A study published in Alzheimer’s and Dementia: The Journal of the Alzheimer’s Association by researchers at Arizona State University has proposed something that the Alzheimer’s field has spent more than a century failing to produce.
A single, unifying explanation for how the disease starts and spreads.
A framework that could account for every known manifestation of Alzheimer’s, from the earliest invisible cellular changes to the devastating cognitive collapse that marks the disease’s final stages, through a single upstream mechanism.
The condition causes widespread disruption of gene behaviour, affecting every known neuropathology and clinical manifestation of the disease.
According to the analysis, the changes caused by the illness may stem from a breakdown in the transport system that shuttles vital molecules between the cell nucleus and cytoplasm, the liquid environment surrounding the nucleus where many essential processes occur.
The culprit the researchers are pointing to is something called stress granules, tiny clumps of protein and RNA that form inside brain cells when those cells are under stress.
And the implications, if the theory holds up, could fundamentally reshape how Alzheimer’s is diagnosed, treated, and most importantly, prevented.
The Scale of the Problem
Before getting into the science, it is worth pausing on what Alzheimer’s disease actually costs.
An estimated 7.4 million Americans aged 65 and older live with clinical Alzheimer’s dementia today, a number that could increase to 13.8 million by 2060 absent medical breakthroughs that prevent or cure the disease.
Official death certificates recorded 116,022 deaths from Alzheimer’s disease in 2024, with the disease officially listed as the sixth-leading cause of death in the United States and the fifth-leading cause among Americans aged 65 and older.
Between 2000 and 2024, deaths from stroke, heart disease, and HIV decreased, whereas reported deaths from Alzheimer’s increased by 134 percent.
More than 12 million family members and other unpaid caregivers provided an estimated 19.6 billion hours of care to people living with Alzheimer’s or other dementias.
Despite more than a century of research and billions of dollars invested, not a single treatment has been developed that can halt or reverse the disease.
The newest drugs, including the monoclonal antibodies lecanemab and donanemab, can slow cognitive decline modestly in some patients.
But they cannot stop it, reverse it, or prevent it.
As of January 2025, 138 Alzheimer’s drugs were being assessed in 182 clinical trials registered with clinicaltrials.gov, with 73 percent seeking to change the underlying biology of the disease.
The sheer volume of concurrent research reflects the field’s fundamental problem: nobody agrees on what Alzheimer’s actually is, at the level of its root cause.
Why Alzheimer’s Has Been So Hard to Crack
The Alzheimer’s field has long been fragmented. Over the decades, there have been a series of hypotheses of biological mechanisms including cholinergic, inflammation, viral, mitochondrial, protein processing, vascular, tau, and amyloid hypotheses, among others. None of these hypotheses have led to halting or reversing disease.
Each of those hypotheses addressed a real observation.
Amyloid plaques really do build up in the brains of Alzheimer’s patients.
Tau tangles really do form.
Neuroinflammation really does occur.
Neurons really do lose their ability to communicate.
The problem was not that any of these observations were wrong.
The problem was that each explained a piece of the disease while leaving the rest unexplained.
Why do all of these things happen together, and at such scale?
A 2022 survey of the literature showed that 91 percent of the known KEGG pathways are involved in Alzheimer’s disease, and that many of these pathways are represented by the known cellular and molecular phenomena of the condition.
The KEGG database catalogues the known biological pathways in cells, the sequences of molecular events that govern how cells function, communicate, and survive.
When 91 percent of those pathways are disrupted in Alzheimer’s, the scale of the chaos becomes clear.
This is not a disease that breaks one or two systems.
It rewires the entire cellular operating system.
That observation raises an urgent question that the Arizona State team decided to pursue directly.
What single mechanism could possibly cause disruption on that scale?
The Answer: Stress Granules and a Broken Communication System
The researchers’ answer centres on two concepts: stress granules and nucleocytoplasmic transport.
