Vesical Imaging–Reporting and Data System for Multiparametric MRI to Predict the Presence of Muscle Invasion for Bladder Cancer

Abstract

Background The Vesical Imaging–Reporting and Data System (VI‐RADS) is a newly developed system of bladder cancer staging with multiparametric MRI (mpMRI), which can be used to predict the presence of muscle invasion for bladder cancer. Purpose To evaluate the accuracy of three mpMRI series (T2WI, diffusion‐weighted imaging [DWI], and dynamic contrast‐enhanced image [DCEI]) and VI‐RADS for diagnosing the muscle invasive bladder cancer (MIBC). Study Type Retrospective. Population In all, 66 pathologically proven bladder cancers in 32 patients. Field Strength/Sequence Before the diagnostic MRI with an intramuscular antispasmodic agent, optimal bladder distension was confirmed. 3.0T MRI with T2WI, DWI, and DCEI. Assessment Three reviewers independently assessed and scored the bladder cancers in T2WI, DWI, and DCEI using a five‐point score system. Based on the scores in the three sequences, reviewers scored each bladder cancer with reference to VI‐RADS categories. We evaluated the diagnostic performance of each of three mpMRI sequences and the final VI‐RADS categorization for diagnosing MIBC. Statistical Tests Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and the area under the curve (AUC) of each of three sequences separately and VI‐RADS categorization for diagnosing the MIBC. Results The diagnostic performances of each of the three mpMRI series and VI‐RADS for diagnosing MIBC were excellent. Especially using the optimal cutoff score >3 for predicting MIBC on DWI, DCEI, and VI‐RADS, the sensitivity, specificity, PPV, NPV, and AUC values were 90% (95% confidence interval [CI]: 0.56, 1.00), 100% (95% CI: 0.94, 1.00), 100% (95% CI: 0.66. 1.00), 98.3% (95% CI: 0.91, 1.00), and 0.95, respectively. Data Conclusion mpMRI based on VI‐RADS can stratify patients with bladder cancer according to the presence of muscle invasion. Level of Evidence 3. Technical Efficacy Stage 2. J. Magn. Reson. Imaging 2020;52:1249–1256.

Systematic Review and Meta-Analysis of Vesical Imaging-Reporting and Data System (VI-RADS) Inter-Observer Reliability: An Added Value for Muscle Invasive Bladder Cancer Detection

“…Patient/tumors description, main findings, and study characteristics of the whole studies included are summarized in Table 1 . Of the 8 included articles [ 8 , 19 , 20 , 21 , 22 , 23 , 24 , 25 ], 3 were conducted in Italy [ 8 , 19 , 24 ] and 2 in Japan [ 20 , 25 ], while the remaining were from China [ 21 ], Egypt, and Korea [ 22 , 23 ], covering an overall time period between February 2019 and April 2020. Overall, the number of patients included was 1016, varying from 32 to 340 in single experiences, while the prevalence of MIBC ranged from n = 6, 10.3% to n = 62, 50%.…”

“…Out of these, the article from Del Giudice et al [ 8 ] was the sole to analyze a large cohort of patients within prespecified limits of enrolling time (i.e., December 2017–May 2019) and more strict inclusion criteria with regard of previous BCa history and pathology related exclusion features (i.e., no CIS or non-urothelial carcinomas), while the other two experiences [ 22 , 24 ] did not report the same information thus suffering from potential higher RoB. The remaining five studies [ 19 , 20 , 21 , 23 , 25 ] were retrospective, assessing both urothelial and other histology variants of BCa. All studies were performed at single centers using either TURBT or a combination of TUR and partial or RC as the reference standard.…”

“…MpMRI had been performed prior to TURBT in all included studies except for one, which consisted of a mixture of pre- and post-TUR patients [ 21 ]. The majority of the studies implemented at a 3T MRI scanner [ 8 , 19 , 21 , 23 , 24 , 25 ] with T2-weighted images including slice thickness of 3 mm [ 8 , 19 , 22 , 24 , 25 ] or only 4 mm [ 20 , 21 , 23 , 25 ] respectively. Diffusion-weighted images were acquired homogeneously among the studies with high b values of 1000 s/mm 2 or higher.…”

“…In all these cases, a κ estimate with 95%CI or SE was reported. Two studies [ 23 , 24 ] enrolled three different readers with increasing levels of experience. In the experience of Marchioni et al [ 24 ], a third and more experienced reader was consulted only in the case of discrepancies among the first two radiologists, while in the study of Hong et al [ 23 ] all three raters were independently involved in the diagnostic VI-RADS assignment.…”

“…Two studies [ 23 , 24 ] enrolled three different readers with increasing levels of experience. In the experience of Marchioni et al [ 24 ], a third and more experienced reader was consulted only in the case of discrepancies among the first two radiologists, while in the study of Hong et al [ 23 ] all three raters were independently involved in the diagnostic VI-RADS assignment. As a consequence, both in the study of Hong et al [ 23 ] and Ueno et al [ 20 ], who involved five different board-certified GU-radiologists, the measure of inter-variability estimate was the ICCs with 95% CI or SE.…”