Artemisinin-resistant Plasmodium falciparum: can the genie be put back in the bottle?
by Timothy Egan, Future Microbiology:
The late 1980s and early 1990s saw what virtually amounted to an international malaria emergency in which the gains of the previous four decades in the fight against malaria were rapidly eroding. The spread of chloroquine resistance, as well as resistance to sulfadoxine-pyrimethamine, eliminated the two cheap, safe and effective options for treatment of the disease. Resistance had also appeared to the much newer and more expensive mefloquine. In addition, resistance to insecticides in the vector mosquitoes further compromised malaria control. In response to this situation, the international community set up a number of initiatives to deal with this neglected disease. These include the Multilateral Initiative on Malaria, the Roll Back Malaria Programme, the Medicines for Malaria Venture and the Global Fund to Fight AIDS, TB and Malaria. These programs have assisted countries in Africa and elsewhere in the control of malaria, with pyrethroid-treated bed nets, for example, as well as promoting and funding the discovery and development of new drugs. Partly as a consequence of these initiatives, the picture with respect to malaria showed a marked improvement after 2000. From the point of view of antimalarial drugs, by far the most important factor has been the introduction of artemisinins and, in particular, artemisinin combination therapy (ACT). ACT, now the WHO-recommended treatment for falciparum malaria in sub-Saharan Africa and elsewhere [101], has been implemented in a number of countries. By 2006, 15 countries in Africa had adopted artesunate and amodiaquine [1], while others, including South Africa and Mozambique, have adopted artemether and lumefantrine ACT. In southeast Asia, artesunate and mefloquine have been used in combination for almost a decade.
In 1999 and 2000, experimental artemisinin resistance was reported in the rodent malaria parasite Plasmodium yoelii for the first time [2,3]. However, the drug-resistant phenotype was found to be unstable and it was widely assumed that clinical artemisinin resistance was unlikely to develop. Then, in 2006, Cravo and coworkers reported the first example of stable artemisinin resistance in the rodent malaria parasite Plasmodium chabaudi chabaudi [4]. In this case, drug pressure with either artesunate or artemisinin induced significant resistance (six- to 15-fold increase in minimum curative dose). This decrease in drug sensitivity remained unchanged following removal of drug pressure through multiple blood passages, survived freezing and thawing of parasites and could be transmitted by mosquitoes. This study clearly demonstrated that the assumption that stable artemisinin resistance is impossible was false. However, it could not exclude the possibility that development of artemisinin resistance by P. chabaudi chabaudi was a peculiarity of this particular species of parasite. Field isolates of Plasmodium falciparum from Cambodia, French Guiana and Senegal with reduced in vitro sensitivity to artemether were reported in 2005 [5] but, as pointed out by Golenser et al.. neither this observation, nor clinical failures alone, definitively demonstrate the appearance of drug resistance [6].
Then, in December 2008, the strongest evidence yet of genuine artemisinin-resistant falciparum malaria was reported in western Cambodia [7]. This study by Noedl and others defined patients as having artemisinin-resistant falciparum malaria when: parasites persisted after 7 days of treatment with artesunate or re-emerged 28 days after the start of treatment; the level of dihydroartemisinin, the major drug metabolite, reached adequate levels; prolonged time to clearance was observed; and reduced susceptibility to dihydroartemisinin was demonstrated. Of 60 patients treated with artesunate in this study, two exhibited artemisinin resistance as defined by the above. This seems to be the first definitive evidence of clinical artemisinin resistance.
At present, there is no evidence of artemisinin resistance occurring anywhere else in the world. The problem appears to be isolated to this one geographic area and seemingly is a result of evolutionary pressure arising from peculiarities of malaria treatment in Cambodia. In particular, the widespread availability of artemisinin monotherapy for more than 30 years, usually sourced through the private sector and often self-administered in inadequate doses [8]. Ominously, the Thai-Cambodian border was also the source of chloroquine and sulfadoxine-pyrimethamine resistance, which subsequently spread throughout the malaria-endemic world. If the problem is not urgently and appropriately addressed, it is likely that artemisinin resistance will similarly spread worldwide. Given the pivotal role that artemisinins now play in malaria treatment, this would be a major disaster as alternative drugs are simply not available, especially since few, if any, compounds have the advantages of artemisinins. These advantages include multiple possible routes of administration, suitability for treating severe malaria without the side effects of quinine, rapid activity and activity against not only trophozoites and blood schizonts, but also ring forms and gametocytes [9]. No other antimalarials offer all of these advantages. Now that artemisinin resistance has appeared, what can be done to put the genie back in the bottle and what lessons can be drawn? Clearly, in the first instance there is a need for surveillance. In addition, the question of whether artemisinin resistance can be squashed before it bursts out of the Thai-Cambodia border area needs careful consideration.
