← Back to Home

Recognize the Brain, Then Attack: New CAR-T Cells Clear Pediatric Brain Tumors in Mice

Using a brain protein as a switch, B-SYNC cell therapy showed anticancer activity in multiple mouse models. But the paper also revealed that another target combination caused fatal neurotoxicity when delivered into the brain's ventricles, reminding researchers that activation in the right location still requires safe targets.

By SURL BioNews

For diffuse midline glioma (DMG), one of the greatest treatment challenges is that cancer cells are closely intermingled with brain tissue essential for life. Treatment options for this aggressive pediatric brain tumor are limited. Enabling immune cells to penetrate tumors while avoiding damage to normal tissue requires more precise control than recognizing cancer cells alone.

To address this, a team at the University of California, San Francisco designed B-SYNC CAR-T cells in which the brain environment controls production of the anticancer receptor. The study was published on October 28 in *Science Translational Medicine*: engineered T cells first sense the brain protein brevican through a synthetic Notch receptor (synNotch), then activate production of a chimeric antigen receptor (CAR) that recognizes EphA2 or IL-13Rα2.

This design separates “where am I?” and “whom should I attack?” into two recognition steps. Brevican provides an environmental signal from the central nervous system; it is not a cancer-specific marker. Whether an attack follows also depends on whether the cell surface carries a target recognized by the CAR. The study aims to restrict CAR expression to the region requiring treatment.

According to ScienceDaily's account of information from the university, in two mouse models implanted with human DMG cells, all mice receiving B-SYNC delivered into the brain's ventricles cleared their tumors, while conventional CAR-T controls failed to control the disease. The paper's abstract also reports that, in immunodeficient models, either intravenous or intraventricular administration promoted cell infiltration into tumors, persistence, and prolonged survival. These findings cannot be interpreted as showing the same success rate for all formulations and administration routes.

The team also tested B-SYNC made with mouse versions of the components and targeting B7-H3 in syngeneic mouse models with intact immune systems. Both administration routes produced durable tumor clearance and enabled the mice to resist reinoculated cancer cells, with no neurotoxicity detected. However, artificial tumor reinoculation and natural recurrence in patients remain different situations.

The other side of the safety findings must also be included. The paper reports that another version, B-SYNC-B/I, which recognizes B7-H3 or IL-13Rα2, improved efficacy when administered intravenously but caused fatal neurotoxicity when administered intraventricularly, involving low levels of B7-H3 in normal central nervous system tissue. This means that restricting activation to the brain cannot, on its own, eliminate the risk of unintended damage to normal brain tissue.

According to the university's public statements, the team is preparing a human trial that could begin as early as April 2027, but this remains a planned timeline. The current findings come from preclinical models and cannot be translated into cure rates for patients. What must be clarified next is which target combination, dose, and administration route can achieve both anticancer activity and safety in the human brain.

References

  1. ScienceDaily Top Health
  2. PubMed