Professional Documents
Culture Documents
Paris System
for Reporting
Urinary Cytology
Eva M. Wojcik
Daniel F.I. Kurtycz
Dorothy L. Rosenthal
Editors
Second Edition
123
The Paris System for Reporting Urinary
Cytology
Eva M. Wojcik • Daniel F.I. Kurtycz
Dorothy L. Rosenthal
Editors
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
This book is dedicated to the memory of
Stefan Pambuccian, MD (1957–2020), one of
the original members of The Paris Group.
From the beginning, Stefan was an integral
part of this project. His expertise and
commitment were always the guiding
principles for all of us to emulate.
Dr. Pambuccian was one of the greatest
pathologists our profession has ever
produced. He was a teacher and researcher
of the highest order. His work was
fundamental for The Paris System (TPS) and
his thoughtful, meticulous approach laid the
ground for the success of TPS.
Stefan was truly a gentle man. Despite his
monumental expertise, his humility and
dedication to work were exemplary. Stefan
left behind an army of his grateful mentees
who have been carrying on his legacy.
Preface
Almost a decade ago, members of two panels started preparing for the International
Congress of Cytology that took place in Paris in 2013. We all had the feeling that
maybe ours would be the last symposia on urine cytology. At that time, the rate of
atypia was raging and credibility with our clinical colleagues was dwindling. Urine
cytology was on the brink of losing clinical relevance, joining the fate of prostate
FNA and breast FNA, at least in the United States. Moreover, for many years, urine
cytology was one of the most frustrating types of cytology specimens, hated by
most cytology practitioners. Whatever we were doing, there was a very high prob-
ability that we were doing it wrong, at least in the eyes of our clinical colleagues.
They would have seen papillary lesions during cystoscopy, but cytology reports
were negative. On the other hand, they could not see any cystoscopic lesions and
cytology reports were positive. Of course, we knew why this happened. Low-grade
carcinomas do not exhibit enough morphologic changes to be distinguished from
normal and reactive urothelium, while flat, high-grade lesions (CIS, carcinoma in
situ), although cytologically detectable, are very difficult to visualize on cystoscopy.
Definitely, it was time for action to save urines!
Once we met in Paris, it was apparent that we were committed and ready to fix
the problem. That is how the original “Paris group” formed. During our first
impromptu meeting, we decided to start working on a standardized reporting system
that would be based on consensus and evidence. Of significance, during the very
first meeting we agreed that the new reporting system should concentrate on the
detection of a clinically significant disease, high-grade malignancy. That was a true
paradigm shift in urine cytology!
The rest is history! However, it did not happen overnight. It took a large group
of committed cytopathologists, surgical pathologists, and urologists who worked
very hard together to create, in record time, a new reporting system that was pub-
lished in 2016. We always knew that we had a good product, the system that we
truly believed in; however, the success of The Paris System (TPS) exceeded every-
one’s wildest imagination. Over the past 5 years, The Paris System has been trans-
lated into Japanese, Russian, and Chinese, and has been accepted throughout the
world as a standard of care. Our credibility in the eyes of clinicians has significantly
improved and, most importantly, now we are able to provide better patient care,
accurately screening for new or recurrent life-threatening high-grade bladder cancer.
vii
viii Preface
From the beginning, we realized that the first edition would only be a start. We
needed real data to confirm that the criteria of The Paris System are appropriate.
Most importantly, we needed prospective studies, based on TPS, to establish rates
of malignancy for each diagnostic category to guide our clinical colleagues.
This second edition fulfills these gaps. After the publication of TPS, a tsunami of
papers from all over the world confirmed the reliability of our established criteria.
Most significantly, the main goal of The Paris System was accomplished: lowering
the rates of the indeterminate categories, particularly diagnoses of “atypia.”
Moreover, after TPS had been published, we quickly realized that there were issues
not sufficiently covered in the first edition, particularly degeneration, subtypes of
high-grade urothelial carcinomas, and squamous dysplasias. TPS 2.0 extensively
discusses these topics. Although the overall concept of the first edition remains the
same, the second one is significantly expanded and much more richly illustrated.
With this continuously evolving project, we are confident that urine cytology is here
to stay. We hope you will agree!
When The Paris System for Reporting Urinary Cytology (TPS) was introduced in 2016,
the new paradigm needed guidelines. Over the past 5 years, experience has provided
evidence that the proposed criteria are effective. As aids for microscopists when dealing
with a cytology slide, the authors of TPS 2.0 present the following in this section:
1. Decision Tree emphasizing changes in N/C ratios as morphologic find-
ings worsen.
The Paris System Approach to Diagnosis in Urinary Cytology
Is there increased
No N/C ratio ( 0.5)? Yes
(i.e., not intermediate
and basal urothelial cells)
Are any one of the following
features present in the atypical
cells?
Is there a reason for the 1. Hyperchromasia
2. Coarse Chromatin
“atypia”? 3. Irregular chromatinic rim
Polyomavirus?
Granuloma? At least 2 features
present and High (~≥ 0.7) N/C ratio
IIeal conduit?
Urolithiasis?
At least 1 feature How many cells with these
Reactive changes? present features?
Graphic algorithm of the Paris System for Reporting Urinary Cytology decision tree.
The system emphasis is on the detection of High-Grade Urothelial Carcinoma (HGUC).
This snapshot conceptual flowchart illustrates the major points in the decision tree including
evaluation of the N/C ratio, nuclear features, and cell quantity. Ref: We’ll Always Have Paris:
The Paris System for Reporting Urinary Cytology 2022, Eva M. Wojcik, Daniel F.I. Kurtycz,
Dorothy L. Rosenthal. J Am Soc Cytopathol. online preprint, DOI: https://doi.org/1.01016/
jasc.2021.12.03
ix
x Introduction
N:C
0.3
0.4
0.5
0.6
0.7
0.8
Negative for High Benign urothelial, glandular, squamous cells, benign tissue 70 - 90% 8 - 24%
Grade Urothelial fragments, changes due to instrumentation, lithiasis,
Carcinoma (NHGUC) polyoma virus, therapy. Low Grade Urothelial Neoplasm
(LGUN)
Atypical Urothelial Required – increased N/C ratio ( > 0.5) and one of: 5 - 15% 24 - 53%
Cells (AUC) Hyperchromasia, Irregular clumpy chromatin or Irregular
nuclear contours
Suspicious for High Required – Few cells (< 5-10) with high N/C ratio (> 0.7) and 0.5 - 3% 59 - 94%
Grade Urothelial hyperchromasia, and/or Irregular clumpy chromatin,
Carcinoma (SHGUC) Irregular nuclear contours
Positive for High Required – Many cells (>10) with high N/C ratio (> 0.7) and 0.1 - 3% 76 - 100%
Grade Urothelial hyperchromasia, Irregular clumpy chromatin, Irregular
Carcinoma (HGUC) nuclear contours
The authors of TPS will continue to monitor the utility of these criteria and will
rely upon our end users to publish their experience with TPS.
Abbreviations
AdCa Adenocarcinoma
AMACR Alpha-methylacyl-coA racemase
ASC American Society of Cytopathology
AUC Atypical Urothelial Cells
AUTF Atypical urothelial tissue fragments
BCG Bacillus Calmette-Guerin
BUTF Benign urothelial tissue fragments
CAP College of American Pathologists
CEP Chromosome enumeration probes
CIS Carcinoma in situ
CS Cytospin™
DAPI 4,6-Diamidino, 2-phenylindole dihydrochloride
DS Direct smear
ERG ETS-related gene
FDA U.S. Food and Drug Administration
Fig. Figure
FISH Fluorescence in situ hybridization
H&E Hematoxylin and eosin stain
HGUC High-grade urothelial carcinoma
HPV Human papillomavirus
IAC International Academy of Cytology
ISUP International Society of Urological Pathology
LBP Liquid-based preparations
LCNEC Large cell neuroendocrine carcinoma
LGPUC Low-grade papillary urothelial carcinoma
LGPUN Low-grade papillary urothelial neoplasm
LGUC Low-grade urothelial carcinoma
LGUN Low-grade urothelial neoplasia
LIS Laboratory Information System
LSI Locus-specific identifier
mag. Magnification
N/C Nuclear-to-cytoplasmic ratio
NE Neuroendocrine
NEC Neuroendocrine carcinoma
xiii
xiv Abbreviations
xv
xvi Contents
Editors
Authors
xvii
xviii Contributors
Bas WG van Rhijn, MD, PhD, FEBU Department of Surgical Oncology
(Urology), Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital,
Amsterdam, The Netherlands
Theodorus H. Van der Kwast, MD, PhD Princess Margaret Cancer Center,
University Health Network, Toronto, ON, Canada
Philippe Vielh, MD, PhD, FIAC Department of Pathology, Medipath & American
Hospital of Paris, Paris, France
Poonam Vohra, MD Department of Pathology, University of California, San
Francisco (UCSF), San Francisco, CA, USA
Yeh-Han Wang, MD, MSS ACT Genomics Co., Ltd., Taipei, Taiwan
Mingjuan Lisa Zhang, MD Department of Pathology/Cytopathology,
Massachusetts General Hospital, Boston, MA, USA
Pathogenesis of Urothelial Carcinoma
1
Kaitlin E. Sundling, Tatjana Antic,
and Stefan E. Pambuccian
Background
For many years it has been known that urothelial carcinoma has two distinct patho-
genetic pathways, a hyperplasia (low-grade urothelial neoplasms) and a dysplasia
(high-grade urothelial carcinoma) pathway [5–7]. More recently, The Cancer
Genome Atlas (TCGA) Research Network on molecular characterization of urothe-
lial bladder carcinoma revealed 302 mutations, 204 segmental copy number altera-
tions, and on average 22 rearrangements per tumor [8]. Thirty two genes showed
recurrent driver mutations. TCGA recognizes four molecular clusters based on dif-
ferential gene expression by RNA sequencing. TCGA also presents a pathway anal-
ysis based on mutation and copy number alterations (Fig. 1.1). It is not surprising,
given the rich and highly complex biological datasets studied, that the cluster and
pathway groupings have significant overlap. The Cluster I of TCGA corresponds to
commonly referred “luminal-type” in previously proposed systems [8]. The pre-
dominant pathway in “luminal-type” low-grade, non-muscle-invasive tumors is the
FGFR3/RAF/RAS signaling pathway with activating mutations in FGFR3 [9].
These tumors may progress from a precursor urothelial proliferation of unknown
significance (previously known as urothelial hyperplasia) with deletion of the gene
cyclin-dependent kinase inhibitor 2A (CDKN2A), located on the short arm of chro-
mosome 9, which encodes the p16INK4A protein. Papillary architecture is predomi-
nant in tumors with FGFR alterations. While these low-grade tumors uncommonly
progress to invasive carcinoma [10], if they do, they retain the molecular signature
involving FGFR3 alterations (about 20% of muscle-invasive bladder cancers). A
smaller group of low-grade urothelial carcinomas showed HRAS/KRAS/NRAS
mutations that are mutually exclusive with FGFR3.
The Cluster II of the TCGA is often referred to as “p53-like” and probably rep-
resents the less common dysplasia pathway. Initial dysplastic lesions harbor TP53
mutations and often progress to flat carcinoma in situ lesions, with concomitant RB
mutations. Lesions on this pathway have a high potential for muscle invasion, lymph
node metastases, and systemic metastases. This pathway has been found in 93% of
muscle-invasive carcinomas [8].
What is of significance is that the key molecular abnormalities associated with
the dysplasia pathway, especially the TP53 mutations, which are strongly associated
with high-grade and high-stage urothelial carcinomas, rarely co-occur with the
molecular abnormalities characterizing the FGFR3 pathway [8].
1 Pathogenesis of Urothelial Carcinoma 3
Normal
N orma
ormall Urothelium
ma Uro
roth
oth
the
thel
elliu
liu
um
Prec
Precursors
Prec
Precur
cur
urso
s rs
s
Hyperplasia Dysplasia
Fig. 1.1 Schematic representation of the major histologic findings in urothelial carcinogenesis.
Key genetic alterations are noted in the noninvasive carcinomas, while pathway alterations from
TCGA analysis are noted in muscle-invasive urothelial carcinomas. Note that there is overlap
among the pathway and cluster designations, and the groupings are not mutually exclusive
4 K. E. Sundling et al.
The Clusters III and IV, which are referred to as “basal,” are tumors enriched in
squamous and sarcomatoid features that often have metastatic disease at presenta-
tion and express high levels of EGFR. About 12% of tumors in this cluster have
activating mutations in ERBB2 (Her2) [8].
The underlying pathways comprise four main groups, with significant overlap.
The RTK/RAS/PI3K pathway is altered in about 72% of urothelial carcinomas,
which contains a subset of tumors with FGFR3 alterations and/or ERBB2 enrich-
ment. The histone modification pathway includes changes in genes effecting meth-
ylation and acetylation and is altered in about 89% of urothelial carcinomas. The
SWI/SNF complex pathway involves nucleosome remodeling genes and is altered in
about 64% of urothelial carcinomas. Finally, the TP53/RB pathway includes many
well-known cell cycle regulatory genes and is altered in 93% of urothelial carci-
nomas [8].
The guiding principle for TPS is to detect HGUC. In line with the proven
strengths of urinary cytology, the negative category includes reactive changes,
infectious and nonneoplastic conditions, as well as cases that may have some cyto-
logic features of LGUN but are negative for HGUC. Therefore, the proposed diag-
nostic category is “Negative for High-Grade Urothelial Carcinoma” (NHGUC).
Despite the fact that we strive to detect all HGUC, we recognize that there will be
cases where a definite diagnosis cannot be made. Therefore, in TPS we include the
categories of “Atypical Urothelial Cells” (AUC) and “Suspicious for High-Grade
Urothelial Carcinoma” (SHGUC). Although the diagnosis of LGUC is not the main
goal of this system, a diagnostic option has been included to define those circum-
stances where cytologic features of HGUC are absent and LGUN are present (see
Chap. 3).
References
1. Sundling KE, Kurtycz DFI. Standardized terminology systems in cytopathology. Diagn
Cytopathol. 2019;47(1):53–63. https://doi.org/10.1002/dc.24103.
2. Lammers RJM, Witjes WPJ, Hendricksen K, Caris CTM, Janzing-Pastors MHC, Witjes
JA. Smoking status is a risk factor for recurrence after transurethral resection of non-
muscle- invasive bladder cancer. Eur Urol. 2011;60(4):713–20. https://doi.org/10.1016/j.
eururo.2011.07.010.
3. Koutros S, Rao N, Moore LE, et al. Targeted deep sequencing of bladder tumors reveals novel
associations between cancer gene mutations and mutational signatures with major risk fac-
tors. Clin Cancer Res. Published online April 13,. 2021; https://doi.org/10.1158/1078-0432.
CCR-20-4419.
4. Fischer EG. Nuclear morphology and the biology of cancer cells. Acta Cytol. 2020;64(6):511–9.
https://doi.org/10.1159/000508780.
5. Koss LG. Bladder cancer from a perspective of 40 years. J Cell Biochem Suppl. 1992;16I:23–9.
https://doi.org/10.1002/jcb.240501305.
6. Spruck CH, Ohneseit PF, Gonzalez-Zulueta M, et al. Two molecular pathways to transitional
cell carcinoma of the bladder. Cancer Res. 1994;54(3):784–8.
7. Lopez-Beltran A, Cimadamore A, Montironi R, Cheng L. Molecular pathology of urothelial
carcinoma. Hum Pathol. 2021;113:67–83. https://doi.org/10.1016/j.humpath.2021.04.001.
1 Pathogenesis of Urothelial Carcinoma 5
8. The Cancer Genome Atlas Research Network, Analysis Working Group: The University of
Texas MD Anderson Cancer Center. Comprehensive molecular characterization of urothelial
bladder carcinoma. Nature. 2014;507(7492):315–22. https://doi.org/10.1038/nature12965.
9. Kamoun A, de Reyniès A, Allory Y, et al. A consensus molecular classification of
muscle-invasive bladder cancer. Eur Urol. 2020;77(4):420–33. https://doi.org/10.1016/j.
eururo.2019.09.006.
10. Vollmer RT. A review of outcomes for stage Ta bladder tumors. Am J Clin Pathol.
2016;146(2):215–20. https://doi.org/10.1093/ajcp/aqw103.
Adequacy of Urine Specimens (Adequacy)
2
Z. Laura Tabatabai, Güliz A. Barkan,
Monique Courtade-Saïdi, Daniel F.I. Kurtycz,
Matthew T. Olson, Toyonori Tsuzuki,
Christopher J. VandenBussche, and Poonam Vohra
Background
collection policies can only be enforced partially. Similarly, the suggestion that the
void occur later than the first void of the morning is a detail not commonly covered
in lab collection systems. Other details including the physical and sexual activity
of the patient immediately prior to specimen collection are useful pieces of infor-
mation that often go unrecorded and therefore only partially appreciated in the
medical literature. The recommendations for adequacy evaluation per TPS stem
from the recognition that all of these variables are important and must be consid-
ered in the context of one another to properly address the questions pertaining to
adequacy.
Adequacy is an essential consideration for cytopathologists. It is worthwhile
repeating the obvious that diagnostic accuracy improves with adequacy.
Unfortunately, due to variable practices that existed before the formulation of
TPS, urine cytology specimens were commonly perceived to only infrequently
yield valuable clinical information. This perception is based on a reasonable
assumption that the limited sampling of cells exfoliating from the urothelial lin-
ing may not reflect the true biologic state of the patient, whether or not neopla-
sia exists.
To provide a quantitative context for sampling the urinary tract, the average
maximally distended healthy human bladder has an approximate volume of
600 mL. Given that a bladder has a roughly spherical shape, the inner surface
area of that bladder would be approximately 350 cm2 (350 × 10−4 m2). The aver-
age urothelial cell has an approximate diameter of 20 μm or a two-dimensional
surface area of 314 μm2 (314 × 10−12 m2), and the urothelium is approximately
five cells thick. Therefore, the total number of urothelial cells lining the bladder
can be estimated to be on the order of 108–109 cells. However, even a highly cel-
lular urine specimen contains only a minute fraction of the urothelial cells
derived from the urinary lining; the chance that a specimen will contain cells
from a clinically significant lesion is therefore a probabilistic event. The larger
the lesion, the greater the involvement of the urothelial surface; the more ana-
plastic and damaged the cellular adhesion structures, the more likely that diag-
nostic cells will be found in the sample. Alternatively, in the absence of disease,
a voided urine will at times contain so few cells that it can be technically consid-
ered “Inadequate for Diagnosis.” However, if the patient has no risk factors and
the volume is adequate, then a diagnosis of “Negative for High Grade Urothelial
Carcinoma (NHGUC)” may be more appropriate, with a limitation statement of
low cellularity.
A false-negative result can be the byproduct of a limited sampling size. Variation
in cellularity among urinary tract samples creates a desire for determining the mini-
mum number of benign cells available for examination that will provide confidence
that the remaining urothelium is benign as well.
2 Adequacy of Urine Specimens (Adequacy) 9
For the purposes of this chapter, the term “adequacy” refers to the usefulness of the
specimen to diagnose or raise suspicion for HGUC. As such, a specimen collected
for the diagnosis of acute bacterial cystitis that shows only neutrophils is inadequate
to detect urothelial carcinoma despite being perfectly useful to answer other specific
non-oncologic clinical questions. Adequacy of urine specimens for the diagnosis of
urothelial carcinoma is determined by the interplay of four specimen characteris-
tics: collection type, cellularity, volume, and cytomorphological findings. Of these
characteristics, cytomorphological findings should be considered first because the
presence of any atypical, suspicious, or malignant findings makes a specimen intrin-
sically adequate regardless of the collection type, cellularity, or volume. Thus, a
decision about specimen adequacy is most relevant when there are no findings
indicative of a disease process. If the cytomorphological findings are negative for
any markers of disease process, the other specimen characteristics should then be
factored into the classification of adequacy.
Since the publication of the first edition of TPS, studies on the role and specific
qualifiers of collection type, cellularity, and volume have provided additional
insights but remain limited, and the opinions expressed are variable. Therefore,
adequacy recommendations in TPS continue to be based on the adequacy algorithm
shown in Fig. 2.1. The purpose of this decision tree is threefold. The first is to com-
municate the relationship that exists among collection type, cellularity, and volume.
The second is to guide individual laboratories in validating appropriate choices for
their own practice settings at each branch in the adequacy algorithm. Finally, the
ultimate purpose is to provide a framework for future investigations dealing with
adequacy of urine specimens whereby each decision point will have clear, evidence-
based, and consensus-determined cutoff values for volume and cellularity in the
context of the collection method.
Due to the incomplete nature of the evidence to determine urine specimen ade-
quacy, the current algorithm does not take into consideration which method is used
for concentrating and preparing the urinary tract specimens. However, the generic
approach to adequacy determination should be uniform regardless of the method
used by a laboratory to prepare a specimen. As increasing data and evidence become
available in the literature, preparation-dependent cutoffs at each decision point in
the algorithm may become more apparent.
10 Z. L. Tabatabai et al.
Volume and Adequacy
Atypical,
NO Suspicious YES
or
Malignant
NO
NO
NO
Adequate Volume
YES
Inadequate Adequate
Fig. 2.1 The adequacy algorithm shows The Paris System’s recommendation for the proper rela-
tionship between specimen source, cytological diagnosis, urine volume, urothelial cellularity, and
obscuring features. Obscuring features include non-urothelial cells such as lubricant, acute inflam-
mation, bacteria, vaginal contaminants, sperm, and crystals which, when copious, may obscure the
finding and characterization of urothelial cells. “*”: Cutoffs for appropriate benign urothelial cel-
lularity should be validated for both instrumented and non-instrumented sources
volume received by a laboratory can result from specimen manipulation at some point
in the collection, e.g., splitting the sample among diagnostic laboratories or discarding
some to fit into an available container.
Urine volume is logically an adequacy criterion, but data and studies remain
sparse. Assessing the role of urine volume as an adequacy criterion in clinical
practice remains challenging for the following reasons: the volume of urine
received in the cytology laboratory may not always be accurately documented,
12 Z. L. Tabatabai et al.
6
Malignant
5 Suspicious for Malignancy
% Malignant or Suspicious
4
3
2
1
0
Fig. 2.2 Relationship of volume to the prevalence of malignant and suspicious diagnosis for
demographically comparable populations. There were no suspicious samples in the lowest volume
bin, and there was a nearly linear increase in the suspicious fraction up to the bin centered on
15 mL. A specimen was nearly twofold more likely to be suspicious when more than 15 mL was
received than when less than 5 mL was received. The global maximum for the malignancy fraction
was seen at 27.5 mL (range: 25–30 mL) which was also the location of a local maximum for the
suspicious fraction. The global maximum for the suspicious curve was centered at 85 mL, and the
difference between the global maximum (5.8, 95% CI: 5.6–5.9) and the local maximum at 27.5 mL
(5.7, 95% CI: 5.5–5.9) was not statistically significant. Based on this analysis, we concluded that
at least 30 mL is necessary to consider a urine fully adequate when processed with SurePath.
(From VandenBussche et al. [4] with kind permission of John Wiley & Sons)
and even if it is accurately documented, it may not reflect the true volume of the
original specimen. A variety of urine specimen types may be received including
voided, washing, and upper tract samples; the addition of a fixative to the sample
may be uncertain, even though it should be a regular part of each laboratory sam-
ple submission document; and the accurate amount of the added fixative is rarely
provided to the laboratory [9].
A recent study suggested that a minimum cutoff of 30 mL is appropriate in a
laboratory that uses SurePath® preparation performed on fresh, unfixed voided
urines [4]. A later, similar study recommended a minimum cutoff of 25 mL for
ThinPrep (Hologic, Inc., MA) preparations, also performed on fresh, unfixed voided
urines. Of note, in the second study, the cytology results were not confirmed by a
biopsy or surgical excision [5]. Yet another recent study justified urine volume as an
adequacy criterion in voided urine and reported that although 10 mL of fresh voided
2 Adequacy of Urine Specimens (Adequacy) 13
Cellularity and Adequacy
Instrumentation of the urinary tract results in cellular samples in which cells and
tissue fragments are forcibly exfoliated from the urothelium; these include bladder
washings (also called “barbotage” specimens). Other instrumented urinary speci-
mens include urine collected after cystoscopy, radiologic procedures, catheteriza-
tion, washings and brushings from the urethra, ureters, and renal pelvises. Of these,
the specimen type most commonly processed is bladder washing, which is often
part of a cystoscopic procedure [11].
The cellularity of instrumented urinary tract specimens obtained via cystoscopy
is affected by various technical factors including those related to the proceduralist’s
skill, method of performing the washing, amount and type of fluid with which the
bladder is washed, and the distance of the cystoscope to the mucosa. Although the
laboratory cannot control most of these factors, setting adequacy criteria will alert
clinicians when the specimen has insufficient cellularity and may be more prone to
a false-negative diagnosis. One factor that the laboratory can influence is the type of
irrigation fluid used. See Chap. 10 (Cytopreparatory Techniques) for a discussion of
optimal cellular preservation during washings.
14 Z. L. Tabatabai et al.
Compared to voided urine specimens, bladder and upper tract washing speci-
mens have a volume that is dependent on the amount of fluid used during the proce-
dure, generally lack contamination by cells originating outside of the urinary tract,
and should have a higher cellularity. In addition, such samples are more often
received fresh or admixed with fixatives. These features may contribute to the higher
sensitivity of bladder washing specimens [6, 12] but also suggest the need for
volume-independent adequacy criteria. A recent study suggested that all instru-
mented specimens should be considered adequate because it is unlikely that the
invasive procedure necessary for a repeat sample will be based only on nondiagnos-
tic cytology [6]. However, since the decision to repeat is based on the overall clini-
cal findings, the urologist has to know that the procedure did not yield diagnostic
material.
Another recent study suggested that for instrumented urine specimens, cases
with no urothelial cells should be considered inadequate, whereas cases with <20
cells per 10 high-power fields (HPF) should be considered as “limited” because they
have a significantly lower sensitivity than samples with higher cell counts. The sen-
sitivities reported for Cytospin preparations and ThinPrep preparations are similar,
suggesting that cell density rather than total cells may be the preferred measure of
cellularity in the setting of urinary cytology [10].
Quantitative adequacy criteria for specimen cellularity have been established
for Pap tests based on evidence obtained by studies using serial dilutions and for
thyroid aspirates based on retrospective review of cytology specimens with histo-
logic follow-up [13, 14]. For urinary cytology, an arbitrary cutoff value of 15
well-preserved and well-visualized urothelial cells has been proposed by some
authors [15, 16].
To date, only a single study has applied both the serial dilution and surgical exci-
sion methods to bladder washing specimens to establish the cell count adequacy
criteria for specimens processed using the ThinPrep method [17]. In this study, an
adequate bladder wash specimen was determined to have at least 20 well-preserved,
well-visualized urothelial cells per 10 HPF. This cutoff was only viable in the
absence of obscuring features. These results were further supported by a more
recent study performed on Cytospin preparations of instrumented urine cytology
specimens which showed that less than 10 urothelial cells per 10 HPF is associated
with significantly increased false-negative rates compared to cases with 10–49 uro-
thelial cells per 10 HPF [10]. Interestingly, this study also noted that cases with 50
or more urothelial cells per 10 HPF also had an increased false-negative rate. In
these high cellularity cases, the neoplastic cells may be diluted by the many benign
urothelial cells in the specimen. In such cases, it is not known whether repeat testing
would resolve this problem.
Another recent study, using voided urine samples from patients who were either
positive or negative for HGUC, compared ThinPrep and Cytospin processing
devices. Specimens from a total of ten individuals positive for HGUC and nine
individuals negative for HGUC were collected and split to compare the two prepara-
tion methods. This study showed that the morphological features of HGUC and
normal cells were comparable in specimens prepared using either the ThinPrep or
2 Adequacy of Urine Specimens (Adequacy) 15
Cytospin systems. The authors concluded that TPS can be applied reliably in diag-
nostic laboratories that use either of these methods [18].
These studies support the observations that instrumented urine cytology cases
where there are 10–20 well-preserved, well-visualized urothelial cells per 10 HPF
are “satisfactory but limited by low cellularity” and cases where there are less than
10 well-preserved, well-visualized urothelial cells per 10 HPF are “unsatisfactory/
nondiagnostic.” Given that these thresholds were determined using ThinPrep and
Cytospin preparations, additional studies are required to determine the thresholds
for other specimen preparation methods. Similarly, since the majority of studies
have been performed on bladder washing specimens, the adequacy criteria for upper
tract and voided specimens are not as rigorously defined. The presence of excessive
lubricant, inflammatory cells, or red blood cells, either as the sole finding or obscur-
ing the urothelial cells, must be interpreted with caution (Fig. 2.3) (see sample
reports). As further data become available for upper tract and voided specimens, the
criteria will reflect the evidence, and this evidence will provide hard quantitative
data to the adequacy algorithm.
a b
Fig. 2.3 Unsatisfactory/nondiagnostic specimen. (a) Acellular specimen showing only lubricant
without the presence of urothelial cells (Bladder wash, TP, medium mag.). (b) Scantly cellular
specimen with obscuring blood (voided urine, TP, medium mag.). (c) Scantly cellular specimen
with obscuring inflammation in a patient with a papillary lesion observed on cystoscopy (bladder
wash, TP, medium mag.)
16 Z. L. Tabatabai et al.
The most uncertain feature of the adequacy algorithm is the classification of voided
urine specimens that have an insufficient or low number of benign urothelial cells but
have an adequate volume. There is currently no evidence that demonstrates the require-
ment for benign urothelial cells in voided urine specimens. In many practices, speci-
mens that meet every adequacy criterion except for urothelial cellularity are assigned a
“less-than-optimal” adequacy. However, there is not a precise cellular metric for when
a voided urine specimen should be classified as “nondiagnostic” or “less than optimal”
[6]. The sensitivity of low cellularity specimens is still relatively high, and in these
cases, the cytology specimen is often one part of an examination that includes clinical
observation and biopsy. As a result, laboratories may prefer to diagnose such cases as
benign/negative with an explanatory note. Alternatively, if sufficient volume were
available, laboratories could make additional preparations in the face of low cellularity
samples which may increase the number of cells found to adequate levels and avoid the
expense and invasiveness of obtaining an instrumented urine specimen.
Previous studies of ThinPrep preparations of urinary bladder wash cytology sug-
gest that specimens with 10–20 cells per 10 HPF should be diagnosed as “less than
satisfactory,” and a more recent study reported that cases with less than 20 cells per
HPF should be considered as “limited” because of significantly lower sensitivity [6,
17]. While the available data are small and limited to a few institutional experiences,
the addition of this category appears to be useful: patients who returned to provide
a repeat voided urine sample usually yielded fully adequate specimens, some of
which resulted in diagnostic findings [4].
Recommendations
Adequacy in urine specimens is a topic for which there remains limited concrete,
quality data in the context of the number of variables at play. As such, the recom-
mendations for adequacy in TPS continue to center on the adequacy algorithm.
When properly validated for each decision point, this algorithm will increase stan-
dardization and quality in reporting across laboratories. As more quantitative results
of validation become available in the literature, we expect this algorithm to adopt
more defined prescriptive qualities that may lead to uniform practice in all urine
adequacy determinations. For now, TPS recommends adequacy as at least 30 mL
voided urine volume when evaluating specimens in the context of the adequacy
algorithm. (Figs. 2.1 and 2.2).
Future Directions
Additional studies are needed to evaluate the correlation between volume and ade-
quacy for voided urines as well as cellularity and adequacy for bladder and upper
urinary tract washings in laboratories that use different cytopreparatory techniques.
Another focus for future studies is the assessment and definition of “less than
2 Adequacy of Urine Specimens (Adequacy) 17
Sample Reports
Example 1
Voided urine:
Adequacy: Satisfactory for evaluation
Interpretation: Atypical urothelial cells present
Example 2
Voided urine:
Adequacy: Less than optimal
Interpretation: Negative for high-grade urothelial carcinoma. See
note.
Example 3
Voided urine:
Adequacy: Unsatisfactory for evaluation
Interpretation: Nondiagnostic sample; see note.
Example 4
Urine, bladder wash:
Adequacy: Less than optimal
Interpretation: Negative for high-grade urothelial carcinoma. See
note.
Example 5
Urine, bladder wash:
Adequacy: Unsatisfactory
Interpretation: Nondiagnostic sample. See note.
References
1. Olson MT, Boonyaarunnate T, Aragon-Han P, Umbricht CB, Ali SZ, Zeiger MA. A tertiary
center’s experience with second review of 3885 thyroid cytopathology specimens. J Clin
Endocrinol Metab. 2013;98:1450–7.
2. Barkan GA, Tabatabai ZL, Kurtycz DFI, Padmanabhan V, Souers RJ, Nayar R, Sturgis
CD. Practice patterns in urinary cytopathology prior to The Paris System for Reporting Urinary
Cytology. Arch Pathol Lab Med. 2020;144(2):172–6.
3. Kurtycz DFI, Brimo F, Rosenthal DL, Siddiqui MT, Tabatabai ZL, VandenBussche CJ,
Wojcik EM, Barkan GA. Perceptions of Paris: an international survey in preparation for
The Paris System for Reporting Urinary Cytology 2.0 (TPS 2.0). J Am Soc Cytopathol.
2021;10(5):S4 (PL06).
4. VandenBussche CJ, Rosenthal DL, Olson MT. Adequacy in voided urine cytology specimens:
the role of volume and a repeat void upon predictive values for high-grade urothelial carci-
noma. Cancer Cytopathol. 2016;124:174–80.
5. Rezaee N, Tabatabai ZL, Olson MT. Adequacy of voided urine specimens prepared by
ThinPrep and evaluated using The Paris System for Reporting Urinary Cytology. J Am Soc
Cytopathol. 2017;6:155–61.
6. Renshaw AA, Gould E. Adequacy criteria for voided urine cytology using cytospin prepara-
tions. Cancer Cytopathol. 2019;127:116–9.
7. Swiderek J, Morcos S, Donthireddy V, Surapaneni R, Jackson-Thompson V, Schultz L, et al.
Prospective study to determine the volume of pleural fluid required to diagnose malignancy.
Chest. 2010;137:68–73.
8. Rooper LM, Ali SZ, Olson MT. A minimum fluid volume of 75 mL is needed to ensure
adequacy in a pleural effusion: a retrospective analysis of 2540 cases. Cancer Cytopathol.
2014;122:657–65.
9. Xing J, Yan Q, Monaco SE, Pantanowitz L. Determination of appropriate urine volume cutoff
values for voided urine specimens to assess adequacy. J Am Soc Cytopathol. 2019;8:89–94.
10. Renshaw AA, Gould EW. Evidence-based adequacy criteria for instrumented urine cytology
using cytospin preparations. Diagn Cytopathol. 2018;46:520–1.
11. Barkan GA, Tabatabai ZL, Sturgis C, Kurtycz DF, Souers RJ, Nayar R. In preparation for
The Paris System for Reporting Urinary Tract Cytopathology (TPSRUTC): observations
from the 2014 supplemental questionnaire of the College of American Pathologists (CAP)
Cytopathology Interlaboratory Comparison Program (CICP) (abstract). Lab Investig.
2015;95(Suppl 1):83A.
12. Murphy WM, Crabtree WN, Jukkola AF, Soloway MS. The diagnostic value of urine versus
bladder washing in patients with bladder cancer. J Urol. 1981;126:320–2.
13. Studeman KD, Ioffe OB, Puszkiewicz J, Sauvegeot J, Henry MR. Effect of cellularity on
the sensitivity of detecting squamous lesions in liquid-based cervical cytolog. Acta Cytol.
2003;47:605–10.
2 Adequacy of Urine Specimens (Adequacy) 19
14. Michael CW, Pang Y, Pu RT, Hasteh F, Griffith KA. Cellular adequacy for thyroid aspirates
prepared by ThinPrep: how many cells are needed? Diagn Cytopathol. 2007;35:792–7.
15. Bastacky S, Ibrahim S, Wilczynski SP, Murphy WM. The accuracy of urinary cytology in daily
practice. Cancer. 1999;87:118–28.
16. Layfield LJ, Elsheikh TM, Fili A, Nayar R, Shidham V. Papanicolaou Society of Cytopathology.
Review of the state of the art and recommendations of the Papanicolaou Society of
Cytopathology for urinary cytology procedures and reporting: the Papanicolaou Society of
Cytopathology Practice Guidelines Task Force. Diagn Cytopathol. 2004;30:24–30.
17. Prather J, Arville B, Chatt G, Pambuccian SE, Wojcik EM, Quek ML, Barkan GA. Evidence-
based adequacy criteria for urinary bladder barbotage cytology. J Am Soc Cytopathol.
2015;4:57–62.
18. Richardson CJ, Pambuccian SE, Barkan GA. Split-sample comparison of urothelial cells in
ThinPrep and cytospin preparations in urinary cytology: do we need to adjust The Paris System
for Reporting Urinary Cytology criteria? Cancer Cytopathol. 2020;128:119–25.
19. Barkan GA. Enough is enough: adequacy of voided urine cytology. Cancer Cytopathol.
2015;124:163–6.
Negative for High-Grade Urothelial
Carcinoma (NHGUC) 3
Christopher J. VandenBussche, Ashish Chandra,
Jonas J. Heymann, Zulfia McCroskey, Christopher L. Owens,
Pawel T. Schubert, and Yeh-Han Wang
Background
Many cytology reporting systems regard the “negative” category as containing only
normal cells from a body site without any evidence of neoplasia. Mild indeterminate
changes are placed in the “atypical” category. If TPS followed that premise, the
“atypical” category would be so large that it would serve no clinical usefulness [1].
Following the lead of The Bethesda System for Reporting Gynecologic Cytology,
TPS includes in the negative category all entities that pose no significant risk to the
patient for developing high-grade urothelial carcinoma (HGUC) based upon current
knowledge. Therefore, if a specific cause of a particular morphologic alteration in
urothelial cells is recognized and the cause is not associated with malignancy, e.g.,
radiation-induced “atypia,” those cases are best classified as “Negative for HGUC”
(NHGUC) and not atypical, unless there are other cellular alterations in the sample
that warrant the “Atypical Urothelial Cells” (AUC) designation. Furthermore, the
goal of TPS is to highlight those cases that are at risk for HGUC. Therefore, the
negative category emphasizes that goal by stating NHGUC.
Many conditions can cause discernible reactive and reparative cytologic changes
but pose no threat to the patient of developing HGUC [2], and they are included in
this chapter. Those entities in which there may be an association with neoplasia will
be so noted and included in both the NHGUC and AUC categories with appropriate
codicils.
Superficial cells are large, shaped like the canopy of an umbrella, with rounded
(convex) luminal surfaces and scalloped (concave) borders onto which the underly-
ing intermediate cells are sometimes attached (Fig. 3.1). The cytoplasm is abundant
and vacuolated or foamy, not to be mistaken for virally induced koilocytes. They are
often bi- or multinucleated or may contain a single large nucleus. The nuclei are
centrally located, round to oval, with smooth nuclear membranes. The chromatin is
fine and an occasionally prominent chromocenter/nucleolus is present.
Characteristically, the nuclear/cytoplasmic (N/C) ratio is low.
In contrast, intermediate (“parabasal-like”) urothelial cells have nuclei with basi-
cally the same size and character as superficial cells, but have less cytoplasm,
imparting a higher N/C ratio (Fig. 3.2). Since they are less mature than superficial
cells, their nuclear chromatin may be a bit coarser than the ubiquitous superficial
cells, and they may have shallow grooves oriented along the long axis of the nucleus;
but otherwise thin, even nuclear membranes and uniform chromatin distribution
will support their totally benign condition. Cytoplasm will not be as vacuolated as
superficial cells but is not completely opaque (homogeneous), the latter feature
being cited as a characteristic of low-grade urothelial neoplasms in some studies
[3–6]. Because of their increased N/C ratios, they are commonly misinterpreted and
placed in the AUC category. Identification of their bland nuclear features can help
avoid that pitfall.
Deep urothelial cells (basal cells) are small cells that form small-to-intermediate
sized fragments (Fig. 3.3). They may appear hyperchromatic at low magnification
due to their elevated N/C ratios. However, the nuclei are uniform and regularly
3 Negative for High-Grade Urothelial Carcinoma (NHGUC) 23
Fig. 3.1 Superficial
urothelial (umbrella) cells. a
(a) The cytoplasm of large
superficial cells is frothy
and abundant, resulting in
a low N/C ratio. Nuclei
have pale, finely granular
chromatin and prominent
nucleoli. Multinucleation
is common, especially in
instrumented samples.
(Washing, TP, medium
mag.). (b) In addition to
superficial (umbrella) cells,
clusters of smaller cells are
seen (arrows). The nuclei
are darker and slightly
smaller than the superficial
cells, but the nuclear b
shapes are round, nuclear
membranes are smooth,
and architecture is
uniform. N/C ratios are
high but, in the context of
the other criteria, should
not be considered atypical.
(Washing, TP, medium
mag.). (c) Umbrella cells
are the most superficial
cells in the bladder,
creating an “umbrella”
over all other urothelial
cells. Their nuclear and
cytoplasmic character is
the same as other
superficial cells, but c
additionally, they possess a
thickened cytoplasmic
edge that doesn’t go all
around the cell. (Washing,
CS, medium mag.)
24 C. J. VandenBussche et al.
Fig. 3.2 Intermediate urothelial cells. These cells have a “fried egg” appearance and resemble the
parabasal squamous cells seen in Pap test specimens, causing them to also be described as
“parabasal-type” or “parabasal-like” cells. They are usually found singly or in small, loose clusters
(as seen here). Note the low N/C ratios, oval-shaped nuclei with regular borders, and bland chro-
matin pattern. The nuclei also are uniform in size. (Voided, SP, high mag.)
arranged within their tissue fragments. The chromatin lacks the coarse, clumpy
quality associated with HGUC. Although they may form papillary fragments, they
lack fibrovascular cores and thus are not “true” papillary fragments. This helps dis-
tinguish fragments of basal cells from low grade urothelial neoplasms.
3 Negative for High-Grade Urothelial Carcinoma (NHGUC) 25
Both men and women can be expected to have benign squamous cells (Fig. 3.4) in
their urine, although they are more common in women. In voided urine from a
woman, the origin may be the urethra but also may be a contaminant arising from
the vagina or perineum. The bladder trigone is subject to estrogenic effect and can
produce squamous cells similar to those found in a cervical sample. Instrumentation
can liberate squamous cells from the urethra and from areas of squamous metapla-
sia in both men and women. Of note, chronic irritation, particularly due to stones, is
often responsible for squamous metaplasia. Although keratinizing squamous meta-
plasia may be a precursor of squamous neoplasia, TPS does not require specifically
reporting the presence of squamous epithelial cells, including parakeratotic or anu-
cleate forms.
Glandular Cells
In voided urine from women, benign glandular cells (Fig. 3.5) may be derived from
the uterine cervix or corpus and are usually few and degenerated. Endometrial-type
cells are characterized as cohesive, three-dimensional aggregates of small glandular
cells with scant cytoplasm, slightly irregular nuclei with vesicular fine chromatin,
and visible small nucleoli. A few clusters in a menstruating woman are of no conse-
quence. However, when the patient is menopausal, and endometrial cells are present
in her voided urine, this is of concern, and clinicians should be alerted (see Chap. 8,
“Non-Urothelial Malignancies and Other Miscellaneous Lesions”).
Glandular-type cells from the urinary tract have small nuclei and vacuolated
cytoplasm. Most often glandular groups are multilayered true tissue fragments but
also can be seen as single cells. There are many potential sites of origin for glandu-
lar cells in urologic samples. Glandular epithelium may arise in the dome of the
bladder (urachal remnant) or trigone as a developmental rest. These changes are not
metaplastic. Endometriosis may involve the urinary tract and present an unexpected
cellular picture of very small hyperchromatic cells, either in voided urine or brush-
ings from the ureter (Fig. 3.5) Cells from Mullerian rests (Mullerianosis) are also
native, albeit rare and may have endometrial, endosalpingial, or endocervical fea-
tures. Conversely, cystitis cystica/glandularis and intestinal metaplasia (Fig. 3.6) are
both metaplastic processes of the urothelium resulting from chronic inflammation.
Although the metaplasia is benign, patients with extensive intestinal metaplasia
involving the urinary tract are at risk of developing subsequent bladder adenocarci-
noma [12]. Therefore, if definitively identified, such metaplasia may be reported as
an additional finding in the NHGUC category.
Renal tubular epithelial cells (RTCs) (Fig. 3.7) will usually appear as degener-
ated, small cells with mildly irregular nuclear contours. When derived from the
distal tubules, the cells will have high N/C ratios; if examined at high magnification,
Fig. 3.5 Benign glandular cells – endometriosis. Glandular cells in a urinary tract specimen may
be native to the urinary collecting system, or external to it. The larger cells are urothelial or squa-
mous. The cluster of small dark cells originated in an endometriosis of the ureter. The patient was
presenting with hematuria and pain coincident with her menstrual periods. (Ureteral brushing, CS,
high mag.)
3 Negative for High-Grade Urothelial Carcinoma (NHGUC) 27
Fig. 3.6 Cystitis cystica/glandularis. (a) Glandular cells from the lining of the bladder can origi-
nate from a focus native to the urothelium, metaplasia from an inflammatory focus (cystitis cystica/
glandularis), or from a glandular neoplasm, either primary or secondary. Unless the cytomorphol-
ogy suggests a neoplasm, glandular cells are considered benign. (Washing, TP, high mag.). (b)
Benign glandular cells. The nuclei are uniform in size and shape, with a bland chromatin pattern,
and the cells maintain a columnar architecture. (Washing, SP, high mag.). (c) Cystitis cystica may
appear as a single layer of glandular cells. They closely resemble endocervical cells and could be
from a case of endocervicosis. This mucosal strip was from a focus of cystitis cystica. (Washing,
TP, high mag.). (d) Another tight glandular group, a BUTF, demonstrates nuclear compression by
relatively large cytoplasmic vacuoles. Regardless of their origin, these cells fulfill the criteria of
benignity, making the diagnosis of NHGUC appropriate. (Washing, TP, high mag.). (e) Benign
glandular cells. The glandular cells are numerous in this field, and the formation of a hyperchro-
matic crowded group may be alarming. It is easiest to identify their columnar morphology at the
edges or in the singly dispersed cells. Note how the nuclei are all relatively similar in size, in addi-
tion to maintaining their apical cytoplasm. (Washing, SP, high mag.). (f) Benign glandular cells. In
this separate field, the glandular cells are mostly individually dispersed and some appear more
cercariform than columnar. The cells have similarly sized, oval-shaped nuclei with bland chroma-
tin. This cytomorphologic pattern has significant overlap with that of a LGUN but without a fibro-
vascular core. Fortunately, there are no features of HGUC, so a diagnosis of NHGUC is not
challenging to make. (Washing, SP, intermediate mag.)
28 C. J. VandenBussche et al.
Fig. 3.6 (continued)
c
f
3 Negative for High-Grade Urothelial Carcinoma (NHGUC) 29
c
30 C. J. VandenBussche et al.
they may be incorrectly interpreted as atypical urothelial cells. Cells from the proxi-
mal tubules, less commonly seen, have lower N/C ratios due to abundant, granular
cytoplasm, resembling degenerated histiocytes. RTCs are usually found in voided
urine specimens as rare, small fragments of cells with scalloped edges that appear
loosely attached at low magnification. The fragments typically contain five to ten
cells. The recognition of this pattern at low magnification can prevent over interpre-
tation at higher magnifications. Depending on the site from which they arise, the
morphology of these cells may vary slightly, and they may also be seen to form casts.
Degenerative Changes
Most voided urine specimens contain urothelial cells that have undergone some
degree of degeneration. These cells have naturally exfoliated into the urinary stream
and remained in the bladder at body temperature until specimen collection. Urine is
a poor cellular preservative, and once a specimen is collected, it may sit at room
temperature for some time until it is refrigerated, processed, or placed into a preser-
vative. Certain conditions may contribute to cellular degeneration, such as immuno-
therapy (described below), urolithiasis, and bacterial infection. Urine specimens
collected following the relief of a urinary tract obstruction may contain numerous
degenerated cells (Fig. 3.8).
Degenerated benign intermediate urothelial cells are often smaller than well-
preserved urothelial cells and have shrunken (condensed), dark nuclei with mildly
irregular nuclear borders. The N/C ratio of these cells typically remains below 0.5
(Fig. 3.9). Characteristic of degeneration are large eosinophilic and/or cyanophilic
Fig. 3.8 Cellular degeneration secondary to stasis. When cells sit in the urinary tract for long
periods of time, they begin to degenerate. Many of the cells here have small, dark pyknotic nuclei
but retain a large amount of cytoplasm. The cells have granular cytoplasm and some cells have
irregular cytoplasmic vacuoles. More severely degenerated cells have lost their nuclear-cytoplasmic
interface. The dark nuclei may be concerning for urothelial carcinoma; however, specimens with
degenerated HGUC cells will often contain some larger cells, and the nuclei will be larger in size
than those seen here. (Catheterized, SP, intermediate mag.)
3 Negative for High-Grade Urothelial Carcinoma (NHGUC) 31
Fig. 3.9 Degeneration. (a) Specimens may contain a large number of degenerated cells, which can
give the impression of a proliferative or neoplastic process. Examination of each individual cell in
this field is reassuring, as the cells contain dark yet very small nuclei and maintain their N/C ratios.
Many are RTCs (voided urine, SP, high mag.). (b) This degenerated benign cell has chromatin that
appears smudged and homogenous. The N/C ratio is low and the cytoplasm is vacuolated (voided
urine, SP, high mag.). (c) These degenerated cells are enlarged and dark, but the nuclear-cytoplasmic
interface is becoming lost in the cell on the right. The nuclei of both cells contain a condensed ball
of chromatin. These cells are too far degenerated for assessment. (Voided urine, SP, high mag.). (d)
At low magnification, this cell may appear to have a high N/C with a prominent nucleolus, but the
cytoplasm is vacuolated and the nuclear-cytoplasmic interface cannot be clearly demarcated. The
specimen diagnosis should not include consideration of this cell, given the severity of its degenera-
tion. (Voided urine, SP, high mag.)
32 C. J. VandenBussche et al.
Fig. 3.9 (continued)
d
Fig. 3.11 (continued)
d
Explanatory Note The use of the phrases “tissue fragments,” “cell clusters,” and
“cellular aggregates” has not been consistent in the literature. According to Frost,
clusters consist of cells that associate together in the fluid, whereas true tissue frag-
ments contain cells that arise and grow together in vivo [14]. Moreover, in a cluster,
cells are loosely arranged with “windows” (spaces) between cells, and nuclei are
round, most likely due to the absence of forces exerted by neighboring cells. This
section refers to “tissue fragments” with the understanding that small tissue frag-
ments meeting the definition of Frost may be referred to as “clusters” by some.
Atypical urothelial tissue fragments (AUTFs) are tissue fragments that contain atyp-
ical cytomorphologic features (Fig. 3.13). While the level of atypia seen may merit
classification of these fragments into the AUC category, an increased threshold is
recommended for making an AUC diagnosis when only AUTFs are seen. Because
HGUC cells tend to be discohesive, it is generally more helpful to examine the sin-
gly dispersed cells in a specimen for atypia rather than tissue fragments. If the only
atypical cells in a specimen are in tissue fragments, one should strongly consider the
possibility that they are from a benign entity (“negative”). The type of specimen,
voided vs. instrumented, is also a factor in the evaluation.
Fig. 3.13 Atypical urothelial tissue fragment (AUTF) adjacent to HGUC cell. The cells in this
urothelial tissue fragment display atypical features: irregular nuclear membranes, increased N/C
ratios, coarse chromatin, and anisonucleosis. It is difficult to assess N/C ratios and chromatin pat-
tern in a three-dimensional tissue fragment; thus, one may have an increased threshold to make a
diagnosis of AUC or SHGUC based on a tissue fragment alone. The more concerning cell is to the
bottom right of the AUTF, which has all the qualitative features of HGUC. This patient had HGUC
on follow-up biopsy. (Bladder washing, SP, high mag.)
38 C. J. VandenBussche et al.
Fig. 3.14 Low-grade urothelial neoplasm (LGUN). (a) A monotonous population of neoplastic
cells fills the field, almost forming a sheet. However, the presence of vascular structures containing
flattened endothelial cell nuclei indicates that this is a papillary lesion. The neoplastic cells have
bland chromatin and are small in size. This finding falls under the NHGUC diagnostic category
(renal pelvic washing, Cytospin, medium mag.). (b) At higher magnification, one can appreciate
the bland chromatin pattern of the neoplastic cells. The nuclei are oval-shaped and have minimal
irregularity in their nuclear contours. These features are not compatible with HGUC category
(renal pelvis washing, Cytospin, high mag.). (c) This papillary fragment contains an endothelial-
lined fibrovascular core. Numerous monomorphic neoplastic cells line the core, and detached neo-
plastic cells can also be seen nearby in the background. The N/C ratio of these cells (which is less
than 0.5) can be more clearly assessed using the detached cells rather than the cells within the
three-dimensional papillary fragment (renal pelvis washing, Cytospin, high mag.). (d) In a separate
field, small fragments of neoplastic cells can be seen which lack the fibrovascular core. They lack
features of HGUC and have significant overlap with benign urothelial tissue fragments. Thus, the
cytomorphologic features of individual LGUN cells and tissue fragments are nonspecific (renal
pelvis washing, Cytospin, high mag.)
40 C. J. VandenBussche et al.
Fig. 3.14 (continued)
c
Explanatory Note 2 Similar to earlier reports [11, 18], a recent study [21] found
that the majority of the features described previously as diagnostic for LGUC were
observed almost equally in patients with or without biopsy-proven LGUC, regard-
less of whether the specimens were from the upper or the lower urinary tract.
Specifically, mild nuclear membrane irregularity was present in 48% of LGUC and
47.2% of negative controls (p = 0.93); mild nuclear enlargement was observed in
42.9% of LGUC patients and 49.1% negative controls (p = 0.26). Although homo-
geneous cytoplasm and three-dimensional papillary structures with fibrovascular
cores were found only in LGUC, there were many that did not show these features.
Hence, these criteria were not statistically significant in this study. The only time a
definitive diagnosis of LGUN can be rendered in instrumented urine is when well-
defined fibrovascular cores (with capillaries) are present [9]; this finding, however,
is exceedingly rare (Figs. 3.16 and 3.17). In these instances, LGUN can be diag-
nosed as a second line diagnosis under a top line diagnosis of NHGUC, or the find-
ing can be described in a note with a top line diagnosis of NHGUC.
Table 3.1 Significance of benign urothelial tissue fragments (BUTFs) and atypical urothelial tis-
sue fragments (AUTFs) in voided urines (Onur I, Rosenthal DL, VandenBussche CJ. Atypical
urothelial tissue fragments in noninstrumented voided urine specimens are associated with low but
significantly higher rates of urothelial neoplasia than benign-appearing urothelial tissue frag-
ments [20]
No. (%) No. (%)
Follow-up diagnosis BUTF AUTF
Histopathologic 29 (10.6) 24 (14.1)
Benign 17 (6.2) 6 (3.5)
Urothelial neoplasia 12 (4.4) 17 (10.0)
PUNLMP 1 (0.4) 0 (0.0)
LGUC 9 (3.3) 2 (1.2)
HGUC 2 (0.7) 15 (8.8)
Cytopathology 45 (16.4) 25 (14.7)
Benign/AUC
Urolithiasis (clinical/radiologic/gross) 45 (16.4) 49 (28.8)
No follow-up 25 (9.1) 10 (5.9)
data support using a higher threshold for classifying AUTFs in the AUC category, as
the ROHM for specimens containing AUTFs is much lower than what has been seen
for the AUC category (8.8% vs. 12.3–69%); (see Chap. 11, Risk of High-Grade
Malignancy).
Explanatory Note Instrumented urine specimens include any specimen that was
obtained by any instrument or whenever force was applied to dislodge individual
cells from the lining urothelium. These include catheterized urines, bladder wash-
ings (barbotage), brushings, or upper urinary tract urine specimens obtained by uri-
nary catheterization. In addition, any immediately post-cystoscopy urine specimen
is also considered instrumented. In general, these specimens are cellular, and the
presence of BUTF in instrumented urine specimens is a normal finding; therefore,
we should avoid the term “atypia” in these types of specimens so long as the nuclear
and architectural features of the UTF do not warrant another diagnosis.
Voided urine specimens from patients who present with hematuria and/or filling
defect on imaging studies are often cellular and consist of three-dimensional urothe-
lial fragments composed of cells that may exhibit significant pleomorphism
(Fig. 3.19). Urinary stones usually abrade the lining causing morphologic changes
to liberated cells and tissue fragments. These three-dimensional urothelial frag-
ments display smooth (community) borders and cytoplasmic collars/collarets, i.e., a
rim of cytoplasm surrounding the nuclei as in BUTF. If a cause of the atypia is
found, such as a stone, then the diagnostic category should be NHGUC.
There may also be sheets or clusters of cells that are not true tissue fragments
(see Fig. 3.11). The cells in the fragments or clusters may also exhibit nuclear
enlargement and atypia, slightly increased N/C ratios, nuclear and cytoplasmic
degeneration, and/or squamous metaplasia (Fig. 3.19). With a confident diagnosis
of lithiasis, and without single HGUC cells in the sample, then the NHGUC cate-
gory is appropriate. If there are any truly atypical well-preserved single cells, the
sample needs to be considered either AUC or SHGUC based upon the criteria for
those categories. Occasionally, crystalline material from the stone will support the
diagnosis (Fig. 3.19c).
3 Negative for High-Grade Urothelial Carcinoma (NHGUC) 45
Fig. 3.19 (continued)
d
In the urinary bladder, these infections may cause generalized reactive changes in
the urothelium similar to those that have been discussed above. Urine specimens
from acute bacterial infections are sometimes cellular, consisting of reactive urothe-
lial cells with slightly enlarged nuclei and prominent nucleoli, but with fine, evenly
distributed chromatin and thin nuclear membranes. The presence of infiltrating neu-
trophils admixed with reactive urothelial cells supports a reactive, benign process,
as do clusters of bacteria in the background. In the absence of a significant number
of neutrophils, clusters of bacteria may also represent specimen contamination dur-
ing collection. If the predominant cells are neutrophils, with rare urothelial cells,
then the specimen should be considered inadequate, with the reason for that desig-
nation included in the report.
decades, one of the most important and most identified viruses in urine specimens is
polyomavirus (Fig. 3.21). Human polyomaviruses are small, non-enveloped double-
stranded DNA viruses that are classified into two main strains that may infect the
urinary tract, named after the initials of the patients from whom they were first identi-
fied (BK and JC) [25]. Polyoma virus infected cells are enlarged with single, usually
homogeneous basophilic inclusions occupying most of the enlarged nuclear area
(Fig. 3.21). Nuclear membranes of those cells are smooth and regular in shape as
compared to the irregular nuclear membranes in high-grade malignant cells, which
they often mimic (“decoy” cells). When these cells degenerate, the basophilia clears
as the chromatin extrudes, leaving a framework or spider web. Usually, only a few
infected cells are found (Fig. 3.21), unless the patient is immunosuppressed.
The presence of cells with well-recognized viral changes should not lead to a
diagnosis of atypia. Primary polyomavirus infections occur during childhood and
are usually subclinical. Over 90% of adults are seropositive for viral antibodies. The
virus generally remains latent in the renal tubular epithelium, but intermittent viru-
ria can be detected in 0.3% of healthy adults. The infection is reactivated in indi-
viduals with various degrees of immunological deficit. In renal transplant recipients,
polyomavirus nephropathy occurs in 3–5% of patients, and loss of transplant occurs
in 50% of those affected [25]. Urines from such patients may contain many cells
with polyoma cytopathic effect.
