Association Between Tumor Necrosis Factor-Α Gene Polymorphisim And Risk Of Oral Cancer-a Meta-analysis

BackgroundOral cancer ranked the sixth in the occurrence of systemic cancers,5-years survival for patients was about50%. The current treatment of oralcancer is still surgery with radiotherapy and chemotherapy. Althoughvarious treatments are evolving updated, but prone to early oral cancerlymph node metastasis, the prognosis is still not optimistic, postoperativefunctional preservation is poor and bring great burden to individuals andsociety.Oral cancer is a complex disease resulting from the common effects ofgenetic and environmental factors. There is close relation between thegenetic susceptibility and development of oral cancer. With the completionof the development of molecular genetics and the human genome project,much attention was paid to single nucleotide polymorphisms (singleneucleotide polymorphism, SNP) which can reflect and determine thegenetic differences between individuals, disease susceptibility and drugresponse. Finding pathogenic susceptibility candidate genes become aresearch hotspot. Role of inflammation in the pathophysiology of oralcancer has also been a growing number of studies, inflammatory cytokines,as one candidate susceptibility genes, its relationship of gene SNP with therisk of oral cancer is also much concerned. Tumor necrosis factor-α (Tumor necrosis factor-α, TNF-α) is animportant proinflammatory cytokine, and its secretion will increase in manypathological conditions. TNF-α expression level in the body is regulated byits gene polymorphisms. Gene promoter region and its role in inflammatorydiseases of the SNP has been a research focus in recent years. TNF-αpromoter region comprising SNPs-308(G�'A),-238(G�'A),-1031(T�'G),-863(C�'A),-851(C�'T), etc., most studies focus on the-308,-238, and less of other studies promoter SNP sites be involved. Generallybelieved,-308G/A polymorphism may affect transcription, A allele canenhance transcription of TNF-a and is involved with the severity orprognosis of disease, while G allele is associated with a low level of TNF-aexpression; TNF-α-308SNP loci may be involved with autoimmunediseases, infectious diseases, cancer and post-transplant rejection, etc., andcan affect the pathological process of some of these diseases.Currently, there are a number of relevant researches about TNF-αpolymorphism and the risk of oral cancer. However, the results of thesestudies are not entirely consistent, even contradictory, which may be relatedto the small sample size of individual studies and clinical heterogeneity andstatistical differences. So, the statistical performance is lacking ofeffectiveness. Therefore, we use Meta-analysis methods, collectcomprehensively relevant literatures and use quantitative methods forstatistical analysis, in order to get more objective and reliable conclusions.ObjectiveTo explore the association between Tumor Necrosis Factor-α genepolymorphisms and risk of oral cancer by carring out a meta-analysis whichwas based on published articles.MethodsWe searched PubMed, Cochrane Library, CNKI database for a comprehensive literature collection published before10August2013aboutall TNF-α single nucleotide polymorphisms and oral cancer, supplementedby hand searches of the literature retrospective.Then, developed strictinclusion/exclusion criteria, screened qualified full-text literature, thedata were extracted by two independent reserachers, the stasticalperformance was done by the software STATA. The odds ratio (odds ratio,OR) and95%confidence intervals (confidence interval, CI) were as theevaluation criteria, the degree of association about TNF-α genepolymorphisms and risk of oral cancer was calculated under four differentgenetic models, including: allele model (Avs.G), additive genetic model(homozygous model GGVS.AA), dominant genetic model (GG+GA vsAA.) and recessive genetic model (GG vs.GA+AA). Q test and I2statisticwas used for heterogeneity test. The appropriate data combined effectsmodel selection was based on heterogeneous situation. If no resultsbetween studies heterogeneity, using Mantel-Haenszel