How a Drug Approved By the FDA Turned Into a Lethal Failure — Despite Early Doubts, Duract Carried a Mild Warning; Then, Some Patients Died — `A Lot of Wishful Thinking’
The Wall Street Journal
Rochelle Sharpe
WASHINGTON — When the Food and Drug Administration approved the potent painkiller Duract last year, Rudolph Widmark couldn’t stop worrying.
Dr. Widmark, an FDA medical officer, had already analyzed mounds of safety data on the medicine from its manufacturer, Wyeth-Ayerst Laboratories. In his judgment, Duract caused more liver-cell damage than any other similar medication available.
“We have seen a `liver flag’ that can be only fully explored through responsible marketing of the drug,” he warned in a memo to his boss. Later, in a meeting with FDA and company officials, he advocated that the agency put its most stringent warning on the drug.
But other experts, including prominent liver specialists hired by Wyeth, had made persuasive arguments that if taken properly, Duract posed no special hazard for short-term use. So the FDA allowed Duract to hit the market in July 1997, with a mild warning about potential dangers printed in tiny type in the 19th paragraph of the 56-paragraph label that accompanied the drug’s packaging.
Just 11 months later, the FDA and Wyeth, a division of American Home Products Corp., Madison, N.J., conceded that they had got it wrong. After doctors linked Duract to four deaths and eight cases in which patients required liver transplants, Wyeth pulled the drug off the market last June.
Duract thus became the fourth drug in just 12 months to be thus removed — a record number for any one-year period. It was the second major debacle for Wyeth, which also marketed Redux, the antiobesity drug withdrawn in September 1997 after it was linked to heart-valve problems. And it came as the FDA struggles to balance two competing mandates: its traditional role of protecting public health and its more-recent effort to ensure that valuable new remedies aren’t delayed by red tape, a legacy of the controversy over the slow approval of early AIDS drugs.
But the rise and fall of Duract isn’t a tale of the dangers of rushing new science to market. The FDA spent 2 1/2 years evaluating it. Rather, it’s a story about the pitfalls of the drug-approval process. Whether done quickly or slowly, judging the safety of new drugs is an inexact process that doesn’t stop when the FDA approval is issued and people start taking pills. In Duract’s case, despite the lengthy screening process, it still was the public that conducted the final trial that led to the drug’s undoing.
“We put the strongest possible emphasis on drug safety,” says Philip de Vane, Wyeth’s vice president of clinical affairs. The serious side effects that led to Duract’s withdrawal occurred in only one of 10,000 or 20,000 patients who took the drug — and then only among people who took it for longer than the label recommended. The deadly side effects weren’t detected in clinical trials, Dr. de Vane says, adding: “They couldn’t have been detected.”
From early on, however, the FDA’s Dr. Widmark says he suspected that something was wrong with Duract. A clinical pathologist and rheumatologist, Dr. Widmark retired from his job as an FDA medical officer last year, at age 73, after he broke his ankle. He says that during his years at the agency, he viewed his mission to be shepherding company officials through the agency’s rigorous approval process. Often, he says, that meant convincing them of the limitations of their proposed products.
“A lot of wishful thinking goes into this,” he says.
Wyeth-Ayerst’s application for Duract, submitted in late 1994, embodied ambitious hopes. The company initially was positioning its new drug to capture the huge osteoarthritis market, encompassing about 40 million people. The company proposed a label that said the analgesic could be used long term to treat arthritis or short term to alleviate surgical or menstrual pain.
Dr. Widmark says he was impressed by clinical data showing the drug’s potency. It appeared to be a “fantastic analgesic,” he says, which acted quickly and was just as effective as some narcotics — without being addictive.
The question was, would Duract have dangerous side effects? Dr. Widmark had developed his own method for detecting problems; it involved elaborate tables that plotted two sets of patient test results against each other. He requested that Wyeth compile such data for Duract. As soon as he looked at the results, he says, he saw trouble.
Of 830 patients who had taken Duract in clinical trials, 23 had unusually high levels of certain liver enzymes that indicated their liver cells were being destroyed too quickly. Four patients were losing cells eight times faster than normal. In December 1995, Dr. Widmark wrote his first safety report on Duract, concluding that the drug caused more liver-cell damage than any other analgesic of its kind.
The FDA asked Wyeth-Ayerst to submit a new label for Duract, eliminating any suggestions that the drug could treat chronic pain, like arthritis. The agency said Duract should be allowed in patients for only a two-week period.
