Evaluation of Diagnostic Accuracy of the Paris System (TPS 2.0) in Urine Cytology Specimens: An Institutional Experience From a Large Cohort of a Tertiary Care Centre.

Objective: The objective of this study is to evaluate the diagnostic performance of urine cytology using The Paris System (TPS 2.0) in comparison with TPS 1.0, and the Four-Tier Reporting System (FTRS) of our institute for identifying high-grade urothelial carcinoma (HGUC).
Methodology: A total of 789 urine cytology specimens from 240 patients including 12 UTUC cases with available histological and clinical details were included. Two hundred twenty-two cases were newly diagnosed, whereas 18 were recurrent tumours. Histopathological evaluation categorised the cases as non-neoplastic (33%, 13.7%), low-grade urothelial neoplasms (LGUNs) (94%, 39.2%), high-grade urothelial carcinoma (HGUCs) (110%, 45.8%) and other malignancies (3%, 1.3%).
Results: TPS 2.0 categorised the cases as 14 (5.7%) non-diagnostic or ND/U, 99 (41.5%) NHGUC, 37 (15.4%) AUC, 24 (10%) SHGUC and 66 (27.5%) HGUC. TPS 1.0 had 14 (5.7%) ND/U, 72 (30%) NHGUC, 61 (24.5%) AUC, 3 (1.3%) LGUC, 24 (10%) SHGUC and 66 (27.5%) HGUC. FTRS classified them as 19 (7.5%) ND/UNS, 44 (18.3%) NEG, 88 (36.7%) INC and 89 (37.1%) POS. The ROHM for TPS 2.0 was 71.4% for ND/U, 12.1% for NHGUC, 29.7% for AUC, 79.2% for SHGUC and 89.4% for HGUC. TPS 1.0 showed a similar ROHM for ND/U, SHGUC and HGUC, whereas had 13.8% for NHGUC and 19.7% for AUCs. FTRS, had 78.9% for UNS, 6.8% for NEG, 35.2% for INC and 73.1% for POS. TPS demonstrated a sensitivity of 70.91% and specificity of 90.77% for identifying HGUC, whereas FTRS showed 59.09% sensitivity and 81.54% specificity. Also, TPS was found to be 75% accurate with 62.5% sensitivity and 100% specificity for UTCC cases separately.
Conclusion: TPS 2.0 exhibits diagnostic accuracy with better performance in comparison to FTRS, making it a more reliable system for clinical practice. Our findings endorse the utility of TPS 2.0 in improving the accuracy of urine cytology in predicting histological diagnosis of HGUC.
(© 2024 John Wiley & Sons Ltd.)