Maintenance immunosuppressive therapy in autoimmune hepatitis: To stop or not to stop, that is the question.

Autoimmune hepatitis (AIH) is a rare and challenging disease. Its presentation ranges from mildly elevated transaminases to jaundice with decompensated liver disease. Blood tests often reveal elevated immunoglobulin G levels and several antibody tests like anti-smooth muscle antibodies, liver-kidney microsomal antibodies or soluble liver antigen antibodies can be positive. Viral hepatitis (including hepatitis E viral infection), drug-induced liver injury and (especially in young patients) Wilson disease should be excluded. A single reliable diagnostic test is lacking, but the (revised or simplified) AIH scoring system allows to establish a probable or definite AIH diagnosis [ ]. A rapid and correct diagnosis of AIH is important, because if left untreated, AIH can lead to cirrhosis and even liver failure with need for transplantation or premature death. It is encouraging to recognize that the modern AIH population is quite different from those in the early years when Dame Sheila Sherlock treated AIH patients (generally icteric with often advanced liver disease) with prednisone monotherapy [ ]. Currently, the diagnosis is established much earlier in most cases, and therapeutic options have expanded greatly. When therapy is initiated with high dose corticosteroids, transaminases often show a rapid response allowing to taper steroids and maintenance immunosuppressive therapy can then be prescribed. Usually azathioprine is used for long-term treatment, either alone or in combination with low-dose corticosteroids. In about 20% of cases, intolerance or insufficient treatment response necessitate second-line treatment with mycophenolate mofetil. Third-line treatment options include cyclosporine, tacrolimus and sirolimus. There is limited experience with infliximab or rituximab. Poor adherence to long-term treatment for chronic diseases such as hypertension, hypercholesterolemia and human immunodeficiency virus (HIV) infection is a frequent phenomenon (especially in asymptomatic patients) with non-adherence rates of 50% or higher, and the same could apply to AIH patients. In general, the patient is happy because -at least according to liver tests- the enigmatic liver inflammation is under control. In the best case the patients asks after some time if therapy can be stopped and in the worst case the patient decides without consultation of the physician, to stop all medication. Unfortunately, relapse is very common after stopping immunosuppressive therapy in AIH patients (25- 89% in previous studies [ , , , , ]), mostly within one year after withdrawal. Risk factors reported to be associated with high relapse rates are slow response to initial immunosuppression, shorter treatment duration, higher ALT and IgG levels at withdrawal, a history of previous relapse, a liver biopsy with residual inflammation at withdrawal, immunosuppressive combination therapy at withdrawal, older age, and concomitant other autoimmune diseases. Long-term treatment with nitrofurantoin, minocycline and Tumor Necrosis Factor alpha (TNF alpha) inhibitors and other medication can lead to drug-induced AIH [ ]. Drug-induced autoimmune-like hepatitis shows a different natural history with very low risk of relapse after drug withdrawal (and often also temporarily prednisone treatment) and is not discussed further. There is no agreement on strict stopping rules for immunosuppression in AIH and reliable non-invasive markers to prove complete remission of AIH do not exist. Recent guidelines from the European Association for the Study of the Liver (EASL) advise that withdrawal could be considered after at least 3 years treatment, including at least two years with normal transaminases and IgG [ ]. In this issue of the Journal, van den Brand et al. present the first prospective study on drug withdrawal with adherence to a strict protocol in patients with AIH in long-term remission [ ]. Although small, this study provides some valuable insights for the clinician. In addition to the EASL criteria, their inclusion criteria were histological activity index (HAI) ≤ 3, absence of cirrhosis and immunosuppressive monotherapy. Of note, persistent histological activity precluded drug withdrawal in 30% of cases (5 of 17 patients). One of these patients had progressed from F0 (no fibrosis) to incomplete cirrhosis (stage 5) with HAI=10 compared to the baseline biopsy. A total of eight patients (67%) remained in treatment-free remission during a median follow-up of 62 (range: 13-75) months. A relapse occurred in four patients (33%).