778. A Novel Genetically Modified Adenovirus Vector Shows Enhanced Infectivity towards Ex Vivo Pancreatic Cancer

Abstract

Top of pageAbstract Conditionally replicating adenovirus (CRAd) represents a potential therapeutic option for pancreatic cancer (PC). Clinical trials have shown safety and specificity of such adenovirus (Ad) vectors. However, transduction of tumor cells is limited by the low expression of the primary Adenovirus Receptor (CAR) on PC. To overcome this problem, Ad vectors were developed with peptides in the fiber that should allow targeting to PC. A peptide ligand for the Ephrin A2 receptor was incorporated and this vector (Ad-YSA) showed an increase of tumor cell transduction in vitro. Ad-RGD served as a positive control. Since animal models have proven to be of little value in predicting the outcome of oncolytic virotherapy, we used the tissue slice system to explore modified Ad behavior in human tissue. Fresh tissue from normal pancreas and pancreatic adenocarcinoma was obtained from surgical specimens (Whipple procedure). Slices, generated with the Krumdieck tissue slicer, were cultured in medium supplemented with growth factors in a 5/95% CO2/O2 environment. The slices were infected with 1e8 to 5e8 viral particles (wildtype and fiber modified virus harboring the green fluorescent reporter protein (GFP) gene under control of a CMV promoter (Ad.CMV.GFP). As an internal control, to correct for slice size, viability and CAR expression, an equivalent amount of wildtype Ad.CMV.dsRED per slice was added. Three days later reporter protein production was measured by fluorimetry. Morphologic features were determined by HE-staining. These experiments were performed both with primary pancreatic tissue and with tumor tissue. According to morphology, pancreas and tumor tissue could be kept alive for at least three days. Normal pancreas did not show increased transduction efficacy with Ad-RGD or Ad-YSA. Compared to wildtype Ad, Ad-RGD and Ad-YSA revealed a 1.5 to 8.6 and 1.8 to 7.6-fold increase in transduction of pancreatic cancer, respectively (n=7). We observed persistent interpatient variability of infection profiles, which may be related to differential receptor expression. We currently examine CAR, integrin and EphA2 receptor expression in the tissue slices. Our results demonstrate that targeting to the ephrin A2 receptor can augment pancreatic cancer transduction. The tissue slicer technology appears to be of great value to study host-virus interactions. (See Figure)