RESEARCH ARTICLE

Efficacy of antimalarial drugs for treatment of

uncomplicated falciparum malaria in Asian
region: A network meta-analysis
Cho Naing ID1,2☯*, Maxine A. Whittaker2☯, Norah Htet Htet ID1, Saint Nway Aye ID1, Joon
Wah Mak1
1 International Medical University, Kuala Lumpur, Malaysia, 2 Faculty of Tropical Heath and Medicine,
James Cook University, Queensland, Australia

a1111111111 ☯ These authors contributed equally to this work.

a1111111111 * cho3699@[Link]
a1111111111
a1111111111
a1111111111
Abstract

Background
OPEN ACCESS The WHO recommends artemisinin-based combination therapies (ACTs) for the treatment
Citation: Naing C, Whittaker MA, Htet NH, Aye SN, of uncomplicated falciparum malaria. Hence, monitoring the efficacy of antimalarial drugs is
Mak JW (2019) Efficacy of antimalarial drugs for a key component of malaria control and elimination. The published randomized trials that
treatment of uncomplicated falciparum malaria in
assessed comparisons of ACTs for treating uncomplicated falciparum malaria reported con-
Asian region: A network meta-analysis. PLoS ONE
14(12): e0225882. [Link] flicting results in treatment efficacy. A network meta-analysis is an extension of pairwise
pone.0225882 meta-analysis that can synthesize evidence simultaneously from both direct and indirect
Editor: Gordon Langsley, Institut national de la treatment comparisons. The objective was to synthesize evidence on the comparative effi-
santé et de la recherche médicale - Institut Cochin, cacy of antimalarial drugs for treatment of uncomplicated falciparum malaria in Asian
FRANCE region.
Received: September 24, 2019

Accepted: November 14, 2019 Methods

Published: December 19, 2019 Relevant randomized trials that assessed efficacy of antimalarial drugs for patients having
uncomplicated falciparum malaria in Asian region were searched in health-related data-
Peer Review History: PLOS recognizes the
benefits of transparency in the peer review bases. We evaluated the methodological quality of the included studies with the Cochrane
process; therefore, we enable the publication of risk of bias tool. Main outcome was treatment success at day 28 as determined by the
all of the content of peer review and author absence of parasiteamia. We performed network meta-analysis of the interventions in the
responses alongside final, published articles. The
trials, and assessed the overall quality of evidence using the GRADE approach.
editorial history of this article is available here:
[Link]
Results
Copyright: © 2019 Naing et al. This is an open
access article distributed under the terms of the Seventeen randomized trials (n = 5043) were included in this network meta-analysis study.
Creative Commons Attribution License, which A network geometry was formed with 14 antimalarial treatment options such as artemether-
permits unrestricted use, distribution, and
lumefantrine (AL), artemisinin-piperaquine, artesunate-amodiaquine, artesunate-meflo-
reproduction in any medium, provided the original
author and source are credited. quine (ASMQ), artesunate-chloroquine, artesunate-mefloquine home treatment, artesu-
nate-mefloquine 2-day course, artesunate plus sulfadoxine-pyrimethamine, chloroquine,
Data Availability Statement: All relevant data are
within the paper and its Supporting Information dihydroartemisinin-piperaquine (DHP), dihydroartemisinin-piperaquine home treatment,
files. dihydroartemisinin-piperaquine 4-day course, dihydroartemisinin-piperaquine and added

PLOS ONE | [Link] December 19, 2019 1 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

Funding: The authors received no specific funding artesunate, sulfadoxine-pyrimethamine. A maximum number of trials included was DHP
for this work. compared to ASMQ (n = 5). In general, DHP had better efficacy than AL at day 28 (DHP vs
Competing interests: The authors have declared AL: OR 2.5, 95%CI:1.08–5.8). There is low certainty evidence due to limited number of stud-
that no competing interests exist. ies and small trials.

Discussion/ Conclusions
The findings suggest the superiority of DHP (3–day course) to AL and other comparator
ACTs are with the overall low/very low quality of evidence judgements. Moreover, one drug
regimen is better than another is only if current drug-resistance patterns are at play. For
example, the AL might be better than DHP in areas where both artemisinin and piperaquine
resistance patterns are prevalent. For substantiation, well-designed larger trials from
endemic countries are needed. In the light of benefit versus harm concept, future analysis
with safety information is recommended.

Introduction
Malaria caused by Plasmodium falciparum is responsible for >90% of malaria cases and almost
all of the malaria deaths worldwide. According to the WHO report, there was no significant
progress in reducing global malaria during the period 2015–2017 [1]. The Global technical
strategy (GTS) 2020 milestones include the elimination of malaria in at least 10 countries that
were malaria endemic in 2015 [2]. In fact, malaria is a preventable and curable disease. The
WHO recommends artemisinin-based combination therapies (ACTs) for the treatment of
uncomplicated malaria caused by P. falciparum. The primary advantage of the combination
therapy is that the artemisinin quickly and drastically reduces the majority of malaria parasites,
and the partner drug clears remaining small number of parasites [3,4]. Thus far, ACTs is the
newest class of antimalarials that are used worldwide including in the Greater Mekong Sub-
region (GMS). There are reports on evidence of artemisinin resistant hotspots in Cambodia,
Thailand and on the Thai-Myanmar border [1,2]. The emergence of falciparum resistance to
artemisinins would not only limit treatment options in the affected areas, but could also com-
promise the management of uncomplicated malaria cases in other areas where ACT is widely
recommended [5]. Containment of parasites developing resistance to anti-malarial drugs is
one of the major goals to progress from malaria control towards elimination [4,5]. Hence,
monitoring the efficacy of antimalarial drugs is a key component of malaria control and subse-
quent elimination.
Five ACTs recommended by WHO for treatment of uncomplicated P. falciparum malaria
are: artemether—lumefantrine (AL), artesunate- amodiaquine (ASAQ), artesunate- meflo-
quine (ASMQ), artesunate plus sulfadoxine-pyrimethamine (ASSP) and dihydroartemisinin-
piperaquine (DHP) [3,5]. Protecting the efficacy of ACTs as the current first- and second-line
treatment for [Link] malaria is one of a top global public health priority [1,3]. However,
the question is which antimalarial drugs offers the greatest benefits (efficacy) for treatment of
uncomplicated P. falciparum malaria at day 28? There are published Cochrane systematic
reviews [6,7] and non-Cochrane systematic reviews/meta-analyses [8,9], assessing head-to-
head comparisons of ACTs for treating uncomplicated [Link] malaria. These reviews
reported conflicting results. For example, a review by Zani and associates reported that in
Africa, there was better efficacy in DHP than AL at day 28. However, such relationship was not

