Diagnostic Value of FPSAD and PI-RADS v2.1 for Clinically Significant Prostate Cancer in Patients with tPSA Levels of 4-10 ng/mL.

Objective: To evaluate the incremental diagnostic value of combining free prostate-specific antigen density (FPSAD) with Prostate Imaging Reporting and Data System version 2.1 (PI-RADS v2.1) for clinically significant prostate cancer (csPCa,defined as Gleason score≥3+4) in the diagnostic gray zone (tPSA 4-10 ng/mL). Methods : This retrospective study analyzed 137 patients (75 with csPCa and 62 with non-clinically significant prostate cancer (ncsPCa) who underwent transperineal prostate biopsy at Xiangtan Central Hospital between January 2022 and January 2024. PI-RADS v2.1 scores were assigned based on magnetic resonance (MR) imaging, and prostate volume (PV) and FPSAD were calculated. Statistical analyses included chi-square/Fisher's exact tests for categorical variables and independent t-tests for continuous variables. Logistic regression identified independent predictors of csPCa, and a nomogram model was developed. Model performance was evaluated using calibration curves and receiver operating characteristic (ROC) analysis. Results: Significant differences were observed in FPSAD, PI-RADS v2.1 scores, and free PSA (fPSA) between the csPCa and ncsPCa groups (P < 0.01). FPSAD (OR = 1.95, 95% CI: 1.22-2.22, P < 0.01) and PI-RADS v2.1 scores (OR = 2.41, 95% CI: 1.57-3.70, P < 0.01) were independent predictors of csPCa. The combined FPSAD and PI-RADS v2.1 model demonstrated superior diagnostic performance (AUC =0.829) compared to FPSAD alone (AUC = 0.69) or PI-RADS v2.1 alone (AUC = 0.773) (P < 0.01), with 91% sensitivity and 32% fewer unnecessary biopsies than PI-RADS≥3 criteria. In PI-RADS 3 subgroup (n=41), FPSAD correctly reclassified 13/18 (72.2%) indeterminate cases. Conclusion: For Asian men with tPSA 4-10 ng/mL, the FPSAD+PI-RADS algorithm (cutoffs: >0.017 and ≥4) provides 15-20% higher accuracy than either marker alone, while reducing biopsies by 25%. This approach is particularly valuable for PI-RADS 3 cases, where it resolved >65% of diagnostic uncertainties in our cohort.