Immunotherapy For Glioblastoma: A Path Forward

By Leah Strickling, Consultant, Back Bay Life Science Advisors with Peter Bak, PhD, Senior Vice President Back Bay Life Science Advisors

Glioblastoma (GBM) is a malignant, fast-growing, and aggressive primary brain tumor. Current standard-of-care (SoC) treatment includes surgery followed by radiation in combination with the chemotherapy agent temozolomide, usually not started until approximately 6 weeks after surgery. Despite numerous attempts to improve treatment outcomes, overall prognosis remains poor for newly diagnosed patients, with a two-year survival rate of <20% and median overall survival (OS) of approximately 12-14 months. Further, GBM has a near 100% relapse rate with a median time to recurrence of only approximately 7 months. Immunotherapy (IO) has revolutionized the treatment of many cancer types, but its promise to transform care across all tumor pathologies has yet to be realized in GBM.

Since the advent of the first IO approvals, various interventions, including immune checkpoint inhibition, tumor vaccines, and cell therapies have been evaluated in GBM. To date, none of these IO modalities have led to meaningful improvements in clinical outcomes. Despite the failures of first generation, single IO agent trials, there is a glimmer of hope on the horizon with encouraging, albeit early data recently released from multiple clinical studies. This article will provide a concise overview of the types of IO interventions that have been trialed in GBM, their associated challenges and shortcomings, and current approaches as a path forward for the field.

IO Failures in GBM and Path Forward

Glioblastoma (GBM) is a malignant, fast-growing, and aggressive primary brain tumor. Each year in the U.S., more than 20,000 patients are diagnosed with GBM, and ~10k recurrent patients are treated (1). Current first-line standard-of-care (SoC) treatment includes surgery followed by radiation in combination with the chemotherapy agent temozolomide, which usually are not started until ~6 weeks after surgery. Recurrent patients have no options available beyond additional chemoradiation and supportive care. Despite numerous attempts to improve treatment outcomes, overall prognosis remains poor for newly diagnosed patients, with a 2-year survival rate of <20% and a median overall survival (OS) of ~12-14 months (2). Further, GBM has a near 100% relapse rate, with a median time to recurrence of only ~7 months (1,3).

IO treatment of several extracranial solid tumors has resulted in significant survival benefit. A variety of IO interventions including immune checkpoint inhibition, tumor vaccines, and adoptive T cell technologies, have been evaluated in GBM. To date, none of these IO modalities have led to meaningful improvements in clinical outcomes for GBM patients. The pathophysiology underlying GBM renders it resistant to IO therapy approaches for several key reasons. Compared to other tumor types for which IO approaches have had success (e.g. melanoma, NSCLC, etc.), GBM is poorly immunogenic with a low mutational burden and neoantigen load (4).

Additionally, during homeostatic conditions, T cells are generally not in the brain in large numbers. While the concept of the brain as an “immune privileged” organ is continuing to evolve, there is consensus that the organ site presents unique challenges to an immune-based therapeutic modality (5). For example, though T cells can extravasate from blood vessels to lesions within the brain in the case of primary malignant brain tumor, the number of intra-tumoral CD8 T cells, which are associated with favorable outcomes and survival, remains small in GBM (0-12% of all cells) when compared to other extracranial tumor types (6). Further, glioblastomas are situated in an immune suppressive micro-environment, which negatively controls anti-tumor CD8 T cell response, hindering the efficacy of IO modalities (7). Lastly, Radiation therapy and chemotherapy used for the treatment of GBM can cause even more immunosuppression (8).

Immune Checkpoint Inhibitors

Immune Checkpoint Inhibitors (CPIs), which block inhibitory receptors and their ligands to elicit an effective and sustained anti-tumor CD8 T cell response, have revolutionized the field of oncology over the past decade, resulting in an impressive extension of survival in a number of different tumor types (9). Several well-known CPIs, mainly PD-1 and PD-L1s have been trialed in GBM across multiple settings such as the primary, recurrent, neo-adjuvant, and adjuvant settings. Unfortunately, results to-date for these molecules as single immune therapies against GBM have been modest at best.

