You are on page 1of 513

Practical Clinical Oncology

Practical Clinical Oncology has been designed as a handson review of all aspects of current practice in clinical oncology. The introductory section to the book provides background information on the four main treatment modalities – radiotherapy, chemotherapy, hormone therapy and biological therapy – plus research, oncological emergencies and palliative care. Subsequent chapters describe the diagnosis and treatment of malignancies, based on tumour site or type. Each chapter follows a standard template, making it easier for the readers to locate information quickly; multiple choice questions are also provided to enable readers to test their knowledge. With an emphasis on practical information that will be useful in day-to-day decision making and treatment, Practical Clinical Oncology is an invaluable resource on clinical care of the cancer patient for all trainees in clinical oncology, medical oncology, surgical oncology and palliative care, as well as for specialist nurses and radiographers. Louise Hanna is a Consultant Clinical Oncologist at the Velindre Cancer Centre, Velindre Hospital, Cardiff. Tom Crosby is a Consultant Clinical Oncologist at the Velindre Cancer Centre, Velindre Hospital, Cardiff. Fergus Macbeth is a Consultant Clinical Oncologist at the Velindre Cancer Centre, Velindre Hospital, Cardiff.

Practical Clinical Oncology
Edited by

Louise Hanna Tom Crosby Fergus Macbeth

CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521618168 © Cambridge University Press 2008 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2008

ISBN-13 978-0-511-37854-6 ISBN-13 978-0-521-61816-8

eBook (NetLibrary) paperback

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate. Every effort has been made to ensure that the information in this text, including drug and radiotherapy doses, is correct. However, the reader is strongly advised to consult the published product information and data sheets provided by manufacturers. The authors and publishers cannot accept any legal responsibility or liability for any errors or omissions which occur in this book or for the misuse or misapplication of the material therein

CONTENTS

List of contributors Preface Acknowledgements Abbreviations

page vii ix x xi

14 Anus
Richard Adams and Tom Crosby

174 183 190 214 222 231 240 252 257 267 278 290 296 304 313 328

15 Gastrointestinal stromal tumours
Kate Parker and Tom Crosby

1 Practical issues in cytotoxic chemotherapy usage
Sian Evans and Philip Savage

1 13 23 39 55 70 83 93 121 132 141 151 159

16 Breast
Nayyer Iqbal and Peter Barrett-Lee

17 Kidney 2 Biological treatments in cancer
Rachel Jones and Robert Leonard Jason Lester and John Wagstaff

18 Bladder 3 Hormones in cancer
Jacinta Abraham and John Staffurth Stephen Williams and Jim Barber

19 Prostate
Jim Barber and John Staffurth

4 Radiotherapy planning
Andrew Tyler and Louise Hanna

20 Testis
Jim Barber and John Staffurth

5 Research in cancer
Robert Hills

21 Penis
Jim Barber

6 Oncological emergencies
Paul Shaw

22 Ovary
Louise Hanna and Malcolm Adams

7 Palliative care
Simon Noble

23 Body of the uterus
Louise Hanna and Malcolm Adams

8 Head and neck
Laura Moss and Chris Gaffney

24 Cervix
Louise Hanna and Malcolm Adams

9 Oesophagus
Tom Crosby

25 Vagina
Louise Hanna and Malcolm Adams

10 Stomach
Michael Button and Tom Crosby

26 Vulva
Louise Hanna and Malcolm Adams

11 Liver, gallbladder and biliary tract
Somnath Mukherjee and Tom Crosby

27 Gestational trophoblast tumours
Philip Savage

12 Exocrine pancreas
Somnath Mukherjee and Tom Crosby

28 Lung
Fergus Macbeth and Carys Morgan

13 Colon and rectum
Richard Adams, Timothy Maughan and Tom Crosby

29 Mesothelioma
Louise Hanna and Fergus Macbeth

v

Contents

30 Soft tissue and bone tumours in adults
Owen Tilsley

335 347 370 382 395 406

36 Neuroendocrine tumours
Atul Kalhan and Aled Rees

418 426 442

31 The lymphomas and myeloma
Eve Gallop-Evans and Chris Poynton

37 Cancer in children
Sally Goodman

32 Central nervous system
Sean Elyan

38 Cancer of unknown primary
Paul Shaw and Tom Crosby

33 Skin cancer other than melanoma
Sankha Suvra Mitra

39 The use of radiotherapy in the treatment of benign conditions
Alison Brewster

449

34 Melanoma
Tom Crosby, Louise Hanna and Dafydd Roberts

35 Thyroid
Laura Moss

Multiple choice questions Multiple choice answers Index

453 468 469

vi

CONTRIBUTORS

Jacinta Abraham
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Eve Gallop-Evans
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Malcolm Adams
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Sally Goodman
Consultant Clinical Oncologist, The Bristol Haematology and Oncology Centre, Bristol, UK.

Richard Adams
Senior Lecturer in Clinical Oncology, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Louise Hanna
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Jim Barber
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Robert Hills
Senior Lecturer in Translational Statistics, Department of Haematology, Cardiff University, Cardiff, UK.

Peter Barrett-Lee
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Nayyer Iqbal
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Alison Brewster
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Rachel Jones
Specialist Registrar in Medical Oncology, South West Wales Cancer Institute, Singleton Hospital, Sketty, Swansea, UK.

Michael Button
Specialist Registrar in Clinical Oncology, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Atul Kalhan
Specialist Registrar, Centre for Endocrine and Diabetes Sciences, University Hospital of Wales, Heath Park, Cardiff, UK.

Tom Crosby
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Sean Elyan
Consultant Clinical Oncologist, Cheltenham General Hospital, Cheltenham, UK.

Robert Leonard
Professor of Medical Oncology, South West Wales Cancer Institute, Singleton Hospital, Sketty, Swansea, UK.

Sian Evans
Chief Pharmacist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Jason Lester
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Chris Gaffney
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Fergus Macbeth
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

vii

Contributors

Timothy Maughan
Professor of Cancer Studies, Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Aled Rees
Senior Lecturer, Centre for Endocrine and Diabetes Sciences, School of Medicine, Cardiff University, Heath Park, Cardiff, UK.

Sankha Suvra Mitra
Consultant Clinical Oncologist, The Sussex Cancer Centre, Royal Sussex County Hospital, Brighton and Sussex University Hospitals NHS Trust, Brighton, East Sussex, UK.

Dafydd Roberts
Consultant Dermatologist, Singleton Hospital, Sketty, Swansea, UK.

Philip Savage
Consultant Medical Oncologist, Department of Health Gestational Trophoblastic Tumour Unit, Charing Cross Hospital, London, UK.

Carys Morgan
Specialist Registrar in Clinical Oncology, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Paul Shaw
Specialist Registrar in Clinical Oncology, Bobby Moore Clinical Research Fellow, School of Bioscience, Cardiff University, Park Place, Cardiff, UK.

Laura Moss
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

John Staffurth
Clinical Senior Lecturer in Oncology, Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Somnath Mukherjee
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Owen Tilsley
Consultant Clinical Oncologist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Simon Noble
Senior Lecturer and Honorary Consultant in Palliative Care, Royal Gwent Hospital, Newport, UK.

Andrew Tyler
Medical Physicist, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

Kate Parker
Specialist Registrar in Clinical Oncology, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

John Wagstaff
Consultant Medical Oncologist, Singleton Hospital, Sketty, Swansea, UK.

Chris Poynton
Senior Lecturer and Honorary Consultant Haematologist, University Hospital of Wales, Heath Park, Cardiff, UK.

Stephen Williams
Specialist Registrar in Clinical Oncology, Velindre Cancer Centre, Velindre Hospital, Whitchurch, Cardiff, UK.

viii

PREFACE

This book is intended primarily for trainees in clinical oncology, but members of other professions such as medical oncology, surgery, palliative care, nursing and radiography will also find it useful. The book started life as a set of lecture notes from the Cardiff Annual FRCR Part II course but has since grown to include more topics than could possibly be covered during the three days of that course. Our approach in producing this volume has been to focus on practical suggestions appropriate to day-to-day decision making during the treatment of oncology patients. We are very grateful to our colleagues from Velindre and elsewhere, who are listed on page xi, for reviewing specific chapters and ensuring that the advice contained within is as widely applicable as possible. The first seven chapters cover ‘generic’ topics that provide background information on cancer treatments. These are chemotherapy, biological and hormonal treatments, radiotherapy planning, research, emergencies and palliative care. The chapters that follow each focus on a tumour site or tumour type. In this latter group, the chapter layout is fairly consistent to help the reader navigate through the book. Thus, each chapter begins with background information on tumour types, anatomy, incidence, epidemiology, risk factors and

aetiology. Next, there are sections on pathology, routes of spread and, where appropriate, screening. These are followed by clinical sections on presentation, investigations, treatment, and prognosis. Most of the chapters also discuss areas of current interest and clinical trials, reflecting the rapidly changing nature of clinical oncology where many areas of practice are open to debate. Where references are given, we have tried as much as possible to include the key publications that have influenced clinical practice. Towards the end of the book, there is a series of ‘single best answer’ multiple choice questions, which give the readers the opportunity to test their knowledge. In a book of this length, it is not possible to provide as much of the subject as would be found, for example, in the larger multivolume oncology textbooks. Nevertheless, an attempt has been made to give an overview of clinical oncology practice at the present time, which we hope will be of interest and benefit to trainees. The idea for writing this book came about several years ago when two of the editors (TC and LH) were studying for their FRCR part II examination. They have since become consultants in Velindre Hospital with FM and all three now teach on the Cardiff Annual FRCR part II course.

ix

ACKNOWLEDGEMENTS We are very grateful to those who have helped by reviewing specific chapters: Dr K. Wolstenholme. Somaiah. Wardley. Blake. Nutting. Prof A. Evans. Thomas. and the staff at Cambridge University Press. Dr N. Dr J. Dr M. Dr G. we acknowledge our respective families for their enduring love and support during the preparation of this book. Fiander. And. Prof M. Williams and Dr V. Perrett. Dr S. Benstead. O’Rourke. Prof H. Rhydian Maggs and Lucy Wills of the Department of Medical Physics. Carter. Mason. Dr P Keeley. Sebag-Montefiore. finally. Gaffney. Glees. Prof R. Dr C. Mr A. Dr D. Dr U. x . Bertelli. Marples. Ms T. Dr C. Betty Fulford and Eleanor Umali who have always responded helpfully to our many queries. Dr M. particularly Pauline Graham. Burnet. Taylor. Our thanks also go to those who have helped with the preparation of this book: Jane Flewitt. Dr N. Falk. Dr A. Dr V. Dr P. Mallick. Dr S. Dr N. . Anne Cleves of the Cancer Research Wales Library. Khoo. Velindre Cancer Centre.

d. size acute myelocytic leukaemia axillary nodal clearance accelerated partial breast irradiation abdominoperineal anterior–posterior activator protein-1 antigen-presenting cell adenomatosis polyposis coli abdominoperineal resection amine precursor uptake and decarboxylation androgen receptor androgen response elements arimidex-nolvadex Administration of Radioactive Substances Advisory Committee American Society of Clinical Oncology autologous stem cell transplant American Society of Hematology A STudy in Endometrial Cancer American Society for Therapeutic Radiology and Oncology Arimidex. extent. metastases. grade. size adjuvant hormone therapy aromatase inhibitor aquired immune deficiency syndrome Adjuvant Interferon in Melanoma HIGH risk American Joint Committee on Cancer thymoma viral proto-oncogene acute lymphoblastic leukaemia acral lentinginous melanoma ALMANAC ALND AMES AML ANC APBI AP A–P AP-1 ApC APC APR APUD AR ARE ARNO ARSAC ASCO ASCT ASH ASTEC ASTRO ATAC ATD ATLAS ATP aTTom AUC b. BC BCG BCL2 BCR Axillary Lymphatic Mapping Against Nodal Axillary Clearance axillary lymph node dissection age. Alone or in Combination amino-terminal domain Adjuvant Tamoxifen Longer Against Shorter adenosine triphosphate adjuvant Tamoxifen – Treatment offer more area under curve twice a day British Columbia Bacille Calmette-Gu´ rin e B-cell CLL/lymphoma 2 breakpoint cluster region xi . extent. Tamoxifen.ABBREVIATIONS General 1D 2D 3D 5AC 5-FU 5-HIAA 5-HT3 5YDFS 5YOS 5YS αFP βhCG AAPM ABCSG ABL ACE ACh ACOSOG ACTH ACTION ADH ADT AF AGES AHT AI AIDS AIM HIGH AJCC AKT ALL ALM one-dimensional two-dimensional three-dimensional MUC subtypes A and C 5-fluorouracil 5-hydroxy-indoleacetic acid 5-hydroxy-tryptamine 3 5-year disease-free survival 5-year overall survival 5-year survival alpha feto-protein beta human chorionic gonadotrophin American Association of Physicists in Medicine Austrian Breast and Colorectal Cancer Study Group Abelson protein tyrosine kinase anticholinesterase acetylcholine American College of Surgeons Oncology Group Adrenocorticotrophic hormone Adjuvant Chemotherapy in Ovarian Neoplasm antidiuretic hormone androgen deprivation therapy activating function age.

subfamily B.Abbreviations BCT BGND BIG BILCAP BIR BMD BMI BMT BNLI BOND BP BRAF BRCA BSA BSC BSO BTOG BTS CA CagA CAIX CALGB Cb CCLG CD CDH-1 CE CEA CgA CHART CHF CI CIN CIS CK CLA CLCG CLL CML cN CNS xii breast conservation therapy bilateral groin node dissection Breast International Group adjuvant capecitabine in biliary tract cancer trial British Institute of Radiology bone mineral density body mass index bone marrow transplant British National Lymphoma Investigation Bowel Oncology with Cetuximab Antibody blood pressure v-raf murine sarcoma viral oncogene homolog B1 breast cancer gene body surface area best supportive care bilateral salpingo-oophorectomy British Thoracic Oncology Group British Thoracic Society cancer antigen cytotoxicity-associated immunodominant antigen carbonic anhydrase IX Cancer and Leukaemia Group B carboplatin Children’s Cancer and Leukaemia Group cluster of differentiation cadherin 1 conversion electron carcino-embryonic antigen chromogranin A continuous hyperfractionated accelerated radiotherapy congestive heart failure confidence interval cervical intraepithelial neoplasia carcinoma in situ cytokeratin common leukocyte antigen Children’s Leukaemia Cooperative Group chronic lymphocytic leukaemia chronic myelocytic leukaemia clinical lymph node stage central nervous system COG COMS CONSORT COPD CR CRC CrCl Creat CRM CRP CRT CRUK CSF CSI CT CTA CTV CTZ CUP CVP CX CXR CYP1B1 D D2 DAHANCA DCC DCIS DDFS DDT DES DFS DHA DHT DLBCL dmax DMC DMSA DMSO DNA DOPA DRE DRR DTC DTIC DVH Children’s Oncology Group of North America Collaborative Ocular Melanoma Study Consolidated Standards of Reporting Trials chronic obstructive pulmonary disease complete response colorectal cancer creatinine clearance serum creatinine circumferential resection margin c-reactive protein chemoradiotherapy Cancer Research UK cerebrospinal fluid craniospinal irradiation computed tomography clinical trials authorisation clinical target volume chemoreceptor trigger zone cancer of unknown primary central venous pressure characteristic X-ray photon chest X-ray cytochrome P450. polypeptide 1 docetaxel dopamine D2 Danish Head and Neck Cancer deleted in colon cancer ductal carcinoma in situ distant-disease-free survival dichloro-diphenyl-trichloroethane diethylstilboestrol disease-free survival dihydroxyandrostenedione 5α dihydrotestosterone diffuse large B-cell lymphoma depth of maximum dose Data Monitoring Committee dimercapto succinic acid dimethyl sulfoxide deoxyribonucleic acid dihydroxyphenylalanine digital rectal examination digitally reconstructed radiograph differentiated thyroid cancer dacarbazine dose volume histogram . family 1.

official name = v-erb-b2 erythroblastic leukemia viral oncogene homolog 2. official symbol = ERBB2.Abbreviations EAU EBC EBCTCG EBRT EBV ECG Echo ECOG EDTA EGF EGFR EIC ELISA ELND EM EMA ENET ENT EORTC EPO EPP ER ERB B1 ERBB2 ERCP ERE ERG ESPAC ESR ESTRO EU EUA EURAMOS European Association of Urology early breast cancer Early Breast Cancer Trialists’ Collaborative Group external beam radiotherapy Epstein Barr virus electrocardiogram echocardiogram Eastern Cooperative Oncology Group ethylenediaminetetraacetic acid epidermal growth factor epidermal growth factor receptor extensive intraductal component enzyme-linked immunosorbent assay elective lymph node dissection electron microscopy epithelial membrane antigen European Neuroendocrine Tumor Network ear nose throat European Organisation for Research and Treatment of Cancer erythropoietin extrapleural pneumonectomy oestrogen receptor official symbol = EGFR. official name = epidermal growth factor receptor (erythroblastic leukemia viral (v-erb-b) oncogene homolog. official name = Ewing sarcoma breakpoint region 1 folinic acid familial adenomatous polyposis full blood count Food and Drug Administration fluorodeoxyglucose Federation Internationale de Gynecologie et d’Obstetrique Finland Herceptin Trial fluorescent in situ hybridisation follicular lymphoma Friend leukemia virus integration 1 Follicular Lymphoma International Prognostic Index fms-related tyrosine kinase 3 peptide deformylase fine needle aspiration fine needle aspiration cytology faecal occult blood Fellow of the Royal College of Radiologists focus skin distance follicle-stimulating hormone free T4 follicular thyroid carcinoma gastrointestinal autonomic tumour giga-Becquerel germinal centre granulocyte-colony stimulating factor good clinical practice germ cell tumour Groupe d’ Etude des Lymphomes de l’ Adulte glomerular filtration rate growth hormone-releasing hormone German Hodgkin Study Group gastrointestinal gastrointestinal stromal tumour glioma-associated oncogene homolog gonadotropin-releasing hormone gastro-oesophageal Gynecologic Oncology Group gastro-oesophageal reflux disease General Practitioner gestational trophoblast tumour gross tumour volume Gray histamine H2 xiii . avian) (alias = HER-2). neuro/glioblastoma derived oncogene homolog (avian) endoscopic retrograde cholangiopancreatogram oestrogen response element v-ets erythroblastosis virus E26 oncogene homolog (avian) European Study Group for Pancreatic Cancer erythocyte sedimentation rate European Society for Therapeutic Radiology and Oncology European Union examination under anaesthetic European and American Osteosarcoma Study Group EUS EWS FA FAP FBC FDA FDG FIGO FinHer FISH FL FLI1 FLIPI FLT3 fms FNA FNAC FOB FRCR FSD FSH FT4 FTC GANT GBq GC G-CSF GCP GCT GELA GFR GHRH GHSG GI GIST GLI GnRH GO GOG GORD GP GTT GTV Gy H2 endoscopic ultrasound EWSR1.

v. IVC IVU JDG JVP KIT KPS KRAS LACE LAK LCIS LD LDH LDL LDR LE International Germ Cell Cancer Collaborative Group intratubular germ cell neoplasia insulin-like growth factor-1 image-guided radiation therapy immunohistochemistry ischaemic heart disease internal jugular vein interleukin-2 invasive lobular carcinoma intraluminal brachytherapy intramuscular internal margin International Mesothelioma Interest Group internal mammary node investigational medicinal product intensity-modulated radiation therapy Iressa NSCLC Trial Assessing Combination Treatment international prognostic factor score international prognostic index Intergroup Rhabdomyosarcoma Studies International Organisation for Standardisation international staging system Italian Tamoxifen Arimidex trial intention to treat intensive therapy unit international units intravenous inferior vena cava intravenous urogram jugulo-digastric jugulo-venous pressure kitten (v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog) Karnofsky performance status v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog Lung Adjuvant Cisplatin Evaluation lymphokine-activated killer lobular carcinoma in situ latissimus dorsi.Abbreviations HAART HAI HART HBF HBL HBV HCC HCG HCV HDC/ASCT HDCT HDR HDU HER-2 HERA HGG HH HHV HIFU HIV HL HMB HNPCC HPOA HPV HR HRT HTLV-1 IBCSG ICAM1 ICC ICG ICON ICRU IDC IDL IES IFNα IFNAR IFNGR IFRT IGCCC highly active antiretroviral therapy hepatic arterial infusion hyperfractionated accelerated radiotherapy heterotopic bone formation hepatoblastoma hepatitis B virus hepatocellular carcinoma human chorionic gonadotropin hepatitis C virus high-dose chemotherapy with autologous stem cell transplant high-dose chemotherapy high dose rate high-dependency unit human epidermal growth factor receptor 2 Herceptin® Adjuvant high-grade glioma hedgehog human herpesvirus high-intensity focussed ultrasound human immunodeficiency virus Hodgkin lymphoma human melanoma black hereditary non-polyposis colorectal cancer hypertrophic pulmonary osteo-arthropathy human papilloma virus hazard ratio hormone replacement therapy human T-cell lymphotropic virus -1 International Breast Cancer Study Group intercellular adhesion molecule 1 interstitial cells of Cajal iodocyanine green International Collaborative Ovarian Neoplasm study International Commission on Radiation Units and Measurements invasive ductal carcinoma indirect laryngoscopy International Exemestane Study interferon alpha interferon (alpha and beta) receptor interferon gamma receptor involved-field radiotherapy International Germ Cell Consensus Classification IGCCCG IGCN IGF-1 IGRT IHC IHD IJV IL-2 ILC ILT i. lymphocyte depleted lactate dehydrogenase low-density lipoprotein low dose rate local excision xiv . IM IMIG IMN IMP IMRT INTACT IPFS IPI IRS ISO ISS ITA ITT ITU IU i.m.

official name = N-myc (and STAT) interactor (NMI) number needed to treat not otherwise specified non-pulmonary visceral metastatic disease xv NNT NOS NPVMD .Abbreviations LEEP LFT LGG LH LHRH LHRHa LLETZ LM LMM LN LOH LR LVI MAB MACIS MAG 3 MAGE MAGIC MALT MAP MAPK MART MBC MBq MC MDM MDR MDT MEN MF MFH MGMT MGUS MHRA MIBG MInT MLC MM MR MRC loop electro-excision procedure liver function test low-grade glioma luteinising hormone luteinising hormone releasing hormone luteinising hormone releasing hormone agonist large loop excision of the transformation zone lentigo maligna lentigo maligna melanoma lymph node loss of heterozygosity local recurrence lymphovascular invasion maximal androgen blockade metastases. age. completeness of surgery. invasion of extrathyroidal tissues. size mercaptoacetyltriglycerine melanoma antigen expression family Medical Research Council Adjuvant Gastric Infusional Chemotherapy mucosa-associated lymphoid tissue mitogen activated protein mitogen-activated protein kinase melanoma antigen recognised by T-cells metastatic breast cancer mega-Becquerel mixed cellularity multidisciplinary meeting medium dose rate multidisciplinary team multiple endocrine neoplasia mycosis fungoides malignant fibrous histiocytoma O6 methylguanine-DNA methyltransferase monoclonal gammopathy of undetermined significance Medicines and Healthcare Products Regulatory Authority meta-iodobenzylguanidine MabThera® International Trial multileaf collimator malignant melanoma magnetic resonance Medical Research Council MRCP MRI mRNA MS MSCC MSH MSI MSU MTC MTD MTI mTOR MTT MTU MUC-1 MUGA scan MV MW MYC NAC NAHT NCCN NCCTG NCI NCICanada NCRI NCRN Nd-YAG NEAT NET NF NHL NHS NICE NIH n-myc magnetic resonance cholangiopancreatogram magnetic resonance imaging messenger ribonucleic acid median survival malignant spinal cord compression DNA mismatch repair gene microsatellite instability midstream urine medullary thyroid carcinoma maximally tolerated dose malignant teratoma intermediate target of rapamycin malignant teratoma trophoblastic malignant teratoma undifferentiated mucin 1 multigated acquisition scan megavoltage molecular weight v-myc myelocytomatosis viral oncogene homolog nipple areola complex neoadjuvant hormone therapy National Comprehensive Cancer Network National Central Cancer Treatment Group National Cancer Institute National Cancer Institute of Canada National Cancer Research Institute National Cancer Research Network neodynium-doped yttrium-aluminium-garnet National Epirubicin Adjuvant Trial neuroendocrine tumour neurofibromatosis non-Hodgkin lymphoma National Health Service National Institute for Health and Clinical Excellence National Institutes of Health official symbol = NMI.

PORT PPE PR PR-A PR-B PRV PS PSA PSTT pT PTC PTCH PTEN PTH PTH-RP PTV PUVA PV PVC QA QART QLQ QOL R0 RAGE RAF RAS Rb RB1 RBE RCC RCR RCT REAL RECIST RFA rhTSH per oral postoperative radiotherapy palmar-plantar erythrodysaesthesia partial response progesterone receptor A progesterone receptor B planning risk volume WHO performance status prostate-specific antigen placental site trophoblast tumour pathological tumour stage papillary thyroid carcinoma (in thyroid cancer).o. percutaneous transhepatic cholangiograph (in hepatobiliary cancer) patched gene phosphatase and tensin homolog (mutated in multiple advanced cancers 1) parathyroid hormone parathyroid hormone-related peptide planning target volume psoralen plus ultraviolet A per vagina poly(vinyl chloride) quality assurance Quality Assurance in Radiation Therapy quality of life questionnaire quality of life complete resection renal antigen expression family v-raf-1 murine leukaemia viral oncogene homolog rat sarcoma viral oncogene homolog retinoblastoma retinoblastoma 1 (including osteosarcoma) radiobiologically equivalent dose renal cell carcinoma Royal College of Radiologists randomised controlled trial revised European-American lymphoma response evaluation criteria in solid tumours radiofrequency ablation recombinant human thyroid-stimulating hormone .Abbreviations NS NSAA NSABP NSAID N/saline NSCLC NSGCT NST NTRAC OAR OC OFS OR OS p450 PAI-1 Pap PCI PCNSL pCR PDA PDC PDD PDGF PDGFR PDN PDR PDT PEI PERCHE PET PFS PgR PI3K PICC PLD PLDH pM PMS pN PNET xvi not significant non-steroidal antiandrogen National Surgical Adjuvant Breast and Bowel Project non-steroidal anti-inflammatory drug normal saline non-small-cell lung cancer non-seminomatous germ cell tumour no specific type National Translational Cancer Research Network organs at risk oesophageal cancer ovarian function suppression odds ratio overall survival cytochrome p450 plasminogen activator inhibitor type 1 Papanicolau prophylactic cranial irradiation primary central nervous system lymphoma pathological complete response poorly differentiated adenocarcinoma poorly differentiated carcinoma percentage depth dose platelet-derived growth factor platelet-derived growth factor receptor poorly differentiated neoplasm pulsed dose rate photodynamic therapy percutaneous ethanol injection Premenopausal Endocrine Responsive Chemotherapy Trial positron emission tomography progression-free survival progesterone receptor phosphatidyl inositol 3 kinase peripherally inserted central catheter post-transplantation lymphoproliferative disease pegylated liposomal doxorubicin hydrochloride pathological metastasis stage PMS1 postmeiotic segregation increased 1 pathological lymph node stage primitive neuroectodermal tumour p.

Abbreviations RMI RMH RMS RPLND RR RT RTOG S-100 SAB SABCS SAE SCC SCF SCLC SCTBG SEER SERM SHH SI S–I SIADH SIGN SIOP SLN SLND SM SMA SLNB SOFT SPECT src SRH SSD SSG SSM SSP SSRS STNI SUSAR SUV SV 40 relative malignancy index Royal Marsden Hospital rhabdomyosarcoma retroperitoneal lymph node dissection response rate radiotherapy Radiation Therapy Oncology Group S-100 calcium-binding protein same as before San Antonio Breast Cancer Symposium serious adverse event squamous cell carcinoma supraclavicular fossa small-cell lung cancer Scottish Cancer Trials Breast Group Surveillance Epidemiology and End Results selective oestrogen receptor modulator sonic hedgehog sacro-iliac superior–inferior syndrome of inappropriate antidiuretic hormone Scottish Intercollegiate Guidelines Network Soci´ t´ Internationale d’Oncologie ee P´ diatrique e sentinel lymph node sentinel lymph node dissection setup margin smooth muscle actin sentinel lymph node biopsy Suppression of Ovarian Function Trial single photon emission computed tomography v-src sarcoma (Schmidt-Ruppin A-2) viral oncogene homolog stigmata of recent haemorrhage source-skin distance Scandanavian Sarcoma Group superficial spreading melanoma statutory sick pay somatostatin receptor scintigraphy subtotal nodal irradiation suspected unexpected serious adverse reaction standardised uptake value simian virus 40 SVC SVCO SWOG SXR T4 TACE TAH TB TBI TCC TD t. TEM TEXT TFT Tg TGF-β Tis TKI TLD TLS TME TMR TNFα TNM TP53 TPR TRAM TRH TRUS TSC TSE TSH TTF-1 TUR TURBT TURP TVS UC U+E UICC UK UKCCSG UKCTOCS UKNET superior vena cava superior vena cava obstruction Southwest Oncology Group superficial X-ray thyroxine transarterial chemoembolisation total abdominal hysterectomy tuberculosis total-body irradiation transitional cell carcinoma teratoma differentiated three times a day transanal endoscopic microsurgery Tamoxifen and EXemestane Trial thyroid function test thyroglobulin transforming growth factor beta carcinoma in situ tyrosine kinase inhibitor thermoluminescence dosimetry tumour lysis syndrome total mesorectal excision tissue maximum ratio tumour necrosis factor alfa tumour nodes metastases tumour protein p53 tissue phantom ratio transverse rectus abdominus myocutaneous thyrotropin-releasing hormone transrectal ultrasound trial steering committee total skin electrons thyroid-stimulating hormone thyroid transcription factor 1 transurethral resection transurethral resection of bladder tumour transurethral resection of the prostate transvaginal ultrasound ulcerative colitits urea and electrolytes International Union Against Cancer United Kingdom United Kingdom Children’s Cancer Study Group UK Collaborative Trial of Ovarian Cancer Screening United Kingdom Neuroendocrine Society xvii .s.d.

cisplatin bleomycin. prednisolone doxorubicin. bleomycin. cisplatin cyclophosphamide. cisplatin. cyclophosphamide doxorubicin. dexamethasone cyclophosphamide. aniridia. doxorubicin. etoposide. vincristine. etoposide. doxorubicin. procarbazine. vinblastine. 5-FU cisplatin. cisplatin. cytarabine. vincristine. prednisolone cyclophosphamide. melphalan bleomycin. dacarbazine daunorubicin. etoposide. bleomycin. cyclophosphamide.Abbreviations ULN uPA US USA US Intergroup USPIO UV UVA UVB VAIN VATS VC VEGF VEGFR VHL VIN VIP VSIM VQ WAGR WBC WBS WHO WLE XRT upper limit of normal urokinase-type plasminogen activator ultrasound scan United States of America United States Intergroup ultra-small-particle iron oxide ultraviolet ultraviolet A ultraviolet B vaginal intraepithelial neoplasia video-assisted thoroscopic surgery vomiting centre vascular endothelial growth factor vascular endothelial growth factor receptor Von Hippel-Lindau vulval intraepithelial neoplasia vasoactive intestinal peptide virtual simulation software ventilation perfusion Wilms’ tumour. capecitabine etoposide. cyclophosphamide. prednisolone cyclophosphamide. thalidomide. methotrexate. prednisolone doxorubicin. vinblastine. bleomycin. vincristine. etoposide. vinblastine cisplatin. paclitaxel. ambiguous genitalia and mental retardation white blood cell whole-body scan World Health Organisation wide local excision X-ray treatment BEAM BEC BEP BOP Bu/Cy CAF CarboPEC CAP CAV C-BOP CG CHOEP CHOP CMF CMV COJEC COPP CTD Chemotherapy regimens ABVD ABVPP AC A→CMF AC→ paclitaxel ACVBP A-T-C AVG BEACOPP doxorubicin. vincristine. cyclophosphamide cyclophosphamide. procarbazine. vincristine. etoposide. carboplatin. vindesine. vinblastine. thioguanine dexamethasone. dexamethasone cyclophosphamide. actinomycin-D – cyclophosphamide. cyclophosphamide doxorubicin. cytarabine. cyclophosphamide then paclitaxel doxorubicin. doxorubicin. methotrexate. bleomycin. cyclophosphamide cyclophosphamide. doxorubicin. doxorubicin. doxorubicin. vincristine carboplatin. cisplatin cisplatin gemcitabine cyclophosphamide. gemcitabine bleomycin. procarbazine. 5-FU doxorubicin. vincristine. prednisolone CVAD CVP CYVADIC DAT DHAP EBVD EC ECF E→CMF ECX EMA-CO carmustine. vincristine. doxorubicin. cyclophosphamide epirubicin. etoposide. cyclophosphamide. vincristine. methotrexate. prednisolone cyclophosphamide. bleomycin. ara-C. dacarbazine doxorubicin. cisplatin epirubicin. carboplatin bleomycin. methotrexate. doxorubicin. methotrexate. 5-FU epirubicin. vinblastine. vincristine. cyclophosphamide. vincristine. vincristine xviii . prednisolone epirubicin. cisplatin busulphan. 5-FU carboplatin. etoposide. prednisolone cyclophosphamide. cyclophosphamide cyclophosphamide. 5-FU epirubicin.

etoposide. carboplatin. doxorubicin. FA. doxorubicin. cisplatin – etoposide. carboplatin gemcitabine. cyclophosphamide 5-FU. oxaliplatin. cisplatin mitomycin. paclitaxel high-dose methotrexate fractionated cyclophosphamide. cyclophosphamide then docetaxel fludarabine. dexamethasone etoposide. cisplatin mitomycin. vincristine. oxaliplatin idarubicin. irinotecan 5-FU. prednisolone mechlorethamine. oxaliplatin gemcitabine. doxorubicin. procarbazine. cyclophosphamide 5-FU. cyclophosphamide 5-FU. prednisolone. cisplatin gemcitabine. cyclophosphamide. prednisolone. doxorubicin. cytarabine. cytarabine. vinblastine. cyclophosphamide. vincristine. dexamethasone methotrexate.Abbreviations EOF EOX EP EP-EMA ESHAP FAC FAMTX FC FEC FEC→ docetaxel FMD FOLFIRI FOLFOX GemCarbo Gem-cis GT HD-MTX Hyper-CVAD ICE JEB MACOP-B m-BACOD MAP MIC MOPP MOPP-ABV MVAC MVP NP epirubicin. vincristine. epirubicin. doxorubicin. carboplatin. etoposide. cyclophosphamide. methotrexate. capecitabine etoposide. cisplatin vinorelbine. vincristine. bleomycin methotrexate. cisplatin etoposide. doxorubicin. ifosfamide. doxorubicin. doxorubicin. bleomycin methotrexate. cisplatin OEPA PIAF PLADO ProMACEcytaBOM R-CHOP R-DHAP R-ICE R-CODOXM/IVAC Stanford V SuperPLADO TAC TIP VAD VIP XELOX Z-DEX vincristine. prednisolone rituximab. 5-FU epirubicin. ifosfamide. cytarabine. vinblastine. vincristine. prednisone PLADO + carboplatin docetaxel. methylprednisolone. doxorubicin. cisplatin 5-FU. vinblastine. vincristine. ifosfamide. dexamethasone. cyclophosphamide paclitaxel. cisplatin mechlorethamine. dexamethasone 5-FU. mechlorethamine. doxorubicin. procarbazine. vincristine. etoposide. FA. doxorubicin. actinomycin-D etoposide. ifosfamide. bleomycin. bleomycin. etoposide carboplatin. bleomycin. cisplatin rituximab. etoposide. cyclophosphamide. cisplatin vincristine. doxorubicin prednisone. interferon. cyclophosphamide. dexamethasone Radioisotopes 60 Co Cs 18 F 123 I 125 I 131 I 111 In 192 Ir 103 Pd 106 Ru 99m Tc 90 Y 137 cobalt-60 caesium-137 fluorine-18 iodine-123 iodine-125 iodine-131 indium-111 iridium-192 palladium-103 ruthenium-106 technetium-99m yttrium-90 xix . epirubicin. ifosfamide doxorubicin. cisplatin capecitabine. oxaliplatin. doxorubicin. vincristine. bleomycin. doxorubicin. cytarabine. 5-FU cisplatin. doxorubicin. doxorubicin cisplatin. mitoxantrone. methotrexate fludarabine. etoposide rituximab. vinblastine methotrexate. prednisone. vincristine. etoposide. dexamethasone ifosfamide. methotrexate. methotrexate. and methotrexate rituximab.

.

these side effects and treatment-related risks are seen as acceptable temporary problems. www. malignant cells can be many times more sensitive to cytotoxic drugs than the cells they have arisen from and. 2001). and standard chemotherapy textbooks (e...bccancer. breast cancer. fortunately.htm) should be consulted for this information. tumour cells are generally more sensitive to cytotoxic drugs than are their parent cells. The sensitivity of different tumour types to the actions of chemotherapy drugs varies widely among the cells of origin and across the range of drugs. and the use of granulocyte-colony stimulating factor (G-CSF) may be required to keep treatment on schedule. r The treatment of patients with advanced incurable malignancies. Whereas the majority of chemotherapy drugs have been developed empirically. where the primary aim is palliation and symptom control.ca/default. When chemotherapy is used curatively. for lymphoma (Lepage et al.g. because there is the clear intent of achieving either cure or.. Alongside this major divide between the differing types of malignancy. including some of the new targeted drugs. 2002. testicular cancer. the mechanisms for the greater effectiveness of some drugs in some tumours 1 . both the prescriber and the patient should be clear about the aims of treatment. In contrast. an increased chance of cure. the regimens used in these treatments have significant side effects including neutropenia. It will not be possible to give a comprehensive description of every drug and regimen.. An increase in overall survival is not usually the primary aim of treatment and very toxic treatments are not usually justified. 1993) and in the adjuvant treatment of Cytotoxic chemotherapy: mode of action Cytotoxic chemotherapy drugs are systemic therapies that aim to kill or slow the growth of tumour cells while being relatively sparing to normal cells. will be discussed. However. more sensitive than the cells of the bone marrow. In this case. Generally. to prescribe chemotherapy safely and to manage the common treatment-related side effects. Aims of chemotherapy treatment There are three main indications for the use of chemotherapy: r The management of patients with curable advanced malignancies including choriocarcinoma. for adjuvant treatment. Before starting a course of chemotherapy.bc. 2003) or specialist websites (e. In the more common malignancies. there is a wide range of activity of the different chemotherapy drugs across the different tumours. Summerhays and Daniels. BC Cancer Agency. Hodgkin lymphoma and high-grade nonHodgkin lymphoma (NHL). patients having palliative chemotherapy should benefit by experiencing an improved quality of life. it is essential to maintain the calculated dose and dosage schedule according to the treatment protocol. where the rate of relapse is higher when the dose intensity is reduced (Wood et al. 1994). In curable cancers. primarily breast cancer and colorectal cancer. the principles of cytotoxic chemotherapy treatment and the appropriate use of anticancer drugs. sometimes without a major expectation of prolonging survival. The importance of this has previously been shown for testicular cancer (Toner et al. r The preoperative or postoperative adjuvant treatment of localised malignancies. maintaining dose intensity is not so important and dose reductions can be made to ensure that the patient safely tolerates the treatment. Allwood et al. but they are insufficiently sensitive to achieve a cure. this chapter should provide chemotherapy prescribers and administrators with enough information to discuss treatments with patients.g.1 PRACTICAL ISSUES IN CYTOTOXIC CHEMOTHERAPY USAGE Sian Evans and Philip Savage Introduction In this chapter. However.

but it causes minimal myelosuppression. the more recent addition of an anti-CD20 monoclonal antibody. r Combinations should avoid drugs of the same class or those with similar modes of action. r The drugs should have different dose-limiting toxicities. fluorouracil and gemcitabine).Sian Evans and Philip Savage are now becoming better understood. The treatment of high-grade B-cell NHL provides a similar example of effective combination chemotherapy and the benefits of adding in a modern drug with a completely different mode of action. In this regimen. the distinction is of some use in anticipating the side effects of chemotherapy. Although the cell-cycle distinction has been of great value in developing drug protocols and combinations. Single-agent therapy does remain a part of palliative chemotherapy for several tumour types. there have been several important trials that have tested more toxic and complex regimens against CHOP.’ The cycle-specific drugs. are combined with vincristine and prednisolone. By combining these drugs with their differing patterns of toxicity. however. such as the antimetabolites (methotrexate. However. etoposide and cisplatin have all been shown to have significant activity as single agents. After this initial approval. 1987). doxorubicin. These drugs are most effective in tumours with high mitotic indices and they produce greater cell kill if given in prolonged exposures. 1993. the principles of choosing combinations of chemotherapy are as follow: r Each drug is active against the tumour as a single agent. The three drugs have different patterns of dose-limiting toxicities. interact predominantly with cells that are actively synthesising new DNA in the S phase. it is usually on the basis of data that have been obtained in trials using single agents.. 1992). and cell killing is more closely linked to the total dose rather than to the duration of administration. which are non-myelosuppressive. each of which is active in the tumour type as a 2 single agent. The bleomycin. 2005). with negative results (Fisher et al. modern research suggests that this distinction is relatively crude and in fact most drugs have actions against both dividing and resting cells. rituximab... and prednisolone (CHOP) chemotherapy has been the standard of care for high-grade B-cell NHL. the dose of cisplatin is limited by renal toxicity. However. capecitabine is metabolised to the active drug fluorouracil by thymidine phosphorylase. the cell-cycle non-specific drugs interact with cells in all parts of the cycle and they can be cytotoxic to more slowly proliferating tumour cells. killing larger numbers as cells move through to the S phase. cyclophosphamide. Bleomycin carries a risk of pulmonary toxicity. but their durations of response are generally short. In contrast. The common cell-cycle non-specific drugs such as the alkylating agents and the antitumour antibiotics have activity at all phases of the cell cycle. The combination of rituximab with CHOP chemotherapy is now standard . etoposide. where the extended use of cell-cyclespecific drugs can lead to major problems with neutropenia and mucosal damage. but it has no significant pulmonary or renal toxicity. bleomycin. to CHOP has brought an increase in the eventfree survival at 5 years from 29% to 47% (Feugier et al. Chemotherapy scheduling and regimens Combination chemotherapy regimens When cytotoxic drugs are developed and licensed. two myelosuppressive drugs. r There are no clinically important drug interactions between the agents. new drugs are generally combined with others into combination chemotherapy schedules. the mode of action of many chemotherapy drugs has been divided between those termed ‘cell-cycle specific’ and those that are ‘cell-cycle non-specific. Over the past 20 years. Rituximab has a completely different mode of action from that of conventional cytotoxic drugs and it rarely causes significant side effects. but it also causes minimal myelosuppression. vincristine. In general. Since its introduction in the 1970s.. the best results from chemotherapy are usually with a combination of drugs. Cell-cycle specificity Historically. and etoposide is highly myelosuppressive. The impact of this approach to chemotherapy treatment has been central in producing an overall cure rate of greater than 80% for patients with advanced testicular cancer (Williams et al. In this regimen. cyclophosphamide and doxorubicin. cisplatin regimen used for the treatment of advanced testicular cancer gives one of the best examples of the benefits derived from combining chemotherapy agents. For example. each can be used at nearly the full single-agent dose. which is present in greater concentrations in some types of tumour cells than in normal cells. Gordon et al.

Paclitaxel also causes hypersensitivity reactions. The most frequent situation where this applies is the combination of paclitaxel and carboplatin in the treatment of patients with ovarian cancer. The original phase III trials. because of the risk of hypersensitivity. In contrast. Other formulae using both electronic and manual methods (nomograms and slide rules) are available. However. mainly because of the need for a solubiliser to allow the drug to dissolve. Another monoclonal antibody. which were completed before licensing. Finally. and so the final volume of the infusion is also critical. 2002). When used in this sequence. However. in routine practice. patients showed a more profound myelosuppression and an approximately 20% decrease in paclitaxel clearance. Carboplatin is a cell-cycle non-specific drug. 1916). paclitaxel is administered before carboplatin. the most commonly used is that of DuBois and DuBois. Attempts to reduce this to 1 hour have resulted in a number of hypersensitivity reactions. which interestingly dates from 1916 and was based on data from only eight adults and one child (DuBois and DuBois. and most centres continue with a 3-hour infusion. and so the method used may vary among centres but there is generally good correlation among the methods. which would indicate administration using a 28day cycle. administered the drug over 24 hours.. it would be suited to bolus administration in a single large dose. These regimens should be familiar to the health professionals who dispense and administer them. there are some situations where drugs must be given in the correct order. and it represents the first major change to highgrade NHL management in 15 years. It too has stability problems and it needs administration through non-PVC lines. Calculation of doses Body surface area The ideal method of calculating a suitable dose of a cytotoxic drug would take into account its pharmacokinetic properties – the ability of the body to deliver the drug to its site of action and the subsequent metabolism and excretion. routine cytotoxic chemotherapy doses continue to be calculated according to the patient’s body surface area (BSA) (Veal et al. in studies when paclitaxel was given after carboplatin. thus. Paclitaxel is cell-cycle specific and so ideally it should be given in multiple fractions over a prolonged period. It is also potentially unstable. Protocols and guidelines The introduction of peer-reviewed treatment policies within the National Health Service (NHS) has led each cancer network and NHS Trust that treats cancer to have formal protocols of their approved chemotherapy regimens. Scheduling and administration of chemotherapy Generally. and ‘off-protocol’ regimens should not generally be prescribed unless there is good evidence in the research literature to do so. and there is a realistic expectation that this will translate to higher overall survival figures as the trial data mature (Piccart-Gebhart et al.. 2003). although more precise and individualised approaches to chemotherapy drug dosing are often advocated. more recent studies such as the International Collaborative Ovarian Neoplasm (ICON) 3 trial have used a 3-hour infusion (ICON Group. especially in poly(vinyl choride) (PVC) containers. Combining the two drugs presents a dilemma about the length of the cycle. this presents logistical problems. Although the precise sequence of giving drugs in many regimens is relatively unimportant. is bringing about exciting results in the treatment of breast cancer. 2005. antihistamine and H2 antagonist. it is given as an infusion over a minimum of 30 minutes rather than as a bolus. Early results suggest that the addition of trastuzumab to conventional adjuvant chemotherapy results in an approximately 50% decrease in the risk of relapse measured at two years. The dose of the drug could then be adjusted further depending on the actual toxicity seen in each patient. 3 . There are several formulae for calculating BSA.Practical issues in cytotoxic chemotherapy usage practice. 2005).. The myelosuppression nadir from carboplatin is between 14 and 21 days. the safety profile is consistent with that reported for single-agent use. indicating a maximum cycle length of 21 days. the scheduling and administration of chemotherapy follow the protocols used in the original clinical trials. The nadir of myelosuppression from paclitaxel occurs after 10 days. but trials have shown that a 21-day cycle does not produce unacceptable myelosuppression and it removes the theoretical possibility of tumour growth attributed to suboptimal scheduling of paclitaxel. However. with a premedication of a corticosteroid. thus. trastuzumab. Romond et al.

1989). or has had previous anthracycline exposure or mediastinal radiotherapy. carboplatin is the only one to have its dose calculated directly according to the renal function. care should be taken when prescribing for tall. some of the newer drugs. Pretreatment investigations and checks Before initiating chemotherapy Informed consent Information on individual drugs or regimens is usually available in a written form and should be used to supplement verbal information. with this drug. Renal function An accurate measurement of renal function. the cisplatin in the regimen can generally be amended by substituting carboplatin. such as an EDTA clearance. such as the monoclonal antibody trastuzumab. for which the dose is calculated according to the GFR. and a formula has been developed (the Calvert equation) which is based on renal function (Calvert et al. is recommended. the creatinine clearance (CrCl) can be calculated from the serum creatinine using the Cockcroft–Gault formula as follows: CrCl = F × (140 − age in years) × weight (kg) Serum creatinine(μM/l) F = 1. dose capping is frequently used. The Calvert equation was developed using the more accurate EDTA clearance as the measure of GFR. and many institutions will use 2. In addition to an explanation of the purpose of treatment and the adverse effects of the drugs. the dose is usually capped at 2 mg. are calculated on body weight alone. such as a MUGA scan. a correction. patients should be given clear advice about monitoring for.1. so although the dose is 1. If there is evidence of a significant decline in renal function. The desired AUC (area under the curve of serum levels against time) is chosen.. and action on. However. Accurate assessment is also recommended for treatments using cisplatin. Repeated monitoring of cardiac function should be performed according to the protocol being followed. Flat dosing Bleomycin is the only commonly used cytotoxic drug for which a fixed dose is employed. . Cardiotoxic drugs For patients on cardiotoxic drugs such as doxorubicin or the other anthracyclines. 4 Doses will need to be amended for drugs that are excreted renally if there is a reduced GFR.4 mg/m2 in the CHOP regimen. a fixed dose of 30 000 units is used irrespective of the patient’s BSA.nlcn. is generally required for carboplatin. To place an upper limit on drug dose. and care should be taken to avoid exceeding the lifetime dose recommendations if more treatment is given at a later date. This must include 24-hour contact telephone numbers at the hospital. is elderly. especially if the patient has a history of cardiac disease. Information on the appropriate dose reduction in renal impairment will be found in most protocols or national guidelines. should be applied. An exception to the 2. suspected neutropenic fever and other serious adverse effects of chemotherapy. usually at an AUC of 4 or 5. such as multiplying by 1.23 (males) Body weight dosing Body weight alone is not used often in calculating doses of cytotoxic drugs. php). pretreatment cardiac assessment. However.Sian Evans and Philip Savage Dose capping Using the calculated BSA in large and obese patients may lead to relative overdosing of chemotherapy with an associated risk of excess toxicity.2 m2 cap is with the use of vincristine. This drug is excreted unchanged by the kidneys. In the treatment of testicular cancer. Area under the curve (AUC) dosage Among the commonly used chemotherapy drugs.2 m2 . all but the smallest patients receive a capped dose of 2 mg. because this drug is significantly nephrotoxic.04 (females) or 1. For other drugs and for continuing assessment of renal function. If the GFR is calculated from serum creatinine using the Cockcroft-Gault formula (see following text).uk/professional.2 m2 as an upper limit for curative and adjuvant treatments and 2 m2 for palliative treatments. and the dose is calculated by the following formula: Dose (mg) = desired AUC × (GFR ml/min + 25) The GFR is the glomerular filtration rate. non-obese individuals because there is a potential risk of underdosing if the BSA is capped at 2. such as those available online from the North London Cancer Network (www.nhs.

Significant neutropenia or thrombocytopenia will mean a treatment delay. bilirubin.. even when there is hepatic impairment. before each course of chemotherapy. alone or accompanied by slight increases in transaminases.Practical issues in cytotoxic chemotherapy usage Hepatic function Most drugs undergo some metabolism by the liver. do not usually require dose reductions. measuring these can be a rapid. cisplatin increases the renal excretion of potassium and magnesium and oral supplementation is required frequently. including doxorubicin. The capacity of the liver to handle drugs. transaminases and alkaline phosphatase should be reviewed before treatment. This can involve direct physical measurement of the tumour. Biochemical. height and weight measurements are needed. which is excreted in the bile. depending on the disease site. Before each cycle Full blood count The patient’s full blood count should be taken close to the actual day of administration. biochemical tests or measurement of tumour markers. line infections. some drugs. other sources of advice include the websites of the BC Cancer Agency and the North London Cancer Network. In response to deteriorating renal or hepatic function. Irinotecan. Patients who are anaemic rarely require a delay in chemotherapy and can be transfused if their haemoglobin level drops below 9. Tumour markers are most useful in gestational 5 Height and weight To calculate BSA. Some drugs have specific toxic effects that require monitoring. renal. has to be dose-reduced in the presence of elevated serum bilirubin. for example. but elevation of bilirubin. and 19%. and again if there is reason to suspect that it has altered by more than 5%. However. some drugs may need dose reduction or a change to alternative therapies. is large. the patient’s weight should be checked regularly. 2000). Patients who are admitted with neutropenic fever or who have had more than one delay in treatment during a course of chemotherapy will require a dose reduction if receiving palliative treatment or support with growth factors if they are receiving curative or adjuvant treatment. and the liver function tests. simple and economic method to monitor the response of the tumour to treatment. in this situation. For example. one should also monitor symptomatic benefit carefully and balance this against the treatment toxicity. do need dose reductions in the presence of hepatic impairment. using a Hickman® or PICC line. smaller changes in weight have a greater impact on the dose and. Central lines Patients who have chemotherapy through ambulatory infusion devices must have central access before treatment. The patient’s body weight should be measured . liver and bone profile A full biochemical profile is required before treatment to ensure that there has been no significant change in renal or hepatic function due to either the treatment or the tumour. The treatment protocols in most units include advice on appropriate dose reductions.5 g/dl or if they develop symptoms of anaemia. if possible. Where the principal aim is palliative. Tumour markers and other tests for response If a tumour secretes circulating tumour markers. Some patients who presented with disease-related hepatic or renal impairment may no longer need a dose reduction if these parameters return to normal with effective treatment. Increases in the alkaline phosphatase. ideally on the day of treatment or the day before. usually requires the dose reduction of drugs that are metabolised in the liver. every three months. particularly if accompanied by increases in transaminases. Patients with poor veins or those who are to receive multiday infusions will also benefit from central lines early on in their treatment. The typical practice in the United Kingdom (UK) is to add 20–40 mM of potassium and about 8 mM of magnesium per day to hydration fluids. Baseline assessments of tumour With the exception of adjuvant treatment. Patients should therefore be monitored regularly for these problems (Minassian et al. studies have shown that up to 11% develop line-related thromboses. Although many patients do not experience any problems with these lines. and the need for dose reductions is relatively uncommon. Where the dose is calculated on the body weight alone. radiological examination. a baseline measurement and regular objective measurement of response are required to assess whether the patient is benefiting from chemotherapy. or if the body weight is thought to have changed by more than 5%.

shock and end-organ failure. Although the nadir of neutropenia varies among drugs (e. it may be cost effective when compared to the costs of a hospital admission for neutropenic fever. Volume replacement with intravenous fluids must be initiated. and the increasing evidence that prophylactic growth factors are clinically and economically effective. urinary tract infection or central-line sepsis. However. Studies have demonstrated Neulasta® to have been more successful in preventing neutropenic sepsis than daily G-CSF and to have improved patient tolerability. 2005). and junior medical staff must be made aware of the need to contact more senior staff. should generally have dose reductions made to their treatment. if necessary. and the nadir for mitomycin C is several weeks after treatment). and the current ASH/ASCO guidelines recommend this approach for regimens with a greater-than-20% risk of neutropenic sepsis in the first cycle (Smith et al. from cycle 1) for patients on very myelosuppressive regimens. These can be used to monitor response or to indicate a change to second-line therapy when the rate of fall is inappropriately slow (Toner et al. Major toxicities and their management This section looks at practical issues of managing chemotherapy toxicities. approximately 60% of tumours make one or both of HCG and alpha feto-protein. which requires only a single administration. In these patients. 11). Myelosuppression Neutropenic fever and neutropenic sepsis The most frequent serious complications of chemotherapy treatment are neutropenic fever and neutropenic sepsis. such as chest infection. 2006).g. are leading many centres to incorporate their use into routine protocols. generally 24 hours after chemotherapy administration. or those who have persistently low neutrophil counts without fever. and they provide an excellent method of following response to treatment. Neutropenic fever and sepsis are also considered in Chapter 6 (see p. . apart from the pyrexia.. reduction in the incidence of admissions for neutropenic fever for patients with solid tumours who were given a prophylactic quinolone antibiotic (Cullen et al. this is not yet standard practice in the UK.. Primary and secondary prophylaxis Patients receiving palliative chemotherapy who are admitted with neutropenic fever. but this has been overcome with the introduction of pegylated G-CSF (Neulasta® ). Other tumour markers are also used in the assessment of some other malignancies. In contrast. should also receive an antibiotic with appropriate additional cover. The treatment plan with intravenous antibiotics will usually include an aminoglycoside and an antipseudomonal antibiotic or a cephalosporin and vancomycin. In regimens that cause prolonged myelosuppression. and consideration should be given to transfer to a high6 Prophylactic antibiotics The Significant trial showed a small. Patients with neutropenic fever or neutropenic sepsis must be treated promptly using the local policies for the empirical treatment.. where human chorionic gonadotrophin (HCG) is constitutively produced by all tumours. as secondary prophylaxis. All patients with neutropenic sepsis can deteriorate very quickly. patients receiving curative (including adjuvant) treatment should receive G-CSF prophylaxis. dependency or intensive-care unit. to prevent further episodes of neutropenia and to maintain dose intensity. Although Neulasta® is more expensive than daily G-CSF. for advice on the management of such patients. such as cancer antigen 125 in ovarian cancer and prostate-specific antigen in metastatic prostate cancer. In advanced testicular cancers. Practical issues with the timing and administration of growth factors can be a problem. Patients with an apparent site of infection. 1990). The data from this study. The main toxicities of individual chemotherapy drugs are summarised in the Appendix (see p.Sian Evans and Philip Savage choriocarcinoma. the risk of neutropenia should be considered at all times during the course of chemotherapy treatment. and they will respond to treatment with intravenous or oral antibiotics. a number of patients can either present or rapidly become seriously unwell with hypotension. 77). but definite.e. patients on docetaxel can become neutropenic two to three days after treatment. rapid assessment and management are essential. patients will benefit from the use of cotrimoxazole to reduce the risk of pneumocystis carinii infection. More recently. However. Most patients with neutropenic fever will initially appear well. there has been a move towards primary prophylaxis (i. particularly lymphoma regimens with long-term steroid administration.

The management of this problem involves considering a benzodiazepine before treatment and/or on the previous evening. 1999). However. and they are used in a step- .org. vindesine or gemcitabine. but they can be reassured that. Cisplatin is the drug that is most frequently linked to renal toxicity. particular attention is needed for drugs that are either renally toxic in themselves or are excreted by the kidneys. newer drugs. In the case of doxorubicin. These factors may require increased prophylaxis above those recommended here. and it is well tolerated.nice. Erythropoietin is also considered in Chapter 2 (see p. Delayed nausea and vomiting Nausea and vomiting The problems of chemotherapy-associated nausea and vomiting have become far less since the introduction of the 5-HT3 antagonist drugs such as ondansetron and granisetron. being female and having a chronic. Good antiemetic prophylaxis. with appropriate use of antiemetics. Predisposing factors The predisposing factors for an increased risk of nausea and vomiting include the previous poor control of nausea/vomiting. Cardiac function should be monitored more closely in patients with cardiac problems or in patients who have received previous treatment with anthracyclines or mediastinal radiotherapy.. 20). Anticipatory nausea and vomiting Anticipatory nausea and vomiting occur up to and during administration of chemotherapy. A blood transfusion is standard practice.. being of a younger age. and the use of aprepitant or palonosetron with subsequent cycles. 2005). For patients who continue to have problems. Oral metoclopramide or domperidone are usually recommended for drugs of low emetogenic potential. Lorazepam 0. 79).uk).5 to 1 mg sublingually or orally is the drug of choice. Erythropoietin may be indicated. a neurokinin-1 receptor antagonist. Renal toxicity Although renal function is monitored before each chemotherapy cycle in most regimens. low alcohol intake (Gralla et al. this is now rare. generally a 5-HT3 antagonist and dexamethasone. cause a gradual reduction in haemoglobin levels over a course of treatment. and are mainly due to the psychological effects associated with previous treatment. Cyclizine may be substituted for metoclopramide for the treatment of delayed nausea and vomiting or during prolonged oral regimens if metoclopramide is ineffective. Anaemia Some commonly used cytotoxic drugs. but it is expensive and under review by the National Institute for Health and Clinical Excellence (www. fluorouracil is a good example that is linked with cardiac ischaemia and arrhythmias. Appropriate treatment modifications should be made if there is a significant rise in the serum creatinine while on 7 Acute nausea and vomiting Acute nausea and vomiting are defined as occurring up to 24 hours after chemotherapy administration. a history of motion sickness. the majority of drugs and regimens require more powerful antiemetics. A number of other drugs can also occasionally cause cardiotoxicity. wise fashion (Herrstedt et al.Practical issues in cytotoxic chemotherapy usage Prophylactic antibiotics are also considered in Chapter 6 (see p. this is 450 mg/m2 . Other dose-limiting toxicities Cardiotoxicity Cumulative cardiac toxicity is a problem associated with the anthracyclines. may also prevent delayed nausea and vomiting. Many new patients still expect nausea and vomiting to be a major problem. The drugs used in prevention depend on the emetogenic potential of the regimen. but it can dramatically affect a patient’s quality of life. are recommended in addition to 5-HT3 antagonists and corticosteroids. The 5-HT3 antagonists are not generally effective in treating delayed nausea and vomiting. including cisplatin. This does not usually need a dose reduction or delay in treatment. and treatment should remain generally within the standard guidelines on total lifetime dose. All chemotherapy units will have guidance on which level of antiemetics to use with each chemotherapy drug or regimen. such as bleomycin. The incidence of delayed nausea and vomiting is increased when there is poor control of the acute phase. including aprepitant. on the day of treatment and for one to two days afterwards.

but it is routine good practice to offer sperm storage for men undergoing chemotherapy. vary in detail among the home countries. epirubicin and docetaxel. More detailed information on cancer treatment. the risk of relapse needs to be taken into account when giving patients advice about the timing of future pregnancies. The situation for women is less satisfactory because techniques for the preservation of oocytes or ovarian tissue are not yet reliable or widely used. in the event of extravasation. but effective prevention is difficult. the alkylating agents and vinca alkaloids.. Hodgkin lymphoma. there are formal standards for the safe prescribing. Diarrhoea can also occur as part of the cholinergic syndrome. Embryo storage is time-consuming and may cause a significant delay in starting treatment. It is important to recognise the symptoms early and to start treatment with fluids.Sian Evans and Philip Savage therapy. These drugs should always be given as bolus injections via fast-flowing drips or through a central line. All prescribers must be aware of the local arrangements for administering chemotherapy and they should refuse to undertake any procedure for which they have not been trained. All healthcare staff involved in chemotherapy must be aware of their local policies. pending a more accurate assessment of renal function such as an EDTA clearance or 24-hour CrCl. Training. High-dose methotrexate can also cause renal toxicity. dispensing. whereas cyclophosphamide and ifosfamide can cause both renal toxicity and haemorrhagic cystitis. Intrathecal chemotherapy must be prescribed and administered only by trained specialist registrars. The regimens used in the treatment of testicular cancer. dispensing and administration of chemotherapy. Fertility and foetal abnormalities (contraception) Chemotherapy can affect the patient’s fertility. and methods of auditing these. and high-grade NHL tend to have a relatively modest impact on fertility. Avoidance of heat can help to prevent PPE. competency and adequate facilities are the keys to safely prescribing. staff grades and consultants. Intrathecal chemotherapy There have been a number of fatalities from the inadvertent administration of vinca alkaloids by the intrathecal route. The problem can be minimised to some extent by the use of scalp hypothermia in patients receiving bolus injections or short infusions of doxorubicin. the incidence of foetal abnormalities born to patients who have previously completed 8 Extravasation A number of cytotoxic drugs are vesicant and. they can cause local tissue necrosis. Diarrhoea Diarrhoea can be a dose-limiting toxicity with the fluoropyrimidines such as capecitabine and 5-FU. Patients are advised to defer pregnancy for 12 months after the completion of treatment but there is little evidence to say whether or not this is too cautious. Patients should be encouraged to use emollients. and administering chemotherapy. Phlebitis and local reactions Phlebitis is a common problem with irritant drugs such as dacarbazine. rehydration salts and loperamide. but it is not clear how effective it is. Pyridoxine 50 mg three times a day is often used as treatment. 2006). many chemotherapy patients do develop significant alopecia and they need to be aware of this possibility. A dose delay is usually required for grade 2 or greater PPE. Surprisingly. In addition to the potential risk of foetal abnormalities. There are a . Arrangements should be offered for wigs and other cosmetic supports such as head scarves. However. Other toxicities Alopecia Alopecia can be a very distressing side effect of chemotherapy treatment for some patients. chemotherapy appears to be similar to that in the normal population. The patient may report pain on injection. which is seen with irinotecan. but every organisation that provides chemotherapy has the responsibility to maintain policies and procedures to ensure that these standards are met. The standards. Palmar–plantar erythrodysaesthesia (PPE) PPE is the dose-limiting toxicity with capecitabine and liposomal doxorubicin. chemotherapy and fertility is readily available (Lee et al. but there may not be any obvious local reaction. Administration issues Safe administration of chemotherapy In the UK.

this involves administration in a hospital setting at a time when staffing levels and skill mix are appropriate to deal with emergencies. can be very distressing for the patient. such as diarrhoea. Full resuscitation facilities must be available for patients receiving first doses of agents that are known to cause hypersensitivity.. and oxygen must be available (see Chapter 6. N. N. Steven. Platinums and taxanes are the cytotoxic drugs that are most commonly implicated. Treatment is symptomatic. Careful patient education is essential and treatment must not be started until the prescriber is sure that the patient fully understands how to take the medication. J. H. Premedication with corticosteroids.. Patients must be educated to recognise these symptoms and to take appropriate action. following established protocols and patterns of administration.. As a result. DuBois. At present. Intern. Generally. and possibly antibiotics. E. the circumstances when treatment should be discontinued. Stanley.Practical issues in cytotoxic chemotherapy usage number of general and individual drug-specific measures for the treatment of suspected extravasation and the problem should be dealt with as an emergency (see Chapter 6. A formula to estimate the approximate surface area if height and weight be known.. their use in the modern multidisciplinary cancer centres.. R. p. 80). 988–98. Cholinergic syndrome Cholinergic syndrome is seen with irinotecan. 4th edn. Billingham. Arch. Prolonging infusion times has been used successfully to reduce the rates of recurrence of oesophageal– pharyngeal syndrome. (2005).. most patients with high-grade NHL. Gumbrell. sweating and diarrhoea. and DuBois. and patients can be reassured if they are warned in advance that this symptom is transient. Subsequent doses may be given at other locations depending on the risk of future reactions. 79). and most monoclonal antibodies. (1916). Carboplatin dosage: prospective evaluation of a simple formula based on renal function. warm drinks often help in mild cases. the introduction of new targeted anticancer drugs such as the monoclonal antibodies and kinase inhibitors will be the main area of change in cancer treatment. and how to obtain help. who may confuse this symptom with respiratory or cardiac arrest. L. A. C. A. J. rehydration. 17... It is likely that these drugs will be increasingly combined with existing therapies not only to produce enhanced results but also to produce differing patterns of side effects that will require ongoing education and training.. Calvert. L. the use of supportive medication. D.. D. Treatment is with atropine. Long-term results . p. Feugier. results in relatively little unpredictable toxicity. can produce hypersensitivity reactions that can range from a ‘flu-like’ syndrome to anaphylactic shock. Oral chemotherapy and overcompliance With oral chemotherapy. (2005). Oxford: Radcliffe Medical Press. A. Although these drugs are highly toxic. the adjuvant treatment of breast and bowel cancer has increased the overall cure rates of these illnesses by at least 10%. and Wright. Sebban. and it is characterised by flushing. F. et al. Oncol. of the R-CHOP Study in the treatment of elderly patients with 9 . 7. Antibacterial prophylaxis after chemotherapy for solid tumors and lymphomas. and premedication with this is recommended for subsequent cycles. Oesophageal–pharyngeal syndrome This is seen with oxaliplatin and. Newell. 353. Late-onset diarrhoea (more than 24 hours postinfusion) is also seen. overcompliance is often a problem when capecitabine is involved: patients may ignore the onset of dose-limiting adverse effects. 1748–56. P Van Hoof. Clin. P (2002). The detailed treatment of anaphylaxis will depend on local policy but will include epinephrine (adrenaline). and then continue treatment and become dehydrated. although rare. A. Avoidance of cold liquids and food and not exposing the body to sudden cold will usually prevent this syndrome. et al. Handbook. Hodgkin lymphoma or testicular cancer are treated with the expectation of cure. Med. et al. M. The Cytotoxics . Hypersensitivity and anaphylaxis Certain cytotoxic drugs. 863–71. Most additional information on the safe and effective use of the current drugs is readily available within each unit or on the expert websites. and it is treated with loperamide. M. Med. REFERENCES Allwood. Engl. Summary The development of effective chemotherapy treatments has been one of the great successes of modern medicine. antihistamines and paracetamol may be recommended. depending on the expected reaction. Alongside this. Cullen. (1989).

739–45. and either vinblastine or etoposide. C. Longterm central venous access in gynecologic cancer patients. Gordon. 1659–72. Med. J.. OH: Lexi-Comp. A. Williams. 24. J.. (2005). Leyland-Jones. 1435–40. D. 353. 1342–9. H. 330. Minassian.. Tan. Prognostic significance of received relative dose intensity in non-Hodgkin’s lymphoma patients: application to LNH-87 protocol. S. 23. Aapro. A. Oncol. The Cytotoxics . V.. C. Cancer Res. N.. Ann. 50. E. Herrstedt. node-positive breast carcinoma. 505–15. Osoba. M. et al.. J. Engl. C. (2005). J. Hudson. Romond. R. 360. and Boddy. J. M. et al. S. 149–56. Andersen. (2005). Oxford: Radcliffe Medical Press. Korzun. G. R. 2006 update of recommendations for the use of white blood cell growth factors: an evidence-based clinical practice guideline. Oncol. clinical practice guidelines.. Clin. 24. Med. V. 1253–9. Australian and New Zealand Germ Cell Trial Group. et al. 2917–31. Comparison of two standard chemotherapy regimens for good-prognosis germ-cell tumours: a randomised trial. J. Coulthard.. Harrington. E. Toner. (2002). H. Gaynor. 328. J. N.. Lepage. J. B. and cisplatin in women with ovarian cancer: the ICON3 randomised trial. (2003). 403–9. Oncol. Practical Chemotherapy – A Multidisciplinary Guide.. Lancet. et al. L.. et al. Oncol. M. T.. New Drugs. 651–6. et al. E. i77–9. Piccart-Gebhart.. Oxford: Radcliffe Medical Press. et al.. Invest. J. I. 316. Perez. American Society of Clinical Oncology. E. M. ICON Group. 357. S. Engl. Med. 191. J. A. Lowe.. J. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. 1673–84. 4117–26. Wood. Einhorn. Paclitaxel plus carboplatin versus standard chemotherapy with either single-agent carboplatin or cyclophosphamide. Practical Chemotherapy – A Multidisciplinary Guide. Recommendations for the use of antiemetics: evidence-based. Engl. P (2002). Lyman. R. Med. Lee. Comparison of a second-generation combination chemotherapeutic regimen (m-BACOD) with a standard regimen (CHOP) for advanced diffuse non-Hodgkin’s lymphoma. et al. Stanley. (2003).. J. M. 5904–10. A. 17.. R. C. 4th edn.. L. (1993). Stockler. (2006). J. 16 (Suppl. Smith. FURTHER READING Allwood. Roila. K. P. Handbook. et al. Med. Geller. Chemotherapy individualization. C. Coll. Fisher. 1). Birch. Oncol. Dose and dose intensity of adjuvant chemotherapy for stage II. H. et al. Haioun. Bryant. et al. Oncol. J. A. and Daniels. (1993). R. Oxford: Radcliffe Medical Press. Toner.. Schover. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin’s lymphoma. Serum tumor marker half-life during chemotherapy allows early prediction of complete response and survival in nonseminomatous germ cell tumors. 10 . H. 4. and Wright. (1992). M. J. Med. American Society of Clinical Oncology recommendations on fertility preservation in cancer patients. (2001)... et al. Clin. et al. N... W.. S. Treatment of disseminated germ-cell tumors with cisplatin. Drug Information Handbook for Oncology. S. Clin. The GELA (Groupe d’Etude des Lymphomes de l’Adulte). (2005).. (1999). et al. C. J. M. D. Engl.. Procter... L. R. Budman. N. and Daniels.. Sood. (1990). (2003).. Khatcheressian. Clin. D. I. Boyer.... G. Kris. bleomycin... N.. Veal. R. Dahlberg. Engl. S. A.. J.. (1987). (2006). J. Engl. J. (2000).. G. G. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. S. Am. 5th edn. ESMO minimum clinical recommendations for prophylaxis of chemotherapy-induced nausea and vomiting (NV). 1002–6. Summerhays. A. H.. doxorubicin. Lancet. M. M. L.. N. D. Gralla. Partridge. A. Surg. D.. Gisselbrecht. Ann. (1994). Summerhays.Sian Evans and Philip Savage diffuse large B-cell lymphoma: a study by the Groupe d’Etude des Lymphomes de l’Adulte. 353.. G. A. F. 21. Solimando. 2971–94.. J. 327. N. 3187–205. J.

give by bolus injection or centrally. Cardiotoxic. Hepatotoxic. Vesicant. Pulmonary. Renal. Avoid in high dose. Fluorouracil Gemcitabine Methotrexate + + + + + + Diarrhoea. dose usually 10 mg. ++ ++ ++ Haemorrhagic cystitis. Exfoliant. As above. Neurotoxic. Needs mesna routinely to prevent haemorrhagic cystitis. Capsules 40 mg. + Renal. As above. Neurotoxic. Vesicant. +++ +++ + + Cardiotoxic. Anthracyclines Doxorubicin Epirubicin Idarubicin +++ +++ +++ ++ ++ ++ Depends on schedule. +++ Delayed myelosuppression. ++ Must be given by intravenous injection only. Vesicant. Encephalopathy. Vinblastine + Minimal. +++ + PPE. Alopecia. Cancer treatment drugs and their major toxicities Emetogenic Drug/class Alkylating agents Cyclophosphamide Ifosfamide Lomustine ++ ++ +++ Delayed myelosuppression. Renal toxicity in Note drug interactions. Exfoliant. Haemorrhagic cystitis. Mucositis. Max. Diarrhoea. Max. (continued ) PPE with long infusions. Minimal. Cardiotoxic. Liposomal doxorubicin Mitomycin Mitozantrone Dacarbazine Antimetabolites Capecitabine Fludarabine + +++ + + PPE. If blood transfusion required by patient. Vinca alkaloids and etoposide Vincristine Minimal. As above. patients with pleural effusions and/or ascites. Myelosuppression potential Other major toxicities Notes 11 . Consider PCP prophylaxis. +++ Must be given by intravenous injection only. Tablets 500 mg and 150 mg. Vesicant. As above. Influenza-like reactions after infusion. Vesicant. need irradiated blood. May need mesna. Oral form available.Practical issues in cytotoxic chemotherapy usage Appendix. dose 2 mg.

Oral form (large capsules) twice i. May need atropine. dose. ++ ++ ++ Less nephrotoxic than cisplatin. dose 60 mg. Neurotoxic. Weak MAOI. Oral form available. ++ + Minimal. PCP = pneumocystis carinii pneumonia. Max. Other major toxicities + Neurotoxic. avoid alcohol. Some alopecia. Alopecia. Carboplatin Oxaliplatin ++ Especially platelets. Vesicant. +++ ++ Pulmonary (fibrosis). Vesicant. Check renal function carefully. MAOI = monoamine oxidase inhibitor. Vesicant.and posthydration essential. Neurotoxic. PPE = palmar–plantar erythrodysaesthesia. (Continued ) Emetogenic Drug/class Vinorelbine Myelosuppression + potential Minimal. Skin. Hypersensitivity reactions. Taxanes Paclitaxel ++ + Alopecia. Cholinergic syndrome. Hypersensitivity reactions – needs premedication.v. Exfoliant. good pre. Etoposide ++ + Alopecia. Exfoliant. i. Neurotoxic.Sian Evans and Philip Savage Appendix. Vesicant. Notes Must be given by intravenous injection only. Hypersensitivity reactions common. Platinums Cisplatin + +++ Nephrotoxic. Docetaxel +++ + Alopecia.v. 50 mg capsules. 12 . ++ +++ ++ + Diarrhoea. Topoisomerase 1 inhibitors Irinotecan Topotecan Other cytotoxics Bleomycin Dacarbazine Procarbazine No. Exfoliant. Hypersensitivity reactions – needs premedication.

where chemotherapy is not very effective – biological agents are the main treatment after surgery.medicines. Renal cell cancer – metastatic disease In phase II trials of high-dose IL-2 there was a response rate of 14% (CR 5%. thereby initiating a number of intracellular signalling cascades that change the functioning of the effector cell. bolus. 1995). Interleukin-2 (IL-2) Background r IL-2 was first identified by its action as a T-cell growth r It is released by T-helper cells in response to antigen presentation and interleukin-1. r Monoclonal antibodies. protocols had higher complete-response rates than continuous infusion. Fyfe et al. Over the next decade. 2004). based on the molecules’ presumed function. For patients with some tumours – such as renal cancer. A systematic review of various IL-2 schedules showed that bolus i.org. It should not be given to patients who are unlikely to respond. There are five recognised major cytokine families: 13 . it is likely that biological therapies will significantly change the management of cancer patients. and it is involved in activation of both T-helper cells and cytotoxic T-cells.v. IL-2 is licensed for use in metastatic renal cell carcinoma. the licensed indications for new drugs is also changing. although chemotherapy has improved survival rates it has been at the expense of more toxicity. r The immunoglobulin supergene family.2 BIOLOGICAL TREATMENTS IN CANCER Rachel Jones and Robert Leonard Introduction Biological agents are becoming increasingly important in the management of cancer patients..v.uk/). and there are several different classification systems that have been used. The median duration of response was 19 months for patients who achieved a partial response. In other solid tumours. r Vaccines. factor in 1976. This group of molecules is complex and increasing in number. such as those with ECOG performance status of 2 or greater or those with ECOG performance status 1. but 4% of patients died of probable treatment-related toxicity. This chapter will concentrate on five areas: r Cytokines. and half the responding patients had responses that lasted for 36 months or more (Baaten et al. their cell of secretion. Because this is a rapidly changing field. only interleukin-2 and interferon alpha have an established role in the management of cancer patients. r Tyrosine kinase inhibitors. They work by binding to specific effector-cell surface receptors. r Haemopoietic colony stimulating factors. The duration of response was highest for i.c.. r Type I cytokines (interleukins). which are able to lyse fresh tumour suspensions. Although there are over a hundred cytokines. although it was difficult to draw firm conclusions because of the heterogeneity of the data. Currently. their target of action or their structure. with more than one site of metastatic Cytokines Cytokines are proteins that are secreted by haemopoietic and non-haemopoietic cells and which play an important role in the immune system. and s. r The chemokines. a useful resource for up-to-date information on licensed drugs can be found in the electronic Medicines Compendium (www. r In vitro incubation of lymphoid cells with recombinant IL-2 leads to the generation of lymphokineactivated killer (LAK) cells. Increasing evidence supports the use of biological agents both as single agents and in combination with chemotherapy. r The tumour necrosis factor family. PR 9%. r Type II cytokines (interferons).

In view of its significant toxicity.. the toxicity associated with this regimen was severe and 2% died from sepsis. Haematological malignancies IFNα has an established role in some haematological malignancies.3 months. and interferon gamma is a type 2 interferon. despite full accrual (Clark et al. Lowdose IFNα. a review of eight clinical trials. endothelial cells and osteoblasts. but with no consistent overall survival data and the associated toxicity.and low-dose treatment. Of the three types. Currently. Five risk factors are associated with short survival: low Karnofsky performance status. Previous nephrectomy and Karnofsky performance status greater than or equal to 80 may be associated with better survival. However. beta and gamma.9 months and overall survival of 11. and it is associated with capillary leak syndrome. 1000 patients (Wirth. is licensed for use in the UK as an adjuvant treatment for melanoma. pulmonary oedema. In metastatic melanoma. fever and death. these factors may help select patients most likely to benefit from IFNα treatment. However. Renal cell carcinoma – metastatic disease IFNα has a response rate of 14% in metastatic renal cell carcinoma. patients need to be selected carefully and only treated in centres with experienced clinicians and facilities adequate enough to provide haemodynamic support. The duration of remission was rarely over two years. fibroblasts. PR 10%. Others In metastatic carcinoid. with a median time to disease progression of 3. which bind to a cell-surface receptor complex called IFNAR. In view of the toxicity associated with this treatment. hypotension. The median duration of complete responses was at least 59 months. high corrected serum calcium and an interval from diagnosis to initiation of therapy of less than 1 year (Motzer et al. particularly in hairy cell leukaemia and chronic myelogenous leukaemia. who have relapsed within 12 months of diagnosis. It acts by stimulating macrophages.Rachel Jones and Robert Leonard disease. IFNα has been used with somatostatin analogues to improve the symptoms associated with hormone secretion in patients who have become resistant or refractory to somatostatin analogues alone. Melanoma – adjuvant setting The use of IFNα in the adjuvant management of melanoma is controversial because of conflicting survival results in trials that look at both high. and its use has not become widespread (see also Chapter 36. IFNα-2a is licensed for use at a starting dose of 3 MU three times a week. no effective adjuvant treatment is presently available. building up to 18 MU three times a week if tolerated. Because of the toxicity and duration of treatment. 423). as shown by a review of more than 14 . natural killer cells and cytotoxic T lymphocytes. 1986). low haemoglobin. including T and B lymphocytes. A randomised trial involving patients with resected locally advanced or metastatic renal cell carcinoma on one course of IL-2 or observation was closed early because an interim analysis found the disease-free survival end point could not be achieved. Interferon alpha (IFNα) Naturally occurring IFNα is produced by many cells. the need for dose modifications was frequently reported. Melanoma – metastatic disease IFNα has been widely tested in the metastatic setting in melanoma. it has not become a standard of care in the UK. There are three types: alpha.. which included 270 patients treated with high-dose recombinant IL-2. 2003). The best responses were noted in soft tissue and small-volume cutaneous disease (Creagan et al.. 2002). high LDH. 2000). A review of three phase II trials showed a response rate of 22%. Interferons Interferons have antiviral. In recurrent or metastatic renal cell cancer. Other situations For patients with high-risk renal cell cancer. however. only interferon alpha has an accepted role in cancer management.. Toxicity of IL-2 IL-2 is a highly toxic drug. high-dose IFNα has become a standard adjuvant treatment in the USA. Atkins et al. which binds to the cell-surface receptor complex IFNGR. uncertainty exists regarding its effectiveness. antiproliferative and immunomodulatory effects. 1993). p. Interferons alpha and beta are type 1 interferons. showed an overall response rate of 16% (CR 6%. cardiac arrhythmias.

2002) resulting in imatinib being licensed for patients with KIT positive. which spans the membrane. Cytokine combination therapy The use of combination biological therapies is the subject of current clinical trials. Side effects associated with long-term therapy include hypothyroidism. Other toxicities include peripheral oedema. The recommended dose is 400 mg daily (see also Chapter 15.. frequently with different. although it can be associated with haematological toxicity. 186). depression. There does not appear to be a clear survival benefit from the addition of TNFα over melphalan alone.1 shows some tyrosine kinase inhibitors in clinical use and in trials. abdominal pain. Patients may be unable to tolerate long-term treatment. This translocation results in an abnormal ‘Philadelphia’ chromosome. which lead to a change in activity of transcription factors and to DNA synthesis.Biological treatments in cancer Toxicity of IFNα The main toxicities associated with treatment are flulike symptoms. skin rash. In untreated. They are oral preparations. fluid retention. Non-receptor tyrosine kinases transmit intracellular signals via recognition of phosphorylated tyrosines by SH domains on other binding proteins. 2002). results in RAS activation. The gene for BCR on chromosome 22 fuses with the gene for the ABL protein (a tyrosine kinase) on chromosome 9. nausea. Gastrointestinal stromal tumours are associated with a mutation in KIT which results in ligand-independent activation of the receptor. IL-2. The recently closed MRC RE04 trial randomised patients with metastatic renal cell carcinoma between IFNα monotherapy and the combination of IFNα. Previously these patients had limited options for treatment. nonoverlapping toxicities from chemotherapy. The development of tyrosine kinase inhibitors has been an important breakthrough in cancer therapy. especially in the treatment of CML. Table 2. neutropenia and reversible hepatotoxicity.. The binding of a ligand to the receptor can activate signalling pathways. diarrhoea and fatigue. 1993). 15 . The recommended dose is 400 mg daily in chronic-phase CML and 600 mg daily in accelerated-phase or blast crisis.. early chronic-phase Philadelphia-positive CML. TNFα has minimal activity in advanced cancer and is highly toxic. invasion. In receptor tyrosine kinases. Toxicity of imatinib Imatinib is generally well tolerated. Imatinib Imatinib is an oral tyrosine kinase inhibitor which acts on more than one type of tyrosine kinase. p. but now several trials have shown consistent clinical responses (Demetri et al. Imatinib has also been shown to be active against the KIT and PDGFR tyrosine kinases. fatigue. The epidermal growth factor receptor (EGFR) is a receptor tyrosine kinase. is an example of how abnormal tyrosine kinase activity can cause malignancy. They act on intracellular signalling pathways that are involved with promoting tumour growth. which allow them to be combined with chemotherapy and radiotherapy or to be given as single agents. It treats Philadelphia-positive CML by inhibiting the tyrosine kinase activity of the BCR-ABL fusion protein. major cytogenetic responses are seen in 80 to 90% of patients (Kantarjian et al. and 5-fluorouracil because of the promising results seen with the combination regimen (Atzpodien et al. Tyrosine kinase inhibitors Tyrosine kinases are enzymes that can transfer phosphate groups from ATP to a tyrosine residue on a protein. Tumour necrosis factor α (TNFα) TNFα has been studied in combination with melphalan in unresectable melanoma of the extremities. unresectable and/or metastatic malignant gastrointestinal stromal tumours. Synthesis of the resultant fusion protein BCRABL is unregulated and the ABL tyrosine kinase activity. which is found in chronic myeloid leukaemia. such as the RAS-RAF MAP kinase pathway and the PI3K-AKT pathway. through further binding to growth factor-related binding protein. The BCRABL fusion protein. The action mechanism of tyrosine kinase inhibitors is to prevent autophosphorylation by competing with ATP to prevent its binding to the intracellular tyrosine kinase region. Tyrosine kinases can be divided into receptor tyrosine kinases and nonreceptor tyrosine kinases. the tyrosine kinase itself forms an integral part of the receptor molecule. angiogenesis and resistance to apoptosis.

Despite some promising early results (Fukuoka et al. In this trial. It is also being studied in other cancers. The recommended dose is 150 mg daily. 2003. EGFR = epidermal growth factor receptor.. FLT = fms-related tyrosine kinase. including breast and head and neck cancer.Rachel Jones and Robert Leonard Table 2. Gefinitib acts by reversibly inhibiting autophosphorylation. vomiting. BRAF = v-raf murine sarcoma viral oncogene homolog B1. Adapted from Arora and Scholar. altering transcription factor activity. . 2002). Kris et al. It causes cell-cycle arrest in G1 phase and apoptosis. Gefitinib is not currently licensed for use in the UK. VEGFR/EGFR Non-small-cell lung cancer EGFR tyrosine kinase inhibitors Gefitinib Gefitinib is an EGFR tyrosine kinase inhibitor that was initially studied in non-small-cell lung cancer (NSCLC). the occurrence and severity of a rash was associated with improved survival. ABL = Abelson protein tyrosine kinase. In progressive NSCLC after chemotherapy it gives a response rate of 12% with a median survival of 8. Name Imatinib Gefitinib Erlotinib Sunitinib Sorafenib Lapatinib Target BCR-ABL/KIT EGFR EGFR VEGFR/PDGFR/KIT/FLT3 BRAF/VEGFR EGFR/HER-2 Tumour type Chronic myelogenous leukaemia Gastrointestinal stromal tumour Non-small-cell lung cancer Non-small-cell lung cancer Renal cell carcinoma Gastrointestinal stromal tumour Renal cell carcinoma Melanoma Breast cancer Non-small-cell lung cancer Colorectal cancer Canertinib EGFR Breast cancer Non-small-cell lung cancer Ovarian cancer Vatalanib VEGFR Colorectal Prostate Renal Glioblastoma multiforme Zactima ZD6474 BCR = breakpoint cluster region. 2002). 2004). KIT = kitten (v-kit: derived from feline sarcoma virus – abbreviated ‘kitten’. Giaccone et al.. Erlotinib is currently licensed for the treatment of patients with locally advanced or metastatic NSCLC after failure of at least one prior chemotherapy regimen. PDGFR = platelet-derived growth factor receptor. and shifting cells from S phase into G0/G1. official name v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog). The most common dose-limiting toxicities associated with erlotinib are acneiform skin rash and diarrhoea. recent studies in NSCLC have shown no benefit from adding gefitinib to chemotherapy (INTACT 1 and 2... HER-2 = human epidermal growth factor receptor 2. 16 Erlotinib Erlotinib is an oral reversible EGFR-targeted tyrosinekinase inhibitor that is highly selective for its receptor at nanomolar concentrations. diarrhoea and an acneiform rash.1.4 months (Perez-Soler. Johnson et al. 2005. VEGFR = vascular endothelial growth factor receptor. The main side effects are nausea. Summary of tyrosine kinase inhibitors in clinical use and trials. 2002. which may require a dose reduction.

chest and back and may resemble folliculitis or an acneiform drug eruption. gastrointestinal toxicity including nausea and diarrhoea.3%) and median progression-free survival (36. Sunitinib is licensed for (a) the management of unresectable and/or metastatic malignant GISTs that are resistant to imatinib and (b) the treatment of advanced and/or metastatic renal cell carcinoma following failure of IFNα or IL-2 therapy. lapatinib and cetuximab. 2006). 2006).5% versus 14. which is targeted by sorafenib.9 weeks versus 19. VEGFR 2 and VEGFR 3). Malignant melanoma can be associated with a mutation in BRAF. 2005). The daily dose should neither exceed 87. The recommended dose is 400 mg twice a day.9%) in favour of sunitinib (Motzer et al.Biological treatments in cancer Lapatinib Lapatinib is an oral dual tyrosine-kinase inhibitor of EGFR and HER-2. taxanes and trastuzumab compared a combination of lapatinib with capecitabine against capecitabine alone. including erlotinib. It is active against many breast cancer cell lines in vitro.d. haemorrhagic events. The dose may be increased or decreased.. or minocycline 100 mg b. The results showed a significant increase in response rate (22. hypertension. Patients with extensive or persistent skin involvement should be referred to a dermatologist. A phase III trial in patients with metastatic HER-2-positive breast cancer who had previously received anthracyclines. sunitinib gives a response rate of 33%. a combination of clindamycin 1% and benzoyl peroxide 5% gel. In a phase III study in refractory GIST patients. or oral antibiotics such as tetracycline 250 mg q. In patients with metastatic renal cell cancer who have failed prior cytokine therapy. VEGFR and PDGFR. and studies are under way to determine whether sorafenib is effective.or IL-2-based therapy or who are considered unsuitable for such therapy. Sorafenib has been associated with tolerable toxicity including hand and foot syndrome. and fatigue.d. In patients with advanced renal cell cancer who had received one previous systemic treatment. and myelosuppression. Dermatological toxicity can include skin discolouration and depigmentation of the hair or skin. 2004).7 weeks) in favour of the combination (Geyer. significant improvement in progression-free survival (47.5 mg. conference presentation. including left ventricular dysfunction. Vascular endothelial growth factor receptor tyrosine kinase inhibitors Sunitinib Sunitinib is an oral multi-targeted kinase inhibitor which inhibits vascular endothelial growth factor receptors (VEGFR1. Further phase III trials are ongoing and are investigating lapatinib as both a single agent and in combination with chemotherapy and hormone therapy.. ASCO. Significant toxicities have been associated with sunitinib.3 versus 24. but advances in genetic engineering have provided better tolerated preparations.5 to 6 months and improved overall survival (Demetri et al.5 mg nor be decreased below 37. produce a characteristic rash as a side effect. Sorafenib Sorafenib is an oral multi-targeted kinase inhibitor that targets RAF kinase. Sunitinib is also active against GIST. based on the patient’s ability to tolerate it. Topical and systemic corticosteroids have produced variable responses. diarrhoea. sorafenib significantly prolonged progression-free survival when compared with best supportive care (Escudier et al. The recommended dose of sunitinib is 50 mg taken orally once a day for four consecutive weeks. and 37% achieve stable disease for three months (Eskens. The rash may respond to clindamycin phosphate 1% gel. Management of the skin rash with anti-EGFR agents Drugs that target the EGFR. followed by a two-week rest period. platelet-derived growth factor receptors (PDGFRα and PDGFRβ). The development of a rash is one of the best predictive factors for disease response. KIT and FLT3. The main toxicities of lapatinib are diarrhoea and skin rash.s.. including trastuzumab-conditioned HER-2-positive cell lines. gefitinib. Sorafenib is indicated for the treatment of patients with advanced renal cell carcinoma who have failed previous treatment with IFNα. sunitinib delayed the median time to progression from 1.9 weeks) and an objective response rate (24. The rash is characterised by inflammatory papules and pustules on the face. 2005). skin rash. hypertension.8% versus 4. 17 . A recent phase III trial that compared sunitinib with IFNα as first-line therapy in metastatic renal cell cancer showed a statistically Monoclonal antibodies Early attempts at using monoclonal antibodies were limited by the strong immune responses evoked by murine antibodies.

IgG. Antibodies can be classified according to their constant region (e. Cardiac function then needs to be monitored every 3 to 6 months because of possible left ventricular impairment. which is the antigen binding site. IgM). 2004): r The antigen should be expressed stably and homogeneously by tumour cells. Each chain has a variable amino acid sequence at its tip.Rachel Jones and Robert Leonard Table 2. It is active against HER-2-positive breast cancer and it has been approved by NICE in metastatic breast cancer. ASCO. r In combination with paclitaxel in patients who have not received chemotherapy for their metastatic disease and for whom an anthracycline is not suitable. Cetuximab Cetuximab is a chimeric IgG1 antibody that targets the EGFR (ERB B1). and the recent 2006 ASCO update confirmed an overall survival benefit in the one-year arm (I. Trastuzumab is a cardiotoxic drug. The mechanisms of antibody action are as follow: r Tumour cells are killed by activation of the immune system through the Fc component. In patients with metastatic irinotecanrefractory colorectal cancer. the BOND study compared irinotecan and cetuximab with cetuximab alone. which is overexpressed in approx18 . 1 year or 2 years of trastuzumab.g. imately 15 to 20% of patients with breast cancer. Classification and nomenclature Antibodies consist of four polypeptides: two heavy chains and two light chains which are joined together by disulphide bonds to form a ‘Y’ shape. including anaphylaxis. The constant region (Fc) determines the mechanism that is used to destroy the antigen. r In combination with docetaxel for patients who have not received chemotherapy for their metastatic disease. Type Murine Chimeric Humanised Origin Mouse 65 to 90% human murine variable regions 95% human 5% murine hypervariable region Human Human “-umab” “-zumab” Nomenclature “-omab” “-ximab” Adapted from Wick. r Receptor protein kinase phosphorylation is inhibited by the antibody binding to the receptor or to the ligand. The following are tumour antigen characteristics that are ideal for targeting by antibodies (Harris. r There should be limited expression of the antigen in normal tissues. 2006).2. trastuzumab is indicated for the treatment of patients with HER-2-positive breast cancer: r As monotherapy in patients who have received at least two chemotherapy regimes (including at least an anthracycline and a taxane unless patients are unsuitable for these treatments). Smith. conference presentation. NICE has approved one year’s treatment with trastuzumab following adjuvant chemotherapy for patients with HER-2-positive tumours (+++ on IHC. In metastatic breast cancer. 2006). Toxicities associated with trastuzumab are an acute infusion reaction and hypersensitivity. Trastuzumab is also being investigated as an adjuvant treatment for breast cancer following chemotherapy. There was a significantly improved response rate and median Trastuzumab Trastuzumab is a humanised IgG1 antibody that targets the HER-2 protein. Monitoring is especially important in the adjuvant setting. Monoclonal antibodies are named according to their components (see Table 2. in which patients were randomised to observation. showed a significant improvement in disease-free survival.2). NICE. and a baseline measurement of cardiac function such as a MUGA scan must be performed. in those who have received prior chemotherapy (NICE. 2002). r There should be minimal or no soluble form of the antigen (to avoid rapid clearance or extracellular binding of the antibody). r The antigen should be present on the cell surface to allow clear access for the antibody. Approximately 10% of patients have deterioration in their left ventricular function and 2 to 4% develop significant heart failure. either in combination with paclitaxel or as a single agent. by either complement-mediated or antibody-dependent cell-mediated cytotoxicity. Interim analysis of the HERA trial. Nomenclature of monoclonal antibodies. 2004.

poor wound healing. but are not licensed for use in the UK. 2002). and they reduce the morbidity and mortality of neutropenia.. cetuximab is currently licensed for use in combination with irinotecan for patients with EGFR-expressing tumours after failure of chemotherapy that included irinotecan. It is licensed for head and neck tumours in combination with radiotherapy for locally advanced squamous cell carcinoma. 78). 19 . proteinuria. Haemopoietic colony stimulating factors Granulocyte colony stimulating factor (G-CSF) Endogenous G-CSF induces the maturation of neutrophils.org) and ASCO (Ozer et al. Kabbinavar et al. r To allow the delivery of dose-dense chemotherapy regimens (chemotherapy regimens in which the cycle length has been reduced). adding bevacizumab to chemotherapy significantly increases progression-free survival but there is no evidence of improved overall survival (Hurwitz et al. In metastatic colorectal cancer. arterial thromboembolism. In patients with head and neck cancer. 2006). haemorrhage. tumour lysis syndrome and mucocutaneous toxicity. cetuximab improves the duration of locoregional control. Febrile patients This is discussed in Chapter 6 (see p. nail disorders and an acneiform rash which may be associated with response.. The management of cetuximab rash is discussed in a previous section (see p.Biological treatments in cancer time to progression for combination therapy (Cunningham et al.and long-acting (pegylated) agents are available. Bevacizumab Bevacizumab is a recombinant humanised monoclonal IgG1 antibody targeted against VEGF. 2000) are available but the reader is advised to be aware that local guidelines may differ for the use of these agents. which can become activated during radiotherapy treatment. A possible mechanism of action is through inhibition of the EGFR/TKI complex. Recombinant preparations are available for clinical use. progression-free survival and overall survival compared to radiotherapy alone (Bonner et al. 2005). r Therapeutically. Recent international guidelines from the National Comprehensive Cancer Network (www. Bevacizumab is licensed for use in combination with intravenous 5-FU/folinic acid or intravenous 5FU/folinic acid with irinotecan as a first-line treatment for patients with metastatic colorectal carcinoma. 2004). to reduce duration of neutropenia during acute infective episode. in follicular lymphoma. They are currently being assessed in clinical trials. 17). and gastrointestinal perforation. 2003.. Radiolabelled antibodies Some monoclonal antibodies are further modified through conjugation with radioisotopes to allow more accurate delivery of radiation with less toxicity to surrounding normal tissue.. which can reduce the risk of infections and can reduce treatment delays in patients treated with chemotherapy. The toxicities include an infusion reaction. The recommended dose is 5 mg/kg body weight given every two weeks as an intravenous infusion.. In metastatic colorectal cancer. vincristine and prednisolone Primary and secondary prophylaxis This is discussed in Chapter 1 (see p. 2004. Both short. It is licensed for use in NHL and has been approved by NICE as a first-line treatment in combination with CHOP chemotherapy in CD20-positive diffuse large B-cell lymphoma and as an option in combination with cyclophosphamide. Potential uses for G-CSF include the following: r Primary prophylaxis with the first cycle of chemotherapy.nccn. Cetuximab is given once a week. r Secondary prophylaxis in subsequent cycles following the first episode of febrile neutropenia. The toxicities associated with treatment are acute infusion reactions. The initial dose is 400 mg/m2 followed by 250 mg/m2 . Several international guidelines have been published that review the use of these agents for the aforementioned indications. 6). Dose reduction for toxicity may be required. Rituximab Rituximab is a chimeric IgG1 antibody that acts against the CD20 antigen on B lymphocytes (Price and Sikora. The toxicities associated with bevacizumab are hypertension.

it is not clear whether erythropoietin improves tumour response and overall survival. cific tumour-associated antigens that will lead to the death of the tumour cell.Rachel Jones and Robert Leonard Dose-density chemotherapy regimens The concept of dose-density chemotherapy is to deliver multiple. r Epoetin beta. The use of erythropoietin is currently under review by NICE. 2004).3 shows tumour-associated antigens that may be suitable targets for vaccination.. secretion of cytokines causing tumour cell destruction). Peptide-based vaccines and carbohydrate-based vaccines Vaccines Vaccination strategies have been investigated as a treatment option in the management and prevention of cancer. and when. such as HPV 31 and 45 (Harper et al. ICAM1). reduced-interval cycles of chemotherapy with G-CSF. which are responsible for genital warts (Munoz. gangliosides) and glycoproteins. to vaccinate. A Cochrane review of 27 trials involving a total of 3287 patients showed that recombinant human erythropoietin significantly decreases the need for red-cell transfusions (Bohlius et al. the main cause of cervical cancer. 2006)..e. r Effector mechanisms need to be generated (e. TGF-β) need to be avoided. 2006). r The longer-acting darbepoetin alpha. malignant lymphoma.g. r Immune-suppressive factors produced by the tumour (e. It is hoped that HPV vaccines will have a major impact on the incidence of cervical cancer in the future. Erythropoietin Most patients who are symptomatic from anaemia are treated with blood transfusions. Discussions are ongoing about who. 2003). Although there is evidence of a benefit with this approach in NHL (Pfreundschuh et al. However. Gardasil® prevents infection by HPV 16 and 18. Three recombinant erythropoietins are licensed for use in patients who have symptomatic anaemia in non-myeloid malignancies and who are receiving chemotherapy: r Epoetin alpha. Antigen vaccines Viral-protein-based vaccines Viral proteins from cancer-causing viruses may be used as antigens to prevent infection by the virus.g.g. 2004) and breast cancer (Citron et al. All are licensed to be used in patients with a haemoglobin level less than 11 g/dl. they are present in significantly greater amounts than in normal cells). HPV virus types 16 and 18 are responsible for about 70% of cervical cancers. Two vaccines have been developed which use non-infectious virus-like particles of the major capsid protein L1. and by the HPV types 6 and 11. Cellular vaccines Cellular vaccines can be based on whole cells.. They may differ from normal cells only in a quantitative manner (i. r The ApCs need to be co-stimulated through members of the B7 family and various adhesion molecules (e. Gangliosides are over expressed in melanoma. The aim of vaccination against an established tumour is to stimulate a cellular immune response against spe20 Peptides derived from tumour-associated antigens or gangliosides (which are sialated glycolipid antigens) can be used as vaccines. Darbepoetin alpha is only licensed for use in patients receiving chemotherapy for solid tumours. Cervarix® prevents infection by HPV types 16 and 18 and shows some cross-protection against other oncogenic HPV viruses.. To achieve this aim: r The tumour antigen needs to be recognised on an antigen-presenting cell (ApC) by interaction with a cytotoxic T-cell receptor.g. Table 2. tumour lysates. Tumour-associated antigens are usually proteins but they can be carbohydrates (e. . A major recent advance in vaccine development for cancer prevention has been the development of two vaccines against human papilloma virus (HPV). although an alternative treatment is to administer recombinant human erythropoietin. The short-acting erythropoietins are licensed for treatment of symptomatic anaemia in patients receiving chemotherapy for solid tumours. use of G-CSF for dosedensity chemotherapy regimens should be restricted to clinical trials. or antigens that have been shed from the surface of tumour cells grown in vitro. or multiple myeloma (epoetin beta is also licensed for CLL).

