The epidemiology of sex steroid hormones and their signaling and metabolic pathways in the etiology of prostate cancer.

The purpose of this review is to discuss the epidemiologic literature on the association of sex steroid hormones and components of their signaling and metabolic pathways with prostate cancer and to describe data evaluating racial variation in sex steroid hormone pathways as a possible explanation for the notably higher risk of prostate cancer in African-American men compared to white or Asian men. Although sex steroid hormones likely contribute to the growth and progression of prostate cancer, associations between hormones and prostate cancer risk across the range of normal levels have been difficult to reliably demonstrate epidemiologically. Methodologic issues no doubt have made the detection of these associations difficult. Of particular importance are (1) the inadequacy of measuring circulating hormones in middle age as a surrogate for the exposure in the target cells in the prostate at the relevant time in life and (2) the current inability to integrate across components of the sex steroid hormone signaling pathway to fully capture target cell androgenic and estrogenic stimulation. Although the approach of evaluating polymorphisms in genes involved in sex steroid hormone signaling or metabolism as a way to minimize some of the issues in the direct measurement of hormones is logical, the findings among these studies are somewhat difficult to reconcile as well. The problems of the changing case mix due to screening for elevated PSA, small sample sizes increasing the likelihood of false negative and false positive results, the controls and their allele frequencies not being representative of the population at risk, and lack of knowledge of the functional consequence of a polymorphism in relation to other polymorphisms in that gene or without consideration of other genes involved in the same pathway may be contributory. The primary result of the Prostate Cancer Prevention Trial confirms that intraprostatic dihydrotestosterone levels in the normal range indeed do contribute to the growth of prostate adenocarcinoma. However, the secondary result of higher-grade disease in cases in the finasteride arm coupled with clinical studies showing higher grade disease in non-metastatic cases with lower serum androgens, if not a pathological artifact or detection bias in the finasteride arm, possibly suggests a complex relationship between androgens and the growth versus differentiation of a prostate tumor. Finally, racial variation in components of the sex steroid hormone pathway do appear to exist, but whether the extent of the variation is adequately great such that it accounts for some of the substantial differences in prostate cancer incidence among blacks, whites, and Asians is unclear.