May 15, 2006
NEJM sat on Vioxx article warnings for years: WSJ
The New England Journal of Medicine had ample warning of misleading claims it published about a Vioxx clinical trial but waited years before acting on them, The Wall Street Journal reported. Less than a year after publishing an article touting the benefits of Vioxx, editors at the NEJM received warning of possible flaws in the November 2000 piece, including a call-in from a pharmacist to a radio show on which the NEJM’s top editor was featured. During the call-in, the pharmacist told the editor, Jeffrey Drazen, about several extra heart attacks not reported in the article but that had occurred during clinical trials. She also submitted a letter to the NEJM about the missing data. The letter went unpublished, even as the article brought in at least $697,000 in revenue on 900,000 reprints sold—most of them to Merck for use as detail aids. A negative letter would have dampened their impact. Drazen told The Wall Street Journal that the article authors, who included Merck employees, were at fault for not fully reporting their data. Merck said the heart attacks occurred after a predetermined trial cut-off date. Other alarm bells sounded, including an article in The Journal of the American Medical Association on Vioxx’s cardiovascular risk and a memo posted to the FDA’s Web site questioning the NEJM article’s theory that naproxen had a protective effect on the heart. These came well before NEJM issued a repudiation of the article in December 2005. The repudiation corrected the mistakes. But the “expression of concern” may have been timed to divert attention from a potentially damaging deposition given by Executive Editor Gregory Curfman in the Merck Vioxx litigation. In it he admitted NEJM’s lax editing of the Vioxx article. Internal e-mails suggest editors feared it could have been played during the trial, bringing attention to NEJM’s role in the Vioxx affair. NEJM says its decision to release the repudiation at that time was tied to Curfman’s discovery that two of the Merck authors knew of the extra heart attacks in advance of the November 2000 article. Vioxx was pulled in September 2004, after a three-year study showed it doubled the risk of heart attack and strokes in patients taking it for at least 18 months. Merck faces more than 11,000 lawsuits associated with the drug. In related news, data released last week by Merck show that Vioxx patients may have remained at high risk of heart attacks or strokes for some time after they stopped taking the drug. Merck said that patients who took Vioxx in the APPROVE study had no greater risk of heart attacks or strokes a year after stopping the medication than those who got placebos. But Steven Nissen, a cardiologist at the Cleveland Clinic and an early Vioxx critic, disputed that interpretation, saying the patients’ relative risk persists during the year following discontinuation. There were 28 heart attacks or strokes in the Vioxx group compared with 16 in the placebo group. That means Vioxx patients were 1.7 times more likely to have a heart attack or stroke than those who took a placebo—numbers that are not statistically significant.