Novel Immune-risk Score Of Gastric Cancer:A Molecular Prediction Model Combining The Value Of Immune-risk Status And Chemosensitivity

Objective: Gastric cancer(GC)is still one of the most common and deadly malignancies in the world.Not all patients could benefit from chemotherapy or chemoradiotherapy due to tumor heterogeneity.Therefore,identifying different subgroups of patients is important trend for obtaining more effective responses.However,few molecular classifications associated with chemosensitivity are based on immune-risk status.In this study,we obtained some key immune-related genes.The aim of this study is to find a chemosensitivity prediction model based on immune risk state.Materials and methods: In this study,we selected we selected GSE 26253 as a development cohort and a combination of The Cancer Genome Atlas(TCGA)and GSE62254 to form a validation cohort.The individual prognostic signature was built based on immune-related genes(IRGs),which were obtained from the ImmPort database.The univariate Cox analysis and multivariate Cox analysis were performed to further select IRGs as the candidates of the individual risk score.Using these genes,we constructed a molecular model related to immune-risk status and calculated an individual immune-risk score.The statistical analyses in this study were all conducted in R.The survival differences of overall survival(OS)and disease-free survival(DFS)were all calculated by the log-rank test.Univariate Cox proportional hazards regression was used to estimate hazard ratios(HRs)between the low immune risk and high immune risk scores within different clinicopathological characteristic subgroups.We compared immune-related infiltration information among the different immune risk groups using the Wilcoxon rank sum test.To search the guiding value of the risk score in individualized clinical decision making,we further conducted log-rank test among the different immune-risk groups.Results: A total of six refined IRGs were selected as the prognostic factors of the prognostic model,including BRD8,CCL25,CMTM3,FPR1,GDF10 and LEPR.An individual risk score was calculated based on a combination of coefficients and expression status of these refined IRGs.The score showed great efficiency and stability in predicting prognosis.The patients could be divided into different risk groups based on the immune-related score.For patients in the low-risk group,both postoperative chemoradiotherapy and chemotherapy could significantly improve prognosis(P<0.001 and P=0.006,respectively)and chemoradiotherapy was significant superior than simple chemotherapy(P=0.027).For patients with an intermediate-risk score,postoperative chemoradiotherapy showed a statistically significant survival advantage over no anticancer treatment(P=0.002),while chemotherapy did not.Compared with no adjuvant treatment,neither postoperative chemoradiotherapy nor chemotherapy made no difference for patients in the high-risk group.Conclusion: Combining the value of immune-risk status and chemosensitivity,the immune-risk score not only could offer us prognostic evaluation and adjuvant treatment guidance,but also improve our understanding about the binding point between chemotherapy or chemoradiotherapy and the immune system,which may be helpful for further expanding the application of immunotherapy.