Adoptive T Cell Therapy Using Educated T Cells Generated by the Sequential Stimulation with DC/Tumor Fusion Cells and Anti-CD3CD28

Hematological malignancies demonstrate susceptibility to T cell mediated killing as evidenced by the potent graft versus disease effect mediated by alloreactive lymphocytes following allogeneic transplantation. A major area of investigation concerns the development of adoptive immunotherapy that would more selectively target and eliminate tumor cells. This strategy is dependent on the expansion of tumor reactive effector cells while minimizing the presence of regulatory T cells and other contributing elements to tumor mediate immunosuppression. Prior studies have demonstrated that ligation of the CD3/CD28 complex results in expansion of T cells ex vivo with phenotypic characteristics that are dependent on the immunologic milieu at the time of stimulation. We have developed a promising cancer vaccine in which autologous tumor cells are fused with dendritic cells (DCs) resulting in the presentation of a wide spectrum of tumor antigens in the context of DC mediated costimulation. We postulated that stimulation with DC/tumor fusions followed by anti-CD3/CD28 would result in the expansion of activated T cells targeting tumor antigens. Tumor cells were obtained from peripheral blood of patients with AML, or bone marrow aspirates of patients with AML and multiple myeloma. DCs were generated from adherent mononuclear cells cultured with rhIL-4, GM-CSF and TNF and fused with tumor cells by coculture in 50% solution of polyethylene glycol. T cells were stimulated by DC/tumor fusions prior to or following exposure to antiCD3/CD28 antibody coated plates for 48 hours. Stimulation by fusions followed by anti-CD3/CD28 resulted in the synergistic expansion of T cells as manifested by the stimulation index (SI) in excess of that seen when cells were stimulated by either pathway alone or first stimulated by anti-CD3/CD28 then fusions. DC/AML fusions induced autologous T cell proliferation with an SI of 3.3 with memory effector cells (CD45RO+) comprising 10% of the total T cell population. In contrast, sequential stimulation with DC/AML fusions followed by anti-CD3/CD28 resulted in a rise in T cell proliferation (SI 8.2) of which 39% of the resultant populations expressed CD45RO. Similarly a rise in CD4+/CD25+ cells was observed following sequential stimulation with DC/AML fusions followed by anti-CD3/CD28 (9.3% vs. 2.7% following stimulation with DC/AML fusions alone). In addition, an increased percentage of CD4+/CD25+ cells expressed IFNγ when exposed to anti-CD3/CD28 following coculture with fusion cells (7% compared to 2% with fusions alone). A significant rise in the percent of Foxp3+ cells was not seen. Expression of granzyme B is up regulated in activated cytolytic CD8+ T cells that confer perforinmediated killing of target cells. As compared to un-stimulated T cells, stimulation with DC/tumor fusions or anti-CD3/CD28 alone resulted in a 3.3 and 3.8 fold increase in CD8+ T cells expressing granzyme B, respectively. In contrast, sequential stimulation with DC/tumor fusions and anti-CD3/CD28 induced a 19-fold expansion of granzyme B+ cells consistent with their enhanced cytolytic capacity. In addition, sequential stimulation with DC/tumor fusions and anti-CD3/CD28 results in heightened capacity to lyse autologous tumor targets as compared to T cells stimulated by fusions alone in a cytotoxicity assay. In spectratyping analysis, T cells undergoing sequential stimulation with DC/tumor fusion cells and anti-CD3CD28 demonstrate greater skewing of CDR3-size usage in the T cell receptor as compared to T cells stimulated by fusions or anti-CD3/CD28 alone consistent with the expansion of a defined clonal population. The pattern of gene expression in T cells stimulated sequentially by DC/AML fusions and anti-CD3CD28 is being assessed by gene arrays to further define the unique nature of this population. In conclusion, we have demonstrated that stimulation of T cells by DC/tumor fusions followed by exposure to anti-CD3/CD28 antibodies results in the expansion of tumor reactive activated T cell populations. A clinical trial evaluating the safety and efficacy of adoptive immunotherapy using T cells generated by sequential stimulation with DC/tumor fusions and anti- CD3CD28 for patients with multiple myeloma following autologous transplantation is planned.