Alzheimer’s disease is usually discussed as a problem of neurons, plaques and memory. A new study suggests that part of the story may begin in a less obvious place: the tiny blood vessels that keep the brain supplied and protected.
Researchers at Mount Sinai have traced one way that APOE4—the strongest common genetic risk factor for late-onset Alzheimer’s disease—damages the brain’s vascular system. In aged mice carrying APOE4, they were also able to reverse that specific damage by blocking a signalling pathway involved in scarring.
That sounds dramatic, but the boundary matters. The scientists did not reverse Alzheimer’s disease in people, restore lost memories or test a new medicine in a clinical trial. This is a preclinical discovery that identifies a potentially treatable mechanism. It could open a new direction for drug research, but it is still near the beginning of that journey.
The cells wrapped around the brain’s smallest vessels
The study, published in Cell on 24 September 2026, focused on pericytes. These cells wrap around capillaries—the body’s smallest blood vessels—and help keep them stable. In the brain, pericytes also support the blood-brain barrier, which tightly controls what can move from the bloodstream into brain tissue.
To investigate what APOE4 does to this system, the researchers assembled a single-cell atlas of human brain vasculature. This allowed them to compare the gene activity of individual vascular cells rather than treating a blood vessel as one uniform structure.
In brain tissue from APOE4 carriers, pericytes were less abundant. At the same time, the researchers found a population of cells resembling myofibroblasts—cells associated with contraction, tissue repair and scar formation. Further experiments indicated that APOE4 was pushing pericytes to change into this scar-producing state. Read the peer-reviewed Cell study.
The altered cells produced fibronectin, a structural protein that can accumulate in scarred tissue. In the models used by the team, this process promoted fibrosis around vessels and encouraged amyloid to collect along their walls. That combination could interfere with blood flow, weaken the blood-brain barrier and make it harder for the brain to clear waste.
The finding reframes vascular injury as more than collateral damage appearing late in Alzheimer’s disease. It suggests that, at least in APOE4 biology, changes in vessel-support cells may actively help drive the unhealthy environment in which neurodegeneration develops.
A signalling pathway became the off switch
The researchers identified increased activity in a pathway called transforming growth factor beta, or TGF-β. This signalling system has many jobs throughout the body, including controlling cell growth, immune responses, wound healing and tissue remodelling.
Here, elevated TGF-β signalling appeared to drive the transformation of pericytes into myofibroblast-like cells.
When the team inhibited the pathway, pericyte coverage around vessels recovered and both vascular fibrosis and amyloid fell to levels seen in their APOE3 comparison models. The effect was reproduced in aged APOE4 mice, where blocking TGF-β reduced scarring and amyloid around brain vessels. Mount Sinai’s research summary describes the work across human tissue, cellular models and animals.
This is the study’s most important result. It indicates that the vascular changes were not necessarily permanent in the experimental models. Once the relevant signalling was interrupted, some of the vessel-support system could recover.
APOE4 raises risk—it does not determine fate
APOE is a gene involved in transporting cholesterol and other fats. People inherit one copy from each parent, and several versions exist. APOE3 is the most common. APOE2 can be protective, while APOE4 raises the probability of developing Alzheimer’s disease.
About one-quarter of people carry one copy of APOE4 and roughly 2% to 3% carry two, according to the US National Institute on Aging. Crucially, carrying APOE4 does not mean someone will inevitably develop Alzheimer’s. Genetics changes risk; it is not a diagnosis. The National Institute on Aging explains APOE4 and Alzheimer’s risk.
That distinction also applies to this research. The study helps explain one mechanism through which APOE4 may create vulnerability in the brain. It does not offer a genetic test that predicts an individual’s future, and it does not show that vascular scarring is the only route from APOE4 to disease.
Why this is not yet a drug
TGF-β might look like an obvious target, but it is not a simple switch that can be turned off throughout the body without consequences. The pathway is involved in immune regulation, tissue repair and normal cell behaviour. Blocking it too broadly or for too long could create serious side effects.
A practical therapy may therefore need to reach the right cells in the brain, suppress only the harmful part of the pathway and preserve TGF-β’s useful functions elsewhere. Researchers will also need to learn whether treatment can protect cognition—not merely improve the appearance of blood vessels or reduce amyloid around them.
No human trial of this strategy was reported with the paper. Before clinical testing, the result will need further replication, dose and safety studies, and evidence that the vascular improvement produces a meaningful neurological benefit.
Mouse models are valuable for testing mechanisms, but they do not reproduce the full complexity or timescale of human Alzheimer’s disease. Even approaches that work convincingly in animals often fail in people.
A broader view of Alzheimer’s biology
The study arrives as Alzheimer’s research is widening beyond a single-minded focus on plaques between neurons. Approved anti-amyloid medicines have shown that removing amyloid can modestly slow decline for some patients, but they do not stop or reverse the disease. Blood vessels, immune cells, lipid metabolism and the brain’s waste-clearance systems are increasingly part of the picture.
That makes the new work significant even if TGF-β inhibition itself never becomes a medicine. It provides a mechanistic bridge between a major genetic risk factor, scar-forming vascular cells and amyloid accumulating around vessels. Each step offers researchers a place to look for safer and more selective targets.
The most responsible conclusion is also the most interesting one: a form of APOE4-linked brain-vessel degeneration that looked like permanent damage proved reversible in experimental models. The next challenge is discovering whether that biological insight can be converted into a safe treatment—and whether repairing the brain’s circulation can actually protect human memory.
Sources
- Cell: A pericyte-to-myofibroblast transition links APOE4 to cerebrovascular degeneration
- Mount Sinai: How APOE4 damages brain blood vessels
- National Institute on Aging: APOE4 and dementia risk
Featured image from the FutureTechDose archive, shown for context: Digital illustration of the human brain against a black background.
Leave a Reply