Phase I non-randomized clinical trial of allogeneic natural killer cells infusion in acute myeloid leukemia patients

INTRODUCTION: A new type of immune cell transplantation called allogeneic NK cell infusion is proposed as a potential universal off-the-shelf cell product for adoptive immune cell therapy in hematologic malignancies. DESIGN: A multicentral phase I non-randomized clinical trial was conducted to assess the safety, feasibility, and potential efficacy of adoptively infused NK cells in patients with refractory/relapsed AML. We evaluated the feasibility of the trial by considering cell production, patient selection, and treatment protocol. METHOD: Allogeneic NK cells were produced from random healthy unrelated donors; 10 patients were selected according to the inclusion criteria and were included in two groups in case of NK cell dose escalation. Two cell infusions were given, spaced 7 days apart, following a lymphodepletion conditioning regimen of fludarabin-endoxan administered 7 days before the first infusion. The intervention safety was scored using Common Terminology Criteria for Adverse Events (CTCAE) based on variations in vital signs due to cell infusion. NK cell chimerism, tumor burden, and duration of relapse were considered to be components of efficacy. The pilot feasibility evaluation was checked using the CONSORT platform. RESULTS: The NK cell infusion procedure was well tolerated, and no grade 2-5 toxicities related (possible or probable) to PB-NK cell infusion were observed. Four patients developed grade 1 transient chills, headaches, vomiting, and bone pain following each PB-NK cell infusion that were not required hospitalization. One of these patients (p01) died because of severe acute respiratory syndrome. Of 9 evaluable patients, 6 (66.6%) showed stable disease (SD) and 3 (33.3%) presented progressive disease (PD). Of 6 SD patients, 2 (p08 and p09) remained alive in SD and 3 patients (p04, p05 and p07) converted to PD at 9 months after infusion of NK cells, and 1 (p03) was not evaluable due to follow-up loss. No patient achieved complete remission. CONCLUSION: The study demonstrated the feasibility and safety of adoptive transfer of random healthy unrelated donor PB-NK cells in refractory/relapsed AML patients and supports continued study in phase II clinical trials in relapsed/refractory AML patients.