S6–8.. J. (2003). kidney Melanoma Head and neck tumours (squamous cell carcinoma) Melanoma Colorectal Breast. Cancer. 2002. G. ERBB2/HER-2 = human epidermal growth factor receptor 2. Their use is currently restricted to clinical trials. and Wagstaff. M. Examples of tumour-associated antigens that are potential targets for cancer vaccines. pancreatic. 354. J. et al. A. Janeway et al. Harari. Eur.. Kunkel. 40. Atzpodien. lung and colorectal Breast. breast (normal testicular proteins) Melanoma.. L. S11–14. Radiotherapy plus . D. r Other nucleotide sequences can be included that code for immune effectors to increase the immunological response to the vaccine. 29A (Suppl. The Cochrane Database of Systematic Reviews. et al. M. 2004.. (1993). MUC1 = mucin 1... Cirrincione. J. Role of tyrosine kinase inhibitors in cancer therapy.6% (Rosenberg et al. et al. Bonner.. A. Langensiepen. J. Baaten. Schwarzer. At present no vaccine for Arora. B. (2004). J. Hanninen. Modified from Espinoza-Delgado. J. Ther. A systematic review of the relation between interleukin-2 schedule and outcome in patients with metastatic renal cell cancer. J. 1127–44. Issue 3. H. r Recombinant viruses (e. G. 5).. 2004). Dendritic cell vaccines Dendritic cells are potent antigen-presenting cells and they can be used to present the tumour-associated antigen to increase the chances of an immune response to the vaccine. P M. Giralt.Biological treatments in cancer Table 2.. M. J. r Heat shock proteins can promote the development of dendritic cells into mature ApCs. and Scholar. Engl. Voogd. (2004). E. established cancer is recommended for use outside a clinical trial.. 2001. αFP = alpha feto-protein.g. 971–9. 6 (Suppl. CD003407. A. Art. et al. Atkins.. C. Randomized trial of dose-dense versus conventionally scheduled and sequential versus concurrent combination chemotherapy as 21 . Tumour-associated antigen Melanoma antigen expression family (MAGE) Renal antigen expression family (RAGE) Melanoma antigen recognised by T cells 1/Melan A (MART) Caspase 8 B catenin TGF-β receptor II MUC1 ERBB2/HER-2 αFP Prostate specific antigen Gangliosides CEA Tumour Melanoma. adenovirus) can be used as vectors for the tumour-associated antigen. Vaccines are likely to play a more important role in cancer prevention as seen with emerging data on recombinant vaccines against HPV 16 and 18 to prevent cervical cancer. Bohlius. Interleukin-2 in combination with interferon-alpha and 5-fluorouracil for metastatic renal cell cancer. Am. E. 567–78. Clinical data on vaccines A recent review of cancer vaccines by the National Cancer Institute (USA) showed that the objective response rate in 440 patients with established tumours was only 2... CEA = carcino-embryonic antigen. Exp. S. and Rosenberg et al. 315. N. C. Sci.. no. M. Pharm. (2005). Eur. L. Interest exists in combining them with tumour-associated antigens. Cancer. 1). REFERENCES Other vaccines r DNA-based vaccines can be specific genes that encode the tumour-associated antigen cloned into a bacterial plasmid. Med. Cancer J. Sznol. High-dose recombinant interleukin-2 therapy in patients with metastatic melanoma: long-term survival update. cetuximab for squamous-cell carcinoma of the head and neck.3.. Citron.. (2006). A. Berry.. L. (2000). J. Kirchner. et al. Erythropoietin for patients with malignant disease. ovarian and lung Hepatocellular Prostate Melanoma Colon TGF-β = transforming growth factor β.

Harris. (2002). Clin. D. A. randomized. D. J. NICE. B. Clin. Phase II trials of recombinant leukocyte A interferon in disseminated malignant melanoma: results in 96 patients. Adjuvant high-dose bolus interleukin-2 for patients with high-risk renal cell carcinoma: a Cytokine Working Group randomized trial. et al. et al. 20–33. Med. (2002). NICE. J. 21. (2004).. 7 (Suppl. et al. M. (2004). Ann. E. H. M. 645–52. Am. J. 619–24. I. J. Wheeler. J. Creagan. Results of treatment of 255 patients with metastatic renal cell carcinoma who received high-dose recombinant interleukin-2 therapy. (1995). Clin. R. et al. C. 3697–705. D.. et al. Hutson. N. L. (2003). Gynecol. (2004). clinical practice guidelines. (2002). Kabbinavar.. 5. J. (2004). N. (2006). Oncol. Eisen. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer.. L. Hurwitz. J. London: National Institute for Clinical Excellence. Ahmann. A. J. N. (2006). Hurwitz. 1–7.. R.. P. Humblet. Med. Espinoza-Delgado. Fehrenbacher..5 years of a bivalent L1 virus-like particle vaccine against human papillomavirus types 16 and 18: follow-up from a randomised control trial. (2000). 337–45. 104. 346. (2001). et al. Abstr.. Clin. Eur. Proc. 107. Tomczak. (2002).. 21. F.. 24. G. Clark. M. 3). F. Obstet. G. Demetri. Oncol. Addition of bevacizumab to bolus fluorouracil and leucovorin in first line metastatic colorectal cancer: results of a randomized phase II trial. Yano. Soc.. B. 10. Engl.. Manegold. 1247–55. G. Oncol... 23. 289–96. 34. Frytak. J. M. fluorouracil and leucovorin for metastatic colorectal cancer... Engl.. Bevacizumab plus irinotecan. A. What’s in a drug name? J. Oncol. Natale. Kloess. 21. 10. (1986). Trumper. A. L. H. Bacik. Oncol. M. Blackstein. Pfreundschuh. H. (2005). Phase III multicenter. J. Clin. 351. 909–15. (2006). D. Lancet Oncol. London: . et al. Ozer.. Hochhaus.. S. 20. J. Treatment of Cancer. Interferon-alfa as a comparative treatment for clinical trials of new therapies against advanced renal cell carcinoma. Soc. 2335–42. Oncol. M. M. 20. M.. 472–80. Blanke. J. Sustained efficacy of up to 4. (2002). A. Bennett. J. The role of erlotinib (Tarceva. E. et al. double-blind. Nat. Szczylik.. Guidance on the Use of Trastuzumab for the Treatment of Advanced Breast Cancer. P. American Society of Clinical Oncology Growth Factors Expert Panel.. Escudier.. H.. Oncol. 292–302. L. D. C. N. 18 [Epub ahead of print]. J. Oncol. 350. A phase II trial of ZD 1839 (Iressa) in advanced non-small cell lung cancer patients (NSCLC) who had failed platinum. M.. D. A phase III clinical trial of ZD 1829 (Iressa) in combination with gemcitabine and cisplatin in chemotherapy-naive patients with advanced non-small-cell lung cancer (INTACT 1). (2002). Phase II randomized trial comparing bevacizumab plus fluorouracil (FU)/leucovorin (LV) with FU/LV alone in patients with metastatic colorectal cancer.. Fehrenbacher. Soc.. Murphy. Immunobiology: The Immune System in Health and Disease. et al.Rachel Jones and Robert Leonard postoperative adjuvant treatment of node-positive primary breast cancer: first report of Intergroup Trial C9741/Cancer and Leukemia Group B Trial 9741. et al.. D. D. ZD 1839 (Iressa) in combination with paclitaxel and carboplatin in chemotherapy-naive patients with advanced non-small-cell lung cancer (NSCLC): results from a phase III clinical trial (INTACT 2). Angiogenesis inhibitors in clinical development. 13. W. J.. Oncol. et al. Abstr. R.. 5th edn. et al. Giaccone. Proc. Giaccone. Price. Rosenberg. Munoz.. T. et al. Biol. Clin. A. Med. Pharm. OSI 774) in the treatment of non-small cell lung cancer. G. Trastuzumab for the Adjuvant Treatment of Early-Stage HER2 Positive Breast Cancer.. Lancet. B. W.. M. 283–95.. Novotny. C. J. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. Immunotherapy for metastatic renal cell . (2004).. Randomized phase III trial of the Raf kinase and VEGFR inhibitor sorafenib (BAY 43–9006) in patients with advanced renal cell carcinoma (RCC).. (2002). G.. I. Urba. et al. 21. (2004). J. R.. Assoc. Cancer immunotherapy: moving beyond current vaccines. et al. 1166. 4510. A. Yang. 626–33... Cancer. L. Cancer Res. Oncol.. 23. Engl. Cancer vaccines. S. 70.. C. (2002). Fisher. J. Two-weekly or 3-weekly CHOP chemotherapy with or without etoposide for the treatment of young patients with good-prognosis (normal LDH) aggressive lymphomas: results of the NHL-B1 trial of the DSHNHL... 1431–9. Schulz. Von Mehren. 2237–46. 13. carcinoma. Oncol. 12–14. Herbst. Johnson. (2005). (2003). et al. Clin. (2003). T. (2006).and docetaxel-based regimens (IDEAL 2). Aug. S. 22 . Urol. Harper.. Giaccone. 347. Clin... et al. LBA3. s308. C. Med. London: National Institute for Clinical Excellence. C. McCleod. Demetri. C. (2004).. et al. Fyfe. 23. Reprod.. Wick. Clin. K. H. J. Hematologic and cytogenetic responses to imatinib mesylate in chronic myelogenous leukemia. Multi-institutional randomized phase II trial of gefitinib for previously treated patients with advanced non-small-cell lung cancer (The IDEAL 1 Trial) [corrected].. Rep. Perez-Soler. placebo controlled trial of SU11248 in patients following failure of imatinib for metastatic GIST. M. Kris. E. 44. Chapman and Hall. Clin. Clin. Travers. 90.... Clin. 4. M. Med. 3133–40. R. Phase III randomized trial of sunitinib malate versus interferon-alfa as first line systemic therapy for patients with metastatic renal cell cancer. Kantarjian. et al. Oncol. Eskens. Wirth. 4238s–40s. B.. P (1993). et al. Van Oosterom.. Armitage. (2002). J. M. R. Kabbinavar. G. Atkins. 21. F. T... Oncol. Johnson. I. North Am. J. 13.. Franco. R. Monoclonal antibodies as therapeutic agents for cancer. Fukuoka. N. Technology Appraisal Guidance No. Siena.. Blood. L.. 18. 127. P Walport. 367. New York: Garland. 688–96. (2004). Proc. . Cancer Treat. T. O. J.. Janeway. where are we now and where are we going? Br. J. Oncol. (2005). N. Cunningham. Am. Sawyers. Clin. The Oncologist.. S.. 2000 update of recommendations for the use of hematopoietic colony-stimulating factors: evidence-based. S. Motzer. 60–5. A review of the phase III clinical data for the quadrivalent human papilloma vaccine (Gardasil® ). S. Am. et al.. Technology Appraisal Guidance No. S. M. Engl. Am.. G. J. Y. H. Clin. Motzer. Rosenberg. 3558–85. P and Sikora.. Herbst. Ann. and Restifo... L. C.

testosterone is converted to oestradiol by the enzyme aromatase. via the enzyme aromatase. the continued growth of the cancer is dependent upon the continued stimulation by hormones. Although this is most clearly seen in breast and prostate cancer. It should be read in conjunction with the relevant site-specific chapters (Chapters 16 and 19). he was the first to identify a link between ovarian function and breast cancer. alone or in combination. metabolism or action.5% of the total daily production of 23 . by doing so. by affecting hormone synthesis. This chapter provides some background knowledge of the production and functioning of hormones and their receptors. and production varies depending on environmental and genetic factors such as obesity. Steroidal hormones have the potential to activate oncogenes or inactivate tumour-suppressor genes. production is cyclical and is controlled by positive and negative feedback via the hypothalamic–pituitary–gonadal axis (see Figure 3. In postmenopausal women. or by altering hormone receptor expression within the cell.3 HORMONES IN CANCER Jacinta Abraham and John Staffurth Introduction Hormonal therapies are some of the oldest active systemic anticancer therapies in use today. oestrogens are synthesised from cholesterol. This chapter will focus primarily on the steroidal hormones that are of particular importance in breast and prostate cancer: the oestrogens. other cancers that may exhibit hormone dependence include endometrial. 1952). 1941). and. The aetiology of hormone-related cancers is discussed in the relevant site-specific chapters. androgens and progestins. in this situation. for example. and they generally act via cell-membrane-localised receptors which trigger second messengers within the cytoplasm. the main site of oestrogen synthesis is adipose tissue. leads to breast and prostate cancers in rats (Noble. Oestrogen synthesis takes place mainly in the granulosa cells of the ovaries. Frequently. r Surgical orchidectomy results in the regression of advanced prostate cancer (Huggins and Hodges. r Steroidal hormones. A summary of the major historical landmarks in anticancer hormonal therapy since Beatson’s observation in 1896 is listed here: r Large doses of oestrogenic steroids inhibit gonadotropin secretion in women (Zondek. In 1896 Beatson demonstrated that surgical oophorectomy resulted in tumour regression in some premenopausal women with metastatic breast cancer. polypeptides or derivatives of amino acids.1). ovarian and testicular cancers. In men. r Adrenalectomy results in regression of advanced breast and prostate cancers (Huggins and Bergenstal. and it inhibits follicle-stimulating hormone (FSH) release by pituitary gonadotrophs. r Prolonged exposure to oestrogens or androgens. Anticancer hormone therapies work in a number of different ways. which will help in the understanding of commonly used therapies. such as oestrogens. progestins and androgens. Although only 0. Substantial evidence now exists that hormones play a key role in both the cause and the outcome of several cancers. which can produce a sequence of changes within the cell that ultimately leads to cancer. Inhibin is a polypeptide which is also produced by ovarian granulosa cells. Hormone synthesis Oestrogen synthesis In premenopausal women. 1940). bind to intracellular receptors to mediate their action. 1980). removing the hormonal stimulus causes the cancer to regress. 30% of plasma oestrogens originate from the testis and 70% arise from peripheral aromatisation of androstenedione and testosterone. Hormones are classified into two groups: r Non-steroidal hormones include peptides.

1. ACTH = adrenocorticotrophic hormone.2. metabolised within the prostate to the more biologically active 5α-dihydrotestosterone (DHT). AR = androgen receptor. FSH = follicle-stimulating hormone. ACTH = adrenocorticotrophic hormone. The hypothalamic–pituitary–gonadal axis in the male. LH = luteinising hormone. The hypothalamic–pituitary–gonadal axis in the female. ERα is often markedly upregulated testosterone is converted. GnRH = gonadotropin-releasing hormone. DHT = 5α-dihydrotestosterone. the main circulating androgen is testosterone. Androgen synthesis In men. For example. such as dihydroxyandrostenedione (DHA).2). Not all ER-positive breast cancers respond to hormonal treatments.Jacinta Abraham and John Staffurth Hypothalamus GnRH Pituitary FSH/LH Ovaries Androgen conversion within ovarian granulosa cells Testosterone Aromatase Oestradiol Negative Feedback Hypothalamus GnRH CRH Negative Feedback CRH Pituitary FSH/LH ACTH ACTH Adrenals Testes Cholesterol Testosterone Adrenals Cholesterol Androstenedione Androgen Peripheral conversion within adipose tissue/skin Androstenedione Aromatase Oestrone 17βOH dehydrogenase Oestradiol Prostate Testosterone DHT Androstenedione Peripheral fat DHT Androstenedione AR activation and nuclear translocation Figure 3. feedback. Testosterone. 90% of which is produced by the Leydig cells of the testis (see Figure 3. CRH = corticotropinreleasing hormone. The remaining circulating androgens. and there is some evidence that different expression of the two ERs in breast cancer cells affects hormone responsiveness and hormone resistance. oestradiol is a much more potent gonadotropin inhibitor than testosterone. and circulating oestrogens contribute to negative Oestrogens Breast Endometrium Ovary Figure 3. The two receptors have a similar structure but they share only 47% amino acid homology. which binds to the androgen receptor (AR) with a three. androstenedione. and DHA sulphate. GnRH = gonadotropin-releasing hormone. androstenedione.to five-times greater affinity than testosterone. FSH = follicle-stimulating hormone. r Oestrogen receptor beta (ERβ) is encoded on chromosome 14. LH = luteinising hormone. All of these androgens may be 24 . DHT. are produced in the adrenal cortex from cholesterol. Hormone receptors Oestrogen receptor There are two oestrogen receptors (ERs) which are encoded by separate genes: r Oestrogen receptor alpha (ERα) is encoded by a gene on chromosome 6. CRH = corticotropin-releasing hormone.

a protein that maintains the AR 25 . co-regulator proteins can also influence ER-mediated transcription. but it is not usual to distinguish between the subtypes. Whether the cancer is sensitive to hormone treatment is related to the degree of ER and PgR positivity. oestradiol binds to the ER in the nucleus via the ligand binding domain on AF2. r Increases the viscosity of cervical mucus. r In contrast. and in doing so they effectively prevent the action of both AF1 and AF2. will have a response rate to hormone therapy of less than 5% and patients with these tumours are likely to derive a greater benefit from chemotherapy. r Reduced low-density lipoprotein (LDL). Tamoxifen blocks AF2 when it competes with oestradiol for ER binding. However. the measurement of AR in patients with prostate cancer is not performed routinely because both hormone-dependent and -independent prostate cancers possess functioning AR. 2004). and it has eight exons. Second. ER−/PgR− cancers. Oestrogen receptor functioning To understand how antioestrogen treatments work. Aromatase inhibitors block the production of oestrogens. Androgen receptor functioning DHT binds to the AR via the ligand binding domain. Oestrogenic effects The following is a list of the effects of oestrogen: r Development of female secondary sexual characteristics.. PgR is important in mammary gland development. It is not clear if this represents a change in ER expression or if it reflects a heterogeneous primary tumour in which the ER-negative component has a survival advantage. which are upstream of oestrogen-responsive genes. it is necessary to be aware of three important domains on the ER: activating functions 1 and 2 (AF1 and AF2) which carry out the effector functions of the receptor and the ligand binding domain. Approximately 75% of postmenopausal and 50% of premenopausal women with breast cancer will have hormone-receptor-positive tumours. and its expression has been implicated in tamoxifen resistance (Miller et al. Progesterone receptor There is one gene for the progesterone receptor (PgR) but it exists as two isoforms. r ER−/PgR+ cancers account for 5% of all breast cancers and they should be regarded as being hormone responsive because they will have response rates of up to 30%. q11-13. AR status provides no prognostic or diagnostic value. the transcriptional activity of the genes that contain EREs is modulated by AF1 and/or AF2. but receptor status alone is not the only determinant of hormone responsiveness. In contrast to the situation in breast cancer. But ERβ may modulate the ability of cells to respond to oestrogen. Main actions of progesterone Progesterone performs the following actions: r Causes endometrial secretory phase. r Facilitates mammary gland development. but AF1 remains functional. and this binding results in separation from heat shock protein 90 (HSP-90. r ER+/PgR+ disease confers the highest response rates of up to 70%. A (PR-A) and B (PR-B). r Bone formation. Androgen receptor The AR gene is on the long arm of the X chromosome.Hormones in cancer in breast cancer and it is the marker that can best predict responsiveness to hormonal treatments. depending on the ligand. This allows the ER to bind to DNA oestrogen response elements (EREs). which have different physiological functions. it is standard practice to measure ER and PgR status in breast cancer. Third. the ER interacts with either co-repressors or co-activators to inhibit or enhance its transcriptional activity on target genes (Milano et al. PR-A can repress PR-B and ER and its expression is related to tamoxifen resistance (Hopp et al. phosphorylation. The response is less for ER+/PgR− cancers. First. and a conformational shape change. 2006). Receptor status in breast cancer and response to hormone therapy At present. with current testing. r Procoagulant. and this causes dimerisation. 2006).. which accounts for the oestrogenic activity seen in some tissues. and an excess production of PR-B is associated with breast cancer risk. r Endometrial growth.. allowing it to metastasise. r An ER-negative status may be detected in metastatic or recurrent disease even when the primary tumour is ER-positive.

Breast cancer: ovarian function suppression (OFS) In premenopausal women. which is currently the treatment of choice. FSH. This alters the balance of co-activator and co-repressor complexes and affects the regulation of activation 26 . In early breast cancer. Current IBCSG trials of ovarian function suppression (OFS) in premenopausal women Trial SOFT Trial design Tamoxifen 5 years or OFS + tamoxifen 5 years or OFS + exemestane 5 years TEXT OFS + tamoxifen 5 years or OFS + exemestane 5 years May or may not have chemotherapy PERCHE OFS + tamoxifen or exemestane 5 years or OFS + chemotherapy + tamoxifen or exemestane 5 years Is chemotherapy necessary in premenopausal women who receive OFS and hormone therapy? Is an aromatase inhibitor superior to tamoxifen in premenopausal women treated with OFS? Trial question Does OFS add to endocrine therapy in premenopausal women? in an inactive state). dimerisation to form a DHT–AR complex.2 summarises some of the hormonal therapies used in breast cancer. and oestradiol. OFS has been shown to produce a survival benefit equivalent to that achieved by chemotherapy (EBCTG. 2005). Chemical ablation is achieved by using gonadotropin-releasing hormone (GnRH/LHRH) agonists such as goserelin. with a median response duration of 9 to 12 months. Mode of action of tamoxifen r SERMs modulate oestrogen activity by binding to the ER. see Chapter 16).1. The various roles of OFS. and a conformational shape change. radiotherapeutic or surgical treatments. resulting in a conformational shape change. r Development of prostate and seminal vesicles. because its effects are reversible. Antioestrogens Tamoxifen Tamoxifen is the most extensively studied and widely used antioestrogen in the treatment of breast cancer. all three techniques are essentially equivalent and they produce response rates of 25 to 30%. cells. In advanced disease. Tamoxifen is described as a selective oestrogen receptor modulator (SERM) because of its agonist and antagonist activities. r Increased bone density. The activity of genes containing AREs can be modulated by transcription factors via the highly variable amino-terminal domain (ATD). particularly in early breast cancer. r Libido and frequency of erection. This complex is translocated to the nucleus of prostate cells. which allows DNA binding of the AR to androgen response elements (AREs) upstream of androgen-regulated genes. phosphorylation. it has been the mainstay of endocrine therapy for breast cancer for many years.Jacinta Abraham and John Staffurth Table 3. endocrine therapy and chemotherapy in early breast cancer are now the subject of clinical trials coordinated by the International Breast Cancer Study Group (IBCSG. which causes a varied effect on genes. Androgenic effects The effects of androgen include the following: r Virilisation and development of secondary sexual characteristics. The hormonal effects following administration of goserelin may be observed by measuring plasma levels of luteinising hormone (LH). and tissues. r Increased muscle mass. the most obvious and direct way to reduce oestrogen production is to suppress ovarian function. Breast cancer: other hormone therapies Table 3. which can be achieved by chemical.

and increased bone mineral density in postmenopausal women. is given with hydrocortisone to prevent adrenal insufficiency. rarely used because of side effects including rash.7 versus 0.o.m. vaginal discharge (30 versus 15%). r The beneficial effects of the weak agonist effect seen with tamoxifen are a reduction in cholesterol levels. It is currently licensed for use in postmenopausal hormone-positive advanced breast cancer that has progressed following antioestrogen therapy. no oestrogen agonist effect has been demonstrated. r Tamoxifen is beneficial in both pre. 2. Fulvestrant is an ER antagonist that competitively binds and downregulates the ER protein. and thrombotic events (1. Fulvestrant Adverse effects and beneficial effects of tamoxifen r Tamoxifen shows increased symptoms when compared with placebo: hot flushes (64 versus 48%).o.o. This is caused by the oestrogen agonist effects of tamoxifen. r An increased risk of endometrial changes exists. r Tamoxifen also reduces plasma levels of the potent cancer mitogen IGF1. 2005). 1 mg daily p.5 mg daily p. 1989). which accounts for the oestrogenic effects of tamoxifen.and postmenopausal women. Side effects such as gastrointestinal 27 . Inhibits the aromatase enzyme.o. 250 mg (5 ml) monthly i. decreased cardiac morbidity. irregular menses (25 versus 19%). r Impaired transcription of oestrogen-dependent genes occurs in the breast. r Further detailed data and discussion on clinical studies and treatment recommendations may be found in Chapter 16 (p. Irreversible aromatase inhibition Downregulation of the ER protein Cellular action not fully understood. Competitive aromatase inhibition Dose/route 20 mg daily p. Tamoxifen should be avoided in individuals at risk of thrombosis. There is a small but statistically significant increase in endometrial cancer..o. Clinical studies of tamoxifen r Five years of postsurgical adjuvant tamoxifen reduces the annual recurrence rate by 41% and annual mortality rate by 34% in ER-positive early breast cancer (EBCTG. and drowsiness ER = oestrogen receptor. downregulation of ovarian steroidogenesis Aminoglutethimide First generation aromatase inhibitor 250 mg twice daily p. nausea. including hyperplasia and polyps. 80–160 mg daily p. The risk is increased further when tamoxifen is co-administered with chemotherapy.4%) (Fisher et al. dizziness.2. 203). AF1 remains active. Hormonal therapies in breast cancer Drug Tamoxifen Anastrazole Letrozole Exemestane Fulvestrant Megestrol acetate Type Antioestrogen Non-steroidal aromatase inhibitor Non-steroidal aromatase inhibitor Steroidal aromatase inhibitor ER antagonist Progestin 25 mg daily p.o. Women on tamoxifen with abnormal vaginal bleeding should be promptly evaluated. Mode of action Competes with oestradiol for ER binding Competitive aromatase inhibition domain AF2. r Thrombotic events include deep vein thrombosis and pulmonary embolism.Hormones in cancer Table 3.

and reduced bone mineral density (BMD). however. Use in premenopausal breast cancer As previously mentioned. r AIs are classified as type 1 steroidal inhibitors. 203) and shown in Table 3. and arthralgia. These studies and the various strategies now available are discussed in Chapter 16 (see p. AIs are contraindicated in premenopausal women unless they are undergoing OFS. A greater understanding of primary resistance to tamoxifen (in particular. r Prolonged secondary amenorrhoea (> 1 year).3. exemestane and formestane). back pain. LH.. resulting in gonadal stimulation. hot flushes and pharyngitis have been reported. r Primary hypogonadism. headache. Risk factors for osteoporosis include the following: r Radiographic evidence of osteopenia and/or vertebral deformity. When AIs are indicated in these patients. AIs alone should be used with caution in premenopausal women who have become amenorrhoeic after chemotherapy. the AIs should be given only after ovarian function suppression. AIs are associated with an increased risk of musculoskeletal events. thoracic kyphosis. Adverse effects of aromatase inhibitors The comparative studies of AIs and tamoxifen have provided excellent comparative toxicity data.. In advanced disease. The clinical relevance of this reduction in BMD and how it should be monitored while the patient receives an AI are matters of ongoing debate. The ATAC bone substudy identified that women who had normal BMD at the outset and were treated with an AI did not become osteopenic or osteoporotic after five years (Howell et al. It is suggested that serial measurements of oestradiol. r Loss of height. Aromatase inhibitors Types r The first generation of AIs includes aminoglutethimide. and FSH levels should be performed.g. and endometrial hyperplasia or cancers. 2005). r Prolonged corticosteroid use. and they significantly reduce serum oestradiol.1). AIs are contraindicated in premenopausal women because ovarian suppression would lead to feedback on the hypothalamic–pituitary–gonadal axis.e. fractures. fractures. the role of concurrent use of OFS and AI is the subject of ongoing clinical trials coordinated by BIG/IBCSG (see Table 28 . accompanied by strong advice that reinforces diet and lifestyle (i. 3. Use in early postmenopausal breast cancer Several studies of early breast cancer have compared AIs with tamoxifen and have confirmed the safety and efficacy of AIs. all show that the use of an AI is associated with an increased rate of disease-free survival. Tamoxifen is associated more with thromboembolic events. 2006). high-calcium diet and smoking cessation). r Third-generation AIs are better tolerated and are used more commonly. which has been available for more than 20 years. being able to predict who is going to have tamoxifen-resistant disease) and longer-term follow-up data may help identify the subgroups that might benefit most from initial or subsequent AIs. these measurements require validated sensitive immunoassays which may not be readily available in all centres. In early premenopausal breast cancer. Mode of action AIs act by inhibiting or inactivating the p450 enzyme aromatase. vaginal discharge. r Previous fragility fracture. A recent report has shown that AIs may promote recovery of ovarian function in some of these women. Monitoring should be performed for individuals who are at a high risk of osteoporosis (see list that follows). Those with a reduced BMD at the start were more at risk of developing a further reduction. It has significant side effects and is now rarely used.Jacinta Abraham and John Staffurth symptoms. These studies have led us to redefine the optimal hormonal strategy for breast cancer patients. which lead to irreversible aromatase inactivation (e. even up to the age of 50 and after many months of amenorrhoea (Smith et al. including arthralgia. whereas the aromatase inhibitors are more linked with osteoporosis. a role clearly exists for OFS and AI use in women who have advanced disease with receptor-positive tumours and who have progressed on tamoxifen. The role of bisphosphonates in early breast cancer patients at risk of osteoporosis is a subject of current clinical trials. and type 2 nonsteroidal competitive inhibitors such as anastrazole and letrozole. r Premature menopause (< 45 years).

05 ABCSG 8/ ARNO 95 Tamoxifen for 5 years vs tamoxifen for 2 years and anastrozole for 3 years.60. inter- fere with steroid synthesis. Breast cancer: mechanisms of resistance Hormone-positive tumours may show resistance to endocrine therapy. Adapted from Coombes et al.3..04 8010 25. r Low body mass index (< 19 kg/m2 ). Their biochemical effects are complex and not well understood. The rate of occurrence of thrombotic events is reduced when using AIs compared to that for tamoxifen use. HR 0. breast tenderness.81. 5 years T or 5 years L or 2 years T. 2005.01) Other No benefit for combination arm. 3 years T IES Tamoxifen for 5 years vs 2–3 years tamoxifen and switch to 2–3 years exemestane 4742 37. Summary of main studies of early breast cancer that compare AIs with tamoxifen Trial ATAC Study design Upfront anastrazole vs tamoxifen for 5 years or in combination (included ER unknown and negative) BIG-98 Letrozole (L) vs tamoxifen (T). p = 0. r Maternal history of osteoporosis.Hormones in cancer Table 3. High doses have been used in premenopausal women but this is also associated with increased side effects.60. The rates of hot flushes are equivalent. 2005. either at first exposure (de novo) or after a period of time (acquired).. Progestins decrease the circulating levels of gonadotropins.0001) Overall survival benefit in the ER+ve group for switching to exemestane: HR 0. HR = hazard ratio. 2005. p = 0.58. The several side effects seen with their use include weight gain. 2005.76.. randomisation for switching at trial entry MA-17 Tamoxifen for 5 years followed by letrozole for 5 years or placebo for 5 years 5187 30 months Improved for letrozole (HR 0. These compounds are thought to act directly via the PgR but they also have an indirect action on the ER.0009) Overall survival benefit in ARNO 95 for switching to anastrazole: n = 979. p = 0. which can lead to corticosteroidal side effects and adrenocorticotrophic hormone suppression. p < 0.. Jakesz et al. p = 0.003) Patients (n) 9366 Median FU 68 months DFS Improved for anastrazole (HR 0.87. p = 0. 3 years L or 2 years L. benefit greater in receptor-positive patients No crossover results available DFS = disease-free survival. Progestins The two most widely used progestins are megestrol acetate and medroxyprogesterone acetate. The optimal progestin dose is controversial. Understanding 29 .4 months Improved for exemestane (HR 0.83. Howell et al. Goss et al.8 months Improved for letrozole (HR 0.045 Overall survival benefit in node positive: HR 0. and hypertension. and exhibit glucocorticoid activity. FU = follow-up.53. fluid retention. vs = versus. p = 0. p = 0. Thurlimann. ER = oestrogen receptor.001) 3224 28 months Improved for anastrozole (HR 0. 2004. nausea.

for example. this bidirectional crosstalk between signalling pathways may be active in endocrine resistance. Crosstalk may also occur between the ER and growth factor receptor pathways such as HER-2. Nearly all patients with advanced breast cancer who initially respond to hormone therapy will ultimately become hormone refractory. and they offer an alternative mechanism that stimulates cell growth. patients initially responding to tamoxifen have a 50% response rate to AIs.Jacinta Abraham and John Staffurth the mechanisms of resistance will help to predict the likely response to specific treatments and will help in the development of new agents targeted at endocrineresistant pathways at the molecular level. De Novo resistance A strongly ER-positive tumour is likely to respond better to tamoxifen than one with a weakly positive ER status. r Expression of ERβ has been seen to activate AP-1regulated genes when bound to tamoxifen. binding of oestrogen. 2006). r ERs located in the membrane can activate growthfactor-signalling pathways. This concept is supported by the fact that withdrawal responses may occur. and so oestrogen or tamoxifen might stimulate these alternative pathways. However. Excessive growth factor signalling or overexpression of HER-2 downregulates expression of the PgR gene. The development of resistance to tamoxifen does not necessarily predict resistance to an AI. This observation was not made in a subgroup analysis in the BIG 1–98 trial. In addition. Increased expression of ERβ is implicated in tamoxifen resistance. tamoxifen may become an ER agonist in breast cancer cells. . Breast cancer: areas of current interest Addition of fulvestrant in patients who relapse on AI The rationale is that patients who relapse on AIs may have tumours that are hypersensitive to the low levels of oestrogen observed in AI treatment. and even tamoxifen in membrane ER can activate the epidermal growth factor family receptors (Milano et al. Enhanced activation of AP-1 transcription factors has been associated with tamoxifen resistance in both human breast cancers and xenografts. the effect of low levels of oestrogen stimulation induced by tamoxifen and AIs may result in sensitisation of the ER. and it remains a continuing area of research. Fulvestrant. This is being studied in the SOFEA trial. a PgR-negative status is linked to a reduced chance of response to tamoxifen but not necessarily to AI resistance. 30 Raloxifene Raloxifene is a nearly pure oestrogen antagonist and it has no stimulatory effects on the endometrium. Acquired resistance Peptide growth factor receptor pathways such as EGFR and HER-2 become selectively upregulated in breast cancer cells that acquire resistance to tamoxifen during prolonged exposure. Tamoxifen resistance may be linked to growth factor receptor pathways such as HER-2. Enhanced expression of EGFR and subsequent downstream MAPK activation have been found in MCF-7 breast cancer cells that become resistant to tamoxifen over time. Initial findings from the STAR breast cancer prevention study have shown that raloxifene is as effective as tamoxifen in reducing the incidence of invasive breast cancer in postmenopausal women and is associated with fewer thromboembolic events and endometrial cancers compared to tamoxifen. Role of the progesterone receptor in hormone resistance In ER-positive disease. may be more effective in an environment of ‘low’ compared to ‘normal’ physiological postmenopausal oestradiol levels. Support for this hypothesis comes from the ATAC study in which the benefit of time to recurrence seen with anastrozole compared with tamoxifen was greater in the ER-positive/PgR-negative subgroup compared to the ER-positive/PgR-positive subgroup (Dowsett et al. The agonist activity of tamoxifen is also an important mechanism of resistance. albeit rarely. Possible mechanisms of resistance include the following: r Some of the actions of oestrogens appear to be mediated through transcription factors such as activator protein 1 (AP-1). 2005).. that is. ER status alone does not accurately predict for intrinsic resistance. a downregulator of the ER..

and celecoxib as single agents and in various combinations in this group of patients. Castration can be achieved surgically (subcapsular orchidectomy) or chemically with LHRH agonists (LHRHa). include diethylstilboestrol (DES.d. Bicalutamide Flutamide Cyproterone acetate Diethylstilboestrol Prednisolone Dexamethasone Non-steroidal antiandrogen Non-steroidal antiandrogen Steroidal antiandrogen Oestrogen Corticosteroid Corticosteroid 50–150 mg daily p. 100 mg t.8 mg every 3 months s. The ongoing MRC STAMPEDE study is investigating the benefits of zoledronic acid. There is evidence that the PSA doubling time before and the PSA nadir following ADT are of prognostic significance (D’Amico et al. often with dramatic reversal of the clinical picture.Hormones in cancer Table 3.75 mg every 28 days or 11.4. Prostate cancer: hormonal therapies The hormonal therapies currently used in prostate cancer are summarised in Table 3. ADT produces responses in approximately 85% of patients.d. downregulation of LHRH receptors and desensitisation of the gonadotroph to the normal pulses of LHRH occur. Eventually. docetaxel. Non-hormonal approaches: signal transduction inhibitors EGFR expression in preclinical breast cancer models is associated with increased proliferation. p. patients will relapse. LHRH = luteinising hormone-releasing hormone. 250 mg t. 2006).. 2006. and it produces serum testosterone levels of less than 50 ng/ml. After an initial surge in LH and FSH and a rise in testosterone (unless temporary antiandrogens are used). 1–5 mg daily 5–10 mg daily 0./i. Alternatives in emergency situations. LHRH agonists: goserelin and leuprorelin Mode of action LHRHa are synthetic analogues of the decapeptide LHRH which have increased affinity for the LHRH receptor and reduced susceptibility to protease degradation.s. and inevitably.25 mg every 3 months s. Leuprorelin LHRH agonist 3. Trials investigating adjuvant therapies that might prolong the time to relapse have generally shown no effect or only minimal benefit..c. Reduced serum testosterone leads to reduced expression of androgen-regulated gene products. p. and a poor outcome. Some cells undergo apoptosis (androgen dependent). There is interest in whether EGFR inhibitors may overcome hormone resistance. LH = luteinising hormone. some 31 Androgen deprivation therapy Androgen deprivation therapy (ADT) can be achieved by either castration or single agent bicalutamide (150 mg daily). Rodrigues et al. LH and FSH fall and castrate testosterone levels are reached in 10 to 20 days. In advanced or recurrent prostate cancer. .o.m.o.s. Hormonal therapies currently used in prostate cancer Drug Goserelin Type LHRH agonist Dose/route 3.6 mg every 28 days or 10. The median duration of response is approximately 18 months. which results in increased binding to the receptor.4. FSH = follicle stimulating hormone. 3 to 5 mg daily).c. such as impending spinal cord compression. resistance to apoptosis.o.5–2 mg daily Competitive AR inhibition Competitive AR inhibition Reduces pituitary production of LH and FSH Reduces pituitary production of LH and FSH Adrenal suppression Adrenal suppression Reduces pituitary production of LH and FSH Mode of action Reduces pituitary production of LH and FSH AR = androgen receptor. cyproterone acetate or ketoconazole.

Anonymous. Improved overall survival has thus far only been seen in high-grade tumours (Hanks et al. . LHRHa are reversible. NAHT) or as an adjuvant following radiotherapy (adjuvant hormone therapy. 2006). Two large studies are investigating the role of intermittent hormone therapy in patients with rising PSA (Intercontinental) and patients with metastatic disease (EORTC 30985). Deferred therapy is an option for patients with locally advanced disease who are not having radical therapy or for those with asymptomatic early metastatic disease (e. However. Intermittent therapy in advanced disease The rationale for intermittent therapy is to allow androgendependent cells to repopulate the tumour. LHRHa can be given as monthly or three-monthly depot injections. Hormone therapy as an adjunct to radical radiotherapy Several studies have shown a benefit of the addition of hormone therapy to radical radiotherapy (Table 3.9 years. 2002). (See Chapter 9 for more information. node-negative disease. and so the balance between quality and quantity of life is very important. Studies looking at these individual strategies have mainly included patients with intermediate.84. or cyproterone acetate 100 mg t. Oral administration is ineffective because LHRHa are rapidly cleared by the liver.d.or high-risk disease and have shown a benefit to either approach. Three years of AHT improves five-year overall survival in patients with locally advanced disease who are treated with prostate and pelvic radiotherapy (Bolla et al. 32 Adjuvant hormonal therapy following surgery Two studies have investigated the role of AHT following radical prostatectomy. although they have been too small to show true equivalence.5). In patients with bulky tumours. Progression-free and cause-specific survival were also improved (Messing et al. NAHT has not been shown to benefit patients with small-volume good-risk localised disease. At a median of 11. Maximum cytoreduction occurs at approximately three months. starting at least one week before the first LHRHa injection. which suggests that men with locally advanced prostate cancer and a Gleason score of greater than or equal to 8 would benefit from a combined course of hormonal treatment.d. which was shown only in RTOG 8610 in low-grade locally advanced tumours. 2003). It may also allow the patient some time free from acute side effects and may delay the development of late side effects of ADT. Patients should remain under close supervision and be made aware of the symptoms of progressive disease. p = 0.). Deferred therapy in locally advanced or asymptomatic metastatic disease Patients with locally advanced or asymptomatic metastatic disease are incurable. but subcutaneous or intramuscular delivery gives 94% bioavailability. as the medical alternative to orchidectomy..01 has shown improved outcome in locally advanced tumours over six months compared to three months. Trials show equivalent response rates and survival. 1997). RTOG 9202 investigated the added benefit of two years of AHT in addition to NAHT and prostate and pelvic radiotherapy in patients with locally advanced disease.s. the second study showed improved overall survival with immediate ADT in patients with involved pelvic lymph nodes (HR 1. They may not need immediate hormonal therapy. and some are unaffected (androgen insensitive). AHT). The temporary rise in testosterone (flare) should be prevented by giving three weeks of antiandrogens (bicalutamide 50 to 150 mg daily.Jacinta Abraham and John Staffurth undergo cell-cycle arrest (androgen sensitive). One study showed no benefit in terms of overall survival for the addition of flutamide in patients with locally advanced. ADT has well-documented acute and late side effects that may be avoidable for a considerable duration in certain clinical situations.g. painless rib metastases only. A similar argument exists for the use of single-agent bicalutamide as an alternative to castration in certain situations. TROG 96. Hormone therapy can be used either neoadjuvantly and concurrently with the radiotherapy (neoadjuvant hormone therapy. and they are the preferred management approach (see Chapter 19). NAHT reduces the volume of rectum irradiated (Zelefsky and Harrison. The optimum duration of NAHT is unknown. flutamide 250 mg t.. the L101 and L200 studies showed no additional benefit to adding AHT in patients who had also received NAHT and prostate radiotherapy.) Neoadjuvant hormone therapy causes cytoreduction and may radiosensitise cells. 1997). Radiosensitisation would be expected to improve local control.s. With regard to combined studies. although it is not licensed for single-agent use in metastatic disease. Clinical use Advanced disease LHRHa have been used for more than 15 years for the management of metastatic prostate cancer.04). thereby delaying the development of androgen-resistant clones..

.59. p < 0.66.002) Delayed time to metastatic disease with orchiectomy Improved LDFS. p = 0.0001 for 3 months.58.. FU = follow-up..3 yrs AHT improved 10 yrs bNED (HR 0. 2004.72. 2005. 2003.85. CSS. HT = hormone therapy.7 yrs NAHT better at 8 yrs (HR 0. p < 0. and OS with AHT.5 yrs AHT better (HR 0. LDFS = local disease-free survival. Hanks et al. 5 yr OS 78 vs 62% (HR 0. OS = overall survival.43. 3 or 10 months of 161 MAB starting 3 months pre-RT PrRT + 5 or 10 months of MAB starting 3 months pre-RT PPRT + 4 months of MAB starting 2 months pre-RT ± 2 yrs of AHT 1554 5. LHRH = luteinising hormone-releasing hormone. p < 0.8 yrs 325 3. p = 0. starting 2 months pre-RT Prostate and seminal vesicle RT + 0. Laverdiere et al. 2005.65. Adapted from Bolla et al. MDFS.Table 3.0001) 401 1370 5.004) AHT improved CSS and OS. vs = versus.77. MAB = maximal androgen blockade. and CSS. 2005. RT = radiotherapy. LDFS. and LDFS.0002) Improved outcome significant on subgroup analysis for locally advanced disease (T3–4) (HR 0. p = 0. Tyrrell et al.53. Denham et al.3 yrs 5. no difference in OS PrRT + 0.0001) AHT better (HR 0. MDFS = metastatic disease-free survival. p = 6.59 and 0.56.004) 818 0. p = 0.9 yrs NAHT better (HR 0.. p = 0.59.009) 277 945 7. 2001. or 6 months of MAB starting 2 or 5 months pre-RT PrRT ± 3 months MAB starting 3 months pre-RT Orchiectomy vs PrRT vs orchiectomy + PrRT PPRT + immediate (in last week of RT) or delayed (at relapse) LHRHa PPRT + concurrent and 3 yrs of AHT or LHRHa at relapse PrRT ± adjuvant bicalutamide 150 mg daily for 2–5 yrs Reduced MDFS. 3. PrRT = prostate radiotherapy.75. p = 0.009) 5 yrs Improved bNED for NAHT months See section on combined studies. Studies of neoadjuvant and/or adjuvant hormonal therapies in prostate cancer Study design n Median FU Disease-free survival Other results/issues Trial Patient group Neoadjuvant studies PPRT ± 4 months of MAB 456 p = 0. 1992. CSS = cause-specific survival. .04) 5 yrs No difference NAHT and NAHT/AHT Both improved bNED compared to no HT (HR 0. MDFS. bNED = biochemical disease-free survival.. HR = hazard ratio. yrs = years. 1997.01 Locally advanced L101 T2–3 Adjuvant studies MRC PR02 T2–4 RTOG 8531 T3–4 or N1 (some pT3) EORTC 22863 T1–4 Early Prostate T1–4. p < 0.. PPRT = prostate and pelvic radiotherapy. NAHT = neoadjuvant hormone therapy.76. below 5. N0–1. Fellows et al.7 yrs No difference NAHT and NAHT/AHT AHT better (HR 0. bNED and CSS seen in low-grade tumours only 3 and 6 months NAHT improved bNED.003). each measure improved with 6 months NAHT vs 3 RTOG 8610 Bulky T2–4 N0–1 TROG 96. Pilepich et al.0001) AHT improved bNED.5. M0 Cancer Trial Combined studies L101 T2–3 L200 T2–3 RTOG 9202 T2c-4 AHT = adjuvant hormone therapy. n = number of patients.. MDFS. 10 yr OS 53 vs 38% (HR 0.0001 for 6 months) 161 (HR 0. HR 0. p = 0. unplanned subgroup analysis showed OS advantage for high-grade tumours (HR 0.

2000).03. including impotence. binding results in a conformational shape change. Tamoxifen may also have a role in patients who have already developed gynaecomastia (Boccardo et al. Adverse events Approximately 70% of men suffer breast pain and gynaecomastia. in men with localised disease.. the outcome for patients with non-metastatic locally advanced disease was not statistically different (Iversen et al. 2004). There is an increased rate of all-site and hip fractures (13 versus 20% and 2 versus 4%.44 to 0. The Early Prostate Cancer Trials have assessed the role of bicalutamide alone. However. The breast-related side effects can be most effectively managed by using adjuvant tamoxifen 10 to 20 mg daily. and standard therapy remains LHRHa alone (Schmitt et al. 2005). and impotence.001) and death by 35% (HR 0. 2006). This response . which leads to the production of testosterone.4 years. MAB has been compared with medical and surgical castration alone in several trials and meta-analyses. although the long-term effects on the cancer and the patient are not yet known.93]. weakness. In patients with locally advanced disease who received radiotherapy (n = 305).Jacinta Abraham and John Staffurth Adverse events The general symptoms of ADT are hot flushes. in three parallel studies including 8113 patients with localised or locally advanced prostate cancer (McLeod et al. and reduced energy levels.. At a median follow-up of 7. 2005. 34 Antiandrogen withdrawal Withdrawing antiandrogen therapy has been reported to give response rates of up to 30% in patients who have been responding to antiandrogen therapy. especially emotional lability and depression. They may notice tiredness. bicalutamide significantly reduced the risk of disease progression by 44% (HR 0. thus reducing many of the side effects associated with LHRHa..65 [95% CI 0. Non-steroidal antiandrogens: bicalutamide and flutamide Mode of action The modes of action of non-steroidal antiandrogens include the following: r Competitive inhibition of the AR prevents the binding of DHT in the prostate... asthenia. but unfortunately bicalutamide cannot be used to reverse the loss of BMD seen with previous use of LHRHa.16 [95% CI 0.99 to 1. p < 0.40 to 0. r Stimulation of the hypothalamus results in LH secretion. Bicalutamide is not currently licensed for use in localised disease. sweats. Patients become impotent and have decreased libido. rash.. r Bicalutamide can be used in combination with LHRHa (dose = 50 mg daily) for maximal androgen blockade (MAB) or as a single agent at 150 mg daily (see Chapter 19).56 [95% CI 0. 2003).07). although their effect on fracture rate has not been reported. Breast bud irradiation and anastrazole are less-effective alternatives. ADT also causes a 4 to 10% loss of BMD within the first year of therapy.. p = 0. respectively) in prostate cancer patients treated with ADT compared to those treated without (Shahinian et al. or as an adjunct. and there is a trend towards decreased survival in patients otherwise undergoing watchful waiting (HR 1..78]. altering the balance of coactivator and co-repressor complexes and reducing the effects of DHT.37]. There are well-documented beneficial effects of bicalutamide on BMD (presumably due to increased testosterone and oestradiol levels). there is no benefit from early or adjuvant bicalutamide. 2003). Bisphosphonates can be used to prevent BMD loss in men who are starting ADT (Smith et al. 2005). Overall. and mood changes. advisory groups on both sides of the Atlantic have questioned the cost-effectiveness of MAB. The depression can be profound and depressive psychoses have been reported. 6% report alopecia and weight gain (Wirth et al.. Perdona et al. The largest benefit may be for patients with minimal disease. 10% report hot flushes. Hussain et al. 2006). p = 0. See and Tyrrell. Men with advanced prostate cancer have been shown to have reduced BMD at presentation compared with controls (40 versus 27%. it seems that there is a small overall survival advantage of approximately 7 months (2 to 5% improvement in 5-year overall survival) when MAB is used (with bicalutamide as the preferred antiandrogen). Clinical uses Single-agent bicalutamide 150 mg daily is less effective than surgical castration in patients with metastatic disease in terms of time to progression and overall survival. 2000). weight gain.

The main clinical use of steroidal antiandrogens is to prevent LHRHa-induced tumour flare. It can cause thromboembolic disease. It should be given with aspirin 75 mg daily or full anticoagulation to prevent the thromboembolic disease. Mode of action The exact mode of action is not understood. Hepatotoxicity can occur with jaundice. Pharmacokinetics DES can be given orally and undergoes first-pass metabolism within the liver to form its active metabolites. with response rates of up to 80%. mental changes including hyperactivity and confusion. and abnormal liver function tests. and up to 30% of patients may get an 80% reduction in PSA. Steroidal antiandrogens are also associated with an increased rate of thromboembolic disease. bicalutamide following flutamide or nilutamide following bicalutamide). hepatitis and abnormal liver function tests. which results in stimulation by the antiandrogen rather than inhibition. Glucocorticoids: prednisolone or dexamethasone Mode of action The mode of action of glucocorticoids involves suppression of adrenal androgen production: r Response rates to prednisolone/prednisone (5 to 20 mg daily) and hydrocortisone (30 mg daily) are in the order of 30% (PSA decrease of at least 50%). with a median duration of response of about 2 months. Nausea and impotence also occur. r Reduced testicular production of testosterone. hepatitis. Adverse effects Typical glucocorticoid toxicities include proximal myopathy. complex. which is prevented with the use of preDES radiotherapy (7 to 10 Gy). and increased risk of osteoporosis. r Progestational and glucocorticoid activities. r Reduced LH/FSH production from the pituitary. r Interference with the intracellular binding of DHT to the androgen receptor. r Dexamethasone (0. Clinical use DES (3 to 5 mg daily) can be used as a single agent to achieve rapid castrate testosterone levels. The median time to progression is approximately 7 months. Reduced libido and erectile dysfunction also occur. Steroidal antiandrogens: cyproterone acetate. Transcutaneous application may bypass the first-pass metabolism and thus reduce the liver. Gynaecomastia occurs. Exogenous oestrogens: diethylstilboestrol and ethinyloestradiol Diethylstilboestrol (DES) is the only oestrogen that is used in standard clinical practice. Adverse events DES causes sodium retention with oedema and can precipitate congestive cardiac failure. thin skin with easy bruising. The lowest effective dose should be prescribed. PSA responders have a median survival of approximately 15 months.g. It can also be used in combination with LHRHa at 1 mg for hormone-refractory disease. but it is likely that DES functions in multiple ways to reduce androgenic activity: r Suppression of adrenal androgen production. and thromboembolic side effects (being assessed in the ongoing MRC PATCH study). cardiac. 35 . Mode of action The biochemical effects of steroidal antiandrogens are multiple. and medroxyprogesterone acetate Cyproterone acetate is licensed for use in advanced prostate cancer and for suppression of tumour flare with initial LHRHa.Hormones in cancer is thought to be due to mutation in the AR. r Reduced steroid synthesis leading directly to reduced testosterone levels. Their hepatotoxicity includes jaundice. Adverse events Fluid retention can occur which can precipitate congestive cardiac failure and cause ankle oedema. cushingoid facies. megestrol acetate.5 to 2 mg daily) is highly active. Re-challenging with an alternative antiandrogen may result in further responses (e. hypertension. r Reduced LHRH and LH release. and not well defined: r Antiandrogenic.

(1896). J. U. 295–300. Urol. whose progeny are primarily androgen sensitive in the presence of androgens. (2005). testicular discomfort.. and cetrorelix lead to immediate competitive inhibition of LHRH receptors. G. but after ADT the less-dependent cells grow more and become the dominant clones. These include the following: r AR mutation – increased activation by low levels of androgens or low-potency androgens. J.e. 2005). and there is a more rapid fall in PSA than occurs with LHRHa. An alternative model is that androgen-sensitive cells initially enter cell-cycle arrest. Oncol. Selective CYP1B1 inhibitors: abiraterone acetate Abiraterone acetate is a steroidal inhibitor of a key adrenal enzyme involved in the androgen production pathway (Attard et al. the progeny become increasingly androgen resistant. M. Battaglia. 96. Toxicities include mild mood variations. Clin. It has good bioavailability at doses greater than 200 mg. Toxicity includes histamine-mediated allergic reactions. and headaches. and de Bono. Belldegrun. Boccardo.. 23. Lancet. 1241–6. the majority of successful hormonal manoeuvres used in LHRHarefractory disease reduce adrenal androgen production. G. They are given as 36 . 2004). Rubagotti. and castrate testosterone levels within one week. 337. J. e. A. The Medical Research Council Prostate Cancer Working Party Investigators Group. N. They are Food and Drug Administration (FDA) approved on a voluntary risk management programme. Selective blockade of androgenic steroid synthesis by novel lyase inhibitors as a therapeutic strategy for treating metastatic prostate cancer. Evaluation of tamoxifen and anastrozole in the prevention of gynecomastia and breast pain induced by bicalutamide monotherapy of prostate cancer. REFERENCES Anonymous. However. Improved survival in . flushing attacks. D. F. T.g. Attard. primarily in hormone-refractory prostate cancer. J. On the treatment of inoperable cases of carcinoma of the mamma. several approaches are being investigated. Non-hormonal approaches In patients with hormone-refractory disease. Bolla. It has been postulated that the malignant cells are prostate cancer stem cells. (1997). including signal transduction inhibitors. via upregulated transcription factors such as those in the EGFR pathway). 104–7. Studies are under way in both hormone-refractory and initial hormone therapy scenarios (Weckermann and Harzmann. Presently these new LHRH antagonists are only licensed in the United States for patients unable or unwilling to be treated with LHRHa and willing to accept the risk of histamine-mediate allergic reactions. After ADT some cells undergo apoptosis (androgen dependent) and some undergo cell-cycle arrest (androgen sensitive) while others are unaffected (androgen insensitive).. r Upregulated survival pathways (i. r Upregulated levels of AR mRNA and protein – this may also alter AR antagonists to give them some agonist function. abarelix. (2005). and immunotherapy. but over time they develop molecular pathways that overcome the low/absent circulating androgens. Engl..Jacinta Abraham and John Staffurth Prostate: mechanisms of resistance The underlying mechanism for hormone resistance is still largely unknown.. Br. Warde. Int... via BCL2 or survivin). Studies are underway. r Upregulation of the ARE in the absence of activated AR (e. 808–15. Med. Active agents may be effective in earlier stages of disease. 79.. abarelix. B.g. Gonzalez. 235–46. S. S. There is no testosterone surge (or flare). M. and cetrorelix Degarelix. But in the absence of androgens or with low circulating levels of androgens. ii. inhibition of upregulated antiapoptotic pathways. et al. P et al. reduced LH and FSH levels within 8 to 24 hours. in 75% of men. J. patients with locally advanced prostate cancer treated with radiotherapy and goserelin. A. Because the adrenal glands produce 5 to 10% of circulating androgens. r Continuously activated AR in the absence of androgens. but there is a fairly large interpatient variability. Beatson. even with MAB. Suggestions for a new method of treatment with illustrative cases. Prostate cancer: areas of current interest New LHRH antagonists: degarelix. reduced apoptosis.. It is thought that before treatment with ADT there is a predominance of androgen-sensitive cells. Immediate versus deferred treatment for advanced prostatic cancer: initial results of the Medical Research Council Trial. patients relapse.. (1997). a subcutaneous injection every 4 weeks because they are not orally bioavailable.

132 (Suppl. 6. Urol. K. A randomized clinical trial evaluating tamoxifen in the treatment of patients with node-negative breast cancer who have oestrogenreceptor-positive tumors. et al. Catalona. D. (2005).. E. W. M. Radiat. Hall. (2006). (2005). 97.17. A.. N. Int. Cancer Res. et al. Nabid. Gnant. 3972–8. Cochrane Database Syst. A... Iversen.. and Tyrrell C.. (2004). Clin. 97.. 7–16.. Cancer Res.. W. A. T. mg plus standard care vs standard care alone for early prostate cancer. Miller. N. C. Phys. J. Dal Lago. H. Jakesz.. et al. P B. Seidenfeld. McLeod.. See... radiotherapy. Production of Nb rat carcinoma of the dorsal prostate and response of estrogen-dependent transplants to sex hormones and tamoxifen. C. C. T. (2000). 61. Oncol. J. (2005). Predictors of .. and Hodges. Denham. P Tyrrell. Oncol. (1952). 60–2. Long-term results with immediate androgen suppression and external irradiation in patients with locally advanced prostate cancer (an EORTC study): a phase III randomised trial. F. Oncol.... J. Cancer Res. Int. 2751–60. M. et al. V. R. Biol. J. 320. L. 1137–40. 2692–705.. et al.. 171. 366. C. S. W. L. (2004). advanced localised prostatic cancer by orchiectomy. Kaisary.. 1. 10. 7. U. G. D. D. J. Int. Short-term androgen deprivation and radiotherapy for locally advanced prostate cancer: results from the Trans-Tasman Radiation Oncology Group 96. Inhibition of human mammary and prostatic cancers by adrenalectomy. T... G. Rev. Beynon.Hormones in cancer Bolla. Oncol. 70. J. Hussain. et al. V. R. et al. Phase III trial of long-term adjuvant androgen deprivation after neoadjuvant hormonal cytoreduction and radiotherapy in locally advanced carcinoma of the prostate: the Radiation Therapy Oncology Group Protocol 92–02. A Medical Research Council Study. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials.. Dowsett. V. 1262–71. et al. R. J. A. (2005). (2006). Breast cancer patients with progesterone receptor PR-A-rich tumors have poorer disease-free survival states. Bennett. Yao. 1687–717. P See. Jonat. H. D. A.. (2006). (2005). Cancer. (2001).. Pilepich. Pilepich. E. Phys. Gibson. Lancet Oncol. Stephenson... Engl. 841–50.. J. C.. M. 7512–17. W. Winter. Lawton.. Ingle. et al. Martino.. et al. 65. B. B. 295–300. J. et al. Biol.. monotherapy compared with castration in patients with nonmetastatic locally advanced prostate cancer: 6. Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: Updated findings from NCIC CTG MA. Studies on prostate cancer. Bicalutamide . et al. G.. Cancer Res.. Med. E. EBCTG.. M... Anderson. J. 21. Wale.. 1579–82. J.. Predictors of mortality after androgen-deprivation therapy in patients with rapidly rising prostate-specific antigen levels after local therapy for prostate cancer. Urol. What clinicians need to know about antioestrogen resistance in breast cancer therapy. et al. Br. Oncol. J. Treatment of . et al. Cancer Res... Results of the ATAC (Arimidex. et al. Clin. L. J. Oestrogen receptor beta and neoadjuvant therapy with tamoxifen: prediction of response and effects of treatment. Redmond. 656–60. Clin. Pajak. W. J. De Bedoya. C. J. Cancer. Phys. (2002). Nat... mortality after prostate-specific antigen failure.. et al. Engl. S. L. S. 50. 360. A. Fisher.... J. Manola. Lancet. Androgen suppression adjuvant to definitive radiotherapy in prostate carcinoma – long-term results of phase III RTOG 85–31. L. 134–41. S. Radiat. 1333–8. Biol.. et al. Oncol. J. Urological Cancer Working Party – Subgroup on Prostatic Cancer. J. Urol. B. 92. et al. J. 690–4. Lancet Oncol. M... Int. (2003). Huggins. Porter. (2006). M. Schmitt. P. Rodrigues. J. Sotiriou. Costantino. L. M. Phase III radiation therapy oncology group (RTOG) trial 86–10 of androgen deprivation adjuvant to definitive radiotherapy in locally advanced carcinoma of the prostate. Retrospective analysis of time to recurrence in the ATAC trial according to hormone receptor status: an hypothesis-generating study. 479–84. K. Bicalutamide 150 . R. Alone or in Combination) trial after completion of 5 years’ adjuvant treatment for breast cancer. CD001526. A. J. V. Efficacy of tamoxifen and radiotherapy for prevention and treatment of gynaecomastia and breast pain caused by bicalutamide in prostate cancer: a randomised controlled trial.. (2006). A. Hilsenbeck. J. The addition of bicalutamide 150 mg to radiotherapy significantly improves overall survival in men with locally advanced prostate cancer. D’Amico. Noble. Laverdiere. Clin. Eur.. V.. E. Fellows. M. Cancer. (2005). N. A. 365. L. J. J. N. Weston. Howell. A.... John. Perdona. R. Cancer Inst. Autorino. Goss.01 randomised controlled trial. Tamoxifen. (1989). Weiss. (2005). L. Maximal androgen blockade for advanced prostate cancer. (2006). Huggins. et al.. Messing. J. 13). Milano. 1285–90. Steigler. Winter. J. Coombes. Med. Int. Lancet. C. (2000). of oestrogen and of androgen injection on serum phosphatases in carcinoma of the prostate. Lancet. Radiat. M. M.1243–52. Kantoff. (1992). 3547–50. M. J. J. 94. L. C.. 40. 1081–92. Blank. Lancet Oncol. W. Baum. (2006). B. and Bergenstal.. Switching of post-menopausal women with endocrine-responsive early breast cancer to anastrozole after 2 years’ adjuvant tamoxifen: combined results of ABCSG trial 8 and ARNO 95 trial. (2003). Dixon. S. Chen. N. 455–62... (2005). 23. J. A. Lancet.3 years of followup. 103–6. 365. et al. and Lamb. 2.. Cuzick. 164. Hopp. (1980). 247–54. 42. P Loffredo. J.. (2004). Br. J. U. A.. The effects of castration. 472–9. (1941). 37 . J. 12. J. R. De Placido. G. M. 6. C. et al. 293–7. A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. Immediate versus deferred androgen deprivation treatment in patients with node-positive prostate cancer after radical prostatectomy and pelvic lymphadenectomy. 350. Iversen. The efficacy and sequencing of a short course of androgen suppression on freedom from biochemical failure when administered with radiation therapy for T2-T3 prostate cancer. et al. Immediate dual energy X-ray absorptiometry reveals a high incidence of osteoporosis in patients with advanced prostate cancer before hormonal manipulation. Clark. 304–9. S. or combined treatment. A. Collette.... 514–20. Cuzick. et al. 107. R. Hanks. C.

J. J.. et al. J. A. Zondek. I. D. Payne. 38–45.. McLeod. Bicalutamide 150 mg in addition to standard care in patients with localized or locally advanced prostate cancer: results from the second analysis of the early prostate cancer program at median followup of 5. 2008–12. M.. L. Wirth. et al. Urol. Y. M. Oncol. 2444–7. Oncol. C.. Neoadjuvant androgen ablation prior to radiotherapy for prostate cancer: reducing the potential morbidity of therapy. 1850–54. S. (2005). Clin. (2005).. Med. A. J... Bicalutamide (‘Casodex’) 150 mg as adjuvant to radiotherapy in patients with localised or locally advanced prostate cancer: results from the randomised Early Prostate Cancer Programme.. Radiother. Engl.. B. J. Eastham. Tyrrell.. Y. et al. J. 154–64. M. D. 49 (Suppl. G. 76. and Harrison A. Freeman.4 years.Jacinta Abraham and John Staffurth Shahinian. D. J. 169. Eur. and Harzmann R. Smith. Kuo. (2005). discussion 283–4. 279–83. A. Smith. J. R. A. B. B. Yap. V. J. A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. J.. (2004). F. (2006). Risk of fracture after androgen deprivation for prostate cancer... Med. 46. M. et al.. Gleason. (1940). Dowsett. (1997). P. N. Effect of prolonged administration of oestrogen on uterus and anterior pituitary of human beings. Zelefsky. 24. 352. Hormone therapy in prostate cancer: LHRH antagonists versus LHRH analogues. Thurlimann.. N.. Adjuvant aromatase inhibitors for early breast cancer after chemotherapy-induced amenorrhoea: caution and suggested guidelines. M. on behalf of the BIG1–98 Collaborative Group. Engl. 1) 1865–70. M. et al. See. See. 114. 38 . J. Urol. 4–10. W. Urol. 3A). 353. Weckermann. W. 172 (5 Pt. (2003).. 2747–57. Urology. Randomized controlled trial of zoledronic acid to prevent bone loss in men receiving androgen deprivation therapy for nonmetastatic prostate cancer. E. (2004). H.

wedges. This chapter will deal with the general principles of developing isodose plans that are suitable for treatment and will use specific examples to highlight particular points. for breast treatments). r Radiotherapy technique. r Prescription. adopting the stance of a cross-country skier (one arm and the opposite leg forward. when the patient may stand up. and calibrated couch movements are all used.e.g. there will be shorter sections on the use of electrons and brachytherapy. providing that a careful set-up.g. 382). suitable reference points. One by Shiu and Mellenberg (2001) includes sections on isodose planning. Another. treatments can be delivered to within ± 5 mm. The following are some commonly used devices: r Simple head. knee or ankle supports which help to stabilise the patient. r Definition of target volumes and organs at risk. orthogonal films. CT scanning. These techniques. including beam arrangements. A wide variety of different immobilisation devices exists. Which is most appropriate depends on the position of the tumour. and total skin electron treatment. and / or during treatment using portal imaging. dose. are used by the treatment centre’s Physics Department for checking the validity of calculations before they are applied to patients. and it is important to understand the methods of immobilisation and the inevitable uncertainties in the delivery of treatment. Radiotherapy is a rapidly developing field and it is important to ensure that new methods satisfy safety and effectiveness requirements before being adopted as routine treatments. This can take place before treatment in the simulator or CT simulator. weighting. size and shape. Some aspects of radiotherapy planning using lower energy (kilovoltage) photons will be discussed in Chapter 33 (see p. by Purdy (2000). and anyone interpreting plans will need to be familiar with them to know whether it is worth adjusting treatment plans during review. when the patient may lie in the lateral position with the arms in front of the chest to provide lung compensation. the other arm and leg back). There are several useful reviews of radiotherapy physics in the literature. and production of isodose plan. energy. the need to protect adjacent structures. Quality assurance (QA) is essential and it is of paramount importance for each member of the planning team to be familiar with the overall process. For patients who do not have very mobile superficial tissues. . including number of phases. Radiotherapy treatments that require unusual treatment positions include total-body irradiation. beam energy. position and immobilisation. provides a perspective on future directions in 3D treatment planning. checking the geometrical set-up of the treatment). and fractionation.4 RADIOTHERAPY PLANNING Andrew Tyler and Louise Hanna Introduction It is important to understand the basic techniques of radiotherapy planning because these will help when developing complex plans. Variables include whether the patient is prone or supine. which are taught at the FRCR Part I level (Royal College of Radiologists). 39 Treatment planning overview Radiotherapy planning can be divided into stages as follows: r Patient preparation. r Localisation method (e. and image co-registration). Patients must be positioned to allow the optimum delivery of the radiotherapy while maintaining comfort. It will focus mainly on external beam radiotherapy with megavoltage photons. and the treatment intent. and the position of the limbs or the neck. r Chest board with cross bar to fix the arms above the head (e. r Verification (i. Patient position and immobilisation For radiotherapy to be effective it must be delivered accurately.

to design the clinical target volume (CTV) and planning target volume (PTV). The narrower the slice width. such as breathing. from which treatments can be set up. in which the patient is scanned in the same position. However.. etc. The target volumes are delineated on orthogonal X-ray images and then transferred onto an outline of the patient taken at one or more levels. lymphoma. may be “frozen” at any point in the cycle on the CT scan and the expansion of the CTV to PTV must take this into account. flat couch. 40 ‘Conventional’ radiotherapy planning Conventional radiotherapy planning has been largely superseded by CT planning. There has been debate whether the use of contrast alters the dosimetry. The wide bore. the field borders themselves are defined on a radiograph or simulator image. Nutting et al. Corrections for MRI distortions must also be validated carefully. for some treatments using simple beam arrangements such as a parallel opposed pair of fields. It is important to be certain that the immobilisation device is effective. but there is software by which the ‘diagnostic’ MRI scan can be co-registered with the planning CT scan to aid in the planning process. It is important to include the proposed target volume and the organs at risk in the scan to allow dose volume histograms (DVHs) to be calculated. such as PET/CT. to be identified securely. The planning CT scan is usually taken using a slice interval of 1. MRI/CT co-registration is of limited value for some sarcomas.g. and to outline critical structures in accordance with ICRU reports 50 and 62 (ICRU. CT simulation A CT simulator is one of the mainstays of 3D conformal planning. allows patients to be scanned in the treatment position and for reference marks. r Vacuum-formed shells. Alternatively. Automatic electronic data transfer to a treatment planning system is required to enable the treatment calculation to be performed. vacuum-formed shells for other anatomical sites do not usually give this reliability.). moulded thermoplastic shells (heated in warm water) or Perspex® shells formed from individualised plaster moulds. and the authors concluded that the treatment accuracy was not improved when compared with the conventional technique (using standard ankle stocks. (2005) have found it makes little difference in dose computation. MRI/CT co-registration is most accurate for tumours that are close to bony structures with characteristic features.Andrew Tyler and Louise Hanna r Vacuum bags offer customised support (e. 1999). It provides precise imaging and positioning information for the treatment planning process. it is increasingly being used to inform the treatment planning process. laser positioning. MRI scans are better than CT scans for delineating the GTV. Tissue definition. and colorectal cancers. whereas a treatment fraction can take longer to deliver. These borders represent the PTV with a small margin to take account of the beam penumbra. For some tumour sites. These also require high-level QA on treatment machines. such as the brain. for example near the skull base. the better will be the resolution in the digital reconstructed radiograph (DRR). 1993. especially with the advent of PET/CT where the automatic co-registration of the images can be used as a direct input into the planning process. particularly . r Complete fixation devices such as stereotactic frames with mouth bites give set up accuracy to within about 1 mm. Imaging the patient Imaging the patient is essential to delineate gross tumour volume (GTV). 2000b). CT scans are taken over a few seconds. Nuclear medicine Although radiopharmaceutical imaging tends to be used for diagnostic purposes. One example of this is during planning for lung cancer. mounts for immobilisation devices. Because of this. This is easier in dual-modality machines. but Lees et al.g. For example.5 to 5 mm. Applications currently in development include lung. normal patient motion. Administration of intravenous contrast helps to distinguish tumour from normal tissues. for limbs when it is necessary to support the limb in a position that allows access for beams). inaccuracies in set up can be reduced to within 3 mm. It is important to make sure that not only the position and scaling but also the rotations have been faithfully applied. It is typically a wide-bore CT scanner with virtual simulation software (VSIM) and radiotherapy accessories (e. It is less helpful for tumours that can move easily relative to bony landmarks or for those not close to bony features. although theoretically providing a slight reduction in image quality. By using wellfitted vacuum-formed shells for head and neck treatments. Therefore. one study looked at the use of a customised immobilisation system (Vacfix® cushion) for the treatment of the prostate.

.Radiotherapy planning for soft tissues. However. r Treated volume – the volume enclosed by an isodose surface. r Planning target volume (PTV) – the CTV and a margin for uncertainties. PTV = planning target volume. 1993. the internal mammary chain lymph nodes lie between 1 and 6 cm deep and up to 4 cm from the midline). which is added to account for involuntary changes in the organs that surround the CTV. r Irradiated volume – the tissue volume that receives a dose that is considered significant in relation to normal tissue tolerance. Senan et al. Some organs are more difficult to delineate than others. Maryland: International Commission on Radiation Units and Measurements. r Clinical target volume (CTV) – the GTV and/or suspected subclinical tumour. 1993: ICRU Report 50. The target volume is delineated with all the relevant clinical information available including medical history. GTV = gross tumour volume. When marGTV CTV PTV Treated volume Irradiated volume Figure 4.1. 1998. CTV = clinical target volume. the tumour volumes may shrink during radiotherapy. Although the treatment machine delivers radiation with an accuracy supported by elaborate QA. Any reduction in the treatment margins spares more normal tissues and introduces an opportunity for considering dose escalation for the tumour. 2000. Defining the target volumes Standards The publications ICRU 50 and ICRU 62 (ICRU. pretreatment diagnostic and staging investigation results. 2003). operation notes and pathology reports. head and neck...g. Jansen et al. it becomes more difficult to be sure that the tumour is being treated. GTV. Recording. it is not possible to identify subclinical disease by imaging or clinical examination. Redrawn with permission from ICRU. Anyone taking part in the planning process should be familiar with the terms defined within. Growth from GTV to CTV By definition. 2000.g.1 shows tumour and target volumes as defined in ICRU 50 (ICRU. As a brief recap. there are many other uncertainties in the planning process. 41 . scan images. 1993. 1999): r Gross tumour volume (GTV) – the demonstrable tumour. Discussion with the multidisciplinary team. tumour eradication. 2004). selected and specified by the radiation oncologist as being appropriate to achieve the purpose of treatment (e. together with the OAR and PRV. in the head and neck. palliation). and lung tumours (Apisarnthanarax et al. the following terms are used to ensure accurate treatment records are kept (ICRU. is not clear when using orthogonal films. 1999) will be referred to throughout this text. which accounts for the uncertainties and lack of reproducibility in setting up the patient day by day. CTV and PTV. r Planning organ at risk volume (PRV) – the organs at risk with a suitable margin to account for movements and uncertainties in set-up (see p. Prescribing. International Commission on Radiation Units volume definitions. which may be systematic or random. Bethesda. 42). dedicated radiotherapy planning meetings and involvement of a specialist radiologist can facilitate this process. but other imaging techniques such as MRI may be helpful.. can be defined. Figure 4. and Reporting Photon Beam Therapy. For example.. Gregoire et al. gins are smaller. r Organs at risk (OARs) – critical normal structures the radiation sensitivity of which may significantly affect treatment planning and/or prescribed dose. considerable variation between GTVs has been observed among different radiation oncologists (Logue et al. 2006. Guidelines have been produced for CTV definition in high-grade gliomas. r the set-up margin (SM). including our inability to define microscopic disease. and are taken into account by r the internal margin (IM). The three volumes. Weiss and Hess. Individual patient variation will add to the uncertainty (e. 1993). Delineating the GTV Delineating the GTV involves outlining the demonstrable tumour. There can be more than one OAR.

1999).g. The dose is expressed as a percentage of the maximum dose deposited by the beam. The effect of radiation on the function of parallel organs is more dependent on the volume irradiated. In general it will be necessary to use a beam that is wider than the PTV to allow the 95% isodose to encompass the PTV. With megavoltage beams the dose is deposited by secondary electrons. active breathing control. the accuracy of treatment planning and delivery are important to ensure that tolerance is not exceeded.. r Changes in the position or size of the CTV (both during and between fractions) must be taken into account. which travel primarily in a forward direction. will move during breathing. r The uncertainties associated with these margins have been estimated (BIR. if an asymmetric beam is created. and. r Uncertainties of patient set-up. up to a centimetre each way anterior–posterior (A–P) and the CTV of tumours in the lung. the shape of the profile may also be noticeably asymmetric. but the size of the variation can to some extent be estimated and accounted for. Techniques to reduce tumour movement have been used in thorax treatments such as breath hold. r Automatic growth tools on the planning computer may help. the dose reduces over the distance of some millimetres. but in detail in other texts (e. these make up the SM. 2003. and so forth also need to be taken into account. the prostate and seminal vesicles can move with changes in the volume of the bladder or rectum. 2000). The width of the radiation beam increases linearly with distance from the treatment head. Beam divergence. the heart). This means that. This maximum dose occurs at a depth dmax . In practice. The penumbra is the distance between the 80 and 20% isodoses. Radial profile. or depressing the chest to reduce the movement. For example. 1976). 1999). Principles of the isodose plan: photons Basic beam data It is important that the concepts of basic beam data are understood. and pelvis. Beam penumbra. The BIR (2003) document uses the concepts of systematic and random errors. some key examples by clinical site provide excellent guidance. As a result there is buildup of dose below the skin surface before dmax is reached. such as the spinal cord. alignment of beams. Lung tumour motion can be in the order of 2 cm during a normal breathing cycle. These form the IM. At the edge of the radiation beam. as part of the document. Others such as the lung may be described as ‘parallel’ and have a low ‘relative seriality’ (ICRU. Williams and Thwaites. depending on the lobe of the lung affected (Seppenwoolde et al. These natural variations are inevitable. These are described briefly here. Khan. r The combination of these two margins around the CTV forms the PTV (ICRU. 2003) and methods of calculation have been introduced that ensure that 95% of the prescribed dose covers the CTV in 90% of cases. r Options such as image-guided radiation therapy (IGRT) and respiratory gating can be employed to reduce the required margins. The radial profile of a megavoltage beam changes with depth due to the differential hardening across the beam. Percentage depth dose. may be described as being ‘serial’ with a high ‘relative seriality. abdomen. Delineating organs at risk OARs are normal structures with a high sensitivity to radiation and it is important that they are delineated on the plan. this applies most commonly to the spinal cord. In a similar manner to the construction of the PTV. The function of serial organs may be seriously affected if even a small portion is irradiated above a tolerance dose. Buildup changes also occur at interfaces between tissues of differing densities. These topics are all covered in part 1 of the FRCR course (RCR) and are discussed in ICRU 24 (ICRU.’ 42 . Within some organs (e. but the mechanism should be understood before using them and they should always be used with care. Some. a margin must be added to make the planning organ at risk volume (PRV). Internal movements are significant in the thorax.Andrew Tyler and Louise Hanna Forming the PTV The dose delivered to the CTV or OARs will not be the same as that shown on the plan if they are not in exactly the same place during all the treatments. particularly those nearest to the diaphragm.g. In some centres the tissue phantom ratio (TPR) or tissue maximum ratio (TMR) is used instead of the percentage depth dose because each is easier to use with isocentric calculations. there may be a combination of serial and parallel structures. 2002). For a serial organ.

2 and 3). it is possible to identify the general shape of the high-dose volume.2. In prostate planning. Beam 2 Choosing beam directions The choice of beam directions will depend on the treatment intent and the locations of the PTV and OARs. with radical oesophagus treatments. It is important to identify potential regions of hot and cold dose. 43 . the treated volume in the plane shown will be approximately the hexagon enclosed by the beam edges. are set at right angles for the floor of mouth and maxillary antrum. Wedges are needed to reduce the inhomogeneities that arise in this situation. The inverse-square law causes the intensity of the radiation beam to decrease but the percentage depth doses below dmax to increase. The shaded area represents the volume that received the highest radiation dose. rather than posterior obliques. It is also important to understand beam weighting so that the relative proportion of dose delivered to the tumour from each beam can be adjusted (Williams and Thwaites. the crosssectional area of the beam increases because of divergence. Changes with field size. By tracing the edges of the beams.g. allows beam shaping to be more effective in reducing the dose to the rectum. r Combinations of beams are usually used for radical treatments in sites such as the chest. pelvis and brain. using oblique laterals instead of the posterior beam would cause an excessive dose to the lungs. such as head and neck. provided that the beams are well balanced and that there is a homogeneous dose distribution across the PTV. Isodose shapes for combinations of beams The directions of the beams will influence the shape of the treated volume. r non-opposing beams. A two-phase technique that combines both of these approaches can prevent tolerance doses from being Beam 1 Beam 3 Figure 4.2. which can improve the dose distribution in the PTV and reduce the dose to organs at risk. breast and limbs. For example. or for tumours that are not at or near the midplane (e. The global hot spot in a balanced multibeam treatment is likely to lie close to the centre of the treated volume. r For two beams. spinal tumours having palliative treatment). for example. Noncoplanar beams are particularly useful in central nervous system (CNS) and head and neck treatments. there is no collision between the treatment head and the couch or the patient). 2000).g. In the scenario presented here there are three beams (beams 1. If the external contour of the patient is irregular in all directions around the central axis of the beam then external tissue compensators or ‘field-in-field’ techniques using multileaf collimators (MLCs) will probably be required. the exit beams should avoid such critical structures as the contralateral eye and spinal cord. by rotating the table and the gantry. the central axis receives more radiation per monitor unit because of increased scatter from the machine head and within the patient. exceeded. In Figure 4. but it is important to ensure that the beam direction can be achieved (e. Non-coplanar beam arrangements can be achieved easily by linear accelerators. Beam arrangements Typical beam arrangements are as follow: r A single beam is used for superficial tumours such as skin cancers. where the use of multiple beams spares adjacent normal tissues. Use of lateral fields.Radiotherapy planning Changes with distance. delivering the dose entirely by anterior and posterior parallelopposed fields would exceed spinal cord tolerance. an anterior beam is combined with two wedged lateral fields. Treatment plans can frequently be improved with the use of non-coplanar beams. Similarly. r opposing beams are used for palliative treatments or for treatments in sites that have a relatively small separation. and it is formed by the edges of the beams. In head and neck treatments. As the distance from the treatment machine to the patient increases. As the field size increases. Beam edges creating the edge of the treated volume.

Examples include the following: r The superior and inferior (S–I) ends of a coplanar plan because of the lack of dose scattering into the volume from outside the beams.. Software can allow co-registration of images from other modalities such as MRI or PET. Use of a field-in-field technique. r Accessory factors (e. Beam modifications: shape. r Field size (output) factor. which is a line graph showing the proportion of the volume that receives at least a specified dose of radiation.g. 1995). r To even out the dose when two or more beams intersect. then it may be necessary to consider using a custom-built tissue compensator. A potential problem is leakage between the abutted leaf ends. this can be reduced if the collimator jaws are positioned as far as possible over the leaf abutments (Klein et al. randomised trials of prostate treatments have shown that the use of conformal radiotherapy reduces rectal toxicity. and tissue compensators Beam shaping Shaping the beam around the PTV spares more normal tissue. in order to avoid hot and cold spots within the treated volume. 1999). trays). 44 . the target volumes and PRVs can be delineated individually on each slice of a crosssectional imaging study such as a CT scan. 2000): r Percentage depth dose (PDD) or tissue maximum ratio (TMR). although they may be needed less often in the future as intensity-modulated radiation therapy (IMRT) develops. After identifying the relevant volumes. and it is not possible to go through each variation in this chapter. They can also be used in other situations such as 3D radiotherapy of the breast. it is important to be familiar with the local method before starting any calculations involving patients. the greatest doses for this type of plan will lie underneath the thin ends of the wedges. Calculation cycle Everybody involved in a patient calculation should know the whole calculation process and where each factor is applied in their own department. The conformity achieved by MLCs is limited by the width of the leaves and their direction of movement. it is then possible to simulate the positions of treatment beams and shielding and to use algorithms to calculate and display the resulting isodose curves. The method of calculation varies from centre to centre. depending on how the factors are defined. Surface obliquity Surface obliquity will cause isodoses to tilt in the same directions as the surface.. with an anterior and a lateral field). where regions of underdose are topped up with small beamlets. wedges. it can be corrected for by placing a wedge in the beam. Because of this. DVHs are a convenient Wedges Wedges are used in two situations where there is an unwanted dose gradient that is relatively uniform in one direction: r To act as a tissue compensator for a sloping external surface (see discussion of surface obliquity that follows). If the obliquity occurs only in one direction.g. but this obviously has implications for the dose to normal tissue on that side. Tissue compensators are used most frequently in head and neck treatments. can help to reduce inhomogeneities of dose. If it occurs in both the S–I and A–P directions. For example. r Wedge factor. r When a wedged pair is used (e. r Inverse-square law. it is common for the postero-medial aspect to be underdosed. even if many of the factors are applied by the planning computer programs. Outside of the treated volume. Making low-melting-point alloy blocks requires time and handling of heavy and hazardous substances. Traditionally. but in many cases field shaping can be achieved with MLCs. It is possible to top up the beam with a small amount of dose from the contralateral side. low-meltingpoint alloy blocks have been shaped individually for each field for megavoltage treatments. The respective doses to the PTV and PRVs can be displayed in the form of a DVH. Factors (and their relative values) that need to be understood follow (Williams and Thwaites. Using a 3D planning system Production of the isodose plan With a 3D planning system. and this has led to studies of dose escalation with a view to improving local control (Dearnaley et al.Andrew Tyler and Louise Hanna There are occasions when it may be difficult to prevent the dose to the PTV from dipping low.

Nearly all planning systems will estimate these effects. Treatment planning systems will make corrections for changes of density pixel by pixel. significant volumes of normal tissue are irradiated to a high dose. In addition to the artefacts seen on CT imaging. 1993). in Figure 4. such as the oesophagus. by using the CT values. from a beam passing through an aerated lung can increase by as much as 15% when a correction is applied. Various methods have been used to minimise any variations in dose that occur when beams are matched together. for example. it is not always necessary to avoid an OAR completely. in the thorax it may be appropriate to pass a posterior beam through the spinal cord to reduce lung dose and meet the appropriate DVH constraint. Selection of the ICRU reference point The recommendation for reporting doses is based on the ICRU reference point. According to ICRU 50 (ICRU. the extra penetration of the beam is usually slightly greater than 2% per centimetre of the lung traversed.3. This can lead to uncertainties in dose when adjacent structures have different densities. bones tend to have little effect on the dose. but sometimes this is not possible. r It should be selected in a region where there is no steep dose gradient. but obviously at a cost of dose to the deeper posterior structures which could include the brain stem. it is important to check the tolerance doses carefully. the planning algorithm may not be able to correct properly for the veryhigh-density material. care must be taken because. If one or more beams exit through an OAR. In general. which means that only the primary beam is corrected. a steep wedge angle on the lateral beam allows extra dose to be given from the anterior beam. although it is possible to produce what looks like a perfect plan with perfect junctions between the beams. When two diverging beams are adjacent to each other. depending on the depth at which the beams meet. This is because in a well-aerated lung. Doses to organs at risk (OARs) Outside the PTV. and the core of long bones is almost unit density. r It should be easy to define in a clear and unambiguous way. For example.Radiotherapy planning way of comparing different planning solutions for the same tumour in order to select the best dose distribution. In practice the ICRU reference point is often at the isocentre or at the intersection of beams. still use 1D corrections. In addition to choosing appropriate beam directions. The OAR dose often limits the highest dose that can be delivered to the tumour. the reference point is selected according to the following criteria: r The point should be clinically relevant and representative of the dose throughout the PTV. or craniospinal axis treatments. Many systems are using a 2D correction which does reduce some of this uncertainty. Artificial hips are very high density and they can create particular problems during planning. in which case it should be selected to be in a place where dose specification is considered to be meaningful. but routine clinical use of a full 3D correction as given by Monte Carlo simulation is still some way off because of the very large computing power required. and skilful use of the beam edge and beam direction can often ensure that the beam passes through the PTV but avoids an OAR. because changes from scatter contributions will be significant. In contrast. it is best to try to avoid letting any part of the beam pass through an OAR. it is also possible to use wedges to reduce the doses in sensitive structures. and they would benefit from receiving a smaller dose. in combined breast and supraclavicular fossa radiotherapy. there will be an overlap and a potential gap. dense bone tends to lie in thin layers. and this can be used to spare a structure such as the contralateral eye or brain. the dose in a central organ. Inhomogeneities of density When organs are of different densities. patients can move between treatments even when they are in an immobilisation shell. However. However. Data from the planning process can be used to create a ‘beam’s eye view’ (BEV) image of the treatment fields. which can be superimposed on the digital reconstructed radiograph and used for verification purposes. there will be changes in the penetration of the beam. 45 . For example. r It should be selected where the dose can be accurately determined. but some of the so-called 3D planning systems Matching adjacent beams in complex treatments Sometimes beams need to be adjacent to each other. It is best to avoid passing beams through metal objects if possible. For example.

as beam conformity increases. WA = wedge angle. but obviously at a cost of dose to the deeper structures posterior to the target volume. If possible. Hornick et al. Although portal imaging helps to verify the setup in relation to bony landmarks. but the possibility of patient movement during treatment must still be considered. For example. using identical simulation and treatment couches. Validity of the setup Indexed positioning. Ultrasound-based guidance can also been used for prostate treatments.3. It will become increasingly important to identify the position of recurrence within the PTV in conformal radiotherapy. One solution has been to increase the number of photon fields with the intention of trying to avoid using the electron fields. Drawing dose away from sensitive structures using wedges. A steep wedge angle on the lateral beam allows extra dose to be given from the anterior beam and can be used to spare a structure such as the contralateral eye or brain. Other ways of minimising dose variations at junctions include half-beam blocking or couch rotations.g. 2000). it is best to minimise these effects by careful 46 . in the prostate) which can be monitored on the portal image. thermoluminescence dosimetry (TLD) is the only practical method of in vivo dosimetry but it is usually only used at the skin surface.. Implanted markers. such as gold grains (e. Techniques such as adjusting the setup at the time of the radiotherapy fraction using information from a CT scanner attached to the gantry will become increasingly important in overcoming these problems. Verification of treatment dose If the CTV does not receive the intended dose then the risk of recurrence is likely to increase. A prostate gland can move more than 5 mm in 7 minutes. there may be a greater risk of a recurrence at the edge of the treatment field. Isodose (%) 95 70 50 40 30 Figure 4. which can pose even more difficulties because the shapes of the isodoses are completely different. Correction of setup errors by moving treatment fields should only be carried out for systematic errors. At present. it is sometimes necessary to have an electron field next to a photon field. This type of treatment is a prime case for complex IMRT. movements may require online megavoltage portal imaging or use of a kilovoltage CT imager attached to the linac gantry. the position of the prostate can be significantly different from the time of scan to the time of treatment (Padhani et al. and/or during treatment by using portal imaging. (1998) have modified treatment couches to automatically compensate for the variations found in rotational and tilt orientations.Andrew Tyler and Louise Hanna Anterior beam WA = 15° Lateral beam WA = 60° immobilisation and the use of accurate and stable marks for beam setup. and internal organs may move during the course of treatment. have been shown to improve the accuracy of treatment setup (Shimizu et al. the junction can be moved twice during treatment to spread the effect of any uncertainty in positioning. By taking serial images it is possible to judge whether a setup error is systematic or random. 2001). both of which can compensate for the effect of beam divergence. the organ of interest may not show up on the image. therefore. Work is Verification Verification of the planned treatment The treatment that is actually delivered will differ from that which has been planned because of setup errors and internal movements during and between fractions. has been developed to aid in quick and repeatable setups. A useful review can be found in the literature (Langen and Jones. When treating neck nodes.. 1999). The treatment position can be verified before treatment by using a simulator or CT simulator. which could include the brain stem. Otherwise. For craniospinal treatments.

Guerrero Urbano and Nutting. IMRT does tend to cause an increase in the low-dose bath of radiation in adjacent tissues. The new management standard (ISO 9000: 2000) ensures that secure working processes are in place and that these are continually reviewed and improvements are incorporated. Useful reviews have been written (McNair et al. the radiation oncologist specifies the required dose distribution in the PVT and OAR. The beam intensity is modulated by adding smaller beams with different weightings to a conformal therapy beam. gynaecological. or ‘dynamic’ mode. A computer algorithm is then used to calculate the beam modulations that will best fulfil the requirement. 2004b). 2004a. Inverse planning For inverse planning. increasing the complexity of manual treatments does lead to more errors. which can be achieved with standard treatment planning software. All staff involved in the process of giving radiotherapy to patients must be fully involved in the QA and management system. IMRT delivers the variations of intensity across the beam either in ‘step-and-shoot’ mode. in which segments are delivered before moving the MLCs to the next position. but IMRT is also used in other sites... Nutting et al. in which the MLC shape is continuously adjusted during the beam portal delivery.. 2005). Quality assurance of treatment planning systems Reducing treatment errors To treat patients safely. 2006). 2000a). technical quality is essential but not enough: QA of the clinical process is also required. This sequence is then re-planned to show the treated dose. 2005a.’ In serial tomotherapy. each with a small number of monitor units. the intensity-modulated beam is delivered as a series of discrete fields. 1999). Because it achieves greater dose homogeneity and conformality. breast and gastrointestinal treatments (Coles et al. Serial tomotherapy involves a linear accelerator beam that moves around a patient and delivers the treatment in narrow ‘slices. depending on whether the treatment will be delivered in step-and-shoot or dynamic mode or by tomotherapy.. In IMRT there is an extra step in the treatment planning process. including CNS. IMRT has the potential to improve tumour control and reduce normal tissue toxicity. the review by Nutting et al. Typically multiple treatment fields are used. 47 . IMRT lends itself to the treatment of irregularly shaped PTVs. Modern treatments have been recognized as being more sensitive to geometric uncertainties than conventional treatments because of the high dose gradients involved (Purdy. such as concave volumes. Forward planning Forward planning is a simpler form of IMRT planning. whereas in helical tomotherapy the couch moves continuously during treatment and the volume is treated in a single spiral (Beavis. The idealised plan is passed to a sequencer which sets up practical MLC positions. Intensity-modulated radiation therapy (IMRT) IMRT involves modulating the intensity profile of the radiation beams to improve spatial agreement (conformity) between the resulting dose distribution and the tumour target volume (Bortfeld. 2005b. 2004. Hong et al. the couch is moved in between treatment slices. In step-and-shoot mode. the idea of beam modulation is not new: tissue compensators and wedges have been used for many years. automated data transfer and automated setups have considerably decreased errors without increasing treatment times. 2002). In fact. includes clinical results. 2003. due to the long beam-on times. The dynamic or slidingwindow method also achieves intensity modulation with MLCs. Most experience to date has been with head and neck and prostate radiotherapy. Systems such as ‘Quality Assurance in Radiotherapy’ (QART) should be put in place to ensure overall control of the activities in the radiotherapy department (Kehoe and Rugg. Increasing the complexity of treatments does not necessarily result in more treatment errors. However. but here the leaves move across the field continuously during the irradiation.Radiotherapy planning currently under way to evaluate electronic portal imaging devices to measure exit doses. and treatment can be performed on any machine without the need for IMRT sequencing software. which may provide some verification of the dose received from each portal. lung. and this should be considered in the treatment decision.

Organs at risk. which will limit the overall delivery of dose to the PTV. Figure 4. Other tools such as the DVH used in 3D planning can cause errors if they are not used correctly. including testing the reliability of the transfer of body outlines. because of the oblique skin surface. provide the basics for this analysis: 48 .Andrew Tyler and Louise Hanna Modern 3D treatment planning systems provide a wealth of information which was previously not available. the 95% isodose lies considerably outside the lateral margins of the PTV. Other normal tissues. of the values for tissue density and of the projection of reconstructions. In plan (a). ‘TG53’ is an important document produced by a topic group of the AAPM Radiation Therapy Committee which describes all the tests to be considered when commissioning a treatment planning system (Fraass et al. The plan should be designed to avoid overtreating any dose-limiting organ. The plan also illustrates the importance of using the correct beam direction. shown in example (b). the quality of a treatment plan can be reviewed and decisions can be made about whether to improve the distribution or whether a compromise must be made between delivery of the dose to the PTV and that to the adjacent OARs. The solution to this problem is to make the anterior beam narrower. two lateral beams are used and the PTV dose homogeneity is not ideal. Care needs to be taken when introducing what seem to be simpler treatments such as replacing physical wedges with dynamic ones. However. Homogenous dose to the PTV. and a full compensator may be needed. as shown in plan (b). is to use fairly small-angled wedged beams (∼15◦ ) to act as tissue compensators. For example. then the jaws will provide extra shielding and reduce the peripheral dose at these points.4 shows two different plans for a radical radiotherapy treatment for carcinoma of the larynx. In example (a). but if the backup collimators can be used. However.g. 1998). Using lateral beams rather than posterior obliques allows for greater sparing of the rectum.5 shows two different plans for radical radiotherapy to the prostate. the doses in normal tissue are not usually a problem. It is expected that the dose under MLCs will increase because of leaf penetration and interleaf leakage. but there will always be higher doses at the entry points of the beams. hollow structures like the rectum) simply by delineating their outer surface. further automation of data transfer is essential. because the beam edge is used to try to provide the maximum dose gradient between the prostate and the rectum. Verification systems have helped the treatment process but. 1998). When the dose is prescribed at the ICRU reference point (typically the crosspoint of the beams). Note that the DVHs for the PTV and rectum in both plans are identical. Examples of plans that require improvement Figure 4. to reduce the risk of errors. When the correct choice of energy and beams is used. backed up by ICRU 50. care must be taken here because any changes in the patient’s contour in the S–I direction will not be accounted for.. but using this information also needs careful QA to ensure that the data are valid (Purdy and Harms. or the computer may be unable to evaluate the dose within a volume where there is a high dose gradient. such as the density of internal body organs and DVHs. but sometimes a compromise may have to be made. it may not be appropriate to define some organs (e. the highest dose in the PTV is usually considerably less than the +7% maximum described in ICRU 50. New tests and phantoms have been designed to investigate the non-dosimetric functions of these systems. it may be more difficult to ensure that the PTV is covered by the 95% isodose while not irradiating too much normal tissue. The solution. Three principles. Critical analysis of treatment plans Using the basic principles discussed earlier. Implementation into the planning systems is complicated and entails significant QA before use in a clinical setting. The high dose volume in the PTV should fall off rapidly in normal tissue. This three-field plan requires 60◦ wedges on the lateral beams to compensate for the dose gradient caused by the anterior beam. the number of dose points used to form the DVH may be low and this will give a false volume. Also. Commissioning 3D treatment planning systems involves a large number of quality checks. The exit sites of all beams should be checked carefully to ensure that the dose to sensitive tissues is at a safe level. This shows that it is important to look at both the plan and the DVHs.

Two different plans for a radical radiotherapy treatment for carcinoma of the larynx (see text for further explanation). Dose (%) (c) Figure 4. the anterior beam avoids the lung.6 shows two plans for radical radiotherapy for carcinoma of the bronchus. The DVHs for both plans are shown in (c) (see text for further explanation). In example (a).4.5. Electrons Electron beam isodoses have a shape that is significantly different from that of photon beams. Figure 4. Part (c) shows the DVH for spinal cord is significantly better for plan (b). by exiting through the cord. The reporting of electron doses should follow ICRU 71 (ICRU. all three beams pass through the spinal cord. 49 . Two different plans for radical radiotherapy to the prostate. As a result only the exit beam from the anterior direction passes through the spinal cord and the dose is now within tolerance. A solution is shown in example (b) where the posterior beam has been moved to a posterior oblique direction and the anterior oblique beam has been moved more laterally. 90 80 70 Wide anterior beam Conformal plan PTV Volume (%) 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100 110 Rectum Figure 4. and cord tolerance is exceeded. 2004) in a way similar to that for photons.Radiotherapy planning % 110 105 102 100 95 70 20 Isodose % 95 70 60 50 35 (a) (a) Isodose % 95 70 60 50 35 % 102 100 95 70 20 (b) 100 (b) Figure 4. and although the lung DVH is worse for plan (b) it is still well within acceptable limits (V20 of less than 35%). Note also that.7 shows an electron isodose for an 18-MeV electron beam. The fairly homogenous dose distribution close to the surface and rapid fall- off at depth make electron doses ideal for treating superficial tumours.

r Inhomogeneities within the patient such as air gaps or high-density structures may cause unexpected hot and cold spots as demonstrated by Figure 4. r High-dose treatment isodoses tend to pinch in. 8 Off-axis distance (crossline) (cm) 6 4 2 0 -2 Al -4 -6 -8 0 5 Depth (cm) 10 Water Isodose (%) 10 20 50 70 80 90 95 100 102 (b) 100 90 80 70 Original plan Cord-sparing plan PTV Volume (%) 60 50 40 30 20 10 0 0 10 20 30 Cord PRV Figure 4. Note the hot spot. skin surface.6. . where an aluminium sheet has been placed in the phantom. esper Low-dose isodoses tend to spread out and can impinge on local OARs. The isodoses have been normalised to the dmax of a beam with no insert.7. Combined Lungs r Bolus is often required to increase the dose at the 80 90 100 110 120 40 50 60 70 Dose (%) (c) Figure 4.8. Two plans for radical radiotherapy for carcinoma of the bronchus. PTV = planning target volume (see text for further explanation).8. The uncertainties are much greater than in photon beam treatments: r Skin apposition rather than point placement of beams is used for setup (cf.Andrew Tyler and Louise Hanna Off-axis distance (inline) (cm) Isodose % 95 80 60 30 20 6 4 2 0 -2 -4 -6 Isodose (%) 10 20 50 70 80 90 95 100 (a) Isodose % 95 80 60 30 20 0 5 Depth (cm) 10 15 Figure 4. 50 Care is required when using electrons: cially with small fields. PRV = planning organ at risk volume. r Oblique surfaces will significantly alter the isodoses. isocentric setups for photons). r Non-standard air gaps between the applicator and patient will affect the scattered dose. Isodoses for a 6 × 6cm 18-MeV electron beam. The thickness of the bolus will need to be taken into account when deciding on the electron energy. while the lower isodoses become wider. Isodoses for a 10 × 10cm 12-MeV electron beam with an aluminium (Al) insert. Note how the higher isodoses become narrower with depth.

medium and high.1. but there is an increasing trend towards the use of remote afterloading. Traditional methods of dosimetry use rules to govern the configuration of radiation sources and methods to calculate the dose received by the tissues. For more detail. Properties of some radionuclides used in brachytherapy. 1978). non-radioactive guides are placed first whenever possible. r Intraluminal – into a lumen.9 days 374 days Disintegration β. 1985) classified radiation dose rates at the point or surface where the dose is prescribed into low. For radiation protection purposes. plaque. CX = characteristic X-ray photon. Brachytherapy treatments are described according to the position of the radioactive source in relation to the body: r Interstitial – in the target (may be permanent.3 years 60 days 16. It is also often accelerated radiotherapy. brachytherapy is a highly conformal type of therapy.. and their centres must be in the same plane 51 Isotopes The properties of some radionuclides used in brachytherapy are shown in Table 4. it is necessary to reduce the total dose and fractionate the radiotherapy to avoid unacceptable late tissue damage. Thus. afterloading Needles. r The radioactive sources must be linear. . as in iridium wires).4 to 2 Gy/h. γ CE. N. r Intravascular – into an artery. B. r High dose rate (HDR): > 12 Gy/h. CX. of moderate size.2 years 74 days 5. and able to be delineated. followed by ‘afterloading’ of the radioactive source. pins. which have important differences in their radiobiological effects: r Low dose rate (LDR): 0. tubes. r Surface applications. or temporary.1. Paris system The Paris system was based on experience in the use of iridium-192 wires (Pierquin et al. tubes. Dose rates The International Commission on Radiation Units and Measurements report 38 (ICRU. as in iodine seeds. and the inversesquare law (dose is proportional to the inverse square of the distance from the source). therefore. afterloading Wires.Radiotherapy planning Table 4. the reader is referred to more comprehensive texts (see Further Reading). Rules for the Paris system are as follow: r The linear activity of the sources must be uniform along the full length of the source and identical for each source. Brachytherapy Introduction Brachytherapy is radiation treatment given by placing a radioactive source near or in the target (usually a malignant tumour). γ β. dose may need to be modified if it exceeds 1 Gy/h. This section will describe the broad principles of brachytherapy. Afterloading may be either manual or remote. γ CX β Average photon energy (keV) 662 380 1250 28 20–23 (mean 21) 3541 Types of source Needles. γ β. afterloading Seeds Seeds Plaque CE = conversion electron. parallel and straight. It is possible to deliver a high dose of radiation to the target while sparing adjacent normal tissues. Approximate Isotope Caesium-137 (137 Cs) Iridium-192 (192 Ir) Cobalt-60 (60 Co) Iodine-125 (125 I) Palladium-103 (103 Pd) Ruthenium-106 (106 Ru) half-life 30. Dosimetry The radiation dose delivered to the tissues is primarily determined by the activity of the source. Tumours suitable for brachytherapy must be accessible. the length of time that the sources are in place. as the dose rate increases. r Intracavitary – in a body cavity. r Medium dose rate (MDR): 2 to 12 Gy/h.

