Glioblastoma (GBM) is an aggressive and devastating primary brain tumor characterized by a high proliferation rate, infiltrative growth, and resistance to traditional interventions. It is the most common primary brain tumor in adults, with an incidence ranging from two to five cases per 100,000 people (1). GBM accounts for approximately 48.3% of all primary malignant brain tumors (2) and affects individuals across various age groups. Its prevalence increases with age, peaking in the 75–84 years of age group, with a median age of 64 at diagnosis (3). GBM is more common in men than women, and more prevalent in Caucasians compared to Afro-Americans, Africans, American Indians, and Asians (3). This malignancy primarily affects the cerebral hemispheres, particularly in subcortical white matter and deep grey matter of the frontal and temporal lobes, although it can arise in any part of the brain (4).

GBM tumors are a type of diffuse astrocytic tumor classified as grade IV by the World Health Organization (WHO). According to the 2021 WHO classification of CNS tumors, they are now required to express isocitrate dehydrogenase (IDH-wild type) to be classified as GBM (5). GBM tumors are generally divided into primary and secondary tumors with the former arising from either progenitor or neural stem cells (90% of GBM), and the latter transforming from pre-existing, less aggressive tumors (10%) (6). Regardless, the pathogenesis of GBM involves wildly variable and complex genetic and molecular alterations. Amplification of the epidermal growth factor receptor (EGFR) gene, mutations in the IDH1 and IDH2 genes, and loss of tumor suppressor genes such as TP53 and PTEN serve as the most identified genetic abnormalities in GBM (7).

These molecular aberrations contribute to uncontrolled cell growth, angiogenesis, and invasion into the surrounding brain tissue (8). Therefore, current treatment modalities for GBM typically involve a combination of surgery, radiation therapy, and chemotherapy. The goal of surgery is to remove as much tumor as possible without compromising neurological function. Adjuvant radiation therapy follows surgical debulking by delivering targeted beams to the remaining cancer cells in the tumor bed to delay recurrence. Concurrently, chemotherapy, usually temozolomide (TMZ), is administered during and after radiation to enhance the treatment outcomes. Despite these aggressive treatment approaches, GBM inevitably recurs, underscoring the challenge of eradicating residual tumor cells and the need for innovative therapeutic interventions. Because these standard therapies only extend survival an average of 15 months (9), with less than 30% of patients living 2 years (10), neuro-oncologists are eager to develop new therapies.

Here, we review major challenges to developing successful treatments against GBM and focus on how hypoxia contributes to these obstacles.

Designing novel therapies against GBM has proven to be remarkably challenging due to the extensive variability of the GBM tumor microenvironment (TME) within brain tissue. GBM tumors metastasize outside the central nervous system (CNS) in less than 0.2% of cases (11), suggesting they require the unique milieu of brain parenchyma to survive. Together, the variety of unique resident brain cells, the favorable blood supply with high delivery of oxygen and nutrients to the brain, and the blood-brain barrier that creates a relatively immune-privileged state create an environment in which GBM tumor cells thrive. Once GBM cells develop and migrate into adjacent tissues, these tumors develop prototypical heterogeneity both within single tumors as well as across tumors in multifocal disease.

GBM is characterized by an overwhelming level of intratumoral heterogeneity, manifesting at various biological levels. At the genetic level, extensive genomic analyses have revealed diverse molecular alterations within individual tumors and across different lesions of multifocal GBM cases. The Cancer Genome Atlas (TCGA) research network has played a pivotal role in uncovering the genetic landscape of GBM, identifying mutations in genes such as EGFR, TP53, and PTEN that contribute to tumor heterogeneity (12, 13). This genetic diversity not only influences the aggressive behavior of GBM, but also poses challenges for developing targeted therapies.

