Familial Aggregation Of Gastric Cancer:Individual And Joint Contribution Of Environmental Exposures And Polymorphisms In The Mismatch Repair Pathway

The risk of gastric cancer(GC)is increased in subjects with GC family history(GCFH).However,the mechanism for familial aggregation of GC is only understood for some inherited syndromes,which accounts for less than 3% of all GC cases.A possible explanation for familial aggregation of GC is that family members of GC patients may have shared environmental exposures and genetic variants that increase GC risk.The verified environmental exposures associated with GC include cigarette smoking,alcohol drinking,high intake of pickled food,Helicobacter pylori(Hp)infection,and so on.Although the interaction between Hp infection and GCFH has been reported in GC carcinogenesis,the associations between GCFH and other environmental exposures still are unclear.In genetic variants that may cause GC,the roles of single nucleotide polymorphisms(SNPs)have been widely studied and verified.Recent studies have reported that SNPs in genes involved with the mismatch repair,one of the key repair mechanism for DNA damages,significantly impact GC risk.And the related effects of these SNPs are particularly strong in population with family history of cancer,indicating that they may contribute to familial aggregation of GC.Therefore,we hypothesized that:1)there are interactions between environmental exposures and GCFH to contribute to familial aggregation of GC;2)SNPs in MMR genes are associated with familial aggregation of GC;3)there are interactions between environmental exposures and the MMR genes SNPs to contribute to familial aggregation of GC.To test our hypothesis,a case-control study was conducted to investigate the following issues: 1)the interactions between environmental exposures and GCFH and their contributions to GC risk;2)the associations of SNPs in MMR genes with GC risk in subjects with and without GCFH respectively,and the relationship of these SNPs with microsatellite instability(MSI)which is the main character of MMR abnormality;3)the interactions between environmental exposures and SNPs of MMR genes in subjects with and without GCFH respectively,and their contributions to GCrisk.In conclusion,this study suggest the individual and joint contribution of environmental exposures and SNPs of MMR genes to familial aggregation of GC,which partly explains the high risk of GC caused by GCFH and may provide theory evidences for early screening and individual prevention of GC.Part I.The relationship of environmental exposures with familial aggregation of GCBackgrounds: Environmental exposures are crucial for GC carcinogenesis.Family members of GC patients may have shared lifestyles and dietary habits associated with GC risk and high prevalence of Hp infection,which are potential causes of familial aggregation of GC.Purpose: To investigate the contribution of environmental exposures to familial aggregation of GC by exploring the interactions between environmental exposures and GCFH and subsequent GC risk induced by such interactions.Methods: A hospital-based case-control study was conducted.With information of GCFH,643 GC cases and 643 matched controls according to age,sex and residence were selected from 1004 GC patients and 1550 healthy people.The associations of GC risk with GCFH,sex,age,smoking,alcohol drinking,intake of:consumption and Hp infection were analyzed by logistic regression.Trend tests were used to study the relationship of GC risk with number of relatives with GC.The interactions between environmental exposures and GCFH were further investigated.Results: GCFH significantly increased GC risk.After adjusted for environmental exposures,subjects with GCFH had a 5.05 fold increased GC risk relative to those without GCFH(OR=5.05,95%CI: 3.32-7.67),and the risk increased with increasing number of relatives with GC(P trend<0.001).Smoking(OR=1.72,95%CI:1.21-2.43),alcohol drinking(OR=1.57,95%CI:1.12-2.19),high intake of pickled vegetables(OR=6.33,95%CI:4.69-8.54)and Hp infection(OR=1.53,95%CI:1.23-2.01)were found to be risk factors,while green tea drinking(OR=0.37,95%CI:0.30-0.48),high intake of milk/dairy products (OR=0.40,95%CI:0.32-0.52),and high fruit consumption(OR=0.32,95%CI:0.26-0.38)were found to be protective factors for GC.No significant interaction was found between GCFH and any single exposure(all P>0.05).However,a significant interaction was observed between GCFH and the number of risk exposures(P interaction=0.025).The risk of GC increased with increasing number of risk exposures,and this effect was much stronger in subjects with than without GCFH.Conclusions: GCFH significantly increased GC risk.GCFH significantly interacted with the number of risk exposures,indicating that the combination of multiple risk factors contributes to the familial clustering of GC.Part II.The relationship of SNPs in MMR genes with familial aggregation of GCBackgrounds: The MMR system,which identifies and repairs incompatible DNA base pairs that are generated during DNA replication and recombination,is crucial for genome stability.Microsatellite instability(MSI),the somatic gain or loss in simple repeat(microsatellite)DNA sequences caused by MMR abnormity,is associated with gastric carcinogenesis.The core MMR genes include MSH2,MLH1,MSH3,MSH6,PMS1 and PMS2,other genes such as PMS2L3、EXO1、TREX1、MGMT、RECQL、POLD3 and TP73 also take part in MMR.SNPs in MMR genes may be associated with individual differences in the MMR capacity,and therefore are important causes for heterogeneous susceptibility of