Flowchart of patients included in the study. Our source study population comprised transfused SCD patients from a cohort from France (FR; N = 325), and a cohort from the Netherlands (NL; N = 342). Of these, 310 patients were eligible for this study (DNA sample available, alloimmunized cases or controls with ≥20 units exposure). An additional 38 cases were excluded because they exclusively had either autoantibodies or naturally occurring antibodies (patients excluded: FR N = 29, NL N = 9). The remaining 272 patients (130 controls and 142 alloimmunized cases) formed our primary study population. We performed 2 sensitivity analyses: (I) excluding controls with <100 units of transfusion exposure (patients excluded: FR N = 36, NL N = 33) and (II) dividing cases into patients who had only antibodies with Rh or K specificity and patients with at least 1 antibody other than Rh or K. A separate analysis was performed to assess the association of FCGR polymorphisms with the occurrence of DHTR. Data on the history of DHTR status was available only in a subset of the FR cohort (N = 157; 28 patients were excluded because of unknown DHTR status, and 12 controls were excluded because they had <20 units of exposure). *Excluded other: because there was a maximum number of samples of the French cohort agreed upon to be shipped, a small fraction of patients from the French source population were excluded on the basis of their chronological entry in the dataset.