Glioblastoma is a lethal brain tumor that is unresponsive to current cancer immunotherapeutic approaches, including immune checkpoint blockade (ICB). This suggests that initial priming of T cells, rather than their expansion and licensing as effectors, is a restricting feature in this tumor setting. To overcome the limited initiation of CD8 + T cell responses, we employed a strong heterologous prime-boost vaccination with the simian adenovirus ChAdOx1 and poxvirus modified vaccinia Ankara (MVA). Vaccination conferred therapeutic efficacy against orthotopic, immune checkpoint-blockade (ICB)-refractory SB28 murine glioblastoma. Vaccination was effective against both the murine tumor antigen, P1A, and a newly identified glioblastoma-associated antigen, Gpr149. Additional treatment with ICB provided no additional benefit. Systemic ChAdOx1/MVA vaccination induced robust infiltration of antigen-specific T cells in tumor-challenged brains, the majority of which exhibited a CD103 + CD69 + CD8 + tissue-resident memory (TRM)-like phenotype. These cells were polyfunctional, durable in brains with sustained tumor control, and mediated tissue-specific immunological memory. Moreover, intracranial adoptive transfer of glioblastoma-derived antigen-specific TRM-like cells was sufficient to protect naïve recipients from subsequent orthotopic tumor challenge. Together, these findings establish that viral vector vaccination can generate tumor-specific TRM-like cells that mediate effective anti-glioblastoma immunity, providing a rationale for clinical evaluation of ChAdOx1/MVA-based strategies in glioblastoma.

Background:Immunotherapy, while effective for many extra-cranial solid tumors, has not shown benefits against glioblastoma. Limited T cell infiltration and an immunosuppressive tumor microenvironment (TME) are among the main barriers to successful immunotherapy. We hypothesized that blocking the chemokine CXCL12 would modulate the TME composition and reverse GBM resistance to immunotherapy. Methods:We used a syngeneic murine SB28 GBM model resistant to immune checkpoint inhibitor (ICI) and compared orthotopic and subcutaneous tumors to differentiate tumor-intrinsic factors from CNS-related barriers. We treated SB28 tumor-bearing mice with vehicle, ICI, CXCL12 inhibitor (NOX-A12), or a combination of NOX-A12 with ICI. We assessed treatment effects on immune cell populations in the blood and the TME, and on tumor growth and mouse survival. Results:ICI alone increased effector CD8+T cells in subcutaneous tumors but did not alter immune subsets in intracranial tumors. The combination of NOX-A12 and ICI increased effector CD8+ and CD4+ T cells in both models. In subcutaneous tumors, combination treatment also reduced MHC-IIlow tumor-associated macrophages (TAM) and slowed tumor growth. However, in orthotopic tumors, TAM populations remained unaffected, and survival was not extended, despite similar T-cell modulation, supporting the role of TAM in mediating GBM resistance to ICI. Conclusion:CXCL12 inhibition demonstrates therapeutic potential in facilitating anti-tumor immune response by dual mechanism, including T cell expansion and TAM reduction. The inability to deplete TAMs and improve survival in intracranial GBM underscores the need to address brain-specific mechanisms underlying TAM persistence to advance immunotherapy in GBM.

