A population pharmacokinetic model of piperaquine in pregnant and non-pregnant women with uncomplicated Plasmodium falciparum malaria in Sudan

Symptomatic patients received a standard dose regimen of the fixed dose oral piperaquine-dihydroartemisinin combination treatment. Densely sampled plasma aliquots were collected and analysed using a previously described LC-MS/MS method. Data from 12 pregnant and 12 non-pregnant women were analysed using nonlinear mixed-effects modelling. A Monte Carlo Mapped Power (MCMP) analysis was conducted based on a previously published study to evaluate the power of detecting covariates in this relatively small study.

A three-compartment disposition model with a transit-absorption model described the observed data well. Body weight was added as an allometric function on all clearance and volume parameters. A statistically significant decrease in estimated terminal piperaquine half-life in pregnant compared with non-pregnant women was found, but there were no differences in post-hoc estimates of total piperaquine exposure. The MCMP analysis indicated a minimum of 13 pregnant and 13 non-pregnant women were required to identify pregnancy as a covariate on relevant pharmacokinetic parameters (80% power and p=0.05). Pregnancy was, therefore, evaluated as a categorical and continuous covariate (i.e. estimate gestational age) in a full covariate approach. Using this approach pregnancy was not associated with any major change in piperaquine elimination clearance. However, a trend of increasing elimination clearance with increasing gestational age could be seen.

The population pharmacokinetic properties of piperaquine were well described by a three-compartment disposition model in pregnant and non-pregnant women with uncomplicated malaria. The modelling approach showed no major difference in piperaquine exposure between the two groups and data presented here do not warrant a dose adjustment in pregnancy in this vulnerable population.