Background: Recent evidence suggests that in addition to calcium homeostasis, vitamin D has immunomodulatory activity. Thus, we hypothesized that genetic variation in the vitamin D metabolism pathway may alter risk of graft-vs-host disease (GVHD) following blood and marrow transplantation (BMT). We evaluated the associations between single nucleotide polymorphisms (SNPs) in vitamin D synthesis (CYP2R1, CYP3A4, CYP27A1, CYP27B1), transport (vitamin D binding protein), gene expression (vitamin D receptor, VDR), and catabolism (CYP24A1) and risk of acute GVHD (aGVHD, grades II-IV) to day 100 and chronic GVHD (cGVHD) to 2 years following BMT. Methods: DNA was obtained for 437 individuals who underwent an allogeneic BMT for hematologic malignancies at the University of Minnesota between 1995-2005. A total of 86 SNPs were selected for genotyping, including Hapmap tagSNPs and those documented as having functional relevance. Proportional hazard regression modeling was used to evaluate associations between SNPs and aGVHD and cGVHD, treating SNPs as continuous variables. Statistical significance was defined as p≤0.01 to minimize the likelihood of reporting false positive findings. Results: After adjustment for race, age at transplant, donor type (sibling/unrelated/cord blood), disease status (standard/high risk), conditioning regimen and GVHD prophylaxis, 5 SNPs in VDR were statistically significantly associated with increased cGVHD risk. Four of the SNPs (rs2525044, 7305032, 11168268, 987849), all located in introns, were in strong linkage disequilibrium (R²>0.8). Rs10783215, located in the 3'UTR, was associated with the highest risk of cGVHD (HR 1.4, 95% CI: 1.1-1.8, p<0.01). VDR rs11168268 was associated with aGVHD incidence (p=0.03), but did not meet the p≤0.01 criteria. Unlike two smaller previous studies, we did not observe an association between the VDR FokI (rs2228570), BsmI (rs1544410), ApaI (rs7975232), and TaqI (rs731236) and risk of aGVHD. Conclusions: While preliminary, our study suggests that VDR genetic variation may alter risk of cGVHD, but not aGVHD. Further studies are needed to confirm these findings, and should consider potential effect modification by vitamin D status.