TMZ, commercially known as Temodar, has received FDA approval for treating glioblastoma in 2005, based on its improvement in overall OS (26). It belongs to the new class of oral alkylating agents with an imidazole ring, with its chemical designation being 3-methyl-4-oxoimidaz[5,1-d][1,2,3,5]tetrazine-8-carboxamide (27–29). TMZ is a BBB penetrating pro-drug, which gets hydrolyzed under physiological pH to its active drug form, methyltriazen-1yl imidazole-4-carboxamide (MTIC). The drug modifies its targets by adding methyl groups in genomic DNA in guanine and adenine, in N7 and O6, and N3 sites respectively. Methylation causes DNA replication errors and disruption of the mismatch repair (MMR) repair system, which leads to DNA double-strand breaks and eventually programmed cell death. Therefore, the clinical benefit of TMZ is significantly influenced by the methylation of MGMT promoter; patients with methylation on the MGMT promoter show a better response to TMZ. While TMZ is generally well-tolerated, it is accompanied with several side effects including hematological and hepatotoxicity and others (30, 31). Furthermore, glioblastoma has several mechanisms of resistance to TMZ, as reviewed elsewhere (15–17).

BCNU (Bischloroethyl Nitrosourea Carmustine) belongs to the class of N-nitrosoureas [1,3-bis(2-chloroethyl)urea], a monofunctional alkylating agent. It is clinically approved in two forms for glioblastoma treatment. Firstly, intravenous (IV) BCNU was approved in 1977 for treating recurrent glioblastoma, however its use has declined due to the availability of more effective treatments will less toxicity. Secondly, Carmustine Wafers (CWs) marketed as Gliadel^®, are biodegradable wafers that are implanted in the surgical site during glioblastoma resection, and they release carmustine locally. They were approved by the FDA for recurrent GBM and malignant glioma in 1996 and 2003 respectively (32). Their use is currently limited and remains a controversial topic among neurosurgeons due to the potential side effect of pulmonary fibrosis, and lack of significant evidence on its impact on the quality of life, infections after surgery and possibility of adjuvant therapy (33). In a recent meta-analysis study, Ricciardi et al. (2022), evaluated the OS and progression-free survival (PFS) in newly high-grade glioma patients that received intraoperative implantation of CWs, and concluded that CWs can significantly improve OS, but patients must be carefully selected based on their age and tumor volume to minimize side effects (34).

Lomustine, also known as CCNU (chloroethyl-cyclohexyl-nitrosourea), is a monofunctional alkylating agent of the nitrosourea family, that alkylates DNA and RNA, which triggers cancer cell death through DNA and RNA cross-linking (35). It is lipid-soluble and thus can successfully cross the BBB. It was FDA-approved for the treatment of brain tumors in 1976, and it is still being widely used for recurrent and progressive glioblastoma, administered orally in 6 to 8 weeks intervals (36, 37). Lomustine is given as monotherapy or in combination with procarbazine and vincristine (PCV regime), and it is considered safe with well-controlled side effects (37, 38). CCNU is increasingly considered as the standard of care option for recurrent glioblastoma, as no other treatment has demonstrated superior outcomes in controlled clinical trials (39).

Bevacizumab, also known as Avastin^®, is a human recombinant monoclonal antibody to vascular endothelial growth factor (VEGF), a signal protein central to angiogenesis, that has been used for the treatment of several cancer types, both in monotherapy and in combination therapies (40). In 2009, bevacizumab was FDA-approved for recurrent glioblastoma following two Phase II trials, in which it was given in combination with irinotecan, based on its safety and improvement of quality of life (41, 42). In a recent scoping review (43), PFS benefits, and well-controlled side effects were supported for recurrent glioblastoma from bevacizumab, but no benefits in OS were identified. In fact, the European Medicines Agency (EMA) has rejected the use of bevacizumab for treating recurrent GBM, due to the lack of positive benefit-risk (44). It was suggested that combining bevacizumab with other therapies, like TTFields, might improve therapeutic outcome, but more studies are needed (43).

