Abstract Glioblastoma multiforme (GBM) is one of the deadliest forms of brain cancer (median survival time of 12-15 months). Since the cancer is unique in its ability to invade diffusely into the surrounding brain tissue, surgical resection is often ineffective. As such, it is imperative to develop methods of understanding GBM behavior and determining novel therapeutic strategies. Currently, there are limited tools to rigorously assess the relationship between the tissue microenvironment, tumor growth and invasion, and therapeutic outcomes. While most invasion studies have utilized 2D substrates or semi-3D chambers, the lack of cell-cell/cell-matrix interactions limits the ability to accurately predict GBM invasive behavior as observed in a 3D tumor environment. GBM cell invasion is influenced by mechanical and chemical heterogeneity within the extracellular matrix (ECM). The presence of hyaluronic acid (HA) and the differences in stiffness within the brain parenchyma contribute to invasive phenotype. Vasculature within the ECM has also been shown to affect the invasive behavior of GBM, as cells migrating along vascular beds experience integrin signaling through proteins in the vascular basement membrane. To understand how these multiple signals regulate GBM cell invasion, we have developed a 3D methacrylamide-functionalized gelatin (GelMA) hydrogel platform that allows for selective incorporation of matrix-immobilized HA as inspired by the varying composition and structure of brain parenchyma. This system can be further modified to incorporate structural, biophysical, and metabolic gradients inspired by GBM tumor margins and the native brain microenvironment. We developed an in vitro invasion assay utilizing GBM cell-coated dextran beads embedded into the hydrogels to effectively characterize GBM invasive phenotype as a function of margin-inspired variations in matrix stiffness and composition within the hydrogels. A family of hydrogels with varying stiffness and variants of matrix-bound HA were explored with the invasion assay. Experiments were conducted using GBM cell line U251, which is known to be highly invasive. Preliminary data showed that stiffness and HA content did not influence cell metabolic activity. Not surprisingly, we saw a significant increase in cell invasion distance in softer hydrogels compared to their stiffer counterparts. Interestingly, cells invaded less in hydrogels with HA than in those containing GelMA only. ELISA showed a higher secretion of soluble HA in GelMA-only hydrogels, which may be due to the necessity of cells to compensate for the lack of bound HA, thereby resulting in a more invasive behavior. Recent efforts have focused on the development of vessel structures within these hydrogels by using a co-culture of human umbilical vein endothelial cells (HUVECs) and normal human lung fibroblasts (NHLFs). The incorporation of vasculature will offer an additional level of complexity to the platform to study cell migration mechanisms in GBM. Preliminary experiments demonstrated that following seven- and ten-day culture periods, the vessel architectures were stable between time points. Vessel development, as measured with metrics such as vessel area, vessel length, and number of junction, was observed more robustly in hydrogels without HA. We conclude that our platform is capable of promoting differential invasive behavior of GBM cells in response to heterogeneous mechanical and chemical properties, and that our platform shows the potential to probe tumor-vessel interactions. In future work, we intend to elucidate how vessel structures, when combined with varying matrix stiffness and HA content, impact GBM invasive behavior within the tumor microenvironment. We predict that more invasive phenotype and more established migratory trajectories will be observed in the presence of vessel structures within the hydrogels. Citation Format: Jee-Wei Emily Chen, Mai Ngo, Brendan Harley. Interfacial migration patterns in glioblastoma are analyzed in structured hydrogel platforms. [abstract]. In: Proceedings of the AACR Special Conference on Engineering and Physical Sciences in Oncology; 2016 Jun 25-28; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2017;77(2 Suppl):Abstract nr A43.