Alzheimer’s Drug Failure Leaves Scientists Seeking New Direction | Hacker News Reader

Alzheimer’s Drug Failure Leaves Scientists Seeking New Direction

bloomberg.com

92 pointsshadykiller76 comments

Healthcare costs for Alzheimer’s is going to be astronomical.

https://www.alzheimers.net/research-spending-vs-annual-care-...

Finding a cure or delaying symptoms by a decade would save a lot of money.

Let’s add $10 billion to the federal research budget and add a motivated czar to oversee the research. Wonder if Bill Gates has a few extra cycles.

I don’t know if any drug could actually cure Alzheimer’s.

We know we can’t completely replace exercise with a drug because mechanical stress is part of the process. So, drugs might work on Alzheimer’s patients, but without knowing the cause it’s still up in the air.

Even if there was only a one in ten chance, it justifies an enormous research budget, when the costs are $200bn a year in the US alone. What we’re actually spending is far too little.

10% chance in the next 50 years seems extremely optimistic.

We have no idea if it’s caused by a pathogen or something else, or even if it’s a single disease or several related diseases.

No one said all the money would be spent on drug research. The idea is that some basic science would also be done.

Outside of basic research there is little to be done. We don’t have a good animal model, and human testing on a disease that takes 30+ years is almost hopeless without a much better understanding of what’s going on.

I'm not exactly sure what your point is? You're trying to tell me how hard the problem is? I never said the problem was easy, never offered a solution, or claimed there's a clear path to a solution.

Alzheimer's is an underfunded disease given the amount that it costs us in healthcare. I'm saying we should spend a lot more money and put someone in charge of the money who will spend it effectively.

Saying it’s underfunded makes assumptions around how likely we are to find a solution. If hypothetically spending 1 trillion dollars gives us a 1 in 1 Billion chance of finding a cure then that’s simply not worth it.

Further spending more money does not give linear increases in value. Bumping up from around 2 bullion a year sounds great from a budget standpoint, but do we have another 8 Billion dollars worth of promising research to do next year and people able to carry that out? We are already ramping up funding rapidly, but there are only so many specialists out there.

We don't need a cure for the investment to pay off. If we can simply slow the disease, we'll save a fortune. If the disease takes an additional decade, that's a lot of healthcare costs. You might even die from something else before it gets worse.

I never claimed that if we doubled the money, we'd get double the research. I simply threw out a SMALL number like $10 billion a year. We spend $28 billion on HIV, for example.

https://www.hiv.gov/federal-response/funding/budget

I've read lots of articles over the decades about how AIDS would never be cured. Now you can live a long productive life with the disease.

HIV has a well known cause which makes research more productive. Faster than light communication would be really useful, but we have no idea who to research it or if it’s even possible. Alzheimer’s research is not quite that bad, but we have already spent 10’s of billions and still have no real clue what’s going on let alone a possible way to treat it.

PS: People talk about the huge improvement in AIDS treatments, but by default people where often living 15+ years after infection. Note infection not detection. Extending that was always considered a viable approach.

You’re wasting everyone’s time with silly comparisons.

No one is trying to violate any laws of physics by researching Alzheimer’s, the brain, etc.

You hope that’s true, but as I previously said you can’t replace exercise with a drug because of physics. So, it may in fact be a hard physical limitation that prevents drugs from working in this case.

I am not arguing it’s likely, only a very real possibility. Because in the end biology is physics, and therefore must deal with it.

Presuming Bill Gates has track record in finding cures?

B&MGF were largely responsible for the eradication of polio

*in the developing world

Polio have been eradicated in developed nations by the 1980s.

Well, to be fair his comment is still correct.

Even Bill Gates can't eradicate polio where it doesn't exist!

He needs to figure out where to effectively invest the money. He's already investing some of his money.

https://www.forbes.com/sites/denizcam/2017/11/13/microsoft-b...

Bill Gates, with the B&MGF, has a decent track record of finding and boosting low-cost solutions to chronic and systemic issues. Basically, choosing where to focus resources to the best effect. I'm not saying he'd revolutionize this field, but there are worse choices than someone with his track record.