CT. The separation between the sources may vary between implants. prostate). the most widespread use of brachytherapy is for gynaecological cancer. second primary tumours or locally recurrent tumours). including the bladder and rectal points. 403). 1985). paracervical tissues. In an implant all the sources must be separated equally from each other. the radiation sources must be longer than the PTV. new imaging techniques. oesophagus). description of sources Manchester interstitial system The Manchester interstitial system was based on experience in the use of radium sources and is also known as the Paterson–Parker system (Meredith. r Palliative treatments (e. r Benign conditions (e.8 to 2 cm. but the separation should be 0. volume implants and moulds. the radiation sources are removed. The reference dose rate is calculated as 85% of the basal dose rate and is the dose rate used for the tumour prescription and calculation time of the implant. the guides for the active sources are placed first. r Once the implant is complete. Remote afterloading is typically used. and. ICRU 58 recommends that the brachytherapy treatment be reported in a standardised way. ultrasound).g. Point B is 5 cm lateral 52 . the technique and the number/size of radiation sources required.g. However. In the UK. keloids). r Re-irradiation within a previously treated volume (e. Other reference points are specified by ICRU 38 (ICRU. prostate cancer. r The active sources are then placed. For this to occur.. Isotopes used include caesium-137 o for low. using orthogonal X-rays.g. r The treatment time is calculated. small tumours of tongue. Its rules governed the distribution of radiation sources for different configurations including planar implants. 1967). r The position of the sources is verified (e. including information on target volume. but the source separation may vary from one implant to another. increasingly. The radiation dose is specified at point A. MRI. bronchus. which is called the central plane. which can be defined on the anterior and lateral radiographs. head and neck cancer. with packing or a rectal retractor to push the rectum away from the high-dose volume.g. Reporting doses to these points allows comparison between centres. per uncrossed end. The volume enclosed in the reference isodose is known as the treated volume. followed by the applicators. Brachytherapy procedures Indications for brachytherapy include the following: r Primary radical treatment (e. Manchester system for gynaecological brachytherapy The applicators are configured as a central uterine tube and two vaginal ovoids. r Treatment in combination with surgery (e. unless the ends are crossed as in iridium hairpins. which is 2 cm lateral to the centre of the uterine canal and 2 cm above the lateral vaginal fornices (this latter level is sometimes described as the level of the cervical os or the level of the flange at the base of the intrauterine tube). 2006). to the midline at the level of point A. and upper vagina. r Treatment in combination with external beam radiotherapy.g. r With afterloading techniques. cervix.Andrew Tyler and Louise Hanna r r r r r r perpendicular to the direction of each source.and medium-dose-rate treatments or cobalt-60 and iridium-192 for high-dose-rate treatments. computerised treatment planning and optimisation open new prospects for individualised treatment and reporting dose to OARs (P¨ tter et al. r The implant procedure itself takes place under general or local anaesthetic. The average length of the source should be 25 to 30% longer than the target volume. For volume implants the array of the source intersections in the central plane should be either in an equilateral triangle or a square configuration.g. the target volume. The dose is calculated as follows: The basal dose rate is calculated in the central plane of the implant and is the arithmetic mean of the local minimum dose rates. Brachytherapy for ocular melanoma is discussed in Chapter 34 (see p. Practical steps in brachytherapy: r The treatment is preplanned to determine the tumour volume. The resulting isodose distribution is a pear-shaped volume encompassing the uterus. For further details see Meredith (1967). An appropriate radionuclide is selected. The isodose surface that corresponds to the reference dose rate is called the reference isodose and this must encompass the PTV. floor of mouth. intraoperative sarcoma treatments).

(1976). Doppke. M. For computerised systems. Reduction of radiotherapy-induced late complications in early breast cancer: the role of intensity-modulated radiation therapy and partial breast irradiation. Determination of Absorbed Dose in a Patient Irradiated by Beams of X or Gamma rays in Radiotherapy Procedures. Any equipment used to measure the actual delivered dose needs to be calibrated against a recognised standard.). G. 267–72. Coche. checks should be carried out that include independent measurements of the activity of the source and tests for leakage. (2004a). London: British Institute of Radiology. (1985). (2005). (R. Pulsed dose rate Pulsed dose rate (PDR) is a hyperfractionated form of brachytherapy that is under evaluation. Tome. C.. 77. A. S.. Med.. ICRU. Litzenberg. Radiat. 17. Areas of current interest in brachytherapy Optimisation Use of cross-sectional imaging. Geometric Uncertainties in Radiotherapy. 17. Selection and delineation of lymph node target volumes in head and neck conformal radiotherapy. Dose and Volume Specification for Reporting Intracavitary Therapy in Gynaecology. Prepared by a Working Party of the British Institute of Radiology. enabling the dose to conform more accurately to the target volume. Clinical use of intensity-modulated radiotherapy: part II.and low-dose volumes (ICRU.. 92. Med.. BIR. But it is still important to take great care in positioning the brachytherapy sources. T. P Khoo. M.. C. Lancet. D. T.. 77. because optimisation cannot make a bad insertion good. Beavis.. Biol. 25. Phys. These rules include source storage and preparation. Radiol. M. R363–79. The dwell times for the source can be varied at different positions. B. 285–95. et al. (2006). Wilson. (1999). C. (2005a). 77. Is tomotherapy the future of IMRT? Br. Clin... written systems of work for staff entering controlled radiation areas and contingency plans for emergency situations such as source sticking. Phys. et al. Coles. A. W. J. Int. Oncol. 353. E. Maryland: International Commission on Radiation Units and Measurements. et al. 88–96. S. W. (1998). et al. 56. (2006).Radiotherapy planning and technique. Clin. Comparison of radiation side-effects of conformal and conventional radiotherapy in prostate cancer: a randomised trial.. Br. 1819–24.. Fraass. Cosnard. Proposal for standardizing terminology and procedure based on the surgical experience.. (2004b).. although differences in relative biological effectiveness must be taken into account. (2004). Defining the Planning Target Volume. Bethesda. S. 98–110. K. Radiol. 678–83. M. M.).. J. 53 . Oncol. et al. REFERENCES Apisarnthanarax. Hong. Coll.. 1739–40. 25. (2003). Reduction of radiotherapy-induced late complications in early breast cancer: the role of intensity-modulated radiation therapy and partial breast irradiation. V. 16–24. It combines the advantages of afterloading and the ability to optimise the treatment in a multiple-fraction highdose-rate treatment and it may provide an alternative to low-dose-rate brachytherapy. Moody. 1773–829. 177–82. C. A. T. K. M. Determining optimal clinical target volume margins in head-and-neck cancer based on microscopic extracapsular extension of metastatic neck nodes.. Coll. computer planning systems. J.. L. Part II – radiotherapy strategies to reduce radiation-induced late effects. El-Nagger. Maryland: International Commission on Radiation Units and Measurements. Part I – normal tissue complications. D. ICRU Report 38. Wilson. 51. Bethesda. D.. W. B. M. Radiother. V. A. Phys.. A. this is especially important so that the transfer of patient image data and the validity of the data and calculations are verified. (1998). Hornick. Ritter. R. M.. Clinical use of intensity-modulated radiotherapy: part I. 64. ICRU Report 24. E. . D. Oncol. and Nutting. Lam. Quality assurance and radiation protection in brachytherapy Local rules and systems of work must be drawn up that enable the centre to comply with current legislation. and Nutting. Radiol. Radiol. Norman. (R. Oncol. E. Med. J. Cancer. Phys. (2005b). American Association of Physicists in Medicine Radiation Therapy Committee Task Group 53: quality assurance for clinical radiotherapy treatment planning.. The quality of calculation methods needs to be properly ensured and a regular quality control programme should be in place. J. Coles. Gregoire. Br. On receipt of a radiation source. dose prescribed and a description of the high. C. Intensitymodulated radiation therapy. C. Guerrero Urbano. et al. K. C. ICRU. et al. et al. Guerrero Urbano. Bortfeld. A. Hunt. Moody. IMRT: a review and preview. T. Br. A. 1997). Radiol. it uses remote afterloading and a single high-activity source that is stepped through dwell positions within the catheters. and treatment machines with moveable stepping sources allows the radiation oncologist to define the target volume. Biol.. emerging cancer treatment technology. Elliott. B. obtain dose volume histograms and achieve the best dose distribution. D.. 135–50. (2000). Dearnaley. A tilt and roll device for automated correction of rotational setup errors.

Journal of the ICRU. Oxford: Oxford University Press.. C. by Ralston Paterson (and others). et al.. J. 33–48. Biol. H. C. ICRU Report 50. and Rugg. (1999). Cowan. Biol. (1967). FURTHER READING Bomford. 1). Senan. 1–9. P. Oncol.. Radiother. (2002). and Reporting Electron Beam Therapy. C. and Kunkler. and Thwaites. H. V. Int. 56.. 2–7.. Radiation Physics.. J. Nutting. A. L. Suckling. Phys. ICRU Report 71. R. (2003). C. Weiss. L. Radiat.. et al. A. W.. ICRU. Edinburgh: Livingstone. E. Biol. Paine. J. S. (1998). Edinburgh: Churchill Livingstone. Practical Radiotherapy Planning. H. Future directions in 3-D treatment planning and delivery: a physicist’s perspective. Radiotherapy Physics in Practice. A. S. H. Dose and Volume Specification in Interstitial Brachytherapy.. Williams. Intensity modulated radiation therapy: a clinical review. Oncol. Implementation of IMRT in the radiotherapy department. Oncol. 4 (No. M. et al. J. (2005). F. Barrett. (1999). L. J. Lagerwaard. Sci. Clinical implementation of a commercial multileaf collimator: dosimetry. Therapy and Oncology. (1998).. Oncol. S. A.. A. J.-J. networking. Radiat.. Williams and Wilkins. Maryland: International Commission on Radiation Units and Measurements. Radiother. Harms. E. (2004).. (2006). Oncol. Radiat. Precise and real-time measurement of 3D tumor motion in lung due to breathing and heartbeat. J. Oxford: Oxford University Press. (2000).. F. I. B. D. C. A. P and Coyle. Shirato. C. H. E. (1999). Shirato. V. Radium dosage: the Manchester system/compiled from articles. 44. Langen. Jansen. M.. 190–5. Oncol. S. Madison. Lees. I. Target . Oncol.. Acta Radiol. P Sharrock. B. ICRU. Flynn. J. Dutreix. Dearnaley. Joslin. (1978). M. Philadelphia: Lippincott. Radiother. Int. Br. 179. J. Maryland: International Commission on Radiation Units and Measurements. E. 308–14. et al. Klein. 3rd edn. Dose-volume specification: new challenges with intensity-modulated radiation therapy.. Haie-Meder. B. Int. Strahlenther. (2003). F. Effect of intravenous contrast on treatment planning system dose calculations in the lung. 28. J. Y. Seppenwoolde. 281–90. ICRU. and Hess. F. Oxford: Oxford University Press. Khoo. M. (1993).. 6th edn. A. Logue. O. Kitamura.. and Hall.. C. J. T. Radiat. K. (2000a). Hoskin.. et al. 41. . Biol. R.. S. Padhani. (1995). Phys. Radiat.. Adams. Bethesda. The impact of gross tumor volume (GTV) and clinical target volume (CTV) definition on the total accuracy in radiotherapy theoretical aspects and practical experiences. o Recommendations from gynaecological (GYN) GEC ESTRO working group (II): concepts and terms in 3D image-based treatment planning in cervix cancer brachytherapy-3D dose volume parameters and aspects of 3D image-based anatomy. 850–6. 265–78. From technical quality assurance of radiotherapy to a comprehensive quality of service management system. Phys. (2001). 46. Hoskin. Chapet. W.. 199–209. 51. and Reporting Photon Beam Therapy. radiobiology. Evaluating the effect of rectal distension and rectal movement on prostate gland position using cine MRI. Prescribing. J. E. Quality assurance for 3D conformal radiation therapy. Phys. Wisconsin: Medical Physics Publishing. J. et al. J. Khan. C. K. External Beam Therapy. measured during radiotherapy. Dobbs. 76. Biol.. Meredith. J. Oncol. Phys. Purdy. van Herk.. Organ motion and its management. Radiother. 48. Holloway. New York: Arnold.. Semin. (2002).. volumes in radiotherapy for high-grade malignant glioma of the brain. 50.. simulation and quality assurance.Andrew Tyler and Louise Hanna ICRU. A. (2005). Kehoe. variability of target volume description in conformal radiotherapy planning. 459–69. (Suppl. Brussels: European Society for Therapeutic Radiology and Oncology. R.. Semin. D. (2002). 54.. Int. A. Phys. A. 12. ICRU Report 62. A. Phys.. 822–34. A. ICRU Report 58. J. 14.. 17. Biol. (2000). Clinical . (2001). 3rd edn. Radiat. J. V. et al. R. London: Arnold. Radiotherapy in Practice. et al. Khoo. R. 21–30. L. H. Biol. (2003). Low. et al. Radiol. P¨ tter. Int. 929–31. J. et al.. Radiat. J. K. and Reporting Photon Beam Therapy (Supplement to ICRU Report 50).. E. Br. P¨ tter. (2003). A randomized study of the use of a customized immobilization system in the treatment of prostate cancer with conformal radiotherapy. A. et al. Kitamura. 3–6. and Ash. et al. The GEC o ESTRO Handbook of Brachytherapy. Van Limbergen.. T. P M. Onkol. Nutting. Oncol. Shimizu. Dewit. Oncol. Purdy. 174. (2001). The Paris system in interstitial radiation therapy. Shiu. and Mellenberg.. Radiat.. P. E. 33. Walter and Miller’s Textbook of Radiotherapy. Gerbaulet. D. (2000). Radiol. Maryland: International Commission on Radiation Units and Measurements. 1195–208. Oncol. (2006). 1591–7. Biol. Eng. D. and Jones. Prescribing. General practice of radiation oncology physics in the 21st century: AAPM Monograph no.. W. Med.. J. 151–6. et al.. Onkol. M. Radiat.. J. J.. 26.. H. 2). Walker. (2000). Principles and Practice of Brachytherapy: Using Afterloading Systems. J.. Oncol. Recording. Fuller. D. E. S. (2004). Phys.. K. Strahlenther. Prescribing. L. Australas. radiation physics. Purdy. Radiotherapy in Practice. 67–77. (1999). Clark. Pierquin. Brachytherapy. Oncol. Bethesda. Recording. C. G. Defining target volumes for non-small cell lung carcinoma. (2000b). and Harms. Oncol. C.. 54 . and Webb. Bethesda. 78. Use of an implanted marker and real-time tracking of the marker for the positioning of prostate and bladder cancers. D. Mazeron. 525–33. Int. Recording. The Physics of Radiation Therapy. Oxford: Oxford University Press. (1997). Radiat. Phys. 53. Int. 73. McNair.

1996) has already been devoted to the theory of the RCT (and much that is less excellent). 2002. it determines that the treatment is safe 55 . it becomes easier to detect smaller treatment effects and to conclude that any differences are not the result of chance. any differences between patients receiving one treatment and those receiving another are purely down to chance. Much excellent literature (e. then it must be due to the only factor that is systematically different between the two groups. But. Thus. In particular. although this avenue may identify promising treatments. unless the action of a particular treatment is both immediate and breathtaking (such as insulin for diabetic coma). it will look at the factors that constitute a successful clinical trial. is effectively a formalisation of common sense.g. Altman. a new treatment (typically a drug) will go through four stages of development. 2000. First. Increasingly. This chapter will concentrate on the underlying principles. whether promising laboratory results translate into activity in humans. This chapter will concentrate on obtaining reliable evidence on the efficacy (whether the treatment works under ideal conditions. a compound may well be developed based on an association seen in a laboratory and backed up by early research involving animals. namely the treatment. Pocock. whether the treatment gives better results when used in clinical practice compared to current best care. can prove misleading. This chapter is intended to provide useful information for researchers planning to conduct an RCT. medical statistics has been well covered in a number of books and articles (Altman et al. and at each stage a clinical trial will be run to ascertain that the treatment still remains sufficiently promising to proceed: Phase I represents the first trials performed in humans. most important. Second. Swinscow and Campbell. At its heart are two principles. if a sufficiently large difference is detected. too.. 2002). health care providers are interested in determining whether or not new treatments represent sufficiently good value for money. This. How does one know precisely what the best care is for a particular patient? In particular. Historical comparisons. but larger numbers give stronger evidence of an unfair coin. through randomisation. there is such a tool: the randomised controlled trial (RCT). If one tosses a coin 10 times and gets 6 heads and 4 tails. The proportion of heads is the same. then one would be concerned that the coin may be biased. and. Clinical trials Introduction The path taken by a new treatment before it comes into common use is typically a long one. this simple statement masks a much more complex issue. convincing evidence that can be used to inform future practice. Duley and Farrell. but can one really be sure that it represents a real improvement on current practice? Generally speaking. Fortunately. it does not tell us the proper dose to give for treatment (or even if it is safe at all to use in humans). but if one saw 6000 heads and 4000 tails from 10 000 tosses. or other database-dependent methods. we cannot be absolutely certain which treatment is best for which people. therefore. how the ideas can be extended to look at the weight of evidence provided by a number of clinical trials (meta-analysis) and how additional laboratory studies can help assess more modern targeted therapies. and also provides useful pointers for clinicians wanting to critically appraise a trial for reliability. it is not out of the ordinary. However. with particular emphasis on their applications in cancer. usually in a highly selected population) and effectiveness (whether a treatment will be beneficial in a real-life setting) of treatments for cancer. Likewise. how does one balance the likely benefits and risks for a particular course of treatment? A new drug may appear promising.5 RESEARCH IN CANCER Robert Hills Introduction It is the responsibility of clinicians to provide the best possible care for their patients. What is required is a method that will provide reliable. the statistical aspect of RCTs. Initially. with large numbers of patients. 1991.

Storer and DeMets. and to look at the long-term balance of risks and benefits. for many non-pharmaceutical 56 treatments. it is studied further in phase IV trials. The actual likelihood of a new intervention having a big treatment effect. in particular. 1984). a randomised design has increasingly been used. Students sitting A levels in 1980 received free school . After a new drug has been identified and demonstrated to provide worthwhile benefit in a phase III trial. For this reason. non-randomised studies do not typically provide robust enough evidence. defined according to the proportion of patients experiencing significant toxicity. the true scale of any ‘breakthrough’ cannot be gauged. There is little point in going to the considerable effort and expense of setting up a large RCT if there is only a small likelihood of success. This four-stage process of treatment development is not set in stone. such trials use short-term surrogate end points. indeed. is found. gives improved clinical outcomes. identifying the ideal dose for a new drug may not be a factor of toxicity alone: similar efficacy may be obtained at drug doses lower than the MTD. Such trials are still relatively small. 1989. which are correlated with more clinically relevant end points such as overall survival. such as tumour shrinkage or remission induction. 1987). Typically. in which care involving the new treatment is compared to previous outcome rates before the new care was introduced. but can form the basis of a larger phase III study if the treatment appears sufficiently promising. and without a comparator group. and that rare or long-term side effects may not have become apparent. early reports of a treatment tend to be extremely positive and raise the possibility of major clinical advances. Yet. it is important to provide reliable evidence about moderate benefits and to be able to reduce as far as possible any systematic biases that might affect the results. some of the phases may be performed simultaneously. and those that are too small to be of any real clinical interest (Yusuf et al. It is more realistic to expect a moderate difference between interventions. Phase II trials identify whether or not the treatment. the earlier phases of development may not be directly relevant. and weed out treatments that are unlikely to be of any benefit in clinical practice. demonstrates sufficient evidence of activity. Phase IV trials allow the study of the drug in standard clinical care in a large number of patients to identify any important safety issues. given either in addition to or instead of current therapy. ranging from the traditional Fibonacci design to more sophisticated techniques based on modelling the predicted toxicity levels based on all available data (Storer. A number of different designs for phase I trials have been proposed. owing to the worldwide prevalence of the conditions being treated. The entirety of the phase I to III process may only have meant that a few hundred patients have been given the drug. randomised pilot is assessed and the decision of whether to proceed to a full-scale large RCT is based on preliminary results. only a few treatments in current practice tend to work overwhelmingly well. yet. is fairly low. and many of the drugs taken for granted today (e. The treatment dose starts off quite low. Thus. tamoxifen in breast cancer) produce only moderate benefits. but still clinically meaningful. Often. Phase III trials tend to be relatively large and randomised. may be seriously misleading. But recently. can provide unreliable estimates of a treatment’s effectiveness.. or moderate effect compared with placebo. historically controlled studies.Robert Hills and identifies the target dose. and is increased until the maximally tolerated dose (MTD). For example. Historically. the history of improvements in outcomes in cancer is one of incremental progress. Similarly. Such trials can involve either healthy volunteers or patients with the condition in question. at a safe dose. and tend to use shortterm surrogate end points. The phase III RCT is dealt with in more detail in the next section. it may make sense to perform a phase I/II trial in which both activity and toxicity are considered simultaneously. Likewise. typically with a few dozen patients. a phase II/III trial could be designed in which a small. at a point where no real toxicity is expected. Case studies. with a corresponding reduction in toxicity levels. These studies can be considered as small screening studies. or being vastly superior to an existing therapy.g. phase II studies were uncontrolled studies. Therefore. Randomised controlled trials The main idea underlying the RCT is the need to distinguish between effects that are moderate. save many hundreds of thousands of lives annually. where all enrolled patients received the study drug. A potentially serious risk of selection bias exists: it is impossible to tell whether only the most promising patients are selected for the new treatment. For example. a comparison of today’s Alevel results with those of 1980 show an improved pass rate. These identify whether the treatment. Also.

2002. Simple randomisation can be achieved using a random number list. so the primary measured outcome in a trial is mortality. irrespective of previous entries into the trial.g.2 gives some thoughts on good practice in randomisation. it is important to ensure that there are no patterns that could allow the next allocation to be predicted. a computer-generated random number or even tossing a coin. and care needs to be taken to ensure that this is made as difficult as possible (Schulz. the improvement has nothing to do with the treatment. Furthermore. but is likely instead to be a result of better supportive care. 2002). the choice of a suitable end point in cancer is less tricky than for many other chronic conditions. in certain instances. the main aim is to prolong life. in permuted block randomisation with a block length of 2. 1995). DAT). can lead to chance imbalances in the numbers allocated to dif- ferent treatment groups.g. (For example. in conditions that are relatively rare and in which outcomes are already very good (e. however. Box 5. where previous allocations do not influence the next. to differences in case mix over time or to a combination of all three. Choice of end points Generally speaking. the general improvement in cancer outcomes over time means that one cannot be sure if any improvements are due to the drug. in other words. after an even number of patients the treatments will be in balance – in a single-centre trial. Some more sophisticated methods also ensure balance across important prognostic factors. The only way to reduce such selection biases as much as possible is to randomise patients: allocate treatments in a way that produces equivalent groups and precludes the chance that the next treatment allocation can be predicted. When comparing SAB with DAT treatment for acute myeloid leukaemia. the following results were obtained (Wheatley. It is possible to extend permuted block randomisation to this scenario or to use allocation algorithms such as minimisation.00007 Resistant disease 23% 24% 0. SAB = same as before (i.1 SAB in acute myeloid leukaemia. 1984). Number of Treatment DAT SAB p-value Period 1984–90 1990–98 patients 167 284 CR rate 47% 61% 0.2 DAT = daunorubicin + Ara-C + thioguanine. This method. to the generally improving prognosis. those sitting them today do not. 2005). It is a fallacy that strict random allocation is required: far more important is concealment of the treatment allocation until the patient is irreversibly committed to the trial. after every 4 or 6 patients). 1999. Likewise. Allocation by date of birth. are we to conclude that free milk makes you less intelligent? More seriously. then those who receive therapy are of worse prognosis than those who don’t. The simplest method is to use so-called simple randomisation in which the chance of receiving a given treatment is the same. may not produce equivalent groups because it is easy to predict what a person’s treatment is going to be and then make a decision on whether to enter the trial based on this knowledge. This appears promising until one realises that SAB stands for ‘same as before’.1). Clinicians can and do attempt to subvert some randomisation approaches. In many cases. for example. So. milk while at primary school. Box 5. acute promyelocytic leukaemia). which generally require the use of a computer. In acute myeloid leukaemia.006 Induction death 30% 15% 0.Research in cancer Box 5. In all except simple randomisation. the outcome measure may be different. For example. a trial assessing mortality may have to both be very large and run for a 57 . so some trials use permuted block randomisation in which the number of patients in each group is required to be in balance at various stages through the trial (e.e. However. a historically controlled study of standard DAT therapy for patients over age 60 against the SAB regimen showed a statistically significant benefit for SAB (Wheatley. A number of ways of allocating patients can produce equivalent groups (Altman and Bland. if clinicians are more likely to give a treatment to patients with more severe symptoms. patients who did not receive their preferred treatment cannot be re-randomised. A straight comparison could therefore show that the treatment appeared worse (Green and Byar. the allocation of every other patient could be predicted precisely).

it is important to choose a measure that is of clinical relevance and that can be used to help guide future practice. clinicians and to organizations such as the National Institute for Health and Clinical Excellence (NICE).bham. Standardised treatment protocols can be introduced. Another alternative is to blind the trial treatments. Similar numbers of different types of patient are ensured to be in each group. it is crucial to identify a small number of primary and secondary outcomes. this could be achieved by a randomisation list for each category.. the simplest outcomes are often the best. to define how big a difference is relevant. it is tempting instead to use a surrogate outcome. For example. one outcome may indicate that one particular treatment is better. either the patient or the clinician/assessor is unaware of the treatment allocation. At its simplest. Sometimes. one treatment may require additional followup visits. These may be clinical outcomes obtained from notes or patient-centred outcomes collected through questionnaires. In choosing a suitable outcome measure. One may consider outcomes relevant to patients. so that all patients receive equivalent supportive care and clinician contact. 1998).Robert Hills Box 5. it may be more relevant to assess quality of life – can one achieve as-good outcomes with less toxicity? This approach has been used in the NCRI AML15 trial (www. tends to minimise the effect of knowledge of treatment allocation on outcome. recurrence to predict survival benefit). however. generally. r Use third-party randomisation. However. Generally speaking. the person recording the outcome is the one who needs to be blinded. and. it is important that there is no foreknowledge of treatment allocation. However.2 Good practice in randomisation. This concept is known as allocation concealment. In these circumstances. Envelopes containing the treatment allocation may appear an attractive option but need to be policed to stop clinicians from opening several envelopes until they find the preferred treatment. but for more than a few variables this becomes cumbersome and a computerised system is preferable and allows simultaneous minimisation across many variables. Trials are typically referred to as either singleblind or double-blind. this approach can be fraught with difficulties.g. A secure computer database held at a central trials office is a good way to ensure allocation concealment. as in the SIGNIFICANT trial of prophylactic antibiotics. Because it is impractical to wait many years for an answer. such as mortality.. and for another 58 . Given enough outcomes. it is usually possible to find a significant result for one of them even in trials of the most unpromising treatments. An independent randomisation service with a degree of separation between clinician and patient provides more security.uk). Therefore. The choice of outcome measure influences both the type of statistical test that is being performed and the size a trial needs to be. In the former. by chance. 2005). outcome the reverse is the case. In double-blind long time. if possible. Blinding To create comparable groups for analysis. The recorded value of a prognostic factor may change or be influenced by treatment allocation. For example. so that supportive care or additional treatments could be provided preferentially to one group or another.g. Similarly. but this difference is not necessarily translated into a significant survival benefit (Specht et al. There are a number of ways of combating this problem. then it is worthwhile to ensure that it is validated – to make sure that the results are meaningful and reproducible and. the real outcome of interest occurs a long time in the future. even after randomisation there are occasions when knowledge of which treatment group a patient is in can influence their treatment or their outcomes. which measured the number of febrile episodes (Cullen et al. in qualityof-life trials) then knowledge of a patient’s treatment may influence the perception of health status (especially if one option is to offer no treatment). Alternatively. It is important. there may be a difference in recurrence-free survival. r Collect all important prognostic factors prior to randomisation.ac.aml15. Also. making such a study impractical. if outcomes are dependent on clinician or patient rating of health state (e. the primary aim of some trials may not be to directly improve mortality but to reduce the incidence of adverse events. which predicts this future outcome (e. which are considered the most important. not to try and collect too many outcome measures: the trial will then lack focus. the use of objective outcome measures. If a questionnaire is being used. r Balance prognostic groups by stratification or minimisation.

negative or positive. It may seem attractive to aim to detect a large treatment effect. Finally. the sample size depends on the type of outcome being measured: dichotomous outcomes (e. it is also important to determine how these patients are likely to be found.g. The number of patients required to answer a question reliably depends on a number of factors. it follows that any observed differences must be the result either of chance or of a genuine difference between the treatments. If the p-value is sufficiently small. although sometimes this will require the participant’s acceptance of an additional. Clearly. but worthwhile. treatment effect (Altman and Bland. to detect a 5% improvement would require about 4000 participants. or to set the power low. But. The use of effect sizes and confidence intervals overcomes this problem. some degree of blinding can be achieved with imagination: in trials of surgical techniques. For other trials. the larger the number of patients one needs to recruit. should not be expected to influence the outcome. it is important to identify precisely who is eligible for the trial.05) and how certain one wants to be of detecting a treatment effect if it really is there. the use of sham surgery has been advocated.g. entered remission. continuous outcomes (e. when deciding which patients 59 . alternatively. pharmacists can make up drug solutions that are indistinguishable irrespective of the treatments actually contained. one can determine an estimate of the size of the treatment effect. survival time. This way. which tells you how often a trial of an ineffective treatment would be expected to produce this type of result. A corollary of this is that a non-significant result does not automatically mean that there is no treatment difference: the trial could simply have been too small to detect a moderate. However. thus making the trial inconclusive. therefore. there is no deception involved (Miller and Kaptchuk.Research in cancer trials. or the higher the power. together with a measure of a likely range. To have a 90% chance of detecting whether a treatment increases the proportion of patients entering remission from 50 to 75% requires only about 150 participants (Machin et al. quality-of-life scores. clinically unnecessary procedure. a study needs about 1000 patients. However. but this means that one runs the risk of the estimate of treatment effect to be sufficiently imprecise as to include both no treatment effect and a clinically relevant effect. the larger the number of patients required. Extending a trial over a number of centres will increase the relevant pool of potential participants. 1995). experienced grade 3 or 4 toxicity). The 95% confidence interval contains the true treatment effect 95% of the time. to detect moderate differences.g. etc. and how many In a randomised trial. which would still be worthwhile. When considering the required number of patients. to detect an improvement from 50 to 60%. The probability that observed differences are the result only of chance differences between groups is given by the pvalue. It has been argued that sham procedures are methodologically necessary to produce valid results. from 50 to 60% entering remission. it is important to specify the level at which statistical testing will be performed (typically set at a significance level of p ≤ 0. both the patient and the clinician/assessor are unaware of which treatment was given. 1997). then one can conclude that the data are inconsistent with the treatment being ineffective. only if this does not include a clinically relevant treatment effect is it fair to conclude that a worthwhile difference is unlikely. First. The smaller the effect one wishes to detect and the smaller the significance level. the number of patients required in a study to detect moderate treatment effects may perhaps be more than it is feasible to recruit. any beliefs about the intervention. Alternatively. However. because the different treatment groups are equivalent. one must discriminate between the signal and the ambient noise: although one cannot reduce differences between people. This latter concept is known as the power and is typically set at 80 or 90%.). based on the natural variability between patients. as long as the participant is informed that they will receive either a real or a sham intervention and that the sham procedure will be indistinguishable from the real treatment under investigation. a difference of 10 points on a scale with a standard deviation of 25. rather in the same way that a jury starts with the presumption of innocence and asks whether there is enough evidence to overturn it beyond reasonable doubt. Second.g. The use of a matching placebo can ensure blinding. size of a laboratory measure) and time-to-event outcomes (e. 2004). larger numbers enable one to improve the signalto-noise ratio. representing a 1-in-5 or 1-in-10 chance of missing a real treatment effect. for relatively uncommon conditions. the size of treatment effect one wishes to detect needs to be specified (e. The smaller the treatment effect one wishes to detect. duration of remission) all need different calculations. Choice of subjects. Patients who could potentially be at risk from one of the trial interventions need to be excluded..

then this means that one is not reliant on clinicians completing forms. randomisation can be considered when both doctor and patient are uncertain about which treatment is preferable. It is helpful for trial organisers to liaise with the statistician and the person designing the database for storing information to ensure that forms are clear and unambiguous and that data are coded wherever possible because coded answers are easier to analyse than free text fields. they should remain in the analyses. Data should be collected on simple. r Keep it seldom – don’t repeat assessments too many times. However. allowing a wide range of patients to be recruited.. Analysing data: the intention-to-treat principle Once patients are randomised into a trial. Data that can be collected only on a minority of patients are unlikely to influence the trial’s conclusions. This sort of wide.Robert Hills to include in a potential trial. it is possible to have selection criteria that are too rigid and exclusive. then ITT analyses tend to be conservative: they underestimate the treatment effect. Collecting data As pointed out earlier. and there can be (virtually) complete ascertainment of the primary outcome measures. national records (e. on participants and clinicians. one needs to be sure that the data are really relevant. Can. additional patient information can be captured. For a given patient. indeed. data are to be analysed. with a view to using them for analyses. Collecting data that will not be analysed puts an extra burden. In this way.g. but only by running large-scale randomised trials with wide entry criteria can questions of which treatment is best for which patient be answered with any confidence. Not only does restricting the population too much reduce the pool of potential participants. would then be applied. even if they stop taking the treatment or. it is important to choose a small number of primary and secondary outcomes to assess the effectiveness of a new treatment. Alternatively. if one sees a difference. But. they should not be collected. clinicians may have differences in opinion. The main idea behind randomisation is the creation of equivalent groups of patients. The ethical imperative then would be to try to reduce uncertainty by contributing to the evidence. but the results of the trial may not be immediately generalisable to other important groups of patients. or indeed whether. perhaps. If it is unclear how. never even receive it. To obtain good follow-up: r Keep it short – don’t ask for unnecessary data. Presented with the same circumstances. well-designed forms. it may be preferable to ask clinicians or patients to complete 60 . not only for the statisticians and data managers but also for clinicians. Keeping data collection to a minimum also increases the likelihood of it being collected. it has been argued that it is preferable to avoid prescriptive inclusion and exclusion criteria and aim for a representative sample from which generalisable results can be obtained (Collins et al. other than diagnostic and safety-related criteria. albeit cautiously. if patients are left out of the analysis. pragmatic entry condition has an added benefit. Of course. without any ultimate purpose. 1996). So. If a wide range of patients is randomised. whether different groups of participants respond differently to the treatment. r Keep it simple – don’t add unnecessary extra tests. it is important to collect data to identify precisely what type of patient has been entered. of mortality) be used to determine outcomes? If so. the groups cease to be equivalent. It is often tempting to include extra ‘nice-to-know’ data. and the most appropriate way of achieving this would be to recruit the patient into a well-designed clinical trial.. Intention-totreat (ITT) analysis analyses every patient according to the treatment to which they were allocated rather than what they received. for example. age can be applied as an exclusion criterion on arbitrary grounds. However. not what was meant to happen. No further restrictions. It would be unethical for a patient to have their treatment chosen at random if either they or their doctor are certain about what treatment they prefer. before becoming committed to extra work. then one can be reasonably sure that it really is there. making it more difficult to achieve target recruitment.3. For this reason. which may require extensive recoding before analysis. as can be seen in Box 5. In cases in which there are protocol deviations. Other sorts of analysis may look attractive because they compare what actually happened. It is also important to determine exactly how these data will be collected. it is possible to examine. 2002). Forms that are well designed and as short as possible increase the likelihood of completion (Edwards et al. there may be uncertainty about the relative merits of different courses of action. these other analyses can lead to misleading results. forms at predetermined time points. For example. additional to the main paper.

the as-treated analysis produces an 8% survival benefit. where direct patient contact may be required. for the ITT analysis.. These drop-outs are not random and may appear more frequently in one arm of the study than another. but it can cause serious problems in quality-of-life studies. For such studies. Because the treatment is toxic but potentially curative. for the as-treated analysis. especially if no further follow-up information is obtainable. Sometimes this is treatment-related. for example. if side effects are distressing or if there is no perceived benefit from the treatment. Allocated new treatment Fit Unfit Survival ratea 50% 50% 50% Allocated control treatment 50% 50% 50% And second. such as a precontact before the questionnaire is sent. which may be enough to change clinical practice. The trial is performed in two types of patients. These survival probabilities are unaffected by the new treatment. In cancer trials in which the outcome is mortality. and ‘unfit’ patients who have a 30% survival probability. making ineffective therapies appear worthwhile. a meta-analysis of methods of improving the response rate to postal questionnaires identified a number of successful strategies (Edwards et al. there will be circumstances in which patients who have committed to the trial do not complete treatment and follow-up. Likewise. Failure to do this will introduce a systematic difference between the groups (known as attrition bias). Receiving new treatment Fit Unfit Survival ratea a Receiving control treatment 40% 60% 46% 60% 40% 54% The survival rate is calculated as 70% times the proportion who are ‘fit’ plus 30% times the proportion who are ‘unfit’ for each group. provision of a prepaid reply envelope and using a short. Consider a trial of chemotherapy with associated toxicity. 2002).Research in cancer Box 5. for a wide range of 61 . however. interesting questionnaire. hence. Whereas an ITT analysis correctly shows that the new chemotherapy is no better in prolonging survival (and. it is more toxic and less attractive than the original option). in addition to obvious monetary incentives. First. In the trial. Ideally. What ITT really measures is the effect of introducing the policy of giving the treatment: what is the effect of introducing a new treatment in the real world? Inevitably. fit patients who draw control treatment may receive the new therapy because they are thought to be fit enough to withstand the side effects. who are recruited in equal numbers: ‘fit’ patients who have 70% survival. so it is crucial to note the reason for non-compliance or withdrawal. this is less of a problem because simple follow-up should be available from clinicians or from national records. 20% of each group receives the opposite chemotherapy. Perils of subgroups A clinical trial sets out to answer whether a particular treatment is effective. The difference arises because fewer poor-risk patients receive the new treatment. some clinicians are not happy to give unfit patients the more-toxic treatment. on average. and they give the less-toxic existing therapy instead. patients who withdraw from treatment should still be encouraged to contribute to data collection. it is a common misconception that a deviation from the protocol necessitates withdrawal. It is possible derive two tables of the proportions of patients surviving. This sort of analysis tends to show an artificial benefit for the treatment received by the better-risk patients.3 Intention to treat.

However. Cascinelli et al. every patient is contributing to two comparisons. but trials that look at subgroups usually need to be much bigger. This method is more complicated if there is evidence of interaction between the treatments. For this reason. Additional topics A common theme in the preceding sections is the need to provide reliable evidence that is likely to influence future clinical practice.1). one will get this kind of spurious result. It is tempting to try and identify subgroups of patients that benefit from an intervention. The first is the factorial design. such qualitative effects need to 62 Table 5. together with a justification of the underlying mechanism and the influence it may have on the outcome.Robert Hills patients. it would be reasonable to give them both together. It is unlikely that. A typical 2 × 2 factorial design. . Thus. one should not conclude. If there are subgroups in which there are legitimate reasons for anticipating a different response to treatment (e. it is possible that. 2001. Again. then if both drugs were shown to work. As a rule of thumb.1. for people born on one particular day of the month. simply because a treatment provides a significant effect in one subgroup and not in another. in a trial of an ineffective treatment.g. So. To test both treatments. prespecified reason why some subgroups should behave differently. but gives greater benefit in one group than another). one can compare patients receiving A + B with patients receiving B. one-quarter receives A only. oestrogen receptor status in trials of hormonal treatments for breast cancer). there is likely to be on average a significant result in one subgroup. Subgroup analyses are therefore typically designed to detect qualitative differences in treatment effects (treatment works in one group. Creutzig et al. Drug B Drug A Control A+B B Control A Control be viewed with scepticism unless there is some plausible. putting brown sugar in their coffee is any better (or worse!) in controlling febrile neutropenia than using white sugar. in addition to receiving either drug B or control (Table 5. Because a similar procedure can be done to assess treatment B. but it is only in a factorial design that such interactions can reliably be investigated. that the treatment works only for one subgroup of patients. With only a finite population of patients from which to choose. if the overall result is positive. to detect a difference reliably within subgroups of the same size as a treatment effect that you have decided is clinically meaningful would require four times as many trial participants as it would to detect an overall effect. (There are a number of examples of this erroneous reasoning in literature: Cascinelli. one can randomise patients in a 2 × 2 factorial trial design to receive drug A or control. If one looks hard enough. a subgroup finding of significant harm is to be relied on. note that a significant result (p < 0. 1994. two design features are widely used in cancer clinical trials so that advances can be made as rapidly as possible. In particular. two research questions can be answered at once.. and those receiving A with control. recruiting enough participants can be difficult. There are statistical techniques to test whether observed effects are real (Assmann et al. they should always be viewed in the context of the overall findings. Factorial designs allow more than one research question to be answered simultaneously. However. it is likely that one will find a statistically significant result – but it is also possible that the result is a fluke. quantitative differences in effect size (treatment works in both groups. 2001. Wheatley and Hills. it does so by considering the population as a whole. This means providing evidence on patients similar to those one would meet in everyday clinical practice and ensuring that the level of evidence is sufficiently strong to enable the making of a decision on treatment. The necessary corollary of this is that trials need to be large. to assess the effect of treatment A. In any case. Although a trial can give some evidence on what treatment to give an individual patient. which will appear purely by chance.05) will be seen by chance 5% of the time. if one looks at 20 different subgroups of patients.. then these should be specified in the trial protocol in advance. It is laughable to say that. one-quarter B only and one-quarter receives control. 1994. there are a number of techniques that can help trials be run more efficiently.. If two different drugs are available and no interactions are expected between them. 2000). One-quarter of patients receives both treatments A and B. Then.) When significant subgroup differences exist. but not in another) as opposed to smaller. and not each separate individual. if one runs a large RCT and ‘data dredges’ through data from enough subgroups.

In such conditions.aml16. and a lively debate.. This ‘pick a winner’ approach should enable useless treatments to be discarded early. 2000).d. with a further randomisation to either 3 or 4 courses of treatment.. For example. see Box 5. it is important that patients are aware of any potential risks. In rarer conditions. various handbooks on GCP. and the recent European Union directive. because relatively large improvements are required from expensive treatments to make them worthwhile. The AML14 trial was intended for patients with AML over the age of 60 and ran in two parts depending on whether patients were deemed fit for intensive chemotherapy. It is therefore paramount that the rights and well-being of participants are considered.4 The AML14 trial. which may tend to reduce the number and size of clinical trials (Burman et al. or an uncertain indication. For conditions with poor prognosis. and treatment lasting in total for three courses vs four courses Factorial designs are common in cancer. and that there is not already good evidence that one or the other treatment is better. the AML14 trial. In particular. it is important to identify as efficiently as possible those treatments that are likely to be the most promising. Other methods can be used to ensure that trials recruit a large. The trial recruited more than 7000 participants to help provide reliable evidence. Ethical considerations Introduction Most clinical research can only be carried out with the cooperation of patients who enter clinical trials.bham. a large trial may take many years to complete. There is a considerable body of literature. on what constitutes good clinical practice (GCP) in clinical trials. For those receiving intensive treatment. one possible way forward is provided by the current NCRI AML16 trial of non-intensive therapy for older patients with AML (see www. which considered the effectiveness of 5-FU/folinic acid–based chemotherapy as well as two different doses of folinic acid and the effect of adding levamisole to chemotherapy (QUASAR Collaborative Group.uk). meaning that many treatments compete for the same small pool of patients. treatments are tested on a small number of patients and those treatments that show no sign of benefit are dropped. To recruit widely. as well as a growing body of guidance and legislation. Here. with only a small chance of inadvertently discarding a good option. Therefore. between two doses of folinic acid and the addition or not of levamisole). One such approach is the pragmatic design adopted by the MRC QUASAR trial of chemotherapy for colorectal cancer. not only on whether to give 5-FU/FA chemotherapy but also to provide evidence on the dose of folinic acid and the role of levamisole. because treatments are usually not without risks. clinically representative group of patients. 2001). Study protocol Perhaps the most important way of achieving goodquality clinical research is by having a well-written trial 63 . the MRC/NCRI trials in acute myeloid leukaemia (AML) typically use factorial designs (e. in which case they were randomised either to receive chemotherapy (randomised in the same 2 × 2 design) or not.Research in cancer Box 5. agree to be treated according to study protocols and who are followed up and their information used to assess the value of a new treatment. clinicians were able to specify if a patient had a clear indication for chemotherapy (in which case patients were randomised in a 2 × 2 factorial design. 2005). the design was a 3 × 2 factorial design for induction. the challenge is perhaps a little different.4. Two courses of induction treatment Daunorubicin 50 mg/m2 vs Daunorubicin 35 mg/m2 vs Daunorubicin 35 mg/m2 + PSC-833 with Ara-C 200 mg/m2 b.g. Burnett et al. One clear area of debate is how to strike the correct balance between protecting the interests of participants and excessive bureaucracy. including the data protection act. vs Ara-C 100 mg/m2 b. giving a total of 12 treatment options (3 × 2 × 2).d.ac.

what data are being collected and how data will be used. the precise way in which they are achieved is still a matter for debate. aimed at academic trialists. 1996). but both stress the importance of having well-written protocols. from the UK MRC Guidelines (available from www. practical reasons. for example.nres . Generally speaking. Another implication of the directive is that trials require standard operating procedures to ensure consistency across the trial and to provide standards against which trials can be audited.mrc. ethical. doses or route of administration. Suspected unexpected serious adverse reactions need to be reported to the ethics committee.int). It is argued that central monitoring can detect anomalies and errors in data more cost-effectively than onsite monitors (Buyse et al. particularly in the academic clinical trials community.Robert Hills protocol.npsa. Strategies may need to be considered to allow participants the time they need (e. onsite monitoring. Also. Various aspects of GCP ensure that the trial is. which involves submitting the study protocol and supporting documentation. Such a protocol is not only a justification for the research being performed. Although these aims can all be agreed on.emea. 64 to the International Committee on Harmonisation (ICH) guidelines adopted by industry-sponsored trials (see www. trials in the EU that involve an investigational medicinal product (IMP) are now subject to the 2001 EU Directive. such as alterations in drug supply. Participation in a clinical trial must be voluntary. are capable of supporting the protocol. an increasing stress on source data verification may use considerable resources for little obvious gain. to a Research Ethics Committee. All trials require ethical approval. and paperwork for ethics committee applications can be found centrally at www. which was transposed into UK law on 1 May 2004. it is sometimes helpful to store copies of all trial documentation on a website so that clinicians can download all the necessary information to keep their own site files up to date. the taking of participants’ consent involves a long period of time. Informed consent All participants in a clinical trial have the right to be adequately informed about the benefits and risks involved in participating in a clinical trial. and informed consent of participants needs to be taken. All serious adverse events (SAEs) need to be reported to the sponsor.nhs. In particular. Clearly.org). Good clinical practice and ethics committees The aims of GCP in clinical trials are twofold: first. it is the trial ‘bible. 1999) and that occasionally monitors in centres have failed to detect problems identified using central checks (Enserink. who must determine whether or not the SAE is expected. Perhaps one of the most important consequences of the EU Directive is that the reporting of adverse events has now been standardised. that the additional work it will entail will reduce the number. and all trials involving IMPs require clinical trial authorisation from the relevant medicines agency (see http://eudract . During the course of a trial.uk). These guidelines differ in the extent to which. to protect the participants and. which explains the research. The protocol needs to be updated with this information and all investigators provided with a copy of the new protocol. For this reason. protocols may be modified as new information becomes available. For studies that include the taking of tissue samples. Information about the trial. is typically given using a patient information sheet.. and the people who will be taking part in the trial. ensuring that a research nurse is present during clinics to help with the process and provide a point of contact for participants). or source data verification is required. and that data are stored securely.g. size and quality of RCTs. Although the aims of the EU directive are laudable. potential participants need to be given time to digest the information they have been given and to ask any questions they may have. second. and a number of guidelines on GCP exist. its arrival has not been met with uniform acceptance.uk. There are concerns. such as treatment schedules and reporting methods. Trials now require a sponsor. consent needs to be obtained for this too. such as patient information sheets. The ethics committee will then assess the protocol and make a judgement. to provide assurance that the trial has been carried out to a sufficiently high standard that results are credible.ich. and remains. thus.ac. it is not .’ It should contain all relevant information and instructions about the trial. and participation in it. and to the Medicines and Healthcare Products Regulatory Agency. or for more prosaic.eu. the possible treatments. each centre is required to undergo a site-specific assessment to ensure that the centre. Additionally. independent Trial Steering Committees (TSC) and set roles and responsibilities of the different people taking part in the study. that data collection is adequate and reliable. that treatments are safe and used appropriately. The ethics application process has been streamlined.

or a parallel study may be set up in a subset of patients or participating centres. Data monitoring committees: is the trial still ethical? As discussed earlier. the safety profile of one treatment may make it unacceptable. For example. This poses a challenge for researchers: how does one recruit enough patients? One option would be to recruit over an extended period. Typically. or. other trial reports and new references). if study recruitment is so slow that a reliable answer is unlikely. These important resources are available for trials that have been funded through the MRC and CRUK. Alternatively.Research in cancer an easy task to strike the right balance in the monitoring and regulation of trials. since the coming of the EU directive. trials can be closed because of safety concerns if the incidence of adverse events in one group is unacceptably high. Some stopping rules set an extreme level of significance (such as p ≤ 0. Increasingly as well. Such commitment to recruitment enables trials to recruit and report quickly. Therefore. Other NCRN initiatives include the development of a computer infrastructure for clinical trials management and the identification of areas where systematic reviews of current evidence are required. Typically. In both cases.ac.g.05. The definition of ‘sufficiently strong’ is one that needs to be determined before the trial starts. In particular the National Cancer Research Network (NCRN) was set up in 2001 to improve the level of recruitment to trials and other well-designed research. notably the Clinical Trials Toolkit (www. Alternatively. but by the time the results are published. and the job of its members is to look at the accumulating data from the trial in light of other external evidence (e. this will involve trying to identify and record all the contacts that study participants have with 65 . the NCRI/MRC trials in AML recruit widely both in the UK and abroad. one group of patients will be receiving care that is known to be suboptimal. then clearly.. the strength of evidence in favour of one treatment over another may be such that an answer is known well before the scheduled end of the trial. and annual recruitment is about 1000 patients. clinical trials in cancer tend to need to be large. then they can recommend closing the trial. but all of them realise that periodically looking at the data increases the chance of a chance positive result. and participants) can identify problems with data collection methods or compliance with treatment.ct-toolkit. 2002). DMCs have another useful purpose: because interim analyses need to be performed. This is something specifically required by NICE both in approving individual interventions for the NHS as part of the appraisal process and in its clinical guidelines programme. out of an incident population of around 2000. Recent initiatives in the UK have recognised the importance of fostering collaboration. making the trial unethical. to provide reliable results.002) throughout the trial. Research networks As we have seen. a number of helpful web sites have been set up to guide trialists through the increasing maze of regulation. There have been a number of different statistical stopping rules proposed (Ellenberg et al. However. In some clinical trials. clinical trials often now include health economic outcomes. trials can be stopped for futility when there is strong evidence that the treatments being compared are not materially different. clinicians. and they therefore require a stronger level of evidence than p < 0. A better approach is to foster the creation of a collaborative group to enable widespread recruitment. The NCRN has set up 34 cancer research networks and employs data managers. a clinical trial only remains ethical while the question being addressed remains relevant. thereby advancing knowledge and practice. If there is good evidence that one treatment is better than another. This is one of the roles of the Data Monitoring Committee (DMC).uk). If they are convinced that the evidence is sufficiently strong and likely to change clinical practice were it revealed to clinicians and participants. up to 200 clinicians take part in these trials. there is a pressing argument to abandon the clinical trial in the face of such strong evidence. as well as for other trials that have been formally reviewed and adopted as part of the NCRN trials portfolio. whereas others adapt depending on how close the trial is to completion. to be kept secret from the trial organisers. and a number of clinical research networks have recently been introduced. the question may no longer be relevant. This group should be independent of the trial organisers. on occasion. nurses and other staff who can provide help and support to clinical trials. the preparation of a DMC report (which needs Health economics There is an increasing need for research to provide evidence of cost-effectiveness as well as clinical effectiveness.

participants in trials will need to know what samples will be taken. therefore. trials with negative or inconclusive results were considered uninteresting and were much less likely to be published (so-called ‘file drawer’ or publication bias.. Of course. Meta-analysis is a powerful tool for doing this: it provides quantitative estimates of a treatment’s effectiveness by combining data from a number of RCTs. and the information used in combination with the effectiveness outcomes to provide information on cost-effectiveness. Of the 28 randomised trials of tamoxifen versus no treatment analysed in the first cycle of the overview. Whereas one RCT may show a significant benefit for a new treatment. the National Translational Cancer Research Network is supporting infrastructure to help facilitate translational research in cancer. Additionally. how these will be stored. This will reduce the chance of repeating previous mistakes and will also stop unnecessary (and therefore unethical) trials from being conducted when there is already enough evidence about the effectiveness of a given treatment. Quality-of-life data may also be collected using a standardised instrument and these data may be used to provide a cost-utility estimate. . Previous trials had been too small to detect such an improvement reliably. the challenge. 1990). the previous research must be in the public domain. as well as wasting resources and abusing the trust placed in investigators by their trial participants’ (Young and Horton. Ethically. Of course. In the past. These can then be costed. As the editors of the Lancet have said: ‘Unnecessary and badly presented clinical research injures volunteers and patients as surely as any other form of bad medicine. This approach can be challenging. 2005). Otherwise. and how results from any samples will be used. others may not. This ‘bench-to-bedside’ approach should lead ultimately to the development of targeted therapies for a wide range of conditions. This new area of laboratory research has led to the development of methods for testing which avenues are the most promising for future research. new trials need to be designed 66 with reference to the findings of a systematic review of previous research. The journal now requires authors to include a clear summary of previous research findings. but it may be possible to carry out a statistical synthesis. One advantage of doing such an analysis is that it can identify important treatment benefits which previous trials had been too small to detect reliably. it is important to be able to synthesise and summarise the results.’ Translational research There is an increasing amount of laboratory cancer research being carried out and with it comes the need to determine whether novel findings in the laboratory can translate into testable hypotheses in patients and ultimately improved outcomes. it is important to consider all the available research evidence. 1987). only 4 showed a statistically significant survival benefit. In most cases this may just be a narrative summary that describes and comments on all the available evidence. Additionally. Synthesising research results: systematic reviews and meta-analysis When deciding the best treatment for a particular patient. Once relevant research has been identified and systematically reviewed for its quality. a search of the published literature would tend only to identify trials that showed significant benefits. such as the ‘quality-adjusted life-year. One such example comes from the Early Breast Cancer Trialists’ Collaborative Group meta-analysis of tamoxifen in breast cancer (Early Breast Cancer Trialists Collaborative Group. a meta-analysis of their findings demonstrated a highly statistically significant (p < 10−6 ) 16% proportional reduction in mortality achieved with tamoxifen. to summarise previous research findings. Such problems have led to the setting up of a number of online journals where such results can be published and put into the public domain. giving an unduly optimistic view of a treatment’s effectiveness. Dickersin et al. Trials need to be designed to collect additional disease markers to identify patients who are likely to benefit from targeted therapies. and a concentration on specific subgroups of patients brings with it the problem of recruiting enough patients to provide reliable and meaningful results. Only by using a meta-analysis was it possible to demonstrate the utility of a treatment that has since annually saved many thousands of lives. gene array technology has dramatically increased the number of markers available for analysis and is a source of hope for identifying prognostic markers that can then be used to help develop targeted therapies.Robert Hills the health services during the trial. is to identify which markers are genuinely prognostic and which are merely artefactual. Yet. it is important to be able to put trial results in context. when there are a large number of potential markers and combinations of markers. In the UK.

with no undue emphasis on particular subgroups or end points. a study needs to be randomised and be of sufficient size to provide reliable evidence. 6. RCTs with definitive results (based on confidence intervals). r Discuss the findings of your study in the context of an updated systematic review.g. Case-control studies. doses. for results of an assessment of a treatment to be reliable. some of the larger medical journals require trials to be reported according to the CONSORT guidance (Moher et al. and reporting of eligibility criteria. a number of journals now allow (and encourage) trialists to publish their study protocols online. 3. 1995): 1.Research in cancer Box 5. 2. There are. Few pieces of research are without some methodological flaws (Altman. A recent monograph (Moher et al. evidence-based practice. As can be seen. A meta-analysis or other review is only as good as the data that go into it and needs to be continually updated in light of new results. 5. dangers in carrying out an inappropriate meta-analysis. Faced with a paper. Results need to be reported fairly. Cross-sectional studies. but do these flaws invalidate the findings? In assessing the merits of different treatments we have already seen that different types of study carry different weight when making evidence-based decisions on health care.. 7. The trial report may demonstrate that the randomisation sequence could have been subverted. hence. For this reason it is important to explore sources of heterogeneity (e. evidence-based practice is impossible. There are many methods of assessing the quality of trials. and overemphasis on any particular trial. the CONSORT statement makes a number of other recommendations. can give misleading findings. Meta-analyses can then be performed with reference to these quality scores to determine whether the quality of the trial affects the result. RCTs with non-definitive results.5 Trials in context. even with perfect reporting.. However. 1999) concluded that more research was needed in this area to determine the effect that trial quality. the totality of trial data that provides the best evidence. Box 5. and the one that will be most familiar. dosing regimens. Assessing published reports of research All the ideas presented in this chapter apply equally well to designing one’s own clinical trial as they do to assessing other people’s research. Perhaps the most important aspect of the CONSORT guidance. equal reliability).. conduct a systematic review or identify a relevant review done by someone else. different end points. Systematic reviews and meta-analyses. after all. adverse events. r When developing the research question. from simple checklists to quality scores. However. The obvious questions that one needs to ask when reading research are whether the results appear believable and whether the methodology is sound. not all randomised trials are of equal quality (and. Conclusions Without proper evidence. it 67 . If there is a large amount of heterogeneity between trials. and a poorly designed trial may well lead to misleading results. it can sometimes be quite difficult to check whether the report matches the original design of the study. have on systematic reviews and. however. However. 2001). the data need to be looked at in relation to previous trials and the impact of the results discussed. then merely reporting the total treatment effect can give misleading results. In assessing the effects of new treatments.5 gives guidelines for putting results in context. It is. r Learn from the achievements (and mistakes) of past trials but read critically. but CONSORT is not a device for improving the design of trials. the method of randomisation and the generalisability of findings. allowing readers to make a direct comparison between the results obtained and the originally proposed methodology. After a new clinical trial has been completed. ultimately. 4. even the one you conducted. There is a generally accepted order of precedence (Guyatt et al. Additionally. This is best achieved by adding the trial into the existing meta-analysis and interpreting any effect that it has on the overall result. showing how many patients did not receive their treatment according to the protocol and how many patients dropped out from (or were lost to) follow-up. and different assessments of trial quality. Cohort studies. including the use of confidence intervals rather than merely p-values. is the requirement to include in a paper a flowchart showing the journey of the patient population through the trial. Case reports. or different patient groups) to determine whether such heterogeneity materially affects the results of the meta-analysis and the estimate of effectiveness. 1994).

V. R. (1987). Design and analysis of Phase I clinical trials.. R. The scandal of poor medical research. 45. R. Lancet. D. R. (2004). (2001). Machin. (1995). Sample Size Tables for Clinical Research. B.. Vol. 283–4. But randomisation alone is not enough. and applying these guidelines. The role of biostatistics in the prevention.. 543. (1994). M. 3. D.. K. R. (2002). (2002). Cohn. (1995).. ed. B. and Farrell. In Oxford Textbook of Medicine. D. Subgroup analysis and other (mis)uses of baseline data in clinical trials. M.. (1995). E. Ritter. and Bland. detection and treatment of fraud in clinical trials. R. 106 (162A). 1499 (letter).. 2nd edn. Chalmers. (1991).. (2000). J. D. T. 121–30. et al. L. A.. J. and meta-analyses of such trials. et al. Assess. et al. 432–3. Haematol. 274. B. D. Campbell. P Roberts. et al. Schulz. Med. Creutzig. Cook.. D. R. 68 . Statistics at Square One. W. 16.. Storer. W. Drug Devel. Clin. (2005). et al. Antibacterial prophylaxis after chemotherapy for solid tumors and lymphomas. Collins. (1987). Altman. 353. M. Pocock. Engl... et al. et al. J. Peto. F. 1064–9. R. Med. 134. Science. How to randomise. M. K. Leukemia. (2000).. 576–8. Moher. F. 319. George. Edwards. 330. S. QUASAR Collaborative Group. Influence of more extensive radiotherapy and adjuvant chemotherapy on long-term outcome of early-stage Hodgkin’s disease: a meta-analysis of 23 randomised trials involving 3. 311. 3rd edn. (1997). Evans. Steven. Soc. A. Warrell. 1191–4. Altman. Lancet. Billingham. Wheatley. (2005). Specht. et al. G. M. M. 1800–4. Trials need to be well designed. S. Ellenberg. It is often said that learning from one’s mistakes is fruitful.. should help produce better research and more reliable interpretations of research. J. together with a healthy scepticism and one’s innate common sense. M. Users’ guides to the medical literature IX: A method for grading healthcare recommendations. Duley. Treatment allocation by minimisation. Dickersin.. A. M. A. and Campbell.. Lancet. Altman.. and DeMets. Absence of evidence is not evidence of absence. 1588–96. Burnett. 2004–5. 152–7. R. J. Comparison of fluorouracil with additional levamisole. Stat. 1. 343. Abstr. D. Oxford: Oxford University Press. and DeMets. Idarubicin improves blast cell clearance during induction therapy in children with AML: results of study AML-BFM 93. and Bland. B.. G.. Jadad. T.. N. Contr.. (2001). T. Assmann. Gray. J.. I. Swinscow. 15. (1994). REFERENCES Altman. K. Clin. T. N. J. Clinical trials: fraud and ethics charges hit stroke drug trial. Trials. B. and Byar. D. Altman. or both. Altman. Statistics with Confidence. registries to compare treatments – the fallacy of omnimetrics. Clinical Trials. (1994). London: Chapman and Hall. Sham procedures and the ethics of clinical trials. D.. M. K. G. A. London: BMJ Books. Statistical Methods for Medical Research. Gray.. (1989). 343–53. Treatment of Early Breast Cancer: Volume 1. Morabito. D. 348–54. Worldwide Evidence 1985–90. C. J. N. G. Moher. Chan. (1999). London: John Wiley and Sons. et al. Ledingham and D. P. et al. use appropriate end points and be properly analysed before their results can be fed into clinical practice. J.. J. U. Machin. M. Oxford: Oxford University Press. 830–43.. 485. L. M. Oxford: Blackwell. Med. D. Clin. G.. rates to postal questionnaires: systematic review. L. Leukemia. Zimmermann. Ann. L. G. S. Cullen.. J. 925–37. 355. et al.. Sinclair. Intern. Med. R. et al. A. A. G. Blood. Wheatley.. 703–4. et al. 1456–8. Cascinelli. 308. 274. D. C. S. London: BMJ Books. and Hills. Subverting randomization in controlled trials.Robert Hills is impossible to overstate the importance of RCTs. for the International Hodgkins Disease Collaborative Group. Med.. Res. 15. 3 (12). A. B. Clinical Trials: A Practical Approach. Fleming. (2001). Breaking the camel’s back: multicenter clinical trials and local institutional review boards. Chichester: Wiley. T. E. et al. M. Current Phase I/II designs: are they adequate? J. Increasing response . This chapter has identified important aspects of researching new treatments in cancer. Guyatt.. and Altman. J. Data Monitoring Committees in Clinical Trials: A Practical Perspective. Inappropriate reporting and interpretation of subgroups in the AML-BFM 93 study. J. 355. Large-scale evidence: trials and overviews.. Lancet. 913–4. Clarke. D. N.. Assessing the quality of reports of randomised trials: implications for the conduct of meta-analyses. J. Bryant. B. M. D. F. H. 18.. G. London: BMJ Books. Fayers. J. 324. (1996). for the CONSORT group.. 1–98. Statist. (2000). J.. S. 274. F. J. Publication bias and clinical trials. (2002) SAB: a promising new treatment for AML in the elderly? Br. K. M. Weatherall. Using observational data from . M. G. Enos.. Milligan. S. Schulz. Miller. higher-dose folinic acid. A. J. (2005). Pocock. D. Lancet. K. D. Sackett. 97. The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials. (1990). Bufalino... Results of adjuvant interferon study in WHO melanoma programme. et al. K.. It is also true that learning from others’ mistakes is less painful. Biometrics. 988–98. M. 2nd edn. Burman. (1998). (1996). D. 357. i–iv. 361–70. L. Prentice.888 patients. (1999). L. G.. and Bland. J. Reves. Adjuvant interferon in melanoma – reply. M. B. Oncol. Buyse. S. J. (1999). Early Breast Cancer Trialists Collaborative Group... B.. Health Technol. J. Enserink. and Kaptchuk. 343.. J. (1996). 1803–4. Modification or dose or treatment duration has no impact on outcome of AML in older patients: preliminary results of the UKNCRI AML14 trial. M. 1183. J. Clarke. Storer. S. D. as adjuvant chemotherapy for colorectal cancer: a randomised trial. 118. (2002). Green. J. 1. D. M. (2001). D. (2002). 843.. 8. 3435–51. N. Cascinelli. so critical reading of existing research is a very good way of starting to design one’s own trial. P (1984). L.

S. Lancet. Assmann et al. Acad. C. 3rd edn. R. Part II: analysis and examples. Peto. M. I. of randomized clinical trials requiring prolonged observation of each patient. J. P et al. 6. Pike. 409–22. A proposed charter for clinical trial data monitoring committees: helping them do their job well. Design and analysis . (2000). The Cochrane Collaboration: preparing. Duley and Farrell (2002). (1984). R. Yusuf. (1976). and Horton. Br. P et al. Y. (1977). (2005). Br. R. 585–612. (1996).Research in cancer Young. 1–39. Putting clinical trials into context. (1993). Ann. 366. Med. 107–8. Peto. N.. (1984). Chalmers. 35. 69 . the following references are valuable resources. Pike. Cancer. Why do we need some large. Collins et al. Part I: introduction and design. M. Additionally. simple randomized trials? Stat. 365. Peto. of randomized clinical trials requiring prolonged observation of each patient.. Oxford: Blackwell. R. C. Lancet. 3. Med. and Yusuf et al. Armitage.. C. 233–44.. Armitage. J. R. Collins. Cancer. 34. (2006). and Peto. (2005). The DAMOCLES Study Group. Sci. How to Read a Paper. maintaining and disseminating systematic reviews of the effects of health care. 156–63.. Greenhalgh. Why do we need systematic overviews of randomized trials? Stat.. Design and analysis . 703. FURTHER READING There are many excellent books and articles on different aspects of clinical trials.. R. T. 711–22.. notably those by Altman (1991). (1987).

Mannitol infusions may be needed for severe symptomatic raised intracranial pressure that does not respond to steroids. blood pressure. oxygen. r Syndrome of inappropriate antidiuretic hormone (SIADH). in which case a previously agreed plan of action will be helpful. Around 10% of advanced solid tumours give rise to malignant hypercalcaemia. 2004). when pain is present it should evoke an immediate response. Sometimes the emergency situation may be predictable. breathlessness. Adequate hydration. r Bleeding. Some seriously ill patients may need to be transferred to a high-dependency unit (HDU) or intensive therapy unit (ITU). breathing. Thought must be given to the appropriateness and value of investigations and treatment because some patients will be in the terminal phase with progressive and treatment-refractory disease. r Tumour lysis syndrome. and circulation are maintained. and monitoring of fluid balance are important in patients with sepsis or tumour lysis syndrome. and distress should be treated as priorities. It is caused by direct bone destruction by bone metastases or. Steroids are used in patients with SVCO and suspected spinal cord compression. r Urinary obstruction. although the evidence base supporting their use is poor. r Cardiac tamponade. r Raised intracranial pressure. resuscitation measures may be needed to ensure that airway. r SVCO. Treatment overview As with any acute medical emergency. Treatment-related emergencies include r Neutropenic fever/sepsis. End-of-life care should be instituted rapidly as a priority to relieve unnecessary distress. requires prompt assessment and action by appropriately experienced staff. r Acute airway obstruction. which may result in serious morbidity or mortality if not treated quickly. An oncological emergency. especially those with a treatable malignancy and a good prognosis and those who develop complications of curative chemotherapy. like any other emergency. r Extravasation of a chemotherapeutic agent. Liaison with specialist colleagues at an early stage is recommended.6 ONCOLOGICAL EMERGENCIES Paul Shaw Introduction An oncological emergency is an acute medical problem related to cancer or its treatment. in patients without . occurring in approximately 5% of all cancer patients. 70 Metabolic emergencies Hypercalcaemia Definition Hypercalcaemia occurs when corrected calcium is greater than 2. It may be secondary to a structural/obstructive.6 mM/l. Spinal cord compression is the commonest neurological complication of cancer. Structural/obstructive emergencies include r MSCC and cauda equina compression. Treatment of pain is considered in Chapter 7. temperature and respiration. r Pain: this has been named the ‘fifth vital sign’ following pulse. particularly for superior vena cava obstruction (SVCO) and malignant spinal cord compression (MSCC). metabolic or treatment-related complication (Cervantes and Chirivella. especially in patients presenting with end-stage cancer and an oncological emergency. The emergency may be the first manifestation of malignant disease. Pain. The WHO pain ladder is a suitable framework to guide appropriate analgesic use. r Anaphylaxis related to a chemotherapeutic agent. Types of emergency Metabolic emergencies include the following: r Hypercalcaemia.

hypothyroidism and adrenal insufficiency are excluded by checking biochemistry. polyuria and polydipsia.9% or glucose 5% over 15 minutes.9% in 24 hours. Treatment r Fluid restriction to 0. Investigations Serum calcium level. Treatment if corrected calcium is greater than or equal to 3. TFTs and a short synacthen test. and monitored regularly. so use of salcatonin is limited. kept mobile. anorexia. flu-like symptoms and exacerbation of metastatic bone pain. muscle cramps.0 mM/l who are about to have chemotherapy or radiotherapy should be rehydrated. lethargy. Treatment if corrected calcium is less than 3. r Mithramycin. r Side effects of bisphosphonates include gastrointestinal upset. Investigation Hyponatraemia and reduced plasma osmolarity occur in the presence of inappropriately concentrated urine (urine osmolarity > 500 mM/kg). Demeclocycline has a diuretic effect by altering the effect of ADH on renal fluid resorption. r Monitor renal function and serum calcium. phosphate and potassium. cautious administration of 71 symptomatic and biochemical improvement. If left untreated it leads to somnolence. Tachyphylaxis develops. Syndrome of inappropriate antidiuretic hormone (SIADH) SIADH is caused by excess levels of antidiuretic hormone (ADH).0 mM/l or patient is symptomatic Fluid replacement: r Give at least 3 l of sodium chloride 0. r Stop thiazide diuretics. r Gallium nitrate. It occasionally occurs in a wide variety of other cancers and can also be caused by chest infections. If the serum sodium is less than 110 mM. Patients with cancer may also have incidental primary hyperparathyroidism. It is most commonly associated with smallcell lung cancer (SCLC). cyclophosphamide and cisplatin. r The underlying malignancy should be treated to reverse the cause of electrolyte imbalance. r Hypocalcaemia occurs in 50% of patients but it rarely causes symptoms because of a compensatory increase in PTH levels secondary to the decreased calcium levels. . and so water retention and low serum sodium levels occur. Renal failure. myeloma and lymphoma. nausea. Osteonecrosis of the mandible can occur after chronic use. somnolence. such as altered consciousness or fitting. but it does increase calcium excretion. electrolytes and renal biochemistry. by circulating factors such as parathyroid hormone-related peptide (PTH-RP).5 to 1 l/day usually results in r Demeclocycline (600 to 1200 mg/day) should be given if the patient does not respond to fluid restriction alone. coma and death. depressed deep tendon reflexes. coma and death may occur. confusion. nausea. Patients who are symptomatic or who are expected to have a slow response to anticancer treatment should be treated as follows. depression and behavioural changes. it may also be an incidental finding. symptoms are fatigue. Bisphosphonates: r Zoledronic acid has a fast onset and a long duration of action. Other drugs used less frequently for hypercalcaemia include the following: r Calcitonin. then give a dose of 4 mg zoledronic acid in at least 50 ml of either sodium choride 0. hypothyroidism and drugs such as antidepressants. seizure.0 mM/l Asymptomatic patients with a corrected calcium of less than 3. anorexia. which leads to failure to excrete dilute urine. vomiting. r Ensure adequate hydration. head and neck carcinomas.Oncological emergencies bone metastases. r In an emergency situation. r Furosemide should not be given until dehydration has been treated. abdominal pain/constipation. ACE inhibitors. Presentation Patients are often asymptomatic but may experience fatigue. renal. breast. Salcatonin has a very rapid onset and may be used in patients with dangerously high serum calcium levels. regardless of fluid status. PTH-RP is associated particularly with lung. Presentation Hypercalcaemia occurs in 10% of patients with advanced solid tumours. pseudobulbar palsy.

However. But upper cervical cord compression can be treated using opposed lateral fields to avoid having the exit beam pass through the throat and mouth. Approximately 60% of patients will have pain. by needle biopsy or during surgical decompression). This may then cause paraparesis.. This study was closed early after the accrual of 123 patients when the surgical arm was superior at interim analysis. In patients who have not been previously diagnosed with cancer it is important to establish a tissue diagnosis if possible (e. r Plan: a single posterior field is used most often. 2001). Patients should receive radiotherapy as soon as is practically possible after arrival in the oncology unit. a rapid rise in sodium may cause osmotic demyelination syndrome and. Asymptomatic cord compression is estimated to occur in one-third of patients with prostate cancer and bone metastases (Bayley et al. Intravenous contrast improves the detection of intradural and intramedullary tumours. the benefit of immediate circumferential decompression of the spinal cord followed by 10 × 3 Gy fractions of radiotherapy has recently been shown to be superior to radiotherapy alone (Patchell et al. reducing the dose after a few days to avoid toxicity. a lower motor neurone pattern.001).. positioning and immobilisation: patient lies prone or supine. urinary or faecal incontinence or retention).g. 2001). Surgery In selected patients with a single area of MSCC from a solid tumour. p = 0. In addition 10 out of 16 paraplegic patients gained the ability to walk following surgery as opposed to only 3 of 16 in the radiotherapy-only group. loss of sensation and bladder or bowel dysfunction. which may be radicular. A typical field Investigation An MRI of the whole spine is performed to detect the level of compression. with arms alongside. predicted survival greater than 3 months and had not been paraplegic for more than 48 hours. After cross-sectional imaging. as confirmed by a retrospective review of 443 patients treated over 10 years (Poortmans et al.8% sodium chloride may be considered. therefore. The centre of the field is in the midline along the spinus processes and has a usual width of 8 cm. It is particularly associated with breast cancer. r Patients require a multidisciplinary team approach including specialist physiotherapy and nursing care. usually ensuring that the inferior and superior limits cross an intervertebral space. Initial treatment Initial treatment includes the following: r Dexamethasone 16 mg is given daily in divided doses with proton pump inhibitor protection. A radiotherapy technique is as follows: r Patient preparation. In this study patients were included if they had a good performance status. Structural/obstructive emergencies Malignant spinal cord and cauda equina compression This condition is caused by pressure from tumour (growing directly between vertebral bodies or growing from bone metastases) or a collapsed vertebral body on the spinal cord or cauda equina. lung cancer and prostate cancer. r Localisation and target volume: simulator films are taken at the level of spinal cord compression. Presentation Symptoms and signs will depend on the level of compression. Because the spinal cord ends at about the L1 level. with the percentage of patients retaining the ability to walk after surgery being greater than those receiving radiotherapy alone (84 versus 57%.Paul Shaw intravenous 1. 72 . or paraplegia. Role of radiotherapy Radiotherapy is the most commonly used treatment of MSCC.. the infusion should be reduced or stopped as soon as the patient’s neurological condition improves. Multiple levels of compression may be found. Polystyrene knee support and/or head support may be used. r The patient should be nursed in the supine position if there is any possibility of spinal instability. the target volume includes the level of compression and one vertebral body above and below this level. It is important that patients would need to be carefully selected for this approach. Muscle weakness often occurs before sensory loss or autonomic dysfunction (impotence. compression above this point will give an upper motor neurone pattern of weakness and below this point. Referrals occur most frequently on Friday afternoons. 2005). Typically the field borders are defined at the time of simulation.

It is usually associated with SCLC. giving oxygen as required and steroids (e. no change in 54%.g. r Thrombosis should be appropriately treated if present. germ cell tumour or SCLC. nasal congestion. 1997). arm oedema and plethora. The symptoms are worse on bending forwards and they include neck and face swelling. and the authors concluded that 8 Gy in a single fraction is an appropriate dose in patients with NSCLC who generally have a poor prognosis. It can also give staging information about the underlying tumour. venous collaterals. and this will often allow time to establish a tissue diagnosis. thymoma and germ cell cancer are rarer causes. Investigation SVCO is rarely an acute emergency. r A stent can be a useful holding measure in patients who require urgent symptom relief. whereas stent insertion provided higher rates of response more rapidly (Rowell and Gleeson. 2001). and 40 Gy in 20 fractions over 4 weeks in patients with non-small-cell lung cancer (NSCLC. 2006). with proton pump inhibitor protection. Sometimes mediastinoscopy or thoracoscopy may be required (Ostler et al. NSCLC and lymphoma (NHL more frequently than HL). A radiotherapy technique is as follows: r Patient preparation positioning and immobilisation: the patient lies supine. and so there is usually time for further investigation aimed at making a histological diagnosis (Ostler et al. conjunctival suffusion. such as solitary plasmacytoma of bone. Examination reveals raised jugulo-venous pressure. and deterioration in 32%. r Dose. fractionation and energy: for patients with metastatic disease. A CT scan provides information about both the site and the cause of the obstruction. unless there is associated airway obstruction and stridor. It seems appropriate that highly responsive tumours (NHL. A Cochrane systematic review concluded that chemotherapy and radiotherapy were equally effective at relieving SVCO secondary to lung cancer. HD. 20 Gy in 5 fractions over 1 week. SCLC) should be treated with chemotherapy first. 1997).. whereas patients with NSCLC and severe SVCO should get a stent first. Presentation The onset is typically insidious over weeks and results in compensatory collateral venous channels in the territory of the SVC. invasion or occasionally intraluminal thrombus. An attempt should be made to establish a histological diagnosis before starting treatment. If opposed lateral beams are used then the same dose may be given but it is prescribed to the ICRU reference point (centre of the intersecting beams). p. which account for more than 90% of cases. Treatment Treatment includes the following: r Initial management: sit the patient up. sputum cytology or biopsy of enlarged neck nodes. For patients with primary tumours. r Radiotherapy is the mainstay of treatment for patients with other solid tumours. epistaxis. A chest X-ray often shows a widened mediastinum. either by CT-guided fine needle aspiration or biopsy or by bronchoscopy. r Chemotherapy is indicated for patients with chemosensitive tumours such as lymphoma. Occasionally there is 73 . Overall there was an improvement in motor function in 14% of patients.. headache. The functional outcome was equivalent for short-course regimens (8 and 20 Gy) as well as for long-course treatment. dizziness and syncope. 366).. by distinguishing external compression from intravascular thrombosis. Rades et al. a higher dose may be required with the aim of achieving a cure (see Chapter 31. sudden occlusion and the patient becomes acutely unwell. r Specific treatment is tailored to the individual underlying disease. 30 Gy in 10 fractions over 2 weeks. dexamethasone 12 to 16 mg daily in divided doses). with a headrest and arms by Superior vena caval obstruction (SVCO) SVCO is caused by compression.Oncological emergencies length for a posterior treatment field would be 10 to 15 cm. typical doses include 20 Gy in 5 daily fractions or 30 Gy in 10 daily fractions given as either an applied dose or prescribed at the depth of the spinal cord (which is determined using the MRI scan) using 6-MV photons. The outcome was related to the development time of motor deficit prior to radiotherapy (> 14 days being better than a shorter time interval). Areas of current interest A recent retrospective study of radiotherapy dose has compared 8 Gy in a single fraction.

efforts should be made to obtain a histological diagnosis. slow pulse rate and raised blood pressure. with arms by sides. 10 Gy in single fraction or 16 Gy in two fractions) if the patient’s performance status and prognosis are poor and if the field size is limited to 12 × 12 cm.v. Investigation Patients with upper airway obstruction should have direct visualisation by laryngoscopy or bronchoscopy according to the level of obstruction. r For specific anticancer treatment. papilloedema and sixth cranial nerve palsy. loss of spontaneous retinal venous pulsations. Focal signs may present depending on the site of the tumour. ataxia and changes in personality and behaviour. Acute airway obstruction Acute airway obstruction is blockage of the main-stem bronchi. seizure. Treatment Treatment is as follows: r Give dexamethasone 12 to 16 mg daily in divided doses with proton pump inhibitor cover. where available.g. if a single metastasis is found. r Provide analgesia for headache (paracetamol with or without NSAID or stronger analgesia as documented in WHO pain ladder). p. r Avoid fluid overload. r Localisation and target volume: simulated treatment fields or. which are most often metastatic. Immediate resuscitation with mannitol and steroids should be instituted. Herniation of the cerebral peduncle can result in hemiplegia.g. 36 Gy in 12 fractions over 2. This may result in quick symptomatic relief with minimal visits to the radiotherapy department. neoadjuvant chemotherapy should also be considered. carina. patients who are most likely to benefit from whole-brain radiotherapy are those who are mobile and have had a good symptomatic benefit from steroids. Special case: obstructive hydrocephalus Patients presenting de novo with tumours causing obstruction to the flow of cerebrospinal fluid and resulting in obstructive hydrocephalus should be considered for ventricular-peritoneal shunt insertion. Raised intracranial pressure Raised intracranial pressure results from the spaceoccupying effect of intracranial tumours. Patients with localised disease and better performance status may benefit from a higher dose (e. energy and fractionation: palliative fractionation regimens may be used (e. Investigation A CT scan of the brain will detect most brain metastases. nausea. r If severe (rapidly falling Glasgow Coma Score or moribund state).5 weeks) in the hope that the tumour and symptomatic control will be prolonged. visual disturbance. If there are multiple metastases. Clinical clues as to the level of obstruction include the presence of neck swelling and stridor with upper obstruction or . the sitting position is used. trachea or larynx and is commonly caused by direct tumour extension from lung cancer or head and neck cancer. neurosurgery plus radiotherapy may be appropriate. 1996).. If the patient is unable to lie flat. In most patients who have not been previously diagnosed with cancer.Paul Shaw his or her sides. r Dose. Occasionally radiotherapy is given with radical intent in appropriately selected patients such as those with NSCLC whose disease can be encompassed in a radical volume. over 15 minutes. CT planning is used with the target volume to cover the superior vena cava. Such cases should be discussed with the neurosurgical team. Presentation Patients present with headache. 442). A biopsy might not be appropriate in frail patients who have multiple cerebral metastases 74 Presentation The patient presents with dyspnoea and stridor. tumour and mediastinum. and possibly other sites of metastatic disease (see Chapter 38. In these patients. give mannitol 0. MRI has a greater sensitivity for detecting small metastases. A randomised controlled trial has shown that 12 Gy in 2 fractions on consecutive days is not inferior to 30 Gy in 10 fractions over 2 weeks in patients with symptomatic cerebral metastases and poor performance status who need treatment (Priestman et al. vomiting. meningeal disease and fourth ventricle obstruction.5 to 1 g/kg i. especially if there is a single brain tumour. Clinical findings may include visual field loss. There may be a reduced consciousness level.

the incidence of GI haemorrhage is estimated at 5% (Maisey et al. Treatment Treatment is as follows: r Heliox (79% He. clotting screen and renal and liver profile. and circulation first. In haematemesis.g. sarcoma and colon cancers. 2004). Haematuria may occur with malignant tumours involving the genitourinary tract but it occurs most commonly in renal. These endobronchial treatments have been reviewed by Morris et al. emergency tracheostomy or endotracheal intubation may be required. This reduces the work required to breathe when the upper airway is obstructed. Patients may present with asymptomatic haematuria. HDR endobronchial brachytherapy (192 Ir) Airway stent Endoscopic intervention Argon plasma coagulator Bleeding Bleeding is more likely to occur in patients treated with anticoagulants. Oesophagogastric tumours rarely present with acute GI haemorrhage. Multivariate analysis identifies shock.1. gastritis. or severe pain caused by clot retention. With the move from surgical treatment towards chemotherapy with or without radiotherapy for the treatment of primary gastric lymphoma (commonest extranodal site of NHL). which can be combined with chemotherapy and/or radiotherapy. . These tests also provide prognostic information in the case 75 External beam radiotherapy Patients are usually treated with 20 Gy in 5 fractions to reduce the chance of larger single fractions increasing oedema. r If the upper airway is acutely compromised. who concluded that ‘good to excellent short term palliation’ may be achieved. age. approximately 2 to 5% of upper GI bleeding is related to malignancy. external beam radiotherapy is all that is required. breathing. Table 6. r Specific intervention should be designed to diagnose and treat the obstruction. which has a lower density and therefore lower specific gravity than oxygen. anaemia and hypovolaemic shock. bladder and prostate cancer. It may be associated with coagulation disorder. nitrogen or air.1 shows the options available (adapted from Freitag. 2004).2).Oncological emergencies Table 6. Then r Perform a full blood count. during turbulent flow. renal failure or advanced malignancy (Palmer. Interventional bronchoscopy For the majority of patients. Chest X-ray or CT scan of neck and thorax or both should be considered. As a result. upper or lower GI endoscopy. co-morbidity and specific endoscopic findings as independent variables predicting re-bleeding and death (Palmer. It may be associated with respiratory difficulty and can lead to rapid deterioration with airway obstruction. 2004). It is defined as expectoration of more than 100 ml of blood in a single episode during 24 to 48 hours. associated symptoms related to the underlying cancer. cystoscopy) to establish diagnosis and identify the site of bleeding. monophonic wheeze (on auscultation) with lower airway obstruction. 2004). Mallory-Weiss tears may be secondary to vomiting induced by chemotherapy. bronchoscopy. However. Even in patients with cancer. (2002).. The Rockall score is a risk assessment tool for GI haemorrhage (Rockall et al. 1996: see Table 6. haematemesis may be caused by benign disease (peptic ulcer disease. Examples Massive haemoptysis is most commonly associated with lung cancer but is also associated with endobronchial metastases from carcinoid. kidney. or fungal infection. Assessment and investigation Secure the airway. Endobronchial intervention in lung cancer Lung pathology Bleeding from central airway tumour Intraluminal tumour Nd-YAG laser Electrocautery Argon plasma coagulator Cryotherapy Photo-dynamic therapy Intramural tumour Extrinsic compression/ airway wall destruction Adapted from Freitag (2004). r Perform a CT scan or endoscopy (e. thrombocytopenia. breast. the flow velocity will be higher when heliox is used. Patients with a Rockall score of six or more have a predicted mortality of around 50%. oesophagitis.. 21% O2 ) contains helium. duodenitis).

proton pump inhibitors. statin). (1996). effective in 90%). Recurrent rectal cancer or pelvic metastases may result . visible or spurting vessel BP = blood pressure. r Radiotherapy to the tumour bed. – All other diagnoses Malignancy of upper GI tract – Renal failure. The Rockall scoring system Variable Age (years) Shock Co-morbidity Score 0 < 60 None None Score 1 60–79 Pulse > 100 bpm. Site-specific interventions include the following: r Haemoptysis – bronchoscopy/radiotherapy as discussed earlier. normal BP – Score 2 ≥ 80 Pulse > 100 bpm. fibrin glue and human thrombin (Palmer. Treatment For patients who require active resuscitation/intervention: r Secure airway. fresh frozen plasma). no SRH Major SRH None – Blood in upper GI tract. systolic BP < 100 mmHg Cardiac failure. r Perform a urine or sputum microscopy and culture as appropriate. r Recognise and treat underlying renal impairment.Paul Shaw Table 6. Adapted from Rockall et al. other major co-morbidity Diagnosis Mallory-Weiss tear. adherent blood clot. because co-morbid conditions can decompensate in the presence of acute haemorrhage. 76 Urinary obstruction This is associated with urological or gynaecological tumours. vitamin K. liver failure. especially carcinoma of prostate or cervix.g. which can be monitored by measuring the CVP. bpm = beats per minute. single fraction of 8 Gy or 20 Gy in five fractions over 1 week. shock and co-morbidity) and two endoscopic variables (diagnosis and major SRH) are each categorised as shown in the table. r Patients with haemoglobin below 10 g/dl should receive a blood transfusion. 2004). r Renal artery embolism for renal tumours bleeding into the urogenital tract. r Actively bleeding and shocked patients should be managed in a high-dependency unit assuming this is appropriate for the individual patient. giving a maximum total score of 11. r Haematemesis options: r Drug therapy (e. SRH = stigmata of recent haemorrhage. For patients in the terminal phase of advanced malignancy who experience massive and uncontrollable bleeding such as carotid blow-out or massive haemoptysis. disseminated malignancy Score 3 – – of upper GI bleeding to direct the appropriate level of care (Palmer.g. r Haematuria options: r Radiotherapy to the prostate or bladder with a single fraction of 8 Gy (or a planned volume for radical treatment).2. somator Endoscopic therapy (e. A score of 0 to 3 points is awarded for each category. Patients with a Rockall score of 6 or more have a predicted mortality of around 50%. and cardiovascular or cerebrovascular disease. ischaemic heart disease. no lesion seen. direct injection of adrenaline into bleeding ulcers. r Use fluid resuscitation with normal saline or colloid to restore blood pressure and urine output. platelet transfusion. Three clinical variables (age. intravenous midazolam and diamorphine provides rapid sedation and palliation. breathing and circulation. 2004) and heat and mechanical devices. r Consider tranexamic acid (with caution in haematuria because of risk of clot retention) or specific measures for a bleeding disorder (e. r Cystoscopy with electrocautery/laser.g.

apy to the pericardium could be considered: 30 Gy in 10 fractions over 2 weeks. Urinary tract infection may occur because of the obstruction. Appropriate early communication with intensive care specialists can be very helpful in selection of patients for intensive organ support. The effect of sepsis-induced vasodilation producing hypotension can rapidly cause end-organ (renal. Constipation is a reversible cause. A CT scan of the pelvis can be used to identify the site and cause of obstruction. short predicted duration of neutropenia. Pain and urinary tract infection should be treated. and chance of response to anticancer treatment.5◦ C). which reduces cardiac filling and leads to circulatory compromise. Treatment The decision of whether to place stents should take into account the patient’s performance status.Oncological emergencies in bilateral ureteric dilatation and hydronephrosis. orthopnoea and weakness. anuria and raised creatinine. Presentation The patient presents with depressed neutrophil cell count (< 1. Septic shock is sepsis with hypotension. increased pulsus paradoxus and oedema.5 × 109 /1 neutrophils) and recent cytotoxic therapy in the presence of raised temperature (> 37.g. good performance status and absence of clinical signs of serious infection are considered to be predictive of a low risk. tachycardia. Partial obstruction may present with alternating polyuria and oliguria. The more common occurrence of febrile neutropenia without shock must still be treated promptly. by percutaneous nephrostomy. Presentation The patient may be asymptomatic or present with flank pain. Treatment Pericardiocentesis and a pericardial window should be performed if fluid re-accumulates. but it is more easily managed with oral or intravenous antibiotics and close observation. Cardiac tamponade is most commonly associated with a malignant pericardial effusion from lung cancer. increased serum lactate or a change in mental status. instigate prompt resuscitative measures which include possible transfer to ITU/HDU. radiother- . In cases of septic shock. Cardiac tamponade Increased intrapericardial pressure occurs due to excess pericardial fluid. Treatment-related emergencies Neutropenic fever/sepsis Neutropenic fever/sepsis may occur after almost any chemotherapy regimen but is mainly associated with cancers that have been treated with intensive myelosuppressive regimens (e. despite adequate fluid resuscitation. decreased urine output. hepatic. Investigation A renal tract ultrasound may show bilateral hydronephrosis. There are signs of haemodynamic compromise – raised JVP. cerebral) damage that can be fatal. Initial assessment Identify immediately any patients with septic shock who have low blood pressure. stage of disease. In patients for whom a pericardial window is not thought suitable. chest pain. ovarian cancer and primary cardiac tumours. especially a retroperitoneal or pelvic mass. if this is not feasible. Factors such as solid tumour. Ureteric stents may be placed under local anaesthetic or. oliguria or an acute alteration in mental status. Risk of complications Patients can be stratified into low or high risk of serious complication. lymphoma or leukaemia). 77 Investigation Two-dimensional echocardiography should be used to diagnose the effusion. hypotension. Symptoms include breathlessness. A low-risk patient may sometimes be treated as an outpatient assuming they are supported. Presentation Two-thirds of patients are asymptomatic. along with the presence of perfusion abnormalities that may include lactic acidosis. have a GP and a telephone and live near a hospital (discussed subsequently). Neutropenia with septic shock is a very serious problem and needs to be managed as an acute emergency. assess the haemodynamic impact and assist with obtaining fluid for cytological examination.

It is important to follow local guidelines and consult with the bacteriologist. the use of G-CSF should be considered. and co-amoxiclav 625 mg t.s.o. r Serum lactate. After 30 minutes. consider transfer to ITU. additional cultures and serology are needed. r Record vital signs every 15 minutes. glucose. blood cultures. Investigate the following: r Full blood count and differential.d.o.1 × 109 /l. vomiting. and clarithromycin 500 mg p. r If serum lactate is greater than 4 mM/l consider transfer to ITU/HDU. In all cases the local policy should be followed. The use of amphotericin B or antiviral agents or non-infectious causes of fever should also be considered. if the patient is still hypotensive. r Start 1000 ml N/saline or 500 ml colloid over 30 minutes. similar to the low-risk protocol. and the microbiology department should be contacted to discuss test results and possible antifungal therapy. r Absolute neutrophil count is not expected to fall below 0. r Blood culture from indwelling venous catheter and peripheral site. start 2 l N/saline or colloid over 1 hour (but use caution in cardiac disease) and consider transfer to ITU/HDU. If output remains below this level. After 1 hour. urine output should be 0. antibiotic treatment.s. blood gases and MSU. For patients without an allergy to penicillin. r The patient is registered with GP surgery or health centre. urea. For patients with a penicillin allergy. for 7 days.d. start with oral ciprofloxacin 750 mg b. Febrile neutropenia Patients may be grouped into low-risk and high-risk categories to decide the appropriate management.d. consideration should be given to removing the catheter. creatinine and LFTs. electrolytes. major organ involvement or other co-morbidity. b. start with a combination such as intravenous piperacillin 4. r Sputum culture. r The patient has a committed live-in caregiver.5 g t. monitor urine output via indwelling catheter. r Chest X-ray may be helpful. r Give intravenous antibiotics according to local hospital policy. give oxygen (24% if COPD or previous bleomycin). If there is evidence of renal impairment. 78 Use of G-CSF For patients with febrile neutropenia and a high risk of complications. the intravenous antibiotics should be continued for an additional 24 hours. along with a CT scan of the chest and bronchoalveolar lavage. FBC. patients who have a central venous catheter and have not yet started teicoplanin should do so. In patients without an allergy to penicillin.to wide-bore cannula. Oral antibiotics are then given for 5 days. If the fever persists beyond 48 to 72 hours. r No highly immunosuppressive therapy is involved.Paul Shaw Hypotension is a systolic blood pressure of less than 90 mmHg or a reduction of greater than 40 mmHg from baseline in the absence of other causes. r Electrolytes. r Investigate serum lactate. For those with a penicillin allergy.d. Such patients include those with the following: . r Swabs of any discharging sites potentially culpable. b. For use of antibiotics in low-risk patients. r If there is evidence of hypoxia. r The patient has a telephone at home.5 to 1 ml/kg per hour. for 7 days. r Initiate a chest X-ray if signs suggest a chest infection.d. and gentamicin loading dose of 6 mg/kg. clarithromycin for chest infection and teicoplanin for a line infection. If a pathogen has not been isolated. line cultures. r Urine microscopy and culture. clotting. r The patient lives less than 30 minutes travel time from a cancer centre. Any patient who does not fulfil the low-risk criteria should be treated as a high-risk patient. r The patient is not involved in any current or recent r The patient is not unwell and there are no signs of shock. for example. the following criteria must be met: r Neutropenia is expected to last fewer than 7 days. renal and liver biochemistry. use teicoplanin and gentamicin. For those with specific localising signs. If the fever settles within 48 to 72 hours. additional antibiotics may be added. use ciprofloxacin 750 mg p. Patients should take the first dose under supervision and can then be discharged with an information sheet. Initial management of septic shock Immediate measures include the following: r Intravenous access with medium.

Prevention Prophylactic steroids and antihistamines reduce the incidence of hypersensitivity reactions to taxanes and carboplatin.9%. with 40% suffering mild symptoms..55 to 0.v.5 ml i. respiratory distress or clinical signs of shock then give adrenaline (epinephrine. lie the patient flat and elevate the patient’s legs if hypotensive.35 to 0.Oncological emergencies r Profound neutropenia (ANC < 0. When first used. Patients receiving chemotherapy for solid tumour or lymphoma and who are at risk of bacterial infection and severe neutropenia (< 500 neutrophils per mm3 ) without G-CSF support should be considered possible candidates for prophylactic levofloxacin. r Those who were hospital inpatients at the time of developing the fever. rash. abdominal pain. L-asparaginase may cause anaphylaxis in 10% of patients treated for acute lym- . A recent meta-analysis of trials of prophylactic antibiotics in neutropenic patients has shown a decrease in the risk of death with their use compared to placebo or no treatment (RR = 0. r Age > 65 years. bronchospasm. and repeat after 5 minutes if there is no improvement. hyperuricaemia. give 1 to 2 l of i. Fluoroquinolone prophylaxis reduced the risk for all-cause mortality (RR = 0. The authors concluded that antibiotic prophylaxis. r Pneumonia. p < 0.m. breathing and circulation. r Uncontrolled primary disease. hypotension.. p < 0. Carboplatin sensitivity is unpredictable and reactions may occur following a prolonged course of treatment – at any time during the infusion or indeed days following its administration.v. carboplatin and docetaxel. should be considered for neutropenic patients (Gafter-Gvili et al. 79 Anaphylaxis related to anticancer drugs Anaphylaxis is associated particularly with paclitaxel.6%. clinically documented infections and microbiologically documented infections. G-CSF is not usually used in afebrile patients with neutropenia or in patients with uncomplicated neutropenic fever such as those with none of the features just described. Levofloxacin significantly reduced the incidence of clinically documented infection (3.7 versus 21.77) as well as infection-related mortality. 2005). These reactions were associated with fast infusion rates and had a tendency to occur after the second infusion. It causes metabolic abnormalities. Tumour lysis syndrome (TLS) Tumour lysis syndrome is caused by sudden tumour necrosis either due to treatment or occurring spontaneously. usually in the first few minutes of treatment and with resolution 15 to 20 minutes after stopping the infusion. taxanes resulted in major hypersensitivity reactions in 30% of patients.001) and hospitalisation for the treatment of neutropenic infection (15. Monoclonal antibodies such as rituximab and cetuximab may cause a cytokine release syndrome. and rash. laryngeal oedema and tongue swelling also occur. r Invasive fungal infections. 2005).004) with few adverse effects (Cullen et al. CI 0. r Multiorgan dysfunction.1 × 109 /l). r For all severe or recurrent reactions give 200 mg i. particularly hyperkalaemia. r If shock fails to respond to drug measures.67. Treatment Treat for anaphylaxis as follows: r Disconnect the drug. r Give oxygen.v. 1:1000 solution) 0. chest tightness. preferably with a fluoroquinolone. hyperphosphataemia and secondary hypocalcaemia.81). r Give 10 mg i. fever. r Hypotension. chlorpheniramine.5 versus 7. crystalloid. r Prolonged neutropenia (> 10 days). Patients may present with agitation. phoblastic leukaemia. wheeze. r If there is stridor.52. r Secure airway. CI = 0. hydrocortisone. r Most deaths due to anaphylaxis are associated with giving adrenaline too late. Angioedema and urticaria. Presentation Prophylactic antibiotics A randomised double-blind placebo controlled trial in patients receiving cyclic chemotherapy for solid tumours and lymphoma and who are at risk of neutropenia (< 500 neutrophils per mm3 ) compared the use of prophylactic levofloxacin or placebo for 7 days during the period of neutropenia.

