7/8 High-Resolution Unrelated Donor HLA Match Rate: Caucasian, African American, Hispanic, and Asian-Pacific Islander

Estimation of the National Marrow Donor Program's Be The Match Registry (BTMR) 8/8 (HLA-A, B, C, DRB1) high resolution (HR) unrelated donor (URD) match rate was determined in a 2009 study for each of the four most frequent patient race/ethnic groups in the United States- Caucasian (CAU), African American (AFA), Hispanic (HIS), and Asian Pacific Islander (API) (Dehn et al, ASBMT/CIBMTR abstract 2011). For patients without an 8/8 matched URD, a 7/8 match is often the minimum acceptable mismatch used by many transplant centers. A follow up to the 8/8 study was designed to determine the 7/8 or better match rate among the 4 major race/ethnic groups, using the same study cohort.

1344 previously high resolution tested URD in the BTMR were randomly selected and treated as pseudopatients (PP) where HR testing was performed to identify an 8/8 matched URD. Following 8/8 potential URD testing, a search was performed on each PP to determine the 7/8 match rate, regardless of whether an 8/8 match was previously identified. The searches used a fixed BTMR file from January of 2012, composed of over 8.6 million URD. The BTMR race/ethnic diversity breakdown for 2012 was 65% CAU, 7% AFA, 10% HIS, 7% API, and 11% miscellaneous categories such as Multiple, Unknown, American Indian- Alaska Native, and Declined to answer. Search results from CAU (N=377), AFA (N=390), HIS (N=307), and API (N=270) PP were evaluated and classified as follow:

1) 7/8 HR matched donor exists on BTMR

2) Potential 7/8 HR donors exist on BTMR

3) No 7/8 potential donors exist on BTMR

PP searches falling into category 2 had an HLA search strategy expert rank potential URD within BTMR in order of their matching likelihood. URD samples were HR HLA tested in order of ranking and evaluated to determine match status. Consecutive rounds of URD sample testing were performed until either a 7/8 matched URD was identified, no potential URD with stored samples remained, or a patient maximum number of URD testing was reached.

Since many PP searches had greater than 100 potential 7/8 matched URD, it was not possible to type every URD. In these cases, up to 35 URD with sample were tested per patient. For analysis of the 7/8 match rate, PP with no 7/8 match identified after URD testing were considered as having no HR match.

98% of CAU and over 80% of the non-CAU race/ethnic groups- AFA, HIS, and API- had at least a 7/8 match identified (Table 1). For cases where an 8/8 matched donor had been previously identified, all but 8 PP also had a 7/8 match (CAU= 1, AFA= 3, API= 3, HIS= 1). Only three PP cases had no 7/8 potential donors on the search prior to URD testing. The range of donors tested for the cases resulting in a match was larger for the non-CAU race groups. The maximum number of URD tested before a 7/8 match was identified was 24 for AFA, 6 for HIS, and 10 for API, while for CAU the maximum number of URD tested was 2 (Table 2). A median of 1 URD was typed for cases resulting in a 7/8 match for each of the four race/ethnic groups. HLA expert review of cases where no 7/8 match was identified but additional 7/8 potential URD remained suggests that few additional cases would likely yield HR matches.

This study estimates a 7/8 or better HR match rate for the BTMR of over 80% for all four broad race/ethnic group categories. These results show that in the majority of cases, after first testing to identify an 8/8 matched URD, a 7/8 matched URD was identified after typing just one URD. However, particularly with non-CAU patients, testing additional URD may be needed to identify a 7/8 match. This study provides a baseline match rate that can be further supplemented using the additional worldwide URD inventory.

No relevant conflicts of interest to declare.