5813

Background: Circulating 1,25-dihydroxyvitamin D (1,25(OH)2D) has potent effects on cell growth and differentiation via binding with the nuclear vitamin D receptor (VDR). Three SNPs in the VDR gene (BsmI, TaqI and ApaI) are in strong linkage disequilibrium and have previously been associated with breast cancer risk in some studies, possibly through linkage with a variable length poly (A) in the 3’ untranslated region. High dietary calcium intake may modify associations between VDR genotype and breast cancer risk by down-regulating circulating 1,25(OH)2D levels. Methods: We conducted a nested case control study of the association of BsmI, ApaI and TaqI VDR genotypes, and their associated haplotypes with postmenopausal breast cancer risk among women in the American Cancer Society’s Cancer Prevention Study II Nutrition Cohort. Participants completed a 10-page questionnaire in 1992 on diet, medical history and breast cancer risk factors. Among 21,965 women who gave blood in 1998-2001 and were cancer free at baseline in 1992, 498 breast cancer cases were diagnosed between 1992-2001; 502 controls were matched on age, race/ethnicity, and date of blood collection. DNA was extracted from buffy coat and genotyping assays were performed using Taqman (Applied Biosystems, Foster City, CA). Unconditional logistic regression was used to examine the association between each polymorphism and breast cancer risk while controlling for matching factors and other possible risk factors. Results: Associations were similar among those with one or two variant alleles or haplotype copies; consequently we present them combined. Breast cancer risk was not statistically different in individuals with a variant allele for BsmI (BB/Bb vs bb): OR=1.21, 95% CI 0.91-1.62 or TaqI (tt/tT vs TT): OR=1.19, 95% CI 0.90-1.57, or the ApaI homozygous wildtype (AA vs Aa/aa): OR=1.20, 95% CI 0.89-1.61. Having at least one copy (compared to no copies) of the BAt haplotype yielded similar results (OR=1.21, 95% CI 0.92-1.60). However, among women with greater than median intake of total calcium (diet plus supplements, ≥902 mg/day vs. <902 mg), we observed increased risk for BsmI (BB/Bb vs bb): OR=1.85, 95% CI 1.21-2.82, p interaction=0.02, TaqI (tt/Tt vs TT): OR=1.86, 95% CI=1.23-2.80, p interaction=0.005 and ApaI (AA vs Aa/aa): OR=1.45, 95% CI 0.94-2.24, p interaction=0.25, as well as for the BAt haplotype (OR=1.88, 95% CI 1.25-2.82, p interaction=0.004). For women with < median calcium intake, the ORs were close to 1.00 and were not statistically significant. Conclusions: We did not observe significant associations between VDR genotypes BsmI, ApaI and TaqI or the BAt haplotype and breast cancer risk. However, dietary factors that influence calcium and vitamin D metabolism may modify associations by genotype.

[Proc Amer Assoc Cancer Res, Volume 46, 2005]