Most commonly. persistent hyperkalaemia. necrosis. Prevention r Early diagnosis requires a high level of suspicion. neuromuscular irritability. K+ . Rasburicase is the first recombinant uricolytic agent (urate oxidase) and it exerts its action on uric acid. and renal function. metabolising it to allantoin. calcium gluconate. sloughing of the skin. it can result in ulceration. r Electrolyte imbalance. r When severe TLS develops. vomiting. 2006).v. with pain and swelling at the site of the intravenous cannula. bulky tumours such as high-grade lymphoma. 2005). In r patients at particular risk. r Monitor ECG. which occurs frequently. which can result in arrhythmias. Patients with lymphoma who have a raised LDH (> 1500 IU/l) are likely to have a high tumour burden and are at increased risk. hyperphosphataemia. It has a role in the prevention and treatment of TLS and is licensed for use immediately before and during the start of chemotherapy. r Therapy for hyperphosphataemia and hypocalcaemia involves oral phosphate binders (aluminium hydroxide 30 ml q. seizure and death. Extravasation of chemotherapy Extravasation is leakage of intravenous drugs from a vein into the surrounding tissue. 14% of patients will still develop renal problems (Bessmertny et al. Unfortunately. and Ca+2 . sulphonate). r The typical biochemical picture is of hyperuricaemia. Table 6. routine uric acid and electrolyte measurements are sensible. myalgia and dark urine. Administer oral sodium bicarbonate to alkalise urine and prevent urate nephropathy in acidic conditions.3 shows cytotoxic agents classified according to the type of reaction they produce. Hydrate the patient with at least 3 l of normal saline per 24 hours to maintain urine output greater than 100 ml per hour with or without diuretic to maintain urate clearance. r Therapy for hyperuricaemia involves sodium bicarbonate to maintain urine pH > 7. More seriously. The safety and efficacy of rasburicase is currently being assessed and compared with allopurinol in a phase III clinical trial (Rampello et al. r Therapy for hyperkalaemia includes cation exchangeresins binding potassium (sodium polystyrene Experience from case reports and small series has resulted in the publication of guidelines for the prevention and treatment of extravasation (Goolsby and Lombardo. Investigation r Check levels of serum electrolytes Na+ . acidosis or volume overload develops. Table 6. the concentration and volume of the chemotherapy drug and the treatment given for the extravasation. injection).4 shows the risk factors for extravasation. intensive care support and continuous monitoring are necessary. r r r Treatment Treatment r Aggressive hydration is required. hypocalcaemia. symptomatic hypocalcaemia. acid/base balance. acute leukaemia and Burkitt lymphoma. The severity of an extravasation will depend on the infusion site. There are some general principles for 80 .. r Renal failure.Paul Shaw TLS is associated with chemosensitive. It is rarely seen in low-grade lymphomas or solid tumours. despite these treatments.) and calcium gluconate (10 ml i. damage to underlying structures and permanent disability. r Renal dialysis is required if hyperphosphataemia. lactic acidosis and renal failure. secondary to hyperuricaemia. Arrhythmias are a common cause of death if left untreated. PO4 −3 . 2006). uric acid.s.0 and allopurinol 600 to 800 mg/day. sodium bicarbonate to correct acidosis and dextrose/insulin injection. hyperuricaemia and anuria/oligouria. The advantage of recombinant urate oxidase is its rapid onset of action. hyperkalaemia.. Presentation Extravasation may present during the administration of chemotherapy or later.d. Pre-existing renal failure may be a contributory factor. Give allopurinol 300 mg/day to prevent hyperuricaemia. Presentation Patients present with the following: r Non-specific symptoms of weakness. nausea. which is five to ten times more soluble in urine than uric acid. it causes pain and localised tissue inflammation.

4. necrosis Examples Anthracyclines Vinca alkaloids Paclitaxel Streptozocin Mechlorethamine Oxaliplatin Mitomycin Irritant Capable of causing irritation and inflammation Platinum compounds Etoposide Irinotecan Topotecan Flurouracil Non-irritant Non-irritant Cyclophosphamide Ifosfamide Bleomycin Fludarabine Gemcitabine Methotrexate Adapted from Goolsby and Lombardo (2006). then give hyaluronidase Topical hydrocortisone cream and a warm compress DMSO = dimethyl sulfoxide. small. r Keep limb elevated with either cold or warm compression as indicated (see Table 6. phlebitis Previous local radiotherapy. the ‘Green Card’ reporting system is in place to collect and analyse data on extravasation. peripheral neuropathy. (www. In the UK. r Provide full documentation and monitoring (consider using a photograph record).org. It is coordinated through the National Extravasation Information Service. or 0.v. based at St. The web address is www. City Hospital in Birmingham. Antidotes for extravasation Drug Anthracyclines Antidote/treatment Topical DMSO 50–99% Topical hydrocortisone cream Cold compress Mitomycin Vinca alkaloids As for anthracyclines Infiltrate the site with hyaluronidase (1500 units of hyaluronidase in 2 ml of water for injection. the management of extravasation.uk) or Allwood and Stanley (2002). blistering. together with specific measures for each chemotherapy drug.Oncological emergencies Table 6.3. r Consider immediate surgical opinion in cases of anthracycline extravasation or when conservative treatment fails to improve symptoms or tissue damage occurs.uk and it provides information on treatment of extravasation.extravasation.9% sodium chloride) using 0. disconnect tubing. Chad’s Unit. aspirate back. inflammation. lymphoedema.5. r Estimate the amount of extravasated drug. sclerosed Duration and chemotherapy dosage exposure to tissue SVCO. For more details see the National Extravasation Information Service Table 6. Cytotoxic agents classified according to the type of reaction they typically produce Class of drug Vesicant Definition Capable of causing pain.org.2 ml injections over and around the affected area Warm compress Paclitaxel Infiltrate site with mixture of hydrocortisone and chlorpheniramine followed by hyaluronidase Warm compression alternating with topical antihistamine cream Cisplatin Infiltrate site with 3% isotonic sodium thiosulphate.5 for examples). radiation recall reactions Needle insertion technique. Table 6.extravasation. cannula in situ. Initial management should include the following: r Stop infusion. r Attempt aspiration of vesicant and administer antidote (see Table 6. 81 .5) if appropriate. Risk factors for extravasation Factor Vein physiology Pharmacological Physiological Radiotherapeutic Mechanical Description Fragile. In addition many centres have developed local protocols. multiple venepuncture sites Adapted from Goolsby and Lombardo (2006). but leave i.

J. Rockall. and Lombardo. Rasburicase: a new approach for preventing and/or treating tumour lysis syndrome. J.. Rev.. 80. 353. 438–47. Engl. M. (1996). Ostler.. et al. .. CD001316. Management of haematemesis and melaena.. Clin. Med. Morris. F. 3. Vulto. radiotherapy. J. F. W. cava obstruction. Cervantes. Nat. 643–8. Pharm. 45 (Suppl.. Oxford: Radcliffe Medical Press. 979–95. Brada.. 4. K. (2004). Freitag. et al.. P. R. J. Risk assessment after acute upper gastrointestinal haemorrhage. Gut. M.. A. 88–91. D. Oncol. N.. N. D. Des. Thorac. 316–21. B. Rampello. A. Direct decompressive surgical resection in the treatment of spinal cord compression caused by metastatic cancer: a randomised trial. Oncol. Ann. P J. Extravasation of chemotherapeutic agents: prevention and treatment. Cochrane Database Syst. and Chirivella.).. Cullen. et al.. Patchell. 4177–85. Goolsby.. 92. Eur. Coll. Antibacterial prophylaxis after chemotherapy for solid tumours and lymphomas. Clin. J. 1928–32.. Fraser. Acta Oncol. Clin. M. Pract. M. et al. A.. Surg. R. Maisey. and Raaijmakers.). Billingham. Budde. Oncol. a modern management strategy. F. Chemotherapy for primary gastric lymphoma: does in-patient observation prevent complications? Clin. M. T. (2004). (R. Always on a Friday? Time pattern of referral for spinal cord compression. 139–43. Postgrad.. Oncol. 988–98.. Clarke. 82 .). Bayley. Schulte. Oncol. 74. 142. 399–404.. 366. Radiol. Interventional endoscopic treatment. M. Blend. Coll. T. et al. Semin. E. Gafter-Gvili. The Cytotoxics Handbook. Coll.. Milosevic. Oncological emergencies. Paul. (2004). A. Priestman. and Stanley. Lung Cancer. P. A. (2005). Oncol. 1052–6. Ann. and Malaguarnera. Lancet. 11. 83–9. et al. 303–10. N. 2). E. and Cairo.. A. S. (2002). Ann. S299–306. . chemotherapy and stents for superior vena caval obstruction in carcinoma of the bronchus. 9. Rades. Logan. Radiol. Palmer. Stalpers. L. Fricia. Dunn. C. A. V. The management of tumour lysis syndrome. Med. Rowell. Radiol. R. (R. Y. Intern. V. (2002). Cancer. et al. O. (2001). Steroids. (2005). M. Superior vena .. Palliative management of malignant airway obstruction. A. T. Godette. M. A prospective study of factors predicting clinically occult spinal cord compression in patients with metastatic prostate carcinoma. Meta-analysis: antibiotic prophylaxis reduces mortality in neutropenic patients... Steven. L. F. 38. (1997). Prior. Med. 42. L. A. 16.. (2006). 48–52. H. 8. M... A. (1996). 15 (Suppl. (2006). L. J. Defining the appropriate radiotherapy regimen for metastatic spinal cord compression in non-small cell lung cancer patients. Curr. Poortmans.. P and Gleeson. (2001). Final results of the Royal College of Radiologists’ trial comparing two different radiotherapy schedules in the treatment of cerebral metastases. et al. R. et al.Paul Shaw REFERENCES Allwood. Regine.. K. 33. N. L. (2005). T. (R. Robitaille.. P Watkinson. Devlin... 40. (2005). (2006). et al. 308–15. Bessmertny. (2004).. 4). Norman. J. Cancer. D. Tibbs. A. A.. (2001). S235–8. F. D.

7

PALLIATIVE CARE
Simon Noble

Introduction
Changing role of palliative care in oncology
The World Health Organisation (WHO) defines palliative care as ‘the active total care of patients whose disease is not responsive to curative treatment, where control of pain, of other symptoms and of psychological, social and spiritual problems is paramount with the achievement of the best possible quality of life for patients and their families as the goal’ (World Health Organisation, 1990). Palliative care should now be considered an integral part of service planning and care delivery in oncology. This chapter covers common problems in symptom control, communication, ethical decision making, and the financial difficulties of patients with advanced cancer.

third have two pains and one-third have three or more pains (Royal College of Physicians, 2000). Pain may be r Related to the cancer itself (e.g. metastatic bone pain). r Treatment related (e.g. neuropathy secondary to chemotherapy). r Related to cancer and debility (e.g. constipation). r Unrelated to the cancer (due to another co-existing condition).

Pharmacological methods
The WHO analgesic ladder (World Health Organisation, 1990) is very useful in managing cancer pain. It can alleviate pain in more than 80% of patients by starting at a level most appropriate to the patient’s pain and increasing in steps until adequate analgesia is achieved (Table 7.1). Adjuvant analgesics may be used at any point depending on the type of pain the patient is experiencing; an adjuvant analgesic is a drug whose primary indication is for something other than pain but which has an analgesic effect for certain types of pain (Table 7.2). When starting someone on an opioid, it is important to prescribe regular laxatives and a suitable antiemetic (such as haloperidol 1.5 mg nocte). Patients should also be written up for adequate breakthrough analgesia, which amounts to one-sixth of the total dose of morphine prescribed for 24 hours (Fallon and McConnell, 2006; see Tables 7.3 and 7.4 for breakthrough doses and conversions to other opiates).

Changing model of palliative care
In the old model of the cancer journey (Figure 7.1(a)), palliative care services would only be involved at the end of life when no further oncological or supportive treatments were available. This was a ‘terminal care’ service for those clearly at the end of life. But, the symptoms that were being controlled occur not only at the end of life but also at different degrees throughout the cancer journey. The new model (Figure 7.1(b)) attempts to dovetail palliative care with active treatment, gradually increasing its involvement as active treatment becomes less appropriate. Some patients, such as those with carcinoma of the pancreas, are likely to need palliative care input early in their illness.

Opioid toxicity
Morphine is glucuronidated in the liver and then excreted through the kidneys. Renal impairment and/or dehydration may lead to accumulation of opioid metabolites and opioid toxicity, which commonly results in cognitive impairment and confusion (Hall and Sykes, 2004). Other causes of confusion such as hyponatraemia, cerebral disease, infection and hypercalcaemia should be considered and excluded. The other features of opioid toxicity include the following:
83

Pain control
Pain is an unpleasant sensory and emotional experience associated with actual or potential tissue damage. It is experienced by up to 70% of patients with advanced cancer and of these, one-third have a single pain, one-

Simon Noble

Table 7.1. The World Health Organisation (WHO) analgesic ladder
Step Step 1 Step 2 Severity of pain Mild Mild to moderate Analgesia Non-opioid ± adjuvant Weak opioid ± non-opioid ± adjuvant Step 3 Moderate to severe Strong opioid ± non-opioid ± adjuvant Morphine Fentanyl Oxycodone Hydromorphone Examples Paracetamol Codeine Dihydrocodeine

Anti-cancer treatments

Terminal care

Table 7.2. Adjuvant analgesics
Drug type Non-steroidal Indication – examples Bone metastases Liver capsule pain Soft tissue infiltration

Diagnosis

(a)

Death

anti-inflammatory drugs (e.g. diclofenac 50 mg t.d.s.) Corticosteroids

Raised intracranial pressure Soft tissue infiltration Liver capsule pain Nerve compression

Anti-cancer treatments Palliative care Grief

(e.g. dexamethasone 8–16 mg daily)

Anticonvulsants/

Nerve compression or infiltration Paraneoplastic neuropathies

Diagnosis

(b)

antidepressants (e.g.

Death

gabapentin 100–300 mg nocte, amitriptyline 25 mg nocte) Bisphosphonates (e.g. pamidronate, zoledronate)

Figure 7.1. (a) Traditional model of palliative care showing set point at which active treatment ceases and comfort-only treatment begins. (b) Current model of palliative care showing gradual introduction of increasing palliative care as active treatment becomes less appropriate.

Malignant bone pain

r Pinpoint pupils. r Myoclonus/metabolic flap. r Visual hallucinations. Patients typically see dark spots
in the periphery of their vision and may think they have seen animals run under the bed. r Drowsiness (severe toxicity). r Respiratory depression (severe toxicity).

r The needs and views of the patient and carers. r Experience of the referring physician. r Availability of the procedure. r Available expertise (e.g. nursing staff) to manage the
intervention. Table 7.5 shows non-pharmacological methods of pain control.

Non-pharmacological methods
Wherever possible, non-pharmacological methods of analgesia should be considered. There are many options available and their use is likely to depend on:

Antiemetic prescribing
Numerous neurotransmitter receptors are involved in transmitting the impulses connected with nausea and

84

Palliative care

Table 7.3. Morphine doses and conversion
Breakthrough dose of morphine = total dose of morphine in 24 hours, divided by 6 (e.g. for someone on MST 60 mg b.d., total 24-hour dose = 60 mg × 2 = 120 mg): Breakthrough dose = {60 mg × 2} ÷ 6 = 120 mg ÷ 6 = 20 mg p.r.n. Converting oral morphine to subcutaneous diamorphine via syringe driver Take the total 24-hour dose of morphine and divide by 3 (subcutaneous diamorphine is three times as potent as oral morphine), e.g. for someone on MST 60 mg b.d.: Subcutaneous diamorphine dose = {60 mg × 2} ÷ 3 = 40 mg diamorphine s.c. over 24 hours

Table 7.4. Morphine potencies
Oxycodone is twice as potent as oral morphine (e.g. 10 mg oxynorm = 20 mg oromorph) Tramadol is stronger than you think! 100 mg tramadol q.d.s. = 80 mg oral morphine = 40 mg MST b.d. Co-codamol 30/500 2 tabs. q.d.s = 24 mg morphine

clopramide may cause restlessness and Parkinsonism). r Consider non-pharmacological measures such as relaxation, acupuncture, and providing small frequent meals that do not have extremes of taste or smell.

Constipation
Constipation is a common symptom facing cancer patients and should be prevented because it can cause: r Abdominal pain and colic. r Intestinal obstruction. r Confusion. r Urinary retention. r Overflow diarrhoea. Consider and, if appropriate, treat the cause of the constipation: r Drugs (e.g. opioids, tricyclics, antacids, phenothiazines). r Dehydration. r Abdominal wall muscle paresis (e.g. from spinal cord compression). r Hypercalcaemia. r Primary or secondary bowel disease. General measures such as increasing fluid intake, high-fibre diet and improved mobility should be introduced. Various laxatives are available and should be chosen for either bowel stimulation or stool softening. Typical starting doses are given but should be titrated up to achieve appropriate results: r Senna (stimulant) 7.5 mg p.o. once a day. r Bisacodyl (stimulant) 5 mg p.o. once a day. r Sodium docusate (softener) 200 mg p.o. twice a day. r Lactulose (osmotic) 10 ml p.o. twice a day. r Magnesium hydroxide (softener) 10 ml p.o. twice a day.

vomiting to the vomiting centre (VC) and chemoreceptor trigger zone (CTZ) in the midbrain (Mannix, 2006). Chemical triggers (such as drugs, metabolites and toxins) are detected at the CTZ, whereas the VC receives input from stretch receptors on nerve terminals (e.g. from the liver capsule being stretched by metastases or bowel dilatation due to obstruction; Baines, 1997) as well as input from higher mental centres (e.g. pain, fear memory) and integrates these with input from the CTZ. Different antiemetics block different receptors and the choice of antiemetic should be guided by the underlying cause of vomiting (Glare et al., 2004; Rhodes and McDaniel, 2001; Tables 7.6 and 7.7). When managing a patient with nausea and vomiting: r Identify possible causes. r Consider probable pathways and neurotransmitters involved to choose the most appropriate antiemetic. r Use an antiemetic regularly and titrate the dose, considering alternatives to the oral route if necessary. r When using combinations of drugs, remember potential interactions (e.g. metoclopramide and cyclizine have an antagonistic effect; haloperidol and meto-

85

Simon Noble

Table 7.5. Non-pharmacological methods of pain control for cancer patients
Procedure Radiotherapy Transcutaneous electrical nerve stimulation Nerve blocks Epidural/intrathecal analgesia For neuropathic pain in clearly identifiable nerve distribution For complex cord-related pains and plexopathies Orthopaedic stabilisation Useful for incident pain (pain on movement), e.g. vertebral disease or prophylaxis of fracture in long bones with metastases Vertebroplasty Useful for stabilisation and analgesia of pain related to vertebral metastatic disease Radiological expertise required Anaesthetic involvement essential Ongoing care difficult if an indwelling catheter is used Orthopaedic opinion required Anaesthetic involvement required Indication Commonly used for pain from bone metastases Used as an adjuvant to other analgesics Other important issues Useful for opiate-sparing Need for transfer to oncology department Requires expertise of physiotherapist

Table 7.6. Different antiemetics
Drug Metoclopramide Mechanism of action Prokinetic, weak D2 (dopamine receptor) antagonist Clinical uses Gastric outflow obstruction Upper GI bleed Liver metastases Carcinoma head of pancreas Haloperidol Good central antiemetic effect (D2 antagonist) Metabolic causes of nausea: Hypercalcaemia Uraemia Antibiotics Opioids Cyclizine ACh (acetylcholine)/H2 (histamine) – central acting Ondansetron or granisetron 5-HT3 antagonist Central causes Raised intracranial pressure Motion sickness/vestibular problems Chemotherapy-induced nausea (Very constipating)

r Co-danthramer (softener and stimulant) caps or suspension twice a day. The choice of laxatives varies in different units. It is our practice to use a combination of softener and stimulant in advanced cancer patients. Start with a combination of senna and magnesium hydroxide, with a typical dose of 10 ml of each, twice a day, although the
86

choice of laxative should be based on individual patients.

Important points
Co-danthramer can cause a sore perianal rash and colours the urine orange. It is not recommended in bedbound patients, especially those with catheters.

Palliative care

Table 7.7. Choices of antiemetics
Cause of vomiting Drug/toxin/metabolic Radiotherapy Chemotherapy Bowel obstruction Choice of antiemetic Haloperidol, levomepromazine Haloperidol Ondansetron, dexamethasone, metoclopramide Cyclizine, hyoscine butylbromide, octreotide, corticosteroids Delayed gastric emptying Raised intracranial pressure Cyclizine, dexamethasone Metoclopramide, domperidone

Docusate acts as a softener at 200 mg twice a day but has softener and stimulant properties at 200 mg three times a day.

Diarrhoea
Consider and treat the causes of diarrhoea. Causes of diarrhoea include: r Clostridium difficile r Radiotherapy. r Drugs (e.g. NSAIDs, laxatives). r Acquired lactose intolerance. Do not forget that opioids cause constipation and that diarrhoea may occur after decreasing the dose or converting to an alternative opioid such as fentanyl. For symptomatic control of diarrhoea consider the following: r Codeine 30 mg p.o. four times a day. r Loperamide 2 mg (1 capsule) p.o. four times a day. Increase dose to 16 mg daily although higher doses have been reported. r Octreotide 300 μg s.c. in 24 hours; up to 1200 μg in 24 hours.

nation to find reversible causes of dysphagia is essential. The following causes of dysphagia in advanced cancer should always be considered: r Obstruction in the oropharynx. r External compression. r Upper motor neurone damage (e.g. cerebral metastases). r Cranial nerve involvement (e.g. metastases at the base of the skull). r Mucosal inflammation (e.g. radiotherapy, infection). r Dry mucosa (e.g. antimuscarinic drugs). The anorexia/cachexia syndrome is very common in cancer patients and it results in decreased appetite, weight loss, decreased energy levels and weakness. The pathophysiology is complex and involves alterations of metabolic, neurohormonal and anabolic mechanisms. Its management should include treating the reversible causes of anorexia, such as nausea, and using appetite stimulants. Corticosteroids and progestagens have been used but there is no strong evidence that their use improves lean body mass or outcome. Nasogastric feeding may be used for carefully selected patients. The decision to offer nutritional support should be a multidisciplinary one involving the patient, their carers and the dietician. As yet, there are no studies to suggest that enteral feeding alters the clinical outcome or quality of life for these patients.

Pleural effusions
Malignant pleural effusions occur in up to 15% of cancer patients, with lung and breast cancers accounting for more than two-thirds of cases. When the primary tumour is chemosensitive, patients with malignant pleural effusions may be treated with systemic chemotherapy. However, the mainstay of treatment is symptomatic drainage. Several options are available: r Repeated thoracocentesis. r Drainage and chemical pleurodesis. r Chronic indwelling pleural catheter. r Pleuroperitoneal shunt. The first two are the most commonly used. However, before embarking upon a potentially dangerous procedure, consider the following questions: 1. Does the effusion account for the patient’s symptoms? 2. Is the effusion recurrent? 3. Is the patient well enough for the procedure? 4. Is the lung likely to re-expand?
87

Anorexia and dysphagia
Dysphagia occurs in more than 20% of patients with advanced cancer. The causes are often complex but may involve alterations in oro-oesophageal anatomy, oral mucosa, paresis of cranial nerves IX, X and XI, and higher cortical functioning (the cerebral cortex is involved in the control of swallowing). A careful history and exami-

Simon Noble

Ascites
Up to 50% of patients with malignancy will develop ascites, especially those with cancer of the ovary, gastrointestinal system or breast. Ascites is associated with a poor prognosis and it is likely to recur. Whenever possible, the cause of the ascites should be established because it may be due to peritoneal seeding, intrahepatic disease or lymphatic obstruction from retroperitoneal disease. The main medical treatment is spironolactone (100 to 400 mg/day) with a loop diuretic such as furosemide (40 to 80 mg/day). Although this management may delay or prevent the need for paracentesis, it must be done with care because depleting the intravascular volume may precipitate renal failure. Paracentesis has been shown to reduce hospital stay when compared to medical treatment for large-volume ascites. Draining the ascites relieves the symptoms but it may precipitate hypovolaemia. Therefore, the fluid should be drained slowly over several hours, clamping the drain for 1 hour after every litre. The use of albumin replacement therapy is not recommended in patients with advanced cancer because it has little sustained effect on long-term albumin levels and there is an increased risk of pulmonary oedema.

Complementary therapies such as aromatherapy may also help the patient (Jackson and Lipman, 2006).

Depression
Depression is a temporary mental state or chronic mental disorder characterised by feelings of sadness, loneliness, despair, low self-esteem and self-reproach. Although as many as 69% of patients with advanced cancer may experience depression, they are often not treated adequately (Barraclough, 1997; Endicott, 1984; Lloyd-Williams et al., 1999). Many of the symptoms (e.g. fatigue and loss of energy) usually used to diagnose depression are very common in cancer patients and so diagnosis can be difficult. Also, thoughts of death may be realistic and a basis for appropriate planning. A number of substitute symptoms have therefore been suggested for the diagnosis of depression in patients with advanced malignant disease: social withdrawal, pessimism and lack of reactivity. The single question ‘Are you depressed?’ has been found to have a high sensitivity, specificity and positive predictive value in some studies but not in others (Lloyd-Williams et al., 2003).

Confusion and delirium
Delirium is characterised by global cerebral dysfunction and by a lowered level of consciousness, combined with disturbed attention, thinking, memory, psychomotor behaviour, emotion and sleep–wake cycle (Lawlor et al., 2000). Delirium is present in up to 42% of patients with advanced cancer. It may often complicate care at the end of life because patients are often unable to give a history or accurately communicate their symptoms. Furthermore, delirium generates distress for patients and their families and impedes communication between them. Managing patients with delirium in the palliative care setting should follow a stepwise approach: 1. Identify the underlying cause and decide whether it is reversible. 2. Identify precipitating and perpetuating factors (e.g. hypercalcaemia or infection) and decide whether they are treatable. 3. Use environmental strategies such as reassurance, re-orientation, the presence of familiar staff and family, limit staff changes and reduce the amount of noise stimulation.

Psychological symptoms
Anxiety
Anxiety is apprehension of danger and dread accompanied by restlessness, tension, tachycardia and dyspnoea without a clearly identifiable cause (Cathcart, 2006). Anxiety may be severe and may also make other symptoms such as pain, breathlessness and nausea worse. It is important to address the underlying factors such as pain or drug withdrawal (including alcohol or nicotine). If nicotine withdrawal is suspected, nicotine patches may be used. Psychological treatment, such as cognitive behavioural therapy, can help patients to identify the thoughts that cause them distress and find ways of challenging them before the anxiety gets worse (Greer et al., 1992; Meyer and Mark, 1995). Drug treatment includes the use of benzodiazepines (e.g. diazepam 1 to 5 mg orally) as required. Selective serotonin reuptake inhibitors may be used to relieve panic attacks if benzodiazepines are ineffective and tricyclic antidepressants in low doses can also be used.
88

Palliative care

4. Benzodiazepines can be used if other options have failed, particularly when agitation and restlessness are prominent. Antipsychotic agents such as haloperidol and levomepromazine can be used, especially when sedation is needed. Elderly patients should receive reduced drug doses because benzodiazepines carry a greater risk of paradoxical agitation. Risperidone and olanzapine should be avoided in older patients with cerebrovascular disease (Breitbart et al., 1996).

Communication issues
The need to communicate sensitive and often upsetting information is not exclusive to palliative medicine. It is an integral part of all health care professionals’ practice, and breaking bad news and handling difficult questions occur commonly in the day-to-day practice of the oncologist. Whether communication occurs within a busy outpatient clinic or on an inpatient ward round, there are several key principles that should form the framework of any difficult consultation. The following ten-step approach can be used as a guide to breaking bad news. 1. Preparation. r Know all facts before the meeting. r Who does the patient want to be present? r Ensure that another health professional (such as a nurse) is present. r Ensure that there is no interruption and you leave your pager with someone. r Ensure that there are enough chairs for everyone. 2. What does the patient know? r Assess what the patient understands is happening at the moment. r Ask for the the patient’s narrative of events. This will give you an idea of how much the patient knows, and his or her preferred vocabulary, etc. 3. Is more information wanted? r Test the waters. r Be aware that it may be frightening for a patient to ask for more information. r Ask, ‘Would you like me to explain a bit more?’ 4. Give a warning shot. r Perhaps start by saying ‘I’m afraid it looks rather serious.’ r Allow time for the patient to respond. 5. Allow denial. r Denial is a powerful coping mechanism. r Denial allows the patient to control the amount of information. 6. Explain (if requested). r Narrow the information gap. r Give the information slowly, step by step. r The details may not be remembered but the way you do it will be. 7. Listen to the patient’s concerns. r Allow time for the patient to express his or her feelings. r Perhaps ask, ‘What are your main concerns at the moment?’
89

Syringe drivers
Syringe drivers are not just for dying patients. They can also be used for patients who: r Are unable to swallow. r Are nauseated. r Have poor absorption because of bowel stasis, pancreatic dysfunction or hypoalbuminaemia, which can cause oedema of the bowel mucosa. r Are too ill to take medicines. Syringe drivers can be used to control pain, nausea, restlessness and colic. It is important that, when combinations are used, the drugs are compatible and do not crystallise. The following commonly used drugs can be mixed together in a syringe driver: r Diamorphine. r Haloperidol. r Cyclizine. r Hyoscine hydrobromide. r Midazolam. r Levomepromazine. r Metoclopramide. A useful website for checking drug compatibilities in syringe drivers is www.pallmed.net.

Care in the last 48 hours of life
Once the team has recognised that a patient is dying, the following should be done: r Stop all unnecessary medicines. r Ensure you write up p.r.n. medicines for all symptomatic eventualities: r Pain – s.c. diamorphine or morphine. r Agitation – s.c. midazolam. r Death rattle – s.c. hyoscine hydrobromide. r Nausea – s.c. cyclizine. r If the patient is on regular analgesics/antiemetics, convert to syringe driver. r Communicate regularly with the patient’s family.

Simon Noble

8. Encourage the patient to express his or her feelings r This is important because it shows your empathy. r Perhaps say, ‘This must all have come as a bit of a shock for you.’ 9. Summary and plan. r Summarise concerns, plan treatment and foster hope. When someone is given devastating news, it is worsened by the thought that there is nothing that can be done. False reassurance is inappropriate, but offering hope in the form of an active approach to symptom control, goal setting and making realistic plans offers some control in a distressing situation. Some clinicians will make a record (written or recorded) of the consultation for the patient to keep and review at a later occasion. 10. Offer your availability. r The patient may need further information because he or she may not have remembered all the details. r The patient may need continuing support because psychological adjustment may take weeks or months. Following are a number of mistakes commonly made during the consultation: r Not checking prior knowledge. r Using jargon. r Giving too much information too quickly. r Not allowing for silence. r Showing discomfort when the patient is distressed.

Not allowing for silence
If there is an uncomfortable pause, we may feel that we have to fill the silence with more information or an inappropriate comment – this should always be avoided. Silence is a valuable tool in communicating. It helps by showing that you are listening and by giving the patient time to assimilate the news, to react to it and to ask questions. As a patient once explained, ‘When you give me bad news, you will pause and all you can hear is silence. Well, all I hear is noise; it’s just that it is internal. When you speak you are interrupting me getting my head together.’

Discomfort with patient distress
BAD NEWS IS BAD NEWS! YOU CAN’T TURN IT INTO GOOD NEWS! By definition, bad news is going to upset the receiver. If a patient cries, he or she is expressing an appropriate emotion in response to devastating news, and we need to be comfortable with that. The problem is that we may feel guilty that we have made them cry and may feel compelled to say something to ‘make it better.’ What happens in practice is that we end up giving patients false hope or reassurance, which often makes it harder for them to come to terms with their illness later on. Sometimes it will pave the way for a highly complicated bereavement.

Giving too much information too quickly
The main task in breaking bad news is taking a patient from a point where he or she knows nothing to a point where he or she understands everything, if he or she wants to. The skill is in slowing down the transition of information so that the patient can handle it. Too much too soon will lead to psychological disturbance and possibly denial. The ‘warning shot’ technique helps test the water and find out how much the patient will allow you to tell. Sometimes the patient will make it quite clear that they do not want any more information; this wish should be respected. When talking to relatives, the principles of communication are the same as talking with patients. However, it is important to acknowledge that their agenda or concerns may differ from those of the patient. Anger and distress are frequently encountered and should be handled sensitively and calmly. Listening to and addressing each of their concerns is essential, although your duty of care must always remain with the patient. Don’t forget that you must respect the patient’s right to

Not checking prior knowledge
One of the commonest mistakes is not checking what the patient already understands about his or her condition. Do not assume anything. Just because one of your colleagues has documented that the patient has been fully informed about the diagnosis and prognosis does not mean he or she has understood or retained the information. Studies have shown that patients can recall only 30% of what is said to them in a consultation and so it is always important that you check beforehand what they know.

Using jargon
When we feel uncomfortable in a consultation, we tend to use more and more technical words to re-establish our status and importance. This commonly happens when patients become upset or angry. Jargon puts up a barrier between us and the patient, reducing the likelihood of clear communication.

90

Palliative care

confidentiality and discuss his or her care only if given permission to do so.

Financial considerations
The impact of advanced cancer is not limited to the physical and emotional. Patients and families commonly experience financial difficulties, especially when the patient or main carer is self-employed. The role of the social worker as part of the multidisciplinary team is important in ensuring that financial aspects of the patient’s life are addressed. Simple things like accessing critical illness payments and pension plans, and arranging mortgage payment holidays, can all contribute to making life more manageable. Most patients will be entitled to financial assistance in the form of benefits. In assessing someone for benefits there are several things that must be known: r The patient’s age (whether he or she is younger or older than 65). r The prognosis of the patient’s condition (i.e. whether his or her prognosis is less than 6 months). r Whether the patient was working and paid National Insurance contributions before he or she became ill. Attendance allowance is paid to individuals over the age of 65 who need help looking after themselves; it pays for carers to ‘attend’ on the patient. The disability living allowance is paid to people under the age of 65 who need help looking after themselves. Incapacity benefit is for people who have worked and had previously made National Insurance contributions. They usually claim it once statutory sick pay (SSP) has ended or if they do not qualify for SSP. Those who have not paid enough National Insurance and who do not qualify for incapacity benefit are entitled to claim income support. The carer’s allowance is not claimable by the patient. Someone, a partner for example, who is looking after the patient can claim it. The DS-1500 is commonly claimed by patients with terminal disease. It is a way of claiming disability benefits for patients under ‘special rules,’ usually when the patient has a terminal condition with a prognosis of less than 6 months.

need to be made for which there is no clear-cut answer. In weighing up the pros and cons of a decision, it is useful to base it around the following ethical domains: r Autonomy. r Beneficence. r Non-maleficence. r Justice.

Autonomy
Autonomy gives the patient the opportunity to make known his or her wishes and to make an informed decision. For a patient to do this, he or she needs the ‘capacity’ to make such a decision. For a patient to have capacity, he or she must be able to: r Take in information. r Understand the information. r Retain it. r Make a decision based on the given information. If a patient has capacity to make a decision, teams should respect his or her wishes. However, doctors are not obliged to give a treatment against their better clinical judgment if they feel it is futile.

Beneficence
Professionals should make decisions with the intention of doing what is most likely to benefit the patient – the intention to do good.

Non-maleficence
Conversely, non-maleficence dictates that when making a decision it should be done with the intention of doing no harm.

Justice
Justice guides us to do what is fair, not only for the patient in question but also for other patients as a whole. It also ensures that we practice within what is legal within our society.

Areas of current interest
The main areas of development in palliative care lie with its place in service delivery. It is now considered an integral part of cancer care, and all patients should be able to access palliative care if it is needed.

Ethical decision making
The management of cancer patients is not solely guided by evidence-based guidelines. Sometimes decisions will

91

Simon Noble

Palliative care teams have an increasing involvement in managing the late effects of curative treatment such as postmastectomy pain or postchemotherapy symptoms such as peripheral neuropathy. There continues to be an interest in the field of opioids’ rotation, by which switching the type of opioids (e.g. morphine, hydromorphone, oxycodone) may improve pain control without an increase in the dose.

Ongoing research
There is an increasing amount of research into the complexities of neuropathic pain and the dorsal horn remodelling that occurs secondary to cancer. Currently, several large clinical trials are looking into the management of the anorexia/cachexia syndrome, breathlessness, and fatigue.

REFERENCES
Baines, M. J. (1997). ABC of palliative care: nausea, vomiting, and intestinal obstruction. B. M. J., 315, 1148–50. Barraclough, J. (1997). ABC of palliative care: depression, anxiety, and confusion. B. M. J., 315, 1365–8. Breitbart, W., Marotta, R., Platt, M. M. et al. (1996). A double-blind trial of haloperidol, chlorpromazine, and lorazepam in the treatment of delirium in hospitalized AIDS patients. Am. J. Psych., 153, 231–7. Cathcart, F. (2006). Psychological distress in patients with advanced cancer. Clin. Med., 6, 148–50. Endicott, J. (1984). Measurement of depression in patients with cancer. Cancer, 53, 2243–9. Fallon, M. and McConnell, S. (2006). The principles of cancer pain management. Clin. Med., 6, 136–9. Glare, P Pereira, G., Kristjanson, L. J. et al.(2004). Systematic ., review of the efficacy of antiemetics in the treatment of

nausea in patients with far-advanced cancer. Support. Care Cancer, 12, 432–40. Greer, S., Moorey, S., Baruch, J. D. et al. (1992). Adjuvant psychological therapy for patients with cancer: a prospective randomized trial. B. M. J., 304, 675–80. Hall, E. J. and Sykes, N. P. (2004). Analgesia for patients with advanced disease: 1. Postgrad. Med. J., 80, 148–54. Jackson, K. C. and Lipman, A. G. (2006). Drug therapy for anxiety in palliative care (Cochrane review). The Cochrane Library, issue 3. Chichester, UK: John Wiley and Sons. Lawlor, P. G., Fainsinger, R. L. and Bruera, E. D. (2000). Delirium at the end of life: critical issues in clinical practice and research. J. A. M. A., 284, 2427–9. Lloyd-Williams, M., Friedman, T. and Rudd, N. (1999). A survey of antidepressant prescribing in the terminally ill. Pall. Med., 13, 243–8. Lloyd-Williams, M., Spiller, J. and Ward, J. (2003). Which depression screening tools should be used in palliative care? Pall. Med., 17, 40–3. Mannix, K. (2006). Palliation of nausea and vomiting in malignancy. Clin. Med., 6, 144–7. Meyer, T. J. and Mark, M. M. (1995). Effects of psychosocial interventions with adult cancer patients: a meta-analysis of randomized experiments. Health Psychol., 14, 101–8. Rhodes, V. A. and McDaniel, R. W. (2001). Nausea, vomiting, and retching: complex problems in palliative care. CA Cancer J. Clin., 51, 232–48. Erratum in: CA Cancer J. Clin., 51, 320. Royal College of Physicians. (2000). Principles of Pain Control in Palliative Care for Adults. Working Party Report. London: Royal College of Physicians. World Health Organisation. (1990). Cancer Pain Relief and Palliative Care. WHO Technical Report Series, 804. Geneva: World Health Organisation.

FURTHER READING
Watson, M., Lucas, C., Hoy, A. et al. (2005). Oxford Handbook of Palliative Care. Oxford: Oxford Medical University Press.

92

salivary glands. middle ear and squamous carcinoma of unknown primary. except for salivary gland tumours. oral cavity. Smoking of cigarettes and cigars is related to cancers of the oral cavity. Buccal mucosa cancer may develop at the site of maximum contact. mandible and hyoid. The UK annual incidence is approximately 14/100 000 cases per year (CRUK. bounded by sternomastoid. pharynx. trapezius and clavicle. in Southeast Asia). nose. 2002): r Level I: Ia = submental. General features of head and neck tumours Types of head and neck tumour The majority of head and neck cancers are squamous carcinomas. with peak incidence occurring in the 60. 2004). r Level II: upper deep cervical. catechu and tobacco. The risk varies among tumour subsites: the association is strongest for pharyngeal cancer. r Sun exposure causes lip cancer.8 HEAD AND NECK Laura Moss and Chris Gaffney Introduction It is not possible in a book with this format to discuss the treatment details for each tumour type. 2004). r Nitrosamines cause nasopharyngeal carcinoma. r Epstein Barr Virus (EBV) causes nasopharyngeal carcinoma. Anatomy The anatomical relations of the head and neck regions are discussed in the various subsections. Tobacco chewing is also a risk factor (e. 93 . extending from skull base to hyoid. Regional variation in the UK is 8 per 100 000 in the Thames region compared to 13 to 15 per 100 000 in Wales and northwest England (NICE. The cervical lymph nodes are described as follows (AJCC. r Level IV: lower deep cervical. Betel quid.to 75-age group. Subsequent sections focus on specific tumour sites. stage and subsite. which is chewed. extending from hyoid to cricoid. r Alcohol acts as both an independent risk factor and synergistically with smoking.1.g. Smoking of pipes is involved particularly in carcinoma of the lip. Ib = submandibular nodes. is a mixture of betel nut and leaf. Risk factors and aetiology for head and neck tumours Prolonged exposure to environmental factors plays a key role. which account for 90% of all cases. Incidence increases from middle age. paranasal sinuses. Approximately 3000 deaths occur per year in England and Wales (CRUK. This chapter highlights treatment principles and controversies. larynx and pharynx. The first section considers general features of head and neck cancer. lime. extending from cricoid to clavicle. 2004). r Level V: posterior cervical. r Previous radiotherapy increases the risk of carcinoma. r Level III: mid deep cervical. This chapter covers the anatomical sites of the lip. The range of tumours is shown in Table 8. larynx. Boundaries include anterior and posterior bellies of digastric. 2004). Incidence and epidemiology of head and neck tumours r Approximately 8000 new cases occur per year in Engr r r r r land and Wales (CRUK. r Wood dust causes adenocarcinoma of nose and paranasal sinuses. Head and neck tumours are more common in males than females (male-to-female ratio of 2:1).

oropharynx. thyroid. Clinical presentation of head and neck cancers Symptoms depend on the anatomical site of primary disease and tumour stage. 2002). r Cranial nerve palsies – VII with parotid gland. r Level VII: inferior to suprasternal notch in the superior mediastinum. r Dysphagia. tongue base and supraglottic primaries. 94 . r Level VI: paratracheal. r Dysphonia – larynx. Range of head and neck tumours Type of tumour Benign Examples Glomus tumour Pleomorphic adenoma Pseudotumour Fibroma Papilloma. cartilage. hypopharynx. approximately 90% are squamous carcinomas arising from mucosal surfaces. r Via the bloodstream. Spread Head and neck cancer can spread: r Locally to involve local soft tissues. odynophagia and aspiration – larynx. extending from the hyoid to the suprasternal notch and laterally to the carotid sheath. middle ear. stridor – larynx. r Anorexia. Pathology Cancers typically arise on mucosal surfaces and may be exophytic or ulcerative. r Otalgia/referred pain – pharyngeal tumours. nasal regurgitation – nose. hypopharynx. r Hearing impairment – nasopharynx. infraorbital nerve with advanced maxillary sinus. r Weight loss due to dysphagia or odynophagia or cachexia. r Dyspnoea. r Epistaxis. breast. Screening for head and neck cancer There is no evidence that screening the general population or high-risk groups leads to a reduction in mortality or morbidity (British Association of Otorhinolaryngologists – Head and Neck Surgeons. bones and liver. prostate or gastrointestinal carcinoma Of microscopic features. Remember that the usual pattern of lymphatic drainage may be disrupted by previous surgery or radiotherapy. IV and VI if cavernous sinus involved. r Neck/facial mass lesion. II. nasal discharge. malignant ameloblastoma Soft tissue sarcoma including Kaposi’s Osteosarcoma Lymphoma: NHL and Hodgkin lymphoma – usually in lymphoid tissue of Waldeyer’s ring Extramedullary plasmacytoma Mucosal melanoma Basal cell carcinoma Merkel cell tumour Malignant secondary Metastatic spread from renal cell. adenoid cystic and acinic cell carcinoma (salivary gland) Ameloblastic carcinoma. and peripheral nerves. nasal obstruction. Bilateral nodal disease is most common with nasopharynx. r Via lymphatics to cervical lymph nodes. paranasal sinuses. pretracheal.1. to the lung. nasopharynx. inverted papilloma Granuloma Haemangioma Juvenile angiofibroma Lymphangioma Angioma Adenoma Chondroma Ameloblastoma Malignant primary Squamous carcinoma Adenocarcinoma Neuroendocrine including olfactory neuroblastoma Mucoepidermoid. bone. which is less common and usually occurs late. r Cervical node enlargement. hypopharynx. lung. Severe dysplasia or carcinoma in situ has a 15 to 30% risk of progression to frank malignancy. III.Laura Moss and Chris Gaffney Table 8. Symptoms related to the primary site include the following: r Persistent mucosal ulcer – oral cavity.

Head and neck Symptoms of metastatic disease (rarely the presenting symptom) include the following: r Haemoptysis. r Nutritional assessment. clinical findings and radiology. consider local anaesthetic spray. Treatment overview: general aspects in head and neck cancer The choice of treatment modality depends on tumour site. asymmetry. If it is difficult to obtain an adequate view because of gagging. CXR lacks sensitivity. r US/CT liver may be indicated if there is locally advanced disease or if there are abnormal liver function tests. r Hypercalcaemia (the primary lesion may produce PTH-like protein or may be due to metastatic disease). r Carry out indirect laryngoscopy or flexible nasendoscopy. bronchoscopy and biopsy depending on the symptoms. anatomical site) in the clinic. reduced hearing or ear discharge. Elective placement of a gastrostomy may be needed if there has been significant weight loss before diagnosis or if there are plans to carry out either an extensive resection resulting in impaired tongue mobility or swallowing function or irradiation of a large volume of mucosa. r An isotope bone scan may be indicated to assess bone pain but is not ideal for assessing mandibular invasion. r US neck. epistaxis. Note that dental amalgam produces an artefact that obscures CT findings in the oral cavity. The consensus guideline standards for audit include CT thorax in 90% of cases prior to treatment planning (British Association of Otorhinolaryngologists – Head and Neck Surgeons. whether normal. r Examine the nose and ear if there are symptoms of nasal obstruction. r Check laryngeal movement over the prevertebral fascia (loss of ‘crepitus’ indicates a locally advanced laryngeal tumour). patient’s wishes. disease stage. presence and extent of cartilage involvement. 95 .e. 2002). cosmetic result and survival rate. or pulmonary metastases. Pretreatment assessment Investigation and staging in head and neck cancer: general considerations Diagnostic and staging investigations will depend on the primary site and associated symptoms. reduced or fixed). r Speech and swallow assessment. which may include panendoscopy. co-morbidity. Assessment in a multidisciplinary clinic is essential. performance status. number and architecture). level. Examination techniques for head and neck cancer are as follow: r Inspect and palpate the mucosal surfaces of the oral cavity (remove dentures) and inspect the pharynx. r Look for tongue and soft palate movement. r Look for trismus (inability to open the jaw due to parapharyngeal space invasion). synchronous primary lesion) and cervical nodes (size. discharge or regurgitation. Local staging r Flexible or rigid nasendoscopy (depending on the r Examination under anaesthesia. It may also be considered for cancers of unknown primary when determining the site of origin may influence management (Intercollegiate Standing Committee on Nuclear Medicine. microlaryngoscopy. r Carefully palpate the cervical lymph nodes (sensitivity and specificity for tumour involvement is 60 to 70%). This may also demonstrate mandibular invasion in advanced oral cavity carcinoma. r Bone pain. If you are in a clinical examination and it is not possible to see the larynx well under examination conditions. The following is a list of commonly used investigations for the head and neck cancer patient. local extension. r 18 FDG-PET is indicated for assessment of possible tumour recurrence where conventional imaging may fail to differentiate between posttreatment fibrosis/ oedema and recurrent malignancy. pooling of saliva in the pyriform fossae and vocal cord movement (i. r US parotid with possible FNAC. pathological subtype. tell the examiner what you are looking for: mucosal ulceration. r CT or MRI to assess primary site (size. 2003). fine needle aspiration cytology (FNAC) of suspicious cervical nodes. r Dental assessment with an orthopantomogram. oesophagoscopy. mediastinal lymphadenopathy. functional outcome. Staging for distant metastases r CT of the thorax is used to look for synchronous primary tumours. direct laryngoscopy. site.

hence. Role of radiotherapy in head and neck cancer: general aspects Radiotherapy planning issues in general: patient positioning (The chapter does not discuss in detail CT-planned conformal radiotherapy or intensity-modulated radiation therapy. Consider nutritional support: will nasogastric tube or gastrostomy/jejunostomy be required? Consider introduction to a patient previously treated in the same manner. laryngeal and pharyngeal cancers. and a straight position for larynx. Beam arrangement and energy: r Opposed lateral photon fields with the patient supine are used to treat the majority of oral cavity. The electron beam Elective nodal treatment Elective nodal treatment is usually performed when the risk of occult node involvement is greater than 20% (British Association of Otorhinolaryngologists – Head and Neck Surgeons. lower alveolar ridge). as well as lymph nodes levels I to V. 96 . If the patient is undergoing microvascular reconstruction and entry to the neck is required then nodal dissection is usually performed. The risk of occult node involvement is greater than 20% in most sites (except lip. Is resection possible with primary closure or is reconstruction of the defect required? The decision whether to proceed with surgery or radiotherapy will depend on the following: r Cure rate: is the survival outcome better with one modality compared to the other? r Operability of the disease: is there skull-base involvement. early glottis. If posterior neck electrons are to be added. The lower the junction and. This is often arranged prior to laryngectomy operations. Measurements are taken from cross-sectional imaging or from the transverse patient outline. which is generally accepted practice but there is no strong evidence. the electron beam energy is chosen following assessment of depth of nodal groups at risk and spinal cord depth. 2002). however. which removes one or more additional lymphatic or non-lymphatic structures. internal jugular vein. an anterior and posterior field arrangement may be needed to provide adequate dose coverage. sternomastoid. Surgery for lymph node areas Lymph node dissection types are as follow: r Radical. which preserves one or more of the r Selective neck dissection. r Consider a mouth bite for oral cavity and paranasal sinus treatments. aforementioned structures. oral cavity and oropharynx and paranasal sinuses. r For posterior neck electron fields. r If the posterior neck nodes are involved in the lower neck. Role of surgery in head and neck cancer Surgery for individual subsites is discussed in the relevant sections. r Extended radical neck dissection. Elective nodal radiotherapy (RT) instead of neck dissection with 50 Gy in 25 fractions can achieve local control rates of 90 to 95%. the longer the lateral fields. and such that it is possible to acquire planning images and verify treatment delivery. involving removal of the spinal accessory nerve. For pN designation following selective neck dissection six or more lymph nodes are needed in the resection. reproducible. the spine is to be kept in a straight position.) Patient positioning: r The patient’s position should be comfortable. swallowing and breathing be better with one of the options? r Cosmetic result/disfigurement/altered body image. r Choosing the angle of orientation of lateral fields is important when taking cervical spine curvature and shoulder position into account. encasement of carotid vessels and so on that would make the disease unresectable? r Patient co-morbidity: is the patient at risk from a prolonged general anaesthetic? r Functional outcome: will physiological functions of speech. r When deciding the position of the junction between lateral and anterior neck fields. the greater the mucosal surface included and the acute mucosal reaction. consider where the high-risk areas lie and try to avoid the junction there. An extended position should be used for nasopharyngeal and parotid tumours.Laura Moss and Chris Gaffney It is necessary to decide on a treatment modality for the primary site as well as the need for elective neck treatment in the absence of clinically apparent cervical nodes. r Modified radical. which involves removal of specified lymph node groups. r Patient preference.

r Lymphovascular invasion. r Follow a soft diet. Lymph node factors: r Extra capsular spread. r Perineural invasion.g.** r Poorly differentiated. is the dose prescription point going to be at the skin surface or at 2 to 3 cm depth where the lymph nodes are? Consider the proportion of the dose from phases 1 and 2 when large lateral fields are required: because spinal cord tolerance is less than the dose required to treat head and neck cancers radically. r Use topical barriers with carmellose (e. Shielding: If central spinal shielding in the anterior neck field is omitted. Alternative methods include half-beam blocking or angling the match line so that the angle exceeds the divergence of the anterior field. antipsychotics. Therefore. r Invasion of adjacent soft tissue. great care must be taken to calculate dose to the spinal cord from the two photon phases and the electron beam. For oral mucositis. For odynophagia. Gelclair® ).* r ≥ 2 nodes. r Sugar-free gum or pastilles may stimulate saliva flow if residual function is present. compound codeine preparation.g. There is a poor correlation between subjective reports of xerostomia and gland function. A combination of two or more intermediate factors would also be classified as high risk. anticholinergics. r Avoid spicy foods. antidepressants. r CIS or dysplasia at resection margins.** .** r Single node > 3 cm.g. r There is conflicting evidence with regard to antibacterial rinses/lozenges. normal saline) and dental hygiene. 97 General indications for postoperative radiotherapy in head and neck cancer Primary site factors: r Positive margin. r Use aqueous cream for erythematous and dry skin. NSAIDs or antihypertensives). r Heat and moisture exchanger base plates will need to be avoided if there is peristomal skin reaction. a double asterisk indicates intermediate risk factors. r Use systemic analgesia (e. r Use sucralfate suspension. The anterior field beam divergence at depth will contribute dose to the spinal cord if the lateral and anterior beams are matched at the skin surface.g. alcohol. r Use benzydamine hydrochloride mouthwash (diluted with water to reduce pain on initial contact with mucosa). Dose: For the anterior neck field. diuretics. r Use raspberry mucilage.** r T3 or T4 disease.** (A single asterisk indicates high-risk factors for locoregional recurrence. r Avoid medications that may exacerbate the situation (e. Check the distance from the skin surface/stoma to the spinal cord to ensure spinal cord tolerance is not exceeded.) Management of acute radiotherapy reactions For management of skin reactions.* r Close margins (≤ 5 mm). r Use hydrocolloids like Intrasite® gel or Geliperm® for moist desquamation. Leaves are often 10 or 3 mm. the full dose cannot be given in a single phase. Orabase® ) or with povidone and sodium hyaluronate (e. Portals are shaped with Cerrobend® blocks or multileaf collimators. insert a small shielding block in the inferior/posterior corner of lateral portals to prevent excessive dose to the cervical spinal cord. then a. r Avoid perfumed soap and toiletries. alcohol-free preparations are better tolerated) if severe.** r Multicentric primary. very hot or cold foods/ liquids and smoking. If higher-energy electrons are used. When deciding on the relative doses between the phases you must stipulate the acceptable dose to the spinal cord from the phase 1 photon field to remain within cord tolerance. For xerostomia: Xerostomia is the subjective sensation of mouth dryness. r Take frequent sips of water and maintain adequate daily fluid intake. b. r Perform regular mouth washing (e. r >1 lymph node level. statins.Head and neck r r r r energy is usually 9 to 12 MeV if the patient has no palpable posterior nodes and has an average neck size. r Avoid sun exposure. NSAID.g. r Treat oral candidiasis. morphine sulphate solution. r Avoid wet shaving. opioids. antispasmodics.

This improvement increased to 8% at 5 years with concomitant chemotherapy use (Pignon et al.g.. 2002).. Same total dose. In nasopharyngeal cancer management.. e. hyperfractionation). Randomised studies fall into three types: 1. Trial data have shown an increase in local control when neoadjuvant chemotherapy is combined with surgery or radiotherapy with some trials also showing an increase in survival. 2004) reported by the RTOG and EORTC respectively.e.e. e. EORTC 22791: 1. 2. 1992). Two phase III trials (Cooper et al. 2003).g. Postoperative concurrent chemoradiotherapy Despite the use of surgery and postoperative radiotherapy. decreased overall treatment time (i.15 Gy twice daily up to 80. treatment-na¨ve patients are far more tolerant ı of the associated treatment toxicities. Use pilocarpine 5 to 10 mg t. i. There is a significant increase in acute toxicity. DAHANCA (six fractions per week). A meta-analysis of randomised studies (>6500 patients) showed a 3% improvement in survival and a 6% improvement in locoregional control for modified fractionation. 2000) but a further analysis of nearly 2500 patients receiving only platinum and 5fluorouracil regimens showed a 16% improvement in relative survival and a 5% absolute survival benefit at 5 years (Monnerat et al. SST® .e.. A meta-analysis showed a non-significant 2% survival benefit when looking at various chemotherapy regimens (Pignon et al. This meta-analysis only addressed concurrent chemotherapy with conventionally fractionated radiotherapy.. Biot` ne Orale balance® gel). Increased total dose. This treatment showed increased acute toxicity but a significant improvement in tumour control probability (Overgaard et al. same overall treatment time (i. high-risk squamous carcinoma (e. Salivix® ). Role of chemotherapy in head and neck cancer: general aspects Neoadjuvant or induction chemotherapy The rationale for neoadjuvant chemotherapy is that drug delivery is better in well-vascularised untreated tumours. and the potential exists for eradication of microscopic metastatic disease. This regimen showed a significant improvement in local control when compared to conventional fractionation (Horiot et al. Acceleration aims to overcome tumour cell repopulation. The use of concurrent chemotherapy is not proven with accelerated or hyperfractionated regimes.. moderate acceleration).s. therefore. r r Radiotherapy fractionation Altering conventional fractionation regimens aims to improve the therapeutic index and to increase dose intensity (high total dose in a short overall treatment time) and may be achieved by increasing the total dose (via hyperfractionation.5 Gy. This treatment showed a marked increase in acute toxicity and decreased late morbidity (Dische et al. 3. CHART (54 Gy in 1. 2004. Decreased total dose and markedly decreased overall treatment time (i. 1997).g.. There are a significant number of contraindications and side effects.g. which looked at the addition of concurrent chemotherapy to Brachytherapy Services should be cancer network based. Follow scrupulous oral hygiene. vice provision for highly specialised treatments (Royal College of Radiologists. 2001). Decreasing the dose per fraction can protect late-responding tissue. 2002). There will often be a need for cross-network supra-regional ser98 . Undergo regular dental review. e. positive margins and extracapsular lymph node spread) of the head and neck frequently recurs in the locoregional area.e. it is important to select patients with good performance status and limited comorbidity.g. Use artificial saliva preparations (e. this treatment approach has shown a reduction in locoregional recurrence and distant metastases as well as an increase in overall survival. Some studies report increased salivary flow but this often fails to translate into improved quality of life..8 weeks). 2000). decreased dose per fraction) and/or decreasing overall treatment time (via acceleration). especially mucositis.d. Concurrent chemoradiotherapy A meta-analysis of 63 randomised trials of chemotherapy used in different clinical settings with more than 10 000 patients showed a survival benefit of 4% at 5 years.. very accelerated).Laura Moss and Chris Gaffney r Use saliva stimulants containing a combination of citr r ric and mallic acids (e. Bernier et al. Hyperfractionation showed the greatest benefit.g. with a 9% improvement in survival at 5 years (Bourhis et al.

The aim is to treat different targets simultaneously with different total doses and fraction sizes. the volume of tissue exposed to high dose is less. Palliative chemotherapy Untreated metastatic or recurrent squamous cell carcinoma has a median survival of 4 months. Collimation can either be static. Patient and organ movement are of great concern. the patient may experience loss of sense of smell. that is. The dose is delivered to the target through a greater amount of surrounding tissue. Although the volume of tissue exposed to a low radiation dose is greater. with the EORTC study also reporting an increase in overall survival (53 versus 40% at 5 years). demonstrated an improvement in locoregional control. chemotherapy is more often used in this setting to palliate symptoms and restore function rather than to prolong life. NICE. r Long-term nutritional assessment and supplementation are often required. Areas of clinical interest include (1) conforming target volumes to be able to treat radically at-risk posterior lymph node groups while avoiding excessive spinal cord irradiation (i. 2002). swallowing and speech. For example. Access to videofluoroscopy is important to assess swallowing function. with most responses being partial and often short lived. impairment of the Valsalva manoeuvre and ability to strain and loss of humidification and air warming.g. the ability to deliver horse-shoe-shaped 99 Rehabilitation following radical treatment for head and neck cancer Many surgical procedures may result in significantly altered physical appearance and physiological function. However. Prognosis in head and neck cancer The status of the cervical nodes may be the single most important prognostic factor. eye). r Detection of second primary. light writers. The prognosis is affected by the number of nodes. electronic larynx.e. allowing a single accelerator beam to function as many smaller beams. Response rates in the order of 20 to 30% are seen. Survival is significantly worse when metastases involve lymph nodes beyond the first echelon of lymphatic drainage (British Association of Otolaryngologists – Head and Neck Surgeons. r Dental follow-up is required. Speech and language therapy input can help in the long term for both voice and swallowing. Areas of current interest in head and neck cancer Intensity-modulated radiation therapy (IMRT) IMRT is a type of conformal therapy that confers a high radiation dose to the target volume and a low dose to sensitive structures. dental implants and prostheses (nose. IMRT is usually inverse planned. including for modification of r Late morbidity and quality-of-life assessment. palm top aids. treatment planning starts with a description of the desired result. The aims of IMRT are to minimise acute and late morbidity and to allow the possibility of dose escalation with improved local tumour control. Follow-up after radical treatment for head and neck cancer Aims of follow-up include the following: r Early detection of residual or recurrent disease with a view to salvage treatment. r Social and psychological support are very important. . r Psychological support. number of nodal levels involved and the presence of extracapsular spread. text messaging and e-mail). even partial response can be associated with a significant improvement in quality of life when these patients are often dealing with cosmetically disfiguring disease and restricted breathing. 2004). r Prosthetics include dentures. with a ‘stepand-shoot’ method or dynamic by changing beam shape during radiation delivery. A highly conformal plan can usually be achieved using five or seven equally spaced fields. oesophageal voice and written communication aids (e. smoking and alcohol behaviour. Although cisplatin-containing chemotherapy has shown a survival advantage in the region of 8 to 10 weeks when compared to best supportive care. following a laryngectomy.Head and neck radiotherapy. The duration of follow-up is usually 5 years (90% recur within first 2 years. The intensity of the beam is varied across the treatment field. unlike conventional treatment planning in which the computer system simulates dose distribution from a treatment planner’s defined set of beams and beam weights. ear. r Communication can be facilitated by surgical voice restoration with speech valves in a tracheooesophageal fistula.

and tumours tend to express high levels of EGFR. an anti-EGFR monoclonal antibody. p = 0. Coronal MRI images can be particularly helpful if partial laryngeal surgery is planned.g. However. arytenoids and false cords. but less specific. Treatment overview for squamous carcinomas of the larynx Glottis It is important to know whether there is supraglottic or subglottic extension and whether the commissures are involved. Investigation and staging Clinical investigation uses IDL or nasendoscopy (to look for laryngeal mucosal appearance – symmetry. in combination with radical radiotherapy in locally advanced squamous carcinoma of the head and neck. preepiglottic space involvement) and cervical nodes.org. r Subglottis: lower border of the glottis to lower border of cricoid cartilage. The situation may arise in which the lesion under review is clinically progressing but histology fails to confirm Clinical presentation Glottic cancers often present at an early stage because of dysphonia (hoarse voice). PARSPORT – a multicentre randomised study of parotid-sparing IMRT in patients with head and neck cancer. the rate of detection of second primaries and unexpected lung metastases is quite low and some clinicians do not feel that CT imaging can be justified in this case.ncrn. accessed September 2006): EaStER Feasibility study – early glottic cancer: endoscopic excision or radiotherapy. r Glottis: vocal cords and anterior and posterior commissures. often present when advanced. Epidermal growth factor receptor (EGFR) signalling EGFR is a transmembrane protein with tyrosine kinase activity that transduces signals from the epithelial cell surface to the intracellular domain. therefore. feasibility study. compared with CT for cartilage involvement. further endoscopic excision or radiotherapy should be considered. exophytic or endophytic disease. an excision biopsy should be performed. in this study cetuximab was not compared to concurrent chemoradiotherapy. Dysphonia can also arise from recurrent laryngeal nerve involvement. Aberrant signalling via EGFR plays an important role in the carcinogenesis of squamous cell carcinoma of the head and neck.. there were two trials registered with the National Cancer Research Network looking at treatment for head and neck cancer (www.2. the entire mucosal abnormality needs to be examined to exclude invasive disease. other ENT primary sites) and examination of the neck for lymphadenopathy/direct tumour extension. In stage Tis (in situ disease). Current clinical trials At the time of this writing. cartilage involvement. a normal CXR does not ‘exclude’ metastases or synchronous primary because of its low sensitivity. However.Laura Moss and Chris Gaffney dose distribution) and (2) use of parotid-sparing techniques to reduce xerostomia. MRI is more sensitive. pain due to cartilage invasion. non-specific sore throat (especially when swallowing) and neck mass. e. has shown promise in increasing median survival (54 months versus 28 months in those receiving radiotherapy alone. stridor.uk. oedema. Comparing CXR and CT of the thorax. Carcinoma of the larynx Anatomy of the larynx Subsites of the larynx include the r Supraglottis: epiglottis. 2004). salivary pooling. Movement artefact may impair MRI images. If there are rapid or multiple recurrences. whereas supraglottic cancers 100 . CT or MRI are used to assess the primary site (extra laryngeal spread. Laryngoscopy/microlaryngoscopy and biopsy are used to assess the site and extent of disease and other primary lesions. Staging classification The staging classification for carcinomas of the larynx is shown in Table 8. vocal cord movement.02) in a phase III study with more than 400 patients (Bonner et al. Cetuximab. aryepiglottic folds. A high degree of EGFR expression correlates with a poorer clinical outcome. Patients can also present with dyspnoea.

r Total laryngectomy and postoperative RT. impaired cord mobility Cord fixation. normal mobility: (a) one cord (b) both cords Supraglottis. normal mobility Mucosa of > 1 adjacent subsite of supraglottis or glottis or adjacent region outside the supraglottis. Tumour and nodal TNM stages for carcinoma of the larynx Stage Glottis T1 T2 T3 T4a T4b Supraglottis T1 T2 T3 T4a T4b Subglottis T1 T2 T3 T4a T4b Nodal stages: all subsites N1 N2a N2b N2c N3 Adapted from UICC (2002). deep/extrinsic muscle of tongue. 101 . N0. soft tissues of neck. carotid artery One subsite. In stage T1 to T2. mediastinal structures. trachea. trachea. thyroid cartilage erosion Through thyroid cartilage. strap muscles. paraglottic space. Ipsilateral single ≤ 3 cm Ipsilateral single > 3 to 6 cm Ipsilateral multiple ≤ 6 cm Bilateral. If the tumour is bulky and causing respiratory compromise or stridor. give radical radiotherapy to the larynx only (there is a low incidence of occult nodes due to sparse lymphatics. r Concurrent chemoradiotherapy.Head and neck Table 8. pre-epiglottic tissues. thyroid. Stage T4 requires total laryngectomy and neck dissection and postoperative RT. there will be a better functional outcome with radiotherapy. oesophagus Prevertebral space. strap muscles. paraglottic space. Stage T3 is a heterogeneous group. soft tissues of neck. deep/extrinsic muscle of tongue. consider surgical debulking before starting radical RT in order to maintain an adequate airway throughout. thyroid cartilage erosion Through thyroid cartilage. mediastinal structures. mediastinal structures. there is no need for elective nodal RT) or conservation surgery (need to preserve more than half of the free margin of contralateral cord). thyroid. If there is extensive superficial T2 disease. thyroid. oesophagus Prevertebral space. contralateral ≤ 6 cm > 6 cm Limited to subglottis Extends to vocal cord(s) with normal/impaired mobility Cord fixation Through cricoid or thyroid cartilage. trachea. In these cases it is often better to treat as for invasive disease rather than continue to observe. without fixation Cord fixation or invades postcricoid area. subglottis. carotid artery Limited to vocal cord(s). Treatment options include: r Primary radiotherapy and salvage surgery. strap muscles.2. If the patient is medically unfit for surgery or declines surgery then proceed with radiotherapy with or without chemotherapy. oesophagus Prevertebral space. carotid artery Description invasion. deep/extrinsic muscle of tongue. therefore.

5 to 7 cm.e. shield the lung apices and the spinal cord. For node-positive disease. r Include the entire thyroid and cricoid cartilages. Shield the oral cavity and salivary glands as much as possible without compromising the nodal areas. bilateral cervical nodes and SCF nodes should be included in the radiation field. If the posterior border of the disease lies close to the spinal cord then a two-phase technique will be needed.Laura Moss and Chris Gaffney Supraglottis There is a high incidence of clinical and occult cervical involvement. A total laryngectomy and postoperative radiotherapy are usually required. those overlying the spinal cord) are treated with electron fields matched at the skin surface to the phase 2 photon fields. The contour change can be compensated for by wedges (early glottic tumours with small lateralopposed fields) or missing tissue compensators when there is a complex contour and variable separations throughout the volume. the nodal stage. include the mediastinum in the radiation field. This technique 102 Supraglottis: radiotherapy There is a high incidence of involvement of the upper deep cervical lymph nodes. N0. To select the appropriate electron beam energy. 1 cm below the palpable thyroid cartilage promontory). If the anterior commissure is involved. the shoulders may obstruct the lateral fields. r Mark any nodes.e. the anterior neck field is matched to the lower edge of the lateral photon fields. If paratracheal nodes are positive. inferiorly. r Margins must allow for movement with respiration and subclinical disease. either use bolus anteriorly or ‘underwedge’ to give the anterior part of the treatment volume between 5 and 10% higher dose than the posterior part. The gantry angle used is usually 45 to 60◦ in order to spare the spinal cord. Choose an electron energy that will cover the depth of the nodal areas sufficiently but without exceeding cord tolerance. 4 to 5 cm. r No tongue bite is required. relevant scars and the stoma. If there is a T3 or T4 glottic or transglottic tumour. and the operation status. For stage T1 to T2. therefore. r Prophylactic nodal RT is not required. Advanced glottis: radiotherapy The PTV depends on the disease extent. Radiotherapy techniques for squamous carcinoma of the larynx General considerations When treating squamous cell carcinoma of the larynx with radiotherapy. the clavicular heads are covered and the lateral field edge lies at the junction of the medial two-thirds and the lateral third of the clavicle. consider using two anterior-oblique wedged fields. especially if there is a midline tumour. In these cases remember to add shielding to the posterior/inferior aspect of the lateral photon field to prevent overdosing the spinal cord as a result of beam divergence from the anterior field. If a tracheostomy is present and the surrounding area is thought to be at risk. r Typical volume sizes: T1 tumours. The posterior neck nodes (i. Early glottis: radiotherapy r Centre the volume at the level of the vocal cords (i. Therefore. preepiglottic space and upper deep cervical lymph nodes. The PTV outline will need to be drawn on a central contour if an orthogonal planning technique is used. consider elective bilateral neck RT. . take the following into account: r Immobilisation/beam direction shell. A two-phase parallel-opposed lateral photon field technique should be used with direct electron fields for the posterior neck and a direct anterior neck photon field (as described earlier). r The cervical spine is straight. r Use parallel-opposed lateral photon beams (6 MV) or anterior-oblique wedged fields. Superiorly. T2 tumours. the central shielding will need to be omitted. If the patient is obese or has a short neck. also reduces the extent of the lateral neck skin reaction. A lower anterior neck field may need to be added. The typical beam energy is 9 to 12 MeV. even for stage N0 clinically. This technique is more difficult if the volume is not parallel to the couch top. refer to the patient’s physics plan and measure the depth of the spinal cord from the skin surface (or measure from CT/MRI images). both sides of the neck should be irradiated. treat the primary tumour. If possible. Use parallel-opposed lateral fields in a two-phase technique to stay within spinal cord tolerance. r Orthogonal films or planning CT are performed. Subglottis Patients often present with late-stage disease. r The patient lies supine.

r Floor of mouth. common fractionation schedules are r 55 Gy in 20 fractions over 4 weeks.Head and neck Subglottis: radiotherapy The target volume includes the primary site. For Tstaging. Prognosis Five-year relative survival for squamous carcinoma of the larynx is shown in Table 8. Cross-sectional imaging may underestimate the extent of gingival and hard palate involvement. r 50 Gy in 25 fractions over 5 weeks as elective nodal irradiation.3. Posterior neck electron fields can be prescribed as given doses or as the dose at dmax . A combination of CT.5 weeks. MRI. r Examining for trismus. Primary radiotherapy doses are as follows: r For early glottis. r Retromolar trigone. Carcinoma of the oral cavity Anatomy of the oral cavity Subsites include r Lip (some classify this area as skin). r Examining range of tongue movement and extent of tongue protrusion.5 weeks to sites of clinical disease. Radiotherapy dose for squamous carcinoma of the larynx Lateral photon fields should be prescribed to the ICRU 50 reference point. r 60 Gy in 30 fractions over 6 weeks to residual disease and high-risk sites. Five-year survival for carcinoma of the larynx Stage 1 2 3 4 Adapted from AJCC (2002). MRI images taken in the follow-up period may show non-specific high signal in postoperative or postradiotherapy areas that may mimic recurrence. r 66 Gy in 30 to 33 fractions over 6 to 6. These findings may directly affect the surgical approach with respect to suitability for partial glossectomy. r For other sites and stages. r Hard palate. Early mandibular invasion may be difficult to detect clinically and radiologically. r Cervical node palpation. Investigation and staging Tumour assessment involves: r Direct and bimanual palpation: tumour size. MRI is more accurate for small primary lesions and when dental amalgam is present because of the artefacts caused by the amalgam on CT. Problems arise with the changing body contour (superior–inferior) and beam obstruction by the shoulders.and paratracheal nodes. It is important to have a speech and language therapist available for rehabilitation and to support surgical voice restoration. r Anterior two-thirds of tongue. An anterior neck photon field can be prescribed as a given dose on the skin surface or at a depth of 2 to 3 cm. r 60 Gy in 25 fractions over 5 weeks. taking contour changes and shoulder position into account. 3D conformal planning is likely to give the best tumour coverage and it allows beam selection and appropriate shielding to be added.3. Table 8. 103 Follow-up Most recurrences happen within 2 years. r For postoperative radiotherapy. MRI of the tongue may show whether the ipsilateral neurovascular bundle is involved or if there is crossing of the midline and submucosal involvement in adjacent areas. whereas others use an isocentric technique. r Upper and lower alveolus and gingiva. r 64 to 66 Gy in 30 to 33 fractions over 6 to 6. . depth of invasion and extension to other subsites. r Buccal mucosa. r 50 Gy in 25 fractions over 5 weeks as elective nodal irradiation. lower deep cervical nodes and superior mediastinum. more than 90% of cases occur on the lower lip. Relative survival for larynx (%) 80 66 55 37 Set-up Some centres use a 100 cm FSD for orthogonally planned volumes. pre.

Suitable buccal mucosa tumours can often be encompassed by a single plane of plastic tube implants. Mohs’ micrographic surgery and transoral laser excision. Temporary tracheostomy may be required if the airway could be compromised because of perioperative bleeding or oedema. T3 or T4 oral cancers r Combination therapy is used. The choice of flap depends on the type and size of defect. Adapted from UICC (2002). inferior alveolar nerve. deep/extrinsic muscle of tongue. skin Through cortical bone.to two-thirds of the lower lip. close primarily.) Brachytherapy is usually only used for small superficial tumours with less than a 10% risk of lymph node metastases. floor of mouth. single ≤ 3 cm Ipsilateral single > 3 to 6 cm Ipsilateral multiple ≤ 6 cm Bilateral. r Surgery is preferred if the periosteum is involved. and plain film radiography may be needed to assess for mandibular invasion. Radiotherapy is used when patients are medically inoperable or when the expected functional or cosmetic outcome is deemed unsatisfactory. Clinically node-negative oral cancer The incidence of occult lymph node metastases is 30 to 35%. Lip cancer Of T1 and T2 lesions. pterygoid plates. internal carotid artery N1 N2a N2b N2c N3 Ipsilateral.4. skull base. Other techniques in use (but which lack evidence) include photodynamic therapy. frozen section evaluation of margins is advisable to ensure adequate resection. r For one. Larger T2. brachytherapy or a combination of the two.Laura Moss and Chris Gaffney Table 8. RT techniques include EBRT and interstitial RT. Tongue lesions may metastasise to level II and III nodes rather than level I. Radiotherapy techniques include external beam (kilovoltage or electron) radiotherapy. if functional compromise is likely. Treatment overview for oral cavity cancer Early stage T1 or small T2 oral cancer Radiotherapy and surgery are equally effective but functional outcome may be better after RT. r More than two-thirds of the lower lip usually requires a microvascular free flap tissue transfer or a regional pedicled flap to reconstruct. Small tumours of the tongue and floor of the mouth may be suitable for a hairpin implant technique whereas thicker tumours would need the plastic loop technique because the separation between the limbs of the hairpin is only 12 mm. maxillary sinus. contralateral ≤ 6 cm > 6 cm With regard to surgery. the impact of the type of reconstruction on function. and the donor site morbidity. Tumour and nodal TNM classification of oral cavity cancer Stage T1 T2 T3 T4a (lip) T4a (oral cavity) Description ≤ 2 cm > 2 to 4 cm > 4 cm Through cortical bone. Primary closure of the surgical defect is the preferred method but. use local flap reconstruction.4. Surgical excision is generally the preferred option provided the commissures are not involved and the functional and cosmetic outcome is acceptable. skin T4b Masticator space. Staging classification The TNM summary for carcinomas of the oral cavity is shown in Table 8. r RT is preferred if the oral commissures are involved. . (Avoid brachytherapy if there is involvement of the tongue tip or the adjacent periosteum. A frozen section is used to assess margins: r For one-third or less of the lip. Gingival shielding helps to reduce mucositis. local flap or free flap reconstruction may be needed. Level IV nodes are also at significant risk. Isotope bone scanning is often not helpful for detecting mandibular invasion. 104 Radical radiotherapy for oral cavity carcinoma Position and immobilisation r The patient should lie supine. 5 to 10% are node positive.

Head and neck r The cervical spine should be kept straight. For early disease. treat bilateral nodes using opposed lateral fields. It is possible for the first site of nodal metastases to arise in the junctional node between levels III and IV. Some tumours may be suitable for interstitial brachytherapy. add an ipsilateral anterior neck field to treat the rest of the cervical node levels. the target volume is the primary disease and bilateral cervical nodes. submental and submandibular). Tongue TNM staging does not take tumour depth into account. and anterior submandibular triangle).5. Five-year relative survival for oral cavity and lip cancers Stage 1 Lip survival (%) 83 73 62 47 Oral cavity survival (%) 68 53 41 27 Lip For stage T1 to T2. TI and T2 control rates are comparable with surgery or RT. for a deeper tumour. The anterior field will. treat all nodes. three or more sources may be needed. Prognosis for oral cavity cancer The 5-year survival is shown in Table 8. Aim to spare the contralateral oral cavity mucosa. however. RT may be preferable because surgery could result in functional morbidity. r Two anterior wedged fields spare the oral mucosa and parotids more than opposed lateral fields. A template technique using rigid needles implanted along the horizontal axis of the lip is used. r There should be a 2 cm margin around the primary tumour in all directions (i. For a large primary. N0. If the primary tumour approaches the midline or if there is nodepositive disease.e. the target volume is the primary plus a 2 cm margin. r Use surgery or RT (EBRT or brachytherapy). result in an acute lip reaction. r Any occult nodes are usually contained in the first echelon nodes (i. 2 3 4 Adapted from AJCC (2002). r Use of a tongue bite keeps lips apart. Buccal mucosa The target volume is the primary tumour and ipsilateral submandibular lymph nodes. Lower alveolus The target volume includes the ipsilateral mandible. Treat with electrons or orthovoltage and use a lead cut-out and shielding intraorally. The target volume includes the primary tumour with a 2 cm margin and ipsilateral submandibular and upper deep cervical nodes if there is well-lateralised disease (because spread to ipsilateral nodes is common but spread to contralateral nodes uncommon).5. A single plane may be used for superficial tumours but. Retromolar trigone The target volume includes the primary tumour and ipsilateral upper deep cervical nodes. Wedged anterior and lateral fields are used. depresses and reduces the mobility of the tongue and can spare mucosal reactions. the parotid. anterior lower alveolus. Floor of mouth For anterior T1 and small T2 lesions. Table 8. the target volume includes the anterior floor of mouth. A tongue bite is often poorly tolerated when oral mucositis has developed. If an early tumour extends on to the ventral aspect of the oral tongue. r It may be possible to irradiate a small anterior tumour with fields lying anterior to the parotids. If lymph nodes are clinically positive. Elective nodal radiotherapy is not indicated because there is a very low risk of occult nodal metastases. If nodes are positive. 105 . including the pterygoid fossa. A lateral wedged and oblique–anterior field arrangement allows maximum sparing of the oral mucosa and lip. The beam arrangement is usually an anterior and lateral wedged field.e. submental and submandibular lymph nodes. and the spinal cord. oral tongue. thereby reducing the risk of xerostomia. the target volume includes the primary and local nodes. For other sites and larger T2 to T4 tumours. Anterior and posterior oblique wedged fields spare the contralateral parotid and mucosa.

Table 8. If the mandible needs to be resected. Sixty percent of patients present with involved lymph nodes but contralateral spread is relatively uncommon unless the tumour involves or crosses the midline. N1 N2a N2b N2c N3 Ipsilateral single ≤ 3 cm Ipsilateral single > 3 to 6 cm Ipsilateral multiple ≤ 6 cm Bilateral. deep/extrinsic muscle of tongue. 60% of oropharyngeal cases). r Impaired tongue movement. T3 or T4 lesions usually require resection with reconstruction of the resulting defect. medial pterygoid. CT is better for imaging cortical bone. Reconstruction of the defect is usually by means of a radial forearm free flap or pectoralis major myocutaneous flap. mandibulotomy is often required. . 106 Surgical treatments Tonsil For the tonsil.Laura Moss and Chris Gaffney Carcinoma of the oropharynx Anatomy of the oropharynx The oropharynx extends from the palate to hyoid. r Referred otalgia. Gadolinium enhancement may demonstrate invasion along nerves. Staging classification Table 8. r Posterior pharyngeal wall. assess local extent for staging and determine whether it crosses the midline and whether there is any tongue base involvement because these two factors are important when considering surgical resection and likelihood of contralateral lymph node involvement. skull base. Oropharyngeal tumours have a higher incidence of distant metastases and it is therefore important to assess for sites of distant disease. r Painful swallowing. contralateral ≤ 6 cm > 6 cm Adapted from UICC (2002). mirror examination. mandible Lateral pterygoid muscle. Tumour and nodal TNM summary of staging for oropharyngeal carcinomas Stage T1 T2 T3 T4a T4b Description ≤ 2 cm > 2 to 4 cm > 4 cm Larynx. carotid artery Clinical presentation Patients present with the following: r Sore throat. r Altered speech (nasal quality when tongue and soft palate are not moving normally). The subsites include: r Tonsil (fossae and pillars. MRI obtains superior discrimination of tumour from muscle and other soft tissue and it is therefore often better for staging the primary tumour. and flexible endoscopy to assess the posterior and inferior extent.6. pterygoid plates. T1 or T2 tumours can be treated by transoral resection. then the risk of bilateral or contralateral lymph nodes is 15%. Investigation and staging It is usually possible to visualise the disease clinically by using a combination of direct inspection. To allow adequate exposure of the tumour site. For EUA and biopsy. Tongue base T1 disease can often be resected with little morbidity but more advanced disease is difficult to manage surgically because of the significant functional morbidity that results.6 shows the TNM classification of oropharygeal cancers. hard palate. including protrusion. r Trismus. r Neck mass. lateral nasopharynx. r Inferior surface of soft palate (10%) and uvula. it is important to try and preserve the rim because this results in a far better functional result than a segmental resection. r Base of tongue (25%) and vallecula. r Altered sensation over the lateral tongue due to involvement of the lingual nerve in the infratemporal fossa. r Impaired sensation over the anterior chin due to involvement of the inferior alveolar nerve.

Radiation fields for postoperative radiotherapy for a stage T3 N2b carcinoma of the tonsil treated with parallel-opposed lateral fields. (a) Tonsil Tonsillar fossa tumours present with more advanced stages than those of the tonsillar pillar or soft palate. 5% for pillar. then involvement of nodal levels I and V is rare (approximately 1%). for example. then the GTV and margins are increased and opposed lateral portals Radical radiotherapy for oropharynx cancers Except for well-lateralised tumours of the tonsil. field on the simulator or by CT planning for a PTV to include the primary site and nodal groups in continuity. 107 . Small (1 to 2 cm) tumours of the anterior pillar or tonsil with limited spread to the adjacent soft palate or retromolar trigone have a small risk of spread to ipsilateral lymph nodes and almost no risk of contralateral nodal metastases. If the tumour is clinically node negative. If malignant lymphadenopathy is apparent upon clinical examination. would require a total glossectomy and. then the risk of level I nodes being involved is approximately 12% (level V nodes. The anterior pillar is approximately 4 cm deep to the external surface of the face in most patients. A wedged-pair technique allows relative sparing of the contralateral parotid gland. a laryngectomy is usually needed as well. the firststation nodes (levels II to IV) are irradiated because of the high risk of microscopic involvement. (a) Phase 1: note how the fields overlap the spinal cord. If the tonsillar mass extends medially to involve the soft palate and approaches the uvula or extends inferiorly to infiltrate the tongue base. 10%). An electron field was matched posteriorly on each side to ensure coverage of the upper posterior neck nodes. (b) Phase 2: note how the posterior borders of the fields are anterior to the spinal cord. The target volume for node-negative T1 or T2 welllateralised primary tumours is the primary tumour plus 2 cm margin and first echelon ipsilateral nodes. Soft palate Tumours tend to arise on the free edge and present relatively early (T1 or T2). most situations require the use of parallel-opposed beams. If there is node-positive lateralised T1 or T2 disease.1.Head and neck Large tumours. If there is no evidence of nodal disease upon clinical examination. one would usually include all ipsilateral neck nodes in the target volume by adding an ipsilateral anterior neck (b) Figure 8. The overall risk of lymph node involvement is high: 60 to 75% for fossa tumours and 45% for pillar tumours. The risk of contralateral nodes is generally low: 10% for fossa. The lower cervical nodes were treated with an anterior photon field (not shown). Resection of these can be effective. but for larger tumours problems arise with surgical treatment because of the functional morbidity associated with reconstruction. Tumours in the posterior tonsillar pillar are more likely to metastasise to the spinal accessory and upper posterior triangle nodes. because of the significant risk of aspiration following this procedure.

Opposed lateral fields were used. which is bounded superiorly by the floor of the vallecula and inferiorly by the inferior border of the cricoid cartilage. treat levels I to IV. The medial extent includes the aryepiglottic folds and the arytenoid and cricoid cartilages. r Glossolaryngectomy and postoperative RT if there is a midline tumour (high functional morbidity). (Surgery may result in problems with nasal regurgitation. NB disruption of normal lymphatic channels by the presence of a tumour or surgery may result in aberrant patterns of spread to levels I and V or to contralateral nodes. The choice of treatment modality is usually based on likely morbidity and RT is often favoured. Carcinoma of the hypopharynx Anatomy of the hypopharynx The hypopharynx extends postero-laterally in relation to the larynx from the aryepiglottic fold at the level of the hyoid to the lower level of the cricoid cartilage. Base of tongue The lymphatics course downward towards the hyoid bone where they pierce the pharyngeal wall and drain to level II. followed by levels III and IV. For T1 cancers. treat with surgery and RT. Local control rates are 80 to 90%. Clinical presentation Carcinoma of the hypopharynx usually presents late. which extends from the level of the arytenoid cartilages and connecting folds to the inferior border of cricoid cartilage. r Chemoradiotherapy Even if there is a clinically node-negative tumour. equivalent local control and survival rates are obtained for surgery or RT. r The postcricoid. Posterior pharyngeal wall First-echelon nodes are the retropharyngeal nodes and nodes at levels II and III.7 shows the 5-year survival of oropharygeal cancer. r Excess salivation. r The pyriform fossae. The subsites are: r The posterior pharyngeal wall. Prognosis Table 8. treat with surgery or RT. Figure 8. With RT alone the local control rate is approximately 30 to 50%. 108 Investigation and staging With indirect laryngoscopy. The risk of LN metastases is 25% for T1 and greater than 75% for T4. Survival – all subsites combined (%) 57 54 43 30 Soft palate For T1 or T2 cancers. r Referred otalgia. Symptoms include: r Neck mass. it may be possible to visualise tumours arising in the pyriform fossae and .Laura Moss and Chris Gaffney are used. r Sensation of something in the throat. r Brachytherapy to the primary and/or EBRT and surgery to the neck. r Frequent coughing bouts due to inefficient closure of the glottis during swallowing. r Pain with swallowing. For T1 or T2 cancers. T3 and T4 tumours the treatment options are as follow: r EBRT and/or brachytherapy. Five-year relative survival of oropharyngeal cancer Stage 1 2 3 4 Adapted from AJCC (2002). which extend from the pharyngoepiglottic fold to the upper end of the oesophagus at the lower border of the cricoid cartilage. consider elective neck RT (NB propensity for bilateral spread). The lateral border is the inner surface of the thyroid cartilage.1 shows the radiation fields for a postoperative radiotherapy treatment for a T3 N2b carcinoma of the tonsil. RT often results in a better functional result.7. Table 8.) For T3 or T4 cancers. The jugulo-digastric node (the largest of the level II nodes) is frequently the first node to be involved. For large T2. For T3 or T4 cancers. r Voice changes (muffled and husky rather than hoarse). treat levels I to III.

unlike lymphadenopathy. Stages T2 to T4 Optimal local control probably results from surgery and postoperative radiotherapy. It is difficult to predict postoperative speech and swallowing function. mediastinal LN 35%. Therefore. extranodal spread. there is a higher risk of distant metastases. extension to the oesophagus.g. anterior displacement of the larynx or thyroid cartilage destruction. arytenoid. Half the larynx and half the hypopharynx are resected and the hyoid bone. If there is palpable neck disease. central compartment soft tissue T4b Prevertebral fascia. Occult lymph node metastases are found in 40% of patients after neck dissection for cN0 disease. saliva pooling or vocal cord oedema. r Radical radiotherapy. quality of life. thyroid gland. However. or more than three lymph nodes. Tumour and nodal TNM classification for carcinomas of the hypopharynx Stage T1 T2 T3 T4a Description ≤ 2 cm and limited to one subsite > 2 to 4 cm or more than one subsite > 4 cm or with hemilarynx fixation Thyroid/cricoid cartilage. CT is better at detecting the presence of thyroid cartilage involvement. Lateral soft tissue plain film may demonstrate widening of the prevertebral space. There is not enough research evidence to make valid comparisons among these options. Surgery for T3 or T4 disease typically requires laryngectomy as part of the procedure. whereas MRI shows soft tissue detail. If there is involvement of the postcricoid region or the apex of the pyriform fossa then level VI nodes are 109 Stage T1 Stage T1 is uncommon. which is indicated for early stage (T1 or T2) tumours in patients who can tolerate some degree of chronic aspiration. thyroid ala. r Conservation surgery (e. N0 disease If the patient is being treated surgically then selective lymph node dissection is performed to include levels II to IV. liver or bone imaging. hyoid bone. it may be possible to see displacement of the larynx anteriorly. Two-thirds of patients are lymph node positive at presentation (especially levels II to IV.Head and neck posterior pharyngeal wall but it is difficult to see the postcricoid area. epiglottis. and bone 30%). contralateral ≤ 6 cm > 6 cm Adapted from UICC (2002). Staging classification The TNM classification for carcinomas of the hypopharynx is shown in Table 8. Table 8. Microlaryngoscopy/oesophagoscopy/bronchoscopy and biopsy are essential to assess the local extent. functional outcome). endoscopic pharyngectomy if a small lesion is easily accessible through the operating laryngoscope). etc. r Partial pharyngolaryngectomy. aryepiglottic fold. levels I and V are rarely involved but level VI may be involved if there is apical pyriform fossa disease or a postcricoid primary). carotid artery. arytenoid eminence and false cord are removed on the affected side. consider lung. mediastinal structures N1 N2a N2b N2c N3 Ipsilateral single ≤ 3 cm Ipsilateral single > 3 to 6 cm Ipsilateral multiple ≤ 6 cm Bilateral. Treatment overview of hypopharyngeal carcinoma There are few comparative data on different treatment modalities and it is often difficult to interpret outcome as the reports use different outcome measures (survival. Neck palpation should be used for cervical nodes and to assess laryngeal crepitus (absence of crepitus signifies laryngeal involvement). oesophagus.). Cancer of the hypopharynx has the highest incidence of distant metastases relative to other head and neck sites (lung 80%. Swelling as a result of direct extension of the primary disease into neck tissue may be palpable and it will move with swallowing. Relative contraindications to surgery include invasion of the prevertebral fascia and carotid involvement. liver 30%. Treatment options include: .8. especially inferiorly (invasion of the trachea.8.

which may be achieved by a floor twist. With orthogonal film planning.Laura Moss and Chris Gaffney also removed. a coronal technique is often needed to treat the inferior aspect of PTV (the 110 . The inferior wall consists of the superior surface of the soft palate. non-keratinising carcinoma. It may be necessary to add an anterior mediastinal field. r The shoulders may obstruct beam entry. r Type 3. associated with a lymphoid infiltrate. If there is a central tumour then a bilateral neck dissection is recommended. plasmacytoma. Other tumour types include lymphoma. Carcinoma of the nasopharynx Types of tumour affecting the nasopharynx The WHO classification divides the commonest tumours into three types (Shanmugaratnam and Sobin. Potential problems r Bulky nodes may overlie the spinal cord. Radical radiotherapy for hypopharyngeal carcinoma Position and immobilisation r An immobilisation/beam direction shell is required. and adenocarcinoma. nodes adjacent to the skull base and nodes in the submastoid region. The lateral wall includes the fossa of Rosenmuller. Table 8. Prognosis for cancer of the hypopharynx The 5-year survival for cancers of the hypophaynx is shown in Table 8. r There may be involvement of mediastinal nodes. cervical oesophagus and mediastinum) and to avoid the shoulders. With 3D conformal planning there is greater ability to cover target volume adequately because non-coplanar planning is possible. If there is oropharyngeal extension then level I nodes are also removed. In these circumstances. r It may not be possible to treat with a radical dose if there is large-volume bulky disease. unless there is a pyriform fossa primary with a large neck mass. In practice this means that the beams need to be angled inferiorly. Partial pharyngectomy with or without partial laryngectomy may be suitable in selected cases. r Type 2. Primary radical radiotherapy with concurrent chemotherapy where possible is indicated for early lesions and in medically inoperable cases. undifferentiated. Target volume Add a 2 cm margin to the GTV in the superior–inferior direction to allow for subclinical disease. Significant contour changes are more easily appreciated and more complex beam shielding is possible. r The cervical spine is normally straight. Most tumours occur in the lateral wall or roof. More advanced disease requires pharyngolaryngectomy with reconstructon of the resulting defect with a free jejunal transfer or gastric pull-up. Survival – all subsites combined (%) 41 36 36 20 Surgical treatments Endoscopic resection of the early posterior wall and pyriform fossa tumours can result in low morbidity and good functional results. Five-year relative survival for cancers of the hypopharynx Stage 1 2 3 4 Adapted from AJCC (2002). r The patient lies supine.9. The DVH and beam’s-eye-view facilities also help planning and verification.9. a large volume needs to be treated and there will be a significant contour change in the superior–inferior direction as well as in the anterior–posterior direction. 1978): r Type 1. If the treatment is for early disease then the first-station nodes are included. keratinising squamous cell carcinoma. Anatomy of the nasopharynx The postero-superior wall extends from the level of the junction of the hard and soft palate to the base of the ¨ skull. in which case it may be preferable to have the chin extended in order to move the oral cavity and mandible out of the field. hence the term lymphoepithelioma.

consider PEG placement before starting treatment. mediastinum. masticator space N1 N2 N3a N3b Unilateral node(s) ≤ 6 cm.4 per 100 000. Tumour and nodal TNM classification for nasopharyngeal carcinomas Stage T1 T2a Description Limited to nasopharynx Soft tissue extension to oropharynx/nasal cavity without parapharyngeal extension T2b T3 T4 Tumour with parapharyngeal extension Bony structures. including diet. therefore. The presence of deep tumour infiltration appears to be of more prognostic significance than the volume of the tumour. r Conductive deafness. Clinical presentation Symptoms are rarely related directly to the nasopharynx itself and are more likely to be due to local or lymph node spread of the tumour: r Nasal blockage. If the patient has presented only with cervical lymphadenopathy and there is no obvious nasopharyngeal lesion. r The hearing should be examined for conductive deafness. tinnitus. MRI is more accurate for evaluating Treatment overview for nasopharyngeal carcinoma Surgery Apart from obtaining tissue for histological verification. infratemporal fossa. Imaging of lung.10. CT is useful for assessing the skull base for cortical bone erosion whereas MRI can often demonstrate involvement of the marrow space. Bilateral nodes do not appear to influence prognosis.10 shows the TNM classification for carcinoma of the nasopharynx. Table 8. bone and liver should be considered to assess sites of distant metastases. A US of neck nodes with FNAC should be performed. above supraclavicular fossa Bilateral node(s) ≤ 6 cm. r Cranial nerve palsies. above supraclavicular fossa > 6 cm In supraclavicular fossa Risk factors and aetiology There are multiple risk factors. Diagnosis and staging r The nose should be examined by anterior rhinoscopy and the postnasal space should be examined by posterior rhinoscopy or via nasendoscopy. then blind biopsies should be taken from the nasopharynx. surgery has no place in the initial management of nasopharyngeal carcinoma. CT or MRI should be used to assess the extent of primary disease. In endemic areas. recognised risk factors are the high intake of salt-cured fish and meat with the associated release of volatile nitrosamines during the cooking process and EBV. r The tympanic membrane should be assessed for middle ear aeration which. cranial nerves. r Cervical lymphadenopathy (often in the posterior triangle). hypopharynx. The incidence in southern China is approximately 50 times higher. viral agents and genetic predisposition. could indicate eustachian tube dysfunction or obstruction. the primary. Surgery may be 111 . orbit. Investigations Investigation involves direct endoscopy with biopsy under anaesthetic. paranasal sinuses Intracranial. Adapted from UICC (2002). if abnormal. r The cranial nerves should be assessed.Head and neck Incidence and epidemiology Carcinoma of the nasopharynx is relatively rare in the Western world. The first-echelon nodes are the parapharyngeal and retropharyngeal nodes. Staging classification Table 8. the UK incidence is 0. r 80% of patients present with palpable cervical lymph nodes: 50% are bilateral. Radical radiotherapy treatment results in significant mucosal toxicity and enteral feeding is often required.

the cervical nodes and the SCF. The technique is discussed later. Radiotherapy Radiotherapy is the primary treatment of choice. General issues in radiotherapy planning r The primary site and bilateral neck need to be treated because there are extensive lymphatics and these cross the midline. 1998). Prechemotherapy investigations are a baseline assessment of GFR. but intracavitary boosts can be used in specialist centres. r The field margins are individually tailored to take into account the primary location. A decision needs to be made whether to use orthogonal films or CT for planning. Advanced disease may extend into cranial fossae. This may result in cold spots over areas of potential disease and hot spots over critical structures. liver and bone profiles. days 1. the pterygoid fossa. and the parapharyngeal space. the floor of the middle cranial fossa. 22 and 43). planning is used.Laura Moss and Chris Gaffney considered in patients who have partial responses in cervical nodes. The extent of oropharyngeal and nasal cavity coverage depends on the individual case. Patients need to be of good performance status and have good renal function. It is necessary to decide whether to treat the neck nodes with anterior and posterior fields or to treat with an anterior field only. r It is impossible to treat an entire volume in continuity r There is no ideal technique. nasopharyngeal carcinoma is usually treated with EBRT. Commonly used chemotherapy strategies include the following: r Two cycles of neoadjuvant cisplatin and 5-fluorouracil (5-FU) prior to radical EBRT. such as an additional 10 Gy boost prescribed to a depth of 1 cm. the retropharyngeal nodes. Positioning r The patient lies supine. the size of the tumour. orbits and normal tissue tolerances may be compromised. the sphenoidal and posterior ethmoidal sinuses. Chemotherapy Chemotherapy is commonly used. Problems in radiotherapy planning to a radical dose in a single phase. days 1. nodal relapse or local recurrence at the primary site. renal. for both the primary site and neck nodes. r Concurrent chemoradiotherapy only (cisplatin 100 mg/m2 . To cover potential microscopic disease. This often achieves a good partial/complete clinical response after two neoadjuvant cycles. EBRT may then be given with concurrent cisplatin (100 mg/m2 . r The neck is extended but the cervical spine is kept r Cervical nodes are marked with wire if orthogonal r It may be useful to mark the lateral orbital margin with r The target volume is the primary disease with a 2 to 3 cm margin. It is important that the catheters are positioned up against the tumour in the vault of the nasopharynx because there is a tendency for the catheters to fall downward. straight. Relapse in the neck nodes without a recurrence at the primary site is rare. the volume must include the skull base. but there is conflicting evidence and there are conflicting opinions and techniques. 22 and 43). This decision will depend on the levels of involved nodes and their bulk. the parapharyngeal space. 112 . r r r r r Radical radiotherapy for carcinoma of the nasopharynx In the UK. r Concurrent cisplatin chemoradiotherapy followed by 3 cycles of adjuvant cisplatin and 5-FU at 28-day intervals (Al-Sarraf et al. The fields covering the primary and neck nodes overlap and there are difficulties matching photon and electron fields. the posterior half of the nasal cavity and orbit. the skull base and the retropharyngeal and cervical lymph nodes bilaterally. There is a large target volume with extensive mucosal coverage. the clivus. and any nodal involvement. the floor of sphenoid sinus. Phase 1 r Large opposing lateral photon fields are used to cover the entire target volume. and FBC. r The volume needs to encompass the nasopharynx. The concerns relate to increased morbidity (especially mucositis) and some positive trials that have shown a benefit to chemotherapy have had poor outcomes in the standard RT arm. audiogram.. usually with chemotherapy. wire for orthogonal planning.

ethmoid sinus and sphenoid sinus. These tumours arise most commonly in the maxillary sinus.11 shows the 5-year relative survival for cancers of the nasopharynx. Symptoms are: 113 . including nasopharynx. Anterior and posterior opposed photon fields.) For subsequent phases (usually two. iii.5 weeks. hence. Carcinoma of the maxillary antrum (and other paranasal sinuses) Anatomy of the maxillary sinus The maxillary antrum is related superiorly to the floor of the orbit.11. 2. NB. ii. the degree of nodal involvement and the position of the nodes must be taken into account. followed by the nasal cavity. The primary tumour. The floor of the antrum is the alveolar process of the maxilla and the hard palate. optic chiasm and the anterior third of the orbit. parapharyngeal space. This line is used to divide the maxillary antrum into the antero-inferior portion (which is associated with earlier presentation and good prognosis) and the supero-posterior portion (which is associated with early involvement of adjacent critical structures and a worse prognosis). The separate contributions from phase 1 and phase 2 differ among centres and the local protocol should be followed. Adenoid cystic carcinoma can demonstrate widespread perineural infiltration. skull base. r The lower cervical lymph nodes and SCF nodes are usually treated with a direct anterior field. When deciding on the optimal technique to treat cervical node disease. Options include: i. posterior third of the orbit.or three-phase techniques).Head and neck r Shield the brain stem. Opposed lateral photon beams with electrons to boost the posterior nodal groups so as to remain within spinal cord tolerance. and any extension into the nasal cavity or oropharynx. Clinical responses tend to be rapid but there is potential for rapid relapse when treatment stops. reduce the acute mucosal reactions at these sites by having a central shielding strip in the anterior beam. Palliative treatments For metastatic disease cisplatin and 5-FU may provide significant symptom palliation. Five-year relative survival for nasopharynx cancer Stage 1 2 3 4 Adapted from AJCC (2002). TLDs should be requested to estimate dose to lens. ¨ Ohngrens line connects the medial canthus of the eye to the angle of the mandible. (The optic nerves exit the orbits just beneath the anterior clinoids and the brain stem lies immediately posterior to the clivus). Anterior photon field only. r Take care not to have the junction of the lateral and anterior photon fields over palpable/radiologically visible disease. Clinical presentation Carcinoma of the maxillary antrum often presents at a locally advanced stage. Request TLDs. and the anterior wall lies deep to the cheek. r By placing the junction between the upper lateral and the lower anterior fields at the level of the hyoid. The neck nodes. Table 8. Adenocarcinomas are associated with hardwood dust exposure. it may be possible to spare much of the larynx and pharynx and. (Some argue that it is acceptable to have a junction through enlarged nodes because the dose at the junction of two diverging photon portals is more likely to be hot than cold. the volume can be divided into two as follow: 1. The medial wall separates it from the nasal cavity. Dose is 66 Gy in 33 fractions over 6. The pterygoid plates and pterygopalatine fossa lie posteriorly. It may be necessary to add an additional anterior photon field to give adequate dose coverage to any nasal cavity disease extension. Pathology Squamous carcinoma is the most common tumour type and is associated with nickel exposure in some cases. Survival (%) 63 52 56 39 Prognosis Table 8.

dura. r Poorly fitting dentures or loose dentition. A contralateral lateral field is used to top up the dose to the postero-medial aspect of the volume. Figure 8. biopsy is associated with low morbidity. Radical radiotherapy for carcinoma of the maxillary sinus Radiotherapy technique An immobilisation shell should be used. pterygoid plates. It is important to avoid the optic chiasm and hypothalamus and beam exit through the contralateral eye. middle cranial fossa. The complex shape of the target volume and surrounding critical structures makes 3D conformal RT desirable. which will allow excellent visualisation and minimum alteration to the tumour and surrounding anatomy. pterygoid fossa and cheek. cranial nerves other than V2. Inflamed mucosa and retained secretions have a similar appearance on CT and can be mistaken for tumours. middle nasal meatus Posterior bony wall maxillary sinus.g. infratemporal fossa. need to be combined with orbital exenteration or craniofacial resection depending on the extent of disease. MRI is better than CT at distinguishing tumours from retained secretions and surrounding soft tissues.12. ethmoid sinus T4a Anterior orbit. total or extended). Table 8. floor/medial wall of orbit. The use of a mouth bite to depress the tongue should be considered. subcutaneous tissues. The resulting defect needs to be filled with an obturator or a free flap. Approximately 15% of patients are LN positive on presentation. contralateral ≤ 6 cm > 6 cm Investigation and staging Clinical examination involves examining the gingivobuccal gutter and palate. r Impaired vision (due to abnormal ocular movement or proptosis). hard palate. soft tissue intracranial extension or perineural spread. sphenoid/frontal sinus T4b Orbital apex. Tumour and nodal TNM summary for carcinomas of the maxillary sinus Stage T1 T2 T3 Description Confined to mucosa Bone erosion/destruction. nasal cavity. r Trismus (due to pterygoid fossa involvement from posterior extension of the disease). Adapted from UICC (2002). Biopsy should be performed via the least-invasive method (e. Staging classification Table 8. It is also helpful in assessing the presence of skull base invasion.Laura Moss and Chris Gaffney r Epistaxis or nasal blockage. combined modality treatment is usually preferred if the patient is of good performance status with limited co-morbidity and can manage with the functional alterations that result from surgical resection. pterygoid fossa. r Cheek swelling. Anterior and lateral wedged fields are often used.12 shows the TNM staging for carcinoma of the maxillary sinus.2 shows a radiotherapy plan for a locally advanced carcinoma of Treatment overview Because this disease typically presents at a locally advanced stage. cribriform plate. examining the nasal cavity and examining the range of jaw movement. clivus N1 N2a N2b N2c N3 Ipsilateral single ≤ 3 cm Ipsilateral single > 3 to 6 cm Ipsilateral multiple ≤ 6 cm Bilateral. subtotal. brain. Maxillectomy may 114 . nasopharynx. looking for cheek swelling and infraorbital nerve involvement. cheek skin. hard palate. The target volume includes the ethmoid sinuses. CT scanning is good for assessing bone detail and determining whether the cancer is confined within the boundaries of the sinus or is eroding through bone. endoscopic biopsy via the nose). gingivobuccal sinus. r Facial or cheek pain. as well as the maxillary antrum. Surgical treatments There are a variety of different maxillectomy operations (medial.

