Abstract

OBJECTIVES:

Multiple trials have shown the high specificity of urine prostate cancer gene 3 (PCA3) compared with serum prostate-specific antigen (PSA) for biopsy detection of prostate carcinoma. We characterized the patterns of use of PCA3 by community urologists and determined the performance of PCA3 testing as a laboratory-developed test in a reference laboratory setting.

METHODS:

The urine PCA3 and PSA mRNA levels after digital rectal examination were determined using transcription-mediated amplification. The cutoff for a positive PCA3 score (PCA3/PSA mRNA x 10(3)) were pre-established at > or = 35. The PCA3 results were correlated with the serum PSA level, previous biopsy history, and the prostate biopsy findings.

RESULTS:

A total of 278 PCA3 tests were performed from December 2006 to June 2007. Of the PCA3 tested patients, 55.5% had previously undergone > or = 1 prostate biopsy; 92.7% had a PSA level > or = 2.5 ng/mL. The PCA3 test informative rate was 97.5%. For 50 samples that were also analyzed at a separate laboratory, concordance was achieved in 94%. The mean and median PCA3 score was 44.3 and 21.1, respectively. No correlation was found with the serum PSA level. The PCA3 test was negative in 16 of 19 patients with negative concurrent biopsy findings and positive in 8 of 11 with positive concurrent biopsy findings (sensitivity 72.7% and specificity 84.2%). Of 32 patients (70% with previous biopsy) who had undergone biopsy an average of 56 days after positive PCA3 test results, prostate carcinoma was detected in 41%.

CONCLUSIONS:

Urine PCA3 testing on the transcription-mediated amplification platform performed well as a laboratory-developed test. The high specificity of PCA3 was confirmed. In patients with elevated PSA levels and negative biopsy findings, PCA3 testing might be useful in choosing between repeat biopsy and more conservative follow-up.