Translational advancements in tumor vaccine therapies for glioblastomas

Abstract

Glioblastoma (GBM) presents significant therapeutic challenges due to the limited efficacy of current treatments. This resistance is multifactorial, stemming from tumor heterogeneity, an immunosuppressive tumor microenvironment, and the restrictive blood-brain barrier, which limits therapeutic access. In response, immunotherapies, particularly tumor vaccines, have emerged as strategies to harness the immune system against these tumors. This review provides an overview of recent advancements and notable clinical trials in tumor vaccine development for GBM. Additionally, it discusses recent preclinical advancements focused on enhancing immune recruitment and response. Identified strategies include peptide, cellular, and nucleic acid vaccines targeting tumor-specific antigens to induce antitumor T-cell responses. Clinical data and preclinical studies exploring various vaccine candidates, adjuvants, and delivery methods demonstrate encouraging results, with some showing improved progression-free and overall survival rates. Despite these advancements, it is clear that further research into personalized vaccines and combination therapies is necessary to enhance immune responses and improve clinical outcomes.

Key Points.

• Vaccines are a promising treatment for glioblastoma.

• Personalized vaccines and combination therapy might enhance efficacy.

Gliomas are the most common malignant primary brain tumors, accounting for approximately 80% of all malignant primary CNS tumors.^1,2 Among these gliomas, IDH-wildtype glioblastoma (GBM) is associated with very poor outcomes; the median overall survival is approximately 10–12 months, with 5% to 10% survival 5 years postdiagnosis, even with optimal standard of care therapies.^1,3 Only 3 therapeutics approved by the U.S. Food and Drug Administration (FDA) have demonstrated a survival benefit: carmustine wafer implants (Gliadel),⁴ temozolomide and tumor-treating field (TTF) devices.^5,6 However, (Gliadel) is costly and has shown high complication rates with only slight improvements in overall survival (OS).⁷ Consequently, the current standard of care for newly diagnosed GBM is centered on maximally safe surgical resection, followed by radiation therapy and temozolomide, with or without TTF.^5,8 Nevertheless, recurrence occurs in nearly all patients, typically at around 7 months posttreatment. Recurrent tumors exhibit increased resistance to treatment, resulting in an even poorer prognosis.⁹

The resistance of GBMs to treatment, including chemotherapy, radiotherapy, and immunotherapy, is attributed to several factors, such as their substantial molecular heterogeneity and the tumor microenvironment. The blood-brain barrier (BBB) further impedes the efficacy of treatments, contributing in part to the high recurrence rate.^10,11 Traditional chemotherapeutics and molecular therapies are only partially and transiently effective against certain subpopulations of the tumor.^12,13 This leads to negative selection for resistant subpopulations and phenotype switching, which further fosters therapy resistance.¹⁴ Additionally, these gliomas are associated with both local and systemic immunosuppressive states, with various mechanisms and pathways acting to prevent robust immune cell infiltration.¹⁵ Notably and relevant to the presentation of tumor antigen(s), endogenous dendritic cells (DC) comprise less than 1% of the total cellular composition surrounding GBM.^15–17

Due to the limited success of current standard-of-care therapies, other therapeutic avenues are being explored for GBM treatment, including immune-targeted therapies. These therapies aim to harness the body’s immune system to selectively eliminate tumor cells by generating a tumor-specific response.¹⁸ Immune therapeutics for cancer fall into several categories, including immune checkpoint inhibitors, T-cell-targeted therapies, oncolytic viruses (OVs), cytokine-mediated therapies, gene therapies, and therapeutic vaccines.¹⁹ We will focus this review primarily on therapeutic vaccines and discuss these grouped by the vaccine type. Tables 1–3 provide an overview of notable clinical trials utilizing tumor vaccine therapies for GBM, some of which we will discuss in further detail.

Table 1.

Notable Phase I Clinical Trials Utilizing Tumor Vaccine Therapies for High-Grade Gliomas