Such a restrictive label would cut into Duract’s market potential. Wyeth readily agreed to drop references to chronic pain, but to contest Dr. Widmark’s other conclusions, the company hired two of the world’s foremost experts on liver disease.
One of Wyeth’s advocates was Hyman Zimmerman, now 84, who had literally written the book drug-induced liver damage: “Hepatotoxicity,” a reference book used around the world. For many years, he was chairman of one of the FDA advisory committees that help the agency grapple with complicated medical questions. He and Willis Maddrey, also hired by Wyeth-Ayerst to defend Duract, were known for their work on proving Tylenol’s potential to cause liver problems.
Dr. Zimmerman commanded respect from top FDA officials. Michael Weintraub, 58, acting director of the FDA division responsible for Duract, says he liked to tell people that he still frequently quoted a Zimmerman article written in the year he was born. Dr. Weintraub was best known for his 1992 study that launched the “fen-phen” diet drug craze. His research showed that taking a two-pill combination of fenfluramine and phentermine could promote weight loss, but fenfluramine, also produced by Wyeth, was subsequently pulled from the market after evidence emerged that it caused heart-valve damage. He has since left the FDA.
During the next seven months, Wyeth-Ayerst sent 46 memos and reports to the FDA about Duract. Then, company officials, the company’s liver experts and agency regulators convened for a critical meeting. For 1 1/2 hours, the scientists pored over the clinical data and had “a very vigorous debate,” recalls Dr. Maddrey, one of Wyeth’s experts. The discussion soon focused on how long patients could safely take Duract as the FDA’s in-house expert, Dr. Widmark, challenged the scientists hired by Wyeth.
Dr. Widmark, citing evidence that Duract destroyed liver cells, argued that the drug should be approved only for use of 14 days or less, and suggested that the drug carry the FDA’s most severe safety warning, a “black box.” Such a warning would highlight the drug’s dangers, literally inside a black box, in the label’s first paragraph. Dr. Widmark says he worried that doctors wouldn’t notice Duract’s usage limits buried in the label. Since there already were about 20 similar drugs on the market that didn’t cause liver problems, he feared that doctors wouldn’t bother reading the fine print.
But Dr. Zimmerman says he disputed Dr. Widmark’s interpretation of the data on the grounds that abnormally high liver enzymes don’t necessarily predict that the drug will cause liver problems, he says. Noting that no one in clinical trials had developed liver disease, Wyeth officials said Duract caused no liver problems for at least 30 days. They proposed that the label state it was for short-term use, and that it tell doctors to monitor patients for liver abnormalities after 30 days, an FDA memo states.
Dr. Weintraub, who chaired the meeting, wanted a specific definition for “short term” and says he focused much of the meeting on that issue. Wyeth disagreed, saying it was more appropriate to list types of short-term painful conditions that Duract could treat, like tooth extractions and sprains.
The debate over label wording continued after the meeting. As discussions dragged on, Dr. Widmark was asked to write a memo about the drug to his boss’s boss, Murray Lumpkin, the FDA’s deputy director for drug evaluation. Dr. Lumpkin didn’t respond to repeated requests for comment.
In that three-page memo, Dr. Widmark defended his work and disagreed with Wyeth’s labeling proposal. The suggestion that doctors monitor patients for liver problems after 30 days, he said, was a ploy that allowed the company to avoid liability for problems. If a patient dies from liver problems after taking Duract for 30 days, he wrote, “the physician will be sued. Wyeth-Ayerst will be in the clear, because `it is in the label.’” He urged Dr. Lumpkin to “to make the right decision in this dispute.”
Wyeth’s Dr. de Vane declined to discuss Dr. Widmark’s memo, but says he viewed the label simply as a “vehicle for rational prescribing,” based on data that supported the drug’s approval. Wyeth officials say they viewed the lengthy discussions about Duract as normal negotiations.
Later, Wyeth officials met with Dr. Lumpkin and Janet Woodcock, director of the agency’s Center for Drug Evaluation and Research. The FDA leaders insisted that short term be defined in the label as “generally less than 10 days.” In July 1997, the agency approved Duract for sale.
For a brief time, it looked like Wyeth had a success. Sales took off, with $29 million in Duract revenue in the third quarter of 1997, as Wyeth advertised Duract in medical journals as an alternative to addictive painkillers such as Percocet. And for all the labor that FDA and Wyeth officials had invested in the language of Duract’s label, the reality was that few people read the fine print.
