The outcomes of patients with AML remain disappointing, with only fewer than half of all adults (and <10% of adults 60 years or older) achieving long-term remissions. Thus, novel therapies are urgently needed. Chimeric antigen receptors (CARs) have been used extensively to redirect the specificity of T cells in patients with lymphoid malignancies. These infusions have been primarily restricted to the autologous setting, but generation of autologous products for each patient is logistically cumbersome and not scalable. A banked allogeneic product could overcome these limitations; but, allogeneic T-cells (even if HLA-matched) carry a major risk of graft-versus-host disease (GVHD). Natural killer (NK) cells provide an attractive alternative to T-cells for CAR therapy as they do not cause GVHD. Cord blood (CB) is a readily available "off-the-shelf" source of allogeneic NK cells that can be expanded to large, highly functional doses for CAR therapy.

CD123 (IL3-receptor α chain) is a potential immunotherapeutic target in AML due to its overexpression on leukemic stem cells and AML blasts, rather than normal hematopoietic stem cells. Thus, we reprogrammed CB-NK cells to target CD123+ve tumor cells by retroviral transduction using synthetic constructs producing either murine-derived (iC9-2A-CD123CAR-CD28-CD3z-2A-IL15) or humanized (iC9-2A-hCD123CAR-CD28-CD3z-2A-IL15) scFv domains. We humanized the ScFv derived from murine antibody clone 26292 to avoid human anti-mouse antibody responses, which could potentially limit the persistence and efficacy of the CAR-NK cells when infused in patients. These constructs also express exogenous soluble IL-15 to support NK cellular survival and proliferation, as well as an inducible caspase-9 (iC9) suicide gene as a safety precaution to ablate CAR+ve NK cells if needed.

Retroviral transduction with murine (CD123CAR) and humanized (hCD123CAR) (n=12) resulted in a median CAR positivity of 75% and 76%, respectively, 14 days post transduction and viability of >90%. Notably, CAR expression was stable over time. iC9/CAR.123/IL15+ and iC9/hCAR.123/IL15+ NK cells expanded for 2 weeks showed no signs of exhaustion, such as downregulation of eomesodermin and T-bet, or upregulation of KLRG1, and exhibited a mature phenotype, similar to ex vivo expanded non-transduced NK cells (NT-NK) using 22 markers. In a 4 hour 51Cr release assay, murine CD123CAR+ve and hCD123CAR+ve NK cells were equally effective at killing CD123+ve AML cell lines (THP-1 and MOLM-14), and exhibited significantly higher cytotoxicity when compared to NT-NK cells at all tested effector to target cell ratios (20:1; 10:1; 5:1 and 1:1). Notably, CAR123 and hCAR123-transduced NK cells were as equally efficient as NT-NK cells in killing CD123 negative targets such as K562 (positive control) and KARPAS (negative control), indicating that the enhanced killing of CD123+ targets by the transduced cells is mediated by the CAR receptor and not due to a non-specific enhancement in NK cytotoxicity. In addition, CD123CAR+ve and hCD123CAR+ve NK cells showed significantly higher production of IFN-γ, TNF-α, and CD107a degranulation in response to MOLM14 and THP in comparison to NT-NK cells. Indeed, using confocal microscopy, we showed significantly higher accumulation of CAR molecules at the immunologic synapse (IS) formed between CD123CAR+ve and hCD123CAR+ve CB-NK cells and CD123+ve targets compared to the diffuse presence of CARs at the IS with K562 target cells, indicating that CAR molecules participate in IS formation in a CD123-dependent manner.

We validated elimination of iC9/CAR.123/IL-15 transduced NK via activation of iC9 in vitro. The addition of 10nM of the small molecule dimerizer AP1903 (Bellicum Pharmaceuticals, Inc) to cultures of iC9/CAR123/IL-15 or iC9/hCAR123/IL-15-transduced NK cells induced apoptosis of transduced NK cells within 4 hours but had no effect on the viability of NT-NK cells. Studies to test the in vivo anti-tumor efficacy of iC9/CAR123/IL-15 or iC9/hCAR123/IL-15-transduced NK cells in an NSG mouse model of MOLM14 are under way.

In summary, our data demonstrate the capacity of CD123CAR+ve and hCD123CAR+ve CB-NK cells to kill CD123+ targets and support the use of CD123-specific CAR NK cells in the clinic for patients with relapsed AML as a means to decrease disease burden, possibly prior to consolidative therapies such as allogeneic transplantation.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution