The current generation of Alzheimer's treatments — monoclonal antibodies like lecanemab and donanemab — represent genuine progress. They slow the disease. They buy time. In a condition where time is the most precious commodity, that matters.
But they have significant limitations. They require infusions, sometimes monthly. They can cause brain swelling and microbleeds in a meaningful proportion of patients. And they produce modest benefits — an average of about 10 additional months of independent function — rather than transformative ones.
The race to find something better is intense. A new approach from researchers at **Washington University School of Medicine in St. Louis**, published in *Science* in March 2026, may represent one of the most promising directions yet — and it starts with a completely different kind of thinking about the brain's own cells.
**Turning Brain Cells Into Cleaners**
The new strategy is directly inspired by CAR-T cell therapy, which has transformed treatment for certain blood cancers. In CAR-T, immune T cells are extracted from a patient, genetically engineered to carry a 'chimeric antigen receptor' (CAR) that allows them to recognise and destroy cancer cells, and then reinfused.
The Washington University team adapted this approach for the brain — but instead of using T cells, they turned to **astrocytes**: the most abundant type of cell in the central nervous system. Astrocytes normally perform a wide range of supporting functions — maintaining the blood-brain barrier, regulating neurotransmitters, providing nutrients to neurons.
By equipping astrocytes with a CAR targeting amyloid beta — the protein that aggregates into the plaques characteristic of Alzheimer's disease — the researchers created what they describe as 'powerful cleaning cells in the brain.' These engineered CAR-astrocytes can identify, latch onto, and destroy amyloid plaques with remarkable specificity.
**The Mouse Results**
In mouse models of Alzheimer's disease, the results were striking:
- When CAR-astrocytes were administered **before** plaque formation began, they **prevented plaques from forming entirely** - When given to mice that **already had plaques**, the therapy **reduced amyloid levels by approximately half** - The treatment was delivered as **a single injection**
A single injection — not a monthly infusion, not a lifetime course of treatment. One intervention that the body's own cells then carry out continuously.
'This study marks the first successful attempt at engineering astrocytes to specifically target and remove amyloid beta plaques in the brains of mice with Alzheimer's disease,' said senior author **Marco Colonna MD**, the Robert Rock Belliveau MD Professor of Pathology at WashU Medicine. 'Although more work needs to be done to optimize the approach and address potential side effects, these results open up an exciting new opportunity.'
**Why Astrocytes?**
The choice of astrocytes is significant. Unlike T cells — which are peripheral immune cells that need to be engineered to cross the blood-brain barrier — astrocytes already live inside the brain. They are native to the environment where Alzheimer's does its damage.
This means a CAR-astrocyte approach doesn't need to solve the blood-brain barrier problem in the same way that many other therapies do. The cells are already there. The question is only whether they can be given the right targeting capability — and the Washington University study suggests they can.
There is also potential beyond Alzheimer's. Colonna noted that the approach could potentially be extended to other neurodegenerative diseases and even brain tumours — any condition where a harmful target in the brain needs to be cleared or destroyed.
**The Distance to Human Treatment**
Mice are not people. The gap between a mouse model result and a human clinical trial is substantial, and the gap between a Phase 1 trial and a licensed treatment wider still. The team acknowledges that optimisation and safety work remain significant challenges.
But the conceptual leap here is important. The existing approved Alzheimer's treatments all work on the same basic mechanism: reducing amyloid levels through monoclonal antibodies. The CAR-astrocyte approach represents a fundamentally different strategy — using the brain's own cellular machinery, rather than a drug infused from outside, to do the cleaning.
For the 55 million people worldwide living with dementia — and the families walking that journey alongside them — this kind of research is the kind that deserves to be watched closely. 🧠
*Sources: ScienceDaily (sciencedaily.com, March 5, 2026) · Washington University School of Medicine in St. Louis (medicine.wustl.edu) · Science journal (science.org) · Professor Marco Colonna, MD*
