FDA Conclave Will Assess
Cardiac Risks of Painkillers

By ANNA WILDE MATHEWS and SCOTT HENSLEY
Staff Reporters of THE WALL STREET JOURNAL
February 7, 2005; Page B1
Subscribe now to The Wall Street Journal and get up to 8 weeks FREE

Medical advisers to the Food and Drug Administration will gather next week in a meeting that could result in tougher warnings or other restrictions on some widely used painkillers.

The FDA advisory committee will confront a medical mystery: What is it about certain painkillers that may increase the risk of heart attacks and strokes? Last fall, Merck & Co. withdrew its painkiller Vioxx because of such a risk, but the FDA must decide what to do about medications still being sold and some not yet approved in the U.S.

To develop recommendations, an advisory committee will sift through information offering mixed evidence on whether popular drugs -- such as Pfizer Inc.'s Celebrex and Bextra and Bayer AG's Aleve -- raise cardiovascular risks beyond benefits of relief from pain and other complications.
RELATED ARTICLES

• Merck Documents Shed Light on Vioxx Legal Battles
 
• Novartis Weighs Cap on Prexige Use
 

The three-day meeting will feature the scientific version of a courtroom clash, with drug makers laying out their own theories on Vioxx's drawbacks -- and why they don't apply to their products. The advisory panel could issue recommendations ranging from beefed-up warnings on the drugs to an outright ban on Celebrex and Bextra -- though that outcome is considered unlikely.

One prominent theory about the drugs is advanced by University of Pennsylvania professor Garret FitzGerald, who is expected to speak at the invitation of the agency. He argues that all the Cox-2 inhibitors, a class that includes Vioxx, Celebrex, Bextra and two drugs not approved in the U.S., Merck's Arcoxia and Novartis AG's Prexige, share characteristics that raise risks.

Cox-2 inhibitors, designed to be easier on the stomach than conventional pain pills, block an enzyme called Cox-2, which is associated with inflammation, but don't block another enzyme called Cox-1.

That, Dr. FitzGerald believes, might create an imbalance of certain substances in the blood, raising the risk of clots, and thus of heart attacks and strokes in patients with cardiovascular risk factors.

If Dr. FitzGerald is right, drugs such as ibuprofen and naproxen, which kill pain by damping both enzymes, shouldn't pose a greater risk of clotting. But Cox-2-only blockers would be implicated. FDA documents released Friday highlight the agency's concerns about cardiovascular results and Arcoxia, but say that studies haven't "definitively answered" how Prexige compared with older painkillers.

John LaMattina, president of Pfizer's global research and development, says, "The bottom line is that there are a lot of interesting theories...and nobody really understands why" some Cox-2 medicines cause cardiovascular problems and others don't. The Pfizer drugs, he says, are safe to use for arthritis pain. "In terms of cardiovascular safety, there's no difference between Celebrex, Bextra and NSAIDs [nonsteroidal anti-inflammatory drugs]," he says, citing a new analysis of company studies of the drugs.

Two other theories about why Vioxx might cause a problem let the remaining Cox-2 medicines off the hook to some degree. One theory is that the higher cardiovascular risks are the result of an old enemy: higher blood pressure.

There is evidence that many painkillers, older ones that affect both Cox-1 and Cox-2, and the newer Cox-2 drugs, tend to elevate blood pressure. But an article in the Archives of Internal Medicine last month, and supported by the marketers of Celebrex, showed that Vioxx had a stronger, more sustained effect on blood pressure than Celebrex and naproxen. "The outlier among the agents so far has been Vioxx," says University of Connecticut professor William White, an author of the study and a Pfizer consultant.

A third theory links increased risks to the molecular structure of some of the drugs. In a paper published last year in the journal Atherosclerosis, researchers wrote that Vioxx and Arcoxia may lead to inflammation of the lining of blood vessels that could raise the risk of heart attacks or strokes. This theory holds that the problem is the chemical structure of the two drugs, which isn't shared by Celebrex, Bextra or older painkillers.

"It's a very compelling mechanism, given the chemical properties," says R. Preston Mason, a Harvard biochemist and an author of the Atherosclerosis study. Mr. Mason, who does consulting work for Pfizer, says he "doesn't see a strong case with the other Cox-2 inhibitors."

The problem for the FDA committee is that each theory has its limits. "None of the theories will account for all of the data," says Gurkirpal Singh, an adjunct clinical professor at Stanford University.

For instance, Dr. FitzGerald's thesis doesn't explain why at least one study raised the possibility that naproxen, which isn't a Cox-2 drug, also might be linked to at least a small risk of cardiovascular problems. Bayer and Roche Holdings AG, which sells a prescription version of naproxen, say that their data show that naproxen more likely protects the heart, rather than raises risks.

Another challenge, raised by Pfizer in materials submitted to the committee, emerges from the trial that prompted Vioxx's removal from the market. That study allowed people at risk of heart problems to take aspirin. Because aspirin blocks both Cox-1 and Cox-2, it should cancel out any risk caused by the suppression of Cox-2 alone. But the patients taking aspirin showed the same problems as those taking only Vioxx, a finding "inconsistent" with Dr. FitzGerald's thesis, Pfizer wrote.

To underscore its point, Pfizer did a new analysis of its own data from dozens of clinical trials and found that Celebrex and Bextra posed no increased risk of heart attacks and related side effects compared with traditional painkillers.

The Pfizer examination didn't include two trials testing the drug for prevention of colon cancer because they involved using high doses for something other than arthritis pain, a Pfizer spokeswoman said. Both trials were stopped in December, after a safety review of one of them found more than double the risk of cardiovascular problems in patients taking Celebrex. The second trial found no significant increase.

Pfizer also analyzed 19 studies of Bextra for chronic pain and found the drug's serious cardiovascular risks weren't statistically different from those of a placebo or older painkillers. But theories shifting the focus from Cox-2 inhibitors also have limits: The study showing Vioxx's risks raised questions about the blood-pressure thesis,but Merck said the trial didn't find "any important association" between blood pressure and cardiovascular problems.

Subscribe now to The Wall Street Journal and get up to 8 weeks FREE