Abstract Glioblastoma is a lethal brain tumor notable for limited spontaneous induction of CD8+ T cell responses. While viral-vector vaccines can drive particularly high magnitudes of tumor-reactive T cells, they have not been investigated for the treatment of glioblastoma. Here, we demonstrate that heterologous prime-boost vaccination with the simian adenovirus ChAdOx1 and poxvirus modified vaccinia Ankara (MVA) treats the orthotopic, syngeneic, checkpoint-inhibitor refractory SB28 murine model of glioblastoma, both in the context of the murine tumor antigen, P1A, and a newly identified tumor-associated antigen expressed by SB28. Adjuvant anti-PD-1 and anti-CTLA-4 did not further improve outcomes. Vaccination induced immunoediting of tumor antigen expression and tumor-specific recruitment of antigen-specific T cells to challenged brains, the majority of which had a CD103+CD69+CD8+ tissue resident memory (TRM)-like phenotype. While induction of TRMs by vaccination has been implicated in superior control of other cancers, their role in mediating anti-glioblastoma immunity was unclear. ChAdOx1/MVA-induced brain TRMs displayed superior polyfunctionality compared to circulating and non-TRM brain antigen-specific CD8+ T cells. Long-term surviving mice maintained consistent levels of antigen-specific TRM cells in their brains at days 70 and 175 post-tumor challenge, despite a three-fold decrease in circulating antigen-specific T cells, indicating durable T cell memory in the brain parenchyma. Furthermore, survivors were protected against a second orthotopic, but not subcutaneous, tumor rechallenge, demonstrating tissue-specific immunological memory. Intracranial adoptive transfer of brain-derived antigen-specific TRMs isolated from vaccinated, tumor-bearing mice was sufficient to prolong the survival of naïve mice challenged with tumors, whereas blood-derived antigen-specific non-TRMs and non-antigen-specific brain-derived CD8+ T cells were not. Despite peripheral administration of the vaccines, antigen-specific CD8+ T cells were induced not only in the brains of tumor-free animals, but in many tissues, including the liver, lung, skin, skull bone marrow, and meninges, with distinct phenotypes observed in different tissues. Overall, we demonstrate that ChAdOx1/MVA vaccination is a potent strategy to induce TRMs to mediate anti-glioblastoma immunity, establishing a basis for further clinical investigation of ChAdOx1/MVA vaccination to treat patients with glioblastoma. Citation Format: Emily Elizabeth Steffke, Laila Latifi, Taijun Hana, Ayaka Hara, Morgan Coombs, Jo Spurgeon, Caitlin Huguely, John Hancock, Brita Anderson, James McAuliffe, Vinnycius Pereira-Almeida, Amanda Wicki, Sara Abdel Malak, Laurine Noblecourt, Meili Zhang, Wei Zhang, Dionne Davis, Nicole Briceno, Hua Song, Chen Cam-El Makranz, Hideho Okada, Mark Gilbert, Carol Leung, Benoit Van den Eynde, Masaki Terabe. Viral vector vaccination induces brain resident memory T cells to drive anti-glioblastoma immunity [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 4363.

BACKGROUND:Triple negative breast cancer (require) new treatment strategies due to poor responses to current therapies. While myeloid SIRPα mediates immunosuppression, its cancer intrinsic role remains poorly understood. METHODS:Human breast cancer scRNAseq profiles were used to examine SIRPα expression across different cell populations and subtypes. TNBC brain-tropic cells were injected into the mouse mammary fat pad for the orthotopic tumor model, and intracardiac-injected for brain metastasis models. Bulk RNA sequencing was used to determine SIRPα-regulated pathway. Stably SIRPα overexpressed and knockout TNBC cell lines were established to determine SIRPα intracellular regulation. Digital spatial profiling was utilized to investigate the orthotopic and brain metastasis tumor immune microenvironment. RESULTS:Human single-cell data showed that SIRPα levels increased in malignant TNBC epithelial cells. We observed that SIRPα is upregulated in patient breast-to-brain metastatic lesions. SIRPα is overexpressed in TNBC brain-tropic cells compared to parental cells. Bulk RNA-Seq showed that targeting SIRPα affects genes involved in mitochondrial dynamics, and that SIRPα upregulates mitochondrial fission and induces metastasis through the SHP2/Erk/Drp1 signaling pathway. In vivo, overexpression of SIRPα in cancer cells significantly increases TNBC systemic metastasis. Next, spatial proteomics revealed changes in the immune microenvironment associated with the SIRPα-regulated ECM protein fibronectin. Fibronectin induces microglial tolerance by impairing inflammatory signaling and metabolic reprogramming, allowing cancer to escape microglial immunosurveillance. Most importantly, SIRPα inhibition reduced TNBC brain metastatic lesions in mouse metastasis models. CONCLUSION:: Cancer-intrinsic SIRPα promotes TNBC brain metastasis through increased mitochondria fission and triggering microglia tolerance, and targeting SIRPα reduces brain metastasis.

The aggressiveness of glioblastoma may be due to its limited induction of anti-tumor T cells; therefore, vaccines are attractive candidates to drive tumor-specific T cells. Heterologous prime-boost vaccination with chimpanzee adenovirus (ChAdOx1) and modified vaccinia Ankara (MVA) viral vectors can generate high tumor-targeted CD8+ T cell frequencies. Therefore, we tested whether this vaccination strategy induces effective T cell responses against murine glioblastoma. We used the orthotopic syngeneic SB28.P1A glioblastoma model expressing P1A, a mouse tumor antigen. Both prophylactic and therapeutic vaccination with ChAdOx1/MVA-P1A significantly prolonged the survival of mice with SB28.P1A tumors. Tumors were enriched for P1A-specific CD8+ T cells, with 40% of CD8+ T cells P1A-specific in the tumor-bearing hemisphere, compared to 20% in the contralateral hemisphere and 5% in the blood. Strikingly, 70% of tumor-infiltrating P1A-specific CD8+ T cells were resident memory T (Trm) cells. The adoptive transfer of sorted P1A-specific Trm cells derived from vaccinated tumor-bearing mice into the brains of naïve mice sufficiently protected mice from later challenge with tumors, while P1A-specific non-Trm cells or non-P1A-specific brain-derived CD8+ T cells did not. The capability of ChAdOx1/MVA vaccination to induce antigen-specific Trms in the brain that are sufficient to protect against glioblastoma is promising for translating this therapy into clinical trials. Supported by NIH intramural research program ZIA BC011877 and the Ludwig Institue for Cancer Research. Vaccines and Immunotherapy (VAC)

