Predictive Model to Identify the Long Time Survivor in Patients with Glioblastoma: A Cohort Study Integrating Machine Learning Algorithms.

We aimed to develop and validate a predictive model for identifying long-term survivors (LTS) among glioblastoma (GB) patients, defined as those with an overall survival (OS) of more than 3 years. A total of 293 GB patients from CGGA and 169 from TCGA database were assigned to training and validation cohort, respectively. The differences in expression of immune checkpoint genes (ICGs) and immune infiltration landscape were compared between LTS and short time survivor (STS) (OS<1.5 years). The differentially expressed genes (DEGs) and weighted gene co-expression network analysis (WGCNA) were used to identify the genes differentially expressed between LTS and STS. Three different machine learning algorithms were employed to select the predictive genes from the overlapping region of DEGs and WGCNA to construct the nomogram. The comparison between LTS and STS revealed that STS exhibited an immune-resistant status, with higher expression of ICGs (P<0.05) and greater infiltration of immune suppression cells compared to LTS (P<0.05). Four genes, namely, OSMR, FMOD, CXCL14, and TIMP1, were identified and incorporated into the nomogram, which possessed good potential in predicting LTS probability among GB patients both in the training (C-index, 0.791; 0.772-0.817) and validation cohort (C-index, 0.770; 0.751-0.806). STS was found to be more likely to exhibit an immune-cold phenotype. The identified predictive genes were used to construct the nomogram with potential to identify LTS among GB patients.

Compartmentalized ocular lymphatic system mediates eye-brain immunity.

The eye, an anatomical extension of the central nervous system (CNS), exhibits many molecular and cellular parallels to the brain. Emerging research demonstrates that changes in the brain are often reflected in the eye, particularly in the retina1. Still, the possibility of an immunological nexus between the posterior eye and the rest of the CNS tissues remains unexplored. Here, studying immune responses to herpes simplex virus in the brain, we observed that intravitreal immunization protects mice against intracranial viral challenge. This protection extended to bacteria and even tumours, allowing therapeutic immune responses against glioblastoma through intravitreal immunization. We further show that the anterior and posterior compartments of the eye have distinct lymphatic drainage systems, with the latter draining to the deep cervical lymph nodes through lymphatic vasculature in the optic nerve sheath. This posterior lymphatic drainage, like that of meningeal lymphatics, could be modulated by the lymphatic stimulator VEGFC. Conversely, we show that inhibition of lymphatic signalling on the optic nerve could overcome a major limitation in gene therapy by diminishing the immune response to adeno-associated virus and ensuring continued efficacy after multiple doses. These results reveal a shared lymphatic circuit able to mount a unified immune response between the posterior eye and the brain, highlighting an understudied immunological feature of the eye and opening up the potential for new therapeutic strategies in ocular and CNS diseases.

STING activation counters glioblastoma by vascular alteration and immune surveillance.

Glioblastoma (GBM) is an aggressive brain tumor with a median survival of 15 months and has limited treatment options. Immunotherapy with checkpoint inhibitors has shown minimal efficacy in combating GBM, and large clinical trials have failed. New immunotherapy approaches and a deeper understanding of immune surveillance of GBM are needed to advance treatment options for this devastating disease. In this study, we used two preclinical models of GBM: orthotopically delivering either GBM stem cells or employing CRISPR-mediated tumorigenesis by adeno-associated virus, to establish immunologically proficient and non-inflamed tumors, respectively. After tumor development, the innate immune system was activated through long-term STING activation by a pharmacological agonist, which reduced tumor progression and prolonged survival. Recruitment and activation of cytotoxic T-cells were detected in the tumors, and T-cell specificity towards the cancer cells was observed. Interestingly, prolonged STING activation altered the tumor vasculature, inducing hypoxia and activation of VEGFR, as measured by a kinome array and VEGF expression. Combination treatment with anti-PD1 did not provide a synergistic effect, indicating that STING activation alone is sufficient to activate immune surveillance and hinder tumor development through vascular disruption. These results guide future studies to refine innate immune activation as a treatment approach for GBM, in combination with anti-VEGF to impede tumor progression and induce an immunological response against the tumor.

Clinical trial links oncolytic immunoactivation to survival in glioblastoma.

