ATPS-24: MULTI-COMPARTMENTAL TARGETING OF GLIOBLASTOMA THROUGH THE EphA3 RECEPTOR

Glioblastoma (GBM) is an aggressive primary brain tumor of dismal prognosis. We have been developing targeted cytotoxic therapies relying on differences primarily between tumor and normal cells. We hypothesize that to make this approach even more efficacious it needs to include other tumor environment compartments. To this end, we further explored EphA3, a protein receptor tyrosine kinase which plays a role in development and is present in GBM stem-like cells. We found high levels of EphA3 protein expression in the majority (8/10) of GBM specimens and GBM cells in culture express (6/9), and the EphA3 presence on the cell surface was demonstrated by flow cytometry. EphA3 receptor was up-regulated nearly 20 fold in GBM explant cells grown under tumorsphere-forming conditions. As GBM is heavily infiltrated by cells of myeloid origin, we examined whether EphA3 is present in those cells. By using confocal microcopy in situ we found that EphA3 co-stains with CD45+ leukocyte cells and also with CD68-expressing (pan-macrophage marker) cells. Nonetheless, the EphA3 receptor was not detected in the hippocampus in mice, but was clearly detected in an orthotropic GBM mouse model. EphrinA5 (eA5), the naturally occurring ligand for EphA3, binds two other Eph receptors; EphB2 and EphA2, and analysis of 10 GBM tumor lysates compared to normal brain confirmed over-expression of at least one of these three receptors in tumor samples. We generated a cytotoxin composed of eA5 chemically conjugated to PE38QQR, a modified Pseudomonas exotoxin A, which effectively killed GBM cells, but did not affect normal endothelial cells and mature dendritic cells. Therefore, EphA3 is an attractive candidate for molecular targeting as it is present on GBM tumor cells, immune cells of myeloid origin in the tumor microenvironment and glioma stem-like cells; while virtually absent in normal brain, mature dendritic cells and the hippocampus.