False presumptions and continued surprises: how much do we really know about nutritional supplements and cancer risk?

Studies of dietary supplementation and cancer risk are full of surprises. Ten years ago, most epidemiologists and nutritionists sincerely believed that fl-carotene was the leading candidate for evaluation in chemoprevention trials. We now know that (3-carotene supplementation among smokers increases the risks of lung cancer, cardiovascular disease, and total mortality (1, 2). Although the story is incomplete, the recent findings on selemum supplementation were equally unanticipated. Until recently, few scientists would have argued that selenium supplementation was likely to be a chemopreventive agent active against a variety of cancers. However, in a large clinical trial, selenium supplementation reduced total cancer incidence by 59% (3). There are other less extreme examples of where our expectations about dietary supplements were off the mark. Consider the Finnish study of (3-carotene and a-tocopherol to prevent lung cancer, in which a-tocopherol supplementation