Identification and Utilization of Pre-Search Killer Cell Immunoglobulin-Type Receptor (KIR) B(x) Unrelated Donors Improves the Selection of KIR Favorable Donors, the Transplant Coordinator Perspective

Background KIR B(x) haplotype donors have been reported to improve outcomes in patients (pts) receiving unrelated donors (URD) for myeloid malignancies, non-Hodgkinu0027s lymphoma (NHL) and multiple myeloma (MM). Donors are categorized as KIR “Best” or “Better”. Our center performs donor KIR genotyping and haplotype assignment routinely as part of our URD evaluation for pts undergoing transplantation for myeloid disorders, NHL and MM. When presented with multiple equivalently HLA matched URDs, our physicians will prioritize KIR B(X) donors over KIR A/A donors. Weisdorf et. al. (BBMT 2018) suggested that pre-search donor KIR genotyping would accelerate/optimize donor selection. Since 2015, donor KIR genotyping data for u003e 3 million donors from the DKMS registry is available. We investigated our ability to enhance KIR-favorable donor selections using the DKMS pre-search KIR. Methods In 2018 we requested access to individual KIR donor typing through the National Marrow Donor Program (NMDP) Scientific Services for donors registered by DKMS Germany, DKMS Poland and obtained access to the DKMS Donor Navigator website. Pts with multiple suitable donors were sorted for KIR, if registered by DKMS. Up to 6 donors per patient were selected from the pre-search KIR B(x) donors who ranked KIR “Best”, “Better” and “Neutral”. For pts whose pre-search screen did not identify an adequate number of KIR B(x) informative donors we selected additional donors (KIR unknown) for confirmatory typing. Results Thirty-five pts had informative KIR B(x) pre-search donors identified. There were 1-27 KIR B(x) available donors per pt. The online KIR B Content Calculator ranked these donors as KIR “Best”, “Better”, or “Neutral”. Donors were selected by the physicians, who were unaware of the pre-search ranking. As of Oct 2019, 23 pts were transplanted, 4 are pending. Of these 27 donors selected, 18 were identified in the pre-search group. Six donors (33%) were ranked KIR B “Best”, 10 (55%) “Better”, and 2 (11%) “Neutral”. Nine pts used donors without a pre-search. Five pts did not proceed to an URD transplant, and 3 pts have ongoing searches. Using the pre-search information, 66% of pts received a KIR B(x) donor and we were able to enrich KIR B(x) “Best” and “Better” donor options. Based on registry data, approximately 11%, 20% and 69% of donors randomly selected from a donor registry will be ranked KIR “Best”, “Better” and “Neutral “respectively. For our pt population, we were able to more than double the selection of KIR “Best” and “Better” donors for transplantation. Conclusion The inclusion of KIR donor genotyping on donor registries enriches the number of preferred KIR B (x) donors available and allows a more efficient selection of KIR “Best” and “Better” donors.