Trial name	Name	Sponsor	Study start date	Completed/expected completion date	Enrollment	Phase	Status
NCT00639639	Vaccine Therapy in Treating Patients With Newly Diagnosed Glioblastoma Multiforme (ATTAC)	Duke University	2006-02-06	2022-06-01	42	Phase I	COMPLETED
NCT00626483	Basiliximab in Treating Patients With Newly Diagnosed Glioblastoma Multiforme Undergoing Targeted Immunotherapy and Temozolomide-Caused Lymphopenia (REGULATe)	Duke University	2007-04-24	2016-07-06	34	Phase I	COMPLETED
NCT01621542	Clinical Study of WT2725 in Patients With Advanced Malignancies	Sumitomo Pharma America, Inc.	2012-07-31	2017-05-17	64	Phase I	COMPLETED
NCT01957956	Vaccine Therapy and Temozolomide in Treating Patients With Newly Diagnosed Glioblastoma	Mayo Clinic	2013-11-11	2016-11-16	21	Phase I	COMPLETED
NCT02010606	Phase I Study of a Dendritic Cell Vaccine for Patients With Either Newly Diagnosed or Recurrent Glioblastoma	Cedars-Sinai Medical Center	2014-01-08	2021-07-10	39	Phase I	COMPLETED
NCT02149225	GAPVAC Phase I Trial in Newly Diagnosed Glioblastoma Patients	Immatics Biotechnologies GmbH	2014-10	2018-06	16	Phase I	COMPLETED
NCT02287428	Personalized NeoAntigen Cancer Vaccine w RT Plus Pembrolizumab for Patients With Newly Diagnosed GBM	Dana-Farber Cancer Institute	2014-11	2026-06	56	Phase I	RECRUITING
NCT02287428	Personalized NeoAntigen Cancer Vaccine w RT Plus Pembrolizumab for Patients With Newly Diagnosed GBM	Dana-Farber Cancer Institute	2014-11	2026-06	56	Phase I	RECRUITING
NCT02498665	A Study of DSP-7888 Dosing Emulsion in Adult Patients With Advanced Malignancies	Sumitomo Pharma America, Inc.	2015-11	2018-09	24	Phase I	COMPLETED
NCT02924038	A Study of Varlilumab and IMA950 Vaccine Plus Poly-ICLC in Patients With WHO Grade II Low-Grade Glioma (LGG)	University of California, San Francisco	2017-04-03	2030-12-31	14	Phase I	ACTIVE, NOT RECRUITING
NCT03223103	Safety and Immunogenicity of Personalized Genomic Vaccine and Tumor Treating Fields (TTF to Treat Glioblastoma	Albert Einstein College of Medicine	2018-03-01	2025-05	13	Phase I	ACTIVE, NOT RECRUITING
NCT03299309	PEP-CMV in Recurrent MEdulloblastoma/Malignant Glioma (PRiME)	Duke University	2018-06-29	2025-04	30	Phase I	ACTIVE, NOT RECRUITING
NCT04015700	Neoantigen-based Personalized DNA Vaccine in Patients With Newly Diagnosed, Unmethylated Glioblastoma	Washington University School of Medicine	2020-07-14	2024-12-31	9	Phase I	ACTIVE, NOT RECRUITING
NCT04573140	A Study of RNA-lipid Particle (RNA-LP) Vaccines for Newly Diagnosed Pediatric High-Grade Gliomas (pHGG) and Adult Glioblastoma (GBM) (PNOC020)	University of Florida	2021-10-26	2027-07	28	Phase I	RECRUITING
NCT05283109	ETAPA I: Peptide-based Tumor-Associated Antigen Vaccine in GBM (ETAPA I)	Duke University	2023-08-30	2028-02	36	Phase I	RECRUITING
NCT05743595	Neoantigen-based Personalized DNA Vaccine With Retifanlimab PD-1 Blockade Therapy in Patients With Newly Diagnosed, Unmethylated Glioblastoma	Washington University School of Medicine	2023-10-27	2027-10-31	12	Phase I	RECRUITING

Table 3.

Notable Phase III Clinical Trials Utilizing Tumor Vaccine Therapies for High-Grade Gliomas

Trial name	Name	Sponsor	Study start date	Completed/expected completion date	Enrollment	Phase	Status
NCT00045968	Study of a Drug [DCVax-L] to Treat Newly Diagnosed GBM Brain Cancer (GBM)	Northwest Biotherapeutics	2006-12	2022-11	348	Phase III	COMPLETED
NCT01480479	Phase III Study of Rindopepimut/GM-CSF in Patients With Newly Diagnosed Glioblastoma (ACT IV)	Celldex Therapeutics	2011-11	2016-11	745	Phase III	COMPLETED
NCT03149003	A Study of DSP-7888 Dosing Emulsion in Combination With Bevacizumab in Patients With Recurrent or Progressive Glioblastoma Following Initial Therapy	Sumitomo Pharma America, Inc.	2017-12-08	2021-08-30	221	Phase III	COMPLETED

Table 2.

Notable Phase II Clinical Trials Utilizing Tumor Vaccine Therapies for High-Grade Gliomas

Trial name	Name	Sponsor	Study start date	Completed/expected completion date	Enrollment	Phase	Status
NCT02465268	Vaccine Therapy for the Treatment of Newly Diagnosed Glioblastoma Multiforme (ATTAC-II)	University of Florida	2016-08-09	2023-11-30	175	Phase II	COMPLETED
NCT02649582	Adjuvant Dendritic Cell-Immunotherapy Plus Temozolomide in Glioblastoma Patients (ADDIT-GLIO)	University Hospital, Antwerp	2015-12	2025-12	20	Phase I/II	RECRUITING
NCT03879512	Autologous Dendritic Cells, Metronomic Cyclophosphamide and Checkpoint Blockade in Children With Relapsed HGG	Wuerzburg University Hospital	2018-02-07	2025-01-31	25	Phase I/II	RECRUITING
NCT02455557	SurVaxM Vaccine Therapy and Temozolomide in Treating Patients With Newly Diagnosed Glioblastoma	Roswell Park Cancer Institute	2015-05-04	2024-12-30	66	Phase II	ACTIVE, NOT RECRUITING
NCT05163080	SurVaxM Plus Adjuvant Temozolomide for Newly Diagnosed Glioblastoma (SURVIVE) (SURVIVE)	MimiVax, LLC	2021-11-18	2024-08-18	247	Phase II	ACTIVE, NOT RECRUITING
NCT02078648	Safety and Efficacy Study of SL-701, a Glioma-Associated Antigen Vaccine to Treat Recurrent Glioblastoma Multiforme	Stemline Therapeutics, Inc.	2014-05	2018-01-22	74	Phase I/II	COMPLETED
NCT04013672	Study of Pembrolizumab Plus SurVaxM for Glioblastoma at First Recurrence	Case Comprehensive Cancer Center	2020-03-19	2024-05	40	Phase II	ACTIVE, NOT RECRUITING
NCT03750071	VXM01 Plus Avelumab Combination Study in Progressive Glioblastoma	Vaximm GmbH	2018-11-21	2022-12-31	30	Phase I/II	ACTIVE, NOT RECRUITING
NCT01920191	Phase I/II Trial of IMA950 Multi-Peptide Vaccine Plus Poly-ICLC in Glioblastoma	University Hospital, Geneva	2013-08	2016-03	19	Phase I/II	COMPLETED
NCT03665545	Pembrolizumab in Association With the IMA950/Poly-ICLC for Relapsing Glioblastoma (IMA950-106)	University Hospital, Geneva	2018-10-25	2023-12-31	18	Phase I/II	ACTIVE, NOT RECRUITING
NCT01280552	A Study of ICT-107 Immunotherapy in Glioblastoma Multiforme (GBM)	Precision Life Sciences Group	2011-01	2015-12	124	Phase II	COMPLETED
NCT04116658	First-in-Human, Phase 1b/2a Trial of a Multipeptide Therapeutic Vaccine in Patients With Progressive Glioblastoma (ROSALIE)	Enterome	2020-07-13	2025-12	100	Phase I/II	ACTIVE, NOT RECRUITING
NCT00905060	HSPPC-96 Vaccine With Temozolomide in Patients With Newly Diagnosed GBM (HeatShock)	University of California, San Francisco	2009-06-29	2014-06-03	70	Phase II	COMPLETED
NCT01814813	Vaccine Therapy With Bevacizumab Versus Bevacizumab Alone in Treating Patients With Recurrent Glioblastoma Multiforme That Can Be Removed by Surgery	Alliance for Clinical Trials in Oncology	2013-05	2023-05-01	90	Phase II	TERMINATED
NCT03018288	Radiation Therapy Plus Temozolomide and Pembrolizumab With and Without HSPPC-96 in Newly Diagnosed Glioblastoma (GBM)	National Cancer Institute (NCI)	2017-09-21	2022-12-20	90	Phase II	COMPLETED

