Skip to main content

Alzheimer's Disease

A Brief Note About Alzheimer’s

Well, there it is. Biogen and Eisai have announced just this morning that they’re halting Phase III trials of aducanumab, their anti-amyloid antibody, after the monitoring committee judged that further treatment would be futile.

I’m not going to do some sort of victory dance, because (once again) this is bad news for Alzheimer’s patients and for their families. I know that I have written many times that I thought this program would fail – I was cautious in 2015, cautious earlier last year, and much more than cautious just a few weeks later. My views on amyloid antibodies are well known, for the little that’s worth. I had no real expectations that aducanumab would work, despite all the attempts at positive spin over the years, and by golly, it doesn’t work. That doesn’t make me a prophet – I think any objective observer would have to have come to the same conclusion. Biogen and Eisai put themselves into this situation, although they definitely made it worse by trying to pretend that things were going differently than they have for every single other amyloid antibody program ever. They have all failed. One after another, again and again.

Amyloid definitely has something to do with Alzheimer’s – there’s far too much evidence to dismiss. But the situation is clearly more complicated than people have hoped, because otherwise, all the attempts to address amyloid (via antibodies and otherwise) would have yielded some tiny bit of clinical benefit. They have not. All the talk, some of it very loud and confident-sounding, about how it’s really another amyloid species, it’s the oligomers, it’s the soluble oligomers, that the drugs would work if they were just dosed earlier, in different patients: none of it has ever worked out. Not once. Something is wrong with the way we’re thinking about Alzheimer’s and amyloid, folks, something is wrong. It’s been wrong for a long time and that’s been clear for a long time. Do something else.

107 comments on “A Brief Note About Alzheimer’s”

  1. McChemist says:

    Biogen’s stock is down 27% in premarket trading, which suggests that a significant number of investors expected this drug to work. Which makes me wonder, who are these investors, and what were they smoking?

    1. Nate says:

      Found one, via Leerink’s analysts:

      “Our model is currently under review as we previously assigned a 35% probability of success for this program despite our many concerns regarding beta amyloid as a target following the failures of multiple other antibodies in recent years.”

      Some very serious analysts thought a 35% chance of success? Whew.

      1. Emjeff says:

        “So you’re telling me there’s a chance…”

        1. rp says:

          Aaah, Dumb and Dumber. Some investors will cry like those two when watching heartbreaking commercials.

      2. analyst voodoo says:

        Yes – 35%, with an uncertainty of +/- 65% ? Investment analysts don’t believe in error bars, especially if they’re fabricating numbers.

      3. Hap says:

        How do you determine to 5% accuracy the chances of success of a drug? Or it is just numerical glitter?

        1. JK says:

          Best guess: You start with a base rate determined by a company’s track record and make adjustments according to Bayesian principles. It’s not a science but it’s better than personal judgement. Just not in the case of Alzheimer’s treatments.

          1. Reverend Tommy B. says:

            Exactly as @JK says, and then you overlay the Bayesian posterior with a /loss function/, which maps correct and incorrect outcomes to dollars. If that stock analyst is being rational, then s/he believes that even with the low probability of success, the shares are underpriced.

            Tech VCs push this idea even further, rationally investing in firms with an even lower probability of success, knowing that a few are likely to be blockbusters.

    2. aairfccha says:

      A version of Pascal’s Wager. An investment the loss of which you can easily tolerate with the potential for a giant profit can be worthwhile even if the odds of success is low. Essentially it’s the stock market version of playing the lottery.

    3. Me says:

      …. and are they interested in any of the lucrative investment opportunities I currently have available?

  2. johnnyboy says:

    This has to be one of the most thoroughly nailed coffin ever for a scientific hypothesis. To be fair, the ‘amyloid as cause of AD’ is probably the toughest zombie theory you’ll ever meet. It has eaten the brains of many scientists

    1. ScientistSailor says:

      The Columbia trial will be the nail in the coffin…

      1. Marc says:

        You’d think that, but you’d be wrong. Like poverty, the amyloid hypothesis will never go away.

  3. Nate says:

    The trial result may not come as a surprise – but perhaps what is a surprise is the stock reaction. Were there truly that many optimists out there? Here I was assuming that a failure was already baked into the price, guess that explains why I’m in the lab and not on Wall Street.

  4. Johannes says:

    What diagnostic category do you put in failure of the brain function regardless of etiology? Alzheimer’s. The amyloid theory is ludicrous

    1. Hap says:

      It’s not ludicrous – it just doesn’t work.

      At some point, you need to actually listen to the data and not treat it (them) like your kid telling you you passed the correct exit on the freeway. “No, I know where I’m going {sprog 1}…be quiet {sprog 1}…I’m talking to Mom {sprog 1}, be quiet…%&*^, I missed the exit. Why didn’t you tell me?”

    2. Billy says:

      Oh how science loves the Monday morning quarterbacks like you. I’m sure in 20 years, you’ll say you had it all figured out by 2019.

