Prognostic Biomarkers of Dendritic Cell Vaccines in Glioblastoma Multiforme

Introduction

Glioblastoma multiforme (GBM) is an incurable primary brain tumor with a median survival of 14 months. Current maximal treatment approaches are gross total resection of the tumor followed by chemotherapy and/or radiotherapy, with nearly all patients undergoing recurrence within 1 year.¹˒² There is a clear need for novel therapies. Recent advancements in cancer immunotherapy, including immune checkpoint inhibitors, chimeric antigen receptor (CAR) T cell therapy, and cancer vaccines, have broadened the scope of treatment options for cancer patients.³ However, immunotherapies for GBM have largely shown disappointing results in clinical trials, despite early promise in animal models. 

A hallmark of GBM is its immunosuppression of the local tumor environment and central immune system, which enables it to evade the body’s ordinary cancer defense mechanisms.⁴ In addition, the tumor neoantigen burden is markedly low, rendering GBM a “cold” tumor.⁵ These characteristics are crucial barriers to overcome in developing all immunotherapies, particularly vaccines, as they rely on a functional immune system to generate a response.  

Cancer vaccines are a potential new therapeutic option for GBM patients, who are faced with a dismal prognosis.⁶ More recently, dendritic cell (DC) vaccine therapy in GBM has aimed to reinvigorate the body’s own anti-tumor immune response. DCs are often described as professional antigen-presenting cells with an important role in both the innate and adaptive immune systems. The focus of this review is autologous DC vaccines, where a patient’s own DCs are harvested, primed with tumor neoantigens ex vivo, then reinfused in the patient to prime reactive T cells to initiate a specific and potent antitumor immune response. Although autologous DC vaccines are the most common, other methods include allogeneic vaccines, in which a healthy donor’s antigens are used to prime DCs and reinvigorate for the patient’s immune system. The only FDA-approved DC vaccine is sipuleucel-T for advanced prostate cancer.⁷ DC vaccines have been tested in clinical trials for various other solid tumors,⁸ but the clinical evidence for their wider efficacy remains heterogeneous. However, DC vaccines in GBM have generally been robustly shown to be well tolerated,⁹˒¹⁰ and more recently have been shown to improve survival in some clinical trials.¹¹˒¹² In this review, we aim to summarize the evidence associating dendritic vaccines with survival outcomes in patients with GBM, with a particular focus on increased immunologic response induced by DC vaccines. 

Methods

In this narrative review, literature was identified via an informal, targeted keyword search of the following databases: PubMed and Web of Science. The searches were conducted between the dates of June 15, 2026 and June 30, 2026. Keywords included dendritic cell vaccine, glioblastoma multiforme, biomarkers, immune response, interferon signaling, prognosis, and immunotherapy. Studies included clinical trials, meta-analyses, and systematic literature review. Studies prior to the year 2000 were excluded based on relevancy to modern immunotherapy techniques. 

Dendritic Cell Vaccines May Improve Survival Metrics in Newly Diagnosed Glioblastoma

Dendritic cell vaccines are one of the most extensively studied vaccination types in glioblastoma management, with demonstrated improvements in both overall survival (OS) and progression-free survival (PFS) in patients with newly-diagnosed glioblastoma.¹⁰˒¹³⁻¹⁶ Most existing studies evaluating efficacy of DC vaccines have been phase I/II trials, with a sample size ranging from 9 to 124 patients with newly-diagnosed glioblastoma. Studies so far have suggested that DC vaccines are the most effective vaccine therapy for glioblastoma. One meta-analysis of 2792 patients examined the efficacy of DC vaccines, peptide vaccines, and viral vector vaccines. Compared to standard of care, DC vaccines provided statistically significant survival benefit in both OS and PFS with a hazard reduction of 42% and 35% respectively, while viral vector vaccines only demonstrated an OS benefit with marginally statistically significant improvement in PFS and peptide vaccines did not show significant improvement.¹⁷ In this section, we discuss the efficacy and survival benefit of DC vaccines in newly diagnosed glioblastoma.

