41BBL-Based Activation and Expansion of Autologous Natural Killer Cells Results in Enhanced Activity Against Leukemia Including ALL

Abstract

Childhood leukemia is the most common pediatric malignancy. There are now excellent cure rates for these patients, however outcomes remain poor for those with refractory disease and for those who relapse after standard salvage therapies, with a disease recurrence of approximately 50%. Therefore, development of novel cellular therapies is essential to treat these refractory patients.Natural Killer (NK) cells generated from an allograft contribute to improved disease free survival after Hematopoietic Stem Cell Transplantation for leukemia when there is a KIR mismatch. This effect appears to be particularly potent in the setting of Acute Myelogenous Leukemia (AML) with less benefit demonstrated in Acute Lymphoblastic Leukemia (ALL). Preclinical studies have also suggested that activation and expansion of resting NK cells can enhance NK cell cytotoxicity and eliminate the need for KIR mismatch due to up-regulation of activating receptors. We are currently testing this approach in the clinic following a fully matched allogeneic transplant platform for leukemia. Our aim is to explore whether 41BB ligand (41BBL) and recombinant IL-15 (rIL-15) mediated ex vivo expansion of autologous NK cells results in enhanced activity against AML and ALL. The activation/expansion process may allow for the use of autologous NK cell infusions, thus eliminating the need for allogeneic NK cell donors.To test this hypothesis, we ex vivo expanded and activated NK cells derived from C57BL/6J (B6) mice using artificial Antigen Presenting Cells (aAPCs) containing 41BBL and rIL-15 for 7-14 days. NK cells were co-cultured with murine AML cells (C1498) and murine ALL cells (E2A-PBX) – both on B6 background. Controls included YAC cells (murine T-cell lymphoma cell line sensitive to NK cell killing) as well as Phorbol Myristate Acetate (PMA)/ionomycin. All cells were co-cultured for 5 hours prior to functional assessment of NK cells via CD107a degranulation.NK cells cultured with 41BBL aAPCs and rIL-15 had a 30-fold expansion in numbers (Figure 1) and an increase in purity to approximately 95-98% (NK1.1+, CD3–) by Day 7. In the absence of cytokine or aAPCs, cultured NK cells underwent rapid apoptosis. Functionally, although resting NK cells (harvested prior to assessment) expressed CD107a when cultured with YAC cells and PMA, only minimal degranulation was observed in the presence of autologous AML cells or ALL cells. In contrast, activated and expanded autologous NK cells displayed enhanced activity against ALL, AML, as well as YAC cells, while only minimal levels of CD107a were seen in the absence of targets (Figure 2). In vivo experiments with a single injection of activated and expanded NK cells did not result in prolonged survival of mice bearing either AML or ALL. Assessment of adoptively transferred NK cells demonstrated very transient persistence (<2 days) with no in vivo expansion, suggesting that repeated injections may be necessary for leukemia eradication. Future murine experiments will investigate the effect repeated injections of activated/expanded NK cells and/or the administration of rIL-15 will have on survival and leukemia eradication. In addition, the ability to activate and expand NK cells in culture provides an opportunity for lentiviral-based transduction with chimeric antigen receptor (CAR) vectors. We are currently testing this with a murine CD19 CAR.These experiments suggest that autologous activated and expanded NK cells may serve as a viable cellular therapy for pediatric patients with refractory/relapsed leukemia. As demonstrated in these in vitro experiments, autologous activated/expanded NK cells still show increased targeting of mouse AML and ALL cell lines despite the lack of KIR mismatch. Thus, they may serve as a potential platform for leukemia therapy, including ALL, which appear to be poor targets for resting NK cells. In addition, these cells demonstrate transient persistence in vivo, a potential advantage in the context of redirected cytotoxicity using CAR constructs that target antigens with broader expression in the hematopoietic compartment. [Display omitted] [Display omitted] DisclosuresNo relevant conflicts of interest to declare.