The possibility that dietary intake of certain
vitamins and minerals may influence the occurrence of human
cancer is receiving considerable scientific attention. One prominent hypothesis, that increased dietary intake of
vitamin A reduces the occurrence of
cancer, has received support from a large number of epidemiologic studies in which an inverse association was observed. The largest body of evidence relates to
lung cancer. However, when examined in further detail, this apparent protective effect appears primarily attributable to higher intakes of green and yellow vegetables, which contain the
carotenoid precursors of
vitamin A. In contrast, there is little evidence to support an association between preformed
vitamin A intake and
cancer risk. In several studies based on prospectively collected sera,
retinol levels were inversely related to subsequent
cancer risk. However, these have not been supported by further investigations and appear to be the result of methodologic artifact. Available evidence thus suggests that factors associated with green and yellow vegetables provides modest protection against certain forms of
cancer;
beta-carotene is a likely candidate and is the focus of considerable research activity. Stimulated by the results of many animal studies and ecologic comparisons, we and other investigators have examined the association of serum
selenium levels with subsequent risk of
cancer. In the three published prospective studies an inverse association was observed, with a 2-to 5-fold increase in overall
cancer risk among those with lowest
selenium levels. As would be predicted by animal studies, the combination of low
selenium and low
vitamin E appears to be particularly deleterious.(ABSTRACT TRUNCATED AT 250 WORDS)