Abstract PL04-04: Vitamin E and selenium for prostate cancer prevention: The SELECT experience

Abstract

Abstract The Selenium and Vitamin E Cancer Prevention Trial (SELECT) tested its namesake agents in the prevention of prostate cancer, a continuing major health threat, especially for African American men, in the United States, including Puerto Rico, and Canada. Designed as a large, simple trial conforming as closely as possible with community standards of care, SELECT was a phase III randomized, placebo-controlled trial of selenium (200 g/day from L-selenomethionine) and/or vitamin E (400 IU/day of all rac -tocopheryl acetate) in men at least 55 years of age (non-African American) or at least 50 years of age (African American) with a serum prostate-specific antigen (PSA) ≤ 4 ng/mL and a digital rectal examination (DRE) not suspicious for prostate cancer. The planned intervention duration was a minimum of 7 years, maximum of 12 years. The strongest evidence supporting the SELECT hypothesis came from secondary clinical results indicating that prostate cancer risk was reduced by vitamin E in the Alpha-Tocopherol, Beta-Carotene (ATBC) Study (designed to prevent lung cancer) and by selenium in the Nutritional Prevention of Cancer (NPC) trial (designed to prevent skin cancer); this evidence was supported by epidemiologic and preclinical data. 35,533 men were randomly assigned at a faster-than-expected accrual rate to the four, well-balanced study arms (selenium, vitamin E, selenium plus vitamin E, placebo) between 2001 and 2004. As a result of a planned interim analysis (at 7 years) and by recommendation of the Data and Safety Monitoring Committee, the interventions were discontinued because of convincing evidence of no benefit from either study agent (p < 0.0001) and no possibility of a benefit to the planned degree with additional follow-up. With a median overall follow-up of 5.46 years (range, 4.17 to 7.33 years), the primary analysis showed the following prostate cancer results (versus placebo, n=416): vitamin E hazard ratio (HR) = 1.13 (n=473; 99% confidence interval [CI], 0.95 - 1.35), selenium HR = 1.04 (n=432; 99% CI, 0.87 – 1.24), combination HR = 1.05 (n=437; 99% CI, 0.88 – 1.25). No significant differences were seen in any prespecified cancer endpoints including lung, colorectal, and overall cancer. A nonsignificant increased risk of prostate cancer occurred in the vitamin E arm (p=0.06; relative risk [RR]=1.13; 99% CI, 0.95 – 1.35), and the selenium arm showed a nonsignificant increased risk of Type 2 diabetes mellitus (p=0.16; RR=1.07; 99% CI, 0.94–1.22); neither of these increases appeared in the combination arm. Selenium, vitamin E, and their combination did not reduce the risk of prostate cancer in SELECT. One important lesson learned from the SELECT experience, building on lessons learned from the earlier ATBC and NPC experiences, is that we should step back and do more small, lead-up preclinical and early-phase clinical studies that increase our understanding of cancer biology and agent effects before moving ahead with large clinical trials of natural or other potential preventive agents. Epidemiologic and secondary clinical data are great pathfinders, but the heaviest supportive evidence for definitive trials may come mainly from more extensive preliminary research, including mechanistic studies that can validate the biologic plausibility of the clinical, risk reduction hypothesis. Another important lesson SELECT has provided for any large prevention trial is built-in opportunity for translational research. The SELECT design comprised sub-studies of molecular and standard epidemiology, diet and nutrition, and other important correlative endpoints that informed the collection of data and biospecimens in the trial and resulted in an invaluable repository for future research. Support for this work was received from Public Health Service Cooperative Agreement grant CA37429 (National Cancer Institute, National Institutes of Health, Department of Health and Human Services) and from the National Center for Complementary and Alternative Medicine (National Institutes of Health). Citation Information: Cancer Prev Res 2010;3(1 Suppl):PL04-04.