­or­ders at risk for MDS/AML

Disease, gene, and inher­i­tance Additional infor­ma­tion includ­ing

pat­tern Risk of MDS/AML Common somatic alter­ations nonhematologic fea­tures
ELANE SCN 15%-20%51; increased inci­dence CSF3R, RUNX1, ASXL1, SUZ12, Acquired CSF3R muta­tions
ELANE, AD of MDS/AML in patients on high EP300, mono­somy 7, tri­somy com­mon (approx­i­ma­tely 50%) in
doses of G-CSF with poor 21.75 patients with­out MDS/AML.75
neu­tro­phil response.54
Cyclic neutropenia AML rarely reported.76 CSF3R, mono­somy 7, tri­somy 21 in Acquired CSF3R muta­tions rarely
ELANE, AD the 1 patient with AML.76 iden­ti­fied in patients with­out
MDS/AML.76
HAX1-SCN MDS/AML reported, but inci­ Somatic alter­ations not published. Acquired CSF3R muta­tions
HAX1, AR dence of leu­ke­mia not pre­cisely com­mon (approx­i­ma­tely 45%) in
reported.77 patients with­out MDS/AML.75

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021

G6PC3 defi­ciency AML rarely reported.78
G6PC3, AR
X-linked neutropenia MDS/AML rarely reported.79
WAS GOF, XL
Glycogen stor­age dis­ease 4% with MDS/AML in the SCNIR.80
type 1b
G6PT1, AR
Poikiloderma with neutropenia MDS/AML reported, but inci­
USB1, AR dence of leu­ke­mia not pre­cisely
reported.82 Myelodysplastic
changes in most patients.
SDS 20%-30%83,84; patients also at risk
SBDS, AR of BMF.16
GATA2 haploinsufficiency 80%86; occur­rence typ­ic
­ ally in later
GATA2, AD ado­les­cence and early adult­
hood and may be presenting
fea­ture.
SAMD9/SAMD9L GOF dis­or­der Risk of MDS and AML unknown.
SAMD9 or SAMD9L, AD Patients also at risk of BMF that
may spon­ta­ne­ously resolve.88
AD, auto­so­mal dom­in
­ ant; AR, auto­so­mal reces­sive, GOF, gain of func­tion; XL, X-linked.
T(8;21)(q22;q22) iden­ti­fied by kar­
yo­type in one patient.78
Somatic alter­ations not published.
Monosomy 7. Short telo­meres also
reported.81
Somatic alter­ations not published.
Biallelic TP53, RUNX1, SETBP1,
BRAF, NRAS, ETNK1, alter­ations
in chro­mo­somes 3, 5, 7, and
20.85
Monosomy 7, tri­somy 8, ASXL1,
der(1;7)(q10;p10).87
Monosomy 7, del(7q), RUNX1,
SETBP1, ETV6.89
Inflammatory bowel dis­ease,
con­gen­i­tal heart and uro­gen­i­tal
defects, and endo­crine dis­or­
ders. Acquired CSF3R muta­tions
iden­ti­fied in patients with­out
MDS/AML.75
Acquired CSF3R muta­tions
iden­ti­fied in patients with­out
MDS/AML.75
Hypoglycemia, hepa­to­meg­aly, and
entero­co­li­tis.
Poikiloderma, nail dys­tro­phy, and
pal­mar/plan­tar hyper­ker­a­to­sis.
Pancreatic exo­crine insuf­fi­ciency,
skel­e­tal abnor­mal­i­ties. Somatic
alter­ations not asso­ci­ated with
MDS/AML include i(7)(q10),
del(20q), and EIF6 muta­tions.
Warts, lymphedema, panniculitis,
deaf­ness, infec­tion, and pul­mo­
nary alve­o­lar proteinosis.
SAMD9: adre­nal hypo­pla­sia,
growth restric­tion, gen­i­tal
abnor­mal­i­ties, enter­op­a­thy,
infec­tions.
SAMD9L: neu­ro­logic symp­toms,
infec­tions.

of the adult patients with pri­mary auto­im­mune or chronic idi­o­
pathic neutropenia, but the response was mostly par­tial and was
lost upon taper­ing or discontinuing the treat­ment.36 In the SCNIR,
almost all­of the 48 adult patients with auto­im­mune or chronic idi­o­
pathic neutropenia on immune mod­u­la­tion (glucocorticosteroids,
gamma glob­u­lin, meth­o­trex­ate, cyclo­spor­ine) discontinued this
ther­apy with­out resump­tion once placed on G-CSF.42 Additional
reports have also been published dem­on­strat­ing the low response
rate and high relapse rate with glucocorticosteroids, gamma glob­
u­lin, cyclo­spor­ine, and rituximab.45 However, treat­ment with these
agents may be con­sid­ered if patients fail to respond to G-CSF.
In con­trast, although rarely requir­ing treat­ment, ­sec­ond­ary
auto­im­mune neutropenia often responds to the immune sup­pres­
sion used to treat other auto­im­mune symp­toms.17 LGL-asso­ci­ated
neutropenia is often severe and requires treat­ment. G-CSF can
increase the neu­tro­phil count but is asso­ci­ated with the exac­er­
ba­tion of spleno­meg­aly and the acute flare of joint symp­toms
while cyclo­phos­pha­mide, meth­o­trex­ate, and cyclo­spor­ine can
be effec­tive.46
Monitoring con­gen­i­tal neutropenia patients at risk
for MDS/AML
Many con­gen­i­tal neutropenia dis­or­ders impose a risk of trans­
for­ma­tion to a mye­loid malig­nancy (see Table 3). The mech­a­
nism of leu­ke­mo­gen­e­sis in these dis­or­ders is still actively under
inves­ti­ga­tion and is unique to each dis­or­der. As knowl­edge of
leu­ke­mo­gen­e­sis in these dis­or­ders increases, the abil­ity to pre­
dict and detect early malig­nant cells will hope­fully be a nat­ur­al
con­se­quence. Historic and cur­rent rec­om­men­da­tions are to per­
form fre­quent (every 3-4 months) CBCs and annual bone mar­row

498  |  Hematology 2021  |  ASH Education Program

­ spi­ra­tion and biop­sies to mon­i­tor for cytopenias and mor­pho­
a has spe­cial con­sid­er­ations in con­gen­i­tal neutropenia. The mea­
logic changes concerning for MDS/AML devel­op­ment in con­gen­i­
tal neutropenias sec­ond­ary to muta­tions in ELANE, HAX1, G6PC3,
SBDS, WAS (X-linked neutropenia), and GATA2.43 However, data
are needed to inform sur­veil­lance prac­tices and their effect on
out­comes for patients with con­gen­i­tal neutropenia.
CLINICAL CASE (Con­t in­u ed)
The patient undergoes CBCs every 3 to 4 months and 2 sur­veil­
lance bone mar­row exam­i­na­tions over the fol­low­ing year. His
CBCs con­tinue to show mild neutropenia and throm­bo­cy­to­pe­
nia, and his bone mar­row results are con­sis­tent with find­ings
from his first exam­i­na­tion. At the age of 22, he undergoes another
bone mar­row biopsy, which dem­on­strates an i(7)(q10) clone on
cyto­ge­net­ics and a TP53 path­o­genic var­i­ant in a somatic NGS
panel. There con­tin­ues to be no mor­pho­logic or flow cytomet-
ric evi­dence of increas­ing dys­pla­sia or an abnor­mal blast count.
Genetic test­ing is revisited, and Sanger sequenc­ing of the SBDS
locus identifies 2 path­o­genic muta­tions inde­pen­dently inherited
from each par­ent. The pres­ence of a chro­mo­some 7 abnor­mal­
ity and a new TP53 clonal hema­to­poi­e­sis raises the con­cern for
a malig­nancy, and the patient is referred for HSCT con­sul­ta­tion.
Author com­men­tary: The phy­si­cian was ­able to diag­nose
SDS in this patient with genetic test­ing, but this case illus­trates
the com­plex­ity of clin­i­cal sequenc­ing in that NGS tech­nol­ogy
has sev­eral short­com­ings, includ­ing the inabil­ity to dif­fer­en­ti­ate
genes from pseudogenes with­out excel­lent map­ping and ana­lyt­
ics and to iden­tify deep intronic var­i­ants. In most clin­i­cal sce­nar­
ios, chro­mo­some 7 aber­ra­tions raise the con­cern for MDS/AML.
But sim­i­lar to del(20q), i(7)(q10) is a somatic change that does not
have an impact on tumor sur­veil­lance machin­ery. TP53 muta­tions
increase cel­lu­lar fit­ness in SDS, but at the expense of decreased
tumor sur­veil­lance, and likely increase the risk of MDS/AML within
the clone. However, TP53 clonal hema­to­poi­e­sis is com­mon in
SDS with­out malig­nancy, and iden­ti­fi­ca­tion by itself is not an indi­
ca­tion for HSCT. After dis­cus­sions with experts in SDS, a deci­sion
is made for con­tin­ued close mon­i­tor­ing that includes CBCs every
3 to 4 months and annual bone mar­row mon­i­tor­ing only.
Curative approach with HSCT for con­gen­i­tal neutropenia
Absolute and pos­si­ble indi­ca­tions to pro­ceed to trans­plant
depend on the under­ly­ing genetic con­di­tion, but com­mon
causes to pro­ceed to HSCT include poor response to G-CSF,
recur­rent and severe infec­tions, BMF, and trans­for­ma­tion to a
mye­loid malig­nancy. The approach to HSCT is not uni­form within
or between genetic dis­or­ders, but each trans­plant method shares
unique obsta­cles to suc­cess­in neutropenic patients. The risk
of bac­te­rial and fun­gal infec­tion is higher early in trans­plant due
to pre­ced­ing neutropenia that rap­idly declines with the dis­con­
tin­u­a­tion of G-CSF. Evaluation for occult infec­tion pretransplant
with sinus, chest, abdo­men, and pel­vis imag­ing is employed,
with more inva­sive mea­sures such as bron­chos­copy or biopsy
indi­cated for areas concerning for infec­tion. Bacterial and mold
pro­phy­laxis at the start of con­di­tion­ing ther­apy is ini­ti­ated to
mit­i­gate the very early neutropenia in these patients.
Chimerism anal­y­sis to dif­fer­en­ti­ate between resid­ual and
engrafted donor hema­to­poi­e­sis is impor­tant for all­HSCTs but
Prakash Singh Shekhawat
sure­ment of hematopoietic progenitor cell (HPC) chi­me­rism is
dif­fi­cult in the clin­i­cal set­ting, and typ­i­cally, the mea­sure­ment of
mature mye­loid cells in the periph­eral blood is used as a sur­ro­
gate for HPC engraft­ment. However, mye­loid mark­ers, such as
CD33, that are pres­ent on neu­tro­phils will not be an accu­rate
proxy for HPC chi­me­rism because recip­i­ent HPCs will have a
sig­nif­i­cant dis­ad­van­tage com­pared to donor HPCs in con­trib­ut­
ing to the periph­eral blood neu­tro­phil pool. Therefore, markers
weakly expressed or absent on neutrophils, such as CD14 (mono-
cytes), CD56 (natural killer cells), CD19 (B cells), CD34 (HPCs),
and CD3 (T cells), should be included in measuring peripheral
blood chimerism, particulalry in disorders such as GATA2 haplo-
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021
insufficiency, which affect multiple cell lineages.
The impact of par­ tial chi­
me­rism and long-term out­ comes
for patients with con­gen­i­tal neutropenia is poorly under­stood,
in large part because myeloablative reg­i­mens have dom­i­nated
early trans­plant approaches, and lit­er­a­ture reports have not rou­
tinely focused on this trans­plant result or relied on neu­tro­phil-
spe­cific mark­ers for chi­me­rism reporting. Mixed chi­me­rism has
been shown to cure the neutropenia in con­gen­i­tal neutropenia
patients.47 Although the per­cent­age required to elim­i­nate the
risk of bac­te­rial infec­tion and G-CSF inde­pen­dence is uncer­tain,
from anec­dotal reports and expe­ri­ence it is likely 20% to 30%
HPC donor chi­me­rism.47
The more dif­fi­cult ques­tion is whether the per­sis­tence of recip­
i­ent hema­to­poi­e­sis fol­low­ing expo­sure to che­mo­ther­apy will con­
tinue to pres­ent a risk for mye­loid malig­nan­cies post trans­plant.
Our cur­rent under­stand­ing of leu­ke­mo­gen­e­sis in each dis­or­der pro­
vi­des the­o­ret­i­cal hypoth­e­ses, but prolonged fol­low-up in patients
with mixed chi­me­rism will be required to answer this ques­tion
defin­i­tively. In patients with ELANE SCN, a mar­row micro­en­vi­ron­
ment with high lev­els of G-CSF pro­vi­des a com­pet­i­tive set­ting to
select out clones with enhanced response to G-CSF. However, in
the posttransplant set­ting with abla­tion of the neutropenia defect
by donor HPCs, the selec­tive envi­ron­ment driven by high lev­els of
G-CSF is elim­i­nated. If this pos­tu­late is cor­rect, the risk of malig­
nant trans­for­ma­tion should also be min­i­mized, if not abolished,
with donor hema­to­poi­e­sis. In patients with SDS, the leu­ke­mo­gen­
e­sis hypoth­e­sis is less reli­ant on exter­nal pres­sures for hema­to­
poi­e­sis but depen­dent on indi­vid­ual cell fit­ness through reliev­ing
ribo­somal stress through TP53 muta­tions. Whether the reduc­tion
of rep­li­ca­tive and inflam­ma­tory stress through donor hema­to­poi­
e­sis is suf­fi­cient to pre­vent leu­ke­mo­gen­e­sis in recip­i­ent stem cells
under ribo­somal stress needs fur­ther study. As leu­ke­mia may not
be a com­mon event in posttransplant patients, research eval­u­at­
ing clonal dynam­ics with a par­tic­u­lar focus on CSF3R in SCN and
TP53 in SDS will be incred­i­bly valu­able to under­stand­ing hema­to­
poi­e­sis dynam­ics in the mixed-chi­mera patient.
Graft rejec­tion for both malig­nant and non­ma­lig­nant indi­ca­
tions in con­gen­i­tal neutropenia is rel­a­tively high com­pared to
other com­pa­ra­ble con­di­tions. Several hypoth­e­ses have been
gen­er­ated to explain this phe­nom­e­non, includ­ing ­com­pe­tent
lym­pho­cyte immu­nity pre­ced­ing con­di­tion­ing ther­apy, an inflam­
ma­tory micro­en­vi­ron­ment induced from recur­rent infec­tion,
alter­ations in the bone mar­row niche sec­ond­ary to the under­ly­
ing defect or high expo­sure to G-CSF,48 and prior expo­sure to for­
eign blood prod­ucts in patients with BMF. Serotherapy is almost
stan­ dard today for non­ ma­lig­
nant trans­ plant to help address
con­cerns for increased alloreactivity, but ques­tions regard­ing
Diagnosis and deci­sion-mak­ing for neutropenia  |  499
the mar­row niche are more dif­fic ­ ult and will require con­tin­ued
research in this area, par­tic­u­larly if pre- and posttransplant mar­
row aspi­ra­tion sam­ples are avail­­able for study.
Congenital neutropenia patients who prog­ress to malig­nancy
pres­ent the most chal­leng­ing and most con­tro­ver­sial sce­nario for
cli­ni­cians. First, the diag­no­sis of trans­for­ma­tion is not straight­for­
ward because dif­fer­en­ti­at­ing hema­to­poi­etic dys­cra­sias pres­ent
in many inborn errors of hema­to­poi­e­sis from dys­plas­tic clones
can be dif­fi­cult even for expe­ri­enced hematopathologists. This
was illus­trated in a study of SDS patients with MDS/AML who
under­went a cen­tral path­o­logic review that showed a 56% dis­
crep­ancy from the orig­i­nal pathol­ogy report.49 Clonal hema­to­
poi­e­sis is also com­mon in con­gen­i­tal neutropenia, and some of
these changes are not believed to be induc­ers of leu­ke­mo­genic
changes. Knowledge of path­o­logic clonal hema­to­poi­e­sis is key
in interpreting these find­ings.
Second, cytoreductive che­mo­ther­apy to reduce blast counts,
as is typ­i­cally done for de novo AML, can be incred­i­bly toxic to
con­gen­i­tal neutropenia patients who have dif­fi­culty recov­er­ing
neu­tro­phils for infec­tion erad­i­ca­tion and tis­sue repair or who
have under­ly­ing organ dys­func­tion from the under­ly­ing genetic
defect or recur­rent infec­tions.50 As such, recent approaches often
elim­ i­
nate cytoreductive ther­ apy com­ pletely or have adopted
reg­i­mens used in med­i­cally frag­ile patients to bridge the patient
while awaiting iden­ti­fi­ca­tion of a suit­able bone mar­row donor.
Driving blast pro­duc­tion with G-CSF has been dem­on­strated in
SCN, and judi­cious use fol­low­ing che­mo­ther­apy is warranted. The
admis­sion of all­ patients fol­low­ing che­mo­ther­apy to mon­i­tor for
infec­tion and insti­tu­tion of G-CSF for evi­dence of severe bac­te­rial
or fun­gal dis­ease is a com­mon approach to bal­ance these risks.
G-CSF is not recommended post HSCT for patients with SCN and
malig­nancy with­out severe infec­tion, but whether G-CSF use fol­
low­ing trans­plant increases the risk of relapse is unknown.
HSCT out­comes for SCN
HSCT for SCN is conducted for G-CSF refrac­to­ri­ness or the devel­
op­ment of MDS/AML. Patients with ELANE muta­tions at high risk
of pro­gres­sion to AML, includ­ing p.G214R, p.C151Y, and p.C223ter,
may be con­sid­ered for pre­emp­tive HSCT, but data to inform
the risks and ben­e­fits of this prac­tice are cur­rently lacking.51,52
Some groups also con­sider high-dose G-CSF (>15 µg/kg/d) as an
indi­ca­tion to offer trans­plant because patients on higher G-CSF
doses with decreased ANC response have shown an increased
fre­quency of leu­ke­mia devel­op­ment.53,54 This approach has not
been uni­ver­sally adopted sec­ond­ary to the desire to avoid HSCT
expo­sure to the major­ity of patients on high-dose G-CSF who
will not develop malig­nancy.
Most lit­er­a­ture on HSCT for SCN is based on case reports
and small case series that have dem­on­strated good out­comes
in patients with­out malig­nancy (over­all sur­vival [OS], approx­i­ma­
tely 90%; event-free sur­vival [EFS], approx­im ­ a­tely 75%) despite
high rates of graft fail­ure (approx­i­ma­tely 20%).55 Historical out­
comes with malig­nancy have been less suc­cess­ful, with an OS of
approx­i­ma­tely 45% and an EFS of approx­i­ma­tely 40%, but many
cases were transplanted prior to the adop­tion of mod­ern prac­
tices to avoid inten­sive cytoreductive ther­apy prior to trans­plant.
A larger report of 136 SCN patients from the Euro­pean Society
for Blood and Marrow Transplantation dem­on­strated a 3-year OS
and EFS of 87% and 77% for patients with­out malig­nancy and
79% and 64% with malig­nancy.56 Graft rejec­tion was 10% for the
500  |  Hematology 2021  |  ASH Education Program
entire cohort, but relapse was only seen in 2 of 14 (14%) patients.
An Amer­i­can Society of Hematology abstract from the Euro­pean
branch of the SCNIR reported out­comes on 51 patients after the
2001 implementation of guide­lines to avoid induc­tion che­mo­ther­
apy and dem­on­strated an OS of 77.8% and 83.3% in SCN patients
treated with and with­out malig­nancy.57 Despite the fact that not
all­patients with SCN and sec­ ond­ ary malig­
nancy pro­ceed to
HSCT in remis­sion, the reported relapse rates are much lower in
com­par­i­son to de novo MDS/AML (10%-15% vs 25%-40% for MDS
and 40%-50% for AML).55,56,58-60 The rea­son for reduced relapsed
rates may be sec­ond­ary to the growth restraint on malig­nant
cells afforded by the germline defect, a fea­ture reflected in some
case reports of SCN patients with “indo­lent” AML.61 Although
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021
reported OS between myeloablation (MA) and reduced inten­
sity con­di­tion­ing (RIC) is sim­i­lar in SCN patients, myeloablative
(MA) con­di­tion­ing is recommended for patients with under­ly­ing
malig­nancy for elim­i­na­tion of the malig­nant clone.
HSCT out­comes for SDS
HSCT for SDS is conducted in patients with BMF or hema­to­logic
malig­nancy. Unlike some patients with SCN, the con­sen­sus is to
not per­form HSCT pre­emp­tively in this dis­or­der. Early expe­ri­
ence in HSCT dem­on­strated high nonrelapse trans­plant mor­tal­
ity, as shown in a recent long-term out­come study reporting that
21% of patients died from tox­ic­ity.62 This real­i­za­tion prompted
many cli­ni­cians to use RIC for all­SDS patients with BMF.63,64 In a
study by the Center for International Blood and Marrow Trans-
plant Research, early deaths were seen in 4 of 13 (31%) SDS BMF
patients receiv­ing a MA con­di­tion­ing reg­i­men com­pared to
4 of 26 (15%) early deaths in patients receiv­ing RIC.65 This report
showed a 5-year OS of 72% in patients with BMF and graft rejec­
tion in 3 of 39 patients (8%). The larg­est study to date published
by the Severe Aplastic Anemia Working Party of the Euro­pean
Society for Blood and Marrow Transplantation on 61 patients
(1988-2016) with BMF dem­on­strated a 5-year OS of 70.7% with a
graft fail­ure rate of 13%.62 OS with RIC and MA was iden­ti­cal, but
the seg­re­ga­tion of con­di­tion­ing ther­apy by indi­ca­tion (BMF vs
leu­ke­mia) was not pro­vided, and the authors con­cluded that MA
reg­i­mens should be reserved for malig­nant trans­for­ma­tion only.
Transplant sur­vival for SDS patients who develop malig­nancy
are abys­mal across lit­er­a­ture reports. A mul­ti­cen­ter study focused
on malig­nant trans­plant out­comes dem­on­strated a 3-year OS
of 51% and 11% for patients with MDS and AML, respec­tively.49
This result is in con­ trast to recent SCN out­ comes for malig­
nancy, indi­cat­ing a higher degree of resis­tance to trans­plant,
per­haps in part related to increased che­mo­ther­apy resis­tance
from mutated TP53, as seen in other can­cers.66 The impor­tance
of sur­veil­lance was suggested by an improved 3-year OS of 62%
vs 28% in patients who had reg­u­lar bone mar­row eval­u­at­ion,
but reg­u­lar CBCs had poor sen­si­tiv­ity in iden­ti­fy­ing mar­row dis­
ease.49 Although early detec­tion may pro­vide an incre­men­tal
improve­ment in sur­vival, novel approaches to avoid exces­sive
tox­ic­ity from con­di­tion­ing ther­apy while pre­vent­ing relapse are
clearly needed for SDS patients who develop malig­nancy.
Gene ther­apy and other poten­tial treat­ment modal­i­ties
for con­gen­i­tal neutropenia
As our under­stand­ing of the mech­a­nism of neutropenia and leu­
ke­mia for­ma­tion evolves, new ther­a­peu­tic modal­i­ties have been
tested in pre­clin­i­cal mod­els with a pri­mary focus on increas­ing
neu­tro­phil mat­u­ra­tion and pro­duc­tion. Investigative ther­a­pies
include nic­o­tin­amide to increase the expres­sion of G-CSF and
G-CSF recep­tors in SCN,67,68 neu­tro­phil elas­tase inhib­i­tors to
pre­vent the unfolded pro­tein response from mutated ELANE,69,70
and gene editing to repair mutated ELANE or reduce trans­la­tion
of the ELANE gene.71-73 Nicotinamide has been used in a small
clin­i­cal trial with pos­i­tive results, but other modal­i­ties have
only reported results in the pre­clin­i­cal set­ting. The use of small-
mol­e­cule inhib­i­tors to pro­mote non­sense sup­pres­sion in
patients with ter­mi­na­tion SBDS muta­tions (c.183-184TA > TC)
has also dem­ on­strated prom­ ise in reduc­ ing apo­pto­sis and
increas­ing neu­tro­phil mat­u­ra­tion in bone mar­row pro­gen­i­
tor cells.74 Inhibition of EIF6, a pro­tein removed by SBDS to
advance ribo­some for­ma­tion, is also an attrac­tive ther­a­peu­tic
tar­get for SDS treat­ment but still requires pre­clin­i­cal inves­ti­
ga­tion.
Summary
The diag­no­sis and man­age­ment of the neutropenia patient are
sig­nif­i­cantly aided by clas­si­fy­ing dis­or­ders into the pres­ence or
absence of ade­quate bone mar­row reserves. Patients with inef­
fi­cient myelopoiesis have a lower thresh­old to ini­ti­ate empiric
treat­ment for fever and may require mon­i­tor­ing for trans­for­
ma­tion into a mye­loid malig­nancy. Intact myelopoiesis patients
with ade­quate mar­row neu­tro­phil reserves are at much lower
risk of these com­pli­ca­tions despite a low periph­eral blood ANC.
G-CSF is com­monly used for poor neu­tro­phil pro­duc­tion and
may pro­vide a ben­e­fit in some sce­nar­ios of competent neu­
tro­phil pro­duc­tion. HSCT offers an oppor­tu­nity for cure in con­
gen­i­tal neutropenias, but ques­tions regard­ing the indi­ca­tions,
the impor­ tance of posttransplant chi­ me­rism, and the tech­
niques to reduce tox­ic­ity and malig­nant relapse in SDS limit
the opti­mi­za­tion of patient selec­tion, con­di­tion­ing approaches,
and out­comes. Gene editing may pro­vide an alter­na­tive option
to HSCT to cure the neutropenia, but the anal­o­gous ques­tion
to HSCT chi­me­rism of resid­ual uncor­rected hema­to­poi­etic
stem and pro­gen­i­tor cells resulting in con­tin­ued can­cer risk is
unknown. Continued col­lab­o­ra­tive research in clin­i­cal reg­is­tries
and basic sci­ence inves­ti­ga­tion are nec­es­sary to answer these
ques­tions and pro­vide addi­tional evi­dence for impor­tant clin­i­
cal deci­sions for neutropenic patients.
Conflict-of-inter­est dis­clo­sure
James A. Connelly: advi­sory board mem­ber: X4 Pharmaceuticals,
Horizon Therapeutics.
Kelly Walkovich: con­ sul­
tancy: Swed­ ish Orphan Biovitrum AB;
advi­sory board mem­ber: Horizon Therapeutics, Pharming; local
prin­ci­pal inves­ti­ga­tor: clin­i­cal trial on mavorixafor spon­sored by
X4 Pharmaceuticals.
Off-label drug use
James A. Connelly: nothing to disclose.
Kelly Walkovich: nothing to disclose.
Correspondence
James A. Connelly, Department of Pediatrics, Vanderbilt Universi-
ty Medical Center, 389 PRB, 2220 Pierce Ave, Nashville, TN 37232;
e-mail: james​­.a​­.connelly@vumc​­.org.
Prakash Singh Shekhawat
References
1. Reich D, Nalls MA, Kao WH, et al. Reduced neu­tro­phil count in peo­ple of
Afri­can descent is due to a reg­u­la­tory var­i­ant in the Duffy anti­gen recep­tor
for chemokines gene. PLoS Genet. 2009;5(1):e1000360.
2. Frater JL. How I inves­ti­gate neutropenia. Int J Lab Hematol. 2020;42(suppl
1):121-132.
3. Schaffner A, Fehr J. Granulocyte demargination by epi­neph­rine in eval­u­a­
tion of hypersplenic states. Scand J Haematol. 1981;27(4):225-230.
4. Papadaki HA, Eliopoulos AG, Kosteas T, et  al. Impaired gran­u­lo­cy­to­poi­
e­sis in patients with chronic idi­ o­pathic neutropenia is asso­ ci­ated with
increased apo­pto­sis of bone mar­row mye­loid pro­gen­i­tor cells. Blood.
2003;101(7):2591-2600.
5. Marins LR, Anizelli LB, Romanowski MD, Sarquis AL. How does pre­eclamp­
sia affect neo­na­tes? Highlights in the dis­ease’s immu­nity. J Matern Fetal
Neonatal Med. 2019;32(7):1205-1212.
6. Melvan JN, Bagby GJ, Welsh DA, Nelson S, Zhang P. Neonatal sep­sis and
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021
neu­tro­phil insufficiencies. Int Rev Immunol. 2010;29(3):315-348.
7. Bousfiha A, Jeddane L, Picard C, et  al. Human inborn errors of immu­
nity: 2019 update of the IUIS phe­no­typ­i­cal clas­si­fi­ca­tion. J Clin Immunol.
2020;40(1):66-81.
8. Larocque A, Hoffman RS. Levamisole in cocaine: unex­pected news from an
old acquain­tance. Clin Toxicol (Phila). 2012;50(4):231-241.
9. Marsh JC, Abboudi ZH, Gib­son FM, et al. Aplastic anae­mia fol­low­ing expo­
sure to 3,4-methylenedioxymethamphetamine (“ecstasy”). Br J Haematol.
1994;88(2):281-285.
10. Powars D. Aplastic ane­ mia sec­ ond­ary to glue sniffing. N Engl J Med.
1965;273(13):700-702.
11. Shi X, DeLucia AL, Bao J, Zhang P. Alcohol abuse and dis­or­der of granulo-
poiesis. Pharmacol Ther. 2019;198(June):206-219.
12. Yarandi SS, Griffith DP, Sharma R, Mohan A, Zhao VM, Ziegler TR. Optic
neu­rop­a­thy, mye­lop­a­thy, ane­mia, and neutropenia caused by acquired
cop­ per defi­ ciency after gas­ tric bypass sur­ gery. J Clin Gastroenterol.
2014;48(10):862-865.
13. Dale DC. How I man­ age chil­ dren with neutropenia. Br J Haematol.
2017;178(3):351-363.
14. Shaver AK, Walkovich K, Connelly JA. Recurrence of neo­na­tal lupus post-
cord blood trans­ plant for severe con­ gen­i­
tal neutropenia. Pediatrics.
2015;136(2):e535-e538.
15. Farruggia P, Fioredda F, Puccio G, et al. Autoimmune neutropenia of infancy:
data from the Ital­ ian Neutropenia Registry. Am J Hematol. 2015;90(12):
E221-E222.
16. Myers KC, Bolyard AA, Otto B, et al. Variable clin­i­cal pre­sen­ta­tion of Shwa-
chman-Diamond syn­drome: update from the North Amer­i­can Shwachman-
Diamond Syndrome Registry. J Pediatr. 2014;164(4):866-870.
17. Gib­son C, Berliner N. How we eval­u­ate and treat neutropenia in adults.
Blood. 2014;124(8):1251-1258, quiz 1378.
18. Michniacki TF, Connelly JA, Sturza J, et al. Neutropenia is an underrecog-
nized find­ing in pedi­at­ric pri­mary immu­no­de­fi­ciency dis­eases: an anal­y­sis
of the United States Immunodeficiency Network Registry. J Pediatr Hema-
tol Oncol. 2020;42(7):e601-e605.
19. Furutani E, Newburger PE, Shimamura A. Neutropenia in the age of genetic
test­ing: advances and chal­lenges. Am J Hematol. 2019;94(3):384-393.
20. Hao R, Saleh M, Liang T, et  al. The prev­a­lence of seri­ous bac­te­rial infec­
tions in neutropenic immu­no­com­pe­tent febrile chil­dren. Am J Emerg Med.
2021;45(July):1-6.
21. Melendez E, Harper MB. Risk of seri­ous bac­te­rial infec­tion in iso­lated and
unsus­pected neutropenia. Acad Emerg Med. 2010;17(2):163-167.
22. Bachur RG, Harper MB. Predictive model for seri­ ous bac­te­
rial infec­tions
among infants youn­ger than 3 months of age. Pediatrics. 2001;108(2):311-316.
23. Pascual C, Trenchs V, Hernández-Bou S, Català A, Valls AF, Luaces C. Out-
comes and infec­tious eti­­ol­o­gies of febrile neutropenia in non-immu­no­com­
pro­mised chil­dren who pres­ent in an emer­gency depart­ment. Eur J Clin
Microbiol Infect Dis. 2016;35(10):1667-1672.
24. Barg AA, Kozer E, Mordish Y, Lazarovitch T, Kventsel I, Goldman M. The risk
of seri­ous bac­te­rial infec­tion in neutropenic immu­no­com­pe­tent febrile
chil­dren. J Pediatr Hematol Oncol. 2015;37(6):e347-e351.
25. Bonadio WA, Stremski E, Shallow K. Clinical char­ac­ter­is­tics of chil­dren with
fever and tran­sient neutropenia who expe­ri­ence seri­ous bac­te­rial infec­
tions. Pediatr Emerg Care. 1989;5(3):163-165.
26. Wittmann O, Rimon A, Scolnik D, Glatstein M. Outcomes of immu­no­com­
pe­ tent chil­dren presenting with fever and neutropenia. J Emerg Med.
2018;54(3):315-319.
Diagnosis and deci­sion-mak­ing for neutropenia  |  501
27. Husain EH, Mullah-Ali A, Al-Sharidah S, Azab AF, Adekile A. Infectious eti­­ol­o­
gies of tran­sient neutropenia in pre­vi­ously healthy chil­dren. Pediatr Infect
Dis J. 2012;31(6):575-577.
28. Tschernin D, Fruchtman Y, Sergienko R, et al. The eti­o­logic, micro­bi­o­logic,
clin­i­cal and out­come char­ac­ter­is­tics of immu­no­com­pe­tent young chil­dren
<2 years of age hos­pi­tal­ized with acute neutropenia. Pediatr Neonatol.
2021;62(1):26-35.
29. Leibovitz E, Kapelushnik J, Alsanaa S, et  al. Comparison of the eti­o­logic,
micro­bi­o­logic, clin­i­cal and out­come char­ac­ter­is­tics of febrile vs. non-fe-
brile neutropenia in hos­pi­tal­ized immu­no­com­pe­tent chil­dren. Eur J Clin
Microbiol Infect Dis. 2020;39(12):2415-2426.
30. Alexandropoulou O, Kossiva L, Haliotis F, et  al. Transient neutropenia in
chil­dren with febrile ill­ness and asso­ci­ated infec­tious agents: 2 years’
fol­low-up. Eur J Pediatr. 2013;172(6):811-819.
31. Karavanaki K, Polychronopoulou S, Giannaki M, et al. Transient and chronic
neutropenias detected in chil­dren with dif­fer­ent viral and bac­te­rial infec­
tions. Acta Paediatr. 2006;95(5):565-572.
32. Palmblad J, Dufour C, Papadaki HA. How we diag­ nose neutropenia in
the adult and elderly patient. Haematologica. 2014;99(7):1130-1133.
33. Kirk SE, Grimes AB, Shelke S, Despotovic JM, Powers JM. The cost of a
“benign” con­di­tion: healthcare uti­li­za­tion and infec­tious out­comes in
young chil­dren with pri­mary auto­im­mune neutropenia. Pediatr Blood Can-
cer. 2020;67(4):e28146.
34. Fioredda F, Calvillo M, Burlando O, et al. Infectious com­pli­ca­tions in chil­
dren with severe con­gen­i­tal, auto­im­mune or idi­o­pathic neutropenia: a ret­
ro­spec­tive study from the Ital­ian Neutropenia Registry. Pediatr Infect Dis J.
2013;32(4):410-412.
35. Newburger PE. Autoimmune and other acquired neutropenias. Hematol-
ogy Am Soc Hematol Educ Program. 2016;2016(1):38-42.
36. Sicre de Fontbrune F, Moignet A, Beaupain B, et al; French Severe Chronic
Neutropenia Registry. Severe chronic pri­mary neutropenia in adults: report
on a series of 108 patients. Blood. 2015;126(14):1643-1650.
37. Farruggia P, Puccio G, Fioredda F, et al. Autoimmune neutropenia of child­
hood sec­ ond­ary to other auto­ im­ mune dis­or­
ders: data from the Ital­ ian
Neutropenia Registry. Am J Hematol. 2017;92(9):E546-E549.
38. Welte K, Zeidler C, Dale DC. Severe con­gen­i­tal neutropenia. Semin Hema-
tol. 2006;43(3):189-195.
39. Fioredda F, Calvillo M, Bonanomi S, et al; Neutropenia Committee of the
Marrow Failure Syndrome Group of the AIEOP (Associazione Italiana Ema-
to-Oncologia Pediatrica). Congenital and acquired neutropenias con­sen­
sus guide­lines on ther­apy and fol­low-up in child­hood from the Neutropenia
Committee of the Marrow Failure Syndrome Group of the AIEOP (Associazi-
one Italiana Emato-Oncologia Pediatrica). Am J Hematol. 2012;87(2):238-
243.
40. Link DC. Mechanisms of leu­ke­mic trans­for­ma­tion in con­gen­i­tal neutropenia.
Curr Opin Hematol. 2019;26(1):34-40.
41. Koch C, Samareh B, Morishima T, et al. GM-CSF treat­ment is not effec­tive
in con­gen­i­tal neutropenia patients due to its inabil­ity to acti­vate NAMPT
sig­nal­ing. Ann Hematol. 2017;96(3):345-353.
42. Dale DC, Bolyard AA. An update on the diag­no­sis and treat­ment of chronic
idi­o­pathic neutropenia. Curr Opin Hematol. 2017;24(1):46-53.
43. Dale DC. How I diag­ nose and treat neutropenia. Curr Opin Hematol.
2016;23(1):1-4.
44. Farruggia P, Dufour C. Diagnosis and man­ age­ment of pri­ mary auto­im­
mune neutropenia in chil­dren: insights for cli­ni­cians. Ther Adv Hematol.
2015;6(1):15-24.
45. Dungarwalla M, Marsh JC, Tooze JA, et al. Lack of clin­i­cal effi­cacy of ritux-
imab in the treat­ment of auto­im­mune neutropenia and pure red cell aplasia:
impli­ca­tions for their path­o­phys­i­­ol­ogy. Ann Hematol. 2007;86(3):191-197.
46. Lamy T, Loughran TP Jr. How I treat LGL leu­ke­mia. Blood. 2011;117(10):2764-
2774.
47. Zeidler C, Welte K, Barak Y, et al. Stem cell trans­plan­ta­tion in patients with
severe con­gen­i­tal neutropenia with­out evi­dence of leu­ke­mic trans­for­ma­
tion. Blood. 2000;95(4):1195-1198.
48. Bardelli D, Dander E, Bugarin C, et  al. Mesenchymal stro­mal cells from
Shwachman-Diamond syn­ drome patients fail to rec­ re­
ate a bone mar­
row niche in vivo and exhibit impaired angio­ gen­ e­sis. Br J Haematol.
2018;182(1):114-124.
49. Myers KC, Furutani E, Weller E, et  al. Clinical fea­tures and out­comes of
patients with Shwachman-Diamond syn­drome and myelodysplastic syn­
drome or acute mye­loid leu­kae­mia: a multicentre, ret­ro­spec­tive, cohort
study. Lancet Haematol. 2020;7(3):e238-e246.
502  |  Hematology 2021  |  ASH Education Program
50. Choi SW, Boxer LA, Pulsipher MA, et al. Stem cell trans­plan­ta­tion in patients
with severe con­gen­i­tal neutropenia with evi­dence of leu­ke­mic trans­for­ma­
tion. Bone Marrow Transplant. 2005;35(5):473-477.
51. Makaryan V, Zeidler C, Bolyard AA, et  al. The diver­sity of muta­tions and
clin­i­cal out­comes for ELANE-asso­ci­ated neutropenia. Curr Opin Hematol.
2015;22(1):3-11.
52. Dale DC, Makaryan V, Kelly ML, et al. Termination and frame­shift muta­tions
in ELANE are asso­ci­ated with adverse out­comes in patients with severe
chronic neutropenia. Blood. 2016;128(22):1326.
53. Rotulo GA, Beaupain B, Rialland F, et al. HSCT may lower leu­ke­mia risk in
ELANE neutropenia: a before-after study from the French Severe Congeni-
tal Neutropenia Registry. Bone Marrow Transplant. 2020;55(8):1614-1622.
54. Rosenberg PS, Zeidler C, Bolyard AA, et al. Stable long-term risk of leu­kae­
mia in patients with severe con­gen­i­tal neutropenia maintained on G-CSF
ther­apy. Br J Haematol. 2010;150(2):196-199.
55. Connelly JA, Choi SW, Levine JE. Hematopoietic stem cell trans­plan­ta­tion
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021
for severe con­gen­i­tal neutropenia. Curr Opin Hematol. 2012;19(1):44-51.
56. Fioredda F, Iacobelli S, van Biezen A, et al; Severe Aplastic Anemia the
Inborn Error; Pediatric Disease Working Parties of the Euro­pean Society
for Blood and Bone Marrow Transplantation and Stem Cell Transplant for
Immunodeficiencies in Europe. Stem cell trans­plan­ta­tion in severe con­gen­
i­tal neutropenia: an anal­y­sis from the Euro­pean Society for Blood and Mar-
row Transplantation. Blood. 2015;126(16):1885-1892, quiz 1970.
57. Zeidler C, Nickel A, Skyora KW, Welte K. Improved out­come of stem cell
trans­plan­ta­tion for severe chronic neutropenia with or with­out sec­ond­ary
leu­ke­mia: a long-term anal­y­sis of Euro­pean data for more than 25 years by
the SCNIR. Blood. 2013;122(21):3347.
58. Choi EJ, Lee JH, Lee JH, et al. Treatment and clin­i­cal out­comes of patients
relaps­ing after allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion for myelodys-
plastic syn­drome. Blood Res. 2018;53(4):288-293.
59. Bitan M, He W, Zhang MJ, et al. Transplantation for chil­dren with acute mye­
loid leu­ke­mia: a com­par­i­son of out­comes with reduced inten­sity and mye-
loablative reg­i­mens. Blood. 2014;123(10):1615-1620.
60. Rautenberg C, Germing U, Haas R, Kobbe G, Schroeder T. Relapse of acute
mye­loid leu­ke­mia after allo­ge­neic stem cell trans­plan­ta­tion: pre­ven­tion,
detec­tion, and treat­ment. Int J Mol Sci. 2019;20(1):228.
61. Connelly JA, Mody RJ, Wu YM, et al. Identification of novel MECOM gene
fusion and per­son­al­ized ther­a­peu­tic tar­gets through inte­gra­tive clin­i­cal
sequenc­ing in sec­ond­ary acute mye­loid leu­ke­mia in a patient with severe
con­gen­i­tal neutropenia: a case report and lit­er­a­ture review. Cold Spring
Harb Mol Case Stud. 2018;4(2):a002204.
62. Cesaro S, Pillon M, Sauer M, et  al. Long-term out­come after allo­ge­neic
hema­to­poi­etic stem cell trans­plan­ta­tion for Shwachman-Diamond syn­
drome: a ret­ ro­
spec­tive anal­ y­
sis and a review of the lit­ er­
a­
ture by the
Severe Aplastic Anemia Working Party of the Euro­pean Society for Blood
and Marrow Transplantation (SAAWP-EBMT). Bone Marrow Transplant.
2020;55(9):1796-1809.
63. Burroughs LM, Shimamura A, Talano JA, et al. Allogeneic hema­to­poi­etic cell
trans­plan­ta­tion using treosulfan-based con­di­tion­ing for treat­ment of mar­
row fail­ure dis­or­ders. Biol Blood Marrow Transplant. 2017;23(10):1669-1677.
64. Bhatla D, Davies SM, Shenoy S, et  al. Reduced-inten­sity con­di­tion­ing is
effec­tive and safe for trans­plan­ta­tion of patients with Shwachman-Dia-
mond syn­drome. Bone Marrow Transplant. 2008;42(3):159-165.
65. Myers K, Hebert K, Antin J, et  al. Hematopoietic stem cell trans­plan­ta­
tion for Shwachman-Diamond syn­drome. Biol Blood Marrow Transplant.
2020;26(8):1446-1451.
66. Hientz K, Mohr A, Bhakta-Guha D, Efferth T. The role of p53 in can­cer drug
resis­tance and targeted che­mo­ther­apy. Oncotarget. 2017;8(5):8921-8946.
67. Skokowa J, Lan D, Thakur BK, et  al. NAMPT is essen­tial for the G-CSF-
induced mye­loid dif­fer­en­ti­a­tion via a NAD(+)-sirtuin-1-depen­dent path­way.
Nat Med. 2009;15(2):151-158.
68. Deordieva E, Shvets O, Voronin K, et al. Nicotinamide (vita­min B3) treat­
ment improves response to G-CSF in severe con­ gen­ i­
tal neutropenia
patients. Br J Haematol. 2021;192(4):788-792.
69. Nanua S, Murakami M, Xia J, et  al. Activation of the unfolded pro­ tein
response is asso­ci­ated with impaired granulopoiesis in trans­genic mice
expressing mutant Elane. Blood. 2011;117(13):3539-3547.
70. Makaryan V, Kelley ML, Fletcher B, Bolyard AA, Aprikyan AA, Dale DC. Elas-
tase inhib­i­tors as poten­tial ther­a­pies for ELANE-asso­ci­ated neutropenia.
J Leukoc Biol. 2017;102(4):1143-1151.
71. Ritter MU, Secker B, Klimiankou M, et al. Efficient cor­rec­tion of ELANE muta­
tions in pri­mary HSPCs of severe con­gen­i­tal neutropenia patients using
 CRISPR/Cas9 and rAVV6 HDR repair tem­ plates. Blood. 2019;134(suppl
1):1036.
72. Nasri M, Ritter M, Mir P, et  al. CRISPR/Cas9-medi­ated ELANE knock­out
enables neu­tro­philic mat­u­ra­tion of pri­mary hema­to­poi­etic stem and pro­
gen­i­tor cells and induced plu­rip­o­tent stem cells of severe con­gen­i­tal neu-
tropenia patients. Haematologica. 2020;105(3):598-609.
73. Rao S, Yao Y, Soares de Brito J, et al. Dissecting ELANE neutropenia path­
o­ge­nic­ity by human HSC gene editing. Cell Stem Cell. 2021;28(5):833-845.
e5845e5.
74. Bezzerri V, Bardelli D, Morini J, et  al. Ataluren-driven res­ to­
ra­
tion of
­Shwachman-Bodian-Diamond syn­drome pro­tein func­tion in Shwachman-
Diamond syn­drome bone mar­row cells. Am J Hematol. 2018;93(4):527-536.
75. Klimiankou M, Uenalan M, Kandabarau S, et al. Ultra-sen­si­tive CSF3R deep
sequenc­ing in patients with severe con­gen­i­tal neutropenia. Front Immu-
nol. 2019;10(Feb­ru­ary):116.
76. Klimiankou M, Mellor-Heineke S, Klimenkova O, et al. Two cases of cyclic
neutropenia with acquired CSF3R muta­ tions, with 1 devel­ op­ing AML.
Blood. 2016;127(21):2638-2641.
77. Carlsson G, Fasth A, Berglöf E, et al. Incidence of severe con­gen­i­tal neu-
tropenia in Sweden and risk of evo­lu­tion to myelodysplastic syn­drome/
leu­kae­mia. Br J Haematol. 2012;158(3):363-369.
78. Desplantes C, Fremond ML, Beaupain B, et al. Clinical spec­trum and long-
term fol­low-up of 14 cases with G6PC3 muta­tions from the French Severe
Congenital Neutropenia Registry. Orphanet J Rare Dis. 2014;9(10 Decem­
ber):183.
79. Ancliff PJ, Blundell MP, Cory GO, et al. Two novel acti­vat­ing muta­tions in the
Wiskott-Aldrich syn­drome pro­tein result in con­gen­i­tal neutropenia. Blood.
2006;108(7):2182-2189.
80. Dale DC, Bolyard AA, Marrero T, et  al. Neutropenia in gly­ co­gen stor­
age dis­ease Ib: out­comes for patients treated with granulocyte col­ony-
stim­u­lat­ing fac­tor. Curr Opin Hematol. 2019;26(1):16-21.
81. Li AM, Thyagu S, Maze D, et al. Prolonged granulocyte col­ony stim­u­lat­ing
fac­tor use in gly­co­gen stor­age dis­ease type 1b asso­ci­ated with acute mye­
loid leu­ke­mia and with short­ened telo­mere length. Pediatr Hematol Oncol.
2018;35(1):45-51.
Prakash Singh Shekhawat
82. Negri G, Crescenzi B, Colombo EA, et al. Expanding the role of the splic­ing
USB1 gene from poikiloderma with neutropenia to acquired mye­loid neo­
plasms. Br J Haematol. 2015;171(4):557-565.
83. Donadieu J, Fenneteau O, Beaupain B, et al; Associated Investigators of
the French Severe Chronic Neutropenia Registry*. Classification of and
risk fac­tors for hema­ to­
logic com­ pli­
ca­
tions in a French national cohort
of 102 patients with Shwachman-Diamond syn­ drome. Haematologica.
2012;97(9):1312-1319.
84. Cada M, Segbefia CI, Klaassen R, et  al. The impact of cat­e­gory, cytopa-
thology and cyto­ ge­net­ics on devel­ op­
ment and pro­ gres­
sion of clonal
and malig­nant mye­loid trans­for­ma­tion in inherited bone mar­row fail­ure
­syn­dromes. Haematologica. 2015;100(5):633-642.
85. Kennedy AL, Myers KC, Bowman J, et al. Distinct genetic path­ways define
pre-malig­nant ver­sus com­pen­sa­tory clonal hema­to­poi­e­sis in Shwachman-
Diamond syn­drome. Nat Commun. 2021;12(1):1334.
86. Spinner MA, Sanchez LA, Hsu AP, et al. GATA2 defi­ciency: a pro­tean dis­or­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/492/1851746/492connelly.pdf by guest on 13 December 2021
der of hema­to­poi­e­sis, lym­phat­ics, and immu­nity. Blood. 2014;123(6):809-
821.
87. Wlodarski MW, Hirabayashi S, Pastor V, et al; Euro­pean Working Groups
of Myelodysplastic Syndromes. Prevalence, clin­i­cal char­ac­ter­is­tics, and
prog­no­sis of GATA2-related myelodysplastic syn­dromes in chil­dren and
ado­les­cents. Blood. 2016;127(11):1387-1397, quiz 1518.
88. Bluteau O, Sebert M, Leblanc T, et al. A land­scape of germ line muta­tions in
a cohort of inherited bone mar­row fail­ure patients. Blood. 2018;131(7):717-
732.
89. Wong JC, Bryant V, Lamprecht T, et al. Germline SAMD9 and SAMD9L muta­
tions are asso­ci­ated with exten­sive genetic evo­lu­tion and diverse hema­to­
logic out­comes. JCI Insight. 2018;3(14).
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000284
Diagnosis and deci­sion-mak­ing for neutropenia  |  503
 NEUTROPENIA: DOING MORE WITH LESS

Understanding neutropenia secondary to intrinsic

or iatrogenic immune dysregulation

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/504/1851486/504walkovich.pdf by guest on 13 December 2021

Kelly Walkovich1 and James A. Connelly2
Department of Pediatrics, University of Michigan, Ann Arbor, MI; and 2Department of Pediatrics, Vanderbilt University Medical Center, Nashville, TN
1

As a key member of the innate and adaptive immune response, neutrophils provide insights into the hematopoietic and
inflammatory manifestations of inborn errors of immunity (IEI) and the consequences of immunotherapy. The facile recog-
nition of IEI presenting with neutropenia provides an avenue for hematologists to facilitate early diagnosis and expedite
biologically rationale care. Moreover, enhancing the understanding of the molecular mechanisms driving neutropenia
in IEI—decreased bone marrow reserves, diminished egress from the bone marrow, and decreased survival—offers an
opportunity to further dissect the pathophysiology driving neutropenia secondary to iatrogenic immune dysregulation,
eg, immune checkpoint inhibitors and chimeric antigen receptor T-cell therapy.

LEARNING OBJECTIVES
• Identify neutropenia as a potential first sign of underlying inborn errors of immunity and utilize a targeted history
and physical exam to assess the need for additional evaluation
• Describe peripheral blood smear and bone marrow findings associated with IEI
• Discuss the benefits of early IEI diagnosis in relation to patient­specific, biologically rationale therapeutic
decision­making
• Explain the pathophysiology of iatrogenic neutropenia secondary to immunotherapy
• Illustrate potential correlations between neutropenia secondary to immunotherapy and known immune disorders

Introduction IEI­mediated neutropenia provide insights into the mecha­

While neutrophils have long been recognized as critical medi­
ators of first­line host defense and inflammation­induced
injury, emerging evidence suggests a more nuanced and
extensive role for neutrophils in both the innate and adap­
tive immune response.1 In parallel, neutropenia, defined as an
absolute neutrophil count (ANC) <1500 cells per microliter for
patients over 1 year of age, is increasingly appreciated as a
clinical feature of inborn errors of immunity (IEI) and immune
dysregulation.2 Patients with IEI also frequently present with
other hematologic clinical features, eg, cytopenias, bone
marrow failure, splenomegaly, and malignancy.3,4
Neutropenia in IEI and other immune dysregulation
states can be a consequence of infection or a drug effect;
however, for several disorders neutropenia is the result of an
intrinsic defect of myelopoiesis or an outcome of an improper
immune response. Understanding the pathophysiology driv­
ing neutropenia in IEI and immune dysregulation provides
an opportunity to intervene in a biologically rational man­
ner. Hematologists adept at distinguishing neutropenia sec­
ondary to IEI or immune dysregulation from other etiologies
can facilitate early diagnosis. Additionally, investigations into
nism and management of neutropenia secondary to the
immune­related therapies used by oncologists that lever­
age the same molecular pathways to enhance immunologic
antitumor activity.
Distinguishing patients with neutropenia related
to IEI from other etiologies
Differentiating neutropenia associated with IEI from more
common postinfectious, drug­related, nutritional, or primary
autoimmune etiologies can be challenging, particularly as
the number of IEI are rapidly expanding and may present
with variable or incomplete clinical features.4 The history
should be inclusive of screening for phagocyte disorders
and humoral and combined immunodeficiency as well as
specific questions to probe for bone marrow failure or
immune dysregulation syndromes (see Table 1). Careful
attention to the exam, particularly the oral cavity (eg, oral
ulcers, gingivitis, abnormal dentition, periodontal disease),
integumentary (eg, eczema, hemangiomas, hypopigmenta­
tion, warts, prominent superficial veins, vasculitis, etc), and
hematologic/lymphatic systems, can provide clues to an

504 | Hematology 2021 | ASH Education Program

Table 1. Targeted screen­ing by his­tory for IEI in patients presenting with neutropenia

Targeted screen­ing by his­tory for IEI Additional stud­ies to con­sider

Phagocyte dis­or­ders
Recurrent fevers Serial CBCPDs
Persistent or recur­rent oral ulcers DHR flow
Bleeding and/or reced­ing gums Genetic sequenc­ing
Gingivitis Echocardiogram
Dental anom­a­lies and/or den­tal infec­tions Renal ultra­sound
Frequent or deep-seated skin infec­tions Skeletal sur­vey
Perirectal infec­tions or inflam­ma­tory bowel dis­ease Genetic sequenc­ing
Developmental delay
Cardiac, renal, or skel­e­tal anom­a­lies
Personal or fam­ily his­tory of MDS/AML

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/504/1851486/504walkovich.pdf by guest on 13 December 2021

Bone mar­row fail­ure syn­dromes
Personal or fam­ily his­tory of bone mar­row fail­ure, MDS, AML, sar­co­mas, fat mal­ab­sorp­tion, Pancreatic amy­lase
warts, lymphedema, pul­mo­nary or hepatic fibro­sis, early graying of hair, nail Fecal elas­tase
abnor­mal­i­ties, stroke, vas­cu­li­tis Pancreatic ultra­sound
Telomere length
Chromosomal break­age
ADA2 level
Quantitative immu­no­glob­u­lins
T/B/NK-cell enu­mer­a­tion
Genetic sequenc­ing
Humoral defi­ciency
Four or more new ear infec­tions in 1 year Quantitative immu­no­glob­u­lins
Two or more sinus infec­tions or pneu­mo­nias requir­ing anti­bi­ot­ics within the past 1 year T/B/NK-cell enu­mer­a­tion
Requirement for prolonged course or mul­ti­ple courses of oral anti­bi­ot­ics to clear “sim­ple” Vaccine anti­body responses
infec­tions BTK flow cytom­e­try
Need for intra­ve­nous anti­bi­ot­ics and/or hos­pi­tal­i­za­tions for infec­tion treat­ment Genetic sequenc­ing
Combined immu­no­de­fi­ciency
Abnormal TREC new­born screen Quantitative immu­no­glob­u­lins
Failure to thrive T/B/NK-cell enu­mer­a­tion
Developmental delay Vaccine anti­body responses
Persistent oral thrush or fun­gal skin infec­tions Lymphocyte mito­gen pro­lif­er­a­tion
Two or more deep-seated infec­tions Lymphocyte anti­gen pro­lif­er­a­tion
Atypical or oppor­tu­nis­tic infec­tions TRECs
CD40L flow cytom­e­try
STAT1 func­tional screen­ing
Toll-like recep­tor-medi­ated sig­nal­ing screen­ing
Genetic sequenc­ing
Immune dysregulation
Personal or fam­ily his­tory of inflam­ma­tory bowel dis­ease, par­tic­u­larly with onset prior to Ferritin
age 6, auto­im­mune cytopenias, pro­gres­sive neu­ro­log­i­cal dys­func­tion, hemophagocytic TCR alpha-beta CD3+CD4−CD8− T cells
lymphohistiocytosis Genetic sequenc­ing
Evidence of non­ma­lig­nant lymphoproliferation with chronic lymph­ade­nop­a­thy or
spleno­meg­aly
Differences in hair or skin pig­men­ta­tion, par­tic­u­larly in rela­tion to imme­di­ate fam­ily
mem­bers
AML, acute mye­loid leu­ke­mia; CBCPD, com­plete blood count with plate­lets and dif­fer­en­tial; DHR, dihydrorhodamine; MDS, myelodysplastic
syn­drome; TCR, T-cell recep­tor; TREC, T-cell recep­tor exci­sion cir­cle.

under­ly­ing IEI. Notably, the hema­tol­o­gist can also raise con­cern
for an under­ly­ing IEI by review of the periph­eral blood smear and,
if com­pleted, bone mar­row biopsy for signs sug­ges­tive of dis­
rupted neu­tro­phil devel­op­ment/func­tion or mar­row stress/infil­
tra­tion (see Table 2). Functional lab­o­ra­tory eval­u­a­tion tai­lored
to the patient pre­sen­ta­tion can pro­vide sup­port­ive evi­dence of
an IEI. The Clinical Immunology Society main­tains a DLI Labo-
ratory Directory that is pub­licly acces­si­ble and can assist with
the iden­ti­fi­ca­tion of labs cur­rently pro­vid­ing Clinical Laboratory
Improvement Amendments-cer­ti­fied immune test­ing (Accessed
30 Sep­tem­ber 2021. https:​­/​­/clinimmsoc​­.org​­/CIS​­/Resources​­/DLI​
­-Lab​­-Directory​­.htm). A defin­i­tive diag­no­sis gen­er­ally requires the
iden­ti­fi­ca­tion of path­o­genic var­i­ants via genetic sequenc­ing.
Prakash Singh Shekhawat
With recent advance­ments in sequenc­ing meth­ods and acces­
si­bil­ity, genetic test­ing is increas­ingly pro­moted as a key tool
in the defin­i­tive diag­no­sis of IEI.5-7 A molec­u­lar diag­no­sis is valu­
able for unequivocally establishing the IEI diag­no­sis, per­mit­ting
appro­ pri­
ate genetic coun­ sel­ing to iden­ tify other at-risk fam­
ily mem­bers and future preg­nan­cies, facil­i­tat­ing geno­type/
phe­no­type cor­re­la­tions to inform patient-spe­cific clin­i­cal expec­
ta­tions, and iden­ ti­
fy­
ing patients most likely to ben­ e­fit from
gene-spe­cific ther­a­pies.
Pathophysiology of neutropenia in IEI
While the gen­eral mech­a­nisms for neutropenia in IEI have been
pro­posed, ie, dimin­ished sur­vival, flawed devel­op­ment, phys­i­cal
Neutropenia and immune dysregulation  |  505
Table 2. Peripheral blood smear and bone mar­row find­ings in IEI asso­ci­ated with neutropenia

Hematopathologic find­ing IEI to con­sider in dif­fer­en­tial Additional com­ments

Absent or decreased neu­tro­phil gran­ules
Atypical mega­kar­yo­cytes
Giant cyto­plas­mic gran­ules
GFI1-related neutropenia
JAGN1-related neutropenia
Specific gran­ule defi­ciency
GATA2 haploinsufficiency
Chediak-Higashi syn­drome
In spe­cific gran­ule defi­ciency, neu­tro­phils with bilobed nuclei are
com­monly observed.
Megakaryocytes may be small or mono­nu­clear or may have sep­a­
rated nuclear lobes. Often asso­ci­ated with mar­row fibro­sis.21
Giant azurophilic gran­ules within the lyso­somes are con­sid­ered
patho­gno­monic for the dis­or­der and are eas­ily vis­i­ble in neu­tro­
phils, eosin­o­phils, and other granulocytes.51
Additional find­ings in the mar­row include exten­sive vacuolization,
aci­do­philic inclu­sion (esp in promyelocytes), spe­cific gran­ules
eas­ily rec­og­nized as early as the mye­lo­cyte stage.9

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/504/1851486/504walkovich.pdf by guest on 13 December 2021

Myelofibrosis DADA215 DADA2 is also asso­ci­ated with lym­phoid aggre­gates in the mar­row.
GATA2 haploinsufficiency21
VPS45-related neutropenia52
Myelokathexis WHIM syn­drome23 Neutrophils are also often hypersegmented with thin, long isthmi,
G6PC3-related neutropenia53 pyknotic nuclei, and cyto­plas­mic vac­u­oles in WHIM syn­drome.
Neutrophil nuclei her­ni­a­tion WDR1 defi­ciency54 In addi­tion to her­ni­a­tion of the nuclear lobes, neu­tro­phils also have
agranular regions within the cyto­sol.
PHA NBAS defi­ciency Pseudo-PHA is asso­ci­ated with var­i­ous med­i­ca­tions (eg, immu­no­
sup­pres­sive agents, anti­mi­cro­bi­als, growth fac­tors). The pseudo-
PHA is gen­er­ally revers­ible with med­i­ca­tion ces­sa­tion.54 PHA also
seen in asso­ci­a­tion with MDS.
Promyelocyte arrest Severe, per­ma­nent arrest:55 Copper defi­ciency can also result in mat­u­ra­tion arrest; revers­ible
ELANE-related neutropenia (severe with cop­per admin­is­tra­tion.
con­gen­i­tal phe­no­type) Promyelocytic AML can mimic the promyelocytic arrest asso­ci­ated
HAX-1-related neutropenia with IEI but is eas­ily dif­fer­en­ti­ated by addi­tional flow and/or
CSF3R-related neutropenia cyto­ge­netic stud­ies.
G6PC3-related neutropenia
WAS-related neutropenia
JAGN1-related neutropenia56
Mild, inter­mit­tent arrest:
GFI1-related neutropenia
Neutropenia asso­ci­ated with
poikiloderma
AML, acute mye­loid leu­ke­mia; MDS, myelodysplastic syn­drome; PHA, Pelger-Huet anom­aly.

replace­ment, or fail­ure to egress from the mar­row,8 defin­i­tive
mech­an ­ isms for neutropenia in the major­ity of IEI out­side of
the con­gen­i­tal dis­or­ders of iso­lated neutropenia (eg, ELANE-
related neutropenia) have not been established. As such, this
arti­cle focuses on IEI with well-established asso­ci­a­tions with
neutropenia accom­pa­nied by the addi­tional immune def­i­cits
most likely to be encoun­tered by hema­tol­o­gists (Table 3).
Chediak-Higashi syn­drome
Chediak-Higashi syn­drome is an auto­so­mal reces­sive dis­or­der
sec­ond­ary to path­o­genic var­i­ants in CHS1/LYST. Patients clas­si­
cally pres­ent with oculocutaneous albi­nism, recur­rent pyo­genic
infec­tions, bleed­ing diath­e­sis, and pro­gres­sive neu­ro­log­i­cal
dys­func­tion and are at high risk of devel­op­ing hemophagocytic
lymphohistiocytosis (HLH) sec­ond­ary to defec­tive lyso­somal traf­
fick­ing. Neutropenia is com­monly observed and is attrib­uted to
abnor­mal bone mar­row reserves and defec­tive granulocyte mobi­
li­za­tion from the mar­row space compounded by intramedullary
granulocyte destruc­tion.9 While not explic­itly stud­ied, it is likely that
infec­tion-related mar­row sup­pres­sion and/or accel­er­ated immune
clear­ance also con­trib­utes to the neutropenia. The periph­eral
smear clas­si­cally dem­on­strates giant azurophilic gran­ules within
granulocyte lyso­somes. The treat­ment of neutropenia is pri­mar­ily
sup­port­ive with anti­mi­cro­bi­als and granulocyte col­ony-stim­u­lat­ing
fac­tor (G-CSF), although hema­to­poi­etic stem cell trans­plant (HSCT)
can resolve the hema­to­poi­etic aspects of the dis­or­der.
Common var­i­able immu­no­de­fi­ciency
Common var­i­able immu­no­de­fi­ciency (CVID) is a het­ero­ge­neous
dis­or­der char­ac­ter­ized by impaired B-cell dif­fer­en­ti­a­tion and
defec­tive immu­no­glob­u­lin (Ig) pro­duc­tion. Autoimmune cytope­
nias are a com­mon clin­i­cal fea­ture of CVID, with immune throm­
bo­cy­to­pe­nia and auto­im­mune hemo­lytic ane­mia more prev­a­lent
than auto­im­mune neutropenia.10 Cytopenias, includ­ing iso­lated
neutropenia, fre­quently pre­cede the diag­no­sis of CVID and her­
ald a worse out­come.11,12 Autoimmune clear­ance sec­ond­ary to
self-reac­tive IgG to hema­to­poi­etic anti­gens is the pri­mary sus­
pected eti­­ ol­
ogy of neutropenia, although hypersplenism and
drug-related (eg, sec­ond­ary to rituximab) and infec­tion-related
neutropenia is documented. Patients may be treated with cor­
ti­co­ste­roids and/or G-CSF. Splenectomy and rituximab are not
use­ful,11 but rapamycin is emerg­ing as an effec­tive ther­apy.13

506  |  Hematology 2021  |  ASH Education Program

Table 3. Characteristics of neutropenia in IEI com­monly encoun­tered by hema­tol­o­gists
Proposed mech­a­nism
IEI Frequency of neutropenia of neutropenia
Chediak-Higashi syn­drome Common 57
-Accelerated granulocyte
turn­over sec­ond­
ary to intramedullary
granulocyte destruc­
tion compounded by
hypersplenism.9
-Abnormal bone mar­row
reserves with defec­tive
granulocyte mobi­li­za­
tion from the mar­row
space.9
CVID 1%-8.4%10,12 -Autoimmune clear­ance as
supported by the iden­ti­
fi­ca­tion of antineutrophil
antibodies and evi­dence
of hyper­plas­tic yet inef­
fi­cient ger­mi­nal cen­ter
responses.11,59
-Hypersplenism
-Postinfectious
-Drug-related
DADA2 7%-15%14,60 -Decreased ADA2 pro­tein
func­tion in severely
del­e­te­ri­ous var­i­ants is
pro­posed to lead to
decreased mar­row
pro­duc­tion.17
-Immune-medi­ated
destruc­tion is also
hypoth­e­sized.
Hyper IgM syn­drome 41%18 -Decreased CD16 expres­
sion and dysregulated
transcriptome resulting in
impaired dif­fer­en­ti­a­tion.62
-Disrupted cyto­kine or
growth fac­tor sup­port in
the bone mar­row.63
GATA2 haploinsufficiency 47%21 -Reduction of the prim­i­tive
HSC pool20
WHIM syn­drome Near uni­ver­sal -Diminished egress from
the bone mar­row
sec­ond­ary to gain of
func­tion muta­tions in
CXCR423
XLA 10%-26%64-66 -Decreased bone mar­row
pre­cur­sors25
-Decreased mat­u­ra­tion
of mye­loid pre­cur­sors
sec­ond­ary to changes in
BTK-related sig­nal trans­
duc­tion25
-Decreased cyto­
kine/chemokine pro­duc­
tion from mono­cytes,
esp decreased IL-1825
AML, acute mye­loid leu­ke­mia; MDS, myelodysplastic syn­drome.
Prakash Singh Shekhawat
Neutropenia-directed
ther­apy Comments
G-CSF -Phagocyte intra­cel­lu­lar
Antimicrobials kill­ing and NK cyto­tox­ic­ity
HSCT58 is impaired.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/504/1851486/504walkovich.pdf by guest on 13 December 2021
Corticosteroids -Neutropenia does not improve
G-CSF with IgG replace­ment
Rapamycin -Neutropenia is asso­ci­ated
with increased infec­
tions, increased poly­clonal
lymphoproliferation, and auto­
im­mu­nity.12
-Increased mor­tal­ity rate asso­ci­
ated with neutropenia in CVID
patients.11,12
Corticosteroids -ADA2 is highly expressed in
Rituximab mye­loid cells and pro­duced
Anti-TNF agents by acti­vated mac­ro­phages,
HSCT61 mono­cytes, and den­dritic cells
when stim­u­lated by an inflam­
ma­tory response.14
G-CSF -Neutropenia may be chronic
or inter­mit­tent.
-Bone mar­row eval­u­a­tion may
dem­on­strate mat­u­ra­tion
arrest.62
G-CSF -Mild chronic neutropenia may
HSCT be the first man­i­fes­ta­tion of
dis­ease with other clin­i­cal fea­
tures, eg, monocytopenia or
MDS/AML presenting later.
-Maturation of neu­tro­phils in the
bone mar­row is gen­er­ally pre­
served.20
G-CSF -Bone mar­row is hypercellular
CXCR4 with full mat­u­ra­tion and clas­sic
antag­o­nists pyknotic nuclei.
G-CSF -BTK expressed in mye­loid and
IVIG B-cell dif­fer­en­ti­a­tion (lim­ited
to hema­to­poi­etic cells).
-Neutropenia may be a
presenting sign of XLA.
-Neutropenia gen­er­ally resolves
with ini­ti­a­tion of IVIG, allowing
G-CSF with­drawal.
-Neutropenia gen­er­ally
documented only in con­junc­
tion with an active infec­tion.
Neutropenia and immune dysregulation  |  507
Deficiency of aden­o­sine deam­i­nase 2
A defi­ ciency of aden­ o­sine deam­i­
nase 2 (DADA2) results from
homo­zy­gous or com­pound het­ero­zy­gous path­o­genic var­i­ants
that result in a loss of func­tion of ADA2. Originally described as
an eti­­ol­ogy for mono­genic vas­cu­li­tis and early-onset stroke, the
phe­no­type is now rec­og­nized to be highly var­i­able and inclu­
sive of hema­to­logic and immune def­i­cits.14 Hematologic pre­sen­
ta­tions include cytopenias—eg, pure red blood cell aplasia and
bone mar­row fail­ure—spleno­meg­aly, and lymph­ade­nop­a­thy,15
with cytopenias often the first sign of dis­ease. Age of pre­sen­ta­
tion is also var­ia­ ble, with 24% of cases diag­nosed in infancy and
77% before the age of 10, but adult-onset symp­toms are increas­
ingly rec­og­nized.14,16 The mech­a­nism of neutropenia in DADA2 is
unclear; how­ever, both immune-medi­ated destruc­tion and bone
mar­row fail­ure are suspected to con­trib­ute. Genotype-phe­no­type
cor­re­la­tions are emerg­ing, with more del­e­te­ri­ous var­i­ants result­
ing in hema­to­logic dis­ease with var­i­able response to anti-tumor
necro­sis fac­tor (TNF) agents.17 Corticosteroids, rituximab, and HSCT
have been uti­lized for the treat­ment of cytopenias.
Hyper IgM syn­drome
Hyper IgM syn­drome is a genetic het­ero­ge­neous group of dis­
or­ders typ­i­fied by the inabil­ity to pro­duce class-switched immu­
no­glob­u­lin. X-linked path­o­genic var­i­ants in CD40L are the most
com­mon eti­­ol­ogy of hyper IgM syn­drome. Patients often pres­
ent in early infancy or child­hood with recur­rent sinopulmonary
infec­tions and/or oppor­tu­nis­tic infec­tions such as Pneumocystis
jirovecii, his­to­plas­mo­sis, or cryp­to­spo­rid­ium. Neutropenia can
be chronic or inter­mit­tent in CD40L defi­ciency and is asso­ci­ated
with an increased risk of oral and rec­tal ulcers, gin­gi­vi­tis, and
proctitis.18 Neutropenia can be the ini­tial clin­ic ­ al sign of hyper
IgM syn­ drome and can mimic severe con­ gen­ it­al neutropenia
sec­ond­ ary to the pres­ ence of mat­ ur­a­
tion arrest in the bone
mar­row. The exact mech­a­nism of neutropenia remains obscure,
although immune clear­ance, defec­tive mye­loid dif­fer­en­ti­a­tion,
and/or flawed bone mar­row egress have been hypoth­e­sized.19
Patients are gen­er­ally respon­sive to G-CSF. Of note, lym­pho­
mas and gas­tro­in­tes­ti­nal-related malig­nan­cies are com­mon in
patients with hyper IgM syn­drome.
GATA2 haploinsufficiency
GATA2 haploinsufficiency occurs sec­ond­ary to auto­so­mal dom­i­
nant path­o­genic var­i­ants in GATA2, an early hema­to­poi­etic tran­
scrip­ tion fac­tor active in mye­ loid devel­op­ ment. Neutropenia
occurs sec­ond­ary to neg­a­tive impacts on the hema­to­poi­etic
stem cell (HSC) pool.20 Patients with GATA2 haploinsufficiency
have var­i­able clin­i­cal pre­sen­ta­tions inclu­sive of neutropenia;
monocytopenia; B-cell, CD4+ T-cell and NK-cell lymphopenia;
myelodysplastic syn­drome; acute mye­loid leu­ke­mia; nontu­
berculous myco­bac­te­rial infec­tions; viral infec­tions (espe­cially
var­i­cella, Epstein-Barr virus [EBV], and human pap­il­loma virus
[HPV]); and/or lymphedema.21 Chronic neutropenia may be the
ini­tial sign of GATA2 haploinsufficiency, par­tic­u­larly in pedi­at­ric
patients.22 G-CSF and con­ sid­
er­ation of HSCT are ther­ ap
­ eu­tic
options to man­age neutropenia.
Warts, hypogammaglobulinemia, infec­tions, myelokathexis
syn­drome
The clin­i­
cal tet­
rad of warts, hypogammaglobulinemia, infec­
tion, and myelokathexis (WHIM) syn­drome is due to auto­so­mal
508  |  Hematology 2021  |  ASH Education Program
dom­i­nant path­o­genic var­i­ants in CXCR4 that result in the abnor­
mal reten­tion of neu­tro­phils in the bone mar­row and sub­se­quent
periph­eral neutropenia. Besides neutropenia, patients with WHIM
syn­drome often have quan­ti­ta­tive defects in mono­cytes, B cells,
and CD4+ T cells. Patients are sus­ cep­
ti­
ble to sinopulmonary
infec­tions due to hypogammaglobulinemia and HPV-related dis­
ease, includ­ing warts and malig­nancy.23 The major­ity of WHIM
patients will spon­ta­ne­ously release neu­tro­phils from the bone
mar­row dur­ing times of “stress,” eg, infec­tion, and do not require
pro­phy­lac­tic G-CSF. However, for patients who suf­fer recur­rent
infec­tions and/or fail to dem­on­strate an ade­quate release of
neu­tro­phils from the bone mar­row dur­ing infec­tious chal­lenges,
G-CSF is recommended.24 CXCR4 antag­o­nists are dem­on­strat­ing
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/504/1851486/504walkovich.pdf by guest on 13 December 2021
prom­ise as ther­ap­ eu­tic options to improve both neu­tro­phil and
lym­pho­cyte counts.
X-linked agam­ma­glob­u­lin­emia
Bruton’s X-linked agam­ma­glob­u­lin­emia (XLA) is sec­ond­ary to
path­o­genic var­i­ants in the sig­nal trans­duc­tion gene BTK that
result in the near absence of CD19+ B cells and severe hypogam­
maglobulinemia. Neutropenia is com­ mon and is most often
iden­ti­
fied dur­ing peri­ ods of mar­ row “stress,” eg, dur­ing an
active infec­tion. The exact eti­­ol­ogy of neutropenia in XLA is
uncer­tain; how­ever, given that Bruton’s tyro­sine kinase (BTK)
is expressed in mye­loid and B-cell dif­fer­en­ti­a­tion, pro­posed
mech­a­nisms include decreased mat­u­ra­tion of mye­loid pre­cur­
sors sec­ond­ary to changes in BTK-related sig­nal trans­duc­tion
compounded by changes in mono­cyte IL-18 pro­duc­tion.25 Neu­
tropenia can be the first clin­i­cal sign of XLA and is respon­sive
to G-CSF. Intravenous immu­no­glob­u­lin (IVIG) is the main­stay of
ther­apy for patients with XLA, often allowing for the with­drawal
of G-CSF.
CLINICAL CASE 1
An 8-month-old boy is admit­ted with severe neutropenia (ANC,
300/µL) and thrombocytosis (plate­let count, 650,000/µL) in
the set­ting of fever and rash. The phys­i­cal exam is nota­ble for
exten­sive impe­tigo with­out lymph­ade­nop­a­thy and the absence
of ton­sil­lar tis­sue. A com­pre­hen­sive met­a­bolic panel reveals
low total pro­tein with nor­mal albu­min. Flow cytometry for
B-cell enu­mer­a­tion dem­on­strates a near absence of CD19+ B
cells. G-CSF and IVIG are ini­ti­ated. Genetic test­ing con­firms
a path­o­genic var­i­ant in BTK con­sis­tent with XLA. Following
sev­eral IVIG infu­sions, his ANC nor­mal­izes, and the G-CSF is
with­drawn.
Author com­men­tary: Humoral deficiencies are often masked
in early infancy due to the pla­cen­tal trans­fer of mater­nal IgG.
A lack of local­ized lymph­ade­nop­a­thy despite an exten­sive skin
infec­tion, an absence of ton­sil­lar tis­sue, and a low total pro­tein
(sur­mised to be due to low immu­no­glob­u­lin, as albu­min is nor­
mal) raise con­cern for a B-cell defect. Understanding that the
neutropenia in X-linked agam­ma­glob­u­lin­emia gen­er­ally occurs
dur­ing peri­ods of mar­row stress and is respon­sive to G-CSF is
key to early man­age­ment. Additionally, aware­ness that the neu­
tropenia in XLA resolves with IVIG ini­ti­at­ ion lends con­fi­dence to
with­draw­ing the G-CSF post infec­tion res­o­lu­tion.

CLINICAL CASE 2
A 16-year-old-girl pres­ents with iso­lated mild neutropenia (ANC,
1100/µL) inci­den­tally iden­ti­fied on a screen­ing sports phys­i­cal. She
has no fam­ily his­tory of neutropenia, and her phys­i­cal exam is unre­
mark­able. Serial test­ing con­firms chronic neutropenia. Her bone
mar­row is mod­er­ately hypocellular, prompting next-gen­er­a­tion
genetic screen­ing, but no path­o­genic var­i­ants are iden­ti­fied. Sub­
sequently, she is admit­ted twice for dehy­dra­tion and cytopenias
related to a sig­nif­i­cant bout of mono­nu­cle­o­sis. Post dis­charge, her
abso­lute lym­pho­cyte and abso­lute mono­cyte counts remain low.
Following dis­cus­sion with a genet­i­cist, addi­tional test­ing is sent to
eval­u­ate for intronic var­i­ants in GATA2, and a path­o­genic var­i­ant is
Patients with prior B-cell dis­or­ders includ­ing a his­tory of auto­
im­mu­nity31 or chronic lym­pho­cytic leu­ke­mia may be at increased
risk of irAEs.28 There is uncer­tainty if PD-1/PD-L1 vs CTLA-4
block­ade results in higher hema­to­logic irAEs, although the devel­
op­ment of HLH may be more com­mon with anti-CTLA-4.27,29,33 The
devel­op­ment of autoreactive T cells and B cells, the increased
secre­tion of cyto­kine with the sub­se­quent infil­tra­tion of CD8+
T cells, and the immune-medi­ated dys­func­tion of HSC mat­u­ra­
tion or pro­lif­er­a­tion have all­ been pos­tu­lated as gen­eral mech­a­
nisms of neutropenia.26 It is prob­a­ble that dif­fer­ent mech­a­nisms
exist, based on bone mar­row biopsy find­ings dem­on­strat­ing that
a minor­ity of patients have gran­u­lo­cytic hypo­pla­sia with lym­pho­
cytic infil­trate (a cen­tral or hypo­plas­tic type), while the major­ity

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/504/1851486/504walkovich.pdf by guest on 13 December 2021

iden­ti­fied con­sis­tent with GATA2 haploinsufficiency.
Author com­men­tary: This case high­lights that many IEI have
sub­tle pre­sen­ta­tions and clin­i­cal symp­toms that evolve over
time, neces­ si­
tat­ing the need for the hema­ tol­
o­gist to remain
vig­i­lant and to con­sider gaps in test­ing, eg, the lim­i­ta­tions of
genetic-sequenc­ ing tech­ niques such as with intronic regions
or pseudogenes, in sus­pi­cious cases. The unusual require­ment
for mul­ti­ple admis­sions for a pri­mary EBV infec­tion should raise
con­cern for immune dys­func­tion. In GATA2 haploinsufficiency,
severe EBV infec­tions are fre­quently encoun­tered as den­dritic
cells, which are reduced or absent in GATA2 haploinsufficiency,
are required for EBV anti­gen pre­sen­ta­tion to T cells. Early diag­
no­sis of GATA2 haploinsufficiency in this patient greatly affects
her sup­port­ive care plan, with rec­om­men­da­tions for G-CSF and
other pre­ven­ta­tive mea­sures, eg, HPV vac­ci­na­tion and malig­
nancy sur­veil­lance. Additionally, early diag­no­sis affords her the
oppor­tu­nity to con­sider cura­tive ther­apy with HSCT.
Pathophysiology of neutropenia sec­ond­ary
to immu­no­ther­apy
The role of the immune sys­tem in tumor sur­veil­lance has long been
appre­ci­ated, but only recently have sci­en­tists and cli­ni­cians har­
nessed the power of T cells to treat can­cer. Two mod­ern immune
ther­a­pies include the use of genet­i­cally engineered T cells pro­
grammed to attack malig­nant cells, eg, chi­me­ric anti­gen recep­tor
(CAR) T cells, and block­ade of sig­nals that drive T-cell exhaus­tion
in the tumor micro­en­vi­ron­ment. The over­all tox­ic­ity from immu­no­
ther­apy is less com­pared to con­ven­tional che­mo­ther­apy, but the
devel­op­ment of immune-related adverse events (irAEs), includ­
ing immune-related neutropenia, can result in severe tox­ic­ity. The
under­stand­ing of the immu­no­logic mech­a­nisms under­pin­ning
immune-related neutropenia is incom­plete, but our knowl­edge of
neutropenia driven by IEI may pro­vide help­ful clues to uncover the
path­o­phys­i­­ol­ogy and sug­gest poten­tial treat­ments (see Table 4).
Immune check­point inhib­i­tors
Immune check­point inhib­i­tors (ICIs) include agents that inhibit
CTLA-4, programmed cell death 1 pro­tein (PD-1), or programmed
cell death 1 ligand 1 (PD-L1) sig­nal­ing and are inte­gral to the
treat­ment of many can­cers.26 The reported inci­dence of all­-grade
hema­to­logic irAEs is 0.4% to 8.3% with high-grade (grade 3-5) at
0% to 1.7%.27-29 High-grade neutropenia in iso­la­tion or in con­cert
with addi­tional cytopenias is reported at a fre­quency of 0.2% to
0.94%.28,30 Neutropenia is almost always severe with ANCs near
zero and can result in severe infec­tion and death. 27,28
The median Cytopenias are the most com­mon grade 3 or greater irAE, with
time to onset is 8.5 to 10.5 weeks after ICI ini­ti­a­tion, but neutro­ neutropenia being the most com­mon defi­ciency.39 CAR stud­ies
penia can occur at any point dur­ing Prakash
ICI treat­ment. Singh
27,31,32 Shekhawat
of patients have nor­mal or hyper­plas­tic bone mar­row with or
with­out the iden­ti­fi­ca­tion of a neu­tro­phil anti­body (periph­eral or
hyper­plas­tic type).31
Patients on ICIs presenting with neutropenia should undergo
bone mar­row eval­u­a­tion to dif­fer­en­ti­ate cen­tral vs periph­eral
neutropenia and to rule out under­ly­ing malig­nancy. A review of
other med­i­ca­tions and screen­ing for infec­tious causes of neutro­
penia are also man­da­tory. Examination of the periph­eral blood
and bone mar­ row for large gran­ u­
lar lym­ pho­ cytes should be
under­taken and immunophenotyped for large gran­u­lar lym­pho­
cyte counts greater than 0.5/µL.27 The pres­ence of high fever
should alert to the pos­si­bil­ity of HLH and pan­cy­to­pe­nia to aplas­
tic ane­mia, 2 diag­noses with poor treat­ment responses and out­
comes.27,33 Therapy for neutropenia is not spe­cif­i­cally addressed
by the Amer­i­can Society of Clinical Oncology or the Euro­pean
Society for Medical Oncology,34,35 but gen­eral lit­er­a­ture rec­
om­ men­ da­tions are to closely mon­ i­
tor patients with grade 1
neutropenia.31 For grade 2 tox­ic­ity, the ICI should be with­held
and low-dose cor­ti­co­ste­roids ini­ti­ated if no recov­ery occurs in
1 week. For grade 3 to 4 tox­ic­ity, ICIs should be discontinued and
high-dose cor­ti­co­ste­roids admin­is­tered with taper­ing over 4 to
6 weeks for responding patients.26,27 G-CSF should be pre­scribed
to has­ten recov­ery for all­grade 2 to 4 toxicities. IVIG is likely
safe and may be more effec­tive for the periph­eral destruc­tion of
neu­tro­phils,31 but there is no con­sen­sus in the lit­er­a­ture regard­
ing whether it should be used as up-front ther­apy or as sal­vage
for cor­ti­co­ste­roid non­re­spond­ers. For patients with hypo­plas­tic/
cen­tral neutropenia, cyclo­spor­ine may be trialed if there is a
poor response to cor­ti­co­ste­roids,31 and antithymocyte glob­u­lin
is an option for aplastic ane­mia.36 Most patients (approx­i­ma­tely
75%-85%) have improve­ment or nor­mal­i­za­tion of the neu­tro­phil
count, with res­o­lu­tion more likely in patients with a hyper­plas­
tic/periph­eral type,31,32 but reinstitution of the ICI fre­ quently
results in neutropenia recur­rence.28,31
CAR T cells
CAR T cells are genet­i­cally engineered to express a mod­i­fied
T-cell recep­tor with costimulatory domains that tar­get com­mon
can­cer anti­gens, such as CD19 and B-cell mat­u­ra­tion anti­gen.
CAR T cells have dem­on­strated activ­ity against a range of B-cell
malig­nan­cies, but effi­cacy has been lim­ited by cyto­kine release
syn­drome (CRS) and neuroinflammation. High lev­els of IL-1, IL-6,
inter­feron gamma, monocyte chemo-attractant protein-1, and
granulocyte-mac­ro­phage (GM) CSF largely pro­duced by mono­
cytes and mac­ro­phages drive this inflam­ma­tory tox­ic­ity.37,38
have dem­on­strated an inci­dence of grade 3 to 4 neutropenia in

Neutropenia and immune dysregulation  |  509

Table 4. Neutropenia sec­ond­ary to ICIs and CAR T-cells with cor­re­la­tion to known immune dis­or­ders

Treatment of Correlate to known Treatment of

Treatment irAE Proposed mech­a­nism cytopenias in irAE immune dis­or­ders cytopenias in IEI
ICI Hypoplastic T-cell infil­tra­tion of the bone Corticosteroids Immune-medi­ated Cyclosporine
(PD-1/PD-L1 and neutropenia/aplastic mar­row with sub­se­quent Cyclosporine aplastic ane­mia67 ATG
CTLA-4 block­ ane­mia destruc­tion of hema­to­poi­ ATG
ade) etic pre­cur­sors26 G-CSF
Immune-medi­ated dys­func­ Corticosteroids HSC dys­func­tion in ADA2 TNF inhi­bi­tion15
tion of HSC mat­u­ra­tion and G-CSF defi­ciency69
pro­lif­er­a­tion68
Hyperplastic/periph­ Peripheral destruc­tion of Corticosteroids Fas/Fas ligand defi­ Corticosteroids
eral neutropenia neu­tro­phils medi­ated by IVIG ciency and devel­op­ Rapamycin
autoantibodies31 G-CSF ment of auto­im­mune Mycophenolate

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/504/1851486/504walkovich.pdf by guest on 13 December 2021

cytopenias70 Mofetil
HLH T-cell and mac­ro­phage Corticosteroids HLH and MAS dis­or­ders Corticosteroids
acti­va­tion, per­haps from Etoposide with nor­mal phago­ Etoposide
immune cell tumor inva­sion cytic and cyto­toxic IFNG anti­body
and malig­nant cell apo­ func­tion; malig­nancy- IL-1 anti­body
pto­sis71 asso­ci­ated HLH72 IL-6R anti­body
Jak inhib­i­tors
ICI Multiple irAE, includ­ing Regulatory T-cell inhi­bi­tion Corticosteroids CTLA-4 haploinsufficiency Corticosteroids
CTLA-4 block­ade neutropenia and/or deple­tion73,74 (to sup­press LRBA defi­ciency CTLA-4-Ig
expanded effec­ IPEX syn­drome (FOXP3 Rapamycin
tor T cells) defi­ciency) Calcineurin
G-CSF CD25 defi­ciency inhib­i­tors
CAR-T Early neutropenia Lymphodepleting ther­apy IL-6R anti­body HLH and MAS syn­dromes Corticosteroids
and CRS75 Corticosteroids with nor­mal phago­ Etoposide
cytic and cyto­toxic IFNG anti­body
func­tion IL-1 anti­body
IL-6R anti­body
Jak inhib­i­tors
Late neutropenia Disrupted CXCL12 bone G-CSF WHIM syn­drome (CXCR4- G-CSF
mar­row gra­di­ent inhibiting GOF) CXCR4 inhi­bi­tion
neu­tro­phil egress46 Late-onset neutropenia
sec­ond­ary to rituximab48
ATG, antithymocyte glob­u­lin; GOF, gain-of-func­tion; IFNG, inter­feron gamma; Ig, immu­no­glob­u­lin; IPEX, immunodysregulation, polyendocrinopathy,
and enter­op­a­thy, X-linked; MAS, mac­ro­phage acti­va­tion syn­drome.

60% to 85% of patients and “late neutropenia” occur­ring 28 days
or later in 3% to 53%.40-44
Early neutropenia has been attrib­uted to immunodepleting
ther­apy with fludarabine and cyclo­phos­pha­mide preinfusion and
hema­to­poi­etic sup­pres­sion from CRS.45 However, late neutrope­
nia, which can per­sist well after che­mo­ther­apy and res­o­lu­tion of
CRS, is more enig­matic. Possible expla­na­tions for late neutrope­
nia include poor bone mar­row reserves and exces­sive inflam­ma­
tory dam­age given the asso­ci­a­tion with prior HCT/mul­ti­ple lines
of ther­apy and higher-grade CRS,45-47 respec­tively, in patients
with prolonged hema­to­logic tox­ic­ity. Interestingly, a sub­set of
patients with late neutropenia show an ini­tial recov­ery of neu­
tro­phils only to be followed by a sec­ond trough with­out any
inter­ven­ing ther­apy.45,46 One group eval­ua ­ ted serum CXCL12 lev­
els and reported a cor­re­la­tion in patients with late neutropenia,
hypoth­e­siz­ing that the con­sump­tion of CXCL12 from an expand­
ing pre­cur­sor B-cell pop­u­la­tion may dis­rupt the nor­mal CXCL12
bone mar­row gra­di­ent required for nor­mal neu­tro­phil egress,
resulting in a recur­rence of periph­eral neutropenia in some CD19
CAR T-cell recip­i­ents.46 A sim­i­lar mech­a­nism has been pro­posed
for late-onset neutropenia fol­low­ing rituximab ther­apy.48
Guidance for the eval­ ua
­t­ion and treat­ ment of neutropenia
fol­low­ing CAR T cells is cur­rently lacking, but many sug­gest a
bone mar­row biopsy for neutropenia persisting 1 month fol­low­ing
infu­sion.45 Both hypocellularity and nor­mal cel­lu­lar­ity have been
reported, per­haps indi­cat­ing dif­fer­ing mech­a­nisms of neutro­
penia.45,46 G-CSF may has­ten neu­tro­phil recov­ery but should be
avoided early (within 3 weeks) of infu­sion given the poten­tial for
height­ened CRS. Data to sup­port this con­cern include higher lev­
els of murine G-CSF in mice cor­re­lat­ing with CRS sever­ity,49 ele­
vated lev­els of G-CSF and GM-CSF in patients with neu­ro­tox­ic­ity,50
and 1 small study dem­on­strat­ing that G-CSF use prox­i­mal to CAR
T-cell infu­sion was asso­ci­ated with higher CRS sever­ity.37 Eltrom­
bopag has been used with suc­cess in patients with aplasia, and
autol­o­gous stem cell infu­sion should be con­sid­ered for very late
neutropenia (greater than 3 months) if cryopreserved cells are
avail­­able.45
CLINICAL CASE 3
A 65-year-old woman with met­as­ tatic mel­a­noma is treated with
sin­gle-agent ipilimumab (anti-CTLA-4). Two weeks fol­low­ing
her third cycle of ther­apy, she pres­ents with fever and an ANC
of 80/µL. A bone mar­row biopsy dem­on­strates increased mye­
loid cells with­out a lym­pho­cytic infil­trate or mat­u­ra­tion block.
Additional cycles of ipilimumab are tem­ po­
rar­
ily discontinued,

510  |  Hematology 2021  |  ASH Education Program

and she is admin­is­tered cor­ti­co­ste­roids, G-CSF, and IVIG, with a
ris­ing ANC within 4 days of treat­ment ini­ti­a­tion. Her ste­roids are
tapered over 4 weeks with a con­tin­ued nor­mal ANC. Her oncol­o­
gist tri­als another cycle of ipilimumab, but her neutropenia recurs.
Author com­men­tary: Neutropenia as an irAE is a rare but seri­
ous com­pli­ca­tion of ICIs. The loss of CTLA-4 through iat­ro­genic
block­ade or genetic haploinsufficiency results in the loss of reg­
u­la­tory T-cell sup­pres­sive activ­ity and immune dysregulation.
Subsequent neutropenia can develop through sev­eral pro­posed
mech­a­nisms, includ­ing T-cell medi­ated attack on the bone mar­
row, auto­an­ti­body for­ma­tion, hypercytokinemia, and immune-
medi­ ated HSC dys­ func­tion. In this case, mye­ loid hyper­ pla­
sia
with­out CD8+ lym­pho­cy­to­sis in the bone mar­row would favor
periph­eral destruc­tion by antineutrophil antibodies, a mech­a­nism
more likely to respond over­all and to IVIG treat­ment. Although
most patients with immune-related neutropenia recover with
treat­ment, recur­rence with reinitiation of the ICI occurs in the
major­ity of patients and should only be trialed if deemed essen­
tial for oncol­ogy care and with care­ful com­plete blood counts
with plate­lets and dif­fer­en­tial mon­i­tor­ing.
Conclusion
As an inte­grated com­po­nent of the immune sys­tem, the neu­tro­
phil func­tions as a strong com­pan­ion with other immune cells to
pro­tect the body under nor­mal phys­i­­ol­ogy. However, with dis­
rup­ tion of tolerogenic path­ ways through genetic alter­ ation or
iat­ro­genic inhib­i­tors, the neu­tro­phil can quickly turn from friend
to foe to unre­strained immune cells. Early and cor­rect iden­ti­fi­ca­
tion of immune dysregulation presenting with or com­pli­cated by
neutropenia is crit­i­cal to improv­ing patient out­comes related to
selecting vastly expanding targeted ther­a­peu­tic options, min­i­miz­
ing infec­tions through the early ini­ti­a­tion of anti­mi­cro­bial pro­phy­
laxis and vac­ci­na­tion strat­e­gies, mon­i­tor­ing for malig­nan­cies, and
lim­it­ing end organ dam­age. Establishing an IEI diag­no­sis is also
nec­es­sary for future preg­nancy coun­sel­ing and HSCT donor eval­u­
a­tions. While many of the mech­a­nisms of neutropenia in IEI remain
incom­pletely under­stood, the com­mon themes of disrupted neu­
tro­phil pro­duc­tion resulting in decreased mar­row reserves, ham­
pered mobi­li­za­tion from the mar­row, and exces­sive destruc­tion
are appar­ent. Leveraging fur­ther mech­a­nis­tic eval­u­a­tion into
IEI-related neutropenia may also illu­mi­nate the path­o­phys­i­­ol­ogy
of neutropenia as a con­se­quence of malig­nancy immu­no­ther­apy.
Through such research, the neu­tro­phil can hope­fully be protected
from the con­se­quences of induced immune acti­va­tion with­out
sac­ri­fic­ing the T-cell onslaught on malig­nant cells.
Conflict-of-inter­est dis­clo­sure
Kelly Walkovich: con­ sul­tancy: Swed­ ish Orphan Biovitrum AB;
advi­sory board mem­ ber: Horizon Therapeutics, AstraZeneca,
Pharming; local prin­ci­pal inves­ti­ga­tor: mavorixafor clin­i­cal trial
spon­sored by X4 Pharmaceuticals.
James A. Connelly: advi­sory board mem­ber: X4 Pharmaceuticals,
Horizon Therapeutics.
Off-label drug use
Kelly Walkovich: therapy with rituximab, rapamycin, CXCR4 inhi­
bition, cyclosporine, ATG, MMG, JAK inhibitors, IL-1 antibodies,
IL-6R antibodies, and CTLA-4-Ig are briefly discussed as options
Prakash
for treating immunodysregulation from Singhetiologies
IEI or iatrogenic Shekhawat
of immune dysregulation. While commonly utilized in these rare
diseases, these conditions are off-label uses from the formal FDA
label.
James A. Connelly: therapy with rituximab, rapamycin, CXCR4 in­
hibition, cyclosporine, ATG, MMG, JAK inhibitors, IL-1 antibodies,
IL-6R antibodies, and CTLA-4-Ig are briefly discussed as options
for treating immunodysregulation from IEI or iatrogenic etiolo­
gies of immune dysregulation. While commonly utilized in these
rare diseases, these conditions are off-label uses from the formal
FDA label.
Correspondence
Kelly Walkovich, Department of Pediatrics, University of Mich­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/504/1851486/504walkovich.pdf by guest on 13 December 2021
igan, 1540 E Medical Center Dr, Ann Arbor, MI 48109; e-mail:
kwalkovi@med​­.umich​­.edu.
References
1. Silvestre-Roig C, Hidalgo A, Soehnlein O. Neutrophil het­ero­ge­ne­ity: impli­
ca­tions for homeo­sta­sis and path­o­gen­e­sis. Blood. 2016;127(18):2173-2181.
2. Michniacki TF, Connelly JA, Sturza J, et al. Neutropenia is an underrecog­
nized find­ing in pedi­at­ric pri­mary immu­no­de­fi­ciency dis­eases: an anal­y­sis
of the United States immu­no­de­fi­ciency net­work reg­is­try. J Pediatr Hema-
tol Oncol. 2020;42(7):e601-e605.
3. Seidel MG. Autoimmune and other cytopenias in pri­ mary immunodefi­
ciencies: pathomechanisms, novel dif­fer­en­tial diag­noses, and treat­ment.
Blood. 2014;124(15):2337-2344.
4. Derpoorter C, Bordon V, Laureys G, Haerynck F, Lammens T. Genes at the
cross­road of pri­mary immunodeficiencies and can­cer. Front Immunol.2018;
9(Novem­ber):2544.
5. Yska HAF, Elsink K, Kuijpers TW, Frederix GWJ, van Gijn ME, van Montfrans
JM. Diagnostic yield of next gen­er­a­tion sequenc­ing in genet­i­cally undi­ag­
nosed patients with pri­mary immunodeficiencies: a sys­tem­atic review. J
Clin Immunol. 2019;39(6):577-591.
6. Heimall JR, Hagin D, Hajjar J, et al. Use of genetic test­ing for pri­mary immu­
no­de­fi­ciency patients. J Clin Immunol. 2018;38(3):320-329.
7. Bonilla FA, Khan DA, Ballas ZK, et al; Joint Task Force on Practice Parame­
ters, representing the Amer­i­can Academy of Allergy, Asthma and Immunol­
ogy, the Amer­i­can College of Allergy, Asthma and Immunology, and the
Joint Council of Allergy, Asthma and Immunology. Practice param­e­ter for
the diag­no­sis and man­age­ment of pri­mary immu­no­de­fi­ciency. J Allergy
Clin Immunol. 2015;136(5):1186-1205.e1.
8. Sullivan KE. Neutropenia as a sign of immu­ no­de­
fi­
ciency. J Allergy Clin
Immunol. 2019;143(1):96-100.
9. Blume RS, Bennett JM, Yan­kee RA, Wolff SM. Defective granulocyte reg­u­
la­tion in the Chediak-Higashi syn­drome. N Engl J Med. 1968;279(19):1009-
1015.
10. Feuille EJ, Anooshiravani N, Sullivan KE, Fuleihan RL, Cunningham-Rundles
C. Autoimmune cytopenias and asso­ci­ated con­di­tions in CVID: a report
from the USIDNET reg­is­try. J Clin Immunol. 2018;38(1):28-34.
11. Guffroy A, Mourot-Cottet R, Gérard L, et al; DEFI Study Group. Neutropenia
in patients with com­mon var­i­able immu­no­de­fi­ciency: a rare event asso­ci­
ated with severe out­come. J Clin Immunol. 2017;37(7):715-726.
12. Ghorbani M, Fekrvand S, Shahkarami S, et al. The eval­u­a­tion of neutropenia
in com­mon var­i­able immune defi­ciency patients. Expert Rev Clin Immunol.
2019;15(11):1225-1233.
13. Bride KL, Vincent T, Smith-Whitley K, et  al. Sirolimus is effec­ tive in
relapsed/refrac­ tory auto­ im­
mune cytopenias: results of a pro­ spec­tive
multi-insti­tu­tional trial. Blood. 2016;127(1):17-28.
14. Meyts I, Aksentijevich I. Deficiency of aden­o­sine deam­i­nase 2 (DADA2):
updates on the phe­no­type, genet­ics, path­o­gen­e­sis, and treat­ment. J Clin
Immunol. 2018;38(5):569-578.
15. Michniacki TF, Hannibal M, Ross CW, et  al. Hematologic man­i­fes­ta­tions
of defi­ciency of aden­o­sine deam­i­nase 2 (DADA2) and response to tumor
necro­sis fac­tor inhi­bi­tion in DADA2-asso­ci­ated bone mar­row fail­ure. J Clin
Immunol. 2018;38(2):166-173.
16. Sharma A, Naidu G, Sharma V, et al. Deficiency of aden­o­sine deam­i­nase 2 in
adults and chil­dren: expe­ri­ence from India. Arthritis Rheumatol. 2021;73(2):
276-285.
Neutropenia and immune dysregulation  |  511
17. Lee PY, Kellner ES, Huang Y, et al. Genotype and func­tional cor­re­lates of
dis­ease phe­no­type in defi­ciency of aden­o­sine deam­i­nase 2 (DADA2). J
Allergy Clin Immunol. 2020;145(6):1664-1672.e101672e10.
18. Leven EA, Maffucci P, Ochs HD, et al. Hyper IgM syn­drome: a report from
the USIDNET reg­is­try. J Clin Immunol. 2016;36(5):490-501.
19. Rezaei N, Moazzami K, Aghamohammadi A, Klein C. Neutropenia and pri­
mary immu­no­de­fi­ciency dis­eases. Int Rev Immunol. 2009;28(5):335-366.
20. Rodrigues NP, Janzen V, Forkert R, et al. Haploinsufficiency of GATA-2 per­
turbs adult hema­to­poi­etic stem-cell homeo­sta­sis. Blood. 2005;106(2):
477-484.
21. Spinner MA, Sanchez LA, Hsu AP, et al. GATA2 defi­ciency: a pro­tean dis­
or­der of hema­to­poi­e­sis, lym­phat­ics, and immu­nity. Blood. 2014;123(6):
809-821.
22. Pasquet M, Bellanné-Chantelot C, Tavitian S, et  al. High fre­ quency of
GATA2 muta­tions in patients with mild chronic neutropenia evolv­ing to
MonoMac syn­drome, myelodysplasia, and acute mye­loid leu­ke­mia. Blood.
2013;121(5):822-829.
23. Heusinkveld LE, Majumdar S, Gao JL, McDermott DH, Mur­phy PM. WHIM
syn­drome: from path­o­gen­e­sis towards per­son­al­ized med­i­cine and cure. J
Clin Immunol. 2019;39(6):532-556.
24. Badolato R, Donadieu J; WHIM Research Group. How I treat warts,
hypogammaglobulinemia, infec­tions, and myelokathexis syn­drome. Blood.
2017;130(23):2491-2498.
25. Jacobs ZD, Guajardo JR, Anderson KM. XLA-asso­ci­ated neutropenia treat­
ment: a case report and review of the lit­er­a­ture. J Pediatr Hematol Oncol.
2008;30(8):631-634.
26. Omar NE, El-Fass KA, Abushouk AI, et  al. Diagnosis and man­age­ment of
hema­to­log­i­cal adverse events induced by immune check­point inhib­i­tors: a
sys­tem­atic review. Front Immunol. 2020;11(21 Octo­ber):1354.
27. Michot JM, Lazarovici J, Tieu A, et  al. Haematological immune-related
adverse events with immune check­point inhib­i­tors, how to man­age? Eur J
Cancer. 2019;122(Novem­ber):72-90.
28. Delanoy N, Michot JM, Comont T, et  al. Haematological immune-related
adverse events induced by anti-PD-1 or anti-PD-L1 immu­ no­ther­
apy: a
descrip­tive obser­va­tional study. Lancet Haematol. 2019;6(1):e48-e57.
29. Kramer R, Zaremba A, Moreira A, et  al. Hematological immune related
adverse events after treat­ment with immune check­point inhib­i­tors. Eur J
Cancer. 2021;147(April):170-181.
30. Petrelli F, Ardito R, Borgonovo K, et  al. Haematological toxicities with
immu­no­ther­apy in patients with can­cer: a sys­tem­atic review and meta-
anal­y­sis. Eur J Cancer. 2018;103(Novem­ber):7-16.
31. Finkel I, Sternschuss M, Wollner M, et al; Israel Lung Cancer Group. Immune-
related neutropenia fol­low­ing treat­ment with immune check­point inhib­i­tors.
J Immunother. 2020;43(2):67-74.
32. Boegeholz J, Brueggen CS, Pauli C, et al. Challenges in diag­no­sis and man­
age­ment of neutropenia upon expo­sure to immune-check­point inhib­i­tors:
meta-anal­y­sis of a rare immune-related adverse side effect. BMC Cancer.
2020;20(1):300.
33. Davis EJ, Salem JE, Young A, et al. Hematologic com­pli­ca­tions of immune
check­point inhib­i­tors. Oncologist. 2019;24(5):584-588.
34. Brahmer JR, Lacchetti C, Thompson JA. Management of immune-related
adverse events in patients treated with immune check­point inhib­i­tor
ther­apy: Amer­i­can Society of Clinical Oncology Clinical Practice Guide­
line Summary. J Oncol Pract. 2018;14(4):247-249.
35. Haanen JBAG, Carbonnel F, Robert C, et al. Management of toxicities from
immu­no­ther­apy: ESMO Clinical Practice Guidelines for diag­no­sis, treat­
ment and fol­low-up. Ann Oncol. 2017;28(suppl 1):iv119-iv142.
36. Zhuang J, Du J, Guo X, et al. Clinical diag­no­sis and treat­ment rec­om­men­
da­tions for immune check­point inhib­i­tor-related hema­to­log­i­cal adverse
events. Thorac Cancer. 2020;11(3):799-804.
37. Gaut D, Tang K, Sim MS, Duong T, Young P, Sasine J. Filgrastim asso­ci­a­tions
with CAR T-cell ther­apy. Int J Cancer. 2021;148(5):1192-1196.
38. Sterner RM, Sakemura R, Cox MJ, et al. GM-CSF inhi­bi­tion reduces cyto­kine
release syn­drome and neuroinflammation but enhances CAR-T cell func­tion
in xeno­grafts. Blood. 2019;133(7):697-709.
39. Yáñez L, Sánchez-Escamilla M, Perales MA. CAR T cell tox­ic­ity: cur­rent man­
age­ment and future direc­tions. Hemasphere. 2019;3(2):e186.
40. Neelapu SS, Locke FL, Bartlett NL, et  al. Axicabtagene ciloleucel CAR
T-cell ther­apy in refrac­tory large B-cell lym­phoma. N Engl J Med. 2017;377
­(26):2531-2544.
41. Abramson JS, Palomba ML, Gordon LI, et  al. Lisocabtagene maraleucel
for patients with relapsed or refrac­tory large B-cell lym­pho­mas (TRAN­
SCEND NHL 001): a multicentre seam­less design study. Lancet. 2020;396
­(10254):839-852.
512  |  Hematology 2021  |  ASH Education Program
42. Raje N, Berdeja J, Lin Y, et  al. Anti-BCMA CAR T-cell ther­apy bb2121 in
relapsed or refrac­tory mul­ti­ple mye­loma. N Engl J Med. 2019;380(18):1726-
1737.
43. Schuster SJ, Bishop MR, Tam CS, et al; JULIET Investigators. Tisagenlecleu­
cel in adult relapsed or refrac­tory dif­fuse large B-cell lym­phoma. N Engl J
Med. 2019;380(1):45-56.
44. Maude SL, Laetsch TW, Buechner J, et al. Tisagenlecleucel in chil­dren and
young adults with B-cell lym­pho­blas­tic leu­ke­mia. N Engl J Med. 2018;378
­(5):439-448.
45. Nagle SJ, Murphree C, Raess PW, et  al. Prolonged hema­to­logic tox­ic­ity
fol­low­ing treat­ment with chi­me­ric anti­gen recep­tor T cells in patients with
hema­to­logic malig­nan­cies. Am J Hematol. 2021;96(4):455-461.
46. Fried S, Avigdor A, Bielorai B, et al. Early and late hema­to­logic tox­ic­ity fol­
low­ing CD19 CAR-T cells. Bone Marrow Transplant. 2019;54(10):1643-1650.
47. Chakraborty R, Hill BT, Majeed A, Majhail NS. Late effects after chi­me­ric
anti­gen recep­tor T cell ther­apy for lym­phoid malig­nan­cies. Transplant Cell
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/504/1851486/504walkovich.pdf by guest on 13 December 2021
Ther. 2021;27(3):222-229.
48. Dunleavy K, Hakim F, Kim HK, et  al. B-cell recov­ery fol­low­ing rituximab-
based ther­apy is asso­ci­ated with per­tur­ba­tions in stro­mal derived fac­tor-1 and
granulocyte homeo­sta­sis. Blood. 2005;106(3):795-802.
49. Giavridis T, van der Stegen SJC, Eyquem J, Hamieh M, Piersigilli A, Sadelain
M. CAR T cell-induced cyto­kine release syn­drome is medi­ated by mac­ro­
phages and abated by IL-1 block­ade. Nat Med. 2018;24(6):731-738.
50. Santomasso BD, Park JH, Salloum D, et al. Clinical and bio­log­i­cal cor­re­lates
of neu­ro­tox­ic­ity asso­ci­ated with CAR T-cell ther­apy in patients with B-cell
acute lym­pho­blas­tic leu­ke­mia. Cancer Discov. 2018;8(8):958-971.
51. Kjeldsen L, Calafat J, Borregaard N. Giant gran­ules of neu­tro­phils in
Chediak-Higashi syn­drome are derived from azurophil gran­ules but not
from spe­cific and gelatinase gran­ules. J Leukoc Biol. 1998;64(1):72-77.
52. Vilboux T, Lev A, Malicdan MC, et al. A con­gen­i­tal neu­tro­phil defect syn­
drome asso­ci­ated with muta­tions in VPS45. N Engl J Med. 2013;369(1):
54-65.
53. McDermott DH, De Ravin SS, Jun HS, et al. Severe con­gen­i­tal neutropenia
resulting from G6PC3 defi­ciency with increased neu­tro­phil CXCR4 expres­
sion and myelokathexis. Blood. 2010;116(15):2793-2802.
54. Kuhns DB, Fink DL, Choi U, et al. Cytoskeletal abnor­mal­i­ties and neu­tro­phil
dys­func­tion in WDR1 defi­ciency. Blood. 2016;128(17):2135-2143.
55. Donadieu J, Fenneteau O, Beaupain B, Mahlaoui N, Chantelot CB. Congen­
ital neutropenia: diag­ no­sis, molec­
ul­ar bases and patient man­ age­ ment.
Orphanet J Rare Dis. 2011;6(May):26.
56. Boztug K, Järvinen PM, Salzer E, et  al. JAGN1 defi­ ciency causes aber­
rant mye­loid cell homeo­sta­sis and con­gen­i­tal neutropenia. Nat Genet.
2014;46(9):1021-1027.
57. Introne W, Boissy RE, Gahl WA. Clinical, molec­ul­ar, and cell bio­log­i­cal
aspects of Chediak-Higashi syn­drome. Mol Genet Metab. 1999;68(2):
283-303.
58. Lozano ML, Rivera J, Sánchez-Guiu I, Vicente V. Towards the targeted man­age­
ment of Chediak-Higashi syn­drome. Orphanet J Rare Dis. 2014;9(August):132.
59. Romberg N, Le Coz C, Glauzy S, et  al. Patients with com­mon var­i­able
immu­no­de­fi­ciency with auto­im­mune cytopenias exhibit hyper­plas­tic yet
inef­fi­cient ger­mi­nal cen­ter responses. J Allergy Clin Immunol. 2019;143(1):
258-265.
60. Pinto B, Deo P, Sharma S, Syal A, Sharma A. Expanding spec­trum of DADA2:
a review of phe­no­types, genet­ics, path­o­gen­e­sis and treat­ment. Clin Rheu-
matol. 2021;40(10):3883-3896.
61. Hashem H, Kumar AR, Müller I, et al; Deficiency of Adenosine Deaminase
Type 2 Foundation. Hematopoietic stem cell trans­plan­ta­tion res­cues the
hema­to­log­i­cal, immu­no­log­i­cal, and vas­cu­lar phe­no­type in DADA2. Blood.
2017;130(24):2682-2688.
62. Cabral-Marques O, França TT, Al-Sbiei A, et  al. CD40 ligand defi­ciency
causes func­tional defects of periph­eral neu­tro­phils that are improved by
exog­e­nous IFN-γ. J Allergy Clin Immunol. 2018;142(5):1571-1588.e91588e9.
63. Solanilla A, Déchanet J, El Andaloussi A, et  al. CD40-ligand stim­ u­
lates
myelopoiesis by reg­u­lat­ing flt3-ligand and thrombopoietin pro­duc­tion in
bone mar­row stro­mal cells. Blood. 2000;95(12):3758-3764.
64. Ochs HD, Smith CI. X-linked agam­ma­glob­u­lin­emia. A clin­i­cal and molec­u­lar
anal­y­sis. Medicine (Baltimore). 1996;75(6):287-299.
65. Farrar JE, Rohrer J, Conley ME. Neutropenia in X-linked agam­ma­glob­u­lin­
emia. Clin Immunol Immunopathol. 1996;81(3):271-276.
66. Winkelstein JA, Marino MC, Lederman HM, et  al. X-linked agam­ma­glob­u­
lin­emia: report on a United States reg­is­try of 201 patients. Medicine (Balti-
more). 2006;85(4):193-202.
67. Wang L, Liu H. Pathogenesis of aplastic ane­mia. Hematology. 2019;24(1):
559-566.
68. Inadomi K, Kumagai H, Arita S, et  al. Bi-cytopenia pos­si­bly induced by 73. Sharma A, Subudhi SK, Blando J, et al. Anti-CTLA-4 immu­no­ther­apy does
anti-PD-1 anti­body for pri­
mary malig­ nant mel­
a­noma of the esoph­ a­
gus: not deplete FOXP3. Clin Cancer Res. 2019;25(4):1233-1238.
a case report. Medicine (Baltimore). 2016;95(29):e4283. 74. Simpson TR, Li F, Montalvo-Ortiz W, et  al. Fc-depen­ dent deple­ tion of
69. Lee PY. Vasculopathy, immu­no­de­fi­ciency, and bone mar­row fail­ure: the tumor-infil­trat­ing reg­u­la­tory T cells co-defi­nes the effi­cacy of anti-CTLA-4
intrigu­ing syn­
drome caused by defi­ ciency of aden­ o­
sine deam­ i­
nase 2. ther­apy against mel­a­noma. J Exp Med. 2013;210(9):1695-1710.
Front Pediatr. 2018;6(18 Octo­ber):282. 75. Neelapu SS, Tummala S, Kebriaei P, et al. Chimeric anti­gen recep­tor T-cell
70. Rieux-Laucat F, Magérus-Chatinet A, Neven B. The auto­im­mune lymphop­ ther­apy—assess­ment and man­age­ment of toxicities. Nat Rev Clin Oncol.
roliferative syn­drome with defec­tive FAS or FAS-ligand func­tions. J Clin 2018;15(1):47-62.
Immunol. 2018;38(5):558-568.
71. Hantel A, Gabster B, Cheng JX, Golomb H, Gajewski TF. Severe hemophago­
cytic lymphohistiocytosis in a mel­a­noma patient treated with ipilimumab +
nivolumab. J Immunother Cancer. 2018;6(1):73.
72. Daver N, McClain K, Allen CE, et  al. A con­sen­sus review on malig­nancy-
asso­ci­ated hemophagocytic lymphohistiocytosis in adults. Cancer. 2017; © 2021 by The Amer­i­can Society of Hematology
123(17):3229-3240. DOI 10.1182/hema­tol­ogy.2021000285

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/504/1851486/504walkovich.pdf by guest on 13 December 2021

Prakash Singh Shekhawat

Neutropenia and immune dysregulation  |  513
 NEUTROPENIA: DOING MORE WITH LESS

Impaired myelopoiesis in congenital neutropenia:

insights into clonal and malignant hematopoiesis

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/514/1851470/514warren.pdf by guest on 13 December 2021

Julia T. Warren1 and Daniel C. Link2
Division of Hematology-Oncology, Department of Pediatrics, and 2Division of Oncology, Department of Internal Medicine, Washington University
1

School of Medicine, St. Louis, MO

A common feature of both congenital and acquired forms of bone marrow failure is an increased risk of developing acute
myeloid leukemia (AML) or myelodysplastic syndrome (MDS). Indeed, the development of MDS or AML is now the major
cause of mortality in patients with congenital neutropenia. Thus, there is a pressing clinical need to develop better strat-
egies to prevent, diagnose early, and treat MDS/AML in patients with congenital neutropenia and other bone marrow
failure syndromes. Here, we discuss recent data characterizing clonal hematopoiesis and progression to myeloid malig-
nancy in congenital neutropenia, focusing on severe congenital neutropenia (SCN) and Shwachman-Diamond syndrome.
We summarize recent studies showing excellent outcomes after allogenic hematopoietic stem cell transplantation for
many (but not all) patients with congenital neutropenia, including patients with SCN with active myeloid malignancy
who underwent transplantation. Finally, we discuss how these new data inform the current clinical management of
patients with congenital neutropenia.

LEARNING OBJECTIVES
• Understand the indications for and interpretation of somatic gene pane sequencing and karyotyping in congenital
neutropenia
• Integrate recent improvements in hematopoietic stem cell transplantation outcome in congenital neutropenia
into clinical management
• Appreciate the contribution of stressors in the development of clonal hematopoiesis and myeloid malignancy in
congenital neutropenia

Introduction
There has been a recent increase in the use of somatic gene
panel sequencing in patients with congenital neutropenia
to identify patients at risk of leukemic transformation. We
present three cases of congenital neutropenia in which
results of molecular testing raise important clinical man-
agement questions. To answer these questions, we review
recent advances on the following topics: (1) the frequency
and molecular features of myeloid malignancy in congenital
neutropenia, (2) clonal hematopoiesis in congenital neutro-
penia and mechanisms of clonal evolution to myeloid malig-
nancy, and (3) recent studies of hematopoietic stem cell
transplantation (HSCT) outcomes in congenital neutropenia.
Finally, we explore how these new data inform the clinical
management of congenital neutropenia, including the case
presentations.
CLINICAL CASE 1
This patient presented in early childhood with recurrent
infections and an absolute neutrophil count (ANC) of <200
cells/mm3. Bone marrow biopsy specimen showed mye-
loid maturation arrest, and genetic analysis revealed a het-
erozygous germline pathogenic variant in ELANE (p.S126L).
Serial surveillance bone marrow biopsy specimens dem-
onstrated no overt clonal abnormalities as detected by
karyotype and fluorescence in situ hybridization. However,
somatic sequencing of a surveillance bone marrow sample
obtained at age 45 years showed a high variant allele fre-
quency (41%) truncating pathogenic variant in CSF3R. Lower
frequency clones in DNMT3A (1.7%) and ASXL1 (2.6%) also
were detected.

514 | Hematology 2021 | ASH Education Program


CLINICAL CASE 2
This patient presented at age 2 years with neutropenia, fail­ure
to thrive with chronic diar­rhea, mild devel­op­men­tal delay, and
pel­vic dysostosis. The ANC was 320 cells/mm3, hemo­glo­bin
was 11.7 g/dL, and plate­lets were 132 000 cells/mm3. Initial bone
mar­row biopsy performed at age 2 years dem­on­strated hypo-
cellularity with a decrease in mature mye­loid cells and nor­mal
cyto­ge­net­ics. Sequencing revealed biallelic germline path­o­genic
var­i­ants in SBDS (c. 183_184TA _CT and c. 258 + 2T>C), confirming
the diag­ no­sis of Shwachman-Diamond syn­ drome (SDS). Over
time, the patient devel­oped mod­er­ate ane­mia with mac­ro­cy­to­sis
and con­tin­ued with inter­mit­tent mild to mod­er­ate neutropenia.
Repeat bone mar­row exam­i­na­tion revealed mild dysmorphology
pre­dom­i­nantly in the mye­loid lin­e­ages with­out dys­pla­sia and
mega­kar­yo­cyte hypo­pla­sia. Serial somatic gene panel sequenc­
ing over a 5-year time span iden­ti­fied mul­ti­ple het­ero­zy­gous low-
abun­dance TP53 path­o­genic mis­sense var­i­ants.
CLINICAL CASE 3
The patient presented in infancy with omphalitis and was noted
to have severe neutropenia with an ANC of <200 cells/mm3.
Before ini­ti­a­tion of granulocyte col­ony-stim­u­lat­ing fac­tor
(G-CSF), the patient expe­ ri­
enced recur­ rent sinopulmonary
infec­tions and cel­lu­li­tis. Germline sequenc­ing revealed a path­
o­genic ELANE var­i­ant (p.V65F). Somatic gene panel sequenc­ing
obtained at age 7 years did not dem­on­strate any severe con­
gen­i­tal neutropenia (SCN)–asso­ci­ated or other clonal hema­to­
poi­e­sis var­i­ants, and serial bone mar­row exam­i­na­tions did not
reveal any dys­pla­sia or chro­mo­somal abnor­mal­i­ties. Although
infec­tious com­pli­ca­tions ini­tially improved while on G-CSF ther­
apy, the patient even­ tu­
ally expe­ ri­
enced severe neutropenia
again with con­com­i­tant recur­rent cel­lu­li­tis despite esca­lat­ing
doses of G-CSF (>10 µg/kg/dose).
Myeloid malig­nancy in con­gen­i­tal neutropenia
SCN is a rare syn­drome char­ac­ter­ized by chronic neutropenia
pres­ent from birth and recur­ring bac­te­rial infec­tions.1 Pathogenic
var­i­ants of ELANE are the most com­mon cause of SCN, account­
ing for approx­i­ma­tely 50% of cases, with var­i­ants in CLPB, HAX1,
and G6PC3 account­ing for an addi­tional 10% to 20% of cases
depending on geo­graphic loca­tion (eg, HAX1 var­i­ants have not
been observed in North America).2-5 Treatment with G-CSF is
the stan­dard of care for SCN; it increases the level of cir­cu­lat­
ing neu­tro­phils in most patients and reduces infec­tion-related
mor­tal­ity.6 The devel­op­ment of mye­loid malig­nancy is now the
major cause of mor­tal­ity in SCN. Indeed, after 15 years of G-CSF
ther­apy, the cumu­la­tive inci­dence of myelodysplastic syn­drome
(MDS) or acute mye­loid leu­ke­mia (AML) in patients with SCN was
esti­mated at 22%.7 In this study, trans­for­ma­tion to MDS/AML was
asso­ci­ated with a higher G-CSF dose (>8 µg/kg/d) and with a
reduced neu­tro­phil response to G-CSF. Similar data from the
French Neutropenia Registry reported an 11% cumu­la­tive inci­
dence of MDS/AML by age 20 years.8 Of note, almost all­genetic
sub­types of SCN have been asso­ci­ated with an increased risk
for trans­for­ma­tion to MDS/AML.9 MDS/AML aris­ing in the set­ting
Prakash Singh Shekhawat
of SCN is asso­ci­ated with dis­tinct molec­u­lar fea­tures com­pared
with de novo MDS/AML.10 Most nota­bly, somatic trun­cat­ing var­
i­ants in CSF3R and mis­sense var­i­ants in RUNX1 are observed in
~80% and 63% of cases, respec­tively.11 In addi­tion, sec­ond­ary
MDS/AML in patients with SCN is fre­ quently asso­ ci­
ated with
mono­somy 7 and tri­somy 21.12-14
SDS is an inherited bone mar­row fail­ure syn­drome char­ac­ter­
ized by a con­stel­la­tion of extra-hema­to­poi­etic man­i­fes­ta­tions
with high var­i­abil­ity (exo­crine pan­cre­atic insuf­fi­ciency, hepa-
topathy, bony abnor­mal­i­ties, poor growth, and other endo­crine
man­i­fes­ta­tions).15,16 Neutropenia is pres­ent in approx­i­ma­tely 90%
of cases of SDS, with ane­mia, throm­bo­cy­to­pe­nia, and pan­cy­to­
pe­nia observed less fre­quently. SDS is caused, in most cases,
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/514/1851470/514warren.pdf by guest on 13 December 2021
by biallelic germline var­i­ants in SBDS.17 The esti­ma­tes of risk for
hema­to­poi­etic malig­nancy vary from 12% to 20% depending in
part on the age of the cohort.8,18,19 MDS/AML that arises in the set­
ting of SDS is char­ac­ter­ized by biallelic TP53 var­i­ants and com­
plex chro­mo­somal abnor­mal­i­ties.8,20,21
Clonal hema­to­poi­e­sis in con­gen­i­tal neutropenia
During hema­to­poi­e­sis, the acqui­si­tion of somatic var­i­ants over
time leads to the devel­op­ment of a genet­i­cally het­ero­ge­neous
blood cell pop­u­la­tion. Some of these var­i­ants con­fer a fit­ness
advan­tage, lead­ing to the expan­sion of hema­to­poi­etic stem/
pro­gen­i­tor cells (HSPCs) car­ry­ing cer­tain var­i­ants over time in
a pro­cess known as clonal hema­to­poi­e­sis.22,23 Variants in genes
impli­cated in mye­ loid malig­ nancy are the most fre­ quent in
age-related clonal hema­to­poi­e­sis (ARCH), includ­ing in the epi­ge­
netic mod­i­fi­ers DNTM3A, TET2, and ASXL1. Of note, in oth­er­wise
healthy per­sons, the pres­ence of clonal hema­to­poi­e­sis con­fers an
increased risk of devel­op­ing hema­to­poi­etic malig­nancy, depend-
ing on the gene involved, clone size, and whether mul­ti­ple var­i­
ants are pres­ent.24
Recent stud­ ies dem­ on­strate that the fre­ quency of clonal
hema­to­poi­e­sis in patients with con­gen­i­tal neutropenia is mark­
edly increased. We showed that in both SCN and SDS, the inci­
dence of clonal hema­to­poi­e­sis was mark­edly increased com­pared
with age-matched healthy donors.25 However, dis­ tinct genes
were respon­si­ble for this increase. In SCN, the increase in clonal
hema­to­poi­e­sis is entirely due to trun­cat­ing var­i­ants in CSF3R;
no increase in clonal hema­to­poi­e­sis due to other genes com­
monly mutated in clonal hema­to­poi­e­sis (eg, DNMT3A, TET2, and
ASXL1) was observed (Figure 1). In con­trast, in SDS, the increase
in clonal hema­to­poi­e­sis is due to a marked increase in TP53 var­i­
ants, with nearly 50% of patients car­ry­ing 1 or more TP53 var­i­ants.
Of note, TP53 var­i­ants were not seen in patients with SCN or in
age-matched healthy donors. Conversely, CSF3R var­i­ants were
not seen in patients with SDS or age-matched healthy donors.
What accounts for the strik­ing dif­fer­ence in clonal hema­to­
poi­e­sis between per­sons with SCN and SDS? Current evi­dence
points to dif­fer­ences in cel­lu­lar stress­ors in healthy per­sons and
per­sons with SCN or SDS. The CSF3R var­i­ants in patients with
SCN typ­i­cally pro­duce a trun­cated G-CSF recep­tor, which trans­
mits a sustained, dysregulated sig­nal in response to G-CSF.26-28
Studies in mice show that expres­ sion of a trun­ cated G-CSF
recep­ tor con­ fers a com­ pet­i­
tive advan­ tage to HSPCs that is
depen­dent on chronic G-CSF treat­ment.29 Thus, the high level
of G-CSF pres­ent in patients with SCN (either through endog­
e­nous pro­duc­tion or phar­ma­co­logic admin­is­tra­tion) may drive
the expan­sion of HSPCs car­ry­ing trun­ca­tion CSF3R var­i­ants
Clonal evo­lu­tion in con­gen­i­tal neutropenia  |  515
Figure 1. Graphical representation of the observed clonal hematopoiesis rates by somatic genetic alteration within healthy age-
matched controls, patients with SCN, or patients with SDS.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/514/1851470/514warren.pdf by guest on 13 December 2021

(Figure 2). Interestingly, a recent study exam­ined clonal hema­
to­poi­e­
sis in 185 patients with chronic idi­ o­
pathic neutropenia
(also termed idi­o­pathic neutropenia of unde­ter­mined sig­nif­i­
cance–neutropenia).30 Clonal hema­to­poi­e­sis was iden­ti­fied in
21 of 185 (11.4%) patients, which is sim­i­lar to that expected for age-
matched con­trols. The most fre­quent somatic var­i­ants observed
were in DNMT3A and TET2, as observed in ARCH. In sharp con­
trast to SCN, somatic CSF3R var­ i­
ants were not detected in
patients with chronic idi­o­pathic neutropenia. Whether this dif­
fer­ence is due to the sever­ity of neutropenia or sec­ond­ary to
cell-intrin­sic changes related to germline var­i­ants in SCN will
require fur­ther study.
Germline var­i­ants in SBDS that are pres­ent in most cases of
SDS result in impaired ribo­some bio­gen­e­sis, which in turn induces
p53 expres­sion and growth arrest.31-33 Pathogenic somatic var­
i­ants in TP53 in HSPCs are predicted to atten­u­ate this growth
arrest, resulting in their selec­tive expan­sion in patients with SDS.
An ele­gant study by Kennedy et al20 pro­vi­des fur­ther sup­port for

Figure 2. Modes of evolution from clonal hematopoiesis to malignant transformation. In the presence of high exogenous or endoge-
nous G-CSF, subclones with truncating CSF3R variants have a competitive advantage and expand over time. With continued pressure
from excessive G-CSF signaling, cooperating RUNX1 variants (*) can contribute to malignant transformation. In SDS, ribosome biogen-
esis stress selects for subclones that have adapted through acquisition of EIF6 or TP53 somatic variants. Chronic ribosome biogenesis
stress can lead to either continued adaptive clonal hematopoiesis or, in TP53 subclones clones that have acquired a second TP53
variant on the other allele, malignant transformation.

