Back in ancient history when I first began trying my hand at deciphering the bizarre clinical trial situation in psychiatry, I looked into the strangest of stories involving a drug called Mifepristone [Corlux®][RU-486]. For the details, see my series of eight consecutive posts starting with Mifepristone 0: $tanford’s $chatzberg redux…. I was relatively new to the amount of corruption in the world of psychopharmacology, and I recall writing that series incredulous at such blatant Conflicts of Interests. The essence of the story is that Dr. Schatzberg, Chairman at Stanford, was the principle Investigator on an NIMH grant to study the effect of RU-486 [the "morning after" pill] on patients with psychotic depression.
The premise was reasonable. These patients had been shown to have hypercortisolemia, and RU-486 blocks the effects of cortisol [among other things]. What was not reasonable was that Dr. Shatzberg and Stanford got a patent on RU-486. Then Dr. Schatzberg and his associate Dr. Joseph Belanoff formed a private company, Corcept, to explore getting FDA approved with plans to capitalize on the drug. It was a bolder time then and the NIMH Study and Schatzberg’s commercial planning were virtually indistinguishable until 2008 when Senator Grassley began to look into Conflicts of Interest and unreported income in psychiatrists in high places [Dr. Schatzberg was in the highest of places – President of the American Psychiatric Association]. In the course of that investigation, he "stepped down" as chairman, and moved off of the Board of Corcept, the company he’d founded with Belanoff – though he remained an investor and advisor. In that same time frame, we learned that he and Dr. Charlie Nemeroff has guest authored a textbook, Recognition and Treatment of Psychiatric Disorders: A Psychopharmacology Handbook for Primary Care, ghost-written by Scientific Therapeutics Incorporated, financed by GSK.
But this is a story that can’t seem to find a way to end. While Corcept has been a company since 1998 and on the Stock Market since 2004, it was only two years ago that it finally had a product – Korlym®, for treating the rare Cushing’s disease. But what about Psychotic Depression? Well that remarkable story is in a recent article in Seeking Alpha:
SeekingAlpha
by Bay Area Biotech Investor
Apr. 15, 2014
Summary
-
Mifepristone in psychotic major depression [PMD] has one positive phase 2 and three negative phase 3 clinical trials.
-
Management has made important trial design changes to the current phase 3 trial, most notably regarding placebo effect and dose.
-
The potential of mifepristone in psychotic major depression [PMD] is underappreciated by the market.
This article is on-line [you have to register with an email address, but they haven’t ever sent out anything, at least to me]. It’s written by
Bay Area Biotech Investor, a graduate student in the Biomedical Sciences Program at UCSF who
enjoys Biotech investiong. So what’s with "
three negative phase 3 clinical trials"? Somehow, this company has raised
half a billion dollars at least $307,000,000 [see below comment #2] along the way from venture capital sources to stay alive and prosper in the absence of any viable product. Their approval for Cushing’s Disease is hardly a great leap forward. The naturally occurring disease is literally as rare as hen’s teeth [I’ve seen
a case, but that’s with a 50 year career spanning two medical specialties]. And Psychotic Depression? It’s not common either [impossible to guage its incidence since the diagnosis was folded into Major Depressive Disorder over thirty years ago].
In this article, Bay Area Biotech Investor does an amazing thing. He goes through each of the previous failed clinical trials and deconstructs them in some detail. If only they didn’t have the European sites in the first trial, they would’ve gotten a significant result. Solution, no more European sites. In another trial, if only they included the patients with high blood levels of Mifepristone, they would’ve had the desired outcome. There was a very high Placebo response – with too many Placebo dropouts. If only that hadn’t happened … Harder to solve, but maybe having a large study would work. So now they’ve launched a trial with only US sites, a high dose of Mifepristone, and are aiming for 450 subjects [finding 450 patients that are really psychotically depressed is going to be quite a challenge]. In fact one of their problems it’s likely that the earlier studies didn’t have a true cohort because the cases are not very common. So maybe that’s why the had such a brisk placebo response.
There’s an old Zen Kõan that’s told in many different ways. In one version, a scholar is lecturing to his students about the moon while pointing at it in the sky, giving elaborate descriptions of its features. The Zen Master watching all of this quips, "That fellow is confusing his finger for the moon." This Seeking Alpha article approaches these three negative trials as a challenge to be overcome – assuming the problem is in the trials, not the basic hypothesis – false negatives. With the just-right design, the truth will out and all will be well in the halls of Corcept [one suspects that if the trials were positive, he would hardly be inclined to slice and dice them with such a fervor]. The question is not about the effectiveness of Mifepristone, it’s how to get a positive trial.
The remarkable thing is that the venture capitalists have been through this three times already, yet they continue to shell out the money without apparently considering the fact that Mifepristone isn’t doing very much for the cases that are being collected for their trials. Imagine the results in the world at large where any kind of depressed patient might be chosen for a pre$cription [anyone with whole lot of loose change]
And there’s something else bizarre going on. The whole hypothesis rests on the finding of hypercortisolemia and the idea that blocking the effects of a high cortisol level with the drug will have an impact on the clinical syndrome. But they’re not measuring cortisol levels in the study. Says Bay Area Biotech Investor when asked:
So, why would reporting cortisol levels in clinical trials be informative? Published literature suggests they will be elevated in PMD patients, and mifepristone does not lower cortisol levels, it is an antagonist, it only prevents cortisol from signaling. There is no rational for them needing to validate the PMD diagnosis by elevated cortisol levels when it has already been proven that PMD patients have elevated cortisol levels during certain times of the day. Do you agree with this reasoning? They do have a stringent entry criteria: DSM-IV TR diagnosis of Major Depressive Disorder with Psychotic Features and pre-specified minimum scores on standardized psychiatric rating scales at baseline, which will be conducted by a centralized rating system Medavante. This entry criteria seems much more practical than a entry criteria based on elevated cortisol levels. Do you agree with this entry criteria or would you rather see entry by elevated cortisol levels?
No, I don’t agree with this reasoning. How about both? If only there were a 1:1 cortisol level/clinical diagnosis relationship like Bay Area Biotech Investor thinks, this would be a 100% biomarker and there would be Champagne flowing in any number of streets…