CAR-T Cell therapy in CNS malignancies and metastatic disease: a quick overview of toxicity profiles and efficacy signals

Aleta Biotherapeutics has developed mono- and dual-targeting CAR-T Engager proteins (CTEs) for solid tumors, CNS metastatic lesions, and CNS primary tumors. They are designed to bind to tumor antigens and to CD19-CAR T cells, the latter providing a persistent and immunologically relevant antigen depot that is independent of the targeted tumor, ie. normal B cells. We and others have published results highlighting the attractiveness of B cell support for the CAR T cells.

CTEs contain the CD19 extracellular domain, the universal target for all anti-CD19 CAR T cells including approved therapeutics such as axi-cel and liso-cel. CTE programs in development target HER2 (ALETA-002), B7-H3/HER2 (ALETA-003) and IL-13Rα2/B7-H3 (ALETA-006). Indications of interest include HER2-positive solid tumors, CNS metastatic disease, and primary CNS cancers.

Here we review CAR-T cells targeting various antigens that have been evaluated in clinical trials  for CNS solid tumors and CNS metastases. Antigens targeted by CAR-T cell therapies include EGFRvIII/EGFR, IL-13Rα2, HER2, GD2, B7-H3, and EphA2. Of particular interest are signals of efficacy and CAR-T-associated toxicities.

In an earlier post we examined some of the cited literature in depth (https://www.sugarconebiotech.com/insights/recent-car-t-clinical-studies-in-glioma-and-glioblastoma-patients).

Toxicities

Cytokine Release Syndrome (CRS)

CRS is the most common type of toxicity after CAR-T cell therapy and is characterized by fever, hypotension, tachycardia, hypoxia, and, in severe cases, multiorgan dysfunction. It typically occurs within hours to days following CAR-T cell administration, though late-onset cases have been reported. The release of inflammatory cytokines such as IL-6, IFN-γ, and TNF-α contribute to its pathogenesis.

A distinctive observation from clinical studies in CNS primary tumors is that severe CRS is paradoxically uncommon. CAR-T toxicity, especially systemic, is much lower in brain tumor patietns than in hematological malignancy patients.  For example, CAR-T cell trials in glioma and GBM patients only rarely show severe CRS1-3. Indeed, moderate to severe CRS has been generally absent in GBM trials. This observation must be tempered with a cautionary note that efficacy in these trials has been very limited. The modest efficacy and minimal toxicities may be due to low antigen burden on CNS tumors as compared to hematological malignancies, limited expansion, and persistence of the CAR-T cells and, potentially, faster CAR-T cell exhaustion.

ICANS (Immune Effector Cell-Associated Neurotoxicity Syndrome)

ICANS is a neurotoxicity syndrome that usually develops around one week after CAR-T infusion and is often preceded by CRS. As with CRS, ICANS is more commonly seen with systemic CAR-T cell therapy administration. ICANS is associated with encephalopathy, tremors, dysgraphia, seizures, apraxia, aphasia, acalculia, and in rare cases, cerebral edema.

TIAN (Tumor Inflammation-Associated Neurotoxicity)

Tumor inflammation-associated neurotoxicity (TIAN) is a CNS-specific adverse event, first observed in glioma patients given CAR-T therapy4,5. TIAN is hypothesized to be caused by intracranial “compartmentalized” cytokine release within the tumor mass. Preliminary data suggested a relationship between CAR-T dose and appearance of TIAN. 

Two mechanistic subtypes have been characterized: Type 1 TIAN (mechanical TIAN) resulting from inflammatory edema at the tumor site leading to mechanical space constraints that can cause obstruction of CSF flow with subsequent hydrocephalus, increased intracranial pressure, and herniation syndrome. Type 2 TIAN (electrophysiologic TIAN) reflects localized dysfunction of tumor-infiltrated neural circuits that can cause transient worsening of neurological deficits. Type 1 TIAN can cause very severe toxicities. In the glioma trials run at Stanford the toxicities were managed with anakinra (IL-1Rα antagonist) and steroids.

