The effect of sickle cell genotype on the pharmacokinetic properties of artemether-lumefantrine in Tanzanian children

Since there are inconsistent data relating to the effect of haemoglobinopathies on disposition of artemisinin antimalarial combination therapy, and none in sickle cell trait (SCT) or sickle cell disease (SCD), the aim of this study was to characterize the pharmacokinetic properties of artemether-lumefantrine (ARM-LUM) in children with SCD/SCT. Thirty-eight Tanzanian children aged 5–10 years with normal (haemoglobin AA; n = 12), heterozygous (haemoglobin AS; n = 14) or homozygous (haemoglobin SS; n = 12) sickle genotypes received six ARM-LUM doses (1.7 mg/kg plus 10 mg/kg, respectively) over 3 days. Sparse venous and mixed-capillary dried blood spot (DBS) samples were taken over 42 days. Plasma and DBS ARM and LUM, and their active metabolites dihydroartemisinin (DHA) and desbutyl-lumefantrine (DBL), were assayed using validated liquid chromatography-mass spectrometry. Multi-compartmental pharmacokinetic models were developed using a population approach. Plasma but not DBS concentrations of ARM/DHA were assessable. The majority (85%) of the 15 measurable values were within 95% prediction intervals from a published population pharmacokinetic ARM/DHA model in Papua New Guinean children of similar age without SCD/SCT who had uncomplicated malaria, and there was no clear sickle genotype clustering. Plasma (n = 38) and corrected DBS (n = 222) LUM concentrations were analysed using a two-compartment model. The median [inter-quartile range] LUM AUC_0–∞ was 607,296 [426,480–860,773] μg.h/L, within the range in published studies involving different populations, age-groups and malaria status. DBS and plasma DBL concentrations correlated poorly and were not modelled. These data support use of the conventional ARM-LUM treatment regimen for uncomplicated malaria in children with SCT/SCD.