ALBERT

Description: Background: The effect of NK-cell alloreactivity on outcome of unrelated stem cell transplantation remains controversial. Some studies have shown a survival benefit for KIR ligand mismatch transplantation. Killer cell immunoglobulin-like receptors [KIRs] are expressed on the surface of NK cell and T-cell subsets. Inhibitory KIRs have specificity for defined alleles of HLA class I and delivered an inhibitory signal to the cells. If the inhibitory signal does not find a corresponding ligand, it is proposed that an activated signaling will lead through an unknown ligand to cell lysis. The KIR-genes are polymorphic. Two broad haplotypes exist which mainly differ in the number of activating KIRs. KIR-haplotype A mainly encode for inhibitory receptors and only for 1 activating [KIR2DS4], whereas the group B haplotype encodes more for activating KIRs [KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS5 und KIR3S1]. Methods: We investigated the impact of KIR ligand mismatch and of donor KIR haplotype genes on outcome of 123 patients (Caucasians) who received standard myeloablative conditioning with ATG followed by unrelated stem cell transplantation from HLA matched [n=87] or mismatched [n=36] donors. The median age of the study population was 35 years [range 20–58]. Diagnoses were CML n=32, AML/MDS n=38, ALL n=39, Hemophagocytosis and inborn errors n=14. The stem cell source was bone marrow in 75 patients [61 %] or PBSC n=48. KIR genotypes were performed by PCR-SSP (Pel-Freez [Dynal Biotech]) and/or PCR-SSOP (Luminexä, OneLambda). KIR haplotypes were determined with the help of the OLITYPE program (Hum Immunol 1991 30:22–26). 38 [31 %] of the donors were homozygous for haplotype A, whereas 28 [20 %] where homozygous for haplotype B. 61 patients [49 %] where heterozygous with A- or B-allele in KIR genotyping. KIR ligand mismatch was found in 15 % of the patients. Results: KIR ligand mismatch had no significant influence on treatment related mortality [p=0.3], relapse [p=0.3], disease free survival [p=0.16], and on overall survival [p=0.3]. In contrast donors with group A haplotype KIR genes had in comparison with other haplotypes a significantly low relapse rate [3% vs 39 %, p=0.005], a better estimated 4 years disease free [72% vs 38 %, p=0.007] and overall survival [72% vs 45 %, p=0.031]. In a multivariate analysis transplantation with donors carrying group A haplotype KIR genes remained an independent factor for relapse [RR: 0.144, p=0.06], for disease free [RR: 0.40, p=0.009] and overall survival [RR: 0.47, p=0.03]. Conclusion: After in vivo T-cell depleted [ATG] unrelated stem cell transplantation with donors carrying multiple inhibitory KIR genes [group A KIR haplotype] the risk of relapse is reduced and resulted in a significantly better disease free and overall survival. These results may have impact on donor selection for unrelated stem cell transplantation.