Meta-analysis of Associations between Interleukin-17 Gene Polymorphisms and Risk of Gastric Cancer

Gastric cancer is well-documented as one of the most common cancers linked to a high incidence of fatality worldwide (Hartgrink et al., 2009; Siegel et al., 2012). Although genetic and environmental factors in addition to Helicobacter pylori infections have been identified to play important roles in the development of gastric cancer (Houghton and Wang, 2005; Merchant, 2005; Dong et al., 2008), the precise etiology of the disease remains unclear. Notably, more and more studies have indicated that genetic variants of interleukin (IL) family genes may be involved in the development of gastric cancer (El-Omar et al., 2000; Xue et al., 2012; Yang et al., 2013; Pan et al., 2014; Zhu et al., 2014). Interleukin-17 (IL-17) is a cytokine secreted exclusively by activated T-cells that bridges the adaptive and innate immune systems (Moseley et al., 2003; Shibata et al., 2009). IL-17A and IL-17F are important members of the IL-17 cytokine family preferentially produced by helper T 17 (Th17) cells, which are responsible for the pathogenic activity of the lineage of CD4+ effector cells and multiple proinflammatory mediators (Rutitzky et al., 2005; Ishigame et al., 2009). A previous study has reported that IL-17A and IL-17F single nucleotide polymorphisms (SNPs) are associated with the risk of gastric cancer (Wu


Introduction
Gastric cancer is well-documented as one of the most common cancers linked to a high incidence of fatality worldwide (Hartgrink et al., 2009;Siegel et al., 2012).Although genetic and environmental factors in addition to Helicobacter pylori infections have been identified to play important roles in the development of gastric cancer (Houghton and Wang, 2005;Merchant, 2005;Dong et al., 2008), the precise etiology of the disease remains unclear.Notably, more and more studies have indicated that genetic variants of interleukin (IL) family genes may be involved in the development of gastric cancer (El-Omar et al., 2000;Xue et al., 2012;Yang et al., 2013;Pan et al., 2014;Zhu et al., 2014).
Interleukin-17 (IL-17) is a cytokine secreted exclusively by activated T-cells that bridges the adaptive and innate immune systems (Moseley et al., 2003;Shibata et al., 2009).IL-17A and IL-17F are important members of the IL-17 cytokine family preferentially produced by helper T 17 (Th17) cells, which are responsible for the pathogenic activity of the lineage of CD4+ effector cells and multiple proinflammatory mediators (Rutitzky et al., 2005;Ishigame et al., 2009).A previous study has reported that IL-17A and IL-17F single nucleotide polymorphisms (SNPs) are associated with the risk of gastric cancer (Wu

Meta-analysis of Associations between Interleukin-17 Gene Polymorphisms and Risk of Gastric Cancer
Hui Yu 1 *, Si Sun 1 , Fang Liu 2 , Qing-Hua Xu 2 et al., 2010).However, subsequent replication studies for the association between IL-17A and IL-17F variants and gastric cancer risk are controversial (Shibata et al., 2009;Arisawa et al., 2012;Rafiei et al., 2013;Zhang et al., 2014;Zhu et al., 2014).The discrepancy may be attributed to the relatively small sample size of previous studies and the genetic heterogeneity of polymorphisms in gastric cancer among different populations.
To further reconcile these conflicting findings and to obtain a more definitive conclusion using multiple genetic statistical models, a meta-analysis using three SNPs within the IL-17A and IL-17F genes was conducted to collectively analyze existing comparative studies.

Literature search
The studies included in the meta-analysis were selected by searching the PubMed, CNKI (http://www.cnki.net), and Wan Fang (http://www.wanfangdata.com.cn)databases through Feb 2014 using the following keywords and subject terms: "Interleukin-17 gene or IL-17" AND "gastric cancer or gastric carcinoma".All references in these studies were examined to identify additional research not indexed by the databases.We selected only published articles written in English or in Chinese.

