Skip to main content

Alzheimer's Disease

An Alzheimer’s Statement

“Despite this latest setback, Eli Lilly remains committed to plunging through this concrete wall headfirst. This is a sad day for our WallBreaker 2020 program, and some of our longtime head bashers will recall similar periods during SkullButt 2012 and ConcreteCracker 2016. But we continue to believe that the only way past this wall is straight through it, with all the force that our craniums can bring to bear.

Our partners in this latest attempt, AstraZeneca, realized in 2014 the importance of plunging with great force into the same structure. But they said “Hey, our heads aren’t just going to bash themselves into this massive edifice – we need to take a big running start and come in with perfect 90 degree orthogonal form.” And they knew to come to the leaders in the field: Eli Lilly. After all, our gamma-secretase inhibitor completely failed in 2010. Then we took our antibody, solanezumab into Phase III trials that failed in 2012. And found out in 2013 that our beta-secretase inhibitor failed. But then we kept plunging on and reported almost identical results for solanezumab in 2015 and 2016. No one knows more about direct, head-on contact with the amyloid hypothesis than Eli Lilly.

So now the Lilly/AstraZeneca beta-secretase inhibitor has failed in turn. But we remain steadfast. If there is any way left to send our ourselves full-tilt into another failed amyloid trial, our pledge to you is that Lilly will find it. There is somehow enough money to fuel the cash-burners that power the steam-driven catapults. We have plenty of helmets, whose dents are even now being pounded out. And most of all, we have our beloved wall, which we shall never forsake. Higher velocity! More power! Once more into the concrete, my friends! Who’s with me?”

Editorial note: as mentioned before, I actually have a lot of respect for Lilly putting as much time, effort, and money into their Alzheimer’s work as they have. But geez. At some point, the dedication to one particular hypothesis, in the face of mounting evidence of trouble, becomes a bit less deserving of admiration

Neither do I intend to make light of the situation faced by Alzheimer’s patients and caregivers (I’ve seen what it can do). But anyone who was holding out for this trial to come through with a cure was sadly misinformed. The amyloid hypothesis is in deep trouble. folks. It’s been in trouble for years. Multiple attempts (and not just by Eli Lilly!) to turn it into something of clinical utility have failed. Biogen, I’m talking to you, too. You have (of all things) an amyloid antibody deep into clinical trials that you have bet a substantial part of your future on. How did you let this happen? Are you happy about it? Are you optimistic?

And all this goes for any other group that’s betting on amyloid inhibition of some sort to make a clinical difference in Alzheimer’s. Yes, yes, unmet medical need, yes, huge market, yes, all that. But look around you. Are you optimistic? And if so: why?

76 comments on “An Alzheimer’s Statement”

  1. JL says:

    Perhaps Eli Lilly can convince prospective patients to take their Alzheimer’s drug candidates (and blunt the cash burn in the process?) under the new ‘Right to Try Act’?

    1. SIrWired says:

      I think this is a huge danger of Right-to-Try. It’s almost begging for charlatans to buy drugs that failed Phase 3 on the cheap, and sell them to desperate patients indefinitely while pretending to run clinical trials destined for inevitable failure, if they were to ever complete, which they won’t.

      1. JL says:

        Indeed! My comment was made with a certain level of sarcasm intended…..

    2. Anon says:

      With 2 Phase 3 trials each required to give a p-value of 0.05, they only need to test about 1/0.05^2 = 400 drugs before they get one approved by random chance alone. They must be pretty close to that by now… so let’s see:

      400 drugs x $500m per drug = $200bn cost, which is probably still less than the money they’d make with the ineffective drug they’d eventually get approved.

      Maybe not so dumb after all.

  2. Some Dude says:

    Biogen still thinks their antibody will work as it some magic additional properties that all the other antibodies don’t. They might have a shot as their antibody comes from a library from healthy older people who might be naturally immune to Alzheimer’s. But it does not look too rosy:

    1. loupgarous says:

      Biogen’s got a hypothesis. They’ll have to go through clinical trials in a human disease with slow disease progression to validate it. Hopefully, they’ll take the hint if/when their hypothesis (not exactly identical to Lilly’s) isn’t affirmed.