Tumor vaccines aim to harness and stimulate the intrinsic immune response by activating effector immune cells to recognize and infiltrate tumors through tumor-specific antigens (TSAs), which are unique to tumor cells, or tumor-associated antigens (TAAs), which are expressed in normal tissues but overexpressed in tumor cells.^19,20 Frequently mutated or highly expressed proteins in GBM, such as the TSA epidermal growth factor receptor variant III (EGFRvIII), or TAAs such as telomerase reverse transcriptase (TERT) have been investigated as potential vaccine targets.²¹ The drawbacks of targeting TSAs include substantial patient heterogeneity, and inconsistent immunogenicity and presentation of neoepitopes, thus often requiring personalized (and therefore expensive and time-consuming) development strategies. On the other hand, TAAs are limited by central tolerance, and cross-reactivity with normal tissues can lead to autoimmunity and side effects including brain inflammation.^21,22 By targeting TSAs and/or TAAs, antitumor immune responses can be amplified to induce antitumor T-cell activity and reverse the immunosuppressive tumor environment. The FDA has approved cancer tumor vaccine therapeutics for solid tumors including metastatic melanomas, prostate cancer, and bladder cancer, all of which have shown significant improvements in outcomes.²³ Tumor vaccines rely on both the tumor antigen and an immune adjuvant to elicit a robust response. Successful construction of tumor vaccines requires the selection of delivery platform (including cell, peptide, nucleic acid, or viral-based), appropriate antigen(s), and adjuvants.

Dendritic Cell Vaccines

Most contemporary research and successes with cellular vaccines for GBM rely on DCs. DCs are among the most potent antigen-presenting cells (APCs), known for their ability to robustly induce innate and acquired immunity as well as immunity conversion.²⁴ Specifically, DCs phagocytose, process, and present antigens to T-cells via major histocompatibility complex (MHC) molecules, initiating antigen-specific T-cell responses. It is important to distinguish between 2 major populations of DCs, DC1 and DC2, which are phenotypically and functionally different. While lymphoid-related DC2 activate mainly T_H2 cells and have shown to generate immune-tolerating responses, myeloid-related DC1 present antigens to MHC class I molecules and exert an immunostimulatory function via T_H1 cells.^25,26 This process links the innate and adaptive immune systems. When T-cells are primed and activated, they migrate to the tumor site to exert antitumor effects, effectively transforming immunologically “cold” gliomas into more “hot” sites.^27,28

The goal of DC vaccines is to amplify this natural process by pulsing autologous or allogeneic DCs with tumor lysate to elicit antigen-specific cytotoxic T-cell responses. For patient-derived approaches, autologous DCs are induced with tumor antigen, treated with cytokines to promote maturation, and then prepared for reinjection into the patient.^29,30 The forms of tumor antigens can vary and have included peptides, polypeptides, mRNA, DNA, and whole-tumor lysates.³¹

DC vaccines can be classified as autologous or allogeneic. Autologous vaccines rely on gathering unique TAAs from the patient’s own tumor.³² This approach theoretically offers tumor specificity, although consistent superior vaccine efficacy has yet to be demonstrated. It also involves more resource utilization and challenges in harvesting and producing sufficient dosages for clinical use. Alternatively, allogeneic vaccines are based on shared antigens found in a high percentage of gliomas, potentially offering a more broadly applicable approach.

