Oct. 6, 2006, Bloomberg
Harvard Researcher Forced Bayer to Disclose Drug Toxicity Study
By Justin Blum and Eva von Schaper
Bayer AG‚s disclosure last week of a study showing that a promising medicine has deadly side effects came only after a Harvard drug safety expert told U.S. regulators that the research existed.
Alexander Walker, a professor at Harvard‘s School of Public Health, says he alerted the U.S. Food and Drug Administration of a 67,000-patient study he helped conduct that analyzed the safety of Bayer‚s Trasylol, used to control bleeding during heart bypass surgery. The study confirmed an earlier trial showing that the Leverkusen, Germany-based company‘s drug increased the risk of kidney failure, heart attacks and strokes.
The FDA is investigating why Bayer, Germany‚s largest drugmaker, didn‘t provide the results for a Sept. 21 agency meeting reviewing the drug‚s safety. Bayer said on Sept. 29 that its failure to alert the FDA was a `„mistake.‘‚ The admission underscores the need for new rules on how companies release study data about medicines already on the market, doctors say.
“It calls into question the honesty of Bayer and the honesty of the pharmaceutical industry in general,‘‚ says John Teerlink, director of the heart failure clinic at the San Francisco Veterans Affairs Medical Center.
„I think the public health has been harmed in two ways,‘‚ said Teerlink, who is a member of the FDA panel that met to review Trasylol. “One, we didn‘t have complete information to make our decision. But secondly, it calls into question a process that all of us depend on.‚‘
In a letter to the panel members on Sept. 29, Karen Weiss, deputy director of the agency‚s office of oncology drugs, said the FDA may call for a second meeting of the advisory panel to review the new data.
`Heavy-Handed‘
Bayer shares fell 12 cents to close at 40.62 euros in Frankfurt. The stock has increased by 54 cents since Sept. 29 when the company announced reporting mistake. The shares are up 15 percent for the year
„This was an extremely heavy-handed action,‚‘ said Alexander Groschke, an analyst at Landesbank Rheinland Pfalz in Mainz, in an interview Oct 2. Groschke said Bayer‚s forecast late last year that Trasylol someday may generate 500 million euros ($634.6 million) annually is “an illusion now.‘‚
Silke Stegemann, also a Landesbank analyst, wrote in a Oct. 2 note to investors that „this problem could draw class action suits.‘‚
The FDA had convened the advisory panel of doctors and other experts after a report published earlier this year showed the drug had a higher risk of toxicity than other treatments. The FDA says it also called the panel together because the number of deaths that may be linked to Trasylol rose to 10 last year from four in 2004.
Sales Decline
Trasylol was approved for U.S. marketing in 1993 to reduce the need for blood transfusions in bypass surgeries. About 4.3 million patients have been given the injectable drug, according to Bayer, which says the treatment generated about $293 million in sales last year, making it the company‘s 11th largest-selling therapy. Revenue declined by 25 percent in the first half of 2006 as compared to the year-earlier period, according to Bayer.
Without seeing the large study‚s data, the 19-member advisory panel recommended to the FDA that Bayer didn‘t have to strengthen a warning to doctors about the drug. The panel also questioned the findings of the side effect report published in the New England Journal of Medicine in February that had led to concerns about the drug and the reduction in sales.
Michael Lincoff, a panel member and vice chairman for cardiovascular research at the Cleveland Clinic in Ohio, said this week the withheld study, if analyzed properly, probably would have played a critical role in the panel‚s decision because it was larger than the other studies the company presented.
Another Outcome?
Had the study‘s findings „been divulged to the panel I believe there could have been another outcome to the panel hearing,‚‘ said Steven Findlay, a Washington-based health-care analyst for Consumers Union and a panel member.
Walker, previously head of Harvard‚s department of epidemiology and a former consultant to the New England Journal of Medicine, contacted the FDA after seeing no mention of his study during the panel meeting Sept. 21, according to Susan Bro, an FDA spokeswoman. Walker helped run the study for i3 Drug Safety, a private clinical research company owned by Minnetonka, Minnesota-based UnitedHealth Group Inc., the largest U.S. health insurer.
Notifying the FDA „seemed like the right thing to do,‘‚ Walker, also an employee of i3, said in a telephone interview Oct. 4. Walker said he also contacted Bayer about the failure to disclose the study, though he declined to say when.
`Regret Our Failure‘
The drugmaker says it confirmed the study‚s existence and its preliminary findings in a phone call to the FDA on the morning of Sept. 27, which is after Walker communicated his concerns to the company and the agency. Two days later, Bayer made its first public disclosure of the study.
“We told the FDA that it was obviously a mistake on our part and we regret our failure to act,‘‚ said Staci Gouveia, a spokeswoman in Bayer‘s U.S. unit based in West Haven, Connecticut.
Bayer says it began the study in June. It was a review of a large database of patients undergoing bypass operations, comparing the side effects of several drugs including Trasylol, used to control excess bleeding, the drugmaker said. The company received a preliminary copy of i3‚s report on Sept. 14 and submitted questions aimed at getting clarifications on the data on Sept. 18, according to company officials in the U.S. and Germany.
Bayer employees who presented information to the FDA panel, which included top research executives, didn‘t know the study results were available, said Gouveia, Bayer‚s spokeswoman. She said Bayer is still analyzing the results and the company believes the drug is safe.
This week‘s British Medical Journal said drugmaker-sponsored studies often come to more favorable conclusions about experimental medicines than independent tests and should be read with caution.
A comparison of separate studies done on the same treatments found that industry-backed drug trials were more likely to support use of the new product, while non-affiliated researchers were more cautious, according to co-author Anders Jorgensen at the Nordic Cochrane Center in Copenhagen.
Regulators have increased pressure on the makers of pharmaceuticals and medical devices to do wider tests and disclose more data to protect the health of patients. Merck & Co. was forced to recall its Vioxx painkiller in 2004 because of increased heart disease.