FDA Advised to Restrict Rezulin Use for Diabetes

BETHESDA, Md. —  Citing concerns about the risk of deadly liver damage, a Food and Drug Administration advisory panel on Friday recommended that the agency restrict use of the popular diabetes pill Rezulin.

The advisors urged the FDA, by an 8-4 vote, to withdraw approval of Rezulin as a stand-alone treatment for adult-onset diabetes while also endorsing the drug’s continued use in combination with insulin and other medications.

“This is not to be used ever as a first-line drug for diabetes,” said one FDA advisor, Dr. Jules Hirsch of Rockefeller University in New York. He added that, if Rezulin continues to be widely prescribed, “in five years there will either be hundreds of people who will die of liver disease or thousands.”

The FDA is not bound by the panel’s recommendations but usually follows its advice.

The vote followed a dramatic presentation regarding the liver toxicity of Rezulin by David J. Graham, a senior FDA epidemiologist. He noted that the FDA has verified Rezulin’s role in 43 reported cases of liver failure.

“The longer you stay on [Rezulin], the longer you compound the risk of sustaining liver failure,” Graham said. “. . . What we see isn’t reassuring.”

Executives of Warner-Lambert Co., the drug’s manufacturer, said that Graham overstated the risk to Rezulin patients. Whereas Graham said he believes only 10% of deaths and transplants are voluntarily reported, the company said the figure is far higher.

“The benefits of Rezulin therapy in Type 2 diabetes clearly outweigh the highly publicized risks,” said Dr. Robert L. Zerbe, the company’s senior vice president of worldwide clinical research. “These [adverse] events are rare.”

Zerbe also said that the company will propose unspecified label modifications in an effort to make safer the stand-alone use of Rezulin. This would mark the fourth label change since Rezulin was introduced in March 1997.

Although Graham did not dispute that Rezulin can lower blood-sugar levels, he presented a multitiered indictment of the drug’s safety: There is no way to know which patients are most at risk of suffering liver failure. The liver-function monitoring now recommended by the FDA and Warner-Lambert, he said, is unreliable. And, Graham said, Rezulin is far more likely to induce death than Glucophage, a top-selling pill for adult-onset diabetes.

Graham also said the risk to Rezulin patients does not recede over time; on the contrary the danger grows after using the drug for six months.

Based on his examination of the drug, Graham told the panel that Rezulin can be expected to cause liver failure in 1 of every 1,800 patients, a rate that clashes with assurances from Warner-Lambert. A company official said at Friday’s hearing that Rezulin is associated with liver failure in 1 of every 57,000 patients.

Rezulin has registered sales exceeding $1 billion and has been prescribed to more than 1 million patients, according to the company. An estimated 15 million Americans have adult-onset, or Type 2, diabetes.

On Friday, Warner-Lambert’s stock price fell 1 11/16 to 67 11/16 on the New York Stock Exchange, where it was the most active issue.

Britain Denies Reintroducing Drug

This week, the FDA’s counterpart in Britain announced that it had refused to allow reintroduction of Rezulin after the drug had been withdrawn from Britain in December 1997. The Medicines Control Agency determined that the risks of Rezulin outweigh its benefits.

The FDA’s unusual decision to reassess Rezulin followed a December Los Angeles Times series which revealed that the agency ignored explicit warnings of danger in granting “fast-track” approval to the drug. The articles reported that the FDA medical officer assigned to review the drug recommended its rejection, based partly on his concern regarding liver toxicity.

The series also has prompted an investigation of the federal government’s top diabetes researcher at the National Institutes of Health. Dr. Richard C. Eastman accepted $78,455 in compensation from Warner-Lambert and affiliates while overseeing a nationwide diabetes prevention experiment that tested the effects of Rezulin.

Indeed, the liver failure and death of one Rezulin-treated participant in that NIH experiment was discussed at Friday’s FDA hearing by Graham. He noted that the liver functions of this participant, an East St. Louis, Ill., high school teacher named Audrey LaRue Jones, were monitored before she developed sudden liver failure.

Graham cited the recent deaths of Jones, whom he did not identify by name, and a patient in a separate clinical trial conducted by Warner-Lambert, as demonstrating the shortcomings of liver monitoring. Both victims had been tested in line with FDA recommendations.

“Lightning struck twice,” Graham said.

FDA’s Plans Are Unknown

It is now up to senior FDA officials to decide whether to follow the advisory committee’s advice.

Dr. Janet Woodcock, director of the FDA’s Center for Drug Evaluation and Research, declined after the meeting to say what course the agency will take.

“We will move expeditiously,” Woodcock said, adding: “We were very pleased to get the advice.”

Earlier this week on Capitol Hill, FDA Commissioner Jane E. Henney said the agency would rely on the advisory committee in taking its “next steps.”

Henney’s name was invoked several times at Friday’s nine-hour hearing by executives of Warner-Lambert, who noted that Henney said earlier this year that before withdrawing a drug, strong attention should be paid to its benefits.

The advisory committee voted, 12-0 and 11-1, to endorse the continued use of Rezulin in combination with insulin and, secondly, with a separate class of drugs that stimulate the pancreas to secrete more insulin.

The panel also decided that Rezulin posed too great a danger to adult-onset diabetics when taken by itself. As a stand-alone treatment, studies show that Rezulin is far less effective, yet all of those patients are still exposed to the risk of liver failure.

Warner-Lambert estimates that 20% of Rezulin patients take the drug as a stand-alone diabetes treatment.

Graham’s presentation appeared to both stun and discomfit the Endocrinologic and Metabolic Drugs Advisory Committee, which scarcely discussed the threat of liver damage while unanimously endorsing approval of the drug in December 1996.

Graham said there is no reliable evidence that compliance with regular monitoring would have saved the lives of patients who, in a matter of days or weeks, plummeted to sudden liver failure.

“For most of these cases, jaundice is the first indication of trouble,” Graham said. “And by then--the horse is out of the barn.”

Moreover, Graham said the FDA staff has found that, after six months of use, fewer than 1% of Rezulin patients are complying with recommended liver-function tests.

Adherence to the recommended liver monitoring has been “abysmal,” said one FDA advisor, Colleen A. Colley, a clinical pharmacist at the Veterans Affairs Medical Center in Portland, Ore.

Other committee members noted that two new drugs of the same chemical class as Rezulin are scheduled to be examined next month by the same panel. It is unknown whether either of these drugs, produced by British drug-maker SmithKline Beecham and Japan’s Takeda Chemical, carries less liver-related danger than Rezulin.

“We’ll let the marketplace tell us which is the best drug,” said Dr. Glenn Braunstein, an FDA advisor who is a diabetes specialist at Cedars-Sinai Medical Center in Los Angeles.

Four of the 12 voting members were granted “full waivers” to participate, despite having potential conflicts of interest, the panel’s executive secretary announced.

The underlying details of those potential conflicts were not disclosed. One of the waivers was provided to the chairman of the committee, Dr. Henry G. Bone III. He declined afterward to elaborate.

The Times reported Thursday that Henney’s office had appointed two new advisory committee members--each of whom had accepted income from a diabetes group established and financed by Warner-Lambert and its Japanese affiliate in developing Rezulin.

One of those appointees, Dr. Saul Genuth of Case Western Reserve University in Cleveland, received a conflict-of-interest waiver.

The other appointee, Dr. Mayer B. Davidson of UCLA, decided at the last minute not to participate, according to a colleague who attended.

Times staff writer Judy Lin in Washington and researcher Janet Lundblad in Los Angeles contributed to this story.