Elevated levels of chromosomal aberrations (CAs) in peripheral blood lymphocytes, widely used as a cytogenetic biomarker of genotoxic effects, have been linked to cancer predisposition. However, tobacco smoking, occupational carcinogen exposure, or time since CA analysis do not appear to explain the cancer predictivity of CAs. Alternatively, the observed CA-cancer association could reflect unidentified exposures or individual susceptibility. We assessed the effects of genetic polymorphisms of DNA repair proteins and xenobiotic-metabolizing enzymes (XMEs) on the levels of CAs and sister chromatid exchanges (SCEs) in peripheral lymphocytes of 145 (CAs) and 60 (SCEs) healthy Caucasians. Genotypes of DNA repair genes X-ray repair cross-complementation group 1 ( XRCC1 codons 194, 280, 399) and 3 ( XRCC3 codon 124), and XME genes glutathione- S-transferase ( GST) M1 and T1 and N-acetyl transferase 2 ( NAT2) were determined using polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP)-based methods. After Poisson regression adjustment for age, sex, smoking, country, and genotypes, a higher frequency of chromosome-type breaks was observed for NAT2 slow acetylators (in nonsmokers) and GSTT1 null subjects (in smokers). Individuals carrying variant alleles for XRCC1 codons 280 and 194 showed a decreased level of chromosome-type breaks. The effect of GSTM1 null and XRCC1 codon 399 genotypes on the frequency of CAs was modified by smoking. In linear regression models adjusting for age, sex, smoking, and genotypes, none of the polymorphisms significantly affected SCE frequency, although GSTT1 null subjects had a slightly elevated SCE level. Our results are in line with earlier findings on the influence of NAT2, GSTT1, and GSTM1 polymorphisms on the level of lymphocyte chromosome damage and suggest that also XRCC1 polymorphism affects CA frequencies, thus apparently influencing DNA repair phenotype. It remains to be examined whether these or other genetic polymorphisms could explain the observed cancer risk predictivity of high CA frequency.