Abstract
BACKGROUND: Sorafenib is the first-line treatment agent for advanced hepatocellular carcinoma (HCC), but it is effective in very few patients. Thus, this study was intended to identify gene signatures associated with sorafenib response in HCC and construct a prognostic risk model based on these gene signatures. METHODS: The gene expression level data of HCC were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. First, we evaluated the HCC sensitivity to sorafenib of the samples and investigated the correlation between HCC sensitivity to sorafenib and clinical prognosis. By Weighted Gene Co-Expression Network Analysis (WGCNA) algorithm, the sorafenib resistance associated genes were obtained. Combined with the sorafenib resistance related data set, genes significantly associated sorafenib responses were screened. Then prognostic signature genes were screened to construct a risk prognosis model. The correlation of risk grouping with immune cells was explored. RESULTS: A total of 399 significantly genes associated with sorafenib response were obtained, which were involved in metabolic and complement pathways. Finally, 5 signature genes (DNASE1L3, ACSL6, ACAN, BRSK1 and CD68) were identified, and a risk prognostic risk prediction model was constructed. The receiver operating characteristic (ROC) curves suggested that the model presented high predictive precision. Additionally, a nomogram was constructed based on pathologic stage and risk model status, which also had good prediction performance in survival prognosis. Moreover, there was a significant correlation between risk groups and immunity. CONCLUSIONS: Our study established a sorafenib response-related prognostic risk prediction model in HCC based on five signature genes (DNASE1L3, ACSL6, ACAN, BRSK1 and CD68), which had high predictive precision on sorafenib response.