The Paris system for Reporting Urinary Cytology: The quest for standardized

terminology

Eva M. Wojcik, MD

Chair and Professor of Pathology and Urology

Loyola University, Chicago, IL, USA

Outlines

• What is the goal of urine cytology?

• Why to standardize, why Paris?

• What is the guiding principle?

• What are diagnostic categories?

• What are the criteria?

• What adjuvant studies?

• What are future clinical and research needs?

The main purpose of urine cytology

To detect bladder cancer

Bladder cancer - current status

• ~ 80,470 new cases in 2019 in the USA (ACS) • ~ 17,670 deaths due to bladder cancer • Average age at dx - 73 • 4th most common ca in men and 9th in women (1 in 27 men, 1 in 89

women) • 9th most common cause of cancer death (F>M) • ~ 75% non-muscle invasive bladder cancers (superficial bladder

cancers), Ta, Tis, T1 • ~ 30% - 70% - recurrence • ~ 5% - 15% - progression (<1% LG Ta*) • > 535,000 people in the US are survivors of this cancer • Highest per patient cost from dx to death of all cancers • $4.1 billion/year spent to tx bladder cancer

*Nielsen ME et al. Trends in stage-specific incidence rates for urothelial carcinoma of the bladder in the United States: 1998 to 2006. Cancer 2014:120:86

Classifications

WHO 1973

WHO/ISUP 2004

Papilloma

Papilloma

Grade I Grade III Grade II

Low Grade High Grade PUNLMP

~ 80-90% ~ 10-20% ~ 50-60%

URINE CYTOLOGY SENSITIVITY

5 Very high probability that we are going to be wrong

• Reproducibility

• Improvement of communication

• Atypical cells

– Wide intraobserver variability

• Nationally rates of atypical vary among institutions

– Range from 2% to 30% (51% atypical + suspicious)

Why to standardize reporting of urinary cytology?

Where did we start?

• 18th International Congress of Cytology, Paris, May, 2013

– “Paris Group” – all participants of two Urine Cytology Symposia

– Outline of the Paris System for Reporting Urinary Cytopathology

– Ultimate goal – detection of HGUC

• Sponsorship by the ASC and IAC

• Contract with Springer

• Numerous face-to-face meetings

I. Pathogenesis of Urothelial Carcinoma II. Adequacy III. Negative for High Grade Urothelial Carcinoma IV. Atypical Urothelial Cells V. Suspicious for High Grade Urothelial Carcinoma VI. High Grade Urothelial Carcinoma VII. Low Grade Urothelial Neoplasm VIII. Other malignancies, both primary and secondary IX. Ancillary Studies X. Clinical management XI. Preparatory techniques relative to Urinary Tract

samples

The Paris Working Group consisted of 49 members, 28 from 12 US states, and 21 from 9 countries including Canada, France, Italy, Japan, Korea,

Luxembourg, Slovenia, Switzerland, and the United Kingdom.

System has to be build based on:

• Consensus • Evidence • Inclusion • Acceptance • Understanding

Urothelial Carcinoma

Normal Urothelium

Hyperplasia Dysplasia

Low Grade Carcinoma High Grade Carcinoma Carcinoma in situ

Invasive Carcinoma

9p-, 9q- p16

Genetically Stable FGFR3 (~85%)

Genetically Unstable p53 (~60%)

<10%

Recurrence Recurrence

RAS (?)

Pathogenesis of Urothelial Carcinoma

Eva M. Wojcik and Stefan E. Pambuccian

Normal Urothelium

Hyperplasia Dysplasia

High Grade Carcinoma Carcinoma in situ

Invasive Carcinoma

Papillary Pathway

80%

Non-Papillary Pathway

20% 9p-, 9q-

p16

Genetically Unstable p53 (~60%)

<10%

Bladder cancer – more then one disease?