I disagree; steering basic science needs more than money to do it, and is different from finding low cost solutions. BG also has some spectacular, costly failures like in education

Your first point may be worth taking, but I'm not sure why you say he's failed in education. The problems with education are systemic and intractable. He isn't taking an approach that attempts an overnight fix. He's taking a long-view with initiatives that attempt to slowly push the needle in the opposite direction. In the short term maybe that seems like failure because money is spent without an easily measure impact, but that is a very short-term view. I've seen first hand the benefits that some of his grants have had in this area, and that hard-won knowledge will take time to build upon and get worked into the broader world of education. It's akin to basic science in this respect.

Edit: Example. In 2010 he gave $3.5 million to charter schools in Newark, NJ. This has contributed to overall increase in success rates in such schools, which has boosted the entire city's graduation rates. They've increased significantly each year since 2013, which is about the first year you'd expect to see the results of such a grant on 4-year programs like HS.

Governments also has some spectacular, costly failures.

I'm glad we have some billionaires trying alternative, independent paths to success!

I'm no Gates Foundation fan, but their unlike every other group that attempted to tackle educational issues, they were relatively honest about good and bad and they documented it.

The problem is that they found what everybody knows but doesn't want to admit. The current public school system is about optimal for the amount of resource being spent. There is no mythical "pile of waste" that you can magically draw from.

They found that focused resource does help outcomes, but in far less proportion than the increase. If you want 5% improvement in achievement, you need closer to 25% increase in resource--and it needs to be sustained over years. More teachers that are better qualified is a good idea, but that is also the most expensive choice. And, your most effective option is to spend almost all your resource on the lowest achievers. All of these are highly politically unpalatable in the US.

"...optimal for the amount of resource being spent."

Belated reply, catching up on my reading.

This is a very interesting point.

I live in a largish school district. I've long believed, but cannot prove, that it's just too big. eg Scale matters. Everyone I've ever met who has ties to the district agrees with me.

I know the CFO. He crunched some numbers, comparing cost per student and outcomes. Although he wouldn't share the data with me (political suicide), he related that the it's a bathtub distribution.

So, I believe, but cannot prove, there's an optimal range for most organizations, and education reformers would get a lot of mileage out of correct sizing their districts.

The Alzheimer's Association's 2018 annual report[0] estimates total healthcare expenditures at $277 billion per year in 2018 alone. That's in addition to the $232 billion cost of ~18.4 billion hours of care by family and unpaid caregivers which itself excludes the immediate expenditures associated with that care, the effects of decline in mental and physical health of the caregivers themselves, changes in workplace participation resulting from providing said caregiving, etc. Or the myriad of other costs, many of which are indirect. Even miniscule increases in R&D funding for Alzheimer's could lead to a truly massive ROI for society.

0. https://www.alz.org/media/HomeOffice/Facts%20and%20Figures/f...

Brush your teeth (and gums): https://www.perio.org/consumer/alzheimers-and-periodontal-di...

The present generation of Alzheimer's researchers has a big, big problem. The best evidence now points to infection as the cause, but they are not infectious-disease specialists. It takes a long, long time to become an infectious-disease specialist, and there are lots already. What will they do when it turns out they have none of the skills needed to advance the study and treatment of Alzheimer's?

The present generation of Alzheimer's patients has a big problem, too. Grants for research into Alzheimer's are controlled by the present generation of researchers, who have no particular desire to learn that Alzheimer's is an infectious disease. So, negligible work is going into establishing an infectious-disease etiology, never mind effective treatment. Publications supporting an infectious-disease cause of Alzheimer's are, likewise, not especially welcomed by established Alzheimer's referees.

The herpes-family observations go back twenty years, and are only now gaining traction. The gingivalis observations are more recent.

This is an area where one wealthy donor could make an enormous difference that might improve the lives of millions in a very short time.

Isn't there a lot of crossover with infectious disease research due to prion disease similarities?

Science has always had the problem of the establishment fighting new theories/insights. Max Planck once said:

> 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.

Science has been able to deal with this problem for a long time.

Halting progress for several decades, and blighting the careers of its best people until the old guard dies is not "dealing with it". That is actively failing to deal with it.

I’m not sure the best evidence for AD is infectious. Do you have sources?

Mutations in the APP (amyloid precursor protein), PSEN1, and PSEN2 genes are shown to cause early onset AD with very high penetrance.

Here’s an overview of AD genetics: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3044597/

One interesting comment from this article: “...which subsequently led to the cloning of the APP type I integral membrane glycoprotein from which Aβ is proteolytically derived. The APP gene was then mapped to chromosome 21q, which accounted for the observation that patients with Down syndrome (trisomy 21) develop amyloid deposits and the neuropathological features of AD when in their 40s”

I'm surprised that a range of existing drugs with neuroprotective capabilities haven't received more attention. The cynic in me says it's because many of them are cheap generics that have little financial upside for large pharmaceuticals.