PLOS ONE | [Link] December 19, 2019 2 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

shown in Asia [7]. This reflects that efficacy of antimalarial is related to whether the area is
with artemisinin-sensitive parasite populations or not [7,9]. Hence, we performed a network
meta-analysis (NMA) that can synthesize evidence simultaneously from both direct and indi-
rect treatment comparisons [10]. For instance, even when no head-to-head trial is available,
studies evaluating A versus B and B versus C can be used to compare A and C indirectly
through the NMA approach. Indirect comparisons must be connected by at least one common
comparator (i.e. treatment B in this example). An assumption required for the NMA is ‘transi-
tivity’ that trials should be comparable in all characteristics [11]. On the whole, the objective of
present study was to synthesize evidence on the comparative efficacy of currently used antima-
larial drugs for treatment of uncomplicated falciparum malaria in Asian region.

Materials and methods

The current study adhered to the preferred reporting items for network meta-analyses (PRIS-
MA-NMA) [12] (S1 Table).

Search strategy
We searched relevant studies in the health-related electronic database including MEDLINE,
Medline-in-Process, OLD Medline, EMBASE, and the Cochrane library using the texts includ-
ing “malaria” “Artemisinin-based combination therapy”, “randomized trial” “humans” with
Boolean operators. The search was done according to guidance provided in the Cochrane
Handbook for Systematic Reviews of Interventions [13] and in consultation with an informa-
tion specialist. The search strategies for the MeSH terms are listed in S2 Table. Additionally,
we searched in, [Link], EU Clinical Trials Register and WHO International Clinical
Trials Registry Platform for ongoing trials. Search was limited to studies published in English
language until June 2019.

Selection of study
Eligible studies were identified through the PICOS format [13].
Study Population (P): Participants having uncomplicated falciparum malaria residing in
Asian region, regardless of gender and age were included.
Uncomplicated malaria caused by the P. falciparum parasite in this study is defined as
patients having symptoms that are non-specific in the presence of P. falciparum, but in the
absence of clinical or laboratory findings of severe organ dysfunction.
An operational definition of the Asian region for this particular study covers countries in
three regions of Southeast Asia, South Asia and East Asia.
Interventions (I): Anti-malarial drugs for the treatment of uncomplicated falciparum
malaria were considered. Different dosages of an antimalarial regimen were considered as
individual treatments.
Comparisons (C): An alternative antimalarial drug or placebo were included.
Study Outcomes (O): Main outcome was cure rate at day 28 (defined as the proportion of
patients with clearance of asexual P. falciparum parasitaemia within seven days of initiation of
trial drug, without subsequent recrudescence within 28 days after initiation of study). Recru-
descence was defined as the existence of positive blood smears after initial clearance of para-
sites from the peripheral blood [14].
Study design (S): Randomised clinical trials (RCT), conducted in Asian region.
Studies were excluded, if they did not meet the inclusion criteria. Studies on pregnant
women and travellers were not considered.

PLOS ONE | [Link] December 19, 2019 3 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

Data extraction and management

One investigator (SNA) screened title and abstracts on the basis of RCTs that assessed human
falciparum malaria. The same investigator extracted information from the RCTs included.
Information collected were study characteristics, intervention and comparators and outcomes.
Information collected were cross-checked by another investigator (CN). Any discrepancies
were settled by discussion.

Methodological quality assessment

The methodological quality of the RCTs was evaluated using the Cochrane risk of bias tool
[13]. Three domains (adequate sequence generation, allocation concealment and blinding of
participants and outcome assessors) were assessed for the risk of bias assessment for each trial.
The ratings were noted (i.e. high risk, unclear risk, low risk) for the risk of bias category in sub-
sequent Grading of Recommendations Assessment, Development, and Evaluation (GRADE)
assessment. [15].

Data synthesis
Main outcome in this review was treatment success (cure rate) by anti-malarial treatment at
day 28. We preferred the intention-to-treat (ITT) analysis over another analysis, whenever it
was available. We performed pair-wise meta-analyses of all available within-study compari-
sons, followed by subsequent network meta-analyses.
Pairwise comparison: When the studies reported in similar ways, we did head-to-head
comparisons as a direct pairwise meta-analyses. An odds ratio (OR) and its 95%confidence
interval (CI) were computed for the dichotomous variables. Between-study heterogeneity was
assessed with the I2 statistic. We pooled ORs with a DerSimonian-Laird random-effects model
in the presence of substantial heterogeneity (I2>50%). Publication bias was investigated with
the contoured-enhanced funnel plot [13].
Network meta-analyses: We performed NMA within a frequentist framework using ran-
dom-effects models [16–18]. We established network mapping. An assumption of NMA is
‘transitivity’ that the trials comparing different sets of interventions should be similar enough
in their characteristics [11]. We also investigated another assumption of NMA such as network
‘inconsistency’ (i.e. disagreement between the different sources of evidence) with the use of the
global Wald test for inconsistency [19,20]. We also checked, if there were concerns with
‘intransivity’ [21]. For a ranking of the effectiveness, we reported ‘Surface Under the Cumula-
tive Ranking Curve’ (SUCRA) [11, 18]. SUCRA = 1 or 0 was indicated the rank of an interven-
tion drug as first or last, respectively. Statistical significance was set at p value �0.05.

