Preclinical Evaluation of CSL362/JNJ-56022473 in Combination with Decitabine or Azacitidine in in Vitro Assays

Acute myeloid leukemia is a heterogeneous clonal disorder characterized by abnormal differentiation and unregulated proliferation of hematopoietic progenitor cells (blast), leading to defects in hematopoiesis. Accumulating evidence has implicated leukemic stem cells (LSCs) as the main cause for perpetuation of the disease based on their profile to rapidly give rise to proliferating blasts, resistance to chemotherapeutics, and ability to self-renew. CD123 (IL-3α) has been demonstrated to be expressed at high levels on LSCs and AML blasts compared to normal hematopoietic progenitor cells (Jordon et al Leukemia 2000), providing a potential strategy to target and eliminate AML cells.

CSL362, a humanized monoclonal antibody that specifically targets CD123, was engineered to have increased affinity for CD16 on natural killer (NK) cells, leading to an enhanced ability to activate NK cell antibody-dependent cellular cytotoxicity (ADCC) against CD123-positive cells (Busfield et al Leukemia 2014). JNJ-56022473 is a CSL362 variant generated from a new line and has highly similar biochemical, biophysical, and biological properties to CSL362.

NK cells are an integral part of the innate immune response to AML cells, and their activity is determined by the balance of activating and inhibitory receptor stimulation (Lanier Ann Rev Immunol 2005). Decitabine and azacitidine, DNA-hypomethylating agents, have been shown to differentially modulate the expression of activating and inhibitory receptors on NK cells at physiologically used doses (Kopp et al Mol Immunol 2013, Schonefeldt et al Blood Cancer J 2013). Decitabine and azacitidine can also activate key epigenetically silenced pathways in AML cells, leading to an arrest of AML cell proliferation. We evaluated the effects of decitabine and azacitidine on CD123 expression and CSL362-mediated cytotoxicity in the TF-1 AML cell line and on NK cell activation in human peripheral blood mononuclear cells (PBMCs).

To assess the effect of decitabine and azacitidine on CD123 expression and NK cell activation, 200nM of each compound was incubated with cells under various conditions. CD123 expression in treated TF-1 cells was quantitated by fluorescence-activated cell sorting, where a significant increase in CD123 expression was detected with the 3 and 5 day treatments for decitabine (p<0.001) and with the 5 day treatment for azacitidine (p<0.05). Decitabine and azacitidine treatment increased expression of the NK cell activating receptors NKG2D+ (p<0.001[decitabine] and p<0.005 [azacitidine]) and NKp46+ (p<0.001 [decitabine] and p=0.6 [azacitidine]) on PBMCs with the 5 day treatment schedule but did not affect the expression of the NK cell inhibitory receptor, KIR. Treatment with azacitidine (200nM and 1µM) reduced NK cell viability (p<0.001) compared to decitabine and also significantly downregulated IFN-gamma production (p<0.001).

The ability of decitabine and azacitidine to increase expression of CD123 and NK cell activating receptors suggests that the compounds may potentiate the ADCC profile of CSL362/JNJ-56022473. TF-1 cells and PBMCs were pretreated with 200nM of each compound under different dosing regimens followed by ADCC assay with CSL362/JNJ-56022473. The percentage of specific lysis was significantly enhanced in the 5 day treatment with decitabine (66.45% lysis compared to control [32.80%]; p<0.005). In contrast, azacitidine reduced JNJ-56022473-mediated cell lysis (22.71% compared to control [30.96%]; p<0.005) with the 5 day treatment. In summary, decitabine and azacitidine pretreatment stimulated the expression of CD123 in TF-1 cells and NK cell activating receptors, NKG2D+ and NKp46+.  The combination of decitabine and CSL362 increased CSL362-induced ADCC activity in TF-1 cells. In contrast, azacitidine inhibited CSL362-mediated ADCC activity, downregulated IFN-gamma production, and reduced NK cell viability.