Optune^® device, also known as TTFields, made by NovoCure is a portable wearable device, that was initially approved in 2011 for treating patients with recurrent glioblastoma and in 2015 for newly diagnosed glioblastoma (45). TTFields are alternating electric fields that disrupt cancer cell replication both in vitro and in vivo ( 46). TTFields combined with TMZ have shown significantly improved OS and PFS, however, they have not been adopted as standard care due to several factors such as high cost and inconvenience (47).

In general, the ECM compromises approximately 20% of the brain mass, and under physiological conditions, it provides structural and biochemical support, as well as regulation of various cellular processes (48). The major components of the ECM, including hyaluronic acid (HA), tenascin-C (TNC), fibronectin (FN), laminin, and collagen, among others, play a vital role in the modulation of invasiveness. In glioblastoma, remodeling, and degrading of the ECM is observed since Matrix metalloproteinases (MMPs), more specifically MMP-2 and MMP-9, are released into the extracellular space. Tenascin-C (TNC), another ECM component, is a matricellular protein (MCP) that is normally expressed at low levels; however, during development and pathological conditions it is highly expressed (49, 50). Glioblastoma cells produce and release TNC and high TNC expression is associated with poor patient survival and disease progression. TNC can interact with multiple proteins (e.g. fibronectin, Toll-like receptors etc.) and thus can promote neovascularization, proliferation, invasiveness, and immunomodulation. Additionally, the presence of TNC stimulates GSCs invasiveness by MMP12 and ADAM metallopeptidase domain 9 (ADAM9) expression and activity, via the c-Jun NH2-terminal kinase pathway. Another ECM component is HA, which activates CD44, a cell surface adhesion protein, stimulating the synthesis and secretion of additional HA, leading to upregulation of MT1-MMP, thus promoting glioblastoma cell infiltration (48). Moreover, FN glycoprotein, also highly expressed in glioblastoma, promotes cell adhesion, differentiation of GSCs, and invasion, and plays a role as a coordinator between ECM and glioblastoma cells (51). Activation of the adhesion kinase/paxillin/Akt signaling pathway is responsible for GSCs adherence and differentiation, while the increase of MMPs’ activity and activation of axis Stat3-ODZ1-RhoA/ROCK, could be responsible for the invasive behavior observed in glioblastoma. Laminin glycoprotein, more specifically laminin-2, -5, and -8, are also found to be highly expressed in glioblastoma patients and they are suspected to have a role in glioblastoma spreading and infiltration (48). Lastly, collagen type I is normally present at low levels in brain tissue, but its expression is slightly elevated in glioblastoma tumors. It is enhanced in the perivascular niche of GSCs, promoting therefore invasiveness via integrin and phosphatidylinositol 3-kinase (PI3K)/Akt signaling pathways.

Physiological factors, such as pH and oxygen concentrations, can affect tumor progression, and immunosuppression (48). The intra-tumoral heterogeneity displayed in glioblastoma alters the nutrient supply and availability of oxygen within the tumor, influencing metabolic properties and energy utilization of cancer cells.

Immune cells found in glioblastoma TME constitute up to 50%; glioblastoma-associated macrophages and microglia (GAMs) are the most prevalent, followed by neutrophils, regulatory T cells (Tregs), dendritic cells (DCs), and myeloid-derived suppressor cells (MDSCs) (54). The immune components present in glioblastoma contribute to the observed immunosuppressive characteristics of the glioblastoma TME.