In addition to genetic heterogeneity, GBM tumors exhibit profound cellular diversity. Amongst a background of mutated glial cells demonstrating rapid proliferation, migration, and infiltration throughout normal brain tissue, the GBM TME also includes cancer stem cells (CSCs) that demonstrate multipotent differentiation and true self-renewal properties (14, 15). Stromal cells, vascular endothelial cells, and both residing and infiltrating immune cells further contribute to specific niches within the TME (16). Traditionally, intratumoral heterogeneity is considered a result of distinct genetic alterations within various tumor cells (17). However, it is more likely the complex interaction of each of these cell types within the extracellular matrix (ECM) that contributes to heterogeneity. For example, within these specific niches, the proliferating or CSC-like tumor cells dynamically interact with resident parenchymal cells, such as microglia and macrophages, to orchestrate a unique milieu. The microscopic result of these various TMEs invariably leads to necrotic regions, endothelial proliferation around vessels, and hypoxic areas (16), all of which are classical pathological findings of GBMs.

The spatial heterogeneity observed within GBM tumors further adds to the complexity of the disease. Intra-tumoral variations in oxygen levels, nutrient availability, and blood flow create distinct microenvironments within the tumor mass. These variations lead to the formation of hypoxic regions, which are known to influence tumor progression, treatment resistance, and the emergence of more aggressive phenotypes (18).

To summarize, the heterogeneity of GBM is a complex phenomenon that covers various aspects like genetics, cells, microenvironment, and spatial diversity. The challenges posed by this heterogeneity highlight the intricate nature of GBM biology, which emphasizes the requirement for personalized and targeted therapeutic approach. The advancements in genomic profiling, single-cell analysis, and a deeper understanding of the TME pave the path for developing innovative strategies to address the diverse components that contribute to GBM heterogeneity.

GBM tumors are notorious for creating an immunosuppressive microenvironment that facilitates their growth and evades the host’s immune response. This immunosuppressive nature is orchestrated by a complex interplay of various cellular and molecular components within the TME ( Figure 1 ). One prominent feature is the recruitment and activation of immunosuppressive cells, including regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs), which inhibit the activity of cytotoxic T lymphocytes (CTLs) and natural killer (NK) cells, crucial effectors of anti-tumor immunity (19, 20). The abundance of these immunosuppressive cells contributes to the establishment of an immune-tolerant milieu that shields GBM from immune-mediated destruction.

VEGF is the master regulator of this immunosuppression. It downregulates intercellular adhesion molecule 1 (ICAM1) and vascular cell adhesion molecule 1 (VCAM1), adhesion molecules that reduce the infiltration of T-effector cells, activate antigen-specific regulator T cells (Tregs), and inhibit the maturation of dendritic cells (DCs) (21). Interestingly, a variety of brain tumors induce reduced expression of cell membrane sphingosine-1-phosphate receptor (S1P1), which causes trapping of T cells in bone marrow, and therefore actively induces an immunosuppressive state (22). Furthermore, Andaloussi et al. found a dramatic reduction in CD4+ and CD8+ cells, a reduction in the size of the thymus, and a decrease in thymic cellularity in the brains of mice implanted with murine glioma GL261 cells (23).

Upon reaching the GBM TME, T cells encounter a harshly immunosuppressive local state due to high concentrations of immunosuppressive modulators such as transforming growth factor β (TGF-β), prostaglandin E-2 (PGE2), and interleukin 10 (IL-10). This cytokine milieu shifts microglia and tumor-infiltrating myeloid cells to a quiescent phenotype that dampens immune responses and supports tumor growth (24). Additionally, GBM tumors exploit the expression of immune checkpoint molecules to dampen the anti-tumor immune response. Programmed cell death protein 1 (PD-1) and its ligand PD-L1, along with cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), serve as key immune checkpoints that, when engaged, downregulate T cell activity (25, 26). GBM cells and immune cells within the TME upregulate these checkpoint molecules, leading to T cell exhaustion and dysfunction (26). The upregulation of these immune checkpoints contributes to the immune escape mechanisms employed by GBM tumors, limiting the effectiveness of immunotherapeutic interventions.

