Bioinformatics in donor search
There are three important areas when it comes to finding the best potential match for a patient: HLA, matching and frequencies.
HLA (Human Leucocyte Antigen)
Histocompatibility testing for stem cell transplantation has become increasingly complex as newly discovered human leukocyte antigen (HLA) alleles are defined continuously. HLA typing results reported by laboratories are used by physicians and donor registries to match donors and recipients. To communicate effectively, there is a common language and terminology for histocompatibility. WMDA ICT Working Group has developed tools to help organisations to stay up to date and be able to validate their HLA. More information can be found here. In addition, WMDA updates members regularly about ongoing developments in the field of HLA-nomenclature/assignment approaches. In a number of ways, WMDA is leading in defining ICT standards for communication/interpretation of HLA typing results. Examples are the following publications:
- Bochtler W., et al. A comparative reference study for the validation of HLA-matching algorithms in the search for allogeneic hematopoietic stem cell donors and cord blood units HLA Immune Responds Genetic doi: 10.1111/tan.12817 (2016 Apr. 22).
- Bochtler W, et al. World Marrow Donor Association framework for the implementation of HLA matching programs in hematopoietic stem cell donor registries and cord blood banks. Bone Marrow Transplantation 46:338-343 (2011).
The accuracy of HLA-matching algorithms is crucial to ensure the correct and efficient identification of optimal unrelated donors and cord blood units for patients in need of a haematopoietic stem cell transplant. In a recent study WMDA has validated five established matching algorithms by comparing the output of five matching algorithms in a study. In addition, a set of validation tools have been established to work on the accuracy of existing matching algorithms. For more information, see this article.
In searches for patients of North-West European ancestry, a 9/10 HLA-matched donor can be identified for 60–80% of the patients. For patients from other populations the percentage is lower. Many transplant centres are using search algorithms based on allele/haplotype frequencies to make a decision to transplant with a mismatched donor, or to select a cord blood unit or a non-transplant strategy. The challenge remains how to reliably predict the individual mismatches for all populations. Therefore, to improve the calculations of predictions, WMDA is working on methods to validate haplo frequencies from all populations.