Stress granules are temporary structures that form inside cells when those cells are under threat.
Stress granules are transient structures that assemble in response to cellular stress, temporarily halting nonessential processes to aid cell recovery. Under normal conditions, they protect the cell and dissipate once the stress is resolved.
Think of them as an emergency shutdown protocol.
When a cell is under threat, whether from a toxin, a genetic mutation, an environmental stressor, or simply the accumulated damage of aging, it assembles these granules to temporarily pause non-essential activities while it manages the crisis.
It is a built-in protective response.
Under normal circumstances, once the crisis passes, the granules dissolve and the cell returns to normal operations.
In Alzheimer’s disease, the theory proposes, the granules do not dissolve.
In Alzheimer’s disease, stress granules persist abnormally, becoming chronic and pathological. They trap essential molecules and disrupt their transport into and out of the cell nucleus, shifting from a protective role to a harmful one that contributes to disease progression.
This is where nucleocytoplasmic transport enters the picture.
Nucleocytoplasmic transport is the process by which molecules move back and forth between a cell’s nucleus, which contains the DNA, and its cytoplasm, the surrounding fluid where proteins are built and most cellular work is done.
This exchange is constant and essential.
Genes in the nucleus send instructions out to the cytoplasm.
Proteins built in the cytoplasm are shuttled back into the nucleus to regulate which genes are active.
The system is the cell’s central communication network.
The nuclear pore complex is a large structure in the membrane surrounding the nucleus that acts as a gateway for this transport. When it malfunctions, the consequences ripple across every system that depends on accurate gene expression.
The Arizona State team proposes that chronic stress granules block this gateway.
They sequester the proteins that keep the nuclear pore complex functioning, effectively jamming the traffic that the cell’s nucleus depends on to do its job.
Upstream, disruption of nucleocytoplasmic transport and gene expression results from Alzheimer’s-induced stress granule sequestration of molecules related to both the nuclear pore complex itself, as well as its chaperones and related molecules. Downstream, this disruption then leads to massive changes in the transcriptome, including the epigenome.
When that communication breaks down, the cascade of consequences is enormous.
Genes that should be active go silent.
Genes that should be silent become active.
The proteins that keep synapses working, that clear cellular debris, that maintain the structural integrity of neurons, all of them are produced in the wrong quantities or at the wrong times.
The result is a progressive, compounding cellular catastrophe that eventually manifests as the cognitive collapse of Alzheimer’s disease.
But Here Is What Makes This Theory Genuinely Different
The obvious objection to any “single cause” theory of Alzheimer’s is that the field has tried this before.
The amyloid hypothesis, which held that the accumulation of amyloid-beta plaques was the primary driver of the disease, dominated Alzheimer’s research for decades and absorbed the lion’s share of research funding.
Drug after drug targeting amyloid failed in clinical trials.
The tau hypothesis, which focused on the protein tangles found inside neurons, has similarly produced treatments that work in laboratory animals but struggle to translate into human benefit.
What the stress granule model offers that those hypotheses did not is an explanation that sits upstream of all of them.
We know Alzheimer’s makes wholesale changes early on, effectively rewiring biological pathways to increase cell stress, block neuron communications, and cause protein abnormalities, such as amyloid-beta clumps.
The critical word there is early.
The amyloid plaques and tau tangles that have been the targets of most Alzheimer’s drug development are downstream consequences of the disease process, not its origin.
The stress granule model suggests that by the time those plaques and tangles appear, the cellular damage has already been building for years, perhaps decades.
A variety of factors, from air pollution to genetic mutations, could be triggering these stress granules to linger longer, and future studies will be able to look in more detail at how they are formed and how they cause damage.
This framing changes what an effective intervention would look like.
Instead of trying to clear the plaques that have already formed, or untangle the tau that has already aggregated, the goal would be to prevent the stress granules from becoming chronic in the first place, interrupting the disease before any of the visible signs of Alzheimer’s ever appear.
