# The Diagnostic Accuracy of the Paris System for Reporting Upper Urinary Tract Cytology: The Atypical Urothelial Cell Conundrum

## TL;DR

This study evaluates how well a classification system can diagnose upper urinary tract tumors using cell samples, finding it effective for aggressive tumors but less reliable for ambiguous cases.

## Contribution

The study provides new insights into the diagnostic accuracy of the Paris System for upper urinary tract cytology, particularly highlighting limitations in classifying atypical urothelial cells.

## Key findings

- The Paris System (TPS 2.0) has high specificity and diagnostic accuracy for high-grade urothelial carcinoma.
- Atypical urothelial cells (AUC) have a low positive predictive value, complicating accurate diagnosis.
- TPS 2.0 is effective in excluding malignancies in non-high-grade urothelial carcinoma cases.

## Abstract

Tumours that arise in the urinary system above the level of the urinary bladder are difficult to diagnose through radiology alone. Endoscopic examination, together with taking a cellular sample or tissue sample, is required to confirm whether there is a tumour. This study aims to determine both the diagnostic accuracy of cellular samples compared to tissue biopsies and how frequently they can correctly predict the presence or absence of cancer using a recently proposed classification system. We found that the classification system works well in terms of excluding the presence of tumours and in confirming the presence of aggressive tumours, but that it is less reliable in differentiating reactive changes due to inflammation from less aggressive tumours. Our study thus adds to knowledge about how best to use cellular samples and their classification system.

Background: The diagnosis of upper urinary tract urothelial carcinoma (UTUC) in cytological specimens is challenging, particularly the designation of atypical urothelial cells (AUC). The application of the Paris System for Reporting Urinary Cytology (TPS) has improved the performance of lower tract urothelial carcinoma specimens but has shown variable results in upper tract specimens, which are frequently instrumented. Methods: This retrospective study analysed upper tract selective cytology samples from January to December 2023. Samples were classified under TPS 2.0 categories. Histological specimens were used where available as the gold standard to calculate statistical metrics including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Results: Out of 122 samples, 12.2% were considered non-diagnostic or insufficient, with 13.1% designated as Non-high-grade urothelial carcinoma (N-HGUC), 53.3% as atypical urothelial cells (AUC) and 21.3% as positive/suspicious for HGUC. Histopathological correlation was available for 48.7% of cases. The risk of malignancy was: NHGUC (0%), AUC (47%) and HGUC (77.7%). The highest PPV was for HGUC (78%), with a diagnostic accuracy of 81.3% and specificity of 88%. In contrast, AUC had a PPV of 47% in instrumented and non-instrumented samples, rising to 58% in combination with the HGUC category. Conclusion: TPS 2.0 is an effective tool with excellent diagnostic accuracy for HGUC and in excluding malignancies in the N-HGUC category, but in our hands, the high rates of the AUC category, together with the low PPV, remain a major challenge and an obstacle to the correct stratification of patients with UTUC.