CPI trials in GBM initially focused on the recurrent setting in an attempt to capture the ‘low hanging fruit’. BMS was the first manufacturer to target GBM, evaluating its PD-1 therapy nivolumab +/- ipilimumab vs. bevacizumab in a Phase 3 trial in patients with recurrent disease (CheckMate 143); the primary endpoint of improved OS was not achieved (10). Several additional failures in GBM followed for nivolumab, as well as disappointing results for other CPIs. A second Phase 3 trial of nivolumab (+ radiotherapy) failed to show an increase in survival in the front-line setting for MGMT-unmethylated GBM (CheckMate 498), and thereafter, nivolumab also failed to show an increase in PFS in a Phase 3 trial in the front-line MGMT-methylated setting (CheckMate 548 – results are still preliminary) (11,12). While Merck’s PD-1 pembrolizumab, the market leader CPI across other tumor types, did demonstrate an increase in OS (gain of ~6 months) in the neoadjuvant setting in a Phase 2 trial, the sample size for the trial was only n=16 patients (13). Pembrolizumab is currently being evaluated across several settings in combination with other investigational products in GBM. Merck KGaA’s PD-L1 avelumab failed its Phase 2 trial in recurrent GBM, and Astra Zeneca’s PD-L1 durvalumab demonstrated only a very modest increase (<3 mos.) in OS vs. SoC for the front-line setting in another Phase 2 trial (9). Single-agent CPI approaches have not been successful in GBM likely given that CPIs alone do not both downregulate the immunosuppressive microenvironment and increase the CD8 T cell influx to the tumor site.

Tumor Vaccines

Tumor vaccines, which induce an immune response directed against tumor antigens, have been trialed in GBM and have also not resulted in substantial clinical benefit as single agents. Tumor vaccines are generally comprised of peptides but may also consist of dendritic cells (DCs) loaded with tumor lysates or engineered to express tumor antigen. Several peptide vaccines and DC vaccines have been evaluated in Phase II and III GBM trials. Variant III of the epidermal growth factor receptor (EGFRvIII) mutation is present in ~20-30% of GBMs, and is a recognized target for many peptide vaccination trials, including Celldex Therapeutics’ Phase 3 trial of rindopepimut for primary GBM which was terminated early due to lack of efficacy despite evidence of immunogenicity in early trials (14). While several other peptide vaccines have elicited immune responses, their use as single agents has not resulted in substantial clinical benefit and no Phase 3 trials have been successful to-date. DC vaccines have also not had impressive results in GBM; while there has been one Phase III trial of a DC vaccination that showed promise in GBM, NW Bio’s DCVax-L for newly-diagnosed disease, maturation of trial data has been ongoing for over three years and KOLs have publicly expressed skepticism (15). While the interim analysis from the trial showed an eight-month OS survival benefit for the addition of DCVax-L to SoC, key stakeholders in the field have indicated potential patient selection bias (7,14). Single agent tumor vaccines trialed to-date may not be considered highly promising given that tumor antigens are scarce in GBM, and many are not shared by a large number of patients (9). Personalized vaccine approaches may be a potential path forward, along with combinational strategies.

Adoptive T Cell Therapy

Adoptive T cell therapy, involving expanding autologous T cells in vitro and returning them to the patient, has been evaluated in early GBM trials with the aim of circumventing the challenge of generating tumor-specific responses in situ. Early adoptive transfer approaches focused on the use of non-engineered therapeutic T cells, such as tumor-infiltrating lymphocytes (TILs), natural killer (NK) cells, T cell clones derived from peripheral T cells, and lymphokine-activated killer (LAK) cells. To-date, early-phase trials of these specific adoptive transfer approaches have not yet demonstrated clear clinical value in primary and recurrent GBM. For example, while GBM-infiltrating lymphocytes can be successfully expanded in vitro, the TILs demonstrate poor function and severe exhaustion (16). More recently, adoptive T cell therapy approaches have focused on utilizing engineered, or genetically modified, T cells expressing chimeric antigen receptors (CARs) or tumor-specific T cell receptors (TCRs). While only a limited number of adoptive transfer trials using engineered T-cells have been conducted so far, there has recently been momentum in the field around the promise of these therapies in GBM.

