Evaluation of tumoral glypican 3 mRNA level as a diagnostic and prognostic biomarker for hepatitis-b virus-associated hepatocellular carcinoma by an integrative transcriptomic meta-analysis and bioinformatics

Abstract

Aim: The aim of this study is to evaluate the potential of GPC3 mRNA level as a diagnostic and prognostic biomarker for HBV-associated HCC. Materials and Methods: GPC3 mRNA expression in HBV-associated HCC tumor tissues compared to matched adjacent tissues was evaluated by integrative transcriptomic meta-analysis. The results were validated in a different patient cohort and the possible associations between GPC3 mRNA level and the clinical variables were evaluated. Results: Transcriptomic data of HBV-associated HCC tissues (n=61) and matched adjacent tissues (n=61) from four datasets (GSE19665;GSE84402;GSE121248;GSE55092) were included in the meta-analysis. GPC3 mRNA level was found to be higher in tumors than adjacent tissues (fold change=12.88; p= 0;FDR=0). The result was validated in GSE14520, (HBV-associated HCC(n)=203; matched adjacent tissue(n)=203), (log-fold-change= 4.82; adj.p=1.43E-79). It was found that GPC3 mRNA level could distinguish HCC from adjacent tissues with high specificity and sensitivity (AUC=0.9108;95%CI=0.08792-0.9424;p