• ~ 75 % Non-Muscle-Invasive (Ta/T1)

– Good prognosis

– Recurrence

– 10%-15% progression (LG Ta - <1%)*

• ~ 25 % Muscle-Invasive (> T2)

– >60% overall survival

*Nielsen ME et al. Trends in Stage-Specific Incidence Rates for Urothelial Carcinoma of the Bladder In the United States: 1998-2006. Cancer 2014:120:86

“Approximately 80% (of Ta bladder tumors) appear to follow a benign course without developing invasive tumors or dying of bladder cancer”

Question…. “Carcinoma”?

GU GI

CARCINOMA

ADENOMA

Question…. “Carcinoma”?

Mr. Smith - You have a bladder cancer

What really matters?

High Grade Urothelial Carcinoma

Diagnostic Categories

Positive Negative Atypical/Suspicious

HGUC Everything else

Hope

Reality

Evolution of the Classification

Owens et al. Cancer Cytopathology 2013

?LG

?HG

NEW paradigm

• It is all about High Grade Urothelial Carcinoma

• Negative for High Grade Urothelial Carcinoma

• AUC SHGUC HGUC

• LGUN – Low Grade Urothelial Neoplasm

Quality and Quantity Quantity

The Process

• INCLUSION – International Working Group consisted of 49 members, 28 from

12 US states, and 21 from 9 countries – Open forum hosted by ASC and IAC seeking an international

input

• EVIDENCE – Review of existing literature – Design and publish studies that address the issue

• CONSENSUS – Numerous face-to-face meetings – Regular conference calls – Daily (almost) emails – Final approval of the manuscript by the group

Adequacy of Urine Specimens (Adequacy)

Matthew T. Olson , Güliz A. Barkan , Monique Courtade-Saïdi , Z. Laura Tabatabai , Yuji Tokuda , Toyonori Tsuzuki , and Christopher J. VandenBussche

• Presence of atypical or malignant cells

• Specimen type – Instrumented (Cellularity,

2600 cells, 2 urothelial cells/10HPF) (*)

– Voided (>30mL more likely “adequate”) (**)

• Obscuring elements (blood, lubricant, etc.)

(*) Prather J, Arville B, Chatt G, et al. Evidence-based adequacy criteria for urinary bladder barbotage cytology. Journal of the American Society of Cytopathology.4: 57-62. (**) 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 carcinoma. Cancer Cytopathol. 2015.

Renshaw AA, Gould EW: Evidence-based adequacy criteria for instrumented urine cytology using cytospin preparations. Diagn Cytopathol 2018

Cytospin preparations of instrumented urine cytology specimens with less than 10 urothelial cells or more than 50 urothelial cells/10 hpfs are both associated with significantly increased false negative rates compared to cases with 10-

49 urothelial cells/10 hpfs.

Positive Negative Atypical Suspicious Positive Negative Atypical Suspicious

Reactive Umbrella Cells

Positive Negative Suspicious Atypical

• Cells - large, often bi- or multinucleated, N/C ratio - low, Cell boarder - often scalloped and sharply demarcated

• Nuclei - centrally located, nuclear membrane is smooth and chromatin is fine. Occasional chromocenters

• Nuclei – occasionally clumped and degenerated – PITFALL – May contain abnormal DNA

Dx. Negative for High Grade Urothelial Carcinoma

Squamous and Glandular Cells

• Squamous cells – GYN tract, trigone, metaplasia, dysplasia • Glandular epithelium – cystitis glandularis, metaplasia, prostatic glandular cells, seminal vesicle cells • Renal tubular cells

Positive Negative Atypical

Dx. Negative for High Grade Urothelial Carcinoma

Squamous dysplasia – Sq papilloma

Squamous dysplasia – LSIL

Mucinous metaplasia

Mucinous metaplasia

Renal tubular cells

Seminal vesicle cells

Suspicious

Benign Urothelial Tissue Fragments – BUTF and Low Grade Urothelial Neoplasia - LGUN

