Abstract PL02-02: Investigating inhibition of the CD47 “don't eat me” signal to enable tumor phagocytic removal and augmented cross presentation to T cells

Abstract

Abstract Following embryonic development, most of our tissues and organs are continuously regenerated from tissue/organ specific stem cells. The principal property that distinguishes such stem cells from their daughter cells is self-renewal; when stem cells divide they give rise to stem cells (by self-renewal) and progenitors (by differentiation). In most tissues only the primitive stem cells self-renew. Stem cell isolation and transplantation is the basis for regenerative medicine. In the 1990s, we isolated mouse and human hematopoietic (blood-forming) stem cells (HSCs), and found that their method of isolation depleted all detectable cancer cells from mobilized peripheral blood (MPB) from patients with breast cancer. We used patients' MPB or cancer-free HSCs in autologous transplantations to rescue the hematopoietic system of women who underwent high dose combination chemotherapy for stage IV breast cancer. The results with MPB autologous transplantation were similar to those reported previously, with a 12-15-year overall survival rate of ∼7%. In contrast, for contemporaneous patients in a small phase I/II trial at Stanford who received autologous transplants with cancer-free HSCs, this survival rate was ∼33%. By chi-square analysis, MPB vs. cancer-free HSC transplantations were significantly different in terms of both progression-free and overall survival at all time points tested from 20 months to 12-15 years after transplantation. In considering stem cells and cancer, self-renewal is dangerous and therefore strictly regulated. Poorly regulated self-renewal can lead to the genesis of cancer stem cells, the only self-renewing cells in the cancerous tumor. We have followed the progression from HSCs to myelogenous leukemias and found that the developing cancer clones progress at the stage of HSCs, until they become fully malignant. At this point, the “leukemia” stem cell moves to a stage of a downstream oligolineage or multilineage progenitor that has evaded programmed cell death and programmed cell removal, while acquiring or keeping self-renewal. In the case of chronic myeloid leukemia, bcr-abl+ HSC clones outcompete normal HSCs in the chronic phase. The transition from the chronic phase to myeloid blast crisis results in the leukemia stem cells appearing in the granulocyte-macrophage progenitor (GMP) stage, and is accompanied by cell intrinsic activation of β-catenin, inhibitable by transfection with axin. In 4 of 7 patients studied, this resulted from stage-specific (GMP) mis-splicing of the glycogen synthase kinase-3β message, deleting the kinase domain. With the Majeti and Quake labs, we have shown that the mutations/indels (insertions or deletions) in 5 FLT3-ITD (internal tandem duplication) acute myeloid leukemias (AMLs) occur successively in HSC clones, including the penultimate FLT3-ITD itself. Of interest, many of the founder mutations are in epigenetic modifier genes such as TET2. While there are many ways to defeat programmed cell death and senescence, there appears to be one dominant method to avoid programmed cell removal_the expression of the cell surface “don't eat me” protein CD47, the ligand for macrophage SIRPα. All cancers we tested express CD47 to overcome expression of “eat me” signals such as calreticulin. In leukemias and lymphomas, calreticulin is the “eat me” signal countermanded by the CD47 “don't eat me” signal. In myelodysplastic syndrome (MDS), we found that MDS-initiating HSC clones outcompete normal HSC clones in the patient and in mice that receive xenotransplants of patient HSC samples. However, the oligolineage progenitors downstream of HSCs (e.g., GMPs) express calreticulin but not CD47, leading to our hypothesis that the blood-cell deficiencies in MDS occur because the progenitors are phagocytosed in vivo. When CD47 is upregulated in the MDS clone, MDS progresses to refractive anemia with excess blasts or AML. Other labs have also found mutations in the calreticulin gene in myeloproliferative neoplasms and MDS. Antibodies that block the CD47-SIRPα interaction enable phagocytosis and killing of the tumor cells in vitro and in vivo. All tested human solid tumors and lymphomas/leukemias/myelomas express CD47 and are susceptible to phagocytosis in the presence of anti-CD47 blocking antibodies, including a humanized antibody of the IgG4 isotype. We expect to have the anti-CD47 therapies in phase I trials in 2014. Citation Format: Irving L. Weissman. Investigating inhibition of the CD47 “don't eat me” signal to enable tumor phagocytic removal and augmented cross presentation to T cells. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr PL02-02. doi:10.1158/1538-7445.AM2014-PL02-02