Association Between XRCC 1 Gene Polymorphisms and Risk of Glioma Development : A Meta-analysis

Gliomas are tumors of glial origin of the central nervous system (CNS), exhibiting various degrees of differentiation inside the same tumor (Kyritsis et al., 2010). Gliomas account for almost 80% of primary malignant brain tumors, and result in more years of life lost than do any other tumors (Schwartzbaum et al., 2006). While the exact molecular causes of gliomas remain unclear, ionizing radiation (IR) and genetic alterations have been demonstrated to be established risk factors for gliomas (Schwartzbaum et al., 2006). It has been well recognized that DNA damage is an important mechanism in the development of various cancer including gliomas. If damaged DNA is not repaired, mutations and development of cancer occurs. In this perspective, polymorphisms of DNA repair genes are plausible candidates which can modify the risk of gliomas. At present, increasing studies have been conducted to investigate the potential association between DNA repair genes and the risk of gliomas (Goode et al., 2002; Felini


Introduction
Gliomas are tumors of glial origin of the central nervous system (CNS), exhibiting various degrees of differentiation inside the same tumor (Kyritsis et al., 2010).Gliomas account for almost 80% of primary malignant brain tumors, and result in more years of life lost than do any other tumors (Schwartzbaum et al., 2006).While the exact molecular causes of gliomas remain unclear, ionizing radiation (IR) and genetic alterations have been demonstrated to be established risk factors for gliomas (Schwartzbaum et al., 2006).It has been well recognized that DNA damage is an important mechanism in the development of various cancer including gliomas.If damaged DNA is not repaired, mutations and development of cancer occurs.In this perspective, polymorphisms of DNA repair genes are plausible candidates which can modify the risk of gliomas.
Wang et al. firstly investigated the association of XRCC1 gene polymorphisms and the gliomas risk (Wang et al., 2004).Since then, a series of studies especially conducted in recent years for the associations between XRCC1 gene polymorphisms and the risk of gliomas have been published.Unfortunately, these epidemiological studies performed in different countries have yielded conflicting results from strong links to no association.The inconsistency of these studies may be explained by the relatively small sample size, difference in population background and study design, etc.Therefore, a timely meta-analysis summarized the results of these studies would provide more accurate estimate of the associations between XRCC1 gene polymorphisms and risk of gliomas.

Literature and search strategy
A computerized literature search was performed to identify the relevant studies from 5 electronic databases including PubMed, ISI Web of Science, China National Knowledge Infrastructure (CNKI), Database of Chinese Scientific and Technical Periodicals (VIP), and China Biology Medical literature database (CBM).The following key words were jointly used: ("X-ray crosscomplementing group 1" or "XRCC1" or "DNA repair gene") and ("glioma" or "glioblastoma" or "astrocytoma") and ("polymorphism" or "gene mutant").The reference lists of review articles and the references cited in the retrieved studies were hand-searched for the collection of omitted relevant studies.If more than one article were published using the same case series, only the study with largest sample size was selected.The literature search was updated on Jun, 2012.

Inclusion criteria
The studies included must meet the following criteria: (1) evaluating the association between XRCC1 Arg399Gln (rs25487) and/or Arg194Trp (rs1799782) and/or Arg280His (rs25489) polymorphisms and the risk of gliomas; (2) case-control or cohort design; (3) the study was published in English or Chinese; (4) providing sufficient data for calculation of odds ratio (ORs) with the corresponding 95% confidence interval (95%CI).All identified studies were reviewed independently by two investigators to determine whether an individual study was eligible for inclusion in this meta-analysis.

Data extraction
Two investigators independently extracted the data from all eligible publications according to the criteria.The following information was extracted from each study: (1) name of the first author; (2) year of publication; (3) country of origin; (4) ethnicity of the study population; (5) source of control subjects; (6) numbers of cases and controls; (7) gender and age of enrolled subjects; and (8) numbers of genotypes in cases and controls.

