Autoimmune hepatitis is an autoimmune chronic liver disease whose treatment includes the use of immunosuppressive drugs, particularly azathioprine, and corticosteroids. When properly treated, patients have a good survival. One of the major problems related to its treatment is the the high rate of relapses after stopping therapy that has lead to biochemical and histological remissions, close to 80%. Many authors recommend continuous treatment throughout life, resulting in the occurrence of many side effects. Chloroquine is a drug with anti-inflammatory properties already used in the treatment of other extrahepatic autoimmune liver diseases. There are some reports in the literature about its beneficial use in liver diseases such as chronic hepatitis B, and a pilot study in patients with autoimmune hepatitis, in which its use was associated with a 6.49 times lower risk of disease recurrence when compared with patients in whom treatment was discontinued after remission. Our purpose is to investigate, in a double-blind randomized trial with placebo, whether chloroquine prevents the recurrence of AIH in patients with histological remission after discontinuation of conventional treatment and to evaluate the occurrence of side effects.

Autoimmune hepatitis (AIH) is a chronic disease with a progressive destruction of hepatic parenchyma, leading to cirrhosis and high mortality in the absence of specific treatment. It has been demonstrated that the treatment with corticosteroids and azathioprine provides clinical and laboratory improvement, reduction of histological inflammatory activity on liver biopsy and an increased survival. The life expectancy of correctly treated patients can approach that of age- and gender-matched controls. One of the features of AIH is the high rate of recurrence after discontinuation of treatment in patients who achieved histological remission. The chances of recurrence reach up to 50% at 6 months and 70% after one year, and the rates of sustained remission, are limited to 10 to 15% in 5 years. Thus, most patients require maintenance treatment to prevent recurrence.

Chloroquine is a drug of the group of 4-aminoquinolines, synthetic derivatives of quinine and constituent of the bark of the Cinchona tree. Chloroquine accumulates in tissues in considerable amounts. In animals, from 200 to 700 times the plasma concentration can be found in liver, spleen, kidneys and lungs. As a weak base, it accumulates intracellularly, particularly in lysosomes with a consequent increase in pH within these organelles, which could contribute to its toxicity. Lysosomal lamellar bodies are observed in tissues affected by chloroquine, such as retina and neuromuscular system. Chloroquine inhibits the absorption and the binding of mitochondrial calcium, alters the membrane permeability and the transport of enzymes to the lysosomes. Apparently there are other mechanisms to explain its anti-inflammatory action; such as the interference with the release of TNF from mononuclear phagocytes by inhibiting gene expression and the down-regulation of TNF receptors, by delaying their transport to the cellular surface. Due to these mechanisms of action, chloroquine has anti-inflammatory activities and therefore is used in diseases such as rheumatoid arthritis and systemic lupus erythematosus. In liver diseases, chloroquine was used in patients with hepatitis B with normalization of the levels of aminotransferases and of the prothrombin time during treatment and relapse after drug discontinuation. Chloroquine was also evaluated in patients with porphyria cutanea tarda and despite the clinical and biochemical improvement, liver biopsies remained unchanged after one year of treatment.

A previous pilot study was performed in our institution, and published in 2005, with chloroquine diphosphate for the maintenance treatment of AIH. In this study, 14 patients with a biochemical and histological remission were treated with chloroquine diphosphate 250 mg/day for at least 12 months or until disease recurrence, and compared with 18 historical controls, which was held in discontinuation of treatment after remission. The chance of relapse was 6.49 times higher in the historical controls when compared with patients in the group treated with chloroquine (72.2% versus 23.5%, p = 0.031). The use of chloroquine was safe in patients with liver cirrhosis without decompensation, and there were no serious adverse events within two years of use.

The most common adverse effects of chloroquine are mild and transient such as gastrointestinal symptoms, headache, dizziness, blurry vision and fatigue. The more severe reactions described are itching, cardiovascular manifestations, dyskinesias, eye injuries, neuromuscular disorders and hearing loss. Among the most feared adverse effects of chloroquine, are the eye injuries, usually associated with chronic treatment. They may consist of changes in the retina, lens, cornea and optic nerve. Usually they remain stable after drug withdrawal, if the drug is discontinued in early stages. However, the retinal damage can increase when found in advanced stages, and may progress even years after cessation of chloroquine. It is believed that the chloroquine retinopathy can be prevented or recognized in an early reversible stage with judicious use, appropriate doses and regular ophthalmologic follow-up. It is recommended that the daily dose does not exceed 250 mg of chloroquine diphosphate or 400 mg of hydroxychloroquine, and ophthalmologic evaluations are carried out every 4 to 6 months. Despite the adverse effects and toxic reactions described above, there is a consensus in most studies with chloroquine that it is a well tolerated drug, provided that the appropriate dosage guidelines and regular eye examinations are followed. With these cautions in mind, its use rarely causes serious side or irreversible effects.