As regards surveillance, molecular biological markers of resistance are better indicators than clinical failure since the latter can arise from many causes other than drug-resistant parasites. A fundamental difficulty at present is that neither the mechanism of artemisinin activity, nor of resistance, is definitively known. One school of thought is that artemisinins act via a radical process involving activation of their endoperoxide group by Fe(II), probably in the form of heme [10,11]. According to this hypothesis, oxidation of Fe(II) by the peroxide bridge generates a highly unstable O-centered radical, which then rearranges to a more stable C-centered radical that diffuses away to alkylate essential proteins or nucleic acids, or alternatively reacts with heme, producing an adduct that then exerts a toxic effect on the parasite. This hypothesis has been criticized on several grounds, the most compelling of which is that the exceptionally low IC50 values of artemisinins cannot be reconciled with the nonspecific diffusion of the inherently rather unstable C-centered radical to sites of alkylation [9]. Others have argued that artemisinins act by specifically inhibiting P. falciparum Ca2+ ATPase (PfATP6) [12]. Indeed, artemisinin has been shown to inhibit this protein when expressed in Xenopus oocytes. Mutation of a key amino acid residue (Leu263), which is close to the putative artemisinin binding site, changes sensitivity to these drugs in this model [13]. For example, the mutation L263E abolishes the activity of artemisinin altogether. However, while such changes seem to correlate with lack of artemisinin activity against mammalian cells, they have not been seen in drug-resistant parasites. Indeed, no evidence was found of any changes in PfATP6 in the artemisinin-resistant parasites discovered in western Cambodia [4], nor were any mutations or changes in copy number in PfATP6 found in artemisinin-resistant P. yoelii. Only in the case of parasites from French Guiana showing reduced in vitro sensitivity was a mutation in PfATP6 found, in this case a S769N mutation [5]. Therefore, it is unclear at present whether surveillance of the PfATP6 gene is likely to be useful. A better understanding of the mechanism of action and resistance to artemisinin is now urgent.
A recent paper by Maude et al. has attempted to address the question of whether artemisinin-resistant malaria can be eliminated before it spreads away from western Cambodia [8]. The authors have developed a mathematical model to try to predict the effects of various interventions. The results are sobering. The model suggests that the immediate implementation of ACT could eliminate artemisinin-resistant malaria from Cambodia in less than 4 years. However, as the parasite population dwindles, the remaining proportion of parasites would be dominated more and more by the resistant strain. It would then become critical to ensure complete eradication of these parasites. Failure to do so would worsen the problem and, ultimately, quite likely lead to more rapid spread of resistant malaria from this source. Other interventions such as mass screening and treatment campaigns or mass drug administration campaigns are predicted to marginally slow the growth of artemisinin resistance, but have no long-term effect. Only the use of treated bed nets may exhibit a synergistic effect in aiding the use of ACT. If the model is an accurate reflection of the true situation, only an unprecedented effort would have a chance of success. In particular, the campaign would have to be sustained until all malaria parasites are eliminated from this geographical region. Whether such a campaign could be mounted and sustained in the middle of a global financial and economic crisis remains to be seen. In any event, it is likely that the window of opportunity to make decisions and take decisive action is short and closing rapidly.
Finally, the lesson that needs to be learned is that it is most unlikely that any drug will provide the final answer to the problem of malaria. Rather, the fight against this parasite is an ongoing battle requiring the constant introduction of new drugs to replace those against which resistance appears. Nonetheless, at present ACT is the best option available, and it is important that concerns about drug resistance should not be used to delay the introduction of ACT in Africa and elsewhere where it is highly effective.
See link below for full text and citations.