Melbourne, Australia - Prominent cardiologist Dr Marvin Konstam (Tufts University Medical Center, Boston, MA) agreed to be lead author on a 2001 Circulation paper about the COX-2 inhibitor rofecoxib (Vioxx, Merck) [1], which was written in-house by Merck scientists, according to claims made in a federal court in Australia last week [2]. The paper was designed to deflect safety criticisms, some experts believe, following the publication of an article in the Journal of the American Medical Association (JAMA) two months previously that first demonstrated an increase in cardiovascular side effects with the drug [3]. Rofecoxib was not withdrawn from worldwide sale until 2004.
| Dr Marvin Konstam (Source: Tufts-New England Medical Center) |
This is very significant, says Dr Steve Nissen (Cleveland Clinic, Cleveland, Ohio), who was an author on the JAMA paper. "During the three years after publication of the Konstam manuscript, millions of patients around the world were prescribed rofecoxib by physicians who believed that the drug was safe. In this case, a ghostwritten article caused great harm to the public health."
The suggestion that Konstam did not significantly contribute to the Circulation paper is not new—it was most recently reported a year ago in a Boston Globe article by Alice Dembner [4]. The most recent questions about Konstam's role have been raised in Australian media coverage, including a report by Kate Hagan in Melbourne's The Age newspaper. The "ghostwriting" issue comes as the Pharmaceutical Research and Manufacturers of America (PhRMA) has issued "revised principles" on the communication of clinical trial results [5], including restricting authorship on medical manuscripts to individuals who have made "substantial contributions."
Nissen told heartwire: "We must not view this situation as 'old news.' We had long suspected that this [Circulation] manuscript was ghostwritten, but definitive proof was lacking. These court documents finally confirm our suspicions."
| Merck documents aired in lawsuits | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Rofecoxib was an anti-inflammatory launched in 1999; Merck ultimately paid almost $5 billion to settle US cases concerning rofecoxib in 2007. The court case in Australia—a class-action lawsuit against Merck and its Australian subsidiary Merck, Sharp and Dohme—alleges that rofecoxib caused the lead plaintiff's heart attack in December 2003 and that Merck downplayed the risks of cardiovascular problems associated with the drug. More than 1000 Australians are included in the class-action suit. So far, only the plaintiffs' lawyers have presented evidence and Merck counsel have been able to cross-examine; the case is expected to last three months. The allegations regarding the Konstam paper come from the testimony in Melbourne last week of expert Dr George Jelinek (Sir Charles Gairdner Hospital, Perth, Australia), an emergency doctor and former medical journal editor (Emergency Medicine Australasia). In his witness statement, obtained by heartwire, Jelinek asserts: "Konstam made no contribution to the design of the study; [he] played no part in the choice of the internal Merck studies used for the report. The Konstam article was ghostwritten by a Merck employee." Under cross-examination by Merck lawyers, Jelinek said his choice of the word ghostwritten was the wrong terminology, but he stood by his contention that Konstam's role was insufficient for him to be described as an author. "I assert that Konstam was a guest author on a paper written in-house by Merck. In my assessment, the study was already complete before he was involved in the study," Jelinek said. Contacted by heartwire, Konstam—who is director of the cardiovascular center at Tufts and was acting chair of the FDA's cardiovascular and renal drugs advisory committee for the prasugrel hearing in February—said he was "not aware" of the claims made in the Australian court case. He maintains he acted properly: "I take full responsibility for everything I have published." Many of the documents tendered in the court case in Australia are freely available through the University of California, San Francisco (UCSF) Drug Industry Document Archive [6]. Publication choices debated Internal company documents, available on the UCSF database, show that at least one Merck employee—Dr Briggs Morrison—had reservations about the Konstam et al paper. Five of the seven authors of that article were Merck employees, and it was alleged in the Australian courts that Merck scientists prepared the article and then shopped around for a lead author, eventually settling on Konstam. But heartwire has learned that Konstam was not the first choice for lead author of the Circulation article, and many people have expressed concerns about his