Indian Journal of Pathology and Microbiology
Home About us Instructions Submission Subscribe Advertise Contact e-Alerts Ahead Of Print Login 
Users Online: 1368
Print this page  Email this page Bookmark this page Small font sizeDefault font sizeIncrease font size

ORIGINAL ARTICLE Table of Contents  
Ahead of print publication
Quest to develop a standard screening method for urothelial carcinoma using liquid-based cytology (The Paris System) and CK20

1 Department of Pathology, IPGME&R, Kolkata, West Bengal, India
2 Department of Urology, IPGME&R, Kolkata, West Bengal, India

Click here for correspondence address and email

Date of Submission28-Jan-2022
Date of Decision09-Mar-2022
Date of Acceptance10-Mar-2022
Date of Web Publication06-Jun-2022


Background: Bladder cancer, the most common malignancy of the urinary tract is a leading cause of morbidity and mortality. But cystoscopy, which is till now the mainstay of screening, is an invasive, high-cost method with low sensitivity especially for flat lesions. Aim: To find a non-invasive and effective screening method with liquid-based cytology (LBC) using The Paris System (TPS) and CK20 immunocytochemistry. Materials and Methodology: It was a prospective study including the patients with clinical or cystoscopic diagnosis of urinary bladder space occupying lesions (SOL). Both conventional (CC) and liquid-based cytology slides were prepared from urine samples. Slides were evaluated by two trained pathologists and categorized according to TPS guidelines. CK20 immunocytochemistry (ICC) was also performed. Consequent formalin-fixed paraffin embedded sections were blindly examined by another pathologist and was taken as gold standard for comparison. Statistical Analysis: All the statistical analysis were done using MedCalc version 15.8 [Mariakerke, Belgium: MedCalc Software 2015]. Results: The study included 150 cases with a mean age of 62.4 years. Five cases revealed unsatisfactory smears. Rest of the cases were categorized as the following: 18.1% as NH-GUC, 8% as LGUN, 22.1% as AUC, 15.4% as SH-GUC, 32.9% as HGUC. Kappa value of CC and LBC were strong (0.854). LBC alone showed very low specificity (58%) and PPV (74.8%) which improved on application of ICC (specificity: 97.4%, PPV: 96.3%). Conclusion: We conclude that CK20 ICC offers potential for accurate, non-invasive detection and surveillance of bladder cancer and is a better tool when combined with liquid-based cytology, reported using The Paris System.

Keywords: CK20, ICC, LBC, Paris system, urothelial carcinoma

How to cite this URL:
Jain K, Datta C, Sengupta M, Pal DK, Chatterjee U. Quest to develop a standard screening method for urothelial carcinoma using liquid-based cytology (The Paris System) and CK20. Indian J Pathol Microbiol [Epub ahead of print] [cited 2023 Jan 29]. Available from:

   Introduction Top

Bladder cancer is known to be one of the most common cancers of the urinary tract and is the sixth-most common malignancy in men.[1] As per the guidelines by the American Urological Association (AUA), cystoscopy, urinary cytology and imaging are recommended for evaluation of hematuria. It is indeed fortunate that bladder cancer happens to be one of the malignancies which is amenable to an early intervention by means of transurethral resection and intravesical chemotherapy, before its progress into a high-stage disease.[2]

Sensitivity of conventional urinary cytology varies from 38%–84% for high-grade bladder tumor and carcinoma in situ but is much less in low grade tumors. Although liquid-based cytology (LBC) is extensively used in the screening of cervical cancer, its potential in evaluation of urothelial malignancy has not been widely explored.[3–6] Unlike the universally accepted Bethesda system for cervical cytology, there was no standardized reporting format for urinary cytology. The Paris system (TPS) evolved from a much-needed quest as a standard and reproducible triage which has been internationally acceptable for reporting of urinary cytology slides. Though introduced in the year 2013, its implementation is still limited.

