Abstract
Currently higher morbidity and mortality rates are observed in cancer diseases, especially sex-dependent cancers. A positive role of endogenous vitamin D concentration in cancer diseases has been reported in many publications. Furthermore, there has been observed a relationship between serum vitamin D and testosterone concentrations in an elderly Caucasian population carrying the vitamin D receptor FokI gene polymorphism. The aim of this study was to investigate whether the vitamin D receptor FokI polymorphism is associated with cancerogenesis in sex-dependent cancers. The MEDLINE and ResearchGate databases were used to search for articles up to January 2017, and 96 articles concerning the FokI polymorphism were chosen. Odds ratios with 95% confidence intervals were used to assess the strength of associations between polymorphisms of vitamin D receptor and cancer risk in the described populations. The fixed-effects model and the DerSimonian–Laird random-effects model (with weights based on the inverse variance) were used to calculate summary odds ratios, and both within- and between-study variation were considered. Generally, the F variant reduces the risk of cancer by 4% (odds ratio = 0.96, p value = 0.0057). This effect is particularly evident in female sex–associated cancers (odds ratio = 0.96, 95% confidence interval: 0.93–0.99, p value = 0.0259), but it is not observed in non-sex-associated cancers. Polymorphism FokI is associated with breast and ovarian cancers.
Introduction
Recent studies indicate the role of vitamin D in such processes as autoimmune diseases, 1 cardiovascular diseases,2,3 cancerogenesis, 4–6 and depressive symptoms in the older population. 7 Currently, higher morbidity and mortality rates are observed in cancer diseases, 8 especially sex-dependent cancers, that is, breast cancer (BC) and ovarian cancer (OC) in women and prostate cancer (PCa) in men. Many authors have reported a protective role of endogenous vitamin D status in cancer diseases.9,10 The protective effect of vitamin D is explained by two cellular mechanisms. The first one is connected with the absorption of ultraviolet (UV) radiation in the skin by 7-dehydrocholesterol. During this process, 7-dehydrocholesterol is converted into cholecalciferol. 11 The second mechanism is connected with complexing of 1,25-dihydroxyvitamin D and the vitamin D receptor (VDR), which stimulates the expression of many gene-encoding enzymes responsible for cell differentiation or apoptosis. 6 Through changes in vitamin D concentration, VDR regulates the expression of more than 200 enzymes acting in different metabolic pathways such as proliferation, differentiation, apoptosis, inflammatory processes, and mutagenesis.6,10 Because vitamin D complexed with VDR regulates the synthesis of sex hormones, its status can also be correlated with BC and OC in women or PCa in men.12,13
The expression level of the VDR gene may be regulated by sequence changes, that is, single nucleotide polymorphisms (SNPs). 14 Only a few of the 5252 VDR gene SNPs (http://www.ncbi.nlm.nih.gov/snp) can influence VDR expression or activity. Currently, the most studied are BsmI (rs1544410), ApaI (rs7975232), and TaqI (rs731236), located in the 3′-untranslated region, responsible for messenger RNA (mRNA) stability, and FokI (rs10735810), located in the coding region of VDR, at the start codon, results in a polymorphic protein form shorter by three amino acids.
Polymorphic forms of the VDR gene are characterized by altered expression levels. This may result in a decrease or increase of vitamin D action in different cells. 14 Many studies have described contrary effects of polymorphic forms of the VDR gene on cancer risk, which may be explained by different numbers of examined patients, the site of cancers, population origin, or applied statistical methods. Verdoia et al. 15 demonstrated that sex significantly affects vitamin D status. Lower 25-hydroxy vitamin D (25 (OH) vitamin D) levels were observed in the female group. Furthermore, there was observed a statistically significant relationship between serum vitamin D and testosterone concentrations in elderly women with the VDR genotypes FF (rs10735810), BB (rs1544410), aa (rs7975232), and tt (rs731236) and in men with genotypes FF, BB, AA, and Tt. 16 In our study, we focused on FokI polymorphism due to another localization than ApaI, BsmI, or TaqI. Other studies have shown a correlation between free androgen index (FAI) and serum vitamin D concentration in elderly FF homozygotes, 16 and many studies have proven its effect on VDR expression in cancer diseases.
The aim of this study was to investigate whether the VDR FokI polymorphism is associated with cancerogenesis in sex-dependent cancers. Therefore, we performed a meta-analysis of studies investigating the association of the VDR FokI polymorphism gene with the risk of sex- and non-sex-associated cancers.
