LAUSR

BACKGROUND Vesical Imaging Reporting and Data System (VI-RADS) score is adopted to provide preoperative bladder cancer (BCa) staging. Repeated transurethral resection of bladder tumor (Re-TURBT) is recommended in most of high-risk non-muscle-invasive bladder cancers (HR-NMIBCs) due to possibility of persistent/understaged disease after initial TURBT. No diagnostic tools able to improve patient's stratification for such recommendation exist. OBJECTIVE To (1) prospectively validate VI-RADS for discriminating between NMIBC and muscle-invasive bladder cancer (MIBC) at TURBT, and (2) evaluate the accuracy of VI-RADS for identifying HR-NMIBC patients who could avoid Re-TURBT and detecting those at higher risk for understaging after TURBT. DESIGN, SETTING, AND PARTICIPANTS Patients with BCa suspicion were offered multiparametric magnetic resonance imaging (mpMRI) before TURBT. According to VI-RADS, a cutoff of ≥3 to define MIBC was assumed. TURBT reports were compared with preoperative VI-RADS scores to assess accuracy of mpMRI for discriminating between NMIBC and MIBC. HR-NMIBC Re-TURBT reports were compared with preoperatively recorded VI-RADS scores to assess mpMRI accuracy in predicting Re-TURBT outcomes. INTERVENTION Multiparametric MRI of the bladder before TURBT. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS Sensitivity, specificity, positive (PPV) and negative (NPV) predictive values were calculated for mpMRI performance in patients undergoing TURBT and for HR-NMIBC patients candidate for Re-TURBT. Performance of mpMRI was assessed by receiver operating characteristic curve analysis. Ƙ statistics was used to estimate inter- and intrareader variability. RESULTS AND LIMITATIONS A total of 231 patients were enrolled. Multiparametric MRI showed sensitivity, specificity, PPV, and NPV for discriminating NMIBC from MIBC at initial TURBT of 91.9% (95% confidence interval [CI]: 82.2-97.3), 91.1% (95% CI: 85.8-94.9), 77.5% (95% CI: 65.8-86.7), and 97.1% (95% CI: 93.3-99.1), respectively. The area under the curve (AUC) was 0.94 (95% CI: 0.91-0.97). Among HR-NMIBC patients (n=114), mpMRI before TURBT showed sensitivity, specificity, PPV, and NPV of 85% (95% CI: 62.1-96.8), 93.6% (95% CI: 86.6-97.6), 74.5% (95% CI: 52.4-90.1), and 96.6% (95% CI: 90.5-99.3) respectively, to identify patients with MIBC at Re-TURBT. The AUC was 0.93 (95% CI: 0.87-0.97). CONCLUSIONS VI-RADS is accurate for discriminating between NMIBC and MIBC. Within HR-NMIBC cases, VI-RADS could, in future, improve the selection of patients who are candidate for Re-TURBT. PATIENT SUMMARY We investigated the accuracy of Vesical Imaging Reporting and Data System (VI-RADS) score to asses bladder cancer staging before transurethral resection of bladder tumors, and we explored the performance of VI-RADS score as a future preoperative predictive tool for the selection of high-risk non-muscle-invasive bladder cancer patients who are candidate for undergoing early repeated transurethral resection of the primary tumor site.