  3. Glen says:

    My hope is that an open review of the data will give someone a new insight. In aggregate, these studies contain a detailed record of the biological/biochemical progression of Alzheimer’s. The proper work now is to sift that data for clues, and turn to different areas of research.

    Perhaps the collective data set will be a good test platform for advanced computer analysis.

    1. HFM says:

      I’m a data scientist who does that kind of thing – and yeah, there have been major wins, where we’ve salvaged a drug that would have cratered (or already has!) in a general-population trial. You’ll semi-routinely see cancer drugs that cause miraculous remissions in a handful of patients, but are useless or worse in the other ninety-some percent. If you characterize the daylights out of these people, it’s usually possible to figure out why it’s happening and identify biomarkers that will let you prospectively identify the people who would benefit from the drug.

      But you actually need a sub-population where the drug is effective! Alzheimer’s isn’t my field, but my understanding is, that’s not the case – these treatments are just generally useless, for everyone. You can’t big-data your way out of that one, sorry.

      1. Glen says:

        You are indeed correct as far as finding successful treatment. I was attempting to suggest using the data to carefully study the progression of the disease(s), and postulate possible causes. Because of the recent nature of the studies, and nature of modern EMR’s, the data set contains recent and accurate information. I do believe that data can be useful in more basic research.
        Until there is a better understanding of the disease process, I really don’t think that expensive drug programs are warranted.

      2. Mansour says:

        It does complicate it!

    2. Datasharingiskey says:

      Here are two examples of data sharing in AD that are already up and running. The second example has a clinical trial simulation tool based on shared historical clinical trial data.

  4. Isidore says:

    Why is the amyloid hypothesis so resilient on the face of multiple failures of drugs that were developed on this basis? Is it because of entrenched interests, i.e. academics with large grants who have championed it for decades and are not willing to part with their hypothesis and the money that it has been generating for their labs and for themselves in consulting fees? Or, if they have no such base motives, is it due to reluctance to admit that they were wrong because of what it will do to their reputation? (Personally I think that concern for money is a much more reasonable consideration than vanity, but then I am not an academic). Or is this obsession (for it appears that this is what it has become) with the amyloid hypothesis a reflection of the dearth of any other mechanisms that would lend themselves to drug development? In which case it becomes something like “if the only tool you have is a hammer than everywhere you look you see nails.”
    These are not rhetorical questions, I am curious as I worked in this field (peripherally) back in its “glory days”, when the secretases were being discovered and before all the failed clinical trials.

    1. Emjeff says:

      I think it’s like that old joke where the guy says, while hunting under a streetlight, that he lost his keys in the house. When asked why he’s looking under the streetlight, he says: ” the light’s better out here.

      Same thing with the amyloid hypothesis. There are no “keys” to be found under that particular light, but by God, we’re gonna keep looking, because it’s so bright.

    2. Harrison says:

      It is really quite unbelievable how much goal-post moving has occurred over the past quarter century. Amyloid Plaques are critical! Nevermind, it’s soluble oligomers! If only we treated earlier! The conclusive demonstration of the hypothesis needs to be in the 1% of patients with APP or PS1 mutations!

      It certainly seems like there are entrenched efforts to keep the amyloid hypothesis alive. At this point I imagine it is a combination of vanity and what’s left of grant money for studying the amyloid cascade.

    3. Ray Truant says:

      It’s difficult to turn a huge ship that has 20 years of dogma invested. Alternative hypotheses of AD are poorly funded, and there is a self-propagating feedback loop of peer reviewers and applicants, funding what they think they understand, circling around model systems designed to show what they want to see.
      Huntington disease was all about proteostasis mechanisms for 25 years, based off the amyloid -like hypothesis of polyglutamine-mediated misfolding. Yet, in HD GWAS finally completed in 2015-17, in over 9,000, proteostasis pathways were absent, the answer was DNA repair and redox control pathways, for which no hypothesis ever existed. Sadly, this has NOT refocused research efforts, and these diseases still continue to silo their research from each other without learning lessons from monogenic age-onset neurodegenerative disorders.

