A large number of studies have reported associations between polymorphisms of xenobiotic-metabolizing enzymes (XMEs) and various cancers. However, the carcinogenic exposures behind such findings have usually been unclear. Information on susceptibility to specific carcinogens could better be obtained by examining situations where the exposure and the endpoint studied are nearer in time, i.e., by studying biomarkers of carcinogen exposure and early (genotoxic) effect in exposed humans. For example, analyses of DNA adducts and cytogenetic endpoints have indicated an increased susceptibility of glutathione S-transferase M1 (GSTM1) null genotype to genotoxicity of tobacco smoking, supporting the view that the associations of the GSTM1 null genotype with bladder and lung cancer are partly related to smoking. In vitro genotoxicity studies with human cells offer an experimental tool that can be used, within the limits of the cell systems, to predict individual sensitivity and genotype-carcinogen interactions. In vitro sensitivity to the genotoxicity of 1,2:3,4-diepoxybutane, an epoxide metabolite of 1,3-butadiene has clearly been shown to depend on GSTT1 genotype, which has also been implicated to modify, along with GSTM1 genotype, the in vitro genotoxicity of 1,2-epoxy-3-butene, another epoxide metabolite of 1,3-butadiene. These genotypes appear to modulate the excretion of 1,3-butadiene-specific mercapturic acids, and influence genotoxicity biomarker levels in 1,3-butadiene-exposed workers. The excretion of specific mercapturic acids (PHEMA) in workers exposed to styrene has clearly been shown to depend on GSTM1 genotype, and GSTT1 genotype seems to modulate the excretion of one PHEMA diastereoisomer. These genotypes have also been implicated to modulate the in vitro genotoxicity of styrene. In general, the genetic polymorphisms potentially important for biomarker response largely depend on the exposing agent, biological material examined, and ethnicity of the population under study. Individual exposure level may vary a lot, and a reliable estimate of the exposure is essential for correct interpretation of genotype-exposure interaction. Besides XME polymorphisms, any polymorphisms that affect cellular response to DNA damage could, in principle, modify individual sensitivity to genotoxins. For instance, those concerning DNA repair proteins are presently being studied by many laboratories.