How the Study Was Conducted
The research team, led by neuroscientist Paul Coleman of the ASU-Banner Neurodegenerative Disease Research Center, conducted a comprehensive review of existing data rather than running new laboratory experiments.
They analysed data from multiple health databases and prior studies, with particular focus on a 2022 paper that catalogued changes in gene expression across the full breadth of Alzheimer’s biology, mapping those changes to the KEGG pathway database.
The review highlights the fact that gene expression, the process by which genes produce the proteins essential for cell function, is altered on an enormous scale in Alzheimer’s. These changes affect the proper functioning of synapses in the brain, as well as metabolism, protein processing and cell survival.
The team then asked whether the known behaviour of stress granules and nucleocytoplasmic transport disruption could account for the breadth of that gene expression change.
Their answer, laid out across the study, was yes, with a detailed mapping of how each stage of the proposed mechanism connects to the downstream biology of Alzheimer’s.
The model provides a rational, evidence-based framework, while also acknowledging the points where additional evidence is still needed.
Crucially, the authors are not claiming to have proven the model.
They are claiming to have built a coherent, testable framework, and to have assembled the existing evidence in a way that supports it more comprehensively than any previous model has managed.
The Findings in Plain Terms
The study’s core claim can be stated simply.
Every known risk factor for Alzheimer’s, whether genetic, environmental, or lifestyle-related, converges on a single cellular response: cellular stress.
That stress triggers the formation of stress granules.
In most people, those granules resolve and cause no lasting harm.
In people who develop Alzheimer’s, they become chronic, jamming the cellular communication system and triggering a cascade of gene expression changes that eventually produce every known feature of the disease.
The prospect of early interventions targeting stress granules offers a potentially transformative approach to combating Alzheimer’s disease.
By identifying and addressing the formation of pathological stress granules at the earliest stages, it might be possible to halt or significantly delay the onset of symptoms such as amyloid plaques and tau tangles, and the devastating cognitive consequences of the disease.
The model also suggests that the same mechanism may underlie other neurodegenerative diseases.
Stress granule proteins are found in the pathological aggregates of an expanding array of age-related neurodegenerative conditions, including amyotrophic lateral sclerosis and frontotemporal dementia, but also spinocerebellar ataxias, Huntington’s disease, and Alzheimer’s disease.
Moreover, nucleocytoplasmic transport defects have also been implicated in the pathogenesis of all these conditions.
If that is correct, the implications extend far beyond Alzheimer’s research.
A drug or intervention capable of preventing chronic stress granule formation could potentially reduce the risk of multiple neurodegenerative diseases simultaneously.
The Limitations That Cannot Be Ignored
The study’s authors are careful and explicit about what their model does not yet prove.
Although the formation of stress granules has been shown for some Alzheimer’s risk factors, all risk factors have not been studied. The profile of molecules sequestered in stress granules under defined circumstances remains to be determined. Although a relationship between sequestration in stress granules and the functioning of the nucleocytoplasmic transport system has been described, the specifics of Alzheimer’s are yet to be determined.
There is also an important complication from the broader scientific literature.
A 2022 study published in PLOS Biology found that nucleocytoplasmic transport defects can occur independently of stress granule assembly, suggesting the relationship between the two may be more complex than the Arizona State model currently captures.
That finding does not disprove the model, but it suggests that stress granules may be one pathway to nucleocytoplasmic transport failure rather than the only one.
The model also faces the challenge that all biological models of Alzheimer’s face: what works in cells and in animal models has repeatedly failed to translate into effective treatments for humans.
Lecanemab and donanemab, the most recently approved Alzheimer’s drugs, both successfully reduced amyloid burden in patients’ brains.
Amyloid reduction was what decades of hypothesis and billions of dollars said should help.
It helped modestly, at best.