Inclusion criteria
Studies included in the current meta-analysis had to meet the following criteria: (1) a case-control study design; (2) a focus on the association of IL-17 SNPs with gastric cancer risk and publication in a peer-reviewed journal; (3) sufficient data to calculate the odds ratio (OR), confidence interval (CI) and p value; (4) specified genotyping methods (including through provision of an appropriate reference); and (5) genotype distribution of all control groups in Hardy-Weinberg equilibrium (HWE).

Data extraction
Data were extracted from eligible publications independently by two of the authors (Hui Yu and Si Sun.) with any disagreement resolved by discussion between them.The following data were collected from each study: first author's name, publication year, sample size, country, ethnicity, age, genotyping method, and SNPs studied.

Quality assessment
The quality of the studies was independently assessed using a set of predetermined criteria that was extracted and modified from previous study (Li et al., 2013) (Table 1).The articles with higher scores (≥12) presented "high quality".The scores of all include articles in this metaanalysis were shown in Table 2.

Statistical analyses
All analyses were calculated using Stata software (V12.0,Stata, College Station, TX, USA) and the RevMan (v.5.1) program (http://www.cochrane.org/revman).To evaluate the association between IL-17 SNPs and gastric cancer, the significance of the pooled odds ratios (OR) were determined by a Z test.The Q-test and I2 statistic were used to investigate the degree of heterogeneity between the studies.Either a random-effect model or a fixed-effect model was chosen to calculate the subtotal OR and 95% CI (for the pooled odds ratio) based upon whether there was significant difference in heterogeneity among these studies.If the Q-test resulted in a p value >0.05, indicating a lack of heterogeneity, the pooled OR estimate of each study was calculated using a fixed-effect model (the Mantel-Haenszel method).If the Q-test resulted in a p value<0.05, the pooled OR estimate of each study was calculated with a random-effect model (the DerSimonian and Laird method).In addition to the pooled OR of the various genetic models, a generalized odds ratio (ORG) was calculated in this analysis by the "ORGGASMA" software (available at http://biomath.med.uth.gr), which is model-free approach and utilizes the complete genotype distribution and provides an estimate of the overall magnitude of the association (Zintzaras, 2010).
To assess the stability of results, sensitivity analysis was performed using a stepwise process in the metaanalysis, namely, a single study in the meta-analysis was deleted each time to reflect the influence of the individual data set to the pooled OR.Publication bias was examined by a Begg's and Egger's tests (p<0.05 was considered representative of statistically significant publication bias) (Begg and Mazumdar, 1994;Egger et al., 1997).

Results of literature search
A total of sixty-eight articles were identified in the initial search (Figure 1).Three of these studies were excluded because they were replicates (n=1) and review articles (n=2).In addition, sixty-five of the remaining studies were excluded after reviewing of the abstracts because they were not IL-17 polymorphism studies (n=57) and not case-control studies (n=2).Finally, six articles were selected for inclusion in the current meta-analysis.The main characteristics of these six are presented in Table 2.

Meta-analysis
In the present study, to obtain more conclusive results, we evaluated whether the SNPs within the IL-17 gene were associated with gastric cancer risk through multiple genetic models combined with a model-free approach, ORG.The results of the statistical analyses showed that rs2275913 polymorphism within the IL-17A was significantly associated with an increased risk of gastric cancer in multiple genetic comparison models (for rs2275913, A vs G: OR=1.30, 95% CI=1.14-1.48,p=0.0001, random-effect model; AA vs GG, OR=1.87, 95% CI=1.36-2.56,p=0.0001, random-effect model; AA+AG vs GG, OR=1.18, 95% CI=1.05-1.33,p=0.005, fixed-effect model; ORG=1.37,95% CI=1.13-1.67,fixed-effect model).However, no significant positive signals were observed with the risk of gastric cancer in the association analysis of the rs3748067 and rs763780 polymorphisms in the IL-17A and IL-17F, respectively (Table 3 and Figure 2).

