157P - Identification and validation of a prognostic 4 genes signature for hepatocellular carcinoma: Integrated ceRNA network analysis

Abstract

BackgroundHepatocellular carcinoma (HCC) is one of the most aggressive malignant tumors, with a poor long-term prognosis worldwide. The functional deregulations of global transcriptome were associated with the genesis and development of HCC. However, reliable molecular signatures predicting overall survival (OS) lacks of systematic research and validation. MethodsA total of 519 postoperative HCC patients were included. We built an interactive and visual competing endogenous RNA (ceRNA) network from The Cancer Genome Atlas (TCGA) database. The prognostic signature was established with the least absolute shrinkage and selection operator (LASSO) algorithm. Multivariate Cox regression analysis and subgroup analysis was used to screen for independent prognostic factors. A time-dependent ROC curve analysis was performed to compare predictive value of the prognostic signature. The robustness of the prognostic signature was validated in validation cohorts. ResultsThere were 39 differentially expressed mRNAs (DEmRNAs), 83 differentially expressed lncRNAs and 20 differentially expressed miRNAs involved in the ceRNA network. Twenty DEmRNAs were found to be significantly associated with OS. We identified a 4-gene signature (PBK, CBX2, CLSPN and CPEB3) using LASSO regression in the training set. Patients in the high-score group exhibited worse survival than those in the low-score group (HR=2.444, P=0.0004), and median OS was significantly shorter in the high-score group than in the low-score group (1005 days versus 2456 days). The 4-gene signature was an independent prognostic factor in multivariate Cox regression and subgroup analysis, particularly for patients with serum AFP ≥ 20ng/ml. The results were validated in internal validation set (P=0.0057) and two external validation cohorts (HR=1.505 and 2.626). The signature (AUCs of one, two, three years were 0.716, 0.726, 0.714, respectively) showed high prognostic accuracy. ConclusionsWe constructed a novel lncRNA-miRNA-mRNA ceRNA network for HCC based on genome-wide analysis. Then we identified a 4-gene signature as a new candidate therapeutic decision marker that yields great promise in the prediction of HCC OS. Legal entity responsible for the studySun Yat-Sen University. FundingNational Natural Science Foundation of China. DisclosureAll authors have declared no conflicts of interest.