Myeloid cells play a pivotal role in establishing an immunosuppressive microenvironment within GBM tumors. myeloid cells, including M2 macrophages, immature monocytes, and myeloid-derived suppressor cells (MDSCs), are the most common non-malignant cells identified in the GBM TME (27). Among these cells, MDSCs are frequently recruited to the GBM microenvironment, where they contribute to the inhibition of both innate and adaptive immune responses, dampening the activity of cytotoxic T lymphocytes (CTLs) and natural killer (NK) cells, which are crucial effectors of anti-tumor immunity (20). These cells play a major role in tumor-mediated immunosuppression through further immune checkpoint molecule expression and anti-inflammatory cytokine release (28). Upregulation of programmed death-ligand 1 (PD-1) on tumor-infiltrating myeloid cells is a crucial GBM-derived method of immune evasion (29, 30). GBM cells themselves release interleukin 6 (IL-6), which is necessary for myeloid PD-L1 induction through a signal transducer and activator of transcription 3 (STAT3)-dependent fashion and disrupting IL6 signaling enhances immune-mediated antitumor effects against GBM (28).

L. Pang et al. recently identified Kunitz-type protease inhibitor TFPI2 as a crucial player in connecting various GBM cell populations, including self-renewing stem cells and immunosuppressive microglia. TFPI2 enhances stem cell renewal and tumor growth by activating the c-Jun N-terminal kinase–signal transducer and activator of transcription (STAT)3 pathway. Simultaneously, secreted TFPI2 triggers the polarization of immunosuppressive microglia via STAT6 signaling. In human GBM, TFPI2 is associated with stemness, immunosuppression, increased microglia, and poor prognosis. Inhibiting the TFPI2-CD51-STAT6 pathway both activates T cells and improves the efficacy of anti-PD1 immunotherapy against GBM in mice (31). Moreover, this same group also characterized legumain (LGMN) as a key protease that is transcriptionally regulated by HIF1α and enriched in tumor-associated macrophages. Increased LGMN activates the GSK-3β-STAT3 signaling pathway to promote TAM immunosuppression. Inhibiting HIF1α and LGMN enhances CD8+ T cell- anti-tumor immunity, impairs GBM tumor progression, and improves immunotherapy responses (anti-PD1 therapy) in mice (32). Altogether, these data suggest that targeting TFPI2, as well as HIF1α and LGMN, are exciting strategies to combat GBM.

Furthermore, myeloid cells within the GBM microenvironment actively promote the expansion and activation of regulatory T cells (Tregs). Tregs are a subset of T lymphocytes that play a crucial role in immune tolerance and homeostasis. The interaction between myeloid cells and Tregs creates a self-reinforcing loop of immunosuppression as Tregs, in turn, support the survival and function of MDSCs. This collaboration contributes to establishing an immunosuppressive network that shields GBM from effective immune surveillance and clearance (19).

Strategies to counteract the immunosuppressive effects of myeloid cells in GBM are actively being explored in the field of cancer immunotherapy. Targeting MDSCs or modulating their function to reduce their immunosuppressive capacity represents a potential avenue for enhancing the anti-tumor immune response. However, the remarkable plasticity and adaptability of myeloid cells in response to the GBM microenvironment pose significant challenges to the development of effective therapeutic interventions. A comprehensive understanding of the intricate interactions between myeloid cells and the immune system within the GBM TME is crucial for devising innovative and targeted immunotherapeutic strategies.