Material and methods
The MEDLINE and ResearchGate databases were used to search for articles up to January 2017, using the following terms: “VDR”(All Fields) AND (“polymorphism, genetic”(MeSH Terms) OR (“polymorphism”(All Fields) AND “genetic”(All Fields)) OR “genetic polymorphism”(All Fields) OR “polymorphism”(All Fields)) AND (“neoplasms”(MeSH Terms) OR “neoplasms”(All Fields) OR “cancer”(All Fields)).” Reference lists and conference reports were also reviewed. Approximately 408 publications were found, among which studies making comparisons between cancer patients and a healthy control group were studied. Accordance with this restriction, 96 articles concerning the FokI polymorphism were chosen. Logistic regression was the basic method of analysis used in the study. Odds ratios with 95% confidence intervals (CIs) were used to assess the strength of associations between polymorphisms of VDR and cancer risk in described populations. Next, publications were divided into three groups: describing male sex–associated cancer (PCa), female sex–associated cancers (OC, BC), and non-sex-associated cancers (colorectal cancer (CRC), urothelial bladder cancer (UBC), renal cell carcinoma (RCC), pancreatic cancer (PC), gastric cancer (GC), hepatocellular carcinoma (HCC), gallbladder cancer (GBC), lung cancer (LC), esophageal squamous cell carcinoma (ESCC), cutaneous melanoma (CM), meningioma (M), glioma (G), multiple myeloma (MM), head and neck squamous cell carcinoma (HNSCC), oral squamous cell carcinoma (OSCC), nasopharyngeal carcinoma (NPC), basal cell carcinoma (BCC), non-Hodgkin lymphoma (NHL), bladder cancer (BlC), osteosarcoma (O), Ewing sarcoma (ES)).
Two models were constructed for the FokI polymorphism: model I (the frequency of occurrence of allele F, nucleotide C in the population) and model II (the frequency of occurrence of allele f, nucleotide T in the population). The frequency of allele F (NF) in model I was calculated from the formula: NF = n × q, where n is the total number of genotypes in the group, and q is the probability of allele F.
Similarly, the number of genotypes in model II in both groups (cancer and control) was calculated from the formula: Nf = n × p, where Nf is the number of genotypes with allele f, n is the total number of genotypes in the group, and p is the probability of allele f. 17
The fixed-effects model and the DerSimonian–Laird random-effects model (with weights based on the inverse variance) were used to calculate summary odds ratios (ORs), and both within- and between-study variations were considered. Publication bias was analyzed using funnel plots of asymmetry, the Begg and Mazumdar test, and Egger’s test. 18 Heterogeneity between studies was evaluated using Cochran’s Q test and I2 estimates. 19 p value less than 0.05 was considered to be statistically significant when comparing trials showing heterogeneity, and random-effects analysis was selected. All statistical analyses were performed using Statistica ver. 10 software (StatSoft, USA) with the Medical Package.
Results
In total, 96 articles describing the association between VDR FokI polymorphism and 129 different cancers which met the inclusion criteria were found. In total, 54,059 cases and 68,983 controls were included in the pooled analysis (Supplementary Table S1).
Results of tests for asymmetry were non-significant—the Begg and Mazumdar test: tau = –0.0407, Z = –0.68, p = 0.4938; the Egger’s test: regression intercept = –0.2385 (95% CI = –0.7335–0.2565), p = 0.3422. Heterogeneity between studies was significant (Q = 208.39, df = 128, p < 0.005, I2 = 38.58%, 95% CI = 23.86–50.45). The significant value of heterogeneity may be due the ethnic diversity of the studied groups as well as the fact that we are dealing with a multifactorial etiology disease, where environmental conditions are also varied. Therefore, we evaluated whether I2 is appropriate. The I2 value suggests that heterogeneity was medium (I2 < 40%), and all the articles can be allowed for analysis.
We choose 14 papers out of 96 which showed a statistically significant odds ratio for carriage of allele F in sex-dependent cancers in male and female groups together 20–33 (p < 0.05). The results were inconsistent: allele F was reported to be associated with cancer risk in eight papers (BC, CRC, CM, LC, GBC) and allele f in six papers (BC, RCC, MM, HCC, BCC, PCa, CM) (Table 1). Cumulative OR was 0.96 (95% CI = 0.93–1.00, p value = 0.0249), which means that allele F reduces the risk of cancer by 4% (p = 0.0249) regardless of the sex.
Statistically significant differences in frequency of FokI polymorphism alleles in different cancers between patients and healthy controls.
OR: odds ratio of allele “F” substitution; CI: confidence interval; BC: breast cancer; CRC: colorectal cancer; RCC: renal cell carcinoma; MM: multiple myeloma; HCC: hepatocellular carcinoma; CM: cutaneous melanoma; LC: lung cancer; BCC: basal cell carcinoma; PCa: prostate cancer; GBC: gallbladder cancer.
NF = n × q, where n is the total number of genotypes in the group, and q is the probability of allele “F”; Nf = n × p, where Nf is the number of genotypes with allele “f,” n is the total number of genotypes in the group, and p is the probability of allele “f.”
p value < 0.05 was considered to be statistically significant.
