An emerging hypothesis suggests that vitamin D metabolites suppress the development of prostate cancer. In a recent epidemiological study, elevated levels of 1,25-dihydroxyvitamin D (1,25-D) in blood were associated with a greatly reduced risk, particularly in older men. We conducted a nested case-control study to evaluate the relationship between plasma levels of the two major vitamin D metabolites, 1,25-D and 25-hydroxyvitamin D (25-D), and subsequent diagnosis of prostate cancer. We also measured vitamin D-binding protein to investigate the influence of free metabolite levels on risk. Plasma samples from 14,916 participants in the Physicians' Health Study were collected and frozen in 1982-1983. This analysis included 232 cases diagnosed up to 1992 and 414 age-matched control participants. Vitamin D metabolite and vitamin D-binding protein assays were conducted without knowledge of case-control status. Median levels of 25-D, 1,25-D, and vitamin D-binding protein were indistinguishable between cases and controls. Analysis of risk for increasing quartiles of total or free metabolites did not reveal a pattern of decreasing risk. For 1,25-D, men in the highest quartile had an odds ratio of 0.88 (95% confidence interval = 0.53-1.45) compared to those in the lowest quartile. Significant reductions in risk were not seen in analyses restricted to older men, to cases occurring > 3 years from blood collection, or to cases presenting as aggressive prostate cancer. Nonsignificant inverse associations for 1,25-D appeared for some groups according to 25-D level, particularly when the cutoff for defining low 25-D was reduced. These results do not support the hypothesis that high circulating levels of vitamin D metabolites reduce prostate cancer risk, although small to moderate effects cannot be excluded.