Mp38-04 the Role of Bladder Epicheck in Patients With Atypical Urothelial Cells in the Voided Urinary Cytology

Abstract

INTRODUCTION AND OBJECTIVE: Continuous follow-up after the treatment of non-muscle invasive bladder cancer (NMIBC) is mandatory for early recurrence detection. This study investigates the clinical relevance of new diagnostic tools such as an mRNA-based urine test (XpertÓ Bladder Cancer Monitor, Cepheid, USA) and Narrow Band ImagingÓ (NBI, Olympus) and compares them to the established follow-up methods.METHODS: This was a prospective single-center study involving patients undergoing NMIBC screening with enrollment from January 2018 to March 2020; the follow-up concluded in January 2021. Aside from performing the standard of care, including conventional white light cystoscopy (WLC), cytology, and sonography, patients also submitted urine for the XPERTÓ urine test and underwent NBI cystoscopy. These latter two tests were compared against WLC and the anatomic pathology in cases of NMIBC recurrence. Follow-up data were included in cases in which (a) WLC findings were ambiguous and pathology was postponed, or (b) in instances where XPERTÓ yielded a positive result while WLC was negative. According to the manufacturer, the XPERTÓ cut-off value (COV) for a positive result was set at a linear discriminant analysis (LDA) value of 0.5. The sensitivity (SE), specificity (SP), positive an negative predictive values (PPV, NPV) and median LDA values for tumor negative and tumor positive low-(LG) and highgrade (HG) lesions were calculated. SE and SP for different XPERTÓ LDA thresholds were also calculated.RESULTS: 301 patients were enrolled during the study period. 49 patients demonstrated NMIBC disease recurrence. In this subset, NBI cystoscopy was congruent with WLC in all patients. Cytology harbored an SE and SP of 27% and 97% (PPV: 65%; NPV 0.87%), respectively, whereas with XPERTÓ, the SE and SP were 58% and 89% (PPV: 51%; NPV: 92%; AUC: 0.79 (0.76-0.871)). Assessment of higher XPERTÓ LDA thresholds elicited higher SE and accordingly lower SP (for e.g., when the LDA was raised to 0.76, the SE was 98% and SP was 40%; when LDA was further raised to 1.0, the SE was 99% while SP was 23%). The overall median LDA values in a tumor negative setting was 0.33 (0.17 e 0.45); in an overall tumor positive bladder, the median LDA was 0.56 (0.35 e 0.95), while LG disease exhibited an LDA of 0.38 (0.26 e 0.52) and HG disease was 0.85 (0.51 e 1.06) (p < 0.001). A subgroup analysis revealed better SE but similar SP (PPV, NPV) for HG compared to LG recurrence, where SE was 74% and SP of 89% (39%, 97%) vs. SE of 33%, and SP of 89% (21%, 94%).CONCLUSIONS: NBI cystoscopy does not necessarily offer any additional benefit to standard WLC, and WLC remains the gold standard for visualization in NMIBC surveillance. Yet, the XPERTÓ Bladder Cancer Monitoring platform may confer better sensitivity and a diagnostic advantage in high grade, but not low grade, disease recurrence.