      1. Isidore says:

        In fairness Derek has been expressing his doubts regarding the amyloid hypothesis and the wisdom of moving forward in the clinic anti-Abeta antibody therapies and in the broader community quite a few people have been questioning this line of research even when the amyloid hypothesis was king, I know, I worked on Alzheimer’s drug discovery, centered on this hypothesis, back in the 1990s and I recall the scorn heaped on those who questioned it.

        1. Ian Malone says:

          Derek has had a very good point, and I suspect lots of people were watching aducanumab more with hope than optimism. But it’s not the same point as, “What diagnostic category do you put in failure of the brain function regardless of etiology? Alzheimer’s.”

          Alzheimer’s is very well categorised pathologically and can be distinguished from other neurological disorders. There is a reasonable rate of misdiagnosis (varies depending where you are) and confusion in the general population between dementia and AD, although that’s decreasing thanks to growing recognition of its importance. Misdiagnosis is also a nuisance for running trials, since you need to be careful about your inclusion criteria.

          Equally, calling the amyloid hypothesis ludicrous is itself ludicrous; the protein has been associated with the disease since it was first identified. It may be we can’t treat AD by removing it, but denying amyloid has any involvement in the disease process requires ignoring quite a lot of evidence.

          1. Hap says:

            I thought the amyloid hypothesis was that amyloid itself (or a lower oligomer) caused Alzheimer’s Disease and its effects, and that reducing amyloid deposition or oligomer formation and liberation could effectively treat Alzheimer’s (prevent its symptoms or ameliorate them). That hypothesis does not require that amyloid not have any effect (because there’s lots of evidence for it to be present and to have an effect) but that either it isn’t causative or that reducing its formation can’t stop Alzheimer’s.

          2. Ian Malone says:

            Hap: yes, I’d say that’s the hypothesis, and in that pure form of amyloid being the sole initiator and driver people are already moving past it. It wasn’t exactly ludicrous though, particularly since that assertion above was made on the basis of Alzheimer’s being some undefined catch-all for “failure of brain function”. There’s a recognisable disease process (or if you prefer, a number of extremely similar ones), which is distinct from, for example, Parkinsonism.

  5. Marc says:

    I’m waiting for the chorus of people saying the drug wasn’t given early enough.

    Or that it didn’t bind the right kind of oligomer…

    1. An Old Chemist says:

      And a group of statisticians will say that the drug worked in a sub-group of people.

      1. Chrispy says:

        ^^ THIS. This will happen.

        1. Marc says:

          When you’re right, you’re right, Chrispy…. sigh

  6. Plan B says:

    Can anyone share what the current alternative hypotheses (and associated targets/pathways) are for treating Alzheimer’s that look promising, combo approaches included? Hopefully, there’s not a bunch more train wrecks ahead, given the acute medical needs.

    1. Charles H. says:

      The problem is, of course, what alternatives are there. As someone totally removed from the field I have often wondered if the plaques are intended as a protective measure, that just loses it’s brakes to the point where it damages things. If that’s so, then the plaques actually *do* cause damage, but only secondarily, and if you remove them the initiating insult continues…so you need to do two things:
      1) fix the initial cause, and
      2) remove the plaques.
      The problem with this theory is that there’s no grounds for assuming that the initial cause is always the same.

    2. Ian Malone says:

      Not a thorough review, but a few ones I know about:
      Anti-sense tau therapies, injected intrathecally tau is implicated in AD, it’s also being tried in diseases that involve tau mutations.
      Losartan, a hypertension treatment targets angiotensin, there’s some observational evidence of reduced risk of AD and supporting evidence in mice. A UK phase 2 study is being run by the University of Bristol.
      Small molecules targeting amyloid, CPHPC different approach to the antibody therapies that have been tried. It’s known to clear plaques in systemic amyloidosis.
      Peripheral involvement in a couple of these.

    3. Dolph says:

      I have reached a point where I believe that the only way to find the cause is to find an effective treatment BY CHANCE.
      There is still the phenomenon of propionic acid derivate analgetics reducing incidence by a whopping ~50% when used long term. I think it would be a good idea to throw a ****load of money on studying these effects and coming to conclusions from there.

      1. loupgarous says:

        Good call.
        I’d double down on that and call for study of all Alzheimer’s disease sufferers’ charts AND those of their close relatives, plotting course of disease versus
        every medication they take,
        – vertical occurrence patterns (grandparents, parents vs children),
        – horizontal occurrence patterns such as difference between siblings in disease susceptiblity/course
        – intercurrent illnesses
        – notable patterns in course of disease versus different geographic locations, jobs, and other things I’m forgetting to consider.

      2. loupgarous says:

        Speaking of propionic acid derivative analgesics, here’s a nice collection of articles on them I just found. Nothing about their activity in Alzheimer’s Disease there, but considerable discussion of new and representative compounds, their adverse event profiles (and mechanisms thereof) and their efficacy as analgesics.

    4. Jane says:

      Look up Dr. Dale Bredesen’s REcode protocol for a much more realistic and wholistic approach to the disease. One drug will NEVER fix it. EVER. He has a book called “The end of Alzheimer’s.”

