Different Immune-related Gene Sets Predict The Prognosis And Drug Response Of Esophageal Adenocarcinoma And Esophageal Squamous Cell Carcinoma

Objective:To screen the immune-related genes that affect the prognosis of esophageal carcinoma(EC)through bioinformatics methods,and construct a prognostic risk scoring model based on immune-related gene signatures to provide prognostic evaluation and individualized treatment of EC patients with clinical decision-making in accordance with.Method:First,download the clinical information and RNA-seq data of 158 EC patients(78 esophageal adenocarcinoma(EAC)samples,80 esophageal squamous cell carcinomas)in the American Cancer Genome Atlas(TCGA)database(esophageal squamous cell carcinoma,ESCC)samples and 10 adjacent non-cancerous tissue samples).Use Perl language script to organize the transcriptome data into a gene expression profile matrix.The differentially expressed genes of EAC and ESCC were calculated and determined through the limma package,and the immune-related genes(IRGs)of EAC and ESCC were screened in combination with the online immune database,IMMPORT.Subsequently,further analysis of GO and KEGG enrichment pathways was performed to explore the biological functions of immune-related genes of EAC and ESCC.Combined with the clinical information of patients,IRGs related to the prognosis of EAC and ESCC were screened out through univariate Cox regression analysis and Lasso regression.Then,through the use of multivariate Cox regression analysis,a prognostic risk scoring model of EAC and ESCC immune-related gene signatures was constructed to score the prognostic risk of EAC and ESCC patients.According to the median of high and low risk scores,patients were divided into high-risk groups and low-risk groups,and the ROC curve and Kaplan-Meier survival analysis were used to analyze each subgroup to verify the prognostic risk score model.Then,analyze the relationship between the risk score and the clinicopathological characteristics of EAC and ESCC,and use univariate and multivariate Cox regression analysis to determine the effectiveness of the risk score as an independent prognostic factor.Finally,use the R package "pRRophetic" to predict drug sensitivity for high-and low-risk patients in EAC and ESCC.Result:Our study included 158 patients with clinical data for esophageal cancer,including 78 samples with esophageal adenocarcinoma,80 samples with esophageal squamous cell carcinoma,and 10 normal controls.266 immune-related differentially expressed genes were identified between esophageal adenocarcinoma tissues and normal controls;318 immune-related differentially expressed genes were identified between esophageal squamous cell cancer tissues and normal controls.Functional enrichment analysis revealed that immune-related differently expressed genes may be involved in several molecular functions and pathways,which are closely linked to esophageal cancer,such as cell chemotaxis,leukocyte migration,positive regulation of secretion,positive regulation of cytokine production Cytokine-cytokine receptor interaction,Chemokine signalling pathway,JAK-STAT signalling pathway,Natural killer cell-mediated cytotoxicity,and NF-kappa B signaling pathway.The Cox regression analysis,by R software,was used to identify immune-related genes that are significantly related to the overall survival of the two esophageal cancer subtype adenocarcinoma and squamous cell carcinoma,respectively.A risk scoring formula was constructed to assess the characteristics of immune-related genes with prognostic value in esophageal cancer.The patients’ overall survival rate with a high-risk score was worse than that of low-risk.Multivariate Cox regression analysis displayed that immune-related gene characteristics were an independent prognostic factor.In addition,this feature can serve as a marker to predict clinical outcomes.Based on the above models,drug sensitivity prediction found that there were differences in drug responses to multiple drugs in patients with high and low risk.Conclusion:Our study demonstrated that the different immune-related gene signatures and may serve as an independent prognostic model for esophageal adenocarcinoma and squamous cell carcinoma respectively.Basedontheimmune-related gene signature score,combined with GDSC database,the low-risk group of esophageal adenocarcinoma is more suitable for 25 kinds of drugs such as KIN001.135,GNF.2 and FTI.277,while the low-risk group of esophageal squamous cell carcinoma is more suitable for 6 kinds of drugs such as PLX4720,AG.014699 and QS11,so as to guide medication and promote individualized treatment.