DCVax-L is a DC vaccine that utilizes an autologous tumor lysate to initiate a robust immune response.³³ The vaccine was evaluated in a phase III clinical trial at multiple sites across multiple countries (NCT00045968), enrolling a total of 331 patients. An initial report from 2018 demonstrated that the median OS was 19.3 months, compared to 16.5 months in a cohort of randomized internal patients and post hoc-introduced synthetic control group.³³ The median OS in patients with recurrent GBM was 13.2 months (compared to 7.8 months in the external control cohort). Furthermore, long-term survival appeared to be improved with a 2-year survival rate of 20.7% (compared to 9.6%) and a 5-year survival rate of 13.0% (compared to 5.7%). The study demonstrated negligible toxicity, with only 5 serious adverse events in over 2000 administered doses. Further analysis suggested that the potential survival benefit was maximal in patients age 65 or older, those with subtotal resection at the time of surgery, and those with increased MGMT promoter methylation.³³ In addition, multiparametric MRI prediction models were developed to assess treatment response.³⁴ Despite these results, the study failed to meet its primary endpoint, PFS, as reported in a second recent report, with a median PFS of 6.2 months (CI: 5.7–7.4) in the DCVax-L group, and 7.6 months (CI: 5.7–7.4) for the placebo group.³⁵ Though the OS data was reported as significant, their analysis relied on post hoc retrospective analysis with cross-trial comparisons. The change of the primary endpoint to OS, raises concern about the validity of the vaccine.³⁶ Thus, clear conclusion of the efficacy of this vaccine on gliomastoma cannot be drawn, and the post hoc nature of these results is largely hypothesis-generating and requires confirmation in further trials.

Peptide Vaccines

Peptide vaccines rely on the injection and presentation of tumor-specific peptide antigens. These short-chain amino acids contain epitopes specific to the tumor, acting as antigenic targets presented to T-cells in lymphoid tissue. Often, these peptides are conjugated to a carrier protein, which increases immunogenicity and strengthens the adaptive immune response to the tumor.^21,37 While single peptides have traditionally been used, recent advances have shifted focus to multi-peptide targets.³⁸

As previously mentioned, GBM is characterized by its tumor heterogeneity and specific genomic mutations, which can lead to antigens that are either unique (TSAs) or expanded in tumors (TAAs) when compared to normal cells. Therefore, these TSAs/TAAs are candidates for eliciting an immune response against tumor cells. Also, single-peptide vaccinations can result in antigen escape, allowing tumors to evade the immune response.³⁹ Strategies to overcome these barriers include multi-peptide proteins and patient-personalized target antigens, which are discussed in further detail below.

Epidermal Growth Factor Receptor Class III variant (EGFR vIII) is a deletion mutation that is heterogeneously expressed in ~1/4–1/3 of GBMs and linked to poor long-term survival, while not found in normal tissue.⁴⁰ Due to the tumor specificity of EGFR vIII, it served as the target of the rindopepimut (CDX-110), a peptide spanning the mutation site of EGFR vIII conjugated to the immune adjuvant hemocyanin.⁴¹ In a phase I clinical trial, the vaccine was shown to eliminate tumor cells while being safe, immunogenic, and tumor-specific.⁴² In a follow-up phase II study with 65 patients, the median OS was 22 months, with a 36-month OS of 26%, with the authors reporting significantly improved survival compared to historical controls. However, the treatment group was a relatively healthy group, especially compared to the historical controls. As a follow-up, rindopepimut was tested in a phase III trial (NCT01480479) with 745 patients. However, the trial was terminated for futility as the median OS did not improve significantly after a preplanned interim analysis. Data from the study showed that a strong anti-EGFR vIII immune response was generated, suggesting that failure to generate a robust immune response was not the primary limitation for the lack of improved survival. Failure of this study may have been due to over selection of patients and comparison of outcomes with potentially inaccurate historical control in the earlier phases of the study. In light of the failure of this study, there have been substantial efforts to identify methods to improve responses. One approach is T-cell bispecific antibodies (TCBs), which bind both EGFRvIII and the T-cell receptor to specifically activate and recruit T-cells.⁴³ Other approaches include tandem chimeric antigen receptor (CAR) T cells that target both EGFR vIII and IL-13Ra2, which has demonstrated some efficacy in in vitro models.⁴⁴ Similarly, bispecific T cell engager (BiTE) antibody constructs, designed to engage CD3+ T cells to EGFR vIII, demonstrated safety tolerance in a phase I trial, though the robustness of its antitumor activity remains to be evaluated in larger studies.⁴⁵

WT1-Based Vaccines

Wilms’ Tumor 1 (WT1) gene is a TAA overexpressed in multiple malignancies, including GBM.⁴⁶ WT2725 is a WT1-derived oligopeptide vaccine designed to include cytotoxic T-cell responses against advanced malignancies known to overexpress WT1.⁴⁷ Several phase I dose-escalation studies (NCT01621542, NCT02498665) in patients with WT1-associated tumors, including GBM, did not identify any dose-limiting toxicities of WT1 vaccines, suggesting they are well-tolerated at therapeutic doses.⁴⁶ Another phase I/II study is currently recruiting, where it aims to evaluate the overall and progression-free survival (PFS) of patients with WT1 mRNA DC vaccination in conjunction with adjuvant temozolomide (NCT02649582). Finally, a large phase III study with 221 patients with GBM is also currently underway, where the DSP-7888 vaccine underwent a dose-escalation trial and is administered in conjunction with bevacizumab (NCT03149003). The interim results suggest that dose-limiting toxicity was not experienced by the cohort, with a larger OS (10.2 [8.2–11.4] months vs. 9.4 [7.4–10.3] months), 1-year OS (37.9 [28.7–47.0]% vs. 31.6 [22.9–40.7]%), and PFS (5.3 [3.9–5.6] months vs. 3.8 [3.7–5.6] months) in patients taking the vaccine, although these differences were not statistically significant.⁴⁸ Analysis of data across these trials has suggested that maintenance of WT1 expression during the vaccination period could be driving the observed trends of increased PFS and OS.⁴⁹

Survivin-Based Vaccines

Survivin is an antiapoptotic protein capable of inhibiting caspase activation and is overexpressed in most cancers, including GBM.⁵⁰ SurVaxM is a peptide vaccine that targets survivin and has been shown to stimulate both T-cell immunity against survivin and inhibit the activity of the survivin pathway.⁵¹ In a phase II clinical trial with 64 patients with newly diagnosed GBM (NCT02455557), SurVaxM was well tolerated, with a median PFS of 11.4 months and a median OS of 25.9 months. A larger phase II trial (NCT05163080) is currently ongoing, with an estimated 247 patients enrolled.

