Background: Activating and inhibitory killer immunoglobulin-like receptors (KIRs) regulate function of NK cells and a subset of T cells. KIR genotype, in particular the content of activating KIR genes is highly polymorphic. NOD2/CARD15 protein is broadly expressed in APCs and lymphocytes. Single nucleotide polymorphisms (SNPs) of this gene have been reported to impair the pathogen elimination and trigger pathologic immunologic reactions like GvHD. The goal of this prospective study was to evaluate the impact of donor’s and recipient’s KIR and NOD2/CARD15 genotypes on outcome after allogeneic hematopoietic stem cell transplantation (alloHSCT).

Pateints and methods: One-hundred-two consecutive patients with hematological malignancies, aged 32(18–58)y, treated with alloHSCT from HLA-matched related (n=34) or matched unrelated donor (MUD) (n=68) were included. The conditioning regimen was myeloablative, GVHD prophylaxis consisted of CsA, Mtx, and, in case of MUD-HSCT, pre-transplant ATG. Donors and recipients were tested for 11 KIR genes as well as SNP8,12,13 of the NOD2/CARD15 gene. In addition, immune reconstitution including KIR expression on T cells, was analyzed on days +28, +56, +100, +180, and +360.

Results: Overall survival (OS) rate at 2y was significantly lower in alloHSCT with at least one activating KIR mismatch compared to transplants with full compatibility (62% vs. 86%, p=0.01). In particular, the presence of at least one activating KIR in the donor with its absence in the recipient (D+R−) was associated with decreased probability of OS (60% vs. 78%, p=0.01) and DFS (58% vs. 82%, p=0.005), as well as increased incidence of non-relapse mortality (NRM) (27% vs. 7%). KIR2DS1 and KIR3DS1 D+R− mismatches resulted in increased risk of grade II–IV acute GvHD, whereas KIR2DS3 and KIR2DS2 D+R− mismatches were associated with increased risk of chronic GvHD. The presence of at least one activating KIR D+R− mismatch was associated with increased CD8+/CD4+ T cell ratio up to day +100. In all cases of incompatibility regarding KIR2DS1, KIR2DS2 and KIR3DS1, T cells with expression of respective receptors could be detected up to 360 days after alloHSCT. The presence of SNP8 of the NOD2/CARD15 gene in the recipient was associated with decreased probability of OS (20% vs. 70%, p=0.005) and DFS (20% vs. 70%, p=0.01) as well as increased incidence of NRM (60% vs. 17%) and grade III–IV acute GvHD (67% vs. 8%). In a multivariate analysis including KIR and NOD2/CARD15 polymorphisms together with other potential risk factors, increasing number of D+R− activating KIR mismatches as a linear variable appeared to independently influence OS (HR: 1.3, p=0.02), DFS (HR: 1.3, p=0.008), NRM (HR: 1.4, p=0.02), grade II–IV acute GvHD (HR: 1.4, p=0.001), and chronic GvHD (HR: 1.2; p=0.02). Recipient SNP8 of NOD2/CARD15 was predictive for OS (HR: 5.5, p=0.003), DFS (HR: 4.4, p=0.008), NRM (HR: 5.9, p=0.006), grade III–IV acute GvHD (HR: 6.1, p=0.02), and chronic GvHD (HR: 3.7; p=0.03).

Conclusions: Both activating KIR D+R− mismatches and recipient SNP8 of NOD2/CARD15 appear to enhance alloreactivity and independently influence survival after alloHSCT. Evaluation of these polymorphisms may contribute to better donor selection and optimization of the alloHCT procedure.

Author notes

Disclosure: No relevant conflicts of interest to declare.