The infected cells can easily be misclassified as malignant. Therefore, they were
referred to as “decoy cells,” analogous to “decoy ducks” used in hunting wild ducks,
by Andrew Ricci, a cytotechnologist working in the cytology laboratory of Memorial
Sloan Kettering Cancer Center. The term was coined and popularized by Leopold
Koss in the second edition of his influential textbook, Diagnostic Cytology and Its
Histopathologic Bases [26, 27]. In addition, urothelial cells infected by polyomavi-
rus can be aneuploid and, thus, be a potential pitfall for any DNA-based tests,
including FISH [28–30].
Once polyoma-infected cells are identified in a specimen, one should be alert and
consider whether other atypical cells in the specimen may simply be artifact induced
by polyomavirus-infected cells as well. Because degenerated HGUC cells may
48 C. J. VandenBussche et al.
b c
Fig. 3.21 Polyomavirus – classic (a), spider web (b–c), benign case (d–f). (a) Classic polyoma
(BK) cytopathic effect includes enlargement of the nucleus and nuclear chromatin homogenization
and/or chromatin margination. The shape of the nucleus is always round or oval with a very smooth
outline. The cell cytoplasm may be absent or, in the case of these cells, form degenerating tails
(“comet cells”). In this case, the chromatin appears somewhat “clumpy” which may initially cause
concern for HGUC. However, all the other features suggest this is simply BK cytopathic effect
(voided, SP, high mag.). (b) Dissolution of the nuclear chromatin is also a characteristic of poly-
omavirus infection. The size and shape of the nucleus are the same as the classic features. (Voided,
SP, high mag.). (c) If the focal plane is changed, then a spider web of the cytoskeleton comes into
view. (Voided, SP, high mag.). (d) The assortment of pale and darker cells is striking on low mag-
nification. (Washing, TP, low mag.). (e) Closer view will demonstrate the reasons for the dark cells
observed on low magnification. Almost all the cells display glassy nuclear inclusions diagnostic of
polyomavirus (washing, TP, high mag.). (f) The BK-infected cells in this field demonstrate positive
nuclear staining for SV40 by immunocytochemistry. In most instances, a morphologic assessment
is sufficient for the diagnosis of BK infection; furthermore, infection with BK (and SV40 positiv-
ity) does not exclude HGUC. (Voided urine, SV40 immunocytochemistry, high mag.)
3 Negative for High-Grade Urothelial Carcinoma (NHGUC) 49
d e
Fig. 3.21 (continued)
Parasites
Rarely, parasites may be seen in urinary tract specimens. These may be extra-
urinary contaminants, such as pinworm (Enterobius vermicularis) ova, or may
arise from within the urinary tract (Schistosoma haematobium) (Fig. 3.22). While
Fig. 3.22 (continued)
c
e
52 C. J. VandenBussche et al.
Radiation
Fig. 3.23 Radiation changes. Patients may receive radiation treatment to the urinary tract to treat
a primary malignancy or areas of the urinary tract may become radiated due to treatment of extra-
urinary malignancies, such as cervical cancer. The changes seen are similar to those seen at other
anatomic sites: nucleomegaly with a corresponding increase in cytoplasm, multinucleation, and
cytoplasmic vacuolization. (Voided urine, SP, high mag.)
3 Negative for High-Grade Urothelial Carcinoma (NHGUC) 53
Immunotherapy
Fig. 3.24 Granulomatous
reaction following BCG a
immunotherapy. (a)
Multinucleation in
superficial urothelial cells
is common. In contrast,
Langhans-type giant cells
resulting from fused
macrophages are
multinucleated but have
their smaller and slightly
hyperchromatic nuclei
clustered at one pole of the
cytoplasm. Clinical history
revealed recent BCG
instillation following
diagnosis of bladder
cancer. (Voided, CS, high
b
mag.). (b) In addition to
Langhans giant cells,
granulomata can be found
in urine following BCG
immunotherapy. These
granulomata are no
different from those in any
other body site, complete
with monocytes,
lymphocytes, and
histiocytes in a tight
mélange. (Washing, TP,
high mag.)
54 C. J. VandenBussche et al.
cells present. Specimens may contain residual malignant cells (usually degenerated)
in patients who have recently received or have failed therapy. Six weeks should
elapse between the last of six BCG instillations and a urine sample to allow malig-
nant cells to die, thus avoiding a mistaken interpretation of “failed therapy.” Other
nonspecific findings include reactive umbrella cells, seen with large forms or in
increased numbers and background granular debris.
Chemotherapy
Intravesical mitomycin and thiotepa usually affect superficial cells and cause
nuclear enlargement, multinucleation, and hyperchromasia of those cells, all of
which are nonspecific but may be worrisome. History of therapy is important to
anticipate and interpret these changes. Systemic cyclophosphamide (Cytoxan),
given for reasons other than urothelial malignancy, often causes hemorrhagic cysti-
tis and has been reported to be associated with urothelial hyperchromasia and
degeneration, plus the presence of large nuclei and increased N/C ratios. In addition
to mimicking HGUC, these changes may indeed reflect the presence of HGUC
caused by systemic cyclophosphamide, albeit rarely [36, 37].
Chemotherapy may be combined with hyperthermia (chemohyperthermia) or
electric currents (electromotive drug administration) to increase drug penetration.
One study has shown cells in these specimens can have increased N/C ratios, hyper-
chromasia, and irregular nuclear membranes, which can simulate HGUC [38].
Explanatory Note Seminal vesicle cells have an abnormal DNA content and poten-
tially are a pitfall for DNA-based adjuvant tests [41]. When seminal vesicle cells are
recognized by the presence of yellow pigment and mature spermatozoa in the back-
ground, there is no need to call the urine specimen AUC.
Urinary diversion specimens are urines obtained from patients who underwent cys-
tectomy and one of the surgical procedures designed to reroute the urine flow (ileal
conduit, Indiana pouch, or neobladder). All of these procedures use a portion of
small bowel (ileum) that is anastomosed to the ureters and/or urethra (Fig. 3.26).
Urine specimens from these neobladders are very cellular and composed mainly
of degenerated enteric glandular cells, either singly or in small clusters, resembling
histiocytes, in a dirty background with mucus and bacteria (Fig. 3.27). In some
specimens, these components may obscure exfoliated urothelial cells, but they
should not distract from examination for HGUC. Often, urothelial cells from upper
tracts are present and show marked degeneration. These degenerated glandular and/
or urothelial cells may lead to an inappropriately increased rate of AUC diagnoses
for these specimens, with one study demonstrating a significantly lower positive
predictive value for the atypical category in patients with ileal neobladders com-
pared to those with native bladders (5.9% vs. 12%, respectively) [42].
56 C. J. VandenBussche et al.
Fig. 3.27 Urinary diversion specimen. (a) Urinary diversion specimens usually contain numerous
small degenerated cells in a background of granular debris. The degenerated cells arise from the
intestinal segment used for diversion. The cells have dark, yet very small, nuclei and granular
cytoplasm. (Catheterized, SP, low mag.). (b) A separate field shows the benign degenerated cells
with apoptotic bodies. These patients usually have a history of aggressive HGUC and are at high
risk for recurrence. HGUC cells would be larger than the small benign degenerated cells seen in
diversion specimens. This field contains strictly benign degenerated cells, though some three-
dimensional areas of cellular piling could appear atypical at lower magnification (catheterized, SP,
high mag.)
The rate of each diagnostic category depends upon the population served by the
laboratory. Referral centers, with oncologic urologists and a greater proportion of
follow-up specimens after diagnosis of malignancy compared to primary screening
specimens, will undoubtedly have much higher rates of SHGUC and outright
HGUC than reference laboratories serving general practitioners and internists.
Having established a baseline figure for each category, every laboratory is wise to
watch for “diagnostic drift,” wherein the indeterminate categories become waste-
baskets, influencing the rate of the adjacent categories; in urinary cytology, AUC
catches the overflow of difficult cases from the negative and suspicious categories.
Careful cytohistologic correlation is strongly recommended to monitor for and pre-
vent “diagnostic drift”.
Category Performance
Sample Reports
References
1. Rosenthal DL, Vandenbussche CJ, Burroughs FH, Sathiyamoorthy S, Guan H, Owens C. The
Johns Hopkins Hospital template for urologic cytology samples: part I-creating the template.
Cancer Cytopathol. 2013;121:15–20.
2. Wojcik EM. What should not be reported as atypia in urine cytology. J Am Soc Cytopathol.
2015;4:30–6.
3. Harris MJ, Schwinn CP, Morrow JW, Gray RL, Browell BM. Exfoliative cytology of the uri-
nary bladder irrigation specimen. Acta Cytol. 1971;15:385–99.
4. Hughes JH, Raab SS, Cohen MB. The cytologic diagnosis of low-grade transitional cell carci-
noma. Am J Clin Pathol. 2000;114(Suppl):59.
5. Raab SS, Slagel DD, Jensen CS, Teague MW, Savell VH, Ozkutlu D, Lenel JC, Cohen
MB. Low-grade transitional cell carcinoma of the urinary bladder: application of select cyto-
logic criteria to improve diagnostic accuracy [corrected]. Mod Pathol. 1996;9:225–32.
6. Raab SS, Lenel JC, Cohen MB. Low grade transitional cell carcinoma of the bladder. Cytologic
diagnosis by key features as identified by logistic regression analysis. Cancer. 1994;74:1621–6.
7. Barkan GA, Wojcik EM. Genitourinary cytopathology (kidney and urinary tract). Cancer Treat
Res. 2014;160:149–83.
8. Wojcik EM, Brownlie RJ, Bassler TJ, Miller MC. Superficial urothelial (umbrella) cells. A
potential cause of abnormal DNA ploidy results in urine specimens. Anal Quant Cytol Histol.
2000;22:411–5.
9. Zhou AG, Hutchinson LM, Cosar EF. Urine cytopathology and ancillary methods. Surg Pathol
Clin. 2014;7(1):77–88.
10. Reynolds JP, Voss JS, Kipp BR, Karnes RJ, Nassar A, Clayton AC, Henry MR, Sebo TJ, Zhang
J, Halling KC. Comparison of urine cytology and fluorescence in situ hybridization in upper
urothelial tract samples. Cancer Cytopathol. 2014;122:459–67.
11. Tapia C, Glatz K, Obermann EC, Grilli B, Barascud A, Herzog M, Schonegg R, Savic S,
Bubendorf L. Evaluation of chromosomal aberrations in patients with benign conditions and
reactive changes in urinary cytology. Cancer Cytopathol. 2011;119(6):404–10.
12. Bullock PS, Thoni DE, Murphy WM. The significance of colonic mucosa (intestinal metapla-
sia) involving the urinary tract. Cancer. 1987;59(12):2086–90.
3 Negative for High-Grade Urothelial Carcinoma (NHGUC) 61
13. Cowan ML, Rosenthal DL, VandenBussche CJ. Improved risk stratification for patients with
high-grade urothelial carcinoma following application of the Paris System for Reporting
Urinary Cytology. Cancer Cytopathol. 2017;125(6):427–34.
14. Frost JK. Concepts basic to general cytopathology. 4th ed. Baltimore: The Johns Hopkins
Press; 1972.
15. Layfield LJ, Elsheikh TM, Fili A, Nayar R, Shidham V. Papanicolaou Society of
Cytopathology: review of the state of the art and recommendations of the Papanicolaou Society
of Cytopathology for urinary cytology procedures and reporting : the Papanicolaou Society of
Cytopathology Practice Guidelines Task Force. Diagn Cytopathol. 2004;30(1):24–30.
16. Renshaw AA. Compassionate conservatism in urinary cytology. Diagn Cytopathol.
2000;22(3):137–8.
17. Renshaw AA, Nappi D, Weinberg DS. Cytology of grade 1 papillary transitional cell carci-
noma. A comparison of cytologic, architectural and morphometric criteria in cystoscopically
obtained urine. Acta Cytol. 1996;40(4):676–82.
18. Zhang ML, Rosenthal DL, VandenBussche CJ. The cytomorphological features of low-grade
urothelial neoplasms vary by specimen type. Cancer Cytopathol. 2016;124(8):552–64.
19. Onur I, Rosenthal DL, VandenBussche CJ. Benign-appearing urothelial tissue fragments in
noninstrumented voided urine specimens are associated with low rates of urothelial neoplasia.
Cancer Cytopathol. 2015;123(3):180–5.
20. Onur I, Rosenthal DL, VandenBussche CJ. Atypical urothelial tissue fragments in nonin-
strumented voided urine specimens are associated with low but significantly higher rates of
urothelial neoplasia than benign-appearing urothelial tissue fragments. Cancer Cytopathol.
2015;123(3):186–92.
21. McCroskey Z, Kliethermes S, Bahar B, Barkan GA, Pambuccian SE, Wojcik EM. Is a con-
sistent cytologic diagnosis of low-grade urothelial carcinoma in instrumented urinary tract
cytologic specimens possible? A comparison between cytomorphologic features of low-grade
urothelial carcinoma and non-neoplastic changes shows extensive overlap, making a reliable
diagnosis impossible. J Am Soc Cytopathol. 2015;4(2):90–7.
22. Zhang ML, Zhou AG, Rosenthal DL, VandenBussche CJ. Urinary tract washing specimens
containing atypical urothelial tissue fragments are significantly associated with urothelial neo-
plasia. Diagn Cytopathol. 2017;45(9):795–9.
23. Blacher EJ. Squamous cell carcinoma of renal pelvis. Urology. 1985:124–6.
24. Li MK, Cheung WL. Squamous cell carcinoma of the renal pelvis. J Urol. 1987;138(2):269–71.
25. Pinto M, Dobson S. BK and JC virus: a review. J Infect. 2014;68(Suppl 1):2.
26. Paquette C, Elhosseiny A. Significance of polyomavirus detection in urine cytology: an update.
Diagn Histopathol. 2012;18:321–6.
27. Koss LG. On decoy cells. Acta Cytol. 2005;49:233–4.
28. Wojcik EM, Miller MC, Wright BC, Veltri RW, O’Dowd GJ. Comparative analysis of DNA
content in polyomavirus-infected urothelial cells, urothelial dysplasia and high grade transi-
tional cell carcinoma. Anal Quant Cytol Histol. 1997;19:430–6.
29. Bakhos R, Shankey TV, Flanigan RC, Fisher S, Wojcik EM. Comparative analysis of DNA flow
cytometry and cytology of bladder washings: review of discordant cases. Diagn Cytopathol.
2000;22:65–9.
30. Halling KC, Kipp BR. Bladder cancer detection using FISH (UroVysion assay). Adv Anat
Pathol. 2008;15:279–86.
31. Galed-Placed I, Valbuena-Ruvira L. Decoy cells and malignant cells coexisting in the urine
from a transplant recipient with BK virus nephropathy and bladder adenocarcinoma. Diagn
Cytopathol. 2011;39:933–7.
32. Geetha D, Tong BC, Racusen L, Markowitz JS, Westra WH. Bladder carcinoma in a transplant
recipient: evidence to implicate the BK human polyomavirus as a causal transforming agent.
Transplantation. 2002;73(12):1933–6.
33. Lu H, Elsheikh TM, Zhang Y. Polyomavirus (BK) cytopathic effect in urine cytology is not
associated with high risk of developing high-grade urothelial carcinoma. J Am Soc Cytopathol.
2020;9(2):84–8.
62 C. J. VandenBussche et al.
34. Allison DB, Olson MT, Lilo M, Zhang ML, Rosenthal DL, VandenBussche CJ. Should the
BK polyomavirus cytopathic effect be best classified as atypical or benign in urine cytology
specimens? Cancer Cytopathol. 2016;124(6):436–42.
35. Martinez-Giron R, Esteban-Sanchis JG. Parasite eggs in urine cytology: fact or artifact? Diagn
Cytopathol. 2009;37(5):353–4.
36. Talar-Williams C, Hijazi YM, Walther MM, Linehan WM, Hallahan CW, Lubensky I, Kerr
GS, Hoffman GS, Fauci AS, Sneller MC. Cyclophosphamide-induced cystitis and bladder can-
cer in patients with Wegener granulomatosis. Ann Intern Med. 1996;124(5):477–84.
37. Knight A, Askling J, Granath F, Sparen P, Ekbom A. Urinary bladder cancer in Wegener's gran-
ulomatosis: risks and relation to cyclophosphamide. Ann Rheum Dis. 2004;63(10):1307–11.
38. Pierconti F, Rossi ED, Straccia P, Fadda G, Larocca LM, Bassi PF, Sacco E, Schinzari
G. The risk of malignancy of atypical urothelial cells of undetermined significance in patients
treated with chemohyperthermia or electromotive drug administration. Cancer Cytopathol.
2018;126(3):200–6.
39. Garret M, Hamm FC. Atypical cells of origin from the seminal vesicles, complicating cyto-
logic evaluation of prostatic secretions. Am J Clin Pathol. 1963;39:265–72.
40. Terada T. Monstrous epithelial cell clusters in the seminal vesicle. Int J Clin Exp Pathol.
2011;4(7):727–30.
41. Wojcik EM, Bassler TJ, Orozco R. DNA ploidy in seminal vesicle cells. A potential diagnostic
pitfall in urine cytology. Anal Quant Cytol Histol. 1999;21(1):29–34.
42. Cimino-Mathews A, Ali SZ. The clinicopathologic correlates of cellular atypia in urinary
cytology of ileal neobladders. Acta Cytol. 2011;55(5):449–54.
43. Nomani L, Abro S, Quek ML, Barkan GA. Guar bean in urinary cytology: a morphologic
pitfall. J Am Soc Cytopathol. 2021;10(1):41–6.
44. Ali-El-Dein B, El-Tabey N, Abdel-Latif M, Abdel-Rahim M, El-Bahnasawy M. Late uro-ileal
cancer after incorporation of ileum into the urinary tract. J Urol. 2002;167:84–8.
45. Brimo F, Vollmer RT, Case B, Aprikian A, Kassouf W, Auger M. Accuracy of urine cytology
and the significance of an atypical category. Am J Clin Pathol. 2009;132:785–93.
46. Raab SS, Grzybicki DM, Vrbin CM, Geisinger KR. Urine cytology discrepancies: frequency,
causes, and outcomes. Am J Clin Pathol. 2007;127:946–53.
47. McIntire PJ, Khan R, Hussain H, Pambuccian SE, Wojcik EM, Barkan GA. Negative predic-
tive value and sensitivity of urine cytology prior to implementation of The Paris System for
Reporting Urinary Cytology. Cancer Cytopathol. 2019;127(2):125–31.
48. McIntire PJ, Kilic I, Pambuccian SE, Wojcik EM, Barkan GA. The Paris System for Reporting
Urinary Cytology reduces atypia rates and does not alter the negative predictive value of urine
cytology. J Am Soc Cytopathol. 2021;10(1):14–9.
49. Ajit D, Dighe SB, Desai SB. Cytology of lleal conduit urine in bladder cancer patients: diag-
nostic utility and pitfalls. Acta Cytol. 2006;50:70–3.
50. Cimino-Mathews A, Ali SZ. The clinicopathologic correlates of cellular atypia in urinary
cytology of ileal neobladders. Acta Cytol. 2011;55:449–54.
Atypical Urothelial Cells (AUC)
4
Güliz A. Barkan, Margaret L. Compton, Tarik M. Elsheikh,
Kim A. Ely, Daniel F.I. Kurtycz, Merce Jorda,
Zahra Maleki, Sachiko Minamiguchi, Hiroshi Ohtani,
Eric Piaton, Bo Ping, Spasenija Savic Prince,
Z. Laura Tabatabai, and Christopher J. VandenBussche
Background
In addition to standardizing the reporting of urine cytology, one of the main objec-
tives of The Paris System (TPS) was to reduce the reported atypia rate, improve
clarity of communication, and thereby aid clinicians in patient management. Prior
to TPS the reported “indeterminant or atypical” rate of urine cytology was quite
variable and often facility dependent. There are reports of excessive atypia rates
reaching as high as 50% [1]. Obviously, such performance characteristics impair the
utility and reputation of urine cytology as a screening and/or a diagnostic tool [2].
The attempt to unify reporting and apply restraint to the reporting of “atypia” has
had some effect. Since the publication of TPS in 2016 [3, 4], there have been numer-
ous publications on urine cytology supporting the concepts behind the system [5–
40]. These studies have shown that the standardized criteria provided by TPS have
reduced the “atypia” rate when comparing the post-TPS to the pre-TPS rates (see
Table 4.1). Additionally, TPS has caused the cases that are interpreted as atypical to
be more meaningful, in that the risk of high-grade malignancy (ROHM) is signifi-
cantly increased in the TPS atypical urothelial cells (AUC) category. Early meta-
analyses exhibited ROHM ranging from approximately 32% in retrospective
Table 4.1 Published reporting rates of the “atypia” and “atypical urothelial cells” category before
and after The Paris System
Pre-TPS % (n) Post-TPS % (n)
Author Year Study location Atypia AUC
Vosoughi et al. [7] 2021 The USA (Miami) 16 (249) 9 (56)
Compton et al. [8] 2021 The USA (Nashville) N/A 12.5 (199)
Stanzione et al. [6] 2020 The USA (Los 59.8 (52) 41.5 (122)
Angeles)
Anbardar et al. [5] 2020 Iran 26 1.2 (22)
Rai et al. [60] 2019 India 16.7 (15) 11.1 (10)
Bakkar et al. [31] 2019 The USA (Los 44 (44) 23 (23)
Angeles)
Vallamreddy et al. [61] 2019 India 21.6 (16) 9.5 (7)
Wang et al. [15] 2018 Canada 18.6 (442) 14.4 (345)
Meilleroux et al. [25] 2018 France 6.1 (100) 5.2 (94)
VandenBussche et al. 2018 The USA (Baltimore) 23.9 (568) 23.0 (589)
[24]
Xing et al. [27] 2018 The USA (Pittsburgh) 34 (52) 24 (37)
Rohilla et al. [22] 2018 India 54.3 (73) 8.5 (114)
Zare et al. [23] 2018 The USA (San Diego) 24.2 (47) 11.9 (23)
Torous et al. [20] 2017 The USA (Boston) 29.5 (328) 21.8 (302)
Roy et al. [62] 2017 India 41.2 (40) 11.3 (11)
Granados et al. [9] 2017 Spain 4.7(7) 20.1 (30)
Suh et al. [17] 2017 Korea 25.4 (36) 14.8 (21)
Hassan et al. [12] 2016 Canada 38.7 (48) 25.8 (32)
AUC atypical urothelial cells, TPS The Paris System, N/A not applicable
(pre-TPS) studies to 43% in prospective (post-TBS) studies. From these works the
estimated ROHM for AUC is placed at 38.5% (±14.3%) as of this edition of TPS
(see Chap. 11). In addition, the use of TPS has resulted in an increase in specificity
and positive predictive values as well as an improved diagnostic concordance of
benign and malignant categories with histologic results [29].
Almost all diagnostic testing has a range of imprecision where values of diseased
and non-diseased states overlap. In clinical chemistry one may see the overlap in
analyte values like glucose or enzyme activity studies; in anatomic pathology it is
the range of increasing morphologic abnormality. There are values that are clearly
derived from a normal or benign state and others that are unmistakably derived from
disease, but the middle values are often a problem and do not yield a clear answer.
This is well explained in the classic 1975 clinical laboratory text Beyond Normality
by Galen and Gambino [41]. In the ideal testing situation there would be a clear set
of features or values that would be discrete for normal and abnormal situations
(Fig. 4.1); however, that is only occasionally the case.
More frequently we are presented with a range of values or morphologic features
that overlap (Fig. 4.2). We can confidently interpret the values or morphologies
without overlap, but at the places where both curves are present, we have a problem.
These overlapping values represent the domain of atypia (Fig. 4.3). In cytology,
cellular degeneration, poor preparations, reparative phenomena, and other artifact
4 Atypical Urothelial Cells (AUC) 65
Fig. 4.1 Diagnostic ideal: in the ideal testing situation, the values derived from a test, be they
numeric or a morphologic impression, would clearly separate the non-diseased population from a
diseased population
Fig. 4.2 Diagnostic reality: in the usual testing situation, some benign states yield values or fea-
tures that overlap with those found in a diseased population
Fig. 4.3 Diagnostic reality: the area of overlap between benign states and diseased states consti-
tutes a region of diagnostic uncertainty and provides a conceptual representation of the domain
of atypia
66 G. A. Barkan et al.
inducing situations can increase the overlap making interpretation more difficult.
On the other hand, well-fixed excellent preparations will tend to lessen the artifact.
Atypical cytology interpretations are frequently criticized as lacking specificity
and reproducibility, leaving clinicians without a clear course of action. Generic des-
ignations of “atypia” are not diagnostic entities but rather interpretations that reflect
the inability to make a definitive diagnosis due to a lack of reliable morphologic
features. Historically, the term “atypia” was introduced to cytopathology by Dr.
George N. Papanicolaou, to convey a very low suspicion of malignancy [42], but the
category leaves clinicians in a quandary and uncertainty in the minds of patients
whose samples generate such a result. Too frequent use of the term will stimulate a
search for another method which will almost certainly be more expensive and may
be more invasive.
Unfortunately, “atypia” is an unavoidable reality in our morphologic interpreta-
tions of cytology.
To reiterate, it fills the gap in the spectrum between what we can recognize as
entirely normal and what we can recognize as being clearly abnormal. Countless
papers lament that one cannot strictly define “atypia,” but these articles are missing
the point. Cytologic atypia is an amorphous concept of not being “typical” or “nor-
mal.” It is the negation of two terms which are themselves difficult to define. It may
be better to think of “atypia” in terms of probability, as Fig. 4.3 above; it represents
an x-axis potential for being more than normal. In other diagnostic systems, atypical
cytologic features are those which are more than can be explained by benign states
and reactive change but less than those which lead to a definitive diagnosis. It is
more than normal, but the cytologist cannot make the call. If a distinct definition
were possible, would the diagnosis still be atypia and not something else? One can
try to squeeze the edges of definition, but like an electron’s orbit, one cannot be
precise. An alternative approach would be to define the boundaries within which
atypia exists. Atypia is chaotic and capable of doing damage; we can set criteria and
keep it constrained. In TPS, the mandate is identifying life-threatening lesions,
especially HGUC; because of the bland nondiagnostic cytology associated with
low-grade, noninvasive neoplasms, AUC has been chosen to indicate worrisome
cells that may be associated with a high-grade invasive urothelial carcinoma
(HGUC). In effect AUC shifts the diagnostic point for atypia to the right on Fig. 4.3
and improves specificity, helping to ensure that HGUC is the target for TPS.
Definition
The general diagnostic category AUC is reserved for specimens that contain urothe-
lial cells with mild to moderate cytologic (not architectural) change that is concern-
ing for HGUC. Non-basal urothelial cells need to exhibit a nuclear to cytoplasmic
(N/C) ratio equal to or greater than 0.5. This feature serves as a warning that the
sample needs special attention and a diligent search for other features, including
nuclear hyperchromasia, coarse chromatin, and an irregular chromatic rim. In AUC
only one of these features needs to be present. Cytologic changes in AUC have to
fall short of a diagnosis of suspicious for high-grade urothelial carcinoma (SHGUC)
or high-grade urothelial carcinoma (HGUC) both of which demonstrate at least two
4 Atypical Urothelial Cells (AUC) 67
Fig. 4.4 Benign urothelial cells. The top left corner shows benign superficial urothelial cells
(umbrella cells), and the bottom right corner has benign intermediate/basal type urothelial cells.
Although the non-superficial urothelial cells have a high N/C ratio, they have a smooth nuclear
contour and do not show nuclear enlargement, placing them in the “negative” category. The follow-
up diagnosis was benign. (Bladder washing, TP, high mag.)
Fig. 4.6 Atypical urothelial cells (AUC). Two groups of urothelial cells are shown. The group on
the top left is orderly, composed of intermediate type urothelial cells with smooth nuclear contours,
and no features of atypia. The group on the bottom right is disorganized; urothelial cells have high
N/C ratio and nuclear contour irregularity. Nuclear chromasia is similar in both groups. Due to the
cytologic atypia seen in the group on the bottom, this case should be categorized as AUC. (Voided
urine, TP, medium mag.)
Fig. 4.7 Atypical
urothelial cells (AUC). A
group of atypical urothelial
cells with variable N/C
ratios and nuclear contour
irregularity. The absence of
hyperchromasia and the
presence of degenerated
clumped chromatin
preclude a diagnosis of
SHGUC. (Bladder
washing, TP, high mag.)
4 Atypical Urothelial Cells (AUC) 69
Fig. 4.8 Atypical
urothelial cells (AUC).
Atypical urothelial cells
with high N/C ratio,
enlarged nuclei (compared
to the neighboring benign
urothelial cells), and mild
nuclear contour
irregularities. The
chromatin is uniform and
hypochromatic, precluding
a diagnosis of
SHGUC. (Bladder
Washing, TP, high mag.)
a b
Fig. 4.9 Atypical urothelial cells (AUC). (a) Atypical urothelial cells with high N/C ratio, enlarged
nuclei (compared to the neighboring benign urothelial cells), and mild nuclear contour irregulari-
ties. (Bladder washing, TP, medium mag.). (b) Another image of the same case shows irregular
nuclear contours in a group of urothelial cells that have degenerative cellular changes. The cellular
changes are worrisome, but the degree of degeneration and lack of marked atypia precludes a
definitive diagnosis. (Bladder washing, TP, medium mag.)
70 G. A. Barkan et al.
Fig. 4.10 Atypical
urothelial cells (AUC).
Urothelial cells display
high N/C ratio,
anisonucleosis, nuclear
contour irregularities, and
alteration of nuclear
polarity. The chromatin is
finely granular with
minimal hyperchromasia.
This patient had renal
urolithiasis on follow-up.
These are reactive changes
and retrospectively
categorize this sample as
NHGUC. (Renal pelvis
washing, TP, high mag.)
a b
Fig. 4.11 Atypical urothelial cells (AUC). (a) Urothelial cells display increased N/C ratio, nuclear
enlargement, and mild hyperchromasia. (Renal pelvis washing, TP, high mag.). (b) The follow-up
nephrectomy showed a low-grade urothelial carcinoma by histology, causing a filling defect on
imaging. (Nephrectomy surgical sample, H&E, medium mag)
4 Atypical Urothelial Cells (AUC) 71
Fig. 4.12 Atypical urothelial cells (AUC). Urothelial cells have increased N/C ratio, enlarged
nuclei, and conspicuous nuclear contour irregularities. The chromatin is coarse and clumped. In
some of the urothelial cells, the N/C ratio is <0.5, and there are degenerative changes (such as fray-
ing of the cytoplasm), but due to the hyperchromasia and coarse chromatin pattern, a diagnosis of
AUC was rendered. Follow-up revealed a high-grade urothelial carcinoma of the kidney. (Bladder
washing, TP, high mag.)
a c
Fig. 4.13 Atypical urothelial cells (AUC). Urothelial cells with high N/C ratio and nuclear hyper-
chromasia. These three figures display all the atypical urothelial cells that are present in the speci-
men. (a) Atypical urothelial cells (AUC) (upper left). Urothelial cells display irregular nuclear
contours and cytoplasmic vacuolization, consistent with degenerative changes. (b) Aggregate of
cells (lower left) have markedly irregular nuclear contours and variation in nuclear size. In com-
parison to the neighboring squamous cells, there is mild nuclear hyperchromasia. There are cellu-
lar degenerative changes, such as partial loss of the cytoplasm and loss of crisp nuclear detail. (c)
Small aggregate of atypical cells adjacent to squamous cells (right). The urothelial cell nuclei also
show degeneration, but the one cell with the high N/C ratio is worrisome. The patient is a 36-year-
old woman with recurrent urolithiasis and no history of urothelial carcinoma. Her age and history
are low-risk factors for bladder cancer; however, without a history of renal calculi, these cytologic
features warrant the diagnosis AUC. (Voided, TP, high mag.). The follow-up was benign
72 G. A. Barkan et al.
Fig. 4.14 Atypical urothelial cells (AUC). The urothelial cell shown here has a high N/C ratio,
nuclear enlargement, and prominent nucleoli. Chromatin is pale and slightly coarse, but the chro-
matinic rim is thin and uniform. Compared to the nucleus of the neighboring inflammatory cells,
the nucleus of the atypical urothelial cell is much larger. Follow-up with cystoscopy and urinary
bladder biopsy within 6 months showed urothelial mucosa with reactive changes, likely due to
inflammation. (Bladder washing, TP, high mag.)
Fig. 4.15 Atypical urothelial cells (AUC). Urothelial cells show high N/C ratio, nuclear enlarge-
ment, and hyperchromasia. Some have prominent nucleoli. The nuclear contours are relatively
regular. The paucity of the atypical cells precluded a HGUC diagnosis. There are more than five
abnormal urothelial cells, and while they do not all show a N/C ratio >0.7, due to the presence of
the other criteria mentioned above, a diagnosis of SHGUC may also be appropriate for this case.
The follow-up showed high-grade urothelial carcinoma in the urinary bladder. (Bladder washing,
TP, high mag.)
4 Atypical Urothelial Cells (AUC) 73
Fig. 4.16 Atypical urothelial cells (AUC). Urothelial cells with high N/C ratio, mild nuclear
enlargement, and hyperchromasia with some degenerative changes in the cytoplasm. The follow-
up showed high-grade urothelial carcinoma of the bladder. (Bladder washing, TP, high mag.)
Fig. 4.17 Atypical urothelial cells (AUC). A single urothelial cell with high N/C ratio, nuclear
enlargement, hyperchromasia, and irregular nuclear contours with some degenerative cytoplasmic
changes. Compare this cell to the surrounding benign urothelial cells with low N/C ratio and regu-
lar chromatin pattern. Since this was the only atypical urothelial cell in the case, either an AUC or
a SHGUC diagnosis would be appropriate for this case. The patient was lost to follow-up. (Bladder
washing, TP, high mag.)
Cellular degeneration has been well described in earlier publications and reflects
forms of both reversible and irreversible cellular injury. Cytoplasmic borders may
be irregular, have a moth-eaten appearance, be variably lost, or display cytoplasmic
vacuolization. The chromatin can be clumped, hazy, smudged, or indistinct; the
chromatin rim may be interrupted or may exhibit variable thickness and irregular
contours but rarely possesses the sharp angles that are often seen in malignancy.
Multinucleated umbrella cells may contain numerous small, degenerate, and
74 G. A. Barkan et al.
pyknotic nuclei of variable sizes that may mimic the hyperchromasia of malignancy.
Urothelial cells, especially in voided urine, may show involutional changes with
poorly preserved urothelial cells possessing single and/or multiple cytoplasmic
eosinophilic inclusions, “Melamed-Wolinska bodies,” were named after Dr. Myron
R. Melamed and Ms. Wanda H. Wolinska CT (ASCP), who first described these
findings [43, 44]. These structures are apparently eosinophilic aggregates of degen-
erate cytoplasmic organelles and other cytoplasmic debris. Recent studies have
documented a reduction in the degenerative features mentioned above with the use
of collection fluids/fixatives [45, 46] (see Chap. 10).
The AUC category also includes specimens where, due to poor preservation and
degeneration, the nature and degree of change in the urothelial cells cannot be well
characterized and there is concern about HGUC; however, the mere presence of
degeneration does not warrant the diagnosis of AUC. Degeneration is an expected
finding in voided urine samples, especially after delayed processing and in urinary
diversion specimens. These changes should be interpretable given a sufficient level
of experience. If one is going to invoke AUC, there has to be a considered reason for
the interpretation. Alternatively, if a few well-preserved atypical cells, worrisome
for but not completely fulfilling criteria for SHGUC, are found in an otherwise
degenerated sample, AUC may be appropriate. It should be reiterated that all inter-
pretations may be modified by history and clinical setting.
For a discussion of this topic, the reader is referred to Chap. 8 (section “Atypical
Squamous Cells”). As stated there, the presence of atypical squamous cells (ASC)
in urine is a rare finding, and its categorization has proven problematic. The cells are
not clearly urothelial and so placement under AUC is inappropriate. They are not
clearly malignant, so inclusion in “non-urothelial malignancy” (NUM) also does
not fit. The TPS solution is to place specimens exhibiting atypical squamous cells in
a free-text (“other”) category with an explanatory note indicating the presence of
ASC and its implications. Sample reports are provided in Chap. 8; example 4 spe-
cifically addresses atypical squamous cells.
One of the significant and as yet unresolved issues regards an important cytomor-
phologic criterion to determine TPS categories, namely, the N/C ratio. This is a
number calculated by dividing the cross-sectional nuclear area by the cross-sectional
area of nucleus
cytoplasmic area of the cell ( ). Typically, this ratio is estimated
area of cytoplasm
by microscopists and is exposed to some degree of subjectivity. An N/C ratio of 0.5
was chosen by the authors of TPS as a discrete value to signal the possibility of a
significant lesion. It is well recognized that the estimation and value chosen is only
4 Atypical Urothelial Cells (AUC) 75
semiquantitative but does provide an observer with a mark at which to aim. A study
by Zhang, Guo, and VandenBussche surveying 137 cytologists showed that the N/C
ratio assessed visually was more reproducible between observers when compared
with other cytological features such as hyperchromasia or coarse chromatin; how-
ever, there was a tendency to overestimate an N/C ratio that approximates 0.5 [47].
Of interest, one of the early studies, performed by Vaickus and Tambouret, after the
publication of the first edition of TPS demonstrated that morphologists are indeed
capable of rendering relatively accurate estimates of the N/C ratio, especially at
higher levels. They also found that the practice of specifying discrete decimal N/C
ratios for atypical cells would be valid [48]. Shortly thereafter, Hang et al. used digi-
tal image analysis validating TPS recommendations for an N/C ratio cutoff at 0.5
for the AUC category [49]. Choosing a discrete value was not without criticism.
Another study by Long et al. showed problems with inter-observer agreement; how-
ever, the article was published shortly after the release of TPS, and its features had
to be somewhat novel to the study group [50]. Subsequently, McIntyre et al., found
mean N/C ratios for the HGUC and SHGUC categories to be 0.57 and 0.53, sug-
gested reducing the N/C ratio below the current TPS threshold of 0.7 for these cat-
egories (see Chaps. 5 and 6); however, the authors recognized that HGUC cases also
contained individual cells with >0.7 ratio [51]. Thus the criteria for HGUC do func-
tion effectively. Overall, it appears that utilizing the N/C ratio criterion of ≥0.5 is
not only feasible, but when present, this criterion also provides a high predictive
power for HGUC in AUC specimens, justifying it as a required criterion for the
AUC category. (See Diagnostic Decision Tree in the Introduction) [47, 52].
that are technique dependent, see Chap. 10. There is some evidence supporting the
common use of TPS criteria among preparation types, but more studies are needed
(see Chap. 10).
In preparation for the second edition of TPS, a survey querying utilization and func-
tionality of the system was compiled by TPS authors. The survey solicited recom-
mendations and suggestions for improvement from the participants. The survey was
sent in July 2020 to the American Society of Cytopathology and International
Academy of Cytology memberships via e-mail, website announcement, and social
media (Twitter and Facebook) for widespread distribution. The survey was open for
2 months following launch and gained responses from 523, of which, 451 passed
the initial screening question and were tabulated as participants. Answers were not
mandated; therefore, not all questions were answered by all participants. The vast
majority of the participants (344/370, 93%) reported processing urine specimens in
their own local cytology laboratories, and of those, a majority (214/261, 82%) were
using The Paris System to report urine cytology results in their institution. As seen
in the previous survey from the CAP, the most common preparation methods used
were ThinPrep (44.6%) followed by Cytospin (37%). The mean reported “atypia”
rates before and after implementation of TPS in the laboratories were 21.6% and
15.98% (p ≥ 0.01%), respectively. Regarding the question of how “atypical squa-
mous cells” should be categorized, 61% of the participants responded, “an atypical
category that is separate and distinct from the AUC category.” The majority of the
participants (156/215, 72.6%) agreed that the set of criteria for diagnosing voided
and instrumented urinary samples should be the same. Similarly, the majority of the
participants (167/215, 77.7%) agreed that the set of criteria for diagnosing lower
versus upper tract specimens should be the same. Again, the majority (nearly 70%)
of participants reported that the cytomorphologic criteria for the AUC category was
applied with relative ease, and was not restrictive [55].
Criteria
For the AUC category, a strict morphological definition and a description of the
characteristics of atypical urothelial cells are not feasible. AUC is multifactorial and
probabilistic as explained in the background paragraphs of this chapter, but the sav-
ing grace to the concept of AUC is that there are boundaries. Practitioners of the
cytologic method recognize that AUC represents a grey zone between reactive
changes and SHGUC, i.e., cytologic changes that fall short of SHGUC but are
beyond what is accepted for NHGUC. Degeneration in the cellular sample is always
a problem; the better the preservation, the more secure the interpretation. Diagnoses
of SHGUC and HGUC are best made on nondegenerated urothelial cells if at all
possible. Many samples interpreted as AUC may exhibit some degree of cellular
degeneration while eliciting worry for HGUC.
4 Atypical Urothelial Cells (AUC) 77
AUC is defined as cellular changes that fulfill the major (required) criterion in
addition to one minor criterion.
Based on the presence of the major criterion and one of the minor criteria noted
above, a diagnosis of AUC may be rendered. The morphology of normal basal uro-
thelial cells, typically observed in instrumented urine specimens, needs to be recog-
nized as “normal” or NHGUC despite the fact that they may have a high N/C ratio
and may appear mildly hyperchromatic (Fig. 4.4). These cells frequently occur in
groups and have uniform, round nuclei, and inconspicuous nucleoli with a finely
dispersed, smooth chromatin. Chapter 3 provides a detailed discussion of the cyto-
morphologic features of the NHGUC category. In cases where there are cells with
significant atypia, but due to extensive degeneration secondary to poor cellular pres-
ervation or obscuring elements a definitive diagnosis could not be rendered, the
diagnosis of AUC is a valid option.
Both the quality and quantity of atypical urothelial cells in a urine specimen are
important for the diagnosis. In a study reviewing subclassification of indeterminate
categories, cases with a benign/negative for HGUC outcome had an average of less
than 9 atypical urothelial cells, compared to cases with a “positive HGUC” outcome
in which >16 atypical urothelial cells were present [56]. Currently, there is no rec-
ommendation for counting the number of atypical urothelial cells for an AUC diag-
nosis; however, as the number of atypical cells with the described features increases,
so does the probability of malignancy.
Explanatory Notes
Explanatory Note 1: High N/C ratio HGUC cells typically show a high N/C ratio
approximating 0.7 [51]. For a diagnosis of AUC, the N/C ratio should be at least 0.5
[49]. The N/C ratio is increased due to nuclear enlargement as a result of increased
chromatin content relative to a smaller or no increase in cytoplasmic volume. If ele-
vated N/C ratio is the sole finding, i.e., there is no nuclear hyperchromasia, coarse
chromatin, or nuclear contour irregularity, then the case is unlikely to fit under the
AUC category but in NHGUC. A proviso is that any concern regarding HGUC is
78 G. A. Barkan et al.
a b
Fig. 4.18 Atypical urothelial cells (AUC). (a) A cohesive group of urothelial cells with high N/C
ratio, mild nuclear enlargement, and hyperchromasia on a Cytospin preparation. The chromatin
pattern is mostly uniform. (Bladder washing, Cytospin, high mag.). (b) Follow-up showed a low-
grade urothelial carcinoma on biopsy (urinary bladder biopsy, H&E, medium mag.)
a b
Fig. 4.19 (a–c) Atypical urothelial cells (AUC). (a) Single cells and groups of urothelial cells
with high N/C ratio and nuclear enlargement are depicted. (b–c) Compared with nearby unbrella
cells there are single urothelial cells with high N/C rtio and nuclear enlargement. However, the
chromatin is smooth and there is a lack of nuclear contour irregularity. The patient was post-BCG
therapy, and the follow-up urinary bladder biopsy showed reactive epithelial changes. (Catheterized
urine, Cytospin, high mag.)
4 Atypical Urothelial Cells (AUC) 79
Explanatory Note 3: Irregular nuclear contours Compared with the round shape
and smooth contours of the nuclei of normal urothelial cells, AUC usually have
irregular, variably thickened chromatin rims. The irregular contours and thickened
rims may relate to the variable amount of chromatin and lengths of DNA segments
attached to the nuclear envelope [44].
The “atypia” diagnosis, prior to TPS, was associated with a risk of subsequent
biopsy-proven HGUC that ranged from 8.3 to 37.5% [1, 3]. Based on results of the TPS
study group practice survey, we believe the rate of AUC category should not exceed
15%, and <10% should be an ultimate goal [55]. As further studies are performed uti-
lizing the criteria set forth in TPS, and as evidence-based data become available, there
will be further recommendations for the frequency and utility of the AUC interpreta-
tion. Post-TPS, the reported ROHM ranged from 24% to 53% (see Chap. 11).
Conclusion
Since the main aim of urine cytology according to TPS is to detect HGUC, a diag-
nosis of “atypia” is inappropriate for known benign cellular changes such as reac-
tive umbrella cells, polyomavirus or other viruses, granulomas, or changes due to
urolithiasis as discussed in Chap. 3. The target for AUC is HGUC, and only cells
worrisome for HGUC but lacking sufficient evidence for a more advanced
80 G. A. Barkan et al.
interpretation should be included in this category. With time and progressive bound-
ing of the AUC category, sensitivity and specificity for HGUC are expected to
improve, and thus increase the ROHM of patients in this diagnostic category while
simultaneously decreasing the rate of indeterminate cases. The exclusion of reac-
tive/reparative causes from AUC has resulted in a reduced atypia rate, sparing these
patients from an unnecessary neoplastic workup. Conversely, since TPS also
includes a “suspicious” category, SHGUC, we expect that some cases formerly
interpreted as AUC will show changes significant enough to be interpreted as
SHGUC, potentially resulting in a lower proportion of HGUC in the follow-up of
AUC. Thus, The Paris System intends to decrease the rate of “atypia” at both ends
of the diagnostic spectrum. For further discussion of SHGUC, see Chap. 5.
Cytologic reports that are unambiguous and issue a specific diagnostic category
mitigate confusion and anxiety among both clinical providers and patients and can
potentially help prevent unnecessary procedures. An article by Barkan, Wojcik, and
Pambuccian72 provides an overview and a roadmap to reduce atypia rates by outlin-
ing measures that can be taken by cytopathology practitioners [59].
Sample Reports
Example 1
Voided urine:
Adequacy: Satisfactory for evaluation
Interpretation: Atypical urothelial cells present
Example 2
Urine, bladder barbotage:
Adequacy: Satisfactory for evaluation
Interpretation: Atypical urothelial cells present. See note.
Note: The urine specimen contains rare urothelial cells with increased nuclear
cytoplasmic ratios, nuclear enlargement, and irregular nuclear membranes. The
paucity of these atypical urothelial cells limits further classification. According to
The Paris System for reporting urinary cytology, the “atypical urothelial cells” cat-
egory is associated with a 24–53% risk of high-grade malignancy. Follow-up is
recommended as clinically warranted.
References
1. Gopalakrishna A, et al. The diagnostic accuracy of urine-based tests for bladder cancer varies
greatly by patient. BMC Urol. 2016;16:30.
2. Bolenz C, West AM, Ortiz N, Kabbani W, Lotan Y. Urinary cytology for the detection of uro-
thelial carcinoma of the bladder—a flawed adjunct to cystoscopy? Urologic Oncol Seminars
Orig Investigations. 2013;31:366–71.
4 Atypical Urothelial Cells (AUC) 81
3. Barkan GA, et al. The Paris System for Reporting Urinary Cytology: the quest to develop a
standardized terminology. J Am Soc Cytopathol. 2016;5:177–88.
4. Rosenthal DL, Wojcik EM, Kurtycz D. The Paris system for reporting urinary cytology. 2016.
https://doi.org/10.1007/978-3-319-22864-8.pdf.
5. Anbardar MH, Monjazeb R. Reclassification of urinary cytology regarding The Paris System
for Reporting Urinary Cytology with cytohistological correlation demonstrates high sensitiv-
ity for high-grade urothelial carcinoma. Diagn Cytopathol. 2020;48:446–52.
6. Stanzione N, et al. The continual impact of the Paris System on urine cytology, a 3-year experi-
ence. Cytopathology. 2020;31:35–40.
7. Vosoughi A, et al. The Paris System “atypical urothelial cells” category: can the current criteria
be improved? J Am Soc Cytopathol. 2020;10:3–8.
8. Compton ML, Weiss VL, Barkan GA, Ely KA. Targeted education as a method for reinforc-
ing Paris System criteria and reducing urine cytology atypia rates. J Am Soc Cytopathol.
2020;10(1):9–13.
9. Granados R, Duarte JA, Corrales T, Camarmo E, Bajo P. Applying the Paris System for
Reporting Urine Cytology increases the rate of atypical urothelial cells in benign cases: a need
for patient management recommendations. Acta Cytol. 2017;61:71–6.
10. Brimo F, Auger M. The atypical urothelial cell category in the Paris System: strengthening the
Achilles’ heel. Cancer Cytopathol. 2016;124:305–6.
11. Glass RE, et al. Two-tiered subdivision of atypia on urine cytology can improve patient follow-
up and optimize the utility of UroVysion. Cancer Cytopathol. 2016;124:188–95.
12. Hassan M, et al. Impact of implementing the Paris System for Reporting Urine Cytology
in the performance of urine cytology: a correlative study of 124 cases. Am J Clin Pathol.
2016;146:384–90.
13. Joudi AM, Pambuccian SE, Wojcik EM, Barkan GA. The positive predictive value of “sus-
picious for high-grade urothelial carcinoma” in urinary tract cytology specimens: a single-
institution study of 665 cases. Cancer Cytopathol. 2016;124:811–9.
14. Miki Y, Neat M, Chandra A. Application of The Paris System to atypical urine cytology sam-
ples: correlation with histology and UroVysion((R)) FISH. Cytopathology. 2017;28:88–95.
15. Wang Y, Auger M, Kanber Y, Caglar D, Brimo F. Implementing The Paris System for
Reporting Urinary Cytology results in a decrease in the rate of the “atypical” category and an
increase in its prediction of subsequent high-grade urothelial carcinoma. Cancer Cytopathol.
2018;126:207–14.
16. Zheng X, et al. The Paris System for urine cytology in upper tract urothelial specimens: a
comparative analysis with biopsy and surgical resection. Cytopathology. 2018;29:184–8.
17. Suh J, et al. Modification of The Paris System for urinary tract washing specimens using diag-
nostic cytological features. Cytopathology. 2017;28:516–23.
18. Malviya K, Fernandes G, Naik L, Kothari K, Agnihotri M. Utility of the Paris System in
Reporting Urine Cytology. Acta Cytol. 2017;61:145–52.
19. Rezaee N, Tabatabai ZL, Olson MT. Adequacy of voided urine specimens prepared by
ThinPrep and evaluated using The Paris System for Reporting Urinary Cytology. J Am Soc
Cytopathol. 2017;6:155–61.
20. Torous VF, Brancely D, VanderLaan PA. Implementation of the Paris System for Reporting
Urinary Cytology results in lower atypical diagnostic rates. J Am Soc Cytopathol.
2017;6:205–10.
21. Mikou P, et al. Evaluation of the Paris System in atypical urinary cytology. Cytopathology.
2018;29:545–9.
22. Rohilla M, et al. Cytohistological correlation of urine cytology in a tertiary centre with appli-
cation of the Paris system. Cytopathology. 2018;29:436–43.
23. Zare S, et al. A single institutional experience with the Paris System for Reporting Urinary
Cytology: correlation of cytology and histology in 194 cases. Am J Clin Pathol. 2018;150:162–7.
24. VandenBussche CJ, Allison DB, Gupta M, Ali SZ, Rosenthal DL. A 20-year and
46,000-specimen journey to Paris reveals the influence of reporting systems and passive peer
feedback on pathologist practice patterns. Cancer Cytopathol. 2018;126:381–9.
82 G. A. Barkan et al.
25. Meilleroux J, et al. One year of experience using the Paris System for Reporting Urinary
Cytology. Cancer Cytopathol. 2018;126:430–6.
26. Bertsch EC, Siddiqui MT, Ellis CL. The Paris system for reporting urinary cytology improves
correlation with surgical pathology biopsy diagnoses of the lower urinary tract. Diagn
Cytopathol. 2018;46:221–7.
27. Xing J, Monaco SE, Pantanowitz L. Utility of The Paris System for Reporting Urinary
Cytology in upper urinary tract specimens. J Am Soc Cytopathol. 2018;7:311–7.
28. Cowan ML, VandenBussche CJ. The Paris System for Reporting Urinary Cytology: early
review of the literature reveals successes and rare shortcomings. J Am Soc Cytopathol.
2018;7:185–94.
29. Kurtycz DFI, et al. Paris Interobserver Reproducibility Study (PIRST). J Am Soc Cytopathol.
2018;7:174–84.
30. Richardson CJ, Pambuccian SE, Barkan GA. Split-sample comparison of urothelial cells in
ThinPrep and cytospin preparations in urinary cytology: do we need to adjust The Paris System
for Reporting Urinary Cytology criteria? Cancer Cytopathol. 2020;128:119–25.
31. Bakkar R, et al. Impact of the Paris system for reporting urine cytopathology on predictive
values of the equivocal diagnostic categories and interobserver agreement. Cytojournal.
2019;16:21.
32. Vlajnic T, Gut A, Savic S, Bubendorf L. The Paris System for reporting urinary cytology
in daily practice with emphasis on ancillary testing by multiprobe FISH. J Clin Pathol.
2020;73:90–5.
33. Barkan GA, et al. Practice patterns in urinary cytopathology prior to the Paris System for
Reporting Urinary Cytology. Arch Pathol Lab Med. 2019;144:172–6.
34. McIntire PJ, et al. Negative predictive value and sensitivity of urine cytology prior to
implementation of The Paris System for Reporting Urinary Cytology. Cancer Cytopathol.
2019;127:125–31.
35. Abro S, et al. Outcome analysis and negative predictive value of the “unsatisfactory/nondiag-
nostic” category of The Paris System for Reporting Urinary Cytology. J Am Soc Cytopathol.
2021;10:64–70.
36. McIntire PJ, Kilic I, Pambuccian SE, Wojcik EM, Barkan GA. The Paris System for Reporting
Urinary Cytology reduces atypia rates and does not alter the negative predictive value of urine
cytology. J Am Soc Cytopathol. 2021;10:14–9.
37. Danakas A, Sweeney M, Cheris S, Agrawal T. Urinary tract cytology: a cytologic-
histopathologic correlation with The Paris System, an institutional study. J Am Soc Cytopathol.
2021;10:56–63.
38. Myles N, et al. Evidence-based diagnostic accuracy measurement in urine cytology using like-
lihood ratios. J Am Soc Cytopathol. 2021;10:71–8.
39. Sahai R, et al. Interobserver reproducibility of The Paris System of Reporting Urine Cytology
on cytocentrifuged samples. Diagn Cytopathol. 2020;48(11):979–85; https://doi.org/10.1002/
dc.24476.
40. Pastorello RG, Barkan GA, Saieg M. Experience on the use of The Paris System for
Reporting Urinary Cytopathology: review of the published literature. J Am Soc Cytopathol.
2021;10:79–87.
41. Galen RS, Gambino RS. Beyond normality: the predictive value and efficiency of medical
diagnoses. Wiley; 1975.
42. Pambuccian SE. What is atypia? Use, misuse and overuse of the term atypia in diagnostic
cytopathology. J Am Soc Cytopathol. n.d.;4:44–52.
43. Melamed MR, Wolinska WH. On the significance of intracytoplasmic inclusions in the urinary
sediment. Am J Pathol. 1961;38:711–9.
44. Frost JK. The cell in health and disease. An evaluation of cellular morphologic expression of
biologic behavior. 2nd, revised edition. Monogr Clin Cytol. 1986;2:1–304.
45. Raistrick J, Shambayati B, Dunsmuir W. Collection fluid helps preservation in voided urine
cytology. Cytopathology. 2008;19:111–7.
4 Atypical Urothelial Cells (AUC) 83
46. Ahmed HG, Tom MA. The consequence of delayed fixation on subsequent preservation of
urine cells. Oman Med J. 2011;26:14–8.
47. Zhang ML, Guo AX, VandenBussche CJ. Morphologists overestimate the nuclear-to-
cytoplasmic ratio. Cancer Cytopathol. 2016;124:669–77.
48. Vaickus LJ, Tambouret RH. Young investigator challenge: the accuracy of the nuclear-
to-
cytoplasmic ratio estimation among trained morphologists. Cancer Cytopathol.
2015;123:524–30.
49. Hang JF, Charu V, Zhang ML, VandenBussche CJ. Digital image analysis supports a nuclear-
to-cytoplasmic ratio cutoff value of 0.5 for atypical urothelial cells. Cancer Cytopathol.
2017;125:710–6.
50. Long T, et al. Interobserver reproducibility of The Paris System for Reporting Urinary
Cytology. Cytojournal. 2017;14:17.
51. McIntire PJ, et al. Digital image analysis supports a nuclear-to-cytoplasmic ratio cutoff value
below 0.7 for positive for high-grade urothelial carcinoma and suspicious for high-grade uro-
thelial carcinoma in urine cytology specimens. Cancer Cytopathol. 2019;127:120–4.
52. Barkan GA, Wojcik EM. Genitourinary cytopathology (kidney and urinary tract). Canc Treat.
2014;160:149–83.
53. McIntire PJ, Elsoukkary SS, Robinson BD, Siddiqui MT. High-grade urothelial carcinoma in
urine cytology: different spaces – different faces, highlighting morphologic variance. J Am Soc
Cytopathol. 2021;10:36–40.
54. Zhang ML, et al. A review of urinary cytology in the setting of upper tract urothelial carci-
noma. J Am Soc Cytopathol. 2020;10:29–35.
55. Kurtycz D, et al. Perceptions of Paris: an international survey in preparation for The Paris
System for Reporting Urinary Cytology 2.0 (TPS 2.0). J Am Soc Cytopathol. 2021;10(5):S4.
56. McCroskey Z, Bahar B, Hu Z, Wojcik EM, Barkan GA. Subclassifying atypia in urine cytol-
ogy: what are the helpful features? J Am Soc Cytopathol. 2015;4:183–9.
57. Renshaw AA, Gould EW. High-grade urothelial carcinoma with hypochromatic chromatin in
urine cytology. J Am Soc Cytopathol. 2021;10:25–8.
58. Pierconti F, et al. Hypochromatic large urothelial cells in urine cytology are indicative of high
grade urothelial carcinoma. APMIS. 2018;126:705–9.
59. Barkan GA, Wojcik EM, Pambuccian SE. A tale of atypia: what can we learn from this?
Cancer Cytopathol. 2018;126:376–80.
60. Rai S, et al. A quest for accuracy: evaluation of The Paris System in diagnosis of urothelial
carcinomas. J Cytol. 2019;36:169–73.
61. Vaheda Begam K, Vallamreddy SKR, Pratima J. Implementation of the Paris system versus
institutional diagnosis in the performance of urinary cytology: a 5 years correlative study of 74
cases. IP Arch Cytol Histopathol Res. 2019;4:193–8.