Background Glioblastoma (GBM) is the most aggressive primary brain cancer in adults and remains incurable. Mucosal-associated invariant T (MAIT) cells are unconventional T cells with a semi-invariant T cell receptor and have been shown to regulate immune responses. However, the role of MAIT cells in glioblastoma (GBM) has not been well characterized. Methods We used flow cytometry, bulk RNA-seq and scRNA-seq, and multiplexed tissue imaging to investigate the role of MAIT cells in GBM. Results Flow cytometry analysis of peripheral blood samples of GBM patients showed a significant reduction of MAIT cell frequency and the ability to produce Th1 cytokines. In bulk RNA sequencing data analysis of GBM tissues, the MAIT cell gene signature significantly correlated with poor patient survival. A scRNA-seq of CD45+ cells from 20 GBM tissue samples showed 12 (60%) were positive for MAIT cells and the enrichment of RORC-expressing MAIT17. The MAIT cell signature significantly correlated with tumor-associated neutrophil (TAN) activities. Multiple immune suppressive genes known to be used by TANs were upregulated in MAIT-positive tumors. Spatial imaging analysis of GBM tissues showed that all specimens were positive for both MAIT cells and TANs, and localized enrichment of TANs. Conclusion These findings revealed an immunosuppressive role of MAIT cells in GBM and highlighted the MAIT-TAN axis as a potential novel therapeutic target to modulate GBM’s immunosuppressive tumor microenvironment.

Background:While chimeric antigen receptor (CAR) T-cells are promising, there is a rapidly growing interest in developing other CAR-expressing immune cells. However, to date, no reported studies evaluated these cells side-by-side in immune-competent glioma models. Methods:We developed a novel C57BL/6-background transgenic mouse strain with all hematopoietic cells carrying the anti-epidermal growth factor receptor (EGFR)vIII-CAR downstream of a Lox-Stop-Lox cassette in the Rosa26 locus. Crossing with mice transgenic for Vav-Cre allowed the expression of anti-EGFRvIII CAR in all hematopoietic cells. In particular, we evaluated CAR-T, CAR-NKT, CAR-NK-cells, and CAR-macrophages in a syngeneic mouse SB28EGFRVIII glioma model. Results:CAR-NK and CAR-NKT-cells demonstrated anti-tumor effects comparable to CAR-T cells in vitro. A single intratumoral administration of CAR-T and CAR-NKT cells in combination mediated superior therapeutic efficacy compared to CAR-T cells or CAR-NKT-cells alone. A single intravenous infusion of CAR-NK cells following lymphodepletion failed to mediate significant anti-glioma effects. Additionally, intratumoral injection of CAR-NK cells did not confer therapeutic benefit. Contrary to previous reports using human macrophages, CAR-macrophages did not demonstrate enhanced antigen-presentation activity against glioma cells compared to non-CAR macrophages. Intratumorally administered CAR-macrophages failed to demonstrate local persistence or anti-tumor effects in vivo. Conclusions:These data provide a valuable basis as to which immune cells can mediate effective anti-glioma response in an immuno-competent glioma environment. Our data also suggest that a combination of CAR-T and CAR-NKT-cells may represent a promising therapeutic strategy.