    4. Mark Thorson says:

      An important reason is that every one of the mutations that cause early-onset AD are in the amyloid precursor protein or the enzymes that process APP. Any alternative hypothesis must explain why that is so.

      1. Crocodile Chuck says:

        B-b-b-but you said it was amyloid all the way down, Mark! 😉

      2. sgcox says:

        There have been numerous studies, for example this one
        that FAD PSEN mutations impair normal psen function, not activate it.

        1. befuddled says:

          Yet even the article you cited confirms the fact that FAD PSEN mutations increase the deposition of beta-amyloid.

          Let’s face it, the genetic data indicates an important role for APP processing (which in AD leads to amyloid deposition) in Alzheimer’s disease. But the clinical data contradicts it. It’s a messy situation, and other than going after Tau, there aren’t many obvious alternatives.

          1. Bad analogy says:

            If the fleas are on rats in the back of a grain cart then stopping either the cart (break a wheel) or the horse (break a leg) would immobilize the cart and prevent the spread of pestis. However, if the fleas are on the horse then stopping just the cart won’t do anything to stop the plague. In that case an effective treatment, stopping the horse, would still leave behind an immobile cart.

            In the end, I don’t think it matters if a proposed target is able to explain amyloid aggregation. What matters is if it works.

          2. They CallMeTheSeeker says:


            If you had invested as much energy in AD as you did in that mess of a post we’d have a cure by now

      3. Andre Brandli says:

        In addition, patients suffering from trisomy 21, will develop invariantly AD by the time they reach their early forties. The APP gene is located on chromosome 21 and thus these patient have three copies of the APP gene instead of two in their genome. In fact, patients with a selective duplication of the APP gene have been reported that will also develop AD. In summary, the genetic evidence is for APP to be central for the development of familial AD is overwhelming in my humble opinion.

        1. Harrison says:

          APP mutations unquestionably lead to the familial AD cases that represent, at most, 5% of all AD cases. The real question is if the pathobiology from FAD cases translates to the 95% of cases that are “sporadic.” Of course genes, such as APOE4, may account for up to 60% of “sporadic” cases, but the amount of research money spent on apoE4 is dwarfed by the amount of money spent on the amyloid cascade.

        2. Ian Malone says:

          The link between Down Syndrome and Alzheimer’s Disease has long been known, and is tantalising for that reason. There’s so much going on genetically with people with DS that it’s hard to draw a definite conclusion however. Despite that it’s another piece of the puzzle. Not mentioned yet, ApoE alleles, ApoE-ε2 is protective, ApoE-ε3 is the common form, ApoE-ε4 raises your risk of AD, with a dose effect of number of copies. It’s the biggest non-autosomal dominant genetic determinant for Alzheimers, and the protein is responsible for breaking down beta-amyloid. (But of course, it’s responsible for other things too, so until we have a complete understanding of the disease process we don’t know for sure why it’s protective.)

    5. fajensen says:

      Max Planck explained it in this way:

      A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.

      Anyone remember stomach ulcers?

      Used to be caused by all sorts of things “stressful”, then “too much acid”, obviously radical surgery was needed and then one day some “crazy Australian charlatans” proposed it to be caused by a bug, which indeed it was, and only 20 years later this was grudgingly accepted as “main-stream” knowledge.

  5. Tourettes of Chemistry says:

    ‘…holding out for this trial to come through with a cure was sadly misinformed…’


  6. Tourettes of Chemistry says:

    ‘…holding out for this trial to come through with a cure was sadly misinformed.’

    Different Day, Same Story:

  7. drsnowboard says:

    I’ve often thought of drug development / clinical trials as being tied to a railway track . You can see the train coming from a long way off, but you are relying on the points set way back to keep you safe. By the time you can feel the vibrations in the track, there’s nothing you can do…

    However, the amyloid hypothesis seems to be running everyone’s train down the track where the bridge is out..

    1. Derek Lowe says:

      Good analogy. I’ve used a similar one, where I talk about having a perverse model of car whose steering wheel you can only use at preset times that you program in at the start of your trip.

