Temporal, Quantitative and Functional Characteristics of Single-KIR Positive Alloreactive NK Cell Recovery Account for Impaired Graft Versus Leukemia Activity after Haploidentical HSCT.

Haploidentical Hematopoietic Stem Cell Transplantation (haplo-HSCT) is a promising therapeutic option for patients lacking a fully compatible donor. Due to extensive T cell depletion, Natural Killer (NK) cell activity represents the only immunological protection against disease relapse for the first months after haplo-HSCT. Clinical studies have associated donor-recipient incompatibility for Human Leukocyte Antigen (HLA) ligands of Killer Immunoglobulin-like Receptors (KIR), with a marked anti-leukemic activity. Alloreactive donor NK cells carrying a single KIR whose ligand is missing in the recipient mediate a potent graft vs. leukemia effect, resulting in reduced incidence of relapse and increased Overall Survival (OS). These exciting results have recently been challenged by conflicting clinical and biological data from different groups. In the present study, we have characterized reconstitution of NK cells, in particular of alloreactive single-KIR⁺ NK cells, in 58 patients who received CD34⁺ selected haplo-HSCT for high-risk hematologic malignancies. One month after haplo-HSCT CD56^bright/CD56^dim NK cell subsets were subverted in their proportions and phenotypic features, accounting for enrichment in maturation intermediates. We show that CD25 and CD117 deregulation by CD56^bright, and NKG2A and CD62L by CD56^dim, are intrinsic to NK cell physiologic differentiation and support a sequential CD56^bright-to-CD56^dim NK cell maturation. Consistently, the in vitro functional potential of these maturation intermediates against leukemic blasts was heavily impaired, both in terms of cytotoxicity and of cytokine release. Full mature receptor repertoire reconstitution took at least three months. Alloreactive single-KIR⁺ NK cells had highly variable frequency ranging from less than 1% to more than 30% of NK cells circulating at 90–120 days after transplantation, independently from predicted NK alloreactivity. Importantly, out of three patients with predicted NK alloreactivity, none had a relative expansion of alloreactive single-KIR⁺ cells, accounting for less than 1% of circulating NK cells in two of them. As demonstrated by flow cytometric analysis of NK cell CD107a mobilization in response to the HLA class I negative target 721.221, single-KIR⁺ NK cells at three months after haplo-HSCT showed a not yet fully developed functional reactivity, which was recovered to donor-levels only at later time-points. In line with these observations, clinical outcome of haplo-HSCT was not affected in any way by the presence of donor NK alloreactivity. The incidence of relapse was virtually identical in patients transplanted from alloreactive or non-alloreactive donors. Taken together, our data shed new light onto the kinetics of NK cell differentiation in vivo and suggest that NK alloreactivity could be best exploited by the use of mature donor single-KIR⁺ selected alloreactive NK cells.