Autoimmune hepatitis displays distinctively high multi-antennary sialylation on plasma N-glycans compared to other liver diseases.

Background: Changes in plasma protein glycosylation are known to functionally affect proteins and to associate with liver diseases, including cirrhosis and hepatocellular carcinoma. Autoimmune hepatitis (AIH) is a liver disease characterized by liver inflammation and raised serum levels of IgG, and is difficult to distinguish from other liver diseases. The aim of this study was to examine plasma and IgG-specific N-glycosylation in AIH and compare it with healthy controls and other liver diseases.

Methods: In this cross-sectional cohort study, total plasma N-glycosylation and IgG Fc glycosylation analysis was performed by mass spectrometry for 66 AIH patients, 60 age- and sex-matched healthy controls, 31 primary biliary cholangitis patients, 10 primary sclerosing cholangitis patients, 30 non-alcoholic fatty liver disease patients and 74 patients with viral or alcoholic hepatitis. A total of 121 glycans were quantified per individual. Associations between glycosylation traits and AIH were investigated as compared to healthy controls and other liver diseases.

Results: Glycan traits bisection (OR: 3.78 [1.88-9.35], p-value: 5.88 × 10), tetraantennary sialylation per galactose (A4GS) (OR: 2.88 [1.75-5.16], p-value: 1.63 × 10), IgG1 galactosylation (OR: 0.35 [0.2-0.58], p-value: 3.47 × 10) and hybrid type glycans (OR: 2.73 [1.67-4.89], p-value: 2.31 × 10) were found as discriminators between AIH and healthy controls. High A4GS differentiated AIH from other liver diseases, while bisection associated with cirrhosis severity.

Conclusions: Compared to other liver diseases, AIH shows distinctively high A4GS levels in plasma, with potential implications on glycoprotein function and clearance. Plasma-derived glycosylation has potential to be used as a diagnostic marker for AIH in the future. This may alleviate the need for a liver biopsy at diagnosis. Glycosidic changes should be investigated further in longitudinal studies and may be used for diagnostic and monitoring purposes in the future.