Meta-analysis And Bioinformatics Analysis Of The Prognosis Of Esophageal Cancer Tumor Infiltrating Immune Cells

Background and AimsIn recent years,people have been looking for biomarker which can predicts prognosis of esophageal cancer patients.Many studies have reported the prognostic role of tumor infiltrating lymphocytes(TILs).The aims of our study are evaluating prognostic relevance of various subsets of tumor infiltrating immune cells and building optimum prognostic prediction model.MethodPubmed,Embase,Cochrane Library and Web of Science were searched and studies assessing prognostic role of different subsets of TILs were retrieved.Data were integrated via meta-analysis for various subsets of TILs at different sites.Pooled hazard ratios(HRs)and 95%confidence interval were calculated.RNA sequencing data of esophageal patients were downloaded form TCGA database.Through single sample gene set enrichment analysis based on signature gene sets of different immune-cell subsets constructed in other research,we calculated enrichment score for immune cell subsets of individual patients.Prognostic role of different immune cell subsets were evaluated.Prognosis related signature genes were found.The final prognostic prediction model was built based on ES of immune cell subsets,expression of prognostic signature genes and clinical data.Result29 studies were found which focused on the association between various subsets of TILs and prognosis of esophageal cancer patients.High overall TILs infiltration in both tumor stroma and nest was associated with better overall survival(OS)(HR = 0.54,95%CI 0.44-0.65,p = 0.00001).High CD8+ T-cell infiltration in tumor stroma(TS)or tumor nest(TN)or both TS and TN were associated with better overall survival(OS)(HR = 0.77,95%CI 0.66-0.89,p = 0.0005;HR = 0.67,95%CI 0.57-0.79,p = 0.00001;HR = 0.83,95%CI 0.75-0.92,p = 0.0006,respectively).For squamous cell carcinoma,activated B cell,activated CD8+ T cell,CD56 dim NK cell,eosinophil,effector memory CD8+ T cell,gamma delta T cell,macrophage,mast cell,NK cell and were associated with worse prognosis.For adenocarcinoma of esophageal cancer,central memory CD8+ T cell and follicular helper T cell were associated with the better prognosis.Combinational use of ES of immune cell subsets,expression of prognosis relevant signature genes and clinical data can help us build the optimum prognosis prediction model.ConclusionOverall TILs and CD8+ T cell are potential biomarker predicting prognosis of esophageal cancer patients.Various subsets of immune cells are related with the prognosis of esophageal cancer patients.Superior prognosis predicting model can be built if we take tumor infiltrating immune cells into account in addition to prognosis relevant genes.