“Only three parties read the whole label: the company, the FDA, and the lawyers,” says Dr. Widmark. “No physician has the time.”
Duract’s label certainly didn’t affect the way Martin Nicholosi, of Brooklyn, N.Y., took the pain pill. The 49-year-old carpenter said his doctor gave him a prescription for 200 pills — 190 over the recommended limit — after he complained about severe headaches. Mr. Nicholosi, who is suing Wyeth as well as his doctor in U.S. District Court in New York, said the Duract helped his headaches, but he became nauseated, lost weight and grew increasingly tired. He says that only after his doctor gave him a second prescription for 360 pills did he learn about the label warnings from his pharmacist, John Cocchiara.
Wyeth declined to comment on the case. Mr. Nicholosi’s doctor, Publius Martins, said in a faxed response to questions: “Based on clinical evaluation, the medication in question was determined to be the most appropriate choice to treat his condition. This medication, at the time it was prescribed, met all prevailing medical standards as na FDA-approved drug.”
Mr. Cocchiara, who works at a Thrifty Drugs store in Brooklyn, says he became concerned about Duract because a number of his patients handed him prescriptions for six months’ worth of the drug. Since he had read the label, Mr. Cocchiara knew what he was supposed to ask: Are you having your liver enzymes monitored? The reply, he says, was “No, what do you mean?”
Mr. Cocchiara says he also challenged a Wyeth-Ayerst saleswoman who visited his store about the dangers of long-term Duract use and the medical problems that arose with a similar drug, Toradol. “`You can use it long term,’” he recalls her saying.
Mr. Cocchiara concluded that physicians in his area had gotten the same message, because within days of the saleswoman’s visits, the pharmacist got so many prescriptions for Duract that he had to increase his inventory. “Most of the physicians are educated by their salesmen . . . unfortunately,” Mr. Cocchiara says.
Dr. de Vane, Wyeth’s vice president, says he can’t comment on the specific incident Mr. Cocchiara describes. But as a matter of company policy, he says, “our sales force is trained to promote the drug within the approved labeling.” As for comparing Duract to Percocet in advertising, Dr. de Vane says that is appropriate because clinical trials had compared the two medications. “We’re only able in any promotional material to refer to data that exist,” he says.
Within two months of Duract’s approval, the existing data on Duract began to change. Some patients’ livers started malfunctioning.
As the problems started to emerge, Wyeth’s difficulties with Redux were escalating. In July, the same month Duract was released, a Mayo Clinic study showed that Redux might cause heart-valve problems. In September, the company withdrew the drug from the market.
That December, the first liver transplant occurred on a patient who was taking Duract. The 36-year-old patient had been taking the drug for osteoarthritis — a condition the label said the drug wasn’t meant to treat — and for two months, much longer than the label suggested.
In January, two more patients needed new livers: a 40-year-old woman who had been taking Duract for three months for her osteoarthritis and a 46-year-old woman who had taken it for three months for her shoulder pain.
At Wyeth-Ayerst, company officials began weekly conversations with FDA officials as soon as they heard about the transplants, Dr. de Vane says. “The postmarketing surveillance system worked,” he says.
By February, the FDA asked Wyeth to issue a new label for Duract, this time with a black-box warning. The company sent letters to 600,000 doctors, notifying them of the change.
After the new label came out, more doctors started limiting prescriptions to 10 days, the company said. But the problems didn’t stop.
Two patients taking Duract died in February, after the label was changed — one woman had taken it for six months for rheumatoid arthritis, and another had taken it for three months for back pain and osteoporosis. (She had received a bulk prescription for the drug, a common practice among health plans that find it a cheaper way of dispensing medication, an FDA official says.)
March brought a report of a third death. When a fourth patient died in April, the FDA and Wyeth executives had another meeting. Dr. de Vane says the May session focused mainly on the impact of putting the black box on the label, but added that withdrawing the drug may have come up during discussions.
On June 22, the company agreed to pull Duract from the market. A Wyeth spokesman wouldn’t disclose the financial impact of Duract’s withdrawal. American Home is in the midst of a planned merger with Monsanto Co., St. Louis. Duract’s withdrawal was in everyone’s best interests, says Dr. de Vane. “You can’t limit patient use,” he says.
The company didn’t change any practices because of Duract, he says, adding, “We continue to follow the strictest procedures.”
The major lesson he learned from the Duract experience: “The process worked,” he says. “The drug was extensively studied. It was extensively reviewed. And once an event that was so rare it couldn’t be detected in premarketing occurred, the response was appropriate.”