EAU 2019: Prediction of Muscle Invasive Bladder Cancer Using the Vesical Imaging-Reporting and Data System and Apparent Diffusion Coefficient Values (VI-RADS/ADC)

Barcelona, Spain (UroToday.com) The Vesical Imaging-Reporting and Data System (VI-RADS) was proposed for predicting muscle invasive bladder cancer (MIBC) 1. The rationale and aim of the VI-RADS scoring system was to define a standardized approach to imaging and reporting mpMRI for bladder cancer, defining the risk of muscle invasion.

The scoring is applicable to untreated patients and to treated patients having only received a diagnostic transurethral resection of the tumor (TURBT), but prior to re-TURBT. mpMRI is best performed before or at least 2 weeks after TURBT, bladder biopsy or intravesical treatment. Administration of an intramuscular antispasmodic agent is recommended, in addition to adequate bladder distention. MRI does not necessarily have the ability to visualize all of the histological bladder wall layers, however, it is able to assess size, location, multiplicity, and morphology. A 5-point VI-RADS score is generated using the individual T2W, DWI, and DCE MRI categories and suggests the probability of muscle invasion. The dominant sequences for risk estimates are DWI (first) and DCE (second, especially if DWI is sub-optimal). The T2 sequence (structural category) is helpful as a first pass guide. 

The VI-RADS scoring is as follows:
  • VI-RADS 1: SC CE and DW category 1 (muscle invasion is highly unlikely)
  • VI-RADS 2: SC, CE and DW category 2; both CE and DW category 2 with SC category 3 (muscle invasion is unlikely to be present)
  • VI-RADS 3: SC, CE, and DW category 3; SC category 3, CE or DW category 3, and the remaining sequence category 2 (the presence of muscle invasion is equivocal) 
  • VI-RADS 4: at least SC and/or DW and CE category 4; the remaining category 3 or 4 SC category 3 plus DW and/or CE category 4; SC category 5 plus DW and/or CE category 4 (muscle invasion is likely)
  • VI-RADS 5: at least SC plus DW and/or CE category 5; the remaining category 4 or 5 (invasion of muscle and beyond the bladder is very likely)
However, no validation study has been reported on the accuracy of VI-RADS. At the Novel Technology and Techniques in Urothelial Cancer Diagnosis at EUA 2019, Dr. Sakamoto and colleagues presented their work validating VI-RADs. Furthermore, the apparent diffusion coefficient values (ADC) on diffusion-weighted MRI are significantly lower in MIBC than those in non-MIBC. As such, they also assessed whether incorporating ADC improved the predictive accuracy of VI-RADS.

This study included 178 patients with primary bladder cancer who underwent a mpMRI before TUR (2013-2018). mpMRI findings were classified into scores 1-5 according to VI-RADS, as noted above. ADC of the tumor and the gluteus maximus muscle were calculated on ADC maps and ADC represented standardized tumor ADC (tumor ADC/gluteus maximus ADC) to overcome the incompatibility between different MRI protocols 2. Cutoff values of VI-RADS and ADC to best predict MIBC were determined by partition analysis. Furthermore, the predictive accuracy of these variables was examined by logistic regression analysis.

Among the 178 patients, 47 (26%) had MIBC. The positive predictive value of MIBC for VI-RADS 1 was 18%, 2 was 17%, 3 was 12%, 4 was 42%, and 5 was 77% (p<0.01). Cutoffs of VI-RADS and ADC were 4 and 0.89, respectively. The positive predictive value of VI-RADS ≤3 was 16% and ≥4 was 60% (AUC 0.70, p<0.01). The positive predictive value of high ADC was 10% and low ADC was 56% (AUC 0.76, p<0.01). Importantly, VI-RADS ≥4 and low ADC were independently associated with MIBC on logistic regression (both p<0.01). When combining VI-RADS and ADC, the positive predictive value of VI-RADS ≤3/high ADC was 6%, for VI-RADS ≤3/low ADC was 39%, for VI-RADS ≥4/high ADC was 30%, and for VI-RADS ≥4/low ADC was 91% (AUC 0.84 p<0.01). Also, the VI-RADS/ADC restored 39% of MIBC understaged with VI-RADS. Based on this work, the following algorithm can be used for predicting MIBC using VI-RADS/ADC:


As with any algorithm, validation in several cohorts enhances the robustness and usability of the model. Dr. Sakamoto and colleagues have validated VI-RADS in their cohort, in addition to adding the value of low/high ADC. 

The authors concluded that VI-RADS was useful in predicting MIBC with and AUC 0.70. Incorporating ADC improved the predictive accuracy of VI-RADS and reduced the risk of understaging for tumors of VI-RADS ≤3.

Presented by: Katsunori Sakamoto, Tokyo Metropolitan Cancer, and Infectious Diseases Center Komagome Hospital, Tokyo, Japan

Written by:  Zachary Klaassen, MD, MSc – Assistant Professor of Urology, Georgia Cancer Center, Augusta University - Medical College of Georgia Twitter: @zklaassen_md at the 34th European Association of Urology (EAU 2019) #EAU19 conference in Barcelona, Spain, March 15-19, 2019.

1. Panebianco V, Narumi Y, Altun E, et al. Multiparametric Magnetic Resonance Imaging for Bladder Cancer: Development of VI-RADS (Vesical Imaging-Reporting and Data System). Eur Urol 2018 Sep;74(3):294-306.
2. Nishizawa T, Yoshida S, Koga F, et al. Standardization of the apparent diffusion coefficient value of bladder cancer across different centers: Applicability in predicting aggressive pathologic phenotypes. Clin Imaging 2017 Jul-Aug;44:121-126.