Prognostic implications of antibodies to Ro/SSA and soluble liver antigen in type 1 autoimmune hepatitis

Abstract

Antibodies to soluble liver antigen are frequently co-expressed with antibodies to ribonucleoprotein/Sjögren's syndrome A (Ro/SSA) in autoimmune hepatitis. Our goals were to evaluate the prognostic implications of antibodies to Ro/SSA in type 1 autoimmune hepatitis and to determine their independence from antibodies to soluble liver antigen. Three hundred and seventy-six serum samples from 170 patients were tested by enzyme immunoassays. Sixty-five patients (38%) had antibodies to Ro52; 11 patients (6%) had antibodies to Ro60; and 27 patients had antibodies to soluble liver antigen (16%). Twenty-six patients with antibodies to Ro52 had antibodies to soluble liver antigen (40%), and 26 patients with antibodies to soluble liver antigen had antibodies to Ro52 (96%). Patients with antibodies to Ro52 and antibodies to soluble liver antigen had a higher frequency of human leucocyte antigen (HLA) DRB1(*) 03 (78 vs 50%, P=0.05) and lower occurrence of HLA DRB1(*) 04 (22 vs 57%, P=0.01) than patients with antibodies to Ro52 alone. Antibodies to Ro52 alone [hazard ratio (HR), 2.90; 95% confidence interval (CI), 1.18-7.14, P=0.02] and antibodies to Ro52 in conjunction with antibodies to soluble liver antigen (HR, 2.98; 95% CI, 1.07-8.43, P=0.04) were independently associated with the development of cirrhosis and hepatic death or liver transplantation. Antibodies to Ro52 alone and antibodies to Ro52 in conjunction with antibodies to soluble liver antigen are independently associated with a poor prognosis in type 1 autoimmune hepatitis. The prognostic implications ascribed to antibodies to soluble liver antigen may reflect their almost invariable concurrence with antibodies to Ro52.