Background
The emergence and spread of multidrug resistant
Plasmodium falciparum is the key factor contributing to complexity in malaria control. To deal with the threat of resistance, artemisinin-based combination therapy (ACT) is being promoted as a strategy to conteract the increasing resistance of the parasite as well as to prevent disease transmission[
1]. Despite the precautionary measures however, artemisinin-resistant
P. falciparum malaria has emerged in western Cambodia and the bordering regions with Thailand, the hotspot of multidrug resistance parasites[
2‐
10], and appears to be emerging in the western border of Thailand[
11,
12].
In Thailand, a three-day, artesunate-mefloquine combination regimen has been used as the first-line treatment for acute uncomplicated falciparum malaria throughout the country[
13]. In a previous study which aimed at monitoring clinical efficacy of this three-day artesunate-mefloquine combination regimen during the year 2009 in 134 patients with acute uncomplicated falciparum malaria in the area along the Thai-Myanmar border, a markedly high failure rate was observed[
11]. The 28- and 42-day cure rates calculated by Kaplan-Meier survival analysis with PCR correction for re-infection were 74.7 and 68.1%, respectively. It is noted that re-appearance of parasitaemia occurred as early as seven days after the first dose. In addition, there was a small but significant delay of parasite clearance in the group with recrudescence response (median (range) parasite clearance time 32.0 (28.0-34.0) h) compared with the sensitive group (26.0 (24.0-26.0) h). Only six (17.6%) and seven (20.5%) patients with recrudescence response, respectively, had parasitaemia and fever cleared within 24 hours. This observation is alarming and is of great concern if resistance of
P. falciparum has actually developed and spread in this area. In the present study, identification of resistance/reduced sensitivity of
P. falciparum in this border area to each component of this three-day, artesunate-mefloquine combination regimen was proposed based on clinico-parasitological response, with confirmed adequacy of anti-malarial systemic drug exposure during acute phase infection, and
in vitro sensitivity of
P. falciparum isolates to each combination partner. In addition, the possible link between the identified “resistance” cases and
P. falciparum multidrug resistance 1
(pfmdr1) copy number, the candidate molecular markers of mefloquine and/or artesunate resistance was investigated.
Discussion
Accumulating evidence suggests a decline in the efficacy and some degree of resistance of
P. falciparum in the Greater Maekong Subregion (GMS) to artemisinins. Early evidence came from western Cambodia and the Thai-Cambodian border in patients following treatment with either artesunate monotherapy or artesunate-mefloquine[
2,
3,
9,
10]. Although results of the containment project in seven provinces of Thailand bordering Cambodia (Buriram, Chantaburi, Sakaew, Srisaket, Surin, and Trat) during 2009–2011, in a total of 1,709
P. falciparum-positive cases, suggest that the therapeutic efficacy of artesunate-mefloquine remains at an acceptable level with cure rate of greater than 90%, continuous monitoring of
P. falciparum resistance in both border areas is critical[
29]. With regard to the Thai-Myanmar border, until recently[
11,
12], there has been no clear evidence of a significant reduction in artemisinin efficacy at either clinical or
in vitro sensitivity level. A longitudinal investigation in a total of 3,202 patients during 2001–2010 from malaria clinics, Shoklo Malaria Research Unit, Tak Province, unveiled that genetically determined artemisinin resistance in
P. falciparum may have emerged along the Thai-Myanmar border at least eight years ago and has since markedly increased[
12]. The clinical efficacy of artesunate-mefloquine combination therapy is beginning to decline in the Thai-Myanmar border and resistance is not only confined to western Cambodia and areas along the Thai-Cambodian border. Due to the limitation of the study design however, it was not possible to attribute treatment failures in these studies to resistance or host-related factors (e g, pharmacokinetics). Furthermore, if resistance actually occurred, it was unclear whether this was due to intrinsic parasite resistance to artesunate alone, mefloquine alone, or both, because of the pre-existing background of mefloquine resistance in these areas. In order to exclude the contribution of host and confounding factors from the partner drug mefloquine, a series of investigation with artesunate monotherapy (2–6 mg/kg body weight/day for seven days) was performed during 2006–2008 using stringent criteria for defining artemisinin resistance with integrated
in vivo-in vitro approach[
4‐
8].
The present study was designed to identify the treatment failure cases due to intrinsic parasite factor to each component of the three-day regimen. Resistance or decline in parasite susceptibility was confirmed based on integrated information on clinicopathological assessment together with
in vitro sensitivity (intrinsic parasite resistance) and systemic drug exposure (pharmacokinetic factor) in 17 out of 29 patients with recrudescence (LPF) following treatment with a three-day combination regimen of artesunate-mefloquine[
11]. All had significant delay in parasite clearance rate (slope half-life) compared with the sensitive cases. Despite the fact that the delay in parasite clearance is influenced by host-related factors, it is proposed as a sensitive marker of reduced susceptibility of artemisinin component of the ACT than recrudescence rate[
30]. The main effect of artemisinins is proposed to be on the slope of the log-linear decline in parasite clearance and thus, the slope half-life[
15].
