The Rise and Fall of the Killer Drug RezulinSeptember 2000
By David Willman
New commissioner seeks reevaluation
Newly appointed FDA commissioner Jane E. Henney ordered a reevaluation of Rezulin in January 1999, in response to a Los Angeles Times investigative series that disclosed at least 33 liver failure deaths attributable to Rezulin.
Graham, the agency’s leading specialist in evaluating and preventing deaths caused by prescription drugs, was assigned the job.
Wiry, with closely cropped auburn hair, Graham, 46, is a rare breed at the FDA. He has nothing to do with reviewing new drugs. His responsibility is to examine medicines already on the market.
At a time when nearly half of all Americans regularly take one or more prescription drugs, Graham’s work is of crucial importance.
Trained at Yale and Johns Hopkins medical schools, Graham’s straight-arrow approach was unaffected by the rough-and-tumble of 15 years inside the FDA: Delivering bad news about a hot-selling drug could stymie a career. Senior FDA officials could not be counted on to provide support. Full-throated opposition often would come from the product’s manufacturer.
While Warner-Lambert tapped an array of specialists to defend Rezulin, Graham had the assistance of one colleague, pharmacist Lanh Green. They began studying the harm done to patients, the extent to which liver-monitoring recommendations had been followed and the ongoing risks of taking Rezulin.
Within two months, Graham amassed an indictment of Rezulin. He presented his research on March 26 to an FDA advisory committee—the same panel that had unanimously endorsed the drug’s approval. Among Graham’s findings:
- An estimated 430 or more Rezulin patients had suffered liver failure.
- Patients incurred 1,200 times more risk of liver failure by taking Rezulin.
- One of every 1,800 Rezulin patients could be expected to suffer liver failure, a far cry from the 1-in-100,000 risk espoused by a Warner-Lambert spokesman.
- Regular liver monitoring offered no safety guarantee, in part because Rezulin could so quickly and unpredictably damage the liver, sometimes within days. And more than 99% of patients taking Rezulin for four months or longer failed to follow the liver-monitoring recommendations.
Graham also described the deaths of Audrey Jones and another woman, Rosa Delia Valenzuela, who had died in a clinical trial despite undergoing monitoring. Valenzuela, 63, of Arcadia, Calif., was struck with liver failure about a month after taking Rezulin as a participant in a Warner-Lambert clinical trial.
Warner-Lambert’s representatives told the advisory committee that Rezulin could not be held responsible for many of the liver failures. They cited factors such as preexisting medical conditions.
The committee was unpersuaded by Graham and voted, 11 to 1, to recommend keeping Rezulin on the market. Three of the panelists had received compensation from Warner-Lambert or an affiliate; they were granted conflict-of-interest waivers by the FDA.
After the meeting, the FDA’s Woodcock promptly distanced herself from Graham’s presentation. She said that it was based on “a very broad range . . . of best guesses.”
Graham also was admonished by his immediate boss for the breadth of his report to the committee, he told acquaintances. The FDA declined to allow Graham to comment for this article.
As the advisory committee members dispersed from the March 26 meeting, tragedy again was unfolding. Another Rezulin patient, despite undergoing monthly monitoring in a Warner-Lambert clinical trial, lay near death.
Three days later, 37-year-old Adrian C. Seay died at nearby Washington Hospital Center. The District of Columbia medical examiner identified the cause of death as liver failure “following treatment with” Rezulin.
On June 16, the FDA agreed to yet another labeling change, the fourth for Rezulin.
Patients and doctors were advised that liver monitoring should be conducted monthly for the entire first year, instead of eight months. The FDA also said that new diabetes patients should no longer use Rezulin initially as a stand-alone treatment.
New recommended use for drug OKd
But the FDA was hardly renouncing Rezulin. In addition to keeping the drug on the market, the agency approved Warner-Lambert’s request for a new recommended use of Rezulin, in combination with two other popular blood-sugar-lowering pills.
The FDA’s handling of Rezulin was ridiculed at a June conference sponsored by Georgetown University Medical Center. Dr. Alastair J.J. Wood, a Vanderbilt University professor who also is drug therapy editor of the New England Journal of Medicine, likened the FDA’s label changes to managing the risk posed by a steep cliff.
“The point was, you don’t keep putting up more and more signs if people continue falling off the cliff,” Wood said. “You try to do something more definitive, like try to prevent them from falling off. You put up a fence.”
By late 1999, the tragedies persisted.
The Times disclosed on Dec. 15 that the FDA had received reports of 21 additional liver failure deaths since Graham’s presentation. Graham, acting with the knowledge of his supervisor, began preparing an updated analysis.
At the start of the new year, Graham told his fellow Termites that he was readying a knockout blow of Rezulin. He had amassed a case so strong that no one at the FDA could resist any longer.
Or so Graham hoped.
On Jan. 6 of this year, Graham shared his latest findings regarding Rezulin’s toll at an FDA staff meeting. In attendance was Dr. Robert J. Temple, one of the agency’s most respected scientists. As director of one of the FDA’s five drug-review offices, Temple was a subordinate, on paper, to Lumpkin and Woodcock. But his reputation, built over 28 years inside the agency, was without peer. In a room of mostly mid-level medical officers, Temple’s every word and gesture counted.
