Malignant melanoma has an unusually high propensity to metastasize to the brain; among stage IV patients 40-50% will develop clinically detectable intracranial disease, while on autopsy brain metastases can be detected in over 70% of patients1. Additionally, melanoma brain metastases are a marker of poor prognosis and despite improvements with systemic targeted therapies and immunotherapies, intracranial control still remains a significant challenge1. In the brain microenvironment, melanoma cells are found predominantly in a perivascular niche where they “co-opt” the vasculature to support their survival and invasion deep into the cortex2-5. Vascular co-optive growth is associated with poor survival in melanoma metastasis, but the molecular mechanisms that drive and sustain vessel co-option are poorly understood and represents a unique opportunity to target the cohesive interactions between melanoma cells and the endothelium to prevent or treat brain metastases6-8. My preliminary data has uncovered a vascular-mediated induction of transforming growth factor beta (TGFβ) and AKT activation in melanoma cells after contact with brain endothelial cells. Since PI3K/AKT activation or loss of phosphatase tensin homolog (PTEN), the negative regulator of AKT, has been strongly implicated in the pathogenesis of melanoma brain metastases, we hypothesize that vascular-induced TGFβ preferentially enhances survival and perivascular invasion of PTEN-null melanoma. Disrupting the cohesive contacts between brain-resident melanoma cells and the vasculature could reduce brain colonization and prevent perivascular dispersal throughout the brain.