Topics in Chemotherapy, Molecular-targeted Therapy, and Immunotherapy for Newly-diagnosed Glioblastoma Multiforme

Chemotherapy for GBM

Historical context of chemotherapy. GBM is considered to be resistant to chemotherapy. There are two main reasons that have been established for this resistance (7); the first is the presence of the blood–brain barrier (BBB), which blocks the transport of most molecules larger than 500 Da. Therefore, the BBB is an obstacle to the adequate delivery of chemotherapy agents to brain tumors, particularly the infiltrating component of the tumors, where the malignant cells have intercalated into normal brain parenchyma (8). Only small and lipophilic molecules are able to cross the BBB to reach their target. The second reason for the chemotherapy resistance of GBM is the intrinsic resistance of the GBM cells, particularly glioma stem cells (9), which show enriched activity of the DNA repair enzyme O⁶-methylguanine-DNA methyltransferase (MGMT) (9) and have demonstrated resistance to alkylating agents (10). Considering these characteristics of GBM, strategies have been developed and evaluated combining chemotherapy with RT.

Historically, intravenous injection of nitrosoureas, which are lipid-soluble agents that cross the BBB, has been considered to have potential for improving the outcomes of patients with GBM receiving RT (11). A meta-analysis by Stewart et al., in which more than 3,000 patients with malignant glioma were included and the bulk of the trials utilized intravenous injection of nitrosoureas, demonstrated a modest increase in MST due to the addition of chemotherapy to RT (12). However, only a 5% increase in survival at two years (from 15% to 20%) was observed using this combination of chemotherapy and RT. In addition, in this meta-analysis, 37% of patients had prognostically favorable, lower-grade gliomas (13). There was no evidence that the effect of intravenous injection of nitrosoureas differed in any group of patients according to age, gender, performance status, or extent of surgical resection. These results drove researchers to explore the use of novel chemotherapy agents in combination with RT.

Temozolomide. At present, temozolomide is the standard agent used for concomitant and adjuvant chemotherapy with RT to treat patients with GBM. Temozolomide is an imidazotetrazine derivative synthesized in a pathway that produces imidazo-1,2,3,5-tetrazines (13). Temozolomide, which is an oral alkylating agent, is rapidly absorbed, with maximum plasma concentrations observed at 30 to 90 min after oral intake. Its plasma half-life is approximately 2 h (14). With regard to pharmacokinetics, the area under the concentration–time curve (AUC) for temozolomide in cerebrospinal fluid corresponds to approximately 20% of the plasma AUC (15). Temozolomide methylates the O⁶-position of guanine, which leads to cell-cycle arrest in the radiosensitive G₂/M phase (16). It is also known that temozolomide can suppress ionizing radiation-induced migration and invasion in glioma cells (17).

In the clinical setting, the efficacy of combined temozolomide and RT in patients with GBM has been shown in a series of studies performed by Stupp et al. (3, 18, 19). In a randomized phase III trial including 573 patients with GBM, the European Organization for Research and Treatment of Cancer (EORTC) 26981-22981/National Cancer Institute of Canada Clinical Trials Group CE3 (NCIC) established temozolomide as the standard chemotherapeutic agent for the treatment of GBM (3). An improvement in MST from 12.1 to 14.6 months was observed due to addition of temozolomide to RT. A survival benefit was observed for all patients, regardless of performance status or age. These results confirmed that the addition of temozolomide to RT is an effective standard therapy for GBM. To date, the standard treatment of patients with GBM is still considered to be postoperative chemoradiotherapy (CRT) with temozolomide followed by six cycles of adjuvant temozolomide.

Carmustine wafers. A carmustine wafer is a form of medication of carmustine that can be placed and left in the cavity after surgery for GBM, acting as a source of local chemotherapy delivery. This wafer circumvents the challenges imposed by the BBB. Recently, Affronti et al. showed the effects of carmustine wafers on the survival of patients with newly diagnosed GBM treated with concurrent temozolomide plus RT plus rotational chemotherapy (20). The overall 1- and 2-year survival times, and MSTs for the non-carmustine wafer cohort versus the carmustine wafer cohort were 69% versus 81%, 29% versus 47%, 72.7 weeks and 89.5 weeks, respectively. Although this study was retrospective, the proportion of patients in the carmustine wafer cohort who lived longer than predicted based upon the results of the Stupp regimen was significantly increased. By contrast, Bock et al. reported that the combination of carmustine wafers and concomitant CRT should be carefully considered due to its significant toxicity (21). In their study of 44 patients with newly diagnosed GBM who received the carmustine wafer-supplemented regimen, 19 (43%) experienced grade 3 or 4 adverse events, including hematotoxic events. This percentage was much higher than that previously reported by Stupp et al. (3). Therefore, prospective trials are required to rigorously compare the carmustine wafer-supplemented regimen to the Stupp regimen.