Other clinical trials have aimed to combine SurVaxM with other immune peptides to increase the efficacy of the elicited immune response. SL-701 is a vaccine comprised of synthetic peptides targeting interleukin-13 receptor alpha-2, ephrinA2, and survivin.⁵² In a phase II study with 74 patients (NCT02078648), SL-701 with or without imiquimod and bevacizumab was well tolerated. Results suggested that the median OS was 11 months with imiquimod and 11.7 months with bevacizumab, with a disease control rate of 22% and 54%, respectively.⁵² Another phase I trial is evaluating the efficacy of a combination of survivin vaccination in conjunction with CMV pp65 and P30-linked EphA2 (NCT05283109). It has enrolled an estimated 36 patients thus far and is estimated to be completed in 2028.⁵³ In another phase II study, 40 patients were treated with SurVaxM and pembrolizumab (NCT04013672). The study is expected to be completed in late 2024 as well.

Heat Shock Protein-Based Vaccines

Heat shock proteins (HSP), which have a role in reassembling misfolded proteins and guiding the degradation of unreducible ones, are found to be upregulated in tumor tissues.⁵⁴ Vaccines based on HSPs may, in theory, lead to more robust T-cell immune responses due to the highly specific interactions between APCs compared to other tumor vaccines. HSPPC-96 is a vaccine that was developed based on HSPs, and its efficacy was evaluated in a phase II trial (NCT00905060) with 70 patients. When administered in combination with temozolomide, the median OS was 23.8 months.⁵⁵ A follow-up randomized phase III trial, where HSPPC-96 was administered with bevacizumab, was completed in 2023, and results are pending (NCT01814813). In addition, there is another phase II trial in progress evaluating the efficacy of HSPPC-96 in conjunction with pembrolizumab (NCT03018288).

Multipeptide Vaccines

As highlighted earlier, vaccines targeting multiple peptides to reduce negative selection and immunologic escape are being developed and tested. IMA950 is a multi-peptide vaccine adjuvanted with poly-ICLC, containing 11 TAA peptides.^56,57 In a phase I/II clinical trial (NCT01920191), a total of 19 patients were recruited and demonstrated a median OS of 21 months. The trial also demonstrated safety and multi-peptide sustained T cell responses in a subset of patients. A follow-up clinical trial was carried out, where IMA950 was combined with pembrolizumab, in 18 patients (NCT03665545).⁵⁸ The study concluded in late 2023, and results are currently pending. Another follow-up trial underway is studying the safety and efficacy of varlilumab and IMA950 plus poly-ICLC in grade II low-grade gliomas (NCT02924038).⁵⁹

ICT-107 is a vaccine based on 6 synthetic peptides targeting those overexpressed in GBM tumor cells.⁶⁰ In a phase II study, 124 patients with newly diagnosed GBM were randomized to receive ICT-107 or unpulsed controls. Results suggested that the vaccine was well tolerated and efficient, with the median OS increased by 2.0 and median PFS increased by 2.2 months compared to controls. Given the possible signal, especially in HLA A2 positive patients, a phase III trial was planned (NCT01280552) but was terminated due to insufficient funding. Finally, EO2401 is a multi-peptide vaccine that is based on the homologies between TAAs and microbiome-derived peptides.⁶¹ These peptides correspond to HLA-A2-restricted peptides with molecular mimicry of 3 TAAs that are often upregulated in GBM. In a phase I/II clinical trial with 100 patients (NCT04116658) where EO2401 was combined with nivolumab and bevacizumab, the interim results suggested that the median OS was 14.5 months, and the median PFS was 5.5 months.⁶¹ The trial is expected to be completed in 2025.

Personalized Peptide Vaccines

Peptide-based personalized vaccines rely on targeting of TSAs, which are often unique to patients. These approaches analyze the transcriptomes and immunopeptides of individual tumors, and the relevant targets from a premanufactured library of neoepitopes are chosen as candidates.⁶² In short, patients were treated with a subset of targets best suited for their tumors. In a phase I clinical trial with 15 patients (NCT02149225), patients were treated with these personalized vaccines derived from biomarker data or tumor genome sequencing, which led to a median PFS of 14.2 months and a median OS of 29.0 months.⁶³ Similarly, in another study using neoantigen prediction, patients were treated with vaccines composed of personalized neoantigen synthetic peptides.⁶⁴ In the study, 8 patients with GBM underwent personalized neoantigen-targeting vaccines, and the investigators found that 2 patients who did not receive dexamethasone during priming showed increased levels of neoantigen-specific T cells targeted to the tumor. The median PFS was 7.6 months, whereas the median OS was 16.8 months. The study has been expanded to a clinical trial (NCT02287428), which has now recruited 56 patients, with the study estimated to be completed in 2026. Multiple other phase I trials are also currently in progress, evaluating the efficacy of personalized approaches. These approaches range from personalized neoantigen DNA vaccines (where tumor-specific mutations are used to create synthetic DNA plasmids, which produce proteins for immune activation) (NCT04015700), personalized genomic vaccines in conjunction with TTFs (NCT03223103), personalized vaccines with checkpoint blockade in children (NCT03879512), neoantigens with pembrolizumab (NCT02287428), and personalized DNA vaccines with retifanlimab PD-1 blockade (NCT05743595).

