Skip to main content

Alzheimer's Disease

The Best Beta-Secretase Inhibitor Yet in Alzheimer’s

Merck has published new results on their latest Alzheimer’s drug candidate, the beta-secretase inhibitor verubecestat (MK-8931). The paper in Science Translational Medicine is quite interesting, both for what it says and what it doesn’t (or can’t, yet).

BACE inhibitors have had quite a time of it over the years. Coming up with a molecule with enough potency on the enzyme is merely the first step. You also have to get it into the brain, and a lot of likely structures have trouble doing that. Then your troubles really begin. Beta-secretase is known to be important in myelination of neurons and in muscle cells, neither of which you will want to mess with, and there have been reports that it’s important in memory as well, which is really unfair. Previous compounds have had problems with myelination, and have also shown liver tox (which to be fair, can happen to any compound at any time).

Merck’s compound has been in the clinic for some years now, and they started the current trial in 2012, with <s>results expected in 2019.</s> Update: turns out that the 2019 readout is for the early stage (prodromal) patient trial. The mild-to-moderate AD patient trial is getting a first readout next year. Lilly and AstraZeneca have another one reporting around that time, and yeah, that’s the timeline for Alzheimer’s in the clinic, all right. So this paper is still an early report of a work in progress. What they’re showing is that verubecestat does indeed reduce cleaved amyloid species in rats and monkeys, without evidence of the severe tox liabilities mentioned above. (There was some fur depigmentation seen in the rats, but nothing of the sort in the monkeys). And in humans, the trend continues: the compound reduces amyloid fragments in the CSF of both the control group patients and those with diagnosed Alzheimer’s disease.

That’s a significant accomplishment – this is definitely the best shot that the beta-secretase mechanism has ever had in the clinic. But what you’ll noticed is that nothing at all is said about actual efficacy of the compound against Alzheimer’s – for that, we’re going to have to wait until the end of the trial in another three years. You wouldn’t know this from the newspaper headlines, particularly the ones in the UK press. (I really don’t know what it is over there, but it’s a tradition that every tiny bit of news about Alzheimer’s gets CURE ON THE WAY coverage, every time). If the amyloid hypothesis is right, and if we’re right about targeting this part of the pathway, and if some ugly tox event doesn’t happen in the next few years, then this has a chance of being a treatment for the disease. For Alzheimer’s disease, that’s literally as good as it gets, but to write things like “Alzheimer’s Cure In Sight” is just not responsible. This field has a success rate of basically zero per cent in the clinic; people have been trying to get beta-secretase inhibitors off the ground for twenty years now.

So looking forward to victory is kind of premature. There’s an awful lot that we don’t know about amyloid and Alzheimer’s, and the only way to discover a lot of it is to take compounds like this into humans, spend hundreds of millions of dollars, and cross our fingers. It could well be that verubecestat goes on to reduce those amyloid fragment species in human patients for the next few years, with no effect on Alzheimer’s at all. We just don’t know. I certainly hope that it actually works, because the number of Alzheimer’s patients is growing every day, and the sooner they get something to slow down the progress of the disease, the better. But telling people that it’s working already is wrong.





36 comments on “The Best Beta-Secretase Inhibitor Yet in Alzheimer’s”

  1. bhip says:

    Note- the compound is also a equi-/more potent inhibitor of BACE2, as if the unknowns associated with BACE1 inhibition were not enough.
    That said, this compound probably does represent the best direct test of the B-amyloid hypothesis.

  2. Anon says:

    Journalism in the UK requires no qualifications, and certainly no scientific training nor effort to do any due diligence. As long as you can string a few words together to write something under a headline that people want to read, you’re in.

    1. loupgarous says:

      Here in the US, it’s the same story. The “Canons of Journalism” once governed how the business was done here, but they were never more than aspirational guidelines. Any time Big Journalism has a dog in the fight here in the US, objectivity and accuracy go out the window – and as in the UK, if they can sell more ad space/time while molding the minds of their readers/viewers, journalists smile and their bosses smile wider.

