Abstract Oxidative stress caused by reactive oxygen species (ROS) enhances tumor carcinogenesis whereas dietary antioxidants such as selenium may protect against this process. Numerous epidemiological studies have suggested an inverse association of selenium levels (measured in blood or toenails) with prostate cancer risk. In 1996, the Nutritional Prevention of Cancer (NPC) trial reported a significant 65% reduction in prostate cancer incidence in men receiving selenium supplementation of 200 mg daily. However, 12 years later, SELECT (the Selenium and Vitamin E Cancer Prevention Trial) found that neither selenium (200 mg daily) nor vitamin E supplementation, alone or in combination, had any effect on incidence of prostate cancer. The null results of SELECT and the concurrent Physicians’ Health Study II trial of vitamin E, together with other previous vitamin supplementation trials, raised concerns about the benefit of nutritional supplements for reducing cancer risk and all-cause mortality. How can we interpret these apparently inconsistent results for selenium and prostate cancer across different studies over the past two decades? Several questions arise. First, can selenium function as a chemopreventive agent? Selenium is an essential trace element and plays a pivotal role in normal physiology. Many experimental and population studies have demonstrated that low selenium status was associated with increased risk of prostate cancer whereas supplementation prevented the cancer, as shown by the NPC trial. One key question is “who is most likely to benefit from the intervention?” This group could include men with low baseline levels of selenium as well as men with a genetic predisposition for greater sensitivity to the ill effects of selenium deficiency. It is well known that there is a narrow range between selenium deficiency and toxicity, and a U-shape effect has been suggested by animal experiments. Indeed, even the first prospective study published in 1983 by Willett, Stampfer, and colleagues showed an apparent threshold in the association between selenium status and cancer; only individuals in the lowest quintile of serum selenium had a significantly increased risk of cancer. This observation was subsequently supported by several prospective studies of prostate cancer and selenium levels measured either in blood or toenails. The NCP trial further confirmed that Se supplementation was beneficial but only among those whose baseline selenium level was below 123 ng/ml. Interestingly, 75% of the subjects enrolled in SELECT had initial plasma levels of selenium higher than this level. Our recent research of genetic polymorphisms on antioxidant pathways such as manganese superoxide dismutase (MnSOD) and 15-kDa selenoprotein (Sep15) further demonstrated that common variants in certain antioxidant genes may influence both the development and progression of prostate cancer by modifying the effects of plasma selenium and other antioxidants, suggesting the possibility that some men may benefit more from antioxidants than others depending on their genotypes. These findings could explain why large randomized trials have failed to show any meaningful results in cancer prevention. Finally, when in the process of tumorigenesis might selenium chemoprevention be effective? It remains unclear at what stage of prostate cancer (initiation or progression) selenium can be most effective. Given our observation of selenium and advanced prostate cancer or cancers with high baseline PSA, it could be that selenium protects against tumor progression. This could largely explain the null effect of selenium on incident prostate cancer in the SELECT trial, in which 86% of men reported receiving PSA tests each year and 98.9% of prostate cancers were localized diseases. The fact that many PSA-detected prostate cancers are indolent and may never turn out to be life-threatening diseases, and that the short follow-up time of SELECT relative to the lead time provided by PSA testing, could also explain the null effect of selenium observed in the SELECT trial. In summary, a better understanding of selenium biology requires additional mechanistic, genetic and genomic approaches, a better characterization of high-risk prostate cancer in the era of PSA screening, which should be informative in designing future clinical studies that could target on a subset of the high-risk population with a certain disease state or selenium status, or even on individuals of a certain genotype, that can benefit most from this micronutrient. Citation Information: Cancer Prev Res 2010;3(12 Suppl):PL01-05.