Extending Current Guideline-recommended Prostate-specific Antigen Density Thresholds for Improving Benefit/Harm Balance in Risk-based Magnetic Resonance Imaging-directed Prostate Cancer Diagnosis, from the Community-based Multicentre MR-PROPER Study

BACKGROUND AND OBJECTIVE: International guidelines recommend a risk-based magnetic resonance imaging (MRI)-directed diagnostic pathway in prostate cancer (PCa) suspected biopsy-naïve men to reduce overdiagnosis. Prostate-specific antigen (PSA) density has been suggested as a stratification tool for MRI or biopsy testing. This study aims to investigate the risk-based MRI-directed pathway with PSA density thresholds indicating MRI and/or targeted biopsy in PCa suspected men, reducing safely unproductive interventions.

METHODS: A post hoc analysis was conducted in the MRI arm of the prospective multicentre MR-PROPER study, originally comparing diagnostic outcomes with a risk calculator-based approach. PSA density thresholding strategies were investigated in PCa-suspected biopsy-naïve men undergoing MRI and targeted biopsy for Prostate Imaging Reporting and Data System (PI-RADS) score ≥3. Outcomes were grade group (GG) ≥2 cancer detection, avoided MRI scans, negative biopsies, and GG1 cancer detection, also addressed by benefit-to-harm ratios.

KEY FINDINGS AND LIMITATIONS: MR-PROPER included 996 PCa-suspected men in the MRI pathway (per-protocol analysis). The GG ≥2 cancer detection rate was 24% (239/996). False MR-positive results were obtained in 47% (217/456; 133 negative biopsies and 84 GG1 cancer cases). In PI-RADS 3 men, post-MRI PSA density thresholding at ≥0.20 ng/ml 2 left 1.3% (0-2.7%; 3/239) GG≥2 cancer cases undetected, while avoiding 29% (22-35%; 38/133) negative biopsies and a 3.6% (0-8.2%; 3/84) GG1 cancer detection rate. To indicate MRI, pre-MRI PSA density thresholding at ≥0.10 ng/ml 2 avoided 38% (31-44%; 50/133) unproductive biopsies, GG1 cancer detection in 12% (6.1-17%; 10/84), and 30% (27-33%; 301/996) MRI scans, incurring 9.6% (5.4-14%; 23/239) undetected GG ≥2 cancer cases.

CONCLUSIONS AND CLINICAL IMPLICATIONS: In a community-based diagnostic setting (24% GG ≥2 disease), applying a post-MRI PSA density threshold of ≥0.20 rather than the recommended threshold of 0.10 ng/ml 2 in men with PI-RADS 3 lesions reduces unnecessary biopsies while maintaining oncological safety. This improved the benefit-to-harm ratio of risk-based PCa diagnostics, particularly in cancer-averse settings. For MRI indication, a pre-MRI PSA density threshold of ≥0.10 ng/ml 2 supports a favourable benefit-to-harm ratio, though with narrow safety margins. These findings may inform risk-adapted strategies, particularly in biopsy-averse settings.