Pancreatic ductal adenocarcinoma (PDAC) is highly malignant disease that is the fourth leading cause of cancer-related death in the US. Gene therapy using AAV vectors to selectively deliver genes to PDAC cells is an attractive treatment option for pancreatic cancer. However, most AAV serotypes display a broad spectrum of tissue tropism and none of the existing serotypes specifically target PDAC cells. This study tests the hypothesis that AAV2 can be genetically re-engineered to specifically target PDAC cells by modifying the capsid surface to display a peptide that has previously been shown to bind plectin-1. Toward this end, a Plectin-1 Targeting Peptide (PTP) was inserted into the loop IV region of the AAV2 capsid, and the resulting capsid (AAV-PTP) was used in a series of in vitro and in vivo experiments. In vitro, AAV-PTP was found to target all five human PDAC cell lines tested (PANC-1, MIA PaCa-2, HPAC, MPanc-96, and BxPC-3) preferentially over two non-neoplastic human pancreatic cell lines (human pancreatic ductal epithelial and human pancreatic stellate cells). In vivo, mice bearing subcutaneous tumor xenografts were generated using the PANC-1 cell line. Once tumors reached a size of ∼1–2 mm in diameter, the mice were injected intravenously with luciferase reporter vectors packaged in the either AAV-PTP or wild type AAV2 capsids. Luciferase expression was then monitored by bioluminescence imaging on days 3, 7, and 14 after vector injection. The results indicate that the AAV-PTP capsid displays a 37-fold preference for PANC-1 tumor xenographs over liver and other tissues; whereas the wild type AAV2 capsid displays a complementary preference for liver over tumors and other tissues. Together, these results establish proof-of-principle for the ability of PTP-modified AAV capsids to selectively target gene delivery to PDAC cells in vivo, which opens promising new avenues for the early detection, diagnosis, and treatment of pancreatic cancer.