Effect of local food on lumefantrine bioavailability and population pharmacokinetics in Ugandan children with malaria

Abstract

Background: Artemether-lumefantrine (AL) is widely adopted as first-line treatment for uncomplicated malaria. Lumefantrine (LUM), the long acting partner drug is critical for cure by eliminating malaria parasites left after artemether exposure. Absorption of LUM is dependent on dietary fat and the basis for the pediatric dose recommendations is unclear.

Aim: To explore effect of local foods on oral bioavailability of LUM and describe its population pharmacokinetics (PPK) among under five year old children in Uganda treated for malaria with the aim of optimizing use and provide basis for AL rational dosage guidelines.

Methods: In an intensive pharmacokinetics (PK) study, 13 healthy adult volunteers were randomized to participate in an open-label four period crossover design and received a single oral dose of AL (80mg A/ 480mg of LUM) with water, milk, maize porridge or maize porridge with oil on separate occasions. Peak concentrations (Cmax) and area under concentration-time curve (AUC) truncated at 48 hours after a single dose (AUC0-48h) were compared using average bioequivalence techniques (I). Relevance of the findings was assessed among children < 5 years with uncomplicated falciparum malaria who were randomized in a parallel study design to receive standard weight-based AL treatment (Coartem®), 6 doses over 3 days, with either milk or maize porridge with oil (n= 33) (III). Parametric two-sample t-test was used to compare relative oral LUM bioavailability, 0 to 8 h after the first dose (AUC0-8h) (III). This bioavailability study (III) was nested in a population pharmacokinetic (PPK) study (IV) in the same pediatric patient group. After treatment, sparse plasma samples were collected during 28 days’ follow up in all children (n=55). NONMEM was used to describe the PPK profile of LUM and its metabolite, desbutyl-lumefantrine (DBL) (IV). A liquid chromatography tandem mass spectrometry (LC-MS/MS) method was developed for determination of low concentrations of LUM and DBL in small amounts of plasma (II).

Results: The LC-MS/MS method was simple, fast and sensitive requiring only 100 μl of plasma with limits of quantification of 21 and 1.7 ng/ml for LUM and DBL respectively (II). Lumefantrine exposure was comparable in milk and maize porridge plus oil study groups (I & III). In adult healthy volunteers, the bioequivalence criteria was met [maize porridge plus oil group ranges for means ratios (90% CI) of 0.84 – 1.88 and 0.85 – 1.69 for Cmax and AUC0-48h respectively, relative to milk (90%CI, 0.80 – 1.25)]. Among pediatric patients, LUM (AUC0-8h) for those dosed with milk (n=16) was comparable to maize porridge plus oil (n=17) arm (GM {95%CI}: 6.01 {3.26-11.1} vs 6.26 {4.5 - 8.43} h*μg/ml, p=0.9). A two-compartment PK model with lag time using first order processes characterized the PPK of LUM (IV). Inter-subject variability in apparent oral clearance (CL/F) was explained by body mass index (BMI) and age, while that in apparent volume of distribution of the central compartment (VC/F) was explained by weight. Lumefantrine population mean CL/F, inter-compartment clearance (Q/F), VC/F and apparent volume of distribution of peripheral compartment (VP/F) were 3.19 L/hr, 0.176 L/hr, 28.1 L, and 58.4 L, respectively. Our results indicate that LUM CL/F decreased with age from two to just less than five years (≈20.6%. p=0.04) and LUM CL/F increased with decreasing BMI.

Conclusions: The LC MS/MS method is suitable for pediatric studies with repeated sampling and long time follow up. Oil fortified maize porridge can be an alternative to milk in augmenting absorption of LUM. Our findings provide a structural basis for consideration of age and BMI in evaluation of rational AL dosing guidelines among under five year old children.