Shorter alleles of a polymorphic [CAG](n) repeat in exon 1 of the androgen receptor (AR) have been associated with increased risk of prostate cancer and decreased risk of breast cancer. We prospectively assessed the association between the [CAG](n) repeat polymorphism in the androgen receptor and breast cancer risk among Caucasian women in a case-control study nested within the Nurses' Health Study cohort (cases, n = 727; controls, n = 969). In addition, we assessed whether androgen receptor genotype influences endogenous steroid hormone levels in women and whether the associations between androgen receptor alleles and breast cancer risk differed according to established breast cancer risk factors. Women with one or more long AR [CAG](n) repeat alleles (>or=22 repeats) were not at increased risk of breast cancer [odds ratio (OR), 1.06; 95% confidence interval (CI), 0.83-1.35]. Significant associations were not observed between AR genotypes comprised of two short alleles ([CAG](n) or=22: OR, 0.92; 95% CI, 0.62-1.36) or two long alleles ([CAG](n) >or= 25 versus both alleles or=22; OR, 1.70; 95% CI, 1.20-2.40; P for interaction = 0.04). In summary, we observed no overall relation of AR genotype with breast cancer risk among mostly postmenopausal Caucasian women. However, these data suggest that longer AR [CAG](n) repeat alleles may increase breast cancer risk among women with a first-degree family history of breast cancer.