CAR-T cell therapies directed to specific antigens

HER2-CAR-T cells for CNS metastatic disease and solid tumors

Epha2-CAR-T cell treatment-induced pulmonary edema demonstrated the issue of on-target, off-tumor toxicity6. HER2-CARs carry a theoretical concern for cardiac toxicity given myocardial expression of HER2 and the adverse-event profiles of anti-HER2 antibodies and ADCs7,8.  However, studies of HER2-CAR-T cell therapy for systemic malignancies have not revealed cardiac-related toxicities, even after multiple CAR-T cell injections9.  Notably, a trial of 17 patients treated with HER2-targeting virus-specific T cells (Adv, CMV, or EBV), no dose-limiting toxicities were seen and ventricular function remained unchanged10. CAR-T cells expanded and persisted in patients in this due to the continued presentation of virus-specific antigen, and modest signals of efficacy were obtained. Intraventricularly delivered HER2-targeted CAR T-cell therapy appeared to be a safe and potentially effective treatment for HER2-positive breast cancer with CNS metastases11,12. Early Phase 1 studies showed manageable toxicity and suggest that locoregional administration and combining therapy with lymphodepletion improved CAR T-cell persistence and activity.

EGFRvIII-Targeted CAR-T cells for Glioblastoma (GBM)

EGFRvIII is a highly oncogenic, constitutively active mutation of the EGFR receptor found in approximately 50% of GBM patients. In a UPenn study, 10 patients with recurrent EGFRvIII+ GBM were treated with a single peripheral dose of EGFRvIII-Targeted CAR-T cells13. No toxicities were reported but there were only limited signs of efficacy, and antigen expression was rapidly lost on the tumor cells. One patient achieved long-term remission14. In a followup study, recurrent injections of CAR-T cells were administered. There were no severe adverse event but efficacy was limited15. An NIH trial investigating dose-escalation of EGFRvIII-targeted CAR-T cells was halted due to instances of severe CRS including lethal hypoxia at higher doses16. An emerging theme is that expression of this antigen is too labile for effective single antigen targeting. Also, toxicity can be related to CAR-T cell dose.

IL-13Rα2-Targeted CAR-T for High-Grade Glioma and GBM

IL-13Rα2 is overexpressed in GBM, DIPG, brainstem gliomas and breast cancer brain metastases. Cityof Hope investigators developed IL-13Rα2-CAR-T cells17. Sixty-five patients with recurrent high grade glioma were enrolled, the majority of which had recurrent GBM. Fifty-eight patients received at least three CAR-T infusions and were evaluable for disease response. One arm of the study used dual intratumoral/intraventricular delivery and a high dose of 2 × 108 CAR-T cells. Additional infusions were allowed until disease progression. In this arm, about 40% of patients survived beyond 1 year and one durable complete response was achieved. Severe adverse events were uncommon.

Dual CAR-T targeting EGFRvIII and IL-13Rα2 for GBM

These CAR-T studies targeting EGFRvIII and IL-13Rα2 showed manageable toxicities but limited efficacy. In an attempt to improve responses, dual targeting has been developed. Bivalent CAR-T cell targeting EGFR and IL-13Rα2 were administered by the intraventricular route18. Of 18 patients, 9 (56%) experienced grade 3 neurotoxicity (ICANS) during dose escalation; the maximum tolerated dose was determined to be 2.5 × 107 CAR-T cells. By TIAN evaluation, 2 of 18 patients (11%) had grade 3 neurotoxicity. One patient with grade 3 ICANS then experienced long-term grade 1 ICANs that spontaneously resolved after 9 months. Notably, that patient also achieved durable stable disease. One patient (6%) had grade 4 TIAN characterized by acute obstructive hydrocephalus in the setting of peritumoral edema. One patient achieved a partial response. Most patients experiencing tumor relapse within the first 1–3 months following CAR-T cell infusion. It is plausible to suggest that EGFRvIII is not a useful second antigen in the dual-CAR setting due to antigen downregulation.

