Prediction of Prognosis in Patients with Hepatocellular Carcinoma Based on Molecular Subtypes of Immune Genes

Abstract

For those patients with hepatocellular carcinoma (HCC), it is really a heavy burden. Herein, the immune genes of HCC were analyzed in groups to determine prognostic biomarkers related to immune genes in HCC. The mRNA data, clinical data in TCGA-LIHC dataset, and immune gene in the ImmPort database were collected for the combining usage with $K$ -means concordance clustering to cluster HCC patients according to the immune gene matrix. Based on ssGSEA analysis result, HCC patients were sorted into high- and low-immune subtypes, and survival curve presented that patients in high-immune subtypes had a better prognosis. Subsequently, differential expression analysis was performed to obtain immune-related differentially expressed genes (IRGs). Cox and lasso analyses were performed for obtaining five optimal immune genes related to prognosis, and a risk assessment model was then established. Patient samples in the training and validation sets were, respectively, divided into high- and low-risk groups. $K$ - $M$ survival curves presented a better prognosis of patients in the low-risk group than in the high-risk group. The ROC curve indicated that this model was finely used for the prediction of prognosis. In addition, immune infiltration assessment revealed that NR0B1 and FGF9 had potential to impact the tumor immune microenvironment. Finally, using qRT-PCR and transwell assays, it was demonstrated that the macrophage chemotaxis was enhanced when NR0B1 and FGF9 were highly expressed in HCC cells. In general, a 5-gene prognostic risk assessment model was constructed based on immune genes and bioinformatics analysis methods, which provides some reference for the prognosis of HCC as well as immunotherapy.