GSCs possess the ability to self-renew and differentiate, contributing to intra-tumoral heterogeneity and playing a key role in tumorigenesis and tumor propagation (48, 54, 66). GSCs activate, regulate, and recruit pro-tumor immune cells. They inhibit T-cell proliferation and cytotoxic T-cell activation while suppressing macrophage-mediated tumor-killing by producing cytokines such as IL-10 and TGFβ. Additionally, GSCs express the TGFβ receptor (TGF-βRII) on their surface, and binding of its ligand triggers the secretion of MMP9 by GSCs. GSCs also interact closely with ECs, creating a perivascular niche, and impacting the glioblastoma progression. Subsequently, GSCs express high levels of proangiogenic growth factors such as VEGF, angiopoietin-1 (Ang-1), bradykinin (BK), IL-8, and stromal cell-derived factor-1 I (SDF-1), which induce their differentiation into ECs and pericytes, further enhancing angiogenesis, migratory abilities, and invasiveness.

PD-1 receptor, expressed on T- and other immune cells, is a dominant negative regulator of T-cell response, when activated by its ligand, PD-L1, which is expressed on tumor cells (117). Anti-PD-1/PDL-1 therapy has been approved for the treatment of several cancer types, including metastatic melanoma, non-small cell lung cancer (NSCLC), head and neck cancers, urothelial carcinoma and others (118). More specifically, there are 6 FDA-approved PD-1/PD-L1 inhibitors, including the PD-1 inhibitors Pembrolizumab (Keytruda), Nivolumab (Opdivo) and Cemiplimab (Libtayo) and the PD-L1 inhibitor Atezolizumab (Tecentriq). PD-1/PDL-1 therapy has been explored in several clinical studies in glioblastoma, and although it is safe, it did not prolong OS (18). Currently, there are no FDA-approved anti-PD-1/PD-L1 inhibitors for glioblastoma (119), but the PD-1 inhibitors Pembrolizumab, Nivolumab, Cemiplimab, Retifanlimab by Zynyz (approved under FDA’s Accelerated Approval Program for Merkel cell carcinoma), and Balstilimab (currently in Phase II clinical trials for several cancers), as well as the PD-L1 inhibitor, Atezolizumab, are being tested in clinical trials for safety, tolerability, feasibility and efficacy in newly diagnosed, recurrent and progressive glioblastoma, both as monotherapy and as part of a therapeutic regime ( Table 1 ).

CTLA-4 belongs to the superfamily of CD28–B7 immunoglobulins and it shares its two ligands (B7.1, B7.2) with its co-stimulatory counterpart CD28, and together these molecules are functioning at the tip of the immunological cascade (120). In glioblastoma, CTLA-4 competes with CD28 for binding to costimulatory molecules (CD80 and CD86) on APCs, thereby inhibiting the activation of T cells. Anti-CTLA-4 as a single form of therapy or in combination with other ICIs, enhances endogenous immune responses to immunogenic tumors. Ipilimumab (MDX-010 and Yervoy^®) is a humanized monoclonal CTLA-4 antibody that has been approved by the FDA as monotherapy in anti-PD-1 refractory cases or in combination with nivolumab as a first-line treatment of advanced melanoma (121). In glioblastoma clinical trials, ipilimumab is administered in combination with nivolumab (NCT06097975, Table 1 ). Botensilimab (AGEN1181), is an Fc-enhanced anti-CTLA-4 antibody, which is considered one of the most advanced-next generation ICIs currently in clinical trials, due to its novel FcyR-dependent mechanism to promote superior priming and activation of T cells (121). In glioblastoma, Botensilimab is being tested in a single clinical trial in combination with Balstilimab and chemotherapy, given through a BBB sonication device (NCT05864534, Table 1 ).

TIM-3 is an immuno-myeloid cell surface marker specific to IFN-γ producing CD4+ and CD8+ T cells, expressed on multiple immune cells and leukemic stem cells (122). In cancer, elevated TIM-3 expression is associated with poor outcome, therefore TIM-3 has become an attractive candidate for immunotherapy. Sabatolimab (MBG453), is a novel high-affinity, humanized, IgG4 antibody targeting the TIM-3 receptor currently under clinical development by Novartis for the treatment of both solid tumors and hematological malignancies (123). Sabatolimab, is tested for safety in a Phase I clinical trial of recurrent glioblastoma, in which is given in combination with spartalizumab (anti-PD-1) (NCT03961971, Table 1 ).

