Radiation therapy synergizes with mRNA vaccination to overcome microglia-mediated suppression of dendritic cell migration and T cell priming in glioblastoma

Keywords

● Radiation therapy

● mRNA cancer vaccines

● conventional dendritic cells

● Glioblastoma

Main Findings

Glioblastoma (GBM) is the most common primary malignant brain tumor and is associated with poor survival outcomes, likely in part because of its highly immunosuppressive tumor microenvironment, which limits responses to many immunotherapies. Improving patient outcomes will therefore require a better understanding of the mechanisms underlying immune dysfunction in GBM and how these mechanisms can be therapeutically overcome.

Dendritic cells (DCs) are key antigen-presenting cells that coordinate systemic immune responses through their specialized roles in antigen acquisition, migration, and T cell priming. In a 2026 preprint by Wisdom et al. (not peer reviewed), the authors investigate DC dysfunction in GBM, hypothesizing that impaired DC function limits effective anti-tumor immunity and therapeutic responses to immunotherapy. To study this, they used the CT2A GBM cell line engineered to express the model antigen ovalbumin (OVA) and the pH-resistant fluorescent reporter ZsGreen. Using this system, the authors identified distinct differences between intracranial GBM tumors and extracranial flank tumors, most notably the accumulation of type 2 conventional dendritic cells (cDC2s) within brain tumors but not the healthy brain. Despite their accumulation, GBM-associated cDC2s exhibited poor antigen-presenting capacity and were dispensable for anti-tumor immunity, in contrast to their type 1 conventional dendritic cell (cDC1) counterparts.

Consistent with previous findings, the authors identified the deep cervical lymph node, which drains the brain tumor, as an important site of initial T cell priming. Using photoconversion experiments, they further demonstrated that DCs within intracranial GBM tumors exhibit impaired migratory capacity compared with DCs in flank tumors. These findings identify a GBM-specific defect in DC migration that may constrain effective T cell-mediated anti-tumor immunity.

To determine the mechanisms driving this dysfunction, the authors used CellChat to analyze ligand-receptor interactions from single-cell RNA-sequencing data. This analysis identified microglia-associated GAS6–AXL signaling as a candidate suppressive pathway limiting DC function. In vivo pharmacologic depletion of microglia using a CSF1R inhibitor, as well as pharmacologic inhibition of AXL, increased MHC expression and expression of the migration-associated receptor CCR7 on DCs, supporting a role for microglia-derived AXL signaling in suppressing DC function.

The authors next investigated whether cancer therapies could overcome DC dysfunction in GBM by testing radiation therapy (RT) and mRNA cancer vaccination. RT alone prolonged survival in a manner dependent on cDC1s and T cells, and DCs isolated from irradiated tumors exhibited enhanced cross-presentation ex vivo.

Combining RT with an mRNA cancer vaccine resulted in tumor regression, expansion of progenitor-exhausted CD8 T cells, and increased CD8 T cell cytotoxicity. In addition to the implantable CT2A model, the authors developed an autochthonous GBM model expressing the OVA model antigen. This model is an important finding as autochthonous tumors better recapitulate early tumor–immune interactions that are absent from implantable models. In this setting, mRNA cancer vaccination improved animal survival; however, the addition of RT did not significantly enhance therapeutic efficacy.

Limitations

Although this study identifies an important mechanism of DC dysfunction in GBM, several limitations remain.

· The authors acknowledge one limitation is the over-simplification of the ovalbumin model antigen. In humans, tumor neoantigens used for mRNA cancer vaccines often do not resemble ovalbumin as they are less abundant, processed by antigen presenting machinery at lower efficiency, and exhibit poorer MHC binding capabilities. These differences limit T cell activity and therapeutic responses in patients. Future work will need to understand if these findings hold true for personalized cancer vaccines.

· Further work is needed to determine how RT and mRNA vaccination affect the specific DC defects identified earlier in the study. Figures 1-4 establish a mechanistic framework in which impaired DC migration limits anti-tumor immunity in GBM. However, Figures 5-7 provide relatively limited mechanistic evidence showing whether RT or mRNA vaccination directly restores DC migration, enhances T cell priming within the draining lymph node, or disrupts microglia-derived GAS6-AXL signaling. Connecting these therapeutic effects to the mechanism identified in the first half of the study would improve the flow of the paper and complete the full story.

· It remains unclear why RT and mRNA cancer vaccination fail to synergize in the autochthonous GBM model despite showing greater efficacy in the implantable model. Understanding the biological basis for this discrepancy would be important, particularly because the autochthonous model may more closely reproduce the immune evolution and microenvironment of naturally arising GBM.

Significance/Novelty

This study is significant because it identifies impaired DC migration as an important mechanism of immune evasion in GBM and links this dysfunction, at least in part, to microglia-associated GAS6–AXL signaling. The findings provide a mechanistic explanation for how the brain tumor microenvironment may restrict the initiation of systemic anti-tumor T cell responses.

The study also provides important preclinical evidence that RT and mRNA cancer vaccination can enhance anti-tumor immunity and improve survival in GBM. Together, these findings highlight DC dysfunction as both a biological barrier to effective immunity and a potentially actionable therapeutic target in GBM.

Preprint rating

· Scientific quality 5/5

· Novelty 4/5

· Significance 5/5

Credit

Reviewed by Annette Wu and Dr. Benjamin R. Schrank as part of a cross-institutional journal club between the Icahn School of Medicine at Mount Sinai, the University of Oxford, the Karolinska Institute, the University of Toronto, and the UT MD Anderson James P. Allison Institute.

The author declares no conflict of interests in relation to their involvement in the review.