A Joint Effort

A column from ACR President Allan Gibofsky, MD, JD

April, 2002

Resource File

Pharmacovigilance: An Update on Leflunomide

Over the years, many of you have asked me why I “bothered” to get a law degree. How could becoming a lawyer possibly assist me in my activities as a rheumatologist?

On March 28, 2002, the group Public Citizen petitioned the FDA to remove leflunomide (Arava®) from the market because of its allegedly unacceptable hepatic toxicity. In addition, advertisements have recently appeared in newspapers nationwide offering legal evaluation for persons potentially injured by this drug (as well as several others). This President’s column is meant to explore some of these issues from both a clinical and legal perspective.

Public Citizen’s petition (available at www.citizen.org) notes that as of September 2001, there had been more than 130 reports of serious hepatic reactions, including 56 hospitalizations and 12 deaths related to the use of leflunomide. Previously, at the request of the European Agency for the Evaluation of Medicinal Products (EMEA is the European agency similar to the U.S. FDA), the manufacturer prepared a summary of all reports of hepatotoxicity among patients treated with leflunomide worldwide through March 2001. This information was reviewed in detail in the August 2001 ACR Hotline (Reports of Leflunomide Hepatotoxicity in Patients with Rheumatoid Arthritis). Recommendations for monitoring liver function tests, managing potential hepatic adverse events, and an information sheet for patients concerning potential hepatic toxicity were included.

According to the manufacturer, from March through September 2001 there have been 21 additional reports of hepatic adverse events possibly or probably related to leflunomide administration among an estimated 65,000 additional patients exposed to the drug. In these reports, when patients were monitored and managed according to guidelines, normalization or improvement in liver function tests was observed.

Adverse effects are important both to patients and to the clinicians who guide their care. Although adverse events observed during clinical studies of a new medication are summarized in the initial product label, certain relevant safety information becomes available only after a drug has been approved and is introduced into the market. The process of collecting safety information for newly marketed products is known as pharmacovigilance. It is an ongoing global activity that includes data derived from pharmaceutical manufacturers, physicians and patients. It includes blinded and open-label clinical trial data, post-marketing safety reports, and data derived from patient registries and other sources.

The strength of pharmacovigilance is that it allows assessment of the safety of drugs in populations much larger and more diverse than those enrolled in clinical trials, and for longer periods of time. This allows the identification of rare adverse events as well as those that may occur preferentially in subsets of exposed patients. Diverse populations may have comorbidities or take other medications that impact the safety of a given drug. For example, in the case of leflunomide, it was observed that included among the most frequently implicated potential hepatotoxins used by leflunomide patients were NSAIDs, including celecoxib and diclofenac and other NSAIDs metabolized by the same cytochrome P450 2C9 pathway by which leflunomide is metabolized. Through pharmacovigilance, product approval and labeling can be appropriately changed over time. For example, dangerous drugs or drugs with an unacceptable risk/benefit ratio can be removed from the market. Alternatively, warnings such as the “black box” can be attached to beneficial drugs with particular significant safety issues, and “Dear Doctor” letters can be sent. These provide information to the clinicians who can then weigh the risk/benefit for and make an informed decision with a given patient.

While pharmacovigilance is a vital and useful process, there are limitations to any data collected from a mixture of sources that affect their interpretation. Under-reporting is very common. It is estimated that less than 1% of serious adverse events are reported to the FDA (although other studies have found rates as high as 10-20%). Importantly, this is not systematic, and factors such as the time a drug has been on the market affect whether adverse events are reported. The peak of spontaneous reporting of adverse events for a drug is at the end of the second year of marketing. Thereafter, reporting of adverse effects tends to decline (the so-called Weber effect). For example, although the safety of methotrexate continues to be monitored by the FDA, it would be anticipated that cases of liver toxicity related to methotrexate would be reported much less frequently than those of a more recently introduced medication; thus, comparisons of ‘safety’ cannot really be made based only on such information. Finally, and perhaps the most important problem with evaluating reports gathered as part of pharmacovigilance, is that causality is often difficult to establish; the accuracy of case details and the presence of confounding variables often remain undefined.

So where does this leave us?

• Adverse effects, some of them severe, can occur with any medication at any time, patients need to understand this and be informed of the risks and benefits of any therapeutic choice as well as the risks and benefits of any available alternatives. Rheumatologists must assess the appropriateness of any treatment based upon factors such as accurate diagnosis, severity of disease, presence of comorbid conditions, and the use of concomitant medications. Ultimately, however, it is the patient who must decide whether to accept or decline a particular medication. Once begun, appropriate monitoring during treatment is required.
  
  
• Product approval and labeling is a dynamic process. A fuller understanding of adverse events often comes only after a drug has been approved. Rheumatologists need to be vigilant regarding toxicities. Not only should patients be encouraged to report adverse events, but clinicians should also report adverse events that are serious, new, or unanticipated to the FDA’s Adverse Event Reporting System (AERS; accessible at www.fda.gov).
  
  
• With specific regard to leflunomide, the FDA is currently reviewing the safety data and is likely to make recommendations concerning appropriate usage based on a complete evaluation of all relevant information. Patients can and should be informed that the rheumatology community is actively monitoring this important issue, and that appropriate action will be taken as soon as possible, based upon thorough and impartial review of the data.
  
  

Law school anyone?

Allan Gibofsky, MD, JD
President, ACR