However, higher doses of Celebrex - more than 200 mg a day - do raise the risk of heart attacks, other research shows.
The results, presented before an FDA advisory committee Thursday morning by agency whistleblower David Graham, MD, also suggest that the arthritis drug Vioxx can increase heart attack risk almost as soon as patients begin taking it. This contradicts findings that led to the removal of Vioxx from the market in September 2004.
Celebrex and Vioxx belong to the group of drugs called Cox-2 inhibitors. An expert panel is meeting this week to advise regulators on what to do about evidence that Cox-2 inhibitors and related medications increase the risk of heart attack and stroke. The agency is considering whether to restrict the use of the drugs, require more studies on their safety, or pull some of them from the market.
Maker of Celebrex Reacts
Graham became well known last fall when he testified to lawmakers that FDA officials had suppressed his attempts to warn the public about safety concerns with Vioxx before the drug was pulled from the market.
Preliminary new data from a study Graham conducted among California Medicaid patients showed that patients taking more than 200 mg of Celebrex per day had a 20% higher risk of heart attacks than those not taking the drug. Patients taking 200 mg a day or less showed no increased risk. A second unpublished study found similar results.
"Above 200 mg, we believe there is evidence of increased risk," Graham said.
The results go against at least five other studies, which found no increased risk of heart attacks with Celebrex. Graham said the two latest studies were the only ones to show the effects of higher doses, which are taken by millions of patients. One major National Institutes of Health study was halted several months ago after researchers noticed a raised heart riskraised heart risk with 400 to 800 mg of Celebrex a day.
Officials from Pfizer, which markets Celebrex, told the panel Wednesday that most studies do not show an increased heart risk with their drug and that safety concerns appear unique to Vioxx.
The results have left experts to figure out whether heart problems are unique to Vioxx or whether they apply to the other drugs in the Cox-2 family. Graham said his conclusions suggest that the whole class, which includes Bextra and two other drugs not available in the U.S., have increased risk.
"I believe based on the evidence that there is a Cox-2 effect and that the Cox-2 effect is dose-dependent," he said.
Several committee members expressed doubt about Graham's conclusions, noting that there were problems with his studies, including the possibility that other factors may have increased heart risk.
Ibuprofen and Heart Attacks
The California study also showed an increased risk of heart attack associated with several older anti-inflammatory drugs, including ibuprofen. That drug appeared to raise the users' risk of heart attack by 11%, an increased risk that Graham referred to as "small but significant" since tens of millions of Americans use the drug regularly.
Other NSAIDs, including indomethacin, a drug often used for gout, and Mobic, an increasingly popular drug since the Cox-2 drugs came under fire, also appeared to raise heart attack risk by 40% to 70%, according to the study.
Graham said the data suggest that researchers must take a closer look at older anti-inflammatory drugs to learn more about their possible safety concerns. Regulators should "weed the garden of bad actors," he said.
Early Vioxx Risks?
Merck scientists presented data Wednesday suggesting that heart risks with Vioxx do not show up until the patients have taken them for at least 18 months. The claim backed up Merck's decision to wait until September 2004 to remove Vioxx from the market because a major study had not yet shown a safety concern.
Graham contradicted those claims, showing a separate analysis in which Vioxx appeared to cause heart attacks much earlier in treatment. Several studies in the analysis show a 30% to 50% increased risk of heart attacks within the first 30 days of use.
Merck's study did not find the risk because the early heart attacks are relatively rare and difficult to detect. Risk of heart attack begins when use of Vioxx begins, says Graham.
Merck officials were not immediately available to comment on the study.