The other day, I was visiting my grandpa and picked up his recent copy of the New Scientist.
There was a short article about anosmia – loss of smell – as an early indicator of Alzheimer’s disease. Nothing unusual there, until I noticed the image they used: what appeared to be a generic stock image of a brain scan showing amyloid beta plaques, with no reference or explanation in the article. That struck me as strange.
Why? Because in 2022, a major scandal broke that called a big part of the so-called “amyloid hypothesis” into question.
The Amyloid Hypothesis
First, a bit about the amyloid hypothesis. It was first championed in 1992 and posits that the accumulation of amyloid-beta protein in the brain is the primary driver of Alzheimer's disease. The plaques were thought to trigger the cascade of neurodegeneration and cognitive decline characterizing the disease.
However, no one had been able to show that these plaques actually cause Alzheimer’s disease. The hypothesis was facing increased scepticism in the early 2000’s after multiple experimental drugs targeting plaques failed to slow cognitive decline in people with Alzheimer’s.
The Perpetuation
Enter Sylvain Lesné and his landmark paper published in Nature in 2006 which claimed to show the first tangible evidence of a cause-effect relationship between amyloid plaques and memory impairment. In the paper, researchers claimed to isolate a specific form of amyloid protein, Aβ*56, that appeared to directly impair memory in mice after injection. This was the first strong evidence of the amyloid hypothesis – the paper blew up, and for years, this led researchers and drug developers down a very specific path: target Aβ*56 proteins, clear them, cure Alzheimer’s.
The paper, which became one of the most-cited Alzheimer’s studies has shaped research funding and drug development (and continues to do so) for 18+ years. In the time since the paper was published in 2006, failures to replicate the results of the study were ignored, and Lesné was continuously awarded coveted multi-million dollar National Institute of Health (NIH) grants. According to Science, annual NIH support for studies with keywords “amyloid, oligomer, and Alzheimer’s” skyrocketed to nearly $300 million in 2021.
Uncovering Fraudulent Foundations
But in 2022, investigations (largely sparked by neuroscientist Matthew Schrag) revealed falsified images in Lesné’s cornerstone paper. Subsequent scrutiny showed dozens of other papers with suspicious images. The doctored so-called “Western blot” images included manipulated protein bands indicating increasing levels of Aβ*56 in aging mice. Schrag’s forensic analysis on the images showed a 0.98 correlation between supposedly distinct protein bands, strongly indicating the images had been tampered with to falsify results.
Several months after whistleblower Matthew Schrag initially raised his concerns about Lesné’s work with NIH, Lesné was awarded a prestigious $765k NIH grant – the program officer of which was a co-author on Lesné’s initial amyloid plaque paper.
It was not until June 2024 that the paper was finally retracted. Finally, in 2025, Lesné quietly resigned from the University of Minnesota – nearly 20 years after the initial publication!
The Aftermath
This timeline is scary – it reflects years of systemic inertia, conflicts of interest, and a lack of adequate safeguards. It leaves me thinking…. How many negative results were buried? How much public funding wasted? How many lives altered due to delayed progress, diverted resources, and tunnel vision? How many people given false hope and needlessly subjected to harmful or even lethal side effects of drugs all based on a sham?
Charles Piller, who wrote the original exposé on the investigation into Lesné et al.’s fraud for Science sought to answer these questions. He continued to investigate the field and found more and more evidence of doctored data and images piling up. One of his findings: an astonishing 132 scientific papers published between 1997 and 2023 by the (former) director of the Division of Neuroscience at the National Institute on Aging were riddled with apparently doctored data – many of which were manipulated Western blot diagrams like in the Lesné paper. Piller has since published a compilation of his and others’ findings in his book, Doctored: Fraud, Arrogance, and Tragedy in the Quest to Cure Alzheimer’s.
How did we even get here?
Let’s rewind and revisit what the field once saw as an unshakable foundation. The amyloid beta hypothesis of Alzheimer’s did not just appear out of thin air. Support for it came from various observations. For instance, postmortem brains of Alzheimer’s patients often showed widespread plaques. People with Down syndrome, who have an extra copy of the amyloid precursor protein (APP) gene, often develop Alzheimer’s-like pathology by middle age. And people with early onset familial Alzheimer’s have mutations which can involve APP genes which affect amyloid processing.