Investigation and staging the maxillary antrum in which 3D conformal planning has been used.13.2. r Local pain. Staging classification Prognosis The 5-year survival for maxillary sinus cancers is shown in Table 8. It is divided in the midline by the septum. Figure 8. such as the eyes. r Sinusitis. This image shows the isodoses at one level only.13. loose teeth and poorly fitting dentures if the disease extends inferiorly. r Proptosis and diplopia if there is superior extension into the orbit. is as low as possible. r Gingival/palatal swelling. The lateral wall is also the medial wall of the maxillary sinus and includes the turbinates and nasolacrimal duct. Pathology Squamous carcinoma is most common. r Epistaxis. A CT-planned radical radiotherapy treatment for a locally advanced carcinoma of the maxillary antrum showing the anterior and two lateral beams.14. The TNM summary is shown in Table 8. Surgical treatments Lateral rhinotomy can be used to treat cancers involving the nasal septum and lateral nasal wall that extend up to. r Trismus from pterygoid muscle involvement may result from lateral extension. mucosal melanoma and olfactory neuroblastoma (a neuroendocrine tumour that arises from the superior aspect of the nasal cavity). Cross-sectional imaging is required to determine the posterior extent of disease. and it is important to study the whole plan to make sure the dose to critical structures.Head and neck Table 8. optic chiasm and hypothalamus. the anterior skull base. The . Five-year relative survival for cancers of the maxillary sinus Stage 1 2 3 4 Adapted from AJCC (2002). Symptoms are: r Nasal obstruction or discharge. Survival (%) 60 50 46 31 superior boundary is the ethmoid sinus and the inferior boundary is the hard palate. The sphenoid sinus lies posterior and superior and the nasopharynx lies directly posterior. Isodose % 95 70 50 20 Spread Lymphatic drainage is to the submandibular and upper deep cervical nodes. Others include adenocarcinoma. Clinical presentation Nasal cavity cancers often do not present until they reach a significant size. but not including. Craniofacial resection can be used to excise olfactory neuroblastoma and the procedure can be extended to 115 Carcinoma of the nose and nasal cavity Anatomy of the nasal cavity The nasal cavity is the most superior part of the upper airway.

followed by the nasal cavity and oral cavity. clivus N1 N2a N2b N2c N3 Ipsilateral single ≤ 3 cm Ipsilateral single > 3 to 6 cm Ipsilateral multiple ≤ 6 cm Bilateral.Laura Moss and Chris Gaffney Table 8. skin of nose/cheek. cranial nerves other than V2. Salivary gland tumours account for 7% of epithelial head and neck cancers. Total rhinectomy is required for extensive tumours of the cartilaginous nasal skeleton and dorsum of the nose. Radical radiotherapy for nose and nasal cavity carcinomas Radiotherapy technique The primary site alone is usually irradiated. because only 20% of these are malignant. pterygoid plates. pituitary and brain. the incidence of malignancy is highest in the submandibular (>50% malignant). an anterior-oblique wedged pair or anterior and opposed laterals will be required. The field extends to the nasal bridge. The lymphatic drainage routes are to the submandibular and upper deep cervical nodes. submandibular.) If the cancer is positioned anteriorly. Consideration should be given to using tissue equivalent bolus in the air cavity. If the cancer is positioned posteriorly in the nose. Adapted from UICC (2002). Pathology There is a wide range of histological types and clinical behaviours. rior) and mastoid process (posterior). Consideration should be given to placing bolus in the nasal cavity to reduce dose inhomogeneity. middle cranial fossa. maxillary sinus. and a prosthesis usually offers a better rehabilitation option. r Megavoltage photons may cause problems because of the dose at depth and the proximity of brain/brain stem. anterior to the styloid process. It is important not to exceed the tolerance doses of the optic nerve and chiasm. a single anterior field with shielding to the eyes (inverted ‘T’ shape) will be sufficient. cribriform plate Anterior orbit. The highest concentration is found on the hard palate. include a maxillectomy or orbital exenteration depending on the extent of the disease. It is often difficult to reconstruct the nose with a flap in a way that leads to a satisfactory cosmetic result. Tumours are most common in the parotid but. the following problems may occur: r There may be dose inhomogeneity in the air cavity with electron treatments. r Lymphatic drainage is to the preauricular and intraparotid nodes. brain. If there is surface disease then bolus may be required. midway to the lip and 1 cm lateral to the ala nasi. The minor salivary glands are small and are found beneath the mucosa of the upper aerodigestive tract.14. Malignant tumour types include the following: . r The plane separating the two lobes is formed by the facial nerve. sphenoid/frontal sinuses T4b Orbital apex. sublingual and minor glands. and adjacent to the parapharyngeal space and the lateral pterygoid muscle. The incidence is approximately 1 in 100 000. Tumour and nodal TNM classification for nasal cavity and ethmoid sinus Stage T1 T2 T3 T4a Description 1 subsite 2 subsites or adjacent nasoethmoidal site Medial wall/floor of orbit. and sublingual glands. Salivary gland Introduction The major salivary glands are the paired parotids. contralateral ≤ 6 cm > 6 cm external nose to enclose the irregular shape. (The area is not routinely irradiated if the disease is confined to the external nose because there is only an approximately 15% risk of involvement. Wax bolus is placed over the 116 Anatomy of the parotid gland r The superficial lobe lies between the masseter (anter The deep lobe is posterior to the mandibular ramus extending medially. nasopharynx. In the postoperative situation following total rhinectomy. dura. palate. anterior cranial fossa (minimal).

without extraparenchymal extension > 2 to 4 cm. The tumour location is also important. an immobile tumour. extensive perineural spread. be better treated with radical radiotherapy. pterygoid plates. r Advanced stage: skin or bone involvement. Neck dissection is indicated if there is clinical or radiological evidence of node involvement. a borderline resectable tumour in an inaccessible area may . Total parotidectomy is indicated if the deep lobe is involved. Salivary duct carcinoma. An example would be a minor salivary gland tumour in a sinus abutting the skull base. Adenocarcinoma. However. It is important to avoid open biopsy of major salivary gland tumours because of the risk of tumour spillage and seeding. without extraparenchymal extension > 4 cm and/or extraparenchymal extension Skin. pain. carotid artery Ipsilateral single ≤ 3 cm Ipsilateral single > 3 to 6 cm Ipsilateral multiple ≤ 6 cm Bilateral. trismus and nodal involvement all suggest malignancy. r Positive or close resection margins. Postoperative radiotherapy of the parotid gland Indications for postoperative radiotherapy Indications for RT include the following: r Inoperable tumours. It is important to determine whether the tumour is arising in the superficial or deep lobe. Imaging Use US and image-guided FNA. Sebaceous carcinoma. 117 Staging classification The TNM classification summary is shown in Table 8.15. Small-cell and undifferentiated carcinoma. r Residual disease. facial nerve Skull. Table 8. Mucoepidermoid (the most common major salivary gland tumour: 4 to 9%). Squamous carcinoma. MRI may be a better choice.Head and neck r Adenoid cystic carcinoma is the most common type. Lymphoma. squamous or undifferentiated. therefore. for example. its relationship with the facial nerve and whether there is any extraglandular extension. Tumour and nodal TNM classification for carcinomas of the salivary glands Stage T1 T2 T3 T4a T4b N1 N2 Description ≤ 2 cm. if the tumour is locally advanced and if the tumour type is high grade or adenocarcinoma. rapid growth. Investigation and staging Fine needle aspiration cytology is reported to be capable of distinguishing between benign and malignant tumours in 90% of cases. dental amalgam may produce artefacts and. Adapted from UICC (2002). Total parotidectomy is also indicated for adenoid cystic tumours because of the propensity for perineural spread. r Positive nodes. ear canal.15. Acinic cell carcinoma (accounts for 3% of parotid tumours). Carcinoma arising in a pleomorphic adenoma. r High-grade type. In a CT scan. it r r r r r r r r r is more common in minor glands than major salivary glands. facial nerve involvement. Treatment overview The extent of the tumour is the single most important factor affecting the choice of treatment modality. The facial nerve can be preserved if there is no direct involvement. mandible. provided there is no involvement of the deep lobe or the facial nerve. contralateral ≤ 6 cm N3 > 6 cm Clinical presentation Clinically it may be difficult to distinguish between benign and malignant tumours. Surgery Superficial parotidectomy can be performed.

It is important to weigh the risk of RT for this benign disease in young patients against the risk of recurrence and risk of late morbidity including second malignancy. r Adenoid cystic type. The parapharyngeal space should be included if irradiating a malignant tumour. The field arrangement is of anterior and posterior oblique wedged fields to treat the parotid or parotid bed in the postoperative setting. .Laura Moss and Chris Gaffney r Recurrence. 3D conformal RT has advantages over 2D planning because complex volume shapes are possible and it helps avoid surrounding critical structures.16. anteriorly into the parotid. space. r Close proximity to facial nerve where the nerve has r For pleomorphic adenoma. Five-year relative survival for major salivary glands combined Stage 1 2 3 4 Adapted from AJCC (2002). Isodose % 100 95 70 50 20 Figure 8. It contains the ossicles and semicircular canals and it communicates posteriorly with the mastoid air space. Survival (%) 86 66 53 32 Radiotherapy technique The entire parotid bed should be included. 118 r Examine the ear. and posteriorly.3. Table 8. extend to include the facial nerve route and skull base) to below the angle of the mandible. Bolus was used because in this case there was concern about tumour seeding in the skin.3 shows a radiotherapy plan for a postoperative treatment following parotidectomy for a high-grade mucoepidermoid carcinoma. extending above the zygomatic arch for malignant tumours (if adenoid cystic. been preserved. Anteriorly. mastoid area and postnasal r Examine cranial nerve function. The middle ear lies between the tympanic membrane and the inner ear.16. The outer portion is cartilaginous and the inner portion is bony. The volume marked on the contralateral side is the other parotid gland. consideration should be given to postoperative radiotherapy if there is incomplete excision or spillage at operation or after reexcision of locally recurrent tumours. The dose is administered as a single phase. Clinical examination Carcinoma of the middle ear Anatomy of the ear The external ear canal is 2. A CT-planned postoperative radiotherapy treatment following excision of a high-grade mucoepidermoid cancer of the parotid. An anterior ipsilateral lower neck field is matched onto the oblique fields if there are involved nodes or a high risk of occult lymph nodes. The beam angles are selected to avoid the contralateral parotid and spinal cord. parotid. the masseter should be included. Prognosis The 5-year survival for malignant tumours is shown in Table 8. Figure 8.5 cm long. superiorly to the middle cranial fossa. Investigation and staging r Perform examination and biopsy under general anaesthesia. Spread Spread occurs medially to the inner ear and facial nerve. posteriorly to the mastoid and laterally to the ear canal. the volume should extend to the mastoid process. r Palpate the cervical and facial nodes.

tympanic membrane. Staging classification There is no specific TNM staging system. A tonsil or tongue base tumour is found in approximately a third of cases. Cervical lymphadenopathy from an unknown primary Introduction This condition occurs when squamous carcinoma is identified in cervical lymph nodes but there is no upper aerodigestive tract primary site evident. Treatment overview Surgery and postoperative RT give the best chance of local control. r MRI is more useful than CT for defining the soft tissue extent. Radical radiotherapy for carcinoma of the middle ear Primary radical radiotherapy is seldom used as the sole treatment modality.and postauricular nodes and the mastoid process and is triangular in shape. extracapsular spread). If the pathology from the neck dissection specimen reports N1 disease. If the cervical node FNAC is positive for squamous carcinoma. The optimal timing is debated: some advocate PET imaging before EUA and blind biopsies to avoid the delay between biopsy and imaging (which is recommended to reduce the chances of a falsepositive result caused by increased uptake at biopsy sites) and to target subsequent biopsies based on the imaging findings. Investigation and staging Panendoscopy and biopsies are undertaken from any suspicious areas plus the postnasal space. the tonsils (or bilateral tonsillectomy) and the tongue base. Clinical presentation The patient presents with a neck mass. upper aerodigestive tract examination will identify the primary lesion. Staging classification No staging system is accepted by UICC or AJCC.g. Others recommend PET only if the EUA and biopsies fail to detect the primary site because of limited resource availability. CT and/or MRI should be used to look for potential primary sites. It is an ideal anatomical site for 3D conformal therapy.Head and neck r CT is the investigation of choice for assessing the temporal bone. As a result. and an EUA with targeted biopsies of potential mucosal primary sites may detect the primary lesion. 18 FDG-PET imaging may also be used to detect a primary site. In 40 to 50% of cases. then proceed with postoperative RT. In practice the neck nodes are staged in the same way as other head and neck cancer subsites (excluding the nasopharynx). It may entail lateral temporal bone resection (removal of osseous and cartilaginous external auditory canal. then observation only is required. Treatment overview for cervical lymphadenopathy of unknown primary If FNAC of the presenting neck mass is inconclusive or negative for malignancy. malleus and incus) with mastoidectomy. It accounts for 5% of head and neck cancers. postoperative radiotherapy is indicated in most cases. It is important to decide 119 Prognosis The 5-year survival is approximately 30%. Consider doing a frozen section at the time of the excision biopsy as well as panendoscopy of the upper aerodigestive tract to try and locate a primary lesion. If pathology reveals N2 or N3 disease. These findings are common because of the limitations on the margins of surgical resection. then an excision biopsy will be required. . Postoperative radiotherapy is indicated for situations where there are close or positive resection margins and perineural invasion. then a neck dissection is indicated. Surgical treatments The required procedure depends on the local extent. The volume includes the pre. unless there are adverse prognostic factors present (e. parotidectomy and neck dissection.

Barrett. et al. 25.. Geneva: World Health Organization. Engl. International Histological Classification of Tumours. M. 32. Specht. CRUK. 44. Med. (1998). Sobin and Ch. Five compared with six fractions per week of conventional radiotherapy of squamous-cell carcinoma of head and neck: DAHANCA 6 and 7 randomised controlled trial. et al. Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. R. London: Royal College Radiologists. UICC. 362. Engl. Bourhis. J. Int.. J. J. J. Oncol..).. Clin. (2000). Oncol. 1945–52. Greene et al.. Lancet. Oncol. N’Guyen.. London: NICE. J... Lancet. Radiother. C. London: CRUK. Oncol. S. Al-Sarraf. MACH-NC Collaborative Group MetaAnalysis of chemotherapy on head and neck cancer. Med. J. (2003). Bourhis. J. Ozsahin. and Sobin. Syz.. Pajak. The Role and Development of Brachytherapy Services in the UK. New York: Wiley-Liss.. Radiat. C. Domenge. Chemotherapy added to locoregional treatment for head and neck squamous-cell carcinoma: three meta-analyses of updated individual data. Effective Head and Neck Cancer Management: Third Consensus Document. ed. p. Intercollegiate Standing Committee on Nuclear Medicine. ed.. Ann. R. 350. N. F. 995–1006. M. (2004). Phys. et al. 13.. 16..Laura Moss and Chris Gaffney whether to treat the ipsilateral neck only or to extend the treatment volume to potential occult mucosal primary sites. M. Bernier. Oncol. J. 120 . H... N. Toms. (2004). A randomised multicentre trial of CHART versus conventional radiotherapy in head and neck cancer. S. J.. Forastiere. J. 22 (14 Suppl. J. ASCO Annual Meeting Proceedings (Post-Meeting Edition). N. T. (2003). London: Royal College of Surgeons.. Cetuximab prolongs survival in patients with locoregionally advanced squamous cell carcinoma of head and neck: a phase III study of high dose radiation therapy with or without cetuximab. L. (2002). 1310–17. Saunders.. A. (2004). TNM Classification of Malignant Tumours. AJCC Cancer Staging Manual... Bonner. et al. Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: phase III randomized intergroup study 0099. et al. Domenge. C. ed. New York: Springer. REFERENCES AJCC. H. Le Fur. (2002). Harari. only a small percentage of patients actually develop a primary tumour. Elective mucosal irradiation has not been shown to increase survival but does have potential for severe morbidity. A. Dische. M. C. Hyperfractionation versus conventional fractionation in oropharyngeal carcinoma: final analysis of a randomized trial of the EORTC cooperative group of radiotherapy. 355. It is advocated by some on the basis that there is a reduced incidence of a primary lesion developing if all the potential mucosal primary sites are irradiated along with the nodal drainage areas. J. Cooper. et al. (2004) CancerStats Monograph 2004: Cancer Incidence. This may occur many years after the RT and raises the possibility that a second primary may have occurred rather than a late appearance of the original primary tumour.. L. (1992). et al.. Overgaard. Pignon. 5507. J. 231–41. 123–36. LeBlanc. L.. Horiot. S. Monnerat. M. et al. Conventional vs modified fractionated radiotherapy. J. T.. Radiother.. (2002). 350. Positron Emission Tomography: A Strategy for Provision in the UK. J.). The evidence for elective mucosal irradiation is conflicting. British Association of Otorhinolaryngologists – Head and Neck Surgeons. (2002). End points for new agents in induction chemotherapy for locally advanced head and neck cancers.. (1997). Royal College of Radiologists. 1937–44. (2004). et al. H. (2001). Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. Giralt.. Overgaard. Vol. Clin. Shanmugaratnam. Hansen. Oncol. However. (2002). No 19: Histological Typing of Upper Respiratory Tract Tumors. P. 6th edn. Biol. Meta-analysis of radiotherapy in head and neck sqamous carcinoma: a meta-analysis based on individual patient data. J. Faivre. Improving Outcomes in Head and Neck Cancers.. 6th edn. P.. 71–2. NICE. A. Wittekind. London: Royal College of Physicians. 933–40. et al. 949–55. Temam. Survival and Mortality in the UK and EU. (1978). 54 (Suppl. (EORTC Trial 22931). Giri. K.

30% of which are oesophageal. although there are also increasing numbers of cases occurring in the young. The incidence of oesophageal cancer increases with age. Siewert describes a system to classify tumours involving the GO junction. Meanwhile. with rates as high as 100 per 100 000. where it would appear to be related to gastrooesophageal reflux disease (GORD) and to be less strongly associated with alcohol and smoking. Types of oesophageal tumour The types of oesophageal tumours are shown in Table 9. The oesophagus is a relatively common site for a second primary cancer. which are defined in Table 9. The risk increases with age and is more common in men. China and Transkei (South Africa). white male population. Carcinoma of the oesophagus Risk factors and aetiology The increase in adenocarcinomas (about 10% per year) of the middle and lower oesophagus appears to be associated more with reflux of acid (gastric and possible bile) and less with alcohol intake and smoking.5-fold increase in men and a two-fold increase in women in the incidence of adenocarcinoma of the lower oesophagus. the incidence of squamous cancer worldwide has remained steady or fallen slightly. especially squamous cancers. with approximately 7000 new cases and approximately 6000 deaths per year in the UK. Type I tumours are predominantly oesophageal and type III predominantly gastric. there has also been an increase in occurrence of adenocarcinoma of the gastric cardia. Squamous cell carcinoma occurs more often in Iran. Chemoradiation is more effective than radiotherapy alone and there is some evidence that preoperative chemotherapy is superior to surgery alone.9 OESOPHAGUS Tom Crosby Introduction In the past few decades there has been a dramatic increase in the incidence of adenocarcinoma of the oesophagus. There is a continuing controversy about the exact role of surgery combined with chemoradiotherapy and which one should be used as the primary therapy. The gastro-oesophageal (GO) junction is usually at about 40 cm. predominantly in the lower oesophagus and gastro-oesophageal junction. 4% of patients per year develop a second primary. Clinical trials are currently trying to define the optimum radiotherapy and chemotherapy regimens. although it has been suggested that tumours may be seeded at the time of enteral feeding. It is often divided into sections.2. In type II tumours the disease equally straddles the junction. There is a large worldwide variation in incidence. there has been a 3.1. though there are large geographical variations. This is likely to be related to a ‘field change’ effect on the mucosa from common aetiological agents. Incidence and epidemiology Tumour incidence in the UK is 11 per 100 000. For instance. following successful treatment for head and neck cancer. This trend has been noted across most patient populations worldwide but is most noticeable in the younger. The majority of patients present with symptoms of locally advanced or metastatic disease. The sternal notch is at 18 cm and the carina is usually at 25 cm but this can vary significantly among patients. In the past three decades. which limits survival from any treatment. Anatomy The oesophagus is usually measured from the central incisors (usually from 15 to 40 cm) at endoscopy. Combined modality therapy is increasingly used in these cases. 121 . During the same time period. particularly among male Caucasians.

and inactivation of p16 and p27. there is only about a 1% lifetime risk of developing an adenocarcinoma. aorta. upper. r Corrosives. middle and lower . r Smoking. The sequence from metaplasia through degrees of dysplasia to invasive adenocarcinoma is associated with genetic changes such as loss of TP53 function. and severe dysplasia is synonymous with carcinoma in situ.g. Infiltration into mediastinal structures occurs. Apart from GORD and Barrett’s oesophagus. A high proportion of adenocarcinomas have foci of gastric. except for those cancers occurring in the upper third of the oesophagus. r Tylosis palmaris. zinc deficiency). r Human papilloma virus. Significant problems exist with interobserver variation between pathologists when classifying dysplastic changes. r Fungally infected cereals. In 10% of those with the disease. overexpression of cyclins D1 and E. r Reduced dietary vitamin C. 1. Types of oesophageal tumour Type Benign Examples Leiomyoma Haemangioma Adenoma Malignant primary Adenocarcinoma (65%) Squamous (∼25%) Others (∼5%) Small cell Lymphoma BCC Melanoma Leiomyosarcoma Kaposi’s sarcoma Adenoid cystic Mucoepidermoid Gastro-intestinal stromal tumour Carcinoid Malignant secondary Lung Breast Melanoma in 3. Of people over 30 years old.or intestinal-type lining in the immediate vicinity of the tumour and elsewhere in the oesophagus. Risk factors are as follow.35 and 0.and H-RAS occur late in the transition to adenocarcinoma. most frequently into the trachea. Therefore. severe. r Coeliac disease. a columnar epithelialisation of the native squamous mucosa. r Plummer-Vinson syndrome. Amplification of MYC and K. in patients with Barrett’s oesophagus.5%. pleura.5% will have long-segment (> 3 cm) Barrett’s oesophagus. Pathological features are shown in Table 9. which are associated with PlummerVinson syndrome. Possible infective causes may be associated with r Helicobacter pylori. They spread circumferentially and longitudinally along submucosal and perineural pathways to form skip lesions of up to 5 to 6 cm from the primary tumour.and high-grade dysplastic change.Tom Crosby Table 9.12% of these patients will develop low. Other related conditions include: r Achalasia. respectively. diaphragm and vertebrae. Spread Spread of primary tumour Oesophageal tumours are usually locally advanced at presentation because there is no peri-oesophageal serosa to inhibit their growth.3. 0. LOH of the Rb gene. Associated conditions An associated condition is Barrett’s oesophagus.1. r Malnutrition (e. the aforementioned conditions are primarily associated with squamous cell carcinoma. Over a period of 2 to 5 years. 10% have symptomatic reflux disease. it will be mild and 122 Lymphatic spread The first-station lymph nodes (N1) of the oesophagus are supraclavicular. Physical/chemical Physical and chemical causes of oesophageal cancer include: r GORD. Pathology Adenocarcinomas occurring in the lower oesophagus usually arise on the background of Barrett’s oesophagus. r Alcohol.

This spread is classified as distant or M1b disease. Clinical presentation Typically. complete dysphagia (including saliva). often solids to liquids over 3 to 6 months. small cells with/without keratinisation) Adenocarcinoma Usually moderately to poorly differentiated. arising in Barrett’s mucosa. paracardial. brain and skin.Oesophagus Table 9. such as bone. The dysphagia grading system is as follows: r Grade 1. r Grade 2. r Grade 4. patient is able to eat a soft diet. splenic artery and coeliac. patient only manages a liquid diet.2. patients present late with progressive dysphagia. Other sites are becoming more common with the increase in adenocarcinomas. patient has difficulty with some foods such as bread and meat. subcarinal. diaphragmatic. divided into intestinal. To tracheal bifurcation To halfway to GO junction To GO junction 18–24 cm 24–32 cm 32–40 cm T3 ∼ T4/5 T5 ∼ T8 T8 ∼ T10 Anatomical description Starts below cricopharyngeus to thoracic inlet Distance from incisors 15–18 cm vertebral bodies C6 – T2/3 Table 9. r Grade 3. Intestinal type shows well-formed glands lined by malignant cells whereas the diffuse type is composed of mucin-producing neoplastic cells para-oesophageal. About 50% of patients experience pain (odynophagia) with or without signs or symptoms of aspiration or weight loss. Investigation and staging A full blood count.3. 123 . aortopulmonary. left gastric. Patients often point to the sternal notch if there is an upper-level obstruction and to the epigastrium if there is a lower-level obstruction. In addition to regional node spread. the rich submucosal plexus means that upper-third tumours can spread to coeliac nodes and lower-third to supraclavicular and deep cervical nodes in up to 30% of cases. Similar growth to above progressing to circumferential mass Microscopic features Squamous cell Usually moderately to well differentiated with typical features of squamous cancer (i.e. Parts of the oesophagus Correlation with Part of oesophagus Cervical oesophagus Thoracic oesophagus Upper Middle Lower GO = gastro-oesophageal. common hepatic. Pathological features of carcinoma of the oesophagus Description Macroscopic features Squamous cell Plaque-like lesion or elevation of mucosa evolving into polypoid fungating lesion or necrotic ulceration Adenocarcinoma Often nodular and multicentric. right and left paratracheal. Haematogenous spread The most frequent sites of haematogenous metastatic spread are the liver and lungs. diffuse or adenosquamous type. biochemical profile and check CEA and CA19–9 tumour markers should be performed.

1997). It may identify more widespread disease and so prevent unnecessary surgery in up to 20% of cases. The TNM staging classification for carcinoma of the oesophagus Stage T1 T2 T3 T4 Description Confined to mucosa Tumour spread through submucosa into muscle layer Tumour extends through muscle layer Tumour infiltrates into surrounding anatomical structures N1 M1a Involvement of regional nodes Involvement of coeliac nodes from lower-thoracic tumours or supraclavicular/deep cervical nodes from upper thoracic tumours (i. A PET scan can show nodal disease with up to 85% accuracy (Kobori et al. PET scanning may have a role in the staging and treatment of oesophageal cancer.. Early oesophageal cancer (T1 or T2. Two cycles of neoadjuvant cisplatin and 5-fluorouracil (5-FU) should be considered standard for fit patients with no metastases (Medical Research Council Oesophageal Cancer Working Group. photodynamic therapy can achieve long-term control Treatment overview For patients with tumours that appear resectable and who are sufficiently fit. Subdivision of this stage is important. In the UK.4.Tom Crosby Diagnostic endoscopy with or without a barium swallow. the incidence of node-positive disease is 16 to 36% for patients with T2 disease. radiotherapy (external beam/intraluminal). Laparoscopy should be considered for lower oesophageal lesions that extend below the diaphragm. still potentially encompassable in radiation portal) M1b Dissemination to non-regional lymph nodes or haematogenously to distant sites Adapted from UICC (2002). Consider bronchoscopy for tumours above the carina or where there are signs suggestive of T4 disease. either alone or in combination with external beam XRT. definitive chemoradiation is used for patients with no metastases and who are unfit for surgery. It may increase radiotherapy target volumes in up to 10% of cases but should not lead to reduced fields because of the significant false-negative rate. Patients with tumours of the upper third are often treated with non-surgical therapy. Palliative chemotherapy. 124 . there is an 80 to 95% concordance with pathology. Endoscopic ultrasound (EUS) is good for detecting the primary and mediastinal lymph nodes. MRI is not used routinely but may be useful for determining infiltration into the aorta or trachea. Table 9. N0) Early oesophageal cancer is more common in countries such as Japan. where there are surveillance and screening programmes. N0) or who are not fit for neoadjuvant therapy. but this needs confirmation in larger studies (Luketich et al. Photodynamic therapy has been used.e. and radiotherapy alone is used for patients unfit for chemotherapy. Surgery alone may be used for patients with early disease (T1 or T2. laser therapy.4. 2002). stent and supportive care alone are all treatment options for patients with advanced disease. It may also have a future role in assessing response to chemoradiation. measure its length (if the scope will pass) and take multiple biopsies. A barium swallow is sometimes done first but if it is abnormal.. an endoscopy is still needed to identify the start of the tumour. 2000). 1999). The disease length defined by EUS is frequently longer than the tumour length measured at endoscopy because both submucosal and nodal disease above and/or below the primary tumour may be identified. Staging classification The TNM staging classification is shown in Table 9. Because there is a rich lymphovascular plexus in the submucosa. As well as being useful for palliation in early inoperable disease.. Surgical resection of these cases is the treatment of choice but chemoradiation may play a role if the patient is not fit enough. Palliative surgical resection should be avoided where possible because there is good evidence that quality of life will never be regained when patients relapse within 2 years of their resection (Blazeby et al. surgery remains the mainstay of therapy in the UK. CT scanning will show the primary tumour as thickening but will often underestimate the tumour length and the presence of nodal disease. The commonest stage at presentation is T3 N1 disease with or without distant metastases.

or threestage. but with neck exploration. 125 . These conflicting data are difficult to explain but the results of the much larger UK trial are supported by a Cochrane review that includes nine additional RCTs showing a small survival advantage that becomes statistically significant some years after treatment but is mitigated somewhat by increased toxicity (Malthaner and Fenlon. Advanced oesophageal cancer (M1) Approximately 60% of patients are not fit for radical therapy. respiratory problems due to single lung ventilation during surgery and postoperative anastomotic leaks. Right thoracotomy is performed for mobilisation and resection and mediastinal lymphadenectomy. 1998) and the UK MRC OE02 trial (Medical Research Council Oesophageal Cancer Working Group.67–0. when used alone. The former study randomised patients to three cycles of cisplatin and 5-FU before and after surgery (if a response was seen) or surgery alone and found no significant survival difference (2-year survival of 35 versus 37%. Adjuvant therapy may improve local control in patients with R1 (microscopic residual) and patients with circumferential margin involvement with low nodal involvement (e. neither pre. co-morbidity and patient preference. and anastomosis. Curative surgery is possible in about 25 to 50% of cases.’ is an ongoing debate about the optimal surgical technique for operable tumours between the two. The transhiatal approach involves resection by an abdominal laparotomy with a cervical anastomosis. 2004). Total thoracic. day 1) and 5-FU (1000 mg/m2 . All patients should be considered for (neo)adjuvant therapy. days 2 to 5). Transthoracic two-stage (Ivor Lewis) surgery involves laparotomy and coeliac lymph node dissection. It has also been given in addition to external beam radiotherapy. Treatment of non-metastatic carcinoma of the oesophagus Surgery Surgery is the mainstay of treatment for most patients with oesophageal cancer who have technically resectable disease and are medically fit. mediastinal trauma.nor postoperative radiotherapy. although most specialist upper GI surgeons in the UK perform the former.Oesophagus when used alone in approximately 40% of cases. This regimen is the standard arm in the current MRC OE05 trial (see below). cisplatin and capecitabine). three-stage (McKeown) surgery is as per Ivor Lewis above.004). a radical resection that can include stomach and spleen. either because of the extent of their disease or because of co-morbidities and poor performance status. rates vary among countries.79. Two cycles of cisplatin (80 mg/m2 . In addition. the outcome from surgery for patients with oesophageal cancer remains poor and only 10 to 30% are alive 5 years after diagnosis. There is no evidence that chemotherapy given after surgical resection improves overall survival. given before surgery. improves survival and neither is routinely recommended. in which it is being compared with four cycles of ECX chemotherapy (epirubicin. There have been no relevant randomised trials of surgery compared to radiotherapy and caution is needed in interpreting comparisons of historical data. HR = 0. Radiotherapy patients have often been turned down for surgery and surgical staging may result in stage ‘migration. who should achieve a postoperative mortality rate of 5 to 10%.g. The UK MRC study randomised 802 patients to two cycles of preoperative cisplatin and 5-FU or surgery alone and found a survival advantage at 2 years for patients who had preoperative therapy (43 versus 34%.. Serious morbidities may arise as a result of moving patients during operations. En-bloc dissection involves thoracoabdominal resection. Locally advanced oesophageal cancer (T3 or T4 and/or N1) Management depends on stage. trials of adjuvant therapy have been undertaken. along with intrathoracic anastomosis. In an attempt to improve these results. is now the accepted standard treatment in the UK. 95% CI = 0. p = 0. although the evidence for this is not strong. performance status. dissection. 2002) are conflicting. fewer than three nodes involved). extended lymphadenectomy and the transhiatal approach. It should be performed by specialist upper GI surgeons working within specialist multidisciplinary teams. p = NS). Postoperative adjuvant therapy Despite improvements in perioperative care. There Preoperative chemotherapy The results from a US Intergroup study (Kelsen et al.93.

r Lower third.. 5-FU and cisplatin. the PTV should be grown manually along the axis of the oesophagus and nodal regions because of the inclination of the target volume. lower age. perhaps due to the regimens tested. about 30%.. the validity of the outcome has been questioned because of relatively poor outcome in the surgery-only arm and possible imbalances between the treatment groups. EUS. Other retrospective studies have found survival rates more consistently between 5 and 10% (Earlam and CunhaMelo. Although there was an improvement in survival. barium swallow). Urba et al. We know that even in 41% of patients who are shown by endoscopic mucosal biopsy to be clear of disease. 1999). to GO junction. Anterior and two lateral tattoos are used with a central radio-opaque marker. patients who are inoperable for medical reasons. 2004.. Definitive chemoradiotherapy Definitive chemoradiotherapy has been widely used to treat patients with oesophageal cancer. treatment is localised using CT planning. Cooper et al.. 1998). 31 to 40 cm. An outline is taken through the central volume. the improved outcome was not statistically significant (Urba et al. perhaps because of an underpowered trial design. to avoid obstruction of the posterior oblique fields. In the longitudinal direction. residual disease will be found at surgery (Bates et al. should be considered for definitive CRT if their disease can be encompassed in a radical radiotherapy volume. Patients who achieve a pathological complete response (pCR) after CRT (30 to 50%) have a better prognosis and such patients may not benefit from surgery. Normal structures (lungs.. 1999).and lower thirds of the oesophagus includes patient preparation. With regard to localisation and target volume. The composite target area is drawn on the central slice. The patients are planned with their arms above the head. CT. Stahl et al. The benefit has only been reported in patients with adenocarcinoma. are taken and the upper and lower borders determined. long-term survival can be achieved in about 20% of cases (Sykes et al. 18 to 31 cm. similar to that seen in surgical series (Crosby et al. preoperative CRT was associated with about a 30% long-term survival rate. it is not yet possible to reliably identify such patients preoperatively.5 cm intervals. Together with the lack of consistent benefit. pCR is associated with higher doses of radiotherapy. slightly arbitrary upper limit. Patients with upper-third tumours can be treated in the same way as patients with a head and neck cancer such as postcricoid carcinoma. 2005). 1996). However. although at the expense of increased toxicity (Cooper et al. this was a result of an increase in postoperative surgical mortality which has also been seen in other studies (leading some to be cautious about using this approach). A shell is needed for planning. Walsh et al. 1980) and the randomised trial of chemoradiation compared to radiotherapy alone had no survivors beyond 3 years with radiotherapy alone (Cooper et al. Studies consistently show a 5-year survival of 126 . The GTV is defined on each slice and the PTV is grown manually or automatically. 2001). 1996). 2003). ideally with 3D conformal capabilities. and a shorter treatment time (Geh et al. In the UK. In both trials.. in whom a complete R0 is unlikely or who decline surgery. cervical oesophagus.. 1997). In a larger multicentre French study of squamous cancer only.Tom Crosby Preoperative chemoradiotherapy Two randomised trials have shown better survival rates for chemoradiation (CRT) given 4 to 6 weeks before radical surgery compared to surgery alone.. used a hyperfractionated schedule but. Using radical radiotherapy as monotherapy for carefully selected patients. 2002. heart and spinal cord) are added in the same way. r Middle third. 15 to 18 cm. Axial slices. at ≤0. Denham et al. 1999. 2000). This technique will not be discussed further here.. and anterior and anterior oblique fields are used as part of combined modality therapy. Concurrent chemoradiation has been shown in a randomised trial and a subsequent systematic review to be significantly better than radiotherapy alone. the following are definitions of oesophageal thirds: r Upper third. means that triple modality therapy should only be used in centres that have experience in this complicated therapy and ideally as part of clinical trials. The radiotherapy technique for the mid... using all diagnostic data available (endoscopy.. an improvement in cancer-specific survival did not translate into an overall survival benefit (Bosset et al. positioning and immobilisation. although the data available are predominantly for squamous cell oesophageal cancer (Bedenne et al. in both phases to avoid movement between phases. Wong and Malthaner.. described a schedule with a relatively high dose per fraction of radiotherapy together with cisplatin/5-FU chemotherapy (Walsh et al. Radiotherapy and chemoradiotherapy technique As a practical guide for radiotherapy planning. 1996.

with the thick end located posteriorly. because the cord will receive 20 to 50% of the dose prescribed in phase 2. r If chemoradiotherapy is the sole curative therapy. plan phase 2 at the same time as phase 1. r Posteriorly as the 0. r To avoid excessive spinal cord dose. whereas the three-field second phase spares the spinal cord and heart. treating each field daily Monday to Friday. The dose. r Disease localisation can be difficult at or around the GO junction on a CT scan and this can be made easier by the use of a barium swallow. It is important to know the spinal cord dose that will be received during the second phase so that you can determine when to change to the three-field plan from phase 1. Siewert types 1 and 2). r The tumour length defined by EUS is nearly always longer than that seen at endoscopy reflecting submucosal spread. V40 ≤ 30%. V20 ≤ 25%. Figure 9. V45 = 0%. r In the second phase. The commonest regimen is probably four three-weekly cycles. if the tumour involves the GO junction (i. r Use 10 MV photons. the target volume stays the same for both phases. An alternative to this is to use a reducing-field technique (5 cm for phase 1 and 2.5 to 1 cm margin around EUSdefined tumour. the position of the oesophagus along its length and the different tissues through which radiation must pass to the PTV. usually the point of intersection of the central axes. Also the disease length is often longer again due to the lymphadenopathy lying more proximal or distal to the tumour. The plane of treatment may be inclined. requiring head rotations. give 45 Gy in 25 fractions.e. Concurrent chemotherapy Cisplatin and 5-FU are the agents most commonly used concurrently with RT. During a single phase or the second phase of the two-phase technique. When EUS has been used to stage the disease. and to treat the mediastinal lymph nodes. In a two-phase technique. and splenic arteries). the inferior margins that define the PTV should be 3 cm below the EUS-defined GTV. The plan should be verified in the simulator using barium before starting treatment and at least one portal image should be taken in the first three fractions of each phase on the linear accelerator and then weekly throughout treatment thereafter. The clinical target volume can be described r Superiorly and inferiorly as the EUS-defined extent of the tumour (primary or nodal) with a 2 cm margin along the axis of the oesophagus. Anterior–posterior fields in the first phase reduce the radiation doses to the lung and ensure adequate coverage of the posterior mediastinum. The PTV minimum should be no less than 95% and the PTV maximum should be no more than 107%. to allow for submucosal spread. They both have reasonable singleagent activity and are potent radiosensitisers. along the lesser curve to include the paracardial and left gastric lymph nodes with or without the common hepatic.Oesophagus r The GTV should encompass the most proximal and distal disease extent identified on any diagnostic procedure (CT. For lower-third tumours. is prescribed to the ICRU 50 reference point. give about 50 to 60% of the dose in phase 1. in 127 . Occasionally the anterior field can be wedged (∼15o ) in the superior–inferior direction.e. reduce the dose to 50 Gy in 25 fractions. Another 1 cm is added in all directions to create the PTV. No point outside the PTV should receive greater than 105%. corresponding approximately to a 10% variation of the ICRU-defined minimum and maximum of the dose prescribed to the ICRU 50 reference point. The aim is to treat the tumour and ‘normal’ oesophagus. Recommendations for normal tissue tolerances are as follow: r Combined lungs. EUS or PET). However. r If chemoradiotherapy is preoperative. Typical radiation doses: r If radiation is the sole treatment. particularly if EUS is not available. r Laterally and anteriorly as the 1 cm margin around EUS-defined tumour.1 shows a 3D conformal radiotherapy plan for carcinoma of the oesophagus. r Heart. the posterior oblique fields are wedged. Planning is important: inhomogeneity can occur because of changing body contour. r Spinal cord. a dose of 60 to 64 Gy (2 Gy per fraction) should be delivered to the target volume. Treatment can be delivered with a single three.or four-field plan throughout treatment or using a twophase technique. usually 15 to 30o . the inferior margin should be grown manually along the pathway of the draining lymph node stations (i. give 26 to 30 Gy in 13 to 15 fractions.5 to 3 cm for phase 2). coeliac. give 20 to 24 Gy in 10 to 12 fractions. r In the first phase.

After surgery. because the non-concurrent chemotherapy is given in a neoadjuvant phase. PTV = planning target volume.5. Side effects from radiotherapy are shown in Table 9. depending on their level of fitness. 1999) and.Tom Crosby Isodose % 95 70 50 20 10 Isodose % 95 70 50 20 10 (a) 100 90 (c) PTV Cord PRV 80 70 Volume (%) 60 Heart 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100 110 Liver Combined Lungs Dose (%) (b) (d) Figure 9. Treatment of recurrent carcinoma of the oesophagus The prognosis of patients with recurrent disease is very poor. which has been achieved by the use of multileaf collimators.. there is time for careful radiotherapy planning and improving the patient’s dysphagia before the radiotherapy starts.1. ing on the initial local therapy given. A 3D conformal plan for radical radiotherapy for carcinoma of the oesophagus. and (d) DVHs for the PTV and organs at risk. the tumour most commonly recurs locally and endoscopic placement of a stent as described in the next section is frequently used to relieve dysphagia. which radiotherapy is given in cycles 3 and 4. (c) coronal section showing how the isodoses have been shaped to conform to the PTV. After definitive chemoradiation. PRV = planning organ at risk volume. The pattern of recurrent disease varies depend128 Palliative treatments and treatment of metastatic carcinoma of the oesophagus Surgery Surgical resection has no role in the palliation of patients with oesophageal cancer. the majority of patients who relapse will do so at distant sites and they should be considered for palliative chemotherapy (as discussed in the next section). (b) sagittal section showing the oblique angle of the PTV relative to the horizontal. . This regimen delivers the same number of planned chemotherapy cycles as used in the RTOG-85–01 study (Cooper et al. In this plan. a single-phase treatment has been used with four fields: (a) transverse section showing how the dose outside the PTV is spread between critical structures.

p. do not require dilatation to be inserted endoscopically. Laser therapy is better for exophytic. avoid stent if possible Pulmonary fibrosis Pericarditis Ischaemic heart disease Tracheo-oesophageal fistula Medical management Medical management Medical management Endoscopic placement of covered stent Mucilage Maintain haemoglobin above 12 g/dl Usually self-limiting. reducing course steroids if severe Management quality of life than other combinations (Webb et al. It may make dysphagia worse at first because of mucositis. r Cisplatin 60 mg/m2 day 1. Endoscopic treatment Stents In experienced hands. chemotherapy has been shown to improve median survival and quality of life. The following is an example of a chemotherapy regimen using epirubicin.s. cisplatin and infusional 5-FU (ECF): r Epirubicin 50 mg/m2 day 1. Side effects of radiotherapy to the oesophagus Side effect Acute Tiredness General advice for fatigue: explanation. dysphagia usually improves after about 10 days. more expensive. moderate exercise if able Mucositis Myelosuppression Pneumonitis Late Benign stricture Endoscopic evaluation.. and have not been shown to improve survival or quality of life. Shorter fractionation schedules. r Metoclopramide 10 mg q.. with highdose-rate Microselectron® . 129 . 1997). but there may be problems related to the UV sensitiser.v. r 5-FU 200 mg/m2 per day. Expandable metal stents provide a wider lumen. Endoscopic laser-thermal Nd-YAG or photodynamic therapy (PDT) Both of these interventions need two or three sessions to provide worthwhile benefit and often need to be repeated every 4 to 8 weeks. In those who do respond. such as 20 Gy in 5 fractions over 1 week.n. flat.d. r 5-HT3 antagonist i. Dilatation Dilators rarely provide more than a few days of relief from malignant dysphagia.5.v.Oesophagus Table 9. may not be effective for long enough. ILT has the advantage of requiring only a single treatment with fewer systemic side effects and its effect may be seen earlier. EBRT has been shown to be effective in 60 to 80% of cases. short lesions of the mid or lower oesophagus. They are. however. It is most useful for patients with minimal dysphagia and a dose that will control the local disease for the majority of their remaining life should be used.d. The dose is 1500 cGy at 1 cm. and 4 mg orally b. goal setting.r. r Antiemetics include r Dexamethasone 8 mg i. The dose is 30 Gy in 10 fractions over 2 weeks with anterior–posterior fields or 40 Gy in 15 fractions over 3 weeks either alone or in combination with chemotherapy.. The benefits of second-line therapy are less certain. and dilatation. Responses can be seen with taxanes and with irinotecan-based combination therapy. Alcohol injection Alcohol injection can be good for tumour overgrowth of stents and bleeding tumours.d. associated with more pain. biopsy. and with significantly better survival and Palliative radiotherapy Both external beam radiotherapy (EBRT) and intraluminal brachytherapy (ILT) are effective in relieving dysphagia. 625 mg/m2 orally b. This regimen is associated with response rates of about 45%. PDT is better for submucosal. days 3 to 7. for infiltrating tumours and for repeated treatments. days 1 to 21). although the benefits are not maximal until 4 to 6 weeks after treatment. and are less likely to move than plastic ones. for 10 days. endoscopic stent insertion is usually successful and provides good palliation of dysphagia. days 1 to 21 (or capecitabine. Palliative chemotherapy For patients with advanced GO cancer.

. B. However.8 Gy) would reduce this finding compared to a modified standard arm (50.8 to 2. Treatment should be similar to that for patients with pulmonary small-cell carcinoma.. The overall median survival is approximately 12 months for patients with limited disease at presentation. et al. cisplatin and capecitabine) in 1300 patients with adenocarcinoma of the oesophagus.4% of all cases of OC. Oncol. But any possible advantage must be balanced against the risks of major surgery in patients with a high chance of treatment failure because of metastatic disease and a relatively poor prognosis. The MRC 0E02 trial (Medical Research Council Oesophageal Cancer Working Group.Tom Crosby Special clinical situations Small-cell carcinoma of the oesophagus Small-cell carcinoma of the oesophagus accounts for 0. 2000). Concurrent radiation therapy and chemotherapy followed by esophagectomy for localized esophageal carcinoma. a significant proportion have had a residual viable tumour (Medgyesy et al. S.4 Gy). Recent important trials in oesophageal cancer Intergroup 0113 (Kelsen et al. 1998) is a 400-patient trial that compared three cycles of preoperative cisplatin/5FU with the option for postoperative chemotherapy if there was evidence of a response with surgery alone. two concomitant with XRT). which showed a benefit for chemoradiation over radiotherapy alone. MS is 5 months without treatment versus 11 months with chemotherapy. following administration of definitive chemoradiation (three cycles cisplatin/5-FU and XRT 50 Gy in 25 fractions starting with cycle 2).. There was no difference in outcome (35 versus 37% 2-year OS). there is a 50% 5-year survival for early tumour (T1 or T2. 2002) studied 802 patients in the UK with operable oesophageal cancer who were randomised to receive surgery with or without two cycles of cisplatin/5-FU chemotherapy. four cycles. N0) and 20% 5-year survival for locally advanced tumours. Like pulmonary small-cell carcinoma.. 1999) involved randomised (1986–90) and non-randomised (1990–1) segments: 129 patients in the randomised and 73 in the follow-on study. A. Survival at 2 years was improved from 34 to 43% in favour of preoperative therapy. In advanced disease. 2002). (2002). it is likely that the cell of origin is a pluripotential stem cell.. F. Michel. It has been suggested recently that patients with localised disease may benefit from surgery following chemotherapy and radiotherapy because. C. Radical radiotherapy results in 20% 5-year survival. Randomized phase . in those who have had a resection. The RTOG-85–01 study (Cooper et al. 14. 130 REFERENCES Bates.. Prognosis of carcinoma of the oesophagus Results from selected series For radical surgery.. There was no difference in the two arms and so the trial closed early. 2000).. P Bouche. L. Of patients who received CRT. Clin. Selective surgery A group in Canada. This option of selective surgery is the subject of an ongoing EORTC trial. whereas there was no long-term survivor in the RT-only arm of the study. Detterbeck. there was still a 45% local failure rate. There were nine deaths (9%) in the high-dose arm. et al. and radical chemoradiotherapy in 30% 5-year survival. INT 0123 was designed to see whether a higher dose of XRT (64. with combined chemotherapy and radiotherapy. although these did not occur during the high-dose therapy. Therefore.5 weeks) with chemoradiotherapy (50 Gy in 25 fractions + cisplatin/5-FU.. together with small-cell features. which compared radiotherapy alone (64 Gy in 32 fractions over 6. only offered surgery to those who have either biopsy-positive disease or 75% or lower rate of regression on CT or with localised recurrence. III trial in locally advanced esophageal cancer: . Current trials in oesophageal cancer The MRC OE05 is a UK multicentre trial that compares the benefit of the chemotherapy schedule from OE02 with four cycles of ECX (epirubicin. This may be why the tumours may be mixed with keratin and mucin production. Bernard. Areas of current interest Dose of radiotherapy In the RTOG-85–01 study. A. O. All cases show ∼7% survival at 5 years. 30% survived 3 years. J.. (1996). 156–63. Bedenne. Two-year survival in the high-dose arm was 24% compared to 33% in the standard arm (Minsky et al. an overview of trials of chemoradiation has suggested that higher doses are associated with better outcomes (Geh et al.

Cunningham.. Br. et al. P.. 161–7. 457–61. Donovan. (2002). Meltzer. Clin. A comparison of multimodal therapy and surgery for esophageal adenocarcinoma. S.. H. Med. R. R. Schauer. 98–108. C. ed. Orringer. Webb. radiotherapy for carcinoma of the oesophagus: an effective alternative to surgery. D. D. 1167–74. 829–37. A. N. 335. (2004). Engl. J. et al. 60–4. J. (2005). (2002). J. M. Ginsberg... et al. 14. Turrisi. Soc. Hollywood... Cancer. W. 261–7. Cancer. J.. A. et al. Clin. A. J. positron emission tomography in staging esophageal cancer. Crosby. Burt.. Medical Research Council Oesophageal Cancer Working Group. F. Clin. Relapse patterns after chemoradiation for carcinoma of the oesophagus. Sykes. Jr. et al. 131 . Am.. Wilke.. 765–9.. Scarffe. 88. (2002).. Oncol. 70–5. 15. Stahl. Oncol.. Putnam. R. Med. J. Steigler. Bentzen. (1996). UICC. 48.. In TNM Classification of Malignant Tumours. N. D. D. Chemoradiotherapy followed by surgery compared with surgery alone in squamous-cell cancer of the esophagus. C. J. M. Gignoux. New York: Wiley-Liss. and methotrexate in advanced esophagogastric cancer. Ann. N.. J. R. et al. M. Farndon. Brewster. In The Cochrane Library. T. et al. J. C. 1623–7. D. T. U. P A.. 90. Radiother. Kirihara. Positron emission tomography of esophageal carcinoma using 11 C-choline and 18 F-flurodeoxyglucose: a novel method of preoperative lymphnode staging. Bond. doxorubicin. Surg. Br. Minsky. Kosaka. A. (1996). J. and Cunha-Melo. J. and fluorouracil versus fluorouracil. Y. Guo. Med. et al. Sobin and Ch.. INT 0123 (Radiation Therapy Oncology Group 94–05) phase III trial of combined-modality therapy for esophageal cancer: high-dose versus standard-dose radiation therapy. M. (R. T. 462–7. Walsh. 88. Oesophageal squamous cell carcinoma: II. J. Geh. Randomized trial of preoperative chemoradiation versus surgery alone in patients with locoregional esophageal carcinoma. Luketich. 67. A.. 15. Engl. Pajak. (2003). J. Earlam. et al. Herskovic. 1638–48. Bosset. Oxford: Update Software. et al. A. Thoracic Surg. S.. J. H. B. Issue 3. H. D. J. (2000). (1980). Oncol. J. N. Clin. A critical view of radiotherapy. G.. Clin. pp. 6th edn. and Fenlon.. Am. D.. 64. 359. cisplatin. E. (1999).. J. Pajak. (2001).. Combined chemotherapy and radiotherapy (without surgery) compared with radiotherapy alone in localized carcinoma of the esophagus (Cochrane Review). J. et al.. et al. R. N...Oesophagus radiochemotherapy followed by surgery versus radiochemotherapy alone (FFCD 9102). Urba. Proc. D.. R. Kobori. N. Fink.. Definitive chemoradiation in patients with inoperable oesophageal carcinoma. et al. Anderson Cancer Center experience and literature review.. Medgyesy. Small cell carcinoma of the esophagus: the University of Texas M. J. (2000). 20. 19. Borley. Kelsen. B. R. D. Oncol. J. Coll. (1998). J. Cancer. Surgical resection with or without preoperative chemotherapy in oesophageal cancer: a randomised controlled trial. et al. Oncol. Triboulet. P R. (2004).. Radiol. Proc. R. J. Role of . A. A. Wong.. et al. Oxford: Update Software. Clin. Randomized trial comparing epirubicin. et al. Combined preoperative chemotherapy and radiotherapy in patients with locally advanced esophageal cancer: Interim analysis of phase II trial. Blazeby. J.. 1979–84.. A prospective longitudinal study examining the quality of life of patients with esophageal carcinoma.. Lancet. 339. Issue 2.. (1997). O. J. 15–21. J. Kilmurray.). 281. B. followed by surgery compared with surgery alone for localized esophageal cancer. Wolff. (1999).. Slevin. Oncol. Radical . S.. P Ginsberg. Noonan. 337..... T. Chemoradiotherapy of locally advanced esophageal cancer: long term follow-up of a prospective randomized trial (RTOG 85–01). F. J. 1727–33. Cooper.. Chemotherapy . Cancer. (1998). Abstr 519.. Preoperative chemotherapy for resectable thoracic esophageal cancer (Cochrane Review). In The Cochrane Library. 305–13. (2000). Oncol. Engl. 1781–7. A.. L. F. A. M. Clin. (1997). 86. Preoperative chemoradiotherapy in esophageal cancer: evidence of dose response. Malthaner. D. Abstr 958. Denham. Oncol.. S. 262–7. (1997). Soc. Wittekind. and Malthaner.

the superior portion of the left kidney. it abuts the diaphragm on the left and the left lobe of the liver on the 132 .1. In some series. and a low incidence in the USA. whereas postoperative chemoradiotherapy is standard treatment in the USA. right gastric and gastroepiploic arteries (from the common hepatic artery) and the left gastro-epiploic and short gastric arteries (from the splenic artery). There is a high incidence of stomach cancer in Japan (>90 in 100 000). gastric cancer remains a major health problem: it is the sixth most common malignancy in the UK and the second most common cause of cancerrelated death worldwide. potential sites of direct invasion) are the spleen. Over several generations. The incidence is falling for endemic distal intestinal-type tumours (see p. approximately 10 000 new cases per year. Proximal tumours now account for 50% of gastric cancer diagnoses in the UK. This difference is probably due to exposure to environmental factors in early life. It is anatomically defined in three parts: the proximal fundus (cardia). The coeliac axis originates at or below the pedicle of T12 in 75% of people and at or above the pedicle of L1 in 25% (Kao et al. The nodal stations are shown in Figure 10.1. the risk in migrant populations changes toward that of the host country. right.10 STOMACH Michael Button and Tom Crosby Introduction There has been a steady decline in the incidence of gastric cancer in most countries in the world in the past 50 years. local recurrence may occur in up to 50% and the 5-year survival is 30 to 40%. the pancreas. the left adrenal gland. which reflects changes in aetiological factors. the body.. adjuvant therapy is being used increasingly. particularly among young white people. although a rich lymphatic plexus complicates drainage routes. Adenocarcinoma accounts for 95% of all malignant tumours. The peak incidence age of stomach cancer is 65 years. even in these cases. The only current curative treatment is surgery. Perioperative chemotherapy is used most commonly in the UK. The vascular supply of the stomach comes from the coeliac axis via the left gastric. Lymphatic drainage follows the vascular supply. but this may reflect diagnostic variations in the classification of tumours around the gastro-oesophageal junction. Types of tumour The types of tumour that affect the stomach are shown in Table 10. and the male-to-female ratio is 3:2. Incidence and epidemiology The UK annual incidence of stomach cancer is 15 to 20/100 000. 133) that are related to environmental factors. but it is rising for proximal intestinal-type tumours that are related to gastro-oesophageal reflux disease (GORD). mostly draining into the coeliac nodal area. Because of this and because response rates to combination chemotherapy are 40 to 50% in patients with advanced disease. Other adjacent organs (and. therefore. Despite the decline in the cancer that was previously most common. 1993). oesophageal cancer is now more common than gastric cancer. but in the UK most patients present late with locally advanced or metastatic disease. there has been a rapid rise in cancers affecting the gastro-oesophageal junction and cardia. Anatomy The stomach begins at the gastro-oesophageal junction and ends at the pylorus. and the transverse colon. and the distal pylorus (antrum). South America and Eastern Europe. It is covered anteriorly by the peritoneum of the greater sac and posteriorly by the peritoneum of the lesser sac. Israel and Kuwait. However. the distal type of cancer. Proximally. and. Only 25 to 40% of cases are amenable to potentially curative surgery. approximately 6500 deaths occur per year.

Helicobacter pylori infection increases cancer risk by three to six times. diets rich in fruit and vegetables or vitamin C and the patient having blood type O. HNPCC. 14 = mesenteric root. and the patient having blood type A. The two histological variants described in the Lauren classification.2 and 10. Atrophic gastritis (secondary to H. 2 = left cardiac. Types of stomach tumour Type Benign Examples Inflammatory fibroid polyp Adenoma (sessile/pedunculated polyps) Leiomyoma Adenoid cystic tumour Hamartoma Malignant primary Carcinomas Adenocarcinoma: diffuse/intestinal types Squamous Small cell Others Lymphoma (most common site of primary GI lymphoma) Carcinoid tumour GI stromal tumour Malignant secondary Rare. 133 Figure 10. 3 = lesser curvature. 12 = hepatoduodenal ligament. for example. Carcinoma of the stomach Risk factors Environmental risk factors include diets low in vitamins A and C. 7 = left gastric artery. pylori. 50% of stomach tumours started in the pyloric region. With regard to inflammation. H. Blair. ASCO Gastrointestinal Cancers Symposium virtual meeting. CDH1 mutation (E-cadherin). However. pylori) is associated with distal disease. 13 = retropancreatic. 1995). 8 = common hepatic artery. Expression of the Cag A virulence factor further increases the risk. 10 = splenic hilus. 9 = coeliac artery. smoking and GORD) is associated with proximal gastric and gastro-oesophageal cancers. inherited cancer syndromes such as FAP. Genetic factors are more associated with the diffuse type. diets high in salty or smoked foods . Factors that protect against stomach cancer include the use of aspirin or NSAIDs. pylori has been classified as a class I carcinogen for gastric cancer. and this tumour may arise in a multistage process from chronic active gastritis through gastric atrophy. intestinal metaplasia and dysplasia to frank malignancy (Correa. 13 6 14 Pathology In the past. 2006. 11 = splenic artery. Of benign adenomas.1. 1 = right cardiac. 5 = suprapyloric. Japanese nodal stations as described by Japanese Gastric Cancer Association (1998). However. are both mucin-secreting adenocarcinomas (see Tables 10.3 for pathological features). 1 12 5 8 9 7 2 10 3 11 4 or nitrates. 25% in the body and 25% in the cardia. the decreasing incidence may be related to improved treatment of H. the frequency of proximal tumours is increasing. which carries a risk of stomach cancer of up to 80% by the age of 80 (V. BRCA2 and Li Fraumeni syndrome. Genetic risk factors include. www. There is a seven-fold increase in the incidence of malignancy in the 5 years after diagnosis of a benign gastric ulcer and a three-fold increase in incidence following the diagnosis of pernicious anaemia. 4 = greater curvature. 10 to 20% of tumours more than 2 cm in size transform into carcinoma. H. tumours in the lesser curve were three or four times more frequent that those in the greater curve.1. Environmental factors are thought to be more important in the aetiology of the intestinal variant. 6 = infrapyloric. low socioeconomic status (which may be related to the previous factors) and radiation exposure at a young age.asco. conference presentation. smoking. involvement of stomach by lobular carcinoma breast GI = gastrointestinal. Barrett’s oesophagitis (related to obesity.org). intestinal and diffuse. pylori infection is associated with both histological types.Stomach Table 10. especially for intestinal-type distal carcinoma.

-5AC and -6) are often positive in oesophago-gastric cancers. there may be haematogenous or lymphatic spread. and those related to Barrett’s oesophagitis. type II. including proximally into the oesophagus. but not usually into the duodenum (‘duodenal block’). type IV. Clinical presentation Symptoms of gastric cancer include anorexia. 1926). Thirty percent of patients have liver involvement at presentation. Pathological features of diffuse-type stomach cancers Features Macroscopic Description Usually endophytic.. vomiting and bleeding (haemetemesis or melaena. venous drainage is initially to the liver (involved in 30% of cases). epigastric discomfort. bone and brain metastases occur in decreasing frequency. type III. The patterns of involvement are difficult to predict because of the rich lymphatic network. periumbilical tumour deposits (Sister Mary Joseph’s nodules). often metastasise to the liver Microscopic Glandular formation predominates with varying degrees of differentiation. renal function if chemotherapy contemplated) tests should also be performed. Once local invasion has occurred. unclassifiable (Borrman. weight loss (indicator of a poor prognosis). polypoid or fungating masses which often ulcerate. FBC (anaemia) and biochemistry (albumin. the lymphatic system. differentiated. rectum or rectal shelf (Blumer’s tumour). It is not clear whether differences in the patterns of staining could be used to differentiate between tumours of the antrum.3. the vascular system or by the transperitoneal route. The Borrman classification divides tumours into five types: type I. In lymphatic spread. not associated with intestinal metaplasia Spread Gastric cancers usually spread either by direct extension. 2002). to the ovaries (Krukenberg tumours). growth penetrates the stomach wall and spreads laterally producing marked thickening (‘leather bottle stomach’ or linitus plastica) with or without ulceration. associated with intestinal metaplasia in nearly 100% of cases Table 10. SCF lymph nodes.2. Endoscopy can be highly sensitive . Systemic metastases are not as common. Investigation and staging A full patient history should be taken and directed clinical examination should be performed. diffusely infiltrating (linitus plastica). ulcerated with elevated borders. the rich submucosal lymphatic plexus helps intramural spread. Spread via the falciform liga134 ment can produce subcutaneous. weight loss. and signs of metastases. Patients may have a poor performance status because of their disease. early satiety. Pathological features of intestinal-type stomach cancers Features Macroscopic Description Usually exophytic.Michael Button and Tom Crosby Table 10. producing nodular. liver function tests. and type V. may include serosal or lymph node involvement Microscopic Diffusely infiltrative with sheets of cells (often ‘signet ring’ due to nuclear compression by mucin). In haematogenous spread. Transcoelomic spread involves peritoneal dissemination after extension through the serosal surface of the stomach. lung. dysphagia. see Smalley et al. polypoid or fungating. often poorly The appearance of gastric cancers can also vary. Immunohistochemistry with cytokeratin markers CK7 and 20 and mucin (MUC-1. Examination findings include epigastric mass. minimal gland formation. cardia. Gastric cancer spreads locally to contiguous structures (as described earlier in the anatomy section) or to the anterior abdominal wall. -2. The initial draining lymph nodes sit on the greater and lesser curves but other lymph node groups are often involved (including the SCF). and around 60% have lymph node involvement (for discussion. ulcerated and invading the gastric wall. 10%).

Laparoscopy assesses the peritoneum. bone scan). lymph nodes and metastases. AJCC Cancer Staging Manual 6th Edition for carcinoma of the stomach (applies to carcinomas only) Stage TX T0 Tis T1 T2 T2a T2b T3 T4 NX N0 N1 N2 N3 MX M0 M1 Description Primary tumour cannot be assessed No evidence of primary tumour Carcinoma in situ: intraepithelial tumour without invasion of the lamina propria Tumour invades lamina propria or submucosa Tumour invades muscularis propria or subserosa Tumour invades muscularis propria Tumour invades subserosa Tumour penetrates serosa (visceral peritoneum) without invasion of adjacent structures Tumour invades adjacent structures Regional lymph nodes cannot be assessed No regional lymph nodes Metastases in 1–6 regional lymph nodes Metastases in 7–15 regional lymph nodes Metastases > 15 regional lymph nodes Distant metastases cannot be assessed No distant metastases Distant metastases Table 10. Stage classification The AJCC classification and stage groupings are shown in Tables 10. CEA has low sensitivity (raised in only 30%) and. although EUS is less sensitive in the assessment of stomach cancer than in the assessment of oesophageal tumours. and the type of resection is tailored to the site of the tumour in the stomach. therefore. in experienced hands but may miss diffuse or intramucosal tumours. which can be missed endoscopically but may show characteristic radiological features. Adapted from AJCC (2002). The aim is to accurately stage patients and to reduce the rate of ‘open-and-close’ laparotomy. respectively. although progress is being made with the 135 .4 and 10. Combination chemotherapy is effective at improving symptoms and offers a survival benefit with improved quality of life in the palliative setting. Tumours are also graded: r GX – grade cannot be assessed. but not in Europe. Endoscopic ultrasound (EUS) is used for assessing the depth of penetration and local lymph node involvement.Stomach Table 10. r G2 – moderately differentiated. Treatment overview Surgery is the mainstay of radical treatment. A barium swallow may be more sensitive for detecting linitis plastica. Stage groupings for carcinoma of the stomach Stage Stage 0 Stage IA Stage IB Description Tis N0 M0 T1 N0 M0 T1 N1 M0 T2a/b N0 M0 Stage II T1 N2 M0 T2a/b N1 M0 T3 N0 M0 Stage IIIA T2a/b N2 M T3 N1 M0 T4 N0 M0 Stage IIIB Stage IV T3 N2 M0 T4 N1–3 M0 T1–3 N3 M0 Any T Any N M1 Adapted from AJCC (2002).g. CT scan of the chest/abdomen/pelvis should be performed to assess the primary tumour. Following the publication of the Intergroup 0116 randomised controlled trial.5. r G3 – poorly differentiated.4. r G1 – well differentiated. is not clinically useful.5. PET is experimental at present but may be useful for lymph node staging and in assessing response to neoadjuvant chemotherapy. the use of adjuvant chemoradiotherapy has become standard in the USA. r G4 – undifferentiated. Extending the benefit of chemotherapy to the adjuvant setting has proved challenging. Other investigations should be performed as clinically indicated (e. Debate exists about the optimal extent of lymph node dissection because of which there are large geographical variations in treatment. It is important to note the total number of nodes examined. liver capsule and mobility of the stomach and is considered a standard investigation before surgical resection.

higher rate of local recurrence. Patterns of failure after surgery for stomach cancer are shown in Table 10. 2003) have not shown an advantage for more extensive lymph node resection. Surgical schemes for total and partial gastrectomy are shown in Figures 10. Surgical scheme for a total gastrectomy: (a) before and (b) after. These differences may be related to surgical experience. 1982) and in up to 90% at autopsy (Lim et al. As a result. proximal tumours usually require total gastrectomy. In the UK. Distal tumours can be treated by partial gastrectomy (if a 6 cm proximal clearance can be achieved). 2005). to stage migration. more than two-thirds of patients diagnosed with gastric cancer will have unresectable disease or metastases at presentation. However. D1 resection includes removal of the perigastric nodes within 3 cm of the tumour. respectively. However. 136 . splenic hilum and also a splenectomy and distal pancreatectomy. which showed a significant survival benefit from the use of perioperative chemotherapy. Patterns of failure following surgery for stomach cancer Incidence (%) Surgery Radical surgery Complete resection is the only curative treatment (Jansen et al. to the fact that patients in Japan are often younger and fitter or even to a difference in disease biology. The majority of specialist upper GI surgeons in the UK perform a modified D2 resection for suitably fit patients.. hepatic artery.. D2 resection involves a more extensive lymph node dissection. 2005). locoregional recurrence is common.2.6.3. only 5% of patients present with T1 disease. (a) (b) Figure 10. There is controversy about the required extent of surgical resection.6. with a Roux-en-Y reconstruction. Surgical scheme for a Billroth II partial gastrectomy: (a) before and (b) after.Michael Button and Tom Crosby recently presented UK MRC ST02 (MAGIC) trial results.3. whereas the more limited D1 resection is more common in Western centres. local and/or regional lymph node recurrence occurs in 54% at second-look laparotomy (Gunderson and Sosin. with removal of lymph nodes around the left gastric artery. Outcomes from more extensive D2 resections performed in Western centres are worse than those obtained in Japan with greater postoperative morbidity and mortality (up to 10% postoperatively). Table 10. to the surgery itself. because more limited resections tend to produce worse functional outcomes and may result in a Pattern of failure Locoregional Gastric bed Anastomosis/ stumps Abdominal wound Lymph nodes Peritoneal seeding Localised Diffuse Distant metastases Clinical Reoperation 38 21 25 67 54 26 5 8 23 42 41 19 22 52 22 Autopsy 80–93 52–68 54–60 52 30–50 49 From Smalley et al. (a) (b) Figure 10. Despite extensive surgery. (2002). the role of adjuvant treatments has been extensively investigated. Of those whose disease is resectable.. Total gastrectomy may not be necessary as long as free resection margins can be obtained with a subtotal resection. D2 resection is more common in Japan.2 and 10. At least three randomised trials and a Cochrane review (McCulloch et al.

1997). with doses of X increasing from 500 mg/m2 (continuous daily dose) to 625 mg/m2 after analysis of fluoropyrimidine toxicity suggested that this would be tolerable (Sumpter et al. and only 10% of patients had local recurrence alone. lower oesophagus or gastro-oesophageal junction. A commonly used regimen in the UK and Europe is ECF (see following discussion). 1997). suggesting that the more-convenient oral capecitabine may safely replace infusional 5-FU in this combination (Cunningham et al. However. 1999). 75% were metastatic and 25% locally advanced. oxaliplatin (O). it is small and does not seem to justify routine use of adjuvant chemotherapy (Earle and Maroun.. A protocol change to include patients with adenocarcinoma of the lower-third oesophagus was made toward the end of the study. Trial accrual was slow and it took over 10 years to complete the study. 5-FU is active as a single agent and is included in most combination regimens.. if there is benefit. This study. Murad et al.. The extent of resection was not specified and was left to the discretion of the surgeon. 2005). Adjuvant chemotherapy should not be routinely offered to patients with gastric cancer except as part of a clinical trial. However..53–0. Perioperative chemotherapy The MRC ST02 ‘MAGIC’ trial (Cunningham et al.and three cycles of postoperative epirubicin. adriamycin and methotrexate. 1995).001). all of which are 5-FU based and give similar median overall survival and 2-year survival rates. 2-year survival 18 versus 9%) at the cost of greater toxicity (Moiseyenko et al. Recently published phase III trials have investigated various chemotherapy regimens for advanced disease (Ajani. It is not clear what combination regimen is the most effective.. 1994. 95% CI = 0.Stomach Chemotherapy Adjuvant chemotherapy Despite gastric carcinomas being the most chemosensitive of gastrointestinal carcinomas.75. 1993. 2005). 36 versus 23%) and progression-free survival (hazard ratio for progression = 0. Of the patients studied. but the median survival in phase III trials is still generally less than 12 months (Ajani. 1993).. Hallisey et al. which has response rates of 40 to 50%. Webb et al. Some newer combination regimens give higher response rates. The role of oxaliplatin awaits further 137 . The results showed that perioperative chemotherapy (three cycles of pre. it can be toxic and take up a significant amount of the patient’s remaining life and so the decision about whether it is appropriate for an individual patient needs to be made carefully... Hermans et al. 1997. EOF and EOX. 2005). 2006b). 2005). The TAX-325 trial (the largest to date) compared cisplatin/5-FU with docetaxel/cisplatin/5-FU in 457 chemotherapy-na¨ve patients and found higher ı response rates and survival in the docetaxel-containing arm (RR of 37 versus 25%. ECX. 5-FU (F) and capecitabine (X) in the following combinations: ECF.66. with high response rates seen in phase II studies.6 months. some centres have yet to adopt perioperative chemotherapy as a standard because of concerns regarding the aforementioned trial and the relative poor overall survival for patients in this study.60–0.81. with a higher overall survival (hazard ratio for death = 0. 2006a) included 503 patients of WHO performance status 0 or 1 with stage II or greater adenocarcinoma of the stomach. This chemotherapy regimen (with capecitabine replacing infusional 5-FU) will be the standard arm of the next NCRN multicentred RCT of adjuvant therapy. 5-year survival rate. p ≤ 0. it has also been shown to improve quality of life and to be cost effective (Glimelius et al. suggesting this regimen was effective at improving local as well as distant control. GOJ and stomach. A randomised trial has shown it to be more effective than FAMTX (5-FU. Resected tumours were smaller and of less advanced T-stage after chemotherapy. has demonstrated equivalence among the treatment arms.. Pyrhonen et al.. 85% had adenocarcinoma and tumours were evenly divided between oesophagus. 1999. p = 0.2 versus 8. from about 3 to 10 months.93.. 95% CI = 0. a phase III trial that evaluates the potential roles of oxaliplatin and capecitabine in chemotherapyna¨ve patients of ECOG performance status 0 to 2 with ı histologically proven oesophago-gastric cancer. cisplatin (C). cisplatin and continuous infusional 5-FU) improved survival compared to surgery alone. median OS 9. Further interesting data have come from the REAL II trial. The four-arm trial evaluated epirubicin (E). prospective randomised controlled trials of postoperative adjuvant chemotherapy have not shown a significant effect on survival (Shimada and Ajani. presented in abstract form in 2006. Palliative chemotherapy Randomised trials comparing palliative chemotherapy with best supportive care have shown a significant improvement in median survival (Glimelius et al.009. Several meta-analyses have shown that.

Michael Button and Tom Crosby

analysis of toxicity, health, economic, and quality-of-life data. ECF regimen: r Epirubicin 50 mg/m2 , day 1. r Cisplatin 60 mg/m2 , day 1. r 5-FU 200 mg/m2 per day, days 1 to 21 (or capecitabine 625 mg/m2 p.o. b.d., days 1 to 21). r Antiemetics: r Dexamethasone 8 mg i.v. then 4 mg p.o. b.d. for 3 days. r 5-HT3 antagonist i.v. r Metoclopramide 10 mg q.d.s. p.r.n. for 10 days.

External beam: technique Patient preparation involves nutritional support, which is very important. Patients should be consuming over 1500 kcal/day before starting treatment; however, this is often difficult when more than three-quarters of the stomach has been resected. Differential renal function should be measured because one kidney often receives an abovetolerance dose of radiation. The patient should be immobilised and positioned supine with arms raised and supported. The planning CT should extend from the sternal notch to L3 to generate adequate dose–volume histograms for organs at risk. The target volume is defined using pre- and postoperative CT scans (ideally, diagnostic quality with oral and i.v. contrast as well as a non-contrast planning scan), a pathology report, an operative note and ideally a personal discussion with the surgeon to identify the areas at highest risk of recurrence. The placing of radio-opaque clips can help. The treatment volume will depend on tumour factors (e.g. proximal or distal location, stage and sites of involved lymph nodes) and the type of operation (total or partial gastrectomy) but needs to cover the tumour bed, gastric remnant (if applicable), anastomosis, duodenal stump (distal resection margin), and regional lymphatics (varying depending on those most at risk). Therefore, with respect to the target volume, the tumour bed, anastomosis, and residual gastric remnant should be adequately covered. The nodal areas at risk include the gastric and gastro-epiploic (usually resected with primary); coeliac nodes, porta hepatis, subpyloric, gastroduodenal, splenic-suprapancreatic and retropanceaticoduodenal nodes. (For a more detailed discussion about this treatment technique, see Smalley et al., 2002.) Margins of 1 cm around the CTV should usually be adequate to generate a PTV that accounts for setup errors and organ motion. In the Macdonald Intergroup 0116 study, anterior– posterior fields were used throughout, giving a dose to the cord of approximately 48 Gy and higher doses to the small bowel. CT planning and individually tailored field arrangements may reduce the dose to organs outside the PTV. The typical field size (superior–inferior extent) is T8/9 or T9/10 interspace to L1/2 interspace (for proximal tumours) or L3/4 interspace (distal tumours), a maximum of 20 cm × 20 cm. For verification, all patients should have a simulator check film prior to starting treatment. Portal images

Second-line palliative chemotherapy
It is not clear whether second-line chemotherapy at relapse following a response to previous chemotherapy is worthwhile. It should only be considered as part of a clinical trial and will be the subject of a UK NCRN trial comparing docetaxel with supportive care. Secondline chemotherapy after a failure to respond to first-line treatment is rarely effective.

Radiotherapy and chemoradiotherapy
Adjuvant chemoradiotherapy
The very high local recurrence rates after radical surgery mean that adjuvant radiotherapy is an important option. The US Intergroup Study 0116 (Macdonald et al., 2001) showed significant improvements in rates of both disease-free and overall survival for patients treated with adjuvant radiotherapy (45 Gy in 5 weeks) and concomitant 5-FU/folinic acid compared to surgery alone. As a result, adjuvant chemoradiotherapy has become a standard therapy in the USA for high-risk patients. However, only 10% of patients had extensive (D2) resection, 30% did not complete the chemoradiation because of toxicity, and more than 30% of the radiotherapy plans had significant errors. This treatment has not been adopted as a standard therapy in the UK or large parts of Europe. Its benefit following more extensive surgery and following preoperative chemotherapy is not known and it should only be carried out in specialist centres with appropriate experience. Despite its occasional use in practice, the principles of radiotherapy planning of gastric tumours should be known. If it is given as an adjuvant treatment, radiotherapy should start within 10 weeks of surgery when possible.

138

Stomach

Table 10.7. Side effects of chemoradiotherapy for gastric cancer
Side effect Acute Nausea Fatigue Diarrhoea Myelosuppression Late Myelopathy The risks of these late sequelae should be kept to a minimum with careful planning Malabsorption Radiation enteritis Antiemetics including 5-HT3 if required Advice re management of fatigue, check haemoglobin Low residue diet, loperamide as required Weekly FBC Comments

trolling pain or bleeding. Treatment is usually well tolerated but can cause short-lived nausea and vomiting, abdominal cramps, or diarrhoea, many of which are prevented by the routine use of 5-HT3 antagonists.

Other palliative treatments
Endoscopic laser photo-coagulation can be useful for controlling bleeding or for debulking a large, obstructing tumour. Expandable metal stents may relieve dysphagia due to gastro-oesophageal tumours. Occasionally, palliative surgical bypass may be required for gastric outlet obstruction or even gastrectomy for uncontrollable bleeding. A coeliac plexus block may palliate severe pain.

Prognosis
Because patients usually present late, the overall 5-year survival is less than 10%, but it is as high as 70% for early (T1) tumours. The median survival for patients with unresectable (or untreated metastatic) disease is around 4 months. However, even in patients with resectable disease, locoregional recurrence and distant metastases are common (see Table 10.6).

should be obtained on at least the first 2 days of treatment and thereafter once weekly. With regard to dose, fractionation and energy, isocentric treatment should be given with a minimum energy of 6 MV and source-axis distance of 100 cm, giving 45 Gy in 25 fractions over 5 weeks to the entire tumour bed, anastomoses and regional lymph nodes in a single phase, treating all fields each day. MLCs, wedges and dual asymmetric collimators are recommended. Doselimiting structures include: r Kidneys – at least three-quarters of one kidney should receive ≤ 20 Gy. r Heart – no more than 30% should receive > 40 Gy. r Liver – no more than 60% should receive > 30 Gy. r Spinal cord – no part should receive > 45 Gy.

Prognostic factors
Prognostic factors include tumour stage at presentation, resectability, morphology (diffuse types have worse prognosis) and poor tumour differentiation.

REFERENCES
Ajani, J. A. (2005). Evolving chemotherapy for advanced gastric cancer. Oncologist, 10 (Suppl. 3), 49–58. AJCC. (2002). AJCC Cancer Staging Manual, ed. F. L. Green et al., 6th edn. New York: Springer. Borrman, R. (1926). Geschwulste des magens und duodenums. In Handbuch der Speziellen Pathogischen Antomie und Histologie, Ed. F. Henske and O. Lubarsch. Berlin: Julius Springer, IV-L864–71. Correa, P (1995). Helicobacter pylori and gastric carcinogenesis. . Am. J. Surg. Path., 19 (Suppl. 1), S37–43. Cunningham, D., Allum, W. H., Stenning, S. P et al. (2006a). . Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N. Engl. J. Med., 355, 76–7. Cunningham, D., Rao, S., Starling, N. et al. (2006b). Randomised multicentre phase III study comparing capecitabine with fluorouracil and oxaliplatin with cisplatin in patients with advanced oesophagogastric (OG) cancer: The REAL 2 trial. J. Clin. Oncol., 2006 ASCO Meeting Proceedings Part I, Vol. 24, No. 18S (June 20 Suppl.), LBA4017.

Concurrent chemotherapy
A regimen of 5-FU/folinic acid was used in the Intergroup Study 0116 but this regimen is probably suboptimal; infusional 5-FU, oral fluoropyrimidines or a combination regimen may be better, and clinical trials are under way to confirm this. Table 10.7 shows the toxicity from chemoradiotherapy.

Palliative radiotherapy
Palliative radiotherapy (30 Gy in 10 fractions or an 8 Gy single fraction, given by antero-posterior fields, simulated with barium contrast) can be very effective in con-

139

Michael Button and Tom Crosby

Earle, C. C. and Maroun, J. A. (1999). Adjuvant chemotherapy after curative resection for gastric cancer in non-Asian patients: revisiting a meta-analysis of randomized trials. Eur. J. Cancer, 35, 1059–64. Glimelius, B., Ekstrom, K., Hoffman, K. et al. (1997). Randomized comparison between chemotherapy plus best supportive care with best supportive care in advanced gastric cancer. Ann. Oncol., 8, 163–8. Gunderson, L. L. and Sosin, H. (1982). Adenocarcinoma of the stomach: areas of failure in a re-operation series (second or symptomatic look) clinicopathologic correlation and implications for adjuvant therapy. Int. J. Rad. Oncol. Biol. Phys., 8, 1–11. Hallisey, M. T., Dunn, J. A., Ward, L. C. et al. (1994). The second British Stomach Cancer Group trial of adjuvant radiotherapy or chemotherapy in resectable gastric cancer: five year follow-up. Lancet, 343, 1309–12. Hermans, J., Bonenkamp, J. J., Boon, M. C. et al. (1993). Adjuvant therapy after curative resection for gastric cancer: meta-analysis of randomized trials. J. Clin. Oncol., 11, 1441–7. Jansen, E., Boot, H., Verheij, M. et al. (2005). Optimal locoregional treatment in gastric cancer. J. Clin. Oncol., 23, 4509–17. Japanese Gastric Cancer Association. (1998). Japanese classification of gastric carcinoma – 2nd English edition. Gastric Cancer, 1, 10–24. Kao, G., Whittington, R. and Coia, L. (1993). Anatomy of the celiac axis and superior mesenteric artery and its significance in radiation therapy. Int. J. Rad. Oncol. Biol. Phys., 25, 131–4. Lim, L., Michael, M., Mann, G. B. et al. (2005). Adjuvant therapy in gastric cancer.J. Clin. Oncol., 23, 6220–32. Macdonald, J. S., Smalley, S. R., Benedetti, J. et al. (2001). Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastro-esophageal junction. N. Engl. J. Med., 345, 725–30.

McCulloch, P., Nita, M. E., Kazi, H. et al. (2003). Extended versus limited lymph nodes dissection technique for adenocarcinoma of the stomach. Cochrane Database Syst. Rev., 18, CD001964. Moiseyenko, V., Ajani, J., Tjulandin, S. et al. (2005). Final results of a randomised phase III trial (TAX 325) comparing docetaxel (T) combined with cisplatin (C) and 5-fluorouracil (F) to CF in patients (pts) with metastatic gastric adenocarcinoma (MGA). J. Clin. Oncol., 23, 308s. Murad, A. M., Santiago, F. F., Petroianu, A. et al. (1993). Modified therapy with 5-fluorouracil, doxorubicin, and methotrexate in advanced gastric cancer. Cancer, 72, 37–41. Pyrhonen, S., Kuitunen, T., Nyandoto, P. et al. (1995). Randomised comparison of fluorouracil, epidoxorubicin and methotrexate (FEMTX) plus supportive care with supportive care alone in patients with non-resectable gastric cancer. Br. J. Cancer, 71, 587–91. Shimada, K. and Ajani, J. A. (1999). Adjuvant therapy for gastric carcinoma patients in the past 15 years: a review of Western and Oriental trials. Cancer, 86, 1657–68. Smalley, S., Gunderson, L., Tepper, J. et al. (2002). Gastric surgical adjuvant radiotherapy consensus report: rationale and treatment implementation. Int. J. Radiat. Oncol. Biol. Phys., 52, 283–93. Sumpter, K., Harper-Wynne, C., Cunningham, D. et al. (2005). Report of two protocol planned interim analyses in a randomised multicentre phase III study comparing capecitabine with fluorouracil and oxaliplatin with cisplatin in patients with advanced oesophagogastric cancer receiving ECF. Br. J. Cancer, 92, 1976–83. Webb, A., Cunningham, D., Scarffe, J. H. et al. (1997). Randomized trial comparing epirubicin, cisplatin, and fluorouracil versus fluorouracil, doxorubicin, and methotrexate in advanced esophagogastric cancer. J. Clin. Oncol., 15, 261–7.

140

11

LIVER , GALLBLADDER AND BILIARY TRACT
Somnath Mukherjee and Tom Crosby

Introduction
Primary liver cancer is one of the most common cancers worldwide, and it predominantly affects people in developing countries. It is often associated with chronic liver infections and it is more common in males. Patients usually present with advanced disease and treatment options are influenced, and often limited, by co-morbidities, especially poor function of the rest of the liver. Tumours of the gallbladder and biliary tract are relatively rare. Patients often present late with symptoms of biliary obstruction which, together with cholangitis, is a common cause of morbidity and death and is the main target for palliative therapies. Gallbladder and biliary tract tumours are moderately chemosensitive. Cytological or histological confirmation of disease is often difficult, and specialist multidisciplinary teams of expert radiologists and pathologists should be involved in the diagnosis and staging. Radical surgery should only be carried out by tertiary surgical teams, and patients should be managed, whenever possible, within clinical trials.

The division between the lateral and medial segments is the gallbladder and IVC and not the falciform ligament. The right part of the liver is made up anteriorly of segments V and VIII (inferior and superior, respectively) and posteriorly of segments VI and VII (inferior and superior, respectively). The latter is the right lateral surface as seen on a CT scan. Contrast in the portal, hepatic, and inferior caval veins on CT allows distinct segmental definition.

Hepatocellular carcinoma
Incidence and epidemiology
Primary hepatocellular carcinoma (HCC) is the eighth most common solid tumour worldwide. Although the incidence is low in the UK and other developed countries, it is rising because of the increase in viral hepatitis. The UK annual incidence rate is 1.8 in 100 000 with 3000 new cases reported per year. The number of deaths is sometimes slightly higher than the incidence rate; this anomaly is probably due to the poor long-term survival and misdiagnosis of secondary malignancies as the primary disease. The male-to-female ratio is 3:1. Areas of the world that are at high risk include East and Southeast Asia and Sub-Saharan Africa. Countries such as Taiwan, China, Malaysia, Japan and those in middle Africa have the highest incidence rates (up to 36 in 100 000) because of the high incidence of hepatitis.

Tumours of the liver
The types of tumour that affect the liver are shown in Table 11.1.

Anatomy of the liver
The liver is divided into right and left lobes by the falciform ligament, but more importantly, in terms of surgical resection, a segmental division can be made and seen with imaging based on the relationship to the hepatic and portal veins. There are four segments in both the left and the right liver. The left liver consists of the caudate lobe (segment I), lateral segments II and III (superior and inferior lateral, respectively, seen extending to the left surface on a CT scan), and medial segment IV.

Risk factors and aetiology
The main causative factors in developing HCC appear to be both chronic liver-cell injury and inflammation; both factors can be additive or synergistic. For instance, chronic hepatitis C virus (HCV) infection is more commonly associated with HCC in chronic alcohol drinkers, and aflatoxin interacts with hepatitis B virus (HBV) infection to increase the risk of HCC threefold.

141

Somnath Mukherjee and Tom Crosby

Table 11.1. The range of tumours affecting the liver
Type Benign Examples Haemangioma Focal nodular hyperplasia Nodular regenerative hyperplasia Hepatic adenoma Hepatic cystadenoma Lipoma Hamartoma Bile duct cystadenoma Malignant primary Hepatocellular (conventional, fibrolamellar) Cholangiocarcinoma (intrahepatic) Mixed hepatocellular-cholangiocarcinoma Hepatoblastoma Hepatic angiosarcoma Malignant secondary Any tumour Adenocarcinoma (e.g. bowel, breast, lung) Sarcoma Lymphoma Carcinoid

Table 11.2. Pathological features of hepatocellular carcinoma
Features Macroscopic Description Yellowish-white nodules with areas of haemorrhage or necrosis; background features of cirrhosis Microscopic Large cells similar to hepatocytes with clear cytoplasm cf. renal cell carcinoma In the fibrolamellar variant the tumour forms cords with collagen strands

Pathology
Tumours may appear as a large solitary lesion (30%) or they may be multicentric (60%). Even what appear to be solitary tumours may have multiple satellite lesions around a central tumour. The pathological features of HCC are shown in Table 11.2.

Clinical presentation
HCC is usually asymptomatic and patients present only rarely with metastatic disease. HCC is often found incidentally in someone with chronic liver disease (hepatitis or cirrhosis) and it may be picked up on imaging or with a sudden rise in alphafetoprotein (αFP), although as many as 30% of tumours do not produce αFP. Patients may present with decompensation of previously stable chronic liver failure with features such as ascites, jaundice, anorexia, GI bleeding, weight loss and encephalopathy. Occasionally, patients may have paraneoplastic syndromes such as hypoglycaemia, hypercalcaemia, erythrocytosis and ectopic gonadotrophic or adrenocorticotrophic hormone. Metastases to lung, bones and adrenal and lymph nodes are rare.

Infective
HCC is associated with HBV and HCV infection. Chronic hepatitis B and C, with or without cirrhosis infection, multiplies the risk of HCC by 100. In Europe the main factor in developing HCC is HCV infection.

Inflammatory
HCC is also associated with other causes of cirrhosis such as hereditary haemochromatosis, Wilson’s disease and type 1 glycogen storage disease.

Chemical injury
Exposure to a number of chemical agents increases the risk of HCC. Alcohol is the most common; others include nitrites, hydrocarbons, solvents and polyvinyl chloride (PVC), which is particularly associated with hepatic angiosarcoma. Aflatoxins, hepatotoxic agents produced by the fungi Aspergillus flavus and A. parasiticus, are causative factors in Africa and Asia.
142

Diagnostic and staging investigations
In addition to diagnosis and staging, investigations are aimed at assessing the patient’s suitability for treatment, particularly, the level of function of the ‘normal’ or nonmalignant liver:

Liver, gallbladder and biliary tract

r Medical and family histories should be noted (prer r r r
vious liver disease, drug exposure, symptoms) and examination (liver failure) should be performed. A full blood count should be performed, along with renal, liver and bone profiles. HBV and HCV serology and αFP levels should be checked; blood clotting tests should be performed and serum albumin level taken. The indocyanine green retention rate at 15 minutes determines the adequacy of liver function for surgery. A biopsy should only be done if patients are inoperable. For patients in whom surgery may be possible, biopsy is not performed so that possible tumour seeding can be avoided. Ultrasound, arterial phase CT, MRI and CT angiography all play a role and such specialist investigations should only be carried out by a specialist multidisciplinary team.

Table 11.3. TNM staging classification of hepatocellular carcinoma
Stage T1 T2 Description Solitary tumour without evidence of vascular invasion Solitary tumour with evidence of vascular invasion; or multiple tumours, none more than 5 cm in greatest dimension T3 Multiple tumours more than 5 cm or tumour involving a major branch of the portal or hepatic vein(s) T4 Tumour(s) with direct invasion of adjacent organs other than the gallbladder or with perforation of visceral peritoneum NX N0 N1 MX M0 M1 Regional lymph nodes cannot be assessed No regional lymph node metastasis Regional lymph node metastasis Distant metastasis cannot be assessed No distant metastasis Distant metastasis

r

Staging classification
The TNM and stage groupings are shown in Tables 11.3 and 11.4, respectively.

Treatment overview
For patients who are fit enough and have resectable tumours (T1 to T3, N0, M0 and some T4), surgery is the treatment of choice. Patients who are unfit for surgery, or who have unresectable tumours, may be suitable for palliative systemic treatment, although response rates are very poor.

Adapted from UICC (2002).

Table 11.4. Stage groupings for hepatocellular carcinoma
Stage Description T1 T2 T3 T4 Any T Any T N0 N0 N0 N0 N1 Any N M0 M0 M0 M0 M0 M1

Surgery
Local resection
Local resection of small isolated tumours is possible but, because hepatitis or chronic liver-cell injury or cirrhosis is multicentric, recurrence is often seen. Local resection is the treatment of choice for patients with non-cirrhotic livers and for patients with Child’s A cirrhosis who have a small solitary tumour and minimal portal hypertension (hepatic venous pressure gradient less than 10 mmHg).

I II IIIA IIIB IIIC IV

Adapted from UICC (2002).

Liver transplantation for HCC
The risk of multicentric recurrence is significantly reduced by the use of liver transplantation (Llovet et al., 1999). Risk factors for recurrence are the presence of HBV infection, multiple tumours, tumour size and vascular invasion (Mazzaferro et al., 1996).

r Transplantation treats hepatoma as well as the underr The selection (Milan) criteria for transplantation
require one tumour between 2 and 5 cm or two or three tumours, each 3 cm or smaller; however, selected patients with HCC larger than 5 cm in size have been
143

lying problem (cirrhosis).

Somnath Mukherjee and Tom Crosby

treated with multimodal therapy (chemoembolisation, surgery, systemic chemotherapy) with a report of an impressive 5-year survival rate of 44% (Roayaie et al., 2002). r A shortcoming of transplantation is a lack of donors.

and diseased liver to radiotherapy is limited. Conformal radiotherapy may have a role in conjunction with chemoembolisation (Seong et al., 1999).

Targeted radiotherapy
Radioactive microspheres containing yttrium-90 given by intra-arterial infusion have shown activity in unresectable HCC. A total radiation dose of 120 Gy resulted in better survival than lower doses of radiation (55.9 and 26.2 Gy, respectively; Lau et al., 1994). A technetium lung scan should be done first to rule out significant shunting to the lungs, which contraindicates yttrium-90 treatment.

Local non-surgical therapy
Transarterial chemoembolisation (TACE), hepatic arterial chemotherapy, radiofrequency ablation, ethanol injection and conformal radiotherapy are all options for the palliation of patients with tumours 3 to 5 cm in diameter. There have been reports of survival similar to that in surgical studies and use of these procedures often reflects local experience and expertise. However, it is not clear from the literature whether or not local non-surgical therapy should be the standard approach for inoperable patients, perhaps because patient populations who have received this treatment have been heterogeneous. TACE combines intra-arterial chemotherapy, often doxorubicin and lipiodol (an oily iodised contrast agent administered manually), exploiting the fact that HCCs are highly vascular and predominantly take their blood supply from the hepatic arteries. A meta-analysis that included seven randomised trials of arterial embolisation has shown that TACE is more effective than conservative therapy or 5-FU chemotherapy and shows a significant improvement in 2-year survival (OR = 0.53, 95% CI = 0.32–0.89) and a median survival of more than 2 years (Llovet et al., 2003). Radiofrequency ablation (RFA) has been shown to be more effective than percutaneous ethanol injection, and for tumours up to 2 cm in diameter, an 85% complete response has been reported. For tumours less than 5 cm in size, the overall survival with laparoscopic RFA is comparable to that following surgical resection, although it is associated with higher incidence of local recurrence. Percutaneous RFA is associated with needle-track seeding, which may preclude its use.

Systemic therapy
HCC is a chemoresistant tumour. The following strategies can also be implemented.

Adjuvant therapy
The role of postoperative adjuvant therapy is not well established. The main site of relapse is the residual liver itself, and one randomised study has shown a lower recurrence rate and prolonged survival by giving postoperative intra-arterial iodine-131-labelled lipiodol (Lau et al., 1999). Interferon in patients with hepatitis C in addition to the use of oral retinoids have also been associated with reduced tumour recurrence (Muto et al., 1996; Yoshida et al., 2004).

Palliative single-agent doxorubicin
Although doxorubicin has never been shown to be better than supportive care alone, it is the most tested agent both by itself and in combination with other agents and it is frequently used as the comparator ‘standard’ arm in trials of novel therapies. It has response rates of 15 to 25%.

Combination treatments
Combination therapy has been reported to give a higher response rate but no improvement in survival. Combinations with biological agents such as interferon, for example in the PIAF regimen (cisplatin, interferon, doxorubicin, and 5-FU), again have given response rates of up to 26% but have significant toxicity and unproven survival benefits (Yeo et al., 2005). However, this regimen rendered 9 of 13 unresectable tumours resectable (Leung et al., 1999) and, in an updated analysis, the authors reported 8 pathological complete responses in 15 patients who underwent surgery (Lau et al., 2001).

Radiotherapy
External beam radiotherapy
Radiotherapy has not been used extensively to treat HCC, but the advent of conformal and intensitymodulated radiotherapy may change this. It is important to remember that there is significant movement of the liver (1 to 2 cm) in the superior–inferior direction with respiration, and that the tolerance of both normal
144

Liver, gallbladder and biliary tract

Table 11.5. The Child-Pugh grading system for cirrhosis. Scores of 1 to 3 are given for clinical investigations and features and summed: Child class A = score of 5 to 6; Child class B = score of 7 to 9; Child class C = score greater than 9
Score 1 2 3
a

Bilirubin (μmol/l) ≤ 34 34–50 > 50

Albumin (g/l) > 35 28–35 ≤ 28

Prothrombin time (s) ≤4 4–6 >6

Hepatic encephalopathy None 1–2 3–4

Ascites None Mild Severe

Encephalopathy score determined by degree of consciousness, intellectual/personality impairment, neurological signs and EEG changes. Adapted from Pugh et al. (1973).