Biomarkers have a great role to play in the cytological evaluation of bladder lesion, especially in high-risk patients and for monitoring of recurrent lesions. Various urinary biomarkers are described in literature, of which the quantitative BTA TRAK, qualitative BTA stat, NMP22, Urovision (FISH) and ImmunoCyt test are approved by US FDA.[7] Cytokeratin 20 (CK20) is a subtype of intermediate filament which is expressed in malignant cells of colon, ovary and umbrella cells of urothelium but not in the deeper layers of benign urothelium.[8–10] CK20 expression in non-umbrella cells is suggested to be an indicator of urothelial malignancy.[11]

The aim of the present study was to compare LBC and conventional cytology (CC) findings using TPS on urine samples. We also aimed to find out the threshold values of CK20 positive cells for optimal sensitivity and specificity in detection of bladder cancer, considering histopathology as the gold standard.

   Materials and Methods Top

It was a prospective, population-based, cross-sectional study performed in the Department of Pathology in collaboration with Department of Urology in a tertiary care hospital in India. The study was performed on the patients admitted with a clinical, imaging and/or cystoscopic suspicion of bladder neoplasm. Institutional ethical committee approval was taken.

Clinico-radiological findings and cystoscopy reports of patients were noted. 50 millilitres (ml) of mid-day urine samples were collected for at least two consecutive days and cytological examination was done on both CC and LBC prepared smears within four hours of collection. These slides were reported by two trained pathologists blindly using TPS.

Four basic steps were used for slide preparation: concentration, fixation, washing, and processing. The sample was centrifuged for 10 minutes at 600 g. 10 ml of BD cytorich was added to the sediment, vortexed and kept for a minimum of 30 minutes. The sample was again centrifuged, supernatant decanted and re-vortexed. When moderate-to-large pellet was identified, five drops of the sample were transferred to a 10-ml tube and 10 ml of distilled water was added. It was then re-centrifuged, supernatant decanted and homogenized. Finally, the centrifuge tube was loaded to BD PrepStain slide processor for processing. Papanicolaou staining was undertaken following the standard procedure and the slides were assessed for cellularity and morphology. Two slides for immunocytochemistry (ICC) staining were preserved at 4 degrees Celsius until stained.

ICC for CK20 was done using DAKO Monoclonal mouse Anti-human CK20 CloneKS20.8 LOT-20046170 antibody using DAB as chromogen. For validation, immunohistochemistry with CK20 on biopsy proven cases of high- and low-grade urothelial carcinoma were used.

Cytological assessment was done using The Paris System (TPS) for reporting of urine cytology. The diagnostic categories used were the following:[12]

  • Unsatisfactory
  • Negative for high grade urothelial carcinoma (NH-GUC)
  • Atypical urothelial cells (AUC)
  • Suspicious for high-grade urothelial carcinoma (SHGUC)
  • High-grade urothelial carcinoma (HGUC)
  • Low grade urothelial neoplasm (LGUN)
  • Others: primary/secondary malignancy.

For immunocytochemistry assessment, the cells with cytoplasmic staining for CK20 were counted in each slide. To compare sensitivity, we used two cut-off values (5% positive and 10% positive cells) while counting.

The transurethral resection of bladder tumor (TURBT) specimen was processed and evaluated blindly by two pathologists separately using WHO Classification of Urothelial Neoplasm. High-grade urothelial carcinoma, adenocarcinoma and malignant spindle cell neoplasm were considered as high-grade malignancy.

Risk of High-Grade Malignancy (ROHM) was calculated under each category.

ROHM = (Number of cases that turned out to be malignant in each category on Hp examination/No. of cases in each category on cytology) × 100%

Statistical analysis:

All the statistical analyses were done using MedCalc version 15.8 (Mariakerke, Belgium: MedCalc Software 2015).

   Results Top

The study included 150 cases with a mean age of 62.4 years. Out of 150 cases, 118 cases were male (79%) and 31 cases were female (21%).

