Detection and Localization of Unfavorable-histology Prostate Cancer Using MRI and Whole-mount Histopathology

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Detecting unfavorable-histology prostate cancer (uhPC), defined by Grade Group (GG) ≥3 or GG2 with specific features, is critical for appropriate intervention.
• Accurate detection and localization of uhPC guide treatment decisions for patients with metastatic potential.

Purpose of the Study:

• To evaluate the utility of Prostate Imaging Reporting and Data System (PI-RADS) and automated Restriction Spectrum Imaging restriction score (RSIrs) for detecting and localizing uhPC.
• To compare the diagnostic performance of PI-RADS and RSIrs in identifying aggressive prostate cancer.

Main Methods:

• A multicenter cohort study evaluated patient-level detection of uhPC using biopsy as the reference standard.
• Lesion-level localization was assessed in a separate cohort using whole-mount histopathology (WMHP) from radical prostatectomy.
• Area under the receiver operating characteristic curve (AUC) was calculated to compare PI-RADS and RSIrs performance.

Main Results:

• Patient-level detection AUC did not differ significantly between PI-RADS and RSIrs (p=0.13) in 1022 patients.
• At the lesion level (n=103), combined PI-RADS and RSIrs sensitivity for the index tumor was 93% and for all uhPC tumors was 90%.
• PI-RADS sensitivity was 81% and RSIrs sensitivity was 86% for all uhPC tumors.

Conclusions:

• MRI, including PI-RADS and automated RSIrs, demonstrates high sensitivity for detecting uhPC, reinforcing its value in identifying aggressive disease.
• Both PI-RADS and RSIrs are potentially useful for targeted biopsy and focal treatment strategies for aggressive intraprostatic lesions.
• These imaging biomarkers aid in guiding intervention for prostate cancer with metastatic potential.