Abstract 611: A complex 3D model of glioblastoma for patient-specific drug response profiling

Approximately 70,000 new cases of brain tumors will be diagnosed this year with glioblastoma (GBM) accounting for about 17% of those cases. Unfortunately the median survival for patients with GBM is only 14.6 months due to the complexity of the tumor, the pattern of diffuse spread within the brain, and the lack of effective therapeutic options. Complex in vitro tumor models developed in 3D better mimic in vivo biology, and, when utilizing patient-derived cells, may offer robust platforms for new drug development and patient-specific drug response profiling. We have previously shown this to be true with both breast and ovarian cancer and we have adopted this approach to modelling GBM ex vivo. We hypothesize that primary derived GBM tissues and stem cells cultured in complex 3D microenvironments can recapitulate the intra-tumor heterogeneity and drug resistance similar to that found in the clinical. To this end, we have developed 3D models of GBM that incorporate tumor cells, endothelial cells, and macrophages within a complex extracellular matrix and cultured them for up to 2 weeks under perfusion flow. We have created these models using both cell lines and primary patient cells and these 3D tissues have been analyzed for changes in viability (PrestoBlue and PicoGreen), marker expression (flow cytometry, Luminex, and protein arrays), macrophage differentiation and invasion (flow cytometry, multiphoton microscopy, and IHC), and methylation patterns (ChIP on chip arrays and methylation specific PCR). Initial optimization work using cell lines has revealed that culturing with endothelial cells and/or macrophages has an effect upon proliferation dependent upon the differentiation state of the macrophages. Additionally, co-culture conditions affect protein secretion and phosphorylation along with patterns of gene methylation. We have also examined how changes in the tumor microenvironment affect these different metrics by examining the role of hypoxic growth conditions. Work using patient cells isolated from primary glioblastoma has focused upon drug response dependent upon the tumor microenvironment and the presence of cancer stem cells utilizing a neurosphere generation assay. We have demonstrated temozolamide resistance in tumors with early relapse after initial chemo-radiation, and sensitivity to EGFR inhibitor afatanib in tumors with EGFR amplication, suggesting clinical and molecular correlation or our ex vivo 3D drug response profiling using patient derived GBM models. Taken together, our data support the further development and use of complex, 3D co-cultures to better mimic GBM in a patient-specific manner. These models are currently being considered for preclinical assessment of novel therapies, including chimeric antigen receptor T cells, and may be a useful adjunct in precision medicine applications to improve patient outcomes. Citation Format: Tessa DesRochers, Lillia Holmes, Lauren O’Donnell, Qi Guo, David Schammel, Jeff Edenfield, Charles C. Kanos, Hal E. Crosswell. A complex 3D model of glioblastoma for patient-specific drug response profiling. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 611.