Although many studies have examined associations between single nucleotide polymorphisms (SNPs) in the CYP1A1, CYP1A2 and CYP1B1 genes and breast cancer risk, no study has examined functional SNPs in the CYP3A5 gene and only a small number of studies have been investigated in Japanese populations. To examine the association between six SNPs, CYP1A1(*)2A, CYP1A1(*)2C, CYP1A2(*)1F, CYP1B1 Arg(48)Gly, CYP1B1 Leu(432)Val and CYP3A5*3 and breast cancer risk, therefore, we conducted hospital-based case-control studies in Nagano, Japan and São Paulo, Brazil including 873 pairs (403 Japanese (JJ), 81 Japanese Brazilians (JB) and 389 non-Japanese Brazilians (NJB)). Although we found no significant association in the three populations combined, subgroup analyses revealed statistically significant associations of CYP1A2*1F in NJB, and CYP1B1 Leu(432)Val and CYP3A5*3 in JJ with breast cancer risk. Compared to women with the AA genotype in CYP1A2*1F, the odds ratio (OR) (95% confidence interval (CI)) for NJB with the CC genotype was 0.54 (0.32-0.90); that for JJ with Leu/Val+Val/Val versus Leu/Leu genotype in CYP1B1 Leu(432)Val was 0.68 (0.48-0.97); and that for JJ with (*)3/(*)1+(*)1/(*)1 versus (*)3/(*)3 genotype in CYP3A5*3 was 1.49 (1.10-2.04). Our findings provide further evidence that genetic polymorphisms related to estrogen metabolism may play a role in the development of breast cancer.