TIGIT belongs to the Ig superfamily, and it is expressed on activated CD4+ and CD8+ T cells, as well as NK cells (124–126). TIGIT interacts with its ligand, CD155, which is expressed mostly on DCs and macrophages, and upon TIGIT/CD155 interaction, immune responses are negatively regulated. Specifically, T-cell receptor expression is reduced, resulting in impairment of the function of CD8+ and NK cells, resulting in immunosuppression. Currently, Domvanalimab, an investigational inhibitor, is being evaluated in clinical trials in combination Zimberelimab (anti-PD-1) in recurrent glioblastoma (NCT04656535, Table 1 ).

Despite extensive research, there is still no ICI approved for the treatment of glioblastoma, since no significant improvement in OS has been observed during clinical trials so far. Despite the disappointing early clinical trial results, the use of ICIs for the treatment of glioblastoma remains under ongoing clinical investigation, aiming to reverse the glioblastoma immunosuppressive TME which is mainly attributed to the upregulation of several immune checkpoint molecules, such as PD-1, PD-L1, CTLA-4, LAG-3, TIM-3 and TIGIT. A variety of ICIs are currently being investigated in clinical trials in glioblastoma (early, aggressive or recurrent), for their safety and effectiveness, often in combination with other therapies to overcome the limitations of monotherapy and improve therapeutic outcomes. The use of novel delivery systems in several trials, like the Optune^® device, suggests a commitment in overcoming delivery limitations especially the BBB. Furthermore, combining ICIs with personalized neoantigen vaccines aims to enhance T-cell activation and specificity. In addition, targeting myeloid cells aims to enhance antigen presentation. Importantly, discovery of new biomarkers to predict response to ICI therapy is essential in GMB; for example, in other types of cancer, patients with low PDL-1 levels also benefit from anti-PD-1 therapy, suggesting other mechanisms are involved in their action (127). The lack of reliable biomarkers for the use of ICIs in GMB is discussed in recent reviews (128, 129). The future of ICIs in glioblastoma, depends on the delivery systems, the drug combination strategies, the tackling of the cold glioblastoma immune microenvironment, and the identification of personalized biomarkers based on molecular signatures or immune profiles that could help tailor ICI therapies to responsive subpopulations of patients. Another important factor is optimizing the timing of ICI administration. For instance, recent findings suggest that administering combination ICI in the neoadjuvant setting can stimulate the infiltration, activation, and proliferation of tumor-specific T cells in patients with newly diagnosed glioblastoma (130). In fact, a clinical trial, based on this, is already ongoing (NCT06816927).

CAR-T cell therapy is at the forefront of T-cell based therapies, providing a powerful tool for cancer treatment (131–133). It involves the inducible expression of a chimeric antigen receptor (CAR), engineered to target a specific antigen of interest on autologous T cells, hence bypassing the need for antigen presentation by Major Histocompatibility Complex (MHC) otherwise required for the activation of endogenous T cells (131–134). Since 2017, six CAR-T cell products have been approved for the treatment of several cancers, including acute lymphoblastic leukemia (ALL), diffuse large B-cell lymphoma (DLBCL), refractory follicular lymphoma (FL), recurrent or refractory mantle cell lymphoma (MCL), and relapsed or refractory multiple myeloma (133). Currently there are no FDA-approved CAR-T cell therapies for glioblastoma, but several clinical trials (mostly in Phase I) are evaluating CAR-T cell therapy in glioblastoma ( Table 2 ; Figure 2B ), including one that is combining CAR-T cell therapy with ICIs (NCT04003649). The glioblastoma antigens currently targeted in the clinic, often in combination, are Cluster of Differentiation proteins (CD133, CD44 & CD70), IL receptors (IL-13Ra2), EGFR and EGFRvIII, ephrin receptor A2 (EphA2), human epidermal growth factor receptor 2 (HER2) and B7-H3 ( Table 2 ).

