The long-term goals of this work are the development of therapeutic strategies to improve the survival of patients with disseminated melanoma. There is already good evidence that mutated BRAF is a bona fide therapeutic target in over 50% of melanomas, and that impressive but short-lived clinical responses can be achieved with small molecule BRAF inhibitors (such as PLX4032). Clinically, BRAF inhibitor resistance follows a course where tumor regression is followed by quiescence and eventual relapse. The overall hypothesis is that BRAF inhibition remodels both the melanoma and host microenvironments to provide a protective "sanctuary" for the minor populations of melanoma cells that escape therapy. Conceptually, it is believed that there are at least two forms of environment-mediated therapeutic escape; the first, defined as "tumor intrinsic", was observed in melanoma cell lines that lack PTEN function (PTEN-) and involved the establishment of autocrine integrin/extracellular matrix (ECM) signaling loops that bypass BRAF signaling and downregulate apoptosis. The second was host-mediated, where BRAF inhibition in primary human fibroblasts paradoxically activated AKT and MEK signaling leading to the expression of PDGF-D, VEGF and the Notch ligand Jagged as well as increasing the deposition of fibronectin, collagens and laminin. The first aim will investigate how BRAF inhibition leads to the acquisition of autocrine ECM-driven signaling loops and will determine how these altered adhesion signals "re-wire" the signaling of the escaping population and leads to the adoption of a phenotype that is slow-proliferating, apoptosis resistant and pro-invasive. We will then test whether therapeutic targeting of the melanoma cell/ECM interactions ameliorates intrinsic resistance by preventing the microenvironment- mediated reorganization of the melanoma signaling network. In aim 2, we will test the hypothesis that BRAF inhibition activates host fibroblasts leading to the creation of a "refuge" vascular microenvironment that allows PTEN+ melanoma cells to escape from therapy. We will address how BRAF inhibition in normal host fibroblasts leads to the establishment of autocrine growth factor signaling loops that drives their activation. We will then use novel 3D melanoma/fibroblast/endothelial cell co-culture and in vivo xenograft models to investigate the mechanisms by which inhibition of BRAF in fibroblasts drives the angiogenic response and will elucidate the role of the vascular niche as a protective "sanctuary" for the escaping melanoma cells. It is expected that knowledge gained from this work will provide novel therapeutic strategies for overcoming BRAF inhibitor resistance in the clinic. 1