Alzheimer’s plaques decline after CAR-T immune cell treatment in preclinical study

By Hugo Francisco de SouzaReviewed by Susha Cheriyedath, M.Sc.Feb 11 2026

A novel CAR-T cell approach targets amyloid plaques in preclinical Alzheimer models, raising the possibility that engineered immune therapies could reshape future treatment strategies while major clinical questions remain.

Study: Engineering chimeric antigen receptor CD4 T cells for Alzheimer’s disease. Image Credit: Andrii Vodolazhskyi / Shutterstock

In a recent study published in Proceedings of the National Academy of Sciences, researchers engineered Chimeric Antigen Receptor (CAR) CD4 T cells, a technology originally revolutionized for cancer treatment, to target fibrillar amyloid-beta (Aβ) plaques in the brain.

The study leveraged murine (mouse) models to demonstrate that these “smart” reprogrammed immune cells can effectively reduce amyloid deposition in distinct anatomical compartments depending on the delivery strategy, including the protective membranes of the brain and the brain tissue itself. This approach represents an early proof-of-concept advance rather than a clinical breakthrough in Alzheimer’s disease (AD) and cellular immunotherapy in neurodegeneration.

Alzheimer’s Pathology and Limits of Current Immunotherapies

Alzheimer’s disease (AD), a progressive neurodegenerative condition characterized by severe cognitive decline and behavioral alterations, remains the leading cause of age-associated dementia. Despite decades of research aimed at mitigating and treating the condition, current ‘gold standard’ antibody treatments are reported to offer only marginal cognitive benefits, although clinical responses vary across trials and patient populations.

Neurodegenerative biology elucidates that the hallmark of Alzheimer’s disease is the toxic accumulation of Aβ plaques in the parenchyma (the functional tissue of the brain), which subsequently triggers neurofibrillary tangles and microglial activation, eventually leading to brain atrophy and memory loss.

Current interventions, such as anti-amyloid antibodies like Lecanemab and Donanemab, have been observed to clear some of these plaques in preclinical and clinical trials, but a growing body of evidence suggests that their clinical efficacy remains limited.

Recent breakthroughs in neuroimmunology have shown that T cells may play a dual role in the brain. While most T cells function primarily in adaptive immune signaling rather than direct phagocytosis, CD4+ T cells (helper T cells) have demonstrated significant potential to regulate inflammation and improve cognitive performance.

Unfortunately, attempts to program these cells to recognize specific Alzheimer’s targets without triggering a broad autoimmune response have been a significant hurdle for neurobiological research.

Study Design: Engineering and Delivery Strategies

The present study aims to overcome these limitations by repurposing a groundbreaking anti-cancer technology called Chimeric Antigen Receptor T-cell (CAR-T) therapy, which enables researchers to genetically engineer a patient’s own T cells to detect and destroy cancer cells. This study specifically aims to leverage this technology to enable antigen-specific targeting of amyloid pathology rather than fully bypassing central nervous system immune barriers.

The study used the 5xFAD mouse model because it mimics the rapid amyloid buildup observed in human Alzheimer’s disease. Simultaneously, the study engineered CD4+ T cells with synthetic CAR receptors featuring a “targeting head” derived from the antibody Lecanemab, fused to internal signaling components that instruct the T cell to activate upon encountering a plaque.

The study subsequently investigated the preclinical efficacy of two primary delivery methods: (1) stable retroviral transduction, which creates “permanently” programmed T cells, thereby providing insights into long-term behavior, and (2) transient messenger ribonucleic acid (mRNA) nucleofection, a technique that uses mRNA to program the cells temporarily.

The latter delivery approach was used to represent a “safety-first” strategy, as the CAR expression has been previously observed to fade naturally, preventing the cells from remaining active indefinitely and potentially causing persistent immune activation or other safety concerns described in CAR-T literature, including neurotoxicity syndromes observed in oncology CAR-T applications.

The study’s primary endpoints included amyloid coverage, microgliosis (activation of the brain’s resident immune cells), and astrogliosis (the expansion of support cells called astrocytes that often occurs in diseased tissue).

Preclinical Findings in Murine Alzheimer’s Model

The study analyses revealed that the Lecanemab-derived CAR (specifically the Lec28z version) proved highly selective, activating only in the presence of fibrillar amyloid (the “sticky” form found in plaques) while remaining inactive in the presence of monomeric amyloid forms that did not trigger CAR signaling in the experimental assays (p < 0.0001).

Specifically, the murine model experiments revealed that stable engineered CAR-T treatment significantly reduced amyloidosis in the dura (the outermost brain membrane), particularly at “exit points” where waste is typically removed (p = 0.0151). However, this stable approach did not significantly reduce parenchymal plaques and was associated with some increases in microglial activation markers, highlighting a complex inflammatory response whose clinical implications remain uncertain.

Furthermore, when observing the transient mRNA-based cells, the study recorded a significant reduction in parenchymal plaque load (Aβ coverage, p = 0.0127; methoxy-stained dense cores, p = 0.0339).

Finally, the CAR-T intervention was shown to reduce markers of neuroinflammation, including microgliosis (Iba1 coverage; p = 0.0220) and astrogliosis (GFAP coverage; p = 0.0055), specifically in the transient-expression condition. Notably, the study authors highlight that the treatment promoted the recruitment of endogenous CD4 T cells into the brain, suggesting that the “living drug” was associated with broader immune engagement, although the precise mechanisms remain uncertain and may involve both direct plaque recognition and indirect modulation of the neuroimmune environment.

Interpretation and Translational Implications

The present study provides the first successful proof of concept demonstrating that CD4+ CAR-T cells can be engineered to specifically target amyloid pathology in a murine Alzheimer’s model, rather than fully neutralizing neurodegenerative disease processes in preclinical systems. The study notably used mRNA for transient expression, thereby mitigating concerns about long-term toxicity and potential CAR-T-associated neuroimmune adverse effects described in oncology literature.

While the results are promising, the authors note that further calibration of receptor signaling strength and persistence is needed before moving to human trials. These findings pave the way for a new generation of cellular immunotherapies that could one day offer a more durable resolution for those living with Alzheimer’s, although substantial safety, mechanistic, and clinical efficacy questions remain before translation to patients, including whether amyloid reduction will translate into meaningful cognitive benefit.