IL-7 is a hematopoietic cytokine that promotes the activation, differentiation and homeostasis of naïve T-cells, as well as the survival, expansion and proliferation of memory T-cells (135). It has been shown that engineered T-cells constitutively expressing the IL-7 receptor alpha (IL-7Rα) have great antitumor efficacies in both breast cancer (136) and glioblastoma models (137), therefore, IL-7 and its receptor are great candidates for immunotherapy. For the clinical trial NCT05577091 ( Table 2 ), autologous T-cells have been genetically modified to express a CAR targeting CD133 and CD44, and a truncated form of the IL-7Rα. These CAR-T cells are believed to have immunostimulating and anti-neoplastic activities since they target CD133 and CD44, two markers of GSCs, associated with the proliferative or invasive state of glioblastoma cells (138). Also, the IL-7Rα-Tris-CAR-T cells induce selective toxicity to tumor cells, and the IL-7/IL-7Ra-mediated signaling promotes the proliferation and survival of T cells.

IL-8 is another important chemokine, that coupled with its receptor, IL-8R, play a role in tumor invasion, proliferation, survival and angiogenesis, as well as in the promotion of the malignant properties of the glioblastoma stem cells (139–142). CD70, is an antigen that is overexpressed in glioblastoma and is associated with poor survival (143). It was also hypothesized that it correlates with a mesenchymal phenotype and immunosuppression via recruitment of macrophages and CD8+ T-cell death. This information on IL-8 and CD70, let to the generation of CD70-targeting CAR with a modified IL-8 receptor (called 8R-70CAR) that let to complete tumor regression of advance cancers in pre-clinical studies, including glioblastoma (144). (NCT05353530 & NCT06946680, Table 2 ).

IL-13Rα2 is a high affinity membrane receptor that is overexpressed in glioblastoma, and is associated with poor outcome, mesenchymal gene profile, immunity, and the tumor microenvironment, which make it an important therapeutic target (145, 146). First-generation of IL-13Rα2-targeted CAR-T cells showed evidence of antitumor efficacy, but limited persistence of T-cells (147), therefore new approaches were explored in the design and production of IL-13Rα2-targeted CAR-T cells. Brown et al. (2018) (148), designed a second-generation of IL-13Rα2-targeted CAR, by engineering memory-enriched T cells to express IL-13Rα2 and 41BB-constimulatory CAR, a member of the tumor necrosis factor (TNF) receptor superfamily that enhances CAR-T survival and persistence (149, 150). Furthermore, the engineered T-cells expressed a truncated form of CD19, a pan B-cell marker (151) that has been targeted for the treatment of several hematological malignancies (152), and was shown to be highly expressed in brain and endothelial cells causing increased BBB permeability and neurotoxicity after targeted CAR-T immunotherapy (153, 154). These second-generation CAR-T cells not only improved anti-tumor activity but also T-cell persistence. Furthermore, it was shown that when delivered intracranially, they had a greater anti-tumor effect compared to intravenous delivery in orthotopic glioblastoma mouse models. This study, led to the protocol design for the NCT02208362 trial ( Table 2 ), in which multiple intracranial infusions of these CAR-T cells, were not only proven to be safe, but also let to the regression of glioblastoma, increased levels of immune cells and cytokines, and persistence of response for up to 7.5 months after treatment initiation (155). Results from this trial, demonstrated that locoregional therapy with IL-13Rα2-targeted CAR-T is safe with promising clinical activity in a subpopulation of patients (88). Later, Chang Xu et al. (2022) (156), developed the first humanized third-generation CAR-targeting IL-13Rα2 that showed great anti-tumor efficacy and reduced expression of immunosuppressive cytokines such as IL-6 (88, 157) Like second and third generation IL-13Rα2-targeted CAR-T cells, HER2-specific, 41BB-costimulatory CAR with a truncated CD19 have been engineered and are being assessed in recurrent or non-responsive glioblastoma (NCT03389230, Table 2 ). HER2 is another important target, since it was found to be overexpressed in glioblastoma and is associated with poor prognosis (158).

