The investigators assessed EPHA2 and EPHB4 by immunohistochemistry in 141 specimens from 116 patients, with 113 included in histological-category analyses. They reported EPHA2 expression across histologies. EPHB4 expression, by contrast, was observed more frequently in higher-grade tumors. Among the glioblastoma/gliosarcoma cohort specifically, 74% of specimens had an EPHA2 staining score of ≥2; 37% had an EPHB4 staining score of ≥2. The article did not provide patient-level demographic breakdowns or additional quantitative thresholds beyond those categorical staining scores.
Tumor-by-tumor variability was present. Some low- and intermediate-grade lesions showed EPHA2 staining, which suggests broad tissue distribution of that receptor in brain tumors. The study team noted that specimens with high expression of both receptors tended to cluster within high-grade groups, indicating potential overlap of targetable antigens in more aggressive disease.
High concurrent expression of EPHA2 and EPHB4 occurred largely in high-grade tumors, according to the authors. This concentration suggests a subset of aggressive brain tumors that may present both antigens for dual-target strategies. The source did not report molecular subtyping, nor did it report correlations with clinical outcomes such as progression-free survival or overall survival. The reported observations therefore describe receptor distribution rather than prognostic associations.
The study team generated EPHRIN-B2-based CAR-T cells using the piggyBac transposon system. The published abstract did not include stepwise manufacturing conditions, cell doses, or vector design specifics beyond the platform name. Investigators used these engineered T cells to probe antigen-specific activity in vitro and to test intracerebroventricular delivery in animal models. The authors reported the generation method to indicate a non-viral gene transfer approach, but operational details required for replication (for example, electroporation parameters or cell expansion timelines) were not provided in the abstract.
EPHRIN-B2 CAR-T reduced viable tumor cell numbers across six brain tumor cell lines. The same CAR-T cells increased secretion of interferon-γ and interleukin-2 when co-cultured with those lines, indicating activation and effector function. A receptor-negative control cell line did not elicit significant differences, supporting antigen-dependent activity in vitro. The abstract does not list the individual cell line names, their receptor expression profiles, or the exact effector-to-target ratios used, so the reader must rely on the reported summary endpoints rather than replicate-specific metrics.
Investigators administered EPHRIN-B2 CAR-T cells via the intracerebroventricular route in two separate intracranial xenograft models. Control arms included mock T cells, control CAR-T cells, and phosphate-buffered saline (saline). The abstract states that CAR-T treatment reduced tumor signals relative to saline-treated controls at selected time points. The design therefore tested local delivery, comparing engineered T cells against both cellular and non-cellular comparators.
Although the team observed reduced tumor signals versus saline at certain time points, differences between EPHRIN-B2 CAR-T and control T cells did not reach statistical significance in the reported analyses. The abstract does not include detailed tumor-signal trajectories, statistical values, or time-point-by-time-point breakdowns, so effect sizes and variance cannot be independently assessed from the text provided. The authors framed these findings as preliminary in vivo activity that did not yet demonstrate clear CAR-specific superiority over control T cell products.
The study characterizes receptor distribution and demonstrates antigen-dependent activity in vitro, but several limitations remain. The abstract omits manufacturing parameters for the piggyBac-generated CAR-T cells, lacks cell line identities and dosing schedules, and does not provide statistical detail for the in vivo comparisons. Investigators concluded the results provide a basis for further studies to establish CAR-specific efficacy in intracranial tumors. Registry information for a clinical trial was listed as N/A. Further preclinical work to define optimal dosing, delivery conditions, and definitive CAR-specific antitumor effects will be required before translation to clinical evaluation.