Nucleic Acid Vaccines for GBM

Nucleic acid vaccines, which are predominantly messenger RNA (mRNA) vaccines, involve incorporating antigen-specific transcripts into APCs, which subsequently express the antigens to generate both innate and adaptive immune responses.⁶⁵ Nucleic acids can be introduced to APCs in either an in vivo or ex vivo fashion; at present, ex vivo loading of DCs with mRNA has been the most commonly used approach. mRNA derived from patient tumor cells can be amplified and pulsed into DCs in vitro, requiring only a relatively small number of cells as the source material.⁶⁶

CMV-Based Vaccines

The cytomegalovirus (CMV) structural protein (pp65) is a phosphoprotein found commonly in GBM (>90%) but not in normal glial tissues.^67,68 As a result, it was a candidate antigen sequence for a vaccine, where DCs were electroporated with mRNA encoding CMV pp65. The vaccine was evaluated in a phase I clinical trial in conjunction with temozolomide in 11 patients with newly diagnosed GBM (NCT00639639).⁶⁸ In this small trial, the vaccine demonstrated safety and improved PFS and overall survival (OS) when compared to historical controls. As a follow-up, 2 clinical trials have been recently completed and are awaiting results, including a phase II trial in conjunction with a tetanus-diphtheria toxoid vaccine (NCT02465268) and a phase I/II trial to identify the vaccine efficacy in patients recovering from lymphopenia caused by temozolomide (NCT00639639).

VBI-1901 is a vaccine utilizing an enveloped virus-like particle targeting the CMV antigens gB and pp65.⁶⁹ In a phase I/IIa trial with 16 patients, the vaccine demonstrated a median OS of 12.9 months and a 12-month OS of 62.5%. In the ongoing part C of the phase I/II trial (NCT03382977), 60 patients across 8 institutions are to be randomized to the active study arm (VBI-1901 + GM-CSF), with the overall survival to be the primary endpoint.⁷⁰

In a recently published paper focusing on the delivery of mRNA (encoding for pp65 and whole-tumor antigens) in intravenously administered multilamellar lipid particle aggregates, the authors found that the tumor microenvironment could be reprogrammed with increased gene signatures for antigen presentation and cytotoxicity.⁷¹ In a feasibility study (NCT04573140), lipid particle aggregates loaded with pp65 mRNA and autologous whole-tumor mRNA, led to periphery immune activation and a proinflammatory reprogramming of the tumor microenvironment in O6-methylguanine-DNA methyl-transferase (MGMT) unmethylated GBM patients.⁷¹ Other trials that are currently active and recruiting include pp65 in conjunction with EphA2 and survivin vaccination (NCT05283109), and pp65 in the treatment of medulloblastoma and malignant glioma (NCT03299309). Furthermore, a recent human trial (NCT04573140) has evaluated feasibility, safety, and activity of multilamellar RNA lipid particle aggregates (LPAs) encoding for pp65, a delivery technique aiming to enhance payload packaging and immunogenicity of vaccines.⁷¹ Their results demonstrated improved and rapid cytokine/chemokine release and immune activation/trafficking.

VEGFR-Based Vaccines

VXM01 is a DNA vaccine consisting of an attenuated Salmonella Typhi Typ21a carrying a plasmid encoding for vascular endothelial growth factor receptor-2 (VEGFR-2).⁷² VEGFR-2 is commonly expressed within tumor microenvironments due to its roles in upregulating angiogenesis and cellular proliferation.⁷³ In a phase I/II clinical trial with 28 included patients (NCT03750071), no treatment-related toxicities were observed when VXM01 was administered with avelumab, an immune checkpoint inhibitor. Interim results suggested that adverse events were mostly unrelated to VXM01, and 3 partial responses with tumor reduction of 58%, 81%, and 95% to baseline were reported in the nonresectable patients (out of 25 total).⁷² Further analysis of the data suggested that 4 specific T cell clones were significantly enriched in the long-term survivor cohort, suggesting that these clones may have predictive value for stratifying patients prior to vaccination.⁷²

Stem Cell-Based Vaccines

Other preclinical approaches have targeted GBM stem cells. This cell population is known for its resistance to chemotherapy and contributes to the aggressive and highly recurrent nature of GBM.⁷⁴ By targeting the stem cells, theoretically, a greater antitumor response could be elicited. Approaches to extracting RNA from CD133+ glioma stem cells to selectively target them with DC vaccines have led to robust and long-lasting immune responses in humanized mouse models.⁷⁵ Mice treated with the approach had a median survival of >60 days, whereas control mice had a median survival of ~40 days. The inhibition of stem cell propagation and tumor growth may lead to decreased immune resistance as these cells are most likely to drive immune resistance.