EGFRs are transmembrane receptors, part of the ErbB family of receptor tyrosine kinases (RTKs), activated by several ligands (e.g. TGFα) binding to the extracellular domain (ECD) (159). Ligand-receptor formation of homo- or hetero-dimers, leads to the activation of several downstream pathways (e.g. MAPK, STAT3 and PI3K) that regulate cell survival, proliferation, angiogenesis and migration. When EGFRs are overexpressed or mutated, they stay constitutively active, which leads to uncontrolled cell proliferation and therefore tumor progression. EGFR is overexpressed in ~60% of primary glioblastoma and 10% of secondary glioblastomas (160). Furthermore, several mutants have been found in glioblastoma, with the most common being EGRvIII (159, 160). Overexpression and mutations in EGFR lead to a more aggressive glioblastoma phenotype and increased tumor heterogeneity, therefore several strategies have been developed to target EGFR ( Table 2 ) (87, 89, 161, 162) B7-H3 (also known as CD276) is a type I transmembrane protein, that expressed in >70% of glioblastoma patients (163, 164), and is associated with progression, metastasis, poor outcome and immune evasion (165), as it is an immune checkpoint molecule expressed on antigen-presenting cells. The antitumor efficacy of B7-H3 CAR-T cells in glioblastoma in vitro and in vivo models was first reported in Xing Tang et al. (2019) (166). Several clinical trials, currently in recruiting phase, in refractory or recurrent glioblastoma, are evaluating the safety, feasibility, dose and efficacy of B7-H3 CAR-T cells (NCT05835687, NCT06482905, NCT05366179 & NCT05474378 Table 2 ). In NCT05241392 ( Table 2 ), the preliminary data were recently published and showed that the use of B7-H3 CAR-T cells is safe and tolerated and holds great promise toward improving patients’ overall survival (167). Regarding the pharmacokinetic profile of B7-H3 CAR-T cells, a study reported that local delivery led to cerebrospinal fluid (CSF) persistence and localized immune activation rather than systemic effects (168). It is worth noting, that in a recent pre-clinical study, B7-H3 CAR-T cells consisting of IL-7Rα, have been shown to suppress tumor growth and prolong overall survival in glioblastoma mouse models (169), suggesting the potential implication of B7-H3-IL-7Rα CAR-T cells in the clinic.

Finally, chlorotoxin (CLTX) is a small 36-amino-acid peptide purified from the venom of the scorpion Leiurus quinquestriatus (170). CLTX was initially characterized based on its inhibition of glioma-specific chloride ion channels (GCC), however recent studies identified MMP-2 as the principal receptor for CLTX on the surface of glioblastoma cells. As mentioned above, MMP-2 expression is increased in glioblastoma TME, contributing significantly to the tumor invasiveness. CLTX has demonstrated specific and selective binding to membrane-bound MMP-2 and minimal binding to normal brain tissue, therefore CAR-T cells engineered to incorporate CTLX as an antigen recognition domain are considered a promising approach for MMP-2 positive glioblastoma treatment, redirecting cytotoxic T cells towards glioblastoma cells (171).