62. Roy M, et al. An institutional experience with The Paris System: a paradigm shift from ambig-
uous terminology to more objective criteria for reporting urine cytology. Cytopathol Off J Br
Soc Clin Cytol. 2017;28:509–15.
Suspicious for High-Grade Urothelial
Carcinoma (SHGUC) 5
Fadi Brimo, Manon Auger, Ivan Chebib, Tarik M. Elsheikh,
Mitsuru Kinjo, Eric Piaton, Marc Pusztaszeri,
and Tatsuro Shimokama
Background
The diagnosis of “suspicious for HGUC” (SHGUC) is meant to reflect the presence
of urothelial cells with severe atypia that falls short for a diagnosis of high-grade
urothelial carcinoma (HGUC) but beyond atypia that is associated with the “atypi-
cal urothelial cells” (AUC) category. Since the publication of the first edition, many
studies have confirmed that a cytologic diagnosis of SHGUC confers a risk for
HGUC that falls in between that of AUC and HGUC cytological diagnoses, provid-
ing justification for keeping it as a separate diagnostic category [1–9].
Definition
This diagnosis is restrictively used in cases with abnormal urothelial cells that quan-
titatively fall short of a definitive diagnosis of HGUC.
Criteria
A diagnosis of SHGUC (Figs. 5.1, 5.2, 5.3, 5.4, 5.5, 5.6, 5.7, 5.8, and 5.9) is defined
as non-superficial and nondegenerated urothelial cells showing increased nuclear
to cytoplasmic (N/C) ratio, at least 0.7, and in addition, at least two of the three
following features:
Until further data become available, the same qualitative morphological criteria
should be applied to different specimen types (voided and instrumented; lower tract
and upper tract). Using all the features listed above, the decision to assign a case
into the SHGUC or the HGUC cytological categories is mainly based on the number
of abnormal cells fulfilling the above criteria. Due to the lack of definitive data
derived from studies specifically addressing this quantitative issue, a strict cutoff
number of abnormal cells above which one can confidently assign a HGUC diagno-
sis cannot be determined [10–11].
Fig. 5.4 SHGUC. A cell cluster composed of six abnormal well-preserved intermediate urothelial
cells displays increased N/C ratios, hyperchromasia, clumpy chromatin, and irregular nuclear
membranes. Note that despite similar nuclear characteristics, there is a range of N/C ratios from
0.5 to 0.8. An HGUC diagnosis may be appropriate in this case, especially in the presence of a
previous history of HGUC. (Catheterized urine, CS, high mag)
88 F. Brimo et al.
Fig. 5.8 SHGUC. One abnormal well-preserved intermediate urothelial cells has increased N/C
ratio, severe hyperchromasia, irregular nuclear membranes but no easily evaluable chromatin pat-
tern. The background contains smaller degenerated cells with hyperchromatic nuclei, red cell
membranes, and necrotic debris, suggesting an invasive lesion. The follow-up biopsy was invasive
high-grade urothelial carcinoma (voided urine, CS, high mag)
Instead, a cutoff range of at least five to ten cells for lower tract samples and at least
ten cells for the upper tract is recommended based on the degree of abnormal nuclear
changes observed. This approach also reflects the results of the first TPS survey in
which 90% of participants reported preferring a range of five to ten cells rather than
setting up a rigid arbitrary cutoff [12]. Accordingly, a HGUC diagnosis should very
rarely be assigned in the presence of less than five abnormal well-preserved cells. In
comparison, in the presence of five to ten abnormal cells, the decision to assign a
“positive for HGUC” diagnosis should take into account the severity of atypia of all
the abnormal cells, the clinical context, as well as the specimen type. For example, in
the presence of a previous history of HGUC and/or in voided specimens which are
by nature less cellular than instrumented specimens and in which cellular degenera-
tion is frequent, as low as five well-preserved and severely abnormal cells with
the features listed above may be sufficient to render a definitive HGUC diagnosis.
90 F. Brimo et al.
On the other hand, at least ten abnormal cells are needed before labelling a case as
HGUC in instrumented specimens derived from the upper urinary tract (see Chaps. 6
and 7 for further discussion).
• The cells are usually seen as single cells although clusters of atypical cells may
also be present.
• Features that may be seen but do not necessarily need to be present are:
–– Eccentric nuclear location (Fig. 5.1).
–– Prominent nucleolus.
–– Necrotic background.
–– Pleomorphism.
–– Mitoses.
–– Apoptotic bodies.
Explanatory Notes
Explanatory Note 3 In the absence of clear and evaluable chromatin details, the
irregular clumpy chromatin pattern is not required in the presence of the other fea-
tures (high N/C ratio, nucleomegaly, irregular nuclear membranes, hyperchromasia)
(Figs. 5.5, and 5.8). Similarly, the presence of nuclear membrane irregularity is not
required in the presence of the other features (increased N/C ratio, nucleomegaly,
hyperchromasia, irregular clumpy chromatin) (Figs. 5.6, and 5.7).
Explanatory Note 4 Intermediate urothelial cells with increased N/C ratio, nucleo-
megaly, and mild hyperchromasia in the absence of evaluable chromatin details and
irregular nuclear membranes should not be labelled as “suspicious” but as AUC
instead. Similarly, cells with increased N/C ratio and irregular nuclear membranes
in the absence of moderate or severe hyperchromasia should generally not be
labelled as “suspicious” (Figs. 5.11, and 5.12).
5 Suspicious for High-Grade Urothelial Carcinoma (SHGUC) 91
Explanatory Note 5 While the category of AUC includes a subset of cases showing
cellular degeneration, a SHGUC diagnosis should not be rendered solely on degen-
erated cells, although some of the severely atypical cells may be degenerated.
Cellular degeneration is often present in voided specimens and can take the form of
incomplete cytoplasm, poorly preserved chromatin details, or discontinuous nuclear
membranes. The resulting altered cellular morphology can be problematic from the
diagnostic standpoint for the following reasons:
• Nuclei may look “blown-up” resulting in a falsely increased N/C ratio (Figs. 5.13,
and 5.14).
• Cytoplasm may be incomplete which makes it difficult to assess the N/C ratio,
often making it appear increased (Fig. 5.14).
• Nuclear membranes may seem irregular from dehydration.
• Nuclei may look hyperchromatic as a feature of degeneration, and not as a result
of an abnormal chromatin (Figs. 5.13, and 5.14).
Fig. 5.10 AUC. Cell clusters of well-preserved intermediate urothelial cells demonstrate variable
N/C ratios and hyperchromasia. The degree of hyperchromasia is mild in comparison to the normal
intermediate cell nucleus (upper right). In addition, the cells do not show clumpy chromatin pattern
or irregular nuclear membranes which preclude the assignment of a SHGUC diagnosis. (Voided
urine, TP, high mag)
92 F. Brimo et al.
Fig. 5.13 AUC. One single and degenerated urothelial cell has an enlarged nucleus, with increased
N/C ratio, mild hyperchromasia, and irregularly distributed clumpy chromatin. Note the presence
of incomplete cytoplasm and poorly preserved chromatin details. A SHGUC diagnosis should not
be rendered on degenerated cells only. However, atypical degenerated hyperchromatic cells are
often seen in the background of cases diagnosed as SHGUC or HGUC. Consequently, the presence
of degenerated atypical hyperchromatic cells warrants careful evaluation in order to rule out the
presence of similar non-degenerated cells in which a SHGUC or a HGUC diagnosis can be ren-
dered. (Voided urine, TP, high mag)
a b
Fig. 5.14 (a) AUC. One single and degenerated urothelial cell displays an increased N/C ratio,
hyperchromasia, and irregularly distributed clumpy chromatin. Note the presence of incomplete
cytoplasm and discontinuous nuclear membranes. In the presence of cellular degeneration, a
SHGUC diagnosis should not be rendered. (Voided urine, TP, high mag.) (b) AUC. One single and
degenerated urothelial cell shows increased N/C ratio, hyperchromasia, and irregularly distributed
clumpy chromatin. Note the presence of incomplete cytoplasm and discontinuous nuclear mem-
branes (Voided urine, TP, high mag)
94 F. Brimo et al.
Fig. 5.15 NHGUC.
Reactive urothelial cells
often show increased N/C
ratios as seen in the current
example. However,
hyperchromasia is absent,
and the nuclei show fine
regularly distributed
chromatin with small
visible nucleoli. Nuclear
membranes are smooth and
regular. (Voided urine, TP,
high mag)
Prevalence of SHGUC
The reported prevalence of SHGUC among all cytologic diagnoses ranges from
0.2% to 6.6% depending on the study [13–15]. As such, suspicious specimens rep-
resent a small percentage of overall cases across studies, and this is likely due to the
strict criteria in TPS for the use of the SHGUC category, leaving this diagnosis only
for cases with a limited number of malignant cells.
Risk of Malignancy
Studies using TPS classification of urine cytology have shown that the risk of high-
grade malignancy (ROHM) that is associated with or subsequent to a diagnosis of
SHGUC ranges from 33.3% to 100%, which is intermediate between that of AUC
(range: 12.3–60.9%) and HGUC (range: 58.8–100%) diagnoses [1–9, 13–15] . In
the four studies with the largest number of patients with histologic follow-up,
ROHMs for the SHGUC category were between 80 and 89%, while they were
between 32.6% and 49% for AUC and all studies above 90% for HGUC [2, 5, 11,
13]. Taking into account all large published studies using TPS, the mean ROHM
weighted by sample size is 76.2% for the SHGUC category, in comparison to 38.5%
and 88.8% for the AUC and HGUC categories, respectively (see Chap. 11 for fur-
ther discussion).
5 Suspicious for High-Grade Urothelial Carcinoma (SHGUC) 95
Management
References
1. Stanzione N, Ahmed T, Fung PC, Cai D, Lu DY, Sumida LC, Moatamed NA. The con-
tinual impact of the Paris system on urine cytology, a 3-year experience. Cytopathology.
2020;31(1):35–40. PMID: 31596979.
2. de Paula R, Oliveira A, Nunes W, Bovolim G, Domingos T, De Brot L, Bezerra S, Cunha I,
Morini M, Saieg M. Two-year study on the application of the Paris system for urinary cytology
in a cancer centre. Cytopathology. 2020;31(1):41–6. PMID:
3. Wang Y, Auger M, Kanber Y, Caglar D, Brimo F. Implementing the Paris system for reporting
urinary cytology results in a decrease in the rate of the “atypical” category and an increase
in its prediction of subsequent high-grade urothelial carcinoma. Cancer Cytopathol. 2018;
PMID: 29278461.
4. Rohilla M, Singh P, Rajwanshi A, Gupta N, Srinivasan R, Dey P, Kakkar N. Cytohistological
correlation of urine cytology in a tertiary centre with application of the Paris system.
Cytopathology. 2018;29(5):436–43. PMID: 29920811.
5. Hassan M, Solanki S, Kassouf W, Kanber Y, Caglar D, Auger M, Brimo F. Impact of imple-
menting the Paris system for reporting urine cytology in the performance of urine cytology.
Am J Clin Pathol. 2016;146(3):384–90. PMID: 27543983.
6. Xing J, Monaco S, Pantanowitz L. The ability of the Paris system to stratify the risk of high-
grade urothelial carcinoma. J Am Soc Cytopathol [Internet]. Elsevier Inc; 2018;7(5):S43.
Aveailable from: https://doi.org/10.1016/j.jasc.2018.06.102.
7. Nguyen L, Nilforoushan N, Krane JF, Bose S, Bakkar R. Should "suspicious for high-grade
urothelial carcinoma" and "positive for high-grade urothelial carcinoma" remain separate cat-
egories?. Cancer Cytopathol. 2020. PMID: 33036060.
8. Zare S, Mirsadraei L, Reisian N, Liao X, Roma A, Shabaik A, Hasteh F. A single institutional
experience with the Paris system for reporting urinary cytology: correlation of cytology and
histology in 194 cases. Am J Clin Pathol. 2018;150(2):162–7. PMID: 29878037.
9. Granados R, Duarte JA, Corrales T, Camarmo E, Bajo P. Applying the Paris system for report-
ing urine cytology increases the rate of atypical urothelial cells in benign cases: a need for
patient management recommendations. Acta Cytol. 2017;61(1):71–6. PMID: 27838683.
10. Brimo F, Xu B, Kassouf W, Ahmadi-Kaliji B, Charbonneau M, Nahal A, Kanber Y, Caglar
D, Auger M. Urine cytology: does the number of atypical urothelial cells matter? A quali-
tative and quantitative study of 112 cases. J Am Soc Cytopathol. 2015;4(4):232–8. PMID:
31051759.
11. McCroskey Z, Bahar B, Hu Z, Wojcik E, Barkan G. Subclassifying atypia in urine cytology:
what are the helpful features? J Am Soc Cytopathol. 2015;4(4):183–9. PMID: 31051752.
12. Daniel F. I. Kurtycz, Fadi Brimo, Dorothy L. Rosenthal, Momin T. Siddiqui, Z. Laura Tabatabai,
Christopher J. VandenBussche, Eva M. Wojcik, Güliz A. Barkan, Perceptions of Paris: An
international survey in preparation for The Paris System for Reporting Urinary Cytology 2.0
(TPS 2.0), accepted, American Society of Cytopathology’s 66rd Annual Scientific Meeting, Las
Vegas, Nevada, November 11–14, 2021.
96 F. Brimo et al.
13. Meilleroux J, Daniel G, Aziza J, d’Aure DM, Quintyn-Ranty ML, Basset CML, Evrard SM,
Courtade-Saidi MM. One year of experience using the Paris system for reporting urinary cytol-
ogy. Cancer Cytopathol. 2018;126(6):430–6. PMID: 29663682.
14. Bakkar R, Mirocha J, Fan X, Frishberg DP, Peralta-Venturina M, de Zhai J, Bose S. Impact
of the Paris system for reporting urine cytopathology on predictive values of the equivocal
diagnostic categories and interobserver agreement. Cytojournal. 2019;16:21.
15. Pastorello RG, Barkan GA, Saieg M. Experience on the use of The Paris System for
Reporting Urinary Cytopathology: review of the published literature. J Am Soc Cytopathol.
2020:S2213–2945(20)30323–9.
High-Grade Urothelial Carcinoma
(HGUC) 6
Momin T. Siddiqui, Derek B. Allison, Guido Fadda,
Jee-Young Han, Patrick J. McIntire, Christopher L. Owens,
Z. Laura Tabatabai, Toyonori Tsuzuki, and Mingjuan Lisa Zhang
• Criteria for cytologic diagnosis of HGUC are internationally accepted and suc-
cessfully applied.
• Spectrum of N/C ratios is noted in malignant cells of HGUC.
• Hypochromasia is recognized as a rare occurrence and as a potential pitfall in
diagnosis of HGUC.
• Degenerative changes in HGUC are addressed.
• HGUC sub-types are described.
• Fibrovascular cores in HGUC are considered.
Background
In the first edition of TPS, the recommended criteria for diagnosing HGUC in urine
samples included a minimum of five to ten malignant cells. This range was estab-
lished to accommodate varying experience, patient populations and previous his-
tory, as well as type of specimen. Considering potential significant consequences of
HGUC diagnosis in the upper tract, higher number of malignant cells (>10) has
been recommended from upper tract specimens. The criteria for malignant cells
included an N/C ratio of 0.7 or greater, i.e., the cross-sectional area of the nucleus
occupying equal to or more than 70% of the cross-sectional area of the cell size,
nuclear hyperchromasia, irregular nuclear membranes, and coarse nuclear chroma-
tin [1]. Diagnosing HGUC is recognized as the gold standard for TPS, and hence the
above criteria have been widely applied by laboratories processing and reporting
urine cytology [2–4].
Introduction
The TPS criteria for HGUC have been widely utilized for reporting urinary cytology
across the globe [2–7]. This fact has been affirmed by peer-reviewed publications
and a 2019 survey/poll conducted by the American Society of Cytopathology [5–7].
In these large cohort studies, with the largest number of patients with histologic
follow-up, the risk of high-grade malignancy for the HGUC category is all above
90%, confirming the high positive predictive value for HGUC by applying TPS
criteria [8–10]. This achievement is notable. And as such, the recommended cyto-
morphologic criteria remain unchanged in the second edition of TPS (Figs. 6.1, 6.2,
6.3, 6.4, 6.5, 6.6, 6.7, 6.8, 6.9, 6.10, 6.11, and 6.12).
Fig. 6.1 HGUC
demonstrating a wide
range of cell sizes and
shapes. Nuclear
hyperchromasia is
pronounced (voided urine,
TP, low mag)
Fig. 6.2 HGUC with numerous malignant cells displaying heterogeneity. The tumor cells vary in
size and shape and are arranged in clusters and single cells. Necrosis is present in the background
(voided urine, SP, low mag)
6 High-Grade Urothelial Carcinoma (HGUC) 99
a b
Fig. 6.3 (a) Malignant cells are arranged in small groups and as single cells. Pleomorphism and
background necrosis are also present (voided urine, SP, medium mag.). (b) Malignant cells with
high N/C ratios and irregular nuclear membranes. The sample is stained lightly, so observers are
cautioned to use benign cells in the background as stain intensity controls. Also, note the presence
of lymphocytes, which can be used as a control for nuclear size (washing, TP, medium mag)
Fig. 6.7 Hyperchromatic
HGUC cells presenting as
a cohesive group. The
majority of tumor cells
have high N/C ratios
(bladder washing, TP,
high mag)
6 High-Grade Urothelial Carcinoma (HGUC) 101
Fig. 6.8 Nuclear
membrane irregularity and
focal thickness of nuclear
membranes are evident in
this group of HGUC cells
(bladder washing, TP,
high mag)
Fig. 6.10 Coarse
chromatin and nuclear
membrane irregularity is
present in HGUC cells
(bladder washing, TP,
high mag)
102 M. T. Siddiqui et al.
Fig. 6.11 Nuclear
membrane irregularity,
hyperchromasia, coarse
chromatin, and
cytoplasmic vacuolation
are present in HGUC. The
vacuoles probably reflect
degeneration but may also
be found in cells with
glandular differentiation.
The follow-up biopsy
confirmed the diagnosis of
HGUC with glandular
differentiation (bladder
washing, TP, high mag)
Reporting the TPS criteria for HGUC has resulted in a very high specificity, i.e.,
decreased false positives, for the diagnosis of HGUC [5]. These criteria may result
in a decrease in the diagnosis of HGUC, especially in degenerated specimens, in
which malignant cells can exhibit a reduction in the N/C ratio to less than 0.7 [5,
11]. Such scenarios lead to malignant cells being under-classified. Likewise, the
high specificity of the criteria may help shift a subset of cases from a suspicious to
a malignant diagnosis [5, 11]. Thus, the high specificity for HGUC ultimately
results in improved risk stratification for patients based on future risk of malignancy
[11]. Malignant cells with degenerative changes and a lower N/C ratio are addressed
in a later section as a variation from the recognized criteria for HGUC.
6 High-Grade Urothelial Carcinoma (HGUC) 103
Hyperchromasia is one of the criteria for diagnosing HGUC [1]. However, after
the publication of TPS first edition, hypochromasia has been reported as a rare fea-
ture in HGUC cells in cytologic samples [12, 13]. Although, unusual, this is an
important variance to be recognized, particularly in instances where it is identified
in all of the malignant cells in a given specimen. Similarly, the first edition of TPS
did not address all of the known sub-types of HGUC, most of which are rarely
encountered. In this second edition, examples of the more aggressive sub-types,
such as plasmacytoid and micropapillary HGUC, are illustrated and discussed [14,
15]. TPS remains focused on a broader discussion of the morphologic criteria of
HGUC with a goal to be able to diagnose as many cases as possible and not be
detracted by potential outliers. Readers are encouraged to keep abreast of peer-
reviewed literature for infrequently encountered morphologic findings and incorpo-
rate them in the diagnosis, if clinically appropriate.
• Cellularity: At least five to ten malignant cells, for lower and upper tract speci-
mens, respectively.
• N/C ratio: 0.7 or greater (cross-sectional area of the nucleus occupying equal to
or more than 70% of the cross-sectional area of the cell size).
• Nucleus: Moderate to severe hyperchromasia.
• Nuclear membrane: Irregular in outline.
• Chromatin: Coarse/clumped.
• Cellular pleomorphism.
• Variation in size and shape of malignant cells, i.e., oval, round, elongated, and
spindle.
• Dense or vacuolated cytoplasm.
• Prominent nucleoli.
• Mitoses.
• Necrosis.
• Inflammation.
N/C Ratio
HGUC can display a very wide range of N/C ratios in malignant cells (Figs. 6.13,
6.14) . However, an increased N/C ratio of at least 0.7 is a recommended benchmark
for diagnosing HGUC. Although this criterion is arguably the most restrictive, it
accomplishes the goal of TPS by producing a very high specificity for urinary
104 M. T. Siddiqui et al.
cytology [5, 11]. In most cases, the majority of HGUC cells will exhibit N/C ratios
greater than 0.7 but may also display a very wide range of N/C ratios, with a signifi-
cant proportion of malignant cells having a N/C ratio between 0.5 and 0.7.
Occasionally, malignant cells can be large with abundant cytoplasm, displaying N/C
ratios of <0.5, which even falls below the threshold for the AUC category (Figs. 6.15,
6.16, and 6.17) [14]. As a result, the use N/C ratio > 0.7 as a requirement for a
malignant diagnosis has been a topic of much debate and further brings up the issue
of a morphologist’s ability to assess N/C ratios accurately and reproducibly. In
almost all samples from HGUC, there will be enough cells meeting criteria with
N/C ratios equal to or greater than 0.7 to make the interpretation with confidence.
Fig. 6.13 Loosely
cohesive HGUC tumor
cells with a very wide
range of N/C ratios. The
nuclear features are helpful
in confirming the
diagnoses of malignant
cells with an N/C ratio
lower than 0.7 (voided
urine, TP, high mag)
The literature shows variable results. For example, Vaickus et al. found that mor-
phologists were relatively accurate in estimating N/C ratios without a measuring
device with greater accuracy achieved as the N/C ratio increased [16]. In contrast,
Layfield et al. concluded the opposite [17]. Additionally, Zhang et al. showed a
trend toward N/C ratio overestimation, especially for values approaching the cutoff
points of 0.5 and 0.7 [18]. As a result, digital image analysis (DIA) has the opportu-
nity to more accurately assess the N/C ratio, but studies are very limited on what the
actual cutoff values should be and how that would translate into traditional morpho-
logical assessment. McIntire et al. found that the HGUC and SHGUC categories
revealed cells with average N/C ratios of 0.57 and 0.53, respectively, suggesting that
the current cutoff value of 0.7 is too high [19]. Yet, this study also noted that malig-
nant cells with N/C ratios higher than 0.7 were also present in the samples. Even if
ideal N/C ratio cutoff values are determined by DIA, these values may not be mor-
phologically distinguishable to the unassisted eye and may not be meaningful, espe-
cially if morphologists tend to overestimate N/C ratios. When otherwise overtly
malignant features are present, where the cells of interest all have N/C ratios <0.7,
those cases can be placed in the “suspicious” category, thereby preserving high
specificity, but also indicating to the treating physician the likelihood of malignancy.
Hypochromasia
HGUC can present with hypochromatic nuclei (Figs. 6.18 and 6.19), usually as a
small proportion of malignant cells along with those exhibiting hyperchromatic
nuclei, which still adheres to the TPS criteria. It is rare to see hypochromatic malig-
nant cells as a solitary feature [12, 13]. Cells with hypochromatic nuclei can be
diagnostic for HGUC based on increased nuclear size, high N/C ratio (>0.7), irregu-
lar nuclear membranes, and coarse (frequently peripheral) chromatin in the absence
of hyperchromasia [12, 13].
Fig. 6.18 Hypochromatic
HGUC tumor cells are
rarely present as a solitary
finding. High N/C ratios
and nuclear membrane
irregularity are also present
to confirm the diagnosis
(voided urine, CS,
medium mag)
6 High-Grade Urothelial Carcinoma (HGUC) 107
Fig. 6.19 Hypochromatic
HGUC tumor cells
predominate along with
rare hyperchromatic
HGUC cells (voided urine,
CS, medium mag)
Degenerative Changes
Degenerative changes are particularly identified in voided urine cytology samples,
where cells are free floating and separated from their nutrient supply for an unknown
period of time. TPS criteria recommend against the assessment of degenerated cells
for inclusion in any category since altered morphology may be misinterpreted, des-
ignating the sample incorrectly [1]. These degenerative changes may include a loss
of cytoplasm and nuclei that appear “blown-up” and enlarged, resulting in increased
N/C ratios. In contrast, some cells may actually pick up additional cytoplasmic vol-
ume in the form of bubbly vacuoles, resulting in a decrease in N/C ratios (Figs. 6.20
and 6.21). Furthermore, nuclear membranes may become irregular because of dehy-
dration or changes in osmotic forces, and nuclei may appear hyperchromatic due to
pyknosis rather than to abnormal chromatin. Therefore, cells with incomplete cyto-
plasm, discontinuous nuclear membranes, and poorly preserved chromatin should
be excluded from evaluation for HGUC and SHGUC categories. Conversely, HGUC
cells may also display degenerative features that can masquerade as benign changes,
such as mimicking BK viral cytopathic effect [5, 20]. Not uncommonly, cells with
overt malignant features may also include one or more degenerative changes; how-
ever, these cells should not count toward the minimum of five to ten nondegenerated
malignant cells to make a diagnosis of HGUC.
108 M. T. Siddiqui et al.
Fig. 6.20 Degenerated
tumor cells with
cytoplasmic vacuolation
and frayed cell membranes
(bladder washing, TP,
medium mag)
Fig. 6.21 Degenerated
tumor cells with irregular
nuclear membranes and
pyknosis (bladder
washing, TP, medium mag)
There are 13 histologic variants of HGUC based on the 2016 WHO classification of
tumors of the urinary system [22]. These can pose a diagnostic challenge in urinary
cytology specimens because the classic features of HGUC, by definition in some
cases, may be absent. TPS has created an understandable diagnostic model, but
some variants of HGUC may not conform to the “required” features, potentially
resulting in a false-negative diagnosis. Fortunately, these variants are extremely rare
and are often associated with conventional components displaying cytomorphologic
features adherent to TPS criteria.
HGUC with squamous differentiation is defined by the presence of keratinization
and/or intercellular bridges as classic morphological features (Figs. 6.22 and 6.23).
Squamous cells are intermixed with malignant cells exhibiting classic features of
HGUC [1, 23]. The squamous cells may display hyperchromatic and spindle nuclei
with clumped chromatin. The cytoplasm is dense, keratinized, and orangeophilic.
Keratin flakes and necrosis are frequently observed in the background [1]. Similarly,
HGUC with glandular differentiation shows the presence of true glandular
formation within groups of tumor cells [1, 23]. Mucin-filled goblet cells with hyper-
chromatic nuclei and coarse chromatin are present (Figs. 6.24 and 6.25). These can
be intermixed with malignant cells exhibiting classic features of HGUC [1, 23].
The nested variant of urothelial carcinoma is a rare malignancy characterized by
bland morphology and a deep and widely invasive pattern of growth. As a result,
even on biopsy, it is often difficult to distinguish superficially invasive nested vari-
ant urothelial carcinoma from von Brunn nests. In a study by Cardillo et al., the
authors were able to retrospectively identify cells in the urine that had a similar
morphology to the patient’s nested variant tumors on resection [24]. These features
included medium-sized, round-to-polygonal cells with abundant granular-to-dense
cytoplasm with distinct cell borders, slightly enlarged N/C ratios, irregular nuclear
contours, and occasional prominent nucleoli. However, given the subtle nature of
these findings, which may overlap with reactive changes, they recommended against
making this diagnosis on urinary cytology samples [24].
Micropapillary urothelial carcinoma is rare variant of HGUC. As a result, it is
important to recognize and sub-type this variant on histology, which, in addition to
Fig. 6.24 Mucin-filled
goblet cells with
hyperchromatic nuclei and
coarse chromatin are
present in HGUC with
glandular differentiation
(bladder washing, TP,
high mag)
high-grade cytologic features, is notable for forming tight infiltrating clusters that
lack fibrovascular cores and that are surrounded by retraction spaces on histologic
samples. Urinary cytology demonstrates tightly packed clusters in a three-
dimensional micropapillary configuration, which may show hyperchromasia, high
N/C ratio, and occasional vacuolated cells (Fig. 6.26) [23, 25, 26].
Plasmacytoid urothelial carcinoma is a poorly differentiated variant of HGUC
that is histologically characterized by infiltrating discohesive cells with eccentric
nuclei and abundant cytoplasm, resembling plasma cells. Rare reports have docu-
mented this variant in urinary cytology specimens, noting similar features to the
histologic correlate [25, 26]. The cytomorphology is characterized by large, disco-
hesive malignant cells with abundant and thick cytoplasm. Thus the tumor cells may
have an N/C ratio < 0.7. The nuclei are eccentrically placed and are hyperchromatic
with irregular nuclear membranes and coarse chromatin. These are clearly different
from benign cells or those seen in LGUN cases (Fig. 6.27) [25, 26].
Fig. 6.27 Large,
discohesive hyperchromatic
malignant cells with
abundant and thick
cytoplasm and an
eccentrically placed
nucleus represent a
plasmacytoid HGUC
(bladder washing, TP,
high mag)
112 M. T. Siddiqui et al.
Fibrovascular Cores
A papillary growth pattern is defined by the presence of three-dimensional cellular
clusters surrounding a central fibrovascular core. True papillary clusters are exceed-
ingly rare in urine cytology specimens, most likely due to the fact that intact papil-
lary tissue fragments do not routinely shed into the urine without some disrupting,
external force. Furthermore, papillary tissue fragments are not specific for a single
entity and may derive from a papilloma, papillary urothelial neoplasm of low malig-
nant potential (PUNLMP), low-grade papillary urothelial carcinoma, or high-grade
papillary urothelial carcinoma. On the other hand, papillary-like clusters, which
lack a central small capillary in the fibrovascular core, are frequently observed in
both voided and instrumented procedures, but their presence is much more common
in washing/barbotage procedures, due to disruptive forces. The diagnostic signifi-
cance of papillary or papillary-like clusters in urinary cytology is completely depen-
dent on the presence or absence of cytologic features for HGUC. Select cases that
contain true papillary fragments but that lack HGUC features may be placed in
NHGUC as a secondary diagnosis, i.e., low-grade urothelial neoplasm (LGUN); the
majority of cases with clusters of benign urothelial cells will also be placed in the
NHGUC category. Although HGUC typically presents as single cells and small
clusters, the presence of true papillary clusters containing cells with high-grade
cytologic features is consistent with a diagnosis of HGUC.
Sample Reports
Example 1
Note: The specimen is cellular with numerous malignant cells. The malignant
cells have a high N/C ratio, hyperchromatic nuclei, irregular nuclear membranes,
and coarse chromatin.
Example 2
Example 3
References
1. Siddiqui MT, Fadda G, Han JH, Owens CL, Tabatabai ZL, Tsuzuki T. High Grade Urothelial
Carcinoma (HGUC). In: Rosenthal DL, Wojcik EM, Kurtycz DFI, editors. The Paris system
for reporting urinary cytology. Springer; 2016. p. 61–74.
2. Koh HH, Lee MJ, Park NJ, Kim H-S, Oh YL. Impact of implementing the Paris system for
reporting urinary cytology: a single-institutional experience with emphasis on diagnostic
yield of high-grade urothelial carcinoma and low-grade urothelial neoplasm. Anticancer Res.
2020;40(6):3477–84.
3. Danakas A, Sweeney M, Cheris S, Agrawal T. Urinary tract cytology: a cytologic-
histopathologic correlation with the Paris system, an institutional study. J Am Soc Cytopathol.
2021;10(1):56–63.
4. Anbardar MH, Monjazeb R. Reclassification of urinary cytology regarding the Paris system
for reporting urinary cytology with cytohistological correlation demonstrates high sensitivity
for high-grade urothelial carcinoma. Diagn Cytopathol. 2020;48(5):446–52.
5. Cowan ML, VandenBussche CJ. The Paris system for reporting urinary cytology: early review
of the literature reveals successes and rare shortcomings. J Am Soc Cytopathol. 2018;7:185–94.
6. Kurtycz DFI, Sundling KE, Barkan GA. The Paris system for reporting urinary cytology:
strengths and opportunities. Diagn Cytopathol. 2020;48:890–5.
7. Pastorello RG, Barkan GA, Saieg M. Experience on the use of the Paris system for report-
ing urinary cytopathology: review of the published literature. J Am Soc Cytopathol.
2021;10(1):79–87.
8. Meilleroux J, Daniel G, Aziza J, Md’Aure D, Quintyn-Ranty ML, Basset CMI, Evrad SM,
Courtade-Saidi MM. One year of experience using the Paris system for reporting urinary cytol-
ogy. Cancer Cytopathol. 2018;126:430–6.
114 M. T. Siddiqui et al.
9. Stanzione N, Ahmed T, Fung PC, Cai D, Lu DY, Sumida LC, Moatamed NA. The con-
tinual impact of the Paris system on urine cytology, a 3-year experience. Cytopathology.
2020;31:35–40.
10. De Paula R, Oliveira A, Nunes W, Bovolim G, Domingos T, Brot LD, Bezerra S, Cunha I,
Morini M, Saieg M. Two-year study on the application of the Paris system for urinary cytology
in a cancer center. Cytopathology. 2020;31:41–6.
11. Cowan ML, Rosenthal DL, VandenBussche CJ. Improved risk stratification for patients with
high-grade urothelial carcinoma following application of the Paris system for reporting urinary
cytology. Cancer Cytopathol. 2017;125:427–34.
12. Pierconti F, Martini M, Straccia P, Fiorentino V, Musarra T, Larocca LM, Lopez-Beltran
A. Hypochromatic large urothelial cells in urine cytology are indicative of high-grade urothe-
lial carcinoma. APMIS. 2018;126:705–9.
13. Renshaw AA, Gould EW. High-grade urothelial carcinoma with hypochromatic chromatin in
urine cytology. J Am Soc Cytopathol. 2021;10:25–8.
14. Renshaw AA, Gould EW. High-grade urothelial carcinoma on urine cytology resembling
umbrella cells. Acta Cytol. 2018;62:62–7.
15. Renshaw AA, Gould EW. High-grade urothelial carcinoma in urine cytology with jet black and
smooth or glassy chromatin. Cancer Cytopathol. 2018;126:64–8.
16. Vaickus LJ, Tambouret RH. Young investigator challenge: the accuracy of the nuclear-
to-
cytoplasmic ratio estimation among trained morphologists. Cancer Cytopathol.
2015;123(9):524–30.
17. Layfield LJ, Esebua M, Frazier SR, Hammer RD, Bivin WW, Nguyen V, Ersoy I, Schmidt
RL. Accuracy and reproducibility of nuclear/cytoplasmic ratio assessments in urinary cytol-
ogy specimens. Diagn Cytopathol. 2017;45(2):107–12.
18. Zhang ML, Guo AX, VandenBussche CJ. Morphologists overestimate the nuclear-to-
cytoplasmic ratio. Cancer Cytopathol. 2016;124(9):669–77.
19. McIntire PJ, Snow JT, Elsoukkary SS, Soong L, Sweeney J, Robinson BD, Siddiqui MT. Digital
image analysis supports a nuclear-to-cytoplasmic ratio cutoff value below 0.7 for positive for
high-grade urothelial carcinoma and suspicious for high-grade urothelial carcinoma in urine
cytology specimens. Cancer Cytopathol. 2019;127(2):120–4.
20. Allison DB, Olson MT, Lilo M, Zhang ML, Rosenthal DL, VandenBussche CJ. Should the
BK polyomavirus cytopathic effect be best classified as atypical or benign in urine cytology
specimens? Cancer Cytopathol. 2016;124(6):436–42.
21. Suh J, Go H, Sung C, et al. Modification of the Paris system for urinary tract washing speci-
mens using diagnostic cytological features. Cytopathology. 2017;28(6):516–23.
22. Moch H. International Agency for Research on C. WHO Classification of Tumours of the
Urinary System and Male Genital Organs. 4th ed. Lyon: International Agency for Research on
Cancer; 2016.
23. Yoo Y, Choi E, Lee DH, Park S, Sung SH. Diagnostic value of urine cytology in detecting
histologic variants of urothelial carcinoma in the urinary bladder: Cytopathologic correlation
of 72 cases. Diagn Cytopathol. 2021;49(3):367–73.
24. Cardillo M, Reuter VE, Lin O. Cytologic features of the nested variant of urothelial carcinoma.
Cancer Cytopathol. 2003;99(1):23–7.
25. Suo L, Vega I, Thrall M. Cyto-histo correlations of plasmacytoid and micropapillary variants
of high-grade urothelial carcinoma: do they fit well in the Paris system for reporting urine
cytology ? J Am Soc Cytopathol. 2021;10:20–4.
26. Straccia P, Martini M, Sacco E, Bassi PF, Pierconti F. Cytological feature of micropapillary
and plasmacytoid variants of urothelial carcinoma. Diagn Cytopathol. 2020;48:111–7.
27. Chan E, Balassanian R, Tabatabai ZL, Lou H, Vohra P. Improved diagnostic precision of urine
cytology by implementation of the Paris system and the use of cell blocks. Cancer Cytopathol.
2018;126(9):809–16.
28. Straccia P, Bizzarro T, Fadda G, Pierconti F. Comparison between cytospin and liquid-based
cytology in urine specimens classified according to the Paris system for reporting urinary
cytology. Cancer Cytopathol. 2016;124(7):519–23.
Cytopathology of the Upper Urinary
Tract 7
Christopher J. VandenBussche, Jen-Fan Hang,
Patrick J. McIntire, Yurina Miki, Stephen Peyton,
Poonam Vohra, and Mingjuan Lisa Zhang
Background
Upper tract (renal and ureteral) urothelial carcinomas (UTUC) account for only
5–10% of all urothelial carcinomas [1–3]. UTUC tend to show higher grade and
stage and have a worse prognosis than urothelial carcinomas of the bladder (UCB)
[4, 5]. Cytological examination of voided urine (VU) or washings (barbotages)
obtained from the upper urinary tract (UUT) is often a component of the diagnostic
workup for UTUC. Tissue biopsies of UUT lesions can be challenging due to the
increased difficulty of visualization and access of these lesions during ureteroscopy.
Studies have shown that instrumented/selective UUT specimens (catheterized,
washings/barbotages, and brushings) are superior to VU for the detection of UTUC,
but the presence of instrumentation artifacts can impede cytological evaluation.
This chapter summarizes the utility of UUT cytology, the current understanding of
UTUC including its molecular biology, and reviews the relevant current literature.
Explanatory Note
In general, the assessment of upper tract lesions utilizes the same principles applied
by TPS to lower tract lesions. One key consideration is that upper tract lesions are
most frequently seen and evaluated on washing/barbotage specimens. However,
these specimens are not expected to differ greatly from those collected from the
bladder. A second key consideration is that UTHGUC cells may be seen in voided
urine (VU) specimens, as VU specimens contain a sampling of cells from through-
out the urinary tract. However, UTHGUC cells in VU specimens are more likely to
be degenerated due to the greater length of time these cells have spent between their
natural exfoliation and specimen collection. A third key consideration is that diag-
noses made on upper tract cytology specimens collected from specific anatomic
sites (“selective cytology”) can have significant clinical implications. As opposed to
VU specimens which cannot localize lesions within the urothelial tract, selective
cytology specimens can localize UTHGUC to a specific kidney or ureter and direct
definitive treatment, potentially without a concurrent or intervening tissue biopsy
specimen. That decision ultimately is made by the urologic surgeon.
UTUC Prevalence
Synchronous UTUC and urothelial carcinomas of the bladder (UCB) are seen in
17% of cases [3]. Patients with primary UTUC have been reported to develop
recurrences in the bladder in 22–47% of cases and in the contralateral UUT in
2–6% of cases [3]. Molecular analysis on paired primary UTUC and recurrent
UCB tissues from the same patient demonstrated a shared clonal origin; both pri-
mary and recurrent tumors were usually high-grade [6]. This finding suggests a
common field cancerization effect or intravesicular tumor seeding. For UTUC,
alterations in FGFR3, KDM6A, and CCND1 were significantly associated with a
higher risk of developing a subsequent UCB, whereas TP53 mutations were asso-
ciated with a lower risk [6].
Both UTUC and UCB have high somatic mutation burdens and predominant apoli-
poprotein B mRNA editing enzyme catalytic polypeptide (APOBEC)-related muta-
tional signatures [6–10]. In UTUC, FGFR3 and HRAS are more frequently altered,
7 Cytopathology of the Upper Urinary Tract 117
whereas TP53, RB1, and ERBB2 are less frequently altered in comparison to UCB
[6, 7, 10]. These differences may be due to environmental factors that are more
specific to carcinogenesis in UTUC (such as phenacetin, Balkan endemic nephropa-
thy, Chinese herb nephropathy, and association with blackfoot disease) [11].
Aristolochic acid (AA), a compound contained in certain Balkan and Chinese herbal
products, is one of the most potent carcinogens associated with UTUC [12–14]. The
reader is referred to Chap. 1, Pathogenesis, for further discussion.
TPS cytomorphological criteria for HGUC is the same for upper and lower tract
specimens (Figs. 7.1, 7.2, 7.3, 7.4, and 7.5), and a detailed discussion can be
found in Chap. 6 [15–19]. A recent study using digital image analysis determined
that UTHGUCs have higher N/C ratios, smaller cell circumference, smaller
nuclei, and less cytoplasm compared to LTHUGC [20]. While these cells would
still qualify as malignant according to the TPS criteria, their smaller size may
cause them to be overlooked, underestimated, or confused with basal cells. Ideally,
a diagnosis of HGUC in a selective cytology specimen should be confirmed with
a tissue diagnosis prior to any definitive treatment. However, because a good-
quality tissue biopsy may be difficult to obtain from some upper tract lesions, a
diagnosis of HGUC should be made with extra caution in upper tract specimens.
See Explanatory Note 2.
118 C. J. Vandenbussche et al.
a b
c d
Fig. 7.1 HGUC arising in the renal pelvis. (a) This field is cellular and contains cells of different
sizes and varied N/C ratios. A cluster of three particularly large cells can be seen in the top center.
Despite having minimal nuclear contour irregularities, the N/C ratios of these cells exceeds 0.7 and
the cells possess coarse chromatin, features concerning for HGUC. (Renal pelvis washing, SP,
medium mag.). (b) At higher magnification, the coarse chromatin pattern of these larger cells can
be better appreciated. Some of the smaller cells in the background also have high N/C ratios and
coarse chromatin and are also likely malignant. A mitotic figure (arrow) can be seen, a finding that
is not entirely specific for HGUC. (Renal pelvis washing, SP, high mag.). (c) Despite having a very
low N/C ratio, this huge cell has all the other features of malignancy: hyperchromasia, coarse
chromatin, and irregular nuclear contours. The abundant cytoplasm is likely a form of degenerative
change. (Renal pelvis washing, SP, high mag.). (d) The corresponding biopsy reveals papillary
HGUC. Note the full thickness of atypical urothelial cells on this papilla, demonstrating hyper-
chromasia, nuclear enlargement, anisonucleosis, and irregular nuclear contours. (Renal pelvis
biopsy, H&E, high mag)
7 Cytopathology of the Upper Urinary Tract 119
a b
Fig. 7.2 HGUC with prominent cercariform cell formation. (a) The specimen is cellular and con-
tains numerous dispersed single atypical cells. At this magnification, great variation in nuclear size
can be seen. (Renal pelvis washing, SP, low mag.). (b) At higher magnification, the atypical cells
can be identified as malignant. They have coarse chromatin and variable N/C ratios. Some cells
have cytoplasmic tails, forming so-called cercariform cells. This feature is seen more often in
washing specimens containing papillary urothelial neoplasms of any grade and is not specific to
HGUC. (Renal pelvis washing, SP, medium mag)
a b
Fig. 7.3 Small fragments of HGUC. (a) This three-dimensional fragment contains cells with
hyperchromatic nuclei and minimal cytoplasm. While HGUC cells tend to be discohesive, frag-
ments of HGUC may be seen in instrumented specimens. Even here, the cells at the edges appear
discohesive. (Renal pelvis washing, TP, medium mag.) (b) A separate fragment of HGUC cells
from the same case demonstrates striking anisonucleosis. The cells have highly irregular nuclear
contours and the chromatin is too dark to assess for the coarse, clumpy chromatin often seen in
HGUC. However, this level of hyperchromasia is just as concerning as the presence of clumpy
chromatin. (Renal pelvis washing, TP, high mag)
120 C. J. Vandenbussche et al.
a b
Fig. 7.4 HGUC in a pelvic washing with cell-in-cell forms. (a and b) The HGUC cells in these
two representative fields are present singly and in small fragments. Both fields show examples of
a HGUC cell’s cytoplasm enveloping a neighboring HGUC cell. This phenomenon has been
described as “clasping” (when the neighboring cell does not appear entirely enveloped) and “cell-
in-cell” (when the neighboring cell is completely enveloped). Other terms used include “cell can-
nibalism” and “hugging.” While this finding can be seen in HGUC, the finding is not specific and
can occasionally be seen with benign, reactive processes. (Renal pelvis washing, Cytospin,
medium mag)
a b
Fig. 7.5 HGUC. (a) In this ureteral washing specimen, the HGUC cells have features of malig-
nancy: high N/C ratios, coarse chromatin, and slightly irregular nuclear contours. The cells do not
demonstrate significant anisonucleosis, which is often seen in other specimens with HGUC but is
not required for a definitive diagnosis of malignancy. (Ureteral washing, SP, medium mag.) (b) The
presence of atypical keratinized cells can indicate squamous differentiation in a HGUC. In this
renal pelvic washing specimen, the keratinized cells have bland-appearing nuclei but brilliant
orange cytoplasm. Overtly malignant cells are seen in small- to medium-sized non-keratinized
fragments adjacent to the keratinized fragment. While their morphology is compatible with urothe-
lial carcinoma, squamous cell carcinoma can metastasize to the kidney from the lung. If uncer-
tainty exists as to whether this is a HGUC, the “Other” diagnostic category may be used. (Renal
pelvis washing, SP, high mag)
7 Cytopathology of the Upper Urinary Tract 121
Explanatory Note 1
In a barbotage/washing specimen that adequately samples a suspicious lesion, many
abnormal cells should be seen in the specimen if the neoplasm is HGUC. In instances
where a distinct lesion cannot be seen on uroscopy and selective cytology is being
used to localize HGUC identified on a prior voided urine specimen, a flat CIS or
small papillary HGUC lesion could be present and yield only a smaller subpopula-
tion of malignant cells. Following treatment and resection, the identification of focal
areas of CIS may be challenging and perhaps not possible on gross and microscopic
examination of the resection specimen.
Explanatory Note 2
When diagnosing HGUC in upper tract specimens, circumspection is recommended,
favoring specificity over sensitivity. In other words, the diagnosis of HGUC in upper
tract specimens should be reserved for those cases with indisputable qualitative cri-
teria (Fig. 7.6 a). Due to the infrequency of diagnosing UTHGUC, review by a
second pathologist is prudent. While a diagnosis of suspicious for HGUC (SHGUC)
is acceptable for specimens that do not clearly contain HGUC cells, contacting the
clinician can allow discussion regarding the amount of suspicion one has for HGUC
as well as how the patient should be managed (Fig. 7.6 b).
a b
Fig. 7.6 Suspicious for HGUC. (a) There are five atypical urothelial cells in the center of the field.
While some of these cells meet the cytomorphologic criteria for HGUC, a higher threshold is rec-
ommended for making a diagnosis of HGUC in upper tract specimens as compared to lower tract
specimens. Based on the overall clinical picture, an upper tract specimen may be quantitatively or
qualitatively insufficient for HGUC. Direct communication with the clinical team in these circum-
stances to convey the level of concern for HGUC may be useful. (Renal pelvis washing, SP, high
mag.) (b) Several singly dispersed malignant cells are intermixed with fungal organisms, debris,
and acute inflammatory cells. While some of the cells meet the cytomorphologic criteria for
HGUC, other cells are obscured and degenerated. This specimen is at least suspicious for HGUC
and a diagnosis of HGUC may be rendered depending on what is seen in other fields and the clini-
cal scenario. (Nephrostomy tube, SP, high mag)
122 C. J. Vandenbussche et al.
Only a few pre-TPS studies have assessed specific cytomorphologic features predic-
tive of UTUC in UUT specimens [21]. Potts et al. [22] found that increased nuclear-
to-
cytoplasmic (N/C) ratio, anisonucleosis, and overlapping nuclei were
distinguishing features for identifying UTHGUC. Witte et al. [23] reviewed a cohort
of renal pelvis washing specimens and determined that high N/C ratio, isolated
cells, anisonucleosis, nuclear hyperchromasia, and coarse chromatin were predic-
tive of UTHGUC (Fig. 7.7).
Several studies have examined the cytomorphologic features of UTHGUC in the
context of TPS and thus provide insight into the usefulness of TPS criteria in this
setting [21]. Simon et al. found that the incidence of cytomorphologic alterations
among UTHGUC in UUT specimens was N/C ratios ≥0.7 (88%), hyperchromasia
(83%), coarse chromatin distribution (67%), and nuclear pleomorphism within cell
clusters (65%) [24]. The most consistent feature associated with a downgrade from
HGUC to SHGUC (60% of cases) was the lack of marked nuclear membrane irreg-
ularity, which was only seen in 23% of HGUC cases. The presence of coarse chro-
matin and bizarre single cells was more commonly seen in HGUC in comparison to
LGUC; these findings were statistically significant. Increased N/C ratios, nuclear
hyperchromasia, nuclear membrane irregularities, pleomorphism, and the presence
of mitoses and/or atypical degeneration were not statistically significant features in
differentiating between low-grade and HGUC of the UUT [43].
Fig. 7.7 Fragment of HGUC from the ureter. Upper tract specimens are obtained through instru-
mentation and thus fragments of HGUC, such as the one seen here, may be forcibly exfoliated
during collection. This contrasts to the HGUC cells seen in voided urine specimens, which natu-
rally exfoliate as predominantly single cells from the urothelial lining. Because the center of this
fragment contains overlapping cells, it can be difficult to discern the morphology of each individ-
ual cell. By looking at the edge of the fragment, the high N/C ratios and coarse chromatin can be
better appreciated. (Ureter washing, SP, medium mag)
7 Cytopathology of the Upper Urinary Tract 123
Explanatory Note 1
While some studies have identified an association of nuclear hypochromasia with
UTHGUC, Renshaw and Gould reported a similar finding in HGUC cells in bladder
washing specimens (A. Renshaw, personal communication, April 15, 2021) [26]. It
is therefore possible that hypochromasia may be associated with instrumented spec-
imens rather than tumor location (Fig. 7.8).
Explanatory Note 2
Due to the discohesive nature of HGUC cells, they are often seen as single malig-
nant cells when they naturally exfoliate into urinary tract specimens. However,
HGUC may be forcibly exfoliated as large papillary fragments in instrumented uri-
nary tract specimens (Fig. 7.9). Therefore, one should not automatically assume that
the presence of urothelial tissue fragments with fibrovascular cores indicates a low-
grade lesion. If HGUC is present, the pathognomonic single cells meeting the crite-
ria of HGUC will confirm the diagnosis (Fig. 7.10).
a b
Fig. 7.8 Hypochromatic HGUC cells. (a) Occasionally HGUC cells may appear paradoxically
hypochromatic, a feature seen in many of the cells in this field. However, the malignant cells are
very large and have elevated N/C ratios, nuclear contour irregularities, and chromatin coarseness.
Because the chromatin stains faintly, these cells may be overlooked. (Renal pelvis washing, SP,
medium mag.) (b) A higher magnification shows the atypical nature of these hypochromatic cells,
with markedly atypical nuclear borders. The chromatin clumps are also small and difficult to dis-
cern, even at this high magnification. Note that some cells with dark chromatin can be seen in the
top right corner. (Renal pelvis washing, SP, high mag)
124 C. J. Vandenbussche et al.
a b
c d
Fig. 7.9 Examples of HGUC in fragments. (a) Tissue fragments containing HGUC cells. While
the tissue fragments have a papillary configuration, they lack fibrovascular cores. Single HGUC
cells with variable amounts of atypia can be seen in the background. (Renal pelvis washing, SP,
medium mag.) (b) Separate field of (a) at higher magnification. The smaller fragment contains cells
with more severe atypia than the cells seen in the lower, larger fragment and demonstrates hyper-
chromasia, coarse chromatin, anisonucleosis, and, for some of the cells, high N/C ratios. (Renal
pelvis washing, SP, high mag.) (c) Numerous tissue fragments of variable sizes containing HGUC
cells. HGUC cells can also be seen dispersed in the background. Examination at higher magnifica-
tion is required to confirm the low power impression of HGUC. (Ureteral washing, SP, low mag.)
(d) Well-preserved HGUC cells in a fragment, with coarse chromatin, hyperchromasia, and high
N/C ratios. Instrumented specimens can obtain freshly exfoliated cells that have not been degener-
ating in the bladder. (Renal pelvis washing, SP, high mag)
7 Cytopathology of the Upper Urinary Tract 125
Fig. 7.10 HGUC cells adjacent to an LGUC fragment. Patients may have both LGUC and HGUC
in the urinary tract, so it is most important to diagnose specimens based on features of HGUC
rather than the identification of low grade urothelial neoplasms. This patient had LGUC and CIS
in one ureter. The papillary lesion was identified by cystoscopy due to its papillary growth, but the
CIS lesion was not identified until subsequent biopsy. This field contains a combination rarely
seen – a papillary fragment of LGUC, identified by a fibrovascular stalk (arising from the right-
hand bottom corner) with a monotonous population of attached neoplastic cells – and a small
fragment of HGUC cells (left-hand bottom corner). (Ureteral washing, SP, medium mag)
The criteria for diagnosing UTHGUC cells in VU specimens are the same criteria
used to diagnose HGUC of the lower urinary tract, as the location of HGUC cannot
be determined in VU specimens. UTHGUC cells are more likely to be degenerated
in VU due to increased transit times and thus are more likely to be diagnosed as
atypical urothelial cells (AUC) or SHGUC rather than HGUC. Zhang et al. found
that among patients with biopsy-confirmed UTHGUC, 50% of preceding VU speci-
mens demonstrated high-grade cytomorphological features (SHGUC/HGUC) com-
pared to almost 90% of preceding UUT cytology specimens; 81% (26/32) of VU
samples were diagnosed as AUC or higher [27].
The instrumentation required to procure UUT cytology specimens can result in the
exfoliation of benign, reactive urothelial tissue fragments along with fragments of
neoplastic cells. This complicates the detection of LGUN, as LGUN tissue frag-
ments can have significant cytomorphologic overlap with tissue fragments contain-
ing benign and reactive urothelial cells (Figs. 7.11, 7.12, 7.13, 7.14, and 7.15). The
presence of urothelial tissue fragments with fibrovascular cores are considered con-
sistent with a clinical impression of LGUN, so long as the diagnostic features of
126 C. J. Vandenbussche et al.
high-grade lesions are not seen in the specimen (Fig. 7.16). The cytomorphologic
features of LGUN are discussed in Chap. 3 (Figs. 7.17 and 7.18) and are applicable
to UT lesions. Small urothelial tissue fragments without fibrovascular cores and
with bland nuclei may arise from either a urothelial neoplasm or the normal urothe-
lial lining [25] [28–30].
a b
c d
Fig. 7.11 Fragments of low-grade urothelial neoplasm without fibrovascular cores. (a–d) This
patient was found to have a low-grade urothelial carcinoma on a concurrent biopsy. In this washing
specimen, there are numerous neoplastic cells both in variably sized fragments (a–c) as well as
dispersed single cells with cercariform features (d). The cells and their nuclei have an elongated,
spindly appearance, possibly due to cautery artefact, which causes some fragments to resemble a
spindle cell neoplasm (a–c). No features of HGUC can be identified. The cells have low N/C ratios
and bland chromatin; the nuclei are all similar in size. In this case, it is appropriate to make a diag-
nosis of NHGUC; the specimen provides a good example of how LGUN can be cellular and visu-
ally striking in washing specimens. (Renal pelvis washings, TP, medium mag)
7 Cytopathology of the Upper Urinary Tract 127
Fig. 7.12 Benign urothelial tissue fragment. Instrumentation often forcibly exfoliates benign uro-
thelial tissue fragments that may have cytomorphologic overlap with LGUN fragments. Benign
urothelial tissue, however, does not form papillary fragments with fibrovascular cores. While some
anisonucleosis can be spotted in this fragment, other features are benign. The cells have pale chro-
matin and low N/C ratios. (Renal pelvis washing, Cytospin, medium mag)
a b
Fig. 7.13 Benign urothelial tissue fragments. (a) The nuclei of benign urothelial cells in frag-
ments may appear slightly darker and have more irregular contours when compared to singly dis-
persed urothelial cells. However, even at this magnification, one can see that these cells have
abundant cytoplasm, supporting a benign diagnosis. (Renal pelvis washing, SP, medium mag.) (b)
One benign feature in the small fragment at the top of the field is the presence of a “cytoplasmic
collar,” formed by centrally placed nuclei surrounded by a peripheral “collar” of granular cyto-
plasm. This configuration represents the asymmetric unit membrane, a pathognomonic feature of
superficial urothelium. Nuclei with prominent chromocenters and condensed rims of chromatin are
characteristic of superficial cells and should not be mistaken for the hyperchromasia and coarse
chromatin pattern seen in HGUC. (Renal pelvic washings, SP, medium mag)
128 C. J. Vandenbussche et al.
a b
Fig. 7.14 Benign urothelial tissue fragments. (a) These are benign urothelial cells from a renal
pelvis washing. Note the presence of multinucleated umbrella cells with abundant cytoplasm,
single cells in the background with low N/C ratios and bland chromatin, and a three-dimensional
benign urothelial tissue fragment at the bottom left-hand side of the field. (Renal pelvis washing,
SP, medium mag.) (b) At higher magnification, the benign features of the tissue fragment seen in
(a) can be appreciated. The cells have nuclei of approximately the same size and abundant, foamy
cytoplasm that forms a peripheral cytoplasmic collar around the fragment. Most of the nuclei have
two to three prominent chromocenters, which could be mistaken for a coarse chromatin pattern.
However, the chromatin clumps are distributed similarly among the cells and do not vary in size.
By contrast, the chromatin clumps in HGUC tend to be nonuniform in size and unevenly distrib-
uted within the nucleus, with dissimilar chromatin patterns found among the carcinoma cells.
(Renal pelvis washing, SP, high mag)
7 Cytopathology of the Upper Urinary Tract 129
a b
c d
Fig. 7.15 Benign urothelial tissue fragments demonstrating stone atypia. (a-d) These tissue frag-
ments originated in a renal pelvis washing from a patient with nephrolithiasis. The fragments are
of variable sizes; the larger three-dimensional fragments (d) make cytomorphology difficult to
assess. Stones in the urinary tract can cause reactive changes in benign urothelial cells and also
forcibly exfoliate urothelial tissue fragments as they pass through the ureter into the bladder. The
fragments contain cells with increased N/C ratios and dark, hyperchromatic nuclei. The nuclei may
be so hyperchromatic that the chromatin pattern cannot be assessed. In addition to increasing one’s
threshold for an atypical diagnosis in patients with urolithiasis, certain features can be reassuring.
