The influence of pregnancy on the pharmacokinetic properties of artemisinin combination therapy (ACT): a systematic review

Abstract Background Pregnancy has been reported to alter the pharmacokinetic properties of anti-malarial drugs, including the different components of artemisinin-based combination therapy (ACT). However, small sample sizes make it difficult to draw strong conclusions based on individual pharmacokinetic studies. The aim of this review is to summarize the evidence of the influence of pregnancy on the pharmacokinetic properties of different artemisinin-based combinations. Methods A PROSPERO-registered systematic review to identify clinical trials that investigated the influence of pregnancy on the pharmacokinetic properties of different forms of ACT was conducted, following PRISMA guidelines. Without language restrictions, Medline/PubMed, Embase, Cochrane Central Register of Controlled Trials, Web of Science, LILACS, Biosis Previews and the African Index Medicus were searched for studies published up to November 2015. The following components of ACT that are currently recommend by the World Health Organization as first-line treatment of malaria in pregnancy were reviewed: artemisinin, artesunate, dihydroartemisinin, lumefantrine, amodiaquine, mefloquine, sulfadoxine, pyrimethamine, piperaquine, atovaquone and proguanil. Results The literature search identified 121 reports, 27 original studies were included. 829 pregnant women were included in the analysis. Comparison of the available studies showed lower maximum concentrations (Cmax) and exposure (AUC) of dihydroartemisinin, the active metabolite of all artemisinin derivatives, after oral administration of artemether, artesunate and dihydroartemisinin in pregnant women. Low day 7 concentrations were commonly seen in lumefantrine studies, indicating a low exposure and possibly reduced efficacy. The influence of pregnancy on amodiaquine and piperaquine seemed not to be clinically relevant. Sulfadoxine plasma concentration was significantly reduced and clearance rates were higher in pregnancy, while pyrimethamine and mefloquine need more research as no general conclusion can be drawn based on the available evidence. For atovaquone, the available data showed a lower maximum concentration and exposure. Finally, the maximum concentration of cycloguanil, the active metabolite of proguanil, was significantly lower, possibly compromising the efficacy. Conclusion These findings suggest that reassessment of the dose of the artemisinin derivate and some components of ACT are necessary to ensure the highest possible efficacy of malaria treatment in pregnant women. However, for most components of ACT, data were insufficient and extensive research with larger sample sizes will be necessary to identify the exact influences of pregnancy on the pharmacokinetic properties of different artemisinin-based combinations. In addition, different clinical studies used diverse study designs with various reported relevant outcomes. Future pharmacokinetic studies could benefit from more uniform designs, in order to increase quality, robustness and effectiveness. Study registration: CRD42015023756 (PROSPERO)