The blood–brain barrier (BBB) plays an active role in generating immunosuppression within the GBM TME. The BBB is a specialized endothelial barrier that separates the bloodstream from the brain tissue, tightly regulating the passage of toxins, drugs, and immune cells into the CNS. Altogether, there are at least five ways in which the BBB affects the neuroimmune status: through BBB permeability, modulation of the BBB transporters, removal of immunoreactive substances by cells of the BBB, transfer of immune cells through the BBB into the brain, and secretion of immunoreactive substances by cells of the BBB (33). In GBM, the stability of the BBB reflects the degree to which the BBB contributes to the establishment of an immunosuppressive milieu within the tumor. The selective permeability of the BBB limits the entry of circulating immune cells, including cytotoxic T lymphocytes (CTLs) and natural killer (NK) cells, into the tumor site. This restriction hampers the effectiveness of the anti-tumor immune response, as these cells are crucial effectors in recognizing and eliminating cancer cells (34). This restrictive behavior also prevents the delivery of many therapeutics into the tumor. Therefore, considerable efforts have been made to modulate or bypass the BBB, either invasively or non-invasively, for the purpose of drug delivery (35). However, the delicate balance between protecting the brain from potential immune-mediated damage and promoting an effective anti-tumor immune response poses a significant challenge in developing targeted and safe therapeutic approaches.

The BBB traditionally was thought to contribute to a relatively immune-privileged state of the brain, creating an environment where immune surveillance is inherently restricted. This immune-privileged status of the brain was thought to be related to the lack of a traditional lymphatic drainage system, limiting immune cells to efficiently traffic to and from the CNS and target tumor cells (36). More recently, however, the brain is no longer considered a truly immune-privileged organ, as microglia act as resident antigen-presenting cells and allow T cells to traffic in and out of the CNS (24). In GBM, this immune privilege is exploited by the tumor, enabling it to thrive in an environment that is less susceptible to immune-mediated destruction (36). The leaky BBB in GBM also leads to the influx of pro-tumoral immune cells into the TME, such as infiltrating MDSCs and T_regs. Once infiltrated in the tumor, these cells contribute to the immunosuppressive network by suppressing effector immune responses and promoting a tumor-favorable microenvironment (20, 37).

The BBB also regulates the transport of immune-modulating molecules into the brain. Endothelial cells within the BBB express various transporter systems on their surface, encompassing both active transport mechanisms and those functioning through facilitated diffusion. The roles of certain transporters undergo alterations in the context of neuroinflammation and are subject to modulation by signaling molecules, with the P-glycoprotein (Pgp) protein being a notable example. Pgp is a BBB efflux transporter that limits drug entry into the CNS through interactions with diverse ligands, including protease inhibitors, opiates, antiepileptics, cyclosporins, glucocorticoids, aldosterone, dexamethasone, and calcium channel blockers. Indeed, Pgp acts as a gatekeeper of the CNS as its functionality undergoes changes during inflammation, primarily resulting in the suppression of its transport activity in vivo. Consequently, induction of Pgp by proinflammatory cytokines such as TNFα, IL-1β, IL-6, IL-2, and IFNγ leads to a reduction in Pgp mRNA expression in CNS endothelial cells, affecting both synthesis and activity of the transporter (38, 39).

Understanding the intricate relationship between the BBB and immunosuppression in GBM is crucial for the development of therapeutic strategies. Strategies that aim to selectively breach the BBB or modulate its permeability are actively being explored to enhance immune cell infiltration into the tumor and improve the efficacy of immunotherapeutic interventions.

GBMs are notoriously vascular tumors in which endothelial cells support tumor growth by delivering nutrients to the tumor; however, blood vessels within GBM tumors are faulty because of their disorganized formation. GBM tumor cells, inflammatory, and stromal cells all contribute to an aggressive production of a disarrayed tangle of leaky, abnormal new blood vessels (40). Angiogenesis plays a major role in the development of these abnormal vessels; however, neovascularization via proliferation of existing endothelial cells and via hypoxia-inducible factor 1α (HIF-1α) recruitment of bone marrow-derived vascular cells also contributes to tumor angiogenesis (41). Glioma stem cells themselves also generate novel vasculature by transforming themselves into pericytes or endothelial cells (42, 43). This pro-angiogenesis drive is supported by the tumor’s inherent metabolic, inflammatory, and hypoxic conditions. Importantly, tumor cells that are often found in the periphery of the hypoxic necrotic core release unregulated levels of VEGF as well as additional paracrine factors that further lead to vascular bed inflammation and increased vascular permeability (44).

