Reply to R Vieth

Departments of Epidemiology and Biostatistics & Urology University of California, San Francisco 3333 California Street Suite 280 San Francisco, CA 94143-1228 E-mail: jchan{at}epi.ucsf.edu

Dear Sir:

In response to Vieth, we state in our article (1) that "in vitro, in vivo, and epidemiologic studies suggest that 1,25(OH)₂D₃ may protect against prostate cancer" and list the references for both experimental and epidemiologic studies. We included references for all the epidemiologic studies (null and statistically significant); some of the nested case-control studies suggest trends in association that were not statistically significant. In 2 paragraphs we reviewed the strength of the experimental literature in contrast with the epidemiologic studies, acknowledging clearly that only 1 of the 4 epidemiologic studies observed a significant inverse association between serum 1,25-dihydroxyvitamin D₃ [1,25(OH)₂D₃] concentrations and the risk of prostate cancer. We concluded that the "results of studies in humans in which a single serum measure of 1,25(OH)₂D₃ was used are conflicting." The evidence for this hypothesis is not solely from ecologic studies, as Vieth indicates.

Vieth asserts that a statistically significant difference of 4.6 pmol/L (1.83 pg/mL) in the mean concentration of 1,25(OH)₂D₃ between the experimental and control groups in the study by Corder et al (2) is physiologically and etiologically unimportant. To the contrary, small differences in population means can translate to large relative risks when quantiles of the population distribution of any given exposure are compared. Indeed, in the study by Corder et al, men in the top 25th percentile of 1,25(OH)₂D₃ concentrations had up to an 85% lower risk of developing prostate cancer than did men in the lowest 25th percentile; the results were statistically significant among men who were concurrently in the lowest 25th percentile of the 25-hydroxyvitamin D distribution. A comparison of mean concentrations between the experimental and control groups, especially in hormone or biomarker concentrations that are etiologically important within clinically normal ranges, can mask biologically meaningful associations. For example, high cholesterol concentrations increase the risk of coronary artery disease, but a comparison of population means between experimental and control groups generally indicates only a 4% difference in total cholesterol. In contrast, a comparison of the extreme quintiles showed an 86% elevated risk of coronary artery disease (3).

Vieth also comments that our article does not address the potential need to balance the benefits of calcium intake for the prevention of osteoporosis and other diseases against the possible increase in the risk of prostate cancer. Obviously, this was beyond the scope of our paper. We believe it is premature to recommend any change in diet for prostate cancer prevention on the basis of these results.

REFERENCES

1. Chan JM, Stampfer MJ, Ma J, Gann PH, Gaziano JM, Giovannucci EL. Dairy products, calcium, and prostate cancer risk in the Physicians’ Health Study. Am J Clin Nutr 2001;74:549–54.
2. Corder EH, Guess HA, Hulka BS, et al. Vitamin D and prostate cancer: a prediagnostic study with stored sera. Cancer Epidemiol Biomarkers Prev 1993;2:467–72.
3. Stampfer MJ, Sacks FM, Salvini S, Willett WC, Hennekens CH. A prospective study of cholesterol, apolipoproteins, and the risk of myocardial infarction. N Engl J Med 1991;325:373–81.