Key Vioxx Research Was Written by Merck, Documents Allege
TUESDAY April 15, 2008 -- Industry documents reveal that pharmaceutical giant Merck & Co. paid academics to put their names on Vioxx research published in top medical journals while company employees often ghost-wrote the studies on the controversial cox-2 painkiller.
The company apparently also exaggerated the safety of the medication in published clinical trials, and the academics frequently did not disclose industry financial support, the documents allege.
Two articles detailing these findings, the latest episode in the Byzantine tangle surrounding the former best-selling pain reliever, appear in the April 16 issue of the Journal of the American Medical Association.
The scope of the analysis was significant. The researchers combed through 250 documents that included 24 clinical trials, 72 review articles and numerous editorials to come to their conclusions. The Vioxx articles appeared in more than a dozen medical journals, including the New England Journal of Medicine and JAMA.
"These are extraordinary manuscripts, and they reveal the inner workings of the promotion of a drug that ultimately turned out to be a hazard," said Dr. Steven Nissen, chairman of cardiovascular medicine at the Cleveland Clinic.
Nissen, who first warned that Vioxx caused heart problems in 2001, was not involved in either paper but said that he agreed with an editorial in the same issue of the journal that "the only people that need to change to stop this are physicians."
Vioxx was pulled from the market in 2004, after a study showed the drug increased users' risks of heart attack and stroke. And last November, Merck agreed to pay $4.85 billion to settle 27,000 lawsuits from plaintiffs who said they or family members were injured or died after taking Vioxx.
Since the findings of both new studies derive largely from documents associated with the litigation surrounding Vioxx, Merck representatives disputed the assertions.
"These allegations are not true. They're not new. A lot of these allegations showed up in litigation years ago, and our evidence successfully rebutted them," said Kent Jarrell, spokesman for Vioxx litigation on behalf of Merck. "This is an example of courtroom antics masquerading as scientific debate."
"We disagree with both articles," added Jim Fitzpatrick, an attorney with Hughes Hubbard & Reed, outside counsel for Merck. "We think the record of disclosure and transparency is really excellent."
Since Vioxx was removed from the market, there have been numerous allegations that Merck withheld and manipulated safety and efficacy data.
The authors of the first JAMA study, who scoured litigation documents as well as the existing medical literature, say the issue is not really about Merck or about the physicians discussed in the paper, but a much larger and more troubling issue.
"We had the opportunity to examine these documents because of litigation [related to Merck and Vioxx], but this is a practice that has been rumored in the physician research world for decades," said Dr. Joseph S. Ross, lead author of the first study and an instructor in geriatrics at Mount Sinai School of Medicine in New York City. "This is about the profession and how we need to move forward to ensure that research is conducted most rigorously." Ross was a paid consultant for plaintiffs in the litigation against Merck.
This study found that, in many cases, Merck employees prepared manuscripts then, later, recruited academics with little intimate knowledge of the research to be lead or second authors. Disclosures of Merck's financial support of the research and of the author was often not included in the article.
The authors of the second study, from the University of Washington, Seattle, also reviewed legal documents and two published articles, finding that mortality information from clinical trials on cognitive impairment was downplayed. They also reported that there had been no data and safety monitoring board in place for at least one Vioxx trial.
The researchers cited two Vioxx studies, one published in 2004 and one published in 2005, in which the company reported more patient deaths in the Vioxx groups but did not analyze the mortality data and simply said the drug was well-tolerated. However, the company's internal analysis in April 2001 had noted a threefold increase in death risk, although an updated safety report it submitted to the FDA in July 2001 used a method of analysis that "minimized the appearance of any mortality risk."
"It's really not appropriate that you only present the data the way that makes the drug look safe," Nissen said. "Taken together, [these papers] paint a very ugly picture of medicine really at its worst. My own hope is that the sunlight that has shone now on this process will cause people to take a closer look, and I hope people will really read those manuscripts and, when they're approached, do things that are ethical."
"This was amazing sleuthing," Nissen continued. "The thing that you can't lose in all of this, amid the corporate misconduct story, is that patients were harmed. We have a right to have all the information on the table, so we can make intelligent decisions. And, in this case, the information we had was not fair, balanced and scientifically appropriate. It was colored in the process of generating these ghost-written manuscripts and by hiding safety data."
In the accompanying editorial, those at the helm of JAMA called for changes that would affect almost everyone involved in the pursuit of medical research, including tightening the requirements of study authorship, publishing all sources of funding for every study, requiring independent statistical analysis of trial data, creating punishments for anyone who fails to follow the rules, and establishing a complete financial separation between the medical profession and the drug industry.
Just before publication of the JAMA studies on Tuesday, Merck issued the following statement: "Merck is disappointed that we did not have an opportunity to respond to the misleading claims made in these articles prior to their publication. We believe that a full, unbiased evaluation of the Merck papers shows that many of the conclusions put forward by the authors of the JAMA papers are incorrect."
At the insistence of the editors of JAMA, documents used for both articles can be viewed at the University of California, San Francisco and the University of Washington School of Public Health and Community Medicine.
Posted: April 2008