SUMMARY
Acute myeloid leukemia (AML) is a clonal hematopoietic stem and progenitor cell malignancy characterized by
poor clinical outcomes. Malignant transformation triggers expression of surface proteins that serve as “danger
signals”, such as MICA and MICB (MICA/B) that are commonly expressed by leukemia cells in response to
cellular stress pathways. Natural killer (NK) cells and cytotoxic T lymphocytes recognize MICA/B with the NKG2D
receptor, which in turn induces cytotoxic functions against leukemia cells. However, they often escape
recognition by shedding MICA/B via an intriguing post-translational modification called proteolytic cleavage. In
previous and revolutionary study, we developed a series of MICA/B-targeted monoclonal antibodies that inhibit
the shedding of MICA/B by blocking the extracellular domain that undergoes an unfolding by disulfide isomerase,
which enables subsequent cleavage by metalloproteases. One of them was further characterized, and it
promotes NK cell-mediated immunity against solid tumors by dual signaling of NKG2D and Fc receptors.
Recently, we discovered that the antibody also promotes macrophage-driven immunity in AML models, by
enabling antibody-dependent phagocytosis of leukemia cells in the blood and bone marrow. In this setting,
MICA/B serve as “leukemia antigens” for Fc receptor-driven immunity, although contributions by NKG2D, NK
cells, and T cells were also detected. Now we developed a new version of this antibody, with three point
mutations in the Fc domain to increase the binding affinity to all three Fc activating receptors while maintaining
low affinity to the Fc inhibitory receptor. This new version is humanized and induces potent NK cell effector
functions against human AML cells, compared to the humanized wild type version. We also established mouse
models of human Fc receptor biology and both murine and human AML models, to establish the anti-leukemia
activity of our now enhanced molecule in vivo. Furthermore, we propose a mechanism-driven drug combination
regimen, whereby a histone deacetylase inhibitor (romidepsin) induces human leukemia cells to express MICA/B
mRNAs that are then translated to MICA/B proteins, which are followed by stabilization on cellular surface by
our antibody that inhibits the shedding. We propose that immunotherapy for AML can be achieved via this drug
combination, which increases the innate immunogenicity of leukemia cells. For these reasons, our studies will
generate important information about how to induce protective immunity against AML. Antibody-mediated
inhibition of MICA/B shedding is a new immunotherapeutic opportunity pioneered by us and now validated
independently by multiple pharmaceutical companies, one of which has recently began phase-I clinical trial
testing an antibody analog to ours, but the wild type version, in patients with advanced solid tumors. Therefore,
our studies will generate compelling rationale to test MICA/B-targeted antibodies in AML trials, by pharmaceutical
companies, ourselves in academia, or both by working in collaboration.