DC vaccines are commonly prepared using autologous tumor homogenate: a mixture of the patient’s own tumor tissue containing tumor-associated antigens, cellular debris, and proteins. These vaccines have been tested in clinical trials in addition to standard GBM treatment with radiotherapy RT), chemotherapy (CT), and temozolomide (TMZ). Autologous DC vaccines were well-tolerated by patients with low rates of DC vaccine-related toxicities, and yielded overall higher OS and PFS as compared to control groups of patients receiving standard therapy in phase I/II trials.¹⁰˒¹³˒¹⁵˒¹⁶˒¹⁸⁻²⁰ One phase II trial evaluating the Audencel autologous DC vaccine concluded that adding the Audencel vaccine did not improve OS or PFS for newly diagnosed GBM patients, with median OS being 564 days in the treatment group as compared to 568 days in the control group (p=0.89).¹⁹ A phase III trial evaluating the autologous DCvax-L vaccine showed extension of overall survival in both new GBM (19.3 vs 16.5 months) and recurrent GBM patients (13.2 vs 7.8 months) with DCVax-L treatment, with no evaluation of PFS, as compared to matched, contemporaneous external controls.²¹

Not all DC vaccines evaluated used autologous tumor homogenate. Lepski et al.’s phase I/II trial evaluated overall survival with allogeneic DC vaccines, which resulted in OS for vaccinated patients with glioblastoma being 27.6 ± 2.4 months, as compared to 16.3 ± 0.7 in control patients receiving standard therapy from the Genomics Data Commons database.²² Other studies included the ICT-107 trial which used autologous DCs pulsed with GBM-associated peptides,¹⁴ and Berneman et al.’s phase II trial which used autologous DCs transfected with WT1 tumor-associated antigen mRNA.⁹ Mitsuya et al. evaluated a peptide cocktail-pulsed alpha-type-1 DC vaccine in a phase II trial, which resulted in significantly increased OS and PFS as compared to a retrospective historical control patient cohort.²³ Table 1 summarizes the findings of studies mentioned in this section, detailing the types of vaccines used and impact on OS and PFS. Overall, DC vaccines are shown to have a favorable safety profile and significantly increase both OS and PFS as an adjuvant to standard therapy in glioblastoma.

Immune Responses to Dendritic Vaccines

A core component of the potential efficacy of the dendritic cell vaccine is its ability to cause a robust immune response. The goal of DC vaccines is to recruit immune cells to attack the targeted tumor through the release of neoantigens, yet their specific ability to do so remains unclear. Post-vaccination functional responsivity can be quantified through peripheral blood metrics, including cytokine levels and T-cell type levels. Monitoring immune phenotypes could also shed light on potential mechanisms by which DC vaccines impact survival, the majority of which remain unknown. In this section, we examine immunological evidence concerning the efficacy of the DC vaccine.  

Early studies of DC vaccines in GBM showed their ability to elicit an increase in CD8+ T-cells and increase their invasion of the tumor.⁹˒²⁰˒²⁴ More recent work has focused on associating the DC-induced immune response with increased survival outcomes. In one phase II study of the α-type DC vaccine, patients showed robust HLA class I expression post-vaccination, confirmed with high levels of peripheral CD8 T cell counts. Peripheral blood studies showed markedly increased chemical markers of T cell activation, such as IFNγ, GZMB, and CD127.⁹ Similarly, Naulaerts et al showed recipients of a DC vaccine with GBM had higher effector memory CD8+ T cells and some evidence of antigen-specific immunity.²⁵

Further research has correlated the metrics mentioned above with survival in GBM patients. More recent work has focused on associating the DC-induced immune response with increased survival outcomes. In one single-arm phase I/II study of patients with various solid tumors receiving dendritic cell vaccines, disease control and higher OS were associated with antigen-specific type 1 CD4+ and/or CD8+ T-lymphocyte responses, mainly induced by WT1-mRNA/DC vaccination.⁹ Berneman et al. show that the majority of patients in the study also demonstrated targeted type 1 T-lymphocyte responses against the targeted neoantigen. 

