Invited commentary: do anthropometric measures predict risk of prostate cancer?

In this issue of the Journal, Schuurman et al. (1) investigate the association between several anthropometric measures and the risk of prostate cancer among approximately 60,000 male participants of the Netherlands Cohort Study. In many ways, their findings echo the results of most previous cohort and casecontrol studies on this topic in that they are largely inconclusive. As discussed by Schuurman et al., most previous studies of prostate cancer and anthropometric measures (including body mass index, body fat distribution, lean body mass, body mass at a young age, change in body mass, and height) have been null, and significant associations that have emerged in individual studies and for certain subgroups of men have not been seen consistently across studies. In this most recent investigation, cohort members were followed from baseline (September 1986) through 1992; during this period, approximately 700 new cases of prostate cancer were diagnosed. In both age-adjusted and multivariate case-cohort analyses, no associations were apparent between baseline measures of height, body mass index, or estimated lean body mass and prostate cancer risk, regardless of whether the tumors were localized or advanced. A modest positive association was observed for body mass index at age 20 years (body mass index >25 vs. <19, relative risk (RR) = 1.33, 95 percent confidence interval (CI): 0.81, 2.10); this association was stronger for localized prostate tumors than for advanced tumors. In contrast, a small (nonsignificant) inverse association with gain in body mass between age 20 years and age at baseline also was seen. With nearly 700 incident cases, this study represents one of the largest prospective investigations of this topic.