Vitamin D Receptor and Calcium Sensing Receptor Polymorphisms and the Risk of Colorectal Cancer in European Populations

Higher blood concentration of vitamin D has been shown to be colorectal cancer (CRC) protective in different populations (1-4). Although the main function of vitamin D is calcium homeostasis, it has also been shown to be involved in the modulation of cell cycle kinetics in the colorectum and other tissues (5-7). Many of the actions of vitamin D are thought to be mediated by the vitamin D receptor (VDR), a member of the nuclear receptor superfamily that is present in different cell types. The VDR can also regulate other vitamin D inducible genes involved in processes of inflammation, immune function, estrogen metabolism, insulin-like growth factor-I signaling, and regulation of intestinal calcium absorption (8-10). Its importance in modulating cancer risk has been experimentally highlighted in vitro and in vivo (8).

It is possible that the effects of vitamin D may differ among individuals depending on variations in the activity of the VDR. A number of VDR polymorphisms have been identified [BsmI, TaqI, Tru91, Apa1, and polyA—all in linkage disequilibrium; Fok1 and CDX2 (11)], and although some are thought to be functionally important (12), much remains unknown about how they may affect carcinogenesis. Nevertheless, given the importance of the gene, a number of studies have explored the role of VDR polymorphisms in modulating cancer risk in various tissues.

With respect to colorectal adenomas, epidemiologic studies conducted primarily in North American populations have shown no associations for the BsmI (13-15), TaqI (15-17), FokI (15, 17, 18), or other VDR genotypes (15). However, in other studies, an effect modification has been observed with lower dietary intake levels of either calcium, vitamin D, or dairy products leading to reduced risk of colorectal adenomas with the BB genotype of the BsmI (14), the ff genotype of the Fok1 (13), or the aA/AA genotype of the ApaI VDR genotypes (15). Considering cancer end points, studies on North American populations suggest that the BsmI BB or PolyA SS VDR genotypes may be associated with a reduced risk of cancer in the colon (19) and possibly in the rectum (20). In addition, the VDR Fok1 FF genotype has been associated with an increased risk of colon cancer in some North American (21) and South Asian populations (22). However, no data are currently available for any of these genotypes from European populations, and it is unclear to what extent any VDR-CRC risk association may be dependent on circulating vitamin D (25OHD) concentrations or level of dietary calcium intake.

Vitamin D metabolism may also be affected by signaling from the calcium sensing receptor (CASR), which is key to extracellular calcium homeostasis, plays a role in cellular growth kinetics, and influences parathyroid hormone secretion (23, 24). It has been suggested that the CASR may have a role in carcinogenesis (25). In a recent study of haplotypes based on three CASR polymoprhisms, the most common haplotype pair was associated with an increased risk of advanced adenoma compared with the next three common pairs (26). But it is not yet clear whether variability in the CASR gene may influence CRC risk.

A nested case-control study was conducted within the European Prospective Investigation into Cancer and Nutrition (EPIC) to assess the role of VDR and CASR polymorphisms in CRC risk and determine any possible interactions with serum vitamin D concentration, dietary calcium intake, and other dietary and life-style parameters.

The rationale and methods of the EPIC study including information on dietary assessment methodology, blood collection protocols, and follow-up procedures have been previously reviewed in detail (27). EPIC is a large prospective study with over 520,000 subjects enrolled and consists of 23 centers in Denmark, France, Greece, Germany, Italy, the Netherlands, Norway, Spain, Sweden, and United Kingdom. Between 1992 and 1998, standardized life-style and personal history questionnaires, detailed dietary intake assessments (validated country-specific questionnaires; diet over previous 12 mo), anthropometric data, and blood samples were collected from most participants. Values for dietary intake of vitamin D and calcium were computed using country-specific food composition tables (no intake data for these variables was available from Greece). The present study was approved by the Ethical Review Board of the IARC and those of all EPIC centers. All EPIC participants have provided written consent for the use of their blood samples and all data.

For the present study, cancers were defined according to the 10th Revision of the International Statistical Classification of Diseases, Injury, and Causes of Death as colon (C18.0-C18.9) and rectum (C19, C20). Anal canal tumors were excluded. CRC is defined as a combination of the colon and rectal cancer cases.

