Background: The aim of this study was to identify predictors that can increase the accuracy of detecting prostate cancer on subsequent biopsies.
Methods: Between 1998 and 2003, a total of 235 men with prostate specific antigen (PSA) levels between 4.0 and 20 ng/mL underwent one or more systematic needle biopsies of the prostate. Of these men, 73 (31.1%) underwent one repeat biopsy and 26 (11.1%) underwent two or more repeat biopsies. We evaluated the results of prostate biopsies in relation to the morbidity of prostate cancer detected on repeat biopsies.
Results: Of the 73 men who underwent repeat biopsy, 16 (21.9%) had prostate cancer. Twenty-six men with one negative re-biopsy underwent two or more repeat biopsies, and five of these patients were found to have early stage prostate cancer. On repeat biopsy, there was a significant difference in percent free PSA between the cancer-detected group and the no-cancer-detected group (P < 0.01). A receiver operating characteristics (ROC) curve gave an optimal cut-off value for percent free PSA of 11%, demonstrating a significant difference in the cancer detection rate on repeat biopsy (P = 0.0009). Analysis of the data for re-biopsies showed that cancer-detected cases showed a raised PSA value and a simultaneously reduced percent free PSA (these differences were statistically significant).
Conclusions: A low percent free PSA level increased the probability of a positive result in repeat biopsy. An increase in the accuracy of detecting cancer, especially on repeat biopsy, will promote the detection of more early stage prostate cancer.