Abstract Biomaterial-based ex vivo models of healthy and diseased brain tissue are among those tools that may help study the complexity of cerebral tissue to better diagnose, prevent, and treat brain diseases. We have established engineered brain tumor biomaterials based on methacrylamide-functionalized gelatin hydrogels and used microfluidic-forming techniques to generate platforms that combine transitions of biophysical and biomolecular properties found in the glioblastoma tumor microenvironment, from the core to the tumor margins (1). Moreover, this biomaterial approach is able to monitor the response of patient-derived xenograft (PDX) cell populations with different molecular signatures; in particular, we have analyzed the amplified and the constitutively activated vIII mutant EGF receptor. We have also characterized PDX response to targeted inhibitors, such as erlotinib, a tyrosine kinase inhibitor that specifically blocks EGFR (2). Patients with glioblastoma (GBM) exhibit poor survival rates, tied both to the significant intra and interpatient heterogeneity of the tumor as well as the complex signaling pathways underlying malignancy. Spatial and temporal gradients regulate cell proliferation, migration, and differentiation during cancer. Therefore, the use of a microfluidic approach to fabricate patterned biomaterials, have the ability to examine transitions between defined environments (e.g., glioma core, periphery and neural tissue). Cells within these structured materials can be fully analyzed by imaging, secretomic and gene expression analyses. Cell morphology evolved to a more spread shape from low to high crosslinking density of gelatin and hypoxic levels depended on cell concentration. Using patient derived GBM tumor samples we are able to generate a miniaturized tumor tissue analog to examine how the heterogeneities within the tumor microenvironment impact glioma malignant phenotype, growth, and therapeutic efficacy. The ability to manipulate tumor extracellular matrix (ECM) can improve therapeutic outcomes and restrict glioma infiltration. Our in vitro tumor model is able to analyze the relationship between HA and the invasive phenotype of GBM tumor cells. We show that cell growth, motility and proteomic responses of GBM cells within our platform were significantly altered by HA molecular weight in response to a tyrosine kinase inhibitor. These results provide additional insights regarding the importance of extracellular microenvironment in the invasive potential of GBM tumors. We have demonstrated that repeated dosing of erlotinib promotes expression of PDGFRb in EGFR vIII mutant cells and that extracellular HA plays a favorable role in the inhibition of EGFR, through STAT3 deactivation (3). (1) Pedron S et al. (2015), Adv. Mater., 27: 1567; (2) Pedron S. et al. (2017), Adv. Healthcare Mater., 6: 1700529; (3) Pedron S, et al. (2019), Biomaterials, 219: 119371. Citation Format: Sara Pedron, Jann N. Sarkaria, Brendan A.C. Harley. Engineered glioblastoma tumor models reveal extracellular matrix signals influence the efficacy of targeted inhibitors [abstract]. In: Proceedings of the AACR Virtual Special Conference on the Evolving Tumor Microenvironment in Cancer Progression: Mechanisms and Emerging Therapeutic Opportunities; in association with the Tumor Microenvironment (TME) Working Group; 2021 Jan 11-12. Philadelphia (PA): AACR; Cancer Res 2021;81(5 Suppl):Abstract nr PO035.