B7-H3-Targeted CAR-T for Diffuse Intrinsic Pontine Glioma (DIPG) and GBM

B7-H3 is nearly universally expressed on the surface of atypical teratoid rhabdoid tumors, DIPG, DMG, medulloblastomas, and HGG. There is also widespread expression on diverse peripheral solid tumors and expression is associated with metastasis. Seattle Children’s Hospital developed B7-H3 CAR T cells for children with recurrent or refractory CNS tumors and DIPG. The BrainChild-03 trial demonstrated preliminary tolerability of repetitive intraventricular doses as a strategy to fractionate dosing and to replenish the B7-H3 CAR T cell population19. The first report of 3 pediatric DIPG patients including two who enrolled after progression. Among these 3 patients, 40 infusions of CAR-T cells were given with no dose-limiting toxicities. One patient had sustained clinical and radiographic improvement through 12 months on study. The second report disclosed 23 DIPG patients enrolled with 21 treated with repeated doses of B7-H3 CAR-T cells20. There was no prior lymphodepletion. The study used intra-patient dose-escalation. There was one dose-limiting toxicity of grade 4 intratumoral hemorrhage reported in a patient who had progressive disease between enrollment and initial infusion. The median survival was 19.8 months for all treated patients which was superior to the historical median survival of 11.2 months. Notably, 3 patients were still alive at 44, 45, and 52 months from diagnosis.

BrainChild Bio's BCB-276 (the commercial version of the Seattle Chlidren’s therapy) was granted FDA Breakthrough Therapy Designation and Regenerative Medicine Advanced Therapy designation for the treatment of DIPG. BrainChild Bio is preparing to advance BCB-276 in a Phase 2 multicenter pivotal registration trial to support a potential Biologics License Application (BLA) to the FDA, with plans to expand into adult CNS tumors including glioblastoma and brain metastases.

Beyond pediatric tumors, B7-H3 CAR T cells have been studied in adults with recurrent GBM21. Thirteen adult GBM patients in China received B7-H3-targeting CAR T cells administered intracavity and/or intraventricularly. There were no dose-limiting toxicities. Among patients eligible for 12-month survival assessment, the 12-month OS rate was 83.3%, with a median OS of 20.3 months. In a seperate North American trial, 4 patients with IDH wild-type glioblastoma received intraventricular B7-H3 CAR T cells in 3 weekly infusions. Several durable partial responses were reported with no severe adverse events22.

The studies using IL-13Rα2 and B7-H3 CAR-T cells demonstrated acceptable sarety and encouraging signs of efficacy. Investigators are exploring multiplex targeting strategies combining B7-H3 with HER2, EGFR, and IL13Rα2 in a single CAR T cell product to overcome tumor antigen heterogeneity and escape.

GD2-Targeted CAR-T for GBM, DIPG and Diffuse Midline Glioma (DMG)

Disialoganglioside GD2 is a cell surface tumor-associated antigen highly expressed in tumors including neuroblastomas, small cell lung cancer, melanomas, retinoblastomas, sarcomas, and GBM. GD2 is highly expressed on H3K27M-mutant glioma cells, and thus is a target in DIPG.

In a Chinese trial, 8 patients with GD2-positive GBM received CAR-T cells either through intravenous administration alone or intravenous combined with intracavitary infusion23. The 4th generation CAR-T cells (with both 4-1BB and CD28 costimulatory domains) expanded for 1-3 weeks and persisted at a low frequency in peripheral blood. There were no severe adverse events reported, including neurotoxicity or off-target effects. Four patients showed a partial response and some responses were durable. For the entire cohort, the median overall survival was 10 months from the infusion. Infiltrated T cells were observed in tumor samples resected after infusion. However, GD2 antigen loss was also demonstrated.

In a pediatric DIPG study, dose escalation was limited by toxicity (CRS) at the higher dose of 3 × 10⁶/kg CAR-T cells delivered IV. On-target, off-tumor toxicity was not observed. All patients exhibited tumor inflammation-associated neurotoxicity that was managed with intensive monitoring and care. Four patients demonstrated major volumetric tumor reductions of 52, 54, 91 and 100%. One patient exhibited a complete response ongoing for over 30 months since enrollment. Nine patients demonstrated neurological benefit, as measured by a protocol-directed clinical improvement score. Thus, sequential intravenous and intracerebroventricular GD2-CAR-T infusions induced tumour regressions and neurological improvements without undue toxicities24,25. 

