Effect of ABO Mismatch and Red Blood Cell Alloimmunization on the Outcome of Hematopoietic Cell Transplantation for Sickle Cell Disease

Abstract

Hematopoietic stem cell transplantation (HCT) remains the sole well-established curative option for patients with sickle cell disease (SCD). Nonmyeloablative conditioning has been used to improve outcomes and reduce toxicities in adult SCD patients. However, recipient-donor ABO incompatibility and alloimmunization may be significant impediments to successful outcomes of HSCT in SCD patients owing to risks of hemolysis, delayed engraftment, poor graft function, and graft failure (GF). Here we report our experience with allogeneic HCT for SCD and the effects of RBC group mismatch and alloimmunization on the outcome of transplant recipients. We conducted a retrospective analysis of all SCD patients age >14 years who underwent HCT between January 2015 and February 2022 at our center. All patients received i.v. alemtuzumab (1 mg/kg divided over 5 days on days -7 to -3) and 300 cGy total body irradiation on day -2 or -1 for conditioning. Pretransplantation preparation consisted of hydroxyurea at the maximum tolerated dose for 2 to 3 months and 1 session of exchange transfusion. Peripherally mobilized CD34 stem cells targeting 10 × 10⁶ /kg of recipient weight were used. For graft-versus-host disease prophylaxis, sirolimus was started on day -1 and continued for 1 year with tapering if lymphoid chimerism was >50%. For patients with major ABO incompatibility, we administered 2 doses of rituximab (375 mg/m²) and 3 sessions of plasmapheresis before starting the conditioning regimen, targeting an isohemagglutinin titer <1/32. The primary objective was to determine the impact of RBC group mismatch and alloimmunization on the outcomes of the HCT recipients. The secondary objective was to assess the impact of GF on overall survival (OS). Logistic regression was done to evaluate predictors for GF. The Kaplan-Meier method was used for survival analysis. A total of 194 patients were included, with a median age of 26 years. Their median baseline hemoglobin and hemoglobin S values were 93 g/L and 71.3%, respectively. Indications for HCT included recurrent vaso-occlusive crisis in 52.5% of the patients, central nervous system events in 19.6%, and acute chest syndrome in 17%. After a median follow-up of 28.8 months (range, 5 to 83 months), 16 patients (8%) experienced GF (3 with primary GF and 13 with secondary GF). On univariate analysis, ABO minor incompatibility and RBC alloantibodies against donor non-ABO antigens were predictive of GF. On multivariate analysis, RBC alloantibodies against donor non-ABO antigens (odds ratio, 8.29; 95% confidence interval, 2.01 to 34.05; P = .0033) was the sole factor predictive of GF. None of the 16 patients with major ABO incompatibility developed GF. The 2-year OS for all patients was 95%. The 2-year OS was 98% in patients without GF, compared to 74% in patients with GF (P < .0001). In our SCD patients who underwent HSCT, the presence of RBC alloantibodies against donor non-ABO antigens was an independent risk factor for GF. Patients with GF had inferior survival, and strategies for decreasing the risk of GF are needed.