Objective To construct a prognostic model for liver hepatocellular carcinoma(LIHC) based on immune-related genes. Methods   LIHC and normal tissue samples were downloaded from the UCSC Xena database and The Cancer Genome Atlas (TCGA) database. Differential analysis was performed on the gene data of LIHC samples and adjacent/normal samples. Enrichment analysis was conducted on differentially expressed genes. Kaplan-Meier survival analysis was performed on liver cancer samples from the TCGA cohort to obtain survival- and immune-related differentially expressed genes. LASSO Cox and multivariate Cox  regression analysis were used to identify hub genes and construct a gene risk prognostic model. Data from a high-throughput gene expression (GEO) database was obtained for external validation. The sensitivity of hub genes to common anticancer drugs was investigated using the CellMiner database. ResultsEnrichment analysis result indicated that differentially expressed genes were mainly associated with metabolic pathways. Through differential analysis and Kaplan-Meier survival analysis, 25 survival- and immune-related differentially expressed genes were obtained. Based on the result  of LASSO Cox and multivariate Cox regression analysis, five hub genes (FYN, CSF3R, HLA-G, FOS, BIRC5) were identified and a nomogram was constructed. The concordance index(C-index) value for the training cohort and validation cohort were 0.739 and 0.625, respectively. Based on the sensitivity of hub genes to anticancer drugs, 12 types of anticancer drugs were selected for subsequent experiments. Conclusion   This model can effectively predict the prognosis of LIHC patients and provide a new insights for immune therapy in LIHC.