Association of Vitamin D Receptor Cdx-2 Polymorphism With Cancer Risk: A Meta-Analysis.

Vitamin D receptor (VDR) Cdx-2 polymorphism (rs11568820) has been indicated to be associated to cancer susceptibility. However, published studies reported mixed results. This meta-analysis was conducted to get a more accurate estimation of the association between Cdx-2 polymorphism and cancer risk.We identified 25 independent studies with a total of 34,018 subjects published prior to March 2015. Summary odds ratios (ORs) and 95% confidence intervals (CIs) were used to evaluate the susceptibility to cancer. Separate analyses were conducted on features of the population such as ethnicity, source of controls, and cancer types.Meta-analysis results showed that Cdx-2 polymorphism significantly increased cancer risk in the homozygous model in overall analysis. According to the further stratified analysis, significant association was found between Cdx-2 variant and cancer risk in American-Africans in the homozygous, recessive, and dominant comparison models. However, no significant associations were found in Caucasians and Asians. When stratified by different cancer types, significant association was observed between Cdx-2 variant and an increased risk of colorectal cancer in the homozygous, recessive, and dominant models. In addition, ovarian cancer susceptibility increased based on the homozygous and dominant comparison models.Our study indicated that VDR Cdx-2 polymorphism was associated with an increased cancer risk, particularly in American-Africans, colorectal, and ovarian cancers. However, other factors may impact on the association. Further multicenter studies are needed to confirm the effects of Cdx-2 polymorphism on cancer susceptibility.

INTRODUCTION

Vitamin D is an essential fat-soluble vitamin which not only maintains the stability of the extracellular calcium and phosphorous concentration by regulating bone turnover and absorption in the gut but also regulates the growth, differentiation, and apoptosis of many normal or tumor tissue cells.[1] Vitamin D receptor (VDR) binds to 1,25-hydroxy vitamin D, the active form of vitamin D, and mediates its biological activity. VDR is active in over 30 different human tissues as an intracellular nuclear receptor and regulates gene expression as a ligand-activated transcription factor.[2] Single nucleotide polymorphisms exist in human genome widely. The polymorphisms of the VDR gene have been reported to exert functional effects on the VDR expression at the transcriptional level and on the receptor affinity.[3] The Cdx-2 (rs11568820) polymorphism of VDR is located in the promoter region of the VDR gene, which carries a G to A sequence change and affects the function of the transcription factor CDX.[4]

Cancer is currently one of the global public health problems, which threatens to human health seriously. Biological and epidemiological studies suggest that carcinogenesis is a multivariate and complicated process because of the interactions between genetic and environmental factors.[5] Previous studies suggested that lower mean serum vitamin D levels or vitamin D deficiency is common in oncology patients and correlates with advanced stage disease.[6,7] Other studies indicated that VDR plays crucial roles in cancers such as regulation of the immune function and modulation of cell proliferation and differentiation.[8,9] Associations of VDR gene variants with different types of cancer have been widely researched. Kupfer et al[10] found a significant association between the VDR polymorphisms and vitamin D intake in colorectal cancer. Ditsch et al[11] proved that the potential anticancer function of vitamin D might be mediated by VDR expression and that VDR can influence cancer predisposition through binding to vitamin D.

Cdx-2 is one of the common polymorphisms within the VDR gene promoter region that may impact on VDR expression and vitamin D intake. It has been hypothesized that the Cdx-2 polymorphism exerts function in carcinogenesis and the association has been investigated between the Cdx-2 variants and several different types of cancer by numerous studies in the past years. However, previous studies results are inconsistent.[12-33] In addition, an earlier meta-analysis found that the Cdx-2 AA homozygote carriers significant elevated cancer risk.[34] But this study only included articles that were performed in Caucasians and African Americans, and was lack of data on Asian populations. In recent years, there were a lot of new literatures published. Therefore, we conduct this meta-analysis on all eligible case–control studies to comprehensively estimate the relationship between the Cdx-2 polymorphism and cancer risk.

MATERIALS AND METHODS

Publication Search and Inclusion Criteria

We searched the Web of knowledge, PubMed, and Chinese National Knowledge Infrastructure databases for studies published prior to March 2015 (last search: March 31, 2015). The keywords searching were conducted with and without MeSH terms for “vitamin D receptor/VDR,” “polymorphism,” and “cancer.” The languages were limited to English and the subjects were human. Studies included in our meta-analysis must meet the criteria as follows: (1) evaluation of the association of VDR Cdx-2 polymorphism and cancer risk; (2) case–control design; (3) available information on genotype frequency. The following were the exclusion criteria: (1) repeat studies, review, and abstracts; (2) study design was based on the family; (3) the genotype distribution of the control population was not consistent with Hardy–Weinberg equilibrium (HWE).

