Elevated Polyreactive Immunoglobulin G in Immune-Mediated Liver Injuries With the Need for Immunosuppressive Therapy.

Background & Aims The distinction of drug-induced liver injury (DILI), drug-induced autoimmune-like hepatitis (DI-ALH), and autoimmune hepatitis (AIH) can be challenging due to overlapping clinical characteristics. Recently, polyreactive immunoglobulin G (pIgG) was identified as a novel biomarker in AIH. This retrospective study aimed to evaluate the diagnostic accuracy of pIgG to distinguish between AIH, DI-ALH, and DILI and thus identify patients in need of immunosuppression. Methods Samples from 120 patients (AIH = 81, DI-ALH = 16, DILI = 23) were compared to a control group (non-AIH-non-DILI-liver disease = 596 and healthy controls = 190). Results No patient in the DILI-group but 98% in the AIH- and 94% in the DI-ALH-group received immunosuppressive treatment. PIgG levels were significantly higher in the AIH-group 1.9 normalised arbitrary units (nAU) compared to DILI (1.1 nAU, p < 0.001), non-AIH-non-DILI-LD (1.0 nAU, p < 0.001) and healthy controls (0.27 nAU, p < 0.001). PIgG levels for DI-ALH (1.7nAU) were significantly higher compared to DILI (p = 0.044) and non-AIH-non-DILI-LD and healthy controls (both p < 0.001). Highest AUC was seen for pIgG (0.818) compared to conventional autoantibodies. The overall accuracy of pIgG to distinguish AIH from DILI (74%) and liver injuries with and without the need for immunosuppression (73%) was like that of ANA (71%/73%) and SMA (74%/69%) at cut-offs of ≥ 1/40. PIgG was positive in up to 79% of patients with AIH that were negative for a conventional autoantibody and was positive in 90% of DI-ALH cases compared to 25% in DILI that were caused by the same drugs. Conclusions PIgG may complement current serologic tests to identify patients with liver injury in need of immunosuppressive treatment.