The effects of pharmacogenetics on pharmacokinetics of artemisinin-based combinations in malaria patients

Malaria is a vector-borne infectious disease caused by protozoan parasites of the genus Plasmodium. If not treated appropriately, human P. falciparum malaria can quickly become life-threatening, leading to an estimated 900’000 annual deaths globally. Key interventions to control malaria include prompt diagnosis and effective treatment with artemisinin-based combination therapies (ACTs), use of insecticide treated nets by people at risk, indoor residual spraying with insecticide to control the vector mosquitoes and intermittent preventive treatment for pregnant women (IPTp) and infants (IPTi).
Whether antimalarial treatments are effective or not, depends on parasite and host factors. The ability to define resistance leading to treatment failure has been greatly enhanced by our understanding of the underlying molecular mechanisms causing resistance in P. falciparum. However, the potential contribution of host genetic factors, particularly those associated with antimalarial drug metabolism, remains largely unexplored. The same applies for the basic mechanisms involved in the pharmacokinetics of antimalarial drugs and the link between antimalarial drug pharmacokinetics and treatment outcomes. Thus, the purpose of this thesis was to quantify the effects of pharmacogenetics on pharmacokinetics of ACTs.
Between 2007 and 2008, three in vivo studies were performed in Cambodia and Tanzania. Patients reporting with fever associated with an infection with Plasmodium falciparum were recruited and treated with ACTs according to the national guidelines in the respective country. In Cambodia, 64 patients were recruited for the treatment with artesunate–mefloquine and 61 for the treatment with dihydroartemisinine–piperaquine. In Tanzania, 150 were treated with artemether–lumefantrine. Blood samples for the pharmacokinetic analysis were taken before treatment and at several time points during and after treatment, e.g. on Days 1, 2 and 7 in all studies and in Cambodia also 1 hour after the first dose and on Day 14.
For the analysis of plasma samples collected during our studies, we developed a broad-range liquid chromatography coupled with tandem mass spectrometry (LC–MS/MS) assay covering 14 of the currently in-use antimalarial drugs and their metabolites. The assay requires only as little as 200 μl of plasma and is a major improvement over previous methods in terms of convenience, sensitivity, selectivity and throughput. The method was validated according to well-established recommendations. The assay was first used for the analysis of the baseline samples collected in our in vivo studies. In all studies more than half of the patients recruited had still antimalarials in their blood. Theses findings enabled us to get a better assessment of the antimalarials circulating in the local population, and hence of the drug pressure on the parasites in both countries.
Single nucleotide polymorphisms (SNPs) in genes encoding enzymes associated with antimalarial drug metabolism, i.e. cytochrome P450 isoenzymes (CYP) and N-acetyltransferase 2 (NAT2), were analyzed. Based on our previous experience, we developed a DNA microarray to affordably generate SNP data. However, after comparison of microarray data and sequencing data, we concluded that the major limit of the microarray technology was lack of robustness which could not be compensated by superior cost-effectiveness. Consequently, the pharmacogenetic profiles of the patients from the three in vivo studies were assessed by direct sequencing of genomic DNA. Whereas for most SNPs allele frequencies were similar in both populations, we found significant inter-ethnic differences in the distribution of genotypes of certain enzymes, namely CYP2D6, CYP3A4/5 and NAT2. Is has been shown that the human CYP3A subfamily plays a dominant role in the metabolic elimination of more drugs than any other biotransformation enzyme. Therefore, our findings might have implications for treatment policies of not only antimalarials and the widely introduced ACTs in particular, but any other drugs metabolized by these enzymes.
To quantify the effect of pharmacogenetics on pharmacokinetics of ACTs we developed population pharmacokinetic models. The pharmacokinetic parameters we estimated in our models were in agreement with those from previous studies. In order to account for parts of the inter-individual variability in drug-metabolizing capacity of the liver we included pharmacogenetic data as covariate. For artemether, we found that 9% of the inter-individual variability in clearance could be explained by the genotype of CYP3A5 (reference allele versus variant allele CYP3A5*3). Heterozygous carriers showed a reduction in clearance of 34%. The alterations in clearance were less pronounced for lumefantrine (increase in clearance of 12% in homozygous carriers of variant allele CYP3A4*1B, explaining 2% of the inter-individual variability in clearance) and mefloquine (decrease in clearance of 14% in carriers of homozygous variant allele CYP3A5*5, explaining 1% of the inter-individual variability in clearance). These data might partially provide an explanation for the differences in drug efficacy observed with artemether–lumefantrine combination treatment.
In conclusion, we were able to show that there is a correlation between the pharmacogenetic profile of the host and the pharmacokinetics of antimalarial drugs administered in malaria patients. These results suggest that pharmacogenetics could be one of the basic mechanisms involved in the pharmacokinetics of antimalarial drugs. The knowledge gained from this study could facilitate the selection process of first-line treatment for malaria and would allow dosing adaptation based on the pharmacogenetic profile of the population. Such adaptations are needed especially in the most vulnerable groups, including infants, pregnant women, and those with prevalent co-morbidities, where often therapeutic antimalarial drug concentrations over time are not achieved.