A Phase I Study of the Vascular Disrupting Combretastatin, OXi4503, in Patients with Relapsed and Refractory Acute Myeloid Leukemia (AML) and Myelodysplastic Syndromes (MDS)

AML depends on blood vessel networks for disease progression and protection from conventional chemotherapy. In prior preclinical work, we demonstrated that the vascular disrupting combretastatin, OXi4503, had significant activity against human AML by dual targeting of nascent blood vessels and malignant myeloblasts. In solid tumor studies, the maximal tolerated dose (MTD) of OXi4503 was 11–14 mg/m2 and the dose limiting toxicity (DLT) was hypertension. Therefore, we initiated a phase I clinical study of OXi4503 in patients with relapsed and refractory AML and advanced MDS.

The primary objectives of the phase I dose escalation study (NCT01085656) were to establish safety and determine MTD in patients with hematological malignancies. Eligibility included history of AML or MDS (RAEB-1, RAEB-2), failure to respond to at least one induction therapy, good performance status (ECOG ≤ 2), adequate kidney and liver functions, blood pressure < 140/90 mmHg, no history of major operative surgery within 28 days, and no history of cerebrovascular accident. DLT was defined as grade 3 or greater non-hematological drug-related toxicity or failure of blood count recovery by Day 42 after the start of Cycle 1. Amlodipine was administered as pre-medication if baseline blood pressure was > 120/80 mmHg. OXi4503 was administered by intravenous infusion over 10 minutes on Days 1, 8, 15 and 22 of each 28-day cycle, and escalated according to a defined dose schema. Continued weekly dosing was permitted until disease progression or unacceptable toxicities.

Between May 2011 and August 2012, 5 patients with refractory AML were enrolled in an ongoing phase I safety study. The median age was 61 years (range, 46–76). Most patients were male (80%). Cytogenetics were adverse in one patient and intermediate in four. The median number of prior therapies was 4 (range, 1–7). Two subjects were assigned to the 2.5 mg/m2 dosing cohort, two received 3.75 mg/m2 and one received 5 mg/m2. No patients developed DLTs. Adverse events attributable to OXi4503 infusion included bone pain, fever, anemia and thrombocytopenia. The mean change in systolic blood pressure after OXi4503 infusion was +11.5 mmHg in the 2.5 mg/m2 cohort, +5.5 mmHg in the 3.75 mg/m2 cohort, and +0.82 mmHg in the 5 mg/m2 cohort. Plasma LDH and uric acid increased by at least two-fold within hours after OXi4503 infusions, suggesting leukemia cell lysis. Overall response was 2/5 (40%), with one patient achieving a marrow complete remission (mCR) and one patient achieving a partial remission (PR). Median number of cycles was 1 (range, 1–6). Three of five subjects discontinued the study due to AML progression and one due to pneumonia. One patient with AML treated with 5 mg/m2 OXi4503 has received 6 months of treatment and continues to receive weekly infusions. Analysis of bone marrow (BM) after OXi4503 infusions showed alterations in endothelial cell (EC) morphology and acute increases in VEGF expression (1.28–1.75 × pre-dose levels).

At low doses the vascular disrupting combretastatin OXi4503 is well tolerated with preliminary evidence of disease response in heavily treated, refractory AML. Biological activity associated with OXi4503 included changes in BM EC morphology and a reactive VEGF response, suggesting that addition of angio-inhibitory agents such as anti-VEGF agents to vascular disruption with OXi4503 may be worthy of study to improve AML regression.

Cogle:OXiGENE: Research Funding; Leukemia & Lymphoma Society: Membership on an entity's Board of Directors or advisory committees, Research Funding. Chaplin:OXiGENE: Consultancy. Balkissoon:OXiGENE: Employment.