Lack of association between MTHFR Ala222Val and Glu429Ala polymorphisms and bladder cancer risk: A meta-analysis of case-control studies

Bladder cancer is a commom malignancy in the urinary tract that is influenced by genetic and environmental factors. The role of functional polymorphisms in the methylenetetrahydrofolate reductase (MTHFR) gene with bladder cancer risk remains to be determined. This meta-analysis was performed to derive a more precise estimation of MTHFR Ala222Val and Glu429Ala polymorphisms and bladder cancer risk. Data were collected with the last report up to September 2013. A total of 3,463 cases and 3,927 controls for Ala222Val, and 3,177 cases and 3,502 controls for Glu429Ala were analyzed. The pooled odds ratios (ORs) and 95% confidence interval (CI) were estimated for the association with bladder cancer risk. Overall, no significant associations of Ala222Val and Glu429Ala polymorphisms with bladder cancer risk were found (for Ala222Val: Val/Val vs. Ala/Ala: OR, 1.02; 95% CI: 0.80–1.29; Val/Ala vs. Ala/Ala: OR, 1.02; 95% CI: 0.92–1.12; dominant model: OR, 1.01; 95% CI: 0.87–1.17; recessive model: OR, 1.00; 95% CI: 0.87–1.15; and for Glu429Ala: Ala/Ala vs. Glu/Glu: OR, 1.11; 95% CI: 0.78–1.58; Ala/Glu vs. Glu/Glu: OR, 1.16; 95% CI: 0.95–1.40; dominant model: OR, 1.15; 95% CI: 0.94–1.41; recessive model: OR, 0.96; 95% CI: 0.79–1.15). In stratified analyses by ethnicity, significant associations were observed for Glu429Ala polymorphism in individuals of Middle Eastern descent (Ala/Glu vs. Glu/Glu: OR, 2.11; 95% CI: 1.26–3.53; dominant model: OR, 2.16; 95% CI: 1.16–4.01; recessive model: OR, 1.82; 95% CI: 1.11–3.01). This meta-analysis demonstrated that overall there was no association of MTHFR Ala222Val and Glu429Ala polymorphisms with bladder cancer risk. However, in the stratified analysis by ethnicity the MTHFR Glu429Ala polymorphism was significantly associated with increased bladder cancer risk in individuals of Middle Eastern descent.


Introduction
Bladder cancer is the most common malignant tumor of the urinary tract, with an estimated 386,300 new cases and 150,200 mortalities from bladder cancer annually (1), ranking seventh in men and seventeenth in women worldwide (2). The etiology of bladder cancer involves the interaction between genetic and environmental factors. Common risk factors include cigarette smoking, occupational exposure to aromatic amines and polycyclic aromatic hydrocarbons, inflammation of the urinary tract and consumption of certain drugs (3). Current evidence supports that molecular alterations and DNA polymorphisms potentially alter individual susceptibility to bladder cancer (4)(5).
Of all the genetic susceptibility factors of bladder cancer, folate metabolic pathway has received increasing attention (5)(6)(7). Since it is the one-carbon groups donor in DNA methylation and synthesis, folate deficiency may induce DNA hypomethylation and potentially induce a proto-oncogene expression leading to cancer.
Methylenetetrahydrofolate reductase (MTHFR) is the key enzyme involved in folate metabolism, which acts as a critical metabolic node in the regulation of methylation reactions. It catalyzes the conversion of 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate. The 5,10-methylenetetrahydrofolate is the methyl donor for de novo thymidine synthesis and the 5-methyltetrahydrofolate is used as a cosubstrate to convert homocysteine into methionine, which is the immediate precursor of S-adenosylmethionine and the primary methyl donor of DNA methylation. A less active form of MTHFR may result in hypomethylation, which is a candidate mechanism for the development of cancer (8).
The functional polymorphisms in the MTHFR gene, Ala222Val (C677T) and Glu429Ala (A1298C), have received increasing attention. Ala222Val homozygotes have been associated with reduced enzyme activity of ~30% of the control value (9), while the Glu429Ala homozygotes exhibit ~60% of the control activity. Heterozygotes for the Ala222Val and Glu429Ala mutations had ~50-60% of control activity (10). Since the two variants result in reduced activity of MTHFR, their associations with the susceptibility of a variety of cancers has been evaluated (11)(12)(13)(14).
MTHFR Ala222Val and Glu429Ala polymorphisms have also been evaluated in relation to bladder cancer risk. However, the results of previous studies have yielded conflicting results. Given the amount of accumulated data, this meta-analysis was performed to derive a more precise estimation of the association of MTHFR Ala222Val and Glu429Ala and bladder cancer risk.

