Prostate Region-Wise Imaging Biomarker Profiles for Risk Stratification and Biochemical Recurrence Prediction
Prostate cancer has highly variable clinical outcomes, and conventional risk-stratification systems do not always accurately identify patients who will relapse after curative treatment. MRI radiomics offers a non-invasive way to capture tumor heterogeneity and tissue characteristics that are not readily visible on conventional imaging.
This study evaluated 128 patients with localized prostate cancer treated with neoadjuvant androgen deprivation therapy and radiotherapy, investigating MRI-derived radiomic, diffusion and perfusion biomarkers for risk stratification and prediction of biochemical recurrence (BCR) 10 years after diagnosis. Importantly, biomarkers were evaluated separately across different prostate regions as well as across the whole gland.
Distinct imaging profiles—mainly driven by radiomic features reflecting tissue heterogeneity—were associated with both higher-risk disease and subsequent BCR. Region-specific analysis proved particularly informative, with models based on individual prostate regions generally outperforming whole-gland approaches.
For BCR prediction, combining imaging biomarkers with clinical variables achieved AUCs of 0.841–0.877, with the peripheral-zone model reaching an AUC of 0.877. In high/unfavorable-intermediate-risk patients, combined models reached AUCs as high as 0.951.
The key novelty is the region-wise approach: rather than treating the prostate as a single homogeneous organ, the study demonstrates that quantitative MRI information from specific anatomical regions can improve prediction of risk and long-term recurrence. It was reported as the first study to demonstrate the prognostic value of such region-wise MRI biomarker profiles specifically in patients receiving radiotherapy and ADT.
Sánchez Iglesias Á, Morillo Macías V, Picó Peris A, Fuster-Matanzo A, Nogué Infante A, Muelas Soria R, Bellvís Bataller F, Domingo Pomar M, Casillas Meléndez C, Yébana Huertas R, et al. Prostate Region-Wise Imaging Biomarker Profiles for Risk Stratification and Biochemical Recurrence Prediction. Cancers. 2023; 15(16):4163.