Open Menu

Principal Investigators

Tracy Batchelor, Brigham & Women’s Hospital
Michelle Monje, Stanford University
Isabel Arrillaga, Massachusetts General Hospital
Patrick Wen, Dana-Farber Cancer Institute

Title

Harvard/Stanford GTN Program: Novel targeted therapeutics for glioblastoma

Project Information

Project 1 - In preliminary studies our group has discovered that glioblastoma integrates into normal neural circuits, and that neuronal activity drives glioblastoma growth and progression through direct glutamatergic synapses between neurons and glioblastoma cells and by paracrine growth factors secreted by glutamatergic synapses. In turn, glioblastoma cells secrete glutamate to increase the excitability and consequently the activity of neurons. This glutamate-fueled, forward-feeding cycle presents a potential drug target. We hypothesize that targeting glutamatergic signaling broadly to disrupt these neuron – glioblastoma interactions will decrease neuronal hyperexcitability, decrease neuron-to-glioblastoma signaling, and decrease glioblastoma growth. This hypothesis makes testable predictions that can be assessed using troriluzole – a brain penetrant drug in advanced phase clinical trials for neurological and neuropsychiatric diseases. In preclinical studies, we find that troriluzole decreases glutamate in the glioblastoma microenvironment and increases survival in a murine model of glioblastoma. Going forward, our study plan has two specific aims. In Aim 1 we will test troriluzole in patient-derived orthotopic xenograft (PDOX) models of IDH WT glioblastoma, and in Aim 2 we will conduct a surgical window-of-opportunity clinical trial of troriluzole in adult subjects with recurrent IDH WT glioblastoma. We will assess effects of troriluzole on glioblastoma electrophysiology in both preclinical models and in the surgical window-of-opportunity trial using intraoperative electrocorticography to determine the effects of troriluzole on neuronal hyperexcitability. Glutamate and drug levels will be measured in both xenograft tissue and in resected human glioblastoma tissue using mass spectrometry imaging; glutamate levels will be further assessed in human subjects using perioperative microdialysis and magnetic resonance spectrometry imaging. We will examine the PDOX tissue and resected human tissue for biomarkers of neuron-glioblastoma signaling, including levels of neuroligin-3 and phosphorylated AMPA receptors (an indicator of synaptic signaling in glioblastoma through a subtype of glutamate receptor). We will assess effects of troriluzole on glioblastoma proliferation in both glioblastoma PDOX models and in resected human tumor tissue, and we will measure overall survival in the preclinical models and progression-free survival in the clinical trial. Together, these studies will elucidate the efficacy of troriluzole to decrease glutamatergic signaling in the glioblastoma microenvironment and disrupt these pathogenic neuron- glioblastoma interactions that robustly promote glioblastoma progression.

