Causal Association Between Modifiable Exposure Factors And Barrett’s Esophagus

Objective: Barrett’s esophagus(BE)is the only known precancerous lesion of esophageal adenocarcinoma(EA),and its incidence rate has continued to rise in recent years.Traditional epidemiological studies demonstrated that modifiable exposure factors are related to BE.However,the results of various studies are inconsistent and susceptible to the influence of confounding bias and reverse causality.Therefore,it is difficult to determine whether this relationship has a causal association.In addition,gastroesophageal reflux disease(GERD)and BE share many common risk factors,but it is not clear whether these risk factors promote the occurrence of BE through GERD.So,this study aims to systematically explore the causal relationship between modifiable factors and BE and the role of GERD in this relationship by using large sample gene sequencing data.Methods: This study obtained data from multiple public gene sequencing databases and constructed instrumental variables for each exposure factor one by one.Then it is divided into two stages for mendelian randomization(MR)analysis.In the first stage,Two Sample Mendelian Randomization(TSMR)and Multivariate Mendelian randomization(MVMR)methods were adopted to verify the causal association between modifiable exposure factors and BE.In the second stage,the mediation effect of GERD was analyzed by using the two-step Mendelian randomization method.This study adopted a variety of statistical methods to complement each other.The Inverse-variance weighted(IVW)is the main statistical verification method.The weighted median(WM),MR-Egger regression,and MR-PRESSO were secondary verification methods.MR-egger intercept,MR-PRESSO,Q test,Leave-One-Out method and Scatter plot were used for sensitivity analysis to ensure the credibility of the results.Results:In this study,the screened exposure factors were divided into two categories for analysis.First,among lifestyle-related exposure factors,TSMR analysis showed that smoking and moderate-vigorous degree physical activity(MVPA)were the risk factor for BE,while education attention level is the protective factor for BE(Smoking: OR=1.4095%CI:1.11-1.75,P<0.01;MVPA: OR=2.59,95%CI:1.10-6.09,P=0.02;Education attention: OR=0.75,95%CI:0.60-0.93,P=0.01).In obesity-related exposure factors,TSMR and MVMR analysis showed that WHR was the best clinical phenotype of obesity closely related to BE.Further subgroup analysis showed that the greater the waist-to-hip ratio in men over 50 years of age,the higher the risk of BE(male WHR: OR=1.37,95%CI: 1.08-1.73,P<0.01;WHR of male over 50 years old: OR=1.48,95%CI: 1.00-2.20,P=0.04).Further analysis of the remaining significant factors showed that except MVPA,other significant factors were causally correlated with GERD(MVPA:P=0.46).In addition,the cause-and-effect relationship of education attention level with BE were significantly affected by GERD,and its correlation disappears after adjusting GERD(P=0.65).Finally,two-step mendelian randomization analysis showed that GERD had a mediating effect of about 25% and 32%,respectively,in the causal effect of smoking and WHR on BE.All the sensitivity analysis methods showed that the main results of this study were robust and reliable.Conclusion: Mendelian randomization analysis found that modifiable factors such as smoking,MVPA,WHR were all risk factors to BE,and the causal association between MVPA and BE was not affected by GERD.GERD plays an important mediating role in the causal effect of smoking and WHR on BE.Therefore,early intervention of these risk factors can reduce the incidence of BE.