ABO blood group incompatibility as an adverse risk factor for outcomes in patients with myelodysplastic syndromes and acute myeloid leukemia undergoing HLA-matched peripheral blood hematopoietic cell transplantation after reduced-intensity conditioning

BACKGROUND: ABO incompatibility is not a contraindication to hematopoietic cell transplantation (HCT), but it has been associated with additional risks including delayed engraftment, pure red cell aplasia (PRCA), and higher transfusion needs. Data on these events and on patient survival after reduced-intensity conditioning (RIC) HCT are limited. STUDY DESIGN AND METHODS: A total of 127 consecutive patients, 86 with acute myeloid leukemia and 41 with myelodysplastic syndromes, who underwent HLA-matched peripheral blood RIC allogenic HCT between 2005 and 2014 were retrospectively analyzed. RESULTS: Eighty ABO-compatible, 26 major/bidirectional, and 21 minor-ABO-mismatch HCT were identified. Compared to the ABO-compatible group, major/bidirectional mismatches had increased red blood cell (RBC) transfusion requirement during the first 100 days (p = 0.009), delayed RBC and PLT engraftment (p = 0.0011 and p = 0.005, respectively), and higher incidence of grade II to IV acute graft-versus-host disease (aGVHD; p = 0.037). In multivariable analysis, major/bidirectional mismatches had significantly higher non-relapse mortality (NRM) and inferior disease-free survival (DFS) and overall survival (OS) compared with ABO-compatible patients (p = 0.01, p = 0.04, and p = 0.035, respectively). Minor ABO mismatch had no impact on survival (p = 0.99). Four (15%) of 26 major/bidirectional mismatches developed PRCA. There was a significant association between fludarabine plus busulfan conditioning and PRCA (p = 0.0046). CONCLUSION: Major/bidirectional ABO mismatch is associated with higher NRM and shortened DFS and OS in the setting of RIC HCT. Increased transfusion need, delayed RBC and platelet engraftment, PRCA, and increased severity of aGVHD are additional complications contributing to the morbidity.