      It makes for few trustworthy news sources, regardless of the political polarization of the source. Katie Couric (who no one here will confuse with a medical journalist) has been very crafty in how she plays to anti-vaxxers’ fears and groundless phobias, and she’s everywhere (most recently, the voice of Yahoo!News, and Jonathan Swift was never more prophetic than when he coined the term “yahoo”).

    2. Mark Thorson says:

      You are so unkind to the Daily Mail. So very very unkind.

      1. Anon says:

        Saying that they can “string a few words together” is a big compliment for The Daily Mail.

    3. pete says:

      re: UK press
      To my surprise the Guardian recently jumped on Merck’s BACE-inhib. data with pretty hype-ish headline & lead-in. Farther in there were the proper caveats.

      Why surprising ? Because I more expect that from The Telegraph, whose sci-med writer’s would have us believe a cancer cure is 1 trial away, but not so much from the Guardian, whose writers generally seem more informed & circumspect about drug candidate data.

      1. Falanx says:

        The Guardian seems to be on the very beginning of the slippery slope these days. No longer does it merely have a problem with correct spelling and punctuation, but is rapidly charging off down some misandrist (Jessica Valenti, pride of place), misunderstood environmentalist angle, with very recently a particularly militant secret moderation of older readers on their CiF pages who simply call them to uphold their previous standards…

    4. John Dallman says:

      Also, the Daily Mail and the Daily Express spend a lot of effort making their readers fearful, and a lot of those readers are elderly. It might be too cynical to claim that they talk up every Alzheimer’s story in the expectation that it will come to nothing and increase fear. Might.

    5. Andy Extance says:

      With due respect, the comment about journalism not requiring qualifications in the UK is very much not true. Mainstream press typically requires NCTJ ( qualifications, and usually it’s graduates vying for what are hard-to-get, much-prized roles. Science journalists in particular are often science graduates who have taken a masters’ in science communication. Yet even writers with science training aren’t experts in every area, and often nonsense slips through when these stories are handled by journalists for which the underlying science isn’t their specialism. And headlines in particular are a thorny area. They’re written by sub-editors, whose primary motivation is to get readers to the article. It’s not unusual that the headline bears little relation to the article itself for this reason, and provides motivation for hype.

      Many factors intertwine to create poor science coverage, and in particular there is great pressure to attract readers in a market where revenue is declining, creating editorial environments where accuracy is often not the most prized commodity. I’d therefore be charitable to the journalists themselves. I know from my time in pharma, where I went because I thought drug discovery was a noble discipline with the purest motivations, that the reality of a job you fought hard to get is often hard to come to terms with.

      Love the idea of a betting market to assess the reliability of news in general. Perhaps we can get people to do that instead of the godawful sports betting that plagues the UK.

  3. Mark Murcko says:

    I’ve not worked on BACE myself, but my hat is off to the many, many teams that have slaved away for a decade or more on this target. The work I’ve seen (in papers and presentations) is generally of an extremely high quality, shows subtle thinking, and — most importantly — required uncommon amounts of risk-taking and resilience. Great role-models for the field. Best of luck to all.

  4. Emjeff says:

    Political journalism has been dead for a long time, and it looks like science journalism is well on its way.

  5. DBP says:

    Any thoughts about Aducanumab? An article in Nature ( showed some interesting stuff. I didn’t feel NBC world news made it seem like a cure is on the way. The absence of plaques is very promising is how they left it. Derek, thoughts?

      1. Anon says:

        sgcox, please,

        the Nature paper (linked above) was not published when the blog you mention was discussed.

        1. sgcox says:

          Precisely. Biogen released the human trial data in July which made a news splash and discussed here. Then the Nature paper with basically same stuff get published.

    1. Lane Simonian says:

      Biogen carried the last observed results forward for Aducanumab. The highest dropout group was among ApoE4 carriers who may progress most rapidly during the early stages of Alzheimer’s disease (some dropped out due to brain swelling). Most in the placebo group made it to the end of the trial (one year), so you are comparing the results of the placebo group most of whom made it to the end of the trial to an ApoE4 some of whom did not make it to the end of the trial all at one year.

