Network-based meta-analysis of gene expression reveals novel prognostic biomarkers for the progression of hepatocellular carcinoma from non-alcoholic fatty liver disease

Abstract

Listen

Translate article

Liver steatosis, also known as non-alcoholic fatty liver disease (NAFLD), is a condition marked by the buildup of fat in the liver. It is frequently linked to obesity, diabetes, and other risk factors such as hypertension, dyslipidemia, and a sedentary lifestyle. It has the potential to progress to non-alcoholic steatohepatitis (NASH), a condition characterized by liver inflammation and fibrosis. Without intervention, NASH can progress, eventually resulting in cirrhosis and, ultimately, hepatocellular carcinoma (HCC). Gaining a greater understanding of the molecular pathways that drive the progression of the disease could facilitate the development of more effective prognostic and monitoring tools. This study utilized a network-based methodology to do a meta-analysis on a large set of gene expression data from three studies, following the guidelines outlined by PRISMA. A study network was created, and differential gene expression (DGE) analysis was conducted to compare disease states. The common differentially expressed genes (DEGs) were identified, and mutual information network analysis unveiled associations among genes such as COL1A1, COL1A2, C8B, and AAMP across several stages. Further, GO and KEGG analyses identified 23 genes, and 21 pathways linked to fatty acid metabolism, inflammation, insulin signaling, cell cycle regulation, growth, apoptosis, and angiogenesis. Furthermore, the Gene Set Enrichment Analysis (GSEA) revealed that there were enriched transcriptional events in pathways such as Nucleotide Excision Repair, Type 2 Diabetes Mellitus, Cell Cycle, and Apoptosis. Nine prognostic genes, namely PGM2L1, ADA, INF2, COL1A1, RPL18A, SLC25A6, SLC39A7, TXN, and ALDH1A1, were identified using the Kaplan-Meier technique in survival analysis. The survival prediction probability was evaluated using regularization regression approaches- LASSO, Ridge, and Elastic-Net. The analysis of the tissue-specific expression data revealed notable expression of SLC39A7, SLC25A6, ALDH1A1, and INF2 in liver hepatocytes. The mutational data indicates alterations in the COL1A1, SLC39A1, INF2, RPL18A, and SLC25A6 genes in cases of hepatocellular carcinoma.