However, despite the amyloid hypothesis’ potential, there was stagnation, and many failed drug trials prior to that 2006 landmark paper. Unfortunately, journals reward novelty and positive results. Influential figures secured large amounts of public funding, reinforcing their ties with grant reviewers. Conflicts of interest seemed to have been overlooked. And several years down the line, and after significant investment in amyloid-based theories, pharmaceutical companies faced increasing sunk costs. Even as drugs failed clinical trials one after the other, the field doubled down. While the field obsessed over amyloid, other hypotheses were severely underexplored.
Like a Phoenix from the flames
After years of chasing a flawed hypothesis, after billions were spent and trust eroded, fortunately the field has not stood still. One such example is work by Dr. Dennis Selkoe (who had previously tried, but never was able to replicate the now-retracted Aβ*56 findings) which showed back in the 1990s that amyloid beta is produced by everyone, throughout their entire lives. So why doesn’t everyone get Alzheimer’s? Instead of focussing on why some people have amyloid, the real question for Dr. Selkoe became why some can’t clear it effectively. This thinking has propelled new targeted trials, such as the A4 study, which aims to help people with familial Alzheimer’s before symptoms appear. While results from the A4 trial showed that the drug tested was unsuccessful at reducing cognitive decline in patients with early onset Alzheimer’s, it highlighted the importance of the correlation between different proteins (amyloid and other types) and the risk of cognitive decline.
The field is finally moving beyond a sole amyloid focus. For instance, recent research indicates that tau tangles may have a stronger association with cognitive decline than amyloid plaques. Tau is a type of protein that helps stabilize the internal skeleton (microtubules) of neurons. In healthy brains, tau supports the transport of nutrients and signals along neurons. But in Alzheimer’s disease, tau becomes altered to a form known as p-tau217, with too many phosphate groups, causing it to detach from microtubules, forming insoluble “tau tangles” inside brain cells. These tau tangles disrupt cell function and eventually lead to cell death. One study developed a biomarker that can detect early tau species which can help us intervene before irreversible tangle formation in a patient. But interestingly, researchers just discovered that the “toxic” p-tau217 protein is actually more abundant in perfectly healthy baby brains than in Alzheimer’s patients! Research that focusses on how babies are able to deal with this protein may lead to important treatment therapies in Alzheimer’s patients. There seems to be a pattern here – much like in the amyloid case, the tau research highlights the importance of keeping an open mind: is the presence of the protein the real problem, or is it something deeper?
Another avenue of Alzheimer’s research involves the exploration of dark microglia. Generally speaking, microglia, the brain’s immune cells, can be both protective and harmful. A recent study identified a neurodegenerative microglia subtype which releases toxic lipids that damage brain cells. Blocking the release pathway reversed Alzheimer’s symptoms in mice.
What’s next for Alzheimer’s research?
The good news is, clinical trials are now exploring multiple target approaches, finally moving away from the amyloid tunnel vision that has plagued the field for so long. Scientists are also investigating how we could use advanced gene editing techniques, such as CRISPR-Cas9 to convert high risk genes associated with Alzheimer’s disease into benign variants.
Recently I came across a good analogy of the amyloid theory, and that was likening it to having a house infested with termites, and focussing solely on trying to sweep up the wood dust – a symptom and not a driver of the problem. At a certain point, you have to deal with the root problem.
Conclusion
So, when I see popular science articles today throwing around images of plaques like it's settled science… I can't help but wonder: Why are we still so attached to this hypothesis? Why can’t we see beyond the wood dust?
Although it took many years, it is promising that the field is now making unprecedented progress. Now that the decades-long debacle is mostly behind us, we are seeing more investment of public funds into holistic and integrated Alzheimer’s research. It is encouraging, for instance, that in the 2025 global Alzheimer’s clinical trial pipeline, more than 60% of trials were not just amyloid focussed. Many scientists around the world have been calling for broader recognition of the complexity of the disease, and diversifying funding and resources will undoubtedly help to make way for real breakthroughs.