      At six months, there was little difference between the results for the placebo group and non-placebo group taking any of the various doses of Aducanumab. Biogen used a statistical slight of hand to convince people that its drug was working.

  6. sgcox says:

    According to there are three active trials.
    First two are phase III, started in 2012 and 2013 and results expecting in june 2017 and may 2019. Perfectly reasonable. But the 3rd, NCT02910739, is basically Phase I trial in patients with hepatic insufficiency. Started in September 2016 and finishes in May 2017, just before the first phase III results are out. I think it is a bit odd and may be there are signs of toxicity. Or may be Merck plays it really safe and want to check any potential liabilities ASAP before the breakthrough news go out.

    1. Scott says:

      My money is on Merck being professionally paranoid and wanting to check for potential toxicity issues before getting their stocks hyped for a Alzheimer’s treatment breakthru.

  7. AnAn says:

    This is a great story in the making – I am sure we all hope it works out.

    If you could please refresh my memory, didn’t other AD drugs previously tested in preclinical species and human also lower amyloid fragments in the CSF? And didn’t those fail to show efficacy in late clinical trials?

  8. Anon says:

    Amgen was an early player, in fact they cloned the BACE1 gene, but lack of serious small-scale molecule drug discovery expertise doomed the effort.

    1. Chrispy says:

      As has doomed every small molecule effort at Amgen (except outright purchase). Repeated failure has not stopped them from trying, though!

      I have heard more than one Amgenner speculating about what would have come from doubling down on their protein drug strengths rather than constantly siphoning off effort into small molecule fiascoes.

    2. john adams says:

      Wait – didn’t Roger Perlmutter build a dynamic, highly productive small molecule discovery team at Amgen. No, sorry ! I messed that up. He DESTROYED a dynamic, highly productive small molecule discovery team at Merck.

  9. Lane Simonian says:

    “The cure for Alzheimer’s disease is just around the corner” articles seem to have hit a new high over the last month. Part of this is due to the fact that the drugs seems to be hitting their target whether amyloid plaques or beta secretases. And then the great leap of faith is made: amyloid is the cause of Alzheimer’s disease–the drugs are preventing the development of or removing amyloid–the cure for Alzheimer’s disease is in reach.

    In most people as Alzheimer’s disease progresses, the secretion of the amyloid precursor protein decreases due to less protein kinase C activity. So beta secretases are being targeted for the early stages of the disease (or put another way this intervention has little chance of helping most people after the early stages of Alzheimer’s disease). Whether the side effect issues have been resolved is certainly up in the air.

    Here is another potential issue. The c-terminal fragment of the amyloid precursor protein (the one produced by the beta secretase) increases g protein signalling and consequent nitro-oxidative stress in Alzheimer’s disease but so too does excessive amounts of the amyloid precursor protein itself. So as pointed out to me last time, mutations that only lead to excess production of the amyloid precursor protein in themselves cause Alzheimer’s disease. Would cutting off the cleavage of the amyloid precursor protein lead to the same result?

    With some reluctance, I will make this point again. The various forms of amyloid (with the possible exception of amyloid plaques) contribute to the nitro-oxidative stress in Alzheimer’s disease. Removing them does not remove the other factors that has caused this stress. At best they only delay the onset of the disease. Reverse the nitro-oxidative stress, however, and you largely reverse the disease.

    C-terminal fragment of amyloid precursor protein induces astrocytosis

    Furthermore, conditioned media derived from CT105-treated astrocytes enhanced neurotoxicity and pretreatment with NO and peroxynitrite scavengers attenuated its toxicity. These suggest that CT-APP may participate in Alzheimer’s pathogenesis through MAPKs- and NF-κB-dependent astrocytosis and iNOS induction.

    1. Matthew K says:

      “With some reluctance”? I for one am glad you overcame your doubts to give us the One Truth about Alzheimer’s yet again.

