Introduction If immunosuppressant therapy (IST) is withdrawn in patients with autoimmune hepatitis (AIH) when remission is achieved, 50–70% will have a disease relapse within 12 months. Thus, many patients receive long-term maintenance IST. However, IST has potential side effects, including an increased cancer risk. Recent studies suggest a lower AIH relapse rate if IST is withdrawn at a later stage in the disease. Data are lacking in regard to long-term outcome following IST withdrawal.

Methods Retrospective audit of long-term outcome of AIH following withdrawal of IST. A total of 282 patients with definite or probable AIH by International AIH Group criteria, had presented to our Unit between 1971–2015 and had achieved normalisation of serum ALT within 12 months of starting IST (prednisolone plus azathioprine, or mycophenolate if azathioprine intolerant). From these, we found 25 patients in whom all IST was subsequently discontinued. We assessed outcome in these patients, in comparison to those who continued IST. Follow-up was until last serum ALT for calculating relapse rate and until death or last known to be alive for calculating survival rates.

Results The 25 patients (21 female, age at diagnosis 50 (8–68) years) had been on IST before withdrawal for 3.9 (0.1–22.4) years. Five had cirrhosis at presentation. All patients had normal serum ALT and globulins at time of withdrawal except one patient with acute pancreatitis. Reasons for IS withdrawal were: cancer (6), infections (3), side effects (4), frailty (6) patient choice (6), doubt about diagnosis (subsequently resolved) (1).

Relapses occurred in 7 patients (28%) after 1.6 (0.5–9.6) years. Relapse rates were 15%, 20% and 28% after 2, 5 and 10 years respectively. All patients achieved re-normalisation of serum ALT within 3 months of re-starting therapy (prednisolone 10–40 mg/day) plus (in 5 patients) azathioprine or mycophenolate, subsequently continued as maintenance therapy. Over follow up (from IST withdrawal) of 4.2 (1.2–25.1 years), no patient developed liver decompensation, variceal bleeding or hepatocellular carcinoma. Six patients died after 4 (3.3–15.9) years; none due to liver disease or following AIH relapse. Patients stopping IST (n = 25) had, compared to those who continued IST (n = 257) higher 10- and 20 year survival (99±4% and 79±9% respectively vs 78±3% and 48±4% p = 0.021).

Conclusion In patients with AIH who achieve remission a trial of withdrawal of IS therapy is associated with (a) a lower disease relapse rate than previously reported, (b) no liver adverse events and (c) no reduction in long-term survival compared with patients who continue on long-term IS therapy.

Disclosure of Interest None Declared