That’s my reaction to yesterday’s news in the field. Merck started things off with news of their last-ditch attempt to see if their beta-secretase (BACE) inhibitor verubecestat (MK-8931) could be useful. I’ve been writing about that one for a while – here’s 2012, when they announced that the compound would move into Phase III, and here’s an update in 2016, making the case that if any beta-secretase inhibitor was going to do something, this would probably be the one. Last February, though, I wrote about that first Phase III trial ended up being stopped for futility: nothing was happening, and nothing ever looked like it was going to happen, either. Here’s a quick look back at that 2012 post:
. . .the headlines are. . .well, they’re mostly just not realistic. “Hope for Alzheimer’s”, “Merck Becomes Bigger Alzheimer’s Player”, and so on. My two (least) favorites are “Merck Races to Beat Lilly Debut” and “Effective Alzheimer’s Drug May Be Just Three Years Away.” Let me throw the bucket of cold water here: that first headline is extremely unlikely, and the second one is insane.
Yeah, that about sums it up. But Merck kept pitching. The usual resort in failed Alzheimer’s trials is the hypothesis that the disease wasn’t being treated early enough. That has never worked, but it’s not a stupid idea, either, and Merck gave it a try in “prodromal” patients, the earliest population you can pick without just calling it a preventative trial and giving it to healthy people. But an interim review of the clinical data apparently was discouraging enough to call a halt to the whole effort. There are a number of other beta-secretase efforts still going on in the clinic, and good luck to them, but these results have to lower their chances of success even further.
Then later on yesterday, Biogen came out with the news that they’re adding more patients to their Phase III trial of their antibody aducanumab. Remember that one? Back in early 2015, there was a rush of enthusiasm for it based on some Phase I data (yeah, Phase I). By July of that year, less compelling data had come out, but the company went straight to Phase III to see if there was something really there. So why are they adding patients? “Variability”, says the company, and it’s easy to assume that they mean “There’s so much noise in the data, because there’s hardly any effect, that we’re going to have to get a bigger N to have any hope of seeing anything at all”. What else? Investors did not take the news well.
I have the same worries about this that I had about Lilly’s failed amyloid-antibody efforts. Back in 2012, while the company was in the process of spending another mountain of cash to make absolutely sure their drug didn’t work, I wondered if we’d all be worse off if they did manage to achieve some sort of statistical significance. I think that the FDA would be under huge pressure to approve anything that reached some sort of endpoint, regardless of whether it actually affected the course of the disease much (or at all). There’s just so much demand for an Alzheimer’s drug – patients and families are desperate, and the first new drug that gets approved will bring in billions upon billions of dollars of revenue, whether it does a damned thing in the real world or not. The moral hazard for drug companies and regulators is the temptation to advance something that only looks as if it’s helping, but really isn’t. So far, the failures have all been definitive enough that this hasn’t been an issue, but it’s waiting for its turn.
What else was in the Alzheimer’s news yesterday? Why, talk of “reversing” the disease itself! Headline after headline used that verb, with only some of them using the word “mice” as well. The paper underlying all this coverage was indeed in an engineered mouse model, and showed regression of amyloid plaques on modulation of the activity of. . .beta-secretase. The same target that Merck had just dropped for futility in human trials again that morning. Even the Daily Mail noted the contrast; that’s how blatant things got.
The study (from a group at the Cleveland Clinic) says that its objective was “To better understand how BACE1 inhibition in adults will benefit AD patients“, but the mass of evidence is that such inhibition does not benefit AD patients in the slightest. Of course, the paper’s abstract starts out by saying that “BACE1 initiates the generation of the β-amyloid peptide, which likely causes Alzheimer’s disease (AD) when accumulated abnormally“, and you know what? I’m pretty sure that I don’t buy that last part any more, either. Not at this point.