Abstract
BACKGROUND AND OBJECTIVE: Prostate-specific antigen (PSA) density (PSAD) improves risk stratification for clinically significant prostate cancer (csPCa) but depends on prostate volume estimation. Transrectal ultrasound is used widely, whereas transabdominal ultrasound (TAUS) is underexplored due to accuracy concerns. We compared PSAD derived from manual magnetic resonance imaging (MRI) segmentation, MRI ellipsoid, and TAUS ellipsoid. METHODS: In a single-centre retrospective cohort, men who underwent biopsy or prostatectomy with histopathological confirmation were included. (1) Manual MRI segmentation versus (2) ellipsoid formulae derived from MRI and TAUS diameters were compared. We evaluated discrimination (area under the curve [AUC]), decision curve net benefit, method agreement, and clinical utility defined by biopsy burden across illustrative PSAD thresholds. KEY FINDINGS AND LIMITATIONS: Of 486 men, 299 had complete MRI and TAUS data (mean age 71.5 [standard deviation 8.3] yr; median PSA 10.3 ng/ml); 98 (32.8%) had csPCa. Both ellipsoid formula underestimated volume, with TAUS marginally closer to the reference. However, at the illustrative threshold of PSAD = 0.15 ng/ml/cc, the estimated biopsy burden was comparable: 53.5 (manual), 57.2 (MRI), and 56.5 (TAUS) biopsies per 100 men, with identical sensitivity (0.837). Discrimination was similar, with AUCs of 0.83 (manual), 0.82 (MRI), and 0.81 (TAUS), and a decision curve analysis demonstrated overlapping net benefits. The main limitations are the retrospective single-centre design and the lack of external validation. CONCLUSIONS AND CLINICAL IMPLICATIONS: Despite volume differences, PSAD derived from TAUS and MRI ellipsoid formulae demonstrated discrimination and net benefit comparable with those of manual segmentation. TAUS-based PSAD represents a pragmatic alternative for risk stratification, particularly in resource-limited settings where MRI segmentation is infeasible. PATIENT SUMMARY: We compared three ways to measure prostate volume to calculate prostate-specific antigen (PSA) density. All three ways showed similar accuracy for finding important prostate cancers, suggesting that ultrasound-based PSA density can be used when detailed magnetic resonance imaging measurements are not practical.