As an example, a range of anticonvulsant drugs exhibit neuroprotective effects. Depakote appears to be the best studief of the lot for this, but that's not saying much.Something like lamotrogine has similar results with lower side effect profile. These are widely used drugs for diseases like epilepsy, but they are also used for a wide range of mental health diseases where psychosis and dementia are issues because they are very effective in those areas as well. A nice bonus in treating another disease with similar neuro symptoms.

This probably wouldn't be front-line treatment, but alongside amyloid reducing medications could form a powerful combination. The biggest obstacle might be the side effects, but judicial use in the earliest stages or in high risk poulations would probably be useful. The problem is right now the data is limited to very small studies or anecdotal case studies of individual patients that were treated off label.

> The cynic in me says it's because many > of them are cheap generics that have > little financial upside for large > pharmaceuticals.

The patent is for problem + solution. You can patent many solutions for one problem and also a solution for many problems.

If they prove, say, that vitamin C stops Alzheimer's then they can patent it for that usage.

You see this a lot in cancer treatment and it makes a lot of sense. In a proper functioning system (i.e. not the USA±) it's a pretty decent solution.

± Seriously, the system there is designed so all in theory competing components are incentivized to work against consumer interests. Same thing with the housing market pushing house prices (5% interest on 500k is more $ then 5% on 200k, house sellers/older owners think they're rich but only benefit when they sell and downsize; only winners are lenders and real estate investors).

> If they prove, say, that vitamin C stops Alzheimer's then they can patent it for that usage.

Here is an article discussing this approach:

http://stm.sciencemag.org/content/6/248/248fs30

I call them "use patents" but I am not sure if that is an official term or just a colloquialism.

I just asked my wife (who's a pharmacist). She says that it's indeed true that drugs companies basically ignore existing drugs because it's not worth the costs to prove that they work even though they can officially patent them for new usages.

Here's an example: https://en.wikipedia.org/wiki/Pregabalin

That's an epilepsy treatment which also worked for nerve damage. They weren't allowed to dispense the generic for nerve problems. However in practise the prescriptions don't include the reason so in practise they provided the generic.

It would cost billions to test the various drugs that you mentioned.

No, you are incorrect. It might cost billions to develop them from scratch, but those are already sunk costs. Additional testing and trials do not cost nearly so much. In fact, while they may cost more in some circumstances, the average is only $19 million [0]

That means that even testing the half dozen or so anticonvulsants known to have neuroprotective benefits it would cost in the range of $100 million.

But for the sake of argument, let's grant your premise anyway and see where it leads. Let's say it costs billions, even $20 billion to do the trials. The annual cost of Alzheimers may top $250 billion. So the cost/benefit analysis still works out either way. In fact, in that context the potential savings make the difference between $100 million and $20 billion pretty negligible if the drugs could normalize even a few years for the average patient that would otherwise be plagued with dementia.

[0] https://www.outsourcing-pharma.com/Article/2018/09/26/Clinic...

> The cynic in me says it's because many of them are cheap generics that have little financial upside for large pharmaceuticals.

Social impact bonds are an interesting idea that might help solve this, although I know little about the practicalities. Specifically, a company would pay for the research and trials, and if it's effective, the government (say, the NHS in the UK) would pay a portion of the money saved to the company.

As someone with firsthand experience in this space, it’s hard to get buy-in even in research world for neuroprotective agents. First, this word neuroprotective gets thrown around a lot. Lots of complex efficacy endpoints. Lots of biases about safety. It’s unfortunately tough sledding.

ot: is tough-sledding an autocorrect morphism or a lnown expression?

Not auto-correct. Like saying it’s a hard slog, or a perilous path to take.

There's no reason to think valproic acid is going to solve the Alzheimer's problem. It's worked in a couple of mice models but that's it. And we could fill a warehouse with drugs that helped Alzheimer's in microwave and did diddly squat for people.

Just to be clear I think it's worth testing but just not especially more than the 500 other drugs that there is credible reason to think they'd work.

You've put forth a straw man. I haven't claimed a specific drug would solve the problem, only that a class of drugs might help.

