Prostate-specific Antigen Density as a Selection Tool Before Magnetic Resonance Imaging in Prostate Cancer Screening: An Analysis from the STHLM3MRI Randomized Clinical Trial

Background and objective: Prostate cancer screening is challenged by overdetection, overtreatment, and high resource use. Prostate-specific antigen (PSA) density (PSAD) correlates more strongly with clinically significant cancer than PSA alone. We compare outcomes using various PSAD cutoffs to select men for magnetic resonance imaging (MRI). Methods: STHLM3MRI is a screening-by-invitation trial with 49 118 men invited and 12 750 screened. In the experimental arm, participants with elevated PSA (≥3 ng/ml) had MRI and, if positive [PI-RADS] score ≥3), systematic and targeted biopsies. Evaluated PSAD cutoffs were 0.075, 0.10, 0.125, and 0.15 ng/ml². Prostate volume was measured by MRI or transrectal ultrasound. The primary endpoint was Gleason ≥3 + 4 cancer detection. The secondary endpoints were Gleason 6 and ≥4 + 3 cancer detection, and the number of MRI scans and biopsies. Relative positive fractions (RPFs) compared the outcomes using PSA ≥3 ng/ml alone. Key findings and limitations: Of 7609 men in the experimental arm, 929 (12.2%) had PSA ≥3 ng/ml, and 846 (91%) underwent MRI, with 38% having a PI-RADS score of ≥3. The median prostate volume was 45 ml (interquartile range [IQR], 33–62 ml) and PSAD 0.10 ng/ml² (IQR, 0.07–0.14 ng/ml²). A PSAD cutoff of 0.075 ng/ml² reduced Gleason 6 cancer detection by 17% (RPF, 0.83; 95% confidence interval [CI], 0.72–0.96), lowered MRI use by 28% (RPF, 0.72; 95% CI, 0.69–0.75), and missed 5% (RPF, 0.95; 95% CI, 0.92–0.98) of Gleason ≥3 + 4 cancer cases, with no difference in Gleason ≥4 + 3 cancer detection. Higher PSAD thresholds reduced Gleason 6 cancer detection and MRI use, but lowered Gleason ≥3 + 4 cancer detection. Limitations include assumed equivalency between MRI and ultrasound volume. Conclusions and clinical implications: The use of PSAD before MRI lowers overdiagnosis and resource use, while maintaining detection of significant cancer cases. Our results suggest that a PSAD cutoff above 0.075 ng/ml² is not advisable, regardless of age, due to the increased risk of missing cancer cases with Gleason scores ≥3 + 4.