After exclusions (56 cases for missing information on fasting status, 31 cases due to missing laboratory 25OHD data from either assay failure or insufficient serum volume in either member of a case-control set), a total of 1,248 first incident CRC cases (number of colon cancer, 785; number of rectal cancer, 463) with existing measures for serum 25OHD were identified for the present study. Cases were not selected from Norway (blood samples only recently collected; few CRCs diagnosed after blood donation) and the Malmo center of Sweden. The distribution of cases (colon/rectum) by country was as follows: Denmark, 186/167; France, 28/8; Greece, 12/14; Germany, 93/55; Italy, 104/42; the Netherlands, 93/48; Spain, 78/41; United Kingdom, 150/64; and Sweden, 41/24. Some subjects were excluded due to incomplete genotyping data. Thus, the final number of complete case-control sets used were as follows: VDR BsmI (number of colon, 689; number of rectal, 402), VDR Fok1 (number of colon, 676; number of rectal, 401), and CASR (number of colon, 729; number of rectal, 431).

Controls were selected by incidence density sampling from all cohort members alive and free of cancer at the time of diagnosis of the cases and were matched by age (± 6 mo at recruitment), gender, study center, time of the day at blood collection, and fasting status at the time of blood collection (<3, 3-6, >6 h). Women were further matched by menopausal status (premenopausal, postmenopausal, perimenopausal/unknown), phase of menstrual cycle at blood collection (to account for potential differences in blood analyte levels by these factors), and usage of hormone replacement therapy at time of blood collection (yes/no). The latter matching criteria were of necessity to other EPIC nested case-control studies that were being conducted using the same matched case-control sets.

Vitamin D status was quantitatively determined by measuring 25OHD in 25 μL of serum (heparin plasma for Swedish samples) using a commercially available enzyme immunoassay kit (OCTEIA 25OHD kit; Immuno Diagnostic Systems) at the Laboratory for Health Protection Research, National Institute for Public Health and the Environment, the Netherlands.

The VDR (BsmI: rs1544410, 60890G>A; Fok1: rs2228570, 27823T>C) and CASR (A986S; rs1801725, G>T) polymorphisms were genotyped by Taqman methodology in 384-well plates read with the Sequence Detection Software on an ABI-Prism7900 instrument, according to the manufacturer's instructions (Applied Biosystems). Primers and probes were supplied by Applied Biosystems (Assays-by-Design). Each plate included a negative control (no DNA). Positive controls were duplicated on a separate plate. For all genotypes, the assay success rate was >97% and the internal study duplicate rate was >99%. Failed genotypes were not repeated. For all analyses, laboratory technicians were blinded to the case-control status of the samples.

For the analysis of VDR and CASR polymorphisms, Hardy-Weinberg equilibrium was tested for each polymorphism in the control subjects using a χ² test. Conditional logistic regression, stratified by the case-control set, was used to estimate the cancer risk association and 95% confidence intervals (95% CI) for each single nucleotide polymorphism (SNP; SAS statistical software, version 9, SAS Institute). In a nested case-control study where controls are selected using incidence density sampling, this procedure estimates the incidence rate ratio (RR), which, given the rarity of the disease, is approximately equal to the odds ratio (28). The two statistical models used were as follows: (a) univariate analyses based on the matching factors and (b) multivariate adjusted analyses incorporating additional adjustments for potential confounders including body mass index (kg/m²), physical activity (combined recreational and household activity; expressed as sex-specific categories of metabolic equivalents or METS), smoking duration/status/intensity (variable categories: never-smokers, ex-smokers who smoked for <10 y, ex-smokers who smoked for ≥10 y, smokers who smoke <15 cigarettes/d, smokers who smoke between 15-25 cigarettes/d, smokers who smoke ≥25 cigarettes/d, and missing), level of schooling (an indicator variable for socioeconomic status), total energy intake (in quartiles), total intake of fruits (in quartiles), total intake of vegetables (in quartiles), total intake of meats/meat products (in quartiles), and total alcohol intake (sex specific categorical cut-points: men, nonconsumers, 1-10, 11-20, 21-40, >40 grams/d; women, nonconsumers, 1-5, 6-15, 16-25, >25 grams/d). For all analyses, the reference genotype was defined as the homozygous (wild-type) allele. All main effects analysis models were run for colon and rectum combined (i.e., CRC) as well as separately. Tests for linear trend were done using a score variable with values from one to three consistent with the VDR genotype groupings. For all polymorphisms, no meaningful differences were noted between the matching factors and multivariate adjusted models so all results refer to the latter, unless otherwise specified.