fixed-effectsmodel; otherwise, using DCTSimoninan-Laird random effects model.Subgroup analyis exploring heterogeneity is according to ethnic origin,source control, and oral cancer types. Use sensitivity analysis to assess theresults of stability. Assess publication bias By Egger’s and Begg’s test.ResultsA total of8studies involving1280cases and1508controls wereincluded in the meta-analysis.8studies have examined the relationshipbetween all studies TNF-α-308polymorphism with oral cancer,4studiesexamined the relationship between TNF-α-238polymorphism with oralcancer, and the results are as follows:Meta-analysis about TNF-α-308G>A polymorphism: TNF-α-308G>A polymorphism was significantly associated with the risk of oral cancerin additive genetic model (GG vs. AA, OR=0.19,95%CI:[0.04,1.00], P= 0.05, I2=68.9%) and dominant genetic model (GG+GA vs. AA, OR=0.22,95%CI:[0.06,0.82], P=0.03, I2=52.4%); however, no significantassociations with risk of oral cancer were found in allele contrast (G vs. A,OR=0.70,95%CI:[0.23,2.16], P=0.54, I2=95.9%) and recessive geneticmodel (GG vs. GA+AA, OR=0.72,95%CI=0.33-1.57, P=0.41, I2=93.1%).All of the calculations were conducted under random-effect model.Subgroup analyses were conducted according to ethnicity, source ofcontrol, and type of oral cancer. In brief, in the subgroups of Asianpopulation, hospital-based case-control studies, and the studies concernedto OSCC patients, no significances were found in four genetic models.However, in the population-based case-control studies, it was found thatTNF-α-308G>A polymorphism was significantly associated with the risk oforal cancer in additive genetic model and dominant genetic model：GG vs.AA, OR=0.03,95%CI:[0.001,0.44, P=0.01；GG vs. GA+AA, OR=0.03,95%CI=0.002-2.52, P=0.02。We performed sensitivity analysis by sequential removal ofindividual studies, the results were not materially altered by leaving outstudies except for the following situations:(1) In the additive geneticmodel, the I2was decreased to18.6%when the study by Liu CJ et al.(2005) was excluded, however, the pooled OR and significance were notmaterially changed under fixed-effect model (OR=0.10,95%CI:[0.06,0.19], P=0.000); however, after the sequential removal of other individualstudies, the pooled ORs were between0.15and0.32, and no statisticalsignificance was found.(2) In the dominant genetic model, the I2wasdecreased to0.0%when the study by Liu CJ et al.(2005) was excluded,however, the pooled OR and significance were not materially changedunder fixed-effect model (OR=0.10,95%CI:[0.06,0.19], P=0.000); however, after the sequential removal of other individual studies, thepooled ORs were between0.22and0.32, and were not statisticallysignificant.Meta-analysis about TNF-α-238G>A polymorphism: For SNPTNF-α-238G>A, significant associations with oral cancer risk were foundin the allele contrast (G vs. A, OR=2.75,95%CI:[1.25,6.04], P=0.01,I2=0.0%) and recessive genetic model (GG vs. GA+AA, OR=2.23,95%CI:[1.18,4.23], P=0.01, I2=0.0%). No heterogeneity was found betweenstudies and the analyses were both conducted under fix-effect model.Additive and recessive genetic model were not presented because therewas no distribution of genotype AA in the three available studies (Liu etal.,2005; Gupta et al.,2008; Yang et al.,2011). Subgroup and sensitivityanalyses were not carried out due to the small number of studies.Begg’s test or Egger’s test, which was designed to indicate publicationbias, proved to be no obvious publication bias for each type of geneticmodel for TNF-α-238G>A and TNF-α-308G>A.Conclusion1. This meta-analysis indicates that TNF-α polymorphisms maycontribute to the risk of oral cancer. Allele G and the GG+GA genotype ofTNF-α-308G>A may decrease the risk of oral cancer, while allele G and theGG genotype of TNF-α-238G>A may cause an increase.2. Given the small sample size and the noteworthy heterogeneity, andnot considering the "gene-gene" and "gene-environment" and otherconfounding factors, more multiple prospective、 large-scale、well-designed studies will be carried out in the future in order to obtain a large sample of more realistic results.