The nuclei are usually small and do not vary greatly in size within the fragments. A cytoplasmic
collar may be present, and well-preserved single cells concerning for HGUC are absent in the
background. (a–b, renal pelvis washing, Cytospin, medium mag.; c, renal pelvis washing, SP,
medium mag)
130 C. J. Vandenbussche et al.
a b
Fig. 7.16 True papillary tissue fragments arising from LGUC. (a) This patient had a LGUC in
their ureter. The washing procedure forcibly exfoliated this papillary fragment with a fibrovascular
stalk and additionally caused the detachment of numerous neoplastic cells (seen in the back-
ground). At this magnification, the neoplastic cells appear monotonous, whereas HGUC cells often
appear more pleomorphic. Because HGUC may coexist with LGUN and because LGUN lesions
may contain foci of HGUC, examination at a higher magnification is warranted to exclude the
presence of HGUC. (Renal pelvis washing, SP, low mag.) (b) Note the monotony of the cells
attached to this papillary fragment. The cells have round-to-oval nuclei with regular contours. The
nuclei are similar in size. Neoplastic cells can also be seen singly in the background and have
eccentrically placed nuclei with low N/C ratios (below 0.5). (Renal pelvis washing, SP, medium mag)
a b
Fig. 7.17 Cercariform cells derived from LGUN. (a) In washing specimens, urothelial cells
derived from papillary neoplasms may exfoliate as single cells. These cells may have cytoplasmic
tails, resulting in a “cercariform” appearance. Here, the nuclei are monomorphic and small, sug-
gesting that the neoplasm is low-grade. No cytomorphologic features of HGUC can be identified.
(Renal pelvis washing, TP, medium mag.) (b) Cells derived from a papillary LGUC appear both
epithelioid and spindled (cercariform). The epithelioid cells have eccentrically placed nuclei with
a bland chromatin pattern and regular nuclear contours. Some cells possess degenerative features,
such as small, pyknotic nuclei. Despite their hyperchromasia, these degenerating cells and their
nuclei are notably smaller than those seen in the well-preserved neoplastic cells. (Ureteral wash-
ing, SP, high mag)
7 Cytopathology of the Upper Urinary Tract 131
a b
Fig. 7.18 LGUN cells with plasmacytoid morphology. (a and b) These are two representative
fields from a renal pelvis washing in a patient with renal LGUC. The neoplastic cells fill the fields
with rare benign umbrella cells seen in the background. The cells are monomorphic and have bland
chromatin. Their nuclei are mostly eccentrically placed, causing a plasmacytoid appearance. Some
cells appear to have irregular nuclear shapes, irregular contours, and nuclear grooves, and many
cells have N/C ratios above 0.5. While these findings technically meet the criteria for the AUC
category, the specimen could be diagnosed as NHGUC if the overall features are more suggestive
of LGUN. (Renal pelvis washing, Cytospin, high mag)
Prior to the publication of TPS in 2016, numerous studies, dating back to the 1960s,
have examined the performance of urinary cytology for the diagnosis of UTUC. Most
studies have primarily focused on the utility of selective UUT specimens (i.e., renal
pelvis and/or ureteric washings/brushings) either in isolation or in comparison with
lower urinary tract cytology specimens, as well as other diagnostic modalities.
Furthermore, some studies have examined the prognostic value of UUT cytology in
predicting disease progression or recurrence in patients who have undergone surgical
treatment for UTUC. Unfortunately, many of these studies focus on the detection of
all urothelial carcinomas regardless of grade. Because of the difficulty in detecting
LGUN in urine cytology specimens, the performance of urine cytology for detecting
all urothelial carcinomas is typically worse than the performance for detecting only
high-grade lesions [31]. The practice of intermingling both low- and high-grade
lesions in these cohorts has resulted in urine cytology being historically associated
with low sensitivities and makes the results of these studies difficult to compare
directly with those obtained from studies in the TPS era. Combined analysis of pre-
TPS studies is further complicated by inconsistencies in the interpretation of an inde-
terminate cytological diagnoses since some studies considered an “atypical” urinary
cytological specimen as a “positive” diagnosis, while others considered it to be a
132 C. J. Vandenbussche et al.
The majority of post-TPS studies on UUT specimens were conducted using retro-
spective slide review and yielded widely variable diagnostic reporting rates of each
category. Some UUT studies showed a decrease in the rate of the AUC category,
whereas others showed no significant change [21, 24, 25, 27, 45–49]. These contra-
dictory trends differ from those seen in studies examining lower urinary tract cytol-
ogy specimens prospectively diagnosed following the implementation of TPS. In
these studies, the AUC rate usually declines following implementation of TPS [50].
The reported sensitivities of UUT cytology for HGUC after implementation of TPS
are not markedly different from those reported in pre-TPS literature (19–82%)
while maintaining excellent specificity (86–100%) and positive predictive value
(92–100%) [24, 25, 45–48]. As expected, the sensitivity of UUT cytology for pre-
dicting subsequent HGUC on tissue biopsy improves when the SHGUC and/or
AUC diagnostic categories are included in the positive cohort [24, 45, 46]. McIntire
et al. reported that application of TPS made UUT cytology nearly twice as likely
(63% vs. 34%) to be concordant with follow-up histology and increased the identi-
fication of HGUC as well as LGUN compared to using pre-TPS diagnoses [47].
The reported risk of malignancy (ROM) for UTUC based on UUT cytology for
each TPS category varies among different studies. These contradictory results may
be attributed to the sparse number of UUT cytology studies, the retrospective design
of most studies, and the limited cohort sizes within each diagnostic category avail-
able for inclusion in these studies [21, 25, 46–48]. Some studies reflected an overall
7 Cytopathology of the Upper Urinary Tract 133
shift to downgrading diagnoses and highlighted how the strict application of TPS
criteria for UUT cytology can lead to a decrease in the frequency of positive diag-
noses and a concurrent increase in SHGUC and NHGUC diagnoses. Cited pre-
sumptive reasons included lack of marked nuclear membrane irregularities and
nuclear hypochromasia rather than irregular nuclear membranes and hyperchroma-
sia, both of which have been considered essential cytological criteria for the diagno-
sis of HGUC in the prior version of TPS [16, 19, 25, 27, 45, 51].
Given that the primary goal of urine cytology is the detection of HGUC, the role of
TPS is limited for the identification of LGUN, which may demonstrate nonspecific
features [31]. The overall sensitivity for LGUN is much lower than that for HGUC,
with reported sensitivities of 46% for LGUN in UUT cytology prior to TPS vs.
0–34% (average 17%) after implementation of TPS [24, 25, 45–49].
The accurate diagnosis and staging of UTUC can be challenging due to the limita-
tions of the endoscopic instrumentation that is utilized to obtain surgical biopsy
specimens. Thus, selective upper tract cytology can have significant impact on the
treatment decision. The decision to proceed with a nephroureterectomy based solely
on a cytology result (HGUC or SHGUC) without an endoscopically or radiographi-
cally visible lesion is generally not recommended; however, these patients should
be monitored closely to identify the source for the abnormal cells and be treated
accordingly in a timely fashion.
Specimen Collection
Fig. 7.19 Indeterminate cells in a voided urine, derived from an upper tract HGUC. Upper tract
HGUC cells naturally exfoliate into the urinary stream but may have degenerated before a voided
specimen is collected. These three urothelial cells are quite large compared to the background
inflammatory cells. The nuclei have irregular contours and dark chromatin, but the chromatin pat-
tern is obliterated, likely due to degeneration. However, these were the only atypical cells seen in
the specimen and the N/C ratios are lower that 0.5; a diagnosis of SHGUC rather than HGUC may
be appropriate depending on overall quantity and quality. (Voided urine, SP, high mag)
Cell block (CB) preparations help detect tissue fragments containing fibrovascular
cores and assess cytomorphologic features, thus aiding the diagnosis of both HGUC
(Fig. 7.20) as well as LGUN (Fig. 7.21). However, the performance of this practice
has not been rigorously studied in UUT cases. Xing et al. examined 12 CBs pre-
pared from UUT specimens and showed that they could provide additional informa-
tion for identifying LGUN, particularly when a biopsy of these UUT lesions was
not feasible [25]. Chan et al. reported that CBs represent a useful adjunctive method
7 Cytopathology of the Upper Urinary Tract 135
a b
c d
Fig. 7.20 Cell block preparation containing HGUC. (a and b) Numerous HGUC cells are found
singly as well as in a tissue fragment. Note the anisonucleosis, elevated N/C ratios, and irregular
nuclear shapes. Many of these cells have a prominent nucleolus rather than clumped chromatin.
(Ureteral washing, TP, intermediate mag.) (c and d) The cell block preparation demonstrates con-
cordant morphology. However, the fibrovascular cores of true papillary fragments can be identified
in the cell block preparation (c). (Renal pelvis washing, H&E, low mag. [c] and high mag. [d])
for classification as “LGUN” within the “Negative for HGUC (NHGUC)” TPS
diagnostic category, but their study included primarily lower urinary tract cytology
specimens [63] They reported that a CB can increase the confidence of a LGUN
diagnosis. Dantey et al. showed that routinely preparing CBs for urine samples did
not improve their laboratory’s specimen adequacy rate [64]. Nonetheless, CBs of
urine samples did help lower the frequency of atypical diagnoses, from 33% to
16–19% when routinely or selectively prepared. Overall, these studies indicate that
CB preparations may be helpful in UUT specimens, as biopsy is not always possi-
ble, and washing/barbotage specimens are generally adequately cellular for CB cre-
ation. However, more data are needed on the performance of CBs in UUT specimens,
specifically. See Chap. 10, Cytopreparatory Techniques, for further discussion.
136 C. J. Vandenbussche et al.
a b
Fig. 7.21 Cell block preparation containing LGUC. (a) Given the bland cytomorphology seen in
their cellular constituents, these benign-appearing tissue fragments could represent either frag-
ments of benign urothelial lining or fragments derived from an LGUN. No features of HGUC can
be identified. (Renal pelvis washing, TP, medium mag.) (b) The cell block preparation shows a
cross section of a papillary fragment containing a fibrovascular stalk. The cells have monomorphic
nuclei, corresponding to LGUN (and, in this case, LGUC). (Renal pelvis washing, H&E,
medium mag)
Adequacy
UUT cytology specimens are usually obtained through washing/barbotage, and thus
assessment for adequacy in these specimens differs from that of VU specimens.
Studies have shown that specimen volume and obscuring factors play an important
role in the sensitivity of voided urinary cytology specimens for HGUC, whereas the
number of urothelial cells per high-power field (HPF) is more important in washing/
barbotage specimens [65–67]. The minimum urothelial cell count predictably
affects adequacy depending on specimen preparation types, with one study showing
an increased false-negative rate in ThinPrep specimens containing less than 20 uro-
thelial cells per 10 HPFs, and another study showing an increased false-negative
rate in Cytospin specimens containing less than 10 urothelial cells per 10 HPFs [67,
68]. Refer to Chap. 2 on Adequacy for suggested guidelines.
of UroVysion in UTUC and reported that combining UUT cytology and UroVysion
can improve the diagnostic accuracy of UTUC, especially for cases diagnosed as
AUC or SHGUC [48]. Further information on ancillary studies is presented in
Chap. 9.
Examples of Reports
Final diagnosis:
Final diagnosis:
References
1. Gupta R, Paner GP, Amin MB. Neoplasms of the upper urinary tract: a review with focus
on urothelial carcinoma of the pelvicalyceal system and aspects related to its diagnosis and
reporting. Adv Anat Pathol. 2008;15(3):127–39.
2. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020 [internet]. CA Cancer J Clin.
2020;70(1):7–30. https://doi.org/10.3322/caac.21590.
138 C. J. Vandenbussche et al.
25. Xing J, Monaco SE, Pantanowitz L. Utility of the Paris system for reporting urinary cytology
in upper urinary tract specimens. J Am Soc Cytopathol. 2018;7(6):311–7.
26. Renshaw AA, Gould EW. High-grade urothelial carcinoma with hypochromatic chromatin in
urine cytology. J Am Soc Cytopathol. 2021;10(1):25–8.
27. Zhang ML, Rosenthal DL, VandenBussche CJ. Upper urinary tract washings outper-
form voided urine specimens to detect upper tract high-grade urothelial carcinoma. Diagn
Cytopathol. 2017;45(8):700–4.
28. Onur I, Rosenthal DL, VandenBussche CJ. Benign-appearing urothelial tissue fragments in
noninstrumented voided urine specimens are associated with low rates of urothelial neoplasia.
Cancer Cytopathol. 2015;123(3):180–5.
29. Onur I, Rosenthal DL, VandenBussche CJ. Atypical urothelial tissue fragments in nonin-
strumented voided urine specimens are associated with low but significantly higher rates of
urothelial neoplasia than benign-appearing urothelial tissue fragments. Cancer Cytopathol.
2015;123(3):186–92.
30. Zhang ML, Zhou AG, Rosenthal DL, VandenBussche CJ. Urinary tract washing specimens
containing atypical urothelial tissue fragments are significantly associated with urothelial neo-
plasia. Diagn Cytopathol. 2017;45(9):795–9.
31. Zhang ML, Rosenthal DL, VandenBussche CJ. The cytomorphological features of low-grade
urothelial neoplasms vary by specimen type. Cancer Cytopathol. 2016;124(8):552–64.
32. Zhang ML, VandenBussche CJ, Hang J-F, et al. A review of urinary cytology in the setting of
upper tract urothelial carcinoma. J Am Soc Cytopathol. 2021;10(1):29–35.
33. Potretzke AM, Knight BA, Vetter JM, et al. Diagnostic utility of selective upper tract urinary
cytology: a systematic review and meta-analysis of the literature. Urology. 2016;96:35–43.
34. Wang L, Pambuccian SE, Wojcik EM, Barkan GA. Diagnosis of upper tract urothelial car-
cinoma—a comparative study of urinary cytology and surgical biopsy [internet]. J Am Soc
Cytopathol. 2015;4(1):3–9. Available from: https://doi.org/10.1016/j.jasc.2014.09.203
35. Messer J, Shariat SF, Brien JC, et al. Urinary cytology has a poor performance for predicting
invasive or high-grade upper-tract urothelial carcinoma. BJU Int. 2011;108(5):701–5.
36. Straub J, Strittmatter F, Karl A, Stief CG, Tritschler S. Ureterorenoscopic biopsy and urinary
cytology according to the 2004 WHO classification underestimate tumor grading in upper
urinary tract urothelial carcinoma. Urol Oncol. 2013;31(7):1166–70.
37. Williams SK, Denton KJ, Minervini A, et al. Correlation of upper-tract cytology, retrograde
pyelography, ureteroscopic appearance, and ureteroscopic biopsy with histologic examination
of upper-tract transitional cell carcinoma. J Endourol. 2008;22(1):71–6.
38. Renshaw AA. Comparison of ureteral washing and biopsy specimens in the community set-
ting. Cancer. 2006;108(1):45–8.
39. Skolarikos A, Griffiths TRL, Powell PH, Thomas DJ, Neal DE, Kelly JD. Cytologic analysis
of ureteral washings is informative in patients with grade 2 upper tract TCC considering endo-
scopic treatment. Urology. 2003;61(6):1146–50.
40. Keeley FX, Kulp DA, Bibbo M, McCue PA, Bagley DH. Diagnostic accuracy of ureteroscopic
biopsy in upper tract transitional cell carcinoma. J Urol. 1997;157(1):33–7.
41. Brien JC, Shariat SF, Herman MP, et al. Preoperative hydronephrosis, ureteroscopic biopsy
grade and urinary cytology can improve prediction of advanced upper tract urothelial carci-
noma. J Urol. 2010;184(1):69–73.
42. Gruschwitz T, Gajda M, Enkelmann A, et al. FISH analysis of washing urine from the upper
urinary tract for the detection of urothelial cancers. Int Urol Nephrol. 2014;46(9):1769–74.
43. Xu C, Zeng Q, Hou J, et al. Utility of a modality combining FISH and cytology in upper
tract urothelial carcinoma detection in voided urine samples of Chinese patients. Urology.
2011;77(3):636–41.
44. Sverrisson EF, Kim T, Espiritu PN, et al. The merits of cytology in the workup for upper
tract urothelial carcinoma - a contemporary review of a perplexing issue. Int Braz J Urol.
2014;40(4):493–8.
45. Simon CT, Skala SL, Magers MJ, et al. The utility of upper urinary tract urine cytology before
and after application of the Paris system. Diagn Cytopathol. 2019;47(5):421–7.
140 C. J. Vandenbussche et al.
46. Zheng X, Si Q, Du D, et al. The Paris system for urine cytology in upper tract urothelial
specimens: a comparative analysis with biopsy and surgical resection. Cytopathology.
2018;29(2):184–8.
47. McIntire PJ, Snow JT, Robinson BD, et al. Improved correlation of urinary cytology specimens
using the Paris system in biopsy-proven upper tract urothelial carcinomas. Cancer Cytopathol.
2018;126(7):498–504.
48. Sassa N, Iwata H, Kato M, et al. Diagnostic utility of UroVysion combined with conven-
tional urinary cytology for urothelial carcinoma of the upper urinary tract. Am J Clin Pathol.
2019;151(5):469–78.
49. Bertsch EC, Siddiqui MT, Ellis CL. The Paris system for reporting urinary cytology improves
correlation with surgical pathology biopsy diagnoses of the lower urinary tract. Diagn
Cytopathol. 2018;46(3):221–7.
50. Cowan ML, Rosenthal DL, VandenBussche CJ. Improved risk stratification for patients with
high-grade urothelial carcinoma following application of the Paris system for reporting urinary
cytology. Cancer Cytopathol. 2017;125(6):427–34.
51. Kurtycz DFI, Barkan GA, Pavelec DM, et al. Paris Interobserver reproducibility study
(PIRST). J Am Soc Cytopathol. 2018;7(4):174–84.
52. Yafi FA, Brimo F, Auger M, Aprikian A, Tanguay S, Kassouf W. Is the performance of urinary
cytology as high as reported historically? A contemporary analysis in the detection and surveil-
lance of bladder cancer. Urol Oncol. 2014;32(1):27.e1–6.
53. Dev HS, Poo S, Armitage J, Wiseman O, Shah N, Al-Hayek S. Investigating upper urinary tract
urothelial carcinomas: a single-Centre 10-year experience. World J Urol. 2016;
54. Raica M, Mederle O, Ioiart I. Upper urinary tract washing in the diagnosis of transitional
cell carcinoma of the renal pelvis and calices. A comparison with voided urine. Romanian J
Morphol Embryol. 1995;41(3–4):101–5.
55. Muus Ubago J, Mehta V, Wojcik EM, Barkan GA. Evaluation of atypical urine cytology pro-
gression to malignancy. Cancer Cytopathol. 2013;121(7):387–91.
56. Asai S, Fukumoto T, Tanji N, et al. Fluorodeoxyglucose positron emission tomography/com-
puted tomography for diagnosis of upper urinary tract urothelial carcinoma. Int J Clin Oncol.
2015;20(5):1042–7.
57. Jovanovic M, Soldatovic I, Janjic A, et al. Diagnostic value of the nuclear matrix protein 22
test and urine cytology in upper tract urothelial tumors. Urol Int. 2011;87(2):134–7.
58. Lodde M, Mian C, Wiener H, Haitel A, Pycha A, Marberger M. Detection of upper urinary tract
transitional cell carcinoma with ImmunoCyt: a preliminary report. Urology. 2001;58(3):362–6.
59. Walsh IK, Keane PF, Ishak LM, Flessland KA. The BTA stat test: a tumor marker for the detec-
tion of upper tract transitional cell carcinoma. Urology. 2001;58(4):532–5.
60. Wu WJ, Liu LT, Huang CN, Huang CH, Chang LL. The clinical implications of telomerase
activity in upper tract urothelial cancer and washings. BJU Int. 2000;86(3):213–9.
61. Dodd LG, Johnston WW, Robertson CN, Layfield LJ. Endoscopic brush cytology of the upper
urinary tract. Evaluation of its efficacy and potential limitations in diagnosis. Acta Cytol.
1997;41(2):377–84.
62. Siemens DR, Morales A, Johnston B, Emerson L. A comparative analysis of rapid urine tests
for the diagnosis of upper urinary tract malignancy. Can J Urol. 2003;10(1):1754–8.
63. Chan E, Balassanian R, Tabatabai ZL, Lou H, Vohra P. Improved diagnostic precision of urine
cytology by implementation of the Paris system and the use of cell blocks. Cancer Cytopathol.
2018;126(9):809–16.
64. Dantey K, Pantanowitz L, Xing J, Cuda J, Nestler R, Monaco SE. Cell block preparation
in urine cytology: examination of utility and workflow in an academic practice. J Am Soc
Cytopathol. 2019;8(2):61–8.
65. VandenBussche CJ, Rosenthal DL, Olson MT. Adequacy in voided urine cytology specimens:
the role of volume and a repeat void upon predictive values for high-grade urothelial carci-
noma. Cancer Cytopathol. 2016;124(3):174–80.
7 Cytopathology of the Upper Urinary Tract 141
66. Rezaee N, Tabatabai ZL, Olson MT. Adequacy of voided urine specimens prepared by
ThinPrep and evaluated using the Paris system for reporting urinary cytology. J Am Soc
Cytopathol. 2017;6(4):155–61.
67. Renshaw AA, Gould EW. Adequacy criteria for voided urine cytology using cytospin prepara-
tions. Cancer Cytopathol. 2019;127(2):116–9.
68. Prather J, Arville B, Chatt G, et al. Evidence-based adequacy criteria for urinary bladder bar-
botage cytology. J Am Soc Cytopathol. 2015;4(2):57–62.
69. Akkad T, Brunner A, Pallwein L, et al. Fluorescence in situ hybridization for detecting upper
urinary tract Tumors—a preliminary report. Urology. 2007;70(4):753–7.
70. Yu Q, Li Y, Li G, et al. Prospective evaluation of FISH for detecting upper tract urothelial
carcinoma in voided urine specimens. Oncol Lett. 2016;12(1):183–8.
71. Johannes JR, Nelson E, Bibbo M, Bagley DH. Voided urine fluorescence in situ hybridization
testing for upper tract urothelial carcinoma surveillance. J Urol. 2010;184(3):879–82.
Non-Urothelial Malignancies and Other
Miscellaneous Lesions 8
Tarik M. Elsheikh, Rana S. Hoda, Stefan E. Pambuccian,
Jae Y. Ro, and Sun Hee Sung
Introduction
Non-urothelial Malignancies
Background
NUM of the bladder are uncommon, accounting for less than 5% of all bladder
tumors, and are seldom detected by urine cytology [1, 2]. They may present as pri-
mary or secondary disease and refer to malignancies other than HGUC. NUM may
demonstrate squamous, glandular, neuroendocrine, or sarcomatoid features among
other components. Approximately 90% of all NUM are epithelial in origin, and
these include carcinomas, such as squamous cell carcinoma (SqCC), adenocarci-
noma (AdCa) and small-cell carcinoma [2]. Non-epithelial malignancies include
sarcoma, melanoma, and lymphoma, with the latter two most often metastatic to the
bladder. The pathogenesis of these cancers is incompletely understood, but several
factors are believed to be contributing, such as metaplasia, chronic inflammation,
and development from multipotent stem cells [3].
Urine cytology findings of NUM have rarely been reported, and frequently pose
diagnostic challenges to the cytologist and pathologist, due to their morphological
overlap with HGUC and its variants. In a College of American Pathologists perfor-
mance comparison study that evaluated participant responses for the reference diag-
noses of HGUC, SqCC, AdCa, and benign diagnoses (including polyomavirus
infection and ileal loop urine) [4], less than half of the participants were able to
accurately identify SqCC or AdCa; and cases recognized as “malignant” were mis-
diagnosed as HGUC by 36–48% of participants. Moreover, the cytological distinc-
tion between HGUC with divergent differentiation from pure non-HGUC may not
be possible in cytological samples or in small biopsy specimens and frequently
requires surgical resection of the tumor with extensive histologic sampling.
Primary NUM usually pursue an aggressive clinical course and often present at
an advanced stage of disease. A multimodal approach using clinical details, imaging
results, and pathological diagnosis is vital for a prompt management decision and
early therapeutic intervention. The overall survival remains poor and, apart from
SqCC, shows similar outcomes to HGUC when controlled for gender, stage, and
grade of malignancy [5, 6]. A recent analysis of 222,435 bladder cancer cases from
the Surveillance, Epidemiology, and End Results registry (SEER, 2004–2016) con-
cluded that primary SqCC had worse survival across all stages, even after multivari-
ate adjustment and matching with urothelial bladder cancers. AdCa, neuroendocrine
carcinomas, and other non-urothelial carcinomas presented at higher stage than UC
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 145
and had worse outcomes than UC when they were organ-confined, but not when
they presented at an advanced stage [7]. It remains unclear, however, whether the
prognosis of NUM is worse than invasive HGUC, based on histologic type alone
[8]. A multi-institutional study of 1131 consecutive patients (1042 HGUC and 89
NUM) reported no differences in 5-year survival following radical cystectomy with
lymph node dissection [6]; multivariate analysis demonstrated that gender, clinical
stage, pathological stage, lymph node involvement, and lymph node dissection were
the independent predictive factors for survival, whereas histological type and grade
had no significant impact on survival.
Secondary tumors of the urinary bladder, upper urinary tract, and urethra repre-
sent rare events and are encountered more commonly in autopsies than in clinical
surgical pathology or cytopathology specimens. Secondary malignancies involving
the urinary tract may result in shedding of malignant cells in the urine by direct
invasion from malignancies arising in adjacent organs (anorectum; cervix and endo-
metrium in women; and prostate in men); transcoelomic spread through the perito-
neum in gastric and ovarian malignancies; lymphatic and hematogenous spread
from breast carcinomas, pulmonary carcinomas, and melanomas; and rarely from
various other malignancies. A multi-institutional study of 1042 UC and 89 NUM
reported no difference in 5-year survival following radical cystectomy [6].
Awareness of the patient’s prior history and comparison and correlation of pathol-
ogy material from prior malignancies is necessary to diagnose metastasis to the
urinary bladder and exclude the possibility of a new primary [9].
Liquid-based preparations (LBPs), such as ThinPrep (TP) (Hologic, Marlborough,
MA) and SurePath (SP) (BD-TriPath Imaging, Burlington, NC), have been increas-
ingly used for processing urinary cytology specimens. Several studies comparing
the older conventional sediment preparations, and the Cytospin (CS) centrifugal
method, to LBPs have shown no significant differences in the sensitivity of detect-
ing malignancies, especially AdCa and SqCC [10–13]. LBPs, however, are associ-
ated with certain subtle artifacts and cytomorphologic changes that are different
from CS, and need to be recognized by cytologists [14] (further discussion is pro-
vided in Chap. 10, Cytopreparatory Techniques). Generally, LBPs provide more
uniform distribution of cells with less cellular overlapping and a cleaner back-
ground, i.e., reduced blood and inflammatory debris, while CS tends to show more
clumped clusters with better preservation of architecture [13]. There is significant
shrinkage of cells associated with LBP, resulting in smaller sized cells compared to
CS. Malignant nuclear features are comparable, including nuclear irregularity and
high N/C ratio, but there appears to be a greater degree of nuclear hyperchromasia
associated with LBP.
Finally, ancillary studies such as immunohistochemistry (IHC) have a limited
role in urine cytology, as most cases are not of sufficient cellularity to allow for cell
block preparation. However, IHC performed on cell block or cytologic preparations
from specimens with adequate cellularity may be helpful in subclassifying or con-
firming NUM in a subset of cases.
146 T. M. Elsheikh et al.
Background
Primary SqCC is the second most common malignant neoplasm of the urinary tract,
accounting for 2–5% of bladder cancers in the Western world [15]. Until recently, it
predominated in some sub-Saharan African countries (western, eastern, and south-
eastern) and Egypt, where Schistosoma haematobium infection was a major risk
factor [16]. SqCC associated with Schistosoma infection was known as bilharzial
SqCC (B-SqCC). The prevalence of B-SqCC appears to be decreasing in recent
years in Egypt and other countries, parallel with the continuing efforts to eradicate
Schistosoma infections [17]. Non-bilharzial SqCC (NB-SqCC), which is found in
areas without Schistosoma infection, has a different epidemiology, risk factor asso-
ciation, pathogenesis, and clinical behavior. In the USA, NB-SCC is more common
in the African American population and has more of a male to female preponder-
ance (4:1) than HGUC (3:1) [18]. In contrast to B-SqCC, which is a disease of
younger age, NB-SqCC occurs more commonly in older individuals in the seventh
decade of life (similar to HGUC of the bladder) [16]. NB-SqCC is associated with
keratinizing squamous metaplasia and leukoplakia secondary to recurrent chronic
infection, neurogenic bladder dysfunction, prolonged irritation from indwelling
catheters, foreign bodies, and stones. Other etiologic factors such as tobacco smok-
ing, cyclophosphamide therapy, and human papilloma virus infection have been
investigated but shown no definitive relationships [19].
Similar to HGUC, primary pure SqCC of the urinary tract most commonly pres-
ents with hematuria, dysuria and irritative symptoms. Most of these cancers are
moderately to poorly differentiated and present at an advanced stage with bladder
muscularis propria invasion. Although pathologic stage is highly correlated with
prognosis, B-SqCC appears to have a better prognosis than NB-SqCC [16].
Definition
Primary SqCC of the urinary tract is a malignant neoplasm that shows exclusively
squamous differentiation, without associated urothelial or glandular elements.
Therefore, the diagnosis of primary pure SqCC should be based on extensive biopsy
sampling or a resection specimen to exclude any squamous differentiation in asso-
ciation with invasive and/or in situ HGUC components (Fig. 8.1).
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 147
a b
Fig. 8.1 Primary pure squamous cell carcinoma (SqCC) of the urinary bladder. (a) Resection
specimen shows a muscle invasive carcinoma composed entirely of squamous components. (b)
Another case of SqCC (upper part of photo) is associated with calcified Schistosoma haematobium
eggs deposited in the stroma (lower part of photo). (H&E, medium mag.)
Cytologic Criteria
a b
Fig. 8.2 SqCC of the urinary bladder. (a) Cellular specimen displays tightly cohesive clusters of
atypical keratinized squamous cells. (Voided urine, TP, medium mag.) (b) Loosely cohesive clus-
ters consist of highly atypical keratinized and non-keratinized squamous cells. (Instrumented
urine, TP, medium mag.)
a b
Fig. 8.3 SqCC of the urinary bladder. (a) Predominately single highly atypical keratinized squa-
mous cells are large, polygonal with high N/C ratio, sharp cytoplasmic borders, and hyperchro-
matic nuclei. (Voided urine, TP, high mag.) (b) Elongated keratinized “fiber” and “tadpole cells”
may also be present, as well as inflammatory cells and some debris in the background. (Bladder
washing, TP, high mag.)
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 149
Fig. 8.4 Primary SqCC of the urinary bladder. Tightly cohesive three-dimensional clusters of
non-keratinized squamous cells. The tumor cells show nuclear enlargement, high N/C ratio,
marked overlapping, and severe hyperchromasia. Occasional nucleoli are observed. In the absence
of a keratinizing component, it is difficult to distinguish this morphology from HGUC or AdCa.
(Bladder washing, CS, high mag.)
a b
Fig. 8.5 SqCC of the urinary bladder. Non-keratinized atypical squamous cells may be arranged
in loose clusters (a) and as single cells (b). The tumor cells have a metaplastic appearance with
rigid basophilic cytoplasm and nuclear elongation with marked hyperchromasia. In the absence of
keratinization, it is difficult to distinguish HGUC from non-keratinizing SqCC of the bladder.
(Instrumented urine, TP, high mag.)
150 T. M. Elsheikh et al.
Explanatory Notes
Fig. 8.8 Atypical squamous cells (ASC). This specimen obtained from a 57-year-old female con-
tains rare clusters of keratinized squamous cells with significant atypia. A diagnosis of “ASC,
suspicious for squamous carcinoma” was rendered. Follow-up transurethral resection (TUR)
revealed Schistosoma-associated SqCC of the bladder. (Instrumented urine, TP, high mag.)
Background
The presence of atypical squamous cells (ASC) in urine cytology is a rare finding
and is seen in approximately 0.3–0.9% of urine specimens [20, 21]. There is a pau-
city of literature regarding the clinical significance of ASC, but what exists suggests
that any degree of squamous atypia should not be discounted. Bladder surface
mucosa frequently shows keratinizing squamous metaplasia with or without squa-
mous dysplasia. The majority of ASC are not associated with malignancy and rep-
resent reactive/inflammatory changes, most commonly due to vaginal contamination
or exfoliation from distal urethra. However, ASC may be associated with bladder or
cervical cancers in up to 20–30% of cases [20]. Association with carcinoma is even
152 T. M. Elsheikh et al.
higher in the presence of severe squamous atypia that is suspicious for carcinoma
[21]. In women, cystoscopy and biopsy and/or colposcopy may be needed to deter-
mine a bladder or lower female genital tract origin of the squamous atypia.
A diagnosis of ASC should be rendered in the presence of significant atypia that falls
short of a definitive diagnosis of malignancy, i.e., significant atypia with insufficient cel-
lularity or equivocal malignant cytologic features. There appears to be no substantial
association between the number of ASC and histological diagnosis, age, sex, or speci-
men type. However, the presence of highly suspicious cytology, i.e., extreme squamous
atypia and pleomorphism, can be associated with cancer in up to 60% of cases [18, 19].
Malignant conditions associated with ASC include primary bladder SqCC, cervical
SqCC, and HGUC with squamous differentiation. Occasionally isolated ASC may be
the only cytologic clue to the presence of an undetected or unsampled HGUC [22].
ASC with cytologic features of low-grade squamous intraepithelial lesion
(LSIL), including koilocytosis/HPV changes, most commonly originate from the
female genital tract and urethra in men [23]. There are no well-defined guidelines
regarding the management of these patients, but patients with persistent abnormali-
ties and/or immunosuppression should be closely followed. In women, LSIL in
urine cytology is highly associated with HPV infection in the genital tract; there-
fore, these findings should be correlated with a corresponding Pap test. Men, with
LSIL in their urine, should be considered for genitourinary examination to evaluate
for genital warts [23]. Benign conditions associated with ASC include squamous
metaplasia of bladder mucosa, radiation therapy effect, and contamination from
distal urethral or genital tract mucosa. Atypical parakeratosis (dyskeratosis) may be
associated with condyloma of bladder or urethra, squamous papilloma, dysplasia,
and bladder irritation or may represent a genital contaminant [24].
More recently, HPV testing using noninvasive techniques such as self-collected
urine and cervicovaginal samplings has been investigated as a possible alternative to
clinician-collected cervicovaginal samples and as means of increasing participation
in screening programs. In countries without well-developed cervical cancer screen-
ing programs, participation rate is low and up to 50–80% of women are not screened
[25]. Noninvasive urine-based high-risk HPV (hrHPV) testing can be an attractive
option for women who are reluctant to undergo gynecologic examination, as self-
collected samples seem to be preferred over clinician-collected ones [26].
Furthermore, urine samples could potentially be collected at home and sent by mail
to the laboratory for hrHPV testing, removing the need for an initial clinic-based
pelvic exam. Whereas most studies have shown that self-collected vaginal samples
appear to have similar sensitivity to clinician-collected cervical samples, results of
urine HPV testing have been variable and inconsistent [27]. Reported sensitivities
of urine hrHPV testing for detection of HGSIL+ ranged from 45% to 80–100%, and
specificities ranged from 25 to 53% among patients referred for colposcopy, which
are significantly lower than self-collected and provider-collected cervicovaginal
samples (94% sensitivity) [28]. Although, based on available literature, urine HPV
testing appears to be a promising secondary option to cervicovaginal sampling, fur-
ther studies are needed to evaluate and improve the accuracy of urine hrHPV testing
for detection of HGSIL+, especially in a primary screening population and/or in
women with limited access to health care.
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 153
Definition
Keratinized squamous cells with large and hyperchromatic nuclei, high N/C ratio,
abnormal nuclear or cytoplasmic shapes, and densely orangophilic cytoplasm. The
atypical cells may occur as groups or isolated cells but lack qualitative and/or quan-
titative unequivocal criteria of malignancy.
Explanatory Notes
Explanatory Note 1 ASC are defined as atypical keratinized cells, since cytologic
distinction between non-keratinized SqCC and HGUC is often difficult and some-
times impossible (Figs. 8.8 and 8.9).
Explanatory Note 2 The presence of ASC should be noted in the report and clini-
cians made aware of their significance. They are particularly clinically significant if
they persist in subsequent specimens.
Fig. 8.9 Atypical squamous cells (ASC), suspicious for squamous cell carcinoma (SqCC). This
patient was recently diagnosed with SqCC of the penile meatal opening. The specimen shows
singly scattered atypical non-keratinized squamous cells with high N/C ratio and marked hyper-
chromasia. Absent the history, it would be impossible to distinguish these atypical non-keratinized
squamous cells from HGUC. (Voided urine, TP, high mag.)
154 T. M. Elsheikh et al.
a b
Fig. 8.10 Atypical squamous cells (ASC). (a) Atypical keratinized and non-keratinized cells have
moderate atypia, and increased N/C ratios. The nuclei are enlarged and hyperchromatic, but have
a smudgy, degenerated appearance. (Bladder washing, TP, medium mag.) (b) Follow-up resection
revealed a noninvasive papillary HGUC with squamous differentiation. Inset shows surface kera-
tinized squamous cells, identical to those observed in the cytology specimen. (H&E, low mag.;
inset, high mag)
a b
Fig. 8.11 Atypical squamous cells (ASC). (a) Rare keratinized ASC with mild to moderate atypia
is present in this specimen from a 50-year-old female. (Voided urine, TP, high mag.). (b) Follow-up
Pap test reveals low-grade squamous intraepithelial lesion and positivity for high-risk HPV. (Pap
test, TP, medium mag.). ASC in this urine specimen represented contamination from the geni-
tal tract
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 155
Primary Adenocarcinoma
Background
AdCa may be primary or metastatic. Primary AdCa is rare, accounting for 0.5–2%
of all bladder malignancies, whereas HGUC with glandular differentiation is more
prevalent [29]. Histologically, primary AdCa has a pure glandular phenotype and
includes bladder AdCa and urachal AdCa, which are indistinguishable cytohisto-
logically. Urachal AdCa, however, tends to be located in the bladder dome and/or
anterior wall with an epicenter in the bladder wall, no association with surface intes-
tinal metaplasia, and typically affects a younger patient population [15, 30]. Primary
AdCa is similar to HGUC in age and gender distribution, with peak incidence in
sixth decade, and is more common in males [31]. Clinical features at presentation
include hematuria, dysuria, urinary frequency, and rarely mucosuria. Risk factors
for primary AdCa include intestinal metaplasia, particularly in patients with bladder
exstrophy, schistosomiasis, diverticula, and chronic irritation and obstruction [15,
30, 32]. Prognosis is best predicted by pathological stage but overall tends to
be poor.
Primary AdCa is histologically subclassified as enteric, mucinous, mixed enteric-
mucinous, and AdCa, not otherwise specified (AdCa-NOS); mixed enteric-mucinous
is the most common subtype. Enteric AdCa is morphologically identical to its intes-
tinal counterpart, while mucinous AdCa shows nests of tumor cells associated with
abundant mucin and may show variable signet ring cell morphology. AdCa-NOS
terminology is used when morphology does not resemble a specific subtype [31,
33]. Clear cell carcinoma is a rare distinctive variant of AdCa that is seen more com-
monly in females and in urethra more than bladder [24]. Divergent glandular dif-
ferentiation has been reported in up to 18% of invasive HGUC, so it should be
excluded before establishing a diagnosis of pure AdCa [34]. Secondary involvement
of the bladder and upper urinary tract by AdCa is more common than primary uri-
nary tract adenocarcinomas and may occur by direct extension from adjacent organs
or metastasis from distant sites (see later discussion of secondary epithelial malig-
nancies). Cytologic examination alone cannot distinguish primary from secondary
AdCa; therefore, clinical and pathologic correlation is warranted.
Definition
Criteria
• Variable cellularity.
• Enteric AdCa shows similar morphology to colonic AdCa (Fig. 8.12), including
columnar cell clusters and single degenerated cells in a background of necrosis
and mucin. Nuclei are large and elongated, vesicular or hyperchromatic, with
irregular shapes, and visible or prominent nucleoli. The cytoplasm may be vacu-
olated (Fig. 8.13a).
• Mucinous AdCa shows rounded three-dimensional clusters of crowded cells,
with variable atypia, small to moderate amount of delicate cytoplasm, occasional
cytoplasmic vacuoles, and medium-sized nuclei with distinct nucleoli. Mucin is
present in the background. Signet ring cells may be present, displaying large
cytoplasmic mucin-containing vacuoles that displace and push crescent-shaped
hyperchromatic nuclei to the periphery of the cells. Signet ring cells may also be
associated with enteric AdCa or AdCa-NOS (Fig. 8.13b).
a b
Fig. 8.12 Primary enteric adenocarcinoma (AdCa) of the urinary bladder. (a) Cytology demon-
strates a cluster of cells with enlarged elongated palisading nuclei, hyperchromasia, irregular
nuclear membranes, and prominent nucleoli. (Voided urine, TP, high mag.). (b) Resected bladder
shows AdCa exhibiting similar morphology to its colorectal counterpart. (H&E, medium mag.)
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 157
a b
Fig. 8.13 Primary enteric AdCa of the urinary bladder. In this case, the malignant cells have more
voluminous cytoplasm with prominent vacuolization, but characteristic palisading morphology is
not demonstrated. (a) There is elongation of the nuclei with significant pleomorphism. (b) Some
of the cells display signet ring appearance, created by a large cytoplasmic vacuole pushing the
crescent-shaped hyperchromatic nucleus to the periphery of the cell. (a, b: Catheterized urine, CS,
high mag.). Although histologic resection demonstrated enteric AdCa, cytologic features in this
urinary cytology do not allow for determination of subtype of AdCa
• Clear cell AdCa presents in clusters and has large cells with abundant hypervacu-
olated cytoplasm, centrally located nuclei, prominent nucleoli, and occasional
hobnail configurations (Fig. 8.14).
• In most instances, AdCa does not show distinctive cytomorphologic features to
allow for determination of the subtype. These cases are best reported as AdCa-
NOS, with appropriate differential diagnosis listed in an accompanying com-
ment (Figs. 8.13 and 8.15).
158 T. M. Elsheikh et al.
a b
Fig. 8.14 Clear cell AdCa of the urethra. (a) Specimens are usually of high cellularity with cohe-
sive clusters of malignant cells. The tumor cells have abundant vacuolated and/or eosinophilic
cytoplasm, centrally placed nuclei, and prominent nucleoli. (Bladder washing, TP, high mag.). (b)
Resection revealed clear cell AdCa of the urethra invading the bladder. The tumor had a tubulo-
papillary architecture, lined by highly pleomorphic cells with abundant eosinophilic or clear cyto-
plasm, and frequent hobnailing. (H&E, high mag.)
Explanatory Notes
a b
Fig. 8.16 Atypical glandular cells (AGC). (a) This urine specimen from a 69-year-old male con-
tains rare clusters of glandular cells with mild atypia, including slightly enlarged palisaded nuclei,
and nuclear hyperchromasia. Inflammatory cells are present in the background. A diagnosis of
AGC was rendered in this case, with a comment/note stating that the differential diagnosis included
benign nonneoplastic conditions versus low-grade AdCa. Follow-up (not shown) revealed florid
cystitis glandularis. (Instrumented urine, CS, high mag.). (b) This specimen, from a 73-year-old
female, contains a single cluster of highly atypical glandular cells. No history was available at the
time of evaluation of the cytology specimen. A diagnosis of “AGC, suspicious for AdCa” was
rendered, with appropriate differential diagnosis listed. Follow-up revealed that the patient had
stage 4 ovarian serous carcinoma; therefore, this most likely represented secondary involvement of
the urinary tract. (Voided urine, TP, high mag.)
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 161
a b
Fig. 8.17 AdCa-NOS. (a) This urine from a 67-year-old female contains single and loosely cohe-
sive atypical cells with enlarged hyperchromatic nuclei, high N/C ratio, and marked pleomor-
phism. The case was signed out as “HGUC.” (Voided urine, CS, high mag.). (b) Follow-up
cystectomy revealed primary AdCa of the urinary bladder, enteric type. (H&E, medium mag.). In
some instances, the AdCa cells tend to round up in cytologic specimens, and it may be extremely
difficult to distinguish them from HGUC. Therefore, rendering HGUC diagnosis was not unrea-
sonable in this case, since it resulted in further workup of the patient
Neuroendocrine Tumors
Background
Neuroendocrine tumors are a diverse group of bladder tumors that include well-
differentiated neuroendocrine tumors (WDNET), neuroendocrine carcinomas
(NEC), and paraganglioma. WDNET has been previously referred to as carcinoid,
but that terminology is currently not recommended [40]. NECs are rare and consist
of small-cell carcinoma (SmCC) and large-cell NEC (LCNEC). SmCC is the most
common of these cancers and accounts for less than 1% of malignant bladder
tumors. In addition to bladder, NEC can involve the kidney and prostate and are
morphologically, immunohistochemically, and ultrastructurally similar to their pul-
monary counterparts. NEC can also involve the bladder secondarily. Only a few
cases of LCNEC and WDNET of bladder have been reported [40].
SmCC is often associated with HGUC and other divergent histologic subtypes,
suggesting a urothelial origin. In over half of the cases, there is association with
other carcinomatous components, such as in situ or invasive HGUC, SqCC, AdCa,
sarcomatoid carcinoma, or mixtures of these components [40, 41]. This supports the
162 T. M. Elsheikh et al.
most widely accepted view that NEC of the bladder originates from a pluripotential
stem cell that can differentiate into various cell types, suggesting a common clonal
origin of the neuroendocrine, urothelial, squamous, and AdCa components. Like
pulmonary SmCC, it is closely associated with tobacco smoking and tends to pursue
an aggressive course and present at an advanced stage. Differential diagnosis
includes metastatic SmCC, which is very rare, lymphoma, and HGUC. IHC profile
commonly includes the expression of chromogranin, synaptophysin, and CD56,
while TTF1 is expressed in about 50% of cases. As opposed to high-grade UC,
SmCC shows variable GATA3 expression and is negative for p63 and CK20 [40].
LCNEC and WDNET are extremely rare and unlikely to be encountered in urine
cytology. LCNEC of the bladder is a high-grade malignancy that exhibits neuroen-
docrine features by light microscopy and increased mitotic activity and expresses
neuroendocrine markers. Only a few cases have been reported in the surgical pathol-
ogy literature, and some of these were associated with HGUC [42]. Most cases have
followed an aggressive course. Differential diagnosis includes HGUC, SqCC, and
AdCa-NOS. WDNETs of the urinary bladder are exceedingly rare with few docu-
mented cases reported in the surgical pathology literature. Age range is similar to
HGUC, and the most common presenting symptom is hematuria. Morphologically
it is similar to its counterparts in other sites and consists of uniform cells with round
nuclei that may be eccentrically placed, salt and pepper chromatin, and minimal to
moderate amount of cytoplasm. WDNET can be confused with well-differentiated
AdCa or nonneoplastic lesions. Other differential diagnostic considerations include
paraganglioma, HGUC, and lymphoma [43]. The tumor cells of WDNET show
positive staining for neuroendocrine markers and occasionally PSA but no other
prostatic markers [40]. Paraganglioma is discussed later with “other miscellaneous
lesions.”
Definition of SmCC
Criteria
a b
Fig. 8.18 Primary small-cell carcinoma (SmCC) of the urinary bladder. (a) Urine cytology speci-
men is highly cellular and shows tightly cohesive clusters of small, individual cells. The tumor
cells have minimal amount of cytoplasm, round to oval nuclei, and salt and pepper chromatin pat-
tern. Nucleoli are absent or inconspicuous. Individual cell necrosis and apoptosis are also noticed.
Background is bloody with necrotic debris. (Voided urine, CS, medium mag.). (b) Resection speci-
men revealed SmCC with histologic appearance similar to its pulmonary counterpart. The tumor
cells were positive for neuroendocrine markers (not shown) and had no associated divergent his-
tologies. (H&E, high mag.)
• Tumor cells are small to medium in size (two to three times the size of lympho-
cytes or red blood cells), with scanty cytoplasm and high N/C ratio (Fig. 8.19).
• Nuclei are round to oval, hyperchromatic with finely granular evenly distributed
or smudged chromatin, ill-defined membranes, and focal to prominent nuclear
molding. Nucleoli are absent or inconspicuous (Fig. 8.19).
• In LBPs, compared to conventional preparations, tumor cells are often smaller
and rounder, and crush artifact or molding may be less apparent [34] (Fig. 8.20).
• Single-cell necrosis (apoptosis) is present and may be extensive.
• Background may be hemorrhagic and necrotic (Fig. 8.18).
164 T. M. Elsheikh et al.
a b
Fig. 8.19 SmCC of the bladder. (a) Cytology displays cells arranged in a loosely cohesive cluster.
Tumor cells are small, show high N/C ratio, round to oval hyperchromatic nuclei with evenly dis-
tributed fine to coarse chromatin, and absent or inconspicuous nucleoli. Nuclear overlapping and
spindling are more pronounced than molding, and the background is clean, compared to Cytospin
preparations (see Fig. 8.18a). (Voided urine, TP, high mag.). (b) In this sample, the cytologic fea-
tures are similar to “a,” but nuclear molding and linear arrangement of the malignant cells are more
pronounced. (Bladder washing, SurePath, high mag.)
Explanatory Notes
Explanatory Note 1 Urine cytology has a low sensitivity and specificity for detect-
ing SmCC, particularly when cellularity is sparse or in cases intermixed with
HGUC. The diagnosis of SmCC, however, can be suggested or made when classic
cytomorphology is appreciated, which is similar to the exfoliative cytology of
SmCC from other sites [10, 41, 44, 45].
a b
Fig. 8.20 Primary SmCC. (a) Urine cytology from a 76-year-old man shows a predominantly
discohesive cell pattern. The tumor cells are small (approximately two to three times the size of
RBCs or neutrophils) and have scant cytoplasm. The nuclei are round and have finely to coarsely
granular chromatin. Single-cell necrosis is present. No nuclear molding or spindling is appreci-
ated. This morphology also raises the differential diagnosis of lymphoma, melanoma, and
HGUC. (Voided urine, TP, high mag.). (b) Immunohistochemistry performed on TP slides demon-
strates positive staining with synaptophysin and chromogranin. (Synaptophysin, high mag.)
Fig. 8.21 Non-keratinizing SqCC. Voided urine from a 48-year-old man with locally advanced
anal squamous cell carcinoma. The cytology consists of tightly cohesive clusters of basaloid tumor
cells with overlapping round to oval, hyperchromatic and irregular nuclei, and ill-defined cyto-
plasm. This cytologic appearance can be confused with SmCC, especially in the absence of clinical
history or previous pathology for comparison. Features favoring SqCC over SmCC are the open
chromatin pattern of nuclei and presence of conspicuous nucleoli. (Voided urine, CS, high mag.)
166 T. M. Elsheikh et al.
Background
Secondary malignancies rarely involve the urinary tract, often presenting as a soli-
tary mass and accounting for approximately 2% of surgically resected bladder
tumors. Most of those malignancies (approximately 70%) represent direct extension
from adjacent organs, such as colorectum, prostate, and female genital tract [48]. A
smaller proportion of secondary tumors (<30%) represent metastases from distant
sites, via lympho-hematogenous spread [49]. The site of the tumor within the bladder
may suggest its origin, for example, cancers of prostate and cervix tend to invade the
bladder neck and trigone, while colorectal cancers more commonly involve the fun-
dus [50]. Distant metastases to the bladder are typically a late manifestation of dis-
seminated disease and more commonly occur with cancers of the breast, stomach,
and lung. Lymphoma and melanoma can also involve the bladder secondarily and are
discussed later with “non-epithelial malignancies.” Other reported primary sites of
origin include the kidney, pancreas, and ovary [49]. The urothelium is usually spared
in metastatic disease, which provides an important clue to the histologic diagnosis,
but may also explain the rarity of exfoliated malignant cells in urine cytology [51].
Whereas the cytohistologic appearance of secondary malignancies may be distinc-
tive, many extravesical cancers can mimic primary urothelial or non-urothelial malig-
nancies. This is especially problematic with AdCa, which represents over 50% of all
secondary bladder malignancies [48]. The rarity of secondary carcinomas and overlap
of their cytomorphologic features with those of primary carcinomas make an accurate
diagnosis of metastasis challenging. In a 10-year retrospective review of all urine
cytologies from biopsy-proven NUM patients, 25 urine samples from 14 patients
showed metastases, accounting for <1% of all urine specimens. Most cases were rec-
ognized as being atypical or malignant but were frequently mistaken for a urothelial
origin [36]. Although the morphologic overlap in some cases may not allow for distin-
guishing NUM from HGUC, especially if there is no previous history of malignancy,
malignant cells from NUM in urine may signify the initial presentation of metastasis
[49]. Therefore, familiarity with characteristic cytologic features that may serve as
clues, review of pertinent clinical information and histologic samples, and application
of ancillary studies can lead to a more specific diagnosis in a subset of these cases.
Colorectal AdCa
Colorectal AdCa is the most frequent of all secondary bladder tumors and tends to
involve the bladder by means of direct extension. Approximately 21% and 12% of
secondary malignancies are from the colon and rectum, respectively [48]. Typical
cytologic features include pleomorphic cells with elongated nuclei, irregular nuclear
borders, coarse chromatin, and often inconspicuous nucleoli. The cytoplasm is usu-
ally vacuolated (Fig. 8.22). Often there is background necrosis, but this is best
observed in conventional preparations, as this tends to be difficult to appreciate and
minimal to absent in LBP [36]. Cytomorphologic and IHC features of colorectal
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 167
a b
Fig. 8.22 Colorectal AdCa. (a) Clusters of pleomorphic columnar cells display elongated irregu-
lar nuclei and coarse chromatin. The cytoplasm is vacuolated, with occasional signet ring cells.
The background shows necrosis and inflammatory cells. Follow-up revealed rectal AdCa directly
invading the bladder. (Bladder washing, SurePath, medium mag.). (b) This patient had metastatic
colon AdCa to the bladder diagnosed via bladder biopsy, collected at the same time with urine
cytology. The malignant cells are columnar with elongated hyperchromatic nuclei and vacuolated
cytoplasm. Notice that the background in TP tends to be cleaner than that of SurePath or Cytospins.
(Bladder wash, TP, high mag.). Cytologic features of colorectal AdCa overlap with those of pri-
mary bladder AdCa, enteric type
AdCa overlap with primary bladder AdCa, enteric and mucinous types [49]. In the
absence of clinical history, high-grade colorectal AdCa is not uncommonly misdi-
agnosed as HGUC. The combination of atypical columnar cells and necrosis are the
best clues to suspecting colorectal carcinoma. IHC markers such as CK20, beta-
catenin, CDX2, and SATB2, all of which are usually positive, may help clarify the
differential diagnosis (see previous discussion of primary AdCa) [9, 52].
Prostate AdCa
Prostate cancer is the second most common non-primary malignancy of the urinary
tract and accounts for approximately 20% of secondary malignancies [48]. It tends
to involve the bladder and urethra by either direct extension or lymphatic spread
[36]. In urine cytology, prostatic AdCa is almost always associated with clinically
advanced stage, high Gleason score, and a known history [53]. Hematuria and outlet
obstruction are often presenting symptoms [54].
Characteristic cytologic features of lower-grade prostatic AdCa include large
cells with uniform appearance, arranged in cohesive groups and syncytia. The
168 T. M. Elsheikh et al.
a b
Fig. 8.23 Prostate AdCa. (a) Characteristic cytologic features of prostatic AdCa are best appreci-
ated in lower Gleason score carcinomas. The tumor cells are large, arranged in cohesive groups,
and have a uniform appearance. The cells have abundant delicate cytoplasm, round to oval nuclei,
fine chromatin, and prominent nucleoli. (Instrumented urine, TP, high mag.). (b) In higher Gleason
score AdCa, the tumor cells tend to have less voluminous cytoplasm but retain the uniform appear-
ance and prominence of nucleoli. There is a suggestion of acinar formation at the periphery of the
cluster. Follow-up biopsy (not shown) revealed prostatic AdCa, Gleason score 8. (Bladder wash-
ing, SurePath, high mag.)
malignant cells have delicate vacuolated cytoplasm, round to oval nuclei that may
be eccentrically placed, fine chromatin, and prominent nucleoli (Fig. 8.23a).
Occasionally, acinar formation is appreciated (Fig. 8.23b). In some cases, however,
cytomorphology may be difficult to distinguish from HGUC (Fig. 8.24), especially
when the prostatic carcinoma is of high Gleason score, of ductal type, or has hyper-
chromatic nuclei without nucleolar prominence [36, 48, 53]. In contrast to HGUC,
however, high-grade prostate AdCa tends to still retain relatively uniform nuclei and
prominent nucleoli. IHC can be of value, as prostate AdCa expresses NKX3.1 and
PSA (Fig. 8.24), while HGUC expresses p63 and GATA3, but the latter two markers
are often lost in poorly differentiated HGUC [49, 50, 54].
Gynecologic Malignancies
a b
c d
Fig. 8.24 Prostate AdCa. This specimen is from an 80-year-old male, who presented with urinary
retention and elevated serum PSA levels. (a) Urine cytology is hypercellular and contains a disco-
hesive population of intermediate sized cells with minimal dense cytoplasm. The nuclei are round
with conspicuous nucleoli and relative uniformity. (Voided urine, TP, high mag.). (b) The cell block
reveals similar cytologic features, including hyperchromatic nuclei and occasional prominent
nucleoli. Differential diagnosis includes HGUC, prostatic AdCa, poorly differentiated carcinoma-
NOS, and melanoma. (Cell block of voided urine, H&E, high mag.). (c) There is strong positive
staining of the tumor cells with NKX3.1 and PSA and negative staining with P63 and GATA3.
(NKX3.1 immunohistochemistry, high mag.). (d) Follow-up prostate biopsies confirm AdCa,
Gleason score 10-grade group 5. (H&E, high mag.)
Fig. 8.25 Cervical
SqCC. This 51-year-old
female presented with
malignant squamous cells
in the urine cytology.
There is focal suggestion
of keratinization. It is not
possible to distinguish
between primary and
secondary SqCC or HGUC
with squamous
differentiation, based on
cytologic exam alone.
Clinical correlation is
needed in such instances.
Clinical follow-up revealed
locally advanced SqCC of
the cervix. (Instrumented
urine, TP, high mag.)
a b
Fig. 8.26 Endometrial serous AdCa. This urine cytology is from a 61-year-old female. (a) There
are many tightly cohesive clusters of large pleomorphic cells, with marked nuclear irregularity and
high N/C ratio. A papillary configuration is suggested. (Voided urine, TP, high mag). (b) An endo-
metrial biopsy, performed 1 week prior, demonstrated serous papillary AdCa, FIGO grade 3.