Molecular-targeted Agents for GBM Treatment

Epidermal growth factor receptor (EGFR) inhibitors. Several molecular-targeted therapy agents have been tested in patients with newly-diagnosed GBM. EGFR is a tyrosine kinase receptor that plays important roles in cell survival, proliferation, migration, and differentiation in many types of cancers (22). Pre-clinical data have suggested that overexpression of EGFR confers radiation resistance in malignant glioma and that the antagonism of EGFR restores radiosensitivity (23). Therefore, the clinical efficacies of small-molecule tyrosine kinase inhibitors (TKIs), such as gefitinib and erlotinib, have been investigated. Mellinghoff et al. critically reviewed a small subset of patients with recurrent malignant glioma enrolled in three multi-Institutional clinical trials who achieved a partial or minor response after treatment with gefitinib or erlotinib (24). However, RTOG 0211, which was a single-arm phase I/II study, showed that the addition of gefitinib to RT did not improve outcome in a general population of patients with newly-diagnosed GBM. EGFR expression analysis was not of prognostic value in patients treated with RT and concurrent and adjuvant gefitinib (25). Taken together, further evaluation of the combined use of RT, temozolomide, and TKIs is warranted in a larger prospective trial, including additional studies to elucidate the underlying functional mechanisms at the molecular level.

Vascular endothelial growth factor (VEGF) inhibitors. GBMs are highly vascularized tumors that rely on angiogenesis and secrete an abundance of VEGF (26). Bevacizumab, which is a humanized monoclonal antibody that targets VEGF, has shown potential antitumor effects in patients with GBM. In a phase II trial performed by Lai et al., 70 patients were treated with bevacizumab, temozolomide, and concurrent RT of 60 Gy in 30 fractions after surgery, followed by maintenance bevacizumab and temozolomide after the completion of concurrent therapy for 24 months or until disease progression, at which time bevacizumab alone was continued (27). Compared to historical controls, the addition of bevacizumab resulted in improved PFS but no significant change in MST; the PFS and MST were 13.6 and 19.6 months, respectively, compared to the EORTC/NCIC trial cohort (3, 18). Toxicity attributable to RT/temozolomide was similar to that observed in the EORTC/NCIC trial, and additional toxicities were consistent with those reported in previous bevacizumab trials (28). To determine whether the addition of bevacizumab to temozolomide and RT improves treatment efficacy, a multi-Institutional randomized phase III trial (RTOG 0825) was performed (29), in which a total of 978 patients were registered, and 637 underwent randomization. Although the median PFS was longer in the bevacizumab group than the placebo group (10.7 months versus 7.3 months), there was no significant difference in the duration of overall survival between these two groups (medians of 15.7 and 16.1 months, respectively). Furthermore, higher rates of neurocognitive decline, increased symptom severity, and a decline in health-related quality of life were observed over time in patients treated with bevacizumab. The AVAglio trial, which was very similar in design to RTOG 0825 (including provisions for crossover) and involved 921 patients, also showed an improvement in PFS, but there was no significant difference in the duration of overall survival for the bevacizumab group (30). Considering the slight benefit on PFS and its higher toxicity rates, the addition of bevacizumab to combined CRT and temozolomide treatment does not seem to represent an effective therapy for patients with newly-diagnosed GBM.

Antagonists of the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) glutamate receptor. Glutamate is a major excitatory neurotransmitter in the central nervous system. Glioma cells release glutamate in concentrations that are toxic to surrounding neurons and to glia (31). Recent studies have suggested that the glutamatergic system also plays a key role in the proliferation, survival, and migration of gliomas, perhaps by the activation of the Akt pathway (32-34). Talampanel is an oral, non-competitive antagonist of the AMPA subtype of glutamate excitatory amino acid receptors, with excellent brain penetration (35). A phase II trial designed to estimate overall survival in patients with newly-diagnosed GBM treated with talampanel in addition to standard RT and temozolomide reported an MST of 18.3 months. This study also showed that talampanel was well-tolerated and did not increase the known hematological or non-hematological toxicities of temozolomide. Most notably, fewer patients in this talampanel study had methylated MGMT compared to the EORTC study (29% versus 43%, respectively). Despite these findings, the overall survival of comparable patients receiving talampanel was superior (2-year survival of 42% versus 27%, respectively). A randomized, placebo-controlled trial is needed to rigorously assess the value of this promising therapeutic approach.

Immunotherapies for GBM

Antitumor vaccines. Recent studies have indicated that antitumor immunotherapy for GBM may be effective and well-tolerated in clinical trials. Antitumor vaccines represent the most established and well-studied immunotherapeutic modality. Several known tumor-associated antigens are currently being targeted in patients with GBM, including HER2, TRP2, GP100, MAGE1, IL13a2, and AIM2 (36). The most extensively studied vaccine strategies have targeted the tumor-restricted neo-antigen EGFR variant III (EGFRvIII) (37). EGFRvIII is a constitutively activated and immunogenic mutation that is not expressed in normal tissues but is widely expressed in GBM (38). For patients with GBM who survive for a year or longer after diagnosis, the expression of EGFRvIII is an independent negative prognostic indicator of survival (39). Thus, EGFRvIII is thought to be an ideal target for antitumor immunotherapy.

