Early diagnosis and prognosis of hepatocellular carcinoma based on a ceRNA array

Objectives Hepatocellular carcinoma (HCC) is one of the most common causes of cancer-related deaths, due to high morbidity, a low early diagnosis rate, and poor prognosis. It is essential to explore competitive endogenous RNA (ceRNA) arrays for early diagnosis and prognosis of HCC. Methods The original gene expression profiles of differentially expressed lncRNA, miRNA and mRNA in HCC were downloaded from TCGA database. Differentially expressed lncRNA-miRNA and miRNA-mRNA interaction pairs were extracted from miRcode and starBase, a ceRNA network was constructed, and GO annotation and KEGG pathway analyses were performed. Cox regression and Kaplan-Meier survival analysis screening identified core genes in the network associated with HCC survival, centering on miRNA, which were screened using ceRNA arrays. qRT-PCR was used to detect the expression level of key genes in clinical samples. Dual luciferase reporter gene assays were used to verify the target binding relationship among lncRNA, mRNA, and miRNA. ROC curves were used to analyze the diagnostic efficacy of the ceRNA array. Results A total of 8 lncRNAs, 5 miRNAs, and 21 mRNAs were used to construct a ceRNA network. Functional enrichment analysis showed that mRNAs in the ceRNA network were mainly enriched in 14 signaling pathways, especially microRNAs in cancer. Survival analysis showed that lncRNA FOXD2-AS1 and miRNA miR-9-5p were related to the prognosis of HCC, and the targeted binding relationships between mRNAs. STMN1, COL15A1, and CCNE2 and miR-9-5p from the TargetScan, starBase, miRDB, and PicTar databases were reliable. qRT-PCR showed that expression levels of FOXD2-AS1, miR-9-5p, STMN1, COL15A1, and CCNE2 were upregulated in HCC tissues. Dual luciferase reporter assays showed that FOXD2-AS1 and STMN1 had a targeted binding relationship with miR-9-5p, but not with COL15A1 or CCNE2. The area under the curve of the candidate ceRNA array (FOXD2-AS1/miR-9-5p/STMN1/COL15A1/CCNE2) was higher than that of each member and ceRNA combination (FOXD2-AS1/miR-9-5p/STMN1). Conclusions The candidate ceRNA array formed by FOXD2-AS1/miR-9-5p/STMN1/COL15A1/CCNE2 could be a biomarker for early diagnosis and prognosis of HCC.