We investigated the association between polymorphism in the androgen receptor (AR) and vitamin D receptor (VDR) genes and breast cancer risk in a large population-based case-control study of genetically homogenous Swedish women. We successfully determined both AR CAG(n) and VDR A(n) genotype in 1,502 women with invasive breast cancer and in 1,510 control women. We did not find any associations between AR or VDR microsatellite lengths and breast cancer when we used a priori determined cutoffs (</=21 or >/=22 repeats for AR and </=18 or >/=19 for VDR) to define long and short alleles. There was statistically significant interaction between VDR genotype and parity, such that women with two short alleles had a halved risk for breast cancer, irrespective of parity, compared with nulliparous women with two long alleles. Homozygosity for the long VDR allele was associated with a more advanced clinical stage at diagnosis. In exploratory analyses, we determined cutoffs based on visual inspection of distributions of allele lengths among cases and controls and found that women carrying two alleles with <20 AR CAG(n) repeats had an increased risk for breast cancer, odds ratio of 1.67 (95% confidence interval, 1.17-2.38), compared with those with two alleles with >/=20 repeats. Women carrying two VDR alleles with <21 A(n) were also at an increased risk, odds ratio of 1.26 (95% confidence interval, 1.04-1.51). Our data do not support major roles for AR or VDR polymorphism as breast cancer risk factors. However, we did find an interaction between VDR genotype and parity that remains to be corroborated.