Immune Subtypes And Functional Effects Of Hepatocellular Carcinoma Based On Multi-Omics Integrated Analysis

Background: The immunotherapy,as a promising treatment,has been investigated in different cancers including hepatocellular carcinoma(HCC).However,due to the tumor heterogeneity and the diversity of the immune microenvironment,not all HCC patients could achieve the desired therapeutic effect from the immunotherapeutic.Presently,we aimed to explore the detailed immune landscape in the immune microenvironment stratification of HCC.Methods: First,we obtained 949 HCC samples from 6 gene expression data through TCGA and GEO datasets and 1172 immune genes from the Immunology Database and Analysis Portal(Imm Port)database.The unsupervised clustering algorithms,i.e.KMeans and t-SNE,were performed to divide HCC samples into immune clusters based on the expression levels of these immune genes.According to the ESTIMATE algorithm,we evaluated the immune status of the immune clusters.Through combining the bulk and single-cell RNA-seq data,we investigate the ESTIMATE score,immune checkpoint levels,and the proportions of infiltrating immune cells among immune clusters.Based on the tumor exome data,we analyzed the tumor mutation burden and the mutation characteristics among clusters.Finally,with the LASSO-Cox regression algorithm,we constructed and assessed the immune risk module.Results: We successfully identified three immune clusters with increasing ESTIMATE scores and redefined them as Cluster A(high immune),B(intermediate immune),and C(low immune).It is indicated that the immune clusters have significant level differences and diversified distribution patterns in immune infiltrating cells and immune checkpoints.According to the immune cells and their gene markers obtained in the single-cell data analysis,the immune clusters also show the immune status consistent with the above analysis.And three clusters have significant differences in the proportions of immune infiltration cells and levels of immune checkpoints.In addition,somatic mutation analysis suggests the statistical differences in tumor mutation burden levels,as well as their abnormal mutation gene characteristics,among immune clusters.Notably,we found weakened immunocompetence discrepancies in immune clusters,especially in Cluster B,between mutation/no-mutation TP53 sample groups.The functional enrichment analysis revealed three mutually exclusive enrichment types of three clusters.Additionally,the immune risk modules,based on immune clusters,revealed the effective performance for the overall survival prediction of HCC.Conclusions: The stratification of tumor immune microenvironment,defined by immune genes,suggests the detailed immune landscape through the multi-omics analysis,which provides a new insight for the immunotherapy in HCC.