Thyroid cancer is the most common endocrine malignancy and ranks as the fifth most common cancer diagnosed in women. Rising incidence of thyroid cancer is reflected by the projected 60,000 new cases in 2013. A majority of patients have differentiated thyroid cancer and are managed successfully with a combination of surgery and radioiodine (RAI) therapy. However, tumors may present or recur as RAI- refractory or metastatic, in which case they have a poorer prognosis and death is common. Anaplastic thyroid cancer, although rare, is typically unresectable at presentation, highly resistant to therapy, uniformly RAI-resistant, and associated with a median survival of less than one year. Approximately 2,000 patients with recurrent or advanced thyroid cancer die each year in the U.S. and many others suffer from progressive, symptomatic disease. More detailed understanding of the pathways involved, and novel targeted therapies are thus desperately needed for patients with advanced thyroid cancer. Numerous clinical data have recently pointed to the PI3K/PTEN/AKT pathway as a crucial player in both differentiated and anaplastic thyroid cancer. During the previous granting cycle, using genetically defined mouse models, we have demonstrated that constitutive activation of PI3K signaling predisposes to the development of metastatic thyroid cancer, that it cooperates with relevant additional relevant genetic alterationsto induce poorly differentiated and anaplastic thyroid tumors, and that these advanced tumors display rapid adaptive resistance to PI3K inhibition. Furthermore, our preliminary data suggest the existence of PI3K-dependent, AKT-independent pathways essential for thyroid transformation. The current application has two broad, long-term objectives. The first goal is to utilize a combination of in vivo, ex vivo, and in vitro approaches to further dissect the early stages of thyroid tumorigenesis and identify key signaling nodes required for thyrocyte transformation and essential for maintenance of established tumors, which may provide novel therapeutic opportunities. The second objective is to characterize in vivo the cell autonomous and non-autonomous mechanisms through which PI3K-active thyroid tumor cells develop adaptive resistance to PI3K inhibitors, and to test the efficacy of combination therapies targetingkey mediators of resistance.