516  |  Hematology 2021  |  ASH Education Program

the key role of impaired ribo­some bio­gen­e­sis in the devel­op­
ment of clonal hema­to­poi­e­sis in patients with SDS. Specifically, in
addi­tion to TP53 var­i­ants, patients with SDS fre­quently develop
inactivating var­i­ants of EIF6. These var­i­ants par­tially res­cue the
ribo­some bio­gen­e­sis defect in SDS, thereby reduc­ing p53 acti­
va­tion. Together, these obser­va­tions illus­trate the impor­tance
of cel­lu­lar stress­ors in the devel­op­ment of clonal hema­to­poi­e­sis
(and likely mye­loid malig­nancy). In SCN, ele­vated lev­els of G-CSF
drive the expan­sion of HSPCs car­ry­ing trun­ca­tion CSF3R var­i­ants,
and in SDS, impaired ribo­some bio­gen­e­sis selects for HSPCs car­
ry­ing TP53 var­i­ants or var­i­ants (ie, EIF6) that reduce p53 acti­va­
tion (Figure 2).
Clonal evo­lu­tion to mye­loid malig­nancy in con­gen­i­tal
neutropenia
The pres­ence of clonal hema­to­poi­e­sis in patients with con­gen­i­
tal neutropenia does not inev­i­ta­bly lead to mye­loid malig­nancy.
Although the pres­ence of trun­cat­ing CSF3R var­i­ants is asso­ci­ated
with the devel­op­ment of mye­loid malig­nancy in patients with
SCN, it is not suf­fi­cient. Indeed, lon­gi­tu­di­nal sequenc­ing stud­ies
show that CSF3R-mutant clonal hema­to­poi­e­sis can per­sist for
many years with­out pro­gres­sion, even in cases with large CSF3R-
mutant clones.34 There is evi­dence that somatic loss-of-func­tion
var­i­ants of RUNX1 may be an early pro­gres­sion event, with acqui­
si­tion of chro­mo­somal abnor­mal­i­ties (mono­somy 7 and tri­somy 21)
late events.11 Consistent with this con­clu­sion, Dannenmann et al35
used SCN patient-derived induced plu­ rip­o­
tent stem cells that
expressed mutant CSF3R with our with­out mutant RUNX1 to show
that these var­i­ants coop­er­ate to block gran­u­lo­cytic dif­fer­en­ti­a­
tion, pro­duc­ing an AML-like phe­no­type.
Variants in TP53 are pres­ent in nearly 50% of patients with SDS
and may per­sist for many years with­out pro­gres­sion.20,25 Thus,
hav­ing a TP53 var­i­ant within the hema­to­poi­etic com­part­ment is
clearly not suf­fi­cient to drive leu­ke­mo­gen­e­sis. We sug­gest that
con­tin­ued ribo­some bio­gen­e­sis stress in SDS HSPCs car­ry­ing a
het­ero­zy­gous path­o­genic TP53 var­i­ant selects for clones that
inac­ti­vate the sec­ond TP53 allele (Figure 2). In con­trast, var­i­ants
in EIF6, by reduc­ing ribo­some bio­gen­e­sis stress, may be pro­tec­
tive. Consistent with this model, biallelic TP53 var­i­ants, but not
EIF6 var­i­ants, are very com­mon in the malig­nant clone in patients
with SDS who develop mye­loid malig­nancy.20
HSCT in con­gen­i­tal neutropenia
Because the devel­op­ment of mye­loid malig­nan­cies is the major
cause of death in con­gen­i­tal neutropenia, a key ques­tion in the
clin­i­cal man­age­ment is when to offer HSCT.
Current indi­ca­tions for HSCT in patients with SCN include
G-CSF refrac­to­ri­ness or devel­op­ment of MDS/AML, with some
inves­­ti­ga­tors also con­sid­er­ing high-dose G-CSF require­ment
(>15 µg/kg/d)36 or cer­tain high-risk ELANE path­o­genic var­i­ants
(p.G214R, p.C151Y, and p.C223ter).37,38 Studies of out­ comes in
patients with SCN after HSCT are con­founded by the dif­fer­ent
eras of trans­plant, con­di­tion­ing reg­i­mens, and donor sources.
Nonetheless, recent reports39,40 dem­on­strate that out­comes are
good for patients who undergo trans­plan­ta­tion with or with­out
mye­loid malig­nancy (Table 1). Also high­lighted in these stud­ies is
the remark­ably low relapse rate after HSCT in patients with SCN
who have mye­loid malig­nancy, espe­cially con­sid­er­ing that many
patients with SCN have active mye­loid malig­nancy at the time
of trans­plant. The rea­son for the very low relapse rate is uncer­
Prakash Singh Shekhawat
Table 1. Overall sur­vival fol­low­ing HSCT with or with­out
malig­nancy in mod­ern-era tri­als
Post-HSCT over­all sur­vival Without malig­nancy With malig­nancy
SCN
 *EBMT39 87% (n = 73) 79% (n = 14)
  *SCNIR (Euro­pean)40 78% (n = 27) 83% (n = 24)
SDS
  ^CIBMTR42 72% (n = 39) 15% (n = 13)
  ^SAAWP-EBMT41 71% (n = 61) 29% (n = 13)
 *SDSR43 — 46% (n = 15)
*Three-year overall survival; ^five-year overall survival.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/514/1851470/514warren.pdf by guest on 13 December 2021
CIBMTR, Center for International Blood and Marrow Transplant Research;
EBMT, Euro­pean Society for Bone Marrow Transplantation; SAAWP-EMBT,
Severe Aplastic Anemia Working Party of the EBMT; SCNIR, Severe Chronic
Neutropenia International Registry; SDSR, Shwachman-Diamond Syndrome
Registry of North America.
tain but sug­gests that SCN leu­ke­mic stem cells have reduced
fit­ness and/or increased sus­cep­ti­bil­ity to allo­genic immune cell
clear­ance.
HSCT for patients with SDS is gen­er­ally performed for bone
mar­row fail­ure or mye­loid malig­nancy. HSCT out­comes are infe­
rior to that that seen with patients with SCN and remains espe­
cially dis­mal for those patients who have already devel­oped a
mye­loid malig­nancy41-43 (Table 1). The rea­ sons for the infe­
rior
out­comes are likely mul­ti­fac­to­rial, includ­ing impaired bone mar­
row stro­mal cell func­tion, nonhematopoietic organ tox­ic­ity, and,
most impor­tant, the intrin­sic resis­tance of TP53 mutant mye­loid
malig­nan­cies to ther­apy. Together, these data sug­gest that an
effec­tive method to iden­tify patients with SDS at high risk of
pro­gres­sion to mye­loid malig­nancy for early HSCT is warranted.
However, this must be bal­anced by the like­li­hood of increased
trans­plant-related mor­bid­ity and mor­tal­ity in patients with SDS
who have under­ly­ing extra-hema­to­poi­etic organ dys­func­tion.
Timing of HSCT in SCN
There are a num­ber of clin­i­cal and molec­u­lar data to con­sider
regard­ing tim­ing of trans­plant for patients with SCN, includ­ing
(1) response to G-CSF, (2) results of sur­veil­lance bone mar­row
stud­ies to iden­tify dys­pla­sia or chro­mo­somal abnor­mal­i­ties, and
(3) somatic var­i­ant detec­tion. Prior stud­ies have dem­on­strated
that G-CSF dose and neu­tro­phil response are pre­dic­tors of pro­
gres­sion to mye­loid malig­nancy.7,19 Based on these data, the
French Severe Neutropenia Registry recommended that trans­
plant be offered to all­patients with SCN requir­ing high-dose
G-CSF (>15 µg/kg/d). Since insti­tut­ing this pol­icy in 2005, none
of the patients in this cohort have devel­oped a mye­loid malig­
nancy.36 Given the excel­lent HSCT out­comes for patients with
SCN who have mye­loid malig­nancy, a wait-and-see approach for
such patients also is rea­son­able. Of note, an inad­e­quate neu­
tro­phil response (ANC <1000 cells/mm3) to G-CSF, despite high
G-CSF dose, is a strong indi­ca­tion for HSCT.
It is com­mon prac­tice to per­form annual bone mar­row biopsy
eval­u­a­tions on patients with SCN to iden­tify chro­mo­somal abnor­
mal­i­ties and/or dys­pla­sia as early signs of malig­nant trans­for­ma­
tion.44-46 Historically, this prac­tice was based on expert opin­ion
rather than strong clin­i­cal or exper­i­men­tal evi­dence. However,
bone mar­row biopsy is an inva­sive and expen­sive pro­ce­dure, and
Clonal evo­lu­tion in con­gen­i­tal neutropenia  |  517
it also car­ries asso­ci­ated risks of seda­tion in youn­ger patients.
In our opin­ion, in light of recent data show­ing good out­comes
for patients with SCN who have mye­loid malig­nancy, the value
of annual sur­veil­lance bone mar­rows is debat­able. That said, a
bone mar­row biopsy for patients with SCN with wors­en­ing blood
counts or decreased respon­sive­ness to G-CSF is warranted.
There has been a recent increased use of gene panel sequenc­
ing to iden­tify somatic var­ia ­ nts in patients with SCN. An advan­
tage of this approach is the abil­ity to ana­lyze blood sam­ples,
obvi­at­ing the needed for an inva­sive bone mar­row biopsy. On
the other hand, the test is expen­sive and insur­ance cov­er­age
var­i­able. Interpreting the sequenc­ing results also is chal­leng­ing.
Truncating CSF3R var­i­ants in SCN are com­mon and may per­sist
for years or even dis­ap­pear with leu­ke­mic pro­gres­sion.34 Thus,
the value of trun­cat­ing CSF3R var­i­ants to iden­tify patients at risk
for pro­gres­sion is lim­ited. The pre­dic­tive value of RUNX1 var­i­ants
appears more prom­ is­
ing. As noted pre­ vi­
ously, the fre­
quency
of RUNX1 var­i­ants occur­ring in the set­ting of SCN-asso­ci­ated
MDS/AML is high (68%),11 and these var­ia ­ nts are often detect­
able in the blood or bone mar­row months prior to pro­gres­sion.
However, in con­trast to CSF3R var­i­ants, var­i­ants in RUNX1 appear
to be rare in patients with SCN with­out mye­loid malig­nancy. In
our study of clonal hema­to­poi­e­sis in 40 patients with SCN with­
out mye­loid malig­nancy, none had detect­able RUNX1 var­i­ants.25
Thus, RUNX1 var­i­ants may be pre­dic­tive of leu­ke­mic pro­gres­sion
in patients with SCN, although pro­spec­tive data supporting this
con­clu­sion are needed. Altogether, con­sid­er­ing the good out­
comes for patients with SCN who undergo trans­plan­ta­tion with
malig­nancy, cur­rent evi­dence does not sup­port the rou­tine use
of sur­ veil­lance somatic gene panel sequenc­ ing (out­
side of a
research study) to strat­ify patients for HSCT.
Timing of HSCT in SDS
One approach for early detec­ tion of clonal pro­ gres­
sion in
patients with SDS is annual bone mar­row sur­veil­lance. Abnormal
find­ings on cyto­ge­netic anal­y­sis are pres­ent in approx­i­ma­tely
50% of patients with SDS, the most com­mon being iso­chro­mo­
some 7 and del(20q).47 Isochromosome 7 results in an extra copy
of the SBDS gene, which, although typ­i­cally car­ry­ing a splice
site var­i­ant, is capa­ble of pro­duc­ing a small amount of full-length
SBDS pro­tein.48 del(20q) deletes 1 copy of EIF6, which, as dis-
cussed ear­lier, atten­u­ates the defec­tive ribo­some bio­gen­e­sis in
SDS.49 Perhaps not sur­pris­ingly, iso­chro­mo­some 7 and del(20q)
are asso­ci­ated with a lower risk of pro­gres­sion to mye­loid malig­
nancy. In con­trast, mono­somy 7 or del(7q) is asso­ci­ated with
rapid pro­gres­sion to mye­loid malig­nancy and is an indi­ca­tion for
trans­plant.50
The role for somatic gene panel sequenc­ing in SDS is uncer­
tain. Clonal hema­to­poi­e­sis due to TP53 var­i­ants is pres­ent in
approx­ i­
ma­ tely 50% of patients with SDS and increases with
age.25 The pres­ence of TP53-mutant clonal hema­to­poi­e­sis is not
asso­ci­ated with more severe cytopenias, and its pre­dic­tive value
for leu­ke­mic pro­gres­sion is likely to be low. Indeed, TP53 var­i­
ants can per­sist for sev­eral years with a sta­ble clone size with­out
pro­gres­sion to mye­loid malig­nancy. Kennedy et al20 showed that
most mye­loid malig­nan­cies aris­ing in patients with SDS carry
biallelic TP53 var­i­ants, rais­ing the pos­si­bil­ity that detec­tion of
hema­to­poi­etic clones car­ry­ing biallelic TP53 var­i­ants may be
more pre­dic­tive of leu­ke­mic pro­gres­sion. Indeed, the authors
used sin­gle-cell sequenc­ing in an infor­ma­tive patient with SDS to
518  |  Hematology 2021  |  ASH Education Program
show that a subclone car­ry­ing biallelic TP53 var­i­ants was detect­
able 4 years prior to malig­nant trans­for­ma­tion, expanded over
time, and was pres­ent in the AML clone. However, sin­gle-cell
sequenc­ing is expen­sive, labor inten­sive, and not yet clin­i­cally
avail­­able, lim­it­ing its cur­rent use to research appli­ca­tions.
CLINICAL CASES (Con­tin­ued)
Case 1
This patient with ELANE-mutated SCN was noted to have a high-
fre­quency CSF3R path­o­genic var­i­ant and low-fre­quency DNTM3A
and ASXL1 var­i­ants at age 45 years. As we have discussed, the
pre­dic­tive value of CSF3R var­i­ants for leu­ke­mic pro­gres­sion is
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/514/1851470/514warren.pdf by guest on 13 December 2021
low. Likewise, the sig­nif­i­cance of the low-fre­quency DNTM3A or
ASXL1 var­i­ants in this older patient is unclear. We favor con­tin­ued
obser­va­tion of this patient, reserv­ing HSCT for the devel­op­ment
of MDS/AML. We would not rec­om­mend repeat somatic gene
panel sequenc­ing out­side of a research study. Ultimately, this
patient devel­oped AML at age 56 years, which was suc­cess­fully
treated by bone mar­row trans­plan­ta­tion.
Case 2
This patient with SDS presented with mul­ti­ple low-abun­dance
TP53 var­i­ants by somatic gene panel sequenc­ing but ini­tially
had no dys­pla­sia or chro­mo­somal abnor­mal­ity. The pre­dic­tive
value for leu­ke­mic pro­gres­sion of TP53 detected by bulk somatic
DNA sequenc­ing is likely low. It is also unknown whether, sim­
i­lar to patients with ARCH, the pres­ence of mul­ti­ple var­i­ants
may con­fer a higher risk of trans­for­ma­tion. We would favor con­
tin­ued obser­ va­
tion in this patient with annual bone mar­ row
eval­u­a­tion for dys­pla­sia and chro­mo­somal abnor­mal­i­ties. The
value of repeat bulk somatic gene panel sequenc­ing to mon­i­
tor TP53-mutant clone size is uncer­tain and not recommended
out­side of a research study. This patient devel­oped AML at age
22 years, and the malig­nant clone dem­on­strated biallelic TP53
var­i­ants detect­able 4 years before trans­for­ma­tion. Although a
pro­spec­tive study is required, this case high­lights the poten­tial
for sin­gle-cell sequenc­ing to detect early biallelic TP53 events
and serve as a bio­marker for con­sid­er­ing early HSCT.
Case 3
This 12-year-old patient with ELANE-mutant SCN was noted to
have decreased G-CSF respon­ sive­
ness and recur­ rent cel­lu­
li­
tis. Although results of cyto­ge­netic and somatic gene panel
sequenc­ing were neg­a­tive, we would favor early HSCT in this
patient. Ultimately, the patient under­went matched unre­lated
donor hema­to­poi­etic stem cell trans­plant at age 13 years and
over 10 years later is alive and well with­out trans­plant-related
com­pli­ca­tions.
Summary and future direc­tions
Recent advances in our under­stand­ing of dis­ease path­o­gen­e­sis
and clonal evo­lu­tion have affected the clin­i­cal man­age­ment of
patients with con­gen­i­tal neutropenia. Cellular stress­ors play a
key role in clonal evo­lu­tion and depend, in part, on the under­ly­
ing genetic eti­­ol­ogy of con­gen­i­tal neutropenia. Thus, iden­ti­fy­ing
the genetic cause of neutropenia is desir­able, and we rec­om­
mend panel-based test­ing rather than sin­gle-gene test­ing as part
of the diag­nos­tic eval­u­a­tion. An effec­tive sur­veil­lance strat­egy
to iden­tify patients at risk for trans­for­ma­tion to mye­loid malig­
nancy is needed, par­tic­u­larly for patients with SDS. Surveillance
strat­e­gies for patients with SCN will need to account for recent
improve­ments in HSCT out­comes even in the pres­ence of active
mye­loid malig­nancy.
Many open ques­tions in the field need to be addressed and
may affect clin­i­cal man­age­ment of patients with con­gen­i­tal neu-
tropenia. Prospective tri­als are needed to deter­mine whether
RUNX1 or biallelic TP53 var­i­ants are pre­dic­tive of leu­ke­mic trans­
for­ma­tion in SCN or SDS, respec­tively. The fre­quency and clin­i­cal
impli­ca­tions of clonal hema­to­poi­e­sis due to chro­mo­somal abnor­
mal­i­ties in patients with CN are unknown. Prospective studies to
detect somatic single nucleotide, small insertions/deletions, or
copy number alterations in patients with CN are needed. Finally,
it will be impor­tant to define risks of mye­loid malig­nancy and
pat­terns of clonal evo­lu­tion in the expanding genetic sub­types
of con­gen­i­tal neutropenia.
Conflict-of-inter­est dis­clo­sure
Julia T. Warren: no com­pet­ing finan­cial inter­ests to declare.
Dan­iel C. Link: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Julia T. Warren: no off-label drug use.
Dan­iel C. Link: no off-label drug use.
Correspondence
Dan­iel C. Link, Division of Oncology, Department of Inter-
nal Medicine, Washington University School of Medicine, 660
S. Euclid Ave, Campus Box 8007, St. Louis, MO 63110; e-mail:
danielclink@wustl​­.edu.
References
1. Skokowa J, Dale DC, Touw IP, Zeidler C, Welte K. Severe con­gen­i­tal neutro-
penias. Nat Rev Dis Primers. 2017;3:17032.
2. Boztug K, Appaswamy G, Ashikov A, et al. A syn­drome with con­gen­i­tal neu-
tropenia and muta­tions in G6PC3. N Engl J Med. 2009;360(1):32-43.
3. Horwitz M, Benson KF, Person RE, Aprikyan AG, Dale DC. Mutations in ELA2,
encoding neu­tro­phil elas­tase, define a 21-day bio­log­i­cal clock in cyclic
haematopoiesis. Nat Genet. 1999;23(4):433-436.
4. Klein C, Grudzien M, Appaswamy G, et  al. HAX1 defi­ciency causes auto­
so­mal reces­sive severe con­gen­i­tal neutropenia (Kostmann dis­ease). Nat
Genet. 2007;39(1):86-92.
5. Xia J, Bolyard AA, Rodger E, et al. Prevalence of muta­tions in ELANE, GFI1,
HAX1, SBDS, WAS and G6PC3 in patients with severe con­gen­i­tal neutrope-
nia. Br J Haematol. 2009;147(4):535-542.
6. Dale DC, Bonilla MA, Davis MW, et al. A ran­dom­ized con­trolled phase III trial
of recom­bi­nant human granulocyte col­ony-stim­u­lat­ing fac­tor (filgrastim)
for treat­ment of severe chronic neutropenia. Blood. 1993;81(10):2496-2502.
7. Rosenberg PS, Zeidler C, Bolyard AA, et al. Stable long-term risk of leu­kae­
mia in patients with severe con­gen­i­tal neutropenia maintained on G-CSF
ther­apy. Br J Haematol. 2010;150(2):196-199.
8. Donadieu J, Leblanc T, Bader Meunier B, et  al.; French Severe Chronic
Neutropenia Study Group. Analysis of risk fac­tors for myelodysplasias, leu­
ke­mias and death from infec­tion among patients with con­gen­i­tal neutro-
penia. Experience of the French Severe Chronic Neutropenia Study Group.
Haematologica. 2005;90(1):45-53.
9. Furutani E, Newburger PE, Shimamura A. Neutropenia in the age of genetic
test­ing: advances and chal­lenges. Am J Hematol. 2019;94(3):384-393.
10. Ley TJ, Miller C, Ding L, et al; Cancer Genome Atlas Research Network.
Genomic and epigenomic land­scapes of adult de novo acute mye­loid leu­
ke­mia. N Engl J Med. 2013;368(22):2059-2074.
11. Skokowa J, Steinemann D, Katsman-Kuipers JE, et  al. Cooperativity of
RUNX1 and CSF3R muta­tions in severe con­gen­i­tal neutropenia: a unique
path­way in mye­loid leu­ke­mo­gen­e­sis. Blood. 2014;123(14):2229-2237.
Prakash Singh Shekhawat
12. Freedman MH, Bonilla MA, Fier C, et  al. Myelodysplasia syn­drome and
acute mye­loid leu­ke­mia in patients with con­gen­i­tal neutropenia receiv­ing
G-CSF ther­apy. Blood. 2000;96(2):429-436.
13. Kalra R, Dale D, Freedman M, et al. Monosomy 7 and acti­vat­ing RAS muta­
tions accom­pany malig­nant trans­for­ma­tion in patients with con­gen­i­tal
neutropenia. Blood. 1995;86(12):4579-4586.
14. Link DC, Kunter G, Kasai Y, et al. Distinct pat­terns of muta­tions occur­ring in
de novo AML ver­sus AML aris­ing in the set­ting of severe con­gen­i­tal neutro-
penia. Blood. 2007;110(5):1648-1655.
15. Burroughs L, Woolfrey A, Shimamura A. Shwachman-Diamond syn­drome: a
review of the clin­i­cal pre­sen­ta­tion, molec­u­lar path­o­gen­e­sis, diag­no­sis, and
treat­ment. Hematol Oncol Clin North Am. 2009;23(2):233-248.
16. Myers KC, Bolyard AA, Otto B, et  al. Variable clin­ i­
cal pre­ sen­ta­
tion of
Shwachman-Diamond syn­ drome: update from the North Amer­ i­
can
Shwachman-Diamond Syndrome Registry. J Pediatr. 2014;164(4):866-870.
17. Shimamura A. Shwachman-Diamond syn­ drome. Semin Hematol. 2006;
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/514/1851470/514warren.pdf by guest on 13 December 2021
43(3):178-188.
18. Hashmi SK, Allen C, Klaassen R, et al. Comparative anal­y­sis of Shwachman-
Diamond syn­drome to other inherited bone mar­row fail­ure syn­dromes and
geno­type-phe­no­type cor­re­la­tion. Clin Genet. 2011;79(5):448-458.
19. Rosenberg PS, Alter BP, Bolyard AA, et al; Severe Chronic Neutropenia
International Registry. The inci­dence of leu­ke­mia and mor­tal­ity from sep­sis
in patients with severe con­gen­i­tal neutropenia receiv­ing long-term G-CSF
ther­apy. Blood. 2006;107(12):4628-4635.
20. Kennedy AL, Myers KC, Bowman J, et al. Distinct genetic path­ways define
pre-malig­nant ver­sus com­pen­sa­tory clonal hema­to­poi­e­sis in Shwachman-
Diamond syn­drome. Nat Commun. 2021;12(1):1334.
21. Lindsley RC, Saber W, Mar BG, et al. Prognostic muta­tions in myelodysplas-
tic syn­drome after stem-cell trans­plan­ta­tion. N Engl J Med. 2017;376(6):536-
547.
22. Genovese G, Kähler AK, Handsaker RE, et  al. Clonal hema­to­poi­e­sis and
blood-can­ cer risk inferred from blood DNA sequence. N Engl J Med.
2014;371(26):2477-2487.
23. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hema­to­poi­e­sis
asso­ci­ated with adverse out­comes. N Engl J Med. 2014;371(26):2488-2498.
24. Warren JT, Link DC. Clonal hema­to­poi­e­sis and risk for hema­to­logic malig­
nancy. Blood. 2020;136(14):1599-1605.
25. Xia J, Miller CA, Baty J, et al. Somatic muta­tions and clonal hema­to­poi­e­sis
in con­gen­i­tal neutropenia. Blood. 2018;131(4):408-416.
26. Hermans MH, Antonissen C, Ward AC, Mayen AE, Ploemacher RE, Touw IP.
Sustained recep­tor acti­va­tion and hyperproliferation in response to gran-
ulocyte col­ony-stim­u­lat­ing fac­tor (G-CSF) in mice with a severe con­gen­
i­tal neutropenia/acute mye­loid leu­ke­mia-derived muta­tion in the G- CSF
recep­tor gene. J Exp Med. 1999;189(4):683-692.
27. Hermans MH, Ward AC, Antonissen C, Karis A, Löwenberg B, Touw IP. Per-
turbed granulopoiesis in mice with a targeted muta­tion in the granulocyte
col­ony-stim­u­lat­ing fac­tor recep­tor gene asso­ci­ated with severe chronic
neutropenia. Blood. 1998;92(1):32-39.
28. McLemore ML, Poursine-Laurent J, Link DC. Increased granulocyte col­ony-
stim­u­lat­ing fac­tor respon­sive­ness but nor­mal rest­ing granulopoiesis in
mice car­ry­ing a targeted granulocyte col­ony-stim­u­lat­ing fac­tor recep­tor
muta­tion derived from a patient with severe con­gen­i­tal neutropenia. J Clin
Invest. 1998;102(3):483-492.
29. Liu F, Kunter G, Krem MM, et al. Csf3r muta­tions in mice con­fer a strong
clonal HSC advan­tage via acti­va­tion of Stat5. J Clin Invest. 2008;118(3):946-
955.
30. Tsaknakis G, Gallì A, Papadakis S, et al. Incidence and prog­no­sis of clonal
hema­to­poi­e­sis in patients with chronic idi­o­pathic neutropenia. Blood.
2021;138(14):1249-1257.
31. Menne TF, Goyenechea B, Sánchez-Puig N, et al. The Shwachman-Bodian-
Diamond syn­drome pro­tein medi­ates trans­la­tional acti­va­tion of ribo­somes
in yeast. Nat Genet. 2007;39(4):486-495.
32. Wong CC, Traynor D, Basse N, Kay RR, Warren AJ. Defective ribo­some
assem­bly in Shwachman-Diamond syn­drome. Blood. 2011;118(16):4305-4312.
33. Zambetti NA, Bindels EM, Van Strien PM, et al. Deficiency of the ribo­some
bio­gen­e­sis gene Sbds in hema­to­poi­etic stem and pro­gen­i­tor cells causes
neutropenia in mice by atten­u­at­ing lin­e­age pro­gres­sion in mye­lo­cytes.
Haematologica. 2015;100(10):1285-1293.
34. Touw IP. Game of clones: the geno­mic evo­lu­tion of severe con­gen­i­tal neu-
tropenia. Hematol Am Soc Hematol Educ Program. 2015;2015:1-7.
35. Dannenmann B, Klimiankou M, Oswald B, et al. iPSC mod­el­ing of stage-spe­
cific leu­ke­mo­gen­e­sis reveals BAALC as a key onco­gene in severe con­gen­i­
tal neutropenia. Cell Stem Cell. 2021;28(5):906.e6-922.e6.
Clonal evo­lu­tion in con­gen­i­tal neutropenia  |  519
36. Rotulo GA, Beaupain B, Rialland F, et al. HSCT may lower leu­ke­mia risk in
ELANE neutropenia: a before-after study from the French Severe Congeni-
tal Neutropenia Registry. Bone Marrow Transplant. 2020;55(8):1614-1622.
37. Dale DC, Makaryan V, Kelley ML, et al. Termination and frame­shift muta­tions
in ELANE are asso­ci­ated with adverse out­comes in patients with severe
chronic neutropenia. Blood. 2016;128(22):1326.
38. Makaryan V, Zeidler C, Bolyard AA, et  al. The diver­sity of muta­tions and
clin­i­cal out­comes for ELANE-asso­ci­ated neutropenia. Curr Opin Hematol.
2015;22(1):3-11.
39. Fioredda F, Iacobelli S, van Biezen A, et al; Severe Aplastic Anemia the
Inborn Error, and the Pediatric Disease Working Parties of the Euro­pean
Society for Blood and Bone Marrow Transplantation (EBMT) and Stem Cell
Transplant for Immunodeficiencies in Europe (SCETIDE). Stem cell trans­
plan­ta­tion in severe con­gen­i­tal neutropenia: an anal­y­sis from the Euro­pean
Society for Blood and Marrow Transplantation. Blood. 2015;126(16):1885-
1892; quiz 1970.
40. Zeidler C, Nickel A, Sykora K-W, Welte K. Improved out­come of stem cell
trans­plan­ta­tion for severe chronic neutropenia with or with­out sec­ond­ary
leu­ke­mia: a long-term anal­y­sis of Euro­pean data for more than 25 years by
the SCNIR. Blood. 2013;122(21):3347.
41. Cesaro S, Pillon M, Sauer M, et  al. Long-term out­come after allo­ge­neic
hema­to­poi­etic stem cell trans­plan­ta­tion for Shwachman-Diamond syn­
drome: a ret­ ro­
spec­tive anal­y­sis and a review of the lit­ er­
at­ure by the
Severe Aplastic Anemia Working Party of the Euro­pean Society for Blood
and Marrow Transplantation (SAAWP-EBMT). Bone Marrow Transplant.
2020;55(9):1796-1809.
42. Myers K, Hebert K, Antin J, et  al. Hematopoietic stem cell trans­plan­ta­
tion for Shwachman-Diamond syn­drome. Biol Blood Marrow Transplant.
2020;26(8):1446-1451.
43. Myers KC, Furutani E, Weller E, et  al. Clinical fea­tures and out­comes of
patients with Shwachman-Diamond syn­drome and myelodysplastic syn­
drome or acute mye­loid leu­kae­mia: a multicentre, ret­ro­spec­tive, cohort
study. Lancet Haematol. 2020;7(3):e238-e246.
520  |  Hematology 2021  |  ASH Education Program
44. Fioredda F, Calvillo M, Bonanomi S, et al; Neutropenia Committee of the
Marrow Failure Syndrome Group of the AIEOP (Associazione Italiana Ema-
to-Oncologia Pediatrica). Congenital and acquired neutropenias con­sen­
sus guide­lines on ther­apy and fol­low-up in child­hood from the Neutropenia
Committee of the Marrow Failure Syndrome Group of the AIEOP (Associazi-
one Italiana Emato-Oncologia Pediatrica). Am J Hematol. 2012;87(2):238-
243.
45. Newburger PE, Dale DC. Evaluation and man­age­ment of patients with iso­
lated neutropenia. Semin Hematol. 2013;50(3):198-206.
46. Porter CC, Druley TE, Erez A, et  al. Recommendations for sur­ veil­
lance
for chil­dren with leu­ke­mia-predisposing con­di­tions. Clin Cancer Res.
2017;23(11):e14-e22.
47. Maserati E, Pressato B, Valli R, et  al. The route to devel­op­ment of myel-
odysplastic syn­drome/acute mye­loid leu­kae­mia in Shwachman-Diamond
syn­drome: the role of age­ing, kar­yo­type insta­bil­ity, and acquired chro­mo­
some anom­a­lies. Br J Haematol. 2009;145(2):190-197.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/514/1851470/514warren.pdf by guest on 13 December 2021
48. Minelli A, Maserati E, Nicolis E, et al. The iso­chro­mo­some i(7)(q10) car­ry­ing
c.258 + 2t>c muta­tion of the SBDS gene does not pro­mote devel­op­ment of
mye­loid malig­nan­cies in patients with Shwachman syn­drome. Leukemia.
2009;23(4):708-711.
49. Pressato B, Valli R, Marletta C, et al. Deletion of chro­mo­some 20 in bone
mar­row of patients with Shwachman-Diamond syn­drome, loss of the EIF6
gene and benign prog­no­sis. Br J Haematol. 2012;157(4):503-505.
50. Donadieu J, Fenneteau O, Beaupain B, et al; Associated inves­ti­ga­tors of
the French Severe Chronic Neutropenia Registry. Classification of and
risk fac­tors for hema­ to­
logic com­ pli­
ca­
tions in a French national cohort
of 102 patients with Shwachman-Diamond syn­ drome. Haematologica.
2012;97(9):1312-1319.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000286
 PERIOPERATIVE CONSULTATIVE HEMATOLOGY: CAN YOU CLEAR MY PATIENT FOR SURGERY ?

Perioperative consultative hematology:

can you clear my patient for a procedure?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/521/1851845/521burnett.pdf by guest on 13 December 2021

Allison Elaine Burnett,1 Bishoy Ragheb,2 and Scott Kaatz3
University of New Mexico, Health Sciences Center, College of Pharmacy, Albuquerque, NM; 2Tennessee Valley Health Systems (TVHS) Veterans
1

Affairs, Nashville, TN; and 3Henry Ford Hospital, Detroit, MI

Periprocedural management of antithrombotics is a common but challenging clinical scenario that renders patients
vulnerable to potential adverse events such as bleeding and thrombosis. Over the past decade, periprocedural anti-
thrombotic approaches have changed considerably with the advent of direct oral anticoagulants (DOACs), as well as
a paradigm shift away from bridging in many warfarin patients. Successfully navigating this high-risk period relies on a
number of individualized patient assessments conducted within a framework of standardized, systematic approaches.
It also requires a thorough understanding of antithrombotic pharmacokinetics, multidisciplinary coordination of care,
and comprehensive patient education and empowerment. In this article, we provide clinicians with a practical, stepwise
approach to periprocedural management of antithrombotic agents through case-based examples of relevant clinical
scenarios.

LEARNING OBJECTIVES
• Explain significant differences between perioperative management of DOAC and warfarin, including the rationale
for these differences
• Determine if temporary interruption of anticoagulant therapy is needed through evaluation of factors, including
bleeding and thrombotic potential of the procedure and the patient
• Identify warfarin patients who do and do not warrant consideration for perioperative bridging
• Develop safe, effective perioperative anticoagulation plans for DOAC and warfarin patients based on existing evi-
dence and expert consensus

Introduction vulnerability to medication discrepancies, therapeutic

Although exact numbers are not known, it is estimated that
6 to 8 million people in the United States are prescribed
oral anticoagulation (OAC) with either warfarin or a direct
oral anticoagulant (DOAC).1,2 Approximately 10% to 15% of
these patients will require temporary interruption of anti-
coagulation for surgery or an invasive procedure, which
equates to 600 000 to 1 200 000 interruptions annually.3,4
The periprocedural period is a time of significant risk for
anticoagulation patients for a multitude of reasons, includ-
ing but not limited to the following:
• A number of complex individualized assessments must
to be made to estimate bleeding and thrombotic risks
of both the procedure and the patient, which are largely
based on expert consensus.5,6
• It is estimated that surgical patients will experience up to
15 transitions of care.7 Each transition is associated with
overlap, failure to resume anticoagulation, communica-
tion errors, and missed coordination of follow-up.8,9
• Anecdotal experience suggests providers lack familiar-
ity with the pharmacokinetics of the DOACs, which differ
dramatically from those of warfarin. It is not uncom-
mon in clinical practice to discover DOAC patients with
periprocedural plans with inappropriately prolonged
hold times and/or overlapping therapy with low molec-
ular weight heparin (LMWH) as providers try to manage
these agents in a similar manner as they would warfarin.
The increased risk for adverse events in the periproce-
dural period has prompted a number of quality improve-
ment and patient safety initiatives and shifts in practice.
• In 2019, the Joint Commission revised its National
Patient Safety Goal (NPSG) 03.05.01 pertaining to anti-

Prakash Singh Shekhawat

Perioperative management of antithrombotics | 521
 co­ag­u­lants to include a new ele­ment of per­for­mance (EP3)
call­ing for hos­pi­tals to use approved pro­to­cols and evi­dence-
based prac­tice guide­lines for periprocedural man­age­ment of
all­ patients tak­ing oral anti­co­ag­u­lants.10
•  The Centers for Medi­care & Med­ic­aid Services offers merit-
based incen­tive pay­ments to phy­si­cians who pro­vide doc­
u­men­ta­tion of periprocedural anticoagulation man­age­ment
plans, includ­ing tim­ing of inter­rup­tion, man­age­ment of con­
com­i­tant antithrombotics, bridg­ing (if indi­cated), lab­o­ra­tory
mea­sure­ments, tim­ing of resump­tion, and dis­cus­sion of plan
with the patient.11
•  An increas­ing num­ber of health sys­tems are using clin­i­cal phar­
ma­ cists and anticoagulation stew­ ard­
ship pro­
grams to opti­
mize devel­op­ment and appli­ca­tion of guide­line-recommended
periprocedural plans.12,13
It is impor­tant to acknowl­edge that perioperative con­sul­
ta­
tion may be accom­ pa­nied by dif­fer­
ences of opin­ ion and
approaches between involved dis­ci­plines that requires thor­
ough dis­cus­sion and con­sen­sus build­ing. In addi­tion, a patient
who is cleared for sur­gery should not be deemed “risk-free”
from adverse events. Multidisciplinary col­ lab­or­a­
tion before,
dur­ing, and after sur­gery to assess for and mit­ig­ ate risk as much
as pos­si­ble is essen­tial for opti­mized patient care.
Atrial fibril­la­tion
CLINICAL CASE
A 65-year-old female patient is to undergo an elec­tive right hip
replace­ment. Her med­i­cal his­tory includes a bioprosthetic aor­
tic valve, nonvalvular atrial fibril­la­tion, hyper­ten­sion, and dia­be­
tes. She is referred for periprocedural DOAC rec­om­men­da­tions.
Laboratory val­ues, includ­ing renal and liver func­tion, are nor­mal.
Step 1: Does anticoagulation need to be interrupted?
Bleeding risk of the pro­ce­dure
The need for OAC inter­ rup­tion is deter­ mined pri­ mar­ily by the
bleed­ing risk of the pro­ce­dure. Unfortunately, a high-qual­ity, evi­
dence-based schema to cat­e­go­rize pro­ce­dural bleed­ing risk has
not been well established and has led to dif­fer­ences across guide­
lines and var­i­a­tions in prac­tice.14-16 Recently, the International Soci-
ety on Thrombosis and Haemostasis issued guid­ance on this with
the intent of pro­mot­ing more stan­dard­ized approaches.17 In our
prac­tice, we also often refer to the very com­pre­hen­sive pro­ce­
dural bleed­ing risk appen­dix published with the 2017 ACC Expert
Consensus Decision Pathway for Periprocedural Management of
Anticoagulation in Patients With Nonvalvular Atrial Fibrillation.18 In
Table 1, we have pro­vided a list (nonexhaustive) of pro­ce­dures with
min­i­mal bleed­ing risk that likely do not require inter­rup­tion that
can greatly sim­plify man­age­ment and mit­i­gate risk for harm. For
other pro­ce­dures, we sug­gest cli­ni­cians use existing pro­ce­dure
cat­e­go­ri­za­tion tools as a frame­work for dis­cus­sion with sur­geons
and other interventionalists when devel­op­ing a perioperative plan.
Bleeding risk of the patient
The intrin­sic bleed­ing risk of the patient should also be con­
sid­
ered dur­ ing periprocedural plan­ ning. Characteristics that
we rou­tinely con­sider when assessing patient bleed­ing risk in-
522  |  Hematology 2021  |  ASH Education Program
Table 1. Minor sur­ger­ies or pro­ce­dures that may not require
inter­rup­tion of OAC
•  Minor den­tal (eg, 1-2 tooth extrac­tion, cleaning)
•  Minor der­ma­to­logic or cuta­ne­ous
• Cataract
•  Cardiac implant­able devices (pace­mak­ers, defi­bril­la­tors)
•  Cardiac abla­tions, cardioversion, elec­tro­phys­i­­ol­ogy stud­ies
•  Endovascular pro­ce­dures (eg, angio­plasty)
•  Endoscopy with­out resec­tion or biopsy
•  Intramuscular vac­ci­na­tion
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/521/1851845/521burnett.pdf by guest on 13 December 2021
•  Percutaneous cor­o­nary inter­ven­tions (radial approach)
clude throm­bo­cy­to­pe­nia, renal dys­func­tion with ure­mia, sig­nif­i­
cant hepatic impair­ment with base­line coagulopathy, his­tory of
bleed­ing, and con­com­i­tant med­i­ca­tions such as antiplatelets or
non­ste­roi­dal anti-inflam­ma­tory drugs. Although many of these
are not mod­ i­
fi­
able, being aware of their pres­ ence may help
antic­i­pate and address com­pli­ca­tions.
Minimal bleed risk pro­ce­dures
Studies have shown that many low bleed­ing risk minor sur­ger­
ies or pro­ce­dures (which con­sti­tute up to 20% of cases) can
be safely done with­out war­fa­rin inter­rup­tion.19 In addi­tion, ran­
dom­ized con­trolled tri­als and meta-ana­ly­ses have shown that
war­fa­rin inter­rup­tion with or with­out LMWH bridg­ing leads to
more adverse events than no war­fa­rin inter­rup­tion.20-23 There is
less cer­tainty with con­tinu­ing DOACs around minor pro­ce­dures.
However, a grow­ing body of evi­dence sug­gests unin­ter­rupted
DOACs may be rea­son­able24-26 and pos­si­bly safer than unin­ter­
rupted war­fa­rin in many pro­ce­dures (Table 1).27,28
Our patient is under­go­ing a hip replace­ment that we would
con­sider a high bleed risk pro­ce­dure, and DOAC should be
interrupted.
Step 2: If OAC inter­rup­tion is needed, does the patient
require bridg­ing?
DOACs
There are sig­nif­i­cant phar­ma­co­ki­netic dif­fer­ences between DOACs
and war­fa­rin, and thus their perioperative man­age­ment requires
dif­fer­ent approaches.26,29 In patients with nor­mal renal func­tion,
DOACs have a much shorter half-life than war­fa­rin (approx­i­ma­
tely 12 hours vs approx­i­ma­tely 40 hours, respec­tively) and much
faster off­set. Thus, with­hold­ing DOACs for a prolonged period
of time (eg, 5 days as is rou­tinely done with war­fa­rin) poten­tially
places the patient at risk for a throm­botic event. Also, given
the rapid off­set of DOACs, it is not nec­es­sary to inter­rupt them
preprocedurally and replace with LMWH. In the postprocedure
set­ting, it is imper­a­tive to rec­og­nize the faster onset of DOAC
anti­co­ag­u­lant action (approx­i­ma­tely 3 hours) com­pared with
war­ fa­rin (approx­ i­
ma­tely 4-5 days) and the need for care­ fully
timed resump­tion to mit­i­gate bleed­ing risk. Very impor­tantly,
DOACs should never be overlapped with par­en­teral anti­co­ag­u­
lants, such as LMWH, as this is not nec­es­sary due to their rapid
onset and stud­ies show­ing sig­nif­i­cantly increased bleed­ing.3,4,25
 The Perioperative Anticoagulation Use for Surgery Evaluation
(PAUSE) cohort study30 used a sim­ ple DOAC inter­ rup­
tion and
resump­tion pro­to­col with­out any bridg­ing in patients with atrial
fibril­la­tion under­go­ing sur­gi­cal pro­ce­dures. This was based in part
on suc­cess­ful use of a sim­i­lar approach in an ear­lier pro­spec­tive
mul­ti­cen­ter trial eval­u­at­ing perioperative man­age­ment of dabiga-
tran.31 Timing of inter­rup­tion and resump­tion was based on DOAC
phar­ma­co­ki­netic prop­er­ties, pro­ce­dural bleed risk, and renal func­
tion. The inves­ti­ga­tors hoped to show a major bleed­ing rate of less
than 2.0% and a stroke or tran­sient ische­mic attack (TIA) rate of
less than 1.5% with 95% cer­tainty. Of note, this study used num­
ber of days and not hours for inter­rup­tion and resump­tion tim­ing,
as shown in Table 2. Because dabigatran is pri­mar­ily elim­i­nated
through the kid­neys (80%), cre­at­i­nine clear­ance was esti­mated
(using the Cockcroft-Gault equa­tion and actual body weight), and
hold times were dou­bled in dabigatran patients with an esti­mated
clear­ance of less than 50 mL/min. Results showed with 95% con­fi­
dence that bleed­ing rates were not greater than 2.00%, 1.73%, and
2.65% with apixaban, dabigatran, and rivaroxaban, respec­tively,
and arte­ rial throm­bo­em­ bo­lism was not greater than 1.5% with
any DOAC. In addi­tion, it was shown that more than 90% of DOAC
patients col­lec­tively had lit­tle to no resid­ual anti­co­ag­u­lant effect
from more than 30 ng/mL (which has been suggested as an accept­
able pre­op­er­a­tive plasma con­cen­tra­tion32), pre­clud­ing the need
for rou­tine pre­op­er­a­tive lab­o­ra­tory assess­ment when the PAUSE
pro­to­col is followed. The Scientific and Standardization Commit-
tee of the International Society on Thrombosis and Haemostasis
sug­gests that if a pre­op­er­a­tive quan­ti­ta­tive DOAC level is 30 ng/mL
or less, it is rea­son­able to pro­ceed with an inva­sive pro­ce­dure.33 To
our knowl­edge, there are no data to sug­gest enhanced util­ity of
mea­sur­ing DOAC con­cen­tra­tions before elec­tive pro­ce­dures com­
pared with a sim­ple stan­dard­ized, phar­ma­co­ki­netic-based pro­to­
col as was used in the PAUSE trial. Urgent or emer­gent pro­ce­dures
pose a chal­lenge, and there may be a role for DOAC mea­sure­ment
in select sit­u­a­tions. However, it is crit­i­cal to bear in mind that these
quan­ti­ta­tive DOAC assays are not avail­­able in many hos­pi­tals, and
if they are, turn­around times may pre­clude any util­ity.
An unanswered ques­tion is what to do in patients with a his­
tory of venous throm­bo­em­bo­lism (VTE), and we extrap­o­late
the PAUSE study tim­ing of inter­rup­tion and resump­tion to these
patients with con­fi­dence in the major bleed­ing rates but uncer­
tainty in VTE recur­rence.
An addi­tional point is the PAUSE study dif­fers slightly from the
rec­om­men­da­tions by the Amer­i­can Society of Regional Anes-
thesia, which rec­om­mend hold­ing DOACs for 72 hours prior to
Table 2. Periprocedural DOAC inter­rup­tion pro­to­col from PAUSE study
Characteristic Apixaban Rivaroxaban
Preprocedural inter­rup­tion Days Days
  Low bleed­ing risk pro­ce­dure 1 1
  High bleed­ing risk pro­ce­dure 2 2
Postprocedural resump­tion
  Low bleed­ing risk pro­ce­dure 1 1
  High bleed­ing risk pro­ce­dure 2-3 2-3
For pro­ce­dural bleed­ing risk strat­i­fi­ca­tion, we sug­gest using soci­ety-based guid­ance in the appen­dix to the ACC con­sen­sus deci­sion path­way18 or
cat­e­go­ri­za­tion from PAUSE study.30
CrCl, cre­at­i­nine clear­ance.
Prakash Singh Shekhawat
neuraxial anes­the­sia and lon­ger for dabigatran if the cre­at­i­nine
clear­ance is low.34
• We would hold the DOAC for 2 days before sur­gery and restart
the DOAC on post­op­er­a­tive day 2 or 3.
•  In the postprocedural interim until ther­a­peu­tic anticoagulation
is resumed, we would ensure ade­quate pro­phy­laxis (eg, enox-
aparin 40 mg subcutaneously [SQ ] once daily) to pre­vent post­
op­er­a­tive VTE.
CLINICAL CASE (Continued)
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/521/1851845/521burnett.pdf by guest on 13 December 2021
The patient does well, and a year later, she is going to have her
other hip replaced. Unfortunately, her insur­ance has changed, and
she can no lon­ger afford a DOAC and is tak­ing war­fa­rin. Her med­
i­cal his­tory is unchanged, and her lab­o­ra­tory tests remain nor­mal.
Warfarin
Due to the long off­set of war­fa­rin (approx­i­ma­tely 5 days), cli­ni­
cians must deter­mine tim­ing of inter­rup­tion to pro­vide the de-
sired resid­ual anti­co­ag­u­lant effect, as well as any role for bridg­ing
with a par­en­teral anti­co­ag­u­lant. For clar­ity, when we use the term
bridg­ing, we mean ther­a­peu­tic-inten­sity anticoagulation, such as
enoxaparin 1 mg/kg twice daily. The deci­sion to bridge or not is
based on the patient’s indi­ca­tion for anticoagulation, under­ly­ing
throm­botic risk, and the throm­botic risk of the pro­ce­dure. Al-
though bridg­ing of war­fa­rin was once rou­tinely employed, ret­ro­
spec­tive obser­va­tional evi­dence published over the past decade
has con­sis­tently shown bridg­ing to be asso­ci­ated with net harm
and sug­gests it should be avoided in most cases.6,35-37
The ran­ dom­ ized con­ trolled Bridging Anticoagulation in
Patients Who Require Temporary Interruption of Warfarin Ther-
apy for an Elective Invasive Procedure or Surgery (BRIDGE) trial
sought to answer whether fore­go­ing LMWH bridg­ing in atrial
fibril­la­tion patients requir­ing war­fa­rin inter­rup­tion for a pro­ce­dure
would result in less major bleed­ing with­out an increase in arte­
rial throm­bo­em­bo­lism. Per stan­dard­ized pro­to­col, war­fa­rin was
held for 5 days preprocedurally, and patients were ran­dom­ized to
pla­cebo or LMWH (dalteparin) with 30-day pri­mary out­comes of
major bleed­ing (supe­ri­or­ity) and arte­rial throm­bo­em­bolic event
(noninferiority). There was no dif­fer­ence in arte­rial throm­bo­em­
bo­lism 0.4% vs 0.3%, and major bleed­ing was reduced with no
bridg­ing (1.3%) vs LMWH bridg­ing (3.2%) (P < .005).37 This trial
Dabigatran CrCl ≥50 Dabigatran CrCl <0 Edoxaban
Days Days Not stud­ied
1 2
2 4
1 1
2-3 2-3
Perioperative man­age­ment of antithrombotics  |  523
prompted a par­a­digm shift away from bridg­ing in most patients and she now has a bileaflet mechan­i­cal aor­tic valve. Her other med­
with atrial fibril­la­tion. i­cal his­tory is unchanged, and lab­o­ra­tory val­ues remain nor­mal. She
In the BRIDGE trial, war­fa­rin was resumed the night of the pro­ pres­ents for perioperative war­fa­rin man­age­ment again.
ce­dure at the patient’s usual home dose. The mean time to rees-
tablish a ther­a­peu­tic inter­na­tional nor­mal­ized ratio (INR) was 8
The Amer­i­can College of Chest Physicians (ACCP) 2012 guide­
days. To min­i­mize this period of sub­ther­a­peu­tic anticoagulation,
lines and the Amer­i­can College of Cardiology (ACC)/Amer­i­can
it is rea­son­able to con­sider a boosted dose of war­fa­rin for 1 to 2
Heart Association 2020 guide­lines are con­cor­dant in their clas­
days after the pro­ce­dure in the absence of high bleed­ing risk.38
si­fi­ca­tion of mechan­i­cal heart valves as (1) high risk if any mitral,
caged-ball, or tilting disk valves or recent (within 6 months)
Thromboembolic risk strat­i­fi­ca­tion
stroke or TIA; (2) mod­er­ate risk with bileaflet aor­tic valve and
We use a throm­bo­em­bolic risk strat­i­fi­ca­tion approach based
any other risk fac­tor, which includes atrial fibril­la­tion, prior stroke
on avail­­able evi­dence and expert con­sen­sus rec­om­men­da­tions
or TIA, hyper­ten­sion, dia­be­tes, con­ges­tive heart fail­ure, or older
from mul­ti­ple orga­ni­za­tions to guide deci­sions on bridg­ing war­
than 75 years; and (3) low risk with bileaflet aor­tic valves with no

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/521/1851845/521burnett.pdf by guest on 13 December 2021

fa­rin patients who require tem­po­rary inter­rup­tion for a pro­ce­
dure (Table 3).18,19,39,40
• We would not use bridg­ing anticoagulation in our atrial fibril­la­tion
patient because her CHADS2-VA2Sc (Congestive heart failure = 1,
Hypertension = 1, age >/= 75 years = 2, Diabetes = 1, Stroke/TIA = 2,
Vascular disease = 1, Age 65-74 = 1, Sex category = 1) score is 4.
•  We would resume her war­fa­rin the night of the pro­ce­dure.
• While the patient is still in the hos­pi­tal, we would ensure ade­
quate pro­phy­laxis with low-dose anticoagulation (eg, enoxa-
parin 40 mg SQ once daily) to pre­vent post­op­er­a­tive VTE until
the INR is 2 or more.
Mechanical heart valve
CLINICAL CASE (Continued)
The patient did well after her sec­ond hip sur­gery and has been
doing so much walk­ing that her left knee now needs replace­ment.
Unfortunately, her bioprosthetic aor­tic valve gave out in the interim,
other risk fac­tors (Table 3).19,39
The only ran­dom­ized trial for LMWH bridg­ing for war­fa­rin inter­
rup­tion in patients with mechan­i­cal heart valves that we are aware
of is the recently published Postoperative low molecular weight
heparin bridging treatment for patients at high risk of arterial
thromboembolism (PERIOP 2) trial.41 Patients with atrial fibril­la­tion
or non-high-risk mechan­i­cal valves who required war­fa­rin inter­rup­
tion for a pro­ce­dure (n = 1471) were ran­dom­ized to LMWH bridg­ing
or pla­cebo and followed for 90 days. LMWH bridg­ing consisted
of ther­a­peu­tic dose dalteparin 200 IU/kg sub­cu­ta­ne­ously on days
−3 and −2 before pro­ce­dure and half the ther­a­peu­tic dose of
100 IU/kg sub­cu­ta­ne­ously the day before the pro­ce­dure. Post-
procedurally (same day or next day), patients in the inter­ven­tion
arm at high bleed risk were given fixed-dose pro­phy­lac­tic dalte-
parin 5000 IU, and those at low bleed risk were given dalteparin
200 IU/kg sub­cu­ta­ne­ously until the INR was more than 1.9.
In the sub­group of 304 patients with non-high-risk mechan­
i­cal valves, no bridg­ing vs bridg­ing major throm­bo­em­bo­
lism event rates were 0% vs 0.7% (P = .49), and major bleed­ing
occurred in 2% vs 0.7% (P = .62) respec­tively. The inves­ti­ga­tors
con­cluded there is no ben­e­fit from postprocedure LMWH bridg­
ing in patients with non-high-risk mechan­i­cal valves.

Table 3. Thromboembolic risk strat­i­fi­ca­tion and bridg­ing con­sid­er­ations

Indication Mechanical heart valve Atrial fibril­la­tion VTE

Guideline(s) ACCP 2012, ACC/AHA 2020 ACCP 2012, ACC 2017 ACCP 2012, ASH 2018
Thrombotic risk Criteria Recommendation Criteria* Recommendation Criteria Recommendation
High •  All mitral valve Suggest bridg­ing/ CHADS2 CHA2DS-VA2Sc Suggest bridg­ing • Within 3 CHEST 2012:
• Caged-ball and rea­son­able to >4 >7 months Suggest
tilting disc bridge • Severe bridg­ing
• Stroke or TIA in thrombophilia ASH 2018: Not
past 6 months addressed
Moderate Bileaflet aor­tic valve Individualized CHADS2 CHA2DS-VA2Sc Individualized • Past 3-12 CHEST 2012:
+ risk fac­tors deci­sion based 3-4 5 or 6 deci­sion based months Individualized
•  Atrial fibril­la­tion on patient and on patient and •  Recurrent VTE deci­sion based
•  Prior stroke or TIA pro­ce­dural pro­ce­dural •  Active can­cer on patient and
• Hypertension nuances nuances • Nonsevere pro­ce­dural
• Diabetes thrombophilia nuances
•  Congestive heart ASH 2018: Do not
fail­ure bridge
•  Age >75 years
Low Bileaflet aor­tic Do not bridge CHADS2 CHA2DS-VA2Sc Do not bridge More than 12 CHEST 2012: Do
valve with­out risk ≤2 <4 months ago not bridge
fac­tors ASH 2018: Do not
bridge
*ACCP 2012 based on CHADS2; 2017 ACC expert con­sen­sus deci­sion path­way for periprocedural man­age­ment of anticoagulation based on CHA2DS-VA2Sc.
ACC, American College of Cardiology; ACCP, American College of Chest Physicians; AHA, Amer­i­can Heart Association.

524  |  Hematology 2021  |  ASH Education Program

Table 4. Special sit­u­a­tions that may influ­ence perioperative antithrombotic man­age­ment

Situation or issue Comments Suggested actions

Urgent/emer­gent • If there is not ade­quate time to allow nat­u­ral nor­mal­i­za­tion
pro­ce­dures of the patient’s coag­u­la­tion sta­tus before sur­gery, use of
rever­sal agents, prohemostatic agents, or spe­cific anti­
dotes may be indi­cated and should be done judi­ciously
and thought­fully.
• Rapidly returning a patient to their native, prothrombotic
state along with any intrin­sic risk of throm­bo­sis posed by
the rever­sal agents or anti­dotes them­selves may increase
the risk for adverse events.
Patients on con­com­i­tant • This is an oppor­tune time to eval­u­ate the over­all clin­i­cal
antiplatelet ther­a­pies neces­sity of con­com­i­tant antiplatelet ther­apy. If not indi­
cated, cli­ni­cians should dis­cuss per­ma­nent dis­con­tin­u­a­tion
• Shared deci­sion mak­ing with the patient, mul­ti­dis­ci­
plin­ary dis­cus­sion, and con­sul­ta­tion with a throm­bo­sis
spe­cial­ist
• Clinicians are referred to existing guid­ance on rever­sal
of anticoagulation.40,45,46
• Shared deci­sion mak­ing with the patient, mul­ti­dis­ci­
plin­ary dis­cus­sion, and con­sul­ta­tion with pre­scriber of
antiplatelet ther­apy (eg, car­di­­ol­o­gist, neu­rol­o­gist) and

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/521/1851845/521burnett.pdf by guest on 13 December 2021

with the patient and pre­scriber.
• Whether to tem­po­rar­ily inter­rupt con­com­i­tant antiplatelet
ther­a­pies is a com­plex deci­sion that is based on sev­eral
fac­tors, includ­ing indi­ca­tion, recency of events, bleed­ing,
and throm­botic risks of the pro­ce­dure and patient.
• Perioperative antiplatelet strat­e­gies should be indi­vid­u­ally
tai­lored based on mul­ti­dis­ci­plin­ary input.
History of hep­a­rin-induced • Patients with a his­tory of HIT should not receive any hep­
throm­bo­cy­to­pe­nia (HIT) a­rin or LMWH prod­ucts, includ­ing small doses such as
flushes or VTE pro­phy­laxis.
Inferior vena cava (IVC) • The esti­mated inci­dence of VTE recur­rence in the first
fil­ters month after an acute event off of anti­co­ag­u­lant ther­apy is
esti­mated to be 40%.50
• If pos­si­ble, delay non­ur­gent/emer­gent pro­ce­dures to
allow at least 3 months of anticoagulation ther­apy fol­low­
ing an acute VTE.
• If the pro­ce­dure can­not be delayed and the VTE occurred
in the pre­vi­ous 30 days, a retriev­able IVC fil­ter may be
con­sid­ered.
Severe renal impair­ment • Warfarin patients with severe renal impair­ment or on
hemo­di­al­y­sis who have a clear indi­ca­tion for bridg­ing can­
not be man­aged with LMWH.
throm­bo­sis spe­cial­ist
• Clinicians are referred to existing guid­ance on
perioperative antiplatelet man­age­ment.47,48
• Use an alter­na­tive, nonheparin anti­co­ag­u­lant such as
bivalirudin, fondaparinux, or a DOAC as appro­pri­ate
based on patient’s clin­i­cal sta­tus and clin­i­cal sit­u­a­tion.
• Clinicians are referred to existing guid­ance on HIT.49
• If the patient is antic­i­pated to be off anticoagulation for
<48 hours, aggres­sive phar­ma­co­logic pro­phy­laxis with
expe­di­ent esca­la­tion to ther­a­peu­tic dos­ing is pre­ferred.
• If a retriev­able fil­ter is con­sid­ered, a plan for timely
removal should be clearly delin­eated prior to place­ment.
• Clinicians are referred to existing guid­ance on IVC fil­ters.51
• These patients may need to have their war­fa­rin held at a
prespecified time in the out­pa­tient set­ting and then be
admit­ted for bridg­ing ther­apy with intra­ve­nous hep­a­rin.

In a sys­temic review of 5 stud­ies of unfractioned hep­a­rin or
LMWH bridg­ing in patients with mechan­i­cal heart valves, major
bleed­ing rates ranged from 4.39% to 10.07%. Bleeding def­i­ni­
tions var­ied, pre­clud­ing the pooling of results. Although obser­
va­tional data, this anal­y­sis sug­gests that bleed­ing risk asso­ci­ated
with bridg­ing in valve patients is not neg­li­gi­ble and under­scores
the need for lim­it­ing to patients at increased throm­botic risk.42
Some obser­va­tional data sug­gest that pro­phy­lac­tic dose
anticoagulation may be a via­ ble perioperative approach for
select mechan­i­cal valve patients requir­ing tem­po­rary inter­rup­
tion in war­fa­rin, as well as those with newly implanted valves as
a bridge to a ther­a­peu­tic INR of 2 or more.43,44 This may mit­i­gate
bleed­ing risk but also pro­vi­des impor­tant post­op­er­a­tive VTE
pro­phy­laxis. However, it can­not be recommended as a rou­tine
approach for all­patients until bet­ter data are avail­­able.
•  We would check her INR about 7 to 10 days prior to sur­gery
and, if in the ther­a­peu­tic range, would hold war­fa­rin 5 days
pre­op­er­a­tively to allow nadir of anti­co­ag­u­lant effect at the
time of the pro­ce­dure.
•  We would have care­ ful shared deci­ sion mak­ ing with the
patient and her ortho­ pe­dic sur­
geon regard­ ing the poten­
tial ben­e­fits and harms of bridg­ing with LMWH because her
throm­bo­em­bolic risk is mod­er­ate, and unlike war­fa­rin inter­
Prakash Singh Shekhawat
rup­tion in atrial fibril­la­tion, there is no strong evi­dence to
guide us.
•  After much dis­cus­sion, it is ulti­mately decided she does not
require bridg­ing ther­apy based on her mod­er­ate throm­bo­
em­bolic risk from her bileaflet mechan­i­cal aor­tic valve.
VTE
CLINICAL CASE (Continued)
•  A year later, the patient is now going to have her other (right)
knee replaced (fourth major ortho­pe­dic sur­gery), and this
con­sul­ta­tion is more com­plex.
•  After her last knee sur­gery, the deci­sion to not use bridg­
ing with ther­a­peu­tic dose LMWH was interpreted as to not
use any LMWH, and the patient never received any venous
thromboembolism pro­phy­laxis post­op­er­a­tively while war­fa­
rin rose to the appro­pri­ate INR tar­get.
•  She devel­oped a deep vein thrombosis, which was treated
in the usual man­ ner, and she remains on war­ fa­
rin for her
mechan­i­cal heart valve and atrial fibril­la­tion.
Perioperative man­age­ment of antithrombotics  |  525
Table 5. Additional clin­i­cal resources for periprocedural tools an anti­co­ag­ul­ant that does not require oral intake or absorp­tion
and guid­ance would be pre­ferred.

Resource Location Special sit­u­a­tions

Anticoagulation Forum Centers of
Excellence
Managing Anticoagulation in the
Perioperative Period (MAPP)
Michigan Anticoagulation Quality
Improvement Initiative (MAQI2)
Thrombosis Canada
https:​­/​­/acforum​­-excellence​­.org
http:​­/​­/mappp​­.ipro​­.org
http:​­/​­/www​­.maqi2​­.org
https://thrombosiscanada​­.ca
Some addi­tional clin­i­cal sit­u­a­tions war­rant dis­cus­sion but are be-
yond the scope this arti­cle. We have sum­ma­rized these in Tables
4 and 5.45-51
Summary
Each year, a large num­ber of patients tak­ing OACs undergo an
inva­sive pro­ce­dure, with many requir­ing tem­po­rary inter­rup­tion
of ther­apy. This is a high-risk period for patients that requires
thought­ ful and method­ i­
cal approaches using the best avail­­
able evi­dence and expert con­sen­sus to bal­ance bleed­ing and

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/521/1851845/521burnett.pdf by guest on 13 December 2021

The ACCP classifies VTE as (1) high, VTE within 3 months or
severe thrombophilia; (2) mod­er­ate, VTE within 3 to 12 months,
nonsevere thrombophilia (eg, het­ero­zy­gous fac­tor V Leiden or pro­
throm­bin gene muta­tion), recur­rent VTE, or active can­cer within 6
months; and (3) low, VTE more than 12 months ago and no other
risk fac­tors. They sug­gest bridg­ing for high risk, no bridg­ing for low
risk, and indi­vid­u­al­ized deci­sion mak­ing for mod­er­ate risk.19 The
Amer­i­can Society of Hematology (ASH) 2018 guide­lines on opti­mal
man­age­ment of anti­co­ag­u­lant ther­apy, which use the same VTE
risk strat­i­fi­ca­tion as the ACCP, dif­fer slightly, in that they sug­gest no
bridg­ing in patients with VTE at mod­er­ate risk (Table 3).40
There are no ran­dom­ized tri­als for bridg­ing in patients with
VTE who require periprocedural war­fa­rin inter­rup­tion that we
are aware of. A sys­tem­atic review of 28 obser­va­tional stud­ies
com­par­ing bridg­ing vs no bridg­ing in 6915 pro­ce­dures showed
no dif­fer­ence in VTE rates (0.7% vs 0.5%), an increase in major
bleed­ing (1.8% vs 0.4%), and an increase in all­bleed­ing (3.9% vs
0.4%), respec­tively. The 95% con­fi­dence inter­vals overlapped for
VTE and major bleed­ing but did not for all­bleed­ing.35
•  There is no clear answer for this patient. She is at low risk from
a purely VTE stand­point based on ACCP and ASH guide­lines.
•  However, we would likely bridge just on prac­ti­cal­ity of her
col­lec­tive risk fac­tors.
•  We would ini­ti­ate perioperative bridg­ing with ther­a­peu­tic
dose LMWH and hold for 24 hours prior to the pro­ce­dure.
•  We would resume ther­a­peu­tic dose LMWH 2 to 3 days post­op­
er­a­tively because knee sur­gery is a high bleed­ing risk sur­gery.
•  This time, how­ever, pro­phy­lac­tic dose LMWH is to be given on
post­op­er­a­tive days 1 and 2 for VTE pre­ven­tion until ther­a­peu­tic
dose LMWH is started for the mod­er­ate-risk mechan­ic ­ al valve
(and will also “cover” the low-risk atrial fibril­la­tion and prior VTE).
Postprocedure VTE pro­phy­laxis
As high­lighted in each of the case vignettes above, most if not
all­ sur­gi­cal patients war­rant post­op­er­a­tive VTE pro­phy­laxis in the
interim until they can be safely increased to ther­a­peu­tic anti-
coagulation at the appro­pri­ate time (ie, 24 hours for low bleed
risk pro­ce­dures, 48-72 hours for high bleed risk pro­ce­dures). This
is termed a “step-up” approach, and the most com­monly used
pro­phy­lac­tic agents are LMWH and hep­a­rin. Although DOACs
could be con­sid­ered a means of VTE pro­phy­laxis in the step-
up period, patients may not imme­di­ately be ­able to tol­er­ate
oral med­i­ca­tions. In addi­tion, some patients may have altered
absorp­tion, includ­ing those with post­op­er­a­tive ileus or patients
who have under­gone bariatric sur­gery. For these patients, use of
throm­botic risks. It is impor­tant for cli­ni­cians to be famil­iar with
pro­ce­dures where OAC does not require inter­rup­tion, as this
will greatly sim­plify man­age­ment and likely min­i­mize adverse
events. For sit­u­at­ ions when inter­rup­tion is indi­cated, it is imper­
a­tive for cli­ni­cians to be famil­iar with key dif­fer­ences in phar­ma­
co­ki­netic prop­er­ties between DOACs and war­fa­rin, as these lead
to sig­nif­i­cantly dif­fer­ent perioperative man­age­ment approaches.
A step­wise sys­tem-level pro­cess, mul­ti­dis­ci­plin­ary col­lab­o­ra­tion,
shared deci­sion mak­ing with patients, and clear com­mu­ni­ca­tion
and doc­um ­ en­ta­tion of the plan are all­key ele­ments of antithrom-
bosis stew­ard­ship nec­es­sary for suc­cess­ful nav­i­ga­tion and im-
plementation of perioperative plans.
Conflict-of-inter­est dis­clo­sure
Allison Elaine Burnett: no rel­e­vant con­flicts of inter­est.
Bishoy Ragheb: no rel­e­vant con­flicts of inter­est.
Scott Kaatz: research funding: Janssen, BMS, Osmosis Research,
National Institutes of Health; consultancy: Janssen, BMS, Alexion/
Portola, Novartis, CSL Behring, Gilead.
Off-label drug use
Allison Elaine Burnett: There are currently no anticoagulants ap-
proved specifically for bridging therapy in warfarin patients, thus
any discussion on this is off-label.
Bishoy Raghe: There are currently no anticoagulants approved
specifically for bridging therapy in warfarin patients, thus any
discussion on this is off-label.
Scott Kaatz: There are currently no anticoagulants approved
specifically for bridging therapy in warfarin patients, thus any
discussion on this is off-label.
Correspondence
Allison Elaine Burnett; University of New Mexico, Health Sciences
Center, College of Pharmacy, 2211 Lomas Blvd NE, Albuquerque,
NM 87106; e-mail: aburnett@salud​­.unm​­.edu.
References
1. Colacci M, Tseng EK, Sacks CA, Fralick M. Oral anti­co­ag­u­lant uti­li­za­tion in
the United States and United Kingdom. J Gen Intern Med. 2020;35(8):2505-
2507.
2. Bleeding Risk Survey. Blood clots. Accessed 24 April 2021. https:​­/​­/www​
­.stoptheclot​­.org​­/news​­/bleeding​­-risk​­-survey​­/
3. Sherwood MW, Douketis JD, Patel MR, et al; ROCKET AF Investigators.
Outcomes of tem­po­rary inter­rup­tion of rivaroxaban com­pared with war­
fa­rin in patients with nonvalvular atrial fibril­la­tion: results from the rivarox-
aban once daily, oral, direct fac­tor Xa inhi­bi­tion com­pared with vita­min K
antag­o­nism for pre­ven­tion of stroke and embolism trial in atrial fibril­la­tion
(ROCKET AF). Circulation. 2014;129(18):1850-1859.
4. Healey JS, Eikelboom J, Douketis J, et al; RE-LY Investigators. Periprocedural
bleed­ing and throm­bo­em­bolic events with dabigatran com­pared with

526  |  Hematology 2021  |  ASH Education Program

 war­fa­rin: results from the ran­dom­ized eval­u­a­tion of long-term anticoagula-
tion ther­apy (RE-LY) ran­dom­ized trial. Circulation. 2012;126(3):343-348.
5. Shaw JR, Kaplovitch E, Douketis J. Periprocedural man­age­ment of oral anti-
coagulation. Med Clin North Am. 2020;104(4):709-726.
6. Rose AJ, Allen AL, Minichello T. A call to reduce the use of bridg­ing antico-
agulation. Circ Cardiovasc Qual Outcomes. 2016;9(1):64-67.
7. Bougeard A-M, Watkins B. Transitions of care in the perioperative peri-
od—a review. Clin Med (Lond). 2019;19(6):446-449.
8. Michota F. Transitions of care in anticoagulated patients. J Multidiscip
Healthc. 2013;6:215-228.
9. Steib A, Mertes P-M, Marret E, Albaladejo P, Fusciardi J. Compliance with
guide­lines for the perioperative man­age­ment of vita­min K antag­o­nists.
Thromb Res. 2014;133(6):1056-1060.
10. Dager WE, Ansell J, Barnes GD, et al. “Reduce the like­li­hood of patient harm
asso­ci­ated with the use of anti­co­ag­u­lant ther­apy”: Commentary from the
Anticoagulation Forum on the Updated Joint Commission NPSG.03.05.01
ele­ments of per­for­mance. Jt Comm J Qual Patient Saf. 2020;46(3):173-180.
11. Centers for Medi­ care & Med­ ic­
aid Services. Quality pay­ ment pro­gram.
Explore mea­sures & activ­i­ties. 2021 improve­ment activ­i­ties. Anticoagulant
man­age­ment improve­ments. Accessed 25 April 2021. https:​­/​­/qpp​­.cms​­.gov​
­/mips​­/explore​­-measures​­?tab=improvementActivities&py=2020
12. Kataruka A, Renner E, Barnes GD. Evaluating the role of clin­i­cal phar­ma­
cists in pre-pro­ce­dural anticoagulation man­age­ment. Hosp Pract (1995).
2018;46(1):16-21.
13. Marquez J, Togami JC, Dant CR, Herrera A, Marshik P, Burnett AE. Peri-pro­
ce­dural antithrombotic man­age­ment: time to burn the bridge? J Thromb
Thrombolysis. 2018;45(3):337-344.
14. Collins R, Scrimgeour A, Yusuf S, Peto R. Reduction in fatal pul­mo­nary
embolism and venous throm­bo­sis by perioperative admin­is­tra­tion of sub­
cu­ta­ne­ous hep­a­rin. Overview of results of ran­dom­ized tri­als in gen­eral,
ortho­pe­dic, and uro­logic sur­gery. N Engl J Med. 1988;318(18):1162-1173.
15. Baron TH, Kamath PS, McBane RD. Management of antithrombotic ther­apy
in patients under­go­ing inva­sive pro­ce­dures. N Engl J Med. 2013;368(22):​
2113-2124.
16. Spyropoulos AC, Douketis JD. How I treat anticoagulated patients under­
go­ing an elec­tive pro­ce­dure or sur­gery. Blood. 2012;120(15):2954-2962.
17. Spyropoulos AC, Brohi K, Caprini J, et al; SSC Subcommittee on Perioper-
ative and Critical Care Thrombosis and Haemostasis of the International
Society on Thrombosis and Haemostasis. Scientific and stan­dard­i­za­tion
com­mit­tee com­mu­ni­ca­tion: guid­ance doc­u­ment on the periprocedural
man­age­ment of patients on chronic oral anti­co­ag­u­lant ther­apy: rec­om­men­
da­tions for stan­dard­ized reporting of pro­ce­dural/sur­gi­cal bleed risk and
patient-spe­cific throm­bo­em­bolic risk. J Thromb Haemost. 2019;17(11):1966–
1972.
18. Doherty JU, Gluckman TJ, Hucker WJ, et al. 2017 ACC expert con­sen­sus deci­
sion path­way for periprocedural man­age­ment of anticoagulation in patients
with nonvalvular atrial fibril­la­tion. J Am Coll Cardiol. 2017;69(7):871–898.
19. Douketis JD, Spyropoulos AC, Spencer FA, et  al. Perioperative man­age­
ment of antithrombotic ther­apy: antithrombotic ther­apy and pre­ven­tion of
throm­bo­sis, 9th ed: Amer­i­can College of Chest Physicians Evidence-Based
Clinical Practice Guidelines. Chest. 2012;141(2, suppl):e326S-e350S.
20. Birnie DH, Healey JS, Wells GA, et al; BRUISE CONTROL Investigators. Pace-
maker or defi­bril­la­tor sur­gery with­out inter­rup­tion of anticoagulation. N
Engl J Med. 2013;368(22):2084-2093.
21. Di Biase L, Burkhardt JD, Santangeli P, et  al. Periprocedural stroke and
bleed­ing com­pli­ca­tions in patients under­go­ing cath­e­ter abla­tion of atrial
fibril­la­tion with dif­fer­ent anticoagulation man­age­ment: results from the
role of coumadin in pre­vent­ing throm­bo­em­bo­lism in atrial fibril­la­tion (AF)
patients under­go­ing cath­e­ter abla­tion (COMPARE) ran­dom­ized trial. Circu-
lation. 2014;129(25):2638-2644.
22. Kowalewski M, Suwalski P, Raffa GM, et al. Meta-anal­y­sis of unin­ter­rupted as
com­pared to interrupted oral anticoagulation with or with­out bridg­ing in
patients under­go­ing cor­o­nary angi­og­ra­phy with or with­out per­cu­ta­ne­ous
cor­o­nary inter­ven­tion. Int J Cardiol. 2016;223:186-194.
23. Shahi V, Brinjikji W, Murad MH, Asirvatham SJ, Kallmes DF. Safety of unin­
ter­rupted war­fa­rin ther­apy in patients under­go­ing car­dio­vas­cu­lar endo-
vascular pro­ce­dures: a sys­tem­atic review and meta-anal­y­sis. Radiology.
2016;278(2):383-394.
24. Birnie DH, Healey JS, Wells GA, et al. Continued vs. interrupted direct oral
anti­co­ag­u­lants at the time of device sur­gery, in patients with mod­er­ate
to high risk of arte­rial thrombo-embolic events (BRUISE CONTROL-2). Eur
Heart J. 2018;39(44):3973-3979.
25. Beyer-Westendorf J, Gelbricht V, Förster K, et al. Peri-interventional man­
Prakash Singh Shekhawat
age­ment of novel oral anti­co­ag­u­lants in daily care: results from the pro­
spec­tive Dresden NOAC reg­is­try. Eur Heart J. 2014;35(28):1888-1896.
26. Steffel J, Collins R, Antz M, et  al. 2021 Euro­pean Heart Rhythm Associa-
tion prac­ti­cal guide on the use of non-vita­min K antag­o­nist oral anti­co­ag­
u­lants in patients with atrial fibril­la­tion [published online 25 April 2021]. EP
Europace.
27. Romero J, Cerrud-Rodriguez RC, Alviz I, et al. Significant ben­e­fit of unin­
ter­rupted DOACs ver­sus VKA dur­ing cath­e­ter abla­tion of atrial fibril­la­tion.
JACC Clin Electrophysiol. 2019;5(12):1396-1405.
28. Sticherling C, Marin F, Birnie D, et  al. Antithrombotic man­ age­ ment in
patients under­go­ing elec­tro­phys­i­o­log­i­cal pro­ce­dures: a Euro­pean Heart
Rhythm Association (EHRA) posi­ tion doc­ u­
ment endorsed by the ESC
Working Group Thrombosis, Heart Rhythm Society (HRS), and Asia Pacific
Heart Rhythm Society (APHRS). Europace. 2015;17(8):1197-1214.
29. Burnett AE, Mahan CE, Vazquez SR, Oertel LB, Garcia DA, Ansell J. Guidance
for the prac­ti­cal man­age­ment of the direct oral anti­co­ag­u­lants (DOACs) in
VTE treat­ment. J Thromb Thrombolysis. 2016;41(1):206-232.
30. Douketis JD, Spyropoulos AC, Duncan J, et al. Perioperative man­age­ment
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/521/1851845/521burnett.pdf by guest on 13 December 2021
of patients with atrial fibril­la­tion receiv­ing a direct oral anti­co­ag­u­lant. JAMA
Intern Med. 2019;179(11):1469-1478.
31. Schulman S, Carrier M, Lee AY, et al; Periop Dabigatran Study Group.
Perioperative man­age­ment of dabigatran: a pro­spec­tive cohort study. Cir-
culation. 2015;132(3):167-173.
32. Godier A, Dincq A-S, Martin A-C, et al. Predictors of pre-pro­ce­dural con­cen­
tra­tions of direct oral anti­co­ag­u­lants: a pro­spec­tive multicentre study. Eur
Heart J. 2017;38(31):2431-2439.
33. Levy JH, Ageno W, Chan NC, et al; Subcommittee on Control of Antico-
agulation. When and how to use anti­dotes for the rever­sal of direct oral
anti­co­ag­u­lants: guid­ance from the SSC of the ISTH. J Thromb Haemost.
2016;14(3):623-627.
34. Horlocker TT, Vandermeuelen E, Kopp SL, Gogarten W, Leffert LR, Benzon
HT. Regional anes­the­sia in the patient receiv­ing antithrombotic or throm­
bo­lytic ther­apy: Amer­i­can Society of Regional Anesthesia and Pain Med-
icine Evidence-Based Guidelines (Fourth Edition). Reg Anesth Pain Med.
2018;43(3):263-309.
35. Baumgartner C, de Kouchkovsky I, Whitaker E, Fang MC. Periprocedural
bridg­ing in patients with venous throm­bo­em­bo­lism: a sys­tem­atic review.
Am J Med. 2019;132(6):722-732.e7732e7.
36. Siegal D, Yudin J, Kaatz S, Douketis JD, Lim W, Spyropoulos AC. Periproce-
dural hep­ a­rin bridg­ing in patients receiv­ ing vita­min K antag­ o­nists: sys­
tem­atic review and meta-anal­y­sis of bleed­ing and throm­bo­em­bolic rates.
Circulation. 2012;126(13):1630-1639.
37. Douketis JD, Spyropoulos AC, Kaatz S, et al; BRIDGE Investigators. Periop-
erative bridg­ing anticoagulation in patients with atrial fibril­la­tion. N Engl J
Med. 2015;373(9):823-833.
38. Schulman S, Hwang H-G, Eikelboom J-W, Kearon C, Pai M, Delaney J. Load-
ing dose vs. main­te­nance dose of war­fa­rin for reinitiation after inva­sive pro­
ce­dures: a ran­dom­ized trial. J Thromb Haemost. 2014;12(8):1254-1259.
39. Otto Catherine M., Nishimura Rick A., Bonow Robert O., et  al. 2020
ACC/AHA guide­line for the man­age­ment of patients with val­vu­lar heart
dis­ease: a report of the Amer­i­can College of Cardiology/Amer­i­can Heart
Association Joint Committee on Clinical Practice Guidelines. Circulation.
2021;​143(5):​e72-e227.
40. Witt DM, Nieuwlaat R, Clark NP, et  al. Amer­i­can Society of Hematology
2018 guide­lines for man­age­ment of venous throm­bo­em­bo­lism: opti­mal
man­age­ment of anticoagulation ther­apy. Blood Adv. 2018;2(22):3257-3291.
41. Kovacs MJ, Wells PS, Anderson DR, et al; PERIOP2 Investigators. Postop-
erative low molec­u­lar weight hep­a­rin bridg­ing treat­ment for patients at
high risk of arte­rial throm­bo­em­bo­lism (PERIOP2): dou­ble blind randomised
con­trolled trial. BMJ. 2021;373:n1205.
42. Torres M. Effects of heparinoid bridg­ing in patients with mechan­i­cal heart
valves [published online 16 Novem­ber 2020]. J Am Assoc Nurse Pract.
43. Mathew JG, Spyropoulos AC, Yusuf A, et  al. Efficacy and safety of early
par­en­teral anticoagulation as a bridge to war­fa­rin after mechan­i­cal valve
replace­ment. Thromb Haemost. 2014;112(6):1120-1128.
44. Hjellström L, Labaf A. Prophylactic doses of low-molec­u­lar weight hep­a­
rin as periprocedural bridg­ing ther­apy in mechan­i­cal heart valve patients.
Thromb Res. 2018;163:180-184.
45. Tomaselli GF, Mahaffey KW, Cuker A, et  al. 2020 ACC expert con­sen­sus
deci­sion path­way on man­age­ment of bleed­ing in patients on oral anti­co­
ag­u­lants. J Am Coll Cardiol. 2020;76(5):594-622.
46. Cuker A, Burnett A, Triller D, et al. Reversal of direct oral anti­co­ag­u­lants:
guid­ance from the anticoagulation forum. Am J Hematol. 2019;94(6):697–
709.
47. Filipescu DC, Stefan MG, Valeanu L, Popescu WM. Perioperative man­age­
ment of antiplatelet ther­apy in non­car­diac sur­gery. Curr Opin Anaesthesiol.
2020;33(3):454-462.
Perioperative man­age­ment of antithrombotics  |  527
48. Roberta R, Giuseppe T, Giuseppe M, et al. A mul­ti­dis­ci­plin­ary approach on 51. Kelkar AH, Rajasekhar A. Inferior vena cava fil­ters: a frame­work for evi­
the perioperative antithrombotic man­age­ment of patients with cor­o­nary dence-based use. Hematology Am Soc Hematol Educ Program. 2020;​
stents under­go­ing sur­gery. JACC: Cardiovasc Interv. 2018;11(5):417-434. 2020(1):619-628.
49. Cuker A, Arepally GM, Chong BH, et al. Amer­i­can Society of Hematology
2018 guide­lines for man­age­ment of venous throm­bo­em­bo­lism: hep­a­rin-
induced throm­bo­cy­to­pe­nia. Blood Adv. 2018;2(22):3360-3392.
50. Kearon C, Hirsh J. Management of anticoagulation before and after elec­tive
sur­gery. N Engl J Med. 1997;336(21):1506-1511. DOI 10.1182/hema­tol­ogy.2021000287

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/521/1851845/521burnett.pdf by guest on 13 December 2021

528  |  Hematology 2021  |  ASH Education Program

 PERIOPERATIVE CONSULTATIVE HEMATOLOGY: CAN YOU CLEAR MY PATIENT FOR SURGERY ?

How to manage bleeding disorders in aging

patients needing surgery

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/529/1851841/529abou-ismail.pdf by guest on 13 December 2021

Mouhamed Yazan Abou-Ismail1 and Nathan T. Connell2,3
Division of Hematology and Hematologic Malignancies, Department of Internal Medicine, University of Utah, Salt Lake City, UT; 2Hematology
1

Division, Department of Medicine, Brigham and Women’s Hospital, Boston, MA; and 3Harvard Medical School, Boston, MA

With improvements in medical care, the life expectancy of patients with bleeding disorders is approaching that of the gen-
eral population. A growing population of older adult patients with bleeding disorders is at risk of age-related comorbidities
and in need of various elective and emergent age-related procedures. The increased risk of thrombosis and volume over-
load in older adults complicates perioperative hemostatic management. Furthermore, antithrombotic treatment such as
antiplatelet or anticoagulant therapy, which is frequently required for various cardiovascular interventions, requires a metic-
ulous individualized approach. Evidence-based guidelines for the management of aging patients with bleeding disorders
are lacking, largely due to the underrepresentation of older adult patients in clinical trials as well as the rarity of many such
bleeding disorders. We discuss the current guidelines and recommendations in the perioperative hemostatic management
of older adult patients with hemophilia and von Willebrand disease as well as other rare bleeding disorders. The optimal
management of these patients is often complex and requires a thorough multidisciplinary and individualized approach
involving hematologists, surgeons, anesthesiologists, and the specialists treating the underlying disorder.

LEARNING OBJECTIVES
• Recognize the challenges unique to the aging patient, such as age-related risks for thrombosis and volume overload
• Review the guidelines for the peri-operative hemostatic management of various bleeding disorders
• Apply an optimal approach in complex management scenarios, utilizing an individualized treatment plan and
multi-disciplinary approach

Introduction teins, have been suggested as mechanisms for the increased

With improvements in medical care, the life expectancy
of patients with bleeding disorders, such as hemophilia,
is approaching that of the general population.1 As a result,
there is a growing population of older adult patients with
bleeding disorders in need of various elective and emer-
gent age-related procedures.1-3 These commonly include
orthopedic procedures such as joint arthroplasties and spi-
nal surgery as well as cardiac interventions such as percuta-
neous coronary intervention (PCI), coronary artery bypass
grafting (CABG), and heart valve replacement, all of which
are increasingly prevalent with age. In addition to the sig-
nificant bleeding and thrombotic risks that many of these
procedures carry, several challenges may further compli-
cate perioperative management in older adult patients in
comparison to children and young adults.
One such major challenge is the increased risk of throm-
bosis that occurs with age. Several age-related physiological
changes, such as vessel wall remodeling, endothelial dys-
function, venous stasis, and an increase in procoagulant pro-
Prakash Singh Shekhawat
thrombotic risk seen in older adults (Figure 1).4-9 Chronic
cardiovascular comorbidities, such as hypertension, diabe-
tes, and hypercholesterolemia, further exacerbate this risk.
Despite having bleeding tendencies, older adult patients with
bleeding disorders are prone to these age-related changes
that contribute to thrombosis and are not considered pro-
tected from these complications. While the incidence of
cardiovascular disease in hemophilia patients, for example,
has been reported to be less than that of the general popu-
lation (15% vs 25.8%),10 events such as myocardial infarction
and stroke still occur and should be prevented.2,3,10-13 Partic-
ular attention should be given to older adults with bleeding
disorders undergoing surgery in order to prevent periop-
erative thrombotic complications as well as bleeding. This
necessitates the judicious use of hemostatic agents, such as
coagulation factors or hemostatic bypass agents, that may
subject patients to a higher risk of thrombosis.14 In addition,
another considerable age-related risk is that of volume over-
load, which may occur as a complication of receiving the
Bleeding disorders and surgery in aging patients | 529

Figure 1. Proposed age-related mechanisms contributing to increased risk of thrombosis with age, based on Virchow’s triad. FVII,
factor VII; FVIII, factor VIII; Fg, fibrinogen; PAI-1, plasminogen activator inhibitor 1; VWF, von Willebrand factor.
large vol­umes of fresh fro­zen plasma (FFP) and plate­let trans­fu­sions
required in man­ag­ing cer­tain bleed­ing dis­or­ders.
Evidence-based guide­ lines for the man­ age­
ment of aging
patients with bleed­ ing dis­ or­
ders are lacking, largely due to
the under­rep­re­sen­ta­tion of older adult patients in clin­i­cal tri­als
as well as the rar­ity of many bleed­ing dis­or­ders. The opti­mal
perioperative man­age­ment of these patients is often com­plex
and requires a thor­ough mul­ti­dis­ci­plin­ary approach that involves
hema­tol­o­gists, sur­geons, anes­the­si­­ol­o­gists, and the spe­cial­ists
treating the under­ly­ing dis­or­der.15
CLINICAL CASE 1
An 88-year-old man with severe hemo­philia A (HA; base­line fac­tor
VIII [FVIII] activ­ity level <1%, with­out inhib­i­tors) has severe, debil­
i­tat­ing left hip arthrop­a­thy. He is oth­er­wise healthy and does not
have any other med­i­cal comorbidities. He self-admin­is­ters pro­
phy­lac­tic fac­tor replace­ment 3 times a week and receives his care
within a hemo­philia treat­ment cen­ter (HTC). He is eval­u­ated by
an ortho­pe­dic sur­geon and deemed a fit can­di­date for elec­tive
left total hip arthroplasty. The hema­tol­o­gist is consulted to assist
with perioperative man­age­ment to pre­vent bleed­ing com­pli­ca­
tions. The patient is treated with 50 U/kg of anti­he­mo­philic fac­tor
(Advate) before sur­gery and achieves a peak FVIII level of 108% and
good post­op­er­at­ive hemo­sta­sis. He is then maintained on daily
fac­tor replace­ment for the rest of the week and begins phys­ic ­ al
reha­bil­i­ta­tion. He is not placed on pro­phy­lac­tic anticoagulation.
530  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/529/1851841/529abou-ismail.pdf by guest on 13 December 2021
Intraoperative and post­op­er­a­tive hemo­sta­sis
In the above sce­nario, metic­u­lous hemo­static sup­port is cru­cial
for the suc­cess of the sur­gi­cal pro­ce­dure. The over­all hemo­static
man­age­ment and avail­­able agents vary sig­nif­i­cantly among
dif­fer­ent bleed­ing dis­or­ders, and we dis­cuss each sep­a­rately
below. However, regard­less of the type of bleed­ing dis­or­der, the
man­age­ment con­cepts are sim­i­lar: to pro­vide effec­tive and suf­
fi­cient hemo­static sup­port to pre­vent sur­gi­cal-related bleed­ing,
to sup­port wound healing, and to avoid side effects and com­
pli­ca­tions related to the ther­apy used. The World Federation of
Hemophilia (WFH) 2020 guide­ lines rec­
om­ mend that patients
with hemo­philia requir­ing sur­gery should always be man­aged
at or in con­sul­ta­tion with rec­og­nized HTCs with appro­pri­ate
hema­to­logic sup­port, reli­able lab­o­ra­tory mon­i­tor­ing, and avail­­
able and suf­fi­cient clot­ting fac­tor con­cen­trates.16 This con­cept
applies to patients with any bleed­ing dis­or­der. The HTC should
be supported by a mul­ti­dis­ci­plin­ary team that includes nurses,
social work­ers, and phys­i­cal ther­a­pists famil­iar with the needs of
hemo­philia patients under­go­ing sur­gery.17 As in the case above,
pro­ce­dures should be performed elec­tively when pos­si­ble, early
in the week, and early in the day by expe­ri­enced sur­geons and
anes­the­si­­ol­o­gists.15 Preoperative in vivo recov­ ery stud­ies are
used to eval­u­ate the phar­ma­co­ki­netic prop­er­ties of coag­u­la­tion
fac­tors, which may dif­fer based on the prod­uct and the patient.
In hemo­philia patients this allows for a more pre­cise esti­ma­tion
of peak FVIII and fac­tor IX (FIX) lev­els and can help tai­lor the dos­
ing and fre­quency of fac­tor replace­ment according to the need
of the indi­vid­ual patient, espe­cially before major sur­gery. The
pres­ence of inhib­i­tors should also be assessed prior to sur­gery.
Table 1. Suggested fac­tor replace­ment sched­ule for older adult per­sons with hemo­philia A or B with­out inhib­i­tors based on WFH
guide­lines
Hemophilia A
Indication Target peak lev­els (%) Duration
Minor sur­gery 50-80 Day of sur­gery
30-80 Days 1-5
Major sur­gery 80-100 Day of sur­gery
60-80 Days 1-3
40-60 Days 4-6
30-50 Days 7-14
Adapted with per­mis­sion from Srivastava et al.16
Hemophilia A
The hemo­static agents avail­­able for HA include desmopressin and
FVIII replace­ment. Desmopressin may be con­sid­ered in patients
with mild HA under­go­ing minor sur­gery in whom a response to
desmopressin has been dem­on­strated, par­tic­u­larly when the cost
or devel­op­ment of inhib­i­tors due to expo­sure to FVIII prod­ucts is
of con­cern. Issues with the use of desmopressin include free water
reten­tion, hyponatremia, fluid shifts, and exac­er­ba­tion of car­dio­
vas­cu­lar dis­ease. For those rea­sons it should be used with con­sid­
er­able cau­tion in older adult patients prone to vol­ume over­load
and car­dio­vas­cu­lar mor­bid­ity and is contraindicated in patients
with active car­dio­vas­cu­lar dis­ease or sei­zure dis­or­der.16,18
FVIII replace­ment ther­apy is the treat­ment of choice for all­
other patients with­out high-titer inhib­i­tors, such as the patient
in our case above. Suggested plasma FVIII tar­ get lev­ els are
pro­vided in Table 1, adapted from WFH guide­lines.16 However,
high-qual­ity data guid­ing this prac­tice are lacking.19 Addition-
ally, patients under­go­ing car­diac pro­ce­dures indi­cated for
antithrombotic treat­ment may require an extended dura­tion of
fac­tor replace­ment to allow for safe admin­is­tra­tion of the for­mer.
This is discussed sep­a­rately below.
Hemophilia B
While the over­all clin­i­cal man­i­fes­ta­tions and man­age­ment
approaches in hemo­philia B (HB) are sim­i­lar to those in HA, a
few dif­fer­ences should be noted. In con­trast to HA, the bleed­
ing phe­no­type in HB may be milder,19-25 and the recommended
tar­get lev­els in the WFH 2020 guide­lines are lower for major sur­
gery (Table 1).16 The dos­ing fre­quency should also con­sider the
lon­ger half-life of FIX. Patients with HB do not have desmopressin
as an avail­­able option, and FIX replace­ment is the treat­ment of
choice for major pro­ce­dures. While the inci­dence of inhib­i­tors
is much rarer in HB than in HA,26 their pres­ence should also be
assessed prior to sur­gery.
Patients with inhib­i­tors
Hemophilia patients with high-titer inhib­ i­
tors to FVIII, FIX, or
fac­tor XI (FXI) require the use of bypass agents, mainly recom­
bi­nant fac­tor VIIa (rFVIIa) or acti­vated pro­throm­bin com­plex con­
cen­trate (aPCC), which pose an addi­tional risk for throm­bo­sis,
espe­cially in older adults. Several stud­ies note that the risk of
throm­bo­sis due to these agents increases with age, and arte­
rial throm­bo­sis is a par­tic­u­lar con­cern.27-29 While the over­all inci­
Prakash Singh Shekhawat
Hemophilia B
Target peak lev­els (%) Duration
50-80 Day of sur­gery
30-80 Days 1-5
60-80 Day of sur­gery
40-60 Days 1-3
30-50 Days 4-6
20-40 Days 7-14
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/529/1851841/529abou-ismail.pdf by guest on 13 December 2021
dence of these events remains low and both rFVIIa and aPCC are
gen­er­ally con­sid­ered safe in older adults,27,30-32 patients should
be very cau­tiously mon­i­tored for throm­bo­sis. The lack of reli­able
ther­a­peu­tic mon­i­tor­ing param­e­ters for bypass agents fur­ther
com­pli­cates their opti­mal dos­ing strat­egy, and metic­u­lous clin­
i­cal eval­u­a­tion of bleed­ing and throm­bo­sis is cru­cial. Data from
the Euro­ pean Acquired Hemophilia Registry sug­ gests a sim­ i­
lar safety and effi­cacy pro­file for both rFVIIa and aPCC27; how­
ever, par­al­lel use of dif­fer­ent bypass agents should be avoided
because it may con­fer a higher risk for throm­bo­sis.33 Suggested
dos­ing of bypass agents is sum­ma­rized in Table 2.
A sim­i­lar approach applies to acquired HA, an auto­im­mune
dis­or­der caused by antibodies against FVIII that is more com­mon
in older adults.34 In addi­tion to bypass agents, these patients
also have the option of recom­bi­nant por­cine FVIII as a hemo­
static agent.34,35 While the recommended ini­tial dose of recom­
bi­nant por­cine FVIII is 200 U/kg, it has been suggested that an
ini­tial dose of 100 U/kg may be suf­fi­cient for most patients and
can be con­sid­ered for those at risk of throm­bo­sis.
Perioperative thromboprophylaxis
The WFH rec­om­mends against the rou­tine use of phar­ma­co­logic
thromboprophylaxis in patients with HA or HB under­go­ing major
non­car­diac sur­gery.16 For ortho­pe­dic pro­ce­dures, such as the case
above, the Amer­i­can Academy of Orthopaedic Surgeons and the
Amer­i­can College of Chest Physicians also sug­gest for­go­ing the
rou­tine use of antithrombotic agents in patients with bleed­ing
dis­or­ders, as does the WFH, with the excep­tion of high plasma
lev­els of coag­u­la­tion fac­tors.36 As pre­vi­ously men­tioned, how­ever,
older adult patients with bleed­ing dis­or­ders are not nec­es­sar­ily
protected from throm­botic out­comes, includ­ing venous or arte­
rial throm­bo­sis, and this risk may increase with age due to age-re-
lated prothrombotic phys­i­o­log­i­cal changes (Figure 1). In older
patients, an indi­vid­u­al­ized risk/ben­e­fit assess­ment is nec­es­sary
when addi­tional throm­botic risk fac­tors are pres­ent, espe­cially
in patients with corrected fac­tor lev­els after ortho­pe­dic sur­gery.
CLINICAL CASE 2
A 79-year-old woman with hyper­ ten­
sion, dia­
be­ tes, and type
2A von Willebrand dis­ ease (VWD; VWF [von Willebrand fac­
tor]:Ag = 35, VWF:RCo = 15, FVIII:C = 110) is sched­uled for elec­tive
Bleeding dis­or­ders and sur­gery in aging patients  |  531
Table 2. Suggested dos­ing reg­i­mens of bypass agents in hemo­philia patients with inhib­i­tors

Agent Dosing for minor sur­gery
rFVIIa 90 µg/kg/dose imme­di­ately before
sur­gery and every 2 hours for 2
days and then every 2-6 hours until
healed.
Dosing for major sur­gery
90 µg/kg/dose imme­di­ately before
sur­gery and every 2 hours for 2 days.
Then every 2-3 hours for 5 days, then
every 4 hours until days 7-10, then
every 6 hours until days 14-21
Comments
• Preferred agent for patients with HB
and high-titer inhib­i­tors per WFH
guide­lines since aPCC con­tains FIX
and may cause or worsen an aller­gic
reac­tion.
• For patients with FIX defi­ciency, low
dose (15-30 µg/kg) in com­bi­na­tion
with TXA has been shown to be
effec­tive for major sur­gery in lim­ited
stud­ies and can be con­sid­ered for
those at risk of throm­bo­sis.33

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/529/1851841/529abou-ismail.pdf by guest on 13 December 2021

Activated pro­throm­bin 50-100 U/kg imme­di­ately before 50-100 U/kg imme­di­ately before sur­ • Maximum: 100 units/kg/dose; 200
con­cen­trate com­plex sur­gery and then every 6-12 hours gery, then every 6-12 hours units/kg/d).
• Less pre­ferred for HB patients with
high-titer inhib­i­tors per WFH guide­
lines.

cho­le­cys­tec­tomy. She is admit­ted and receives 60 U/kg of VWF
con­cen­ trate prior to sur­ gery, achiev­ing VWF:RCo  = 104, and
FVIII:C = 225. Her sur­gi­cal course is uncom­pli­cated, but the next
morn­ing she com­plains of substernal chest pain. She is found to
have non-ST ele­va­tion myo­car­dial infarc­tion. With her VWF activ­
ity lev­els still nor­mal, she undergoes PCI and is found to have
severe tri­ ple ves­sel dis­ease. She is recommended for elec­ tive
CABG. The patient’s man­age­ment is ­discussed in a mul­ti­dis­ci­plin­
ary approach involv­ing car­di­­ol­ogy, car­dio­tho­racic sur­gery, and
hema­tol­ogy.
Von Willebrand dis­ease
The 2021 guide­lines by the Amer­i­can Society of Hematology,
International Society of Thrombosis and Hemostasis, National
Hemophilia Foundation, and WFH rec­om­mend a goal VWF activ­
ity and FVIII level of >50% for at least 3 days for major sur­ger­ies
using VWF/FVIII con­cen­trates.18 This is often extended to a dura­
tion of 7 to 14 days based on the indi­vid­ual type of sur­gery. In
older adults, spe­cial pre­cau­tion must be taken to avoid overdos-
ing of fac­tor replace­ment, as prolonged supraphysiologic FVIII
lev­els may lead to throm­botic com­pli­ca­tions. Available options
include plasma-derived for­ mu­la­
tions of VWF/FVIII as well as
recom­bi­nant VWF when avail­­able. Older adult patients may have
nor­mal or ele­vated base­line FVIII lev­els, and this should be taken
into con­sid­er­ation when choos­ing the appro­pri­ate VWF replace­
ment prod­ uct. This includes plasma-derived prod­ ucts with a
higher VWF to FVIII ratio, such as Humate-P, or dose adjust­ment
of con­com­i­tant FVIII replace­ment given with recom­bi­nant VWF.
Patients with type 1 VWD respon­sive to desmopressin with­out
active car­dio­vas­cu­lar dis­ease may be con­sid­ered for desmopressin
for minor sur­gery, with a goal VWF level >50 per the 2021 guide­
lines.18 The same risks and pre­cau­tions for desmopressin described
under HA above apply for VWD. Close mon­i­tor­ing for throm­botic
com­pli­ca­tions is nec­es­sary for older adult patients, espe­cially if
con­com­i­tant tranexamic acid (TXA) is given. Patients with mild
type 1 VWD under­go­ing minor muco­sal pro­ce­dures may be con­
sid­ered for TXA alone in some cases,18 which would min­i­mize the
risks asso­ci­ated with fac­tor replace­ment or desmopressin.
Several stud­ies on patients with type 1 VWD have dem­on­strated
that VWF lev­els may nor­mal­ize with age due to phys­i­o­log­i­cal
increases in VWF.37-42 The 2021 guide­ lines sug­
gest reconsider-
ing the diag­no­sis in these patients, as opposed to remov­ing it.43
Whether the nor­mal­i­za­tion of VWF lev­els in older patients results
in an ame­lio­ra­tion of bleed­ing symp­toms has not been established.
While that may be pos­si­ble in some patients, it has been sug-
gested that supranormal VWF lev­els may be required in oth­ers.44
Those with nor­mal­ized VWF lev­els in whom bleed­ing symp­toms
have resolved may be harmed by the unnec­ es­sary admin­
is­tra­
tion of VWF con­cen­trates, espe­cially if they have car­dio­vas­cu­lar
risk fac­tors.44 There are cur­rently insuf­fic
­ ient data to guide opti­mal
man­age­ment in these cases. An indi­vid­u­al­ized risk/ben­e­fit assess­
ment is cru­cial in patients with nor­mal­ized VWF lev­els, with care­ful
clin­i­cal eval­u­a­tion of bleed­ing and throm­botic risk fac­tors.
Acquired von Willebrand syn­drome may result from the shear­
ing of  VWF (eg, across a ste­notic car­diac valve or through mechan­
i­cal cir­cu­la­tory sup­port devices) or less com­monly may arise as an
auto­im­mune dis­or­der with antibodies directed against VWF, often
seen in patients with lymphoproliferative dis­or­ders and plasma
cell dys­cra­sias. When due to an auto­an­ti­body, acquired von Wil-
lebrand syn­drome can be effec­tively man­aged with intra­ve­nous
immune glob­ u­lin in addi­
tion to fac­tor replace­ ment for major
sur­gery, as well as the use of continuous-infusion factor replace-
ment.45,46,47 In older adult patients with a risk of throm­bo­sis or renal
dys­func­tion, approaches such as ther­a­peu­tic plasma exchange,
rFVIIa, and TXA have been uti­lized, although data are lacking.45
Cardiovascular inter­ven­tions
Percutaneous cor­o­nary inter­ven­tion
PCI is a com­mon pro­ce­dure among older adult patients, and its
chal­lenges include the need for intraoperative heparinization
as well as post­op­er­a­tive antiplatelet ther­apy. In patients with
bleed­ing dis­or­ders, a best effort should be made to min­i­mize
the dura­tion of dual antiplatelet ther­apy (DAPT). Earlier rec­om­
men­da­tions favored the use of bare metal stents over drug-elut­
ing stents to limit the dura­tion of DAPT to 1 month.48,49 However,
more recent data sug­gest that 1 month of DAPT may be con­
sid­ered with newer-gen­er­a­tion drug-elut­ing stents with biore-
sorbable poly­mers,50-52 which was the approach reported in a
recent case series of hemo­philia patients under­go­ing PCI.53 The
indi­vid­ual cor­o­nary anat­omy, risks of restenosis, and bleed­ing
pro­pen­sity should all­ be fac­tored into a mul­ti­dis­ci­plin­ary deci­

532  |  Hematology 2021  |  ASH Education Program

sion-mak­ing pro­cess. Due to the bleed­ing risk of DAPT, patients
with bleed­ing dis­or­ders may require pro­phy­lac­tic hemo­static
ther­apy through­out the dura­tion of DAPT, even if they had not
been receiv­ ing it before. For hemo­ philia patients, the WFH
guide­lines rec­om­mend trough lev­els of 15% to 30% of the defi­
cient fac­tor dur­ing the dura­tion of DAPT and lev­els of 1% to 5%
dur­
ing sin­ gle antiplatelet agent ther­ apy.16 Similarly, the 2021
VWD guide­lines rec­om­mend pro­phy­laxis with VWF con­cen­trate
dur­ing DAPT in patients with VWD and a severe bleed­ing phe­no­
type.18 In the patient above under­go­ing PCI, VWF:RCo lev­els >30
may be con­sid­ered through­out the dura­tion of DAPT.
Cardiac sur­gery
acquired coagulopathy. In high-risk patients, off-pump var­i­a­tions
of such pro­ce­dures should be con­sid­ered, or less inva­sive alter­
na­tives such as multivessel PCI in lieu of CABG. As with PCI, an
effort should be made to min­i­mize the dura­tion of DAPT when
pos­si­ble by weighing the risks and ben­e­fits of shorter dura­tions.
In patients requir­ing valve replace­ment, bioprosthetic valves
should be used instead of mechan­i­cal valves that require indef­i­
nite anticoagulation. An approach uti­liz­ing low-molec­u­lar-weight
hep­a­rin for 10 days post­op­er­a­tively followed by 3 months of
vita­min K antag­o­nist, with con­com­i­tant coag­u­la­tion fac­tor cor­
rec­tion through­out that dura­tion, has been suggested for hemo­
philia patients.54,55 A sim­i­lar approach can be implemented in
patients with other bleed­ing dis­or­ders, with cor­rec­tion of the

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/529/1851841/529abou-ismail.pdf by guest on 13 December 2021

Data on car­dio­vas­cu­lar pro­ce­dures such as CABG, valve replace­
ment, or device implan­ta­tion in patients with bleed­ing dis­or­ders
are lim­ited. Management is more com­plex than other pro­ce­dures
due to the fre­quent need for post­op­er­a­tive antithrombotic ther­
apy or the need for car­dio­pul­mo­nary bypass in cer­tain cases.
Procedures that may require car­dio­pul­mo­nary bypass, such as
CABG, con­sti­tute sev­eral high risks for patients with bleed­ing dis­
or­ders, includ­ing the need for sys­temic anticoagulation as well as
the pos­si­ble occur­rence of con­sump­tive coagulopathies.54 Con-
tinuous replace­ment of the defi­cient fac­tor may bet­ter main­tain
steady lev­els com­pared to IV boluses, and addi­tional hemo­static
prod­ucts such as FFP or plate­lets may be nec­es­sary in cases of
hemo­static defect applied through­out the dura­tion of anticoag-
ulation. An addi­tional con­sid­er­ation in HB patients is the height­
ened sever­ity of the dis­ease that can occur as a result of vita­min
K antag­o­nist ther­apy lead­ing to decreased FIX lev­els, as well as
the inter­fer­ence with inter­na­tional nor­mal­ized ratio mon­i­tor­ing
caused by FIX replace­ment.2
The lit­er­a­ture sug­gests that despite the major hemo­static
chal­lenges of var­i­ous car­diac inter­ven­tions, care­ful, mul­ti­dis­
ci­
plin­ary plan­ ning can help achieve opti­ mal out­ comes with
min­i­mal mor­bid­ity and mor­tal­ity in patients with bleed­ing dis­
or­ders.53,54,56-60 However, these data may be sub­ject to poten­tial
pub­li­ca­tion bias, and older adult patients are under­rep­re­sented.

Table 3. Hemostatic man­age­ment of rare bleed­ing dis­or­ders and pre­cau­tions in older adult patients

Recommended hemo­static Recommended hemo­static Suggested tar­get trough

Bleeding dis­or­der treat­ment for minor sur­gery treat­ment for major sur­gery level of defi­cient fac­tor
Rare fac­tor deficiencies
 II FFP (15-20 mL/kg) FFP (15-20 mL/kg) >30%
PCC* (20-30 U/kg)
 V FFP (15-20 mL/kg) FFP (15-20 mL/kg) >20%
  V + VIII FFP (15-20 mL/kg) FFP (15-20 mL/kg) Factor V: >20% FVIII:
as in Table 1
 VII rFVIIa (15-30 µg/kg rFVIIa (15-30 µg/kg every 4-6 hours) >20%
every 4-6 hours) PCC* (20-30 mL/kg)
 X Factor X con­cen­trate (10-20 U/kg) Factor X con­cen­trate (10-20 U/kg) >20%
FFP 15-20 mL/kg FFP (15-20 mL/kg)
PCC* (20-30 U/kg)
 XIII Plasma-derived fac­tor XIII con­cen­trate Plasma-derived fac­tor XIII con­cen­trate 5%
(40 U/kg) (40 U/kg)
Recombinant fac­tor XIII con­cen­trate Recombinant fac­tor XIII con­cen­trate
(35 U/kg) (35 U/kg)
Cryoprecipitate (1 bag per 10 kg) Cryoprecipitate (1 bag per 10 kg)
Other rare bleed­ing dis­or­ders
  Platelet func­tion defects Intravenous or oral TXA (500-1000 mg Platelet trans­fu­sions (4-6 units)
every 6-12 hours) Bypass agents†
Desmopressin
 Hypo/dysfibrinogenemia† Cryoprecipitate (1 bag per 10 kg) Cryoprecipitate (1 bag per 10 kg) >100 mg/dL
Fibrinogen con­cen­trate (20-30 mg/kg) Fibrinogen con­cen­trate (20-30 mg/kg)
FFP (15-20 mL/kg)
  Disorders of fibri­no­ly­sis Intravenous or oral TXA (500-1000 mg Intravenous TXA (10-15 mg/kg)
every 6-12 hours) every 6-12 hours
*After PCC treat­ment, fac­tors II, VI, IX, and X should not exceed 150%.
†Dysfibrinogenemia often pres­ents with a throm­botic phe­no­type. Hypofibrinogenemia and afibrinogenemia usu­ally pres­ent with bleed­ing, although
both venous and arte­rial throm­bo­sis can occur. Replacement rec­om­men­da­tions apply to patients with a bleed­ing phe­no­type, and con­com­i­tant
antithrombotic pro­phy­laxis may be nec­es­sary in some, par­tic­u­larly older adults.
Adapted with per­mis­sion from Mannucci et al.67
Prakash Singh Shekhawat
Bleeding dis­or­ders and sur­gery in aging patients  |  533
Management of such patients under­ go­
ing car­
diac inter­ ven­
tions requires a metic­u­lous, indi­vid­u­al­ized, and mul­ti­dis­ci­plin­ary
approach, with care­ful mon­i­tor­ing of hemo­static param­e­ters.
CLINICAL CASE 3
An 86-year-old man is eval­ua ­ ted by a urol­o­gist for con­sid­er­ation
of transurethral resec­tion of the pros­tate for a new diag­no­sis of
early-stage pros­tate can­cer. He has a his­tory of con­gen­i­tal FXI defi­
ciency, and his most recent tests show an FXI level of 24%. He has
a his­tory of hyper­ten­sion, chronic kid­ney dis­ease, aor­tic ste­no­sis,
and advanced heart fail­ure with reduced ejec­tion frac­tion. Because
of his prior his­tory of cir­cu­la­tory over­load with FFP, the option of
TXA in com­bi­na­tion with sin­gle low-dose rFVIIa is discussed.
FXI defi­ciency
Compared to HA and HB, FXI defi­ciency has its own unique chal­
lenges in man­age­ment. The poor cor­re­la­tion between FXI lev­els
and bleed­ing phe­no­type lim­its the abil­ity to pre­dict bleed­ing out­
comes.61 In gen­eral, any FXI level <20% is con­sid­ered severe.61 The
lack of FXI con­cen­trates in the United States neces­si­tates the use
of FFP infu­sions (usu­ally 10-20 mL/kg) to replace FXI periopera-
tively or the use of bypass agents or antifibrinolytic agents.59,61-63
In older adult patients, large vol­umes of FFP are a major risk fac­
tor for cir­cu­la­tory over­load, espe­cially in those with valve dis­ease
or con­ges­tive heart fail­ure. In such patients, major sur­gery would
require thor­ough fac­tor replace­ment and large vol­umes of plasma,
which may pose a substantial risk. Ther­a­peu­tic plasma exchange
has been uti­lized as a method to over­come this risk and should be
con­sid­ered as a poten­tially safer method of fac­tor cor­rec­tion.64,65
The opti­mal tar­get FXI level is unknown; how­ever, lev­els of >45%
have been recommended for major sur­gery, with con­sid­er­ation
of antifibrinolytic agents in areas of high fibri­no­lytic activ­ity, such
as the blad­der, endo­me­trium, and oro­phar­ynx.61 For patients who
can­not receive blood prod­ucts or are at risk of cir­cu­la­tory over­
load, as in the patient above, the use of a sin­gle low-dose rFVIIa
(Table 2) in com­bi­na­tion with TXA has been shown to be effec­tive
in lim­ited stud­ies.66
Other rare bleed­ing dis­or­ders
Very lim­ited data exist to guide the opti­mal perioperative man­
age­ment of older adults with rare bleed­ing dis­or­ders.66 The avail­
abil­ity of reli­able lab­o­ra­tory test­ing and appro­pri­ate hemo­static
prod­ucts is nec­es­sary in the man­age­ment of these patients as
guided by a hemo­sta­sis expert. Table 3 sum­ma­rizes some of the
con­cerns that may arise in using spe­cific hemo­static agents in
each of these dis­or­ders in older adult patients, and nec­es­sary
pre­cau­tions should be implemented.
Conflict-of-inter­est dis­clo­sure
Mouhamed Yazan Abou-Ismail: no com­pet­ing finan­cial inter­ests
to declare.
Nathan T. Connell: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Mouhamed Yazan Abou-Ismail: The use of bypass agents men-
tioned in this article, namely aPCC and rFVIIa, is considered
off-label in various rare hematologic conditions due to paucity
of data.
534  |  Hematology 2021  |  ASH Education Program
Nathan T. Connell: The use of bypass agents mentioned in this
article, namely aPCC and rFVIIa, is considered off-label in various
rare hematologic conditions due to paucity of data.
Correspondence
Nathan T. Connell, Hematology Division, Brigham and Women’s
Hospital, SR322, 75 Francis St, Bos­ton, MA 02115; e-mail: ntcon-
nell@bwh​­.harvard​­.edu.
References
1. Hassan S, Monahan RC, Mauser-Bunschoten EP, et al. Mortality, life expec­
tancy, and causes of death of per­sons with hemo­philia in the Netherlands
2001-2018. J Thromb Haemost. 2021;19(3):645-653.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/529/1851841/529abou-ismail.pdf by guest on 13 December 2021
2. Mannucci PM, Schutgens RE, Santagostino E, Mauser-Bunschoten EP. How
I treat age-related morbidities in elderly per­sons with hemo­philia. Blood.
2009;114(26):5256-5263.
3. Franchini M, Tagliaferri A, Mannucci PM. The man­age­ment of hemo­philia in
elderly patients. Clin Interv Aging. 2007;2(3):361-368.
4. Engbers MJ, Van Hylckama Vlieg A, Rosendaal FR. Venous throm­bo­sis in
the elderly: inci­dence, risk fac­tors and risk groups. J Thromb Haemost.
2010;8(10):2105-2112.
5. Nurmohamed MT, Büller HR, ten Cate JW. Physiological changes due to
age: impli­ca­tions for the pre­ven­tion and treat­ment of throm­bo­sis in older
patients. Drugs Aging. 1994;5(1):20-33.
6. Yamamoto K, Takeshita K, Kojima T, Takamatsu J, Saito H. Aging and
plas­min­o­gen acti­va­tor inhib­i­tor-1 (PAI-1) reg­u­la­tion: impli­ca­tion in the
path­o­gen­e­sis of throm­botic dis­or­ders in the elderly. Cardiovasc Res.
2005;66(2):276-285.
7. Kiechl S, Willeit J; Bruneck Study Group. The nat­u­ral course of ath­ero­scle­ro­
sis, II: vas­cu­lar remodeling. Arterioscler Thromb Vasc Biol. 1999;19(6):1491-
1498.
8. Balleisen L, Bailey J, Epping PH, Schulte H, van de Loo J. Epidemiological
study on fac­tor VII, fac­tor VIII and fibrin­o­gen in an indus­trial pop­u­la­tion, I:
base­line data on the rela­tion to age, gen­der, body-weight, smok­ing, alco­
hol, pill-using, and men­o­pause. Thromb Haemost. 1985;54(2):475-479.
9. Sagripanti A, Carpi A. Natural anti­co­ag­u­lants, aging, and throm­bo­em­bo­
lism. Exp Gerontol. 1998;33(7-8):891-896.
10. Sood SL, Cheng D, Ragni M, et al. A cross-sec­tional anal­y­sis of car­dio­vas­
cu­lar dis­ease in the hemo­philia pop­u­la­tion. Blood Adv. 2018;2(11):1325-
1333.
11. Girolami A, Ruzzon E, Fabris F, Varvarikis C, Sartori R, Girolami B. Myocardial
infarc­tion and other arte­rial occlu­sions in hemo­philia a patients: a car­di­o­
log­i­cal eval­u­a­tion of all­ 42 cases reported in the lit­er­a­ture. Acta Haematol.
2006;116(2):120-125.
12. Aledort LM. Comparative throm­ botic event inci­ dence after infu­ sion of
recom­bi­nant fac­tor VIIa ver­sus fac­tor VIII inhib­i­tor bypass activ­ity. J Thromb
Haemost. 2004;2(10):1700-1708.
13. Srámek A, Kriek M, Rosendaal FR. Decreased mor­tal­ity of ischaemic heart
dis­ease among car­ ri­
ers of haemophilia. Lancet. 2003;362(9381):351-
354.
14. O’Donnell JS, Lavin M. Perioperative man­age­ment of patients with von Wille-
brand dis­ease. Hematology Am Soc Hematol Educ Program. 2019;2019(1):
604-609.
15. Mensah PK, Gooding R. Surgery in patients with inherited bleed­ing dis­or­
ders. Anaesthesia. 2015;70(suppl 1):112-120, e39-40.
16. Srivastava A, Santagostino E, Dougall A, et al; WFH Guidelines for the Man-
agement of Hemophilia pan­el­ists and coau­thors. WFH guide­lines for the
man­age­ment of hemo­philia, 3rd edi­tion. Haemophilia. 2020;26 (suppl 6):1-
158.
17. Rodríguez-Merchán EC. The role of ortho­pae­dic sur­gery in haemophilia:
cur­rent ratio­nale, indi­ca­tions and results. EFORT Open Rev. 2019;4(5):165-
173.
18. Connell NT, Flood VH, Brignardello-Petersen R, et al. ASH ISTH NHF WFH
2021 guide­lines on the man­age­ment of von Willebrand dis­ease. Blood Adv.
2021;5(1):301-325.
19. Coppola A, Windyga J, Tufano A, Yeung C, Di Minno MN. Treatment for
pre­vent­ing bleed­ing in peo­ple with haemophilia or other con­gen­i­tal
bleed­ing dis­or­ders under­go­ing sur­gery. Cochrane Database Syst Rev.
2015;(2):Cd009961.
20. Ludlam CA, Lee RJ, Prescott RJ, et al. Haemophilia care in cen­tral Scotland
1980-94, I: demo­graphic char­ac­ter­is­tics, hos­pi­tal admis­sions and causes
of death. Haemophilia. 2000;6(5):494-503.
21. Soucie JM, Cianfrini C, Janco RL, et  al. Joint range-of-motion lim­i­ta­tions
among young males with hemo­philia: prev­a­lence and risk fac­tors. Blood.
2004;103(7):2467-2473.
22. Schulman S, Eelde A, Holmström M, Ståhlberg G, Odeberg J, Blombäck
M. Validation of a com­pos­ite score for clin­i­cal sever­ity of hemo­philia. J
Thromb Haemost. 2008;6(7):1113-1121.
23. Santagostino E, Mancuso ME, Tripodi A, et al. Severe hemo­philia with mild
bleed­ing phe­no­type: molec­u­lar char­ac­ter­iza­tion and global coag­u­la­tion
pro­file. J Thromb Haemost. 2010;8(4):737-743.
24. Nagel K, Walker I, Decker K, Chan AK, Pai MK. Comparing bleed fre­quency
and fac­tor con­cen­trate use between haemophilia A and B patients. Hae-
mophilia. 2011;17(6):872-874.
25. Mahlangu J, Powell JS, Ragni MV, et al; A-LONG Investigators. Phase 3 study
of recom­bi­nant fac­tor VIII Fc fusion pro­tein in severe hemo­philia A. Blood.
2014;123(3):317-325.
26. Ehrenforth S, Kreuz W, Scharrer I, et al. Incidence of devel­op­ment of fac­tor
VIII and fac­tor IX inhib­i­tors in haemophiliacs. Lancet. 1992;339(8793):594-
598.
27. Baudo F, Collins P, Huth-Kühne A, et al; Euro­pean Acquired Haemophilia
Registry Contributors. Management of bleed­ing in acquired hemo­philia A:
results from the Euro­pean Acquired Haemophilia (EACH2) Registry. Blood.
2012;120(1):39-46.
28. Levi M, Levy JH, Andersen HF, Truloff D. Safety of recom­bi­nant acti­vated
fac­tor VII in ran­dom­ized clin­i­cal tri­als. N Engl J Med. 2010;363(19):1791-1800.
29. Rajpurkar M, Croteau SE, Boggio L, Cooper DL. Thrombotic events with
recom­bi­nant acti­vated fac­tor VII (rFVIIa) in approved indi­ca­tions are rare
and asso­ci­ated with older age, car­dio­vas­cu­lar dis­ease, and con­com­i­tant
use of acti­vated pro­throm­bin com­plex con­cen­trates (aPCC). J Blood Med.
2019;2019(10):335-340.
30. Leebeek FW, Kappers-Klunne MC, Jie KS. Effective and safe use of recom­
bi­nant fac­tor VIIa (NovoSeven) in elderly mild haemophilia A patients with
high-titre antibodies against fac­tor VIII. Haemophilia. 2004;10(3):250-253.
31. Sumner MJ, Geldziler BD, Pedersen M, Seremetis S. Treatment of acquired
haemophilia with recom­bi­nant acti­vated FVII: a crit­i­cal appraisal. Haemo-
philia. 2007;13(5):451-461.
32. Yin EB, Tan B, Nguyen T, et al. Safety and effec­tive­ness of fac­tor VIII inhib­
i­tor bypassing activ­ity (FEIBA) and fresh fro­zen plasma in oral anti­co­ag­u­
lant-asso­ci­ated intra­cra­nial hem­or­rhage: a ret­ro­spec­tive anal­y­sis. Neurocrit
Care. 2017;27(1):51-59.
33. Ingerslev J, Sørensen B. Parallel use of by-pass­ing agents in haemophilia
with inhib­i­tors: a crit­i­cal review. Br J Haematol. 2011;155(2):256-262.
34. Janbain M, Leissinger CA, Kruse-Jarres R. Acquired hemo­philia A: emerg­ing
treat­ment options. J Blood Med. 2015;6(8 May):143-150.
35. Tiede A, Collins P, Knoebl P, et  al. International rec­ om­men­ da­ tions on
the diag­no­sis and treat­ment of acquired hemo­philia A. Haematologica.
2020;105(7):1791-1801.
36. Lieberman JR, Pensak MJ. Prevention of venous throm­bo­em­bolic dis­ease
after total hip and knee arthroplasty. J Bone Joint Surg Am. 2013;95(19):1801-
1811.
37. Sanders YV, Giezenaar MA, Laros-van Gorkom BA, et al; WiN Study Group.
von Willebrand dis­ease and aging: an evolv­ing phe­no­type. J Thromb Hae-
most. 2014;12(7):1066-1075.
38. Borghi M, Guglielmini G, Mezzasoma AM, et al. Increase of von Willebrand
fac­tor with aging in type 1 von Willebrand dis­ease: fact or fic­tion? Haema-
tologica. 2017;102(11):e431-e433.
39. Abou-Ismail MY, Ogunbayo GO, Secic M, Kouides PA. Outgrowing the lab­o­
ra­tory diag­no­sis of type 1 von Willebrand dis­ease: a two decade study. Am
J Hematol. 2018;93(2):232-237.
40. Flood VH, Christopherson PA, Gill JC, et al. Clinical and lab­o­ra­tory var­i­abil­
ity in a cohort of patients diag­nosed with type 1 VWD in the United States.
Blood. 2016;127(20):2481-2488.
41. Atiq F, Meijer K, Eikenboom J, et al; WiN Study Group. Comorbidities asso­
ci­ated with higher von Willebrand fac­tor (VWF) lev­els may explain the
age-related increase of VWF in von Willebrand dis­ease. Br J Haematol.
2018;182(1):93-105.
42. Rydz N, Grabell J, Lillicrap D, James PD. Changes in von Willebrand fac­tor
level and von Willebrand activ­ity with age in type 1 von Willebrand dis­
ease. Haemophilia. 2015;21(5):636-641.
43. James PD, Connell NT, Ameer B, et al. ASH ISTH NHF WFH 2021 guide­lines
on the diag­no­sis of von Willebrand dis­ease. Blood Adv. 2021;5(1):280-300.
44. Seaman CD, Ragni MV. The effect of age on von Willebrand fac­tor and
bleed­ing symp­toms in von Willebrand dis­ease. Thromb Haemost.
2020;120(8):1159-1165.
Prakash Singh Shekhawat
45. Abou-Ismail MY, Rodgers GM, Bray PF, Lim MY. Acquired von Willebrand
syn­drome in mono­clo­nal gammopathy—a scop­ing review on hemo­static
man­age­ment. Res Pract Thromb Haemost. 2021;5(2):356-365.
46. Tiede A. Diagnosis and treat­ment of acquired von Willebrand syn­drome.
Thromb Res. 2012;130(suppl 2):S2-S6.
47. Meizlish ML, et al. A neutrophil activation signature predicts critical illness
and mortality in COVID-19. Blood Adv. 2021;5(14):2813-2816.
48. Jabbar AY, Baydoun H, Janbain M, Ferdinand KC. Current con­cepts in the
man­age­ment of sta­ble ische­mic heart dis­ease and acute cor­o­nary syn­
drome in patients with hemo­philia. Ann Transl Med. 2018;6(15):299.
49. Staritz P, de Moerloose P, Schutgens R, Dolan G; ADVANCE Working Group.
Applicability of the Euro­pean Society of Cardiology guide­lines on man­
age­ment of acute cor­o­nary syn­dromes to peo­ple with haemophilia—an
assess­ment by the ADVANCE Working Group. Haemophilia. 2013;19(6):833-
840.
50. Varenne O, Cook S, Sideris G, et al; SENIOR Investigators. Drug-elut­ing
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/529/1851841/529abou-ismail.pdf by guest on 13 December 2021
stents in elderly patients with cor­o­nary artery dis­ease (SENIOR): a ran-
domised sin­gle-blind trial. Lancet. 2018;391(10115):41-50.
51. Urban P, Meredith IT, Abizaid A, et al; LEADERS FREE Investigators.
Polymer-free drug-coated cor­o­nary stents in patients at high bleed­ing risk.
N Engl J Med. 2015;373(21):2038-2047.
52. Valgimigli M, Patialiakas A, Thury A, et al; ZEUS Investigators. Zotarolimus-
elut­ ing ver­ sus bare-metal stents in uncer­ tain drug-elut­
ing stent can­di­
dates. J Am Coll Cardiol. 2015;65(8):805-815.
53. Cohen OC, Bertelli M, Manmathan G, et al. Challenges of antithrombotic
ther­apy in the man­age­ment of car­dio­vas­cu­lar dis­ease in patients with
inherited bleed­ing dis­or­ders: a sin­gle-cen­tre expe­ri­ence. Haemophilia.
2021;27(3):425-433.
54. Ferraris VA, Boral LI, Cohen AJ, Smyth SS, White GC II. Consensus review of
the treat­ment of car­dio­vas­cu­lar dis­ease in peo­ple with hemo­philia A and B.
Cardiol Rev. 2015;23(2):53-68.
55. Mannucci PM, Mauser-Bunschoten EP. Cardiovascular dis­ease in haemo-
philia patients: a con­tem­po­rary issue. Haemophilia. 2010;16(suppl 3):58-66.
56. Lin PS, Yao YT. Perioperative man­age­ment of hemo­philia A patients under­
go­ing car­diac sur­gery: a lit­er­a­ture review of published cases. J Cardiotho-
rac Vasc Anesth. 2021;35(5):1341-1350.
57. Fogarty PF, Mancuso ME, Kasthuri R, et al; Global Emerging Hemophilia
Panel. Presentation and man­age­ment of acute cor­o­nary syn­dromes among
adult per­sons with haemophilia: results of an inter­na­tional, ret­ro­spec­tive,
10-year sur­vey. Haemophilia. 2015;21(5):589-597.
58. Boehnel C, Rickli H, Graf L, Maeder MT. Coronary angi­og­ra­phy with or with­
out per­cu­ta­ne­ous cor­o­nary inter­ven­tion in patients with hemo­philia—sys­
tem­atic review. Catheter Cardiovasc Interv. 2018;92(1):1-15.
59. Shalabi A, Kachel E, Kogan A, et al. Cardiac sur­gery in patients with hemo­
philia: is it safe? J Cardiothorac Surg. 2020;15(1):76.
60. Lim MY, Abou-Ismail MY. Left atrial append­age occlu­sion for man­age­ment
of atrial fibril­la­tion in per­sons with hemo­philia. Thromb Res. 2021;206(Octo­
ber):9-13.
61. Bolton-Maggs PHB. Factor XI defi­ciency—resolv­ing the enigma? Hematol-
ogy. 2009;2009(1):97-105.
62. Gay ND, Azar S, Salomon O, Taylor JA. Management of severe fac­tor XI defi­
ciency in car­diac sur­gery: a case report and review of the lit­er­a­ture. Hae-
mophilia. 2017;23(6):e512-e514.
63. Royal College of Obstetricians and Gynecologists. Management of inher-
ited bleed­ing dis­or­ders in preg­nancy: green-top guide­line no. 71 (joint
with UKHCDO). BJOG. 2017;124(8):e193-e263.
64. Pagano MB, Konkle BA, Wu Y, Josephson N. Preoperative man­age­ment of
fac­tor XI defi­ciency with ther­a­peu­tic plasma exchange: a case report and
lit­er­a­ture review. J Clin Apher. 2016;31(6):579-583.
65. Alsammak MS, Ashrani AA, Winters JL, Pruthi RK. Therapeutic plasma
exchange for perioperative man­age­ment of patients with con­gen­i­tal fac­
tor XI defi­ciency. J Clin Apher. 2017;32(6):429-436.
66. Salomon O, Budnik I, Avishai E, et al. Single low dose of rFVIIa com­bined
with antifibrinolytic agent is a sim­ple and safe treat­ment for fac­tor XI-
defi­cient patients under­go­ing sur­gery. Thromb Haemost. 2019;119(12):1927-
1932.
67. Mannucci PM, Duga S, Peyvandi F. Recessively inherited coag­u­la­tion dis­or­
ders. Blood. 2004;104(5):1243-1252.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000288
Bleeding dis­or­ders and sur­gery in aging patients  |  535
 PERIOPERATIVE CONSULTATIVE HEMATOLOGY: CAN YOU CLEAR MY PATIENT FOR SURGERY ?