suitability to author this type of paper—Konstam is a heart-failure doctor and had no particular expertise in meta-analyses or in COX-2 inhibitors. When these questions were put to Konstam by heartwire, he did not address them, responding: "I submitted this paper to Circulation because I felt that a paper considering the cardiovascular effects of an anti-inflammatory agent would be appropriate for that journal." Konstam's paper was published in Circulation after it had been rejected by JAMA for fast-track publication. An internal Merck e-mail dated August 25, 2001 and entitled "Urgent decision needed about CV meta-analysis journal selection" notes the meta-analysis is "not the usual type of trial published by Circulation. Marv Konstam just added to editorial board of Circulation." When the content of this email was put to Konstam by heartwire, he replied saying only: "That's absurd." Also mentioned in the same e-mail is the American Heart Journal, put forward as another possibility for publication. It notes: "Robert Califf from Duke is the editor. It is possible that a call from Konstam could expedite the review process." Collins was first choice Dr Joseph Ross (Mount Sinai School of Medicine, New York, NY), who is the lead author of a paper on ghostwriting published in JAMA last year, told heartwire: "This was a paper we had suspected was ghostwritten." He said he found that a draft of the paper "was originally sent to another investigator before Konstam was involved." That investigator was Dr Rory Collins (Clinical Trial Services Unit, Oxford UK). In an interview with heartwire, Collins said that Merck conducted the meta-analysis and approached him to be an author. He explained that he had attended a consultants' meeting in New York in October 2000 to discuss the Vioxx Gastrointestinal Outcomes Research (VIGOR) trial, results of which gave the first inkling of an increased cardiovascular risk with rofecoxib. Konstam was also at that meeting, along with Dr Robert M Califf (Duke Clinical Research Institute, Durham, NC), Dr Myron L Weisfeldt (Johns Hopkins University Medical School, Baltimore, MD), and a number of Merck employees. "It may well have been me who recommended a meta-analysis be done," Collins told heartwire. "Merck went away and did one; they had the data and they pulled it together and sent it to me for my comments. I recall having conversations about further analysis they might do and gave some advice on interpretation. They asked me if I would like to join them as a coauthor on the paper, but I said I was too busy to make a serious commitment. There was no reason other than that." But a Merck internal e-mail dated March of 2001—the Circulation paper was published in early November 2001—somewhat contradicts Collins's version of events. It says: "Rory [Collins] would not agree to be an author on the paper unless he felt he contributed significantly to the paper." It then discusses reviewing the manuscript before it is dispatched in a package to Collins and says Collins has indicated he will not schedule a teleconference "until he has the background package." There is no mention of Konstam in this e-mail. No documentation on Konstam's involvement prior to summer 2001 In fact, there is no correspondence relating to Konstam's involvement until July 2001, when letters in mid-July from Merck to Konstam and also to Dr Matthew Weir (University of Maryland Hospital, Baltimore), a nephrologist who is the second author on the disputed paper, refer to conference calls to discuss "plans for a rofecoxib CV meta-analysis paper." One of the letters referred to an attached document, stating, "Attached you will find . . . a first draft of the CV meta-analysis publication." Merck spokesperson Ronald Rogers disputes suggestions that Konstam was somehow added to the manuscript late in the game, telling heartwire: "Dr Konstam was in fact very much involved with the design and authorship of the study that bears his name. [His] involvement in the pooled analysis that Circulation published began over a year prior to the publication of the paper [at the] October 2000 Merck consultants' meeting. This, among other things, made him a particularly appropriate author." But neither Merck nor Konstam were able to provide heartwire with any documentation relating to Konstam's involvement with the meta-analysis prior to the summer of 2001, other than a list of consultants who had attended the New York meeting in October 2000. Numerous other pieces of correspondence, obtained by heartwire and available at the UCSF website, point to Merck scientists preparing the initial draft of the paper and first consulting Collins regarding the possibility of him being an author before then approaching Konstam in July 2001. Konstam was involved after this in