Identification Of A Novel Tumor Mutation Burden And Major Histocompatibility Complex-Related Score To Predict Immunotherapy Response And Overall Survival In Gastric Cancer

Background: Gastric cancer is one of the most common malignancies worldwide.The advent of immunotherapy has greatly extended the overall survival(OS)of gastric cancer patients,but the overall response rate is less than 15%.The U.S.Food and Drug Administration(FDA)recommends immunotherapy for patients with solid tumors with a high tumor mutational burden(TMB)(TMB ≥ 10 mutations per megabase),but some studies have shown that patients with high TMB gastric cancer do not benefit significantly from immunotherapy.Recent studies have found that some patients with high TMB have loss of heterozygosity(LOH)at the human leukocyte antigen locus,and that HLA LOH reduces the recognition of new antigens by cytotoxic T lymphocytes(CTL).Thus,we hypothesize that HLA LOH may be a potential mechanism for the poor immunotherapeutic response in gastric cancer patients with high TMB.Therefore,in this study we will combine TMB and Major histocompatibility complex(MHC)to establish a score TM-Score for predicting immunotherapy efficacy and OS in gastric cancer patients,in order to be able to provide a method for screening the population with superior immunotherapy for gastric cancer.Materials and methods: In this study,the TMB and MHC-I and MHC-II scores of gastric cancer patients were first obtained by sequencing and mutation public database of the cancer genome atlas(TCGA).The R package "maxstat" was then used to calculate the cut-off values,and three sets of differentially expressed genes(DEGs)were obtained by comparing the high TMB group with the low TMB group,the high MHC-I group with the low MHC-I group,and the high MHC-II group with the low MHC-II group.To explore the potential mechanism of action of TM-Score,we compared the mutated genes,HLA gene expression,immunity and stromal cell infiltration in the low TM-Score group with the high TM-Score group(calculated by three methods: TIMER,CIBERSORT and x CELL),immune checkpoint gene expression and gene pathway enrichment differences.The ability of TM-Score to predict the efficacy of immunotherapy was first assessed by the Tumor immune dysfunction and exclusion(TIDE)score and the Immunophenotype Score(IPS)and validated in a clinical cohort using immunotherapy(the Kim cohort)for validation.In exploring the role of the six genes comprising the TM-Score in influencing tumor immune response,we analyzed the correlation of expression of the six genes with immune cell infiltration,ESTIMATE score,TIDE score,and immune-related marker score.The ability of TM-Score to predict OS in gastric cancer patients was first assessed by KM curve and subject operating characteristic curve(ROC)and validated in external validation sets(GSE15459 and GSE62254).Subsequently,seven clinical indicators(age,sex,T,N,M,and tumor stage and grade)of gastric cancer patients were evaluated in the Surveillance,Epidemiology,and End Results(SEER)database of the National Cancer Institute,Finally,the SEER CP-Score and TM-Score were combined in TCGA to improve the prediction of OS in gastric cancer patients.OS efficacy.Results: We identified TMB,MHC-I and MHC-II as protective factors in gastric cancer patients,and a total of 2393 differential genes were obtained.After further downscaling we finally constructed a TM-Score consisting of 6 genes and divided the patients into 2 groups(high TM-Score group vs.low TM-Score group)by median values.Gene mutations,TMB and HLA gene expression were higher in the low TM-Score group than in the high TM-Score group.In terms of TM-Score predicting immunotherapy efficacy,we found that the low TM-Score group had a lower TIDE score and a higher IPS score in TCGA-STAD,implying that patients in the low TM-Score group had better efficacy with immunotherapy.In the Kim cohort using immunotherapy,patients in the low TM-Score group had better immunotherapy outcomes,with better predictive accuracy than TMB,MSI and EBV,and the highest accuracy when combined with PDL1.Meanwhile,we found that gastric cancer patients with low TM-Score had higher immune checkpoint gene expression and higher infiltration of CD8+ T cells,CD4+ helper T cells and M1 macrophages,indicating that TM-Score was significantly associated with tumor immunogenicity and tumor immune microenvironment.In exploring the regulation of tumor immune response by the six genes comprising the TM-Score,we found that AKAP5 positively correlated with antitumor immune response and individually predicted immunotherapy efficacy and OS in gastric cancer patients.In TCGA-STAD,OS was significantly better in the low TM-Score group than in the high TM-Score group.In the validation datasets GSE15459 and GSE62254,the OS of the low TM-Score group was also significantly better than that of the high TM-Score group.To further optimize the ability of TM-Score to predict OS in gastric cancer patients,we established the SEER CP-Score,and by combining it with TM-Score,we more accurately predicted OS in gastric cancer patients.Conclusion: TM-Score,as an optimization of TMB,is a more precise biomarker for predicting the immunotherapeutic efficacy of the gastric cancer patients,and AKAP5 is a promising therapeutic target for gastric cancer.