Optimizing Hydroxyurea Therapy with Reduced Laboratory Monitoring for Children with Sickle Cell Anemia in Sub-Saharan Africa: The Reach Experience

Introduction: Realizing Effectiveness Across Continents with Hydroxyurea (REACH, NCT01966731) is an open-label study of hydroxyurea for children with sickle cell anemia (SCA) in sub-Saharan Africa (Angola, DR Congo, Kenya, and Uganda). Initial results documented the feasibility, safety, and benefits of hydroxyurea for SCA in sub-Saharan Africa but guidance for optimizing hydroxyurea therapy is needed. We describe 5 years of hydroxyurea dosing and monitoring in the largest prospective cohort of children with SCA receiving hydroxyurea to date.

Methods: Children 1-10 years of age with SCA were enrolled. The hydroxyurea dose was fixed at 15-20 mg/kg/day for the first 6 months with monthly complete blood counts (CBCs) to ensure safety. From month 6-24, the dose was escalated (5 mg/kg every 8 weeks) to maximum tolerated dose (MTD), defined as mild myelosuppression with absolute neutrophil count (ANC) <4.0 x 10⁹/L on 2 consecutive CBCs without hematological toxicities. CBCs were performed monthly until MTD or a stable dose was achieved, then subsequently every 3 months. Dose limiting toxicities (DLT) requiring a temporary treatment hold were defined as ANC <1.0 x 10⁹/L, Hb <4.0 g/dL, reticulocyte count <80 x 10⁹/L unless Hb >7.0 g/dL, or platelets <80 x 10⁹/L. Known genetic modifiers of SCA, including G6PD deficiency and α-thalassemia trait, were determined for all participants.

Results: A total of 606 children initiated hydroxyurea and currently 555 (92%) remain on treatment, with average treatment duration of 48 ± 12 months and a total of 2,441 patient-years of hydroxyurea treatment. Over 85% achieved MTD with an average hydroxyurea dose of 22.5 ± 5.0 mg/kg/day, ranging from 19.0 mg/kg/day in Angola to 25.4 mg/kg/day in Uganda. With dose increases over time, the most recent average dose is 23.9 ± 5.4 mg/kg/day (site range 22.9-24.6 mg/kg/day). Lab benefits have been sustained; Hb increased from 7.3 g/dL at baseline to 8.4 g/dL at MTD and remains 8.3 g/dL at Month 60. Similarly, the average HbF increased from 11% baseline to 25% at MTD and remains 23% at Month 60. The average ANC decreased from 6.8 x 10⁹/L at baseline to 3.2 x 10⁹/L at MTD and remains 3.5 x 10⁹/L at Month 60. Lab toxicities are infrequent, transient, and mostly incidental. Of 19,730 CBCs obtained during the treatment phase, 421 (2.1%) in 225 participants included a DLT. The most common DLT was thrombocytopenia (33%), with only 4 platelet counts <20 x 10⁹/L and no bleeding. Anemia was the second most common DLT (26%), most commonly associated with a high reticulocyte count and malarial infection, unrelated to hydroxyurea. Severe neutropenia (ANC <0.5 x 10⁹/L) was rare (5 events) with no neutropenic infections. Over 2/3 of DLT events were identified incidentally during a scheduled visit when the study participant was asymptomatic, including all 5 severe neutropenic episodes. Weight-for-age and height-for-age Z-scores were not associated with higher rates of DLT during hydroxyurea treatment. Children with two-gene deletional α-thalassemia trait tolerated significantly lower hydroxyurea doses than the normal genotype (MTD dose 20.0 vs. 24.0 mg/kg/day, p <0.001) and had significantly different treatment responses at Month 60 including lower HbF (19.5 vs 24.3%, p <0.0001) and MCV (72 vs 99 fL, p<0.001) but higher hemoglobin (8.5 vs 8.1 g/dL, p=0.016). DLT frequency was unaffected by α-thalassemia status. Males with G6PD A^- deficiency did not demonstrate significant differences in dosing, response, or toxicity.

Conclusions: Hydroxyurea is safe, well-tolerated, and effective for children with SCA living in sub-Saharan Africa. Treatment responses are robust and sustained in REACH across all 4 clinical sites and unaffected by baseline Z-score. Hydroxyurea optimization requires periodic dose escalation for weight gain and titration to mild myelosuppression. Deletional α-thalassemia trait significantly influences the hydroxyurea dose and treatment responses, but the lab benefits with optimized dosing are still robust regardless of the α-globin genotype. Lab toxicities from hydroxyurea are uncommon and typically asymptomatic, suggesting that routine CBC monitoring is needed only at 3-month intervals once a stable dose is achieved, more to optimize the dose than to identify incidental toxicities. This approach to optimizing hydroxyurea therapy will allow more widespread utilization in low-resource settings with limited laboratory monitoring.