Multiparametric magnetic resonance imaging (mpMRI) for bladder cancer (BCa) is expanding and becoming increasingly accurate in providing high tissue contrast resolution, getting able to finely differentiate bladder wall layers. VI-RADS score is a novel diagnostic tool adopted to provide preoperative BCa staging demonstrated to be a reliable image-guided approach to assess presence of muscle invasiveness in the pre-TURBT setting. Aim of this study is to validate the diagnostic accuracy of VI-RADS scoring system in discriminating NMIBC from MIBC in a prospective single-center cohort of patients who undergo mpMRI of the bladder as initial diagnostic tool before primary TURBT.
This prospective study received formal Institutional Review Board and Ethical Committee approval. Between December 2017 and May 2019 all patients referred to our institutions for BCa suspicion were offered mpMRI before TURBT as per institution protocol. Each patient underwent mpMRI of the bladder to evaluate diagnostic accuracy of VI-RADS score in NMI-and-MIBC discrimination at initial TURBT. All exams are reviewed by two urogenital radiologists, blinded to clinical history. Both readers assign a VI-RADS score (1 to 5) to each lesion (up to three per patient) and for each patient only the one with the highest VI-RADS score is considered. A cutoff score of VI-RADS ≥3 to define MIBC is assumed. Sensitivity, specificity, positive and negative predictive values (PPV, NPV) are calculated to assess accuracy of mpMRI in discriminating NMI-and-MIBC, using TURBT for low-risk NMIBC, repeated TURBT (Re-TURBT) for high-risk NMIBC and radical cystectomy (RC) for MIBC as standard of reference. MpMRI performance is assessed by receiver operating characteristics (ROC) curve analysis. Ƙ statistics to estimate inter- and intra-reader variability.
231 patients were enrolled. MpMRI showed sensitivity, specificity, PPV and NPV, in discriminating NMI-from-MIBC at initial TURBT of 91.9% (95% 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. Area under curve (AUC) was 0.94 (95%CI: 0.91 – 0.97). Inter-reader agreement was overall good (Ƙ= 0.81, 95% CI: 0.65-0.93) with only 17 cases reporting disagreement between readers. Intra-reader agreement was near-perfect for both readers (Ƙ>0.92).
VI-RADS score is a novel imaging tool leading urologist to properly differentiate patients with NMIBC vs. MIBC before TURBT. Future prospective, larger, multicentric trials are mandatory to definitively validate clinical reliability of VI-RADS score in pre-TURBT setting.