Prostate imaging-reporting and data system version 2 in combination with clinical parameters for prostate cancer detection: a single center experience.

Abstract

The diagnostic performance of the Prostate Imaging Reporting and Data System version 2 (PI-RADS v2) has been challenged due to its lower diagnostic accuracy and higher false-positive rates for prostate cancer detection. This study aimed to analyze the diagnostic performance of PI-RADS v2 in combination with clinical parameters in patients with suspected prostate cancer. A total of 424 men with suspicion of prostate cancer were retrospectively analyzed. Logistic regression analyses were performed to identify predictors of clinically significant prostate cancer defined as a Gleason score of 3 + 4 or greater. The prediction performance was compared with prostate specific antigen (PSA), free/total PSA ratio (f/t PSA), PSA density (PSAD), PI-RADS v2 alone, and PI-RADS v2 plus PSAD using receiver operating characteristics (ROCs). In total, 231 out of 424 cases (54.48%) were pathologically diagnosed as prostate cancer. The percentage of clinically significant prostate cancer was higher in patients with PI-RADS v2 score of 4 or greater compared to PI-RADS v2 score of less than 4 (90.86% vs. 55.88%, P < 0.001). Age, PSA level, f/t PSA, PSAD, and PI-RADS v2 were significant independent predictors of clinically significant prostate cancer. The ROC area under the curve of PI-RADS v2 plus PSAD (0.952) was larger compared with PSA (0.845), f/t PSA (0.719), PSAD (0.920), and PI-RADS v2 alone (0.885). PI-RADS v2 in combination with PSAD may help detect clinically significant prostate cancer and provide benefit in making the decision to biopsy men at suspicion of prostate cancer.