Serious Liver Injury

FDA Consumer, May, 2001 by Michelle Meadows

Leading Reason For Drug Removals, Restrictions

People who experience acute liver failure become critically ill in a matter of days, with most cases resulting in the need for a liver transplant or even death Drug-induced liver injury has become the number one cause of this rare syndrome. It's also the leading reason that drugs are removed from the market or require restricted use and special monitoring of patients.

During a drug's development, manufacturers conduct animal tests that include an assessment of potential toxic effects of the new drug on the liver. Additionally, liver function testing is conducted on humans to detect adverse reactions--some of which may keep the drug from making it to the market. But acute liver failure due to a new drug is a very rare event, and may show up only after a drug has been approved.

Because clinical trials typically involve several thousand subjects, they pick up common problems--affecting 1 person in 500 or 1 in 1,000, says Peter Honig, M.D., director of the Food and Drug Administration's office of postmarketing drug risk assessment. "But they aren't designed to pick up rare adverse events, occurring at a rate of 1 per 50,000 exposures," such as acute liver failure, he says. Liver injury can also be hard to predict because genetics may make one patient more susceptible to liver problems than others. Or, a drug may have toxic effects when used with other substances or when used too long.

Because liver failure also can occur rarely in people not exposed to drugs, it is often hard to know if early cases reported for a new drug were actually caused by the drug. When cases of liver injury are caused by a drug, whether before drug marketing or after, experts weigh the risk against the value of the drug. "Unless the drug is treating a life-threatening illness, a significant rate of severe injury (greater than 1 in 50,000 exposures) will lead to limiting the drug's use or withdrawing it from the market," Honig explains.

Most recently, the FDA asked Parke Davis/Warner Lambert to voluntarily withdraw the diabetes drug Rezulin (troglitazone) from the market in March 2000. FDA officials had reviewed safety data showing that Rezulin is more toxic to the liver than two newer, similar drugs on the market, Avandia (rosiglitazone) and Actos (pioglitazone). And in 1998, the FDA asked Wyeth-Ayerst Laboratories to voluntarily remove the pain medication Duract (bromfenac) from the market. The FDA had received reports of liver failure associated with the drug when it was used for longer periods than the 10 days specified in the labeling.

Trovan (trovafloxacin/alatrofloxacin) is an example of a drug with significant limitations of use because of the potential for serious liver injury. The antimicrobial therapy treats a variety of infections, from mild to life threatening. Though no cases of liver failure, liver transplant, or death were reported in the 7,000 patients who took part in premarketing clinical trials for Trovan, the FDA began receiving reports of liver failure after Pfizer began marketing the drug in 1998. As a result, the FDA and Pfizer agreed to restrictions, which include limiting distribution of Trovan to inpatient facilities (hospitals, nursing homes) so doctors can closely monitor patients taking the drug. Trovan use was also limited to the treatment of patients with serious, life- or limb-threatening infections.

The FDA is working with other organizations to identify potential liver problems before drugs have been approved. And when that's not possible, experts want to spot problems quickly after marketing through spontaneous adverse event reporting. In February 2000, representatives from pharmaceutical companies, universities, and the FDA met in Chantilly, Va., to explore these goals.

Sponsored by the FDA, the American Association of the Study of Liver Diseases, and the Pharmaceutical Research and Manufacturers of America, the two-day workshop provided a forum to discuss pre-clinical, clinical, and postmarketing approaches related to drug-induced liver injury. These approaches include such areas as improving the quality of adverse event reports, conducting the best tests to detect liver abnormalities during a drug's development, and better interpreting laboratory findings.

One possible strategy for researchers, Honig says, will involve going back and evaluating animal test databases of those drugs known to cause severe liver damage in humans to see if certain laboratory test combinations are predictors of liver injury.

"We go over data rigorously and regularly examine possible signals that can predict liver injuries," Honig says. "But we need to learn more about the performance characteristics of the tests we are using." Performance characteristics rear to certain elements of liver function tests, alone or in combination, which can predict serious liver injury. Minor changes in one or more of the liver tests are very common in people in general, including patients enrolled in clinical trials.