Epidemiologic studies have shown an inverse association between isoflavones and breast cancer risk. Because isoflavones bind estrogen receptors, we hypothesized that polymorphisms in the estrogen receptor genes might modify the association between isoflavone intake and breast cancer risk. We conducted hospital-based case-control studies of patients aged 20-74 years with primary, incident, histologically confirmed invasive breast cancer, and matched controls from among medical checkup examinees in Nagano, Japan, and from cancer-free patients in São Paulo, Brazil. A total of 846 pairs (388 Japanese, 79 Japanese Brazilians and 379 non-Japanese Brazilians) completed validated food frequency questionnaires, and provided blood samples. Five single nucleotide polymorphisms in the estrogen receptor alpha (rs9340799, rs1913474, and rs2234693) and beta (rs4986938 and rs1256049) genes were genotyped. We found no consistent association between the five single nucleotide polymorphisms and breast cancer risk among the three populations. In analyses of combinations of isoflavone intake and single nucleotide polymorphisms, an inverse association between intake and risk was limited to women with the GG genotype of the rs4986938 polymorphism for postmenopausal Japanese (odds ratio for highest versus lowest tertile = 0.47; P for trend = 0.01), Japanese Brazilians (odds ratio for highest versus lowest median = 0.31) and non-Japanese Brazilians (odds ratio for consumers versus non-consumers = 0.37) (P for interaction = 0.11, 0.08, and 0.21, respectively). We found no remarkable difference for the other four polymorphisms. Our findings suggest that polymorphisms in the estrogen receptor beta gene may modify the association between isoflavone intake and breast cancer risk.