VEGF is a master regulator of the vascular aberrations in GBM, which in addition to driving faulty neo-angiogenesis, also result in the breakdown of the BBB. The BBB is comprised of astrocytes, endothelial cells, and tight junctions by pericyte feet, which together tightly regulate the transfer of molecules between the blood and brain parenchyma (45). The breakdown of the BBB results in increased vessel permeability with the influx of fluid into tumor tissue and surrounding brain tissue, resulting in cerebral edema and increased intracranial pressure. Simultaneously, the slowing of blood flow through these compromised blood vessels results in patchy oxygen delivery within the tumor bed. These focal areas of hypoxia develop from occluded vessels into pseudopallisading necrosis, a pathognomonic characteristic of GBM. These necrotic niches also recruit macrophages and other innate immune cells that further propagate angiogenesis and immunosuppression.

More recently, Pang et al., have shown that the circadian regulation of glioma stem cells alters the immunosuppressive TME in GBM tumors through paracrine and autocrine mechanisms (46, 47). One way the stem cells interact with the TME is via circadian regulation through CLOCKs, which include both transcriptional activators (such as CLOCK-BMAL1 complex) as well as inhibitors (such as cryptochrome 1 and 2, period 1,2,3 and REV-ERBα (48). CLOCK-BMAL1 complex in glioma stem cells leads to the upregulation of periostin (POSTN) that ultimately activates the TANK-binding kinase 1 (TBK1) in endothelial cells. The inhibition of POSTN and TBK1 in GBM mouse models, as well as pathological analysis in GBM patient samples, reveal that this pathway may be an interesting therapeutic strategy for targeting GBM angiogenesis (49).

GBM tumors are characterized by areas of tissue hypoxia (50–53), which not only render them more aggressive (51) but are also critical for the maintenance of glioma stem cells, the tumoral cell population that is responsible for resistance to therapies (54, 55) and tumor recurrence (56). Hypoxic and inflammatory responses overlap powerfully in the GBM microenvironment (51), as hypoxic conditions restrict cytotoxic T lymphocyte development while promoting the proliferation and expression of inflammatory cytokines (57). Although not yet completely understood, glioma stem cells, which thrive in hypoxic conditions, preferentially inhibit the proliferation of activated T cells, as compared to differentiated tumor cells (58). Hypoxic conditions within GBM tumors activate the STAT3 signaling pathway, which mediates HIF-1α and leads to immunosuppression (59). Finally, the hypoxic TME recruits immunosuppressive cells such as tumor-associated macrophages, myeloid-derived suppressor cells, and T_regs (60), which all reduce the immune response through the expression of inhibitory molecules on the surface of effector immune cells (51).

Immunotherapies are one of the newest frontiers in combating cancer ( Figure 2 ). These therapies take advantage of a patient’s existing functional immune system by harvesting immune cells and retraining them to attack and kill tumor cells. These treatments have been FDA-approved for multiple blood-borne cancers (61) and are used investigationally for brain tumors. Immunotherapeutic strategies cover a wide range of treatments and include vaccines (62), adoptive cell therapy (63), monoclonal antibodies (64), oncolytic viruses (65, 66), and immune checkpoint blockers (67, 68).

Tumor Treating Fields (TTFields) therapy is an innovative treatment modality for GBM, that is non-invasive and can be administered alongside standard treatments such as chemotherapy and radiation therapy. TTFields therapy involves the use of low intensity alternating electric fields delivered via transducer sticker arrays placed on the patient’s scalp. TTFields therapy is an antimitotic approach that disrupts the division of glioblastoma cells and their organelle assembly. This is achieved through the application of low intensity alternating electric fields directly to the tumor site. TTFields therapy is its favorable safety profile, as it primarily targets dividing cancer cells while sparing healthy brain tissue from damage.

