A recent review of anti-amyloid drugs to treat Alzheimer’s Disease was critical of their meagre performance. It drew indignation from some, but shouldn’t have. The link between reducing amyloid plaques (see pic above) and improving dementia has been questioned for some time. I dig into the original hypothesis, the landmark proof and the recent review, and find more than I bargained for.
Main Points
- The recent review was by the independent Cochrane Collaboration;
- The results were welcomed by some and dismissed by others;
- The review’s criticism seems more than justified in light of other evidence.
By the way, if you’ve read the post ‘Good News For Boomers From 17 Dementia Drug Trials’, you can skip the first part about the review and go straight to the history of amyloid: ‘Not Exactly New News’.
About The Review
It was run by the Cochrane Collaboration, a global, independent, non-profit network of health researchers and professionals whose main concern is verifying results of clinical trials.
Scientific American describes its work as ‘gold-standard, independent analysis of the evidence for and against specific health interventions or treatments’. In other words, Cochrane’s reviews are generally well accepted.
The review was a ‘systematic analysis’ of 17 clinical trials of 15 amyloid-reducing drugs (one drug, aducanemab, had three trials that met the review’s criteria).
The 15 drugs included 12 that had not been approved for use and three that had, including the two current drugs, lecanemab (Leqembi) and donanemab (Kisunla). The third approved drug was aducanemab (the drug with three trials), which was later withdrawn due to adverse effects. The Cochrane review analysed data from 20,342 trial participants and, to be clear, it was a ‘study of studies’ not a new clinical trial.
To be included in this review:
- The trials had to be of the highest standard (Randomised Controlled Trials or RCTs) and conducted over at least 18 months (results were compared at the 18-month point);
- The drugs had to be monoclonal antibodies (the ‘mab’ part of the drug name) that focused on reducing beta-amyloid;
- The mode of action (how they reduced amyloid) didn’t have to be the same.
In other words, the authors wanted to know if reducing amyloid affected AD, regardless of the mechanism or whether the drugs had been approved or not.
The work of the Cochrane Collaboration is the ‘gold-standard, independent analysis of the evidence for and against specific health interventions or treatments’
What The Review Said
The conclusions were pretty clear.
The review’s lead author Francesco Nonino of the Institute of Neurological Sciences of Bologna said: ‘Unfortunately, the evidence suggests that these drugs make no meaningful difference to patients…’
Edo Richard, a co-author on the Cochrane review and professor of neurology at Radboud University Medical Centre in the Netherlands, told The Guardian that the analysis found ‘no clinically meaningful effect on cognitive decline or dementia severity, and these effect sizes are too small for patients and caregivers to notice.’
In other words, reducing amyloid didn’t make a difference in patients that was big enough to notice.
Regarding the currently approved drugs, lecanemab and donanemab, Richard said ‘they caused more swelling and bleeding in the brain than the placebo.’ Robert Howard, professor of old age psychiatry at UCL, said: ‘The sad truth is that even the best-performing drugs don’t do anything that’s clinically meaningful.‘
In plain English, the two drugs that gained approval didn’t perform much better than those that didn’t.
Nonino and his co-authors argue that ‘future clinical trials of drugs designed to remove amyloid proteins are unlikely to have benefits for patients, and they instead recommend other avenues of research.’
In other language, reducing amyloid did not make a clinical difference to AD and did cause adverse effects so, maybe, future research look elsewhere for solutions.
Unfortunately, the evidence suggests that these drugs make no meaningful difference to patients…’
Francesco Nonino, Institute of Neurological Sciences of Bologna, Co-author, Cochrane Collaboration Review
Not Exactly New News
Ben Adams, business, science and healthcare journalist from Fierce Biotech, reached a similar conclusion 5 years ago.
The difference is, in 2026, it’s the first time that the published clinical trials of 15 similar drugs have been analysed independently, side by side.
After the 2021 approval of aducanemab (the multiple-trial, now-withdrawn drug), Adams wrote: ‘The FDA’s decision on aducanumab will drag down innovation for years to come.’
In other words, by approving aducanemab, the FDA validated the ‘amyloid causes Alzheimer’s Disease’ belief, thereby ensuring more research would be channeled into this approach. That’s the opposite of what Cochrane concluded 5 years later.
Yet, the demise of aducanemab shouldn’t have been a surprise.
