Erythrocyte Alloimmunization Analyzed by Gender and Transfusion Status

Abstract 4407

(Purpose) Prior investigations have looked for trends and patterns in post-transfusion erythrocyte alloantibody formation (e.g., Walker et al., Tormey et al., and Hoeltge et al.). While anti-E has consistently been one of the most frequently observed alloantibodies following transfusion, the relative frequency of other antibodies is variably reported. In fact, some authors have reported no relationship between transfusion and alloantibody prevalence (Coles et al. and Domen et al.). Different outcomes might be attributed to antibody detection methods, cohort size, and cohort composition by race (including differences in the racial composition of donor and patient populations). As yet, the significance of gender has not been explicitly investigated in a large cohort of transfused vs. non-transfused patients. This study reports gender differences in erythrocyte alloantibody formation in a nation where blood donor and patient populations are phenotypically similar and relatively homogeneous. (Methods) In response to an open invitation made in November, 2008, 25 Japanese institutions contributed data on 248,785 patients (male:female=1:1.19). Each institution reported results from the previous 3 years. Patients known to have been transfused were placed in the transfused group, whereas patients known not to have been transfused were placed in the non-transfused group. Patients for whom transfusion history was uncertain were excluded. Erythrocyte alloantibodies against D, C, c, E, e, f, Ce, P₁, M, N, S, s, Mi^a, Le^a, Le^b, Jk^a, Jk^b, Jk3, Fy^a, Fy^b, K, k, Kp^a, Kp^b, Js^a, Js^b, Di^a, Di^b, Lu^a, Lu^b, Xg^a and H were analyzed. A patient investigated multiple times was counted as one case. Multiple antibodies in one patient were separately summarized. Antibody incidence was calculated as the percentage of patients in each group with the corresponding antibody. (Results) Erythrocyte antibodies were detected in 3,554 patients (1.43% of all patients, 1.32% of males and 1.52% of females). Including 655 patients with antibodies reported by four more institutions, a total of 4,219 patients with at least one antibody were analyzed. From these, anti-E was observed in 34.7% and 41.3% of male and female transfused patients vs. 8.0% and 21.9% in those not transfused, respectively. Anti-Di^a was observed in 3.5% and 4.8% of male and female transfused patients vs. 2.7% and 2.8% in those not transfused. Jk^a was observed in 2.6% and 4.7% of male and female transfused patients vs. 0.4% and 0.1% in those not transfused. Anti-E+c was observed in 6.6% and 8.4% of male and female transfused patients vs. 0.4% and 2.3% in those not transfused. Emergence of anti-E, anti-Di^a and anti-Jk^a following transfusion was significantly different between males and females (p<0.05, p<0.05 and p<0.01, respectively). (Conclusion) The frequencies of anti-E, anti-Di^a, anti-Jk^a and others depend not only on transfusion history, but also gender. Discordance with other studies suggests that population genetics may also be relevant. These are especially important results in the emerging era of personalized medicine. Hematologists should continue large-scale investigations of transfusion-related alloimmunization to elucidate contributory factors and to optimize patient care.

No relevant conflicts of interest to declare.