One promising area of current research in chemoprevention is the possibility that micronutrients, including vitamin A analogues, may decrease cancer incidence. The term "vitamin A" refers either to retinol (preformed vitamin A) and its synthetic analogues, or to certain carotenoids (provitamin A), which are converted to retinol in the body, as needed. Retinol and the other retinoids are integrally involved in cell growth and differentiation, which may affect carcinogenesis. Such a role has been supported by a large number of in vitro and animal experiments. Data from studies among humans are sparse, in part because most dietary studies have been conducted in populations in which the vast majority of vitamin A intake is actually beta-carotene and other carotenoids, found in carrots and other vegetables and fruits. Although the carotenoids do not have the hormone-like properties of retinol, they do have a potent antioxidant effect and could thus reduce cancer risk by preventing tissue damage due to oxidation. This possibility is supported by data from a large number of observational epidemiologic studies, most of which have consistently found an inverse relation between consumption of carotene-rich vegetables and cancer risk. However, the only direct way to determine whether carotenoids have a beneficial effect is through large, carefully conducted randomized trials. Several such studies are currently underway and should provide sound evidence on which future medical policy and practice can be based.