Five cases revealed smears unsatisfactory for evaluation. Out of the rest 145 cases, as per conventional cytology, 29 cases (20%) as NH-GUC, 12 cases (8.27%) as LGUN, 32 cases (22%) as AUC, 19 cases (13.1%) as SH-GUC, 53 cases (36.5%) were reported as HGUC. During LBC evaluation, following diagnoses were made as per LBC: 27 cases (18.1%) as NH-GUC, 12 cases (8%) as LGUN, 33 cases (22.1%) as AUC, 23 cases (15.4%) as SH-GUC, 49 cases (32.9%) as HGUC. Calculated kappa values between LBC and CC values using TPS was 0.854 which implies a strong agreement between both the procedures (CC and LBC). [Table 1] shows our findings as per CC and LBC, CK20 (cases with >5% stained cells) staining and their corresponding histological diagnosis. One case was lost to follow up and consecutive biopsy was not obtained.
Table 1: Distribution of cases on CC and LBC along with their immunostain status and histological diagnosis

Click here to view

A total of five cases were reported as unsatisfactory. This category consisted of smears that had a sparse cell population, necrotic debris or dense hemorrhage in the background. On histopathology, these cases ranged from completely benign lesion like cystitis to malignant ones like low-grade urothelial carcinoma.

A total of 27 cases were placed in the category of negative for high-grade urothelial carcinoma (NHGUC). On histopathology, twenty cases were reported as benign amongst which only one case showed false positive CK20 staining on ICC. Six cases were categorized as premalignant to malignant on histology and one case has inadequate tissue for final diagnosis. ROHM in this category was 3.7%.

Cases showing presence of papillary clusters with fibrovascular cores having none to minimal atypia were categorized as Low grade Urothelial Neoplasm (LGUN). A total of 12 cases with majority of them (75%) diagnosed as completely benign were included in this category. ROHM in this category was 8.3% in our study.

Atypical urothelial cells (AUC) was the second-most commonly reported category after HGUC. The ROHM in this group was 30.3%, and 60.6% cases in this category were diagnosed histologically under premalignant-to-malignant category. CK20 ICC was positive in >95% of histologically diagnosed high-grade lesions of this category.

A total of 23 cases were reported as suggestive of high-grade urothelial carcinoma (SHGUC). Only four cases amongst all were diagnosed as benign, which included one case of endometriosis on histopathology. All the cases which fell in the malignant category showed positivity by CK20 immunostaining. ROHM in this category was calculated to be 60.86%.

Majority of cases (32.6%) were under the category of high-grade urothelial carcinoma (HGUC) in our study. These comprised of major number of cases (32.6%). All the cases in this category were histologically diagnosed as malignant with ROHM as 65%. On formalin fixed paraffin embedded tissue, majority (30 cases) were diagnosed as invasive urothelial carcinoma. Other categories included cases of non-invasive papillary LGUC, non-invasive papillary HGUC and one case each of adenocarcinoma and leiomyosarcoma. Photomicrographs of cytology, ICC and the final histopathology are illustrated in [Figure 1],[Figure 2],[Figure 3].
Figure 1: (a) LBC showing presence of cells with high N:C ratio, pleomorphic nuclei in clusters (HGUC). (Pap, ×100). (b) Cells showing positive cytoplasmic staining for CK20. (c) HPE showing invasive urothelial carcinoma involving lamina propria. Inset shows tumor cells around blood vessels. (H and E, ×40). (d) HPE showing muscle invasion by the tumor cells. Stage –pT2. Inset shows a higher magnification of the same. (H and E, ×100)

Click here to view
Figure 2: (a) Stained smear showing monomorphic cells arranged in papillae; category LGUN. (Pap, ×400). (b) Positive cytoplasmic staining of cells by CK20 immunostain. (c) Low power view showing non-invasive low-grade papillary urothelial carcinoma; Stage-pTa. (H and E, ×100). (d) High-power view of the same tumor. (H and E, ×400)

Click here to view
Figure 3: (a) Conventional cytology showing few dispersed atypical urothelial cells (arrow) in a background of red blood cells. (Pap, ×100). (b) Hpe showing non-invasive high-grade urothelial carcinoma with clear cells and necrosis; stage pTa (H and E, ×400). (c) Cells showing cytoplasmic positivity for CK20. (d) Hpe showing deep muscle sent separately shows no invasion by tumor cells. (H and E, ×400)

