The New York Times
February 17, 2005
By GARDINER HARRIS
GAITHERSBURG, Md., Feb. 16 - A panel of federal drug advisers who gathered here on Wednesday to judge the safety of the heavy-selling pain pills Celebrex, Bextra and Vioxx were tough, sometimes even caustic, in their remarks about the drugs, suggesting that the panel may take a hard line on their uses.
One panel member accused Pfizer of hiding data. Another asked a Pfizer researcher if she would take Bextra if she had heart risks. Dr. Karen Seibert, the Pfizer researcher, declined to answer. Dr. Vibeke Strand, an arthritis specialist from Stanford University who consults for Pfizer, then stepped to the microphone.
"Would I recommend that a patient with high cardiovascular risk take these products?" she asked. "I think the answer would be no."
The panel was convened by the Food and Drug Administration after Merck announced in September that it would withdraw Vioxx from the market because a study had found that the medicine doubled the risks of heart attacks and strokes.
In the weeks after the announcement, Pfizer insisted that it had no evidence that its very similar drugs, Celebrex and Bextra, caused heart problems. Pfizer has since disclosed that it had, at the time of those statements, studies that indeed demonstrated heart problems among patients taking Celebrex or Bextra.
In December, the National Institutes of Health announced that a very large trial had found that high doses of Celebrex more than tripled the risk of heart problems.
The committee has been charged with telling the Food and Drug Administration what to do. The options include forcing the withdrawal of Celebrex and Bextra or strengthening the label warnings. Few people believe that Pfizer will be forced to withdraw the drugs, but tough warnings on labels could severely hurt sales. Sales are already about half what they were in the fall.
Pfizer contends that most studies demonstrate that its drugs are safe.
Wednesday was the first of three days of hearings at the Hilton here, and it was largely taken up by presentations from Merck and Pfizer.
The meeting was mobbed with so many spectators and interested parties that the Food and Drug Administration set up an overflow room that itself overflowed. The hotel eventually had clusters of people scattered all over its lobby, gathered around closed-circuit televisions. Representatives of drug regulatory agencies from all over the world were here.
Dr. Ned Braunstein, a senior director at Merck Research Laboratories, said Merck now believed that the cardiovascular risks seen in Vioxx are shared by "other drugs in the class." The class, COX-2 inhibitors, includes Vioxx, Celebrex and Bextra, as well as Prexige, an experimental drug from Novartis, and Arcoxia, an experimental drug from Merck.
Dr. Kenneth Verburg, a Pfizer vice president, gave lengthy presentations about the safety of Celebrex and Bextra.
A panel member, Dr. Curt Furberg of Wake Forest University, responded, "I'm troubled by some inconsistencies that I have found in the briefing document from Pfizer."
Dr. Furberg suggested that all of Pfizer's mistakes seemed to benefit the company. "So I wonder how much trust can we put in these presentations," he said.
Such a charge is unusual in the genteel confines of advisory committee meetings.
Pfizer said it would recheck its numbers.
Dr. Alastair Wood of Vanderbilt University, the chairman of the committee, criticized Dr. Verburg's talk for its exclusion of the results of some trials that found increased risks of heart problems for Celebrex.
On another issue, another panel member, Dr. Byron Cryer of the University of Texas Southwestern Medical Center, described part of the Pfizer presentation as "misleading."
Dr. Verburg responded simply, "Point taken."
Committee members were also deeply critical of Merck. Dr. Wood noted that after more than a year of contentious negotiations with the F.D.A., the company in 2002 placed on the physician prescribing information for Vioxx a statement that caution should be used in giving Vioxx to patients with heart disease.
"Tell me what I'm supposed to do with that information?" Dr. Wood asked. "Do I tell patients to take it slowly, or without the lights on?"
Dr. Braunstein said doctors and their patients needed to take the information into account when choosing therapies.
He acknowledged that an early Vioxx study had found that patients taking it had more than four times as many heart attacks as those taking naproxen, another common pain pill. But he said the company believed that the difference might have resulted because naproxen protected against heart problems in a manner similar to aspirin.
Dr. Steven Nissen, a cardiologist at the Cleveland Clinic, dismissed this explanation. He noted that aspirin reduced the risk of heart attacks by about 20 percent, a fraction of the suggested protection of naproxen.
"Naproxen would have to be far better than aspirin" at preventing heart problems for this explanation to make sense, Dr. Nissen said.
Dr. Braunstein explained that the number of heart attacks in the study was small, potentially confounding the results.
Dr. Nissen was not convinced.
"This story about naproxen doesn't stand up to any kind of scientific rigor," he said.
A medical officer for the Food and Drug Administration gave a detailed account of the agency's analyses of the mountain of information that Merck submitted on Vioxx. "My goal is to point out that we were not sleeping behind the wheel," the officer, Dr. Lourdes Villalba, said.
Dr. Garret A. FitzGerald of the University of Pennsylvania described to the committee his years of research to try to uncover how COX-2 drugs affected body chemicals that determine the "stickiness" of blood and, thus, the likelihood that someone taking them might experience a heart attack, stroke or similar problem.
Dr. FitzGerald tried to undermine many of the arguments Pfizer has made to defend the safety of Celebrex and Bextra. He said all drugs in the COX-2 class caused heart problems.
"It seems to me most rational people would accept a class-based mechanism, as they did for efficacy," Dr. FitzGerald said.
His statement referred to Pfizer's longtime contention that all COX-2 drugs protected against ulcers, an important point because Pfizer has little evidence that Celebrex and Bextra are safer on the stomach than older pain pills but has long insisted that the drugs had differing effects on the heart.
Dr. FitzGerald also said that simply reducing the doses of Celebrex or Bextra would not be effective in eliminating the risks of the drugs.
Two speakers presented data showing that Celebrex, known generically as celecoxib, was no better at inhibiting the COX-2 enzyme than diclofenac, also known as Voltaren, and might even be slightly worse. Diclofenac is now sold as a generic drug for pennies a pill and is one of the most widely used pain medications in the world; Celebrex can often cost $3 a pill.
One slide presented by Dr. FitzGerald stated simply, "diclofenac = celecoxib."
In an interview after his presentation, Dr. FitzGerald said he believed that the only way to safeguard against the risks was to insist that patients be allowed to take the drugs only for short-term therapy, a few weeks at most. If more research suggests that longer therapy is safe, that prohibition could be lifted, he said.