Because it is postulated that the VDR enact many functions of vitamin D, particularly those affecting cell cycle kinetics, a potential interaction of VDR genotypes and serum 25OHD concentration on CRC risk was explored by including a single degree of freedom interaction term formed by the product of VDR genotype (designated as 0, 1, or 2) and the value of three categories of 25OHD concentration [cut points determined on the basis of proposed levels of vitamin D insufficiency/sufficiency (29-32): category 1, <50.0 nmol/L (<20.0 ng/mL); category 2, ≥50.0 to <75.0 nmol/L (≥20.0 to <30.0 ng/mL; category 3, ≥75.0 nmol/L (≥30.0 ng/mL)]. For these analyses, the multivariate-adjusted model was used.

Potential interaction with the level of dietary calcium intake (in tertiles) was similarly assessed for all the polymorphisms. In addition, given that body weight and gender may modify the VDR-CRC risk association (33-35), any possible interactions between these variables and all the polymorphisms were carefully assessed. In all analyses, P values for interaction were assessed by the likelihood ratio test.

At each level of the BsmI VDR genotype, differences in serum 25OHD concentration between cases and controls were assessed by t tests. To further examine the association between genotype and serum 25OHD concentration, a nonzero linear trend was tested in serum 25OHD concentration against BsmI VDR minor allele count, separately in cases and controls. Similar approaches were taken for the BsmI VDR genotype and dietary calcium.

Baseline characteristics and description of the study population are shown in Table 1. All polymorphisms were in Hardy-Weinberg equilibrium. Serum 25OHD concentration and dietary calcium levels did not vary across genotypes for any of the polymorphisms (data for VDR BsmI summarized in Table 3; data for VDR Fok1 and CASR were not shown). In the multivariate adjusted model, the BB genotype of the VDR BsmI showed a significant negative association with risk of CRC (RR, 0.76; 95% CI, 0.59-0.98; P_trend = 0.10). This association was observed for colon cancer (RR, 0.69; 95% CI, 0.45-0.95; P_trend = 0.05) but not rectal cancer (RR, 0.97; 95% CI, 0.62-1.49; P_trend = 0.81). No meaningful or statistically significant cancer risk associations were noted for either the VDR Fok1 or CASR polymorphisms (Table 2).

The association between VDR BsmI polymorphism and CRC risk did not show a statistically significant interaction with serum 25OHD concentration (P = 0.43) or level of dietary calcium intake (P = 0.72). However, given that the actions of vitamin D are mediated by the VDR and that calcium homeostasis is a main function of vitamin D, interaction analyses were done anyway as described in the statistical methods section. The negative association of the BB genotype was most apparent at a serum 25OHD concentration of ≥75.0 nmol/l (RR, 0.41; 95% CI, 0.24-0.69; Table 3). For dietary calcium, interaction analysis shows that higher intake is associated with a decreased CRC risk, particularly for the BB genotype of the VDR BsmI polymorphism (Table 3). No significant linear trends were observed in cases and controls between genotype and serum 25OHD concentration or dietary calcium levels. At each level of the VDR BsmI polymorphism, serum 25OHD concentration was significantly higher in controls than cases. There were no significant differences between cases and controls in dietary calcium intake (Table 3).

No statistically significant CRC risk interactions were observed for the other polymorphisms with either serum 25OHD concentration (VDR Fok1 P = 0.40; CASR P = 0.36) or dietary calcium (VDR Fok1 P = 0.97; CASR P = 0.22). Interaction analyses for these variables were not meaningful (data not shown). None of the polymorphisms showed any interactions with body weight or gender.

The results of this nested case-control study show an inverse CRC risk association for the BB genotype of the BsmI (rs1544410) VDR polymorphism. This association was observed in the colon and not the rectum anatomic site. Although no statistically significant interactions were identified, the inverse CRC risk associations of either higher serum 25OHD concentration or increased intake level of dietary calcium were more clearly observed with the BB than the other VDR BsmI genotypes. No CRC risk associations were observed for the VDR Fok1 (rs2228570) or CASR (rs1801725) polymorphisms.

