Restriction Spectrum Imaging as a Quantitative Biomarker for Prostate Cancer With Reliable Positive Predictive Value
Mariluz Rojo Domingo, D. Deondre, Christopher C. Conlin, Aditya Bagrodia, Tristan Barrett, Madison T Baxter, Matthew R. Cooperberg, Felix Y. Feng, Michael E. Hahn, Mukesh G. Harisinghani, Gary Hollenberg, Juan Javier‐Desloges, Karoline Kallis, Sophia C. Kamran, Christopher J. Kane, Dimitri A. Kessler, Joshua Kuperman, Kang-Lung Lee, Jonathan M. Levine, Michael A. Liss, Daniel Margolis, Ian Matthews, P. Murphy, Nabih Nakrour, Michael A. Ohliger, Courtney Ollison, Thomas Osinski, Anthony J. Pamatmat, Isabella R. Pompa, Rebecca Rakow‐Penner, Jacob Roberts, Ahmed Shabaik, Yuze Song, David Song, Clare M. Tempany, Shaun Trecarten, Natasha Wehrli, Eric Weinberg, Sean Woolen, George Xu, Allison Y. Zhong, Anders M. Dale, Tyler M. Seibert
Abstract
PURPOSE: The positive predictive value of the Prostate Imaging Reporting and Data System (PI-RADS) for clinically significant prostate cancer (csPCa, grade group [GG] ≥2) varies widely between radiologists. The restriction spectrum imaging restriction score (RSIrs) is a biophysics-based metric derived from diffusion MRI that could be an objectively interpretable biomarker for csPCa. We aimed to evaluate performance of RSIrs for patient-level detection of csPCa in a large and heterogenous dataset, and to combine RSIrs with clinical and imaging parameters for csPCa detection. MATERIALS AND METHODS: At 7 centers, participants underwent prostate MRI between January 2016 and March 2024. We calculated patient-level csPCa probability based on maximum RSIrs in the prostate and compared patient-level csPCa detection to apparent diffusion coefficient (ADC) and PI-RADS using AUC. We also evaluated csPCa discrimination by GG and combining RSIrs with clinical risk factors through multivariable regression. RESULTS: = .31). RSIrs and PI-RADS combined outperformed either alone. The model with RSIrs, PI-RADS, age, and PSA density achieved the best discrimination of csPCa. CONCLUSIONS: RSIrs is an accurate and reliable quantitative biomarker that performs better than conventional ADC and comparably with expert-defined PI-RADS for patient-level detection of csPCa. RSIrs provides objective estimates of probability of csPCa that do not require radiology expertise.