News of the Week

Massive Trial of Celebrex Seeks to Settle Safety Concerns

Science  23 Dec 2005:
Vol. 310, Issue 5756, pp. 1890a-1891a
DOI: 10.1126/science.310.5756.1890a

Since the COX-2 inhibitor Vioxx was yanked off the market more than a year ago, the remaining anti-inflammatory painkillers have been under a cloud of suspicion. Which are the safest, the least likely to contribute to heart attacks and strokes? And which are the most dangerous?

Pfizer, maker of the COX-2 inhibitors Celebrex and Bextra (which was pulled in April), is placing a $100 million bet on a 20,000-person, international trial led by the Cleveland Clinic in Ohio. But some experts are concerned that the design of the trial, announced last week, could load the dice in Celebrex's favor and put patients at risk. European Union (E.U.) countries have declined to participate because of their concerns about Celebrex's safety.

Three-way race.

Pfizer is putting up at least $100 million for elebrex to take on naproxen (above, right) and ibuprofen.


The clinical trial is unusual for focusing on patients with heart disease, including those who recently underwent bypass surgery and those at risk of cardiac problems. The approach is meant to mirror conditions in the real world. “If you have arthritis and you have heart disease, we can't ask you to tolerate the pain. So what do I give you?” says Steven Nissen of the Cleveland Clinic, who's leading the trial. “In the absence of knowledge, we're just guessing.” Nissen has criticized Vioxx and other COX-2 drugs, although at a U.S. Food and Drug Administration (FDA) meeting last February, he voted to keep Bextra on the market.

Patients in the Celebrex trial will be randomly and blindly assigned to receive either Celebrex or one of two older anti-inflammatory drugs—ibuprofen or naproxen. The trial will end after 715 “events”—heart attacks, strokes, or deaths—have occurred, says Nissen. That's expected to take roughly 4 years.

But some scientists wonder whether the study will really resolve questions about the drug's safety. “The important thing in science is to make sure you've controlled all your variables,” says Alastair Wood, a drug-safety expert and associate dean of Vanderbilt University School of Medicine in Nashville, Tennessee. “Here, there's another variable in the room that potentially could affect some of the outcomes.”

That variable is aspirin, used by heart disease and at-risk patients to reduce clotting. Previous trials have often excluded those on aspirin, which will be given in low doses to all the volunteers in the Pfizer trial because they're at higher risk.

The catch, says Garret FitzGerald, a pharmacologist and cardiologist at the University of Pennsylvania, is that aspirin reduces clotting by acting on COX-1. That's one of the molecules targeted by ibuprofen and naproxen, but mostly ignored by Celebrex. Previous studies in animals and humans have suggested that both ibuprofen and naproxen, but not COX-2 inhibitors, “can interfere to undermine the cardiovascular protection of aspirin,” says FitzGerald. If so, a finding that heart attacks and strokes are the same in all three drug groups might actually mean that Celebrex is less safe, because the cardiovascular benefits of aspirin may be decreased for those taking ibuprofen or naproxen but not for those in the Celebrex group.

The solution, say both FitzGerald and Wood, is to banish aspirin from the study and give patients clopidagrel, or Plavix, a more expensive drug made by Bristol-Myers Squibb that has cardiovascular benefits similar to aspirin but doesn't work through COX molecules. Nissen disputed that approach in an e-mail, noting that clinically, chronic clopidagrel use isn't indicated for heart disease patients, and its effects are not known. He also said the interaction between aspirin and ibuprofen remains speculative.

The ethics of the new trial are also getting mixed reviews. Although some clinical trials are faulted for relying on the healthiest patients, this one has garnered criticism for planning to enroll the sickest. “Why take the highest-risk people?” asks Curt Furberg, an epidemiologist at Wake Forest University School of Medicine in Winston-Salem, North Carolina, who suggests instead tracking them through health databases of hundreds of thousands of patients like the one kept by Kaiser Permanente. The E.U. will not participate because its drug regulatory agency contraindicates Celebrex for heart disease patients.

Still, “the trouble in the real world is that people have multiple illnesses,” says oncologist Richard Goldberg of the University of North Carolina, Chapel Hill, whose ongoing trial of whether Celebrex could prevent colon polyps ground to a halt earlier this year after Vioxx was pulled for lack of new recruits. It “wouldn't surprise me” if this latest study faces the same problem, he says.

But Nissen's Cleveland Clinic colleague Eric Topol, who is not involved in the study, isn't worried. “It'll recruit very quickly,” he predicts. “It's not like you're doing a trial to hurt anybody.”

Navigate This Article