Artesunate plus pyronaridine for treating uncomplicated Plasmodium falciparum malaria

Hasifa Bukirwa; B Unnikrishnan; Christine V Kramer; David Sinclair; Suma Nair; Prathap Tharyan

doi:10.1002/14651858.CD006404.pub2

. 2014 Mar 4;(3):1–115. doi: 10.1002/14651858.CD006404.pub2

Artesunate plus pyronaridine for treating uncomplicated Plasmodium falciparum malaria

Hasifa Bukirwa ^1,^✉, B Unnikrishnan ², Christine V Kramer ³, David Sinclair ⁴, Suma Nair ⁵, Prathap Tharyan ⁶

PMCID: PMC4448218 PMID: 24596021

Abstract

Background

The World Health Organization (WHO) recommends that people with uncomplicated Plasmodium falciparum malaria are treated using Artemisinin-based Combination Therapy (ACT). ACT combines three-days of a short-acting artemisinin derivative with a longer-acting antimalarial which has a different mode of action. Pyronaridine has been reported as an effective antimalarial over two decades of use in parts of Asia, and is currently being evaluated as a partner drug for artesunate.

Objectives

To evaluate the efficacy and safety of artesunate-pyronaridine compared to alternative ACTs for treating people with uncomplicated P. falciparum malaria.

Search methods

We searched the Cochrane Infectious Diseases Group Specialized Register; Cochrane Central Register of Controlled Trials (CENTRAL), published in The Cochrane Library; MEDLINE; EMBASE; LILACS; ClinicalTrials.gov; the metaRegister of Controlled Trials (mRCT); and the WHO International Clinical Trials Search Portal up to 16 January 2014. We searched reference lists and conference abstracts, and contacted experts for information about ongoing and unpublished trials.

Selection criteria

Randomized controlled trials of artesunate-pyronaridine versus other ACTs in adults and children with uncomplicated P. falciparum malaria.

For the safety analysis, we also included adverse events data from trials comparing any treatment regimen containing pyronaridine with regimens not containing pyronaridine.

Data collection and analysis

Two authors independently assessed trial eligibility and risk of bias, and extracted data. We combined dichotomous data using risk ratios (RR) and continuous data using mean differences (MD), and presented all results with a 95% confidence interval (CI). We used the GRADE approach to assess the quality of evidence.

Main results

We included six randomized controlled trials enrolling 3718 children and adults.

Artesunate-pyronaridine versus artemether-lumefantrine

In two multicentre trials, enrolling mainly older children and adults from west and south-central Africa, both artesunate-pyronaridine and artemether-lumefantrine had fewer than 5% PCR adjusted treatment failures during 42 days of follow-up, with no differences between groups (two trials, 1472 participants, low quality evidence). There were fewer new infections during the first 28 days in those given artesunate-pyronaridine (PCR-unadjusted treatment failure: RR 0.60, 95% CI 0.40 to 0.90, two trials, 1720 participants, moderate quality evidence), but no difference was detected over the whole 42 day follow-up (two trials, 1691 participants, moderate quality evidence).

Artesunate-pyronaridine versus artesunate plus mefloquine

In one multicentre trial, enrolling mainly older children and adults from South East Asia, both artesunate-pyronaridine and artesunate plus mefloquine had fewer than 5% PCR adjusted treatment failures during 28 days follow-up (one trial, 1187 participants, moderate quality evidence). PCR-adjusted treatment failures were 6% by day 42 for these treated with artesunate-pyronaridine, and 4% for those with artesunate-mefloquine (RR 1.64, 95% CI 0.89 to 3.00, one trial, 1116 participants, low quality evidence). Again, there were fewer new infections during the first 28 days in those given artesunate-pyronaridine (PCR-unadjusted treatment failure: RR 0.35, 95% CI 0.17 to 0.73, one trial, 1720 participants, moderate quality evidence), but no differences were detected over the whole 42 days (one trial, 1146 participants, low quality evidence).

Adverse effects

Serious adverse events were uncommon in these trials, with no difference detected between artesunate-pyronaridine and comparator ACTs. The analysis of liver function tests showed biochemical elevation were four times more frequent with artesunate-pyronaridine than with the other antimalarials (RR 4.17, 95% CI 1.38 to 12.62, four trials, 3523 participants, moderate quality evidence).

Authors' conclusions

Artesunate-pyronaridine performed well in these trials compared to artemether-lumefantrine and artesunate plus mefloquine, with PCR-adjusted treatment failure at day 28 below the 5% standard set by the WHO. Further efficacy and safety studies in African and Asian children are required to clarify whether this combination is an option for first-line treatment.

