An Efferocytosis-Associated Gene Signature for Identifying At-Risk MASH: Transcriptomic and Exploratory Plasma Biomarker Assessment

Background/Objectives: At-risk metabolic dysfunction-associated steatohepatitis (MASH) is associated with increased risks of cirrhosis, hepatocellular carcinoma, and liver-related mortality. Because impaired efferocytosis contributes to persistent hepatic inflammation and fibrotic remodeling in MASH, we investigated whether efferocytosis-associated molecular signatures could identify at-risk MASH. Methods: Bulk RNA-sequencing datasets from Gene Expression Omnibus (GSE135251 and GSE174478) were analyzed to identify differentially expressed efferocytosis-related genes and characterize associated pathways and immune infiltration patterns. Machine learning-based feature selection was used to identify hub genes, which were incorporated into a transcriptomic nomogram. Experimental validation involved reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and Western blotting of liver tissues from a Western diet-induced murine metabolic dysfunction-associated steatotic liver disease (MASLD) model. Plasma proteomic data were analyzed to explore the discriminatory performance of hub gene products. Results: A total of 17 efferocytosis-related genes (ERGs) associated with at-risk MASH were identified and enriched in pathways related to efferocytosis, inflammation, and immune regulation. Five hub genes, CD24, CHI3L1, TREM2, PTGS2, and LGR6, were shared by all three feature-selection approaches and significantly upregulated in at-risk MASH. A transcriptomic nomogram yielded area under the curve (AUC) values of 0.866 and 0.805 in the training and external cohorts, respectively. In the murine MASLD model, all five hub genes showed increased mRNA and protein expression in advanced disease. Exploratory plasma proteomic analysis identified elevated circulating TREM2 and CHI3L1 levels in at-risk MASH, and a simplified plasma-based model achieved an AUC of 0.736. Conclusions: This study identified a five-gene efferocytosis-associated signature and developed models for identifying at-risk MASH, suggesting that these genes warrant further evaluation as candidate biomarkers.