The FDA approved the drug, despite two failed clinical trials and unanimous opposition from the expert group it appointed to assess the third. Aducanemab was the first disease-modifying dementia drug in two decades to pass Phase III trials and, by 2021, dementia had become a pandemic. Did the FDA feel that it had to approve something? We may never know.
Adam’s article starts like this: ‘The amyloid theory just won’t die. In fact, by approving Biogen’s new Alzheimer’s disease drug (aducanemab), the FDA has just resurrected it for a long and healthy future. This … will decimate Alzheimer’s research, faith in the FDA and the healthcare budget for decades to come.’
And that is what happened.
The amyloid theory just won’t die. In fact, by approving Biogen’s new Alzheimer’s disease drug (aducanemab), the FDA has just resurrected it for a long and healthy future. This … will decimate Alzheimer’s research, faith in the FDA and the healthcare budget for decades to come.
Not Much Effect on AD
By 2021, many researchers knew that breaking down amyloid in Alzheimer’s brains wasn’t improving the disease.
That’s why dozens of drug candidates using his mechanism failed over two decades: they couldn’t show that the drugs improved cognition or arrested the disease’s progression.
It’s also why Cochrane included failed drugs as well as approved and withdrawn drugs: to see if reducing amyloid had a clinical impact regardless of the drugs’ status. That is, does reducing amyloid improve dementia?
This inclusion of approved and failed drugs was what drew the most criticism of Cochrane’s review. Dr Susan Kohlhaas at Alzheimer’s Research UK told The Guardian: ‘This study is attempting to paint an entire class of drugs with the same brush even though we know different anti-amyloid treatments can act in different ways.’
She argued that lecanemab and donanemab were the only two approved drugs covered in the review, while the other 12 had never gained approval and the modes of action were different. (Aducanemab made the number up to 15.)
That wasn’t the point for Cochrane’s authors. They didn’t care how the amyloid was reduced; they just wanted to know:
- Does reducing amyloid make a clinical impact on AD?
- Does reducing amyloid cause any harm?.
You can check dozens of dismissive quotes from dementia experts at the UK Dementia Research Institute website.
This study is attempting to paint an entire class of drugs with the same brush even though we know different anti-amyloid treatments can act in different ways.
The Shaky Case For Amyloid
The original idea, ‘Amyloid Cascade Hypothesis’ in 1992, proposed that plaques of amyloid in people’s brains caused dementia, but it hadn’t been proven in vivo (in live animals). At that point, it was just a theory.
In 2006, Sylvain Lesné of the Uni Minnesota, showed that a beta-amyloid protein caused the symptoms linked to Alzheimer’s, in an experiment using mice as the model. This became a landmark paper in dementia research and has since been cited thousands of times. (I can’t link to it because, as you’ll see below, it’s been retracted.) This paper also sparked generous funding from drug makers to explore this vein; hence the two decades of amyloid-reducing drugs.
Then, in 2022, investigative journalist, Charles Piller, found evidence that the Lesné experiment contained falsified data. Piller later published a book about it called ‘Doctored‘. Here what Piller said to the New York Times:
‘After most of its authors conceded that technical images were doctored, the paper was finally retracted. Days later, a City University of New York scientist behind a well-financed, controversial Alzheimer’s drug was indicted on charges alleging research fraud.’
In other words, the basis of 20 years of research may have been very shaky indeed.
Neurobiologist, Matthew Schrag from Vanderbuilt Uni, reached a similar conclusion, so it wasn’t just journalistic enthusiasm.
Schrag was less direct in his summation, calling it ‘potential fabrication’. He gently concluded: ‘Even if misconduct is rare, false ideas inserted into key nodes in our body of scientific knowledge can warp our understanding.’ And maybe twenty years of dementia research?
A six-month investigation by Science into the question of fraud showed that Piller and Schrag were right.
It concluded that the authors of the 2006 paper ‘appeared to have composed figures by piecing together parts of photos from different experiments. The obtained experimental results might not have been the desired results, and that data might have been changed to … better fit a hypothesis.’ In question were 70 apparently doctored images, not just the one above.
The long-term impact of this falsified landmark study is clear.
‘The immediate, obvious damage is wasted NIH funding and wasted thinking in the field because people are using these results as a starting point for their own experiments,‘ says Stanford University neuroscientist Thomas Südhof, a Nobel laureate and expert on Alzheimer’s and related conditions.
