USA Today | May 22, 2007
Nine months ago, the drug company Glaxo-SmithKline posted a study along with dozens of others on an obscure company website.
The study indicated that the company’s blockbuster diabetes drug, Avandia, raised patients’ heart disease risk by 30%. At about the same time, company officials say, they told the Food and Drug Administration what they had found.
That apparently wasn’t the first time Glaxo warned the agency about its drug. FDA’s Susan Cruzan confirmed Monday that the company had warned the agency of a potential safety problem at least as early as 2005.
Yet neither the company nor the FDA took additional steps to warn the public until Monday, when a fast-tracked study released on the Internet by a major medical journal prompted the agency to issue a Safety Alert warning of a “potentially significant” excess risk of heart attack and heart-related deaths.
Rep. Henry Waxman, D-Calif., immediately called a June 6 hearing of the House Oversight and Government Reform Committee to examine the FDA’s oversight of Avandia. It unleashed a swarm of questions from doctors and other politicians. In a floor statement, Sen. Chuck Grassley, R-Iowa, asked: “Do we have another Vioxx on our hands?”
The study that prompted the FDA to act linked Avandia to a 43% increase in the risk of heart attack and a 64% increase in the risk of death from all cardiovascular causes. The data were mainly Glaxo’s own, culled from 42 studies large and small, many of them posted on the company’s website.
“This is important because cardiovascular disease is the leading cause of death in diabetes and is responsible for between 65% and 80% of all deaths in diabetics,” says Steven Nissen of the Cleveland Clinic, lead author of the study released Monday by The New England Journal of Medicine.
“The reasons we give drugs to lower blood sugar is to prevent the complications of diabetes, the most important of which is heart disease. When a drug increases that risk, it can have profound public health consequences.”
Taken together, the studies involved more than 27,000 patients. There were 86 heart attacks and 39 deaths from heart disease among Avandia patients; there were 72 heart attacks and 22 deaths among patients not taking the drug.
Doctors in the USA wrote 13 million prescriptions for Avandia last year, generating more than $2 billion in sales, according to the medical research firm IMS Health.
Ronald Krall, Glaxo’s medical director, shot back that Nissen himself had carefully pointed out weaknesses in his analysis, not the least of which is that it mingles data from scores of different studies that weren’t designed to be lumped together.
Krall added that previous clinical trials, a review of data from an ongoing European trial and an analysis of 30,000 patients in a managed-care database found no reason for concern. “We are confident of the safety profile of Avandia and believe in its benefit for type 2 diabetic patients.”
Nissen concedes: “Our study has weaknesses. It was based on an analysis of publicly available material, not patient-level data for these trials. Nevertheless, we think the results will stand up over time.”
Is it Vioxx redux?
If Nissen’s research does stand up, Avandia may become the latest of a series of disgraced blockbusters whose risks have been found to outweigh their benefits. Baby boomers still are reeling from the loss of Vioxx, the infamous arthritis drug withdrawn in September 2004 for heart safety concerns.
Then there’s Pfizer’s $800 million gamble, torceptrapib, an experimental drug that appeared to be medicine’s best hope of clearing clogged arteries because it raised good cholesterol by 60%. Torceptrapib caused heart problems rather than preventing them. And hormone therapy for postmenopausal women caused the heart attacks it was intended to prevent.
But torceptrapib and hormone therapy don’t raise the same questions about drug regulation that have cropped up with Vioxx and Avandia, because regulators apparently knew of potential problems with Vioxx and Avandia long before they took action.
“What did the FDA know and when?” says Yale cardiologist Harlan Krumholz. “That’s an open question. Here, somebody’s gone through data that’s existed for some time. What’s going on at FDA? What’s going on at the company?”
Stuart Seides of Washington Hospital Center echoed Krumholz’s concern. “One has to wonder why the FDA was sitting on this information,” he says, expressing concern that “an excess incidence of cardiac events had been sitting out there for the better part of a year and has not been brought to our attention.”
FDA’s Robert Meyer said Monday that the agency has not completed its review of the data posted on Glaxo’s website last August. “We wanted to do a more robust analysis,” he says. “What we’d like to do is complete our analysis and take this to a public discussion at an advisory committee meeting as soon as possible.”
FDA advisory committees consider a range of drug and device issues, but their opinions are not binding. It would be “a couple of months” before such a meeting could be held, Meyer says.
The agency could post a “black box” warning on the drug’s label — the strongest type of safety alert — or withdraw the drug from the market. In 2006, the FDA added a new warning about a potential increase in heart attacks and heart-related chest pain in patients taking Avandia. The warning stemmed from a trial in patients who already had congestive heart failure.
The new findings shook medical experts because Avandia is so widely used — and because the results challenge one of the pillars of diabetes treatment.
“It’s a shocker because the whole field has been built on the notion that if you increase insulin sensitivity, cardiovascular events would go down,” says Yale diabetes expert Silvio Inzucchi.
Avandia and its chemical cousin Actos, made by Takeda Pharmaceuticals, are central to the approach. They enable the body to make the most of the insulin it produces.
Thought to be safer
That’s critical for the 18 million people in the USA who suffer from type 2 diabetes. Diabetes occurs when the pancreas can’t produce enough insulin. Without insulin, cells can’t absorb the sugar they need for energy. Body tissues begin to break down. That’s why the disease can cause kidney failure, blindness, amputations, heart attacks and strokes.
Older diabetes drugs, such as sulfonylureas, force the pancreas to make more insulin. Another drug, metformin, reduces the amount of sugar produced in the liver.
Unlike older drugs, insulin sensitizers such as Avandia and Actos can be taken by people with kidney problems. They don’t cause dangerous plunges in blood sugar levels associated with some drugs, says University of Miami diabetes expert Ronald Goldberg.
The drugs aren’t without side effects, he says. Avandia and Actos can cause weight gain and swelling. Because they prompt the kidneys to retain water, they’re used cautiously in patients with heart failure, because weakened hearts can’t pump large amounts of fluid.
There’s another issue as well. Although Avandia, approved in 1999, has been shown to reduce blood sugar by about 60%, there’s no evidence to prove that the drug actually prevents heart attacks and strokes. Those studies would take much longer and require many more volunteers.
A study of Actos called PROactive, on the other hand, seemed to indicate the drug may protect the heart, though no one knows exactly how. Based on the results of that study, Yale researchers including Inzucchi are testing Actos to see whether it can prevent strokes. The Cleveland Clinic’s Nissen is carrying out a trial to see whether the drug can reduce artery blockages.
The new findings may, however, sweep Actos into the Avandia controversy. The FDA has asked Takeda to conduct an analysis similar to Nissen’s. “Ultimately, it would be nice to have that sort of data for other drugs as well,” Meyer says. Another member of the class, Rezulin, was pulled off the market in 2000 because of liver toxicity.
Nevertheless, if the Nissen analysis is shown to reveal genuine risks, Avandia “represents a major failure of the drug-use and drug-approval processes in the United States,” Bruce Psaty of the University of Washington and Curt Furberg of Wake Forest University wrote in a journal editorial.
Doctors urged worried patients not to stop taking the drug without talking to their doctors. “People should not panic,” says Seides, noting that just 158 people in the study suffered heart attacks, and only 61 deaths resulted from cardiovascular causes. He said quitting the drug abruptly may be dangerous, too, causing a sudden increase in blood sugar.
“The big question,” he says, “is whether the data here is sufficient for the drug to cease being used at all. That’s a difficult question.”