Predictive value of total and percent free prostate specific antigen in high grade prostatic intraepithelial neoplasia lesions: Results of the Tyrol Prostate Specific Antigen Screening Project

Purpose: We evaluate the predictive values of total and percent free prosta te specific antigen (PSA) in regard to high grade intraepithelial lesions i n volunteers who participated in the Tyrol PSA Screening Project. Materials and Methods: Between June 1995 and December 1998, 1,474 patients undergoing transrectal biopsy of the prostate were evaluated. The primary d etection rates of prostate cancer and high grade intraepithelial lesions we re evaluated. In addition, the rate of prostate cancer detected on biopsy i n patients diagnosed with high grade prostatic intraepithelial neoplasia on the previous biopsy was assessed. Mean total PSA values and mean percent f ree PSA levels were determined for each study group and compared using the Mann-Whitney U test. Results: A total of 1,077 (73.1%) volunteers had benign prostatic hyperplas ia or prostatitis, and 327 (22.2%) had prostate cancer. The primary detecti on rate for high grade intraepithelial lesions was 4.7% (70 patients) and o n repeat biopsy was 38.6% (27). Mean total PSA for the benign prostatic hyp erplasia, prostate cancer, high grade and intraepithelial cancer groups wer e 6.0, 8.7, 5.9 and 5.2 ng./ml., respectively. Mean percent free PSA values for the various groups were 21.9, 12.1, 15.0 and 12.0, respectively. In re gard to total PSA there was a statistically significant difference between the prostate cancer and high grade prostatic intraepithelial neoplasia grou ps (p = 0.016), as well as the prostate cancer and intraepithelial cancer g roups (p = 0.028). However, the high grade and intraepithelial cancer group s did not differ significantly. In regard to percent free PSA there were st atistically significant differences between the prostate cancer and high gr ade prostatic intraepithelial neoplasia groups (p = 0.0001), and the high g rade and intraepithelial cancer groups (p = 0.013). Conclusions: In regard to percent free PSA our data indicate a significant difference between high grade intraepithelial lesion and intraepithelial ca ncer. Due to a substantial overlap in percent free prostate specific antige n between the 2 groups, a clinically useful cutoff point could not be estab lished. Therefore, we recommend repeat biopsy in all patients with high gra de intraepithelial lesions regardless of the percent free PSA.