The Path Forward

Gene Engineered Adoptive T Cell Approaches — CAR-T approaches have achieved tremendous successes in treatment of hematological malignancies. The potent clinical responses from CAR-T cells in blood cancers have sparked interest in exploring the approach in solid tumors, including GBM. In early GBM trials, CAR-Ts have been shown to migrate from peripheral blood to the tumor bed. Several clinical case studies using GBM specific CAR-Ts have been reported. For example, in one patient with recurrent multifocal GBM, intracranial administration of CAR-Ts targeting interleukin-13 receptor alpha 2 (IL-13Rα2) resulted in robust anti-tumor immunity and complete tumor regression, sustained for 7.5 months (17). Despite the nascent data, this concept has generated interest from large pharma. In June, Bayer-backed Century Therapeutics bought Ontario-based Empirica Therapeutics to create CAR-Ts against GBM. Empirica’s CAR-Ts which target CD133, expressed on the surface of brain tumors, have shown strong single agent activity in preclinical models. Empirica is exploring combination strategies for its CAR-Ts as well, and there are several ongoing Phase 1 trials combining CAR-Ts with CPIs. Besides CD133, other CAR-T targets being explored for GBM include CSPG4, NKG2DL, and EGFRvIII, among others (18). On the TCR T cell therapy side, no clinical trial results have been reported for GBM to-date, though the National Cancer Institute (NCI) recently initiated two Phase II studies to assess clinical response after adoptive transfer of T cells genetically engineered to express TCRs reactive against neoantigens in patients with GBM.

CPI and Tumor Vaccine Combinations — It is likely that combination approaches are the future of IO in GBM, considering that single-agent IO approaches have not been successful in late-stage trials. One such combination being evaluated is combining tumor vaccines with CPIs. A few peptide vaccination trials have shown that vaccine-induced T cells, predominantly CD4 T cells, are able to migrate to the tumor site in the brain and express multiple immune checkpoints, suggesting that combining vaccines with a CPI could be an effective strategy (19). As such, this combination approach is being evaluated in early GBM trials. For example, Merck KGaA is currently evaluating avelumab in combination with VAXIMM’s VEGFR-2 DNA vaccine VXM01 in a Phase 1/2 trial of recurrent GBM, and early promising results from the trial were presented in May. Three of the first nine patients responded, two of which were progression-free at >6 months. Personalized tumor vaccines in combination with CPIs are also being evaluated as a promising strategy. Merck is currently studying pembrolizumab in combination with a personalized neoantigen vaccine and radiation therapy in a Phase 1 trial for newly diagnosed GBM.

Novel Cancer Vaccination Approaches — Several companies have recently been in the spotlight for their work on novel approaches to cancer vaccination in GBM. Imvax, a biotech company based in Philadelphia, raised a $112M series C to take its autologous tumor cell vaccine IGV-001 into a Phase 2 GBM trial. The company raised $40M across its series A and B rounds in 2017 and 2019, and between the rounds presented interim Phase 1b data linking its vaccine to better OS vs. a historical control based on SoC. Imvax’s technology is an autologous, cell-based treatment that activates a full array of patient-specific antigens leading to an immune response. Imvax produces the vaccine by sampling cancer cells during the surgical removal of the patient’s tumor and treating the cancer cells with an oligodeoxynucleotide against IGF-R1. The antigenic cells are then added to a delivery device in the abdomen under the skin, wherein the cancer cells release antigens that ‘take the brakes off’ of the immune system so that it can attack the post-surgery remnants of the brain tumor. Early results from this novel approach have been impressive, reinforced by the large rounds that Imvax has raised over the past few years. Further, new categories of antigens are being considered as targets for this difficult to treat cancer. For example, Enterome recently initiated a Phase 1/2 trial for EO2401, an OncoMimic, in combination with an immune checkpoint inhibitor as a potential new treatment for progressive or recurrent GBM. Enterome’s OncoMimics are microbiome-derived peptide antigens that closely mimic antigens expressed by tumor cells, which are selected based on their ability to trigger the rapid activation of memory T-cells.

While IO approaches have generally had disappointing results in GBM, there is a path forward. Gene-engineered CAR-Ts, rationally designed combination immunotherapeutic approaches, and novel vaccine approaches that sensitize GBM to IO therapies have promise in this hard-to-treat tumor type. Early results are encouraging, but in a space littered with monotherapy failures, these innovative combination approaches may hold the promise to unleash the immune system in the treatment of this deadly cancer.

