Identification of fatty acid metabolism-related clusters and immune infiltration features in hepatocellular carcinoma.

Abstract

Hepatocellular Carcinoma (HCC) is a type of liver cancer which is characterized by inflammation-associated tumor. The unique characteristics of tumor immune microenvironment in HCC contribute to hepatocarcinogenesis. It was also clarified that aberrant fatty acid metabolism (FAM) might accelerate tumor growth and metastasis of HCC. In this study, we aimed to identify fatty acid metabolism-related clusters and establish a novel prognostic risk model in HCC. Gene expression and corresponding clinical data were searched from the Cancer Genome Atlas (TCGA) and the International Cancer Genome Consortium (ICGC) portal. From the TCGA database, by unsupervised clustering method, we determined three FAM clusters and two gene clusters with distinct clinicopathological and immune characteristics. Based on 79 prognostic genes identified from 190 differentially expressed genes (DEGs) among three FAM clusters, five prognostic DEGs (CCDC112, TRNP1, CFL1, CYB5D2, and SLC22A1) were determined to construct risk model by least absolute shrinkage and selection operator (LASSO) and multivariate cox regression analysis. Furthermore, the ICGC dataset was used to validate the model. In conclusion, the prognostic risk model constructed in this study exhibited excellent indicator performance of overall survival, clinical feature, and immune cell infiltration, which has the potential to be an effective biomarker for HCC immunotherapy.