First, he raised the precedent of the “imminent hazard” withdrawal in 1977 of Phenformin, the diabetes drug Misbin had dealt with at Ceiba-Geigy.
This was useful context. But where, his colleagues wanted to know, did Temple stand on Rezulin? All eyes were fixed on him.
Temple briskly slashed two fingers across his throat.
Specialists asked to look at two other drugs
As the session ended, the Termites were emboldened. A consensus had formed for Rezulin’s prompt withdrawal.
But any such hope was doused on Jan. 13, when Lumpkin met with the same FDA specialists.
Lumpkin directed them to shift focus and assess the safety of two newer chemical cousins of Rezulin. The FDA had granted rapid approval to these diabetes drugs, Avandia and Actos, in mid-1999 after they were found to be far less toxic to the liver.
Several FDA physicians viewed Lumpkin’s approach as diversionary. Rezulin, they said, should stand or fall on its own merits.
Lumpkin created further delay by scheduling two private meetings between FDA staff members and Warner-Lambert executives on Feb. 2 and March 1.
At the first meeting, Graham pointed out that the FDA had received reports of liver failure among patients who had taken Rezulin for eight to 18 months. This clashed with the company’s earlier claim that the risk “substantially declines after six to eight months of therapy.”
"How many more unnecessary deaths will it take before you take action?" Misbin asked the company.
Some of the most pointed questions came from Temple and Misbin. They were skeptical of the company’s claims that liver monitoring worked.
As of early February, the FDA press office confirmed 85 cases of liver failure, including 58 deaths. This was nearly twice the number of liver failures acknowledged by the FDA a year earlier.
Nevertheless, the FDA brass continued to endorse Rezulin.
On Feb. 24, Woodcock issued a statement reaffirming the agency’s confidence in the drug, saying that “in many patients it has proven to be very effective.” Her remarks showed that Rezulin continued to enjoy deep and well-connected support.
And doctors kept prescribing the drug. According to the pharmaceutical information company IMS Health, Rezulin during the preceding year generated $674 million in sales.
“People who felt that the drug was too risky really shouldn’t have prescribed it,” Woodcock said in an interview. “The information was available to them. . . . Everything we’ve done [has been] out in the open.”
FDA session with firm seen as pivotal
The Termites, their ranks now swollen to about a dozen agency specialists, did not retreat. They made known to colleagues their conclusion that Rezulin should be withdrawn.
The March 1 meeting with Warner-Lambert loomed pivotal. The Termites were bracing for more resistance from the top. Senior FDA officials were determined to keep the session shrouded in confidentiality.
Saying that they suspected a Times reporter was inside the agency’s high-security headquarters in Rockville, Md., officials at midday shifted the 4 p.m. meeting a few miles north, to a conference room next to the sixth-floor offices of Woodcock and Lumpkin.
Behind closed doors, Lumpkin broached the option of bringing Rezulin for a third time to the FDA advisory committee, the panel that had overwhelmingly endorsed the drug in December 1996 and again in March 1999.
Lumpkin’s suggestion, several participants said, meant that the drug might stay on the market indefinitely.
Warner-Lambert proposed that Rezulin remain on the market for as long as two years while the company conducted new studies measuring the frequency of liver failure.
Dr. Solomon Sobel, until recently director of the FDA’s endocrine drug division, questioned the company’s claim that doctors and patients at last had complied with the monitoring recommendations.
Sobel, a hulking, soft-spoken man, had helped supervise the original fast-track approval of Rezulin. Now he too was telling colleagues that the drug should be withdrawn.
Another mid-level medical officer with responsibility over diabetes drugs, Dr. Saul Malozowski, also joined the Termites in pushing for Rezulin’s withdrawal.
Lumpkin indicated to Warner-Lambert that no regulatory conclusion had been reached. But, Lumpkin made clear afterward, he was unconvinced that withdrawal of Rezulin was warranted. On March 2, one day after the meeting, Lumpkin wrote in an e-mail that his subordinates were relying on “soft hypotheses.”
Lumpkin then left the agency for several days, traveling to the French Riviera for a meeting of the Drug Information Assn., an industry group on whose board he sits.
The Termites feared the FDA would stand pat and that avoidable deaths would continue.
By March 3, Graham had seen and heard enough.
He dropped a bomb: an e-mail addressed to Lumpkin and 13 other FDA officials. Graham’s message could embarrass, if not render untenable, the position of Rezulin’s defenders.
He wrote that Warner-Lambert’s claims of safety, long accepted by senior FDA officials, were “contradicted” by the scientific record.
“There are no existing data anywhere to suggest or support the hypothesis that monthly monitoring can or in fact does prevent drug-induced liver failure,” he declared. “This idea, translated into policy through labeling, is entirely unproven and represents an imagined, artificial hope, not reality.”
“At each juncture in the management of Rezulin’s liver failure risk, hindsight shows that we had little or no effect and that [Warner-Lambert’s] assertions that the liver failure problem was solved, were proved false. . . . The data at hand should persuade us that Rezulin is unsafe compared to other available therapies and that its marketing be stopped.”