A phase II multi-center trial was undertaken to assess the immunogenicity and efficacy of an EGFRvIII-targeted peptide vaccine called PEPvIII (40) (which is also known as rindopepimut or CDX-110). In this trial, adults with newly-diagnosed EGFRvIII-expressing GBM with gross total resection (>95%) and a Karnofsky performance status of ≥80% who had no evidence of progression after standard concurrent CRT using temozolomide were eligible for vaccination. No significant adverse effects were reported, and the MST among this cohort was 26.0 months. At recurrence, 82% of patients had lost EGFRvIII expression. Although the effects of RT and chemotherapy on EGFRvIII have not yet been fully-characterized (41), EGFRvIII-targeted vaccination in patients with GBM is a promising strategy. A phase III randomized trial to investigate the efficacy of the addition of the EGFRvIII-targeted peptide to the current standard of care in patients with newly diagnosed GBM is ongoing (NCT01480479).

The Wilms' tumor 1 (WT1) peptide vaccine has also been tested in patients with GBM. The WT1 gene was identified as a gene responsible for Wilms' tumor. It encodes a zinc finger transcription factor that is involved in cell proliferation and differentiation, apoptosis, and organ development (42). The wild-type WT1 gene was shown to be overexpressed in various types of solid tumors. Thus, the WT1 protein was found to be an attractive target antigen for immunotherapy to treat these malignancies (43). Izumoto et al. reported the results of a phase II trial of WT1 peptide vaccination for 21 patients with WT1/human leukocyte antigen (HLA)-A*2402–positive recurrent GBM (44). The protocol was well-tolerated, and only local erythema occurred at the WT1 vaccine injection site. The disease control rate (cases with complete or partial response as well as those in which disease was stable) was 57.1%. The median PFS was 20.0 weeks, and the 6-month PFS was 33.3%. WT1 vaccine therapy may represent a promising treatment strategy for patients with WT1/HLA-A*2402-positive newly-diagnosed GBM. Further clinical studies of WT1 vaccine therapy or the addition of WT1 to the combination CRT/temozolomide treatment in patients with newly diagnosed GBM are warranted.

Immune cell therapy. To enhance the therapeutic response of malignant gliomas, an adoptive cell transfer therapy using ex vivo-activated autologous lymphocytes has been considered to be a promising approach. Autologous lymphokine-activated killer (LAK) cells have been administered to patients with glioma in clinical studies with some positive results (45, 46). However, LAK cell therapy alone is not considered very efficient for clinical use because of its low killing activity, non-specificity, and the side-effects associated with simultaneous IL2 administration (45, 46).

Administration of tumor antigen-loaded dendritic cells (DCs) has the theoretical advantage of activating host lymphocytes while bypassing the requirement for antigen processing by host antigen-presenting cells, which are often dysfunctional in patients with cancer (47). In comparison with LAK therapy, DC administration may be a good strategy from the tumor specificity point of view. Phase I and II studies have demonstrated the efficacy and safety for the use of DC vaccines for the treatment of GBM (48, 49). Ardon et al. reported the results of an HGG-2006 phase I/II, single-arm trial (50). This study evaluated the clinical efficacy of the full integration of DC-based tumor vaccination into standard postoperative CRT in 77 patients with newly-diagnosed GBM. Tumor cell lysates and autologous DCs were administered every four weeks after a 6-week course of concomitant CRT. The results showed tolerable toxicity, and the MST was 18.3 months, which compares favorably with the survival data reported by Stupp et al., reporting a median OS of 14.6 months (3). These results were used to develop the currently running phase II randomized clinical trial (NCT01280552). However, the immune profiles in the peripheral lymphocyte or delayed-type hypersensitivity skin tests did not reveal any correlation with clinical outcome in the previous studies. A possible explanation for this lack of correlation between the immunological and clinical responses might be that the peripheral immune status does not mirror the immune response that occurs in the tumor itself. Establishing immunological predictors of clinical outcome appears to be necessary to enhance treatment strategies involving the use of DCs.

Immune checkpoint inhibitors. The discovery of immune checkpoints and the development of agents targeting these pathways have led to a novel strategy in cancer therapy. Cytotoxic T lymphocyte antigen-4-blocking antibodies, programmed cell death-1 (PD1)-blocking antibodies, and PD1 ligand antibodies have shown promising results in the treatments of several types of tumors (51-53). These immune checkpoint molecules have not yet been used for the treatment of patients with GBM in the clinical setting. Zeng et al. reported that synergistic responses could be achieved using a combination of an antibody to PD1 and RT in a mouse model of glioma (54). RT causes the release of multiple tumor-associated antigens, acting as a vaccine and leading to the generation of immunological memory. Therefore, it is quite reasonable to administer immunotherapy in combination with RT. Zeng et al.'s study provided pre-clinical evidence to support the combination of immunotherapy and RT for the treatment of patients with GBM. Further studies are required to evaluate the efficacy of these immune checkpoint agents in the clinical setting.