Actual trials haven't been done, but anecdotal data from individual case studies show promise. It's absolutely not a silver bullet or potential full cure. It could however be part of a cocktail of drugs that provide significant improvement. Or maybe not. But the incremental cost if studying an already approved drug is negligible compared to creating new drugs from scratch or continuing research on other experimental drugs. But by all means, both kinds of research should occurr.

Read an interesting anecdote on amyloids in Why We Sleep. During a certain sleep phase, the brain tissues change size and open up pathways for cerebrospinal fluid to pass this. This washes out amyloids.

Over time with age and with brain damage, older adults lose the ability to generate this type of sleep, and so there is less cleaning.

Does anyone know any more about this? It seems plausible that the cause could be partly due to a loss of this sleep cleaning function. Indeed lack of sleep is correlated with alzheimers.

But that's just a book anecdote. I'd be curious to read more.

I am currently reading Why We Sleep as well, and Matthew Walker's research on the topic is really interesting to consider within the larger constellation of approaches to Alzheimer's—particularly in context of the dietary/metabolic angle.

The neural maintenance that occurs during sleep, the body's circadian mechanisms and quality of diet all seem to be highly complementary and ultimately point towards a more holistic and fundamental understanding of the disease (or brain physiology in general). It's been demonstrated pretty comprehensively that quality of diet and timing of eating have a profound effect on how well the body functions as a whole, so the sleep component seems to be an important thread in tying all of these approaches together.

One alternative theory: Alzheimer as a metabolic disease.

"We conclude that the term “type 3 diabetes” accurately reflects the fact that AD represents a form of diabetes that selectively involves the brain and has molecular and biochemical features that overlap with both type 1 diabetes mellitus and T2DM."

[1] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2769828/

Significant impact in 30 days. Without changing diet to reduce glucose. Can’t achieve 1mmol/L of Ketones in the blood, though, without a Ketogenic diet. The brain lights up in the PET scan on a Ketogenic diet:

https://youtu.be/OU26epaihmw?t=1469

Wasn't that PET scan showing ketone uptake? So it showed that brains can use ketones... but nothing about the activity or ability of the brain otherwise?

Yes. I have no idea what OP assumed by "the brain lights up."

The researcher claimed to have observed declining energy use for years before symptoms occurred, showing decreasing energy consumption leading to cell death regardless of fuel availability, finally leading to Alzheimer’s symptoms.

Provision of ketones alone reversed this energy starvation. The PET scan literally lit up like a Christmas tree — turned from all blue/black to red!

It was near the end of the presentation...

If neural cellular degeneration is the target of any Alzheimer’s management strategy, wouldn’t dramatic energy consumption declines over decades before cell death, and reversal of these declines after a short treatment be of specific interest?

Could alzheimer's be more like cancer, where there are several variants and several causes? It seems like beta amyloid were looked at because they produce similar protein manifolding to prions. Other comments on here mention herpes (which is incredibly common. I've never had cold-sores, but dated enough people who have where I may just have it and not present, and it can't easily be tested for since it produced very similar antibodies to chickenpox).

I think this may be the case for many brain-related illnesses. My son is severely autistic and through schools etc I know a lot of autistic children and adults. They are all very different but still labelled as "autistic spectrum".

I think there are many causes with many similar symptoms, which is why no-one has discovered a common cause.

This is definitely true for autism. It’s clearly somewhat heritable, but no one has found a gene or mutation that all people with autism share. Study A picks out this gene from one set of families, Study B picks out a totally different gene (and can’t replicate A’s findings).

Alzheimer’s is a little odd in that most (maybe all) patients seem to have the plaques. Removing the plaques seems to do nothing though so it might be a side effect of something else going wrong.

That's my opinion have a special needs brother. I'm down on the side that these are mostly subtle developmental disorders. Both structural and in neural 'presets'. The latter I think explains kids with specific deficits that disappear over time.

There's a novel theory that herpes is the culprit for a significant number of Alzheimer's patients.

This line of thought was originally discounted, under the (mistaken) belief that the blood/brain barrier made this impossible.

What's the evidence?

(I'm not an expert) Isn't this evidence correlated to the presence of beta-amyloid in the brain and the increased time required to clean them with herpes? As far as I read it, the evidence that beta-amyloid are a cause and not the consequence of Alz is not there. Also, I think that making such a statement for institutions needs solids theory, because it could effect a major part of population

I think more money should be directed towards tools, sensors and technology that can be used so map and study brain and biochemistry in general, and not just particular pathologies. Expansion microscopy, optogenetics and deep brain stimulation via temporally interfering electric fields are good examples of that. The surprising result that you can clear amyloids from the brain (only in transgenic mice) with 40 Hz light therapy is good example of surprising findings you get when you build new tools.