Immunotherapy failures can result from the highly suppressive tumour microenvironment that characterizes aggressive forms of cancer such as recurrent glioblastoma (rGBM)1,2. Here we report the results of a first-in-human phase I trial in 41 patients with rGBM who were injected with CAN-3110-an oncolytic herpes virus (oHSV)3. In contrast to other clinical oHSVs, CAN-3110 retains the viral neurovirulence ICP34.5 gene transcribed by a nestin promoter; nestin is overexpressed in GBM and other invasive tumours, but not in the adult brain or healthy differentiated tissue4. These modifications confer CAN-3110 with preferential tumour replication. No dose-limiting toxicities were encountered. Positive HSV1 serology was significantly associated with both improved survival and clearance of CAN-3110 from injected tumours. Survival after treatment, particularly in individuals seropositive for HSV1, was significantly associated with (1) changes in tumour/PBMC T cell counts and clonal diversity, (2) peripheral expansion/contraction of specific T cell clonotypes; and (3) tumour transcriptomic signatures of immune activation. These results provide human validation that intralesional oHSV treatment enhances anticancer immune responses even in immunosuppressive tumour microenvironments, particularly in individuals with cognate serology to the injected virus. This provides a biological rationale for use of this oncolytic modality in cancers that are otherwise unresponsive to immunotherapy (ClinicalTrials.gov: NCT03152318 ).

Tumor-secreted lactate contributes to an immunosuppressive microenvironment and affects CD8 T-cell infiltration in glioblastoma.

Introduction: Glioblastoma is a malignant brain tumor with poor prognosis. Lactate is the main product of tumor cells, and its secretion may relate to immunocytes' activation. However, its role in glioblastoma is poorly understood. Methods: This work performed bulk RNA-seq analysis and single cell RNA-seq analysis to explore the role of lactate in glioblastoma progression. Over 1400 glioblastoma samples were grouped into different clusters according to their expression and the results were validated with our own data, the xiangya cohort. Immunocytes infiltration analysis, immunogram and the map of immune checkpoint genes' expression were applied to analyze the potential connection between the lactate level with tumor immune microenvironment. Furthermore, machine learning algorithms and cell-cell interaction algorithm were introduced to reveal the connection of tumor cells with immunocytes. By co-culturing CD8 T cells with tumor cells, and performing immunohistochemistry on Xiangya cohort samples further validated results from previous analysis. Discussion: In this work, lactate is proved that contributes to glioblastoma immune suppressive microenvironment. High level of lactate in tumor microenvironment can affect CD8 T cells' migration and infiltration ratio in glioblastoma. To step further, potential compounds that targets to samples from different groups were also predicted for future exploration.

Microbial peptides activate tumour-infiltrating lymphocytes in glioblastoma.

Microbial organisms have key roles in numerous physiological processes in the human body and have recently been shown to modify the response to immune checkpoint inhibitors1,2. Here we aim to address the role of microbial organisms and their potential role in immune reactivity against glioblastoma. We demonstrate that HLA molecules of both glioblastoma tissues and tumour cell lines present bacteria-specific peptides. This finding prompted us to examine whether tumour-infiltrating lymphocytes (TILs) recognize tumour-derived bacterial peptides. Bacterial peptides eluted from HLA class II molecules are recognized by TILs, albeit very weakly. Using an unbiased antigen discovery approach to probe the specificity of a TIL CD4+ T cell clone, we show that it recognizes a broad spectrum of peptides from pathogenic bacteria, commensal gut microbiota and also glioblastoma-related tumour antigens. These peptides were also strongly stimulatory for bulk TILs and peripheral blood memory cells, which then respond to tumour-derived target peptides. Our data hint at how bacterial pathogens and bacterial gut microbiota can be involved in specific immune recognition of tumour antigens. The unbiased identification of microbial target antigens for TILs holds promise for future personalized tumour vaccination approaches.

IL13Rα2-targeted third-generation CAR-T cells with CD28 transmembrane domain mediate the best anti-glioblastoma efficacy.