Statistical analysis
χ 2 analysis with exact probability was used to test departure from Hardy-Weinberg equilibrium (HWE) for the genotype distribution in controls.The association between XRCC1 gene polymorphisms and risk of gliomas was estimated by calculating pooled ORs and 95%CI.We estimated the risk for mutant alleles, variant homozygous genotypes and heterogeneous genotypes compared with the wild-type homozygous genotypes, respectively, and then for the combination of homozygous and heterogeneous genotypes compared with the wild type genotypes.The significance of the pooled effect size was determined by Z test.Heterogeneity among studies was assessed using Q test as well as the I 2 statistic (Higgins and Thompson, 2002).The DerSimonian and Laird random effect model (REM) was used as the pooling method when I 2 > 50%, otherwise, the Mantel-Haenszel fixed effect model (FEM) was considered to be the appropriate choice (Higgins and Thompson, 2002).Subgroup analyses were stratified by ethnicity, source of control, and the methods for genotyping.Sensitivity analysis was undertaken by removing an individual study each time to check whether any of single study could bias the overall estimate (Tobias, 1999).Begg's funnel plots and Egger's regression test were undertaken to assess the potential publication bias (Harbord et al., 2006).Probability less than 0.05 was judged significant except for the I 2 statistic.Data analysis was performed using STATA version 11 (StataCorp LP, College Station, Texas, USA).

Characteristics of studies
A total of 47 relevant studies concerning XRCC1 gene polymorphisms and the risk of gliomas were identified.36 studies were excluded, while 11 studies mainly published in 2007-2012 were finally included in the analysis (Wang et al., 2004;Felini et al., 2007;Kiuru et al., 2008;Liu et al., 2009;McKean-Cowdin et al., 2009;Rajaraman et al., 2010;Yosunkaya et al., 2010;Custodio et al., 2011;Hu et al., 2011;Liu et al., 2011;Zhou et al., 2011).The study by Kiuru et al. enrolled subjects from four countries including Demark, Finland, Sweden, and UK, but did not provide the genotypes of cases and controls of each country, thus was still considered as one study (Kiuru et al., 2008).Among these relevant studies, 3 studies investigated 2 polymorphisms, while 4 studies investigated 3 polymorphisms, and thus a total of 11, 7, and 4 studies were finally included in the analysis of associations between XRCC1 Arg399Gln, Arg194Trp and Arg280His polymorphisms and the risk of gliomas, respectively.All the included studies were case-control design, and used peripheral blood samples for DNA extraction.Genotyping was performed by using polymerase chain reactionrestriction fragment length polymorphism (PCR-RFLP), TaqMan, and etc.Most of the studies did not mention the ionizing radiation of enrolled subjects, while higher ionizing exposure rate in cases was reported in 4 studies (Liu et al., 2009;Hu et al., 2011;Liu et al., 2011;Zhou et al., 2011).The detailed characteristics of the included studies were shown in the Table 1.

