Optimal designs for population pharmacokinetic studies of the partner drugs administered with artemisinin derivatives

Background: The World Health Organization recommends artemisinin-based combination therapy (ACT) as first line treatment for uncomplicated falciparum malaria1. ACT consists of a highly effective but short lived artemisinin derivative and a less effective but longer lasting partner drug(s). We have recently proposed optimal designs for future population pharmacokinetic studies of the main artemisinin derivative (Artesunate), and are now focusing on the partner drugs.

Aim: To determine and assess optimal designs for future population pharmacokinetic studies of the partner drugs administered with artemisinin derivatives: Mefloquine, Lumefantrine, Piperaquine and Sulphadoxine-Pyrimethamine.

Methods: Optimal designs for each partner drug were determined using POPT3, and were based on (i) reported population pharmacokinetic models for non-pregnant adults, pregnant women and children, and (ii) key sampling constraints identified in a questionnaire sent to active malaria researchers (five samples per patient, at least 15 minutes between samples and a minimum of 100 patients). The expected %RSEs from POPT and empirical %RSEs from simulation-estimation for the optimal designs were compared with those from fixed sampling designs that collected eight samples per patient at non-optimal times.

Results: The optimal designs yielded acceptable expected precision of the pharmacokinetic parameters (%RSEs < 30%), between subject-variability (%RSEs < 60%, with the exception of peripheral volume and clearance, <75%) and residual variability (%RSEs < 35%). Preliminary results from simulation-estimation of the optimal designs generally support these findings, with the exception of the empirical %RSE of ka for Lumefantrine being quite large (>100%). The optimal designs displayed similar expected precision to the fixed designs. Furthermore, the designs were fairly consistent across the drugs for the first two sampling times, but not for the latter three.

Conclusions: The proposed optimal designs for Mefloquine, Lumefantrine, Piperaquine and Sulphadoxine-Pyrimethamine are efficient and performed similarly to designs with more intensive blood sampling. This reduction in samples reduces the cost of the study and is more acceptable to children, who are the most vulnerable population and less studied in terms of pharmacokinetics.


  1. World Health Organization. Guidelines for the treatment of malaria, second edition. Geneva, Switzerland. ISBN 978 92 4 154792 5, 2010.
  2. F Mentre, A Mallet, and D Baccar. Optimal design in random-effects regression models. Biometrika, 84(2):429–442, 1997.
  3. SB Duffull. POPT – Installation and user guide. Ver 3.0. www.winpopt.com, 2006.

Kris Jamsen

  • Model Answers Pty Ltd