      1. Simona says:

        He’s a quack.

    5. In answer to your question about a hypothesis other than amyloid for a cause of Alzheimer’s there is one but it is not well excepted and it should be. Amyloid plaque as well as tau are symptoms of Alzheimer’s. My husband who is a chemist several years ago began researching ways to help his mother who has late onset Alzheimer’s. Once he learned there was no treatment he went to the scientific literature as he knew he needed to find a cause in order to find treatment. He achieved his goal. Research has reached a tipping point and Aluminum is a causal factor of Alzheimer’s. The good news is drinking silica rich mineral water is effective at removing aluminum from our bodies. Amyloid plaque, tau and ROS are all caused by aluminum. What my husband learned compelled him to write 2 evidence based books about aluminum as a causal factor of Alzheimer’s. The book titles are Prevent Alzheimer’s Autism and Stroke with 7 Supplements, 7 Lifestyle choices and a Dissolved Mineral and Silica water the Secret of Healthy Blue Zone Longevity in the Aluminum Age by Dennis N Crouse. Check out the research being done at Keele University by Dr. Chris Exley.

      1. SomeIdiot says:

        And how many peer-reviewed papers did your husband’s work result in?

  7. Matt says:

    To date, there is only one robust pharmacologic test of the amyloid hypothesis (if broadly defined to include oligomers etc.) as the cause of organ (and neuro) degeneration in humans…Tafamidis / transthyretin. In that case the small molecule clearly reduces amyloid/oligomers/protein aggregation and the patient benefits are manifold. Anyone choosing to now ignore the amyloid hypothesis (again, broadly defined) better have a great alternative explanation for that story.

    There is a lot of validity to the arguments that the approaches to reduce amyloid that have so far been clinically applied in Alzheimer’s have major defects and do not represent a robust test of the amyloid hypothesis. Again, look to the transthyretin story before making the (likely) misguided decision to walk away from the concept.

    1. Derek Lowe says:

      I think that TTR-based amyloid pathology is sufficiently different from the cerebral beta-amyloid pathology of Alzheimer’s to make the comparison difficult. But I know that others disagree.

      1. Lane Simonian says:

        I am not commenting on Alzheimer’s disease but on transthyretin-related amyloidosis. In fact, I am not even going to comment on that, but read this article carefully and you will get some answers.

      2. Christopher G. Janson says:

        Could you please be more specific on what you think is different? How about British Familial Dementia or other forms of amyloidosis with superimposed dementia? Is the dementia incidental in your view?

  8. DTX says:

    One intriguing talk I heard last year noted that the amyloid theory may be still right, but that we are starting clinical trials much too late. That once the cascade starts, it’s impossible to stop.

    The speaker suggested that we may need to start Alzheimer’s trials 20 years before signs of the disease are actually evident.

    He noted that the rate of mental decline for individuals that have a specific genetic sensitivity to Alzheimer’s is no different than the rate of decline for that of the rest of the population. It’s just that for those with genetic susceptibility, the decline starts many years earlier, but the curves parallel each other (he showed this graphically. Hopefully, the above text explains this adequately).

    It seems like it would be impossible to test a drug candidate for a disease that might happen in 20 years. (Maybe a candidate could be tried in those with genetic susceptibility – but how predictive are the genetic markers for early onset Alzheimer’s? And how big is this patient population?)

    1. Derek Lowe says:

      A lot of people have been making this argument, but it’s very close to unfalsifiable. Antibodies are already in trials in populations with genetic susceptibility, though, so maybe that will clear things up a bit?

      1. Ian Malone says:

        One hopes. There is also the possibility of the autosomal-dominant forms being a somewhat different disease, this recent paper on PET imaging in ADAD for example you might look at this and convince yourself of a sort of dominoes effect with amyloid, where you can pour in quite a lot until you either exceed some threshold (ADAD, note non-carriers at the age investigated don’t have detectable amyloid, unlike older people with AD) and things kick off (that sudden appearance of Tau) or something else goes wrong and they interact (sporadic variants).

    2. Matt says:

      Actually, they saw exactly that in the TTR clinical trials with Tafamidis, check the papers on the clinical trial results.

      It’s true that TTR cardio- and neuro-pathies are not Alzheimer’s, but the TTR/Tafamidis data clearly show that protein aggregation at least can drive the pathology of an amyloid disease. Giving up on this concept in Alzheimer’s, particularly given that the only serious clinical success to date in any amyloid disease specifically prevents protein aggregation, could be a big mistake.

      1. Hap says:

        When does it get falsified? At some point, sticking with a hypothesis when it doesn’t work stops being “persistence” and starts being “immune to data”.