Meta-analysis results
A total of 11 studies containing 3810 cases and 6079 controls were included for the analysis of XRCC1 Arg399Gln polymorphism and the risk of gliomas, while 7 studies containing 2928 cases and 5048 controls, and 4 studies containing 1461 cases and 2593 controls were  included in the analysis of associations between XRCC1 Arg194Trp and Arg280His polymorphisms and the risk of gliomas, respectively.We found that the mutant allele frequency for Arg399Gln polymorphism in controls of Asians was similar to that of Caucasians (35.3% vs. 34.6%).In contrast, the mutant allele frequencies of Arg194Trp and Arg280His polymorphisms in control subjects of Asians were obviously larger than those in Caucasians (allele frequency, 25.8% vs. 7.2%, and 18.1% vs. 5.3%, respectively).
Results of pooled analysis on the associations between XRCC1 Arg399Gln polymorphism and the risk of gliomas were shown in Table 2.As significant between study heterogeneity was detected (I 2 >70%), thus REM was used for the analysis.The pooled results showed that 399Gln allele was not significantly associated with the gliomas risk compared with the Arg allele (OR = 1.150, 95%CI 0.975-1.356).However, significant association was found between the genotypes comparisons (ArgGln vs. ArgArg, OR = 1.234, 95%CI 1.018-1.497;GlnGln/ ArgGln vs. ArgArg, OR = 1.231, 95%CI 1.008-1.504).The subgroup analysis stratified by ethnicity showed that significant associations also existed in Asians (GlnGln vs. ArgArg, OR = 1.490, 95%CI 1.031-2.153;GlnGln/ArgGln vs. ArgArg, OR = 1.321, 95%CI 1.037-1.684),whereas no significant association was found in Caucasians.The between study heterogeneity in Asians was moderately decreased.In the subgroup analysis stratified by the source of control, no significant association was found in allele comparisons as well as in the genotype comparisons.In regard to the subgroup analysis stratified by genotyping methods, significant association was found in the subgroup of other genotyping methods (Glu vs. Arg, OR=1.129, 95%CI 1.043-1.223;GlnGln/ArgGln vs. ArgArg, OR=1.192, 95%CI 1.068-1.330)(Table 2 and Figure 1).
Results of pooled analysis on the associations between XRCC1 Arg194Trp and Arg280His polymorphisms and glioma risk were shown in Table 3.For the Arg194Trp polymorphism, no significant association was revealed in each genetic contrasts in the worldwide population, except for the homozygous genotype comparisons (TrpTrp vs. ArgArg, OR = 2.0209, 95%CI 1.398-2.945).
As apparent difference in allele distribution between Asians and Caucasians, we only performed subgroup analysis based on ethnicity.No significant association was found in Asians as well as in Caucasians.Extreme between study heterogeneity existed for the analysis of the worldwide population, which was markedly decreased in the subgroup analysis of Asians and Caucasians.For the Arg280His polymorphism, no significant association was found in each contrast (Table 3 and Figure 1).

Sensitivity analysis and publication bias
Sensitivity analysis was undertaken by removing one individual study each time to check whether any of single

Discussion
Glioma is the most common type of primary brain malignancy in adults.Despite recent advances in cancer diagnosis and therapy, the prognosis of patients with gliomas remains dismal (Sathornsumetee et al., 2007).DNA repair genes play important roles in maintaining the genome integrity, and thus polymorphisms of DNA repair genes are plausible candidates which can modify the risk of gliomas.XRRC1 is one of the most important DNA repair genes responsible for BER pathway and fixes base damage and DBS caused by IR (Tudek, 2007).Up to now, a series of studies have been performed to address the association between XRCC1 gene polymorphisms and the risk of gliomas, but yielded conflicting results.Because of the above-mentioned conflicting results from relatively small studies underpowered to detect the effects, a metaanalysis should be an appropriate approach to obtain a more definitive conclusion.
To our knowledge, this was the first meta-analysis addressing the association between XRCC1 gene polymorphisms and the gliomas risk.In this study, a total of 11 studies (3810 cases and 6079 controls), 7 studies (2928 cases and 5048 controls), and 4 studies (1461 cases and 2593 controls) were included in the analyses of the associations between XRCC1 Arg399Gln, Arg194Trp, Arg280His polymorphisms and gliomas risks, respectively.The pooled results showed that GlnGln carrier was associated with moderately increased risk of gliomas in Asians (GlnGln vs. ArgArg, OR = 1.490, 95%CI 1.031-2.153;GlnGln/ArgGln vs. ArgArg, OR = 1.321, 95%CI 1.037-1.684),whereas only marginal association was revealed in Caucasians.However, it should be noticed that all the three studies in Asians enrolled Chinese subjects, thus the effects on people of other Asian countries such as Japan, Korea, etc were still unclear.For the Arg194Trp polymorphism, no significant association was revealed in each genetic contrasts in the worldwide population, except for the homozygous genotype comparisons (TrpTrp vs. ArgArg, OR = 2.209, 95%CI 1.398-2.945),which might be due to the inclusion of the study by Custodio et al. (2011).In the subgroup analysis, no significant association was found in Asians as well as in Caucasians.In regard to the Arg280His polymorphism, no significant association was found in each contrast.
The XRCC1 gene encodes the XRCC1 protein, which serves as a scaffold for two other proteins, DNA ligase III and polymerase β, and also serves as a single-strand break sensor by its interaction with poly (ADP-ribose) polymerase (PARP) ( Caldecott et al., 1994;Caldecott et al., 1996;Masson et al., 1998).The observed association between XRCC1 Arg399Gln gene polymorphism and the gliomas risk is biologically plausible.The Arg399Gln is located at the carboxylic acid terminal side of the polyadenosine diphosphate-ribose polymerase interacting domain, and the variant Gln allele has been shown to reduce DNA repair capacity, and thereby, increase the risk of developing glioma (Duell et al., 2000).However, the XRCC1 codon 194 and codon 280 polymorphisms located in the linker region were not found to be significantly associated with risk of gliomas, which was consistent with almost all the individual studies.
Similar to other systematic reviews and meta-analyses, our study also has some limitations.First, the present meta-analysis was based primarily on unadjusted effect estimates and CIs, thus the effect estimates were relatively imprecise.Second, relatively small sample size existed for some subgroup analyses especially for Arg194Trp and Arg280His polymorphisms, as limited studies were included.Third, cancer is known as a multifactor disease, however, the gene-gene and gene-environment interactions were not addressed in this meta-analysis, and thus the potential roles of the above gene polymorphism may be masked or magnified by other gene-gene/geneenvironment interactions.Lastly, although we did not detect publication bias, selection bias may exist because only studies published in English or Chinese were retrieved.
In summary, this meta-analysis systematically analyzed the association between XRCC1 polymorphisms and the gliomas risks.The pooled results showed that the XRCC1 Arg399Gln polymorphism was moderately associated with increased risk of gliomas in Asians.In contrast, XRCC1 Arg194Trp and Arg280His polymorphisms were not significantly associated with gliomas risks.Due to the limited studies and the potential confounders, further studies are needed to confirm these results.

