Comprehensive Analysis of Sorafenib Druggable Targets on Differential Gene Expression and ceRNA Network in Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is the predominant pathological type of liver cancer. Several therapeutic treatments, including sorafenib and regorafenib, have only modestly improved survival in patients with HCC. The aim of this study was to investigate the expression profiles and the regulation of competitive endogenous RNAs (ceRNAs) of the sorafenib-related target genes in HCC. Based on clinical information and expression profiles of HCC clinical samples from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases, shared differentially expressed genes (DEGs) were analyzed and identified. Sorafenib-associated DEGs (SADs) were obtained by intersecting the DEGs with the sorafenib target genes from SuperTarget database. The expression patterns of SADs were verified in the Oncomine database. The biological functions of the SADs were annotated by gene set enrichment analysis (GSEA). In addition, a ceRNA network associated with SADs was constructed. Long non-coding RNAs (lncRNAs) in network that were significantly associated with overall survival were identified as prognosis of patients by Cox regression analysis. Finally, the expression levels of prognostic genes in HCC tissues and cell lines were verified using qRT-PCR. Gene expression differential analysis yielded a total of 146 common DEGs were obtained, including 21 upregulated and 125 downregulated DEGs. Among them, ten SADs were detected to be differentially expressed between tumor and normal tissues, including AXL, CYP2C19, CYP2C8, CYP2C9, CYP3A4, FGFR2, GMNN, PDGFRA, and TTK. GSEA analysis grouped them into three categories by function. The first category (CYP2C19, CYP2C8, CYP2C9 and CYP3A4) and second category (GMNN, TTK and EGER2) had the opposite roles in the enriched terms and pathways, while the third class (AXL and PDGFRA) has enrichment terms and pathways that intersect with those of the first and second categories. A ceRNA network associated with SADs was also constructed including 49 lncRNAs, 14 miRNAs, and 8 mRNAs. Three of these lncRNAs, SNHG7, GAS5 and HCP5, were found upregulated in HCC tissues and to be independent predictors in HCC patients. Significant correlations were found in expression between the prognostic lncRNAs and SADs. Ten SADs were systematically identified using expression data from HCC and normal tissues from TCGA and GEO datasets. GSEA analysis provided us with insight into the function of SADs. In the future, we will continue to explore the mechanisms of coordinated regulation of SADs-related prognostic lncRNAs and SADs at the ceRNA axis level and their potential functions in the development of HCC.