(H&E, high mag.) Therefore, the cytologic features are consistent with metastatic high-grade
serous carcinoma
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 171
Breast AdCa
Breast AdCa is one of the more common primary sites responsible for distant metas-
tases to the bladder and may be of lobular or ductal type. It accounts for approxi-
mately 2.5% of secondary malignancies involving the urinary bladder [48] and is
seldom reported in urine cytology [32, 58]. In some instances, the bladder may be
the sole organ involved by metastatic breast carcinoma. Lobular carcinoma tumor
cells may be arranged in a linear (single file) configuration but more often present
as single cells or small groups. Tumor cells are small with high N/C ratio, eccentri-
cally placed nuclei, and small or inconspicuous nucleoli (Fig. 8.27). Occasionally,
the center of the cytoplasmic vacuole shows mucin condensation (“bull’s eye” or
“targetoid” body). Lobular carcinoma of breast can resemble primary signet ring
cell AdCa, plasmacytoid UC, melanoma, neuroendocrine tumors, and plasmacy-
toma. Ductal carcinoma is rarely encountered in the urine and presents as tightly
cohesive clusters of atypical cells with somewhat uniform appearance (Fig.8.28).
a b
Fig. 8.27 Breast lobular carcinoma. This patient is a 48-year-old woman with a remote history of
lobular carcinoma of the breast. (a) Urine cytology reveals discohesive small to intermediate sized
cells, with high N/C ratios, hyperchromatic nuclei, and small nucleoli. Some cells have eccentri-
cally placed nuclei with a plasmacytoid appearance. Immunohistochemistry performed on the cell
block (not shown) revealed positive staining of the tumor cells for gross cystic disease fluid pro-
tein-15 (GCDFP-15). (Voided urine, TP, high mag.). (b) Follow-up transurethral resection revealed
metastatic lobular carcinoma diffusely invading the urinary bladder and undermining and penetrat-
ing the surface urothelium. (H&E, medium mag.)
172 T. M. Elsheikh et al.
a b
Fig. 8.28 Metastatic breast ductal AdCa. Both of these patients had a previous history of invasive
breast ducal carcinoma. (a) In this first patient, the tumor cells are forming a cellular sphere (“mor-
ula”), which is commonly seen in metastatic breast ductal carcinomas involving exfoliative fluid
cytologies. The cluster has a sharp community border, and tumor cells are relatively uniform.
(Voided urine, TP, high mag.). (b) In this second case, the tumor cells are more loosely cohesive
with suggestion of acinar formation but are not forming spheres. The neoplastic cells have a mod-
erate amount of cytoplasm, round to oval nuclei, and prominent nucleoli. (Voided urine, CS,
high mag.)
Gastric AdCa
Metastatic gastric AdCa to the urinary bladder and upper urinary tract is rare,
accounting for less than 5% of secondary malignancies of the bladder. The bladder
tends to be involved in advanced stages of disease, and most cases are associated
with peritoneal dissemination. Sporadic case reports describe urine cytologic
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 173
Fig. 8.29 Metastatic
gastric signet ring cell
AdCa from a 78-year-old
male with advanced stage
disease, including
peritoneal dissemination.
The tumor cells are
small- to medium-sized,
are discohesive, and have
cytoplasmic vacuoles that
indent and displace the
nuclei to the periphery.
(Catheterized urine, TP,
high mag.)
Lung Carcinoma
While lung carcinoma is a common form of cancer, metastases to the bladder are
relatively uncommon, accounting for <3% of secondary malignancies. AdCa is the
most common histologic type of pulmonary carcinoma metastatic to the bladder and
upper urinary tract, followed by SqCC [61]. The efficacy of urine cytology in assess-
ment of metastatic lung carcinoma is unclear, as it may be associated with signifi-
cant false negative results [62]. Cytomorphologic features alone cannot distinguish
metastatic lung AdCa or SqCC from primary bladder or other metastatic carcino-
mas; therefore, careful evaluation of the clinical history is essential. IHC may be
helpful in a subset of cases, as TTF1 and napsin A positivity are consistent with lung
AdCa, whereas primary bladder AdCa is not expected to stain for these markers
(Fig. 8.30).
174 T. M. Elsheikh et al.
a b
Fig. 8.30 Lung AdCa. This specimen is from a 91-year-old man with stage 4 lung AdCa. (a) Urine
cytology reveals a few clusters of large pleomorphic malignant cells with abundant delicate cyto-
plasm. The nuclei are round to oval shaped, have open chromatin, and have prominent nucleoli.
Based on cytologic evaluation alone, it is not possible to determine the primary origin of AdCa.
(Voided urine, TP, high mag.). (b) TTF1 IHC, performed on cell block, shows positive staining of
the malignant cells, consistent with clinical history of lung AdCa. (TTF1, high mag.)
Metastatic renal cell carcinoma (RCC) to the urinary bladder is unusual and it is
exceedingly uncommon for the malignant cells to be detected in urine cytology.
This is in large part due to the fact that most RCCs are currently detected by imag-
ing studies before they erode and shed cells into the urine [53]. Therefore, urine
cytology plays no significant role in the diagnosis of RCC [63]. As expected, most
reported metastatic RCCs to the bladder are of clear cell type [9]. Cytomorphologic
findings include large tumor cells with abundant hypervacuolated clear cytoplasm,
round to oval nuclei, and occasionally prominent nucleoli (Fig. 8.31). Granular
eosinophilic cells with pyknotic nuclei and indistinct cytoplasmic borders are
believed to be the results of degenerative changes caused by the urinary environ-
ment. Differential diagnosis includes HGUC with clear cell features and clear cell
variant of primary AdCa. In many instances the malignant cells are very degener-
ated and difficult to separate from benign renal tubular cells or reactive urothelial
cells [63] (Fig. 8.32).
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 175
a b
Fig. 8.31 Renal cell carcinoma (RCC). This urine specimen is from a patient with an established
history of advanced RCC. (a) On lower power exam, there are cohesive clusters of large tumor
cells with abundant cytoplasm that has a hypervacuolated and/or granular appearance. (b) High
power illustrates the cytoplasmic quality and enlarged round nuclei with large prominent nucleoli.
(a, b: Catheterized urines, CS, medium and high mag.)
a b
Fig. 8.32 Atypical cells suspicious for RCC. The specimen is from a 79-year-old man without
clinical history at the time of evaluation of urine cytology. (a) Rare discohesive large cells with
abundant granular cytoplasm are seen. The nuclei have small, distinct nucleoli, but no significant
pleomorphism. The background is bloody. This case was initially reported out as reactive/repara-
tive changes. (Voided urine, CS, high mag.). (b) Subsequently provided clinical history and review
of surgical resection revealed an advanced RCC that eroded into the renal pelvis. (H&E, high
mag.). In retrospect, these atypical cells most likely represented RCC
176 T. M. Elsheikh et al.
Non-epithelial Malignancies
Background
Non-epithelial malignancies account for less than 0.5% of all bladder tumors and
may primarily or secondarily involve the urinary tract. They include sarcoma, mela-
noma, and lymphoma [8]. Due to their rarity, inexperience and/or lack of awareness
of cytologists with their occurrence in this setting, and tendency for these tumors to
involve the bladder wall rather than the mucosal surface, they are seldomly diag-
nosed in urine cytology specimens. Pathogenesis of primary non-epithelial malig-
nancies is incompletely understood, but several factors are believed to be contributing
agents, such as metaplasia, chronic infection, and development from multipotent
stem cells [64].
Sarcoma
Overall, primary sarcomas of the urinary tract are exceptionally rare, except for
embryonal rhabdomyosarcoma in the pediatric age group.
Leiomyosarcoma is the most common urinary bladder sarcoma in adults,
accounting for 1% of all bladder malignancies, and most often has high-grade mor-
phology [64]. Patients usually present with hematuria and a palpable pelvic mass,
and mean age at presentation is 65 years. Leiomyosarcoma has been reported to be
often associated with prior cyclophosphamide therapy for neoplastic and nonneo-
plastic conditions.
In adults, rhabdomyosarcomatous differentiation occurs more commonly as a
component of sarcomatoid HGUC; only rare cases of urinary tract pure rhabdomyo-
sarcoma or angiosarcoma have been reported in the literature [65]. Adult sarcomas
generally share an extremely aggressive biologic behavior, regardless of the histo-
logic subtype.
Diagnosis of sarcoma by urine cytology is seldom made, taking into account the
rarity of those lesions in the urinary tract, infrequent exfoliation, and the lack of
experience of cytologists with their appearance. Only a few cases have been reported
in the English literature [66, 67]. High-grade sarcoma is easier to detect since the
tumor cells tend to have a spindle or epithelioid appearance and show significant
cytologic atypia and pleomorphism. In the absence of significant atypia, differential
diagnosis of spindle cell lesions includes low-grade sarcomas and benign nonneo-
plastic and neoplastic processes such as postoperative spindle cell proliferation and
inflammatory myofibroblastic tumor (also known as inflammatory pseudotumor),
which cannot be distinguished by urine cytologic evaluation.
Definition
Criteria
a b
Fig. 8.33 Leiomyosarcoma of the urinary bladder. (a) Urinary cytology shows fragments of atypi-
cal spindle cells amid stroma containing inflammatory cells. The cells contain large elongated
(“cigar”-shaped) mildly hyperchromatic nuclei with a moderate amount of ill-defined cytoplasm.
(Bladder washing, CS, medium mag.). (b) Histology of the resected leiomyosarcoma demonstrates
bundles of spindle-shaped cells in profile and cross section with elongated and pleomorphic nuclei.
(Biopsy, H&E, medium mag.)
178 T. M. Elsheikh et al.
a b
Fig. 8.34 Pleomorphic rhabdomyosarcoma of the bladder. (a) Cytology demonstrates numerous
highly pleomorphic and bizarre malignant cells with spindle and epithelioid features. Individual
cell necrosis is evident in the background. Differential diagnosis based on this morphology
includes any pleomorphic malignancy, including sarcoma, sarcomatoid carcinoma, and melanoma.
(Bladder washing, SP, medium mag.). (b) Histology shows pleomorphic rhabdomyosarcoma.
(H&E, high mag.)
Explanatory Notes
Explanatory Note 3 In general terms, the presence of atypical spindle cells in urine
indicates the need for a more invasive diagnostic procedure to arrive at a precise
histologic diagnosis or subclassification, as a comprehensive panel of IHC stains
may not be feasible on a limited cytological sample [68].
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 179
Fig. 8.35 Sarcomatoid
HGUC. This 63-year-old
male presented with gross
hematuria. Urine cytology
contains atypical spindle
cells and individual cell
necrosis. This case was
reported as “high grade
spindle cell malignancy,
can’t exclude HGUC.”
(Voided urine, TP, high
mag.). Follow-up
cystoscopy and TURP
showed a muscle invasive
sarcomatoid HGUC
Malignant Melanoma
Background
Malignant melanoma of the urinary tract is very rare, accounting for less than 1% of
all melanomas. Most melanomas are metastatic from another site and tend to involve
the kidney and bladder [36]. Some cutaneous melanomas of the glans penis or vulva
extend into the urethra and urinary bladder. Primary melanoma is postulated to arise
from neural crest cells and most commonly arises in the urethra, whereas only few
cases have been reported in the urinary bladder [69]. Metastatic melanoma accounts
for approximately 4% of all secondary malignancies of the bladder and generally
has a poor prognosis [48]. Hematuria is the most common presenting symptom.
Definition
Criteria (Fig. 8.36)
• Individually scattered large atypical cells with round to oval nuclei and abundant
cytoplasm.
• The malignant cells may have epithelioid or spindle appearance.
• Nuclei are large, pleomorphic, and may be eccentrically placed.
• Prominent nucleoli with occasional nuclear pseudoinclusions.
• The cytoplasm may contain dark finely dusty brown to black melanin pig-
ment [70].
• Pigment granules may be present in the background or in macrophages in cases
of widely metastatic melanoma with melanuria [71].
180 T. M. Elsheikh et al.
a b
Fig. 8.36 Malignant melanoma. (a) Metastatic cutaneous melanoma displays scattered large atyp-
ical cells with round to oval nuclei and variable amount of cytoplasm. The cytoplasm contains dark
dusty brown melanin pigment. The nuclei are enlarged and have prominent nucleoli. (Voided urine,
TP, high mag). (b) In this primary urethral melanoma, the malignant cells are huge and show
extremely pleomorphic nuclei that are eccentrically placed and have macronucleoli. In contrast to
specimen “a,” the cytoplasm is dense and amelanotic. (Urethrovaginal fluid, TP, high mag.)
Explanatory Notes
Explanatory Note 5 IHC melanocytic markers are needed for confirmation, includ-
ing positive staining with S100, HMB45, Melan A, and SOX10; and generally
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 181
n egative staining with cytokeratins and lymphoid markers (although rare aberrant
expression of cytokeratin staining has been reported in melanoma) [72].
Hematopoietic Malignancies
Background
Lymphoma
Primary lymphoma accounts for 0.2% of all extranodal lymphomas and <1% of all
bladder tumors. Secondary involvement of the urinary tract occurs more often [74].
The majority of primary bladder lymphomas (80%) are low-grade, extranodal mar-
ginal zone B-cell lymphoma of mucosa-associated lymphoid tissue (MALT) origin.
High-grade lymphomas are rarer, with diffuse large B-cell lymphoma (DLBCL)
being most common. Other types, such as anaplastic large-cell lymphoma (ALCL)
and Burkitt’s and mantle cell lymphomas, have been seldom reported. Primary
MALT lymphoma is commonly associated with chronic cystitis and affects adults
older than 60 years of age. It is usually localized at time of diagnosis and often has
excellent prognosis following local resection [73, 74]. On the other hand, primary
high-grade types and secondary lymphomas tend to have a poor prognosis and may
be treated with various therapeutic options, including chemotherapy, radiotherapy,
and surgery, alone or in combination.
The cytologic features of lymphoma in the urine are similar to those seen in
other sites. DLBCL is characterized by large cells with increased N/C ratio, round
to oval nuclei, dense chromatin, and prominent nucleoli (Fig. 8.37). The cells are
most commonly arranged singly, but they may form cohesive clusters, especially in
ALCL. Abundant necrotic debris and numerous apoptotic bodies are common. The
atypical lymphoid cells resemble centroblasts with multiple smaller peripherally
located nucleoli, or immunoblasts with central large nucleoli. Multinucleated
forms and plasmacytoid cells may be seen in ALCL. MALT lymphoma shows
medium-sized cells with round nuclei, dispersed chromatin, and inconspicuous
nucleoli [75]. They may resemble small lymphocytes or have a plasmacytic appear-
ance (Fig. 8.38).
Cytologic diagnosis of lymphoma in urine specimens is very challenging, espe-
cially if there are no clinical symptoms or known clinical history. Also, because of
182 T. M. Elsheikh et al.
a b
Fig. 8.37 Diffuse large B-cell lymphoma (DLBCL). (a) This 59-year-old woman presented with
gross hematuria, urinary frequency and lower abdominal pain. Urine cytology demonstrates a dis-
persed bi-morphic population of small and large lymphoid cells. The larger cells predominate and
show clumped chromatin, prominent single or multiple nucleoli, and variably stripped cytoplasm.
Follow-up biopsy (not shown) revealed a primary DLBCL in the bladder dome. (Voided urine, CS,
medium mag.). (b) This is a case of secondary DLBCL involving the bladder, showing a monomor-
phic population of large cells with round to oval nuclei, vesicular chromatin, and prominent nucle-
oli. Some of the cells have a plasmacytoid appearance. Differential diagnosis includes high-grade
carcinoma and melanoma. (Voided urine, TP, high mag.)
its rare occurrence, lymphoma may not be entertained in the differential diagnosis
of abnormal cytology. High-grade lymphoma is easily identified as abnormal or
malignant but can be misdiagnosed as HGUC because of the high N/C ratio and
hyperchromasia of the nuclei (Fig. 8.37a). The initial clinical impression may be
misleading and suggestive of HGUC [74], since most lymphomas present as a soli-
tary mass. Background tumor diathesis and numerous apoptotic bodies should
raise the possibility of lymphoma. Differential diagnosis of high-grade lymphoma
includes HGUC, SmCC, melanoma, and sarcoma. IHC positivity for lymphoid
markers such as CD20, CD79a, and bcl-2 help support the diagnosis, but ulti-
mately histologic confirmation is needed. MALT lymphoma is more challenging to
diagnose than high-grade lymphoma, as it may be difficult or impossible to distin-
guish it from reactive lymphoid cells (Fig. 8.38). The diagnosis of MALT lym-
phoma should be considered in a lymphoid-rich urine specimen, if the patient has
a previous history of lymphoma, or there is a clinical presentation of a blad-
der mass.
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 183
Plasmacytoma
Plasmacytoma of the urinary bladder is very rare, accounting for <5% of all plasma
cell neoplasms. For a diagnosis of primary plasmacytoma, there must be no evi-
dence of plasmacytoma elsewhere, including a negative bone marrow biopsy [73].
An occasional case report documenting plasmacytoma in urine can be found in the
literature [76]. Cytologic features are similar to plasmacytoma at other sites. The
neoplastic cells are large, singly dispersed, with dense basophilic cytoplasm and
eccentrically placed nuclei. The nuclear chromatin has peripheral clumping in a
clockface distribution and may show prominent nucleoli. Differential diagnosis
includes HGUC, plasmacytoid UC, lymphoma, melanoma, NET, and sarcoma. IHC
or surgical biopsy is needed to establish a definitive diagnosis. Prognosis of primary
plasmacytoma is improved with early detection and combination radiation therapy
and surgery. Chemotherapy is utilized in high-risk patients and radiation-resistant
tumors [73].
Two lesions that are extremely rare to encounter in urine cytology are briefly dis-
cussed in this section. They include paraganglioma and nephrogenic adenoma.
Paraganglioma
a b
Fig. 8.39 Paraganglioma. (a) Urine cytology shows a loose cluster of large epithelioid cells with
abundant delicate cytoplasm, anisonucleosis, round to oval nuclei with fine chromatin, and incon-
spicuous nucleoli. (Bladder washing, SurePath, high mag.). (b) Bladder tumor resection displays
cells arranged in a nested (zellballen)/trabecular pattern, separated by delicate vasculature. The
cells are large and have acidophilic cytoplasm, round to oval nuclei, and prominent nucleoli.
(H&E, high mag.)
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 185
Nephrogenic Adenoma
a b
c d
Fig. 8.40 Nephrogenic adenoma (NA), from a 64-year-old man. (a) Histology shows characteris-
tic tubulocystic and papillary architecture of NA, which involves the urothelial surface and lamina
propria. The tubules are closely packed and associated with background inflammation. A small
uninvolved portion of the urothelium is present at left top corner of photo (*). (H&E, low mag.).
(b) NA shows PAX8 positivity but spares the portion of uninvolved urothelium (*). (c) Urine cytol-
ogy demonstrates rare groups of large cells with abundant dense cytoplasm, round nuclei, and
small nucleoli. There is no significant atypia. (Instrumented urine, TP, high mag.). (d) Lesional
cells of NA have abundant eosinophilic dense cytoplasm with round nuclei and small nucleoli
(right part of photo); these cells have similar cytologic appearance to those observed in corre-
sponding urine cytology. However, it would be very difficult, in isolation, to distinguish between
the lesional cells and adjacent uninvolved urothelial cells (left part of photo). (H&E, high mag.)
Sample Reports
If urine cytology is interpreted as positive for “NUM,” it is implied that the sample
is adequate for evaluation and diagnostic of malignancy. “Suspicious for NUM”
implies that the cytologic findings are highly suspicious but not diagnostic of malig-
nancy. “Atypical cells present” implies that the atypia is of uncertain nature or sig-
nificance, and it would be best to categorize the cell type, if possible, i.e., squamous,
glandular, spindle, lymphoid, etc. Narrative notes or comments should be used to
explain or subclassify the malignant, suspicious, or atypical findings and offer a
potential differential diagnosis. A microscopic description is optional.
Example 1
Malignant
Keratinizing squamous cell carcinoma
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 187
Example 2
Malignant
High-grade carcinoma with squamous features
Note: There is admixture of malignant keratinizing squamous cells and malig-
nant non-keratinizing cells. The latter may represent high-grade urothelial carci-
noma versus non-keratinizing squamous carcinoma component. Differential
diagnosis includes urothelial carcinoma with squamous differentiation, primary
pure squamous carcinoma, and secondary squamous cell carcinoma extending from
the genital tract. Clinical correlation is recommended.
Example 3
Example 4
Example 5
Malignant
Adenocarcinoma
Note: The malignant cells have columnar appearance with elongated hyperchro-
matic nuclei, and necrosis in the background. Major diagnostic considerations
include primary adenocarcinoma, urothelial carcinoma with divergent glandular
differentiation, and metastatic adenocarcinoma from other sites such as colorectum.
Clinical correlation is recommended.
188 T. M. Elsheikh et al.
Example 6
Malignant
Adenocarcinoma with signet ring cell features
Note: The malignant cells show signet ring cell features. Signet ring cells may be
associated with several carcinomas, including plasmacytoid urothelial carcinoma,
primary adenocarcinoma, and metastatic adenocarcinoma from other sites such as
gastrointestinal tract and breast. Clinical correlation is recommended.
Example 7
Malignant
Adenocarcinoma
Note: There are cohesive clusters of highly atypical glandular cells, consistent
with adenocarcinoma. Although high-grade urothelial carcinoma cells are not seen,
urothelial carcinoma with glandular differentiation cannot be excluded. Differential
diagnosis also includes primary and secondary adenocarcinoma.
Example 8
Example 9
Malignant
Small-cell (high-grade neuroendocrine) carcinoma
Note: The specimen contains many clusters and single highly atypical small cells
with prominent nuclear molding/spindling and apoptosis. Immunohistochemistry
(IHC) performed on the cell block demonstrated positive staining with synaptophy-
sin and chromogranin and negative staining with CD45, p63, and SOX10. The com-
bined cytologic and IHC findings are consistent with small-cell carcinoma.
Example 10
Malignant
Adenocarcinoma, consistent with prostate primary
Note: Previous history of high-grade prostatic adenocarcinoma is noted. Rare
groups of atypical glandular cells with relatively uniform enlarged nuclei and prom-
inent nucleoli are present. Immunohistochemistry performed on cytologic prepara-
tions showed positive staining with NKX3.1 and PSA and negative staining with
GATA3 and p63, consistent with prostatic adenocarcinoma.
8 Non-Urothelial Malignancies and Other Miscellaneous Lesions 189
Example 11
References
1. Chalasani V, Chin JL, Izawa JI. Histologic variants of urothelial bladder cancer and nonurothe-
lial histology in bladder cancer. Can Urol Assoc J. 2009;3:S193–8.
2. Dahm P, Gschwend JE. Malignant non-urothelial neoplasms of the urinary bladder: a review.
Eur Urol. 2003;44:672–81.
3. Kunze E. Histogenesis of nonurothelial carcinomas in the human and rat urinary bladder. Exp
Toxicol Pathol. 1998;50:341–55.
4. Barkan GA, Laucirica R, Auger M, et al. Performance characteristics of urinary tract cytology:
observations from the College of American Pathologists Interlaboratory Comparison Program
in Nongynecologic Cytopathology. Arch Pathol Lab Med. 2015;139:1009–13.
5. Abol-Enein H, Kava BR, Carmack AJ. Nonurothelial cancer of the bladder. Urology.
2007;69:93–104.
6. Nishiyama H, Habuchi T, Watanabe J, et al. Clinical outcome of a large-scale multi-institutional
retrospective study for locally advanced bladder cancer: a survey including 1131 patients
treated during 1990–2000 in Japan. Eur Urol. 2004;45:176–81.
7. Deuker M, Martin T, Stolzenbach F, et al. Bladder cancer: a comparison between non-urothelial
variant histology and urothelial carcinoma across all stages and treatment modalities. Clin
Genitourin Cancer. 2021;19:60–8.
8. Quinn DI. Non-urothelial bladder cancer. Web site: https://www.uptodate.com/contents/non-
urothelial-bladder-cancer. Published 2019. Updated 02/13/2019. Last Accessed 11/01/2020.
9. Xiao GQ, Chow J, Unger PD. Metastatic tumors to the urinary bladder: clinicopathologic
study of 11 cases. Int J Surg Pathol. 2012;20:342–8.
10. Hoda RS. Non-gynecologic cytology on liquid-based preparations: a morphologic review of
facts and artifacts. Diagn Cytopathol. 2007;35:621–34.
11. Laucirica R, Bentz JS, Souers RJ, et al. Do liquid-based preparations of urinary cytology per-
form differently than classically prepared cases? Observations from the College of American
Pathologists Interlaboratory Comparison Program in Nongynecologic Cytology. Arch Pathol
Lab Med. 2010;134:19–22.
12. Luo Y, She DL, Xiong H, Yang L, Fu SJ. Diagnostic value of liquid-based cytology
in urothelial carcinoma diagnosis: a systematic review and meta-analysis. PLoS One.
2015;10:e0134940.
13. Straccia P, Bizzarro T, Fadda G, Pierconti F. Comparison between Cytospin and liquid-based
cytology in urine specimens classified according to the Paris System for Reporting Urinary
Cytology. Cancer Cytopathol. 2016;124:519–23.
14. Elsheikh TM, Kirkpatrick JL, Wu HH. Comparison of ThinPrep and Cytospin preparations in
the evaluation of exfoliative cytology specimens. Cancer. 2006;108:144–9.
15. Koss LG, Hoda RS. Koss’s cytology of the urinary tract with histopathologic correlations.
New York: Springer Press; 2012.
16. Shen SS, Al-Ahmadie HH, Mahfouz SM. Squamous cell neoplasms. In: Moch H, Humphrey
PA, Ulbright TM, Reuter VE, editors. WHO classification of tumors of the urinary system
and male genital organs. Lyon, France: International Agency for Research on Cancer; 2016.
p. 108–10.
190 T. M. Elsheikh et al.
41. Ciesla MC, Guidos BJ, Selvaggi SM. Cytomorphology of small-cell (neuroendocrine) car-
cinoma on ThinPrep cytology as compared to conventional smears. Diagn Cytopathol.
2001;24:46–52.
42. Alijo Serrano F, Sanchez-Mora N, Angel Arranz J, Hernandez C, Alvarez-Fernandez E. Large
cell and small cell neuroendocrine bladder carcinoma: immunohistochemical and outcome
study in a single institution. Am J Clin Pathol. 2007;128:733–9.
43. Chen YB, Epstein JI. Primary carcinoid tumors of the urinary bladder and prostatic urethra: a
clinicopathologic study of 6 cases. Am J Surg Pathol. 2011;35:442–6.
44. Helpap B. Morphology and therapeutic strategies for neuroendocrine tumors of the genitouri-
nary tract. Cancer. 2002;95:1415–20.
45. van Hoeven KH, Artymyshyn RL. Cytology of small cell carcinoma of the urinary bladder.
Diagn Cytopathol. 1996;14:292–7.
46. Ainechi S, Pambuccian SE, Wojcik EM, Barkan GA. Cytomorphologic features and differ-
ential diagnosis of neoplasms with small cell features in liquid-based urinary tract cytologic
specimens. J Am Soc Cytopathol. 2015;4:295–306.
47. Acs G, Gupta PK, Baloch ZW. Cytomorphology of high-grade neuroendocrine carcinoma of
the urinary tract. Diagn Cytopathol. 2000;23:92–6.
48. Bates AW, Baithun SI. Secondary neoplasms of the bladder are histological mimics of non-
transitional cell primary tumours: clinicopathological and histological features of 282 cases.
Histopathology. 2000;36:32–40.
49. Grignon DJ, Epstein J, Oliva E, Ro JY, Shen SS. Metastatic tumors In: Moch H, Humphrey PA,
Ulbright TM, Reuter VE, editor. WHO classification of tumors of the urinary system and male
genital organs. Lyon: International agency for research on cancer; 2016. p. 130.
50. Velcheti V, Govindan R. Metastatic cancer involving bladder: a review. Can J Urol.
2007;14:3443–8.
51. Murphy WM, Grignon DJ, Perlman EJ. Tumors of the urinary bladder. Tumors of the kidney,
bladder, and related urinary structures: American registry of pathology; 2004. p. 241–361.
52. Rao Q, Williamson SR, Lopez-Beltran A, et al. Distinguishing primary adenocarcinoma of the
urinary bladder from secondary involvement by colorectal adenocarcinoma: extended immu-
nohistochemical profiles emphasizing novel markers. Mod Pathol. 2013;26:725–32.
53. Renshaw AA. Urine and bladder washings. In: Cibas E, Ducatman B, editors. Cytology: diag-
nostic principles and clinical correlates. Philadelphia: Elsevier; 2021.
54. Sathiyamoorthy S, Ali SZ. Urinary cytology of prostatic duct adenocarcinoma - a clinico-
pathologic analysis. Acta Cytol. 2013;57:184–8.
55. Feldman A, Borak S, Rais-Bahrami S, Gordetsky J. Secondary malignancies of the bladder:
avoiding the diagnostic pitfall. Int J Surg Pathol. 2018;26:120–5.
56. Blochin EB, Park KJ, Tickoo SK, Reuter VE, Al-Ahmadie H. Urothelial carcinoma with prom-
inent squamous differentiation in the setting of neurogenic bladder: role of human papilloma-
virus infection. Mod Pathol. 2012;25:1534–42.
57. Gordetsky J, Spieker AJ, Pena M, et al. Squamous cell carcinoma of the bladder is not associ-
ated with high-risk HPV. Urology. 2020;144:158–63.
58. Zhao P. Urine cytology of metastatic breast lobular carcinoma. Diagn Cytopathol.
2019;47:738–9.
59. Liu H, Shi J, Wilkerson ML, Lin F. Immunohistochemical evaluation of GATA3 expression in
tumors and normal tissues: a useful immunomarker for breast and urothelial carcinomas. Am
J Clin Pathol. 2012;138:57–64.
60. Vigliar E, Marino G, Matano E, Imbimbo C, de Rossella C, Insabato L. Signet-ring-cell car-
cinoma of stomach metastatic to the bladder: a case report with cytological and histological
correlation and literature review. Int J Surg Pathol. 2013;21:72–5.
61. Sanguedolce F, Loizzi D, Sollitto F, et al. Bladder metastases from lung cancer: clinical and
pathological implications: a systematic review. Oncology. 2017;92:125–34.
62. Karle W, Barazani Y, Tareen B. A rare case of metastatic lung cancer to the bladder. Can Urol
Assoc J. 2012;6:E147–9.
63. Layfield LJ, Elsheikh TM, Fili A, Nayar R, Shidham V, Papanicolaou Society of C. Review of
the state of the art and recommendations of the Papanicolaou Society of Cytopathology for uri-
192 T. M. Elsheikh et al.
nary cytology procedures and reporting: the Papanicolaou Society of Cytopathology Practice
Guidelines Task Force. Diagn Cytopathol. 2004;30:24–30.
64. Cheville J, Folpe A. Mesenchymal tumors In: Moch H, Humphrey PA, Ulbright TM, Reuter
VE, editor. WHO Classification of tumors of the urinary system and male genital organs. Lyon:
International agency for research on cancer; 2016. p. 122–127.
65. Iwa N, Ito S, Takegaki Y, et al. Cytologic features of sarcomatoid carcinoma of the urinary
bladder: a case report. Diagn Cytopathol. 2013;41:536–41.
66. Hemachandran M, Nada R, Rajwanshi A. Leiomyosarcoma of the urinary bladder: a diagnos-
tic challenge in urine cytology. Diagn Cytopathol. 2004;31:281–2.
67. Mitra S, Kaur G, Kakkar N, Singh P, Dey P. Sarcoma in urine cytology; an extremely rare
entity: a report of two cases. J Cytol. 2017;34:171–3.
68. Westfall DE, Folpe AL, Paner GP, et al. Utility of a comprehensive immunohistochemical
panel in the differential diagnosis of spindle cell lesions of the urinary bladder. Am J Surg
Pathol. 2009;33:99–105.
69. Algaba F, Eble J. Melanocytic tumors. In: Moch H, Humphrey PA, Ulbright TM, Reuter VE,
editor. WHO classification of tumors of the urinary system and male genital organs. Lyon:
International Agency for Research on Cancer; 2016. p. 120–21.
70. Khalbuss WE, Hossain M, Elhosseiny A. Primary malignant melanoma of the urinary bladder
diagnosed by urine cytology: a case report. Acta Cytol. 2001;45:631–5.
71. McIntire PJ, Kilic I, Wojcik EM, Barkan GA, Pambuccian SE. The color of urine: then and
now-a comprehensive review of the literature with emphasis on intracytoplasmic pigments
encountered in urinary cytology. J Am Soc Cytopathol. 2020;9:9–19.
72. Chen N, Gong J, Chen X, et al. Cytokeratin expression in malignant melanoma: potential
application of in-situ hybridization analysis of mRNA. Melanoma Res. 2009;19:87–93.
73. Osunkoya AO, Humphrey PA. Hematopoietic and lymphoid tumors. In: Moch H, Humphrey
PA, Ulbright TM, Reuter VE, editor. WHO classification of tumors of the urinary system
and male genital organs Lyon, Lyon: International Agency for Research on Cancer; 2016. p.
128–129.
74. Bhutani DN, Goel DV, Kajal DP, Pawar DD, Sharma DP, Sen DR. Primary extra nodal Non-
Hodgkin's lymphoma of urinary bladder presenting as a bladder tumor: a case report. Ann Med
Surg (Lond). 2020;56:68–71.
75. Quinn AM, Flanigan R, Sienko A, Wojcik EM. Cytologic features of recurrent lymphoma
involving the urinary bladder. Diagn Cytopathol. 2004;31:185–8.
76. Farinola MA, Lawler LP, Rosenthal D. Plasmacytoma with involvement of the urinary bladder.
Report of a case diagnosed by urine cytology. Acta Cytol. 2003;47:787–91.
77. Park S, Kang SY, Kwon GY, et al. Clinicopathologic characteristics and mutational status of
succinate dehydrogenase genes in paraganglioma of the urinary bladder: a multi-institutional
Korean study. Arch Pathol Lab Med. 2017;141:671–7.
78. Lee J, Park S, Cho MS, Sung SH, Kim KH. Urine cytology findings of primary paraganglioma
of the urinary bladder: case report. Diagn Cytopathol. 2017;45:350–3.
79. Pina-Oviedo S, Shen SS, Truong LD, Ayala AG, Ro JY. Flat pattern of nephrogenic adenoma:
previously unrecognized pattern unveiled using PAX2 and PAX8 immunohistochemistry. Mod
Pathol. 2013;26:792–8.
80. Rutgers JL, Young RH. Nephrogenic adenoma of the urinary bladder: a comparison of its
cytologic and histopathologic features in ten cases. Diagn Cytopathol. 1988;4:210–6.
81. McCroskey Z, Pambuccian SE, Wojcik EM, Barkan GA. Can we identify nephrogenic ade-
noma in urine cytology specimens? A study evaluating previously described cytomorphologic
features in correlation with PAX8 immunohistochemical staining results. Am J Clin Pathol.
2016;145:373–8.
82. Koss LG. Diagnostic cytology of the urinary tract with histopathologic and clinical correla-
tions. Philadelphia: Lippincott-Raven; 1996.
Ancillary Studies in Urinary Cytology
9
Lukas Bubendorf, Nancy P. Caraway, Andrew H. Fischer,
Ruth L. Katz, Fernando Schmitt, Margareta Strojan Fležar,
Theodorus H. Van der Kwast, and Philippe Vielh
• Updated literature review on ancillary testing including new markers and assays.
• Recent data on ancillary testing in the era of TPS.
• New section on the promise of next-generation sequencing in urinary cytology.
Background
Urine cytology and cystoscopy with biopsy still remain the only means of definitive
diagnosis of urothelial carcinoma (UC). Though cytology and cystoscopy are com-
plementary, together they fail to detect a significant number of patients with UC. The
difficulties and challenges of urinary cytology have been driving the search for bio-
markers and development of commercial diagnostic assays to improve its sensitivity
for detecting urothelial carcinoma. In the past two decades, many different assays,
or ancillary tests, have been developed to overcome the limitations of urinary cytol-
ogy and improve the timely detection of UC. In parallel, the number of review
articles on urinary markers for detection of bladder cancer (UC) has steeply risen.
Despite the countless markers that have been proposed, only a few have reached the
stage of US Food and Drug Administration (FDA)-approved assays for diagnostic
application. The ancillary tests that have recognized utility include UroVysion®
Fluorescence in situ Hybridization (U-FISH; Abbott Laboratories, Abbott Park, IL,
USA); BTA™ (Polymedco Inc., Cortlandt Manor, NY, USA); and NMP22™
(Bladder Check) test (Alere Inc., Waltham, MA, USA). Several other tests for
tumor-associated antigens and PCR-based tests for mutations have been proposed,
and within the past several years, next-generation sequencing (NGS) has begun to
be applied to detect mutations or epigenetic changes of UC.
Evaluation of individual markers has been hindered by the complexity of the disease
diagnosis and behavior, variable clinical endpoints, and the different types of treatment
dependent upon tumor grade and stage. Importantly, many studies have been done with-
out reference to cytology results classified according to TPS, making interpretation dif-
ficult because cytological findings and results of ancillary tests often cannot be linked
back to particular stages or grades of UC. These factors make it at least challenging if
not impossible to define an optimal role for ancillary studies through review of the litera-
ture. Concerns about cost-effectiveness of ancillary testing are another source of ongo-
ing discussions. For these reasons, no single ancillary test is clearly being recommended
as part of the routine evaluation in the guidelines of the American Urological Association
and the European Association of Urology at this time [1, 2].
There are basically two types of ancillary tests: those that are based on cytologi-
cal preparations (cell-based tests) and those that rely on a non-morphologic analysis
of urinary fluids (liquid-based tests). Liquid-based tests, such as NMP22, are meant
to be applied in the office of the urologist to identify patients with high risk of pri-
mary urothelial carcinoma or recurrence. Cell-based tests dovetail with the activi-
ties of the Cytopathology Laboratory and generally benefit from input from the
cytologist about whether a particular sample should be tested. Cell-based tests also
offer the potential to provide feedback to cytologists on a per cell basis about the
significance of atypical cytologic findings, potentially leading to future improve-
ment in criteria and the performance of cytologic diagnosis.
Many of the controversies on the value of ancillary testing in urinary cytology are
caused by a lack of clear indications for testing. Major unresolved questions are the
threshold risk of UC to invoke ancillary testing, whether ancillary studies should be
routinely performed on patients with an atypical cytology, whether cytologists or urol-
ogists are in the best position to order ancillary cell-based testing, and whether ancil-
lary tests have value as a quality improvement measure for the cytology laboratory.
TPS provides an opportunity to define and narrow down the diagnostic catego-
ries in which ancillary testing of urinary specimens is most rewarding and to avoid
unnecessary testing in areas with minimal or no added value. TPS has laid the
ground for prospective studies in search of cost-effective combinations of urine
cytology with ancillary testing to improve diagnosis and clinical outcome of
UC. According to a survey by the College of American Pathologists (CAP) covering
the time period from 2014 to 2016, ancillary testing of urinary tract cytology was
performed by 128 of 596 laboratories (21.4%), and UroVysion® fluorescence in
situ hybridization (U-FISH; Abbott Laboratories, Abbott Park, IL, USA) was the
most commonly used ancillary test (83.6%) followed by FISH tests from other man-
ufacturers (14.8%) [3]. While we will summarize several proposed diagnostic mark-
ers, we will therefore place a special emphasis on U-FISH that has been approved
by the FDA for diagnosis of UC in patients with hematuria and/or monitoring for
tumor recurrence in patients previously diagnosed with bladder cancer. To date,
U-FISH is considered the most promising ancillary test in urinary cytology and
therefore described in a separate section of this chapter [1, 4, 5]. This test is based
on the detection of numerical and structural chromosomal aberrations, which are a
hallmark of cancer, but only rarely seen in benign cells. ImmunoCyt/UCyt+®
(uCyt; DiagnoCure Inc., Quebec, Canada), another cell-based test, relies on the
9 Ancillary Studies in Urinary Cytology 195
The commercially available multitarget multicolor U-FISH assay contains four sin-
gle-stranded DNA probes. Three probes are chromosome enumeration probes (CEP)
targeting pericentromeric regions of chromosomes 3, 7, and 17. Another probe is a
locus-specific identifier (LSI) probe targeting 9p21 locus on gene p16. All probes are
directly labeled with fluorescent dyes, specifically CEP 3 SpectrumRed, CEP 7
SpectrumGreen, CEP 17 SpectrumAqua, and LSI 9p21 SpectrumGold. There is a
similar FISH test that is labeled as an in vitro diagnostic (IVD) product (ZytoLight®
SPEC CDKN2A/CEN 3/7/17 Quadruple Color Probe; ZytoVision, Bremerhaven,
Germany). The same technical principles apply as for U-FISH. However, there are
no published data on its performance in urinary tract cytology.
Specimen Preparation
The U-FISH test was originally intended for use on voided urine specimens.
However, later studies have proved its use on other fluid specimens from the urinary
tract, e.g., bladder washings and upper urinary tract washings. Fresh urinary speci-
mens without pre-fixation must be processed to cell preparations on glass slides as
soon as possible after collection (within 3 hours). If necessary, fluid specimens may
be prefixed with 2% polyethylene glycol in 50% ethanol (Carbowax), PreservCyt®
(Hologic, Marlborough, MA), or other preservatives (e.g., an equal volume of 50%
ethanol) and preferably processed within 72 hours. Different techniques of slide
preparation can be used for U-FISH test. See Chap. 10 for appropriate specimen
preparation methods. There should be 100 to 200 well-preserved cells in the target
area for hybridization, spread in a moderately cellular monolayer, and fixed by
alcohol-based fixatives.
Unstained slide preparations are often used for the assay. However, U-FISH is
also applicable to original slides that have been stained according to the Papanicolaou
method to enable cytomorphological evaluation before the FISH assay. A disadvan-
tage of performing FISH on a slide that has been pre-stained is that the Eosin Azure
(EA) dye can cause fluorescence: if the stain is not completely washed away during
FISH, high background can be seen. Decreasing the EA concentration for samples
destined for FISH can be helpful to keep background fluorescence low. Areas with
abnormal urothelial cells on the Papanicolaou preparation can be relocated after
FISH, as described in more detail in the section on automation.
The basic principles of the U-FISH assay are illustrated in Fig. 9.1. In addition,
the technical procedures including slide pretreatment and hybridization are
described in the Appendix.
196 L. Bubendorf et al.
Fig. 9.1 Basic principles of U-FISH assay. The assay contains directly fluorescent dye-labeled
single-stranded DNA probes. Pre-analytical procedures include slide preparation (including pre-
treatment of pre-stained slides). UroVysion assay starts with probe preparation and specimen –
target DNA denaturation followed by DNA hybridization of probes to target DNA sequences.
Analysis of FISH signals is performed under epi-fluorescent microscope. CEP centromere enu-
meration probe, LSI locus-specific identifier probe, chr chromosome, loc locus
a b c
Fig. 9.2 (a–c) Schematic illustration of U-FISH findings. (a) Nuclei of a FISH-negative benign
cell with 2 signals for the chromosomes 3 (red), 7 (green), and 17 (aqua) and for 9p21 (gold). (b)
U-FISH-positive cell with increased and unbalanced number of chromosomes 3 (red) and 17
(aqua) and heterozygous loss of 9p21 (gold). (c) Tetraploid FISH pattern with four signals of each
FISH probe. This pattern is not specific for urothelial carcinoma but commonly seen in reactive
urothelial cells
reactive cells (e.g., umbrella cells) and may give rise to false-positive FISH results.
Along this line, it is specifically mentioned in the package insert that results at or
near the cut-off point should be interpreted with caution. A schematic illustration of
typical U-FISH findings is shown in Fig. 9.2.
There are several conditions that may prevent accurate FISH examination,
including the presence of lubricant jelly, corpora amylacea, degenerated cells, crys-
tals, sperm, and bacteria.
In the case of performing FISH on conduit urines, interference by inflammation
and cell debris may occur. Squamous contamination, especially in urines from
female subjects, may obscure urothelial cells. If there are only a few atypical cells
present, then a targeted FISH approach, using the same Papanicolaou stained prepa-
ration as the cytology slide, is advisable.
Because manual FISH analysis has a relatively slow throughput and the fluorescent
signals fade over time [10], automated imaging systems are increasingly being uti-
lized, especially in institutions with high volumes of FISH tests. Reviewing repre-
sentative digitized images instead of a FISH slide at the microscope can lead to
dramatic savings in pathologist time [11]. Imaging systems may also be imple-
mented for improved productivity, quality control, archiving of images, and
increased accuracy. We estimate that almost 50% of the laboratories performing
U-FISH analysis in the United States use automated systems. In some labs there
was a combination of both manual and automated use. In non-US countries, auto-
mated imaging systems for U-FISH analysis are less commonly used.
198 L. Bubendorf et al.
Studies have shown a high concordance (>98%) between the manual method and
the automated method of reviewing cells. In several cases deemed negative by the
manual method, machine-assisted interpretation showed a positive result for abnor-
mal cells which was subsequently verified within a short interval by cystoscopy
[11]. In addition, the ability to enlarge cells with questionable signals and enhance
weak signals is advantageous in reducing false positives or false negatives.
Disadvantages of automation may include a higher number of unsatisfactory
specimens due to scant cellularity or clumping, in which case the manual system of
review is employed. Factors to consider before purchasing an imaging system
include financial cost, space requirements, information technology system needs or
modifications, validation of system, training programs for personnel, continuing
education, maintenance of system, workflow and patterns of reimbursement [11].
Fig. 9.3 Automated scanning instrument showing workstation with concurrent acquisition of
DAPI images and FISH images illustrated on screen
9 Ancillary Studies in Urinary Cytology 199
Fig. 9.4 Urine from a patient with a history of high-grade urothelial carcinoma. Histogram of
U-FISH abnormal cells showing 14 out of 25 cells with polysomy for at least 2 chromosomes per
cell. This is an abnormal result and consistent with high-grade urothelial carcinoma. Although a
total of 99 cells were scored, reporting is based on 25 of the most morphologically abnormal
cells scored
System (Ikonsys, Inc., New Haven, CT). Technical details of these imaging systems
are described in the Appendix. Automated systems allow for interactive rejection or
acceptance of cells, so that overlapping cells or degenerated cells can be rejected
[12]. The imaged cells are then formatted in a gallery on a screen for review. Normal
and abnormal cells can be segregated and then can be reclassified as needed [11].
Extraneous cells such as neutrophils and lymphocytes are automatically excluded
from analysis.
Target FISH
Fig. 9.5 (a–d) Examples of U-FISH and corresponding urinary cytology (Papanicolaou stain).
Chromosomes 3 in red, 7 in green, 17 in aqua, and 9p21 in gold. (a) Benign bi- or multi-nucleate
umbrella cell with pronounced reactive changes in a patient with irritative bladder (bladder wash-
ing). Highly increased number of all FISH signals due to repeated doubling of the genome (i.e.,
endoreplication), but no unbalanced copy number changes and no 9p21 deletion. No evidence of
urothelial carcinoma after 8-year follow-up. (b) Sheets of mildly atypical urothelial cells in a
washing of the renal pelvis. Normal FISH result (two copies per chromosome), consistent with
reactive changes. No histological follow-up. (c) Atypical urothelial cells suspicious of high-grade
urothelial carcinoma (SHGUC) in a washing of the renal pelvis. Positive U-FISH result with
increased copy numbers of all three chromosomes and a homozygous deletion of 9p21. Note the
normal cell nucleus with retained 9p21 signals as internal normal control (right lower corner).
Nephrectomy revealed noninvasive high-grade papillary urothelial carcinoma of the renal pelvis.
(d) U-FISH-positive atypical urothelial cells (SHGUC) with increased copy number of all five
chromosomes but no 9p21 deletion. Note the normal cell nucleus with normal copy numbers (right
lower corner). Histology revealed invasive high-grade papillary urothelial carcinoma (pT1) with
focal CIS
9 Ancillary Studies in Urinary Cytology 201
Fig. 9.5 (continued)
enables the reviewer to associate aneuploidy with cytologic features on a per cell
basis. This also provides an opportunity to the cytopathologists to improve their
skills in differentiating morphological features of benign/reactive and neoplastic
cells using the chromosomal status as a reference.
Reporting of Results
One or several pathologists can sign out cases by using remote computers situated
away from the main automated scanning microscope equipped with appropriate
software [11, 16]. In manual screening a minimum of 25 abnormal cells is evalu-
ated. In contrast, the automated systems require a minimum of 100 urothelial cells
without any chromosomal aberrations to be classified as normal [11]. Using either
method, only 25 of the most abnormal cells are reported.
202 L. Bubendorf et al.
U-FISH testing of voided urine has been approved by the FDA as an aid for initial
diagnosis of bladder cancer in patients with hematuria and/or subsequent monitor-
ing for tumor recurrence in patients previously diagnosed with bladder cancer. In
routine practice, there are several additional situations in which U-FISH is fre-
quently used [14]. In a meta-analysis, the pooled sensitivity and specificity of FISH
were 72% and 83%, respectively, as compared with 42% and 96%, respectively, for
cytology [17]. However, there is a range of U-FISH results in the literature [9, 18–
21]. Like urine cytology, performance of FISH is better in patients with high-grade
UC. Analyzing these published results is challenging due to great variability of
critical parameters including type of material (voided urine versus bladder wash-
ing), clinical situation (history or no history of bladder cancer), concurrent cystos-
copy findings, length and type of follow-up, proportion of low-grade and high-grade
tumors, local experience in FISH analysis, and the definition of a positive result,
among others. This emphasizes the need for well-controlled and independently
funded studies to clarify the performance of FISH in clearly defined situations and
to analyze the cost-benefit ratio for both the laboratory and patient management.
Atypical urinary cytology has emerged as the most rewarding application of U-FISH
analysis [7, 15, 21–30]. U-FISH is currently regarded as the most useful marker in
the setting of a negative cystoscopy and atypical cytology, according to the
International Consultations on Urological Diseases, the European Society of
Urology, and the Guidelines of the American Urological Association [1, 4]. In an
early pivotal study, U-FISH in 120 urine specimens revealed a sensitivity of 100%,
89%, and 60% in patients with suspicious, atypical, and negative cytology, respec-
tively, while the overall specificity was 97% [31]. This is in line with another study
showing that the sensitivity of U-FISH to detect UC in patients with atypical cytol-
ogy and equivocal or negative cystoscopy was 100% with a specificity ranging from
60% to 100% depending on whether or not there was a history of UC or a lesion at
9 Ancillary Studies in Urinary Cytology 203
cystoscopy [28]. The low positive predictive value in patients with a history of UC
and a negative cystoscopy in this study was explained by early (“anticipatory”)
detection of neoplastic cells that preceded the appearance of established cancer. In
fact, other studies have shown that such anticipatory positive U-FISH results predict
recurrence in patients under surveillance with atypical or suspicious cytology and a
negative cystoscopy [23, 29, 32]. Previous data on U-FISH results in conjunction
with atypical cytology samples are hampered by the inconsistent definitions and
variable prevalence of “atypia” in the pre-TPS era. In a recent study with prospec-
tively determined TPS categories, 40/90 (44%) of AUC cases were U-FISH posi-
tive, and 25/28 (89%) of these U-FISH-positive patients with follow-up were
diagnosed with UC [15].
Despite these promising data, performing U-FISH on a urine sample with atypi-
cal cytology at the time of a negative surveillance cystoscopy is not yet generally
recommended since the U-FISH result is still unlikely to change the currently estab-
lished surveillance strategies in non-muscle-invasive UC.
U-FISH is particularly helpful in the notoriously difficult construct of atypical
cytology after intravesical bacillus Calmette-Guerin (BCG) treatment of high-grade
non-muscle-invasive UC [1, 27, 33–35]. Patients with a positive post-BCG cytology
(HGUC) or FISH result have a substantially higher risk of recurrence when com-
pared to the patients with negative results (NHGUC) [27, 36, 37]. In a recent meta-
analysis, the preferred timing of post-BCG U-FISH analysis was proposed at
3 months after initial transurethral resection of non-muscle-invasive bladder cancer
[37]. Washing cytologies of the upper urinary tract (UUT) can also be challenging
due to instrumentation-related changes that might lead to false-positive cytology
results. On the other hand, accurate diagnosis of UUT UC is critical to avoid delay
of diagnosis. Published reports suggest that U-FISH also appears as a helpful tool to
increase the sensitivity for the detection of UC of the UUT [14, 38–41]. In our expe-
rience, FISH is very helpful in atypical UUT washing cytology. However, one needs
to be familiar with tetraploid U-FISH patterns that may occur in reactive conditions,
e.g., renal calculi, in order to avoid false-positive FISH diagnoses (see Pitfalls of
UroVysion® FISH Analysis). Examples of U-FISH in atypical urinary cytology are
shown in Fig. 9.5.
There have been concerns about the cost-effectiveness of the relatively expensive
U-FISH testing, mainly because of the low positive predictive value. If applied to
unselected patients with hematuria (in whom the prevalence of UC is about 13%)
[42] and recognizing the limited clinical value of enhanced detection of low-grade
UC, FISH results cost the patient a high price [14]. However, recent data suggest
cost-effectiveness of U-FISH in patients with atypical cytology and equivocal or
negative cystoscopy by avoiding unnecessary biopsies in patients with a negative
FISH result [22]. Decreasing anxiety for the patient and clinician is also beneficial.
Despite the evident utility of FISH in atypical urinary cytology, there is a possibility
of false-positive results. Besides “anticipatory positive” FISH results, low specific-
ity and low PPV in some studies might also be due to an increased number of tetra-
ploid or dividing cells under reactive conditions. Although tetraploid cells are not
specifically mentioned in the scoring guidelines of the manufacturer, it has now
been recognized that rare cells with a tetraploid signal pattern (i.e., four signals of
each probe) should not be taken as an unequivocally FISH-positive result but inter-
preted with caution [8, 43–47]. In contrast, unbalanced numerical changes of one or
more chromosomes (e.g., 2, 3, 5) or loss of 9p21 is virtually specific for neoplasia
in bladder cytology. An exception is that pelvic irradiation (e.g., of the prostate or
the uterus) often results in permanent chromosomal aberrations carrying a risk of a
false-positive diagnosis by U-FISH. This is particularly the case for polysomies,
whereas homozygous or heterozygous deletion of 9p21 remains highly specific for
neoplasia even after irradiation. This emphasizes the need of appropriate clinical
information and consideration of the typical post-irradiation cytological changes.
Decoy cells, i.e., polyoma-infected urothelial cells, can easily mislead to a false-
positive cytological diagnosis of UC if one is not familiar with the morphological
spectrum of decoy cells. Although there have been reports on rare FISH-positive
decoy cells, most data suggest that polyomavirus infection does not interfere with
U-FISH results [48]. While rare, tumors other than UC involving the urinary tract
such as adenocarcinoma, small cell carcinoma, and renal cell carcinoma have been
found to have positive U-FISH results in urinary specimens [32].
9 Ancillary Studies in Urinary Cytology 205
Liquid- or non-cell-based tests can be applied to voided urine samples either with-
out the need for further preparation or on sediments of centrifuged urine samples.
The two most commonly applied liquid-based urine tests on pure voided urine sam-
ples are the BTA™ (Bladder Tumor Antigen) (Polymedco Inc., Cortlandt Manor,
NY, USA) and the NMP22™ (Nuclear Matrix Protein) BladderChek test (Alere
Inc., Waltham, MA, USA), both approved by the FDA, both for detection of UC in
symptomatic patients and for monitoring of patients with a history of UC. For both
tests, a qualitative point-of-care test (BTA stat™ and NMP22-BladderChek™) and
a quantitative version (NMP22™ and BTA TRAK ™) are available. The latter tests
require a dedicated laboratory with specialized personnel. The advantage of a point-
of-care test is that it gives an immediate result just prior to cystoscopy. A positive
urine test may lead to a diagnostic bias, i.e., with the knowledge of a positive urine
test result, the urologist may be more likely to detect a bladder neoplasm at cystos-
copy [49]. These tests have been developed for both voided urine samples (not first-
morning urine) and catheter-collected urines but not for barbotage fluids.
BTA Test
The bladder tumor antigen (BTA) is a complement factor H-related protein secreted
by malignant cells, which was shown to be elevated in urine samples of patients
with bladder cancer [50]. There are two approved versions of this test for bladder
cancer follow-up in concurrent use with cystoscopy: the quantitative ELISA test
BTA TRAK and the qualitative point-of-care test BTA stat (Polymedco Inc.,
Cortlandt Manor, New York, USA) [51]. For the BTA™ test, the urine should be
collected without preservatives or fixatives in a clean urine cup and labeled
appropriately.
In different reviews and meta-analysis, the BTA stat has an overall sensitivity
and specificity of 64% (range 58–69%) and 77% (range 73–81%), respectively [51].
The specificity of the BTA™ tests may be overrated by some studies by their exclu-
sion of patients with urinary bladder infection and renal or bladder stones. The BTA
TRAK™ test has an overall sensitivity of 65% (range 54–75%) and specificity of
74% (range 64–82%) respectively [51].
206 L. Bubendorf et al.
Nuclear matrix proteins (NMPs) are a family of nuclear structural proteins [55].
Several of them are overexpressed in UC, like NMP22, and released into the urine
upon apoptosis of tumor cells. However, elevated NMP22 levels may correspond to
the cellularity and cell turnover in the bladder rather than representing a tumor-
specific protein [56]. Both quantitative ELISA NMP22 Test Kit for bladder cancer
and qualitative NMP22 BladderChek point-of-care (POC) tests (Matritech, Newton,
Massachusetts) have been approved by the US FDA, the first one in 1996 and the
second one in 2002 [57]. They are used in the context of diagnosis and monitoring
for bladder cancer recurrence. The test should not be performed on voided urine
samples obtained within 5 days after instrumentation of the urinary bladder as this
may evoke wound repair potentially resulting in a false-positive result. Variation has
been reported among studies with some very low sensitivity values [54]. Moreover,
false-positive results of the NMP22 tests are more common in patients with benign
bladder conditions such as infection, stones, inflammation, and hematuria that need
to be distinguished clinically from bladder cancer [54].
A wide variety of other potential ancillary tests have emerged based on protein
detection, microRNAs, gene expression profiling, epigenetic changes, and PCR-
based detection. A few of these tests are described in this section, and Table 9.1
summarizes their performance [58–62].
9 Ancillary Studies in Urinary Cytology 207
Table 9.1 Performance characteristics of selected ancillary tests in comparison with cytology for
bladder cancer monitoring. Methodological differences in the studies make it difficult to compare
sensitivity, specificity, and negative predictive value (NPV)
Test Target molecule Sensitivity Specificity NPV Comment
Cytology [54, NA 34–42 96–99 68–99 Performance depends
89–91] on grade of
UC. Interlaboratory
variation
UroVysion Chromosome 3, 72 83 – Performance is lower
FISHa [17] 7, 17 in low-grade UC
enumeration
and 9p21 locus
BTA stat and Human 65 74 – Performance is lower
TRAKa [51, 92] complement in low-grade
factor H-related UC. Lower specificity
protein in benign non-
neoplastic urologic
disorders
NMP22a [47, 54] Nuclear matrix 52–59 87–89 – Variation in
protein performance between
studies. False positives
in patients with
non-neoplastic
urologic diseases
ADXBLADDER Single protein 45–88 70–73 74–100 ELISA test, adversely
[54] (MCM5) affected by
inflammation or
instrumentation
CxBladder mRNA (5 gene 93–95 96–97 Not influenced by
Monitor [59] expression previous BCG therapy
[93] markers
combined with
2 clinical
features)
(continued)
208 L. Bubendorf et al.
Table 9.1 (continued)
Test Target molecule Sensitivity Specificity NPV Comment
Bladder Methylated 62–95 82–88 79–99 Not influenced by
EpiCheck [54, DNA (15 inflammation or
68, 94] markers) hematuria. Lower
sensitivity for
low-grade UC
compared to
high-grade UC
Uromonitor-V2a PCR DNA (3 73–93 86 95 Not influenced by
[75, 76] markers; TERT, inflammation.
