This study examined the possible effect of cytochrome P450 (CYP1A1), glutathione S-transferase (GSTM1 and T1) and N-acetyltransferases 2 (NAT2) polymorphisms on DNA-protein crosslinks (DPC) formation in the white blood cells of breast cancer patients, and assessed the levels of DPC detected. Sixty cases of breast cancer were examined, all involving women diagnosed with primary, histopathologically confirmed breast cancer at the Chinese Medical College Hospital in central Taiwan. Additionally, 60 healthy women without breast cancer were selected as a control group, matched by age, cigarette smoking habits, and history of breast cancer among first-degree relatives. Known risk factors for breast cancer, including menarche before 13 years of age (OR=3.2; CI, 1.1-9.5), no history of breast-feeding (OR=4.7; CI, 1.5-14.4) and use of oral contraceptives (OR=9.1; CI, 2.8-29.8), were found to be significantly associated with breast cancer. For the CYP1A1 MspI polymorphism, 16.7 and 18.3% of cases and controls, respectively contained both alleles with the MspI restriction fragment length polymorphism (RFLP). Regarding the NAT2 allele, 25.0 and 21.7% of cases and controls carried slow genotypes. For GSTM1 and GSTT1, 56.7 and 45.0% of cases, as well as 58.3 and 43.3% of controls, contained the null genotype. Meanwhile, chi(2)-tests found no significant differences between the groups. After controlling for confounders such as cigarette smoking and family history of breast cancer, the DPC value of the case group significantly exceeded that of the control group (1.62% versus 0.98%, P<0.001). In conclusion, our findings were inconsistent with those of previous studies that showed polymorphism genes (CYP1A1, NAT2, GSTM1 and GSTT1) were associated with cancer risk. However, this study indicated that genotypic variants of these polymorphisms did not elevate the risk for breast cancer, individually or interactively. Additionally, this investigation represents the first description of the use of DPC as a biomarker to assess the level of DNA damage of breast cancer patients. Our data suggest that the DPC method is a useful tool for detecting DNA damage, and DPC formation may be associated with the induction of breast cancer.