Targeted cell therapy shows promise in lab, mouse models of MG

An experimental cell therapy designed to selectively eliminate immune cells that produce the most common disease-driving antibodies in myasthenia gravis (MG) showed promise in laboratory and mouse models.

The treatment involves engineering immune T-cells to selectively kill B-cells producing self-reactive antibodies (autoantibodies) against the acetylcholine receptor (AChR). This protein is found on the surface of muscle cells at the neuromuscular junction, the site where nerve cells transmit signals to muscle fibers, triggering contraction.

In laboratory experiments, the engineered T-cells — called chimeric autoantibody receptor (CAAR) T-cells — recognized and killed cells mimicking human B-cells that produce anti-AChR antibodies. In mice, treatment reduced these cells and anti-AChR antibodies both in circulation and bound at the neuromuscular junction.

“Collectively, these findings support AChR CAAR T cells as a promising precision immunotherapy for patients with MG and support their advancement into first-in-human clinical evaluation,” the researchers wrote.

The study, “Bispecific chimeric autoantibody receptor T cells eliminate acetylcholine receptor-specific B cells in myasthenia gravis models,” was published in Nature Communications.

Recommended Reading

Aug. 24, 2026 Columns by Shawna Barnes

3 symptoms of myasthenia gravis that are often misunderstood

Engineering cells

MG is caused in most cases by self-reactive antibodies that target proteins at the neuromuscular junction, most commonly AChR. Disrupted nerve-cell communication leads to the hallmark MG symptoms of muscle weakness and fatigue.

CAR T-cell therapies have emerged as a potential strategy to leverage the body’s immune system to kill B-cells, thereby lowering MG-driving antibodies. Some of these therapies, including Descartes-08 and miv-cel (formerly KYV-101), are currently being tested in MG clinical trials.

These therapies involve collecting a person’s immune T-cells and engineering them in the lab to produce a chimeric antigen receptor (CAR) protein that directs them to a specific target on the surface of B-cells. The engineered T-cells are then expanded in the lab and infused back into the patient, where they can seek out and eliminate B-cells.

CAR T-cell therapies, however, can also eliminate healthy B-cells that produce protective antibodies. This broad B-cell depletion can leave patients with low antibody levels, increasing their susceptibility to infections.

To make this approach more selective, a team led by researchers in Berlin developed CAAR T-cells engineered to selectively target B-cells producing anti-AChR antibodies.

The T-cells were engineered to produce two different CAARs that bind to autoantibodies targeting the alpha-1 and beta-1 subunits of the AChR protein — the two major targets of MG-driving antibodies. Because these antibodies are also found at the surface of the B-cells that produce them, the CAAR T-cells can selectively recognize and eliminate those B-cells.

The researchers tested whether the CAAR T-cells could respond to and kill lab-grown human cells engineered to carry alpha-1- or beta-1-targeting antibodies on their surface, mimicking the self-reactive B-cells found in MG.

Both alpha-1- and beta-1 CAAR T-cells became activated, multiplied, and released immune signaling molecules when they encountered their intended targets. They also efficiently killed the corresponding target cells, while sparing cells carrying an unrelated antibody. Similar responses were seen when individual T-cells were engineered to carry both CAARs.

Because people with AChR-related MG also have anti-AChR antibodies freely circulating in their blood, the researchers asked whether these antibodies could interfere with the CAAR T-cells. To mimic this in the lab, they added free-floating, lab-produced AChR-targeting antibodies to cultures containing the CAAR T-cells and their target cells.

Even at concentrations within or above those reported in people with MG, the free-floating autoantibodies did not interfere with ability of CAAR T-cells to kill their target cells, either when using separate alpha-1 and beta-1 CAAR T-cells or dual-targeting CAARs T-cells.

The team next tested the approach in severely immunodeficient mice that were injected with lab-grown human cells producing anti-AChR antibodies, allowing them to assess whether CAAR T-cells could eliminate their intended targets in living animals.

Initial experiments with alpha-1 CAAR T-cells supported the approach, reducing the target autoantibody-producing cells and antibody deposits at the neuromuscular junction while increasing levels of AChR protein at the neuromuscular junction.

The researchers then tested the administration of separate alpha-1 and beta-1 CAAR T-cells and dual-targeting CAAR T-cells in mice carrying a mixture of cells producing either alpha-1- or beta-1-targeting antibodies.

Both approaches markedly reduced target cell populations, significantly lowered both types of autoantibodies in the blood, and reduced antibody deposits at the neuromuscular junction.

However, the mouse model “did not permit evaluation of clinical disease reversal, as animals lacked a functional complement system and did not develop overt muscle weakness,” the researchers wrote. The complement system is a part of the immune system that contributes to MG-related damage.

“Future studies using models with complement activity may enable direct assessment of clinical efficacy,” the scientists said.

Still, they said, the findings “establish AChR CAAR T cells as a precision immunotherapy with the potential to achieve durable remission in [hard-to-treat] MG.”