  8. luysii says:

    Well how about using flickering light to treat Alzheimer’s. The following is about a paper that made its way into Nature 2 years ago

  9. luysii says:

    Here’s one reason the amyloid hypothesis won’t die

    [ Nature vol. 487 pp. 153 ’12 ] A mutation in APP protects against Alzheimer’s disease. First the genome sequence APP of 1,795 Icelanders were studied to look for low frequency variants. They found a mutation A673T adjacent to the site that is cleaved by beta secretase 1 (BACE1) which doesn’t vary — it’s gamma secretase which cleaves at variable sites leading to Abeta40, Abeta42 formation. The mutation is at position 2 in Abeta. The mutation results in a 40% reduction in the formation of amyloidogenic peptides (which?) in vitro (293T cells transfected with variant and normal APP). Amazingly different variant at 673 (A673V) — increases Abeta formation. Alternative processing of APP at the alpha site prevents Abeta formation (as the alpha site is located within Abeta).
    The mutation was more common in the controls (.62%) than the Alzheimer group (.13%) but these were Scandinavians. It was found in 1/7,020 exomes sequenced genes in NHLBI group and 3/31,714 from a North American population using a SNP chip array.
    The variant increased the odds of reaching 85 by 1.47 fold. Cognitive function in 41 carriers of A673T ranging in age from 80 – 100, and 3,673 noncarriers living in nursing homes showed the carriers to perform better.
    A group of cognitively intact 85 year olds had slight enrichment (.79% vs. .62% in the general population).
    30 coding mutations in APP are known, of which 25 are pathogenic.

  10. cynical1 says:

    Actually, it’s not only Alzheimer’s. Look at how much money has been thrown at MBP as the auto-antigen in multiple sclerosis? And they are still doing it, to this day! Talk about beating a dead horse. And the NMSS continues to piss away countless money to the same twits for funding work on MBP as the auto antigen.

    The world would be a better place if stupid hurt.

    1. zero says:

      It hurts, just not the right people.

      1. cynical1 says:

        Point taken…..

    2. befuddled says:

      I’d be interesting in hearing your alternative theory for the immune reactivity in MS. Serious question.

  11. Eric says:

    Lilly and AZ started the AMARANTH phase 3 trial in September of 2014. Given the time it takes to gear up for a study like this, preparations probably began well before that. Did we know enough 4 years ago to say that BACE was a poor target? Or perhaps more relevant – did Lilly/AZ run this BACE trial at the expense of some better validated target? It’s much easier to see the failures in retrospect. I don’t honestly know what they should have done differently given the information they had at the time. Perhaps they could have killed the trial a year ago when Merck’s BACE inhibitor failed, but at that point Lilly/AZ was approaching their own interim analysis time point.

  12. AndrewD says:

    Derek, have you seen the article on Alzheimer’s in the June edition of Chemistry World? There is an interesting attempt to use PET scanning as a tool early detection of Alzheimer’s which could be a useful support tool for Pharmaceutical studies

  13. Daniel Barkalow says:

    I still think the reason the amyloid hypothesis hasn’t gone away is that people haven’t been publishing studies that should that their anti-amyloid treatments cleared the amyloid effectively without having any useful effects. By now, it’s almost certain that somebody’s treatment has the direct effect they intended, and the mechanism just doesn’t work, but there hasn’t been a lot of publishing results that exclude the possibility that amyloid is the culprit and it’s just a really hard target.

  14. Christophe Verlinde says:

    One piece in the amyloid accumulation puzzle that is under appreciated is the innate ability of the brain to REMOVE amyloid. A few years ago Maiken Nedergaard, a brilliant Danish neurologist, discovered the Glymphatic system in the brain (G stands for Glia). This system clears the brain of tons of undesirables, including amyloid. Remarkably this system operates ONLY at night.
    see: Xie, L., Kang, H., Xu, Q., Chen, M. J., Liao, Y., Thiyagarajan, M., … Nedergaard, M. (2013). Sleep Drives Metabolite Clearance from the Adult Brain. Science (New York, N.Y.), 342(6156), 10.1126/science.1241224.

