The impact of 5‐alpha reductase inhibitors on multiparametric MRI diagnostic performance and prostate‐specific antigen density triage: A systematic review

Objectives To evaluate the impact of 5‐alpha reductase inhibitor (5‐ARI) exposure on multiparametric MRI (mpMRI) diagnostic performance for clinically significant prostate cancer (csPCa) and to assess whether standard prostate‐specific antigen density (PSAD)‐gated biopsy thresholds remain safe in this population. Subjects/patients and methods A systematic review followed PRISMA 2020 (PROSPERO: CRD420251050507). PubMed/MEDLINE, EMBASE and the Cochrane Library were searched from January 2015 to December 2024. Eligible studies reported PI‐RADS, PSAD or lesion‐level apparent diffusion coefficient (ADC) data by exposure status, biopsy‐verified csPCa (ISUP grade group ≥2) and an unexposed comparator (≥100 patients). Risk of bias was assessed using QUADAS‐2 and ROBINS‐I; heterogeneity precluded meta‐analysis. Results Seven comparative studies met inclusion criteria. Patient‐level csPCa diagnostic accuracy was largely maintained: Sensitivity at a PI‐RADS ≥3 threshold was 93.8% in exposed versus 96.1% in unexposed men, csPCa detection was 39.1% versus 39.3% (p = 0.9), and 5‐ARI exposure was not an independent negative predictor of detection. On blinded matched review, peripheral zone lesion ADC was higher in exposed men (869 vs. 765 mm 2 /s, p = 0.04) with lower conspicuity (1.59 vs. 1.78, p = 0.03); that cohort also reported lower adjusted per‐lesion yield (OR 0.32, 95% CI 0.18–0.57), confined to the peripheral zone. A PSAD threshold of >0.15 ng/ml/cc avoided biopsy in 37.0% of exposed versus 30.3% of unexposed men, with missed csPCa in 12.3% and 11.0%, respectively. Conclusion Across seven heterogeneous observational studies, patient‐level mpMRI performance for csPCa appears largely preserved in men receiving 5‐ARIs, though this lacks pooled support and was not reproduced in the one lesion‐level, zone‐stratified analysis. Standard PSAD‐gated thresholds, evidenced by a single cohort, may perform comparably across populations. Prospective validation is required.