Construction and experimental verification of user-friendly molecular subtypes mediated by immune-associated genes in hepatocellular carcinoma

Abstract

Hepatocellular carcinoma (HCC) accounts for third most cancer death globally, and its prognosis continues to be poor even with many novel therapeutic approaches emerging. The advent of immunotherapy seems to offer new hope, but low response rates are an unresolved problem. To gain further knoeledge of the effect of immune-related genes in HCC, we examined the connection between immune-related genes and the immune microenvironment in HCC through the HCC transcriptome dataset. The study also aimed to construct and experimentally validate user-friendly molecular subtypes mediated by immune-related genes in HCC. The immune cell infiltration patterns differs in HCC adjacent non-disease tissues and cancerous tissues. Patients with HCC could be classified into 2 subtypes: subtype A and subtype B. Specifically, subtype A shows characteristics of a hot tumor, in which the infiltration of cells exhibiting antigens and the expression of other crucial factors associated with immune function are higher than in a cold tumor. In addition, we identified Hub genes for the different subtypes and constructed a prognostic prediction model based on six genes (KLRB1, KLF2, S100A9, MSC, ANXA5, and IMPDH1). Further experimental analysis of HCC samples exhibited that the expression levels of KLF2 and ANXA5 were associated with immune cell infiltration and expression of PD-L1 in cancer tissues. Our work suggests that the expression of immune-related genes has crucial effect on the tumor microenvironment and prognosis of HCC patients and may be associated with immunotherapeutic response, which provides new clues for the widespread and effective application of immunotherapy in HCC.