Pancreatic adenocarcinoma is the fourth leading cause of cancer-related deaths in the US. Conditionally replicative oncolytic adenoviruses (CRAds) have been developed and tested in clinical trial, but fell short of clinical expectation. We have developed new CRAds for pancreatic cancer with RGD modification to facilitate efficient transduction in CAR negative pancreatic cancer cells and Cox2 promoter-based replication control, which eliminates toxicity in normal organs including liver and normal pancreas. A preclinical challenge that oncolytic viruses face is the absence of suitable animal model which permits in vivo viral replication for toxicology. Given that human adenoviral shows replication tropism to natural host, human cells, the optimal model to use is human tissue. In this study, we used tissue resected from patients undergoing operative procedures to access our adenoviral vectors. METHODS: After obtaining IRB approval, we collected normal liver, normal pancreas, and pancreatic cancer tissue from specimens resected from patients undergoing planned resections. The tissue was immediately sliced into 200 micron slices using the Krumdieck Tissue Slicer and incubated overnight. They were then infected with Ad RGD wild type or RGDCox2CRAdF vectors. On day 3, samples were collected and DNA was extracted. E4 copy number and B-actin was determined for each sample by PCR. Statistical analysis was performed by Student’s T-test. RESULTS: After compensation with B-actin, the Ad RGD Cox2 vector showed high replication in pancreatic adenocarcinoma tissues while its replication in normal liver and normal pancreas was minimal. In contrast, the replication of RGDWt was high regardless of the origin of the tissues. When the replication of these two vectors were compared in each type tissue, the percentage of compensated viral copy number RGDCox2CRAdF in pancreatic adenocarcinoma tissue was significantly higher than those in normal pancreas and normal liver (Table). CONCLUSION: RGDCox2CRAdF vectors show high specificity for pancreatic adenocarcinoma tissue compared to normal pancreas and normal liver. By using human tissue specimens, we can effectively evaluate human adenoviral replication. This model provides us with valuable preclinical data to determine specificity and toxicity of Ad vectors prior to proceeding to Phase I clinical trials.
Relative Viral Copy Number (compensated with beta-actin)

	RGDCOX2CRAdF	RGDWt	COX2CRAdF/Wt (%)
Normal Liver	1.22x10(6)	9.08x10(6)	13.49% +/- 6.09*
Normal Pancreas	2.11x10(6)	1.61x10(7)	13.09% +/- 10.05*
Pancreatic Adenocarcinoma	3.33x10(7)	3.18x10(7)	104.71% +/- 25.45

*p<0.01 when compared to Pancreatic Adenocarcinoma