Aim: Recently, serum levels of anti-programmed cell death-1 (anti-PD-1) antibodies have been reported to be useful for the discrimination of type 1 autoimmune hepatitis (AIH) from drug-induced liver injury (DILI) and to be associated with clinical features of type 1 AIH. This multicenter study aimed to validate the usefulness of serum anti-PD-1 antibody as a serological marker for type 1 AIH.
Methods: Serum samples before the initiation of corticosteroid treatment were obtained from 71 type 1 AIH patients and 37 DILI patients. Serum levels of anti-PD-1 antibodies were measured by indirect enzyme-linked immunosorbent assay.
Results: Serum levels of anti-PD-1 antibodies were higher in type 1 AIH patients than in DILI patients (P < 0.001). The receiver-operator curve analysis showed that serum levels of anti-PD-1 antibodies were useful for the discrimination of type 1 AIH from DILI (area under the curve, 0.80). On the other hand, the multivariate Cox proportional hazard model showed that positivity for serum anti-PD-1 antibody, probable diagnosis based on the revised scoring system proposed by the International Autoimmune Hepatitis Group, and prothrombin activity of less than 60% were associated with the later normalization of serum transaminase levels. During the clinical course, the disease relapsed more frequently in patients positive for serum anti-PD-1 antibody (36% vs 11%).
Conclusion: This study suggests that serum anti-PD-1 antibody is useful for the diagnosis of type 1 AIH as an auxiliary diagnostic marker, and that serum levels of anti-PD-1 antibodies reflect clinical features of type 1 AIH.
Keywords: autoantibody; autoimmune hepatitis; drug-induced liver injury; programmed cell death-1; validation.
© 2014 The Japan Society of Hepatology.