    Now, here is my speculation. One of the ways amyloid accumulation could lead to plaques is if sleep is disturbed because the glymphatic system clearance will not be in operation. A burning question in this context is whether this system operates if sleep is induced by benzodiazepines and the like.

    1. luysii says:

      Another way to look at Nedegaard’s work is that the plaques in some way gum up the plumbing system (Glymphatics) in which case the Lilly work should have helped.–

  15. leftscienceawhileago says:

    Everything around “Once more into the concrete, my friends!” is amongst the funniest lines on this site. Right up there with this comment:

  16. Peter S. Shenkin says:

    If you do inhibit the disease, you will almost certainly inhibit aggregate-formation as well, since individuals who don’t have Alzheimer’s don’t exhibit the aggregates.

    So eliminating the aggregates may well be a necessary consequence of any treatment that inhibits Alzheimer’s; but now we know that elimination of aggregates is not sufficient to inhibit the disease.

    So if I have a drug candidate that eliminates aggregates, it could still become an effective drug, but not for that reason. And if I have a candidate that does not eliminate aggregates, it it probably reasonable to assume that it will not become an effective drug.

    I don’t know anything about current and upcoming drug candidates for Alzheimer’s, but if a drug candidate does inhibit aggregation formation, particularly in vivo, by some new and even unknown mechanism further back in the chain of events that causes aggregates to form, I think I would have a right to feel encouraged.

    1. Harrison says:

      I’m not sure what you mean by aggregates, but there are many individuals with amyloid plaques that do not suffer from dementia. It’s also well-known that amyloid imaged with PET does not correlate well with cognitive decline. The field is shifting to look at tau aggregates instead as more proximal to measures of cognitive decline.

  17. ignorant says:

    I ask out of total ignorance…
    Pharma continues to get ridiculed for working on amyloid drugs. Is it possible that the critics are missing something?

  18. Mark Thorson says:

    The problem is obviously the two previous PIII clinical trials came too late. You can’t treat a forest fire after the forest has burned. That’s why Lilly bought the company that makes florbetapir. With a study population that is earlier in the course of the disease — mild cognitive impairment — victory may yet be had. How sad if the effective drug was abandoned before being given a proper trial! Once more into the breach!

    1. Harrison says:

      I cannot tell if you wrote this with tongue firmly in cheek or with the upmost seriousness.

  19. Lame Simian says:

    something something peroxynitrites

  20. John Wayne says:

    We need enough people tossing things into the clinic until somebody gets lucky; only then will we start learning what will really move the needle for patients.

    Everybody who tells you that they knows the answer should not be believed; at best they have a reasonable hypothesis (at worst, a lot less.)

  21. Dr Llort Mudaphuk says:

    Alzheimer Disease is quite tricky from a therapeutic standpoint. Pfizer’s peptide, BR4556, never took off just because of BBB impermeability. Also, keep in mind of several faked data coming out soon and in the near future. But most importantly – is it tau or amyloid that causes the darn disease in the first place. Unless we figure these conundrums, I’m afraid people will continue to die from AD. Biggest market related question is if it’s worth saving people that old to begin with. In the long scheme of things AD tends to afflict people so old that they’re best recourse is to just check out…sad but true

  22. BiotechFanatic says:

    Hate to play devil’s advocate on a sinking ship – but isn’t there still room for a last-ditch hypothesis where amyloid plaques cause the damage underlying AD, but by the time they appear, the damage has been done and is essentially irreversible. I.e. treatments may have clinical benefit in some patients in a preventative capacity.

    1. Matthew K says:

      This is pretty much the mainstream position; it’s undermined (as johndoe below notes) by people whose brains accumulate significant amounts of amyloid during healthy ageing without it causing noticeable cognitive problems. However it is still a valid argument in my opinion.
      There is just no way around the fact that APP and PS1 mutations, and trisomy 21, bring down the disease at tragically early age accompanied by a flurry of amyloid plaques. Just because we don’t know and can’t drug the specific mechanism doesn’t mean the amyloid isn’t causal – it may well be causal in ways we simply fail to understand at this point.

    2. Barry says:

      causality requires that the responsible agent appear before the effect.