  9. IrishNeuro says:

    I tend to agree that not treating early enough is the problem. There is too much genetic and pathological evidence to dismiss Abeta, in my humble opinion. However I also believe that Abeta will never be viable as a therapeutic target. It starts to accumulate up to 20 years before disease onset, and presumably sets off a series of events that means by the time we come to treat patients, Abeta is no longer the problem. Its like removing a thorn after you have already got sepsis. Not going to help. Even the most ambitious preventative trials in fAD populations are starting way after ABeta has started to accumulate. Also, just from the perspective of somebody in drug discovery – good luck trying to find a CNS drug with so few side effects that somebody “healthy” will happily take it for 10 years before they are symptomatic for AD and feel the need to take it.

  10. ElonMusk says:

    AI will figure this out… computer learning can find the trends we cannot. Maybe the value of the Ph. 3 failures will be the clinical data and associated metadata that goes into the AI models??

    1. buzzwords says:

      I think big data will definitely help.
      Maybe we should set up and AI to investigate the connection between the microbiome and Alzheimers.
      Then use photoredox chemistry for late-stage functionalization of drugs to prevent alzheimers.

      So many nature papers are to be had

      1. PhotoDeTox says:

        Very disruptive innovation!! Who wants to invest in your company?

      2. Emjeff says:

        Buzzwords, take my money!! ;^))

      3. mfernflower says:

        photoredox peroxynitrite chemistry for late-stage functionalization of drugs to prevent alzheimers thank you very much

    2. Christopher G. Janson says:

      This is not a problem of pattern recognition, which would be the application of AI. The real problem is that we lack empirical high throughout models to test pathways and networks of diverse cells. Similarly, we lack the exhaustive level of fine detail (the input) to make AI analysis useful.

  11. anon the II says:

    I’m still trying to figure out how these antibodies were supposed to work on the other side of the blood-brain barrier. Antibodies in the blood, beta amyloid in the brain, separated by the blood brain barrier and never the twain shall meet. What in the world am I missing here?

    1. Fuh Dge says:

      I have the exact same question.

      I know people are working on pretty clever ways on getting large molecules across the blood-brain barrier. This is outside of my wheelhouse, but people have used bispecific antibodies with low affinity to the transferrin receptor and high affinity at their target to shuffle large molecules across the BBB via transcytosis. Here’s a paper from Genentech and a pretty funny youtube animation they made about the discovery:

      However, I don’t think think aducanumab is one of these antibodies, so I have no idea.

    2. 10 Fingers says:

      It is a weird fact – but a fact nonetheless – that antibodies have a reliable B/P of 0.001. And, often, that they can be often be safely dosed wildly higher than their effective affinity for their target and the concentration of that target in brain.

      So, for an amyloid clearance antibody, shoving enough of it into the brain will actually result in it binding amyloid and dragging it back out of the brain. They reduce amyloid burden. That’s not the problem.

      The problem is that doing so doesn’t help people with Alzheimer’s Disease.

  12. Fuh Dge says:

    How much of these antibodies get into the brain to engage their target? Or are they made to bind up peripheral amyloid beta oligomers and create a chemical gradient promoting more efflux from the CNS? I saw in the 2016 Nature paper that, regardless of how it works, it does lower plaques in the brain – just curious as to how and if that mechanism puts a ceiling on its efficacy in reducing plaques.

    As a point of reference, I know that only fractions of the plasma concentration of anti-CGRP antibodies like galcanezumab, which ostensibly need to act in the PNS/meninges because of the low plasma stability of CGRP, get into the CNS/PNS in preclincal models (<1% into the CNS, ~4.5% in the trigeminal ganglion and ~11% in the dura mater).

    1. MrXYZ says:

      In healthy individuals, probably about 0.1 to 1% of antibody gets into the CSF but infinitesimal amounts get into the CNS. This is, of course, without any targetting, e.g. using transferrin receptor to mediate transcytosis.

      The argument is that in individuals with substantial amounts of brain inflammation, the blood-brain barrier will be compromised and that therapeutically useful amounts of antibody will get into the brain. My understanding is that some of these antibodies have been shown to reduce plaques (using PET imaging) but have just not shown any clinical improvement. Given this, I would say the failure of anti-amyloid therapies is mechanistic, not pharmacokinetic, in origin.

      1. johnnyboy says:

        That 2016 Nature paper showed a reduction in amyloid plaques in the brain of prodromal or mild AD patients, by PET. I guess it depends how much you trust the authors, and how much you trust PET (which is not exactly a high resolution imaging approach) as a reliable method to measure microscopic changes in microscopic lesions like amyloid plaques. I would assume that the subtleties of PET scanning methodologies and their interpretation would go over the heads of 99%+ of Nature peer reviewers.
        Note that the same paper also showed a reduction in cognitive decline in these patients, which now has been disproven. So make your own conclusions about the validity of the paper, Nature or no Nature.

  13. steve says:

    Not in the field but it seems to me that all the amyloid data are consistent with it being a result of the underlying pathology, not a cause. One can scrape off all the psoriatic plaques you want but you won’t cure the disease. Why would one think that removing amyloid will cure Alzheimer’s? Clearly some new thinking is needed:

  14. Wavefunction says:

    The amyloid hypothesis has turned into the Norwegian Blue of drug development. It is an ex-hypothesis.

