Ad hoc analyses suggest that the association between finasteride and high-grade prostate cancer could have been spurious.
The Prostate Cancer Prevention Trial (PCPT) was
designed to show whether finasteride (Proscar and generics), a 5
-reductase inhibitor, can
prevent prostate cancer. At 7 years, finasteride, compared with placebo, was
associated with lower overall risk for prostate cancer (18.4% vs. 24.4%), but
with higher risk for high-grade cancers (6.4% vs. 5.1%;
Journal Watch Jul 18 2003). Two new analyses of PCPT data suggest that these
paradoxical results might be artifacts.
In one analysis, researchers (most of whom are employed by a manufacturer of finasteride) examined whether the increased rate of high-grade cancers reflected "sampling density bias." They showed that the likelihood of detecting high-grade prostate cancer was inversely proportional to prostate volume. Because finasteride recipients had smaller glands than did placebo recipients (a known effect of finasteride), high-grade cancer was more likely to be found by random biopsies in the finasteride group.
In another analysis, researchers compared specimens from diagnostic biopsies and radical prostatectomy in patients who had both procedures. Among patients whose radical prostatectomy specimens showed high-grade cancer, the high-grade histology was more likely to have been seen on previous diagnostic biopsies in the finasteride group than in the placebo group. This finding suggests that high-grade disease was underdetected in the placebo group, perhaps because of sampling density bias.
Comment:
Concern that finasteride might induce high-grade cancers is one reason that
cancer prevention with finasteride was not embraced after the PCPT results were
published. These post hoc analyses are consistent with the hypothesis that the
observed increase in high-grade disease with finasteride was spurious. In
another cancer prevention trial, the 5-reductase inhibitor dutasteride
(Avodart) currently is being evaluated; if this study shows a decrease in
overall prostate cancer incidence, without any worrisome increase in aggressive
cases, chemoprevention for prostate cancer could gain momentum after all.
Published in Journal Watch General Medicine October 11, 2007
Cohen YC et al. Detection bias due to the effect of finasteride on prostate volume: A modeling approach for analysis of the Prostate Cancer Prevention Trial. J Natl Cancer Inst 2007 Sep 19; 99:1366.