An epitope-based approach of HLA-matched platelets for transfusion: a noninferiority crossover randomized trial

Abstract Platelet transfusion refractoriness results in adverse outcomes and increased health care costs. Managing resulting from HLA alloimmunization necessitates the use of antigen–matched platelets but requires a large platelet donor pool does not guarantee full matching. We report first randomized, double-blind, noninferiority, crossover trial comparing epitope–matched (HEM) with standard (HSM) transfusions. Alloimmunized, platelet-refractory, thrombocytopenic patients aplastic anemia, myelodysplastic syndrome, or acute myeloid leukemia were eligible. HEM selected using HLAMatchMaker epitope (specifically eplet) Patients received up to 8 prophylactic HSM transfusions provided random order. The primary outcome was 1-hour posttransfusion count increment (PCI). Forty-nine randomized at 14 UK hospitals. For intention treat, numbers evaluable 107 112 for methods, respectively. Unadjusted mean PCIs methods 23.9 (standard deviation [SD], 15) 23.5 (SD, 14.1), respectively (adjusted difference, ?0.1; 95% confidence interval [CI], ?2.9 2.8). Because lower limit CI greater than predefined noninferiority limit, approach declared noninferior approach. There no differences secondary counts, requirements, bleeding events. Adequate PCI more frequently observed, number 3.2 mismatches, compared 5.5 mismatches inadequate increments. every additional mismatch, likelihood an adequate decreased by 15%. Epitope-matched should be considered support alloimmunized patients. This registered www.isrctn.com as #ISRCTN23996532.