Pharmacokinetic (PK)-guided vs weight-based dosing of hydroxyurea for tanzanian children with sickle cell anemia

Abstract Introduction: Hydroxyurea titrated to achieve mild myelosuppression affords the greatest benefits for children with sickle cell anemia (SCA), but dose adjustments can be challenging to achieve in regions where laboratory monitoring is less available and undetected drug toxicities might be more consequential. Personalized pharmacokinetic (PK)-guided hydroxyurea dosing facilitates a higher initial starting dose and a shorter titration period. We designed the Stroke Prevention with Hydroxyurea Enabled through Research and Education (SPHERE, NCT03948867) Extension trial to explore the feasibility, safety, and benefits of PK-guided hydroxyurea dosing for children with SCA in a low-resource setting in northwest Tanzania. Methods: SPHERE is a prospective, phase 2 trial of open-label hydroxyurea in Tanzanian children, 2-16 years old, with SCA. During the first phase of SPHERE, children with elevated (>170 cm/sec) transcranial Doppler (TCD) velocities were treated with hydroxyurea at 20 mg/kg/day and dose escalated every 2 months to achieve mild myelosuppression; children with normal TCD velocities were placed into an Observation Arm. During the PK Extension phase, children in the Observation Arm started PK-guided hydroxyurea treatment. During screening, a single 500mg hydroxyurea oral test dose was administered, and venous blood was collected 15, 30, 60, and 180 minutes later. Hydroxyurea serum concentrations were measured using a diacetylmonoxime detection method using a portable high performance liquid chromatography machine (SmartLife, PolyLC), and then analyzed with the HdxSimTM PK calculator to generate a daily dose that would achieve an area under the curve of 115 mg*h/L. If the PK process failed or the calculator predicted a dose <15 or >35 mg/kg/day, a 25 mg/kg/day default dose was initiated. Doses were subsequently escalated to maximum tolerated dose (MTD) and continuously optimized to maintain mild myelosuppression with an absolute neutrophil count <3.0 x 109/L. Results: Between May 2023 and April 2024, 100 children (56 female, 44 male) were enrolled at a mean age of 11.4±3.8 years of age. All 100 children had PK testing attempted using an average test dose of 20.3± 6.2 mg/kg (range 9.6-41.7 mg/kg/day). In 71 participants, an average PK-guided dose of 24.4±4.7 mg/kg/day (range 13.3-34.0) was started. In 29 participants, the PK process failed due to problems with lab testing (N=12), or the generated dose was too low (N=2) or high (N=15); these children began a weight-based dose of 25.0±2.1 mg/kg/day (range 18.8-30.1). Laboratory values were similar in those who started the PK-guided dose versus those who started a weight-based dose including baseline hemoglobin (Hb), mean corpuscular volume (MCV), fetal hemoglobin (HbF), white blood cell count (WBC), platelets, absolute neutrophil count (ANC), and absolute reticulocyte count (ARC). After at least 12 months of treatment with protocol-guided dose escalation, the PK-guided cohort completed 86.7 patient-years of treatment with a final mean dose of 29.0±4.5 mg/kg/day (range 17.6-34.9), while the weight-based cohort completed 43.0 patient-years with a final dose of 29.7±3.9 mg/kg/day (range 21.5-34.6). Laboratory benefits were not statistically different in the two groups: Hb = 9.3±1.1 versus 9.3±1.4 g/dL; MCV = 98±12 versus 99±12 fL; HbF = 22.9±8.6 versus 22.3±9.1%; ANC = 3.6±2.4 versus 3.5±2.0 x 109/L. Dose-limiting toxicities (DLT) were also similar with 0.15 per patient-year in the PK-guided dosing cohort and 0.19 per patient year in the weight-based dosing cohort (IRR 0.74, 95% CI 0.31-1.8, p=0.51). The incidence of any clinical adverse events (AE) was slightly lower, especially in the first 6 months of treatment (0.94 versus 1.19 per 100 patient-years, IRR 0.88, 95% CI 0.55-1.41, p=0.60) as was all sickle-related events (IRR 0.92, 95% CI 0.54-1.57, p=0.75). Grade 3 or higher AE were also slightly lower (0.17 versus 0.23 per patient-year, 0.57, 95% CI 0.27-1.22, p=0.15) and the transfusion rate was lower (0.09 versus 0.19 per patient-year, IRR 0.46, 95% CI 0.17-1.22, p=0.12). Conclusion: Initiating hydroxyurea in Tanzanian children with SCA using a PK-guided dosing algorithm is feasible, safe, and effective and confers both laboratory and clinical benefits. Similar safety and benefits can be achieved with dose initiation at 25 mg/kg/day, above the usual weight-based starting dose. Dose escalation to MTD at 25-35 mg/kg/day is appropriate for all children with SCA.