The major thrust is to determine whether treatment of patients with Primary Biliary Cirrhosis (PBC) with Ursodiol (Ursodeoxycholic Acid-UDCA) plus methotrexate (MTX) is more effective than treatment with UDCA alone.

PBC is a chronic cholestatic liver disease, predominantly of women, in which interlobular and septal bile ducts undergo inflammation and destruction. Once initiated, the disease persists and progresses at varying rates. Neither the initiating nor perpetuating mechanisms are well understood. Current concepts of pathogenesis include (1) destruction of bile ducts is maintained and perhaps initiated by autoimmune mechanisms; (2) hydrophobic bile acids which accumulate in serum and liver cause functional and cytotoxic liver injury; (3) cytokines and lymphokines released at sites of inflammation may contribute to cell damage and fibrosis. A considerable body of evidence indicates that UDCA when fed orally leads to improvement in liver tests, in pruritus and in liver histology. There exist differences in opinion as to whether development of complications of liver disease, liver transplantation or transplant-free survival is affected. UDCA, a relatively non-toxic bile acid, when administered orally, alters the composition of the bile acid pool in factor of its enrichment with UDCA and appears to protect against the cytotoxic effects of endogenous bile acids that accumulate as a result of bile acid destruction. MTX is being shown to improve liver tests, symptoms and liver histology in a small number of precirrhotic patients with PBC. The mechanism of action is unknown but felt to be related to anti-inflammatory immunosuppressive effects of MTX. The current trial explores whether MTX improves the therapeutic effects of UDCA in PBC. Patients with PBC whose serum bilirubin is less than 3 mg%, who have been on UDCA for at least 6 months, and who satisfy a series of inclusion and exclusion criteria are stratified into 2 groups on the basis of liver histologic stage (Ludwig classification), i.e., early (Stages I and II) versus late (Stages III or IV). They are then randomized to receive either methotrexate or its placebo as a second drug while continuing to receive UDCA. The relative value of the two treatment arms is assessed by comparing their effects on symptoms, results of laboratory tests, development of complications of liver disease, histologic changes in liver, liver transplantation, and on transplant-free survival. The safety of each therapeutic regimen is also being determined.

Inclusion Criteria:

• Chronic cholestatic liver disease of at least 6 months' duration.
• Serum alkaline phosphatase levels at least 1.5 times the upper limit of normal prior to treatment with UDCA.
• Serum bilirubin less than 3.0 mg% prior to treatment with UDCA.
• Serum albumin of 3.0 gram% or greater prior to treatment with UDCA.
• Positive antimitochondrial antibody test
• Liver biopsy within the previous 6 months after at least 6 months on UDCA (available for review, and at least 2 cm long if cirrhosis not detected) compatible with the diagnosis of PBC.
• Ultrasound, computed tomography (CT) or cholangiography of the biliary tree which excludes biliary obstruction.

Exclusion Criteria:

• Treatment with immunosuppressive agents including azathioprine, chlorambucil, colchicine, corticosteroids, or d-penicillamine in the preceding 3 months; or with cyclosporine, FK-506 or methotrexate in the preceding 6 months.
• Treatment with rifampin in the preceding 3 months.
• Serum bilirubin of 3.0 mg% or greater.
• Serum albumin less than 3.0 gm%.
• WBC 2,500 mm3; granulocytes 1,500 mm3; platelets 80,000mm3.
• Ascites, hepatic encephalopathy, variceal bleed.
• Findings by clinical, serologic and histologic evidence of liver disease of other etiology (such as chronic hepatitis B or C, autoimmune chronic active hepatitis, alcoholic liver disease, sclerosing cholangitis, drug-induced liver disease, symptomatic or obstructive gallstones).
• Pregnancy, or if not pregnant and in the reproductive period, unwillingness to utilize an adequate form of birth control.
• Age less than 20 or greater than 69 years.
• Epilepsy requiring use of dilantin.
• Malignant disease within the past 5 years (except skin cancer)
• Anti-HIV positive. Major illnesses that could limit life span.
• History of alcoholism during the previous 2 years.
• Creatinine clearance less than 60 ml per minute.
• Severe lung disease, defined as a diffusion capacity or vital capacity of less than 50 percent of predicted.
• Patients who are both asymptomatic and have Stage I histology on liver biopsy (Ludwig classification).