Click here to view

Cytological diagnosis as per TPS and its comparison with the gold standard histological diagnosis has been described in details in [Table 2].
Table 2: Distribution of cases and risk of high-grade malignancy in each Paris category

Click here to view

In our study, LBC alone had sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of 93.9%, 58%, 74.8%, and 87.8%, respectively. However, it was noteworthy that CK20 ICC showed significant increase in specificity, PPV, and NPV of 93.8%, 91.8%, and 96.4% respectively. Raising the threshold value of immunopositivity from 5% to 10% led to a reduction in number of false positives, increasing the specificity and PPV of the test further. The cases with >10% CK20 positive cells had specificity of 97.4% (92.5%–99.4%) and PPV of 96.3% (89.5%–99.2%) as opposed to specificity and PPV of 93.8% (87.6%–97.5%) and 91.8% (83.9%–96.6%) of the other category.

   Discussion Top

Bladder cancer is one of the few cancers with the best scope of early screening as it can help the patients undergo an early, complete treatment by means of transurethral resection of a bladder tumor (TURBT) and intra-vesical chemotherapy. At present, diagnosis and surveillance of bladder cancer requires imaging, cystoscopy, and cytological confirmation.[13] However, cells which are retrieved via urine samples are sparse and degenerated or may be present in small clusters, which is not always easy to interpret via morphology alone, especially by SurePath method where true mono layering is not achieved. Straccia et al.[14] concluded in his study that cases of AUC was better recognized by ThinPrep method. In addition, dire inter-observer variability and a lack of standardized reporting format complicate the matter further. Therefore, in the present scenario, there is a need for supplementary diagnostic tools.

LBC has the added advantage of using innovative technology like parallel processing of residual cytology material for cell block preparation, which may aid in improved diagnostic precision. Cell blocks also have the advantage of increasing the cellular yield and for serving archival purpose. In addition, cytology combined with ICC may have a great potential for evaluating post-BCG therapeutic effect and recurrence without imparting recurrent trauma.

Pastorello RG et al. has reviewed 23 studies describing the worldwide experience of TPS cytology reporting.[12] The distribution of cases among various categories by different studies ranged from 50.5% to 95.3% for NH-GUC, 1.2% to 23% for AUC, 0.2% to 6.6% for SHGUC and 2.2% to 14.1% for HGUC. In our study, HGUC comprised the majority of cases (32.6%) followed by AUC (22.1%), NHGUC (18%), SHGUC (15.3%) and LGUN (8%). Four percent of cases were unsatisfactory on cytology for opinion. The difference in different categories from other studies could be due to selection bias especially in limited cohorts. Implied risk of high-grade malignancy (ROHM) was also calculated in our study. High-grade urothelial carcinoma, adenocarcinoma, metastases and malignant spindle cell tumors were categorized as high-grade malignancy. Highest ROHM was 65.3% in HGUC category and lowest (3.7%) in NHGUC category. We found ROHM in SHGUC, AUC and LGUN category to be 60.86%, 30.3% and 8.3% respectively. Stanzione et al.,[15] found ROHM in different TPS reported categories as 14.4%, 44.3%, 88.9% and 97.4% in NHGUC, AUC, SHGUC and HGUC, respectively. The three largest studies showed ROHM to be more than 90% in HGUC categories. Although, our study had highest ROHM in HGUC category, it was still found to be less than most other studies as non-invasive papillary urothelial carcinomas also constituted a significant portion of final histopathologic examination.[15–17]

As a non-invasive, definitive and objective method, ICC on urinary samples has an important role to play. In our study, statistics shows that the application of ICC significantly improved the performance of screening (increased sensitivity and specificity to 100% and 95.4% respectively). Also, when the cut-off value of positivity was raised, there was a reduction in the false positive cases. These findings are significant and would require standardization on larger cohorts. CK20 expression is found in urothelial and gastro-intestinal epithelia, with its expression confined to superficial umbrella cells only.[9] Its extension into deeper cells is abnormal and is reported in carcinoma in situ (81%) and urothelial tumors (100%). Significantly, CK20 expression was undisturbed even in the presence of inflammation.[18–20] Wadhwa et al.[11] found positivity of CK20 in non-umbrella cells in cytologically inconclusive cases as potentially useful adjunct in the diagnosis of low-grade urothelial carcinoma.