Data Extraction

Initially, 2 investigators (Z-MD and Y-LF) independently checked all potentially relevant studies and disagreements were resolved through discussion with a third researcher (Z-JD). We extracted the following items for each article: first author, years of publication, original country, subjects’ ethnicity, cancer types, source of control, genotyping method, total number of cases and controls, and number of different genotypes in cases and controls. All data came from published articles. All cancers were confirmed by histology or pathology. The noncancer controls had no present evidence of any malignant disease. Controls were identified through random selections and the source of controls was either population based or hospital based. Controls and cancer cases used the same gene detection method and was matched in age and sex and so on. Therefore, all the case and control groups were well controlled. The Newcastle–Ottawa Scale (NOS) was used to assess the quality of included studies.[35]

Statistical Analysis

We used odds ratios (OR) and 95% confidence intervals (CIs) to evaluate cancer risk associated with the VDR Cdx-2 polymorphism. Statistical heterogeneity between studies was evaluated with the I2 test, higher I2 values means higher levels of heterogeneity (I2 = 75% to 100%: extreme heterogeneity; I2 = 50% to 75%: large heterogeneity; I2 = 25% to 50%: moderate heterogeneity; I2 < 25%: no heterogeneity). In heterogeneity evaluation, when the P value ≥ 0.10, the fixed-effects model would be used; if P < 0.10, a random-effects model was used. To estimate the cancer site-specific and ethnicity-specific effects, subgroup analyses were performed by cancer types, source of controls, and ethnicity. Sensitivity analysis was performed to assess the stability of the final results. In order to assess the influence of each study to the pooled OR, risk assessment was tested by sequentially omitting 1 individual study at a time. Sensitivity analysis determines whether the individual data in fact have a major effect on the results of the review. The Egger's test and Begg's test were adopted to assess publication bias. The meta-analysis assessed the following genetic models: dominant model (AA + GA vs GG), recessive model (AA vs GG + GA), homozygote comparison (AA vs GG), heterozygote comparison (GA vs GG), and allele comparison (A vs G). All analysis was performed by the Stata version 12.0. All the P values were 2-sided.

RESULTS

Study Characteristics

There were 584 articles obtained by keyword search and manual search. The flow chart for studies selection process was shown in Figure 1. Overall, our study included 27 independent studies that contained detailed genotype distribution data, among which 2 different studies were excluded from our meta-analysis because the genotype distributions of the control group departed from HWE.[16,21] Finally, a total of 25 studies from 22 published articles, involving 16,269 cases and 17,749 cancer-free controls were included in this meta-analysis. Eligible studies presented data for several different cancer types including colorectal, prostate, skin, breast, ovary, brain, and esophageal cancer. Among these studies, 15 studies were based on Caucasian, 3 on Asian, and 2 on African-American and 5 on mixed ethnicities. The NOS score of all articles are not <6, so that each included literature was a high-quality study. The characteristics of the eligible studies are presented in Table 1. The genotype distributions are shown in Table 2, and the frequency of the minor allele distributed widely across the 25 eligible studies, ranging from 0.17 to 0.84. The average frequency of the minor allele was 0.28.

FIGURE 1

Preferred reporting items for systematic reviews and meta-analyses flow diagram of the literature review process for Cdx-2 polymorphism and cancer.

TABLE 1

Characteristics of the Studies Included in the Meta-Analysis

TABLE 2

Cdx-2 Polymorphism Genotype Distribution and Allele Frequency in Cases and Controls

