Phase I-II Trial of Adoptive Immunotherapy with Haploidentical KIR Ligand-Mismatched Natural Killer Cells in High Risk Acute Myeloblastic Leukemia Patients.

Natural killer (NK) cells play a critical role in innate immunity and are relevant for the immune control of tumors. Their effector function is regulated by a number of activating and inhibitory receptors, termed killer immunoglobulin receptors (KIRs). In haploidentical T-cell depleted transplantation the donor/recipient KIR mismatch significantly impacts on tumor cell killing, particularly in acute myeloid leukemia (AML). Therefore, haploidentical KIR-mismatched NK cells may be used as adoptive immunotherapy for high risk AML patients in non-transplantation settings. Thirty-one high risk AML patients entered a phase I-II study of NK-cell based immunotherapy and were screened for the avalilability of one haploidentical KIR ligand mismatched donor. Ten of them resulted as having one suitable donor. NK cells were enriched from steady-state leukaphereses by using a double-step immunomagnetic separation system (Miltenyi Biotec, Germany), consisting in

1. depletion of CD3⁺ T cells followed by

2. positive selection of CD56⁺ NK cells, and then cryopreserved.

CD56⁺ NK cells were enriched from 8.28±3.7% to 85.9±3.2% (recovery 67.9±15%, viability >92%). So far, 8 patients (1 partial remission, 4 progressions and 3 complete remissions) received NK cell infusion which was preceeded by immunosuppressive chemotherapy, including fludarabine 125 mg/m² and cyclophosphamide 4g/m². AML patients received 6 subcutaneous injections of interleukin 2 (IL-2) after adoptive immunotherapy. The median number of reinfused NK cells was 3.6 × 10⁶/Kg and contaminating CD3⁺ T cells were always less than 1×10⁵/Kg. NK cells were capable to kill NK-sensitive K562 cells in a flow-cytometry-based cytotoxicity assay. The procedure was well-tolerated and no significant toxicity, including GVHD, related to NK cell infusion was observed. The patient in partial remission obtained a complete remission, which lasted for 6 months; among the 4 patients with active/progressive disease 2 patients showed the persistence of disease and one patient died during the aplastic phase. The 3 patients in complete remission are stable after 5, 3 and 4 months of follow up. By using a RT-PCR-based chimerism assay, we demonstrated the presence of adoptively transferred NK cells in 2/4 patients with a peak in circulating NK cell number at day 12 after infusion. Functional biological studies demonstrated the presence of alloreactive NK cell clones in all the donors (at the frequencies of 11 ± 4%) and in 3/4 patients (at the frequences of 16 ± 5%) only at day 3 after infusion of NK cells. In summary, a two-step enrichment of CD56⁺ NK cells allows the collection of a suitable number of target cells to be used as adoptive immunotherapy in AML patients. Infusion of cryopreserved NK cells is feasible and safe and adoptively transferred NK cells can be detected in the peripheral blood after infusion.