Abstract
Introduction: Immune cells play a dual role inhepatocellular carcinoma (HCC) – it can both restrain and promote tumor growth, which is the significant component of the tumour immune microenvironment (TIME). This study aimed to develop a immune-related signature model to predict the prognosis and TIME of patients with HCC. Methods: Data for the TCGA-LIHC and GSE14520 cohorts were downloaded from The Cancer Genome Atlas and the Gene Expression Omnibus databases. Single-cell RNA-sequencing data for HCC samples were retrieved from the GSE140228 cohort. The Least Absolute Shrinkage and Selection Operator algorithm was employed to develop a Immune-related signature (IRSig). The predictive value of the IRSig was determined using Kaplan-Meier, Cox regression and Receiver Operating Characteristic curves. Gene set enrichment analysis (GSEA), Gene Set Variation Analysis (GSVA) and gene ontology (GO) analysis were performed to explore the functional enrichment of the IRSig. Finally, the TIMER platform, single sample Gene Set Enrichment Analysis and the Estimation of STromal and Immune cells in MAlignant Tumour tissues using Expression data algorithms were performed to determine the TIME landscape. Results: The immune-related signature demonstrated its superior ability to predict the clinical outcome of patients with HCC. TMB, immune score, stromal score, and ESTIMATE score were higher in the high-risk group compared to the low-risk group. Additionally, most immune checkpoints, including CTLA4, PD1 and PD-L1, were expressed at significantly higher levels in high-risk group. Conclusions: Our study established an immune-related signature based on single-cell RNA-sequencing and machine learning for optimal prognosis prediction and treatment decisions in hepatocellular carcinoma, and verified by TCGA and GEO databases. Besides, we found immune-related cells and pathways were significant differences in high- and low-risk group, which might be helpful for illustrating the application of immunotherapy for HCC patients.