Monday, November 20, 2006


One of the innovations in stealth infomercials that pharma has developed is the basic science play. Here’s how it works, illustrated in masterly style by the good people at Corcept Therapeutics.

Since 2001, Corcept has been touting a drug called mifepristone (that’s right, RU486) for treatment of patients with psychotic major depression (PMD). The drug blocks cortisol receptors, and high cortisol was theorized to be a cause of the delusions and hallucinations experienced by patients with PMD.

Corcept got fast track approval for their drug from the FDA on the somewhat specious ground that there are no FDA-approved treatments for PMD. They have published a series of small clinical trials that had negative outcomes. One tiny study (N = 5) had no statistically significant result. A second study of 30 patients had no statistical analyses whatsoever. In another study of 30 patients, the statistical analysis was incorrect. These failures did not stop the Corcept team from talking up the drug’s potential in review articles, book chapters, Continuing Medical Education programs, news reports, and press releases. In quite a few of these promotional “product placements,” the financial ties of the authors with the company were not disclosed.

All that is pretty much par for the course in the academic-industrial complex: clinical key opinion leaders were bought and paid for long ago. Now comes the basic science play. In the August 2006 issue of Journal of Neuroendocrinology, a report describes how Corcept’s drug can reverse the deleterious effect of high glucocorticoid levels on neurogenesis in the hippocampus of rats (1). This finding ties in with some inconclusive evidence that some patients with depression may have reduced hippocampal volume. So, this new basic finding is potentially a big deal for Corcept. Indeed, Corcept provided funding for the study. One fully expects that the company will highlight this new information in future statements.

Looking at the new report, one sees in the Acknowledgements that the first author was supported by Corcept. One also sees that a co-author, E. Ronald de Kloet, failed to disclose his relationship to the company: he is a member of Corcept’s scientific advisory board and, unless he has sold any, the owner of 60,000 shares of Corcept stock. One also sees that this basic science article is careful to follow the company’s marketing message and branding language on the putative efficacy of mifepristone for PMD. For instance, it states, “The glucocorticoid receptor antagonist mifepristone has been shown to rapidly and effectively ameliorate symptoms of psychotic major depression.” These basic scientists also stated, “recent clinical studies have shown that the glucocorticoid-receptor (GR) antagonist mifepristone relieves symptoms of psychotic depression after a remarkably brief treatment period of 4 or 8 days.” None of the cited studies shows anything of the sort. We then read, “… similarly to its clinical efficacy, mifepristone’s effects on adult neurogenesis are rapid and positive, and may therefore be important for its mechanism of action.”

What’s going on here? Could it be that one of Corcept’s basic science consultants has been pulled off the bench to pinch hit for the company? The usual readership of Journal of Neuroendocrinology will know nothing about the clinical trials of the drug and will accept at face value the exaggerated statements of mifepristone’s efficacy in PMD. And now the clinical spokespeople for Corcept can point to a new piece of basic science “validation” of their product. Look for it to be highlighted in the next round of book chapters, review articles in clinical journals, CME events, and press releases.

The company needs pinch hitters right now because they have recently had to announce that two large Phase III trials failed. The house of cards is collapsing: they have been informed by NASDAQ that the company is in danger of being delisted because the share price has gone from $12 at issue in April 2004 to around 80 cents today. All in all, Corcept is a case study in bad outcomes for academic entrepreneurs and their backers. Many investors have been burned. Worse, after going through around $95 million of venture capital and IPO proceeds, not to mention several million dollars of NIMH funding, they have failed to answer the original question: does mifepristone help seriously ill patients with PMD accompanied by elevated cortisol production? Now, as the company enters what may be its final spasm of activity, comes this new corruption of the basic medical science literature.

We have long known that citations of the medical literature are not holy writ, but now they are becoming advertising copy. As George Winokur, former chair of psychiatry at the University of Iowa, liked to joke, the medical literature is like the Bible: people can find in it whatever they are looking for. Some of us learned in our training that when we cite an article in one of our manuscripts, we need to actually read that article, understand its methods, and agree with its conclusions. Had the authors of this stealth infomercial in Journal of Neuroendocrinology scrutinized the clinical trials reports that they cited, even as basic scientists they would have recognized the weakness of the evidence. Thus does the corporate mandate to put lipstick on the pig vitiate even the basic medical science literature.


(1) Brief Treatment With the Glucocorticoid Receptor Antagonist Mifepristone Normalises the Corticosterone-Induced Reduction of Adult Hippocampal Neurogenesis. By: Mayer, J. L.; Klumpers, L.; Maslam, S.; de Kloet, E. R.; Joëls, M.; Lucassen, P. J.. Journal of Neuroendocrinology, Aug2006, Vol. 18 Issue 8, p629-631, 3p, 1 graph; DOI: 10.1111/j.1365-2826.2006.01455.x; (AN 21447609)

No comments: