Who Decided to Change the Outcome Variable of the ENHANCE Trial of Ezetimibe?

We have posted before (as have many others, see below) about problems with the ENHANCE trial of ezetimibe (Zetia, by Schering-Plough, and one component of Vytorin, by Merck), and how the trial was designed and implemented seemed meant to increase the likelihood of a favorable result for the sponsors' interests. Particularly controversial was the sponsors' decision to change the definition of the trial's outcome variable after the data was collected. That decision was then reversed.

Today, Matthew Herper, writing in Forbes, described how the decision to change the outcome variable was made. First, the decision was made apparently entirely without any involvement by the ostensible principle investigator of the study.

Forbes has learned that the lead investigator charged with conducting the study, John J. P. Kastelein of the University of Amsterdam, was not even present when the companies made the controversial decision to change its goals, an unusual circumstance in such situations.

It's 'shocking' that Kastelein would not be party to discussion of the ENHANCE trial, says Harlan Krumholz, a cardiologist at Yale University. 'There should be a scientific committee that's independent running a study. He should be taking a leadership role.'

Asked about Kastelein not being present, Lee Davies, a Schering-Plough spokesman, said only, 'I will confirm that this was an independent expert panel.' Kastelein could not be reached for comment Wednesday or Thursday, but in the past he has said he disagreed with the company's decision to change the endpoint of the trial. In an interview with The Wall Street Journal in December, he said he breathed a 'sigh of relief' when it was changed back.

Next, Herper listed the members of the supposedly independent panel.

The list of those on the panel, obtained by Forbes, is filled with well-regarded experts in the field and should satisfy critics left wondering who was behind the recommendation to change the goals. It's not a wide-ranging committee of experts on clinical trials, but instead a group with expertise on studying the arteries with imaging technology. Four are top experts in the field of using ultrasound to take pictures of artery plaque. David Orloff, as a former Food and Drug Administration official, has expertise in exactly what regulatory challenges might be posed by changes to a clinical trial.

The panel: J. Robin Crouse and Gregory W. Evans of Wake Forest University, who conducted a study using Crestor, a rival cholesterol drug from AstraZeneca; David G. Orloff, medical director of Medpace, a company that conducts clinical trials for drug firms and the former head of the FDA division that handled the approvals of Zetia and Vytorin; Michiel L. Bots of the Julius Center for Health Sciences and Primary Care University Medical Center Utrecht, Netherlands, who has worked on similar studies and James H. Stein of the University of Wisconsin, who has conducted studies on the treatment of heart disease in patients with HIV and how various substances affect artery walls.

But is this panel really so independent? Using PubMed, I searched for conflict of interest disclosures made by its members. Guess what - you can see this one coming, can't you -

• J. Robin Crouse - "Dr Crouse reported receiving grant or salary support from Merck, Merck-Schering Plough, Pfizer, AstraZeneca, and Kos Pharmaceuticals; and giving lectures for Merck, Merck-Schering Plough, Pfizer, AstraZeneca, Abbott, and Kos Pharmaceuticals." (1)
• Gregory W. Evans - "Mr Evans reported receiving grant support and honoraria from AstraZeneca, Organon, and Pfizer; and being a consultant to AstraZeneca and Pfizer."(1)
• David G. Orloff - "David G. Orloff, MD, serves as a consultant to Amylin, AstraZeneca Pharmaceuticals LP, Bristol-Myers Squibb, Glaxo-SmithKline, ISIS, Merck & Co., Inc., Novo Nordisk, Pfizer Inc, Roche, Takeda Pharmaceuticals, Vivus, and Wyeth; and owns stock in Abbott Laboratories Inc., Amgen, Genentech, Genzyme, Johnson & Johnson, Eli Lilly, Merck & Co., Inc., Pfizer Inc, and Wyeth."(2)
• Michiel L. Bots - "Dr Bots reported receiving study grants for studies on carotid intima-media thickness and/or honoraria for professional input on carotid intima-media thickness issues from AstraZeneca, Icelandic Heart Foundation, Organon, Pfizer, the Netherlands Heart Foundation, the Netherlands Organisation for Health Research and Development, Servier, and Unilever."(1)
• James H. Stein - " Dr Stein has received research or grant funding from AstraZeneca, Bristol-Myers Squibb, Kos, LipoScience, PreMD, and Sanofi-Aventis, in addition to serving as a consultant to or on the speakers’ bureau for Kos, LipoScience, Merck, Pfizer, Schering-Plough, and Takeda."(3)
  

I presume that all the panel members were paid by Merck or Schering-Plough for their service on it. But all five panel members had other financial arrangements with multiple pharmaceutical companies, and three had other financial arrangements with Merck, or both Merck and Schering-Plough. So this panel may have been independent of the principal investigator of the study, but it did not seem independent of the two corporate sponsors of the study.

Thus, the case of the ENHANCE trial of ezetimibe appears to be an even stronger example of how commercial sponsors of clinical research may try to manipulate studies in ways likely to produce results favorable to their products and vested financial interests. We have posted frequently about cases in which a sponsor manipulated the design, implementation, analysis, or dissemination of clinical research to increase the likelihood that the results would favor the sponsor's product (e.g., recently here and here). One important study showed the contractual basis for such manipulation when studies are done in academic medical environments (see post here). In extreme cases, sponsors have attempted to suppress the research results entirely when they turn out to be unfavorable to their interests (e.g., see recent post here). Such manipulation, of course, is bad for patients and doctors, since it impedes their ability to make decisions about tests, treatments etc based on the best clinical evidence. It also betrays the trust of people who volunteer to participate in research thinking that the goal is to improve health care and advance science.

Although this case is striking, and has received considerable attention in the media and the blogsphere, it has been largely ignored in the health care and medical literature, the dominant source of information for physicians and health care professionals. On the Scientific Misconduct Blog, Dr Aubrey Blumsohn cataloged how extensively this case has been discussed on blogs, listing 35 related posts in the top 100 medical blogs since November. But he could find not a single mention of it in the best known English language medical journals, the British Medical Journal, Lancet, JAMA, or the New England Journal of Medicine. He asked, "One wonders how long our key medical journals will retain any credibility as honest impartial portals for discussion of science and the principles of good medicine." Thus this case also seems to be yet another example of what we have called the anechoic effect, the failure of stories about questionable ethics or bad leadership in health care to create the echoes one might expect (see post here).

See also coverage on the PharmaLot blog, updated here.

References

1. Crouse JR, Raichlen JS, Riley WA, Evans GW, Palmer MK, O'Leary DH, Grobbee DE, Mots ML, for the METEOR Study Group. Effect of rosuvastatin on progression of carotid intima-media thickness in low-risk individuals with subclinical atherosclerosis: The METEOR Trial. JAMA. 2007;297:1344-1353. (Link here.)

2. Orloff DG. Regulatory considerations in the development of high-density lipoprotein therapies. Am J Cardiol 2007; 100[suppl]: 10N-14N.

3. Keevil JG, Cullen MW, Gangnon R, McBride PE, Stein JH. Implications of cardiac risk and low-density lipoprotein cholesterol distributions in the United States for the diagnosis and treatment of dyslipidemia: data from National Health and Nutrition Examination Survey 1999 to 2002. Circulation. 2007;115:1363-1370. (Link here.)