Conserved hepatic RNA signatures across multiple cohorts reveal novel mechanistic clues for advanced fibrosis in human MASLD.

Fibrosis development in patients with metabolic dysfunction-associated steatotic liver disease (MASLD) is a key indicator of disease progression and clinical outcome. Bulk and single-cell transcriptomics on human tissue have advanced understanding of fibrosis progression, but interstudy heterogeneity and limited sample size hinder the identification of consistent and targetable fibrogenic mechanisms. To identify conserved fibrogenic mechanisms, we performed a comprehensive meta-analysis of hepatic transcriptomic data with fibrosis stage characterized from ~1000 patients with MASLD. Our meta-analysis revealed 846 differentially regulated genes associated with fibrosis progression (F3-4 vs. F0-1). Pathway analysis showed that these genes are involved in matrix organization (THBS2, ADAMTSL2), inflammation (CXCL6, CCL19), solute transport (SLC13A5, SLC16A10), and metabolism (AADAT, GRAMD1B). scRNA-seq-based deconvolution revealed increased proportions of immune (CD4+ T cells), endothelial (HA endo cells), and mesenchymal (myofibroblasts) cell types, and loss of LSECs in patients with advanced fibrosis in the meta-analysis. Finally, analysis of ligand-receptor pairs identified putative cell-matrix interactions (MMP7-CDH6), cell signaling (PDGFD-PDGFRA), and immune cell interactions (ANXA1-FPR1) associated with fibrosis conserved across multiple datasets and enriched in patients with fibrosis. We identified conserved transcriptomic changes and gene networks in patients with advanced fibrosis, as well as putative ligand-receptor interactions that facilitate cell-cell interactions that drive fibrosis. These findings will inform mechanistic studies and the development of anti-fibrotic therapies.