As we saw, that’s what happened.
Is there a glaring reason why proving that ‘reducing amyloid improves dementia‘ is so elusive? That it’s not actually been established? Maybe the tiny improvements of the approved dementia drugs are just statistical noise, and not truly significant? One day, we may know.
The immediate, obvious damage is wasted NIH funding and wasted thinking in the field because people are using these results as a starting point for their own experiments.
Not Robinson Crusoe
It seems that Lesné’s scientific fraud was not alone in dementia research.
Piller writes that it ‘suffered a litany of ostensible fraud and other misconduct by world-famous researchers and obscure scientists alike, all trying to ascend in a brutally competitive field.’
He cites the example of neuroscientist Eliezer Masliah who, in 2016, was appointed to lead the National Institute on Ageing’s project to address Alzheimer’s.
‘With roughly 800 papers to his name, many of them considered highly influential, Dr. Masliah seemed a natural choice to steer the project, with millions in new funding. He hailed the moment as the dawning of “the golden era of Alzheimer’s disease research” ’, Piller writes.
in 2024, Piller wrote for Science Magazine, where he showed that ‘for decades Dr. Masliah’s research had included improperly manipulated photos of brain tissue and other technical images’, a clear sign of fraud. Many of Masliah’s studies contained apparently falsified images to show the presence of proteins in blood or tissue samples.
Eventually the National Institutes of Health conceded that Dr. Masliah had engaged in research misconduct and announced that he had been removed from his position at the NIA. Piller stresses that Masliah’s misconduct was merely the tip of a vast iceberg of Alzheimer’s researchers, ‘including some once considered luminaries who have recently faced credible allegations of fraud or misconduct.’
I worry about it giving science a further black eye, just as the public’s confidence in science and scientists is sinking to new depths.
‘The Amyloid Mafia’
The tenacious journalist, Piller, also discovered what he called the ‘amyloid mafia’, led by Dennis Selkoe, a Harvard professor who is one of the most celebrated Alzheimer’s researchers.
He describes Selkoe’s ‘role in strong-arming the field to favor a focus on amyloids above all else — “the scientific equivalent of the Ptolemaic model of the Solar System,” in which the Sun and planets revolve around Earth … ‘
Piller reports that proponents of the amyloid hypothesis claim that there is ‘no amyloid mafia trying to control things behind the scenes, and there is no group of exalted scientists who are trying to suppress other ideas, and yet this is exactly what they were doing.’
How many possible causes of dementia did this dogged focus on amyloid bury for two decades? How many people with dementia waited in vane for dementia treatments that might actually work? It‘s a question Charles Piller answers in his post All the Alzheimer’s Research We Didn’t Do.
In a nutshell, the 2026 Cochrane review seems to have confirmed what many researchers knew, but preferred to ignore: reducing amyloid plaques in the brains of AD sufferers is unlikely to affect the disease.
You might imagine that Cochrane’s new review might shine the spotlight on the questionable science behind amyloid and bury the hypothesis once and for all. Don’t hold your breath. I won’t.
Instead, legions of authors from revered institutions will write learned rebuttals that will outnumber supporters of Cochrane by orders of magnitude; Cochrane’s independence and work will be called into question; the amyloid myth will continue and the amyloid gravy train will go on its merry way. Sadly, that seems to be the pattern these days.
That said, there are still independent, trustworthy researchers in medicine and I rely on them. (Read more about research integrity here.)
***
If you’d like to move away from dementia drugs based on dubious science and towards positive steps to avoid dementia, you may like to read a little further.
I wrote a couple of eBooks to try to make sense of all of this, translating the latest dementia prevention research into plain English for non-scientists.
If this sounds interesting, you may like to see what’s inside:
- Dementia: Keep Your Marbles shows you the many causes of dementia and how to avoid them;
- Anti-Dementia Food (the companion eBook) shows you the difference between foods that promote and those protect against dementia – and how to make the protective ones into delicious meals.
If you decide not to read further, that’s fine. It’s up to you. No pressure and no deadlines.
The good news is, many credible, independent sources say there is a pile of stuff we can all do to prevent, delay or avoid dementia.
ANTI-DEMENTIA FOOD
How you can feed your brain with real food
