Perspectives on investigational drugs and immunotherapies for glioblastoma

KEYWORDS:

• Glioblastoma
• IDH
• immunotherapy
• immune checkpoint inhibition
• vaccination

1. Introduction

Glioblastoma is the most common primary brain tumor in adults and one of the most complex cancers to treat. With current standard of care, including surgery, radiation and chemotherapy, the prognosis remains poor, and the median overall survival ranges from 15–17 months [1]. In recent years there has been significant advancement in the understanding of the molecular underpinnings of glioma pathogenesis. Various molecular and genetic markers have defined glioblastoma subclasses associated with distinct prognosis and response to therapy [2]. This knowledge is starting to influence clinical practice and clinical trial design. Novel immunotherapeutic and tumor-targeting strategies are on the horizon with hope for improved patient outcomes.

2. Immunotherapy

As cancer immunotherapy is making rapid clinical inroads in solid and hematologic malignancies, there is increasing interest in therapeutic immune manipulation in glioblastoma. Patients with glioblastoma exhibit extreme immunosuppression, both systemically as well as within the tumor microenvironment. Examples include T cell lymphopenia, T cell dysfunction, disproportionately high numbers of circulating suppressive regulatory T-lymphocytes, high intratumoral levels of vascular endothelial growth factor (VEGF), tumor cell expression of PD-L1, and a suppressive tumor-associated macrophage (TAM) orientation [3]. These factors, as well as impaired immune effector trafficking across the blood–-brain barrier, present challenges for effective immunotherapy in glioblastoma. However, there have been advances in ways of how to overcome tumor related immunosuppression. Such approaches include efforts to improve and expand immune recognition of tumor associated antigens, typically by vaccination, and to minimize negative immune regulation, such as by using immune checkpoint inhibitors.

2.1. Vaccination

The well-known cellular and molecular heterogeneity in glioblastoma represents both a challenge and opportunity for immunotherapy. A single peptide target likely does not exist for immune targeting of glioblastomas. Nevertheless, peptide vaccination against mutant EGFR (EGFRviii) in EGFRviii-positive glioblastomas increased survival of patients in a multicenter phase II study [4]. However, interim analysis of a randomized phase III study (ACT IV) showed no overall survival benefit following vaccination, though survival in both treatment and control groups was unexpectedly high (www.celldex.com).

Conceptually similar, cytomegalovirus (CMV) antigens frequently found in glioblastomas have been targeted via dendritic cell (DC) vaccination [5]. In this approach, leukapheresis derived DCs are pulsed in vitro with the CMV pp65 antigen and injected subcutaneously into the patient. Pulsed DCs traffic to draining lymph nodes, where they encounter, educate, and stimulate naïve T lymphocytes, which then traffic to tumor cells that are bearing antigenic targets. In a small, randomized study, preconditioning of vaccinated patients with tetanus toxoid improved patient survival. Corresponding animal studies demonstrated that tetanus toxoid preconditioning drives enhanced trafficking of CMV antigen-pulsed DCs in a CCL3-dependent fashion [6].

Alternative approaches to DC vaccination take advantage of glioblastoma cell heterogeneity. Rather than being used to target a single peptide antigen, DCs can also be loaded with multiple peptides or whole tumor lysates. For example, ICT-107 is a DC vaccination platform in which DCs are pulsed with an array of six-peptide antigens, in addition to being modified to express interleukin-12 [7]. In a randomized phase II study in patients with newly diagnosed glioblastoma, ICT-107 treatment improved overall survival of HLA-A2⁺ haplotype patients, and a phase III study with this subset has now been initiated (NCT02546102). DC-VAX is another approach utilizing whole tumor lysates as antigenic material and is currently under investigation in newly diagnosed glioblastoma patients (NCT00045968), after promising survival results were seen in single arm early-phase studies.

An alternative approach to targeting the whole tumor antigenic repertoire is the ex vivo derivation of heat shock proteins (HSPs) from tumor tissues. These HSPs are naturally complexed to and chaperone tumor-associated antigens into antigen presenting cells. Heat shock protein complexes have been used as subcutaneous vaccination in patients with newly diagnosed and recurrent glioblastoma, and are currently being examined in combination with bevacizumab in a multicenter randomized study in recurrent glioblastoma patients (NCT01814813).

In an alternative approach, vaccination of recurrent glioblastoma patients with irradiated autologous tumor cells mixed with an irradiated GM-CSF expressing bystander line has shown consistent activation of T-lymphocytes and generation of specific antitumor immunity [8].

While definitive survival results remain to be established, a number of vaccination strategies that are focusing on driving responses against unique cancer peptides and targeting a broader array of antigens, are being studied. To date, valuable insights into the feasibility and the optimization of biological anticancer activity have been established. Future studies will examine these immunotherapeutic products in combination with other existing therapies.