Heparin-induced thrombocytopenia
and cardiovascular surgery

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/536/1851441/536pishko.pdf by guest on 13 December 2021

Allyson M. Pishko1 and Adam Cuker1,2
Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA; and 2Department of Pathology and Laboratory
1

Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA

Clinicians generally counsel patients with a history of heparin-induced thrombocytopenia (HIT) to avoid heparin prod-
ucts lifelong. Although there are now many alternative (nonheparin) anticoagulants available, heparin avoidance remains
challenging for cardiac surgery. Heparin is often preferred in the cardiac surgery setting based on the vast experience
with the agent, ease of monitoring, and reversibility. To “clear” a patient with a history of HIT for cardiac surgery, hema-
tologists must first confirm the diagnosis of HIT, which can be challenging due to the ubiquity of heparin exposure and
frequency of thrombocytopenia in patients in the cardiac intensive care unit. Next, the “phase of HIT” (acute HIT, sub-
acute HIT A/B, or remote HIT) should be established based on platelet count, immunoassay for antibodies to platelet
factor 4/heparin complexes, and a functional assay (eg, serotonin release assay). As long as the HIT functional assay
remains positive (acute HIT or subacute HIT A), cardiac surgery should be delayed if possible. If surgery cannot be
delayed, an alternative anticoagulant (preferably bivalirudin) may be used. Alternatively, heparin may be used with either
preoperative/intraoperative plasma exchange or together with a potent antiplatelet agent. The optimal strategy among
these options is not known, and the choice depends on institutional experience and availability of alternative anticoag-
ulants. In the later phases of HIT (subacute HIT B or remote HIT), brief intraoperative exposure to heparin followed by an
alternative anticoagulant as needed in the postoperative setting is recommended.

LEARNING OBJECTIVES
• Recognize the phases of HIT and implications for heparin reexposure for CV surgery
• Understand the indications and potential alternative (nonheparin) anticoagulants for use in CV procedures and
surgeries

Introduction
Heparin­induced thrombocytopenia (HIT) is a highly pro­
thrombotic state resulting from pathogenic antibodies to
platelet factor 4/heparin (PF4/H) complexes.1 Clinicians
generally counsel patients who experience this potentially
life­threatening adverse reaction to never receive heparin
again. With the development of many alternative (non­
heparin) anticoagulants, avoiding heparin in most circum­
stances (eg, venous thromboembolism treatment) is not
difficult.2 Cardiovascular (CV) surgery is a unique scenario
in which heparin is highly preferred given the vast experi­
ence with the drug, the ease of monitoring with a point­of­
care assay (activated clotting time), and a readily available
reversal agent (protamine).3 It is not uncommon for hema­
tologists to be asked to “clear” a patient with a history of
HIT for CV surgery. Here we present our approach to eval­
uating and managing such patients.
CLINICAL CASE
A 45­year­old man with ischemic cardiomyopathy and a
history of left ventricular thrombosis receiving warfarin is
admitted with worsening dyspnea. Warfarin is held and
an unfractionated heparin infusion is started. He develops
acute thrombocytopenia on hospital day 7, and a lower
extremity ultrasound reveals a new popliteal vein throm­
bosis (Figure 1A). A 4Ts score is calculated to be 7 points
(high probability). The clinical team switches the heparin
to bivalirudin and sends HIT laboratory testing. The immu­
noglobulin G–specific PF4/H enzyme­linked immunosor­
bent assay (ELISA) is 2.2 optical density (OD) units (positive
result ≥0.4 units). A few days later, the serotonin release
assay (SRA) returns positive.

536 | Hematology 2021 | ASH Education Program

 A B C D
Phases of HIT

Acute HIT Subacute HIT A Subacute HIT B Remote HIT

HIT lab
tesng

PF4/H ELISA 2.2 OD 1.5 OD 1.0 OD <0.4 OD

SRA POSITIVE POSITIVE NEGATIVE NEGATIVE
Ancoagulant

Warfarin
Heparin Bivalirudin

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/536/1851441/536pishko.pdf by guest on 13 December 2021

250 Bivalirudin for PCI
225
200
Platelet count x

175
Plan for intraoperave
150
109/L

heparin for cardiac

125 Bivalirudin for
transplant then
100 LVAD placement
alternave
75 (nonheparin)
50 ancoagulant
postoperavely
25
0
1 2 3 4 5 6 7 8 9 10 15 16 17 18 19 20 21 22 52 58 75 100
Days from iniaon of heparin

Figure 1. Management of a patient undergoing PCI and cardiac surgery during multiple phases of HIT. (A) Patient develops a fall in
platelet count and lower extremity deep vein thrombosis 7 days after initiation of unfractionated heparin. The 4Ts score is 7. Heparin
is stopped and the patient is started on bivalirudin. HIT laboratory testing reveals a positive PF4/H ELISA and positive SRA. Acute HIT
is diagnosed. (B) At hospital day 15, the platelet count has recovered. The PF4/H ELISA and SRA remain positive, meeting criteria for
subacute HIT A. The patient undergoes left heart catheterization with bivalirudin. At hospital day 20, he remains in subacute HIT A
and requires LVAD placement that cannot be delayed. He receives bivalirudin during LVAD placement. Postprocedurally, he continues
receiving bivalirudin and is bridged to warfarin for discharge to home. (C) The patient is subsequently referred to a hematology clinic
for cardiac transplant evaluation. Repeat anti-PF4/H testing remains positive by ELISA (1.0 OD units) 45 days post-HIT diagnosis, but
the SRA is now negative, satisfying criteria for subacute HIT B. (D) Approximately 3 months after index admission for HIT, both PF4/H
ELISA and SRA are negative. The patient is listed for cardiac transplant with planned brief intraoperative heparin exposure followed
by treatment with an alternative anticoagulant postoperatively. PCI, percutaneous cardiac intervention; LVAD, left ventricular assist
device; PF4/H ELISA, Platelet factor-4/heparin Enzyme linked immunoassay; SRA, serotonin release assay.

Diagnosis of HIT in patients with CV dis­ease or can­celed. In extreme cases, HIT may be the comorbidity that
HIT is a highly feared iat­ro­genic com­pli­ca­tion of CV sur­gery, dur­ deters cli­ni­cians from pro­ceed­ing with life­sav­ing mea­sures such
ing which patients are nearly uni­ver­sally exposed to hep­a­rin.4 as list­ing for car­diac trans­plant. Although the diag­no­sis of HIT
Indeed, when HIT occurs post-CV sur­gery, it is asso­ci­ated with alone should never be an abso­lute con­tra­in­di­ca­tion to nec­es­sary
excess mor­bid­ity and mor­tal­ity. In a pro­pen­sity score–matched urgent car­diac inter­ven­tions, it undoubt­edly com­pli­cates oper­
study of 11 820 CV sur­gery patients, 29.1% of patients who devel­ a­tive man­age­ment. Thus, it is of utmost impor­tance to get the
oped HIT after car­diac sur­gery had a throm­bo­em­bolic event diag­no­sis “right.” The first step of any eval­u­a­tion for “his­tory of
(com­pared with 2.9% who did not develop HIT), and post­op­er­a­ HIT” is to obtain and review the clin­i­cal his­tory and lab­o­ra­tory
tive mor­tal­ity was 21.8% in patients with HIT (vs 5.3% in patients test­ing that led to the diag­no­sis. Heparin aller­gies placed on the
who did not develop HIT). Fortunately, in this study, as in oth­ers, chart for suspected HIT are often not removed after HIT has been
HIT was uncom­mon, occur­ring in 1.1% of CV sur­gery patients.5 ruled out.7 Clarifying the HIT diag­no­sis is vital to deter­min­ing the
Educational ini­tia­tives have increased aware­ness of what can anti­co­ag­u­lant strat­egy for sur­gery.
be a hor­ren­dous com­pli­ca­tion fol­low­ing even the most rou­tine In car­diac inten­sive care unit or CV sur­gery patients, cau­tion
of CV sur­ger­ies. However, they have also con­trib­uted to an epi­ is needed when cal­cu­lat­ing the 4Ts score (Table 1). First, most
demic of overdiagnosis.6 The ram­i­fi­ca­tions of overdiagnosis in patients will have a fall in plate­let count fol­low­ing place­ment of
patients with CV dis­ease are seri­ous; as such, patients fre­quently intra­vas­cu­lar devices, car­dio­pul­mo­nary bypass (CPB), or extra­cor­
require sur­ger­ies or pro­ce­dures neces­si­tat­ing hep­a­rin use. A po­real mem­brane oxy­gen­a­tion that can mimic the degree of fall
patient with a his­tory of HIT may havePrakash
urgent pro­cSingh Shekhawat
e­dures delayed typ­i­cal of HIT (≥30%-50% from base­line).8 With intra-aor­tic bal­

HIT and CV sur­gery  | 537

Table 1. Special con­sid­er­ations in appli­ca­tion of the 4Ts score to car­diac sur­gery patients

Component of 4Ts score Cardiac sur­gery pop­u­la­tion con­sid­er­ations

Timing of plate­let count fall
Thrombocytopenia
Thrombosis or other sequelae
Other causes of throm­bo­cy­to­pe­nia
A biphasic plate­let count pat­tern is typ­i­cal of HIT after CPB sur­gery
Early-onset and per­sis­tent throm­bo­cy­to­pe­nia after CPB is rarely HIT
Percent fall in plate­let count is cal­cu­lated from the highest plate­let count that imme­di­ately pre­cedes the
puta­tive HIT-related plate­let count decline, to the nadir value
The nadir plate­let count need not fall below 150 × 109/L, par­tic­u­larly in sur­gi­cal patients who have post­op­
er­a­tive thrombocytosis
Digital ische­mia due to hypo­ten­sion, vaso­pres­sors, non-HIT DIC, and/or under­ly­ing periph­eral arte­rial
dis­ease may mimic HIT-asso­ci­ated small-ves­sel throm­bo­em­bo­lism
Common alter­na­tive causes in car­diac patients include CPB, IABP, ECMO, LVAD, infec­tion, and other med­i­
ca­tions that can cause throm­bo­cy­to­pe­nia (eg, gly­co­pro­tein IIb/IIIa antag­o­nists)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/536/1851441/536pishko.pdf by guest on 13 December 2021

Adapted from Pishko and Cuker4 with per­mis­sion from Georg Thieme Verlag KG.
CPB, cardiopulmonary bypass; IABP, intra-aortic balloon pump; DIC, dis­sem­i­nated intra­vas­cu­lar coag­u­la­tion; ECMO, extra­cor­po­real mem­brane oxy­
gen­a­tion; LVAD, left ventricular assist device.

loon pump place­ment, the plate­let count falls a mean 40% from
base­line shortly after place­ment.9 Left ven­tric­u­lar assist devices
(LVADs) also cause plate­let acti­va­tion and shear stress, lead­ing
to throm­bo­cy­to­pe­nia.10 Following CPB sur­gery, 30% to 50% of
patients develop throm­bo­cy­to­pe­nia, with a typ­i­cal decrease
of 50% and a pre­dict­able nadir 48 to 72 hours post­op­er­a­tively.8
Thrombocytopenia that occurs soon after cardiac surgery and
persists is rarely indicative of HIT in cardiac surgery patients.11
Rather, a “biphasic fall,” in which the plate­let count recov­ers after
the 48- to 72-hour post-CPB fall and then sud­denly drops around
days 5 to 10 after hep­a­rin expo­sure is more indic­a­tive of HIT.12 Sec­
ond, “dusky dig­its” are com­monly due to prolonged vaso­pres­sor
use, arte­rial line place­ment, periph­eral arte­rial dis­ease, non-HIT
dis­sem­i­nated intra­vas­cu­lar coag­u­la­tion, and/or poor car­diac out­
put, not throm­bo­sis.4 There are other pre­dic­tion scores devised
for the CV sur­gery pop­u­la­tion (Lillo–Le Louet score)13 or that incor­
po­rate fea­tures such as the use of an intra-aor­tic bal­loon pump or
CPB within the past 96 hours (HIT Expert Probability [HEP] score).14
These scores have not been con­sis­tently proven to have higher
diag­nos­tic accu­racy than the 4Ts score in pro­spec­tive stud­ies.14
However, they may be use­ful in con­junc­tion with the 4Ts score,
par­tic­u­larly in high­light­ing “other” causes of throm­bo­cy­to­pe­nia.
Determining an accu­rate pre­test prob­a­bil­ity of HIT is cru­cial
as HIT anti­body test­ing can be mis­lead­ing in the CV sur­gery pop­
u­la­tion. Up to 20% of patients screened before CPB sur­gery had
detect­ able anti-PF4/H antibodies (most were the non­ patho­
genic non–immu­ no­ glob­ u­
lin G type). Detection of these anti­
bodies was not asso­ci­ated with an increase in adverse events.15
PF4/H antibodies also fre­ quently develop post­ op­
er­a­tively; a
mul­ti­cen­ter study of approximately 1000 CV sur­gery patients
reported a 50% sero­con­ver­sion rate as mea­sured using a poly­
specific PF4/H ELISA at 30 days after sur­gery.16 Seroconversion
was not asso­ci­ated with increased death or throm­bo­em­bo­lism.
Thus, it is essen­tial to send HIT lab­o­ra­tory test­ing only in patients
with suf­fi­ciently high clin­i­cal sus­pi­cion of HIT.
CLINICAL CASE (Con­t in­u ed)
The patient con­tin­ues tak­ing bivalirudin for con­firmed HIT. His
plate­let count recov­ers to base­line approx­i­ma­tely 1 week after
diag­no­sis. Unfortunately, the patient’s car­diac sta­tus wors­ens.
An echo­car­dio­gram shows new left ven­tric­u­lar wall motion
abnor­mal­i­ties. Cardiology rec­om­mends a left heart cath­e­ter­i­za­
tion (Figure 1B).
Phases of HIT: selec­tion of anti­co­ag­ul­ant for car­diac
pro­ce­dures/sur­gery
HIT gen­er­ally fol­lows a pre­dict­able course, with the plate­let
count returning to base­line in most patients within 7 days fol­low­
ing dis­con­tin­u­a­tion of hep­a­rin, followed by neg­a­tive func­tional
assay at a median of 50 days, and dis­ap­pear­ance of anti-PF4/H
antibodies at a median of 85 days.17,18 The risk of hep­ar­in reex­
posure depends on the “phase of HIT” (Table 2).19,20 The highest
risk is before plate­let count recov­ery and when the func­tional
assay remains pos­i­tive. Recurrence of HIT with hep­a­rin reexpo­
sure in the “remote HIT” phase after the immu­no­as­say becomes
neg­a­tive appears to be low.19 However, as most patients avoid
hep­a­rin after a HIT diag­no­sis, this risk is not well char­ac­ter­ized.
PCI: acute HIT or his­tory of HIT
There are robust data for alter­na­tive anti­co­ag­u­lants, par­tic­u­
larly bivalirudin, for per­cu­ta­ne­ous car­diac inter­ven­tions (PCIs)
(Table 3). The Amer­i­can College of Cardiology/Amer­i­can Heart
Association guide­ line gives both hep­ ar­in and bivalirudin a
class I indi­ca­tion for use dur­ing pri­mary PCI in patients with
ST-ele­va­tion myo­car­dial infarc­tion (STEMI).21 A recent unpub­
lished meta-anal­y­sis iden­ti­fied 8 ran­dom­ized con­trolled tri­als
com­par­ing bivalirudin to hep­ar­in for PCI for STEMI or non-ST
ele­va­tion myo­car­dial infarc­tion. In the pooled data, bivaliru­
din was asso­ci­ated with a reduc­tion in seri­ous bleed­ing rates
com­pared with hep­a­rin and, in the STEMI sub­group, a 30-day
mor­tal­ity ben­e­fit as well.22,23 One pro­spec­tive study of bivaliru­
din dur­ing PCI exclu­sively in patients with HIT reported “pro­
ce­dural suc­cess” in 98% of patients; only 1 of 52 patients (1.9%)
had major bleed­ing.24 There are also data supporting the use
of bivalirudin in other min­i­mally inva­sive car­diac pro­ce­dures,
includ­ing transcathether aor­tic valve replace­ment. The effect
of Bivalirudin on Aortic Valve Intervention 3 trial ran­dom­ized
802 patients (HIT and non-HIT) to bivalirudin vs hep­a­rin dur­
ing transcathether aor­tic valve replace­ment and found sim­i­lar

538  |  Hematology 2021  |  ASH Education Program

Table 2. Phases of acute HIT
Phases Platelet count Functional assay
Suspected HIT Decreased
Acute HIT Decreased
Subacute HIT A Normal
Subacute HIT B Normal
Remote HIT Normal
Adapted from Cuker19 with per­mis­sion from Georg Thieme Verlag KG.
NA, not appli­ca­ble.
rates of major bleed­ing and adverse CV out­comes.25 Among
patients with a his­tory of HIT, there was no sig­nif­i­cant dif­fer­
ence in bleed­ing rates in the hep­a­rin and bivalirudin groups
(9.0% vs 10.4%).25
Other alter­na­tive anti­co­ag­u­lants stud­ied for PCI include arg­
atroban and danaparoid (not avail­­ able in the United States).
These agents have less data than bivalirudin but may be con­sid­
ered when there is a lack of avail­abil­ity or pro­ce­dural expe­ri­ence
with bivalirudin. In 1 open-label study of 91 patients with HIT who
received argatroban for PCI, the “pro­ce­dural suc­cess” rate was
94%.26 There are reports of 61 cases of danaparoid use for inva­
sive vas­cu­lar pro­ce­dures (mostly PCI), but out­comes were not
reported.27
In acute HIT/sub­acute HIT A, Amer­i­can Society of Hematol­
ogy (ASH) 2018 HIT guide­lines sug­gest using bivalirudin over
other alter­na­tive anti­co­ag­u­lants for PCI. If bivalirudin is not avail­­
able, argatroban may also be con­sid­ered.2 In later phases of HIT,
sub­acute HIT B/remote B, the panel also sug­gests bivalirudin
over hep­a­rin use for PCI because mul­ti­ple stud­ies show sim­i­lar
safety and effi­cacy to hep­a­rin. However, this is sub­ject to the
avail­abil­ity of the agent and insti­tu­tional expe­ri­ence.
CLINICAL CASE (Con­t in­u ed)
The patient undergoes car­diac cath­e­ter­i­za­tion while receiv­ing
bivalirudin. No intervenable cor­o­nary lesion is iden­ti­fied. The
patient remains crit­i­cally ill. The car­diac team rec­om­mends a
LVAD. On hos­pi­tal day 18, HIT lab­o­ra­tory test­ing is repeated.
The PF4/H ELISA is 1.5 OD units, and the SRA remains pos­i­tive
(Figure 1B).
Cardiac sur­gery: acute HIT/sub­acute HIT A
Although data sup­port bivalirudin for PCI, hep­a­rin remains the
highly pre­ferred agent for most other CV pro­ce­dures/sur­ger­
ies. Thus, hema­tol­o­gists are asked to deter­mine the safety of
reexposure to hep­a­rin. Figure 2 describes our approach to se­
lecting an anti­co­ag­u­lant strat­egy for CV sur­gery in patients with
HIT. Due to the pref­er­ence for hep­a­rin in car­diac sur­ger­ies, ASH
Prakash Singh Shekhawat
Time from dis­con­tin­u­a­tion
Immunoassay
of hep­a­rin
? ? NA
+ + NA
+ + Majority recover plate­let
count by 7 days
− + Median 50 days for func­tional
assay to become neg­a­tive
− − Median 85 days for PF4/
H antibodies to become
unde­tect­able
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/536/1851441/536pishko.pdf by guest on 13 December 2021
HIT guide­lines rec­om­mend delaying CV sur­gery until patients
enter at least the sub­acute HIT B phase, if fea­si­ble. For car­diac
sur­gery that can­not be delayed, these guide­lines sug­gest 1 of
3 options2:
1. intraoperative anticoagulation with bivalirudin,
2. intraoperative hep­a­rin after treat­ment with pre­op­er­a­tive
and/or intraoperative ther­a­peu­tic plasma exchange (TPE), and
3. intraoperative hep­a­rin with a potent antiplatelet agent (eg,
pros­ta­cy­clin/tirofiban).
There is grow­ing expe­ri­ence with direct thrombin inhibitors
(DTIs), par­tic­u­larly bivalirudin, in CV sur­gery (Table 3). The Evalua­
tion of Patients dur­ing Coronary Artery Bypass (EVOLUTION-ON)
study com­pared hep­a­rin with bivalirudin for patients under­go­ing
coronary artery bypass graft (CABG) with CPB (his­tory of HIT was
not required).28 “Procedural suc­cess” was not sig­nif­i­cantly dif­fer­
ent between study arms, nor was cumu­la­tive median blood loss
within the first 24 hours. Notably, 6.1% of patients who received
bivalirudin required repeat explor­ atory oper­ a­
tions com­ pared
with 1.9% of patients who received hep­a­rin. Limitations of the
study were small sam­ple size and var­i­a­tion in trans­fu­sion thresh­
olds between insti­tu­tions. Similarly, the EVOLUTION-OFF study
showed sim­i­lar rates of major adverse events in patients under­
go­ing elec­tive “off-pump” car­diac sur­gery with bivalirudin com­
pared with hep­a­rin.29 A pro­spec­tive sin­gle-arm study of bivalirudin
only in patients with a his­tory of HIT who required CPB sur­gery
enrolled 49 patients (43 with acute HIT and 6 with a his­tory of HIT).
“Procedural suc­cess” was achieved in 42 of 49 patients (85.7%).30
In a “real-world” ret­ro­spec­tive study, 13 patients received a DTI
for CPB sur­gery, mostly CABG and/or valve sur­gery.31 Patients
who received DTI for CPB had no dif­fer­ence in mor­tal­ity, throm­
bo­sis, or hem­or­rhage com­pared with those who received hep­a­
rin. The authors cau­tioned that patients treated with a DTI may
have been at intrin­si­cally lower bleed­ing risk or may have under­
gone lower-risk pro­ce­dures than those reexposed to hep­a­rin.31
Another strat­egy is to reexpose patients with acute HIT/sub­
acute HIT A to hep­a­rin intraoperatively fol­low­ing TPE +/− intra­
ve­nous immu­no­glob­u­lin (IVIG). In a clin­i­cal cohort of 24 patients
under­go­ing TPE for HIT before hep­a­rin use for CPB, 3 non-HIT-
related deaths and 3 throm­bo­em­bolic events occurred.32 Nota­
bly, a National Inpatient Sample data­ base study iden­ ti­
fied 90
patients with HIT who received TPE; less than one-fourth under­
HIT and CV sur­gery  | 539
 Table 3.  Alternative anti­co­ag­u­lants for car­diac inter­ven­tions and CV sur­gery

Recommendation on use by pro­ce­dure*

Drug Mechanism T1/2 Dosing† Monitoring Additional pre­cau­tions
PCI CPB sur­gery Non-CPB sur­gery
Bivalirudin Direct 25 min (nor­mal to Preferred alter­na­ Preferred alter­na­tive Preferred alter­na­ PCI: 0.75 mg/kg PCI: weight-based Prolonged half-life in
throm­bin mildly impaired tive anti­co­ag­u­lant anti­co­ag­u­lant if tive anti­co­ag­ bolus followed by dos­ing, ACT patients with renal
inhib­i­tor renal func­tion) in patients with a acute HIT/sub­ u­lant if acute 1.75 mg/kg/h for CPB†: ACT >2.5 times impair­ment
34 min (mod­er­ate his­tory of HIT acute HIT A and HIT/sub­acute dura­tion of pro­ce­ base­line Avoid sta­sis in the CPB
renal impair­ sur­gery can­not be HIT A and sur­ dure Non-CPB†: ACT cir­cuit
ment) delayed gery can­not be CPB‡: 1 mg/kg bolus >300 s Hypothermia should
57 min (severe renal delayed then 2.5 mg/kg/h be min­i­mized to
impair­ment) +50 mg added to avoid drug accu­mu­
3.5 h (dial­y­sis) prim­ing solu­tion la­tion
for CPB
Non-CPB‡: 0.75 mg/kg
bolus then
1.75 mg/kg/h
Argatroban Direct 39-51 min (nor­mal Bivalirudin pre­ferred Not gen­er­ally Bivalirudin pre­ PCI: begin infu­sion of PCI: ACT 300-450 s Prolonged half-life
throm­bin hepatic func­tion) but may be con­ recommended ferred but may 25 µg/kg/min and CPB: NR in patients with
inhib­i­tor 181 min (hepatic sid­ered based for use in CPB be con­sid­ admin­is­ter bolus of Non-CPB†: ACT hepatic impair­ment

540  |  Hematology 2021  |  ASH Education Program

impair­ment) on insti­tu­tional ered based on 350 µg/kg (over 3- 200-300 s Case reports describe
expe­ri­ence insti­tu­tional 5 min). Then adjust prolonged bleed­
expe­ri­ence infu­sion rate based ing after stop­ping
on ACT drawn 10 argatroban at end
mins following of CPB
bolus as fol­lows: Case reports describe
ACT <300 s: give clot for­ma­tion in
addi­tional cir­cuit with use of
150 µg/kg bolus, argatroban despite
and increase rate ther­a­peu­tic ACT
to 30 µg/kg/min
(recheck ACT in 5-
10 mins)
ACT >450 s: decrease
infu­sion rate to
15 µg/kg/min
(recheck ACT in
5-10 min)
CPB: NR
Non-CPB‡: NR. Infusion
of 2-5 µg/kg/min
with­out bolus
30-60 min prior to
sur­gery to main­tain
ACT 200-300 s has
been described

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/536/1851441/536pishko.pdf by guest on 13 December 2021

 Table 3.  Alternative anti­co­ag­u­lants for car­diac inter­ven­tions and CV sur­gery (Continued)

Recommendation on use by pro­ce­dure*

Drug Mechanism T1/2 Dosing† Monitoring Additional pre­cau­tions
PCI CPB sur­gery Non-CPB sur­gery
Danaparoid§ Indirect 25 h (nor­mal renal Not recommended Not recommended Case reports PCI: NR Anti-Xa Inappropriately long
fac­tor Xa func­tion) have been CPB: NR half-life for CPB
inhib­i­tor 29-35 h (renal described, but Non-CPB‡: NR. Bolus sur­gery
impair­ment) not a pre­ferred of 40 U/kg for non-
agent when CPB sur­gery has
other alter­na­ been described.
tive anti­co­ag­u­
lants avail­­able

Prakash Singh Shekhawat

*Recommendation based on opin­ion of authors as well as the ASH 2018 HIT guide­lines (note certainty of evi­dence was “low” for PCI and “very low” for CPB and non-CPB).
†
Dosing as described in US Food and Drug Administration (FDA) label for approved indi­ca­tions. For off-label use, exam­ple of dos­ing reg­i­men described in selected stud­ies and/or case reports.
Dose adjustments may be needed for hepatic or renal dysfunction depending on the agent. Refer to insti­tu­tional dos­ing algo­rithms when avail­­able.
‡
Off-label use of the med­i­ca­tion. Not FDA approved for this indi­ca­tion.
§
Not avail­­able in the United States.
ACT, activated clotting time; NR, not recommended.

HIT and CV sur­gery  | 541

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/536/1851441/536pishko.pdf by guest on 13 December 2021
 Paent with reported acute HIT or history of HIT in need of
cardiovascular surgery

Review inial HIT presentaon and HIT lab tesng

Current (or prior) presentaon and lab tesng consistent with No Proceed with CV surgery with

Not HIT
diagnosis of HIT? heparin exposure +
postoperave heparin use
Yes
Obtain Platelet Count, Immunoassay (ex. PF4/H ELISA),
and Funconal Assay (ex. Serotonin Release Assay)

Platelets LOW or RECOVERED,

POSITIVE Funconal Assay & Platelets RECOVERED &

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/536/1851441/536pishko.pdf by guest on 13 December 2021

POSITIVE Immunoassay NEGATIVE Funconal Assay &

Subacute HIT B/Remote HIT

POSITIVE OR NEGATIVE Immunoassay
Yes Delay unl
Acute HIT/subacute HIT A

Can surgery be delayed?

funconal assay
NEGATIVE Intraoperave ancoagulaon: Heparin
No
Intraoperave ancoagulaon: one of 3
opons
1. Bivalirudin Postoperave ancoagulaon: alternave
2. TPE before and/or during CPB with (nonheparin) ancoagulant
heparin
3. Heparin + potent anplatelet agent

Postoperave ancoagulaon: alternave

(nonheparin) ancoagulant

Figure 2. Approach to the management of HIT or history of HIT in cardiac surgery patients. For patients with a history of HIT who
require cardiac surgery, the records should first be reviewed to confirm the diagnosis. The platelet count, anti-PF4/H immunoassay,
and functional assay (e.g., SRA) determine the “phase of HIT” and the risks of heparin reexposure.

went TPE for car­diac sur­gery (CPB or non-CPB). Outcomes fol­
low­ing car­diac sur­gery were not reported, but over­all, TPE with
HIT com­pared with not receiv­ing TPE with HIT was asso­ci­ated
with a higher like­li­hood of major bleed­ing (mostly gas­tro­in­tes­ti­
nal bleed­ing) and higher length of stay (20.5 vs 10 days).31 Other
cen­ters have suc­cess­fully used IVIG and TPE before emer­gent CV
sur­gery.33,34
A third strat­egy for patients with acute/sub­acute HIT A requir­
ing CV sur­gery is intraoperative hep­a­rin in com­bi­na­tion with plate­
let inhi­bi­tion. Multiple case stud­ies have described this approach,
using a vari­ety of potent antiplatelet agents.35,36 A recent report
describes suc­cess­ful hep­a­rin reexposure in a patient with acute
HIT for CPB sur­gery com­bin­ing antiplatelet ther­apy and IVIG.37
As the 3 afore­men­tioned strat­e­gies have not been directly
com­pared, the ASH guide­lines do not rec­om­mend 1 strat­egy
over the oth­ers. Rather, the selec­tion of strat­egy should depend
on insti­tu­tional expe­ri­ence and sur­gi­cal pref­er­ence. If the sec­ond
or third strat­egy is cho­sen, patients’ expo­sure to hep­a­rin should
be strictly lim­ited to the intraoperative set­ting. Before and after
sur­gery, they should receive an alter­na­tive anti­co­ag­u­lant.
CLINICAL CASE (Con­t in­u ed)
With a pos­ i­
tive immu­
no­
as­
say and func­ tional assay but nor­
mal plate­let count, the patient is diag­nosed with “sub­acute
HIT A.” After dis­cus­sion with CV sur­gery and anes­the­sia, the
patient undergoes LVAD place­ment with bivalirudin infu­sion.
Postprocedurally, he con­tin­ues on bivalirudin and is ulti­mately
discharged on war­fa­rin. The patient pres­ents to the out­pa­tient
hema­tol­ogy clinic for car­diac trans­plant eval­u­a­tion. Repeat
test­ing 45 days post-HIT diag­no­sis shows PF4/H ELISA remains
pos­i­tive (1.0 OD units), but the SRA is now neg­a­tive (Figure 1C).
Two months later, both PF4/H ELISA and SRA are neg­at­ ive (Fig­
ure 1D). The patient is listed for car­diac trans­plant with planned
brief intraoperative hep­a­rin expo­sure.
Reexposure in sub­acute B/remote HIT
After a patient with HIT no lon­ger has a pos­it­ive func­tional
assay and/or no lon­ger has PF4/H ELISA antibodies, the risk
of devel­op­ing HIT with intraoperative hep­a­rin expo­sure ap­
pears to be low.38 Some groups have hypoth­e­sized that the
very high doses of intraoperative intra­ve­nous hep­a­rin do not
invoke the same plate­let-acti­vat­ing response as lower con­cen­
tra­tions, even if HIT antibodies are pres­ent.39 Warkentin and
Sheppard38 described 17 patients with a his­tory of HIT who
received intraoperative hep­a­rin (but no post­op­er­a­tive hep­a­
rin) for car­diac or vas­cu­lar sur­gery between 8 weeks and 13.5
years after HIT diag­no­sis. All patients had a neg­at­ive PF4/H
ELISA and SRA pre­op­er­a­tively. Anti-PF4/H anti­body pos­i­tiv­
ity and SRA pos­i­tiv­ity were com­mon fol­low­ing sur­gery (9/17

542  |  Hematology 2021  |  ASH Education Program

Table 4. Alternative anti­co­ag­u­lants for venous throm­bo­em­bo­lism pro­phy­laxis and con­sid­er­ations in the post­op­er­a­tive set­ting
in patients with a his­tory of HIT

Alternative Dosing for venous throm­bo­em­bo­lism

Considerations in post­op­er­a­tive set­ting
anti­co­ag­u­lants pro­phy­laxis
Fondaparinux 2.5 mg sub­cu­ta­ne­ously once daily Elimination half-life 17-21 h
Not read­ily revers­ible
Use with cau­tion in patients with renal impair­ment (CrCl 30-50 mL/min). Avoid use
if CrCl <30 mL/min
Reduced clear­ance in patients <50 kg
Caution against use in unsta­ble patients who may require urgent sur­ger­ies/
pro­ce­dures
Apixaban* 2.5 mg oral twice daily Elimination half-life 12 h
Reversal agent may not be read­ily avail­­able and is not cur­rently FDA approved for

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/536/1851441/536pishko.pdf by guest on 13 December 2021

use prior to sur­gery
Use with cau­tion in hepatic impair­ment (contraindicated in Child Pugh C cir­rho­sis)
Caution against use in unsta­ble patients who may require urgent sur­ger­ies/
pro­ce­dures
Rivaroxaban* 10 mg oral daily Elimination half-life 5-9 h
Caution with renal impair­ment, contraindicated in acute renal fail­ure
Use with cau­tion in hepatic impair­ment (contraindicated in Child Pugh C cir­rho­sis)
Caution against use in unsta­ble patients who may require urgent sur­ger­ies/
pro­ce­dures
*Off-label use. Drug is not approved by the US Food and Drug Administration (FDA) for venous thromboembolism pro­phy­laxis in hos­pi­tal­ized
nonorthopedic sur­gi­cal patients.

[53%] and 8/17 [47%], respec­tively). Despite the high rate of
sero­con­ver­sion, recur­rent HIT occurred in only 1 patient, and
this appeared to be a case of “auto­im­mune HIT” char­ac­ter­ized
by strong HIT antibodies that also acti­vated plate­lets in the
absence of hep­a­rin.
Given this evi­dence and the vast expe­ri­ence with hep­a­rin in
CV sur­gery, ASH HIT guide­lines rec­om­mend using intraoperative
hep­a­rin in patients with remote HIT/sub­acute HIT B.2 Again, hep­
a­rin expo­sure should be lim­ited to the intraoperative set­ting and
scru­pu­lously avoided before and after sur­gery.19,20 The expected
period for HIT recur­rence is 5 to 10 days after intraoperative hep­
a­rin expo­sure, and thus the plate­let count should be mon­i­tored
dur­ing this time.20
Acknowledgment
Allyson M. Pishko is supported by a 2019 HTRS Mentored Re­
search Award from the Hemostasis and Thrombosis Research
Society (HTRS), which was supported by an educational grant
from Sanofi Genzyme.
Conflict-of-inter­est dis­clo­sure
Allyson M. Pishko has received research funding from an edu­ca­
tional grant from Sanofi Genzyme and Novo Nordisk. Adam Cuker
has served as a con­sul­tant for Synergy and has received roy­al­ties
from UpToDate, and his insti­tu­tion has received research sup­port
on his behalf from Alexion, Bayer, Novartis, Novo Nordisk, Pfizer,
Sanofi, and Spark.

Off-label drug use

Postoperative anticoagulation in patients with a his­tory
of HIT
Postoperatively, regard­less of the phase of HIT, hep­a­rin should
be avoided. Table 4 out­lines alter­na­tive anti­co­ag­u­lants for ve­
nous throm­bo­em­bo­lism pro­phy­laxis and con­sid­er­ations in the
post­op­er­a­tive set­ting.
Conclusions
Clinicians should be aware of the many chal­lenges in diag­nos­ing
HIT in the CV pop­u­la­tion and order HIT lab­o­ra­tory test­ing judi­
ciously. Ideally, CV sur­ gery that requires hep­ a­rin is delayed in
a patient with acute HIT/sub­acute HIT A until the plate­let count
recov­ers and the func­tional assay becomes neg­a­tive. If delay is
not pos­si­ble, options include intraoperative bivalirudin, TPE prior
to and/or dur­ing sur­gery with intraoperative hep­a­rin, or intraop­
erative hep­a­rin in com­bi­na­tion with a potent antiplatelet agent.
Once the patient progresses to sub­acute HIT B and beyond, the
ben­e­fits of intraoperative hep­a­rin use out­weigh the risks of brief
hep­a­rin expo­sure. In all­cases, hep­a­rin should be avoided pre- and
post­op­er­a­tively.
Prakash Singh Shekhawat
Allyson M. Pishko: Off-label use of fondaparinux and direct oral
anticoagulants for treatment of heparin-induced thrombocyto­
penia are discussed in the manuscript. Off-label use of alterna­
tive anticoagulants in cardiac bypass surgery discussed.
Adam Cuker: Off-label use of fondaparinux and direct oral anti­
coagulants for treatment of heparin-induced thrombocytopenia
are discussed in the manuscript. Off-label use of alternative anti­
coagulants in cardiac bypass surgery discussed.
Correspondence
Allyson M. Pishko, Department of Medicine, Perelman School of
Medicine, University of Pennsylvania, 3400 Spruce St, 3rd Floor Dull­
es, Philadelphia, PA 19104; e-mail: allyson​­.pishko@pennmedicine​
­.upenn​­.edu.
References
1. Arepally GM. Heparin-induced throm­bo­cy­to­pe­nia. Blood. 2017;129(21):2864-
2872.
2. Cuker A, Arepally GM, Chong BH, et  al. Amer­i­can Society of Hematology
2018 guide­lines for man­age­ment of venous throm­bo­em­bo­lism: hep­a­rin-
induced throm­bo­cy­to­pe­nia. Blood Adv. 2018;2(22):3360-3392.
HIT and CV sur­gery  | 543
3. Shore-Lesserson L, Baker RA, Ferraris V, et  al. STS/SCA/AmSECT clin­ic ­ al
prac­tice guide­lines: anticoagulation dur­ing car­dio­pul­mo­nary bypass. J
Extra Corpor Technol. 2018;50(1):5-18.
4. Pishko AM, Cuker A. Heparin-induced throm­bo­cy­to­pe­nia in car­diac sur­gery
patients. Semin Thromb Hemost. 2017;43(7):691-698.
5. Brown JA, Aranda-Michel E, Kilic A, et al. Outcomes with hep­a­rin-induced
throm­bo­cy­to­pe­nia after car­diac sur­gery. Ann Thorac Surg. 2021;112(2):487-
493.
6. Cuker A. Heparin-induced throm­bo­cy­to­pe­nia (HIT) in 2011: an epi­demic of
overdiagnosis. Thromb Haemos. 2011;106(6):993-994.
7. McMahon CM, Tanhehco YC, Cuker A. Inappropriate doc­ u­men­ta­tion of
hep­a­rin allergy in the med­i­cal record because of mis­di­ag­no­sis of hep­a­
rin-induced throm­bo­cy­to­pe­nia: fre­quency and con­se­quences. J Thromb
Haemost. 2017;15(2):370-374.
8. Griffin BR, Bronsert M, Reece TB, et al. Thrombocytopenia after car­dio­pul­
mo­nary bypass is asso­ci­ated with increased mor­bid­ity and mor­tal­ity. Ann
Thorac Surg. 2020;110(1):50-57.
9. Vonderheide RH, Thadhani R, Kuter DJ. Association of throm­bo­cy­to­pe­nia
with the use of intra-aor­tic bal­loon pumps. Am J Med. 1998;105(1):27-32.
10. Muslem R, Caliskan K, Leebeek FWG. Acquired coagulopathy in patients
with left ven­tric­u­lar assist devices. J Thromb Haemost. 2018;16(3):429-440.
11. Selling S, Malowsky B, Strobel U, et al. Early-onset and persisting throm­bo­
cy­to­pe­nia in post-car­diac sur­gery patients is rarely due to hep­a­rin-induced
throm­bo­cy­to­pe­nia, even when anti­body tests are pos­i­tive. J Thromb Hae-
most. 2010;8(1):30-36.
12. Pouplard C, May MA, Regina S, Marchand M, Fusciardi J, Gruel Y. Changes
in plate­let count after car­diac sur­gery can effec­tively pre­dict the devel­
op­ment of path­o­genic hep­a­rin-depen­dent antibodies. Br J Haematol.
2005;128(6):837-841.
13. Louet AL-L, Boutouyrie P, Alhenc-Gelas M, et al. Diagnostic score for hep­
a­rin-induced throm­bo­cy­to­pe­nia after car­dio­pul­mo­nary bypass. J Thromb
Haemost. 2004;2(11):1882-1888.
14. Pishko AM, Fardin S, Lefler DS, et  al. Prospective com­par­i­son of the HEP
score and 4Ts score for the diag­no­sis of hep­a­rin-induced throm­bo­cy­to­pe­
nia. Blood Adv. 2018;2(22):3155-3162.
15. Selleng S, Selleng K. Heparin-induced throm­bo­cy­to­pe­nia in car­diac sur­
gery and crit­i­cally ill patients. Thromb Haemost. 2016;116(5):843-851.
16. Welsby IJ, Krakow EF, Heit JA, et al. The asso­ci­a­tion of anti-plate­let fac­tor
4/hep­a­rin antibodies with early and delayed throm­bo­em­bo­lism after car­
diac sur­gery. J Thromb Haemost. 2016;15(1):57-65.
17. Warkentin TE, Kelton JG. Temporal aspects of hep­a­rin-induced throm­bo­cy­
to­pe­nia. N Engl J Med. 2001;344(17):1286-1292.
18. Mattioli AV, Bonetti L, Zennaro M, Ambrosio G, Mattioli G. Heparin/PF4
antibodies for­ma­tion after hep­a­rin treat­ment: tem­po­ral aspects and long-
term fol­low-up. Am Heart J. 2009;157(3):589-595.
19. Cuker A. Management of the mul­ti­ple phases of hep­a­rin-induced throm­bo­
cy­to­pe­nia. Thromb Haemost. 2016;116(05):835-842.
20. Warkentin TE, Anderson JAM. How I treat patients with a his­tory of hep­ar­ in-
induced throm­bo­cy­to­pe­nia. Blood. 2016;128(3):348-359.
21. O’Gara PT, Kushner FG, Ascheim DD, et  al. 2013 ACCF/AHA guide­ line
for the man­age­ment of ST-ele­va­tion myo­car­dial infarc­tion. Circulation.
2013;127(4):e362-e425.
22. Stone GW. Individual patient data pooled anal­y­sis of ran­dom­ized tri­als of
bivalirudin ver­sus hep­a­rin in acute myo­car­dial infarc­tion. Presented at: TCT
Connect; Octo­ber 2020; (virtual meeting).
23. Bikdeli B, McAndrew T, Crowley A, et al. Individual patient data pooled anal­
y­sis of ran­dom­ized tri­als of bivalirudin ver­sus hep­a­rin in acute myo­car­dial
infarc­tion: ratio­nale and meth­od­ol­ogy. Thromb Haemost. 2019;120(2):348-
362.
24. Mahaffey KW, Lewis BE, Wildermann NM, et al. The anti­co­ag­u­lant ther­apy
with bivalirudin to assist in the per­for­mance of per­cu­ta­ne­ous cor­o­nary
544  |  Hematology 2021  |  ASH Education Program
inter­ven­tion in patients with hep­a­rin-induced throm­bo­cy­to­pe­nia (ATBAT)
study: main results. J Invasive Cardiol. 2003;15(11):611-616.
25. Dangas GD, Lefèvre T, Kupatt C, et al. Bivalirudin ver­sus hep­a­rin anticoag­
ulation in transcatheter aor­tic valve replace­ment the ran­dom­ized BRAVO-3
trial. J Am Coll Cardiol. 2015;66(25):2860-2868.
26. Lewis BE, Matthai WH, Cohen M, et al. Argatroban anticoagulation dur­
ing per­cu­ta­ne­ous cor­o­nary inter­ven­tion in patients with hep­a­rin-induced
throm­bo­cy­to­pe­nia. Catheter Cardiovasc Intev. 2002;57(2):177-184.
27. Magnani H, Gallus A. Heparin-induced throm­bo­cy­to­pe­nia (HIT). A report of
1,478 clin­i­cal out­comes of patients treated with danaparoid (Orgaran) from
1982 to mid-2004. Thromb Haemost. 2006;95(6):967-981.
28. Dyke CM, Smedira NG, Koster A, et al. A com­par­i­son of bivalirudin to hep­
a­rin with prot­amine rever­sal in patients under­go­ing car­diac sur­gery with
car­dio­pul­mo­nary bypass: the EVOLUTION-ON study. J Thorac Cardiovasc
Surg. 2006;131(3):533-539.
29. Smedira NG, Dyke CM, Koster A, et  al. Anticoagulation with bivaliru­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/536/1851441/536pishko.pdf by guest on 13 December 2021
din for off-pump cor­ o­nary artery bypass grafting: the results of the
EVOLUTION-OFF study. J Thorac Cardiovasc Surg. 2006;131(3):686-692.
30. Koster A, Dyke CM, Aldea G, et  al. Bivalirudin dur­ing car­dio­pul­mo­nary
bypass in patients with pre­vi­ous or acute hep­a­rin-induced throm­bo­cy­to­
pe­nia and hep­a­rin antibodies: results of the CHOOSE-ON trial. Ann Thorac
Surg. 2007;83(2):572-577.
31. Carlson D, Bartholomew J, Gomes M, McCrae K, Chaturvedi S. Outcomes
of car­dio­vas­cu­lar sur­gery uti­liz­ing hep­a­rin ver­sus direct throm­bin inhib­i­
tors in car­dio­pul­mo­nary bypass in patients with pre­vi­ously diag­nosed HIT.
Thromb Haemost. 2019;120(2):300-305.
32. Moreno-Duarte I, Cooter M, Onwuemene OA, Ghadimi K, Welsby IJ.
Clinical out­comes of car­diac sur­gery patients under­go­ing ther­a­peu­
tic plasma exchange for hep­a­rin-induced throm­bo­cy­to­pe­nia. Vox Sang.
2021;116(2):217-224.
33. Soares Ferreira Júnior A, Boyle SH, Kuchibhatla M, Akinyemiju T, Onwuemene
OA. Use of ther­a­peu­tic plasma exchange in hep­a­rin-induced throm­bo­cy­to­
pe­nia: a pop­u­la­tion-based study. J Clin Apher. 2021;36(3):398-407.
34. Liu VC, Klompas AM, Stulak JM, Yalamuri SM. The tri­ple HIT: periopera­
tive man­age­ment of hep­a­rin-induced throm­bo­cy­to­pe­nia using plasma
exchange, intra­ve­nous immu­no­glob­u­lin, and prot­amine infu­sion for left
ven­tric­u­lar assist device implan­ta­tion [published online March 30, 2021].
J Cardiothor Vasc Anesth.
35. Palatianos G, Michalis A, Alivizatos P, et  al. Perioperative use of iloprost
in car­diac sur­gery patients diag­nosed with hep­a­rin-induced throm­bo­cy­
to­pe­nia-reac­tive antibodies or with true HIT (HIT-reac­tive antibodies plus
throm­bo­cy­to­pe­nia): an 11-year expe­ri­ence. Am J Hematol. 2015;90(7):608-
617.
36. Koster A, Kukucka M, Bach F, et  al. Anticoagulation dur­ing car­dio­pul­mo­
nary bypass in patients with hep­a­rin-induced throm­bo­cy­to­pe­nia type II
and renal impair­ment using hep­a­rin and the plate­let gly­co­pro­tein IIb-IIIa
antag­o­nist tirofiban. Anesthesiology. 2001;94(2):245-251.
37. Koster A, Nazy I, Birschmann IE, Smith JW, Sheppard JI, Warkentin TE. High-
dose IVIG plus cangrelor plate­let “anes­the­sia” dur­ing urgent hep­a­rin-CPB
in a patient with recent SRA-neg­a­tive HIT-throm­bo­sis with persisting plate­
let-acti­vat­ing antibodies. Res Pract Thromb Haemost. 2020;4(6):1060-
1064.
38. Warkentin TE, Sheppard J-AI. Serological inves­ti­ga­tion of patients with a
pre­vi­ous his­tory of hep­a­rin-induced throm­bo­cy­to­pe­nia who are reexposed
to hep­a­rin. Blood. 2014;123(16):2485-2493.
39. Warkentin TE. Acute intraoperative HIT dur­ing heart sur­gery: why so rare?
Thromb Res. 2016;146(6, suppl):110-112.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000289
 PREGNANCY IN SPECIAL POPULATIONS: CHALLENGES AND SOLUTIONS

How to evaluate and treat the spectrum of TMA

syndromes in pregnancy

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/545/1851434/545scully.pdf by guest on 13 December 2021

Marie Scully
Department of Haematology, University College London Hospitals NHS Foundation Trust and Cardiometabolic Programme-NIHR UCLH/UC BRC,
London, UK

Thrombotic microangiopathy (TMA) is the broad definition for thrombocytopenia, microangiopathic hemolytic anemia,
and end-organ damage. Two important categories are thrombotic thrombocytopenic purpura (TTP) and complement-
mediated hemolytic-uremic syndrome (CM-HUS). Pregnancy and the immediate postpartum period are associated with
TMAs specific to pregnancy in rare situations. These include pregnancy-induced hypertension, preeclampsia, and hemo-
lysis, elevated liver enzymes, and low platelets. TTP and CM-HUS may present in pregnancy. However, the diagnosis
may not be immediately obvious as they share characteristics of pregnancy-related TMAs. Within this review, we discuss
investigations, differential diagnosis of TMAs in pregnancy, and management. The importance is a risk of maternal mor-
tality but also poor fetal outcomes in relation to TTP and CM-HUS. Treatment of these disorders at presentation in preg-
nancy is discussed to achieve remission and prolong fetal viability if possible. In subsequent pregnancies, a treatment
pathway is presented that has been associated with successful maternal and fetal outcomes. Critical to this is a multidis-
ciplinary approach involving obstetricians, the fetal medicine unit, and neonatologists.

LEARNING OBJECTIVES
• TTP and CM-HUS are acute life-threatening disorders, require prompt diagnosis and treatment, and may be com-
plicated by PE or HELLP.
• Congenital TTP needs considering in pregnancy. Subsequent pregnancies require multidisciplinary team monitor-
ing and ADAMTS-13 replacement.

Introduction TTP or CM-HUS needs urgent therapy, initially with plasma

Thrombotic microangiopathies (TMAs) presenting in preg-
nancy or the postpartum period can be difficult to diagnose
and have an impact on both maternal and fetal outcomes,
and therapies are dependent on the clinical condition. The
treatment for pregnancy-related TMAs, such as pregnancy-
induced hypertension (PIH), preeclampsia (PE) or hemolysis,
elevated liver enzymes, and low platelets (HELLP), includes
control of blood pressure and delivery. Very rarely, acute
fatty liver of pregnancy (AFLP) may present with some TMA
features.2 Diagnosis and treatment of TMA in pregnancy may
be challenging, as it is dependent on the time of presen-
tation during pregnancy and potential fetal viability at this
stage. Furthermore, presentation, for example, of PE can
occur in 1% to 20% of women in the postpartum period,
also typical of complement-mediated hemolytic-uremic
syndrome (CM-HUS).3 Alternatively, features suggestive of
PIH, PE, or HELLP may in fact be the result of thrombotic
thrombocytopenic purpura (TTP) or CM-HUS.4 Diagnosis of
Prakash Singh Shekhawat
exchange but thereafter specific ongoing treatment, a dis-
integrin and metalloproteinase with a thrombospondin type
1 motif–member 13 (ADAMTS-13) replacement, immunosup-
pression, or complement inhibitor therapy, respectively.5
CLINICAL CASE
A 21-year-old woman presented at her 28-week checkup
in her first pregnancy. Routine bloods revealed a reduced
platelet count (21 × 109/L) and raised blood pressure of
135/110 mm Hg. She was admitted for further investigation.
Cardiotocography (CTG) revealed fetal distress. Remain-
ing relevant bloods included alanine aminotransferase
(ALT) of 100 IU (normal range [NR], 10–35 IU) and urine
protein/creatinine ratio of 432 mg/mmol (NR, <15).
What is the diagnosis? How was this established, and
what was the management in this situation?
Pregnancy TMAs | 545
Diagnosis of preg­nancy-related TMA
TMA in preg­ nancy is suggested by plate­ let counts less than
100 × 109/L, a raised LDH (lactate dehydrogenase) typ­i­cally twice
the upper limit of nor­mal, and blood film find­ings suggesting
ane­mia, polychromasia, throm­bo­cy­to­pe­nia, and fragmented red
blood cells. There is often organ involve­ment related to ische­mic
dam­age affect­ing the kid­neys, pri­mar­ily in CM-HUS, and brain
and heart in TTP (Table 1). However, multiorgan involve­ment can
be seen in any of the TMAs asso­ci­ated with preg­nancy. There is
the addi­tional impact on pla­cen­tal func­tion, which may man­i­fest
as evi­dence of reduced fetal growth, impaired uter­ine Dopp­ler
flow, abnor­mal CTGs, or even in utero fetal death (IUFD).6
Further lab­o­ra­tory anal­y­sis to assist with the diag­no­sis of
TMAs include raised bil­i­ru­bin, which is uncon­ju­gated; evi­dence of
renal impair­ment; urine pro­tein/cre­at­i­nine esti­mate; a coag­u­la­
tion screen; a direct anti­glob­ul­in test; urate lev­els; retic­ul­o­cytes;
and liver func­tion tests (Table 2). Autoimmune screen—includ­ing
ANA and ENA; com­ple­ment lev­els, spe­cif­i­cally C3 and C4; and
ADAMTS-13 activ­ity lev­els—may aid dif­fer­en­ti­at­ing the under­ly­
ing mech­a­nism.
In rela­tion to case 1, her retic­u­lo­cyte count was 5.32% (NR,
0.45% –2.42%); cre­at­i­nine, 80    µmol/L (NR, 49-92); eGFR, 83   mL/min
(NR, >90); LDH, 558 (214 IU upper limit of normal); hemo­glo­bin,
100 g/L (NR, 115-155); white blood cell count,  15.8 × 109/L  (NR, 3-10);
and ADAMTS-13 activ­ity, 14 IU/dL (NR,>60). Anti-nuclear antibody
and extractable nuclear antigen were neg­a­tive, with C3 and C4
within the NR.
Pregnancy-asso­ci­ated TMAs
The diag­no­sis of pre­eclamp­sia is defined by a blood pres­sure
of 140/90 mm Hg or higher after 20 weeks’ ges­ta­tion and evi­
dence of acute kidney injury (serum cre­ at­i­
nine >90   
µmol/L)
or pro­tein­uria.7 Serum urate lev­ els are increased. Pregnancy
induced hypertension occurs in patients with blood pres­sure
lev­els not in the pre­eclamp­sia range but above the patients’
base­line lev­els.8 HELLP is con­sid­ered in patients with epi­gas­
tric or right upper quad­ rant pain and ala­ nine ami­no­ trans­
fer­
ase (ALT) or aspar­tate ami­no­trans­fer­ase (AST) lev­els more than
40  IU/L. Alkaline phos­pha­tase lev­els are increased phys­i­o­log­i­
cally in preg­nancy (Table 2). Eclampsia is a severe con­di­tion in
which there may be altered men­tal state, severe head­aches,
blind­ness, stroke, or sei­zures. AFLP is a rare and life-threat­en­ing
con­di­tion asso­ci­ated with liver fail­ure, multiorgan involve­ment,
and coagulopathy. It is an impor­tant dif­fer­en­tial given its acute
pre­sen­ta­tion and multiorgan involve­ment but dif­fer­en­ti­ated by
Table 1. Key fea­tures dif­fer­en­ti­at­ing preg­nancy-asso­ci­ated and preg­nancy-related TMAs
Incidence/
Characteristic preg­nan­cies Thrombocytopenia
Preeclampsia 1/20 +++
HELLP 1/1000 ++++
TTP 1/200,000 +++
CM-HUS 1/25,000 +++
Note: the degree of involvement is demonstrated by an increasing number of +.
Table adapted from Scully.9
HBP, hypertension.
546  |  Hematology 2021  |  ASH Education Program
the coag­u­la­tion abnor­mal­i­ties. The con­di­tion appears to be
the result of mito­chon­drial dys­func­tion in the oxi­da­tion of fatty
acids in the liver. Diagnosis can be con­firmed by test­ing for the
enzyme long-chain 3-hydroxyacyl-CoA dehy­dro­ge­nase.10
Thrombocytopenia in preg­nancy, defined as below the nor­mal
lab­o­ra­tory range, typ­i­cally 150 to 400 × 109/L, is not uncom­mon
and 70% to 80% of all­cases of throm­ bo­cy­to­
pe­
nia in preg­
nancy are due to ges­ta­tional throm­bo­cy­to­pe­nia, which rarely
decreases below 75 × 109/L or is asso­ci­ated with immune dis­ease
(eg, immune thrombocytopenic purpura [ITP] in 3%–4%).11 Throm-
bocytopenia asso­ci­ated with hyper­ten­sive dis­ease accounts for
20% of all­cases,12 but plate­let lev­els less than 30 × 109/L would
be a sig­nal to exclude CM-HUS or TTP. Similarly, with increas­ing
cre­at­i­nine and wors­en­ing AKI with oliguria/anuria, one should
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/545/1851434/545scully.pdf by guest on 13 December 2021
con­sider a diag­no­sis of CM-HUS, and neu­ro­logic fea­tures should
prompt con­sid­er­ation of TTP. AKI can be dif­fi­cult to diag­nose in
preg­nancy, given the reduc­tion in cre­at­i­nine lev­els dur­ing the
preg­nant state. However, guid­ance sug­gests AKI in preg­nancy
is diag­nosed with a dou­bling in cre­at­i­nine from base­line at the
begin­ning of preg­nancy or a 26-µmol/L increase and, there­fore, a
cre­at­i­nine more than 90   µmol/L or a 25% increase from base­line.13
TTP presenting in preg­nancy
TTP is asso­ci­ated with severe ADAMTS-13 defi­ciency. ADAMTS-13
is required for cleav­age of von Willebrand fac­tor (VWF) into mul-
timeric forms. In preg­nancy, there are increases in von Willebrand
factor and fac­tor VIII through­out each tri­mes­ter of preg­nancy,
which remain raised in the post­par­tum period. There is a rel­a­tive
decrease in ADAMTS-13 lev­els but usu­ally within the NR lev­els:
this is a phys­i­o­logic effect.14,15 Pregnancy is asso­ci­ated with LDH
lev­els within the NR, and increases are related to cell dam­age or
ische­mia.16 Presentation of TTP in preg­nancy can be con­fused by
the pres­ence of hyper­ten­sion and throm­bo­cy­to­pe­nia, but symp­
toms may also be like those of patients with TTP out­side of preg­
nancy, such as head­aches, transient ischemic attack, and stroke.
Pregnancy-asso­ci­ated TTP accounts for approx­i­ma­tely 5% to
10% of all­TTP cases. What is unusual is that there appears to be
an increased chance of con­gen­i­tal TTP in preg­nancy pre­sen­ta­
tions rather than immune anti­body-medi­ated TTP, which would
be more typ­i­cal.17,18 Presentation of immune TTP (iTTP) and con-
genital TTP (cTTP) is 95% and 5%, respec­tively, but in preg­nancy,
cTTP has been reported at an increased rate, seen in 24% to 66%
of cases. Abnormal liver func­tion tests, as in HELLP, are not nor-
mally asso­ci­ated with pregnancy associated-TTP. TTP can pres­
ent at any stage of preg­nancy but is more fre­quently diag­nosed
Renal Neurologic ADAMT-13
HBP impair­ment fea­tures activ­ity, IU/dL
+++ +/− ++ >20
+ +/− +/− >20
+ + +++ <10
++ +++ + >20
Table 2. Investigations in preg­nancy-asso­ci­ated TMA

Pregnancy-related TMAs (eg, PE,

Test TTP CM-HUS HELLP, AFLP)
Hemoglobin Reduced Reduced Reduced
Platelet count Very reduced Reduced Reduced/very reduced
Reticulocytes Increased Increased Normal/increased
Fragmentation on blood film Yes +++ Yes ++ Yes +
Serum cre­at­i­nine Normal/increased Increased Normal/increased
Urine pro­tein/cre­at­i­nine ratio Normal/increased Increased Increased
Coagulation screen Normal Normal Normal (except AFLP; coag­u­la­tion screen
prolonged—decreased fibrin­o­gen)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/545/1851434/545scully.pdf by guest on 13 December 2021

Bilirubin Increased Increased Mild increase
ALT/AST Normal Normal Increased
Alkaline phos­pha­tase Normal Normal Increased in HELLP and AFLP
Urate Normal Normal Increased
C3/C4 Normal Reduced C3 in some cases Normal
ADAMTS-13 Severely reduced Mildly reduced Mildly reduced
Note: the degree of involvement is demonstrated by an increasing number of +.

in the third tri­mes­ter and post­par­tum period. Undiagnosed TTP,

espe­cially cTTP, is related to sec­ond-tri­mes­ter IUFD.
CM-HUS presenting in preg­nancy
There are nor­ mal phys­ i­
o­logic changes in both the immune
and com­ple­ment path­ways dur­ing preg­nancy with the aim of
protecting the fetus. The increase in com­ple­ment acti­va­tion is
asso­ci­ated with a com­pa­ra­ble esca­la­tion in con­trol­ling com­
ple­ment pro­teins such as CD46 to pre­vent excess com­ple­ment
acti­va­tion.19
Presentation of CM-HUS was thought to be only in the post­
par­tum period. Indeed, 80% of cases are diag­nosed dur­ing this
period, but like TTP, CM-HUS pre­sen­ta­tion can, rarely, occur from
the first tri­mes­ter. The detec­tion of com­ple­ment-asso­ci­ated muta­
tion is com­pa­ra­ble to non­preg­nant CM-HUS cases, and 30% to
50% have no detect­able abnor­mal­ity. Some patients with HELLP
syn­drome show fea­ tures, on serum sam­ ples, of com­ ple­
ment
acti­va­tion,20 and increased pla­cen­tal com­ple­ment acti­va­tion
asso­ci­ated with com­ple­ment muta­tions has been iden­ti­fied in
pre­eclamp­sia.21,22
Vascular endo­the­lial fac­tor pro­tects against aber­rant com­
ple­ment acti­va­tion and is inhibited by sol­u­ble fms-like tyro­sine
kinase 1 (sFLT1) in other organs. There is a cor­re­late with sFLT1
expres­sion, suggesting a role for com­ple­ment in PE.23 A role for
mea­sur­ing sFLT1 and pla­cen­tal growth fac­tor to aid dif­fer­en­ti­a­ TTP or CM-HUS is con­sid­ered (Figure 1). The inves­ti­ga­tions in Table
tion of hyper­ten­sive dis­or­ders in preg­nancy by mea­sur­ing the
ratio of sFLT1/placental growth factor has been suggested.
Before the use of com­ple­ment inhib­i­tor ther­apy, there was
mater­nal and fetal mor­tal­ity with, nearly two-thirds of women
requir­ing acute renal replace­ment ther­apy, half of all­cases devel­
op­ing end-stage renal dis­ease, and an CM-HUS relapse rate of
30%. There was also a risk of spon­ta­ne­ous abor­tion and still­birth
that in the past has not been fully appre­ci­ated.22
Unlike TTP, CM-HUS in preg­nancy may not be asso­ci­ated with
throm­bo­cy­to­pe­nia (ie, <100  ×  109/L), with AKI being the over­rid­ ini­ti­a­tion of com­ple­ment inhib­i­tor ther­apy can be con­sid­ered.
ing fea­ture.
Prakash Singh Shekhawat
CLINICAL CASE (Con­tin­ued)
There were fea­tures sug­ges­tive of hyper­ten­sion/PE and poten­
tially HELLP. Indeed, the plan for this woman had been emer­gency
cae­sar­ean sec­tion with plate­let trans­fu­sion cover. However, the
diag­no­sis of TTP was raised. When the patient was trans­ferred, she
received imme­di­ate plasma exchange and the baby was deliv­ered
at 28+ weeks’ ges­ta­tion by cae­sar­ean sec­tion under gen­eral anes­
thetic, with no plate­let trans­fu­sion required. Blood loss related to
sur­gery was 500 mL. High-dose ste­roid ther­apy was given before
and after the cae­sar­ean sec­tion. The baby weighed 0.836  kg and
was trans­ferred to the neo­na­tal unit. Blood pres­sure was con­
trolled with labetalol, and plasma exchange con­tin­ued daily until
plate­let nor­mal­i­za­tion, requir­ing 6 in total. ADAMTS-13 activ­ity
level at the refer­ring hos­pi­tal was 14  IU/dL and less than 5  IU/dL
preexchange. There was no immu­no­glob­u­lin G (IgG) anti­body to
ADAMTS-13, and sub­se­quent muta­tional anal­y­sis for ADAMTS-13
con­firmed a homo­zy­gous mis­sense muta­tion R1060W.24 Diagnosis
in the clin­i­cal case was con­gen­i­tal TTP, presenting in preg­nancy.
Treatment of preg­nancy-asso­ci­ated TMA
In many cases, the pre­cise diag­no­sis may not be imme­di­ately clear,
so plasma exchange should be started as soon as a diag­no­sis of
1 should be under­taken, with ADAMTS-13 activ­ity lev­els dif­fer­en­ti­at­
ing CM-HUS and TTP (Figure 1). It may not be pos­si­ble to con­firm
if TTP is con­gen­i­tal or immune medi­ated; there­fore, ini­tial immu­no­
sup­pres­sion with ste­roids may be nec­es­sary. In the cur­rent era, the
nanobody caplacizumab is ini­ti­ated on the diag­no­sis of acute TTP.25
If the diag­ no­sis of TTP is fol­ low­ing deliv­
ery/post­par­
tum, this is
accept­ able, but it is not licensed in preg­
n ancy, and given the small
molec­u­lar size, it should be avoided if the preg­nancy is ongo­ing.
It may be obvi­ous the diag­no­sis is CM-HUS, in which case
Eculizumab is a human­ized IgG4κ chi­me­ric anticomplement C5
Pregnancy TMAs  |  547

Delivery
PIH
Control blood
pressure and
PE monitor cTg
HELLP
for pregnancy-
related TMA
and
deterioraon
HUS TTP
AKI Cardiac
Neurologic +/-
renal symptoms
PEX
Figure 1. Summary of treatment of acute TMA in pregnancy. Highlighted are the conditions that may present as TMA in pregnancy.
Pregnancy-related TMAs generally require delivery. TTP and CM-HUS may present features of pregnancy-related TMAs but require a
different pathway in part dictated by the ADAMTS-13 activity level. TTP is diagnosed when ADAMTS-13 activity is less than 10 IU/dL but
may be up to 20 IU/dL before TTP can be excluded. Repeat ADAMTS-13 activity samples and ADAMTS-13 antibody should be checked.
anti­body, pre­vents ongo­ing acti­va­tion of the mem­brane attack
com­plex, but has no effect on the com­ple­ment sys­tem up to C3.
Eculizumab has been used in preg­nancy, pri­mar­ily in parox-
ysmal nocturnal haemoglobinuria,26,27 but there are data now of
its suc­cess­ful use in CM-HUS.28 The global atypical HUS Registry
has con­firmed the ben­e­fit of eculizumab in women with preg­
nancy-asso­ci­ated HUS, and it has sig­nif­i­cantly improved renal
out­comes.29
Meticulous con­trol of blood pres­sure and sup­port­ive ther­
apy should be insti­gated, such as renal replacement therapy in
CM-HUS.
Unlike preg­nancy-related TMAs, for which deliv­ery is asso­
ci­ated with nor­mal­i­za­tion of the con­di­tion within 48 hours, in
preg­ nancy-asso­ ci­
ated TMAs (TTP and CM-HUS), achiev­ ing a
remis­sion is pos­si­ble, which may be impor­tant for women pre-
senting ear­lier in preg­nancy and deliv­ery would be asso­ci­ated
with fetal com­pro­mise. This deci­sion needs to be under­taken in
con­junc­tion with obste­tri­cians and neo­na­tol­o­gists.
What are the risks and treat­ment options in sub­se­quent
preg­nan­cies?
Subsequent preg­nan­cies in patients with both TTP and CM-HUS
can be supported, assum­ing women and their part­ners are fully
aware of the risks to them and the fetus (Figure 2). Management
in sub­se­quent preg­nan­cies requires a mul­ti­dis­ci­plin­ary approach,
with reg­u­lar mon­i­tor­ing of mother and baby dur­ing preg­nancy
and into the post­par­tum period. Empirically, women are started
on low-dose aspi­rin (e.g., 75/100    mg/d)30 and thromboprophy-
548  |  Hematology 2021  |  ASH Education Program
ADAMTS
-13
acvity
level
=CM-HUS =TTP
ADAMTS-13
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/545/1851434/545scully.pdf by guest on 13 December 2021
Eculizumab replacement/
immunosuppression
ADAMTS-13 anbody
Complement mutaon
analysis if negave
analysis/CFH anbody
ADAMTS-13 mutaon
analysis
laxis. This is based on evi­dence of pla­cen­tal dys­func­tion his­to­log­
i­cally and impaired uter­ine artery Dopp­ler flow dur­ing preg­nancy
from affected cases (Figure 2).
The risk of relapse of CM-HUS in sub­se­quent preg­nan­cies is
approx­i­ma­tely 25%. The risk appears to be lower in patients for
whom no com­ple­ment muta­tion has been iden­ti­fied.31 Even with
a com­ple­ment abnor­mal­ity, the risk remains var­i­able. The chance
of sub­se­quent acute CM-HUS in preg­nancy is greatest in those
women with com­ple­ment fac­tor H or I muta­tions. The use of the
com­ple­ment inhib­i­tor, eculizumab, has sig­nif­i­cantly improved
the out­come for CM-HUS. Currently, ther­apy may be ini­ti­ated
dur­ing preg­nancy if there is a sig­nal to sug­gest exac­er­ba­tion of
CM-HUS. There are no data on starting treat­ment for high-risk
cases once preg­nancy is con­firmed, but this should be con­sid­
ered in indi­vid­ual cases.31
In women in whom iTTP was con­firmed, mon­i­tor­ing of ADAMTS-
13 activ­ity lev­els as well as rou­tine lab­o­ra­tory param­e­ters should
be under­taken, iden­ti­fy­ing the base­line level at the begin­ning
of preg­nancy. Testing of ADAMTS-13 activ­ity is advised in each
tri­mes­ter and in the post­par­tum period. If lev­els decrease, there
are sev­eral options, includ­ing ste­roids, aza­thi­op ­ rine, or plasma
exchange, if ADAMTS-13 activ­ity lev­els drop below 10   IU/dL. Rit-
uximab has been used in preg­nancy, but it is pref­er­ab ­ le to wait
until the post­par­tum period.4
In con­gen­i­tal TTP, ADAMTS-13 replace­ment should begin
once preg­nancy is con­firmed, ini­tially every 2 weeks (approx­i­
ma­tely 10 mL/kg). This may need to be increased to weekly from
the sec­ond tri­mes­ter, targeting nor­mal plate­let count and LDH
and an absence of symp­toms. Further evi­dence for the role of
 iTTP

ADAMTS-13 prepregnancy to ensure in normal range

Iniate LDA or confirmaon of pregnancy

Iniate thromboprophylaxis if confirmed thrombophilia

or previous placental dysfuncon

ADAMTS-13 measurements at least each trimester/postpartum Regular fetal scans/measurement of uterine artery
flow by Doppler

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/545/1851434/545scully.pdf by guest on 13 December 2021

Severely
Abnormal:
Closely
reduced: Risks: TTP Normal: Abnormal:
Normal: Iniaon of Risks: IUGR,
monitor +/- relapse, Connue Ensure no
Delivery PEX complicaon IUFD,
iniaon of down clinical/ preterm
as per frequency s of therapy
steroids/
depending
obstetric ADAMTS- delivery, fetal
obstetric azathioprine eg, PEX,
on roune immunosupp
pathway to 13 TTP prematurity
plan if earlier in
lab ression delivery relapse
pregnancy
parameters

Will need regular laboratory parameters, FBC, LDH and renal funcon

Congenital

Confirmaon of pregnancy and iniate plasma

infusion 10 mL/kg every 2 weeks

Start LDA and thromboprophylaxis

Regular fetal
scans/measurement of uterine
Increase plasma infusion to 10 mL/kg artery flow by Doppler
weekly from second trimester or if
platelets decrease/LDH increase
outside of the normal lab range

Plan delivery 37/38 weeks’ gestaon

Connue regular plasma infusions

for at least 8 weeks postpartum

Will need regular laboratory parameters, FBC, LDH and renal funcon

Figure 2. TTP in subsequent pregnancies, immune mediated and congenital pathways. Proceeding with subsequent pregnancy
requires a discussion of the maternal and fetal risks. Close management throughout pregnancy is required for patients with both
immune-mediated and congenital TTP. Regular fetal imaging ultrasound and uterine artery Doppler measurements would be advised.
IUGR, in utero growth retardation; PEX, plasma exchange.
Prakash Singh Shekhawat
Pregnancy TMAs  |  549

Figure 3A- From the 1st pregnancy
Figure 3. Placental histology from the clinical case. (A) Distal villous hypoplasia and infarction at 28 weeks’ gestation in untreated
congenital TTP. (B) Normal villi in the subsequent delivery from the same mother after treatment throughout pregnancy. ADAMTS-13
replacement was via plasma infusion.
plasma infu­sion and improved out­comes for both mother and
fetus has been presented.32
In all 3 of these sub­groups of TMA, deliv­ ery needs to be
planned, aiming for 37 to 38 weeks’ ges­ta­tion.
CLINICAL CASE (Con­t in­u ed)
The patient did have a fur­ther preg­nancy. She had ini­ti­ated a
pro­gram of reg­u­lar plasma infu­sion because of symp­tom­atol­
ogy and con­tin­ued with fort­nightly plasma infu­sion until the
end of her sec­ond tri­mes­ter. She received weekly plasma infu­
sion until 8 weeks post­par­tum. Her sec­ond child was deliv­ered
by elec­tive cae­sar­ean sec­tion at 38 weeks’ ges­ta­tion, with no
com­pli­ca­tions, and the baby weighed 2.75  kg. A com­par­i­son
of his­tol­ogy from her first preg­nancy and sec­ond preg­nancy is
pro­vided in Figure 3.
In con­ clu­
sion, TTP and CM-HUS are acute life-threat­ en­
ing
con­di­tions that require diag­nos­tic con­sid­er­ation and imme­di­ate
ther­apy. Presentation within preg­nancy com­pli­cates the diag­
no­sis fur­ther. Pregnancy has its own TMAs (eg, PE and HELLP).
Features of these preg­nancy-related TMAs may be part of the
TTP/CM-HUS pre­sen­ta­tion. Aside from the mater­nal mor­tal­ity,
there are sig­nif­i­cant risks for the fetus, includ­ing in utero growth
retar­da­tion and IUFD, normally evi­dent in the sec­ond tri­mes­ter.
Pathways have been presented to aid remis­sion dur­ing preg­
nancy but also to ensure improved out­ comes in sub­ se­
quent
preg­nan­cies. This includes close obser­va­tion, with reg­u­lar fetal
scans to ensure no impact on pla­cen­tal func­tion.
Conflict-of-inter­est dis­clo­sure
Marie Scully: speak­
ers fees and advi­ sory boards for Sanofi,
Takeda, Octapharma, Novartis, Alexion.
550  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/545/1851434/545scully.pdf by guest on 13 December 2021
Figure 3B-From the 2nd pregnancy
Off-label drug use
Marie Scully: rituximab for TTP is an off-label use of therapy.
Correspondence
Marie Scully, Department of Haematology, University College
London Hospitals NHS Foundation Trust and Cardiometabolic
Programme-NIHR UCLH/UC BRC, 250 Euston Rd, London NW1
2PG, UK; e-mail: m​­.scully@ucl​­.ac​­.uk.
References
1. Scully M, Cataland S, Coppo P, et al; International Working Group for
Thrombotic Thrombocytopenic Purpura. Consensus on the stan­dard­i­
za­tion of ter­mi­nol­ogy in throm­botic throm­bo­cy­to­pe­nic pur­pura and
related throm­botic microangiopathies. J Thromb Haemost. 2017;15(2):
312-322.
2. Nelson DB, Byrne JJ, Cunningham FG. Acute fatty liver of preg­nancy. Clin
Obstet Gynecol. 2020;63(1):152-164.
3. Sibai BM. Etiology and man­a ge­m ent of post­p ar­tum hyper­ten­s ion-
pre­eclamp­sia. Am J Obstet Gynecol. 2012;206(6):470-475.
4. Neave L, Scully M. Microangiopathic hemo­ lytic ane­
mia in preg­ nancy.
Transfus Med Rev. 2018;32(4):230-236.
5. Scully M, Goodship T. How I treat throm­botic throm­bo­cy­to­pe­nic pur­pura and
atyp­i­cal haemolytic uraemic syn­drome. Br J Haematol. 2014;164(6):759-766.
6. Thomas MR, Robinson S, Scully MA. How we man­age throm­botic microan-
giopathies in preg­nancy. Br J Haematol. 2016;173(6):821-830.
7. O’Gorman N, Wright D, Poon LC, et  al. Multicenter screen­ ing for pre-
eclamp­sia by mater­nal fac­tors and bio­mark­ers at 11-13 weeks’ ges­ta­tion:
com­par­i­son with NICE guide­lines and ACOG rec­om­men­da­tions. Ultra-
sound Obstet Gynecol. 2017;49(6):756-760.
8. Umesawa M, Kobashi G. Epidemiology of hyper­ten­sive dis­or­ders in preg­
nancy: prev­a­lence, risk fac­tors, pre­dic­tors and prog­no­sis. Hypertens Res.
2017;40(3):213-220.
9. Scully M. Thrombotic throm­bo­cy­to­pe­nic pur­pura and atyp­i­cal hemo­lytic
ure­mic syn­drome microangiopathy in preg­nancy. Semin Thromb Hemost.
2016;42(7):774-779.
10. Gutiérrez Junquera C, Balmaseda E, Gil E, et al. Acute fatty liver of preg­
nancy and neo­na­tal long-chain 3-hydroxyacyl-coen­zyme A dehy­dro­ge­nase
(LCHAD) defi­ciency. Eur J Pediatr. 2009;168(1):103-106.
11. Gernsheimer T, James AH, Stasi R. How I treat throm­bo­cy­to­pe­nia in preg­
nancy. Blood. 2013;121(1):38-47.
12. Eslick R, McLintock C. Managing ITP and throm­bo­cy­to­pe­nia in preg­nancy.
Platelets. 2020;31(3):300-306.
13. Kidney Disease: Improving Global Outcomes (KDIGO) Acute Kidney Injury
Work Group. KDIGO clin­i­cal prac­tice guide­line for acute kid­ney injury. Kid-
ney Int. 2012;2(suppl):1-138.
14. Mannucci PM, Canciani MT, Forza I, Lussana F, Lattuada A, Rossi E. Changes
in health and dis­ease of the metalloprotease that cleaves von Willebrand
fac­tor. Blood. 2001;98(9):2730-2735.
15. Sánchez-Luceros A, Farías CE, Amaral MM, et al. von Willebrand fac­tor-cleav­ing
pro­te­ase (ADAMTS13) activ­ity in nor­mal non-preg­nant women, preg­nant and
post-deliv­ery women. Thromb Haemost. 2004;92(6):1320-1326.
16. Jaiswar SP, Gupta A, Sachan R, Natu SN, Shaili M. Lactic dehy­dro­ge­nase: a
bio­chem­i­cal marker for pre­eclamp­sia-eclamp­sia. J Obstet Gynaecol India.
2011;61(6):645-648.
17. Scully M, Thomas M, Underwood M, et al; col­lab­o­ra­tors of the UK TTP Reg-
istry. Thrombotic throm­bo­cy­to­pe­nic pur­pura and preg­nancy: pre­sen­ta­tion,
man­age­ment, and sub­se­quent preg­nancy out­comes. Blood. 2014;124(2):
211-219.
18. Moatti-Cohen M, Garrec C, Wolf M, et al; French Reference Center for Throm-
botic Microangiopathies. Unexpected fre­quency of Upshaw-Schulman syn­
drome in preg­nancy-onset throm­botic throm­bo­cy­to­pe­nic pur­pura. Blood.
2012;119(24):5888-5897.
19. Regal JF, Gilbert JS, Burwick RM. The com­ple­ment sys­tem and adverse preg­
nancy outcomes. Mol Immunol. 2015;67(1):56-70.
20. Vaught AJ, Gavriilaki E, Hueppchen N, et  al. Direct evi­dence of com­ple­
ment acti­va­tion in HELLP syn­drome: a link to atyp­i­cal hemo­lytic ure­mic
syn­drome. Exp Hematol. 2016;44(5):390-398.
21. Salmon JE, Heuser C, Triebwasser M, et al. Mutations in com­ple­ment reg­
u­la­tory pro­teins pre­dis­pose to pre­eclamp­sia: a genetic anal­y­sis of the
PROMISSE cohort. PLoS Med. 2011;8(3):e1001013.
22. Gaggl M, Aigner C, Csuka D, et al. Maternal and fetal out­comes of preg­
nan­cies in women with atyp­i­cal hemo­lytic ure­mic syn­drome. J Am Soc
Nephrol. 2018;29(3):1020-1029.
23. Yonekura Collier A-R, Zsengeller Z, Pernicone E, Salahuddin S, Khankin E-V,
Karumanchi S-A. Placental sFLT1 is asso­ci­ated with com­ple­ment acti­va­tion
Prakash Singh Shekhawat
and syncytiotrophoblast dam­age in pre­eclamp­sia. Hypertens Pregnancy.
2019;38(3):193-199.
24. Alwan F, Vendramin C, Liesner R, et al. Characterization and treat­ment of
con­gen­i­tal throm­botic throm­bo­cy­to­pe­nic pur­pura. Blood. 2019;133(15):
­1644-1651.
25. Scully M, Cataland SR, Peyvandi F, et al; HERCULES Investigators. Capla-
cizumab treat­ment for acquired throm­botic throm­bo­cy­to­pe­nic pur­pura.
N Engl J Med. 2019;380(4):335-346.
26. Miyasaka N, Miura O, Kawaguchi T, et al. Pregnancy out­comes of patients
with par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria treated with eculizumab: a
Jap­a­nese expe­ri­ence and updated review. Int J Hematol. 2016;103(6):703-
712.
27. Kelly RJ, Höchsmann B, Szer J, et al. Eculizumab in preg­nant patients with
par­ox­ys­mal noc­tur­nal hemo­glo­bin­uria. N Engl J Med. 2015;373(11):1032-
1039.
28. Servais A, Devillard N, Frémeaux-Bacchi V, et  al. Atypical haemolytic
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/545/1851434/545scully.pdf by guest on 13 December 2021
uraemic syn­drome and preg­nancy: out­come with ongo­ing eculizumab.
Nephrol Dial Transplant. 2016;31(12):2122-2130.
29. Fakhouri F, Scully M, Ardissino G, Al-Dakkak I, Miller B, Rondeau E. Pregnancy-
trig­gered atyp­i­cal hemo­lytic ure­mic syn­drome (aHUS): a Global aHUS Regis-
try anal­y­sis [published online 7 April 2021]. J Nephrol. 2021.
30. Bujold E, Roberge S, Nicolaides KH. Low-dose aspi­rin for pre­ven­tion of
adverse out­comes related to abnor­mal pla­cen­ta­tion. Prenat Diagn. 2014;
­34(7):642-648.
31. Fakhouri F, Scully M, Provôt F, et al. Management of throm­botic microangi-
opathy in preg­nancy and post­par­tum: report from an inter­na­tional work­ing
group. Blood. 2020;136(19):2103-2117.
32. Sakai K, Fujimura Y, Nagata Y, et al. Success and lim­i­ta­tions of plasma treat­
ment in preg­nant women with con­gen­i­tal throm­botic throm­bo­cy­to­pe­nic
pur­pura. J Thromb Haemost. 2020;18(11):2929-2941.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000290
Pregnancy TMAs  |  551
 PREGNANCY IN SPECIAL POPULATIONS: CHALLENGES AND SOLUTIONS

Pregnancy in special populations: challenges

and solutions practical aspects of managing von
Willebrand disease in pregnancy

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/552/1851794/552turan.pdf by guest on 13 December 2021

Ozlem Turan1,2 and Rezan Abdul Kadir1,2
Katharine Dormandy Haemophilia and Thrombosis Unit and Department of Obstetrics and Gynecology, Royal Free Hospital NHS Trust, London, UK;
1

and 2EGA Institute for Women’s Health, University College London, London, UK

Pregnancy and childbirth pose an important hemostatic challenge for women with von Willebrand disease (VWD) and
can be associated with an increased risk of maternal and neonatal bleeding complications. VWD is a genetically
and clinically heterogeneous bleeding disorder caused by a deficiency or an abnormality in the function of von Wille-
brand factor. Understanding inheritance pattern, hemostatic response to pregnancy, and response to treatment is essen-
tial for provision of individualized obstetric care and optimal outcome. A multidisciplinary approach to management with
a close liaison between the obstetric team and the hemophilia treatment center is required for continuity of care from
preconception counseling through to antenatal, peripartum, and postpartum care. Delivery plan must be coordinated
by the multidisciplinary team and include decisions on place and mode of delivery, implementation of safe analgesia/
anesthesia, and peripartum hemostasis. In this clinical case-based review, we aim to deliver evidence-based practical
guidance for challenges encountered during pregnancy and management of childbirth and puerperium.

LEARNING OBJECTIVES
• Evaluate the clinical challenges in the management of pregnancy and childbirth in VWD
• Recognize heterogeneity in the maternal and fetal risks in different types of VWD
• Discuss multidisciplinary management and best practice for optimal maternal and neonatal outcome

diagnosis, perform genetic testing when applicable, and assess

CLINICAL CASE
A 20­year­old woman with severe type 2M von Willebrand
disease (VWD) is pregnant at 10 weeks’ gestation. This
is her first pregnancy, and she is seen in the multidisci­
plinary clinic for women with bleeding disorders at the
hemophilia treatment center (HTC) to discuss and plan
management of pregnancy and delivery.
Preconceptional care and prenatal diagnosis
Understanding different types of inherited bleeding disorders
(IBDs), their inheritance pattern, and maternal and fetal/neo­
natal bleeding risks is important for appropriate counseling
in relation to prenatal diagnosis (PND) and planning manage­
ment of pregnancy and childbirth. Ideally, women with IBDs
should be seen preconceptionally, to review their historic
their bleeding risk and response to treatment.
The inheritance of most cases of type 1, 2A, 2B, and
2M VWD is autosomal dominant, with a 50% chance of an
affected fetus in each pregnancy. Types 2N and 3 are inher­
ited in an autosomal recessive manner,1 and there is a 25%
risk of having an affected child in each pregnancy when
both parents are carriers of the genetic mutation. Genetic
testing and PND are not routinely performed in VWD but
are considered in families with type 3 VWD.2 PND in late
pregnancy (usually at 34 weeks’ gestation) is also consid­
ered in women with severe types of VWD if the mutation is
known to assist management of delivery3,4; for example, if
the baby is affected, delivery can be planned in a tertiary
center, or delivery can be managed without any restric­
tions and possibly in the local maternity unit if the baby is
unaffected.

552 | Hematology 2021 | ASH Education Program


CLINICAL CASE (Con­t in­u ed)
The patient was seen for pre­con­cep­tion coun­sel­ing in the pre­vi­
ous year. Her diag­no­sis was con­firmed to be type 2M. Her base­
line fac­tor VIII (FVIII) level was 0.58 IU/mL, von Willebrand fac­tor
(VWF):antigen (Ag) was 0.28 IU/mL, VWF:Ristocetin cofactor (RCo)
was <0.4 IU/mL, and VWF:collagen binding was 0.18 IU/mL, with
nor­mal multimeric anal­y­sis. Genetic anal­y­sis had con­firmed dele­
tion on allele 4222_4224delAAG. She was coun­seled by a mul­ti­dis­
ci­plin­ary obstet­ric/hema­tol­ogy team about inher­i­tance, bleed­ing
risks for her and her baby, and man­age­ment of preg­nancy.
Changes in coag­u­la­tion fac­tors in preg­nancy
There are sev­eral phys­i­o­log­i­cal adap­ta­tions from early stages of
preg­nancy in prep­a­ra­tion for labor and deliv­ery. Hypercoagula­
bility is one of the phys­i­o­logic changes that involves a pro­gres­
sive increase in lev­els of VWF and FVIII through­out preg­nancy,
with a peak at the time of deliv­ery.5 This preg­nancy-induced
rise in VWF and FVIII is also seen in most preg­nant women with
VWD but not to the same extent that is seen in healthy preg­nant
con­trols.6 In addi­tion, due to the het­ero­ge­ne­ity of geno­typic
and phe­no­typic fea­tures of VWD, each type exhib­its a dif­fer­ent
response to preg­nancy.
Women with type 1 VWD with a prepregnancy base­line VWF
and FVIII level of >0.3 IU/mL are likely to achieve fac­tor lev­els
within the nor­mal range by the end of preg­nancy.6 In type 2 VWD,
WVF and FVIII lev­els increase sig­nif­i­cantly, but lack of high molec­
ular weight multimers does not change and the VWF:RCo remains
reduced.7 Pregnancy may also exac­er­bate preexisting throm­bo­
cy­to­pe­nia in type 2B VWD.8 Both VWF and FVIII lev­els go up in
type 2N VWD, but FVIII is usu­ally lower in com­par­i­son to VWF in
most cases.9 Women with type 3 VWD typ­i­cally do not show any
increase in FVIII and VWF dur­ing preg­nancy10 (Figure 1).
CLINICAL CASE (Con­t in­u ed)
In our patient, changes in her fac­tor lev­els were con­gru­ent with
changes described in type 2 VWD; VWF:Ag and FVIII lev­els
increased, whereas VWF:RCo remained unchanged.
VWF:RCo
FVIII
VWF:Ag
Baseline First Second Third 8 weeks
trimester trimester trimester postpartum
First tri­mes­ter: FVIII: 0.57 IU/mL, VWF:RCo: <0.4 IU/mL, and
VWF:Ag: 0.29 IU/mL
Second tri­mes­ter: FVIII: 0.64 IU/mL, VWF:RCo: <0.4 IU/mL,
and VWF:Ag: 0.34 IU/mL
Third tri­mes­ter: FVIII: 0.97 IU/mL, VWF:RCo: <0.4 IU/mL, and
VWF:Ag: 0.50 IU/mL
Prakash Singh Shekhawat
Antenatal man­age­ment
Women with all­types of IBDs require a mul­ti­dis­ci­plin­ary team
(MDT) approach for man­age­ment of preg­nancy and deliv­ery, as
well as a close col­lab­o­ra­tion between the obstet­ric team and
HTC.11,12 Women with VWD do not appear to have an increased
risk of mis­car­riage or antepartum hem­or­rhage7,13-15 and do not
usu­ally require pro­phy­lac­tic treat­ment in preg­nancy; thus, ante­
na­tal care can be pro­vided in their local obstet­ric units in col­
lab­o­ra­tion with the HTC, except those with severe type 2 and
3 VWD, who require care in a ter­tiary cen­ter.16
There are a num­ber of com­mon early preg­nancy com­pli­ca­
tions asso­ci­ated with high risk of hem­or­rhagic sequalae. Mis­
carriage is the most prev­a­lent one that affects up to 15% of all­
preg­nan­cies before 12 weeks’ ges­ta­tion.17 In women with VWD,
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/552/1851794/552turan.pdf by guest on 13 December 2021
the preg­nancy-induced rise in fac­tor lev­els is not achieved in
the first half of preg­nancy, and they remain at an increased risk
of bleed­ing after a mis­car­riage or inva­sive PND pro­ce­dures. In
these cir­cum­stances, it is vital to assess their fac­tor lev­els and
con­sider pro­vi­sion of hemo­static sup­port.16
Antenatal mon­i­tor­ing should include checking VWF:Ag and
FVIII:procoagulant activity at least dur­ing their first visit and the
third tri­mes­ter of preg­nancy (32-34 weeks), unless first-tri­mes­ter
lev­els are already within the nor­mal range.10,15,18 VWF:RCo is the
best pre­dic­tor of bleed­ing risk and should be checked at least
once dur­ing the third tri­mes­ter, espe­cially in women with type 2
VWD. Women with type 2B VWD should also have mon­i­tor­ing of
their plate­let count.
Women with VWD are not at an increased risk of obstet­
ric com­pli­ca­tions.19 A national data­base from the United States,
includ­ing 4067 deliv­er­ies, found no increase in fetal growth restric­
tion, pla­cen­tal abrup­tion, or pre­term birth in women with VWD.19
Therefore, they do not require addi­tional obstet­ric assess­ment or
mon­i­tor­ing.
Iron defi­ ciency and iron-defi­ ciency ane­ mia, sec­ond­ary to
heavy men­strual bleed­ing, is a com­mon diag­no­sis among women
with VWD. There is an increased iron require­ment, espe­cially at
late stages of preg­ nancy. Anemia is asso­ ci­
ated with adverse
fetal and mater­nal out­come, includ­ing risk of pri­mary post­par­
tum hem­or­rhage (PPH).20-22 In a cohort study of 506 women with
VWD, 28% of those who expe­ri­enced PPH were ane­mic ante­na­
tally.23 Therefore, reg­u­lar mon­i­tor­ing and cor­rec­tion of ane­mia
and iron defi­ciency are a very impor­tant part of ante­na­tal care.
Women with IBDs should be reviewed by the MDT team in
the third tri­ mes­ter of preg­ nancy (32-34 weeks) ahead of the
expected date of deliv­ery for a deliv­ery plan. The plan should
include details of place and mode of deliv­ery (MOD), hemo­static
cover for deliv­ery and post­par­tum period, pain relief and anes­
the­sia, and man­age­ment of mother and baby dur­ing labor and
after deliv­ery. The plan is agreed on with the mother and com­
mu­ni­cated to all­involved in her care.
Management of labor and deliv­ery
Women with type 1 VWD who have a VWF:RCo level >0.5 IU/mL
at 32 to 34 weeks’ ges­ta­tion and a mild bleed­ing phe­no­type
can plan deliv­ery in their local mater­nity unit unless they have
addi­tional bleed­ing risk such as pre­vi­ous PPH.16 However, for
women with a severe IBD, includ­ing type 2 and 3 VWD, or those
car­ry­ing a baby poten­tially at risk of a severe dis­or­der, it is rec­
ommended that they deliver in a mater­nity unit with an affil­i­
ated HTC, where the nec­es­sary exper­tise in man­age­ment and
Management of von Willebrand dis­ease in preg­nancy  |  553

Figure 1. Modifications of FVIII and VWF levels during normal pregnancy and in women with the more frequent types of VWD.10
resources for lab­o­ra­tory test­ing and clot­ting fac­tor replace­ment
are read­ily avail­­able.16,24
The aim of plan­ning the MOD is to achieve the least trauma
to the mother and infant. VWD is usu­ally not an indi­ca­tion for a
cesar­ean sec­tion, and the deci­sion for MOD should be guided by
obstet­ric indi­ca­tions.18,25 Unlike hemo­philia, there is a lack of data
on MOD and risk of cra­nial bleed­ing for fetuses at risk of severe
VWD (types 2 and 3). Labor and deliv­ery are asso­ci­ated with an
increase in VWF and FVIII in neo­na­tes,26 and thus neo­na­tes with
mild type 1 VWD have a neg­li­gi­ble bleed­ing risk.4 Neonates with
type 2 and 3 and severe type 1 VWD still have reduced lev­els of
VWF and FVIII at birth and have a the­o­ret­i­cal risk of increased
bleed­ing.27 Despite this the­o­ret­i­cal risk, bleed­ing risks in neo­na­
tes with VWD, includ­ing type 3 VWD, are uncom­mon, and life-
threat­en­ing intra­cra­nial hem­or­rhage is very rare. Small case series,
includ­ing 6 from Canada and 2 from the United Kingdom, show
that there were no neo­ na­
tal cases of intra­ cra­
nial hem­ or­
rhage
in infants with VWD.27,28 However, cephalohematoma has been
reported.29,30 In a cohort of 113 pedi­at­ric patients (aged 0-16 years)
with VWD, cephalohematoma at birth was reported in 10 of 113
(9%) chil­dren29; 50% of them were born by a ventouse deliv­ery. In
the same study, bleed­ing from inva­sive mon­i­tor­ing using a fetal
scalp elec­trode was reported in 2 cases. In another study, 5 of 7
(71%) chil­dren had a cephalohematoma after being born by a ven­
touse deliv­ery.30 Therefore, pre­cau­tions to reduce hem­or­rhagic
554  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/552/1851794/552turan.pdf by guest on 13 December 2021
risk in the new­borns need to be implemented. These mea­sures
are presented in Table 1, based on risk strat­i­fic
­ a­tion and rec­om­
men­da­tions by the Royal College of Obstetricians and Gynaecolo­
gists guide­lines on man­age­ment of IBDs in preg­nancy.16
Neuraxial anal­ge­sia/anes­the­sia
Neuraxial anal­ge­sia pro­vi­des effec­tive pain con­trol in labor, but
there is a con­cern of a the­o­ret­ic ­ al increased risk of spi­nal canal
hema­toma in women with bleed­ing dis­or­ders due to their
coag­u­la­tion defect. In total, 161 550 epi­du­rals were sited for
obstet­ric anal­ge­sia dur­ing a 2-week national audit period in the
United Kingdom, with a reported inci­dence of 5 cases of spi­nal
canal hema­toma. There were zero cases of spi­nal canal hema­toma
out of a total of 133 525 obstet­ric spi­nal anes­the­sia sited dur­ing
the same audit period.31 On the other hand, in the event of an
oper­a­tive deliv­ery, the alter­na­tive to neuraxial anes­the­sia would
be gen­eral anes­the­sia, which is asso­ci­ated with an 8 times higher
risk of uter­ine atony and PPH, a 50 times higher risk of failed intu­
ba­tion, and low neo­na­tal Apgar scores.32 In women with IBDs, the
data on the risk of spi­nal canal hema­toma are scarce and lim­ited
to case reports and small case series, all­show­ing a safe admin­
is­tra­tion of neuraxial anal­ge­sia/anes­the­sia when the coag­u­la­tion
fac­tors are nor­mal­ized due to phys­i­o­logic changes of preg­nancy
or with appro­pri­ate hemo­static cover. Two of the larg­est series
included 15 and 33 cases of VWD,33,34 with no hemo­static treat­ment
Table 1. Risk strat­i­fi­ca­tion of bleed­ing risk for the fetus/neo­nate in women with VWD16

Risk level Possible or con­firmed fetal diag­no­sis Suggested man­age­ment of deliv­ery

High Type 3 VWD • Discuss MOD, tak­ing mater­nal and fetal fac­tors into con­sid­er­ation, but avoid
midcavity for­ceps, ventouse deliv­ery; FBS, FSE, rota­tional for­ceps, and exter­nal
cephalic ver­sion
Medium Type 2 VWD • Avoid midcavity for­ceps, ventouse, rota­tional for­ceps, and exter­nal cephalic ver­sion
• Judicious use of FBS and FSE to facil­i­tate vag­i­nal deliv­ery
Mild Clinically mod­er­ate or severe type • Consider avoid­ance of ventouse and exter­nal cephalic ver­sion
1 VWD in fam­ily • Judicious use of rota­tional for­ceps, FBS, and FSE
Unlikely to be at risk Clinically mild type 1 VWD in fam­ily •  No spe­cial pre­cau­tions
FBS, fetal blood sampling; FSE, fetal scalp electrode.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/552/1851794/552turan.pdf by guest on 13 December 2021

or only tranexamic acid (TXA) for those with mild VWD but fac­tor
con­cen­trate admin­is­tra­tion for mod­er­ate and severe cases.
The UK guide­line on the man­age­ment of women with IBDs
rec­om­mends avoid­ance of neuraxial block in women with VWD
unless VWF activ­ity is more than 0.5 IU/mL and the hemo­static
defect has been corrected.16 A tar­get VWF activ­ity level of 0.50
to 1.50 IU/mL is recommended to allow neuraxial block, and VWF
activ­ity should be maintained above 0.50 IU/mL while the neurax­
ial cath­e­ter is in place and for at least 6 hours after removal.35 In
cases with a mild risk of bleed­ing (ie, type 1 VWD with corrected
VWF activ­ity >0.5 IU/mL), hemo­static sup­port with TXA is suf­fi­
cient.16 In women with type 2 and 3 VWD, ade­quate hemo­sta­
sis may not be achieved despite fac­tor replace­ment and nor­mal
lab­o­ra­tory VWF activ­ity level; such cases should be indi­vid­u­ally
assessed by an MDT with exper­tise in VWD.
Management of obstet­ric anal­ge­sia and anes­the­sia in women
with VWD requires a prior indi­vid­u­al­ized risk assess­ment and
plan­ning by the MDT. Risks and ben­e­fits of neuraxial block and
its alter­na­tives should be con­sid­ered to help the mother make an
informed deci­sion. Table 2 sum­ma­rizes con­di­tions for the use of
neuraxial block in women with IBDs.
Postpartum hem­or­rhage
PPH is defined as blood loss of 500 mL or more within 24 hours
of child­birth and com­pli­cates on aver­age 3% to 6% of all­deliv­
er­ies.35,37 Lack of con­sis­tency in the def­i­ni­tion used for PPH
in the VWD lit­er­a­ture was rec­og­nized by a mul­ti­dis­ci­plin­ary
panel work­ing on the recent VWD guide­line38; the panel has
pro­posed a PPH def­i­ni­tion as “blood loss of 1000 mL or more
within 24 hours of birth or any blood loss with the poten­tial to
pro­duce hemo­dy­namic insta­bil­ity” to be used in future VWD
research.38 Women with VWD have a greater risk of expe­ri­enc­
ing PPH,39 and a recent sys­tem­atic review reported an inci­
dence of pri­mary PPH of 34% among 811 deliv­er­ies in women
with VWD.40 In a large national cohort study includ­ing 4067
women with VWD,19 the risk of PPH was 6% in women with
VWD com­pared with 4% in those with­out VWD, and women
with VWD were 5 times more likely to receive a blood trans­
fu­sion after child­birth.19
Although the eti­­ol­ogy for PPH is often mul­ti­fac­to­rial, uter­
ine atony is the commonest obstet­ric cause.41 Other com­mon
obstet­ric causes are retained pla­cen­tal tis­sue and gen­i­tal
tract trauma. In women with IBDs, coag­u­la­tion defect can be
the pri­mary or con­trib­ut­ing fac­tor to PPH. Prevention of PPH
should include obstet­ric mea­sures to min­i­mize risk of uter­ine
atony and birth trauma as well as hemo­static mea­sures to
cor­rect their coag­u­la­tion defect. Active man­age­ment of the
third stage of labor, which incor­po­rates pro­phy­lac­tic use of
uterotonic (oxy­to­cin), early clamping, and con­trolled cord
trac­tion to deliver the pla­centa, reduces the risk of severe pri­mary

Table 2. Conditions for the use of regional block in women with IBDs dur­ing labor and deliv­ery36

Multidisciplinary man­age­ment involv­ing hema­tol­o­gists, anes­the­tists, obste­tri­cians, and the mother

Detailed coun­sel­ing on the ben­e­fits and risks of regional block and its alter­na­tives to help the mother make an informed deci­sion
Careful assess­ment of coag­u­la­tion sta­tus, includ­ing assess­ment of clot­ting fac­tor dur­ing the third tri­mes­ter, and bleed­ing phe­no­type, includ­ing per­
sonal and fam­ily bleed­ing his­tory
Availability of ther­a­peu­tic prod­ucts and lab­o­ra­tory facil­i­ties to ensure ade­quate response to treat­ment
Plan of man­age­ment made ante­na­tally dur­ing the third tri­mes­ter, clearly documented, and read­ily avail­­able to pro­fes­sion­als attend­ing the woman in
labor
Normalization of coag­u­la­tion defect by either a preg­nancy-induced rise in coag­u­la­tion fac­tors or the use of appro­pri­ate pro­phy­lac­tic treat­ment prior
to regional block pro­ce­dures
Meticulous tech­ni­cal skills in the admin­is­tra­tion of regional block by an expe­ri­enced anes­the­tist
Where an epi­du­ral cath­e­ter is placed, ade­quate hemo­sta­sis should be maintained prior to cath­e­ter removal, as the risk of bleed­ing is no less than
with inser­tion
Awareness and sur­veil­lance for symp­toms and signs of poten­tial com­pli­ca­tions

Prakash Singh Shekhawat

Management of von Willebrand dis­ease in preg­nancy  |  555
PPH greater than 1000 mL42 and should be implemented for
all­women with IBDs. Additional uterotonic agents such as
pros­ta­glan­din E2 (misoprostol) have been recommended for
pro­phy­lac­tic use in women with mod­er­ate and severe IBDs.25
Obstetric man­age­ment should also include avoid­ance of fac­
tors that increase the risk of PPH, such as prolonged labor,
mater­nal birth tract injury dur­ing vag­i­nal deliv­ery, and assur­
ance of ade­quate sur­gi­cal hemo­sta­sis dur­ing repair or dur­ing
cesar­ean deliv­ery.
Hemostatic man­age­ment includes prior assess­ment of PPH
risk and plan­ning an appro­pri­ate level of hemo­static cover. In
women with VWD, risk assess­ment should be indi­vid­u­al­ized,
tak­ing into con­sid­er­ation the VWF level dur­ing the third tri­
mes­ter, bleed­ing phe­no­type, pre­vi­ous response to treat­ment,
and obstet­ ric risk fac­tors for PPH (eg, mul­ ti­
ple preg­ nancy,
pla­centa previa). Several stud­ies have shown a high rate of
PPH in women with VWD, with a third-tri­mes­ter VWF of <0.5
IU/dL despite pro­phy­lac­tic treat­ment with VWF con­cen­trates
(plasma derived or recom­bi­nant).6,43-45 However, these stud­ies
pro­vide no data on treat­ment regimes, tar­get VWF activ­ity and
fac­tor-level mon­i­tor­ing dur­ing treat­ment, cause of PPH, and the
use of other treat­ments such as uterotonics and TXA.
There is an increased sys­ temic and local fibri­ no­ly­
sis after
child­birth, and TXA inhib­its fibri­no­ly­sis and reduces blood loss
and mor­tal­ity with PPH.46 Prophylactic use of TXA is recom­
mended for all­women with VWD: alone for those with a VWF
level >0.5 IU/mL and in con­junc­tion with desmopressin or VWF
con­cen­trate if VWF:RCo is <0.5 IU/mL.16,35 This can be started at
the onset of established labor and given 6 hourly and con­tin­ued
for sev­eral weeks until lochia is light. The intra­ve­nous route may
be pre­ferred due to slower gas­tric emp­ty­ing and increased risk
of vomiting in labor.
Desmopressin increases cir­cu­la­tory lev­els of VWF and FVIII
due to release of an endo­the­lial store of VWF. It is a valid
treat­ment option for VWD, pri­mar­ily type 1 VWD and selected
type 2 cases. It is effec­tive in patients with base­line VWF and
FVIII lev­els higher than 0.10 IU/mL, but due to var­ia ­ bil­ity in
response, a prior test dose is recommended.10,35 Desmopres­
sin is gen­er­ally contraindicated in type 2B VWD as it can tran­
siently exac­er­bate throm­bo­cy­to­pe­nia.47 It is not used in type
3 VWD because the clin­i­cal response is very poor or absent.9
A sys­tem­atic review of 216 preg­nan­cies in 30 stud­ies dem­
on­strated a safe and effec­tive use of desmopressin for the
pre­ven­tion and treat­ment of bleed­ing in preg­nancy and child­
birth in most cases.48 The most com­mon indi­ca­tion for its use
was as a pro­phy­lac­tic hemo­static agent for the pre­ven­tion of
PPH in 172 cases with no sig­nif­i­cant bleed­ing in 167 deliv­er­ies.
There were no adverse fetal out­comes. Maternal side effects
included well-tol­er­ated facial flash­ing and head­ache and only
1 case of water intox­i­ca­tion sei­zure.48 Thus, fluid intake should
be mon­i­tored and lim­ited to 1 to 1.5 L in 24 hours, and elec­
tro­lytes should be mon­it­ored if addi­tional fluid is required.
Repeated dos­ing of more than 2 to 3 doses 12 to 24 hours
apart should be avoided. Desmopressin should also be
avoided if fluid restric­tion or mon­i­tor­ing is not fea­si­ble and in
women with pre­eclamp­sia due to its vaso­con­stric­tive effect,
exac­er­bat­ing vas­cu­lar resis­tance and hyper­ten­sion.47
Factor replace­ment is required when desmopressin is inef­
fec­tive or contraindicated. Treatment with fac­tor con­cen­trate
is given when the mother is in active labor and as near to
556  |  Hematology 2021  |  ASH Education Program
deliv­ery as pos­si­ble or prior to inser­tion of a neuraxial cath­e­
ter. The peak tar­get VWF activ­ity level is aimed at 1.0 IU/mL,
and the level should be maintained above 0.5 IU/mL until
hemo­sta­sis is secured.16 Monitoring of pre- and posttreatment
lev­els of VWF activ­ity and fac­tor VIII lev­els is recommended
after deliv­ery, if labor is prolonged, or in case of exces­sive
bleed­ing.16 There are a num­ber of plasma-derived prod­ucts
with dif­fer­ent VWF/FVIII ratios or recom­bi­nant VWF, which is
devoid of FVIII. The choice of fac­tor con­cen­trate will depend
on the local avail­abil­ity and FVIII lev­els in the third tri­mes­ter.9
An increased level of FVIII is asso­ci­ated with a risk of venous
throm­bo­em­bo­lism (VTE). Products that limit a patient’s expo­
sure to FVIII allows for more fre­quent dos­ing of VWF if needed,
with­out the risk of VTE sec­ond­ary to accu­mu­la­tion of FVIII.49
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/552/1851794/552turan.pdf by guest on 13 December 2021
Plasma-derived con­cen­trates with a low FVIII are advis­able for
women with a high third-tri­mes­ter FVIII level, such as women
with type 2 VWD. Recombinant VWF con­cen­trates con­tain no
FVIII with a lon­ger half-life, now licensed for use in sev­eral
countries, and they can also be used in these women.9,44 In
women with type 2B VWD, plate­let count should be checked
when in labor and plate­let trans­fu­sion is recommended to
main­tain a plate­let count of >50 × 109.18
Women with VWD are also at risk of sec­ond­ary PPH and pro­
longed and heavy lochia loss dur­ing the puer­pe­rium.50 A preg­
nancy-induced rise in VWF and FVIII returns to prepregnancy
lev­els within a few days after deliv­ery6,51 (Figure 2). For women
who are receiv­ing fac­tor con­cen­trates, it is impor­tant to ensure
VWF and FVIII lev­els are maintained above 0.5 IU/mL for 3 to
5 days after an uncom­pli­cated vag­i­nal deliv­ery and 5 to 7 days
fol­low­ing oper­a­tive deliv­er­ies.52 In women with severe VWD,
espe­cially type 3 VWD, VWF con­cen­trate may be required for
a few weeks after deliv­ery.16 In addi­tion, TXA 1 g 6 hourly dur­ing
the puer­pe­rium should be con­sid­ered in all­women with VWD
until the lochia has stopped. All women should be briefed, with a
safety net put in place before dis­charge regard­ing increased risk
of sec­ond­ary PPH, and they are advised to report any exces­sive
vag­i­nal bleed­ing.
Pregnancy and the post­par­tum period are asso­ci­ated with an
increased risk of VTE. Women with VWD were no less likely to
Figure 2. Postnatal changes in VWF.6
expe­ri­ence a pul­mo­nary embolism after child­birth than women
with­out VWD in a large cohort study.19 Therefore, women should
be care­fully assessed for throm­botic risk post­de­liv­ery against
their risk of bleed­ing, and appro­pri­ate post­na­tal VTE pro­phy­laxis
should be advised. Low-molec­u­lar-weight hep­a­rin can be used
for women with ade­quate cor­rec­tion of VWF and FVIII lev­els, but
mechan­i­cal meth­ods should be employed when fac­tor lev­els are
less than 0.5 IU/mL.16
Management of neo­na­tes
VWF level is raised at birth, mak­ing diag­no­sis of mild VWD dif­
fi­
cult in the neo­ na­tal period and reduc­ ing neo­ na­tal bleed­ ing
risk. However, fac­tor lev­els in neo­na­tes with severe type 1 and
types 2 and 3 remain low, and thus these neo­na­tes are at risk of
bleed­ing com­pli­ca­tions. It is recommended that a cord blood
sam­ple is obtained at the time of deliv­ery for assess­ment of FVIII
and VWF anti­gen and activ­ity in these neo­na­tes. Cranial imag­ing
and short-term pro­phy­laxis with fac­tor con­cen­trate should be
con­sid­ered in neo­na­tes with type 3 VWD, espe­cially if deliv­ery
was poten­tially trau­matic.16 Neonates with reduced VWF are also
at risk of mus­cle bleeds fol­low­ing intra­mus­cu­lar injec­tion, and
vita­min K should be admin­is­tered orally in neo­na­tes with type
2 and 3 VWD. Bleeding post­vac­ci­na­tion has also been reported
in these neo­na­tes, and it was the sec­ond most fre­quent pedi­
at­ric bleed­ing symp­tom in 1 study.30 Therefore, all­ immu­ni­za­
tions should be given sub­cu­ta­ne­ously with the smallest gauge
needle.53 Review at the HTC with the pedi­at­ric team should be
planned and arranged before dis­charge.
CLINICAL CASE (Con­t in­u ed)
The mother opted for a planned cesar­ean sec­tion for fear of
labor and to avoid any risk for her baby. After appro­pri­ate coun­
sel­ing, a planned cesar­ean sec­tion at 39 weeks’ ges­ta­tion was
performed under spi­nal anes­the­sia. As her FVIII was high dur­
ing the third tri­mes­ter, a plasma-derived con­cen­trate with a
low FVIII level was cho­sen for pro­phy­lac­tic treat­ment dur­ing
cesar­ean sec­tion and post­na­tally. First dose was given prior to
inser­tion of a spi­nal cath­e­ter with 1 g intra­ve­nous TXA. Factor
con­cen­trate was given 8 hourly for 24 hours, then 12 hourly on
day 2, and then daily for 5 days. Tranexamic acid 1 g was given
intra­ve­nously in the first 24 hours and then orally for 4 weeks or
until lochia com­pletely stopped.
She deliv­ered a baby boy weighing 2936 g under spi­nal anes­
the­sia with no com­pli­ca­tions. Estimated blood loss at deliv­ery
was 450 mL. Cord blood sam­ple indi­cated that her infant was
affected by VWD. Vitamin K was admin­is­trated orally. Mother and
baby were discharged home on day 3 post­na­tally with a hemo­
philia nurse and a com­mu­nity mid­wife to visit daily for a week.
Conflict-of-inter­est dis­clo­sure
Ozlem Turan: no com­pet­ing finan­cial inter­ests to declare.
Rezan Abdul Kadir: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Ozlem Turan: Misoprostol is currently only approved by the Food and
Drug Administration (FDA) in the US for the prevention and treatment
of gastric ulcers secondary to non-steroidal anti-inflammatory drugs
Prakash Singh Shekhawat
(NSAIDS) use. Misoprostol is also widely used for the prevention and
treatment of postpartum haemorrhage.
Rezan Abdul Kadir: Misoprostol is currently only approved by the
Food and Drug Administration (FDA) in the US for the prevention
and treatment of gastric ulcers secondary to non-steroidal anti-
inflammatory drug (NSAID) use. Misoprostol is also widely used
for the prevention and treatment of postpartum haemorrhage.
Correspondence
Rezan Abdul Kadir, Royal Free Hospital NHS Foundation Trust,
Pond Street, Hampstead, London NW3 2QG, UK; e-mail:rezan​
­.abdul-kadir@nhs​­.net.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/552/1851794/552turan.pdf by guest on 13 December 2021
References
1. James PD, Goodeve AC. von Willebrand dis­ease. Genet Med. 2011;13(5):365-
376.
2. Ludlam CA, Pasi KJ, Bolton-Maggs P, et al; UK Haemophilia Centre Doc­
tors’ Organisation. A frame­work for genetic ser­vice pro­vi­sion for haemo­
philia and other inherited bleed­ing dis­or­ders. Haemophilia. 2005;11(2):
145-163.
3. Cutler J, Chappell LC, Kyle P, Madan B. Third tri­mes­ter amnio­cen­te­sis for
diag­no­sis of inherited bleed­ing dis­or­ders prior to deliv­ery. Haemophilia.
2013;19(6):904-907.
4. Leebeek FWG, Duvekot J, Kruip MJHA. How I man­age preg­nancy in car­ri­ers
of hemo­philia and patients with von Willebrand dis­ease. Blood. 2020;136
­(19):2143-2150.
5. Nowak-Göttl U, Limperger V, Kenet G, et al. Developmental hemo­sta­sis: a
lifespan from neo­na­tes and preg­nancy to the young and elderly adult in a
Euro­pean white pop­u­la­tion. Blood Cells Mol Dis. 2017;67:2-13.
6. James AH, Konkle BA, Kouides P, et al. Postpartum von Willebrand fac­tor
lev­els in women with and with­out von Willebrand dis­ease and impli­ca­tions
for pro­phy­laxis. Haemophilia. 2015;21(1):81-87.
7. Castaman G, Tosetto A, Rodeghiero F. Pregnancy and deliv­ery in women
with von Willebrand’s dis­ease and dif­fer­ent von Willebrand fac­tor muta­tions.
Haematologica. 2010;95(6):963-969.
8. Kruse-Jarres R, Johnsen JM. How I treat type 2B von Willebrand dis­ease.
Blood. 2018;131(12):1292-1300.
9. Castaman G, James PD. Pregnancy and deliv­ery in women with von Wille­
brand dis­ease. Eur J Haematol. 2019;103(2):73-79.
10. Castaman G, Goodeve A, Eikenboom J; Euro­pean Group on von Willebrand
Disease. Principles of care for the diag­no­sis and treat­ment of von Wille­
brand dis­ease. Haematologica. 2013;98(5):667-674.
11. Kadir RA, Davies J. Hemostatic dis­or­ders in women. J Thromb Haemost.
2013;11(suppl 1):170-179.
12. van Galen K, Lavin L, Skouw-Rasmussen N, et  al. Euro­pean prin­ci­ples of
care for women and girls with inherited bleed­ing dis­or­ders. 2021. Accessed
6/18/2021. https:­/­/eahad­.org­/eahad­-wgbd­-poc­-webinar­-on­-9­-march­/
13. Kadir RA, Lee CA, Sabin CA, Pollard D, Economides DL. Pregnancy in
women with von Willebrand’s dis­ease or fac­tor XI defi­ciency. Br J Obstet
Gynaecol. 1998;105(3):314-321.
14. Lak M, Peyvandi F, Mannucci PM. Clinical man­i­fes­ta­tions and com­pli­ca­tions
of child­birth and replace­ment ther­apy in 385 Ira­nian patients with type
3 von Willebrand dis­ease. Br J Haematol. 2000;111(4):1236-1239.
15. Skeith L, Rydz N, O’Beirne M, Goodyear D, Li H, Poon M-C. Pregnancy loss
in women with von Willebrand dis­ease: a sin­gle-cen­ter pilot study. Blood
Coagul Fibrinolysis. 2017;28(5):393-397.
16. Pavord SR, Rayment R, Madan B, et al. Management of inherited bleed­ing
dis­or­ders in preg­nancy. Green-top Guideline No. 71 (joint with UKHCDO).
BJOG. 2017;124(8):e193-e263.
17. Quenby S, Gallos ID, Dhillon-Smith RK, et al. Miscarriage mat­ters: the epi­
de­mi­o­log­i­cal, phys­i­cal, psy­cho­log­i­cal, and eco­nomic costs of early preg­
nancy loss. Lancet. 2021;397(10285):1658-1667.
18. Millar CM. Von Willebrand dis­ease. In: Kadir RA, James PD, Lee C, eds.
Inherited Bleeding Disorders in Women. 2nd ed. John Wiley; 2019:83-94.
19. James AH, Jamison MG. Bleeding events and other com­pli­ca­tions dur­ing
preg­nancy and child­birth in women with von Willebrand dis­ease. J Thromb
Haemost. 2007;5(6):1165-1169.
20. Kavle JA, Stoltzfus RJ, Witter F, Tielsch JM, Khalfan SS, Caulfield LE. Associa­
tion between anae­mia dur­ing preg­nancy and blood loss at and after deliv­ery
Management of von Willebrand dis­ease in preg­nancy  |  557
 among women with vag­i­nal births in Pemba Island, Zanzibar, Tanzania. J
Health Popul Nutr. 2008;26(2):232-240.
21. Mirza FG, Abdul-Kadir R, Breymann C, Fraser IS, Taher A. Impact and man­
age­ment of iron defi­ciency and iron defi­ciency ane­mia in women’s health.
Expert Rev Hematol. 2018;11(9):727-736.
22. Nair M, Choudhury MK, Choudhury SS, et al. Association between mater­
nal anae­mia and preg­nancy out­comes: a cohort study in Assam, India. BMJ
Glob Health. 2016;1(1):e000026.
23. Malec LM, Moore CG, Yabes J, Li J, Ragni MV. Postpartum haemorrhage in
women with von Willebrand dis­ease: an obser­va­tional study of the Penn­
sylvania Health Care Cost Containment Council (PHC4) data­base. Haemo-
philia. 2015;21(5):e442-e445.
24. Johnsen JM. Current approaches to preg­nancy and child­birth in women
with von Willebrand dis­ease. Clin Adv Hematol Oncol. 2018;16(4):254-257.
25. Huq FY, Kadir RA. Management of preg­ nancy, labour and deliv­ ery in
women with inherited bleed­ing dis­or­ders. Haemophilia. 2011;17(suppl 1):
20-30.
26. Kulkarni AA, Osmond M, Bapir M, et al. The effect of labour on the coag­ul­a­
tion sys­tem in the term neo­nate. Haemophilia. 2013;19(4):533-538.
27. Labarque V, Stain AM, Blanchette V, Kahr WH, Carcao MD. Intracranial hae­
morrhage in von Willebrand dis­ease: a report on six cases. Haemophilia.
2013;19(4):602-606.
28. Chalmers E, Alamelu J, Collins P, et al; Paediatric and Rare Disorders
Working Parties of UKHCDO. Intracranial haemorrhage in chil­dren with
inherited bleed­ing dis­or­ders in the UK 2003-2013. J Thromb Haemost.
2018;24(4):641-647.
29. Biss TT, Blanchette VS, Clark DS, et al. Quantitation of bleed­ing symp­toms
in chil­dren with von Willebrand dis­ease: use of a stan­dard­ized pedi­at­ric
bleed­ing ques­tion­naire. J Thromb Haemost. 2010;8(5):950-956.
30. Sanders YV, Fijnvandraat K, Boender J, et al; WiN Study Group. Bleeding
spec­trum in chil­dren with mod­er­ate or severe von Willebrand dis­ease: rel­
e­vance of pedi­at­ric-spe­cific bleed­ing. Am J Hematol. 2015;90(12):1142-1148.
31. Cook TM, Counsell D, Wildsmith JA; Royal College of Anaesthetists Third
National Audit Project. Major com­ pli­
ca­
tions of cen­tral neuraxial block:
report on the Third National Audit Project of the Royal College of Anaes­
thetists. Br J Anaesth. 2009;102(2):179-190.
32. Delgado C, Ring L, Mushambi MC. General anaes­the­sia in obstet­rics. BJA
Educ. 2020;20(6):201-207.
33. Boyd SC, O’Connor AD, Horan MA, et al. Analgesia, anaes­the­sia and obstet­
ric out­come in women with inherited bleed­ing dis­or­ders. Eur J Obstet
Gynecol Reprod Biol. 2019;239:60-63.
34. Chi C, Lee CA, England A, Hingorani J, Paintsil J, Kadir RA. Obstetric anal­ge­
sia and anaes­the­sia in women with inherited bleed­ing dis­or­ders. Thromb
Haemost. 2009;101(6):1104-1111.
35. Connell NT, Flood VH, Brignardello-Petersen R, et al. ASH ISTH NHF WFH
2021 guide­lines on the man­age­ment of von Willebrand dis­ease. Blood Adv.
2021;5(1):301-325.
36. Bouthors AS, England A, Kadir RA. Analgesia and anaes­the­sia for preg­nant
women with inherited bleed­ing dis­or­ders. In: Kadir RA, James PD, Lee C,
eds. Inherited Bleeding Disorders in Women. 2nd ed. John Wiley; 2019:191-
202.
37. Knight M, Callaghan WM, Berg C, et al. Trends in post­par­tum hem­or­rhage
in high resource countries: a review and rec­om­men­da­tions from the Inter­
national Postpartum Hemorrhage Collaborative Group. BMC Pregnancy
Childbirth. 2009;9:55.
558  |  Hematology 2021  |  ASH Education Program
38. Connell NT, James PD, Brignardello-Petersen R, et al. von Willebrand dis­ease:
pro­pos­ing def­i­ni­tions for future research. Blood Adv. 2021;5(2):565-569.
39. James AH. More than men­or­rha­gia: a review of the obstet­ric and gynaeco­
logical man­i­fes­ta­tions of bleed­ing dis­or­ders. Haemophilia. 2005;11(4):295-
307.
40. Punt MC, Waning ML, Mauser-Bunschoten EP, et  al. Maternal and neo­
na­tal bleed­ing com­pli­ca­tions in rela­tion to peripartum man­age­ment in
women with von Willebrand dis­ ease: a sys­ tem­
atic review. Blood Rev.
2020;39:100633.
41. Chandraharan E, Krishna A. Diagnosis and man­age­ment of post­par­tum
haemorrhage. BMJ. 2017;358:j3875.
42. Begley CM, Gyte GM, Devane D, McGuire W, Weeks A, Biesty LM. Active
ver­sus expec­tant man­age­ment for women in the third stage of labour.
Cochrane Database Syst Rev. 2019;2:CD007412.
43. Kadir RA, Chi C. Women and von Willebrand dis­ease: con­tro­ver­sies in
diag­no­sis and man­age­ment. Semin Thromb Hemost. 2006;32(6):605-615.
44. Machin N, Ragni MV. Recombinant vs plasma-derived von Willebrand fac­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/552/1851794/552turan.pdf by guest on 13 December 2021
tor to pre­vent post­par­tum hem­or­rhage in von Willebrand dis­ease. Blood
Adv. 2020;4(14):3234-3238.
45. Stoof SC, van Steenbergen HW, Zwagemaker A, et al. Primary post­par­tum
haemorrhage in women with von Willebrand dis­ease or carriership of hae­
mophilia despite specialised care: a ret­ro­spec­tive sur­vey. Haemophilia.
2015;21(4):505-512.
46. Shakur H, Elbourne D, Gülmezoglu M, et al. The WOMAN Trial (World Mater­
nal Antifibrinolytic Trial): tranexamic acid for the treat­ment of post­par­tum
haemorrhage: an inter­ na­tional randomised, dou­ ble blind pla­cebo con­
trolled trial. Trials. 2010;11:40.
47. Laffan MA, Lester E, O’Donnell JS, et al. The diag­no­sis and man­age­ment of
von Willebrand dis­ease: a United Kingdom Haemophilia Centre Doctors
Organization guide­line approved by the Brit­ish Committee for Standards
in Haematology. Br J Haematol. 2014;167(4):453-465.
48. Trigg DE, Stergiotou I, Peitsidis P, Kadir RA. A sys­tem­atic review: the use of
desmopressin for treat­ment and pro­phy­laxis of bleed­ing dis­or­ders in preg­
nancy. Haemophilia. 2012;18(1):25-33.
49. Coppola A, Franchini M, Makris M, Santagostino E, Di Minno G, Mannucci
PM. Thrombotic adverse events to coag­u­la­tion fac­tor con­cen­trates for
treat­ment of patients with haemophilia and von Willebrand dis­ease: a sys­
tem­atic review of pro­spec­tive stud­ies. Haemophilia. 2012;18(3):e173-e187.
50. Chi C, Bapir M, Lee CA, Kadir RA. Puerperal loss (lochia) in women with or
with­out inherited bleed­ing dis­or­ders. Am J Obstet Gynecol. 2010;203(1):56.
e1-56.e5.
51. Huq FY, Kulkarni A, Agbim EC, Riddell A, Tuddenham E, Kadir RA. Changes
in the lev­els of fac­tor VIII and von Willebrand fac­tor in the puer­pe­rium.
Haemophilia. 2012;18(2):241-245.
52. National Hemophilia Foundation. MASAC Document 265—MASAC Guide-
lines for Pregnancy and Perinatal Management of Women with Inherited
Bleeding Disorders and Carriers of Hemophilia A or B. National Hemophilia
Foundation; 2021.
53. Carcao M, Bouskill V. The new­born. In: Kadir RA, James PD, Lee C, eds.
Inherited Bleeding Disorders in Women. 2nd ed. John Wiley; 2019:205-221.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000321
 PREGNANCY IN SPECIAL POPULATIONS: CHALLENGES AND SOLUTIONS

Prevention and management of venous

thromboembolism in pregnancy: cutting through
the practice variation

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021

Leslie Skeith
Division of Hematology and Hematological Malignancies, Department of Medicine, Foothills Medical Centre, University of Calgary, Calgary, Canada

There is clinical practice variation in the area of prevention and management of venous thromboembolism (VTE) in
pregnancy. There are limited data and differing recommendations across major clinical practice guidelines, especially
relating to the role of postpartum low-molecular-weight heparin (LMWH) for patients with mild inherited thrombophilia
and those with pregnancy-related VTE risk factors. This chapter explores the issues of practice variation and related data
for postpartum VTE prevention. Controversial topics of VTE management in pregnancy are also reviewed and include
LMWH dosing and the role of anti-Xa level monitoring, as well as peripartum anticoagulation management around labor
and delivery.

