Phase 1 Clinical Trial of Adoptive Immunotherapy Using "Off-the-Shelf" Activated Natural Killer Cells (aNK) in Patients with Refractory/Relapsed Acute Myeloid Leukemia

Introduction: Natural killer (NK) cells mediate direct anti-tumor effects via their cytotoxic and cytokine-secreting functions. We and others have demonstrated that acute myeloid leukemia (AML) patients have a low NK-cell frequency and significantly depressed NK-cell activity. Activated NK cells (aNK) are generated by expansion of a human, interleukin-2 (IL-2)-dependent NK cell line (NK-92). aNK cells express activating receptors, lack most of the inhibitory killer-cell immunoglobulin-like receptors (KIRs) and mediate cytotoxicity against leukemia cell lines as well as primary leukemia blasts. Anti-leukemia effects of aNK cells were demonstrated in vivo in SCID mice xenografted with human leukemia cells. Advances in ex-vivo expansion of aNK cells under cGMP conditions and their strong anti-leukemia activity provided a rationale for conducting a phase 1 clinical trial (NCT00900809) with adoptively transferred aNK cells in patients with refractory/relapsed AML.

Method: We conducted a phase 1, open label clinical trial at the University of Pittsburgh to evaluate the safety of aNK cells in patients ≥ 18 years old withrefractory and relapsed AML. Clinical-grade aNK cells were provided by NantKwest and expanded in the GMP facility at the Center for Cell and Gene Therapy at the Baylor College of Medicine, Houston, TX. Two cell-dose levels were used: 1 x 10⁹ cells/m² and 3 x 10⁹ cells/m². Patients were enrolled to dose levels based on the traditional 3 + 3 dose-escalation design. aNK cells were administered in the outpatient setting. One treatment course consisted of a total of 2 infusions of the same cell dose, each cell infusion delivered 24h apart (Days 1&2). Bone marrow biopsy was performed 21 days after a treatment course with aNK cells. Patients who had stable disease or reduction of leukemia blasts were eligible to receive additional aNK infusions.

Results: Seven patients with refractory/relapsed AML were treated with a total of 20 aNK cell infusions. The median age was 71 years (range, 56-80 years). All patients had previously received multiple therapies for AML. Three patients were treated with the cell dose of 1 x 10⁹ aNK /m² and four patients with 3 x 10⁹ aNK /m². All seven patients completed the first course of therapy. None of the 7 patients experienced dose limiting toxicities during the aNK administration or during the 21 days observation period post-infusion. No grade 3-4 toxicities related (probable or definite) to the aNK infusion occurred. One patient developed fever, rigors, and hypotension 2h post infusion that required hospitalization; these known side effects were reversible with supportive care, intravenous hydration, and antipyretics. Three of the seven patients exhibited signs of clinical activity after the first course of treatment; in one patient, the blast percentage was reduced from 70% to 48% after a course of treatment, and the patient received an additional course of aNK cells. In two patients, the blast percentage remained stable after the first course of treatment; one of these 2 patients received additional two courses of treatment with aNK cells.

Conclusion: The trial demonstrated the safety and feasibility of therapy with "off-the-shelf" aNK cells, and provided early evidence of efficacy in heavily pretreated patients with refractory/relapsed AML. No grade 3-4 toxicity occurred with the maximal cell dose used. These data provide the foundation for combination immunotherapy trials for the optimization of aNK cell therapy in patients with AML.