Assessing the quality of evidence

We assessed the quality of evidence derived from the pairwise and NMA, following the GRADE
approach, as described elsewhere [15,17,21,22]. For direct comparison, we rated evidence on the
five categories; study limitations (risk of bias), precision, consistency of results, directness of evi-
dence and publication bias, using the standard GRADE approach. We then evaluated the overall
confidence in estimates of effect for treatment efficacy for each direct comparison as ‘high’, ‘mod-
erate’, ‘low’ or ‘very low’ quality of evidence. For indirect comparison, we rated evidence from the
most dominant first-order loop by first taking the lowest certainty of direct comparisons. We did
not rate on intransitivity [21] in the absence of important imbalance in the distribution of effect
modifiers (e.g. age, gender) across included trials. For NMA mixed estimates, we started with the
higher quality of the two certainty ratings and rated down certainty for incoherence (degree of

PLOS ONE | [Link] December 19, 2019 4 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

inconsistency between direct and indirect effect estimates) in the final quality rating [21,22]. Data
analysis was employed with STATA 15.0 (StataCorp, TX).

Results
Trials included
Fig 1 illustrates the study selection process. The initial search produced 13640 hits. After
removal of duplicates and screening of titles and abstracts, 34 full-text papers were evaluated
and 17 studies, incorporating 5043 total number of patients were finally selected for this review
[14, 23–38]. The highest number of trials include DHP compared to ASMQ (n = 5), followed
by artemisinin-piperaquine (AMPQ) to DHP (n = 3) and AL to ASMQ (n = 2). A summary of
the 17 excluded studies is provided in S3 Table.
Table 1 provides the key characteristics of the studies identified. The number of participants
varied from 47 [14] to 769 [36]. The majority of participants in the studies included were
males (range from 51% to 96%) with mean age between 5.9 to 29 years. The distribution of
studies is presented in S4 Table. Three studies were three-arm RCTs [23,24,30] and one study
was a four-arm RCT [29] and the remaining 13 studies were two-arm RCT. Studies were con-
ducted in 8 countries such as Cambodia, India, Indonesia, Laos, Myanmar, Nepal, Thailand
and Vietnam. One multi-country study was conducted in India, Laos and Thailand [36].
The number of studies with unclear risk of bias in sequence generation (53%) and that with
high risk of bias in the blinding status was 58.8%. Overall risk of bias assessment revealed that
most studies had an unclear/high risk of bias due to insufficient information on allocation con-
cealment and the blinding status of the RCTs (S5 Table).

Fourteen-node analysis
Fig 2 shows a network plot of treatment success with 14 antimalarial treatment options. These
options of antimalarial regimens included AL, AMPQ, ASAQ, artesunate plus chloroquine
(ASCQ), ASMQ, artesunate-mefloquine home treatment/not supervised (ASMQh), artesunate-
mefloquine 2-day course (ASMQ2), ASSP, chloroquine (CQ), DHP, dihydroartemisinin-piper-
aquine home treatment/not supervised (DHPh), dihydroartemisinin-piperaquine 4-day course
(DHP4), dihydroartemisinin-piperaquine and added artesunate (DHPAS), sulfadoxine-pyri-
methamine (SP). As seen in the present network map, none of the studies compared ASSP and
CQ directly, but each had been compared with a common comparator AL. We may assume an
indirect comparison of ASSP and CQ on the direct comparison of ASSP and AL and the direct
comparison of CQ and AL. Fig 3 presents forest plot with effects for each study, estimates from
direct pairwise meta-analysis and mixed estimate from the network meta-analysis.
Pairwise-analysis of the relative efficacy of antimalarial drugs for treating uncomplicated P.
falciparum malaria was reported that there were comparable cured rates between the treatment
regimens, spanning both benefit and harm, except one comparison (i.e. AL versus CQ) (S1
Fig). For instance, DHP compared to AL in a single trial and there was a superiority of DHP in
cure rate at day 28 (OR 2.5, 95%CI: 1.08 to 5.8) (Table 2). The results of the network meta-
analysis are presented in Fig 4. In general, DHP was better than many comparators in terms of
efficacy at day 28. For instance, DHP was superior to ASCQ (OR: 11.21,95% CI 3.4–36.89). Of
note is that there was small number of studies in many comparisons and 95%CIs were (very)
wide. For instance, DHP versus AL was done in a single study.
The global Wald test showed the presence of consistency in the network [Chi2 (4) = 3.02,
p = 0.8089] (Fig 3). Tests of local incoherence did not show any inconsistent loops for efficacy
at day 28 (S2 Fig). The comparison adjusted funnel plots of the network meta-analysis [39] for
efficacy at day 28 was not suggestive for publication bias (S3 Fig).

PLOS ONE | [Link] December 19, 2019 5 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

Fig 1. Study selection process.

[Link]

Treatment relative ranking

Treatment-relative ranking in this network meta-analysis is presented in (S6 Table). DHP had
the highest probability of being the best choice for treating patients with uncomplicated P. fal-
ciparum (S4 Fig). The SUCRA and ranking results are subjected to the small number of partic-
ipants in some studies and wide estimates spanning from benefit to harm. As the evidence on
which the SUCRA rankings are of very low quality, they are untrustworthy [40]. Predictivity
intervals of mixed estimates are presented in Fig 5. Although their confidence intervals suggest
an association, the respective predictive interval crosses the line of no effect and suggests that
future studies might favour either treatment. We therefore made overall evidence through the
GRADE approach rather than the SUCRA rankings (Table 2).
Overall, we observed a low certainty evidence whether any antimalarial regimens included
in this study were better in clearance of parasitemia at day 28 since the certainty of the evi-
dence was assessed as low.

