The objective of this project is to develop a multivalent dendritic cell (DC) vaccine to be used to generate an anti-tumor cellular immune response in patients with pancreatic adenocarcinoma. Pancreas cancer is a common malignancy with an extremely poor prognosis. Surgical resection is the only curative treatment, and even after resection of localized tumor, the majority of patients relapse quickly and die from systemic metastasis. Despite many trials of multimodality therapy, effects of adjuvant or systemic therapy are minimal. We hypothesize that a multivalent DC vaccine designed by peptide pulsing with known HLA- A2 restricted pancreas cancer tumor associated antigens will generate a more effective population of tumor-specific cytotoxic T-lymphocytes (CTL) when compared with that generated by a single antigen. Additionally, we seek to compare the efficacy of multi-specific CTL against autologous pancreas cancer cells using two different methods of CTL generation: stimulation with peptide pulsed DCs using three known HLA-A2 restricted antigens and transfection of DCs with a minigene containing these three epitopes. Specifically, we aim to: (1)Test the in vitro efficacy of CTL generated from stimulation with peptide pulsed DCs using three known HLA-A2 restricted peptides derived from three different antigens (HER2/neu, mutated K-ras and telomerase). (2) Compare cytotoxicity of CTL generated by different DC vaccine strategies, i.e., peptide-pulsed DCs (exogenous loading) vs. DCs transfected with a minigene encoding all three peptide antigens (endogenous processing). (3) Test efficacy of peptide-specific CTL generation and toxicity in an HLA-A2 transgenic murine model (4) Conduct a phase I clinical trial of a multivalent DC vaccine in pancreas cancer patients.