Introduction
Cell therapies have introduced a critically needed therapeutic option for pediatric patients with high-grade glioma (HGG), providing hope to many patients and their families.1 There are no known curative therapies for HGG, a devastating central nervous system (CNS) tumor that occurs in children and young adults. Early signals suggest that treatment with chimeric antigen receptor T (CAR T) cells delivered directly to the CNS may induce tumor regression. A pioneering clinical trial evaluating lentiviral GD2-directed CAR T cells in patients with diffuse midline glioma (DMG) demonstrated favorable tumor response in multiple patients.2 However, 100% of patients developed tumor inflammation-associated neurotoxicity with intracerebroventricular (i.c.v.) injection.2 The long-term side effect profile from CNS-delivered lentiviral CAR T therapies is not yet known, and preclinical murine experiments suggest that neurocognitive decline emerges following lentiviral CAR T cell treatment.3
Alternative strategies that preserve efficacy while minimizing acute and long-term neurological function in pediatric patients are needed. To address this therapeutic gap, in our recently published paper in Neuro-Oncology, we developed mRNA CAR T cells that enable a tunable dosing strategy for CAR T therapy.4 This treatment carries the goal of providing effective CAR T therapy to children with CNS tumors with lower toxicity burden. The transient nature of mRNA CAR T cells allows for direct-to-CNS delivery of CAR T cells with the ability to provide dose-adjusted therapeutic delivery (Figure 1).
Figure 1.
Schematic showing production of lentiviral versus mRNA CAR T therapies and their anticipated cerebrospinal exposure upon treatment
“Created in BioRender. Pomaville, M. (2026) https://BioRender.com/4nfzuxp.”
Main
CAR T cells are traditionally generated using a modified lentiviral vector. Virus delivers a gene encoding a CAR designed to recognize targets present on tumor cells. Lentiviral CAR T cells are engineered to possess properties to enable long-term durability and induce a robust anti-tumor immune response.5 While anti-tumor effect is demonstrated in multiple models, lentiviral CAR T cells introduced to the CNS have induced toxic inflammatory responses in preclinical models and in patients, and even inducing death in some preclinical HGG models.6 The sensitive location of midline CNS tumors may predispose patients and animal models used in preclinical studies to inflammatory-induced toxicities.6 Mice with DMGs located in the thalamus have shown particular toxicity with lentiviral CAR T injection.
GD2 is a disialoganglioside glycosphingolipid present on the cell surface across various subtypes of pediatric-type HGG, including DMG, and other pediatric tumors.7 We have found that CAR T cells can be effectively engineered by transfecting T cells with mRNA encoding the GD2 CAR. Transfection using electroporation in this approach showed high efficiency (90%) and potency. mRNA CAR T cells exhibited maximal cell kill activity at 1 day, with minimal activity by 9 days in HGG cell culture methods. Functional assays confirmed presence of interferon gamma degranulation. In addition to inducing toxicity in cell line models, GD2 mRNA CAR T cells induced cytotoxicity in three-dimensional DMG organoid models. In vivo studies confirmed a dose-dependent improvement in survival with mRNA CAR T cells, with no mRNA CAR T cell deaths in mice with pontine tumors. This effect was repeated in specific thalamic murine models of DMG, where aggressive dose titration recapitulated toxicity seen with lentiviral CAR T therapy, and downward mRNA CAR T titration reduced toxicity while maintaining efficacy.
Summary and future directions
GD2-directed mRNA CAR T cells represent an innovative strategy to reconcile toxicity with anti-tumor efficacy for patients with HGG. This study showed that multiple subtypes of pediatric HGG express the tumor antigen GD2, and that mRNA CAR T cells could be generated with high expression of GD2 CAR via electroporation. Addition of GD2 mRNA CAR T cells to cell culture models of HGG induced potent cytotoxicity. Intratumoral (i.t.) injection of GD2 mRNA CAR T therapy into murine HGG xenografts led to improved overall survival and toxicity profiles in relation to previously published lentiviral-transduced GD2 CAR T cells. The i.t. delivery of immunotherapies is emerging as a more frequent strategy to administer these therapies to patients with HGG, making these findings translationally relevant to patient care. Ultimately, CAR T therapy, including those directed at GD2, holds promise for patients with HGG, as demonstrated by early clinical trial findings that some patients exhibit decreased tumor size after treatment with lentiviral GD2 CAR T cells. We have an immediate goal to translate these findings to patients through the development of a phase 1 clinical trial evaluating i.t.-delivered mRNA GD2 CAR T cells in pediatric patients.