Clinical studies evaluating the efficacy of TTFields therapy in GBM have shown relatively promising results. The landmark EF-14 was a randomized clinical trial involving 695 patients with GBM who had completed initial radiochemotherapy, were randomized to the addition of TTFields to maintenance temozolamide. The results demonstrated a significant improvement in both progression-free survival (6.7 months in the TTFields plus temozolomide group and 4.0 months in the temozolomide-alone group) and overall survival (median overall survival was 20.9 months in the TTFields-temozolomide group vs 16.0 months in the temozolomide-alone group (HR, 0.63; 95% CI, 0.53–0.76; P < .001)) in newly diagnosed glioblastoma patients who received TTFields therapy in addition to standard chemotherapy compared to those who received chemotherapy alone (92). Additionally, TTFields therapy has been associated with a favorable quality of life profile, with main risk.

While TTFields therapy represents a significant advancement in treating glioblastoma, ongoing research aims to optimize its efficacy and explore its potential in other cancer types. Studies are investigating various aspects of TTFields therapy, including optimal treatment schedules, combination with other therapeutic modalities, and mechanisms of action to enhance its anti-tumor effects further. As TTFields therapy continues to evolve, it holds promise as a valuable addition to the armamentarium of treatments for glioblastoma and potentially other malignancies, offering patients a novel and well-tolerated therapeutic option.

Recently, cytokines such as IFN-α, TNF-α, and IL-12 have been used as novel therapeutic approaches to treat GBM treatment, aiming to alter the immunosuppressive TME to evoke an effective antitumor immune response (93). IFN-α, for instance, has been reported to enhance T cell and macrophage activity and inhibit tumor angiogenesis and immune-suppressive gene expression (94). Conversely, TNF-α fosters dendritic cell maturation and subsequent T cell activation, while IL-12 is associated with improved CAR-T cell functionality, increased infiltration of CD4+ T cells, and reduced T-regulatory cell frequency within the tumor microenvironment (95, 96).

However, IFN-α therapy is hindered by its significant systemic toxicity and limited efficacy at maximum tolerated doses, as evidenced by clinical trials reporting adverse effects such as hyperthermia, shivering, headaches, gastrointestinal symptoms, and orthostatic hypotension (97). This restricts its current use, highlighting the need for future strategies that combine IFN-α with other treatments to enhance both efficacy and tolerability (93).

IL-6 is a multifaceted cytokine that orchestrates several biological processes, such as the acute phase response, immune defense against infections, leukocyte infiltration at inflammation sites, leukocyte maturation, and endothelial cell characteristics (98). The IL-6 signaling pathway initiates with IL-6 binding to membrane-bound IL-6R (gp80) present on various cells like hepatocytes, neutrophils, monocytes, B cells, and T cells. Subsequently, the IL-6-IL-6R complex binds to membrane-bound gp130, expressed ubiquitously by all cells, forming an activated IL-6 receptor that triggers intracellular signaling. This classic IL-6 signaling pathway leads to anti-inflammatory responses. Alternatively, IL-6 can also act via the trans-signaling pathway, binding to soluble forms of IL-6R, and then binding to membrane-bound gp130, activating target cells lacking membrane-bound IL-6R and inducing pro-inflammatory responses (99, 100). Both pathways activate Janus kinases (JAKs) and signal transducer and activator of transcription (STATs) to modulate cellular responses (98).

Overexpression of IL-6 has been linked to poor survival in GBM patients based on data from The Cancer Genome Atlas (TCGA) and the Repository of Molecular Brain Neoplasia Data (REMBRANDT) (101). Within the GBM TME, IL-6 is secreted by tumor cells, glioma-associated microglia/macrophages (GAMs), and tumor-associated endothelial cells (102, 103), with its expression induced by treatments such as chemotherapy (104) and radiotherapy (105). IL-6 exerts its effects through IL-6R expressed on tumor cells and GAMs, promoting immunosuppressive GAMs and impairing T cell functions. Treatment with anti-IL-6 siltuximab, anti-IL-6R tocilizumab, or STAT3 inhibitor Stattic has been shown to mitigate PD-L1 expression and T cell apoptosis, underscoring the role of IL-6 signaling in regulating T cell responses via GAMs (28). Additionally, IL-6 sustains tumor progression by directly influencing glioblastoma cells to activate anti-apoptotic pathways, induce autophagy, and promote invasion through mechanisms involving MMPs and fascin-1 (106).

