Glaxo executives were steaming in early 1999 over the work of biostatistician Michael Elsahoff and other FDA reviewers who had examined the company's new flu drug, Relenza.
The reviewers found that Relenza was no more effective than a placebo in treating common flu symptoms among American patients. The drug showed better results in foreign studies. But the reviewers also found that Relenza, a powdery inhalant, was potentially unsafe for flu patients with asthma or other respiratory disease.
Relenza is not a vaccine. It is designed to be taken within two days of the onset of flu-like symptoms such as fever, cough, sore throat or headache. Relenza may reduce by about one day a patient's symptoms.
The drug underwhelmed members of the FDA's Antiviral Drugs Advisory Committee. It voted, 13 to 4, on Feb. 24, 1999, to reject it.
"There isn't sufficient efficacy to warrant me recommending this drug for my family or myself," said Dr. John D. Hamilton, a professor of medicine at Duke University. Said another committee member, Dr. Sharilyn Stanley of the Texas Health Department: "I have significant concerns."
Glaxo reacted quickly.
Dr. James Palmer, the company's director of medical, regulatory and product strategy, told an FDA administrator in a March 2, 1999, letter that the staff's position on Relenza "is completely at odds with the will of Congress that drug development and approval proceed swiftly and surely." The letter was addressed to Dr. Heidi M. Jolson, director of the FDA's antiviral drugs division.
The Glaxo executive accused the reviewers of "blindsiding" the company. He said one FDA medical officer "exerted considerable and, we believe, misguided and inappropriate influence on the review." He decried the "total silence" at the advisory meeting of another agency physician. He termed Elashoff's analysis "extreme." He said the "advisory process was distinctly biased against fair and open consideration of [Relenza]." A copy of the letter was obtained by The Times.
Elashoff said his superiors told him he would no longer make presentations to the advisory committee. He said he was asked at least five times to delete the anti-Relenza recommendation from his review. He refused. The FDA declined to comment on these matters.
The agency approved Relenza on July 26, 1999. In a memo dated that same day, Jolson provided a mixed assessment.
Relenza, she said, had not been shown effective for patients over 65 or those "with a variety of respiratory, cardiovascular and other medical conditions." She said "special precautions are warranted" if Relenza is used by patients with respiratory disease. These groups would encompass patients most vulnerable to death from the flu. On the other hand, Jolson said, "the totality of the data" suggested that some Relenza patients could expect modest benefit and that their influenza A or B symptoms might improve an average of one day sooner by taking the drug.
The FDA medical officer first assigned to review Relenza, Dr. Barbara Styrt, wrote that "a rationale can be constructed either for non-approval or for approval." She ultimately backed approval, saying that concerns could be addressed through "label language" and later studies. Analysts at Merrill Lynch & Co. predicted the drug would generate sales topping $400 million within four years.
Glaxo, aiming for customers as widespread as the flu itself, last fall placed ads for Relenza on network television. The lighthearted spots featured an actor from the "Seinfeld" sitcom.
Problems emerged quickly.
Following the voluntarily reported deaths of seven Relenza patients, the FDA issued an unusual "public health advisory" to doctors on Jan. 12, 2000, warning of the limited role of Relenza and another recently approved flu-symptom drug. Two of the dead had bacterial infections and should have been treated with antibiotics.
The agency said it had received "several reports of deterioration of respiratory function following inhalation of Relenza in patients with underlying asthma" or another breathing problem. Reports filed through June show that Relenza was cited as a suspect in 22 deaths.
In July, Glaxo issued a warning letter to health professionals, noting "reports of serious respiratory adverse events when Relenza was used in patients with known airways disease." The letter also said that patients with no history of such disease had suffered "decline in respiratory function."
The company added: "Some adverse events have required immediate treatment or hospitalization, and some patients . . . have had fatal outcomes." The Glaxo letter said it was "difficult to determine" whether Relenza caused the deaths.
None of which has surprised Elashoff.
"Even if you accept the company line, that it knocks a day off the flu, a day is not much when you compare it to your life," he said, adding that the approval "was certainly a top-down decision."
Woodcock said the FDA's actions reflect a balancing of risks and benefits. She noted that "strengthened" warning language was added to the product label in April and that a newer study had shown the drug to be effective and safe for children age 7 and older.
Glaxo spokeswoman Ramona DuBose said Relenza "is an important tool for physicians to have to reduce the risk of the disease."
She said the drug can reduce symptoms from both A and B flu strains "by at least a day."
The Glaxo executive wrote the letter in March 1999, DuBose said, to "vehemently protest" the FDA staff's performance at the advisory committee meeting. "We weren't given the opportunity to fully prepare our responses to the FDA's questions."
DuBose said Glaxo would not comment on the volume of reported deaths, adding: "No causal relationship has ever been established between Relenza and any death, anywhere in the world."