Abstract 3239

Red blood cell (RBC) transfusion is a key treatment for patients with sickle cell disease (SCD), but remains complicated by the high incidence of RBC immunization. Despite provision of phenotypic Rh D, C, and E antigen-matched donor RBCs, patients continue to develop Rh antibodies. In many cases, these antibodies are considered autoantibodies with specificities for D, C, and e antigens because the patient's own RBCs type serologically positive for the corresponding antigen. Recent evidence is emerging that Rh alloimmunization within populations of African origin is complicated by the genetic diversity of this locus. Individuals of African ancestry often carry RH alleles that differ from those defined as conventional alleles that are common in Europeans and other ethnic groups. These “variant” alleles encode Rh proteins, often with multiple amino acid changes, that either lack common epitopes or produce novel immunogenic epitopes. The clinical significance of Rh alloimmunization in patients with SCD with variant RH genes is largely unknown. In the present study, we performed RH genotyping in 212 patients with SCD to determine prevalence of RH variants, the association with Rh alloimmunization, and the clinical significance measured by changes in hematologic parameters at time of antibody detection. RH genotyping was performed by polymerase chain reaction (PCR) amplification using RHD-specific and RHCE-specific primers designed in the flanking intronic regions and analyzed by direct sequencing of exons, and/or a combination of multiple PCR-restriction fragment length polymorphism (RFLP) assays and by RHD and RHCE BeadChip arrays. We identified variant RH alleles in 88.7% (188/212) of patients with SCD. Twelve different RHD and 13 RHCE alleles encoding variant Rh D, C, and e antigens were represented in this cohort of patients. In 172 patients with >1 RBC transfusion (median 125 units), 55 antibodies were identified with Rh specificity despite antigen-positive status (28 anti-D, 16 anti-e, 9 anti-C, and 2 anti-E). RH genotypes revealed 47.3% of these antibodies developed in patients who lacked the corresponding conventional RH allele, and would be considered Rh alloantibodies. In 43.6%, RH genotypes predicted expression of the conventional antigen to which the antibody was directed, suggesting these were potentially autoantibodies. However, these patients had at least one other RH allele that was altered. This may suggest that the presence of one variant protein potentially changes the Rh complex in the membrane, and carrying at least one conventional RH allele is not necessarily protective against production of Rh specific antibodies. In the remaining 9.1%, we detected no variant RHD or RHCE alleles, and complete gene sequencing is in progress to confirm the absence of novel mutations. Importantly, to determine clinical significance, we evaluated whether Rh antibodies in patients carrying variant alleles caused decreased transfused RBC survival by comparing the patient's hematologic parameters at the time of antibody detection with the baseline pre-transfusion parameters. In all but 4 cases, Rh antibodies in antigen-positive patients occurred in chronically transfused patients. Therefore, baseline values were determined from the average pre-transfusion hemoglobin, hematocrit and % hemoglobin S level for the 6–12 months preceding antibody detection. Compromised transfused RBC survival was determined by a lower hemoglobin/hematocrit or higher % hemoglobin S level at time of antibody formation compared to the patient's baseline. Forty percent of antibodies were associated with delayed hemolytic transfusion reactions or decreased transfused RBC survival. Our data suggest that the high prevalence of variant RH alleles in patients with SCD is associated with clinically significant immunization. Discrimination of allo- versus auto- antibodies in this chronically transfused patient population presents a significant technical challenge and suggests a role for RH genotyping in the clinical evaluation of Rh antibodies and to improve RBC matching. Importantly, in this study Rh antibodies in patients with variant RH often compromised transfused RBC survival and, therefore, were clinically significant and may be targets for prevention strategies analogous to standard phenotype matching for C, E, and K.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract