Dual E-Selectin/CXCR4 Antagonist GMI-1359 Exerts Efficient Anti-Leukemia Effects in a FLT3 ITD Mutated Acute Myeloid Leukemia Patient-Derived Xenograft Murine Model

Background: Acute myeloid leukemia (AML) is a clinically and molecularly heterogeneous disease with poor outcome. Targeted therapy of FMS-like tyrosine kinase-3 (FLT3)-mutated AML patients using small molecular inhibitors including sorafenib showed clinical success in reducing leukemia blasts in peripheral blood, but has limited effect on leukemic stem cells in the bone marrow (BM) microenvironment (1, 2). The BM is presumed to be the reservoir for leukemia stem cells (LSCs) which persist under targeted therapy and mediate disease relapse. The interaction of leukemic blasts with the BM microenvironment mediated by receptor-ligand axes such as CXCR4/SDF-1, E-selectin/HECA-452, and cell-cell contact have also been associated with drug resistance in FLT3 mutated AML (3-5). We recently reported that targeting CXCR4/E-selectin with the dual inhibitor GMI-1359 (GlycoMimetics, Inc., Rockville, MD) showed efficient mobilization of leukemia cells into the circulation and significant prolongation of survival of mice in FLT3-ITD mutant AML cells-xenografted murine model (Zhang et al., ASH Abstract #3790, 2015; Zhang et al., AACR Abstract #3284, 2016).

Methods: In the present study, we further evaluated anti-leukemia effects of the dual CXCR4/E-selectin inhibitor GMI-1359 in a patient-derived xenograft (PDX) murine model. Primary FLT3 ITD mutated AML cells were isolated from BM aspirates and injected into irradiated NSG mice. Human CD45+/CD33+ cells were isolated from spleens of secondary engrafted mice after 3-6 months. The PDX leukemic cells were injected into irradiated NSG mice via tail vein (2x10⁶ cells/mouse) and mice received either GMI-1359 (40mg/kg) or sorafenib (10mg/kg), or combination treatment starting at Day 18 after injection of leukemia cells. Vehicle treated mice served as controls. Leukemic cell burden was evaluated by flow cytometry (hCD45+ cells), histology and immunohistochemistry.

Results: GMI-1359 treatment alone efficiently mobilized leukemic cells into blood circulation (absolute hCD45+ cells: 23.24k/UL vs. 9.06k/UL, respectively, in GMI-1359 vs. vehicle treatment groups after 71 days of treatment). Sorafenib alone or combined with GMI-1359 enhanced apoptosis induction and markedly reduced hCD45+ cells in circulation, BM and liver. The combination treatment profoundly reduced spleens leukemia cell infiltration (immunohistochemistry of CD45). A novel and unexpected finding of this study was that the combination of GMI-1359 with sorafenib significantly increased the regeneration of normal bone marrow cell populations (erythropoiesis, myelopoiesis and megakaryopoiesis) in the PDX model analyzed by histomorphology. Morphology and frequency of the hematopoiesis were identified in normal bone marrow of non-tumor bearing mice, suggesting a partial treatment response to the combination of GMI-1359 and sorafenib. In addition, GMI-1359 alone, sorafenib alone, or the combination treatment profoundly extended survival of PDX mice compared to vehicle treatment (98, 98 and 98 days vs 60.5 days, respectively (p = 0.04)).

Conclusions: The dual CXCR4/E-selectin inhibitor GMI-1359 alone or in combination with the FLT3 ITD inhibitor sorafenib demonstrated anti-leukemia effects in a PDX model of FLT3 ITD mutated AML. Interestingly, combining GMI-1359 with sorafenib strikingly increased normal hematopoiesis in the bone marrow. The underlying mechanisms are under investigation.