ESTRO 2025 - Abstract Book

S1970

Clinical - Urology

ESTRO 2025

California San Diego, San Diego, USA. 7 Department of Urology, Northwestern Feinberg School of Medicine, Chicago, USA

Purpose/Objective: Multiparametric prostate MRI (mpMRI) is widely used for diagnosing and staging prostate cancer (PCa), but its role in predicting outcomes and risk of reclassification during active surveillance remains unclear. This study aims to evaluate the prognostic significance of baseline mpMRI findings and the risks of disease reclassification in patients with favorable-risk PCa undergoing active surveillance. Material/Methods: A systematic search of Ovid MEDLINE (PubMed), Cochrane Library, Ovid Embase, Scopus, Google Scholar, and Web of Science databases was conducted on July 2024 to identify studies evaluating the association between baseline MRI PI-RADS score and the risk of biopsy reclassification in patients on active surveillance (AS) for PCa (PROSPERO: CRD42024567762). The primary endpoint was reclassification-free survival (rFS), defined as the time to biopsy reclassification, death, or censoring. Biopsy reclassification was defined as an increase in Gleason grade on surveillance biopsy relative to the initial biopsy. Hazard ratios (HR) with 95% confidence intervals (95% CI) were pooled using a random-effects model meta-analysis, to compare rFS in patients with baseline Prostate Imaging Reporting and Data System (PI-RADS) 4-5 lesions to those with PI-RADS 1-3. Additional analyses were conducted among subsets with PI-RADS 3, 4, and 5, as compared with PI-RADS 1-2 findings. Risk of bias was assessed using the ROBINS-I tool for non-randomized studies. Results: Out of 2,197 articles screened, one prospective trial (n = 100), three prospective cohort studies (n = 1,480), and seven retrospective cohort studies (n = 4,729) were included in the meta-analysis. Baseline PI-RADS 4 or 5 was associated with significantly higher risks of rFS compared to those with PI-RADS 1-3 lesions (HR: 2.21, 95% CI: 1.66 2.93, p < .001). Compared to PI-RADS 1-2, the presence of lesions assessed as PI-RADS 3 (HR: 1.88, 95% CI: 1.29 – 2.74), PI-RADS 4 (HR: 2.73, 95% CI: 2.08-3.58), or PI-RADS 5 (HR: 3.69, 95% CI: 2.50-5.45) was associated with increased risk of upgrade. The 11 studies in the meta-analysis had a moderate risk of bias, primarily due to inconsistencies in PI-RADS scoring, differences in active surveillance protocols, and lack of data imputation. Conclusion: Baseline PI-RADS scores are associated with risk of Gleason upgrade among patients with favorable risk PCa managed with initial AS. These findings support further study of tailored surveillance strategies based on MRI findings.

Keywords: PI-RADS, biopsy reclassification, prostate cancer