Chimeric antigen receptor (CAR)-modified T (CAR-T) cell therapy has been proven to be a powerful tool for the treatment of cancer, however, the limits are obvious, especially for solid tumors. Therefore, constantly optimizing the structure of CAR to improve its therapeutic effect is necessary. In this study, we generated three different third-generation CARs targeting IL13Rα2, with the same scFv, but different transmembrane domains (TMDs) from CD4, CD8 or CD28 (IL13-CD4TM-28.BB.ζ, IL13-CD8TM-28.BB.ζ and IL13-CD28TM-28.BB.ζ). CARs were transduced into primary T cells using retroviruses. The anti-GBM efficacy of CAR-T cells was monitored by flow cytometry and real-time cell analysis (RTCA) in vitro and examined in two xenograft mouse models. The differentially expressed genes related to different anti-GBM activity were screened by high throughput RNA sequencing. We observed that T cells transduced with these three CARs have similar anti-tumor activity when co-cultured with U373 cells which expressed higher IL13Rα2 but exhibited different anti-tumor activity when co-cultured with U251 cells that expressed lower IL13Rα2. All the three groups of CAR-T cells can be activated by U373 cells, but only IL13-CD28TM-28.BB.ζ CAR-T cells could be activated and expressed increased IFN-γ after co-culturing with U251 cells. IL13-CD28TM-28.BB.ζ CAR-T cells exhibited the best anti-tumor activity in xenograft mouse models which can infiltrate into the tumors. The superior anti-tumor efficacy of IL13-CD28TM-28.BB.ζ CAR-T cells was partially owing to differentially expressed extracellular assembly, extracellular matrix, cell migration and adhesion-related genes which contribute to the lower activation threshold, increased cell proliferation, and elevated migration capacity.

Single-cell spatial immune landscapes of primary and metastatic brain tumours.

Single-cell technologies have enabled the characterization of the tumour microenvironment at unprecedented depth and have revealed vast cellular diversity among tumour cells and their niche. Anti-tumour immunity relies on cell-cell relationships within the tumour microenvironment1,2, yet many single-cell studies lack spatial context and rely on dissociated tissues3. Here we applied imaging mass cytometry to characterize the immunological landscape of 139 high-grade glioma and 46 brain metastasis tumours from patients. Single-cell analysis of more than 1.1 million cells across 389 high-dimensional histopathology images enabled the spatial resolution of immune lineages and activation states, revealing differences in immune landscapes between primary tumours and brain metastases from diverse solid cancers. These analyses revealed cellular neighbourhoods associated with survival in patients with glioblastoma, which we leveraged to identify a unique population of myeloperoxidase (MPO)-positive macrophages associated with long-term survival. Our findings provide insight into the biology of primary and metastatic brain tumours, reinforcing the value of integrating spatial resolution to single-cell datasets to dissect the microenvironmental contexture of cancer.

Phagocytosis of Glioma Cells Enhances the Immunosuppressive Phenotype of Bone Marrow-Derived Macrophages.

Tumor-associated macrophages (TAM) play a crucial role in immunosuppression. However, how TAMs are transformed into immunosuppressive phenotypes and influence the tumor microenvironment (TME) is not fully understood. Here, we utilized single-cell RNA sequencing and whole-exome sequencing data of glioblastoma (GBM) tissues and identified a subset of TAMs dually expressing macrophage and tumor signatures, which were termed double-positive TAMs. Double-positive TAMs tended to be bone marrow-derived macrophages (BMDM) and were characterized by immunosuppressive phenotypes. Phagocytosis of glioma cells by BMDMs in vitro generated double-positive TAMs with similar immunosuppressive phenotypes to double-positive TAMs in the GBM TME of patients. The double-positive TAMs were transformed into M2-like macrophages and drove immunosuppression by expressing immune-checkpoint proteins CD276, PD-L1, and PD-L2 and suppressing the proliferation of activated T cells. Together, glioma cell phagocytosis by BMDMs in the TME leads to the formation of double-positive TAMs with enhanced immunosuppressive phenotypes, shedding light on the processes driving TAM-mediated immunosuppression in GBM. SIGNIFICANCE: Bone marrow-derived macrophages phagocytose glioblastoma cells to form double-positive cells, dually expressing macrophage and tumor signatures that are transformed into M2-like macrophages and drive immunosuppression.

RNA-binding protein DHX9 promotes glioma growth and tumor-associated macrophages infiltration via TCF12.