Main Results

Overall, our result suggested that the VDR Cdx-2 polymorphism was significant associated with increased cancer risk in the homozygote comparison (AA vs GG: OR = 1.23, 95% CI = 1.01–1.48, P = 0.03, Figure 2) and the allele comparison (A vs G: OR = 1.07, 95% CI = 1.01–1.13, P = 0.01). In the subgroup analysis by ethnicity, no significant correlation was observed between the Cdx-2 variation and cancer risk in Caucasians and Asians. However, significant association was found in American-Africans in 4 comparison models (AA vs GG: OR = 1.84, 95% CI = 1.19–2.85, P = 0.006; AA vs GG + GA: OR = 1.31, 95% CI = 1.07–1.61, P = 0.01; AA + AG vs GG: OR = 1.73, 95% CI = 1.12–2.65, P = 0.01; A vs G: OR = 1.32, 95% CI = 1.11–1.57, P = 0.002). When stratifying by source of controls, Cdx-2 polymorphism was detected to be significantly associated with an increased cancer risk in the following genetic models (AA vs GG: OR = 1.39, 95% CI = 1.12–1.72, P = 0.003; AA + AG vs GG: OR = 1.32, 95% CI = 1.09–1.60, P = 0.004; AA + GA vs GG: OR = 1.08, 95% CI = 1.03–1.14, P = 0.002; A vs G: OR = 1.10, 95% CI = 1.04–1.16, P = 0.001) in population-based case–control studies, whereas no statistical significance was found in hospital-based case–control studies. All comparisons are shown in Table 3.

FIGURE 2

Forest plots of Cdx-2 polymorphism and cancer risk in overall population (AA vs GG).

TABLE 3

Meta-Analysis Results

In the cancer-specific analyses, there were 6 studies with 4327 cases and 4744 controls for colorectal cancer. The results showed significant correlation between Cdx-2 polymorphism and an increased risk of colorectal cancer in different comparison models (AA vs GG: OR = 1.30, 95% CI = 1.08–1.57, P = 0.006; AA vs GG + GA: OR = 1.27, 95% CI = 1.05–1.52, P = 0.01; AA + GA vs GG: OR = 1.12, 95% CI = 1.02–1.21, P = 0.01; A vs G: OR = 1.12, 95% CI = 1.04–1.20 P = 0.002). Furthermore, we identified 3 studies with 1475 cases and 2055 controls for ovarian cancer. The result showed significant association between the Cdx-2 polymorphism and ovarian cancer susceptibility in the dominant model (AA + GA vs GG: OR = 1.19, 95% CI = 1.04–1.37, P = 0.01) and the heterozygote comparison (GA vs GG: OR = 1.21, 95% CI = 1.05–1.41, P = 0.01) and the allele comparison (A vs G: OR = 1.13, 95% CI = 1.01–1.28, P = 0.04), but not in other genetic models.

There were 5 studies with 3080 cases and 2630 controls for prostate cancer, 5 studies with 3938 cases and 5100 controls for breast cancer and 3 studies with 2108 cases and 1815 controls for skin cancer (cutaneous melanoma, basal cell carcinoma, and squamous cell carcinoma), respectively. However, no statistical significance was found between the Cdx-2 polymorphism and prostate, breast or skin cancer susceptibility in any genetic model.

Tests of Heterogeneity

As shown in Table 4, statistically significant heterogeneity was observed between trials of the following analyses using Q statistic. When the P value of the heterogeneity test was more than 0.1 (P ≥ 0.1), a fixed-effects model was performed. Otherwise, the random-effects model was used.

TABLE 4

Heterogeneity-Analysis Results

Publication Bias

Begg's funnel plot and Egger's test were used to evaluate the publication bias. The Begg's funnel plots’ shape seemed symmetrical (Figure 3). Therefore, there was no significant evidence for publication bias in our meta-analysis (P = 0.45).

FIGURE 3

Begg's funnel plot assessing evidence of publication bias from the eligible studies.

Sensitivity Analysis

Sensitivity analysis was performed by sequentially omitting 1 individual study at a time, in order to reflect the influence of each study on the overall meta-analysis. As shown in Figure 4, sensitivity tests suggested that no single study greatly influenced the estimates of overall risk for the VDR Cdx-2 polymorphism, thus the results of our meta-analysis were stable.

FIGURE 4

Sensitivity analysis of association between Cdx-2 polymorphism and cancer.

DISCUSSION

According to the World Health Organization, an estimated 14.1 million new cancer cases and 8.2 million cancer-related deaths occurred in 2012.[36] Malignant tumors therefore become the leading cause of death worldwide. Biological as well as epidemiologic data suggest that vitamin D level could modulate the risk of some cancer and play a role in cancer prevention.[9] VDR can influence cancer predisposition through binding to vitamin D. Polymorphisms of the VDR gene promoter region have been associated with several forms of cancer.[37,38] Although a large number of previous studies have been published to investigate the association between Cdx-2 polymorphism and cancer risk, these studies always reported mixed results. To further precisely evaluate the association, the current meta-analysis is performed on all eligible case–control studies.