[Table 3] demonstrates the comparative analysis of efficacy of cytology alone and CK20 immunostain with various prior published literature.[11],[21],[22],[23] Most of the studies were retrospective and immunocytochemistry was performed on biopsy proven cases of bladder cancer. As compared to other studies, sensitivity of urine cytology was higher (93.9%) in our study. Song Yi choi et al.[21] and M. Lenos et al.[22] performed dual immunostaining using p53 and CK20 on LBC slides. However, the sensitivity and specificity of CK20 immunostain alone in their study was higher than cytology alone.
Table 3: A comparative analysis of efficacy of cytology and CK20 immunostain with prior published studies

Click here to view

To our knowledge, no other study has endeavored to determine the effect of changing the cut-off values on performance of the tests. Also, we calculated the individual and combined sensitivity and specificity of cytology and ICC, and found the combined value to be greatly surpassing individual values. We recommend the validation of our findings in larger cohorts and further evaluation regarding the cut-off value of CK20+ cells to standardize the immunostain in the application of cytology.

Although, our study was limited by a small population size, lack of follow up and cost of the immunostains, the strengths lay in the well-defined study population and a cross-sectional study design, a population in each of which a histology diagnosis was available and also in the usage of the same sample for CC, LBC and ICC.

   Conclusion Top

Urinary bladder malignancies are notorious for their high recurrence rate and the need of prolonged follow up with multiple episodes of instrumentation. From this study, we conclude that combination of The Paris System for reporting of urinary cytology along with CK20 immunocytochemistry is an effective, non-invasive method for diagnosis and surveillance of urothelial carcinoma where it can avoid repeated trauma of instrumentation occurring during cystoscopy. It is also especially useful in routine screening to avoid missed diagnosis of flat lesions and in samples with low cellularity.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

