Genetic Association Analysis Between KIR2DS4 and Its Variant KIR1D With Pediatric Acute Lymphoblastic Leukemia

The killer immunoglobulin-like receptor (KIR) gene family encodes receptors of natural killer (NK) cells and some T cell subsets and is thereby involved in the regulation of anti leukemia effects. Every human being possesses an individual pattern of KIR genes, which can be assigned to either KIR haplotype A (which lacks KIR2DL2, KIR2DL5, KIR2DS1, KIR2DS2, KIR2DS3, KIR2DS5 and KIR3DS1) or haplotype B (which contains at least one of the activating KIRs, KIR2DL2 or KIR2DL5). KIR2DS4 is the only activating receptor with a short cytoplasmic tail, which can be found in individuals with KIR haplotype A. Interestingly, besides the cell membrane-anchored receptor, a truncated soluble protein (designed as KIR1D) is produced once the 22 bp deletion in exon 5 of KIR2DS4 is present. Previously, Giebel et al., Leukemia 2008, reported the association of KIR1D with leukemia, in particular with chronic myelogenous leukemia (CML). The aim of our study was to clarify whether the KIR1D allelic variant plays any role for innate immunosurveillance against acute lymphoblastic leukemia (ALL) and whether KIR1D subclassification is helpful for donor selection in the case of pediatric ALL before undergoing haploidentical hematopoietic cell transplantation (HHCT).

312 pediatric patients with ALL and 222 healthy controls were analyzed for the presence of KIR2DS4 and its variant KIR1D and then further subdivided into the following groups: 2DS4-/1D+ (homozygous for the deletion variant KIR1D), 2DS4+/1D+ (heterozygous), 2DS4+/1D- (two intact KIR2DS4 alleles) and 2DS4-/1D- (no KIR2DS4 allele). Furthermore, we investigated the influence of the KIR2DS4 variants on event-free and relapse of 69 patients with ALL receiving HHCT. The 22 bp deletion of KIR2DS4 was determined by PCR and agarose gel electrophoresis using primers described by Alves et al., Tissue Antigens 2009.

Of the 312 patients with ALL, n=160 (51.3%) were identified as 2DS4-/1D+, n=101 (32.4%) as 2DS4+/1D+, n=35 (11.2%) as 2DS4+/1D- and n=16 (5.1%) as 2DS4-/1D-. The frequencies in the control group were n=136 (61.3%) for 2DS4-/1D+, n=55 (24.8%) for 2DS4+/1D+, n=25 (11.3%) for 2DS4+/1D- and n=6 (2.7%) for 2DS4-/1D-. In our study, the occurrence of 2DS4+ alleles, as well as 1D+ alleles were similar for patient and control group (p>0.050, respectively).

In order to discover any possible impact of the different KIR2DS4 variants on outcome after HHCT, a retrospective analysis using cumulative incidence adjusted for competing risk (CICR, Scrucca et al., Bone marrow transplantation 2007) was performed. As previously reported by our group, patients with ALL transplanted from a KIR B haplotype donor were found to have lower relapse rates (Oevermann et al., ASH 2011). In the presented study, n=34 (49.3%) stem cell donors showed 2DS4-/1D+, n=20 (29.0%) 2DS4+/1D+, n=5 (7.2%) 2DS4+/1D-, n=10 (14.5%) 2DS4-/1D-. We found that the various KIR2DS4 variants did not influence incidence of relapse rate (2DS4-/1D+ 42.5%, 2DS4+/1D+ 55.0%, 2DS4+/1D- 20.0%, 2DS4-/1D- 30.0%, p=0.514), as well as event-free survival of ALL patients (2DS4-/1D+ 45.7%, 2DS4+/1D+ 40.0%, 2DS4+/1D- 60.0%, 2DS4-/1D- 50.0%, p=0.870).

We conclude from our results that KIR1D seems to have no meaningful influence on immunosurveillance against ALL. In contrast to KIR haplotype B donor selection, the preference of a KIR1D positive donor has no additional beneficial impact on the outcome of patients with ALL receiving HHCT.

No relevant conflicts of interest to declare.