The performance of prostate specific antigen for predicting prostate cancer is maintained after a prior negative prostate biopsy.

Abstract

It has been suggested that prostate specific antigen has no predictive value for prostate cancer after a first negative biopsy has been performed. We compared the performance operating characteristics of prostate specific antigen for prostate cancer between a first and subsequent prostate biopsy in a group of men with complete verification of cancer status. From the 18,882 participants in the Prostate Cancer Prevention Trial we examined men in the placebo group who had only a first biopsy or a first and second prostate biopsy with a prostate specific antigen and digital rectal examination within 1 year before each biopsy. The receiver operating characteristic curve was estimated for prostate specific antigen for detection of prostate cancer on the first biopsy compared to the second, and the C-statistics were compared. Of this group 5,608 men had a first biopsy and 687 of those with a negative first biopsy underwent a second biopsy. The C-statistic was 0.650 (95% CI 0.632, 0.668) for the first biopsy and 0.664 (95% CI 0.607, 0.721) for the second biopsy. The C-statistic for the second biopsy was statistically significantly greater than 0.5 (p <0.001) and overlapped with that from the first biopsy. Prostate specific antigen does not lose predictive value for the detection of prostate cancer even after a first biopsy shows no evidence of cancer, and its performance characteristics are only slightly decreased.