PWE-079 Diagnostic utility of autoimmune serology profiling

Introduction Immunoserologic investigation is of clinical use in the discrimination of liver disease of autoimmune aetiology from alternate causes. We sought to investigate the diagnostic utility of immune serology in patients investigated for liver disease. Method We analysed the immunoprofile of patients investigated for liver disease at a tertiary centre between 2001 and 2017. We compared Results for patients clinically coded with a diagnosis of autoimmune hepatitis (AIH), primary biliary cholangitis (PBC), or primary sclerosing cholangitis (PSC) against those without, amongst patients investigated by our liver services. Overlap features (o/l) were included if clinically documented. Results We evaluated Results from 2874 patients. Patients with AIH (n=556; incl. o/l (19.3%)) were predominantly female (3 F: 1 M), with a bimodal age distribution at presentation (≈18, 62 years). Antinuclear (ANA) and smooth muscle antibodies (SMA) were both sensitive (92.3%, 86.6% respectively) for a diagnosis of AIH, but poorly specific (23.6%, 27.2%). Amongst patients with T1AIH, 39 patients had antibodies to soluble liver antigen (SLA). Of these, 36 (92%) had concordance to Ro-52 antibodies. 2.3% of patients with AIH had T2AIH, defined by antibodies to liver-kidney microsome-1 (LKM); of the 13 patients, 4 had SMA and 8 have ANA reactivity. Patients with PBC (n=527; incl. o/l (18.3%)) were primarily female (9 F: 1 M), and in their 6th decade (mean=56 years). Anti-mitochondrial antibody (AMA) testing was both sensitive (70.6%) and specific (91.2%). In patients who were AMA negative at testing, anti-glycoprotein 210 (gp210) (sens 40.7%, spec 95%) and anti-SP100 (sens 40.3%, spec 96.3%) proved useful diagnostic adjuncts. Patients with PSC (n=291; incl. o/l (10.1%)) were predominantly male (3 M: 2 F) with a bimodal age distribution at presentation (≈25, 58 years). Perinuclear anti neutrophil cytoplasmic antibody was seemingly specific for PSC (93.9%), as compared to non-autoimmune liver patients, and found in 23.3% of patients. To differentiate between PSC and AIH, anti-SLA, -LC1, -Ro-52, ds-DNA and f-actin all had a PPV >95% for AIH;>85% in patients who were negative for SMA. Similarly, to distinguish PBC from PSC, AMA, anti Ro-52, -gp210 and -SP100 all have a PPV >90% for PBC; this was also true for anti-Ro52 and -SP100 in patients who were AMA-negative. Conclusion The optimal use of immunoserology requires careful evaluation of test parameters in the population served by a liver disease programme.