LEARNING OBJECTIVES
• Describe the practice variation and limited data in the area of postpartum VTE prevention
• Describe management of VTE in pregnancy and an approach to anticoagulation management around labor and
delivery

There is an increased risk of VTE (deep vein thrombosis

CLINICAL CASE 1
You are seeing a 30-year-old G1P0 woman who is 32
weeks pregnant and is known to be heterozygous for the
factor V Leiden gene mutation. She has no personal his-
tory of venous thromboembolism (VTE), but her mother
has a history of an unprovoked VTE. She has no other VTE
risk factors (RFs) or medical history. She is referred to you
for counseling relating to the role of postpartum throm-
boprophylaxis.
CLINICAL CASE 2
A 25-year-old G2P2 woman had an urgent unplanned
cesarean delivery (CD) overnight because of failure to
progress in labor. Her CD was uncomplicated, and she
had an estimated blood loss of 750 mL. Should the patient
receive postpartum thromboprophylaxis?
[DVT] and pulmonary embolism [PE]) during pregnancy
that affects approximately 1.2 per 1000 deliveries, which is
a 5- to 10-fold increased risk compared with the nonpreg-
nant population.1-3 The antepartum period and postpartum
period each carry a similar VTE risk (0.6 per 1000 deliveries).1
VTE is a leading cause of direct maternal mortality and can
have important long-term consequences, including post-
thrombotic syndrome, post-PE syndrome associated with
functional limitations, and a reduced quality of life.4-9 One of
the greatest predictors of postthrombotic syndrome after
a pregnancy-associated VTE is having a postpartum DVT.5
The risk of VTE during pregnancy and the postpartum
period is higher in patients with additional VTE RFs, such
as an inherited thrombophilia or pregnancy-related RFs
(eg, a woman with a combination of prolonged antepartum
immobility and elevated body mass index ≥25kg/m2 has an
adjusted odds ratio of 40 to develop a postpartum VTE).10,11
Because the VTE incidence in the postpartum period is still
relatively low (even in these higher-risk groups) and large
randomized controlled trials have been difficult to conduct,

Prakash Singh Shekhawat

VTE prevention and management in pregnancy | 559
clin­i­cal prac­tice guide­line rec­om­men­da­tions regard­ing thrombo-
prophylaxis are largely based on obser­va­tional data and expert
opin­ion. Information about the VTE risk, bleeding risk, other
downsides of low-molecular-weight heparin, related mortality,
and patient preferences are taken into account when formulating
guideline recommendations.12
LMWH thromboprophylaxis is indi­ cated dur­ ing preg­nancy
for indi­vid­u­als with a prior unpro­voked or hor­mone-asso­ci­ated
VTE, and LMWH thromboprophylaxis is indi­cated in the 6-week
post­par­tum period for indi­vid­u­als with any prior VTE (includ­
ing unpro­voked, hor­mone-asso­ci­ated, or pro­voked VTE).13 In
patients with­out a per­sonal his­tory of VTE, there is more uncer­
tainty about the role of thromboprophylaxis. There is gen­eral
guide­line agree­ment (but not com­plete con­sen­sus) on LMWH
thromboprophylaxis for an indi­vid­ual with a potent thrombo-
philia. There are more differences across clinical practice guide-
lines for the role of LMWH use between suggested LMWH use
and clin­i­cal prac­tice guide­lines for indi­vid­u­als who have a more
com­mon inherited thrombophilia (eg, het­ero­zy­gous fac­tor V
Leiden) or those with preg­nancy-related RFs (eg, pre­eclamp­sia,
urgent CD, post­par­tum hem­or­rhage).13-17
LMWH thromboprophylaxis rec­om­men­da­tions dif­fer across
clin­i­cal prac­tice guide­lines for patients with inherited thrombo-
philia and are outlined in Table 3 of the 2018 Amer­ic ­ an Society
of Hematology (ASH) guide­lines for man­age­ment of VTE in the
con­text of preg­nancy.13 There has been less empha­sis placed on
the role and prac­tice var­i­a­tion of LMWH thromboprophylaxis for
preg­nancy-related RFs, includ­ing after CD. A sum­mary of thrombo-
prophylaxis guide­line state­ments for LMWH use post-CD is listed
in Table 1. In a US data­base that cap­tured over 1.2 mil­lion women
post-CD, the pro­por­tion of patients who would have the­o­ret­i­cally
received LMWH thromboprophylaxis based on VTE RFs was 0.3%,
16.2%, 73.4%, and 0.2%, according to dif­fer­ing rec­om­men­da­tions
by the 2011 Amer­ic ­ an College of Obstetricians and Gynecologists
(ACOG), 2012 Amer­i­can College of Chest Physicians (ACCP), 2015
Royal College of Obstetricians and Gynaecologists (RCOG), and
the 2018 ASH guide­lines, respec­tively22 (Table 2). Similarly, in a
ret­ro­spec­tive review in Geneva, Switzerland, of 344 post­par­tum
women who deliv­ered in 1 month at 1 cen­ter, the the­o­ret­i­cal use
of post­par­tum LMWH thromboprophylaxis after CD included 35%
(ACOG), 40% (ACCP), 89% (RCOG), and 0% (ASH) (Table 2 for all­
deliv­er­ies). For both exam­ples, the ASH guide­lines do not focus
on this spe­cific area of rec­om­men­da­tion.23
Why do clin­i­cal prac­tice guide­lines dif­fer? The some­what
obvi­ous answer is because lower-qual­ity evi­dence is avail­­able to
inform deci­sions, which includes the lim­ited data on the actual
ben­e­fits and risks of thromboprophylaxis in this pop­u­la­tion. In an
updated Cochrane sys­tem­atic review of avail­­able ran­dom­ized tri­
als, the lim­ited evi­dence remains “very uncer­tain about the ben­e­
fits and harms of VTE thromboprophylaxis” dur­ing preg­nancy and
the post­par­tum period.24 More spe­cif­i­cally, there are dif­fer­ences
in how the lim­ited data are interpreted, which can lead to dif­fer­
ent guide­line rec­om­men­da­tions. This includes issues of pub­li­ca­
tion date, what sta­tis­ti­cal strat­e­gies are used to com­bine mul­ti­ple
VTE RFs together, what VTE risk thresh­old is set to rec­om­mend
LMWH, if a VTE scor­ing sys­tem is used, and the role of expert and
patient opin­ion.12,25 When should older clin­i­cal prac­tice guide­
lines be con­sid­ered “retired”? Earlier guide­lines, such as those
from the ACCP and the Aus­tra­lian and New Zealand Journal of
Obstetrics and Gynaecology (ANZJOG), were published almost
560  |  Hematology 2021  |  ASH Education Program
10 years ago, and so data used to inform deci­sions may be out of
date.15,17 Experts do not agree on what VTE risk thresh­old to use,
which is high­lighted in the clin­i­cal prac­tice guide­line meth­od­ol­
ogy described.13,26 Although most abso­lute VTE risk thresh­olds in
guide­lines (if reported at all­) are set some­where between 1% and
3% to con­sider LMWH use, some experts have recommended a
lower VTE thresh­old of 0.2% post-CD based on deci­sion mod­el­
ing.27 Others argue that set­ting a higher VTE thresh­old or num­
ber needed to treat is required to bal­ance a higher bleed­ing case
fatal­ity rate (com­pared with the VTE case fatal­ity rate) or other
com­pli­ca­tions seen, which is pri­mar­ily extrap­o­lated from non-
pregnancy data.12,28,29 Understanding what mat­ters to patients is
still poorly under­stood. Bates et al30 interviewed preg­nant patients
and iden­ti­fied that patients placed sim­i­lar health state val­ues on
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021
recur­rent VTE and obstet­ric bleed­ing, but this also var­ied among
indi­vid­u­als. Furthermore, how patient val­ues and pref­er­ences are
incor­po­rated into the guide­line rec­om­men­da­tions remains var­i­
able and unclear, and fur­ther work is needed in this area.
Even after mak­ing sta­tis­ti­cal assump­tions about the lim­ited
data and decid­ing on a VTE risk thresh­old to rec­om­mend LMWH,
we are still learn­ing how this risk is reflected in real-life prac­tice.
For exam­ple, the 2018 ASH guide­line panel recommended a 1%
VTE risk thresh­old for post­par­tum LMWH use.13 In a large US data­
base study of over 1.2 mil­lion cesar­ean deliv­er­ies, patients who
were iden­ti­fied as “ele­vated risk” by the 2018 ASH guide­lines had
an actual VTE inci­dence of 20.0 (14.9-25.7) per 1000 cesar­ean
deliv­er­ies (2% risk) at 6 weeks post­par­tum, whereas other clin­i­cal
prac­tice guide­lines’ “high-risk” categories had lower actual VTE
inci­dences (Table 2).22 In con­trast, in a sin­gle-cen­ter study, the
highest cal­cu­lated VTE risk was approx­im ­ a­tely 0.5% using a risk
score derived by Sultan et al31, which is lower than the VTE risk
threshold cutoff suggested by the ASH guidelines.23
There have been preg­nancy and post­par­tum VTE risk scores
derived and exter­nally val­i­dated from large reg­is­try data­bases,
but these scores have yet to be stud­ied pro­spec­tively or incor­
po­rated into guide­lines.31-33 Several lim­i­ta­tions exist in cohort and
data­base stud­ies, includ­ing dif­fer­ent RF and VTE def­i­ni­tions used
in admin­is­tra­tive data­bases, miss­ing data for some VTE RFs, low
num­bers of patients with high-risk con­di­tions, and, most impor­
tant, the actual LMWH pro­phy­laxis use is often not accu­rately
cap­tured.
Although esti­mat­ing what the true ben­e­fit of LMWH is remains
unknown, what is equally chal­leng­ing is esti­mat­ing the bleed­
ing risk and pos­si­ble wound com­pli­ca­tions, cost, and bur­den
of LMWH injec­tions. Introducing an anti­co­ag­u­lant post-CD may
the­ o­ret­
i­
cally affect wound healing of the inci­ sion by caus­ ing
local­ized bleed­ing, but lit­tle data remain in this area. In a large
ret­ro­spec­tive cohort study of 24 229 deliv­er­ies, VTE and bleed­ing
out­comes were assessed before and after a stan­dard­ized throm-
boprophylaxis hos­pi­tal pro­to­col was implemented. There was
1.2% and 15.6% of anti­co­ag­u­lant use before and after pro­to­col
implementation, respec­tively. There were no dif­fer­ences in VTE
rates seen before and after pro­to­col implementation, but there
was a 2-fold higher rate of super­fi­cial wound hema­to­mas.31 With
com­pet­ing risks and pos­si­ble com­pli­ca­tions, under­stand­ing the
patient per­spec­tive on LMWH use, as well as how we best com­
mu­ni­cate the benefit and risk of LMWH to our patients, is still
needed.
Two pilot tri­als were conducted that assessed the fea­si­bil­
ity of ran­dom­iz­ing postpartum women with VTE RFs to LMWH
 Table 1. LMWH pro­phy­laxis rec­om­men­da­tions across major clin­i­cal prac­tice guide­lines after cesar­ean deliv­ery
Characteristic ASH 201813
Elective For women with no or 1
CD alone clin­i­cal RF (exclud­ing a
known thrombophilia or
his­tory of VTE), the ASH
Emergent
guide­line panel sug­gests
CD alone
against antepartum
post­par­tum pro­phy­laxis
(con­di­tional rec­om­men­
da­tion, low cer­tainty in
evi­dence about effects).
Elective Not stated
CD + RF
Prakash Singh Shekhawat
VTE pre­ven­tion and man­age­ment in preg­nancy  |  561
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021
RCOG 201516 SOGC 201418 ACCP 201215 ANZJOG 201217
No No For women under­go­ing No; early mobi­li­za­tion
cesar­ean sec­tion with­ and avoid­ance of
out addi­tional throm­ dehy­dra­tion
bo­sis RFs, we rec­om­
All women who have had cesar­ean No Following emer­gency
mend against the use of
sec­tions should be con­sid­ered cesar­ean sec­tion
throm­bo­sis pro­phy­laxis
for thromboprophylaxis with thromboprophylaxis
other than early mobi­li­
LMWH for 10 days after deliv­ with LMWH or UFH is
za­tion (grade 1B).
ery apart from those hav­ing recommended for at
an elec­tive cesar­ean sec­tion least 5 days or lon­ger
who should be con­sid­ered for until recov­ery of full
thromboprophylaxis with LMWH mobil­ity (group con­sen­
for 10 days after deliv­ery if they sus level 1).
have any addi­tional RFs
(grade C).
See above; con­sid­ered for Postpartum thromboprophylaxis For women at increased ≥1 major and ≥2 minor
thromboprophylaxis with LMWH should be con­sid­ered in the risk of VTE after cesar­ RFs: Postpartum
for 10 days after deliv­ery pres­ence of mul­ti­ple clin­i­cal or ean sec­tion because of thromboprophylaxis
(grade C). preg­nancy-related RFs when the pres­ence of for ≥5 days or until fully
the over­all abso­lute risk is 1 major or at least 2 mobile; 1 major or 2
esti­mated to be greater than minor RFs, minor RFs: Consider
1% drawn from the fol­low­ing we sug­gest phar­ grad­u­ated com­pres­sion
group­ings: ma­co­logic stock­ings.
Any 2 of the fol­low­ing RFs (emer­ thromboprophylaxis
gency cesar­ean sec­tion counts (pro­phy­lac­tic LMWH) or
as 1 RF) (II-2B) mechan­i­cal pro­phy­laxis
LMWH until dis­charge up to 2 (elas­tic stock­ings or
weeks if 2 RFs inter­mit­tent pneu­matic
com­pres­sion) in those
with con­tra­in­di­ca­tions
to anti­co­ag­u­lants while
in the hos­pi­tal fol­low­ing
deliv­ery rather than no
pro­phy­laxis (grade 2B).
 Table 1. (continued)

Characteristic ASH 201813 RCOG 201516 SOGC 201418 ACCP 201215 ANZJOG 201217
Emergent Not stated All women who have had cesar­ean Postpartum thromboprophylaxis Further details in Table 3: Following emer­gency
CD + RF sec­tions should be con­sid­ered should be con­sid­ered in the Requires pres­ence of cesar­ean sec­tion,
for thromboprophylaxis with pres­ence of mul­ti­ple clin­i­cal or at least 1 major RF OR thromboprophylaxis
LMWH for 10 days after deliv­ preg­nancy-related RFs when pres­ence of at least with LMWH or UFH is
ery apart from those hav­ing the over­all abso­lute risk is esti­ 2 minor RFs (planned recommended for at
an elec­tive cesar­ean sec­tion mated to be >1% drawn from cesar­ean sec­tion) OR 1 least 5 days or lon­ger
who should be con­sid­ered for the fol­low­ing group­ings: minor RF in the set­ting until recov­ery of full
thromboprophylaxis with LMWH Any 3 or more of the fol­low­ing of an emer­gency cesar­ mobil­ity (group con­sen­
for 10 days after deliv­ery if they RFs (elec­tive cesar­ean sec­tion ean sec­tion sus level 1).
have any addi­tional RFs counts as 1 RF) (II-2B)
(grade C). LMWH until dis­charge up to 2
weeks if 1 RF
Note: These mostly include preg­nancy-related RFs and do not review inherited thrombophilia guid­ance unless stated below.
ACCP15: Major RFs: Immobility, post­par­tum hem­or­rhage ≥1000 mL with sur­gery, pre­vi­ous VTE, pre­eclamp­sia with fetal growth restric­tion, thrombophilia (AT defi­ciency, fac­tor V Leiden, pro­throm­
bin gene muta­tion), med­i­cal con­di­tions (SLE, heart dis­ease, sickle cell dis­ease), blood trans­fu­sion, post­par­tum infec­tion. Minor RFs: BMI >30 kg/m2, mul­ti­ple preg­nancy, PPH >1 L, smok­ing >10
cig­a­rettes/d, fetal growth restric­tion, thrombophilia (pro­tein C and S defi­ciency), pre­eclamp­sia.
RCOG16: See Table 1 of the 2015 RCOG guide­lines.

562  |  Hematology 2021  |  ASH Education Program

SOGC18: Any 2 of the fol­low­ing RFs: BMI ≥30 kg/m2, smok­ing >10 cig­a­rettes/d, pre­eclamp­sia, intra­uter­ine growth restric­tion, pla­centa previa, emer­gency cesar­ean sec­tion, peripartum or post­
par­tum blood loss of >1 L or blood prod­uct replace­ment, any low-risk thrombophilia (pro­tein C or pro­tein S defi­ciency), het­ero­zy­gous fac­tor V Leiden, or pro­throm­bin gene muta­tion 20210A,
mater­nal car­diac dis­ease, SLE, sickle cell dis­ease, inflam­ma­tory bowel dis­ease, var­i­cose veins, ges­ta­tional dia­be­tes, pre­term deliv­ery, still­birth. Any 3 or more of the fol­low­ing RFs: age >35 years,
par­ity ≥2, any assisted repro­duc­tive tech­nol­ogy, mul­ti­ple preg­nancy, pla­cen­tal abrup­tion, pre­ma­ture rup­ture of mem­branes, elec­tive cesar­ean sec­tion, mater­nal can­cer.
ANZJOG17: Major RFs: Elective cesar­ean sec­tion, BMI ≥30 kg/m2, immo­bi­li­za­tion, med­i­cal comorbidity (eg, inflam­ma­tory bowel dis­ease, SLE, pneu­mo­nia), pre­eclamp­sia, sys­temic infec­tion. Minor
RFs: Age >35 years, prolonged labor (>24 hours), smoker, PPH >1000 mL, exten­sive per­i­neal trauma and prolonged repair, gross var­i­cose veins.
Descriptions used for evi­dence grad­ing and con­sen­sus:
ASH 2018: The panel used the GRADE approach to assess the cer­tainty in the evi­dence and for­mu­late rec­om­men­da­tions.
RCOG 2015: Described in a sep­a­rate meth­od­ol­ogy arti­cle: “using the SIGN meth­od­ol­ogy, the qual­ity of the evi­dence used and the direct­ness of its appli­ca­tion should be incor­po­rated into the
for­mu­la­tion and grad­ing of the rec­om­men­da­tion.”
SOGC 2014: “The qual­ity of evi­dence in this doc­u­ment was rated using the cri­te­ria described in the Report of the Cana­dian Task Force on Preventative Health Care.”
ACCP 2012: “We followed the approach artic­u­lated by Grades of Recommendations, Assessment, Development, and Evaluation for for­mu­la­tion of rec­om­men­da­tions.”
ANZJOG 2012: “To assess the level of con­sen­sus with the rec­om­men­da­tions, all­authors were sent a spread­sheet list­ing all­the rec­om­men­da­tions and were asked to indi­cate whether they agreed
or disagreed with each state­ment. Recommendations were then graded with the fol­low­ing lev­els of con­sen­sus: Group Consensus Level 1—com­plete con­sen­sus: all­ten authors in agree­ment;
Group Consensus Level 2—par­tial con­sen­sus: eight of ten authors in agree­ment; Group Consensus Level 3—no con­sen­sus—two or more authors disagreed with rec­om­men­da­tion.”
AT, anti­throm­bin; BMI, body mass index; GRADE, Grading of Recommendations Assessment, Development and Evaluation; PPH, post­par­tum hem­or­rhage; SLE, sys­temic lupus erythematosus;
SOGC, Society of Obstetricians and Gynaecology of Canada.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021

Table 2. Differences in clin­i­cal prac­tice guide­lines for the use of post­par­tum thromboprophylaxis and asso­ci­ated VTE risk

Author, year, and Population n (% who Proportion Statements of actual LMWH Incidence of VTE in
loca­tion under­went CD) meet­ing cri­te­ria use women who met
for LMWH guide­line cri­te­ria for
pro­phy­laxis LMWH pro­phy­laxis
Pamerola,19 2016, USA Post-CD; cross-sec­tional 293 ACOG: 1.0% “At the cen­tre where this study NR
chart review at 2 time CD: 100% RCOG: 85% was performed, hep­a­rin is
points in 2013-2014 ACCP: 34.8% admin­is­tered empir­i­cally to
all­women after CD unless
there is a spe­cific con­tra­in­
di­ca­tion.”
Omunakwe,20 2016, UK All deliv­er­ies; 4 weeks in 227 RCOG: 46.6% NR NR
Sep­tem­ber-Octo­ber 2015 CD: 35%

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021

O’Shaughnessy,21 2019, All deliv­er­ies; cross-sec­tional 21 019 ACOG: 8% NR NR
Ireland study of pro­spec­tively col­ CD: 32% RCOG: 37%
lected data; Jan­u­ary 2015 ACCP: 7%
to Decem­ber 2017 Australia/New
Zealand: 23%
SOGC: 15%
Federspiel et al,22 2021, Post-CD; Nationwide 1 390 603 ACOG: 0.3% Unknown VTE inci­dence per 1000
USA Readmissions Database, CD: 100% RCOG: 73.4% deliv­er­ies:
Octo­ber 2015 to Decem­ ACCP: 16.2% ACOG: 19.5 (14.9-23.9)
ber 2017 ASH: 0.2%* RCOG: 1.9 (1.8-2.0)
ACCP: 4.2 (3.9-4.6)
ASH: 20.0 (14.9-25.7)
Gassmann et al,23 2021, All deliv­er­ies in Jan­u­ary 2019; 344 ACOG: 8.7% 24% Calculated based on
Switzerland ret­ro­spec­tive chart review CD: 23.3% RCOG: 40.1% “Standard of care is to pre­ Sultan risk score:
(34.9%-45.5%) scribe thromboprophylaxis ACOG: 0.20%
ACCP: 9.9% to women with CD. For RCOG: 0.12%
ASH: 0%* women with vag­i­nal deliv­ ACCP: 0.20%;
ery, LMWH is restricted to
those with thrombophilia or
prior VTE, fol­low­ing indi­vid­
ual hemo­sta­sis con­sul­ta­tion.”
*The 2018 ASH guide­lines did not spe­cif­i­cally com­ment on post-CD thromboprophylaxis use (see Table 1).
NR, not reported.

vs pla­cebo/no LMWH for 10 to 21 days. Unfortunately, both pilot
tri­als were not fea­si­ble due to low par­tic­i­pant recruit­ment rates
that aver­aged <1 par­tic­i­pant enrolled per month per cen­ter.32,33
In a sur­vey of 306 patients who were eli­gi­ble but did not par­tic­i­
pate, 32% were too overwhelmed or pre­oc­cu­pied in the post­par­
tum period to con­sider research, and 28.4% declined because
they wanted to avoid LMWH injec­tions.36 Although par­tic­i­pant
num­bers were too small in the pilot PROSPER (PostpaRtum PrO-
phylaxiS for PE Randomized Control Trial Pilot) tri­als to com­ment
on the effi­cacy or safety of post­par­tum LMWH use, there was 1
major bleed­ing event and 2 clin­i­cally rel­e­vant non­ma­jor bleed­ing
events among 16 par­tic­i­pants, which high­lights that high-qual­ity
tri­als are still needed to deter­mine the true risk and ben­e­fit of
post­par­tum thromboprophylaxis.36 The pilot PARTUM (Postpar-
tum Aspirin to Reduce Thromboembolism Undue Morbidity) trial
is ongo­ing, to see the fea­si­bil­ity of low-dose aspi­rin for 6 weeks
to pre­vent VTE in post­par­tum women with VTE RFs, com­pared
with pla­cebo (clinicaltrials​­.gov ID NCT04153760).
CLINICAL CASE 1 (Con­t in­u ed)
Given the lim­
ited data in this area, the patient should be
approached to par­tic­i­pate in a research study such as the pilot
Prakash Singh Shekhawat
PARTUM trial. In the absence of an avail­­able research study, the
2018 ASH guide­line panel sug­gests against thromboprophylaxis
use for this patient, even with the pres­ence of a fam­ily his­tory
of VTE. Because other clin­i­cal prac­tice guide­lines rec­om­mend
thromboprophylaxis in this sce­nario, fur­ther dis­cus­sion is needed
with the patient before mak­ing an informed deci­sion, includ­ing
high­light­ing this prac­tice var­i­a­tion and reviewing the patient’s
val­ues and pref­er­ences. Regardless of the choice to use LMWH
or not post­par­tum, symp­toms of VTE and when to seek med­i­cal
atten­tion should be reviewed. A review of addi­tional preg­nancy-
related RFs should also be com­pleted at the time of deliv­ery.
CLINICAL CASE 2 (Con­tin­ued)
The abso­lute risk of post­par­tum VTE after urgent CD is less
than 1%, and it is still uncer­tain what the true ben­e­fit of LMWH
thromboprophylaxis is in this sit­ u­a­
tion, weighing the side
effects and sys­tem-level cost. Although I would not rec­om­
mend LMWH thromboprophylaxis in this case, oth­ers would
rec­om­mend LMWH use, such as in the 2015 RCOG guide­lines.
Instead, I would advo­cate for early mobi­li­za­tion and reassess
the role of LMWH if addi­tional post­par­tum VTE RFs develop.
VTE pre­ven­tion and man­age­ment in preg­nancy  |  563
Management of VTE in preg­nancy
CLINICAL CASE 3
You are see­ing a 29-year old G2P1 woman who devel­oped a symp­
tom­atic PE at 20 weeks’ ges­ta­tion based on a high-prob­a­bil­ity
V/Q scan with large mismatched defects in the left lower lobe.
Her vitals are nor­mal, and she has no exam­i­na­tion find­ings of DVT.
Her preg­nancy has been unre­ mark­able, and she has no other
med­i­cal his­tory. Her only med­i­ca­tion is a pre­na­tal vita­min. What
anti­co­ag­u­lant reg­i­men do you rec­om­mend for her preg­nancy?
Therapeutic-dose LMWH is the recommended treat­ ment
for preg­nant patients with acute VTE because it does not cross
the pla­centa and has an improved safety pro­file com­pared with
unfractionated hep­ a­
rin (UFH). Vitamin K antag­ o­
nists are not
recommended in preg­nancy due to known ter­a­to­ge­nic­ity, and
direct oral anti­co­ag­u­lants still have lim­ited safety infor­ma­tion
(with an International Society on Thrombosis and Haemostasis
reg­is­try ongo­ing).37,38 Side effects of LMWH use dur­ ing preg­
nancy include a small risk of impor­tant bleed­ing antepartum
(~0.5%), injec­tion site bruis­ing and reac­tions, and a very rare
pos­si­bil­ity of hep­a­rin-induced throm­bo­cy­to­pe­nia.39 Unlike UFH,
pro­phy­lac­tic or inter­me­di­ate-dose LMWH does not reduce bone
min­eral den­sity dur­ing preg­nancy, but less data are avail­­able
for ther­a­peu­tic doses.40,41 Systemic thrombolysis should only be
reserved for patients with PE who have life-threat­en­ing hemo­dy­
namic com­pro­mise or car­diac arrest given the exces­sive bleed­
ing risk (among 83 women treated with sys­temic thrombolysis:
mater­ nal sur­ vival, 94%; fetal sur­vival, 88%; antepartum major
bleed­ing, 17.5%; post­par­tum major bleed­ing, 58.3%).42 Beyond
these state­ments, there remains sig­nif­i­cant prac­tice var­i­a­tion on
the details of anticoagulation man­age­ment in preg­nancy.43
As preg­nancy advances, there is an increased vol­ume of dis­
tri­bu­tion and glo­mer­u­lar fil­tra­tion rate, so there is the poten­
tial for increased clear­ance of LMWH.38,44 Because of these
preg­nancy-spe­cific changes, there remains con­tro­versy on the
use of once- vs twice-daily LMWH dos­ing and the role of anti-
Xa level mon­i­tor­ing of LMWH. In a 2014 Cana­dian sur­vey of 69
hema­tol­o­gists, inter­nists, and obste­tri­cians, there was con­sid­
er­able prac­tice var­i­a­tion in acute VTE man­age­ment. Within the
first month of a VTE event, par­tic­i­pants used either once-daily
(36%) or twice-daily (62%) LMWH, and anti-Xa level mon­i­tor­ing
was com­pleted weekly (20%), monthly (26%), weekly or monthly
in spe­cial pop­u­la­tions only (23%), or never (20%).40
A sys­tem­atic review in non­preg­nant patients showed sim­i­lar
out­comes of once- vs twice-daily LMWH dos­ing.45 To date, there
has been no clear out­come dif­fer­ence in recur­rent VTE seen in
ret­ro­spec­tive cohorts of preg­nant patients treated with dif­fer­
ent reg­i­mens.46-47 Information related to anti-Xa level mon­i­tor­ing
is lim­ited to small cohorts of preg­nant patients who received
dif­fer­ent ther­a­peu­tic-dose LMWH reg­i­mens in which phy­si­cians
titrated LMWH doses based on var­i­ous anti-Xa level tar­gets; no
major safety sig­nals were iden­ti­fied across dif­fer­ent strat­e­gies,
albeit with lim­ited data from a small num­ber of patients.48,54 One
chal­lenge in interpreting anti-Xa lev­els is that a spe­cific “anti-
Xa level range” in preg­nancy is not known, because there is no
data avail­­able to cor­re­late anti-Xa lev­els with VTE or bleed­ing
564  |  Hematology 2021  |  ASH Education Program
out­comes. In the larg­est non-ran­dom­ized study of 26 par­tic­i­
pants (11 received anti-Xa level mon­i­tor­ing and 15 did not receive
anti-Xa level mon­i­tor­ing), there was no dif­fer­ence in recur­rent
VTE or bleed­ing reported.53 Based on lim­ited evi­dence, the 2018
ASH guide­line panel sug­gests either a once-daily or twice-daily
LMWH reg­i­men be used and suggested against anti-Xa level
mon­i­tor­ing with the poten­tial excep­tion of higher-risk sce­nar­ios,
such as in those patients with obe­sity or advanced renal dys­
func­tion.13 Table 3 sum­ma­rizes var­i­ous clin­i­cal prac­tice guide­line
rec­om­men­da­tions for once- vs twice-daily LMWH, anti-Xa level
mon­i­tor­ing, and peripartum anticoagulation man­age­ment (dis-
cussed fur­ther in case 4).
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021
CLINICAL CASE 3 (Con­tin­ued)
Whatever anti­co­ag­u­lant reg­i­men is cho­sen, it is impor­tant to
have close fol­low-up and mon­i­tor­ing of symp­toms, espe­cially in
the first month of treat­ment. The rec­om­men­da­tions presented
by the 2018 ASH guide­lines are a patient-focused approach
that includes min­i­miz­ing the num­ber of injec­tions and lab­o­ra­
tory vis­its. Although this aligns with my gen­eral prac­tice and
approach for the major­ity of patients, I would con­sider esca­
lat­ing to a twice-daily LMWH and/or anti-Xa level mon­i­tor­ing
in the first month of VTE treat­ment in those with higher-risk
fea­tures, such as those with exten­sive VTE bur­den, recur­rent
VTE despite anticoagulation, or patients with a high-risk throm-
bophilia such as antiphospholipid syn­drome.
Peripartum man­age­ment of anticoagulation
CLINICAL CASE 4
A 38-year-old G0 woman had a large PE 2 years ago after start-
ing a com­bined oral con­tra­cep­tive pill. After receiv­ing 6 months
of anticoagulation, she stopped anticoagulation and the con­
tra­cep­tive pill and had a pro­ges­ter­one-only intra­uter­ine device
placed. She is now inter­ested in get­ting preg­nant. You meet
her for a pre­con­cep­tion coun­sel­ing visit and recommended
starting pro­phy­lac­tic-dose LMWH dur­ing preg­nancy and the
6-week post­par­tum period to pre­vent recur­rent VTE. She has
sev­eral ques­tions about what this means for her deliv­ery and if
she will be ­able to get an epi­du­ral for pain con­trol.
One of the more chal­ leng­ing areas with lit­ tle high-qual­ity
data is when and how to stop anticoagulation around labor and
deliv­ery, as well as when to resume post­par­tum anticoagulation.
Systematic reviews report the var­i­able and low-qual­ity ret­ro­
spec­tive data avail­­able that phy­si­cians rely on to make deci­sions
about anti­co­ag­u­lant man­age­ment.37,56,57 One par­tic­u­lar chal­lenge
in the lit­er­a­ture has been the dif­fer­ent def­i­ni­tions of peripartum
bleed­ ing; peripartum bleed­ ing def­i­
ni­tions may be based on
esti­mated blood loss alone (variably mea­sured across cen­ters),
a decrease in hemo­glo­bin, blood trans­fu­sion, hos­pi­tal readmis-
sions, repeat sur­gery, or wound com­pli­ca­tions.57-62 To min­i­mize
this var­i­at­ ion, the International Society on Thrombosis and Hae-
mostasis Committee of Women’s Health Issues in Thrombosis
 Table 3. Recommendations of VTE man­age­ment in preg­nancy across major clin­i­cal prac­tice guide­lines
Characteristic ASH 201813
Once- vs twice- For preg­nant women with
daily LMWH acute VTE treated with
LMWH, the ASH guide­
line panel sug­gests either
once-per-day or twice-
per-day dos­ing reg­i­mens
(con­di­tional rec­om­men­da­
tion, very low cer­tainty in
evi­dence about effects).
Anti-Xa level For preg­nant women receiv­
mon­i­tor­ing ing ther­a­peu­tic-dose
LMWH for the treat­ment
of VTE, the ASH guide­line
panel sug­gests against
rou­tine mon­i­tor­ing of anti-
Prakash Singh Shekhawat
FXa lev­els to guide dos­ing
(con­di­tional rec­om­men­
da­tion, low cer­tainty in
evi­dence about effects).
VTE pre­ven­tion and man­age­ment in preg­nancy  |  565
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021
RCOG 201552
LMWH should be given in
doses titrated against
the woman’s book­ing or
early preg­nancy weight.
There is insuf­fi­cient evi­
dence to rec­om­mend
whether the dose of
LMWH should be given
once daily or in two
divided doses (grade C).
Routine mea­sure­ment of
peak anti-Xa activ­ity for
patients tak­ing LMWH for
treat­ment of acute VTE in
preg­nancy or post­par­
tum is not recommended
except in women at
extremes of body weight
(<50 and 90 kg or more)
or with other com­pli­cat­
ing fac­tors (eg, with renal
impair­ment or recur­rent
VTE) (grade C).
SOGC 201418
For the treat­ment of acute
VTE in preg­nancy, we
rec­om­mend adher­ing
to the man­u­fac­turer’s
recommended dos­ing for
indi­vid­ual LMWH based
on the woman’s cur­rent
weight. (II-1A).
LMWH can be admin­is­
tered once or twice a day
depending on the agent
selected (III-C).
No rec­om­men­da­tion
In the text: con­sid­er­ation
should be given to ini­tial
mon­i­tor­ing of anti-Xa activ­
ity, dur­ing the first month
of treat­ment only, to tar­get
a level of 0.6 to 1.0 U/mL
4 hours after injec­tion,
bear­ing in mind that tar­get
lev­els will vary with the
LMWH used. However, the
cost of the assay, the lack
of cor­re­la­tion with clin­i­cal
events, and the var­i­abil­ity
between assays make the
util­ity of mon­i­tor­ing anti-
FXa activ­ity in preg­nancy
con­tro­ver­sial.
ACCP 201215
No rec­om­men­da­tion
No rec­om­men­da­tion
In the text:
Given the absence of large
stud­ies using clin­i­cal end
points that dem­on­strate an
opti­mal ther­a­peu­tic anti-Xa
LMWH range or that dose
adjust­ments increase the
safety or effi­cacy of ther­apy,
the lack of accu­racy and
reli­abil­ity of the mea­sure­
ment, the lack of cor­re­la­tion
with risk of bleed­ing and
recur­rence, and the cost of
the assay, rou­tine mon­i­
tor­ing with anti-Xa lev­els is
dif­fi­cult to jus­tify.
ANZJOG 201217
Treatment of acute VTE in
preg­nancy should be with
LMWH given once daily or
twice daily at ther­a­peu­tic
doses. There is cur­rently
insuf­fi­cient evi­dence to
favor one dose reg­i­men
over the other (group con­
sen­sus level 1).
Women with PE or more
exten­sive DVT (ie,
iliofemoral throm­bo­sis)
dur­ing preg­nancy should
receive ini­tial treat­ment
with twice-daily LMWH for
at least 8 to 12 weeks, after
which time a reduc­tion to a
once-daily reg­i­men may be
con­sid­ered (group con­sen­
sus level 2).
There is insuf­fi­cient evi­dence
to rec­om­mend mon­i­tor­ing
of anti-Xa lev­els to guide
dos­ing in women on ther­a­
peu­tic dose LMWH (group
con­sen­sus level 1).
 Table 3. (continued)

Characteristic ASH 201813
Peripartum For preg­nant women receiv­
anticoagulation ing ther­a­peu­tic-dose
man­age­ment LMWH for the man­age­ment
of VTE, the ASH guide­line
panel sug­gests sched­uled
deliv­ery with prior dis­
con­tin­u­a­tion of anti­co­ag­
u­lant ther­apy (con­di­tional
rec­om­men­da­tion, very low
cer­tainty in evi­dence about
effects).
For preg­nant women receiv­
ing pro­phy­lac­tic-dose
LMWH, the ASH guide­line
panel sug­gests against
sched­uled deliv­ery with
dis­con­tin­u­a­tion of pro­
phy­lac­tic anticoagulation
RCOG 201552
The woman tak­ing LMWH
for main­te­nance ther­apy
should be advised that
once she is in established
labor or thinks that she is
in labor, she should not
inject any fur­ther hep­a­rin.
When VTE occurs at term,
con­sid­er­ation should
be given to the use of
intra­ve­nous UFH, which is
more eas­ily manip­u­lated
(grade D).
Where deliv­ery is planned,
either by elec­tive cesar­
ean sec­tion or induc­tion
of labor, LMWH main­te­
nance ther­apy should be
SOGC 201418
Women receiv­ing pro­
phy­lac­tic, inter­me­di­
ate-dose, or ther­a­peu­tic
anticoagulation should
have a dis­cus­sion about
options
for anal­ge­sia/anes­the­sia
prior to deliv­ery (III-B).
Switching from
thromboprophylactic
LMWH to a pro­phy­lac­tic
dose of UFH at term (37
weeks) may be con­sid­ered
to allow for more options
with respect to labor anal­
ge­sia (III-L).
Discontinue pro­phy­lac­tic or
inter­me­di­ate-dose LMWH
ACCP 201215 ANZJOG 201217
For preg­nant women receiv­ No rec­om­men­da­tion
ing adjusted-dose LMWH
ther­apy and where deliv­ery
is planned, we rec­om­mend
dis­con­tin­u­a­tion of LMWH
at least 24 hours prior to
induc­tion of labor or cesar­
ean sec­tion (or expected
time of neuraxial anes­the­
sia) rather than con­tinu­ing
LMWH up until the time of
deliv­ery (grade 1B).

566  |  Hematology 2021  |  ASH Education Program

com­pared with allowing
spon­ta­ne­ous labor (con­
di­tional rec­om­men­da­tion,
very low cer­tainty in evi­
dence about effects).
Time of neuraxial
anes­the­sia after
last dose of
LMWH
See Table 1 for descrip­tions used for evi­dence grad­ing and con­sen­sus.
discontinued 24 hours
prior to planned deliv­ery
(grade D).
Regional anes­thetic or anal­
ge­sic tech­niques should
not be under­taken until
at least
24 hours after the last
dose of ther­a­peu­tic
LMWH (grade D).
or UFH upon the onset
of spon­ta­ne­ous labor or
the day before a planned
induc­tion of labor or cesar­
ean sec­tion (II-3B).
For women tak­ing LMWH,
neuraxial anes­the­sia can
be admin­is­tered as a:
(a) Prophylactic dose: a min­i­
mum of 10 to 12 hours after
the last dose (III-B)
(b) Therapeutic dose: after
24 hours since the last
dose (III-B)
Neuraxial anes­the­sia must be
avoided in a woman who
is fully anticoagulated or in
whom there is evi­dence of
altered coag­u­la­tion (II-3A).
For preg­nant women receiv­
ing adjusted-dose LMWH
ther­apy and where deliv­ery
is planned, we rec­om­mend
dis­con­tin­u­a­tion of LMWH
at least 24 hours prior to
induc­tion of labor or cesar­
ean sec­tion (or expected
time of neuraxial anes­the­
sia) rather than con­tinu­ing
LMWH up until the time of
deliv­ery (grade 1B).
Table 3: LMWH—pro­phy­lac­tic
dose: ensure a min­i­mum of
12 hours after LMWH dose
before the per­for­mance of
a neuraxial block or removal
of a neuraxial cath­e­ter.
LMWH—ther­a­peu­tic dose:
pref­er­a­ble to avoid ther­a­
peu­tic dos­ing with cath­e­ter
in situ; wait at least 24 hours
after the last dose of LMWH
before performing neuraxial
block­ade or remov­ing a
neuraxial cath­e­ter.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021

and Haemostasis devel­oped and published stan­dard­ized preg­
nancy and peripartum bleed­ ing out­ come def­i­ni­tions, which
focus on the inter­ven­tions needed to treat impor­tant blood loss
rather than on esti­mated blood loss alone.63
There is con­tro­versy on what to do for a patient tak­ing pro­
phy­lac­tic-dose LMWH because there is a rel­a­tively low bleed­ing
risk, and access to neuraxial anes­the­sia can occur 12 hours after
the last anti­ co­
ag­u­
lant dose, according to sev­ eral guide­ lines,
includ­ing the Amer­i­can Society of Regional Anesthesia and the
Society for Obstetric Anesthesia and Perinatology (Table 3).64,65
The 2018 ASH guide­line panel sug­gests a spon­ta­ne­ous labor
over a timed induc­tion of labor for preg­nant patients tak­ing pro­
phy­lac­tic-dose LMWH, but acknowl­edg­ing this deci­sion is also
based on indi­vid­ual patient val­ues and pref­er­ences.13 In a ret­ro­
spec­tive cohort study of 199 patients tak­ing pro­phy­lac­tic-dose
LMWH in preg­nancy, approx­i­ma­tely 90% of those in spon­ta­ne­
ous labor were eli­gi­ble for neuraxial anes­the­sia, which did not
dif­fer based on if the patient was pri­mip­a­rous (first preg­nancy)
or mul­tip­a­rous.66 In this ret­ro­spec­tive study, there was nota­bly
less time off of anticoagulation for patients in the spon­ta­ne­ous
labor group com­pared with the induc­tion of labor group (last
LMWH injec­tion to deliv­ery inter­val: 25.8 vs 48.2 hours, respec­
tively).66 If a patient val­ues access to neuraxial anes­the­sia for pain
con­trol, then other options include a planned induc­tion of labor
or stop­ping LMWH early depending on the sce­nario and asso­ci­
ated VTE risk. For exam­ple, the last dose of LMWH can be given
the day prior to an induc­tion of labor to ensure that at least 12
hours have passed since the last anti­co­ag­u­lant dose was given
for the patient to be eli­gi­ble for neuraxial anes­the­sia. However, in
prac­tice, this is often chal­leng­ing because the induc­tion of labor
tim­ing may not be known until the actual day due to hos­pi­tal
logis­tics, and the induc­tion of labor dura­tion is often var­i­able and
can be prolonged in some patients. A dis­cus­sion between mul­ti­
dis­ci­plin­ary care pro­vid­ers can be help­ful to coor­di­nate logis­tics
and the details of the induc­tion of labor, to pro­vide accu­rate risks
and ben­e­fits of each approach with the patient.
For patients receiving ther­a­peu­tic-dose LMWH, because of
the poten­tial for excess bleed­ing risk dur­ing deliv­ery and delayed
access to neuraxial anes­the­sia of 24 hours after the last LMWH
dose,64,65 the 2018 ASH guide­line panel rec­om­mends a planned
labor (eg, induc­tion of labor), with the last dose of LMWH given
24 hours prior to induc­tion/deliv­ery.13 If the VTE is not acute (>3-6
months), then an alter­na­tive approach (albeit with lim­ited data)
includes reduc­ing the LMWH dose closer to deliv­ery from a ther­
a­peu­tic dose to a pro­phy­lac­tic dose, to allow for spon­ta­ne­ous
deliv­ery. If the VTE is within 2 weeks of deliv­ery, then using a
more cau­tious anti­co­ag­u­lant reg­i­men may be con­sid­ered to
min­i­mize time off of anticoagulation, such as starting or switch-
ing to a twice-daily LMWH or arranging admis­sion to the hos­pi­
tal for intra­ve­nous UFH. An inferior vena cava (IVC) fil­ter inser­tion
may be con­sid­ered if the VTE is within 2 weeks of deliv­ery, but
this is largely based on expert opin­ion. IVC fil­ter inser­tion should
­otherwise be avoided when pos­si­ble due to the chal­lenges of
inser­tion and pos­si­ble migra­tion relat­ing to the gravid uterus and
IVC changes, and the deci­sion should be made in con­sul­ta­tion
with a mul­ti­dis­ci­plin­ary spe­cial­ist team. The fail­ure-to-retrieve
rate of IVC fil­ters among 80 preg­nant patients was 11.3%, so it is
impor­tant to remove the IVC fil­ter after deliv­ery once anticoagu-
lation is safely resumed.67
Prakash Singh Shekhawat
Two mul­ti­cen­ter inter­na­tional pro­spec­tive cohort stud­ies are
ongo­ing (PREP and GO and the PANDA stud­ies) and will use stan­
dard­ized VTE and bleed­ing def­i­ni­tions to pro­vide more infor­ma­
tion on the risk esti­ma­tes of VTE, bleed­ing, patient-focused and
health care uti­li­za­tion out­comes that mat­ter to patients, pro­vid­
ers, and other stake­hold­ers.
Postpartum, ret­ro­spec­tive cohort data report an increased
bleed­ing risk if ther­a­peu­tic-dose anticoagulation is resumed too
soon after deliv­ery. Initiating ther­a­peu­tic-dose LMWH too soon
may lead to a seri­ous post­op­er­a­tive bleed­ing com­pli­ca­tion in
which the patient may be off of anticoagulation for longer, which
could then increase the risk of VTE. In a ret­ro­spec­tive cohort study
of 232 con­sec­u­tive women on ther­a­peu­tic anticoagulation who
deliv­ered in Que­bec, Canada, between 2003 and 2015, resum­ing
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021
ther­a­peu­tic anticoagulation within approx­i­ma­tely 15 hours68 after
CD and approximately 9 hours within vag­i­nal deliv­ery was asso­
ci­ated with a higher risk of a com­pos­ite of major hem­or­rhagic
com­pli­ca­tions (requir­ing trans­fu­sion, hos­pi­tal­i­za­tion, vol­ume
resus­ci­ta­tion, trans­fer to an inten­sive care unit, or sur­gery) and
major wound com­pli­ca­tions. Starting with a pro­phy­lac­tic-dose
LMWH post­de­liv­ery or delaying ther­a­peu­tic anticoagulation for
lon­ger appears safer. Practically, for most patients, resum­ing
daily pro­phy­lac­tic-dose LMWH the fol­low­ing day after deliv­ery
(~12-24 hours post­de­liv­ery) can be done, such as for case 4.
I follow the approach of starting with pro­phy­lac­tic-dose LMWH
before esca­lat­ing to ther­a­peu­tic-dose LMWH on day 2 or 3
post­de­liv­ery if needed. For higher-risk cases, such as an acute
VTE within 30 days, this approach should be mod­i­fied and may
include starting with twice-daily LMWH or intra­ ve­
nous UFH
6 to 12 hours post­de­liv­ery. Postpartum, either LMWH or vita­min K
antag­o­nists are safe with breastfeeding. Data indi­cate that direct
oral anti­co­ag­u­lants cross into breastmilk and should be avoided
for breastfeeding women until more data are avail­­able.69-70
Conflict-of-inter­est dis­clo­sure
Leslie Skeith has received research funding from CSL Behring,
and hon­o­rar­ium from Leo Pharma and Sanofi.
Off-label drug use
Leslie Skeith: the use of low-molecular-weight heparins is
off-label use for the prevention and treatment of VTE during
pregnancy and the postpartum period. The use of aspirin for pre-
vention of postpartum VTE is off-label and should only be used in
the setting of a clinical trial.
Correspondence author
Leslie Skeith, Division of Hematology and Hematological Malig-
nancies, Department of Medicine, Foothills Medical Centre, Uni-
versity of Calgary, Room C210, 1403-29th Street NW, Calgary
T2N 2T9, Alberta, Canada; e-mail: laskeith@ucalgary​­.ca.
References
1. Kourlaba G, Relakis J, Kontodimas S, Holm MV, Maniadakis N. A sys­tem­
atic review and meta-anal­y­sis of the epi­de­mi­­ol­ogy and bur­den of venous
throm­bo­em­bo­lism among preg­nant women. Int J Gynaecol Obstet.
2016;132(1):4-10.
2. Heit JA, Kobbervig CE, James AH, Petterson TM, Bailey KR, Melton LJ III.
Trends in the inci­dence of venous throm­bo­em­bo­lism dur­ing preg­nancy
or post­par­tum: a 30-year pop­u­la­tion-based study. Ann Intern Med. 2005;
143(10):697-706.
VTE pre­ven­tion and man­age­ment in preg­nancy  |  567
3. Pomp ER, Lenselink AM, Rosendaal FR, Doggen CJ. Pregnancy, the post­
par­tum period and prothrombotic defects: risk of venous throm­bo­sis in the
MEGA study. J Thromb Haemost. 2008;6(4):632-637.
4. Knight M, Bunch K, Tuffnell D, et  al. Saving Lives, Improving Mothers’
Care: les­sons learned to inform mater­nity care from the UK and Ireland
Confidential Enquiries into Maternal Deaths and Morbidity 2016-2018,
published Dec. 2020. https://www.nhsbmenetwork.org.uk/wp-content
/uploads/2021/03/MBRRACE-UK_Maternal_Report_Dec_2020_v10-1.pdf.
5. Wik HS, Jacobsen AF, Sandvik L, Sandset PM. Prevalence and pre­dic­tors
for post-throm­botic syn­drome 3 to 16 years after preg­nancy-related venous
throm­bo­sis: a pop­u­la­tion-based, cross-sec­tional, case-con­trol study. J
Thromb Haemost. 2012;10(5):840-847.
6. Kahn SR, Hirsch AM, Akaberi A, et  al. Functional and exer­cise lim­i­ta­tions
after a first epi­sode of pul­mo­nary embolism: results of the ELOPE pro­spec­
tive cohort study. Chest. 2017;151(5):1058-1068.
7. Hunter R, Noble S, Lewis S, Bennett P. Long-term psy­cho­so­cial impact
of venous throm­bo­em­bo­lism: a qual­i­ta­tive study in the com­mu­nity. BMJ
Open. 2019;9(2):e024805.
8. van Es J, Exter PL Den, Kaptein AA, et al. Quality of life after pul­mo­nary embo-
lism as assessed with SF-36 and PEmb-QoL. Thromb Res. 2013;132:500-505.
9. Kahn SR, Ducruet T, Lamping DL, et al. Prospective eval­u­a­tion of health-
related qual­ity of life in patients with deep venous throm­bo­sis. Arch Intern
Med. 2005;165(10):1173-1178.
10. Croles FN, Nasserinejad K, Duvekot JJ, Kruip MJ, Meijer K, Leebeek FW.
Pregnancy, thrombophilia, and the risk of a first venous throm­bo­sis: sys­
tem­atic review and Bayes­ian meta-anal­y­sis. BMJ. 2017;359:j4452.
11. Jacobsen AF, Skjeldestad FE, Sandset PM. Incidence and risk pat­terns of
venous throm­bo­em­bo­lism in preg­nancy and puer­pe­rium-a reg­is­ter-based
case-con­trol study. Am J Obstet Gynecol. 2008;198(2):233.e1-233.e7.
12. Skeith L. Preventing venous throm­bo­em­bo­lism dur­ing preg­nancy and
post­par­tum: cross­ing the thresh­old. Hematol Am Soc Hematol Educ Pro-
gram. 2017;2017(1):160-167.
13. Bates SM, Rajasekhar A, Middeldorp S, et al. Amer­i­can Society of Hematology
2018 guide­lines for man­age­ment of venous throm­bo­em­bo­lism: venous throm­
bo­em­bo­lism in the con­text of preg­nancy. Blood Adv. 2018;2(22):3317-3359.
14. Amer­i­can College of Obstetricians and Gynecologists’ Committee on Prac-
tice Bulletins—Obstetrics. ACOG Practice Bulletin No. 196: throm­bo­em­bo­
lism in preg­nancy. Obstet Gynecol. 2018;132(1):e1-e17.
15. Bates SM, Greer IA, Middeldorp S, Veenstra DL, Prabulos A-M, Vandvik P-O.
VTE, thrombophilia, antithrombotic ther­apy, and preg­nancy: Antithrom-
botic Therapy and Prevention of Thrombosis, 9th ed: Amer­i­can College
of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest.
2012;141(2, suppl):e691S-e736S.
16. No GG. Reducing the Risk of Venous Thromboembolism During Pregnancy
and the Puerperium Pregnancy and the Puerperium. Green-Top Guide-
line No. 37a. Royal College of Obstetricians and Gynaecologists; 2015.
https://www.rcog.org.uk/globalassets/documents/guidelines/gtg-37a
.pdf. Accessed Nov 4 2021.
17. McLintock C, Brighton T, Chunilal S, et al; Councils of the Society of
Obstetric Medicine of Australia and New Zealand; Australasian Society
of Thrombosis and Haemostasis. Recommendations for the pre­ ven­
tion
of preg­nancy-asso­ci­ated venous throm­bo­em­bo­lism. Aust N Z J Obstet
Gynaecol. 2012;52(1):3-13.
18. Chan W-S, Rey E, Kent N-E, et al; VTE in Pregnancy Guideline Working
Group; Society of Obstetricians and Gynecologists of Canada. Venous
throm­bo­em­bo­lism and antithrombotic ther­apy in preg­nancy. J Obstet
Gynaecol Can. 2014;36(6):527-553.
19. Palmerola KL, D’Alton ME, Brock CO, Friedman AM. A comparison of
­recommendations for pharmacologic thromboembolism prophylaxis after
caesarean delivery from three major guidelines. BJOG. 2016;123:2157-2162.
20. Omunakwe HE, Roberts LN, Patel JP, et al. Re: A comparison of the recom-
mendations for pharmacologic thromboembolism prophylaxis after cae-
sarean delivery from the major guidelines: Impact on thromboprophylaxis
rates of implementing Royal College of Obstetricians and Gynaecologists’
guidance for reducing the risk of ante- and postnatal venous thromboem-
bolism. BJOG. 2017;124:831-832.
21. O’Shaughnessy F, Donnelly JC, Bennett K, et al. Prevalence of postpartum
venous thromboembolish risk factors in an Irish urban obstetric popula-
tion. J Thromb Haemost. 2019;17:1875-1885.
22. Federspiel JJ, Wein LE, Addae-Konadu KL, et al. Venous throm­bo­em­bo­lism
inci­dence among patients recommended for phar­ma­co­logic throm­bo­em­
568  |  Hematology 2021  |  ASH Education Program
bo­lism pro­phy­laxis after cesar­ean deliv­ery in selected guide­lines. J Thromb
Haemost. 2021;19(3):830-838.
23. Gassmann N, Viviano M, Righini M, Fontana P, Martinez de Tejada B, Blon-
don M. Estimating the risk thresh­olds used by guide­lines to rec­om­mend
post­par­tum thromboprophylaxis. J Thromb Haemost. 2021;19(2):452-
459.
24. Middleton P, Shepherd E, Gomersall JC. Venous throm­bo­em­bo­lism pro­phy­
laxis for women at risk dur­ing preg­nancy and the early post­na­tal period.
Cochrane Database Syst Rev. 2021;3:CD001689.
25. Sibai BM, Rouse DJ. Pharmacologic thromboprophylaxis in obstet­ rics:
broader use demands bet­ter data. Obstet Gynecol. 2016;128(4):681-684.
26. Bates SM, Middeldorp S, Rodger M, James AH, Greer I. Guidance for the
treat­ment and pre­ven­tion of obstet­ric-asso­ci­ated venous throm­bo­em­bo­
lism. J Thromb Thrombolysis. 2016;41(1):92-128.
27. Blondon M. Thromboprophylaxis after cesar­ean sec­tion: deci­sion anal­y­sis.
Thromb Res. 2011;127(suppl 3):S9-S12.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021
28. Kotaska A. Postpartum venous throm­bo­em­bo­lism pro­phy­laxis may cause
more harm than ben­e­fit: a crit­i­cal anal­y­sis of inter­na­tional guide­lines
through an evi­dence-based lens. BJOG. 2018;125(9):1109-1116.
29. Rodger M. Pregnancy and venous throm­bo­em­bo­lism: “TIPPS” for risk strat­
i­fi­ca­tion. Hematology Am Soc Hematol Educ Program. 2014;2014(1):387-
392.
30. Bates SM, Alonso-Coello P, Tikkinen KA, et  al. Women’s val­ues and pref­
er­ences and health state val­ua ­ ­tions for thromboprophylaxis dur­ing preg­
nancy: a cross-sec­tional inter­view. Thromb Res. 2016;140:22-29.
31. Sultan AA, West J, Grainge MJ, et  al. Development and val­i­da­tion of risk
pre­dic­tion model for venous throm­bo­em­bo­lism in post­par­tum women:
mul­ti­na­tional cohort study. BMJ. 2016;355:i6253.
32. Dargaud Y, Rugeri L, Vergnes MC, et al. A risk score for the man­age­ment of
preg­nant women with increased risk of venous throm­bo­em­bo­lism: a multi-
centre pro­spec­tive study. Br J Haematol. 2009;145(6):825-835.
33. Lu B, Sultan AA, West J, et  al. External val­i­da­tion of a model to pre­dict
women most at risk of post­par­tum venous throm­bo­em­bo­lism: mater­nity
clot risk [published online 8 June 2021]. Thromb Res.
34. Lu MY, Blanchard CT, Oglesby KR, et al. Safety and effi­cacy of a risk-based
obstet­ric hep­a­rin-based thromboprophylaxis reg­i­men. Am J Obstet Gyne-
col. 2021;224(2):S41-S42.
35. Rodger MA, Phil­lips P, Kahn SR, James AH, Konkle BA; PROSPER Investi-
gators. Low-molec­u­lar-weight hep­a­rin to pre­vent post­par­tum venous
throm­bo­em­bo­lism: a pilot randomised pla­cebo-con­trolled trial. Thromb
Haemost. 2015;113(1):212-216.
36. Rodger MA, Phil­lips P, Kahn SR, et al; PROSPER Investigators. Low molec­
u­lar weight hep­a­rin to pre­vent post­par­tum venous throm­bo­em­bo­lism:
a pilot study to assess the fea­si­bil­ity of a ran­dom­ized, open-label trial.
Thromb Res. 2016;142:17-20.
37. Schaefer C, Hannemann D, Meister R, et  al. Vitamin K antag­on ­ ists and
preg­nancy out­come: a multi-cen­tre pro­spec­tive study. Thromb Haemost.
2006;95(6):949-957.
38. Beyer-Westendorf J, Tittl L, Bistervels I, et al. Safety of direct oral anti­co­
ag­u­lant expo­sure dur­ing preg­nancy: a ret­ro­spec­tive cohort study. Lancet
Haematol. 2020;7(12):e884-e891.
39. Greer IA, Nelson-Piercy C. Low-molec­u­lar-weight hep­a­rins for thrombo-
prophylaxis and treat­ment of venous throm­bo­em­bo­lism in preg­nancy: a
sys­tem­atic review of safety and effi­cacy. Blood. 2005;106(2):401-407.
40. Rodger MA, Kahn SR, Cranney A, et al; TIPPS inves­ti­ga­tors. Long-term
dalteparin in preg­nancy not asso­ci­ated with a decrease in bone min­eral
den­sity: substudy of a ran­dom­ized con­trolled trial. J Thromb Haemost.
2007;5(8):1600-1606.
41. Galambosi P, Hiilesmaa V, Ulander V-M, Laitinen L, Tiitinen A, Kaaja R.
Prolonged low-molec­u­lar-weight hep­a­rin use dur­ing preg­nancy and
sub­se­quent bone min­eral den­sity. Thromb Res. 2016;143:122-126.
42. Martillotti G, Boehlen F, Robert-Ebadi H, Jastrow N, Righini M, Blondon M.
Treatment options for severe pul­mo­nary embolism dur­ing preg­nancy and
the post­par­tum period: a sys­tem­atic review. J Thromb Haemost. 2017;15(10):
1942-1950.
43. Gándara E, Carrier M, Rodger MA. Management of preg­nancy asso­ci­
ated venous-throm­bo­em­bo­lism: a sur­vey of prac­tices. Thromb J. 2014;
12:12.
44. Lebaudy C, Hulot JS, Amoura Z, et al. Changes in enoxaparin phar­ma­co­ki­
net­ics dur­ing preg­nancy and impli­ca­tions for antithrombotic ther­a­peu­tic
strat­egy. Clin Pharmacol Ther. 2008;84(3):370-377.
45. Bhutia S, Wong PF. Once ver­sus twice daily low molec­u­lar weight hep­a­rin
for the ini­tial treat­ment of venous throm­bo­em­bo­lism. Cochrane Database
Syst Rev. 2013;(7):1-23.
46. Lepercq J, Conard J, Borel-Derlon A, et al. Venous throm­bo­em­bo­lism dur­
ing preg­nancy: a ret­ro­spec­tive study of enoxaparin safety in 624 preg­nan­
cies. BJOG. 2001;108(11):1134-1140.
47. Voke J, Keidan J, Pavord S, Spencer NH, Hunt BJ; Brit­ish Society for Hae-
matology Obstetric Haematology Group. The man­age­ment of ante­na­tal
venous throm­bo­em­bo­lism in the UK and Ireland: a pro­spec­tive multicentre
obser­va­tional sur­vey. Br J Haematol. 2007;139(4):545-558.
48. Rey E, Rivard GE. Prophylaxis and treat­ment of throm­bo­em­bolic dis­eases
dur­ing preg­nancy with dalteparin. Int J Gynaecol Obstet. 2000;71(1):19-24.
49. Barbour LA, Oja JL, Schultz LK. A pro­spec­tive trial that dem­on­strates that
dalteparin require­ments increase in preg­nancy to main­tain ther­a­peu­tic lev­
els of anticoagulation. Am J Obstet Gynecol. 2004;191(3):1024-1029.
50. Jacobsen AF, Qvigstad E, Sandset PM. Low molec­ u­
lar weight hep­ a­
rin
(dalteparin) for the treat­ment of venous throm­bo­em­bo­lism in preg­nancy.
BJOG. 2003;110(2):139-144.
51. Rodie VA, Thomson AJ, Stewart FM, Quinn AJ, Walker ID, Greer IA. Low
molec­u­lar weight hep­a­rin for the treat­ment of venous throm­bo­em­bo­lism
in preg­nancy: a case series. BJOG. 2002;109(9):1020-1024.
52. Smith MP, Norris LA, Steer PJ, Savidge GF, Bonnar J. Tinzaparin sodium
for throm­bo­sis treat­ment and pre­ven­tion dur­ing preg­nancy. Am J Obstet
Gynecol. 2004;190(2):495-501.
53. Ní Ainle F, Wong A, Appleby N, et al. Efficacy and safety of once daily low
molec­u­lar weight hep­a­rin (tinzaparin sodium) in high risk preg­nancy. Blood
Coagul Fibrinolysis. 2008;19(7):689-692.
54. Gib­son PS, Newell K, Sam DX, et al. Weight-adjusted dos­ing of tinzaparin in
preg­nancy. Thromb Res. 2013;131(2):e71-e75.
55. Nelson-Piercy C, MacCallum P, Mackillop L, et al. Thromboembolic Disease
in Pregnancy and the Puerperium: Acute Management. Green-Top Guide-
line No. 37b. Royal College of Obstetricians and Gynaecologists; 2015.
56. Romualdi E, Dentali F, Rancan E, et  al. Anticoagulant ther­apy for venous
throm­bo­em­bo­lism dur­ing preg­nancy: a sys­tem­atic review and a meta-
anal­y­sis of the lit­er­a­ture. J Thromb Haemost. 2013;11(2):270-281.
57. Sirico A, Saccone G, Maruotti GM, et al. Low molec­u­lar weight hep­a­rin use
dur­ing preg­nancy and risk of post­par­tum hem­or­rhage: a sys­tem­atic review
and meta-anal­y­sis. J Matern Fetal Neonatal Med. 2019;32(11):1893-1900.
58. Arbuthnot C, Browne R, Nicole S, Erb SJ, Farrall L, Borg A. A dou­ble cen­tre
ret­ro­spec­tive study into rates of post­par­tum haemorrhage in women on
low molec­u­lar weight hep­a­rin. Br J Haematol. 2017;176(1):141-143.
59. Galambosi PJ, Kaaja RJ, Stefanovic V, Ulander V-M. Safety of low-molec­u­lar-
weight hep­a­rin dur­ing preg­nancy: a ret­ro­spec­tive con­trolled cohort study.
Eur J Obstet Gynecol Reprod Biol. 2012;163(2):154-159.
Prakash Singh Shekhawat
60. Knol HM, Schultinge L, Veeger NJ, Kluin-Nelemans HC, Erwich JJ, Meijer
K. The risk of post­par­tum hem­or­rhage in women using high dose of low-
molec­u­lar-weight hep­a­rins dur­ing preg­nancy. Thromb Res. 2012;130(3):334-
338.
61. Roshani S, Cohn DM, Stehouwer AC, et al. Incidence of post­par­tum haem-
orrhage in women receiv­ing ther­a­peu­tic doses of low-molec­u­lar-weight
hep­ a­
rin: results of a ret­ ro­spec­tive cohort study. BMJ Open. 2011;1(2):​
e000257.
62. Santoro R, Iannaccaro P, Prejanò S, Muleo G. Efficacy and safety of the
long-term admin­is­tra­tion of low-molec­u­lar-weight hep­a­rins in preg­nancy.
Blood Coagul Fibrinolysis. 2009;20(4):240-243.
63. Tardy B, Chalayer E, Kamphuisen PW, et al; SSC Subcommittee on Con-
trol of Anticoagulation of the ISTH. Definition of bleed­ing events in stud­
ies eval­u­at­ing pro­phy­lac­tic antithrombotic ther­apy in preg­nant women: a
sys­tem­atic review and a pro­posal from the ISTH SSC. J Thromb Haemost.
2019;17(11):1979-1988.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/559/1851791/559skeith.pdf by guest on 13 December 2021
64. Horlocker TT, Vandermeuelen E, Kopp SL, Gogarten W, Leffert LR, Benzon
HT. Regional anes­the­sia in the patient receiv­ing antithrombotic or throm­
bo­lytic ther­apy: Amer­i­can Society of Regional Anesthesia and Pain Med-
icine Evidence-Based Guidelines (fourth edi­tion). Reg Anesth Pain Med.
2018;43(3):263-309.
65. Leffert L, Butwick A, Carvalho B, et al; mem­bers of the SOAP VTE Taskforce.
The Society for Obstetric Anesthesia and Perinatology con­sen­sus state­
ment on the anes­thetic man­age­ment of preg­nant and post­par­tum women
receiv­ing thromboprophylaxis or higher dose anti­co­ag­u­lants. Anesth
Analg. 2018;126(3):928-944.
66. Rottenstreich A, Zacks N, Kleinstern G, et  al. Planned induc­tion ver­sus
spon­ta­ne­ous deliv­ery among women using pro­phy­lac­tic anticoagulation
ther­apy: a ret­ro­spec­tive study. BJOG. 2020;127(10):1241-1248.
67. Harris SA, Velineni R, Davies AH. Inferior vena cava fil­ters in preg­nancy: a
sys­tem­atic review. J Vasc Interv Radiol. 2016;27(3):354-360.e8.
68. Cote-Poirier G, Bettache N, Cote A-M, et al. Evaluation of complications
in postpartum women receiving therapeutic anticoagulation. Obstet
­Gynecol. 2020;00:1-8.
69. Zhao Y, Arya R, Couchman L, Patel JP. Are apixaban and rivaroxaban dis­trib­
uted into human breast milk to clin­i­cally rel­e­vant con­cen­tra­tions? Blood.
2020;136(15):1783-1785.
70. Daei M, Khalili H, Heidari Z. Direct oral anti­co­ag­u­lant safety dur­ing breast-
feeding: a nar­ra­tive review. Eur J Clin Pharmacol. 2021;77(10):1465-1471.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000291
VTE pre­ven­tion and man­age­ment in preg­nancy  |  569
 SURVIVORSHIP AFTER ALLOGENEIC HEMATOPOIETIC CELL TRANSPLANT

Psychosocial and financial issues after

hematopoietic cell transplantation

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/570/1851740/570buchbinder.pdf by guest on 13 December 2021

David Buchbinder1 and Nandita Khera2
CHOC Hospital, Orange, CA; and 2College of Medicine, Mayo Clinic in Arizona, Phoenix, AZ
1

With improvement in survival after hematopoietic cell transplantation (HCT), it has become important to focus on sur-
vivors’ psychosocial issues in order to provide patient-centered care across the transplant continuum. The goals of
this article are to describe updates in the literature on certain psychosocial domains (emotional/mental health and
social/financial) in HCT survivors, offer a brief overview of the status of the screening and management of these compli-
cations, and identify opportunities for future practice and research. An evidence-based approach to psychosocial care
can be broken down as primary (promoting health, raising awareness, and addressing risk factors), secondary (screening
and directing early pharmacological and nonpharmacological interventions), and tertiary (rehabilitating, limiting disabil-
ity, and improving quality of life) prevention. Implementing such an approach requires close coordination between mul-
tiple stakeholders, including transplant center staff, referring hematologist/oncologists, and other subspecialists in areas
such as palliative medicine or psychiatry. Innovative models of care that leverage technology can bring these stakehold-
ers together to fulfill unmet needs in this area by addressing barriers in the delivery of psychosocial care.

LEARNING OBJECTIVES
• Understand the burden of psychosocial complications after hematopoietic cell transplantation
• Describe the approach to screening and managing these complications
• Recognize barriers in the delivery of psychosocial care

syndrome at 10 months of age. His parents emigrated from

CLINICAL CASES
J. B. is a 47-year-old non-Hispanic White female who is 6 years
post C antigen mismatched unrelated-donor peripheral
blood hematopoietic cell transplant (HCT) for acute mye-
loid leukemia. She had chronic graft-versus-host disease
(GVHD) of the skin, fascia, and joints and is currently on
treatment with prednisone, sirolimus, and extracorporeal
photopheresis. She also had multiple other complications,
including a chronic nonhealing ulcer of the left shin, left-
lower-extremity deep vein thrombosis, chronic renal insuf-
ficiency, and a fracture of the left femur due to osteopo-
rosis, which then ended up as a mycobacterial infection
of the hip joint after surgery. She got divorced 1 year after
hematopoietic cell transplantation (HCT) and lost her job
3 years after the HCT due to missing work frequently. She
is extremely depressed due to the ongoing medical and
financial issues and was recently hospitalized for express-
ing suicidal intent.
A. R. is a 14-year-old Hispanic male who received a matched
unrelated-donor bone marrow transplant for Wiskott-Aldrich
Mexico without extended family just before his transplant.
Medicaid coverage was adequate, with minimal out-of-
pocket expenses or deductibles, but the stress of having
barely enough money for food and necessities was high,
especially during the early posttransplant period. Soon
after the HCT, the father ran out of paid time off and was
forced to take unpaid time off work. The parents shared
duties during and after transplant, often neglecting their
own health care needs. A. R. did well from a medical stand-
point but has lasting cognitive and academic deficits and
behavioral problems that make it difficult for him to man-
age school. This has led to a great deal of stress for his par-
ents, especially since they have little social support.
Introduction
J. B. and A. R. illustrate some of the issues faced by 100 000
HCT survivors in the United States, a population projected
to increase 5-fold by 2030.1 The cost of cure is the physi-
cal, psychological, and social sequelae of HCT that have a

570 | Hematology 2021 | ASH Education Program

Table 1. Physical, psy­cho­log­i­cal, and social sequelae of HCT
Physical sequelae Psychological sequelae Social sequelae
Poor phys­i­cal and Depression/anx­i­ety Financial dis­tress
men­tal health
Fatigue, sleep, pain Perceived stress Social rein­te­gra­tion
Sexual dys­func­tion Adverse cop­ing Return to work
Symptom bur­den Cognitive dys­func­tion Social sup­port/
mar­i­tal prob­lems
det­ri­men­tal impact on the qual­ity of life of these sur­vi­vors after
HCT (Table 1). HCT sur­vi­vors and their care­giv­ers have iden­ti­fied
these con­cerns as “what mat­ters to them,” and addressing them
is an impor­tant com­po­nent for deliv­ery of high-qual­ity care.2 Un-
met needs in emo­tional health and finan­cial bur­den in HCT sur­vi­
vors have been recently described.3
The goal of this arti­cle is to describe the updates in the lit­er­
a­ture for the last 5 years for psy­cho­so­cial domains (emo­tional/
men­ tal health and social/finan­ cial) across the HCT tra­jec­ tory
describ­ing the bur­den of the prob­lem, with a spe­cial focus on
the pedi­at­ric/ado­les­cent and young adult (AYA) group. Many
sem­ i­
nal stud­ ies conducted ear­ lier are not included. We also
do not include phys­i­cal and sex­ual func­tion­ing con­cerns and
psy­cho­so­cial out­comes for HCT care­giv­ers, though they may
be highly related to the domains of inter­est. We con­clude by
reviewing the cur­rent sta­tus of screen­ing and man­age­ment of
these com­pli­ca­tions and oppor­tu­ni­ties for improve­ment, includ­
ing pos­si­ble incor­po­ra­tion of recently tested psy­cho­so­cial inter­
ven­tions in prac­tice, and high­light the bar­ri­ers and mod­els of
care to address these com­pli­ca­tions.
Emotional and men­tal health and cog­ni­tive prob­lems
after HCT
Psychological sequelae fol­low­ing allo­ge­neic HCT, the impor­tance
of which can­ not be underestimated, have been documented
(depres­sion, anx­i­ety, per­ceived stress, adverse cop­ing, and neu-
rocognitive dys­func­tion). Potential risk fac­tors for psy­cho­log­i­cal
sequelae fol­low­ing HCT are numer­ous; how­ever, female gen­der,
youn­ger age, not liv­ing with a part­ner, and chronic GVHD appear
to be salient pre­dic­tors of impair­ment. Among allo­ge­neic HCT
sur­vi­vors, the most fre­quently reported unmet needs are focused
on psy­cho­log­i­cal sequelae.4 Despite this, few stud­ies of psy­cho­
log­i­cal sequelae have been conducted in long-term HCT sur­vi­
vors in a con­tem­po­rary cohort. These stud­ies are often sub­ject to
a vari­ety of meth­od­o­log­i­cal lim­i­ta­tions (small or often het­ero­ge­
neous sam­ples, a lack of lon­gi­tu­di­nal data with short fol­low-up,
and the use of var­i­able mea­sures).
Depression and anx­i­ety have been reported in 12% to 30%
of HCT sur­vi­vors.4,5 Even at very dis­tal time points as long as
>20 years fol­low­ing HCT, depres­sion and anx­i­ety remain a fre­
quent prob­lem.6 While 22% to 43% of HCT sur­vi­vors suf­fer from
emo­tional dis­tress, post­trau­matic stress dis­or­der is rel­a­tively
infre­quent (3%–13%).7 Importantly, adap­tive or maladaptive cop­
ing strat­e­gies play a cru­cial role in the occur­rence and impact of
psy­cho­log­i­cal sequelae.8,9
Estimates sug­gest that 10% to 40% of HCT sur­vi­vors expe­
ri­ence neurocognitive dys­func­tion when mea­sured with neuro-
Prakash Singh Shekhawat
psychological test­ing; how­ever, esti­ma­tes are higher when self-
reported (40%–60%).4 A pro­por­tion of HCT sur­vi­vors may expe­
ri­ence per­sis­tent neurocognitive dys­func­tion, while other
sur­vi­vors may expe­ri­ence improve­ments over time. Age, sex,
edu­ca­tion, income, cog­ni­tive reserve, type of HCT, and con­di­
tion­ing inten­sity are pre­dic­tive fac­tors for post-HCT cog­ni­tive
impair­ment. Specific domains (ie, motor func­tion­ing) appear to
be dis­pro­por­tion­ately impaired to a greater extent, whereas
oth­ers may be spared.10
Social and finan­cial prob­lems after HCT
Medical com­pli­ca­tions after HCT along with the phys­i­cal and
psy­cho­log­i­cal prob­lems described above can adversely affect
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/570/1851740/570buchbinder.pdf by guest on 13 December 2021
the social well-being as well as the employ­ment and finan­cial
sta­tus of sur­vi­vors. In turn, finan­cial hard­ship due to reduced
income, loss of employ­ment, and the cost of care for ongo­ing
active med­i­cal issues can exac­er­bate the stress and worsen the
psy­cho­log­i­cal prob­lems. Financial tox­ic­ity has emerged as an
impor­tant com­pli­ca­tion of HCT, as mul­ti­ple stud­ies have dem­
on­strated patients’ poor finan­cial well-being after HCT. A lack
of well-val­ i­
dated instru­ ments to mea­ sure social and finan­ cial
out­comes, the use of mostly small insti­tu­tional stud­ies, and the
absence of lon­gi­tu­di­nal data are some of the gaps in research
for this com­pli­ca­tion of HCT. The prev­a­lence of finan­cial hard­
ship ranges from 20% to 70% of HCT sur­vi­vors.4,11-15 In addi­tion
to the sociodemographic fac­tors, chronic GVHD in allo­ge­neic
HCT sur­vi­vors is a major fac­tor asso­ci­ated with finan­cial stress,
as two-thirds of chronic GVHD patients endure finan­cial hard­
ship despite health insur­ance, and these are likely to be patients
with lower house­hold income.16 Denials and delays in care due to
insur­ance and loss of cov­er­age are com­mon stress­ors faced by
HCT sur­vi­vors. Financial hard­ship in HCT sur­vi­vors is known to be
asso­ci­ated with poor qual­ity of life, higher per­ceived stress, non-
adherence to treat­ment, higher symp­tom bur­den, and low sat­
is­fac­tion with care but, for­tu­nately, no impact on sur­vival.11,12,16,17
The return to work (RTW) is an indi­ca­tor of func­tional recov­
ery for trans­plant sur­vi­vors, and the inabil­ity to RTW may be
asso­ci­ated with finan­cial bur­den. Only about half of the patients
who pre­vi­ously con­trib­uted to house­hold earn­ings may be ­able
to RTW by 2 years after HCT.18 Chronic GVHD is a major fac­tor
influ­enc­ing RTW/sick leave, with one-quar­ter of patients with
chronic GVHD reporting being “dis­abled/unable to work” at a
median of 2 years post HCT.16,19 Lower phys­i­cal func­tion­ing, mul-
timorbidity, being female, and the receipt of periph­eral blood
stem cells for an unre­lated HCT are asso­ci­ated with a lower like­li­
hood of RTW post HCT.20-22 A lack of stan­dard guide­lines for RTW
lead­ing to poor com­mu­ni­ca­tion between pro­vid­ers, patients,
and employers can be a bar­rier to suc­cess­ful rein­te­gra­tion of
patients in the work­place after HCT.23
While a major­ ity of HCT sur­ vi­
vors show rea­ son­able lev­
els of social adjust­ment, those with fatigue may be at risk for
maladjustment and may need effec­tive reha­bil­i­ta­tion strat­e­
gies.24 Social rein­te­gra­tion and social sup­port, includ­ing emo­
tional, infor­ma­tional, and logis­ti­cal sup­port, can help mit­i­gate
the effects of psy­cho­so­cial prob­lems, as illus­trated by stud­ies
in can­cer patients. One of the pos­i­tive sequelae of over­com­ing
trau­matic stress­ors caused by life-threat­en­ing dis­ease and inten­
sive treat­ment reported in HCT sur­vi­vors is the phe­nom­e­non of
“post­trau­matic growth.” The roles of emo­tional and instru­men­tal
social sup­port, social con­straints, and sup­port from health care
Psychosocial com­pli­ca­tions of HCT  |  571
pro­fes­sion­als in pro­mot­ing post­trau­matic growth have been elu­
ci­dated fur­ther in the last few years.14,25
Specific con­sid­er­ations for pedi­at­ric/AYA pop­u­la­tions
Pediatric patients are vul­ner­a­ble to psy­cho­so­cial sequelae after
HCT. Among pedi­at­ric HCT sur­vi­vors, esti­ma­tes of phys­i­cal, psy­
cho­log­i­cal, and social dys­func­tion vary due to het­ero­ge­ne­ity in
meth­od­o­log­i­cal aspects. Longitudinal eval­u­a­tion of pedi­at­ric
HCT sur­vi­vors is vital because the abil­ity to assess spe­cific con­
structs in chil­dren changes over time; how­ever, pro­spec­tive lon­
gi­tu­di­nal data in pedi­at­ric HCT sur­vi­vor­ship care are infre­quent.
Symptom bur­den is greater and health-related qual­ity of life is
poorer when com­pared to healthy peers, though com­pa­ra­ble
with con­ven­tion­ally treated can­cer sur­vi­vors.26 Health prob­lems
may pre­vent 16% of sur­vi­vors from attend­ing school; how­ever, the
need for spe­cial edu­ca­tion ser­vices at school was sim­i­lar between
pedi­at­ric HCT sur­vi­vors and con­ven­tion­ally treated sur­vi­vors.27
Depression as well as anx­ie ­ ty is often noted among pedi­at­ric HCT
sur­vi­vors as com­pared to healthy peers. A recent study found
that 16% of pedi­at­ric HCT sur­vi­vors endorsed depres­sion, where-
as 10% of sur­vi­vors endorsed anx­i­ety.28 Neurocognitive dys­func­
tion is prev­a­lent in about 50% of pedi­at­ric HCT sur­vi­vors.28 The
domain of motor devel­op­ment was affected to the greatest ex-
tent; how­ever, lan­guage, social, and emo­tional devel­op­ment was
also impaired com­pared to healthy peers.
AYA sur­vi­vors of HCT are a unique pop­u­la­tion that require
spe­cial care con­sid­er­ations. HCT and its sequelae may under­
mine the abil­ity of AYA sur­vi­vors to achieve mile­stones such as
grad­u­at­ing from col­lege, selecting a career, and establishing
employ­ment as well as achiev­ing socio­eco­nomic inde­pen­dence
from par­ents, serv­ing as a risk fac­tor for adverse psy­cho­so­cial
out­comes.29,30 Despite the impor­tance of devel­op­ing a greater
under­stand­ing of the AYA expe­ri­ence within the con­text of HCT
sur­vi­vor­ship, few stud­ies have focused on psy­cho­so­cial sequelae
among AYA sur­vi­vors of HCT. When we com­pare AYA sur­vi­vors
to healthy peers, dec­re­ments in phys­i­cal and social func­tion­ing
and health become appar­ent.31,32 AYA sur­vi­vors of HCT report
feel­ing left behind their peers, which some­times leads them to
rush back to work and school and take on too many respon­si­
bil­i­ties too quickly, resulting in over­whelm­ing demands.33 RTW
and employ­ment issues con­tinue to be major psy­cho­so­cial chal­
lenges for AYA sur­vi­vors of HCT.34
Risk fac­tors for adverse psy­cho­so­cial out­comes among AYA
sur­ vi­vors of HCT include sociodemographic (age, edu­ ca­
tion
Table 2. Examples of psy­cho­so­cial and finan­cial screen­ing tools
Scale
Psychosocial Assessment of Candidates for Transplant41
Transplant Evaluation Rating Scale42
Psychosocial Assessment Tool—Hematopoietic Cell Transplantation43
Stanford Integrated Psychosocial Assessment for Transplantation 44
Comprehensive Score for Financial Toxicity45
InCharge Financial Distress/Financial Well-Being Scale46
572  |  Hematology 2021  |  ASH Education Program
level, etc), trans­plant-related (con­di­tion­ing reg­i­mens with total
body irradiation), and posttransplant fac­ tors such as chronic
health con­di­tions and chronic GVHD. AYA sur­vi­vors of HCT have
a myr­iad of unmet psy­cho­so­cial needs that dif­fer from other HCT
sur­vi­vors, indi­cat­ing a neces­sity to develop inter­ven­tions spe­cif­i­
cally tai­lored to them.35
Screening for and inter­ven­tions to help improve
psy­cho­so­cial health in HCT sur­vi­vors
Most trans­ plant cen­ ters per­
form some form of psy­ cho­so­cial
eval­u­a­tion, includ­ing finan­cial assess­ment of HCT can­di­dates
prior to HCT. Table 2 pro­vi­des exam­ples of some of the screen­
ing instru­ments used in clin­i­cal prac­tice, though these may not
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/570/1851740/570buchbinder.pdf by guest on 13 December 2021
nec­es­sar­ily be lim­ited to the HCT pop­ul­a­tion. However, there
is a lot of var­i­abil­ity in terms of fre­quency and types of these
assess­ments before and after HCT at var­io ­ us trans­plant cen­ters.
Also, the approach when addressing these issues is usu­ally reac­
tive rather than pro­ac­tive. Increasing aware­ness is vital to iden­
tify sur­vi­vors and their fam­ily mem­bers who may ben­e­fit from
addi­tional in-depth assess­ment and refer­ral for resources for
these issues. Guidelines for post-HCT care rec­om­mend peri­odic
screen­ing for psy­cho­so­cial dif­fi­cul­ties, espe­cially depres­sion,
after trans­plant, at 6 and 12 months, and annu­ally there­af­ter.36
They also rec­om­mend men­tal health pro­fes­sional coun­sel­ing
for those with rec­og­nized def­i­cits and reg­ul­ar assess­ments of
spou­sal/care­giver psy­cho­log­i­cal adjust­ment and fam­ily func­
tion­ing while encour­ag­ing robust sup­port net­works. Universal
screen­ing for finan­cial dis­tress should be added to the usual
assess­ments since it is prev­a­lent and affects patient out­comes.37
Developing a set of stan­dard patient-reported out­come mea­
sures to be col­lected at peri­odic time points and to address the
domains most rel­e­vant to sur­vi­vors is cru­cial. Such mea­sures
can help under­stand the pop­u­la­tion-based prev­a­lence of these
psy­cho­so­cial issues and screen for prob­lems that would need
inter­ven­tion in real time. Electronic med­i­cal records can be lev­
er­aged to rou­tinely col­lect a stan­dard­ized set of social, psy­cho­
log­i­cal, and behav­ioral deter­mi­nants of health to inte­grate this
pro­cess with rou­tine clin­i­cal care. Screening rec­om­men­da­tions
and pos­si­ble man­age­ment options may also be incor­po­rated
into sur­vi­vor­ship care sum­ma­ries to facil­i­tate the deliv­ery of psy­
cho­so­cial care across the tra­jec­tory of HCT care.38 It is inter­est­
ing to note that these indi­vid­ua ­ l­ized care plans as edu­ca­tional
tools can lead to reduced dis­tress and improve­ment in the men­
tal domain of health-related qual­ity of life, as shown by a recent
ran­dom­ized clin­i­cal trial.39
Description
Describes psy­cho­so­cial func­tion­ing before organ and HCT
Describes psy­cho­so­cial func­tion­ing before organ and HCT
Describes the fam­ily psy­cho­so­cial risk for fam­i­lies of a child under­go­ing HCT
Measures psy­cho­so­cial read­i­ness for trans­plant
Describes the finan­cial dis­tress expe­ri­enced by can­cer patients 18 years of
age and older dur­ing the past 7 days
Measures responses to one’s finan­cial state on a con­tin­uum rang­ing from
over­whelm­ing finan­cial dis­tress to no finan­cial dis­tress
 Table 3. Psychosocial inter­ven­tions for HCT sur­vi­vors

Author Sample Study design Intervention Measures Findings

Depression/anx­i­ety
  Copeland n = 20, allo­ge­neic HCT, >18 years Single-arm pro­spec­tive Weekly col­lab­o­ra­tive care Hospital Anxiety and Anxiety decreased and depres­sion
et al47 of age pre-HCT, day 0, +14, meet­ings with HCT cli­ni­cians Depression Scale decreased (pre- to day +60).
+30, +60 and psy­chi­a­trist, review
case and pro­vide phar­ma­co­
logic rec­om­men­da­tions and
psy­cho­so­cial coun­sel­ing refer­rals
  Kim et al48 n = 40, mean age 59.2 years Randomized (inter­ven­tion Digital sto­ry­tell­ing—four 3-min­ute POMS sub­scales for depres­ Perceived social sup­port increased
or infor­ma­tion con­trol) per­sonal/emo­tion­ally rich stories sion and anx­i­ety, PROMIS for inter­ven­tion and decreased for
pre- and post assess­ment about post-HCT care Social Support scale con­trol. Anxiety and depres­sion
improved with time for both.
Coping/stress
  Balck et al49 n = 91, allo­ge­neic and autol­o­gous, Randomized (inter­ven­tion Problem-solv­ing train­ing HADS, SCL-9, Brief Cope, Anxiety, dis­tress, pain, stress were
>18 years of age or con­trol group) pre- Social Problem-Solving lower and active cop­ing bet­ter
HCT, day +11, +23 Inventory-R, Questions with inter­ven­tion. Better ­able to
of Pain, NCCN Distress reduce neg­a­tive ori­en­ta­tion and
Thermometer prob­lem-solve with inter­ven­tion.
  Somers n = 36, allo­ge­neic and autol­o­gous Randomized (inter­ven­tion Mobile pain-cop­ing skills: website Brief pain inven­tory, pain dis­ Intervention with improved pain, self-
et al50 (83%), mean age 56 years or usual treat­ment) with per­son­al­ized mes­sages abil­ity index, pain sub­scale effi­cacy, and on the 2MWT, and
pre- and post assess­ment and pain assess­ment activ­ity. of the chronic pain self-effi­ no improvement in the control
Materials (hand­outs, vid­eos, cacy scale, PROMIS Fatigue group. Changes in pain ­dis­abil­ity
audio files) about pain-cop­ing scale, FACT well-being and fatigue in both groups;
advice from patients and how to scale, 2MWT effect sizes were larger for the
apply pain-cop­ing skills inter­ven­tion group. No changes

Prakash Singh Shekhawat

in pain sever­ity in either group.
  Majhail n = 458, allo­ge­neic (48%) inter­ven­tion Randomized (inter­ven­tion Survivorship care plan CSI, CTXD, Knowledge of Intervention—lower dis­tress scores
et al39 arm, (44%) con­trol arm, median or con­trol) pre- and post Transplant Exposures, at 6 months and greater men­tal
age 59 years for both arms assess­ment (6 months) Health Care Utilization, SF-12 health scores
  Syrjala et al40 n = 755, allo­ge­neic and autol­o­gous Randomized (inter­ven­tion INSPIRE—tai­lored web page with CTXD, SCL-90-R Depression INSPIRE+ prob­lem-solv­ing train­ing
Syrjala et al51 (27%), mean age 51 years or inter­ven­tion + prob­lem- top­ics (1) lift mood, reduce fatigue, Scale, SF-36, FSI more likely to improve in dis­tress
Yi et al52 solv­ing train­ing or con­ boost health; (2) self-care tips and than con­trols. Male, age <40
trol—delayed access to tools; (3) tai­lored care guide­lines; years, Afri­can Amer­i­cans were
inter­ven­tion) pre- and (4) forum for post­ing expe­ri­ences less likely to enroll. Engagement
post assess­ment and for input; (5) resource list did not dif­fer by race, edu­ca­tion,
(6 months) prob­lem-solv­ing—focus on income, rural/urban res­i­dence,
prob­lems and goal set­ting with com­puter expe­ri­ence, donor
psy­chol­o­gists (>4 ses­sions) type, or pres­ence of depres­sion.
Cognitive/devel­op­men­tal
  Ferraro n = 110 Feasibility study of screen­ing MoCA, BACH Neurocognitive screen­ing is
et al53 at pre-HCT, day +100, 2 fea­si­ble to do prior to and post
years, 5 years post HCT trans­plant.
  Hoodin n = 9, allo­ge­neic HCT, mean age Ancillary study and Vorinostat + Tac + MTX for GVHD Cogstate, PHQ-9, GAD-7, Neurocognitive func­tion, depres­
et al54 53 years, con­trol cohorts— fea­si­bil­ity of assessing pro­phy­laxis FACT-General sion, anx­i­ety, quality of life did
con­cur­rent lon­gi­tu­di­nal study vorinostat, cog­ni­tive not dif­fer across time. Modeling—
func­tion, and QOL neurocognitive func­tion in
base­line, day 30, 100, 160 vorinostat patients com­pared to
auto con­trols was equiv­a­lent and

Psychosocial com­pli­ca­tions of HCT  |  573

bet­ter than allo con­trols.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/570/1851740/570buchbinder.pdf by guest on 13 December 2021

 Transplant center/ Paent-level Health care system
health care provider • Lack of informaon, • Lack of/inadequate
level knowledge, and skills reimbursement
• Socioeconomic and • Lack of resources and
• Lack of resources

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/570/1851740/570buchbinder.pdf by guest on 13 December 2021

cultural barriers specialists
• Lack of interest and
• Heterogeneity of late • Poor coordinaon of care
compeng priories
complicaons • Compeng priories
• Lack of awareness of
• Logiscal challenges • Inadequate quality
needs and screening and
prevenon guidelines • Different psychosocial assurance and
needs at different mes in accountability
• Lack of tools to address
transplant connuum
psychosocial concerns
• Inadequate
communicaon

Figure 1. Barriers to psychosocial care delivery.

There is a need for inter­ven­tions spe­cific to HCT sur­vi­vors that
are evi­dence based and lever­age novel plat­forms to improve
acces­si­bil­ity, with the goal of miti­gat­ing risks and improv­ing out­
comes fol­low­ing HCT. Table 3 sum­ma­rizes some of the inter­ven­
tions that have attempted to address the psy­cho­so­cial domains
in HCT sur­ vi­vors. These inter­ ven­tions were largely aimed at
depres­sion, anx­ie ­ ty, stress, and cop­ing, with only one inter­ven­
tion focused on cog­ni­tive dys­func­tion. To our knowl­edge, in the
last 5 years no inter­ven­tions have been published focus­ing on
finan­cial hard­ship or social health among HCT sur­vi­vors spe­cif­i­
cally, suggesting a concerning gap in the care of HCT sur­vi­vors
and their fam­i­lies. Most inter­ven­tion stud­ies have been by sin­gle
insti­tu­tions employing small sam­ple sizes to assess accept­abil­ity
as well as fea­si­bil­ity. Most stud­ies were not appro­pri­ately pow-
ered to assess effi­cacy and require future eval­u­a­tion within the
con­text of larger ran­dom­ized con­trolled tri­als.
Barriers to psy­cho­so­cial care deliv­ery and mod­els
to address bar­ri­ers
Many bar­ri­ers at dif­fer­ent lev­els need to be over­come to rec­
og­nize and ade­quately treat the psy­cho­so­cial con­cerns of HCT
sur­vi­vors in rou­tine clin­i­cal prac­tice (Figure 1). Even if we iden­
tify HCT sur­vi­vors in need of psy­cho­so­cial sup­port, pro­vid­ing in-
depth assess­ments with con­nec­tiv­ity to avail­­able resources in a
per­son­al­ized fash­ion and trans­lat­ing the research in psy­cho­so­cial
inter­ven­tions to actual clin­i­cal prac­tice remains a chal­lenge.
To over­come this chal­lenge, we need to develop and imple­
ment effec­tive mod­els for deliv­er­ing psy­cho­so­cial care. These
mod­els should be based on the 3-step approach to pre­ven­
tion of these com­pli­ca­tions. The first step or pri­mary pre­ven­
tion would include health pro­mo­tion activ­i­ties, includ­ing using
trans­ plant as a “teach­ able moment” for pro­ mot­ing health
behav­ior change, increas­ing aware­ness, and pre­emp­tively try­
ing to address risk fac­tors for these com­pli­ca­tions. Secondary
pre­ven­tion would include screen­ing that, as reviewed above,
should be com­pre­hen­sive, performed mul­ti­ple times post HCT,
and lead to action­able care path­ways pos­si­bly incor­po­rat­ing
the tested inter­ven­tions in rou­tine care. Tertiary pre­ven­tion tar-
geted at decreas­ing mor­bid­ity and mor­tal­ity from these com­
pli­ca­tions would include reha­bil­i­ta­tion strat­e­gies and dis­abil­ity
lim­i­ta­tion.
Implementing the above approach requires inte­gra­tion of
all­the rel­e­vant stake­hold­ers, such as the trans­plant team, pri­
mary care or oncol­ogy phy­si­cian, psy­chol­o­gist or psy­chi­a­trist,
social work­ ers, and finan­ cial coun­ sel­
ors (Table 4). With the
increased focus on vir­tual med­i­cine because of the COVID-19
pan­demic, there is an oppor­tu­nity to lever­age vir­tual tech­nol­
o­gies to deliver well-rounded, risk-strat­i­fied sur­vi­vor­ship care

574  |  Hematology 2021  |  ASH Education Program

Table 4. HCT psy­cho­so­cial pro­vid­ers and roles

Category Role
Transplant phy­si­cian Provide edu­ca­tion to patients and fam­i­lies regard­ing signs of psy­cho­so­cial dis­tress asso­ci­ated with HCT
Recognize signs of psy­cho­so­cial dis­tress and assess for psy­cho­so­cial dis­tress asso­ci­ated with HCT
Transplant nurse Provide edu­ca­tion to patients and fam­i­lies regard­ing signs of psy­cho­so­cial dis­tress asso­ci­ated with HCT
Recognize signs of psy­cho­so­cial dis­tress and assess for psy­cho­so­cial dis­tress asso­ci­ated with HCT
Social worker/case man­ager Educate, rec­og­nize, and assess for psy­cho­so­cial dis­tress.
Facilitate patient and fam­ily adjust­ment to the pro­cess of HCT and its sequelae.
Refer patients and fam­i­lies with dis­tress for spe­cial­ized ser­vices and com­mu­nity resources.
Address a range of psy­cho­so­cial and finan­cial needs (eg, insur­ance ben­e­fits, coor­di­na­tion of care, nav­i­gat­ing
health sys­tem, peer sup­port).
Psychologist Provide con­sul­ta­tion and man­age­ment of psy­cho­so­cial con­cerns iden­ti­fied among HCT patients and fam­i­lies

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/570/1851740/570buchbinder.pdf by guest on 13 December 2021

Psychiatrist
Chaplain, clergy, pas­to­ral care
Patient finan­cial ser­vices
Primary care phy­si­cian/pri­mary
hema­tol­o­gist/oncol­o­gist
using var­i­ous psy­cho­ther­a­peu­tic tech­niques.
Assist in the con­trol of symp­toms (eg, anx­i­ety, depres­sion) that may impact patient and fam­ily well-being.
Diagnose and treat psy­cho­logic and psy­chi­at­ric dis­or­ders that arise dur­ing HCT using psy­cho­phar­ma­co­logic
inter­ven­tions.
Assist in the con­trol of symp­toms (eg, anx­i­ety, depres­sion) that may impact patient and fam­ily well-being.
Assist patients and fam­i­lies with cop­ing, spir­i­tual coun­sel­ing.
Assist patient and fam­i­lies with prior autho­ri­za­tion, pay­ment plans to put together an appeal for deni­als to help
with costs of care.
Work closely with trans­plant team to screen and address psy­cho­so­cial issues and pro­vide sup­port.

incor­po­rat­ing psy­cho­so­cial screen­ing and inter­ven­tions. This

can help encour­ age health-related self-effi­ cacy in patients,
increase health sys­tem capac­ity, and pro­mote adap­tive HCT
sur­vi­vor­ship. Online pro­grams may help address some of the
bar­ri­ers in pro­vid­ing psy­cho­so­cial care such as dis­tance, vary­
ing needs, and lack of stan­dard fol­low-up care. Recently, the
INSPIRE (Inter­net-Based Survivorship Program with Information
and Resources) study found no dif­fer­ences between the study
arms in the pri­mary end point of aggre­gated out­comes of can­
cer and treat­ment dis­tress, depres­sive symp­toms, phys­i­cal
dys­func­tion, and fatigue, though there was sig­nif­i­cant improve­
ment in dis­tress alone for those in the INSPIRE+ prob­lem-solv­ing
treat­ment arm.40
CLINICAL CASES (Con­tin­ued)
In addi­tion to the man­age­ment of J. B.’s med­i­cal com­pli­ca­tions,
a mul­ti­dis­ci­plin­ary team that included her trans­plant phy­si­
cian, nurse, and social worker; patient finan­cial ser­vices; and
pro­vid­ers from pal­li­a­tive med­i­cine and psy­chi­a­try/psy­chol­ogy
came together to help her man­age her var­i­ous issues and pro­
vide coor­di­nated care. Frequent touch points with her using
vir­tual vis­its made her feel that her team was engaged in her
over­all care. Participating in a sup­port group for chronic GVHD
patients helped her feel emo­tion­ally supported.For the finan­
cial hard­ship, A. R.’s fam­ily was referred to a local char­i­ta­ble
orga­ni­za­tion that pro­vided grant funds for rent assis­tance. Due

Table 5. Recommended areas of focus for future research pri­or­i­ties and clin­i­cal prac­tice

Area
Study design and mea­sures
Screening and assess­ment
Rehabilitation and inter­ven­tion
Education and aware­ness
Focus
Proactively col­lect psy­cho­so­cial and finan­cial health data among HCT patients and fam­i­lies across the entire
tra­jec­tory of trans­plant care (lon­gi­tu­di­nal) using large, rep­re­sen­ta­tive sam­ples.
Increase focus on spe­cial pop­u­la­tions—chil­dren, ado­les­cents, and young adults, older adults, racial/eth­nic minor­ity
patients.
Use well-val­i­dated and stan­dard patient-reported out­come mea­sures.
Develop evi­dence-based rec­om­men­da­tions for uni­ver­sal screen­ing and assess­ment of HCT patients and fam­i­lies
at defined time points across the entire trans­plant tra­jec­tory using a stan­dard set of patient-reported out­come
mea­sures focused on psy­cho­so­cial and finan­cial dis­tress.
Leverage tech­nol­ogy to ensure robust con­nec­tiv­ity to HCT cen­ter and com­mu­nity-based resources.
Develop and test effec­tive reha­bil­i­ta­tion strat­e­gies, social sup­port (emo­tional, infor­ma­tional, and logis­ti­cal)
inter­ven­tions, and finan­cial sup­port inter­ven­tions.
Employ novel plat­forms for deliv­ery and devel­op­men­tally tai­lored for spe­cial pop­u­la­tions.
Raise aware­ness regard­ing psy­cho­so­cial and finan­cial health through var­i­ous means includ­ing edu­ca­tional efforts
directed at all­lev­els includ­ing, but not lim­ited to, patients and fam­i­lies as well as trans­plant care pro­vid­ers and
pol­icy-mak­ers.
Develop met­rics to exam­ine the effec­tive­ness of these mea­sures.
Prakash Singh Shekhawat
Psychosocial com­pli­ca­tions of HCT  |  575
to his cog­ni­tive and aca­demic chal­lenges, A. R. was referred for
for­mal neuropsychological eval­u­a­tion. Based upon the detailed
eval­ua
­ ­tion, the HCT clin­ic
­ al team and A. R.’s par­ents worked
closely with the child’s school, includ­ing his teach­ers and the
admin­is­tra­tion, to develop an appro­pri­ate, indi­vid­u­al­ized edu­
ca­tion plan for A. R. With addi­tional sup­port in the class­room,
A. R. was ­able to make mod­est gains in his aca­demic suc­cess.
Future direc­tions
Table 5 pro­vi­des a suggested out­line of research pri­or­i­ties and
areas of clin­i­cal focus when con­sid­er­ing poten­tial psy­cho­so­cial
risk and finan­cial dis­tress among HCT sur­vi­vors. With advances
in treat­ment prac­tices and sur­vival after HCT, there is an increas­
ing rec­og­ni­tion of the adverse impact of trans­plant on psy­cho­
so­cial aspects of sur­vi­vor­ship, includ­ing the need to proactively
col­lect and inter­vene in patient-reported out­comes after HCT.
Multiple ini­tia­tives in the field of HCT are begin­ning to iden­tify
the gaps in prac­tice and research and rec­og­nize the impor­tance
of inte­grat­ing patient-cen­tered out­come screen­ing and inter­
ven­tions with the goal of improv­ing long-term health out­comes
after HCT.2,4 The role of edu­ca­tional ini­tia­tives to empower and
engage patients and their care­giv­ers in improv­ing their psy­cho­
so­cial health is extremely valu­able. A mul­ti­pronged approach to
address psy­cho­so­cial con­cerns uti­liz­ing all­ rel­e­vant stake­hold­ers
can go a long way in ensur­ing the long-term suc­cess of this med­
i­cally com­pli­cated, resource-inten­sive pro­ce­dure.
Conflict-of-inter­est dis­clo­sure
David Buchbinder: no com­pet­ing finan­cial inter­ests to declare.
Nandita Khera: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
David Buchbinder: nothing to disclose.
Nandita Khera: nothing to disclose.
Correspondence
Nandita Khera, College of Medicine, Mayo Clinic in Arizona, 5777 E
Mayo Blvd, Phoenix, AZ 85054; e-mail: khera​­.nandita@mayo​­.edu.
References
1. Majhail NS, Tao L, Bredeson C, et  al. Prevalence of hema­to­poi­etic cell
trans­plant sur­vi­vors in the United States. Biol Blood Marrow Transplant.
2013;19(10):1498-1501.
2. Burns LJ, Abbetti B, Arnold SD, et al. Engaging patients in set­ting a patient-
cen­tered out­comes research agenda in hema­to­poi­etic cell trans­plan­ta­
tion. Biol Blood Marrow Transplant. 2018;24(6):1111-1118.
3. Hahn T, Paplham P, Austin-Ketch T, et al. Ascertainment of unmet needs and
par­tic­i­pa­tion in health main­te­nance and screen­ing of adult hema­to­poi­etic
cell trans­plan­ta­tion sur­vi­vors followed in a for­mal sur­vi­vor­ship pro­gram.
Biol Blood Marrow Transplant. 2017;23(11):1968-1973.
4. Bevans M, El-Jawahri A, Tierney DK, et  al. National Institutes of Health
hema­to­poi­etic cell trans­plan­ta­tion late effects ini­tia­tive: the patient-
cen­tered out­comes work­ing group report. Biol Blood Marrow Transplant.
2017;23(4):538-551.
5. Barata A, Gonzalez BD, Sutton SK, et al. Coping strat­e­gies mod­ify risk of
depres­sion asso­ci­ated with hema­to­poi­etic cell trans­plant symp­tom­atol­
ogy. J Health Psychol. 2018;23(8):1028-1037.
6. Seneviratne AK, Wright C, Lam W, Lipton JH, Michelis FV. Comorbidity pro­
file of adult sur­vi­vors at 20 years fol­low­ing allo­ge­neic hema­to­poi­etic cell
trans­plan­ta­tion. Eur J Haematol. 2021;106(2):241-249.
7. Liang J, Lee SJ, Storer BE, et al. Rates and risk fac­tors for post-trau­matic
stress dis­or­der symp­tom­atol­ogy among adult hema­to­poi­etic cell trans­
576  |  Hematology 2021  |  ASH Education Program
plant recip­i­ents and their infor­mal care­giv­ers. Biol Blood Marrow Trans-
plant. 2019;25(1):145-150.
8. Jacobs JM, Fishman S, Sommer R, et  al. Coping and mod­i­fi­able psy­cho­
so­cial fac­tors are asso­ci­ated with mood and qual­ ity of life in patients
with chronic graft-ver­ sus-host dis­ease. Biol Blood Marrow Transplant.
2019;25(11):2234-2242.
9. Kusaka K, Inoguchi H, Nakahara R, et  al. Stress and cop­ing strat­e­gies
among allo­ge­neic haematopoietic stem cell trans­plan­ta­tion sur­vi­vors: a
qual­i­ta­tive study. Eur J Cancer Care (Engl). 2020;29(6):e13307.
10. Sharafeldin N, Bosworth A, Patel SK, et  al. Cognitive func­ tion­
ing
after hema­to­poi­etic cell trans­plan­ta­tion for hema­to­logic malig­nancy:
results from a pro­spec­tive lon­gi­tu­di­nal study. J Clin Oncol. 2018;36(5):
463-475.
11. Abel GA, Albelda R, Khera N, et al. Financial hard­ship and patient-reported
out­comes after hema­to­poi­etic cell trans­plan­ta­tion. Biol Blood Marrow
Transplant. 2016;22(8):1504-1510.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/570/1851740/570buchbinder.pdf by guest on 13 December 2021
12. Hamilton JG, Wu LM, Austin JE, et al. Economic sur­vi­vor­ship stress is asso­ci­
ated with poor health-related qual­ity of life among distressed sur­vi­vors of
hema­to­poi­etic stem cell trans­plan­ta­tion. Psychooncology. 2013;22(4):911-
921.
13. Abel GA, Albelda R, Salas Coronado DY, et al. Prospective assess­ment of
famil­ial finan­cial hard­ship after hema­to­poi­etic cell trans­plan­ta­tion. Biol
Blood Marrow Transplant. 2015;21(suppl 2):S72.
14. Jeon M, Yoo IY, Kim S, Lee J. Post-trau­matic growth in sur­vi­vors of allo­ge­neic
hema­to­poi­etic stem cell trans­plan­ta­tion. Psychooncology. 2015;24(8):871-
877.
15. Denzen EM, Majhail NS, Stickney Ferguson S, et  al. Hematopoietic cell
trans­ plan­ ta­
tion in 2020: sum­ mary of year 2 rec­ om­ men­ da­ tions of the
national mar­row donor pro­gram’s sys­tem capac­ity ini­tia­tive. Biol Blood
Marrow Transplant. 2013;19(1):4-11.
16. Khera N, Hamilton BK, Pidala JA, et al. Employment, insur­ance, and finan­
cial expe­ri­ences of patients with chronic graft-ver­sus-host dis­ease in North
America. Biol Blood Marrow Transplant. 2019;25(3):599-605.
17. Khera N, Albelda R, Hahn T, et al. Financial hard­ship after hema­to­poi­etic
cell trans­plan­ta­tion: lack of impact on sur­vival. Cancer Epidemiol Biomark-
ers Prev. 2018;27(3):345-347.
18. Denzen EM, Thao V, Hahn T, et al. Financial impact of allo­ge­neic hema­to­
poi­etic cell trans­plan­ta­tion on patients and fam­i­lies over 2 years: results
from a mul­ti­cen­ter pilot study. Bone Marrow Transplant. 2016;51(9):1233-
1240.
19. Eriksson L, Wennman-Larsen A, Bergkvist K, Ljungman P, Winterling J.
Important fac­tors asso­ci­ated with sick leave after allo­ge­neic haematopoi-
etic stem cell trans­plan­ta­tion—a 1-year pro­spec­tive study. J Cancer Surviv.
Published online 8 Jan­ua ­ ry 2021.
20. Lee SJ, Logan B, Westervelt P, et al. Comparison of patient-reported out­
comes in 5-year sur­vi­vors who received bone mar­row vs periph­eral blood
unre­lated donor trans­plan­ta­tion: long-term fol­low-up of a ran­dom­ized clin­
i­cal trial. JAMA Oncol. 2016;2(12):1583-1589.
21. Kirchhoff AC, Leisenring W, Syrjala KL. Prospective pre­dic­tors of return to
work in the 5 years after hema­to­poi­etic cell trans­plan­ta­tion. J Cancer Sur-
viv. 2010;4(1):33-44.
22. Winterling J, Johansson E, Wennman-Larsen A, Petersson LM, Ljungman P,
Alexanderson K. Occupational sta­tus among adult sur­vi­vors fol­low­ing allo-
SCT. Bone Marrow Transplant. 2014;49(6):836-842.
23. Salit RB, Lee SJ, Burns LJ, et  al. Return-to-work guide­lines and pro­grams
for post-hema­to­poi­etic cell trans­plan­ta­tion sur­vi­vors: an ini­tial sur­vey. Biol
Blood Marrow Transplant. 2020;26(8):1520-1526.
24. Park J, Wehrlen L, Mitchell SA, Yang L, Bevans MF. Fatigue pre­dicts impaired
social adjust­ment in sur­vi­vors of allo­ge­neic hema­to­poi­etic cell trans­plan­ta­
tion (HCT). Support Care Cancer. 2019;27(4):1355-1363.
25. Nenova M, DuHamel K, Zemon V, Rini C, Redd WH. Posttraumatic growth,
social sup­port, and social con­straint in hema­to­poi­etic stem cell trans­plant
sur­vi­vors. Psychooncology. 2013;22(1):195-202.
26. Yen HJ, Eissa HM, Bhatt NS, et al. Patient-reported out­comes in sur­vi­vors of
child­hood hema­to­logic malig­nan­cies with hema­to­poi­etic stem cell trans­
plant. Blood. 2020;135(21):1847-1858.
27. Wilhelmsson M, Glosli H, Ifversen M, et al; Nor­dic Society of Pediatric
Hematology and Oncology. Long-term health out­comes in sur­vi­vors of
child­hood AML treated with allo­ge­neic HSCT: a NOPHO-AML study. Bone
Marrow Transplant. 2019;54(5):726-736.
28. Luo Y, Yang P, Yang Y, et al. Cognitive and psy­cho­log­i­cal out­comes of pedi­
at­ric allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion sur­vi­vors in a sin­
gle cen­ter in China. Medicine (Baltimore). 2019;98(39):e17307.
29. Brauer E, Pieters HC, Ganz PA, Landier W, Pavlish C, Heilemann MV. Com-
ing of age with can­cer: phys­i­cal, social, and finan­cial bar­ri­ers to inde­pen­
dence among emerg­ing adult sur­vi­vors. Oncol Nurs Forum. 2018;45(2):
148-158.
30. Tremolada M, Bonichini S, Basso G, Pillon M. Perceived social sup­port and
health-related qual­ity of life in AYA can­cer sur­vi­vors and con­trols. Psy-
chooncology. 2016;25(12):1408-1417.
31. Pulewka K, Wolff D, Herzberg PY, et al. Physical and psy­cho­so­cial aspects
of ado­les­cent and young adults after allo­ge­neic hema­to­poi­etic stem-cell
trans­plan­ta­tion: results from a pro­spec­tive mul­ti­cen­ter trial. J Cancer Res
Clin Oncol. 2017;143(8):1613-1619.
32. Walsh CA, Yi JC, Rosenberg AR, Crouch MV, Leisenring WM, Syrjala KL.
Factors asso­ci­ated with social func­tion­ing among long-term can­cer sur­vi­
vors treated with hema­to­poi­etic stem cell trans­plan­ta­tion as ado­les­cents
or young adults. Psychooncology. 2020;29(10):1579-1586.
33. Brauer ER, Pieters HC, Ganz PA, Landier W, Pavlish C, Heilemann MV. “From
snail mode to rocket ship mode”: ado­les­cents and young adults’ expe­ri­
ences of returning to work and school after hema­to­poi­etic cell trans­plan­
ta­tion. J Adolesc Young Adult Oncol. 2017;6(4):551-559.
34. Bhatt NS, Brazauskas R, Tecca HR, et al. Post-trans­plan­ta­tion employ­ment
sta­tus of adult sur­vi­vors of child­hood allo­ge­neic hema­to­poi­etic cell trans­
plant: a report from the Center for International Blood and Marrow Trans-
plant Research (CIBMTR). Cancer. 2019;125(1):144-152.
35. Pulewka K, Strauss B, Hochhaus A, Hilgendorf I. Clinical, social, and psy­
cho-onco­log­i­cal needs of ado­les­cents and young adults (AYA) ver­sus older
patients fol­low­ing hema­to­poi­etic stem cell trans­plan­ta­tion. J Cancer Res
Clin Oncol. 2021;147(4):1239-1246.
36. Majhail NS, Rizzo JD, Lee SJ, et al; Center for International Blood and Marrow
Transplant Research; Amer­i­can Society for Blood and Marrow Transplanta-
tion; Euro­pean Group for Blood and Marrow Transplantation; Asia-Pacific
Blood and Marrow Transplantation Group; Bone Marrow Transplant Society
of Australia and New Zealand; East Med­i­ter­ra­nean Blood and Marrow Trans-
plantation Group; Sociedade Brasileira de Transplante de Medula Ossea.
Recommended screen­ing and pre­ven­tive prac­tices for long-term sur­vi­
vors after hema­to­poi­etic cell trans­plan­ta­tion. Bone Marrow Transplant.
2012;47(3):337-341.
37. Khera N, Holland JC, Griffin JM. Setting the stage for uni­ver­sal finan­cial
dis­ tress screen­ ing in rou­ tine can­
cer care. Cancer. 2017;123(21):4092-
4096.
38. Denzen EM, Preussler JM, Mur­phy EA, et  al. Tailoring a sur­vi­vor­ship care
plan: patient and pro­vider pref­er­ences for recip­i­ents of hema­to­poi­etic cell
trans­plan­ta­tion. Biol Blood Marrow Transplant. 2019;25(3):562-569.
39. Majhail NS, Mur­phy E, Laud P, et al. Randomized con­trolled trial of indi­vid­
u­al­ized treat­ment sum­mary and sur­vi­vor­ship care plans for hema­to­poi­etic
cell trans­plan­ta­tion sur­vi­vors. Haematologica. 2019;104(5):1084-1092.
40. Syrjala KL, Yi JC, Artherholt SB, et al. An online ran­dom­ized con­trolled trial,
with or with­out prob­lem-solv­ing treat­ment, for long-term can­cer sur­vi­vors
after hema­to­poi­etic cell trans­plan­ta­tion. J Cancer Surviv. 2018;12(4):560-
570.
41. Hong S, Rybicki L, Corrigan D, et  al. Psychosocial Assessment of Candi-
dates for Transplant (PACT) as a tool for psy­cho­log­i­cal and social eval­u­
a­tion of allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion recip­i­ents. Bone
Marrow Transplant. 2019;54(9):1443-1452.
Prakash Singh Shekhawat
42. Solh MM, Speckhart D, Sol­o­mon SR, et al. The Transplant Evaluation Rat-
ing Scale pre­dicts over­all sur­vival after allo­ge­neic hema­to­poi­etic stem cell
trans­plan­ta­tion. Blood Adv. 2020;4(19):4812-4821.
43. Pai ALH, Swain AM, Chen FF, et al. Screening for fam­ily psy­cho­so­cial risk in
pedi­at­ric hema­to­poi­etic stem cell trans­plan­ta­tion with the psy­cho­so­cial
assess­ment tool. Biol Blood Marrow Transplant. 2019;25(7):1374-1381.
44. Mishkin AD, Shapiro PA, Reshef R, Lopez-Pintado S, Mapara MY. Standard-
ized semi-struc­tured psy­cho­so­cial eval­u­a­tion before hema­to­poi­etic stem
cell trans­plan­ta­tion pre­dicts patient adher­ence to post-trans­plant reg­i­­
men. Biol Blood Marrow Transplant. 2019;25(11):2222-2227.
45. de Souza JA, Yap BJ, Wroblewski K, et al. Measuring finan­cial tox­ic­ity as a
clin­i­cally rel­e­vant patient-reported out­come: The val­i­da­tion of the COm-
prehensive Score for Financial Toxicity (COST). Cancer. 2017;123(3):476-
484.
46. Prawitz AD, Garman TE, Sorhaindo B, O’Neill B, Kim JI, Drentea P. InCharge
finan­cial dis­tress/finan­cial well-being scale: devel­op­ment, admin­is­tra­tion,
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/570/1851740/570buchbinder.pdf by guest on 13 December 2021
and score inter­pre­ta­tion. Financial Couns Plan. 2006;17(1):34-49.
47. Copeland AC, Tan X, Nash RP, et al. Collaborative care for depres­sion and
anx­i­ety in the bone mar­row trans­plant pop­u­la­tion: a pilot fea­si­bil­ity study.
Psychooncology. 2021;30(1):118-122.
48. Kim WS, Langer S, Todd M, et al. Feasibility of a dig­i­tal sto­ry­tell­ing inter­ven­
tion for hema­to­poi­etic cell trans­plant patients. J Cancer Educ. Published
online 3 Jan­u­ary 2021.
49. Balck F, Zschieschang A, Zimmermann A, Ordemann R. A ran­dom­ized con­
trolled trial of prob­lem-solv­ing train­ing (PST) for hema­to­poi­etic stem cell
trans­plant (HSCT) patients: effects on anx­i­ety, depres­sion, dis­tress, cop­ing
and pain. J Psychosoc Oncol. 2019;37(5):541-556.
50. Somers TJ, Kelleher SA, Dorfman CS, et al. An mHealth pain cop­ing skills
train­ing inter­ven­tion for hema­to­poi­etic stem cell trans­plan­ta­tion patients:
devel­op­ment and pilot ran­dom­ized con­trolled trial. JMIR Mhealth Uhealth.
2018;6(3):e66.
51. Syrjala KL, Crouch ML, Leisenring WM, et al. Engagement with INSPIRE, an
online pro­gram for hema­to­poi­etic cell trans­plan­ta­tion sur­vi­vors. Biol Blood
Marrow Transplant. 2018;24(8):1692-1698.
52. Yi JC, Sullivan B, Leisenring WM, et al. Who enrolls in an online can­cer sur­
vi­vor­ship pro­gram? Reach of the INSPIRE ran­dom­ized con­trolled trial for
hema­to­poi­etic cell trans­plan­ta­tion sur­vi­vors. Biol Blood Marrow Trans-
plant. 2020;26(10):1948-1954.
53. Ferraro CS, Williams C, Dixon B, Bernhard L, Majhail NS, Hamilton BK.
Neurocognitive and phys­i­cal func­tion­ing screen­ing in hema­to­poi­etic
cell trans­plant (HCT) sur­vi­vor­ship clinic. Biol Blood Marrow Transplant.
2020;26(suppl 3):S365.
54. Hoodin F, LaLonde L, Errickson J, et al. Cognitive func­tion and qual­ity of
life in vorinostat-treated patients after matched unre­lated donor myeloab-
lative con­di­tion­ing hema­to­poi­etic cell trans­plan­ta­tion. Biol Blood Marrow
Transplant. 2019;25(2):343-353.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000292
Psychosocial com­pli­ca­tions of HCT  |  577
 SURVIVORSHIP AFTER ALLOGENEIC HEMATOPOIETIC CELL TRANSPLANT

Noninfectious complications of hematopoietic

cell transplantation

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/578/1851735/578williams.pdf by guest on 13 December 2021

Kirsten M. Williams
Blood and Marrow Transplant Program, Aflac Cancer and Blood Disorders Center, Emory University School of Medicine, Children’s Healthcare of
Atlanta, Atlanta, GA

Noninfectious lung diseases contribute to nonrelapse mortality. They constitute a spectrum of diseases that can affect
the parenchyma, airways, or vascular pulmonary components and specifically exclude cardiac and renal causes. The dif-
ferential diagnoses of these entities differ as a function of time after hematopoietic cell transplantation. Specific diagno-
sis, prognosis, and optimal treatment remain challenging, although progress has been made in recent decades.