Resistance of the
P. falciparum isolates to mefloquine or aretsunate was defined according to
in vitro sensitivity criteria. Existing data on the relationship between
in vitro sensitivity and artemisinin are controversial. Although lack of significant correlation between
in vitro sensitivity of artemisinins and clinical response was found in some studies[
4,
31], good correlation was observed in most studies[
2,
3,
9,
26,
32,
33]. Adequacy of mefloquine and dihydroartemisinin systemic drug exposure was defined based on the lower limits of 95% CI of the median AUC
0-7 days and AUC
0-24h, respectively, in the sensitive group. Based on this defined criteria, results suggest that low level of a decline in sensitivity of
P. falciparum to artesunate (in terms of a small increase in IC
50 of artesunate and number of identified cases with reduced sensitivity to artesunate) exists in this area on a background of pre-existing mefloquine resistance, and both parasite in conjunction with host-related factors significantly contributed to high failure rate in this group of patients. There was only one (out of 17) confirmed case with reduced sensitivity to artesunate alone, while there were three cases with reduced sensitivity/resistance to both artesunate and mefloquine. Pharmacokinetic factor contributed to about 58.8% (10 of 17 cases) of the total recrudescence cases. Inadequacy of mefloquine and dihydrortemisinin systemic drug exposure was observed in five and five cases, respectively. However, there was no significant difference in the systemic exposure of both mefloquine and dihydroartemisinin (AUC
0-7 days and AUC
0-24h) in patients with treatment failure compared with those with sensitive response.
Host-related factor contributing to treatment failure in one case could not be definitely defined. It is noted that the
in vitro sensitivity of the only one isolate with identified resistance/reduced sensitivity to artesunate alone (No 16) was markedly low compared with others (IC
50 5.2 nM). Furthermore, in one (No 11) of the three cases with
in vitro resistance/reduced sensitivity to both mefloquine and artesunate, mefloquine concentration of as high as 1,250 ng/ml was detected on the day of recrudescence (day 17). It is of note that this high level was still inadequate to completely eliminate residual parasites on background of resistance/reduced sensitivity to both mefloquine and artesunate. In the other two cases with contribution of mefloquine pharmacokinetic factor (Nos 1 and 2), variable whole blood mefloquine concentration on the days of recrudescence of 610 and 100 ng/ml was observed. The relatively low drug concentrations in some of the recrudescence cases could be due to variability in pharmacokinetics of mefloquine and artesunate/dihydroartemisinin, and these concentrations were no longer adequate once the level of resistance to either mefloquine or artesunate was aggravated. Systemic exposure of both drugs during the acute phase infection was used as a criterion to define adequacy of drug levels. Whole blood mefloquine concentration on day 1 of treatment has been reported to be an important determinant of successful treatment[
34], but there has been no defined threshold levels of artesunate/dihydroartemisinin for treatment of falciparum malaria. Sensitivity of
P. falciparum isolates in this area to mefloquine was still at the resistance level (43.3% of the isolates with recrudescence) after a certain period of improvement[
27]. Mefloquine was used as monotherapeutic treatment of acute uncomplicated falciparum in this area long before the introduction of the combination regimen and thus mefloquine resistance had already reached a level too extreme to protect the development and spread of artesunate resistance. The sensitivity of
P. falciparum to artemisinins would be compromised by intensifying resistance to mefloquine. Decline of
in vitro sensitivity of parasite in this, as well as other areas to artesunate has been demonstrated[
4,
7,
35] and it is noted for a ten-fold decrease in
in vitro artemisinin sensitivity observed over a 10-year period in north-western Thailand[
36]. In a previous study, decreased
in vitro susceptibility to dihydroartemisinin (IC
50 21.2 nM) and artesunate (16.3 nM) was reported in a patient returning from south of Laos and north of Thailand[
26]. In view of the short half-life of oral artesunate on the other hand, it is expected that the drug exerts little drug pressure, provided the treatment generally results in parasitological cure. Compared with western Cambodia and areas along the Thai-Cambodian border, intensity of malaria transmission is relatively low in the north-western border areas of Thailand. This would lead to lower selective drug pressure and drug resistance. Although artemisinin resistance starts to gradually emerge[
37,
38], relatively good parasitologic responses to artemisinins are observed in this border area even after almost 20 years of intensive use[
31].
