I briefly mentioned Sarepta and etiplirsen, their proposed therapy for Duchenne muscular dystrophy (DMD) in September. In that post, I made reference to the "delirious fun of investing in biotech". Well, the company recently got some regulatory news that illustrates that point even more clearly. The FDA told Sarepta that it would not get accelerated approval for the drug, and that sent the stock into a mineshaft (and infuriated the DMD community, as you might well think).
Matthew Herper at Forbes has some good background on the story here Etiplirsen is one of these drugs aimed at a small market (one particular DMD mutation). And the clinical data were pretty thin:
It can be hard to imagine saying no to a plea like that – but sometimes that is the FDA’s job. As one muscular dystrophy expert told me when I wrote about Sarepta’s results earlier this year, it was always possible that it might be “too much to hope for” to think that eteplirsen could be approved based on the data so far. Eteplirsen was studied in only twelve boys, half of whom received the medicine immediately, the other half of whom initially got placebo but then switched to taking the drug. Those who started on the medicine earlier have higher levels of dystrophin, at least according to muscle biopsies, and appeared to be able to walk a greater distance in six minutes, a sign that their muscles are deteriorating less quickly.
Unfortunately, great results from small trials have a history of not bearing out in larger studies. Even for rare disease drugs, this study was tiny. Worse, the Sarepta results only look good when two of the 12 patients are excluded – two boys were too sick to be helped by the drug. The FDA usually insists that clinical trials be presented in what is known as an “intent-to-treat” analysis, which means that if you even thought about treating a patient they need to be included when you do the math on the study’s results. This is intended to keep scientists from lying to themselves, convincing themselves that a drug works when it doesn’t. One biotech executive with a great deal of experience in rare diseases told me recently that this issue meant the data “would never fly” with the FDA. The recent failure of a similar, but less effective, drug from Prosensa and GlaxoSmithKline GSK -1.64% made the odds dimmer.
And Adam Feuerstein of TheStreet.com, who thought that the company would get the accelerated designation, has a look at the decision here. He spoke with a bearish investor who made this case:
The FDA's issues with trial design are so wide-ranging that it seems like wishful thinking that Sarepta will be able to agree on a study design and start enrolling by the second quarter 2014.
Major questions with dystrophin quantitative assay. Questions with results of anything less than two years. Need for a larger study to power the six-minute walk test (6MWT) data. Possible need to expand study population both high and low and go beyond 6MWT as primary endpoint. The FDA is very deeply skeptical and Sarepta will have a difficult time coming to a study design that the company thinks they can do and that the FDA will be satisfied with.
And any trial seems likely to last 2 years. Seems to me that even if all goes well, approval would get pushed out much more than two years. They're going to spend 9 months arguing over study design and probably won't start enrolling until early 2015. Two-year trial plus filing and approval. Sounds like early 2018 approval at best.
I have to say, this is consistent with worries resulting from the Prosensa trial that the market ignored. But it's actually even more negative that I expected. I thought the FDA would just say, do a larger trial along the same lines. What they're saying is much more confused than that. What is a valid marker and what do you need to get the data to support it?
It makes you wonder whether the FDA really changed their minds lately or if Sarepta misrepresented (through wishful thinking or worse) what the FDA had been telling them all along.
The agency really did made things much trickier than most people had been expecting. They're talking about completely new endpoints, rather than just shoring up the data collected so far (which was already more than the company's boosters were willing to think about, in some cases). I don't envy the folks at the FDA, but then, I never do. They get to look like heartless bureaucrats, bleating about numbers while children are suffering. The flip side, though, is trying to keep people from raising their hopes for something that does no good. If we approve things that just look as if they might work, all sorts of charlatans will rush in, human nature being what it is.
But we're not talking about approval here, just an accelerated protocol for it. Surely they could have at least agreed to fast-track this one? I think, though, that the FDA saw itself being put into an untenable position. They did not think that there was enough solid evidence to approve the drug as it stood, and accelerated approval would ensure that no more was going to be forthcoming. All that would do would be to get everyone's hopes up even more, for what still would looked very much like a rejection based on insufficient data. And in that case, why not tell the company now and get it over with?