FGFR3 and Roughly equivalent
KRAS) performance for
low-grade and
high-grade UC
uCAPP-Seq [87] NGS-based test 84 96 – Somewhat better
assessing 20 sensitivity for
mutations high-grade UC
compared to
low-grade UC
AssureMDx [95] NGS-based 93 86 – Better test
test, assessing performance for
mutations in high-grade than
TERT, FGFR3, low-grade UC.
HRAS, and
DNA
methylation of
OTX1,
ONECUT2, and
TWIST1
UroSeek [81, 96] NGS-based 83 99 – Complementary to
test, mutations cytology. Equivalent
in 11 genes, performance for
and detection of low-grade and
aneuploidy high-grade UC. Lower
sensitivity in a cohort
followed for
recurrence
FDA-approved tests
a
[84], which are often denuded due to exfoliation of CIS cells. Thus, NGS may be an
important ancillary test in patients with a positive urine cytology when searching for
genetic alterations that might predict the response to intravesical BCG treatment.
There are many approaches for detecting tumor cell mutations when the propor-
tion of tumor cells is relatively low. One approach is using cell-free DNA (DNA in
suspension or not associated with cells) within a urine sample. The appeal of cell-
free DNA is the expectation that tumor cells have a relatively higher turnover than
benign cells, such that fragments of their genome will be enriched in the fluid com-
pared to the genomes of normal cells. In fact, the performance of cellular DNA- and
cell-free DNA-based NGS assays has been comparable [82, 85].
For cell-based assays, microfluidic techniques have been developed attempting
to enrich tumor cells based on physical properties or surface antigens. Initial posi-
tive results using microfluidics to enrich tumor cell DNA have not been validated
prospectively [86]. Cytologists are aware that UC cells can have a wide variety of
physical shapes and sizes, making microfluidic approaches seem difficult.
Enrichment of tumor DNA can also be accomplished through genetic techniques.
For example, uCAPP-Seq enriches mutant alleles by pre-hybridizing to immobi-
lized probes that recognize the mutation [87]. When applied to cell-free urine sam-
ples, uCAPP-Seq had a sensitivity of 84% when blinded to tumor mutation status
[87]. Laser capture microdissection of individual cells, as guided by cytologic
assessment, has been proposed as another means of enriching tumor cells or exclud-
ing obviously normal cells [83]. This “cytologically targeted NGS” would ulti-
mately allow morphologic features to be correlated on a per cell basis—analogous
to target FISH.
At least 25 publications of ancillary NGS using a wide variety of platforms have
been published [79]. The performances of the various studies are difficult to com-
pare since prevalence of disease in different studies varies widely, and some studies
only assessed urine samples in which there were known mutations in a bladder
tumor sample from the same patient. Table 9.1 lists selected ancillary tests in rela-
tion to conventional cytology.
Conclusions
The existing evidence strongly points towards a utility of U-FISH in the setting of
atypical urinary cytology. Since the definition of “atypia” and its boundaries with
“suspicious” have been variable in the past, further studies are needed to better
determine the performance of UroVysion® FISH and other ancillary tests in the
more stringently and precisely defined AUC and SHGUC categories of TPS. Other
urine-based biomarkers for diagnosis and follow-up of urothelial neoplasia have
limited value for the time being, and many require further validation. Liquid-based
tests done outside of the cytology laboratory such as BTA and NMP22 tests have a
sensitivity comparable to that of cytology for high-grade tumors and are superior for
the clinically less important low-grade tumors. Although urine marker-guided fol-
low-up of patients with bladder cancer appears feasible, studies proving the efficacy
212 L. Bubendorf et al.
of this concept and demonstrating an added value for patients or the health system
are lacking. Therefore, based on current levels of evidence, systematic marker-
guided follow-up is still not recommended [1, 2, 4]. However, this might change in
the near future given the promising results of new molecular tests. According to the
current EAU guidelines, diagnostic markers might be most useful in patients with
negative cystoscopy and UUT work-up to identify patients at increased risk of dis-
ease recurrence and possibly progression [88]. Until there are new international
consensus recommendations on ancillary testing in urinary cytology, U-FISH
should best be used judiciously on the basis of cytological findings in an individual
case after prior consultation with the treating physicians or as a reflex test. For the
future, assessing multiple biomarkers simultaneously, for example, by using next-
generation sequencing, is likely to be most sensitive, therefore having a sufficiently
high negative predictive value to help avoid unnecessary cystoscopies. There are
several opportunities for future studies to specify the role of ancillary testing in
urinary cytology. Well-controlled and independently funded studies are needed to
clarify the performance of U-FISH and other tests in clearly defined situations and
to analyze the cost-benefit ratio for both the laboratory and patient management. For
example, there is still insufficient clinical evidence to change the currently estab-
lished surveillance strategies in non-muscle-invasive UC based on U-FISH or other
tests in urines with atypical cytology at the time of a negative surveillance cystos-
copy. It also remains to be studied to what degree targeted U-FISH of atypical or
suspicious cytology after automated relocation can help the cytopathologists to
improve their skills in differentiating morphological features of benign/reactive and
neoplastic cells using the chromosomal status as a reference. Whether U-FISH or
other molecular test results should be integrated in new TPS categories will be a
subject of future discussions.
Sample Reports
Voided urine: Several degenerated urothelial cells with nuclear atypia of uncertain
significance.
TPS category: AUC.
Results of UroVysion FISH Testing (Chromosomes 3, 7, 17, and 9p21): Positive
(10/25 cells).
Final diagnosis (cytology & FISH): Consistent with urothelial neoplasia
(see note).
Note: Ten of the 25 atypical cells that were scored revealed unequivocal chromo-
somal aberrations including polysomy of chromosomes 3 and 17 and homozygous
deletion of 9p21. This positive test result is in favor of urothelial neoplasia. A dis-
tinction between low-grade and high-grade lesion is not possible. We advise further
investigations.
214 L. Bubendorf et al.
Appendix
UroVysion® Assay
The slide pretreatment with protease uncovers target DNA and is recommended in
Pap-stained specimens. Decolorization is not mandatory since the stain is removed
during further phases of FISH procedure. If using archival slides, remove the cover-
slip and mounting medium in xylene. Place the slides in 1% acid alcohol (HCL and
70% alcohol) overnight or until decolorized. The U-FISH assay can subsequently
be conducted either manually or automatically. The first step is denaturation of
specimen DNA to expose single-stranded target DNA. U-FISH probes should be
prepared accordingly and applied to the selected area of the slide. The area should
be coverslipped and sealed immediately to ensure optimal conditions. Hybridization
of probes to target DNA sequences follows under appropriate conditions. The pro-
cedure is finished with post-hybridization washes to remove excessive probes.
Slides should be dried in a dark area. The exact procedure of the FISH assay is
described in the UroVysion kit datasheet. The procedure should be validated in each
individual laboratory, together with positive and negative controls, to ensure opti-
mal hybridization. Afterwards the specimen chosen for analysis is stained by DAPI
solution. Slides are coverslipped and stored at −20 °C in the dark until analysis.
analysis work station. It features automated slide loading and handling, low and
high magnification scanning to identify cells of interest, and digital image acquisi-
tion [10]. The MetaSystems uses an automated fluorescent scanning microscope
and analysis software with “tile-sampling” method [16].
The BioView Duet System™ scans cells that are imaged at high resolution
(under oil immersion) both in bright light illumination and in fluorescent illumina-
tion. Cells are classified by the system according to their morphological features,
according to their staining on bright field (Giemsa or Papanicolaou stains, if target
FISH is used), and according to the pattern of fluorescent signals. The automated
microscope has micrometer-level precision in the X, Y, and Z axes which allows it
to focus on cells and retain coordinate information for target cells. There are two
modes of operation: [1] automatic scanning, which provides a gallery of all fields of
view, and [2] manual scanning, which provides interactive control allowing the user
to select the fields of view using either bright-field or fluorescent illumination.
Similar to manual scoring, the automated system scans the FISH slides by locat-
ing and scoring the nuclei exhibiting abnormalities such as enlargement, irregular
borders, and patchy DAPI staining. As identification of abnormal or malignant cells
based solely on aberrant morphology may be misleading, the BioView System™
classifies cells both by morphology on the DAPI fluorescence and by superimposed
FISH signals. Cells are ranked based on a combination of nuclear features, includ-
ing size, shape, DAPI intensity, and DAPI standard deviation inside the nucleus.
References
1. Chang SS, Boorjian SA, Chou R, Clark PE, Daneshmand S, Konety BR, et al. Diagnosis
and treatment of non-muscle invasive bladder cancer: AUA/SUO guideline. J Urol.
2016;196(4):1021–9.
2. Babjuk M, Burger M, Comperat EM, Gontero P, Mostafid AH, Palou J, et al. European
Association of Urology guidelines on non-muscle-invasive bladder cancer (TaT1 and carci-
noma in situ) - 2019 update. Eur Urol. 2019;76(5):639–57.
3. Barkan GA, Tabatabai ZL, Kurtycz DFI, Padmanabhan V, Souers RJ, Nayar R, et al. Practice
patterns in urinary cytopathology prior to the Paris system for reporting urinary cytology. Arch
Pathol Lab Med. 2020;144(2):172–6.
4. Kamat AM, Hegarty PK, Gee JR, Clark PE, Svatek RS, Hegarty N, et al. ICUD-EAU interna-
tional consultation on bladder cancer 2012: screening, diagnosis, and molecular markers. Eur
Urol. 2013;63(1):4–15.
5. Soria F, Droller MJ, Lotan Y, Gontero P, D'Andrea D, Gust KM, et al. An up-to-date catalog of
available urinary biomarkers for the surveillance of non-muscle invasive bladder cancer. World
J Urol. 2018;36(12):1981–95.
6. Yang M, Zheng Z, Zhuang Z, Zhao X, Xu Z, Lin H. ImmunoCyt and cytology for diagnosis of
bladder carcinoma: a meta analysis. Chin Med J. 2014;127(4):758–64.
7. Bubendorf L, Piaton E. UroVysion(R) multiprobe FISH in the triage of equivocal urinary
cytology cases. Ann Pathol. 2012;32(6):e52–6. 438-43.
8. Halling KC, Kipp BR. Bladder cancer detection using FISH (UroVysion assay). Adv Anat
Pathol. 2008;15(5):279–86.
9. Mischinger J, Guttenberg LP, Hennenlotter J, Gakis G, Aufderklamm S, Rausch S, et al.
Comparison of different concepts for interpretation of chromosomal aberrations in uro-
216 L. Bubendorf et al.
thelial cells detected by fluorescence in situ hybridization. J Cancer Res Clin Oncol.
2017;143(4):677–85.
10. Marganski WA, El-Sirgany Costa V, Kilpatrick MW, Tafas T, Yim J, Matthews M. Digitized
microscopy in the diagnosis of bladder cancer: analysis of >3000 cases during a 7-month
period. Cancer Cytopathol. 2011;119(4):279–89.
11. Smith GD, Riding M, Oswald K, Bentz JS. Integrating a FISH imaging system into the cytol-
ogy laboratory. Cytojournal. 2010;7:3.
12. Smith GD, Bentz JS. "FISHing" to detect urinary and other cancers: validation of an imaging
system to aid in interpretation. Cancer Cytopathol. 2010;118(1):56–64.
13. Daniely M, Rona R, Kaplan T, Olsfanger S, Elboim L, Zilberstien Y, et al. Combined analysis
of morphology and fluorescence in situ hybridization significantly increases accuracy of blad-
der cancer detection in voided urine samples. Urology. 2005;66(6):1354–9.
14. Bubendorf L. Multiprobe fluorescence in situ hybridization (UroVysion) for the detection of
urothelial carcinoma - FISHing for the right catch. Acta Cytol. 2011;55(2):113–9.
15. Vlajnic T, Gut A, Savic S, Bubendorf L. The Paris system for reporting urinary cytol-
ogy in daily practice with emphasis on ancillary testing by multiprobe FISH. J Clin Pathol.
2020;73(2):90–5.
16. Furrer D, Jacob S, Caron C, Sanschagrin F, Provencher L, Diorio C. Validation of a new clas-
sifier for the automated analysis of the human epidermal growth factor receptor 2 (HER2) gene
amplification in breast cancer specimens. Diagn Pathol. 2013;8:17.
17. Hajdinjak T. UroVysion FISH test for detecting urothelial cancers: meta-analysis of diagnostic
accuracy and comparison with urinary cytology testing. Urol Oncol. 2008;26(6):646–51.
18. Gomella LG, Mann MJ, Cleary RC, Hubosky SG, Bagley DH, Thumar AB, et al. Fluorescence
in situ hybridization (FISH) in the diagnosis of bladder and upper tract urothelial carcinoma:
the largest single-institution experience to date. Can J Urol. 2017;24(1):8620–6.
19. Lavery HJ, Zaharieva B, McFaddin A, Heerema N, Pohar KS. A prospective comparison of
UroVysion FISH and urine cytology in bladder cancer detection. BMC Cancer. 2017;17(1):247.
20. McHale T, Ohori NP, Cieply KM, Sherer C, Bastacky SI. Comparison of urinary cytology
and fluorescence in situ hybridization in the detection of urothelial neoplasia: an analysis of
discordant results. Diagn Cytopathol. 2019;47(4):282–8.
21. Miki Y, Neat M, Chandra A. Application of the Paris system to atypical urine cytology sam-
ples: correlation with histology and UroVysion((R)) FISH. Cytopathology. 2017;28(2):88–95.
22. Gayed BA, Seideman C, Lotan Y. Cost-effectiveness of fluorescence in situ hybridiza-
tion in patients with atypical cytology for the detection of urothelial carcinoma. J Urol.
2013;190(4):1181–6.
23. Kim PH, Sukhu R, Cordon BH, Sfakianos JP, Sjoberg DD, Hakimi AA, et al. Reflex fluo-
rescence in situ hybridization assay for suspicious urinary cytology in patients with bladder
cancer with negative surveillance cystoscopy. BJU Int. 2014;114(3):354–9.
24. Kipp BR, Halling KC, Campion MB, Wendel AJ, Karnes RJ, Zhang J, et al. Assessing the
value of reflex fluorescence in situ hybridization testing in the diagnosis of bladder cancer
when routine urine cytological examination is equivocal. J Urol. 2008;179(4):1296–301; dis-
cussion 301.
25. Lotan Y, Bensalah K, Ruddell T, Shariat SF, Sagalowsky AI, Ashfaq R. Prospective evalu-
ation of the clinical usefulness of reflex fluorescence in situ hybridization assay in patients
with atypical cytology for the detection of urothelial carcinoma of the bladder. J Urol.
2008;179(6):2164–9.
26. Montalbo R, Izquierdo L, Ingelmo-Torres M, Galve P, Sole M, Franco A, et al. Urine cytology
suspicious for urothelial carcinoma: prospective follow-up of cases using cytology and urine
biomarker-based ancillary techniques. Cancer Cytopathol. 2020;128(7):460–9.
27. Savic S, Zlobec I, Thalmann GN, Engeler D, Schmauss M, Lehmann K, et al. The prognos-
tic value of cytology and fluorescence in situ hybridization in the follow-up of nonmuscle-
invasive bladder cancer after intravesical bacillus Calmette-Guerin therapy. Int J Cancer.
2009;124(12):2899–904.
9 Ancillary Studies in Urinary Cytology 217
28. Schlomer BJ, Ho R, Sagalowsky A, Ashfaq R, Lotan Y. Prospective validation of the clinical
usefulness of reflex fluorescence in situ hybridization assay in patients with atypical cytology
for the detection of urothelial carcinoma of the bladder. J Urol. 2010;183(1):62–7.
29. Seideman C, Canter D, Kim P, Cordon B, Weizer A, Oliva I, et al. Multicenter evaluation of
the role of UroVysion FISH assay in surveillance of patients with bladder cancer: does FISH
positivity anticipate recurrence? World J Urol. 2014.
30. Virk RK, Abro S, de Ubago JMM, Pambuccian SE, Quek ML, Wojcik EM, et al. The value
of the UroVysion(R) FISH assay in the risk-stratification of patients with "atypical urothelial
cells" in urinary cytology specimens. Diagn Cytopathol. 2017;45(6):481–500.
31. Skacel M, Fahmy M, Brainard JA, Pettay JD, Biscotti CV, Liou LS, et al. Multitarget fluores-
cence in situ hybridization assay detects transitional cell carcinoma in the majority of patients
with bladder cancer and atypical or negative urine cytology. J Urol. 2003;169(6):2101–5.
32. Yoder BJ, Skacel M, Hedgepeth R, Babineau D, Ulchaker JC, Liou LS, et al. Reflex UroVysion
testing of bladder cancer surveillance patients with equivocal or negative urine cytology: a pro-
spective study with focus on the natural history of anticipatory positive findings. Am J Clin
Pathol. 2007;127(2):295–301.
33. Fritsche HM, Burger M, Dietmaier W, Denzinger S, Bach E, Otto W, et al. Multicolor FISH
(UroVysion) facilitates follow-up of patients with high-grade urothelial carcinoma of the blad-
der. Am J Clin Pathol. 2010;134(4):597–603.
34. Kipp BR, Karnes RJ, Brankley SM, Harwood AR, Pankratz VS, Sebo TJ, et al. Monitoring
intravesical therapy for superficial bladder cancer using fluorescence in situ hybridization. J
Urol. 2005;173(2):401–4.
35. Whitson J, Berry A, Carroll P, Konety B. A multicolour fluorescence in situ hybridization test
predicts recurrence in patients with high-risk superficial bladder tumours undergoing intravesi-
cal therapy. BJU Int. 2009;104(3):336–9.
36. Mengual L, Marin-Aguilera M, Ribal MJ, Burset M, Villavicencio H, Oliver A, et al. Clinical
utility of fluorescent in situ hybridization for the surveillance of bladder cancer patients treated
with bacillus Calmette-Guerin therapy. Eur Urol. 2007;52(3):752–9.
37. Liem E, Oddens JR, Vernooij RWM, Li R, Kamat A, Dinney CP, et al. The role of fluores-
cence in situ hybridization for predicting recurrence after adjuvant bacillus Calmette-Guerin
in patients with intermediate and high risk nonmuscle invasive bladder cancer: a systematic
review and meta-analysis of individual patient data. J Urol. 2020;203(2):283–91.
38. Freund JE, Liem E, Savci-Heijink CD, de Reijke TM. Fluorescence in situ hybridization in 1
mL of selective urine for the detection of upper tract urothelial carcinoma: a feasibility study.
Med Oncol. 2018;36(1):10.
39. Jin H, Lin T, Hao J, Qiu S, Xu H, Yu R, et al. A comprehensive comparison of fluorescence in
situ hybridization and cytology for the detection of upper urinary tract urothelial carcinoma: a
systematic review and meta-analysis. Medicine (Baltimore). 2018;97(52):e13859.
40. Reynolds JP, Voss JS, Kipp BR, Karnes RJ, Nassar A, Clayton AC, et al. Comparison of urine
cytology and fluorescence in situ hybridization in upper urothelial tract samples. Cancer
Cytopathol. 2014;122(6):459–67.
41. Sassa N, Iwata H, Kato M, Murase Y, Seko S, Nishikimi T, et al. Diagnostic utility of
UroVysion combined with conventional urinary cytology for urothelial carcinoma of the upper
urinary tract. Am J Clin Pathol. 2019;151(5):469–78.
42. Sutton AJ, Lamont JV, Evans RM, Williamson K, O'Rourke D, Duggan B, et al. An early
analysis of the cost-effectiveness of a diagnostic classifier for risk stratification of haematuria
patients (DCRSHP) compared to flexible cystoscopy in the diagnosis of bladder cancer. PLoS
One. 2018;13(8):e0202796.
43. Huysentruyt CJ, Baldewijns MM, Ruland AM, Tonk RJ, Vervoort PS, Smits KM, et al.
Modified UroVysion scoring criteria increase the urothelial carcinoma detection rate in cases
of equivocal urinary cytology. Histopathology. 2011;58(7):1048–53.
44. Tapia C, Glatz K, Obermann EC, Grilli B, Barascud A, Herzog M, et al. Evaluation of chromo-
somal aberrations in patients with benign conditions and reactive changes in urinary cytology.
Cancer Cytopathol. 2011;119(6):404–10.
218 L. Bubendorf et al.
45. Zellweger T, Benz G, Cathomas G, Mihatsch MJ, Sulser T, Gasser TC, et al. Multi-target fluo-
rescence in situ hybridization in bladder washings for prediction of recurrent bladder cancer.
Int J Cancer. 2006;119(7):1660–5.
46. Zhou AG, Liu Y, Cyr MS, Garver J, Woda BA, Cosar EF, et al. Role of Tetrasomy for the diag-
nosis of urothelial carcinoma using UroVysion fluorescent in situ hybridization. Arch Pathol
Lab Med. 2016;140(6):552–9.
47. Wang J, Batourina E, Schneider K, Souza S, Swayne T, Liu C, et al. Polyploid superfi-
cial cells that maintain the urothelial barrier are produced via incomplete cytokinesis and
Endoreplication. Cell Rep. 2018;25(2):464–77. e4
48. Hossain D, Hull D, Kalantarpour F, Maitlen R, Qian J, Bostwick DG. Does polyomavirus
infection interfere with bladder cancer fluorescence in situ hybridization? Diagn Cytopathol.
2014;42(3):225–9.
49. van der Aa MN, Steyerberg EW, Bangma C, van Rhijn BW, Zwarthoff EC, van der Kwast
TH. Cystoscopy revisited as the gold standard for detecting bladder cancer recurrence: diag-
nostic review bias in the randomized, prospective CEFUB trial. J Urol. 2010;183(1):76–80.
50. Kinders R, Jones T, Root R, Bruce C, Murchison H, Corey M, et al. Complement factor H
or a related protein is a marker for transitional cell cancer of the bladder. Clin Cancer Res.
1998;4(10):2511–20.
51. Oliveira MCD, Caires HR, Oliveira MJ, Fraga A, Vasconcelos MH, Ribeiro R. Urinary bio-
markers in bladder cancer: where do we stand and potential role of extracellular vesicles.
Cancers. 2020;12(6):1400.
52. van Rhijn BW, van der Poel HG, van der Kwast TH. Urine markers for bladder cancer surveil-
lance: a systematic review. Eur Urol. 2005;47(6):736–48.
53. Leyh H, Marberger M, Conort P, Sternberg C, Pansadoro V, Pagano F, et al. Comparison of
the BTA stat test with voided urine cytology and bladder wash cytology in the diagnosis and
monitoring of bladder cancer. Eur Urol. 1999;35(1):52–6.
54. Wolfs JRE, Hermans TJN, Koldewijn EL, van de Kerkhof D. Novel urinary biomarkers
ADXBLADDER and bladder EpiCheck for diagnostics of bladder cancer: a review. Urol
Oncol. 2021;
55. Soloway MS, Briggman V, Carpinito GA, Chodak GW, Church PA, Lamm DL, et al. Use of
a new tumor marker, urinary NMP22, in the detection of occult or rapidly recurring transi-
tional cell carcinoma of the urinary tract following surgical treatment. J Urol. 1996;156(2 Pt
1):363–7.
56. Miyake M, Goodison S, Giacoia EG, Rizwani W, Ross S, Rosser CJ. Influencing factors on the
NMP-22 urine assay: an experimental model. BMC Urol. 2012;12:23.
57. Ng K, Stenzl A, Sharma A, Vasdev N. Urinary biomarkers in bladder cancer: a review of the
current landscape and future directions. Urol Oncol. 2021;39(1):41–51.
58. Allison DB, VandenBussche CJ. A review of urine ancillary tests in the era of the Paris system.
Acta Cytol. 2020;64(1–2):182–92.
59. Lotan Y, O'Sullivan P, Raman JD, Shariat SF, Kavalieris L, Frampton C, et al. Clinical com-
parison of noninvasive urine tests for ruling out recurrent urothelial carcinoma. Urol Oncol.
2017;35(8):531 e15–22.
60. Sapre N, Anderson PD, Costello AJ, Hovens CM, Corcoran NM. Gene-based urinary biomark-
ers for bladder cancer: an unfulfilled promise? Urol Oncol. 2014;32(1):48 e9–17.
61. Tan WS, Tan WP, Tan MY, Khetrapal P, Dong L, De Winter P, et al. Novel urinary biomarkers
for the detection of bladder cancer: a systematic review. Cancer Treat Rev. 2018;69:39–52.
62. Carolina BBoN. Urinary Tumor Markers for Bladder Cancer AHS – G2125: BlueCross
BlueShield of North Carolina; 2020 Available from: https://www.bluecrossnc.com/sites/
default/files/document/attachment/services/public/pdfs/medicalpolicy/urinary_tumor_mark-
ers_for_bladder_cancer.pdf.
63. Roupret M, Gontero P, McCracken SRC, Dudderidge T, Stockley J, Kennedy A, et al.
Diagnostic accuracy of MCM5 for the detection of recurrence in nonmuscle invasive blad-
der cancer Followup: a blinded, prospective cohort, Multicenter European study. J Urol.
2020;204(4):685–90.
9 Ancillary Studies in Urinary Cytology 219
64. Gontero P, Montanari E, Roupret M, Longo F, Stockley J, Kennedy A, et al. Comparison
of the performances of the ADXBLADDER test and urinary cytology in the follow-up of
non-muscle-invasive bladder cancer: a blinded prospective multicentric study. BJU Int.
2021;127(2):198–204.
65. Robertson AG, Kim J, Al-Ahmadie H, Bellmunt J, Guo G, Cherniack AD, et al. Comprehensive
molecular characterization of muscle-invasive bladder cancer. Cell. 2017;171(3):540–56. e25.
66. Wolff EM, Chihara Y, Pan F, Weisenberger DJ, Siegmund KD, Sugano K, et al. Unique DNA
methylation patterns distinguish noninvasive and invasive urothelial cancers and establish an
epigenetic field defect in premalignant tissue. Cancer Res. 2010;70(20):8169–78.
67. Beukers W, Kandimalla R, Masius RG, Vermeij M, Kranse R, van Leenders GJ, et al.
Stratification based on methylation of TBX2 and TBX3 into three molecular grades predicts
progression in patients with pTa-bladder cancer. Mod Pathol. 2015;28(4):515–22.
68. Mancini M, Righetto M, Zumerle S, Montopoli M, Zattoni F. The Bladder EpiCheck Test as a
Non-Invasive Tool Based on the Identification of DNA Methylation in Bladder Cancer Cells in
the Urine: A Review of Published Evidence. Int J Mol Sci. 2020;21(18).
69. Trenti E, D'Elia C, Mian C, Schwienbacher C, Hanspeter E, Pycha A, et al. Diagnostic predic-
tive value of the bladder EpiCheck test in the follow-up of patients with non-muscle-invasive
bladder cancer. Cancer Cytopathol. 2019;127(7):465–9.
70. Pierconti F, Martini M, Fiorentino V, Cenci T, Capodimonti S, Straccia P, et al. The combina-
tion cytology/epichek test in non muscle invasive bladder carcinoma follow-up: Effective tool
or useless expence? Urol Oncol. 2021;39(2):131 e17–21.
71. Allory Y, Beukers W, Sagrera A, Flandez M, Marques M, Marquez M, et al. Telomerase
reverse transcriptase promoter mutations in bladder cancer: high frequency across stages,
detection in urine, and lack of association with outcome. Eur Urol. 2014;65(2):360–6.
72. Leao R, Lee D, Figueiredo A, Hermanns T, Wild P, Komosa M, et al. Combined genetic and
epigenetic alterations of the TERT promoter affect clinical and biological behavior of bladder
cancer. Int J Cancer. 2019;144(7):1676–84.
73. van Rhijn BW, Vis AN, van der Kwast TH, Kirkels WJ, Radvanyi F, Ooms EC, et al.
Molecular grading of urothelial cell carcinoma with fibroblast growth factor receptor 3 and
MIB-1 is superior to pathologic grade for the prediction of clinical outcome. J Clin Oncol.
2003;21(10):1912–21.
74. Jebar AH, Hurst CD, Tomlinson DC, Johnston C, Taylor CF, Knowles MA. FGFR3 and Ras
gene mutations are mutually exclusive genetic events in urothelial cell carcinoma. Oncogene.
2005;24(33):5218–25.
75. Batista R, Vinagre J, Prazeres H, Sampaio C, Peralta P, Conceicao P, et al. Validation of a novel,
sensitive, and specific urine-based test for recurrence surveillance of patients with non-muscle-
invasive bladder cancer in a comprehensive Multicenter study. Front Genet. 2019;10:1237.
76. Sieverink CA, Batista RPM, Prazeres HJM, Vinagre J, Sampaio C, Leao RR, et al. Clinical
Validation of a Urine Test (Uromonitor-V2((R))) for the Surveillance of Non-Muscle-Invasive
Bladder Cancer Patients. Diagnostics (Basel). 2020;10(10).
77. Kandimalla R, Masius R, Beukers W, Bangma CH, Orntoft TF, Dyrskjot L, et al. A 3-plex
methylation assay combined with the FGFR3 mutation assay sensitively detects recurrent
bladder cancer in voided urine. Clin Cancer Res. 2013;19(17):4760–9.
78. Roperch JP, Hennion C. A novel ultra-sensitive method for the detection of FGFR3 mutations in
urine of bladder cancer patients - Design of the Urodiag(R) PCR kit for surveillance of patients
with non-muscle-invasive bladder cancer (NMIBC). BMC Med Genet. 2020;21(1):112.
79. Hentschel AE, van der Toom EE, Vis AN, Ket JCF, Bosschieter J, Heymans MW, et al. A
systematic review on mutation markers for bladder cancer diagnosis in urine. BJU Int.
2021;127(1):12–27.
80. Avogbe PH, Manel A, Vian E, Durand G, Forey N, Voegele C, et al. Urinary TERT promoter
mutations as non-invasive biomarkers for the comprehensive detection of urothelial cancer.
EBioMedicine. 2019;44:431–8.
220 L. Bubendorf et al.
81. Springer SU, Chen CH, Rodriguez Pena MDC, Li L, Douville C, Wang Y, et al. Non-invasive
detection of urothelial cancer through the analysis of driver gene mutations and aneuploidy.
elife. 2018;7
82. Ward DG, Gordon NS, Boucher RH, Pirrie SJ, Baxter L, Ott S, et al. Targeted deep sequenc-
ing of urothelial bladder cancers and associated urinary DNA: a 23-gene panel with utility for
non-invasive diagnosis and risk stratification. BJU Int. 2019;124(3):532–44.
83. Harris T, Sheel A, Zong Y, Hutchinson LM, Cornejo KM, Bubendorf L, et al. Cytologically
targeted next-generation sequencing: a synergy for diagnosing urothelial carcinoma. J Am Soc
Cytopathol. 2021;10(1):94–102.
84. Scott SN, Ostrovnaya I, Lin CM, Bouvier N, Bochner BH, Iyer G, et al. Next-generation
sequencing of urine specimens: a novel platform for genomic analysis in patients with non-
muscle-invasive urothelial carcinoma treated with bacille Calmette-Guerin. Cancer Cytopathol.
2017;125(6):416–26.
85. Hentschel AE, Nieuwenhuijzen JA, Bosschieter J, Splunter APV, Lissenberg-Witte BI, Voorn
JPV, et al. Comparative Analysis of Urine Fractions for Optimal Bladder Cancer Detection
Using DNA Methylation Markers. Cancers (Basel). 2020;12(4).
86. Chen A, Fu G, Xu Z, Sun Y, Chen X, Cheng KS, et al. Detection of urothelial bladder carci-
noma via microfluidic immunoassay and single-cell DNA copy-number alteration analysis of
captured urinary-exfoliated tumor cells. Cancer Res. 2018;78(14):4073–85.
87. Dudley JC, Schroers-Martin J, Lazzareschi DV, Shi WY, Chen SB, Esfahani MS, et al. Detection
and surveillance of bladder cancer using urine tumor DNA. Cancer Discov. 2019;9(4):500–9.
88. Babjuk M, Compérat E, Gontero P, Mostafid AH, Palou J, van Rhijn BWG, Rouprêt M, Shariat
SF, Sylvester R, Zigeuner R. Non-muscle-invasive Bladder Cancer: european association of
urology; Available from: https://uroweb.org/guideline/non-muscle-invasive-bladder-cancer/.
89. Guo A, Wang X, Gao L, Shi J, Sun C, Wan Z. Bladder tumour antigen (BTA stat) test com-
pared to the urine cytology in the diagnosis of bladder cancer: a meta-analysis. Can Urol Assoc
J. 2014;8(5–6):E347–52.
90. McIntire PJ, Khan R, Hussain H, Pambuccian SE, Wojcik EM, Barkan GA. Negative predic-
tive value and sensitivity of urine cytology prior to implementation of the Paris system for
reporting urinary cytology. Cancer Cytopathol. 2019;127(2):125–31.
91. McIntire PJ, Kilic I, Pambuccian SE, Wojcik EM, Barkan GA. The Paris system for reporting
urinary cytology reduces atypia rates and does not alter the negative predictive value of urine
cytology. J Am Soc Cytopathol. 2021;10(1):14–9.
92. Chou R, Gore JL, Buckley D, Fu R, Gustafson K, Griffin JC, et al. Urinary biomarkers
for diagnosis of bladder cancer: a systematic review and meta-analysis. Ann Intern Med.
2015;163(12):922–31.
93. Kavalieris L, O'Sullivan P, Frampton C, Guilford P, Darling D, Jacobson E, et al. Performance
characteristics of a multigene urine biomarker test for monitoring for recurrent urothelial car-
cinoma in a Multicenter study. J Urol. 2017;197(6):1419–26.
94. Witjes JA, Morote J, Cornel EB, Gakis G, van Valenberg FJP, Lozano F, et al. Performance of
the bladder EpiCheck methylation test for patients under surveillance for non-muscle-invasive
bladder cancer: results of a Multicenter, prospective. Blinded Clinical Trial Eur Urol Oncol.
2018;1(4):307–13.
95. van Kessel KE, Beukers W, Lurkin I. Ziel-van der made a, van der Keur KA, Boormans JL,
et al. validation of a DNA methylation-mutation urine assay to select patients with Hematuria
for cystoscopy. J Urol. 2017;197(3 Pt 1):590–5.
96. Rodriguez Pena MDC, Springer SU, Taheri D, Li L, Tregnago AC, Eich ML, et al. Performance
of novel non-invasive urine assay UroSEEK in cohorts of equivocal urine cytology. Virchows
Arch. 2020;476(3):423–9.
Cytopreparatory Techniques
10
Donna K. Russell, Willam N. Crabtree, and Gary W. Gill
Background
Specimen Collection
Assessment
Gross examination of the fluid is noted upon receipt in the cytopathology laboratory.
Physical features such as volume of sample, color, clarity, opalescence, and viscos-
ity are then documented in the cytopathology report [1]. If the cell sample is low in
volume, is clear in appearance, and contains only a few cells, the method of prepara-
tion should include a concentrating cytocentrifugation or a liquid-based preparation
method. The method of cytopreparation may be selected on the basis of the physical
properties of the specimen and the patient’s clinical history. For example, highly
viscous specimens may require a smear technique or mucolytic agents, and bloody
specimens will benefit from methods that segregate epithelial cells from red blood
cells. Highly cellular specimens are ideal for smeared preparations, whereas sparsely
cellular specimens will require multiple centrifugation steps and special cell con-
solidation processing.
Processing
A collection method should harvest well-preserved cells that reliably represent any
urinary tract lesion that might be present. As urine itself may degrade cells that have
been sitting in the bladder for extended periods of time, cells expelled during the
first void of the morning are less than optimal; it is best to discard the initial voided
urine and use the next voided fresh urine after the patient drinks 8–16 oz. of water.
Clinically the laboratory usually receives sterile, pyrogen-free balanced salt solu-
tions from specimens associated with bladder instrumentation. Normal saline (i.e.,
0.9% NaCl w/v) is discouraged [2] as it may ruin cell morphology [3]. Saline may
be isotonic, but it is not physiologic.
Ideally, freshly collected urine specimens should be processed promptly after
collection. Cells in urine specimens may be degenerated at the time of collection,
and no amount of preservative can save them and may even increase cellular degen-
eration [4, 5]. Process urine specimens within about 4 hours, and if this is not fea-
sible, specimens may be refrigerated for up to 12 hours without noticeable effects
on specimen quality. However, refrigeration accelerates the precipitation of salts
in urines.
10 Cytopreparatory Techniques 223
Preparation Methods
8.14%
44.41%
37.29%
8.81%
Other
The ThinPrep® 5000 processor (Fig. 10.4) delivers a fully automated slide prep-
aration option for non-gynecologic urine samples. It is a continuous hands-free pro-
cessor and prepares up to 20 samples per batch, which allows 45 minutes of free
walk-away time for the cytopreparatory technologist. SOP for ThinPrep preparation
is documented in the laboratory procedure manual (Appendix B). (For exact details,
refer to manufacturer’s procedure manual.) [8, 9].
10 Cytopreparatory Techniques 225
Fig. 10.2 ThinPrep®
liquid-based preparation
technique
1. Combine into
50 mL tube
2. Concentrate
by Centrifugation
3. Pour Off
Supernatant
and Resuspend
Cell Pellet
4. Add an Appropriate
Amount of Specimen
to the PreservCyt
Solution Vial
5. Run on ThinPrepTM
2000 Processor Using
Sequence 2 or
ThinPrepTM 5000
Processor Using
sequence Non-Gyn
226 D. K. Russell et al.
Fig. 10.3 High-grade
urothelial carcinoma.
(ThinPrep® preparation,
voided urine, Papanicolaou
stain, high magnification)
Fig. 10.4 Hologic
ThinPrep® 5000®
Processor
Cytocentrifugation
Cytocentrifugation with the Cytospin deposits the cells onto a 6 mm circle on a
vertical slide while the fluid is removed by the surrounding absorbent paper. A slide
(label at the top) is placed in the clip with an absorbent paper on top and the plastic
funnel chamber on top of both. The Cytospin clip is closed by swinging the wire
10 Cytopreparatory Techniques 227
toward the clip back and hooking it closed. The circle of the funnel and the slide
must match up exactly. Slide clips are placed in the centrifuge opposite one another
for balancing, and specimen is added to the chamber. Five to eight drops are placed
in the chamber for clear fluid and one to two drops for cloudy fluids (Fig. 10.5). The
sample chambers are capped, and the cytocentrifuge unit cover is closed (Fig. 10.6).
Specimens are spun at 800 rpm for 5 minutes.
Fig. 10.5 Thermo
Scientific™ Cytospin™
preparation technique
1. Combine into
a 50 mL tube
2. Concentrate
by Centrifugation
3. Pour Off
Supernatant
and Resuspend
Cell Pellet
4. Run on Shandon
Cytocentrifuge
228 D. K. Russell et al.
The program will begin when “Start” is selected, and the unit will chime when
complete. Once complete, the top of the cytocentrifuge is removed, the slide clip
unit opened, and the paper and slide are removed. The slide is placed in 95% alcohol
and is stained with the Papanicolaou stain (Fig. 10.7). (For exact details, refer to
manufacturer’s procedure manual.) [10].
10 Cytopreparatory Techniques 229
1. Combine into
50 mL tube
3. Run on the
BD TotalysTM
SlidePrep
10 Cytopreparatory Techniques 231
Fig. 10.11 High-grade
urothelial carcinoma. (BD
Totalys™ SurePath®
preparation, bladder
washing, Papanicolaou
stain, high magnification)
Most anatomic pathology laboratories prefer to perform ancillary tests on cell block
material, since cell block processing for ancillary studies is similar to standard his-
tology. Laboratories should have alternative validated protocols for pure cytology
samples. The standard staining methods used are hematoxylin and eosin (H&E) for
formalin-fixed, paraffin-embedded (FFPE) specimens. Cell blocks (CBs) have
many advantages and play a pivotal role in cytology [13, 14]. Currently, the utility
of CBs is recognized due to their role in ancillary testing, particularly in immuno-
chemistry, as well as capturing tissue fragments akin to biopsy material.
CB can be made from aggregation of exfoliated cells from fluid cytology. Cells
are resuspended in fluids by agitation immediately prior to processing the fluid to
ensure homogeneous distribution of the cell types throughout the specimen. The
architecture present in CB is more reminiscent of histology and is therefore familiar
to surgical pathologists (Fig. 10.13). Multiple sections for tests and controls can be
obtained from a single block. CB preparation methods vary among laboratories.
Evaluation in urine cytology can contribute to a final diagnosis in a range of cases
including benign, atypical, high-grade urothelial carcinoma and metastatic disease
[15]. Diagnostic architectural features can be observed in specimens that have a
large sediment when centrifuged. CB can demonstrate LGUC with fibrovascular
cores (Fig. 10.14).
In a study by Dantey et al., data revealed preparing CB for urine samples rou-
tinely did not improve the laboratory’s specimen adequacy rate. However, they can
assist to decrease the frequency of indeterminate diagnoses. Routine preparation of
CB, however, is expensive and provides minimal added clinical benefits, the authors
concluded [16].
Wilson et al. prepared CB cytology when a visible pellet was observed after
centrifugation [17]. Despite low urine volume in a number of cases in the study,
sufficient cellularity was available to produce diagnostic material. This recent study
234 D. K. Russell et al.
Fig. 10.13 High-grade
urothelial carcinoma. (Cell
block preparation, bladder
washing, hematoxylin &
eosin stain, high
magnification)
a b
Fig. 10.14 Low-grade urothelial carcinoma with fibrovascular cores. (a) Millipore® filter prepa-
ration, bladder washing, Papanicolaou stain, low magnification. (b) Cell block preparation, bladder
washing, hematoxylin & eosin stain, low magnification
Urine samples are poured into two 50 mL conical centrifuge tubes labeled with
patient identification. The tubes are placed in the centrifuge and spun at 2400 rpm
for 5 minutes to achieve a concentrated cell pellet. The supernatant is poured off. If
the cell pellet is not cohesive, formalin can be added, and the sample is spun at
2400 rpm for an additional 5 minutes. Allow the tube to sit for an additional
30–60 minutes to harden. The hardened button is removed with a metal spatula by
loosening the edges so it breaks free of the tube. The cell pellet is placed on histol-
ogy tissue paper moistened with neutral buffered formalin. Tissue paper is folded
around the cell pellet and placed in a histology cassette labeled with patient identi-
fication (Fig. 10.15). Once all steps are complete, the tissue cassette is placed in a
Fig. 10.15 Centrifugation
sediment cell block
preparation technique
CENTRIFUGE
10% FORMALIN
a b
Fig. 10.16 Metastatic adenocarcinoma. (a) Cell block preparation, bladder washing, hematoxylin
& eosin stain, high magnification. (b) Cell block preparation, bladder washing, Kreyberg mucin
histochemical stain, high magnification
formalin filled cup for additional time in formalin. The time is noted on the
requisition.
Time in formalin should be at least 6 hours of fixation for specific ancillary test-
ing IHC protocols [18, 19]. Quality control procedures documenting correct patient
identification and accession number, as well as any additional instructions, e.g.,
immunochemical protocol, may be incorporated in the cell block SOP. The speci-
men is then processed routinely in the histopathology laboratory and stained with
H&E stain according to the laboratory procedure for cell block preparation
(Appendix D). Immunostains can be ordered on the cell block material (Fig. 10.16).
If the cell pellet is extremely small after spinning the sample, HistoGel™/agar, col-
lodion, or Cellient® automated cell block preparation methods will conserve the
cellular material [20].
Fig. 10.17 Cellient®
automated cell block
preparation technique
1. Combine into
a 50 mL tube
10 min
2. Concentrate
by Centrifugation
3. Pour Off
Supernatant
and Resuspend
Cell Pellet
4. Add an Appropriate
Amount of Specimen
to the PreservCyt
Solution Vial
5. Run on Cellient®
Automated Cell
Block System
10 Cytopreparatory Techniques 239
Discussion
A cursory review of the scientific literature reveals no generally accepted best mate-
rials and methods of collecting and processing urine to detect urothelial malignan-
cies. Various citations come to different conclusions about the pros and cons of
urine that is voided spontaneously versus catheterized versus bladder washings
[22]. The number of urine specimens [23], whether urine should be collected fresh
or preserved [5], and whether to keep urine at room temperature or refrigerated are
debated pre-analytical steps [5, 22–24]. Additionally pondered is the question of
hours at these temperatures, the number of urine specimens needed to improve sen-
sitivity [24], which processing method is best [25–27], and the number of cells to
define adequacy [28, 29]. Some basic principles appear to have broad consensus and
are practiced by laboratories worldwide.
Conclusions and Recommendations
The urine collection and processing guidance in this chapter must be followed by
consistently good staining and mounting techniques for best results. Readers are
referred elsewhere for that guidance [1, 4, 15, 30, 31].
The Johns Hopkins Hospital Cytopreparatory Laboratory is named in honor and
memory of Gary Gill, cytotechnologist extraordinaire. Inscribed on the bronze
plaque at the entrance are words chosen by Gary: “Quality begins here!”
Title
Centrifugation Process
Purpose
Responsibilities
Approval and signatories for this document are determined in accordance with the
cytopathology laboratory.
Responsibilities specific to the process outlined in this document are listed below.
Role(s) Responsibilities
Cytopreparatory Technicians Follow procedure.
Cytotechnologists Follow procedure
Equipment
Procedure
1. After gently dispersing the specimen by shaking, it is poured into 50 mL centri-
fuge tubes. If there is not enough of the sample to fill two like centrifuge tubes
with equal volumes of fluid, a balance tube filled with water is used to balance
the centrifuge.
2. There are four separate holding chambers for the centrifuge tubes. Each chamber
is self-contained and covered by a plastic cover that is snapped into place. Each
chamber holds four 50 mL conical tubes.
3. Open the centrifuge lid, remove the plastic domes, and place specimens in to the
chamber being careful to properly balance the centrifuge
(a) To balance the centrifuge, place an identical volume tube into the directly
opposite chamber in the spot.
4. Close the lid and press the start button. The setting is preset at 2400RPM for
5 minutes.
5. When complete the centrifuge will alert that the process is done. The lid is
opened and the specimens are removed and can proceed with processing
Training
Role(s) Responsibilities
List role(s) of applicable staff Note which of the following are applicable: Read/review, Quiz (Knowledge Check),
and/or Skills Assessment.
Cytopreparatory Technicians Read/review
Cytotechnologists Read/review
10 Cytopreparatory Techniques 241
Revision History
Previous Version Revised Document
Reason for Revision
Number Version Date
Title
ThinPrep™ NonGynecologic Processing
Purpose
Scope
Responsibilities
Approval and signatories for this document are determined in accordance with the
cytopathology laboratory.
Responsibilities specific to the process outlined in this document are listed below.
Role(s) Responsibilities
Cytopreparatory Technicians Follow procedure.
Cytotechnologists Follow procedure
Procedure
Training
Role(s) Responsibilities
List role(s) of applicable staff Note which of the following are applicable: Read/review, Quiz
(Knowledge Check), and/or Skills Assessment.
Cytopreparatory Technicians Read/review
Cytotechnologists Read/review
Revision History
Previous Version Revised Document
Reason for Revision
Number Version Date
10 Cytopreparatory Techniques 243
Title
SurePath™ Preparation Using Totalys™ Multiprocessor for Non-Gynecologic Processing
Purpose
A SurePath preparation is a monolayer technology that uses sedimentation and sam-
ple enrichment techniques to produce a specimen for cytologic examination.
Totalys MultiProcessor can be used to automate the front-end processing with
chain of custody and error reconciliation prior to slide processing.
Scope
All cytology staff
Definitions
The specimens are from a variety of body sites including major body cavity fluids,
bronchoscopic specimens, urines, brushings and FNA rinses, etc.
Reagents
1. Density reagent
2. GYN preservative
3. Tris buffered saline working solution
4. Alcohol blend rinse
5. SurePath prep Stain
6. Cleaning alcohol (95% ETOH)
7. Deionized water
8. Xylene
9. 50/50 xylene/absolute ETOH mix
10. 100% ETOH
Equipment/instrumentation/consumables
1. SurePath vials
2. SlidePrep instrument
3. Multi-mix vortex
4. SurePath pre-coat microscopic slides
5. Slide and tube racks
6. Centrifuge tubes
7. Centrifuge
8. Settling chambers
9. Disposable transfer/aspirator tips
244 D. K. Russell et al.
Procedure
Note: All procedures involving the specimen should be performed underneath
the biohazard fume hood.
1. Access the specimen into the laboratory computer system to obtain a cytology
specimen number (see procedure on accessioning). Label the specimen con-
tainer with a printed 2-D Sato label.
2. It is important to ensure that the SurePath vial BDVA barcode is overla-
beled with the appropriate CoPath 2-D accession label.
3. Thoroughly mix fluid submitted by vigorously agitating bottle or specimen
container.
4. Any brush accompanying a specimen should be removed from the submit-
ted container and placed directly in the large opening of the SurePath vial.
The original container should be centrifuged with any pellet formed also
added to the same vial to ensure complete transfer.
5. Label a centrifuge tube with unique identifiers for the specimen using 1 inch
Sato labels.
6. Attempt to aliquot at least 50 ml of specimen into a 50 ml centrifuge tube when-
ever possible. For smaller volumes the entire specimen may be used.
7. In a balanced centrifuge, centrifuge the specimens on program 3 in the Jouan
centrifuge.
8. When finished. Remove specimens and decant supernatant into formalin discard
container.
9. Add 5 drops of centrifuged pellet (or in scant specimens all available) directly to
previously labeled SurePath vial using a transfer pipette.
Title
Cytocentrifuge Cell Block Standard Process
Purpose
The following procedure for preparation of cell blocks can be applied to any non-
gynecologic or gynecologic specimen received by the Cytopathology Department.
This procedure will be used most commonly on serous effusions, pelvic/abdominal
washes from the female genital tract, and fine needle aspirates. A cell block may be
performed on virtually any sediment that exists in a centrifuge tube after a specimen
has been spun down. Cell blocks are the preparation of choice for any specimen that
10 Cytopreparatory Techniques 245
Responsibilities
Approval and signatories for this document are determined in accordance with the
cytopathology laboratory. Responsibilities specific to the process outlined in this
document are listed below.
Role(s) Responsibilities
Cytotechnologist Follow policy.
Cytopreparatory technician Follow policy.
General Guidelines/Policy
Patient name and accession number should be verified and documented by two
people on the cell block submission form before submitting to Histology. An
accessioning sticker must be used to identify patient and Cytology accession num-
ber for each cassette.
All cell block material is fixed in 10% formalin (unless lymphoma protocol
is requested) for at least 6–72 hours.
246 D. K. Russell et al.
Training
Role(s) Responsibilities
List role(s) of applicable staff Note which of the following are applicable: Read/review, Quiz
(Knowledge Check), and/or Skills Assessment.
Cytotechnologist Read/review
Cytopreparatory technician Read/review
Revision History
Previous Version Revised Document
Reason for Revision
Number Version Date
References
1. Weidmann JE, Keebler CM, Facik MS. Cytopreparatory techniques. In: Bibbo M, Wilbur
DC, editors. Comprehensive cytopathology. 4th ed. Philadelphia: Elsevier Saunders; 2015.
p. 851–2.
2. Awad S, Allison SP, Lobo DN. The history of 0.9% saline. Clin Nutr. 2008;27:179–88.
3. Pomerat CM, Overman RR. Electrolytes and plasma expanders – I. Reaction of human cells in
perfusion chambers with phase contrast, time-lapse cine records. Z Zellforsch Mikrosk Anat.
1956;45:2–17.
4. Gill GW. Cytopreparation: principles & practice. In: Rosenthal DL, editor. Essentials in cyto-
pathology, vol. 12. New York: Springer; 2013.
5. Crabtree WN, Murphy WM. The value of ethanol as a fixative in urinary cytology. Acta Cytol.
1980;24:452–5.
6. Barkan GA, Tabatabai LZ, Kurtycz DFI, Padmanabhan V, Soures RJ, Nayar R, Sturgis
CD. Practice patterns in urinary cytopathology prior to the Paris System for reporting urinary
cytology. Arch Pathol Lab Med. 2020;144:172–6.
7. Kurtycz DFI, Brimo F, Rosenthal DL, Siddiqui MT, Tabatabai LZ, VandenBussche CJ, Wojcik
EM, Barkan GZ. Perceptions of Paris: An international survey in preparation for The Paris
System for Reporting Urinary Cytology 2.0 (TPS 2.0). J Am Soc Cytopathol. 2021;10:S4.
8. Hologic, Inc. ThinPrep® 2000 processor: Instructions for Use. 2017. https://www.hologic.
com/sites/default/files/package-insert/MAN-02624-001_005_002.pdf.
9. Hologic, Inc. ThinPrep® 5000 processor: Operator’s Manual. 2017. https://www.hologic.com/
sites/default/files/package-insert/MAN-02624-001_005_002.pdf.
10. Thermo Scientific™. CytoSpin® Preparation System: Instructions for Use. 2015. www.ther-
moscientific.com/pathology.
11. Becton Dickinson and Company (BD). BD Totalys™ Multiprocessor. 2016. https://www.
bd.com/en-u s/offerings/capabilities/cervical-c ancer-s creening/cytology-i nstruments/
totalys-system/totalys-multiprocessor.
12. Becton Dickinson and Company (BD). BD Totalys™ SlidePrep. 2016. https://www.bd.com/
en-us/offerings/capabilities/cervical-cancer-screening/cytology-instruments/totalys-system/
totalys-slideprep.
13. Nambirajan A, Jain D. Cell blocks in cytopathology: an update. Cytopathology.
2018;29(6):505–24.
14. Jain D, Mathur SR, Iyer VK. Cell blocks in cytopathology: a review of preparative methods,
utility in diagnosis and role in ancillary studies. Cytopathology. 2014;25(6):356–71.
10 Cytopreparatory Techniques 247
15. Chan E, Balassanian R, Tabatabai LZ, Lou H, Vohra P. Improved diagnostic precision of urine
cytology by implementation of the Paris System and the use of cell blocks. Cancer Cytopathol.
2018;126:809–16.
16. Dantey K, Pantanowitz L, Xing J, Cuda J, Nestler R, Monaco SE. Cell block preparation
in urine cytology: examination of utility and workflow in an academic practice. J Am Soc
Cytopathol. 2019;8:61–8.
17. Bennett WL, Russell DK, Evans SK, Agrawal T. Cell blocks in urine cytopathology: do they
add value to the diagnosis? A pilot study. J Am Soc Cytopathol. 2021;10:47–55.
18. Hammond ME, Hayes DF, Dowsett M, et al. American Society of Clinical Oncology/College of
American Pathologists guideline recommendations for immunohistochemical testing of estro-
gen and progesterone receptors in breast cancer. Arch Pathol Lab Med. 2010;134(6):907–22.
19. Wolff AC, Hammond ME, Hicks DG, et al. Recommendations for human epidermal
growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/
College of American Pathologists clinical practice guideline update. Arch Pathol Lab Med.
2014;138(2):241–56.
20. Hologic, Inc. Cellient® Automated Cell Block System: Operator’s Manual. 2015. https://
www.hologic.com/sites/default/files/package-insert/MAN-02078-001_003_02.pdf.
21. Thermo Scientific. Thermo Scientific Shandon Cytoblock Cell Block Preparation System
Instructions for Use. 2015. http://tools.thermofisher.com/content/sfs/manuals/231803%20
Cytoblok%20IFU.pdf/
22. Murphy WM. Current status of urinary cytology in the evaluation of bladder neoplasms. Hum
Pathol. 1990;21:886–96.
23. Murphy WM, Crabtree WN, Jukkola AF, Soloway MS. The diagnostic value of urine versus
bladder washing in patients with bladder cancer. J Urol. 1981;126:320–2.
24. Beyer-Boon ME, de Voogt HJ, van der Velde EA, Brussee JA, Schaberg A. The efficacy of
urinary cytology in the detection of urothelial tumours. Sensitivity and specificity of urinary
cytology. Urol Res. 1978;6:3–12.
25. Beyer-Boon ME, Voorn-den Hollander MJ. Cell yield obtained with various cytopreparatory
techniques for urinary cytology. Acta Cytol. 1978;22:589–93.
26. Beyer-Boon ME, van der Voorn-Den Hollander MJ, Arentz PW, Cornelisse CJ, Schaberg A,
Fox CH. Effect of various routine cytopreparatory techniques on normal urothelial cells and
their nuclei. Acta Pathol Microbiol Scand A. 1979;87:63–9.
27. Voss JS, Kipp BR, Krueger AK, Clayton AC, Halling KC, Karnes RJ, et al. Changes in
specimen preparation method may impact urine cytologic evaluation. Am J Clin Pathol.
2008;130:428–33.
28. Hundley AF, Maygarden S, Wu JM, Visco AG, Connolly A. Adequacy of urine cytology
specimens: an assessment of collection techniques. Int Urogynecol J Pelvic Floor Dysfunct.
2007;18:997–1001.
29. Prather J, Arville B, Chatt G, Pambuccian SE, Wojcik EM, Quek ML, Barkan GA. Evidence-
based adequacy criteria for urinary bladder barbotage cytology. J Am Soc Cytol. 2015;4:57–62.
30. Gill GW. Chapter 28: The laboratory. In: DeMay RM, editor. The Art & science of cytopathol-
ogy, vol. 3. 2nd ed. Chicago: ASCP Press; 2011. p. 1539–92.
31. Gill GW. Chapter 6: Fixation and specimen processing. In: Gupta PK, Baloch ZW, edi-
tors. Cytohistology of small tissue samples. New York: Cambridge University Press; 2011.
p. 148–61.
Risk of High-Grade Malignancy (ROHM)
11
Mauro Saieg, Güliz A. Barkan, Fadi Brimo,
Ashish Chandra, Tarik M. Elsheikh, Ricardo G. Pastorello,
Marcus L. Quek, Jianyu Rao, Momin T Siddiqui,
Z. Laura Tabatabai, Christopher J. VandenBussche,
and Philippe Vielh
Background
Since its first edition in 2016, The Paris System for Reporting Urinary Cytology
(TPS) [1] has been adopted worldwide in hospitals and laboratories for the cytologi-
cal diagnosis of urothelial malignancies. This has resulted in a large number of
publications that reported varied experiences. Besides providing clarity of commu-
nication, one of the major goals of a standardized reporting system is to yield an
estimated risk of malignancy (ROM) for each diagnostic category in order to prop-
erly guide patient management. Since the main goal of urine cytology is the detec-
tion of clinically significant cancers, particularly high-grade urothelial carcinoma
(HGUC), it seems more appropriate to utilize the term “risk of high-grade malig-
nancy (ROHM)” rather than ROM. Low-grade urothelial neoplasms (LGUN), on
the other hand, by definition have minimal cytologic atypia, and accurate diagnosis
of these lesions by cytomorphology alone is practically impossible. Thus, this sec-
ond TPS edition incorporates the term “ROHM,” rather than “ROM.”