Materials and methods
Publication search. An exhaustive search of the literature was performed using the electronic databases: PubMed, EBSCO-Medline, Elsevier ScienceDirect and BIOSIS Previews for relevant articles published (up to September 2013). The articles were identified by using the search terms 'methylenetetrahydrofolate reductase', 'MTHFR', 'NADPH2', 'urinary bladder' and 'bladder'. The search was limited to human studies with no language restrictions being applied. Additional studies were obtained through the references cited in retrieved articles on the association between the MTHFR Ala222Val and Glu429Ala polymorphisms and bladder cancer. The searching of the electronic databases and reviewing of the references in retrieved articles were independently achieved by two investigators.
Inclusion and exclusion criteria. Inclusion criteria for the studies were: case-control studies on the association between MTHFR Ala222Val and Glu429Ala polymorphisms and bladder cancer risk, and the data of each study was required to be sufficient for statistical analysis of odds ratio (OR) and 95% confidence interval (CI). Studies that had duplicated data and in which genotypes could not be ascertained were excluded.
Data abstraction. Information was extracted from all the eligible publications independently by two investigators as per the inclusion and exclusion criteria. For disagreements, consensus was reached by discussion of the two investigators. The data abstracted from each study were as follows: first author's name, publication date, country, ethnicity, source of controls, number of cases and controls, and number of cases and controls for MTHFR Ala222Val and Glu429Ala polymorphisms. Individuals of different descents were classified as Chinese, European, American and Middle Eastern.
Statistical analysis. ORs with 95% CIs were applied to evaluate the strength of association between MTHFR Ala222Val and Glu429Ala polymorphisms and bladder cancer risk. Ala/Ala + Ala/Glu vs. Glu/Glu for Glu429Ala) and recessive model (Val/Val vs. Val/Ala + Ala/Ala for Ala222Val; Ala/Ala vs. Ala/Glu + Glu/Glu for Glu429Ala), respectively. The Chi-square test-based Q-statistic (Q test) was applied to assess heterogeneity among the studies. The fixed-effects model was used to calculate the pooled ORs if no heterogeneity was detected (Ph≥0.05 by Q test) (15). Otherwise, the random-effects model was applied (Ph<0.05 by Q test) (16).
Subgroup analyses were performed by ethnicity and source of controls. The sensitivity analyses were conducted by excluding each study at a time to determine its effect on the overall estimation, since all the studies indicated that they conformed to the Hardy-Weinberg equilibrium (HWE). The publication bias was estimated by the funnel plot, in which the standard error of log (OR) was plotted against its log (OR) for each study. Egger's linear regression test was applied to assess the asymmetry of the funnel plot, with P<0.05 indicating an asymmetric plot and a possible publication bias (17). In this meta-analysis, the Stata version 11.0 (StataCorp, College Station, TX, USA) was applied for all the statistical tests.

Results
Flow of included studies. A total of 420 articles potentially relevant to the searching terms were screened, including PubMed, 25; EBSCO-Medline, 24; Elsevier ScienceDirect, 338; and BIOSIS Previews, 33. Based on the inclusion criteria, a total of 13 studies (5-7,18-27) with full-text articles on polymorphisms of MTHFR Ala222Val and Glu429Ala and bladder cancer risk were identified as eligible. Of the 13 studies, two studies (22,24) were excluded due to duplicated data with the study by Rouissi et al (6).
Quantitative data synthesis. Overall, no significant associations of Ala222Val and Glu429Ala polymorphisms with bladder cancer risk were identified by this meta-analysis (Table III) Fig. 1. In the subgroup analysis by the source of controls, no significant associations were found in any of the genetic models for the two polymorphisms (Table III).
Heterogeneity and sensitivity analysis. In the heterogeneity analysis, the Val/Val vs. Ala/Ala model and dominant genetic model for the Ala222Val polymorphism, as well as the Ala/Ala vs. Glu/Glu, Ala/Glu vs. Glu/Glu and dominant genetic models for the Glu429Ala polymorphism were found to be significant (Ph<0.05 by Q test, Table III). The Ph value of the subgroup analysis showed that the heterogeneity was effectively decreased in some of the comparisons and the major source of heterogeneity may stem from the hospital-based controls and ethnicity, such as the Chinese and American subgroups. In the sensitivity analyses, with each study been excluded one at a time during the analysis, the overall results were not altered and no different conclusions were obtained, although the heterogeneity of the analysis was obviously decreased during the exclusion. Table I. Characteristics of the case-control studies considered in the meta-analysis.