BACKGROUND: Glioma is the most common malignant tumor of the central nervous system, with high heterogeneity, strong invasiveness, high therapeutic resistance, and poor prognosis, comprehending a serious challenge in neuro-oncology. Until now, the mechanisms underlying glioma progression have not been fully elucidated. METHODS: The expression of DExH-box helicase 9 (DHX9) in tissues and cells was detected by qRT-PCR and western blot. EdU and transwell assays were conducted to assess the effect of DHX9 on proliferation, migration and invasion of glioma cells. Cocultured model was used to evaluate the role of DHX9 on macrophages recruitment and polarization. Animal study was performed to explore the role of DHX9 on macrophages recruitment and polarization in vivo. Bioinformatics analysis, dual-luciferase reporter assay and chromatin immunoprecipitation (ChIP)-qPCR assay was used to explore the relation between DHX9 and TCF12/CSF1. RESULTS: DHX9 was elevated in gliomas, especially in glioblastoma multiforme (GBM). Besides promoting the proliferation, migration, and invasion of glioma cells, DHX9 facilitated the infiltration of macrophages into glioma tissues and polarization to M2-like macrophages, known as tumor-associated macrophages (TAMs). DHX9 silencing decreased the expression of colony-stimulating factor 1 (CSF1), which partially restored the inhibitory effect on malignant progress of glioma and infiltration of TAMs caused by DHX9 knockdown by targeting the transcription factor 12 (TCF12). Moreover, TCF12 could directly bind to the promoter region of CSF1. CONCLUSION: DHX9/TCF12/CSF1 axis regulated the increases in the infiltration of TAMs to promote glioma progression and might be a novel potential target for future immune therapies against gliomas.

SLC1A5 enhances malignant phenotypes through modulating ferroptosis status and immune microenvironment in glioma.

Glioma is the most common type of primary malignant tumor in the central nervous system with limited treatment satisfaction. Finding new therapeutic targets has remained a major challenge. Ferroptosis is a novel and distinct type of programmed cell death, playing a regulatory role in the progression of tumors. However, the role of ferroptosis or ferroptosis-related genes (FRGs) in glioma progression has not been extensively studied. In our study, a novel ferroptosis-related prognostic model, including 7 genes, was established, in which patients classified into the high-risk group had more immuno-suppressive status and worse prognosis. Among these 7 genes, we screened solute carrier family 1 member 5 (SLC1A5), an FRG, as a possible new target for glioma treatment. Our results showed that the expression of SLC1A5 was significantly upregulated in glioblastoma tissues compared with the low-grade gliomas. In addition, SLC1A5 knockdown could significantly inhibit glioma cell proliferation and invasion, and reduce the sensitivity of ferroptosis via the GPX4-dependent pathway. Furthermore, SLC1A5 was found to be related to immune response and SLC1A5 knockdown decreased the infiltration and M2 polarization of tumor-associated macrophages. Pharmacological inhibition of SLC1A5 by V9302 was confirmed to promote the efficacy of anti-PD-1 therapy. Overall, we developed a novel prognostic model for glioma based on the seven-FRGs signature, which could apply to glioma prognostic and immune status prediction. Besides, SLC1A5 in the model could regulate the proliferation, invasion, ferroptosis and immune state in glioma, and be applied as a prognostic biomarker and potential therapeutic target for glioma.

Identification and validation of an anoikis-associated gene signature to predict clinical character, stemness, IDH mutation, and immune filtration in glioblastoma.

Background: Glioblastoma (GBM) is the most prominent and aggressive primary brain tumor in adults. Anoikis is a specific form of programmed cell death that plays a key role in tumor invasion and metastasis. The presence of anti-anoikis factors is associated with tumor aggressiveness and drug resistance. Methods: The non-negative matrix factorization algorithm was used for effective dimension reduction for integrated datasets. Differences in the tumor microenvironment (TME), stemness indices, and clinical characteristics between the two clusters were analyzed. Difference analysis, weighted gene coexpression network analysis (WGCNA), univariate Cox regression, and least absolute shrinkage and selection operator regression were leveraged to screen prognosis-related genes and construct a risk score model. Immunohistochemistry was performed to evaluate the expression of representative genes in clinical specimens. The relationship between the risk score and the TME, stemness, clinical traits, and immunotherapy response was assessed in GBM and pancancer. Results: Two definite clusters were identified on the basis of anoikis-related gene expression. Patients with GBM assigned to C1 were characterized by shortened overall survival, higher suppressive immune infiltration levels, and lower stemness indices. We further constructed a risk scoring model to quantify the regulatory patterns of anoikis-related genes. The higher risk score group was characterized by a poor prognosis, the infiltration of suppressive immune cells and a differentiated phenotype, whereas the lower risk score group exhibited the opposite effects. In addition, patients in the lower risk score group exhibited a higher frequency of isocitrate dehydrogenase (IDH) mutations and a more sensitive response to immunotherapy. Drug sensitivity analysis was performed, revealing that the higher risk group may benefit more from drugs targeting the PI3K/mTOR signaling pathway. Conclusion: We revealed potential relationships between anoikis-related genes and clinical features, TME, stemness, IDH mutation, and immunotherapy and elucidated their therapeutic value.