(γδ)T-cells comprise a unique type of innate immune T cells that express a γδ T cell receptor (TCR), and are found abundant in several tissues, including lymphocytes that infiltrate solid tumors (172, 173). They can directly kill tumor cells through (a) cytokine-mediated cytotoxicity (e.g. TNFα and IFNγ), (b) perforin (PFN) & granzyme (GzmB) and FasL & TRAILR mediated target cell apoptosis, (c) antibody-dependent cell-mediated cytotoxicity, and (d) antigen processing and presentation. They can also have indirect anti-tumor effects through interaction with immune cells (e.g. NK cells and αβ T-cells). In addition, they have no autologous limitations, can be derived from healthy donors, and can be easily expanded. These properties make them a powerful tool in immunotherapy. The first therapeutic mechanism involves the selective amplification of (γδ)T-cells in vivo using antibodies or bisphosphonate antigens. In the second mechanism, which is adoptive cell therapy, tumors are treated with allogeneic (γδ)T-cells (natural or genetically engineered) or (γδ)T-cells that have been expanded in vitro. The first-in-human Phase I clinical trial (NCT04165941, Table 2 ) involves the intracranial infusion of TMZ-resistant (γδ)T-cells in newly diagnosed glioblastoma patients. During TMZ treatment, the natural killer group 2, member D (NKG2D) receptor ligands (NDG2DL), the master activators of NK cells (174), are upregulated, but this immune response is impaired due to lymphodepletion (172, 175). Therefore, genetic modification of (γδ)T-cells, naturally expressing the NKG2D receptor, were genetically modified and expanded ex vivo with an MGMT-expressing lentivector that provided resistance to TMZ, allowing therefore the simultaneous infusion of (γδ)T-cell with chemotherapy, and targeting therefore the tumor when NKG2DL are maximally expressed ( Figure 2C ). Modifying the cells to be resistant to TMZ is crucial, as they need to be able to resist the cytotoxic effects of TMZ and be able to survive and function on the presence of chemotherapy. In another trial, gene modified allogeneic or autologous (γδ)T-Cells (DeltEx), again conferring TMZ resistance, are evaluated for safety, tolerance and ability to delay recurrence in newly or recurrent glioblastoma (NCT05664243, Table 2 ; Figure 2C ).

The use of CAR-T cells in glioblastoma is based on targeting the tumor-specific or tumor-enriched antigens like IL-13Rα2, EGFR/EGFRvIII, HER2, CD70, B7-H3, and CD133/CD44. Apart from antigen-specificity, the use of innovative constructs in active clinical trials, such as next-generation CAR-T that integrate co-stimulatory domains, can improve T-cell persistence, activation, and tumor selectivity. In some trials, locoregional delivery of CAR-T cells through intracranial administration aims to enhance the efficacy and reduce systemic toxicity, although is more invasive and complex. Since glioblastoma is characterized by profound heterogeneity that limits the durability of response due to downregulation of target antigens, CAR-T cells that target multiple antigens have been developed that aim to reduce escape via antigen heterogeneity. (γδ)T-cells use a different strategy from CAR-T, exploiting innate-like immune response. Based on their recognition of ligands independently of MHC, they are ideal for glioblastoma that is immune-evasive. Another advantage is that they are multifunctional, since use different mechanisms for killing tumor cells. On the other hand, the safety and efficacy of (γδ)T-cells is still not well established and they also require complex genetic engineering. Although highly innovative, both CAR-T and (γδ)T-cells need to overcome antigen escape, and the suppressive microenvironment. Moving forward, further clinical exploration is needed, and their success will depend on effectively reshaping the TME.

Peptide vaccines are composed of 8–30 amino acid including tumor-specific or tumor-associated antigens that elicit an anti-tumor T cell response (177, 178). As they are not highly immunogenic, they can be combined with other forms of immunotherapy or chemoradiation. They can be individualized, single-targeted or multi-targeted. One target of peptide vaccines is the survivin protein; a member of the inhibitor of apoptosis (IAPs) family, that is overexpressed in glioblastoma and is associated with poor prognosis (179). The survivin peptide vaccine, SurVaxM, is currently being evaluated in two clinical trials in combination with TMZ (NCT02455557 & NCT05163080, Table 3 ). Preliminary results showed that SurVaxM is safe and well-tolerated and its combination with TMZ is very promising (96). pp65 and survivin are also being targeted together with EphA2 through a multi-peptide vaccine (ETAPA I) in HLA-A*0201 positive patients with a newly diagnosed, unmethylated, and untreated glioblastoma (NCT05283109, Table 3 ). pp65 is overexpressed in high grade gliomas and medulloblastomas, but not in adjacent brain, and plays a significant role in glioblastoma progression. When tested in children and young adults, it was proven to be well-tolerated and promoted antigen-specific immune responses (180). Other targets include Telomerase Reverse Transcriptase (TERT), Protein tyrosine phosphatase receptor type Z1 (PTPRZ1) and Toll-like receptors (TLRs), that are being targeted in combination in NCT06622434 ( Table 3 ). PTPRZ1 is a clinically relevant antibody in glioblastoma associated with stemness (181), and telomerase (TRT) is a major oncogene, whose promoter is mutated in approximately 80% of glioblastoma patients and is associated with tumor progression (96, 182, 183). TLRs are ubiquitously expressed receptors that recognize pathogens and lie at the first line of defense in the innate immune system (184). In several tumor types, upon ligand recognition, TLRs activate downstream intracellular signaling pathways either supporting or suppressing tumor growth, thus they are a great candidate for immunotherapy. Finally, several autologous or allogeneic multipartite vaccines, designed to induce a variety of neoantigen-specific immune responses, are tested in clinical trials in combination with other therapies, such as ICIs and nucleic acid vaccines (NCT03223103 & NCT02287428, Table 3 ).

