Pd35-12 accuracy of Prostate Specific Antigen Density at Detecting Prostate Cancer

Abstract

INTRODUCTION AND OBJECTIVES: PSA density (PSAD) is a PSA derivative that has been used to help predict for prostate cancer on biopsy. PSAD cut offs were formulated from cohorts consisting of mostly Caucasian (CA) men who underwent a sextant biopsy. We sought to evaluate PSAD in a more modern cohort of men to predict prostate cancer, to predict significant cancer, and to compare the performance of PSAD in African Americans (AA) and CA.METHODS: After IRB approval, we performed a retrospective chart review of all men who underwent a 12 core prostate biopsy between 2015 and 2017 at LSU Shreveport. Data collected included age, race, BMI, finasteride use, PSA, prostate volume (PV), biopsy pathology, T stage and Gleason score. PSA was corrected for finasteride use. PSA density was calculated by PSA divided by PV. Receiver operator characteristic (ROC) curve analysis was performed using the entire cohort, the AA cohort, and the CA cohort to predict for prostate cancer and significant cancer. Significant cancer was defined two ways using the AUA risk groups: 1. Intermediate (favorable and unfavorable) and high risk groups 2. Unfavorable intermediate and high risk groups.RESULTS: Of the entire cohort (n[239), 68.2% (n[163), 28.8% (n[69), and 2.9% were AA, CA, and other, respectively. Including 91 AA and 27 CA, 49.3% had a positive biopsy.The area under the curve (AUC) to predict prostate cancer was 0.784 (95% CI 0.726-0.841), 0.761 (95% CI 0.688-0.834), and 0.814 (95% CI 0.711-0.916) using the entire, AA, and CA cohorts, respectively.Using definition 1 for significant cancer, the AUC for the entire, AA, and CA cohorts was 0.796 (95% CI 0.738-0.854), 0.777 (95% CI 0.705-0.85), and 0.824 (95% CI 0.714-0.935), respectively.Using definition 2 for significant cancer, the AUC for the entire, AA, and CA cohorts was 0.841 (95% CI 0.786-0.895), 0.814 (95% CI 0.745-0.883), and 0.905 (95% CI 0.83-0.98), respectively.Table 1 lists the sensitivity (Se) and specificity (Sp) for selected PSAD cut off scores to predict a positive biopsy for the entire, AA, and CA cohorts.CONCLUSIONS: PSAD is still a useful tool and can be used to help counsel patients about undergoing a prostate biopsy. PSAD performed better in Caucasian men than African American men in predicting prostate cancer and significant cancer.