• Instrumentation, stones

• Can be present in voided urines after DRE

• LGUN – Fibrovascular cores

• Second line diagnosis after the NHGUC

Positive Negative Atypical

Dx. Negative for High Grade Urothelial Carcinoma

Suspicious

Treatment/Procedure Effects

Positive Negative Atypical

Dx. Negative for High Grade Urothelial Carcinoma

S/P Brachytherapy S/P XRT

S/P TURBT – Cautery artifact S/P Cystectomy - neobladder S/P BCG immunotherapy - granuloma

S/P Cystoscopy – voided urine

Suspicious

Polyoma Virus

Positive Negative Atypical

Dx. Negative for High Grade Urothelial Carcinoma

Suspicious

“Negative, NOT atypia”

Wojcik EM: What should not be reported as atypia in urine cytology: JASC 2015;4;3;30-36

Negative for High-Grade Urothelial Carcinoma (Negative)

Dorothy L. Rosenthal, Michael B. Cohen, Hui Guan, Christopher L. Owens, Yuji Tokuda, and

Eva M. Wojcik

Definition of Negative for High-Grade Urothelial Carcinoma

• A sample of urine, either voided or instrumented, may be considered benign, i.e., NHGUC, if any of the following components are present in the specimen:

– Benign urothelial, glandular, and squamous cells

– Benign urothelial tissue fragments (BUTF) and urothelial sheets or clusters

– Changes associated with lithiasis

– Viral cytopathic effect; polyoma virus (BK virus—decoy cells)

– Post-therapy effect, including epithelial cells from urinary diversions

Umbrella cells

BUFT

Ilial conduit Umbrella cells and basal/intermediate cells Polyoma virus

BCG - Granuloma

Ilial conduit

LSIL – GYN contamination

Positive Suspicious Atypical Negative

What is Atypia ?

Findings in literature

1. High nuclear cytoplasmic ratio (>0.7)

2. Nuclear hyperchromasia

3. Coarse, clumped chromatin

4. Irregular nuclear membranes

Atypia Suspicious Positive

Atypical Urothelial Cells

• Non-superficial and non-degenerated urothelial cells with an high N/C ratio > 0.5 (required)

and one of the following: • Hyperchromasia (compared to the umbrella cells or

the intermediate squamous cell nucleus) • Irregular clumpy chromatin • Irregular nuclear contours

Positive Negative Suspicious Atypical

Dx. Atypical Urothelial Cells

Suspicious for High Grade Urothelial Carcinoma

• Non-superficial and non-degenerated urothelial cells with an high N/C ratio > 0.7 (required)

• Hyperchromasia (compared to the umbrella cells or the intermediate squamous cell nucleus) (required)

and one of the following: • Irregular clumpy chromatin • Irregular nuclear membranes

Positive Negative Suspicious Atypical

Dx. Suspicious for High Grade Urothelial Carcinoma

Positive vs. Suspicious for High Grade Urothelial Carcinoma

• “The number of atypical urothelial cells is an

important criterion to classify urine cytology specimens into the ‘positive’ or the ‘suspicious’ categories. A cut-off number of >10 cells to render a definitive diagnosis of HGUCA seems valid from the clinical standpoint .”

Positive Negative Suspicious Atypical

Dx. High Grade Urothelial Carcinoma

JASC 2015;4(4)232–238

5 – 10 cells – gray zone, based on experience, history, individual threshold, etc

High-Grade Urothelial Carcinoma (HGUC) Momin T. Siddiqui, Guido Fadda, Jee-Young Han, Christopher L. Owens,

Z. Laura Tabatabai, and Toyonori Tsuzuki

• Cellularity: At least 5–10 abnormal cells

• N/C ratio: 0.7 or greater

• Nucleus: Moderate to severe hyperchromasia

• Nuclear membrane: Markedly irregular

• Chromatin: Coarse/clumped

Other Notable Cytomorphologic Features

• Cellular pleomorphism

• Marked variation in cellular size and shapes, i.e., oval, rounded, elongated, or plasmacytoid (Comet cells)

• Scant, pale, or dense cytoplasm

• Prominent nucleoli

• Mitoses

• Necrotic debris

• Inflammation

Suspicious HGUC AUC

This could be a HGUC I think, this is a HGUC

I know, this is a HGUC

Positive Negative Suspicious Atypical

TPS – didn’t eliminate the GRAY ZONE – it DEFINED it!