Larger post-vaccine shifts in peripheral T cell counts were correlated with increased survival in GBM.³ Moreover, Mitsuya et al. showed that DC vaccination also reduces the amount of IL-12 needed to enhance the immune response from cytotoxic T-lymphocytes. This study indicated that DC vaccines influence the IL-12-IFNγ-Th1 pathway, suggesting a more functional immune communication pathway post-vaccination.²³˒²⁶ Everson et al. demonstrated that peripheral CD8+ T cells were more responsive to IL-2 post vaccination, and that increased responsivity was correlated with improved survival, further highlighting the importance of cytokines as biomarkers for vaccine efficacy. However, the study also found that patients with decreased survival had a lower pSTAT1 ratio, a surprising contrast to Mitsuya et al above, who showed increased responsiveness to IFNγ as correlated with increased survival.²³˒²⁷ Prins et al. compared the survival and immune response between antigen peptide-loaded (GAA-DC) and autologous tumor lysate-loaded dendritic cell (ATL-DC) vaccinations in glioblastoma patients. Survival was unable to be compared meaningfully due to the low sample size, and ATL-DC vaccination was associated with decreased fractions of activated natural killer (NK) cell populations, which was a prognostic factor for increased survival. In both vaccine groups, decreased regulatory T-cell (Treg) frequencies were associated with prolonged survival.²⁸ Taken together, these studies show that dendritic vaccines as an adjunct therapy may contribute to increased survival in patients with glioblastoma through an induced T cell response.

Genetic and Immune Factors Impacting Survival in Dendritic Cell Vaccine-Treated Patients

Dendritic cell vaccine immunotherapy has been associated with modest survival benefits in glioblastoma patients in most early trials, with increased immune response linked to improved overall and progression-free survival. Beyond clinical characteristics, molecular and immunologic biomarkers are increasingly recognized as important determinants of prognosis and therapeutic response in glioblastoma, particularly in the context of DC vaccine therapy. While some biomarkers have established prognostic value irrespective of DC vaccine treatment, others may specifically predict responsiveness to immunotherapy by influencing tumor immunogenicity, antigen presentation, or the tumor immune microenvironment. The available evidence suggests that both established glioblastoma biomarkers and emerging molecular signatures may help to identify patients most likely to benefit from DC vaccination.

MGMT promoter methylation is a leading determinant of prognosis in GBM, and is both an independent factor associated with longer overall survival and associated with longer survival in patients receiving standard therapy with temozolomide.²⁹˒³⁰ There is conflicting data on association with MGMT protein expression with methylation or prognosis: some evidence suggests favorable prognosis with loss of MGMT protein expression, while others suggest no association between MGMT expression with methylation or improved prognosis.³¹˒³² These single-arm cohort trials show that MGMT is prognostic in vaccinated patients. Whether it specifically predicts incremental benefit from DC vaccination (as opposed to from temozolomide) would require a treated-versus-untreated comparison demonstrating a treatment-by-methylation interaction.¹³˒¹⁴˒¹⁶˒²¹˒³³ However, evidence is heterogeneous: other studies showed no significant association of MGMT promoter methylation status with survival following DC vaccine therapy.³⁴

Another group of biomarkers implicated in both glioblastoma prognosis and immunotherapy response comprises the genes encoding the human leukocyte antigens (HLA).³⁵ Wen et al.’s trial of the ICT-107 DC vaccine showed an increase in median PFS associated with HLA-A2 (24.1 months compared to 8.5 months in the control group) in MGMT-methylated groups, with no significant benefit in unmethylated groups and no significant OS benefit in either group. Increase in median OS was associated with HLA-A1 in MGMT-methylated groups (47.6 months compared with 25.8 months for the control group), with no significant OS benefit in unmethylated groups or PFS benefit in either methylated or unmethylated groups. In addition, HLA-A2 patients showed higher immune response after administration of DC vaccine.¹⁴ ICT-107's randomized design makes HLA-A2 a plausible predictive marker (benefit versus a control arm, confined to the MGMT-methylated subgroup). Takei et al.'s treated-only HLA-A finding is prognostic. Takei et al. showed improved prognosis after DC vaccine therapy associated with low HLA-A, but not HLA-B or HLA-C expression, extending overall survival.³⁴ 

Alterations in tumor metabolism may also influence prognosis with glioblastoma and response to immunotherapy. Isocitrate dehydrogenase (IDH), a critical protein in energy metabolism, is one of the most mutated genes in human cancer, with somatic mutations in IDH1 and IDH2 frequently identified in glioblastoma.³⁶ IDH mutations have been associated with longer survival in glioblastoma, and Bota et al.'s phase 2 trial of the DC-ATA vaccine showed a positive association of IDH mutation with vaccine response.¹³ However, Yao et al.'s trial contradicted this, showing significantly prolonged overall survival in IDH-wild type patients and suggested that IDH-wild type patients were the most responsive to DC-vaccines.³⁷ Both are vaccinated-only cohorts, so these are prognostic associations. Since IDH mutation is independently prognostic in glioma, a predictive claim needs a randomized comparison.