A tailored, dose-adjusted approach to immunotherapy may represent an important clinical strategy to effectively treat patients. Importantly, disease-specific characteristics may define the efficacy and tolerability of novel therapies. Lentiviral-transduced CAR T therapies that expand in patients have been curative for the treatment of certain hematologic malignancies such as relapsed B cell acute lymphoblastic leukemia (BALL), with cohorts of patients experiencing durable disease responses with minimal toxicity.8 For patients that do experience toxicity with disease such as cytokine release syndrome (CRS), therapies directed at mechanism-defined toxicities have helped to bridge patients from resolution of toxicity to disease response.9 In contrast, i.c.v.-delivered lentiviral GD2 CAR T cells induce neurotoxicity in nearly every patient through alternative inflammatory mechanisms, with signal of effect limited to some patients. Furthermore, repeated doses are typically required for effect, even with lentiviral CAR T constructs.
mRNA CAR T therapies may unveil a suite of solutions to common barriers to effective and accessible CAR T therapy. mRNA CAR T therapy is both less expensive to manufacture and more efficiently generated, with no risk of DNA integration mutagenesis. The short-lived, dosed CAR T approach may allow quick, active CAR T presence that repeatedly capitalizes on the effector function of T cells in repeated doses. Ultimately, a more integrated understanding of tumor-neural-immune system approach is necessary to develop safe, effective therapies against high-grade gliomas.10 To advance this work further, a combination strategy that enhance efficacy of GD2 mRNA CAR T cells are under investigation and more nuanced strategies leveraging the microenvironment of HGG. Lastly, generation of mRNA CAR T therapies against alternative targets are also being tested, which may reveal whether the effects demonstrated in this manuscript are generalizable across multiple mRNA CAR T constructs in pediatric HGG.
Declaration of interests
The authors have no conflicts of interest to disclose related to this article.
References
- 1.Ronsley R., Bertrand K.C., Song E.Z., Timpanaro A., Choe M., Tlais D., Vitanza N.A., Park J.R. CAR T cell therapy for pediatric central nervous system tumors: a review of the literature and current North American trials. Cancer Metastasis Rev. 2024;43:1205–1216. doi: 10.1007/s10555-024-10208-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Monje M., Mahdi J., Majzner R., Yeom K.W., Schultz L.M., Richards R.M., Barsan V., Song K.W., Kamens J., Baggott C., et al. Intravenous and intracranial GD2-CAR T cells for H3K27M(+) diffuse midline gliomas. Nature. 2025;637:708–715. doi: 10.1038/s41586-024-08171-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Geraghty A.C., Acosta-Alvarez L., Rotiroti M.C., Dutton S., O'Dea M.R., Kim W., Trivedi V., Mancusi R., Shamardani K., Malacon K., et al. Immunotherapy-related cognitive impairment after CAR T cell therapy in mice. Cell. 2025;188:3238–3258.e25. doi: 10.1016/j.cell.2025.03.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Foster J.B., Madsen P.J., Harvey K., Griffin C., Stern A., Patterson L., Joshi N., Dickson C., McManus O., Beaubien E., et al. Transient mRNA CAR T cells targeting GD2 provide dose-adjusted efficacy against diffuse midline glioma and high-grade glioma models. Neuro Oncol. 2025;27:2684–2696. doi: 10.1093/neuonc/noaf115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Ramesh P., Hui H.Y.L., Brownrigg L.M., Fuller K.A., Erber W.N. Chimeric antigen receptor T-cells: Properties, production, and quality control. Int. J. Lab. Hematol. 2023;45:425–435. doi: 10.1111/ijlh.14121. [DOI] [PubMed] [Google Scholar]
- 6.Mount C.W., Majzner R.G., Sundaresh S., Arnold E.P., Kadapakkam M., Haile S., Labanieh L., Hulleman E., Woo P.J., Rietberg S.P., et al. Potent antitumor efficacy of anti-GD2 CAR T cells in H3-K27M(+) diffuse midline gliomas. Nat. Med. 2018;24:572–579. doi: 10.1038/s41591-018-0006-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Fernández-Rilo A.C., de Billy E., Del Baldo G., De Angelis B., Rossi S., Del Bufalo F., Quintarelli C., Locatelli F., Mastronuzzi A., Vinci M. GD2: hopes and challenges for the treatment of pediatric patients with tumors of the central nervous system. npj Precis. Oncol. 2025;9:295. doi: 10.1038/s41698-025-01079-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Maude S.L., Laetsch T.W., Buechner J., Rives S., Boyer M., Bittencourt H., Bader P., Verneris M.R., Stefanski H.E., Myers G.D., et al. Tisagenlecleucel in Children and Young Adults with B-Cell Lymphoblastic Leukemia. N. Engl. J. Med. 2018;378:439–448. doi: 10.1056/NEJMoa1709866. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hughes A.D., Teachey D.T., Diorio C. Riding the storm: managing cytokine-related toxicities in CAR-T cell therapy. Semin. Immunopathol. 2024;46:5. doi: 10.1007/s00281-024-01013-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Song K.W., Lim M., Monje M. Complex neural-immune interactions shape glioma immunotherapy. Immunity. 2025;58:1140–1160. doi: 10.1016/j.immuni.2025.04.017. [DOI] [PubMed] [Google Scholar]