PLOS ONE | [Link] December 19, 2019 6 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

Table 1. Characteristics of the studies included in the network meta-analysis.

Author, Study Country Setting Study Interventions Participants Male Age, PCR Under Funding
Year of period arms mean in year supervision
publication (±SD) or
[[Link]] median and
range
1 Rachmawati, 1/2009 Indonesia. H 2 AL, 47 59.5% 5.9 ±3.46 No Yes Not reported
2010 [14] - 7/ (in-pt) ASSP
2009
2 Ashley, 7/ Thailand H 3 ASMQ, DHP, 731 78.1% Adults No Yes Holleykin
2004 [23] 2002– (in-pt) DHPAS (25.3± 8.2) Pharmaceutical
4/2003
3 Ashley, 4/ Thailand OPD 3 ASMQ, DHP, 499 60.5% Any age Yes Yes MMV;
2005 2003– DHP4 (21, Wellcome
[24] 4/ 2004 3–57) -Mahidol;
Wellcome Great
Britain.
4 Kshirsaga, 6/ India H 2 AL, CQ 179 96% 29 Yes Yes
2000 1996– (in-pt) (17–66)
[25] 1/1997
5 Lefevre, 9/ Thailand H 2 AL, 219 70% 50 Yes Yes Novartis
2001 1998- ASMQ (12–71) Pharma AG.
[26] 1/1999
6 Huong, NA Vietnam H 2 ASSP, ASCQ 123 51% Any age, Yes Yes male% & age in
2003 (in-pt) 10.3 ± 11.3 the ASSP gr
[27] (4–65)
7 Silachamroon, NA Thailand H 2 ASMQ, 120 70.8% Adults No Yes WHO/RBM/
2005 (in-pt) ASMQ2 (25.6± 10.1) Mahidol
[28] University
8 Smithuis, 11/ Myanmar OPD 4 ASMQ, 652 52% 3 age-gr; Yes Yes (Gr1 MSF (Holland);
2006 2003 - ASMQh, DHP, 58.2%(5–14 No (Gr 2)
[29] 4/2004 DHPh yr)
9 Song, 7/2005 Cambodia H 3 AL, 220 73% 3 age-gr; 80% Yes Yes Science &
2011 - (in-pt) AMPQ,DHP (>15 yr). Technology
[30] 10/ Planning Project,
2005 MOST/China
10 Thanh, 9/2006 Vietnam Health 2 DHP, 116 63.8% Any age; Yes Yes People’s Army
2009 - station AMPQ (20.6± 12.4) Department of
[31] 12/ Military Medicine
2007.
11 Thanh, 5/ Vietnam Commune 2 ASAQ, DHP 128 70.1% Any age; Vietnam
2012 2008– centre (18.9± 12.7) People’s Army
[32] 12/ Department of
2009, Military Medicine

12 Thapa, 8/ Nepal H 2 AL, SP 99 53% >5 yr; Yes Yes Not reported
2007 2005– (in-pt) (AL); (26.5 ± 13.8)
[33] 10/ 73%
2005. (SP)
13 Tjitra, 2007– Indonesia 4 Hs 2 ASSP, SP 105 60% 83.8% Yes Yes Nicholson-Hill
2001 2008 (under 12 yr) Malaria Research
[34] Fund & Tudor
Foundation.
14 Trung, NA Vietnam, treatment 2 DHP, AMPQ 103 61.2% 25.8±13.9 Yes Yes Science and
2009 center (in- Technology
[35] pt) Research Projects
of
Guangdong
Province
(Continued )

PLOS ONE | [Link] December 19, 2019 7 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

Table 1. (Continued)

Author, Study Country Setting Study Interventions Participants Male Age, PCR Under Funding
Year of period arms mean in year supervision
publication (±SD) or
[[Link]] median and
range
15 Valecha, NA Multi OPD 2 DHP, ASMQ 1150 78.4% mainly adults, Yes Yes MMV,
2010 country (25.46± Sigma Tau. &
[36] (India, 13.3) Oxford University
Laos,
Thailand)
16 van Vgt,2000 11/ Thailand H & health 2 AL, 200 73.5% Adults & Wellcome Trust
[37] 1997- camp ASMQ children of Great Britain
3/1998 (23, 13–63)
17 Wilairatana,2002 ?? Thailand H 2 DHP, 352 66.8% 24.8 No Yes Tonghe
[38] (in-pt) ASMQ (±13.3) Phramaceutical
Co. Ltd

AL: Artemether-lumefantrine; AMPQ; artemisinin-piperaquine; ASMQ: artesunate-mefloquine; ASMQh: artesunate-mefloquine home treatment/not supervised;
ASMQ2: artesunate-mefloquine 2-day course; ASAQ; artesunate-amodiaquine; ASCQ: artesunate-chloroquine; ASSP: artesunate plus sulfadoxine-pyrimethamine; CQ:
chloroquine; DHP: dihydroartemisinin-piperaquine; DHP4: dihydroartemisinin-piperaquine 4-day course; DHPh; dihydroartemisinin-piperaquine home treatment/
not supervised; DHPAS dihydroartemisinin-piperaquine & artesunate added; SP: sulfadoxine-pyrimethamine; gr: group(s); H; hospital; In-pt: Inpatients; MMV:
Medicines for Malaria Venture; MOST/China: Ministry of Science and Technology of the People’s Republic of China; MSF: Medecins Sans Frontieres; OPD: outpatient
department/centre; WHO/RBM: World Health Organization/Roll Back Malaria’ yr: year.

