"Why I Sent It Is a Mystery" - More About Avandia, Conflicts of Interest, and Confused Thinking

We have posted frequently about the controversy about rosiglitazone (Avandia, by GlaxoSmithKline). Some of the distressing aspects of the case were various efforts made to keep information and opinions unfavorable to Avandia away from physicians, patients, and the public, including efforts involving obfuscation, deception, and intimidation. (For example, see this post about the "spinning" of Avandia and the ridiculing of Avandia critics, and this post about attempts to silence an early Avandia critic.)

So it is ironic that the latest Avandia controversy is about the premature release of a manuscript submitted for publication.

Recall how the story began, with the publication in the New England Journal of Medicine by Nissen and Wolski of a meta-analysis focused on cardiac adverse effects of rosiglitazone (Avandia, by GlaxoSmithKline) (see post here). The main achievement of the Nissen and Wolski meta-analysis [Nissen SE, Wolski K. Effects of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med 2007; 356, online here] was to be the first published article to combine data from all relevant clinical trials of rosiglitazone completed to date. Although two major trials of Avandia had been published, its manufacturer, GlaxoSmithKline, had performed many other smaller trials of the drug, most of which have not been published to date. They did eventually appear on a web-site run by GSK. However, this web-site was relatively obscure, and it was not created voluntarily, but in response to a settlement of legal action that alleged GSK had suppressed clinical research about its antidepresant paroxetine (Paxil). (See Steinbrook R. Registration of clinical trials - voluntary of mandatory. N Engl J Med 2004; 351: 1820-1822, link here and our post here). Nissen and Wolski found it, compiled the results of trials on Avandia, and combined their results with those of the few published trials in their meta-analysis.

Now it turns out one of the NEJM peer reviewers of the Nissen and Wolski manuscript leaked it to GSK well before the manuscript was published. From the News section of Nature,

17 days earlier [before the article was published], the reviewer, diabetes researcher Steven Haffner of the University of Texas Health Science Center at San Antonio, had faxed his copy of the article to Alexander Cobitz, a GSK employee whom Haffner knew from working on an earlier clinical trial of the drug.

This was a serious breach of the reviewers' code of conduct. Having reviewed many articles for many medical journals, (but not the NEJM), I can say with conviction that reviewers are forbidden from releasing unpublished manuscripts to anyone, with only one major exception. Reviewers can seek reviewing help from a colleague, as long as this is disclosed to the journal, and the colleague does not further release the manuscript.

As pointed out by an article by Stephanie Saul in the New York Times,

Under The New England Journal’s rules, reviewers are prohibited from disclosing an article’s contents before publication, as a way of protecting the exclusivity of the journal’s material and protecting the intellectual property of scientists who submit articles.

Besides violating The New England Journal’s rules, disclosing a pending article would also be considered a breach of professional ethics, according to Dr. Jerry Avorn, a professor of medicine at Harvard Medical School. Dr. Avorn said that he was not familiar with the specific allegations against Dr. Haffner.

So why did Dr Haffner do it? His explanation of why he sent the manuscript was at best incoherent.

'Why I sent it is a mystery,' Haffner told Nature . 'I don't really understand it. I wasn't feeling well. It was bad judgement.' Haffner says that Cobitz did not ask to see the draft and was 'probably a bystander'.

According to a report in theHeart.org, a GSK spokesperson rationalized Haffner's conduct thus,

Haffner, who consulted for GSK on Avandia, had concerns and questions regarding the methodology of the analysis and sent the article to GSK for advice from company statisticians.

This begs the question of why he did not simply note in his review his lack of expertise on the relevant statistical issues, or seek help from a statistician at the University of Texas.

So perhaps we should seek other explanations. Dr Haffner, it turned out, had a number of financial ties to GSK. As reported in Nature.

Haffner had earlier served on the steering committee of a GSK-sponsored clinical trial of Avandia. He says that he has given many talks for the company, although he declined to say how much he had earned from them. 'I've got a considerable amount of money. I didn't do it to raise my income or anything like that,' he says.

Furthermore, in the NY Times article,

Dr. Haffner, who had been involved in a clinical study that found Avandia worked better at controlling blood sugar than two other treatments, was quoted last year in the online medical publication TheHeart.org criticizing the publication of Dr. Nissen’s study and of editorials that supported it in two other journals.

'The three major medical journals are becoming more like British tabloid newspapers. All they lack is a bare-chested woman on Page 3,' Dr. Haffner was quoted as saying.

In fact, Dr Haffner was one of the co-investigators for the ADOPT study, which compared rosiglitazone to metformin or glyburide, and was sponsored by GSK. [Kahn SE, Haffner SM, Heise MA et al. Glycemic durability of rosiglitazone, metformin, or glyburide monotherapy. N Engl J Med 2006; 355: 2427-43. See link here.] In that article, Haffner disclosed,

Dr. Haffner reports receiving consulting fees from AstraZeneca and Takeda; consulting fees and grant support from Novartis and Pfizer; grant support and consulting and lecture fees from GlaxoSmithKline; consulting and lecture fees from Merck; and lecture fees from Sanofi-Aventis.

Furthermore, an article in the Philadelphia Inquirer noted,

Haffner is a national expert in diabetes. His university Web site calls him 'one of the highest-funded investigators, in terms of [National Institutes of Health] funding, in Health Science Center history.'

And we know that in this age of "greed is good" as the reigning philosophy in medical schools and academic medical centers, the biggest "tax-payers" like Haffner can do no wrong in the eyes of the administrators of these institutions.

So the Avandia story gets more complex, involving more people with more conflicts of interest, and not only suppression of research, but the breach of confidentiality of an unpublished manuscript submitted for publication.

Dr Haffner's nearly nonsensical explanation of why we went running to his fax machine to send the confidential manuscript to his part-time GSK employers is another reminder of the confused reasoning often displayed by people with conflicts of interest, even full professors of medicine who are supposed to be national authorities. As noted before, "people who have conflicts of interest often find giving clear advice (or opinions) particularly difficult." Apparently becoming a consultant for a pharmaceutical company makes one susceptible to faxing confidential manuscripts in a fugue state, something like the sleep-driving pheonomenon reportedly caused by sleep medications like Ambien.

This latest development in the Avandia case is yet another argument that physicians should not only not accept pens, coffee mugs, and free pizzas from companies trying to sell their medical and health care products and services, but that academic physicians should shun all financial ties to these companies. (A physician who wants to work for a pharmaceutical, biotechnology, or device company should do so full-time, or not at all, in my humble opinion.)

Otherwise, prepare to appear on the national news blinking like a deer in the headlights, saying "Why I did it is a mystery."

ADDENDUM (31 January, 2008) - See also this post on the Clinical Psychology and Psychiatry Blog.