Infusion of Umbilical Cord Blood-Derived Natural Killer Cells in Immune-Deficient Mice for the Treatment of Resistant Acute Myeloid Leukaemia

Adoptive transfer of allogeneic natural killer (NK) cells represents a promising treatment approach against acute myeloid leukaemia (AML). Success of this NK cell immunotherapy is dependent on obtaining high numbers of functional NK cells that have the potential to survive in vivo. The use of umbilical cord blood (UCB) CD34+ cells as a source of allogenic NK cells is an interesting method that can generate a readily available, non-invasive, off the shelf cellular product.

We developed a cytokine-based culture method for the generation of NK cell products derived from CD34+hematopoietic progenitor cells (HPC) isolated from fresh UCB units. Immuno-phenotyping of ex vivo expanded NK cells showed typical inhibitory and activating NK receptors except for CD16 and the KIR receptors.

UCB-derived NK cells displayed good cytolytic activity against NK-sensitive K562 cells with a percentage of specific lysis varying from 30 to 50%. Cytolysis was directly correlated to CD94 expression since CD94-sorted NK cells were responsible for all the in vitro cytolytic function of differentiated NKs against K562 cells. There was an inconstant susceptibility of patient-derived primary AML cells to UCB-derived NK lysis in vitro with a specific lysis ranging from 0 to 25%.

We further characterized UCB-derived NK cells by investigating their toxicity, biodistribution, and anti-leukemic potential in vivo. As adoptive transfer of NK cells is an attractive approach for treating refractory leukemia, immune deficient mice were engrafted with a patient derived AML strain resistant to NK-mediated lysis and doxorubicin.

After successful engraftment and randomization, leukemic mice were injected with either UCB- derived NK cells or NK cells from healthy donors (NKhds) or doxorubicin, with one control group that didn't receive any treatment. Mice were sacrificed after 2 weeks of treatment and leukemia load along with NK distribution were evaluated by flow cytometry in the blood, bone marrow (BM) and spleen.

There was no evidence of toxicity of UCB-derived or healthy donors NK cells in mice. Both types of cells were preferentially found in the blood and in the spleen, even though NKhds reached much higher levels than UCB-derived NKs.

As for treatment efficacy, none of our treatment showed anti-leukemic potential based on the absence of decrease of leukemic cells in BM, blood, and spleen. In vivo microenvironment didn't overcome resistance of the patient derived AML cell to NK lysis or to doxorubicin.

Remarkably, all of the UCB derived NK cells founded in vivo expressed the CD94 whereas not more than 20% of the injected cells were positive for this marker. Whether it was by in vivo selection or by in vivo differentiation must be investigated.

Interestingly, a small cell population with CD56 and CD34 double staining was distinguished in UCB-derived NK and NK healthy donor treated leukemic mice suggesting in vivo interaction between leukemic and NK cells. Further characterization of this population may help to understand the molecular mechanism of leukemic recognition by NK cells and resistance of leukemic cells to cytolysis.

In conclusion, UCB-derived NK generation is feasible. Investigation of the role of CD94 in these cells is needed, as cell sorting by CD94 selection in addition to the CD56 could be an interesting approach in the future to select highly functional expanded NK cells before therapeutic use. Furthermore, infusion of UCB-derived NK cells into immune-deficient mice is achievable and non-toxic. However, in vivo environment didn't overcome primary in vitro resistance of AML cells despite an established interaction. Additional elucidation of AML resistance mechanisms to NK lysis is mandatory before therapeutic application.