[Link]

Discussion
Summary of main results
The present network meta-analysis, including 14 different antimalarial interventions from 17
RCTs studies, provided both direct and indirect evidences regarding the relative efficacy at the
end of 28-day follow-up time. This approach provided both direct and indirect information
through the use of a common comparator to obtain estimates of the relative effects on multi-
ple-intervention comparisons. To our knowledge, this is the first comprehensive synthesis of
data from available antimalarial interventions for treatment of uncomplicated falciparum
malaria in Asia.
If parasite populations are not completely eradicated from the patient, low-level replication
continues until it reaches the microscopic or clinical detection threshold to cause a treatment
failure (i.e. recrudescent infection). The parasitological cure rates at day 28 is potentially a
more sensitive marker for in vivo efficacy of an antimalarial drug compared to the cure rates at
day 14. This is particularly for drugs with long half-lives in plasma. But, it is also applicable to a
certain me extent to drugs with short plasma half-lives. Therefore, the end point of PCR-cor-
rected cure rate at day 28 in the current analysis is an acceptable outcome measure to increase
the sensitivity of the test as well as to generate appropriate data for a comparison of the differ-
ent treatment regimens (artesunate alone versus the combination of artesunate and AMQ)
[41]. Overall, the results of this NMA provided low quality evidence that no regimen could
provide better rates of treatment success, except DHP.
Artemisinin resistance to parasites have shown reduced susceptibility and is clearly associ-
ated with increasing rate of failure of ACT in Cambodia [42] and Thailand [43]. K13-propeller
mutation was identified as a key determinant of artemisinin resistance in Southeast Asia [44].
Many administrative regions in Myanmar had combined K13-propeller mutation prevalence
of more than 20% [45], including regions on the Myanmar India border areas.

PLOS ONE | [Link] December 19, 2019 8 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

Fig 2. Network plot of the antimalarials for treating P. falciparum malaria.

[Link]

ACT is recommended to be given for three days along with slowly eliminated partner drug.
In a three-day regimen, the artemisinin component stays in body for two asexual parasite life-
cycles, except for P. malariae. In each asexual cycle, artemisinin and its derivatives reduce par-
asite counts by a factor of almost 10,000 [47,48]. The current NMA showed that there is a very
low-certainty evidence that DHP (3-day course) was superior to other ACTs (ASAP, AL,
ASAQ, ASMQ). A published NMA of antimalarial treatments for uncomplicated falciparum
malaria in African children that included 12 RCTs showed superiority of DHP among cur-
rently WHO recommended ACTs [46], but it did not report an overall quality of evidence.
The utility of DHP is an important information in drug compliances as AL has to be adminis-
tered twice daily for three days and it also need to take fatty food to be effective. Moreover, the
treatment cost of AL is also more expensive (>10 US$ per treatment course) [46].

Study limitations
Studies in other languages may have been missed, if abstracts in English language are not
available. Future study, addressing the cost-effectiveness of particular antimalarial drug

PLOS ONE | [Link] December 19, 2019 9 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

Fig 3. All direct and mixed comparisons.

[Link]

Table 2. GRADE quality of evidence for the comparative efficacy of antimalarial.

Treatment Direct estimate; OR (95% CI) Quality of Indirect estimate; Quality of Network estimate; OR Quality of
comparison evidence OR (95% CI) evidence (95% CI) evidence
DHP vs AL 1.29 ⊕◯◯◯d,e 2.5 ⊕⊕◯◯a 2.5 ⊕⊕◯◯a
(0.74 to 2.23) VERY LOW (1.08 to 5.8) LOW (1.08 to 5.8) LOW
1 fewer per 1,000 3 fewer per 1,000 3 fewer per 1,000
(from 2 fewer to 1 fewer) (from 6 fewer to 1 fewer) (from 6 fewer to 1 fewer)
DHP vs ASCQ NA NA 11.21 ⊕◯◯◯ 11.21 (3.4–36.89) ⊕◯◯◯a,b,c
(3.40 to 36.89) VERY LOW 11 fewer per 1,000 VERY LOW
(from 37 fewer to 3 fewer)
DHP vs CQ NA NA 16.54 ⊕⊕◯◯ 16.54 ⊕⊕◯◯a,b
(5.02–54.56) LOW (5.02 to 24.56) LOW
17 fewer per 1,000
(from 25 fewer to 5 fewer
DHP vs ASSP NA NA 0.01 (0.00–0.04) ⊕⊕◯◯ 0.01 ⊕⊕◯◯a,b
LOW (0.00 to 0.04) LOW
0 fewer per 1,000
(from—to 0 fewer)
ASCQ vs AL NA NA 0.22 (0.06–0.79) ⊕⊕◯◯ 0.22 (0.06–0.79) ⊕⊕◯◯a
LOW 0 fewer per 1,000 LOW
(from 1 fewer to 0 fewer)
ASAQ vs AL NA NA 5.55 (0.26–119.75) ⊕◯◯◯ 5.55 (0.26–119.75) ⊕◯◯◯a,b,d
VERY LOW 6 fewer per 1,000 VERY LOW
(from 120 fewer to 0 fewer)

CI: Confidence interval; OR: Odds ratio; Explanations: a. studies at unclear and high risk of bias; b. wide predictive interval; c. very wide CI; d. wide 95%CI and it
crossed a null value; e: a singular study at high risk of bias; f. a singular study

[Link]

PLOS ONE | [Link] December 19, 2019 10 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

Fig 4. Network meta-analysis of antimalarial treatments.