Tumor Microenvironment and Immune Escape in the Time Course of Glioblastoma.

Glioblastoma multiforme (GBM) is the most aggressive primary brain tumor with a malignant prognosis. GBM is characterized by high cellular heterogeneity and its progression relies on the interaction with the central nervous system, which ensures the immune-escape and tumor promotion. This interplay induces metabolic, (epi)-genetic and molecular rewiring in both domains. In the present study, we aim to characterize the time-related changes in the GBM landscape, using a syngeneic mouse model of primary GBM. GL261 glioma cells were injected in the right striatum of immuno-competent C57Bl/6 mice and animals were sacrificed after 7, 14, and 21 days (7D, 14D, 21D). The tumor development was assessed through 3D tomographic imaging and brains were processed for immunohistochemistry, immunofluorescence, and western blotting. A human transcriptomic database was inquired to support the translational value of the experimental data. Our results showed the dynamic of the tumor progression, being established as a bulk at 14D and surrounded by a dense scar of reactive astrocytes. The GBM growth was paralleled by the impairment in the microglial/macrophagic recruitment and antigen-presenting functions, while the invasive phase was characterized by changes in the extracellular matrix, as shown by the analysis of tenascin C and metalloproteinase-9. The present study emphasizes the role of the molecular changes in the microenvironment during the GBM progression, fostering the development of novel multi-targeted, time-dependent therapies in an experimental model similar to the human disease.

Sp1 induced gene TIMP1 is related to immune cell infiltration in glioblastoma.

Tumor immune microenvironment exerts a profound effect on the population of infiltrating immune cells. Tissue inhibitor of matrix metalloproteinase 1 (TIMP1) is frequently overexpressed in a variety of cells, particularly during inflammation and tissue injury. However, its function in cancer and immunity remains enigmatic. In this study, we find that TIMP1 is substantially up-regulated during tumorigenesis through analyzing cancer bioinformatics databases, which is further confirmed by IHC tissue microarrays of clinical samples. The TIMP1 level is significantly increased in lymphocytes infiltrating the tumors and correlated with cancer progression, particularly in GBM. Notably, we find that the transcriptional factor Sp1 binds to the promoter of TIMP1 and triggers its expression in GBM. Together, our findings suggest that the Sp1-TIMP1 axis can be a potent biomarker for evaluating immune cell infiltration at the tumor sites and therefore, the malignant progression of GBM.

All Glioblastoma Are Not Equal: Distinct Spatial Immune Profiles Between De Novo and Recurrent Tumors.

Lack of insight into the immunosuppressive microenvironment of glioblastoma (GBM) hinders successful application of immunotherapy. In this issue, Alanio and colleagues identify a distinct T-cell localization profile between recurrent and de novo GBM highlighting the importance of spatial heterogeneity and providing new avenues to explore to improve GBM outcome. See related article by Alanio et al., p. 800 (3).

T-cell dysfunction in the glioblastoma microenvironment is mediated by myeloid cells releasing interleukin-10.

Despite recent advances in cancer immunotherapy, certain tumor types, such as Glioblastomas, are highly resistant due to their tumor microenvironment disabling the anti-tumor immune response. Here we show, by applying an in-silico multidimensional model integrating spatially resolved and single-cell gene expression data of 45,615 immune cells from 12 tumor samples, that a subset of Interleukin-10-releasing HMOX1+ myeloid cells, spatially localizing to mesenchymal-like tumor regions, drive T-cell exhaustion and thus contribute to the immunosuppressive tumor microenvironment. These findings are validated using a human ex-vivo neocortical glioblastoma model inoculated with patient derived peripheral T-cells to simulate the immune compartment. This model recapitulates the dysfunctional transformation of tumor infiltrating T-cells. Inhibition of the JAK/STAT pathway rescues T-cell functionality both in our model and in-vivo, providing further evidence of IL-10 release being an important driving force of tumor immune escape. Our results thus show that integrative modelling of single cell and spatial transcriptomics data is a valuable tool to interrogate the tumor immune microenvironment and might contribute to the development of successful immunotherapies.