Tamoxifen
Despite the fact that 30% of tumours express oestrogen receptors, randomised trials have not shown any benefit from the use of tamoxifen.

in clinical trials and to help compare outcomes from such studies.

Current trials
The UK NCRN trial Hep-1 is a randomised clinical trial that evaluates the benefits of doxorubicin chemoembolisation compared to systemic doxorubicin in patients with unresectable, advanced HCC.

Newer agents
Newer agents that exhibit activity in HCC include gemcitabine (RR = 18%; Yang et al., 2000), capecitabine (RR = 13%; Lozano et al., 2000), erlotinib (RR = 10%; Philip et al., 2004), irinotecan (RR = 7%; O’Reilly et al., 2001), and bevacuzimab (Schwartz et al., 2004).

Angiosarcoma of the liver
Angiosarcomas of the liver are rare tumours but they are notable because of their association with exposure to PVC monomers. They occur 10 to 20 years after exposure and not always in the most heavily exposed individuals. They have also been described in patients who received thorotrast (a diagnostic radioactive contrast agent used between 1930 and 1950). Patients present with a large, often painful, hepatic mass. Treatment consists of surgical resection, where possible, or palliative anthracycline chemotherapy.

Prognosis
Surgical series
The 5-year survival (5YS) rates for surgery are as follows: r Stage I – 60%. r Stage II – 45%. r Stage III – 20%. r Stage IV – 10%. The 5-year survival after resection is 41 to 74% (Fong et al., 1999; Llovet et al., 1999), whereas the 5-year survival after transplantation (following Milan criteria) is 70 to 80% (Bismuth et al., 1999; Llovet et al., 1999).

Carcinoma of the gallbladder and biliary tree
Types of tumour
The types of tumour affecting the gallbladder and biliary tree are shown in Table 11.6; more than 80% are adenocarcinomas.

Non-surgical series
The overall 5-year survival for non-surgical series is 10% or less. The prognosis of patients with HCC is significantly influenced by the severity of cirrhosis present in the non-malignant liver. This can be graded according to the Child-Pugh classification (see Table 11.5); for example, patients with Child’s C cirrhosis with HCC have a median survival of approximately 3 months. This system is also used for patient selection and stratification

Incidence and epidemiology
In England and Wales, 1200 new cases are reported each year. The 1-year survival is 22%, and the 5-year survival is 5 to 10%; the male-to-female ratios are 2:3 and 1:1
145

Somnath Mukherjee and Tom Crosby

Table 11.6. Types of tumour affecting the gallbladder and biliary tree
Anatomical site Carcinoma gallbladder Examples Adenocarcinoma Adenosquamous Squamous cell Small cell Carcinoma biliary tree Cholangiocarcinoma Nodular Diffuse Papillary

Table 11.7. Bismuth classification for perihilar tumours
Type Type I Type II Type III Type IV Description Tumour is below confluence of right and left ducts Confined to the confluence Extension into right or left hepatic ducts Extension into right and left hepatic ducts or multicentric Adapted from Bismuth and Corlette (1975).

for gallbladder cancer and cholangiocarcinoma, respectively. The cases are distributed as follow: r Gallbladder cancer, 40%. r Cholangiocarcinoma (intra- and extrahepatic), 43%. r Periampullary, 13%. r Others, 4%.

Left hepatic duct Right hepatic duct Common hepatic duct Common bile duct Gallbladder Middle third Upper third including hilus

Risk factors and aetiology
Carcinoma of the gallbladder
Risk factors for gallbladder cancer include: r Obesity. r Gallstones (especially those larger than 3 cm). r Chronic typhoid and paratyphoid carriers. r Polyps (greater than 1 cm in diameter). r Ulcerative colitis.

Duodenum Pancreas

Lower third

Figure 11.1. Locations of the upper, middle and lower thirds of the extrahepatic biliary tree (as described in Chamberlain and Blumgart, 2000). Perihilar tumours are classified further according to the Bismuth classification (see Table 11.7).

Cholangiocarcinoma
Risk factors for cholangiocarcinoma include r Primary sclerosing cholangitis (lifetime risk of 10%), which can be associated with ulcerative colitis. r Clonorchis sinensis infestation. r Polycystic liver disease, gallstones (somewhat weaker correlation than with carcinoma of gallbladder). r Choledochal cyst. r Caroli’s disease, a rare congenital condition of multiple saccular dilatations of the intrahepatic bile ducts. r Chemical carcinogens (aflatoxin, vinyl chloride, methylene chloride, thorotrast).

Clinical presentation
The most common presentation of patients with cholangiocarcinoma is obstructive jaundice (gallbladder, bile duct). Patients also present with fluctuating jaundice (periampullary carcinoma) and constitutional symptoms such as weight loss (> 50% of cases), anorexia, and fatigability. Hepatomegaly and pain in the right upper quadrant (especially for hilar tumours and carcinoma of the gallbladder) also occur. A palpable, nontender gallbladder can indicate an obstructive tumour below the level of the cystic duct (Courvoisier’s law), also gallbladder carcinoma. Symptoms and signs of metastasis include ascites and pleural effusion. Carcinoma

Anatomy
The anatomy of the gallbladder and main bile ducts is shown in Figure 11.1. The Bismuth-Corlette classification is shown in Table 11.7.
146

Liver, gallbladder and biliary tract

of the gallbladder can also present as acute or chronic cholecystitis; incidental diagnosis of the cancer is made at surgery.

Clinical examination LFT: obstructive picture US shows dilated bile duct

Investigations
Blood tests
Following a full blood count, renal and liver profiles and test for CA19–9 tumour marker, LFT derangement consistent with obstructive jaundice (raised alkaline phosphatase, gamma-glutamyl transferase, bilirubin) and persistent elevation of CA19–9 despite decompression of the biliary tree suggest biliary duct/pancreatic malignancy. CA19–9 is elevated in up to 85% of patients with cholangiocarcinoma.
Proximal obstruction Distal obstruction

PTC Cytology ± stent

ERCP / MRCP Cytology ± stent

Figure 11.2. Investigation strategy for a patient with obstructive jaundice. ERCP = endoscopic retrograde cholangiopancreatogram; LFT = liver function tests; MRCP = magnetic resonance cholangiopancreatogram; PTC = percutaneous transhepatic cholangiograph; US = ultrasound.

Imaging
Ultrasonography is the first-line investigation for suspected biliary tract obstruction; it is used to find intrahepatic duct dilatation. Extrahepatic ducts may be dilated in carcinoma of the distal bile duct/pancreatic head tumours. Ultrasonography is useful in ruling out choledocholithiasis. A CT scan is useful in assessing the extent of local disease and regional lymphadenopathy (gallbladder, intrahepatic and perihilar cholangiocarcinoma) and to rule out metastatic disease. It can establish resectability in approximately 60% of cases. ERCP and PTC are used to image the intra- and extrahepatic bile ducts and to localise the site of obstruction. PTC is useful for proximal tumours (perihilar cholangiocarcinomas), whereas ERCP is most useful for distal bile duct tumours. Although invasive, the procedures allow for obtaining a cytological diagnosis (positive in 30%), biopsy, and insertion of endobiliary stents. EUS is useful for imaging the gallbladder, distal extrahepatic bile duct and lymph nodes, and it facilitates a guided biopsy. Laparoscopy rules out peritoneal metastases before curative surgery. MRI and magnetic resonance cholangiopancreatogram (MRCP) are non-invasive investigations that are used to evaluate the biliary tree in suspected cholangiocarcinoma.

adjacent liver, vessels and lymph nodes) and they detect liver metastasis.

Disadvantage
A disadvantage is that imaging does not give histological/cytological diagnosis. PET scanning is under investigation and has been shown to detect cholangiocarcinomas that are more than 1 cm in diameter and to detect metastases (not detected by other imaging) in 30% of cases (Anderson et al., 2004). Figure 11.2 shows an investigation strategy for a patient with obstructive jaundice.

Treatment overview
Fewer than 20% of patients have resectable tumours at the time of diagnosis and there is a high risk of recurrence after surgery. The success of adjuvant treatment is unproven and treatment should only be considered in the context of a clinical trial. Radiotherapy and chemoradiation, both as adjuvant and primary therapy, should be regarded as investigational and are not widely practised in the UK. Palliative chemotherapy is an option for fitter patients (performance status 0 to 2) with incurable disease although there is no randomised trial showing a survival advantage of palliative chemotherapy over that of active supportive care. A plastic or metallic stent may be useful in palliating obstructive jaundice, a common feature of the disease. Given these uncertainties about management, patients should always be considered for clinical trials.
147

Advantages
The advantages of imaging are that the procedures are non-invasive, they show the extent of duct involvement (as in MRCP), they indicate local staging (invasion of

Intrahepatic cholangiocarcinoma requires a resection of involved segments/lobes.5% (MaindraultGoebel et al. 2003). Acceptable treatment options include the following: 1. Single-agent 5-FU (modified by leucovorin) has been reported to produce response rates of up to 32% (Choi et al. mitomycin. lymph nodes and Roux-en-Y hepaticojejunostomy. weeks 1 to 7. There is a high risk of local recurrence at the porta hepatis and of distant metastases. Metallic stents have a lower rate of complications (e. The combination of gemcitabine and oxaliplatin has demonstrated a response rate of 35.. Gemcitabine 1000 mg/m2 . 2001). Other drugs that seem to be active in cholangiocarcinoma include epirubicin.. Advanced stages require radical cholecystectomy. an infusional regimen through a central access device is more commonly used in the UK. 2004). Plastic stents are placed if surgery is planned or if the prognosis is poor. then 1 week off. Surgery is an option for perihilar tumours: r Bismuth classifications I and II require en bloc resection of the bile duct. Radical surgery for carcinoma of the bile duct Hepaticojejunostomy is usually possible for tumours below the first division of the left or right main duct with uninvolved vessels. Oral capecitabine has also shown activity in biliary tract tumours. 5-FU (modified by leucovorin) with or without mitomycin C. r Bismuth III is as above. gallbladder. Chemotherapy Adjuvant chemotherapy One randomised trial has shown there is no survival advantage obtained from giving adjuvant chemotherapy using infusional 5-FU and mitomycin followed by a prolonged course of oral fluoropyrimidine over surgery alone (Takada et al. 2005) and mitomycin (Kornek et al. Cisplatin and 5-FU. Palliative chemotherapy There is no evidence from randomised trials that chemotherapy is superior to supportive care alone. 2001) and one randomised phase II trial showed superior survival and response rate for the addition of cisplatin to 5-FU (median survival 7. A pilot study has shown a drop in bilirubin levels and an increase in survival rates that compares favourably to historical series (Ortner. docetaxel. either as a single agent or in combination with cisplatin (Kim et al. 148 Stent insertion Biliary obstruction and cholangitis are common causes of morbidity and death. The addition of cisplatin may enhance the efficacy of gemcitabine and this question is being addressed in the ABC 02 study. from infection or blockage) but are more expensive. plus right or left hemihepatectomy. which involves nodal dissection and may entail excision of adjacent liver tissue. Gemcitabine monotherapy is associated with response rates of 22 to 45% and a median survival of 6 to 20 months. 2000). 2002). subsequent cycles weekly for 3 out of 4 weeks. Photodynamic therapy Photodynamic therapy involves injection of a photosensitiser followed by direct endoscopic illumination of the tumour using light of a specific wavelength.. release of oxygen free radicals and tumour death. Laparoscopic procedures are contraindicated in patients with known carcinoma because of the high risk of needletrack seeding. r Bismuth IV is as Bismuth I and II. The combination of 5-FU and mitomycin has been reported to give a response rate of 26% (Chen et al. 3..8 versus 5.. which is evaluating the role of adjuvant capecitabine in patients with completely resected biliary tract cancer. Surgery for the lower third involves a pancreaticoduodenectomy.. gemcitabine. which is randomising patients with inoperable biliary tract carcinoma to single-agent gemcitabine with or without cisplatin. Although most trials have used bolus 5-FU. 2. carcinoma in situ and tumours limited to the muscle layer may be effectively treated by cholecystectomy alone. plus extended right or left hemi-hepatectomy. which causes activation of the photosensitising compound. 2002). interferon and the novel tumour antibiotic rebeccamycin. 19 versus 7%. gemcitabine (Cho et al.g.3 months. oxaliplatin. 2003). Mitry et al. Patients should be considered for entry into clinical trials like BILCAP.Somnath Mukherjee and Tom Crosby Surgery Radical surgery for carcinoma of the gallbladder Carcinoma of the gallbladder is found incidentally in fewer than 1% of cases of routine cholecystectomies for cholelithiasis.. Patients should be prescribed ursodeoxycholic acid in ..

et al. Ho. Clin. Seo. Intrahepatic cholangioenteric anastomosis in carcinoma of the hilus of the liver. L. (2003). Llovet. and Adam. Effects of 5-fluorouracil and leucovorin in the treatment of pancreatic-biliary tract adenocarcinomas. Rice.. et al. (1999). R. W.. (1999). 1434–40. S.. A randomised phase II trial of weekly high-dose 5FU (HD-FU) with and without folinic acid (FA) and cisplatin (P) in patients (pts) with advanced biliary tract carcinoma: the EORTC 40955 trial. Y. Leung.. Jarnagin. (1999). Final results. Prognosis Prognosis after surgery Lymph node metastasis. Leung.. Hepatology. W. (2002).. 229. Bismuth. Chen. Kim. 19. PTC can be attempted with a view to external biliary drainage and thereafter by placing the stent internally across the stricture.. Lin. V. (2003). Llovet. et al. et al. 693–9. 5.... A phase II study of gemcitabine and oxaliplatin (GEMOX) in advanced biliary adenocarcinoma (ABA). Surg. Pre-operative systemic chemoimmunotherapy and sequential resection for unresectable hepatocellular carcinoma. J. T.. positive resection margin and perihilar tumours are all associated with a poorer prognosis. Med. 236–41. Oncol. C. J. W. 2003). S. Lau. Prevention of second primary tumors by an acyclic retinoid. 22.. randomisation is between biliary stenting with or without photodynamic therapy. 170–8. The PHOTOSTENT 2 trial is for patients with inoperable biliary tract cancers. Patt. Schuell. M. T... A. Soc. Phase II study of capecitabine plus cisplatin as first-line chemotherapy in advanced biliary cancer. Mitomycin C in combination with capecitabine or biweekly high-dose gemcitabine in patients with advanced biliary tract cancer: a randomised phase II trial. J. (2000). Ann. E. Kornek. Choi. M. E. Am. Mitry. R. B. Obstet.. Am. W. Am. polyprenoic 149 . E. D. 2753–8. In one series a 5-year survival rate of 43% was reported for patients who have resected intrahepatic cholangiocarcinoma (Nakagohri et al. 425–8. 14. 1907–17.. H. Majno. 90–7. Surg. Fluorodeoxyglucose PET imaging in the evaluation of gallbladder carcinoma and cholangiocarcinoma. N.. I. et al.. On progression. SWOG 0202 is a phase II study of gemcitabine and capecitabine in patients with unresectable locally advanced or metastatic gallbladder cancer or cholangiocarcinoma. Maindrault-Goebel.. Proc. Abstr. Surg. M. (1999). Surg. Sun. In the BILCAP trial. patients with operable biliary tract cancer are randomised to surgery alone or to surgery and adjuvant capecitabine chemotherapy. et al. Cancer. J. 21. Oncol. 15.. W. W. patients with advanced inoperable biliary tract cancers are randomised to single-agent gemcitabine or a gemcitabine and cisplatin combination. 55–66. J.. S.. et al. Br. 478–83. Ann. Burroughs. W. O. for hepatocellular carcinoma. H. F. Oncol. Soc. J. C. Rosmorduc. et al. 70. and Bruix. 311–22. Bismuth. Chamberlain. Engl. (2001).. M.. F. 12. Pinson.. Oral capecitabine (Xeloda) for the treatment of hepatobiliary cancers (hepatocellular carcinoma. Clin. Lau. 797–801. Hassan. Clin. B. Lozano. Chang. H. Fong. Clin. M.. S. W. Liver Dis. Lancet. Liver transplantation . Kang. Y. V. 19. Capecitabine combined with gemcitabine (CapGem) as first-line treatment in patients with advanced/metastatic biliary tract carcinoma. (1975). An analysis of 412 cases of hepatocellular carcinoma at a Western center. J.. van Custem. R. et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. J. J. J.. J. Clin. (1996). 339–43.. (1999). Y. (1994). C. 353. 790–800. S. M. cholangiocarcinoma and gallbladder cancer).. Am... Lau. Y. F. 334. Intention-to-treat analysis of surgical treatment for early hepatocellular carcinoma: resection versus transplantation.. Selle. (2005). W. et al. K. Surg. Cancer Res. R. W. et al. (2004). Ann. H. Leung. H. Chang. Lau. Paik.. perineural invasion. Cho. K. et al. Y. S.. (2000). Hepatocellular carcinoma. REFERENCES Anderson. Van Laethem. P E. (2000). Oncol. Y. Lai. 104. Moriwaki. Mazzaferro. W. Cancer. G. Y. Y. If the obstruction cannot be relieved endoscopically from below. 696. Oncol. 1025. and Blumgart. T... Abstr. Ann. B. Regalia. et al. 362... Abstr. Ho. C. Gynecol. et al. Y. J. Doci. H. (2001). 1676–81. J... Z. Patt. T. 7. Leung. Hilar cholangiocarcinoma: a review and commentary. R.. Proc. Gastrointest. Complete pathological remission is possible with systemic combination chemotherapy for inoperable hepatocellular carcinoma. Mitomycin C with weekly 24-h infusion of high-dose 5-fluorouracil and leucovorin in patients with biliary tract and periampullar carcinomas.. Muto. Oncol. Ann. 23. T. (2003). Semin. Adjuvant intra-arterial iodine-131-labelled lipiodol for resectable hepatocellular carcinoma: a prospective randomised trial. M. 140. Y. Treatment of inoperable hepatocellular carcinoma with intrahepatic arterial yttrium-90 microspheres: a phase I and II study. 30. Ninomiya. 8.. et al. Choi. Lancet. and Corlette. 1115–20. Oncol.. Laengle. Y. patients are eligible for the ABC 02 study.. 1178. D. (1996)... H.. M. 233. Jan. et al. gallbladder and biliary tract an attempt to keep the stent open.Liver. Anticancer Drugs. Y. T. Proc. J. and Bruix. H. Fuster. Y. Z. Y. L. 994–9. Ongoing trials in biliary tract cancer In the ABC 02 study. W. Soc. (2004).

T. Yeo. N. Roayaie. et al.. 89. S. ed. 393–7. 1532–8. Schwartz. Cancer. K. Frischer. R. erlotinib (OSI-774) in patients with hepatocellular or biliary cancer. K. M. 750–6. pp. S. L. New York: Wiley-Liss. T. Transection of the oesophagus for bleeding oesophageal varices.. 1685–95. Goldman. E. Surg. 53... in patients with hepatocellular carcinoma. (2002). Murray-Lyon. Takada. (2000). 4088. Cancer. (2004).. J. V. Oncol. Asano. J. Yang. (2001).Somnath Mukherjee and Tom Crosby acid. Tateishi. (2005). 235. Surg. Y. Benefits of interferon therapy in hepatocellular carcinoma prevention for individual patients with chronic hepatitis C. I. J. (1999). H.. 2004 ASCO Meeting Proceedings (Post-Meeting Edition). M. O’Reilly. et al. Clin.. J. et al. E.. H. 101–5. Phys. 137–9. Philip. Cancer Inst. Natl. 43. J. 8. H. J. H. (1973). 2004 ASCO Meeting Proceedings (Post-Meeting Edition). et al. et al. Wittekind.. B. 1561–7. Zee. 22 (no. S. Is postoperative adjuvant chemotherapy useful for gallbladder carcinoma? A phase III multicenter prospective randomized controlled trial in patients with resected pancreaticobiliary carcinoma. H. et al. H. 425–30. 22 (14S July 15 Suppl. 95. Phase II trial of . Mahoney. J. Yoshida. Chen. Br... M. Surg. 150 . D. N.. H. et al. Keum. J. J. 6th edn. Cancer. 97. J. (2002). Pugh. Amano. C. Han. Int.. Y. H. Nakagohri.. 334.. Dawson. Sanz-Altamira. T.).. UICC. Thomas. Biol. 27.. Med. A phase II study of irinotecan in patients with advanced hepatocellular carcinoma. Clin. Schwartz. (2004).. Lin. C. Radiat. Oncol.. (2002). 14S July 15 Suppl. T.. J. W. M. et al. T. S. Ortner. Emre. S. 533–9. Yasuda. A randomized phase III study of doxorubicin versus cisplatin/interferon alpha-2b/doxorubicin/fluorouracil (PIAF) combination chemotherapy for unresectable hepatocellular carcinoma.. Ann. Combined transcatheter arterial chemoembolization and local radiotherapy of unresectable hepatocellular carcinoma... Pancreat. (2001). Mok. P. P A. M. K.).. Seong. 289–93.. Hepatobil. Surg. Photodynamic therapy for cholangiocarcinoma. Arakawa. TNM Classification of Malignant Tumours.. World J. R. et al... 646–9. S. 81–3. Long-term results with multimodal adjuvant therapy and liver transplantation for the treatment of hepatocellular carcinomas larger than 5 centimetres. Stuart. (2003). Bevacizumab in hepatocellular carcinoma in patients without metastasis and without invasion of the portal vein. Kinoshita. (2004)... L. J. J. Gut. Sobin and Ch. M. Phase II study of gemcitabine in patients with advanced hepatocellular carcinoma. 60. 4025. Oncol. et al. Engl. 91. Aggressive surgical resection for hilar-invasive and peripheral intrahepatic cholangiocarcinoma.. J. et al.

The types of pancreatic tumour are shown in Table 12. The annual incidence is 8. Chemoradiation is an option for locally advanced disease. invasion or compression of these structures leads to obstructive jaundice and pancreatic insufficiency. body and tail. Pathology The pathological features of carcinoma of the pancreas are shown in Table 12.12 EXOCRINE PANCREAS Somnath Mukherjee and Tom Crosby Introduction Nearly 7000 new cases of pancreatic carcinoma are diagnosed in the UK each year and nearly as many patients die from the disease. and other chemotherapy agents (oxaliplatin and capecitabine). Anatomy The pancreas is a retroperitoneal structure (peritoneum on the anterior surface only) and it lies in front of the first and second lumbar vertebrae. The pancreatic duct and common bile duct pass through the head to open in the ampulla of Vater. benzidine. diet and a previous total gastrectomy. can measure 5 to 10 cm and are often found adjacent to the pancreas in the region of the tail. Lymphatic drainage involves the pancreaticoduodenal. erlotinib and bevacizumab). inherited diabetes mellitus and ataxia telangiectasia syndrome. A Westernised diet (rich in protein and carbohydrates and poor in fruit and fibre) increases the risk. The major risk factors include smoking. Pseudocysts are loculated collections of fluid arising from necrosis. Gemcitabine-based chemotherapy is the mainstay of treatment for fit patients with advanced or metastatic cancers. Carcinoma of the exocrine pancreas Risk factors and aetiology Three percent of pancreatic cancers may be inherited. Most cases of the disease (80%) occur in the head of the pancreas. biological agents (cetuximab. Cancer family syndromes include inherited chronic pancreatitis. There is also a positive association between long-standing diabetes and pancreatic cancer. peak incidence occurs for men in their eighties and for women in their nineties. Pancreatic cancer is the sixth leading cause of cancer death overall: the sixth most common cause of death in men (4% of cancer mortality) and the fifth most common cause of death in women (5% of cancer mortality). but fewer than 20% of patients are suitable for this approach. Surgery is the only curative option. Other risk factors include long-standing diabetes mellitus. They are often solitary. . suprapancreatic. inflammation or haemorrhage.5 in 100 000. It has four parts: head (including uncinate process). Benign cystadenomas occur in elderly women and are found incidentally at autopsy or during other investigations. and DDT exposure.2. Types of pancreatic tumour Benign cysts can be congenital and they arise from anomalous development of the pancreatic ducts. Microcystic and papillary-cystic variants are found in younger women.1. There is a male preponderance toward the disease (male-to-female ratio of 4:3). There has been growing interest in combining gemcitabine with radiotherapy (including newer radiotherapy techniques like IMRT). and cigarette smoking doubles the risk. total gastrectomy (two to five times the risk) and pernicious anaemia. neck. Toxic chemical risk factors are 2-naphthylamine. pyloric and pancreaticosplenic 151 Incidence and epidemiology There are about 7000 new cases of pancreatic cancer in the UK each year and almost as many die from the disease.

the portal vein. the head of the pancreas lies within the C-loop of the duodenum and in close proximity to the stomach and jejunum. r Back pain (spread to retroperitoneum and coeliac plexus). non-Hodgkin lymphoma) More frequently metastases to retropancreatic lymph nodes which mimic pancreatic tumour Table 12. . left upper quadrant for tail lesions).1. r Cholangitis.Somnath Mukherjee and Tom Crosby Table 12. peritoneum. rarely bone) is common at presentation. r Obstructive jaundice. which drain to coeliac and superior mesenteric nodes. Clinical presentation Pain (epigastric for head lesions. Types of pancreatic tumour Type Benign tumours Examples Congenital cysts Pseudocysts Serous cystadenoma Pancreatic cysts Papillary-cystic/microcystic Primary tumours of exocrine pancreas Adenocarcinoma Ductal Acinar Anaplastic Cystadenocarcinoma Adenoacanthoma Squamous cell carcinoma Sarcoma Solid and papillary neoplasms Neuroendocrine tumours of pancreas (see Chapter 36) Islet cell tumour Gastrinoma Glucagonoma VIPoma Carcinoid Somatostatinoma Metastasis Rare (breast. melanoma. r Altered bowel habit (infiltration of colon). the splenic vein and the coeliac artery and its branches. r Blumer’s shelf (peritoneal metastasis in pouch of Douglas. leading to back pain. lung. r Abdominal pain and ascites. the liver and the spinal cord are likely to be included in the radiation fields when planning radiotherapy to the pancreas. anorexia and weight loss are the most commonly presented symptoms. In relation to the surrounding organs. posterior extension of a tumour can involve the first and second coeliac ganglia. which can be palpated rectally). Clinical features of pancreatic cancer nodes. occasional adenosquamous growth pattern Occasionally arise in cysts (cystadenocarcinoma) or from acinar cells (acinar cell carcinoma).2. Pathological features of pancreatic cancers Features Macroscopic Microscopic Description White. Tumour extension to these structures is often a sign of inoperability and can lead to splenic or portal vein thrombosis. Vagus and splanchic nerves form the coeliac and superior mesenteric plexus. The tail abuts the hilum of the spleen and the head is close to the right kidney. Metastatic disease (liver. infiltrative margins Clearly recognisable glandular structures. may be mucinous. Clinical examination should also include careful palpation for supraclavicular lymphadenopathy (Virchow’s node). Vessels in close proximity to the pancreas include the superior mesenteric vessels. Symptoms and signs from metastatic disease include: r Jaundice (extensive liver metastases). r Oesophageal varices (portal vein thrombosis). scirrhous. lung. r Fatty diarrhoea (obstruction of bile duct by primary or malignant lymphadenopathy). The aforementioned structures along with 152 Symptoms and signs from primary tumour and local spread include: r Gastric outlet obstruction (duodenal spread). Back pain relieved by leaning forward is due to infiltration of the retroperitoneal structures and usually suggests inoperability. r New-onset diabetes mellitus.

.Exocrine pancreas r Shortness of breath (pulmonary metastasis. A coagulation profile is necessary for patients who present with obstructive jaundice. pleural effusion). Intraoperative US may detect small liver metastases not seen on conventional imaging. Blood tests A full blood count.4. a radiologist. Investigation and staging Investigations are directed towards diagnosis and staging of the tumour. 1997). and anorexia. stage (locally advanced or metastatic) and performance status are important factors guiding management and overall prognosis. CT. Rarer histological variants like neuroendocrine tumours can be missed if a histological diagnosis is not established. Levels of the CA19–9 tumour marker are raised in 70% of patients with pancreatic cancer. r Weber-Christian disease (subcutaneous fat necrosis. r Virchow’s node (malignant left supraclavicular node). a histological diagnosis is needed before treatment: it may be radiologically difficult to distinguish chronic pancreatitis from carcinoma. 3D reconstruction may give additional information about the involvement of vascular structures. Paraneoplastic features include: r Migratory thrombophlebitis (Trousseau’s sign). A dual-phase helical CT scan (arterial phase to show the pancreas and venous phase to look for liver metastasis) has been reported to have a 79% positive predictive value and a 96% negative predictive value in predicting tumour resectibility. eosinophilia). polyarthralgia. and CA19–9 may be spuriously elevated in the presence of obstructive jaundice. Patients with metastatic disease have a poorer prognosis: a median survival of approximately 6 months and only 20% surviving 12 months from diagnosis (Burris et al. an oncologist and a pathologist to decide the most appropriate management. The endoscopic retrograde cholangiopancreaticogram is useful in patients presenting with obstructive jaundice. About 20% of patients are resectable at diagnosis with a median survival of about 20 months (Neoptolemos et al. weight loss. and needle-track seeding is extremely rare. US or EUS-guided FNA are safe procedures. The CA19–9 level needs to be interpreted with caution in patients with obstructive jaundice because the jaundice can cause high levels of the marker. In the past. especially in patients with a small tumour in whom a standard ultrasound (US) or CT-guided biopsy may be difficult. which is associated with acinar cell tumour. liver function tests. brushings obtained during the procedure can establish a malignant diagnosis (although the yield is low) and an endobiliary stent can be placed to relieve biliary obstruction. respectively. Imaging A CT scan is usually used for local staging and to exclude metastatic disease. Further staging and diagnostic procedures For patients who are otherwise suitable for radical surgery. pancreatic lesions have often been treated as ‘cancers’ based on a radiological diagnosis in the presence of raised tumour markers. EUS is also useful for obtaining a biopsy. Patients with localised disease should be discussed at a multidisciplinary meeting that includes a hepato-biliarypancreatic surgeon. imaging to assess vascular involvement. 2004). Chemotherapy is appropriate 153 . Histologically benign disease is found in up to 10% of radical resections. Staging classification The TNM definitions and staging classification are shown in Tables 12. r Dermatomyositis/polymyositis. preoperative laparoscopy may find unsuspected peritoneal metastases.3 and 12. A low initial level and a fall in marker with therapy are associated with a better prognosis. MRI has been shown to be equivalent to a good-quality CT scan but may be limited by cost and availability. both chemotherapy alone and chemoradiation are acceptable options with a median survival of about 10 months. Ideally. For patients with locally advanced non-metastatic cancer. Constitutional symptoms include fatigue. The ‘double-duct’ sign (occlusion of common bile duct and pancreatic duct) is almost diagnostic of pancreatic cancer. Endoscopic ultrasound (EUS) is a relatively new test which can be used before surgery along with standard Treatment overview The prognosis of pancreatic cancer is poor. urea and electrolytes are checked. For patients who are inoperable..

adjuvant chemoradiation or adjuvant chemoradiation followed by chemotherapy. M0 T1. N1. N1. Perioperative mortality and morbidity is inversely related to a surgeon’s case load (Birkmeyer et al. Distal pancreatectomy is the surgery of choice and total pancreatectomy may be necessary to achieve clear margins for large tumours. adjuvant chemotherapy alone. 1997) and patients should therefore be diagnosed and treated by a specialist surgical team (NHS Executive.05) (Neoptolemos et al. > 2 cm in greatest dimension Tumour extends beyond the pancreas but without involvement of the coeliac axis or the superior mesenteric artery T4 Tumour involves the coeliac axis or the superior mesenteric artery (unresectable primary tumour) NX N0 N1 MX M0 M1 Regional lymph nodes cannot be assessed No regional lymph node metastasis Regional lymph node metastasis Distant metastasis cannot be assessed No distant metastasis Distant metastasis Table 12. Patients with cancer of the body or tail present late and. The surgical complications include delayed gastric emptying. any N. N0. every 4 weeks for six cycles) is standard practice in the UK. therefore. any N. which entails a choledochojejunostomy and/or gastrojejunostomy. 2004). M0 T2.Somnath Mukherjee and Tom Crosby Table 12. M0 T1. M0 T2.3. p = 0.. TNM definitions of carcinoma of the pancreas Stage TX T0 Tis T1 T2 T3 Description Primary tumour cannot be assessed No evidence of primary tumour Carcinoma in situ Tumour limited to the pancreas. Patients on adjuvant chemotherapy had a significant survival advantage (median survival 20. 2005). 2001).. 154 . M1 Adapted from UICC (2002). N0. N1. N0. haemorrhage.5 months. palliative bypass surgery is required. Chemotherapy and chemoradiation Postoperative adjuvant chemotherapy Postoperative adjuvant chemotherapy (5-FU with folinic acid. sepsis. based on the results of the ESPAC 1 study. M0 III IV T4. 1999. Pylorus-preserving pancreatico-duodenectomy may be as effective but can result in delayed gastric emptying (Lin et al.0009) and chemoradiation was found to be detrimental (median survival 15. Neoptolemos et al. which randomised patients to no adjuvant therapy. M0 T3. Stage groups for carcinoma of the pancreas Stage 0 IA IB IIA IIB Description Tis.4. Adapted from UICC (2002). pancreatic fistula.1 months compared to 15. p = 0. for patients with WHO performance status 0 to 2. M0 Any T. The radiotherapy arms of the study have been criticised for using a low-dose Surgery Radical surgery Pancreatico-duodenectomy (Whipple’s procedure) is the standard operation for tumours in the head of the pancreas with an operative mortality of 1 to 16%. palliative radiotherapy and a coeliac plexus nerve block (for control of local pain due to retroperitoneal infiltration) are useful measures. Endobiliary stenting (for obstructive jaundice). days 1 to 5...9 months. ≤ 2 cm in greatest dimension Tumour limited to the pancreas. M0 T3. malabsorption and diabetes mellitus.9 months compared to 17. are less likely to have resectable disease. Palliative surgery In some cases of biliary and/or gastrointestinal obstruction. Nutritional support and close collaboration with the palliative care team is needed to ensure good symptom control. N0.

used what would be considered a suboptimal dose and lacked quality assurance (Klinkenbijl et al.. fractionation are as follow: give 45 to 50.4 Gy in 25 to 28 fractions) with infusional 5-FU (200 mg/m2 per day).5 months. One randomised study showed gemcitabine to be superior to 5FU-based chemoradiation . Trials comparing radiotherapy and chemotherapy (5-FU) to either treatment alone favour the use of combined-modality therapy. 3 out of 4 weeks. the maximum dose to the small bowel is 50 Gy. 3. respectively... energy. to support its use and this approach is not favoured in Europe or the UK. is an acceptable treatment option for patients with locally advanced non-metastatic disease. Gemcitabine weeks 1 to 7. The trial was underpowered. Concurrent chemotherapy additionally involves infusional 5-FU at 200 mg/m2 per day throughout radiotherapy. However. 1985).5 months compared to 19. respectively (p ≤ 0. p = 0. treating all fields daily. 1. The early results from the CONKO 001 study. the equivalent of at least 67% of one functional kidney must receive less than 20 Gy. which randomised patients between 5-FU with folinic acid and gemcitabine. Gemcitabine is a potent radiosensitiser and there has been a growing interest in its use in chemoradiation. Gemcitabine and capecitabine combination: gemcitabine 1000 mg/m2 weekly and capecitabine 1660 mg/m2 daily. lacking a detailed protocol and providing inadequate quality assurance. 2003). Acceptable treatment options are as follow: 1.2 and 7. There was a small but not statistically significant improvement in survival with combined modality therapy in patients with pancreatic cancer (median survival 24.099). The following are tolerance doses for organs at risk: r Small bowel – no more than 10% to receive > 45 Gy. However. using 10 MV photons. which randomised patients to gemcitabine or no adjuvant therapy. alone or in combination with radiotherapy. CT localisation should be performed with the patient in the supine position with arms raised. The median disease-free survival in the two arms was 14. 2-year survival 34% compared to 26%. intravenous contrast should be used. nausea and vomiting.001). 2005. there is only one randomised trial. for the purpose of verification. of the dose prescribed at ICRU 50. Dose. The additional benefit of chemoradiation in patients receiving gemcitabine-based chemotherapy is unknown and should be evaluated formally in a randomised study. 2005). (Li et al. were presented in ASCO 2005 (Neuhaus et al. Technique for chemoradiotherapy for locally advanced disease As a preplanning measure. which includes only 49 patients (Kalser and Ellenberg. 218 patients with pancreatic head and periampullary tumours were randomised to either adjuvant chemoradiotherapy (split-course) or surgery alone. 1981). r Kidney – one fully functioning kidney should be excluded from treatment if possible.Exocrine pancreas split-course technique. Louvet et al. has completed accrual and results are awaited. 3 out of 4 weeks. which has a median survival of about 10 months (Gastrointestinal Tumour Study Group. and then weekly thereafter. Postoperative adjuvant chemoradiation Adjuvant chemoradiation is widely practised in the USA. dehydration. such trials were carried out in the 1980s and the reported outcome from recent trials in patients treated with gemcitabine-based chemotherapy alone is similar but with less toxicity (Cunningham et al.. 2005). ESPAC 3. r Liver – 60% of the liver should receive no more than 30 Gy. 2. In an EORTC trial. patients should have an EDTA clearance and functioning renogram to assess total and individual renal function.0 months. subsequent cycles weekly. Three or four coplanar fields and anterior–posterior beams are used. DVHs for critical structures at risk should be obtained and the dose to the critical structures should be considered before accepting the final plan. then 1 week off.4 Gy in 25 to 28 fractions to the ICRU 50 reference point. The planning target volume is the gross tumour volume (GTV) with a 2 cm margin.. Moertel et al.8 Gy per fraction. 1999). Toxicity of chemoradiotherapy include gastrointestinal: anorexia.. 1988. Maximum and minimum allowable dose within the PTV should be 107 and 95%. lateral fields are angled anteriorly to reduce the dose to the kidneys. and the whole liver no more than 20 Gy. r Spinal cord – maximum of 45 Gy. Monday to Friday. Radiotherapy (45 to 50. 155 Chemotherapy and chemoradiotherapy for locally advanced disease Chemotherapy. Portal beam imaging should be taken during the first three days of treatment.

1997) showed that gemcitabine was more effective than bolus 5-FU in terms of median survival (5. capecitabine.91 months. progression-free survival (2. however. p = 0. Cunningham et al... none of the other drugs (alone or in combination) has resulted in better overall survival than single-agent gemcitabine (Cunningham. Plastic stents are placed if surgery is planned or if the prognosis is poor. 2005) and capecitabine (7. et al. conventional simulation with the patient supine and given oral contrast is acceptable and may be more practical.37 versus 5. the inferior border of the L3 vertebra.0002). oxaliplatin. 3 out of every 4 weeks. Anterior–posterior parallel opposed fields are used. Gemcitabine and erlotinib combination (erlotinib is licensed for use but currently not available in the NHS). Technique for palliative radiotherapy CT localisation in a CT simulator as described earlier is ideal. 2005). trials have. Radiotherapy Palliative radiotherapy A short course of radiotherapy may be used to palliate local symptoms such as pain in patients with metastatic disease or in symptomatic patients with locally advanced disease not suitable for chemoradiation. and the right border. Endoscopic relief of obstruction Obstructive jaundice is a major cause of morbidity in patients with pancreatic cancer.0025). p = 0. 5 cm to the right of the vertebral-body edge. irinotecan. General care for patient toxicity should include prophylactic antiemetics (according to local hospital policy) and weekly clinical review. 156 . p = 0. However. The target volume for CT simulation is a GTV with a 2 to 3 cm margin. Patients should be given ursodeoxycholic acid to help keep the stent patent. 3. Fixed dose-rate gemcitabine The intracellular effect of gemcitabine is limited by its intracellular phosphorylation. 80% have locally advanced. p = 0. 2005). 2. shown an increase in median survival when gemcitabine is used in combination with erlotinib (6. Moore. pemetrexed. 2 cm to the left of the vertebral-body edge.3 versus 0. a combination of oxaliplatin and 5-FU has a survival advantage compared to active support as a second-line therapy in patients progressing on gemcitabine. then 1 week off with subsequent cycles weekly for 3 out of every 4 weeks. blockage) but are more expensive. Metallic stents have a lower complication rate (infection. and 1-year overall survival (18 versus 2%. inoperable or metastatic disease at presentation. Palliative chemotherapy for metastatic or inoperable disease Of patients with pancreatic cancer.Somnath Mukherjee and Tom Crosby gastrointestinal bleeding (may occur a few weeks after the end of chemoradiation) and duodenal stenosis.026. Dose and fractionation involve the administration of 20 Gy in 5 fractions (4 Gy per fraction) or 30 Gy in 10 fractions.0002). marimastat and tipifarnib) have all failed to show a survival advantage. exatecan. One randomised trial (Burris et al.4 versus 6 months. the target volume is delineated by the superior border or middle of the T11 vertebra. the left border. p = 0. or thrombocytopenia (especially with gemcitabinebased chemoradiation). One trial has shown that a regimen using a lower dose rate infusion (1500 mg/m2 given at 10 mg/min) has greater activity than one that uses a higher dose rate (2200 mg/m2 given over 30 minutes) but with more toxicity (Tempero et al. Gemcitabine 1000 mg/m2 days 1.. leucopenia. Trials comparing gemcitabine alone with gemcitabine in combination with a variety of drugs (irinotecan.. Gemcitabine as above combined with capecitabine 1660 mg/m2 daily.4 months. Apart from capecitabine. Acceptable treatment options are as follow: 1. For conventional simulation (head of pancreas).9 months.025. Areas of current interest New chemotherapeutic agents under evaluation Oxaliplatin. single-agent gemcitabine has been the drug of choice for patients with a Karnofsky performance score of 50 or more. Since its recommendation by NICE in May 2001. 8 and 15.6 versus 4. every 4 weeks or once weekly for weeks 1 to 7. orathecin (oral topoisomerase 1 inhibitor) and docetaxel have demonstrated activity in the treatment of pancreatic cancer. Haematological toxicity includes anaemia. 5-FU. 2003).

Int. 23. 157 . Oncol.. Garofalo.5. Bramhall.. Surg. cetuximab and erlotinib (EGFR antagonists).. Sahmoud. Burris. A median survival of 6 to 16 months has been reported. 751–5. (2005).. 13. Int. J.. Hepatogastroenterology. Shan. J. Biol. M. (2003). J. The toxicity of such regimens depends on the size of the radiotherapy field as well as the dose and schedule of gemcitabine. et al.. .. Cancer. C. Tosteson A. M. S.Exocrine pancreas Table 12. 250–6. Low . Treatment of locally unresectable carcinoma of the pancreas – comparision of combined-modality therapy (chemotherapy plus radiotherapy) to chemotherapy alone. and bevacizumab (VEGF antagonist) have all shown activity in phase I and II studies. Hahn. Semin. Brief communication: a new combination in the treatment of advanced pancreatic cancer. 57. Compared to conventional radiation. (2005). 1601–4. Surg. (1997). Intensity-modulated radiation therapy (IMRT) IMRT delivers a radiation dose more conformally than conventional 2D or 3D radiotherapy. Phase III randomised comparison of gemcitabine (GEM) versus gemcitabine plus capecitabine (GEM-CAP) in patients with advanced pancreatic cancer. 98–104. Neoptolemos.. 899–903. Therapy of locally unresectable pancreatic carcinoma.. D. 59.. 48. (1988). Natl. et al. Li. Chau. S. 8).. Pancreaticoduodenectomy for pancreatic head cancer: PPPD versus Whipple procedure. Surg. Both low-dose gemcitabine (200 to 400 mg/m2 weekly. and Ellenberg. kidneys. (1985). H. Klinkenbijl. Anderson. J. et al. Moore. Labianca. Lin. Moertel. Chi. S.. S. 776–82. 445–53. 1370–6. Y. R. Kalser. C. Gemcitabine-based chemoradiation Several phase I and II studies and one phase III study (Li et al. III. P et al. G.. Radiat...5.. Cancer Inst. Ongoing/planned trials in pancreatic cancer A list of ongoing or upcoming trials is shown in Table 12. 84. Chmura. P Russell. P W. R. et al. Erlotinib is licensed for first-line use in conjunction with gemcitabine in advanced pancreatic cancer. H. Phys. Y. Radiat. T. IMRT reduces the mean dose to the liver. (1997). Oncol.. G. H. Biol. Ongoing/planned trials in pancreatic cancer Study group CALGB SWOG ECOG University of Chicago NCRN (UK) Trial arms Gemcitabine ± bevacizumab Gemcitabine ± cetuximab Docetaxel/irinotecan ± cetuximab Gemcitabine + bevacizumab + (cetuximab or erlotinib) TELOVAX – gemcitabine ± telomerase vaccine 3 2 Phase 3 3 2 REFERENCES Birkmeyer. UK Pancreatic Cancer Group. (1981). K.. J. Abstr. Adjuvant radiotherapy and 5-fluorouracil after curative resection of cancer of the pancreas and periampullary region: phase III trial of the EORTC gastrointestinal tract cancer cooperative group. Surgery. C. Gemcitabine in . and high dose radiation + 5-fluorouracil: The Gastrointestinal Tumor Study Group. 120. Arch. M. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial.. Intensity-modulated radiotherapy in treatment of pancreatic and bile duct malignancies: toxicity and clinical outcome. H. moderate dose radiation (4000 rads + 5-fluorouracil). C. (1999). Oncol. J. J. Phys. combination with oxaliplatin compared with gemcitabine alone in locally advanced or metastatic pancreatic cancer: results of a GERCOR and GISCAD phase III trial. R. Novel agents Pancreatic cancer cell lines express RAS mutations and mutations in EGF receptors and overexpress receptors for VEGF ligands and secrete VEGF-A. Hammel. Clin. M.. Frytak. 32 (6 Suppl. ECCO. Louvet. Milano. 1705–10. Jeekel. et al. (2005). J.. Concurrent . stomach and small bowel without compromising local control (Milano et al.. N. Lin. 52. Stoken. et al. S. 2003) have shown that gemcitabine-based chemoradiation is a feasible treatment. chemoradiotherapy treatment of locally advanced pancreatic cancer: gemcitabine versus 5-fluorouracil. J. 5–6. et al. 15. R. 3509–16.. Finlayson. 40 to 50 mg/m2 twice weekly) with full-dose radiation and full-dose gemcitabine with low-dose radiation (39 Gy) have been reported to have acceptable toxicity. a randomized comparison of high dose (6000 rads) radiation alone.. (2005). P Chao. J. Y. (2004). A. Tipafarnib (a farnasyl transferase inhibitor). a randomized controlled study. J. S. et al. Clin.. Gastrointestinal Tumor Study Group. Cunningham. 80.. mortality following resection for pancreatic and periampullary tumours in 1026 patients: UK survey of specialist pancreatic units. J. J.. PS11. Oncol. 230. 2403–13. D. 125. M. Ann. J. Pancreatic cancer. et al.. 2004). J. (1999). S.. C. I. Adjuvant combined radiation and chemotherapy following curative resection.. Oncol. Effect of hospital volume on in-hospital mortality with pancreaticoduodenectomy. Proc. D. Moore. Br.

4013. Sobin and Ch. et al. V. multicenter. (2005). 21. Post. N. 6th edn. I). A randomised. et al. J. prospective.. Oncol. P Stocken. H. W.. H. 1200–10. . New York: Wiley-Liss. L. Plunkett. Tempero...Somnath Mukherjee and Tom Crosby Neoptolemos. M. Med. Clin.. Clin. 93–6. H. phase III trial of adjuvant chemotherapy with gemcitabine vs. Wittekind. Guidance on Commissioning Cancer Services – Improving Outcomes in Upper Gastro-intestinal Cancers. ed. Randomized phase II comparison of dose-intense gemcitabine thirty-minute infusion and fixed dose rate infusion in patients with pancreatic adenocarcinoma. A random. (2003). ASCO Annual Meeting Proceedings. Engl. pp. J. et al. J. (2002). 158 . TNM Classification of Malignant Tumours. P Oettle. 23 (16 Suppl. 3402–8. Friess. Oncol.. ized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. observation in patients with resected pancreatic cancer. (2001). UICC. S.. NHS Executive. NHS Executive Catalogue Number 23943. J. Ruiz Van Haperen. Neuhaus.. London: Department of Health. D. D. (2004). 350.

which allow gene mutations to accumulate. There is a higher incidence of CRC in affluent Western countries and a lower incidence in Japan. is associated with an increased risk. and 98% of all cancers in the large intestine are adenocarcinomas. Many other countries are also considering such a programme.to ten-fold higher incidence in the developed world compared to that in the developing world suggests environmental causes. Screening is currently being adopted in the UK with the roll-out of a programme of faecal occult blood (FOB) testing. HNPCC is divided into two types: r The Lynch I classification describes CRC that is colon site-specific. Timothy Maughan and Tom Crosby Introduction Colorectal cancer (CRC) is second in incidence in Europe only to lung cancer. multiple tumours and an early age of onset. accessed September 2006). Africa and South America.1. Incidence and epidemiology The annual incidence of CRC is 54 per 100 000 in the UK. pancreatic and renal malignancies and has a propensity for right-sided tumours. Surgery is the only curative treatment and total mesorectal excision (TME) is now well established as the best way of managing rectal carcinoma. Targeted therapy such as epithelial growth factor receptor (EGFR) inhibitors and vascular endothelial growth factor inhibitors have been tested in patients with advanced CRC.cancerresearchuk. which are often mucin-producing and less aggressive. There are two well-recognised inherited cancer syndromes: hereditary non-polyposis colorectal cancer (HNPCC) and familial adenomatous polyposis (FAP). http://info. The peak incidence of CRC occurs at ages 60 to 70 years.13 COLON AND RECTUM Richard Adams. of neoadjuvant and adjuvant chemotherapy. with small but incremental improvements in outcome. but the role of these therapies in routine management has not yet been established. ovarian. The male-to-female ratios for colonic and rectal tumours are 2:3 and 2:1. and it causes around 204 000 deaths each year. tumours of the colon outnumber those of the rectum by 3 to 2. r Lynch II CRC involves increased risk of carcinoma of the colon and endometrium. Carcinoma of the colon and rectum Risk factors and aetiology Family history A family history of CRC.. gastric. particularly in first-degree relatives younger than age 40. and of new agents for advanced disease. right-sided. More than 16 100 deaths (50% of rectal origin) occur per year in the UK. CRC accounts for . 1999) and is then 159 Types of colorectal tumours The range of tumours that affect the colon and rectum is shown in Table 13. but the eight. and 35 000 new cases are reported per year (CRUK. HNPCC (Lynch syndrome) is an autosomal dominant condition involving defective mismatch-repair genes. HNPCC carries an associated increased risk of endometrial. and genetic causes probably account for around 15% of all CRCs. Around 15 to 20% of CRCs are of familial origin. around 13% of non-skin cancers. making CRC the second most common cause of cancer death. and affected gene carriers have an 80% lifetime risk of CRC. In general. The aetiology of CRC is still unclear. The clinical diagnosis of HNPCC is made using Modified Amsterdam Criteria (Vasen et al. involves multiple tumours and is more often mucinous. followed by colonoscopy if FOB testing is positive.org. respectively. But the last 10 years have also seen a rapid increase in the use of preoperative radiotherapy. HNPCC probably accounts for approximately 2% of colorectal cancers.

. causes a linear increase in the risk of colon cancer. Secondary prevention Strategies for secondary prevention are reviewed in a recent article by Agrawal and Syngal (2005). There is a higher incidence in affluent Western countries and a lower incidence in Southeast Asia. The range of tumours that affect the colon and rectum Type Non-neoplastic tumours (no malignant potential) Examples Benign polyps Hyperplastic polyps Hamartomatous polyps: Juvenile polyps and Peutz-Jegher polyps Inflammatory polyps Lymphoid polyps Neoplastic epithelial tumours Benign polyps Tubular adenoma Tubulo-villous adenoma Villous adenoma Malignant polyps Adenocarcinoma Carcinoid tumour Anal zone carcinoma Mesenchymal tumours Benign Leiomyoma Lipoma Neuroma Angioma Malignant Leiomyosarcoma Liposarcoma Malignant spindle-cell tumour Kaposi’s sarcoma Other Lymphoma r Attenuated FAP involves fewer adenomas and a slightly lower risk of colonic cancer. 2002). 2003. to a lesser extent. r Turcot’s syndrome involves colonic polyps associated with CNS tumours. which results in an increased risk of adenomas but with no proven increase in CRC. Crohn’s colitis are associated with CRC.1.. Other risk factors include smoking. Other factors Epidemiological studies on the cause of CRC now provide compelling evidence for environmental factors. infiltrating and poorly differentiated. Familial adenomatous polyposis (FAP) is an autosomal dominant condition characterised by defects in the APC gene on chromosome 5. high in refined carbohydrates and fat content. flat. Conditions associated with colorectal cancer Ulcerative colitis (UC) and. FAP accounts for fewer than 1% of all CRCs. Africa. The risk is greatest in patients who develop early onset UC with a large proportion of the colon affected. which is presumably due to dietary factors. osteomas and desmoid tumours. but cohort studies in the USA provide less evidence for this finding. and much of South America. Jacobs et al. Chemoprevention There is currently no evidence that any drugs are effective in preventing CRC. . 160 The EPIC (European Prospective Investigation into Cancer and Nutrition) study has concluded that diets deficient in fibre are associated with CRC and a doubling of fibre intake could reduce the risk of CRC by 40%. Possible dietary risk factors include a diet low in indigestible starch. There are three variants of FAP: r Gardner’s syndrome involves colorectal adenomatous polyps. More recent results from this study have suggested that diets high in fish and low in red meat content may protect against CRC (Bingham et al. including ependymomas and medulloblastoma. Screening and prevention Primary prevention confirmed by laboratory testing for MSH1 and 2 and PMS1 and 2 mutations. Timothy Maughan and Tom Crosby Table 13. Increased fruit and vegetable intake results in a protective effect for colon and other cancers as suggested by case-control studies. It often manifests as a carpet of adenomatous polyps throughout the large intestine. 90% of affected individuals will have an invasive cancer by the age of 45. 1998. and a decreased intake of protective micronutrients.Richard Adams. between 23 and 30 kg/m2 . Increased body mass index.. These cancers are often multifocal. Key et al.

2. 1996). may be removed. 2003. See Fearon and Vogelstein (1990). These procedures also have a preventative role because polyps. Normal epithelium 5q-mutation / loss FAP Hyperproliferative epithelium DNA hypomethylation Early adenoma 12p mutation KRAS Intermediate adenoma 18q loss ?DCC Late adenoma 17p loss TP53 Carcinoma Sigmoidoscopy or colonoscopy Flexible sigmoidoscopy is both sensitive and specific. which account for approximately 65% of colorectal carcinomas. r Polyps greater than 2 cm carry a 50% risk. FAP = familial adenomatous polyposis. which might be premalignant. When used as part of a screening programme. This multistage concept of carcinogenesis includes mutation or deletion of the tumour suppressor gene adenomatous polyposis coli (APC). Transcoelomic spread occurs with diffuse peritoneal disease.. the bladder. Adapted Vogelstein model of carcinogenesis for CRC. The risk of cancer varies with the size and number of polyps: r Polyps ≤ 1 cm carry a ≤ 1% risk of invasive malignancy. In the Nottingham trial. 161 . 2005). the false-positive rate was high. 1990).1 and it defines the pathway to chromosomal instability. polypectomy has shown a 59% reduction in mortality from bowel cancer. through adenomas. Pathology The adenoma-carcinoma sequence Vogelstein described a model of transition from normal bowel. those with a positive family history). Hardcastle et al. Then the deleted in colon cancer (DCC) tumour suppressor gene probably helps to initiate metastatic potential and. Spread Carcinoma of the colon and rectum can spread to adjacent structures such as the small or large bowel. Figure 13. at around 90%. Morphology The morphology of colorectal adenocarcinoma is shown in Table 13. Metastases Strategies for patients with inflammatory bowel disease Screening with regular colonoscopies is offered to high-risk patients and prophylactic panproctocolectomy is appropriate in selected cases. however. the uterus and so on. Colonoscopy is better at picking up proximal lesions but is more expensive and incurs higher morbidity. FOB testing has been launched in areas of the UK with a plan for general phased roll-out in late 2006 (Autier et al.e. but it will only pick up abnormalities in the rectum and sigmoid colon. A separate pathway characterised by MSI occurs in HNPCC along with an acquired loss of DNA mismatch repair (Rowan et al. there was a 15% reduction in cumulative CRC mortality in the screened group. r 3% of CRCs are multicentric. followed at some stage by mutation in the KRAS protooncogene. TP53 gene mutation and DNA microsatellite instability (MSI). TP53 = tumour protein p53.1. Polypectomy via sigmoidoscopy or colonoscopy is generally performed on at-risk patients (i. KRAS official name = v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog. DNA = deoxyribonucleic acid.. eventually. r Polyps between 1 and 2 cm carry a 10% risk. including one from Nottingham. DCC = deleted in colon cancer. Vogelstein’s model of carcinogenesis is shown in Figure 13.3. to carcinoma (Fearon and Vogelstein.Colon and rectum Faecal occult blood (FOB) There have been three large population-based studies in people between the ages of 50 and 80. Features associated with an increased risk of malignant potential within colonic polyps are shown in Table 13..

Clinical presentation Symptoms and signs from the primary tumour Symptoms that occur more commonly in rightsided lesions include unexplained anaemia. stenosing tumours Microscopic Well.2. Rectal examination can identify 75% of rectal tumours. r A palpable rectal mass.Richard Adams. 2000). giving macroscopic firmness Many tumours produce mucin. r Iron-deficiency anaemia (haemoglobin concentration ≤ 11 g/dl in men or ≤ 10 g/dl in postmenopausal women) without obvious cause. moderately or poorly differentiated Spectrum from tall columnar cells akin to those found in benign polyps but with evident invasion to frankly anaplastic carcinomas Strong desmoplastic reactions are common. hypogastric nodes and presacral nodes).5 cm) Sessile or flat Severe dysplasia present Squamous metaplasia present Villous architecture present Multiple polyps Low risk Small size (≤ 1 cm) Pedunculated Mild dysplasia present No metaplasia present Tubular architecture present Single polyp Table 13. ulcerative or diffusely infiltrating Proximal tumours often grow as polypoid or fungating masses extending along one wall of the caecum or colon Carcinomas in the distal colon tend to be annular. an abdominal mass. Haematogenous spread is most often to the liver. Criteria for referral to a specialist Symptoms that require urgent referral to a specialist have been reviewed (NHS Executive. lungs and bone. the lymph node chain usually follows the supplying blood vessels (for rectal cancers these will be pararectal lymph nodes. In 25 to 30% of patients at presentation. 162 . but many other sites are possible. small-cell undifferentiated tumours are of neuroendocrine origin A few have signet ring appearances more commonly found in stomach or breast carcinomas Distal colon tumours occasionally have foci of squamous differentiation and are known as adenosquamous carcinomas Regional lymph nodes are involved in 40 to 70% of cases at operation. for all ages: r A palpable right-sided. Additional urgent symptoms in patients over age 60 include: r Persistent rectal bleeding without anal symptoms. the tumour will have spread either locally or to distant sites and will be unsuitable for radical treatment. abdominal mass. nodes at the bifurcation of the inferior mesenteric artery. namely diarrhoea or constipation. tenesmus and mucoid discharge. which carry a poor prognosis 10% of adenocarcinomas of the colon have neuroendocrine differentiation. Pathological features of colorectal adenocarcinoma Features Macroscopic Description Polypoid. occurring in that order of frequency. They include. Features associated with an increased risk of malignant potential within colonic polyps Risk category High risk Feature Large size (> 1. More common in left-sided lesions are a change in bowel habit. r Rectal bleeding with a change in bowel habit to more frequent defecation or looser stools (or both) that is persistent over 6 weeks.3. including the brain. rarer mucinous tumours produce sheets/pools of mucin. obstruction. r A change of bowel habit to more frequent defecation or looser stools (or both). weight loss and rectal bleeding with altered blood. and persistent for 6 weeks. ill-defined abdominal pain. Timothy Maughan and Tom Crosby Table 13. fungating. rectal bleeding. without rectal bleeding.

patients should be 163 . Colonoscopy or double-contrast barium enema are used because 5% of lesions are synchronous. the ability to predict the threat of tumour to the CRM (MERCURY Study Group. resect the tumour if possible and then consider the patient for adjuvant (or palliative) therapy. A PET scan is only relevant when resection of liver metastases is considered.5. A CT scan of the thorax and abdomen or CXR and US of liver can be performed. especially if local excision of a rectal T1 lesion is contemplated. penetration and nodal involvement of low rectal tumours and may be appropriate. M0 40% II Stage (AJCC) 0 I TNM Tis N0 M0 T1 N0 M0 T2 N0 M0 T3 N0 M0 T4 N0 M0 Five-year survival (%) 100% 95% 90–95% 70–80% Investigation Staging investigations A full blood count and liver function tests should be performed. nodal assessment is relatively poor (accurate in 50 to 80%). Treatment for non-metastatic disease: colonic carcinoma The general principle of non-metastatic colon cancer management is to obtain staging information. diverticular mass or abscess. Higher CEA values are associated with a worse prognosis when measured preoperatively but they do not change clinical management. Endo-anal US gives good information on size. The TNM classification is shown in Table 13. Pelvic MRI should be performed for patients with rectal tumours. 2006). but it is poor at assessing nodal disease. Pathological staging of colorectal cancer Dukes’ staging has undergone various modifications and was originally used to describe rectal cancers. be accurate in over 90% of cases (can miss microscopic T3 disease). M1 5% III Any T. therefore. Biopsy is essential because of the wide differential diagnosis: inflammatory bowel disease/colitis (ischaemic or inflammatory). infective diarrhoea.Colon and rectum Table 13. Any N. but it is now also accepted for colonic cancers and is shown in Table 13. Dukes’. 65% of the liver can be involved before an abnormality is detected. Local staging of rectal cancer Digital rectal examination (DRE) can be accurate for T staging in up to 80% of cases. to assess the likely circumferential resection margin (CRM) and the need for preoperative therapy (see discussion that follows). no LN spread Through the muscularis propria with no nodes involved Nodes positive but not the apical node Apical node positive Metastatic IV Any T. EUS is very user-dependent but it can Surgery Surgery is an essential part of curative treatment for colon cancer. TNM and AJCC staging classifications shown with approximate 5-year survival by stage Dukes’ stage (modified) Description In situ A B(1) B(2) C(1) C(2) D Cancer confined to submucosa or muscularis propria but not through it Into but not beyond muscularis propria. N1–2. CEA is raised in 85% of patients with CRC.4. Whenever possible. The MERCURY study showed concordance between MRI and histological findings in the spread of tumour beyond the lamina propria and. Recent studies have suggested that MRI can be used to identify the mesorectal fascia (the only investigation able to visualise this plane used in TME dissection) and to predict clearance at the CRM. Once again.4.

Fluoropyrimidines All adjuvant regimens are based on fluoropyrimidines.. NHS Executive.. 2006). extramural vascular invasion and involvement of the CRM – although these selection criteria have not yet been prospectively verified.. There are certain poor-risk features identifiable in node-negative patients that may be used to select those suitable for adjuvant treatment – serosal involvement. 2000) and has become the standard bolus 5-FU regimen in the UK. 2004). the benefit is about 6% at 5 years (range 2 to 10%. perforated tumours. r When using a standard 5-FU and FA regimen. heavily node-positive cancers. r Levamisole is ineffective (Haller et al. 2005) and is useful in patients with poor venous access. 1998). stomatitis and leucopenia. although it remains controversial whether all such patients should be offered adjuvant therapy. r All 5-FU-based regimens are associated with a risk of toxicity. N0 tumours. TNM classifications of CRC Stage T1 T2 T3 Definition Tumour invades submucosa Tumour invades muscularis propria Tumour invades through muscularis propria and into subserosa or into non-peritonealised pericoloic or perirectal tissue T4 N1 N2 M1 Tumour directly invades other organs/structures and/or perforates visceral peritoneum 1–3 lymph nodes involved 4 lymph nodes involved (12 should be examined) Presence of metastases Dukes’ C and to selected patients with Dukes’ B colon cancer.. This means that patients presenting as an emergency with obstruction or perforation should be resuscitated and first have a defunctioning colostomy. but in a non-randomised comparison in the QUASAR trial the 5-FU/FA regimen showed no reduction in effectiveness (Kerr et al. vascular pedicle and draining lymph nodes. T3 or node-positive tumours Two recent metaanalyses have shown that prolonged use of a 5FU-based regimen for longer than 3 months can improve survival in colorectal cancer. For patients with node-negative colon cancer. See the Appendix for a list of different adjuvant chemotherapy treatments (p. notably diarrhoea. 2000). Andre et al. r The weekly bolus 5-FU and FA regimen is less toxic than the conventional 5-day 4-week (Mayo) regimen. followed by elective removal of the tumour. Current standard practice in the UK is to offer adjuvant chemotherapy to patients with 164 Oxaliplatin as adjuvant Trials comparing 5-FU and FA plus oxaliplatin versus 5-FU and FA alone have both shown an improvement in disease-free survival at 3 years of about 6% (MOSAIC and NSABP C-07 trials.5. 2005). operated on by the specialist team. r Neutropenia is more prevalent in patients over age 70 but they are just as likely to complete treatment as younger patients and they have a similar survival benefit from treatment (Lichtman. 173).. the increase in overall survival is 1 to 5% after a median follow-up of 4. such as those with T4.2 years (Gray et al. Adjuvant chemotherapy for colon cancer T1 or T2. Twelves et al. but with a greater risk of neuropathy.. For patients with node-positive colon cancer. also has a proven role in adjuvant therapy for rectal cancers. Oral fluoropyrimidines Capecitabine is as effective as and less toxic than the Mayo clinic regimen (X-ACT trial. Wolmark et al. 2004). N0 tumours Adjuvant chemotherapy has shown no benefit in patients with T1 or T2. Timothy Maughan and Tom Crosby Table 13. Stenting an obstructing left-sided colonic or rectal tumour can also be useful in an emergency (Athreya et al. 2004. Adapted from UICC (2002). Irinotecan as adjuvant Three trials have shown no benefit from adding irinotecan to adjuvant 5-FU and . there is no advantage to extending the duration of therapy past 6 months (O’Connell et al. In the 1990s the following guidelines were established: r Low-dose folinic acid (FA) is just as effective as highdose FA (Anonymous..Richard Adams. tegafururacil. Another oral fluoropyrimidine. r Infusional (de Gramont) chemotherapy is no more effective than bolus 5-FU/FA in the adjuvant setting. Definitive surgery involves removing the appropriate bowel segment with its mesentery.. 1997). 2005). Oxaliplatin with a fluoropyrimidine has therefore become the standard adjuvant therapy in patients who are fit to receive this more toxic regimen and whose risk of relapse is high.

Early stage rectal cancer Treatment for non-metastatic disease: rectal carcinoma Patients whose tumours are predicted to be likely to recur after surgery alone can be selected for preoperative therapies. r High risk (> 30%) – G3. 2004.or four-field volume. the fact that they do not tend to spread distally means an anterior resection is possible. MacFarlane et al. which has been advocated for years by Heald. https://www. T2 or T3. r Intermediate risk (10 to 20%) – T2. This technique.jsp). The risk categories are not universally agreed but can be considered to be r Favourable: rT1 to rT3a (minimal extension into rectum). Adjuvant online Data from clinical trials in colon cancer can be used to help estimate the benefit of adjuvant chemotherapy attributable to oxaliplatin plus fluoropyrimidine combination therapy through the ‘adjuvant on-line’ service (adjuvantonline. stage for stage.4 Gy in 25 to 28 fractions). 1999). and it reduces the risk of an involved resection margin (defined as ≤ 1 mm. It is currently recommended that patients with rectal cancer be given the same adjuvant systemic therapy as those with colon cancer. Anterior resection is made easier by the use of a staple gun to form the anastomosis. A local excision is an alternative to APR in low rectal cancers that are unlikely to have lymph node metastases.64.. both given to a planned three. In experienced surgical centres local excision for low-risk T1 tumours can be associated with excellent results (Willett. 95% CI = 0. T2. but this is still the subject of ongoing clinical trials.Colon and rectum FA in colon cancer (Saltz et al. the mesorectum.. N0. 1993).. the meta-analysis of trials of adjuvant chemotherapy for CRC showed a greater benefit in rectal cancer than in colon cancer (OR for mortality = 0. MRI can be used to identify three prognostic groups based on stage. Surgery Surgery is technically more difficult for rectal than colonic tumours because the pelvis is narrow. G2. 1997). Ychou et al. 165 . TME has become a standard procedure for rectal cancer surgery..85) and estimated the benefit to be a 9% increase in survival at 5 years for rectal cancer patients (Dube et al. 36% of patients were CRM +ve and had a 64% chance of local recurrence. Patients with adverse risk factors after local excision should be considered for further resection. Based on the results of the MERCURY study described earlier. password required. r Unfavourable: > rT3a (significant mesorectal contamination). Very few data exist that support the use of postoperative chemoradiotherapy after local excision alone. involves sharp dissection of Radiotherapy and chemoradiotherapy in rectal cancer There is good evidence for the use of either shortcourse preoperative radiotherapy (25 Gy in 5 fractions) or longer-course chemoradiotherapy (45 to 50. or N +ve with margin not at risk. R1 resection) do worse with local excision (LE) or abdomino-perineal resection (APR). Van Cutsem et al.. Very low tumours require an APR.48– 0. 2005. and the lymph node status. the predicted relationship of the tumour to the CRM. In practice this procedure is for well-differentiated T1 tumours: r Low risk (≤ 10% chance of nodal disease) – T1. The risk of local recurrence following surgery is related to the CRM and the nodal status. which is consistent with the findings of phase III clinical trials such as the Dutch Rectal Cancer Trial and CR07. G1. In the pre-TME era.. The most common approach to local excision is transanal endoscopic microsurgery (TEM). For tumours that are higher. r Advanced: T4 or CRM ≤ 1 mm. 1994). However. Low anterior resection is associated with increased morbidity including bladder and sexual dysfunction and increased bowel frequency than for patients with higher rectal tumours. 2005). Retrospective series have shown that low rectal cancers with poor prognostic features (G3. The patients with CRM −ve surgery had a local recurrence rate of only a 9% (Adam et al. especially in the male. lymphovascular invasion.com/colon. Adjuvant chemotherapy for rectal cancer The clinical benefit of adjuvant chemotherapy alone for patients with rectal cancer has been difficult to prove because many trials have assessed the combination of chemotherapy and radiotherapy together. even when the surgeon felt macroscopic clearance had been achieved.adjuvantonline. LVI +ve.

r The inferior border is 3 to 5 cm below the lower limit of macroscopic tumour. which showed an 8% improvement in overall survival (38 versus 30%. The local relapse rates reduced from 17 to 5% overall at 5 years in favour of preoperative radiotherapy. There was no survival benefit. Marijnen et al. and the effect was seen in patients with CRM −ve tumours. Node-positive patients who 166 Practicalities of radiotherapy for rectal cancer Patient position and immobilisation are as follows for the external beam radiotherapy technique.3%. Camma et al. 2005). Folkesson et al.5 versus 58.6 versus 13. p ≤ 0. 2001). Preoperative radiotherapy and chemoradiotherapy An overview of preoperative radiotherapy trials showed a significant reduction in the odds of death or local recurrence (OR = 0. and a comfortably full bladder to displace the small bowel. The patient lies prone with toes together.3%). local lymphatics and the presacral region: r The superior border is defined as the superior aspect of the sacrum or 3 cm above the upper limit of the macroscopic tumour.001. early trials used a two-field technique and were associated with a risk of cardiovascular mortality. There was also more acute and late grade 3 or 4 toxicity in the postoperative group (acute. this effect was not seen in later trials. Either CT simulation or orthogonal plain X-ray films are taken. r The posterior border encompasses the presacral nodes and sacral hollow. Long-term follow-up (median 13 years) was provided by the Swedish Rectal Cancer Trial. whichever is highest. This trial showed 10% CRM positivity with a 60-day postoperative mortality of 1.. respectively. late.. The Dutch TME trial looked at preoperative shortcourse radiotherapy in more than 1800 patients undergoing TME surgery. but there was an increased risk of short-term morbidity (perineal wound infection/non-healing) and late morbidity (increased stool frequency and incontinence of loose stools. postoperative radiotherapy is reserved for patients who are found to have CRM involvement at surgery and who did not receive preoperative radiotherapy. The target volume includes the primary tumour. For those trials that employed higher radiation doses (RBE > 30). or surgery followed by selective postoperative chemoradiotherapy for CRM-positive tumours (95% had TME). 24 versus 14%). Birgisson et al. 2006). Oral contrast Gastrografin may be taken orally 1 hour before simulation to delineate small bowel volume within the field. with localisation marks (tattoos posterior and two laterals) and rulers. but the benefit is smaller than for preoperative radiotherapy (18. Once again there was no significant effect on either rectal-cancer-specific or overall survival (Colorectal Cancer Collaborative Group. p = 0. 1 cm lateral to the pelvic brim or 3 cm to the most lateral portion of macroscopic disease.04). 40 versus 27%.8% (Colorectal Cancer Collaborative Group.. had CRM-negative tumours and who did not receive postoperative radiotherapy had a high local relapse rate of 17% at 5 years.49. A–P and lateral. The anal canal is only included in low rectal tumours and perineal scar in postoperative AP resections. Within this group. The MRC CR07 trial randomised 1350 rectal cancer patients to preoperative radiotherapy (25 Gy in 5 fractions) followed by surgery.74. which used a threeor four-field technique.008) and a reduction in local recurrence from 26 to 9% (p = 0.. heels apart. there was a marginal survival advantage (56. Such patients should be considered for postoperative chemoradiotherapy if no neoadjuvant therapy has been given (Sebag-Montefiore et al.03 and OR = 0.001. p = 0. 2005). but the local recurrence rate was significantly lower in patients who had radiotherapy (11. The role of postoperative radiotherapy in node-positive patients is less certain. Timothy Maughan and Tom Crosby A recent German study compared long-course preand postoperative chemoradiotherapy and showed an improvement in local control rate (13 versus 6%) in those given preoperative treatment (Sauer et al.Richard Adams. r The anterior border is 2 to 3 cm anterior to the macroscopic tumour or anastomosis but at least to an area 2 cm anterior to the sacral promontory in a horizontal . 2004). Barium may be instilled into the rectum to assist in localisation and a radio-opaque marker placed at the anal margin. A metaanalysis published in 2001 showed that local recurrence is reduced from 17 to 10%. p = 0. whereas for trials in which a radiobiologically equivalent dose (RBE) of more than 30 Gy was given. In the UK. 2000).. urgency and emptying difficulties. Postoperative radiotherapy A meta-analysis of postoperative radiotherapy trials in rectal cancer indicates an effect on local recurrence.9%. head resting on the hands.4 versus 5. whichever is the most lateral. 2001). Preoperative radiotherapy schedules have become standard practice in the UK. r The lateral borders are the pelvic side walls and internal iliac regions.. 2005). recurrence was reduced from 20 to 9.8%.

and patients with low tumours may get a severe and painful perineal reaction.5 to 7 weeks to a patient with macroscopic disease. When explaining the procedure to the patient undergoing radiotherapy or chemoradiotherapy. and cystitis. dose for days 1 to 5 and days 29 to 34: should be 5-FU 350 mg/m2 per day i. Radiotherapy with or without chemotherapy can give useful palliation of pain and control of local disease. Be careful if patients have adhesions. red cell transfusion can be performed (keep Hb > 10. The plan is a three-field plan with one posterior and two lateral fields.. prescribed to the ICRU reference point. r For neutropenia. There is a small risk of long-term bladder effects and risk of small bowel obstruction. The indication for treatment should be reviewed and treatment fields should be kept as small as possible to reduce toxicity. apply aqueous cream topically b. a small-volume boost of 10 to 15 Gy can be given in 5 to 7 fractions. 7 days per week for the 5 weeks of radiotherapy.d.v. Trans-anal tumour ablation using laser. prescribed to the ICRU reference point. apply Instillagel® to areas of moist desquamation (low rectal tumours). In women. Concurrent chemotherapy in combination with longcourse radiotherapy involves continuous infusion of 5-FU at 200 mg/m2 per day throughout radiotherapy treatment. treat the infection.0 g/dl). There is no evidence that palliative resection improves survival. using 6 to 10 MV photons. and 12 to 15 cm (S–I) × 10 to 12 cm for the lateral fields. bolus. the patient receives 45 Gy in 25 fractions over 5 weeks. r For skin reaction. using 6 to 10 MV photons.. there is also evidence from phase II and ongoing phase III studies for the use of capecitabine at a dose of 825 mg/m2 b. and a plan is produced by the Department of Physics. inflammatory bowel disease. 167 . Explain the procedure and its possible side effects: r Acute: the patient may experience tiredness. Typically 45◦ wedges are used on the two lateral fields. The following information should be used to support the patient: r For diarrhoea. diverticular disease or diabetes mellitus. Portal beam imaging should be taken on the first day of treatment. give the patient metoclopramide. alternatively. r For anaemia. bolus and FA 20 mg/m2 per day i. bowel dysfunction and urge incontinence may occur. Dose. if not CT simulated. r For nausea. or 1% hydrocortisone cream if severe. for preoperative long-course chemoradiotherapy. and encourage the patient to drink fluids. An outline of the patient is taken. Though an unlicensed indication. Myelosuppression is rare. consider a boost as just mentioned to maximise local control. a defunctioning colostomy may provide useful palliation but will not relieve tenesmus. clinically unresectable tumours may become operable after radiotherapy. r Inoperable or recurrent disease can be dosed with 45 to 50 Gy in 25 fractions. diarrhoea. give written information and refer the patient to a specialist nurse counsellor if one is available. Lead shielding may be applied to the posterior sacrum and anterior to the anal margin (if not a low tumour). r In case of residual disease in postoperative treatment. Lasers should be used to align the patient. give imodium and advise a low-residue diet.Colon and rectum plane. electrocoagulation or resectional techniques may provide better palliation and should be considered in selected cases. In some patients. r Late: menopause (in females) and infertility/sterility. energy and fractionation procedures are as follow: r For preoperative short-course radiotherapy without chemotherapy. bleeding or mucous discharge. r The postoperative dose is 45 to 50 Gy in 25 fractions over 5 weeks. set to the ICRU 50 reference point. treat all fields isocentrically each day. the patient receives 25 Gy in 5 fractions over 1 week. followed by surgery at around 1 week. r For cystitis. the anterior border should include the posterior vaginal wall. Treatment for advanced/inoperable local disease: colon and rectal carcinoma Inoperable primary disease Inoperable primary disease is most common in the rectum. Field sizes are typically 12 to 15 cm (S–I) × 14 to 17 cm for the posterior fields. impotence. Practically this includes the posterior two-thirds of the femoral heads or comes to the posterior symphysis pubis in T4 lesions.d. r Or. treat the patient as per local neutropenic sepsis protocol. r The total dose should be 55 to 60 Gy over 6. In a patient with an inoperable obstructing rectal cancer. send MSU. followed by surgery at 6 weeks.v.

In very unfit and frail patients some surgeons will attempt to debulk rectal tumours trans-anally. Some evidence exists that resection of locally recurrent cancer may improve survival but this has not been proved in a randomised trial.Richard Adams. those with a higher number of liver metastases. discharge. Around 15 to 20% of patients with liver metastases may benefit from primary resection and. Patients with fixed caecal tumours may benefit from local radiotherapy (8 Gy in a single fraction). Fitter patients may benefit from a higher dose of 45 Gy in 25 fractions. Systemic treatment is usually the only 168 Palliative chemotherapy A number of agents have demonstrated activity in metastatic CRC: the fluoropyrimidines including 5-FU. In situ destructive therapies for liver metastases Interstitial laser ablation. A retrospective review of 2040 patients with metachronous isolated hepatic metastases showed that those who underwent resection had a mean survival of 31 months (projected 5-year survival of 26%) whereas those patients who did not undergo resection had a mean survival of 11 months (projected 5-year survival of 2%. Wade et al. and those with a higher CEA (Fong and Salo. Timothy Maughan and Tom Crosby Palliative radiotherapy Patients with rectal cancer who are medically unfit for surgery. Treatment with chemoradiotherapy should be considered if it has not already been given. oral prodrugs capecitabine and combination UFT. may be treated with palliative external beam therapy to the pelvis. p ≤ 0. 2004. However. Case series of selected patients have reported long-term survival. particularly if they have bleeding and recurrent anaemia. Progression of disease while the patient is on chemotherapy is a poor prognostic factor. or bleeding. this may not be possible. Palliative radiotherapy is also beneficial for bony metastases or local bony erosion (8 Gy in a single fraction). hepatectomy can achieve a 5-year survival of 40 to 50%. 1996).0001.. the . option although. Locoregional recurrence The 3-year survival of patients with locoregional recurrence of CRC is about 10%.. surgical excision of metastases or in situ destructive therapy may be feasible. but there is no evidence to support routine use of these therapies. with careful patient selection. Mid and low rectal tumours can be complicated by stent migration and tenesmus. cryotherapy and radiofrequency ablation have been used for isolated colorectal liver metastases. those with a heavy positive lymph node burden with the primary tumour. Symptom relief occurs in around 70 to 80% of patients treated with radiotherapy but the median duration of relief is only 3 months. even in patients with liver metastases undergoing staged removal of metastases. The survival rate is worse for patients presenting with metastatic disease that is synchronous with their primary tumours (compared with a prolonged progression-free interval). Bismuth et al. Endoscopic therapy Stenting of colonic and high rectal tumours is feasible and effective either for immediate relief before radical surgery or as a purely palliative procedure. Preoperative chemoradiation before surgical resection of recurrent disease has increased resectability rates by up to 60%. 1999). Palliative surgery Even in the palliative setting. Some patients with solitary pulmonary metastases may also be suitable for resection. all the hepatic veins are involved or not enough viable liver tissue will be left after resection. Treatment for metastatic disease: colonic and rectal carcinoma Surgery for isolated liver metastases The liver and lung are the most common sites for metastatic CRC.. Such patients should be selected on an individual basis with involvement of the specialist lung multidisciplinary team. Patients should not generally be considered for surgery if there is disease outside the liver. 1996). for a few patients. either as a single fraction of 8 Gy or fractionated to 36 Gy in six fractions over 6 weeks. and who have symptoms of local pain. a procedure which has little morbidity. Neoadjuvant chemotherapy for metastatic disease Non-randomised evidence supports the use of preoperative chemotherapy prior to resection in patients with potentially operable liver metastases but which are considered inoperable at presentation (Adam et al. excision of a symptomatic tumour and re-anastomosis is preferable to a palliative bypass because excision is associated with a better survival rate and quality of life.

The use of capecitabine has not been compared to optimal infusional 5-FU regimens. 2006). 2003). 2005). issued in August 2005. 1 and 0. Experience is required in deciding whether patients should be given chemotherapy. However. slowly progressive metastatic disease. Palliative chemotherapy: irinotecan and oxaliplatin There is evidence to support the use of irinotecan and oxaliplatin as a first-line treatment in combination with 5-FU.8 to 44. high alkaline phosphatase and liver involvement are independent predictors of progression. Poor performance status. low serum albumin.3 months in the firstline setting (Hurwitz et al.. 2006). high γ -glutamyl transferase and high CEA predict for poor survival.Colon and rectum chemotherapy agents oxaliplatin. Tournigand et al. Fluoropyrimidine monotherapy is used for patients who are unfit for combination chemotherapy or who have low-volume. Median overall survival has improved. low serum albumin. The optimal way of delivering 5-FU is by infusion over a minimum of 24 hours. The sequencing of chemotherapy has been investigated in a number of trials. which is now no longer used.’ Meanwhile.. including the MRC CR08 (FOCUS) trial (Maughan. the same guidance stated that ‘raltitrexed is not recommended for people with advanced colorectal cancer. to 10 to 14 months for those treated with fluoropyrimidines alone.8% and median survival increased from 15. irinotecan. especially 5FU in metastatic disease or via portal vein infusion for adjuvant therapy. or given on its own to people who have already had chemotherapy. There is no evidence that any particular sequence of oxaliplatin. Bevacizumab is a monoclonal antibody for use against the vascular endothelial growth factor ligand and is licensed for first-line metastatic use. raltitrexed. early chemotherapy before clinical deterioration improves survival by 3 to 6 months without any adverse impact on quality of life (NHS Executive... oxaliplatin and bevacizumab. 2004). 2005. NICE guidance 93. One trial of 813 patients randomised to first-line irinotecan and fluoropyrimidine with or without bevacizumab found that response rates increased from 34. The oral prodrug capecitabine is also approved for monotherapy and has been shown to be less toxic than the Mayo Clinic regimen. 10 and Regional chemotherapy Hepatic arterial infusion of chemotherapy. if they are used in these ways: r Irinotecan is given with 5-fluorouracil and folinic acid to people who have not had chemotherapy for advanced colorectal cancer before.. for patients with advanced disease. Cetuximab is an EGFR monoclonal antibody licensed for use with irinotecan in second. and fluoropyrimidine is superior to another. This practise is based on evidence from the MRC CR06 study (Maughan et al. median survivals were 4.or subsequent line treatment of metastatic CRC. 2002) and the OPTIMOX studies (Tournigand et al. When used as a combination therapy. and the biological agents bevacizumab. This is being further explored in the MRC COIN and AEROMC04 trials. r Oxaliplatin is always given with 5-fluorouracil and folinic acid. and then start again when there is evidence of progressive disease. Performance status is a particularly strong indicator. The MRC CR06 trial showed a non-statistically significant trend towards superior results in patients who receive intermittent rather than continuous chemotherapy. from 6 to 8 months for patients managed with supportive care alone. The evidence suggests that. states that ‘irinotecan and oxaliplatin are recommended as possible treatments for people with advanced colorectal cancer. respectively (Thirion et al. In a meta-analysis of patients treated in trials of 5-FUbased chemotherapy. were both evaluated in the 1990s but have not been shown to be effective (James et al. 14 months for patients with ECOG performance status scores of 2. 1999). 2005). and mitomycin.6 to 20.0002).. unless they are taking part in a clinical trial’ (NICE. cetuximab and panitumumab. irinotecan. Palliative chemotherapy: fluoropyrimidines A meta-analysis of five trials of palliative chemotherapy has shown an improved survival with chemotherapy compared with best supportive care (p = 0. Monoclonal antibodies At least two monoclonal antibodies are effective when given with standard chemotherapy regimens in metastatic CRC: bevacizumab and cetuximab. The optimum duration of treatment for chemotherapy is not known. the standard therapy for fit patients is now combination chemotherapy with a fluoropyrimidine and either oxaliplatin or irinotecan (NICE. cetuximab plus irinotecan has been found to double the response 169 . to 22 to 24 months for patients receiving combination regimens including irinotecan. 1997). A standard approach in the UK is to give chemotherapy for 12 weeks followed by a break.

. which included 5 trials but only 1342 patients. Another area of interest is in evaluating the role of ultra-small-particle iron oxide–assisted MRI. cost and patient preference must be taken into account 170 . Timothy Maughan and Tom Crosby rate.uk/ portfolio/dbase. 2004). Current clinical trials All NCRN trials can be found at www. For adjuvant chemotherapy for rectal cancer. Toxicity.Richard Adams. The CHRONICLE study tests adjuvant chemotherapy (capecitabine/oxaliplatin) in patients with completely resected locally advanced rectal cancer who had received preoperative long-course chemoradiotherapy. Follow-up for colorectal carcinoma after radical therapy There is controversy over the intensity of follow-up that should be used after resection of a primary colorectal carcinoma and this question is being addressed in an ongoing UK trial (FACS) that looks at options of intensive surveillance with hospital clinical and radiological surveillance to GP-based follow-up.. The evidence pointed towards a reduction in cancer-related mortality of 9 to 13% with an average earlier pickup rate of metastatic disease of 8.4. including QUASAR 2 (capecitabine ± bevacizumab). For colon cancer.gov/clinicaltrials. 2002). is a randomised crossover clinical trial comparing bolus fluorouracil/leucovorin to capecitabine as treatment for moderate.5 months. increasing time to disease progression and trending towards a survival advantage (Cunningham et al. showed a significant benefit from intensive follow-up. The CAPP2 study is a randomised controlled trial of colorectal polyp and cancer prevention using aspirin and resistant starch in carriers of HNPCC (Lynch syndrome). An overview by Renehan et al. Panitumumab. it is hoped that trials such as the large (> 2400 patients) phase III COIN trial will give further evidence for the combination of oxaliplatin. New biological therapies are being studied to treat metastatic disease.ncrn. a human anti-EGFR monoclonal antibody has been licensed as monotherapy for chemorefractory disease. Prognosis of colorectal carcinoma The 5-year survival by stage is shown in Table 13.cancer. In adjuvant therapy for colon cancer.. In the UK. AVANT (FOLFOX ± bevacizumab) and FFCD-PETACC-8 (combination chemotherapy ± cetuximab). because small phase II trials that used this combination have shown response rates of up to 81% (Van Cutsem et al. 5-FU and cetuximab. when considering which therapy is most appropriate. ongoing trials in the UK and abroad are evaluating the additional benefit of adding biological therapies to chemotherapy. The COIN trial is a recently launched trial that compares continued chemotherapy plus the EGFR antibody cetuximab versus intermittent chemotherapy with standard palliative combination treatment using oxaliplatin and a fluoropyrimidine in first-line treatment of patients with metastatic CRC. OxaliCap-RT integrates intravenous oxaliplatin plus oral capecitabine with pelvic radiation for rectal cancer. generally every 3 months for the first year and subsequently spreading out to every 6 months or annually (Renehan et al. Genetic Factors in Colorectal Cancer studies the role of genetic factors in clinical outcome for CRC patients. Areas of current interest No consistent predictive molecular or chromosomal biomarkers has been found to date for prognostic or therapeutic gain. the ongoing Cancer Research UK-sponsored CHRONICLE trial will attempt to define the benefit of adjuvant chemotherapy in those patients who have received neoadjuvant chemoradiotherapy.. PACT. FACS is a randomised controlled trial that assesses the cost-effectiveness of intensive versus no scheduled follow-up in patients who have undergone resection for CRC with curative intent. which has yet to adopt either of these monoclonal antibodies as standard therapy. 2004). this is being further examined in the UK PACT trial.org. oral fluoropyrimidines probably exert a toxicity profile that differs slightly compared to weekly bolus 5-FU and FA. NSCCG stands for the National Study of Colorectal Cancer Genetics. or Patient Preferences in Adjuvant Colorectal Cancer Therapy. The FOCUS 2 trial is a randomised trial to assess the role of irinotecan and oxaliplatin in advanced incurable CRC for less fit patients.to high-risk resected CRC.asp? A list of global trials is available at www. which may enhance the positive predictive value of nodal involvement in rectal cancer.

163. study of fluorouracil. 707–11. J. J. Lancet. Pahlman. T. H. Allen. Lancet. QUASAR: a randomised study of adjuvant chemotherapy (CT) vs. I. 350.. (2001). Ann. 1588–96. Med. O. Lancet.. S.. 1200–12. (2003). 1472–7. S. Ryall. J.. as adjuvant chemotherapy for colorectal cancer: a randomised trial. 2335–42. (2005). Hurwitz. The Wales Polyp study is a screening study that looks at genes and multiple colorectal adenomas. QUASAR Collaborative Group. M. Abstr. et al. Lancet. and Heald. Clin. Nutr. S. 31. (2005).. H. H. 337–45. Med. (1996).. 509–20.. Oncol.. A. N. Preoperative radiotherapy for resectable rectal cancer: a meta-analysis. Slavin. 457–60. 23. J. 60. colonoscopy or barium enema for the diagnosis of CRC in older symptomatic patients. Day. G. Eur. Radiol. W. Birgisson. Boni. J. K. (2005). Lancet. followed by excision for locally advanced rectal cancer. Cancer. 284. Oncol. fluorouracil and leucovorin for metastatic colorectal cancer.. Giunta. (1990). C. et al. (1994). L. Gray.. or both.. Engl.. 947–55. Engl. D. Colorectal Cancer Collaborative Group. J. Long term results. J. R. Birgisson. et al. 1008–15.. Donaldson. 358. Hills.. Mohamdee. Barnwell.. J. 224. (2002). J.. Dis. 351. complications. 759–67. Maughan. Colon cancer screening strategies. Haller. F. Robinson.. Peeters. Urquhart. 11. O. No. Oncol.. Fiorica. Oncol. Gastroenterol. 5644–50. Fong. 2343–51. Camma. Mounedji-Boudiaf. 23. QUASAR Colorectal Cancer Study Group. et al.. S. P Buyse. 59–63. D. G.. J. N. (1999). C. 35–41. . Agrawal. Athreya. et al. Levi. J. Chemotherapy in the elderly. Clin. et al. Kerr. Adjuvant chemotherapy with 5-fluorouracil. M.. (2005). 362. 1496–501. Adjuvant radiotherapy for rectal cancer: a systematic overview of 8507 patients from 22 randomised trials. R. same gain. D. Y. 348. 350.. 644–57. et al. (1998). James. Fearon. et al. Anonymous.. Gray. J. 14S (July 15 Suppl. E. Soc. Adjuvant chemotherapy in colorectal carcinoma: results of a meta-analysis.. et al.. Jr. The QUASAR 2 trial studies postoperative adjuvant chemotherapy capecitabine plus bevacizumab in colon cancer. A genetic model for colorectal tumorigenesis.. higher-dose folinic acid..Colon and rectum PICCOLO is a phase II study that will look at irinotecan (Ir) versus Ir plus panitumumab versus Ir plus cyclosporin in fluorouracil-resistant advanced CRC. J. 3501. Catalano. (1996). R... G.. Proc. D. R. Oncol. and Vogelstein. 90. the answer for lowering mortality in colorectal cancer? Recent Results Cancer Res. Dietary fibre in food and protection against colorectal cancer in the European Prospective Investigation into Cancer and Nutrition (EPIC): an observational study. 8671–8. et al. N. Semin. J. D. palliation and quality of life with three chemotherapy 171 . Delvart. Ann. Erratum in Lancet. R. R. I. Spencer. T.. Y. et al... Pahlman. Adam. 1000.. H. 254–63. Macdonald. S. Med. Curr. et al... et al. Autier. 341. Novotny. Engl. 355.. Oxaliplatin. R. and leucovorin as adjuvant treatment for colon cancer. Resection of nonresectable liver metastases from colorectal cancer after neoadjuvant chemotherapy. 91–4.. Oncol. Pascal. Ann. Bismuth. Surgical therapy of hepatic colorectal metastasis. 361. (2003). R. D. (2004).. 23. and Jenicek. 2004 ASCO Annual Meeting Proceedings (Post-Meeting Edition). (2000). (2003).M. E.. 30. Opin. E. (2000).. Andre. et al. M. Oncol. Surg.. D.. D. (2002). Randomised clinical trial of adjuvant radiotherapy and 5-fluorouracil infusion in colorectal cancer (AXIS). et al. Cell. et al. Hardcastle. J. S. J..). L. J. Colon Rectum. (2005).. C. R. Comparison of fluorouracil with additional levamisole. Lichtman. Randomised controlled trial of faecal-occult-blood screening for colorectal cancer. Am. Putter. J. 240. Role of circumferential margin involvement in the local recurrence of rectal cancer. L. 861–8. et al.. toxicity and quality of life in the TME trial.. Key. J. M. V. J. P J. L. leucovorin and levamisole in high-risk stage II and III colon cancer: final report of Intergroup 0089. 40. and Syngal. and capecitabine. Surg. Swedish Rectal Cancer Trial: long lasting benefits from radiotherapy on survival and local recurrence rate. G. Gunnarsson. B. R. RICE is a phase I/II study of radiotherapy.. 8697–705. Dube. 22. (2000). effectiveness and outcome – 5-year review. Lancet. T. Clin. U.A. R. 360. R.. SIGGAR 1 involves CT colonography. F. The effect of diet on risk of cancer. et al. MacFarlane. M. REFERENCES Adam. G. 85–96. Folkesson. 160–74. N. McConkey.. Phase III .. Br. Clin. J. (2006). Comparison of survival. (2004). et al. Chamberlain. Marijnen. Is FOB screening really . 166. 514–23. 1291–304. J. Vol. D. Marquart. Fehrenbacher. Martin. H. James. Siena. L.. Semin. Bevacizumab plus irinotecan.. J. Lancet. P Boyle. N. D. Oncol. et al. irinotecan. Bingham. 344. Gray. Cunningham. Humblet. (1993). Mesorectal excision for rectal cancer.. M. L-folinic acid and levamisole for patients with colorectal cancer: non-randomised comparison of weekly versus four-weekly schedules – less pain.. Luben. (2004). E. Kerr. (2004). Adam. (2004). Rescue surgery for unresectable colorectal liver metastases downstaged by chemotherapy: a model to predict long-term survival. S. fluorouracil. A.. (2004). J. Heyen. Colorectal stenting for colonic obstruction: the indications. J. Clin. F. et al. Surg.. 61. XERXES is a phase I/II study that looks at cetuximab and capecitabine in preoperative chemoradiotherapy for rectal cancer.A.. Jacobs.. Moss. H. Adverse effects of preoperative radiation therapy for rectal cancer: long-term follow-up of the Swedish Rectal Cancer Trial. (1997). Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. Gastrointestinal Cancer Symposium. R. observation including 3238 colorectal cancer patients. C.. 21. Whole-grain intake and cancer: an expanded review and meta-analysis. and Salo.. K. S. 26.

.). Thirion. Engl. J. 2006 ASCO Meeting Proceedings Part I. OPTIMOX1: a randomised study of FOLFOX4 or FOLFOX7 with oxaliplatin in a stop-and-go fashion in advanced colorectal cancer – a GERCOR study. J. 3511. Oncol. 24. Oncol. Figer.. T. Vol. 3500. P. Diagnostic accuracy of preoperative magnetic resonance imaging in predicting curative resection of rectal cancer: prospective observational study. D. T. B... C. E. Douillard. (1999). Refining molecular analysis in the pathways of colorectal carcinogenesis.. Egger... J. Li. Renehan. Lynch syndrome) proposed by the International Collaborative group on HNPCC. (2004). adjuvant treatment for stage III colon cancer. 1731–40. No. 359. Randomised phase III trial comparing infused irinotecan/5-fluorouracil (5-FU)/folinic acid (IF) versus 5-FU/FA (F) in stage III colon cancer patients (pts).. NHS Executive.. Watson. No. Hepatol. Med. Kahn. 10.).. 333. Routine short-course pre-op radiotherapy or selective post-op chemoradiotherapy for resectable rectal cancer? Preliminary results of the MRC CR07 randomised trial. Oncol. Clin. N. Niedzwiecki.. et al. E. 18S (July 15 Suppl. Rowan. No. 172 . (2004). Oncol. Ir) use and sequencing in advanced colorectal cancer (ACRC): the UK MRC FOCUS (CR08) trial. Int. oxaliplatin. J. et al. Gastrointestinal Cancers Symposium. CPT-11 (irinotecan. 24. J. NHS Executive. G.. Halford. M. Meta-Analysis Group in Cancer. Oncol. Oncol. E. (2006). 1107–8. Clin. Prospectively randomised trial of postoperative adjuvant chemotherapy in patients with high-risk colon cancer. J. A. Survival impact of . P. (1999). J. S. 23. Haddad. M. Ychou. et al. 22. F. (2005). J. Cervantes.. Part I of II (June 1 Suppl. chemotherapy in patients with colorectal metastases confined to the liver: a re-analysis of 1458 non-operable patients randomised in 22 trials and 4 meta-analyses. D. Wieand. Van Cutsem. Vol. (OX). (1999). Oncol. Oncol. (2000). 813.. Oxaliplatin and Raltitrexed (Review): Guidance. Gaasenbeek. Technical advances in the treatment of patients with rectal cancer. Gastroenterology. Vol. TNM Classification of Malignant Tumours. N... Oncol. 45.. Hohenberger.J.M. K. Clin. Tabernero. London: Department of Health. J. 1115–23. J.. Part I of II (June 1 Suppl.. J. B. R. 182. Vol. Sobin and Ch. Guidance Improving Outcomes in Colorectal Cancer: The Research Evidence. Coll. 8.M. (2004).. (2005). Steele. C.. J. (2005). J.J. A phase II randomized trial of LV5FU2+CPT-11 vs. Outcomes after detection of metastatic carcinoma of the colon and rectum in a national hospital system. (1996). (2006). et al.. MERCURY Study Group. Kuebler. Sebag-Montefiore. (1997). Twelves. 3500. Clin... J. Radiat. Labianca. P. Vasen.. A. pp. et al. Van Cutsem. A phase III trial comparing FULV to FULV + oxaliplatin in stage II or III carcinoma of the colon: results of NSABP Protocol C-07. Am. Sauer. Gastro. Wittekind. Lancet. 165. et al. J.. Capecitabine as . Laurie. Clin. M. P Wolmark. Colorectal Cancer (advanced) Irinotecan. H. 295–300. Clin. Raoul.. Nowacki. Surg.). Biol.). Oncol. New York: Wiley-Liss. London: Department of Health. 72–6. J. N. Preoperative versus postoperative chemoradiotherapy for rectal cancer. Dianz-Rubio. 324. Hossfeld. A. 23 (16 Suppl. S.. P et al. 116. H. 16. 353–61. M.. Ann. 16S. New clinical . R. 2005 ASCO Meeting Proceedings.. 1453–6. 1317–20. J. P et al. 16S.Richard Adams. J. Clin. B.. 1555–63.. R. An international phase II study of cetuximab in combination with oxaliplatin/5-fluorouracil (5-FU)/folinic acid (FA) (FOLFOX4) in the first-line treatment of patients with metastatic colorectal cancer (CRC) expressing epidermal growth factor receptor. Ann. Wade. 339. Referral Guidelines for Suspected Cancer. ASCO Annual Meeting Proceedings. G. D.. N. Quirke.. (1998). Wolmark. P. et al.. Saltz.. (2005). M.. P. C. Tournigand. L. et al. Phys. 351. Am. 6th edn. Virgo. Technology Appraisal 93. M. 23. Proc. H. 2004 ASCO Annual Meeting Proceedings (Post-Meeting Edition). 3502. (2002). 2696–704. Fluorouracil (FU). et al.. S. ed. UICC. Irinotecan plus fluorouracil/leucovorin (IFL) versus fluorouracil/leucovorin alone (FL) in stage III colon cancer (intergroup trial CALGB C89803).. 3). et al. 779.. London: NICE. NICE. Maughan. Saunders. M.. criteria for hereditary nonpolyposis colorectal cancer (HNPCC. Soc. D. G. J. Abstr. 394–400. et al. Willett. Med. A. H. Timothy Maughan and Tom Crosby regimens in metastatic colorectal cancer: a multicentre randomised trial. A.. O’Connell.. Cancer Guidance Sub-group of the Clinical Outcomes Group. 352.. (2005). Clin. Clin. 3. ASCO Annual Meeting Proceedings. (2005). et al. (PETACC 3). I). M. L. Oncol. 15 (Suppl.. Engl. A. (2006). J. Becker. LV5FU2 alone in adjuvant high risk colon cancer (FNCLCC Accord02/FFCD9802). Impact on survival of intensive follow up after curative resection for colorectal cancer: systematic review and meta-analysis of randomised trials. (2002).. Hollis. E. W. No. (2005). Wong. 18S (June 20 Suppl. Mecklin. et al.

o.Colon and rectum Appendix. b. days 1–14 plus oxaliplatin 130 mg/m2 over 2 hours 1 day 2 weeks 3 weeks 3 weeks 2nd line Unlicensed metastatic 175 mg 2 days 14 days 175 mg 14 days 2 days 3 weeks 14 days 1st line 1st line with potentially resectable liver mets 2nd line Bolus/infusion 5-FU 400 mg/m2 bolus then 2.d.8 g/m2 over 46 hours 5-FU 300 mg/m2 per day 12 weeks 1st line 175 mg 2 days 14 days 1st line leucovorin 20 mg/m2 20 mg/m2 200 mg/m2 Duration Day 1–5 30 weeks 2 days Interval 4 weeks 1 week 14 days Use 1st line Adjuvant 1st line 173 .d.4 g/m2 over 46 hours plus irinotecan 180 mg/m2 350 mg/m2 Capecitabine 1000 mg/m2 p. 2. 5-FU 400 mg/m2 bolus then 2. Colorectal chemotherapy regimens Dose of Regimen Mayo QUASAR de Gramont Mode Bolus Bolus Bolus/infusion Doses of chemotherapy 5-FU 425 mg/m2 5-FU 370 mg/m2 (5-FU 400 mg/m2 bolus then 600 mg/m2 over 22 hours) D1 and 2 Modified de Gramont Lokich Capecitabine Oxaliplatin modified de Gramont Irinotecan modified de Gramont Irinotecan XELOX (capecitabine and oxaliplatin) Short infusion Oral/infusion Bolus/infusion Continuous infusion Oral Bolus/infusion 1250 mg/m2 b.4 g/m2 over 46 hours plus oxaliplatin 85 mg/m2 400 mg/m2 bolus.

