Yesterday's announcement came as something of a surprise to regulators because previous large trials of Celebrex did not show worrisome cardiovascular risk. Before the withdrawal of Vioxx, for instance, FDA safety officer David Graham, who has since sharply criticized FDA's handling of safety issues, had found that patients using Celebrex were significantly less likely to develop heart attacks and strokes than those using Vioxx.
"These clinical trial results are new," Pfizer's chairman and chief executive, Hank McKinnell, said in a statement. "The cardiovascular findings in one of the studies are unexpected and not consistent with the reported findings in [a] second study." Results from that ongoing study, to test Celebrex for prevention of colon polyps, have not shown a similar increased rate of cardiovascular disease.
"Pfizer is taking immediate steps to fully understand the results and rapidly communicate new information to regulators, physicians and patients around the world," McKinnell said.
Celebrex has been one of Pfizer's biggest moneymakers, bringing in $1.9 billion in sales last year. Yesterday's news sent stock in the company -- the world's largest drug manufacturer -- down by more than 11 percent.
Many former users of Vioxx switched in October to Celebrex, which Pfizer aggressively advertised this fall as not posing the same risks as its former competitor. The company says more than 27 million Americans have used Celebrex.
Pfizer also got bad news yesterday about Bextra, a newer COX-2 inhibitor that has been used by 7 million people worldwide. In a soon-to-be-published letter released early by the New England Journal of Medicine, three researchers recommended that doctors stop prescribing the drug because of its cardiovascular risks.
"We believe the doubts raised about the safety of [Bextra] constitute a potential imminent hazard to public health and thus require action," three Vanderbilt University School of Medicine researchers wrote.
Two recently published clinical trials of patients who had undergone heart bypass surgery showed that those using Bextra were three times as likely to have serious cardiovascular problems as those who did not take the drug.
Based on those findings, the FDA required an urgent "black box" warning on the Bextra label warning that the drug should not be used in patients who have undergone bypass surgery. But the journal authors said the warning is not enough.
Celebrex, Vioxx and Bextra are the three COX-2 inhibitors that have come onto the market so far, but other related drugs are in drug companies' pipelines. The medications block the action of the COX-2 enzyme, which is associated with inflammation and pain. Researchers found in early studies that the enzyme has heart-protective qualities, and so eliminating or reducing its activity could, in theory, also reduce its beneficial effects.
Sidney Wolfe, director of Public Citizen's Health Research Group and a longtime critic of COX-2 drugs, said yesterday that it is time to take all of them off the market. "The company and the FDA are saying these results with Celebrex are surprising, but they shouldn't be," he said. "There have been warning signs going back to the early animal studies of COX-2s."
John Jenkins, director of the FDA's Office of New Drugs, said the agency did not have any clear indications that Celebrex might cause heart and stroke problems until Thursday night, when Pfizer and the NCI reported that their Celebrex trial was being stopped. He also said COX-2 inhibitors have helped many people because they do not cause the same kind of potentially serious gastrointestinal problems as other painkillers such as aspirin and ibuprofen.
Rejecting criticism that the agency does not have a strong system for identifying potential safety problems with drugs already on the market, Jenkins said the problems with both Vioxx and Celebrex came out during clinical trials designed to learn more about the drugs and their potential for preventing disease.
"It wasn't by chance that these problems came out in these trials," he said. "We can't do studies on patients with arthritis because that wouldn't be ethical."
Also yesterday, the FDA announced that the lung cancer drug Iressa had been shown to be ineffective in prolonging survival of patients with non-small-cell lung cancer in a clinical trial by its manufacturer, AstraZeneca. The drug was approved in 2003 under an accelerated review process that required the company to continue conducting trials after it came on the market. The FDA recommended that patients using Iressa consult with their doctors about further treatment.
