Many cancer patients are still found with advanced/spread diseases upon initial diagnosis. Particularly, pancreatic cancer is the 4th leading cause of cancer related death, and less than 10% of the patients are the subject of surgical resection. Treatment of spread disease, which cannot be handled solely by locoregional therapy, is still a big challenge. Thus, novel systemic treatment of advanced cancer is in high demand. Compared to loco-regional therapy, systemic application of cancer gene therapy mandates more efficient and selective gene delivery, and also needs to embody sufficient antitumor effect even with low initial transduction. Oncolytic viruses have been tried in systemic diseases. Adenovirus (Ad) shows efficient in vivo transduction mediated by specific protein-to-protein binding, and its exponential replication lead to oncolysis. If the vector distribution and transduction can be controlled, such a vector would overcome the obstacles for systemic delivery and enable the treatment of advanced cancer. However, incorporation of functional ligand into virus virion is extremely difficult due to interference of virus assembly or ligand affinity. One promising way to overcome this obstacle is high-throughput screening of the high-diversity ligand- peptide library in the format of Ad capsid from the beginning. Recently, we have achieved a drastic increase of Ad library diversity (5 x 109) by employing our novel Ad production system. High throughput screening of this library with a cell line expressing a pancreatic cancer marker, mesothelin (msln), identified a targeting motif showing potent and selective infectivity to msln expressing cells. Thus, high throughput screening of high diversity Ad library is a powerful and practical way to development transductionally targeted vectors for various cancers. Our system has two major benefits over existing oncolytic Ad systems: 1) transductional targeting and reduction of sequestration to non-cancer cells provides tumor targeting at systemic level as well as in situ cancer cell targeting in the tumor, and 2) a high diversity Ad-library in replication competent platform permits the high-throughput direct screening in vitro and in vivo. In this project, we will develop oncolytic Ads enabling systemic treatment of pancreatic cancer. The Ad fiber libraries with vector backbones showing reduced liver sequestration will undergo high throughput screening in vitro and in vivo. The identified capsid structures will be combined with promoter-based replication control for stringent cancer selectivity. The resultant oncolytic viruses will be assessed first in vitro and next in subcutaneous tumor model in vivo. Next, the lead vectors will be assessed in orthotopic pancreatic cancer models with cell lines and patient tumor transplants. The development of the potent and systemically-injectable oncolytic Ad will be able to improve the clinical outcome of the cancer patients, including those with unrespectable pancreatic cancer and other neoplasms.