Thanks as so often to Dr. Roy Poses over at Health Care Renewal--
--we are alerted to the recent publication by S. Swaroop Vedula and colleagues:
These authors were part of the team that testified in lawsuits brought against Pfizer related to the drug Neurontin (gabapentin). As a result they had access to a trove of internal e-mails and documents. Here they report their analysis of these documents, from two separate legal actions in 2004 and 2008, providing a longitudinal account of how the company approached publication of clinical trial data on gabapentin for off-label uses. (The drug was approved by the FDA for two relatively narrow indications, epilepsy and post-herpetic neuralgia;and the company wanted to extend sales by showing that it worked also for migraine, bipolar disorders, neuropathic pain, and nociceptive pain.)
First the authors documented: "For each of the four off-label indications we included in our study, a document titled ‘marketing assessment,’ designed to examine the financial impact of seeking FDA approval for a new indication versus other methods of increasing sales for the indication, preceded clinical trials sponsored by Pfizer ..." That is, before the company or independent academic scientists had a go at designing and conducting a clinical trial, the marketing people already had their innings and decided on the overall strategy. In this instance, the key decision was not to try to gather data that would allow an aplication to the FDA to expand the approved indications for Neurontin, but to find data that would support off-label prescribing for that usage. And that key decision was made not based on the scientific results of clinical trials, but before the clinical trials were even begun. A critical difference is that if the company had approached the FDA, they would have been legally required to hand over all study results, both favoring the drug and not; while by going the off-label route, they could cherry-pick what they published, as we'll come to. (Reminder: a company cannot legally market a drug for off-label use directly, but its physician "key opinion leaders" are free to recommend off-label use so long as it seems to be at arm's length from the company. Today, drug reps can also hand out trial report reprints that support off-label use, hence the need to publish trials that are favorable.)
The other main reason, besides more control over publication, was that the patent on Neurontin was about the expire and it made no business sense for the company to expand its use just as the drug was about to go generic. In one instance, the decision as to publication of negative results seems to have been driven by the form's "evergreening" strategy, which was to roll out the "new" drug pregabalin as a me-too drug to replace gabapentin as their expensive, heavily-marketed brand-name drug. As pregabalin was about to be launched, the company may have wanted negative trials published about Neurontin, so as to lay the groundwork for later being able to claim that pregabalin was really head and shoulders better than its predecessor.
The next major finding from the authors was, "According to the internal company documents, 'affiliate-driven manuscripts' were written for Pfizer ... by [Medical Action Communications, a medical communication company] and sent to the authors for approval. Each article was coordinated by a manuscript team, consisting of representatives from the medical and marketing divisions of the company." This suggests that ghostwriting--having paid writers spin the research papers the way the company wanted them, and only then letting the scientific "authors" see the drafts of the papers--was simply standard operating procedure within the company. The entire process showed that Pfizer, and especially its marketing division, kept everyone involved in the publication of a paper on a short leash.
Finally, the authors demonstrated that the journals in which papers were published, and the timing of publication, were also chosen for maximum marketing impact--for instance, with positive results published in high-impact journals and published quickly, and with negative results showing up in low-circulation journals years later.
As the authors summarize: "In this study we also observed that publication occurred in journals with higher or lower circulations related to statistical significance of findings; delay in publishing statistically non-significant findings; tailoring of publication content to reflect key marketing messages; adding spin to scientific publications such that conclusions favoring gabapentin were emphasized and conclusions that did not favor gabapentin were explained away; and indicators of ghost authorship. Each form of bias, and spin, on its own, could be seen as a relatively minor issue. The value of our findings is in the overall picture that emerges from what appears to be the simultaneous use of many forms of reporting bias and spin, all within the context of a pharmaceutical company’s publication strategy, implemented for marketing purposes."
So-- it's all about the marketing, and not about the science. No big surprise here, but further documentation of what we've had reason to believe from all previous reports of this type.
Now, I sort of hate to keep beating up on editor Jeffrey Drazen of the New England Journal:
--who as previously reported said we should not judge articles based on who sponsored the research but on the quality as depicted in the published paper. What Dr. Vedula and colleagues show us is that if a paper that's positive about a new drug shows up in Dr. Drazen's high-impact journal, we can be reasonably sure that there are 3-4 other studies with much less impressive, or even negative results, which are kept aside to be not published at all, or else published in much-lower-circulation journals many years later--or at least that's the standard business plan within the industry. And of course there's no way on this green earth that even the most fastidious reader of NEJM could ever know that. So much for being able to read a paper in the journal and then draw your own conclusions.