[Link]

intervention is needed. Some RCTs included were done with small sample size and they also
had low methodological quality. Hence, there was a concern to the confidence in the effect esti-
mates. Due to lack of blinding in some RCTs included, the risk of performance bias is a con-
cern. Myanmar has substantially more malaria than any other country in the Southeast Asia
[45]. The current analysis included only one RCT from Myanmar, indicating a limited geo-
graphical representativeness and an interpretation of the findings was limited with regard to
generalizability.

Implications
There is evidence that DHP has better treatment outcome than monotherapy such as SP and
CQ, which are almost completely ineffective in Southeast Asia due to drug resistance.
Evidence from RCTs may not apply to real-world practice, where people in need of antima-
larial treatment are often limited to compliance due to unsupervised treatment and lack of
monitoring. Further studies should be directed to detect the effectiveness in real-world prac-
tice and should also focus on monitoring and treatment regimens. Moreover, in the light of
benefit versus harm concept, a NMA of the relative safety of different antimalarial treatment is
needed.

PLOS ONE | [Link] December 19, 2019 11 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

Fig 5. Predictive intervals plot for the antimalarial network.

[Link]

Conclusions
The findings suggest the superiority of DHP to AL and other comparator ACTs are with the
overall low/very low quality of evidence judgements. Moreover, one drug regimen is better
than another is only, if current drug-resistance patterns are at play. For example, the AL might
be better than DHP in areas where both artemisinin and piperaquine resistance patterns are
prevalent [7,9]. For substantiation, well-designed larger trials from endemic countries are
needed. In the light of benefit versus harm concept, future analysis with safety information is
recommended.

Supporting information
S1 Table. PRISMA NMA checklist.
(PDF)
S2 Table. Search terms.
(PDF)

PLOS ONE | [Link] December 19, 2019 12 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

S3 Table. Excluded studies and reasons for exclusion.

(PDF)
S4 Table. Distribution of studies and comparisons.
(PDF)
S5 Table. Risk of bias assessment by the review authors.
(PDF)
S6 Table. Treatment relative ranking.
(PDF)
S1 Fig. Forest plot of direct pairwise comparison of antimalarial regimens.
(PDF)
S2 Fig. Inconsistency plot with loop-specific heterogeneity.
(PDF)
S3 Fig. Comparison-adjusted funnel plot of the placebo-controlled antimalarial trials.
(PDF)
S4 Fig. Cumulative probability curves for the antimalarial network.
(PDF)

Acknowledgments
The authors thank the patients and researchers of the primary studies. We also thank our insti-
tutions for allowing us to perform this study. We also thank the anonymous reviewers and edi-
tors for giving us the comments and valuable inputs to improve the quality of our manuscript.

Author Contributions
Conceptualization: Maxine A. Whittaker, Joon Wah Mak.
Data curation: Cho Naing, Norah Htet Htet, Saint Nway Aye.
Formal analysis: Cho Naing, Norah Htet Htet, Saint Nway Aye.
Funding acquisition: Joon Wah Mak.
Investigation: Cho Naing.
Methodology: Cho Naing, Maxine A. Whittaker, Norah Htet Htet, Joon Wah Mak.
Project administration: Cho Naing, Joon Wah Mak.
Resources: Joon Wah Mak.
Supervision: Joon Wah Mak.
Writing – original draft: Cho Naing, Maxine A. Whittaker.

References
1. WHO. World malaria report 2018. Geneva: World Health Organization; 2018.
2. WHO. Global technical strategy for malaria 2016–2030. Available at [Link]
global_technical_strategy/en/
3. WHO. Antimalarial drug efficacy and drug resistance. 2018. Available at [Link]
areas/treatment/drug_efficacy/en/

PLOS ONE | [Link] December 19, 2019 13 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