Variables That Correlate with Toxicity

Route of Administration

The studies persented show a shift away from peripheral delivery of CAR-T cells and toward locoregional delivery, into either the ventricles or tumor cavity. Although intracranial delivery bypasses the blood-brain barrier and decreases the risk of some systemic toxicities, studies have shown that central infusions can still lead to neurotoxic effects and systemic cytokine release syndromes. Intraventricular delivery is associated with more direct TIAN due to immediate engagement of tumor tissue, while intravenous delivery is associated with more systemic CRS.

CAR Construct Design — Costimulatory Domain

Whether the costimulatory domain impacts toxicity in CNS CAR-T studies is unclear, with claims made in support of both the CD28 and 4-1BB domains.

Tumor Burden and Antigen Load

High tumor antigen burden drives more rapid and intense CAR-T activation, particularly with systemic delivery, leading to CRS. CAR-T cell neurotoxicity has been correlated with pre-treatment disease burden, extent of CAR-T cell expansion, early/severe CRS, and CAR-T cell dose. Faster CAR-T cell expansion in vivo has also been related to the onset and severity of ICANS. 

Tumor Location in the CNS

Tumor location shapes the character of TIAN toxicity. Midline and brainstem tumors pose the highest risk of mechanical TIAN from edema in anatomically constrained compartments. For example, CAR-T cell-induced brainstem inflammation can result in obstructive hydrocephalus and increased intracranial pressure. As such, neurocritical care precautions were incorporated from the outset in the DIPG GD2-CART trial. 

Lymphodepletion Regimen

The type of costimulatory domain in the CAR construct and lymphodepletion regimens have been implicated in ICANS risk. Pretreatment inflammatory markers including IL-6, C-reactive protein, and fibrinogen are associated with high ICANS risk. In the HER2-CNS metastasis trial, adding lymphodepletion increased toxicity (grade 3 headache as DLT) without improving response durability.

References

1. Palazzo et al. 2024. CAR-T Cells for the Treatment of Central Nervous System Tumours: Known and Emerging Neurotoxicities. Brain Sci. 14:1220. https://doi.org/10.3390/brainsci14121220

2. Gatto et al. 2023. CAR-T cells neurotoxicity from consolidated practice in hematological malignancies to fledgling experience in CNS tumors: fill the gap. Front. Oncol. 13:1206983. https://doi.org/10.3389/fonc.2023.1206983

3. Genoud & Migliorini. 2023. Novel pathophysiological insights into CAR-T cell associated neurotoxicity. Front Neurol. 14:1108297. https://doi.org/10.3389/fneur.2023.1108297

4. Majzner et al. 2022. GD2-CAR T cell therapy for H3K27M-mutated diffuse midline gliomas. Nature 603, 934–941. https://doi.org/10.1038/s41586-022-04489-4

5. Monje et al. 2025. Intravenous and intracranial GD2-CAR T cells for H3K27M+ diffuse midline gliomas. Nature 637, 708–715. https://doi.org/10.1038/s41586-024-08171-9

6. Lin et al. 2021. First-in-Human Trial of EphA2-Redirected CAR-T cells in Patients With Recurrent Glioblastoma: A Preliminary Report of Three Cases at the Starting Dose. Front Oncol. 11:694941. https://doi.org/10.3389/fonc.2021.694941

7. Bouwer, Jager & Liesting. 2020. Cardiac monitoring in HER2-positive patients on trastuzumab treatment: A review and implications for clinical practice. The Breast 52: 33-44. https://doi.org/10.1016/j.breast.2020.04.005

8. Seth et al. 2025. Cardiotoxic Effects of Antibody Drug Conjugates vs Standard Chemotherapy in ERBB2-Positive Advanced Breast Cancer - A Systematic Review and Meta-Analysis. JAMA Netw Open 8: e2540336. https://doi.org/10.1001/jamanetworkopen.2025.40336

9. Hegde et al. 2024. Autologous HER2-specific CAR-T cells after lymphodepletion for advanced sarcoma – A Phase I trial. Nat Cancer. 5: 880-894. https://doi.org/10.1038/s43018-024-00749-6

