Natural Killer Cells Enhance Transplantation Outcomes

Natural killer (NK) cells, the first lymphocyte subset to reconstitute after hematopoietic cell transplantation (HCT), may enhance transplant outcomes by killing virally-infected or malignant cells to reduce relapse and treatment-related mortality by promoting engraftment and by reducing graft vs. host disease. The function of NK cells is regulated by the net balance of signaling via several families of activating or inhibitory receptors. The killer-cell immunoglobulin-like receptor (KIR) family is of particular importance in HCT because of its interactions with class I human leukocyte antigen (HLA) molecules. HLA-C, HLA-Bw4 and some HLA-A allotypes function as ligands for the inhibitory KIR receptors, thus mediating tolerance to self. HLA-mismatched HCT may generate alloreactive NK cells in the recipient when there is a KIR-ligand mismatch at HLA-C, B or A. The proof of concept for this principle was established by the Perugia group in haploidentical transplants, where KIR-ligand mismatch was associated with reduced relapse in patients with acute myeloid leukemia. Subsequent groups have studied this and other algorithms, including KIR ligand absence or KIR-KIR ligand genotyping to evaluate the role of NK cells in mediating protection after allogeneic HCT. Human KIR haplotypes are defined by gene content, where the A haplotype contains mainly inhibitory KIR, and B haplotypes contain more activating KIR. Unrelated donors with more KIR B haplotype genes have been associated with protection from relapse and improved survival in AML, an effect which is enhanced in recipients expressing HLA-C1. Although NK cells mediate beneficial effects after HCT, compared to adult donor NK cells, engrafting NK cells are immature and hyporesponsive when exposed to tumor targets. Recently human cytomegalovirus (CMV) has been shown to drive the expansion of a population of long-lived CD57⁺NKG2C⁺ NK cells with attributes of immunologic memory. These cells have heightened capacity for cytokine production or cytolytic response to tumor targets. Several groups have shown that CMV reactivation correlates with reduced relapse after allogeneic HCT, suggesting that these CMV “adaptive” NK cells may contribute to the graft vs. leukemia effect. Strategies to isolate and expand “adaptive” NK cells without clinical CMV reactivation are being explored. Lastly, adult donor NK cells are an appealing population to exploit for adoptive cellular therapy. Donors can be chosen based on predicted NK cell alloreactivity or based on KIR gene content. Adoptively transferred NK cells which expand and proliferate in vivo in response to cytokine stimulation (IL-2 or IL-15) are potent anti-tumor effectors which do not mediate graft vs. host disease. Several groups are using haploidentical adult NK cell products to treat acute myeloid leukemia, multiple myeloma, lymphoma, myelodysplasic syndrome and a variety of solid tumors. Various approaches using adoptive transfer of NK cells, together with HCT, targeting antibodies which mediate antibody dependent cellular cytotoxicity, or cytokine stimulation are being tested clinically. In summary, we are just beginning to define the complexity of NK cell interactions with HLA and other ligands and to describe different properties of various NK cell subsets to develop more sophisticated strategies to exploit NK cells to treat cancer.