LEARNING OBJECTIVES
• Describe the different pulmonary toxicities that occur after HCT and their timing post HCT as well as risk factors
and recommended treatments
• Describe the workup of lung toxicity and the current challenges to diagnosis and treatment of noninfectious lung
injury after HCT

Introduction Pulmonary differential of Case 1

Noninfectious lung complications after hematopoietic cell
transplantation (HCT) are uncommon but associated with
significant morbidity and mortality.1 Lung compromise
may be due to intrapulmonary injury to the parenchyma,
airways, or vascular structures and excludes that due to
extrapulmonary, renal, or cardiac causes. Identifying the
correct diagnosis is critical but can be challenging with
available tools, though timing after HCT narrows the differ­
ential. This review thus presents the differential diagnoses
and key findings as a function of time post transplant.
CLINICAL CASE 1
A 9­month­old, who was transplanted for a nonmalignant
condition with a mismatched cord­blood donor after a
reduced­intensity busulfan­based preparative regimen and
had recently engrafted, developed acute hypoxic respira­
tory distress 28 days after HCT, in the setting of sinusoidal
obstructive syndrome treated with defibrotide. Fresh blood
was noted on bronchoalveolar lavage (BAL) that did not
clear, as well as a new infiltrate on chest x­ray and rapid
progression to respiratory failure. An infectious and cardiac
workup was not revealing. Sadly, despite attempting high­
frequency oscillatory ventilation, this patient succumbed.
The differential diagnosis of noninfectious lung injury early
after HCT largely reflects alveolar injury, many mediated by
endothelial dysfunction. These early lung injuries include
peri­engraftment respiratory distress syndrome (PERDS),
capillary leak syndrome (CLS), diffuse alveolar hemorrhage
(DAH), acute interstitial pneumonitis (AIP), or toxic pneu­
monitis/injuries, collectively termed idiopathic pneumonia
syndrome (IPS). Additionally, vascular injury, such as pulmo­
nary veno­occlusive disease (PVOD), can lead to respira­
tory decline early after HCT (Table 1 and Figure 1). These
diagnoses are associated with respiratory distress, include
radiographic abnormalities, and although pulmonary func­
tion tests (PFTs) are rarely performed, would demonstrate
a restrictive defect. The diagnosis of Case 1 is DAH.
Diffuse alveolar hemorrhage
Hallmarks of this diagnosis include hypoxia, respiratory
distress, and new radiographic infiltrates in the perihilar
and lower lobes, with the diagnosis being confirmed by
bloody BAL that does not clear, along with hemosiderin­
laden macrophages. The incidence is 2% to 5% in alloge­
neic recipients and 1% in autologous. This patient had risk
factors for DAH, including status as a cord­blood recipi­
ent, defibrotide treatment, delayed engraftment, though
not sirolimus prophylaxis, total body irradiation (TBI), and
myeloablation.2,3 Patients often present <30 days after HCT

578 | Hematology 2021 | ASH Education Program

Table 1. Diagnoses, key points, and treat­ments of non­in­fec­tious lung injury after HCT

Peak time post HCT/

Diagnosis struc­ture Key find­ings Additional diag­no­sis Proposed treat­ments
DAH <day 30 post HCT Hypoxia, new infil­trate, Hemosiderin-laden Inhaled transexamic acid,
endo­the­lial bloody BAL mac­ro­phages intrapulmonary
recom­bi­nant fac­tor VIIa
PERDS (IPS) Pre-engraft­ment Respiratory dis­tress/ Multilobar infil­trates Steroids/etanercept
Alveolar (from endo­the­lial hyp­oxia
dam­age)
CLS (IPS) Engraftment to day 15 Dyspnea/hyp­oxia, >3% Multilobar infil­trates Steroids/etanercept
Alveolar (from endo­the­lial weight gain
dam­age)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/578/1851735/578williams.pdf by guest on 13 December 2021

AIP and toxin-related lung 50 days post HCT Fever, dyspnea, cough Ground glass on imag­ing/​
injuries, DPTC Alveolar bilat­eral infil­trates,
restric­tive PFT
IPS (PERDS, CLS, DAH, 45 days post HCT Hypoxia, pul­mo­nary Steroids/etanercept
AIP, DPTC) Alveolar infil­trates
CRS <7 days postcellular Respiratory com­pro­mise, Infiltrates Tociluzumab
ther­apy hyp­oxia
Alveolar
PVOD, PCT, TA-TMA 15-120 days post HCT Hypoxia, dyspnea, vas­cu­lar Sildenafil (PVOD, TA-TMA),
Endothelial occlu­sion, may prog­ress nitric oxide (TA-TMA) pros­
to pul­mo­nary hyper­ten­sion ta­cy­clins, cal­cium chan­nel
Biopsy diag­no­sis block­ers (PVOD) Steroids
(PCT)?
RLD after HCT Day 100-1 year Decreased FEV1, nor­mal PPFE, NSIP inter­sti­tial Etanercept
Alveolar FEV1/VC ratio, fibro­sis on pneu­mo­nia; can include Poor response to ste­roids
CT (upper lobes) other diag­noses
BOS Day 100-1.5 years FEV1 < 75%, ≥10% decline, Newer modal­i­ties: PRM and FAM: inhaled fluticasone,
Airway FEV1/VC LL of the 90% xenon-129 MRI azithromycin, montelukast
CI, absence of infec­tion +1 mg/kg/d pred­ni­sone
and either preexisting with rapid taper; ECP,
cGVHD, air trap­ping by etanercept, GERD tx, nutri­
expi­ra­tory CT or by PFT, or tion, infec­tion pro­phy­laxis
cir­cum­fer­en­tial fibro­sis of
bron­chi­­oles on biopsy
Non-HCT-spe­cific Alveolar (ARDS, PTLD) TRALI/TACO: tem­po­ral A1AT: obstruc­tive dis­ease, TRALI/TACO: sup­port­ive care
com­pli­ca­tions: Airway (A1AT) asso­ci­a­tion blood emphy­sema, PE: anti­co­ag­u­lants
TRALI/TACO, Endothelial (TRALI/ prod­ucts, fever, acute bron­chi­ec­ta­sis ARDS: treat under­ly­ing cause
PE, ARDS, A1AT, TACO, PE) dyspnea PTLD: nod­ules/EBV+ (often anti­mi­cro­bial) and
can­cer/PTLD, PE: dyspnea/hyp­oxia, V/Q+ sup­port­ive care
pneu­mo­tho­rax ARDS: fever, dyspnea, A1AT: A1AT infu­sion, inhaled
hyp­oxia, bilat­eral infil­trates ste­roids
PTLD: rituximab, cel­lu­lar
ther­apy
Pneumothorax: chest tube
EBV, Epstein-Barr virus; ECP, extracorporeal photopheresis; FAM, inhaled fluticasone, azithromycin, and montelukast + 1 mg/kg/d steroid burst and
rapid taper; GERD tx, gastro-esophageal reflux disease treatment; PRM; para­met­ric response map­ping; V/Q , ven­ti­la­tion/per­fu­sion scan.

and do poorly, with <25% of adults and <16% of chil­dren sur­viv­
ing a year.4,5 While the path­o­gen­e­sis is incom­pletely under­stood,
DAH is thought to reflect accu­mu­lated dam­age from con­di­tion­
ing reg­i­mens, aber­rant immune response, and sub­se­quent dam­
age to the endo­the­lium, resulting in hem­or­rhage into alve­o­lar
spaces. DAH can occur as a result of infec­tions, and thus a full
infec­tious workup is advised (Table 2).6 Treatment his­tor­i­cally
included ste­roids, though mul­ti­ple stud­ies have dem­on­strated
a lack of ben­e­fit, and some have shown harm.6-8 Similarly, amino­
caproic acid and sys­temic recom­bi­nant fac­tor VIIa have not ben­
efited DAH.8,9 In con­trast, small series have supported the use
of intrapulmonary admin­is­tra­tion of recom­bi­nant fac­tor VIIa and
Prakash Singh Shekhawat
inhaled transexamic acid. A small case series showed improved
sur­
vival in patients receiv­ ing intrabronchi recom­ bi­
nant fac­
tor
VIIa com­pared with his­tor­i­cal patients who received ste­roids,10,11
which is con­sis­tent with my expe­ri­ence. For patients too sick
to inter­rupt the ven­ti­la­tory cir­cuit, inhaled transexamic acid, an
antifibrinolytic agent, can be used with con­ven­tional ven­ti­la­tion.
We have used this in sev­eral DAH patients with improve­ment
(with­out ste­roids), based on published ret­ro­spec­tive data.12 Fi­
nally, extra­cor­po­real mem­brane oxy­gen­a­tion has been rarely
employed.13 DAH is con­sis­tently diag­nosed, while its eti­­ol­ogy
and opti­mal treat­ment are unknown, though local ther­a­pies may
offer the greatest ben­e­fit, and high-dose ste­roid use has waned.
Noninfectious pul­mo­nary tox­ic­ity after HCT  |  579

Figure 1. Illustrated HCT noninfectious lung complications in
terms of structure and peak timing after HCT. The time after
HCT is shown on the x-axis (in months) with the diseases. Light
blue bubbles, parenchymal processes (CRS; PERDS; CLS, cap­
illary leak syndrome from endothelial damage; AIP; rP; DPTS;
RLD). Red vessels, the endothelial processes (DAH; PVOD; PCT;
TA-TMA). Green airway, airway disease (BOS). IPS, which is most
often defined early post HCT, includes these diagnoses (defined
by hypoxia, infiltrates, alveolar injury, and excludes later diagno­
ses that often do not present with these features). White box,
pulmonary noninfectious diagnoses not specific to HCT (TRALI;
TACO; COP; PE; PTLD), which can occur throughout the HCT tra­
jectory. Above the table are key events occurring during HCT
aGVHD and cGVHD.
AIP and toxin-related lung injury
Pneumonitis, an inflam­ ma­ tion of the alveoli, can occur early
after HCT due to infec­tions, the pre­par­a­tive reg­i­men, or be idi­o­
pathic. The pre­sen­ta­tion is non­spe­cific, includ­ing fever, dyspnea,
cough, and ground-glass opac­ i­
ties on imag­ ing. These toxin-
­induced injuries have had sev­eral terms linked to them, includ­ing
AIP, radi­a­tion pneu­mo­ni­tis, carmustine pneu­mo­ni­tis, or delayed
pul­mo­nary tox­ic­ity syn­drome (DPTS from che­mo­ther­apy). Over­
all, the inci­dence of radi­a­tion and che­mo­ther­apy pneu­mo­ni­tis has
decreased with mod­i­fi­ca­tions to che­mo­ther­apy reg­i­mens and
the advent of lung blocking and frac­tion­ated TBI.14,15 The recent
over­all inci­dence of non­in­fec­tious pneu­mo­ni­tis was 5% one year
after HCT, with a median onset of 1.6 months in chil­dren.5 Risk
fac­tors included can­cer indi­ca­tion and myeloablative con­di­tion­
ing.5 There was also an asso­ci­a­tion with cord blood trans­plan­ta­
tion, which may reflect the increased use in TBI in this cohort,
though it was not discussed. After myeloablative TBI in adults,
the inci­dence of non­in­fec­tious radi­a­tion-induced IP is 7.4%, with
a median onset of 2.5 months, and less after che­mo­ther­apy con­
di­tion­ing.14 The over­all sur­vival for adults and chil­dren is 50%
and 40%, respec­tively.5,14 Reducing lung expo­sure dur­ing TBI has
­decreased the rates of radi­a­tion-induced IP fur­ther.16 Other toxin-
medi­ated early lung com­pli­ca­tions can include cyclo­phos­pha­
mide-induced, eosin­o­philic pneu­mo­nia linked to fludarabine; IP
or DAH with rituximab; DAH after alemtuzumab; or pneu­mo­ni­tis
after check­point inhib­i­tors.15,17 While ste­roids are often cited for
treat­ment of these pneu­mo­ni­tis and toxin-induced injuries, data
sug­gest that they have fallen out of favor, given the sparce data
580  |  Hematology 2021  |  ASH Education Program
for ben­e­fit and the rel­at­ive risks of infec­tion and can­cer recur­
rence. Collectively, these AIP- or toxin-induced lung injuries are
uncom­mon com­pli­ca­tions of HCT that would ben­e­fit from spe­
cific diag­nos­tic cri­te­ria and inves­ti­ga­tion into ste­roid-spar­ing
ther­a­peu­tic approaches.
PERDS and CLS
PERDS and CLS lead to alve­ o­lar dam­ age through endo­ the­
lial
injury, with pul­mo­nary edema as a down­stream event. The man­
i­fes­ta­tions include respi­ra­tory dis­tress and hyp­ox­emia, the pres­
ence of multilobar infil­trates by imag­ing, and the absence of infec­
tions, with PERDS typ­i­cally occur­ring within 5 days of engraft­ment,
and some suggesting even a week ear­lier, and CLS occur­ring in
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/578/1851735/578williams.pdf by guest on 13 December 2021
the first 15 days and accom­pa­nied by fluid over­load (>3% weight
gain). Recent data sug­gest PERDS occurs in n ­ early 5% of autol­
o­gous HCTs and 7% of allo­ge­neic HCTs.18,19 ­Several ele­va­tions in
sys­temic cyto­kines have been asso­ci­ated with engraft­ment syn­
drome, includ­ing ST2 protein (an IL-33 receptor), IL-2Rα, tumor
­necro­sis fac­tor 1 between base­line and day 14, and procalcitonin
at day 0.19,20 Engraftment syn­drome is asso­ci­ated with increased
mor­tal­ity after HCT, though PERDS was not ana­lyzed sep­a­rately.19
A sim­i­lar inci­dence is reported for CLS, which affected approx­i­
ma­tely 5% of allo­ge­neic HCT patients in a case series, though all­
but 0.6% of these cases were due to sep­sis and asso­ci­ated with
mor­tal­ity.21 A risk fac­tor for CLS may be preexisting vas­cu­lar dys­
regulation and cord blood trans­plant.22 PERDS and non­in­fec­tious
CLS are rarer causes of lung dys­func­tion early after HCT, char­ac­ter­
ized by endo­the­lial dys­func­tion and cyto­kine abnor­mal­i­ties, with
increased mor­tal­ity.
IPS—a uni­fy­ing diag­no­sis
IPS is an umbrella diag­no­sis that has his­tor­i­cally incor­po­rated
sev­eral of the above diag­noses. Most def­i­ni­tions have included
PERDS, CLS, DAH, AIP, and DPTC in this ter­mi­nol­ogy, with diag­
nos­tic cri­te­ria of non­in­fec­tious, noncardiogenic, nonrenal pul­mo­
nary injury after HCT with asso­ci­ated pul­mo­nary infil­trates and
hyp­ox­emia due to alve­o­lar injury.23-26 The term is an impor­tant
uni­fy­ing diag­no­sis, per­mit­ting key clin­i­cal tri­als of rare diag­noses
that col­lec­tively com­prise early non­in­fec­tious lung injury. The
inci­dence of IPS is 3.7% of allo­ge­neic adult HCT patients, with
a median onset of 43 days, though later onset is described, and
risk fac­tors of myeloablative con­di­tion­ing, high-dose TBI, lower-
respi­ra­tory tract viruses pre-HCT, and inborn errors of metab­o­
lism as indi­ca­tion for HCT.25,27 In chil­dren, the most recent data
(2014) showed a rate of 6.7% in allo­ge­neic recip­i­ents.28 Approxi­
mately 5% of autol­o­gous trans­plant recip­i­ents incur IPS.29 Nota­
bly, this rate is lower than the rate of the com­pos­ite diag­noses
com­bined and may reflect the het­ero­ge­ne­ity of dis­or­ders and
patients included under the IPS umbrella. Outcomes remain
poor after allo­ge­neic HCT, with <16% to 30% of chil­dren and
adults sur­viv­ing a year.25,28,30,31 Many stud­ies have linked ele­vated
BAL lev­els of tumor necro­sis growth fac­tor, ST2, and IL-6 to IPS
at diag­no­sis.26,32-34 Some stud­ies, though not all­, have shown a
ben­e­fit from etanercept with cor­ti­co­ste­roids, although less of
an impact on long-term sur­vival.26,31,35 The risk of death was high­
er in patients with prior pul­mo­nary impair­ment, those who re­
quired ven­ti­la­tory sup­port, or those who had con­cur­rent renal
or hepatic injury.25 IPS is a ben­e­fi­cial umbrella diag­no­sis of early
life-threat­en­ing non­in­fec­tious pul­mo­nary injury after HCT with­
out con­sen­sus on its opti­mal treat­ment.
Table 2. Workup con­sid­er­ations of non­in­fec­tious lung injury after HCT

Type Test Considerations

Radiographic CT-inspi­ra­tory Noncontrast suf­fi­cient for most; bron­chi­ec­ta­sis, enlarged
pul­mo­nary ves­sels may be seen in some diag­noses
CT-added expi­ra­tory Valuable to assess air trap­ping (BOS or RLD)
MRI/MRA May iden­tify vas­cu­lar dis­ease
V/Q scan Evaluate for PE
Chest x-ray Less valu­able if air trap­ping is pres­ent
BAL PCP (PJP) PCR PCR tests are increas­ingly valu­able
Explify PCR Explify can detect mul­ti­ple path­o­gens by NGS and may
Respiratory virus PCR detect organ­isms at low lev­els not evi­dent in cul­ture,
Gram stain/cul­ture which may be most valu­able for pretreated patients or
Fungal stain/cul­ture dis­tal lesions

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/578/1851735/578williams.pdf by guest on 13 December 2021

Cytology (sil­ver stain) Blood return that does not clear is diag­nos­tic for DAH
Nocardia cul­ture Visual inspec­tion can iden­tify fun­gal lesions or other
AFB smear/cul­ture indi­ca­tors
CMV PCR Because BAL fluid quan­tity is per kilo­gram and return is
HSV PCR dimin­ished in severe lung injury, it may be impor­tant to
Mycoplasma pneumoniae PCR pri­or­i­tize tests in chil­dren
Legionella pneumophilia cul­ture
Galactomannan
Blood Blood cul­ture Karius uses NGS and can detect lung organ­isms with good
Karius sen­si­tiv­ity, though data sug­gest that some pul­mo­nary
Blood gas infec­tions may be missed (eg, fun­gal nod­ules)
Consider met­he­mo­glo­bin if hyp­oxic and med­i­ca­tions at risk
Urine Histoplasmosis anti­gen
PFTs FEV1, FEV1/VC ratio, RV and RV/TLC, DLCO, Lung clear­ance index may be valu­able 6 min­utes’ walk time
+/−albuterol is a mea­sure of endur­ance for BOS
FEV1 slope of decline can be valu­able for prog­no­sis in BOS
Lung biopsy Pathology, stains, and cul­tures High mor­bid­ity should prompt care­ful con­sid­er­ation of
risk/ben­e­fit anal­y­sis
Echocardiogram/+/−angi­og­ra­phy Pulmonary hyper­ten­sion May also iden­tify peri­car­dial effu­sions
AFB, acid fast bacil­lus; CMV, cyto­meg­a­lo­vi­rus; DLCO, dif­fu­sion capac­ity of lung for car­bon mon­ox­ide; HSV, her­pes sim­plex virus; MRA, mag­netic ­
res­o­nance angi­og­ra­phy; PJP/PCP, Pneumocystis jirovecii pneu­mo­nia (for­merly Pneumocystis carinii pneu­mo­nia); TLC, total lung capac­ity.

Cytokine release syn­drome
A new diag­no­sis has emerged related to the admin­is­tra­tion of
genet­i­cally mod­i­fied T cells (chi­me­ric anti­gen recep­tor T-cell
ther­apy; CAR-T). Cytokine release syn­drome (CRS) can include
respi­ra­tory com­pro­mise, with infil­trates, hyp­oxia, and endo­the­
lial dys­func­tion that closely mim­ics CLS.36-38 A large inter­na­tional
study showed that 25% of patients with CAR-T had sig­nif­i­cant
pul­mo­nary com­pro­mise.39 Risk fac­tors for CRS include high dis­
ease bur­den, con­cur­rent infec­tions, and lymphodepletion prior
to CAR-T.37,38 The role of IL-6 in this pro­cess has been established
and block­ade (via tocilizumab) has miti­gated tox­ic­ity, some­times
with ste­roid coad­min­is­tra­tion.37,38 These data beg the ques­tion of
whether IL-6 block­ade could be valu­able in IPS as well, though as
yet this treat­ment remains untested.
PVOD, pul­mo­nary cyto­lytic thrombi, and trans­plant-­
asso­ci­ated throm­botic microangiopathy
PVOD dis­ease, PCT, and trans­plant-asso­ci­ated throm­botic micro­
angiopathy (TA-TMA) are endo­the­lial lung com­pli­ca­tions rarely
described after HCT that are iden­ti­fied by pathol­ogy and usu­
ally result in poor out­comes. PVOD is due to inti­mal fibro­sis and
venule hyper­tro­phy and obstruc­tion.40 While the path­o­logic
find­ings are sim­i­lar to those found in the liver of hepatic VOD
(sinu­soi­dal obstruc­tive syn­drome; SOS), the lit­er­a­ture is lacking
Prakash Singh Shekhawat
regard­ing co-inci­dence, and the peak tim­ing dif­fers for these
dis­eases (ear­lier after HCT for SOS and later for PVOD).41 PVOD
is dif­fi­cult to diag­nose, with dyspnea and non­spe­cific find­ings
on imag­ing that even­tu­ally man­i­fest as right ventricular dys­func­
tion.42 PVOD has been linked to myeloablative HCT and acute
graft-­ver­sus-host dis­ease (aGVHD) and other air­way dis­eases.40
Treatment with nitrates may be del­e­te­ri­ous, while cal­cium chan­
nel block­ers, pros­ta­cy­clins, and sildenafil have suggested pos­
si­ble ben­e­fit.42 Defibrotide, an antithrombotic and fibri­no­lytic
agent that sta­bi­lizes the endo­the­lium, has yet to be inves­ti­gated
for PVOD after HCT, though the mech­a­nism of action and suc­
cess in SOS would sup­port its use. Rare reports of pul­mo­nary
cyto­lytic thrombi (PCT) after HCT appeared in the early 2000s,
with pre­sen­ta­tion marked by fever, often in patients with ac­
tive aGVHD who had nod­ules on imag­ing and biopsy show­ing
leu­ko­cytes and baso­philic occlu­sions.43 TA-TMA is a syn­drome
dis­or­der ­after HCT of endo­the­lial dys­func­tion most com­monly
affect­ing the kid­ney and is dif­fi­cult to diag­nose with­out biopsy.
It can lead to pul­mo­nary arte­rial hyper­ten­sion and hyp­ox­emia,
has been dem­on­strated on biopsy with fibrin occlu­sion of small
pul­mo­nary ves­sels, and has been treated with sildenafil and ni­
tric oxide.44,45 Given the sim­i­lar­i­ties among these diag­noses, it is
unclear if they are sep­a­rate enti­ties or a spec­trum of dis­ease pro­
cesses. Collectively, these rare endo­the­lial dis­eases would ben­e­
Noninfectious pul­mo­nary tox­ic­ity after HCT  |  581
fit from fur­ther study, refined diag­nos­tic cri­te­ria, and sys­tem­atic
eval­u­a­tion of treat­ment approaches.
Other pul­mo­nary diag­noses after HCT
HCT recip­i­ents are at risk for pul­mo­nary prob­lems that com­
monly occur in crit­i­cal ill­ness. These include inflam­ma­tory con­
di­tions, acute respi­ra­tory dis­tress syn­drome (ARDS), trans­fu­sion-
asso­ci­ated acute lung injury (TRALI), trans­fu­sion-asso­ci­ated
cir­cu­la­tory over­load (TACO), and cryp­to­genic orga­niz­ing pneu­
mo­ nia (COP; pre­ vi­
ously termed bronchiolitis obliterans and
orga­niz­ing pneu­mo­nia), which is respon­sive to ste­roids and
restric­tive on PFTs and often linked to infec­tions. Other con­di­
tions can con­trib­ute to pul­mo­nary dys­func­tion, includ­ing
­alpha-1-antitrypsin defi­ciency (A1AT), pneu­mo­tho­rax, pul­mo­nary
embolism, and, finally, can­cer relapse or posttransplant lymph­
oproliferative dis­or­ders (PTLDs).
CLINICAL CASE 2
One year after matched-sib­ling HCT for relapsed acute leu­ke­
mia, with busul­fan con­di­tion­ing, com­pli­cated by severe aGVHD
and cyto­meg­a­lo­vi­rus reactivation before day 100, a 12-year-old
girl presented with dyspnea on exer­tion and ocu­lar chronic
GVHD (cGVHD). She under­went lung biopsy, which con­firmed
bronchiolitis obliterans (forced expi­ra­tory vol­ume in 1 sec­ond
[FEV1]/VC ratio >0.7). Although she had mycobacteria, FEV1
failed to fully recover with protracted anti­mi­cro­bi­als con­sis­tent
with biopsy find­ings of bronchiolitis obliterans syn­drome (BOS),
improv­ing from 46% to 59% but then plateauing while still on
sirolimus treat­ment. She was treated in a trial with montelukast
with ris­ing FEV1 (to 73%, with FEV1 of 1.56-1.87 L at 2 years).
Now, over a decade later, her FEV1 is 2.08 L, 85% predicted, and
she remains on montelukast and treat­ment for mycobacteria
(which recurred after 2 anti­mi­cro­bial ces­sa­tions).
Pulmonary dif­fer­en­tial in Case 2
Late-onset non­in­fec­tious lung dis­ease affects nearly 20% of 100-
day HCT sur­vi­vors.46 Overall risk fac­tors include chest r­adi­a­tion
prior to HCT, lung infec­tion within the first 100 days after HCT, and
a reduced mean forced expi­ra­tory flow at 25% to 75% (FEF25-75)
at 100 days after HCT.46 BOS is the diag­no­sis of Case 2.
Bronchiolitis obliterans syn­drome
BOS is the accepted pul­mo­nary man­i­fes­ta­tion of cGVHD, diag­
nosed by an FEV1 < 75% predicted and an irre­vers­ible ≥10% decline
in <2 years, an FEV1/vital capac­ity (VC) ratio <0.7 or the lower
limit (LL) of the 90% CI of the ratio, an absence of infec­tion, and ei­
ther preexisting cGVHD, air trap­ping by expi­ra­tory com­put­er­ized
tomog­ra­phy (CT) or by PFT, or cir­cum­fer­en­tial fibro­sis of bron­chi­­
oles on biopsy.47,48 This case high­lights sev­eral impor­tant aspects
of the revised def­i­ni­tion of BOS, includ­ing the abil­ity to diag­nose
obstruc­tive dis­ease in chil­dren for whom the LL of nor­mal is far
higher than 0.7. Now this child could be diag­nosed with­out biopsy.
As the FEV1/VC ratio declines with age, this is also a key mod­i­fi­ca­
tion for older indi­vid­u­als for whom 0.7 may be a nor­mal value, and
a diag­no­sis of BOS is not warranted. In addi­tion, this case high­
lights the value of BAL, which could have diag­nosed mycobac­
teria and excluded rare dis­eases, eg, bronchial obliterans.49 Case
582  |  Hematology 2021  |  ASH Education Program
2 has sev­eral risk fac­tors linked to BOS, includ­ing busul­fan con­di­
tion­ing, viral infec­tions early after HCT, and aGVHD, with periph­
eral blood stem cell donor and ABO incom­pat­i­bil­ity not pres­
ent.47,50 A recent study sug­gests that pre-HCT viruses may con­fer
increased risk as well.27 The cur­rent inci­dence is approx­i­ma­tely 3%
in pedi­at­ric and 3% to 6% in adult HCT.1,51-53 A chal­lenge of this dis­
ease has been its insid­i­ous nature lead­ing to mod­er­ate decline at
diag­no­sis. The fre­quent use of home devices to detect early de­
clines in lung func­tion and the use of FEF25-75 as an indi­ca­tor may
aid in the ear­lier iden­ti­fi­ca­tion of BOS.52,54,55 Newer imag­ing and
serum mark­ers show prom­ise for BOS diag­no­sis, includ­ing para­
met­ric response map­ping (with recent data for machine-learn­ing
algo­rithms), xenon-129 mag­netic res­o­nance imag­ing (MRI), and
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/578/1851735/578williams.pdf by guest on 13 December 2021
serum MMP3, though the data are insuf­fic ­ ient to replace PFTs.56-59
This may be most crit­i­cal for chil­dren who can­not per­form PFTs
in the future.60 Therapy for BOS con­sis­tently has striven for dis­
ease sta­bi­li­za­tion because the nat­ur­al his­tory of BOS is steady
decline, with severe FEV1 declines asso­ci­ated with highest mor­
tal­ity.61 A pro­spec­tive, multi-insti­tu­tional study of inhaled flutica­
sone, azithromycin, and montelukast with a brief ste­roid burst
(to 1 mg/kg/d pred­ni­sone equiv­a­lent with rapid taper­ing off in
1 month) showed sta­bi­li­za­tion or improve­ment in 64% of patients
with new-onset BOS at 6 months.62 Immunosuppression with cal­
cineurin inhib­i­tor or sirolimus was con­tin­ued. This approach has
been endorsed by the Euro­pean Society for Blood and Marrow
Transplantation as first-line ther­apy for BOS after HCT.63 One of
these agents, azithromycin, came under scru­tiny due to increased
relapse rates in a ran­dom­ized trial of azithromycin prior to HCT,64
which were not con­firmed in a larger study of azithromycin in
established BOS, though a higher rate of sec­ond­ary neo­plasms
was observed.65 At this time, the ben­e­fits of azithromycin likely
out­weigh the risks for most BOS patients, though ste­roids and
other risks for sec­ond­ary neo­plasms should also be min­i­mized
and per­son­al­ized risks eval­u­ated. In patients with ongo­ing pul­mo­
nary decline, extra­cor­po­real photopheresis may improve sur­vival,
which builds upon prior ret­ro­spec­tive data of a response rate of
approx­i­ma­tely 60%.47,66 Etanercept has shown some suc­cess for
BOS (32%), while ruxolitinib has not had an impact on BOS pro­
gres­sion.67 A chal­lenge to eval­u­at­ing ther­a­peu­tic suc­cess is the
relaps­ing-remit­ting nature of BOS, though rapid declines have
con­sis­tently been asso­ci­ated with poorer out­comes, and the FEV1
slope can thus guide prog­no­sis.68 Lung trans­plan­ta­tion remains a
final option, and recent data sug­gest improved sur­vival of 80%
at a median of 5 years.69 Overall, the sur­vival of patients with BOS
has improved to 40% to 50% at 5 years, which likely reflects the
incor­po­ra­tion of newer ther­a­pies, a reduc­tion of protracted high-
dose ste­roids, and improve­ments in sup­port­ive care.47 As high­
lighted by our case, chil­dren may expe­ri­ence an unique tra­jec­tory
of recov­ery, which I have observed after sig­nif­i­cant growth and
which was recently published in a ret­ro­spec­tive cohort.51
Restrictive lung dis­ease
Restrictive lung dis­ease (RLD) after HCT is less well char­ac­ter­ized.
RLD has a much larger dif­fer­en­tial to exclude, such as infec­tions,
COP, and extrapulmonary con­straints, eg, scle­ro­sis or myo­si­tis.
Intraparenchymal pathol­ ogy can include elas­ tic fibro­
sis of the
pleura or paren­chyma, typ­i­cally in the upper lobes, termed pleu­
ral paren­chy­mal fibroelastosis (PPFE) or non­spe­cific inter­sti­tial
pneu­mo­nia (NSIP). A study of lung explants suggested that late
RLD was due to paren­chy­mal fibro­sis with 20% to 30% of small
air­ways obstructed (a much lower pro­por­tion than patients with
BOS, in which 70% were obstructed).70 These RLD find­ings can oc­
cur with many other HCT-asso­ci­ated lung pathol­o­gies, includ­ing
BOS, PVOD, and lym­pho­cytic bronchiolitis.71 Because che­mo­ther­
apy can induce PPFE or NSIP, the con­tri­bu­tion of cGVHD has been
dif­fi­cult to dis­cern. However, a recent meta-anal­y­sis com­pared
the inci­dence of PPFE in patients with che­mo­ther­apy or autol­
o­gous HCT to those with allo­ge­neic HCT and showed a higher
rate in allo­ge­neic HCT (71% vs 25%), suggesting that alloimmunity
plays a role in this pro­cess.72 A long-term sur­vi­vor pedi­at­ric study
showed an asso­ci­a­tion between RLD and GVHD as well as pre-
HCT pul­mo­nary dys­func­tion.73 There are no diag­nos­tic cri­te­ria for
RLD after HCT, though these find­ings have been reported: fibro­sis
The chang­ing land­scape of lung injury after HCT
Infections and lung injury
The def­i­ni­tion of non­in­fec­tious lung injury is a mov­ing tar­get.
Polymerase chain reac­ tion (PCR) stud­ ies have improved the
detec­tion of viral, bac­te­rial, and fun­gal path­o­gens, which has
been fur­ther enhanced by next-gen­er­a­tion sequenc­ing (NGS)
with high sen­si­tiv­ity and spec­i­fic­ity.78-81 Using such PCR meth­ods
from a ret­ro­spec­tive cohort, 57% of IPS patients had a detect­
able path­o­gen in the BAL.82 By improv­ing infec­tion diag­nos­tics,
the role of infec­tions will be bet­ter elu­ci­dated.
We may need to mod­ify our approach to the con­sid­er­ation of
infec­tions with lung injury. Some non­in­fec­tious lung injuries are
likely late sequelae from an acute infec­tious pro­cess, eg, BOS,

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/578/1851735/578williams.pdf by guest on 13 December 2021

on CT (espe­cially the upper lobes) and reduced dif­fu­sion capac­ity
and an increased resid­ual vol­ume (RV)/total lung capac­ity ratio,
a 20% decline in FEV1 with a nor­mal FEV1/VC ratio, and reduced
VC.74,75 From adult series, the inci­dence of RLD would seem rarer
than BOS, up to 2.6% to 5%.46,74 However, inter­est­ingly, pedi­at­ric
reports of long-term sur­vi­vors (>6 years) sug­gest RLD is com­mon,
in up to 50% of sur­vi­vors, high­light­ing the need for con­sen­sus
diag­nos­tic cri­te­ria.73,76 Little is known to guide treat­ments of RLD;
approx­ i­
ma­tely one-third of patients responded to etanercept,
while ruxolitinib did not affect dis­ease.67,77 While newer stud­ies
have granted insight into the var­i­able pathol­ogy of this dis­ease,
the clin­i­cal diag­nos­tic cri­te­ria and eti­­ol­ogy remain ill-defined.
linked to viral infec­tions after HCT, and pro­gres­sion has been
linked to infec­tions as well.83
Challenges to diag­no­sis, treat­ment, and pre­ven­tion
There are sig­nif­i­cant diag­nos­tic chal­lenges to non­in­fec­tious
lung injuries. The diag­ nos­tic cri­te­
ria for BOS and IPS have
changed over time, lead­ing to chal­lenges in com­par­ing stud­
ies, though both have published con­sen­sus cri­te­ria.15,23,24,48,84-86
RLD lacks diag­nos­tic cri­te­ria alto­gether. It will be crit­i­cal to
con­tinue to use these stan­dard­ized diag­nos­tic cri­te­ria for these
chal­leng­ing diag­noses and refine with con­sen­sus to per­mit
broad appli­ca­tion.

Figure 2. Flow diagram for workup and diagnosis of noninfectious lung diseases after HCT. Blue box, diseases commonly diagnosed
in the first 100 days after HCT. Gray box, those diseases occurring usually beyond day 100 after HCT. Noninfectious lung injury workup
and diagnoses are in boxes. *Additional pulmonary diagnoses are denoted by stars and listed in the column in which these diagnoses
would be included in the differential. Notably, other processes can exhibit low DLCO, which is often reduced in RLD, but isolated
DLCO reduction should prompt evaluation for vascular diseases of the lung. For the workup of obstructive disease, infection is often
diagnosed, which should prompt repeat testing after treatment to ascertain BOS diagnosis (arrows). bx, biopsy; DLCO, diffusion
capacity of lung for carbon monoxide; ID, infectious disease workup; PE, pulmonary embolism; toxin-IP, toxin associated interstitial
pneumonitis, including that from radiation or chemotherapy.
Prakash Singh Shekhawat
Noninfectious pul­mo­nary tox­ic­ity after HCT  |  583
 Similarly, many of these diag­noses lack spe­cific diag­nos­tic
cri­te­ria. The endo­the­lial syn­dromes of PERDS, CLS, and TA-TMA
are par­tic­u­larly plagued by this issue, such that a recent review
made a “plea” for con­sen­sus def­i­ni­tions.86 While there are sub­
tle dif­fer­ences in the find­ings and risk fac­tors of PERDS/CLS and
idi­o­pathic or toxin-related pneu­mo­ni­tis, there is no way to dis­
tin­ guish these diag­ noses clin­ i­
cally. It is also unclear whether
these lung injuries result from a final com­mon immu­no­logic path­
way or result from dis­tinct mech­a­nisms. Addressing this crit­i­cal
ques­tion could enhance our abil­ity to iden­tify bio­mark­ers and
targeted ther­a­pies for these com­po­nents of IPS as well as the
pul­mo­nary vas­cu­lar diag­noses.
Accurate and non­in­va­sive diag­nos­tic tools that iden­tify lung
diag­noses remain an elu­sive goal. Lung biopsy yields the diag­
no­sis in 50% of patients but is asso­ci­ated with mor­bid­ity and
mor­tal­ity.87 BAL yields fewer diag­noses with fewer com­pli­ca­tions
and likely a higher yield with newer modal­i­ties (eg, NGS).87 Fig­
ure 2 and Table 2 sum­ma­rize a data-driven workup approach. As
is high­lighted in Figure 2, many pro­vid­ers attempt a diag­no­sis
through non­in­va­sive means (imag­ing, BAL, serum test­ing) and
reserve lung biopsy for those patients with non­spe­cific results
from these tests. Notably, these tests typ­ic ­ ally require over a
week to yield results, prompting most prac­ti­tion­ers to pre­emp­
tively treat broadly for infec­tions and for a non­in­fec­tious diag­
no­sis of high sus­pi­cion. Biopsy is still often required to diag­nose
vas­cu­lar com­pli­ca­tions, though some cen­ters have tried treat­
ments for the pul­mo­nary VOD or TA-TMA with­out biopsy and
used response as a method of diag­no­sis (often while awaiting
results from the BAL tests). Given the high rate of pneumotho­
raxes in patients with BOS, attempting to make this diag­no­sis
with­out biopsy is advan­ta­geous. Because this remains a clin­i­cal
chal­lenge, hope­fully, imag­ing or blood bio­mark­ers will be valu­
able for non­in­va­sive diag­no­sis in the future.
The pre­ven­tion of lung injuries after HCT is largely an unat­
tain­able end point because of the lack of knowl­edge of the
mech­a­nisms under­pin­ning these dis­eases. The reduc­tion of lung
injury through targeted dos­ing (of radi­a­tion or che­mo­ther­apy)
has dimin­ished the pneu­mo­ni­tis related to these insults. How­
ever, most of the non­in­fec­tious lung injuries lack known mod­
i­fia
­ ble incit­ing fac­tors. Further, the lack of objec­tive spe­cific
diag­nos­tic cri­te­ria, the rar­ity of these dis­eases, and the rel­a­tive
pau­city of ani­mal mod­els that mimic the clin­i­cal con­di­tion have
ham­pered prog­ress in under­stand­ing these dis­eases to develop
pre­ven­tive strat­e­gies. Finally, the unin­tended con­se­quences of
diminishing some known risk fac­tors, eg, aGVHD, can lead to
increased life-threat­en­ing risks, eg, relapse and infec­tions, con­
straining some efforts due to the need to bal­ance these risks.88,89
Conclusion
Noninfectious lung injury con­ trib­
utes to nonrelapse mor­ tal­
ity
and mor­bid­ity. It usu­ally devel­ops in the first year after HCT and
can affect the paren­chyma, air­ways, or vas­cu­lar struc­tures. Tim­
ing, pre­sen­ta­tion, and imag­ing can aid in distinguishing the diag­
no­sis, and mul­ti­ple diag­noses can coex­ist. However, diag­no­sis
and treat­ment are areas of crit­i­cal need. A bet­ter under­stand­ing
of the eti­­ol­ogy and immune dysregulation may inform both new
diag­nos­tic tools and bet­ter treat­ments to improve out­comes.
584  |  Hematology 2021  |  ASH Education Program
Acknowledgments 
Drs. Jennifer Holter Chakrabarty, James George, Gregory Yan­
ik, and Guang-Shing Cheng are acknowl­edged for their expert
coun­sel dur­ing this arti­cle’s prep­a­ra­tion.
Conflict-of-inter­est dis­clo­sure
Kirsten M. Williams: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Kirsten M. Williams: All drugs recommended in this manuscript
constitute off-label use with the exception of tociluzumab for
CRS. Drugs that are used off label in non-infectious lung treat­
ments include steroids, etanercept, fluticasone, azithromycin,
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/578/1851735/578williams.pdf by guest on 13 December 2021
montelukast, inhaled transexamic acid, recombinant factor VII.
Correspondence
Kirsten M. Williams, Blood and Marrow Transplant Program,
Aflac Cancer and Blood Disorders Center, Emory University
School of Medicine, Children’s Healthcare of Atlanta, 1760 Hay­
good Dr, 3rd floor W362, Atlanta, GA 30322; e-mail: kirsten​­.marie​
­.williams@emory​­.edu.
References
1. Duque-Afonso J, Ihorst G, Waterhouse M, et al. Impact of lung func­tion on
bronchiolitis obliterans syn­drome and out­come after allo­ge­neic hema­to­
poi­etic cell trans­plan­ta­tion with reduced-inten­sity con­di­tion­ing. Biol Blood
Marrow Transplant. 2018;24(11):2277-2284.
2. Fan K, McArthur J, Morrison RR, Ghafoor S. Diffuse alve­o­lar hem­or­rhage after
pedi­at­ric hema­to­poi­etic stem cell trans­plan­ta­tion. Front Oncol. 2020;10
(9 Sep­tem­ber):1757.
3. Keklik F, Alrawi EB, Cao Q , et al. Diffuse alve­o­lar hem­or­rhage is most often
fatal and is affected by graft source, con­di­tion­ing reg­i­men tox­ic­ity, and
engraft­ment kinet­ics. Haematologica. 2018;103(12):2109-2115.
4. Zhang Z, Wang C, Peters SG, et  al. Epidemiology, risk fac­tors, and out­
comes of dif­fuse alve­o­lar hem­or­rhage after hema­to­poi­etic stem cell trans­
plan­ta­tion. Chest. 2021;159(6):2325-2333.
5. Broglie L, Fretham C, Al-Seraihy A, et al. Pulmonary com­pli­ca­tions in pedi­
at­ric and ado­les­cent patients fol­low­ing allo­ge­neic hema­to­poi­etic cell
trans­plan­ta­tion. Biol Blood Marrow Transplant. 2019;25(10):2024-2030.
6. Majhail NS, Parks K, Defor TE, Weisdorf DJ. Diffuse alve­o­lar hem­or­rhage and
infec­tion-asso­ci­ated alve­o­lar hem­or­rhage fol­low­ing hema­to­poi­etic stem
cell trans­plan­ta­tion: related and high-risk clin­i­cal syn­dromes. Biol Blood
Marrow Transplant. 2006;12(10):1038-1046.
7. Haider S, Durairajan N, Soubani AO. Noninfectious pul­mo­nary com­pli­ca­
tions of haematopoietic stem cell trans­plan­ta­tion. Eur Respir Rev. 2020;
29(156):190119.
8. Rathi NK, Tanner AR, Dinh A, et  al. Low-, medium- and high-dose ste­
roids with or with­ out aminocaproic acid in adult hema­ to­
poi­etic SCT
patients with dif­fuse alve­o­lar hem­or­rhage. Bone Marrow Transplant. 2015;
50(3):420-426.
9. Elinoff JM, Bagci U, Moriyama B, et al. Recombinant human fac­tor VIIa for
alve­o­lar hem­or­rhage fol­low­ing allo­ge­neic stem cell trans­plan­ta­tion. Biol
Blood Marrow Transplant. 2014;20(7):969-978.
10. Baker MS, Diab KJ, Carlos WG, Mathur P. Intrapulmonary recom­bi­nant fac­
tor VII as an effec­tive treat­ment for dif­fuse alve­o­lar hem­or­rhage: a case
series. J Bronchology Interv Pulmonol. 2016;23(3):255-258.
11. Park JA, Kim BJ. Intrapulmonary recom­bi­nant fac­tor VIIa for dif­fuse alve­ol­ar
hem­or­rhage in chil­dren. Pediatrics. 2015;135(1):e216–e220.
12. O’Neil ER, Schmees LR, Resendiz K, Justino H, Anders MM. Inhaled
tranexamic acid as a novel treat­ment for pul­mo­nary hem­or­rhage in crit­
i­
cally ill pedi­ at­
ric patients: an obser­ va­
tional study. Crit Care Explor.
2020;2(1):e0075.
13. Fan K, Hurley C, McNeil MJ, et  al. Case report: man­age­ment approach
and use of extra­cor­po­real mem­brane oxy­gen­a­tion for dif­fuse alve­o­lar
hem­or­rhage after pedi­at­ric hema­to­poi­etic cell trans­plant. Front Pediatr.
2020;8(13 Jan­u­ary):587601.
14. Chiang Y, Tsai CH, Kuo SH, et al. Reduced inci­dence of inter­sti­tial pneu­
mo­ni­tis after allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion using a
mod­i­fied tech­nique of total body irra­di­a­tion. Sci Rep. 2016;6:36730.
15. Ahya VN. Noninfectious acute lung injury syn­dromes early after hema­to­
poi­etic stem cell trans­plan­ta­tion. Clin Chest Med. 2017;38(4):595-606.
16. Durie E, Nicholson E, Anthias C, et al. Determining the inci­dence of inter­
sti­tial pneu­mo­ni­tis and chronic kid­ney dis­ease fol­low­ing full inten­sity
haemopoetic stem cell trans­plant con­di­tioned using a for­ward-planned
inten­sity mod­u­lated total body irra­di­a­tion tech­nique. Radiother Oncol.
2021;158(May):97-103.
17. Vande Vusse LK, Madtes DK. Early onset non­in­fec­tious pul­mo­nary syn­
dromes after hema­to­poi­etic cell trans­plan­ta­tion. Clin Chest Med. 2017;​
38(2):​233-248.
18. Wieruszewski PM, May HP, Peters SG, et al. Characteristics and out­come
of peri-engraft­ment respi­ra­tory dis­tress syn­drome fol­low­ing autol­o­gous
hema­to­poi­etic cell trans­plant. Ann Am Thorac Soc. 2020;18(6):1013-1019.
19. Chang L, Frame D, Braun T, et al. Engraftment syn­drome after allo­ge­neic
hema­to­poi­etic cell trans­plan­ta­tion pre­dicts poor out­comes. Biol Blood
Marrow Transplant. 2014;20(9):1407-1417.
20. Shah NN, Watson TM, Yates B, et  al. Procalcitonin and cyto­kine pro­files
in engraft­ment syn­drome in pedi­at­ric stem cell trans­plan­ta­tion. Pediatr
Blood Cancer. 2017;64(3):e26273.
21. Lucchini G, Willasch AM, Dan­iel J, et  al. Epidemiology, risk fac­tors, and
prog­no­sis of cap­il­lary leak syn­drome in pedi­at­ric recip­i­ents of stem cell
trans­plants: a ret­ro­spec­tive sin­gle-cen­ter cohort study. Pediatr Transplant.
2016;20(8):1132-1136.
22. Reikvam H, Grønningsæter IS, Ahmed AB, Hatfield K, Bruserud O. Metabolic
serum pro­files for patients receiv­ing allo­ge­neic stem cell trans­plan­ta­tion:
the pretransplant pro­file dif­fers for patients with and with­out posttrans­
plant cap­il­lary leak syn­drome. Dis Markers. 2015;2015(Octo­ber):943430.
23. Yanik G, Kitko C. Management of non­in­fec­tious lung injury fol­low­ing hema­
to­poi­etic cell trans­plan­ta­tion. Curr Opin Oncol. 2013;25(2):187-194.
24. Clark JG, Hansen JA, Hertz MI, Parkman R, Jensen L, Peavy HH. NHLBI work­
shop sum­mary: idi­o­pathic pneu­mo­nia syn­drome after bone mar­row trans­
plan­ta­tion. Am Rev Respir Dis. 1993;147(suppl 6, pt 1):1601-1606.
25. Wenger DS, Triplette M, Crothers K, et al. Incidence, risk fac­tors, and out­
comes of idi­o­pathic pneu­mo­nia syn­drome after allo­ge­neic hema­to­poi­etic
cell trans­plan­ta­tion. Biol Blood Marrow Transplant. 2020;26(2):413-420.
26. Yanik GA, Ho VT, Levine JE, et  al. The impact of sol­u­ble tumor necro­sis
fac­tor recep­tor etanercept on the treat­ment of idi­o­pathic pneu­mo­nia syn­
drome after allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion. Blood.
2008;112(8):3073-3081.
27. Versluys B, Bierings M, Murk JL, et  al. Infection with a respi­ra­tory virus
before hema­to­poi­etic cell trans­plan­ta­tion is asso­ci­ated with alloimmune-
medi­ated lung syn­dromes. J Allergy Clin Immunol. 2018;141(2):697-703.
e8703e8.
28. Sano H, Kobayashi R, Iguchi A, et al. Risk fac­tor anal­y­sis of idi­o­pathic pneu­
mo­nia syn­drome after allo­ge­neic hema­to­poi­etic SCT in chil­dren. Bone
Marrow Transplant. 2014;49(1):38-41.
29. Afessa B, Abdulai RM, Kremers WK, Hogan WJ, Litzow MR, Peters SG. Risk
fac­tors and out­come of pul­mo­nary com­pli­ca­tions after autol­o­gous hema­to­
poi­etic stem cell trans­plant. Chest. 2012;141(2):442-450.
30. Thompson J, Yin Z, D’Souza A, et al. Etanercept and cor­ti­co­ste­roid ther­apy
for the treat­ment of late-onset idi­o­pathic pneu­mo­nia syn­drome. Biol Blood
Marrow Transplant. 2017;23(11):1955-1960.
31. Yanik GA, Horowitz MM, Weisdorf DJ, et  al. Randomized, dou­ ble-blind,
pla­cebo-con­trolled trial of sol­u­ble tumor necro­sis fac­tor recep­tor: enbrel
(etanercept) for the treat­ment of idi­op ­ athic pneu­mo­nia syn­drome after allo­
ge­neic stem cell trans­plan­ta­tion: Blood and Marrow Transplant Clinical Trials
Network pro­to­col. Biol Blood Marrow Transplant. 2014;20(6):858–864.
32. Seo S, Yu J, Jenkins IC, et al. Diagnostic and prog­nos­tic plasma bio­mark­ers
for idi­o­pathic pneu­mo­nia syn­drome after hema­to­poi­etic cell trans­plan­ta­
tion. Biol Blood Marrow Transplant. 2018;24(4):678-686.
33. Bhargava M, Viken KJ, Dey S, et al. Proteome pro­fil­ing in lung injury after
hema­to­poi­etic stem cell trans­plan­ta­tion. Biol Blood Marrow Transplant.
2016;22(8):1383-1390.
34. Varelias A, Gartlan KH, Kreijveld E, et al. Lung paren­chyma-derived IL-6 pro­
motes IL-17A-depen­dent acute lung injury after allo­ge­neic stem cell trans­
plan­ta­tion. Blood. 2015;125(15):2435-2444.
35. Tizon R, Frey N, Heitjan DF, et al. High-dose cor­ti­co­ste­roids with or with­out
etanercept for the treat­ment of idi­o­pathic pneu­mo­nia syn­drome after allo-
SCT. Bone Marrow Transplant. 2012;47(10):1332-1337.
Prakash Singh Shekhawat
36. Gavriilaki E, Sakellari I, Gavriilaki M, Anagnostopoulos A. A new era in endo­
the­lial injury syn­dromes: tox­ic­ity of CAR-T cells and the role of immu­nity. Int
J Mol Sci. 2020;21(11):3886.
37. Lee DW, Santomasso BD, Locke FL, et  al. ASTCT con­sen­sus grad­ing
for cyto­kine release syn­drome and neu­ro­logic tox­ic­ity asso­ci­ated with
immune effec­tor cells. Biol Blood Marrow Transplant. 2019;25(4):625-638.
38. Maus MV, Alexander S, Bishop MR, et  al. Society for Immunotherapy of
Cancer (SITC) clin­i­cal prac­tice guide­line on immune effec­tor cell-related
adverse events. J Immunother Cancer. 2020;8(2):e001511.
39. Azoulay E, Castro P, Maamar A, et al; Nine-I Investigators. Outcomes in
patients treated with chi­me­ric anti­gen recep­tor T-cell ther­apy who were
admit­ted to inten­sive care (CARTTAS): an inter­na­tional, multicentre, obser­
va­tional cohort study. Lancet Haematol. 2021;8(5):e355–e364.
40. Gazourian L, Spring L, Meserve E, et  al. Pulmonary clin­ i­
co­
path­
o­log­
i­
cal
cor­re­la­tion after allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion: an
autopsy series. Biol Blood Marrow Transplant. 2017;23(10):1767-1772.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/578/1851735/578williams.pdf by guest on 13 December 2021
41. Mandel J, Mark EJ, Hales CA. Pulmonary veno-occlu­ sive dis­ease. Am J
Respir Crit Care Med. 2000;162(5):1964-1973.
42. Bunte MC, Patnaik MM, Pritzker MR, Burns LJ. Pulmonary veno-occlu­sive
dis­ease fol­low­ing hema­to­poi­etic stem cell trans­plan­ta­tion: a rare model of
endo­the­lial dys­func­tion. Bone Marrow Transplant. 2008;41(8):677-686.
43. Gulbahce HE, Pambuccian SE, Jessurun J, et al. Pulmonary nod­u­lar lesions
in bone mar­row trans­plant recip­i­ents: impact of his­to­logic diag­no­sis on
patient man­age­ment and prog­no­sis. Am J Clin Pathol. 2004;121(2):205–
210.
44. Jodele S, Hirsch R, Laskin B, Davies S, Witte D, Chima R. Pulmonary arte­rial
hyper­ten­sion in pedi­at­ric patients with hema­to­poi­etic stem cell trans­plant-
asso­ci­ated throm­botic microangiopathy. Biol Blood Marrow Transplant.
2013;19(2):202-207.
45. Nakamura Y, Mitani N, Ishii A, et al. Idiopathic pneu­mo­nia syn­drome with
throm­botic microangiopathy-related changes after allo­ge­neic hema­to­poi­
etic stem cell trans­plan­ta­tion. Int J Hematol. 2013;98(4):496-498.
46. Bergeron A, Chevret S, Peffault de Latour R, et al. Noninfectious lung com­
pli­ca­tions after allo­ge­neic haematopoietic stem cell trans­plan­ta­tion. Eur
Respir J. 2018;51(5).
47. Williams KM. How I treat bronchiolitis obliterans syn­drome after hema­to­
poi­etic stem cell trans­plan­ta­tion. Blood. 2017;129(4):448-455.
48. Williams KM, Chien JW, Gladwin MT, Pavletic SZ. Bronchiolitis obliterans
after allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion. JAMA. 2009;302(3):
306-314.
49. Yokoi T, Hirabayashi N, Ito M, et al; Nagoya Bone Marrow Transplant Group.
Broncho-bronchiolitis obliterans as a com­pli­ca­tion of bone mar­row trans­
plan­ta­tion: a clin­i­co­path­o­log­i­cal study of eight autopsy cases. Virchows
Arch. 1997;431(4):275-282.
50. Zhou X, O’Dwyer DN, Xia M, et al. First-onset herpesviral infec­tion and lung
injury in allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion. Am J Respir Crit
Care Med. 2019;200(1):63-74.
51. Walther S, Rettinger E, Maurer HM, et  al. Long-term pul­mo­nary func­tion
test­ing in pedi­at­ric bronchiolitis obliterans syn­drome after hema­to­poi­etic
stem cell trans­plan­ta­tion. Pediatr Pulmonol. 2020;55(7):1725-1735.
52. Jamani K, He Q , Liu Y, et al. Early post-trans­plan­ta­tion spi­rom­e­try is asso­ci­
ated with the devel­op­ment of bronchiolitis obliterans syn­drome after allo­
ge­neic hema­to­poi­etic cell trans­plan­ta­tion. Biol Blood Marrow Transplant.
2020;26(5):943-948.
53. Yomota M, Yanagawa N, Sakai F, et  al. Association between chronic
bac­te­rial air­way infec­tion and prog­no­sis of bronchiolitis obliterans syn­
drome after hema­to­poi­etic cell trans­plan­ta­tion. Medicine (Baltimore).
2019;98(1):e13951.
54. Turner J, He Q , Baker K, et al. Home spi­rom­e­try telemonitoring for early
detec­tion of bronchiolitis obliterans syn­drome in patients with chronic
graft-ver­sus-host dis­ease. Transplant Cell Ther. 2021;27(7):616.e1-616.e6.
55. Abedin S, Yanik GA, Braun T, et al. Predictive value of bronchiolitis obliter­
ans syn­drome stage 0p in chronic graft-ver­sus-host dis­ease of the lung.
Biol Blood Marrow Transplant. 2015;21(6):1127-1131.
56. Liu X, Yue Z, Yu J, et  al. Proteomic char­ac­ter­iza­tion reveals that MMP-
3 cor­re­lates with bronchiolitis obliterans syn­drome fol­low­ing allo­ge­
neic hema­to­poi­etic cell and lung trans­plan­ta­tion. Am J Transplant.
2016;16(8):2342-2351.
57. Cheng GS, Selwa KE, Hatt C, et  al. Multicenter eval­u­a­tion of para­met­
ric response map­ ping as an indi­ ca­
tor of bronchiolitis obliterans syn­
drome after hema­to­poi­etic stem cell trans­plan­ta­tion. Am J Transplant.
2020;20(8):2198-2205.
Noninfectious pul­mo­nary tox­ic­ity after HCT  |  585
58. Sharifi H, Lai YK, Guo H, et  al. Machine learn­ing algo­rithms to dif­fer­en­ti­
ate among pul­mo­nary com­pli­ca­tions after hema­to­poi­etic cell trans­plant.
Chest. 2020;158(3):1090-1103.
59. Walkup LL, Myers K, El-Bietar J, et al. Xenon-129 MRI detects ven­ti­la­tion def­
i­cits in pae­di­at­ric stem cell trans­plant patients unable to per­form spi­rom­e­
try. Eur Respir J. 2019;53(5):1801779.
60. Cuvelier GDE, Nemecek ER, Wahlstrom JT, et al. Benefits and chal­lenges
with diag­nos­ing chronic and late acute GVHD in chil­dren using the NIH
con­sen­sus cri­te­ria. Blood. 2019;134(3):304-316.
61. Kwok WC, Liang BM, Lui MMS, et  al. Rapid ver­sus grad­ual lung func­tion
decline in bronchiolitis obliterans syn­drome after haematopoietic stem
cell trans­plan­ta­tion is asso­ci­ated with sur­vival out­come. Respirology.
2019;24(5):459-466.
62. Williams KM, Cheng GS, Pusic I, et  al. Fluticasone, azithromycin, and
montelukast treat­ment for new-onset bronchiolitis obliterans syn­drome
after hema­to­poi­etic cell trans­plan­ta­tion. Biol Blood Marrow Transplant.
2016;22(4):710-716.
63. Penack O, Marchetti M, Ruutu T, et  al. Prophylaxis and man­age­ment of
graft ver­sus host dis­ease after stem-cell trans­plan­ta­tion for haematolog­
ical malig­nan­cies: updated con­sen­sus rec­om­men­da­tions of the Euro­
pean Society for Blood and Marrow Transplantation. Lancet Haematol.
2020;7(2):e157-e167.
64. Bergeron A, Chevret S, Granata A, et al; ALLOZITHRO Study Investigators.
Effect of azithromycin on air­ flow decline-free sur­ vival after allo­
ge­ neic
hema­to­poi­etic stem cell trans­plant: the ALLOZITHRO ran­dom­ized clin­i­cal
trial. JAMA. 2017;318(6):557-566.
65. Cheng GS, Bondeelle L, Gooley T, et al. Azithromycin use and increased
can­cer risk among patients with bronchiolitis obliterans after hema­to­poi­
etic cell trans­plan­ta­tion. Biol Blood Marrow Transplant. 2020;26(2):392–
400.
66. Hefazi M, Langer KJ, Khera N, et  al. Extracorporeal photopheresis
improves sur­vival in hema­to­poi­etic cell trans­plant patients with bron­
chiolitis obliterans syn­drome with­out sig­nif­i­cantly impacting mea­sured
pul­mo­nary func­tions. Biol Blood Marrow Transplant. 2018;24(9):1906-
1913.
67. Bondeelle L, Chevret S, Hurabielle C, et al. Effect of ruxolitinib on lung func­
tion after allo­ge­neic stem cell trans­plan­ta­tion. Biol Blood Marrow Trans-
plant. 2020;26(11):2115-2120.
68. Cheng GS, Storer B, Chien JW, et al. Lung func­tion tra­jec­tory in bronchi­
olitis obliterans syn­drome after allo­ge­neic hema­to­poi­etic cell trans­plant.
Ann Am Thorac Soc. 2016;13(11):1932-1939.
69. Kliman DS, Kotecha SR, Abelson DC, Snell GI, Glanville AR, Ma DDF. Favor­
able out­ come of lung trans­ plan­ta­
tion for severe pul­ mo­nary graft ver­
sus host dis­ease: an Aus­tra­lian mul­ti­cen­ter case series. Transplantation.
2019;103(12):2602-2607.
70. Verleden SE, McDonough JE, Schoemans H, et  al. Phenotypical diver­sity
of air­way mor­phol­ogy in chronic lung graft vs. host dis­ease after stem cell
trans­plan­ta­tion. Mod Pathol. 2019;32(6):817-829.
71. Takeuchi Y, Miyagawa-Hayashino A, Chen F, et  al. Pleuroparenchymal
fibroelastosis and non-spe­cific inter­sti­tial pneu­mo­nia: fre­quent pul­mo­nary
sequelae of haematopoietic stem cell trans­ plan­
ta­
tion. Histopathology.
2015;66(4):536-544.
72. Higo H, Miyahara N, Taniguchi A, Maeda Y, Kiura K. Cause of pleuroparen­
chymal fibroelastosis fol­low­ing allo­ge­neic hema­to­poi­etic stem cell trans­
plan­ta­tion. Respir Investig. 2019;57(4):321-324.
73. Madanat-Harjuoja LM, Valjento S, Vettenranta K, Kajosaari M, Dyba T, Taski­
nen M. Pulmonary func­tion fol­low­ing allo­ge­neic stem cell trans­plan­ta­tion in
child­hood: a ret­ro­spec­tive cohort study of 51 patients. Pediatr Transplant.
2014;18(6):617-624.
586  |  Hematology 2021  |  ASH Education Program
74. Namkoong H, Ishii M, Mori T, et al. Clinical and radio­log­i­cal char­ac­ter­is­tics
of patients with late-onset severe restric­tive lung defect after hema­to­poi­
etic stem cell trans­plan­ta­tion. BMC Pulm Med. 2017;17(1):123.
75. Yang JY, Oh SY, Song JW, et al. Restrictive chronic lung func­tion decline
after haematopoietic stem cell trans­plan­ta­tion. Eur Respir J. 2016;47(1):336–
339.
76. L’excel­lent S, Yakouben K, Delclaux C, Dalle JH, Houdouin V. Lung eval­ua ­­
tion in 10 year sur­vi­vors of pedi­at­ric allo­ge­neic hema­to­poi­etic stem cell
trans­plan­ta­tion. Eur J Pediatr. 2019;178(12):1833-1839.
77. Yanik GA, Mineishi S, Levine JE, et al. Soluble tumor necro­sis fac­tor recep­
tor: enbrel (etanercept) for sub­acute pul­mo­nary dys­func­tion fol­low­ing
allo­ge­neic stem cell trans­plan­ta­tion. Biol Blood Marrow Transplant. 2012;
18(7):1044-1054.
78. Blauwkamp TA, Thair S, Rosen MJ, et al. Analytical and clin­i­cal val­id ­ a­tion
of a micro­bial cell-free DNA sequenc­ing test for infec­tious dis­ease. Nat
Microbiol. 2019;4(4):663-674.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/578/1851735/578williams.pdf by guest on 13 December 2021
79. Gwinn M, MacCannell D, Armstrong GL. Next-gen­ er­
a­
tion sequenc­ ing of
infec­tious path­o­gens. JAMA. 2019;321(9):893-894.
80. Armstrong AE, Rossoff J, Hollemon D, Hong DK, Muller WJ, Chaudhury
S. Cell-free DNA next-gen­er­a­tion sequenc­ing suc­cess­fully detects infec­
tious path­o­gens in pedi­at­ric oncol­ogy and hema­to­poi­etic stem cell trans­
plant patients at risk for inva­sive fun­gal dis­ease. Pediatr Blood Cancer.
2019;66(7):e27734.
81. Rossoff J, Chaudhury S, Soneji M, et al. Noninvasive diag­no­sis of infec­tion
using plasma next-gen­er­a­tion sequenc­ing: a sin­gle-cen­ter expe­ri­ence.
Open Forum Infect Dis. 2019;6(8):ofz327.
82. Seo S, Renaud C, Kuypers JM, et al. Idiopathic pneu­mo­nia syn­drome after
hema­to­poi­etic cell trans­plan­ta­tion: evi­dence of occult infec­tious eti­­ol­o­
gies. Blood. 2015;125(24):3789-3797.
83. Zinter MS, Hume JR. Effects of hema­to­poi­etic cell trans­plan­ta­tion on the
pul­mo­nary immune response to infec­tion. Front Pediatr. 2021;9(26 Jan­u­
ary):634566.
84. Panoskaltsis-Mortari A, Griese M, Madtes DK, et al; Amer­i­can Thoracic
Society Committee on Idiopathic Pneumonia Syndrome. An offi­cial Amer­
i­can Thoracic Society research state­ment: non­in­fec­tious lung injury after
hema­to­poi­etic stem cell trans­plan­ta­tion: idi­o­pathic pneu­mo­nia syn­
drome. Am J Respir Crit Care Med. 2011;183(9):1262-1279.
85. Klein OR, Cooke KR. Idiopathic pneu­mo­nia syn­drome fol­low­ing hema­to­
poi­etic stem cell trans­plan­ta­tion. J Pediatr Intensive Care. 2014;3(3):147–
157.
86. Pagliuca S, Michonneau D, Sicre de Fontbrune F, et al. Allogeneic reac­tiv­ity-
medi­ated endo­the­lial cell com­pli­ca­tions after HSCT: a plea for con­sen­sual
def­i­ni­tions. Blood Adv. 2019;3(15):2424-2435.
87. Chellapandian D, Lehrnbecher T, Phil­lips B, et al. Bronchoalveolar lavage
and lung biopsy in patients with can­cer and hema­to­poi­etic stem-cell trans­
plan­ta­tion recip­i­ents: a sys­tem­atic review and meta-anal­y­sis. J Clin Oncol.
2015;33(5):501-509.
88. Williams KM, Inamoto Y, Im A, et al. National Institutes of Health con­sen­sus
devel­op­ment pro­ject on cri­te­ria for clin­i­cal tri­als in chronic graft-ver­sus-
host dis­ease, I: the 2020 Etiology and Prevention Working Group report.
Transplant Cell Ther. 2021;27(6):452-466.
89. Kitko CL, Pidala J, Schoemans HM, et  al. National Institutes of Health
con­sen­sus devel­op­ment pro­ject on cri­te­ria for clin­i­cal tri­als in chronic
graft-ver­sus-host dis­ease, IIa: the 2020 Clinical Implementation and Early
Diagnosis Working Group report. Transplant Cell Ther. 2021;27(7):545-557.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000293
 SURVIVORSHIP AFTER ALLOGENEIC HEMATOPOIETIC CELL TRANSPLANT

Infectious complications and vaccines

Per Ljungman

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/587/1851732/587ljungman.pdf by guest on 13 December 2021

Division of Hematology, Department of Medicine, Huddinge, Karolinska Institutet, Stockholm, Sweden; and Department of Cellular Therapy and
Allogeneic Stem Cell Transplantation, Karolinska Comprehensive Cancer Center, Karolinska University Hospital, Huddinge, Stockholm, Sweden

Infections are a major cause of morbidity and can result in mortality in long-term survivors after allogeneic hematopoietic
cell transplantation. Chronic graft-versus-host disease and delayed immune reconstitution are recognized risk factors.
Different strategies must be utilized depending on the individual patient’s situation but include prolonged antimicrobial
prophylaxis and vaccination. Some important infections due to pathogens preventable by vaccination are pneumococci,
influenza, varicella-zoster virus, and SARS-CoV-2. Despite the fact that such recommendations have been in place for
decades, implementation of these recommendations has been reported to be poor.

LEARNING OBJECTIVES
• Learn about important pathogens causing disease in allogeneic HCT survivors
• Increase awareness of the benefits and risks of vaccination of allogeneic HCT survivors

Introduction
Infections are major causes of morbidity and can result in
mortality in long-term survivors (more than 1 year) after
allogeneic hematopoietic stem cell transplantation (HCT).
Table 1 shows some important pathogens in this population.
The risk is increased with active chronic graft-versus-host
disease (GVHD) due to both incomplete immune reconsti-
tution and ongoing immunosuppressive therapy. Although
the risk for nonrelapse mortality has decreased over time,1
in a retrospective analysis infections represented 20.7% of
all deaths. Furthermore, infection was the second most fre-
quent cause of death in patients still alive 1 year after HCT.2
Two pandemics during the last 12 years have illustrated the
vulnerability of this population. The 2009 H1N1 influenza
“swine flu” pandemic and the recent COVID-19 pandemic
both caused significant mortality in the allogeneic-HCT
population in long-term survivors. It is to be expected that
new infections will appear from time to time and that those
who have undergone allogeneic HCT might be more vul-
nerable than the general population. Therefore, awareness
and preventive measures are indicated.
CLINICAL CASE
A 63-year-old man who had undergone an allogeneic
unrelated-donor transplantation 3 years previously came
Prakash Singh Shekhawat
to the emergency room with a high fever that had lasted
12 hours. He had a history of moderate chronic GVHD and
was still on tacrolimus. His wife reported that during the
last hour at home he had become lethargic. At examination
he was unconscious, was febrile, and had a blood pressure
of 79/50. He was admitted. C-reactive protein was 150 and
the white blood cell count was 11.5. A computed-tomogra-
phy scan was performed but showed no pathology. Blood
cultures were taken and a lumbar puncture performed. The
latter showed a cell count of 70, prominently consisting of
neutrophils. A gram stain showed gram-positive cocci, and
he was started on ceftriaxone.
Bacterial infections
Bacterial infections are the most common type of infection
causing mortality in both 1- and 5-year survivors after allo-
geneic HCT.2 The cause of infections occurring late after
transplant is frequently not identified, especially if occur-
ring away from transplant centers, but bacterial infections
likely represent many of these “unknown” infections. The
incidence of invasive pneumococcal disease (IPD) after
allogeneic HCT has been reported to be 80 times that in a
healthy population,3 can occur many years after allogeneic
HCT, and has a high mortality. The risk is increased in pa-
tients with chronic GVHD. Vaccination is an effective way
of reducing IPD in both young children and older adults.
Other vaccine-preventable bacterial infections that might
Infectious complications and vaccines | 587
Table 1.  Some impor­tant path­o­gens in allo­ge­neic HCT
sur­vi­vors
Bacteria Pneumococci
Hib
Meningococci
Viruses Influenza A and B
SARS-CoV-2
Respiratory syncytial virus
Varicella-zoster virus
Hepatitis B, C, and E
Fungi Pneumocystis jirovecii
Endemic myco­ses
cause severe late infec­tions are Haemophilus influenzae type
B (Hib) and menin­go­cocci. Multiresistant bac­te­ria can col­o­nize
allo­ge­neic HCT patients, and it has been reported that patients
becom­ing long-term car­ri­ers have an increased risk for late mor­
tal­ity.4 Some cen­ters give anti­bac­te­rial pro­phy­laxis pri­mar­ily
targeting pneu­mo­cocci to patients with chronic GVHD. No con­
trolled data exist regard­ing this approach, and what agent to
use and its effi­cacy is likely to depend on the anti­bac­te­rial resis­
tance sit­u­a­tion of pneu­mo­cocci in the com­mu­nity.
Viral infec­tions
Viral infec­tions have been reported to be the sec­ond most com­
mon cause of infec­tious dis­ease mor­tal­ity in long-term sur­vi­vors.2
Community-acquired respi­ra­tory virus infec­tions are impor­tant
causes of severe infec­tion occur­ring years after an allo­ge­neic
HCT, espe­cially in patients with delayed immune recon­sti­tu­tion.
The most impor­tant of these infec­tions until recently have been
caused by influ­enza A and B, para­in­flu­enza virus, and respi­ra­
tory syn­cy­tial virus. However, COVID-19 emerged dur­ing the last
year and has become a major cause of late mor­tal­ity after HCT.5,6
Ljungman et al reported from the European Society for Blood
and Marrow Transplantation reg­is­try a mor­tal­ity of 28.7% in pa-
tients diag­nosed with COVID-19 at 1 to 2 years after allo­ge­neic
HCT and 20.9% in patients at >2 years after the pro­ce­dure.6 Shar-
ma et al reported in a Center for International Blood and Marrow
Transplant Research anal­y­sis a mor­tal­ity of 15.6% in patients with
COVID-19 more than 1 year after HCT.5 It is likely, how­ever, that
both these stud­ies have a selec­tion bias in that milder infec­tions
in patients with­out comorbidities have been undi­ag­nosed.
Other impor­ tant viral infec­ tions in long-term sur­ vi­
vors are
reactivated var­i­cella-zoster virus (VZV) and chronic hep­a­ti­tis virus
infec­tions. Reactivated VZV infec­tion can be severe even dur­ing
appro­pri­ate anti­vi­ral pro­phy­laxis, espe­cially if the patient is still
receiv­ing immu­no­sup­pres­sion.7 Patients infected with hep­a­ti­tis B
virus or who have infected donors should receive pro­phy­laxis a
least dur­ing the time of ongo­ing immu­no­sup­pres­sion.8 Hepati-
tis E virus has emerged as a poten­tial cause of liver cir­rho­sis in
patients on chronic immu­no­sup­pres­sion, such as solid-organ and
allo­ge­neic HCT trans­plant recip­i­ents. Further stud­ies are needed.
Fungal infec­tions
Invasive fun­gal infec­tions were reported to be the third most fre­
quent cause of infec­tious dis­ease mor­tal­ity in 1-year sur­vi­vors but
588  |  Hematology 2021  |  ASH Education Program
the sec­ond in 5-year sur­vi­vors after allo­ge­neic HCT.2 Late infec­
tions occur with Pneumocystis jirovecii, espe­cially in patients with
delayed immune recon­sti­tu­tion, but can also occur with other
fungi, par­tic­ul­arly in the set­tings of relapse of the under­ly­ing dis­
ease or severe chronic GVHD. Pneumocystis pro­phy­laxis is there­
fore indi­cated. Ibrutinib has been asso­ci­ated with an increased
risk of inva­sive fun­gal infec­tion, and a recent small case series of
patients treated for chronic GVHD illus­trates this risk.9 Ruxolitinib
has also been asso­ci­ated with inva­sive fun­gal infec­tions. Systemic
anti­fun­gal pro­phy­laxis should be con­sid­ered in patients with se-
vere chronic GVHD receiv­ing inten­sive immu­no­sup­pres­sion.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/587/1851732/587ljungman.pdf by guest on 13 December 2021
CLINICAL CASE (Con­tin­ued)
The patient was treated with ceftriaxone and improved; how­
ever, he had devel­oped hear­ing loss and some addi­tional neu­
ro­log­i­cal def­i­cit. Blood and CSF cul­tures showed growth of
Streptococcus pneumoniae. Questioning his wife pro­vided the
infor­ma­tion that his phy­si­cians had recommended vac­ci­na­tion
against pneu­mo­cocci after the HCT, but it had never been per-
formed.
Vaccines
Table 2 shows cur­rent rec­om­men­da­tions for vac­ci­na­tion of HCT
recip­i­ents. These can be broadly divided into com­mon vac­cine-
pre­vent­able infec­tions caus­ing severe dis­ease after allo­ge­neic
HCT and vac­cines against rare but severe infec­tions when a broad
immu­nity in the pop­u­la­tion is desir­able. Examples of the for­mer
are pneu­mo­coc­cal infec­tions, influ­enza, and COVID-19, while
exam­ples of the lat­ter are tet­a­nus and diph­the­ria. In addi­tion,
travel vac­cines are indi­cated for trans­plant recip­i­ents want­ing to
travel to areas where they might encoun­ter region­ally com­mon
impor­tant path­og ­ ens, such as yel­low fever or Jap­a­nese enceph­
a­li­tis. This review discusses some exam­ples of the 3 sit­u­a­tions.
An impor­tant aspect of vac­ci­nat­ing allo­ge­neic HCT recip­i­ents
is safety. Currently avail­­able data with nonlive vac­cines sup­port
a high level of safety with side effects sim­i­lar to those in a nor­mal
pop­u­la­tion. Knowledge about risks with new vac­cine plat­forms
such as those used in the recently licensed COVID-19 vac­cines
is less and will require con­tin­ued and care­ful sur­veil­lance. Live
atten­u­ated vac­cines can cause vac­cine-induced dis­ease, which
can be fatal in severely immunosuppressed indi­vid­ua ­ ls. Careful
weighing of risk/ben­e­fit is there­fore needed.
Although inter­na­tional rec­om­men­da­tions regard­ing vac­ci­
na­tion have existed for sev­eral years,10,11 sev­eral reports state
that the implementation is poor, as illus­trated by the clin­i­cal
case.12-14 The rea­sons vary, includ­ing com­pli­ance, unclear respon­
si­bil­i­
ties, and reluc­tance to immu­ nize patients with GVHD or
ongo­ing immu­no­sup­pres­sion. Data clearly sug­gest, how­ever,
that patients with GVHD should be vac­ci­nated with nonlive vac­
cines, although it must be rec­og­nized that the immune response
might be lower.
Pneumococcal vac­cines
Pneumococcal poly­sac­cha­ride vac­cine is a poor inducer of an
immune response, espe­cially in patients with chronic GVHD. On
the other hand, stud­ies with the pneu­mo­coc­cal con­ju­gate vac­
cines (PCV) showed that they could induce strong and dura­ble
Table 2.  Recommended vac­cines after allo­ge­neic HCT

Vaccine Recommendation Comments

Nonlive vac­cines
Tetanus tox­oid + diph­the­ria tox­oid Yes Three doses (DT) starting 6 mo after trans­plan­ta­tion
Inactive influ­enza Yes Seasonal, begin­ning 4-6 mo after trans­plan­ta­tion depending on sea­son
Inactivated polio­vi­rus Yes Three doses starting 6 (-12) mo after trans­plan­ta­tion
Conjugated Hib Yes Three doses starting 6 (-12) mo after trans­plan­ta­tion
Pneumococcal con­ju­gate Yes Three doses starting 3 (-6) mo after trans­plan­ta­tion; booster at 12 mo in patients
with chronic GVHD
Pneumococcal poly­sac­cha­ride Yes Booster at 12 mo in patients with­out GVHD no ear­lier than 8 weeks after

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/587/1851732/587ljungman.pdf by guest on 13 December 2021

con­ju­gate
Acellular per­tus­sis Yes Children <7 starting 6 (-12) mo after trans­plan­ta­tion
Hepatitis B virus Yes In countries where it is recommended to the gen­eral pop­u­la­tion, starting 6 (-12)
mo after trans­plan­ta­tion
Papillomavirus Yes As in the gen­eral pop­u­la­tion, starting ear­li­est 6-12 mo after trans­plan­ta­tion;
three doses
Meningococcal con­ju­gate Yes As in the gen­eral pop­u­la­tion, starting 6 mo after trans­plan­ta­tion; two doses
Recombinant zoster vac­cine Can be con­sid­ered Limited data after stem cell trans­plan­ta­tion
Vaccines against COVID-19 Yes Limited data but risk/ben­e­fit favors vac­ci­na­tion
Live vac­cines
MMR Individual con­sid­er­ation Children and sero­neg­a­tive adults, not before 24 mo after HCT; not to be given
to patients with GVHD
Varicella Individual con­sid­er­ation Seronegative patients, not before 24 mo after BMT; not to be given in patients
with GVHD
Live zoster Not recommended
DT, diph­the­ria and tet­a­nus tox­oids; MMR, mea­sles, mumps, and rubella. Adapted from Cordonnier et al.10

immune responses. The cur­rent rec­om­men­da­tions are to start
vac­ci­na­tion with 3 doses of PCV at 3 to 4 months after HCT fol-
lowed by either a pneu­mo­coc­cal poly­sac­cha­ride vac­cine V23
dose in patients with­out or a fourth PCV dose in patients with
chronic GVHD.10,11 The intro­duc­tion of PCV in the man­age­ment
strat­egy has reduced the risk of IPD after allo­ge­neic HCT.15 The
long-term retainment of antibodies against pneu­ mo­cocci has
been stud­ ied after vary­ing vac­cine sched­ ules. However, the
major­ity received at least 3 PCV doses, as cur­rently recommend-
ed. Fifty per­cent of patients were protected against all 7 ana­lyzed
sero­types at a median of 9.3 years after trans­plan­ta­tion while 70%
were protected against 5 of 5 sero­types.16 A lack of pro­tec­tive an-
tibodies was asso­ci­ated with chronic GVHD, relapse of under­ly­
ing malig­nancy, and cord blood trans­plan­ta­tion. This shows that
seroprotection ought to be reg­u­larly assessed to define the need
for booster doses. New con­ju­gated vac­cines cov­er­ing addi­tional
sero­types are in devel­op­ment.
Hib
Immunization against Hib is recommended after HCT. However,
nontypable Haemophilus influenzae strains has become more
com­mon, and there is cur­rently no vac­cine against these strains.
Influenza vac­cine
Influenza A and B infec­tions can be severe and life-threat­en­ing,
and fatal infec­tions can occur sev­eral years after HCT. Therefore,
yearly influ­enza vac­ci­na­tion with the inactivated vac­cine is rec-
ommended for all­allo­ge­neic HCT recip­i­ents starting at 6 months
Prakash Singh Shekhawat
after HCT but could be ini­ti­ated as early as 4 months after HCT
dur­ing a com­mu­nity out­break. Children 6 months to 8 years of
age should be given a sec­ond dose. The live atten­u­ated influ­
enza vac­cine should not be used in HCT recip­i­ents. Family mem­
bers and hos­pi­tal staff should receive influ­enza vac­cine, thereby
reduc­ing the risk for trans­mis­sion.10,11
Several stud­ ies have shown poorer immune sero­ log­i­
cal
responses after vac­ci­na­tion in allo­ge­neic HCT recip­i­ents, both
adults and chil­ dren, com­ pared to healthy con­ trols. The time
after trans­plan­ta­tion is the most impor­tant fac­tor for vac­cine
effi­cacy, with patients vac­ci­nated later responding bet­ter. Two-
dose vac­cine sched­ules and the use of adjuvanted vac­cines have
been stud­ied with vary­ing results.
Despite sub­op­ti­mal sero­log­i­cal responses, there might be
clin­i­cal effec­tive­ness of vac­ci­na­tion since pro­tec­tive anti­body
lev­els against severe influ­enza dis­ease are poorly defined. Kumar
et al showed in a pro­spec­tive cohort study that influ­enza vac­
ci­na­tion dur­ing the rel­e­vant sea­son reduced the risk of severe
dis­eases such as pneu­mo­nia and the need for admis­sion to an
inten­sive care unit.17 Piñana et al showed in a sim­i­larly designed
cohort study that influ­enza vac­ci­na­tion reduced the risk of lower-
respi­ra­tory tract influ­enza and hos­pi­tal admis­sions.18 Thus, there
is a sub­stan­tial clin­i­cal ben­e­fit to reg­u­lar influ­enza vac­ci­na­tion of
HCT recip­i­ents.
COVID-19 vac­cines
COVID-19 is asso­ci­ated with sub­stan­tial mor­bid­ity and mor­tal­ity
after allo­ge­neic HCT. Thus, vac­ci­na­tion is indi­cated but must be
Infectious com­pli­ca­tions and vac­cines  | 589
eval­u­ated regard­ing the risk of pos­si­bly induc­ing GVHD or other
immune acti­va­tion phe­nom­ena. Currently, no infor­ma­tion is avail­­
able addressing these issues in allo­ge­neic HCT recip­i­ents. Data
in patients with hema­to­log­i­cal malig­nan­cies, espe­cially chronic
lymphocytic leukemia and mul­ ti­
ple mye­ loma, and after solid-
organ trans­plan­ta­tion sug­gest lower rates of response than in
healthy con­trols.19-24 It is cur­rently unclear how to inter­pret the
results of anti­body assays in severely immu­no­com­pro­mised indi­vid­
u­als such as HCT patients regard­ing pro­tec­tion against COVID-19.
Thus, the con­tin­ued use of pro­tec­tive mea­sures is recommended.
HPV vac­cine
HCT patients are prone to develop pap­il­lo­ma­vi­rus-driven com­
pli­ca­tions such as cer­vi­cal dys­pla­sia. Stratton et al showed that
strong immune responses can be elicited by the quad­ri­va­lent
human pap­il­lo­ma­vi­rus (HPV) vac­cine with­out sig­nif­i­cant side ef-
fects.25 It would be log­i­cal to vac­ci­nate HPV-sero­neg­a­tive indi­
vid­u­als to pre­vent acqui­si­tion of HPV, but whether vac­ci­na­tion of
HCT recip­i­ents already infected with HPV would result in a clin­
i­cal ben­efi
­ t is unknown. This could be a topic for future stud­ies.
Varicella vac­cine
Primary var­i­cella can be severe after HCT. The existing vac­cine
is live and atten­ua ­ ted and should not be used soon after HCT. A
sero­neg­a­tive patient should, if pos­si­ble, be immu­nized before
trans­plan­ta­tion, pro­vid­ing that enough time can elapse after
the vac­ci­na­tion. Vaccination of sero­neg­a­tive fam­ily mem­bers is
recommended. A few uncon­trolled stud­ies have reported that
var­i­cella vac­ci­na­tion is safe and can result in sero­con­ver­sion
if performed more than 2 years after HCT in patients with­out
chronic GVHD or ongo­ing immu­no­sup­pres­sion.
A high pro­por­tion of HCT patients develop her­pes zoster that
occa­sion­ally becomes severe. Both live and nonlive recom­bi­nant
vac­cines exist. Issa et al gave 1 dose of live atten­u­ated zoster
vac­cine to 58 allo­ge­neic HCT recip­i­ents and noted no sig­nif­i­cant
side effects.26 However, fatal infec­tions have been documented
after autol­o­gous HCT. Due to the proven effi­cacy and safety of
acy­clo­vir pro­phy­laxis, vac­ci­na­tion with the live zoster vac­cine is
not recommended dur­ing at least the first few years after HCT.
The recently licensed recom­ bi­
nant zoster vac­ cine showed
high immu­no­ge­nic­ity and decreased the rate of her­pes zoster in
autol­o­gous HCT recip­i­ents.27 It has also been stud­ied in allo­ge­neic
HCT recip­i­ents but not in a con­trolled trial. Camargo et al dem­
on­strated that 2 doses of recombinant zoster vaccine were less
immu­no­genic in allo­ge­neic HCT recip­i­ents com­pared to autol­o­
gous recip­i­ents and that VZV reactivations occurred.28 Baumrin et
al showed that a 2-dose reg­i­men was safe and tol­er­a­ble, with high
rates of mild to mod­er­ate local and sys­temic side effects but with­
out increas­ing the rates of GVHD; how­ever, VZV reactivations were
seen after vac­ci­na­tion.29 Additional stud­ies are there­fore needed.
Measles vac­cine
The major­ity of allo­ge­neic HCT patients will become sero­neg­
a­tive to mea­sles dur­ing extended fol­low-up. There are docu-
mented cases of fatal mea­sles in BMT recip­i­ents.30,31 The loss of
immu­nity is more rapid in patients pre­vi­ously vac­ci­nated against
mea­sles com­pared to those hav­ing expe­ri­enced nat­u­ral infec­
tion. Vaccination can only be con­sid­ered in a patient with­out
chronic GVHD or ongo­ing immu­no­sup­pres­sion. The reported
590  |  Hematology 2021  |  ASH Education Program
effect of vac­ci­na­tion varies between dif­fer­ent stud­ies, with a
seem­ingly higher response rate in adults than in chil­dren.
Travel vac­cines
Vaccines are avail­­able for HCT patients liv­ing in cer­tain areas
of the world or trav­el­ing to areas where cer­tain infec­tions are
endemic. Nonlive vac­ cines include those against tick-borne
enceph­a­li­tis32,33 and Jap­a­nese enceph­a­li­tis.34 The response to
hep­a­ti­tis A vac­cine was shown to be poor in both pre­vi­ously
sero­neg­a­tive and sero­pos­i­tive indi­vid­u­als.35 Yellow fever is en-
demic in some areas of the world, and the only vac­cine is live
atten­u­ated. In a ret­ro­spec­tive French mul­ti­cen­ter study, 21 pa-
tients who received yel­low fever vac­ci­na­tion a median of 39
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/587/1851732/587ljungman.pdf by guest on 13 December 2021
months after HCT dem­on­strated a high response rate with­out
documented side effects.36 For patients who live in or must vis-
it areas where yel­low fever is endemic, immu­ni­za­tion could be
con­sid­ered at least 2 years after HCT and if the patient is with­out
GVHD or ongo­ing immu­no­sup­pres­sion.
Conflict-of-inter­est dis­clo­sure
Per Ljungman: research funding, con­sul­tancy, lec­tur­ing: Pfizer.
Off-label drug use
Per Ljungman: nothing to disclose.
Correspondence
Per Ljungman, Department of Cellular Therapy and Allogeneic
Stem Cell Transplantation, M75, Karolinska University H­ ospital,
Hud­ dinge, SE-14186 Stockholm, Sweden; e-mail: per​­
.ljungman
@ki​­.se.
References
1. Penack O, Peczynski C, Mohty M, et  al. How much has allo­ge­neic stem
cell trans­plant-related mor­tal­ity improved since the 1980s? A ret­ro­spec­tive
anal­y­sis from the EBMT. Blood Adv. 2020;4(24):6283-6290.
2. Styczyński J, Tridello G, Koster L, et al; Infectious Diseases Working Party.
Death after hema­to­poi­etic stem cell trans­plan­ta­tion: changes over cal­en­
dar year time, infec­tions and asso­ci­ated fac­tors. Bone Marrow Transplant.
2020;55(1):126-136.
3. van Aalst M, Lötsch F, Spijker R, et al. Incidence of inva­sive pneu­mo­coc­cal
dis­ease in immu­no­com­pro­mised patients: a sys­tem­atic review and meta-
anal­y­sis. Travel Med Infect Dis. 2018;24(July-August):89-100.
4. Heidenreich D, Kreil S, Jawhar M, et  al. Course of col­on ­ i­za­tion by mul­ti­
drug-resis­tant organ­isms after allo­ge­neic hema­to­poi­etic cell trans­plan­ta­
tion. Ann Hematol. 2018;97(12):2501-2508.
5. Sharma A, Bhatt NS, St Martin A, et al. Clinical char­ac­ter­is­tics and out­comes
of COVID-19 in haematopoietic stem-cell trans­ plan­ta­tion recip­ i­
ents: an
obser­va­tional cohort study. Lancet Haematol. 2021;8(3):e185-e193.
6. Ljungman P, de la Camara R, Mikulska M, et  al. COVID-19 and stem cell
trans­plan­ta­tion: results from an EBMT and GETH mul­ti­cen­ter pro­spec­tive
sur­vey. Leukemia. 2021:1-10. Published online 2 June 2021.
7. Baumrin E, Cheng MP, Kanjilal S, Ho VT, Issa NC, Baden LR. Severe her­pes
zoster requir­ing intra­ve­nous anti­vi­ral treat­ment in allo­ge­neic hema­to­poi­
etic cell trans­plan­ta­tion recip­i­ents on stan­dard acy­clo­vir pro­phy­laxis. Biol
Blood Marrow Transplant. 2019;25(8):1642-1647.
8. Mallet V, van Bömmel F, Doerig C, et al; 5th Euro­pean Conference on Infec-
tions in Leukaemia. Management of viral hep­a­ti­tis in patients with haema-
tological malig­nancy and in patients under­go­ing haemopoietic stem cell
trans­plan­ta­tion: rec­om­men­da­tions of the 5th Euro­pean Conference on
Infections in Leukaemia (ECIL-5). Lancet Infect Dis. 2016;16(5):606-617.
9. Kaloyannidis P, Ayyad A, Bahaliwah Z, et al. Ibrutinib for ste­roid refrac­tory
chronic graft-ver­sus-host dis­ease: ther­a­peu­tic effi­ciency can be lim­ited
by increased risk of fun­gal infec­tion. Bone Marrow Transplant. 2021;56(8):​
2034-2037.
10. Cordonnier C, Einarsdottir S, Cesaro S, et al; Euro­pean Conference on Infec-
tions in Leukaemia Group. Vaccination of haemopoietic stem cell trans­plant
 recip­i­ents: guide­lines of the 2017 Euro­pean Conference on Infections in Leu-
kaemia (ECIL 7). Lancet Infect Dis. 2019;19(6):e200-e212.
11. Rubin LG, Levin MJ, Ljungman P, et al; Infectious Diseases Society of Amer-
ica. 2013 IDSA clin­i­cal prac­tice guide­line for vac­ci­na­tion of the immu­no­
com­pro­mised host. Clin Infect Dis. 2014;58(3):e44-100.
12. Dyer G, Gilroy N, Brice L, et al. A sur­vey of infec­tious dis­eases and vac­ci­
na­tion uptake in long-term hema­to­poi­etic stem cell trans­plant sur­vi­vors in
Australia. Transpl Infect Dis. 2019;21(2):e13043.
13. Ariza-Heredia EJ, Gulbis AM, Stolar KR, et al. Vaccination guide­lines after
hema­to­poi­etic stem cell trans­plan­ta­tion: prac­ti­tion­ers’ knowl­edge,
­atti­tudes, and gap between guide­lines and clin­i­cal prac­tice. Transpl Infect
Dis. 2014;​16(6):878-886.
14. Gouveia-Alves F, Gouveia R, Ginani VC, et  al. Adherence and immune
response to revaccination fol­low­ing hema­to­poi­etic stem cell trans­plan­ta­
tion at a pedi­at­ric onco-hema­tol­ogy ref­er­ence cen­ter. Transpl Infect Dis.
2018;20(4):e12903.
15. Roberts MB, Bak N, Wee LYA, et al. Clinical effec­tive­ness of con­ju­gate pneu­
mo­coc­cal vac­ci­na­tion in hema­to­poi­etic stem cell trans­plan­ta­tion recip­i­
ents. Biol Blood Marrow Transplant. 2020;26(2):421-427.
16. Robin C, Bahuaud M, Redjoul R, et al. Antipneumococcal seroprotection
years after vac­ci­na­tion in allo­ge­neic hema­to­poi­etic cell trans­plant recip­i­
ents. Clin Infect Dis. 2020;71(8):e301-e307.
17. Kumar D, Ferreira VH, Blumberg E, et  al. A 5-year pro­spec­tive mul­ti­cen­
ter eval­u­a­tion of influ­enza infec­tion in trans­plant recip­i­ents. Clin Infect Dis.
2018;67(9):1322-1329.
18. Piñana JL, Pérez A, Montoro J, et  al. Clinical effec­ tive­ ness of influ­ enza
vac­ci­na­tion after allo­ge­neic hema­to­poi­etic stem cell trans­plan­ta­tion: a
cross-​sec­tional, pro­spec­tive, obser­va­tional study. Clin Infect Dis. 2019;​
68(11):1894-1903.
19. Herishanu Y, Avivi I, Aharon A, et  al. Efficacy of the BNT162b2 mRNA
COVID-19 vac­cine in patients with chronic lym­pho­cytic leu­ke­mia. Blood.
2021;137(23):3165-3173.
20. Waissengrin B, Agbarya A, Safadi E, Padova H, Wolf I. Short-term safety
of the BNT162b2 mRNA COVID-19 vac­cine in patients with can­cer treated
with immune check­point inhib­i­tors. Lancet Oncol. 2021;22(5):581-583.
21. Pimpinelli F, Marchesi F, Piaggio G, et al. Fifth-week immu­no­ge­nic­ity and
safety of anti-SARS-CoV-2 BNT162b2 vac­cine in patients with mul­ti­ple mye­
loma and mye­lo­pro­lif­er­a­tive malig­nan­cies on active treat­ment: pre­lim­i­nary
data from a sin­gle insti­tu­tion. J Hematol Oncol. 2021;14(1):81.
22. Monin L, Laing AG, Muñoz-Ruiz M, et al. Safety and immu­no­ge­nic­ity of one
ver­sus two doses of the COVID-19 vac­cine BNT162b2 for patients with can­
cer: interim anal­y­sis of a pro­spec­tive obser­va­tional study. Lancet Oncol.
2021;22(6):765-778.
23. Marion O, Del Bello A, Abravanel F, et al. Safety and immu­no­ge­nic­ity of anti-
SARS-CoV-2 mes­sen­ger RNA vac­cines in recip­i­ents of solid organ trans­
plants. Ann Intern Med. 2021:M21-1341. Published online 25 May 2021.
Prakash Singh Shekhawat
24. Boyarsky BJ, Chiang TP, Ou MT, et  al. Antibody response to the Janssen
COVID-19 vac­cine in solid organ trans­plant recip­i­ents. Transplantation.
2021;105(8):e82-e83.
25. Stratton P, Battiwalla M, Tian X, et  al. Immune response fol­low­ing quad­
ri­va­lent human pap­il­lo­ma­vi­rus vac­ci­na­tion in women after hema­to­poi­etic
allo­ge­neic stem cell trans­plant: a nonrandomized clin­i­cal trial. JAMA Oncol.
2020;6(5):696-705.
26. Issa NC, Marty FM, Leblebjian H, et al. Live atten­u­ated var­i­cella-zoster vac­
cine in hema­to­poi­etic stem cell trans­plan­ta­tion recip­i­ents. Biol Blood Mar-
row Transplant. 2014;20(2):285-287.
27. Bastidas A, de la Serna J, El Idrissi M, et al; Zoster Efficacy Study in Patients
Undergoing HCT Study Group Collaborators. Effect of recom­bi­nant zoster
vac­cine on inci­dence of her­pes zoster after autol­o­gous stem cell trans­plan­
ta­tion: a ran­dom­ized clin­i­cal trial. JAMA. 2019;322(2):123-133.
28. Camargo JF, Lin RY, Natori Y, et al. Reduced immu­no­ge­nic­ity of the adju-
vanted recom­bi­nant zoster vac­cine after hema­to­poi­etic cell trans­plant: a
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/587/1851732/587ljungman.pdf by guest on 13 December 2021
pilot study. Blood Adv. 2020;4(19):4618-4622.
29. Baumrin E, Izaguirre NE, Bausk B, et  al. Safety and reactogenicity of the
recom­bi­nant zoster vac­cine after allo­ge­neic hema­to­poi­etic cell trans­plan­ta­
tion. Blood Adv. 2021;5(6):1585-1593.
30. Kaplan LJ, Daum RS, Smaron M, McCarthy CA. Severe mea­sles in immu­no­
com­pro­mised patients. JAMA. 1992;267(9):1237-1241.
31. Nakano T, Shimono Y, Sugiyama K, et  al. Clinical fea­tures of mea­sles in
immu­no­com­pro­mised chil­dren. Acta Paediatr Jpn. 1996;38(3):212-217.
32. Harrison N, Grabmeier-Pfistershammer K, Graf A, et  al. Humoral immune
response to tick-borne enceph­a­li­tis vac­ci­na­tion in allo­ge­neic blood and
mar­row graft recip­i­ents. NPJ Vaccines. 2020;5(1):67.
33. Einarsdottir S, Nicklasson M, Veje M, et al. Vaccination against tick-borne
enceph­a­li­tis (TBE) after autol­o­gous and allo­ge­neic stem cell trans­plan­ta­
tion. Vaccine. 2021;39(7):1035-1038.
34. Assawawiroonhakarn S, Apiwattanakul N, Pakakasama S, et al. Immunoge-
nicity of vero cell cul­ture-derived Jap­a­nese enceph­a­li­tis vac­cine in pedi­at­
ric and young hema­to­poi­etic stem cell trans­plan­ta­tion recip­i­ents. Pediatr
Infect Dis J. 2021;40(3):264-268.
35. Adati EM, da Silva PM, Sumita LM, et al. Poor response to hep­a­ti­tis A vac­ci­
na­tion in hema­to­poi­etic stem cell trans­plant recip­i­ents. Transpl Infect Dis.
2020;22(3):e13258.
36. Sicre de Fontbrune F, Arnaud C, Cheminant M, et al. Immunogenicity and
safety of yel­low fever vac­cine in allo­ge­neic hema­to­poi­etic stem cell trans­
plant recip­i­ents after with­drawal of immu­no­sup­pres­sive ther­apy. J Infect
Dis. 2018;217(3):494-497.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000294
Infectious com­pli­ca­tions and vac­cines  | 591
 THE CHANGING LANDSCAPE OF α - AND β -THALASSEMIA DIAGNOSIS AND TREATMENT

Advances in the management of α-thalassemia

major: reasons to be optimistic

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/592/1851406/592horvei.pdf by guest on 13 December 2021

Paulina Horvei,1 Tippi MacKenzie,2 and Sandhya Kharbanda1
Division of Pediatric Allergy, Immunology and Bone Marrow Transplantation, and 2Division of Pediatric Surgery and Fetal Treatment Center, UCSF
1

Benioff Children’s Hospital, University of California, San Francisco, CA

α-Thalassemia major (ATM) is a severe disease resulting from deletions in all 4 copies of the α-globin gene. Although it is
usually fatal before birth, the advent of in utero transfusions has enabled survival of a growing number of children. Post-
natal therapy consists of chronic transfusions or stem cell transplantation, similar to patients with β-thalassemia major.
In this review, we discuss the experience with postnatal stem cell transplantation in patients with ATM, as well as the
ongoing phase 1 clinical trial of in utero stem cell transplantation for this condition.

LEARNING OBJECTIVES
• Review HCT for patients with ATM
• Discuss optimal conditioning regimens for HCT in patients with ATM

CLINICAL CASE patterns such that in California 1 in 10 000 newborns have

A pregnant woman of Southeast Asian ancestry under-
went routine prenatal ultrasound and was found to have a
fetus with hydrops fetalis, manifesting as ascites and pla-
centamegaly. She and her partner were both known to
be carriers of 2 α-globin deletions, giving them a 25%
chance of a pregnancy with deletion in all 4 α-globin genes
(α-thalassemia major [ATM]). Given the concern for ATM
in the fetus, the couple were counseled regarding their
options for the pregnancy, including pregnancy termina-
tion or fetal therapy with intrauterine transfusions (IUTs).
They opted for IUTs, which were performed starting at
23 weeks of gestation and repeated every 3 weeks until
birth. The baby was born at term and stayed in the hospi-
tal for 2 weeks. Postnatal transfusions were given every 3
weeks. Neurological development was normal at the time
of testing at 8 months.
a clinically significant α-thalassemia, and the rate of ATM
is 0.2 per 100 000 state births.2 α-Thalassemia has varying
degrees of severity depending on the number of deleted or
mutated genes and remaining functional α-globin genes.
Carriers of this condition can have (a) 1 α-globin gene
deleted/inactivated in each chromosome (α−/α−), known
as the trans form of the α-thalassemia trait (common in
people of African descent), or b) 2 missing/inactivated
α-globin genes on the same chromosome (αα/−−), known
as the cis form of the α-thalassemia trait (common in peo-
ple of Asian descent).
There are 2 clinically significant forms: ATM, also known
as α0-thalassemia or hemoglobin (Hb) Bart’s hydrops feta-
lis, caused by deletion/inactivation of all 4 α-globin genes
(−/−), and HbH disease, most frequently caused by deletion/
inactivation of 3 α-globin genes (−/−α).3

Introduction Normal and abnormal developmental Hb

Epidemiology and genetics
α-Thalassemia is a recessively inherited hemoglobinopathy
caused by mutations in the α-globin genes located on the
short arm of chromosome 16. It is one of the most common
monogenic disorders in the world, affecting approximately
5% of the population; prevalence is highest in China, South-
east Asia, the Middle East, India, and Africa.1 There is a rising
incidence in the Western United States due to immigration
and pathology of hydrops
Fetal oxygen exchange is accomplished by embryonic Hb
(composed of 2 zeta and 2 gamma chains, or ξ2γ2) until 2
months of gestation (Figure 1); thereafter, fetal Hb (α2γ2)
performs this function.1 A reduction of α-globin results in
the aggregation of γ-globin tetramers in utero (Hb Bart’s)
and after birth, of β-globin tetramers (HbH). Both Hb Bart’s
and HbH have increased oxygen affinity, resulting in poor

592 | Hematology 2021 | ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/592/1851406/592horvei.pdf by guest on 13 December 2021
Figure 1. Hb switching at the human α- and β-globin loci. Green, embryonic globins; blue, fetal globins; red, adult globins. Primitive
erythropoiesis, derived from the yolk sac, is characterized by the expression of ζ-globin (from the α-globin locus) and ɛ-globin (from
the β-globin locus). These are silenced at approximately 8 weeks’ gestation. α-Globin then accounts for the entirety of the transcrip-
tional output from the α-globin locus. At the β-globin locus, there is a switch to fetal globin (γ-globin) during fetal life and then a sec-
ond switch to the adult β-globin. The predominant type of Hb corresponding to each developmental stage is shown below. Adapted
with permission from King and Higgs 2018.4

oxy­gen deliv­ery; they cause the pre­ma­ture destruc­tion of mature
red blood cells (RBCs), lead­ing to hemo­ly­sis, and dam­age matur­
ing ery­throid pre­cur­sors, lead­ing to inef­fec­tive eryth­ro­poi­e­sis.5
While most patients with a 3 α-glo­bin gene dele­tion do not have
severe symp­toms, some patients with nondeletional muta­tions,
such as Hb Constant Spring, require chronic trans­fu­sions and
may ben­e­fit from stem cell trans­plan­ta­tion.6
ATM is the most severe form: patients have a pre­ na­tal
onset of non­im­mune hydrops fetalis as a result of heart fail­ure
induced by severe ane­mia. Embryonic Hb Portland (ξ2γ2) is
the only func­tional oxy­gen-car­ry­ing Hb in these infants, which
is short-lived, as the ξ-glo­bin gene is then silenced, switch-
ing to the non­ func­
tional Hb Bart’s. Extramedullary eryth­ ro­
poi­e­sis with marked hepatosplenomegaly and an enlarged
pla­centa are com­mon. With few excep­tions, ATM is fatal in
utero or shortly after birth with­out IUTs to treat the ane­mia.
In untreated patients, hydrops can lead to mater­nal com­pli­
ca­tions such as ane­mia, polyhydramnios, pre­term labor, and
pre­eclamp­sia.3
Prenatal diag­no­sis and coun­sel­ing
Couples who are car­ri­ers for α-glo­bin dele­tions and are at risk
of hav­ing a child with ATM should receive genetic coun­sel­ing
to review the nat­u­ral his­tory of the dis­or­der and dis­cuss pre­
na­tal genetic test­ing, repro­duc­tive options, and pre­na­tal and
post­na­tal treat­ment options and their risks. During preg­nancy,
a defin­i­tive fetal diag­no­sis can be obtained using chorionic vil­
lus sam­pling, amnio­cen­te­sis, or fetal blood sam­pling.7 Given the
neces­sity of IUTs to enable sur­vival, early diag­no­sis (pref­er­a­bly
prior to the onset of hydrops fetalis) is impor­tant. Although some
infants have, report­edly, sur­vived with­out IUTs, they are almost
always born pre­term and have neo­na­tal com­pli­ca­tions due to
the effects of untreated fetal hyp­oxia.5
Prakash Singh Shekhawat
IUTs and impact on clin­i­cal out­comes
Given the sever­ity of ATM and the lack of ther­a­peu­tic options
uni­ ver­sally avail­­
able until recently, par­ ents have most often
elected to undergo ter­ mi­na­
tion of preg­ nancy. However, IUTs
are now com­monly performed for mul­ti­ple fetal ane­mias and
are increas­ingly used to treat fetuses with ATM. Although the
ini­tial IUT can treat the crit­i­cally low ane­mia, sub­se­quent IUTs
are gen­ er­ally given every 3 weeks to main­ tain an appro­ pri­
ate Hb level.1 Most fetal cen­ters start IUTs at 18 weeks of ges­
ta­tional age and fol­low pro­to­cols sim­i­lar to those for treat­ment
of isoimmunization,8 with low com­pli­ca­tion rates (1.2% per pro­
ce­ dure, 3.3% per fetus).9 Since most fetuses are diag­ nosed
after 18 weeks, the most com­mon approach is an intra­ve­nous
infu­sion through the umbil­i­cal vein. The pro­to­col is to trans­fuse
O neg­a­tive blood with a high hemat­o­crit so that the ane­mia
may be corrected with­out vol­ume overloading the fetus. Several
reports indi­cate that this fetal ther­apy pro­vi­des ben­e­fits dur­ing
the peri­na­tal and neo­na­tal period: it reverses ane­mia, fetal growth
restric­tion, and hydrops; decreases pre­term deliv­er­ies; and con­
trib­utes to infants hav­ing higher Apgar scores and a shorter dura­
tion of neo­na­tal ven­ti­la­tion com­pared to untreated patients.5
Several case series have also dem­on­strated a long-term pos­i­tive
impact on growth and devel­op­ment.1,5,10,11 After birth, infants with
ATM con­tinue to require trans­fu­sions, using a pro­to­col sim­i­lar to
patients with β-thal­as­se­mia major (BTM). Given the improv­ing
sur­vival of patients with ATM, par­ents can be given the option of
fetal ther­apy dur­ing a non­di­rec­tive pre­na­tal coun­sel­ing ses­sion.
Allogeneic hema­to­poi­etic stem cell trans­plan­ta­tion
for ATM
Allogeneic hema­to­poi­etic stem cell trans­plan­ta­tion (HCT) is cur­
rently the only avail­­able cura­tive treat­ment option for patients
with ATM. The under­ly­ing prin­ci­ple of allo­ge­neic HCT in ATM is
Advances in ther­a­pies for α-thal­as­se­mia major  |  593
sim­i­lar to that in other hemo­glo­bin­op­a­thies and involves the
replace­ment of recip­i­ent hema­to­poi­etic stem cells (HSCs) with
donor HSCs, resulting in the pro­duc­tion of donor-derived RBCs
that do not har­bor the α-thal­as­se­mia muta­tion.
In 1998 a 21-month-old girl with ATM received a bone mar­row
(BM) HCT from a matched sib­ling donor (MSD) after con­di­tion­ing
with busul­fan (Bu), cyclo­phos­pha­mide (Cy), and horse-derived
antithymocyte glob­u­lin (hATG). Graft-ver­sus-host dis­ease (GVHD)
pro­phy­laxis included meth­o­trex­ate (MTX) and cyclo­spor­ine
A (CSA). Neutrophil engraft­ment was detected at day +17. No
major HCT-related com­pli­ca­tions devel­oped. Hb lev­els remained
>10 g/dL with­ out blood trans­ fu­sions despite the pres­ ence of
resid­ ual host HSCs.12 Since this first case published in 1998,
14 other patients who received an HCT for ATM have been
shared with the sci­en­tific com­mu­nity (Table 1): 5 patients (33%)
did not receive IUTs, and all­were born pre­ma­ture, with lower
Apgar scores at 1 and 5 min­utes, had evi­dence of hydrops
(3 mild, 2 severe), and required lon­ger inten­sive neo­na­tal care.
Ten patients (66%) received IUTs and con­ se­quently had less
hydrops and were born at term (50%) or late pre­term. Despite
some patients expe­ri­enc­ing a chal­leng­ing ini­tial neo­na­tal course,
with chronic trans­ fu­
sions, inten­sive care, and close fol­ low-up
they were ­able to be bridged to trans­plant. Ten patients (66%)
received HCT at an age ≤24 months, with the youn­gest being
5 months of age. In total, 16 HCTs were performed; 2 patients had
graft fail­ure, 1 of which was sal­vaged with a sec­ond trans­plant,
while the other con­tin­ued on chronic trans­fu­sions. In 5 patients
the choice of con­di­tion­ing reg­i­men was not reported (NR). Of
the remaining, the major­ ity received myeloablative con­ di­
tion­
ing, which in 8 patients (53%) was a Bu-based reg­i­men along with
other agents, includ­ing fludarabine (Flu) and Cy. With respect to
types of donors and stem cell source, 4 patients received an MSD
(3 BM, 1 NR), 1 received related mismatched umbil­i­cal cord blood
cells (UCB), 4 received matched unre­lated donor stem cells (1 BM,
3 periph­eral blood stem cells [PBSCs]), 4 got mismatched unre­
lated donor HSCs (2 UCB, 2 NR), and 1 patient received αβ T-cell-
depleted haploidentical PBSCs. Only 1 patient, who was 13 years
old at the time of trans­plant—the oldest patient reported—died
from trans­plant-related com­pli­ca­tions. Considering those patients
that engrafted, all­became trans­ fu­
sion inde­
pen­ dent. Chimerism
data were reported in 13 patients, and approx­im ­ a­tely half of the
patients (6/13) were found to have mixed chi­me­rism and became
trans­fu­sion inde­pen­dent, sim­i­lar to those with full chi­me­rism.
From those patients with known iron over­ load prior to HCT
(9 total; 1 mild, 4 mod­er­ate, 4 severe), only 1 patient was reported
to develop severe and 1 patient to develop mild veno-occlu­sive
dis­ease of the liver (VOD). Of the patients who suc­ cess­ fully
engrafted and sur­vived, 6/13 (53%) had no devel­op­men­tal delays
reported (66% of whom had pre­vi­ously received IUTs), and 46%
(6/13) had a mild delay. For 1 patient this was NR.1,12-20
In con­trast to ATM, the results of allo­ge­neic HCT for a closely
related hemo­glo­bin­op­a­thy, BTM, are widely reported and well
stud­ied in large patient pop­u­la­tions. Patients transplanted after
the year 2000 have had an out­stand­ing 2-year over­all sur­vival
(OS) and event-free sur­vival (EFS) with the use of MSD (93% and
85%, respec­tively) and BM (91% and 83%, respec­tively).22 How-
ever, in recent years the use of Bu-based myeloablative reg­i­mens
have sig­nif­i­cantly improved out­comes with alter­na­tive donors,
and they are now com­pa­ra­ble to those of MSD (5-year OS and
EFS for MSD and HLA-matched unre­lated donors of 89% vs 87%
594  |  Hematology 2021  |  ASH Education Program
and 86% vs 82%, respec­tively).23 Evidence also sup­ports that
patients who receive a trans­plant before 2 years have the best
out­comes, with a 2-year OS and EFS of 95% and 93%.24
Traditional myeloablative con­di­tion­ing reg­i­mens such as
Bu/Cy typ­i­cally result in sustained donor mye­loid engraft­ment;
how­ever, they are asso­ci­ated with sig­nif­i­cant poten­tial tox­ic­ity
that can result in organ dys­func­tion, such as VOD and treat­ment-
related mor­tal­ity (TRM). Therefore, the use of reduced-tox­ic­ity
con­di­tion­ing (RTC) reg­i­mens, with the goal of min­i­miz­ing toxic-
ities with­out jeopardizing engraft­ment, is an appeal­ing option
and a con­tin­ued sub­ject of inves­ti­ga­tion in the field. Studies done
in patients with high-risk BTM showed that sub­sti­tu­tion of Flu for
Cy or Bu/Flu-based con­di­tion­ing reg­i­mens led to myeloablation
but were asso­ci­ated with decreased tox­ic­ity com­pared to the
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/592/1851406/592horvei.pdf by guest on 13 December 2021
con­ven­tional Bu/Cy reg­i­mens, hence the term RTC.22,25 Further
strides toward improved safety pro­files of RTC reg­i­mens have
been made with the use of phar­ma­co­ki­netic (PK) model-based
dos­ing of con­di­tion­ing agents, which allows for reduced tox­ic­ity
while maintaining the reg­i­men’s effi­cacy. The drug’s expo­sure,
mea­sured as a cumu­la­tive area under the curve (cAUC) over the
treat­ment course, can be adjusted in real time using ther­a­peu­tic
drug mon­i­tor­ing. A drug expo­sure below a cer­tain thresh­old car­
ries an increased risk of graft fail­ure, whereas an expo­sure above
the desired range is asso­ci­ated with an increased risk of organ
tox­ic­ity. In a study com­par­ing PK-targeted Bu and Flu and con­
ven­tional Bu and Cy, the for­mer approach was shown to be less
toxic while maintaining effi­cacy in pedi­at­ric HCT patients.26 In
sev­eral non­ma­lig­nant dis­eases, includ­ing thal­as­se­mia, this expo­
sure could be fur­ther reduced due to the accept­abil­ity of mixed
but sta­ble mye­loid chi­me­rism, which often leads to ther­a­peu­tic
ben­e­fit. More recently, our group has shown that a less toxic
and even lower expo­sure to Bu with a cAUC of 70 mg·h/L was
ade­ quate for dis­ ease cor­rec­ tion and achieve­ ment of dura­ ble
full or sta­ble mixed mye­loid engraft­ment (chi­me­rism of >20%) in
pedi­at­ric non­ma­lig­nant con­di­tions, with rare cases of severe Bu-
related tox­ic­ity and GVHD.27 Further, stud­ies aiming to under­
stand the rela­tion­ship between graft rejec­tion and Flu dose in
chil­dren under­go­ing HCT have been conducted to opti­mize the
immu­no­sup­pres­sive ele­ment of con­di­tion­ing based on which
a Flu cAUC of 16 to 20 mg·h/L is deemed safe and effec­tive
to over­come the immune bar­rier while show­ing no ben­e­fit of
increased expo­sure with respect to sur­vival.28,29 In a recent study
by Contreras et al,30 chil­dren receiv­ing RTC for non­ma­lig­nant
HCT, a total of 62 patients, were transplanted (23% had a diag­
no­sis of hemo­glo­bin­op­a­thies, 7, BTM, and 7, sickle cell dis­ease
[SCD]). Using targeted Bu and Flu in com­bi­na­tion with alemtu-
zumab (2012-2018), the cumu­la­tive inci­dence of graft fail­ure was
6.9% (none occurred in patients with an under­ly­ing diag­no­sis of
hemo­glo­bin­op­a­thies), the 3-year cumu­la­tive inci­dence of TRM
was 3.4%, and the 3-year OS was 96.6%, with a low inci­dence of
acute GVHD (aGVHD) grade 2 to 4 and chronic GVHD (7% and
5%, respec­tively). These results sug­gest that the use of targeted
Bu and Flu offers a well-tol­er­ated option for chil­dren with non­
ma­lig­nant dis­or­ders to achieve sustained engraft­ment with a low
inci­dence of trans­plant-related com­pli­ca­tions.30
The thresh­old chi­me­rism needed to ren­der trans­fu­sion inde­
pen­dence fol­low­ing trans­plan­ta­tion in patients with thal­as­se­
mias is not clearly known. However, the lim­ited data in BTM
sug­gest that com­plete donor chi­me­rism is not essen­tial for sus-
tained engraft­ment or for achiev­ing trans­fu­sion inde­pen­dence,
 Table 1. Summary of patients receiv­ing HCT for ATM

Conditioning reg­i­men and

Age at HCT Donor source GVHD pro­phy­laxis Toxicity and com­pli­ca­tions Outcomes Authors
21 months MSD BM Bu, Cy, hATG, MTX, CSA None Neutrophil engraft­ment at Chik et al12
day +17. Stable mixed
chi­me­rism (75%-90%).
Transfusion inde­pen­dent.
20 months 5/6 sib­ling UCB (α-thal trait) Bu, Cy, hATG, MTX, CSA HHV-6 vire­mia, grade 2 skin Neutrophil engraft­ment Zhou et al14
aGVHD, ITP at day +26. 100% donor
chi­me­rism. Transfusion
inde­pen­dent.
23 months MSD BM (α-thal and HbE trait) Cy, TBI (1400 cGy), MTX None Neutrophil engraft­ment at Thornley et al15
day +26. Stable mixed chi­
me­rism (66%). Transfusion
inde­pen­dent.
24 months MSD Bu, Flu, ATG, TLI Mild VOD Initially engrafted but then Joshi et al16
had graft fail­ure at 7
months after HCT. On
chronic trans­fu­sions with
che­la­tion ther­apy.
8 months, 18 months #1: 4/6 unre­lated UCB #1 HCT: Bu, Flu, TLI, CSA, EBV + PTLD #1 HCT: graft fail­ure and on Yi et al17
mismatched at B and DRB; MMF; #2 HCT: Bu, Flu, ATG, chronic trans­fu­sions.
#2 MUD 6/6 PBSCT, CD34+ TBI (200 cGy), CSA, MMF #2 HCT: neu­tro­phil engraft­
selec­tion ment at day +18. Stable
mixed chi­me­rism (90%-
97%). Transfusion

Prakash Singh Shekhawat

inde­pen­dent.
44 months 5/6 unre­lated UCB Bu, Cy, ATG Grade 2 skin aGVHD, 100% donor chi­me­rism. Gumuscu et al13
can­dida sep­sis Transfusion inde­pen­dent.
19 months MSD BM NR NR Neutrophil engraft­ment at Pongtanakul et al20
day +17. Initially, 100%
donor at day +25 then sta­
ble mixed chi­me­rism (41%
donor at day +112) with sta­
ble Hb of 9.1. Discontinued
immu­no­sup­pres­sion and
received DLI × 5. Converted
to 100% donor chi­me­rism.
Transfusion inde­pen­dent.
5 months MUD 10/10 BM Bu, Flu, ATG, MTX, CSA Severe VOD Neutrophil engraft­ment at Elsaid et al19
day +25. Stable mixed
chi­me­rism. Transfusion
inde­pen­dent.
13 years 9/10 MMUD NR (RIC) Invasive fun­gal infec­tion, Died 10 months post trans­ Pecker et al20
grade 4 skin and gut GVHD plant from trans­plant-
related com­pli­ca­tions.
10 years 9/10 MMUD NR (RIC) None 100% donor chi­me­rism. Pecker et al.20
Transfusion inde­pen­dent.

Advances in ther­a­pies for α-thal­as­se­mia major  |  595

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/592/1851406/592horvei.pdf by guest on 13 December 2021
 Table 1. (continued)

Conditioning reg­i­men and

Age at HCT Donor source GVHD pro­phy­laxis Toxicity and com­pli­ca­tions Outcomes Authors
12 months MSD NR NR Transfusion inde­pen­dent by Kreger et al1
18 months.