Recent Development in NKT-Based Immunotherapy of Glioblastoma: From Bench to Bedside.

Glioblastoma multiforme (GBM) is an aggressive and dismal disease with a median overall survival of around 15 months and a 5-year survival rate of 7.2%. Owing to genetic mutations, drug resistance, disruption to the blood-brain barrier (BBB)/blood-brain tumor barrier (BBTB), and the complexity of the immunosuppressive environment, the therapeutic approaches to GBM represent still major challenges. Conventional therapies, including surgery, radiotherapy, and standard chemotherapy with temozolomide, have not resulted in satisfactory improvements in the overall survival of GBM patients. Among cancer immunotherapeutic approaches, we propose that adjuvant NKT immunotherapy with invariant NKT (iNKT) and cytokine-induced killer (CIK) cells may improve the clinical scenario of this devastating disease. Considering this, herein, we discuss the current strategies of NKT therapy for GBM based primarily on in vitro/in vivo experiments, clinical trials, and the combinatorial approaches with future therapeutic potential.

Immune Activity and Response Differences of Oncolytic Viral Therapy in Recurrent Glioblastoma: Gene Expression Analyses of a Phase IB Study.

PURPOSE: Previously, clinical trials of experimental virotherapy for recurrent glioblastoma multiforme (GBM) demonstrated that inoculation with a conditionally replication-competent Δγ134.5 oncolytic herpes simplex virus (oHSV), G207, was safe. Following the initial safety study, a phase Ib trial enrolled 6 adult patients diagnosed with GBM recurrence from which tumor tissue was banked for future studies. PATIENTS AND METHODS: Here, we analyzed tumor RNA sequencing (RNA-seq) data obtained from pre- and posttreatment (collected 2 or 5 days after G207 injection) biopsies from the phase Ib study patients. RESULTS: Using a Spearman rank-order correlation analysis, we identified approximately 500 genes whose expression pattern correlated with survival duration. Many of these genes were enriched for the intrinsic IFN-mediated antiviral and adaptive immune functional responses, including immune cell chemotaxis and antigen presentation to T-cells. Furthermore, we show that the expression of several T-cell-related genes was highest in the patient with the longest survival after G207 inoculation. CONCLUSIONS: Our data support that the oHSV-induced type I IFN production and the subsequent recruitment of an adaptive immune response differed between enrolled patients and showed association with survival duration in patients with recurrent malignant glioma after treatment with an early generation oHSV.

Identification of Tumor Antigens and Immune Subtypes of Glioblastoma for mRNA Vaccine Development.

The use of vaccines for cancer therapy is a promising immunotherapeutic strategy that has been shown to be effective against various cancers. Vaccines directly target tumors but their efficacy against glioblastoma multiforme (GBM) remains unclear. Immunotyping that classifies tumor samples is considered to be a biomarker for immunotherapy. This study aimed to identify potential GBM antigens suitable for vaccine development and develop a tool to predict the response of GBM patients to vaccination based on the immunotype. Gene Expression Profiling Interactive Analysis (GEPIA) was applied to evaluate the expression profile of GBM antigens and their influence on clinical prognosis, while the cBioPortal program was utilized to integrate and analyze genetic alterations. The correlation between antigens and antigen processing cells was assessed using TIMER. RNA-seq data of GBM samples and their corresponding clinical data were downloaded from the Cancer Genome Atlas (TCGA) and the Chinese Glioma Genome Atlas (CGGA) for further clustering analysis. Six overexpressed and mutated tumor antigens (ARHGAP9, ARHGAP30, CLEC7A, MAN2B1, ARPC1B and PLB1) were highly correlated with the survival rate of GBM patients and the infiltration of antigen presenting cells in GBMs. With distinct cellular and molecular characteristics, three immune subtypes (IS1-IS3) of GBMs were identified and GBMs from IS3 subtype were more likely to benefit from vaccination. Through graph learning-based dimensional reduction, immune landscape was depicted and revealed the existence of heterogeneity among individual GBM patients. Finally, WGCNA can identify potential vaccination biomarkers by clustering immune related genes. In summary, the six tumor antigens are potential targets for developing anti-GBMs mRNA vaccine, and the immunotypes can be used for evaluating vaccination response.