The impact of duffy-null phenotype on neutrophil counts in patients with HbSC/ HbSβ+-thalassemia treated with hydroxyurea Free

Introduction: The Duffy-null phenotype is characterized by the lack of Duffy antigen expression on red cells and is associated with lower absolute neutrophil counts (ANCs) in individuals of African ancestry. We previously reported that the Duffy-null phenotype is associated with lower baseline ANCs, and higher frequency of hydroxyurea (HU)-associated neutropenia in pediatric patients with sickle cell disease (SCD). Of interest, the impact of the Duffy-null phenotype on ANCs varies with SCD genotype, and patients with HbSC/HbSβ⁺-thalassemia have a more pronounced ANC reduction than those with HbSS/HbSβ⁰-thalassemia. HbSC and HbSβ⁺ are common forms of SCD with milder clinical phenotypes compared to HbSβ⁰-thalassemia, however individuals remain at risk for complications such as vaso-occlusive crisis, chronic organ damage and shortened life expectancy. HU has been shown to be effective in reducing SCD-related complications in both children and adults with HbSC whereas HU-indued cytopenia occurs in 20-30% of the patients. Since Duffy-null patients with HbSC/HbSβ⁺ had markedly lower ANCs than Duffy-positive patients, we investigated the impact of the Duffy-null phenotype on ANCs and HU treatment and toxicities in patients with HbSC/HbSβ⁺while receiving HU.

Methods:The inclusion criteria included pediatric patients with HbSC or HbSβ⁺ who 1) were enrolled in the St. Jude Children's Research Hospital observational cohort study, the Sickle Cell Clinical Research and Intervention Program (SCCRIP; NCT02098863); 2) had available Duffy genotype; 3) reached maximal tolerated dose (MTD) of HU before 12/31/2023. MTD was considered achieved at a dose when ANC was between 1000 and 3000 /mm³. Baseline ANCs were defined as values collected 30 days from an acute care visit or hospitalization. To examine the impact of the Duffy-null phenotype on patients receiving HU, patients who started HU between ages 2 and 16 years were included. Data was analyzed every two years up to 8 years after HU initiation. The ANC closest to the end of each treatment period was used. The Wilcoxon rank-sum test or two-sample t-test was used for continuous variables, Fisher's exact test for categorical variables. Data normality was checked using Shapiro-Wilk test. Generalized linear mixed models with Gaussian or binomial link functions were used to assess the associations between Duffy status and log transformed ANC levels and risk of ANC <1500/mm³, adjusting for relevant covariates.

Results: A total of 24 patients with HbSC/HbSβ⁺ met the inclusion criteria, and 16 (66.7%) patients had the Duffy null phenotype. Female patients accounted for 45.8% (n= 11). Duffy-null patients had a significantly lower MTD than Duffy-positive patients (18.74 ± 6.14 vs. 25.04 ± 4.72 mg/kg/day, p=0.012), and tended to start HU at an older age (10.26 ± 4.4 vs. 7.71 ± 2.6 years old, p=0.09). The time to MTD was similar for the two groups (Duffy-null: 1.40 ± 2.13 vs. Duffy-positive:1.45 ± 1.01 years, p=0.31). The Duffy null phenotype was associated with a significantly lower baseline ANCs before HU initiation (3560 ± 1570 /mm³ vs. 6188 ± 4239 /mm³; p = 0.021), and with HU treatment (overall p = 0.004 for the total of 8 years of treatment). When each treatment interval was examined individually, the Duffy-null associated reduction was significant with 4-6 years of HU treatment (2904 ± 894 /mm³ vs. 5561 ± 5020 /mm³; p = 0.0081). Overall, Duffy-null patients were more likely to develop neutropenia while receiving HU than Duffy-positive patients (overall p = 0.027 for the total of 8 years of treatment). When each treatment period was examined individually, the percentage of patients with ANC <1500 /mm³ was significantly higher for Duffy-null than those of Duffy-positive patients for 2-4 years of HU treatment (50% vs. 0%; p = 0.022). Notably, MCV and HbF increased significantly with HU, and the increases were similar between the two groups (MCV: Duffy-null 12.58 ± 9.39 fL vs. Duffy-positive 16.68 ± 8.32 fL, p= 0.29; HbF: Duffy-null 3.25 ± 3.14% vs. Duffy-positive 5.06 ± 4.30, p= 0.25), suggesting comparable HU adherence.

Conclusions: Duffy-null patients with HbSC/HbSβ⁺ had lower ANCs and were more likely to experience HU-induced neutropenia than Duffy-positive patients despite receiving lower dosage of HU. Larger studies are warranted to determine the impact of Duffy status and HU management on clinical outcomes.