Reflections on therapeutic trials in primary biliary cirrhosis


  • See Article on Page 1184

  • Potential conflict of interest: Nothing to report.

We have had the results of our randomized, placebo-controlled treatment trial of methotrexate (MTX) in patients with primary biliary cirrhosis (PBC) accepted for publication in this issue of HEPATOLOGY,1 and have been offered the opportunity to reflect on this and other therapeutic trials in PBC by the editors of this journal.


MTX, methotrexate; UCDA, ursodeoxycholic acid.

The time span from protocol development to this first publication1 of our major results has been approximately 12 years. Over 30 investigators, a large number of support staff including clinical coordinators, GCRC nurses and aides, each with secretarial support, have been involved in extensive dedicated collection and validation of data. It is apparent that an extensive investment has been committed to carrying out this clinical trial.

The conclusion that weekly MTX, when added to ursodeoxycholic acid (UDCA), has no appreciable effect on the course of PBC treated with UDCA alone leaves us with a relatively barren therapeutic armamentarium for this disease. This is particularly so because in our judgment, UDCA, the only compound approved by the FDA for treatment of PBC, has yet to be shown to prolong life free of liver transplantation, or to decrease the need for liver transplantation, because appropriate numbers of randomized, placebo-controlled patients have not been followed simultaneously for the same lengths of time as UDCA-treated patients who have comparable severity of disease.2–6

It is now generally conceded that advanced PBC will not respond to medical therapy. The focus has shifted to treating early disease not already encumbered with hepatic complications. The dilemma posed by this is that early disease has a long natural history. In our current study that randomized patients with only moderately advanced PBC,1 we estimated eight year survival probability as 93%, and 8-year transplant-free survival probability as 84%. In order to demonstrate efficacy of any medical intervention, when either overall survival or transplant-free survival are the primary endpoints, an inordinately large number of patients, including simultaneously followed appropriate controls, would have to be enrolled.

What pressure does this place on investigators in selecting new agents to assess in the treatment of PBC? In the absence of surrogate markers for transplantation and death that could lead to earlier termination of trials, even feasibility studies which are of most use in excluding agents with high toxicity, or that show improvement in results of laboratory tests, or of decreased inflammation in liver, would involve commitment of large resources to assess the efficacy of any medication.

A strong case has been made for an autoimmune etiopathogenesis of PBC. The evidence has been reviewed in detail in a recent publication.7 Loss of immune tolerance (to self), however initiated, is felt to induce progressive immune destruction of small bile ductules and ducts in the liver. It is not surprising then that most therapeutic trials have involved the use of immunosuppressive agents. Yet, none has been shown to favorably affect the course of the disease.

Vierling has recently provided a detailed list of future treatment options for patients with PBC.8 It is clear that it will not be possible to test more than a few of these in the next several decades, and even then it will be difficult to decide on which ones to assess. Regrettably, this is a pessimistic point of view. We await a seminal breakthrough that is not yet evident.