A Novel Immune‑Related Gene-Based Model: Prognosis and Immunotherapy Response in Hepatocellular Carcinoma

Abstract Background: Liver hepatocellular carcinoma (LIHC) remains a significant global health challenge. Immunotherapy for LIHC is a very promising treatment modality. However, some patients with LIHC have a poor response to tumor immunotherapy. Thus, it is imperative to fully understand liver cancer immune subtypes, by which individualized and optimized immunotherapy would be developed. Methods: We downloaded transcriptome and clinical data of TCGA-LIHC from the Cancer Genome Atlas (TCGA) databases, GSE52018 and GSE72467 from the Gene Expression Omnibus (GEO). The immune cell infiltration of the tumor microenvironment was calculated. Based on the results of CIBERSORT, the LIHC patients were clustered into three subtypes. To establish the prognostic model, we employed LASSO-COX regression analysis. The differences in immune cell infiltration, tumor stemness, somatic mutations, and sensitivity to the immune checkpoint inhibitors in patients with different risk groups were systematically assessed.Results: According to the results of CIBERSORT, LIHC samples were divided into three individual subtypes: Cluster-A, B, and C. These three subtypes have distinct characteristics with regard to immune cell infiltration, prognosis, and response to immune checkpoint inhibitors (ICIs) and chemotherapy drugs. Cluster-B belongs to hot tumors, and Cluster-A and Cluster-C belong to cold tumors. We trained a prognostic model based on differential genes between cold and hot tumors. We then attempted to explore the differences among different risk groups in the immune microenvironment, oncogenic biology, genetic variation, and prognosis. The characteristics of immune activation, enhanced immunotherapy response, and prolonged survival was observed in the low-risk group. Finally, we combined Riskscore and other clinicopathological features to build a nomogram to predict the 1-, 3-, and 5-year survival rates of LIHC patients.Conclusion: In this study, an immune‑related gene-based model of predicting the prognosis and immunotherapy response in hepatocellular carcinoma was established.