  23. nonchemist says:

    Here’s a nitpick from a different direction: This “maybe our drug can treat Alzheimer’s, if only we could provide it early enough!” line of thought is something of a slippery slope. I mean, it’s not like they’re taking in far-gone patients already, many of these trials (to my limited understanding) focus on mild forms to begin with.

    It’s possible to go even earlier, yes, but how far back do you really want to shoot for, especially with trial costs in mind? You can take in asymptomatic patients with aggregates and monitor the onset of Alzheimer’s, and if that doesn’t work you can hand a lifetime supply of pachycephalomab to some toddlers and see how they do after 60 years, and if that also falters you can always administer it to mothers and figure if epigenetic inheritance has anything to do with it…

    1. Mark Thorson says:

      After 60 years it would be way off-patent, so heck with that. Any drug that takes more than 5 or 10 years to test is a non-starter.

    2. Ian Malone says:

      I mean, it’s not like they’re taking in far-gone patients already, many of these trials (to my limited understanding) focus on mild forms to begin with.

      It depends what you mean by “far-gone”. Clinically, psychologically, no, my impression is recent big industry studies do not recruit very affected people. In part that’s simply because people with severe disease will have difficulty with compliance and long term follow-up, but there has for a while now been a strong desire to try to treat earlier in the disease process.

      Pathologically however, by the time someone has diagnosable mild Alzheimer’s disease, you can see it on a structural MRI scan. Enough atrophy has occurred that loss of tissue is noticeable cross-sectionally, particularly in the hippocampus, regardless of variation in size and shape across the population. People are still trying to nail down the sequence of events, but the Jack biomarker model is a start: regardless of what’s causative, there are macro changes years prior to a clinical diagnosis, and micro and metabolic ones possibly a decade or even two earlier. (See also papers on the DIAN study, Dominantly Inherited Alzheimers Network.)

      Hence the desire to try for earlier stages such as MCI, or even based on biomarkers such as PET or CSF, which may be the stage you have to intervene at regardless of whether the amyloid hypothesis is correct or not. Unfortunately that means long follow up times and large sample sizes to see if you made a difference or not.

      1. nonchemist says:

        Thank you for the detailed answer to this complaint.

        And I suppose the take-home message here is that Alzheimer’s drugs are essentially shooting in the dark until someone gets a proper handle on the degradation mechanisms (and given the scale of the problem, progress on that front will surely be painful and incremental, so sudden revelations here). Amyloid plaques, it would seem, are such a late symptom that clearing them is futile once they have shown up, but it is still possible that they’re the byproduct of an entirely different process that leads to cognitive loss, so that even early removal has little merit.

        All of this is also very concerning for patients who already display AD symptoms, who must face two dismal but near-certain facts: (1) an effective Alzheimer’s therapy may take a decade or more to validate, and (2) they may have missed the window for intervention even if it were available right now. Drug companies, likewise, must contend with the fact that already long-and-expensive Alzheimer’s trials must by necessity become even longer and more expensive.

  24. David Edwards says:

    What all of this is telling me, quite simply, is that once again, we don’t know enough about the underlying biology.

    Does anyone actually know, in a proper, rigorous manner, why amyloid plaques appear in the brain when Alzheimer’s starts wreaking its cognitive havoc? Or what they do when they form? For that matter, does anyone genuinely know what tau protein does?

    If these happen to be sideshows that simply make themselves manifest as a result of the real disease process doming something else, then going after those sideshows will achieve nothing. But do we actually know the difference between misleading symptomatic sideshows and actual disease mechanism in the case of Alzheimer’s? I’m hoping someone with a far better understanding of brain chemistry than I have, will step up to the plate and enlighten me, because although my understanding of brain chemistry is woefully limited, one piece of knowledge I have in this field that is reasonably secure, is that brain chemistry is achingly intricate. Understanding brain chemistry properly, makes writing 10 million lines of assembly language code and debugging it all, look simple by comparison, and debugging assembly language code happens to be one of the strings to my bow, so to speak, and I can tell you some fun stories about the woes involved if anyone cares to listen (though not many here might).