Other developments have focused on utilizing mRNA from cancer stem cells and transfecting monocyte-derived autologous DCs.⁷⁶ In this approach, harvested tumor cells are grown as tertiary tumor spheres and whole-cell mRNA is then processed and transfected into DCs. This method has been tested clinically, albeit in a small number of patients (n = 7) who had significantly longer PFS compared to a matched control cohort (694 days vs. 236 days).⁷⁶

Whole Tumor Lysate Approaches

Though many of the vaccine approaches discussed above have attempted to predict or identify antigens associated with GBM, whole-tumor cell approaches allow investigators to take antigens directly from the tumor. Advantages of this approach include the ability to load DCs with a wide range of antigens, which can provide a more comprehensive immune response and lessen the likelihood of tumor escape due to heterogeneity. In addition, these approaches allow DCs to display nonmutated, but nonetheless posttranslationally modified, antigens that are likely to be missed with conventional peptide prediction vaccination strategies.⁷⁷ Preclinical studies have found these methods to be superior to peptide-pulsed DC approaches.⁷⁸ Nonetheless, a challenge with this approach is selectivity in purifying, preparing, and quantifying immunogenic antigens from the whole-tumor lysates. In addition, it is difficult to measure the dose of relevant antigens pulsed into DCs, thus making the standardization and safety evaluation in patients obscure.

A phase I trial used allogenic human GBM tumor cell lysate in conjunction with mature autologous DCs (NCT01957956). Specifically, GBM cell lines derived from patient biopsies were grown, lysed, and loaded into DCs with their optimized manufacturing strategy.⁷⁹ Of the 21 patients who were enrolled, the median PFS was 9.7 months, the median OS was 19 months, and the OS was 25% and 10% at 2 and 4 years, respectively.⁷⁹ In another study, autologous DCs were pulsed with lysate derived from a GBM stem-like cell line.⁸⁰ In this clinical trial (NCT02010606), 11 patients had a median PFS of 8.75 months and OS of 20.36 months. The cell line was safe and well tolerated, and the results suggest potential efficacy.

Strengthening Tumor Vaccine Responses

A prominent limitation of current vaccines is the lack of robust immune system recruitment and response. GBMs are considered “nonsupportive niches,” where even with DC vaccination, T cell states are often immature or insufficiently activated, leading to an antiproliferative state.⁸¹ Furthermore, while DC vaccines have been known to partially correct the tolerogenicity of GBM, this alone still remains insufficient. To this end, multiple preclinical approaches have attempted to strengthen the resultant T cell responses following vaccination, including effective homing and persistence of T cells. The GIFT-7 fusokine tumor vaccine is an engineered fusokine of IL-7 and GM-CSF.⁸² Application of this vaccine to aged mice led to long-term antitumor immunity and a durable long-term increase in proinflammatory cytokines, resulting in hyperactivation of DCs and increased T cell trafficking to tumor sites.⁸² In another study, authors found that Tc17-1 cells (CD8+ T cells that produced IL-17 and IFN-γ) combined with poly-ICLC and Pmel-1 peptide vaccine significantly prolonged the survival in animal models.⁸³ Another vaccine platform (ITI-1001) utilized a DNA vaccine that codes for the HCMV protein pp65, gB, and IE-1. In a syngenic orthotopic GBM model, ITI-1001 resulted in strong cellular and humoral immune response, antitumor activity, and enhanced survival of tumor-bearing mice (56%).⁸⁴ Specifically, the authors observed increased activation of Th1, cytotoxic CD8, reg T cells, and IFN-γ+ CD4 T cells. Finally, an autologous and allogenic mixed glioma cell lysate vaccine in a rat model was developed to reverse tumor-associated immunosuppression.⁸⁵ The vaccine not only inhibited glioma cell proliferation, it also promoted their apoptosis, leading to a significantly improved survival time for rats. The results were thought to be due to the enhanced secretion of T cell chemokines monocyte chemotactic protein-2 (MCP-2), interferon gamma (IFN-γ), and interleukin 2 (IL-2), which stimulated the proliferation of T cells and NK cells.

Adjuvants

Adjuvant immune response modifiers may potentiate vaccine efficacy and immunogenicity. The addition of Type I interferon and synthetic TLR3 activator poly (I:C) improves DC function, leading to sustained cytokine production postvaccination and a corresponding survival benefit.⁸⁶ Additional strategies to improve the inflammatory milieu at the tumor site include the administration of granulocyte-macrophage colony-stimulating factor (GM-CSF),⁶⁸ poly(I:C),⁸⁶ cytokine IL-12,⁸⁷ and other TLR agonists.²⁴ Bevacizumab, a monoclonal antibody against VEGF, may lead to improvement in antigen presentation, lymphocyte trafficking, and DC maturation for cell-based vaccines.⁸⁸ Montanide is an emulsion that has been shown to prolong the release of vaccine antigens. Two trials (NCT02455557, NCT02454634) have shown no associated serious adverse events.^51,89 Prevaccination with the tetanus-diphtheria toxoid vaccine has been shown to improve DC migration and efficacy, suppress tumor growth, better lymph node homing, and prolong survival.⁹⁰ Similarly, synthetic immunomodulatory agents, such as resiquimod and imiquimod, have been shown to enhance cytokine production and skew immunity toward a Th1 response.^91,92 Combination with immune checkpoint inhibitors, which aim to reverse the immunosuppressive microenvironment and improve non-exhausting T-cell homing to the glioma site is also currently being investigated in multiple clinical trials.⁹³

An adjuvant that has been explored in the preclinical setting is lenalidomide.⁹⁴ In a murine GL261 intracranial glioma model, multi-epitope peptides in combination with lenalidomide and anti-PD-1 prolonged the survival of mice by suppressing tumor growth. An increase in the percentage of regulatory T cells was seen, as well as an enhanced cytotoxicity against GL261. Another study investigated the effectiveness of DCs pulsed with protein antigens and anti-PD-1 and found enhanced expression of MHC and costimulatory markers, resulting in increased IFN-γ+ effector T cells and enhanced lysis of GBM target cells.⁹⁵ Finally, another preclinical study showed that combinatorial treatment with branched multi-peptide and peptide adjuvants such as PADRE and poly-ICLC, anti-PD, and radiation could enhance the antitumor effects of radiotherapy in a GBM mouse model.⁹⁶