The lesson the field has taken from those disappointments is that the biology of Alzheimer’s in living human brains is more complex, more entangled, and more resistant to single-target intervention than laboratory models suggest.
A recent review published in Science China Life Sciences argues that focusing on a single cause has not worked because Alzheimer’s is far more complex, arising from the combined effects of amyloid-beta buildup, tau protein tangles, genetic risk factors, aging-related changes, and broader health conditions. Because of this complexity, future treatments must take a more comprehensive and coordinated approach.
The stress granule model is not immune to that critique.
It proposes a single upstream mechanism, but the mechanism it proposes produces an extraordinarily complex downstream cascade.
Whether targeting that upstream event proves easier in practice than targeting downstream consequences is a question that only clinical trials will answer.
What This Could Mean for the Future
Despite those caveats, the potential implications of a validated stress granule model are significant.
The most immediate practical application would be in early detection.
Coleman says the key questions are when Alzheimer’s can first be detected and when intervention should begin, both of which have profound implications for society and future medical approaches.
If stress granule dysregulation precedes the appearance of amyloid plaques and tau tangles, it could provide a detectable biomarker that appears years or even decades before current diagnostic tools can identify the disease.
Current Alzheimer’s diagnosis relies heavily on cognitive testing and brain imaging that can only detect the disease after significant neurological damage has already occurred.
A biomarker that flags the earliest cellular disruption would give clinicians a window to intervene while the brain still has the capacity to recover or compensate.
The second implication is therapeutic.
Because stress granule dysregulation occurs before noticeable Alzheimer’s symptoms, it offers a potential window for early intervention. Scientists may be able to develop ways to block the disease in its earliest stages by targeting this initial cause.
Several compounds that modulate stress granule formation are already known.
None have been tested specifically in the context of Alzheimer’s prevention, but the new model gives researchers a clear rationale for doing so.
The third and perhaps most profound implication concerns who should be treated and when.
If Alzheimer’s begins at the cellular level long before any symptoms appear, then the population that could benefit from intervention is not just those already showing cognitive decline.
It is anyone carrying the risk factors, whether genetic variants like APOE4, environmental exposures, or lifestyle factors, that the model suggests trigger the initial stress granule cascade.
That is a very large and very different group of potential patients.
The Question This Research Leaves Open
More than a century after Alois Alzheimer first described the disease that bears his name, the fundamental question of what causes it remains technically unresolved.
The Arizona State model does not close that question.
It proposes a compelling, evidence-supported answer and maps out a research agenda to test it.
“Our proposal, focusing on the breakdown of communication between the nucleus and cytoplasm leading to massive disruptions in gene expression, offers a plausible framework to comprehensively understand the mechanisms driving this complex disease,” says neuroscientist Paul Coleman. “Studying these early manifestations of Alzheimer’s could pave the way for innovative approaches to diagnosis, treatment, and prevention, addressing the disease at its roots.”
What makes this model worth paying attention to is not just its ambition.
It is the fact that it does not ignore the existing evidence.
It incorporates the amyloid hypothesis, the tau hypothesis, the inflammation hypothesis, and every other partial explanation that has accumulated over a century of fragmented research, and proposes a mechanism that explains why all of them are partially right while none of them are sufficient on their own.
Whether that mechanism proves to be the answer, a piece of the answer, or ultimately another step in the long process of understanding a devastatingly complex disease, will only become clear with time, experiments, and eventually, clinical trials.
But for the first time in a long time, researchers have a framework that could account for everything Alzheimer’s does.
The study, titled Massive changes in gene expression and their cause(s) can be a unifying principle in the pathobiology of Alzheimer’s disease, was published in Alzheimer’s and Dementia: The Journal of the Alzheimer’s Association in February 2025 by Paul D. Coleman, Elaine Delvaux, Ashley Boehringer, Carol Huseby, and Jeffrey Kordower of the ASU-Banner Neurodegenerative Disease Research Center at Arizona State University.