596  |  Hematology 2021  |  ASH Education Program

24 months NR NR NR Transfusion inde­pen­dent by Kreger et al1
age 3.
22 months MUD 12/12 PBSCT Bu, Cy, hATG, MTX, CSA Grade 2 skin and grade 3 gut Neutrophil engraft­ment at Chan et al18
aGVHD, HHV-7 vire­mia day +15. 100% donor chi­
me­rism. Transfusion inde­
pen­dent.
28 months MUD PBSC HU, azacitadine, Cy, Bu, TT, Grade 2 skin aGVHD. HHV-7 Neutrophil engraft­ment at Chan et al18
Flu, rATG, MTX, MMF, CSA and EBV vire­mia, cen­tral day +11. Mixed chi­me­rism
line sep­sis (98%). Transfusion inde­
pen­dent.
60 months AB TCD haploidentical PBSC HU, azacitadine, Cy, TT, Fu, Grade 2 skin and gut aGVHD. Neutrophil engraft­ment at Chan et al18
Treo, rATG Klebsiella bac­ter­emia. day +13. Mixed chi­me­rism
(99%). Transfusion inde­
pen­dent.
AB TCD, αβ T-cell depleted; cGVHD, chronic graft-ver­sus-host dis­ease; DRB, HLA DR isotype, beta chain; DLI, donor lym­pho­cyte infu­sions; EBV, Epstein-Barr virus; HbE, hemoglobin E;
HHV, human her­pes­vi­rus; HU, hydroxy­urea; ITP, immune throm­bo­cy­to­pe­nia; Mel, mel­pha­lan; MMF, mycophenolate mofetil; MUD, matched unre­lated donor; MMUD, mismatched unre­lated donor;
PBSCT, periph­eral blood stem cell trans­plant; PTLD, post-transplant proliferative disease; rATG, rab­bit-derived ATG; RIC, reduced-inten­sity con­di­tion­ing; TBI, total body irra­di­a­tion; TLI, total
lym­phoid irra­di­a­tion; Treo, treosulfan; TT, thio­tepa.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/592/1851406/592horvei.pdf by guest on 13 December 2021

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/592/1851406/592horvei.pdf by guest on 13 December 2021

Figure 2. Prenatal screening for ATM. (1) This tool is not a replacement for referral to genetic counseling, which may happen at
any time in this pathway. Genetic counseling provides guidance for genetic testing and management options for families with
­pregnancies at risk for severe forms of thalassemia. (2) The sensitivity and specificity are not definitive, and not all carriers will be
detected by this screening. (3) The American College of Obstetricians and Gynecologists recommends that all pregnant women
have a CBC with a ­ ssessment of MCV. (4) Rare mutations, such as nondeletional α-thalassemia and others, may not be captured in this
­algorithm. In high-risk cases, or where Hb electrophoresis is abnormal, consultation with a genetic counselor and/or h ­ ematologist
is ­recommended. (5) The presence of HbA2 > 3.5 does not exclude a coexisting α0-thalassemia trait. In individuals of Southeast
Asian, ­Filipino, or ­Chinese descent who have microcytic hypochromic anemia, perform α-globin gene deletion and common vari-
ant studies irrespective of HbA2 level. CBC, complete blood count; CVS, chorionic villus sampling; HPLC, high-performance liquid
­chromatography; MCA, ­middle c ­ erebral artery; MCH, mean corpuscular hemoglobin; MCV, mean corpuscular volume; MoM, multiples
of the median; PCR, ­polymerase chain reaction; PSV, peak systolic velocity; PUBS, percutaneous umbilical blood sampling. Adapted
Prakash Singh Shekhawat
with permission from the University of California, San Francisco hemoglobinopathy screening algorithm.
Advances in ther­a­pies for α-thal­as­se­mia major  |  597
and this is thought to be due to inef­fec­tive eryth­ro­poi­e­sis,
which allows the donor RBCs to have a sur­vival advan­tage. The
per­sis­tence of resid­ual host cells at >25% early fol­low­ing trans­
plant (<60 days) was shown to be a risk for graft rejec­tion, while
a level <10% was deemed low risk for rejec­tion.31 Patients with
per­sis­tent mixed chi­me­rism (PMC), defined as a sta­ble mixed
chi­me­rism for >2 years, were a ­ ble to remain trans­fu­sion inde­
pen­dent even with >25% resid­ual host cells, and their graft was
func­tional and suf­fi­cient to lead to trans­fu­sion inde­pen­dence.31
This is fur­ther supported by evi­dence from split chi­me­rism in
patients with PMC, where Andreani et  al.32 observed that the
pro­por­tion of donor-derived cells was equally dis­trib­uted in the
dif­fer­ent cell lines, both in the periph­eral blood and BM, with
the excep­tion of the eryth­ro­cyte com­part­ment. Despite the
pres­ence of few donor-engrafted nucle­ated cells, the eryth­
ro­cytes were almost com­pletely donor in ori­gin (80%-100%).
This enrich­ment of donor RBCs in the blood was not observed
in ery­throid pre­cur­sors from the mar­row, suggesting that the
inef­fec­tive eryth­ro­poi­e­sis pre­sum­ably respon­si­ble for this phe­
nom­e­non works at a later stage of ery­throid devel­op­ment.32
Consequently, since ATM is also char­ac­ter­ized by inef­fec­tive
eryth­ro­poi­e­sis due to the imbal­ance of α and β chains,5 it is
likely that com­plete donor chi­me­rism is not crit­i­cal, and PMC
would be suf­fi­cient to ren­der trans­fu­sion inde­pen­dence fol­low­
ing allo­ge­neic HCT.31
Rationale for in utero trans­plan­ta­tion in patients with ATM
In utero HSC trans­plan­ta­tion (IUHCT) is a prom­is­ing ther­apy for
dis­eases that can be treated by post­na­tal HCT. IUHCT takes
advan­tage of the unique immune sta­tus in the fetus to allow
engraft­ment of transplanted cells with­out con­di­tion­ing.33 Our
team is cur­rently performing a phase 1 clin­i­cal trial of IUHCT
(NCT02986698) using mater­nal stem cells as the donor: this
strat­egy takes advan­tage of 2 aspects of mater­nal-fetal biol­
ogy resulting from the nat­u­ral traf­fick­ing of cells between the
mother and the fetus: first, the pres­ence of mater­nal cells in the
fetus results in fetal tol­er­ance to noninherited mater­nal anti­gens
dur­ing preg­nancy34; sec­ond, the mater­nal immune sys­tem can
medi­ate rejec­tion of third-party cells transplanted into the fetus
in mouse mod­els.35 Thus, trans­plan­ta­tion of mater­nal cells pro­
vi­des the highest like­li­hood of suc­cess. We are cur­rently enroll­
ing patients with ATM for IUHCT between 18 and 26 weeks of
ges­ta­tion (Figure 2). It is impor­tant to note that although there
are immune advan­tages to transplanting mater­nal stem cells
into the fetus, the cur­rent lack of an appro­pri­ate con­di­tion­ing
reg­i­men to cre­ate space in the hema­to­poi­etic niche results in
low lev­els of engraft­ment in most ani­mal mod­els unless a very
high dose of cells is infused. Therefore, the most likely sce­nario
is that the pre­na­tal trans­plan­ta­tion is part of a 2-step pro­to­col
in which the in utero trans­plant achieves enough engraft­ment
to sus­tain tol­er­ance to mater­nal anti­gens after birth followed
by a post­na­tal “boost” using mater­nal HSC with an RTC reg­
i­men with min­i­mal or no immunoablation, improv­ing engraft­
ment to clin­i­cally mean­ing­ful lev­els. If suc­cess­ful, this pro­to­col
would over­come the lack of donor avail­abil­ity as well as offer an
improved safety pro­file com­pared to tra­di­tional post­na­tal HCT.
Indeed, prom­is­ing out­comes have been seen in murine mod­els
of SCD and BTM in which IUHCT resulted in the devel­op­ment
of donor-spe­cific tol­er­ance and PMC. A post­na­tal boost was
given, and chi­me­rism was fur­ther enhanced to high lev­els with
598  |  Hematology 2021  |  ASH Education Program
near-com­plete Hb replace­ment (donor Hb >90%), ulti­mately
correcting SCD and β-thal­as­se­mia phe­no­types.36
There have been mul­ti­ple pre­vi­ous reports of IUHCT in mul­
ti­ple dis­ease set­tings, includ­ing in fetuses with ATM with low
engraft­ment.37,38 However, none of these pre­vi­ous reports has
used the strat­egy of a high dose of mater­nal cells, infused intra­
ve­nously, which is the pro­to­col of our ongo­ing clin­i­cal trial for
ATM. Of note, engraft­ment has been seen in the uniquely per­mis­
sive set­ting of immunodeficiencies such as bare lym­pho­cyte syn­
drome and severe com­bined immu­no­de­fi­ciency.39,40 We chose
to start our clin­i­cal trial in patients with ATM since they require
in IUTs for sur­vival so there is no addi­tional pro­ce­dural risk from
the trans­plan­ta­tion pro­to­col. If the pro­to­col is safe, it could be
applied to other hemo­glo­bin­op­a­thies or other dis­eases that can
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/592/1851406/592horvei.pdf by guest on 13 December 2021
be treated with HCT, such as Fanconi ane­mia or inborn errors
of metab­o­lism. Given the impor­tance of cell source, dose, and
other var­i­ables in engraft­ment, it is vital to per­form IUHCT in the
con­text of well-designed clin­i­cal tri­als.41
Conclusions and future direc­tions
Although ATM was pre­vi­ously seen as a fatal dis­ease, there are
now mul­ ti­
ple reports of patients who have sur­ vived to birth
with excel­lent qual­ity of life after IUTs. Since these patients still
need chronic trans­fu­sions after birth, it is impor­tant to con­tinue
to develop strat­eg ­ ies for a defin­i­tive cure using HCT. Postnatal
HCT can be cura­tive in ATM and can be safely performed using
RTC, espe­cially with PK model-based dos­ing to main­tain effi­cacy
while min­i­miz­ing tox­ic­ity. It should be performed early in life to
decrease the risk of TRM and other com­pli­ca­tions. In the future,
nongenotoxic anti­body-medi­ated con­di­tion­ing reg­i­mens could
fur­ther improve safety and effi­cacy.42 IUHCT is another prom­is­
ing approach that we are test­ing in a clin­i­cal trial. For IUHCT, it
is impor­tant to use mater­nal stem cells to take advan­tage of the
preexisting tol­er­ance between the mother and fetus. This strat­
egy will likely still require a post­na­tal “boost” trans­plan­ta­tion,
and defin­ing safe and effec­tive pro­to­cols to reduce trans­plant-
related com­pli­ca­tions will be impor­tant. In the future, gene ther­
apy or gene editing approaches may also be devel­ oped for
these patients. Several strat­e­gies devel­oped for β-thal­as­se­mia
could be employed, includ­ing ex vivo lentiviral gene ther­apy to
replace α-glo­bin or gene editing to intro­duce α-glo­bin into a
geno­mic safe har­bor. Given the improved sur­vival of patients
with ATM as a result of in IUTs, increas­ing num­bers of patients
with ATM will require chronic care: improved HCT or gene ther­
apy options could be trans­ for­
ma­tional in enabling defin­i­
tive
cures.
Conflict-of-inter­est dis­clo­sure
Paulina Horvei: no com­pet­ing finan­cial inter­ests to declare.
Tippi MacKenzie: sci­en­tific advi­sory board mem­ber: Acrigen.
Sandhya Kharbanda: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Paulina Horvei: nothing to disclose.
Tippi MacKenzie: nothing to disclose.
Sandhya Kharbanda: nothing to disclose.
Correspondence
Sandhya Kharbanda, Division of Pediatric Allergy, Immunolo-
gy and Bone Marrow Transplantation, UCSF Benioff Children’s
Hospital, University of California, 1975 4th St, San Francisco, CA
94158; e-mail: sandhya​­.kharbanda@ucsf​­.edu.
References
1. Kreger EM, Singer ST, Witt RG, et  al. Favorable out­comes after in utero
trans­fu­sion in fetuses with alpha thal­ as­
se­mia major: a case series and
review of the lit­er­a­ture. Prenat Diagn. 2016;36(13):1242-1249.
2. Feuchtbaum L, Carter J, Dowray S, Currier RJ, Lorey F. Birth prev­a­lence of
dis­or­ders detect­able through new­born screen­ing by race/eth­nic­ity. Genet
Med. 2012;14(11):937-945.
3. Tamary H, Dgany O. Alpha-Thalassemia. 2005 Nov 1 [updated 2020 Oct 1].
GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle;
1993–2021.
4. King AJ, Higgs DR. Potential new approaches to the management of the
Hb Bart’s hydrops fetalis syndrome: the most severe form of α-thalassemia.
Hematology Am Soc Hematol Educ Program. 2018 Nov 30;2018(1):353-360.
5. Songdej D, Babbs C, Higgs DR; Bart’s Hydrops Fetalis Syndrome Inter-
national Consortium. An inter­na­tional reg­is­try of sur­vi­vors with Hb Bart’s
hydrops fetalis syn­drome. Blood. 2017;129(10):1251-1259.
6. Lal A, Goldrich ML, Haines DA, Azimi M, Singer ST, Vichinsky EP. Heteroge-
neity of hemo­glo­bin H dis­ease in child­hood. N Engl J Med. 2011;364(8):710-
718.
7. Amer­i­can College of Obstetricians and Gynecologists Committee on Obstet-
rics. ACOG Practice Bulletin No. 78: hemo­glo­bin­op­a­thies in preg­nancy.
Obstet Gynecol. 2007;109(1):229-237.
8. Society for Maternal-Fetal Medicine; Mari G, Norton M, Stone J, et al. Soci-
ety for Maternal-Fetal Medicine (SMFM) Clinical Guideline #8: the fetus
at risk for ane­mia—diag­no­sis and man­age­ment. Am J Obstet Gynecol.
2015;212(6):697-710.
9. Zwiers C, Lindenburg ITM, Klumper FJ, de Haas M, Oepkes D, Van Kamp IL.
Complications of intra­uter­ine intra­vas­cu­lar blood trans­fu­sion: les­sons
learned after 1678 pro­ce­dures. Ultrasound Obstet Gynecol. 2017;50(2):
180-186.
10. Chan WY, Leung AW, Luk CW, Li RC, Ling AS, Ha SY. Outcomes and morbid-
ities of patients who sur­vive haemoglobin Bart’s hydrops fetalis syn­drome:
20-year ret­ro­spec­tive review. Hong Kong Med J. 2018;24(2):107-118.
11. Zhang HJ, Amid A, Janzen LA, et  al. Outcomes of haemoglobin Bart’s
hydrops fetalis fol­low­ing intra­uter­ine trans­fu­sion in Ontario, Canada. Arch
Dis Child Fetal Neonatal Ed. 2021;106(1):51-56.
12. Chik KW, Shing MM, Li CK, et al. Treatment of hemo­glo­bin Bart’s hydrops
with bone mar­row trans­plan­ta­tion. J Pediatr. 1998;132(6):1039-1042.
13. Gumuscu B, Thompson EI, Grovas AC, Zach TL, Warkentin PI, Coccia PF. Suc-
cessful unre­lated cord blood trans­plan­ta­tion for homo­zy­gous α-thal­as­se­mia.
J Pediatr Hematol Oncol. 2013;35(7):570-572.
14. Zhou X, Ha SY, Chan GC, et al. Successful mismatched sib­ling cord blood
trans­plant in Hb Bart’s dis­ease. Bone Marrow Transplant. 2001;28(1):105-
107.
15. Thornley I, Lehmann L, Ferguson WS, Davis I, Forman EN, Guinan EC.
Homozygous alpha-thal­as­se­mia treated with intra­uter­ine trans­fu­sions and
post­na­tal hema­to­poi­etic stem cell trans­plan­ta­tion. Bone Marrow Trans-
plant. 2003;32(3):341-342.
16. Joshi DD, Nickerson HJ, McManus MJ. Hydrops fetalis caused by homo­zy­
gous alpha-thal­as­se­mia and Rh anti­gen alloimmunization: report of a sur­vi­
vor and lit­er­a­ture review. Clin Med Res. 2004;2(4):228-232.
17. Yi JS, Moertel CL, Baker KS. Homozygous alpha-thal­as­se­mia treated with
intra­uter­ine trans­fu­sions and unre­lated donor hema­to­poi­etic cell trans­
plan­ta­tion. J Pediatr. 2009;154(5):766-768.
18. Chan WYK, Lee PPW, Lee V, et  al. Outcomes of allo­ge­neic trans­plan­ta­
tion for hemo­glo­bin Bart’s hydrops fetalis syn­drome in Hong Kong. Pediatr
Transplant. 2021;25(6):e14037.
19. Elsaid MY, Capitini CM, Diamond CA, Porte M, Otto M, DeSantes KB. Suc-
cessful matched unre­lated donor stem cell trans­plant in hemo­glo­bin
Bart’s dis­ease. Bone Marrow Transplant. 2016;51(11):1522-1523.
20. Pongtanakul B, Sanpakit K, Chongkolwatana V, Viprakasit V. Normal cog­ni­
tive func­tion­ing in a patient with Hb Bart’s hydrops suc­cess­fully cured by
hema­to­poi­etic SCT. Bone Marrow Transplant. 2014;49(1):155-156.
21. Pecker LH, Guerrera MF, et al. Homozygous α-thalassemia: Challenges sur-
rounding early identification, treatment, and cure. Pediatr Blood Cancer.
2017 Jan;64(1):151-155.
22. Anurathapan U, Pakakasama S, Mekjaruskul P, et al. Outcomes of thal­as­se­
mia patients under­go­ing hema­to­poi­etic stem cell trans­plan­ta­tion by using
a stan­dard myeloablative ver­sus a novel reduced-tox­ic­ity con­di­tion­ing
Prakash Singh Shekhawat
reg­i­men according to a new risk strat­i­fi­ca­tion. Biol Blood Marrow Trans-
plant. 2014;20(12):2066-2071.
23. Li C, Mathews V, Kim S, et  al. Related and unre­lated donor trans­plan­ta­
tion for β-thal­as­se­mia major: results of an inter­na­tional sur­vey. Blood Adv.
2019;3(17):2562-2570.
24. Baronciani D, Angelucci E, Potschger U, et al. Hemopoietic stem cell trans­
plan­ta­tion in thal­as­se­mia: a report from the Euro­pean Society for Blood
and Bone Marrow Transplantation Hemoglobinopathy Registry, 2000-
2010. Bone Marrow Transplant. 2016;51(4):536-541.
25. Sheth V, Grisariu S, Avni B, et al. Fludarabine-based reduced tox­ic­ity yet
myeloablative con­di­tion­ing is effec­tive and safe par­tic­u­larly in patients
with high-risk thal­as­se­mia under­go­ing allo­ge­neic trans­plan­ta­tion. Pediatr
Blood Cancer. 2018;65(11):e27312.
26. Bartelink IH, van Reij EM, Gerhardt CE, et  al. Fludarabine and expo­sure-
targeted busul­fan com­pares favor­ably with busul­fan/cyclo­phos­pha­mide-
based reg­i­mens in pedi­at­ric hema­to­poi­etic cell trans­plan­ta­tion: maintaining
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/592/1851406/592horvei.pdf by guest on 13 December 2021
effi­cacy with less tox­ic­ity. Biol Blood Marrow Transplant. 2014;20(3):
345-353.
27. Apsel Winger B, Shukla P, Kharbanda S, et  al. The rela­tion­ship between
busul­fan expo­sure and achieve­ment of sustained donor mye­loid chi­me­rism
in patients with non-malig­nant dis­or­ders. Transplant Cell Ther. 2021;27(3):
258.e1-258.e6.
28. Long-Boyle JR, Green KG, Brunstein CG, et al. High fludarabine expo­sure and
rela­tion­ship with treat­ment-related mor­tal­ity after nonmyeloablative hema­
to­poi­etic cell trans­plan­ta­tion. Bone Marrow Transplant. 2011;46(1):20-26.
29. Ivaturi V, Dvorak CC, Chan D, et  al. Pharmacokinetics and model-based
dos­ing to opti­mize fludarabine ther­apy in pedi­at­ric hema­to­poi­etic cell
trans­plant recip­i­ents. Biol Blood Marrow Transplant. 2017;23(10):1701-1713.
30. Contreras CF, Long-Boyle JR, Shimano KA, et al. Reduced tox­ic­ity con­di­
tion­ing for non­ma­lig­nant hema­to­poi­etic cell trans­plants. Biol Blood Mar-
row Transplant. 2020;26(9):1646-1654.
31. Andreani M, Testi M, Lucarelli G. Mixed chi­me­rism in haemoglobinopa-
thies: from risk of graft rejec­tion to immune tol­er­ance. Tissue Antigens.
2014;83(3):137-146.
32. Andreani M, Testi M, Battarra M, Lucarelli G. Split chi­me­rism between nucle­
ated and red blood cells after bone mar­row trans­plan­ta­tion for haemoglo-
binopathies. Chimerism. 2011;2(1):21-22.
33. Witt R, MacKenzie TC, Peranteau WH. Fetal stem cell and gene ther­apy.
Semin Fetal Neonatal Med. 2017;22(6):410-414.
34. Mold JE, Michaëlsson J, Burt TD, et al. Maternal allo­an­ti­gens pro­mote the
devel­ op­ ment of tolerogenic fetal reg­ u­
la­
tory T cells in utero. Science.
2008;322(5907):1562-1565.
35. Nijagal A, Wegorzewska M, Jarvis E, Le T, Tang Q , MacKenzie TC. Maternal
T cells limit engraft­ment after in utero hema­to­poi­etic cell trans­plan­ta­tion in
mice. J Clin Invest. 2011;121(2):582-592.
36. Peranteau WH, Hayashi S, Abdulmalik O, et al. Correction of murine hemo­
glo­bin­op­a­thies by pre­na­tal tol­er­ance induc­tion and post­na­tal nonmye-
loablative allo­ge­neic BM trans­plants. Blood. 2015;126(10):1245-1254.
37. Cowan MJ, Golbus M. In utero hema­to­poi­etic stem cell trans­plants for
inherited dis­eases. Am J Pediatr Hematol Oncol. 1994;16(1):35-42. Accessed
23 Octo­ber 2021. https:­/­/pubmed­.ncbi­.nlm­.nih­.gov­/7906103­/
38. Hayward A, Ambruso D, Battaglia F, et al. Microchimerism and tol­er­ance
fol­low­ing intra­uter­ine trans­plan­ta­tion and trans­fu­sion for alpha-thal­as­se­mia
-1. Fetal Diagn Ther. 1998;13(1):8-14.
39. Touraine JL, Raudrant D, Royo C, et al. In-utero trans­plan­ta­tion of stem cells
in bare lym­pho­cyte syn­drome. Lancet. 1989;1(8651):1382.
40. Flake AW, Roncarolo MG, Puck JM, et  al. Treatment of X-linked severe
com­bined immu­no­de­fi­ciency by in utero trans­plan­ta­tion of pater­nal
bone mar­row. N Engl J Med. 1996;335(24):1806-1810.
41. MacKenzie TC, David AL, Flake AW, Almeida-Porada G. Consensus state­
ment from the first inter­na­tional con­fer­ence for in utero stem cell trans­
plan­ta­tion and gene ther­apy. Front Pharmacol. 2015;6(10 Feb­ru­ary):15.
42. Palchaudhuri R, Saez B, Hoggatt J, et  al. Non-genotoxic con­di­tion­ing
for hema­to­poi­etic stem cell trans­plan­ta­tion using a hema­to­poi­etic-cell-
spe­cific inter­nal­iz­ing immunotoxin. Nat Biotechnol. 2016;34(7):738-745.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000295
Advances in ther­a­pies for α-thal­as­se­mia major  |  599
 THE CHANGING LANDSCAPE OF α- AND β-THALASSEMIA DIAGNOSIS AND TREATMENT

β-Thalassemia: evolving treatment options

beyond transfusion and iron chelation

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/600/1851627/600langer.pdf by guest on 13 December 2021

Arielle L. Langer1 and Erica B. Esrick2
Division of Hematology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA; and 2Division of Hematology/Oncology, Boston
1

Children’s Hospital, Harvard Medical School, Boston, MA

After years of reliance on transfusion alone to address anemia and suppress ineffective erythropoiesis in β-thalassemia,
many new therapies are now in development. Luspatercept, a transforming growth factor–β inhibitor, has demonstrated
efficacy in reducing ineffective erythropoiesis, improving anemia, and possibly reducing iron loading. However, many
patients do not respond to luspatercept, so additional therapeutics are needed. Several medications in development
aim to induce hemoglobin F (HbF): sirolimus, benserazide, and IMR-687 (a phosphodiesterase 9 inhibitor). Another group
of agents seeks to ameliorate ineffective erythropoiesis and improve anemia by targeting abnormal iron metabolism in
thalassemia: apotransferrin, VIT-2763 (a ferroportin inhibitor), PTG-300 (a hepcidin mimetic), and an erythroferrone anti-
body in early development. Mitapivat, a pyruvate kinase activator, represents a unique mechanism to mitigate ineffective
erythropoiesis. Genetically modified autologous hematopoietic stem cell transplantation offers the potential for lifelong
transfusion independence. Through a gene addition approach, lentiviral vectors have been used to introduce a β-globin
gene into autologous hematopoietic stem cells. One such product, betibeglogene autotemcel (beti-cel), has reached
phase 3 trials with promising results. In addition, 2 gene editing techniques (CRISPR-Cas9 and zinc-finger nucleases) are
under investigation as a means to silence BCL11A to induce HbF with agents designated CTX001 and ST-400, respectively.
Results from the many clinical trials for these agents will yield results in the next few years, which may end the era of
relying on transfusion alone as the mainstay of thalassemia therapy.

LEARNING OBJECTIVES
• Understand the strengths and limitations of luspatercept for the treatment of TDT and NTDT
• Understand the potential benefits and toxicity of genetically modified autologous HSCT for TDT
• Describe therapies under development for the treatment of TDT and NTDT

After many years without novel disease-modifying ther-
apeutics, numerous agents are now in development for
β-thalassemia. We review therapies that have been recently
approved or are in development for transfusion-dependent
thalassemia (TDT) and non-transfusion-dependent thalasse-
mia (NTDT) β-thalassemia using 4 patient cases (Table 1).
of approximately 9.5g/dL is indicated because of end-organ
damage. However, she has been unable to tolerate the
transfusion volumes, as attempts at transfusion frequently
trigger a heart failure exacerbation and acute worsening of
her pulmonary pressures. She is started on luspatercept,
which leads to a rise in hemoglobin to 9.2 to 10.1g/dL with
drops below this range only in the setting of acute illness.

Luspatercept is a recombinant fusion protein that blocks

CLINICAL CASE 1: USE OF LUSPATERCEPT
A 54-year-old woman with congestive heart failure, parox-
ysmal atrial fibrillation on warfarin, and pulmonary hyper-
tension due to β-thalassemia intermedia has a hemoglobin
ranging from 7.5 to 8.5g/dL without packed red blood cell
(RBC) transfusion. Transfusion to maintain hemoglobin nadir
transforming growth factor-β (TGF-β) superfamily ligands
and promotes late-stage erythroid maturation, resulting
in effective reduction in transfusion requirements in some
patients with TDT.1 In a randomized, placebo-controlled
trial, 21.4% of patients met the primary end point of at
least a 33% reduction in transfusion volume during weeks

600 | Hematology 2021 | ASH Education Program

Table 1. Current lim­i­ta­tions of thal­as­se­mia care
Limitation
Regular trans­fu­sion appoint­ments
Iron over­load
Iron che­la­tor tox­ic­ity
Lack of allo­ge­neic stem cell donor
Potential solu­tions
Reduce trans­fu­sion require­ment
Stem cell trans­plan­ta­tion
Reduce trans­fu­sion require­ment
Block intes­ti­nal iron absorp­tion
Escalate che­la­tion
Reduce trans­fu­sion require­ment
Block intes­ti­nal iron absorp­tion
Additional che­la­tion options
Genetically mod­i­fied autol­o­gous
trans­plan­ta­tion
Nontransplant ther­a­pies that
dence of throm­bo­sis may com­pound other risk fac­tors, such as
prior sple­nec­tomy, and this may affect the risk-ben­e­fit ratio of
ini­ti­at­ing luspatercept. In case 1, the patient was already on life­
long anticoagulation because of atrial fibril­la­tion, so this risk was
miti­gated. Finally, published data pri­mar­ily address the TDT pop­
u­la­tion; forth­com­ing results will help clar­ify how best to tai­lor the
use of luspatercept for patients with NTDT.
Further fol­low-up will be required to deter­mine if luspater-
cept may improve iron over­load, although pre­lim­i­nary results
are prom­is­ing.6 Early find­ings that fer­ri­tin and liver iron con­cen­
tra­tion (LIC) improved in both those who met the pri­mary end
point and those who did not sug­gest that reduced iron load­
ing from the gut is likely playing a role, rather than just reduced

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/600/1851627/600langer.pdf by guest on 13 December 2021

pre­clude desir­abil­ity of trans­plant
13 through 24 com­pared with 4.5% receiv­ing pla­cebo (P < .001).
This equated to a least mean square dif­fer­ence in trans­fu­sion
bur­den of 1.35 fewer RBC trans­fu­sions in weeks 13 through 24
in the TDT pop­u­la­tion. Follow-up data show trans­fu­sion reduc­
tion con­tin­ued at 48 weeks and up to 4.8 years into treat­ment.2,3
Common side effects included head­ache, myal­gia, and bone
pain. Although rare, 8 of 223 (3.6%) patients in the luspatercept
arm had throm­bo­em­bo­lism com­pared with 1 of 109 patients in
the pla­cebo arm (Table 2).1
In the NTDT pop­u­la­tion, data from a phase 1/2 nonrandomized
trial showed an increase in hemo­glo­bin of at least 1.5 g/dL in 58%
of patients receiv­ing higher dose lev­els.4 A ran­dom­ized, pla­cebo-
con­trolled trial for luspatercept in patients with NTDT has fin­ished
accrual, and pre­lim­i­nary results showed hemo­glo­bin of at least
1.5 g/dL in 52.1% in the luspatercept arm com­pared with 0% in
the pla­cebo arm and no throm­bo­em­bolic events in either arm
(NCT03342404).5
Case 1 dem­on­strates sev­eral strengths and lim­i­ta­tions of luspa-
tercept. This med­i­ca­tion is most likely to be use­ful in patients for
whom a mod­er­ate increase in hemo­glo­bin or a mod­er­ate reduc­
tion in trans­fu­sion vol­ume would be impactful. This may be the
case because of dif­fi­culty with vol­ume sta­tus, as was the case in
this patient; for patients with alloimmunization for whom obtain-
ing RBC units is more dif­fi­cult; or for patients whose qual­ity of
life would improve by spend­ing less time at trans­fu­sion appoint­
ments. The risk of throm­bo­em­bo­lism should be con­sid­ered when
ini­ti­at­ing this med­i­ca­tion; although low in clin­i­cal tri­als, the inci­
trans­fu­sion bur­den. If this bears out, it would rep­re­sent a sig­
nif­i­cant addi­tional ben­e­fit, par­tic­u­larly for indi­vid­u­als in poor
iron bal­ance. Although luspatercept rep­re­sents an excit­ing new
ther­a­peu­tic option and is the first novel approved ther­apy in
many years, cau­tion is merited in set­ting expec­ta­tions, as most
patients will not be ­able to forgo trans­fu­sion, and increas­ing the
inter­val between trans­fu­sions is often not prac­ti­cally fea­si­ble or
appeal­ing since vis­its every 3 weeks are required for this sub­cu­
ta­ne­ous med­i­ca­tion. As such, a great unmet need remains.
CLINICAL CASE 2: AUTOLOGOUS GENETIC THERAPIES
A 17-year-old boy with TDT on a sta­ble trans­fu­sion reg­i­men and
in good iron bal­ance wishes to con­sider cura­tive intent ther­apy.
He is con­sid­er­ing his goals for adult­hood, includ­ing attend­ing
col­lege that is not near a large med­i­cal cen­ter, and is inter­ested
in avoiding fre­quent med­i­cal and trans­fu­sion appoint­ments. He
has no avail­­able donor for allo­ge­neic stem cell trans­plan­ta­tion.
He opts to enroll in a clin­i­cal trial of genet­i­cally mod­i­fied autol­
o­gous hema­to­poi­etic stem cell trans­plan­ta­tion (HSCT) using a
len­ti­vi­rus vec­tor.
Patients may desire a cura­tive intent ther­apy given the long-
term impact on mor­bid­ity, mor­tal­ity, and qual­ity of life asso­ci­
ated with chronic trans­fu­sion ther­apy. Until recently, the only
cura­tive option was allo­ge­neic HSCT. Allogeneic HSCT has pro­
vided a cure to many patients with TDT, with the best out­comes

Table 2. Luspatercept con­sid­er­ations1-6

Subcutaneous admin­is­tra­tion
Reduction of trans­fu­sion in TDT
Increase in hemo­glo­bin in NTDT
Risk of throm­bo­em­bo­lism
Impact on iron load­ing
Prakash Singh Shekhawat
Strengths Limitations
Does not require intra­ve­nous access Requires infu­sion cen­ter visit every 3 weeks
21.4% had >33% reduc­tion Most patients with­out sig­nif­i­cant reduc­tion
7.6% had >50% reduc­tion Majority still have vis­its every 3 weeks
52.1% had 1.5-g/dL increase Many patients with­out sig­nif­i­cant increase
Requires increased appoint­ment bur­den
Not yet approved by reg­u­la­tory agencies
Rates low: 8 in 223 (3.6%) in trial Patients with thal­as­se­mia already at increased
risk of throm­bo­em­bo­lism, espe­cially if prior
sple­nec­tomy
Preliminary data showed lower fer­ri­tin and LIC Additional data needed to ver­ify find­ings
in some patients, includ­ing some who did
not meet pri­mary end point
Evolving ther­a­pies in β-thal­as­se­mia  |  601
 Table 3. Gene ther­apy tri­als for β-thal­as­se­mia

602  |  Hematology 2021  |  ASH Education Program

Also in devel­op­ment for
Therapy Gene mod­i­fi­ca­tion tech­nique Genetic mech­a­nism Phase of devel­op­ment sickle cell dis­ease? Comments
Betibeglogene autotemcel Lentiviral vec­tor Insertion of mod­i­fied β-glo­bin Phase 3 tri­als Yes Reopened after tem­po­rary halt due
(beti-cel; for­merly to AML cases in sickle cell patients
BB305)8-11,17,18
GLOBE12 Lentiviral vec­tor Insertion of human β-glo­bin Phase 1/2 trial com­pleted No Intrabone admin­is­tra­tion
CTX00114,15 CRISPR-Cas9 Targets BCL11A enhancer → Phase 1/2 trial Yes
HbF induc­tion
ST-40016 ZFN Targets BCL11A enhancer → Phase 1/2 trial No
HbF induc­tion

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/600/1851627/600langer.pdf by guest on 13 December 2021

reported for matched sib­ling donors and recip­i­ents who under­
went trans­plan­ta­tion in teen­age years.7 However, the risk of
acute and chronic com­pli­ca­tions and the lack of avail­­able donors
have lim­ited its use.
Ex vivo mod­i­fi­ca­tion of CD34+ hema­to­poi­etic stem cells using
dif­fer­ent tech­niques has opened the pos­si­bil­ity of autol­o­gous
HSCT as a cura­tive ther­apy, thereby elim­i­nat­ing the need for a
donor as well as the risks spe­cific to allo­ge­neic HSCT, such as
graft-ver­sus-host dis­ease. In this con­text, genet­i­cally mod­i­fied
autol­o­gous HSCT is an area of active research with sig­nif­i­cant
advance­ments in the past few years.
One ex vivo gene ther­apy, betibeglogene autotemcel (beti-
cel; for­merly LentiGlobin), uses a lentiviral vec­tor to add a mod­
i­fied β-glo­bin gene into CD34+ cells for infu­sion by autol­o­gous
HSCT after myeloablative con­di­tion­ing. In phase 1/2 tri­als, beti-
cel treat­ment led to trans­fu­sion inde­pen­dence (TI) in 12 of 13
non-β0/β0 geno­type and 3 of 9 β0/β0 geno­type patients with
TDT, or 68% over­all.8 The remaining patients had a reduc­tion in
trans­fu­sion vol­ume. Toxicity was pri­mar­ily related to busul­fan
used for HSCT con­ di­tion­
ing. The orig­ i­
nal beti-cel trial estab-
lished the poten­tial for genet­i­cally mod­i­fied autol­o­gous HSCT,
although most patients with the β0/β0 geno­type did not achieve
TI. Preliminary data from 2 phase 3 tri­als of beti-cel for β0/β0
(NCT03207009) and non-β0/β0 (NCT02906202) patients have
dem­on­strated TI at >12 months of fol­low-up in 88% of 34 eval-
uable patients, includ­ing 24 of 27 non-β0/β0 and 6 of 7 β0/β0
patients, with no addi­tional toxicities noted.9 Preliminary reports
of long-term fol­low up of 44 patients with >2 years of fol­low-up
after beti-cel treat­ment (23-76 months) showed a dura­ble hemo­
glo­bin response, reduc­tion in iron bur­den, and favor­able safety
pro­file.10 Initial eval­u­a­tion of 27 pedi­at­ric patients treated with
beti-cel, 16 of whom were under 12 years old, showed a com­pa­
ra­ble TI rate of 84.6% (Table 3).11
Another phase 1/2 trial uses a lentiviral vec­tor (the GLOBE
vec­tor) to add a β-glo­bin gene to autol­o­gous hema­to­poi­etic
stem cells, with intrabone admin­is­tra­tion of the trans­duced cells.
Three of 4 evaluable pedi­at­ric patients achieved TI, whereas 3
adult patients had a reduced trans­fu­sion bur­den with­out achiev­
ing TI.12 At pres­ent, a sub­se­quent trial is not reg­is­tered.
In con­trast to the gene addi­tion approach, another strat­egy
for autol­o­gous genetic ther­a­pies for β-hemo­glo­bin­op­a­thies is to
decrease BCL11A expres­sion in order to induce expres­sion of γ-
glo­bin and hemo­glo­bin F (HbF). The first trial targeting BCL11A
uses a lentiviral vec­tor, BCH-BB694, to intro­duce short hair­pin
RNA to decrease BCL11A expres­sion.13 Six patients with sickle cell
dis­ease who received BCH-BB694 dem­on­strated safety and fea­
si­bil­ity and had sub­stan­tial induc­tion of HbF. This vec­tor has not
been eval­u­ated in patients with TDT. Trials are cur­rently open for
patients with thal­as­se­mia, in whom a gene editing tech­nique,
either CRISPR-Cas9 or zinc-fin­ger nucle­ase (ZFN), is employed
to dis­rupt an ery­throid enhancer of BCL11A. CTX001 is a CRISPR-
Cas9–mod­i­fied autol­o­gous HSCT prod­uct being inves­ti­gated
in TDT as well as sickle cell dis­ease (NCT03655678). Preliminary
results from the first 10 patients treated with CTX001 for TDT
showed sub­stan­tial HbF induc­tion, broad dis­tri­bu­tion of HbF,
and achieve­ment of TI in all­patients.14,15 Toxicity was related to
busul­fan con­di­tion­ing. A phase 1/2 study of ST-400, an autol­o­
gous HSCT prod­uct in which the BCL11A enhancer is disrupted
by a ZFN, is cur­rently under way (NCT03432364), but only very
pre­lim­i­nary data on the first 2 patients have been reported.16
Prakash Singh Shekhawat
Case 2 dem­on­strates the clin­i­cal demand for cura­tive intent
ther­apy and its poten­tial impact on qual­ity of life as well as
health. Although genet­i­cally mod­i­fied autol­o­gous HSCT may
offer a more fea­si­ble and less toxic option for cura­tive intent
ther­apy com­pared with allo­ge­neic HSCT, lim­i­ta­tions are likely to
remain. HSCT of any kind is a costly ther­apy and not likely to be
avail­­able in low-resource set­tings. In addi­tion to mon­e­tary costs,
under­go­ing autol­o­gous HSCT requires a sig­nif­i­cant time com­
mit­ment and inter­rup­tion of other life events. The poten­tial for
reduced fer­til­ity or infer­til­ity due to expo­sure to myeloablative
alkylator con­di­tion­ing is an impor­tant con­sid­er­ation for many
patients. Finally, 2 cases of acute mye­ loid leu­
ke­
mia (AML) in
patients with sickle cell dis­ease treated with beti-cel have raised
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/600/1851627/600langer.pdf by guest on 13 December 2021
con­cerns.17,18 Evaluation sug­gests that nei­ther case was caused
by inser­tional muta­gen­e­sis. To date, no patients with thal­as­se­
mia have devel­ oped myelodysplastic syn­ drome or AML after
genet­i­cally mod­i­fied autol­o­gous HSCT, but addi­tional data will
be nec­es­sary before final con­clu­sions may be drawn about what
fac­tors influ­ence malig­nancy risk.
CLINICAL CASE 3: TARGETING ABNORMAL IRON
METABOLISM
A 34-year-old man with NTDT is concerned about ongo­ing iron
over­load. His hemo­glo­bin aver­ages 8.4 g/dL, and he receives
trans­fu­sion a few times per year. His LIC is esti­mated at 8 mg
per gram of dry weight on mag­netic res­o­nance imag­ing. In
addi­tion to inten­si­fy­ing his iron che­la­tion, he inquires whether
there are any ther­a­pies that may raise his hemo­glo­bin or reduce
his iron absorp­tion. He is open to par­tic­i­pa­tion in clin­i­cal tri­als.
Currently, there are no established ther­a­pies to address this
patient’s ane­mia. However, sev­eral med­i­ca­tions in devel­op­ment
may improve ane­mia in patients with NTDT that also tar­get iron
metab­o­lism. Apotransferrin has been shown to upregulate hep-
cidin and downregulate trans­ fer­
rin recep­ tor 1.19,20 This leads to
decreased iron absorp­tion from the gut,19 as well as poten­tially less
car­diac iron load­ing.21 The cor­rec­tion of path­o­logic iron metab­o­
lism led to more effec­tive eryth­ro­poi­e­sis in mouse mod­els.20 These
pre­clin­i­cal find­ings have spurred a phase 2 trial of intra­ve­nous apo-
transferrin every 2 weeks in patients with β-thal­as­se­mia intermedia
that seeks to raise hemo­glo­bin and reduce trans­fu­sion require­
ments (NCT03993613).
Similarly, a ferroportin inhib­i­tor (VIT-2763) has begun a phase 2
trial (NCT04364269) after dem­on­strat­ing improve­ment in ane­mia
and dysregulated iron metab­o­lism in a mouse model of thal­as­
se­mia.22 Like the apotransferrin trial, the VIT-2753 is focused on
cor­rec­tion of ane­mia but includes sec­ond­ary out­comes mea­sures
of iron overloading. Notably, VIT-2753 is admin­is­tered orally. In
con­sid­er­ing treat­ment for patients with NTDT, this is par­tic­u­larly
impactful for qual­ity of life, as many of these patients do not oth­
er­wise have recur­ring infu­sion cen­ter appoint­ments (Table 4).
Hepcidin itself has also been a tar­get for drug devel­op­ment,
because increas­ing hepcidin has been shown to both reduce
iron absorp­tion and ame­lio­rate inef­fec­tive eryth­ro­poi­e­sis.23 A
hepcidin mimetic, PTG-300, has com­pleted accrual for a phase 2
study, includ­ing both NTDT and TDT arms (NCT03802201). A
Evolving ther­a­pies in β-thal­as­se­mia  |  603
 Table 4. Novel med­i­ca­tions for β-thal­as­se­mia

Route and fre­quency Current tar­get patient In devel­op­ment for

Medication Mechanism of action of admin­is­tra­tion Phase of devel­op­ment pop­u­la­tion sickle cell dis­ease? Comments
1,5
Luspatercept TGF-β inhib­i­tor Subcutaneous, every EMA and FDA approved NTDT, TDT No Full data for NTDT pend­ing;
3 weeks in use for MDS
Sotatercept36,37 TGF-β inhib­i­tor Subcutaneous, every Development halted NTDT, TDT No In devel­op­ment for pul­mo­nary
3 weeks hyper­ten­sion

604  |  Hematology 2021  |  ASH Education Program

Sirolimus25-28 mTOR inhib­i­tor; HbF Oral, daily Phase 2 tri­als TDT No In use for other dis­or­ders
induc­tion
Benserazide31,32 HbF induc­tion Oral, daily Phase 1 trial NTDT Yes In use for Parkinson dis­ease
IMR-68729,30 PDE-9 inhib­i­tor; HbF Oral, daily Phase 2 trial NTDT, TDT Yes
induc­tion
Apotransferrin19-21 Hepcidin upregulation Intravenous, every Phase 2 trial NTDT No
2 weeks
VIT-276322 Ferroportin inhib­i­tor Oral, one or twice daily Phase 2 trial NTDT Yes
PTG-30023 Hepcidin mimetic Subcutaneous, once Phase 2 tri­als NTDT, TDT No In devel­op­ment for
weekly hemo­chro­ma­to­sis and
poly­cy­the­mia vera
Mitapivat33-35 Pyruvate kinase acti­va­tor Oral, twice daily Phase 3 tri­als NTDT, TDT Yes In devel­op­ment for pyru­vate
kinase defi­ciency
Ruxolitinib38 JAK 1/2 inhib­i­tor Oral, twice daily Development halted TDT No Spleen size reduc­tion;
no change in trans­fu­sion
EMA, Euro­pean Medicines Agency; FDA, Food and Drug Administration; JAK, Janus-asso­ci­ated kinase; MDS, myelodysplastic syn­drome; mTOR, mech­a­nis­tic tar­get of rapamycin.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/600/1851627/600langer.pdf by guest on 13 December 2021

recent mouse study of an erythroferrone anti­ body show­ ing
improved ane­mia as well as iron homeo­sta­sis pro­vi­des another
poten­tial ther­a­peu­tic tar­get in the iron reg­u­la­tion path­way.24
Case 3 empha­sizes the large unmet need of ther­a­pies for
NTDT and the appeal of a ther­apy that may also address non-
transfusional hemosiderosis in these patients. These tri­als not
only high­light the role of altered iron metab­o­lism in inef­fec­tive
eryth­ro­poi­e­sis but also pro­vide sev­eral prom­is­ing ther­a­peu­tic
tar­gets for ame­lio­rat­ing ane­mia.
CLINICAL CASE 4: FETAL HEMOGLOBIN INDUCTION
AND OTHER MECHANISMS
A 21-year-old woman with TDT and in good iron bal­ance recently
tried luspatercept with only a min­i­mal reduc­tion in trans­fu­sion
bur­den. She inquires if there are other ther­ a­
pies that might
reduce her trans­fu­sion require­ment. She is not inter­ested in allo­
ge­neic or genet­i­cally mod­i­fied autol­o­gous HSCT, as she is wor­
ried about the upfront risk and dis­rup­tion in her life, because
she is doing quite well over­all on her cur­rent trans­fu­sion and
che­la­tion reg­i­men.
Several agents cur­rently in clin­i­cal tri­als could pro­vide good
options for this patient. Sirolimus, the mech­a­nis­tic tar­get of
rapamycin inhib­i­tor, is being inves­ti­gated as a method of increas­
ing HbF expres­sion. Sirolimus appears to upregulate the expres­
sion of HbF in ery­throid cell cul­tures derived from patients with
β-thal­as­se­mia, as well as sickle cell patients, and may increase
clear­ance of α-glo­bin in RBC pre­cur­sors.25-28 The lat­ter find­ing
may have the poten­tial to reduce inef­fec­tive eryth­ro­poi­e­sis sep­
a­rate from HbF induc­tion. In this con­text, 2 ongo­ing phase 2 tri­als
of sirolimus in patients with TDT (NCT03877809, NCT04247750)
will track mark­ers of inef­fec­tive eryth­ro­poi­e­sis in addi­tion to the
pri­mary end point of HbF induc­tion.
Another path­way for HbF induc­tion may be phos­pho­di­es­ter­
ase 9 inhi­bi­tion with IMR-687.29,30 A phase 2 trial of IMR-687 for both
patients with TDT and NTDT is cur­rently under way (NCT04411082).
Benserazide, which is cur­rently used in com­bi­na­tion with levo­
dopa for the treat­ment of Parkinson dis­ease, showed prom­is­
ing induc­tion of HbF, although the study was not of thal­as­se­mia
mod­els,31 but it did not induce HbF in patients on this med­i­ca­tion
for Parkinson dis­ease.32 A phase 1 trial of benserazide in patients
with NTDT is ongo­ing (NCT04432623).
Mitapivat (for­merly AG-348), an allo­ste­ric acti­va­tor RBC-spe­cific
pyru­ vate kinase, rep­ re­
sents a dis­ tinct and novel mech­ a­
nism.
RBC-spe­cific pyru­vate kinase acti­va­tion increases aden­o­sine tri­
phos­phate syn­the­sis, as well as reduces the pro­duc­tion of reac­tive
oxy­gen spe­cies and con­cen­tra­tion of 2,3-diphosphoglycerate.33
Preliminary results of a phase 2 trial of mitapivat in α and β NTDT
showed a response, defined as an increase in hemo­glo­bin of at
least 1 g/dL in 16 of 20 patients, includ­ing all 5 patients with α-
thal­as­se­mia.34 Phase 3 stud­ies of mitapivat in TDT (NCT04770779)
and NTDT (NCT04770753) patients are under way.35 Like the phase
2 trial, these tri­als include patients with α-thal­as­se­mia, a patient
pop­ u­la­
tion that is under­ served but out­ side the scope of this
review.
Prakash Singh Shekhawat
Case 4 is a reminder that even in well-man­aged patients,
there is sig­nif­i­cant room for improve­ment in care. All 4 patients
presented high­light the nuanced needs of patients with thal­as­
se­mia and the impor­tance of tai­lor­ing ther­a­peu­tic choices to
the indi­vid­ual as new treat­ments emerge. Although the med­i­
ca­tions and inter­ven­tions reviewed here are grounded in strong
pre­clin­i­cal research on thal­as­se­mia, the results of these tri­als are
needed to under­stand which agents will be clin­i­cally impact-
ful and which agents will not be fruit­ful for patient care. If early
stud­ies are borne out, it is likely that both patients with TDT
and NTDT will have sev­eral options beyond trans­fu­sion and iron
che­la­tion in the com­ing years.
Conflict-of-inter­est dis­clo­sure
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/600/1851627/600langer.pdf by guest on 13 December 2021
Arielle L. Langer: no com­pet­ing finan­cial inter­ests to declare.
Erica B. Esrick: steering com­ mit­tee (con­ sul­ting) for blue­bird
bio and research funding to insti­tu­tion from Celgene (site for
luspatercept trial).
Off-label drug use
Arielle L. Langer: nothing to disclose. The majority of the drugs
discussed are not yet approved.
Erica B. Esrick: nothing to disclose. The majority of the drugs
discussed are not yet approved.
Correspondence
Erica B. Esrick, Division of Hematology/Oncology, Bos­ton Children’s
Hospital, 300 Longwood Avenue, Bos­ton, MA 02115; e-mail: erica​
­.esrick@chil­drens​­.harvard​­.edu.
References
1. Cappellini MD, Viprakasit V, Taher AT, et al; BELIEVE Investigators. A phase 3
trial of luspatercept in patients with trans­fu­sion-depen­dent β-thal­as­se­mia.
N Engl J Med. 2020;382(13):1219-1231.
2. Taher AT, Porter JB, Hermine O, et al. Fewer red blood cell trans­fu­sion units
and vis­its across base­line trans­fu­sion bur­den lev­els in patients with beta-
thal­as­se­mia treated with luspatercept in the phase 3 believe trial. Paper
presented at the 26th Congress of the Euro­pean Hematology Association
(EHA), June 9-17, 2021. Abstract 325024.
3. Piga A, Tartaglione I, Gamberini R, et  al. Long-term effi­cacy and safety
out­comes in the phase 2 study of luspatercept in Β-thal­as­se­mia. Paper
presented at the 26th Congress of the Euro­pean Hematology Association
(EHA), June 9-17, 2021. Abstract EP1304.
4. Piga A, Perrotta S, Gamberini MR, et al. Luspatercept improves hemo­glo­
bin lev­els and blood trans­fu­sion require­ments in a study of patients with
β-thal­as­se­mia. Blood. 2019;133(12):1279-1289.
5. Taher AT, Cappellini MD, Kattamis A, et al. The BEYOND study: results
of a phase 2, dou­ble-blind, ran­dom­ized, pla­cebo-con­trolled mul­ti­cen­ter
study of luspatercept in adult patients with non-tranfusion depen­dent
beta-thal­as­se­mia. Paper presented at the 26th Congress of the Euro­pean
Hematology Association (EHA), June 9-17, 2021. Abstract S101.
6. Hermine O, Cappellini MD, Taher AT, et al. Longitudinal effect of luspater-
cept treat­ment on iron over­load and iron che­la­tion ther­apy (ICT) in adult
patients (Pts) with β-thal­as­se­mia in the believe trial. Blood. 2020;136(suppl 1):
47-48.
7. Baronciani D, Angelucci E, Potschger U, et al. Hemopoietic stem cell trans­
plan­ta­tion in thal­as­se­mia: a report from the Euro­pean Society for Blood
and Bone Marrow Transplantation Hemoglobinopathy Registry, 2000-
2010. Bone Marrow Transplant. 2016;51(4):536-541.
8. Thompson AA, Walters MC, Kwiatkowski J, et al. Gene ther­apy in patients
with trans­fu­sion-depen­dent β-thal­as­se­mia. N Engl J Med. 2018;378(16):1479-
1493.
9. Locatelli F, Kwiatkowski JL, Walters MC, et al. Betibeglogene autotemcel in
patients with trans­fu­sion-depen­dent β-thal­as­se­mia: updated results from
HGB-207 (NORTHSTAR-2) and HGB-212 (NORTHSTAR-3). Paper presented
Evolving ther­a­pies in β-thal­as­se­mia  |  605
 at the 26th Congress of the Euro­pean Hematology Association (EHA), June
9-17, 2021. Abstract S266.
10. Yannaki E, Locatelli F, Kwiatkowski JL, et  al. Betibeglogene autotemcel
gene ther­apy for the treat­ment of trans­fu­sion-depen­dent β-thal­as­se­mia:
updated long-term effi­cacy and safety results. Paper presented at the 26th
Congress of the Euro­pean Hematology Association (EHA), June 9-17, 2021.
Abstract S267.
11. Kulozik AE, Thuret I, Kwiatkowski JL, et al. Interim results of betibeglogene
autotemcel gene ther­apy in pedi­at­ric patients with trans­fu­sion-depen­dent
Β-thal­as­se­mia (TDT) treated in the phase 3 NORTHSTAR-2 and NORTHSTAR-3
stud­ies. Paper presented at the 26th Congress of the Euro­pean Hematology
Association (EHA), June 9-17, 2021. Abstract EP1301.
12. Marktel S, Scaramuzza S, Cicalese MP, et al. Intrabone hema­to­poi­etic stem
cell gene ther­apy for adult and pedi­at­ric patients affected by trans­fu­sion-
depen­dent ß-thal­as­se­mia. Nat Med. 2019;25(2):234-241.
13. Esrick EB, Lehmann LE, Biffi A, et al. Post-tran­scrip­tional genetic silenc­
ing of BCL11A to treat sickle cell dis­ease. N Engl J Med. 2021;384(3):205-
215.
14. Frangoul H, Altshuler D, Cappellini MD, et al. CRISPR-Cas9 gene editing for
sickle cell dis­ease and β-thal­as­se­mia. N Engl J Med. 2021;384(3):252-260.
15. Locatelli F, Ailinca-Luchian S, Bobruff Y, et al. CTX001 for trans­fu­sion-depen­dent
β-thal­as­se­mia: safety and effi­cacy results from the ongo­ing CLIMB THAL-111
study of autol­o­gous CRISPR-CAS9-mod­i­fied CD34+ hema­to­poi­etic stem and
pro­gen­i­tor cells. Paper presented at the 26th Congress of the Euro­pean Hema-
tology Association (EHA), June 9-17, 2021. Abstract EP733.
16. Smith AR, Schiller GJ, Vercellotti GM, et al. Preliminary results of a phase 1/2
clin­i­cal study of zinc fin­ger nucle­ase-medi­ated editing of BCL11A in autol­
o­gous hema­to­poi­etic stem cells for trans­fu­sion-depen­dent beta thal­as­se­
mia. Blood. 2019;134(suppl 1):3544-3544.
17. Hsieh MM, Bonner M, Pierciey FJ, et al. Myelodysplastic syn­drome unre­lated
to lentiviral vec­tor in a patient treated with gene ther­apy for sickle cell dis­
ease. Blood Adv. 2020;4(9):2058-2063.
18. blue­bird bio Provides updated find­ings from reported case of acute mye­loid
leu­ke­mia (AML) in LentiGlobin for Sickle Cell Disease (SCD) Gene Therapy
Program—blue­bird bio, Inc. Accessed 6 July 2021.
19. Gelderman MP, Baek JH, Yalamanoglu A, et al. Reversal of hemo­chro­ma­to­sis
by apotransferrin in non-trans­fused and trans­fused Hbbth3/+ (het­ero­zy­gous
B1/B2 glo­bin gene dele­tion) mice. Haematologica. 2015;100(5):611-622.
20. Li H, Choesang T, Bao W, et al. Decreasing TfR1 expres­sion reverses ane­mia
and hepcidin sup­pres­sion in β-thal­as­se­mic mice. Blood. 2017;129(11):1514-
1526.
21. Garbowski MW, Evans P, Vlachodimitropoulou E, Hider R, Porter JB.
Residual eryth­ro­poi­e­sis pro­tects against myo­car­dial hemosiderosis in
trans­fu­sion-depen­dent thal­as­se­mia by low­er­ing labile plasma iron via tran­
sient gen­er­a­tion of apotransferrin. Haematologica. 2017;102(10):1640-1649.
22. Manolova V, Nyffenegger N, Flace A, et al. Oral ferroportin inhib­i­tor ame­
lio­rates inef­fec­tive eryth­ro­poi­e­sis in a model of β-thal­as­se­mia. J Clin Invest.
2019;130(1):491-506.
23. Schmidt PJ, Flem­ing MD. Modulation of hepcidin as ther­apy for pri­mary and
sec­ond­ary iron over­load dis­or­ders: pre­clin­i­cal mod­els and approaches.
Hematol Oncol Clin North Am. 2014;28(2):387-401.
24. Arezes J, Foy N, McHugh K, et  al. Antibodies against the erythroferrone
N-ter­mi­nal domain pre­vent hepcidin sup­pres­sion and ame­lio­rate murine
thal­as­se­mia. Blood. 2020;135(8):547-557.
606  |  Hematology 2021  |  ASH Education Program
25. Fibach E, Bianchi N, Borgatti M, et al. Effects of rapamycin on accu­mu­la­
tion of alpha-, beta- and gamma-glo­bin mRNAs in ery­throid pre­cur­sor cells
from beta-thal­as­sae­mia patients. Eur J Haematol. 2006;77(5):437-441.
26. Mischiati C, Sereni A, Lampronti I, et al. Rapamycin-medi­ated induc­tion of
gamma-glo­bin mRNA accu­mu­la­tion in human ery­throid cells. Br J Haema-
tol. 2004;126(4):612-621.
27. Pecoraro A, Troia A, Calzolari R, et al. Efficacy of rapamycin as inducer of Hb
F in pri­mary ery­throid cul­tures from sickle cell dis­ease and β-thal­as­se­mia
patients. Hemoglobin. 2015;39(4):225-229.
28. Lechauve C, Keith J, Khandros E, et  al. The autophagy-acti­vat­ing kinase
ULK1 medi­ates clear­ance of free α-glo­bin in β-thal­as­se­mia. Sci Transl Med.
2019;11(506):eaav4881.
29. Almeida CB, Traina F, Lanaro C, et al. High expres­sion of the cGMP-spe­cific
phos­pho­di­es­ter­ase, PDE9A, in sickle cell dis­ease (SCD) and the effects
of its inhi­bi­tion in ery­throid cells and SCD neu­tro­phils. Br J Haematol.
2008;142(5):836-844.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/600/1851627/600langer.pdf by guest on 13 December 2021
30. McArthur JG, Svenstrup N, Chen C, et al. A novel, highly potent and selec­
tive phos­pho­di­es­ter­ase-9 inhib­i­tor for the treat­ment of sickle cell dis­ease.
Haematologica. 2020;105(3):623-631.
31. Pace BS, Perrine S, Li B, et  al. Benserazide race­mate and enan­tio­mers
induce fetal glo­bin gene expres­sion in vivo: stud­ies to guide clin­i­cal devel­
op­ment for beta thal­as­se­mia and sickle cell dis­ease. Blood Cells Mol Dis.
2021;89:102561.
32. Santos MEHP, Olops L, Vendrame F, et al. Benserazide as a poten­tial novel
fetal hemo­glo­bin inducer: an obser­va­tional study in non-car­ri­ers of hemo­
glo­bin dis­or­ders. Blood Cells Mol Dis. 2021;87:102511.
33. Kung C, Hixon J, Kosinski PA, et al. AG-348 enhances pyru­vate kinase activ­
ity in red blood cells from patients with pyru­vate kinase defi­ciency. Blood.
2017;130(11):1347-1356.
34. Kuo KHM, Layton DM, Lal A, et al. Results from a phase 2, open-label, mul­ti­
cen­ter study of the oral pyru­vate kinase activatorr mitapivat in adults with
non–trans­fu­sion-depen­dent alpha- or beta-thal­as­se­mia. Paper presented
at the 26th Congress of the Euro­pean Hematology Association (EHA), June
9-17, 2021. Abstract S267.
35. Kuo KHM, Layton DM, Al-Samkari H, et al. ENERGIZE and ENERGIZE-T: two
phase 3, ran­dom­ized, dou­ble-blind, pla­cebo-con­trolled stud­ies of mitapi-
vat in adults with non–trans­fu­sion-depen­dent or trans­fu­sion-depen­dent
alpha- or beta-thal­as­se­mia. Paper presented at the 26th Congress of the
Euro­pean Hematology Association (EHA), June 9-17, 2021. Abstract PB1805.
36. Cappellini MD, Porter J, Origa R, et  al. Sotatercept, a novel transforming
growth fac­tor β ligand trap, improves ane­mia in β-thal­as­se­mia: a phase II,
open-label, dose-find­ing study. Haematologica. 2019;104(3):477-484.
37. Humbert M, McLaughlin V, Gibbs JSR, et al; PULSAR Trial Investigators.
Sotatercept for the treat­ment of pul­mo­nary arte­rial hyper­ten­sion. N Engl J
Med. 2021;384(13):1204-1215.
38. Taher AT, Karakas Z, Cassinerio E, et  al. Efficacy and safety of ruxolitinib
in reg­u­larly trans­fused patients with thal­as­se­mia: results from a phase 2a
study. Blood. 2018;131(2):263-265.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000313
 THE CHANGING LANDSCAPE OF α - AND β -THALASSEMIA DIAGNOSIS AND TREATMENT

Optimal strategies for carrier screening

and prenatal diagnosis of α- and β-thalassemia

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/607/1851620/607mensah.pdf by guest on 13 December 2021

Cheryl Mensah1 and Sujit Sheth2
Division of Hematology and Oncology, Department of Medicine, Weill Cornell Medicine, New York, NY; and 2Division of Hematology and Oncology,
1

Department of Pediatrics, Weill Cornell Medicine, New York, NY

The thalassemias are inherited quantitative disorders of hemoglobin synthesis with a significant worldwide burden, which
result in a wide spectrum of disease from the most severe transfusion-dependent form to the mildest asymptomatic
carrier state. In this article, we discuss the importance of carrier, prenatal, and newborn screening for thalassemia. We
examine the rationale for who should be screened and when, as well as the current methodology for screening. Deficien-
cies in the newborn screening program are highlighted as well. With the advent of inexpensive and rapid genetic testing,
this may be the most practical method of screening in the future, and we review the implications of population-based
implementation of this strategy. Finally, a case-based overview of the approach for individuals with the trait as well as
prospective parents who have a potential fetal risk of the disease is outlined.

LEARNING OBJECTIVES
• Understand the current guidelines and recommendations for screening and diagnosis of thalassemia syndromes

increasing awareness of the disease, with the potential for

CLINICAL CASE
A 24-year-old woman of Thai descent is planning to start
a family with her partner, who is of Chinese descent. She
recalls being told she is anemic all her life and has been on
iron supplements off and on for many years, without correc-
tion of her anemia. She recalls that her mother has the same
medical history. Her internist refers her to hematology for
testing, and she is diagnosed with β-thalassemia trait. She
is then referred to genetics with her partner (a recent immi-
grant who has never been tested) for testing and counsel-
ing regarding the risk of thalassemia in their offspring.
reducing transmission. Current newborn screening identi-
fes affected individuals with the more common, but not
all, severe α- and β-thalassemias, but most carriers may
not be identifed in this manner. In time, with advances in
technology for genetic testing, the platform for screening
may include direct gene analysis using diagnostic chips or
whole-genome or exome sequencing, providing a more
complete genetic picture and allowing for more defnitive
carrier identifcation. A brief overview of pathophysiology
and a rational approach to testing and screening are exam-
ined based on the why, who, when, and how.

Introduction
The thalassemias are a group of genetic diseases with a
high prevalence and signifcant morbidity. The broad range
of clinical manifestations and complications, as well as
high burden of disease, from the quality-of-life as well as
fnancial standpoint, underscores the importance of min-
imizing its prevalence and optimizing outcomes in those
who are affected, through early diagnosis and compre-
hensive care. Identifcation of individuals with mild disease
or carriers, prenatal screening, and counseling are key to
Pathophysiology
Several hundred mutations have been described in the
α- and β-globin gene clusters, and the resulting imbalance
between the complementary globins results in intramed-
ullary apoptosis of the maturing erythroid precursors and
leads to the hallmark ineffective erythropoiesis (IE).1 These
genotypes are summarized in Figure 1.
Mutations or deletions of the α-globin genes result in
an excess of γ-globin in the fetus. When all 4 α-genes are
deleted, HbF cannot be produced in the second trimes-
ter and results in severe anemia and hydrops fetalis. When

Prakash Singh Shekhawat

Optimal strategies for thalassemia screening | 607

Figure 1. Thalassemia syndromes. Adapted from Sheth and Thein.1
3 genes are deleted or mutated, there is for­ma­tion of tet­ra­mers of
γ-glo­bin (hemo­glo­bin Barts [HbBarts]) ini­tially and β-glo­bin (HbH)
later. When both β-glo­bin genes are mutated, there is decreased
or absent HbA pro­duc­tion in the third tri­mes­ter and pro­gres­sive
ane­mia post­na­tally. The char­ac­ter­iza­tion of com­pound het­ero­zy­
gotes with β-thal­as­se­mia and HbE muta­tions leads to a sim­i­lar
pic­ture but with the addi­tional pres­ence of HbE.
As a result of IE and ane­mia, clin­i­cal man­i­fes­ta­tions vary, with
the most severely affected indi­vid­u­als requir­ing life­long reg­u­
lar trans­fu­sions and iron che­la­tion, with poten­tial for a vari­ety
of com­pli­ca­tions.1,2 Less severely affected indi­vid­u­als may not
608  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/607/1851620/607mensah.pdf by guest on 13 December 2021
require reg­u­lar trans­fu­sions but may still develop man­i­fes­ta­tions
sec­ond­ary to the IE such as extramedullary hema­to­poi­e­sis, bone
changes, and vas­cu­lar dis­ease, also lead­ing to sig­nif­i­cant mor­
bid­ity.1,2 Treatment and mon­i­tor­ing are cum­ber­some, are expen­
sive, and sig­nif­i­cantly affect qual­ity of life. Thus, the thal­as­se­mia
syn­dromes have a heavy bur­den of dis­ease and are good can­di­
dates for screen­ing.
Scope of the prob­lem
About 1.5% of the world’s pop­u­la­tion car­ries a thal­as­se­mia gene
muta­tion, approx­i­mat­ing 80 to 90 mil­lion peo­ple world­wide

Figure 2. Thalassemia distribution map showing prevalence of thalassemia syndromes.
(Figure 2).3,4 The car­rier state is found in Africa, around the Med­i­
ter­ra­nean, and the Middle East, but most affected indi­vid­u­als live
in the Indian sub­con­ti­nent, Southeast Asia, and China.1,3,4 How-
ever, chang­ing immi­gra­tion pat­terns have led to a world­wide
dis­tri­bu­tion of the dis­ease.5
Why we should screen
Screening for thal­as­se­mia syn­dromes is impor­tant in 3 main set­
tings: for the cor­rect diag­no­sis and man­age­ment of ane­mia, pre­
na­tal coun­sel­ing for par­ents at risk of hav­ing an affected child,
and new­born screen­ing for early diag­no­sis.6
Thalassemia car­ri­ers are com­monly misdiagnosed with iron-
defi­ciency ane­mia and often receive prolonged courses of iron,
occa­sion­ally par­en­ter­ally. Appropriate screen­ing would obvi-
ate unnec­es­sary use of iron in car­ri­ers and mon­i­tor­ing pre­vents
sequelae and chronic com­pli­ca­tions in those with thal­as­se­mia
intermedia.
Prenatal screen­ing can be used to iden­tify adult car­ri­ers inter­
ested in starting fam­i­lies. For known car­ri­ers, screen­ing of part­ners
leads to pre­na­tal risk assess­ment and fam­ily plan­ning coun­sel­ing.
Pregnant women diag­nosed with thal­as­se­mia trait should have
part­ner screen­ing to assess fetal risk, par­tic­u­larly for fetuses with
hydrops fetalis and still­birth, and receive edu­ca­tion and coun­sel­
ing.7 Earlier diag­no­sis of hydrops fetalis and still­birth can mit­i­gate
the men­tal and phys­i­cal con­se­quences for preg­nant moth­ers.
Newborn screen­ing pro­grams allow early iden­ti­fi­ca­tion of
affected infants, who can be mon­i­tored closely for the devel­op­
Prakash Singh Shekhawat
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/607/1851620/607mensah.pdf by guest on 13 December 2021
ment of ane­mia, and start trans­fu­sions at the appro­pri­ate time,
min­i­miz­ing the risk for com­pli­ca­tions.
Who should be screened
Individuals—based on demo­graph­ics and his­tory
Adults and chil­dren of Asian, Afri­can, or Med­i­ter­ra­nean ances­try,
with a fam­ily his­tory of thal­as­se­mia, fam­ily his­tory of life­long ane­
mia, and those with a his­tory of micro­cytic ane­mia unre­spon­sive
to iron sup­ple­men­ta­tion or micro­cytic ane­mia with­out iron defi­
ciency, should be screened for thal­as­se­mia.6
Prospective par­ents—pre­na­tal screen­ing
The Amer­i­can College of Obstetrics and Gynecology (ACOG)
rec­og­nizes that patients with Afri­can, Med­i­ter­ra­nean, and South-
east Asian ances­try are at increased risk for hemo­glo­bin­op­a­
thies, includ­ing sickle cell and thal­as­se­mia.7 As a result of more
wide­spread mixing of pop­u­la­tions, lim­it­ing screen­ing to these
ancestries may not be suf­fi­cient, and broader guide­lines are
nec­es­sary.8
Prospective par­ents who are iden­ti­fied as thal­as­se­mia car­ri­
ers should receive pre­na­tal genetic test­ing with coun­sel­ing. If
one part­ner is con­firmed to be a car­rier, the other should be
exten­sively tested. Couples who are both thal­as­se­mia car­ri­ers
can be offered in vitro fer­til­iza­tion and pre­im­plan­ta­tion genetic
diag­no­sis.7 For expec­tant par­ents, antepartum genetic test­ing
can be offered prior to deliv­ery along with coun­sel­ing.
Optimal strat­e­gies for thal­as­se­mia screen­ing  |  609
Newborns
At-risk new­borns should be screened for thal­as­se­mia syn­dromes.6
However, this poses tech­ni­cal chal­lenges because the meth­od­ol­
ogy cur­rently used for new­born hemo­glo­bin­op­a­thy screen­ing (in
all­50 states in the United States) is spe­cif­i­cally designed to detect
sickle hemo­glo­bin. The detec­tion of HbBarts (tet­ra­mers of γ-glo­bin)
on the new­ born screen would facil­ i­
tate the early diag­ no­sis of
α-thal­as­se­mia trait or HbH dis­ease, but this requires hemo­glo­bin
quan­ti­ta­tion, not performed uni­ver­sally. The most severe form of
β-thal­as­se­mia, β0/β0-thal­as­
se­mia, could be diag­ nosed by the
absence of HbA on the new­born screen. Less severe forms of β-thal­
as­se­mia could also be detected but not usu­ally the trait. In regions
where thal­as­se­mia is more com­mon, par­tic­u­larly Asia, implement-
ing new­born screen­ing meth­od­ol­ogy to be ­able to detect thal­as­se­
mia would be an impor­tant strat­egy for early diag­no­sis, coun­sel­ing,
and reg­u­lar com­pre­hen­sive care where appro­pri­ate.
Infants
The Amer­i­can Academy of Pediatrics had a longstanding rec­om­
men­da­tion for a screen­ing hemo­glo­bin/hemat­o­crit for all­ infants
at 9 months of age.9 To pre­vent iron defi­ciency, iron-for­ti­fied for­
mula is now recommended, and screen­ing is done at 12 months of
age for for­mula-fed infants. Severe thal­as­se­mia, β0/β0-thal­as­se­mia,
usu­ally pres­ents well before this age with pro­gres­sive ane­mia and
other clin­i­cal man­i­fes­ta­tions. However, more inter­me­di­ate forms,
includ­ing β+/β+-thal­as­se­mia, HbE/β0-thal­as­se­mia, or HbH dis­ease,
may have few clin­i­cal man­i­fes­ta­tions besides mod­er­ate ane­mia and
may be missed until the 9- to 12-month hemo­glo­bin/hemat­o­crit
is done. Screening these infants would be impor­tant to estab­
lish a diag­no­sis, pre­clude unnec­es­sary iron sup­ple­men­ta­tion, and
begin mon­i­tor­ing and com­pre­hen­sive care to opti­mize out­comes
and min­i­mize mor­bid­ity.
How we screen
Newborn thal­as­se­mia screen­ing
With the high prev­a­lence and wide dis­tri­bu­tion of hemo­glo­bin­
op­a­thies, test­ing for these was incor­po­rated into the new­born
screen.6 The nor­mal new­born at term has approx­i­ma­tely 80%
to 90% HbF and 10% to 20% HbA, with trace amounts of other
hemo­glo­bins such as HbA2, which are typ­i­cally not reported.
α-Thalassemia new­born screen­ing.  α-Thalassemia syn­dromes
com­pat­i­ble with life can be detected on new­born screen­ing.
Fetuses with 3 α-gene dele­ tions or 2 α-gene dele­ tions and
1 α-gene with a Constant Spring muta­tion (or other nondeletion-
al α muta­tion) will pro­duce excess unbound γ-glo­bin, which will
result in the for­ma­tion of γ-tet­ra­mers, des­ig­nated HbBarts, iden­
ti­fi­able on new­born screen­ing. The amount of HbBarts cor­re­lates
with the sever­ity of α-thal­as­se­mia, although not in all­instances,
and can be detected antepartum and at birth.9,10 Newborns with
2 α-gene dele­tions may have 3% to 6% HbBarts at birth, and a
silent car­rier miss­ing only 1 α-gene may have 1% to 4% at birth.11
β-Thalassemia new­born screen­ing. Fetuses with β0/β0 muta­
tions will pro­duce no HbA and will have only HbF on the new­
born screen, only oth­er­wise seen in extreme pre­ma­tu­rity, where
β-glo­bin expres­sion has not begun yet. Thus, ges­ta­tional age is
an impor­tant con­sid­er­ation, and nor­mal ranges should be estab-
lished at var­i­ous ges­ta­tional ages for eas­ier inter­pre­ta­tion. Fe-
tuses with 1 β0-muta­tion and 1 β+-muta­tion or 2 β+-muta­tions will
have some HbA on the new­born screen. If this is quan­ti­fied, it is
610  |  Hematology 2021  |  ASH Education Program
less than the nor­mal range—approx­i­ma­tely 3% to 5%—and may
be flagged as pos­si­ble β-thal­as­se­mia.6 Individuals with β0/HbE
may have no HbA and only have HbF and HbE, whereas those
with β+/HbE will have some HbA. HbA2, which is ele­vated in car­
ri­ers, is pro­duced in very small quan­ti­ties in the fetus (0.5%) and
is not typ­i­cally reported on the new­born screen.1,2
Newborn screen­ing in the United States. There is cur­rently no
stan­dard for new­born screen­ing across the United States, and
rec­om­men­da­tions for test­ing vary from state to state.12 In 2017,
the Amer­ic ­ an Academy of Pediatrics reviewed new­born hemo­
glo­bin­op­a­thy screen­ing across the 50 states.13 Thirty-one states
recommended ane­mia screen­ing beyond rou­tine care or sickle
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/607/1851620/607mensah.pdf by guest on 13 December 2021
cell screen­ing. Five of the 31 states recommended hemo­glo­bin
elec­tro­pho­re­sis and 4 recommended hema­tol­ogy refer­ral.13 Only
1 rec­om­mend ini­tial genetic test­ing. The other 19 states do not
have thal­as­se­mia screen­ing rec­om­men­da­tions beyond rou­tine
care. Twelve of these 19 states reported α-thal­as­se­mia trait with
no addi­tional fol­low-up, and 7 did not report α-thal­as­se­mia or
car­rier sta­tus at all­.13
Initial new­born screen­ing in most states is by high-per­for­mance
liq­uid chro­ma­tog­ra­phy (HPLC) or iso­elec­tric focus­ing (IEF) on a
spot of blood. These tests detect hemo­glo­bin var­i­ants and may
pro­vide lim­ited data to sup­port diag­no­sis of a thal­as­se­mia syn­
drome. HPLC, either by cat­ion exchange or reverse phase, quan-
tifies hemo­glo­bin var­i­ants such as hemo­glo­bin A, A2, F, S, H, and
Barts.13,14 However, dif­fer­ent hemo­glo­bin var­i­ants may coelute,
mak­ing it dif­fi­cult to dis­tin­guish one var­i­ant from another. Thus,
HPLC is used for ini­tial screen­ing and is followed by a con­fir­ma­tory
test, either IEF or molec­u­lar test­ing.10 IEF is more labor inten­sive but
identifies and quantifies hemo­glo­bin var­i­ants more pre­cisely than
stan­dard hemo­glo­bin elec­tro­pho­re­sis.13 Molecular test­ing includes
poly­mer­ase chain reac­tion (PCR)–based test­ing or DNA sequenc­
ing.1 Allele-spe­cific PCR-based meth­ods are used to detect point
muta­tions in α- and β-glo­bin genes. Gap PCR is used to detect
dele­tions, α dupli­ca­tions, and deletional hemo­glo­bin var­i­ants.1
DNA sequenc­ing is more expen­sive and more com­pre­hen­sive and
can be used to diag­nose pre­vi­ously uniden­ti­fied muta­tions.1
There is no nation­wide stan­dard of reporting HbBarts per­cent­
age to pro­vid­ers, patients, or their fam­i­lies.10 A Centers for Dis-
ease Control and Prevention sur­vey found that HbBarts reporting
varies among new­born screen­ing pro­grams, and some pro­grams
do not report quan­ti­ta­tive HbBarts or rec­om­mend fol­low-up.13
Thalassemia screen­ing in chil­dren and adults
Prospectively screen­ing all­patients for thal­as­se­mia is nei­ther cost-
effec­tive nor indi­cated because iron defi­ciency is rel­a­tively much
more com­mon and explains a micro­cytic ane­mia. Establishing the
diag­no­sis of thal­as­se­mia trait is rel­e­vant, and avoid­ance of unnec­
es­sary iron sup­ple­men­ta­tion is impor­tant. Individuals with an inter­
me­di­ate-sever­ity thal­as­se­mia syn­drome with IE have com­pli­ca­tions
that could be prevented if com­pre­hen­sive care is pro­vided early.
In the fol­low­ing sit­u­a­tions, screen­ing for thal­as­se­mia is
recommended:
a. Family his­tory of thal­as­se­mia—trait or dis­ease
b. Microcytic ane­mia with a neg­a­tive his­tory for iron defi­ciency—
ade­quate die­tary iron intake, absence of blood loss
c. Persistent micro­ cytic ane­ mia despite an ade­ quate trial of
sup­ple­men­tal iron
Testing of such indi­vid­u­als can be approached in a step­
wise man­ner (Figure 3).
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/607/1851620/607mensah.pdf by guest on 13 December 2021

Figure 3. Ideal algorithm for thalassemia screening, developed to screen for thalassemia syndromes in adults based on history,
physical examination, and laboratory parameters.

Prenatal screen­ing deter­min­ing fetal risk and is based on hemo­glo­bin elec­tro­pho­

ACOG rec­om­mends that women with a low mean corpuscular re­sis or genetic test­ing as appro­pri­ate for β- and α-thal­as­se­mia,
volume (MCV) should have serum fer­ri­tin assess­ment. Those with respec­tively.8 Couples who are both iden­ti­fied as car­ri­ers should
nor­mal lev­els and micro­cytic ane­mia should have hemo­glo­bin elec­ receive genetic coun­ sel­
ing to deter­ mine risk of an affected
tro­pho­re­sis test­ing.7 For patients with a nor­mal hemo­glo­bin elec­ fetus, and fam­ily deci­sion sup­port should be pro­vided. The most
tro­pho­re­sis and Asian ances­try, genetic test­ing for α-thal­as­se­mia severe form of α-thal­as­se­mia—dele­tion of all 4 α-genes—results
is recommended. In each sit­ u­
a­tion,Prakash
part­
ner test­
ing is key
Singh Shekhawat to in hydrops fetalis, with severe ane­mia (hemo­glo­bin rang­ing from

Optimal strat­e­gies for thal­as­se­mia screen­ing  |  611

3-8 g/dL), organomegaly, and edema, usu­ally lead­ing to intra­uter­
ine fetal demise from heart fail­ure, with pre­term labor, still­birth,
and a det­ri­men­tal effect on mater­nal health.1,11 Prenatal screen­ing
would iden­tify the at-risk fetus, and amnio­cen­te­sis or chorionic vil­
lus sam­pling would con­firm the diag­no­sis.11 In a suspected preg­
nancy, Dopp­ler assess­ments of cere­bral ves­sel flow veloc­i­ties or
direct fetal blood sam­pling to quan­tify the ane­mia could be used.
If intra­uter­ine trans­fu­sions are insti­tuted in time, such fetuses
may sur­vive but would be trans­fu­sion depen­dent post­na­tally—
des­ig­nated α-thal­as­se­mia major.11
It is impor­tant to keep in mind that 2 α-gene dele­tions are not
uncom­mon in Afri­can ances­try indi­vid­u­als, but these muta­tions
tend to occur on oppo­site chro­mo­somes (trans con­fig­u­ra­tion),
and hence the risk of hav­ing a fetus with all 4 genes deleted is
lower than in indi­vid­u­als with East Asian ances­try, in whom the
2 gene dele­tions are more com­monly on the same chro­mo­some
(cis con­fig­u­ra­tion). For women who are iden­ti­fied as hav­ing
β-thal­as­se­mia trait or HbS trait, part­ner test­ing of hemo­glo­bin
frac­tion­ation is imper­a­tive to assess fetal risk for β-thal­as­se­mia
or sickle cell dis­ease (HbS-β thal).1
Preimplantation genetic test­ing.  Preconception, ACOG rec­om­
mends genetic test­ing and coun­sel­ing for cou­ples at high risk for
thal­as­se­mia as noted.7 For cou­ples inter­ested in avoiding elec­
tive ter­mi­na­tion, in vitro fer­til­iza­tion and pre­im­plan­ta­tion genetic
diag­no­sis can be offered.7 For cou­ples who were not diagnosed
as high risk for thalassemia prior to preg­nancy, DNA test­ing via
chronic vil­lus sam­pling or amnio­cen­te­sis should be offered. Coun-
seling should be offered in all­such sit­u­a­tions.7 This approach has
resulted in a major reduc­tion in the birth rate of infants with thal­as­
se­mia through­out the Med­i­ter­ra­nean region and the Middle East,
as well as in parts of the Indian sub­con­ti­nent and Southeast Asia.
Novel approaches con­tinue to be explored in an attempt to avoid
the use of inva­sive pro­ce­dures such as chronic vil­lus sam­pling and
amnio­cen­te­sis, includ­ing harvesting fetal DNA from fetal cells in
mater­nal blood or mater­nal plasma.15,16
Suggested approach for screen­ ing of indi­
vid­
u­als at risk, all­
ages.  As the demo­graph­ics con­tinue to change glob­ally and in
the United States, a com­pre­hen­sive approach for early diag­no­sis
will improve the health of patients and min­i­mize risk. We sug­gest
the fol­low­ing, starting pre­con­cep­tion and through adult­hood:
1. Screening and iden­ti­fi­ca­tion of thal­as­se­mia car­rier sta­tus and
coun­sel­ing early for all­pro­spec­tive moth­ers at risk based
on eth­nic­ity, fam­ily his­tory, med­i­cal his­tory, and lab­o­ra­tory
param­e­ters. Confirmatory genetic test­ing of both part­ners
and genetic coun­sel­ing must be offered and, if avail­­able, pre­
im­plan­ta­tion genetic diag­no­sis when appro­pri­ate.
2. If a preg­nancy has already occurred in a mother at risk, the
same screen­ing should be done, and if thal­as­se­mia trait is
con­firmed as described pre­vi­ously, part­ner test­ing and ge-
netic coun­sel­ing are indi­cated. Testing of the fetus should be
offered if both par­ents are car­ri­ers or if the father’s sta­tus can­
not be ascertained. More advanced tech­niques such as fetal
DNA in mater­nal blood need to be tested more exten­sively to
avoid more inva­sive test­ing pro­ce­dures. If the fetus is found
to be affected, appro­pri­ate coun­sel­ing should be pro­vided.
3. Newborn screen­ing should be stan­dard­ized to include hemo­
glo­bin char­ac­ter­iza­tion and frac­tion­ation for all­. Ideally, as
612  |  Hematology 2021  |  ASH Education Program
genetic test­ing becomes more widely avail­­able, tech­ni­cally
eas­ier to per­form, and cheaper, mov­ing test­ing to a genome/
exome or chip tech­nol­ogy would be most effec­tive in defin­
i­tively iden­ti­fy­ing indi­vid­u­als with thal­as­se­mia. We rec­og­
nize this may be a future-ori­ ented goal and a not widely
fea­si­ble approach. Until uni­ver­sal genetic test­ing can be per-
formed, we rec­om­mend the opti­mal strat­egy of performing
a hemo­ glo­
bin elec­ tro­
pho­ re­sis, HPLC, or IEF on all­infants;
reporting the hemo­glo­bin var­i­ants, espe­cially HbBarts, and
unsta­ble hemo­glo­bin var­i­ants; and quan­ti­fy­ing HbA. Additions
to the new­born screen are jus­ti­fied if a treat­ment intro­duced
early would reduce mor­bid­ity and mor­tal­ity. For thal­as­se­mia,
this remains con­tro­ver­sial, and pro­spec­tive data dem­on­strat­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/607/1851620/607mensah.pdf by guest on 13 December 2021
ing the ben­e­fits of this approach could pro­vide a ratio­nale.
. We rec­om­mend screen­ing all­ indi­vid­u­als (adults and chil­dren)
4
with appro­pri­ate his­to­ries, cer­tain phys­i­cal exam­i­na­tion find­
ings, and micro­cytic ane­mia for thal­as­se­mia using the algo­
rithm in Figure 3.
CLINICAL CASE (Con­tin­ued)
The cou­ple is at risk for an affected child given that the pro­
spec­tive mother has β-thal­as­se­mia trait, and the pro­spec­tive
father is of Chi­nese descent. He should have car­rier test­ing for
β-thal­as­se­mia trait. If he is a car­rier, then the poten­tial fetus is
at risk of hav­ing the dis­ease. Preimplantation diag­no­sis should
be offered before con­cep­tion, or test­ing of the fetus should be
offered early in the preg­nancy. Some fam­i­lies may choose nei­
ther option.
Conclusion
Due to increas­ingly chang­ing demo­graph­ics and migra­tion from
areas where thal­as­se­mia is more prev­a­lent to those where it is less
so, the prev­a­lence of thal­as­se­mia is chang­ing. It has become an
impor­tant health issue to iden­tify patients with thal­as­se­mia syn­
dromes and thal­as­se­mia trait to be ­able to not only pro­vide early
com­pre­hen­sive care but also avoid unnec­es­sary inter­ven­tions.
Early diag­no­sis can pre­pare at-risk cou­ples, as well as iden­tify
at-risk fetuses and new­borns. Newborn screen­ing for hemo­glo­bin
dis­or­ders, pioneered with a sickle cell focus, has come a long way,
but there is still room for advances to achieve the goal of opti­mal
screen­ing for thal­as­se­mia and estab­lish­ment of an early diag­no­sis
to opti­mize man­age­ment.
Acknowledgments
C.M and S.S. have received funding from the Centers for Disease
Control and Prevention and the Health Resources and Services
Administration.
Conflict-of-inter­est dis­clo­sure
Cheryl Mensah has served as a consultant to Bluebird bio and
Chiesi. She currently serves on adjudication committee for CRISPR/
Vertex for thalassemia.
Sujit Sheth is a con­sul­tant to Agios, Bluebird bio, Bristol Myers
Squibb, and Chiesi and serves on a clin­i­cal trial steering com­mit­
tee for CRISPR/Vertex CTX001 for thal­as­se­mia.
Off-label drug use
Cheryl Mensah: nothing to disclose.
Sujit Sheth: nothing to disclose.
Correspondence
Sujit Sheth, Division of Pediatric Hematology Oncology, Weill
Cornell Medical College, 525 E 68th Street, P695, New York, NY
10065; e-mail: shethsu@med​­.cornell​­.edu.
References
1. Sheth S, Thein S. Thalassemia: a dis­or­der of glo­bin syn­the­sis. In: Kaushan-
sky K, Prchal JT, Burns LJ, Lichtman MA, Levi M, Linch DC, eds. Williams
Hematology. 10th ed. McGraw-Hill; 2021.
2. Taher AT, Weatherall DJ, Cappellini MD. Thalassaemia. Lancet. 2018;391(10116):
155-167.
3. Modell B, Darlison M. Global epi­de­mi­­ol­ogy of haemoglobin dis­or­ders and
derived ser­vice indi­ca­tors. Bull World Health Organ. 2008;86(6):480-487.
4. Williams TN, Weatherall DJ. World dis­tri­bu­tion, pop­u­la­tion genet­ics, and
health bur­den of the hemo­glo­bin­op­a­thies. Cold Spring Harb Perspect Med.
2012;2(9):a011692.
5. Sayani FA, Kwiatkowski JL. Increasing prev­a­lence of thal­as­se­mia in America:
impli­ca­tions for pri­mary care. Ann Med. 2015;47(7):592-604.
6. Centers for Disease Control and Prevention. Hemoglobinopathies: Current
Practices for Screening, Confirmation and Follow-up. Association of Public
Health Laboratories; 2015.
7. ACOG Committee on Obstetrics. ACOG Practice Bulletin No. 78: hemo­glo­
bin­op­a­thies in preg­nancy. Obstet Gynecol. 2007;109(1):229-237.
Prakash Singh Shekhawat
8. Shang X, Xu X. Update in the genet­ics of thal­as­se­mia: what cli­ni­cians need
to know. Best Pract Res Clin Obstet Gynaecol. 2017;39:3-15.
9. Kohli-Kumar M. Screening for ane­mia in chil­dren: AAP rec­om­men­da­tions—a
cri­tique. Pediatrics. 2001;108(3):e56.
10. Bender M, Yusuf C, Davis T, et al. Newborn screen­ing prac­tices and alpha-
thal­as­se­mia detec­tion—United States, 2016. MMWR Morb Mortal Wkly
Rep. 2020;69:1269–1272.
11. Vichinsky EP. Alpha thal­as­se­mia major—new muta­tions, intra­uter­ine man­
age­ment, and out­comes. Hematology Am Soc Hematol Educ Program.
2009;2009(1):35–41.
12. Benson JM, Therrell BL. History and cur­rent sta­tus of new­born screen­ing for
hemo­glo­bin­op­a­thies. Semin Perinatol. 2010;34(2):134-144.
13. Fogel BN, Nguyen HLT, Smink G, Sekhar DL. Variability in state-based rec­
om­men­da­tions for man­age­ment of alpha thal­as­se­mia trait and silent car­rier
detected on the new­born screen. J Pediatr. 2018;195(2, pt 2):283-287.
14. Hoppe CC. Newborn screen­ing for non-sick­ling hemo­glo­bin­op­a­thies.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/607/1851620/607mensah.pdf by guest on 13 December 2021
Hematology Am Soc Hematol Educ Program. 2009;2009(1):19-25.
15. Cheung M-C, Goldberg JD, Kan YW. Prenatal diag­no­sis of sickle cell ane­
mia and thal­as­se­mia by anal­y­sis of fetal cells in mater­nal blood. Nat Genet.
1996;14:264-268.
16. Hung ECW, Chiu RWK, Lo YMD. Detection of cir­cu­lat­ing fetal nucleic acids:
a review of meth­ods and appli­ca­tions. J Clin Pathol. 2009;62(4):308-313.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000296
Optimal strat­e­gies for thal­as­se­mia screen­ing  |  613
 THE COVID CRASH: LESSONS LEARNED FROM A WORLD ON PAUSE

How to recognize and manage

COVID-19-associated coagulopathy

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/614/1851465/614gerber.pdf by guest on 13 December 2021

Gloria F. Gerber and Shruti Chaturvedi
Division of Hematology, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD

COVID-19 is frequently associated with abnormalities on coagulation testing and a coagulopathy driven by inflamma-
tion, intravascular coagulation activation, and microvascular thrombosis. Elevated D-dimer is the most common finding
and is a predictor of adverse outcomes including thrombosis, critical illness, and death. Although COVID-19-associated
coagulopathy has some similarities to disseminated intravascular coagulation, the platelet count is usually preserved,
coagulation times are usually normal or minimally prolonged, and thrombosis is more common than bleeding, at least
in noncritically ill patients. Bleeding is uncommon but may be a significant problem in critically ill patients, includ-
ing those who may develop a consumptive coagulopathy with frank disseminated intravascular coagulation and those
on extracorporeal membrane oxygenation. Blood product support to correct coagulopathy is reserved for bleeding
patients or those requiring invasive procedures. Current recommendations suggest that all hospitalized patients should
receive at least a prophylactic dose of anticoagulation. Results from a multiplatform randomized clinical trial suggest
that therapeutically dosed anticoagulation may improve outcomes, including the need for organ support and mortality
in moderately ill patients but not in those requiring critical care. The results of ongoing trials evaluating the impact of
different antithrombotic strategies (therapeutic agents and intensity) on COVID-19 outcomes are eagerly awaited and
are expected to have important implications for patient management. We also discuss COVID-19 vaccine-associated
cytopenias and bleeding as well as vaccine-induced thrombotic thrombocytopenia, in which thrombosis is associated
with thrombocytopenia, elevated D-dimer, and, frequently, hypofibrinogenemia.

LEARNING OBJECTIVES
• Recognize the coagulation derangements commonly encountered in COVID-19 associated with outcomes includ-
ing thrombosis, need for critical care support, and mortality
• Appropriately monitor and manage coagulation abnormalities in hospitalized patients with COVID-19
• Recognize and manage the rare complications of COVID-19 vaccination-associated thrombocytopenia with bleed-
ing and vaccine-induced thrombotic thrombocytopenia (VITT)

Introduction cause mortality.5,6 In critically ill patients, this coagulopathy

The COVID-19 pandemic has contributed to significant
mortality and morbidity across the world. Despite unprec-
edented public health efforts including the development of
multiple effective vaccines, the pandemic continues to rage
in many parts of the world, with novel viral variants leading
to recurrent outbreaks. While COVID-19 is a predominantly
respiratory illness, hemostatic and thrombotic complica-
tions are common.1,2 Early in the pandemic, a high preva-
lence of thromboembolic complications was observed in
patients with COVID-19.3,4 Many patients exhibit laboratory
evidence of a coagulopathy reminiscent of, but not identi-
cal to, disseminated intravascular coagulation (DIC), which
is associated with thrombosis risk, critical illness, and all-
may evolve into frank DIC. While less common than labora-
tory coagulopathy or thrombosis, bleeding occurs in up to
5% of hospitalized patients and has important implications
for management.3 In this review we discuss our approach
to COVID-19-associated coagulopathy (CAC) and bleeding.
CLINICAL CASE
A 51-year-old woman with hypertension and asthma pre-
sented with fever, cough, and dyspnea; polymerase chain
reaction for SARS-CoV-2 was positive, and she was hospital-
ized due to the need for supplemental oxygen. Laboratory

614 | Hematology 2021 | ASH Education Program

Table 1.  Similarities and dif­fer­ences between CAC the fact that intra­vas­cu­lar coag­u­la­tion acti­va­tion and micro­vas­cu­
and DIC/SIC lar throm­bo­sis are rel­a­tively local­ized to the pul­mo­nary cir­cu­la­tion
in CAC versus the more sys­temic acti­va­tion of coag­u­la­tion in DIC.
CAC DIC SIC Additionally, fibri­no­ly­sis is suppressed in CAC, which may con­trib­
Platelet count ↑↓ ↓↓ ↓ ute to the prothrombotic phe­no­type.11,12 Iba et al have pro­posed
diag­nos­tic cri­te­ria for CAC (Table 2).13
D-dimer ↑ ↑↑ ↑
Thrombocytopenia and PT pro­lon­ga­tion are less com­mon in
PT/aPTT ↔↑ ↑↑ ↑ COVID-19 but pre­dict poor clin­i­cal out­comes. In an anal­y­sis of 201
Fibrinogen ↑ ↓↓ ↓ hos­pi­tal­ized patients from China, prolonged PT on admis­sion was
asso­ci­ated with the devel­op­ment of acute respi­ra­tory dis­tress
Antithrombin ↔ ↓ ↓
syn­drome (ARDS; haz­ard ratio, 1.56 [1.32-1.83]; P < .001).14 Inter-
VWF ↑ ↑ ↑ estingly, throm­bo­cy­to­pe­nia was pres­ent in only 18.8% of these
FVIII ↑ ↑↓ ↑ patients com­pared to >50% in patients with ARDS from other

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/614/1851465/614gerber.pdf by guest on 13 December 2021

LA/aPL + - - infec­tious causes. Compared to non-COVID ARDS, COVID-19 is
Thrombosis ↑↑ ↑ ↑
Bleeding ↔↑ ↑↑ ↑ stud­ies includ­ing 1779 COVID-19 patients, the plate­let count was
test­ing on admis­sion revealed ele­vated dimerized plas­min frag­
ment D (D-dimer; 1.4 mg/L [0.00-0.49]) and fibrin­o­gen (345 mg/dL).
Platelet count (238  ×  109/L) and pro­throm­bin time (PT; 10.8 sec­
onds) were nor­mal. Activated par­tial throm­bo­plas­tin time (aPTT)
was prolonged (43 sec­onds [23.1-30.9]).
CAC
Laboratory find­ings in CAC
Coagulation derange­ments suggesting hyper­co­ag­u­la­bil­ity are
com­mon in COVID-19. Markedly ele­vated D-dimer is the most com­
mon find­ing and is asso­ci­ated with adverse out­comes, includ­ing
throm­bo­sis, the need for crit­i­cal care sup­port, and over­all mor­tal­
ity.3,5,7 Elevated D-dimer lev­els cor­re­late with lev­els of inflam­ma­tory
mark­ers, includ­ing C-reac­tive pro­tein, sed­i­men­ta­tion rate, procal-
citonin, and fer­ri­tin, supporting an asso­ci­a­tion with the inflam­
ma­tory state of COVID-19.3 However, fibrin­o­gen lev­els are usu­ally
ele­vated/nor­mal, a plate­let count <100  ×  109/L is uncom­mon,3,8 and
the PT is nor­mal at pre­sen­ta­tion in the major­ity of patients. Thus,
in most cases CAC is dis­tinct from the pro­to­typ­i­cal con­sump­tive
coagulopathies, such as sep­sis-induced coagulopathy (SIC) and
DIC (Table 1).9,10 Moreover, CAC is less likely to be asso­ci­ated with
bleed­ing than throm­bo­sis. These dif­fer­ences are prob­a­bly due to
asso­ci­ated with higher PT, anti­throm­bin, fibrin­o­gen, and plate­let
count and lower aPTT and D-dimer lev­els.15 In a meta-anal­y­sis of 9
sig­nif­i­cantly lower in patients with severe dis­ease and patients
who died.16 Similarly, a marked decrease in plasma fibrin­og ­ en has
been reported shortly before death in a num­ber of patients.5 An
early study by Tang et al noted a higher prev­a­lence of DIC by
International Society on Thrombosis and Haemostasis (ISTH) cri­
te­ria on admis­sion in nonsurvivors vs sur­vi­vors (71.4% vs 0.6%).
However, sub­se­quent stud­ies report a much lower inci­dence
of DIC at ini­tial pre­sen­ta­tion. For exam­ple, none of the 150 con­
sec­u­tive patients with COVID-19 and ARDS and only 2.2% of 388
con­sec­u­tive patients hos­pi­tal­ized with COVID-19 in Milan met the
cri­te­ria for DIC at pre­sen­ta­tion.15,17 Thus, overt DIC is uncom­mon
in COVID-19 but may develop in crit­i­cally ill patients.3
Based on the spec­trum of coag­u­la­tion abnor­mal­i­ties in COVID-
19 that par­al­lels dis­ease sever­ity, three stages of COVID-19 coagu-
lopathy have been pro­posed (Figure 1).18 In stage 1, patients exhibit
mild symp­toms and mild local­ized inflam­ma­tion and ele­vated D-
dimer. Stage 2 is char­ac­ter­ized by more severe symp­toms requir­
ing sup­ple­men­tal oxy­gen, with pul­mo­nary inflam­ma­tion and intra­
vas­cu­lar coag­u­la­tion acti­va­tion (ele­vated D-dimer, fibrin­o­gen).
Patients who prog­ress to stage 3 require crit­i­cal care sup­port and
dem­on­strate severe sys­temic inflam­ma­tion and coagulopathy
with mark­edly ele­vated D-dimer and fibrin­o­gen, prolonged PT,
and throm­bo­cy­to­pe­nia. A sub­set of these patients devel­ops overt
DIC that may be due to COVID-19 or bac­te­rial super­in­fec­tion and
is asso­ci­ated with a high risk of bleed­ing and death.

Table 2.  Proposed cri­te­ria for CAC com­pared with ISTH cri­te­ria for SIC and DIC

Proposed CAC cri­te­ria13 ISTH SIC cri­te­ria10 ISTH DIC cri­te­ria9

≥2 of the fol­low­ing: ≥4 points: ≥5 points:
1) Platelet count <150 × 109/L 1) Platelet count, × 109/L 1) Platelet count, × 109/L
2) D-dimer ele­va­tion >2 × ULN   <150 but ≥100 = 1  <100 = 1
3) PT prolonged >1s or INR >1.2   <100  =  2  <50 = 2
4) Presence of micro and/or macrovascular thrombosis 2) INR 2) D-dimer
**Risk fac­tors for the devel­op­ment of CAC include ele­vated fibrin­o­gen and    >1.2 but ≤1.4 = 1     increased <5 × ULN = 2
VWF and pres­ence of lupus anti­co­ag­u­lant or antiphospholipid antibodies   >1.4 = 2    ≥5 × ULN = 3
3) SOFA score 3) PT
  1 = 1    prolonged ≥3 s but
  ≥2 = 2 <6 s = 1
   prolonged ≥6 s = 2
4) Fibrinogen
   ≤1.0 g/L = 1
INR, international normalized ratio; SOFA, Sequential Organ Failure Assessment; ULN, upper limit of nor­mal.
Prakash Singh Shekhawat
COVID-19 coagulopathy | 615
 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/614/1851465/614gerber.pdf by guest on 13 December 2021
Figure 1.  Stages of CAC. The spectrum of coagulation abnormalities in COVID-19 parallels disease severity. In stage 1, patients exhibit
mild symptoms and evidence of mild localized inflammation and coagulopathy with elevations in D-dimer approximately 2 to 3 times
the upper limit of normal. Stage 2 is characterized by more severe symptoms requiring supplemental oxygen with pulmonary inflam-
mation and intravascular coagulation activation in the lungs as well as some systemic involvement. This is reflected in further D-dimer
elevation to 3 to 6 times the upper limit of normal. Patients progressing to stage 3 require critical care with respiratory and other
organ support and demonstrate severe systemic inflammation and coagulopathy with markedly elevated D-dimer and fibrinogen,
prolonged PT, thrombocytopenia, and high incidence of thromboembolism. A proportion of these patients will develop overt DIC as
a result of their COVID-19 illness or other superimposed infections.18

Antiphospholipid antibodies in COVID-19 Other coag­u­la­tion param­e­ters

The aPTT is prolonged in some patients with COVID-19. A like-
ly expla­na­tion is the pres­ence of a lupus anti­co­ag­u­lant (LA) or
antiphospholipid antibodies (aPL). In a series of 56 hos­pi­tal­ized
patients with COVID-19, LA was detected in 44.6%; 8.9% also
had either anticardiolipin (aCL) or anti-beta-2-gly­co­pro­tein anti-
bodies.19 Both tra­di­tional aPL (such as immu­no­glob­u­lin [Ig] G and
IgM aCL) and var­io ­ us “noncriteria” aPL (includ­ing anti-PS/PT as
well as IgA isotypes of aCL and anti-beta-2-gly­co­pro­tein) have
been detected in COVID-19. Whether these are tran­sient aPL, as
reported in other viral infec­tions,20 or per­sis­tent aPL, more likely
to be asso­ci­ated with long-term throm­botic risk, is unknown.
Moreover, LA assays must be interpreted with cau­tion in crit­i­
cally ill patients due to poten­tial confounding by anti­co­ag­u­lant
ther­apy. High lev­els of acute phase reac­tants such as C-reac­tive
pro­tein can also cause false pos­i­tives by prolonging in vitro clot­
ting times. Prolonged aPTT due to LA may neces­si­tate the use of
the anti-factor Xa assay for mon­i­tor­ing hep­a­rin ther­apy.
In addi­tion to the com­monly eval­u­ated lab­o­ra­tory param­e­ters
discussed above, ele­vated coag­u­la­tion fac­tor VIII (FVIII) and von
Willebrand fac­ tor (VWF) are fre­ quent find­ ings in hos­pi­tal­
ized
COVID-19 patients and cor­re­late with ele­va­tions in inflam­ma­tory
mark­ers.15,21-23 Nonsurvivors have lower anti­throm­bin lev­els than
sur­vi­vors (84% vs 91%), but lev­els gen­er­ally do not drop below
80%.5,15,22 In a study com­par­ing 48 inten­sive care unit (ICU) to
30 gen­eral ward patients, fibrinolytic pathway parameters, includ­
ing plas­min­o­gen acti­va­tor inhib­i­tor, tis­sue plas­min­o­gen acti­va­tor,
and throm­bin activatable fibri­no­ly­sis inhib­i­tor, were sig­nif­i­cantly
higher in the ICU patients.22 Viscoelastic test­ing (thromboelas-
tography, rotational thromboelastometry) reveals increased clot
strength with con­tri­bu­tions of both ele­vated fibrin­o­gen and plate­
lets.21,24 Fibrinolysis is also suppressed.11,12 Thrombin gen­er­a­tion is
nor­mal to increased even in patients receiv­ing pro­phy­lac­tic anti-
coagulation.25 However, the clin­ic ­ al util­ity of these assays is uncer­
tain, and they are not recommended out­side of a research set­ting.

616  |  Hematology 2021  |  ASH Education Program

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/614/1851465/614gerber.pdf by guest on 13 December 2021
Figure 2.  Pathogenesis of CAC. SARS-CoV-2 binds the angiotensin-converting enzyme 2 receptor on respiratory epithelium and endo-
thelial cells. Viral replication and shedding lead to pneumocyte and endothelial cell apoptosis, triggering an inflammatory response,
which in a portion of patients leads to pathogenic cytokine storm. These cytokines and endothelial injury result in procoagulant
changes, including the release of VWF and FVIII and the upregulation of TF, P-selectin, and fibrinogen. Anticoagulant proteins such as
endothelial protein C receptor, antithrombin, and thrombomodulin are downregulated. Fibrinolytic impairment also contributes to
hypercoagulability. Neutrophil activation and the release of neutrophil extracellular traps further stimulate thrombosis, and platelets
in COVID-19 may exhibit marked hyperreactivity. SARS-CoV-2 spike proteins may directly amplify complement, and contribute to vas-
cular injury and microthrombosis. The coagulation derangements observed in COVID-19 are a direct result of microthrombosis, which
is initially localized to the lungs, causing elevations in D-dimer and fibrinogen. As the disease progresses, more systemic inflammation
and the activation of coagulation lead to thrombocytopenia and prolongation of clotting times. In some patients this may progress
to a frank consumptive coagulopathy.

Pathogenesis of CAC
Thromboinflammation is cen­tral to the path­o­phys­i­­ol­ogy of CAC,
which is driven by a com­bi­na­tion of direct effects of viral infec­tion
on pul­mo­nary vas­cu­lar endo­the­lium, a local and sys­temic inflam­
ma­tory response, and cross talk between the inflam­ma­tory and
coag­u­la­tion path­ways (Figure 2). Pathogenic mech­a­nisms con­trib­
ut­ing to CAC have been reviewed in detail by Iba et al13: (1) direct
viral effects and inflam­ma­tion: SARS-CoV-2 enters respi­ra­tory epi­
the­lium and endo­the­lium via the angiotensin-converting enzyme 2
recep­tor. Direct endo­the­lial effects include procoagulant changes
such as sur­face adhe­sion mol­e­cule expres­sion and the release of
VWF and FVIII. Viral rep­li­ca­tion and shed­ding trig­ger an inflam­ma­
tory response, or “cyto­kine storm.” Proinflammatory cyto­kines IL-
1β, IL-6, and tumor necro­ sis fac­ tor upregulate procoagulant
mol­e­cules, includ­ing tis­sue fac­tor (TF), P-selectin, fibrin­o­gen, FVIII,
and VWF; (2) neu­tro­phil extra­cel­lu­lar traps: neu­tro­phil acti­va­tion
and release of neu­tro­phil extra­cel­lu­lar traps pro­mote plate­let acti­
va­tion, phosphatidylserine expo­sure, acti­va­tion of fac­tor V and
fac­tor XI, and throm­bin gen­er­a­tion26; (3) com­ple­ment acti­va­tion:
SARS-CoV-2 spike pro­teins directly acti­vate com­ple­ment,27 and
com­ple­ment depo­si­tion and microthrombosis are observed in the
lungs, skin, kid­neys, and hearts of patients with severe dis­ease28;
Prakash Singh Shekhawat
(4) effects on coag­u­la­tion and fibri­no­lytic sys­tems: hyp­oxia and
vaso­con­stric­tion induce the expres­sion of TF and plas­min­o­gen
acti­va­tor inhibitor disrupting the bal­ance between coag­u­la­tion
and fibri­no­ly­sis.29 Patients with COVID-19 have increased throm­bin
and plas­min gen­er­a­tion poten­tial and an increase in fibrin for­ma­
tion,25 con­sis­tent with other reports of decreased fibri­no­ly­sis.11,12
Perturbation of the bal­ance between throm­bin gen­er­a­tion (fibrin
clot for­ma­tion) and plas­min gen­er­a­tion (fibri­no­lytic path­ways) is
crit­i­cal to the path­o­phys­i­­ol­ogy of COVID-19.11,21 In severe COVID-19,
the lag time to throm­ bin, plas­ min, and fibrin for­ ma­tion is in-
creased, suggesting a loss in coag­u­la­tion-acti­vat­ing mech­a­nisms
in severe dis­ease.25 While coag­u­la­tion acti­va­tion is ini­tially restrict-
ed to the lungs and man­i­fests as an increase in D-dimer, severe
dis­ease is char­ac­ter­ized by sys­temic inflam­ma­tion and dis­sem­i­
nated microthrombosis caus­ing a con­sump­tive coagulopathy.
Bleeding in COVID-19
Bleeding is infre­quent in the set­ting of CAC, which has a pre­dom­
i­nantly prothrombotic phe­no­type. Helms et al reported bleed­
ing com­pli­ca­tions in 2.7% of 150 patients with COVID-related
ARDS.15 Among 2773 hos­pi­tal­ized COVID-19 patients from New
York, major bleed­ing was reported in 1.9% not on ther­a­peu­tic
COVID-19 coagulopathy | 617
anticoagulation and in 3% of those on ther­a­peu­tic anticoagula-
tion.30 However, in an anal­y­sis of 400 patients hos­pi­tal­ized with
COVID-19 in Bos­ton, Al-Samkari et al found a higher over­all bleed­
ing rate of 4.8%; major bleed­ing occurred in 2.3%, includ­ing 5.6%
of crit­i­cally ill patients, and was asso­ci­ated with a plate­let count
<150 × 109/L at admis­sion (Odds Ratio 2.90 [1.05-7.99]).3 Prelimi-
nary data from a multiplatform, ran­dom­ized con­trolled trial found
a 0.9% and 1.9% rate of major bleed­ing in noncritically ill par­
tic­i­pants on pro­phy­lac­tic or ther­a­peu­tic anticoagulation, respec­
tively.31 Among crit­ i­
cally ill patients, how­ ever, major bleed­ ing
occurred in 3.1% on ther­ap ­ eu­tic anticoagulation and 2.4% on
thromboprophylaxis.32
Among patients with COVID-19 ARDS treated with extra­cor­
po­real mem­brane oxy­gen­at­ ion (ECMO), major bleed­ing occurs in
>40%, includ­ing intra­cra­nial bleed­ing in up to 12%, which is sig­nif­
i­cantly higher than the 2% to 6% bleed­ing rate reported in non-
COVID-19 ECMO cohorts.33,34 DIC, shear stress-induced acquired
von Willebrand syn­drome, COVID-19-asso­ci­ated endothelitis, and
inten­sive anticoagulation are impli­cated in severe bleed­ing on
ECMO.34 The anticoagulation required for ECMO and the acti­va­
tion of coag­u­la­tion from arti­fi­cial sur­faces may con­found inter­pre­
ta­tion of coag­u­la­tion stud­ies. Careful inter­pre­ta­tion is required
to bal­ance anticoagulation, bleed­ing, and throm­bo­sis in patients
receiv­ing ECMO.
Management of CAC
Which lab­o­ra­tory param­e­ters should be eval­u­ated in CAC?
In out­pa­tients with COVID-19, there is cur­rently insuf­fic ­ ient evi­
dence to rec­om­mend mon­i­tor­ing of coag­u­la­tion mark­ers, includ­
ing D-dimer, plate­let count, PT, and fibrin­o­gen, since these are
unlikely to have an impact on man­ age­ment. For hos­ pi­tal­
ized
patients, we agree with the interim rec­om­men­da­tions from the
ISTH and the Amer­i­can Society of Hematology to seri­ally mon­i­
tor D-dimer lev­els, plate­let count, PT, and fibrin­o­gen since these
help with risk strat­i­fi­ca­tion.35 Worsening of these param­e­ters may
war­rant more aggres­sive crit­i­cal care sup­port and con­sid­er­ation
of inves­ti­ga­tional ther­a­pies. D-dimer ele­va­tion is non­spe­cific and
need not prompt imag­ing for venous throm­bo­em­bo­lism (VTE)
in the absence of clin­i­cal symp­toms or signs of VTE.36 However,
increas­ing D-dimer in the set­ting of decreas­ing C-reac­tive pro­
tein may war­rant screen­ing for VTE.
Role of blood prod­uct trans­fu­sion
There is no evi­dence that blood com­po­nent ther­apy to cor­rect
abnor­mal lab­o­ra­tory param­e­ters improves out­comes in the ab-
sence of bleed­ing. In patients with bleed­ing or those need­ing
inva­sive pro­ce­dures, we sug­gest trans­fu­sion of plate­lets to main­
tain a plate­let count ≥50 × 109/L, cryoprecipitate if fibrin­og ­ en is
<150 mg/dL, and plasma for an inter­na­tional nor­mal­ized ratio ≥1.8
after correcting fibrin­ o­
gen. Four fac­ tor prothrombin complex
concentrate (25 units/kg) may be used instead of plasma in pa-
tients with severe coagulopathy and bleed­ing to avoid vol­ume
over­load that can worsen respi­ra­tory sta­tus.
Anticoagulation
As with other coagulopathies, treat­ment of the under­ly­ing con­
di­tion is the only defin­i­tive solu­tion. Based on the asso­ci­a­tion of
D-dimer and other coag­u­la­tion mark­ers with severe dis­ease, mor­
tal­ity, and throm­bo­sis, at least pro­phy­lac­tic-dose anticoagulation
is recommended in hos­pi­tal­ized patients with COVID-19.1,36,37 No-
618  |  Hematology 2021  |  ASH Education Program
tably, this rec­om­men­da­tion is iden­ti­cal to that for other acute-
ly ill inpa­tients. In the absence of sig­nif­i­cant renal impair­ment,
we encour­age the use of low-molec­u­lar-weight hep­a­rin (LMWH)
over unfractionated hep­ ar­in (UFH) to avoid pos­ si­
ble hep­a­
rin resis­tance with UFH. If UFH is used, we sug­gest mon­i­tor­ing
anti-factor Xa lev­els due to poten­tial confounding of the aPTT due
to ele­vated FVIII and fibrin­o­gen or LAs. Further, anticoagulation
rec­om­men­da­tions may need to be mod­i­fied for patients with
severe throm­bo­cy­to­pe­nia, bleed­ing risk, and extremes of body
weight. Early in the pan­demic, there were reports of throm­bo­sis
despite stan­dard thromboprophylaxis. Based on these reports
and extreme lab­o­ra­tory derange­ments, cli­ni­cians fre­quently
esca­lated anticoagulation to inter­me­di­ate or even ther­a­peu­tic
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/614/1851465/614gerber.pdf by guest on 13 December 2021
inten­sity. However, whether a higher inten­sity of anticoagulation
reduces mor­bid­ity or mor­tal­ity with­out an unac­cept­able increase
in bleed­ing risk remains uncer­tain and is being eval­u­ated in mul­
ti­ple tri­als.
Preliminary data from a multiplatform, ran­ dom­ized con­
trolled trial com­pris­ing 3 global stud­ies (REMAP-CAP, ATTACC,
and ACTIV-4A) com­par­ing the effects of stan­dard-dose vs full-
dose hep­a­rin on the pri­mary out­come, a com­pos­ite of 21-day
“organ sup­port-free” days and in-hos­pi­tal mor­tal­ity, were
recently made avail­­ able as pre­ prints.31,32 The enroll­
ment of
mod­er­ately ill patients was stopped early for supe­ri­or­ity; ther­
a­peu­tic anticoagulation reduced the require­ment for organ
sup­ port and mor­ tal­
ity (OR 1.29 [1.04-1.61]) irrespective of
base­line D-dimer.31 Meanwhile, the enroll­ment of crit­i­cally ill
patients was halted due to futil­ity, with no decrease in mor­tal­
ity and a higher rate of major bleed­ing with ther­a­peu­tic anti-
coagulation.32 These appar­ently dis­crep­ant results sug­gest
that per­haps ther­a­peu­tic hep­a­rin works best if started early.
The INSPIRATION study reported that in crit­i­cally ill patients
with COVID-19, inter­me­di­ate- vs pro­phy­lac­tic-dose LMWH did
not result in a decrease in the com­pos­ite end point of throm­
bo­sis, extra­cor­po­real mem­brane oxy­gen­a­tion, or mor­tal­ity
within 30 days.38 The mul­ti­na­tional prag­matic RAPID COVID
COAG trial (NCT04362085) is eval­u­at­ing ther­a­peu­tic vs pro­
phy­lac­tic doses of LMWH or UFH in hos­pi­tal­ized, noncritically
ill COVID-19 patients with ele­vated D-dimer on the com­pos­ite
out­come of ICU admis­sion, non­in­va­sive pos­i­tive pres­sure ven­
ti­la­tion, mechan­i­cal ven­ti­la­tion, and death at 28 days. The
ACTION trial, a prag­matic trial at 31 sites in Brazil, com­pared
in-hos­pi­tal ther­a­peu­tic anticoagulation with rivaroxaban
(92%) or enoxaparin followed by rivaroxaban until day 30 vs
pro­phy­lac­tic-dose enoxaparin or UFH in hos­pi­tal­ized patients
with ele­vated D-dimer. There was no dif­fer­ence in the pri­mary
effi­cacy out­come of time to death and dura­tion of hos­pi­tal­i­za­
tion and sup­ple­men­tal oxy­gen use through 30 days but sig­nif­
i­cantly more bleed­ing in the rivaroxaban group.39 Pending the
final results of these and other stud­ies, the Amer­i­can Society
of Hematology guide­line panel sug­gests pro­phy­lac­tic over
inter­me­di­ate or ther­a­peu­tic dos­ing in mod­er­ately and severely
ill patients with COVID-19 with­out con­firmed or suspected
VTE. Higher-inten­ sity anticoagulation may be pre­ ferred in
mod­er­ately ill (non-ICU, stage 2) patients deemed at high
throm­botic and low bleed­ing risk. Therapeutically dosed anti-
coagulation is appro­pri­ate for patients with preexisting indi­
ca­tions such as atrial fibril­la­tion and mechan­i­cal heart valves
and, if needed, to main­tain extra­cor­po­real cir­cuits and vas­cu­
lar access devices.
COVID-19 vac­cines: cytopenias, bleed­ing, and throm­bo­sis
The COVID-19 vac­ cines have proven to be highly safe and
effec­tive and remain the most prom­is­ing tool in containing the
COVID-19 pan­demic. Cases of immune throm­bo­cy­to­pe­nia and
bleed­ing, includ­ing life-threat­en­ing hem­or­rhage, have been diag­
nosed after expo­sure to the mRNA-based vac­cines pro­duced by
Moderna (mRNA-1273) and Pfizer-BioNTech (BNT162b2).40 Fortu-
nately, the major­ity of these respond to immune throm­bo­cy­to­
pe­nia-directed treat­ments such as intra­ve­nous Ig and cor­ti­co­ste­
roids.40 It is unclear whether this rep­re­sents a causal rela­tion­ship
or a coin­ci­den­tal asso­ci­a­tion; how­ever, a recent anal­y­sis from the
Vaccine Adverse Event Reporting System reported that the inci­
dence of throm­bo­cy­to­pe­nia after either mRNA vac­cine was 0.8
per mil­lion doses, which is no higher than the back­ground rate of
immune throm­bo­cy­to­pe­nia (3.3 per 100 000 adults).41
More recently, rare cases of a vac­ cine-induced throm­
botic throm­bo­cy­to­pe­nia (VITT) were reported after admin­is­
tra­tion of the ade­no­vi­ral vec­tor vac­cines (ChAdOx1 nCoV-19
and Ad26.COV2.S).42 The esti­ mated inci­ dence is 1 case per
100,000 expo­ sures with a slight female pre­ pon­der­ ance, and
it occurs more com­ monly in youn­ ger (<55-60 years) indi­
vid­ u­
als, although this could reflect the demo­ graph­ ics of
indi­vid­u­als vac­ci­
nated early. Occurring 4 to 30 days post­
vac­ci­na­tion, VITT is asso­ci­ated throm­bo­cy­to­pe­nia and
throm­bo­sis, par­tic­u­larly at unusual sites such as the cere­bral or
splanch­nic cir­cu­la­tions, along with high lev­els of plate­let-acti­vat­
ing IgG antibodies that rec­og­nize plate­let fac­tor 4 (PF4) in the
absence of hep­a­rin. Anti-PF4/hep­a­rin antibodies are detect­able
by enzyme-linked immu­no­sor­bent assays, but rapid immu­no­as­
say may yield false-neg­a­tive results. Functional assays should be
performed in the pres­ence and absence of PF4. The stan­dard
sero­to­nin release assay may yield false-neg­a­tive results as VITT
antibodies exhibit PF4-depen­dent activ­ity and may dem­on­strate
decreased reac­tiv­ity in the pres­ence of hep­a­rin. Patients with
VITT fre­quently have a con­sump­tive coagulopathy with ele­vated
D-dimer and hypofibrinogenemia. VITT is asso­ci­ated with high
mor­ tal­
ity, with cere­ bral hem­ or­
rhage as the lead­ ing cause of
death. Treatment with a nonheparin anti­co­ag­u­lant and intra­ve­
nous Ig is recommended, with plate­let trans­fu­sions reserved for
bleed­ing. While much is still to be learned about the inci­dence,
path­o­gen­e­sis, and opti­mal man­age­ment of VITT, given the mil­
li­ons of indi­vid­u­als vac­ci­nated world­wide, these hema­to­logic
com­pli­ca­tions are van­ish­ingly rare and should not hin­der vac­ci­
na­tion efforts.
CLINICAL CASE (Con­t in­u ed)
Our patient had find­ings typ­i­cal of CAC with an ele­vated d-dimer
and fibrin­o­gen with nor­mal plate­lets and PT. An LA was detected,
account­ ing for her prolonged aPTT. She received a pro­ phy­
lac­tic dose of LMWH in addi­tion to dexa­meth­a­sone. D-dimer
trended up to a peak of 4.1 mg/dL, and plate­let count dropped
to 118 × 109/L on day 5 of hos­pi­tal­i­za­tion. PT remained nor­mal.
Oxygen require­ments peaked at 4 L by nasal can­nula on day 9
and grad­u­ally improved over the fol­low­ing week. No throm­botic
events occurred. She was discharged home with­out out­pa­tient
thromboprophylaxis on day 15. Repeat test­ ing to deter­
mine
whether her LA is per­sis­tent was pend­ing at discharge.
Prakash Singh Shekhawat
Summary
Inflammation, intra­vas­cu­lar coag­u­la­tion acti­va­tion, and micro­vas­
cu­lar throm­bo­sis cause coag­u­la­tion derange­ments in COVID-19.
Elevated D-dimer and fibrin­ o­
gen lev­ els are the most com­ mon
find­ing and are pre­dic­tive of adverse out­comes, includ­ing the
need for crit­i­cal care and mor­tal­ity. In con­trast to DIC, throm­bo­
cy­to­pe­nia is rare, and the PT/aPTT are usu­ally nor­mal or min­i­mally
prolonged. Some patients exhibit aPTT pro­lon­ga­tion due to aPL.
Bleeding, while rare, may occur in crit­i­cally ill patients that develop
a con­sump­tive coagulopathy and those on ECMO. Blood-prod­uct
sup­port to cor­rect the coagulopathy is reserved for bleed­ing pa-
tients or those requir­ing inva­sive pro­ce­dures. The results of ongo­
ing tri­als eval­u­at­ing the impact of higher-inten­sity anticoagulation
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/614/1851465/614gerber.pdf by guest on 13 December 2021
on COVID-19 out­comes are eagerly awaited. Until then, all­hos­pi­
tal­ized patients should receive stan­dard thromboprophylaxis.
Acknowledgments
This work was supported by National Institutes of Health National
Heart, Lung, and Blood Institute grant K99HL150594 (to S.C) and
T32HL007525 (to G.F.G.) and an Amer­i­can Society of Hematology
Scholar Award (to S.C.).
Conflict-of-inter­est dis­clo­sure
Glo­ria F. Gerber: no com­pet­ing finan­cial inter­ests to declare.
Shruti Chaturvedi: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Glo­ria F. Gerber: the off-label use of prothrombin complex
­concentrate is discussed.
Shruti Chaturvedi: the off-label use of prothrombin complex
­concentrate is discussed.
Correspondence
Shruti Chaturvedi, Johns Hopkins University, Ross Research Bldg,
Rm 1025, 720 Rutland Ave, Baltimore, MD 21205; e-mail: schatur3@
jhmi​­.edu.
References
1. Iba T, Levy JH, Connors JM, Warkentin TE, Thachil J, Levi M. Managing
throm­bo­sis and car­dio­vas­cu­lar com­pli­ca­tions of COVID-19: answer­ing the
ques­tions in COVID-19-asso­ci­ated coagulopathy. Expert Rev Respir Med.
2021;15(8):1003-1011.
2. Iba T, Levy JH, Levi M, Connors JM, Thachil J. Coagulopathy of coronavirus
dis­ease 2019. Crit Care Med. 2020;48(9):1358-1364.
3. Al-Samkari H, Karp Leaf RS, Dzik WH, et  al. COVID-19 and coag­u­la­tion:
bleed­ing and throm­botic man­i­fes­ta­tions of SARS-CoV-2 infec­tion. Blood.
2020;136(4):489-500.
4. Nopp S, Moik F, Jilma B, Pabinger I, Ay C. Risk of venous throm­bo­em­bo­lism
in patients with COVID-19: a sys­tem­atic review and meta-anal­y­sis. Res Pr
Thromb Haemost. 2020;4(7):1178-1191.
5. Tang N, Li D, Wang X, Sun Z. Abnormal coag­u­la­tion param­e­ters are asso­
ci­ated with poor prog­no­sis in patients with novel coronavirus pneu­mo­nia.
J Thromb Haemost. 2020;18(4):844-847.
6. Zhou F, Yu T, Du R, et  al. Clinical course and risk fac­tors for mor­tal­ity of
adult inpa­tients with COVID-19 in Wuhan, China: a ret­ro­spec­tive cohort
study. Lancet. 2020;395(10229):1054-1062.
7. Zhang L, Yan X, Fan Q , et al. D-dimer lev­els on admis­sion to pre­dict in-hos­pi­tal
mor­tal­ity in patients with Covid-19. J Thromb Haemost. 2020;18(6):1324-
1329.
8. Guan WJ, Ni ZY, Hu Y, et al; China Medical Treatment Expert Group for
Covid-19. Clinical char­ac­ter­is­tics of coronavirus dis­ease 2019 in China. N
Engl J Med. 2020;382(18):1708-1720.
9. Taylor FB Jr, Toh CH, Hoots WK, Wada H, Levi M; Scientific Subcommit-
tee on Disseminated Intravascular Coagulation (DIC) of the International
Society on Thrombosis and Haemostasis (ISTH). Towards def­i­ni­tion, clin­i­cal
COVID-19 coagulopathy | 619
 and lab­o­ra­tory cri­te­ria, and a scor­ing sys­tem for dis­sem­i­nated intra­vas­cu­
lar coag­u­la­tion. Thromb Haemost. 2001;86(5):1327-1330.
10. Ding R, Wang Z, Lin Y, Liu B, Zhang Z, Ma X. Comparison of a new cri­te­ria
for sep­sis-induced coagulopathy and International Society on Thrombosis
and Haemostasis dis­sem­i­nated intra­vas­cu­lar coag­u­la­tion score in crit­i­cally
ill patients with sep­sis 3.0: a ret­ro­spec­tive study. Blood Coagul Fibrinoly-
sis. 2018;29(6):551-558.
11. Hardy M, Michaux I, Lessire S, et al. Prothrombotic dis­tur­bances of hemo­
sta­sis of patients with severe COVID-19: a pro­spec­tive lon­gi­tu­di­nal obser­
va­tional study. Thromb Res. 2021;197(Jan­u­ary):20-23.
12. Wright FL, Vogler TO, Moore EE, et  al. Fibrinolysis shut­down cor­re­la­tion
with throm­bo­em­bolic events in severe COVID-19 infec­tion. J Am Coll Surg.
2020;231(2):193-203.e1203e1.
13. Iba T, Warkentin TE, Thachil J, Levi M, Levy JH. Proposal of the def­i­ni­tion for
COVID-19-asso­ci­ated coagulopathy. J Clin Med. 2021;10(2):191.
14. Wu C, Chen X, Cai Y, et al. Risk fac­tors asso­ci­ated with acute respi­ra­tory
dis­tress syn­drome and death in patients with coronavirus dis­ease 2019
pneu­mo­nia in Wuhan, China. JAMA Intern Med. 2020;180(7):934.
15. Helms J, Tacquard C, Severac F, et al; CRICS TRIGGERSEP (Clinical Research
in Intensive Care and Sepsis Trial Group for Global Evaluation and Research
in Sepsis) Group. High risk of throm­bo­sis in patients with severe SARS-
CoV-2 infec­tion: a mul­ti­cen­ter pro­spec­tive cohort study. Intensive Care
Med. 2020;46(6):1089-1098.
16. Lippi G, Plebani M, Henry BM. Thrombocytopenia is asso­ci­ated with severe
coronavirus dis­ease 2019 (COVID-19) infec­tions: a meta-anal­y­sis. Clin Chim
Acta. 2020;506(July):145-148.
17. Lodigiani C, Iapichino G, Carenzo L, et al; Humanitas COVID-19 Task
Force. Venous and arte­rial throm­bo­em­bolic com­pli­ca­tions in COVID-19
patients admit­ted to an aca­demic hos­pi­tal in Milan, Italy. Thromb Res.
2020;191(July):9-14.
18. Thachil J, Cushman M, Srivastava A. A pro­ posal for stag­ ing COVID-19
coagulopathy. Res Pract Thromb Haemost. 2020;4(5):731-736.
19. Harzallah I, Debliquis A, Drenou B. Lupus anti­co­ag­u­lant is fre­quent in patients
with Covid-19. J Thromb Haemost. 2020;18(8):2064-2065.
20. Abdel-Wahab N, Talathi S, Lopez-Olivo MA, Suarez-Almazor ME. Risk of
devel­op­ing antiphospholipid antibodies fol­low­ing viral infec­tion: a sys­tem­
atic review and meta-anal­y­sis. Lupus. 2018;27(4):572-583.
21. Panigada M, Bottino N, Tagliabue P, et al. Hypercoagulability of COVID-19
patients in inten­sive care unit: a report of thromboelastography find­ings and
other param­e­ters of hemo­sta­sis. J Thromb Haemost. 2020;18(7):1738-1742.
22. Nougier C, Benoit R, Simon M, et al. Hypofibrinolytic state and high throm­
bin gen­er­a­tion may play a major role in SARS-COV2 asso­ci­ated throm­bo­
sis. J Thromb Haemost. 2020;18(9):2215-2219.
23. Blasi A, von Meijenfeldt FA, Adelmeijer J, et al. In vitro hyper­co­ag­u­la­bil­ity
and ongo­ing in vivo acti­va­tion of coag­u­la­tion and fibri­no­ly­sis in COVID-19
patients on anticoagulation. J Thromb Haemost. 2020;18(10):2646-2653.
24. Ranucci M, Ballotta A, Di Dedda U, et  al. The procoagulant pat­tern of
patients with COVID-19 acute respi­ ra­
tory dis­tress syn­drome. J Thromb
Haemost. 2020;18(7):1747-1751.
25. Bouck EG, Denorme F, Holle LA, et al. COVID-19 and sep­sis are asso­ci­ated
with dif­fer­ent abnor­mal­i­ties in plasma procoagulant and fibri­no­lytic activ­
ity. Arterioscler Thromb Vasc Biol. 2021;41(1):401-414.
26. Middleton EA, He XY, Denorme F, et al. Neutrophil extra­cel­lu­lar traps con­
trib­ute to immunothrombosis in COVID-19 acute respi­ra­tory dis­tress syn­
drome. Blood. 2020;136(10):1169-1179.
27. Yu J, Yuan X, Chen H, Chaturvedi S, Braunstein EM, Brodsky RA. Direct acti­
va­tion of the alter­na­tive com­ple­ment path­way by SARS-CoV-2 spike pro­
teins is blocked by fac­tor D inhi­bi­tion. Blood. 2020;136(18):2080-2089.
620  |  Hematology 2021  |  ASH Education Program
28. Magro C, Mulvey JJ, Berlin D, et al. Complement asso­ci­ated micro­vas­cu­lar
injury and throm­bo­sis in the path­o­gen­e­sis of severe COVID-19 infec­tion: a
report of five cases. Transl Res. 2020;220(June):1-13.
29. Connors JM, Levy JH. COVID-19 and its impli­ca­tions for throm­bo­sis and anti-
coagulation. Blood. 2020;135(23):2033-2040.
30. Paranjpe I, Fuster V, Lala A, et al. Association of Treatment Dose Anticoagu-
lation With In-Hospital Survival Among Hospitalized Patients With COVID-
19. J Am Coll Cardiol. 2020;76(1):122-124.
31. Lawler PR, Goligher EC, et al. Therapeutic anticoagulation in non-crit­i­cally
ill patients with Covid-19. medRxiv. Preprint posted online 17 May 2021.
32. Zarychanski R. Therapeutic anticoagulation in crit­i­cally ill patients with
Covid-19—pre­lim­i­nary report. medRxiv. Preprint posted online 12 March
2021.
33. Schmidt M, Hajage D, Lebreton G, et al; Groupe de Recherche Clinique en
Reanimation et Soins intensifs du Patient en Insuffisance Respiratoire aigue
(GRC-RESPIRE) Sorbonne Université; Paris-Sorbonne ECMO-COVID Inves-
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/614/1851465/614gerber.pdf by guest on 13 December 2021
tigators. Extracorporeal mem­brane oxy­gen­a­tion for severe acute respi­ra­
tory dis­tress syn­drome asso­ci­ated with COVID-19: a ret­ro­spec­tive cohort
study. Lancet Respir Med. 2020;8(11):1121-1131.
34. Lebreton G, Schmidt M, Ponnaiah M, et al; Paris ECMO-COVID-19 Investi-
gators. Extracorporeal mem­brane oxy­gen­a­tion net­work orga­ni­sa­tion and
clin­i­cal out­comes dur­ing the COVID-19 pan­demic in Greater Paris, France:
a multicentre cohort study. Lancet Respir Med. 2021;9(8):851-862.
35. Thachil J, Tang N, Gando S, et  al. ISTH interim guid­ance on rec­og­ni­tion
and man­ age­ ment of coagulopathy in COVID-19. J Thromb Haemost.
2020;18(5):1023-1026.
36. Cuker A, Tseng EK, Nieuwlaat R, et  al. Amer­i­can Society of Hematology
2021 guide­ lines on the use of anticoagulation for thromboprophylaxis
in patients with COVID-19. Blood Adv. 2021;5(3):872-888.
37. Barrett CD, Moore HB, Yaffe MB, Moore EE. ISTH interim guid­ance on rec­
og­ni­tion and man­age­ment of coagulopathy in COVID-19: a com­ment. J
Thromb Haemost. 2020;18(8):2060-2063.
38. Sadeghipour P, Talasaz AH, Rashidi F, et al; INSPIRATION Investigators.
Effect of inter­me­di­ate-dose vs stan­dard-dose pro­phy­lac­tic anticoagulation
on throm­botic events, extra­cor­po­real mem­brane oxy­gen­a­tion treat­ment,
or mor­tal­ity among patients with COVID-19 admit­ted to the inten­sive care
unit: the INSPIRATION ran­dom­ized clin­i­cal trial. JAMA. 2021;325(16):1620-
1630.
39. Lopes RD, de Barros E, Silva PGM, Furtado RHM, et al; ACTION Coalition
COVID-19 Brazil IV Investigators. Therapeutic ver­sus pro­phy­lac­tic anticoagu-
lation for patients admit­ted to hos­pi­tal with COVID-19 and ele­vated D-dimer
con­cen­tra­tion (ACTION): an open-label, multicentre, randomised, con­trolled
trial. Lancet. 2021;397(10291):2253-2263.
40. Lee EJ, Cines DB, Gernsheimer T, et al. Thrombocytopenia fol­low­ing Pfizer
and Moderna SARS-CoV-2 vac­ci­na­tion. Am J Hematol. 2021;96(5):534-537.
41. Welsh KJ, Baumblatt J, Chege W, Goud R, Nair N. Thrombocytopenia
includ­ing immune throm­bo­cy­to­pe­nia after receipt of mRNA COVID-19 vac­
cines reported to the Vaccine Adverse Event Reporting System (VAERS).
Vaccine. 2021;39(25):3329-3332.
42. Arepally GM, Ortel TL. Vaccine-induced immune throm­botic throm­bo­
cy­to­pe­nia: what we know and do not know. Blood. 2021;138(4):293-298.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000297
 THE COVID CRASH: LESSONS LEARNED FROM A WORLD ON PAUSE

COVID-19 and thrombosis: searching for evidence

Bright Thilagar,1 Mohammad Beidoun,2 Ruben Rhoades,3 and Scott Kaatz1

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/621/1852155/621thilagar.pdf by guest on 13 December 2021

Division of Hospital Medicine, Henry Ford Health System, Detroit, MI; 2Department of Internal Medicine, Henry Ford Health System, Detroit, MI;
1

Division of Hematology, Thomas Jefferson University, Philadelphia, PA

3

Early in the pandemic, COVID-19-related increases in rates of venous and arterial thromboembolism were seen. Many
observational studies suggested a benefit of prophylactic anticoagulation for hospitalized patients using various dosing
strategies. Randomized trials were initiated to compare the efficacy of these different options in acutely ill and critically
ill inpatients as the concept of immune-mediated inflammatory microthrombosis emerged. We present a case-based
review of how we approach thromboembolic prophylaxis in COVID-19 and briefly discuss the epidemiology, the patho-
physiology, and the rare occurrence of vaccine-induced thrombotic thrombocytopenia.