Sensitivity analysis and publication bias
By using a stepwise process, meta-analysis was performed repeatedly when each particular study was sequentially excluded.For rs2275913, the removal of  any study one by one did not alter the statistical results.
Begg's funnel plot and Egger's test were performed to assess publication bias of the literature.The shape of the funnel plot was symmetrical and the statistical results did not show any publication bias (Begg's test: p=0.188,Egger's test: p= 0.263 for rs2275913, Figure 3).

Discussion
Previous reports indicated that chronic inflammation may be involved in the development of gastric cancer, because IL-17A and IL-17F are expressed by Th17 cells which mediate chronic inflammation and various diseases (Park et al., 2005;Hu et al., 2013;Lee et al., 2013).Thus far, some studies have investigated the association between IL-17A and IL-17F polymorphisms and susceptibility to gastric cancer, but the results were unclear.Therefore, we conducted a comprehensive meta-analysis of large samples to provide insight into the relationship between the IL-17A and IL-17F gene SNPs and gastric cancer and obtain a more reliable conclusion.
The findings from the current meta-analysis suggested that the IL-17A gene polymorphism is associated with gastric cancer risk.Importantly, the rs2275913 polymorphism is located in the 5' region near the IL-17A gene which may regulate the gene transcription.Hence, the functions of this SNP need to be further investigated in future studies.
In interpreting the results of the meta-analysis, some limitations should be addressed.First, although we observed a significant association between the IL-17A polymorphism and the risk of gastric cancer in multiple genetic models and stratified analysis by clinicopathological features, the small number of references included in our study precluded a thorough evaluation of publication bias.Second, environmental factors are also important in the development of tumor disease.The potential interactions between genetic and environmental factors may also modify the development of gastric cancer.Failing to consider these factors may influence the effects of IL-17 polymorphisms on the susceptibility to gastric cancer.Third, different genotyping methods may also have a possible effect on the frequency of the allele.Fourth, we have only reviewed published studies, but some unpublished studies, especially negative results, may affect the final conclusion.Fifth, the current data were only from Asian populations.Moreover, the 'A' allele frequency of the rs2275913 polymorphism in Chinese gastric cancer patients is 48.8% and 28.8% in Wu's study and Zhang's study, respectively, suggesting that the diversity in geography, environment, and demography might also cause genetic heterogeneity in different populations even within a country (Pan et al., 2013).
Taken together, despite some limitations, the present meta-analysis provided a more precise estimation of the relationship between the IL-17 gene SNPs and gastric cancer risk compared with the individual studies, which also supports the previous findings (Dai et al., 2014).Furthermore, other possible causative variants in the IL-17 gene should be explored.

Figure 2 .
Figure 2. Forest plot of the association between the IL-17 gene SNPs and gastric cancer.a: for rs2275913, A vs G; b: for rs2275913, AA+AG vs GG; c: for 3748067, T vs. C; d: for 3748067, TT+TC vs CC; e: for 763780, C vs T; f: for 763780, CC+CT vs TT

Figure 3 .
Figure 3. Begg's Funnel Plot of Publication Bias for the Association of the 'A' allele Carriers of IL-17A SNP rs2275913 with the Risk of Gastric Cancer.Each point represents a separate study for the indicated association

Table 2 . Characteristics of Case-control Studies of between IL-17 SNPs and Gastric Cancer Included in the Meta-analysis
*SD, standard deviation; PCR-SSCP, polymerase chain reaction-single strand conformation polymorphism; PCR-RFLP, polymerase chain reaction-restriction fragment length polymorphism Figure 1.

Table 3 . Summary Odds Ratio (OR) and 95% Confidence Intervals (CI) of IL-17 SNPs and Gastric Cancer
*Estimates for random-effects model; † Estimates for fixed-effects model; # p value of Z-test for pooled odds ratio; § Added 0.01 to zero frequency of cell; ORG, generalized odds ratio; ‡ p value for Q-test to estimate heterogeneity between studies, p< 0.05 indicates significant heterogeneity

Table 4 . Association between the 'A' Allele Carriers of IL-17A SNP rs2275913 and Clinicopathological Features of Gastric Cancer
*Estimates for random-effects model; † Estimates for fixed-effects model; # p value of Z-test for pooled odds ratio