Friday, August 22, 2008

The ADVANTAGE Revelations: Seeding Caught in the Act

Apologies to faithful readers for my slowness in posting the big news of the week (that darn day job again). The Annals of Internal Medicine made rather a big deal of their publication of Merck company documents showing that the ADVANTAGE trial of 2003, comparing gastrointestinal side effects in Vioxx (rofecoxib) vs. naproxen, was really a seeding trial. You might wonder why the big deal, given that the phenomenon of the seeding trial (a clinical trial whose real purpose is to stimulate many practitioners to prescribe a new drug, rather than to answer any scientifically pertinent question) has been widely described previously, including in HOOKED. Was it merely a mea culpa because Annals was hoodwinked into publishing ADVANTAGE in the first place? However, a closer look shows that it is indeed valuable to have the smoking gun laid out for inspection.

Merck did the usual sputtering when questioned by the media. According to Bloomberg (, Jonathan Edelman, Merck scientific affairs director, said: "As with all Merck clinical research, there is a commercial interest ... The documents included in this article include an example of the marketing use for the data. The allegation by the authors that the primary goal of the study was marketing is simply wrong.''

Let's see how the original documents brought to light in the Annals article (all resulting from the Vioxx litigation, and seen initially by the authors because of their role as paid consultants to the plaintiffs in that litigation) stack up against this Merck protest. A key memo dated 1/4/99 was a nomination of the designers of ADVANTAGE for the company's Best Physician Program Award. This is one of the marketing annual awards and the people nominated are all from the Marketing division. The memo goes on to justify the nomination by saying: "The trial was designed and executed in the spirit of the Merck marketing principles... The sales force nominated potential investigators and completed intake forms, allowing a very large number of sites to be evaluated and enrolled and ensuring equal distribution of investigators across the business groups. ... Finally, the results of the trial are being carefully tracked. An analysis performed at 6 months post launch demonstrated a significantly higher level of prescribing for VIOXX among primary care ADVANTAGE investigators compared to a control group..."

Okay, let's see if I've got it so far. The investigators for the trial were chosen by the sales force. The main goal was to insure that these investigators represented as broad as possible a selection from those to whom the company looked for its later profits. When the company set out to analyze the data from the trial, they did not look at the way Vioxx performed against naproxen in terms of stomach symptoms; they looked at how many Vioxx prescriptions the investigators wrote. So far no mention at all of anyone or anything related to the scientific trials division of Merck.

ADVANTAGE was designed so that each "investigator" enrolled around 6 subjects in the trial, meaning that 600 doctors were eventually recruited to involve 5557 patient/subjects. That might seem an inefficient way to run a scientific trial. But it's a great way to assure that more docs get used to writing prescriptions for Vioxx.

Mr. Edelman also insisted to Bloomberg that the study was conducted by the medical and scientific affairs unit, not the marketing unit. The next original document laid out for our inspection is a slide from a presentation that describes the role of Merck's Clinical Development Program, a part of the marketing division, in ADVANTAGE:
  • Design protocol and oversee execution of trial
  • Select investigator sites
  • Run investigator meetings
  • Choose and manage [contract research organization]
  • Perform data analysis
  • Prepare publications
All right, I give up. Exactly what is left over for the "scientific" division to do, after Marketing has done all of this? It walks like a seeding trial, it looks like a seeding trial, it quacks like a seeding trial...

The first author of ADVANTAGE was Dr. Jeffrey R. Lisse of the University of Arizona. Dr. Lisse had already felt heat (as explained in HOOKED) when it was shown several years ago that ADVANTAGE conveniently concealed a few cases of coronary disease that occurred among the Vioxx subjects. Dr. Lisse, contacted by the New York Times, was then forced to admit that he had had nothing to do with the data analysis or the writing of the first draft of the trial report, and that this was the first time he had heard about any cases suspicious for coronary disease. That is, he had to admit that the ADVANTAGE trial report was ghostwritten. When Bloomberg tried to contact Dr. Lisse for comments about these new seeding-trial revelations, they discovered that he was unavailable and on medical leave.

So in the end, the protestations of Merck seem pretty lame when laid aside the documents revealed. I have not checked out all the industry-friendly blogs, but I would imagine that the company's next line of attack will be to accuse the authors of an unacceptable conflict of interest (funny--sounds familiar) because they were paid by the plaintiff's attorneys who were suing Merck. That is indeed a problem (though the paper makes clear that a rigorous method was followed in analyzing the documents). The solution to the problem is for the drug industry to be more open and forthcoming with this sort of documentation. If they did not do such a good job of hiding these internal documents from scrutiny, then independent investigators who were not part of a lawsuit might someday get to see them.

Lisse JR, Perlman M, Johansson G, et al. ADVANTAGE Study Group. Gastrointestinal tolerability and effectiveness of rofecoxib versus naproxen in the treatment of osteoarthritis: a randomized, controlled trial. Ann Intern Med 139:539-46, 2003.

Hill KP, Ross JS, Egilman DS, Krumholz HM. The ADVANTAGE seeding trial: a review of internal documents. Ann Intern Med 149:521-58, 2008.

Sox HC, Rennie D. Seeding trials: just say "no." [editorial] Ann Intern Med 149:279-80, 2008.

1 comment:

Dr.M.Scott said...

It’s interesting to see that even a boring topic like medicine can capture such media space and have so many words devoted to it. To quote an instance we can mention chantix. Its one of those drugs that almost represents the whole concept of the quit smoking drugs, or rather publicity has made it that. Irrespective of the fact how chantix works or how effective it is it has managed to gain a fair amount of attention and make way to the people’s homes. Thanks to publicity or rather publicity agenda superstardom for a drug was an unthought of concept a few years back.