2.2. Immune checkpoint blockade

The development, preclinical analysis, and clinical trial demonstration of disease response have brought antibody-based inhibition of immune checkpoints into the spotlight of current clinical investigations. These drugs, best exemplified by blockade of CTLA-4, PD-1, or PD-L1, allow for persistent stimulation and proliferation of antigen-specific T-lymphocytes. With acceptable autoimmune toxicities, durable responses have been demonstrated in multiple cancer types, and several immune checkpoint blockade approaches have received Food and Drug Administration (FDA) approvals. Clinical application in glioblastoma is in early stages, and preclinical demonstrations of efficacy, while clear, remain relatively scarce.

In early results from the Checkpoint 143 trial (NCT02017717) in patients with recurrent glioblastoma, PD-1 blockade with nivolumab combined with CTLA-4 blockade with Ipilimumab was less tolerable than treatment with nivolumab alone. Median overall survival in the nivolumab group was 10.5 months, suggestive of a modest response. This study is presently ongoing. Similarly, PD-1/PD-L1 blockade is being examined in single and multicenter trials (e.g., NCT02336165; NCT02359565; NCT02337491; NCT02311582; NCT02529072) [9].

In summary, as in other fields in clinical oncology, the investigation of immunotherapy in glioblastoma is advancing at a rapid rate. Several approaches to vaccination drive systemic antitumor immune responses with enhanced recognition of glioma antigens with manageable toxicity. Existing immune checkpoint inhibitors hold the promise of overcoming one aspect of negative immune regulation by bolstering existing T-lymphocyte responses. The immediate future will likely include various combination immunotherapies, development of new immune checkpoint inhibitors, and application of new technologies such as chimeric antigen receptor (CAR) T cells [10].

3. Targeting IDH mutation

Among the multitude of potentially targetable molecular-genetic alterations in gliomas are mutations in the gene isocitrate dehydrogenase (IDH). IDH mutations are present in 70–90% of WHO grade II and III glial tumors and 10% of glioblastomas that evolved from lower-grade tumors [11,12]. A genome-wide mutational analysis demonstrated that heterozygous mutations in IDH1 are frequently found in glioblastoma [13].

Mutations in the related IDH2 occur at a much lower frequency [11]. Presence of IDH mutations is associated with prolonged overall survival [12,13].

IDH1 and IDH2 are homodimeric metabolic enzymes, which catalyze the NADP⁺ -dependent oxidative decarboxylation of isocitrate to α-ketoglutarate (α-KG). The mutant enzyme acquires a neoenzymatic activity, resulting in elevated levels of 2-hydroxyglutarate (2-HG) [14]. Several small molecule inhibitors designed to block IDH mutant activity and to limit 2-HG production are presently under investigation (e.g. NCT02073994; NCT02273739; NCT02381886). A second-generation IDH inhibitor, targeting both IDH1 and IDH2 mutant tumors, is currently tested in a phase I study (NCT02481154).

The specificity of the IDH mutation appears to represent an ideal drug target, as the mutant enzyme is only found in tumor cells. Moreover, the IDH mutation is considered an early event in tumor pathogenesis, resulting in near ubiquitous expression throughout a tumor mass [15]. Several other agents which may selectively target IDH mutant cancers have been identified in preclinical models and are entering clinical investigation, including 5-azacytidine [16], decitabine [17], glutaminase inhibitors [18,19], and NAMPT inhibitors [20].

4. Expert opinion

Glioblastoma remains one of the most difficult cancers to treat, in part because of its tumor cell heterogeneity and widely infiltrative growth pattern. Promising therapeutic strategies are aiming to exploit tumor cell heterogeneity, to targeting tumor specific antigens and to incorporate immunotherapeutic approaches into the armament of cancer therapy. Drugs designed to target the IDH mutation, as one of the potentially targetable mutations in gliomas, take advantage of its likely function as an early mutation in tumorigenesis, and its presence throughout all phases of tumor progression.

Immune checkpoint inhibitors and vaccine-based therapies make use of antigenic substrates provided by tumor diversity, which could enhance the robustness of the immune attack with formation of immune memory. The promise of these strategies also lies in their increasing specificity towards targeting tumor cells preferentially over healthy tissue, either through tumor-specific antigens, or modulation of immune checkpoint molecules.

Several challenges remain with the clinical development of immune checkpoint inhibitors and vaccination strategies in glioblastoma, including the exceptional immune properties of the central nervous system, the unique requirements on imaging assessment of treatment response and a limited insight into the compatibility of immunotherapeutic approaches with existing antineoplastic agents.

Several novel biomarkers have increased the current understanding of malignant gliomas and their varying responses to therapy, and are being incorporated into tumor classification and clinical trial design. Powerful treatment strategies are emerging and are currently in early clinical trial stages. There is hope that such therapies will result in significant antitumor effects. Lastly, individualized therapies with selection of specific treatments to individual patients promise to change the management of this devastating disease in the near future.

Declaration of interest

J Dietrich also received generous philanthropic support from the family foundations of Sheila McPhee, Ronald Tawil and Bryan Lockwood. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Additional information

Funding

J Dietrich has received funding and research support from the American Academy of Neurology Foundation [1200-219466], the American Cancer Society [1200-222362] and the National Institute of Health [RO1CA190901].