Establishment of an immune microenvironment-based prognostic predictive model for gastric cancer.

ABSTRACT

AIMS: The prognoses of patients with gastric cancer(GC) vary in different stages, which is mainly due to the great differences in tumor and tumor microenvironment. This study is aimed to explore the specific differences. MAIN METHODS: Based on RNAseq-based expression data from The Cancer Genome Atlas database and GSE15459 and the latest biological process genelist, stage-related biological processes in gastric cancer were screened out. GSVA, LASSO-COX, univariate and multivariate Cox regression analysis, Kaplan-Meier survival analysis, and pearson correlation analysis were performed for prediction model construction, verification and functional annotation. KEY FINDINGS: The immune system process was enriched at advanced stages of gastric cancer. The tumor immune microenvironment-based prognostic risk score could be used to predict the overall survival and disease-free survival of patients with gastric cancer. The prognostic risk score was significantly associated with gastric cancer subtypes, inflammatory factors, and immune processes and a higher risk score indicated stronger tumor immunosuppression. SIGNIFICANCE: We found immune system processes were significantly elevated in advanced gastric cancer and established an immune-based prognostic predictive risk model for gastric cancer, which could reflect the degree of tumor immunosuppression and might be beneficial for clinical decision-making.