. These comprise the majority of tumours. The presence of the virus.5) but cancer of the anal margin is slightly more common in men. but there is no evidence to support this or an association with ulcerative colitis. Therefore. 174 Carcinoma of the anus Risk factors and aetiology Anal cancer was previously believed to be associated with chronic inflammation from haemorrhoids. have an excellent outcome with the chemoradiotherapy approach. Men with anal cancer were more likely to be single or to have had homosexual relationships or to have practiced anal-receptive intercourse. Anal-canal tumours occur more often in women (male-to-female ratio of 1:2. The sexually transmitted agent implicated in anal cancer is HPV. like carcinoma of the cervix. approximately 200 deaths occur per year from this cancer.1. There are approximately 720 new cases reported per year in England (National Statistics. Other factors for disease include having multiple sexual partners. high incidence in areas of Brazil) with similar high rates of penile. are often moderately or poorly differentiated. with the evident physical and psychosocial benefits of organ preservation. T4 or N ≥ 1 disease and they have a significant risk of both locoregional and subsequent distant failure. The anal verge is the lower end of the anal canal.g. and carry a worse prognosis. wide geographical variation in incidence mirrors the relative risk of HPV infection (e. except during defaecation. Anal-margin tumours are usually small and well differentiated and are more common in men. vulval and cervical cancer. It is 3 to 4 cm long and its walls are kept in apposition by the sphincter muscles. 2005). especially types 16 and 18. having other sexually transmitted viral and . Incidence and epidemiology The annual incidence of anal cancer in the UK is approximately 1. The tumours can be divided into three types: 1. HPV positivity or negativity does not appear to affect survival. Half of patients still die within 5 years of treatment.5 in 100 000. ages 35 to 40 years.. which constitute approximately 10% of anorectal tumours. Anatomy The anal canal extends from the rectum to the junction of the hair-bearing skin of the perianal region. treatment has swung away dramatically from primary surgery to definitive chemoradiotherapy. and fissures. N0. Anal-canal tumours are more common than those at the anal margin.5). efforts are ongoing to improve outcomes via better local and systemic therapy. 1997). In the past few decades. There has been a slight but steady increase in incidence in Western countries. The peak incidence age for anal cancer is 60 to 65 years but there is a bimodal distribution that includes a younger group. Dual-component tumours have both anal-margin and -canal components and it is not possible to define from where the tumour originated. 3. Unfortunately. can be found in more than 90% of cases (Tilston. Types of anal tumour Types of anal tumour are shown in Table 14. The anal margin is the perianal skin immediately adjacent to the distal limit of the anal canal. fistulae. Tumours of lower stage T1 or T2. many patients present with T3.14 ANUS Richard Adams and Tom Crosby Introduction Anal cancer. 1987). as well as to have had sexually transmitted diseases (Daling et al. 1982). is strongly associated with human papilloma virus (HPV) infection. 2. They are more common in women (male-to-female ratio ∼1:2. A case-control study performed between 1978 and 1985 established a link in women between anal cancer and genital warts (Daling et al.

vagina. etc. However. The tumour is usually flat or raised. an increased incidence of anal cancers is associated with HIV/AIDS. Spread There are abundant lymphatic channels in the anus. Overall. and ulceration is more likely to represent invasive disease. laterally. The pathological features of squamous carcinoma of the anus are shown in Table 14. and iatrogenic immunosuppression for organ transplant. bacterial infections. A negative FNA does not exclude malignancy. in women. 175 Pathology Squamous carcinomas represent 90% of anal cancers. Types of anal tumour Type Benign Examples Condylomata Conditions that may cause swelling/ ulceration Haemorrhoids Fissure Fistula Abscess Crohn’s disease Primary malignant Anal canal Squamous cell carcinoma: Large-cell keratinising Large-cell non-keratinising (transitional) Basaloid Adenocarcinoma Small-cell carcinoma Malignant melanoma Lymphoma Carcinoid GIST Undifferentiated carcinoma Anal margin Squamous cell carcinoma Basal cell carcinoma Bowen’s disease Paget’s disease Secondary malignant Metastatic spread from other tumours: case reports only Direct spread from rectum. submucosally to the rectum and bladder. to the prostate and downward to the perianal skin. Cloacogenic. keratinising and HPV +ve in about two-thirds of cases of squamous carcinoma. And. excision biopsy is rarely performed in clinical practice. they may be collectively described as epidermoid tumours.Anus Table 14. vagina or vulva. despite this fact. Anal intraepithelial neoplasia (AIN) Anal intraepithelial neoplasia is graded from one to three as in CIN (cf. although anal cancer is not an AIDSdefining malignancy. to the vagina/urethra. Table 14. but ulcerative and infiltrative when larger Canal tumours tend to be composed of small cells with basaloid features and non-keratinisation and HPV +ve ∼100%. basaloid and transitional are all variants . and in men. Spread direct from primary Cancer spread from the primary site is upward. which may also be keratinising and non-keratinising.2. of these. Because the natural histories and prognoses for all these subtypes are similar. Perianal tumours are usually large-cell. GIST = gastrointestinal stromal tumour. cervix. cigarette smoking.1. a history of cancer or intraepithelial neoplasia of the cervix. carcinoma of the cervix) and is considered precancerous (the precursor of squamous cell carcinoma of the anus).2. into the ischio-rectal fossa and sphincter muscles. Pathological features of squamous carcinoma of the anus Features Macroscopic Microscopic Description Nodular or plaque-like when small. only about 50% have malignant lymphadenopathy cytologically confirmed upon fine needle aspiration (FNA). approximately 25% of patients have palpable inguinal nodes at presentation.

4. A chest X-ray. bladder). 176 TNM classification and stage groupings The TNM classification and stage grouping are shown in Tables 14. The staging applies to anal-canal tumours.3 and 14. counselling and an HIV test should be performed. MRI is more effective at imaging the primary tumour and perirectal disease. rarely. It is rare for anal cancers to present with metastases outside the pelvis. including the hypogastric and obturator nodes and not infrequently to paraaortic/retroperitoneal nodes. A full blood count. followed by the inguinal nodes. Treatment overview Radiotherapy replaced surgery as the initial treatment in the 1980s. In patients with extensive tumour or excessive pain and tenderness. less frequently to the lungs and bones.Richard Adams and Tom Crosby Table 14. respectively. more commonly. and. urethra. Metastatic spread Haematogenous metastatic spread is to the liver. A tumour is occasionally found during an investigation of malignant inguinal lymphadenopathy. examination under anaesthetic (EUA) may be required. Stage grouping for carcinomas of the anal canal Stage Stage 0 Stage I Stage II Description Tis T1 T2 T3 Stage IIIA T1–3 T4 Stage IIIB T4 Any T Any T Stage IV Any T N0 N0 N0 N0 N1 N0 N1 N2 N3 Any N M0 M0 M0 M0 M0 M0 M0 M0 M0 M1 Adapted from UICC (2002). involvement of sphincter muscles alone is not T4 N1 N2 N3 Metastasis in perirectal lymph node(s) Metastasis in unilateral internal iliac and/or unilateral inguinal lymph node(s) Metastasis in perirectal and inguinal lymph nodes and/or bilateral internal iliac and/or bilateral inguinal lymph nodes M1 Distant metastasis present Table 14. The non-specific nature of the tumour can lead to a reporting delay of up to 6 months in one-third of patients. TNM classification of carcinomas of the anal canal Stage T1 T2 T3 T4 Definition 2 cm or less Greater than 2 cm. Bleeding.4. and then to external iliac vessels and the common iliac/para-aortic chain. Investigation and staging Examination without anaesthetic is often adequate to stage local disease (in ∼80% of patients). biochemical profile and. Lymphatic spread Low anal tumours. Biopsy of both the primary tumour and suspicious nodes in the inguinal region should be performed or. anal verge tumours and anal-margin tumours spread to perirectal nodes. not more than 5 cm Greater than 5 cm Of any size invading adjacent organs (e. and MRI can be performed. but the false-negative rate is higher with FNA. Further investigations should be performed as clinically or biochemically indicated. Evidence that concurrent chemoradiotherapy (CRT) improves local disease control compared to . computed tomography (CT) of the abdomen and pelvis. FNA. Adapted from UICC (2002). Stage classification Clinical presentation The tumour occurs as a lump or mass. Mid and upper anal-canal tumours spread via the internal iliac nodes to the pelvis.3. anal-margin tumours are staged in the same way as skin tumours. either found by the patient on wiping or causing the patient discomfort.g. vagina. to the brain. discharge and anal discomfort occur in about 50% of patients and about 25% are aware of a mass. if risk factors are present.

On examination of the surgical specimen. consisted of 5-FU alone. The chemotherapy regimen was 5-FU. Two further trials in the USA and Europe confirmed the superiority of local control of CRT over radiotherapy alone. 1999). treating to a dose of 30 Gy (Hu et al. as opposed to 94% treated with continuous-phase chemoradiation and 93% in the splitcourse CRT group. two cycles.. delivered a preoperative CRT schedule.Anus radiotherapy alone emerged in the 1990s. patients went on to an APR. 1 g/m2 on days 1 to 4 and mitomycin C..1997. 1 gm/m2 and mitomycin C. A 4-week gap was therefore introduced between the two treatment phases. local control was achieved in 60% of the patients treated with radiotherapy alone. If there was less than a 50% response. However.. If the tumour recurs following CRT. patients received a boost. albeit at the expense of increased acute toxicity (Bartelink et al. although this was not mandatory. 1996). The radiation given was 45 to 50 Gy in 25 fractions. 12 mg/m2 on day 1. 1983). 2004). Radical surgery for anal carcinoma Surgery for primary tumour Well-differentiated margin tumours less than 2 cm in diameter may be locally excised if margins of 5 mm or larger are possible. no more than two clinical oncologists should take responsibility for the care of patients with anal cancer and these should be core members of the anal cancer MDT (NICE. Within each radiotherapy facility. The incidence of nodal disease is less than 5%. The second cycle. The dose schedule was 45 Gy in 25 or 20 fractions with two cycles of 5-FU and mitomycin C following a 6-week break for assessment. Abdominoperineal resection (APR) is used for patients who are unsuitable for CRT. 1996). Marked acute toxicity was seen when radiotherapy was given with 5-FU and mitomycin C (two cycles of 5-FU.. either with brachytherapy 10 Gy/day with iridium-192 (25 Gy) or EBRT 15 Gy in six fractions. Radiotherapy technique Consider the use of a temporary defunctioning colostomy if the patient has a large obstructing tumour or incontinence. 50% were salvaged with surgery as opposed to 60% of patients who failed after radiotherapy alone. Simlarly. and ∼30% of all patients with anal cancer in the UK took part. The radiotherapy technique was as follows: the superior border was the bottom of the S–I joints and the inferior border covered the tumour by at least 3 cm. Flam et al. Local failure occurred in 39% of patients in the CRT arm versus 61% who underwent radiotherapy alone. (1984) described a cohort of patients treated with CRT and compared them to a similar group of patients treated with radiation alone. Where possible this type of cancer should be treated by local excision. The patient is simulated and treated 177 . but overall survival was not significantly different. from the Wayne State University Cancer Center in Michigan. the UKCCR published a randomised control trial of 585 patients (ACT I) that compared radiotherapy and CRT (Anal Cancer Trial Working Party. (1983). In 1996. APR is now rarely used as a primary treatment. The dose schedule was 30 Gy in 15 fractions with two cycles of chemotherapy (5-FU and mitomycin C. the only curative salvage therapy is APR. 7 out of 12 patients had no viable tumour (Nigro et al. Cummings et al. for patients with persistent/recurrent local disease or for patients who have unmanageable late toxicity after CRT. Chemoradiotherapy was associated with significantly greater toxicity. Adjuvant radiotherapy or CRT should be considered for patients who have undergone an excisional biopsy for an invasive tumour with positive margins. Radical radiotherapy and chemoradiotherapy Nigro et al. ACT I was a very successful trial. 4 weeks apart). and APR should be reserved for symptomatic uncontrolled multifocal disease. 1974. if radiotherapy is contraindicated – because the patient has previously had radiotherapy to the pelvis. Current studies are under way to find the most effective chemoradiotherapy regimen. there is a functioning transplanted kidney present in the pelvis. It was recommended that the inguinal nodes were included with the use of large parallel-opposed fields. NICE service guidance on colorectal cancer suggests that anal cancer should be managed by a network-based specialist team that includes an oncologist and a surgeon and that has established links with gynaecological and plastic surgical services. 10 mg/m2 ). Of patients who failed CRT. or there is already loss of continence due to the tumour – an end colostomy may be appropriate. given during the last week of radiotherapy. Anal cancer deaths were reduced by 29%. If there was greater than a 50% response. The management of high-grade AIN is uncertain. At 6 months.

a dose of 45 Gy in 25 fractions to the pelvic field including inguinal lymph nodes and a brachytherapy boost of 25 Gy are used. Dose. Large parallel-opposed fields are used to treat the phase 1 target volume. with hands under the head. 5-FU should be commenced at least 1 hour before the first fraction of radiotherapy.4 Gy in 28 fractions over 5. Mitomycin C. For phase 2 give 19. Treatment is dependent on the presence or absence of significant lymphadenopathy determined by CT or MRI imaging.8 Gy in 11 fractions to the ICRU reference point for macroscopic disease. In the absence of lymph node disease. Orthogonal films are taken. bolus on day 1 (max.) should be given throughout treatment. Anal-margin tumours may be treated with a single direct field with a 3 cm margin. When brachytherapy is used alone it is common to give a dose in the region of 55 to 60 Gy to the PTV. usually the point of the beam intersection using 10 MV photons.5 weeks to the ICRU 50 reference point. The total dose should be 50. it includes the known tumour and involved lymph nodes with at least a 3 cm margin and also treats potential lymph node involvement within the true pelvis and up to the common iliac nodes in higher tumours using a shrinking field technique.Richard Adams and Tom Crosby prone. The side effects of radiotherapy are shown in Table 14. is given by i. The phase 2 target volume is to treat macroscopic disease with at least a 3 cm margin. Field sizes are typically 14 to 16 cm (S–I) × 17 to 20 cm for large opposed fields. The bladder should be comfortably full. 12 mg/m2 . in which case the border is recommended to extend 3 cm above the upper limit of macroscopic disease. a three. which is the current UKCCCR anal cancer trial (see p. Unscheduled interruptions should be avoided. and prophylactic antibiotics (ciprofloxacin 250 mg b. and portal check films be taken on the set weekly and for the first three fractions of each phase.to four-portal plan should be used for all anal-canal tumours. 2003). r Occupy less than 50% of the circumference of the canal. If used in combination with external beam radiotherapy.5. the verification technique requires that simulator check films taken before treatment. 181). Brachytherapy should only be done in specialist centres with a skilled operator.d. r Are 1 cm or less in thickness. unless pelvic lymph nodes or the primary tumour extend to within 3 cm. For phases 1 and 2.6 Gy in 17 fractions to the ICRU reference point (usually midpoint) for microscopic and macroscopic disease. Bolus is applied to the natal cleft in all analmargin tumours and anal-canal tumours that extend to within 2 cm of the anal verge. such as small elderly women. the true pelvis and the known tumour with large anterior–posterior parallel-opposed fields. . The GTV should be marked on the film before the phase 1 and phase 2 field borders are defined. If GFR is 50 ml/minute or less (confirmed by EDTA clearance). Weekly FBC should be obtained.v. may have a spuriously low GFR as estimated by using the Cockcroft-Gault formula. fractionation and energy for phase 1 involves 30. r Lateral borders – lateral to the femoral heads to cover inguinal nodes completely. single dose 20 mg). which may lead to radionecrosis. The target volume for phase 1 includes inguinal nodes. Phase 1 and 2 volumes as described next are extracted from the ACT II protocol. If lymph node disease is evident. The PTV should be individually adapted. Implant/brachytherapy boost Implantation alone may be suitable as a boost for T1 and T2 tumours if they: r Lie below the anal-rectal ring. It is useful to instil a little barium in the anus and rectum and to use a radio-opaque marker (wire) to localise the anal verge. Planning using orthogonal films or CT planning and radio-opaque markers is recommended (for rectal contrast and an anal verge marker). Palpable lymph nodes should be marked with wire. or 3 cm below macroscopic disease for marginal tumours. r Have no nodal involvement. Some patients. r Inferior border – 3 cm below the anal margin. There is a high risk of dose inhomogeneity with implants in this region. omit mitomycin C. This technique was developed under the rationale that 30 Gy of CRT should be adequate to treat microscopic disease and 50 Gy is required to treat gross tumour (Melcher and Sebag-Montefiore. The following fields should be expanded to ensure coverage of the GTV plus 3 cm: r Superior border – 2 cm above the inferior aspect of the S–I joints. anterior–posterior 178 parallel-opposed fields are used to include the GTV plus a margin of 3 cm. Chemotherapy regimen in combination with radiotherapy The dose regimen is 5-FU 1000 mg/m2 per day by continuous 24-hour infusion on days 1 to 4 and days 29 to 32.

Side effects of radiotherapy for anal cancer Affect Acute side effects Lethargy Cutaneous toxicity from dry. postpone until under control Crampy abdominal discomfort Late side effects Sterility HRT for premenopausal females Offer sperm banking to males MUST BE DISCUSSED AT CONSENT Anal stenosis Faecal incontinence Telangiectasia of the rectum/bladder leading to bleeding Bowel fistula Chronic diarrhoea Small bowel resection if severe. Age. usually to moist. potassium citrate/cranberry juice Loperamide/codeine/low-residue diet. Provide adequate analgesia for the patient on ward. Haematological toxicity Advice re: coping with fatigue (e. Treatment for locally recurrent anal carcinoma Formal restaging is essential. often multiple loops affected Loperamide AP resection if failure to control conservatively AP resection if failure to control conservatively Surgery if failure to control conservatively Suspend treatment if guarding or significant tenderness Management Technique for brachytherapy The patient is prepared for theatre with twice-daily phosphate enemas. It is essential to have details of the initial tumour. EUA is performed to decide on the arrangement of the implant.. In a report from a single centre in the UK involving 254 patients with radiotherapy alone or with concurrent chemotherapy. pacing. In theatre the patient lies in the lithotomy position and is catheterised. 6 hours apart (alternatively 16.0 × 109 /l or platelet count ≤ 50 × 109 /l Cystitis Diarrhoea. delay RT if diarrhoea > 7 times per day and requiring parenteral support. Dose with high-dose-rate iridium-192. 15 Gy in two fractions. The second treatment. including diamorphine subcutaneous infusion. drinking fluids. hair dryer blowing cold can be soothing FBC once a week. Codeine is sometimes used over several days before treatment. Put in tubes with guide needles. and place index finger into the anus and to guide needles subcutaneously. and should be 30% longer than the tumour length. tenesmus Alkalinisation of urine.5.5 Gy in three fractions) or at a low dose rate of 20 to 25 Gy at 0. Apply a plastic obturator to prevent coning of needles at the apex. is done without anaesthetic but is usually tolerated reasonably well. usually between five and seven hollow applicators. because the tumour is likely to have responded substantially after the first phase of therapy. 2nd Skin® application is difficult in this area.Anus Table 14. Suture a circular or horseshoe template to the perianal skin. and removal of tubes. or incontinence. 2005). dose and stage were significantly associated with 179 . use hydrocortisone 1% cream Occasionally. delay if neutrophil count ≤ 1. all but 5 being within 3 years of primary therapy (Renehan et al. separated by 1 cm and not more than two-thirds circumference of the anal canal. Once satisfied with the position.5 Gy/ hour.g. there were 99 local disease failures. desquamation between buttocks. localised disease can be considered for radical resection (APR). tighten the screw to keep the needles in place. goal setting) Expose to air.

The average time of survival after diagnosis with metastatic disease is 8 to 12 months. reduced PTV or a reduced radiotherapy dose should be considered for this group. patients should be treated as if they have a low rectal adenocarcinoma and have an APR. the evidence for their use comes from studies in metastatic carcinoma of the cervix rather than anal cancer. but it can be rapidly progressive. EUA is reserved for patients who find clinical examination too uncomfortable or for those in whom local disease remains an issue. Node-positive patients have a 10 to 20% lower 5-year survival rate than node-negative patients. Recurrent disease in the inguinal nodes (in patients previously treated with CRT to this site and in whom there is local control of the primary) can be treated with block dissection.. respectively. Consider a radiotherapy boost/ implant if there is less than a 50% response or consider salvage surgery with APR after restaging. in more frail patients. preoperative radiotherapy with or without chemotherapy is frequently given. Patients treated with radiotherapy with or without chemotherapy are at increased risk of mucosal and cutaneous toxicity. Special problems Anal cancer in patients with HIV/AIDS Patients with HIV/AIDS are at increased risk of developing anal cancer. the 5-year survival rate with radical CRT 65%. However. Follow-up after radical treatment Clinically assess the patient at 6 weeks after the end of CRT and subsequently every 1 to 3 months. Patients with established AIDS or a CD4 count ≤ 200 cells/mm3 appear more likely to suffer local and haematological toxicity and it would seem appropriate to optimise retroviral therapy (HAART) prior to treatment. However. Palliative treatment for anal carcinoma Palliative radiotherapy/chemoradiotherapy Radiotherapy alone can be useful for patients who are unfit to receive chemotherapy. Of 73 patients who underwent surgical salvage the 3. Effective palliation/local control can also be achieved with a low-dose CRT regimen consisting of 30 Gy to the GTV plus a 3 cm margin with concomitant 5-FU (Charnley et al.or postoperative radiation. this treatment provides local control in about 80% of patients. respectively. If the malignant lymphadenopathy is mobile/non-fixed. it is 180 Prognostic factors The principal prognostic factor of anal cancer is disease stage (i. If possible one should avoid a large biopsy in a previously irradiated area because of the risk of radionecrosis. and the 5-year local control rate and rate of colostomy-free survival is 75%. such as cisplatin with 5-FU or mitomycin C with 5-FU. TNM). Adenocarcinoma Although long-term stoma-free disease control has been described following radiotherapy or CRT.Richard Adams and Tom Crosby recurrence. For anal-canal tumours. The procedure is associated with a significant risk of lymphoedema. giving treatment not previously used during primary CRT. All patients who are HIV +ve and have anal cancer (including those with CD4 count > 200 cells/ mm3 ) should be on highly active antiretroviral therapy (HAART). Useful palliation and even long-term disease control can be achieved with radiotherapy with a dose of 45 to 54 Gy in 25 to 30 fractions. the rates are 80 and 85%. Combination chemotherapy regimens are often used. . Consideration of a modified chemotherapy dose regimen. But in an attempt to reduce involvement of the circumferential margin and the risk of subsequent local recurrence. Such patients should be monitored closely. using 4 to 6 MV photons with bolus or. essential to confirm persistent or recurrent disease if an APR is being considered. Prognosis Survival Radiotherapy and radical surgery have similar survival and local control rates but radiotherapy has the advantage of being sphincter-sparing. Palliative chemotherapy Haematogenous metastatic disease tends to occur late. 6 Gy per fraction weekly for 5 to 6 weeks. Fungating inguinal nodes can be treated with 30 Gy in ten fractions.and 5-year survival rates were 55 and 29%. Distant metastases clearly have the greatest impact on survival. which is increased by pre. whereas the T stage determines the chance of local control.e. For anal-margin tumours. 2005).

Treatment differences between early (T1 to 2. Radiotherapy dose. Adapted from the ACT II protocol (available from www. Eschwege. 5-fluorouracil and mitomycin.1. (1997). J. Current clinical trials The current UKCCCR trial (ACT II) has been extended due to excellent recruitment and now includes more than 600 patients with anal cancer. 15... Bartelink. 2040–9. The EORTC is running a phase II/III randomised study (EORTC 22011) that looks at radiation therapy. Cummings. Roelofsen.uk/). Cancer. F.. N0) and locally advanced (T3 to 4 or N ≥ 1) disease and treatment in the elderly are also areas of interest. this trial addresses the issue of whether mitomycin C (MMC) or cisplatin with 5-FU and radiotherapy produce a greater complete 181 . but it is uncertain whether these are independent of the stage of the tumour.. 1049–54. GFR = glomerular filtration rate. Effective . et al. 92. and DNA aneuploidy are also factors affecting prognosis. EUA = examination under anaesthetic. Clin. Choudhury. et al. Oncol. Staged and biopsied by EUA and CT scan GFR > 50 ml/min Randomise 5-FU and MMC + Radiotherapy 5-FU and MMC + Radiotherapy 5-FU and CDDP + Radiotherapy 5-FU and CDDP + Radiotherapy No maintenance therapy Maintenance therapy of 5-FU and CCCP No maintenance therapy Maintenance therapy of 5-FU and CDDP Figure 14. N. Br. TP53 mutation. H. Lancet. (1984). Keane. Thomas.1 for the ACT II randomisation schedule. The ACT II study is investigating the role of ‘maintenance therapy’ for two cycles after CRT in squamous cell cancers. mitomycin C and either 5-FU or cisplatin in treating patients with locally advanced anal cancer.org. B.. 54.. Epidermoid anal cancer: results from the UKCCCR randomised trial of radiotherapy alone versus radiotherapy. G. T. Areas of current interest Adjuvant therapy has no proven role. MMC = mitomycin C. F.. CDDP = cisplatin. 5-FU = 5-fluorouracil. Concomitant radiotherapy and chemotherapy is superior to radiotherapy alone in the treatment of locally advanced anal cancer: results of a phase III randomized trial of the European Organization for Research and Treatment of Cancer Radiotherapy and Gastrointestinal Cooperative Groups. Ki-67. (2005). A. Results and toxicity of the treatment of anal canal carcinoma by radiation therapy or radiation therapy and chemotherapy. The prognosis in females is generally better than in males. The randomisation schedule of the ACT II trial. P et al. 348. Cancer. ACT II also looks at the role of maintenance therapy after CRT using two cycles of cisplatin and 5-FU. See Figure 14. 2062–8.ncrn. remission rate. REFERENCES Anal Cancer Trial Working Party. (1996). J. Charnley. mitotic rate. although patients with adenocarcinoma of the anus may receive this treatment in line with rectal cancers. Chesser. treatment of anal cancer in the elderly with low-dose chemoradiotherapy. CT = computed tomography. 1221–5.Anus Patients with confirmed primary epidermoid anal cancer. adjuvant/neoadjuvant therapy and the role of cisplatin (ACT II) and capecitabine in anal cancer are currently of interest.

et al. and Sebag-Montefiore. 1988–90. D. Cancer. England. 354–6. John. Klopfenstein. Cohen. Hu. UICC. L. 50. D. National Statistics. Role of mitomycin in combination with fluorouracil and radiotherapy. Engl. (1974).. Hislop. B. N. Oncol. F. Seydel. Jr. . et. 182 . 317. (2005). Sobin and Ch. Improving Outcomes in Colorectal Cancers. G. (2003).. (2002). Saunders. TNM Classification of Malignant Tumours. K. Combined preoperative radiation and chemotherapy for squamous cell carcinoma of the anal canal. (1982). Minsky. V. J. Renehan. al. J. R. A.. N. et.Richard Adams and Tom Crosby Daling. B.. J. et al. A.. Guidance on Cancer Services. J.. Surg. K.. 71–7.. (1996). 6th edn. Cancer. F. 70. H. Clin. 2527–39. 77–80... and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: results of a phase III randomized intergroup study. B. M. J. M. J. M. (1999). M. 34. A. (2005). 88. sexually transmitted diseases and the incidence of anal cancer. Surg. A. Weiss. Patterns of local disease failure and outcome after salvage surgery in patients with anal cancer. 605–14. Manual Update. . al. (1987). R.. Nigro. T. 625–34. Registrations of Cancer Diagnosed in 2003. (1983). et al. Oncol. M. New York: Wiley-Liss. G. P Schofield. Clin.. A. Pajak. 92. G. 14. ed. 17. N. Flam. S. London: Office for National Statistics. Cancer Statistics Registrations. Series MB1 no. 247. J.. Br. J. Correlates of homosexual behavior and the incidence of anal cancer. Anal human papilloma virus and anal cancer. A. Br. D. Nigro. 1352–7. Dis. 1826–9. London: NICE. Weiss.. N. pp.. Combined therapy for cancer of the anal canal: a preliminary report.. Pathol. P (1997). Sexual practices. NICE. Wittekind. Melcher. N. H. Concurrent chemoradiotherapy for squamous cell carcinoma of the anus using a shrinking field radiotherapy technique without a boost. and Considine. (2004).. S. L. P. L. Med. Daling. 973–7. D. Vaitkevicius. Considine. 30 Gy may be an adequate dose in patients with anal cancer treated with excisional biopsy followed by combined modality therapy. Colon Rectum. Tilston. et al.. 51. T... J.

There are rare familial forms of GIST. dropping to 35% at 10 years (note that these figures are before imatinib treatment). GISTs and ICC share morphological features 183 . there was also very little to offer GIST patients in the way of treatment. The majority (95%) of GISTs stain positively with CD117. Kindblom et al. GIST occurrence is rare before age 40. Some studies show a slight male predominance but most show an equal gender distribution of the disease. 2002). Of GISTs. Oncogenesis appears to be related to dysregulation of the proto-oncogene KIT. 2003). which may change the perceived incidence (Novartis. Histogenesis and pathology Histogenesis GISTs are thought to originate from the precursor of the ICC. which may be due to germline mutation in the genes for KIT or PDGFRα (Joensuu et al. and rare mutations are found in exons 13 and 17. 2002). There are 600 to 1200 new cases reported per year in England and 30 to 60 new cases per year in Wales. Of tumours that are morphologically similar to GISTs. The gastrointestinal autonomic tumour (GANT) is a phenotypic variant of GIST with neuroendocrine differentiation. Types of tumour Of gastrointestinal mesenchymal tumours. suggesting GISTs are an entity distinct from smooth muscle tumours. 2003). 2004). a growth-factor receptor tyrosine kinase (de Silva and Reid.. in the platelet-derived growth factor receptor alpha (PDGFRα. it is difficult to obtain an accurate count because of the cancer’s rarity and because of previous difficulty in diagnosis. However. GISTs represent 1% or fewer of all primary tumours of the GI tract and 5% of soft tissue sarcomas (Duffaud and Blay. GIST is the most common. In the past. the aetiology is largely unknown. 2003). GISTs originate from the interstitial cells of Cajal (ICC) – pacemaker cells that control gut motility.. 5% are CD117 negative. including a rare syndrome – the Carney triad – and a possible association with neurofibromatosis type 1. makes their incidence hard to estimate.. recent understanding of the molecular pathology involved in GISTs has made diagnosis more accurate and effective treatments are now available with molecular-targeted therapy. the median age at diagnosis is 50 to 60 years. True smooth muscle tumours also occur. The tumours are diagnosed by a combination of morphological features and immunohistochemistry staining. As research continues.15 GASTROINTESTINAL STROMAL TUMOURS Kate Parker and Tom Crosby Introduction Gastrointestinal stromal tumours (GISTs) are rare mesenchymal tumours that can occur anywhere in the gastrointestinal tract. is in exon 11 of KIT. GISTs have been difficult to diagnose in the past which. GISTs have become more reliably diagnosable since KIT expression has become detectable by immunohistochemistry. (2002) for more information on incidence and epidemiology. 60 to 70% express CD34. along with their rarity. Ten percent have a mutation in exon 9. Approximately 30% of these have mutations Risk factors and aetiology Because the tumour is rare. which is found in 70% of GISTs. Incidence and epidemiology The annual incidence in the UK is between 10 and 20 per million. The most common mutation. but Schwann cell tumours are rare. The 5-year survival is roughly 50%. it is likely that more molecular-targeted therapies will become available for this condition. and the prognosis was extremely poor. See Kindblom et al. an antibody that recognises an extracellular epitope of KIT (Corless et al.

They also both stain positive for CD117 (KIT). Microscopic appearance Microscopically. which is also known as the mast cell growth factor.2.Kate Parker and Tom Crosby Table 15.2. including melanoma.1. Tumours are 2 to 30 cm in diameter at the time of diagnosis but may be 1 cm or smaller if found incidentally. Studies have reported associations among mutation. and various other locations (omentum. histological appearance. KIT = kitten (v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog). Mutations in KIT often cause a change in these signals. The molecular classification of GISTs Type KIT (95%) Description Exon 11: 70% of patients Exon 9: 10% of patients Exons 13 and 17: rare (2% of patients) PDGFRα (∼1. Larger lesions may display cystic degeneration or necrosis.. Spread Modes of GIST spread involve spread by the primary tumour to involve contiguous structures. retroperitoneum. KIT is a transmembrane receptor tyrosine kinase. Adapted from Fletcher et al. Mutation in KIT is important but not felt to be solely sufficient to produce malignant transformation in a GIST. Metastatic spread commonly includes the liver and peritoneal .. GISTs can occur as three different types: r Spindle cell (70%) – relatively uniform eosinophilic cells arranged in short fascicles or whorls.5%) Wild type Familial GIST Paediatric KIT and PDGFRα mutations rare Exon 18: uncommon (6% of patients) Exons 12 and 14: rare (1. but it may occur following surgery. Pathology Macroscopic features GISTs are located in the stomach (60 to 70%). Ligand binding causes dimerisation of adjacent KIT molecules. fibroid-like appearance. KIT = kitten (v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog). 2002). The ligand involved is the stem cell factor. Immunohistochemical features of GIST Immunohistochemical feature CD117 (KIT) positive CD34 positive SMA positive S-100 protein positive Frequency (%) 90% 60–70% 30–40% 5% Table 15. The main role of the ICC is as pacemaker cells that control gut motility. ≤ 5%). which results in a gain of function – constitutive activation (Heinrich et al. Adapted from Corless et al. 184 Molecular classification The molecular classification of GISTs is shown in Table 15. small intestine (20 to 30%). Most GISTs are submucosal. (2004).. and anatomical location of GISTs (Kindblom et al. and oesophagus (5 to 15%). 2003). grow endophytically. Spread to lymph nodes is rare even when there is metastatic disease. SMA = smooth muscle actin. colon. 2002) and between mutation and prognosis (Fletcher et al. and seminoma. adhesion.1 shows the immunohistochemical features of GISTs. Immunohistochemistry Table 15.4% of patients) Molecular aetiology unclear Germline KIT or PDGFRα mutation CD = cluster of differentiation. angiosarcoma. PDGFR = platelet-derived growth factor receptor. (2002). and are often well circumscribed with a whorled. often via the signal transduction intermediates AKT and MAPK. S-100 = S-100 calcium binding protein. either mingled together or with an abrupt transition between them. and differentiation. which activates the intracellular signalling cascades affecting cell proliferation. r Mixed – areas of each of the aforementioned types. r Epithelioid (20%) – rounded cells with variably eosinophilic or clear cytoplasm. rectum. and contain a similar form of embryonic smooth muscle myosin. Metastatic deposits from other tumours can also be positive for CD117. mesentery.

Of all GISTs. Percutaneous biopsy is not advised. depending on the clinical context. The aim is macroscopic removal of the intact tumour mass regardless of its size. A proportion of these have a very low risk of malignant transformation and a firm diagnosis is not always made. GIST should be classified as localised or metastatic. Some evidence exists that preoperative imatinib may be beneficial and trials are likely to open in this area at some point (Eisenberg and Judson. The approach depends on the presentation. gastrointestinal obstruction (10 to 30%). 185 . 40 to 90% recur postoperatively. For tumours of the bowel. continue giving imatinib until the disease progresses or becomes operable.Gastrointestinal stromal tumours cavity. extended lymphadenectomy is not required. The primary aim is complete resection while avoiding tumour rupture. 2004). In the case of stable disease or patient response. because there is a risk of necrotic tumour leakage from the biopsy site. but rarely. but currently it should be noted that the role of imatinib in localised disease. Since GISTs rarely metastasise to nodes. Tumours larger than 2 cm Wide local excision is required for tumours larger than 2 cm. is experimental. Stage classification Malignant GIST lesions are not classified using TNM nomenclature. Lymph node involvement is classified as metastatic. surgery will amount to a formal segmental resection with the accompanying mesentery. abdomen and pelvis should be used to look for metastases. and the role of surgery in metastatic disease. If the tumour is fully resected. then a biopsy should be performed to confirm the diagnosis. If these tumours are not resected. then it should be resected following standard oncological principles. bone or subcutaneous sites are involved. Consider assessment of endocrine and αFP and βhCG levels to rule out adrenal tumours and teratomas. NICE. if the tumour is not operable. en bloc dissection is necessary with the help of appropriate specialist surgeons. If the tumour is inoperable or residual disease is present. If the tumour is operable. Symptoms include gastrointestinal bleeding (50%). then start imatinib therapy (if the patient is fit enough). 2006) and performed by a surgeon who is fully trained and experienced in cancer surgery in the relevant body area. either locally or with metastatic disease. Treatment Treatment overview Treatment involves staging investigations and discussion at a multidisciplinary team (MDT) meeting to decide operability. and general malaise. they should be re-imaged at 6 and 12 months with a CT scan or endoscopic ultrasound. with a margin of 1 to 2 cm. GIST is an incidental finding and patients are asymptomatic in 20 to 30% of cases. When adjacent organs are involved. It should be discussed within the relevant MDT (both the upper GI team and sarcoma team. Clinical presentation The patient initially presents typically with an intraabdominal tumour or as an emergency with intestinal haemorrhage or obstruction. Progressive disease should also be discussed at the MDT meeting (see the following section for options). Histopathological review by an experienced pathologist and MDT review should be done to confirm whether the disease is CD117-positive GIST and to assess the tumour’s aggressiveness. tiredness. abdominal pain (20 to 50%). Surgery Surgical resection remains the best hope for a cure. follows) and the patient should be considered for entry into adjuvant trials. Investigation and staging CT (or MRI) of the thorax. Tumours arising outside GI tract The tumour should be dissected from adjacent organs and then excised with a margin of normal tissue. Assess patient response at 3-month intervals and discuss the CT scan in an MDT meeting. PET scanning and endoscopic ultrasound may add useful information for the management of some patients. Lung. then follow-up should be done according to assessed risk (see discussion that Small tumours The management of asymptomatic small (≤ 2 cm) tumours is controversial.

The patient’s medication should be checked for drugs that interact via the p450 system. Radiotherapy could be considered as a palliative measure in advanced disease (Eisenberg and Judson. sunitinib). 2003a). If the tumour cannot be excised. there is no role for radiotherapy in primary treatment. The patient’s performance status and cardiac status should be assessed and baseline liver and renal function tests and FBC should be performed... excision of disease fol¨ lowing response to imatinib (Bumming et al. assess toxicity. Advanced disease Surgery may have a role in excision of recurrent disease. Patients on imatinib can present with acute gastrointestinal haemorrhage due to tumour response.. Verweij et al. These patients should be managed intensively because the prognosis of patients on imatinib may be reasonable. See van Oosterom et al. 2004). 2003b) and inoperable recurrent or metastatic disease confirmed by the MDT. Informed patient consent is needed (including advice to the patient on avoiding 186 . The patient should also avoid caffeine and grapefruit for 1 hour before and after receiving the dose and should avoid lying down for 1 hour afterward. 2004). 2003). Zalcberg et al. Upon stopping treatment with imatinib. In patients who have initially responded to imatinib but then become resistant.. once a day continuously (van Oosterom et al. debulking of advanced disease.. 2004). dose escalation should be discussed at the MDT meeting (dose escalation is controversial. and each tumour should be treated as appropriate. and perform a physical and symptomatic assessment of response. 2004).to 6-week intervals.Kate Parker and Tom Crosby Multiple tumours Patients with more than one primary GIST should be referred to the cancer genetics service for advice. Baseline investigations The disease should be assessed with a CT scan to allow accurate monitoring. even with progressive disease.g. and procedures for palliation of symptoms. Consider a second-line treatment (e. Confirmed disease progression NICE guidance dictates that imatinib should be discontinued when the disease progresses (NICE. Repeat this procedure at 4. If there is evidence of response or stable disease then continue treatment according to toxicity parameters.o. there is some evidence that therapy should continue despite disease relapse. with good quality of life. Repeat a CT or PET scan after 3 months and discuss the results at an MDT meeting to assess the patient’s response. The patient’s weight should also be checked. check LFT and FBC. Treatment The patient should be given imatinib 400 mg p. there are some cell populations responsive to imatinib). measure the patient’s weight (to determine fluid retention). Patients then require an appropriate staging scan and an MDT discussion. 2001. therefore. Repeat the assessment with CT every 6 months or earlier if clinically indicated. Emergency presentations Patients may present with an obstruction and need an emergency laparotomy. Monitoring Two weeks after starting imatinib. Radiotherapy GIST is a radioresistant tumour. Imatinib Indications Indications for use of imatinib are a confirmed KIT(CD117-)positive GIST (Verweij et al. it is important to watch for tumour flare (which suggests that. pregnancy and breast feeding). Cytotoxic chemotherapy GISTs are resistant to conventional cytotoxic chemotherapy but do respond to targeted therapy as outlined in the next section (Eisenberg and Judson. one study showed a 33% response to dose escalation but the median duration of response was only 3 months. In view of the tumour flare phenomenon and research showing an increase in FDG metabolism of GIST deposits after stopping imatinib treatment. a bypass procedure should be undertaken and the tumour should be biopsied. which should be taken with a large glass of water (to avoid gastric irritation). (2001) for more information on patient monitoring.

and the presence or absence of solid nodules within the tumours. Assessing response using contrast-enhanced CT A response to therapy can be seen on a CT scan as early as 1 month after treatment starts. but it can be severe. or has been discontinued because of progressive disease. These new criteria may make the determination of response assessment via CT more accurate. All patients should have a CT scan at 3 months. Patients with very-low-risk tumours need an annual review and no further CT scan unless clinically indicated. ascites).3). GISTs can decrease in size but may also increase in size (because of haemorrhage or myxoid degeneration). Treatment options should be discussed on a case-by-case basis at the appropriate MDT meeting. 2002). For patients with low-risk tumours. 2003). Follow-up Routine follow-up after surgery depends on the risk of recurrence (see the prognostic factors in Table 15. repeat a CT scan at 12 months.. and tumour vessels and solid enhancing nodules may disappear.. 2002). Imatinib is only currently licensed for use in recurrent or metastatic disease that is inoperable. it is usually mild and selflimiting. imatinib treatment should be stopped. Chemotherapy and radiotherapy are unlikely to be helpful. and although the mechanism is poorly understood it may be explained by mutational analysis (Blanke et al. ment withheld if levels are three times the upper limit of normal. 2001). If a skin rash occurs. Occasionally neutropenia can occur (monitor FBC). In addition. As well as changing in size. Surgery may be considered for these ‘tumour-resistant clones’ but the benefit of this approach is uncertain. patient management involves active supportive care that includes palliation of symptoms. Diuretic therapy is helpful but. and that they are discussed in the MDT meeting. the tumour can become more homogeneous and hypoattenuated on CT. However. Therefore. diarrhoea and dyspepsia. KIT-negative GIST The histological diagnosis should be confirmed by an experienced pathologist. the NICE guidance says that imatinib should only be used for KIT-positive GIST. and oedema. The standard RECIST criteria for assessing tumour response are therefore inadequate as the sole method to assess response (Silberman and Joensuu. 20% of GISTs do not take up appreciable FDG at baseline and so cannot be monitored in this way. However. Adjuvant therapy Currently there is no evidence for adjuvant use of imatinib. degree and extent of enhancement. Criteria to help predict response on CT. Where imatinib is inappropriate. FDG-PET is not routinely available in many UK centres and international response criteria for FDGPET have not been standardised. for these tumours. Gastrointestinal bleeding also occurs.Gastrointestinal stromal tumours Sometimes nodules of resistant disease grow in previously controlled metastases. CT remains the routine form of radiological assessment. it is important that the CT scans used to assess response are reported by radiologists experienced in assessing response in GIST. which include size and number of tumours. presence or absence of tumour vessels. A proportion of KIT-negative GIST patients still respond to imatinib. if oedema is severe. change in Hounsfield unit. Special clinical situations Positive surgical resection margins Residual disease should be excised whenever possible. Palliation Imatinib is used in the palliative setting as described earlier in the chapter. Perform an annual review thereafter and a CT scan only if indicated clinically. have been evaluated and were found to be better than RECIST criteria for assessing the response of GISTs to imatinib. often periorbital (occasionally pleural effusions. often due to response and tumour regression (van Oosterom et al. Role of PET Metabolic responses to imatinib can be reported as early as 24 hours after treatment and changes in FDG-PET images at 1 week have been reported compared to those seen on a CT scan at 2 months (Stroobants et al.. Abnormal LFTs should be checked and treat- 187 . Toxicities Imatinib toxicities include nausea and vomiting (level 1 antiemetics are usually enough to help).

GIST can recur many years after a curative resection. The median time to progression on imatinib is around 24 months (Von Mehren et al. repeat a CT scan at 9 months and then annually for 5 years in addition to an annual clinical review. 2006). C. with a significant benefit for sunitinib in progression-free survival of 6. Prognostic factors for GIST are shown in Table 15.3 months compared to 1. 188 New drugs: other targeted inhibitors of KIT activity Sunitinib Sunitinib is a small-molecule inhibitor of all VEGFRs. The Radiation Therapy Oncology Group phase II trial investigates the use of neoadjuvant imatinib to debulk tumours prior to surgery or to make inoperable tumours resectable. C. For patients with high-risk tumours. Dasatinib Dasatinib is a synthetic small-molecule inhibitor of tyrosine kinases including KIT. FLT3 and KIT that is being investigated for use in imatinib-resistant patients. ACOSOG Z9001 is a trial involving 732 patients receiving 400 mg imatinib daily for 1 year following R0 or R1 resection for GIST. Prognosis The 5-year survival rate is about 50% and the 10-year survival is about 35%. Prognostic factors for GIST Mitotic count per 50 Risk group Very low risk Low risk Intermediate risk Size < 2 cm 2–5 cm < 5 cm 5–10 cm High risk > 5 cm > 10 Any size high-powered fields <5 <5 6–10 <5 >5 Any rate > 10 Areas of current interest and clinical trials Clinical trials Trials looking at the role of adjuvant imatinib are currently open and recruiting in both Europe (EORTC and SSG) and the USA (ACOSOG): EORTC 62024 aims for 400 patients to have 400 mg imatinib daily for 2 years following R0 or R1 resection for non-metastatic GIST. Trials are needed to address the following questions: r What is the role of dose escalation of imatinib in patients who initially respond and then progress (median time to progression is 2 years)? A proposed trial involves imatinib 800 mg versus sunitinib. 2006. r What is the role of other new molecular therapies? A proposed trial involves imatinib ± bevacizumab. These figures are for patients treated before imatinib was available and before improvements in diagnostic accuracy by the introduction of CD117 immunohistochemistry. r What is the role of neoadjuvant imatinib? r What is the role of imatinib in KIT-negative tumours? Adapted from Fletcher et al.Kate Parker and Tom Crosby Table 15. For patients with intermediate-risk tumours. Phase III dose-randomized study of imatinib mesylate (STI571) for . 2004). (2003).. A clinical review should be performed every 6 months. R. PDGFR. 2002). PDGFRβ. Verweij et al.. perform a CT scan every 6 months for 3 years. and the median time to achieving CR or PR is 13 weeks (Demetri et al. It is active when given orally. The response to imatinib can be predicted in part according to mutational analysis.5 months in the placebo arm of the study (Faivre et al. PDGFRα. REFERENCES Benjamin. A phase III trial has recently been closed early. and it may have activity in imatinib-resistant patients. et al. Ranking. due to efficacy end points being reached. compared to 48% for exon 9 and 0% for PDGFRα mutation. 2003.. Demetri et al.3.. Benjamin et al. then annually over the next 5 years. BCR-ABL and SRC. (2002).. in one study exon 11 mutation was associated with an 84% partial response rate. around 67% partial responses and the remainder of patients with stable disease. Fletcher. S.. The Scandinavian trial SSGXVII involves 240 patients on 400 mg imatinib daily following R1 surgery or for a high-risk group for either 1 or 3 years. The response rate to imatinib is initially between 80 and 90% (which is derived from a very small number of complete responses.. 2002).3..

. Oncol. (2002).. Joensuu. Goeminne. Zalcberg.. (2002). and Judson. 364. adjuvant and palliative treatment of gastrointestinal stromal tumours (GIST) with imatinib: a centre-based study of 17 patients.. 18FDG-Positron Emission tomography for the early prediction of response in advanced soft tissue sarcoma treated with imatinib mesylate. van Oosterom. C. Oncol. pharmacokinetic. Cancer. Soc. Update of phase I study of imatinib (STI571) in advanced soft tissue sarcomas and gastrointestinal stromal tumors: a report of the EORTC Soft Tissue and Bone Sarcoma Group. S83–7. Verweij. Verweij. Safety. 465–75. 347.. et. survival in gastrointestinal stromal tumours with high-dose imatinib: randomized trial. P Andersson. et al. Pathol. L.. J. C. Zalcberg. comparison of two doses of Imatinib for the treatment of advanced gastrointestinal stromal tumors (GIST): interim results of a randomised phase III trial from the EORTC-STBSG. L. 2. 65. Casali. 4342–9. patients with advanced gastro-intestinal stromal tumours (GIST) crossing over to a daily imatinib dose of 800 mg (HD) after progression on 400 mg – an international. Fletcher. L.. H.. C. T. and Joensuu. NICE.Gastrointestinal stromal tumours GIST: Intergroup S0033 early results. Abstr. J. Dimitrijevic.. Berman. (2003). ISG and AGITG. Clin. Clin. and Blay.-Y. M. Diagnosis of gastrointestinal stromal tumors. Eur. (2003b). Abstr. C. P G.. A consensus approach. 21. G. M. G.. van Oosterom.. et al. Verweij. Am. (2006).. Gastrointestinal stromal tumors: biology and treatment. L. et al. Soc. Oncol. High incidence of durable responses induced by Imatinib mesylate (Gleevec) in patients with resectable and metastatic gastrointestinal stromal tumors (GISTs). I. S. Stroobants. 5). Casali. Am. Outcome of . (2001). Oncol. Proc. Incidence. Oncol. 2006–11. Imatinib for the Treatment of Unresectable and/or Metastatic Gastrointestinal Stromal Tumours.. 655–64. Casali. Soc.. A. (2001). The Manual. A. Abstr. 11. Demetri. 38 (Suppl.. Med. Clin. Am. J. and Heinrich.. Am. I. Delbaldo. 187–97. A. Verweij. et al. Guidance on Cancer Services. Pathol. Eisenberg. S. et al. C. et al. M. H. Early efficacy . and antitumour activity of SU11248. Am. 13 (Suppl. Oncol. Ann. J. et al. 33. (2004). 157. Novartis UK Ltd. Verweij.. J. J. (2003). Duffaud. Oncol. Corless. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. et al. Safety and efficacy of Imatinib (STI571) in metastatic gastrointestinal stromal tumours: a phase I study. Abstr. 9. G. P G. 358. I. (2006). Treat. Oncol. (2004).. Meis-Kindblom.. (2004). STI571. Curr.. C. Eur. Corless. (2004).. S. van Oosterom.. and Reid.. Proc. Evaluation of the safety and the efficacy of an oral molecularly-targeted therapy. A. Cancer. M. J.. Eur. D.al. Blanke. Vera. G. D. C. 1127–34. Fletcher. Engl. Joensuu. Silberman. Open. (2002). London: NICE. P G. et al.. J. Seegers. 472–80. Proc.. Proc. Lancet. CD117 expression. Eisenberg. in patients with unresectable or metastatic gastrointestinal stromal tumours (GISTs) expressing c-KIT (CD117). . a novel oral multitarget tyrosine kinase inhibitor. Blanke. et al. M. Overview of issues related to Imatinib therapy of advanced gastrointestinal stromal tumors: a discussion amongst experts. 460–4. J.. Cancer. (2002). Faivre. C. NICE. 22. Surg. Clin. 189 . Ann. (2003). ¨ Kindblom. J.. in patients with cancer.. J. J. Demitri. Fletcher. 9004. 1608. (2003). Soc.. A. M. Blanke. S. 15 December. 5770. phase II study of Glivec r in patients with unresectable or metastatic malignant gastrointestinal stromal tumors expressing c-kit. Cancer. Bumming. 89. Kinase mutations and imatinib response in patients with metastatic gastrointestinal stromal tumor. J. 5). and Heinrich. J.. Judson. phenotype and biologic spectrum of gastrointestinal stromal cell tumors (GIST) – a population-based study of 600 cases.. D. ¨ Bumming. Biology of gastrointestinal stromal tumors. 39. Gastrointestinal stromal tumours (GIST): C-kit mutations.. 1421–3.. R. Abstr. Neoadjuvant. 3813– 25. (2003). Soc. F. Br. J. J. T.. Hum. J. von Mehren. (2003). Demetri. Novartis. 38 (Suppl. et al.. Res. 13–19.. J. differential diagnosis and targeted cancer therapy with Imatinib. Proc. Oncol. Lancet. S66–9. Imatinib mesylate (STI-571 Glivec. (2002). Y. randomized. de Silva. Oncol. Oncology. Technology Appraisal 86.. Surgery and imatinib in the management of GIST: emerging approaches to adjuvant and neoadjuvant therapy. London: NICE. et al. B. H. Oncol. Judson. Von Mehren. 39. Improved survival and sustained clinical benefit with SU11248 (SU) in pts with GIST after failure of imatinib mesylate (IM) therapy in a phase III trial. 3. and Garrett. N.. Clin. C.. Management of malignant gastrointestinal stromal tumours. Clin.. J. K. (2002). Gastrointestinal Cancer Symposium. 5). Clin. et al. J. 485–91. Lancet Oncol. 2012–20. Options Oncol. Zalcberg. Heinrich. 4–5. J. C. Eur. J. D. J. B. M. D. Abstr. C. ISG and AGITG. Cancer. (2004). Gleevec) is an active agent for gastrointestinal stromal tumours but does not yield responses in other soft tissue sarcomas that are unselected for molecular target: results form an EORTC Soft Tissue and Bone Sarcoma Group phase II study. 24. J. J. et al. 8.. Corless. Verweij. Study No STI51B2222. 459–65.. D. C.. J. P. 3271. prevalence. Meis-Kindblom.. J. T. Improving Outcomes for People with Sarcoma. 3272.. (2002). L. C. Clin. Blay.. intergroup study of the EORTC. Progression-free . van Oosterom. (2006). (2003a). Report Data on File.

1 shows the range of tumours that can affect the breast. and this is believed to be due to screening. such as those in North America and Northern Europe.). the highest mortality is in Eastern followed by Western Europe. with one of the highest rates of decline occurring in the UK (Coleman et al. occurrence is fivefold higher in Western countries. the lowest in the Nordic countries (Coleman et al. patients were treated with radical mastectomy. although the validity of the comparisons made has been questioned and the reasons for any real differences are unknown. adjuvant hormone treatment. There are approximately 125 cases of male breast cancer each year in the UK. In the past. 1 in 29 before age 60 and 1 in 12 by age 80. There also appears to be a significant geographic variation in breast cancer mortality. 2003). patients are usually treated with breast-conserving techniques – wide local excision and radiotherapy. Types of tumour affecting the breast Table 16. Breast cancer is also rare among women younger than age 20. compared to Asian and African countries such as Japan. A few lymphatic channels also communicate with those in the opposite breast and in the abdominal wall. Each breast is divided into a central portion and four quadrants. increased use of better adjuvant chemotherapy) and the MDT (Thomson et al. The management of breast cancer has changed dramatically. RR = 2) and late age at first birth . Breast cancer is the most common malignancy in women in the UK. The breast cancer mortality rate has fallen since 1990. The risk of disease is 1 in 39 before age 50. The lymphatic drainage from the breast is primarily to the axillary lymph nodes but also to the internal mammary nodes. The upper outer quadrant also contains the axillary tail. improved treatment (improved surgical techniques. chemotherapy and. Reproductive factors include a younger age at menarche (if menarche is before the age 11. Breast cancer most commonly presents as a lump in the breast. relative risk [(RR)] = 3. The nipple is surrounded by the areola.7% of all male cancers).. 2004). the monoclonal antibody trastuzumab.5% of all breast cancers occur in men (0. Anatomy The female breast extends from the 2nd to the 6th rib. 190 Risk factors and aetiology Breast cancer occurs predominantly in women and is rare among men. when the rate of increase slows. China. The incidence increases with age. India and Zimbawe. doubling every 10 years until menopause. late natural menopause (if the menopause occurs after age 54. recently. As a result. whereas today.. 2003). and the breast tissue does not usually extend beyond the areola. More than 40 000 women in the UK develop breast cancer each year (which accounts for 29% of all cancer in women).. and it is made up of 15 to 20 lobes which radiate out from the nipple. there has been a steady fall in the mortality from breast cancer. The past few decades have also seen the wider use and development of systemic therapies: hormonal treatments.16 BREAST Nayyer Iqbal and Peter Barrett-Lee Incidence and epidemiology Introduction The lifetime risk of breast cancer in women is one in nine. which lie in the thorax alongside the internal thoracic artery. within Europe. There is a wide geographic variation in the ageadjusted incidence of breast cancer. the nipple and areola are small. but the use of screening has also allowed very early cancers to be diagnosed before they can be detected clinically. In the male and prepubertal female. 0.

The HABITS trial (hormonal replacement after breast cancer – is it safe?) showed that the use of HRT in the early years after diagnosis can increase the rate of breast cancer recurrences. and moderate or florid epithelial hyperplasia may have a slightly higher risk of breast cancer (1.24 (Anonymous. In a large meta-analysis. because the results suggested an unacceptably high risk for women with previous breast cancer who took HRT (Holmberg et al.0001). Women with palpable cysts. In this trial. When younger women are exposed to ionising radiation to the chest (e. after a median follow-up of 2.0001) (Beral and Million Women Study Collaborators. RR = 4. Tumours that affect the breast Type Benign Examples Fibroadenoma Solitary cyst Intraduct papilloma Adenomas Duct ectasia Epithelial hyperplasia (without atypia) Sclerosing adenosis Radial scar Complex sclerosing lesion Atypical ductal hyperplasia Atypical lobular hyperplasia Malignant primary Carcinoma in situ Tis (ductal carcinoma in situ) Tis (lobular carcinoma in situ) Tis (Paget) Invasive carcinoma Invasive ductal carcinoma NOS or NST (75% of cases) Invasive lobular carcinoma (10%) Medullary carcinoma (5%) Tubular carcinoma Mucinous (colloid) carcinoma Cribriform carcinoma Papillary carcinoma Adenoid cystic carcinoma Apocrine carcinoma Secretory carcinoma Squamous cell carcinoma Inflammatory carcinoma Paget’s carcinoma Metaplastic carcinoma Others Phyllodes tumour Sarcomas Vascular tumours Lymphoma Malignant secondary Melanoma Lung Ovary Kidney Stomach Thyroid (medullary) Rhabdomyosarcoma (alveolar) NOS = not otherwise specified.75]. RR = 9. In benign proliferative breast disease. however. RR = 2. RR = 3).e. the Million Women study showed that HRT users were more likely to develop breast cancer than non-users (adjusted relative risk 1. The data for oral contraceptives (OCs) are less consistent than those for HRT. The relative risk of breast cancer in current users was 1. Nulliparity has a RR of 3 compared to high parity. for atypical epithelial hyperplasia alone. use of OC within 10 years was associated with a slightly increased risk of having breast cancer diagnosed. sclerosing adenosis. 2004). Women who undergo bilateral oophorectomy before the age of 35 have a reduced risk (RR = 0. 1996). If diethylstilbestrol is used during pregnancy. 434 women were randomised to HRT or no HRT. complex fibroadenomas.g. Dietary factors in breast cancer include obesity and alcohol intake. If atypical hyperplasia is found in a woman with a family history of breast cancer in a first-degree relative. the breast cancers that developed were of favourable prognosis and did not affect the disease mortality. For exogenous hormone administration involving hormone replacement therapy (HRT). 26 women in the HRT group developed a new breast cancer event compared with only 7 women in the non-HRT group.66 [95% CI 1.Breast Table 16. The magnitude of risk was greater for patients taking oestrogen and progestin in combination than for oestrogen alone (p < 0. for Hodgkin lymphoma) then RR = 3.5 to 3 times) than women without these changes.1. Obesity increases breast cancer risk 191 .1 years. because of its selective population (i. The trial was stopped early.58–1. NST = no special type.6) compared to women who undergo natural menopause. p < 0.. This study has been criticised. However. women who accepted an invitation for mammographic screening). (if the first pregnancy is after age 40. duct papillomas. 2003).

5 to 1% of the general population and are probably breast-cancer prone. Ataxia-telangiectasia (A-T) is a rare autosomal recessive disorder that renders homozygous individuals more vulnerable to cancer because of an elevated risk of mutation. A moderate alcohol intakes increases breast cancer risk by about 10 to 20%. They are associated with early onset breast cancer. Of the different locations in the breast. and 5% are multicentric (i. Up to 10% of breast cancer cases in Western countries are thought to result from a genetic predisposition. r Mitotic index. The Nottingham combined histological grading of breast carcinoma (Elston-Ellis modification of the Scarff-Bloom-Richardson system) is based on: r Degree of tubule formation – > 75% gives a score of 1. in different quadrants). Half of affected females have fibrocystic disease of the breast and breast cancer. respectively. The BRCA2 mutation accounts for 1% of breast cancer. such as prostate. RR = 2. r The androgen-receptor gene is found on the Y chromosome. When a patient has previously had breast cancer in one breast. They are typically ER. scores 1 to 3. A-T heterozygotes are about 0. r The BRCA2 gene is located on a region of chromosome arm13q.e. Mutation of this gene has been associated with several cases of male breast cancer. Breast cancer may result from a mutation in one or more critical genes: r The BRCA1 gene is located on chromosome 17q21. ERBB2 [c-erb b2. r Nuclear pleomorphism – mild. often with a basal-like phenotype (Sorlie et al. lower inner quadrant (10%). PgR and HER-2/neu-negative. The gene is associated with a 6% lifetime risk of male breast cancer. However. 2003). the risk of contralateral cancer is estimated at 0. 2 or 3. BRCA1 mutations account for 2% of breast cancers and increase the chances of breast cancer by 35 to 85%. Conversely. although these terms are often interchanged. neu or HER-2]. and non-Hodgkin lymphoma.. r Grade II = total score = 6 to 7 points. r Grade III = total score = 8 to 9 points. An inherited mutation in TP53 (Li-Fraumeni syndrome) is associated with a 50% lifetime risk of developing breast cancer. TP53 [p53] and NME1 [NM23]). probably by increasing the serum concentration of oestradiol and decreasing the serum concentration of sex hormone binding globulin. 2006c). In socio-economic groups I and II. The lifetime risk of ovarian cancer is up to 60%. The anatomical distribution of breast tumours includes the upper outer quadrant (50%). The oestrogen receptor (ER) and progesterone receptor (PgR) status can be established via immunohistochemistry (IHC) and a semiquantitative measure is 192 . MYC) or inactivation of tumoursuppressor genes (RB1. the RR is 2 or more when breast cancer occurs in a first-degree relative at a younger age. which depends on the number of mitoses per ten high-powered fields. Cowden’s syndrome (PTEN) is an autosomal dominant condition. pancreas and bladder cancer. Other syndromes with increased risk of breast cancer include Muir-Torre syndrome and Peutz-Jeghers syndrome. There are referral criteria for referral from primary care to secondary care and from secondary care to tertiary care (NICE.5 to 1. Multifocal means two or more foci of disease exist in the same quadrant. r The TP53 gene is located on chromosome 17p13. tumours in the upper outer quadrant carry the best prognosis. 10 to 75% scores 2 and < 10% scores 3. and it increases the chance of breast cancer by 20 to 60%. causing a substantial increase in bio-available oestradiol. BRCA1 tumours tend to have aggressive pathologic features. lower outer quadrant (10%) and central (20%). upper inner quadrant (10%). It is also associated with elevated risk for the development of other cancers.0% per year. of which onethird are bilateral. Pathology Almost all breast carcinomas are adenocarcinomas arising from the epithelial cells that line the terminal duct lobular unit (TDLU). moderate or severe give a score of 1. BCL2. the risk is much higher if the first breast cancer occurred at a younger age and in a known gene carrier. With regard to inherited susceptibility. The grades are as follow: r Grade I = total score = 3 to 5 points. The types of genetic damage may be activation and overexpression of proto-oncogenes (EGFR [c-erb b1]. Diffuse tumours comprise 3%.Nayyer Iqbal and Peter Barrett-Lee in postmenopausal women by around 50%. obesity in premenopausal women reduces breast cancer risk. A high intake of saturated fat or smoking are not yet proven to be risk factors. BRCA2 tumours typically express ER and PgR and tend to be of higher grade with less tubule formation.

uk/. Options include monthly breast selfexamination.Breast provided by the ‘H-score. 2+ or 3+).nhs. internal mammary and supraclavicular nodes. by measuring the number of gene copies with fluorescent in situ hybridisation (FISH) or by the level of circulating receptor protein. For women at high risk of breast cancer. although they will not be routinely invited to attend. Mammography is less effective in women under the age of 35 because the breast is relatively radiodense. Screening and prevention Screening The aim of screening is to reduce mortality from breast cancer by detecting and treating it at an earlier stage. Involvement of axillary nodes occurs in up to 50% of symptomatic breast cancer and in 10 to 20% of screen-detected breast cancer.cancerscreening. 25 to 30 years). the H-score is zero). Tumours are ER positive in more than two thirds of postmenopausal patients but in fewer than half of those who are premenopausal. Tumours from any site in the breast can spread to any nodal site. HER-2/neu expression can be assessed by IHC with an antibody that recognises cell surface receptors (the result is scored –.1 or greater. In the UK the current target screening age is 50 to 70 years.5% will have cancer. 1+. More information about the UK breast cancer screening programme can be found at www. clinical assessment and ultrasound. HER-2/neu positivity is established by either an IHC score of 3+ or a FISH amplification of 2. Internal mammary nodes have evidence of tumour in 26% of patients with inner quadrant tumours and 15% with outer quadrant tumours. Lymph nodes To obtain an adequate assessment of nodal staging. ER-negative PgR-positive tumours account for roughly 5% of all tumours and are likely to be hormone responsive. Any other lymph node metastasis is coded as a distant metastasis.5 mGy is standard. histological sampling of the internal mammary chain reveals occult disease in 20 to 50% of cases. liver. histopathological assessment of the excised lymph node is required: only 70% of involved nodes are detectable clinically or radiologically. An IHC of 2+ should be confirmed by FISH. most have tumours involving the medial half of the breast. Women below the age of 50 are not invited for screening because the incidence of cancer is lower and the density of the breast tissue makes cancers more difficult to detect. 10% of women are expected to be recalled because of some mammographic abnormality. Spread Routes of cancer spread include direct local spread to the chest wall. annual clinical examination and annual mammographic screening with ultrasound. Women over 70 years are still entitled to breast screening. options include chemoprevention and prophylactic surgery. Haematogenous spread moves to bone.’ About 20% of patients are true ER negative by these criteria (i. More than 90% of women with metastases to the internal mammary nodes have axillary node involvement and. 1. skin and so forth. another oestrogen antagonist with fewer endometrial side effects.g. and aromatase inhibitors are under investigation. Taking two views rather than one increases the detection rate by 43%. after additional mammographic views. PgRs can be identified in some breast cancers. Internal mammary nodes 193 . Tamoxifen is known to reduce the incidence of breast cancer. their presence depends on an intact ER pathway. such as those with germ-line mutations in BRCA1 or 2. lymphatic spread and haemotogenous spread. infraclavicular. brain. In patients with axillary-node-positive breast cancer. Women at very high risk of breast cancer. There should be two views on the mammogram: mediolateral oblique and cranio-caudal. Lymphatic spread first involves the regional lymph nodes: the ipsilateral axillary (levels I. are more likely to be involved in patients who also have axillary nodal disease.5% will require diagnostic biopsy and about 0. Raloxifene. but its use is associated with potentially serious adverse effects such as endometrial cancer and thromboembolic disease. Prophylactic mastectomy dramatically reduces the risk of breast cancer. Prevention Currently there is no proven role for breast cancer prevention in the general population. may opt for breast surveillance from an early age (e. lung. During the initial prevalence screen. About 20% of ER-negative tumours are PgR-positive and possible reasons for their ER-negative status include a very low-level expression of ER or a false-negative result.e. in the remaining 5 to 10%. With modern screening a radiation dose of less than 1. II and III).

A persistent scaly or eczema-like lesion may be an indication of Paget’s disease. In patients with advanced (T3/T4) disease there is an incidence of metastatic disease of 15 to 20%. r B4 – suspicious. Investigations Triple assessment After a complete history and thorough clinical examination. lung. Regional disease Malignant axillary lymphadenopathy is usually caused by the presence of an ipsilateral breast cancer. Core biopsy has the advantage of providing a histological diagnosis and can differentiate between invasive and in situ carcinoma. a CXR is performed first. patients should be managed the same as they would be for the equivalent stage of primary breast cancer where the primary has been locally resected (see Chapter 38. unilateral. In the absence of evidence of another primary source of disease. breast distortion or pain alone should be taken seriously but are less likely to be associated with an underlying malignancy. liver function tests and serum calcium level should be taken. PR and HER-2 status can also be tested. The ER. Skin changes The skin may show thickening. Other tests A full blood count. r C2 – benign. r B3 – benign but may be associated with more malignant lesions nearby (i. redness. Patients with T1 and T2 primary breast tumours have an incidence of metastatic disease of only 2% and ‘routine’ staging for asymptomatic patients is not indicated (SIGN. Skin dimpling or a change in contour is present in up to 25% of patients with breast cancer. A painful lump or lumpiness. dimpling. 194 . are sometimes the presenting feature of breast cancer. solid. r B2 – benign. a ‘routine’ CT scan is not indicated (RCR. e. A provisional type and grade is usually given if possible. Cytology is reported as follows: r C1 – inadequate. 2006). Screening Breast cancer can present as a result of a primary screening programme in the general population. staging investigations are performed for these patients if it will affect management. either DCIS or invasive carcinoma. Metastatic disease Signs or symptoms from metastatic disease. no lesion to account for imaging findings is present (i. probably malignant. HER-2 testing should be performed on the surgical sample. irregular. followed by a CT scan if the CXR is normal and lymphangitis is suspected (RCR.e. Nipple changes Thickening and loss of elasticity. but core is crushed or there is too little tumour present (insufficient evidence for malignant diagnosis).e. r C4 – suspicious probably malignant. and/or inflammation. 2006). including chest Xray. r C3 – suspicious probably benign. hard. non-mobile and non-tender. probably missed). For patients with breathlessness. brain and skin. p.Nayyer Iqbal and Peter Barrett-Lee Clinical presentation Breast lump A malignant breast nodule is usually solitary. liver ultrasound and bone scan. 2005). liver. diagnostic excision is necessary. triple assessment is performed. Biopsy is reported as follows: r B1 – normal breast tissue. r B5 – malignant tumour present. which includes clinical examination (score 1 to 5).g. to be accurate. Spontaneous discharge in patients older than 50 years of age is likely to be caused by carcinoma. A diffuse dimpling or peau d’orange (like orange peel) is caused by infiltration of the tumour into the subcutaneous lymphatic channels. during follow-up from a previously treated breast cancer or during screening carried out because of a strong family history. r C5 – malignant epithelial cells. is suspicious if it is a new finding in a breast. 442). atypical ductal hyperplasia or papillary lesions). However. imaging by bilateral mammography (score R1 to R5) or ultrasonography (score U1 to U5) and cytology (fine needle aspiration cytology) or histology (Tru-cut® biopsy). most commonly affecting bone. Even for patients with advanced disease. causing flattening or inversion of the nipple.

3 shows the N and M staging for breast cancer. The overall survival is not affected. The prognosis is associated with size. low-grade DCIS of the solid. 2003). ulceration. unicentric.. If a mastectomy is not done. unless invasive cancer cannot be definitively excluded. However. Treatment options in DCIS Wide local excision alone may be an option in very small (< 5 mm). may occur in T1. 195 .5 to 1 cm > 1 to 2 cm > 2 to 5 cm > 5 cm Chest wall /skin Chest wall a a sive breast cancer in the ipsilateral breast within 10 years of diagnosis.to four-fold after breast-conservative surgery compared with BCT alone but has no effect on overall survival (Bijker et al. Because of this. Dimpling of skin. CA27–29 and CEA tumour marker levels can be useful to follow response to treatment in advanced disease. Radiotherapy reduces the relative risk of local failure by approximately one-half. LCIS = lobular carcinoma in situ.2 shows the T staging for breast cancer. T2 or T3 without affecting the classification. HER-2/neu amplification and TP53 mutation are often found. there is a proven risk reduction of recurrent disease (both malignant and premalignant) in patients treated with radiotherapy and/or tamoxifen (in ER-positive cases). Tis (LCIS). distance to the resection margin and age. The blood CA15–3. DCIS accounts for 20 to 25% of all cancer detected by screening. for example when microcalcifications are incompletely removed or if there is extensive microcalcification or multicentric disease and a mastectomy is planned. Between 30 and 50% of women with untreated DCIS will develop inva- Role of adjuvant radiotherapy Radiotherapy has been shown to reduce local recurrence by two. The TNM classification of breast cancer T Stage Tis T1 T1mic T1a T1b T1c T2 T3 T4 T4a T4b T4c T4d a Description Tis (DCIS). Table 16.. 2005. Bone marrow aspiration is performed if there is an unexplained cytopenia or a leucoerythroblastic blood smear. Table 16. DCIS requires local therapy. DCIS = ductal carcinoma in situ. there is not a consensus on the management of this condition. or other skin changes. 1991.4 shows the staging groupings for breast cancer.5. A scoring index for DCIS (Van Nuys Prognostic Index) is shown in Table 16. Adapted from UICC (2002). except those in T4b and T4d. grade. Tis (Paget) ≤ 2 cm ≤ 0. Skin oedema (including peau d’orange).Breast Table 16. 1998). Mastectomy with or without reconstruction is recommended in patients with widespread disease (disease in two or more quadrants) or in cases in which surgical margins free of carcinoma cannot be obtained. Staging The staging of breast cancer is determined by the American Joint Committee on Cancer (AJCC) and is a clinical and pathologic staging system based on the TNM classification (UICC.2. Treatment: non-invasive carcinoma of the breast (Tis) Ductal carcinoma in situ (DCIS) DCIS is a true premalignant condition. intercostal muscles and serratus anterior muscle but not pectoral muscle. especially in patients up to 50 years of age. Wide local excision with adjuvant whole-breast radiotherapy is the standard treatment for most women with DCIS. or papillary subtypes. and a suggested management strategy based on the scoring system is shown in Table 16. and Table 16. The most common mammographic finding in DCIS is branching microcalcification localised to a small region of breast. nipple retraction. Role of axillary nodal surgery Axillary dissection or sampling is not recommended in cases of pure DCIS. 2002). Fisher et al. satellite skin nodules Both 4a and 4b Inflammatory carcinoma Chest wall includes ribs.1 cm > 0.1 to 0. Sentinel node biopsy is generally not needed in cases of DCIS. cribriform. The use of a radiation boost applied to the tumour bed may be considered to maximise local control.6 (Silverstein.5 cm > 0.

Ongoing studies are evaluating the role of aromatase inhibitors (AIs) in ER-positive DCIS. LCIS is found predominantly in premenopausal women. Table 16. The invasive cancer is usually ductal and may be present in the same or the opposite breast. in the US Intergroup trial RTOG 9804.4. clinically apparent. LCIS cells are commonly ER-positive.2 mm ≤ 2 mm 1–3 axillary nodes Internal mammary nodes with microscopic metastasis by sentinel node biopsy but not clinically apparent 1–3 axillary nodes and internal mammary nodes with microscopic metastasis by sentinel node biopsy but not clinically apparent N2a N2b Fixed axillary Internal mammary clinically apparent N3a N3b Infraclavicular Internal mammary and axillary N3c M stage MX M0 M1 Supraclavicular Description Distant metastasis cannot be assessed No distant metastasis Distant metastasis pN3c pN3a pN3b ≥ 10 axillary nodes or infraclavicular node(s) Internal mammary nodes. 196 . Description N0 N0 N1 N1 N0 N1 N0 N2 N2 N2 N1. with axillary node(s) or > 3 axillary nodes and internal mammary nodes with microscopic metastasis by sentinel node biopsy but not clinically apparent Supraclavicular pN2a pN2b 4–9 axillary nodes Internal mammary nodes. The role of tamoxifen for DCIS in the absence of RT is also being evaluated. without axillary nodes Adapted from UICC (2002). Stage groupings for breast cancer Stage Stage 0 Stage I Stage IIA Tis T1a T0 T1a T2 Stage IIB T2 T3 Stage IIIA T0 T1a T2 T3 Stage IIIB Stage IIIC Stage IV a Role of adjuvant tamoxifen The benefit of tamoxifen in patients with clear resection margins after breast conservation therapy (BCT) and radiotherapy appears small and it is not recommended routinely (Fisher et al. Adapted from UICC (2002). it is almost always multicentric in the breast and bilateral in about one-third of cases. 1999).1 cm microinvasion. clinically apparent. whereas the overexpression of T4 Any T Any T T1 includes T1mic < 0. N2 N0. N1.3.Nayyer Iqbal and Peter Barrett-Lee Table 16. It is associated with approximately a 30% lifetime risk of developing an invasive carcinoma.. N2 N3 Any N M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M0 M1 Lobular carcinoma in situ (LCIS) LCIS is not thought to be a true premalignant condition but rather a marker of increased risk of breast cancer. N and M staging of breast cancer N stage N1 Description Movable axillary pN stage PN1mi pN1a pN1b pN1c Description Micrometastasis > 0.

3% versus 39. or lumpectomy plus axillary dissection and RT. There is no uniform consensus if the margin is 1 to 10 mm. p < 0. WLE = wide local excision. Adapted from Silverstein (2003). 1995). In cases of a positive margin. but if this would lead to an unacceptable cosmetic outcome. LCIS is not clinically palpable or detectable by mammography but is identified incidentally in about 1% of benign breast biopsies. observation alone is the preferred option because the risk of developing invasive cancer is low (21% over 15 years). it is inadequate if it is < 1 mm. 2002a. 1995. necrosis present Grade 3 ≤ 1.001). nipple apy (BCT) consists of wide local excision (WLE) and postoperative RT. the patient should have a further excision to achieve a clear margin. or lumpectomy plus axillary dissection alone.. Bilateral mastectomy with or without reconstruction is preferred in special circumstances (such as a strong family history of invasive breast cancer or nongenetic predisposition) because the risk of invasive breast cancer after a diagnosis of LCIS is equal in both breasts.5. The NSABP B-06 trial (Fisher et al.. For treatment of LCIS. Current practice aims to achieve microscopically disease-free margins of at least 1 mm and it is unacceptable to have tumour cells at the surgical margin without further excision. no necrosis Grades 1–2.2%. Management of early breast cancer Surgery in breast cancer Primary disease Modified radical mastectomy Modified radical mastectomy involves removal of the entire breast. Systematic reviews have shown similar outcomes for BCT and mastectomy (EBCTCG. ERBB2 and TP53 are uncommon. a mastectomy should be done. 2002). It may be a reasonable option to treat selected cases with BCT using a microscopically focally positive margin in the absence of extensive intraductal tumour by RT together with a boost. 2002a) randomised 1851 women with invasive breast cancer and negative margins to modified radical mastectomy. Breast-conserving surgery Breast conservation ther> 60 years 40–60 years < 40 years The margin in DCIS is considered negative if it is > 10 mm. Fisher et al. The absolute contraindications for BCT and RT include: 197 .6–4 cm > 4 cm RT = radiotherapy. the deep margin or skin margin cannot be easily increased). There was no significant difference in the OS or DFS at 20 years but there was a significant difference in local recurrence (LR) at 20 years between lumpectomy plus RT and lumpectomy alone (14.6. The histology of the invasive cancer tends to be favourable. and areola and it is usually combined with an axillary node dissection (see discussion that follows). Management of DCIS Van Nuys Prognostic Index Ten-year local recurrence rate 3% 27% 66% Management WLE WLE + RT Mastectomy Score 4–6 7–9 10–12 Definition ≥ 10 mm 1–9 mm < 1 mm Grades 1–2. The acceptable size of the margin is influenced by which margin is the closest and the need to minimise the degree of mutilation (e..5 cm 1. These results were confirmed in the meta-analysis performed by the EBCTCG (EBCTCG.g. A scoring system for DCIS Score components Marginsa 1 2 3 Histological subtype 1 2 3 Size 1 2 3 Age 1 2 3 a Table 16. or the patient chooses.. 1997.Breast Table 16. Adapted from Silverstein (2003). Veronesi et al. The most important surgical factor influencing LR is the completeness of excision. Morris et al.

Axillary surgery Until recently. involving removal of approximately four level 1 nodes. Axillary node clearance involves clearing the axillary contents from the volume bounded by the axillary skin laterally. r Focally positive margin. Different choices exist for breast reconstruction procedures: r Myocutaneous flap reconstruction. A tissue expander is gradually inflated. unpublished data. r Tumour larger than 5 cm. using the latissimus dorsi (LD) muscle or the transverse rectus abdominus myocutaneous flap (TRAM). posterior to pectoralis minor. Veronesi et al. Increased awareness and screening have resulted in more women being diagnosed with early stage breast cancer and 60 to 80% of patients are node-negative. There is no absolute contraindication for immediate reconstruction. When it is an adequate size it is removed and replaced by a permanent prosthesis which often contains silicone. inferolateral to pectoralis minor. teres major and subscapularis posteriorly. into 198 the dermis (subdermal) or under the nipple (subareolar). Breast reconstruction Breast reconstruction should be offered to women with breast cancer who are undergoing mastectomy. . smoking and metastatic disease. The reconstruction is done as a second procedure after the histology of the WLE with a complete excision. decreased range of arm movement. Lymph node dissection is considered overtreatment and is associated with significant morbidity. and reduces psychosocial morbidity compared with delayed reconstruction. such as axillary pain and numbness. Mansel et al. In reality. Although this causes less morbidity. at 6 months. 1999. and level 3. these nodes are in continuity with each other but the concept of axillary node levels is useful when discussing the extent of axillary node surgery. which may or may not require an implant. The combination of blue dye and radioisotope gives better results than either agent alone.Nayyer Iqbal and Peter Barrett-Lee r Previous RT to the breast or chest wall. latissimus dorsi. Localisation rates of over 95% and false-negative rates of less than 5% are possible after an initial learning curve. diabetes. and chronic lymphoedema. Delayed reconstruction should be considered when there is a concern about tumour clearance or when postoperative RT is planned.. r The node should be clinically negative at the time of diagnosis. steroid treatment. produces better results. but only peritumoural injections map accurately to the internal mammary lymph nodes. level 2. r The LD muscle and fat without overlying skin can also be used to fill defects in the breast following WLE for large tumours.. r An extended LD reconstruction combined with a skinsparing mastectomy. Relative contraindications include: r Active connective tissue disease involving the skin (especially scleroderma and lupus). calcifications. r RT during pregnancy. Immediate reconstruction has economic benefits. overall arm morbidity is less and quality of life is better. Localisation of SLN is possible by injecting blue dye and a radioactive colloid tracer around the tumour (peritumoural). sentinel lymph node biopsy (SLNB) has emerged as a minimally invasive procedure that accurately assesses axillary node status (Cody. r Insertion of a breast prosthesis. The SLN (or nodes) is the first node in the regional lymphatic basin to which a tumour drains. However. pectoralis muscles anteriorly. Some centres have chosen to use axillary nodal sampling. Therefore. it is important that SLNB should meet all the following criteria: r It should be performed by an experienced surgical team. r Positive pathologic margin. Relative contraindications include ischaemic heart disease. obesity. The first nodal station or SLN are detectable either as visible blue nodes or as radioactive nodes detected by a hand probe. r Appearance of diffuse suspicious or malignant micror Widespread disease that cannot be removed by WLE through a single incision to achieve clear margins and a satisfactory cosmetic result. the lower border of the axillary vein superiorly. superomedial to petoralis minor. 2003). The levels of axillary nodes are defined as: level 1. 1991). further surgery or postoperative radiotherapy would be needed if involved nodes are found. and the chest wall medially. axillary dissection (levels 1 to 3) was the preferred technique for all women with invasive breast cancer and at least 10 lymph nodes were needed for pathologic evaluation to accurately stage the axilla (NIH Consensus Conference. r The patient should have no previous chemotherapy or hormonal therapy. Compared to conventional axillary dissection.

Morris et al. Indications for postmastectomy RT are if the tumour is stage T3 or T4. For conventional planning with orthogonal films. Conventional orthogonal X-ray films or CT scanning is used. the use of systemic therapy may add to the local control given by RT. 1995.. NSABP B-21 and the Scottish trial. 1997.. Isodose % 100 95 70 10 Figure 16. and a target volume is marked with approximate field borders: r Medial – midline. but it may not fit in a narrowbore CT scanner). A plan is provided by the department of physics.1. r 40 Gy in 15 fractions over 3 weeks. 2004). Radiation given concurrently with anthracycline-based regimens or taxanes is not recommended. NIH Consensus Conference. r Inferior – 1 cm below the breast tissue. which is the subject of an ongoing study (Hughes et al. small ‘filler fields’ have been added to improve the dose distribution. it is not considered necessary in patients who have a low risk of recurrence. r Deep – incorporating a maximum of 2 to 3 cm of lung.5 Gy in 16 fractions over 3. r 42. and the PTV . glandular breast tissue cannot be visualised. This has not been seen in recent RT trials using modern treatment techniques. Forrest et al. 2005).1 shows an RT plan for a postoperative treatment for carcinoma of the breast. a prone position is useful. Veronesi et al. 2004. trials have not substantiated this concern (Ahn et al. such as those with one to three axillary nodes involved. Fisher et al. 1991. 2004. Ragaz et al. usually allowing 5 mm skin sparing. 2002b. there is no GTV. Partial-breast radiation is still investigational. Fyles et al. However. 2000). The magnitude of the benefit of RT is greater in node-positive patients. For postoperative RT to the breast. 3D isocentric radiotherapy plan for postoperative carcinoma of the breast. is the CTV with a 1 cm margin. 2005. Figure 16. It remains uncertain whether RT may be safely omitted for women over 70 years of age with clear margins and who also receive adjuvant endocrine treatment.. Doses are prescribed to the ICRU reference point. Wolmark et al. Ongoing clinical studies will determine the role of this treatment in patients with a lower risk of local recurrence. 2003)... Harris et al. It has been shown in a randomised trial that chemotherapy and tamoxifen are not appropriate substitutes for RT (NSABP B-06. 2005). In addition to the two main tangential fields. 2002). Postmastectomy radiotherapy Although several individual trials have shown a survival advantage for using RT after mastectomy (Overgaard et al. the CTV is the glandular breast tissue. multifocal disease and tumours with 199 Radiotherapy technique: postoperative radiotherapy to breast The patient is positioned supine on an angled ‘breast board’ or flat. The EBCTCG (2005) has shown that adjuvant RT gives significantly better control of local recurrence with little effect on breast cancer mortality during the first 5 years. Determining the optimal dose is a matter of ongoing clinical trials... if four or more lymph nodes are involved. For patients with very large breasts. and if there is a positive excision margin (< 1 mm)..Breast Role of adjuvant radiotherapy Post-BCT radiotherapy Adjuvant whole-breast RT reduces the risk of local recurrence by two-thirds (EBCTCG.. 1996. A retrospective review has shown increased risk of LR with a delay in RT (Huang et al.. r Superior – suprasternal notch. During the subsequent decade there is a lower breast cancer mortality rate but in previous analyses this has been balanced by an excess of deaths due to cardiac disease.5 weeks.. One or both arms are abducted. Doses in current use include: r 50 Gy in 25 fractions over 5 weeks. Although some centres avoid the use of concurrent tamoxifen with RT because of theoretical concerns that residual tumour cells may be put into growth arrest and therefore be resistant to radiation. Usually two tangential parallel-opposed fields are used. depending on the method of localisation (a breast board has the advantage that the sternum is more horizontal.. RT is usually given after the patient has finished adjuvant chemotherapy. r Lateral – mid-axillary line.

4% after axillary clearance alone. is angled to be in line with the rib cage. r Lateral – for supraclavicular fossa only. the junction between the medial two-thirds and lateral third of the clavicle.3% (95% CI 3. The posterior field for the axilla. Because of the morbidity of axillary irradiation and because recurrence in the axilla is rare. and the RT fields that include them irradiate more normal tissue. most of the dose is given through an anterior field that also treats the supraclavicular nodes. RT is only given to the part of the breast closest to the site of the excised tumour. p < 0. The main difference is the clinical target volume. 1986). However. One anterior field is used. the usual dose is 40 Gy. Doses used are 50 Gy in 25 fractions over 5 weeks and 40 Gy in 15 fractions over 3 weeks.. because of the lack of data from phase III trials on long-term safety..3% (95% CI 6.3% after axillary clearance and RT compared with 7. efficacy and issues about treatment of regional nodes. however. higher doses are only used in patients who have extensive axillary soft tissue disease or who have not had axillary dissection. and ongoing trials will determine whether this justifies the extra time and expense involved in this planning process. 1980. MammoSite® . The technique for chest wall RT following mastectomy is similar to that for breast RT. and combined treatment to the axilla is associated with a much higher risk of lymphoedema (38. Le et al. The 5-year actuarial rates of LR were 7. The technique for RT to the axilla and supraclavicular fossa depends on whether the breast or chest wall is also treated..001. One study. A posterior photon field is then used to 200 supplement the mid axilla to the desired target dose. Other radiotherapy issues Three-dimensional planning using IMRT usually produces a more homogenous dose distribution across the breast.Nayyer Iqbal and Peter Barrett-Lee extensive lymphovascular infiltration. 2001). particularly in those in whom there is concern regarding the adequacy of surgical clearance. Partial-breast irradiation can be done by CT planning. possibly increasing the risk of cardiac complications (Arriagada et al.001. Axillary RT is not necessary after a complete axillary clearance. r Inferior – cover the lower border of clavicle.8–4. With conventional localisation.. A boost of 10 to 16 Gy is therefore considered in younger patients. 1990)..8–7. Delivery of 50 Gy from the anterior field results in a dose of 35 to 37 Gy to the midline axillary structures.7%) between two groups (p < 0. It is difficult to define what this should be but it is usually the chest wall excluding ribs and muscle. giving the same applied dose as that used for the breast/chest wall therapy. brachytherapy or intraoperative RT. When the axilla is irradiated. The IMNs are difficult to treat because their exact position is often uncertain. especially in patients whose risk of recurrence is low. Less irradiation of normal tissue may lead to less morbidity. Fisher et al.6%) and 4. In practice. RT to the internal mammary nodes (IMNs) is controversial because recurrence at this site is very rare and most patients at risk receive adjuvant systemic therapy. cover the humeral head. Matching diverging field borders can be difficult and requires a gap at the skin surface. and there is a need to develop ways of identifying their position. for SCF and the axilla. especially in patients younger than age 50 who received a boost of 16 Gy to the tumour bed. Bartelink et al. if used. the approximate field borders for the anterior field are as follow: r Medial – 1 cm lateral to the midline. Role of a radiation boost to the breast The role of a radiation boost given after whole-breast RT with external beam RT or a brachytherapy boost is uncertain and is not routinely recommended. The lymph node failure rates are very low. The position of the lymph nodes varies among patients. Radiotherapy to the regional lymphatics RT to the ipsilateral supraclavicular fossa is associated with a reduced risk of disease failure at this site and is used in patients with the highest risk of disease such as those with more than four positive axillary lymph nodes. the field borders and doses are similar to those defined for breast RT. the C7/T1 junction. RT should only be considered if axillary clearance is not performed (patient refuses or is medically unfit) or for patients with extensive extranodal tumour spread. 1988. The use of half-beam blocking can help to reduce dose inhomogeneities. A high-grade tumour and young patient age are not sole indications for treatment. APBI should . It has been associated with a slight improvement in local control but needs to be balanced against a higher morbidity. In accelerated partial-breast irradiation (APBI). showed a significantly reduced risk of LR. Kissin et al. r Superior – 3 cm above the clavicle.

2004).. and the HER-2/neu gene is overexpressed or amplified. r Grade 2 to 3. r HER-2/neu gene neither overexpressed nor amplified. national and local guidelines should be followed. the last update was in January 2005. r Patient preference. or node-positive with four or more involved nodes. r The menopausal status. international. Wallner et al. Adapted from Goldhirsch et al. Endocrine responsiveness In endocrine-responsive breast cancer. intermediate or high. r The potential benefit from the different types of adjuvant treatment – chemotherapy. the aforementioned risk categories and the type of 201 . r Age < 35. comorbidities and performance status. Role of adjuvant chemotherapy The benefit of adjuvant chemotherapy in terms of OS and DFS depends on the endocrine responsiveness. St. 2. r Absence of peritumoral vascular invasion. (2005). with one to three involved nodes. Adjuvant treatment: general For the adjuvant treatment in early breast cancer. r The cancer is node-positive (one to three involved nodes) and HER-2/neu gene is neither overexpressed nor amplified. r Grade 1. r The risk category – low. ET = endocrine therapy. r HER-2/neu gene overexpressed or amplified. The following occur when endocrine response is uncertain: r Low levels of steroid hormone receptor (<10% of cells positive). Endocrine-non-responsive breast cancer cells have no detectable expression of steroid hormone receptors. r A high number of involved lymph nodes. The patient is considered to be at high risk when the cancer is node-positive. Risk categories The patient is at low risk if the cancer is node-negative and all of the following features apply: r pT ≤ 2 cm. Gallen made a fundamental change in the algorithm for the selection of adjuvant systemic therapy for early breast cancer (Goldhirsch et al. 3. the 9th St. r High tumour levels of urokinase-type plasminogen r Increased proliferation marker. PgR). only be used as part of a clinical trial (Orecchia.7. The National Institutes of Health (NIH) guidelines were started in 1980 and are updated every 5 years. 2005). 2005. the cells express steroid hormone receptors (ER.7. age.Breast Table 16. Choice of adjuvant treatment in breast cancer Treatment if endocrine Risk category Low risk Intermediate risk High risk Treatment if endocrine responsive ET or Nil ET alone or CT → ET (CT + ET) CT → ET (CT + ET) response uncertain ET or Nil CT → ET (CT + ET) CT → ET (CT + ET) Treatment if endocrine non-responsive Not applicable CT CT CT = chemotherapy. activator inhibitor type 1. r Age ≥ 35. endocrine non-responsive or the endocrine response is uncertain. trastuzumab and radiotherapy. r Presence of peritumoral vascular invasion. r HER-2 overexpression. The choice of systemic treatment (2005) is shown in Table 16. endocrine therapy. There are three international guidelines: 1. r Lack of PgR. Gallen guidelines were started in 1978 and have been updated every 2 years since 2001. The National Comprehensive Cancer Network (NCCN) guidelines were started in 1996 and are updated every year. The patient is at intermediate risk if the cancer is node-negative and at least one of the following features applies: r pT > 2 cm. The important points to consider are: r Whether the tumour is endocrine-responsive. In January 2005..

The 5-year Anthracycline. 1990. epirubicin 60 mg/m2 i. Mamounas et al. The Oxford overview compared a number of anthracyclinebased regimens with CMF and showed an 11% further reduction in the odds of recurrence (p = 0. The Oncotype DX Recurrence ScoreTM (RS) has been developed in an attempt to further inform the decision regarding the benefit of chemotherapy in node-negative and ER-positive patients for whom genomic information is available. EBCTCG. Non-anthracycline non-taxane-containing regimens Chemotherapy with cyclophosphamide. days 1 and 8. Results from a number of studies have found a modest improvement in DFS but not a consistently meaningful improvement in OS with anthracyline. 2003. methotrexate. taking comorbidities into account. doxorubicin 50 mg/m2 and cyclophosphamide 500 mg/m2 . methotrexate and 5-FU (CMF) repeated every 28 days for six cycles (cyclophosphamide 100 mg/m2 orally days 1 to 14.v. r FEC (5-FU 600 mg/m2 i. Also. A number of individual studies showed a small survival advantage for anthracycline regimens over that of CMF (Fisher et al.Nayyer Iqbal and Peter Barrett-Lee chemotherapy planned. 5-FU). Martin et al. r AC (doxorubicin.02. day 1) repeated every 21 days for six cycles (Budman et al.v.. Clearly these will be wider when fewer trial data are available. r FEC→docetaxel (5-FU. the lower the likely benefit from adding chemotherapy. The RS is based on the expression profile from a select panel of 21 genes in the patient’s tumour tissue based on retrospective analysis of two trials (NSABP B-14 and B-20). 2001. 2003. day 1. cyclophosphamide 500 mg/m2 i. Some examples of anthracycline-containing combinations are as follow: r Canadian CEF (5-FU 500 mg/m2 i. epirubicin.and taxanecontaining regimens (Buzdar et al. It is not clear which. cyclophosphamide). day 1. epirubicin 100 mg/ m2 i.. r E→CMF (epirubicin then cyclophosphamide. 2004). 1998). Roche et al. r TAC (docetaxel.v. 2002. cyclophosphamide. It is important to understand that the predicted benefits of therapy are estimates and the confidence intervals around such estimates are not usually given. There is no consensus about how individual chemotherapy drugs should be sequenced (SIGN. cyclophosphamide). In its most recent publication. cyclophosphamide 75 mg/m2 orally days 1 to 14) repeated every 28 days for six cycles (Levine et al. doxorubicin 50 mg/m2 and cyclophosphamide 500 mg/m2 ) reduced the relative risk of recurrence by 28% and death by 30% compared to FAC (5FU 500 mg/m2 . The BCIRG-001 trial showed that TAC (docetaxel 75 mg/m2 .v. then paclitaxel). r FEC (5-FU.. 2005). Anthracycline. r A→CMF (doxorubicin then cyclophosphamide. epirubicin. 2001). the higher the degree of endocrine responsiveness... if any. further prospective validation is needed (Paik et al. 1998). cyclophosphamide).. methotrexate. Adjuvant treatment in women over age 70 should be individualised.v. r French FEC (5-FU 500 mg/m2 i. cyclophosphamide). days 1 and 8. day 1. 202 . methotrexate 40 mg/ m2 i.. 2005).v. days 1 and 8. non-taxane-containing regimens Examples of non-taxane-containing regimens of anthracycline include: r FAC (5-FU. These regimens are preferred over CMF for use in node-positive disease and HER-2-positive cases. day 1) repeated every 21 days for six cycles (French Adjuvant Study Group. 1995. 2003). then docetaxel). 2005).v.com) and must be discussed with patients thoroughly at the time of diagnosis and staging. epirubicin 60 mg/m2 i.. 2005).and taxane-containing regimens Examples of regimens that use anthracycline and taxane include: r AC→paclitaxel (doxorubicin.. doxorubicin.. 5-FU). Henderson et al. of these combinations is the most effective. cyclophosphamide).006) and a 16% further reduction in the odds of death (p = 0. cyclophosphamide. doxorubicin. In general.adjuvantonline. 2005). The relative and absolute benefits can be estimated using systems such as Adjuvant! Online (www. Poole et al. day 1. However. the higher the likely benefit from adding chemotherapy to other adjuvant treatments.v. the younger the patient. the EBCTCG has shown that women receiving chemotherapy had a relative reduction in the risk of breast cancer recurrence by 23% and death by 17% in any year compared with the control group (EBCTCG. 2004)..v. and 5-FU 600 mg/m2 days 1 and 8) reduces the risk of recurrence and death from breast cancer (Bonadonna et al. cyclophosphamide 600 mg/m2 i. r EC (epirubicin.

or for patients of any age who are at high risk. ovarian function suppression (OFS) with goserelin is an acceptable alternative. The EBCTCG has shown that tamoxifen. and PERCHE are addressing the issues of OFS. 2005).01). NICE has approved the use of TAC as an adjuvant therapy for women with nodepositive breast cancer (NICE. Joint. They should receive tamoxifen as a standard adjuvant treatment for 5 years. 2003). and the recently closed trials aTTOM and ATLAS have addressed the question of continuing tamoxifen beyond 5 years. OFS with tamoxifen appears to be as effective as CMF chemotherapy. ongoing trials such as SOFT. but in patients receiving chemotherapy and tamoxifen it is not clear if OFS has an additional benefit. letrozole) alone for 5 years is based on the results from RCTs. 2006a). AIs and chemotherapy in endocrine-responsive breast cancer in young women (Dellapasqua et al. The results of this study are awaited. 2004a). if chemotherapy did not induce OFS. randomised more than 4000 patients between the control arm (FEC or E-CMF) and the experimental arm (four cycles of FEC followed by four cycles of docetaxel 100 mg/m2 ). It is not yet clear whether this regimen is more effective than standard regimens (Citron et al. Again the results are awaited. Tamoxifen should be avoided during pregnancy owing to its teratogenicity. which converts circulating androgens to oestrogen.. ‘Dose dense’ describes an increased dose intensity achieved by dosing at more frequent intervals than normal. see Chapter 3. At 68-month follow-up. Postmenopausal patients Oestrogen receptors have two activation domains: AF1 and AF-2. there are now several options for postmenopausal women who require adjuvant endocrine therapy: r In women with low-risk disease or with comorbidities such as osteoporosis or heart disease. OFS may be discussed individually. But no benefit in OS has yet been seen. muscle and bone pain. Premenopausal patients Premenopausal patients whose tumours are not shown to have absent ERs or PgRs should be considered for adjuvant endocrine therapy (SIGN.. 26). randomised patients to paclitaxel with or without gemcitabine (GT) after four cycles of epirubicin with cyclophosphamide (all every 3 weeks). 2005a) showed that 5 years of anastrozole significantly increases DFS compared to tamoxifen and the benefit appears to increase with longer follow-up. Pritchard. TACT. r Upfront use of an AI (anastrozole. AIs are active in endocrineresponsive tumours and may be beneficial in cases of tamoxifen resistance. The benefit was largest in those with PgRnegative tumours.. when given to women with ERpositive breast cancer over 5 years. Because of the results of recently reported RCTs. TEXT. p = 0. were more frequent with anastrozole. paclitaxel. The BIG 1– 98 Trial with 8010 patients (Thurlimann et al. 2005. If tamoxifen is contraindicated. Five years of tamoxifen is more effective than two. On the basis of this trial. there were fewer recurrences with anastrozole (HR 0. AIs exert their antioestrogenic effects through inhibition of the aromatase enzyme. Dose-dense regimens Examples of dose-dense regimens include dose-dense AC→paclitaxel (doxorubicin. TANGO. tamoxifen should be given for 5 years. and using growth factor support to achieve fortnightly cycles. The ATAC trial with 9366 patients (Howell et al. Anastrozole and letrozole are nonsteroidal competitive inhibitors of aromatase whereas exemestane is a steroid-irreversible AI and inhibits aromatisation in vivo by 98%. cyclophosphamide. reduces the annual recurrence rate by 41% and annual mortality rate by 34% (EBCTCG. . However.. Tamoxifen is useful in the adjuvant treatment of breast cancer in both pre. Tamoxifen blocks AF-2 but not AF-1 so there remains oestrogenic activity in some tissues. Tamoxifen should be given sequentially after adjuvant chemotherapy (Albain et al. 2005) 203 Role of adjuvant endocrine treatment Tamoxifen is a selective oestrogen receptor modulator and blocks the action of oestrogen by binding to one of the activating regions of the ER. whereas gynaecological and vascular events occurred more frequently with tamoxifen. then paclitaxel) and dose-dense A-T-C (doxorubicin.87. 2005). especially for very young patients. 2005. especially bone fractures. It should be noted that this arm of the study was associated with a high incidence of neutropenic sepsis despite routine use of G-CSF.and postmenopausal women. p. Another UK study. A large UK multicentre study.Breast DFS was 75 versus 68% and that for OS was 87 versus 81%. cyclophosphamide)..

