Outcomes of Unrelated Cord Blood Transplantation for Non-Malignant Diseases in Children Are Not Associated with KIR-Ligand Matching Between Donor and Recipient: a Study On Behalf of Eurocord.

Abstract 2301

Poster Board II-278

NK-cell alloreactivity in the setting of allogeneic transplantation is largely determined by the specificity of the killer immunoglobulin receptors (KIR) on donor NK cells for recipient HLA class I. We have recently shown that KIR ligand mismatch in the graft-versus-host (GvH) direction between donor and recipient is associated with improved outcomes after unrelated cord blood transplantation (UCBT) for patients with acute leukemia in complete remission. However, this finding was not confirmed in a series of reduced intensity conditioning and UCBT using 2 unrelated cord blood units. Therefore, those controversial results can raise some questions on the donor choice based on KIR ligand matching status. It is not known whether KIR ligand matching is associated with outcomes after UCBT for non-malignant disorders.

To address this question we have analyzed patients with non-malignant disorders who had received a UCBT using a single cord blood unit and were reported to the Eurocord registry with sufficient HLA-A, -B, -C typing to assign KIR ligand groups. In agreement with previous studies in malignant diseases, KIR ligand mismatch (KIR+) in the GvH direction occurred when the donor expressed one or several of the following KIR ligands which were lacking in the recipient: HLA-Bw4, and HLA-C group 1 or 2. In addition patients and cord blood units were grouped according to HLA A-3/A-11 alleles, which are putative KIR ligands. One-hundred-thirty-seven eligible patients were selected, mostly children (n=124). UCBT was performed from January 2000 to December 2008 in 47 EBMT-centers. The frequency of KIR ligand mismatched (KIR+) pairs reported was 23% (n=31). Forty-three percent of patients had bone marrow failure syndromes (BMFS) (n=60), other diagnosis included metabolic disorders (n=41), SCID (n=32), hemoglobinopathies (n=2), autoimmune diseases (n=2). Median follow up was 24 months (4-101). Median recipient age was 3 years. UCB units were HLA matched at 6 of 6 (n=38), 5 of 6 (n=62), 4 of 6 (n=34) and 3 of 6 (n=3). Median cell dose infused was 5.6 ×10E7 TNC/Kg and 2.4 ×10E5 CD34 cells/Kg. Conditioning regimen was myeloablative in 48% of cases and reduced intensity in 52% and included ATG in 90% and fludarabine in 42%. Graft-versus-host disease (GvHD) prophylaxis consisted of cyclosporine (CsA) and corticosteroids, CsA alone and CsA and MMF in 47%, 23% and 14% of patients, respectively.

Cumulative incidence (CI) of the day-60-neutrophil recovery was 76±3% (median time: 22 days) and it was not associated with KIR mismatch [(74±4% for KIR- and 80±6% for KIR+ group (p=0.9)]. CI of aGvHD was 21±3%, only 3 patients out of 31 with KIR+ presented aGvHD (II-IV). The estimated probability of 2-year overall survival (OS) was 63±4%; it was 64±5% in the group of KIR- and 63±3% in the KIR+ (p=0.5). The main cause of death was graft failure associated with infections (n=16, 32%). There was no statistical difference in frequency of causes of death between the 2 KIR ligand groups.

We have also looked the association of KIR ligand mismatch in the host versus graft (HvG) direction. The frequency of KIR mismatched pairs in the HvG direction was 22% and it was not associated with neutrophil recovery (76±4% for KIR ligand matched in HvG direction versus 73±6% for KIR ligand mismatched in HvG direction, p=0.9) or OS (64±5% for KIR ligand matched in HvG direction versus 58±7% for KIR ligand mismatched in HvG direction, p=0.6)

In conclusion KIR ligand mismatching in the GvH or HvG direction is not associated with outcomes after UCBT for children with non malignant disease. Therefore it is not useful in the algorithm of donor choice in this setting.