Figure 2 .
Figure 2. Begg's Funnel Plot with the Egger's Test for Publication Bias of XRCC1 Polymorphisms and the Risk of gGliomas.(A) GlnGln/ArgGln vs. ArgArg; (B) TrpTrp/ArgTrp vs. ArgArg; (C) HisHis/ArgHis vs. ArgArg.The horizontal line in the funnel plot indicates the fixed-effects summary estimate, whereas the diagonal lines pseudo-95% CI limits about the effect estimate.In the absence of publication bias, studies will be distributed symmetrically above and below the horizontal line

Table 1 . Characteristics of Individual Studies for the Association Between XRCC1 Polymorphisms and Gliomas Risk
DOI:http://dx.doi.org/10.7314/APJCP.2012.13.9.4783  XRCC1 Gene Polymorphisms and Glioma Risk: A Meta-analysis a Sex were shown as the percentage of male in cases and controls; b Age were shown as the mean age of case and controls; c the IR (ionizing radiation) were shown as the percentage of the subjects with IR exposure history in cases and controls; d Mixed means that population and hospital based controls were used; e 11,12,22 represent the homozygous wild genotypes, heterozygous genotypes, and the homozygous mutant genotypes for the three SNPs, respectively; f genotyping methods: MassARRAY, genotyping was performed using the Sequenom MassARRAY iPLEXTM platform2; PCR-CTPP, polymerase-chain-reaction with the confronting-two-pair primer; PCR-RFLP, polymerase chain reaction-restriction fragment length polymorphism; NA, means not available

Table 3 . Summary of ORs and 95%CI for the Association Between XRCC1 Arg194Trp (rs1799782, C/T) and Arg280His (rs25489, G/A) Polymorphisms and Risks of Gliomas
a M, model for meta-analysis; F, fixed effect model; R, random effect model; b p value for heterogeneity based on Q test; " *", represents p<0.05; "-", means not available Figure 1.

Meta-analysis for XRCC1 Polymorphisms and the Gliomas Risks.
(A) Random effect model.Top: