A New Class of Antigen T-Cells That Redirect Bystander T-Cells to CD19 Positive Malignancies

Background: Immunotherapy with T cells expressing chimeric antigen receptors (CARs) has shown promise for the immunotherapy of CD19+ malignancies in early clinical studies. However, clinical efficacy depends on significant in vivo expansion of adoptively transferred T cells, which can be difficult to achieve. Genetically modifying T cells with bispecific T-cell engagers, which are able to recruit other T cells locally, amplifying antitumor effects, could potentially overcome this problem. Consistent and prolonged synthesis of engagers by T cells should also be superior to the intermittent direct infusion of the protein, both because these molecules have short half-lives and do not accumulate at tumor sites. The goal of this project was to generate T cells secreting CD19-specific T-cell engagers (CD19-ENG T cells) and to evaluate their effector function in vitro and in vivo.

Methods: A CD19-specific T-cell engager gene, consisting of two single chain variable fragments specific for CD3 and CD19, was synthesized and subcloned into a SFG retroviral vector in front of an IRES and mOrange. CD19-ENG T cells were generated by retroviral transduction and we determined their effector function in coculture and cytotoxicity assays, and in the Ph+ ALL BV173/xenograft model.

Results: Post transduction 50-60% of T cells were positive for transgene expression. In coculture assay CD19-ENG T cells recognized CD19+ lymphoma (Daudi, Raji) and acute leukemia (BV173) cells as judged by IFN-γ and IL-2 secretion in contrast to CD19- K562 cells. None of the targets were recognized by non-transduced (NT) T cells or T cells secreting engagers specific for an irrelevant antigen (EphA2-ENG T cells). Antigen-dependent recognition was confirmed in standard cytotoxicity assays. In transwell assays containing inserts that do not allow T-cell migration, only CD19-ENG T cells redirected NT T cells in the bottom well to CD19-positive tumor cells, demonstrating the ability of a diffusible product from CD19-ENG T cells to redirect NT T cells to CD19-positive tumor cells. To assess the anti-tumor activity of CD19-ENG T cells in vivo we used BV173 cells that were genetically modified with fire fly luciferase (ffLuc; ffLuc-BV173) to allow for serial bioluminescence imaging. NSG mice were injected iv with ffLuc-BV173 cells, and received an iv dose of CD19-ENG or EphA2-ENG T cells and an ip dose of IL2 on days 7, 14, and 21 post leukemia cell injection. Untreated mice served as controls. CD19-ENG T cells had potent anti-leukemia activity in contrast to EphA2-ENG T cells resulting in a significant survival advantage of treated animals.

Conclusions: We have generated CD19-ENG T cells with the unique ability to direct bystander T cells to CD19+ malignancies. CD19-ENG T cells had potent anti-leukemia activity, and may present a promising alternative to current CD19-targeted immunotherapy approaches.