4. WHO. Guidelines For The Treatment of Malaria - 3rd edition. 2015. Available at [Link]
iris/bitstream/handle/10665/162441/9789241549127_eng.pdf?sequence=1
5. Satimai W, Sudathip P, Vijaykadga S, Khamsiriwatchara A, Sawang S, Potithavoranan T, et al. Artemi-
sinin resistance containment project in Thailand. II: responses to mefloquine-artesunate combination
therapy among falciparum malaria patients in provinces bordering Cambodia. Malar 2012; 11:300.
6. Bukirwa H, Unnikrishnan B, Kramer CV, Sinclair D, Nair S, Tharyan P. Artesunate plus pyronaridine for
treating uncomplicated Plasmodium falciparum malaria. Cochrane Database Syst Rev. 2014;(3):
CD006404. [Link] PMID: 24596021
7. Zani B, Gathu M, Donegan S, Olliaro PL, Sinclair D. Dihydroartemisinin-piperaquine for treating uncom-
plicated Plasmodium falciparum malaria. Cochrane Database Syst Rev. 2014;(1):CD010927. https://
[Link]/10.1002/14651858.CD010927 PMID: 24443033
8. Naing C, Racloz V, Whittaker MA, Aung K, Reid SA, Mak JW, et al. Efficacy and safety of dihydroartemi-
sinin-piperaquine for treatment of Plasmodium vivax malaria in endemic countries: meta-analysis of
randomized controlled studies. PLoS One. 2013; 8(12):e78819. [Link]
0078819 PMID: 24312446
9. WWARN Parasite Clearance Study Group. Baseline data of parasite clearance in patients with falcipa-
rum malaria treated with an artemisinin derivative: an individual patient data meta-analysis. Malar J.
2015; 14:359. [Link] PMID: 26390866
10. Lu G. and Ades A.E. Combination of direct and indirect evidence in mixed treatment comparisons. Stat
Med. 2004; 23: 3105–24. [Link] PMID: 15449338
11. Salanti G. Indirect and mixed-treatment comparison, network, or multiple-treatments meta-analysis:
many names, many benefits, many concerns for the next generation evidence synthesis tool. Res
Synth Methods. 2012; 3:80–97. [Link] PMID: 26062083
12. Cornell JE. The PRISMA extension for network meta-analysis: bringing clarity and guidance to the
reporting of systematic reviews incorporating network meta-analyses. Ann Intern Med. 2015; 162:797–
8. [Link] PMID: 26030637
13. Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version
5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. [Link].
14. Rachmawati R, Rampengan N, Tatura S, Rampengan T. Comparison of the efficacy of artemether-
lumefantrine vs. artesunate plus sulfadoxine-pyrimethamine in children with uncomplicated falciparum
malaria. PI.2010; 50(2):113 Available at [Link]
indonesiana/article/view/1216
15. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, et al. GRADE: an emerging
consensus on rating quality of evidence and strength of recommendations. BMJ. 2008; 336(7650):924–
6. [Link] PMID: 18436948
16. Salanti G, Ades AE, Ioannidis JPA. Graphical methods and numerical summaries for presenting results
from multiple-treatment meta-analysis: an overview and tutorial. J Clin Epidemiol. 2011; 64:163–71.
[Link] PMID: 20688472
17. Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R, Brozek J, et al. GRADE guidelines: 3.
Rating the quality of evidence. J Clin Epidemiol. 2011; 64(4):401–6. [Link]
2010.07.015 PMID: 21208779
18. Shim S, Yoon B-H, Shin I-S, Bae J-M. Network meta-analysis: application and practice using Stata. Epi-
demiol Health. 2017; 39:e2017047. [Link] PMID: 29092392
19. White IR, Barrett JK, Jackson D, Higgins JPT. Consistency and inconsistency in network meta-analysis:
model estimation using multivariate meta-regression. Res Synth Methods 2012; 3(2):111–25. https://
[Link]/10.1002/jrsm.1045 PMID: 26062085
20. Dias S, Welton NJ, Caldwell DM, Ades AE. Checking consistency in mixed treatment comparison meta-
analysis. Stat Med 2010; 29(7–8):932–44. [Link] PMID: 20213715
21. Brignardello-Petersen R, Bonner A, Alexander PE, Siemieniuk RA, Furukawa TA, Rochwerg B, et al.
Advances in the GRADE approach to rate the certainty in estimates from a network meta-analysis. J
Clin Epidemiol 2018; 93:36–44. [Link] PMID: 29051107
22. Salanti G, Del Giovane C, Chaimani A, Caldwell DM, Higgins JPT. Evaluating the quality of evidence
from a network meta-analysis. PLoS One 2014; 9(7):e99682. [Link]
0099682 PMID: 24992266
23. Ashley EA, Krudsood S, Phaiphun L, Srivilairit S, McGready R, Leowattana W et al. Randomized con-
trolled dose-optimization studies of dihydroartemisinin-piperaquine for the treatment of uncomplicated
multidrug-resistant falciparum malaria in Thailand. J Infect Dis 2004; 190:1773–82. [Link]
1086/425015 PMID: 15499533

PLOS ONE | [Link] December 19, 2019 14 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

