FDA Officials Delayed, Discounted Study on Safety Risks of Vioxx, Agency Researcher Says
Dr. David Graham, associate director for science in the Office of Drug Safety at FDA, alleged in a "series of testy e-mail exchanges" that agency officials "delayed and demeaned" the results of a study he conducted that found Vioxx -- a COX-2 inhibitor that Merck withdrew from the market in September over safety concerns -- can increase patient risk for cardiovascular problems, the New York Times reports. In an Oct. 25 e-mail, Graham said that Dr. Anne Trontell, one of his supervisors, referred to the study as "nothing more than a scientific rumor." The "exchanges are part of a stream of e-mails suggesting a tense environment" in the office, the Times reports.
In his notes from an Oct. 29 meeting with Dr. Paul Seligman, director of the office, Graham wrote that he, Dr. Andrew Mosholder -- an FDA researcher whom supervisors would not allow to testify at a public meeting in February on a stidy he conducted that found most antidepressants increase suicidality in children -- and others in the office maintained "the review and clearance process had been turned into a battleground, full of contention and intimidation because our managers, the people who fill out our performance evaluations, had created a system where it was taking a great risk to stand firm in our scientific beliefs."
In addition, according to the notes, Seligman said in the meeting that he had prepared a letter of resignation he planned to submit on Tuesday because he had concluded "the staff don't trust me" (Harris, New York Times, 11/4).
For his study, Graham and colleagues examined the records of 1.39 million Kaiser Permanente members, 26,748 of whom took Vioxx and 40,405 of whom took Celebrex, a COX-2 inhibitor manufactured by Pfizer. The study found that high doses of Vioxx, or rofecoxib, tripled patient risk for heart attack and sudden cardiac death.
Graham planned to present the study at an epidemiology conference on Aug. 25, but his supervisors said that the results were "too preliminary" and recommended that he first submit them to a medical journal for peer review or present them at the conference with an alternative FDA opinion.
On Aug. 25, the abstract of the study said that "this and other studies cast serious doubt on the safety of rofecoxib" at doses higher than 25 milligrams "and its use by physicians and patients." However, when Graham submitted a revised, final version of the study to FDA on Sept. 30, the agency announcement of the release of the results did not mention specific data on cardiovascular risks related to Vioxx.
FDA on Tuesday published a memo based on the study that said Vioxx might have contributed to an estimated 27,785 heart attacks and sudden cardiac deaths between 1999 and 2003 (California Healthline, 11/3). FDA posted the memo after the Wall Street Journal published parts of the study. Graham has submitted the study to the Lancet for peer review.
Graham wrote in an Oct. 26 e-mail, "For all the center claims in its operating principles that respect for others is a core value, my experience with rofecoxib was just the opposite." Dr. David Campen, a co-author of the study, added, "There were a number of e-mails that were generated within FDA that were really an effort to delay publication of our study."
In addition, Mosholder recently requested the removal of his name from a textbook chapter on risk management "because under the current climate, he's afraid to do anything that will set him up for retaliation or adverse disciplinary action," according to notes that Graham took at the Oct. 29 meeting.
However, according to Steve Galson, director of the Center for Drug Evaluation and Research at FDA, the e-mail exchanges between Graham and his supervisors are part of peer scientific review and are "a totally normal process that takes place all over the government dozens of times a day."
In related news, Graham said that he hopes to begin a study on the safety of two other COX-2 inhibitors, Bextra and Mobic. The study would use data collected from the California Medicaid program.
FDA officials have said that they have "no convincing evidence" that Bextra or the COX-2 inhibitor Celebrex are unsafe, but some independent researchers have said that all COX-2 inhibitors might increase patient risk for cardiovascular problems. FDA will hold an advisory panel meeting on the issue in January (New York Times, 11/4).