Analysis of the contribution of in vitro parasite sensitivity and drug concentrations and relationship with pfmdr1 copy number in the group with sensitive response was performed in 21 of 69 cases. Results indicate contribution of both parasite and pharmacokinetic factors in treatment response, but with different magnitude. Pharmacokinetic factor contributed to 71.4% (15 of 21 cases) of variation in mefloquine and/or dihydroartemisinin concentrations in this group of patients, of which 35.3% (six of 17 cases) were due to variable mefloquine pharmacokinetics (alone or together with dihydroartemisinin) and 64.7% (11 of 17 cases) were due to variable artesunate/dihydroartemisinin pharmacokinetics (alone or together with mefloquine). In vitro resistance/reduced sensitivity to mefloquine and aresunate was found in 36.7% (nine of 17 cases) and 43.3% (12 of 17 cases) of all cases, respecrtively. In the group with recrudescence response, pharmacokinetic factor contributed to 58.8% (ten of 17 cases) of variation in melfoquine and/or dihydroartemisinin concentrations; 50% (five of ten cases) were due to variable mefloquine pharmacokinetics (alone or together with dihydroartemisinin) and 50% (five of ten cases) were due to variable artesunate/dihydroartemisinin pharmacokinetics (alone or together with mefloquine). Resistance/reduced sensitivity to mefloquine and aresunate was found in 23.8% (five of 21 cases) and 42.9% (nine of 21 cases) of all cases, respectively. Altogether, these findings may suggest that the influence of pharmacokinetic variability of mefloquine influences more on treatment response than of artesunate/dihydroartemisinin on background of mefloquine resistance together with decreasing susceptibility of parasite to artesunate. Although statistical significance could not be achieved due to a marked variability, the IC50 of both artesunate (medians of 49.9 vs 21.1 nM) and mefloquine (medians of 2.8 vs 1.8 nM) tended to be higher in isolates obtained from patients with recrudescence response. In the two cases with sensitive response (Nos 18 and 20), sensitivity of the parasite to both artesunate and mefloquine is relatively high and even with low systemic exposure of mefloquine, were still adequate to completely eliminate all the parasites. In the other three cases (Nos 31, 32 and 33) with in vitro resistance/reduced sensitivity to mefloquine and artesunate, radical cure could be obtained as far as adequate systemic exposure of both mefloquine and dihydroartemisinin were achieved.
A consensus for the molecular and cellular mechanisms of artemisinin resistance has not emerged. Despite continuous efforts to uncover definite molecular markers for resistance to artemisinins in
P. falciparum, no valid marker has been identified and confirmed, which precludes efficient monitoring of emerging and spreading resistance. Results from various molecular studies varied, probably reflecting a multigenetic nature of artemisinin resistance. Among these, polymorphisms of PfMDR1 (an ATP-binding cassette (ABC) transporter residing on the digestive vacuolar membrane of the parasite, where it transports solutes, including anti-malarial drugs into the digestive vacuole) encoded by the
pfmdr1 gene have received the most attention. Several mutations (N86Y, Y184F, S1034C, N1042D and D1246Y) and copy number variation that occurred in
pfmdr1 from field isolates collected from several endemic areas are associated with altered sensitivity to artemisinins and their combination partner mefloquine[
39‐
42]. Amplification of PfMDR1 copy number has been proposed as a key determinant for both
in vivo and
in vitro resistance to both artemisinins and mefloquine in Cambodia and areas along the Thai-Cambodian and Thai-Myanmar borders[
26,
35,
43‐
47]. Nevertheless, some clinical trials could not confirm the link between PfMDR1 polymorphism/copy variation and artemisinin resistance[
4‐
8]. There appeared to be no association between the mutation or amplification of
pfatp6, the sarco/endoplasmic reticulum Ca
2+-ATPase, in
P. falciparum isolates in Thailand[
38]. In this study, increased copy number of
pfmdr1 was shown to be associated with treatment failure and a decline in
in vitro susceptibility of
P. falciparum isolates in this area following artesunate-mefloquine combination, with approximately 40% of the isolates carrying
pfmdr1 copy number between two and six copies.
Pfmdr1 copy number correlated well with clinical treatment response in both groups of patients with sensitive and recrudescence response. Approximately 70.1% (12 of 17 cases) and 71.4% (15 of 21 cases) in the groups with sensitive and recrudescence response carried gene copy number >1 (two to six) and one, respectively. Resistance to ACT may evolve even when the two drugs within the combination are taken simultaneously and amplification of the
pfmdr1 gene may partly contribute to this phenotype.
Competing interests
The authors have declared that they have no competing interests.
Authors’ contributions
KN and JK conceived and designed the experiments. PM, WC and RR performed the experiments. PM and WC analysed the data. KN and JK wrote the paper. All authors read and approved the final manuscript.