Future Directions

Many candidate vaccines failed to impact clinically meaningful endpoints, likely due to nonrobust and nonsustained immunogenicity. Alternative strategies to increase immune activation or increase the potency of vaccines may augment their efficacy. Analysis from personalized vaccine clinical trials has suggested that corticosteroids should be avoided, and additional measures to combat T-cell exhaustion are required.⁹⁷ The lack of success with immune monotherapies has suggested that combination therapies that target multiple immunological mechanisms in coordination may lead to a stronger immune response characterized by improved APC immunogenicity, leading to increased cytokine production, migration to the tumor, and increased effectiveness at the tumor site.⁹⁷

Combinatorial Approaches

Due to the lack of success observed with monotherapies, combining multiple different immunological therapies could represent a promising avenue for clinical success. The combination of vaccines with OVs and chimeric antigen receptor T-cell therapy (CAR-T) represents a coordinated approach that comprehensively utilizes multiple immune mechanisms and pathways. CAR-T therapy relies on modified T cells to initiate cytotoxic attacks on targeted antigen-bearing tumor cells, thereby bypassing the MHC presentation pathway.⁹⁸ Thus, vaccines in conjunction with CAR-T therapy represent tumor treatment via complementary immunologic arms. Multiple clinical trials have been carried out with CAR-T therapy, where although safety and feasibility have been shown, limited efficacy has been demonstrated.^99–101 This is likely again in part due to tumor heterogeneity and antigen escape. Targeting multiple epitopes via personalized approaches could, in theory, expand the coverage of GBM and provide notable clinical efficacy. OVs selectively infect and replicate in tumor cells to both lyse tumors and stimulate an antitumor immune response.¹⁰² Multiple clinical trials have demonstrated safety and efficacy of these modalities,¹⁰³ though there remains ongoing exploration in vector and delivery design to improve efficacy of these novel treatment modalities. Combination with vaccine therapies may promote synergy in mechanisms, where OVs could traffic more easily to cells being targeted by vaccines, whereas oncolysis may limit immunological escape against vaccines. Notably, in a recent Nature Cancer publication, Chen et al. showed that OV can be used to induce the expression of bystander T cell epitopes in tumor cells. These are tumor-irrelevant antigens (often viral) that are recognized by bystander T cells, leading to antitumoral cytotoxicity and improved tumor control in preclinical models of numerous malignancies.¹⁰⁴ Robust preclinical and clinical investigations are required to assess these possible interactions.

Delivery

The majority of vaccination trials in GBM rely on subcutaneous, intradermal, or intravenous delivery methods. However, there are recent advancements aiming to improve the efficacy of local therapy delivery, including polymeric wafers, nanofibrous scaffolds, and hydrogels.^105,106 In one study, a hydrogel vaccine system was designed containing granulocyte-macrophage colony-stimulating factor (GM-CSF) and tumor cell lysates and implanted at the surgical site.¹⁰⁷ The personalized hydrogel was able to inhibit tumor recurrence, and coupled with findings from another earlier study implementing polypeptide hydrogels,¹⁰⁸ suggests that hydrogels can be used as an effective sustained delivery platform for vaccines in GBM. Another example is mitoxantrone-loaded PEGylated PLGA-based nanoparticles (NP-MTX).¹⁰⁹ A study utilizing intratumoral NP-MTX delivery showed that it increased immune cell infiltration, specifically increased frequency of IFN-y secreting CD8 T cells and M1-like macrophages. More importantly, the study was able to demonstrate that vaccines administered via this approach increased the median survival of GL261-bearing mice. Thus, this delivery method may lead to more effective immune cell recruitment, allowing for possibly increased median survival.

Conclusion

Therapeutics for GBM continue to be limited due to tumor heterogeneity, immunosuppressive microenvironment, and the BBB. Vaccine therapeutics employing multiple approaches have been investigated in the preclinical and clinical settings. Whereas monotherapy targeting single epitopes formed the foundation of earlier investigations, the nature of immunologic escape and lack of clinical success is guiding the field away from single-peptide vaccines. Vaccines targeting multiple targets likely form the cornerstone of future advancement, though optimization to improve potency is paramount. Furthermore, these multiple targets are not consistently conserved across patients and tumors, and thus personalized approaches, with hierarchical identification of best target mutations, could improve vaccine efficacy and potency. Finally, adjuvants and combination therapy to improve vaccine delivery and stimulate the immune system are critical to improving possible clinical responses. Though there are encouraging results from preclinical and Phase I clinical investigations, further Phase II and III investigations are required to optimize and realize the clinical potential of vaccines for GBM. Vaccines are a promising treatment for GBM; while significant clinical efficacy has yet to be conclusively demonstrated, new therapeutic strategies to overcome tumor heterogeneity and immune escape may soon bring them into more widespread clinical use.

Supplement sponsorship

This article appears as part of the supplement “Immunotherapy for Brain Tumors,” sponsored by the Wilkins Family Chair in Neurosurgical Brain Tumor Research.

Conflict of interest statement

The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. J.D.B. has an equity position in Treovir Inc. and UpFront Diagnostics. J.D.B. is also a cofounder of Centile Bioscience and on the NeuroX1 and QV Bioelectronics scientific advisory boards. J.A.C. is a cofounder and holds equity in Verge Genomics. J.A.C. is also an advisor and holds equity in Gravity Medical Technology. F.A.G. received honoraria from Signus, BBraun, Aesculap, and AstraZeneca.