Next, we selected publications about male sex–associated cancer (PCa) and female sex–associated cancers (OC, BC) separately. Cumulative OR for allele F in the female sex–associated cancer group was 0.96 (95% CI = 0.92–0.99) and was statistically significant (p value = 0.0259). Male sex–associated cancer and non-sex-associated cancer cumulative ORs were 0.97 (95% CI = 0.91–1.03) and 0.97 (95% CI = 0.90–1.02), respectively, and did not reach statistical significance (p value = 0.3059, p value = 0.1905) (Table 2). Summarizing the data on 50,234 subjects (41 populations) shows that the occurrence of allele F reduces the female sex–associated cancer (i.e. BC, OC) risk by 4%, but we did not find a significant association in male sex–associated cancer.
Cancer type risk connected with FokI genotype.
OR: odds ratio of allele “F” substitution; CI: confidence interval; PCa: prostate cancer; OC: ovarian cancer; BC: breast cancer; CRC: colorectal cancer; UBC: urothelial bladder cancer; RCC: renal cell carcinoma; PC: pancreatic cancer; GC: gastric cancer; HCC: hepatocellular carcinoma; GBC: gallbladder cancer; LC: lung cancer; ESCC: esophageal squamous cell carcinoma; CM: cutaneous melanoma; M: meningioma; G: glioma; MM: multiple myeloma; HNSCC: head and neck squamous cell carcinoma; OSCC: oral squamous cell carcinoma; NPC: nasopharyngeal carcinoma; BCC: basal cell carcinoma; NHL: non-Hodgkin lymphoma; BlC: bladder cancer; O: osteosarcoma; ES: Ewing sarcoma.
Male sex–associated cancer: PCa; female sex–associated cancers: OC and BC; and non-sex-associated cancers: CRC, UBC, RCC, PC, GC, HCC, GBC, LC, ESCC, CM, M, G, MM, HNSCC, OSCC, NPC, BCC, NHL, BlC, O, ES.
p value < 0.05 was considered to be statistically significant.
Discussion
The VDR gene with allele F is more expressed than allele f. 14 In our meta-analysis, we found that the F variant reduces risk of cancer by 4%, irrespective of the location of the cancer in 125,951 persons from 135 populations (OR = 0.96, p value = 0.0057). This effect is particularly evident in female sex–associated cancers (i.e. BC, OC) (OR = 0.96, 95% CI = 0.93–0.99, p value = 0.0259). We did not observe any statistically significant effect in non-sex-associated cancers (OR = 0.96, 95% CI = 0.90–1.02, p value = 0.1947) or male sex–associated cancer.
Our previous study showed a correlation between serum vitamin D3 and testosterone concentrations in elderly persons with allele F of the VDR FokI polymorphism gene. 16 Polymorphisms of VDR, the serum sex hormone level, free estrogen index (FEI), and FAI as well as vitamin D were evaluated in 766 persons (362 women and 404 men) selected from a population of 5695 Polish persons, aged 65–90 years, from the PolSenior survey. We found a positive correlation between testosterone concentration and serum level of vitamin D in genotype FF of the rs10735810 (FokI) polymorphism regardless of gender. This may be due to the fact that the receptor isoform (M4) encoded by this variant of the gene is shorter by three amino acids and has a greater efficiency with the transcription factor II B (TFIIB).114,115
In another study, we did not observe a statistically significant direct association between FokI polymorphism genotype and testosterone serum concentration. 116 The conclusion of our publications was that allele F associated with the higher VDR expression level may stimulate VDR in the regulatory pathway of testosterone production. In a reanalysis of nine prospective studies by the Endogenous Hormones and Breast Cancer Group (2002), it was found that levels of endogenous sex hormones are associated with BC risk in postmenopausal women. Similar data were reported by other authors, 117 who pointed out the indirect influence on BC risk through its conversion to estradiol in postmenopausal women. Another study 117 found that testosterone has a modest indirect influence on the risk of BC, through its conversion to estradiol. In turn, prostate cancer appears to be unrelated to endogenous testosterone levels. 118 Another meta-analysis20,93 obtained contrary results to ours, but different numbers of patients and ethnic groups were studied. In our meta-analysis, we did not observe an association between non-sex-associated and VDR FokI gene polymorphism cancers studied in a group of 67,892 subjects from 68 populations. In conclusion, we observed that polymorphism FokI showed stronger evidence of an association with sex-associated cancer disease in comparison to non-sex-associated cancers, especially in the female group.
Generally, the F variant allele in our meta-analysis reduces risk of cancer by 4% (OR = 0.96, p value = 0.0057). This effect is particularly evident in female sex–associated cancers (OR = 0.96, 95% CI = 0.93–0.99, p value = 0.0259), but it is not observed in non-sex-associated cancers.
Conclusion
Polymorphism FokI is associated with BC and OC.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The publication was supported by Wroclaw Center of Biotechnology, the Leading National Research Centre (KNOW) program for years 2014–2018.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
