We wish to add an additional suggestion to the “AASLD Practice Guidelines: Primary Biliary Cirrhosis” that was published last month in HEPATOLOGY1 and of which we were coauthors.
Approximately 40% of patients with primary biliary cirrhosis (PBC) respond incompletely to treatment with ursodiol. Because the optimal management of these patients is unclear, there was no discussion of these patients in the Practice Guidelines. Nonresponding patients with PBC can be identified by their persistently abnormal liver biochemical tests and worsening liver histology despite treatment with ursodeoxycholic acid (UDCA) for at least 12 months.2 Such patients should ideally be treated within the context of a clinical study, but this is not possible for most patients.
Among the authors of the Practice Guidelines, most recommend the continuation of ursodiol and symptomatic treatment. They advise patients that they may eventually become candidates for liver transplantation. The two of us try immunosuppressive agents such as glucocorticoids combined with azathioprine or mycophenolate mofetil, colchicine, and methotrexate because of our favorable experience with these drugs in some patients with PBC.3–10
The rationale for the use of glucocorticoids is based on the following arguments. Glucocorticoids and specifically budesonide have been shown to provide benefit in patients treated with UDCA.8–10 Combination of both affords better biochemical response and better histological response in terms of inflammation and fibrosis. Budesonide is given at 6–9 mg/day in patients without cirrhosis. The drug is contraindicated in patients with cirrhosis. Preliminary studies in our patients indicate that patients with a suboptimal response to UDCA benefit from the combination of UDCA and glucocorticoids (prednisone or budesonide) in terms of survival without cirrhosis.
Methotrexate, 0.25 mg/kg body weight/week orally plus folic acid, 1 mg daily, improved biochemical test results and liver histological findings when it was added to UDCA in patients who had an incomplete response to UDCA5 and was associated with sustained remission in other patients.4 However, other studies found no efficacy of methotrexate when it was used alone or in combination with UDCA.11 In a 10-year study, survival was the same in patients taking methotrexate and UDCA as in those taking colchicine and UDCA, and survival was similar to that predicted by the Mayo model.11 However, one-third of the patients had few signs of PBC after 10 years of treatment.11 No patient who was in the precirrhotic stage at baseline showed progression to cirrhosis. Colchicine was effective in some patients who failed to respond to ursodiol and methotrexate6 and improved survival in one prospective double-blind trial.12
These drugs have had few side effects in patients with PBC and might benefit some patients with PBC who fail to respond to ursodiol. We believe the risk-to-benefit ratio of these drugs favors their use in patients with PBC who respond poorly to ursodiol.