The purpose of this study is to investigate whether chloroquine prevents the risk of recurrence of AIH in patients in remission after discontinuation of conventional treatment, in prospective, randomized, double-blind trial, and to evaluate the occurrence of side effects from the use of chloroquine.

To be included patients had to satisfy the following criteria simultaneously: a diagnosis of probable / definite AIH (according to the criteria of the international AIH Group), normal liver function and absence of clinical signs of decompensated liver disease ( ascites, hepatic encephalopathy, gastrointestinal bleeding and hepatocellular carcinoma); biochemical (aminotransferase levels within the normal value for at least 18 months) and histological remission in liver biopsy (periportal inflammatory activity less than 2) in the presence of immunosuppressive treatment; men or non-pregnant women and women with no intention to become pregnant; willing to participate in the study. An explanation of this study will be made to patients. For their enrollment, it is necessary that the patients are in accordance with the proposed study, following the precepts of the Declaration of Helsinki. If patients refuse to participate in the study, they will be treated following the traditional guidelines of our service. We will exclude patients who stop the drug before six months for side effects, patient's desire or loss of follow up. Treatment will be discontinued in case of pregnancy, patient's desire, side-effects or relapse of AIH. Recurrence was defined as a sustained increase in liver enzymes, greater than or equal to two times the upper normal value (according to the criteria of the International Autoimmune Hepatitis Group).

The sample size was calculated by exact Fisher test to compare the treatment and placebo groups considering the relapse percentage varying between 25% to 40% and 65% to 80%, respectively. Thus, the sample size was obtained between from 30 to 128 patients for each group considering the power of 80% and the statistical significance level of 5%. The statistical analysis will present mean and standard error for quantitative variables and percentages for qualitative variables as descriptive measures. The survival analysis will be performed,through of simple Cox regression and multiple Cox regression with treatment as a fixed covariate, to test the effect of treatment considering patients' characteristics on the relapse risk from the beginning of treatment to the relapse in a period of 3 years. In a second step, survival analysis will be repeated for patients that finished or stopped the treatment for side effects to verify if the absence of relapse during treatment is due the patients' characteristics. Furthermore, the proportions of side effects will be compared between the groups by exact Fisher test. A p-value less than 5% will be considered of statistical significance.

Patients will be randomized to receive placebo or chloroquine, under the name of drug A or drug B. Information on the composition of the tablets is kept secret with the Pharmacy's staff of Hospital das Clinicas, University of Sao Paulo School of Medicine. The treatment regimen that led to the histological remission will be maintained for a month, in association with drug A or B, at which time it will be withdrawal. Patients who were taking ursodeoxycholic acid to achieve remission will remain in use of this drug during the study. Drugs A or B will be maintained until disease recurrence (persistent changes of aminotransferases greater than twice the normal value) or until three years if biochemical tests are normal, or until the time that patient want to remain in the study. If there is recurrence of the disease, the patient will be excluded from the study and will receive the treatment they received previously. All patients will have visits every 30 days during the first six months after using drug A or drug B in monotherapy, with routine blood tests performed. After then, consultations will be every two months. All complaints will be recorded and the number of tablets will be counted at each visit to assess adherence and to account for failures in taking the medication. Every patient will be treated by the doctors responsible for the study, laboratory tests will be performed in the Central Laboratory of the hospital in which the protocol will be performed and will be available online for consultation. In the case of side effects, the drug may be stopped at the request of the patient or according to medical advice. All patients will undergo to initial ophthalmologic evaluation followed by six-monthly evaluations. The drug will be withdrawal, if changes suggestive of retinopathy are observed. After the end of treatment, patients will be followed over the next five years due to the risk of late onset of retinopathy.

Inclusion Criteria: - diagnosis of autoimmune hepatitis according to Autoimmune Hepatitis International Group

• histological remission during treatment with immunosuppressive drugs (Liver biopsy with periportal inflammatory activity less than 2)
• No evidence of decompensated liver cirrhosis
• Non-pregnant women and women with no intention to become pregnant
• Willing to participate in the study

Exclusion Criteria:

• patients who needed to suspend the drug under six months of the medication because of side effects or the patient's desire
• cases of loss of follow up