Genetic metabolic polymorphisms and the risk of cancer: a review of the literature

Abstract The purpose of this paper is to systematically analyse the design and results of epidemiological studies on the association between various types of cancer (lung, bladder, breast, colon, stomach) and four genetically-based metabolic polymorphisms, involved in the metabolism of several carcinogens (glutathione-S-transferase M1, debrisoquine hydroxylase, N acetyltransferase, aryl hydrocarbon hydroxylase). These inherited polymorphisms usually cause modifications in the quality or quantity of the relevant enzymes. Such enzymes are involved in the activation/inactivation of known carcinogens and seem to modify the extent to which carcinogens interact with DNA in target tissues. Two enzymes, debrisoquine hydroxylase and aryl hydrocarbon hydroxylase, activate procarcinogens to carcinogens (phase I enzymes). The other two, glutathione-S-transferase M1 and N-acetyltransferase, mainly detoxity carcinogenic substances (phase II enzymes). Because of their role as host factors (modulating the action of carcinogens), it has been hypothesized that subjects presenting a specific phenotype for such polymorphisms could be at a greater risk of developing various types of cancer. A number of epidemiological studies have investigated such associations, often with discordant results. We examine and discuss the design of the studies, and present a meta-analysis of the available data.