Immunotherapy as a strategy to deal with cancer is increasingly being used clinically, especially in hepatocellular carcinoma (HCC). We aim to create an immunotherapy-related signature that can play a role in predicting HCC patients’ survival and therapeutic outcomes. Immunotherapy-related genes were discovered first. Clinical information and gene expression data were extracted from GSE140901. By a series of bioinformatics methods to analyze, overlapping genes were used to build an immunotherapy-related signature that could contribute to predict both the prognosis of people with hepatocellular carcinoma and responder to immune checkpoint blockade therapy of them in TCGA database. Differences of the two groups in immune cell subpopulations were then compared. Furthermore, A nomogram was constructed, based on the immunotherapy-related signature and clinicopathological features, and proved to be highly predictive. Finally, immunohistochemistry assays were performed in HCC tissue and normal tissue adjacent tumors to verify the differences of the four genes expression. As a result of this study, a prognostic protein profile associated with immunotherapy had been created, which could be applied to predict patients' response to immunotherapy and may provide a new perspective as clinicians focus on non-apoptotic treatment for patients with HCC.