      1. Buck turgidson says:

        Please, attack the data or his claims not his message. And yes I too have been watching this merry-go-round turn for years but back-of-the classroom snark doesn’t help. There are points there that oppose the pharma stupidity in this area.

  10. luysii says:

    It’s worth noting that another trial is in progress, with a drug (Lopid) which has already cleared the FDA. As the link below will show I’ve been back and forth with the authors. Results aren’t expected for another year or so (which is better than the 3 years mentioned above)

  11. Rule Follower says:

    Two interesting nuggets in the paper were in vitro hERG of 2.2 uM yet no QT prolongation and PGP ratio of 11 yet useful amounts in CSF and apparent CNS amyloid lowering. Congrats to the team for pressing forward with a few rules broken.

  12. luysii says:

    Sorry, the reference to the guys doing the lopid work is in here —

  13. luysii says:

    Sorry, the following is a link to the guys doing the lopid work

  14. Lane Simonian says:

    Beginning in the early stages of cognitive decline, beta secretase activity is increased due to the upregulation of caspase 3 and the alternative alpha secretase is downregulated by the tyrosine nitration of the phosphatidyinositol 3-kinase. At a certain point in time likely very early in the process, all that you are doing by inhibiting beta secretase is increasing levels of the amyloid precursor protein and that alone can kill neurons.

    alpha- and beta-secretase: profound changes in Alzheimer’s disease.

    “Here, we describe a decrease in alpha-secretase (81% of normal) and a large increase in beta-secretase activity (185%) in sporadic Alzheimer’s disease temporal cortex… Eighty percent of Alzheimer brains examined had an increase in beta-secretase, a decrease in alpha-secretase, or both; which may account for the means by which the majority of people develop Alzheimer’s disease.”

    Alzheimer’s Disease Without Amyloid Plaques

    Amyloid plaques have long been thought to be the cause of neuron loss in Alzheimer’s disease. Now researchers report that excess of mutated amyloid precursor protein (APP) inside the neurons is sufficient to induce neuron death. The report challenges the notion that amyloid deposits outside of the cells are necessary for neuron death in Alzheimer’s disease. Their results also indicate that by reducing the amount of APP within neurons it may be possible to rescue them.

    But even if you remove excess amyloid precursor proteins, many other factors can cause the death of neurons. Indeed this is the problem with trying to remove or prevent the development of amyloid in any form; it is at best only part of the problem in Alzheimer’s disease.

    I suppose that we have to wait another three years to find out about the cognitive effects of beta secretase inhibitors. In the meantime, it might be a good idea to start working with compounds based on a different hypothesis of Alzheimer’s disease.

  15. tangent says:

    The Brits ought to do a betting market in whether drug candidates will be approved as effective. The long odds would filter in to journalism and public consciousness. And since the odds probably wouldn’t be long enough, cynics could make a little money.

    I actually think this might be brilliant.

  16. Lane Simonian says:

    I probably need to take a step backward: it appears that it is not excessive amount of the amyloid precursor protein, but the c terminal-fragment cleaved from the amyloid precursor protein that is responsible for neurotoxicity via g protein signalling–this occurs whether there is a subsequent production of amyloid monomers, oligomers, and plaques or not.

    “It can be inferred from these data that the mutation Asp664 “rescues” multiple aspects of neuropathology and impaired learning that are normally caused by the Swedish and Indiana mutations in APP. In other words, if C31 cannot be generated, the FAD APP mutations cannot cause certain pathological and behavioral abnormalities. What are the implications of these findings? First of all, note that Asp664 selectively rescues the neurodegeneration and the learning abnormalities of the PDAPP mice without decreasing the production of Aβ40 or Aβ42. Thus, the rescue is independent of the production of Aβ. Secondly, the C31 region of APP encompasses the binding sites for nearly all of the signaling proteins, including APP–BP1 and PAK3, that have been shown to bind to the intracellular domain of APP.”