In addition to tumor-intrinsic molecular alterations, immune checkpoint biomarkers have also been investigated as predictors of outcome following DC vaccination. The immune checkpoint inhibitor programmed death-ligand 1 (PD-L1) is often expressed in glioblastoma, with limited and controversial data on prognostic value, with high PD-L1 expression suggested to be a prognostic factor associated with increased survival in some studies but poor survival in others.³⁸  Peres et al.’s phase II trial revealed that PD-L1 expressed by immune cells in the tumor microenvironment was a  was associated with longer overall survival among vaccinated patients.³⁹ Conversely, Jan et al. suggested that patients with lower PD-1 ratios (PD-1 being the receptor for PD-L1) experienced improved outcomes following DC vaccination.³³

Beyond these established biomarkers, exploratory molecular profiling studies have identified additional candidates for prognostic survival biomarkers of DC vaccine therapy in glioblastoma. Takei et al. conducted comprehensive molecular profiling, including transcriptome and exome analyses, as part of their phase I/II trial evaluating 28 patients who all underwent DC vaccination. They observed better prognosis associated with CCDC88A, KRT4, TACC2, and TONSL mutations in tumor cells.³⁴ Yao et al. observed an association between low expression of the immune checkpoint protein B7-H4 and better responsiveness to DC vaccine immunotherapy.³⁷ Erhart et al. conducted a proteomics and miRNomics study of DC-vaccine-treated glioblastoma patients. They observed associations of huntingtin interacting protein 1 (HIP1), retinol-binding protein 1 (RBP1), ferritin heavy chain (FTH1) and focal adhesion kinase 2 (FAK2) associated with poor prognosis in their protein analysis. MicroRNA analysis yielded positive associations of miR-216b, miR-216a, miR-708 and let-7i with longer survival.⁴⁰ Most candidates come from single-arm cohorts and are thus prognostic within vaccinated patients. Only the randomized ICT-107 subgroup data offer preliminary predictive evidence, thus true predictive value awaits trials demonstrating a treatment-by-biomarker interaction.

Conclusion

While results vary across studies, the available evidence indicates that dendritic cell vaccine immunotherapy is a promising adjunct to standard therapy for newly diagnosed glioblastoma. Most phase I/II trials demonstrated improvements in both overall and progression-free survival, while maintaining favorable safety profiles. Although not all trials have shown a survival benefit, there has been overall consistency of data suggesting improved survival across trials of both allogeneic and autologous vaccine trials. So far, only one phase III study has been performed on DC vaccine therapy, which showed improved survival rates.  These clinical benefits appear to be mediated by enhanced antitumor immune activity, including increased CD4+ and CD8+ T-cell responses, activation of Th1-associated cytokine pathways, and favorable changes in immune cell populations. The relationship between immune response and survival remains inconsistent across studies. Although these early trials seem promising, further innovation is needed to provide more substantial survival benefits to patients. 

Current research also highlights the importance of biomarkers in the prognosis of glioblastoma and determining which patients are most likely to benefit from DC vaccine therapy. Established glioblastoma prognostic factors such as MGMT promoter methylation, HLA subtype, and IDH mutation status may affect vaccine efficacy, and exploratory studies focusing on miRNomics and proteomics may identify further biomarkers that influence treatment response. 

Despite encouraging findings, many challenges remain before DC vaccines can be integrated into standard therapy for glioblastoma. Many studies are limited by small sample sizes and variability in vaccine manufacturing, immune monitoring, and outcome reporting, making direct comparisons between studies difficult. Future adequately powered clinical trials and preclinical studies with standardized immunologic endpoints will be essential to validate prognostic biomarkers, clarify mechanisms of response, and identify the patients most likely to benefit from treatment. Continued integration of molecular profiling with immune monitoring will be critical for advancing personalized DC vaccine immunotherapy and improving outcomes for patients with glioblastoma.

Conflicts of Interest/Disclosures

SI has no conflicts of interest. 

DPN is a member of the Board of Editors for the Virginia Journal of Medicine.