DC are the most superior APC cells of the immune system, thus playing a vital role in presenting antigens in the lymph nodes eliciting T-cell priming and distant anti-tumor response (185, 186). DC vaccines are generated by culturing hematopoietic progenitor cells or monocytes ex vivo in the presentation of a cytokine cocktail to induce their maturation. Following maturation, DCs are loaded into tumor antigens and subsequently injected into patients. Not only they can achieve priming of CD4+ T cells by peptide-MHCII complex, but also, they elicit CD8+ T-cell antitumor responses. Several clinical trials are underway in glioblastoma investigating the safety and efficacy of DC vaccines in newly diagnosed or recurrent glioblastoma ( Table 3 ). Some are loaded with autologous tumor lysates (NCT03395587, NCT04523688 & NCT04801147, Table 3 ), and even “double-loaded” (NCT06805305), and others are loaded with multiple tumor neoantigens (NCT04968366 & NCT06253234, Table 3 ). They are administered either alone or in combination with chemoradiation or ICIs. Finally, other DC vaccines are loaded with mRNAs for specific proteins, including Survivin/hTERT derived from autologous GSCs (NCT03548571, Table 3 ), the tumor-associated antigen Wilm’s tumor 1 (WT1) (NCT02649582, Table 3 ) and the human CMV immunodominant protein pp65-LAMP (NCT03688178, Table 3 ).

Other than the DC and peptide vaccines, a few other vaccines are being exploited in glioblastoma immunotherapy trials based on the delivery of nucleic acids ( Table 3 ). Nucleic acid-based vaccines introduce a segment of DNA or RNA that encodes a specific or multiple tumor antigens, to elicit and immune response (176). The current vaccines clinical trials include DNA vaccines (NCT05698199, & NCT04015700), an mRNA vaccine (NCT05938387), and two vaccines that use lipid nanoparticle technology (187) to deliver RNA (NCT04573140 & NCT06389591). Apart from the mRNA vaccine that encodes several GBM peptides (188), all the others are targeting the immunogenic and viral antigens of CMV.

Cancer vaccines are a growing field in glioblastoma immunotherapy, and they hold a great promise, since they aim to evoke tumor-specific immune response through tumor associated or neoantigen presentation. In addition, they have shown great tolerability and minimal toxicity in early-phase trials (e.g. SurVaxM). Furthermore, they allow personalized and/or multi-antigen targeting to address the heterogeneity present in glioblastoma. Especially DC vaccines, that are loaded with autologous tumor lysates, offer T-cell priming, overcoming the limitations of peptide vaccines (priming of weak immune response). The efficacy of vaccines might be limited the high infiltration of Tregs, and antigenic heterogeneity, therefore combining vaccines with other immunotherapies (CAR-T cells, ICIs and oncolytic viruses) may further improve immunogenicity and efficacy. Future progress hinges on optimizing vaccine formulations for better delivery, refining antigen targets, and leveraging synergies with other immunotherapies.