The Role of Missing Killer Cell Immunoglobulin-Like Receptor Ligands in T Cell Replete Peripheral Blood Stem Cell Transplantation from HLA-Identical Siblings

Abstract

The contribution of natural killer (NK) cells to graft-versus-malignancy (GVM) effects following hematopoietic stem cell transplantation (HSCT) remains uncertain, particularly in the HLA-identical setting. A model considering missing HLA ligands to the donor's inhibitory killer cell immunoglobulin-like receptor (KIR), termed the missing KIR ligand model, has been established in T cell depleted bone marrow transplantation (BMT), but lacks validity in other cohorts with different treatment characteristics. We hypothesized that the impact of missing KIR ligands on relapse-free survival (RFS) and overall survival (OS) in T cell replete peripheral blood SCT (PBSCT) differs from that in the T cell depleted BMT setting, and retrospectively evaluated 100 consecutive, HLA-identical sibling transplantations for hematologic malignancies. In addition to KIR ligand status, we considered the donors' activating KIRs and grafted NK, T, and CD34(+) cell doses. Our findings demonstrate noninferiority for OS (P = .005) and RFS (P = .002) for the heterozygous HLA-C group KIR ligand status (C1/2; n = 47) compared with patients missing either C1 or C2 (n = 53). Similarly, OS (P = .031) and RFS (P = .034) of Bw4-positive patients was noninferior to that of patients missing a Bw4 ligand to KIR3DL1. By multivariate analysis, C1/2 heterozygous patients had a favorable risk ratio (RR) for relapse (RR = 0.28; P = .003), RFS (RR = 0.56; P = .046), and acute graft-versus-host disease grade II-IV (RR = 0.36; P = .05). Following reduced-intensity conditioning (RIC), but not standard-intensity conditioning, myeloablative (MA) transplantation, a grafted NK cell dose above the median (3.4 x 10(7)/kg) was associated with a lower risk of relapse (RR = 0.57; P = .003) and improved survival (RR = 0.78; P = .03). Overall, our findings support a role for NK alloreactivity in HLA-identical HSCT, but argue against a favorable impact of missing KIR ligands in the given setting. We conclude that the mechanism favoring the missing KIR ligand constellation in T cell depleted BMT may not operate in T cell replete PBSCT. The reasons for this differential effect remain unresolved.