There is epidemiologic evidence for a role of some VDR polymorphisms not only in CRC risk, but in several other cancers as well (36). Results from studies on adenomas and CRC primarily indicate either no association or a reduced risk for the B allele of the VDR BsmI (13-15, 20). Similarly, previous findings for the f allele of the VDR Fok1 are either null (15, 17-19) or conflicting (decreased risk of large adenoma, ref. 13; decreased, ref. 22; or increased risk of CRC, ref. 37). Although these polymorphisms may have functional significance, their role in CRC risk is far from clear. The VDR BsmI polymorphism analyzed here is intronic and is in strong linkage disequilibrium with the poly(A) sequence in the 3′ untranslated region whose length may determine mRNA stability and hence affect local VDR protein levels (19, 38). But it is not clear whether in fact the BsmI allele has an effect on the expression level or activity of the translated VDR protein (12). The Fok1 polymorphism is in the 5′ promoter region of the VDR, within the DNA binding domain, and its f allele results in the production of a VDR protein with reduced effectiveness as a transcriptional activator (8). It is clear that a better understanding of VDR-mediated vitamin D function is necessary (39). In fact, limitations in the current understanding of the VDR leave open the possibility that any observed associations with cancer risk may be due to the functions of another nearby site or even to a different gene (12).

In the present study, the potential interaction between the VDR polymorphisms and serum 25OHD concentration was assessed. Although the interactions were not statistically significant, the inverse CRC risk association of the BB genotype of the VDR BsmI polymorphism seemed to be more clearly observable with higher 25OHD concentration or increased intake of dietary calcium. It is possible that despite the relatively large sample size of the present study, which is sufficient for observing the main effects, it still lacks statistical power to clearly observe any possible or subtle gene-nutrient interactions. The same is true for previous studies that have also considered an interaction between VDR and circulating 25OHD levels (16). Although many of the functions of vitamin D are thought to be enacted via the VDR, vitamin D may also have some important non VDR-mediated functions (39). As well, the possibility of alternative receptors for vitamin D metabolites has been suggested (40). Thus, it may be speculated that the vitamin D–CRC risk association is somewhat VDR independent. Nevertheless, the potential interaction is biologically meaningful and should be explored in greater detail with respect to the risk of CRC and other cancers.

Other studies in North American (20) and Singapore-Chinese (37) populations have also considered a gene-nutrient interaction between VDR and dietary calcium. The results of the latter study showed that the ff genotype of VDR Fok1 is associated with a significantly higher CRC risk only with low calcium and low fat intake (37), whereas the findings in the North American population showed little interaction (20), in line with the present results.

It is possible that analysis of VDR haplotypes may reveal further information about the VDR-CRC risk association, but results to date from various populations have been inconsistent (11, 41-43) and further research is required in this area.

In addition to the VDR polymorphisms, a polymorphism in the CASR gene was also analyzed in the present study. The CASR is potentially important because it is key to extracellular calcium homeostasis, plays a role in cellular growth kinetics, influences PTH secretion, and may affect vitamin D metabolism (23, 24). The promoter region of the CASR gene contains a vitamin D response element, suggesting a potential regulation by vitamin D (44). CASR polymorphisms have been shown to be associated with the risk of colorectal adenomas (26) and increasing stage of rectal cancer (45). However, findings were overall null in a recent comprehensive study of various CASR polymorphisms and haplotypes and the risk of colon cancer (46). The present results also showed no association with cancer risk, nor any meaningful interactions with serum 25OHD or dietary calcium levels. However, the important functions of the CASR indicate its potential association with CRC risk should be studied greater detail.

In summary, this relatively large study is the first in a Western European population to assess the association of polymorphisms in the VDR and CASR genes with risk of CRC. The results show an inverse CRC risk association for the BB genotype of the VDR BsmI polymorphism, but not for the other polymorphisms assessed, none of which showed a statistically significant interaction with circulating 25OHD levels, dietary calcium intake, or other dietary and life-style factors in modifying CRC risk. Better-powered studies are necessary to more clearly explore any possible gene-diet–life-style interactions.

No potential conflicts of interest were disclosed.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.