10. Ahmed et al. 2017. HER2-Specific Chimeric Antigen Receptor-Modified Virus-Specific T Cells for Progressive Glioblastoma: A Phase 1 Dose-Escalation Trial. JAMA Oncol. 3: 1094-1101. https://doi.org/10.1001/jamaoncol.2017.0184

11. Priceman et al. 2018. Clin Cancer Res. 24: 95-105. https://doi.org/10.1158/1078-0432.CCR-17-2041

12. Portnow et al. 2025. A phase 1 study of intraventricularly administered autologous her2-targeting chimeric antigen receptor t cells (her2-car t cells), with and without lymphodepletion, in her2-positive breast cancer patients with recurrent brain and/or leptomeningeal metastases. Neuro-Oncology 27 Supplement 5: v115.  https://doi.org/10.1093/neuonc/noaf201.0464

13. O'Rourke et al. 2017. A single dose of peripherally infused EGFRvIII-directed CAR T cells mediates antigen loss and induces adaptive resistance in patients with recurrent glioblastoma. Sci Transl Med 9: eaaa0984. https://doi.org/10.1126/scitranslmed.aaa0984

14. Durgin et al. 2021. Case Report: Prolonged Survival Following EGFRvIII CAR T Cell Treatment for Recurrent Glioblastoma. Front Oncol 11: 669071. https://doi.org/10.3389/fonc.2021.669071

15. Bagley et al. 2023. Phase I study of repeated peripheral infusions of anti-EGFRvIII CAR T cells in combination with pembrolizumab in patients with newly diagnosed, MGMT-unmethylated glioblastoma. Research Square. https://doi.org/10.21203/rs.3.rs-2742648/v1

16. Goff et al. 2019. Pilot Trial of Adoptive Transfer of Chimeric Antigen Receptor-transduced T Cells Targeting EGFRvIII in Patients With Glioblastoma. J Immunother 42: 126-135. https://doi.org/10.1097/CJI.0000000000000260

17. Brown et al. 2024. Locoregional delivery of IL-13Rα2-targeting CAR-T cells in recurrent high-grade glioma: a phase 1 trial. Nat Medicine 30: 1001-1012. https://doi.org/10.1038/s41591-024-02875-1 

18. Bagley et al. 2025. Intracerebroventricular bivalent CAR T cells targeting EGFR and IL-13Rα2 in recurrent glioblastoma: a phase 1 trial. Nat Med. 31: 2778-2787. https://doi.org10.1038/s41591-025-03745-0

19. Vitanza et al. 2023. Intraventricular B7-H3 CAR T Cells for Diffuse Intrinsic Pontine Glioma: Preliminary First-in-Human Bioactivity and Safety. Cancer Discov. 13: 114-131. https://doi.org/10.1158/2159-8290.CD-22-0750

20. Vitanza et al. 2025. Intracerebroventricular B7-H3-targeting CAR T cells for diffuse intrinsic pontine glioma: a phase 1 trial. Nat Med. 31: 861–868. https://doi.org/10.1038/s41591-024-03451-3

21. Zhang et al. 2024. Safety and efficacy of B7-H3 targeting CAR-T cell therapy for patients with recurrent GBM. J Clin Oncol. 42(suppl): 2062.

22. Rauf Y et al. 2025. Safety and tolerability of intraventricular infusion of B7-H3.CAR-T cells in recurrent glioblastoma. Neuro-Oncology 27, Supplement 5: v121.  https://doi.org/10.1093/neuonc/noaf201.0489

23. Liu et al. 2023. Safety and antitumor activity of GD2-Specific 4SCAR-T cells in patients with glioblastoma. Mol Cancer. 22: 3. https://doi.org/10.1186/s12943-022-01711-9

24. Majzner et al. 2022. GD2-CAR T cell therapy for H3K27M-mutated diffuse midline gliomas. Nature 603: 934–941. https://doi.org/10.1038/s41586-022-04489-4

25. Monje et al. 2025. Intravenous and intracranial GD2-CAR T cells for H3K27M+ diffuse midline gliomas. Nature 637: 708–715. https://doi.org/10.1038/s41586-024-08171-9