Diagnostics of autoimmune hepatitis enabled by non-invasive clinical proteomics

Autoimmune hepatitis (AIH) can be challenging to distinguish from other chronic liver diseases such as metabolic dysfunction-associated steatotic liver disease (MASLD). We applied advanced proteomic analysis to paired liver and plasma samples from newly diagnosed AIH patients, MASLD patients, and healthy controls to uncover disease-specific pathways and biomarkers. We quantified 7632 liver and 556 plasma proteins, identifying significant differences in 2521 liver and 227 plasma proteins between AIH and healthy individuals. A machine learning model based on plasma proteomics accurately distinguished AIH from MASLD and healthy controls (AUC = 0.91), with validation in the UK Biobank confirming diagnostic value for several candidate biomarkers. This study highlights distinct metabolic and immune signatures in AIH and supports the utility of plasma proteomics in non-invasive disease classification.