Drugs that target tumor angiogenesis can increase response rates and survival in patients with advanced malignancies, although the percentage of patients responding is generally low. Recent clinical trials suggest that most advanced pancreatic exocrine carcinomas are refractory to angiogenesis inhibitors, and this may be due to the hypovascularity and other biological features of these tumors. In contrast to the more common exocrine pancreatic cancers, malignancies originating from neuroendocrine cells in the pancreas, called pancreatic neuroendocrine tumors (PNET), lack a desmoplastic fibroinflammatory capsule and are highly vascularized. While metastatic PNET are not cured by surgery or chemotherapy, objective responses of PNET to anti-angiogenic agents have been seen in mouse models and in randomized phase III clinical trials. Since pancreatic cancers often express multiple angiogenic factors, an objective of this proposal is to determine if combinations of inhibitors targeting different angiogenic factors are effective in mouse models of PNET. The inhibitors to be studied include a novel, highly specific, orally available small molecule inhibito of the angiogenic factor thymidine phosphorylase (TP) which we have synthesized, and two highly specific peptibodies which neutralize the angiogenic factors angiopoietin-1 (Ang-1) and Ang-2. The second objective of this proposal is to determine the role of TP, and the mechanisms of action of our TP inhibitor, in vivo. Included in this objective are studies that are based on the unexpected discovery of a direct link between the angiopoietin/TIE system and TP, observed in a subset of tumor-associated macrophages (TAMs) called TEMs (TIE2-expressing monocytes/ macrophages). These studies could form the basis for new therapies to treat pancreatic cancer, and could be readily extrapolated to other solid tumor types. The specific aims are: 1) to conduct pharmacokinetic and pharmacodynamic (PK/PD) studies of AEAC in normal and tumor-bearing mice, and to determine if it produces host toxicity; 2) to determine the anti- angiogenic and antitumor activities of AEAC when used in combination inhibitors of angiopoietin and/or VEGF pathways; 3) to determine the mechanisms of action of TP and AEAC in vivo, including effects on downstream signaling events in tumor endothelial cells and effects on vasculature morphology and normalization, and to determine if TP expression is regulated in vivo by Ang-2; and 4) to use genetic approaches to assess the effect of TP on tumor vascularization and progression in vivo, including the manipulation of the pancreatic cancer cells and tumor stromal cells, particularly TAMs and TEMs. Among the in vivo models to be used is a genetic mouse model which has a conditional knockout of the Men1 gene in the endocrine pancreatic cells (Pdx1- Cre;Men1 mice). These mice develop pancreatic endocrine tumors that closely mimic the human disease both in genetic origin and phenotype, including increased expression of angiogenic factors, and respond therapeutically to treatment with angiogenesis inhibitors.