BACKGROUND:To develop effective therapies for glioblastoma (GBM), a deeper understanding of its underlying immunoregulatory mechanisms is needed. Invariant natural killer T (iNKT) cells are unconventional T cells that recognize lipid antigens and are known to regulate tumor immunity in other cancer types. Given the lipid-rich nature of the brain and the unique metabolic activity of GBM cells, we hypothesized that GBM-enriched lipids could direct iNKT cells to contribute to the immunosuppressive nature of the disease. METHODS:Lipid levels of multiple human GBM stem-like cell (GSC) lines, low grade-glioma lines, and normal human astrocytes were determined using liquid-chromatography mass spectrometry. GBM stem-like cell-enriched lipids were tested in iNKT stimulation assays, with either human iNKT cell lines or PBMC samples from both healthy donors and GBM patients, to determine antigenicity and characterize the nature of iNKT activation. RESULTS:Multiple lipid species were found to be uniquely enriched in GSCs. Many of these lipids, called sulfatides, were recognized by and activated iNKT cells in a dose-dependent manner when presented by CD1d. Pharmaceutical and genetic targeting of the sulfatide synthetic pathway within GSCs resulted in an altered ability to activate iNKT cells. However, one lipid, lyso-sulfatide, inhibited the activation of iNKT cells and suppressed activation induced by a cognate antigen, α-galactosylceramide. CONCLUSIONS:The modulation of iNKT cell functions by GSC-enriched glycosphingolipids may contribute to the immunosuppression of GBM and highlights sulfatide production as a potential therapeutic target for GBM treatment.

Background:Glioblastoma (GBM) prognosis remains poor, and although immune checkpoint inhibitors (ICI) have transformed the treatment of many tumors, they are ineffective in GBM. However, response to ICIs occurs in high-tumor-mutational-burden (TMB) GBMs. To address the immunological impact of high TMB in GBM, we created a high TMB syngeneic mouse model from the low TMB SB28 GBM cell line. Methods:We used CRISPR-Cas9 to target murine Msh2, Mlh1, CXCL10, and CCL5. Single-cell-sorted clones were characterized by whole exome, bulk RNA-sequencing, and neoantigen prediction. Clones were injected subcutaneously or intracranially with or without anti-PD-1/anti-CTLA4 and dexamethasone. Results:Loss of mismatch repair (MMR) proteins Msh2 or Mlh1 increased nonsynonymous mutations. A fraction of mice with intracranial Msh2KO but not Mlh1KO SB28 showed long-term survival with anti-PD-1/anti-CTLA4 treatment plus dexamethasone. Long-term surviving mice from Msh2 KO SB28 rejected rechallenged subcutaneous tumors. Subcutaneous tumors from clones with increased TMB grew more slowly. This was fully abrogated in Rag1 null mice for Msh2KO but only partially for Mlh1KO SB28. Hypermutant Msh2KO clones spontaneously secreted CXCL10, CCL5, and increased pro-inflammatory chemokines after IFN-γ stimulation. Knockout of CXCL10 or CCL5 in the highest TMB Msh2KO clone restored flank tumor growth, indicating loss of immune response despite elevated TMB. Conclusion:Mismatch repair-deficient SB28 tumors were more immunogenic, but this was not completely correlated with TMB. Rather, rejection depended on increased secretion of pro-inflammatory chemokines. Msh2 and Mlh1 loss was not equivalent, suggesting that additional studies are needed to elucidate germline and somatic mismatch repair gene-specific immune alterations.

Immunotherapy, while effective for many solid tumors, has not shown benefits against glioblastoma (GBM). This is postulated to result from limited T cell infiltration and an immunosuppressive tumor microenvironment (TME). We hypothesized that the brain tissue environment, distinct from extracranial tissues, contributes to treatment resistance. To test this hypothesis, we established a mouse subcutaneous (s.c.) SB28 GBM model, representing an extracranial tissue environment, and compared it to orthotopic (intracranial) SB28. Treatment with Immune check-point blockade (ICB) increased overall survival in s.c. SB28, but not in intracranial (i.c.) SB28 model. We then analyzed ICB effects on the immune landscape in the circulation and TME in each model, using flow cytometry. ICB increased circulating CD8 and CD4 T cells subsets in both i.c. and s.c. tumor models compared to a vehicle control group. Yet, antigen-experienced CD8 and CD4 T cells increased only in the TME of the s.c. model. To enhance T cell migration into i.c. tumors, we targeted CXCL12, leveraging its synergistic effect with ICB on T cell infiltration observed in other extracranial tumors. We compared treatment-induced changes in the immune landscape in mice receiving CXCL12 inhibitor NOX-A12, ICB, the combination of NOX-A12 and ICB, or vehicle control. As expected, combination therapy significantly increased antigen-experienced CD8 T cells in i.c. and s.c. SB28 compared to ICB or NOX-A12 monotherapy. Combination treatment also increased antigen-experienced CD4 T cells in i.c. tumors, but not in s.c. tumors, indicating a tissue-dependent treatment effect. These treatment effects were not evident in the blood, indicating a local mechanism of action within the TME. However, the increase in intra-tumoral T-cells induced by combination treatment was not translated into tumor growth reduction, implying that additional factors contribute to GBM resistance to immunotherapy. Elucidating those factors is imperative for the development of innovative strategies to augment immunotherapy efficacy in GBM.

