How not to detect prostate cancer.

摘要

In their excellent study of the effect of finasteride on serum prostate-specific antigen (PSA) concentrations, D'Amico and Roehrborn referto the usual recommendation of doing a prostate biopsy when the PSA value is 4 ng/mL or over. The accompanying commentary endorses the conventional wisdom that men with a raised serum PSA concentration should undergo prostate biopsy. This approach to the detection of prostate cancer needs to be challenged, not least because of the huge number of men with a baseline PSA concentration that might be considered abnormal. For example, around 17% of men older than 60 years of age have a PSA concentration of over 2.5 ng/mL. The justification for biopsy is that men with a raised PSA concentration are at a high risk of prostate cancer. However, the Prostate Cancer Prevention Trial (PCPT) has shown that cancer is also a common biopsy finding in men with normal PSA concentrations. In the PCPT study over 5000 men aged more than 55 years underwent prostate biopsy. Of those with aPSA concentration between 1.1 and 2.0 ng/mL, no fewer than 21% were found to have prostate cancer. This compares with 30% of men with a PSA concentration of 3.1-4.0 ng/mL, and 43% of men with a PSA concentration of over 6 ng/mL. These data argue against the use of an arbitrary PSA threshold to select men for prostate biopsy. Rather, serum PSA concentration should be considered as a continuous variable that has an association, albeit a modest one, with the chance of finding prostate cancer on biopsy.