A literature search yielded 11 studies with more than 100 samples each and avail-
able results on subsequent biopsy and/or resection for patients with a previous cyto-
logical diagnosis using TPS. These data were included to estimate ROHM for each
category of TPS (Table 11.1). Overall, a total of 2292 cases were available. The
ROHM ranged from 8.7% to 36.8% for NHGUC, 12.3% to 60.9%% for AUC,
33.3% to 100% for SHGUC, and 58.8% to 100% for HGUC. Mean ROHM weighted
by sample size (and standard deviations) from all studies was calculated at 15.7%
(± 7.8%), 38.5% (± 14.3%), 76.2% (± 17.2%), and 88.8% (± 12.7%) for NHGUC,
AUC, SHGUC, and HGUC, respectively. It is important to note that the majority of
these studies retrospectively applied TPS classification to cases that were signed out
prior to the introduction of TPS (Table 11.1, “reclassified pre-TPS”). Hence, these
Table 11.1 Distribution of risks of high-grade malignancy within each diagnostic category after
the implementation of TPS across included 11 studies
Study Number of cases with ROHM (%)
Author, year cohort available histology NHGUC AUC SHGUC HGUC
Stanzione et al., P 294 14.4 44.3a 88.9 97.4
2020 [3]
de Paula et al., P 499 11 32.6 80 92.3
2019 [4]
Bakkar et al., 2019 R 100 29.7 60.9 100 100
[6]
Wang et al., 2018 P 167 17.7 50 76.4 89.1
[7]
Meilleroux et al., P 229 8.7 49 87 91
2018 [2]
Xing et al., 2018 P 150 27 59 93 100
[8]
Rohilla et al., 2018 R 244 11.6 12.3 33.3 58.8
[9]
Zare et al., 2018 R 194 9.9 17.4 72 96.3
[10]
Suh et al., 2017 R 142 13.5 38.1 59.1 68
[11]
Granados et al., R 149 36.8 40 75 87.1
2017 [12]
Hassan et al., 2016 R 124 17.8 53.1 83.3 100
[13]
TPS The Paris System for Reporting Urine Cytology, ROHM risk of high-grade malignancy,
NHGUC negative for high-grade urothelial carcinoma, AUC atypical urothelial cells, SHGUC sus-
picious for high-grade urothelial carcinoma, HGUC high-grade urothelial carcinoma, R reclassi-
fied pre-TPS, P post-TPS
a
AUC”, “AUC, NOS/reactive,” and “AUC, favor LG” grouped together
publications might suffer from review bias, as pathologists may be more inclined to
upgrade indeterminate diagnoses in a research setting when there is no associated
clinical implication. Accordingly, when ROHM was calculated separately for pre-
and post-TPS studies, mean ROHM was significantly lower for the AUC and
SHGUC categories in the retrospective (pre-TPS) studies (32% and 65%, respec-
tively) versus prospective (post-TPS) studies (43 and 84%, respectively). Another
potential bias in ROHM may be attributed to differences in age and gender. However,
data regarding these possible differences is not extractable from the available stud-
ies, and different numbers by age and gender should be incorporated in future
reports for further refinement of the ROHM of the distinct categories in the third
version of TPS. In addition, all of these studies used histological follow-up to esti-
mate ROHM, which may overestimate ROHM due to selection bias. Therefore, the
actual ROHMs should be expected to be in the mid-range of what has been pub-
lished in the literature. Nonetheless, most studies have shown progressively higher
ROHM across the diagnostic categories. Within the three studies with the largest
number of patients that had histological follow-up, ROHMs for the HGUC were
11 Risk of High-Grade Malignancy (ROHM) 251
above 90%, emphasizing the high positive predictive value of TPS [2–4]. Indeed, it
would be important for future studies to have a standardized way to determine
ROHM, making them easier to compare and reflecting a final mean number closer
to actual clinical practice. Of interest, a recent study proposed the use of likelihood
ratios (LR) and pre-test probability (based on clinical and epidemiological param-
eters) to assess a more personalized post-test probability of having malignancy. This
would minimize the bias in number when comparing multiple studies, but is depen-
dent upon the clinical parameters pre-test, not always available [5].
Data on ROHM associated with a nondiagnostic/unsatisfactory (ND/U) urine
cytology is scant. As these cases represent a minority of specimens, their estimated
risks may be particularly prone to overestimation in limited cohorts [8]. In addition,
several authors have opted to exclude them from their analyses. In these few efforts
that reported ND/U cases with histological follow-up, the associated ROHM ranged
from 0 to 15.6% [2, 4, 14]. Of particular interest, a recent, larger study detailed
outcomes of 116 ND/U specimens from 106 patients, including histological and
cytological follow-up, with a NPV of 84.4% [14].
Among cases classified as atypical (AUC), a decrease in the use of this category
was noted in the majority of reports following implementation of TPS [2, 3, 6, 7,
15–18]. This was likely due to the clearly defined cytomorphological features of
atypia by TPS and inclusion of reactive conditions such as previous radiation, BCG
therapy, lithiasis, or manipulation within the NHGUC category [1, 19]. The major-
ity of the studies have also shown an increased ROHM for AUC category post-TPS,
reflecting a more meticulous use of this term, based on the clear recommendations
by the first edition of TPS [2, 3, 6–8, 13].
Table 11.2 Distribution of TPS categories initially assigned to cases ultimately reported as LGUN
on histology
Number of TPS categories % (n)
cases with
histologic
Study diagnosis of
Author, year cohort LGUN NHGUC AUC SHGUC HGUC LGUN
Stanzione et al., P 58 41.4 (24) 56.9 (33)a 1.7 (1) 0 (0) a
2020 [3]
Anbardar et al., R 22 4.5 (1) 18.2 (4) 31.8 (7) 40.9 (9) 4.5 (1)
2020 [21]
de Paula et al., P 88 93.2 (82) 4.5 (4) 0 (0) 2.3 (2) 0 (0)
2019 [4]
Vallamreddy R 32 18.8 (6) 21.9 (7) 6.3 (2) 0 (0) 53.1 (17)
et al., 2019 [17]
Rai et al., 2019 R 16 37.5 (6) 12.5 (2) 25 (4) 0 (0) 25 (4)
[22]
Wang et al., P 46 69.6 (32) 23.9 (11) 4.3 (2) 2.2 (1) b
2018 [23]
Meilleroux P 62 72.6 (45) 16.1 (10) 1.6 (1) 0 (0) 8.1 (5)
et al., 2018 [2]
Xing et al., P 27 85.2 (23) 14.8 (4) 0 (0) 0 (0) 0 (0)
2018 [8]
Zare et al., 2018 R 52 80.8 (42) 9.6 (5) 3.8 (2) 1.9 (1) 3.8 (2)
[10]
Roy et al., 2017 R 34 14.7 (5) 23.5 (8) 5.9 (2) 11.8 (4) 29.4 (10)
[20]
Granados et al., R 40 60 (24) 25 (10) 7.5 (3) 7.5 (3) 0 (0)
2017 [12]
Malviya et al., R 5 40 (2) 20 (1) 20 (1) 20 (1) 0 (0)
2017 [24]
Suh et al., 2017 R 27 33.3 (9) 11.1 (3) 37 (10) 18.5 (5) 0 (0)
[11]
Hassan et al., R 25 72 (18) 28 (7) 0 (0) 0 (0) 0 (0)
2016 [13]
TPS The Paris System for Reporting Urine Cytology, LGUN low-grade urothelial neoplasm, ND/U
nondiagnostic/unsatisfactory, NHGUC negative for high-grade urothelial carcinoma, AUC atypical
urothelial cells, SHGUC suspicious for high-grade urothelial carcinoma, HGUC high-grade uro-
thelial carcinoma, R reclassified pre-TPS, P post-TPS
a
“AUC”, “AUC, NOS/reactive,” and “AUC, favor LG” grouped together
b
The LGUN category was excluded from this study
11 Risk of High-Grade Malignancy (ROHM) 253
Several studies have reported diagnostic test measures for urinary cytology after
implementation of TPS. Overall sensitivity ranged from 40% to 84.7%, specificity
from 73% to 100%, PPV from 68% to 100%, and NPV from 46% to 88.2% [2, 4, 6,
9–12]. It is noteworthy that these studies can be heterogeneous in their methods. The
calculation of accuracy measures assumes that a test has only two possible results:
benign or malignant. However, urine cytology is significantly more complex, as it
includes additional diagnostic categories that are not simply “benign or malignant,”
such as ND, AUC, SHGUC, and LGUN. Consequently, different authors have used
distinct approaches on converting multiple reporting categories into a dichotomous
system for these analyses. Despite this limitation, most studies that report their expe-
rience before and after implementation of TPS exhibit an increased diagnostic sensi-
tivity for HGUC after putting TPS into practice [2, 6, 10–12]. This is a great advantage
as urinary cytology is primarily used for symptomatic screening and urothelial can-
cer surveillance. In addition, calculation of ROHM mostly relies on surgical pathol-
ogy interpretation, which in itself is not immune to bias, depending on multiple
factors. Also, it is important to remember that in any study reporting a cytology-his-
tology correlation, a verification bias exists; that is, not all cytology cases are fol-
lowed up by surgical procedures and there are other clinical factors that influence
management strategy.
Conclusion
Table 11.3 Estimated risk of high-grade malignancy (ROHM) for each category of The Paris
System for Reporting Urinary Cytology
TPS category Risk of high-grade malignancy
ND 0–16%
NHGUC 8–24%
LGUN 0–44%
AUC 24–53%
SHGUC 59–94%
HGUC/malignant 76–100%
254 M. Saieg et al.
References
1. Rosenthal DL, Wojcik EM. Kurtycz DFI. The Paris System for Reporting Urinary Cytology:
Springer; 2016.
2. Meilleroux J, Daniel G, Aziza J, d’Aure DM, Quintyn-Ranty ML, Basset CML, Evrard SM,
Courtade-Saidi MM. One year of experience using the Paris System for Reporting Urinary
Cytology. Cancer Cytopathol. 2018;126(6):430–6. PMID: 29663682.
3. Stanzione N, Ahmed T, Fung PC, Cai D, Lu DY, Sumida LC, Moatamed NA. The con-
tinual impact of the Paris System on urine cytology, a 3-year experience. Cytopathology.
2020;31(1):35–40. PMID: 31596979.
4. de Paula R, Oliveira A, Nunes W, Bovolim G, Domingos T, De Brot L, Bezerra S, Cunha I,
Morini M, Saieg M. Two-year study on the application of the Paris system for urinary cytology
in a cancer centre. Cytopathology. 2020;31(1):41–6. PMID: 31654587.
5. Myles N, Auger M, Kanber Y, Caglar D, Kassouf W, Brimo F. Evidence-based diagnos-
tic accuracy measurement in urine cytology using likelihhod ratios. J Am Soc Cytopathol.
2021;10:71–8.
6. Bakkar R, Mirocha J, Fan X, Frishberg DP, de Peralta-Venturina M, Zhai J, Bose S. Impact
of the Paris system for reporting urine cytopathology on predictive values of the equivocal
diagnostic categories and interobserver agreement. Cytojournal. 2019;16(21)
7. Wang Y, Auger M, Kanber Y, Caglar D, Brimo F. Implementing The Paris System for
Reporting Urinary Cytology results in a decrease in the rate of the “atypical” category
and an increase in its prediction of subsequent high-grade urothelial carcinoma. Cancer
Cytopathol. 2018; PMID: 29278461.
8. Xing J, Monaco S, Pantanowitz L. The ability of the Paris system to stratify the risk of high-
grade urothelial carcinoma. J Am Soc Cytopathol [Internet]. Elsevier Inc; 2018;7(5):S43.
Available from: https://doi.org/10.1016/j.jasc.2018.06.102.
9. Rohilla M, Singh P, Rajwanshi A, Gupta N, Srinivasan R, Dey P, Kakkar N. Cytohistological
correlation of urine cytology in a tertiary centre with application of the Paris system.
Cytopathology. 2018;29(5):436–43. PMID: 29920811.
10. Zare S, Mirsadraei L, Reisian N, Liao X, Roma A, Shabaik A, Hasteh F. A single institutional
experience with the Paris system for reporting urinary cytology: correlation of cytology and
histology in 194 cases. Am J Clin Pathol. 2018;150(2):162–7. PMID: 29878037.
11. Suh J, Go H, Sung C, Baek S, Hwang H, Jeong S, Cho Y. Modification of The Paris System
for urinary tract washing specimens using diagnostic cytological features. Cytopathology.
2017;28(6):516–23. PMID: 28816366.
12. Granados R, Duarte JA, Corrales T, Camarmo E, Bajo P. Applying the Paris system for report-
ing urine cytology increases the rate of atypical urothelial cells in benign cases: a need for
patient management recommendations. Acta Cytol. 2017;61(1):71–6. PMID: 27838683.
13. Hassan M, Solanki S, Kassouf W, Kanber Y, Caglar D, Auger M, Brimo F. Impact of imple-
menting the Paris system for reporting urine cytology in the performance of urine cytology.
Am J Clin Pathol. 2016;146(3):384–90. PMID: 27543983.
14. Abro S, Pambuccian S, Wojcik E, Barkan G. Outcome analysis and negative predictive value
of the “unsatisfactory/nondiagnostic” category of the Paris system for reporting urinary cytol-
ogy. J Am Soc Cytopathol. 2020;8(5):S15.
15. Vosoughi A, Ordobazari A, Lora Gonzalez MA, Guido LP, Skiba M, Campuzano-Zuluaga G,
Kryvenko ON, Gomez-Fernandez C, Garcia-Buitrago M, Jorda M. The Paris System “atypical
urothelial cells” category: can the current criteria be improved? J Am Soc Cytopathol. 2020;
16. Redman R, Zalaznick H, Mazzaferri EL, Massoll NA. The impact of assessing specimen
adequacy and number of needle passes for fine-needle aspiration biopsy of thyroid nod-
ules. Thyroid [Internet]. 2006;16(1):55–60. Available from: http://www.ncbi.nlm.nih.gov/
pubmed/16487014. PMID: 16487014.
11 Risk of High-Grade Malignancy (ROHM) 255
17. Vallamreddy SKR, Begam KV, Pratima J. Implementation of the Paris system versus institu-
tional diagnosis in the performance of urinary cytology: a 5 years correlative study of 74 cases.
IP Arch Cytol Histopathol Res. 2019;4(3):193–8.
18. VandenBussche CJ, Allison DB, Gupta M, Ali SZ, Rosenthal DL. A 20-year and
46,000-specimen journey to Paris reveals the influence of reporting systems and passive peer
feedback on pathologist practice patterns. Cancer Cytopathol. 2018;126(6):381–9. PMID:
29757495.
19. Wojcik EM. What should not be reported as atypia in urine cytology. J Am Soc Cytopathol
[Internet]. Elsevier Inc; 2015;4(1):30–36. Available from: https://doi.org/10.1016/j.
jasc.2014.08.001
20. Roy M, Kaushal S, Jain D, Seth A, Iyer VK, Mathur SR. An institutional experience with The
Paris System: a paradigm shift from ambiguous terminology to more objective criteria for
reporting urine cytology. Cytopathology. 2017;28(6):509–15. PMID: 28833848.
21. Anbardar MH, Monjazeb R. Reclassification of urinary cytology regarding The Paris System
for Reporting Urinary Cytology with cytohistological correlation demonstrates high sensi-
tivity for high-grade urothelial carcinoma. Diagn Cytopathol. 2020;48(5):446–52. PMID:
31976626.
22. Rai S, Lali BS, Venkataramana CG, Philipose CS, Rao R, Laxman Prabhu GG. A quest for
accuracy: evaluation of the Paris system in diagnosis of urothelial carcinomas. J Cytol. 2019;
23. Wang Y, Auger M, Kanber Y, Caglar D, Brimo F. Implementing The Paris System for
Reporting Urinary Cytology results in a decrease in the rate of the “atypical” category and an
increase in its prediction of subsequent high-grade urothelial carcinoma. Cancer Cytopathol.
2018;126(3):207–14. PMID: 29278461.
24. Malviya K, Fernandes G, Naik L, Kothari K, Agnihotri M. Utility of the Paris System in
Reporting Urine Cytology. Acta Cytol. 2017;61(2):145–52. PMID: 28380477.
Clinical Management
12
Marcus L. Quek, Trinity J. Bivalacqua, Ashish M. Kamat,
Mauro Saieg, Alexander I. Sankin, Yuji Tokuda,
and Bas WG van Rhijn
Background
The clinical scenarios in which urine cytology is performed include the evalua-
tion of hematuria and irritative voiding symptoms, initial work-up for a suspected
urothelial malignancy, and surveillance following treatment for urothelial cancer.
The decision to obtain a voided or instrumented (barbotage/washing/brushing)
cytology depends on the clinical setting. Since the introduction of TPS in 2016,
several publications have reported the risk of high-grade malignancy associated
with the various diagnostic categories (see Chap. 11). Herein, we review the clinical
management considerations for each of these diagnoses.
radiotherapy, chronic indwelling foreign body in the urinary tract, or family history
of urothelial carcinoma or Lynch syndrome. Depending on the risk category, the
diagnostic work-up might include upper tract imaging (renal ultrasonography, CT
or MR urography) and cystoscopy.
Urine cytology plays a critical role in the ongoing surveillance for urothelial
recurrences following therapy. While most bladder recurrences will often be
detected by routine surveillance cystoscopy, upper urinary tract and prostatic or
urethral recurrences can be more difficult to diagnose in a timely manner. Anterior
bladder wall and bladder neck lesions can occasionally be missed, while carcinoma
in situ may not always be readily distinguishable cystoscopically from other benign
inflammatory conditions. Thoughtful utilization of contrast-enhanced imaging
along with urine cytology represents the primary initial diagnostic modalities when
direct endoscopic visualization can be difficult or impractical.
The current AUA/Society of Urologic Oncology (SUO) Guidelines recommend
a risk-stratified approach in surveillance of non-muscle-invasive bladder cancer [2].
The risk categories are based on factors such as tumor grade, stage, multifocality,
size, and recurrence. Urine cytology or other urinary biomarkers are not recom-
mended for patients with a history of low-risk cancer and a normal cystoscopy.
Surveillance for intermediate-risk disease should include both cystoscopy and
cytology every 3–6 months for 2 years, then 6–12 months for 2 years, and then
annually thereafter. High-risk patients who have a negative first surveillance cystos-
copy should be followed with cystoscopy and cytology every 3–4 months for
2 years, then 6 months for 2 years, and then annually thereafter.
Urine cytology is also an essential component in the surveillance of patients who
undergo radical cystectomy and urinary diversion, as these patients remain at risk
for recurrences in the remnant urothelium (upper tracts and urethra) [6, 7]. Prior to
TPS, the diagnostic yield of cytology in the setting of a urinary diversion was vari-
able due to the interpretive challenges related to glandular and inflammatory cells
and necrotic debris or mucous from the bowel segment. While some diagnostic
challenges remain [8], the rates of “atypical” diagnoses in urinary diversion patients
should approximate the rates in those without an interposed bowel segment [6].
Voided specimens for those with orthotopic neobladders, catheterized specimens
from incontinent and continent cutaneous diversions, and urethral washings provide
important diagnostic information some times before lesions become radiographi-
cally or symptomatically evident. Risk factors for urethral and upper tract recur-
rences following radical cystectomy have previously been described [9–12]. A
NHGUC diagnosis is reassuring, and the patient may continue routine surveillance
at intervals commensurate with the risk of recurrence.
The management of “atypical urothelial cells (AUC)” has always presented a diag-
nostic dilemma for the urologist. Traditionally, this category has included a wide
spectrum of benign and malignant conditions lacking definitive cytologic criteria.
260 M. L. Quek et al.
The use of TPS has decreased the rate of AUC diagnoses, as known benign condi-
tions such as reactive/inflammatory changes and cellular changes associated with
polyomavirus and urolithiasis are now categorized as NHGUC. As a result, there
may be an increased risk of malignancy associated with the AUC diagnosis due to
the defined criteria for this category (see Chap. 11). The importance of communica-
tion between the cytopathologist and clinician cannot be emphasized enough in
relaying the degree of suspicion for an underlying malignancy.
From the standpoint of the urologist, the work-up for AUC should be individual-
ized based on the risk assessment of the patient. Those with hematuria or persistent
irritative voiding symptoms still require a thorough evaluation with upper tract
imaging and cystoscopy. A patient with known risk factors for urothelial carcinoma
and an atypical cytology should prompt consideration of performing further testing
to rule out malignancy. For patients with a prior history of urothelial malignancy,
the extent of the work-up is dependent on the clinical suspicion for recurrent dis-
ease. Evaluation of the upper urinary tract and urethra may be considered if not
recently performed. The utility of random bladder biopsies of “normal”-appearing
urothelium is controversial, but may identify unsuspected carcinoma in situ [13, 14].
The role of additional molecular testing, such as UroVysion FISH and other
urinary biomarkers, remains to be determined (see Chap. 9) [3, 4, 15]. Several cen-
ters have instituted reflex UroVysion FISH testing for adjudication of AUC diagno-
ses, whereby a positive FISH assay is managed similarly to a suspicious diagnosis
and a negative FISH test is followed expectantly [16–19]. Other urinary biomarker
assays (such as ImmunoCyt, Cxbladder) have also been studied as adjunctive tests
when an AUC diagnosis is encountered [20, 21]. Further study is needed to confirm
the utility of these and other biomarker tests in this setting.
lower urinary tract. Biopsies and formal transurethral resection of mucosal abnor-
malities are indicated. For those with negative upper tract imaging and cystoscopy,
selective cytologic sampling from both upper tracts may help further localize lesions
such as CIS which may be difficult to visualize radiographically. Enhanced direct
endoscopic visualization techniques, such as fluorescence (“Blue Light”) cystos-
copy with hexaminolevulinate or narrowband imaging, have demonstrated improved
ability to detect more subtle lesions compared with conventional white light cystos-
copy [22–28].
For the subset of patients being followed post-cystectomy and urinary diversion,
a finding of suspicious or positive for HGUC also warrants a thorough investigation.
Recurrences in the intestinal segment itself are extremely unlikely [29] .
Investigations should focus on the upper tracts as well as the urethra. Prostatic stro-
mal invasion and anterior vaginal wall involvement with urothelial carcinoma por-
tend a high risk for urethral recurrences following cystectomy, especially in those
diverted by means of a cutaneous diversion (ileal conduit or continence cutaneous
reservoir) [9, 10, 12]. Urethral wash cytologies are sometimes employed in the rou-
tine follow-up of these patients (if a prophylactic urethrectomy is not performed)
based on the specific risk factors (multifocal CIS, tumors at the bladder neck or
prostatic urethra). For those with orthotopic neobladders, a positive/suspicious
voided cytology may be further investigated with cystoscopy and biopsies as indi-
cated. Risk factors for upper tract recurrences have previously been reported [7, 11].
Patients with suspected upper tract recurrences following cystectomy and diversion
often require direct percutaneous access to the upper tract for antegrade ureterore-
noscopy since retrograde access through the ureteroenteric anastomosis may be
challenging. Management of lesions identified in the remnant urothelium can be
treated via endoscopic resection (antegrade/retrograde), instillation of topical intra-
cavitary (pyelocaliceal) immuno- or chemotherapeutics, and surgical resection
(nephroureterectomy, urethrectomy) depending on the clinical scenario [9, 30].
Histologic evidence of divergent differentiation (such as squamous, glandular,
micropapillary, plasmacytoid, and sarcomatoid) in urothelial carcinoma is often
associated with a more aggressive phenotype. Several studies have shown a lower
success rate of intravesical therapy for these tumors, and hence the option of “early”
radical cystectomy for non-muscle-invasive disease should be discussed with the
patient [31].
Conclusions
Since 2016, The Paris System has been widely adopted and now represents the
standard nomenclature for urine cytology reporting. As the end-users of the urine
cytology report, we applaud the international cytopathology community in stan-
dardizing the criteria and methodology and bringing uniformity to the diagnostic
reporting. The evaluation and management described above represent only general
recommendations to provide insight for the cytopathologist to understand how a
given cytologic diagnosis may impact clinical decision-making. The ultimate care
of the patient should be individualized based on all available clinical information.
Communication between the clinician and cytopathologist is essential in order to
optimally utilize this powerful diagnostic tool.
References
1. Kurtycz DFI, Barkan GA, Pavelec DM, Rosenthal DL, Wojcik EM, VandenBussche CJ,
Mangiulli K, Olson MT. Paris interobserver reproducibility study (PIRST). J Am Soc
Cytopathol. 2018;7:174–84.
2. Chang SS, Boorjian SA, Chou R, Clark PE, Daneshmand S, Konety BR, Pruthi R, Quale DZ,
Ritch CR, Seigne JD, Skinner EC, Smith ND, McKiernan JM. Diagnosis and treatment of non-
muscle invasive bladder cancer: AUA/SUO guideline. J Urol. 2016;196:1021–9.
3. Xylinas E, Kluth LA, Rieken M, Karakiewicz PI, Lotan Y, Shariat SF. Urine markers for detec-
tion and surveillance of bladder cancer. Urol Oncol. 2014;32:222–9.
264 M. L. Quek et al.
4. Yafi FA, Brimo F, Steinberg J, Aprikian AG, Tanguay S, Kassouf W. Prospective analysis of
sensitivity and specificity of urinary cytology and other urinary biomarkers for bladder cancer.
Urol Oncol. 2015;33:66.e25–31.
5. Barocas DA, Boorjian SA, Alvarez RD, Downs TM, Gross CP, Hamilton BD, Kobashi KC,
Lipman RR, Lotan Y, Ng CK, Nielsen ME, Peterson AC, Raman JD, Smith-Bindman R, Souter
LH. Microhematuria: AUA/SUFU Guideline. J Urol. 2020;204:778–86.
6. Chen H, Liu L, Pambuccian SE, Barkan GA, Quek ML, Wojcik EM. Urine cytology in moni-
toring recurrence in urothelial carcinoma after radical cystectomy and urinary diversion.
Cancer Cytopathol. 2016;124:273–8.
7. Picozzi S, Ricci C, Gaeta M, Ratti D, Macchi A, Casellato S, Bozzini G, Carmignani L. Upper
urinary tract recurrence following radical cystectomy for bladder cancer: a meta-analysis on
13,185 patients. J Urol. 2012;188:2046–54.
8. Nomani L, Abro S, Quek ML, Barkan GA. Guar bean in urinary cytology: a morphologic
pitfall. J Am Soc Cytopathol. 2021;10:41–6.
9. Mitra AP, Alemozaffar M, Harris BN, Schuckman AK, Skinner EC, Daneshmand S. Outcomes
after urothelial recurrence in bladder cancer patients undergoing radical cystectomy. Urology.
2014;84:1420–6.
10. Clark PE, Stein JP, Groshen SG, Miranda G, Cai J, Lieskovsky G, Skinner DG. The manage-
ment of urethral transitional cell carcinoma after radical cystectomy for invasive bladder can-
cer. J Urol. 2004;172:1342–7.
11. Sanderson KM, Cai J, Miranda G, Skinner DG, Stein JP. Upper tract urothelial recurrence fol-
lowing radical cystectomy for transitional cell carcinoma of the bladder: an analysis of 1,069
patients with 10-year followup. J Urol. 2007;177:2088–94.
12. Stein JP, Penson DF, Wu SD, Skinner DG. Pathological guidelines for orthotopic urinary diver-
sion in women with bladder cancer: a review of the literature. J Urol. 2007;178:756–60.
13. Otsuka M, Taguchi S, Nakagawa T, Morikawa T, Maekawa S, Miyakawa J, Matsumoto A,
Miyazaki H, Fujimura T, Fukuhara H, Kume H, Igawa Y, Homma Y. Clinical significance of
random bladder biopsy in primary T1 bladder cancer. Mol Clin Oncol. 2018;8:665–70.
14. Takamatsu K, Matsumoto K, Kikuchi E, Ogihara K, Hayakawa N, Tanaka N, Takeda T, Morita
S, Kosaka T, Mizuno R, Asanuma H, Mikami S, Oyama M, Oya M. Can random bladder biop-
sies be eliminated after bacillus Calmette-Guerin therapy against carcinoma in situ? Int Urol
Nephrol. 2021;53:465–9.
15. Kamat AM, Hegarty PK, Gee JR, Clark PE, Svatek RS, Hegarty N, Shariat SF, Xylinas
E, Schmitz-Drager BJ, Lotan Y, Jenkins LC, Droller M, van Rhijn BW, Karakiewicz PI,
International Consultation on Urologic Disease-European Association of Urology Consultation
on Bladder Cancer 2012. ICUD-EAU International Consultation on Bladder Cancer 2012:
screening, diagnosis, and molecular markers. Eur Urol. 2013;63:4–15.
16. Bubendorf L, Piaton E. UroVysion(R) multiprobe FISH in the triage of equivocal urinary
cytology cases. Ann Pathol. 2012;32:e52–6, 438–43.
17. Gayed BA, Seideman C, Lotan Y. Cost-effectiveness of fluorescence in situ hybridiza-
tion in patients with atypical cytology for the detection of urothelial carcinoma. J Urol.
2013;190:1181–6.
18. Skacel M, Fahmy M, Brainard JA, Pettay JD, Biscotti CV, Liou LS, Procop GW, Jones JS,
Ulchaker J, Zippe CD, Tubbs RR. Multitarget fluorescence in situ hybridization assay detects
transitional cell carcinoma in the majority of patients with bladder cancer and atypical or nega-
tive urine cytology. J Urol. 2003;169:2101–5.
19. Yoder BJ, Skacel M, Hedgepeth R, Babineau D, Ulchaker JC, Liou LS, Brainard JA, Biscotti
CV, Jones JS, Tubbs RR. Reflex UroVysion testing of bladder cancer surveillance patients with
equivocal or negative urine cytology: a prospective study with focus on the natural history of
anticipatory positive findings. Am J Clin Pathol. 2007;127:295–301.
20. Odisho AY, Berry AB, Ahmad AE, Cooperberg MR, Carroll PR, Konety BR. Reflex
ImmunoCyt testing for the diagnosis of bladder cancer in patients with atypical urine
cytology. Eur Urol. 2013;63:936–40.
12 Clinical Management 265
21. Konety B, Shore N, Kader AK, Porten S, Daneshmand S, Lough T, Lotan Y. Evaluation
of cxbladder and adjudication of atypical cytology and equivocal cystoscopy. Eur Urol.
2019;76:238–43.
22. Daneshmand S, Schuckman AK, Bochner BH, Cookson MS, Downs TM, Gomella LG,
Grossman HB, Kamat AM, Konety BR, Lee CT, Pohar KS, Pruthi RS, Resnick MJ, Smith
ND, Witjes JA, Schoenberg MP, Steinberg GD. Hexaminolevulinate blue-light cystoscopy in
non-muscle-invasive bladder cancer: review of the clinical evidence and consensus statement
on appropriate use in the USA. Nat Rev Urol. 2014;11:589–96.
23. Daneshmand S, Bazargani ST, Bivalacqua TJ, Holzbeierlein JM, Willard B, Taylor JM, Liao
JC, Pohar K, Tierney J, Konety B, Blue Light Cystoscopy with Cysview Registry Group. Blue
light cystoscopy for the diagnosis of bladder cancer: results from the US prospective multi-
center registry. Urol Oncol. 2018;36:361.e1–6.
24. Herr HW. Randomized trial of narrow-band versus white-light cystoscopy for restaging
(second-look) transurethral resection of bladder tumors. Eur Urol. 2015;67:605–8.
25. Lerner SP, Goh A. Novel endoscopic diagnosis for bladder cancer. Cancer. 2015;121:169–78.
26. Lotan Y, Bivalacqua TJ, Downs T, Huang W, Jones J, Kamat AM, Konety B, Malmstrom PU,
McKiernan J, O'Donnell M, Patel S, Pohar K, Resnick M, Sankin A, Smith A, Steinberg G,
Trabulsi E, Woods M, Daneshmand S. Blue light flexible cystoscopy with hexaminolevulinate
in non-muscle-invasive bladder cancer: review of the clinical evidence and consensus state-
ment on optimal use in the USA – update 2018. Nat Rev Urol. 2019;16:377–86.
27. Stenzl A, Burger M, Fradet Y, Mynderse LA, Soloway MS, Witjes JA, Kriegmair M, Karl A,
Shen Y, Grossman HB. Hexaminolevulinate guided fluorescence cystoscopy reduces recur-
rence in patients with nonmuscle invasive bladder cancer. J Urol. 2010;184:1907–13.
28. Witjes JA, Babjuk M, Gontero P, Jacqmin D, Karl A, Kruck S, Mariappan P, Palou
Redorta J, Stenzl A, van Velthoven R, Zaak D. Clinical and cost effectiveness of
hexaminolevulinate-guided blue-light cystoscopy: evidence review and updated expert rec-
ommendations. Eur Urol. 2014;66:863–71.
29. Doshi CP, Barkan GA, Quek ML. Urothelial carcinoma recurrence in an orthotopic neobladder
without urethral or upper urinary tract involvement. Case Rep Urol. 2019;2019:8458706.
30. Kleinmann N, Matin SF, Pierorazio PM, Gore JL, Shabsigh A, Hu B, Chamie K, Godoy G,
Hubosky S, Rivera M, O'Donnell M, Quek M, Raman JD, Knoedler JJ, Scherr D, Stern J,
Weight C, Weizer A, Woods M, Kaimakliotis H, Smith AB, Linehan J, Coleman J, Humphreys
MR, Pak R, Lifshitz D, Verni M, Adibi M, Amin MB, Seltzer E, Klein I, Konorty M, Strauss-
Ayali D, Hakim G, Schoenberg M, Lerner SP. Primary chemoablation of low-grade upper
tract urothelial carcinoma using UGN-101, a mitomycin-containing reverse thermal gel
(OLYMPUS): an open-label, single-arm, phase 3 trial. Lancet Oncol. 2020;21:776–85.
31. Willis D, Kamat AM. Nonurothelial bladder cancer and rare variant histologies. Hematol
Oncol Clin North Am. 2015;29:237–52.
32. Hall MC, Chang SS, Dalbagni G, Pruthi RS, Seigne JD, Skinner EC, Wolf JS Jr, Schellhammer
PF. Guideline for the management of nonmuscle invasive bladder cancer (stages Ta, T1, and
Tis): 2007 update. J Urol. 2007;178:2314–30.
33. Babjuk M, Burger M, Zigeuner R, Shariat SF, van Rhijn BW, Comperat E, Sylvester RJ,
Kaasinen E, Bohle A, Palou Redorta J, Roupret M, European Association of Urology. EAU
guidelines on non-muscle-invasive urothelial carcinoma of the bladder: update 2013. Eur Urol.
2013;64:639–53.
34. Kamat AM, Witjes JA, Brausi M, Soloway M, Lamm D, Persad R, Buckley R, Bohle A,
Colombel M, Palou J. Defining and treating the spectrum of intermediate risk nonmuscle inva-
sive bladder cancer. J Urol. 2014;192:305–15.
35. Quek ML, Nichols PW, Yamzon J, Daneshmand S, Miranda G, Cai J, Groshen S, Stein JP,
Skinner DG. Radical cystectomy for primary neuroendocrine tumors of the bladder: the uni-
versity of southern California experience. J Urol. 2005;174:93–6.
266 M. L. Quek et al.
36. Siefker-Radtke AO, Dinney CP, Abrahams NA, Moran C, Shen Y, Pisters LL, Grossman HB,
Swanson DA, Millikan RE. Evidence supporting preoperative chemotherapy for small cell
carcinoma of the bladder: a retrospective review of the M. D. Anderson cancer experience. J
Urol. 2004;172:481–4.
37. Meijer RP, Meinhardt W, van der Poel HG, van Rhijn BW, Kerst JM, Pos FJ, Horenblas S, Bex
A. Local control rate and prognosis after sequential chemoradiation for small cell carcinoma
of the bladder. Int J Urol. 2013;20:778–84.
The History of Urinary Cytology:
The Long and Winding Road to Paris 2.0 13
Stefan E. Pambuccian
History has a good way to tell us, not only where we have been but also where we
are going, as William Osler [1] liked to quote the English churchman and historian
Thomas Fuller (1608–1661) [2]:
History maketh a young man to be old, without either wrinkles or gray hair; privileging him
with the experience of age, without the infirmities or inconveniences thereof. Yea, it not
only maketh things past, present, but enableth one to make rational conjecture of things to
come …. Old actions return again, furbished over with some new and different
circumstances.
Ancient History
Urine has had many uses in history. In Ancient Rome, urine was used to launder the
emblematic Roman togas, to fertilize fields and grow juicier pomegranates, and
even to whiten the teeth. Other nonmedical uses encountered through history
included the preparation of cosmetics, leather softening, textile dyeing, and the
preparation of gunpowder.1 Medical uses of human and animal urine as an ingredi-
ent of various remedies are found in all ancient cultures, starting with the Hindu
culture, to the Greco-Roman world, and even persisted into the late Middle Ages
and early Enlightenment, when urine was still recommended by some of the most
famous medical personalities. They included the French surgeon Ambroise Paré
(1510–1590), the English philosopher and chemist Robert Boyle (1627–1691), and
English physician Thomas Willis (1621–1675), who recommended urine to disin-
fect wounds and to treat cuts and burns and various external and internal ailments
1
Gunpowder to Teeth Whitener: The Science behind Historic Uses of Urine M Kumar - Smithsonian
Magazine (smithsonian. com), July–August. 2013 accession 3.1.2021
https://www.smithsonianmag.com/search/?q=Gunpowder%20to%20Teeth%20Whitener:%20
The%20Science%20behind%20Historic%20Uses%20of%20Urine
[3, 4]. While we may be surprised, shocked, repulsed, or disgusted by these and
other historical scatological therapeutic practices [5], we cannot be but fascinated
by the diagnostic use of urine through the ages. In many areas it is still a common
but not medically recommended practice to urinate on a sea urchin sting, jellyfish
sting, or a stingray wound.2 Through five millennia, witnessing the rise and fall of
empires, the examination of urine was considered one of the most informative diag-
nostic tools. Urine, after all, is an easily accessible fluid. The examination of urine
reached its apogee in the middle ages, when uroscopy was essentially the diagnosis.
The practice of uroscopy in the Middle Ages has inspired numerous manuscript
illuminations, paintings, and other works of art [6] and was considered the symbol
of the physician and its patron saint, Saint Cosmas [7] (Figs. 13.1, 13.2, 13.3,
and 13.4).
All major cultures left written and archeological evidence of the diagnostic value
of the examination of urine. Sickness and death have been plaguing humankind
since its earliest days, and humans have always struggled to understand and cure or
alleviate disease. Specialized “healers” have probably existed since the dawn of
civilization. The writings from Mesopotamia, Egypt, China, India, and Ancient
Greece all attest to the presence of healers. Depending on the culture, such healers
had different concepts of disease and employed a combination of magical/religious
and empiric/rational approaches, the relative importance of which varied in cultures
and different times. The oversimplified concept of a magical/religious healer, priest,
or exorcist of the Mesopotamian and Egyptian cultures changes abruptly in Ancient
Greece to a rational physician practicing Hippocratic medicine. In all these cultures,
just like today, sick people have sought help in multiple ways, starting from self-
help, prayer, and seeking the advice of friends and family and ultimately moving on
a b c
Fig. 13.1 Stained glass showing uroscopy: (a) Chartres Cathedral, thirteenth-century France.
Emperor Constantine, who has leprosy, consults a physician. (b, c) The monkey as a doctor exam-
ining a vial of urine. Detail from the Pilgrimage window. York Minster, fourteenth century
https://www.scientificamerican.com/article/fact-or-fiction-urinating/
2
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 269
Fig. 13.2 Left: A surgeon letting blood from a woman’s arm and a physician examining a urine
flask. Oil painting by a Flemish painter, 18th (?) century. Credit: Wellcome Collection. Attribution
4.0 International (Creative Commons (CC) BY 4.0). Right: doctor inspecting a urine flask in his
“home office.” Notice the wicker basket in which the urine was received. Evert Oudendijck, Dutch
painter (1648–1695)
Fig. 13.3 St. Cosmas and St. Damian (the patron saints of surgeons, physicians, and pharmacists).
St Cosmas (left) is frequently depicted holding a matula (urine flask). Woodcut attributed to Hans
Wechtlin from the 1517 book Feldbuch der Wundarzney of Hans von Gersdorff (1455–1529). The
framed image depicts Saint Cosmas and Saint Damian in an oil painting. Credit: Wellcome
Collection, London, UK. Attribution 4.0 International (Creative Commons (CC) BY 4.0)
270 S. E. Pambuccian
Fig. 13.4 Illuminations from medieval manuscripts showing physicians performing uroscopy. (a)
Mystère de la Vengeance by Eustache Marcadé, Bruges, c. 1465 (London, British Library). (b) De
proprietatibus rerum by Bartholomaeus Anglicus, fifteenth century, France (Biblioteque Nationale
Francaise)
Fig. 13.5 Map of the countries surrounding the Mediterranean Sea which played a role in the
spread of the knowledge about urine diagnosis
Egypt and Kairouan (currently Tunisia), and further westwards to the Iberian
Peninsula, to Cordoba (Al-Andalus), where they were included in the works of
Maimonides [24, 25] (Fig. 13.5).
In the following centuries, the Arabic texts on uroscopy authored by these physi-
cians were translated into Greek and Latin in Constantinople [26, 27] and Salerno
[28–31] in Southern Italy, from where uroscopy diffused to the major European
medical schools of Montpellier and Paris. Throughout this journey, the “judgement
of urines” became more and more complex, the number of colors of urine described
became more numerous, and the relative importance of uroscopy in the diagnostic
process increased. As the number of urine colors increased, naming them became
insufficient, and hand-painted urine wheels, depicting the various nuances of urine
color, were included in uroscopy manuscripts and the first printed uroscopy manu-
als. These urine wheels were considered so important that they were included in
foldable vade mecums (Latin, “go with me”) or pocket references that were placed
in bags or pouches hanging from the belts of the physicians of the time [32]. The
ownership of a urine wheel and the ability to interpret it constituted the closest
approximation to a medical license at the time (Fig. 13.6).
In the Middle Ages, when monastic medicine, societal norms, and concepts of
“modesty” made physical examination of patients, and in particular of women, dif-
ficult or impossible, the objective diagnostic methods were reduced to taking the
pulse and looking at the urine. Doctors educated at one of the rare universities of the
Middle Ages were few, were sometimes related to monasteries, and ordinarily did
not make house calls. The habit therefore ensued to send off urine in a rounded-
bottom flask that mimicked the shape of the bladder, called matula, which was
transported in a specially made basket. This practice, probably the first example of
274 S. E. Pambuccian
Fig. 13.6 A portable physician’s folding almanac from the fifteenth century containing tables use-
ful for field practice, including a table of urine colors, next to zodiacal tables with moon and sun
phases and zodiacal-humoral correspondences, and bloodletting points (British Library, Harley
MS 5311). Hand-colored urine wheel from Fasciculo de medicina, attributed to Johannes de
Ketham, a professor of medicine in Vienna about 1460, and published in Venice in 1493 by Zuane
& Gregorio di Gregorii. This is the first illustrated medical book, a compendium of several differ-
ent, independently authored medical treatises covering a wide spectrum of the medical knowledge
of the time, including uroscopy, bloodletting, wound treatment, anatomical dissection, and wom-
en’s health
Fig. 13.7 Depiction of the matula after Joannes Zacharias Actuarius (1275–1328) [34]. Note that
the regions of the matula are corresponding to the parts of the human body. Regions 1–4, numbered
from the fundus of the matula, correspond to the belly and the urinary tract and regions 3–8 to the
thorax and 10–11 to the head. Human-shaped vessel to distill the urine (“Vase Distillatorio ad
Urinam”) made of “white venetian blown glass, substantially transparent,” known as anatomical
alembic or anatomic furnace (“Furnace Anatomica”), from the “Aurora Thesaurusque
Philosophorum, Theophrasti Paracelsi,” published in Basel in 1577, a treatise attributed to the
Belgian alchemist Gerard Dorn (1530–1584)
Galenism. Teaching in his native German rather than the customary Latin of his
times, he told his students that his shoe buckles contained more wisdom than Galen
and Avicenna [39]. Paracelsus refuted the diagnosis by uroscopy and pulse taking
practiced by the physicians of his times as impostures, commenting “their ignorance
cannot justify their fantastic theories. All they can do is gaze at piss.” In fact, prac-
titioners of uromancy acquired the appellation “Pisse Prophets.” Paracelsus none-
theless continued examining the urine of patients, through what he called chemical
“dissection” of the urine. He used alchemical extraction, coagulation, and distilla-
tion of the urine as diagnostic methods. Using such methods, he and his followers
would determine if the patient had bladder or kidney stones and what kind of stones
they were. They demonstrated albuminuria by precipitation of proteins when acid
276 S. E. Pambuccian
(vinegar) was added to the urine [40, 41] and determined the weight (specific grav-
ity) of urine. They also introduced the quantification of substances it contained [42],
the fractionated distillation of urine, and the observation of the colors of the flames
after burning of different distillation residues.
One of his followers, the Swiss Leonard Thurneysser (1531–1595), used the
“anatomical furnace” to have an even more detailed analysis of the urine by frac-
tionated distillation at different temperatures to determine the presence and quantity
of elements of sulfur, salt, and mercury in urine. For a time, he had a very successful
practice and established the first “telemedicine” center in Berlin, where he was
examining patients’ urines sent from all over Europe and sent them back the pre-
scribed medications (Fig. 13.8).
Another alchemist, Hennig Brand (1630–1692) of Hamburg, studying urine,
accidentally discovered a chemical element, which he named phosphorus (Greek
for “light-bearer”) because the substance that he isolated from the heated resi-
dues from boiled-down urine emitted a pale-green glow. Despite the arcane,
strange, and complex theoretical underpinnings, the alchemical understanding of
disease was a step forward and laid the foundations of iatrochemistry, which
evolved into clinical chemistry. By the middle of the nineteenth century, the pres-
ence and quantity of most of the normal chemical constituents of urine were
known, and chemical methods of examining urine started to be routinely used in
the clinical diagnosis.
Fig. 13.8 Stained-glass window panel made by Christoph Murer in1579, showing Leonhard
Thurneysser zum Thurn (1531–1596) testing urine for a potentate in the Middle East (Kunstmuseum
Basel, Isobel Leybold-Johnson in Basel) http://www.swissinfo.ch/eng/culture/basel-s-16th-century-
superstar/28863378
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 277
The excesses of uroscopy and the fact that it was based on the Galenic theory of
humors can lead to the impression that the Late Antiquity, Byzantine, Islamic, and
Medieval Latin Western civilizations were a period of stagnation of medicine. However,
this period should not be perceived only as a relay station for the transmission of the
wisdom of Ancient Greek medicine. The authors of this period, while striving to main-
tain tradition, added new empirical data and practical therapeutic measures that lead to
improvements in medical and surgical care. Even while they claimed to follow Galen’s
theory, in their comments and explanations, they made subtle changes in its interpreta-
tion [28] and established a rational, almost algorithmic system of diagnostic decision-
making. This elaborate diagnostic system could be conveyed by formal instruction,
through textbooks and taught in the classroom [18], and could even be written in verse,
for ease of memorization [43]. Uroscopy also arguably constituted the first laboratory
test and highlighted the importance of observation. In today’s extensively technolo-
gized laboratory, it reminds us that sometimes simple observations, such as the change
in the color of urine, may have very important diagnostic connotations. The color of
urine, its viscosity and specific gravity, its clarity, and the presence of precipitates and
floating particles, all features that were assessed by uroscopy, are still diagnostically
useful. Modern-day urine wheels can be devised to depict the significance of urine
color for diagnosis [44], and observing the color of urine in the bag of catheterized
patients can still give important diagnostic clues [45] (Fig. 13.9).
The sixteenth century saw not only the birth of medical alchemy, which morphed
into iatrochemistry (or chemiatry), which in turn gave rise to clinical chemistry and
biochemistry, but also the invention of a device that was destined to change the
practice of medicine completely, the microscope. Towards the end of the sixteenth
century, simple compound microscopes were made by combining lenses and put-
ting them at the ends of a telescoping tube, by the Dutch spectacle makers, Hans
Janssen and his son Zaccharias. Such very primitive devices had 9x magnification,
and the blurry images were mostly used for the entertainment of nobility. In time,
bigger and better instruments were built, and in the early seventeenth century, lay-
men and scientists used them to extend the reach of visual observation of all kind of
objects: plants, insects, and human tissues and body fluids. Despite the difficulties
that must have been encountered in examining a transparent liquid under the primi-
tive microscopes of the time, urine was one of the first body fluids to be subjected
to microscopic analysis. Initially the microscopic examination of urine was proba-
bly done mostly out of curiosity, and outside the clinical context, it likely had little
impact on the medical field. It is worth mentioning that the microscopic examina-
tion of urine was not mentioned by some of the most prominent microscopists of the
time, who were interested in the urinary tract, including Marcello Malpighi
(1628–1694) and Lorenzo Bellini (1643–1704).
The first to describe the microscopic appearance of crystals (most likely uric acid
or oxalate crystals) in the urinary sediment as “a heap of rhomboidal bricks” was the
Provençal polymath Nicolas-Claude Fabri de Peiresc (1580–1637), according to his
biographer [46]. In 1630 he examined “sandy urine” and thought that he found the
cause of urinary calculi. A few decades later, in 1665, the English physicist Robert
Hooke (1635–1703) collated his microscopic observations in a beautifully illus-
trated tome “Micrographia.” In addition to coining the name “cells” to the spaces he
saw looking at a sliver of cork through his brass microscope, he also illustrated
crystals seen in the urine sediment and crystals formed by freezing urine [47]
(Fig. 13.10).
Starting around 1673, the Dutch draper (cloth retailer) Anton van Leeuwenhoek
(1632–1723) started reporting to the scientific community of the Royal Society of
London his observations made with his simple single-lens, yet powerful micro-
scopes capable of magnifications of up to 200–370 times. The metal construction of
his microscopes made the examination of fluids difficult, but he did report observa-
tions on urine crystals and other body fluids [48] (Fig. 13.11).
The first physician to perform microscopic studies on urine was probably the
Danish physician and member of the Leopoldina Academy Jørgen Hermansen Hahn
of Copenhagen (1647–1699). Better known under his Latinized name Georgius
Hannaeus, he published in 1687 an account of the microscopic appearance of “sand”
from urinary calculi, obtained after drying the urine sediment on paper and obtain-
ing a “farinaceous” powder. Also interested in urinary stones, the Leyden physician
and pioneer of bedside teaching, Hermann Boerhaave (1668–1738), studied the
urine of normal people to determine if they were predisposed to urolithiasis [49].
Primitive microscopes were expensive, were technically difficult to use, and pro-
duced poor images that could easily lead to misinterpretations. Their use was not
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 279
Fig. 13.10 Hooke’s 1665 depiction of urinary sediment crystals and crystals formed by freezing
urine. From “Micrographia: or some physiological descriptions of minute bodies made by magni-
fying glasses with observations and Inquiries thereupon” by Robert Hooke, Fellow of the Royal
Society, 1665. (Credit: Engraving from Micrographia, 1665, by Robert Hooke. Credit: Project
Gutenberg License)
accepted by the medical community until more and more sophisticated and power-
ful microscopes were built, assuring physicians that the objects seen under the
microscope were “real” and not artifacts.
In the meantime, the eighteenth century, which was relatively quiet in regard to
microscopy, saw a dramatic change in the way disease was conceived. Humoral
medicine still conceived all diseases as involving the whole body as imbalances
between humors, the visible manifestations reflecting the local impact of this imbal-
ance. As indicated above, these concepts changed due to the publication in 1761 of
the results of 640 autopsies with clinicopathologic correlations (historiae anatomo-
medicae) by the 79-year-old Giovanni Battista Morgagni (1682–1771) [8, 50]. In
his influential work, The Seats and Causes of Diseases, the professor of anatomy at
the University of Padua proposed that most diseases do not diffusely involve the
whole body, but originate and reside locally in its organs. Despite the fact that
Morgagni was a pupil of Antonio Maria Valsalva (1666–1723), who had been men-
tored by the great microscopy pioneer Marcello Malpighi (1628–1694), Morgagni
only rarely used the microscope.
It is frequently assumed that the study of cells in the urine, or urinary tract cytology,
started with the works of Papanicolaou. This is hardly surprising, since Papanicolaou
himself was unaware of the rich history of cytology and entitled his first clinical
cytology contribution “New Cancer Diagnosis.”
The history of urine cytology started in the nineteenth century, but was “discon-
tinuous” [51] and followed a tortuous path. It had many promising starts that were
cut short by the researchers’ career changes, personal tragedies, historical events
like revolutions and wars, or merely lack of interest on the part of the medical estab-
lishment. During the 1830s, urine microscopy was “rediscovered” [52] as an impor-
tant clinical tool in Paris and then quickly disseminated to German-speaking
countries and Great Britain.
In order to understand the background against which this rediscovery has
occurred, we need to review the changes in medical thinking brought about by the
Paris clinical school in the first decades of the nineteenth century, which refined the
concept of disease and pioneered the nosology and classification of diseases. The
concept of disease endured another major change through the work of Marie-
Francois Xavier Bichat (1771–1802), chief physician at the Hotel-Dieu, Paris. A
product of the widespread changes in medical education and practice brought about
by the French revolution, Bichat, a member of the Paris clinical school, introduced
the notion of tissues as distinct entities. In his 1800 “Treatise on Membranes” [53]
Bichat distinguished 21 types of tissues, different combinations of which formed
the organs of the body, and maintained that diseases attacked tissues rather than
whole organs (“…we must consider disease not from the standpoint of the com-
pound organs but from the standpoint of their different textures, which are almost
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 281
always attacked separately”). Remarkably, Bichat had these insights without the use
of a microscope through performing numerous autopsies, which may have contrib-
uted to his untimely death at age of 31. Half a century later, the German pathologist
Rudolf Ludwig Carl Virchow (1821–1902) further narrowed down the site of dis-
eases to the cell [54]. Although Virchow advocated this cellular view and promoted
it to the scientific medical community, both the aphorism “omnis cellula e cellula”
usually attributed to him and the concept of the cell as the site of disease had been
previously stated by the French scientist and revolutionary François-Vincent Raspail
(1794–1878) in 1825 [55] and 1843 [56].
In the aftermath of the reforms brought by the French Revolution, medicine was
taught at the bedside in hospitals, rather than in auditoria and lecture halls. The clinical
experience was fostered in contrast to the sterile memorization of “classic” textbooks.
This practice encouraged the observation of the patient throughout the entire course of
the disease, and, if the disease proved fatal, correlation with autopsy findings. This led
to the birth of the clinique (clinic or teaching hospital) in Paris, a medical teaching
model still used to this day. The success of this observation-based clinical teaching
model, with the motto “peu lire et beaucoup voir” (“read little but see a lot”), was made
possible by the very large number of patients (up to 5000 per year) seen by doctors and
students at that time. The observation of patients was coupled with the examination of
patients aided by new or improved methods of physical examination (auscultation and
percussion), statistical thinking, and the increasing importance of pathologic correla-
tion [57]. Based on radical empiricism and opposition to any abstract theory, system, or
philosophical thinking, it was so successful in patient care that it made Paris the mecca
of the medical world, attracting students and physicians from other European countries
and North America. This led to the dissemination of clinicopathologic or anatomo-
clinical, as it was called at the time, thinking to other countries.
Gross pathologic examination was the cornerstone of the scientific method of the
time, but as soon as microscopes became more reliable, microscopic examination
was seen by some as an extension of macroscopic examination and was rapidly
embraced. This occurred in the 1820s when achromatic lenses, small enough to be
used in a microscope, were created through the synchronous achievements of com-
peting and collaborating artisans and physicists across Europe, including Giovanni
Battista Amici (1786–1863) in Italy, Joseph Jackson Lister (1786–1869) in England,
Charles Louis Chevalier (1804–1859) in France, and Joseph Von Fraunhofer
(1787–1826) in Bavaria.
Achromatic lenses reduced the distortion factor from 20% to 3%; controlled iri-
descence, removing the fringes of color around objects; and eliminated the reticular
image, a historical artifact that plagued early microscopists [58]. Therefore, despite
their initial significant cost, which may have reached a person’s yearly salary, such
microscopes were quickly embraced in Paris, Berlin, Vienna, and London and revo-
lutionized biologic and medical thinking through the development of the cell theory
around 1840, which states that the cell is the quantum minimum of all plant and
animal life.
Before 1837, when Gabriel Valentin (1810–1883), a Swiss-German professor at
the University of Bern, developed a double-bladed knife for this purpose, obtaining
282 S. E. Pambuccian
Fig. 13.12 Example of daguerreotype of urinary sediment included in Donne and Foucault’s atlas
showing “white filaments” (left) and red blood cells, some crenated (right) in the urine
284 S. E. Pambuccian
Around the same time, Eugène-Napoléon Vigla (1813–1872) (Fig. 13.13), the
assistant of the founder of French nephrology, Pierre François Olive Rayer
(1793–1867), published his findings of the microscopic appearance of the crystals,
red cells, white cells, and epithelial cells most often seen of the urinary sediment in
various forms of kidney diseases [67, 68]. Vigla, who had taken Donné’s micros-
copy course [69], described the normal epithelial elements that “desquamate con-
tinuously” (“éprouve une desquamation continuelle dans l’état naturel”) and do not
represent a sign of a disease of the kidney or bladder. In women, when in doubt
about the urinary or genital tract origin of the epithelial cells, he recommended the
use of catheterized urine.
In the following years, urine microscopy was used more frequently, and a series
of new findings were described in rapid succession. Dysmorphic red blood cells of
Bright’s disease were described in 1841 by the French physician Louis Alfred
Becquerel (1814–1862) [70]. Urinary casts were described in 1844 concomitantly
by Julius Vogel (1814–1880), Johann Joseph Scherer (1814–1869), and Johann
Franz Simon (1807–1843) [71]. Simon’s drawings of casts were included by the
Guy’s Hospital physician Golding Bird (1814–1854) in his influential book on
Fig. 13.13 Eugène-Napoléon
Vigla, Lithograph by
Bornemann after Charles
Reutlinger Wellcome
Library, London
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 285
Fig. 13.14 Charles Moore and the title page of his 1852 paper. (From Robinson [81])
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 287
part of the Austro-Hungarian Empire. In his study, first presented at the meeting of
the College of Physicians of Prague in November 1854 and published in 1856,
Lambl describes the microscopic findings of ten patients with bladder cancer,
mostly from the surgical clinic of Franz (Jiří František) Pitha (1810–1875). Some
tumors had been studied at autopsy, but some were diagnosed intra vitam in urinary
specimens, usually obtained through catheterization. The article contains extensive
descriptions of the microscopic findings and four lithographs beautifully depicting
the cytology of the tumors. In his paper, Lambl, who is remembered for his descrip-
tion of Giardia and of aortic valve excrescences that bear his name, not only states
that a diagnosis of bladder cancer is possible through the microscopic examination
of the urine but also that such a diagnosis can be performed at the patient’s bedside.
Contrary to the general thinking of the time, the diagnosis of tumors was usually
made based on their gross appearance. Even for pathologists with experience with
the use of the microscope, the diagnosis of cancer was made based on the demon-
stration of invasion. In addition, the question whether papillary bladder tumors
without demonstrable invasion are benign or malignant had not been solved and
would haunt pathologists to this day. One of the greatest pathologists of that time,
Rokitansky in Vienna considered all papillary (villous) tumors of the bladder as
cancer (“Zottenkrebs”) [78], while Virchow thought that there are both benign and
malignant papillary bladder tumors, but papillary tumors are mostly benign [82,
83]. Lambl’s paper elicited a negative reaction, both because he was recommending
the diagnosis of bladder cancer by a cytologic method that cannot demonstrate inva-
sion and the fact that he considered, like Rokitansky, that all papillary bladder
tumors are malignant (“Harnblasenkrebs”). Ernst Leberecht Wagner (1829–1888),
professor of pathology in Leipzig, wrote a detailed and thoughtful review of Lambl’s
paper for the “1856 Year Book in Medicine” [84]. He argues that neither cell
enlargement and rounding, nor nuclear enlargement or multinucleation, which
could be seen in the urine sediment, can answer the question if we are dealing with
a cancerous or non-cancerous papillary bladder tumor. In his view, the diagnosis of
cancer could only be made in the presence of fragments of the flat bladder mucosa
showing cancer, in addition to the papillary structures. He adds that, based on these
criteria, he had diagnosed two cases of cancer in the urine, a primary cancer of the
bladder and a metastatic gastric cancer. A similar position was held by Gustaf
Vilhelm Johan von Düben (1822–1892), the chair of Pathological Anatomy at the
Karolinska Institute [85], and despite Lambl’s insistence on the possibility of diag-
nosing bladder cancer microscopically in the urine in a follow-up paper in Virchow’s
Archiv [86], this idea was not generally accepted by the pathology community
(Fig. 13.15).
