bp-VI-RADS combined with normalized apparent diffusion coefficient enhances diagnostic efficacy for muscle-invasive...

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Magnetic resonance imaging (MRI) staging of bladder cancer uses Vesical Imaging-Reporting and Data System (VI-RADS).
• VI-RADS qualitative assessment leads to diagnostic uncertainty for muscle-invasive bladder cancer (MIBC) in score 3 cases.
• Biparametric MRI (bp-MRI) offers comparable accuracy to multiparametric MRI without contrast agents.

Purpose of the Study:

• To evaluate the diagnostic value of normalized apparent diffusion coefficient (nADC) in bp-MRI for MIBC.
• To investigate if integrating nADC with VI-RADS improves bladder cancer staging performance.
• To specifically assess performance in indeterminate VI-RADS score 3 cases.

Main Methods:

• Retrospective analysis of 172 bladder cancer patients undergoing bp-MRI.
• VI-RADS scoring and region of interest (ROI) delineation for apparent diffusion coefficient (ADC) calculation.
• Normalization of ADC using bladder urine or obturator internus muscle (OIM) to derive nADC.
• Statistical analysis including Wilcoxon rank-sum, ROC curve, and logistic regression for model comparison.

Main Results:

• VI-RADS alone had an AUC of 0.87.
• Urine-normalized ADC (nADCurine) showed the best individual performance (AUC=0.82).
• Composite models integrating VI-RADS with nADC metrics significantly outperformed VI-RADS alone (highest AUC 0.93 with nADCurine).
• In VI-RADS score 3 cases, nADCurine achieved 84.6% accuracy, correctly identifying all MIBC cases.

Conclusions:

• Normalized ADC values, particularly nADCurine, enhance VI-RADS assessment for muscle invasion.
• nADC provides an objective quantitative basis for indeterminate bladder cancer staging.
• Integrating nADC with VI-RADS improves diagnostic accuracy for MIBC, especially in equivocal cases.