Oncolytic viruses (OVs) are weakly pathogenic viruses designed to selectively infect, replicate within, and kill tumor cells through apoptosis while avoiding infection of healthy cells (107). OVs stimulate the innate immune system through pathogen-associated molecular patterns and pattern recognition receptors that recruit immune cells such as neutrophils, macrophages, natural killer cells, Th1 cells, leading to cell lysis (108). This response also leads to an adaptive immune reaction to new cancer antigens, therefore possibly propagating long-term anti-tumor immunity (109). Moreover, OVs are able to convert the GBM TME from “cold” to “hot” by triggering inflammation, and improve the outcome of immune checkpoint inhibitors (110). OVs utilizing modified herpes simplex virus, adenovirus, measles virus, parvovirus, reovirus, and poliovirus, amongst others, are being studied in clinical trials against GBM worldwide (93). Because OVs are noted to have oncolytic activity but low specificity, various methods have been employed to improve tumor-specificity such as deleing virulence genes, adding tumor-specific promotors, or adding tumor suppressor gene miRNAs (111).

More than 20 oncolytic viruses have been evaluated in clinical trials for GBM. Key examples include HSV-1, adenovirus, reovirus, measles virus, Newcastle disease virus, and poliovirus (112). To circumvent the limitations imposed by the BBB, innovative delivery methods such as convection-enhanced delivery (CED) have been utilized. CED is particularly used for the recombinant nonpathogenic polio-rhinovirus chimera (PVSRIPO) and works by using a pressure gradient within a catheter to deliver therapeutic agents into the interstitial spaces of the brain parenchyma (113). Achieving efficient and safe delivery of oncolytic viruses is critical for successful virotherapy. The difficulty of delivering these viruses to the CNS and their potential elimination by the immune system has made intratumoral delivery the preferred approach. However, for optimal outcomes, the oncolytic virus should ideally be administered systemically to reach both primary and metastatic tumor sites, especially in the cases of multifocal GBM.

The predominant immune cells in the GBM TME are macrophages originating from peripheral monocytes, known as tumor-associated macrophages (TAMs) (114). TAMs primarily express the M2 surface marker and produce IL-10 and TGF-β with STAT3 expression, promoting an immunosuppressive “cold” TME (115). Additionally, T cell exhaustion, characterized by the upregulation of inhibitory molecules and increased Treg cells, is a hallmark of GBM (116). Moreover, expression levels of immunomodulatory chemokines CXCL2, CX3CL1, CCL5, and CCL2 are elevated in GBMs (117). Consequently, OVs could enhance alternative therapies by reversing these conditions to promote greater immune infiltration and improved tumor destruction (118). For example, delivering concomitant OVs with ICIs has shown promising outcomes. Infection with the measles virus in GBM models resulted in increased PD-L1 expression (119); and using both measles as well as anti-PD-1 antibodies resulted in longer survival in murine glioma models compared to survivals in animals treated with each therapy alone (120). Similarly, OVs can improve CAR T cell therapy against solid tumors (112). Care must be taken however, as application of a vesicular stomatitis virus encoding IFNβ resulted in the reverse outcome on CAR T cells targeting EGFRvIII in B16 murine glioma tumors (121). Further understanding of the complex interplay between OVs and other immunological treatments must be further characterized prior to use in larger-scale clinical trials.