   References Top

Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin 2015;65:87-108.  Back to cited text no. 1
Canter DJ, Revenig LM, Smith ZL, Dobbs RW, Malkowicz SB, Issa MM, et al. Re-examination of the natural history of high-grade T1 bladder cancer using a large contemporary cohort. Int Braz J Urol 2014;40:172-8.  Back to cited text no. 2
Raisi O, Magnani C, Bigiani N, Cianciavicchia E, D'Amico R, Muscatello U, et al. The diagnostic reliability of urinary cytology: A retrospective study. Diagn Cytopathol 2012;40:608-14.  Back to cited text no. 3
Lu DY, Nassar A, Siddiqui MT. High-grade urothelial carcinoma: Comparison of SurePath liquid-based processing with cytospin processing. Diagn Cytopathol 2009;37:16-20.  Back to cited text no. 4
Nassar H, Ali-Fehmi R, Madan S. Use of ThinPrep monolayer technique and cytospin preparation in urine cytology: A comparative analysis. Diagn Cytopathol 2003;28:115-8.  Back to cited text no. 5
Wright RG, Halford JA. Evaluation of thin-layer methods in urine cytology. Cytopathology 2001;12:306-13.  Back to cited text no. 6
Herman MP, Svatek RS, Lotan Y, Karakiewizc PI, Shariat SF. Urine-based biomarkers for the early detection and surveillance of non-muscle invasive bladder cancer. Minerva Urol Nefrol 2008;60:217-35.  Back to cited text no. 7
Srivastava R, Arora VK, Aggarwal S, Bhatia A, Singh N, Agrawal V. Cytokeratin-20 immunocytochemistry in voided urine cytology and its comparison with nuclear matrix protein-22 and urine cytology in the detection of urothelial carcinoma. Diagn Cytopathol 2012;40:755-9.  Back to cited text no. 8
Moll R, Löwe A, Laufer J, Franke WW. Cytokeratin 20 in human carcinomas. A new histodiagnostic marker detected by monoclonal antibodies. Am J Pathol 1992;140:427-47.  Back to cited text no. 9
Buchumensky V, Klein A, Zemer R, Kessler OJ, Zimlichman S, Nissenkorn I. Cytokeratin 20: A new marker for early detection of bladder cell carcinoma? J Urol 1998;160:1971-4.  Back to cited text no. 10
Wadhwa N, Diwaker P, Lotha N, Arora VK, Singh N. Cytokeratin 20 immunocytochemistry on urine sediments: A potential low-cost adjunct to cytology in the diagnosis of low-grade urothelial carcinoma. Cytopathology 2017;28:531-5.  Back to cited text no. 11
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.  Back to cited text no. 12
Ro JY, Staerkel GA, Ayala AG. Cytologic and histologic features of superficial bladder cancer. Urol Clin North Am 1992;19:435-53.  Back to cited text no. 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.  Back to cited text no. 14
Stanzione N, Ahmed T, Fung PC, Cai D, Lu DY, Sumida LC, et al. The continual impact of the Paris System on urine cytology, a 3-year experience. Cytopathology 2020;31:35-40.  Back to cited text no. 15
de Paula R, Oliveira A, Nunes W, Bovolim G, Domingos T, De Brot L, et al. Two-year study on the application of the Paris system for urinary cytology in a cancer centre. Cytopathology 2020;31:41-6.  Back to cited text no. 16
Meilleroux J, Daniel G, Aziza J, d'Aure DM, Quintyn-Ranty ML, Basset CM, et al. One year of experience using the Paris system for reporting urinary cytology. Cancer Cytopathol 2018;126:430-6.  Back to cited text no. 17
Bhatia A, Dey P, Kumar Y, Gautam U, Kakkar N, Srinivasan R, et al. Expression of cytokeratin 20 in urine cytology smears: A potential marker for the detection of urothelial carcinoma. Cytopathology 2007;18:84–6.  Back to cited text no. 18
Harnden P, Allam A, Joyce AD, Patel A, Selby P, Southgate J. Cytokeratin 20 expression by non-invasive transitional cell carcinomas: Potential for distinguishing recurrent from non-recurrent disease. Histopathology 1995;27:169-74.  Back to cited text no. 19
McKenney JK, Desai S, Cohen C, Amin MB. Discriminatory immunohistochemical staining of urothelial carcinoma in situ and non-neoplastic urothelium: An analysis of cytokeratin 20, p53, and CD44 antigens. Am J Surg Pathol 2001;25:1074-8.  Back to cited text no. 20
Choi SY, Kim KH, Suh KS, Yeo MK. Diagnostic significance of dual immunocytochemical staining of p53/cytokeratin20 on liquid-based urine cytology to detect urothelial carcinoma. Cytojournal 2020;17:3.  Back to cited text no. 21
Lenos M, Katafigiotis I, Vrettou K, Papaioannou D, Malta F, Trigka EA, et al. Evaluation of double immunocytochemical staining for CK20 and P53 as a potential adjunct to cytology for urothelial cancer diagnosis. Cytopathology 2017;28:96-102.  Back to cited text no. 22
Li HX, Li M, Li CL, Ma JH, Wang MR, Rao J, et al. ImmunoCyt and cytokeratin 20 immunocytochemistry as adjunct markers for urine cytologic detection of bladder cancer: A prospective study. Anal Quant Cytol Histol 2010;32:45-52.  Back to cited text no. 23

Correspondence Address:
Moumita Sengupta,
Flat No A11, Snehaneer Apartment, 1185 Chakgaria, Kolkata - 700094, West Bengal
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/ijpm.ijpm_95_22


  [Figure 1], [Figure 2], [Figure 3]

  [Table 1], [Table 2], [Table 3]


 Article in PDF
     Search Pubmed for
    -  Jain K
    -  Datta C
    -  Sengupta M
    -  Pal DK
    -  Chatterjee U

    Materials and Me...
    Article Figures
    Article Tables

 Article Access Statistics
    PDF Downloaded30    

Recommend this journal