What happened to LGUC??

• Almost Practically impossible to diagnose without a mini-biopsy with fibrovascular core

• Cytologically normal nuclei

• Is it truly a carcinoma?

• More common than HGUC

• BUT, not life threatening

Low-Grade Urothelial Neoplasia (LGUN) Eva M. Wojcik, Tatjana Antic, Ashish Chandra, Michael B. Cohen, Zulfia McCroskey,

Jae Y. Ro, and Taizo Shiraish

• LGUN - combined cytologic term for low grade papillary urothelial neoplasms (LGPUN) (which include urothelial papilloma, PUNLMP and LGPUC) and flat, low grade intraurothelial neoplasia

LGUC LGUN

Cytologic Criteria of Low Grade Urothelial Neoplasia (LGUN) (regardless of the specimen type: voided or

instrumented):

• Three-dimensional cellular papillary clusters (defined as clusters of cells with nuclear overlapping, forming "papillae") with fibrovascular cores with capillaries

Nucle

ar

/ cyto

logic

aty

pia

Probability of high grade UC

low moderate/high certain

AUC/SHGUC

8%-30%

HGUC NFHG

Implementation

Barkan G et al: The Paris System for Reporting Urinary Cytology: The quest to develop a standardized terminology Acta Cytol. 2016;60(3):185-97; Adv Anat Pathol. 2016;23(4):193-201; JASC. 2016;5, 177-188

Visual aids for diagnostic criteria in the lab:

• In the sign out/fellows room

•In the cytotech screening room

%AUC (blue) and %SUSP (red) at LUMC 2008-2016

0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

12.00%

14.00%

2008 2009 2010 2011 2012 2013 2014 2015 2016

9.28% 10.08%

13.57%

10.16%

6.57%

9.73%

11.05%

10.01%

6.35%

Barkan et al.

Rate of Atypia at Loyola per pathologist

0.00%

5.00%

10.00%

15.00%

20.00%

25.00%

30.00%

35.00%

2008 2009 2010 2011 2012 2013 2014 2015 2016

Barkan et al.

%AUC and %SUSP before and After TPS Implementation at LUMC

0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

12.00%

14.00%

16.00%

Non-Paris CP Pre Non-Paris CP Post Paris CP Pre Paris CP Post

15.08%

11.82%

8.19%

5.94%

AUC

SUSP

Non-Paris CP Post vs pre %AUC, p<0.05 Paris CP Post vs pre %AUC, p<0.05

Were We Able To Reduce The Atypia Rate? Barkan et al.

• 5 institutions; pre-TPS - 15,589, post-TPS - 15,311

Mean Pre-Paris

Mean Post-Paris

P value

UNSAT 0.47% 0.59% 0.16

NEG/NHGUC 82.20% 85.59% <0.0001

ATYPICAL/AUC 13.49% 9.40% <0.0001

SUSPICIOUS/ SHGUC 2.53% 2.04% 0.05

POSITIVE/ HGUC 2.26% 2.34% 0.68

LGUN 0.00% 0.01% n/a

OTHER MALIGNANCIES 0.05% 0.02% n/a

Pre- TPS Post -TPS Pre- TPS

Post -TPS

Pre- TPS

Post -TPS

Pre- TPS

Post -TPS

NHGUC NR NR 64.3% 70.7% 64.9% 66.1% 75.4% 80.0%

AUC 39% 26% 29.5% 21.6% 23.9% 23.0% 18.6% 14.4%

SHGUC NR NR 3.3% 4.4% 5.8% 4.5% 3.0% 2.4%

HGUC NR NR 2.9% 3.0% 3.8% 5.0% 3.0% 3.2%

Hassan et al. VanderBussche et al. Torous et al. Wang et al.

