Our goal is to improve survival in older acute myeloid leukemia (AML) patients using optimized, genetically engineered, autologous AML cell vaccines. Compelling evidence for the efficacy of immunotherapy in eliminating minimal residual disease (MRD) is provided by the superior outcomes of allogeneic hematopoietic stem cell transplants (HSCT) and donor leukocyte infusion (DLI) due to graft vs leukemia (GVL) effects. However, patients > 60 yo are often ineligible for allo-HSCT, due to co-morbidities or lack of a donor, and have dismal outcomes. Our hypothesis is that co-expression of IL-15 and IL-15Rα by autologous AML vaccines combined with CD80 expression, will elicit potent anti-leukemic responses that will improve relapse-free survival in older patients after remission induction. Mechanisms of resistance to immunotherapy include ineffective presentation of leukemia-specific or -associated antigens and inadequate effector cell activation. Previous efforts to improve antigen presentation of AML cells by in vitro differentiation into AML-derived dendritic cells were unsuccessful. However, transduction of AML to co-express the missing CD80 co-stimulator, and IL2 or GM-CSF, efficiently stimulates specific anti-leukemic immunity in murine AML and in human ex vivo assays. Despite approval in the UK, of a Phase-I trial using a CD80/IL2 expressing irradiated AML vaccines, concerns remain that IL2 expression may stimulate not only the activation of cytolytic T and NK cells, but also immune inhibitory T regulatory cells (Treg). Recently IL15, a γc chain cytokine that shares with IL2 the ability to stimulate NK and CD8+ memory T cells, has shown unique properties suited to stimulating anti-tumor immunity. IL15 is primarily trans-presented by cells expressing IL-15 receptor alpha (IL15Rα) to responding cells expressing IL15R subunits (IL2/15Rα and γ common (γc) sub-units). IL15 improves memory CD8+ T cell expansion, shows less Treg induction than IL2, and can protect immune effectors from Treg suppression. Unlike IL2, IL15 does not cause activation induced cell death in stimulated T cells. Co-expression of IL15 with membrane-bound IL15Rα (mIL-15Rα), as a secreted complex (IL15/sIL15Rα), or as an IL15/IL15Rα fusion protein, greatly increases IL15 stability and efficacy. We have generated and will test tricistronic lentiviral vectors expressing CD80 and either 1) IL15/mIL-15Rα, 2) IL-15/sIL-15Rα, or 3) IL-15/IL-15Rα fusion protein to determine which is optimal for clinical trial. Ex vivo studies in Aim 1 will provde a head-to-head comparison of the specificity as well as the nature, and magnitude of anti-leukemia responses induced by co-culture of human immune effectors with irradiated, autologous AML cells, transduced with the test human IL-15 lentiviral vectors vs a control vector expressing human CD80/IL-2. In vivo studies in Aim 2 will test the safety, toxicity, and efficacy of parallel vaccines transduced with murine IL-15 vectors vs the IL-2 control vector against murine 32Dp210 leukemia. The AML vaccine with best safety and survival outcomes will then be tested in aged mice that recapitulate features of immune senescence. Studies will directly inform the design of a Phase 1 clinical AML vaccine trial at UCSF for patients >60, ineligible for HSCT.