SUMMARY
Treatment of glioblastoma (GBM) represents an unmet need in medicine. We have been pursuing a
therapeutic approach of delivering potent targeted and specific cytotoxins using continuously evolving
convection-enhanced delivery. Patients with GBM over-express interleukin 13 receptor alpha 2 (IL-13RA2),
EphA2, EphA3 and EphB2 receptors that are present in various pathophysiological compartments of GBM
and all four are expressed in tumor cells of the core of tumor, and in locally-infiltrating tumor cells, while EphA2
is also found in tumor neovasculature. Further, IL-13RA2, EphA2, and EphA3 are associated with, and play
crucial roles in, the pathobiology of glioma stem-like cells. Finally, the EphA3 receptor are found in M2 GBM-
associated macrophages. Thus, collectively, IL-13RA2, EphA2, EphA3 and EphB2 are over-expressed in
principal GBM compartments shown to be involved in tumor progression and/or resistance to therapies. In a
first-of-kind approach, we performed Phase I clinical trial in dogs with spontaneous gliomas, which represents
a faithful model of human disease, using a cocktail of cytotoxins targeting IL-13RA2 and EphA2 receptor. We
observed exceptional anti-tumor responses, including several near complete regressions, prolongation of
survival and excellent quality of life in this dose-finding trial, at no toxicity. In addition, we found evidence for
immune system activation during the therapy. Encouraged by these results, we pursued the novel idea of
targeting all four receptors instead of two with one pharmaceutical compound. One of the Eph receptor ligands,
ephrinA5 (eA5), binds EphA2, EphA3 and EphB2 receptors. We have thus generated an agent based on eA5
and IL-13 mutants targeting all four receptors using an IgG1 scaffold (QUAD). In our initial experiments, the
QUAD was conjugated to derivatives of Doxorubicin (Dox) or a derivative of Pseudomonas exotoxin A,
PE38QQR, to generate single pharmaceutical agents and these drug conjugates retained their binding
affinities towards the targeted receptors while demonstrating prominent killing activity on GBM cells. QUAD-
Dox and QUAD-PE38QQR conjugates have already shown prominent, long-lasting anti-tumor effects in dogs
with spontaneous glioma at no toxicity: 60, 88, and 91% of tumor volume regression in the treated dogs,
respectively. Recently, we have conjugated QUAD to DM1, a microtubule-disrupting agent. The QUAD-DM1 is
extremely potent on GBM cells with IC50s in low femtomolar range, ~50x better than the Dox/PE conjugates.
Therefore, we will continue this exciting line of research through Specific Aims as follows. In Specific Aim 1, we
will treat dogs with spontaneous newly diagnosed and recurrent high-grade gliomas with QUAD-DM1. In
Specific Aim 2, we will examine immune responses and the phenotype and genotype of recurring tumors in the
course of QUAD-DM1 therapy. Our approach addresses crucial issues of inter- and intra-tumoral heterogeneity
and evokes an in situ vaccination or so called “tumor inflaming” effect. We envision that this all-out assault,
termed by us “molecular resection”, will result in a more effective management of GBM.