The present study addressed the question as to whether prostate-specific antigen reverse transcriptase-polymerase chain reaction (PSA RT-PCR) could be used to identify prospectively men who have prostate cancer and to help determine which patients with an initially negative biopsy would benefit from rebiopsy. PSA RT-PCR was performed prospectively on 90 patients who were to have a prostate biopsy because of an elevated PSA level, an abnormal digital rectal examination, or both. PSA RT-PCR was performed, and the sensitivity of the test was enhanced by hybridization of the PCR with a 32P-labeled PSA cDNA probe (exons 3-5). Of the 90 men, 36 (40%) had prostate cancer on biopsy. Of these 36 men, 5 (13.9%) had a positive PSA RT-PCR finding, whereas 31 (84.1%) tested negative. Of 54 men with negative biopsies, 8 (14.8%) had a positive PSA RT-PCR result. The sensitivity of PSA RT-PCR for the detection of biopsy-proven prostate cancer was 13.9% and the specificity was 85.2%. Only 3 of 12 (25%) patients with advanced disease had a positive test result. The sensitivity of PSA RT-PCR for the detection of biopsy-proven prostate adenocarcinoma in men suspected of having prostate cancer is poor. Indeed, men without biopsy-proven prostate cancer are just as likely to have a positive result in the PSA RT-PCR as are men with cancer. Whether these men with negative prostate biopsies and positive PSA RT-PCR findings may eventually develop prostate cancer remains to be determined. At this time, PSA RT-PCR for the prospective detection of prostate cancer should be considered investigational.