Multiplexed digital spatial protein profiling reveals a unique protein signature for advanced liver fibrosis

Abstract Background and Aims: Intrahepatic mononuclear phagocytes (MPs) are critical for the initiation and progression of liver fibrosis. In this study, using multiplexed digital spatial protein profiling, we aimed to derive a unique protein signature predicting advanced liver fibrosis. Methods Snap-frozen liver tissues from various chronic liver diseases were subjected to spatially defined protein-based multiplexed profiling (Nanostring GeoMX™). Single-cell RNA sequencing analysis was performed using Gene Expression Omnibus (GEO) datasets from normal and cirrhotic livers. Results Sixty-four portal regions of interest (ROIs) were selected for the spatial profiling. Combined analysis of single-cell RNA sequencing data from GEO datasets (GSE136103) and spatially-defined, protein-based multiplexed profiling revealed that most proteins upregulated in F0–F2 livers in portal CD68+ cells were specifically marked in tissue monocytes whereas proteins upregulated in F3 and F4 livers were marked in SAMacs and tissue monocytes. Internal validation using mRNA expression data with the same cohort tissues demonstrated that mRNA levels TREM2, CD9, and CD68 are significantly higher in livers with advanced fibrosis. Using the results from the CD68+ area, a highly sensitive and specific immune-related protein signature (CD68, HLA-DR, OX40L, phospho-c-RAF, STING, and TIM3) was developed to predict advanced (F3 and F4) fibrosis. Conclusions In patients with advanced liver fibrosis, portal MPs consist heterogeneous populations composed of SAMacs, Kupffer cells, and tissue monocytes. Using digital spatial protein profiling, the protein signature with high accuracy in predicting advanced fibrosis was developed.