FLT3 mutated AML confers poor prognosis despite treatment with standard chemotherapy, attributable to bone marrow (BM) induced tumor cell resistance in the microenvironment via molecules such as CXCR4 and E-selectin. We recently reported that simultaneous targeting of CXCR4 and E-selectin with GMI-1359, a novel small molecule dual antagonist, mobilized leukemic cells into the circulation; further, combination with AraC and DNR resulted in prolonged survival in a FLT3-ITD AML murine xenograft model. Here we report a novel profile of leukemic cell mobilization induced by GMI-1359, its impact on survival when combined with chemotherapy, and toxicity profile supporting entry to the clinic.
A comparison of tumor mobilization by GMI-1359 to E-selectin or CXCR4 antagonists was determined in NCR mice with established BM infiltrative MV4-11 FLT3 ITD human AML tumor (modified with mCherry and luciferase). Mice (n=18/group) were injected with GMI-1359 (40 mg/kg), a potent E-selectin antagonist (40 mg/kg), or plerixafor (5 mg/kg) and blood samples were assessed by flow cytometry at multiple times for human AML cells and murine WBC. Appearance of human AML cells in peripheral blood increased ~16-fold compared to untreated mice 8 h post treatment with GMI-1359, and remained elevated for the duration of the study (72 h). Tumor cell mobilization following treatment with the E-selectin antagonist was gradual, reaching a 9.7-fold increase 48 h post injection and trending down at 72 h. Administration of plerixafor induced a rapid 11-fold mobilization of tumor cells by 1 h returning to baseline at 24 h. Murine WBC count did not change in any group at the time points assessed. These data establish that the profile of GMI-1359 induced tumor cell mobilization is distinct from both plerixafor (rapid mobilization of short duration) and an E-selectin antagonist (gradual mobilization of long duration). As E-selectin but not CXCR4 has been reported to be critical for tumor cell entry into the BM, the sustained presence of tumor cells in circulation after GMI-1359 treatment could be the result of blocking not only CXCR4 but also E-selectin, thereby inhibiting their re-entry into the BM.
The therapeutic impact of tumor mobilization by GMI-1359 was evaluated for therapeutic consequence in combination with AraC and DNR in MV4.11 tumor bearing mice. Mice (n=10/group) were treated with saline; GMI-1359 (40 mg/kg) alone; chemotherapy alone; or GMI-1359 and AraC/DNR. GMI-1359 was initiated at onset of chemotherapy and given for 1, 3 or 14 days. Treatment of mice with GMI-1359 alone or together with AraC/DNR was well-tolerated. Median survival times (MST) of mice treated with saline, GMI-1359 or AraC/DNR was 47, 50 and 50 days, respectively; all mice succumbed to progressive disease by study conclusion (Day 71). In contrast, the MST of tumor bearing mice treated with GMI-1359 qdx3 or qdx14 and AraC/DNR was 63.5 days (p<0.001 vs. AraC/DNR) with a 40% survival rate. The MST of mice given a single dose of GMI-1359 at the start of AraC/DNR was statistically indistinguishable from chemotherapy alone (49.5 and 50 days, respectively) even though a 20% survival advantage was noted for the GMI-1359 combination group.
GMI-1359 was tested for off-target interactions against 87 potential targets, with no significant activity observed. Toxicology studies showed a safety profile consistent with that of plerixafor. The pharmacokinetic (PK) profile was characterized in mice, rats, dogs, and monkeys, and was linear with t1/2 <2h in monkey and no accumulation in plasma. GMI-1359 is not mutagenic or clastogenic. Based on microsomal and liver S9 stability testing, GMI-1359 is not expected to be highly metabolized in any species tested. GMI-1359 did not inhibit P-gp, is moderately bound to human plasma proteins, and is stable in human plasma at 37C.
In summary, GMI-1359, an innovative dual E-selectin and CXCR4 antagonist, more rapidly mobilized tumor cells and maintained them in circulation for a longer time when compared to either an antagonist of E-selectin or CXCR4 alone in a murine model of FLT3 ITD AML. This novel kinetic signature of mobilization by dual inhibition of two important adhesion molecules and accompanying improvement in survival over standard chemotherapy support the potential for combination with current treatment to improve outcomes for patients with FLT3-mutated AML. First-in-human trials are being initiated.
Fogler:GlycoMimetics, Inc.: Employment. Flanner:GlycoMimetics, Inc.: Employment. Wolfgang:GlycoMimetics, Inc.: Employment. Smith:GlycoMimetics, Inc.: Employment. Thackray:GlycoMimetics: Employment, Equity Ownership. Magnani:GlycoMimetics: Employment, Equity Ownership, Membership on an entity's Board of Directors or advisory committees.
Asterisk with author names denotes non-ASH members.