individuals for GC.Recently,a study found that SNPs of MSH2 gene was associated with GC risk of subjects with cancer family history,suggesting that SNPs of MMR genes may contribute to the familial clustering of GC.Purpose: To explore the contribution of SNPs in MMR genes to familial aggregation of GC by studying the associations of these SNPs with GC risk in subjects with and without GCFH respectively.To investigate potential functions of GC-related SNPs by exploring relationships between these SNPs with MSI.Methods: The studied subjects were the same with part I.Genomic DNA was extracted from peripheral blood.Genotyping was done by Taqman allelic discrimination assays for SNPs of MMR genes that was found to impact GC risk in our previous studies,including rs1800734 in MLH1,rs2303428 and rs3732183 in MSH2,rs735943 in EXO1,and rs11797 in TREX1.The associations of SNPs with GC risk in subjects with and without GCFH respectively were analyzed by logistic regression and trend tests.Genomic DNA from available tumor samples was further extracted.MSI status was evaluated by the single fluorescent multiplex PCR and the correlations of SNPs with MSI status were analyzed.Results: All genotype distributions among the controls were in Hardy-Weinberg equilibrium.Compared with the GG genotype,thers11797 GA or AA genotype of TREX1 significantly decreased GC risk in subjects with GCFH(OR = 0.48;95% CI:0.23-0.99;P trend= 0.037),but this effect was not significant in those without GCFH(OR = 0.91;95% CI: 0.65-1.28;P trend= 0.560).For the rs1800734 of MLH1,the GA or GG genotype no significantly increased GC risk in subjects with GCFH(OR = 1.46;95% CI: 0.69-3.09;P trend= 0.475)but significantly deceased GC risk in those without GCFH(OR = 0.70;95% CI: 0.49-1.01;P trend= 0.045)relative to the AA genotype.According to OR,we defined at-risk genotypes as those with increased GC risk in subjects with GCFH,including rs1800734 GA or GG,rs2303428 CT or CC,rs11797 GG,rs735943 TC or TT,andrs3732183 AA.The risk of GC in subjects with GCFH also significantly increased with increasing number of at-risk genotypes(P trend=0.041).In contrast,the combined at-risk genotypes no significantly decreased GC risk in those without GCFH(P trend= 0.426).Compared with the GG genotype,the rs11797 GA or AA genotype significantly decreased the MSI phenotype in patients with GCFH(P = 0.036)and no significantly decreased it in patients without GCFH(P = 0.462).Compared with the AA genotype,the rs1800734 GA or GG genotype decreased the MSI phenotype in patients without GCFH(P = 0.036)while increasing it in patients with GCFH(P = 0.003).Conclusion: The rs11797 of TREX1 and the rs1800734 of MLH1 polymorphisms may be associated with GC risk of subjects with GCFH by inducing MSI.A combined effect was observed for at-risk genotypes in subjects with GCFH that GC risk was increased with increasing number of at-risk genotypes.Together,SNPs of MMR genes may contribute to familial aggregation of GC.Part III.The interactions between environmental exposures and MMR genes SNPs and the relationship of such interactions with familial aggregation of GCBackgrounds: The interactions between environmental and genetic factors are crucial for GC carcinogenesis.Recently,we reported that there were synergistic interactions between environmental exposures and SNPs of MMR genes to impact GC risk.However,it is still not clear that whether such interactions can be found in subjects with GCFH and whether there are differences for such interactions between subjects with and without GCFH.Purpose: To investigate the combined effects between environmental exposures and SNPs of MMR genes on familial aggregation of GC through studying their interactions in subjects with and without GCFH respectively and the impact of such interactions to GC risk.Methods: Based on data from part I and II,the interactions between environmental exposures and SNPs of MMR genes and their impacts to GC risk were evaluated.Results: In subjects with GCFH,the TREX1 rs11797 polymorphism significantly interacted with the intake level of pickled vegetables(P interaction=0.013).Compared with carriers of the GG genotype,the carriers of rs11797 GA+AA genotype had no increased GC risk in subjects with GCFH(OR = 0.99;95% CI: 0.11-9.17)but had an increased GC risk in those without GCFH(OR = 2.42;95% CI: 1.02-5.76),indicating an antagonistic effect of the rs11797 GA or AA genotype in subjects with GCFH on the role of high intake of pickled vegetables which increases GC risk.In subjects without GCFH,the rs11797 GA or AA genotype significantly interacted with both green tea drinking(P interaction=0.037)and soy food intake(P interaction=0.045).In detail,green tea drinkers carrying rs11797 AG or AA genotype were at decreased risk of GC as compared to nondrinkers carrying rs11797 GG genotype(OR = 0.29;95% CI: 0.16-0.50).And compared with carriers of rs11797 GG genotype with low intake of soy food,the carriers of rs11797 GA+AA genotype with high intake of soy food had a reduced GC risk(OR = 0.24;95% CI: 0.09-0.67).Above results suggest a joint effect of the rs11797 GA or AA genotype with green tea drinking and soy food consumption on decreased GC risk in subjects without GCFH,which was not observed in subjects with GCFH.No significant interaction was observed between other SNPs of MMR genes and environmental exposures in subjects with or without GCFH.Conclusions: Environmental exposures may interact with some genotypes of SNPs in MMR genes,which impacts GC risk in subjects with GCFH,suggesting their combined effect on familial aggregation of GC.