Clinical impact of MRI-based risk calculators for prostate cancer diagnosis: a systematic review and meta-analysis

Background: Prostate cancer (PCa) is the second most common cancer among men worldwide. Current diagnostic methods often lack sufficient sensitivity and specificity, leading to unnecessary biopsy. With growing use of MRI and EAU guideline recommendations, this review synthesised evidence on MRI-based risk calculators (RCs) for PCa diagnosis and compared their performance with traditional clinical RCs. Methods: A systematic search of Embase, Medline, Scopus, Cochrane Library, and Web of Science databases assessed the discriminatory ability of MRI-based RCs using Area Under the Curve (AUC). A meta-analysis was conducted to pool AUC estimates, assess heterogeneity, and compare the differences in discriminatory ability. Results: Of 2049 papers, 16 met the inclusion criteria. MRI-based RCs showed increased discrimination, with an AUC of 0.84 (95% CI: 0.81–0.86) for clinically significant PCa (csPCa), compared to 0.76 (95% CI: 0.73–0.79) for clinical models, and an AUC of 0.81 (95% CI: 0.78–0.84) for all PCa, compared to 0.74 (95% CI: 0.68–0.79). The pooled logit(AUC) difference was 0.49 units for csPCa and 0.37 units for all PCa. High heterogeneity was noted, likely due to PCa variability, and 31% of the studies had a high or unclear risk of bias, potentially affecting generalisability. Conclusions: MRI-based RCs improve the diagnostic accuracy for PCa with the potential to reduce unnecessary biopsies and optimise healthcare resources, thereby supporting their integration into clinical practice.