LEARNING OBJECTIVES
• Appreciate the incidence, risk factors, and pathophysiology of thrombosis in COVID­19 infection
• Accurately prescribe anticoagulation prophylaxis before, during, and after hospitalization
• Be able to recognize and diagnose vaccine­induced immune thrombotic thrombocytopenia

P <.001).4 Increased arterial thrombotic events were also

CLINICAL CASE
A 62­year­old physician had been rounding on the
COVID­19 ward early in the pandemic (April 2020) and con­
tracted the disease. After 1 week, he became hypoxic and
was admitted to the general medical floor, requiring 3L
nasal oxygen. He had no major comorbidities and received
prophylactic­dose low­molecular­weight heparin (LMWH).
Introduction
COVID­19 infection clearly increases the risk of thrombotic
events for hospitalized patients, but rates of reported inci­
dence have varied. A meta­analysis of retrospective studies
involving 64 503 patients showed that deep vein thrombo­
sis (DVT) had an overall prevalence of 11.2% and pulmonary
embolism (PE) of 7.8% in those needing hospitalization.1
Pooled rates of venous thromboembolism (VTE) were
higher in the intensive care unit (ICU) setting (27.9% vs 7.1%
in the ward).2 Studies screening patients for VTE reported
a prevalence rate of 25.2% compared to a rate of 12.7%
in those testing only symptomatic patients.1 When more
than 95% of the hospitalized patients received pharmaco­
logic VTE prophylaxis, rates were lower, at 3.1% for non­
ICU patients and 7.6% for ICU patients.3 Venous or arterial
thrombosis was independently associated with higher
mortality risk (hazard ratio [HR], 1.82; 95% CI, 1.54-2.15;
Prakash Singh Shekhawat
reported during hospitalization, with a prevalence of 3.9%
for coronary artery events and 1.6% for stroke.1 Acute
limb ischemia was reported in 0.3% to 1% of hospital­
ized patients, predominantly affecting men, with 18% of
patients suffering limb loss.5
Rates of postdischarge VTE in COVID­19 patients, how­
ever, were not as high. In a cohort from California, VTE
rates in COVID­19 patients were no higher than patients
testing negative for COVID­19 (1.8% vs 2.2%; P =.16).6 A
pre­/postpandemic comparison of postdischarge VTE did
not show a statistically significant risk of VTE in COVID­19
patients (odds ratio, 1.6; 95% CI, 0.77-3.1).7
Pathophysiology of thrombosis in COVID-19 infection
The coagulopathy of COVID­19 infection is complex, involv­
ing an interplay between endothelial cell injury, inflammation,
and coagulation. Infection of pulmonary alveolar cells causes
severe endothelial injury and is marked by a local inflamma­
tory infiltrate and microthrombi.8 Microthrombi in alveolar
capillaries and vascular congestion were present in nearly
45% of patients who died from acute respiratory complica­
tions due to COVID­19.9 Alveolar capillary microthrombi were
9 times more prevalent in those who died from COVID­19
infection compared to similar severe H1N1 infections, and
increased thrombotic complications were mostly noted in
those with longer hospital stays.8 In severe infection, this
COVID­19 and thrombosis: searching for evidence | 621
Table 1. Risk fac­tors to remem­ber in COVID-19-related throm­bo­sis4,37
COVID-19 infec­tion
Older age (>75)
Male
Obesity
Hypertension
Prior car­diac dis­ease
Active can­cer or recent anti­can­cer treat­ment
Elevated D-dimer, FDP, LA
VTE in lower-extrem­ity deep veins and PE com­monly seen
FDP, fibrin deg­ra­da­tion prod­ucts.
immune response can be exag­ger­ated and cause sys­temic hyper­
inflammation, marked by high lev­ els of proinflammatory cyto­
kines.10,11 The down­stream effects of endo­ the­
lial cell acti­
va­tion
and a sys­temic inflam­ma­tory response include a hyper­co­ag­u­la­ble
state and both micro- and macrovascular throm­bo­sis, with mul­ti­
ple inter­re­lated mech­a­nisms.
Activation of endo­the­lial cells, plate­lets,
and leu­ko­cytes
Direct infec­tion of vas­cu­lar endo­the­lial cells with resultant apo­pto­
sis has been described in an autopsy series of COVID-19 patients.12
This, along with com­ple­ment acti­va­tion,13 con­trib­utes to a local
inflam­ma­tory reac­tion and fur­ther endo­the­lial cell acti­va­tion, with
the expres­sion of tis­sue fac­tor, release of von Willebrand fac­tor
(VWF), and decreased syn­the­sis of nitric oxide and pros­ta­cy­clin,14
all­ of which pro­mote coag­u­la­tion. An increase in VWF cou­pled
with a rel­a­tive imbal­ance in ADAMTS-13 has been dem­on­strated
in patients with COVID-19 of vary­ing sever­ity,15 and ele­vated VWF
anti­gen and low ADAMTS-13 activ­ity seem to cor­re­late with the
devel­op­ment of VTE.16 Platelet acti­ va­
tion increases in severe
COVID-19 infec­tion, trig­ger­ing increased tis­sue fac­tor expres­sion
and coag­u­la­tion.14,17,18 Similarly, neu­tro­phil acti­va­tion increases, with
the release of neu­tro­phil extra­cel­lu­lar traps that pro­mote throm­bo­
sis in these patients.19
Coagulation and fibri­no­ly­sis
The increase in tis­ sue fac­ tor expres­sion on a vari­ ety of cells
induces extrin­sic coag­u­la­tion path­way acti­va­tion, while the
release of fac­tor VIII from dam­aged endo­the­lium, NETosis, and
plate­let and com­ple­ment acti­va­tion all­ con­trib­ute to intrin­sic
path­way acti­va­tion. These events result in throm­bin gen­er­a­
tion and the for­ma­tion of a fibrin clot. COVID-19 infec­tion is also
marked by high lev­els of fibrin­o­gen,10,11 an acute phase reac­tant,
and the inhi­bi­tion of fibri­no­lytic path­ways. Patients with severe
infec­tion and shut­down of fibri­no­ly­sis—as mea­sured by a lack
of clot lysis on vis­co­elas­tic test­ing—were found to have greater
dimerized plas­min frag­ment D (D-dimer) and fibrin­o­gen lev­els
and a sig­nif­i­cantly greater inci­dence of clin­i­cal VTE (Table 1).20
Heparin-induced throm­bo­cy­to­pe­nia
Given the pro­pen­sity for throm­bo­sis, patients admit­ted with
COVID-19 are fre­quently man­aged with anticoagulation pro­phy­
laxis. This, com­bined with the sys­temic inflam­ma­tory response,
622  |  Hematology 2021  |  ASH Education Program
COVID-19 (ade­no­vi­rus) vec­tor vac­cines
Middle age (18-49)
Female
Platelet count <150
Splanchnic vein and cere­bral vein throm­bo­sis reported in higher rates
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/621/1852155/621thilagar.pdf by guest on 13 December 2021
increases the like­li­hood of devel­op­ment of hep­a­rin-induced
throm­bo­cy­to­pe­nia (HIT), another poten­tial eti­­ol­ogy of VTE in
these patients. Rates of HIT in severe COVID-19 infec­tion are
var­i­able across stud­ies, with rates rang­ing from 0.16% in all­
admit­ted patients to 8.1% in an ICU pop­u­la­tion. Anti-plate­let
fac­tor 4 (PF4) antibodies are fre­quently detected even in the
absence of clin­i­cal HIT,21,22 and 1 group has dem­on­strated non-
hep­a­rin-depen­dent plate­let-acti­vat­ing immune complexes that
may con­trib­ute to throm­bo­sis.23
Antiphospholipid antibodies
A final poten­ tial mech­ a­
nism for throm­ bo­sis in patients with
COVID-19 is the devel­op­ment of antiphospholipid antibodies,
an appeal­ ing the­ory given the hyperinflammatory response
and poten­tial for anti­body pro­duc­tion. As many as 90% of crit­
i­cally ill patients have been found to be pos­i­tive for a lupus
anti­co­ag­u­lant (LA).24,25 The clin­ic
­ al sig­nif­i­cance of this is unclear,
as an asso­ci­a­tion with throm­bo­sis has not been con­sis­tently
dem­on­strated,25 assays may be sus­ cep­ ti­
ble to inter­ fer­
ence
by C-reac­tive pro­tein and unfractionated hep­a­rin, and rates
of anticardiolipin and anti-β2-gly­co­pro­tein seem to be sig­nif­
i­cantly lower.24-26 However, 1 study at a ter­tiary care cen­ter in
the US found a sig­nif­i­cant asso­ci­a­tion between LA pos­i­tiv­ity
and venous or arte­rial throm­bo­sis.26
CLINICAL CASE (Con­tin­ued)
Our patient was a real one and in fact is the senior author of this
arti­cle. He received pro­phy­lac­tic-dose enoxaparin while on the
med­i­cal floor.
Anticoagulation pro­phy­laxis
When choos­ing a pro­phy­lac­tic approach, the risk of bleed­ing must
be con­sid­ered, par­tic­u­larly if the plate­let count is <50 000, and
mechan­i­cal pro­phy­laxis is likely the best approach until the bleed­
ing risk abates. The dose and choice of anti­co­ag­u­lant also depend
on renal func­ tion, and it may be rea­ son­ able to give “obe­ sity-
adjusted” doses—eg, enoxaparin 40 mg twice daily or a daily total
dose of 0.5 mg/kg based on lim­ited evi­dence.27 Of note, obe­sity-
adjusted pro­phy­lac­tic dos­ing dif­fers from inter­me­di­ate dos­ing.
Noncritically ill patients
The Amer­ i­
can Society of Hematology (ASH) draft guide­
lines (updated 8 Feb­ru­ary 2021) sug­gest pro­phy­lac­tic over
inter­me­di­ate- or ther­a­peu­tic-dose anticoagulation for patients
with acute (not crit­i­cal) ill­ness and acknowl­edge pend­ing results
of the com­bined anal­y­sis of the REMAP-CAP, ACTIV-4, and ATTAC
multiplatform ran­dom­ized con­trolled tri­als (mpRCT).28 These
3 mpRCTs har­mo­nized their pro­to­cols to accel­er­ate the bat­tle
against COVID-19 infec­tion and reported their pooled results in
sep­a­rate papers for crit­i­cally ill and noncritically ill patients.29,30
The first patient was ran­dom­ized on 21 April 2020, and these
tri­
als were stopped on 22 Jan­ u­
ary 2021. The pre­ print report
appeared on 17 May 2021 when the prespecified supe­ri­or­ity stop­
ping rule thresh­old was attained.30 The pri­mary anal­y­sis pop­u­la­
tion had 2219 par­tic­i­pants with con­firmed COVID-19 who did not
require ICU-level organ sup­port, which was defined as high-flow
oxy­gen, mechan­i­cal ven­ti­la­tion (inva­sive or non­in­va­sive), vaso­pres­
sors, or inotropes. They were ran­dom­ized in an open-label man­
ner to ther­a­peu­tic-dose hep­a­rin (94.7% received LMWH, mostly
enoxaparin) or usual care thromboprophylaxis (71.7% low dose
and 26.5% inter­me­di­ate dose) for up to 14 days. The pri­mary out­
come of sur­vival to hos­pi­tal dis­charge with­out organ sup­port (as
defined above) through 21 days occurred in 76.4% of par­tic­i­pants
receiv­ing the usual care and increased by 4.6% with ther­a­peu­tic-
dose anticoagulation, with a median adjusted odds ratio of 1.29%
and a 99% prob­a­bil­ity of the ther­a­peu­tic dose being effec­tive. The
prob­a­bil­ity of sur­vival to hos­pi­tal dis­charge with ther­a­peu­tic-dose
hep­a­rin was 87.1%, with a median abso­lute improve­ment of 1.3%.
Major bleed­ing occurred in 1.9% and 0.9% of ther­a­peu­tic-dose and
usual care par­tic­i­pants, respec­tively. A prespecified anal­y­sis based
on D-dimer lev­els showed a slightly bet­ter prob­a­bil­ity of supe­ri­or­
ity in patients with high lev­els (97.3%) vs low lev­els (92.9%).
The ACTION trial ran­dom­ized 615 hos­pi­tal­ized patients (who
were pri­mar­ily sta­ble and not crit­i­cally ill) in 31 Brazilian sites with
ele­vated D-dimer lev­ els above the upper limit of nor­ mal from
24 June 2020 to 26 Feb­ru­ary 2021 to ther­a­peu­tic doses of rivarox­
Figure 1. ASH recommendations for VTE prophylaxis in COVID patients with critical illness. Reproduced with permission from
Cuker et al.28
Prakash Singh Shekhawat
aban (20 mg or renally adjusted 15 mg daily) for 30 days or stan­dard
VTE pro­phy­laxis.31 Clinically unsta­ble patients (10%) ran­dom­ized to
rivaroxaban first received enoxaparin or unfractionated hep­a­rin
(1 patient) followed by rivaroxaban when clin­i­cally sta­ble. Patients
ran­dom­ized to stan­dard pro­phy­laxis primarily received enoxaparin
(84%), and 13% con­tin­ued postdischarge pro­phy­laxis at their cli­ni­
cian’s dis­cre­tion. The hier­ar­chi­cal anal­y­sis of time to death, dura­
tion of hos­pi­tal­i­za­tion, and dura­tion of sup­ple­men­tal oxy­gen was
performed using the win ratio, and there was no dif­fer­ence in effi­
cacy with a win ratio of 0.86; P  =  .40. There was also no dif­fer­ence in
each com­po­nent of this com­pos­ite out­come or throm­bo­em­bolic
event. However, major or clin­i­cally rel­e­vant non­ma­jor bleed­ing
was increased with ther­a­peu­tic anticoagulation (8%) com­pared to
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/621/1852155/621thilagar.pdf by guest on 13 December 2021
stan­dard dos­ing (2%) with a rel­a­tive risk of 3.64 ( P  =  .001).
Until the 3 mpRCTs are fully published, it is rea­son­able to use
either ther­a­peu­tic or pro­phy­lac­tic doses of LMWH (pre­ferred if
renal func­tion is accept­able) or hep­a­rin. Direct oral anti­co­ag­u­
lants (DOACs) should not be used unless there is another indi­ca­
tion, such as atrial fibril­la­tion.
How we treat noncritically ill hos­pi­tal­ized patients
We dis­cuss the risks and ben­e­fits of pro­phy­lac­tic vs ther­
a­peu­tic doses of anticoagulation and their impact on the
need for organ sup­port in noncritically ill patients hos­pi­tal­
ized with COVID-19. We use ther­a­peu­tic dos­ing of LMWH
or unfractionated hep­a­rin based on these early results and
eagerly await peer-reviewed pub­ li­
ca­
tion and guide­ line
updates.
CLINICAL CASE (Con­tin­ued)
Oxygen require­ ment increased for our patient, and a high
flow was needed with trans­fer to the ICU, where LMWH was
increased to inter­me­di­ate dose. Several days later, intu­ba­tion
COVID-19 and throm­bo­sis: searching for evi­dence  |  623
and mechan­i­cal ven­ti­la­tion were required. Meanwhile, the
patient’s father was also crit­i­cally ill with COVID-19, and his cli­
ni­cians used ther­a­peu­tic-dose LMWH: 2 patients with the same
last name, 1 floor apart, with dif­fer­ent doses to pre­vent the
same com­pli­ca­tions—all­ were searching for evi­dence.
Critically ill patients
ASH guide­lines sug­gest pro­phy­lac­tic- over inter­me­di­ate-dose
anticoagulation in crit­i­cally ill patients, and rec­om­men­da­tions
for ther­a­peu­tic dose vs pro­phy­lac­tic dose are forthcoming
(Figure 1).
Intermediate dose.  Among 600 ran­dom­ized COVID-19 patients in
the ICU, 562 were included in the pri­mary anal­y­sis of the INSPIRATION
trial.32 Intermediate enoxaparin doses of 1 mg/kg once daily vs
40 mg daily (with adjust­ment for weight and cre­at­i­nine clear­ance)
were com­pared for the pri­mary com­pos­ite effi­cacy out­come of
venous or arte­rial throm­bo­sis, treat­ment with extra­cor­po­real
mem­brane oxy­gen­a­tion, or mor­tal­ity. No dif­fer­ences were noted in
30-day out­comes between inter­me­di­ate and pro­phy­lac­tic doses
(45.7% vs 44.1%; odds ratio, 1.06; P   =   .70). Major bleed­ing occurred
in 2.5% and 1.4% of patients who received inter­ me­ di­
ate and
­pro­phy­lac­tic dos­ing, respec­tively, and the risk dif­fer­ence failed
to meet the noninferiority mar­gin. Severe throm­bo­cy­to­pe­nia
occurred in 6 patients, all­of whom received an inter­me­di­ate dose.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/621/1852155/621thilagar.pdf by guest on 13 December 2021

Therapeutic dose.  Data for crit­i­cally ill par­tic­i­pants ran­dom­ized
while receiv­ing organ sup­port as pre­vi­ously described in the How we treat crit­i­cally ill patients
3 mpRCTs were posted on 12 March 2021.29 This prespecified
cohort stopped recruit­ ment for crit­ i­
cally ill patients on 19 We use pro­phy­lac­tic- and not inter­me­di­ate- or ther­a­peu­tic-
Decem­ber 2020 because of futil­ity. The pri­mary out­come of sur­ dose LMWH or hep­a­rin in crit­i­cally ill patients with COVID-19.
vival until dis­charge and the median num­ber of organ-sup­port-
free days was 3 in those ran­dom­ized to ther­ap ­ eu­tic-dose LMWH
or hep­a­rin and 5 days with the usual care (41% pro­phy­lac­tic-
dose and 51% inter­me­di­ate-dose LMWH or hep­a­rin; ­adjusted CLINICAL CASE (Con­tin­ued)
odds ratio, 0.87; 95% cred­i­ble inter­val, 0.70-1.08; like­li­hood of After a cou­ple of weeks in the ICU with respi­ra­tory fail­ure and
not achiev­ing mean­ing­ful rel­at­ ive improve­ment of at least 20%, sep­tic shock, our patient was extubated but suf­fered a small PE
99.8%). while his cen­tral line was being removed. He was treated for

Table 2. List of selected anticoagulation tri­als in COVID-19

Patient Trial name and

pop­u­la­tion trial iden­ti­fier Intervention Comparison Outcome Status
Hospitalized ACTION Trial 31
Therapeutic rivaroxaban Usual care thromboprophylaxis Mortality, length of stay, and oxy­gen use Published
(non-ICU) or LMWH
ATTACC30 Therapeutic hep­a­rin Usual care thromboprophylaxis Hospital dis­charge with­out need for Preprint
NCT04372589 or LMWH organ sup­port
ACTIV-4a30 Therapeutic hep­a­rin Usual care thromboprophylaxis Hospital dis­charge with­out need for Preprint
NCT04505774 or LMWH organ sup­port
REMAP-CAP30 Therapeutic hep­a­rin Usual care thromboprophylaxis Hospital dis­charge with­out need for Preprint
NCT02735707 or LMWH organ sup­port
Hospitalized ATTACC29 Therapeutic hep­a­rin Usual care thromboprophylaxis Hospital dis­charge with­out need for Preprint
(ICU) NCT04372589 or LMWH organ sup­port
ACTIV-4a29 Therapeutic hep­a­rin Usual care thromboprophylaxis Hospital dis­charge with­out need for Preprint
NCT04505774 or LMWH organ sup­port
REMAP-CAP29 Therapeutic hep­a­rin Usual care thromboprophylaxis Hospital dis­charge with­out need for Preprint
NCT02735707 or LMWH organ sup­port
INSPIRATION32 Intermediate-dose Prophylactic-dose hep­a­rin or Composite of VTE, arte­rial throm­bo­sis, Published
NCT04486508 hep­a­rin or LMWH LMWH or ECMO
Post dis­charge ACTIV-4 Prophylactic-dose Placebo Composite out­come of symp­tom­atic DVT, Recruiting
Convalescent apixaban PE, other VTE, ische­mic stroke, acute
NCT04650087 MI, other ATE, and all­-cause mor­tal­ity
Outpatient ACTIV-4c Prophylactic-dose Placebo Composite of symp­tom­atic DVT or PE, Recruiting
NCT04498273 apixaban, ther­a­peu­tic- ATE, MI, ische­mic stroke, hos­pi­tal­i­za­tion
dose apixaban, ASA for car­dio­vas­cu­lar/pul­mo­nary events,
and all­-cause mor­tal­ity
PREVENT-HD Prophylactic-dose Placebo Composite of symp­tom­atic VTE, MI, Recruiting
NCT04508023 rivaroxaban ische­mic stroke, sys­temic embolism,
acute limb ische­mia, all­-cause
hos­pi­tal­i­za­tion, and all­-cause mor­tal­ity
ASA, aspi­rin; ECMO, extra­cor­po­real mem­brane oxy­gen­a­tion; MI, myo­car­dial infarc­tion.

624  |  Hematology 2021  |  ASH Education Program

3 months with a DOAC, which pre­cluded the deci­sion regard­
ing post-hos­pi­tal dis­charge anti­co­ag­u­lant pro­phy­laxis.
Postdischarge pro­phy­laxis
This patient had symp­toms of PE with acute chest pain and
wors­ened hyp­ox­emia and was sta­ble for com­puted tomog­ra­
phy, ren­der­ing a straight­for­ward workup. We only pur­sue diag­
nos­ing VTE based on clin­i­cal sus­pi­cion and do not screen or
use D-dimer thresh­olds to screen in accor­dance with guid­ance
state­ments.33
As discussed pre­vi­ously, rates of VTE are rel­a­tively low post
hos­pi­tal­i­za­tion; how­ever, mit­i­ga­tion of arte­rial and VTE along
with reduc­tions in all­-cause mor­tal­ity deserves inves­ti­ga­tion. A
pro­spec­tive reg­is­try of 4906 patients with COVID-19 were con­
tacted a mean of 92 days after dis­charge, and 1.55% devel­oped
VTE, arte­rial throm­bo­em­bolic events (ATE) occurred in 1.71%,
and all­-cause mor­tal­ity was 4.83%. There was a 46% rel­a­tive
reduc­tion in the com­pos­ite out­come of VTE, ATE, and mor­tal­ity
(P = .0046) in the mul­ti­var­i­able anal­y­sis.34 The ACTIV-4b National
Institutes of Health-spon­sored trial is inves­ti­gat­ing whether apix­
aban at pro­phy­lac­tic or ther­a­peu­tic doses and low-dose aspi­rin,

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/621/1852155/621thilagar.pdf by guest on 13 December 2021

Figure 2. VITT diagnostic flowchart from ISTH interim guidance published 20 April 2021. aPTT, activated partial thromboplastin time;
IVIG, intravenous immunoglobulin, PT, prothrombin time. Reproduced with permission from the ISTH.41
Prakash Singh Shekhawat
COVID-19 and throm­bo­sis: searching for evi­dence  |  625
com­pared to pla­cebo, can reduce these com­pli­ca­tions after hos­
pi­tal­i­za­tion. The stud­ies are sum­ma­rized in Table 2.
The US Food and Drug Administration has approved rivarox­
aban and betrixaban for postdischarge VTE pro­phy­laxis; how­
ever, betrixaban is not com­mer­cially avail­­able. The pro­phy­lac­tic
dose of 10 mg of rivaroxaban for a total of 31 to 39 days has been
approved but should not be used in patients with cre­a­tine clear­
ance less than 30 mL/min, with drug-drug interactions and high
bleeding-risk conditions.27
How we treat patients after dis­charge
We only use postdischarge pro­phy­laxis in selected patients
in accor­dance with US Food and Drug Administration
approval for med­i­cally ill patients with­out increased bleed­
ing risk using a DOAC.
“Prehospitalization” pro­phy­laxis
The trial results above sug­gest that higher doses of anticoag­
ulation are not effec­tive late in the dis­ease course, and it is a
rea­son­able hypoth­e­sis that anticoagulation very early in the dis­
ease may be ben­e­fi­cial to decrease immuno-micro pul­mo­nary
throm­bo­sis. Two tri­als with DOACs are under­way to help answer
this ques­tion (Table 2).
How we treat non­hos­pi­tal­ized out­pa­tients
We do not use anticoagulation pro­phy­laxis early in COVID-
19 for patients who do not require hos­pi­tal­i­za­tion.
Postvaccination: vac­cine-induced immune throm­botic
throm­bo­cy­to­pe­nia
The approval and wide­spread admin­is­tra­tion of vac­cines for
COVID-19 brought yet another ele­ment into the dis­cus­sion of
COVID-19 and throm­bo­sis. Vaccine-induced immune throm­botic
throm­bo­cy­to­pe­nia (VITT) or throm­bo­sis with throm­bo­cy­to­pe­nia
syndrome is the name given to the throm­botic com­pli­ca­tions
cur­rently being reported across the world fol­low­ing the admin­
is­tra­tion of the ade­no­vi­rus vec­tor vac­cines (ChAdOx1 nCoV-19
[AstraZeneca/COVIDSHIELD] and Ad26.COV2.S [Johnson and
Johnson/Janssen]) against SARS-CoV-2.35 The under­ly­ing path­
o­phys­i­­ol­ogy of VITT is not fully under­stood. One hypoth­e­sis is
that the polyanionic con­stit­u­ents of the ade­no­vi­rus vec­tor vac­
cines cause the for­ma­tion of antibodies to PF4. These antibodies
against PF4 then induce plate­let acti­va­tion, caus­ing both throm­
bo­cy­to­pe­nia and throm­bo­sis. VITT is sim­i­lar to auto­im­mune hep­
a­rin-induced throm­bo­cy­to­pe­nia (aHIT), with the pres­ence of
anti-PF4 polyanion antibodies induc­ing plate­let acti­va­tion in the
absence of and inde­pen­dently of hep­a­rin expo­sure; how­ever,
VITT and aHIT dif­fer in the dis­tri­bu­tion of the thrombi.36,37
VITT gen­er­ally pres­ents between 4 to 30 days after ini­tial vac­
ci­na­tion with either ade­no­vi­rus vec­tor vac­cine.35 Early reports
showed that VITT pri­mar­ily presented in youn­ger females. Data
on the inci­dence range from 1 case per 26 000 to 1 case per
127 000 doses of ChAdOx1 nCoV-19 vac­cine admin­is­tered and
1 case per 500 000 doses of the Ad26.COV2.S. No spe­cific risk
fac­tors have been iden­ti­fied. In VITT, throm­bo­sis tends to man­i­
fest in rare sites, most nota­bly in the cere­bral venous sinuses and
in the splanch­nic and hepatic veins.35,38
626  |  Hematology 2021  |  ASH Education Program
On pre­sen­ta­tion, lab­o­ra­tory test­ing gen­er­ally shows vary­
ing lev­els of throm­bo­cy­to­pe­nia, high D-dimer lev­els, and low
fibrin­o­gen lev­els.35,38,39 Diagnostic test­ing for PF4 antibodies via
enzyme-linked immu­no­sor­bent assay (ELISA) results is one of
the more sen­si­tive test­ing modal­i­ties for VITT, while the rapid
HIT tests such as the chemi­lu­mi­nes­cent immunoassay and latex
immunoturbidometric assay are not sen­ si­
tive and can yield
false-neg­a­tive results.38 Principles of man­age­ment mir­ror those
of aHIT, with the goal of inhibiting Fcγ recep­tor-medi­ated plate­
let acti­va­tion through the use of intra­ve­nous immune glob­u­lin
(with suggested daily dos­ing of 0.5-1 g/kg of ideal body weight
for at least 2 days) and the use of either oral or paren­tal nonhep­
arin anti­co­ag­u­lants.37 (See Figure 2 for man­age­ment sug­ges­tions
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/621/1852155/621thilagar.pdf by guest on 13 December 2021
from the International Society on Thrombosis and Hemostasis
[ISTH].)40 Vaccination with Ad26.COV2.S car­ries a warn­ing about
the poten­tial devel­op­ment of VITT, espe­cially in women aged 18
to 49.36
Conclusion
The throm­bo­em­bolic bur­den of COVID-19 seems to be decreas­
ing as the pan­demic evolves, and a mul­ti­tude of mech­an ­ isms
are pur­ported to explain the ele­vated risk with this infec­tion.
Key to under­stand­ing the tim­ing, dos­ing, and patient set­ting for
pro­phy­laxis is an appre­ci­a­tion of immune-induced inflam­ma­tory
microthrombosis. Robust ran­dom­ized tri­als are the best tools to
under­stand the appro­pri­ate use of anticoagulation to pre­vent
throm­bo­em­bolic com­pli­ca­tions, and the final results of many
ongo­ing stud­ies are eagerly awaited.
Conflict-of-inter­est dis­clo­sure
Bright Thilagar: no competing financial interest to declare.
Mohammad Beidoun: no competing financial interest to declare.
Ruben Rhoades: no competing financial interest to declare.
Scott Kaatz: research funding: Janssen, BMS, Osmosis R ­ esearch,
National Institutes of Health; con­sul­tancy: Janssen, BMS, ­Alexion/
Portola, Novartis, CSL Behring, Gilead.
Off-label drug use
Bright Thilagar: nothing to disclose.
Mohammad Beidoun: nothing to disclose.
Ruben Rhoades: nothing to disclose.
Scott Kaatz: nothing to disclose.
Scott Kaatz’s institution received research support from Osmosis
Research for one of the multi-platform trials using therapeutic
dose anticoagulation prophylaxis.
Correspondence
Scott Kaatz, Division of Hospital Medicine, CFP 413, 2799 W
Grand Blvd, Detroit, MI 48202; e-mail: skaatz1@hfhs​­.org.
References
1. Tan BK, Mainbourg S, Friggeri A, et al. Arterial and venous throm­bo­em­bo­
lism in COVID-19: a study-level meta-anal­y­sis. Thorax. Published online 23
Feb­ru­ary 2021.
2. Jiménez D, García-Sanchez A, Rali P, et al. Incidence of VTE and bleed­ing
among hos­pi­tal­ized patients with coronavirus dis­ease 2019: a sys­tem­atic
review and meta-anal­y­sis. Chest. 2021;159(3):1182-1196.
3. Al-Samkari H, Karp Leaf RS, Dzik WH, et  al. COVID-19 and coag­u­la­tion:
bleed­ing and throm­botic man­i­fes­ta­tions of SARS-CoV-2 infec­tion. Blood.
2020;136(4):489-500.
4. Bilaloglu S, Aphinyanaphongs Y, Jones S, Iturrate E, Hochman J, Berger
JS. Thrombosis in hos­pi­tal­ized patients with COVID-19 in a New York City
Health System. JAMA. 2020;324(8):799-801.
5. Etkin Y, Conway AM, Silpe J, et  al. Acute arte­ rial throm­
bo­ em­bo­lism in
patients with COVID-19 in the New York City Area. Ann Vasc Surg. 2021
(Jan­u­ary);70:290-294.
6. Roubinian NH, Dusendang JR, Mark DG, et al. Incidence of 30-day venous
throm­bo­em­bo­lism in adults tested for SARS-CoV-2 infec­tion in an inte­grated
health care sys­tem in Northern California. JAMA Intern Med. 2021;181(7):997-
1000.
7. Roberts LN, Whyte MB, Georgiou L, et al. Postdischarge venous throm­bo­
em­bo­lism fol­low­ing hos­pi­tal admis­sion with COVID-19. Blood. 2020;136(11):
1347-1350.
8. Ackermann M, Verleden SE, Kuehnel M, et  al. Pulmonary vas­cu­lar endo­
thelialitis, throm­bo­sis, and angio­gen­e­sis in COVID-19. N Engl J Med.
2020;383(2):120-128.
9. Menter T, Haslbauer JD, Nienhold R, et  al. Postmortem e ­ xam­
i­
na­tion of
COVID-19 patients reveals dif­fuse alve­o­lar dam­age with severe ­cap­il­lary
con­ges­tion and var­ie­gated find­ings in lungs and other organs ­suggesting
vas­cu­lar dys­func­tion. Histopathology. 2020;77(2):198-209.
10. Tang N, Li D, Wang X, Sun Z. Abnormal coag­u­la­tion param­e­ters are asso­
ci­ated with poor prog­no­sis in patients with novel coronavirus pneu­mo­nia.
J Thromb Haemost. 2020;18(4):844-847.
11. Ranucci M, Ballotta A, Di Dedda U, et  al. The procoagulant pat­tern of
patients with COVID-19 acute respi­ra­tory dis­ tress syn­drome. J Thromb
Haemost. 2020;18(7):1747-1751.
12. Varga Z, Flammer AJ, Steiger P, et al. Endothelial cell infec­tion and endo­
theliitis in COVID-19. Lancet. 2020;395(10234):1417-1418.
13. Yu J, Yuan X, Chen H, Chaturvedi S, Braunstein EM, Brodsky RA. Direct acti­
va­tion of the alter­na­tive com­ple­ment path­way by SARS-CoV-2 spike pro­
teins is blocked by fac­tor D inhi­bi­tion. Blood. 2020;136(18):2080-2089.
14. Canzano P, Brambilla M, Porro B, et al. Platelet and endo­the­lial acti­va­tion as
poten­tial mech­a­nisms behind the throm­botic com­pli­ca­tions of COVID-19
patients. JACC Basic Transl Sci. 2021;6(3):202-218.
15. Mancini I, Baronciani L, Artoni A, et al. The ADAMTS13-von Willebrand fac­tor
axis in COVID-19 patients. J Thromb Haemost. 2021;19(2):513-521.
16. Delrue M, Siguret V, Neuwirth M, et al. von Willebrand fac­tor/ADAMTS13 axis
and venous throm­bo­em­bo­lism in mod­er­ate-to-severe COVID-19 patients.
Br J Haematol. 2021;192(6):1097-1100.
17. Hottz ED, Azevedo-Quintanilha IG, Palhinha L, et al. Platelet acti­va­tion and
plate­let-mono­cyte aggre­gate for­ma­tion trig­ger tis­sue fac­tor expres­sion in
patients with severe COVID-19. Blood. 2020;136(11):1330-1341.
18. Althaus K, Marini I, Zlamal J, et al. Antibody-induced procoagulant plate­lets
in severe COVID-19 infec­tion. Blood. 2021;137(8):1061-1071.
19. Middleton EA, He XY, Denorme F, et al. Neutrophil extra­cel­lu­lar traps con­
trib­ute to immunothrombosis in COVID-19 acute respi­ra­tory dis­tress syn­
drome. Blood. 2020;136(10):1169-1179.
20. Wright FL, Vogler TO, Moore EE, et  al. Fibrinolysis shut­down cor­re­la­tion
with throm­bo­em­bolic events in severe COVID-19 infec­tion. J Am Coll Surg.
2020;231(2):193-203.e1203e1.
21. Delrue M, Siguret V, Neuwirth M, et al. Contrast between prev­a­lence of HIT
antibodies and con­firmed HIT in hos­pi­tal­ized COVID-19 patients: a pro­spec­
tive study with clin­i­cal impli­ca­tions. Thromb Haemost. 2021;121(7):971-975.
22. Daviet F, Guervilly C, Baldesi O, et al. Heparin-induced throm­bo­cy­to­pe­nia
in severe COVID-19. Circulation. 2020;142(19):1875-1877.
23. Nazy I, Jevtic SD, Moore JC, et  al. Platelet-acti­vat­ing immune complexes
iden­ti­fied in crit­i­cally ill COVID-19 patients suspected of hep­a­rin-induced
throm­bo­cy­to­pe­nia. J Thromb Haemost. 2021;19(5):1342-1347.
24. Bowles L, Platton S, Yartey N, et al. Lupus anti­co­ag­u­lant and abnor­mal coag­
u­la­tion tests in patients with COVID-19. N Engl J Med. 2020;383(3):288-290.
25. Siguret V, Voicu S, Neuwirth M, et  al. Are antiphospholipid antibod­
ies asso­ci­ated with throm­botic com­pli­ca­tions in crit­i­cally ill COVID-19
patients? Thromb Res. 2020;195(Novem­ber):74-76.
Prakash Singh Shekhawat
26. Reyes Gil M, Barouqa M, Szymanski J, Gonzalez-Lugo JD, Rahman S, Billett
HH. Assessment of lupus anti­co­ag­u­lant pos­i­tiv­ity in patients with coronavi­
rus dis­ease 2019 (COVID-19). JAMA Netw Open. 2020;3(8):e2017539.
27. Anticoagulation Forum, Inc. Venous throm­bo­em­bo­lism (VTE) pro­phy­
laxis in acutely ill med­i­cal patients. Accessed 19 Sep­tem­ber 2021. https://
acforum-excellenceorg/Resource-Center/resource_files/1702-2021-03-
03-142440pdf.
28. Cuker A, Tseng EK, Nieuwlaat R, et  al. Amer­i­can Society of Hematology
2021 guide­lines on the use of anticoagulation for thromboprophylaxis in
patients with COVID-19. Blood Adv. 2021;5(3):872-888.
29. Zarychanski R. Therapeutic anticoagulation in crit­i­cally ill patients with
COVID-19—pre­lim­i­nary report. medRxiv. Preprint posted online 12 March
2021.
30. Lawler PR, Goligher EC, Berger JS, et  al. Therapeutic anticoagulation
in non-crit­ i­
cally ill patients with COVID-19. N Engl J Med. 2021;385(9):
790-802.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/621/1852155/621thilagar.pdf by guest on 13 December 2021
31. Lopes RD, de Barros E, Silva PGM, Furtado RHM, et al; ACTION Coalition
COVID-19 Brazil IV Investigators. Therapeutic ver­sus pro­phy­lac­tic antico­
agulation for patients admit­ted to hos­pi­tal with COVID-19 and ele­vated
D-dimer con­cen­tra­tion (ACTION): an open-label, multicentre, randomised,
con­trolled trial. Lancet. 2021;397(10291):2253-2263.
32. Sadeghipour P, Talasaz AH, Rashidi F, et al; INSPIRATION Investigators.
Effect of inter­me­di­ate-dose vs stan­dard-dose pro­phy­lac­tic anticoagulation
on throm­botic events, extra­cor­po­real mem­brane oxy­gen­a­tion treat­ment,
or mor­ tal­
ity among patients with COVID-19 admit­ ted to the inten­ sive
care unit: the INSPIRATION ran­dom­ized clin­i­cal trial. JAMA. 2021;325(16):
1620-1630.
33. Spyropoulos AC, Levy JH, Ageno W, et al; Subcommittee on Perioperative,
Critical Care Thrombosis, Haemostasis of the Scientific, Standardization
Committee of the International Society on Thrombosis and Haemosta­
sis. Scientific and Standardization Committee com­mu­ni­ca­tion: clin­i­cal
guid­ance on the diag­no­sis, pre­ven­tion, and treat­ment of venous throm­
bo­em­bo­lism in hos­pi­tal­ized patients with COVID-19. J Thromb Haemost.
2020;18(8):1859-1865.
34. Giannis D, Allen SL, Tsang J, et  al. Postdischarge throm­bo­em­bolic out­
comes and mor­tal­ity of hos­pi­tal­ized patients with COVID-19: the CORE-19
reg­is­try. Blood. 2021;137(20):2838-2847.
35. Cines DB, Bussel JB. SARS-CoV-2 vac­ cine-induced immune throm­ botic
throm­bo­cy­to­pe­nia. N Engl J Med. 2021;384(23):2254-2256.
36. MacNeil JR, Su JR, Broder KR, et al. Updated rec­om­men­da­tions from the
advi­sory com­ mit­
tee on immu­ ni­
za­
tion prac­tices for use of the Janssen
(Johnson and Johnson) COVID-19 vac­cine after reports of throm­bo­sis with
throm­bo­cy­to­pe­nia syn­drome among vac­cine recip­i­ents—United States,
April 2021. MMWR Morb Mortal Wkly Rep. 2021;70(17):651-656.
37. Greinacher A, Thiele T, Warkentin TE, et al. Thrombotic throm­bo­cy­to­pe­nia
after ChAdOx1 nCov-19 vac­ci­na­tion. N Engl J Med. 2021;384(22):2092-2101.
38. See I, Su JR, Lale A, et al. US case reports of cere­bral venous sinus throm­bo­
sis with throm­bo­cy­to­pe­nia after Ad26.COV2.S vac­ci­na­tion, March 2 to April
21, 2021. JAMA. 2021;325(24):2448-2456.
39. Scully M, Singh D, Lown R, et al. Pathologic antibodies to plate­let fac­tor 4
after ChAdOx1 nCoV-19 vac­ci­na­tion. N Engl J Med. 2021;384(23):2202-2211.
40. Nazy I, Sachs UJ, Arnold DM, et al. Recommendations for the clin­i­cal and
lab­o­ra­tory diag­no­sis of VITT against COVID-19: com­mu­ni­ca­tion from the
ISTH SSC sub­com­mit­tee on plate­let immu­nol­ogy. J Thromb Haemost.
2021;19(6):1585-1588.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000298
COVID-19 and throm­bo­sis: searching for evi­dence  |  627
 THE COVID CRASH: LESSONS LEARNED FROM A WORLD ON PAUSE

Passive immune therapies:

another tool against COVID-19

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021

Lise J. Estcourt
NHS Blood and Transplant, John Radcliffe Hospital, Oxford, UK

Passive immune therapy consists of several different therapies, convalescent plasma, hyperimmune globulin, and severe
acute respiratory syndrome coronavirus 2 (SARS-CoV-2) neutralizing monoclonal antibodies. Although these treatments
were not part of any pandemic planning prior to coronavirus disease 2019 (COVID-19), due to the absence of high-quality
evidence demonstrating benefit in other severe respiratory infections, a large amount of research has now been per-
formed to demonstrate their benefit or lack of benefit in different patient groups. This review summarizes the evidence
up to July 2021 on their use and also when they should not be used or when additional data are required. Vaccination
against SARS-CoV-2 is the most important method of preventing severe and fatal COVID-19 in people who have an intact
immune system. Passive immune therapy should only be considered for patients at high risk of severe or fatal COVID-19.
The only therapy that has received full regulatory approval is the casirivimab/imdevimab monoclonal cocktail; all other
treatments are being used under emergency use authorizations. In Japan, it has been licensed to treat patients with mild
to moderate COVID-19, and in the United Kingdom, it has also been licensed to prevent infection.

LEARNING OBJECTIVES
• Summarize current evidence on the use of passive immune therapy to prevent infection—whether it reduces risk
of death or hospitalization
• Summarize current evidence on the use of passive immune therapy for high-risk patients who have mild COVID-19
symptoms—risk of death
• Can state whether passive immune therapy reduces all-cause mortality for hospitalized patients and whether any
subgroups will beneft

(1854-1917) was awarded the frst Nobel Prize in Medicine

CLINICAL CASE 1
A 74-year-old man (case) had a test for severe acute respiratory
syndrome coronavirus 2 (SARS-CoV-2) after a family gathering
for the christening of his granddaughter. He had had the test
because he was a close contact for the index case at the chris-
tening. His 49-year-old nephew (index case) had been admit-
ted to hospital the day after the christening and was found to
be SARS-CoV-2 positive. The nephew had had some symp-
toms at the time of the christening but had not thought that
these were due to coronavirus disease 2019 (COVID-19). The
74-year-old man self-isolated with his wife while awaiting the
results of the test.
Introduction
Passive antibody therapy is one of the oldest treatments
for infectious diseases that is still in use. Emil von Behring
“for his work on serum therapy, especially its applica-
tion against diphtheria.”1 Today, passive antibody therapy
involves treatment with polyclonal antibodies derived from
humans (convalescent plasma [CP] or hyperimmune glob-
ulin), animals (antisera), or antigen-specifc monoclonal
antibodies (mAbs).2 In this review, I focus on CP and neu-
tralizing mAbs and the current evidence for their use. There
are no published trials on the use of hyperimmune globulin
in COVID-19.3
CP after infection, particularly after severe illness, may
contain high levels of polyclonal pathogen-specifc anti-
bodies. These antibodies may confer passive immunity to
recipients and in viral diseases are thought to have their
main action via neutralization of viral particles.4 Collection
and transfusion of CP can occur rapidly after the onset
of a pandemic, with collection of plasma occurring from
14 days after a patient has recovered from the infection.

628 | Hematology 2021 | ASH Education Program

Table 1. Comparison between dif­fer­ent pas­sive anti­body ther­a­pies

Characteristic CP Human hyper­im­mune glob­u­lin mAb

Speed of pro­duc­tion after start of
pan­demic
Can adapt to viral var­i­ants
Number of anti–SARS-CoV-2
antibodies prod­uct con­tains
Amount of anti­body contained
within the prod­uct
Rapid—weeks
Can be pro­duced once patients
have recov­ered from infec­tion
(14 to 28 days after recov­ery)
Yes
Many—poly­clonal
Very var­i­able. Variability can be
reduced by using mini-pools—
Slow—months
Needs time to col­lect plasma from
a large num­ber of peo­ple who
have recov­ered from infec­tion
Yes—more slowly than CP
Many—poly­clonal
Fixed amount of total anti­body
Slowest—months
Need to iden­tify poten­tial
antibodies that would be use­ful
to develop as a mAb and then
man­u­fac­ture anti­body
No
One to 2 antibodies
Fixed amount of neu­tral­iz­ing SARS-
CoV-2 neu­tral­iz­ing anti­body

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021

used in Argentina but not all­
countries allowed to pro­duce
mini-pools
Route of admin­is­tra­tion Intravenous Subcutaneous, intra­mus­cu­lar, or Subcutaneous, intra­mus­cu­lar, or
intra­ve­nous intra­ve­nous
Derived from blood* Yes Yes No
Availability Able to be pro­duced in any Not ­able to be pro­duced in every Limited sup­ply due to com­plex
coun­try in which peo­ple have coun­try but manufactur­ing manufactur­ing require­ments
recov­ered from the infec­tion require­ments are not com­plex
and are a­ ble to pro­duce plasma
Cost Relatively cheap—$100 to $200 More expen­sive than CP but Expensive—$1000s per dose
per dose cheaper than mAb—may be Cannot be afforded by low- and
Can be used in low- and mid­dle- ­able to be pro­duced locally, mid­dle-income countries
income countries in low- and mid­dle-income Can only be pro­duced at a lim­ited
countries num­ber of manufactur­ing sites
*Any prod­uct derived from human blood requires viral test­ing to ensure that the trans­fused prod­uct does not cause a trans­fu­sion-trans­mit­ted infec­
tion. In high-income countries with good screen­ing sys­tems, trans­fu­sion-trans­mit­ted infec­tion is very rare. In the United Kingdom, there were no
trans­fu­sion-trans­mit­ted infec­tions reported to the national hemovigilance sys­tem in 2020 (https:​­/​­/www​­.shotuk​­.org​­/wp​­-content​­/uploads​­/myimages​
­/TTI​­-Supplementary​­-material​­-2020​­.pdf).

The anti­body response in CP donors adapts with the virus, either
due to donors becom­ing infected with new var­i­ants of the virus
or due to vac­ci­na­tion after an ini­tial nat­u­ral infec­tion. How-
ever, the lev­els of neu­tral­iz­ing anti­body can vary sig­nif­i­cantly
from 1 unit to the next, and a min­i­mum thresh­old of anti­body is
required within each unit to ensure that the trans­fu­sion con­tains
a suf­fi­cient level of neu­tral­iz­ing anti­body (Table 1).
Hyperimmune glob­u­lin is cur­rently used to pro­tect vul­ner­
a­ble indi­vid­u­als from other viral infec­tions, includ­ing var­i­cella
zoster.5 It is pro­duced by pooling thou­sands of dona­tions from
peo­ple who have recov­ered from an infec­tion or have been vac­
ci­nated against an infec­tion and have high lev­els of antibodies.
It pro­duces a con­sis­tent prod­uct that always has a defined level
of anti­body within it. However, it takes time to pro­duce, so it
can­not be used as early in a pan­demic. It will also evolve with
changes with the viral var­i­ant but not as rap­idly as CP.
Neutralizing mono­ clo­ nal anti­ body ther­ apy can be derived
from humans who have had an infec­tion, or been vac­ci­nated, or
from human­ized mice that have been exposed to SARS-CoV-2
anti­gens.2 Monoclonal anti­body pro­duc­tion can iden­tify antibod-
ies with a high level of neu­tral­iz­ing activ­ity and can be pro­duced
with­out the need for blood donors. mAbs will, how­ever, not
adapt to viral var­i­ants, and so over time, the virus can become
resis­tant to the mono­clo­nal anti­body. This has already hap­pened
with the SARS-CoV-2 virus (Table 2). Monotherapy with bam-
lanivimab has had its emer­gency use autho­ri­za­tion (EUA) with­
drawn due to devel­op­ment of viral resis­tance.6 This may mean
Prakash Singh Shekhawat
that mono­clo­nal cock­tails are effec­tive and less likely to lead
to resis­tance than use of a sin­gle mono­clo­nal. However, even
a mono­clo­nal cock­tail could become inef­fec­tive in the future,
as shown by the pro­ spec­tive map­ ping of viral var­ i­
ants that
detected a poten­tial muta­tion (E406W) that could escape neu­
tral­i­za­tion by both com­po­nents of the casirivimab/imdevimab
mono­clo­nal cock­tail, as well as both com­po­nents of the bam-
lanivimab/etesevimab cock­tail (Table 2).7 Monoclonal ther­apy is
expen­sive and requires very spe­cial­ized manufactur­ing units. It is
there­fore a treat­ment that most low- and mid­dle-income coun-
tries can­not afford and will find more dif­fi­cult to man­u­fac­ture
locally. CP can be pro­duced in many countries and is much more
afford­able.8 Therefore, when think­ing about whether to use pas­
sive immu­ni­za­tion ther­apy, con­sid­er­ation needs to be made not
just on its effec­tive­ness but also on its acces­si­bil­ity.
Prophylactic ther­apy
The main­stay of pre­vent­ing infec­tion in peo­ple with an intact
immune sys­tem is vac­ci­na­tion. Active immu­ni­za­tion against SARS-
CoV-2 has been shown to be effec­tive at pre­vent­ing severe and
fatal COVID-19, as well as reduc­ing the risk of symp­tom­atic COVID-
19.9 In the United States, by the end of June 2021, ­vac­ci­na­tion had
averted an esti­mated 279 000 deaths and up to 1.25 mil­lion
hos­pi­tal­i­za­tions.10 However, vac­ci­na­tion is not as effec­tive in
patients with an impaired immune sys­tem, either due to immu­
no­sup­pres­sive ther­apy or an under­ly­ing dis­ease that affects the
immune sys­tem. Patients with hema­to­logic ­malig­nan­cies mount
Passive immune ther­a­pies: tool against COVID-19  |  629
Table 2. Different types of mAb in clin­i­cal use

Clinical In clin­i­cal use Competent

Route of trial (out­side of Viral resis­tance author­ity
Name of mono­clo­nal Site of action admin­is­tra­tion results* clin­i­cal tri­als) detected approval
Bamlanivimab Blocks bind­ing of SARS- IV Yes Previously—FDA Yes EUA (now revoked
CoV-2 to the ACE2 EUA revoked Marked—gamma (P.1) in United States)
recep­tor by targeting April 2021 and beta (B.1.351)
the RBD on the spike VoCs
pro­tein of SARS-CoV-2 Modest—delta (B.1.617.2)
VoC
Bamlanivimab plus Blocks bind­ing of IV Yes Yes—use paused Yes EUA†
etesevimab SARS-CoV-2 to the in United Marked—gamma (P.1)
ACE2 recep­tor by States and beta (B.1.351) VoC

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021

targeting dif­fer­ent but Modest—Delta (B.1.617.2)
overlapping epi­topes VoC
on the spike pro­tein
RBD of SARS-CoV-2
Casirivimab plus Blocks bind­ing of SARS- IV or S/C Yes Yes Yes—only to casirivimab Yes—MHRA and
imdevimab CoV-2 to the ACE2 Marked—beta (B.1.351) Jap­a­nese
recep­tor by targeting VoC reg­u­la­tory
dif­fer­ent epi­topes on agency†
the spike pro­tein RBD
of SARS-CoV-2 (do not
over­lap)
Sotrovimab Blocks bind­ing of SARS- IV, trialing IM Yes Yes No EUA†
CoV-2 to the ACE2
recep­tor by targeting
an epi­tope on the RBD
of the spike pro­tein
that is con­served
between SARS-CoV
and SARS-CoV-2
Regdanvimab Blocks bind­ing of SARS- IV Yes Yes Yes EUA‡
CoV-2 to the ACE2 Marked—beta (B.1.351)
recep­tor by targeting VoC
an epi­tope on the RBD
of the spike pro­tein of
SARS-CoV-2
*Results avail­­able in Table 2 for all­-cause mor­tal­ity and need for hos­pi­tal­i­za­tion (by day 30) in Table 3 for out­pa­tients.
†EUA in the United States as well as other countries. In the United States, the indi­ca­tion is for mild or mod­er­ate COVID-19 not requir­ing oxy­gen ther­apy.
‡EUA in South Korea and Indonesia, but indi­ca­tions for use dif­fer.
FDA, Food and Drug Administration (United States); IM, intra­mus­cu­lar; IV, intra­ve­nous; MHRA, Medicines and Healthcare prod­ucts Regulatory Agency
(United Kingdom); RBD, recep­tor-bind­ing domain; S/C, sub­cu­ta­ne­ous; VoC, var­i­ant of con­cern.

blunted anti­ body responses to SARS-CoV-2 vac­ ci­

na­tion, and
those most at risk appear to be patients who are actively treated
with Bruton tyro­sine kinase inhib­i­tors, ruxolitinib, venetoclax, or
anti-CD20 anti­body ther­a­pies.11 Consideration there­ fore needs
to be given to pre­vent­ing infec­tion in peo­ple who are unvac­ci­
nated or par­tially vac­ci­nated and are at high risk of devel­op­ing
severe and fatal COVID-19 due to comorbidities or those who
have been fully vac­ci­nated but are unable to mount an effec­
tive immune response to vac­ci­na­tion. There have been 2 com­
pleted tri­als assessing mono­clo­nal ther­apy12,13 (Table 3), with 1
trial assessing the effect of 1.2 g of the casirivimab/imdevimab
cock­tail (sub­cu­ta­ne­ously via 4 injec­tions) given to house­hold
con­tacts of a pos­i­tive case.13 However, most of these indi­vid­
u­als were not at high risk of severe dis­ease, with fewer than
10% older than 65 years and only 1% with immune sup­pres­sion.
The other trial used bamlanivimab, which showed a ben­e­fit in
nurs­ing home res­i­dents who received the inter­ven­tion in pro­
gres­sion to mod­er­ate or severe dis­ease but no evi­dence of a
dif­fer­ence in mor­tal­ity (Figure 1). However, bamlanivimab is not
effec­tive against many of the viral var­i­ants cur­rently in cir­cu­la­
tion. Ongoing stud­ies of CP, hyper­im­mune glob­u­lin, and mAbs
assessing high-risk pro­phy­laxis will be assessed within liv­ing sys­
tem­atic reviews, so addi­tional infor­ma­tion may be avail­­able over
the next few months.23,24 It is espe­cially impor­tant to know their
addi­tional impact over and above ­vac­ci­na­tion for those indi­vid­
u­als who have a poor response to vac­ci­na­tion, for exam­ple, the
immunosuppressed pop­u­la­tion.11
CLINICAL CASE (Con­tin­ued)
The patient and his wife received the results of the poly­mer­
ase chain reac­tion test the next day, and both had tested pos­i­
tive for SARS-CoV-2. In total, 15 of the 45 peo­ple who attended
the chris­ten­ing tested pos­i­tive for SARS-CoV-2 within the days
fol­
low­ing the chris­ ten­
ing. He started to develop symp­ toms

630  |  Hematology 2021  |  ASH Education Program

 Table 3. Completed ran­dom­ized con­trolled tri­als of anti-SARS-CoV-2 mAb ther­apy

Number Need for hos­pi­

ran­dom­ized Viral var­i­ants tal­i­za­tion at 30
up to data Number Patient con­sid­ered in Mortality at 30 days or death
Study Country Intervention(s) cut­off ana­lyzed pop­u­la­tion Primary out­come Type of anal­y­sis ana­ly­ses* days (GRADE)† (GRADE)†
Prophylaxis
O’Brien et al‡13 Moldova, S/C casirivimab/ 753 1505 Age ≥12 years Incidence of Interim No NR NR
NCT04452318 Romania, imdevimab 1.2 g Household con­tact symp­tom­atic Planned Recruited
United (0.6 g + 0.6 g) of con­firmed COVID-19 recruit­ment ??-Jan­u­ary 2021
States SARS-CoV-2 case 3750
Placebo 752
SARS-CoV-2 PCR
and anti­body
neg­a­tive

BLAZE-212 United States Bamlanivimab 588 966 Residents and Incidence of Final No RR, 0.83 NR
NCT04497987 (4.2 g) staff at 74 skilled COVID-19— Recruited 95% CI,
nurs­ing viro­log­i­cal or August- 0.25-2.70
Placebo 587
and assisted liv­ing clin­i­cal Novem­ber (low)
facil­i­ties 2020
Within 7 days
of a reported
con­firmed SARS-
CoV-2 case
Outpatients (asymp­tom­atic)
O’Brien et al14 United States S/C casirivimab/ 155 311 Age ≥12 years Incidence of Interim July 2020 to NR RR, 0.14
NCT04452318‡ imdevimab 1.2 g Confirmed SARS- symp­tom­atic Planned Jan­u­ary 2021 95% CI, 0.01-2.76

Prakash Singh Shekhawat

(0.6 g + 0.6 g) CoV-2 COVID-19 recruit­ment (low)
SARS-CoV-2 3750
Placebo 156
anti­body
neg­a­tive
Asymptomatic
Outpatients (mild dis­ease)
Weinreich et al Chile, Mexico, Casirivimab/ 838 4057 Adult Clinical—COVID- Interim No RR, 1.02 RR, 0.30
(phase 3)15 Romania, imdevimab 1.2 g Confirmed SARS- related Planned Recruited 95% CI, 0.06- 95% CI, 0.13-0.68
NCT04425629‡ United (0.6 g + 0.6 g) CoV-2 hos­pi­tal­i­za­tion recruit­ment Sep­tem­ber 16.22 (low)
States ≤7 days from onset or death 6420 2020 to (low)
symp­toms (day 29) Subgroup— Jan­u­ary 2021
Casirivimab/ 1529 RR, 0.33 RR, 0.29
≥1 risk fac­tor anti­body
imdevimab 2.4 g 95% CI, 0.01- 95% CI, 0.17-0.48
COVID-19 neg­a­tive
(1.2 g + 1.2 g) 3.94 (mod­er­ate)
Mild symp­toms
(low)
(WHO 2-3)
Placebo 1500 Excluded — —
Hospitalized (any
rea­son)
Vaccinated or plan
to be vac­ci­nated

Passive immune ther­a­pies: tool against COVID-19  |  631

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021
 Table 3. (continued)

Number Need for hos­pi­

ran­dom­ized Viral var­i­ants tal­i­za­tion at 30
up to data Number Patient con­sid­ered in Mortality at 30 days or death
Study Country Intervention(s) cut­off ana­lyzed pop­u­la­tion Primary out­come Type of anal­y­sis ana­ly­ses* days (GRADE)† (GRADE)†
COMET-ICE‡16 Brazil, Sotrovimab (0.5 g) 291 583 Adult Clinical—dis­ease Interim No RR, 0.33 RR, 0.14
NCT04545060 Canada, Confirmed SARS- pro­gres­sion Stopped Recruited 95% CI, 0.01- 95% CI,
Placebo 292
Spain, CoV-2 recruit­ment August 2020 8.18 0.04-0.48
United ≤5 days from onset due to to March (low) (low)
States symp­toms effi­cacy 1057 2021
>55 years or par­tic­i­pants
comorbidities
Mild symp­toms
(WHO 2-3)
Excluded
Likely to require
hos­pi­tal­i­za­tion
≤24 h
Likely to die ≤7
days

632  |  Hematology 2021  |  ASH Education Program

Severely immu­no­
com­pro­mised
Vaccinated or plan
to be vac­ci­nated
within 4 weeks
BLAZE-117 Puerto Rico, Bamlanivimab 104 577 Adult Virological—viral Interim No No events RR, 0.17
NCT04427501 United (0.7 g) Confirmed SARS- clear­ance Planned Recruited June- (low) 95% CI, 0.02-1.33
States CoV-2 recruit­ment Sep­tem­ber (low)
Mild symp­toms 3160 2020
Bamlanivimab 109 RR, 0.32
(WHO 2-3)
(2.8 g) 95% CI, 0.07-1.47
Excluded
(low)
Requires oxy­gen
Bamlanivimab 104 ther­apy RR, 0.34
(7.0 g) Any seri­ous 95% CI,
con­com­i­tant 0.08-1.56
sys­temic dis­ease (low)
Pregnant or
Bamlanivimab/ 114 breastfeeding See updated See updated
etesevimab results results below
(2.8 g + 2.8 g) below
Placebo 161 — —
BLAZE-118 United States Bamlanivimab/ 518 1035 Adult Clinical—COVID- Interim No RR, 0.05 RR, 0.30
NCT04427501 etesevimab Confirmed SARS- 19–related Planned Sep­tem­ber- 95% CI, 95% CI, 0.16-0.59
(2.8 g + 2.8 g) CoV-2 hos­pi­tal­i­za­tion recruit­ment Decem­ber 0.00-0.81 (low)
Mild symp­toms (acute care 3160 2020 (low)
Placebo 517
(WHO 2 to 3) for ≥24 hours)
Excluded or death from
Requires oxy­gen any cause by
ther­apy day 29
Any seri­ous
con­com­i­tant
sys­temic dis­ease
Pregnant or
breastfeeding

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021

 Table 3. (continued)

Number Need for hos­pi­

ran­dom­ized Viral var­i­ants tal­i­za­tion at 30
up to data Number Patient con­sid­ered in Mortality at 30 days or death
Study Country Intervention(s) cut­off ana­lyzed pop­u­la­tion Primary out­come Type of anal­y­sis ana­ly­ses* days (GRADE)† (GRADE)†
Eom et al‡19 Romania, Regdanvimab 101 327 Adult Clinical—time Interim No No events RR, 0.45
NCT04602000 South 0.04 g/kg Confirmed SARS- to clin­i­cal Planned Recruited (low) 95% CI, 0.14-1.42
Korea, CoV-2 recov­ery recruit­ment Octo­ber- (low)
Spain, ≤7 days from onset Virological—viral 1172 Decem­ber
Regdanvimab 103 RR, 0.56
United symp­toms clear­ance 2020
0.08 g/kg 95% CI, 0.19-1.60
States Excluded
(low)
Hospitalized
Placebo 103 —
Weinreich et al Chile, Mexico, Casirivimab/ 92 275 Adult Virological—viral Interim No NR RR, 0.43
(phase 1/2)20 Romania, imdevimab 2.4 g Confirmed SARS- clear­ance Recruited 95% CI, 0.08-2.19
NCT04425629 United (1.2 g + 1.2 g) CoV-2 June- August (low)
States ≤7 days from onset 2020
Casirivimab/ 90 RR, 0.21
symp­toms
imdevimab 8 g 95% CI, 0.02-1.79
Excluded
(4 g + 4 g) (low)
Hospitalized
Placebo 93 —
Inpatients (mod­er­ate or severe dis­ease)
RECOVERY‡21 United Casirivimab/ 4839 9185 Any age Clinical— Complete No RR, 0.94 NA
NCT04381936 Kingdom imdevimab 8 g Suspected or con­ all-cause Subgroup— Recruited 95% CI, 0.87-
(4 g + 4 g) firmed COVID-19 mor­tal­ity anti­body Sep­tem­ber 1.02
Median 9 days day 28 neg­a­tive 2020 to May (mod­er­ate)
Standard care 4946

Prakash Singh Shekhawat

symp­tom onset 2021
>90% requir­ing
oxy­gen ther­apy
(WHO score ≥5)
ACTIV-322 Denmark, Bamlanivimab (7 g) 314 Inpatients, Adult Clinical—time Interim anal­y­sis, No RR, 1.39 NA
NCT04501978 India, mod­er­ate Confirmed SARS- to sustained bamlanivimab Recruited 95% CI, 0.07-
Poland, dis­ease CoV-2 recov­ery arm stopped August to 1.47
Singapore, ≤12 days from onset for futil­ity, Octo­ber (low)
Spain, symp­toms recruit­ment 2020
Switzerland, Excluded ongo­ing for
United Requiring organ other arms
Kingdom, sup­port
United Pregnant/breast
States feed­ing
*Was the type of virus (eg, alpha, beta, gamma, delta) detected at base­line taken into con­sid­er­ation as a sub­group anal­y­sis?
†GRADE assess­ment—assesses cer­tainty of the evi­dence. High cer­tainty: very con­fi­dent that the true effect lies close to that of the esti­mate of the effect. Moderate cer­tainty: mod­er­ately
con­fi­dent in the effect esti­mate: The true effect is likely to be close to the esti­mate of the effect, but there is a pos­si­bil­ity that it is sub­stan­tially dif­fer­ent. Low cer­tainty: con­fi­dence in the effect
esti­mate is lim­ited: the true effect may be sub­stan­tially dif­fer­ent from the esti­mate of the effect. Very low cer­tainty: very lit­tle con­fi­dence in the effect esti­mate: The true effect is likely to be
sub­stan­tially dif­fer­ent from the esti­mate of effect.
‡Pre–peer review.
NA, not applicable; NR, not reported; PCR, poly­mer­ase chain reac­tion; WHO, World Health Organization.

Passive immune ther­a­pies: tool against COVID-19  |  633

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021

Figure 1. SARS-CoV-2 mAb vs placebo or standard care—all-cause mortality at 28 days or hospital discharge.
of fever and a dry cough the day after his pos­it­ive test, but
his symp­toms were mild, and he treated the fever with parac-
etamol and rested at home.
Treatment with asymp­tom­atic infec­tion
or mild symp­toms
Vaccination is known to decrease the risk of pro­ gres­
sion to
mod­er­ate or severe COVID-19 dis­ease, although it has less of an
impact on the devel­op­ment of asymp­tom­atic or mild symp­toms.
All of the cur­rent evi­dence for pas­sive immune ther­apy is prior to
vac­ci­na­tion, or patients who were vac­ci­nated or about to be vac­
ci­nated were excluded from the tri­als. Data from Public Health
England have shown that vac­ci­na­tion decreases the risk of hos­
pi­tal­i­za­tion by over 90%, even with the delta var­i­ant. It is there­
fore more dif­fi­cult to know the real addi­tional effect of pas­sive
immune ther­apy over and above vac­ci­na­tion (Tables 3 and 4).
In the stud­ies that assessed bamlanivimab alone or regdan-
vimab, there were no deaths in any of the study arms; there­
fore, any effect on all­-cause mor­tal­ity can­not be assessed.42 The
only study that showed a reduc­tion in mor­tal­ity was the BLAZE-1
trial17 arm that used the bamlanivimab/etesevimab mono­ clo­
nal cock­tail (Figure 1); the casirivimab/imdevimab cock­tail trial
showed a trend in the direc­tion of effect, but it was not clin­i­cally
sig­nif­i­cant,15,20 nor was the effect of high-dose CP (Figure 2).3,8
Therefore, although it is sug­ges­tive that pas­sive immune ther­
apy if given early could save lives and the reduce risk of severe
dis­ease, addi­tional data are required before it can be used in
rou­tine prac­tice. None of the stud­ies have performed a cost-
effec­tive­ness anal­y­sis, but it is likely that those patients who do
not respond to vac­ci­na­tion and are at high risk of severe or fatal
COVID-19 are the group that will dem­on­strate a ben­e­fit.
CLINICAL CASE (Con­t in­u ed)
Seven days after the chris­ ten­
ing, he devel­ oped increas­ ing
short­ness of breath and called an ambu­lance. On exam­i­na­tion
by the para­med­ics, he was pyrex­ial (tem­per­a­ture 38.9°C), had a
respi­ra­tory rate of 25 breaths per min­ute, and was hyp­oxic with
an oxy­gen sat­u­ra­tion of 91% on room air. He was given oxy­gen
634  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021
by the para­med­ics and taken to the local emer­gency room. In
the emer­gency room, he was given ste­roids and admit­ted to
the hos­pi­tal for oxy­gen ther­apy.
Treatment with mod­er­ate or severe symp­toms
Two tri­als have assessed mono­clo­nal anti­body use in mod­er­ately
or severely unwell patients; nei­ther showed a ben­e­fit for patients
over­all (Figure 1), but there was a ben­e­fit for patients who had not
yet devel­oped a detect­able anti­body response (Figure 3). Those
patients who did not develop an anti­body response had a much
higher mor­tal­ity rate than those who did not.19,21 This is partly
because patients who have a delayed anti­body response are
older and have more comorbidities. This does mean that reduc­
ing mor­tal­ity in this group will mean a larger abso­lute reduc­tion
in mor­tal­ity; for exam­ple, based on the RECOVERY data, 54 lives
per 1000 patients treated would be saved (95% CI, 24-80 lives
saved per 1000 patients treated). Some SARS-CoV-2 var­i­ants are
resis­ tant to bamlanivimab monotherapy, whereas the cur­ rent
major var­i­ants in cir­cu­la­tion are still sen­si­tive to neu­tral­i­za­tion by
the casirivimab/imdevimab cock­tail. However, this may change
with the devel­op­ment of new var­i­ants.7 As var­i­ant screen­ing can­
not be done in real time, pas­sive anti­body treat­ments need to
be used that are effec­tive against all­cur­rent var­i­ants in a par­tic­
u­lar region of the world. CP use does not show a ben­e­fit over­all
for patients with mod­er­ate or severe COVID-19.29,30,39 CP has also
been assessed in anti­body-neg­at­ive patients in 2 major tri­als,
RECOVERY and REMAP-CAP.29,39 This shows a sim­i­lar trend in the
direc­tion of a pos­si­ble effect in CP, but it does not reach sta­
tis­ti­cal sig­nif­i­cance (risk ratio [RR], 0.93; 95% CI, 0.86-1.01). This
may partly be because CP is a much more var­i­able prod­uct, with
some units hav­ing much lower anti­body lev­els than oth­ers, so
even if a min­i­mum titer is used within the tri­als, this may not have
been suf­fi­cient. Several tri­als have shown an effect in a sub­group
of par­tic­i­pants who received a higher titer prod­uct.8,32
CLINICAL CASE (Con­tin­ued)
Over the next 2 days, the 74-year-old man con­tin­ued to dete­ri­o­
rate and was admit­ted to inten­sive care for non­in­va­sive ven­ti­la­tory
 Table 4. Completed ran­dom­ized con­trolled tri­als of CP and hyper­im­mune glob­u­lin ther­apy

Viral var­i­ants
Randomized Number Neutralizing anti­body con­sid­ered in
Study Country Intervention(s) per arm ana­lyzed Patient pop­u­la­tion Primary out­come titer ana­ly­ses*
Prophylaxis
No com­pleted stud­ies
Outpatients (asymp­tom­atic)
No com­pleted stud­ies
Outpatients (mild dis­ease)
Libster et al8 Argentina 250 mL CP on 80 160 Age >74 or 65 to 74 and Development of severe Anti–S IgG SARS-CoV-2 No
NCT04479163 day 1 comorbidity dis­ease—defined as (COVIDAR IgG) Recruited
Confirmed SARS-CoV-2 RR ≥30 breaths/min min­i­mum titer 1:1000 June-Octo­ber 2020
Saline (Placebo) 80
Mild ill­ness, not requir­ing or oxy­gen
hos­pi­tal­i­za­tion sat­u­ra­tions <93%
≤48 h from symp­tom onset on air
Inpatients (mod­er­ate dis­ease (WHO 4 or 5)
Agarwal et al25 India Two doses 235 464 Adult Composite all­- NAb not used to select No
CTRI/2020/04/024775 200 mL CP, Confirmed SARS-CoV-2 cause mor­tal­ity plasma, tested at Recruited
24 h apart, SpO2 ≤93% and RR >24/min or pro­gres­sion to end of study—63% April-July 2020
pref­er­a­bly or PaO2/FiO2 200-300 severe dis­ease of donors had NAb
dif­fer­ent Excluded (PaO2/FiO2 <100) titer >1:20 with
donors Critically ill (PaO2/FiO2 within day 28 median titer 1:40
<200 or shock requir­ing
Standard care 229
vaso­pres­sors)

Prakash Singh Shekhawat

Simonovich et al26 Argentina 10-15 mL/kg 228 333 Adult Clinical sta­tus at day Anti–S IgG SARS-CoV-2 No
NCT04383535 Mini-pools (5-10 Confirmed SARS-CoV-2, 30 ordi­nal categories (COVIDAR IgG) Recruited
donors) requir­ing hos­pi­tal­i­za­tion, 1: death Median titer of 1:3200 May-August 2020
pneu­mo­nia, plus SpO2 2: inva­sive ven­ti­la­tory (IQR, 1:800 to
Saline (pla­cebo) 105
<93% or PaO2/FiO2 <300 sup­port 1:3200)
Excluded 3: hos­pi­tal­ized with
MV or NIV sup­ple­men­tal
oxy­gen require­ments
4: hos­pi­tal­ized with­out
sup­ple­men­tal
oxy­gen require­ments
5: discharged with­out
full return of base­line
phys­i­cal func­tion
6: discharged with full
return of base­line
phys­i­cal func­tion

Passive immune ther­a­pies: tool against COVID-19  |  635

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021
 Table 4. (continued)

Viral var­i­ants
Randomized Number Neutralizing anti­body con­sid­ered in
Study Country Intervention(s) per arm ana­lyzed Patient pop­u­la­tion Primary out­come titer ana­ly­ses*
27
Avendaño-Solà et al† Spain 250-300 mL CCP 38 81 Adult Proportion of patients NAb not avail­­able No
NCT04345523 on day 1 Confirmed SARS-CoV-2 in cat­e­gory 5, 6, to select plasma, Recruited
Radiological changes or or 7 of 7-cat­e­gory all­ dona­tion on April-July 2020
Standard care 43
clin­i­cal fea­tures plus SpO2 COVID-19 ordi­nal sub­se­quent test­ing
<94%, <12-day symp­tom scale at day 15 had VMNT-ID50 > 1:80
onset
Excluded
MV, high-flow O2
AlQahtani et al28 Bahrain Two doses 20 40 Age ≥21 Requirement for NR No
NCT04356534 200 mL CCP, Confirmed SARS-CoV-2 ven­ti­la­tion Recruited
24 h apart Pneumonia, plus SpO2 <92% (NIV or MV) April-June 2020
or PaO2/FiO2 <300
Standard care 20
Excluded
MV or MOF
Inpatients (mod­er­ate or severe dis­ease—WHO 4 to 7)

636  |  Hematology 2021  |  ASH Education Program

RECOVERY 202129 United Two doses 5795 11 558 Any age Clinical—all­-cause Minimum Euroimmun 6 Partially
NCT04381936 Kingdom 275 mL CCP, Suspected or con­firmed mor­tal­ity day 28 Used sur­ro­gate
24 h apart COVID-19 of ran­dom­ized
Median 9 days symp­tom before/after
Standard care 5763
onset Decem­ber 2020
>90% requir­ing oxy­gen for WT/alpha
ther­apy (WHO score ≥5) var­i­ant
Recruited
May 2020 to
Jan­u­ary 2021
CONCOR-1†30 Brazil, Canada, 500 mL CCP 614 921 Adult Intubation or death Viral neu­tral­iz­ing No
NCT04348656 United Confirmed SARS-CoV-2 by day 30 antibodies titer Recruited
Standard care 307
States Requiring oxy­gen ther­apy >1:160 or antibodies May 2020 to
(WHO score ≥5) against the RBD Jan­u­ary 2021
Excluded of the SARS-CoV-2
Symptoms >12 days spike pro­tein titer
MV >1:100
O’Donnell et al31 Brazil, United 200-250 mL CCP 150 223 Adult Clinical sta­tus at Anti–SARS-CoV-2 No
NCT04359810 States Confirmed SARS-CoV-2 day 28 fol­low­ing antispike IgG Recruited
200-250 mL 73
Requiring oxy­gen ther­apy ran­dom­i­za­tion anti­body titer ≥1:400 April-Novem­ber
non­im­mune
(WHO score ≥5) (7-point ordi­nal scale 2020
plasma
based on WHO)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021

 Table 4. (continued)

Viral var­i­ants
Randomized Number Neutralizing anti­body con­sid­ered in
Study Country Intervention(s) per arm ana­lyzed Patient pop­u­la­tion Primary out­come titer ana­ly­ses*
CAPSID†32 Germany CCP (250- 53 105 Age 18-75 years Composite end point Median PRNT50 No
NCT04433910 325 mL) on Confirmed SARS-CoV-2 of sur­vival and no neu­tral­i­za­tion titer Recruited
days 1, 3, and 5 RR >30 or requir­ing oxy­gen lon­ger fulfilling 1:160 (IQR, 1:80 to August-Decem­ber
ther­apy (WHO score ≥5) cri­te­ria of severe 1:320) 2020
Standard care, 52
COVID-19 (day 21)
cross­over to
CCP at day 14
if not improved
Gharbharan et al33 The 300 mL CCP on 43 86 Adult Mortality until Minimum of PRNT50 No
NCT04342182 Netherlands day 1 Confirmed SARS-CoV-2 dis­charge or titer of ≥1:80 Recruited
within 96 h max­i­mum of 60 days April-June 2020
Standard care 43
Excluded
Patients on MV >96 h
Ray et al†34 India Two doses 40 80 Adult All-cause mor­tal­ity at Euroimmun ≥1.5
CTRI/2020/05/025209 200 mL CCP, Confirmed SARS-CoV-2 30 days
24 h apart RR >30, SpO2 <90%,
PaO2/FiO2 <300
Standard care 40
Excluded
preg­nant, MV
Bennett-Guerrero et al35 United States 480 mL (2 units) 59 74 Adult Number of days patient Ideally >1:320, but No
NCT04344535 CCP Confirmed SARS-CoV-2 remained ven­ti­la­tor meet­ing min­i­mum Recruited
Excluded free (up to 28 days) titer per FDA April-August 2020
480 mL (2 units) 15
Pregnant/breastfeeding
non­im­mune

Prakash Singh Shekhawat

plasma
Pouladzadeh et al36 Iran 500 mL CCP 30 60 Confirmed SARS-CoV-2 Improvement in the NR No
IRCT20200310046736N1 WHO score >4 lev­els of cyto­kine Recruited
Standard care 30
Excluded storm indi­ces March-May 2020
Pregnant/breastfeeding
Comorbidities (eg, heart,
liver, kid­ney dis­ease)
Smokers
Hamdy Salman and Ail Egypt 250 mL CP on 15 30 Adult At least 50% NAb not used to select No
Mohamed37 day 1 Confirmed SARS-CoV-2 improve­ment of the plasma Recruited
NCT04530370 2 or more of RR ≥24, SpO2 sever­ity of ill­ness at June-August 2020
Saline (pla­cebo) 15
≤93%, PaO2/FiO2 <300, any time dur­ing 5-
pul­mo­nary infil­trates day study period
Excluded
MOF, sep­tic shock
Bajpai et al†38 India Two doses 14 29 Age 18-65 years Proportion of patients Variable No
NCT04346446 250 mL CCP, Confirmed SARS-CoV-2 remaining free Recruited
24 h apart Pneumonia, plus SpO2 <93% of mechan­i­cal April-May 2020
or PaO2/FiO2 <300 ven­ti­la­tion day 7
Nonimmune 15
Excluded
plasma
Comorbidities (kid­ney,
heart or liver dis­ease,
COPD)

Passive immune ther­a­pies: tool against COVID-19  |  637

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021
 Table 4. (continued)

Viral var­i­ants
Randomized Number Neutralizing anti­body con­sid­ered in
Study Country Intervention(s) per arm ana­lyzed Patient pop­u­la­tion Primary out­come titer ana­ly­ses*
Inpatients (severe dis­ease—WHO 6 to 7)
REMAP-CAP†39 Australia, Two doses 1084 2000 Adult Organ sup­port–free Minimum Euroimmun 6 No
NCT02735707 Canada, 275 mL CCP, Confirmed SARS-CoV-2 days at day 21 Recruited
United 24 h apart ≤48 h since admis­sion to ICU May 2020 to
Kingdom, Jan­u­ary 2021
Standard care 916
United
States
Gonzalez et al†40 Mexico Two doses 130 90 Adult (16-90 years) Duration of NAb not used to select No
NCT04381858 200 mL CCP, Suspected or con­firmed hos­pi­tal­i­za­tion plasma Recruited
24 h apart COVID-19 All-cause mor­tal­ity May-Octo­ber 2020
Severe respi­ra­tory fail­ure day 28
IVIg 0.3 g/kg 60
daily for 5 days
Li et al41 China 4-13 mL/kg of CP 52 103 Confirmed SARS-CoV-2 Time to clin­i­cal Minimum of S-RBD– No
ChiCTR2000029757 Excluded improve­ment spe­cific IgG of 1:640 Recruited

638  |  Hematology 2021  |  ASH Education Program

Standard care 51
high-titer S-RBD–spe­cific (patient dis­charge or (approx­i­ma­tely Feb­ru­ary-April 2020
IgG (≥1:640) reduc­tion 2 points equiv­a­lent to NAb
on 6-point dis­ease of 1:40)
sever­ity scale)
*Was the type of virus (e.g., alpha, beta, gamma, delta, etc.) detected at base­line taken into con­sid­er­ation as a sub­group anal­y­sis?
†Pre–peer review.
CCP, COVID-19 con­va­les­cent plasma; COPD, chronic obstruc­tive pul­mo­nary dis­ease; FDA, Food and Drug Administration; FiO2, fraction of inspired oxygen; ICU, inten­sive care unit; IQR,
interquartile range; IVIg, intra­ve­nous immu­no­glob­u­lin; MOF, multiorgan fail­ure; MV, mechan­i­cal ven­ti­la­tion; NAb, neu­tral­iz­ing anti­body; NIV, non-inva­sive ven­ti­la­tion; PaO2, partial pressure of
oxygen; PRNT50, 50% reduction in plaque count using the plaque reduction neutralization test; RR, respiratory rate; SpO2, oxygen saturation; VMNT, virus microneutralization test; WT, wild type.

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021

 Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021
Figure 2. CP vs placebo or standard care—all-cause mortality at 28 days or hospital discharge.

Figure 3. SARS-CoV-2 mAb vs placebo or standard care—all-cause mortality at 28 days or hospital discharge by antibody status
at baseline.

Prakash Singh Shekhawat

Passive immune ther­a­pies: tool against COVID-19  |  639
sup­port. An emer­gency use access request for CP was made, and
he received CP on day 10 after the chris­ten­ing.
Treatment of crit­i­cally unwell par­tic­i­pants
Fewer tri­ als have spe­ cif­i­
cally assessed inter­ ven­
tions for the
crit­i­cally unwell patients (requir­ing respi­ra­tory or car­dio­vas­cu­
lar organ sup­port) with an inten­sive care level of care (Tables 1
and 2). Several tri­ als exclude patients requir­ ing mechan­ i­
cal
ven­ti­la­tion or organ sup­port of any type. One of the tri­als that
has focused on this patient group is the REMAP-CAP trial.39 It
showed no ben­e­fit of CP over­all (Figure 2), but a prespecified
sub­group showed poten­tial ben­e­fit of CP. This trial, based on
Bayes­ian sta­tis­tics, showed an 89.9% pos­te­rior prob­a­bil­ity of
ben­e­fit in this sub­group. This was a broad group of immunosup-
pressed patients based on the Acute Physiology and Chronic
Health Evaluation (APACHE) score def­i­ni­tion of immu­no­sup­pres­
sion.39 However, as it was a small sub­group, addi­tional research
is required to con­firm whether or not this is a true find­ing. The
RECOVERY trial did include patients receiv­ing non­in­va­sive and
inva­sive ven­ti­la­tion, and no evi­dence of a dif­fer­ence was seen
for those patients receiv­ing inva­sive (RR, 0.71; 95% CI, 0.35-1.47;
70 par­tic­i­pants) or non­in­va­sive ven­ti­la­tion (RR, 0.86; 95% CI,
0.68-1.08; 673 par­tic­i­pants), but the CIs are wide.
CLINICAL CASE (Con­t in­u ed)
Unfortunately, the 74-year-old man con­tin­ued to dete­ri­o­rate and
sub­se­quently required inva­sive ven­ti­la­tion. Fourteen days after
the chris­ten­ing, he died due to COVID-19. In total, 5 of the guests
at the chris­ten­ing were admit­ted to hos­pi­tal, includ­ing all 4 of the
grand­par­ents of the baby and the uncle (the index case), who had
dia­be­tes and hyper­ten­sion. Both grand­fa­thers died of COVID-19.
Key points (see Visual abstract)
Up to now (July 2021), most of the evi­dence is based on tri­als
performed prior to the intro­duc­tion of vac­ci­na­tion. Vaccination
is likely to be the most cost-effec­tive strat­egy for pre­vent­ing
infec­tion in the gen­eral pop­u­la­tion, includ­ing healthy indi­vid­u­
als with high-risk expo­sure (eg, health care work­ers). There are
insuf­fi­cient data on the out­comes of pro­phy­lac­tic pas­sive anti­
body ther­apy in immunosuppressed indi­vid­u­als.
Passive immune ther­apy (mono­clo­nal ther­apy and CP) may
be ben­ e­fi­
cial for high-risk patients who have mild COVID-19
symp­toms, but more data are required. Some countries, includ­
ing the United States, are using pas­sive mono­clo­nal ther­apy for
this indi­ca­tion under EUA.
High-dose pas­sive immune ther­apy (casirivimab/imdevimab)
reduces all­-cause mor­tal­ity for hos­pi­tal­ized patients who have
not yet devel­oped a detect­able anti­body response (SARS-CoV-2
IgG anti­body). This mono­clo­nal cock­tail has just been approved
by the Medicines and Healthcare prod­ucts Regulatory Agency
based on this evi­dence, but indi­ca­tions for its use are cur­rently
not avail­­able.
CP may be ben­e­fi­cial for immunosuppressed indi­vid­u­als who
are severely or crit­i­cally unwell, but more data are required.
640  |  Hematology 2021  |  ASH Education Program
The virus con­tin­ues to change, and so although treat­ments
may be effec­tive against cur­rent SARS-CoV-2 viral var­ia ­nts of
con­cern, this may not be true in the future. Passive immune ther­
a­pies will either have to be very broad spec­trum or adapt with
the virus.
Conflict-of-inter­est dis­clo­sure
Lise J. Estcourt: author on Cochrane liv­ing sys­tem­atic reviews
of mono­clo­nal ther­ap
­ ies and CP. Investigator on the RECOVERY
and REMAP-CAP tri­als.
Off-label drug use
Lise J. Estcourt: nothing to declare.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021
Correspondence
Lise J. Estcourt, NHS Blood and Transplant, Level 2, John Radcliffe
Hospital, Oxford, OX3 9BQ , UK; e-mail: lise​­.estcourt@nhsbt​­.nhs​
­.uk.
References
1. The Nobel Prize. Emil von Behring Nobel Lecture, Serum Therapy in Ther-
apeutics and Medical Science. Accessed 1 June 2021. https:­/­/www­.nobel
prize­.org­/prizes­/medicine­/1901­/behring­/lecture­/
2. Taylor PC, Adams AC, Hufford MM, De La Torre I, Winthrop K, Gottlieb RL.
Neutralizing mono­clo­nal antibodies for treat­ment of COVID-19. Nat Rev
Immunol. 2021;21(6):382-393.
3. Piechotta V, Iannizzi C, Chai KL, et  al. Convalescent plasma or hyper­im­
mune immu­no­glob­u­lin for peo­ple with COVID-19: a liv­ing sys­tem­atic
review. Cochrane Database Syst Rev. 2021(5):CD013600.
4. Casadevall A, Pirofski L-A. The con­va­les­cent sera option for containing
COVID-19. J Clin Invest. 2020;130(4):1545-1548.
5. Lachiewicz AM, Srinivas ML. Varicella-zoster virus post-expo­sure man­age­
ment and pro­phy­laxis: a review. Prev Med Rep. 2019;16:101016.
6. Food and Drug Administration. Revocation let­ter bamlanivimab 04162021.
2021. Accessed 1 June 2021. https://www.fda.gov/news-events/press-an
nouncements/coronavirus-covid-19-update-fda-revokes-emergen
cy-use-authorization-monoclonal-antibody-bamlanivimab
7. Starr TN, Greaney AJ, Addetia A, et  al. Prospective map­ ping of viral
muta­ tions that escape antibodies used to treat COVID-19. Science.
2021;371(6531):850-854.
8. Libster R, Pérez Marc G, Wappner D, et al; Fundación INFANT–COVID-19
Group. Early high-titer plasma ther­apy to pre­vent severe Covid-19 in older
adults. N Engl J Med. 2021;384(7):610-618.
9. Haas EJ, Angulo FJ, McLaughlin JM, et al. Impact and effec­tive­ness of mRNA
BNT162b2 vac­ cine against SARS-CoV-2 infec­ tions and COVID-19 cases,
hospitalisations, and deaths fol­low­ing a nation­wide vac­ci­na­tion cam­paign
in Israel: an obser­va­tional study using national sur­veil­lance data. Lancet.
2021;397(10287):1819-1829.
10. Galvani A, Moghadas SM, Schneider EC. Deaths and hos­ pi­tal­i­
za­
tions
averted by rapid U.S. vac­ci­na­tion roll­out. The Commonwealth Fund, Issue
Briefs. Accessed 15 July 2021. https:­/­/www­.commonwealthfund­.org­/publi
cations­/issue­-briefs­/2021­/jul­/deaths­-and­-hospitalizations­-averted­-rapid­
us­-vaccination­-rollout
11. Maneikis K, Šablauskas K, Ringelevičiūtė U, et  al. Immunogenicity of the
BNT162b2 COVID-19 mRNA vac­cine and early clin­i­cal out­comes in patients
with haematological malig­ nan­cies in Lithuania: a national pro­ spec­ tive
cohort study. Lancet Haematol. 2021;8(8):e583-e592.
12. Cohen MS, Nirula A, Mulligan MJ, et al; BLAZE-2 Investigators. Effect of bam-
lanivimab vs pla­cebo on inci­dence of COVID-19 among res­i­dents and staff
of skilled nurs­ing and assisted liv­ing facil­it­ies: a ran­dom­ized clin­i­cal trial.
JAMA. 2021;326(1):46-55.
13. O’Brien MP, Forleo-Neto E, Musser BJ, et al. Subcutaneous REGEN-COV anti­
body com­bi­na­tion for Covid-19 pre­ven­tion. N Engl J Med. 2021;385:1184-
1195.
14. O’Brien MP, Forleo-Neto E, Sarkar N, et al. Subcutaneous REGEN-COV anti­
body com­bi­na­tion in early SARS-CoV-2 infec­tion. medRxiv. 2021.
15. Weinreich DM, Sivapalasingam S, Norton T, et  al. REGEN-COV anti­body
com­bi­na­tion and out­comes in out­pa­tients with Covid-19 [published online
29 Sep­tem­ber 2021]. N Engl J Med.
16. Gupta A, Gonzalez-Rojas Y, Juarez E, et al. Early Covid-19 treat­ment with
SARS-CoV-2 neu­tral­iz­ing anti­body sotrovimab. medRxiv. 2021.
17. Gottlieb RL, Nirula A, Chen P, et al. Effect of bamlanivimab as monotherapy
or in com­bi­na­tion with etesevimab on viral load in patients with mild to
mod­er­ate COVID-19: a ran­dom­ized clin­i­cal trial. JAMA. 2021;325(7):632-644.
18. Dougan M, Nirula A, Azizad M, et al. Bamlanivimab plus etesevimab in mild
or mod­er­ate Covid-19 [published online 14 July 2021]. N Engl J Med.
19. Eom JS, Ison M, Streinu-Cercel A, et al. Efficacy and safety of CT-P59 plus
stan­dard of care: a phase 2/3 ran­dom­ized, dou­ble-blind, pla­cebo-con­
trolled trial in out­pa­tients with mild-to-mod­er­ate SARS-CoV-2 infec­tion
[published online 16 March 2021]. Res Square.
20. Weinreich DM, Sivapalasingam S, Norton T, et al; Trial Investigators. REGN-
COV2, a neu­tral­iz­ing anti­body cock­tail, in out­pa­tients with Covid-19. N
Engl J Med. 2021;384(3):238-251.
21. Horby PW, Mafham M, Peto L, et al. Casirivimab and imdevimab in patients
admit­ted to hos­pi­tal with COVID-19 (RECOVERY): a randomised, con­trolled,
open-label, plat­form trial. medRxiv. 2021.
22. ACTIV-3/TICO LY-CoV555 Study Group. A neu­tral­iz­ing mono­clo­nal anti­body
for hos­pi­tal­ized patients with Covid-19. N Engl J Med. 2021;384(10):905-914.
23. Hirsch C, Valk SJ, Piechotta V, et al. SARS-CoV-2-neutralising mono­clo­nal
antibodies to pre­vent COVID-19 [published online May 20, 2021]. Cochrane
Database Syst Rev.
24. Valk SJ, Piechotta V, Kimber C, et al. Convalescent plasma and hyper­im­
mune immu­no­glob­u­lin to pre­vent infec­tion with SARS-CoV-2. Cochrane
Database Syst Rev. 2021;1:CD013802.
25. Agarwal A, Mukherjee A, Kumar G, Chatterjee P, Bhatnagar T, Malhotra P.
Convalescent plasma in the man­age­ment of mod­er­ate Covid-19 in adults in
India: open label phase II multicentre randomised con­trolled trial (PLACID
Trial). BMJ. 2020:m3939.
26. Simonovich VA, Burgos Pratx LD, Scibona P, et al; PlasmAr Study Group. A
ran­dom­ized trial of con­va­les­cent plasma in Covid-19 severe pneu­mo­nia. N
Engl J Med. 2021;384(7):619-629.
27. Avendaño-Solá C, Ramos-Martínez A, Muñez-Rubio E, et  al. Convalescent
plasma for COVID-19: a mul­ti­cen­ter, ran­dom­ized clin­i­cal trial. medRxiv. 2020.
28. AlQahtani M, Abdulrahman A, Almadani A, et  al. Randomized con­trolled
trial of con­va­les­cent plasma ther­apy against stan­dard ther­apy in patients
with severe COVID-19 dis­ease. Sci Rep. 2021;11(1):9927.
29. RECOVERY Collaborative Group. Convalescent plasma in patients admit­
ted to hos­ pi­
tal with COVID-19 (RECOVERY): a randomised con­ trolled,
open-label, plat­form trial. Lancet. 2021;397(10289):2049-2059.
30. Bégin P, Callum J, Jamula E, et al. Convalescent plasma for hos­pi­tal­ized
patients with COVID-19 and the effect of plasma antibodies: a ran­dom­
Prakash Singh Shekhawat
ized con­trolled, open-label trial [published online 9 Sep­tem­ber 2021].
Nat Med.
31. O’Donnell MR, Grinsztejn B, Cummings MJ, et  al. A ran­dom­ized dou­ble-
blind con­trolled trial of con­va­les­cent plasma in adults with severe COVID-
19. J Clin Invest. 2021;131(13):e150646.
32. Körper S, Weiss M, Zickler D, et  al. High dose con­va­les­cent plasma in
COVID-19: results from the ran­dom­ized trial CAPSID. medRxiv. 2021.
33. Gharbharan A, Jordans CCE, GeurtsvanKessel C, et  al. Effects of potent
neu­tral­iz­ing antibodies from con­va­les­cent plasma in patients hos­pi­tal­ized
for severe SARS-CoV-2 infec­tion. Nat Commun. 2021;12(1):3189.
34. Ray Y, Paul SR, Bandopadhyay P, et al. Clinical and immu­no­log­i­cal ben­e­fits
of con­va­les­cent plasma ther­apy in severe COVID-19: insights from a sin­gle
cen­ter open label randomised con­trol trial. medRxiv. 2020.
35. Bennett-Guerrero E, Romeiser JL, Tal­bot LR, et al. Severe acute respi­ra­tory
syn­drome coronavirus 2 con­va­les­cent plasma ver­sus stan­dard plasma in
coronavirus dis­ease 2019 infected hos­pi­tal­ized patients in New York: a dou­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/628/1852150/628estcourt.pdf by guest on 13 December 2021
ble-blind ran­dom­ized trial. Crit Care Med. 2021;49(7):1015-1025.
36. Pouladzadeh M, Safdarian M, Eshghi P, et al. A ran­dom­ized clin­i­cal trial eval­
u­at­ing the immu­no­mod­u­la­tory effect of con­va­les­cent plasma on COVID-19-
related cyto­kine storm [published online 10 April 2021]. Intern Emerg Med.
37. Hamdy Salman O, Ail Mohamed HS. Efficacy and safety of trans­ fus­ ing
plasma from COVID-19 sur­vi­vors to COVID-19 vic­tims with severe ill­ness:
a dou­ble-blinded con­trolled pre­lim­i­nary study. Egypt J Anaesth. 2020;​
36(1):264-272.
38. Bajpai M, Kumar S, Maheshwari A, et  al. Efficacy of con­va­les­cent plasma
ther­apy com­pared to fresh fro­zen plasma in severely ill COVID-19 patients:
a pilot ran­dom­ized con­trolled trial. medRxiv. 2020.
39. The REMAP-CAP Investigators. Convalescent plasma in crit­i­cally ill patients
with Covid-19. medRxiv. 2021.
40. Gonzalez JLB, González Gámez M, Mendoza Enciso EA, et al. Efficacy and
safety of con­va­les­cent plasma and intra­ve­nous immu­no­glob­u­lin in crit­i­cally
ill COVID-19 patients: a con­trolled clin­i­cal trial. medRxiv. 2021.
41. Li L, Zhang W, Hu Y, et al. Effect of con­va­les­cent plasma ther­apy on time to
clin­i­cal improve­ment in patients with severe and life-threat­en­ing COVID-
19: a ran­dom­ized clin­i­cal trial. JAMA. 2020;324(5):460-470.
42. Kreuzberger N, Hirsch C, Chai KL, et al. SARS-CoV-2-neutralising mono­clo­
nal antibodies for treat­ment of COVID-19. Cochrane Database Syst Rev.
2021;9:CD013825.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000299
Passive immune ther­a­pies: tool against COVID-19  |  641
 UPDATE IN GRAFT- VERSUS - HOST DISEASE

Acute GVHD: think before you treat

Laura F. Newell1 and Shernan G. Holtan2

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/642/1852146/642newell.pdf by guest on 13 December 2021

Knight Cancer Institute, Hematology and Medical Oncology, Oregon Health and Science University, Portland, OR; and 2Division of Hematology,
1

Oncology, and Transplantation, Department of Medicine, University of Minnesota, Minneapolis, MN

The treatment of acute graft-versus-host disease (aGVHD) has become more nuanced in recent years with the develop-
ment of improved risk classification systems and a better understanding of its complex, multisystem pathophysiology.
We review contemporary approaches to the risk stratification and initial treatment of aGVHD, including ongoing clinical
trials. We summarize the findings that led to the first US Food and Drug Administration approval for steroid-refractory
aGVHD (SR-aGVHD), ruxolitinib, as well as some of the challenges clinicians still face in treating SR-aGVHD. Finally, we
discuss the evaluation and management of steroid-dependent aGVHD, which affects approximately one-third of patients
who have long-term, waxing and waning symptoms distinct from chronic GVHD. Future clinical trials for aGVHD treat-
ment may identify steroid-sparing approaches for patients who have a high likelihood of response and approaches to
improve tissue repair and dysbiosis for those unlikely to respond to immunosuppression alone.

LEARNING OBJECTIVES
• Identify risk­stratifed approaches to the initial management of aGVHD
• Identify potential contributing factors for persistent or recurring symptoms after the initial treatment of aGVHD

Introduction rolimus and methotrexate for GVHD prophylaxis. Day 11

Recent advancements have improved overall survival and
decreased the incidence of grades 3 and 4 acute graft­
versus­host disease (aGVHD) after allogeneic hematopoi­
etic cell transplantation (HCT), but substantial challenges
remain.1-3 High­dose corticosteroids are standard therapy
for aGVHD, but this approach does not consider individ­
ual factors and may lead to over­ or undertreatment. In
this review we describe current risk­adapted approaches
to aGVHD management, highlighting opportunities for
improvement.
Sixty years elapsed between the description of sec­
ondary disease as an immunologic complication of murine
bone marrow transplantation (1959) and the frst US Food
and Drug Administration (FDA) approval for SR­aGVHD
treatment, ruxolitinib (2019).4,5 With a deeper understand­
ing of the pathophysiology of aGVHD, including a need for
tissue repair and reversal of dysbiosis, clinical advance­
ments should be tested at a much faster pace.
CLINICAL CASE
A 47­year­old man with acute myeloid leukemia under­
went matched unrelated donor HCT after myeloablative
busulfan and cyclophosphamide conditioning with tac­
methotrexate was 50% dose reduced, and leucovorin
rescue was added for severe mucositis. On day 21, he
developed a patchy maculopapular skin rash involving
a body surface area (BSA) of <20% and was started on
0.1% triamcinolone topical cream. On day 25, he devel­
oped fever, nausea and vomiting, and worsening rash
involving his face, anterior/posterior torso, and lower
extremities to his knees (60% BSA). He had stage 3 skin,
stage 1 upper­gastrointestinal (GI), and overall grade 2
Minnesota standard­risk aGVHD.6,7 Treatment included
2 mg/kg/d prednisone and continued tacrolimus. A com­
plete response (CR) by day 28 of therapy was sustained
at 8 weeks. Although steroids were discontinued by day
52 of therapy, he had hyperglycemia requiring insulin and
became cushingoid during treatment.
Risk-based assessment and initial treatment
Clinical severity-based risk assessment
The grading criteria for aGVHD are well established for the
skin, liver, and intestinal tract, with higher grades associated
with worse transplant outcomes.7 However, discrepancies
have remained in clinical staging across centers, which can
influence multicenter trial outcome reporting.8 In order to
standardize staging, consensus guidelines were developed

642 | Hematology 2021 | ASH Education Program

at the University of Michigan and sub­ se­ quently tested in the
mul­ ti­
cen­
ter Mount Sinai Acute GVHD International Consortium
(MAGIC).9,10 These guide­lines pro­vide more pre­cise def­i­ni­tions for
aGVHD organ stag­ing as well as con­fi­dence lev­els for diag­no­sis
(con­firmed, prob­a­ble, pos­si­ble, neg­a­tive) based on supporting
evi­dence (eg, his­to­logic con­fir­ma­tion, clin­i­cal action). Experts from
the Euro­pean Society for Blood and Marrow Transplantation, the
National Institutes of Health, and the Center for International Blood
and Marrow Transplant Research Task Force have recommended
the MAGIC cri­te­ria for the diag­no­sis and scor­ing of aGVHD.11
Simpler clas­si­fi­ca­tion schemes may fur­ther delin­eate aGVHD
risk in order to test per­son­al­ized treat­ment strat­e­gies in patients
with dif­fer­ing organ stag­ing who are likely to have sim­i­lar out­
comes. The Minnesota aGVHD risk score incor­ po­rates sites
of organ involve­ment and stage of dis­ease to prog­nos­ti­cate a
response to ini­tial aGVHD ther­apy and the risk of transplant-
related mor­tal­ity.6,12 A web-based cal­cu­la­tor is avail­­able online to
deter­mine the Minnesota aGVHD score (https:​­/​­/redcap​­.ahc​­.umn​
­.edu​­/surveys​­/​­?s=bNmFhseJIf). Together, MAGIC stag­ing and the
Minnesota score are used in the risk assess­ment of patients with
newly diag­nosed aGVHD in many cur­rent clin­i­cal tri­als.
Biomarker-based risk assess­ment
In addi­tion to opti­miz­ing clin­i­cal GVHD scor­ing sys­tems, efforts
to improve the deter­mi­na­tion of sever­ity and like­li­hood of re­
sponse have focused on iden­ti­fy­ing periph­eral blood bio­mark­
ers. Criteria and phases of bio­marker devel­op­ment, as well as
the dif­fer­ent types of bio­mark­ers for acute and chronic GVHD,
have recently been reviewed by Adom et  al.13 Because some
avail­­able bio­mark­ers used to diag­nose aGVHD are con­founded
by organ dam­age or infec­tions, they are of lim­ited use in aGVHD
risk assess­ment.
The MAGIC group iden­ti­fied 2 serum bio­mark­ers of GVHD
that in com­bi­na­tion pre­dict severe GVHD and nonrelapse mor­
tal­ity (NRM): a sup­pres­sor of tumor­i­gen­e­sis, a mem­ber of the
inter­leu­kin 1 recep­tor fam­ily and the sol­u­ble recep­tor for inter­
leu­ kin 33, and regenerating islet-derived 3-alpha, pro­ duced
by dam­aged Paneth cells with anti­mi­cro­bial and epi­the­lial-
pro­tec­tive prop­er­ties.14 In a mul­ti­cen­ter val­i­da­tion study, the
MAGIC algo­rithm prob­a­bil­ity (MAP) and clin­i­cal responses were
deter­mined at the time of treat­ment ini­ti­a­tion and again after 4
weeks of ther­apy. Using the ini­tial MAP, patients could be cat­
e­go­rized by Ann Arbor (AA) score,1-3 each with a dis­tinct risk of
NRM; after 4 weeks, changes in the MAP fur­ther refined the esti­
ma­tion of NRM risk in all­AA groups. Importantly, the MAP was
found to more accu­rately pre­dict 6-month NRM than a change
in clin­i­cal symp­toms.
Additional bio­mark­ers of inflam­ma­tion and tis­sue dam­age
have been inves­ ti­
gated. Amphiregulin (AREG), an epi­ der­mal
growth fac­tor (EGF)-like mol­e­cule involved in type 2 immune
responses and tis­sue repair, is another novel bio­marker asso­ci­
ated with GVHD risk. Elevated lev­els of AREG have been iden­ti­
fied in late aGVHD.15 Additionally, AREG lev­els have been shown
to fur­ther refine the Minnesota aGVHD risk score, using sam­ples
col­lected at aGVHD diag­no­sis as part of the Blood and Marrow
Transplant Clinical Trials Network (BMT CTN) 0302 and 0802.16
An AREG thresh­old of ≥33 pg/mL iden­ti­fied patients with a lower
like­li­hood of day-28 response and a higher 2-year NRM. Patho­
logically ele­vated AREG lev­els decline over time in patients who
respond to aGVHD ther­apy.17 Although bio­marker test­ing is avail­­
Prakash Singh Shekhawat
able to cli­ni­cians, novel mark­ers may yet be devel­oped, and the
role of bio­mark­ers remains an area of active research inves­ti­ga­
tion. The assess­ment of bio­marker-based risk strat­i­fi­ca­tion and
mon­i­tor­ing should con­tinue in the con­text of pro­spec­tive clin­i­cal
tri­als.
Risk-adapted ini­tial treat­ment
For patients with grade 2a man­i­fes­ta­tions of aGHVD (defined as
upper-GI symp­toms, stool out­put <1 L/d, rash <50% BSA, with­
out hepatic involve­ment), treat­ment with lower-dose ste­roids
(0.5 mg/kg/d vs 1.0 mg/kg/d) has been shown to be effec­tive
with­out increas­ing the risk of sec­ond­ary immu­no­sup­pres­sion.18
However, for patients with grade 2b or higher man­i­fes­ta­tions
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/642/1852146/642newell.pdf by guest on 13 December 2021
(defined as stool vol­ ume ≥1  L/d, rash ≥50% BSA, or hepatic
involve­ment), treat­ment with lower-dose ste­roids (1.0 mg/kg/d
vs 2.0 mg/kg/d) was asso­ci­ated with an increased like­li­hood of
requir­ing sec­ond­ary immu­no­sup­pres­sive ther­apy.
Recently, the BMT CTN reported (trial 1501) a ran­dom­ized
phase 2 study test­ing the ste­roid-free ini­tial treat­ment of Minne­
sota stan­dard risk aGVHD (N = 127) with sirolimus vs pred­ni­sone.19
In this study, the day-28 response rate was sim­i­lar, at 65% for
sirolimus (90% CI, 54%-76%) vs 73% (90% CI, 64%-82%) for pred­
ni­sone. However, patients on the sirolimus arm had fewer side
effects from ther­apy and an improved patient-reported qual­ity
of life, suggesting that ste­roid-free treat­ment of stan­dard-risk
aGVHD is fea­si­ble. A note of cau­tion is that <10% of the par­tic­
i­pants enrolled in BMT CTN 1501 had lower-GI GVHD, and thus
we still do not know with con­fi­dence how sirolimus ther­apy
would per­form as the upfront ther­apy in this patient pop­u­la­tion.
Additionally, it is unclear how widely these results have been
adopted across clin­i­cal cen­ters.
Patients with a favor­able like­li­hood of response to aGVHD
treat­ment may be can­di­dates for ste­roid-spar­ing approaches; in
con­trast, those patients at highest risk of fatal aGVHD may be
can­di­dates for novel agents. GVHD treat­ment tri­als within the
last 5 years incor­po­rat­ing risk-assess­ment algo­rithms are sum­
ma­rized in Table 1.
CLINICAL CASE
A 61-year-old woman with a his­tory of acute mye­loid leu­ke­mia
in sec­ond com­plete remis­sion under­went fludarabine and mel­
pha­lan con­di­tioned matched unre­lated (male) donor peripheral
blood stem cell transplant. She received GVHD pro­phy­laxis with
tacrolimus and meth­o­trex­ate. Her trans­plant course was com­pli­
cated by neutropenic fever, pro­tein/cal­o­rie mal­nu­tri­tion requir­
ing total par­en­teral nutri­tion (TPN), and grade 1 mucositis. On
day 11 she devel­oped a maculopapular skin rash involv­ing 36%
BSA, and her anti­bi­ot­ics were changed due to con­cern about
a pos­si­ble drug reac­tion. On day 14 she devel­oped increas­ing
diar­rhea with a neg­a­tive infec­tious workup. On day 23, she had
a rash with 30% BSA involve­ ment, nau­
sea, eme­sis, and diar­
rhea (<1500 mL/d). She was diag­nosed with stage 2 skin, stage 1
upper-GI, stage 2 lower-GI, and over­all grade 3 Minnesota high-
risk aGVHD and started on 2 mg/kg/d meth­yl­pred­nis­o­lone. Her
skin rash improved; how­ever, she con­tin­ued to have nau­sea and
watery diar­rhea (1000-1500 mL/d) progressing to ileus. Her hos­pi­
tal course was also com­pli­cated by cytomegalovirus reactivation
Acute GVHD  | 643
 Table 1. First-line risk-adapted treat­ment tri­als for aGVHD

Transplant eli­gi­
Study num­ber GVHD risk cat­e­gory Study type/inter­ven­tion Treatment tar­get Study sta­tus Primary out­come mea­sures
bil­ity cri­te­ria
Standard-risk aGVHD
NCT02806947 Standard-risk (Minnesota Randomized, mul­ti­cen­ter, phase 2 mTOR inhib­i­tor Any donor type, Completed Day-28 CR/PR rates sim­i­lar
BMT CTN risk), pre­vi­ously untreated Sirolimus vs pred­ni­sone 2 mg/kg con­di­tion­ing for sirolimus vs pred­ni­sone
1501(19) aGVHD requir­ing sys­temic Child, adult, (64.8% vs 73%)
ther­apy, older-adult Sirolimus asso­ci­ated with
AAscore 1-2 bio­marker ages reduced ste­roid expo­
assess­ment (performed sure and hyper­gly­ce­
at ran­dom­i­za­tion) mia, improved QOL, and
increased risk for TMA
NCT03846479 Low-risk aGVHD, Minnesota Single arm, monotherapy JAK1 inhib­i­tor Any donor type, Recruiting Day-28 Minnesota stan­dard-risk
stan­dard risk, AA score 1 Itacitinib con­di­tion­ing clin­i­cal cri­te­ria (CR, PR, TRM)
Ages 12-75 Day-28 AA score 1
years
NCT04144036 Standard-risk aGVHD, Single-cen­ter nonrandomized Humanized mono­clo­nal Any donor type, Recruiting Day-28 SAE, MTD of
Minnesota risk, AA score phase 1 anti­body bind­ing to con­di­tion­ing neihulizumab, plasma Cmax
1-2 Neihulizumab dose esca­la­tion CD162 (PSGL-1) Ages ≥18 years

644  |  Hematology 2021  |  ASH Education Program

High-risk aGVHD
NCT02133924 High-risk aGVHD, Multicenter, phase 2 Humanized recom­bi­nant Any donor type, Recruiting Day-28 CR
AA score 3 Natalizumab + pred­ni­sone anti­body bind­ing to con­di­tion­ing
alpha-4 integrin Ages ≥18 years
NCT0252502917 Arm 1: high-risk aGVHD Nonrandomized phase ½ uhCG, which may pro­ Any donor type, Active, not Phase 1: MTD of uhCG/EGF in
(Minnesota risk, or Standard of care IST + Pregnyl mote immune tol­er­ con­di­tion­ing recruiting com­bi­na­tion with stan­dard IST
AA score 3, or AREG ance, pro­vi­des source Ages 0-76 years Phase 2: day-28 CR, PR, MR,
≥33 pg/mL) of EGF for epi­the­lial and no response
Arm 2a: ste­roid-depen­dent repair (Pregnyl®)
aGVHD
Arm 2b: SR-aGVHD
NCT03158896 De novo high-risk aGVHD Phase 1, dose esca­la­tion Immunosuppressive Ages 18-75 Active, not Day-45 treat­ment-related SAE
SR-aGVHD Umbilical cord-derived, ex vivo cul­ prop­er­ties: sup­press years recruiting
tured, and expanded Wharton’s pro­lif­er­a­tion of acti­vated
Jelly MSC T cells, increase pro­duc­
tion of reg­u­la­tory T cells,
shift immune response
toward tol­er­ance
NCT04061876 High-risk grade 2-4 aGVHD Randomized, phase 2 JAK1/2 inhib­i­tor Any donor type Recruiting Day-28 ORR
(based on ST2, REG3α, or Ruxolitinib + cor­ti­co­ste­roids vs Ages 14-65
other exper­i­ment object) cor­ti­co­ste­roids alone years
NCT04167514 High-risk aGVHD requir­ing Randomized, dou­ble-blind, pla­cebo- Serine pro­te­ase inhib­i­tor Any donor type, Recruiting Day-28 CR, PR
cor­ti­co­ste­roids (first-line con­trolled, mul­ti­cen­ter phase 3 con­di­tion­ing
ther­apy) AAT + cor­ti­co­ste­roids vs cor­ti­co­ste­ Ages ≥18 years
roids alone
NCT04291261 High-risk aGVHD, AA score Single-arm phase 2 Induces apo­pto­sis of Any donor type, Active, not Day-28 CR
2-3 ECP with methoxsalen +2 mg/kg mono­nu­clear cells con­di­tion­ing recruiting
cor­ti­co­ste­roids Ages ≥18 years
NCT04397367 Grade 2-4 or high-risk aGVHD Nonrandomized JAK1/2 inhib­i­tor Any donor type Recruiting Day-28 CR
(based on ST2, REG3α, or Ruxolitinib + cor­ti­co­ste­roids Ages 14-65
other exper­i­ment object) years
AAT, alpha-1 antitrypsin; ECP, extra­cor­po­real photopheresis; IST, immu­no­sup­pres­sive ther­apy; JAK1, Janus kinase 1; MR, mixed response; MSC, mes­en­chy­mal stem cell; MTD, max­i­mum tol­er­ated dose;
mTOR, mech­a­nis­tic tar­get of rapamycin; PR, par­tial response; PSGL-1, P-selectin gly­co­pro­tein 1; QOL, qual­ity of life; REG3α, regenerating islet-derived 3-alpha; SAE, seri­ous adverse event; ST2, sup­
pres­sor of tumor­i­gen­e­sis; TMA, throm­botic microangiopathy; TRM, trans­plant-related mor­tal­ity.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/642/1852146/642newell.pdf by guest on 13 December 2021
treated with ganciclovir, a vancomycin-resis­tant Enterococcus
faecium uri­ nary tract infec­tion treated with linezolid, and BK
cys­ti­tis pal­li­ated with pyridium. The plan was to start ruxolitinib
after insur­ance approval and with the con­trol of cytomegalovirus
reactivation; how­ever, she was never a ­ ble to start ther­apy. On
day 32 post-HCT, she devel­oped a pro­duc­tive cough, hypo­ten­
sion, and tachy­car­dia and required oxy­gen and was trans­ferred
to the inten­sive care unit for the man­age­ment of sep­tic shock.
Labs showed pan­cy­to­pe­nia and new lung infil­trates. She was
given high-flow oxy­gen, nor­epi­neph­rine, and broad-spec­trum
anti­bi­ot­ics. There was some ini­tial improve­ment, but on day 36
she devel­oped atrial fibril­la­tion with rapid ven­tric­u­lar response,
was started on an amiodarone drip, and devel­oped respi­ra­tory
dis­tress. Her respi­ra­tory sta­tus wors­ened, and she died on day 41.
SR-aGVHD
Definition of SR-aGVHD
The EBMT-NIH-CIBMTR Task Force suggested the fol­low­ing def­i­
ni­tions for SR-aGVHD or ste­roid-resis­tant aGVHD:1 pro­gres­sion of
aGVHD within 3 to 5 days of treat­ment with ≥2 mg/kg/d pred­ni­sone
equiv­a­lent, or2 fail­ure to improve with 5 to 7 days of treat­ment, or
incom­plete response after more than 28 days of immu­no­sup­pres­
sive ther­apy includ­ing ste­roids.3,11 SR-aGVHD has also been rec­og­
nized as (a) wors­en­ing GVHD man­i­fes­ta­tions in patients receiv­ing
≥1 mg/kg/d pred­ni­sone equiv­a­lent ≥2 days prior to ste­roid dose
taper­ing; (b) per­sis­tent grade 2 to 4 GVHD with­out improve­ment
≥7 days dur­ing con­tin­ued treat­ment with >0.4 mg/kg/d pred­ni­
sone equiv­a­lent, or (c) ini­tial improve­ment followed by exac­er­
ba­tion ≥3 days dur­ing ste­roid taper at any dose of >0.4 mg/kg/d
pred­ni­sone equiv­a­lent.20 In prac­tice, it can be very chal­leng­ing to
know how long to wait before adding a sec­ond-line agent.
Treatment options for SR-aGVHD
In May 2019, the FDA approved the first treat­ment for SR-aGVHD:
ruxolitinib, an inhib­i­tor of Janus kinase 1 and 2, for pedi­at­ric and
adult patients 12 years of age or older.21 FDA approval was based
upon the REACH-1 trial (NCT02953678, study INCB 18424-271), a
phase 2, mul­ti­cen­ter, open-label, sin­gle-arm study of 71 patients,
49 of whom had grades 2 to 4 SR-aGVHD (includ­ ing patients
who failed ste­roid treat­ment with or with­out receiv­ing addi­tional
GVHD ther­apy).21 A starting dose of ruxolitinib, 5 mg twice daily,
was admin­is­tered with meth­yl­pred­nis­o­lone.5 At day 28, the over­
all response rate (ORR) was 55% (83% for grade 2, 41% for grade
3, and 43% for grade 4 aGVHD), includ­ing 27% with a CR, with a
median time to first response of 7 days (range, 6-49). Ruxolitinib
was fur­ther inves­ti­gated in a phase 3, mul­ti­cen­ter, open-label ran­
dom­ized trial com­par­ing the effi­cacy of ruxolitinib vs inves­ti­ga­tor’s out celiac dis­ease. Fecal elas­tase was mildly low at 168 µg/g of
choice of ther­apy in patients 12 years of age and older with SR-
aGVHD (REACH-2, NCT02913261).22 Day-28 ORR was higher in the
ruxolitinib group vs the con­trol group (62% vs 39%, P < .001), with
34% and 19% CR, respec­tively; dura­ble day-56 ORR was also higher
in the ruxolitinib group vs the con­trol group (40% vs 22%, P < .001).
Ruxolitinib was asso­ci­ated with a higher inci­dence of throm­bo­cy­
to­pe­nia and a mod­est increase in ane­mia and cyto­meg­a­lo­vi­rus
infec­tion. In the REACH-1 study, the median time to death or new
aGVHD ther­apy was 5.7 months; in the REACH-2 study, median
fail­ure-free sur­vival was 5.0 months. Previous series have shown
21,22
that most patients die from SR-aGVHD within 6 months of diag­no­
sis, and only 25% to 30% of patients sur­ Prakash Singh
vive beyond Shekhawat
2 years. 23,24
Other exper­ i­
men­tal approaches to SR-aGVHD with vary­
ing mech­ a­
nisms of action are being tested in clin­ i­
cal tri­
als,
recently sum­ma­rized by Abedin and Hamadani and Martin.20,25
With admi­ra­ble humil­ity in his “How I Treat” pub­li­ca­tion, Dr
Martin admit­ted, “Truth be told, I do not know how to treat
SR-aGVHD.” This state­ment under­scores the tre­men­dous dif­fi­
culty in over­com­ing this largely fatal dis­ease despite decades
of intense study. Truth be told, we do not fully under­stand the
path­o­phys­i­­ol­ogy of SR-aGVHD. Most novel ther­a­peu­tics have
been potent immu­no­sup­pres­sants, yet that may not be the cor­
rect or only approach. In a pre­clin­i­cal model, there were no dif­
fer­ences in T-cell expan­sion, acti­va­tion, or enhanced cyto­kine
pro­duc­tion in donor T cells in mice with respon­sive vs refrac­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/642/1852146/642newell.pdf by guest on 13 December 2021
tory aGVHD.26 Recently, murine SR-aGVHD was described as
driven by inter­leu­kin 22-depen­dent dysbiosis, with a pro­tec­tive
role of CXCR3hi mono­nu­clear phago­cytes in reduc­ing bac­te­rial
trans­lo­ca­tion.27 Gene expres­sion stud­ies from human colo­rec­tal
biop­sies showed that human SR-aGVHD is char­ac­ter­ized by tis­
sue response to dam­age, cel­lu­lar stress, and mac­ro­phage accu­
mu­la­tion, not T-cell pro­lif­er­a­tion.28 Collectively, these recent
stud­ies sug­gest that future ther­a­peu­tic efforts in SR-aGVHD,
in addi­tion to targeting the ini­tial T-cell-medi­ated dam­age and
inflam­ma­tion, might also con­sider stud­ies of agents designed
to enhance tis­sue repair and to cor­rect dysbiosis while try­ing to
avoid broad immu­no­sup­pres­sion and its inher­ent risks of infec­
tion17,29-34. Recently described tar­gets such as CD83 sug­gest this
may be fea­si­ble.35
CLINICAL CASE
A 23-year-old man under­went a myeloablative, matched unre­
lated donor peripheral blood stem cell transplant for high-risk
T-cell acute lym­pho­blas­tic leu­ke­mia and was diag­nosed with
stage 2 skin, stage 3 GI, over­all Minnesota high-risk aGVHD.
He was treated with high-dose ste­ roids, uri­
nary-derived
human chorionic gonad­o­tro­pin (uhCG), and EGF in clin­i­cal trial
NCT02525029. He achieved a CR at 4 weeks after the ini­ti­a­tion
of ther­apy, which was sustained at 8 weeks. Over the ensu­ing
2 years, he had 6 hos­pi­tal­i­za­tions with a recur­rence of nau­sea,
vomiting, and diar­rhea. With each epi­sode he was tested for
infec­tions (1 epi­sode was due to norovirus) and other com­pli­ca­
tions, yet he would tran­siently improve with a mod­est increase
in pred­ni­sone dose. At 3 years posttransplant, he was read­
mitted with a body mass index of 16 and recal­ci­trant anorexia
and diar­rhea. Repeat intes­ti­nal biop­sies showed vil­lous blunt­
ing, but the patient and donor absence of HLA-DQ2/8 ruled
stool, sug­ges­tive of mild/mod­er­ate pan­cre­atic exo­crine insuf­
fi­ciency. Fecal calprotectin was mildly ele­vated at 72.8 mg/kg,
suggesting the pres­ ence of inflam­ ma­tion/neu­ tro­
phils in the
intes­ti­
nal lumen (nor­ mal, 0-49.9  mg/kg). Plasma AREG was
mildly ele­vated at 29.1 pg/mL, suggesting tis­sue dam­age and
mild sys­temic inflam­ma­tion (nor­mal, <5 pg/mL; active severe
intes­ti­nal aGVHD is typ­i­cally >33  pg/ml). Plasma cit­rul­line was
low at 0.9 µmol/dL (nor­mal, 1.3-6.0 µmol/dL; val­ues <1 µmol/dL
are pre­dic­tive of total vil­lous atro­phy).36 D-xylose absorp­tion
was low, with a serum level of 7 pg/mL at 2 hours (nor­mal, 32-
58  pg/mL), suggesting mal­ ab­sorp­tion. Budesonide and TPN
were resumed. After a week­long hos­pi­tal­i­za­tion, he was dis­
Acute GVHD  | 645

Figure 1. Considerations for therapy of GVHD in the first line and beyond.
charged from the hos­pi­tal tol­er­at­ing some oral intake but likely
to be depen­dent on TPN long-term given the find­ings of low
enterocyte mass (low cit­rul­line) and mal­ab­sorp­tion (low D-xy­
lose absorp­tion).
Steroid-depen­dent aGVHD
This patient high­lights a com­mon clin­i­cal sce­nario of a prolonged
aGVHD course of wax­ ing and wan­ ing symp­ toms. “Steroid-
depen­dent” aGVHD can be defined as (a) only achiev­ing a par­
tial (not com­plete) response to ste­roids after 8 weeks, (b) still
requir­ing >10 mg/m2 pred­ni­sone after 8 weeks or any pred­ni­sone
at all­after 10 weeks, or (c) a flare of aGVHD symp­toms requir­ing
at least a 25% increase in pred­ni­sone dose.37 Steroid-depen­dent
aGVHD is expe­ri­enced by 31% of patients with aGVHD.37 While it
is not asso­ci­ated with increased mor­tal­ity, it may be asso­ci­ated
with mor­bid­ity and a prolonged health care bur­den, as this case
descrip­tion high­lights. This patient had high-risk aGVHD with an
ini­tial response to high-dose cor­ti­co­ste­roids and uhCG/EGF but
sub­se­quently flared, suggesting that one or more under­ly­ing
causes had not been com­pletely addressed by his ini­tial ther­apy.
Figure 1 shows the dif­fer­en­tial diag­noses and con­sid­er­ations
in eval­u­at­ing aGVHD. An increase in immu­no­sup­pres­sion may
not be the right treat­ment for all­patients with wors­en­ing symp­
toms. Several ques­tions must be answered in the com­ing years
to improve out­comes. Can bio­mark­ers be used repeat­edly over
weeks to months as a guide to taper­ing immu­no­sup­pres­sion?
Which patients need dif­fer­ent modes of sup­port­ive care (eg,
reme­di­at­ion of dysbiosis vs tis­sue dam­age), and can this even
be dis­tin­guished bio­log­i­cally? How long should adjunct repair-
based ther­ap ­ ies such as uhCG/EGF be con­tin­ued to achieve
max­i­mal muco­sal healing? What other tar­gets of aGVHD (eg,
the endo­the­lium) should be treated? Additional clin­i­cal tri­als are
urgently needed to address these ques­tions.
646  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/642/1852146/642newell.pdf by guest on 13 December 2021
Conflict-of-inter­est dis­clo­sure
Shernan G. Holtan: advi­sor: Incyte, Generon.
Laura F. Newell: no conflicts to disclose.
Off-label drug use
Laura F. Newell: no off-label drug use.
Shernan G. Holtan: Off-label drug use for SGH: Urinary-derived hu­
man chorionic gonadotropin has Orphan Drug Designation from
the Food and Drug Administration for the treatment of acute GVHD.
Correspondence
Shernan G. Holtan, Division of Hematology, Oncology, and Trans­
plantation, Department of Medicine, University of Minnesota,
55455 Minneapolis, MN; e-mail: sgholtan@umn​­.edu.
References
1. Hahn T, McCarthy PL Jr, Hassebroek A, et  al. Significant improve­ment in
sur­vival after allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion dur­ing a period
of sig­nif­i­cantly increased use, older recip­i­ent age, and use of unre­lated
donors. J Clin Oncol. 2013;31(19):2437-2449.
2. Gooley TA, Chien JW, Pergam SA, et al. Reduced mor­tal­ity after allo­ge­
neic hema­to­poi­etic-cell trans­plan­ta­tion. N Engl J Med. 2010;363(22):2091-
2101.
3. Khoury HJ, Wang T, Hemmer MT, et  al. Improved sur­ vival after acute
graft-ver­sus-host dis­ease diag­no­sis in the mod­ern era. Haematologica.
2017;102(5):958-966.
4. van Bekkum D, Vos O, Weyzen WW. The path­o­gen­e­sis of the sec­ond­ary
dis­ease after for­eign bone mar­row trans­plan­ta­tion in x-irra­di­ated mice.
Blood. 2020;135(20):1739-1749.
5. Jagasia M, Perales MA, Schroeder MA, et al. Ruxolitinib for the treat­ment of
ste­roid-refrac­tory acute GVHD (REACH1): a mul­ti­cen­ter, open-label phase 2
trial. Blood. 2020; 135(20):1739-1749.
6. MacMillan ML, Robin M, Harris AC, et al. A refined risk score for acute graft-
ver­sus-host dis­ease that pre­dicts response to ini­tial ther­apy, sur­vival, and
trans­plant-related mor­tal­ity. Biol Blood Marrow Transplant. 2015;21(4):761-
767.
7. Przepiorka D, Weisdorf D, Martin P, et  al. 1994 con­sen­sus con­fer­ence on
acute GVHD grad­ing. Bone Marrow Transplant. 1995;15(6):825-828.
8. Weisdorf DJ, Hurd D, Carter S, et al. Prospective grad­ing of graft-ver­sus-
host dis­ease after unre­lated donor mar­row trans­plan­ta­tion: a grad­ing algo­
rithm ver­sus blinded expert panel review. Biol Blood Marrow Transplant.
2003;9(8):512-518.
9. Levine JE, Hogan WJ, Harris AC, et al. Improved accu­racy of acute graft-
ver­sus-host dis­ease stag­ing among mul­ti­ple cen­ters. Best Pract Res Clin
Haematol. 2014;27(3-4):283-287.
10. Harris AC, Young R, Devine S, et  al. International, mul­ti­cen­ter stan­dard­i­
za­tion of acute graft-ver­sus-host dis­ease clin­i­cal data col­lec­tion: a report
from the Mount Sinai Acute GVHD International Consortium. Biol Blood
Marrow Transplant. 2016;22(1):4-10.
11. Schoemans HM, Lee SJ, Ferrara JL, et al; EBMT (Euro­pean Society for Blood
and Marrow Transplantation) Transplant Complications Working Party and
the “EBMT−NIH (National Institutes of Health)−CIBMTR (Center for Interna­
tional Blood and Marrow Transplant Research) GvHD Task Force.” EBMT-
NIH-CIBMTR task force posi­tion state­ment on stan­dard­ized ter­mi­nol­ogy
and guid­ ance for graft-ver­ sus-host dis­ ease assess­ ment. Bone Marrow
Transplant. 2018;53(11):1401-1415.
12. MacMillan ML, DeFor TE, Holtan SG, Rashidi A, Blazar BR, Weisdorf DJ. Val­
idation of Minnesota acute graft-ver­sus-host dis­ease risk score. Haemato-
logica. 2020;105(2):519-524.
13. Adom D, Rowan C, Adeniyan T, Yang J, Paczesny S. Biomarkers for allo­ge­neic
HCT out­comes. Front Immunol. 2020;11:673.
14. Srinagesh HK, Özbek U, Kapoor U, et al. The MAGIC algo­rithm prob­a­bil­ity is
a val­i­dated response bio­marker of treat­ment of acute graft-ver­sus-host dis­
ease. Blood Adv. 2019;3(23):4034-4042.
15. Holtan SG, Khera N, Levine JE, et al. Late acute graft-ver­sus-host dis­ease:
a pro­spec­tive anal­y­sis of clin­i­cal out­comes and cir­cu­lat­ing angio­genic fac­
tors. Blood. 2016;128(19):2350-2358.
16. Holtan SG, DeFor TE, Panoskaltsis-Mortari A, et al. Amphiregulin modifies
the Minnesota acute graft-ver­sus-host dis­ease risk score: results from BMT
CTN 0302/0802. Blood Adv. 2018;2(15):1882-1888.
17. Holtan SG, Hoeschen AL, Cao Q , et al. Facilitating res­o­lu­tion of life-threat­
en­ing acute GVHD with human chorionic gonad­o­tro­pin and epi­der­mal
growth fac­tor. Blood Adv. 2020;4(7):1284-1295.
18. Mielcarek M, Furlong T, Storer BE, et al. Effectiveness and safety of lower
dose pred­ni­sone for ini­tial treat­ment of acute graft-ver­sus-host dis­ease: a
ran­dom­ized con­trolled trial. Haematologica. 2015;100(6):842-848.
19. Pidala J, Hamadani M, Dawson P, et al. Randomized mul­ti­cen­ter trial of siro­
limus vs pred­ni­sone as ini­tial ther­apy for stan­dard-risk acute GVHD: the
BMT CTN 1501 trial. Blood. 2020;135(2):97-107.
20. Martin PJ. How I treat ste­roid-refrac­tory acute graft-ver­sus-host dis­ease.
Blood. 2020;135(19):1630-1638.
21. Przepiorka D, Luo L, Subramaniam S, et  al. FDA approval sum­mary: rux­
olitinib for treat­ment of ste­roid-refrac­tory acute graft-ver­sus-host dis­ease.
Oncologist. 2020;25(2):e328-e334.
22. Zeiser R, von Bubnoff N, Butler J, et al; REACH2 Trial Group. Ruxolitinib for
glu­co­cor­ti­coid-refrac­tory acute graft-ver­sus-host dis­ease. N Engl J Med.
2020;382(19):1800-1810.
Prakash Singh Shekhawat
23. Xhaard A, Rocha V, Bueno B, et al. Steroid-refrac­tory acute GVHD: lack of
long-term improved sur­vival using new gen­er­a­tion anticytokine treat­ment.
Biol Blood Marrow Transplant. 2012;18(3):406-413.
24. Rashidi A, DeFor TE, Holtan SG, Blazar BR, Weisdorf DJ, MacMillan ML. Out­
comes and pre­dic­tors of response in ste­roid-refrac­tory acute graft-ver­sus-
host dis­ease. Biol Blood Marrow Transplant. 2019;25(11):2297-2302.
25. Abedin S, Hamadani M. Experimental phar­ma­ceu­ti­cals for ste­roid-refrac­
tory acute graft-ver­sus-host dis­ease. J Exp Pharmacol. 2020;12:549-557.
eCollection 2020.
26. Toubai T, Rossi C, Tawara I, et al. Murine mod­els of ste­roid refrac­tory graft-
ver­sus-host dis­ease. Sci Rep. 2018;8(1):12475.
27. Song Q , Wang X, Wu X, et al. IL-22-depen­dent dysbiosis and mono­nu­clear
phago­cyte deple­tion con­trib­ute to ste­roid-resis­tant gut graft-ver­sus-host
dis­ease in mice. Nat Commun. 2021;12(1):805.
28. Holtan SG, Shabaneh A, Betts BC, et al. Stress responses, M2 mac­ro­phages,
and a dis­tinct micro­bial sig­na­ture in fatal intes­ti­nal acute graft-ver­sus-host
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/642/1852146/642newell.pdf by guest on 13 December 2021
dis­ease. Nature. 2015;528(7583):560-564.
29. Lindemans CA, Calafiore M, Mertelsmann AM, et  al. Interleukin-22 pro­
motes intes­ti­nal-stem-cell-medi­ated epi­the­lial regen­er­a­tion. Nature. 2015;
528(7583):560-564.
30. Bruce DW, Stefanski HE, Vincent BG, et  al. Type 2 innate lym­phoid cells
treat and pre­vent acute gas­tro­in­tes­ti­nal graft-ver­sus-host dis­ease. J Clin
Invest. 2017;127(5):1813-1825.
31. Norona J, Apostolova P, Schmidt D, et al. Glucagon-like pep­tide 2 for intes­ti­
nal stem cell and Paneth cell repair dur­ing graft-ver­sus-host dis­ease in mice
and humans. Blood. 2020;136(12):1442-1455.
32. Michonneau D, Latis E, Curis E, et al. Metabolomics anal­y­sis of human acute
graft-ver­sus-host dis­ease reveals changes in host and microbiota-derived
metab­o­lites. Nat Commun. 2019;10(1):5695.
33. Beckman MF, Morton DS, Bahrani Mougeot F, Mougeot JC. Allogenic stem
cell trans­plant-asso­ci­ated acute graft ver­sus host dis­ease: a com­pu­ta­tional
drug dis­cov­ery text min­ing approach using oral and gut microbiome sig­
na­tures. Support Care Cancer. 2021;29(4):1765-1779.
34. Rashidi A, Ebadi M, Shields-Cutler RR, et al. Early E. casseliflavus gut col­
o­ni­za­tion and out­comes of allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion.
PLoS One. 2019;14(8):e0220850.
35. Shrestha B, Walton K, Reff J, et al. Human CD83-targeted chi­me­ric anti­gen
recep­tor T cells pre­vent and treat graft-ver­sus-host dis­ease. J Clin Invest.
2020;130(9):4652-4662.
36. Crenn P, Vahedi K, Lavergne-Slove A, Cynober L, Matuchansky C, Messing
B. Plasma cit­rul­line: a marker of enterocyte mass in vil­lous atro­phy-asso­ci­
ated small bowel dis­ease. Gastroenterology. 2003;124(5):1210-1219.
37. El Jurdi N, Rayes A, MacMillan ML, et  al. Steroid-depen­dent acute GVHD
after allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion: risk fac­tors and clin­i­cal
out­comes. Blood Adv. 2021;5(5):1352-1359.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000300
Acute GVHD  | 647
 UPDATE IN GRAFT- VERSUS - HOST DISEASE

Updates in chronic graft-versus-host disease

Betty K. Hamilton

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/648/1851934/648hamilton.pdf by guest on 13 December 2021

Blood and Marrow Transplant Program, Hematology and Medical Oncology, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH

Chronic graft-versus-host disease (GVHD) is the leading cause of late morbidity and mortality after allogeneic hema-
topoietic cell transplantation. Symptoms and manifestations of chronic GVHD are heterogeneous and pleomorphic,
and there are no standard treatments beyond corticosteroids. Therapy is typically prolonged, and chronic GVHD and
its treatment are associated with adverse effects that have a significant impact on long-term quality of life and func-
tional status. Several advances have been made over the last 2 decades to define the diagnosis of chronic GVHD as
well as its severity and response criteria for clinical trials. Further understanding into the biologic mechanisms of the
development of chronic GVHD has led to the investigation of several novel immunomodulatory and targeted therapies.
Multi-institutional collaboration and pharmaceutical support in the development of therapies based on sound biologic
mechanisms and clinical trials with defined end points and responses have led to several promising agents on the hori-
zon of approval for treatment of chronic GVHD. This article reviews advances in our knowledge of chronic GVHD and its
biologic framework to improve approaches to prevention and treatment.

