Prostate Region-Wise Imaging Biomarker Profiles for Risk Stratification and Biochemical Recurrence Prediction

Background: Identifying prostate cancer (PCa) patients with a worse prognosis and higher risk of biochemical recurrence (BCR) is essential to guide treatment choices. Here, we aimed identify possible imaging biomarker (perfusion/diffusion + radiomic features) profiles extracted from MRIs that were able discriminate according their or the occurrence BCR 10 years after diagnosis, as well evaluate predictive value without clinical data. Methods: Patients localized PCa receiving neoadjuvant androgen deprivation therapy radiotherapy retrospectively evaluated. Imaging features for each region whole gland. Univariate multivariate analyses conducted. Results: 128 (mean [range] age, 71 [50–83] years) included. Prostate region-wise mainly composed allowed discriminating groups experiencing BCR. Heterogeneity-related increased in Overall, biomarkers retained good ability (AUC values superior 0.725 most cases), which generally improved when data included (particularly evident prediction BCR, AUC ranging 0.841 0.877 combined models sensitivity above 0.960) built per vs. Conclusions: region-aware enable identification retaining variables.