With ever-advancing technologies to study the TME, intratumoral bacterial and fungal biomasses have been characterized in multiple tumor types (122, 123). Intratumoral microbiota generates anti-tumorigenic effects by increasing antigen presentation, activating NK and T cells, improving immunosurveillance, and ultimately generating metabolites that suppress tumor progression (124). Intratumoral bacteria can induce potent immunogenic antitumor effects and significantly extend survival rates with effective immunological memory in various murine cancer models (125), such as against melanoma (126). Memory T cells pre-stimulated by bacteria-derived or tumor-derived antigens within GBM are found in peripheral blood and tumors (127). There is cross-reactivity between human tumor antigens and bacterial antigens due to molecular mimicry, allowing tumor antigen recognition by CD4+ T cells (127) and CD8+ cytotoxic T cells (128). Indeed, GBM contains distinct microbiota compared to adjacent normal tissues (129). Bacteria-specific peptides obtained from normal cell lines or GBM tumor samples are presented via MHC II molecules. After stimulation by these bacterial peptides, tumor-infiltrating lymphocytes secrete more pro-inflammatory cytokines and recruit CD8+ T cells to enhance tumor-killing (130). For patients where intratumoral bacterial nucleic acids or peptides are detected in very low quantities, subcutaneous injection of tumor antigen-engineered commensal bacteria to elicit distant anti-tumor immune responses is a novel and safe approach (126), and is an exciting option for GBM treatment (131). Through the subcutaneous delivery of lysed-engineered bacterial peptides, epidermal Langerhans cells further present tumor peptides in MHC molecules, allowing distant primed CD8+ T cells to enter the circulation and infiltrate tumors. This vaccination approach avoids HLA restriction (132) and is an exciting option to be used alone or in combination with other tumor vaccine strategies (131).

Various approaches have been suggested to address hypoxia in cancer, including inhibition of HIF signaling, the use of hypoxia-activated prodrugs (HAPs), targeting downstream pathways like the unfolded protein response (UPR) and mTOR, as well as metabolic interventions (133). Considering hypoxia’s central role in regulating tumor progression and immune suppression, it is plausible to explore hypoxia as a potential target in combined cancer immunotherapy. This section delves into the manipulation of hypoxia to enhance the effectiveness of cancer immunotherapy, drawing insights from both pre-clinical and clinical studies. Please refer to Box 1 for more details.

Future advances leveraging RNA-seq and single-cell RNA sequencing techniques, along with bioinformatics and AI technology, hold immense potential in unraveling the intricacies of GBM TME. RNA-seq provides a high-throughput method to analyze gene expression profiles, allowing researchers to decipher the molecular patterns associated with different cell types within the TME. Moreover, sc-RNA-seq enables the dissection of individual cells, providing unprecedented insights into the heterogeneity and cellular interactions driving GBM progression and therapy resistance (53).

By harnessing the power of bioinformatics and AI, researchers can sift through vast amounts of genomic data generated from RNA-seq and sc-RNA-seq experiments to identify key molecular players and signaling pathways implicated in GBM pathogenesis. Advanced computational algorithms can integrate multi-omics data, such as genomics, transcriptomics, proteomics and epigenomics, to construct comprehensive models of the GBM TME. These models not only enhance our understanding of tumor biology but also facilitate the identification of novel therapeutic targets and the development of personalized treatment strategies. Furthermore, AI-driven approaches enable predicting patient outcomes and therapeutic responses, paving the way for precision medicine in GBM management.

It’s important to understand the dynamics of hypoxia in the GBM TME to develop targeted therapies that address the complexities of such an environment. The unique features of the TME, such as the presence of CSCs, interactions with stromal and immune cells, and the vascular and hypoxic niches, can be targeted to develop more effective and personalized treatments. However, developing clinically viable therapies requires overcoming the challenges posed by the remarkable heterogeneity inherent to GBM. Advances in precision medicine, immunotherapy, and innovative drug delivery systems are being explored to design therapies that account for the diverse elements of the GBM TME and to improve patient outcomes.

LF: Writing – original draft, Writing – review & editing.