CD47 Is An Independent Prognostic Factor and Therapeutic Antibody Target on Human Acute Myeloid Leukemia Stem Cells

A growing body of evidence has added to our fundamental knowledge by demonstrating that human acute myelogenous leukemia (AML) is organized as a cellular hierarchy initiated and maintained by rare self-renewing leukemia stem cells (LSC). One implication of this cancer stem cell model is that in order to eradicate the leukemia and cure the patient, therapies must target and eliminate the leukemia stem cells. For the development of such LSC-targeted therapies, it is necessary to identify molecules that are preferentially expressed in LSC compared to their normal counterparts and that are critical for their function. We report here the identification of increased expression of CD47 on human AML LSC compared to normal hematopoietic stem cells (HSC). Furthermore, we demonstrate that differential CD47 expression on Lin-CD34+CD38− cells can be utilized to prospectively separate normal (CD47lo) from leukemic (CD47hi) progenitors from the same patient sample. CD47 serves as the ligand for signal regulatory protein alpha (SIRP-alpha) on phagocytic cells, which in turn delivers an inhibitory signal for phagocytosis. We hypothesize that increased CD47 expression on human AML contributes to pathogenesis by inhibiting phagocytosis of leukemia cells through the interaction of CD47 with SIRP-alpha. One prediction from this hypothesis is that increased expression of CD47 will be associated with a worse clinical outcome. We analyzed microarray gene expression data and associated clinical outcomes for a cohort of 123 AML patients with normal cytogenetics. Patients were stratified into low CD47 and high CD47 expression groups, based on expression relative to the median. Increased CD47 expression was found to be associated with worse event-free and overall survival. Patients in the low CD47 expression group had a median event-free survival of 18.2 months compared to 9.9 months in the high CD47 expression group, corresponding to a hazard ratio of 1.61 (p=0.04). For overall survival, patients in the low CD47 expression group had a median of 24.3 months compared to 14.1 months in the high CD47 expression group, corresponding to a hazard ratio of 1.70 (p=0.02). In multivariable analysis considering age, FLT3-ITD status, and CD47 expression as a continuous variable, increased CD47 expression remained associated with worse event-free survival with a hazard ratio of 1.33 (p=0.03) and overall survival with a hazard ratio of 1.31 (p=0.05). A second prediction from our hypothesis is that disruption of the CD47-SIRP-alpha interaction with a monoclonal antibody directed against CD47 will enable the phagocytosis of AML LSC. We obtained mouse anti-human CD47 monoclonal antibodies that are able to disrupt the CD47-SIRP-alpha interaction and tested their ability to enable phagocytosis of AML LSC both in vitro and in vivo. Unlike isotype control or an anti-CD45 antibody, anti-CD47 antibodies enabled phagocytosis of AML LSC by both mouse and human macrophages in vitro. Furthermore, unlike isotype control or anti-CD45 antibody, coating of AML LSC with anti-CD47 antibody completely eliminated in vivo engraftment upon xenotransplantation into NOG mice. Finally, treatment of NOG mice already engrafted with human AML LSC with daily intraperitoneal injections of anti-CD47 antibody for two weeks, resulted in complete elimination of circulating leukemia and a statistically significant reduction in bone marrow leukemia compared to control IgG. In summary, increased CD47 expression is an independent poor prognostic factor that can be targeted on AML stem cells with a monoclonal antibody capable of stimulating phagocytosis and in vivo elimination of LSC.