Blood cell counts and the duffy null phenotype: Beyond neutropenia

Abstract Background: The absence of Duffy antigen expression on red blood cells, known as the Duffy null phenotype [Fy(a-b-)], results from a single nucleotide polymorphism in the DARC/ACKR1 gene and provides selective protection against malaria. This phenotype is present in over 60% of the African American population and is associated with lower neutrophil counts. Recently, reference intervals for neutrophil counts have been proposed specifically for Duffy-null individuals with African ancestry. In a prior study in Brazil, the Duffy null phenotype proved to be more accurate than self-reported race in identifying benign ethnic neutropenia. However, hematological reference ranges specific to Duffy phenotypes have not been systematically studied in admixed populations. This study aimed to establish reference intervals for absolute neutrophil counts based on Duffy phenotyping and to investigate other hematologic parameters potentially influenced by the Duffy null status. Methods: We conducted a prospective study involving 311 healthy adult blood donors from two independent blood banks in São Paulo, Brazil. Participants were ≥18 years old and met all eligibility criteria for blood donation. EDTA blood samples were collected for complete blood counts and Duffy phenotyping. CBCs were processed using harmonized Sysmex XN series analyzers to ensure standardized and comparable results. Duffy phenotyping was performed by indirect antiglobulin test (IAT) using gel cards (Grifols S.A.) with polyclonal anti-Fya and anti-Fyb reagents (Grifols S.A.), following the manufacturer's protocol. Individuals were categorized into two groups: Duffy null [Fy(a-b-)] and Duffy non-null [Fy(a+b+), Fy(a+b-), or Fy(a-b+)]. Statistical comparisons were performed using chi-square, likelihood ratio, and unpaired Student's t-tests. The study was approved by both institutional review boards and written informed consent was obtained from all participants. Results: Among the 311 participants, 51.4% were male, with a median age of 37 years (range 18–69); 10.6% reported a history of chronic disease, and 23.6% were on chronic medications. Self-identified Black or mixes individuals accounted for 46.6% of the cohort. A total of 146 individuals (46.9%) had the Duffy null phenotype. This group had a significantly higher proportion of females (54.8% vs. 45.2%, p = 0.038), higher prevalence of Black ethnicity (74% vs. 22.4%, p < 0.001), and lower use of chronic medications (6.8% vs. 38.2%, p < 0.001). Several hematologic differences were observed. In the white blood cell lineage, the Duffy null group had significantly lower median total leukocyte counts (5,678 vs. 7,366/uL; p < 0.001) and absolute neutrophil counts (median 2,638/uL; IQR: 547–4,728; range: 780–6,540/uL vs. 4,368/uL; IQR: 903–7,834; range: 1,610–15,460/uL; p < 0.001). Neutrophil percentages were also reduced (45.5% vs. 58.1%; p < 0.001), while relative lymphocytes (40.7% vs. 31.2%; p < 0.001), monocytes (9.3% vs. 7.7%; p < 0.001), eosinophils (3.1% vs. 2.4%; p = 0.002), and basophils (1.3% vs. 0.6%; p < 0.001) were increased. Absolute lymphocyte counts did not differ significantly (2,165 vs. 2,220/uL; p = 0.68). In the red cell lineage, Fy(a-b-) individuals had higher erythrocyte counts (5.15 vs. 4.90 109/uL; p < 0.001), hemoglobin (14.8 vs. 14.3 g/dL; p = 0.002), and hematocrit (45.4% vs. 42.8%; p < 0.001), with differences persisting across both genders. Platelet parameters also differed significantly: the Duffy null group had higher platelet counts (286.6 vs. 247.2 × 10³/uL; p < 0.001) and mean platelet volume (11.6 vs. 10.4 fL; p < 0.001). Conclusion: Duffy null individuals exhibit not only lower neutrophil counts but also significant alterations across multiple hematologic parameters. These findings support the establishment of Duffy-specific reference intervals to prevent misclassification and avoid unnecessary clinical investigations. This is particularly relevant in genetically admixed populations, such as Brazil, where race-based classifications may be inadequate for accurate medical interpretation.