24. Ashley EA, McGready R, Hutagalung R, Phaiphun L, Slight T, Proux S, et al. A Randomized controlled
study of a simple once-daily regimen of dihydroartemisinin-piperaquine for the treatment of uncompli-
cated, multidrug-resistant falciparum malaria. Clin Infect Dis 2005; 41(4):425–32. [Link]
1086/432011 PMID: 16028147
25. Kshirsagar NA, Gogtay NJ, Moorthy NS, Garg MR, Dalvi SS, Chogle AR, et al. A randomized, double
blind, parallel group, comparative safety, and efficacy trial of co-artemether versus oral chloroquine in
the treatment of acute uncomplicated Plasmodium falciparum malaria in adults in India. Am J Trop Med
Hyg. 2000; 62:402–8. [Link] PMID: 11037786
26. Lefevre G, Looareesuwan S, Treeprasertsuk S, Krudsood S, Silachamroon U, Gathmann et al. A clini-
cal and pharmacokinetic trial of six doses of artemether-lumefantrine for multidrug-resistant Plasmo-
dium falciparum malaria in Thailand. Am J Trop Med Hyg, 2001; 64:247–56
27. Houng NM, Davis TM, Cox-Singh J, Hewitt S, Tran QT, Tran BK, et al. Treatment of uncomplicated fal-
ciparum malaria in southern Vietnam: can chloroquine or sulfadoxine-pyrimethamine be reintroduced in
combination with artesunate? Clin Infect Dis 2003; 37(11):1461–6. [Link]
PMID: 14614668
28. Silachamroon U, Krudsood S, Thanachartwet W et al. An open, randomized trial of three-day treatment
with artesunate combined with a standard dose of mefloquine divided over either two or three days, for
acute, uncomplicated falciparum malaria. Southeast Asian J Trop Med Public Health 2005; 36: 591–6.
PMID: 16124422
29. Smithuis F, Kyaw MK, Phe O, Aye KZ, Htet L, Barends M, et al. Efficacy and effectiveness of dihydroar-
temisinin-piperaquine versus artesunate-mefloquine in falciparum malaria: an open-label randomised
comparison. Lancet 2006; 367(9528):2075–85. [Link] PMID:
16798391
30. Song J, Socheat D, Tan B, Seila S, Xu Y, Ou F, et al. Randomized trials of artemisinin-piperaquine,
dihydroartemisinin-piperaquine phosphate and artemether-lumefantrine for the treatment of multi-drug
resistant falciparum malaria in Cambodia-Thailand border area. Malar J 2011; 10(1):231.
31. Thanh NX, Trung TN, Phong NC, Thien NX, Dai B, Shanks GD, et al. Open label randomized compari-
son of dihydroartemisinin-piperaquine and artesunate-amodiaquine for the treatment of uncomplicated
Plasmodium falciparum malaria in central Vietnam. Trop Med Int Health. 2009; 14(5):504–11. https://
[Link]/10.1111/j.1365-3156.2009.02269.x PMID: 19320869
32. Thanh NX, Trung TN, Phong NC, Quang HH, Dai B, Shanks GD, et al. The efficacy and tolerability of
artemisinin-piperaquine (Artequick®) versus artesunate-amodiaquine (Coarsucam™) for the treatment
of uncomplicated Plasmodium falciparum malaria in south-central Vietnam. Malar J 2012; 11:217.
[Link] PMID: 22741618
33. Thapa S, Hollander J, Linehan M, Cox-Singh J, Bista MB, Thakur GD, et al. Comparison of artemether-
lumefantrine with sulfadoxine-pyrimethamine for the treatment of uncomplicated falciparum malaria in
eastern Nepal. Am J Trop Med Hyg. 2007; 77(3):423–30. PMID: 17827354
34. Tjitra E, Suprianto S, Currie BJ, Morris PS, Saunders JR, Anstey NM. Therapy of uncomplicated falcipa-
rum malaria: a randomized trial comparing artesunate plus sulfadoxine-pyrimethamine versus sulfadox-
ine-pyrimethamine alone in Irian Jaya, Indonesia. Am J Trop Med Hyg. 2001; 65(4):309–17. [Link]
org/10.4269/ajtmh.2001.65.309 PMID: 11693875
35. Trung TN, Tan B, Van Phuc D, Song JP. A randomized, controlled trial of artemisinin-piperaquine vs
dihydroartemisinin-piperaquine phosphate in treatment of falciparum malaria. Chin J Integr Med. 2009;
15(3):189–92. [Link] PMID: 19568711
36. Valecha N, Srivastava P, Mohanty SS, Mittra P, Sharma SK, Tyagi PK, et al. Therapeutic efficacy of
artemether-lumefantrine in uncomplicated falciparum malaria in India. Malar J. 2009; 8(1):107.
37. van Vugt M, Looareesuwan S, Wilairatana P, McGready R, Villegas L, et al. Artemether-lumefantrine
for the treatment of multidrug resistant falciparum malaria. Trans R Soc Trop Med Hyg 2000; 94: 545–
8. [Link] PMID: 11132386
38. Wijeyaratne PM, Chand PB, Valech N, Shahi B, Adak T, Ansari MA, et al. 2005. Therapeutic efficacy of
antimalarial drugs along the eastern Indo-Nepal border: a cross-border collaborative study. Trans R
Soc Trop Med Hyg 99: 423–9. [Link] PMID: 15837354
39. Chaimani A, Higgins JPT, Mavridis D, Spyridonos P, Salanti G. Graphical tools for network meta-analy-
sis in STATA. PLoS ONE 2013; 8(10): e76654. [Link] PMID:
24098547
40. Naing C, Poovorawan Y, Tong KS. Comparative effectiveness of anti-viral drugs with dual activity for
treating hepatitis B and HIV co-infected patients: a network meta-analysis. BMC Infect Dis. 2018; 18
(1):564. [Link] PMID: 30428847

PLOS ONE | [Link] December 19, 2019 15 / 16

 A network meta-analysis of treatment efficacy for uncomplicated falciparum malaria

41. Borrmann S, Adegnika AA, Missinou MA, Binder RK, Issifou S, Schindler A, et al. Short-course artesu-
nate treatment of uncomplicated Plasmodium falciparum malaria in Gabon. Antimicrob Agents Che-
mother 2003; 47(3):901–4. [Link] PMID: 12604519
42. Leang R, Barrette A, Bouth DM, Menard D, Abdur R, Duong S, et al. Efficacy of dihydroartemisinin-
piperaquine for treatment of uncomplicated Plasmodium falciparum and Plasmodium vivax in Cambo-
dia, 2008 to 2010. Antimicrob Agents Chemother 2013; 57(2):818–26. [Link]
00686-12 PMID: 23208711
43. Carrara VI, Lwin KM, Phyo AP, Ashley E, Wiladphaingern J, Sriprawat K, et al. Malaria burden and arte-
misinin resistance in the mobile and migrant population on the Thai–Myanmar border, 1999–2011: An
observational study. PLoS Med 2013; 10(3):e1001398. [Link]
PMID: 23472056
44. Straimer J, Gnadig NF, Witkowski B, Amaratunga C, Duru V, Ramadani AP, et al. Drug resistance.
K13-propeller mutations confer artemisinin resistance in Plasmodium falciparum clinical isolates. Sci-
ence. 2015; 347(6220):428–31. [Link] PMID: 25502314
45. Hay SI, Okiro EA, Gething PW, Patil AP, Tatem AJ, Guerra CA, et al. Estimating the global clinical bur-
den of Plasmodium falciparum malaria in 2007. PLoS Med 2010; 7(6):e1000290. [Link]
1371/[Link].1000290 PMID: 20563310
46. Youdom SW, Tahar R,Basco LK. Comparison of anti-malarial drug efficacy in the treatment of uncom-
plicated malaria in African children and adults using network meta-analysis. Malar. J. 2017; 16: 311.
[Link] PMID: 28774303
47. WHO. Antimalarial drug combination therapy: report of a WHO technical consultation. 2001.
48. Tjitra E, Hasugian AR, Siswantoro H, Prasetyorini B, Ekowatiningsih R, Yusnita EA, et al. Efficacy and
safety of artemisinin-naphthoquine versus dihydroartemisinin-piperaquine in adult patients with uncom-
plicated malaria: a multi-centre study in Indonesia. Malar J 2012; 11(1):153.

PLOS ONE | [Link] December 19, 2019 16 / 16