Doctor takes on role of whistle-blower
Until early March, the Termites had confined their opposition to within the FDA.
That was about to change.
Misbin, the longtime champion of Rezulin, had concluded that the drug must go.
“I consistently underestimated the rapidity with which Rezulin could damage the liver . . . ,” Misbin recalled in a recent interview. “I have underestimated the virulence of Rezulin.”
Misbin seized the role of whistle-blower. He reached for an audience the FDA could not ignore: Congress.
Misbin wrote to Rep. Henry A. Waxman (D-Los Angeles) and seven other lawmakers. He turned over internal e-mails regarding the agency’s handling of Rezulin. He shared damning correspondence sent to him by a St. Louis physician who conducted early research for Warner-Lambert.
The doctor, Janet B. McGill, alleged that the company “deliberately omitted reports of liver toxicity and misrepresented serious adverse events experienced by [Rezulin] patients in their clinical studies.”
Misbin soon, however, found himself under an FDA internal-affairs investigation for allegedly disseminating confidential agency materials. The inquiry was initiated based on a complaint by Warner-Lambert that “someone had leaked nonpublic information” from agency files, according to Melinda K. Plaisser, an FDA associate commissioner.
On March 13, a senior FDA official warned Misbin:
“You are required to cooperate with the investigation and failure to cooperate may result in disciplinary actions up to and including dismissal from federal service.”
Misbin was undeterred. He refused to answer investigators’ questions unless they were posed in writing.
Another agency medical officer, 72-year-old Dr. Leo Lutwak, also was targeted. Two internal affairs agents asked Lutwak if he had given The Times a Jan. 24 e-mail written by Misbin. After the interrogation, Lutwak said, the agents warned that if his statements were proved to be untruthful he was at risk of imprisonment.
First the defrocking of Gueriguian. Then the admonishing of Graham. Now an investigation of Misbin and Lutwak. For many inside the FDA, the message was unmistakable: Oppose Rezulin at your peril.
FDA briefing yields no pronouncement
The FDA at this point had linked 89 voluntarily reported liver failures, including 61 deaths, to the use of Rezulin.
On March 15, Woodcock and Lumpkin were summoned to the 14th-floor office of FDA Commissioner Henney for a confidential discussion of Rezulin.
On the job since June 1998, Henney had said virtually nothing publicly about Rezulin. Her detachment contrasted with her predecessor, Dr. David A. Kessler, whose activism made him a formidable national figure.
Following the briefing, Henney again made no pronouncement about Rezulin. The drug’s day-to-day fate remained in the hands of Woodcock and Lumpkin. They scheduled another staff meeting to discuss Rezulin.
For the Termites, this was perhaps their last chance.
At 2:30 p.m. on Tuesday, March 21, Woodcock, Lumpkin, Misbin, Graham, Temple, Sobel and nearly a dozen other agency specialists gathered for a round-table discussion at the FDA’s “Woodmont 2” office building in Rockville.
Graham focused on how, in his view, patients incurred increased risk of liver failure the longer they stayed on Rezulin. He estimated that 20 liver failures were occurring each month.
After two hours, Woodcock adjourned the meeting without declaring her or the agency’s position. She then huddled in her office with a handful of subordinates, including Lumpkin and Temple.
The moment reflected more than one drug’s destiny: Lumpkin and Woodcock had captained the FDA’s shift to accelerated approvals and less-adversarial relations with drug companies, a new paradigm epitomized by Rezulin.
By this point, the FDA had overseen the withdrawal of seven medications in 2½ years. The fate of Rezulin posed unique sensitivity: It was the only therapy approved on a fast-track--and by the same FDA officials who were now sitting in final judgment of the drug.
Lumpkin and Woodcock had cited the absence of more withdrawals as evidence that the agency’s faster approvals were not compromising safety. “The steady, if not declining, rate of withdrawals is particularly reassuring,” they wrote in a May 1999 issue of the Journal of the American Medical Association.
All of which meant that, if Lumpkin and Woodcock sought the withdrawal of Rezulin, they risked further discrediting the FDA’s faster, less-adversarial approach.
For Rezulin, the agency had confirmed 63 liver failure deaths by the deliberations of March 21.
Nightfall was approaching. Lumpkin made one last attempt to avert Rezulin’s immediate withdrawal, according to officials familiar with the discussion. He suggested scheduling another meeting with the advisory committee to reassess Rezulin. Temple objected, saying that more delay was unjustified.
The decision was now Woodcock’s. It had been 29 months and 11 days since the FDA received the first reports of liver failure. She phoned executives at Warner-Lambert’s headquarters in Morris Plains, N.J.
By 7:30 p.m., the FDA issued a statement disclosing that the company had agreed to immediately withdraw Rezulin. In the statement, Woodcock observed:
“We are now confident that patients have safer alternatives.”
Times researchers Janet Lundblad in Los Angeles and Sunny Kaplan in Washington contributed to this story. Story originally appeared in the L. A. Times. Copyright, 2000, Los Angeles Times. Reprinted with permission.