LEARNING OBJECTIVES
• Describe biologic phases and mechanisms of chronic GVHD development
• Describe novel approaches to chronic GVHD prevention and treatment
• Understand the continued need for high-quality multicenter clinical trials based on biologic rationale and using
chronic GVHD end points

CLINICAL CASE
A 64-year-old male with a history of relapsed acute mye-
logenous leukemia in second complete remission under-
went a reduced-intensity matched unrelated-donor
peripheral blood stem cell transplant. Graft-versus-host
disease (GVHD) prophylaxis consisted of tacrolimus and
mycophenolate mofetil (MMF). In the setting of tapering
immunosuppression, 4 months after transplant he devel-
oped worsening fatigue, anorexia, skin rash, dry eyes, and
dry mouth. Skin biopsy and ocular evaluation confrmed
chronic GVHD. Tacrolimus was increased to therapeutic
levels, and he was started on 1mg/kg prednisone. Topical
therapies including dexamethasone oral rinses and ocu-
lar agents were used, including scleral lenses. The GVHD
symptoms resolved on steroids; however, he had several
adverse effects to steroids, such as hyperglycemia, irrita-
bility, edema, and cytomegalovirus reactivation. Systemic
corticosteroids were subsequently successfully tapered
off. The patient was eager to get off all immunosuppres-
sion, but again, after a taper of tacrolimus, he had a flare of
GVHD with oral, eye, and skin involvement. There was some
improvement with maximizing topical therapy and increas-
ing tacrolimus. The patient was very reluctant to consider
systemic corticosteroids or other medications. He devel-
oped worsening fatigue, anorexia, shortness of breath
(pleural effusions), lower-extremity swelling, dysphagia,
and anemia/thrombocytopenia. Additional workup to rule
out other etiologies was consistent with a GVHD flare. He
was restarted on 0.5mg/kg prednisone with plans for a
quick taper and started on ibrutinib. He had an excellent
response; his counts improved and symptoms resolved. He
was able to taper off prednisone without a flare in GVHD.
He remains on low-dose tacrolimus and ibrutinib.
Introduction
Allogeneic hematopoietic cell transplantation (HCT) is
the only potentially curative therapy for many high-risk
hematologic malignancies, metabolic and immunode-
fciencies, and bone marrow failure syndromes, and the
number of transplant procedures continues to increase.
Despite signifcant advances in donor selection, condi-
tioning regimens, and supportive care, chronic GVHD
remains the leading cause of late morbidity and mortal-

648 | Hematology 2021 | ASH Education Program

ity.1,2 The inci­dence of chronic GVHD ranges from 30% to 50%,
depending on GVHD pro­phy­laxis reg­i­mens and the donor graft
source.1 Chronic GVHD, once established, is a het­ero­ge­neous
and pleo­ mor­phic syn­drome in which most patients have at
least 3 involved organs, and treat­ment typ­i­cally requires the
prolonged (median 2-3.5 years) use of immu­ no­sup­ pres­ sive
agents.3 Chronic GVHD and its treat­ment are thus often asso­ci­
ated with late mor­tal­ity as well as sev­eral adverse effects with
sig­nif­i­cant impact on long-term qual­ity of life and func­tional
sta­tus among HCT sur­vi­vors.2,4
The National Institutes of Health (NIH) chronic GVHD pro­jects
of 2005 and 2014 have pro­vided a crit­i­cal frame­work for sub­
stan­tial advances in the trans­plant field over the last decade,5,6
outlining stan­dard guide­lines and def­i­ni­tions for chronic GVHD
diag­no­sis, grad­ing, and response cri­te­ria for clin­i­cal tri­als.7,8 Fur-
ther refine­ment of response def­i­ni­tions has enhanced the reli­
abil­ity and prac­ti­cal util­ity of these mea­sures in clin­i­cal tri­als,
ulti­mately lead­ing to the first US Food and Drug Administration
(FDA) approval of treat­ment for chronic GVHD.9 In 2020, the NIH
Chronic GVHD Consensus Conference fur­ ther set a for­ ward-
think­ing agenda to address gaps and needs in future research
over the next 3 to 7 years. This arti­ cle sum­ ma­ rizes recent
advances in our under­stand­ing of chronic GVHD and its (1) eti­­
ol­ogy and pre­ven­tion, (2) devel­op­ment and diag­no­sis, and (3)
novel treat­ments.
Etiology and pre­ven­tion
Advances in our under­stand­ing of chronic GVHD have led to the
rec­og­ni­tion of sev­eral donor, recip­i­ent, and trans­plant fac­tors
that con­trib­ute to its ini­ti­a­tion and devel­op­ment. Established
clin­i­cal risk fac­tors include growth fac­tor-mobi­lized periph­eral
blood stem cells, mismatched or unre­lated donor grafts, female-
to-male trans­plan­ta­tion, older recip­i­ent age, and acute GVHD
his­tory.10 Preclinically, although no murine model ade­ quately
mim­ics the clin­i­cal spec­trum of chronic GHVD,11 bio­logic insights
from mul­ti­ple mod­els with dis­tinct pathophysiologies have laid
a foun­da­tion lead­ing to fur­ther clin­i­cal stud­ies and new treat­
ment par­a­digms. Several detailed and com­pre­hen­sive reviews
of chronic GVHD biol­ogy and pre­clin­i­cal mod­els already exist
and are out­side the scope of this arti­cle.12-14 Key ele­ments in the
path­o­phys­i­­ol­ogy, how­ever, will be briefly sum­ma­rized here to
pro­vide a frame­work for the updates on novel ther­a­peu­tic tar­
gets for pre­ven­tion and treat­ment.
Etiology and path­o­gen­e­sis
Chronic GVHD is a result of com­plex and dynamic mech­a­nisms
and can be thought to occur within 3 bio­logic phases: (1) early
inflam­ma­tion due to tis­sue injury, (2) thy­mic injury and T- and B-
cell dysregulation, and (3) tis­sue repair and fibro­sis (Figure 1).14
Although these pro­cesses are often overlapping and depen­dent,
they do not nec­es­sar­ily need to occur in sequence, or even occur
at all­. As with acute GVHD, exper­i­men­tal mod­els sug­gest that the
ini­tial phase of chronic GVHD also begins with dam­age of host tis­
sues by pretransplant con­di­tion­ing and the sub­se­quent release of
inflam­ma­tory cyto­kines. Damage to gut tis­sues and the release of
micro­bial con­tents result in the acti­va­tion of anti­gen-presenting
cells, and inflam­ma­tory cyto­kines stim­u­late the acti­va­tion of do-
nor alloreactive T cells, driv­ing helper T (Th1)/T-cyto­toxic 1 (Tc1),
and Th17/Tc17 dif­fer­en­ti­a­tion and expan­sion of effec­tor T cells,
caus­ing fur­ther cyto­tox­ic­ity to host tar­get cells.12
Prakash Singh Shekhawat
Tissue dam­age to the thy­mus and the sec­ond­ary lym­phoid
organs pre­dis­poses patients to sub­se­quent chronic GVHD. The
sec­ ond phase is thus char­ ac­
ter­ized by thy­ mic injury, which
allows for the emer­gence of and the selec­tion of auto- and allo-
reactive T-cell pop­u­la­tions.13 A loss of cen­tral and periph­eral tol­
er­ance leads to impaired or defi­cient reg­u­la­tory T (Treg) and
B cells.15 Donor T fol­lic­u­lar helper (Tfh) cells have been shown to
expand in the sec­ond­ary lym­phoid organs and pro­mote the sur­
vival, expan­sion, and dif­fer­en­ti­a­tion of donor B cells into aber­rant
anti-host-immu­no­glob­u­lin-pro­duc­ing plasma cells via cyto­kines
such as IL-21 and B-cell acti­vat­ing fac­tor.16 Chronic inflam­ma­tion
is thought to be maintained by Th17 cells that have escaped
immune reg­u­la­tion.12
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/648/1851934/648hamilton.pdf by guest on 13 December 2021
The third phase is marked by aber­rant tis­sue repair. Activated
mac­ro­phages depen­dent on IL-17 and col­ony-stim­u­lat­ing fac­tor
(CSF-1) have been shown to play an impor­tant role in this by
pro­duc­ing transforming growth fac­ tor β and plate­ let-derived
growth fac­tor alpha.17 This leads to fibro­blast acti­va­tion, result-
ing in chronic GVHD man­ i­
fes­
ta­tions such as sclero­ derma or
bronchiolitis obliterans. Thus, it is increas­ingly clear that sub­clin­
i­cal path­o­genic pro­cesses crit­i­cal to the devel­op­ment of chronic
GVHD begin long before the spe­cific clin­i­cal man­i­fes­ta­tions
become evi­dent.
Chronic GVHD pre­ven­tion
Calcineurin inhib­i­tor (CNI)-based pro­phy­laxis, typ­i­cally in com­bi­
na­tion with meth­o­trex­ate (MTX) or MMF, has been stan­dard prac­
tice over the past 3 decades for GVHD pre­ven­tion.18 Although
these reg­i­mens may have rel­a­tive ben­e­fit in lim­it­ing acute GVHD,
they are not effec­tive against chronic GVHD and, par­a­dox­i­cally,
may sup­port its devel­op­ment by blocking thy­mic cen­tral tol­er­
ance and Treg func­tion.19-21 The need for improve­ments in GVHD
pre­ven­tion strat­e­gies has been rec­og­nized as a pri­or­ity by the
field,22 and sev­eral novel strat­e­gies and end points have been
inves­ti­gated.
Although sev­eral pro­phy­laxis strat­e­gies have been shown
to decrease the inci­dence of chronic GVHD, their effects on
immune recon­ sti­
tu­tion and dis­ease relapse remain prob­ lem­
atic. Because donor-derived effec­tor T cells are crit­i­cal ini­ti­a­
tors of chronic GVHD, approaches to deplete T cells from the
hema­to­poi­etic cell graft have been inves­ti­gated, dem­on­strat­ing
sig­nif­i­cant reduc­tions in the inci­dence and sever­ity of chronic
GVHD. While the ex vivo removal of T cells by the pos­i­tive selec­
tion of CD34+ cells has become the pre­dom­i­nant approach in
clin­i­cal use,23 other novel approaches, such as αβ-T-cell graft
deple­tion, have dem­on­strated sig­nif­i­cant reduc­tions in mod­er­
ate and severe chronic GVHD.24 Although T-cell deple­tion has
not been shown to increase relapse risk, it is asso­ci­ated with
delayed immune recon­sti­tu­tion and higher rates of infec­tion-
related mor­tal­ity,23 which was recently con­firmed in a phase 3
mul­ti­cen­ter Blood and Marrow Transplant Clinical Trials Network
(BMT CTN) trial.25 In vivo T-cell deple­tion with antithymocyte
glob­ u­lin (ATG) has also been effec­ tive in reduc­ ing the inci­
dence of chronic GVHD. Several phase 3 stud­ies have con­firmed
reduc­tions in chronic GVHD with ATG, with sub­se­quent supe­
rior qual­ity of life and GVHD-free sur­vival.26-28 However, con­tro­
versy remains on its effect on over­all sur­vival due to increases
in relapse and nonrelapse mor­tal­ity, as well as opti­mal dos­ing,
for­mu­la­tion, and depen­dency on lym­pho­cyte count at the time
of admin­is­tra­tion.28
Chronic graft-ver­sus-host dis­ease | 649

Figure 1.  Proposed biologic phases of chronic GVHD. A few novel therapeutic approaches targeting biologic pathways are highlight-
ed. Reproduced with permission from the Cleveland Clinic Center for Medical Art and Photography.
The role of donor B cells in chronic GVHD biol­ogy has also
pro­vided a strong ratio­nale for the eval­u­a­tion of B-cell deple­
tion strat­eg­ ies for chronic GVHD pre­ven­tion. Prior stud­ies have
incor­po­rated mono­clo­nal anti-B-cell ther­apy as part of the con­
di­tion­ing reg­im­ en or the peritransplant period with the intent
of reduc­ing relapse of B-cell malig­nan­cies.29 Although sig­nif­i­cant
deple­tion of B cells and reduced rates of chronic GVHD have
been reported,30 con­cerns over the delayed recon­sti­tu­tion of
B-cell immu­nity and excess infec­tions remain a con­cern.
Posttransplant cyclo­ phos­ pha­
mide (PTCy) is a promising
approach used suc­ cess­fully to pre­
vent GVHD in the HLA-​
mismatched/haploidentical set­ting.31 Recent work has pro­
posed that the effi­cacy of PTCy is due to the reduc­tion of
alloreactive CD4+ effec­tor T-cell pro­lif­er­a­tion, impaired func­
tion­ al­
ity of sur­ viv­
ing CD4+ and CD8+ effec­ tor T cells, and
pref­er­en­tial recov­ery of CD4+ Treg cells.32 PTCy has been inves­
ti­gated both with and with­out CNI in the fully HLA-matched
650  |  Hematology 2021  |  ASH Education Program
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/648/1851934/648hamilton.pdf by guest on 13 December 2021
set­ting and has shown itself to be a poten­tially effec­tive strat­
egy for both acute and chronic GVHD pre­ven­tion.
The opti­ mal GVHD pro­ phy­ laxis reg­
i­
men for the pre­ ven­
tion of chronic GVHD remains elu­sive. A bench­mark anal­y­sis of
novel GVHD pre­ven­tion was conducted through the Center for
International Blood and Marrow Transplant Research to select
prom­is­ing agents and new, clin­i­cally mean­ing­ful out­come mea­
sure­ ments for fur­ ther inves­
ti­
ga­tion through the BMT CTN.33
Several com­pos­ite GVHD end points were explored, includ­ing
chronic GVHD relapse-free sur­vival (CRFS), which includes sur­
vival with­out devel­op­ment of chronic GVHD, dis­ease relapse, or
death, and GVHD relapse-free sur­vival (GRFS), which includes
sur­vival with­out acute grade 3 to 4 GVHD, chronic GVHD, dis­
ease relapse, or death. The results of these ana­ly­ses led to the
devel­op­ment of 2 mul­ti­cen­ter BMT CTN clin­i­cal tri­als. PROG-
RESS 1 (BMT CTN 1203) was a ran­dom­ized phase 2, 3-arm trial
inves­ti­gat­ing bortezomib/Tac/MTX, maraviroc/Tac/MTX, and
PTCy/Tac/MMF in reduced-inten­sity HCT, eval­u­at­ing GRFS as
a pri­mary end point. PTCy was asso­ci­ated with supe­rior GRFS
and had the low­est inci­dence of chronic GVHD requir­ing immu­
no­sup­pres­sion com­pared to other arms.34 This led to the ongo­
ing BMT CTN trial 1703 (NCT03959241) phase 3 trial com­par­ing
PTCy/Tac/MMF with Tac/MTX, with the pri­mary end point of
GRFS, poten­tially establishing a new stan­dard of care for GVHD
pre­ven­ tion. PROGRESS 2 (BMT CTN 1301) was a ran­ dom­ ized
phase 3, open-label, 3-arm trial com­par­ing CNI-free approaches
of CD34 selec­tion, PTCy, and Tac/MTX, with a pri­mary end point
of CRFS. While there were no sta­tis­ti­cally sig­nif­i­cant dif­fer­ences
in CRFS between the 3 approaches, CD34+ selec­tion was found
to be asso­ci­ated with worse over­all sur­vival, driven by increased
infec­tion-related mor­tal­ity.25
Other novel strat­e­gies
As we work to fur­ther under­stand the devel­op­ment of GVHD, we
now rec­og­nize sev­eral addi­tional poten­tial tar­gets. For exam­ple,
instead of blan­ket B- or T-cell deple­tion, pre­ven­tion strat­e­gies
to block the devel­op­ment of path­o­genic B cells, or to inhibit
Tfh cells, may bet­ter pre­vent chronic GVHD with­out affect­ing
other out­comes. Further, strat­e­gies to aug­ment Tregs, as have
been reported with novel agents such as IL-2,35 ruxolitinib,36
hypomethylating agents,37 extra­cor­po­real photopheresis,38 pro-
teasome inhib­i­tors,39 and adop­tive trans­fer of Tregs,40 may need
addi­tional explo­ra­tion. Table 1 pro­vi­des a sum­mary of sev­eral
recently published and ongo­ ing stud­ ies that include chronic
GVHD as a pri­mary or as part of a pri­mary com­pos­ite end point.
Development and diag­no­sis
While the 2005 and 2014 NIH Chronic GVHD Consensus Con-
ference papers established and refined stan­dard def­i­ni­tions for
chronic GVHD diag­no­sis,5,6 the 2020 Chronic GVHD Consen-
sus Conference fur­ther high­lighted sev­eral impor­tant con­sid­
er­ations in the devel­op­ment and diag­no­sis of chronic GVHD,
focus­ ing espe­ cially on the neces­ sity of early rec­og­ni­tion.41
These include the need for ear­lier clin­i­cal iden­ti­fi­ca­tion by both
pro­vid­ers and patient/care­giv­ers, poten­tially with the use of
tech­nol­ogy such as telehealth and apps, as well as the need
for bet­ter iden­ti­fi­ca­tion of pre­ced­ing symp­toms or bio­mark­
ers asso­ci­ated with sub­se­quent more severe/mor­bid forms
of chronic GVHD—includ­ing bio­mark­ers from blood and tis­
sue or find­ings from imag­ing or func­tional test­ing. Trials spe­
cif­i­cally inves­ti­gat­ing the devel­op­ment of chronic GVHD with
increased mon­i­tor­ing using blood, fluid, tis­sue, imag­ing, and
patient-reported symp­toms at time points through­out trans­
plant to detect early signs of dis­ease are cur­rently ongo­ing
(NCT04372524, NCT04188912).
The early iden­ti­fi­ca­tion of symp­toms or sub­clin­i­cal indi­ca­tors
of chronic GVHD is nec­es­sary for the pro­posed approach of pre­
emp­tive treat­ment of chronic GVHD,42 poten­tially allowing for
the pre­ven­tion of the more severe and mor­bid phe­no­types of
chronic GVHD. In a pre­ven­ta­tive/pre­emp­tive approach with rit-
uximab posttransplant, a mod­est decrease in chronic GVHD and
ste­roid-requir­ing chronic GVHD was found.30 Ideally, future pre­
emp­tive approaches will have the advan­tage of directing treat­
ment at patients who have been iden­ti­fied as at greatest risk,
allowing for early, more targeted and sub­se­quently less long-
term, dam­ag­ing ther­apy.
Prakash Singh Shekhawat
Treatment of established chronic GVHD
Given the het­ero­ge­neous man­i­fes­ta­tions of established chron-
ic GVHD, clear guide­lines for treat­ment are lacking. High-dose
cor­ti­co­ste­roids, typ­i­cally 0.5 to 1 mg/kg/d, remain the first-line
stan­dard treat­ment for chronic GVHD.3 While there is no cur­
rent evi­dence to sup­port any ben­e­fit from addi­tional agents
in the up-front set­ting, they are often con­sid­ered in severe
cases. The response rate to ste­roids alone is about 50%, with
more than half of patients requir­ing sec­ond-line ther­apy with-
in 2 years. A num­ber of sec­ond-line agents of vary­ing effi­cacy
and toxicities have been inves­ ti­
gated over the past sev­ eral
decades, high­light­ing the het­ero­ge­ne­ity in prac­tice, var­i­abil­
ity in response rates, and chal­lenges in deter­min­ing the best
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/648/1851934/648hamilton.pdf by guest on 13 December 2021
treat­ment options.43 Corticosteroids thus remain the stan­dard
despite con­tinu­ing to be asso­ci­ated with sev­eral short- and
long-term toxicities, includ­ ing infec­tion, myop­ a­
thy, edema,
hyper­gly­ce­mia, bone loss, avas­cu­lar necro­sis, cat­a­racts, and
sleep/mood dis­tur­bances. In clin­i­cal prac­tice, top­i­cal ther­a­pies
and sup­port­ive mea­sures to min­i­mize these adverse effects are
of crit­i­cal impor­tance.44
The ideal treat­ment for chronic GVHD con­tin­ues to be elu­
sive. Goals of ther­apy are sev­eral in num­ber but should include
symp­tom bur­den reduc­tion and improve­ment of qual­ity of life,
pre­ven­tion of pro­gres­sion and inflam­ma­tory activ­ity, pre­ven­tion
of fibro­sis and dis­abil­ity, pres­er­va­tion of response to allow for
with­drawal of immu­no­sup­pres­sion, repair and mod­u­la­tion of the
immune sys­tem, and, ulti­mately, improve­ment of chronic GVHD
mor­tal­ity.
With these aims in mind, the focus of novel chronic GVHD
treat­ ment has thus shifted from the use of broad, long-term
immu­no­sup­pres­sion toward the inves­ti­ga­tion of immu­no­mod­u­
la­tory agents that tar­get path­ways rel­e­vant to the path­o­phys­i­­ol­
ogy of the dis­ease (Figure 1). Several mul­ti­cen­ter phase 2 and 3
clin­i­cal tri­als have now inves­ti­gated mul­ti­ple agents in the treat­
ment of up-front or ste­roid-refrac­tory chronic GVHD, and a few
recently reported tri­als are briefly reviewed here.
Ibrutinib is a small-molec­u­lar tyro­sine kinase inhib­i­tor target-
ing Bruton’s tyro­sine kinase (BTK), which plays an impor­tant role
in B-cell recep­tor sig­nal­ing and acti­va­tion of B cells and T cells.
A phase 2 trial of ibrutinib in 42 patients with ste­roid-refrac­tory
chronic GVHD led to the first FDA approval of a sec­ond-line ther­
apy for chronic GVHD, dem­on­strat­ing an over­all response rate
of 67%. Approximately 71% of respond­ers con­tin­ued to show
a response for at least 20 weeks, with responses gen­ er­
ally
observed across all­organ sys­tems involved.9 Although the num­
ber of patients treated on this trial was rel­a­tively small, the FDA
approval of this drug high­lights the suc­cess of multi-insti­tu­tional
col­lab­o­ra­tion and phar­ma­ceu­ti­cal sup­port in devel­op­ing a trial
based on pre­clin­i­cal data and clear bio­logic mech­a­nisms, with
defined clin­i­cal end points.
Ruxolitinib is a JAK 1/2 inhib­i­tor that suppresses T-cell acti­va­
tion by inhibiting cyto­kine recep­tor-medi­ated sig­nal­ing for a vari­
ety of proinflammatory cyto­kines and is cur­rently FDA approved
for the treat­ment of ste­roid-refrac­tory acute GVHD.36 A phase 3
study of ruxolitinib vs best avail­­ able ther­apy (BAT) in ste­ roid-
refrac­tory chronic GVHD was recently reported,45 dem­on­strat­ing
an over­all response rate of 50% com­pared to 26% in the BAT arm.
Failure-free sur­vival was sig­nif­i­cantly lon­ger for ruxolitinib-treated
patients, and symp­tom improve­ment was greater with ruxolitinib
Chronic graft-ver­sus-host dis­ease | 651
Table 1.  Recently published and ongo­ing pre­ven­tion stud­ies with chronic GVHD as a pri­mary end point

Phase Prevention reg­i­men Primary end point/out­come Reference or NCT num­ber

Phase 3 ATG vs no ATG with CSA/MTX Chronic GVHD at 2 years: 32.2% (ATG) vs 68.7% (no ATG), P < .001 26
Phase 3 ATG vs no ATG with CNI/MTX Freedom from sys­temic immu­no­sup­pres­sion with­out resump­tion 27
up to 12 months: 37% (ATG) vs 16% (no ATG), OR 4.25, P = .0006
Phase 3 ATG vs no ATG with Tac/MTX Moderate-severe chronic GVHD-free sur­vival: 48% (ATG) vs 44% 28
(no ATG), P = .47
Phase 2 PTCy/CSA Chronic GVHD requir­ing sys­temic treat­ment: 16% (95% CI, 5%- 48
28%)
Phase 2 PTCy/Tac/MMF vs bortezomib/Tac/MTX GRFS: HR 0.72, P = .04 (PTCy/Tac/MMF) vs HR 0.98, 49
vs maraviroc/Tac/MTX P = .92 (bortezomib/Tac/MTX) vs HR 1.10, P = .49
(maraviroc/Tac/MTX)

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/648/1851934/648hamilton.pdf by guest on 13 December 2021

Phase 2 Rituximab (D100, 6, 9, 12 months) post-HCT Chronic GVHD: 48%, chronic GVHD requir­ing ste­roids: 31% 30
Phase 3 CD34 selec­tion vs PTCy vs Tac/MTX CRFS: 60.2% (CD34), 60.3% (PTCy), 56.6% (Tac/MTX), P = .72 25
Phase 1/2 Ixazomib/Tac/Sir Moderate/severe chronic GVHD NCT03225417
Phase 2 Palifermin Chronic GVHD NCT02356159
Phase 2 Obinutuzumab Chronic GVHD requir­ing ste­roids NCT02867384
Phase 2 Ixazomib/CNI/MTX Acute or chronic GVHD NCT03082677
Phase 2 PTCy/Tac/MMF GRFS NCT03128359
Phase 2 PTCy/Sir/MMF Extensive chronic GVHD NCT03192397
Phase 2 Tocilizumab/Tac/MTX GRFS NCT03699631
Phase 2 Ex vivo deple­tion of CD45RA +
GRFS NCT03779854
cells/Tac/MTX
Phase 3 Atorvastatin/Tac/MTX Chronic GVHD NCT03066466
Phase 3 PTCy/Tac/MMF vs Tac/MTX GRFS NCT03959241
CSA, cyclosporine.

com­pared to BAT (24% vs 11%). Based on the results of this study,
ruxolitinib has been granted pri­or­ity review by the FDA for the
treat­ment of chronic GVHD, and an FDA deci­sion on drug approval
is expected before the end of the year.
Belumosudil is an oral selec­ tive rho-asso­ ci­
ated coiled-coil
kinase 2 (ROCK) inhib­i­tor targeting Tfh-cell gen­er­a­tion and thus
B-cell dif­fer­en­ti­at­ion and sur­vival. Results of the phase 2 trial
of belumosudil in ste­ roid-refrac­ tory chronic GVHD were also
recently reported,46 dem­on­strat­ing an over­all response rate of
74% in heavily pretreated patients. Responses were seen across
all­affected organs, and of respond­ers, 49% maintained response
for at least 20 weeks. Failure-free sur­vival was 77% at 6 months.
Based on the results of this study, belumosudil has also been
granted pri­or­ity review by the FDA.
The need to inves­ti­gate other novel ther­a­pies either targeted
toward fibro­sis or already being devel­oped for other fibrotic dis­
eases has been rec­og­nized. For exam­ple, pirefenidone, which
inhib­its transforming growth fac­tor β sig­nal­ing, is cur­rently FDA
approved for idi­o­pathic pul­mo­nary fibro­sis and has been shown
in mouse mod­els to restore pul­mo­nary func­tion and reverse lung
fibro­sis. Its study in chronic GVHD-related bronchiolitis obliter-
ans is ongo­ing (NCT03315741). Axatilimab is a mono­clo­nal anti­
body against CSF-1, which reg­u­lates mono­cyte dif­fer­en­ti­a­tion
into mac­ro­phages pro­mot­ing fibro­sis, and has been shown to
have poten­tial effi­cacy in heavily pretreated chronic GVHD. A
phase 2 study fur­ther eval­ua ­ t­ing this is ongo­ing (NCT03604692).
Table 2 sum­ma­rizes sev­eral other up-front and ste­roid-refrac­tory
chronic GVHD tri­als.
Despite our improved under­stand­ing of chronic GVHD, it
remains a com­plex dis­ease marked by sig­nif­i­cant immune dys­
func­tion, dis­abil­ity, and toxicities, and chronic GVHD-related
non-relapse mortality keeps ris­ing over time.47 Practically, the
choice of first- and sec­ond-line ther­apy con­tin­ues to depend on
phy­si­cian pref­er­ences and clin­i­cal expe­ri­ence, patient dis­ease
his­tory and comorbidities, cost/avail­abil­ity of treat­ment, and
access to clin­i­cal tri­als. As we prog­ress away from broad immu­
no­ sup­pres­sion, there remains a need to inves­ ti­gate agents
that tar­get the bio­logic path­ways rel­e­vant to the path­o­phys­
i­­ol­ogy of dis­ease and chronic GVHD symp­toms; to ulti­mately
under­stand which drug will work when. The iden­ti­fi­ca­tion and
use of blood and tis­sue bio­mark­ers, imag­ing, and other novel
approaches can bet­ter iden­tify spe­cific clin­i­cal and bio­logic
phe­no­types to dif­fer­en­ti­ate chronic GVHD and work toward a
more tai­lored, per­son­al­ized ther­apy for patients.
Reflecting back to our clin­i­cal case, this sce­nario dem­on­
strates sev­eral aspects of chronic GVHD devel­op­ment, evo­lu­
tion, and treat­ment chal­lenges and high­lights the impor­tance of
under­stand­ing the goals of ther­apy from both a patient and pro­
vider per­spec­tive. While broad immu­no­sup­pres­sion may reduce
symp­ tom bur­ den and decrease inflam­ ma­tory activ­ ity in the
short term, there remains a need to fur­ther inves­ti­gate targeted
agents to pre­serve func­tion, pro­vide a prolonged response to
allow for the with­drawal of med­i­ca­tions, pre­vent dis­abil­ity, and
ulti­mately improve qual­ity of life and over­all mor­tal­ity. The con­
tin­ued under­stand­ing of bio­logic mech­a­nisms through multi-
insti­tu­tional col­lab­o­ra­tive pre­clin­i­cal studies, clin­i­cal tri­als with

652  |  Hematology 2021  |  ASH Education Program

Table 2.  Active ongo­ing stud­ies for sys­temic treat­ment in up-front or ste­roid-refrac­tory chronic GVHD

Drug mech­a­nism/
Phase Treatment Indication Enrollment Primary end point NCT num­ber
tar­get
Pilot Itacitinib JAK inhib­i­tor Steroid refrac­tory 40 ORR at 6 months NCT04200365
Phase 2 Itacitinib and ECP JAK inhib­i­tor, Treg Up-front 58 ORR at 24 weeks; NCT0446182
dose-lim­it­ing tox­ic­ity
Phase 1 SHR0302 JAK inhib­i­tor Up-front 30 Adverse events NCT04146207
Phase 1 Abatacept T-cell costim block­ Steroid refrac­tory 22 Maximum tol­er­ated dose NCT01954979
ade
Phase Itacitinib JAK inhib­i­tor Up-front 431 Dose deter­mi­na­tion; NCT03584516
2/3 ORR at 6 months

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/648/1851934/648hamilton.pdf by guest on 13 December 2021

Phase 2 Mesenchymal stem cells Immunomodulatory Up-front 152 ORR at 12 weeks NCT04692376
Phase 2 IL-2 and ECP Treg Steroid refrac­tory 24 ORR at 4 weeks NCT03007238
Phase 1/2 Ofatumumab CD20 B-cell anti­body Up-front 44 ORR at 6 months; max­i­ NCT010680965
mum tol­er­ated dose
Phase 2 KD025 (belumosudil) ROCK inhib­i­tor Steroid refrac­tory 166 ORR at 6 months NCT03640481
Phase 2 Acalabrutinib BTK inhib­i­tor Recurrent/pro­gres­sive 50 Best response at 6 NCT04198922
months
Phase 1/2 Glasdegib Hedgehog sig­nal­ing Sclerosis/ste­roid- 21 Maximum tol­er­ated dose NCT03415867
inhib­i­tor refrac­tory
Phase 2 Ibrutinib/rituximab BTK inhib­i­tor/CD20 Up-front 35 Off immu­no­sup­pres­sion NCT04235036
B-cell anti­body at 8 weeks
Phase 1 Leflunomide Immunomodulatory Steroid depen­dent 24 Dose-lim­it­ing tox­ic­ity NCT04212416
Phase 2 Axalitimab CSF-1 inhib­i­tor Steroid refrac­tory 210 ORR at 6 months NCT04710576
Phase 2 Ibrutinib BTK inhib­i­tor Up-front 40 ORR at 6 months NCT04294641
Phase 1/2 Baricitinib JAK inhib­i­tor Steroid refrac­tory 31 Safety and effi­cacy NCT02759731
Phase 3 Ruxolitinib JAK inhib­i­tor Steroid refrac­tory 330 ORR at day 168 NCT03112603
ECP, extracorporeal photopheresis; ORR, overall response rate; ROCK, rho-asso­ci­ated coiled-coil kinase.

care­ful cor­rel­a­tive stud­ies of blood, fluid, tis­sue sam­ples, imag­
ing, and patient-reported symp­toms—are nec­es­sary to achieve
a more per­son­al­ized approach to the pre­ven­tion and treat­ment
of chronic GVHD.
Conflict-of-inter­est dis­clo­sure
Betty K. Hamilton: con­sul­tancy: Syndax, Equilium.
Off-label drug use
Betty K. Hamilton: none.
Correspondence
Betty K. Hamilton, Blood and Marrow Transplant Program, Hema-
tology and Medical Oncology, Taussig Cancer Institute, Cleve-
land Clinic, 9500 Euclid Ave, Cleveland, OH 44195; e-mail: ham-
iltb2@ccf​­.org.
References
1. Arai S, Arora M, Wang T, et al; Graft-ver­sus-Host Disease Working Com-
mittee of the CIBMTR. Increasing inci­dence of chronic graft-ver­sus-host
dis­ease in allo­ge­neic trans­plan­ta­tion: a report from the Center for Inter-
national Blood and Marrow Transplant Research. Biol Blood Marrow Trans­
plant. 2015;21(2):266-274.
2. Pidala J, Kurland B, Chai X, et  al. Patient-reported qual­ity of life is asso­
ci­ated with sever­ity of chronic graft-ver­sus-host dis­ease as mea­sured by
NIH cri­te­ria: report on base­line data from the Chronic GVHD Consortium.
Blood. 2011;117(17):4651-4657. Prakash Singh Shekhawat
3. Flowers ME, Martin PJ. How we treat chronic graft-ver­sus-host dis­ease.
Blood. 2015;125(4):606-615.
4. Wingard JR, Majhail NS, Brazauskas R, et  al. Long-term sur­vival and late
deaths after allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion. J Clin Oncol.
2011;29(16):2230-2239.
5. Filipovich AH, Weisdorf D, Pavletic S, et al. National Institutes of Health
con­sen­sus devel­op­ment pro­ject on cri­te­ria for clin­i­cal tri­als in chronic
graft-​ ver­
sus-host dis­ ease, I: Diagnosis and Staging Working Group
report. Biol Blood Marrow Transplant. 2005;11(12):945-956.
6. Jagasia MH, Greinix HT, Arora M, et al. National Institutes of Health con­sen­
sus devel­op­ment pro­ject on cri­te­ria for clin­i­cal tri­als in chronic graft-ver­
sus-host dis­ease, I: the 2014 Diagnosis and Staging Working Group report.
Biol Blood Marrow Transplant. 2015;21(3):389-401.e1.
7. Pavletic SZ, Martin P, Lee SJ, et al; Response Criteria Working Group. Mea-
suring ther­a­peu­tic response in chronic graft-ver­sus-host dis­ease: National
Institutes of Health con­sen­sus devel­op­ment pro­ject on cri­te­ria for clin­i­cal
tri­als in chronic graft-ver­sus-host dis­ease, IV: Response Criteria Working
Group report. Biol Blood Marrow Transplant. 2006;12(3):252-266.
8. Lee SJ, Wolff D, Kitko C, et al. Measuring ther­a­peu­tic response in chronic
graft-ver­sus-host dis­ease. National Institutes of Health con­sen­sus devel­
op­ment pro­ject on cri­te­ria for clin­i­cal tri­als in chronic graft-ver­sus-host
dis­ease, IV: the 2014 Response Criteria Working Group report. Biol Blood
Marrow Transplant. 2015;21(6):984-999.
9. Miklos D, Cutler CS, Arora M, et al. Ibrutinib for chronic graft-ver­sus-host
dis­ease after fail­ure of prior ther­apy. Blood. 2017;130(21):2243-2250.
10. Flowers ME, Inamoto Y, Carpenter PA, et al. Comparative anal­y­sis of risk
fac­tors for acute graft-ver­sus-host dis­ease and for chronic graft-ver­sus-
host dis­ease according to National Institutes of Health con­sen­sus cri­te­ria.
Blood. 2011;117(11):3214-3219.
11. Zeiser R, Blazar BR. Preclinical mod­els of acute and chronic graft-ver­sus-
host dis­ease: how pre­dic­tive are they for a suc­cess­ful clin­i­cal trans­la­tion?
Blood. 2016;127(25):3117-3126.

Chronic graft-ver­sus-host dis­ease | 653

12. MacDonald KP, Hill GR, Blazar BR. Chronic graft-ver­sus-host dis­ease: bio­log­
i­cal insights from pre­clin­i­cal and clin­i­cal stud­ies. Blood. 2017;129(1):13-21.
13. Zeiser R, Blazar BR. Pathophysiology of chronic graft-ver­sus-host ­dis­ease
and ther­a­peu­tic tar­gets. N Engl J Med. 2017;377(26):2565-2579.
14. Cooke KR, Luznik L, Sarantopoulos S, et al. The biol­ogy of chronic graft-ver­
sus-host dis­ease: a task force report from the National Institutes of Health
con­sen­sus devel­op­ment pro­ject on cri­te­ria for clin­i­cal tri­als in chronic graft-
ver­sus-host dis­ease. Biol Blood Marrow Transplant. 2017;23(2):211-234.
15. Khoder A, Sarvaria A, Alsuliman A, et  al. Regulatory B cells are enriched
within the IgM mem­ory and tran­si­tional sub­sets in healthy donors but are
defi­cient in chronic GVHD. Blood. 2014;124(13):2034-2045.
16. Sarantopoulos S, Stevenson KE, Kim HT, et al. High lev­els of B-cell acti­
vat­ing fac­tor in patients with active chronic graft-ver­sus-host dis­ease.
Clin Cancer Res. 2007;13(20):6107-6114.
17. Alexander KA, Flynn R, Lineburg KE, et al. CSF-1-depen­dent donor-derived
mac­ro­phages medi­ate chronic graft-ver­sus-host dis­ease. J Clin Invest.
2014;124(10):4266-4280.
18. Hamilton BK, Liu Y, Hemmer MT, et al. Inferior out­comes with cyclo­spor­ine
and mycophenolate mofetil after myeloablative allo­ge­neic hema­to­poi­etic
cell trans­plan­ta­tion. Biol Blood Marrow Transplant. 2019;25(9):1744-1755.
19. Sorror ML, Leisenring W, Deeg HJ, Martin PJ, Storb R. Twenty-year fol­low-
up of a con­trolled trial com­par­ing a com­bi­na­tion of meth­o­trex­ate plus
cyclo­spor­ine with cyclo­spor­ine alone for pro­phy­laxis of graft-ver­sus-host
dis­ease in patients admin­is­tered HLA-iden­ti­cal mar­row grafts for leu­ke­mia.
Biol Blood Marrow Transplant. 2005;11(10):814-815.
20. Jenkins MK, Schwartz RH, Pardoll DM. Effects of cyclo­spor­ine A on T cell
devel­op­ment and clonal dele­tion. Science. 1988;241(4873):1655-1658.
21. Zeiser R, Nguyen VH, Beilhack A, et al. Inhibition of CD4+CD25+ reg­ul­a­tory
T-cell func­tion by calcineurin-depen­dent inter­leu­kin-2 pro­duc­tion. Blood.
2006;108(1):390-399.
22. Ferrara JL, Anasetti C, Stadtmauer E, et al. Blood and Marrow Transplant
Clinical Trials Network State of the Science Symposium 2007. Biol Blood
Marrow Transplant. 2007;13(11):1268-1285.
23. Tamari R, Oran B, Hilden P, et al. Allogeneic stem cell trans­plan­ta­tion for
advanced myelodysplastic syn­drome: com­par­i­son of out­comes between
CD34+ selected and unmod­i­fied hema­to­poi­etic stem cell trans­plan­ta­tion.
Biol Blood Marrow Transplant. 2018;24(5):1079-1087.
24. de Witte MA, Janssen A, Nijssen K, et  al. αβ T-cell graft deple­tion for
allo­ge­neic HSCT in adults with hema­to­log­i­cal malig­nan­cies. Blood Adv.
2021;5(1):240-249.
25. Pasquini M, Luznik L, Logan B, et  al. Calcineurin inhib­i­tor-free graft-ver­
sus-host dis­ease (GVHD) pro­phy­laxis in hema­to­poi­etic cell trans­plan­ta­tion
(HCT) with myeloablative con­di­tion­ing reg­i­mens (MAC) and HLA-matched
donors: results of the BMT CTN 1301 PROGRESS II trial [ab­stract]. T ­ ransplant
Cell Ther. 2021;27(3 suppl). TCT abstract LBA1. Accessed 30 August 2021.
https:​­/​­/www​­.hematologyandoncology​­.net​­/supplements​­/highlights​­-in​
­-graft​­-vs​­-host​­-disease​­-from​­-the​­-2021​­-transplantation​­-cellular​­-therapy​­-tct​
­-meetings​­-of​­-the​­-astct​­-and​­-the​­-cibmtr​­/.
26. Kröger N, Solano C, Wolschke C, et al. Antilymphocyte glob­u­lin for pre­ven­
tion of chronic graft-ver­sus-host dis­ease. N Engl J Med. 2016;374(1):43-53.
27. Walker I, Panzarella T, Couban S, et al; Cana­ dian Blood and Marrow
Trans​plant Group. Pretreatment with anti-thy­mo­cyte glob­u­lin ver­sus no
anti-thy­mo­cyte glob­u­lin in patients with haematological malig­nan­cies
under­go­ing haemopoietic cell trans­plan­ta­tion from unre­lated donors:
a randomised, con­trolled, open-label, phase 3, multicentre trial. Lancet
Oncol. 2016;17(2):164-173.
28. Soiffer RJ, Kim HT, McGuirk J, et  al. Prospective, ran­dom­ized, dou­ble-
blind, phase III clin­i­cal trial of anti-T-lym­pho­cyte glob­u­lin to assess impact
on chronic graft-ver­sus-host dis­ease-free sur­vival in patients under­go­ing
HLA-matched unre­lated myeloablative hema­to­poi­etic cell trans­plan­ta­
tion. J Clin Oncol. 2017;35(36):4003-4011.
29. Laport GG, Wu J, Logan B, et al; Blood and Marrow Transplant Clinical Trials
Network. Reduced-inten­sity con­di­tion­ing with fludarabine, cyclo­phos­pha­
mide, and high-dose rituximab for allo­ge­neic hema­to­poi­etic cell trans­plan­
ta­tion for fol­lic­u­lar lym­phoma: a phase two mul­ti­cen­ter trial from the Blood
and Marrow Transplant Clinical Trials Network. Biol Blood Marrow Trans­
plant. 2016;22(8):1440-1448.
30. Cutler C, Kim HT, Bindra B, et al. Rituximab pro­phy­laxis pre­vents cor­ti­co­
ste­roid-requir­ing chronic GVHD after allo­ge­neic periph­eral blood stem cell
trans­plan­ta­tion: results of a phase 2 trial. Blood. 2013;122(8):1510-1517.
31. Luznik L, Jalla S, Engstrom LW, Iannone R, Fuchs EJ. Durable engraft­ment
of major his­to­com­pat­i­bil­ity com­plex-incom­pat­i­ble cells after nonmyeloab-
lative con­di­tion­ing with fludarabine, low-dose total body irra­di­a­tion, and
posttransplantation cyclo­phos­pha­mide. Blood. 2001;98(12):3456-3464.
654  |  Hematology 2021  |  ASH Education Program
32. Nunes NS, Kanakry CG. Mechanisms of graft-ver­sus-host dis­ease pre­ven­
tion by post-trans­plan­ta­tion cyclo­phos­pha­mide: an evolv­ing under­stand­
ing. Front Immunol. 2019;10(Novem­ber):2668.
33. Pasquini MC, Logan B, Jones RJ, et al. Blood and mar­row trans­plant clin­ic ­ al
tri­als net­work report on the devel­op­ment of novel end­points and selec­
tion of prom­is­ing approaches for graft-ver­sus-host dis­ease pre­ven­tion tri­
als. Biol Blood Marrow Transplant. 2018;24(6):1274-1280.
34. Bolanos-Meade J, Reshef R, Fraser RB, et  al. Novel approaches for graft-
ver­ sus-host dis­ ease (GVHD) pro­ phy­ laxis: pri­ mary results of prog­ ress I
mul­ti­cen­ter trial of matched allo­ge­neic hema­to­poi­etic cell trans­plan­ta­
tion (alloHCT) using reduced inten­sity con­di­tion­ing (RIC) BMT CTN 1203.
Abstract presented at: BMT Tandem Scientific Meeting; 25 Feb­ru­ary 2018;
Salt Lake City, UT. Late Breaking Abstracts. Accessed 30 August 2021.
https:​­/​­/bmt​­.confex​­.com​­/tandem​­/2018​­/meetingapp​­.cgi​­/Paper​­/11767.
35. Matsuoka K, Koreth J, Kim HT, et al. Low-dose inter­leu­kin-2 ther­apy restores
reg­u­la­tory T cell homeo­sta­sis in patients with chronic graft-ver­sus-host dis­
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/648/1851934/648hamilton.pdf by guest on 13 December 2021
ease. Sci Transl Med. 2013;5(179):179ra43.
36. Zeiser R, Socié G. The devel­op­ment of ruxolitinib for glu­co­cor­ti­coid-refrac­
tory acute graft-ver­sus-host dis­ease. Blood Adv. 2020;4(15):3789-3794.
37. Goodyear OC, Dennis M, Jilani NY, et al. Azacitidine aug­ments expan­sion of
reg­u­la­tory T cells after allo­ge­neic stem cell trans­plan­ta­tion in patients with
acute mye­loid leu­ke­mia (AML). Blood. 2012;119(14):3361-3369.
38. Bruserud O, Tvedt TH, Paulsen PQ , et  al. Extracorporeal photopheresis
(photochemotherapy) in the treat­ment of acute and chronic graft ver­
sus host dis­ease: immu­no­log­i­cal mech­a­nisms and the results from clin­i­cal
stud­ies. Cancer Immunol Immunother. 2014;63(8):757-777.
39. Blanco B, Pérez-Simón JA, Sánchez-Abarca LI, et  al. Treatment with bor-
tezomib of human CD4+ T cells pre­serves nat­ur­al reg­u­la­tory T cells and
allows the emer­gence of a dis­tinct sup­pres­sor T-cell pop­u­la­tion. Haema­
tologica. 2009;94(7):975-983.
40. Di Ianni M, Falzetti F, Carotti A, et  al. Tregs pre­vent GVHD and pro­mote
immune recon­sti­tu­tion in HLA-haploidentical trans­plan­ta­tion. Blood. 2011;
117(14):3921-3928.
41. Kitko CL, Pidala J, Schoemans HM, et  al. National Institutes of Health
con­sen­sus devel­op­ment pro­ject on cri­te­ria for clin­i­cal tri­als in chronic
graft-ver­sus-host dis­ease, IIa: the 2020 Clinical Implementation and Early
Diagnosis Working Group report: NIH cGVHD WG2a. Transplant Cell Ther.
2021;27(7):545-557.
42. Pidala J, Kitko C, Lee SJ, et al. National Institutes of Health con­sen­sus devel­
op­ment pro­ject on cri­te­ria for clin­i­cal tri­als in chronic graft-ver­sus-host dis­
ease, IIb: the 2020 Preemptive Therapy Working Group report. Transplant
Cell Ther. 2021;27(8):632-641.
43. Wolff D, Schleuning M, von Harsdorf S, et al. Consensus con­fer­ence on
clin­i­cal prac­tice in chronic GVHD: sec­ond-line treat­ment of chronic graft-
ver­sus-host dis­ease. Biol Blood Marrow Transplant. 2011;17(1):1-17.
44. Carpenter PA, Kitko CL, Elad S, et  al. National Institutes of Health con­
sen­sus devel­op­ment pro­ject on cri­te­ria for clin­i­cal tri­als in chronic graft-
ver­sus-host dis­ease, V: the 2014 Ancillary Therapy and Supportive Care
Working Group report. Biol Blood Marrow Transplant. 2015;21(7):1167-1187.
45. Zeiser RP, Ram N, Hasmi R, et al. Ruxolitinib (RUX) vs best avail­­able ther­apy
(BAT) in patients with ste­roid-refrac­tory/ste­roid-depen­dent chronic graft-
vs-host dis­ease (cGVHD): pri­mary find­ings from the phase 3, ran­dom­ized
REACH3 study. Blood. 2020;136(suppl 1):22-24.
46. Cutler C, Lee SJ, Arai S, et  al. Belumosudil for chronic graft-ver­sus-host
dis­ease (cGVHD) after 2 or more prior lines of ther­apy: the Rockstar Study
(KD025-213). Blood. 2020;136(suppl 1):45-46.
47. DeFilipp Z, Onstad L, Arai S, et  al. Non-relapse mor­tal­ity among patients
diag­nosed with chronic graft-ver­sus-host dis­ease: an updated anal­y­sis from
the chronic GVHD con­sor­tium. Transplant Cell Ther. 2021;27(3):S79-S80.
48. Mielcarek M, Furlong T, O’Donnell PV, et al. Posttransplantation cyclo­phos­
pha­mide for pre­ven­tion of graft-ver­sus-host dis­ease after HLA-matched
mobi­lized blood cell trans­plan­ta­tion. Blood. 2016;127(11):1502-1508.
49. Bolaños-Meade J, Reshef R, Fraser R, et al. Three pro­phy­laxis ­reg­im ­ ens (tac-
rolimus, mycophenolate mofetil, and cyclo­phos­pha­mide; t­acrolimus,
meth­o­trex­ate, and bortezomib; or tacrolimus, meth­o­trex­ate, and maraviroc)
ver­sus tacrolimus and meth­o­trex­ate for pre­ven­tion of graft-ver­sus-host dis­
ease with haemopoietic cell trans­plan­ta­tion with reduced-inten­sity con­di­
tion­ing: a randomised phase 2 trial with a non-randomised con­tem­po­ra­ne­ous
con­trol group (BMT CTN 1203). Lancet Haematol. 2019;6(3):e132-e143.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000301
 UPDATE IN GRAFT- VERSUS - HOST DISEASE

What else do I need to worry about when treating

graft-versus-host disease?

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/655/1851925/655jawahri.pdf by guest on 13 December 2021

Areej El-Jawahri
Hematology­Oncology, Massachusetts General Hospital, Boston, MA

Graft-versus-host disease (GVHD) is the main cause of morbidity and mortality in allogeneic hematopoietic stem cell
transplant survivors. Patients with acute and chronic GVHD often endure substantial symptom burden and quality of
life (QOL) and functional impairments. Living with GVHD affects multiple domains of patient-reported QOL, physical
functioning, and psychological well-being. Patients describe living with GVHD as a life-altering “full-time job” requiring
unique knowledge, personal growth, and resilient coping strategies. Managing the supportive care needs of patients
living with GVHD must include (1) monitoring of patient-reported QOL and symptom burden; (2) routine screening for
psychological distress and implementing therapeutic strategies to treat depression, anxiety, and posttraumatic stress
symptoms; (3) a systematic review of care needs by a multidisciplinary team experienced in managing transplant-related
complications and organ-specific GVHD symptoms; and (4) ensuring optimal prevention and management of infection
complications in this highly immunocompromised population. Improving the QOL in patients with GVHD requires a mul-
tidisciplinary approach with emphasis on aggressive symptom management, psychological coping, and promoting phys-
ical activity and rehabilitation in this population living with immense prognostic uncertainty and struggling to adapt to
this difficult and unpredictable illness.

LEARNING OBJECTIVES
• Understand the impact of GVHD on quality of life, functional status, symptom burden, and psychological distress
in allogeneic HCT survivors
• Develop strategies to address the supportive care needs of patients with GVHD

The supportive care needs of patients

CLINICAL CASE
Emma is a 39­year­old woman, a middle school teacher,
who was initially diagnosed with acute myeloid leukemia
and achieved a complete remission after induction che­
motherapy. Subsequently, she underwent a myeloablative
matched unrelated donor peripheral stem cell transplan­
tation during a 4­week hospitalization. Two weeks after
discharge, she developed diarrhea, prompting a hospital
readmission. She was diagnosed with acute skin and gas­
trointestinal graft­versus­host disease (GVHD) and initiated
first­line therapy with methylprednisolone 2mg/kg/d. She
had a complete response within 16 days of therapy and
was discharged from the hospital 3 weeks later. Due to
prolonged corticosteroid exposure, Emma developed ste­
roid myopathy and weakness, and she also struggled with
some lower extremity edema during the first few months
following her rehospitalization.
Prakash Singh Shekhawat
with acute GVHD
Acute GVHD is a common complication of allogeneic
hematopoietic stem cell transplant (HCT), which has a sig­
nificant impact on patient­reported quality of life (QOL),
physical functioning, and clinical outcomes.1,2 Classic acute
GVHD typically occurs within the first 100 days posttrans­
plant and predominantly affects the skin, liver, and gastro­
intestinal tract.1,2 Recent advances in the assessment and
management of acute GVHD rely on identification and risk
stratification of patients based on clinical staging of the
disease and blood biomarkers.3­5 Despite these advances,
there are substantial gaps in our understanding of the sup­
portive care needs of this population, particularly when
it comes to their QOL and overall functioning.6 Patients
with acute GVHD struggle with substantial symptom bur­
den, including nausea, anorexia, diarrhea, and decline in
functional status.6 As in Emma’s case, treatment of acute
GVHD typically involves high­dose corticosteroid therapy
Supportive care for GVHD | 655
along with other immu­no­sup­pres­sive agents, which may fur­ther
increase the risk of com­pli­ca­tions and symp­toms such as infec­
tions and weak­ness.
Studies com­pre­hen­sively exam­in­ing the sup­port­ive care
needs of patients with acute GVHD are lacking in part due to
(1) the lack of a valid and reli­able tool spe­cific for symp­toms of
acute GVHD; (2) the need for fre­quent patient-reported assess­
ments, which can place sub­stan­tial bur­den on this acutely ill
pop­u­la­tion; and (3) lack of robust stud­ies cor­re­lat­ing the objec­
tive response cri­te­ria with clin­i­cally mean­ing­ful changes in
QOL and acute GVHD symp­toms.6 Nonetheless, mul­ti­ple stud­
ies have shown that acute GVHD is asso­ci­ated with a decline
in patient-reported QOL, phys­i­cal func­tion­ing, role func­tion­ing,
social func­tion­ing, men­tal health, and gen­eral health.7-11 As high­
lighted in Emma’s clin­i­cal case, patients with acute GVHD often
spend prolonged peri­ods in the hos­pi­tal, which fur­ther con­trib­
utes to their social iso­la­tion and psy­cho­log­i­cal dis­tress. Given
the extent of trauma and iso­la­tion these patients expe­ri­ence dur­
ing their ill­ness course, stud­ies are also needed to exam­ine their
exis­ten­tial dis­tress and demor­al­iza­tion.
When car­ing for patients with acute GVHD, cli­ni­cians should
at min­i­mum screen for psy­cho­log­i­cal dis­tress. Prior work has
shown the fea­si­bil­ity of patient-reported out­comes mon­i­tor­ing
and psy­cho­so­cial dis­tress screen­ing in allo­ge­neic HCT recip­i­
ents.12,13 Studies inte­ grat­ing serial patient-reported out­ comes
mon­i­tor­ing dur­ing the acute GVHD course will help cli­ni­cians
guide clin­i­cal care in addressing the unmet needs of this pop­u­la­
tion, as well as mon­i­tor poten­tial response to ther­apy. In addi­tion,
clin­i­cal tri­als focused on the man­age­ment of acute GVHD must
also incor­po­rate serial patient-reported out­come assess­ments
that allow for robust exam­i­na­tion of response to ther­apy and
com­pli­ca­tions related to prolonged ste­roid use such as weak­
ness, func­tional decline, fatigue, insom­nia, and psy­cho­log­i­cal
dis­tress. Moreover, max­i­mal sup­port­ive care mea­sures, includ­ing
top­i­cal treat­ments for skin GVHD as well as anti­di­ar­rheal ther­apy
for patients with gas­tro­in­tes­ti­nal GVHD, should be implemented
rou­tinely in clin­i­cal prac­tice. Given the poten­tial risk of func­tional
decline, phys­i­cal ther­apy and reha­bil­i­ta­tion are also essen­tial
com­po­nents to max­i­mize the QOL and func­tion­ing of patients
with acute GVHD.14,15 Cumulative expo­sure to high-dose cor­ti­co­
ste­roids is asso­ci­ated with adverse side effects, includ­ing infec­
tion com­pli­ca­tions, avas­cu­lar necro­sis, oste­o­po­ro­sis, edema,
and ste­roid myop­a­thy.16 In fact, the inci­dence of ste­roid myop­a­
thy in patients with acute GVHD is as high as 41%.17 Studies have
shown that patients with acute GVHD have base­ line impair­
ments in their func­tion, which are wors­ened within 14 days of
receiv­ing cor­ti­co­ste­roid ther­apy.14 These find­ings under­score the
poten­tial need for early super­vised con­sul­ta­tion with phys­i­cal
ther­apy, occu­pa­tional ther­apy, and phys­i­cal med­i­cine and reha­
bil­i­ta­tion to closely mon­i­tor and inter­vene to improve phys­i­cal
func­tion­ing in this pop­u­la­tion.14 As in Emma’s case, patients can
also develop some lower extrem­ity edema and fluid reten­tion
due to cor­ti­co­ste­roid use, poor nutri­tion and hypoalbuminemia,
and their inflam­ma­tory state from acute GVHD. Prompt man­age­
ment of these com­pli­ca­tions and close atten­tion to nutri­tional
needs in this pop­u­la­tion can improve patients’ over­all QOL and
func­tion­ing. Notably, recent stud­ies have shown an asso­ci­a­tion
between the gut microbiota and the risk of devel­op­ing acute
GVHD as well as over­all trans­plant out­comes.18 Future ther­a­pies
for acute GVHD, includ­ing probiotics, nutri­tional sup­ple­ments,
656  |  Hematology 2021  |  ASH Education Program
and fecal microbiota trans­plan­ta­tion, are cur­rently being tested
in clin­i­cal tri­als.18
In a ran­ dom­ ized clin­
i­
cal trial, the inte­ gra­
tion of spe­ cialty
pal­ li­
at­ive care has been shown to improve a wide range of
patient-reported out­comes, includ­ing QOL, symp­tom bur­den,
fatigue, depres­sion, anx­i­ety, and post­trau­matic stress symp­toms
both acutely dur­ing HCT hos­pi­tal­i­za­tion and up to 6 months
post-HCT.19,20 Although these data are not spe­cific to patients
with acute GVHD, admis­sion for acute GVHD is a rea­son­able trig­
ger for spe­cialty pal­li­a­tive care given the role of pal­li­a­tive care in
man­ag­ing com­plex symp­toms, facil­i­tat­ing effec­tive cop­ing, cul­
ti­vat­ing prog­nos­tic aware­ness, and addressing the psy­cho­log­i­
cal and exis­ten­tial dis­tress in patients liv­ing with a seri­ous ill­ness.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/655/1851925/655jawahri.pdf by guest on 13 December 2021
Although com­pre­hen­sive geri­at­ric assess­ments have not been
tested in patients with acute GVHD, they focus on numer­ous
domains that are affected by patients with acute GVHD. A com­
pre­hen­sive geri­at­ric assess­ment model of care that incor­po­rates
a focus on func­tional capac­ity, cog­ni­tion, mood, polypharmacy,
social sup­port, finan­cial con­cerns, and nutri­tion, as well as over­
all goals of care, may be a ben­e­fic ­ ial frame­work to address the
needs of this pop­u­la­tion.
CLINICAL CASE (Con­tin­ued)
Emma recov­ered from her acute GVHD symp­toms and con­tin­ued
to make prog­ress toward her HCT recov­ery. At 8 months post-
HCT, Emma devel­oped symp­toms of chronic GVHD affect­ing her
skin, mouth, eyes, and vagina while taper­ing her immu­no­sup­pres­
sion ther­apy. Overall, she devel­oped mod­er­ate chronic GVHD
based on the num­ber of organ sys­tems involved. She received
treat­ment with sys­temic cor­ti­co­ste­roids, tacrolimus, ruxolitinib,
and ibrutinib. Given the sever­ity of her symp­toms, Emma has not
been ­able to return to work. She also strug­gles to man­age her
daily rou­tine with top­i­cal ther­a­pies for her oral, ocu­lar, skin, and
vag­i­nal GVHD.
QOL and symp­tom bur­den in chronic GVHD
Chronic GVHD is a debil­ i­
tat­
ing immu­ no­logic syn­
drome that
attacks mul­ti­ple organs and is the major cause of post-HCT mor­
bid­ity and mor­tal­ity.21-25 Patients with chronic GVHD strug­gle to
man­age their ill­ness, which often results in sub­stan­tial phys­i­cal
symp­toms, func­tional lim­i­ta­tions, and impaired QOL.9,21,22,26,27
Treatment of chronic GVHD, such as cor­ti­co­ste­roids, is lim­ited
in effi­cacy and has sig­ nif­i­
cant toxicities, thus con­trib­
ut­
ing to
the unpre­dict­able tra­jec­tory of this ill­ness and its unre­lent­ing
course.27 Studies have shown that the sever­ity of chronic GVHD
is asso­ci­ated with the extent of QOL, phys­i­cal, and func­tional
impair­ments seen in this pop­u­la­tion.28 The Lee Symptom Scale
is a reli­able and valid mea­sure for chronic GVHD symp­toms that
cor­re­lates with National Institutes of Health cri­te­ria as well as
patient-reported chronic GVHD sever­ity.29
Given the global and mul­ ti­
di­
men­
sional impact of chronic
GVHD in allo­ge­neic HCT sur­vi­vors, patients with chronic GVHD
should be reviewed by a team expe­ri­enced in man­ag­ing trans­
plant-related com­pli­ca­tions and the sup­port­ive care needs of
this pop­u­la­tion.30 Furthermore, trans­plant cen­ters should estab­
lish a clin­i­cal net­work of spe­cial­ists with an inter­est in chronic
GVHD to allow for mul­ti­dis­ci­plin­ary man­age­ment.30 Assessment also be a cause of psy­cho­log­i­cal dis­tress in patients liv­ing with
of QOL and chronic GVHD symp­toms is recommended for all­ chronic GVHD and deal­ing with an unre­lent­ing ill­ness. Although
patients liv­ing with chronic GVHD.28 Improving QOL in this pop­ stud­ies have focused on finan­ cial dis­
tress in allo­
ge­
neic HCT
u­la­tion requires a mul­ti­dis­ci­plin­ary approach with a focus on recip­i­ents,40 future assess­ment of the extent of finan­cial dis­tress
aggres­sive symp­tom man­age­ment, psy­cho­log­i­cal cop­ing, and and tox­ic­ity is needed for patients with chronic GVHD along with
pro­mot­ing phys­i­cal activ­ity and reha­bil­i­ta­tion.30 con­sid­er­ation for finan­cial nav­i­ga­tion pro­grams and other evi­
dence-based inter­ven­tions to reduce finan­cial tox­ic­ity and psy­
Psychological dis­tress in patients with chronic GVHD cho­log­i­cal dis­tress in this pop­u­la­tion.
Given the phys­ i­
cal symp­ tom bur­ den and uncer­ tainty about
the ill­ness course in patients with chronic GVHD, psy­cho­log­i­ Organ-spe­cific sup­port­ive care man­age­ment
cal dis­tress is prev­a­lent in this pop­u­la­tion, with 50% and 30% of for chronic GVHD
patients with chronic GVHD reporting depres­sion and anx­i­ety Management of organ-spe­ cific issues is a core com­ po­
nent
symp­toms, respec­tively.31,32 Notably, self-reported depres­ sion of opti­ mal sup­ port­ ive GVHD care that entails top­ i­
cal ther­

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/655/1851925/655jawahri.pdf by guest on 13 December 2021

symp­toms have been asso­ci­ated with lower sur­vival in patients
with chronic GVHD.28 In prior stud­ies, the use of adap­tive cop­
ing strat­e­gies was asso­ci­ated with lower psy­cho­log­i­cal dis­tress
in patients with chronic GVHD.29 Coping style is a mod­i­fi­able
con­struct that, if altered to be more adap­tive in the con­text of
a chronic med­i­cal stressor, could poten­tially change the tra­jec­
tory of QOL and psy­cho­log­i­cal dis­tress in patients with chronic
GVHD.32 Therefore, psy­cho­log­i­cal inter­ven­tions, such as cog­
ni­tive behav­ioral ther­apy, to pro­mote adap­ta­tion and enhance
effec­tive cop­ing have a prom­is­ing poten­tial for improv­ing QOL
and mood in this pop­u­la­tion.33-39
Similar to acute GVHD, screen­ing for psy­cho­so­cial dis­tress
and prompt refer­ral to sup­port­ive care ser­vices such as psy­chol­
ogy, psy­chi­a­try, and/or social work for pro­ac­tive man­age­ment
of depres­sion and anx­i­ety are crit­i­cal to opti­miz­ing the qual­ity
of care for patients with chronic GVHD.31,32 In addi­tion to ther­apy
with a licensed clin­i­cal social worker or psy­chol­o­gist, patients
may ben­e­fit from a phar­ma­co­logic approach for mood man­age­
ment with anti­de­pres­sants or anxi­o­lyt­ics. Financial dis­tress can
apy and other inter­ven­tions directed at organ spe­cific–con­trol
of symp­toms.30,41 The National Institutes of Health Consensus
Development Project and the Brit­ish Society for Bone Marrow
Transplantation have both insti­tuted rec­om­men­da­tions for the
sup­port­ive care and man­age­ment of organ-spe­cific com­pli­ca­
tions of chronic GVHD.42 We review the spe­cific ther­a­pies for
the most com­mon chronic GVHD symp­toms, includ­ing skin, oral,
ocu­lar, and gen­i­tal man­i­fes­ta­tions of chronic GVHD.
Skin chronic GVHD
Supportive care for skin chronic GVHD focuses on pre­ven­tion
and man­age­ment of chronic GVHD symp­toms, includ­ing pruritus,
rash, pain, dyspigmentation, lim­ited range of motion, ero­sions,
ulcer­a­tions, and super­in­fec­tions (Table 1).
Sun pro­tec­tion is impor­tant to reduce the risk of ultra­vi­o­let
radi­a­tion caus­ing an exac­er­ba­tion in skin chronic GVHD symp­
toms as well as risk of sec­ond­ary skin can­cers.43 Photopro­
tection mea­sures include sun avoid­ance, pro­tec­tive cloth­ing,
phys­i­cal sun blocks, and high-potency sun­screens that pro­

Table 1. Supportive care rec­om­men­da­tions for chronic GVHD of the skin

Preventative mea­sures
  Photoprotection (UVA and UVB block­ade)
  Avoidance of sun expo­sure
  Use of sun­screens (≥SPF 20 with broad-spec­trum UVA and UVB pro­tec­tion)
Treatment
  Intact skin
     Symptomatic treat­ments with emol­lients and anti­pru­ritic agents
     Topical cor­ti­co­ste­roids
     Light ther­apy (PUVA, UVAI, UVB, nar­row­band UVB)
     Topical calcineurin inhib­i­tors (pimecrolimus, tacrolimus)
  Erosion and ulcer­a­tions
     Wound dress­ings and debride­ment
    Control of edema
  Sclerotic man­i­fes­ta­tion with joint stiff­ness or con­trac­tures
     Deep mus­cle/fas­cial mas­sage to improve range of motion
     Referral to phys­i­cal ther­apy, occu­pa­tional ther­apy, or phys­i­cal med­i­cine and reha­bil­i­ta­tion
     Daily stretching exer­cises to improve range of motion
     Strengthening, iso­tonic, iso­met­ric, and isokinetic exer­cises
  Other
     Dyspigmentation: trial of depigmenting creams containing hydro­qui­none, top­i­cal tre­tin­oin, or cor­ti­co­ste­roids
     Hair loss: dermovate scalp lotion once or twice daily
     Xerosis: reg­u­lar mois­tur­iz­ers
     Pruritus: tepid water rather than hot water for bath­ing, top­i­cal cor­ti­co­ste­roids, oral anti­his­ta­mines, doxepin, or gabapentin
PUVA, psoralen + ultraviolet light; SPF, sun pro­tec­tion fac­tor; UVA, ultraviolet light therapy.
Prakash Singh Shekhawat
Supportive care for GVHD  |  657
tect against both ultra­vi­o­let A (UVA) and ultra­vi­o­let B (UVB)
radi­a­tion. Routine lubri­ca­tion of dry skin with emol­lients may
decrease pruritus and enhance skin integ­rity.42 Topical ste­roids
and emol­lients can be used to treat nonsclerotic skin lesions
with­out ero­sions or ulcer­a­tions. Mid-strength top­i­cal ste­roids
(eg, tri­am­cin­o­lone 0.1% cream or ointment) can be used to
treat skin areas from the neck down. Higher-potency top­i­cal
ste­roids such as clobetasol 0.5% can be used in skin areas that
are refrac­tory to mid-strength top­i­cal treat­ment. In unre­spon­
sive cases, short-term occlu­sion of mid-strength ste­roids with
damp tow­els (“wet wraps”) increases skin hydra­tion and ste­
roid pen­e­tra­tion.30,41,42 Lower-potency top­i­cal cor­ti­co­ste­roids
such as top­ic ­ al hydro­cor­ti­sone 1.0% to 2.0% are pre­ferred
for long-term use on the face, axil­lae, and groin. Other adju­
vant treat­ments that can be par­tic­u­larly help­ful for pruritus
related to GVHD include top­i­cal hydro­cor­ti­sone/pramoxine or
menthol-based cream/lotions, as well as sys­temic anti­his­ta­
mines or the tri­cy­clic agent doxepin.30,41,42 Other inter­ven­tions
such as top­i­cal calcineurin inhib­i­tors have been reported to
improve ery­thema and pru­ri­tus. For patients with ulcerated
skin lesions, wound dress­ings main­tain a moist envi­ron­ment
that enhances repair of the epi­the­lium. Protective films can
also be use­ ful to pre­ vent break­ down of com­ pro­mised but
nonulcerated skins. Topical anti­ mi­
cro­
bi­als may also have
some util­ity to pre­vent recur­rent infec­tions. For patients with
major wounds related to skin GVHD, mul­ti­dis­ci­plin­ary exper­
tise from der­ ma­ tol­
ogy, plas­tic sur­gery, and wound care is
often needed.
Patients with scle­ro­sis affect­ing the skin and sub­cu­ta­ne­ous
tis­sue, includ­ing fas­cia, joints, and the mus­cu­lo­skel­e­tal sys­tem,
strug­gle with sub­stan­tial func­tional impair­ments depending on
the sever­ity of the scle­ro­sis.30,41,42 Sclerotic fea­tures are one of
the most fre­quent and most dif­fi­cult man­i­fes­ta­tions of chronic
GVHD. Often patients with scle­ rotic skin GVHD require pro­
longed treat­ment with sys­temic ther­a­pies, includ­ing cor­ti­co­ste­
roids, which fur­ther causes mus­cu­lar atro­phy, osteopenia, and
func­tional lim­i­ta­tions. Table 1 high­lights sup­port­ive care inter­
ven­tions to address the needs of patients with scle­rotic chronic
GVHD of the skin.30,41,42
Table 2. Supportive care rec­om­men­da­tions for oral chronic
GVHD
Mild to mod­er­ate muco­sal dis­ease
  Localized appli­ca­tion of high-potency top­i­cal cor­ti­co­ste­roids
  Generalized appli­ca­tion of upper mid-strength top­i­cal cor­ti­co­ste­roid
  Topical anal­ge­sics
Moderate to severe muco­sal dis­ease
  Localized appli­ca­tion of upper mid-strength top­i­cal cor­ti­co­ste­roids
  Topical appli­ca­tion of tacrolimus 0.1% ointment
  Intralesional ther­apy with high-potency ste­roids for refrac­tory lesions
  Topical appli­ca­tion of cyclo­spor­ine rinses
  Oral phototherapy
Salivary gland dis­ease
  Home fluo­ride ther­apy
  Frequent water sip­ping and saliva sub­sti­tutes
  Salivary stim­u­lants (sugar-free gum, sugar-free candy)
  Sialogogues: cevimeline, pilo­car­pine
658  |  Hematology 2021  |  ASH Education Program
Oral GVHD
Oral GVHD involv­ing the mouth and oral mucosa has 3 com­
po­nents: muco­sal involve­ment, sal­i­vary gland involve­ment,
and scle­rotic involve­ment of the mouth and surrounding tis­
sues.30,41,42 Oral chronic GVHD can result in dry­ness, pain, ody­
nophagia, and taste impair­ment. New oral lesions should also
be eval­u­ated for sec­ond­ary can­cers given that they are more
com­mon in allo­ge­neic HCT recip­i­ents. Table 2 sum­ma­rizes man­
age­ment strat­e­gies for oral chronic GVHD.30,41,42 Treatment of
oral cav­ity chronic GHVD is mainly focused on using cor­ti­co­ste­
roid rinses to reduce inflam­ma­tion in the oral cav­ity. Dexameth­
asone or other cor­ti­co­ste­roid rinse for­mu­la­tions are held and
swished in the mouth for approx­i­ma­tely 5 min­utes as often as
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/655/1851925/655jawahri.pdf by guest on 13 December 2021
4 to 6 times per day. For patients with severe symp­toms, top­i­
cal anal­ge­sia with vis­cous lido­caine can also be help­ful. Tacro­
limus rinses are also an alter­na­tive that can be effec­tive.30,41,42
Tacrolimus ointment is espe­cially use­ful for treat­ment of lip
GVHD involve­ment. Intralesional injec­tions with tri­am­cin­o­lone
can also be help­ful for dis­crete oral lesions that fail to respond
to top­i­cal ther­apy.
Patients with chronic GVHD affect­ing the sal­i­vary gland often
report dry mouth and sensitivities to hot, cold, spicy, and acidic
foods. Patients may also develop mucoceles, which often pres­
ent as pain­less blis­ters on the pal­ate.30,41,42 Supportive care mea­
sures for dry mouth include fre­quent liq­uid intake, use of sal­i­vary
stim­u­lants such as sugar-free gum, oral mois­tur­iz­ing agents, and
saliva sub­sti­tutes. Mouth dry­ness can increase the risk of car­ies
and tooth decay, which can be prevented with top­i­cal fluo­ride
use. In patients with severe symp­toms, treat­ment with cho­lin­
er­gic ago­nists such as cevimeline or pilo­car­pine may enhance
sal­i­vary secre­tions.30,41,42
Ocular GVHD
Ocular GVHD may pres­ent with acute con­junc­ti­val inflam­ma­tion,
pseudomembranous con­junc­ti­vi­tis, or ker­a­to­con­junc­ti­vi­tis sicca
syn­drome (or dry eye syn­drome). Keratoconjunctivitis sicca syn­
drome is the most com­mon pre­sen­ta­tion of chronic ocu­lar GVHD
and is diag­nosed by the pres­ence of symp­toms, tear pro­duc­tion
aver­ag­ing ≤5 mm (Schirmer’s test), and clin­i­cal signs of ker­a­ti­
tis.30,41,42 Most treat­ments of ocu­lar GVHD are aimed at reliev­ing
dry eye symp­toms, includ­ing burn­ing, irri­ta­tion, pain, for­eign
body sen­sa­tion, blurred vision, pho­to­pho­bia, and exces­sive tear­
ing. Table 3 sum­ma­rizes the rec­om­men­da­tions for sup­port­ive
care man­age­ment of ocu­lar GVHD.
Preservative-free arti­fi­cial tears are often used in patients
with chronic GVHD to increase lubri­ca­tion and reduce ocu­lar
symp­toms.30,41,42 Oral med­i­ca­tions can also be used to increase
lubri­ca­tion by stim­u­lat­ing aque­ous tear flow with selec­tive
mus­ca­rinic ago­nists such as cevimeline or pilo­car­pine. Man­
agement of ocu­lar GVHD also depends on decreas­ing evap­
o­ra­tion.30,41,42 Warm com­ pres­sors and lid care can max­ i­
mize
meibomian gland out­put that pro­duces the oil layer of the tear
film, which pro­tects against evap­o­ra­tion. Certain pro­tec­tive
eye wear such as mois­ture cham­ber googles may also help
decrease evap­o­ra­tion.30,41,42 Doxycycline is often pre­scribed for
patients with ocu­lar GVHD to treat rosacea blepharitis, thereby
reduc­ing the inflam­ma­tion of the lid. Severe cases may also
ben­e­fit from scleral lenses, but these are avail­­able only in a few
spe­cial­ized cen­ters. To decrease drain­age from the ocu­lar sur­
faces, tem­po­rary and per­ma­nent occlu­sions of the tear ducts
Table 3. Supportive care rec­om­men­da­tions for ocu­lar chronic
GVHD
Topical and oral ther­a­pies
  Artificial tears, pre­ser­va­tive free
  Viscous ointment at bed­time/vis­cous tears dur­ing the day
  Cyclosporine and top­i­cal ste­roid eye drops
  Oral agents such as cevimeline and pilo­car­pine
  Doxycycline to reduce inflam­ma­tion
Surgical
  Punctal occlu­sion (tem­po­rary using sil­i­cone plugs)
  Permanent occlu­sion using ther­mal cau­tery
  Superficial debride­ment of fil­a­men­tary ker­a­ti­tis
  Partial tarsorrhaphy
Eye wear/envi­ron­men­tal strat­egy
  Occlusive eye wear (mois­ture cham­ber gog­gles)
  Lid care/warm com­pres­sors/humid­i­fied envi­ron­ment
  Bandage con­tact lens
Treatments not widely avail­­able
  Autologous serum eye drops
  Gas-per­me­able con­tact lens (scleral lens pros­the­sis)
Table 4. Supportive care rec­om­men­da­tions for vul­var
and vag­i­nal chronic GVHD
Vulvar dis­com­fort
  Avoid mechan­i­cal and chemical irri­tants
  Cleanse gen­i­tal area with warm water, allow air cir­cu­la­tion, and
wipe front to back
  Sparing use of sim­ple emol­lients to vulva
  Water-based lubri­cants
Vulvovaginal symp­toms due to low estro­gen sta­tus
  Topical estro­gen with or with­out dila­tor ther­apy
Topical ther­apy for vulvovaginal GVHD
  High- and ultra-high-potency cor­ti­co­ste­roids
  ○ Clobetasol gel 0.05% (vagina)
  ○ Betamethasone dipropionate augmented gel (vagina) or ointment
(vulva)
  ○ Tacrolimus ointment 0.1% (vulva)
Surgical ther­apy
  Surgery for stric­tures
may help patients with mod­er­ate to severe ocu­lar dis­ease.30,41,42
In addi­tion to these mea­ sures, patients with ocu­ lar sur­face
inflam­ma­tion may ben­e­fit from the judi­cious use of top­i­cal ste­
roid ther­apy to reduce inflam­ma­tion. Topical cyclo­spor­ine is
also com­monly used. Autologous tears can also reduce ocu­lar
Table 5. Supportive care rec­om­men­da­tions for pul­mo­nary chronic GVHD
   Pulmonary func­tion test and high-res­o­lu­tion expi­ra­tory phase chest com­put­er­ized tomog­ra­phy to assess for chronic GVHD of the lung
   Routine mon­i­tor­ing of pul­mo­nary func­tion test in high-risk pop­u­la­tion is warranted
   Systemic ther­apy with cor­ti­co­ste­roid and other chronic GVHD agents
   Macrolides, inhaled ste­roids, and leu­ko­tri­ene inhib­i­tors are help­ful as an adjunc­tive ther­apy (ie, FAM ther­apy)
   Consideration of use of intra­ve­nous immu­no­glob­u­lins, par­tic­u­larly those with low IgG lev­els
   Adequate vac­ci­na­tions against pneu­mo­coc­cus and sea­sonal influ­enza
   Referral to pulmonology and con­sid­er­ation for pul­mo­nary reha­bil­i­ta­tion
FAM, fluticasone, azithromycin, and montelukast.
Prakash Singh Shekhawat
sur­face inflam­ma­tion, but these are often avail­­able only in spe­
cial­ized cen­ters.30,41,42
Vulvar and vag­i­nal GVHD
Vulvar and vag­i­nal chronic GVHD pres­ents with muco­sal abnor­
mal­i­ties that can sub­se­quently evolve into scle­rotic changes.30,41,42
Patients with vulvovaginal chronic GVHD pres­ent with symp­toms
of vag­i­nal dry­ness, dyspareunia, and dys­uria. If left untreated, vul­
vovaginal chronic GVHD can result in scle­ro­sis of the vul­var and
vag­i­nal tis­sues, resulting in changes such as agglu­ti­na­tion of the
cli­to­ral hood, narrowing of the introitus, and short­en­ing of the
vag­i­nal canal. Table 4 sum­ma­rizes the sup­port­ive care and man­
age­ment strat­e­gies for patients with vulvovaginal GVHD.30,41,42
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/655/1851925/655jawahri.pdf by guest on 13 December 2021
Itching and irri­ta­tion can be relieved by the appli­ca­tion of
emol­lients to the exter­nal gen­i­ta­lia. Water-based lubri­cants
can also be used in the vagina to reduce itching and irri­ta­tion.
Patients with vulvovaginal GVHD often also have low estra­diol
lev­els and vag­i­nal atro­phy.30,41,42 Topical estro­gen ther­apy with
or with­out the use of vag­i­nal dila­tors can be extremely effec­tive
in the absence of abso­lute con­tra­in­di­ca­tions. For treat­ment of
vulvovaginal GVHD symp­toms, top­i­cal treat­ment with ultra-high-
potency cor­ti­co­ste­roid treat­ment rep­re­sents the main ther­apy,
although top­i­cal calcineurin ointments can be used in this pop­
u­la­tion as well.30,41,42
Late effects and infec­tion pro­phy­laxis and vac­ci­na­tions
for patients with chronic GVHD
Patients with chronic GVHD are at high risk of late effects, includ­­
ing skel­e­tal com­pli­ca­tions, sec­ond­ary can­cers, car­dio­vas­cu­lar
dis­ease, and throm­bo­em­bolic events.44,45 Due to both chronic
GVHD and its treat­ ment, patients often develop met­ a­
bolic
com­pli­ca­tions, includ­ing hyper­ten­sion, hyper­lip­id­emia, dia­be­
tes, and met­a­bolic syn­drome.44,45 These, in turn, increase the
risk of car­dio­vas­cu­lar events. Careful atten­tion to car­dio­vas­cu­lar
risk fac­tors and these com­pli­ca­tions is nec­es­sary in pro­vid­ing
high-qual­ity sur­vi­vor­ship care for this pop­u­la­tion.44,45 Patients
with chronic GVHD are highly immu­no­com­pro­mised with def­
i­cits in mac­ro­phage func­tion, immu­no­glob­u­lin pro­duc­tion, and
T-cell func­tion.30,41,42 A review of infec­tion pro­phy­laxis and vac­
ci­na­tion strat­e­gies for patients with chronic GVHD is beyond
the scope of this chap­ter, but these issues are well outlined in
com­pre­hen­sive guide­lines for the pre­ven­tion of oppor­tu­nis­
tic infec­tions fol­low­ing HCT published by the Centers for Dis­
ease Control and Prevention, the Infectious Diseases Society
of America, and the Amer­i­can Society for Blood and Marrow
Transplantation.46 Table 5 out­lines strat­e­gies for mon­i­tor­ing and
man­age­ment of other chronic GVHD com­pli­ca­tions, includ­ing
pul­mo­nary GVHD.
Supportive care for GVHD  |  659

CLINICAL CASE (Con­t in­u ed)
Despite her grat­i­tude that she is still alive and ­able to spend
qual­ity time with her chil­dren, Emma often mourns the immense
losses she has expe­ri­enced as a result of her ill­ness and its impact
on her day-to-day life. She also strug­gles with the uncer­tainty
regard­ing her future health and prog­no­sis. She is grate­ful for her
trans­plant cli­ni­cians who often val­i­date her emo­tional strug­gles,
allow her the space to dis­cuss her emo­tional jour­ney with this ill­
ness, and pro­vide sup­port­ive resources, includ­ing psy­cho­so­cial
coun­sel­ing to pro­cess this dif­fi­cult ill­ness.
Summary
Living with GVHD affects mul­ti­ple domains of patient-reported
QOL, phys­i­cal func­tion­ing, and psy­cho­log­i­cal well-being. Patients
describe liv­ing with GVHD as a life-alter­ing “full-time job” requir­
ing unique knowl­edge, per­sonal growth, and resil­ient cop­ing
strat­eg
­ ies. Management of GVHD must include opti­mal sup­port­
ive care mea­sures to address GVHD symp­toms, pro­mote effec­
tive cop­ing, and reduce psy­cho­log­i­cal and exis­ten­tial dis­tress
in a pop­u­la­tion liv­ing with immense prog­nos­tic uncer­tainty and
strug­gling to adapt to a dif­fi­cult and unpre­dict­able ill­ness.
Conflict-of-inter­est dis­clo­sure
Areej El-Jawahri: no com­pet­ing finan­cial inter­ests to declare.
Off-label drug use
Areej El-Jawahri: nothing to disclose.
Correspondence
Areej El-Jawahri, Hematology-Oncology, Mas­sa­chu­setts General
Hospital, 55 Fruit St, Yawkey 9E, Bos­ton, MA 02114; e-mail:ael-
jawahri@part­ners​­.org.
References
1. Holtan SG, Pasquini M, Weisdorf DJ. Acute graft-ver­sus-host dis­ease: a bench-
to-bed­side update. Blood. 2014;124(3):363-373.
2. Choi S, Reddy P. Graft-ver­sus-host dis­ease. Panminerva Med. 2010;52(2):
111-124.
3. MacMillan ML, Robin M, Harris AC, et al. A refined risk score for acute graft-
ver­sus-host dis­ease that pre­dicts response to ini­tial ther­apy, sur­vival, and
trans­plant-related mor­tal­ity. Biol Blood Marrow Transplant. 2015;21(4):761-
767.
4. Levine JE, Braun TM, Harris AC, et al. A prog­nos­tic score for acute graft-
ver­sus-host dis­ ease based on bio­ mark­ ers: a multicentre study. Lancet
Haematol. 2015;2(1):e21-e29.
5. Major-Monfried H, Renteria AS, Pawarode A, et al. MAGIC bio­mark­ers pre­
dict long-term out­comes for ste­roid-resis­tant acute GVHD. Blood. 2018;
131(25):2846-2855.
6. Patel SS, Lapin B, Majhail NS, Hamilton BK. Patient-reported out­comes in
acute graft-ver­sus-host dis­ease: opti­miz­ing patient care and clin­i­cal trial
end­points. Bone Marrow Transplant. 2020;55(8):1533-1539.
7. Kurosawa S, Yamaguchi T, Mori T, et al. Patient-reported qual­ity of life after
allo­ge­neic hema­to­poi­etic cell trans­plan­ta­tion or che­mo­ther­apy for acute
leu­ke­mia. Bone Marrow Transplant. 2015;50(9):1241-1249.
8. Lee SJ, Kim HT, Ho VT, et al. Quality of life asso­ci­ated with acute and chronic
graft-ver­sus-host dis­ease. Bone Marrow Transplant. 2006;38(4):305-310.
9. Syrjala KL, Chapko MK, Vitaliano PP, Cummings C, Sullivan KM. Recovery
after allo­ge­neic mar­row trans­plan­ta­tion: pro­spec­tive study of pre­dic­tors
of long-term phys­i­cal and psy­cho­so­cial func­tion­ing. Bone Marrow Trans-
plant. 1993;11(4):319-327.
660  |  Hematology 2021  |  ASH Education Program
10. Chiodi S, Spinelli S, Ravera G, et al. Quality of life in 244 recip­i­ents of allo­
ge­neic bone mar­row trans­plan­ta­tion. Br J Haematol. 2000;110(3):614-619.
11. Worel N, Biener D, Kalhs P, et  al. Long-term out­come and qual­ity of life
of patients who are alive and in com­plete remis­sion more than two years
after allo­ge­neic and syn­ge­neic stem cell trans­plan­ta­tion. Bone Marrow
Transplant. 2002;30(9):619-626.
12. Shaw BE, Brazauskas R, Millard HR, et al. Centralized patient-reported out­
come data col­lec­tion in trans­plan­ta­tion is fea­si­ble and clin­i­cally mean­ing­
ful. Cancer. 2017;123(23):4687-4700.
13. Lee SJ, Loberiza FR, Antin JH, et  al. Routine screen­ing for psy­cho­so­cial
dis­tress fol­low­ing hema­to­poi­etic stem cell trans­plan­ta­tion. Bone Marrow
Transplant. 2005;35(1):77-83.
14. Ngo-Huang A, Yadav R, Bansal S, et al. An explor­atory study on phys­i­cal func­
tion in stem cell trans­plant patients under­go­ing cor­ti­co­ste­roid treat­ment
for acute graft-ver­sus-host-dis­ease. Am J Phys Med Rehabil. 2021;100(4):
402-406.
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/655/1851925/655jawahri.pdf by guest on 13 December 2021
15. Hamada R, Kondo T, Murao M, et al. Effect of the sever­ity of acute graft-ver­
sus-host dis­ease on phys­i­cal func­tion after allo­ge­neic hema­to­poi­etic stem
cell trans­plan­ta­tion. Support Care Cancer. 2020;28(7):3189-3196.
16. Schäcke H, Döcke WD, Asadullah K. Mechanisms involved in the side
effects of glu­co­cor­ti­coids. Pharmacol Ther. 2002;96(1):23-43.
17. Lee HJ, Oran B, Saliba RM, et al. Steroid myop­a­thy in patients with acute
graft-ver­sus-host dis­ease treated with high-dose ste­roid ther­apy. Bone
Marrow Transplant. 2006;38(4):299-303.
18. Yoshioka K, Kakihana K, Doki N, Ohashi K. Gut microbiota and acute
graft-ver­sus-host dis­ease. Pharmacol Res. 2017;122:90-95.
19. El-Jawahri A, LeBlanc T, VanDusen H, et al. Effect of inpa­tient pal­li­a­tive care
on qual­ity of life 2 weeks after hema­to­poi­etic stem cell trans­plan­ta­tion: a
ran­dom­ized clin­i­cal trial. JAMA. 2016;316(20):2094-2103.
20. El-Jawahri A, Traeger L, Greer JA, et al. Effect of inpa­tient pal­li­a­tive care dur­
ing hema­to­poi­etic stem-cell trans­plant on psy­cho­log­i­cal dis­tress 6 months
after trans­plant: results of a ran­dom­ized clin­i­cal trial. J Clin Oncol. 2017;
35(32):3714-3721.
21. Duell T, van Lint MT, Ljungman P, et al. Health and func­tional sta­tus of long-
term sur­vi­vors of bone mar­row trans­plan­ta­tion. EBMT Working Party on
Late Effects and EULEP Study Group on Late Effects. Euro­pean Group for
Blood and Marrow Transplantation. Ann Intern Med. 1997;126(3):184-192.
22. Socié G, Stone JV, Wingard JR, et al. Long-term sur­vival and late deaths
after allo­ge­neic bone mar­row trans­plan­ta­tion. Late Effects Working Com­
mittee of the International Bone Marrow Transplant Registry. N Engl J Med.
1999;341(1):14-21.
23. Sullivan KM, Witherspoon RP, Storb R, et al. Prednisone and aza­thi­op ­ rine
com­pared with pred­ni­sone and pla­cebo for treat­ment of chronic graft-v-
host dis­ease: prog­nos­tic influ­ence of prolonged throm­bo­cy­to­pe­nia after
allo­ge­neic mar­row trans­plan­ta­tion. Blood. 1988;72(2):546-554.
24. Wingard JR, Piantadosi S, Vogelsang GB, et  al. Predictors of death from
chronic graft-ver­sus-host dis­ease after bone mar­row trans­plan­ta­tion.
Blood. 1989;74(4):1428-1435.
25. Loughran TP Jr, Sullivan K, Morton T, et al. Value of day 100 screen­ing stud­
ies for predicting the devel­op­ment of chronic graft-ver­sus-host dis­ease
after allo­ge­neic bone mar­row trans­plan­ta­tion. Blood. 1990;76(1):228-234.
26. Sutherland HJ, Fyles GM, Adams G, et  al. Quality of life fol­low­ing bone
mar­row trans­plan­ta­tion: a com­par­i­son of patient reports with pop­u­la­tion
norms. Bone Marrow Transplant. 1997;19(11):1129-1136.
27. Lee SJ, Flowers MED. Recognizing and man­ag­ing chronic graft-ver­sus-host
dis­ease. Hematology. 2008;2008(1):134-141.
28. Pidala J, Kurland B, Chai X, et  al. Patient-reported qual­ity of life is asso­
ci­ated with sever­ity of chronic graft-ver­sus-host dis­ease as mea­sured by
NIH cri­te­ria: report on base­line data from the Chronic GVHD Consortium.
Blood. 2011;117(17):4651-4657.
29. Lee Sk, Cook EF, Soiffer R, Antin JH. Development and val­i­da­tion of a scale
to mea­sure symp­toms of chronic graft-ver­sus-host dis­ease. Biol Blood Mar-
row Transplant. 2002;8(8):444-452.
30. Dignan FL, Scarisbrick JJ, Cor­nish J, et al; Haemato-oncol­ogy Task Force of
Brit­ish Committee for Standards in Haematology; Brit­ish Society for Blood
and Marrow Transplantation. Organ-spe­cific man­age­ment and sup­port­ive
care in chronic graft-ver­sus-host dis­ease. Br J Haematol. 2012;158(1):62-78.
31. El-Jawahri A, Pidala J, Khera N, et al. Impact of psy­cho­log­i­cal dis­tress on
qual­ity of life, func­tional sta­tus, and sur­vival in patients with chronic graft-
ver­sus-host dis­ease. Biol Blood Marrow Transplant. 2018;24(11):2285-2292.
32. Jacobs JM, Fishman S, Sommer R, et al. Coping and mod­i­fi­able psy­cho­so­
cial fac­tors are asso­ci­ated with mood and qual­ity of life in patients with
 chronic graft-ver­sus-host dis­ease. Biol Blood Marrow Transplant. 2019;
25(11):2234-2242.
33. Traeger L, Greer JA, Fernandez-Robles C, Temel JS, Pirl WF. Evidence-based
treat­ment of anx­i­ety in patients with can­cer. J Clin Oncol. 2012;30(11):1197-
1205.
34. Duncan M, Moschopoulou E, Herrington E, et al; SURECAN Investigators.
Review of sys­tem­atic reviews of non-phar­ma­co­log­i­cal inter­ven­tions to
improve qual­ity of life in can­cer sur­vi­vors. BMJ Open. 2017;7(11):e015860.
35. DuHamel KN, Mosher CE, Winkel G, et al. Randomized clin­i­cal trial of tele­
phone-admin­is­tered cog­ni­tive-behav­ioral ther­apy to reduce post-trau­
matic stress dis­or­der and dis­tress symp­toms after hema­to­poi­etic stem-cell
trans­plan­ta­tion. J Clin Oncol. 2010;28(23):3754-3761.
36. Stanton AL, Luecken LJ, MacKinnon DP, Thompson EH. Mechanisms in psy­
cho­so­cial inter­ven­tions for adults liv­ing with can­cer: oppor­tu­nity for inte­
gra­tion of the­ory, research, and prac­tice. J Consult Clin Psychol. 2013;81(2):
318-335.
37. Bower JE. Cancer-related fatigue—mech­ a­
nisms, risk fac­ tors, and treat­
ments. Nat Rev Clin Oncol. 2014;11(10):597-609.
38. Gudenkauf LM, Antoni MH, Stagl JM, et al. Brief cog­ni­tive-behav­ioral and
relax­a­tion train­ing inter­ven­tions for breast can­cer: a ran­dom­ized con­
trolled trial. J Consult Clin Psychol. 2015;83(4):677-688.
39. Duijts SF, van Beurden M, Oldenburg HS, et al. Efficacy of cog­ni­tive behav­
ioral ther­apy and phys­i­cal exer­cise in alle­vi­at­ing treat­ment-induced men­o­
pausal symp­toms in patients with breast can­cer: results of a ran­dom­ized,
con­trolled, mul­ti­cen­ter trial. J Clin Oncol. 2012;30(33):4124-4133.
Prakash Singh Shekhawat
40. Abel GA, Albelda R, Khera N, et al. Financial hard­ship and patient-reported
out­comes after hema­to­poi­etic cell trans­plan­ta­tion. Biol Blood Marrow
Transplant. 2016;22(8):1504-1510.
41. Couriel DR. Ancillary and sup­port­ive care in chronic graft-ver­sus-host dis­
ease. Best Pract Res Clin Haematol. 2008;21(2):291-307.
42. Carpenter PA, Kitko CL, Elad S, et  al. National Institutes of Health Con­
sensus Development Project on Criteria for Clinical Trials in Chronic Graft-
ver­sus-Host Disease: V. The 2014 ancil­lary ther­apy and sup­port­ive care
work­ing group report. Biol Blood Marrow Transplant. 2015;21(7):1167-1187.
43. Kitajima T, Imamura S. Graft-ver­sus-host reac­tion enhanced by ultra­vi­o­let
radi­a­tion. Arch Dermatol Res. 1993;285(8):499-501.
44. Carpenter PA. Late effects of chronic graft-ver­sus-host dis­ease. Best Pract
Res Clin Haematol. 2008;21(2):309-331.
45. Majhail NS. Long-term com­pli­ca­tions after hema­to­poi­etic cell trans­plan­ta­
tion. Hematol Oncol Stem Cell Ther. 2017;10(4):220-227.
46. Centers for Disease Control and Prevention; Infectious Disease Society of
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/655/1851925/655jawahri.pdf by guest on 13 December 2021
America; Amer­i­can Society of Blood and Marrow Transplantation. Guide­
lines for pre­vent­ing oppor­tu­nis­tic infec­tions among hema­to­poi­etic stem
cell trans­plant recip­i­ents. MMWR Recomm Rep. 2000;49(RR-10):1-125, CE1-7.
© 2021 by The Amer­i­can Society of Hematology
DOI 10.1182/hema­tol­ogy.2021000302
Supportive care for GVHD  |  661
 WHAT ’ S NEW IN PLASMA CELL DISORDERS ?

Advances in MGUS diagnosis, risk stratification,

and management: introducing myeloma-defining
genomic events

Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/662/1851672/662landgren.pdf by guest on 13 December 2021

Ola Landgren
Myeloma Program, Sylvester Comprehensive Cancer Center, University of Miami, Miami, FL

In the 1960s, Dr Jan Waldenström argued that patients who had monoclonal proteins without any symptoms or evidence
of end-organ damage represented a benign monoclonal gammopathy. In 1978, Dr Robert Kyle introduced the concept of
“monoclonal gammopathy of undetermined significance” (MGUS) given that, at diagnosis, it was not possible with avail-
able methods (ie, serum protein electrophoresis to define the concentration of M-proteins and microscopy to determine
the plasma cell percentage in bone marrow aspirates) to determine which patients would ultimately progress to multiple
myeloma. The application of low-input whole-genome sequencing (WGS) technology has circumvented previous prob-
lems related to volume of clonal plasma cells and contamination by normal plasma cells and allowed for the interrogation
of the WGS landscape of MGUS. As discussed in this chapter, the distribution of genetic events reveals striking differences
and the existence of 2 biologically and clinically distinct entities of asymptomatic monoclonal gammopathies. Thus,
we already have genomic tools to identify “myeloma-defining genomic events,” and consequently, it is reasonable to
consider updating our preferred terminologies. When the clinical field is ready to move forward, we should be able to
consolidate current terminologies—from current 7 clinical categories: low-risk MGUS, intermediate-risk MGUS, high-risk
MGUS, low-risk smoldering myeloma, intermediate-risk smoldering myeloma, high-risk smoldering myeloma, and multi-
ple myeloma—to future 3 genomic-based categories: monoclonal gammopathy, early detection of multiple myeloma (in
which myeloma-defining genomic events already have been acquired), and multiple myeloma (patients who are already
progressing and clinically defined cases). Ongoing investigations will continue to advance the field.

LEARNING OBJECTIVES
• Review of current clinical risk scores for myeloma precursor conditions, which are limited indirect measures of
disease burden/activity
• Show how low-input WGS reveals striking differences and existence of 2 biologically and clinically distinct entities

CLINICAL CASE confrmed normal complete blood count and compre-

In the beginning of January 2016, a 46-year-old previously
healthy male lawyer was applying for a new life insurance
policy. As part of the application process, the insurance
company required a standard health questionnaire and
laboratory blood tests. The results from the blood work
showed a normal complete blood count and a normal
comprehensive metabolic panel with the exception of an
elevated total protein of 8.9 g/dL (normal reference range,
6.0–8.3 g/dL). The patient was referred to a hematologist,
who ordered repeat laboratory blood tests, including
additional serum immune assays, to further investigate
the elevated total protein levels. The repeated results
hensive metabolic panel, whereas a serum protein elec-
trophoresis (SPEP) revealed evidence of a monoclonal
band in the gamma region and immunofxation showed
immunoglobulin G (IgG) κ isotype. The concentration of
the monoclonal (M)–protein was determined to 0.9 g/dL.
Quantitative immunoglobulins showed normal IgG, immu-
noglobulin A (IgA), and immunoglobulin M (IgM) levels.
Also, serum free light chains (FLCs) were evaluated,
and they showed normal levels. Given the low concen-
tration of the M-protein, the IgG isotype, the absence
of an abnormal FLC ratio, and normal IgA (150 mg/dL)
or IgM (173 mg/dL) concentrations, per current clinical

662 | Hematology 2021 | ASH Education Program

­guide­lines,1 the ­hema­tol­o­gist recommended no bone mar­row
biopsy and no imag­ing. Instead, the patient was recommended
to repeat the bloodwork in 6 months. The patient was given
the diag­no­sis of mono­clo­nal gammopathy of unde­ter­mined
sig­nif­i­cance (MGUS), and he was told that the life­time risk of
pro­gres­sion to mul­ti­ple mye­loma (MM) was very low.
When the patient returned for a fol­low-up visit in July 2016, all­
the results were vir­tu­ally unchanged. Per cur­rent clin­i­cal guide­
lines,1 the hema­tol­o­gist now recommended fol­low-up annu­ally
with repeated bloodwork. When the patient returned in July
2017, the IgG κ M-pro­tein had increased to 1.1 g/dL, whereas all­
other results were sim­i­lar to the year before. The patient had
no symp­toms. In July 2018, he returned for a fol­low-up visit, and
the bloodwork now revealed an IgG κ M-pro­tein of 1.4 g/dL.
The FLC κ lev­els were slightly ele­vated at 5.3 mg/dL (nor­mal
ref­er­ence range, 0.33-1.94 mg/dL), FLC λ was 0.8 mg/dL (nor­
mal ref­er­ence range, 0.57-2.63 mg/dL), and the FLC ratio was
6.63 (nor­mal ref­er­ence range, 0.26-1.65). Also, the IgM con­cen­
tra­tion was decreased at 25 mg/dL (nor­mal ref­er­ence range,
37-286 mg/dL), whereas IgA was within the nor­mal ref­er­ence
inter­val at 78 mg/dL (61-356 mg/dL). Given the grad­ual wors­
en­ing of serum immune mark­ers, the hema­tol­o­gist discussed
with the patient that he may choose to undergo a bone mar­
row biopsy and imag­ing to rule out MM, but the patient asked
his doc­tor if it was totally nec­es­sary since he felt com­pletely
healthy, ate healthy food, and exer­cised sev­eral days every
week. After a lon­ger dis­cus­sion, the patient and the hema­tol­
o­gist decided to hold off with addi­tional test­ing and con­tinue
with annual lab­o­ra­tory results.
In June 2019, about 3.5 years after ini­tial diag­no­sis of MGUS,
the patient came for a visit to review his annual bloodwork.
Now the SPEP showed an IgG κ M-pro­tein of 1.6 g/dL. The FLC κ
lev­els were again ele­vated now at 8.6 mg/dL (nor­mal ref­er­ence
range, 0.33-1.94 mg/dL), FLC λ was 0.5 mg/dL (nor­mal ref­er­ence
range, 0.57-2.63 mg/dL), and the FLC ratio was 17.2 (nor­mal ref­
er­ence range, 0.26-1.65). The IgM con­cen­tra­tion was fur­ther
decreased at 18 mg/dL (nor­mal ref­er­ence range, 37-286 mg/dL),
and IgA was decreased at 40 mg/dL (61-356 mg/dL). Hemoglo-
bin was slightly decreased at 13.2 g/dL (nor­mal ref­er­ence range,
13.5-17.5  g/dL), whereas all­ other lab­o­ra­tory results (­includ­ing
cal­cium, cre­at­i­nine, albu­min, lac­tate ­dehy­dro­ge­nase, and β2-
microglobulin) were nor­mal. The hema­tol­o­gist recommended
the patient to undergo a bone mar­row biopsy and a pos­i­tron
emis­sion tomog­ra­phy/com­puted tomog­ra­phy to rule out MM.
Immunohistochemistry staining of the core biopsy spec­i­men
showed 30% κ light chain–restricted plasma cells in the bone
mar­row. Also, whole-body pos­i­t ron emis­sion tomog­ra­phy/
com­puted tomog­ra­phy revealed a 1.1-cm diam­e­ter lytic lesion
in ileum on the right side of the pel­vis with an standardized
uptake value (SUV) of 6.7, a 1.2-cm diam­e­ter lytic lesion in the
left femur with a SUV of 8.2, and 2 small (<5 mm diam­e­ter)
lytic lesions in the left fifth and sixth ribs with SUVs of 2.5 and
3.1, respec­tively. Fluorescence in situ hybrid­iza­tion and sin­gle-
nucle­o­tide poly­mor­phism array test­ing of the bone mar­row
aspi­rate did not cap­ture any high-risk char­ac­ter­is­tics. Ten-
color flow cytom­e­try of the bone mar­row aspi­rate con­firmed
the immunophenotype of κ light chain–restricted plasma
cells expressing CD56 and CD117 but neg­a­tive for CD20. The
patient was diag­ nosed with stan­ dard-risk MM and started
com­bi­na­tion ther­apy 2 weeks later, after he had under­gone
Prakash Singh Shekhawat
rou­tine workup with a clin­i­cal exam and base­line echo­car­dio­
gram and elec­tro­car­dio­gram.
Because most MGUS cases will never prog­ress, sta­tis­ti­cally
speak­ing, this patient case illus­trates a rel­a­tively uncom­mon sit­u­
a­tion with pro­gres­sion to MM within a few years from ini­tial MGUS
diag­no­sis. However, most phy­si­cians who mon­i­tor large num­bers
of patients with MGUS have seen these kinds of cases in their
clinic. This is reflec­tive of the fol­low­ing 2 facts: (1) most cases
with MGUS remain sta­ble over time, but at the same time, (2) all­
MM cases are con­sis­tently pre­ceded by MGUS (but the major­ity
do not know they had MGUS prior to MM because no test­ing was
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/662/1851672/662landgren.pdf by guest on 13 December 2021
done).2 When it comes to cur­rent clin­i­cal risk scores to pre­dict
pro­gres­sion from MGUS to MM, it is impor­tant to empha­size that
today’s risk scores only pro­vide the aver­age risk of pro­gres­sion
for all­indi­vid­u­als with a given score. None of the avail­­able risk
scores pro­vide the abso­lute risk of pro­gres­sion for an indi­vid­
ual patient. Furthermore, cur­rent clin­i­cal risk scores are based
on tumor bur­den, and they are unable to detect progressors
among MGUS cases with lower dis­ease bur­den. Indeed, cur­rent
clin­i­cal risk scores show that indi­vid­u­als with higher dis­ease bur­
den (higher plasma cell per­cent­age and/or higher serum immune
marker con­cen­tra­tions) on aver­age have a higher prob­a­bil­ity of
progressing com­pared with the aver­age indi­vid­ual with lower
dis­ease bur­den. Although lower dis­ease bur­den has lower risk of
pro­gres­sion, just like the above patient case, there are indi­vid­u­als
with lower dis­ease bur­den progressing to MM. Conversely, many
indi­vid­u­als with higher dis­ease bur­den will never prog­ress.
There is an unmet clin­i­cal need for bet­ter assays allowing
the phy­si­cian to deter­mine the indi­vid­ual patient’s risk of pro­
gres­sion to MM. We do not yet have any such established assays
avail­­able in the clinic. The best tools we have today include lon­
gi­tu­di­nal mon­i­tor­ing and reassessments, as discussed in detail in
this chap­ter. This review addresses cur­rent sta­tus of sci­ence and
clin­i­cal man­age­ment, novel and emerg­ing tech­nol­o­gies, recent
dis­cov­er­ies, and future direc­tions.
Initial obser­va­tions and emer­gence of dif­fer­ent
schools of thought
Early dis­cov­ery work focus­ing on abnor­mal serum pro­teins was
pioneered by Dr Jan Waldenström and col­leagues. In 1944, he
presented his paper “Incipient Myelomatosis or ‘Essential’ Hyper-
globulinemia With Fibrinogenopenia—A New Syndrome?” in
which he described 3 patients with refrac­tory ane­mia and bleed­
ing ten­dency whose sera exhibited a very high vis­cos­ity. Walden-
ström spec­u­lated that a spe­cial glob­u­lin frac­tion was the cause
of the increased vis­cos­ity, and by ultra­cen­tri­fu­ga­tion stud­ies,
he was ­able to dem­on­strate that the sera of 2 of these patients
contained mac­ro­glob­u­lins (ie, plasma glob­u­lins of high molec­u­
lar weight)—sub­se­quently referred to as “Waldenström’s mac­ro­
glob­u­li­ne­mia.”3 Together with Dr Carl-Bertil Laurell—a world-class
clin­i­cal chem­ist—Waldenström started explor­ing con­di­tions with
gammaglobulin derange­ments in a sys­tem­atic man­ner, as well as
clin­i­cal cor­re­lates of mono­clo­nal and poly­clonal gammopathies.4
Through their trans­la­tional research efforts, they delin­eated the
occur­rence and clin­i­cal sig­nif­i­cance of so-called mono­clo­nal
com­po­nents. Gradually, obser­va­tions from the lab­o­ra­tory cou­
pled with clin­i­cal data led to the emer­gence of 2 major schools
of thought. In the 1960s, Waldenström pro­posed that there were
Myeloma pre­cur­sor con­di­tions  |  663
patients who had mono­clo­nal (M)–pro­teins with­out any symp­
toms or evi­dence of end-organ dam­age, representing a benign
mono­clo­nal gammopathy (MG).5-8 Waldenström was of the firm
belief that benign MG was unre­lated to MM. Conversely, the alter­
nate opin­ion was that some patients with asymp­tom­atic mono­
clo­nal pro­teins nev­er­the­less progressed over time to MM and
that it was impor­tant to not term the pro­cess entirely benign.
In 1978, Dr Robert Kyle published his obser­va­tions from a ret­ro­
spec­tive chart review of all­indi­vid­u­als (N = 241) diag­nosed with a
MG at the Mayo Clinic prior to Jan­u­ary 1, 1971. In brief, among the
241 cases, he found that after a 5-year fol­low-up period, (1) the
M-pro­tein remained sta­ble in 137 (57%) patients; (2) the M-pro­tein
increased by 50% or more in 22 (9%) patients; (3) onset of MM,
Waldenström mac­ro­glob­u­li­ne­mia, or amy­loid light chain amy­
loid­osis occurred in 27 (11%) patients; and (4) 55 (23%) patients
died with­out 5-year fol­low-up serum.6 When he com­pared the
mean M-pro­tein con­cen­tra­tion and per­cent bone mar­row plasma
cell infil­tra­tion at base­line for the 4 groups, there were no dif­fer­
ences (mean con­cen­tra­tions were 1.6-1.8 g/dL and mean plasma
cell infil­tra­tion in bone mar­row aspi­rates was 3%-4% across the
4 groups).6 Based on these small num­bers, Kyle6 observed that
among indi­vid­u­als with an M-pro­tein and who later devel­oped
MM, the size of the mono­clo­nal peak increased along with symp­
toms prior to onset of MM. Therefore, Kyle argued in his sem­i­nal
paper from 1978 that “mono­clo­nal gammopathy of u ­ nde­ter­mined
sig­nif­i­cance” (MGUS) is a bet­ter term because one can­not tell
whether the mono­clo­nal pro­tein will remain unchanged or
whether the patient has an evolv­ ing MM. The word unde­ter­
mined was used to reflect that, at diag­no­sis, it was not pos­si­ble
with avail­­able meth­ods (ie, SPEP to define the con­cen­tra­tion of
M-pro­teins and micros­copy to deter­mine the plasma cell per­
cent­age in bone mar­row aspi­rates) to deter­mine which patients
would ulti­mately prog­ress to MM.
Diagnostic cri­te­ria and types of abnor­mal serum pro­teins
The cur­rent def­i­ni­tion of MGUS is char­ac­ter­ized by the pres­ence
of M-pro­teins or an abnor­mal FLC ratio in periph­eral blood.1 For
an indi­vid­ual to be diag­nosed with MGUS, per cur­rent def­i­ni­
tions, the con­cen­tra­tion of the mono­clo­nal spike (M-spike) has
to be less than 3 g/dL, and for an indi­vid­ual to be diag­nosed
with light chain MGUS, the FLC ratio has to be abnor­mal (nor­
mal ref­er­ence for κ/λ FLC ratio, 0.26-1.65), but the involved/
unin­volved ratio has to be less than 100.9 It should be empha­
sized that ele­vated FLC con­cen­tra­tions are not unique to plasma
cell dis­or­ders, and a clin­i­cal inter­pre­ta­tion of the results is always
required. For exam­ple, ele­vated FLC lev­els can be reflec­tive of
under­ly­ing renal insuf­fi­ciency, auto­im­mune con­di­tions, sys­temic
inflam­ma­tion, infec­tion, and other causes. Furthermore, if the
indi­vid­ual undergoes a bone mar­row biopsy, based on cur­rent
diag­ nos­tic cri­te­ria, the plasma cell involve­ ment of the bone
mar­row must be less than 10%.1 Last, the clin­i­cal workup must
be neg­a­tive for evi­dence of end-organ dam­age from plasma
cell dys­cra­sia, hyper­cal­ce­mia, ane­mia, renal fail­ure, lytic bone
lesions, or mul­ti­ple (2 or more) focal lesions in the skel­et­on by
mag­netic res­o­nance imag­ing.9 If 1 or more of those abnor­mal­
i­ties are iden­ti­fied, unless there is another expla­na­tion for the
abnor­mal­ity (eg, ane­mia due to bleed­ing, renal fail­ure due to
car­dio­vas­cu­lar dis­ease), then the patient would be diag­nosed
with MM. If none of the above-listed abnor­mal­i­ties are pres­ent,
but the M-spike is 3 g/dL or more and/or there is over 10% (but
664  |  Hematology 2021  |  ASH Education Program
less than 60%) plasma cells in the bone mar­row; then, based on
cur­rent cri­te­ria, the diag­no­sis would be smol­der­ing mye­loma.1,9
Using a com­bi­na­tion of serum-based pro­tein assays (SPEP,
immunofixation, and serum FLC assays), MGUS cases can be
clas­si­fied based on the isotype of M-pro­teins pres­ent. So-called
non-IgM MGUS is the most com­mon type and is defined by the
pres­ence of IgG, IgA, and, rarely, immu­no­glob­u­lin D or immu­no­
glob­u­lin E M-pro­teins.10 IgM MGUS is defined by the pres­ence of
IgM M-pro­teins.11 Light chain MGUS is defined by an abnor­mal FLC
ratio, indi­cat­ing an excess of mono­clo­nal FLCs in the absence of
M-pro­teins.12 Non-IgM MGUS and light chain MGUS are caused by
mono­clo­nal bone mar­row plasma cells, and they are pre­cur­sors
of MM.2 IgM MGUS is com­monly caused by mono­clo­nal lymph-
Downloaded from http://ashpublications.org/hematology/article-pdf/2021/1/662/1851672/662landgren.pdf by guest on 13 December 2021
oplasmacytic cells and is a pre­cur­sor to other lymphoprolifera-
tive dis­or­ders, most nota­bly Waldenström mac­ro­glob­u­li­ne­mia,
chronic lym­pho­cytic leu­ke­mia, and, only in very rare instances
(<0.5%), MM.11 In addi­tion, MGUS of all­types, espe­cially λ light
chain MGUS, can pre­cede amy­loid light chain amy­loid­osis.13 It
should be noted that in the gen­eral MM pop­u­la­tion, around 80%
of patients have an M-pro­tein and approx­i­ma­tely 20% have light
chain secre­ tory MM (ie, with­ out evi­
dence of an M-pro­ tein).14
Among patients with MM who have an M-pro­tein, most of them
also have an abnor­mal FLC ratio caused by over­pro­duc­tion of
either κ or λ FLCs.14 Among patients diag­nosed with MGUS with
an M-spike, around 30% also have an abnor­mal FLC ratio.15
Epidemiologic stud­ies: novel insights
In ret­ro­spec­tive stud­ies with long-term fol­low-up, Kyle et al10
and oth­ers16 have reported 0.5% to 1.0% annual aver­age risk of
pro­gres­sion from MGUS to MM. Importantly, most ret­ro­spec­tive
stud­ies seek­ing to iden­tify risk fac­tors for pro­gres­sion are based
on sta­tis­ti­cal mod­els using risk fac­tor data from a sin­gle time
point (usu­ally the ini­tial workup).10,16 Data from this kind of mod­
el­ing have been used to develop clin­i­cal con­sen­sus guide­lines
recommending annual periph­ eral blood mon­ i­tor­
ing of serum
pro­tein mark­ers and other assays for patients with inter­me­di­ate-
risk and high-risk MGUS.1 Inspired by the obser­va­tions by Kyle
in 1978,6 a few smaller ret­ro­spec­tive stud­ies17,18 have pro­posed
evolv­ing changes in M-pro­tein lev­els are asso­ci­ated with pro­
gres­sion to MM. However, in clin­i­cal prac­tice, most patients are
typ­i­cally coun­seled based on their risk pro­file cap­tured at ini­tial
workup. Regarding cases with light chain MGUS, there is only lim­
ited infor­ma­tion avail­­able regard­ing the risk of pro­gres­sion from
light chain MGUS to light chain MM, and con­se­quently, clin­i­cal
guide­lines for this con­di­tion are lacking.12 However, the big­gest
lim­i­ta­tion of defin­ing risk in this fash­ion is lead-time bias because
there are no pri­ mary care screen­ ing guide­ lines. Therefore,
patients are often found to have MGUS after an inci­den­tal rou­tine
lab­o­ra­tory find­ing (eg, as part of the workup for ele­vated total
pro­tein). Alternatively, patients may be diag­nosed with MGUS
after a workup for non-mye­loma-defin­ing signs/symp­toms (eg,
unex­plained periph­eral neu­rop­at­hy or slight increase in serum
cre­at­i­nine).
In 2009, the first pro­spec­tive pop­ul­a­tion-based MGUS screen­
ing study was published. Using stored periph­eral blood sam­
ples that were col­lected as part of the large (N = 77 469) National
Cancer Institute Prostate, Lung, Colorectal, and Ovarian Cancer
Screening Trial (NCI PLCO), this large study was a ­ ble to show
that MM is con­sis­tently pre­ceded by the MGUS pre­cur­sor stage.2
This impor­tant obser­va­tion defin­i­tively links MM to its ­pre­cur­sor
dis­ease. This study has set the stage for future inves­ti­ga­tions limit of nor­mal; 1-2 points). For the first time, a scor­ing sys­tem
seek­ing