FDA advisory committees recommend marketing of COX-2 inhibitors with 'black box' warning

GAITHERSBURG, MD-Despite unanimity in the belief that the COX-2 selective nonsteroidal anti-inflammatory drugs (NSAIDs) significantly increase cardiovascular risk, FDA's Arthritis and Drug Safety and Risk Management Advisory Committees have advised FDA that the risk/benefit profile of the previously approved COX-2 inhibitors support the marketing of these agents in the United States.

A week after the committees' recommendations, The New York Times reported that 10 of the voting committee members had consulted in recent years with the manufacturers of rofecoxib, celecoxib (Celebrex, Pfizer), or valdecoxib (Bextra, Pfizer). FDA representatives have previously stated that the agency tries to balance expertise with potential conflicts of interest, and that those with knowledge of a specific drug or class of drugs often gain that experience in part through research funded by drug companies.

A potential "class effect" with respect to cardiovascular risk associated with the COX-2 inhibitors was discussed by the advisory panels. The difficulty in assigning a class effect lies in the heterogeneity of these agents, with some being more selective for COX-2 than the others, said Peter Gross, MD, chair of FDA's Drug Safety and Risk Management Advisory Committee. This difference may also account for differing degrees of gastric protection afforded by these agents.

The committees voted 31-1 that the risk/benefit ratio supports continued marketing of celecoxib, with the aforementioned black box warning.

The limited cardiovascular safety database of valdecoxib was cited as a reason for a panel that was divided over the question of its continued marketing-17 members voted that evidence supported continued marketing, 13 voted that it did not, and 2 members abstained. Valdecoxib's prodrug parecoxib was shown in a placebo-controlled trial to increase the incidence of thromboembolic cardiovascular events when used intravenously in patients undergoing coronary artery bypass graft surgery.

Steven L. Shafer, MD, FDA consultant and professor of anesthesia at Stanford University in Palo Alto, Calif, said that the data thus far support an excess cardiovascular risk with valdecoxib only in patients undergoing cardiac revascularization. There is a "level of trespass... in applying data from cardiopulmonary bypass to other populations," Dr Shafer said.

The panel was nearly split in its decision on the marketing of rofecoxib, with 17 voting "yes" and 15 voting "no." In voting "yes" to its availability, Susan M. Manzi, MD, MPH, a member of the Arthritis Advisory Committee and associate professor of medicine at the University of Pittsburgh, noted benefits unique to rofecoxib. Rofecoxib is the only COX-2 inhibitor to be approved for the treatment of juvenile rheumatoid arthritis, it is the only one to demonstrate a reduction in the incidence of gastrointestinal (GI) events compared with other NSAIDs, and it can be used in sulfa-allergic patients.

Several members of the committees voted for availability of rofecoxib only with certain restrictions. "If we did bring [rofecoxib] back, I would only be comfortable with the 12.5-mg dose," said Steven Nissen, MD, a consultant to FDA and medical director of the Cleveland Clinic Cardiovascular Coordinating Center. Dr Nissen also said that the cardiovascular risk appears to increase with increasing doses. Others requested that labeling include language that rofecoxib be considered as an option only after other NSAIDs have failed to provide adequate analgesia.

For celecoxib and valdecoxib, most of the committee members favored removal of the recommended "black box" warnings only if further adequately powered clinical studies of sufficient duration demonstrated cardiovascular safety.

In their risk/benefit assessment, the advisory panels found that the COX-2 inhibitors were no better than nonselective NSAIDs in their anti-inflammatory actions and that only rofecoxib has shown conclusively that it offers superior GI safety, although somewhat less than originally believed.

Byron Cryer, MD, an FDA consultant, said that rofecoxib is associated with a reduction in GI events of about 30%, a figure that depends on the NSAID comparator. "Concomitant use of low-dose aspirin clearly increases the risk and mitigates the potential beneficial effects of COX-2 inhibitors on the GI tract," said Dr Cryer, associate professor of medicine, division of gastroenterology, University of Texas Southwestern Medical School, Dallas.

Dr Graham's presentation on day 2 was much-anticipated given the controversy surrounding his data, which have not yet been published. The epidemiologic data he presented suggest that cardiovascular safety is compromised early with rofecoxib and that higher doses of celecoxib appear to increase the risk of acute myocardial infarction. The database he used to draw his conclusions was the California Medicaid population (Medi-Cal), which covered 2.4 million person-years of NSAID use, including COX-2 inhibitors.

Cardiovascular risk with rofecoxib "begins early in therapy, and is apparent during days 1 to 30 of use," Dr Graham said. Controlled clinical trials of rofecoxib did not expose this risk because they were underpowered and included healthier patients than "real-world" users of the agent, he said. The Medi-Cal database also suggests that celecoxib risk occurs at dosages greater than 200 mg/d.

Many of the nonselective NSAIDs are also associated with excess cardiovascular risk, said Dr Graham, with the biggest culprits appearing to be indomethacin, meloxicam (Mobic, Boehringer Ingelheim), and sulindac. Of 7 epidemiologic studies of naproxen, 4 suggest a cardioprotective effect; however, these 4 studies suffered from major flaws, he said, including selection bias and failure to adjust for confounders.

Meeting chair Alastair Wood, MD, professor of medicine and pharmacology at Vanderbilt University in Nashville, Tenn, said that the advisory committee hearings were likely the first time "that the FDA has had to deal with a drug that has increased the frequency of a common problem, myocardial infarction, unlike something like acute liver failure or torsades de pointes." He defined the risk associated with the COX-2 inhibitors as "a much bigger problem than the FDA has seen with the 16 drugs it has withdrawn."

SOURCE Harris G, Berenson A. 10 voters on panel backing pain pills had industry ties. The New York Times. February 25, 2005.