It's enough to make your head spin. In September, Merck & Co. (MRK ) pulled its blockbuster painkiller Vioxx from the market because a study linked it to heart attacks and strokes. Another study fingered Celebrex, a similar drug made by rival Pfizer Inc. (PFE ). The Food & Drug Administration quickly came under attack for failing to protect the public from these dangerous drugs. At a three-day FDA advisory committee hearing in late February, 32 outside experts agreed that these relatively new nonsteroidal anti-inflammatory drugs (NSAIDs) do pose serious risks.
But wait. Despite the hazards, the panel also concluded that some older NSAIDs could be just as dangerous as Celebrex -- and that all should stay on the market. The committee even decided that Vioxx, which may have caused thousands of deaths, is useful enough that it shouldn't be banned.
Is this Solomonic wisdom or simply more confusion? Both. The saga starkly illuminates larger underlying problems in drug regulation and use -- and the implications go far beyond painkillers. Here are some of the key insights and issues:
MEDICINES ARE NEVER HARMLESS "Clearly all drugs have risks. That is the price we pay for the benefits," says Dr. Alastair J.J. Wood, associate dean of the Vanderbilt University Medical Center and chair of the advisory meeting. That point has been often forgotten as new drugs appear on the market promising to make life better for tens of millions of Americans. And while the FDA is charged with ensuring that benefits exceed the risks, it's not easy to do. Even when done right, people will be hurt. "Drugs may have a positive risk balance but cause grievous harm," says Dr. Steven Galson, acting director of the FDA's Center for Drug Evaluation & Research.
A classic case: The agency yanked Lotronex, GlaxoSmithKline PLC's (GSK ) drug for irritable bowel syndrome, off the market in 2000 after five deaths -- only to allow limited use two years later when patients demanded it. The thorny question for the Cox-2 painkillers like Vioxx is whether reductions they may offer in stomach and intestinal bleeding, compared with older NSAIDs, outweigh their increased cardiovascular risks.
KNOWLEDGE GAPS MAKE THE TASK HARDER "For most drugs, we know little about how well they work and less about how safe they are," says Dr. David J. Graham, the FDA drug safety official who helped blow the whistle on Vioxx. The FDA's advisory committee was faced with the uncomfortable fact that the risks of the older NSAIDs aren't known either. In fact, epidemiological studies suggest that some of those drugs, such as diclofenac (Cataflam and Voltaren), meloxicam (Mobic), or even over-the-counter ibuprofen (Advil or Motrin), are as dangerous as Celebrex. Putting limits or warnings on Celebrex thus "risks bringing about an enormous migration from this drug to others we don't know much about," warned New York University rheumatologist Dr. Steven Abramson at the hearing.
PATIENTS MISCALCULATE THE ODDS Studies show that people typically overestimate the benefits of drugs and underestimate their risks, especially for heavily advertised medicines. That's why direct-to-consumer ads contribute to the mistaken notion that there is a safe pill for every problem. The FDA's Cox-2 panel called for an end to ads for the drugs.
ANSWERS ARE HARD TO COME BY On Feb. 16, another FDA panel advised putting black-box warnings -- the strongest possible -- on two eczema creams: Novartis' (NVS ) Elidel and Fujisawa's Protopic. Recent monkey studies show that the drugs, which suppress the immune system, may promote lymphoma. So will Elidel, the leading prescription eczema drug in the U.S., cause cancer? We won't know unless researchers follow its users for years or decades.
Rare cancers take years to develop, but at least an increase in cases is relatively easy to spot. Uncovering dangers with Cox-2 painkillers is tougher, because heart attacks and strokes are common. And the increase in risk seen in the studies is small. Often doctors don't consider a safety problem proven unless there is a two- or three-fold increase in a side effect, Dr. Robert J. Temple, FDA's associate director for medical policy, said. But with some of the painkillers, "we're talking about [10%] differences. What do we make of these small differences?"
To pin down the actual risk, a clinical trial would have to be long and huge. Vioxx and Celebrex can't be tested against placebos in arthritis patients, because it's unethical to deny these people pain relief. And testing newer drugs against older NSAIDs is problematic, since their risks and benefits aren't known. "Asking for clinical trials is not the answer," says Arthur Levin, director of the Center for Medical Consumers.
Is there a way out of this mess? Doctors need to better inform patients about risks and benefits, and ads should be more balanced. But part of the solution is getting out the message that regulators don't have all the answers. One benefit of the meeting, says FDA's Dr. John Jenkins, is that "it's good for the public to see how the science evolves and how challenging these decisions can be." Agency watchers also see an increased willingness by the FDA to acknowledge uncertainty by slapping black-box warnings on drugs even when risks aren't proven. "Putting concerns on labels treats physicians and consumers in a more adult fashion," says pediatrician Dr. Richard Gorman.
The next part of the solution is working harder to get answers. The FDA should require companies to collect information on patients receiving medicines and put far more effort into analyzing the data. Beyond that, new genetic technologies offer the promise of being able to identify individuals who will be most helped or hurt by any particular drug. Until then, the best advice comes from Wood. "Hopefully this will make people think about taking any drug," he says. All drugs are potent substances that must be used with respect and caution. The more we take that to heart, the safer we will be.
By John Carey with Kerry Capell in London