Abstract 3599


Despite the development of many new agents for the treatment of acute myelogenous leukemia (AML), the prognosis for most patients remains dismal. The interleukin-3 receptor α chain (IL-3Rα, CD123) is highly expressed on the surface of acute and chronic leukemias compared with normal hematopoietic stem cells, providing a rational therapeutic target. CSL362 is a fully humanized, second generation, neutralizing monoclonal antibody targeting CD123 that contains a modified Fc region to enhance human natural killer (huNK) cell-mediated antibody dependent cellular cytotoxicity (ADCC). The purpose of this study was to test the in vivo efficacy of CSL362 in combination with an induction-type chemotherapy regimen and adoptive transfer of huNK cells against xenograft models of human AML.


Direct patient explants of 5 AML samples with defined genetic lesions were established as continuous xenografts in sub-lethally irradiated immune-deficient NOD/SCID or NSG mice. In order to complement the clinical development of CSL362 we optimized an induction-type regimen of cytarabine (AraC, 25 mg/kg IP daily × 4 × 2 weeks) combined with daunorubicin (DNR, 0.62 mg/kg IV weekly × 2 weeks) to simulate the complex biology of AML in a post-chemotherapy setting. For adoptive transplantation into AML engrafted mice, primary huNK cells were expanded 600- to 700-fold in vitro by culturing on feeder cells in the presence of IL-2 for 21 days. In efficacy experiments, an event was defined a priori to occur when the % human CD45 cells in the peripheral blood (%huCD45) reached 25%.


Human AML xenografts consistently and predictably produced progressive and disseminating disease in the peripheral blood, bone marrow, spleen, liver, kidneys, lungs and brain of mice between 42 and 200 days post-transplantation. Disease progression represented by increasing percentages of human CD45+ cells in the peripheral blood (%huCD45) inversely correlated with murine red blood cell and platelet counts, and mouse hematocrit and hemoglobin levels. The immunophenotype and gene expression profiles of each xenograft were comparable with the respective original patient samples, indicating that the xenografts provide a valid representation of the human disease. In vitro co-cultures of AML xenograft cells with CD56+/CD16+ huNK cells (5 h at a 1:10 ratio) resulted in an average AML cell death of 18% (range 7–38%). The addition of CSL362 (10 μg/ml) resulted in an average 26% (range 2–64%) increase in AML cell death compared with huNK alone. In vivo AraC/DNR treatment significantly extended the event-free survival (EFS) of mice engrafted with 3/3 AML xenografts by between 32 and 41 days compared with vehicle-treated control mice (P < 0.0001). The addition of CSL362 (300μg IP thrice weekly × 2) significantly extended the median EFS of AraC/DNR-treated mice by 5.9 days (P = 0.002). Moreover, while the adoptive transfer of huNK cells (20 million IV × 2) resulted in no significant survival advantage, huNK cells administered with CSL362 further prolonged mouse EFS by 15.2 days compared with chemotherapy alone (P = 0.0004), providing evidence of therapeutic enhancement.


AML xenografts established in NOD/SCID or NSG mice from direct patient explants provide a clinically relevant experimental model for preclinical drug testing, and reflect the immunophenotype and gene expression profiles of the original biopsy specimen. CSL362 treatment of AML xenografted mice augments the efficacy of an induction-type therapy, with additional therapeutic enhancement achieved by the adoptive transfer of huNK cells. These data support the rationale for CSL362 treatment of AML patients post remission induction in an upcoming clinical trial.


Lee:CSL Limited: Research Funding. Yee:CSL Limited: Research Funding. Busfield:CSL Limited: Employment. Vairo:CSL Limited: Employment. Lock:CSL Limited: Consultancy, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.