Recent clinical results attest that adoptive T cell therapy is a promising treatment modality, at least for some cancers. However, it is yet to be determined how anti-tumor cytotoxic T lymphocyte (CTL) grafts for optimal adoptive therapy should be generated ex vivo. Our long-term goal is to establish a standardized T cell culture system that enables the widespread use of adoptive T cell therapy for cancer. We previously developed an artificial antigen-presenting cell (aAPC) that can generate ex vivo large numbers of long-lived CTL with a memory phenotype. An ongoing clinical trial demonstrates that, after transfer, these CTL can expand and persist in vivo for longer than 6 months. This suggests that ex vivo generated CTL with a memory phenotype can indeed persist and function as memory T cells in vivo. However, complete clinical responses are yet to be seen in this early phase clinical trial. The overall objective of this proposal is to develop a second generation aAPC that can generate anti-tumor CTL with improved tumor recognition. We hypothesize that anti-tumor CTL with high functional avidity selectively enriched and expanded ex vivo using a modified aAPC can efficiently reject tumor cells in vivo. To generate highly avid anti-tumor CTL and test this hypothesis, we will modify our aAPC by incorporating CD8 independent T cell receptor (TCR) stimulation (Aim 1), tumor necrosis factor receptor (TNFR) costimulation (Aim 2), and CD4+ T cell help (Aim 3). In Aim 1, we will generate anti-tumor CTL using a mutant aAPC, which expresses mutated HLA incapable of binding CD8, and study the effects of CD8 coreceptor independent TCR engagement on the ex vivo generation of highly avid anti-tumor CTL. In Aim 2, we will generate anti-tumor CTL using an aAPC individually expressing various ligands for the TNFR family and study the effects of TNFR costimulation on the ex vivo enrichment, expansion, and survival of highly avid anti-tumor CTL. In Aim 3, using an aAPC that co-expresses HLA class I and II molecules, we will generate anti-tumor CTL in the presence of CD4+ T cells and study the effects of CD4+ T cell help on the ex vivo acquisition and maintenance of a memory phenotype by highly avid anti-tumor CTL. Finally, in Aim 4, we will establish a second generation aAPC by incorporating all three modifications to HLA class I (Aim 1), TNFR ligand(s) (Aim 2), and HLA class II (Aim 3). Highly avid anti-tumor CTL generated using this improved aAPC will be tested for enhanced in vitro and in vivo anti-tumor activity. Taken all together, these studies will provide critical information concerning the ex vivo generation of anti-tumor CTL that can effectively eradicate tumor cells and induce sustained clinical responses in cancer patients.