    Now I can understand that in a field that complicated, you’ll reach for any handle that looks as if it will bring a task even slightly closer to tractability. When one’s concern is a disease that is [1] a hideous spider web of intertwined enigmas, and [2] exacts an even more hideous human cost from those it affects, then there’s an entirely comprehensible tendency to reach for the best handle you happen to have at a given moment, even if that handle is only marginally better than useless. Progressing from that stage, especially if doing so successfully will constitute a research project covering decades, won’t just require a rethink of the utility of the handle you’re grasping, it’ll also require a rethink of your entire strategic approach.

    1. fajensen says:

      If these happen to be sideshows that simply make themselves manifest as a result of the real disease process doming something else, then going after those sideshows will achieve nothing.

      Not Nothing – Demonstrating that a well known model is wrong and maybe even explain how, that is also a very useful scientific result.

      I find “resistance to failure” to be the weak-spot in many Chinese research papers on Machine Learning, BTW. They don’t want to present a null-result so they jump through many (flaming!) statistical hoops to somehow show that their hypothesis “worked”.

    2. Ian Malone says:

      On the topic of tau, we do actually know what its normal role is. It builds microtubules that stabilise cells, particularly axons, and is normally inside cells. However if cells die or it gets phosphorylated or there are mutant forms, it aggregates, clogs things up, kills cells from the inside and escapes into the environment. What triggers that, how it spreads, why it’s involved in distinct neurodegenerative diseases (tau mutations cause frontotemporal dementia, not Alzheimers, amyloid and tau appear to interact in AD, which is more common than pure tau pathologies), I think it’s safe to say is less well understood.

  25. fajensen says:

    I think we are perhaps being too hard on Ely Lilly & Co.

    Isn’t that kind of endless fruitless pursuit, that there is someone willing to go out and be coming home tired for 10 years every day while reeking of mouse pee, driven by the hope of perhaps some winning result sometime in the future (and a 1.0 x 10^-18: 1.0 odds of a Nobel Price) the very foundation of our civilisation?

    All the easy* stuff was cracked a log time ago making research very, very hard work.

    *) Easy because someone did all the equally hard work back then and put it in a book I can read while sitting on my sofa ^__^.

  26. aairfccha says:

    Semi-OT: Did the methylene blue/LMTX idea go anywhere?

  27. Andre Brandli says:

    Regarding the clinical prospects of Biogen’s aducanumab antibody, I had the opportunity to ask Dennis J. Selkoe, a leading AD specialist at Harvard Medical School (, for his opinion at a public lecture given in Munich last December. He was extremely optimistic that the phase 3 trial would result in a positive outcome. He was basing his comments on a previous publication published in Nature in 2016 (

    1. Harrison says:

      Asking Dennis Selkoe if an amyloid-based drug will work is like asking Donald Trump if Ivanka is a good daughter.

      1. Andre Brandli says:

        Sounds like a very harsh comparison, which Dennis Silkoe may not deserve.You are implying he is not capable of making an critical assessment based on experimental facts, which may or may not kill his favorite AD hypothesis. I would give him the benefit of doubt.

        I would like to pick up here Matthew K’s comments from above: “Just because we don’t know and can’t drug the specific mechanism doesn’t mean the amyloid isn’t causal.” or in other words: “Failure to see therapeutic effects after interfering with Abeta plaque formation (via antibodies, BACE or gamma-secretase inhibitors) may indicate that these drugging strategies were not optimal for in vivo applications. For example, insufficient penetration across the blood brain barrier may pose considerable problem as mentioned by others. In closing, this calls for alternative approaches to interfere with Abeta plaque deposition but not for complete abandonment of the targets.

        1. Harrison says:

          Perhaps the analogy was a bit harsh, but even the most objective scientist will have some degree of bias. Nonetheless, apologies to Dennis Selkoe.

          While APP/PS1 mutations are causal for familial AD cases that are 5% of all AD cases, that does not mean FAD pathobiology directly translates to the 95% of cases that are “sporadic.” While not “necessary and sufficient,” the APOE4 allele has likely contributed to far more cases of AD than all of the familial mutations combined. However, the link between apoE4 and AD is not well understood, despite being first identified 25 years ago. Which brings us back to what was funded and not funded for the entirety of the 1990s, which led the amyloid treatments we are now discussing.

