The goal of the proposed research project is to develop and apply modular strategies to the development and application of positron emission tomography (PET) imaging agents for the diagnosis and staging of pancreatic adenocarcinoma (PDAC), which ranks as the fourth leading cause of cancer mortality nationally. Current diagnostic methodology lacks the specificity and sensitivity to efficiently identify PDAC lesions, and this lack of adequate diagnostic tools contributes to the high mortality of PDAC, representing a significant unmet medical need. The use of antibodies in PET imaging is a proven method for achieving high sensitivity and selectivity, and, thus, is a logical choice for th development of a diagnostic tool for PDAC that overcomes those shortfalls. 5B1, a fully-human antibody that targets an sialyl Lewisa (sLea), an antigen that is expressed on the surface of PDAC tumor cells, has shown promise in initial studies as an immunoPET imaging agent. However, antibody fragments can offer certain advantages as imaging agents, compared to their full-length counterparts. Using antibody engineering, diabodies (5B1Db) engineered from the fully-human 5B1 antibody have been developed by MabVax Therapeutics, which will be developed as a PET imaging agent using a modular radiolabeling strategy. Two strategies will be applied in order to identify the most promising candidate for eventual clinical translation: a direct approach and a pre-targeting approach will be assessed in SCID mice with tumors induced by BxPC3 pancreas cancer cells as well as a genetically engineered mouse model expressing sLea in pancreatic tumors. In both approaches, series of radionuclides (18F and 64Cu) will be conjugated to 5B1Db in order to find the best combination of radionuclide and labeling strategy that complement the pharmacokinetic, pharmacodynamics, and biodistribution properties of 5B1Db. We will compare tumor uptake and biodistribution of the combinations of 5B1Db via PET imaging in order to identify one optimal candidate(s) for further pre-clinical development. In the first approach, 5B1Db constructs will be conjugated to chelators that may be radiolabeled immediately prior to use in PET imaging. In the second approach, radiolabeling will be achieved in vivo, using the inverse [4+2] cycloaddition reaction between a chelator-conjugated trans-cylcooctene (TCO) and a tetrazine (Tz)-conjugated 5B1Db. Our pre-targeted method will involve four steps: i) the injection into the bloodstream of 5B1-Tz fragment; ii) accumulation of the antibody fragment in the tumor and concomitant clearance from the blood; iii) the injection into the bloodstream of the radiolabeled TCO conjugate; and iv) the binding of the radiolabeled TCO conjugate to the 5B1-Tz fragment followed by the rapid clearance of unbound radioactive material. These two approaches will provide the best chance of identifying a construct for further development.