    This would seem to be good news for the developers of BACE inhibitors to treat Alzheimer’s disease, but not really. Amyloid precursor protein production decreases as Alzheimer’s progresses so inhibiting BACE would make little difference. Secondly oxidative stress is only partly dependent upon the c-terminal fragment of the amyloid precursor protein. If you lower oxidative stress by limiting the formation of the c-terminal fragment of the amyloid precursor protein through BACE inhibition, you can perhaps delay the onset of Alzheimer’s disease. If you lessen oxidative stress–no matter what its source–you delay the onset of Alzheimer’s disease and treat it.

  17. Lane Simonian says:

    Another recorrection. Mutations that lead to an excess amount of the amyloid precursor protein can increase g protein signalling leading to apoptosis, but this signalling is amplified by the subsequent formation of c terminal fragments of the amyloid precursor protein. Other amyloid precursor protein mutations make the amyloid precursor protein more susceptible to the cleavage that leads to c terminal fragments and this is what contributes to the onset of Alzheimer’s disease in those cases.

    A recent critical finding:

    Alzheimer’s disease-associated mutations increase amyloid precursor protein resistance to γ-secretase cleavage and the Aβ42/Aβ40 ratio

    “Unexpectedly, most familial AD (FAD)-linked APP mutations make APP partially resistant to γ-secretase.”

    With limited y-secretase activity there are few amyloid oligomers and few amyloid plaques, but the individuals with these mutations still develop Alzheimer’s disease. And y-secretase inhibitors actually made Alzheimer’s disease worse, suggesting that the amyloid cascade actually works in reverse–each form of amyloid is less toxic than the one that precedes it.

    One more wrinkle, caspase-3 appears to be able to cleave the amyloid precursor protein directly.

    Cell death induced by a caspase-cleaved transmembrane fragment of the Alzheimer amyloid precursor protein

    “The Alzheimer amyloid precursor protein (APP) is a transmembrane protein whose abnormal processing is associated with the pathogenesis of Alzheimer’s disease. Activated caspases cleave APP and generate its carboxyl-terminally truncated fragment (APPDeltaC31)…These results suggest that accumulation of wild-type APP activates neuronal caspase-3 to generate APPDeltaC31 that mediates caspase-3-independent cell death.”

    If you target caspase-3 instead of BACE, you take care of both enzymes that are apparently responsible for the cleavage of the amyloid precursor protein (as caspase-3 stabilizes BACE) and you reduce neuronal cell death. Even better if you scavenge peroxynitrite, you limit caspase-3 activity.

    Peroxynitrite induces apoptosis of HL-60 cells by activation of a caspase-3 family protease.

    So to repeat the findings in the study referenced in a previous post:

    “Furthermore, conditioned media derived from CT105-treated astrocytes enhanced neurotoxicity and pretreatment with NO and peroxynitrite scavengers attenuated its toxicity.”

    No clinical trial to date targeting various forms of amyloid have succeeded in the treatment of Alzheimer’s disease. Several small-scale trials using peroxynitrite scavengers have produced benefits for Alzheimer’s patients, including improvements in some forms of memory and in some cases behavior. Draw your own conclusions.

  18. Andre says:

    The recently published data on Merck’s the beta-secretase inhibitor verubecestat (MK-8931) are indeed encouraging. Two question however arise:

    1. Will treatment of the mouse model carrying a knocked-in mutant form of amyloid precursor protein with verubecestat reduce or abolish amyloid plaque formation?

    In my opinion, this is onto only mouse model reproducing the early pathological hallmarks of AD (see

    2. Why doesn’t Merck perform the clinical trials in patients with diagnosed familial AD. All these patients will with 100% certainty develop AD and thus they should profit most from verubecestat as there is no approved drug on the market that will slow disease progression or cure them?

  19. david says:
    says that when they examine the brains of 90 year olds (who were superior in memory) after their death they find the same Alzheimer’s pathology as those with Alzheimer’s. If this holds up, it may be that the model is just plain wrong.

Comments are closed.