ABOUT THE AUTHORS

Leah Strickling is a Consultant at Back Bay Life Science Advisors, an integrated life science strategy and investment banking firm based in Boston, MA. Ms. Strickling leads projects spanning an array of therapeutic areas, with emphasis on oncology and rare disease. Leah has advised clients developing novel approaches to address unmet needs in the oncology space, from novel surgical visualization technologies for brain tumor resection to microbial based targets to treat colorectal cancer.

Ms. Strickling’s background includes Biomedical Science studies with a focus in Pharmacology at McGill University in Montreal, Canada. 

Peter Bak, PhD is a senior vice president at Back Bay Life Science Advisors, with over a decade of experience in a broad range of research approaches and fields—from immunology and infection through oncology. Dr. Bak has published several white papers and peer-reviewed articles on cancer vaccinology, the molecular mechanisms of myeloid-cell-mediated immune-suppression, costimulatory requirements for intertumoral T-cell activation, and immune editing within the tumor environment.

To contact the authors and learn more about Back Bay’s expertise in oncology development, visit www.bblsa.com. 

REFERENCES

1. Kantar Epidemiology Reports from Prior Back Bay Projects
2. Mitrofan et al., Clinical characteristics of glioblastoma multiforme (GBM) patients who reached 400 days post diagnostic from a retrospective real-world data. Annals of Oncology CNS Tumors. 2018; 29(8).
3. Weller et al., Standards of care for treatment of recurrent glioblastoma—are we there yet? Neuro-Oncology. 2013; 15(1): 4-27.
4. Dutoit et al., Immunotherapy of malignant tumors in the brain: How different from other sites? Front. Oncol. 2016; 6, 256.
5. Orrego et al., Distribution of tumor-infiltrating immune cells in glioblastoma. CNS Oncol. 2018; 7, CNS21.
6. Karolina et al., T-cell Dysfunction in Glioblastoma: Applying a New Framework. 2018; 24(16).
7. Quail, D.F., Joyce, J.A. The microenvironmental landscape of brain tumors. Cancer Cell. 2017; 31, 326–341.
8. Sengupta et al., Impact of Temozolomide on Immune Response during Malignant Glioma Chemotherapy. Clin Dev Immunol. 2012; ID: 831090.
9. Weenink et al., Immunotherapy in Glioblastoma: Current Shortcomings and Future Perspectives. Cancers. 2020; 12, 751.
10. Reardon et al., Effect of Nivolumab vs Bevacizumab in Patients With Recurrent Glioblastoma – The CheckMate 143 Phase 3 Randomized Clinical Trial. JAMA Oncol. 2020; 6(7):1003-1010.
11. BMS Press Release, May 2019: https://news.bms.com/press-release/corporatefinancial-news/bristol-myers-squibb-announces-phase-3-checkmate-498-study-did
12. BMS Press Release, September 2019: https://news.bms.com/press-release/corporatefinancial-news/bristol-myers-squibb-provides-update-phase-3-opdivo-nivolumab-
13. Cloughesy et al., Neoadjuvant anti-PD-1 immunotherapy promotes a survival benefit with intratumoral and systemic immune responses in recurrent glioblastoma. Nat Med. 2019; 25(3): 447-486.
14. Celldex Press Release, March 2016: https://ir.celldex.com/news-releases/news-release-details/data-safety-and-monitoring-board-recommends-celldexs-phase-3
15. Pharmaceutical Technology Report – Part of GlobalData, June 2018: https://www.pharmaceutical-technology.com/comment/northwest-bios-dcvax-l-hype-warranted-glioblastoma-patients/
16. Woroniecka et al., T-cell exhaustion signatures vary with tumor type and are severe in glioblastoma. Clin. Cancer Res. 2018; 24, 4175–4186.
17.  Brown et al., Regression of glioblastoma after chimeric antigen receptor t-cell therapy. N. Engl. J. Med. 2016; 375, 2561–2569.
18.  Fierce Biotech: Startup targets glioblastoma tumors with CAR-T therapy, May 2020: https://www.fiercebiotech.com/research/canadian-startup-targets-glioblastoma-brain-tumors-car-t
19. Keskin et al., Neoantigen vaccine generates intratumoral t cell responses in phase ib glioblastoma trial. Nature 2019; 565, 234–239.