**6. Vectors**

Clinical artemisinin resistance was first detected in western Cambodia [92, 93, 96, 97] but is now detected in other GMS regions including Thailand, Laos, Vietnam, Southern Myanmar, and the China-Myanmar border area [94, 95, 98–103]. Out of fear of a catastrophic spread of artemisinin resistance to Africa, WHO deployed an artemisinin resistance containment plan in Cambodia [104]. Later, with the finding that artemisinin resistance has emerged independently in many areas of the GMS [105], the containment plan has been revised to a regional

The principle of ACTs is that the fast-acting artemisinins rapidly reduce the parasite biomass, leaving the slow-eliminating partner drugs to clear the residual parasites. The emergence of artemisinin resistance means that a larger parasite mass is left for the partner drugs to clear after the usual 3-day ACT course, which increases the chance of resistance development to the partner drugs. Indeed, in the short period of time since the deployment of ACTs, clinical resistance to two ACTs, first artesunate/mefloquine [106] and more recently dihydroartemisinin/piperaquine (DHA/PPQ), has emerged in the GMS.These are the two most popular ACTs deployed in the GMS countries (**Table 1**). Since promising new antimalarials are still in the development pipeline, possible solutions to this problem include introduction of new ACTs, rotation of different ACTs, use of longer course of ACT treatment, and introduction of triple ACTs (artemisinin derivatives with two slow-eliminating partner drugs) [112]. To mitigate the threat of spread of artemisinin-resistant *P. falciparum* parasites, heightened surveillance is needed in sentinel sites of the GMS [113].

Tools for monitoring the epidemiology of antimalarial drug resistance include ex vivo or in vitro drug assays and molecular surveillance, which complement in vivo drug efficacy studies. It is noteworthy that the slow-clearance phenotype of clinical artemisinin resistance does not correspond to the 50% inhibitory concentrations of artemisinin drugs estimated from the conventional DNA replication-based in vitro assay but is better reflected in the newly developed ring-stage survival assay, which quantifies the number of early ring-stage parasites (0–3 h) that can survive the exposure to 700 nM of DHA for 6 h [114]. The discovery of mutations in the *kelch* domain protein K13 associated with artemisinin resistance provides a convenient molecular marker for a large-scale surveillance purpose [115]. To date, the correlations of K13 mutations with delayed parasite clearance have been established in several studies [95, 105, 115–117] but only a very limited number of K13 mutations were confirmed to confer in vitro artemisinin resistance through genetic manipulations [118, 119]. The K13 gene in the world *P. falciparum* populations harbors more than 108 nonsynonymous mutations, which showed marked geographic disparity in frequency and distribution [120]. Similarly, K13 mutations in the GMS also showed highly heterogeneous distribution [103, 121–125], possibly reflecting different drug histories and evolutionary origins of the parasite populations [126]. Clinical failures of DHA/ PPQ have been associated with increased in vitro PPQ resistance and the molecular markers of PPQ resistance in western Cambodia include amplification of the aspartic protease genes *plasmepsin 2–3* and point mutation E415G in an exonuclease gene (PF3D7\_12362500) [127, 128]. Molecular surveillance of artemisinin resistance in western Cambodia, Thailand, and Laos has detected the spread of a parasite clone with a long K13 haplotype carrying the C580Y mutation (the artemisinin-resistant mutation reaching near fixation in western Cambodia) to northeastern Thailand and southern Laos, which indicates a transnational selective sweep [129]. Importantly, this parasite lineage also harbors the *plasmepsin 2* amplification, which may preclude further use of DHA/PPQ in this region. In addition, this situation also necessitates implementation of

malaria elimination strategy [3, 4].

186 Towards Malaria Elimination - A Leap Forward

LLINs and IRS are the key vector-based malaria interventions that have been found to be highly effective in sub-Saharan Africa. However, these measures are much less efficient in the GMS [130]. The GMS has a complex vector system; most of the malaria vectors belong to species complexes or groups such as Dirus, Minimus, Maculatus, and Sundaicus, which vary significantly in terms of geographic distribution, ecology, behavior, and vectorial competence [131–133]. At least 19 species are known malaria vectors, some of which comprise cryptic species complexes [132]. In order to apply the appropriate control approaches in relation to the biology of the vector species, we first need to identify the mosquitoes to their species level and to differentiate the vector from nonvector species, which requires molecular assays [134]. These vector species display significant variations in geographical distribution and seasonal dynamics, and accordingly their roles in malaria transmission also vary in space and time [135]. In many endemic areas of the GMS, perennial malaria transmission is maintained by *Anopheles dirus* during the rainy season and *An. minimus* during the drier periods of the year [132, 136]. Environmental changes such as deforestation have caused changes in the vector species composition [137, 138] and benefited the survivorship of major vectors [40]. Since many of these vector species exhibit early evening and outdoor biting preferences, LLINs alone are not sufficient for interrupting malaria transmission [140]. In addition, the emergence and spread of insecticide resistance further compromise the effectiveness of the mosquito control measures [141–143].