    1. 10 Fingers says:

      “No, it’s resting….”’

      Laughed out loud at the reference…

    2. Christopher G. Janson says:

      Yes, but Always Look On The Bright Side Of Life, as they say…maybe the Norwegian Blue Parrot is like Schrodinger’s Cat..the parrot is alive and dead at the same time. Something to think about, Wavefunction. Cheers!

  15. C99 guy says:

    All the genetic arguments in favor of beta-amyloid also apply to C99, an intermediate in the production of amyloid. C99 levels increase only in vulnerable areas in AD patients and correlate with cognitive decline, very different from amyloid which levels increase everywhere in the brain. Counter-argument is that BACE trials have also failed.

  16. Hap says:

    I keep hearing “Does the plug ever come out, or not?” when I read this post.

  17. Peterw says:

    Sunk cost fallacy + decision paralysis in these large orgs + most importantly, immense rewards in store for those who – even marginally – push the needle on AD go a long way here.

  18. An Old Chemist says:

    Today’s Biospace has also covered this news, although, as always, the write up above by Derek is more scientific and elaborate:

  19. Gene says:

    CNN is spinning this one real hard:

    Another promising Alzheimer’s drug trial ends in failure: ‘This one hurts’

  20. dave w says:

    I wonder if amyloid is a downstream manifestation, and the actual causative process is something else – some upstream inflammatory (?) damage that results in the appearance of both amyloid and senility. (So going after the amyloid itself might be like trying to treat a heart attack by giving analgesics for the chest pain…)

  21. Marc says:

    What I dislike most is that alot of the new genetic hits that track with AD, i.e. Trem2, ABCA7, are mostly evaluated on what they do to amyloid-beta.

    Genetics of AD are pretty interesting, in that they point to APP-processing, microglia, and lipid metabolism. Why this all must boil down to “Are plaques up?” or “Is amyloid getting cleared?” in follow-up cell biology studies is beyond me.

    1. Diver Dude says:

      Because you need a take home bullet point for senior management.

  22. TheUnknownChemist says:

    Of course everyone is thinking wrong on amyloid. Where’s Lane S? Looking at amyloid is like waking up after a wild party with a house littered with beer bottles and blaming the mess on the beer bottles. Instead of the drunken mitochondria that generated them.

    Put a nail in it already.

  23. Martin Hofmann-Apitius says:

    You may want to have a look at this paper first-authored by Alpha Tom Kodamullil:

    Nat Rev Drug Discov. 2017 Dec;16(12):819. doi: 10.1038/nrd.2017.169. Epub 2017 Oct 23.

    It is time to think of alternatives.

  24. Mr Stink says:

    This isn’t exactly on topic, but it’s interesting none the less. Parkinsons sufferers apparently have a distinctive smell that is detectable years before other symptoms are visible. It might form the basis of an early diagnosis.

  25. PorkPieHat says:

    Some 50 plus comments in, and no one has mentioned the potential neuroinflammatory (bacterial, particularly P. gingivalis) origin of Alzheimer’s disease.

    The work Cortexyme and several universities have very recently added to their publications in this regard, and reported finding the two toxic enzymes that P. gingivalis uses to feed on human tissue in 99 and 96 per cent of 54 human Alzheimer’s brain samples taken from the hippocampus (Science Advances, These protein-degrading enzymes, gingipains, are found in higher levels in brain tissue that also had more tau fragments and more cognitive decline.

    This appears to be a most promising line of research.

    1. I was wondering the same thing. The data and conclusions from that study lead one to wonder if AD is not a constellation of diseases with bacterial infection being one of the modalities.

      1. PorkPieHat says:

        There should be a way to also do correlative population studies: are large numbers of older people with gum disease for a given period of time going to have a proclivity towards Alzheimer’s disease?

        1. I’m a dentist that started testing patients with OralDNA for p.gingivalis. Oddly, there are patients who do not have gum disease, that do have the bacteria. It is possible to have resistance to gums disease, but it can still cause other problems systematically. I happen to test myself, and found it to be true personally.

          TO my knowledge, this bacteria is the only one that directly produced gingipains. Once the bacteria is there, there is no way good homecare, or dental scalings can remove it, since it embeds into the tissue.

          And it is also possible to spread the bacteria thru kissing, or even if toothbrushes touch. So once it is gone with treatment, which can be 2 antibiotics at the same time, your partner could reinfect you.

  26. luysii says:

    There are 3 possible explanations for any microscopic change seen in any neurologic disease. The first is that they are causative (the initial reasonable assumption). The second is that they are a pile of spent bullets, which the neuron uses to defend itself against the real killer. The third is they are tombstones, the final emanations of a dying cell, a marker for the cause of death rather than the cause itself.

    There is evidence that the aggregates of SOD1 (superoxide dismutase) found in ALS are actually protective — Proc. Natl. Acad. Sci. vol. 115 pp. 4661 – 4665 ’18 . For more see —

  27. Surfactrant says:

    Worked in this field for some time now, and I’m always struck by the ‘silver bullet’ thinking that much of the community engages in. That AD has one causitive agent and removing that one agent will lead to disease regression or halt of progression.

