Molecular subtypes based on cell death genes predict prognosis and immunotherapy effect in patients with hepatocellular carcinoma

Abstract Background Hepatocellular carcinoma (HCC) is the leading cause of cancer-related death in many regions of the world. Cell death plays an important role in the development of tumor, and there is no comprehensive analysis of the role of cell death genes in HCC. This study aims to identify different HCC molecular subtypes based on cell death genes, and compare the prognosis and immunotherapy effect of different subtypes. Methods Cell death genes associated with HCC prognosis were screened by univariate COX analysis from the cancer Genome Atlas (TCGA), International Cancer Genome Consortium (ICGC) and Gene Expression Database (GEO), respectively. Based on the expression of cell death genes, NMF consensus clustering was used to subtype HCC. The differences of tumor microenvironment score, cytolytic immune response, inflammation score, tumor mutation load, immune checkpoint expression level and TIDE score among diffrent subtypes were analyzed using R software. Results HCC was divided into two subtypes based on the expression values of eight cell death genes associated with prognosis of HCC in three data sets. The prognosis and immunotherapy response of C1 subtype were better. There were differences in tumor microenvironment score, cytolytic immune response, inflammation score, immune checkpoint expression level and TIDE score between the two HCC subtypes. Conclusions we classified different subtypes of HCC based on cell death genes, determined the relationship between different HCC subtypes and prognosis, studied the tumor microenvironment of different HCC subtypes and predicted their response to immune checkpoint inhibitors (ICIs) therapy, providing molecular basis for immunotargeted therapy.