Red Cell Antigen Genotyping Compared to Standard Serological Phenotyping in Sickle Cell Disease Patients in Canada: Potential for Reducing Alloimmunization

Introduction:

Red blood cell (RBC) transfusion is the cornerstone of management in many patients with sickle cell disease (SCD). However, RBC transfusion can be complicated by alloimmunization and hemolytic transfusion reactions in this population despite providing extended phenotype-matched RBC transfusions. This is due to heterogeneity of RBC antigens, unique variant mutations in this population, and genetic mismatch between the blood donor pool and SCD patients in North American settings. In this study, we evaluated the level of discrepancy between RBC antigen genotyping and traditional phenotyping methods and the association of these discrepancies with the presence of RBC alloantibodies in SCD patients at our centre in Canada.

Methods:

Commencing in January 2015, RBC antigen genotyping has been included in the care for patients with SCD treated at our Hemoglobinopathy Clinic in an academic medical centre. Patient blood samples are sent to a reference laboratory to perform genotyping of RhCE, Kell, Kidd, Duffy, and S antigens. RBC antigen phenotyping was performed locally using both tube and automated solid phase assays. Additional clinical data, demographic and transfusion-related data were obtained from a local transfusion registry databse and thorough clinical chart reviews. Approval from our centre's research ethics board was obtained prior to commencement of data collection.

Results:

To date, RBC antigen genotyping has been performed on 45/88 SCD patients treated at our centre. The mean age of these patients was 25, and 58% were female. The majority of patients had HbSS SCD genotype (64.4%), or HbSC (26.7%).

Overall, 32/45 (71%) of patients had variant mutations detected by genotyping, including 9 (20%) patients with more than one variant mutation. The most common mutation detected was the GATA mutation (n= 23; 51%) resulting in loss of Fy^b antigen expression on RBCs, but associated with expression of Fy^b on non-erythroid tissues. The RhCE system showed variant mutations resulting in partial expression of antigens in 9 (20%) patients. Alloantibodies were found in 9/36 (25%) patients with either a GATA mutation or no variant mutations. Alloantibodies were found in 2/9 (22.2%) patients with mutations resulting in partial antigen expression. The proportion of patients with any discrepancy between genotyping and phenotyping was 34/45 (75.6%). The largest rates of discordance were seen in the RhCE system, with the c antigen having a kappa of 0.68 and e antigen having a kappa of 0.32 (Table 1).

Conclusion:

Our results showed a high prevalence of variant mutations and significant discrepancies between genotyping and phenotyping methods, most notably in the RhCE antigen system. Mutations resulting in partial antigen expression were associated with development of alloantibodies in 22.2% of patients in our study, which may have been prevented with a genotype-based antigen-matching strategy. Additionally, knowledge of presence of GATA mutation will enhance feasibility of antigen matching for affected patients, who may have otherwise required RBC units negative for Fy^b based on local policies.

To our knowledge these results represent the first published data from a Canadian centre, showing similar rates of discrepancy between traditional phenotyping methods and RBC antigen genotyping as reported in other regions. Although phenotype-based matching strategies are used in many centres, these strategies can place patients with partial RBC antigen variant mutations at a direct increased risk of alloimmunization. Thus genotype-based antigen-matching strategies should be considered for transfusion of matched RBCs in patients with SCD.