High Efficacy of Liposomal Annamycin (L-ANN) in Combination with Cytarabine in Syngeneic p53-Null AML Mouse Model

Background: Acute myeloid leukemia (AML) is a heterologous hematological malignancy in which the p53-mutated subset is associated with the most guarded prognosis. Induction therapy for AML with cytarabine (cytosine arabinoside, ARA-C) is routinely used in combinations with anthracyclines.

Annamycin (ANN) is an antitumoral anthracycline whose anti-leukemia activity, in contrast to doxorubicin (DOX) and daunorubicin, is unaffected by P-glycoprotein (ABCB1)-related multidrug resistance (MDR1). Unlike conventional anthracyclines, ANN accumulates in multidrug resistant cell lines, inducing DNA damage and apoptosis. Additionally, in preclinical toxicology studies, ANN displayed a greatly reduced cardiotoxicity profile compared to DOX.

A liposomal formulation of ANN, termed L-Annamycin (L-ANN), is being evaluated in patients with acute myeloid leukemia (AML) in two phase Ib/IIa clinical trials in both the US and Europe. The patients are stringently followed for cardiotoxicity (Shephard et al., ASH 2020, submitted). There was no decrease in ejection fractions observed in any of the 20 patients treated to date and all cardiotoxicity biomarkers including troponin remained normal. In addition, echocardiogram analyses by the Duke cardio-oncology lab, and independently by a cardio-oncologist at Cleveland Clinic, were all normal. Considering that cardiotoxicity and MDR1-limited anthracyclines like daunorubicin are routinely used in a combination with ARA-C as induction therapy in AML, we here evaluated the combination of Ara-C with L-ANN pre-clinically.

Objective: The objective of the study was to assess in vivo efficacy of the combination of L-ANN with ARA-C in a pre-clinical model of AML.

Methods: The efficacy of L-ANN alone or in combination with ARA-C was investigated in a highly aggressive AML mouse model characterized by the p53-/-, MLL, ENL-FLT3, ITD mutations and genetically tagged with the cyan fluorescent protein mTurquoise2 for flow cytometry and microscopic visualization. L-ANN was intravenously administered at different dosing regimens (days 1, 2, 3 or 1, 3, 5 weekly, or 1 dose of 4 mg/kg once a week). ARA-C was administered by 5 daily intraperitoneal injections at 50 mg/kg, which was repeated every other week up to 3 times. The level of leukemia cells in peripheral circulation was analyzed by flow cytometry and AML cell presence in organ tissues was imaged by thick-mount fresh tissue confocal microscopy, in various disease stages.

Results: In all tested L-ANN administration regimens (days 1, 2, 3 or 1, 3, 5 weekly, or 1 dose of 4 mg/kg once a week), we observed a significant increase in the survival of ANNARAC cohorts (combination of L-ANN with ARA-C), when compared with the respective single agents. Specifically, upon intravenous infusion of 1x10⁵ AML1-mTurq2 cells into syngeneic immunocompetent C57BL6 mice, lethal AML disease developed with median survival of 14 days. Administration of L-ANN on a weekly basis significantly delayed leukemia progression, as evaluated by flow cytometry and fluorescence microscopy, resulting in survival increase to 34 to 40 days, in multiple experiments. The mice treated with 50 mg/kg of ARA-C daily for 5 days a week every other week using intraperitoneal injections showed moderate to limited response to the therapy with median survival ranging from 17 to 30 days. In contrast, the median survival of animals treated with the L-ANN/ARA-C combination using different schedules ranged from 44 to 76 days, with a fraction of animals living more than 180 days after implantation of AML cells. Remarkably, imaged on day 36, the bone marrow, spleen and lungs of mice receiving combination of L-ANN (4 mg/kg once a week) with ARA-C (50 mg/kg five times per week) showed no residual disease. These results are consistent with the increased survival observed for this combination.

Conclusion: This study demonstrated vastly higher efficacy of the L-ANN/ARA-C combination (ANNARAC) over that of the single agents in an immune-competent setting of an aggressive, p53-null AML model. Overall, these experiments indicate that L-ANN has the capacity to sensitize AML cells to the ARA-C induction regimen and support initiation of clinical development of L-ANN in combination with ARA-C in AML patients.