4 versus 3.003) compared to tamoxifen.6%) and diarrhoea (4. There was no increased benefit in ER-positive and PgR-negative patients.0001). Piccart-Gebhart. It should be clear at the outset whether breast conservation is intended.. This recommendation is supported by the results of a recent meta-analysis that shows that postmenopausal patients who switch to an AI after 2 or 3 years of tamoxifen had improved disease-free survival (HR 0. 2006d). 2005. The IES trial (4742 patients) tested a switch to exemestane after 2 to 3 years of tamoxifen (Coombes et al. exemestane) are recommended as options within their licensed indications. The choice of treatment should be made after a discussion between the responsible clinician and the patient about the risks and benefits of each option (NICE. The risk is related to both age and baseline left ventricular ejection fraction. The optimum duration of treatment with trastuzumab may be known when the results of the 2-year arm of the HERA trial are available. RT can be given before or concurrent with trastuzumab. Letrozole improved DFS over placebo (HR 0. r Another option is to switch the patient to an AI (letrozole) after 5 years of tamoxifen (extended adjuvant treatment).59 [95% CI 0. M0). in postmenopausal women with breast cancer. based on two RCTs. r The patient could switch to an AI (exemestane. p < 0. The MA-17 trial (Goss et al.76. because conservation is not appropriate for patients with central tumours when the cosmetic outcome will be unacceptable.74]. 2006). which may allow for breast-conserving surgery rather than mastectomy (Hanrahan et al. p = 0. The benefit was seen in all analysed subgroups.. p = 0.0009). 2006b). r Stage IIIC (any T.0001) and overall survival (HR 0. Consideration should be given to switching to an AI after 2 to 3 years or after 5 years of tamoxifen (SIGN. 2006. 2005) – the joint analysis of the Austrian and German trials of 3224 patients – showed that anastrozole treatment after 2 years of tamoxifen gave a significantly better relapse-free survival compared with continuing the patient on tamoxifen (HR 0... Role of adjuvant trastuzumab Trastuzumab is a chimeric antibody (95% human and 5% murine) developed against the HER-2/neu transmembrane epidermal growth factor receptor.71 [95% CI 0.3 versus 2. tamoxifen remains the initial choice of adjuvant treatment unless there are relative contraindications to its use. However. Again there was no difference in OS. 2005) compared letrozole with placebo after completion of about 5 years of tamoxifen in 5187 women. N3). The ABCSG trial 8 and ARNO 95 trial (Jakesz et al. 2005) and is possible in up to 80% of suitable patients. r Stage IIIB (T4. p = 0. Perez et al. N0–2. with multifocal tumours. Slamon et al. The results of five reported RCTs that used trastuzumab as an adjuvant therapy for early breast cancer have shown a dramatic improvement in recurrence-free survival and a 50% overall reduction in the risk of recurrence when trastuzumab was given in combination with or following chemotherapy as shown in Table 16. 2005). Overall. 2004). cardiovascular events were more common in patients receiving letrozole. avoidable and reversible. Romond et al.8 (Joensuu et al. Management Women who are fit enough receive primary chemotherapy.001). or with inflammatory breast cancer.98]. M0). Trastuzumab has now been recommended by NICE for treatment of patients with HER-2-positive tumours in the adjuvant setting (NICE. but it is not known whether sequential or anthracycline-free regimens present a lower risk of cardiac events. The AIs (anastrozole. 2005... At 3 years there was a highly significant benefit in disease-free survival in the exemestane arm (HR 0. exemestane was associated with higher incidence of arthralgia (5.. Management of locally advanced breast cancer Locally advanced breast cancer is defined as inoperable non-metastatic disease and occurs in the following stages: r Stage IIIA (T0–3.58.Nayyer Iqbal and Peter Barrett-Lee showed that letrozole significantly increases DFS (HR 0. letrozole. In addition to the aforementioned side effects. p = 0.52–0. anastrozole) after 2 to 5 years of tamoxifen to complete 5 years of therapy. 2005. 2005).. The rate of radiological complete response (CR) is best assessed using 204 . p = 0. predictable. The risk of cardiac dysfunction associated with adjuvant trastuzumab is acceptable.3%). N2..04.81. NICE has recently published a guidance for hormonal therapies for the adjuvant treatment of early ER-positive breast cancer. Jonat et al.60.48–0.

RT may be used instead of surgery but it is associated with a higher LR rate.64 1 0. V = vinorelbine.1 0.8 4.5 0 0.6 n/a 0. Adapted from Joensuu et al. In the case of radiological CR. or definitive RT to the breast. Summary of the trastuzumab trials Trial (no.49 0. mastectomy plus ALND with or without RT. the complete excision rate is better.3 1. bone/soft tissue disease only. Patients with HER-2positive tumours should be considered for trastuzumab. w = weekly. The aim of treatment should be to improve survival and to improve quality of life with minimal toxicity. PgR and HER-2 assays if biopsy material is available from a metastasis. Therefore.6 0. Chemo = standard chemotherapy. because the receptor status of the metastasis can be different from that of the primary tumour. Perez et al. of patients) NSABP B-31 (2030) NCCTG N9831 (3505) 2 Median FU (years) 2 Arms AC (×4) → P (×4) AC (×4) → P (×4) + Hw (×12) → Hw (×40) AC → Pw (×12) AC → Pw (×12) → Hw (×52) AC → Pw (×12) + Hw (×12) → Hw (40) HERA (5090) 1 Chemo Chemo → H3w × 1 year Chemo → H3w × 2 year BCIRG 006 (3222) 2 AC (×4) → D (×4) AC (×4) → D (×4) + Hw (×12) → H3w (×13) DCarbo (×6) + Hw (×19) → H3w (×11) Fin Her (1010) 3 D (×3) or Vw (×8) → CEF (×3) D (×3) + Hw (×9) or V (×8) + Hw (×9) → CEF (×3) HR 1 0. hormone or delayed breast reconstruction. with or without hormone. Neoadjuvant endocrine therapy has been evaluated recently in less fit/older patients with endocrinesensitive disease who are unlikely to tolerate chemotherapy. or asymptomatic and minimal visceral disease. 205 . (2005). F = 5-FU. Letrozole is licensed for neoadjuvant therapy (followed by surgery and/or RT) in postmenopausal women with large operable or locally advanced ERpositive breast cancer. Piccart-Gebhart (2005). P = paclitaxel. E = epirubicin. CHF = congestive heart failure. Carbo = carboplatin. and the rates of local recurrence are low.87 0. C = cyclophosphamide.48 1 0.Breast Table 16.8. H = trastuzumab. Letrozole or anastrozole treatment results in more patients able to undergo BCT. (2005) and Slamon et al.42 CHF (%) 0. Letrozole gives a higher response rate than tamoxifen in patients with HER-2-positive tumours (Bhatnagar. Management of metastatic breast cancer Consideration should be given to repeating ER. axilla and SCF.54 n/a 1 0. If there is no response to neoadjuvant chemotherapy.4 1 0 Cardiac death (n) 1 0 1 1 0 1 0 n/a 0 0 0 0 0 A = doxorubicin.5 3. 2006). then alternative additional chemotherapy and/or preoperative RT should be considered. Primary hormone therapy Hormone therapy alone is often given if the patient has ER/PgR-positive disease. ultrasound ( ± MRI) and is higher than the rate of pathological CR. treatment is neoadjuvant chemotherapy (usually anthracycline-based ± taxane) followed by: BCT plus ALND plus RT with or without hormone therapy (which is not suitable in inflammatory breast cancer). (2006).3 2. D = docetaxel. Adjuvant hormone treatment in postmenopausal women should be with an AI. Romond et al. (2005).61 1 0. HR = hazard ratio for death.

etoposide. should be considered if the tumour is HER-2-positive and if there is symptomatic visceral or hormone-refractory disease (Albain et al. Other active agents are cisplatin. Cobleigh et al. Taxanes can be given weekly or every 3 weeks. 2002). the use of AIs produces better results than tamoxifen (Mauri et al. After second-line hormonal therapy. trastuzumab is continued as a single agent after a period of combined treatment with a cytotoxic agent. It would be normal to start with an anthracycline if it is not already being used as an adjuvant treat206 . Single-agent treatments include docetaxel.Nayyer Iqbal and Peter Barrett-Lee In postmenopausal women. In premenopausal women who have had previous antioestrogen therapy and who are within 1 year of antioestrogen exposure. initial treatment with an antioestrogen with or without LHRH agonists is preferred. The choice of regimen depends on the firstline history of received adjuvant treatment. A regimen of bevacizumab (a humanised monoclonal antibody against VEGF receptor) plus paclitaxel gives a better DFS and OS than paclitaxel alone (Miller et al. which lacks the oestrogen agonistic activity of tamoxifen and is a monthly gluteal intramuscular injection. 2005). sequential single-agent treatments are still considered appropriate. r Progestin (megestrol acetate).. capecitabine.. fulvestrant. It is at least as effective as anastrozole in patients whose disease progressed on previous endocrine therapy and has fewer side effects [Howell et al. O’Shaughnessy et al. r Trastuzumab + vinorelbine.. Slamon et al. should be considered for chemotherapy and/or trastuzumab if the disease is HER-2 positive. Treatment with trastuzumab should Primary chemotherapy Primary chemotherapy should be considered for suitably fit patients particularly if they have ER/PgRnegative. This regimen may be suitable as a first-line chemotherapy for women with recurrent breast cancer but it will need to be reviewed by regulatory bodies such as NICE before decisions to use this combination are possible. In premenopausal women without previous exposure to an antioestrogen. vinblastine and continuous-infusion 5-FU. Combinations are: r Trastuzumab + docetaxel. r Steroidal AI (exemestane). the preferred second-line therapy is ovarian ablation (by surgery. 2005.. In this situation. Second-line hormone therapy Postmenopausal women whose breast cancer responds to an endocrine manoeuvre should receive additional endocrine therapies for second-line and subsequent therapy as outlined here. NICE recommends the use of trastuzumab in patients with HER-2-positive advanced disease in combination with taxotere after anthracycline use or as a single agent after the use of a taxane (NICE. RT or LHRH agonists) with or without an antioestrogen. 2002). and are suitably fit. Patients should receive appropriate endocrine maintenance treatment between courses of palliative chemotherapy. Combination chemotherapy is better than sequential single agents but is associated with greater toxicity.g.. 2000. doxorubicin plus docetaxel/paclitaxel. 2006). Targeted therapy Trastuzumab with or without chemotherapy. paclitaxel. FAC. r Pure antioestrogen (e. symptomatic visceral or hormone-refractory disease. Patients who develop endocrine-resistant disease. Suitable regimens include FEC. CMF. r Androgens (fluoxymesterone). 1999. 2004b. r Trastuzumab + paclitaxel. carboplatin. and gemcitabine plus paclitaxel (Leonard and Howell... vinorelbine. ment followed by a taxane single agent or in combination. with or without endocrine therapy. There is evidence that trastuzumab plus chemotherapy may have a better PFS and response rate. AC. albumin-bound paclitaxel. little evidence exists to assist in selecting the optimal sequence of hormonal therapy. EC. unless or until endocrine resistance develops. Premenopausal women with ER-positive disease should have ovarian ablation/suppression and then follow postmenopausal guidelines. 2005b]). epirubicin. r High oestrogen (ethinyl oestradiol).. pegylated liposomal doxorubicin and gemcitabine. docetaxel plus capecitabine. Single-agent trastuzumab in HER-2-positive metastatic disease gives a 35% clinical response rate as a firstline treatment. r Trastuzumab + capecitabine. doxorubicin. 2001). Options are: r Non-steroidal AIs (anastrozole and letrozole). Marty et al.

should be treated intensively and. and adjuvant endocrine treatment can only be given postpartum. Those with an isolated chest wall recurrence. especially lytic ones. although the DFS at 5 years may be 25 to 30%. Locoregional recurrence may present as chest wall. Giving RT with hyperthermia is also a promising approach. trastuzumab (risk of oligohydramnios). . and an ultrasound scan of the liver. adjuvant RT. intrathecal methotrexate for leptomeningeal carcinomatosis). Their prognosis appears to be much better than for HER-2-negative patients with CNS metastases. supraclavicular or axillary recurrence. 2006). half of the patients will still be alive after 5 years. Patients also need oral calcium and vitamin D supplementation. After surgery and RT.Breast be stopped in patients who develop progressive extracerebral disease. side effects including flulike symptoms. Patients on trastuzumab have a relatively high rate of isolated relapse in the CNS because trastuzumab does not effectively cross the blood–brain barrier. surgery or regional chemotherapy (e.3% in the second and third trimesters). Taxane. a complete surgical resection with a clear margin should be aimed for followed by RT (chest wall and areas of lymphatic spread) if it was not given previously. Late diagnosis may be related to the endocrine effects of pregnancy or missed diagnosis caused by the tumour being difficult to detect within swollen breast tissue. There should be very close liaison with the obstetrician who is monitoring foetal growth and development. 30%). and larger tumours. Such patients should be considered for radiotherapy and/or further chemotherapy. 207 Other treatments Patients with metastatic breast cancer develop a number of localised problems that can be treated with local RT. clodronate. particularly after a long disease-free interval. Possible investigations include a mammogram. Management depends on the timing of the diagnosis. Special clinical situations Pregnancy and breast cancer Breast cancer during pregnancy is uncommon (Eedarapalli and Jain. is the other option. Management depends on previous treatment: r For patients who had previous BCT. During the first trimester the options are to continue pregnancy and treat with mastectomy and ALND. Chemotherapy cannot be given during the first trimester because of the high risk of foetal malformations (whereas the risk is only 1. It is not clear that there is any difference in prognosis. followed by standard treatment. Use of bisphosphonates Women with bone metastases. Management of recurrent disease Half of the patients with locally recurrent disease have evidence of visceral metastasis at presentation. The patient’s dental hygiene should also be assessed before treatment because there are increasing cases of osteonecrosis of the jaw being reported in patients treated with zoledronic acid. Termination. should be given a bisphosphonate. ibandronate (intravenous or oral) and zoledronic acid (intravenous) have all been shown to be effective in patients with breast cancer. Sometimes neoadjuvant chemotherapy or hormone therapy is given prior to surgical excision. Intravenous zoledronic acid is one of the most powerful inhibitors of osteoclast activity and has been shown to be better than pamidronate for controlling malignant hypercalcaemia and skeletalrelated events. r For patients who had previous mastectomy. which can be safely performed with appropriate shielding of the foetus. ER/PgR-negative and HER-2-positive (slightly. stage for stage.g. Pamidronate. either clinically obvious or on restaging. Patients with a locoregional recurrence can still be cured but treating them may be difficult. 2002). a chest X-ray with shielding. an ultrasound scan of the breast. chemotherapy and endocrine treatment should be considered. Patients may be diagnosed late and often have involved axillary lymph nodes. Adjuvant chemotherapy can begin in the second trimester. renal toxicity and abnormal calcium metabolism need to be monitored to decide whether necessary doses should be reduced or alternative bisphosphonates should be used. which are more likely to be poorly differentiated. which reduces the frequency of skeletal-related events and relieves pain (Brown and Coleman. However. mastectomy is the usual treatment (or further WLE in highly selected cases). compared to women who are not pregnant.

r Excision of DCIS + excision of the NAC with wholebreast RT. Examination and imaging of the breast should be performed and MRI considered if no abnormality is found.75%. with or without adjuvant endocrine treatment stage for stage. in FAC chemotherapy). ultrasound and. LCIS is not found in men. 66 to 68% are DCIS. The disease is usually unilateral and occurs most frequently in the fifth or sixth decade. Chemotherapy should not be given later than 35 weeks to avoid haematological complications at delivery. often ER. it is usually ER-positive. Management involves neoadjuvant chemotherapy (16% complete response. Standard treatment involves modified radical mastectomy with or without adjuvant RT. options include: r Mastectomy + axillary staging. the options are mastectomy or BCT plus ALND followed by adjuvant chemotherapy. may be required. Full-thickness skin biopsy of the involved NAC should be performed. mastectomy or BCT plus ALND can be performed followed by postpartum adjuvant chemotherapy. If the involved NAC is associated with underlying invasive breast cancer.g. 45% partial response) followed by mastectomy and ALND plus adjuvant RT with or 208 . It is a rare but virulent form of breast cancer with poor prognosis. Inflammatory breast cancer Inflammatory breast cancer accounts for 3% of all breast cancers. and adjuvant endocrine treatment as indicated. and infiltrating lobular carcinoma is unusual. rapid (within 3 months). A skin biopsy usually shows dermal lymphatic invasion but a skin biopsy is not essential to establish the diagnosis. Male breast cancer The median age of occurrence of breast cancer in men is in the sixth decade. Inflammatory breast cancers tend to be poorly differentiated. management options include: r Mastectomy + axillary staging. In patients with unilateral breast cancers. but methotrexate should be avoided. Paget’s disease Paget’s disease is a rare manifestation of breast cancer (1 to 4% of all breast cancers) characterised by the presence of neoplastic Paget’s cells in the epidermis of the nipple areola complex (NAC). similar to treatment for female breast cancer. Mammograms. with or without adjuvant chemotherapy. sometimes MRI of the breast. The aforementioned staging investigations should be carried out to look for metastases. r Excision of the breast tumour and excision of the NAC with whole-breast RT. adjuvant RT. Anthracyclines and alkylating agents can be used (e.Nayyer Iqbal and Peter Barrett-Lee During the second trimester and early third trimester. For palpable tumours 90 to 94% are invasive cancers. adjuvant RT and adjuvant endocrine treatment can only be given postpartum. options include: r Mastectomy ± axillary staging.and PgR-negative and more likely to have HER-2 overexpression. brawny indurations of the skin with an erysipeloid edge. with or without adjuvant RT and adjuvant endocrine treatment. It is characterised by diffuse. Another option is neoadjuvant chemotherapy followed by postpartum mastectomy or BCT plus ALND with or without adjuvant trastuzumab (risk of oligohydramnios). Appropriate adjuvant systemic therapy depends on the histology of the underlying breast cancer. Diagnosis is often delayed. Adjuvant trastuzumab (risk of oligohydramnios). the annual incidence of a contralateral primary is about 0. r Excision of a NAC with whole-breast RT. Bilateral breast cancer Of patients who present with breast cancer. The skin changes are caused by extensive involvement of dermal lymphatics. ulceration and itching of the nipple. Paget’s disease is usually (more than 90%) associated with an underlying breast cancer. and about 50% of underlying tumours are palpable. Half the patients with inflammatory breast cancer have an underlying mass. which may not be adjacent to the NAC. The patient presents with eczema of the areola. if not palpable. If there is only involvement of the NAC with no evidence of underlying malignancy. bleeding. Local recurrence occurs in about 20% of patients. If the involved NAC is associated with underlying DCIS. During the late third trimester. adjuvant trastuzumab. 1% may present with bilateral primaries in the breast.

cribriform. score 3 (if four or more LNs are positive). radical RT may be discussed as an option.9. Although not routinely evaluated. score 2 (if one to three LNs are positive). Gene amplification or overexpression of HER-2 carries an adverse prognosis. r Aneuploid as opposed to diploid tumours. Many of the so-called special types of invasive breast carcinoma (invasive tubular. Tumour grade The grade of the tumour is an important predictor of both disease-free and overall survival. r Proteases and second-messenger systems – high concentrations of cyclic AMP. involvement of four or more positive lymph nodes. Patients with ER. BCL2 and p21 WAF1. Lymphatic or vascular invasion Lymphatic or vascular invasion is present in 25% of breast cancer patients and is associated with a doubling of the rate of local recurrence and high risk of short-term systemic relapse. and/or RT should be considered. than cancer of no special type. the presence of positive internal mammary nodes is associated with a worse prognosis. Tumour size The tumour size directly correlates with 10-year survival. Younger women are at a higher risk of relapse with a relative risk of 1. Prognosis The 5-year survival according to stage of the disease is shown in Table 16. Age Younger women (< 35 years) have a poorer prognosis than older patients with cancer of an equivalent stage. score 1 (if N0). Other markers of poor prognosis Other indicators of poor prognosis include the following: r High proliferation rates (measured by fraction of cells in S-phase. microinvasive. alternative chemotherapy. score 2 (for grade 2) and score 3 (for grade 3). 20% overexpress the HER2 oncogene. The inflammatory type has a poor prognosis. cathepsin B and matrix metalloproteinase 2. especially in the presence of abnormal expression of other markers such as pglycoprotein.. and the risk of metastasis. it is also a predictive factor and patients are more likely to respond to anthracycline-based chemotherapy and to be resistant to tamoxifen. In cases of progressive disease. r TP53 accumulation. with or without adjuvant trastuzumab. The rate of local recurrence after 5 years is 17% in patients under age 35 compared with 6% in those over 50. and the presence of cathepsin D. For grading.and PgR-negative tumours have a higher complete response rate (> 30%) following neoadjuvant chemotherapy but survival time is shorter (Colleoni et al. Hormone receptor status Patients with ER-positive tumours live longer than those with ER-negative tumours and they are more likely to respond to hormonal treatment. score 1 (for grade 1). mucinous. EIC is associated with increased local recurrence after BCT. hormone therapy. Ki67 and MIB-1 monoclonal antibodies. adenoid cystic and medullary) are associated with a much better prognosis 209 . Histological type Histological type is one of the best predictors of longterm survival.6 for distant metastases.2 × size) + lymph node stage + grade For lymph node stage. 2004). Prognosis Prognostic factors Regional lymph node status The status of the regional lymph nodes is the most important prognostic factor and is directly related to survival and the best predictor of systemic micrometastases. papillary. In cases of complete response.Breast without adjuvant hormone. An extensive intraduct component (EIC) is defined as an infiltrating ductal cancer in which greater than 25% of the tumour volume is DCIS and the DCIS extends beyond the invasive cancer into surrounding normal breast parenchyma. or bromodeoxyruridine). HER-2 oncogene Of breast cancer patients. The Nottingham prognostic index (NI) is calculated as follows: NI = (0.

Radiother. (2004b). Bartelink.5 to ≤ 4. HOT is a randomised phase II trial of hyperbaric oxygen therapy for patients with chronic arm lymphoedema following RT for early breast cancer. P et al.9.. The BISMARK trial explores the cost-effective use of bisphosphonates in metastatic bone disease using a comparison of bone-marker-directed zoledronic acid therapy versus a standard schedule.ncrn. Barlow. 1). Recurrence . rates after treatment of breast cancer with standard radiotherapy with or without additional radiation. W. Lannin. (2005). et al.4 2. Poortmans. see Blamey (1996). G. Nag.. K. H.. Five-year survival according to stage of disease Stage I II III IV Adapted from Miller et al. (2001). Lancet.Nayyer Iqbal and Peter Barrett-Lee Table 16. 1378–87. Global phase III study of gemcitabine plus paclitaxel (GT) vs. 88 (Suppl. T. Results of a multivariate analysis of 1195 patients with operable breast cancer and positive axillary nodes. SOFEA is a study of fulvestrant with or without concomitant anastrozole versus exemestane following progression on non-steroidal AIs. 510. Sequence of radiotherapy with tamoxifen in conservatively managed breast cancer does not affect local relapse rates. 1713–27..4 > 5. O’Malley. A-37. Treat. node-positive. et al. doxorubicin.. et al. J.4 3. Clin.10. Am. Horiot. FAST is a prospective randomised clinical trial testing 5. H.4 4..5 to ≤ 5.. Breast Cancer Res. N. Proc. H. M. K. J. Albain. P. 11. 213–22.. Abstr. 17–23. C. Survival at 5 years 84% 71% 48% 18% Table 16...-C. Clin. Oncol. Albain.. TACT 2 is a trial of accelerated adjuvant chemotherapy with capecitabine in early breast cancer. localised breast cancer.. Engl. 23. H. POSH is a prospective study of outcomes in sporadic versus hereditary breast cancer. Ongoing clinical trials Details of current trials in breast cancer can be found at www. DIETCOMPLYF explores the role of diet. F. paclitaxel (T) as frontline therapy for metastatic breast cancer (MBC): first report of overall survival. Oncol. estrogen (ER) and/or progesterone (PgR) receptor-positive breast cancer: mature outcomes and new biologic correlates on phase III intergroup trial 0100 (SWOG-8814). TEXT is a phase III trial evaluating the role of exemestane plus GnRH analogue as an adjuvant therapy 210 for premenopausal women with endocrine-responsive breast cancer.5 to ≤ 3. 5-fluorouracil. 347. Anonymous. J. R. (1994).uk.4 Ten-year survival (%) 94 83 70 51 19 NI = Nottingham prognostic index. Soc. Oncol. Le. BBC is a British breast cancer study of risk factors for contralateral breast cancer. et al. tamoxifen) versus T alone for postmenopausal. . D. Concurrent (CAFT) versus sequential (CAF-T) chemo hormonal therapy (cyclophosphamide. Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53 297 women with breast cancer and 100 239 women without breast cancer from 54 epidemiological studies. Vu.10. S. 345. (1988). ZICE is a comparative evaluation of zoledronate (intravenous) versus ibandronate (oral).org. The 10-year survival according to the Nottingham prognostic index is shown in Table 16. REFERENCES Ahn. Long-term effect of internal mammary chain treatment.0 Gy fractions of whole-breast RT in terms of late normal tissue responses and tumour control. DCIS II is a randomised trial (currently being set up) that will test observation (no RT) against RT in women with low-risk completely excised ER-positive DCIS of the breast on adjuvant endocrine therapy. (2004a). S. 23 (5).7 and 6.. Calderillo-Ruiz. Ten-year survival according to the prognostic index Prognostic group Excellent Good Moderate I Moderate II Poor NI scoring ≤ 2... complementary treatment and lifestyle in breast cancer survival – a study of phytoestrogens in breast cancer. (1996). Med. Mouriesse. DEVA is a multicentre randomised trial of sequential epirubicin and docetaxel versus epirubicin in nodepositive postmenopausal breast cancer patients. Arriagada. AZURE attempts to answer whether adjuvant zoledronic acid reduces recurrence in patients with highrisk.

Breast. Engl.. Tripathy. The design and clinical use of the Nottingham Prognostic Index in breast cancer. N. (2004). the results of 20 years of follow-up.. (2005).. SABCS 2005. Med. Lumpectomy and radiation therapy for the treatment of intraductal breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-17.. Med. D. 4141–9. 332. 353. v128–49.. (2001).. (2002b). A. Fisher. 24–9. S. S3–13. 350. EBCTCG. J. N. Adjuvant cyclophosphamide. 14 (Suppl. (1980). 963–70. 708–13.. D. Gelber. E. et al. R. Fisher. Oncol.. EBCTCG. Budman. E. J. Surg.. W. Breast. Clin.. The present and future role of bisphosphonates in the management of patients with breast cancer. Oncol. et al. A. J. Oncol. (2006). Engl. N. R. J. Coombes. 1687–717. R. Clin. Valagussa. A. A. 602–11. H. (2003). J. Conservative management of intraductal carcinoma (DCIS) of the breast. Med. V. Cancer Res. Fisher. Clin.. Lancet. 1431–9. 5). D. et al. Forrest. Clin. 211 . (2005). Clin. Oncol. et al. N. Breast cancer and hormone replacement therapy in the Million Women Study. Redmond. C. P H. (2004). Leeming. B-04-comparison of radical mastectomy with alternative treatments for primary breast cancer. French Adjuvant Study Group. Randomised . (2001). C.M. 20. Breast cancer in pregnancy. P. P Gatta. Fisher. 17. T.. G. Cobleigh. L. et al. Blamey. Clin. L.. Verdecchia. Brown. Bijker. (2005). et al. J. N... Dignam. G. J.. S. S. Wolmark. M. Eedarapalli. Zambetti. J.. J. B. 15 (Suppl. 931–42. 90. 348. R. H.. J. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. L. Meijnen. 347... (1998). and Coleman. 351. 4. Obstet. et al. J.. N. Bonadonna. Randomized trial of dose-dense versus conventionally scheduled and sequential versus concurrent combination chemotherapy as postoperative adjuvant treatment of node-positive primary breast cancer: first report of Intergroup Trial C9741/Cancer and Leukemia Group B Trial 9741. breast-conserving treatment for ductal carcinoma in situ (DCIS): ten-year results of European Organisation for Research and Treatment of Cancer (EORTC) randomized trial 10853. J. N. Berry.. A.. Abstr. E... S. Radiation compliance and its relation to treatment outcome. U. Oncol. Anderson. 13. estrogen receptor-negative breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-23. B. Everington. (1995). 19.. 8. Viale.. M. Dignam. S. J. D. Singletary. B. E. Dimitrov. Fisher. Tamoxifen and chemotherapy for axillary node-negative. 21.. M. Bonadonna. J... Anderson. Adjuvant endocrine therapy for premenopausal women with early breast cancer. Cody. Moliterni... J. Breast Cancer Res. Benefit of a high-dose epirubicin regimen in adjuvant chemotherapy for node positive breast cancer patients with poor prognostic factors: 5-year follow-up results of French Adjuvant Study Group 05 randomized trial. M. B. Clin. Buzdar.. summary: cancer survival in Europe at the end of the 20th Century. E.. 6622–8. and fluorouracil in positive-node breast cancer patients with tamoxifennonresponsive tumors: results from the National Surgical Adjuvant Breast and Bowel Project B-15. Cirrincione. Fisher. 1–13. C. et al. Fisher. 1081–92.. P Stewart. 217. Oncol.. Lancet. Coleman. S. B. Tamoxifen in treatment of intraductal breast cancer: National Surgical Adjuvant Breast and Bowel Project B-24 randomised controlled trial. A. A. 1205–11. Cancer. Colleoni. M. 139–47. Findings from NSABP Protocol No. G. (2002). Wolmark. Gibson. Hall. J. Sentinel lymph-node mapping in breast cancer. Lancet. 419–27. 1–4.. 19. J. et al. (1999). methotrexate. B. and Million Women Study Collaborators. V. Cancer Inst. E. Dellapasqua. Engl. controlled trial of conservation therapy for breast cancer: 6-year analysis of the Scottish trial. 2639–48. C. 47. Gynaecol. Valero.. (1995). V. R. A randomized trial of exemestane after two to three years of tamoxifen therapy in postmenopausal women with primary breast cancer. D. Twenty-year follow up of a randomized trial comparing total mastectomy. (1990). Med.. Tamoxifen with or without breast irradiation in women 50 years of age or older with early breast cancer. Multinational study of the efficacy and safety of humanized anti-HER2 monoclonal antibody in women who have HER2overexpressing metastatic breast cancer that has progressed after chemotherapy for metastatic disease. D. 5. Review of the development of letrozole and its use in advanced breast cancer and in the neoadjuvant setting.. 441–52. 156–7. et al. A... Brown.. et al.. J. EUROCARE-3 . Manchul. B. Dignam. Dose and dose intensity as determinants of outcome in the adjuvant treatment of breast cancer. Oncol. et al. 10. (2003).. 23. D. N. N. 1993–2000. Engl. methotrexate. M. 8.. 1). 901–6. Moliterni.Breast Beral. Citron. et al. J. Tan-Chui. Natl. Cirrincione. J. Fisher. Clin. 333. (2004). Clin.. A.. Anderson. Berry. Colleoni. (1991).. Cancer Res. J. . et al. J. et al. or both for prevention of ipsilateral breast tumor recurrence after lumpectomy in women with invasive breast cancers of one centimeter or less. R. Oncol.. Zahrieh. Bogaerts. radiation therapy. Oncol. Evaluation of paclitaxel in adjuvant chemotherapy for patients with operable breast cancer: preliminary data of a prospective randomized trial. An overview of the randomized trials.. Oncol. D. lumpectomy. R.. et al. McCready. et al. The Cancer and Leukemia Group B. 26. et al. 1073–9. 330. M... (2006). Montague. C. J. G. Bhatnagar. (1998). et al. (2005).. and fluorouracil in node positive breast cancer. Engl. Bryant. (2002a). Collaborating NSABP investigators. S. 1233–41. L. Oncology (Williston Park). 7. J. 30 years’ follow-up of randomised studies of adjuvant CMF in operable breast cancer: cohort study. (1999). 365. Med. I. (1996).. Ann... S. 1444–55. (1996). E. J. Chemotherapy is more effective in patients with breast cancer not expressing steroid hormone receptors: a study of preoperative treatment. (2003). et al. Lancet. Effects of radiotherapy and surgery in early breast cancer. San Anton.. B. A.. 46. et al. Tamoxifen. J. A. 362. 1483–96. Clin. 25–34. (1999). 16. and lumpectomy plus irradiation for the treatment of invasive breast cancer.. (2002). R. W. P and Jain.Vogel. J. Two months of doxorubicin-cyclophosphamide with and without interval reinduction therapy compared with 6 months of cyclophosphamide. 1736–50.. Fyles. III. Bryant.. Radiotherapy in .

I. Effectiveness of switching from adjuvant tamoxifen to anastrozole in postmenopausal women with hormone-sensitive early-stage breast cancer: a meta-analysis. Improved outcomes from adding sequential paclitaxel but not from escalating Doxorubicin dose in an adjuvant chemotherapy regimen for patients with node-positive primary breast cancer. M. Surg.. Breast-conserving therapy vs mastectomy in early-stage breast cancer: a meta-analysis of 10-year survival. Demetri. Arriagada.. (2006).. et al. Elledge. Risk of lymphoedema following the treatment of breast cancer. Nat. 2. A randomized phase III trial of paclitaxel plus bevacizumab as first-line therapy for locally recurrent or metastatic breast cancer: a trial coordinated by Eastern Cooperative Oncology Group (E2100). Clin. (1994). Jonat. Sci. Switching of postmenopausal women with endocrine-responsive early breast cancer to anastrozole after 2 years’ adjuvant tamoxifen: combined results of ABCSG trial 8 and ARNO 95 trial. A.. Wang. Clin. C.. Familial Breast Cancer. Neoadjuvant systemic therapy for breast cancer: an overview and review of recent clinical trials.. 2302–13. J. Gnant. 363. D. J. Randomized trial of intensive cyclophosphamide. N.. (2002). NICE. M. (1986).. R. Survival with .. J. Prognostic factors. 21. 6. W. P. Lumpectomy plus tamoxifen with or without irradiation in women 70 years of age or older with early breast cancer. Br. Kissin.Nayyer Iqbal and Peter Barrett-Lee Goldhirsch.. G.. F. 21. G.. M. A. Oncol. et al. W. J. de Vathaire. 1–3. Pippen. C. F. A. A. L. Le. London: National Institute for Health and Clinical Excellence. (2006b). J. Expert Opin. 580–4. 236–9. S.. et al. C. H. NICE. Hanrahan. paclitaxel. Clin. L. Cancer Inst. NICE. (2006a). et al. (2004). Lancet Oncol. et al. (2005).. J. N. J. (2006)....17. E. P. Cuzick. and fluorouracil in premenopausal women with node-positive breast cancer. D. 7.. et al. Abstr. 1477–91. Easton. Goss. D. 352. 23. Berry. (2005). E. Martino. R.. (1990). P. Goyal.. Oncol. Cancer. Soc. W. J. Results of the ATAC (Arimidex. H. D. Bramwell. Tamoxifen. K. 6–12. A systematic review of docetaxel. et al.. A.. 73. Med. Does delay in starting treatment affect the outcomes of radiotherapy? A systematic review. Levine. (2006). J. Cancer J. Oncol. Gralow. Henderson. NICE. Baum. 1262–71. Querci della Rovere. (2000).. ABC of breast diseases. Med. 1285–91. D. 809–20. and Howell.nice. Mamounas.. M. Holmberg.. Engl. 97. 351. D. Harris.. K. Med. L.. B. D... Paclitaxel following doxorubicin/cyclophosphamide as adjuvant chemotherapy for node-positive breast cancer: results from NSABP B-28. et al. Bryant. Boccardo. Jakesz. I.M.. Mauri. et al. M. Oncol. F. R. Impact of concurrent versus sequential tamoxifen with radiation therapy in early-stage breast cancer patients undergoing breast conservation treatment.. C. Clin. (2005b). Trastuzumab for the Adjuvant Treatment of Early-Stage HER2-Positive Breast Cancer. et al. Pritchard. (2006c). 16... G. Guidance on the Use of Trastuzumab for the Treatment of Advanced Breast Cancer. 212 . et al. 309. Ingle. J. M. Adjuvant docetaxel for node-positive breast cancer. S. J. 455–62. Hwang. Docetaxel for the Adjuvant Treatment of Early Node-Positive Breast Cancer. (2005). Schnaper. SABCS 2005. 991–6. and fluorouracil chemotherapy compared with cyclophosphamide. 555–63. Hughes. O.. 354. aromatase inhibitors and inactivators versus standard hormonal therapy in advanced breast cancer: a meta-analysis. M. M. J. J. (2004). NICE Technology Appraisal Guidance 107.. R. 1573–6... E. P. 16. et al. Meeting highlights: international expert consensus on the primary therapy of early breast cancer 2005.uk/guidance/cg41. (2005a). 60–2. W. 22 (4)... Am. Oncol. J. B. Breast Cancer. Hennessy. P et al... Fulvestrant versus anastrozole for the treatment of advanced breast carcinoma: a prospectively planned combined survival analysis of two multicenter trials. (2005). Ellis. 66. M.. Part review of NICE Technology Appraisal Guidance 30. Oncol. Miller. 971–7.. Maraninchi. 12. L. E. Howell. et al. HABITS (hormonal replacement therapy after breast cancer – is it safe?). Adv. 23. and HABITS Steering and Data Monitoring Committees. M. Randomized phase II trial of the efficacy and safety of trastuzumab combined with docetaxel in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer administered as first line treatment: the M77001 Study Group. W. et al. Engl. (2003). Lancet.. F. E. (2005). Lancet.. Joensuu... Mansel. Oncol. London: National Institute for Health and Clinical Excellence. Gnant. London: NICE. Jonat. 4265–74. Martin.. H. D.. V. Huang. 366. J. Natl. London: National Institute for Health and Clinical Excellence. Engl. 11–16. T. R. Mackey. Proc. R. 98. R. (2005). Pavlidis. Hormonal Therapies for the Adjuvant Treatment of Early Oestrogen-Receptor-Positive Breast Cancer. NICE Technology Appraisal Guidance 109..org. (1998). B. Pienkowski.. N. J. Gelber. 104. Clin. M.. E.. Ann. Fallowfield. Howell.. Anderson. 365. Brouwers. a randomised comparison: trial stopped. 1569–83. V. K. 3. H. et al. Morris. (unpublished data). www. London: National Institute for Health and Clinical Excellence. Adjuvant docetaxel or vinorelbine with or without trastuzumab for breast cancer. and vinorelbine in the treatment of advanced breast cancer. Barbera. 453–5. M. D. N... (2005). V. Marty. Technology Appraisal Guidance 34. T. (2003). Lancet.. A. (2003). J. Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: updated findings from NCIC CTG MA. A.. et al. A. Polyzos. NICE. 2313–18. Cancer Inst. (1997).. J. Cognetti. N. J. R. Wilson. 2651–8. et al. Glick. NICE Clinical Guideline. Berry. Can internal mammary chain treatment decrease the risk of death for patients with medial breast cancers and positive axillary lymph nodes? Cancer. methotrexate. Sainsbury. Leonard. N. and Valero. Alone or in Combination) trial after completion of 5 years’ adjuvant treatment for breast cancer. Christensen. I.. et al. Clin. J. Am. J. Bono... Morris. et al. Abstr. J.. Kellokumpu-Lehtinen. Sentinel node biopsy versus standard axillary treatment in breast cancer: results of the randomized multicentre ALMANAC trial. et al. (2005). D. Miller. N. O. J. epirubicin. T. Lembersky. R. (2006d). Pharmacother. 3. et al. 976–83.

Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that over expresses HER2. Natl. et al.. J. J. 20. Surg. ASCO. Wallner. T. Thomson. J. H. 391–5. 6th edn.org. J. (2005).. 1227–32. Silverstein. (2003). Slamon.. 84.. Margolese.M. C. Abstr. C. 2005. Locoregional radiation therapy in patients with high-risk breast cancer receiving adjuvant chemotherapy: 20-year results of the British Columbia randomized trial J. SIGN. (2005). Amsterdam. ESTRO 2004.. Proc. Orecchia. M. Recommendations for Cross-Sectional Imaging in Cancer Management. Leyland-Jones. Veronesi. Clin. Cancer Inst..org. 13. et al.. Breast. (2002). Presented at the ASCO Annual Meeting. Viale. Cancer. J. J. G. Oncol. A271.. Perez. A. A. node-negative breast cancer. N. (2004). J. Engl. S. Edinburgh: Scottish Intercollegiate Guidelines Network. 97.. RCR. N. Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. Vukelja. G. Clin. et al. Shak. K. N. Proc. et al. R. Natl.. Abstr. HERA trial. S. N. Sorlie.. D. 344. (2005). A. Dunn. London: Royal College of Radiologists. J. J. 213 . 131–41. P Arthur. Suman. J. M. Workshop on . J. A. Earl. Thurlimann. Treat. 40. E. G. A multigene assay to predict recurrence of tamoxifen-treated. A. N.. S. H. 353. B. Brewster. U. S. 89. Roche. Olivotto. (1991). Parker. Adjuvant endocrine therapies for pre/perimenopausal women. Engl.. 14 (Suppl... Dewar. Poole. presented at SABCS.virtualmeeting). I. Pritchard. D. M. 349. Piccart-Gebhart. H.. Eiermann. S.asco. M.. The University of Southern California/Van Nuys prognostic index for ductal carcinoma in situ of the breast.). Nielsen. H. 2747–57. et al. S16. Natl. Spinelli. Presented at the ASCO Annual Meeting. 33.. (2005). (2004).A. Engl. et al.. M. Treatment of early-stage breast cancer. Breast. et al. Paik.Breast NIH Consensus Conference.. 1. H. partial breast irradiation: state of the art and the science. R.org. N.virtualmeeting). New York: Wiley-Liss.. Shak. FL (available at www. Soc. carboplatin and trastuzumab (TCH) in HER2 positive early breast cancer patients: BCIRG 006 study.0 cm treated with lumpectomy: preliminary results of NSABP protocol B-21. J. E. D. S8. (2003).. Proc. Med. Bryant. Meet. Sci. NEAT and SCTBG BR9601 phase III adjuvant breast trials show a significant relapse-free and overall survival advantage for sequential ECMF. E... 116–26. Presented at the ASCO Annual Meeting. TNM Classification of Malignant Tumours. J. A. No. (2001). as recommended in international consensus reports? A subgroup analysis of the DBCG 82 b and c randomized trials. O’Shaughnessy. A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. 14 (Suppl. Med. (2005). V.. Cascinelli. UICC. S. (2003). ed.asco. Ragaz. W. Med.. et al. Ann. Joint analysis of NSAPB-B-31 and NCCTG-N9831. L. Sobin and Ch. Davidson. A randomized comparison of sentinel-node biopsy with routine axillary dissection in breast cancer.. D. (2002). J. Intraoperative RT to the breast. Management of Breast Cancer in Women.. S9. J.. Miles. et al. 2812–23. J. FL (available at www. 96. L. Improvements in survival for women with breast cancer in Scotland between 1987 and 1993: impact of earlier diagnosis and changes in treatment. J... J. D. Is the benefit of post mastectomy irradiation limited to patients with 4 or more positive nodes. Wolmark. Veronesi. Engl. U. A. and Overgaard.virtualmeeting).). Bartelink. J. (2000). Wittekind. Eur. Overgaard. Oncol. Superior survival with capecitabine plus docetaxel combination chemotherapy in anthracycline-pretreated patients with advanced breast cancer: phase III trial results. (2005). 337–43. FL (available at www. Am. 743–53.. et al. Perez. 175–84. 100. (2006). 2817–26. J. B. 19. R. 8418–23.. 186. N. Phase III randomized trial comparing doxorubicin and cyclophosphamide followed by docetaxel (ACT) with doxorubicin and cyclophosphamide followed by docetaxel and trastuzumab (ACTH) with docetaxel.. Orlando. analysis of the PACS 01 trial: 6 cycles of FEC100 vs 3 cycles of FEC100 followed by 3 cycles of docetaxel (D) for the adjuvant treatment of node positive breast cancer. P Spielmann. USA.. The role of radiotherapy and tamoxifen in the management of node-negative breast cancer -1. N. et al. Engl. J. Orlando. Fumoleau. M.. Romond. (2005). Tibshirani.. Med. Acad. (2003). Keshaviah. pp...asco.. Coates. Abstr. Abstr. A. Cancer Inst. 546–53. Five years . Dignam. Med. (2002).. (2005). (2004). Repeated observation of breast tumor subtypes in independent gene expression data sets. 783–92.. et al. Robert. Slamon. H. et al. J.. Mariani. 347. J. Am. Paganelli. N. (2005).. et al. Tang. A National Clinical Guideline. Further analysis of NCCTG-N9831. et al. (2004). J. Orlando. J. et al. (2004).. Breast Cancer Res. 351. 265.

resulting in approximately 3500 deaths. which was used as a contrast agent in the 1920s. and there seems to be a dose-response effect). Acquired cystic kidney disease. malignant primary or metastatic. The right kidney abuts the liver and stomach and the left. Renal cell carcinoma Risk factors and aetiology Risk factors for RCC include smoking (up to one-third of cancers may be due to smoking. which occurs in nearly half of patients on dialysis. and type II. their Types of kidney tumour Kidney tumours can be benign. Occupational factors include leather tanning.cancerresearchuk. but they are correctly termed renal cell carcinoma (RCC). especially in women. More than 30% of patients present with metastatic disease. with phaeochromocytoma. is associated with hepatocellular carcinoma and cholangiocarcinoma as well as RCC. infarction). sarcoma and renal pelvis tumours (5% of malignant renal cancers arise from the renal pelvis. inflammatory (infection. lymphoma. not only because of an increase in incidental findings. There are two subtypes: type I. the death rate is still rising. approximately 6500 new cases of kidney cancer are diagnosed per year in the UK (Cancer Research UK. info. Tuberous sclerosis and adult polycystic disease both involve autosomal dominant inheritance. arising from the proximal renal tubular epithelium. Unlike some other cancers. and adult polycystic disease. adenoma. Genetic risk factors for RCC include Von HippelLindau disease. 214 . is also a factor. It is the 10th most common cancer in men and 15th most common cancer in women in the UK. Also thorotrast. on the right draining to paracaval and aortocaval nodes. Transitional cell carcinoma of the renal pelvis accounts for 5% of all renal malignancies. accessed September. and on the left to the para-aortic region. 2006). There are about 6500 new cases per year diagnosed in England and Wales. Disease incidence has increased over the past 30 years. the right lying slightly lower than the left. The majority of malignant tumours are adenocarcinomas. and more than 90% of these are transitional cell carcinoma). Men are more frequently affected than women. The lymphatics drain along the renal vessels. Kidney cancer occurs most commonly in people 50 to 80 years of age. and is covered separately at the end of the chapter.17 KIDNEY Jason Lester and John Wagstaff Introduction Cancer of the kidney represents 3% of adult malignancies. Incidence and epidemiology Representing 3% of adult malignancies. multiple). stomach. without phaeochromocytoma. this disease occurs in 1 in 36 000 births and is associated with clear cell adenocarcinomas. org. Malignant primary tumours include RCC. the male-to-female ratio is 5:3. which are often multiple and bilateral. Benign tumours include cysts (simple. each approximately 11 cm in length. obesity. These tumours were previously called hypernephroma because it was believed that they originated from adrenal rests. the spleen. Anatomy The kidneys are retroperitoneal structures that lie between the 11th rib and the transverse process of the 3rd lumbar vertebral body. Each kidney is surrounded by perinephric fat which in turn is covered by Gerota’s fascia. In Von Hippel-Lindau disease. and pancreas. and oncocytoma. complex. occurring most commonly between the ages of 50 and 70 years. The risk is up to 30 times greater in dialysis patients with cystic changes compared to the general population. the gene involved is on the short arm of chromosome 3. shoe working and asbestos exposure. and use of phenacetin analgesics. Men are more frequently affected than women. tuberous sclerosis.

1 (Kovacs et al. malaise. For example. It spreads to lymph nodes at the renal hilum. via lymphatics. and via the blood to lung. Regions that are frequently lost are 3p12–14. Pathology Classification of renal cell tumours The major morphological classifications discriminate eight subtypes of primary renal cell tumours that are related to the basic cell types of the nephron from which they are derived. Several studies involving large numbers of patients have shown good correlation with survival (Murphy et al. polycythaemia (EPO-like molecule). 1997). Spread Renal cancer can spread locally. inferior vena cava. and skin. Multinucleated giant tumour cells are seen only in grade IV tumours.infobiogen . 1982). Grade I tumours demonstrate small. soft tissues. 50% of patients will have two of these symptoms.1. Other symptoms include fever. and paracaval regions. (1997).Kidney Table 17. respectively. unclassified Sarcomatoid differentiation can occur in both clear cell and papillary variants of renal cancer.. Clinical presentation The tumour may remain clinically silent until metastases develop. The pathological features of RCC are shown in Table 17. the highest grade is always assigned.2. and via the blood. When tumour heterogeneity is present. nuclear contour. 3p21 and 3p25. This will be left-sided because of the different anatomy of the left and right renal veins. weight loss. the liver. Adapted from Kovacs et al. Cytogenetics Histological differentiation of benign from malignant primary renal tumours can be difficult. uniform nuclei without nucleoli. The Heidelberg classification is shown in Table 17. and those secondary to metastatic spread such as bone pain. Gerota’s fascia. disease incidences are 1 in 10 000 and 1 in 1000. Cytogenetic analysis can be a useful aid.. abdominal para-aortic. because it varies among tumours and it does not correlate well with prognosis. The Heidelberg classification of renal cell tumours Type Benign Examples Metanephric adenoma and metanephric adenofibroma Papillary adenoma Renal oncocytoma Malignant Common or conventional (clear cell) RCC: most common form of RCC. accessed September 2006). and perinephric tissue. renal vein. hypertension (renin). Mitotic activity is not considered Paraneoplastic syndromes Paraneoplastic syndromes include hypercalcaemia (PTH-related peptide). 215 . A varicocele can occur in up to 2% of males due to obstruction of the testicular vein. clear cell carcinoma is characterised by loss of part of the short arm of chromosome 3. Local spread is to the adrenal gland.. and haematuria occurs in only 19% of patients and predicts a poor prognosis. The classic triad of pain. and hepatic dysfunction (unknown mechanism). www. Four grades are recognised based on nuclear size. flank mass. Grade The Fuhrman nuclear grading system is widely used (Fuhrman et al.fr/services/chromcancer/Tumors/RenalCellCarcinID 5021. it was previously considered to be a separate entity but is not RCC = renal cell carcinoma. and the presence of nucleoli. comprising 70–75% of cases Papillary (formerly chromophilic) carcinoma: 10–15% of cases Chromophobe carcinoma: 2–5% of cases Collecting duct carcinoma: 1% of cases RCC. Other aberrations include trisomy of chromosome 5 (Infobiogen. bone. the CNS. which occasionally may demonstrate spindling and severe nuclear anaplasia resembling a sarcoma. However. sweats. 1994).html.

cytokine therapy or palliative radiotherapy. Contrast-enhanced CT scans of the thorax.Jason Lester and John Wagstaff Table 17. metastectomy. Postmortem results have shown that approximately 50 percent of people over age 50 have one or more renal cysts. Treatment overview Localised or locally advanced resectable disease (stages I to III) Surgical resection is the treatment of choice where appropriate. Staging classification The regional lymph nodes are the hilar. therefore. haemorrhage. In 1950. In addition. MRI is useful for imaging the vena cava. abdomen and pelvis are used to look at perirenal extension. This classification system has been shown to be useful in separating tumours requiring surgery from those that can be safely followed up. In 1986. full blood count and a biochemical profile. and so a biopsy may be misleading. an increasing number of small tumours that had already metastasised were reported. abdominal paraaortic. a postmortem study reported that metastases were rare when renal tumours were less than 3 cm in size. 1992). Radical nephrectomy includes removal of the kidney. renal vein/caval involvement.4 show the TNM classification and stage groupings for RCC. the tumour may contain single or multiple fluid-filled cysts Microscopic Clear cell carcinoma: clear cytoplasm. and these tumours should be considered benign adenomas (Bell. Surgery can be extended to remove the Staging investigations Renal ultrasound can be useful in evaluating questionable cystic renal tumours if CT imaging is inconclusive. Surgery Radical nephrectomy Surgical resection is the only potentially curative therapy for RCC. Inoperable or metastatic disease Treatment for inoperable or metastatic RCC is individualised.3 and 17.. adrenal gland and perirenal fat within Gerota’s fascia. are also useful. cytoplasm crowded with glycogen deposits Collecting duct carcinoma: tubular growth pattern. Bosniak created a four-part classification of cystic renal masses that are found on CT scans (Bosniak. reticulated cytoplasm. 1986). and other studies indicate that almost one-third have at least one renal cyst that is identifiable on CT (Tada et al. With the advent of CT scans. Adjuvant treatment is under investigation and should not be used routinely outside clinical trials. Consider palliative nephrectomy. including serum calcium and alkaline phosphatase. there are types I and II with different molecular defects Chromophobe carcinoma: polygonal tumour cells with a transparent. histological features of adenoma and RCC overlap. pulmonary metastases. anaplastic nuclei Investigation and staging Differentiating benign from malignant tumours There are no reliable criteria to distinguish benign renal adenomas from RCC. and paracaval nodes. The system uses Hounsfield units to categorise these lesions in order of increasing probability of malignancy. A bone scan. lymph node enlargement and 216 . condensed hyperchromatic nuclei Papillary carcinoma: centrally located small nuclei and ground glass eosinophilic staining cytoplasm. solid. Pathological features of renal cell carcinoma Features Macroscopic Description Clear cell carcinoma typically arises from the cortex. the average size at diagnosis is 7 cm. lobulated and yellow. Tables 17. tumour size is no longer considered a reliable criterion to distinguish RCC (Aso and Homma. with or without a regional lymph node dissection. necrosis and/or calcification are common. A renogram (DMSA or MAG 3) should be performed if renal impairment is involved. 1983). 1950). a pseudocapsule is often present. Abdominal ultrasound and CT detect incidental renal cysts in a significant proportion of people. basophilic cytoplasm. papillary architecture.2.

may be a preferred alternative to bilateral nephrectomy with dialysis or transplantation. and technical difficulties in defining surgical margins. Stage grouping of renal cell cancer Stage I II III Description T1 N0 M0 T2 N0 M0 T3 N0 M0 or T1–3 N1 M0 IV T4 N0–1 M0 or Any T N2 M0 or Any T Any N M1 Adapted from UICC (2002). the right atrium. It may alleviate pain and bleeding and may improve paraneoplastic syndromes such as hypercalcaemia. and survival is equivalent to an open operation. and long-term survival has been reported. provided that their performance status is adequate. Spontaneous regression of metastases following surgery occurs in fewer than 1% of cases (Montie et al. Both of these trials showed significantly better survival for patients undergoing nephrectomy. Increasing evidence suggests that a partial nephrectomy is curative in selected cases (tumours < 4 cm in size). The need for pain medications is reduced. confined to the kidney ≤ 4 cm > 4 cm and ≤ 7 cm > 7 cm limited to the kidney Extends into major veins or directly invades adrenal gland or perinephric tissues but not beyond Gerota’s fascia T3a Directly invades adrenal gland or perinephric tissues but not beyond Gerota’s fascia T3b T3c T4 N0 N1 N2 M0 M1 Grossly extends into renal vein or vena cava or its wall below the diaphragm Tumour grossly extends into vena cava or its wall above the diaphragm Directly invades beyond Gerota’s fascia No regional nodal metastases Single regional node More than one regional node No distant metastasis Distant metastasis Table 17. patients with a long disease-free interval following surgery survive longer than those presenting with a solitary metastasis synchronous with a primary lesion (O’Dea et al. tumour or tumour thrombus from the infra. secondary to spillage) is no lower with this approach. 217 . A transperitoneal or retroperitoneal approach may be used.4. demonstrated clearly that the local recurrence rate (i. when technically feasible. therefore.Kidney Table 17. Adapted from UICC (2002). In patients with bilateral stage I tumours (concurrent or subsequent). Laparoscopic surgery tends to incur less morbidity and is associated with a shorter recovery time and less blood loss.. bilateral partial nephrectomy or unilateral partial nephrectomy with contralateral radical nephrectomy. Mickisch et al.. Nephrectomy should. 1977). 1978).and supradiaphragmatic vena cava and.3. be performed in all patients who are to undergo immunotherapy. 2001. Many believe that this approach should be used wherever possible. Two randomised trials have examined the role of nephrectomy in patients who were subsequently treated with interferon for metastatic disease (Flanigan et al.. TNM staging of renal cell carcinoma Stage T1 T1a T1b T2 T3 Description ≤ 7 cm.e. but operating room time and costs are higher. As expected. cardiopulmonary bypass is required for the latter. Palliative nephrectomy Palliative nephrectomy can be performed if the burden of metastatic disease is small and the patient is fit. 2001). Laparoscopic (partial) nephrectomy Laparoscopic nephrectomy is a less invasive procedure. limited experience. rarely. Disadvantages include concerns about spillage. Metastectomy Solitary metastases can be resected. It is not used in routine clinical practice. Studies of laparoscopic nephrectomy have now Radiotherapy Preoperative radiotherapy Studies of preoperative radiotherapy have not shown any evidence of benefit..

on a linear accelerator.5% in ten non-immunotherapy arms and 4. Adjuvant radiotherapy should not be used in routine clinical practice. intermediate (70%) and poor (11%). but results are disappointing. 95% CI 0. An 8 Gy single fraction or 20 Gy in five fractions are suitable dose fractionations. Palliative radiotherapy technique The patient lies supine on the X-ray simulator couch with arms by sides. In this analysis the addition of anything to single-agent interferon did not significantly affect the outcome. 1987). demonstrated that immunotherapy (interferon and/or IL-2) produced an objective response rate (RR) of 12. 14. Because tachyphylaxis usually develops.6. In four studies (644 patients) the odds ratio for death at 1 year was significantly reduced by interferon therapy (OR 0. 2003. The median survival of these groups was 28. respectively. cytokine therapies are the basis of current systemic treatment. published in the Cochrane library (Coppin et al.4–0. 218 .8-month improvement compared to non-immunotherapy study arms. Treatment field edges are positioned to cover part or all of the affected kidney. Adjuvant vaccine therapy A German trial randomised 558 radical nephrectomy patients to autologous renal tumour cell vaccine or surveillance and reported a modest improvement in progression-free. showing a limited antitumour effect. The EORTC 30955 trial has randomised high-risk postoperative patients (T3+ and/or N1+) to the Atzpodien regimen (see discussion later in this chapter) or to surveillance. Side effects of radiotherapy include tiredness. particularly bone pain.77). nausea (give prophylactic antiemetics) and diarrhoea. 1973. Based on these numbers. HYDRA is an ongoing NCRN study of similar design. Adjuvant therapy is not currently used in routine clinical practice. Kjaer et al. but not overall. It can be used to control bleeding. however. Bone lesions from kidney cancer are lytic and can be aggressive. These trials predated conformal therapy.9% (CR 3. Use parallel-opposed anterior and posterior fields. In defining the target volume...Jason Lester and John Wagstaff Adjuvant radiotherapy Two old randomised trials showed no benefit from postoperative radiotherapy (Finney...) For large tumours.6 and 4. 2005). CT planning allows more accurate delineation of the disease and can reduce the dose to normal tissues. 14. The addition of two other prognostic variables (prior radiotherapy and none or one versus two or more metastatic sites) to these initial five factors divided the patients into favourable (37%). No definite dose-response relationship has been proven for interferon therapy.6%). (Make sure you are treating the correct kidney. but results are often disappointing when it is used for pain relief. This expensive treatment (18 000 Euros) is not widely used in routine clinical practice. intravenous contrast is given to outline the kidney.5 months. Five factors (Table 17. This study is closed to recruitment and results are awaited.3% in two placebo arms. a widely used protocol is to begin with 3 million units subcutaneously three times weekly. which was a 3. respectively.. 2004). Messing et al. The median survival was 13. Palliative cytokine therapy: interferon α At present.5) were used to divide patients into three prognostic groups: favourable (19% of patients).56. The dose is 20 Gy in five fractions over 1 week prescribed to the midplane.3 months. interferon therapy should not be offered to patients in the poor prognostic group. survival with the vaccine (Jocham et al.3 months. 6 to 10 MV photons.. Radiotherapy for metastases Palliative radiotherapy can alleviate the symptoms of metastatic disease. Fatal liver damage was reported and the complication rates were high. 2003). knees supported by a polystyrene wedge and without further immobilisation. intermediate (35%) and poor (28%) with median survival times of 26. Palliative radiotherapy to the primary tumour Radiotherapy plays no role in the curative treatment of RCC.4 and 7. An overview. These prognostic groupings have been used to select patients for clinical trials of the new targeted agents that are discussed later in this chapter. Biological treatment Adjuvant immunotherapy There is no evidence that adjuvant interferon α (IFNα) or interleukin-2 (IL-2) monotherapy improves survival following potentially curative radical nephrectomy (Clark et al. This was compared with an RR of 2. 2005). Several prognostic scoring systems have been developed that can help select patients for treatment with immunotherapy (Mekhail et al.

One recent study (Atkins et al. nausea. but it is not used in routine clinical practice in the UK. with 80 to 85% of patients being alive after 10 years and probably cured of their disease. It should not be used in routine clinical practice outside a clinical trial. Risk groups are defined as follows: favourable. High-dose bolus IL-2 is the only therapy that has been shown to cure metastatic renal carcinoma. Patients may. but the treatment is associated with significant toxicity and two other phase II trials failed to confirm the high response rate. Side effects include flu-like symptoms. renal impairment. (2005).5 mM/l) Poor prognosis < 80 < 12 IX (CAIX) have an almost 40% chance of long-term survival and possible cure. and whose tumours stain positively for CAIX should be referred for this therapy. Prognostic variables for renal cell cancer Factors KPS Time from diagnosis to treatment with interferon α (months) Haemoglobin < lower limit of laboratory’s reference range Lactate dehydrogenase > 1. anorexia and depression. Hormonal treatments Palliative progesterones Progesterones were used for many years based on responses seen in animal models and a lack of any other effective systemic therapy. nausea. Auto-immune thyroid disease also occurs frequently. one or two poor prognostic factors present. capillary-leak syndrome. with no proven clinical benefit (Kjaer. Palliative combination cytokine therapy To date.Kidney Table 17. 1988). Other treatments In patients who are not candidates for surgery. no poor prognostic factors present.5× upper limit of laboratory’s reference range Corrected serum calcium (mg/dl) > 10.5. hypersensitivity reactions. suicidal behaviour. poor. These responses have been shown to be very durable. bone marrow suppression and. Conventional chemotherapy has not been shown to give a survival advantage.0 (equivalent to > 2. IL-2 and 5-FU). Treatment should be continued until progressive disease is documented. and response rates are poor (Yagoda et al. rarely. and hypertriglyceridaemia can occur. The median time to response is 6 months. Patients who are young. High-dose therapy is associated with higher complete response rates than IFN (7 to 8%). Chemotherapy escalating to 6 and then 9 million units at fortnigtly intervals. skin rashes.. but this has been shown to be a good phenomenon because its presence predicts a better outcome. Adapted from Mekhail et al. RCC is chemoresistant. arterial embolisation or radiofrequency ablation can provide 219 . diarrhoea. A small randomised trial has shown a survival advantage of using the Atzpodien regimen compared to tamoxifen in metastatic RCC (Atzpodien et al. This treatment is no longer recommended. there is no evidence that IFNα/IL-2 combinations with or without conventional chemotherapy show a survival advantage over monotherapy. have an excellent performance status. This regimen is currently being compared with single agent IFNα in a randomised trial in advanced RCC (REO4). IL-2 seems to incur a response rate and median survival similar to that of IFN-α... cardiovascular. 2005) has shown that patients whose tumours stain positively for the expression of carbonic anhydrase Palliative tamoxifen Tamoxifen may have a low level of activity in high doses (100 mg/m2 per day). KPS = Karnofsky performance status. have increased appetite and an improved feeling of well-being when on these agents. Side effects include influenza-like symptoms. Rarely. Response rates are 0 to 2%. more than two poor prognostic factors present. lethargy and malaise. however. cardiac toxicity and confusion. 2001). Palliative cytokine therapy: interleukin-2 In the Cochrane overview. renal and hepatotoxicity. 1995). intermediate. High response rates have been reported with the Atzpodien regimen (IFNα.

and 5-fluorouracil (5-FU) in patients with a high risk of relapse (T3b to T4 or N1/2 or positive margins or vascular invasion) after surgical treatment for RCC. r III – 20 to 60%. absence of weight loss. Side effects are simi220 . lar to those of sunitinib. Ongoing clinical trials HYDRA is a phase III trial that tests adjuvant IL-2.Jason Lester and John Wagstaff palliation. The majority of patients present with frank haematuria. Following radical nephrectomy. Carcinoma of the renal pelvis Renal pelvis carcinoma is a relatively rare tumour. constituting 5% of all renal tumours. work in the dye and textile industry. Ureteroscopy allows direct visualisation and biopsy of the majority of tumours. Bayer) are oral multitargeted TKIs which show promise in RCC. Follow-up with urine cytology and cystoureteroscopy is recommended. r II – 90%. nausea and stomatitis. Temsirolimus is an mTOR inhibitor that has been tested in a randomised trial comparing temsirolimus alone. Progression-free survival was significantly prolonged compared to placebo (24 versus 12 weeks. more than 90% are transitional cell carcinomas (TCCs). with inhibition of VEGF and PDGF receptors. r IV – 0 to 20%. In metastatic disease. 2006b). REO4 is a phase III trial involving IFNα versus IFNα.. An IVU will often demonstrate a filling defect in the collecting system. Localised/locally advanced disease can be treated by radical nephrectomy including total removal of the ipsilateral ureter. TCC is often multifocal. Median survival was 7 months for IFN alone and 11 months for the temsirolimus-alone arm (p = 0. r II – 50 to 70%. increased survival is associated with good PS. progression-free and overall survival (Motzer et al. r IV – 33%. r III– 55%. IFNα. removal of the primary tumour and a long disease-free interval between nephrectomy and the appearance of metastases. IL-2 and 5-FU in metastatic RCC. Overall survival Five-year survival by stage is r I – 60 to 90%. because up to 50% of patients will Areas of current interest Tyrosine kinase inhibitors (TKI) Sunitinib (SU011248. Both of these agents are licensed in the UK for use after cytokine failure but have not yet been evaluated by NICE. Temsirolimus has been licensed as an orphan drug in the European Union. The major risk factor is smoking. There was no statistically significant difference in overall survival between the two-drug combination and IFN alone. diarrhoea. They have both direct antitumour effects and antiangiogenic activity. and up to 50% of patients with TCC may also develop bladder cancer. p < 0.0069). urban residence. IFN alone and the combination as a first-line therapy in patients with a poor prognosis according to the Motzer scoring system (Hudes et al. It is more common in men (male-to-female ratio of 2:1) and usually occurs between the ages of 50 and 70. presence of only pulmonary metastasis. and CT or MRI is used to assess local extent and distant spread. Early stage low-grade cancers are often treated with a more conservative local surgical excision. accounts for the majority of the remaining malignant tumours. Phase II studies using sunitinib in patients failing prior cytokine therapy have shown overall response rates of up to 40% (Motzer et al. chronic inflammation and Balkan nephropathy. The drug appeared to be well tolerated and side effects included fatigue.000001) and overall survival data is awaited. Sorafenib has been tested in a phase II randomised placebo-controlled trial in patients who had failed at least one prior systemic therapy (Escudier et al. A phase III trial randomising patients to sunitinib or IFN has shown superiority in terms of response rate. and 5-year survival (M0 disease) by Furhman grade is r I – 90%. Prognosis Prognostic factors Survival is dependent on the stage and grade of the tumour. which is usually associated with chronic inflammation or infection of the renal pelvis... 2005). Other factors include phenacetin use.. 2006). Squamous cell carcinoma. Pfizer) and sorafenib (BAY 43– 9006. 2006a). renal vein involvement alone does not seem to adversely affect 5-year survival but does reduce 10-year survival. The beneficial effect on longer-term disease control and palliation is uncertain.

Int.. LBA3. C.. 158. F. et al. A. Yamagishi. Hoffmann.. et al. Motzer. S. Nephrectomy followed by interferon alfa-2b compared with interferon alfa-2b alone for metastatic renal-cell cancer. T. 23.W. M. Limas. Ed. 1130–6. P L. E. J. pp. et al. M. Urol.. Tada. 24 (18 Suppl. Clin. Radiotherapy in the treatment of hypernephroma: a clinical trial. Eisen. D.. randomised controlled trial. J.. (2005). 665–72. J. 832–41. Illiger. Lancet. Prognostic significance of morphological parameters in renal cell carcinoma. J.. Semin. Radiol. Kovacs. R.. Biol. J. J. J.. The role of medroxyprogesterone acetate (MPA) in the treatment of renal adenocarcinoma.. H. (2006b). Radiat. 131–3. Abou-Jawde. M. T. A. Atkins.). Activity of SU11248.. R. Clin. Kobayashi. Tomczak. Oncol.. G. et al. poor-risk patients with advanced renal cell carcinoma (adv RCC). (1950). 193–5. M. H. in patients with metastatic renal cell carcinoma. 3714–21. (2003). Richter.. 6. (1987). 2nd edn. Carbonic anhydrase IX expression predicts outcome of interleukin 2 therapy for renal cancer. Lasky... S. J. et al. Lancet. J. Renal Diseases. (1978). Clin. A survey of incidental renal cell carcinoma in Japan. M. A. (2005)... et al.. Hudes.. Br. Atkins. Oncol. 85. 21. van Poppel. Tumors of the kidney. B. M. Tomczak. Phase III . J. et al. B. Akhtar. 3-arm study of temsirolimus (TEMSR) or interferon-alpha (IFN) or the combination of TEMSR + IFN in the treatment of first-line. (1983). Michaelson. J. REFERENCES Aso. N. Wilding.. C. Radical nephrectomy plus interferon-alfa-based immunotherapy compared with interferon-alfa alone in metastatic renal-cell carcinoma: a randomised trial. (2005). Rev. IL-2 in combination with IFN-alpha and 5-FU versus tamoxifen in metastatic renal cell carcinoma: long-term results of a controlled randomized clinical trial. J. Stewart.. Y. Clin. Messing. Randomized phase III trial of the Raf kinase and VEGFR inhibitor sorafenib (BAY 43–9006) in patients with advanced renal cell carcinoma (RCC). J. J.. PA: Lea and Febiger. Y. Murphy. and Farrow. Carducci. Mekhail. Bosniak. a multitargeted inhibitor of vascular endothelial growth factor receptor and platelet-derived growth factor receptor. T. D. J.. (1982). G. J. Awa. Clin. Kjaer. P et al. 380s. G. R. D. H. Garin.. Zincke. L. M. Oncol. 23. Pathol. R. G. E. 340–3. Urol. Escudier. et al.. B.. Validation and extension of the Memorial Sloan-Kettering prognostic factors model for survival in patients with previously untreated metastatic renal cell carcinoma... Porzsolt. Clark. (2006). C. S. M. Abi-Rached. B. Blumenstein. Pathol. 16–24. (2002). S. 221 . Oncol. and related urinary structures. In: Atlas of Tumor Pathology. UICC. and Homma. 966–70. (2003). 540–2. (1995). B. H. Oncol. Oncol. 1655–9. A. J. et al. Manola. C.. Kirchner. Postoperative radiotherapy in stage II and III renal adenocarcinoma: a randomised trial by the Copenhagen Renal Cancer Study Group.. (2001). 26–40. 6th edn. D. 594–9. 345. A. J. J. et al... G. 34... In TNM Classification of Malignant Tumours. A. The role of adjunctive nephrectomy in patients with metastatic renal cell carcinoma. Oncol. et al. randomised trial of sunitinib maleate (SU11248) versus interferon-alfa (IFN-a) as first-line systemic therapy for patients with metastatic renal cell carcinoma (mRCC).Kidney develop another renal tract cancer.. (2004). E.. (2001).. 272–5. Regan.. H. 117. J. H... (1977). Phase III study of interferon alfa-NL as adjuvant treatment for resectable renal cell carcinoma: an Eastern Cooperative Oncology Group/Intergroup trial. (1986). Oncol. Soc. MD: Armed Force Institute of Pathology. LBA4.. Bethesda.. 24 (18 Suppl. et al. M. W. Coppin. Montie. Yagoda. which will greatly reduce the risk of subsequent cancers developing. A. 13. J. Beckwith. G.). R. Atzpodien. (1994). Szczylik. Frederiksen. Oncol. The patient MUST be advised to stop smoking. (1988). Clin. Am. M. C. 45. (1992). Cancer Res. P. Flanigan... 22. et al. 15. Engl... 1214–22. 437–9. CD001425. (1997). H. T. J. R. Finney. 120.. Mickisch. O’Dea. 147. Surg. (2001). 11.. A. J. M. Rev. 358. Proc.. et al. Motzer. et al. Clin. A phase 3. Beckwith. J. Br. G. Urol. 21. Kjaer. Bell. Salmon. J. (2006a). Boumerhi. Straffon. E. Philadelphia. Am.. randomized. 3133–40. L. Utz. 655–63. The Heidelberg classification of renal cell tumours. Wittekind. Urol. The treatment of renal cell carcinoma with solitary metastasis. C. Radiology. Immunotherapy for advanced renal cell cancer. Jocham. et al. Chemotherapy for advanced renal-cell carcinoma: 1983–1993. Adjuvant autologous renal tumour cell vaccine and risk of tumour progression in patients with renal-cell carcinoma after radical nephrectomy: phase III. Clin. Clin. McDermott. B.. J. 24. M. et al. 3rd edn. B. et al. D. Med. Cancer. Urba. The incidence of simple renal cyst by computed tomography.. Petrylak. Fuhrman. Phys.. 183. 363. L. M. Adjuvant high-dose bolus interleukin-2 for patients with high-risk renal cell carcinoma: a cytokine working group randomized trial.. and Engelholm. (1973). 1–10... M. A. The current radiological approach to renal cysts. bladder. Hutson. (2005). 1. Redman... New York: Wiley-Liss... et al. 195–209. 42–60. Cancer Treat. E. Sobin and Ch. Cochrane Database Syst. H. .

the vasa deferentia.org/). The peak age of incidence is 69 to 71 years. the bladder has two inferolateral surfaces and one superior surface. It lies in the anterior pelvis behind the pubic bone and its apex is joined to the umbilicus via the urachus. from superficial tumours. These include chronic urinary stasis (e. and in men it is separated from the rectum by the seminal vesicles.5% of cases). and the rectovesical fascia. 2-amino-1-naphthol and acrolein (rubber dye). When empty. to highly aggressive and infiltrative tumours. and in men the bladder neck rests on the prostate gland. Types of bladder tumour The types of bladder tumour that can occur are shown in Table 18. Anatomy The shape and size of the bladder change depending on how much urine it contains. which increases the risk about four-fold. and nearly half of those affected will die from their disease. routine screening for microscopic haematuria may be indicated . Historically. continuing to smoke increases the progression rate of non-invasive cancers. The ureters join the bladder at the base. Conditions associated with ‘squamous metaplasia’ can lead to the development of squamous cell carcinomas. Bladder cancer is more common in white people. transitional cell cancers were associated with exposure to various industrial chemicals including analine dyes. in the subtropics. Each year in the UK. The neck of the bladder 222 Screening Screening for asymptomatic haematuria is not recommended because its positive predictive value is too low (0. Although radiotherapy is still an important part of radical and palliative treatment. More important. www. connects with the urethra. 4-aminobiphenyl. Bladder cancers have a wide range of biological behaviours. the prevalence in men is second only to prostate cancer. Incidence and epidemiology Bladder cancer is the fourth most common cancer in men (6% of cancer cases) and the eighth most common cancer in women (2. which can be treated by local resection. chronic infection with schistosomiasis – a major public health problem in the Middle East. around 10 000 people develop a bladder cancer. longterm catheter. Carcinoma of the bladder Risk factors and aetiology The majority of transitional cell bladder cancers are related to a history of smoking (probably because of the excretion of carcinogens such as 4-aminobiphenyl).5%) to warrant mass screening. The treatment of invasive disease has traditionally been radical radiotherapy that has been given in an effort to avoid cystectomy. and when full. evidence is emerging that combined treatment modalities may improve outcomes. Stopping smoking reduces the risk of developing cancer but it takes many years for the risk to return to normal.1. There has been a steady rise in the occurrence of bladder tumours over the past 20 years (33% in the USA). In women. However. The incidence is 17 per 100 000 of the population (CRUK. 2-naphthylamine. Health and safety procedures should now prevent exposure to these chemicals. it becomes ovoid.g. bladder stones or paraplegia) and.18 BLADDER Stephen Williams and Jim Barber Introduction Bladder cancer is an important cause of morbidity and mortality and has a high incidence rate in industrialised countries. 4-nitrobiphthol. the bladder is related posteriorly to the uterus and vagina. Because of recurrences. where squamous cell carcinomas of the bladder are relatively common.cancerresearchuk. benzidine.

On cystoscopy. They are most common in the base of the bladder. A TCC of high malignant potential is more often solid or flat and is likely to grow both into and outside the muscle wall of the bladder. small-cell or sarcomatoid elements. Apart from some benign conditions such as the bladder papilloma. many of them arising from such areas. dysuria and increased urinary frequency. for populations exposed to bladder carcinogens.g. adenocarcinoma. but they may also occur around the trigone (possibly originating from cystic glandularis). Even if an obvious cause is present (e.1. A conservative approach with local resection is the usual treatment. but this is a relatively rare event.g. This is a condition that must be taken seriously. Histologically these tumours are of low grade (G1 to 2) with mild cellular dysplasia. they fall into two broad groups (with some overlap). vagina) Distant spread from tumours at other sites invaded the muscle layer have a high chance of local recurrence. from prostate. stones and any structural 223 . Overall at least half these patients will die of their cancers. A TCC of low malignant potential is likely to be papillary and superficial. Even tumours that have not Clinical presentation Patients usually present with haematuria and all cases of haematuria should be referred urgently as suspected cancers to a urologist for further investigation. Such tumours are normally treated in the same way as TCCs. where prompt and efficient investigations can be performed. Patients with bladder cancer may also complain of urgency. Investigation Investigations for haematuria of unknown cause It is clearly important to screen the whole of the urinary tract for tumours. Pathology Urothelial tumours Urothelial tumours are derived from the transitional cell epithelium and are termed transitional cell carcinomas (TCCs). they may become muscle invasive.Bladder Table 18. ‘Carcinoma in situ’ is a natural precursor of highgrade cancers. because it will progress quickly to a muscleinvading cancer over a period of a few years. Types of bladder tumour Type Benign Examples Inflammatory plaques Transitional cell papilloma Inverted papilloma Leiomyoma Malignant primary Carcinomas Transitional cell carcinoma (90%) Squamous cell carcinoma (approx. and must be treated aggressively. where they are thought to originate from a persistent urachus. Adenocarcinomas are extremely rare. Death from cancer is relatively rare. 5%) Adenocarcinoma Mixed Small-cell carcinoma Others Sarcoma Lymphoma Malignant secondary Direct spread (e. Other cancers TCCs commonly contain components of squamous cell carcinoma. cytologically cells are similar to G3 TCC but without any evidence of invasion. minimal haematuria associated with a proven urinary tract infection in a female patient). including heavy smokers. Pure squamous cell cancers are rare in the absence of a predisposing factor for squamous metaplasia. classified by their behaviour: tumours with low or high malignant potential. Small-cell (oat cell) carcinoma is well recognised and should be managed in the same way as small-cell carcinoma of the lung. carcinoma in situ appears as flat erythematous patches and often leads to irritating urinary tract symptoms. and multiple tumours are frequent (up to 40%). As they grow larger. cervix. a co-existent cancer may be present. but it may be difficult to distinguish between a TCC with squamous differentiation and a true ‘pure’ squamous cell carcinoma. Histologically these tumours are high-grade cancers (G3). even of metastasising. Histologically the epithelium loses its orderly structure and may be thickened. They are occasionally seen in the dome of the bladder. Most urologists run rapid-access haematuria clinics.

uterus or vagina Pelvic or abdominal wall Nodes free of tumour Single node positive (< 2 cm) Single node positive (2–5 cm) or multiple nodes positive (< 5 cm) Node positive (> 5 cm) No distant metastases Distant metastases Further investigations if bladder tumour is confirmed The first stage in management is to perform transurethral resection (TURBT). a tumour may appear small but. Local staging should be completed by MRI if available because it is agreed that MRI scanning is better than CT for assessing the extent of the primary tumour. For example. Concordance across cystoscopic findings. Lung metastases may be the first radiological sign of metastatic disease. Urinalysis should be performed for cytology and culture. but sometimes the findings may not agree. and multiple or recurrent tumours have a higher . Ideally there should be some muscle in the biopsy. TURBT includes resection of all visible tumour (including muscle in the specimen if possible) and a separate resection biopsy from the border of the resected area and the tumour base.2. obturator. the tumour is confined to the bladder mucosa and tumour growth has not been detected beyond the lamina propria. 70% are superficial. and bone metastases may be found in around 5% of cases. PET scanning is of limited use because of interference from urinary excretion of contrast in the bladder. internal. MDT review is then particularly important for deciding the stage of the tumour. the pathology. 224 Adapted from UICC (2002).e. Treatment overview: superficial TCC of the bladder Of urothelial bladder tumours. and the radiological findings increases the confidence in the results of staging. A CT scan of the abdomen and pelvis is an alternative but this is not standard practice in the UK. using a resectoscope. CT scanning may give similar information about lymph node enlargement. Table 18. The risk of lymph node metastases is proportional to the depth of tumour invasion (i. TNM classification for carcinoma of the bladder Stage Tis Ta T1 T2a T2b T3a T3b T4a T4b N0 N1 N2 N3 M0 M1 Description Carcinoma in situ Noninvasive papillary tumour Subepithelial connective tissue Inner half of muscularis Outer half of muscularis Perivesical tissues (micro) Perivesical tissues (macro) Prostate. penetrating through the bladder wall into perivesical fat and adjacent organs. can be growing outside the bladder. and the pathologist will be able to comment on the presence or absence of muscle invasion. Staging Tumours originate in the bladder epithelium and infiltrate deeply into the muscle layers. 20% for lamina propria invasion. but caution is needed in attributing significant haematuria to a urinary tract infection. a large solid tumour may be apparently deeply invasive at operation but there may be no muscle invasion on the biopsy.2. staging investigations involve a CT scan of the chest and abdomen and a bone scan. Lymphatic spread is first to the hypogastric. However. and so care must be taken when comparing published results that use the old systems. usually under general anaesthetic. on imaging. external and common iliac lymph nodes. For patients being considered for radical treatment.Stephen Williams and Jim Barber abnormalities. In other cases. The TNM staging classification for carcinomas of the bladder is shown in Table 18. After a successful resection. 30% for superficial muscle invasion and 60% with full-thickness muscle invasion). The TURBT will give information about the grade of cancer and the depth of invasion. together with imaging of the upper urinary tracts by an ultrasound of kidneys and bladder and an IVU. there should be no mass to feel on EUA but the results of bimanual examination must be considered in the context of the other staging investigations. The staging system has been changed several times over the past few years. All superficial bladder tumours have a tendency to recur. A flexible cystoscopy is generally required to diagnose bladder tumours. that is.

2004). This trial showed little difference in the survival outcome between cystectomy and bladder preservation. outcomes similar to those after surgery have been reported following the use of neoadjuvant chemotherapy and chemoradiotherapy 225 High risk of progression (mostly TCCs of high malignant potential) pTa–pT1. albeit considerably more toxic (Shelley et al. but it contacts tumour cells by a fibronectin attachment protein which causes internalisation into the tumour cells. Treatment of invasive bladder cancer: management of the primary For patients with tumours that invade muscle (T2 and above) there are high recurrence and metastasis rates following TURBT alone.. frank . This strategy results in about 30% of patients being longterm survivors. 2005. long-term cancer control rates are only 90%.Bladder risk (Parmar et al. In the UK this approach would usually be reserved for those patients that fail BCG therapy. Although it is not standard therapy. followed by regular cystoscopic follow-up. but radiotherapy remains an option for patients who have failed intravesical therapy and who are unfit for cystectomy. Usually around six weekly treatments are given. Radiotherapy is used less today as a method of avoiding cystectomy in patients with recurrent pT1 G3 disease. 2005). G3. G1–2 A single dose of intravesical mitomycin C (20 mg) within 6 hours of resection significantly decreases recurrences (Sylvester et al. but the majority of patients with local failure develop metastases at the same time. On the other hand..5◦ C). 1989). and radical treatment with surgery or radiotherapy is advised. A randomised trial of radiotherapy (MRC BS06 study) failed to show any clear benefit for radiotherapy in this situation (Harland and the UK NCRI Bladder Clinical Studies Group. However. which would require antituberculous antimicrobial therapy. Radiotherapy fails to give local control of the tumour in up to half the patients.. BCG therapy BCG is an attenuated mycobacterium used for vaccinations. The exact mechanism of its effect is unknown. Although recurrence can be dealt with by a repeat TUR. If there are further high-grade recurrences or resistant CIS the risk of progression to muscle-invasive disease is high and early cystectomy should be considered. The aim of treatment is to deliver 10 000 000 organisms per instillation in 50 ml of normal saline 2 to 4 weeks post-TURBT. This approach was justified for many years because of the high mortality and morbidity rates for radical cystectomy and by the evidence from one small randomised trial from the Institute of Urology (London). Until the early 1990s. which means that salvage cystectomy is only needed for a few patients. adjuvant therapy should be considered to prevent the risk of progression to muscleinvasive disease and a potentially fatal cancer.5◦ C for > 24 hours or a pyrexia of > 39. in the UK. as follows.. a metaanalysis of this trial and three other similar trials from around the same period shows that this conclusion was flawed: the trials do indeed show a substantial and significant benefit for elective surgical management (Shelley et al. haematuria and bacterial infection are contraindications. systemic BCG infection can occur (pyrexia of > 38. a substantial proportion with their bladders intact. The solution should be retained in the bladder for about 2 hours. Low risk of progression (mostly TCCs of low malignant potential) pTa or pT1. Five-year survival rates of just less than 50% have been recorded even in multicentre studies. Immunosuppression. lasting several days. a policy of radical radiotherapy and surgical salvage was pursued. because some patients have metastases. Canada and elsewhere. CIS. Even with this approach. The side effects of BCG therapy are considerably worse than with mitomycin and include dysuria.. particularly if there is co-existent carcinoma in situ. cystectomy is sometimes performed for patients with T1 G3 disease. Shelley et al. but the best results reported recently have been achieved by using neoadjuvant chemotherapy combined with cystectomy and pelvic nodal clearance. multiple recurrences of G2 tumours Consider adjuvant intravesical BCG therapy (Harland and the UK NCRI Bladder Clinical Studies Group. Unfortunately. 2004) and is standard practice in the UK. 2001). urgency and frequency in most patients. In rare cases. 2003). this is a relatively arduous and toxic treatment and many patients presenting with bladder cancer are on the borderline of fitness for this approach. To what extent these data can be extrapolated to present-day practice is debated. There is evidence that intravesical BCG is somewhat more effective than mitomycin.

Patients particularly at risk are those with a large. a wider margin should be allowed around those areas. large tumours or tumours in the dome of the bladder (the bladder moves more superiorly). if nodal irradiation is considered. and so a stoma is still necessary. Disadvantages include nocturnal leakage and problems with voiding requiring intermittent self-catheterisation. a complete resection at TURBT.to 25-cm-long segment of ileum. Drawing a wider . or a complete response with chemotherapy have a good relative outcome after radiotherapy. On the other hand. There are three options for surgery: ileal conduit. Patients tolerate neoadjuvant chemotherapy poorly if their GFR is much below 50 ml/min. The pouch is emptied by inserting a soft silicone catheter into the stoma four to eight times a day. radiotherapy is more often given to elderly patients or those considered to be unfit for cystectomy or to patients with inoperable tumours. Important information about the site and spread of the tumour should be available both from radiological investigation (CT or MRI) and from the cystoscopy and EUA. However. continent urinary diversion or orthotopic neobladder. Patient fitness and attitude It is important to decide at an early stage whether patients are fit enough for a cystectomy and/or chemotherapy treatment. dilated rectum. no ureteric obstruction. such as the prostate and seminal vesicles in men and the uterus and adnexa in women. In many cases those with neobladders learn how to completely empty their bladders and do not require intermittent catheterisation. if it is found. The ileal conduit is made from a 15.Stephen Williams and Jim Barber regimens (Shipley et al. today. Some degree of hydronephrosis is common. The orthtropic neobladder. the patient empties the bladder by abdominal straining or clean intermittent catheterisation. also known as ‘continent orthotopic urinary reconstruction. The ureters are implanted into the closed end of the conduit. The reservoir is constructed from small bowel and is anastomosed to the top of the urethra. and the other end is brought to the surface of the abdomen to form a stoma. Hydronephrosis is often seen on CT scanning. the patient’s renal function must be carefully evaluated.. For continent urinary diversion. Removal of the urethra is indicated if the bladder neck is involved in women or the prostate is involved in men. patients often have co-existent cardiovascular and pulmonary disease and they may also have poor renal function due to obstructive uropathy. CT scanning should be used to simulate the patient with an empty bladder. the risks and benefits of surgery and chemotherapy should be weighed up carefully and the patient’s preference considered. skin tattoos are placed anteriorly over the pubic symphysis and laterally over the iliac crests to prevent lateral rotation. 2005). It appears that patients who respond well to initial neoadjuvant chemotherapy may be good candidates for bladder preservation with radiotherapy or chemoradiotherapy and a randomised study (SPARE) will be starting shortly in the UK to test this hypothesis. the patient should have a full bladder during the pelvic treatment so that the small bowel is pushed out of the field. Radical cystectomy Radical cystectomy consists of removal of the bladder and distal ureters and neighbouring organs. An ileal conduit is the quickest and simplest method but requires the patient to wear a urostomy bag. Target volume Standard practice is to treat the whole bladder with a 1 to 2 cm margin without making any attempt to treat the pelvic lymph nodes (see section on nodal treatment). however. the continent urinary pouch is made from approximately 2 feet of ileum. Applications Younger patients with small tumours. Standard margins may be inadequate to allow for bladder motion in many patients. 226 Setup With the patient in a supine position. 2005). but overall the complication rate is low. One end of the conduit is closed with stitches or staples. and whether they are prepared to have such an operation. Role of radical radiotherapy: a bladder preservation strategy The role of radiotherapy in bladder cancer has been reviewed (Logue and McBain. For tumours with an extravesical component. For patients with a GFR below 60 ml/min. this is a learned response and not guaranteed. Because bladder cancer is so strongly associated with smoking. also with 5-year survival at around 50%. The operative mortality of radical cystectomy has decreased substantially over the past 20 years and is about 3% in experienced centres.’ appeals especially to younger patients who wish to avoid a stoma on the abdomen and/or wearing a urostomy bag.

progression-free survival. Results comparable to cystectomy have been achieved with intensive bladder preserving regimens using neoadjuvant chemotherapy (gemcitabine and cisplatin) and concurrent chemoradiotherapy with cisplatin (Shipley et al. in which a combination of 5-fluorouracil and mitomycin C is used concurrently with radiotherapy. the effectiveness of concurrent chemoradiation is uncertain. The long term survival rates of patients who have microscopic nodal involvement (around 25% 5-year survival) support the use of nodal surgery.1 shows a radiotherapy plan for treating bladder cancer. but there are undoubtedly a few longterm survivors after radical cystectomy and nodal clearance. Figure 18.Bladder Isodose % 95 60 50 whether extending the lymph node dissection proximally is useful. Unfortunately. 2005). it is standard practice in the UK to avoid nodal irradiation. because doses are limited by the tolerance of the bladder itself.1. the whole bladder is included in a first phase followed by a boost to the tumour alone.. these regimens have also included pelvic nodal irradiation.’ For example. see p. Patients with good performance status and who show nodal enlargement on MRI can be considered for radical therapy with neoadjuvant chemotherapy followed by a cystectomy/node clearance if a good response is obtained. Combined modality treatment (chemoradiotherapy) Concurrent chemoradiotherapy may have a role in the treatment of bladder cancer (Sherwood et al. margin should be considered for these patients but is often impractical because of the size of the resulting fields. 231). The only other randomised trial of concurrent chemoradiation is the UK BC2001 study. This trial shows a 19% improvement in local control with concurrent cisplatin but no significant improvement in overall survival. because it allows a significant reduction in the amount of small bowel that is irradiated and may also reduce the rectal dose. radical therapy is not possible.. Palliative chemotherapy is considered if the patient is fit enough and.5 weeks or 55 Gy in 20 fractions over 4 weeks. 2005). Use of CRT/IMRT 3D planning with the use of conformal blocking or an MLC is strongly recommended. It is unlikely that macroscopic nodal disease can be cured with the relatively modest radiotherapy doses currently possible in this situation and radiotherapy is frequently palliative. because only one small prospective randomised controlled phase III trial has been published. IMRT may be applicable but has not yet been investigated. 1996). because the benefits are unclear. in the event of a good response. but it is not clear . Treatment of invasive bladder cancer: management of the nodes Clinically node negative Radical cystectomy generally includes elective dissection of the regional pelvic lymph nodes. For many patients. dose escalation is probably not practical with the standard whole-bladder technique. The regimen (based on the ‘Nigro’ anal cancer regimen) showed good results in a small 227 Beams One anterior and two posterior oblique wedged fields are used. A radiotherapy plan for treating bladder cancer. For patients treated with bladder preservation. For the nodal irradiation technique (not often performed) please refer to the prostate chapter (Chapter 19. depending on the bladder contour. in the BC2001 trial. Clinically node positive Figure 18. Before starting treatment a careful search for metastatic disease should be carried out. Doses Single-phase treatments involve 64 Gy in 32 fractions over 6. Most patients with macroscopic nodal disease relapse with metastases. A more sophisticated approach to irradiation of the bladder is needed in which a higher dose of radiation is given to the tumour with relative sparing of uninvolved areas: a move away from the ‘whole-bladder treatment. Importantly. is followed by consolidation radiotherapy treatment to the bladder and possibly nodal areas. despite the high risk of nodal disease.. However. or the distant metastasis rate (Coppin et al.

2005. 2003. Chemotherapy: neoadjuvant and adjuvant It is now clear from several randomised trials and two meta-analyses that the addition of neoadjuvant chemotherapy to either cystectomy or radiotherapy provides a modest overall survival benefit of around 5% (Advanced Bladder Cancer Meta-analysis Collaboration. its use should be encouraged. Adjuvant or neoadjuvant? Neoadjuvant chemotherapy is generally favoured because it is more practicable. For a variety of reasons. Sadly. some clinicians feel that the delay in definitive local treatment might be detrimental and. Many centres have adopted Gem-Cis as a standard regimen for palliative treatment. 1999. In fit patients with good renal function. McLaren. Despite this. Large-scale trials with novel agents have not yet been performed for bladder cancer.. 2001). 2003. because randomised data show that in metastatic disease Gem-Cis (cisplatin 80 mg/m2 day 1. surgery may be easier in patients who have a good response to chemotherapy. carboplatin can be substituted for cisplatin. Ureteric obstruction is common and stent insertion should be considered first if there is obstruction present. the toxicity of this treatment is at the limit of what can be tolerated for an adjuvant therapy for most patients. The recovery period after cystectomy can be very long and it may be weeks or months before it is possible to give chemotherapy. Patients with good performance status and renal clearance above 60 ml/min tolerate chemotherapy 228 . better. It is also an option for elderly or unfit patients with clinically localised disease. 2004). despite the lack of evidence. 2005. for the majority. The same regimens are in use for palliative and adjuvant chemotherapy.. in patients with poor renal function. gemcitabine 1000 mg/m2 day 1 and 8) provides a rate of survival similar to that of CMV but with considerably less toxicity (von der Maase et al. Chemotherapy regimens Combination regimens containing cisplatin appear more effective than single-agent cisplatin.. International Collaboration of Trialists. 2000). the benefits of adjuvant chemotherapy are unlikely to be quantified in the near future because a meta-analysis of adjuvant chemotherapy trials was inconclusive. 1996). Although this outcome is similar to the survival benefit from adjuvant therapy in other cancers.. Unfortunately. particularly when delivered with a radiotherapy plan. The target volume may vary widely among patients and so the technique must be determined on a caseby-case basis. and consideration should be given to prophylactic anticoagulation. Alternatively. Many patients are elderly or in relatively poor general health and so the decision to give chemotherapy needs a careful clinical assessment and time to discuss the pros and cons of the treatment with the patient. because of the small numbers of patients and an EORTC trial answering this question is recruiting poorly. Treatment limited to the bladder only may improve urinary symptoms and has the least toxicity. Accelerated MVAC (given with GCSF support) is considered an alternative to Gem-Cis and has similar toxicity (Sternberg et al. Its use is also increasing in the adjuvant setting.. Duchesne et al. median survival is around 12 months. and there is a consensus that CMV and MVAC have similar efficacy and toxicity. Both of these regimens are decreasing in use. they favour an adjuvant approach. Moreover. Phase II trials of combination regimens including taxanes with cisplatin have shown promising results but randomised data are awaited. the results of palliative chemotherapy are disappointing. but response rates are consistently inferior (Petrioli et al. however.. Treatment to the whole pelvis with parallel fields Palliative chemotherapy A few patients with a relatively slow-growing TCC may achieve long remissions (sometimes multiple remissions) with combination chemotherapy treatment. therefore. the incidence of thromboembolic disease in these patients is extremely high.Stephen Williams and Jim Barber phase II study but it will be several years before the results of BC2001 are available. Winquist et al. A randomised trial has shown similar palliation rates with a short. Palliative radiotherapy Palliative radiotherapy may be highly effective at temporarily relieving haematuria and urinary symptoms in locally advanced and metastatic bladder cancer. hypofractionated treatment (21 Gy in 3 fractions) compared to a higher-dose palliative treatment (35 Gy in 10 fractions. 2000). Grossman et al.

859–66. Duchesne. Oncol. radical cystectomy in patients with pT3-pT4 and/or N+ M0 TCC of the bladder. Freedman.. Neoadjuvant chemotherapy in invasive bladder cancer: a systemic review and meta-analysis. (2004). (2000). Current trials See www.uk/ for information regarding ongoing trials. such as the bladder only. 4505. J. Lancet.. Frediani. Comparison between a cisplatin-containing regimen and a carboplatincontaining regimen for recurrent or metastatic bladder cancer patients. Lancet. M. low dose hypofractionated treatment is effective (e. Alonzi. A. 17. et al. Oncol. 17. 47. Coll. Griffiths.. r The role of adjuvant chemotherapy. Oncol. 387–98. B. Coppin. 23. BCON is a multicentre randomised trial of radical radiotherapy with carbogen in locally advanced bladder cancer.ncrn. REFERENCES Advanced Bladder Cancer Meta-analysis Collaboration. J. A randomized trial of hypo-fractionated schedules of palliative radiotherapy in the management of bladder carcinoma: results of Medical Research Council trial BA09.. r The use of novel drugs such as anti-EGFR agents (Alonzi and Hoskin. Natale. prospective. .org. Grossman. (2005). 349. 14. 284–8. McLaren. M. r The use of imaging to overcome the effects of bladder movement during radiotherapy (Muren et al. Neoadjuvant chemotherapy in transitional-cell carcinoma of the bladder. Prognostic factors for recurrence and follow-up policies in the treatment of superficial bladder cancer: report from the British Medical Research Council Subgroup on Superficial Bladder Cancer (Urological Cancer working party). Med. 354. J.Bladder permits the inclusion of pelvic lymph nodes but can lead to severe rectal toxicity. Cancer. Cystograms are rarely used today because of easy access to CT planning and/or simulation. Urol. Eur.g.. Clin. J. A randomised trial of radical radiotherapy in pT1G3 NXM0 bladder cancer (MRC BS06). r Dose escalation in radiotherapy. M. J. Even a single fraction of 10 Gy may provide relief from haematuria in a patient with poor performance status. 73. B..). 503–7. (2005). Novel therapies in bladder cancer. and Hoskin. G. et al. and Dahl. 344–51. (1999). C. BC2001 is a 2 × 2 factorial randomised phase III study comparing standard versus whole-bladder radiotherapy with tumour boost with or without synchronous chemotherapy in muscle-invasive bladder cancer. Advanced Bladder Cancer Meta-analysis Collaboration. G.. J. 77. B. D. (R. Urol. Oncol. (1989). Petrioli. S. 21 Gy in 3 fractions over 1 week). methotrexate. Clin. L. M. A dose/fractionation schedule of 35 Gy in 10 fractions over 2 weeks or 30 to 36 Gy in 5 to 6 fractions weekly can be delivered using a CT plan but this dose would be considered on the borderline of tolerance if parallel fields are used. 508–13. C. et al. and the UK NCRI Bladder Clinical Studies Group. J. Improved local control of invasive bladde