    I think that if this were the case maybe at least some progress would have been made in the last hundred years, perhaps onset may/must be triggered by multiple (still unknown?) factors.

    1. Roger Q says:

      Completely outside the field, but,
      1.) in a very limited anecdotal observation, it seems many old doctors end this way.
      2.) the cure of ulcers via the correct antibiotic overturned a near 100 year belief that “stress” was the precipitating cause.
      3.) Remember, I’m a Geologist, so I don’t fear looking stupid.. My money is something is going on in the guts, (where, I’m told serotonin originates).

    1. StumpedByTheCaptchaMath says:

      Truly mind-boggling! If I were a long holder, I’d probably just start demanding they return cash on hand to shareholders and dissolve the company. They obviously can’t be trusted to put the capital to get use!

  28. RB says:

    You’ve removed the vandals from your house and cleaned up the destruction and mess… but the wiring and plumbing are gone. Clean house though.

  29. RB says:

    A colleague of mine described how back in the early ’90s he visited the McGill Brain Bank, where he was shown an array of brain slices (light microscopy) with varying loads of beta-amyloid deposition, some of which were truly packed with the stuff. There was no correlation, however, as to which brains came from dementia patients. Some of the largest beta-amyloid depositions were from non-demented “normal” brains.

  30. rb says:

    A colleague of mine described a visit to the McGill Brain Bank in the early ’90s where he was shown brain slices from a range of beta-amyloid loads (light microscopy). There was no correlation between the load and degree of dementia. Some of the brains with the heaviest burdens of beta-amyloid were from non-demented, normal aged brains.

  31. us10 says:

    I am closely following the cortisol hypothesis and the recently completed and highly anticipated (results within a month) phase II study of people with mild alzheimers from Actinogen Medical Australia, drug name Xanamem. Their drug targets cortisol levels in the brain. High cortisol levels have been well correlated with cognitive decline, dementias, memory loss and numerous other related conditions.

  32. R. Sarangarajan says:

    Given the complexity and multifactorial etiology leading up to the fairly well documented observation of amyloid in AD, the surprising part is the focus on developing a magic bullet (clearly this does not appear to be the case) and trying it make it work in many different ways. The most disheartening aspect of the failures is the reaction to the amyloid hypothesis and the potential poisoning of investments in future drug development remotely connected to the theory even in the presence of novel/new science to support it in the distant future, reminiscent of the storyline for the heat shock proteins in oncology. Agree on the need to think and do things differently!!

  33. rtah100 says:

    For anybody approaching the disease mechanism with an open mind, you might want to invite the cell cycle theory to the “Amyloid is Dead” party.

    The cell cycle theory is that A-beta outside the cell and phospho-tau inside the cell are downstream phenomena that are the result of CNS neurons re-entering the cell cycle (which they do as part of normal synaptic remodelling). In healthy patients, the neurons cannot pass out of the G1 phase and shuttle back to G0; in Alzheimer patients, an incompetent G1/S checkpoint permits the neuron to pass irreversibly into the S phase. The characteristic altered biochemistry of the S-phase, when the cell prepares for mitosis by disassembling the cytoskeleton (increased [tau]), producing extracelular APP and upregulating kinases: lo and behold, tau and APP get phosphorylated and, after years stuck in what should be a transient S-phase, tau tangles and beta-amyloid result. There is extensive neuropathology evidence to show these phenomena, including S-phase DNA replication, and their correlation with the progression of AD through the different brain structures.

    There is also evidence that the G1/S checkpoint incompetence is present in Alzheimer patients genetically in all their cell types and can be the basis of diagnostic tests based on phenotype, e.g. culturing viable lymphocytes and then challenging them with rapamycin and cells in AD patients keep dividing because their brakes have been cut, and with genetic testing. A whole bunch of cell cycle modifying drugs may have repurposing potential in AD and it is certainly worth trawling through the effect on AD incidence of many common drugs with cell cycle side effects.

  34. Kelvin says:

    Regardless of disease, amyloid forms beta-barrel pores in the cell membrane which causes all sorts of problems as the cells struggle to maintain homeostasis. No antibody will be able to remove these pores once formed.

  35. Chris says:

    There has been some recent studies suggesting that reverse transcriptase might be relevant—even that it might be one of the reasons why the many trials focusing on AB are failing (there was an article in Nature by Jerold Chun on this, and then recent study from Brown using HIV drugs pointed to related issues). But I don’t see people talking much about that in response to the failed AB trials. Chun noted that people taking HIV drugs that inhibit reverse transcriptase don’t seem to get Alzheimers. Is there some reason that this information seems more interesting to me as a layperson than the specialists? I believe the study noted that during a review of 100,000 HIV drug takers over 65, only 1 person has been documented with AZ (when statistics would suggest 2000-3000). Along with the fact that reverse transcriptase in the brain also suggests a rationale for why the AB approaches are failing, I was wondering if someone might offer some thoughts on why this is not getting more discussion as these other drugs fail?

