AUTHORS
Parveen Shiraz, MD, Carlos Jose Rodriguez Santiago, MS2, Sonia Fonseca, BS3, Kathy Phuong Le, MS4, Jessica Poyser, BS5, Hao Yan, PhD6, Judith Anne Shizuru, MD, PhD7
AFFILIATIONS
1 Stanford University, Stanford, California, United States
2 Stanford University, Stanford, California, United States
3 Stanford University, Stanford, California, United States
4 Stanford University, Stanford, California, United States
5 Stanford University, Stanford, California, United States
6 Stanford University, Stanford, California, United States
7 Stanford University, Stanford, California, United States
ABSTRACT
Introduction
Chemotherapy refractory myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML) have high morbidity and mortality. Therefore, innovative immunotherapeutic strategies are needed to improve outcomes. CD117 and CD33 are the two most highly expressed antigens on myeloid leukemia stem cells (LSC). Yet, to date, there is no single therapeutic modality that targets both surface markers simultaneously. We have developed a novel tri-specific antibody (Ab) construct (CM1/3/16) that targets CD117 and CD33 on myeloid LSC. We designed CM1/3/16 to engage and activate monocytes, macrophages and natural killer (NK) cells via an active IgG1 Fc and additional CD16 binding scFv domains linked to the C-terminus of the Fc.
Methods
CM1/3/16 was evaluated in vitro and in vivo using CD117 and/or CD33 expressing AML cell lines or primary patient samples and compared to the monospecific CD117 (CM1/1), CD33 (Gemtuzumab-Gem), and bispecific CD117/CD33 (CM1/3) binding Abs. Binding studies were performed on CD117high(h)/CD33h AML cell line Kasumi-1, CD117low(l)/CD33h AML cell line MOLM-13 with flow cytometry and to purified CD16a protein with ELISA. In vitro cytotoxicity assays were performed on Kasumi-1, MOLM-13, and CD117l/CD33h primary AML cells, using healthy donor NK cells as effector cells. In vivo efficacy of CM1/3/16 was evaluated in immune deficient NBSGW mice xenograft models injected with luciferase expressing (Luc) Kasumi-1 or primary patient MDS/AML cells. Mice were treated with CM1/3/16 at 0.5 or 1mg with or without 2x10e6 NK cells that were fresh or cytokine (IL-2, IL-15 and IL-21) activated. Luc Kasumi-1 models were assessed with weekly bioluminescence imaging (BLI) and primary patient MDS/AML models were assessed with weekly blood and a terminal 4-week blood, bone marrow and spleen human myeloid leukemia chimerism (MLC).
Results
At 1.23nM dilution, both CM1/3/16 and CM1/3 demonstrated high-affinity, dose-dependent and saturable binding to Kasumi-1 and MOLM-13 cells, with 1.49- and 3.35-fold higher maximum binding capacity (Bmax) compared to CM1/1 and Gem respectively. Tri-specific CM1/3/16 demonstrated superior binding to CD16a antigen (KD6.62nM) with 5.82-fold higher affinity than bi-specific CM1/3 (KD38.54nM). In vitro cytotoxicity profiles of these constructs against Kasumi-1 and MOLM-13 were reflective of their binding affinities with CM1/3/16 performing better than CM1/3 and the mono-specific Abs CM1/1 and Gem. CM1/3/16 showed >70% killing of Kasumi-1 cells at 0.001nM and >40% killing of MOLM-13 cells and primary patient derived AML cells at 1nM dilution. In xenograft models of primary patient derived MDS/AML, CM1/3/16 depleted CD117+ and/or CD33+ LSC and maintained survival at 4 weeks post-treatment, while control mice treated with fresh or cytokine activated NK cells demonstrated persistence of malignant disease. In NBSGW mice models of Luc Kasumi-1, CM1/3/16 treated mice remained disease free at 4 weeks post treatment evaluated by BLI and blood, bone marrow and spleen human MLC. Untreated and NK cell treated mice showed leukemia progression by BLI at 2 weeks at which time we performed a cross over study and treated them with CM1/3/16 at 0.5mg and 1mg doses and demonstrated decrease in leukemia burden at one week post-treatment. CM1/3/16 was therefore able to decrease leukemia burden in mice with advanced leukemia within one week.
Conclusions
CM1/3/16, a tri-specific Ab, designed to target CD117 and CD33 on myeloid LSC and engage multiple Fc receptors to activate monocytes, macrophages and NK cells demonstrated significantly higher myeloid leukemia cell binding and in vitro cytotoxicity compared to the respective mono- and bi-specific antibodies. Treatment with CM1/3/16 as a single agent led to deep depletion of myeloid leukemia in vivo outperforming cytokine activated NK cells. Clinical evaluation of CM1/3/16 to debulk disease burden as a bridge to hematopoietic stem cell transplant in chemotherapy refractory MDS/AML is warranted.
DISCLOSURES
Parveen Shiraz, MD: No relevant conflicts of interest to declare.