Plasmodium falciparum resistance to amodiaquine in monotherapy and in combination therapy with artesunate

Abstract

In response to the increasing resistance to antimalarial monotherapies, artemisinin basedcombination therapy (ACT) is now recommended as first line therapy against uncomplicatedPlasmodium falciparum malaria. However, the choice of partner drug to artemisinin (ART) iscritical for ACT efficacy to endure. One main partner drug option for ACT is amodiaquine(AQ), with its active long half-life metabolite desethyl-amodiaquine (DEAQ). AQ is related tochloroquine (CQ) and has been widely used in Africa for decades, but despite widespread CQresistance it has remained relatively effective. CQ resistance has been associated with mutationsin the P. falciparum CQ resistance transporter (pfcrt) gene and the P. falciparum multiple drugresistance 1 (pfmdr1) gene. Possible mutations associated with AQ/DEAQ resistance haveremained unclear. In this thesis we explore whether mutations, especially in the pfcrt andpfmdr1 genes, are associated with tolerance/resistance to AQ/DEAQ in monotherapy and inACT, as well as a possible associated parasite fitness cost.

The thesis is based on (a) in vivo clinical trials in East-Africa with AQ monotherapy or ARTplus AQ (ASAQ) combination therapy and (b) in vitro studies on isolates from Colombia and reference clones in which the pfmdr1 gene has been modified by allelic exchange. Mutationanalyses were done by PCR followed by either RFLP and/or DNA pyrosequencing or fullsequencing and gene amplification analysis was done by TaqMan probe based Real-Time PCR.The genetic results were related to drug susceptibilities determined by an HRP2-ELISA assayand parasite growth in competition experiments in vitro and/or the clinical outcome in vivo.

The treatment failure rate after AQ therapy was relatively high (20%), while after ASAQtherapy it fulfilled the efficacy criteria of WHO (<10%). AQ/DEAQ tolerance/resistance wasfound to be associated with pfmdr1 1246Y in addition to pfcrt (a.a. 72-76) CVIET/SVMNTand possibly in a synergistic or compensatory relation with pfmdr1 86Y, 184Y, 1034C and1042D. Possibly pfcrt 326S/D and 356T/L, as well as a newly identified mutation pfcrt 334N,are also involved. Treatment failure after AQ monotherapy was not found to be associated withrare findings of pfmdr1 amplifications or variable DEAQ blood concentrations. No newmutations could be verified in the pfcrt and pfmdr1 genes in recrudescent parasites after AQ orASAQ therapy. The partial cross-resistance with CQ is probably conferred to mainly DEAQ through the pfcrt gene, while resistance to AQ may be more dependent on the pfmdr1 gene. Pfmdr1 1246Y was associated with a substantial fitness cost to the parasite. The relative growth for parasites with a particular mutation represents a concentration dependent balance between the fitness cost and the specific drug selection benefit. The in vitro estimated cost-benefit from pfmdr1 1246Y correlated with the allele dynamics after AQ and ASAQ therapy in vivo. The added effect of ART will potentially prevent against a selection of pfmdr1 1246Y by a more effective reduction of parasite biomass and an opposite pfmdr1 allele selection and in the absence of AQ exposure pfmdr1 1246Y will possibly incur too substantial fitness cost to sustain in competition with wild type parasites. We conclude that several factors counteract a selection of AQ resistance, which support sustained efficacy of ASAQ. Thus, we estimate that AQ represents a valuable partner drug option in ACT in East Africa.