Population pharmacokinetic modelling and simulation of antimalaria drugs to optimize dosing in neglected populations

Abstract

Malaria is a mosquito-borne infectious disease which affects humans and other animals. In 2016 the World Health Organization (WHO) estimated 212 million malaria cases and 429,000 deaths occurred globally in the year 2015. This results in one child dying after every two minutes. Historically, children experience higher burden of the disease due in part to their lower immunity. The prevention of malaria disease relies on vector control. Two forms of vector control are currently available; insecticide-treated mosquito nets and indoor residual spraying. While for treatment, artemisinin combination therapy has been recommended by the WHO. Children are at risk of suboptimal dosing of antimalarials, with lower drug exposure related to malaria recrudescent in young children. Optimal curative dosing in adults is informed by extensive safety and efficacy clinical trials, but for children, dosing is currently based on extrapolation from adult doses. Often linear relationships between dose and body weight are assumed, which is the same as treating a child as a small adult. However, children differ from adults in regard to their response to drugs, changes in body weight, and changes in other factors which determine pharmacokinetic (PK) and pharmacodynamics (PD) relationships. These factors include delayed maturation of drug-metabolizing enzyme and maturation of renal function, and expression of receptors and protein involved in PD. To address these factors and improve dosing, we developed population pharmacokinetic models for several antimalarials and used simulation to recommend optimal dosing in children for two antimalarials. Two antimalarial drugs; amodiaquine (AQ) and naphthoquine (NPQ) were investigated in this thesis. Pharmacokinetic data of amodiaquine collected via 8 clinical studies were used, involving both adult and children uncomplicated malaria patients in African countries and in one Asian country. For both amodiaquine and naphthoquine, population pharmacokinetics (pop-PK) of drug concentration-time data were analysed using non-linear mixed effects methods with NONMEM statistical software. Covariates were analysed and the goodness of fit of each model was assessed via statistical tests and diagnostic plots. Individual parameter estimates from the final pop-PK models were used to simulate day 7 concentration (exposure) and maximum plasma concentration (Cmax) for both antimalarials. The simulated exposure values were compared across different weight bands, defined according to the manufacturer’s recommendation and were evaluated for each available tablet strength. For each weight band a new dose was recommended if the median exposure for each weight was less than 80% of median day 7 concentration of the typical patient. Both desethylamodiaquine and naphthoquine concentration time profiles were best described by three disposition kinetic models (3 compartment). For amodiaquine the kinetics were best described by two disposition kinetics (two compartment). Amodiaquine bioavailability was found to increase with the time (day) after drug administration and clearance maturation with age. Children at birth were determined to have 27% of adult clearance for amodiaquine and 24% for desethylamodiaquine. In addition, children were determined to reach 50% of the weight-adjusted adult clearance values at 3 and 4 months after birth for amodiaquine and desethylamodiaquine and 100% values at about 3 years of age. No covariate was found to be associated with pharmacokinetic parameters of the naphthoquine. Simulation results indicated that the day 7 concentrations of amodiaquine and naphthoquine in children were on average 25-30% lower than a typical 50 kg patient and hence higher doses are needed compared to current recommended doses. The optimized dosing regimen recommended includes higher doses per kg for younger children. In particular, higher mg/kg doses are recommended in smaller children, consistent with the nonlinear effect of body size on clearance described by allometric scaling. The optimized regimen proposed in this thesis aims to achieve more a balanced exposure across weight bands without any risk of toxicity. They will need to be further assess in light of operational, safety, and clinical studies. This is the first study to investigate the suboptimal plasma exposures level of amodiaquine and naphthoquine antimalarials and hence confirm the previous findings of other antimalarials that dosing in young children is not adequate when based on allometric scaling alone. Simulation of well calibrated pop-PK, models, indicates that for young children and patients of weight above 60 kg require higher doses than the current manufacturer recommended dose. The optimized dose will ensure similar exposure levels over all patient’s weight range and hence reduce the selective pressure of the development of drug resistance.