Demographic characteristics Genotyping
Control

Discussion
The MTHFR Ala222Val and Glu429Ala polymorphisms have been found to be a risk factor for a variety of cancers including colon cancer (11), acute lymphoblastic leukemia (12), gastric cancer (13) and head and neck squamous cell carcinoma (14). As for bladder cancer, several epidemiological studies (5-7, 18-27) have been investigated for their association with cancer risk. However, the results are not conclusive due to small sample-sized association studies that lack statistical power. In the current meta-analysis, a more precise estimation of MTHFR Ala222Val and Glu429Ala and bladder cancer was derived by including a pooled total of 3,463 cases and 3,927 controls for Ala222Val in 11 studies and 3,177 cases and 3,502 controls for Glu429Ala in 9 studies. The data showed that the Ala222Val and Glu429Ala polymorphisms were not significantly associated with bladder cancer susceptibility in the entire population. Furthermore, the role of Ala222Val and Glu429Ala polymorphism was evaluated in different subgroups (Chinese, European, American and individuals of Middle Eastern descent), and the results indicated that the Ala222Val polymorphism was not significantly associated with bladder cancer risk in any of the genetic models. Notably, Glu429Ala polymorphism was significantly associated with bladder cancer risk in individuals of Middle Eastern descent. Individuals of Middle Eastern descent who had the Ala/Ala allele were ~82% more likely to have bladder cancer than those who had Ala/Glu or Glu/Glu genotype. Our Table II. Distribution of methylenetetrahydrofolate reductase (MTHFR) gene Ala222Val and Glu429Ala genotypes for bladder cancer patients and controls.
As a potential problem when interpreting the results of all meta-analyses (29), heterogeneity was reduced in this study. Significant between-study heterogeneity existed in the Val/Val vs. Ala/Ala and dominant genetic models of the Ala222Val polymorphism, as well as Ala/Ala vs. Glu/Glu, Ala/Glu vs. Glu/Glu and the dominant genetic model for the Glu429Ala polymorphism when all studies were pooled. Following the subgroup analyses by source of control, the heterogeneity among the Val/Val vs. Ala/Ala and dominant genetic models of the Ala222Val polymorphism arose from the hospital-based subgroup. By contrast, heterogeneity of the Ala/Ala vs. Glu/Glu, Ala/Glu vs. Glu/Glu and dominant genetic models of the Glu429Ala polymorphism arose from hospital-based subgroups, as well as from ethnicity, which was effectively decreased in the Chinese, European and American subgroup analyses. There are two reasons for the heterogeneity: i) hospital-based controls did not represent the entire population and ii) differences of genetic backgrounds and environmental factors potentially exist among different ethnical subgroups.
The associations between the Ala222Val and Glu429Ala polymorphisms and bladder cancer risk have been previously studies. For the variant genotype of Ala222Val polymorphism, Lin et al (5), Cai et al (23) and Wang et al (25) reported that it was associated with a higher risk of bladder cancer. Safarinejad et al (27) reported that the polymorphism was Table III. Results of meta-analysis for methylenetetrahydrofolate reductase (MTHFR) gene Ala222Val and Glu429Ala polymorphism and bladder cancer risk.     not associated with bladder cancer but was associated with increased risk of muscle-invasive bladder cancer. By contrast, Moore et al (19) reported conflicting results, suggesting a lower risk of bladder cancer was observed in individuals carrying either the Ala/Val or Val/Val genotype compared to those carrying the Ala/Ala genotype. The remaining 6 studies reported that there were no statistically significant associations between the Ala222Val polymorphism and bladder cancer risk (6)(7)18,(20)(21)26). For the variant genotypes of Glu429Ala polymorphism, Rouissi et al (6) reported that it was associated with a higher risk of bladder cancer. However, Safarinejad et al (27) reported that it was not associated with bladder cancer but associated with increased risk of muscle-invasive bladder cancer. The remaining 7 studies reported that there were no statistically significant associations between the Glu429Ala polymorphism and bladder cancer risk (5,7,(19)(20)(21)23,25). Although Wang et al (25) previously published a case-control study with meta-analysis, in which the data included were up to 2007 and 7 case-control studies were analyzed with conclusions suggesting no significant associations between Ala222Val and Glu429Ala polymorphisms and bladder cancer risk, which was in accordance with results of this meta-analysis. However, in the demographic characteristics of that study (25), all the cases and controls of different studies, including individuals from the USA, Argentina and Tunis (mistaken for Turkey in that study) were considered Caucasian, rendering the subgroup analyses by ethnicity impossible, thereby discounting the statistical power of analysis for the estimation of genetic effects. In the present study, the Glu429Ala polymorphism was found to be significantly associated with increased bladder cancer risk in individuals of Middle Eastern descent.
Given the data was abstracted from these publications, there are some limitations that should be addressed with regard to this meta-analysis. Firstly, the ethnicity of the cases and controls were not uniformly defined. The Glu429Ala polymorphism for the Middle Eastern individuals, which was found to be significant, had only 2 publications with 343 cases and 507 controls in total. Secondly, the controls may not truly represent the populations, since most of the controls included in this study were hospital-based controls and the Ala222Val and Glu429Ala polymorphism is known to be potentially associated with other diseases. Thirdly, some of the environmental risk factors of bladder cancer, such as cigarette smoking habits and occupational exposure were not uniformly provided by the original studies. Furthermore, some inevitable publication bias might exist in the publications. Therefore, the results of this study should be interpreted with caution and additional studies with a large amount of data are needed for evaluation.
In conclusion, the results of the present meta-analysis suggest that heterozygous and homozygous variant carriers of the MTHFR Glu429Ala polymorphism are significantly associated with increased bladder cancer risk in individuals of Middle Eastern descent, with no evidence of an association between this polymorphism and bladder cancer risk in individuals of Chinese, European and American descent. Since the Ala/Ala genotype is relatively infrequent in the population, more well-designed studies with larger sample sizes are required to confirm the findings of the present study.