In a comprehensive analysis of common polymorphic variation in vitamin D pathway genes, we found only weak associations with prostate cancer risk overall. However, in men with low serum 25(OH)-vitamin D status, tag SNPS in or near the 3′ UTR of VDR were associated with prostate cancer risk, suggesting an important gene–environment interaction. In addition, SNPS in GC, which encodes the major carrier protein for 25(OH)D, were moderately associated with serum 25(OH)D concentration, but not with prostate cancer risk.
, a key mediator of the biological actions of 1,25(OH)2
D, possesses adenylate/uridylate-rich elements (AUUUA) in the 3′ UTR that regulate messenger RNA stability (35
), potentially altering VDR
expression and the cellular response to changing circulating vitamin D levels. SNPS in this region have been linked to higher VDR
messenger RNA stability and higher gene transcription activity in some but not all in vitro
). Of the three SNPS (rs11574143, rs757343 and rs1544410) in the 3′ UTR region related to prostate cancer in men with low serum 25(OH)D in our study, findings were strongest for rs11574143 (P
0.0007). However, our sliding window haplotype analysis suggested that the effect might be due to additional unmeasured risk variants, in LD with rs11574143, rs757343 and rs1544410. Future studies are warranted to examine comprehensively the effect of SNPS in the 3′ UTR on VDR expression.
Our results for rs1544410 (BsmI) are consistent with the Physician's Health Study's findings, both showing that carriage of the rs1544410 G allele is associated with higher risks of prostate cancer only in men with lower 25(OH)D levels (23
), whereas in the Health Professionals Follow-up Study risks tended also to be increased with carriage of the G allele, but differences by serum 25(OH) were not seen (10
A recent prostate cancer case–control study by Holick et al.
) examined 22 tag SNPS across the VDR
region and observed associations with two tag SNPS and prostate cancer risk: rs2107301, which we genotyped and found to be null, and rs2238135, which is in high LD with our genotyped rs2853564 (D′
1.0; ~300 bp distance) and also found to be null. rs2107301 is, however, located in intron 4 and in weak LD with the risk-related SNPS in our study (), whereas rs2238135 is located upstream in a different LD block. Our findings of no association between tag SNPS in CYP27B1
with prostate cancer risk are consistent with the Holick et al.
). Other genetic studies have been limited to selected VDR
RFLP SNPS [Bsm1 (rs1544410), ApaI (rs7975232), Fok1 (rs10735810) and Cdx2 (rs11568820)] and have shown inconsistent results (20
Interactions between two additional VDR
RFLPs (FokI and Cdx2) and serum vitamin D levels have been examined in relation to prostate cancer risk (10
). Fok I (rs10735810), located in exon 2 of the VDR
gene, may confer altered VDR transcriptional activity (41
). The Cdx2
SNP (rs11568820), located in the 5′ regulatory region of the VDR gene, may alter the affinity of the Cdx2 transcription factor for the VDR promoter, potentially affecting VDR expression (42
). We found no overall association or significant interactions between FokI (rs10735810) and serum vitamin D in relation to prostate cancer risk. Other studies have been inconsistent, finding no association (10
), lower risks related to the variant allele, in men with low serum 25(OH)D status (11
), and stronger risk in men with high sunlight exposure, presumably correlated with higher 25(OH)D status (42
). We also found no overall association or interactions with 25(OH)D for Cdx2 (rs11568820), although the Health Professionals Follow-up Study reported lower risks related to the Cdx2 (rs11568820) variant allele in men with low 25(OH)D (10
Serum 25(OH)D concentration in blood is largely determined by non-genetic factors (15
), including sunlight and diet; we found only modest differential in season-standardized 25(OH)D concentrations by variants in GC
, which encodes vitamin D-binding protein, the major protein carrier of 25(OH)D. Our finding that associations remained for rs2282679 and rs1155563 in the simultaneously adjusted model may indicate additional unmeasured variants linked to rs2282679 and rs1155563 may play an important role in determining serum 25(OH)D concentrations. Serum 1,25(OH)2
D is tightly regulated (15
), and we found no relationship with the studied gene variants; however, with only a subset of controls for this analysis (n
150), we may not have had sufficient power to detect SNP–serum 1,25(OH)2
Although tag SNPS in GC
were related to serum 25(OH)D, these SNPS were not associated with total or aggressive prostate cancer risk. We recently reported no statistically significant differences in the PLCO Trial in the risk of prostate cancer with increasing serum 25(OH)D concentration, although there was some evidence of increasing risks for aggressive disease (12
). Considering the modest impact of GC
polymorphisms on serum 25(OH)D (i.e. mean difference: 8 nmol/l for GC rs2282679 homozygous versus variant homozygous), substantial gene–prostate cancer associations would not be expected.
This is the first large-scale evaluation of both serum vitamin D status and candidate genes in the vitamin D pathway in relation to prostate cancer. Since this study was conducted in a cancer screening trial, study participants had the same protocol for prostate cancer detection irrespective of lifestyle factors, substantially reducing the likelihood of screening-related detection bias. Other strengths include the use of prediagnostic serum vitamin D measurements. A potential limitation of our study is measurement of only a single serum sample; 25(OH)D measures at multiple time points would have resulted in more precise estimates of exposure.
In a broad evaluation of genetic variation in vitamin D-related genes, we found no overall associations with prostate cancer; however, genetic variants in the 3′ region of VDR, particularly rs11574143, may be associated with risk of this disease in men with low vitamin D status, indicating that knowledge of both serum 25(OH)D levels and VDR genotype may be needed to understand the complex relationship between vitamin D and prostate cancer risk.