Nonetheless, some of Lambl’s ideas and drawings were taken up by the German
pathologist Karl Julius Vogel (1814–1880), who included them in his very success-
ful book on the chemical and microscopic analysis of urine written together with the
German chemist Carl Theodor Ludwig Neubauer (1830–1879). In the 1830s, Vogel
had travelled to Paris during his medical studies, where he became familiar with the
use of the microscope, probably by attending Donné’s clinical microscopy course.
Vogel and Neubauer’s book became a standard reference manual on the laboratory
288 S. E. Pambuccian
Fig. 13.15 Wilhelm Lambl and the depiction of papillary tumor cells and fragments in the urine
in his 1856 paper
examination of urine for 50 years and was translated into English, French, and
Russian. The third edition of this book, published in 1858 [87], contains drawings
of cancer cells in the urinary sediment made by Vogel “in part from the valuable
work of Dr. Lambl,…, and partly from my own observation.” He describes the blad-
der cancer cells as “isolated cells, mostly irregular in form, partly caudate, and
partly branched, rather large, containing a large nucleus.” (Fig. 13.16).
A few years later, in 1864, the Edinburgh pathologist William Rutherford Sanders
(1828–1881), who had learned microscopy in Heidelberg with Friedrich Gustav
Jacob Henle (1809–1885), reports the case of a 43-year-old man with an extensive
“dendritic” (papillary) bladder cancer that discharged fragments into the urine, but
had no accompanying hematuria [88]. The tumor fragments formed plugs that inter-
mittently obstructed the urethra and under the microscope showed “peculiar con-
centric globular bodies,” which most likely represented squamous pearls.
The next case of bladder cancer diagnosed in the urine was reported by William
Howship Dickinson (1832–1913), an “all-around physician” [89], pathologist, and
specialist in childhood disease and kidney diseases, who would become president of
both the Pathological Society of London (1889–1990) and the Royal Medical and
Chirurgical Society (1897). In 1869, he reported an unusual case, in which tumor
fragments as large as a bean were passed in the urine without accompanying symp-
toms [90]. Microscopic examination of the small quantity of bloody urine in which
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 289
Fig. 13.16 Julius Vogel and depiction of bladder cancer fragments and single cells, partially
drawn after Lambl, in his book
the masses lay “abounded with large cells, presenting a great variety of shapes;
some were spheroidal, some spindle-shaped, others irregularly angular, and curved.
There could be no doubt that the masses described formed parts of an organised
cellular growth, such as would belong to a malignant tumour, apparently of the
encephaloid kind.” He confesses that he had never previously found anything diag-
nostic in cases of cervical or bladder cancer that he had examined microscopically.
After he acknowledges Sanders’ prior contribution, Dickinson states that, while
passage of cancer tissue fragments is rare, the passage of fragments of “villous
growth” (which he considers benign) is not, as he has observed it a number of times.
He states that: “When, however, the bladder has become the seat of a villous growth,
it not unfrequently happens that loops and fragments of blood vessels become
detached and pass out with the urine, and give conclusive evidence of the nature of
the disease,” since they are “easily recognised under the microscope.” Dickinson
also cautions that the microscopic suspicion of malignant disease of the bladder or
kidney is often a source of erroneous diagnosis of cancer. He writes: “Under cir-
cumstances of irritation and disturbance large cells, presenting quite the ideal of
cancer cells, of all shapes and characters, regular and irregular, rounded, angular,
and elongated, are occasionally thrown off in abundance from various parts of the
290 S. E. Pambuccian
urinary membrane. No isolated or detached cells, therefore, found in the urine, can
be regarded as proof of cancer in the urinary tract, while the absence of such cells
presents no argument in favour of the freedom of the bladder and kidney from
malignant growths.” (Fig. 13.17).
In the following years, bladder tumor cells were occasionally described and
depicted in the urine by authors from France, Austria, and Germany. In Vienna,
which had become a center of medical progress, the surgeon/urologist Robert
Ultzmann (1842–1889) and the chemist Karl Berthold Hofmann (1842–1922) pub-
lished a beautifully illustrated atlas on urinary sediments in 1871 [91]. The book
includes drawings depicting papillae and single cells found in the urine in a case of
Vincenz Czerny (1842–1916) and states that this is a very rare finding. It also
advises that when only single cells (frequently unusually large, caudate cells, with
multiple nuclei) are present, a probable diagnosis should only be made when they
are seen in great quantity (Fig. 13.18).
In his 1874 book, the Paris pathologist Charles Philippe Robin (1821–1885) and
former student of Donne writes that: “The papillary epithelia, or fungating vegeta-
tions of the bladder neck, may be recognized by the character of this epithelium,
when these vegetations detach themselves from their surface, and form sediments in
the urine. The cells are sometimes isolated, sometimes united in tatters or papillary
sheaths. Sometimes papillae or clusters of papillae containing a vascular loop, and
covered with their epithelium, are also seen.”
In 1877, the German pathologist Max Perls (1843–1881), the inventor of the iron
stain that bears his name, published a beautifully illustrated textbook of pathology.
This included drawings made by his former assistant, Ziesing, depicting a urinary
bladder villous tumor and cancer cells (Fig. 13.17). Perls died at the age of 38 of
typhus acquired through the practice of pathology, and the second edition of his
book was published under the care of Friedrich Neelsen (1854–1898), the co-
developer of the Ziehl-Neelsen stain. The drawings of the cytology of bladder
Fig. 13.17 Drawing from Dickinson’s paper depicting bladder cancer in the urine. Portrait of
William Howship Dickinson from his obituary in Br Med J Jan 18, 1913
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 291
Fig. 13.18 Bladder cancer in the urinary sediment depicted in Ultzmann and Hofmann’s atlas
a b c
Fig. 13.19 Drawings by Dr. Ziesing from Perls’ book showing a fresh papillary tumor of the blad-
der, magnified 50× (a) and 250× (b), and cells from a papillary carcinoma of the bladder (c)
suitable. for examination, are sometimes withdrawn with the catheter” [93]
(Fig. 13.20).
Towards the end of the nineteenth century, the Danish surgeon Niels Thorkild
Rovsing (1862–1927) published a paper on the diagnosis of kidney cancer [94], in
which he emphasized the diagnostic value of finding abnormal large round or
spindle-shaped cells sometimes with fatty degeneration in the urine.
The last decades of the nineteenth century witnessed extraordinary progress in
the study of bladder tumors, especially due to the introduction of cystoscopy into
routine urologic practice [95–97] and progress in surgical technique, anesthesia,
and antisepsis, which allowed the performance of cystectomies and other bladder
operations. During the same period, microscopes were significantly improved
through the development of focusable condensers, homogeneous immersion, and
apochromatic lenses at Carl Zeiss’s [98] workshop, based on Ernst Abbe’s research.
The last two decades of the nineteenth century were also the time in which most of
the histopathology techniques that are still in use today were perfected. Edwin
Klebs (1834–1913) introduced paraffin embedding and fixation with Zenker’s solu-
tion in 1869; aniline dyes (methylene blue and others, including eosin) were first
used as cytologic stains by Paul Ehrlich (1854–1915) in 1878. This was around the
same time that multiple researchers used the combinations of hematoxylin and
eosin [99]. Formalin fixation was introduced by Isaac Blum (1833–1903) in 1893,
and before the end of the nineteenth century, Paul Meyer (1848–1923) formulated
the currently used hematoxylin-eosin. Dmitri Leonidovich Romanowsky
(1861–1921) formulated a stain consisting of mixture of eosin and aged methylene
blue that was modified multiple times and proved to be useful in hematology, cytol-
ogy, and bacteriology [100–102]. By the turn of the century, most histologic
Fig. 13.20 James Tyson and the depiction of urinary sediment cells in his 1878 book
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 293
techniques that we use today, including formalin fixation, paraffin embedding, and
cutting thin sections with a rotary microtome, were already in widespread use in
laboratories on both sides of the Atlantic and histopathology began to dominate
pathology. At the same time, the cytologic methods, which had been used until then,
were largely abandoned by pathologists.
In these conditions, the microscopic diagnosis of urinary sediments was prac-
ticed predominantly by clinicians with an interest in nephrology or urology or by
“hybrid” clinician-pathologists. Towards the end of the nineteenth century, two
American clinician/pathologists practicing in New York City published an interest-
ing work on the urinary cytologic diagnosis of bladder tumors. The first of these was
the New York University Hospital Professor of Pathology and Clinical Medicine,
Farquhar “Frank” Ferguson (1851–1907). In 1892, he presented the results of his
new method of microscopic diagnosis of bladder tumors at the New York
Pathological Society. This was the first reported use of cellblock preparations,
4 years before Bahrenberg’s use of cellblock preparation obtained through paraffin
embedding clotted ascitic fluid [103]. However, his method, which he considered
superior to that of direct smears from the urinary sediment, was very tedious and
consisted of filtering the urine collected over 24 hours through a cheesecloth, hard-
ening the resulting sediment with alcohol, embedding it in celloidin, and cutting
100–200 sections from this. The sections were then screened at low power; if “any-
thing of importance” was found at screening, the sections were stained and exam-
ined at high power. Despite the fact that Dr. Ferguson stated that he has made the
diagnosis of bladder tumors on several occasions by this method and that “a greater
adoption of this method would soon enlarge our knowledge of tumors of the blad-
der,” this method was too cumbersome for clinical use and was to our knowledge
not adopted by others. Ferguson also acknowledged that one of his New York col-
leagues, Dr. Louis Heitzmann, was also engaged in the cytologic diagnosis of blad-
der cancer for several years. Louis Heitzmann (1864–1939) had been taught
microscopy by his father, the Austrian pathologist Carl Heitzmann (1836–1896)
who immigrated to the United States after failing in his attempt to succeed
Rokitansky as chair of pathology in Vienna. He was a consulting physician at Lenox
Hill Hospital, New York, and Professor of Pathology at the New York Medical
College. Louis Heitzmann published the results of his experience with urinary
cytology in the diagnosis of bladder tumors in a successful book first printed in
1899 [104], successive editions of which followed until 1936. The criteria Heitzmann
uses for the diagnosis of bladder papilloma (the equivalent of low-grade urothelial
neoplasms in our current classification) are very similar to those used in The Paris
System, i.e., the presence of fibrovascular cores: “The characteristic features of a
papilloma are peculiar connective tissue shreds that are, as a rule, abundant. They
are long, or extremely irregular, frequently branching in different directions and
often assume the shape of coils or knobs… in which capillary blood vessels filled
with blood corpuscles are seen coursing in various directions.” “The connective tis-
sue shreds found in villous cancer may be even larger and more irregular than those
seen in papilloma…. A number of these shreds contain large irregular cancer epithe-
lia, sometimes even small nests, a feature never found in papilloma.” In 1911, the
294 S. E. Pambuccian
a b c
Fig. 13.21 (a, b) Urine sediment findings in papilloma (left) and papillary (“villous”) carcinoma
(right) depicted in Heitzmann’s book. (c) Urine sediment findings in a carcinoma of the renal pel-
vis diagnosed preoperatively by Wyeth
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 295
Fig. 13.22 Drawings of renal cell carcinoma and papillary bladder tumor from Eastes’ article
Fig. 13.24 Ulrik Quensel and the urinary sediment finding in bladder cancer depicted in his paper.
J Ulrik T Quensel, https://sok.riksarkivet.se/sbl/artikel/7457, Swedish biographical lexicon (art by
Margareta Åman). (Retrieved 2020-07-02)
to his cytology study were that the cystoscopically obtained bladder washings are
useful in the diagnosis of malignant (but not benign) papillomas and invasive papil-
lary and solid carcinomas. He considers the presence of epithelial cells with large
nuclei (over 13×7 μ) and nucleoli over 2–3 μ as suspicious for malignancy and cells
with nucleoli over 2–3 μ as malignant.
In the late 1920s, the London pathologist Leonard Stanley Dudgeon (1876–1938),
working at the St. Thomas Hospital, introduced a form of intraoperative diagnosis
based on smears made from scraping the surface of fresh operative specimens [118].
The smears were wet-stained with Schaudinn’s solution (a mix of 2 volumes of
mercuric chloride, one volume of alcohol, and glacial acetic acid) and then stained
with Mayer’s hemalum and counterstained with a weak eosin solution. A paper co-
authored by the young Australian-born surgery trainee, Norman Rupert Barrett
(1903–1979) [119], who would attain eponymic fame for his description of the
columnar-lined esophagus, summarized over 1000 cases examined by this tech-
nique. The paper provides a detailed description and depiction of the cytologic cri-
teria of malignancy that can hardly be improved upon even today. The paper finally
laid to rest the misconception started by Virchow and repeated almost verbatim by
surgeons and pathologists that cancer cannot be diagnosed from the appearance of
its cells. This was still the prevailing opinion during Dudgeon’s times, as stated by
the influential surgeon and cancer specialist Sir John Bland-Sutton (1855–1936):
“in the appearance of a cell from a cancer there is nothing characteristic of the dis-
ease, nothing that would lead a pathologist to identify it as a malignant cell. Cancer
can only be identified in sections showing the relation of cells to each other in a
group.” [120]
In addition, Dudgeon and Barrett’s paper also stresses the value of the method in
the diagnosis of bladder papillomas, in which early diagnosis of malignant change
can be made (Fig. 13.25).
298 S. E. Pambuccian
Fig. 13.25 Professor Leonard S. Dudgeon and Norman Barrett, the latter as a surgical resident, at
St. Thomas Hospital, London. (Bamforth [167]; Lord [168]). Below, images from Dudgeon and
Barrett’s paper with the following captions. Left: Bladder papilloma showing the regular size of
the individual cells, placards of cells, and the abundance of isolated cells, all with a large amount
of cytoplasm. Right: On the right side of the circle, bladder carcinoma showing cells which have a
large nucleus and very little cytoplasm
(1858–1924) in Vienna developed around 1900 a cell block technique and applied it
to pus, body fluids, and urine. In 1917, he published the details of the technique,
which was very similar to that used today [121]. He gave the task to review the
accumulated experience with these cases to the young surgeon Abraham Philip
Zemansky, Jr. (1900–1928). In 1928, Zemansky reported the results of 914 of these
cases that had been reported since 1912, which included 46 cases of urine or bladder
washings [122]. He concluded that the cell block method is valuable in the study of
pleural and peritoneal fluids but of doubtful value in the diagnosis of urines/bladder
wash. His conclusion is based on the fact that only two of 46 cases showed tumor
cells in the cell blocks from urinary tract specimens, a conclusion the validity of
which is difficult to assess without knowing the prevalence of urinary tract malig-
nancy in the population from which these samples were obtained. In addition, cyto-
histologic correlation showed that, of the cases with surgical or autopsy follow-up,
only one of five cases of carcinoma of the bladder and none of six kidney tumors
showed tumor cells in the urine cell blocks. To explain the lack of sensitivity of this
method, he mentions the small quantity of urine received in the laboratory
(30–50 ml), which is a small fraction of the 24 hour urine output and is not sufficient
for the detection of tumor cells. He also mentions the technical difficulties encoun-
tered in embedding and cutting the minute amount of centrifugate obtained from
these specimens and the cellular degeneration caused by the hypotonic urine.
Unfortunately, Zemansky died the same year from septicemia after a mastoid opera-
tion at the age of 28.
In his study published in 1925, the University of Buffalo urologic surgeon
Frederick J. Parmenter (1880–1962) concluded that “the cytological examination of
the urine is a valuable aid to diagnosis and should be practiced as a routine proce-
dure,” that “its greatest value is for suspected tumors of the upper part of the
tract”….., and finally that “only positive results are of value” [123]. These conclu-
sions are largely valid to this day.
Despite this evidence and prior studies documenting the usefulness of urine cytol-
ogy for the diagnosis of bladder tumors, no widespread use of urinary cytology is
documented until 1945. While it is very important to ask the question why it took
urinary cytopathology so long to thrive [124], it was not for lack of studies docu-
menting its value, as can be surmised by the preceding examples. Also in doubt: “the
interest was there; the findings were striking; but the impetus of a devoted, painstak-
ing scientist was lacking to elevate the subject to consistent scientific duplicability”
[51]. At least in the case of urine cytology, “painstaking scientists” had been per-
forming study after study reaching the same conclusion that cytology was a valid and
valuable method for bladder cancer diagnosis. The problem was most likely that they
were ahead of their times. In his insightful analysis, Dr. James R. Wright, Jr. con-
cluded that the most important factor was the need for paradigm shifts in the under-
standing of cancer [124], from gross to microscopic and from the level of tissues to
the level of cells. According to Thomas Kuhn, such paradigm shifts are fundamental
changes in the basic concepts, experimental practices, techniques, and values shared
by members of a scientific community (in this case the medical community). Such
shifts do not occur suddenly, but gradually. Although the new paradigm, in this case
300 S. E. Pambuccian
the diagnosis of cancer by microscopic examination of tissues and then of cells, may
gain followers, its adoption may take time. As the German physicist Max Planck
(1858–1947) puts it: “a new scientific truth does not triumph by convincing its oppo-
nents and making them see the light, but rather because its opponents eventually die,
and a new generation grows up that is familiar with it.”
For Virchow, the father of “cellular pathology,” “cancer” was defined by four major
characteristics: local progression, recurrence after resection, lymphadenopathy, and
distant metastases, all of which are clinical or gross features [125]. The only micro-
scopic feature that Virchow accepted was the presence of invasion. He did not accept
cellular malignancy criteria. He wrote “Indeed, there are no cells or nuclei that would
be characteristic of cancer” and “although I do not deny that [...] the size of the nuclei
and nucleoli could give meaningful moments for the diagnosis, it was impossible for
me to find anything specific in them124”. The enormous influence that Rudolf Virchow
had on the pathology of the nineteenth century, amplified by his network of former
trainees who occupied the chairs of almost all pathology departments in German-
speaking universities, drowned out the opposing views voiced in the middle of the
nineteenth century by the Danish pathologist Adolph Hannover (1814–1894) and the
former trainees of Alfred Donné, Hermann Lebert, and John Hughes Bennett, who
believed that the cellular changes in cancer are specific enough to allow a diagnosis.
Therefore, the paradigm shift could only occur with a new generation of pathologists,
who were not trained in Virchow’s dogma of nonspecificity of cancer cells and open
to the idea that a diagnosis of cancer is possible through microscopic examination of
the cells. This occurred in the late 1920s and 1930s, when cytologic diagnosis of can-
cer was brought back into the scientific discourse by pathologists from different coun-
tries, focusing on different areas of pathology. Leonard Dudgeon in London used
cytology to perform intraoperative diagnoses, Aurel A. Babeş (1886–1961) in
Bucharest, Romania, pioneered the cytologic diagnosis of cervical cancer [126], while
Fred Stewart in New York diagnosed thousands of cases of various cancers in needle
aspiration biopsies [127]. The explanation for why it took another two decades until
cytologic diagnosis was finally embraced is complex. For one thing, the proponents of
the cytologic method were not self-promoting, were also involved in other activities,
and did not exclusively focus on cytologic diagnosis. They failed to seek or gain the
support of medical societies interested in fighting cancer, which would prove to be
crucial in the future success of the method.
Finally, there was no perceived need for this method, which may have required
technical and interpretive skills that few pathologists of the time possessed, espe-
cially in areas that were relatively easily amenable to surgical biopsy. In the case of
urine cytology, with the widespread use of cystoscopy, which allowed for an easy
tissue diagnosis, many doubted the value of the cytologic diagnosis for clinical
decision-making. The fact that cytology was of little use in the diagnosis of well-
differentiated papillary tumors classified by some as papillomas and by others as
carcinomas was another argument against the use of this method, especially since
these tumors were easily diagnosed by cystoscopy. An additional obstacle in the
acceptance of urine cytology was the fact, alluded to by some of the early
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 301
proponents of cytology, that the lack of pathologists’ cytology training and experi-
ence had resulted in false-positive diagnoses, which made clinicians distrust
the method.
The criteria of malignancy that he used in urine sediments were the same, namely,
“modifications and abnormalities of the nucleus (enlargement, anisonucleosis, frag-
mentation, hyperchromatosis [sic], granular arrangement of the chromatin, promi-
nence of nucleoli, etc.), changes affecting the cytoplasm (basophilia, vacuolation,
leucocytic infiltration, etc.), and significant deviation of cells from their normal size
and form.” Papanicolaou also notes that catheterized urine specimens are generally
preferable to voided urine specimens. He also clearly states that “the smear
302 S. E. Pambuccian
technique should not be considered as a substitute for biopsy, curettage or any other
standard method of diagnosis, but as an adjunct to them.” (Fig. 13.26).
It is important to note that these results were taking advantage of the fact that
the Memorial Hospital (currently the Memorial Sloan Kettering Cancer Center)
and the New York Hospital had the practice of interpreting low-grade papillary
Fig. 13.26 George Papanicolaou, with his chief cytotechnologist, Charlotte Street, and Dr. John
Seybolt, his collaborator on urine cytology studies. (From the George Papanicolaou, MD
Collection, Cornell University Medical Center)
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 303
Fig. 13.27 Dr. Leopold Koss. (From Steven I. Hajdu and Hormoz Ehya. Foundation of Diagnostic
Cytology Annals of Clinical & Laboratory Science, vol. 38, no. 3, 2008). Myron Melamed, from
M. B. Zaman. Myron R. Melamed, MD, 1927–2013 Journal of the American Society of
Cytopathology 2014, 3 (1), 2–4)
(UroVysion® test), the latter deletion relates to the loss of the p16 (CDKN2A) sup-
pressor gene. These tests typically have higher sensitivity but lower specificity than
urine cytology and frequently have problems similar to urine cytology, with lower
sensitivity in the detection of low-grade papillary neoplasms (see TPS 2.0 Chap. 10).
The initially high enthusiasm in using these tests in combination with, or instead of,
urine cytology has been dampened by their lower analytical sensitivity and specificity
in real-world use than was reported in the initial studies. In addition, their use has been
stymied by relatively high costs and lack of a clear place in the management of patients
for the diagnosis or surveillance of bladder cancer.
In recent years, an ever-growing number of multiplex molecular assays using
reverse transcription quantitative polymerase chain reaction or massive parallel
sequencing have been proposed or introduced in practice. These assays assess for
various genetic or epigenetic abnormalities that are found in urothelial carcinomas,
including mutant FGFR3 which characterizes low-grade papillary urothelial neo-
plasms (CertNDx). On the other hand, TERT promoter and TP53 usually characterize
high-grade urothelial carcinoma (UroSEEK). Despite promising initial results, the
value of these tests, cost-effectiveness, and their place in the management of the
patient suspected of, or under surveillance for, urothelial carcinoma is still unclear.
In addition to the difficulty diagnosing low-grade neoplasms was the large propor-
tion of unsatisfactory specimens, especially for voided urines. This was addressed
with a number of technical advances meant to increase the concentration of the cells
306 S. E. Pambuccian
present in the sample. These methods include filter techniques (Millipore, Nuclepore),
cytocentrifugation, and liquid-based cytology. Even in the face of improved method-
ology that readily increased the cellularity of the samples prepared by these tech-
niques, it was still not entirely clear what constituted an adequate urine sample.
A nagging and persistent problem, which seemed to get worse rather than better
with time, was the large proportion of indeterminate diagnoses rendered in urine
cytology. In cytopathology, the rates of atypical and suspicious diagnoses are in gen-
eral associated with factors that are related to the patient (prior instrumentation or
therapy), the specimen (poor preparation, poor staining), and those attributable to the
pathologist (inadequate training and ability to make a diagnosis) [157]. In urinary
cytopathology, additional factors that can explain the high rate of atypical and suspi-
cious diagnoses are the low comfort level of pathologists with urine specimens, the
inherent difficulty and subjectivity of the diagnosis, as attested by relatively high
interobserver variability in urinary cytology interpretations, and, most importantly,
the lack of clear agreed-upon diagnostic categories and definitions. It is difficult to
communicate findings between facilities when local diagnostic opinions differ. It is
difficult to train cohorts of residents without clear consistent landmarks.
A number of classifications have been proposed starting in the 1980s [158–
161], all representing an improvement over Papanicolaou’s numeric class system,
but none of them was widely adopted or solved the problem of indeterminate
diagnoses. In 2013, two proposals, one representing a radical rethinking of uri-
nary cytology reporting, by the Johns Hopkins Hospital group led by Dr. Dorothy
L. Rosenthal [162, 163], and the other proposing a subclassification of the atypi-
cal group, made by a French group led by Dr. Eric Piaton [164], renewed the
interest in a new, reproducible, and practical classification of urine cytology diag-
noses. In the immediate aftermath of these proposals, Drs. Dorothy L. Rosenthal
and Eva M. Wojcik convened a meeting at the 18th International Congress of
Cytology held on May 26–30, 2013, in Paris, to discuss the possibility of a new
classification of urine cytology. On May 28, 2013, in a small room of the Palais
Des Congrès, in which the congress was held, the speakers of the two sessions on
urinary cytology met to discuss the potential new classification. They included, in
addition to Eva Wojcik and Dorothy Rosenthal, Güliz Barkan, Lukas Bubendorf,
Margareta Strojan Fležar, Drs. Rana Hoda, Daniel Kurtycz, Stefan Pambuccian,
Eric Piaton, Marcus Quek, and Momin T. Siddiqui. The president of the American
Society of Cytopathology (ASC), Ritu Nayar, and the president of the International
Academy of Cytology (IAC), Philippe Vielh, were also present at this inaugural
Paris meeting. The group agreed that a new, standardized terminology for report-
ing urinary cytology based upon histopathology and clinical outcomes was
needed. The agreed-upon guiding principle for the new classification would be
that urine cytology is meant to diagnose the clinically important high-grade uro-
thelial carcinoma capable of local invasion and distant spread. Cellular changes
that did not indicate a significant risk for high-grade urothelial carcinoma would
be classified under and relegated to the negative diagnostic category (“Negative
for high grade urothelial carcinoma”), including changes possibly attributable to
a low-grade neoplasm. An important principle of this system was that all
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 307
diagnostic categories should be clearly defined and have strict diagnostic criteria,
restricting the use of “atypical” to cases in which the cytologic features cannot be
attributed to a known etiology [165]. It was also agreed that the new classification
would be designated The Paris System (TPS), analogous to The Bethesda System,
to reflect the location of the first meeting that took place on the subject. Chapter
lead authors were chosen and contributing authors were recruited. Subsequent
meetings to clarify the details of this classification were held at the ASC meeting
in Orlando, FL, in November 2013; the USCAP meeting in San Diego, CA, in
March 2014; the European Congress of Cytopathology, Geneva, Switzerland, in
September 2014; and the ASC meeting in Dallas, TX, in November 2014. During
these meetings, the participants discussed the details of the classification and the
clinical significance of the cytologic diagnosis. After considering the results of
web surveys, additional discussions, and teleconferences, the authors decided on
the final diagnostic categories, their definition, and diagnostic criteria. Finally,
these were included in a book authored by 49 cytopathologists, cytotechnologists,
surgical pathologists, and urologists from 10 countries. The Paris System for
Reporting Urinary Cytology was edited by Drs. Rosenthal, Wojcik, and Kurtycz
[166], with an introduction by Drs. Ritu Nayar and Philippe Vielh and a very
thoughtful prologue written by Dr. William M. Murphy [166], one of the pioneers
in urinary cytology. The Paris System was well received and was adopted by most
institutions in the United States and many institutions throughout the world. One
of the most important outcomes was the reduction of the “atypical” diagnoses in
almost all institutions that have adopted The Paris System.
Another of the important features of the first edition TPS was a call to research.
A host of questions needed to be answered. In the afterword to The Paris System
book, the editors admitted that much of the published information on which the
reporting system was built was based on poorly defined, inadequately studied, unin-
vestigated, and/or anecdotal evidence. They also hoped that the publication of TPS
would stimulate investigations that would fill gaps in our knowledge base. At the
time of the publication, there was no or only scant evidence on some of the issues
covered by TPS, especially regarding the use of the recommended criteria in differ-
ent types of preparation and the applicability of the diagnostic criteria for urothelial
carcinoma variants. It was also obvious that one of the most important issues regard-
ing this classification was assigning a risk of malignancy to every diagnostic cate-
gory to aid clinical management. Given that the Paris categories were transformative
and had not been used before in the manner recommended by the Paris system,
studies had to be done after the implementation of TPS to define these risks. As
such, from the beginning, it was decided that would be revised as soon as more
information became available. Preparations for the second edition of The Paris
System (2.0) were started by Drs. Wojcik, Rosenthal, and Kurtycz in 2019, and
meetings were held at the ASC meeting in Salt Lake City in November 2019 and
USCAP meeting in Los Angeles in March 2020.
In the midst of the COVID-19 pandemic, the authors of TPS 2.0 have been com-
mitted to completing this volume despite the additional demands on their professional
time and energies. The gratifying and near universal acceptance of the thinking behind
308 S. E. Pambuccian
TPS continues the centuries-old tradition of turning to urine to unravel many of the
mysteries of the human body. TPS will continue to be modified by new evidence and
evolving concepts of urothelial neoplasia. The authors and editors of TPS look for-
ward to advances in genetics and associated biomarkers that will help the laboratory
refine accuracy and advance precision in the decades ahead (Fig. 13.27).
Fig. 13.27 The participants at the initial meeting in Paris 2013 from left to right: Upper row –
Dorothy L Rosenthal, Eva M. Wojcik, Rana Hoda, Momin T. Siddiqui, Stefan E. Pambuccian. Lower
picture – Lukas Bubendorf, Margareta Strojan Fležar, Güliz A. Barkan, Eric Piaton, and Marcus Quek
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 309
References
1. Osler W. A note on the teaching of the history of medicine. Br Med J. 1902;2:93.
2. Fuller T. The historie of the holy warre. Cambridge: Thomas Buck; 1639.
3. Hörl WH. The medicinal use of urine. Am J Nephrol. 1999;19:111–3.
4. Savica V, Calo LA, Santoro D, Monardo P, Mallamace A, Bellinghieri G. Urine therapy
through the centuries. J Nephrol. 2011;24(Suppl 17):S123–5.
5. Harris WV. Scatological Asklepios: the use of excrement in Graeco-Roman Healthcare1. J
Hist Med Allied Sci. 2019.
6. Kiefer JH. Uroscopy: the artist's portrayal of the physician. Bull N Y Acad Med.
1964;40:759–66.
7. Friedlaender GE, Friedlaender LK. Saints Cosmas and Damian: patron saints of medicine.
Clin Orthop Relat Res. 2016;474:1765–9.
8. Morgagni G. De sedibus et causis morborum; The Seats and Causes of Diseases, 1761,
Translation, Alexander B., 1768, Classics of Medicine Library, 1983, Gryphon Editions,
Birmingham Alabama.
9. Diamandopoulos AA, Goudas PC. The late Greco-Roman and Byzantine contribution to the
evolution of laboratory examinations of bodily excrement. Part 1: urine, sperm, menses and
stools. Clin Chem Lab Med. 2003;41:963–9.
10. Diamandopoulos AA, Goudas PC. The late Greco-Roman and Byzantine contribution
towards the evolution of laboratory examinations of bodily excrement. Part 2: sputum, vomit,
blood, sweat, autopsies. Clin Chem Lab Med. 2005;43:90–6.
11. Eknoyan G. Beginnings-the kidney and nephrology in ancient mesopotamian culture. Semin
Dial. 2016;29:236–46.
12. Shokeir AA, Hussein MI. The urology of pharaonic Egypt. BJU Int. 1999;84:755–61.
13. Gordetsky J, O'Brien J. Urology and the scientific method in ancient Egypt. Urology.
2009;73:476–9.
14. Scurlock J. Mesopotamian beginnings for Greek science? In: Keyser JT, Scarborough J, edi-
tors. The Oxford handbook of science and medicine in the classical World. Oxford: Oxford
University Press; 2018. p. 1–14.
15. Jungersen K. Reception of uroscopy in Scandinavia. J Nephrol. 2009;22(Suppl 14):50–4.
16. Bloom DA, Milen MT, Heininger JC. Claudius Galen: from a 20th century genitourinary
perspective. J Urol. 1999;161:12–9.
17. Temkin O. Galenism; rise and decline of a medical philosophy. Ithaca N.Y: Cornell University
Press; 1973.
18. Wallis F. Inventing diagnosis: Theophilus' De urinis in the classroom. Dynamis.
2000;20:31–73.
19. Kouba E, Wallen EM, Pruthi RS. Uroscopy by Hippocrates and Theophilus: prognosis versus
diagnosis. J Urol. 2007;177:50–2.
20. Androutsos G. Theophilus Protospatharius: byzantine forerunner of urology. Hist Sci Med.
2007;41:41–8.
21. Eknoyan G. Arabic medicine and nephrology. Am J Nephrol. 1994;14:270–8.
22. Collins KE, Sussman M. Isaac Israeli and his book of urine. Scott Med J. 1999;44:86–8.
23. Massry SG. Isaac Judaeus Israeli: a Jewish founder of the origins of nephrology. J Nephrol.
2009;22(Suppl 14):8–11.
24. Rosner F, Muntner S. Moses Maimonides' aphorisms regarding analysis of urine. Ann Intern
Med. 1969;71:217–20.
25. Massry SG. Maimonides: physician and nephrologist. Am J Nephrol. 1994;14:307–12.
26. Touwaide A. Arabic urology in Byzantium. J Nephrol. 2004;17:583–9.
27. Pardalidis N, Kosmaoglou E, Diamantis A, Sofikitis N. Uroscopy in Byzantium (330-1453
AD). J Urol. 2008;179:1271–6.
28. Dal Canton A, Castellano M. Theory of urine formation and uroscopic diagnosis in the
Medical School of Salerno. Kidney Int. 1988;34:273–7.
310 S. E. Pambuccian
53. Bichat, M F X, Treatise on Membranes (Traité des membranes), 1801 France, translation by
Joseph Houlton, July 1821, (Classics of Medicine, project Gutenburg). https://www.guten-
berg.org/ebooks/52987.
54. Virchow RLK. Cellular pathology as based upon physiological and pathological histol-
ogy. Twenty Lectures delivered in the pathological institute of Berlin during the months of
February, March and April, 1858. Transelated from the second edition of the original by frank
chance. London: John Churchill, 1860.
55. Raspail F-V. Développement de la fécule dans les organes de la fructification des céréales.
Ann Sci Nat. 1825;6:224–39.
56. Raspail F-V. Histoire naturelle de la santé et de la maladie chez les végétaux et chez les
animaux en général, et en particulier chez l'homme; suivie du formulaire pour une nouvelle
méthode de traitement hygiénique et curatif. Paris: Alphonse Levasseur; 1843.
57. de Saint-Maur PP. The birth of the clinicopathological method in France: the rise of mor-
bid anatomy in France during the first half of the nineteenth century. Virchows Arch.
2012;460:109–17.
58. Majno G, Joris I. The microscope in the history of pathology. With a note on the pathology of
fat cells. Virchows Arch A Pathol Pathol Anat. 1973;360:273–86.
59. Bracegirdle B. Dyeing for the microscope. J Soc Dye Colour. 1993;109:54–6.
60. Bracegirdle B. The history of histology: a brief survey of sources. Hist Sci. 1977;15:77–101.
61. Wick MR. Diagnostic histochemistry: a historical perspective. Semin Diagn Pathol.
2018;35:354–9.
62. Donne A. Tableau des differents depots de matiere saline et des substances organisees, qui se
font dans les urines, presentant les characteres propres a les distinguer entre eux et a recon-
naitre leur nature. Paris: Crochard et Comp; 1839.
63. Donné A. Cours de microscopie complémentaire des études médicales anatomie
microscopique et physiologie des fluides de l'économie. Paris: J.-B. Bailliere; 1844.
64. Donné A, Foucault L. Cours de microscopie complementaire des études médicales anato-
mie microscopique et physiologie d es fluides de l'économie. Atlas exécute d'apres nature au
microscope-daguerréotype. Paris: J.-B. Bailliere; 1845.
65. Donné A. Cours de microscopie complémentaire des études médicales. Paris: J. B. Baillière;
etc.; 1844.
66. Daremberg G. Les grands médecins du XIXe siècle. ÉDITEURS: PARIS MASSON ET
Cie; 1907.
67. Vigla EN. Ētude microscopique de l’urine, éclairée par l’analyse chimique L’Expérience.
1837: 177–190.
68. Vigla EN. Ētude microscopique de l’urine, éclairée par l’analyse chimique L’Expérience.
1838: 193–204.
69. Cameron JS. Towards the millennium: a history of renal anaemia and the optimal use of epo-
etin. Nephrol Dial Transplant. 1999;14:10–21.
70. Becquerel A. Séméiotique des urines: ou, Traité des altérations de l'urine dans les maladies,
suivi d'un traité de la maladie de Bright aux divers ages de la vie. Paris: Fortin, Masson; 1841.
71. Simon F, Chemie VfPuP. Beiträge zur physiologischen und pathologischen Chemie
und Mikroskopie in ihrer Anwendung auf die praktische Medizin. Verlag von August
Hirschwald, 1844.
72. Beale LS. On urine, urinary deposits, and calculi: their microscopical and chemical examina-
tion, including the chemical and microscopical apparatus required, and tables for the practical
examination of the urine in health and disease : the anatomy and physiology of the kidney :
with upwards of sixty original analyses of the urine in disease, and general remarks on the
treatment of certain urinary diseases. London: J. Churchill, 1861.
73. Hajdu SI. A note from history: landmarks in history of cancer, part 2. Cancer.
2011;117:2811–20.
74. Mehlhausen D. Charité-Annalen. Berlin: A. Hirschwald; 1877.
75. J. M. Ueber den feinern Bau und die Formen der krankhaften Geschwülste. Berlin:
G. Reimer, 1838.
312 S. E. Pambuccian
76. Bennett JH. On cancerous and cancroid growths. Edinburgh: Sutherland and Knox; etc.; 1849.
77. Der GJ. Zottenkrebs: Eine pathologisch-anatomische Untersuchung. Mainz: Verlag von
Eduard Janitsch; 1851.
78. von Rokitansky K. Über den Zottenkrebs. Sitzungsberichte der Mathematisch-
Naturwissenschaftlichen Classe der Kaiserlichen Akademie der Wissenschaften.
1852;8:513–35.
79. Moore CH. An account of a case of pulsating tumour in which the urine contained cancer
cells. Med Chir Trans. 1852;35:459–69.
80. Anonymous. Charles Hewitt Moore. Br Med J. 1870;1:641–2.
81. Robinson JO. Treatment of breast cancer through the ages. Am J Surg. 1986;151:317–33.
82. Virchow R. Über Kankroide und Papillargeschwülste. Verhandlungen der Physikalisch-
medizinische Gesellschaft zu Würzburg. 1850;1:106–16.
83. Virchow R. Die krankhaften Geschwülste, Band. I. Berlin: August Hirschwald; 1863.
84. Wagner E. 808. Ueber Harnblasenkrebs. Ein Beitrag zur mikroskopischen Diagnostik;
von Dr. Lambl in Prag. (Prag. Vjhrschr. XIII. 1. 1856.). In: Richter HE, Winter A, editors.
Schmidt's Jahrbücher der in- und ausländischen gesammten Medicin. Leipzig: Verlag von
Otto Wigand, 1856:39–41.
85. von Düben GVJ. Treatise on microscopical diagnosis. Translated by Louis Bauer. New York:
John Wiley; 1859.
86. Lambl W. Mikroskopische Untersuchungen der Darm·Excrete. Beitrag zur Pathologie
des Darms und zur Diagnostik am Krankenbette. Mit lithographirten Abbildungen.
Vierteljahrschrift für die praktische Heilkunde. 1859;61:1–58.
87. Neubauer CTL, Vogel J. Anleitung zur qualitativen und quantitativen Analyse des Harns:
sowie zur Beurtheilung der Veränderungen dieses Secrets ; mit besonderer Rücksicht auf die
Zwecke des praktischen Arztes. Kreidel & Nieder: Wiesbaden; 1858.
88. Sanders WR. Cancer of the bladder. Fragments forming urethral plugs discharged in the
urine. Concentric colloid bodies. Edinb Med J. 1864;10:273–4.
89. Anonymous. William Howship Dickinson, M.D.Cantab., F.R.C.P.Lond. Br Med
J. 1913;1:141–3.
90. Dickinson WH. Portions of cancerous growth passed by urethra. Trans Pathol Soc London.
1869;20:233–7.
91. Fogazzi GB. Austria 19th century. An atlas on urinary sediment written by a surgeon and a
chemist still of interest today. Nephrol Dial Transplant. 1999;14:2038–40.
92. Senn N. The pathology and surgical treatment of tumors. Philadelphia: W. B. Saunders; 1895.
93. Tyson J. A guide to the practical examination of urine for the use of physicians and students.
Philadelphia: Lindsay and Blakiston; 1878.
94. Rovsing T. Ueber die Diagnose und die Behandlung der bösartigen Nierengeschwülste bei
Erwachsenen. Archiv fur klinische Chirurgie. 1895;49:407–50.
95. Herr HW. Max Nitze, the cystoscope and urology. J Urol. 2006;176:1313–6.
96. Moran ME. The light bulb, cystoscopy, and Thomas Alva Edison. J Endourol. 2010;24:1395–7.
97. Reuter MA, Reuter HJ. The development of the cystoscope. J Urol. 1998;159:638–40.
98. Louw DF, Sutherland GR, Schulder M. From microscopic to astronomic, the legacy of Carl
Zeiss. Neurosurgery. 2003;52:668–74. discussion 672-664
99. Cook HC. Origins of ... tinctorial methods in histology. J Clin Pathol. 1997;50:716–20.
100. Krafts K. The Malachowski-Wright-Giemsa stain: a many-splendored thing. Biotech
Histochem. 2011;86:5–6.
101. Krafts K, Hempelmann E, Oleksyn BJ. In search of the malarial parasite: biographical
sketches of the blood stain contributors. Parasitol Res. 2011;109:521–9.
102. Krafts KP, Pambuccian SE. Romanowsky staining in cytopathology: history, advantages and
limitations. Biotech Histochem. 2011;86:82–93.
103. Bahrenburg LPH. On the diagnostic results of the microscopical examination of the ascitic
fluid in two cases of carcinoma involving the peritoneum. Cleveland Med Gaz. 1896;11:274–8.
104. Heitzmann L. Urinary analysis and diagnosis by microscopical and chemical examination.
New York: W. Wood & Co.; 1899.
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 313
105. Wyeth GA. The location of pathological lesions of the genitourinary tract from epithelia
found in urine, proved by operation. New York Med J. 1911;94:473–7.
106. Eastes GL. The histology of the urinary tract in its relationship to morbid urinary deposits. Br
Med J. 1901;1:570–3.
107. CABOT AT. Surgical treatment of cancer of the bladder. The Boston Med Surg
J. 1909;160:65–8.
108. Gaillard A. Epithelia found in urine, and their differentiation, as an aid to correct diagnosis.
Med Rec (1866–1922). 1912;81:362.
109. Gaillard AT. Modern laboratory methods in the diagnosis of surgical diseases of the genito-
urinary tract. Ann Surg. 1914;59:267–77.
110. Bauer A. Erkennung einer Blasenzottengeschwulst aus dem Harnsediment mittels der
Alizarinmethode. Zeitschrift für urologische Chirurgie. 1932;35:219–21.
111. Bauer A. Zur Bewertung des basischen Charakters von Geschwulstzellen. Ärztliche
Rundschau. 1934;44:76–7.
112. Widal F, Sicard JA, Ravaut P. Cytodiagnostic de la méningite tuberculeuse. C R hebd Soc
Biol. 1900;52:838–43.
113. Le RP. Diagnostic de la nature des épanchements séro-fibrineux de la plèvre; Cytodiagnostic
(thèse). Paris: Durand; 1901.
114. Labbé M. Le cytodiagnostic; les méthodes d'examen des sérosités pathologiques et du liquide
céphalo-rachidien. Paris: Baillière; 1903.
115. Quensel U. Untersuchungen über die Morphologie des organisierten Harnsediments bei
Krankheitene der Nieren und der Harnwege und über die Entstehung der Harnzylinder
Nordiskt Medicinskt Arkiv. 1918;50: 319–660.
116. Stenius F. Zur Bewertung des epithelialen Harnsedimentes für die Diagnose der
Blasengeschwülste. Acta Chir Scand. 1925;58:240–55.
117. Stenius F. Studien über Pathologie und Klinik der Papillome und Karzinome der Harnblase.
Arb Pathol Inst Univ Helsingfors. 1925;3:27–190.
118. Dudgeon LS, Patrick CV. A new method for the rapid microscopical diagnosis of tumours:
with an account of 200 cases so examined. Brit J Surg. 1927;15:250–61.
119. Dudgeon LS, Barrett NR. The examination of fresh tissues by the wet-film method. Brit J
Surg. 1934;22:4–22.
120. Bland-Sutton J. Tumours, innocent and malignant: their clinical characters and appropriate
treatment. 7th ed. London: Cassell; 1922.
121. Mandelbaum FS. The diagnosis of malignant tumors by paraffin sections of centrifuged exu-
dates. J Lab Clin Med. 1917;2:580.
122. Zemansky APJ. The examination of fluids for tumor cells: an analysis of 113 cases checked
against subsequent examination of tissue. Am J Med Sci. 1928;175:489–503.
123. Parmenter FJ. The cytological examination of the urine as an aid to the diagnosis of tumors
of the genito-urinary tract. Surg Gynec Obstet. 1925;40:531–4.
124. Wright JR Jr. Cytopathology: why did it take so long to thrive? Diagn Cytopathol.
2015;43:257.
125. Hauptmann S, Schnalke T. Rudolf Virchows' view of malignant growth. Pathologe.
2001;22:291–5.
126. Babes A. Diagnosis of cancer of the uterine cervix by smears. Acta Cytol. 1967;11:217–24.
127. Stewart F. The diagnosis of tumours by aspiration. Am J Pathol. 1933;9:801–12.
128. Papanicolaou GN, Marshall VF. Urine sediment smears as a diagnostic procedure in cancers
of the urinary tract. Science. 1945;101:519–20.
129. Traut HF, Papanicolaou GN. Cancer of the uterus: the vaginal smear in its diagnosis. Cal West
Med. 1943;59:121–2.
130. Chantziantoniou N, Donnelly AD, Mukherjee M, Boon ME, Austin RM. Inception and devel-
opment of the Papanicolaou stain method. Acta Cytol. 2017;61:266–80.
131. Papanicolaou GN. The sexual cycle in the human female as revealed by vaginal smears. Am
J Anat. 1933;52:519–616.
132. Koss LG, Hoda RS. Koss's cytology of the urinary tract with histopathologic correlations.
New York: Springer Science+Business Media, LLC; 2012.
314 S. E. Pambuccian
133. Papanicolaou GN. Cytology of the urine sediment in neoplasms of the urinary tract. J Urol.
1947;57:375–9.
134. Stewart FW. Introduction. Bladder tumors: a symposium. Philadelphia: Lippincott; 1956.
135. Silberblatt JM. Exfoliative cytology as a screen test for urinary tract malignancy. Bull N Y
Acad Med. 1953;29:889–97.
136. Fremont-Smith M, Graham RM, Meigs JV. Early diagnosis of cancer by study of exfoliated
cells. J Am Med Assoc. 1948;138:469–74.
137. Daut RV, Mc DJ. Diagnosis of malignant lesions of the urinary tract by means of microscopic
examination of centrifuged urinary sediment. Proc Staff Meet Mayo Clin. 1947;22:382–6.
138. McDonald JR. Exfoliative cytology in genitourinary and pulmonary diseases. Am J Clin
Pathol. 1954;24:684–7.
139. Hazard JB, McCormack LJ, Belovich D. Exfoliative cytology of the urine with special refer-
ence to neoplasms of the urinary tract: preliminary report. J Urol. 1957;78:182–7.
140. Foot NC, Papanicolaou GN, Holmquist ND, Seybolt JF. Exfoliative cytology of urinary sedi-
ments; a review of 2,829 cases. Cancer. 1958;11:127–37.
141. Boorjian S, Vaughan ED Jr, Pitts WR Jr, Muecke EC. Victor Fray Marshall: twentieth century
renaissance urologist. J Urol. 2006;175:43–5.
142. Schmidlapp CJ, Marshall VF. The detection of cancer cells in the urine: a clinical appraisal
of the Papanicolaou method. J Urol. 1948;59:599–603.
143. Crabbe JG. Exfoliative cytological control in occupational cancer of the bladder. Br Med
J. 1952;2:1072–6.
144. Presti JC, Weyrauch HM. Papanicolaou examination of urine in diagnosis of urinary cancer.
I. As a test in mass screening. J Urol. 1955;73:430–4.
145. Mostofi FK, Sobin LH, Torloni H, World Health O. Histological typing of urinary bladder
tumours / F. K. Mostofi, in collaboration with L. H. Sobin, H. Torloni and pathologists in
fourteen countries. Geneva: World Health Organization, 1973.
146. Halling KC, King W, Sokolova IA, et al. A comparison of cytology and fluorescence in situ
hybridization for the detection of urothelial carcinoma. J Urol. 2000;164:1768–75.
147. Yafi FA, Brimo F, Auger M, Aprikian A, Tanguay S, Kassouf W. Is the performance of uri-
nary cytology as high as reported historically? A contemporary analysis in the detection and
surveillance of bladder cancer. Urol Oncol. 2014;32:27 e21–6.
148. Koss LG, Bartels PH, Bibbo M, Freed SZ, Taylor J, Wied GL. Computer discrimination
between benign and malignant urothelial cells. Acta Cytol. 1975;19:378–91.
149. Koss LG, Bartels PH, Bibbo M, et al. Computer analysis of atypical urothelial cells.
I. Classification by supervised learning algorithms. Acta Cytol. 1977;21:247–60.
150. Koss LG, Bartels PH, Sychra JJ, Wied GL. Computer analysis of atypical urothelial cells.
II. Classification by unsupervised learning algorithms. Acta Cytol. 1977;21:261–5.
151. Koss LG, Bartels PH, Sychra JJ, Wied GL. Computer discriminant analysis of atypical uro-
thelial cells. Acta Cytol. 1978;22:382–6.
152. Koss LG. Urinary cytology: the subjective diagnostic clues and their evaluation by computer.
Anal Quant Cytol. 1979;1:202–6.
153. Koss LG, Bartels PH, Sherman A, et al. Computer identification of degenerated urothelial
cells. Anal Quant Cytol. 1980;2:107–11.
154. Collste LG, Darzynkiewicz Z, Traganos F, et al. Cell-cycle distribution of urothelial tumour
cells as measured by flow cytometry. Br J Cancer. 1979;40:872–7.
155. Badalament RA, Hermansen DK, Kimmel M, et al. The sensitivity of bladder wash flow
cytometry, bladder wash cytology, and voided cytology in the detection of bladder carcinoma.
Cancer. 1987;60:1423–7.
156. Wojcik E, Miller C, O'Dowd G, Veltri RW. Value of computer-assisted quantitative nuclear
grading in differentiation of normal urothelial cells from low and high grade transitional cell
carcinoma. Anal Quant Cytol Histol. 1998;20:69–76.
157. Pambuccian SE. What is atypia? Use, misuse and overuse of the term atypia in diagnostic
cytopathology. J Am Soc Cytopathol. 2015;4:44–52.
13 The History of Urinary Cytology: The Long and Winding Road to Paris 2.0 315
158. Murphy WM, Soloway MS, Jukkola AF, Crabtree WN, Ford KS. Urinary cytology and blad-
der cancer. The cellular features of transitional cell neoplasms. Cancer. 1984;53:1555–65.
159. Koss LG, Bartels PH, Sychra JJ, Wied GL. Diagnostic cytologic sample profiles in patients
with bladder cancer using TICAS system. Acta Cytol. 1978;22:392–7.
160. Ooms EC, Veldhuizen RW. Cytological criteria and diagnostic terminology in urinary cytol-
ogy. Cytopathology. 1993;4:51–4.
161. Layfield LJ, Elsheikh TM, Fili A, Nayar R, Shidham V, Papanicolaou Society of C. Review
of the state of the art and recommendations of the Papanicolaou Society of Cytopathology
for urinary cytology procedures and reporting : the Papanicolaou Society of Cytopathology
Practice Guidelines Task Force. Diagn Cytopathol. 2004;30:24–30.
162. Rosenthal DL, Vandenbussche CJ, Burroughs FH, Sathiyamoorthy S, Guan H, Owens C. The
Johns Hopkins Hospital template for urologic cytology samples: part I-creating the template.
Cancer Cytopathol. 2013;121:15–20.
163. VandenBussche CJ, Sathiyamoorthy S, Owens CL, Burroughs FH, Rosenthal DL, Guan
H. The Johns Hopkins Hospital template for urologic cytology samples: parts II and III:
improving the predictability of indeterminate results in urinary cytologic samples: an out-
comes and cytomorphologic study. Cancer Cytopathol. 2013;121:21–8.
164. Piaton E, Decaussin-Petrucci M, Mege-Lechevallier F, Advenier AS, Devonec M, Ruffion
A. Diagnostic terminology for urinary cytology reports including the new subcategories
'atypical urothelial cells of undetermined significance' (AUC-US) and 'cannot exclude high
grade' (AUC-H). Cytopathology. 2014;25:27–38.
165. Wojcik EM. What should not be reported as atypia in urine cytology. J Am Soc Cytopathol.
2015;4:30–6.
166. Rosenthal DL, Wojcik EM, Kurtycz DFI. The Paris system for reporting urinary cytology.
Cham Heidelberg New York Dordrecht London: Springer International Publishing; 2016.
167. Bamforth J. Pioneer work by Professor Dudgeon in cytological diagnosis. J Clin Pathol.
1963;16:395.
168. Lord RVN. Norman Barrett, “Doyen of Esophageal Surgery”. Ann Surg. 1999;229(3):428–39.
fterword – Honoring the Past, Anticipating
A
the Future
authors of this book has been the most gratifying collaboration in my 50-year career.
They have devoted enormous time, energy, and expertise during the devastating
COVID pandemic, despite the demands of their roles as essential workers develop-
ing and providing new tests for patient care. Their dedication to our field is my
legacy. I couldn’t be more proud of them.
July 2021.
Dorothy L. Rosenthal, MD.
Emerita Professor of Pathology/Cytopathology.
The Johns Hopkins School of Medicine.
Baltimore, Maryland, USA
Chapters 1 and 9: Pathogenesis and ancillary tests are intertwined by their shared
molecular bases.
Chapter. 1: Pathogenesis
1. In TPS 1.0: Will the suspected distinction between the hyperplasia and dysplasia
pathways remain following more intense molecular studies?
2. In TPS 2.0: The molecular classification of The Cancer Genome Atlas (TCGA)
(2014) and subsequent studies largely upheld the distinction between the precur-
sors of Low-Grade Urothelial Carcinoma (LGUC) (TCGA Cluster I) and of
High-Grade Urothelial Carcinoma (HGUC) (predominantly TCGA Cluster II).
Development of invasive carcinoma from a LGUC remains rare.
3. In TPS 3.0: Further studies will refine the molecular basis of HGUC and identify
biomarkers with the potential for high clinical utility in AUC and SHGUC cases.
In order to relate morphologic classifications of cytology and histology to
defined molecular groups, large-scale detailed studies are needed. Results would
influence clinical management decision.
1. Prospectively compare the performance of novel tests on the sensitivity and neg-
ative predictive value of AUC and SHGUC categories.
2. Explore to what degree the correlation between morphological features and tar-
geted FISH analysis can improve the diagnostic skills of cytopathologists in
resolving AUC and SHGUC categories.
3. Investigate if the results of ancillary testing could be added to the defining crite-
ria of specific TPS categories in addition to morphological features.
4. Determine whether surveillance guidelines can be changed using currently
approved ancillary tests (e.g., U-FISH and uCyt) for patients with urothelial car-
cinoma depending on individual risk factors.
Afterword – Honoring the Past, Anticipating the Future 319
Chapters 2 and 10: Specimen adequacy and preparatory methods are aimed at
the same goal: providing a sample that is optimized to provide the platform on
which a reliable diagnosis can be based.
Chapter 2: Adequacy
1. In TPS 1.0: What is adequate volume and cellularity in the context of various
collection and preparation methods?
2. Questions answered in TPS 2.0: Volume recommendation for voided urine sam-
ples in the context of the adequacy algorithm.
3. Questions still unanswered for TPS 3.0: What is adequate cellularity in the con-
text of various collection and preparation methods?
Chapters 3 through 8: The diagnostic categories define the criteria for placing
cytomorphologic changes into logical and reproducible bins.
Chapter 3: Negative for HGUC (NHGUC)
1. Determine whether the same quantitative and qualitative criteria of the “suspi-
cious for HGUC” category should be used in different specimen types (upper
versus lower tract, voided versus instrumented).
2. Determine whether the association of the “suspicious for HGUC” and “positive
for HGUC” cytological categories with a histologically confirmed HGUC is sta-
tistically different enough to justify keeping those two categories separate.
1. Data are needed from more institutions regarding the performance of TPS for
upper tract lesions.
2. Determine whether upper tract lesions have any distinctly different cytomorpho-
logic features compared to their bladder counterparts, being careful to exclude
method of procurement (e.g., instrumentation) as a cause for these differences.
3. Define improved criteria for making a diagnosis of HGUC in the upper tract vs.
the bladder, particularly whether the ten-cell cut-off is appropriate for upper
tract HGUC.
1. Evaluate clinical data from major academic centers to assess the success of mor-
phology and immunohistochemistry for cytological detection of NUM of the
urinary tract.
2. Investigate application of innovative molecular and genetic tests to aid in the
identification of sources of non-urothelial cancers of the urinary tract as well as
determine the cell of origin in primary non-urothelial tumors.
3. Correlate cytologic diagnoses with clinical data to evaluate the clinical signifi-
cance of rendering a specific diagnosis of NUM.
4. Determine risk of malignancy associated with the diagnosis of NUM.
Afterword – Honoring the Past, Anticipating the Future 321
1. Assess ROHM for all categories based on the novel recommendations by the
second edition.
2. Compare new metrics such as likelihood ratios and pre-test probability (based on
clinical and epidemiological parameters) to classical ROHM, in order to assess
a more personalized post-test probability of having malignancy.
3. Assess potential bias in ROHM according to age, gender, and ethnical
distribution.
1. Since the first edition of TPS, clinical trials have been performed demonstrating
improved accuracy with fluorescence-assisted cystoscopy in bladder cancer
diagnoses and in performing a more complete resection, thereby decreasing
recurrences.
2. For the future:
• Plan clinical trials integrating these new techniques with cytology results.
• Explore the role of novel urine-based molecular marker assays to see how
these tests can be optimally integrated with cytology to enhance diagnostic
accuracy and surveillance.
Index