revisions to the paper, but these appear mainly to have been carried out by Merck staff. Indeed, one email provided to heartwire by Merck, dated August 14, 2001, from Konstam to a Merck employee, relates to Konstam's misspelling of the drug. Konstam writes: "I guess I should learn to spell rofecoxib." "Paper was written to defer safety criticisms" Ross says: "Dr Konstam could distance himself however he wants. He can say that he played major roles in editing the paper and so deserved to be an author, but my major criticism of this practice [is that] the person who makes and writes the first draft makes key decisions about how the paper is written and about what to focus on. You can tell that that paper was written to defer safety criticisms, to deflect them." He adds: "To really discern what has been going on between medical professionals and industry [requires] a tremendous amount of reading between the lines to try to document how guest authorship and ghostwriting take place—where the company runs the trial from start to finish and then identifies authors to get involved."  Konstam is quite prominent in the cardiology community. He's not a small fish."It seems clear that that's what happened with this meta-analysis," Ross continues. "They [Merck] had already made the decision about what the analyses were going to be and what the outcomes were going to be and then they got people involved afterward. Ghostwriting is a major disservice to patients, when you can't trust the medical literature." Ross continues: "What makes this different than just about any [other] person [involved in guest authorship] is that Dr Konstam is quite prominent in the cardiology community. He's not a small fish." Collins begs to differ: Paper was "scientifically sound" Collins maintains the Circulation paper is "scientifically sound" but says if he had authored it he would have changed the wording slightly. "I did reread the paper, about a year or so ago, and I thought if I'd written the paper, I would perhaps have used slightly different language in one particular sentence, to make it clear one can't refute [a possible cardiovascular effect of rofecoxib], but it was a perfectly reasonable description of the results," he said. As to the suitability of Circulation as a vehicle for its publication, he says: "I can see the advantages of JAMA in that you are trying to get the results out to a wide audience, and this drug was used by noncardiologists. . . . a generalist publication would have been preferable. You might ask why JAMA rejected it. Why would they not do an expedited publication?" Contacted by heartwire, Jann Ingmire, director of media relations for JAMA, said that JAMA staff do not comment on papers that have been rejected by the journal. Collins maintains that, at that time, "It was uncertain whether there was any real [cardiovascular] signal [with rofecoxib]. Merck continued to monitor cardiovascular outcomes in their other studies and some years later had more evidence." As to Konstam being the lead author of the paper and the appropriateness of this, Collins said, "I don't know him at all." But he pointed out that Konstam was at the consultants' meeting to discuss the VIGOR trial in October 2000 and added that doctors very often get involved in areas that are not considered their main field: "You do stuff that is interesting, and he may have gotten interested." For example, "I don't have any particular expertise in COX-2," Collins noted. But Nissen is not swayed: "Accountability is now essential. Circulation may want to consider whether to formally withdraw the [Konstam] manuscript. The universities where the academic authors worked may want to consider formal disciplinary actions. This was a tragic example for American medicine. Through ongoing scrutiny and vigilance, we must work to ensure that this type of activity never again endangers our patients." Ross received payment as a consultant at the request of plaintiffs in litigation against Merck & Co related to rofecoxib two to four years ago. In the 2001 Circulation paper, Konstam and Weir report having been paid consultants to Merck. Sources
Your comments
 Family docs and internists mistakenly think they do fine treating high blood pressure May 5, 2009 15:00 EDT CMS on warfarin-responsiveness genetic test: No Medicare coverage except within trials May 5, 2009 09:00 EDT Protocol stressing uninterrupted compressions can improve survival after out-of-hospital cardiac-arrest May 4, 2009 16:00 EDT CTA safe and effective to rule out ACS in emergency room for low- to moderate-risk patients—cheaper too May 4, 2009 12:30 EDT HeartMate 2 gains experience, bid for destination-therapy approval filed with FDA May 4, 2009 11:30 EDT   Previews  | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