    1. Chris says:

      Actually, I believe Chun noted the point about the number of patients taking HIV drugs with AZ disease in this video (near 3:10), rather than in article:

  36. An Old Chemist says:

    More on Alzheimer’s from the Latest Biospace:

    Now, What’s Next in Alzheimer’s? For Biogen and Eisai, Another Phase III Trial

  37. Claudiu Bandea says:

    “Something is wrong with the way we’re thinking about Alzheimer’s and amyloid, folks, something is wrong. It’s been wrong for a long time and that’s been clear for a long time.” A putative explanation:

  38. Christopher G. Janson says:

    I’m in agreement with Derek that these antibody drugs always had a negligible chance of working, and it is unfortunate that so many resources have been spent on this narrow avenue. However, in my opinion, this failure of drug development has nothing to do with a failure of the amyloid hypothesis, and much to do with a failure of basic pharmacokinetics. Plus a failure of mechanism, as “Mr. XYZ” pointed out above. It was a failure to understand that the blood-brain barrier and blood-CSF barrier are not as simple as imagined, and the physiology of amyloid clearance is years away from being well understood. At least it can be said that this drug failure was not a “Failure of Imagination,” in the words of Arthur C. Clarke. I’m reminded here of his book “Profiles of the Future” in the chapter titled “Hazards of Prophecy.” He posits “Clarke’s Law” which says: “When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong.” My hunch here (like Biogen’s), speaking as a rather young scientist, is that amyloid does in fact play a critical role in Alzheimer’s disease, but it is far more complicated than tossing some antibodies into the bloodstream and expecting a major impact on the buildup of peptides in a multitude of discrete compartments in the most complex organ of the body. There are missing links in this story which await clarification. This failure is an opportunity to reflect a bit harder on the physiology of the brain barriers, which may give clues as to the best solution of this problem. While I think it is the height of arrogance to dismiss the amyloid hypothesis, it is also the height of arrogance to propose that it adequately explains the complex pathophysiology of Alzheimer’s disease. But ignoring it entirely comes with other perils. I predict that the amyloid hypothesis will continue to play a key role in the search for effective drugs, but that underlying metabolic and energy-dependent processes are equally important, as well as other currently unknown factors. This most recent drug failure, if nothing else, should be a clarion call for more funding on basic brain research. So if Derek means by “do something else” that we need to expand our horizons, I agree 100%. Call me an optimist, but I tend to look on “failure” in the same way as Dr. Luis Perez-Breva. If done frequently and properly, it leads in the right direction. It is merely an opportunity for future success. Is that a Norwegian Blue I hear singing?

    1. Derek Lowe says:

      I think that we agree that the problems with the amyloid hypothesis in the clinic should send people back to doing more basic research. That’s not a sexy or immediately profitable recommendation, but I think it’s the right one. Remember, though, that it’s not just antibodies: every secretase inhibitor (presumably upstream of all the oligomers, etc.) has failed as well. We’re missing something, and how much of the classic amyloid hypothesis will be left standing once we grasp the missing details is a matter of debate. My own opinion: I think that a lot of it will come down, leaving a non-causative role for amyloid in the disease. But we’ve got to go back to the proverbial drawing board to find out, for sure. . .

  39. Christopher G. Janson says:

    Agreed. But is “failure” of a secretase inhibitor due to off target effects and intolerance of a potentially therapeutic dose the same as “failure” due to a total lack of effect? One challenge of small molecule drugs is that they often have myriad and dose dependent effects. My therapeutic bias is toward gene therapy and other biologics, while realizing that delivery will be the challenge for those “drugs.” By the way, I think something that naysayers need to explain is the common pattern of dementia among other diseases in which amyloid and tau are involved, looking beyond Alzheimer’s (viz. my comment above on BRI). I think there are more unknowns than knowns with amyloid beta and it’s definitely part of the problem. Time will tell. We’ll see if “Clarke’s Law” is predictive. Exciting time to be a part of this story, and eventually we’ll get it right.

  40. MrXYZ says:

    For the folks who think that the various AD clinical trial failures are due to failures of the molecules and not failures of the therapeutic hypothesis, what would a (realistic) clinical trial have to show in order to convince you that the hypothesis was incorrect? By therapeutic hypothesis, I mean that aberrant Abeta production and plaque formation causes AD and that reduction of plaques should lead to reduction of disease.