Project 2 - The epidermal growth factor receptor (EGFR) gene is mutated or amplified in over half of glioblastomas, and its mutation and focal amplification correlate with a more aggressive disease course. However, EGFR-directed tyrosine kinase inhibitors (TKIs) have failed to show efficacy in this disease and these failures cannot be attributed strictly to poor brain penetrance. We posit that the failure to date of EGFR TKIs for glioblastoma reflects the lack of a therapeutic window. A lesson learned from application of EGFR inhibitors in non-small cell lung cancer (NSCLC) is that mutant-selectivity is absolutely required. Without selectivity, systemic inhibition of wild-type (WT) EGFR signaling is the dose-limiting toxicity. In NSCLC, activating mutations in the tyrosine kinase domain confer enhanced sensitivity to certain EGFR TKIs relative to WT EGFR, allowing true mutant-selective inhibition. The EGFR genetic aberrations in glioblastoma create constitutive, ligand-independent signaling via signal generating domains that are almost exclusively WT in structure. Our objective is to create EGFR TKIs with a therapeutic window for aberrant EGFR signaling in glioblastoma. We have two specific aims in this project. In Aim 1 we will test the hypothesis that an EGFR TKI with an allosteric mechanism of action will selectively block ligand-independent EGFR signaling in glioblastoma while sparing ligand-activated EGFR systemically, thereby providing a therapeutic window that allows for effective treatment of EGFR-driven glioblastomas. In preliminary studies, we have developed small-molecule allosteric inhibitors that potently inhibit WT EGFR (IC50 < 100 nM in biochemical assays) and have a good oral mouse PK profiles and are brain-penetrant. Guided by efficacy studies in patient-derived glioblastoma neurosphere and xenograft models, we expect to identify a compound suitable for clinical development in the first grant year, to enable clinical translation in the out years. In Aim 2 we will study CM93, a novel third generation EGFR TKI that is highly brain-penetrant – so much so that it displays a positive brain/plasma ratio. We will test the hypothesis that CM93 can provide a de facto “tissue-based” therapeutic window allowing effective inhibition of EGFR in the tumor with relative sparing of the receptor systemically. Towards this end, we will conduct a first-in-human, phase 1, dose-escalation and dose-expansion study, as well as a surgical “window of opportunity” study to determine the maximum tolerated dose, evaluate the safety, pharmacokinetics, pharmacodynamics, and clinical effects of orally administered CM93 in subjects with recurrent glioblastoma characterized by EGFR mutation or amplification. Studies on clinical materials will be facilitated by our Pharmacological and Genomics Imaging Core (PGIC). The PGIC will allow us to quantify intra-tumoral accumulation of CM93 using matrix-assisted laser desorption ionization mass spectrometry imaging mass spectrometry imaging (MALDI-MSI) and to define the impact of CM93 treatment on tumor heterogeneity using single cell sequencing. Collectively, these studies promise to yield new targeted therapeutics for EGFR-driven glioblastomas and provide a molecular understanding of determinants of sensitivity and resistance to these agents.

Project 3 - The standard treatment for newly diagnosed glioblastoma has not changed since 2005 and no new medical therapies have been approved for adult glioblastoma in the last decade. In response to this challenge, we have devised a new way to treat glioblastoma that have a mutation in either of the IDH1 or IDH2 genes (now termed, astrocytoma, grade 4, IDH-mutant). Collectively, IDH mutations are present in ~20% of adult diffuse gliomas, indicating that any treatment advance in this patient population would have broad impact. Based on our knowledge that IDH mutations cause profound metabolic reprogramming in glioma cells, we used a novel pharmacological screening platform to systematically identify vulnerabilities that result from this process. We discovered that a class of drugs targeting nucleotide metabolism preferentially kill glioma cells with IDH mutations, thereby revealing an avenue for tumor-selective, biomarker-guided therapy that is poised for rapid clinical translation. To build on this discovery and translate exploitation of this vulnerability to the clinic, we propose a phase 0 surgical window clinical trial of a brain-penetrant nucleotide metabolism inhibitor in patients with astrocytoma, grade 4, IDH-mutant. We will characterize response to this agent by addressing three Specific Aims. In Aim 1 we will use MALDI-MSI, conventional liquid chromatography-mass spectrometry, and magnetic resonance spectroscopy to comprehensively characterize the pharmacokinetic and pharmacodynamic properties of this targeted therapeutic in glioma patients. In Aim 2 we will investigate how this inhibitor alters the biology of astrocytoma, grade 4, IDH-mutant tumors at the molecular and cellular levels by analyzing resected primary tissue samples via single-cell RNA sequencing and immunohistochemistry. Finally, in Aim 3 we will evaluate the safety and tolerability of this drug in a focused cohort of patients. Taken together, our work will outline and test a new treatment strategy for astrocytoma, grade 4, IDH-mutant patients that could be expanded to a larger multi-center phase 2 study if our trial is successful. Furthermore, our efforts to elucidate key components of the mechanism of action of this nucleotide metabolism inhibitor are expected to inform the rational design of combination therapies centered on this agent that can be explored in future studies.