In the present meta-analysis, we have included a total of 16,269 cases and 17,749 controls from 25 eligibility studies and found that Cdx-2 polymorphism significant associated with overall cancer risk. The results showed significant association for the Cdx-2 variant and cancer risk under the homozygote and allele models, respectively. The data indicated that the Cdx-2 A allele may exert substantial biological impact on the development of cancer. Dysfunction of vitamin D metabolism pathways may be involved in the carcinogenesis.

Because of different gene–environment interplays may exist in different ethnicity backgrounds, we conducted an ethnicity-based subgroup analysis and the results demonstrated that Cdx-2 was significant association with an increased risk of cancer in the following comparison models (AA vs GG: OR = 1.84, 95% CI = 1.19–2.85; AA vs GG + GA: OR = 1.31, 95% CI = 1.07–1.61; AA + GA vs GG: OR = 1.73, 95% CI = 1.12–2.65) in African-Americans. However, no significant association was observed between the Cdx-2 variant and Caucasians or Asians. Fang et al[39] revealed that the Cdx-2 A allele occurs more commonly in African (74%) populations than among the Caucasians (19%). Therefore, we suggested that these data should be explained with caution considering the heterogeneity of the ethnicity subgroup. Dietary habits, sun exposure and genetic backgrounds can influence the correlation between polymorphisms and cancer susceptibility.

In the cancer-specific analyses, we were able to provide a comprehensive evaluation of the role of the Cdx-2 polymorphism in several different cancers susceptibility. Our meta-analysis included 7 new independent studies, based on prostate cancer,[30] colorectal cancer,[23] glima,[26] esophageal cancer,[31] hepatocellular carcinoma,[32] and breast cancer,[33] which were not included in a previous meta-analysis.[34] Furthermore, 2 different studies for ovarian carcinoma and prostate cancer were excluded in our study because the HWE was insufficient.[16,21] In another meta-analysis that discussed the association between Cdx2 polymorphism and breast cancer susceptibility, significance was only detected on Africans.[40] Our study included more eligible researches performed in Asians. The results also showed no significant association between Cdx-2 polymorphism and breast cancer susceptibility in overall populations. In the subgroup analysis by ethnicity, significant correlation was observed between Cdx-2 polymorphism and cancer risk in African-Americans but not in Caucasians or Asians. More notable was that our study included more independent studies involved other different cancer types,[30-33] and the results provided stronger evidence that Cdx-2 polymorphism might associate with cancer risk in Africans.

Our meta-analyses showed that the Cdx-2 AA genotype significantly increased colorectal cancer risk by 30% and 27% versus GG and (GG + GA), respectively. In addition, this current meta-analysis showed that GG heterozygote genotype had a 21% and 19% decreased risk of ovary cancer than GA homozygote and (AA + GA), respectively. Our results suggested that Cdx-2 GG genotype may contribute to a protective effect in ovary cancer and it need to be confirmed by studies with larger sample sizes. However, no significant associations were found among studies of skin, breast, and prostate cancer in any genetic models. According our meta-analysis results, the inconsistent results might be due to different functional mechanisms of vitamin D in different cancer tissues. The results also suggested that the influence of the VDR genetic polymorphism may be ambiguous due to the presence of other factors, such as sun exposure, vitamin D intake, and involvement, in different cancers.

There were several limitations that should be noted in this meta-analysis. First, our study was lack of original information of Cdx-2 gene transcription and expression. The Cdx-2 polymorphism located in the promoter region of VDR gene might affect the transcription and expression of VDR, which might further influence the absorption of vitamin D.[4] Vitamin D exerted functions in the immune, neural, and endocrine systems and were involved in the regulation of tumor growth.[41] Second, our meta-analysis was based on unadjusted estimates so that we could not assess the risk of cancer according to stratification of age, diet, calcium and vitamin D intake, UV exposure, and other risk factors of cancer. The lack of such data for the meta-analysis may cause confounding bias. Third, there were only 2 studies on Africans and 3 studies on Asians included in this meta-analysis, thus the conclusion should be interpreted with caution at overall population. Fourth, for some cancers there was only 1 study, which may lead to heterogeneity in quantitative analysis. In addition, the potential influence on genotype-cancer associations by environmental factors is worthy of consideration.

CONCLUSIONS

This meta-analysis provides an updated and comprehensive meta-analysis about the role of the VDR Cdx-2 polymorphism in cancer susceptibility. Our results showed that the Cdx-2 polymorphism is associated with an increased risk of some cancers. Further studies were needed to confirm the prediagnostic effect of Cdx-2 gene polymorphism in carcinogenesis.