Recent studies determined that in glioblastoma, tumor cells integrate into neural networks, forming synapses with neurons, enhancing neuronal firing facilitating tumor growth. We sought to understand how neuronal activity alters the function of immune cells in glioblastoma using both murine models and freshly acquired human glioblastoma tissue from surgical specimens. We first used chemogenetic stimulation (AAV5-SYN1-hM3Dq) of an RCAS-tva mouse glioblastoma model (Ntv-a Ink4a-Arf-/-) to identify which tumor immune populations are impacted by enhanced neuronal firing in vivo. Using high-dimensional spectral flow cytometry, we found an increase in microglia following stimulation. We then used human glioblastoma tissue to study a possible neuron-myeloid interaction. Immunofluorescent staining of these tumors showed that approximately 50% of the Iba1+ myeloid cells colocalized with the presynaptic neuronal marker synapsin 1, confirming the neuron-myeloid interaction. We then sorted CD11b+ myeloid cells from these patient samples. Bulk RNA-seq revealed high expression of neurotransmitter receptors, predominantly the Gi protein-coupled P2Y12 receptor for ADP. Calcium fluorescence imaging, calcium flow cytometry, and a phagocytic assay showed that the cells were primarily responsive to ADP among the neurotransmitters examined, including glutamate, norepinephrine, GABA, and acetylcholine. Analysis of bulk RNA-seq from 286 brain tumor patients found that P2Y12 had a higher positive correlation with M1-like markers and a negative correlation with M2-like markers than all other neurotransmitter receptor genes and others such as ADRB2, CX3CR1, TMEM119, TREM2, and CSF1R. To understand how the ADP-P2Y12 axis affects myeloid phenotypic function in vivo, we conducted flow cytometry on our SB28 mouse brain tumor tissues. We found that P2Y12-positive microglia had an increased M1-like and decreased M2-like phenotype compared to P2Y12-negative microglia, based on functional marker expression. Overall, we conclude that neuronal excitation affects microglia in the glioblastoma microenvironment, primarily through an ADP-P2Y12 axis, which could be an important target for enhancing response to immunotherapy.

In a structure-function study of sulfatides that typically stimulate type II NKT cells, we made an unexpected discovery. We compared analogs with sphingosine or phytosphingosine chains and 24-carbon acyl chains with 0-1-2 double bonds (C or pC24:0, 24:1, or 24:2). C24:1 and C24:2 sulfatide presented by the CD1d monomer on plastic stimulated type II, not type I, NKT cell hybridomas, as expected. Unexpectedly, when presented by bone marrow-derived DCs (BMDCs), C24:2 reversed specificity to stimulate type I, not type II, NKT cell hybridomas, mimicking the corresponding beta-galactosylceramide (beta GalCer) without sulfate. C24:2 induced IFN-gamma-dependent immunoprotection against CT26 colon cancer lung metastases, skewed the cytokine profile, and activated conventional DC subset 1 cells (cDC1s). This was abrogated by blocking lysosomal processing with bafilomycin A1, or by sulfite blocking of arylsulfatase or deletion of this enyzme that cleaves off sulfate. Thus, C24:2 was unexpectedly processed in BMDCs from a type II to a type I NKT cell-stimulating ligand, promoting tumor immunity. We believe this is the first discovery showing that antigen processing of glycosylceramides alters the specificity for the target cell, reversing the glycolipid's function from stimulating type II NKT cells to stimulating type I NKT cells, thereby introducing protective functional activity in cancer. We also believe our study uncovers a new role for antigen processing that does not involve MHC loading but rather alteration of which type of cell is responding