Cowan and VanderBussche, Cancer Cytopatol 2018;7;185

Author Journal Year No Cases Pre-Paris Post-Paris

Long T. et al. Cytojournal 2017 357 N/A 22%

Granados R. et al. Acta Cytol 2017 149 6.60% 15.80%

Suh J. et al. Cytopathology 2017 142 25.30% 14.80%

Malviya K. et al. Acta Cytol 2017 176 11.90% 5.30%

Torous VF et al. JASC 2017 2295 29.50% 20.10%

Roy M. et al. Cytopathology 2017 225 41.20% 11.30%

Wang Y. et al. Cancer Cytol 2017 4764 18.60% 14.40%

Hassan et al. AJCP 2016 124 39% 26%

Post-Paris publications - AUC

Barkan et al.

Challenges and Shortcomings

• N/C Ratio

– Overestimated

– Underestimated

– High interobserver variability

• Degeneration

• Fibrovascular cores

Why >0.5

Digital Image Analysis Supports a Nuclear-To-Cytoplasmic Ratio Cutoff Value of 0.5 for Atypical Urothelial Cells Jen-Fan Hang, MD; Vivek Charu, PhD; M. Lisa Zhang, MD; and Christopher J. VandenBussche, MD, PhD

Cancer Cytopathol 2017;125:710-6.

N:C threshold 0.486

Digital Image Analysis Supports a Nuclear-To-Cytoplasmic Ratio Cutoff Value of 0.5 for Atypical Urothelial Cells Jen-Fan Hang, MD; Vivek Charu, PhD; M. Lisa Zhang, MD; and Christopher J. VandenBussche, MD, PhD

Cancer Cytopathol 2017;125:710-6.

Challenges – N/C ratio

SHGUC + HGUC

AUC

Zhang ML et al. Cancer Cytopathol 2016;124(9):669-677

Basal cells

Challenges – N/C ratio – Many faces of HGUC

HGUC - is N:C > 0.7 too high?

• Great pleomorphism

• Squamous differentiation

N/C ratio in UTUC and LTUC Patrick McIntire et al. USCAP 2019

Conclusions: • The N:C ratios for HGUC (0.57) and SHGUC (0.53) categories are lower

than those previously suggested in TPS – should the current threshold be lowered?

• At the same time – all cases contained cells with N:C ratio > 0.7

Shortcoming - Degeneration

Challenges – Fibrovascular cores

Fibrovascular cores – very rarely seen LGUN – separate diagnostic entity?

HGUC have fibrovascular cores too!

What to do if we suspect LGUN on cytology?

LGUN may be considered in correlation with cystoscopic or biopsy findings First Line Diagnosis - NHGUC

• Three-dimensional cellular clusters without fibrovascular cores

• Increased numbers of monotonous single (non-umbrella) cells

Negative for HGUC Comment – Suggestive of LGUN

Purpose of Urine Cytology

Follow up patients with history of urothelial carcinoma

Risk of malignancy Category Risk of

Malignancy Management

Unsatisfactory/Nondiagnostic ? (<5%) Repeat cytology, cystoscopy in 3 months if increased clinical suspicion

Negative for HGUC 0-6% Clinical follow up as needed

Atypical Urothelial Cells (AUC) 8-35% Clinical follow up as needed. Use of ancillary testing.

Suspicious for HGUC 50-90% More aggressive follow up, cystoscopy, biopsy

LGUN ~10% Need biopsy to further evaluate grade and stage

High Grade UC >90% More aggressive follow up, cystoscopy, biopsy, staging

Other malignancy >90% More aggressive follow up, cystoscopy, biopsy, staging

Take home message

• HGUC – this is the one that matters – Negative for HGUC

• Atypia rate has been decreasing since the introduction of TPS but it is a slow and challenging process

• LGUN – new diagnostic category or under NHGUC?, based on presence of fibrovascular cores; HGUC also have F/V cores

• N/C Ratio – the most significant challenge • Degeneration – need to be better addressed

TPS

Atypia