Objectives. The effects of finasteride on prostate-specific antigen (PSA) variability and usefulness in prostate cancer detection were examined.

Methods. Percent change and crossover of PSA levels between the low (1.0 to 3.9 ng/mL) and high (4.0 to 10.0 ng/mL) ranges were evaluated in 72 men with benign prostatic hyperplasia (BPH) and 77 men with both BPH and prostate cancer (PCa) treated with finasteride or placebo for 6 months. Patients with PCa were studied as a model for evaluating the effects on PSA levels in patients with BPH and latent PCa. As recommended on the product label, PSA levels for finasteride-treated patients were doubled for interpretation.

Results. In patients with BPH, most placebo- and finasteride-treated patients with low PSA levels at baseline had subsequent PSA levels below 4.0 ng/mL throughout the study. Among patients with high baseline PSA levels, only 1 of 17 finasteride-treated patients, compared with 8 of 13 placebo-treated patients, crossed into the low range. In the BPH/PCa study, most placebo-treated patients maintained PSA levels in the same range (15 of 19 less than 4.0 ng/mL; 14 of 16 greater than 4.0 ng/mL). Almost one third of finasteride-treated patients with low PSA levels at baseline crossed into the high range (8 of 22), whereas most patients with high PSA levels at baseline were not masked with treatment, with PSA levels remaining high (12 of 15).

Conclusions. PSA levels cross between the low and high PSA ranges in both finasteride- and placebo-treated patients with BPH and those with both BPH and PCa. Doubling the PSA levels in finasteride-treated patients allows appropriate interpretation of PSA values and does not mask the detection of PCa.