          1. Andre Brandli says:

            Harrison, it’s nice so see that people on this blog are able to revise their statements. This is not standard in the age of the internet rage and bad behaviour. Chapeaux!

            By the way, also mutations in presenilin 2 (PS2) cause familial AD. Regarding the link between AD and ApoE4, the Südhof lab at Stanford put forward last year an interesting hypothesis in Cell magazine suggesting that ApoE2, ApoE3, and ApoE4 differentially stimulate APP transcription and Aβ secretion ( This proposed molecular mechanism involves a novel signal transduction pathway in neurons whereby ApoE activates a non-canonical MAP kinase cascade that enhances APP transcription and amyloid-β synthesis. If this all holds up then there are ample targets in this signal transduction pathway that drugable. The work has to my knowledge not received lots of attention AD field. Any corrections or further insights are welcome.

  28. biocqr says:

    What about the leaky BBB/bacterial infection hypothesis?

    Cortexyme (private) is running a Phase I trial targeting a pathogen identified in the brain tissue and cerebral spinal fluid of patients with AD. Cortexyme’s AD study inclusion criteria…periodontitis.

    A Multiple Ascending Dose Study of COR388

    Casey Lynch: Targeting Pathogens to Treat the Brain

    Periodontitis and Alzheimer’s Disease: A Possible Comorbidity between Oral Chronic Inflammatory Condition and Neuroinflammation

  29. Dave says:

    I note this article:
    says that they have successfully reversed Alzheimer’s in mice. It is certainly early days, but they show reverse of cognitive impairment, “They show, for the first time in an animal model, that tau pathology — the second-most important lesion in the brain in patients with Alzheimer’s disease — can be reversed by a drug.”

    Of course it is MICE, not humans, so it may be meaningless….

  30. rtah100 says:

    I ran a company developing biomarkers for stratifying AD patient populations based on the cell cycle dysregulation hypothesis of AD. Brain neurons shuttle between G0 and the G1 phase cell cycle as part of neuronal plasticity. In AD brains, a significant population of neurons show markers of the S and G2 phases, i.e. they have crossed the irreversible G1/S checkpoint. As they are not mitotically competent, they are stuck in G2, the very particular biochemistry of which leads to AD pathology (cytoskeleton disassembly ==> high free tau, kinases phosphorylating this and APP, producing phospho-tau and promoting beta-cleavage of the APP). Different causes of G1/S checkpoint incompetence are correlated with different AD sub-populations – this is a syndrome, not a disease! Tau and amyloid are just downstream events in the AD pathophysiology.

  31. Len hudyma says:

    TORONTO, ONTARIO and CAMBRIDGE, MA – June 12, 2018 – ProMIS Neurosciences, Inc., a biotechnology company focused on the discovery and development of precision treatments for neurodegenerative diseases, today announced the initiation of producer cell line development for PMN310, its lead therapeutic antibody candidate for treatment of Alzheimer’s disease (AD). Selexis, SA will carry out this critical first step in the manufacturing of antibody therapeutics using the proprietary Selexis SUREtechnology Platform™.
    ProMIS’ lead antibody product candidate, PMN310, is a humanized antibody that binds with high affinity and selectivity for the toxic oligomers of amyloid beta (Aβ), a recognized root cause of AD. PMN310 is anticipated to enter phase I clinical trials in the second half of 2019.

  32. Bruce Grant says:

    The amyloid thesis is Lilly’s Vietnam

  33. MStriker says:

    I realize this is not an ideal forum, but: points out that air pollution in Mexico City is causing AD (or AD symptoms) in babies. Could some form of chelation or molecular neutralization be a solution for their AD?

  34. Jow says:

    Selective BET inhibitor RVX-208 is undergoing a nearly completed Ph III trial for patients with Diabetes Mellitus and is trying to show a 30% RRR for MACE versus palcebo, and a subset of patients over 70 will have the MOCA test done to see if the drug (also known as apabetalone) has any impact on improving cognitive function.

Comments are closed.