Abstract Glioblastoma is an aggressive primary brain tumor highly resistant to currently available immunotherapies. A deeper understanding of its underlying immunoregulatory mechanisms is paramount to developing future immunotherapeutic treatments. Invariant natural killer T (iNKT) cells are unconventional T cells that recognize lipid antigens presented by an MHC-like molecule called CD1d. Although iNKT cells have been shown to regulate tumor immunity in other cancer types, their role in glioblastoma is not well characterized. Given the lipid-rich nature of the brain and the unique metabolic activity of glioblastoma cells, we hypothesized that interactions between glioblastoma and iNKT cells through glioblastoma-produced lipids could contribute to the immunosuppressive nature of the disease. We report multiple lipid species enriched in aggressive human glioblastoma stem-like cell (GSC) lines that activate human iNKT cells and modulate their functions. Lipidomic LC-MS analysis of GSCs, low-grade glioma stem-like cell lines, and normal human astrocytes revealed various glycosphingolipid species, including sulfatides, highly enriched in the GSCs. Multiple enriched sulfatide species, when presented by CD1d, were recognized by and activated iNKT cells in a dose-dependent manner. Additionally, cytokine analysis of stimulated human PBMC-derived iNKT cells demonstrated that the enriched sulfatides did not induce Th1 cytokine production, but rather many non-Th1 cytokines that potentially counteract Th1-type immune responses. This modulation of iNKT cell function by glioblastoma-enriched glycosphingolipids may contribute to the immunosuppression of glioblastoma and could highlight sulfatide production as a potential therapeutic target for glioblastoma treatment.

Background Glioblastomas are immunologically 'cold' tumors with sparse cytotoxic T-cell infiltration. Therefore, viral vector vaccines may be a promising approach to boost the induction of glioblastoma-targeted T cells. It has been previously shown that heterologous prime-boost vaccination with chimpanzee-derived adenovirus ChAdOx1 and modified vaccinia Ankara (MVA) vectors can induce a high magnitude of CD8+ T cells specific for cancer-associated antigens and have therapeutic effects against mouse models of cancer.1 Therefore, we aimed to evaluate whether treating mice with ChAdOx1 and MVA vaccines targeting model tumor antigens and endogenous antigens could be beneficial in the prophylactic and therapeutic settings against syngeneic, intracranial murine glioblastoma. Methods Murine glioblastoma cell lines were developed to express model tumour antigens. In addition, endogenous tumor-associated antigens and neoantigens were identified in murine glioblastoma lines using our novel Mouse nEoanTigen pRedictOr (METRO) antigen discovery pipeline. We then created ChAdOx1 and MVA vectors expressing model tumor antigens or endogenous antigen candidates, and we confirmed their immunogenicity via intracellular cytokine staining of ex vivo stimulated peripheral blood mononuclear cells or splenocytes. In the prophylactic tumor setting, mice were vaccinated with ChAdOx1 and MVA vectors expressing endogenous antigens or a model tumor antigen, then challenged with syngeneic intracranial wild-type or model antigen expressing tumors, respectively. In the therapeutic setting, mice bearing intracranial tumors were treated with vaccines in combination with checkpoint inhibitors after confirming tumor formation. Results ChAdOx1 and MVA heterologous prime-boost vaccination generated a high magnitude of antigen-specific CD8+ T cells against a model tumor antigen. Furthermore, we confirmed the immunogenicity of some of the antigens identified by the METRO pipeline. Prophylactic vaccination targeting a model tumor antigen significantly increased the survival time of mice bearing intracranial tumors engineered to express the same antigen. Ongoing studies are investigating the efficacy of this vaccination strategy in the therapeutic setting. Conclusions Our heterologous prime-boost strategy generates a high magnitude of antigen-specific CD8+ T cells that provide protection against the development of orthotopic glioblastoma tumors in mice. It remains to be seen whether these vaccines can provide a therapeutic benefit to mice bearing intracranial tumors, and the effects of vaccination on the remodeling of the tumor microenvironment and tumor-draining lymph nodes are yet to be determined. Preclinical data generated using our vaccine and tumor models may provide proof-of-concept to move these vaccines into clinical trials to treat patients with glioblastoma. Reference 1. McAuliffe J, Chan HF, Noblecourt L, Ramirez-Valdez RA, Pereira-Almeida V, Zhou Y, Pollock E, Cappuccini F, Redchenko I, Hill AV, Leung CSK, Van den Eynde BJ. Heterologous prime-boost vaccination targeting MAGE-type antigens promotes tumor T-cell infiltration and improves checkpoint blockade therapy. J Immunother Cancer . 2021 Sep; 9 (9):e003218. Ethics Approval All animal work was approved by either the University of Oxford Animal Care and Ethical Review Committee and experimental procedures were carried out in accordance with the terms of the UK Animals (Scientific Procedures) Act Project Licenses P0D369534 and PB050649E; or by the National Cancer Institute-Bethesda Animal Care and Use Committee and experimental procedures were carried out in accordance with the terms of Protocol NOB-024.