Donors with group B KIR haplotypes improve relapse-free survival after unrelated hematopoietic cell transplantation for acute myelogenous leukemia

Abstract

Survival for patients with acute myeloid leukemia (AML) is limited by treatment-related mortality (TRM) and relapse after unrelated donor (URD) hematopoietic cell transplantation (HCT). Natural killer (NK)-cell alloreactivity, determined by donor killer-cell immunoglobulin-like receptors (KIRs) and recipient HLA, correlates with successful HCT for AML. Hypothesizing that donor KIR genotype (A/A: 2 A KIR haplotypes; B/x: at least 1 B haplotype) would affect outcomes, we genotyped donors and recipients from 209 HLA-matched and 239 mismatched T-replete URD transplantations for AML. Three-year overall survival was significantly higher after transplantation from a KIR B/x donor (31% [95% CI: 26-36] vs 20% [95% CI: 13-27]; P = .007). Multivariate analysis demonstrated a 30% improvement in the relative risk of relapse-free survival with B/x donors compared with A/A donors (RR: 0.70 [95% CI: 0.55-0.88]; P = .002). B/x donors were associated with a higher incidence of chronic graft-versus-host disease (GVHD; RR: 1.51 [95% CI: 1.01-2.18]; P = .03), but not of acute GVHD, relapse, or TRM. This analysis demonstrates that unrelated donors with KIR B haplotypes confer significant survival benefit to patients undergoing T-replete HCT for AML. KIR genotyping of prospective donors, in addition to HLA typing, should be performed to identify HLA-matched donors with B KIR haplotypes.

Figure 1

Transplantation using donors with a KIR B haplotype improves relapse-free survival (RFS) irrespective of the recipient KIR haplotype status. Donor (D) and recipient (R) DNA samples from 448 URD transplants were genotyped for 16 KIR loci. KIR gene content was used to identify haplotypes (A or B), from which KIR genotypes were assigned (KIR A/A or KIR B/x). The Kaplan-Meier curves demonstrate RFS for each donor/recipient genotype pairing (A/A into A/A, A/A into B/x, B/x into A/A, and B/x into B/x). RFS was significantly better after transplantation using KIR B/x donors (28% [95% CI: 23-33]; n = 317) compared with A/A donors (17% [11%-24%]; n = 131; P = .003).The recipient KIR genotype had no effect (P = .75).

Figure 2

Transplantation using donors with KIR B haplotypes improves overall survival. The univariate Kaplan-Meier curves for overall survival (OS) are shown for patients receiving transplants from donors with (KIR B/x; n = 317) or without (KIR A/A; n = 131) KIR B haplotypes. Survival is significantly better if the donor had a KIR B haplotype (31% [95% CI: 26-36] vs 20% [95% CI: 13-27]; P = .007).

Figure 3

Relapse-free survival benefit from KIR B haplotype donors after HLA-matched and HLA-matched/KIR ligand–matched transplantation. The univariate Kaplan-Meier curves demonstrate the relapse-free survival (RFS) for patients receiving transplants from donors with or without KIR B haplotypes who were (A) 10/10 HLA matched, (B) HLA mismatched and KIR-ligand matched, or (C) HLA mismatched and KIR-ligand mismatched.

Figure 4

The survival benefit of a donor KIR B haplotype is not dependent on the presence of KIR2DS2. Donors with KIR B haplotypes were divided into 2 groups based on the presence (2DS2⁺, n = 252) or absence (2DS2⁻, n = 65) of KIR2DS2. Survival was significantly better for patients who received transplants from donors with either KIR B haplotype (2DS2⁺: 31% [95% CI: 25-37] or 2DS2⁻: 31% [95% CI: 20-42]) versus donors homozygous for the A haplotype (A/A: 20% [95% CI: 13-27]; P = .02).

Figure 5

Transplantation using donors with KIR B haplotypes leads to less relapse and TRM, and equivalent acute GVHD, but increased chronic GVHD. The cumulative incidences of relapse, TRM, acute GVHD, and chronic GVHD are shown for transplantation using donors with (KIR B/x) or without KIR B (KIR A/A) haplotypes. The incidence of chronic GVHD was higher using KIR B/x donors (39% [95% CI: 34-45]) than when using KIR A/A donors (28% [95% CI: 20-35], P = .02). Acute GHVD, relapse, and TRM were not significantly different between donor KIR haplotype groups (P > .15).