    1. Christopher Janson says:

      If other types of APP modulation (including alpha secretase modulation) were to have no beneficial effect in humans, and if modulating other identified amyloid interacting targets within specific regions of the CNS were to have no effect, including approaches which indirectly improve brain clearance of certain abeta isoforms, then I would be convinced that the fundamental underlying theory needs revision. But to focus on a couple of narrow anti-amyloid approaches and then to just give up is premature. There are literally thousands of other approaches which should have been looked at before putting all of our eggs in the basket of antibody therapy against abeta, or the broad sword of gamma/beta secretase inhibition. Anyone who considers these data to “disprove” the modified amyloid theory simply lacks imagination. By modified amyloid theory, I mean that deposition of “amyloid” peptides both intracellularly and extracellularly in neurons and other cells is causally linked with the initiation and progression of dementia. I’m curious, if you believe that abeta peptides are a benign epiphenomenon, how would you explain similar diseases in which amyloid and tau appear to play a role in neurodegeneration, such as British Familial Dementia, where the “amyloid” is not generated by abeta? There are common themes and biochemical similarities which are compelling and support the role of diverse types of amyloidosis in neurodegeneration. The biochemistry is distinct, but a number of early triggers appear to lead to dementia. The data from familial Alzheimer’s cases and Down syndrome is likewise supportive of a common pathway. There are many testable hypotheses of merit.

  41. MrXYZ says:

    I don’t dispute that there is a relation between Abeta deposition and AD – the familial forms of the disease certainly point in that direction; what I want to know is whether the deposition is what causes the disease or whether there is an upstream process that causes both the cognitive decline AND the Abeta deposition (i.e. deposition tracks with disease but is not causal per se). Despite all the criticism (much justified) of the various antibody and BACE trials, some of them showed substantial decreases in plaques without any corresponding decrease in symptoms. Perhaps we are treating too late but at some point you have to say that decreasing Abeta deposition is not a therapeutic approach. Or to put it another way, a successful AD treatment may reduce plaque formation but plaque reduction won’t necessarily lead to a successful AD treatment.

    Personally, I am intrigued by the microbial hypotheses that are floating around with Abeta essentially playing a role in the innate immune response.

    And yes, there are many other diseases that show various types of amyloid formation. I find Huntington’s a particularly compelling case. However, with the possible exception of TTR given above, is there a case of a therapy that reduces amyloid deposition and that leads to believable and meaningful therapeutic response (even if it’s in just an early clinical trial)?

    FYI, my comment was not meant to be snarky but was trying to find a way of stating the Abeta hypothesis of AD in a form that can be disprovable by a realistic set of experiments/clinical trials.

    1. Claudiu Bandea says:

      I think the points you are making, along with those by Christopher Janson above on the involvement of ‘amyloid’ in several other neurodegenerative diseases that have enigmatic etiologies and lack effective treatments, might be one of the strongest clues in understanding their putative nature:

  42. Chris Janson says:

    Another key point here is the issue of extracellular “plaques” vs. intraneuronal or intrasynaptic abeta species (which antibodies have no effect on), which have been shown to be much more predictive of cognitive decline. Not to mention the fundamental issue of all of the different isoforms and terminally modified forms and complex physical conformations of abeta and how they interact and are processed and transported. When we talk about “amyloid” we are actually talking about potentially thousands of different peptides. There are many questions which remain unsolved. Regarding the different “flavors” of abeta, look at work by William Klein and Karen Ashe and David Teplow and Charles Glabe and many other great scientists. Abeta is legion! To focus exclusively on so-called “amyloid plaques” rather than abeta oligomers or conformers is a common fallacy. It’s like a chef lumping “chocolate” into a single monolithic category, without specifying if it’s white chocolate or bittersweet chocolate or chocolate sauce or chocolate bricks or chocolate flakes or whatever. You can’t make a decent chocolate souffle without knowing your ingredients and properly controlling all the variables in the oven. One final word of wisdom — a few years ago, there were multiple “failed” endovascular device trials published in leading journals such as NEJM, the Holy Grail of medical publishing. All the neurology pundits had already written off endovascular devices for stroke therapy, except for practitioners in the field who actually saw lives being saved every day. These multiple study “failures” and a widespread, acrimonious call to stop using the devices in clinical practice were suddenly followed by a string of positive trials which have now revolutionized the entire field of stroke and vascular neurology. The same may be true for Alzheimer’s drug development. Pundits are often wrong.

  43. former biogen employee says:

    Woohoo, Go BIIB!!!

  44. Brian says:

    Looks like the study was halted too soon?

  45. KS says:

    What the hell just happened…

  46. DrSAR says:

    Can someone please explain how Biogen is going to get it approved in a year? Derek, maybe you could write a follow-up if there is more information (which I haven’t found with minimal search) on aducanumab’s status as a viable AD drug?

    1. Derek Lowe says:

      Tomorrow’s blog post, absolutely. . .

  47. Chris Janson says:

    As everyone has heard by now, the latest data on one of the monoclonal antibody drugs show a probable treatment effect. Witness the sudden reversal of course by Biogen after re-analysis of their own “futility” analysis. Looks like the Norwegian Blue is alive. Derek has updated his blog on a separate posting, and tempered his pessimism a bit. At the end of the day, I agree with Derek that proving a clinically meaningful effect of any monoclonal antibody treatment still has a long way to go. In my opinion, we need to fully explore other specific mechanisms of brain clearance, aside from this one alone, which may not be the best. Exciting times to be in the field!

Comments are closed.