The performance of prostate specific antigen, prostate specific antigen density and transition zone density in the era of extended biopsy schemes.

ABSTRACT

PURPOSE: Prostate specific antigen, prostate specific antigen density and transition zone density have been previously identified as prostate cancer detection tools. Recent studies suggest that prostate specific antigen may be increasingly accurate for detecting clinically significant high grade prostate cancer (Gleason grade 7 or greater). We defined the performance of these measures in a referral based population undergoing an extended prostate biopsy scheme. MATERIALS AND METHODS: We retrospectively reviewed prospectively collected data on 1,708 men referred for prostate needle biopsy. All participants were men who had not undergone biopsy in the past. From these data ROC curves were constructed for prostate specific antigen, prostate specific antigen density and transition zone density for the presence of cancer, high grade (Gleason 3 + 4 or greater) and high volume (50% or greater of cores positive) disease. RESULTS: Prostate specific antigen density had a statistically higher AUC than prostate specific antigen for detecting all prostate cancers (0.737 vs 0.633, p <0.001) as well as high grade (0.766 vs 0.673, p <0.001) and high volume (0.843 vs 0.755, p <0.001) disease. Additionally, prostate specific antigen and prostate specific antigen density performed better for detecting high grade and high volume disease compared to overall prostate cancer detection. The performance of transition zone density was similar to that of prostate specific antigen density. CONCLUSIONS: Prostate specific antigen and prostate specific antigen density show improved performance characteristics for detecting clinically significant high grade and high volume prostate cancer in referral populations undergoing extended scheme prostate needle biopsy. Prostate specific antigen density shows better performance characteristics than prostate specific antigen. No advantage was seen when using transition zone density over prostate specific antigen density.