The effect of single or multiple mismatches at each HLA locus on outcomes after cord blood transplantation (CBT) is controversial. We analyzed the effects of single or multiple HLA locus mismatches on the outcomes after single CBT using Japanese registry data from the Japan Society for Hematopoietic Cell Transplantation. Patients age ≥16 years with acute leukemia and myelodysplastic syndromes who underwent their first CBT between 2003 and 2017 (n = 4074) were included. The effect of the number of HLA locus mismatches (0, 1, or 2 for the HLA-A, -B, -C, and -DRB1 loci) on outcomes was analyzed after adjusting for other significant variables. The patient cohort had a median age of 54 years. The median total nucleated and CD34 cell doses were 2.6 × 107/kg and .8 × 105/kg, respectively. The number of CBTs with single or double mismatches were 2099 and 292, respectively, for the HLA-A locus, 2699 and 341 for the HLA-B locus, 2555 and 609 for the HLA-C locus, and 2593 and 571 for the HLA-DRB1 locus. Single and double HLA-DRB1 mismatches were associated with a higher risk of grade II-IV acute graft-versus-host disease (GVHD; single: hazard ratio [HR], 1.29, P < .001; double: HR, 1.49, P < .001; P for trend <.001). Single and double mismatches at HLA-DRB1 as well as single mismatches at HLA-A and HLA-B also were associated with grade III-IV acute GVHD. Single and double HLA-B mismatches and double HLA-DRB1 mismatches were associated with a high risk of nonrelapse mortality (NRM). On the other hand, double mismatches at HLA-A or HLA-DRB1 and single mismatches at HLA-B were associated with a lower risk of relapse. HLA-DRB1 double mismatch was associated with high risk of grade II-IV and grade III-IV acute GVHD and NRM but lower risk of relapse. Not only the locus mismatch, but also the number of mismatches, should be considered in cord blood unit selection.
!!!All Science Journal Classification (ASJC) codes