Determining the clinicopathological significance of the VI-RADS true-positive group: A retrospective study

Abstract Background The Vesical Imaging Reporting and Data System (VI-RADS) is widely used as a diagnostic method for predicting muscle-invasive bladder cancer (MIBC). This study aimed to determine the clinicopathological significance of the VI-RADS true-positive (TP) group. Methods We performed 333 transurethral resections of bladder tumors (TURBT) at our hospital between January 2019 and April 2021, and included 125 patients who underwent preoperative Magnetic resonance imaging (MRI) into this study. The patients were pathologically diagnosed with urothelial carcinoma (UC). We first compared the results of patients with VI-RADS scores of 3 and 4 to determine the cut-off values for MIBC; thereafter, the 125 patients were divided into TP, false-positive (FP), true-negative (TN), and false-negative (FN) groups using VI-RADS. The clinicopathological significance of the TP group was examined by retrospectively comparing the characteristics of each group. Results A total of 125 cases were examined, out of which 29 were pathologically diagnosed with MIBC. Of the 29 MIBC cases, 4 (13.8%) had a VI-RADS score of ≤ 3 and 25 (86.2%) had a VI-RADS score of ≥ 4. Of the 96 Non-muscle-invasive bladder cancer (NMIBC) cases, 83 (86.5%) had a VI-RADS score of ≤ 3 and 13 (13.5%) had a VI-RADS score of ≥ 4. The diagnostic performance of the VI-RADS with a cutoff value of 4 was 87.1% for sensitivity, 86.5% for specificity, and an area under the curve (AUC) of 0.87. In contrast, for a cutoff value of 3, the sensitivity was 90.3%, specificity was 61.5%, and AUC was 0.76. In this study, a VI-RADS score of ≥ 4 could predict MIBC. In the TP group, all 25 patients had high-grade tumors. The TP group had significantly more high-grade bladder cancers than the other groups (TP vs. TN, p = 0.001; TP vs. FP, p = 0.009; and TP vs. FN, p = 0.0278). In addition, the TP group had more tumor necrosis (TP vs. TN, p = 0.001; TP vs. FP, p = 0.0045) and more UC variant cases (TP vs. TN, p = 0.0012; TP vs. FP, p = 0.0217) than the TN and FP groups. Conclusion This study suggests that VI-RADS has a high diagnostic performance in predicting MIBC, and that VI-RADS could diagnose high-grade tumors, necrosis, and UC variants.