Application of prostate specific antigen density and related indicators in the diagnosis of prostate cancer and construction of risk prediction model

Objectives: To investigate the diagnostic value of prostate-specific antigen density (PSAD), prostate volume (PV), free prostate-specific antigen (fPSA), total prostate-specific antigen (tPSA), and fPSA/tPSA in prostate cancer (PCa) and to construct a risk prediction model based on analyzed risk factors. Methods: Age, Gleason score of pathological biopsy, PV, fPSA, tPSA, fPSA/tPSA, and PSAD were retrospectively analyzed, and accuracy of the predictive model was evaluated using receiver operating characteristic curve and logistic regression analyses. An age × PSAD product term was introduced to test the interaction between age and PSAD, and the predictive efficacy of PSAD stratified by age was evaluated. Results: Multivariate logistic regression analysis suggested that PV, fPSA, tPSA, fPSA/tPSA, and PSAD were independent risk factors for PCa ( P < 0.05). The area under the curve of nomogram model constructed on this basis was 0.855, indicating that the model has good discrimination ability. There was a significant interaction effect between age and PSAD (P for interaction = 0.011). Stratified analysis revealed that predictive effect of PSAD was strongest in the 60- to 70-year-old group (odds ratio [OR] = 46.55, 95% confidence interval [CI]: 24.24 to 89.40) and relatively weaker in the <60-year-old group (OR = 30.93, 95% CI: 10.36 to 92.38) and the >70-year-old group (OR = 33.54, 95% CI: 19.38 to 58.05) (all P < 0.001). Conclusion: The nomogram model constructed based on PV, fPSA, tPSA, fPSA/tPSA, and PSAD can effectively predict the probability of PCa occurrence. Among these factors, the predictive value of PSAD is age dependent and particularly prominent in individuals aged 60 to 70 years.