Circulating vitamin D, vitamin D-related genetic variation, and risk of fatal prostate cancer in the National Cancer Institute Breast and Prostate Cancer Cohort Consortium.
Shui IM., Mondul AM., Lindström S., Tsilidis KK., Travis RC., Gerke T., Albanes D., Mucci LA., Giovannucci E., Kraft P., Breast and Prostate Cancer Cohort Consortium Group None.
BACKGROUND: Evidence from experimental animal and cell line studies supports a beneficial role for vitamin D in prostate cancer (PCa). Although the results from human studies have been mainly null for overall PCa risk, there may be a benefit for survival. This study assessed the associations of circulating 25-hydroxyvitamin D (25(OH)D) and common variations in key vitamin D-related genes with fatal PCa. METHODS: In a large cohort consortium, 518 fatal cases and 2986 controls with 25(OH)D data were identified. Genotyping information for 91 single-nucleotide polymorphisms (SNPs) in 7 vitamin D-related genes (vitamin D receptor, group-specific component, cytochrome P450 27A1 [CYP27A1], CYP27B1, CYP24A1, CYP2R1, and retinoid X receptor α) was available for 496 fatal cases and 3577 controls. Unconditional logistic regression was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for the associations of 25(OH)D and SNPs with fatal PCa. The study also tested for 25(OH)D-SNP interactions among 264 fatal cases and 1169 controls. RESULTS: No statistically significant relationship was observed between 25(OH)D and fatal PCa (OR for extreme quartiles, 0.86; 95% CI, 0.65-1.14; P for trend = .22) or the main effects of the SNPs and fatal PCa. There was evidence suggesting that associations of several SNPs, including 5 related to circulating 25(OH)D, with fatal PCa were modified by 25(OH)D. Individually, these associations did not remain significant after multiple testing; however, the P value for the set-based test for CYP2R1 was .002. CONCLUSIONS: Statistically significant associations were not observed for either 25(OH)D or vitamin D-related SNPs with fatal PCa. The effect modification of 25(OH)D associations by biologically plausible genetic variation may deserve further exploration.