Artificial intelligence to improve cytology performance in urothelial carcinoma diagnosis: results from validation phase of the French, multicenter, prospective VISIOCYT1 trial.

PURPOSE: Cytology and cystoscopy, the current gold standard for diagnosing urothelial carcinomas, have limits: cytology has high interobserver variability with moderate or not optimal sensitivity (particularly for low-grade tumors); while cystoscopy is expensive, invasive, and operator dependent. The VISIOCYT1 study assessed the benefit of VisioCyt® for diagnosing urothelial carcinoma. METHODS: VISIOCYT1 was a French prospective clinical trial conducted in 14 centers. The trial enrolled adults undergoing endoscopy for suspected bladder cancer or to explore the lower urinary tract. Participants were allocated either Group 1: with bladder cancer, i.e., with positive cystoscopy or with negative cystoscopy but positive cytology, or Group 2: without bladder cancer. Before cystoscopy and histopathology, slides were prepared for cytology and the VisioCyt® test from urine samples. The diagnostic performance of VisioCyt® was assessed using sensitivity (primary objective, 70% lower-bound threshold) and specificity (75% lower-bound threshold). Sensitivity was also assessed by tumor grade and T-staging. VisioCyt® and cytology performance were evaluated relative to the histopathological assessments. RESULTS: Between October 2017 and December 2019, 391 participants (170 in Group 1 and 149 in Group 2) were enrolled. VisioCyt®'s sensitivity was 80.9% (95% CI 73.9-86.4%) and specificity was 61.8% (95% CI 53.4-69.5%). In high-grade tumors, the sensitivity was 93.7% (95% CI 86.0-97.3%) and in low-grade tumors 66.7% (95% CI 55.2-76.5%). Sensitivity by T-staging, compared to the overall sensitivity, was higher in high-grade tumors and lower in low-grade tumors. CONCLUSION: VisioCyt® is a promising diagnostic tool for urothelial cancers with improved sensitivities for high-grade tumors and notably for low-grade tumors.

Artificial intelligence to improve cytology performances in bladder carcinoma detection: results of the VisioCyt test.

OBJECTIVE: To explore the utility of artificial intelligence (AI) using the VisioCyt® test (VitaDX International, Rennes, France) to improve diagnosis of bladder carcinoma using voided urine cytology. PATIENTS AND METHODS: A national prospective multicentre trial (14 centres) was conducted on 1360 patients, divided in two groups. The first group included bladder carcinoma diagnosis with different histological grades and stages, and the second group included control patients based on negative cystoscopy and cytology results. The first step of this VISIOCYT1 trial focussed on algorithm development and the second step on validating this algorithm. A total of 598 patients were included in this first step, 449 patients with bladder tumours (219 high-grade and 230 low-grade) and 149 as negative controls. The VisioCyt test was compared to voided urine cytology performed by experienced uro-pathologists from each centre. RESULTS: Overall sensitivity was highly improved by the VisioCyt test compared to cytology (84.9% vs 43%). For high-grade tumours the VisioCyt test sensitivity was 92.6% vs 61.1% for the uro-pathologists. Regarding low-grade tumours, VisioCyt test sensitivity was 77% vs 26.3% for the uro-pathologists. CONCLUSION: In comparison to routine cytology, the results of the first phase of the VISIOCYT1 trial show very clear progress in terms of sensitivity, which is particularly visible and interesting for low-grade tumours. If the validation cohort confirms these results, it could lead to the VisioCyt test being considered as a very useful aid for pathologists. Moreover, as this test is in fact software based on AI, it should become more and more efficient as more data are collected.