COX-2 Inhibitors: Are Patients Any Safer Now?

April 7, 2005
Internal Medicine World Report, April 2005,

COX-2 Inhibitors: Are Patients Any Safer Now?

FDA Advisory Committee Answers—and Raises—Questions

GAITHERSBURG, Maryland—As a general rule, regulatory agencies usually follow the advice of their advisory committees. But how should the FDA act when the experts are split on the issue of whether or not some of the cyclooxygenase (COX)-2 selective nonsteroidal anti-inflammatory drugs (NSAIDs) should be banned? And what does it mean for your individual patient?

Occurring during the widely covered “event,” the split was evident on February 18, 2004, during a series of votes that capped the unprecedented 3-day joint meeting of the FDA’s Arthritis Advisory and Drug Safety and Risk Management Advisory committees.

The assembled experts all agreed that, for COX-2 inhibitors marketed in the United States—celecoxib (Celebrex), rofecoxib (Vioxx), and valdecoxib (Bextra)—evidence is suggestive of increased cardiovascular (CV) risk, even if that evidence is only in select patient populations.

But when asked whether the risk/benefit profile of rofecoxib makes it market-worthy, the committee was nearly evenly split, with 17 saying yes and 15 saying no. Nevertheless, much of the media portrayed this as a majority vote that reflected “a victory” for rofecoxib, even thought an FDA action plan was to follow only some weeks after the committee’s deliberations.

In contrast, John Jenkins, MD, director of the FDA’s Center for Drug Evaluation and Research Office of New Drugs, said that the agency “heard a message” that rofecoxib has a CV risk “possibly larger than the other 2, or better documented perhaps,” and that only a “very narrow margin” of committee members thought that the drug should be on the market.

“We will have to take that under very careful consideration in making decisions about how to proceed,” Dr Jenkins noted.

Asked about the risk/benefit profile of valdecoxib, 17 committee members said that the data support continued marketing (several added “with restrictions”), while 13 voted no, and 2 abstained.

Compare this to the 31 to 1 vote recommending that the risk/benefit profile of celecoxib supports its marketing in the United States. A committee member said that the available data do suggest an increased risk at higher-than-recommended doses of 800 and possibly 400 mg/day, but that he could not find “any evidence” of excess risk at the recommended 200 mg dose.

Committee Recommends Strong Warnings

One clear message emerged from the meeting: the CV risk signal is not exclusive to rofecoxib. “It appears there is a class effect,” said Eric Larson, MD, chair of the board of regents of the American College of Physicians and a drug safety researcher and internist in private practice in Seattle, Wash. “There may be some differences in degree [of CV risk] in terms of different drugs…. Certainly, there is a dose-response effect: the higher the dose, the more likely you are to have that kind of a side effect,” Dr Larson said.

Even the majority of committee members who voted in favor of continued marketing of COX-2 selective NSAIDs recommended issuing strong warnings and restrictions, including “carefully crafted” black box warnings, prescribing restrictions that would assure the lowest possible dose, possible second- or even third-line status for the drugs, and specific contraindications against use of these agents in patients with known increased CV risk.

“It would be a brave man or woman who started a person with a clear history of heart disease in an at-risk group on these drugs without having some strong reason to do that, [such as] failure of other therapy,” said Alastair J. J. Wood, MD, of the Vanderbilt Medical Center, Nashville, Tenn, and committee chair.

Committee members specifically recommended that valdecoxib be contraindicated for cardiac surgery patients based on data from 2 separate investigations suggesting an increased risk in that population.

Clear Ranking Emerges

The vote on rofecoxib (which Merck voluntarily withdrew from the market last September) was particularly poignant in light of statements made at the FDA meeting saying that the company would “consider” the implications of new data suggesting that excess CV risk is a “class effect.” (Should Merck decides to bring the drug back to the market, many new questions would emerge, no doubt.)

Indeed, many committee members thought that the CV risks associated with the COX-2 inhibitors indicate a class effect; however, the magnitude of that risk appears to be different for the different drugs.

“There was a clear ranking of the drugs, and that ranking came out in various ways,” Dr Wood said. “It came out in a [nearly] uniform vote to leave celecoxib on the market, and then a split vote on whether to leave valdecoxib and rofecoxib on the market.”

According to Dr Wood, that vote fits reasonably well with data from randomized trials and observational studies that “clearly” show a CV hazard associated with rofecoxib and valdecoxib, “and if a hazard had been shown with celecoxib, the hazard seemed to be relatively low,” he added.

Thomas Fleming, PhD, chairman of biostatistics, University of Washington, Seattle, noted that with rofecoxib, excesses of CV risk were “strongly seen” in the VIGOR (VIOXX GI Outcomes Research) and APPROVe (Adenomatous Polyp Prevention on VIOXX) trials. In addition, there are data from a placebo-controlled Alzheimer’s disease prevention study showing that, although rofecoxib appeared “neutral” with regard to CV events, a “statistically significant increase” in mortality was evident.

“We have some significant concerns in each of the trials—even the trial that was neutral in cardiovascular events is very unfavorable in mortality,” Dr Fleming said.

Clinical Implications for Intenists

Dr Larson said that COX-2 inhibitors clearly still have a place in clinical practice—but they must be used more judiciously. He recommended that, if a COX-2 is indicated for a particular patient, an internist should first monitor the symptom that is being treated to ensure that the medication is providing symptomatic relief. In addition, keep in mind that given the apparent dose-response relationships observed, the lowest possible dose of a COX-2 inhibitor—or any NSAID for that matter—should be used.

Finally, you must make sure that the patient fully understands the potential risks and benefits of using a COX-2 inhibitor, he added. “In the case of someone getting a COX-2 inhibitor because other [NSAIDs] have not worked, you have to sit down with the patient and say, ‘how much is it worth to you to have something that might work better?’” Dr Larson emphasized.

Related Content:

Publications