Chimeric antigen receptors against CD33/CD123 antigens efficiently target primary acute myeloid leukemia cells in vivo.

As significant numbers of acute myeloid leukemia (AML) patients are still refractory to conventional therapies or experience relapse, immunotherapy using T cells expressing chimeric antigen receptors (CARs) might represent a valid treatment option. AML cells frequently overexpress the myeloid antigens CD33 and CD123, for which specific CARs can be generated. However, CD33 is also expressed on normal hematopoietic stem/progenitor cells (HSPCs), and its targeting could potentially impair normal hematopoiesis. In contrast, CD123 is widely expressed by AML, while low expression is detected on HSPCs, making it a much more attractive target. In this study we describe the in vivo efficacy and safety of using cytokine-induced killer (CIK) cells genetically modified to express anti-CD33 or anti-CD123 CAR to target AML. We show that both these modified T cells are very efficient in reducing leukemia burden in vivo, but only the anti-CD123 CAR has limited killing on normal HSPCs, thus making it a very attractive immunotherapeutic tool for AML treatment.

rhG-CSF does not affect the phenotype of adult donor peripheral blood NK cells.

Considerable evidence in preclinical models as well as in human transplantation now suggests that donor-derived natural killer (NK) cells can contribute to alloimmune recognition of recipient residual tumour cells. This makes the NK cell population an attractive target for in vitro or in vivo manipulations, in order to improve the antitumour effect of allogeneic transplantation. However, conditions in which allogeneic donor cells are collected vary; several reports have emphasised the different phenotypic and functional properties of T cells derived from marrow, cord blood or mobilised peripheral blood grafts; others have demonstrated different clinical outcomes following blood or marrow transplantation after myeloablative conditioning regimens. NK cells have been examined in this setting; the availability of new tools to study the expression of a variety of surface antigens that are involved in the control of NK cell activity offered us an opportunity to extensively characterise the phenotypic properties of NK cells from donors, before and after administration of pharmacological doses of rhG-CSF used for haematopoietic progenitor mobilisation. Our study suggests that rhG-CSF does not reproducibly alter blood NK cell phenotype in normal individuals, and thus that donor-derived cells are fully equipped to exert their potential antitumour effect.