"I'm going for my biannual brain plaque checkup" will be a sentence I won't be surprised hearing. I think it's just a matter of time when we figure out brains need a bit of human-assisted maintenance just like teeth.

I'm taking every preventative measure I can at a young age to reduce chance of Alzheimer's later in life, starting with low carb intake (+ no sugar) and IF.

I can't think of anything more depressing than working your whole life so you can retire only to go into retirement and be diagnosed with Alzheimer's.

News flash: we're all going to die, and time flies. At the typical retirement age of 65, most people have 10 or 15 years left.

I'd advise a sensible diet and enjoying life while you are young. Don't spend time ruminating about what might kill you when you're near the end of your life regardless.

Life expectancy is around 20 years at 65 in most developed countries.

Fine. Point being, 20 years sounds like a long time when you're 25. When you're 55 or 60, and reflect on how fast that time went by, 15 or 20 years doesn't sound like very long. Certainly not enough time to make up for a lifetime of denying yourself life's pleasures thinking you're going to make up for it when you're old.

Edit to add: none of us are guaranteed anything. You could get cancer and be dead at 40. Or younger. Avoiding reckless risks is one thing. Thinking you really have much control over diseases you might get in 40 years from now is another.

I watched two people get it side by side... It felt like I was watching them catch a cold. It is only anecdotal but it feels like it was a virus or something catching.

What if the presence of beta amyloid is not the cause of Alzheimer’s but rather the symptom? In that case, these drugs should be ineffective.

It would be interesting to see how brain function continues to be impaired even while the beta amyloids are suppressed using these drugs, helping us get to the real process.

What if it's both?

For those willing (or those with power of attorney), would nanoparticle research experiments be worth the wont, per OP's point? While two major drug trials are being dropped towards the amyloid target (based on aducanumab) on public display, why not look towards AI and neuroplasticity measures to find the next set of targets

Disclaimer: I am not a doctor, just try to follow up on big things impacting many people.

According to my understanding of the amyloid cascade hypothesis, the amyloid beta accumulation is not a cause but indeed a symptom (of gene mutations/something). Then, the amyloid beta basically clumps up and other mechanisms happen and neurons get damaged and this collective damage is known as Alzheimer's Disease.

So the amyloid beta is just one link in the chain.

Even with a complete removal of that link (removing/reducing amyloid beta), one cannot expect the disease to be reversed, since removal of something in-between does not revert the already occured damage. But accumulation of new damage might slow down. Coupled with improved early detection even this would be very helpful to many people.

As far as I know, some modest slowing down was observed in some of the clinical trials. This could mean the amyloid cascade hypothesis isn't completely wrong.

Now, what I found really, really interesting is that first Biogen gets ca. 20 % of its market value vaporized after the aducanumab (a joint venture betwen Biogen and Eisai). This is nearly 20 billion USD. Then, Eisai informs that they still go forward with another substance called BAN2401.

The BAN2401 also targets amyloid beta, but in a different way than others (protofibrils).

So... I'm thinking here that this is either a genius move or a really dumb one. It cannot really be anything in-between. I honestly hope it'll be a genius move.

> 20 % of its market value vaporized after the aducanumab

You're missing a few words here. Was it "the aducanumab trial failed"?

Yes, you are correct, thank you!

Yes, obviously that is possible and now even likely. Given what was known at the time, it was reasonable to target amyloid. There are reams of papers studying beta amyloid and all the myriad downstream effects on neurophysiology, but it could turn out to just be an intermediary step.

It's the maddening thing about biology. It's a brutal intellectual undertaking. You could have spent your life studying amyloid. Imagine sitting in your office with your stacks of papers now, considering what you have done with your life.

You contributed one more thing to the set of knowledge we have about the world.

Science is about discovery, not outcome.

https://outline.com/rfdTx9

This book https://www.amazon.com/dp/0735216207

I read it, it is interesting but it tells only a way to mitigate the disease, not to cure it.

And I would love if there were independent studies that could confirm Bredesen's writings (I do not say there are none, just that I am not aware of them).

Ultrasound to break up plaques in the brain: https://www.fusfoundation.org/diseases-and-conditions/neurol...

Reply on news.ycombinator.com