Plain Language summary

Artesunate plus pyronaridine for treating uncomplicated Plasmodium falciparum malaria

What is uncomplicated malaria and how might artesunate-pyronaridine work

Uncomplicated malaria is the milder form of malaria which usually causes fever, with or without headache, tiredness, muscle pains, abdominal pains, nausea, and vomiting. If left untreated, uncomplicated malaria can rapidly develop into severe malaria with kidney failure, fitting, unconsciousness, and eventually death. Plasmodium falciparum is the most common parasite causing malaria in sub-Saharan Africa and causes most of the severe malaria worldwide.

The World Health Organization currently recommends countries use one of five different artemisinin-based combination therapies (ACTs) to treat malaria. These combinations contain an artemisinin component (artemether, dihydroartemisinin, or artesunate), which works quickly to clear the parasite from the person's blood, and a longer-acting drug which clears the remaining parasites from the blood and may prevent new Plasmodium infections for several weeks. Artesunate plus pyronaridine is a new combination and in this review we evaluate its effectiveness and safety compared to the other ACTs.

After examining the research published up to 16 January 2014, we included six randomized controlled trials, enrolling 3718 children and adults.

What the research says

Based on studies of mostly older children and adults living in Africa and Southeast Asia, artesunate-pyronaridine is probably as effective as artemether-lumefantrine at treating uncomplicated malaria and preventing further malaria infections after treatment (moderate quality evidence).

In a study primarily of older children and adults in Asia, artesunate-pyronaridine is probably as effective as artesunate plus mefloquine at treating P. falciparum malaria and preventing recurrent parasitaemias (moderate quality evidence).

Serious adverse events were rare in people treated with either artesunate-pyronaridine or other ACTs. However, short-lasting liver toxicity was more frequent in people treated with artesunate-pyronaridine than with the other antimalarials (moderate quality evidence).

Authors' conclusions

Artesunate-pyronaridine performed well compared to the other two ACT with which it has been compared, but further studies in African and Asian children are required to help clarify whether this combination is an option for first-line treatment.

Background

Description of the condition

Malaria continues to pose a serious global health challenge despite considerable progress over the past decade to control and eliminate malaria in some parts of the world. In 2010, there were an estimated 219 million malaria illness episodes, resulting in around 660,000 deaths (WHO ²⁰¹²).

Five species of Plasmodium parasite cause malaria in humans; Plasmodium falciparum and P. vivax are the most common, and P. falciparum causes most of the severe disease cases (WHO ²⁰¹²). Uncomplicated malaria is the mild form of the disease, typically characterized by fever with or without associated headache, tiredness, muscle pains, abdominal pains, rigors, nausea, and vomiting (WHO ^2010a). If left untreated, uncomplicated malaria can rapidly develop into severe, life-threatening forms of the disease, particularly in people that have not acquired immunity. Effective immunity generally requires repeated infections over five to 10 years, and is reduced during pregnancy. Consequently in highly endemic settings, as seen in many areas of rural sub-Saharan Africa, young children and pregnant women are most at risk, while in settings with low or seasonal transmission, all age groups can be equally at risk (WHO ^2010a).

In many parts of the world, P. falciparum has developed resistance to most antimalarial drugs used as monotherapy (White 2004; WHO ^2010b). Consequently, the World Health Organization (WHO) now recommends that P. falciparum malaria is always treated with a combination of two drugs that act at different biochemical sites within the parasite (WHO ^2010a). If a parasite mutation producing drug resistance arises spontaneously during treatment, the parasite should then be killed by the partner drug, thus reducing or delaying the development of resistance and increasing the useful lifetime of the individual drugs (White 1996; White ¹⁹⁹⁹).

Five artemisinin-based combination therapies (ACTs) are now recommended for the first-line treatment of uncomplicated malaria; artemether-lumefantrine (AL), artesunate plus amodiaquine (AS+AQ), artesunate plus mefloquine (AS+MQ), artesunate plus sulfadoxine-pyrimethamine (AS+SP), and dihydroartemisinin-piperaquine (DHA-P) (WHO ^2010a). The artemisinin components (artemether, artesunate, or dihydroartemisinin) are highly effective schizonticides, and over three days of treatment rapidly eliminate up to 90% of the blood stage asexual forms of P. falciparum. The partner drugs are longer-acting and are used to clear any residual infection (Nosten 2007; Kurtzhals ²⁰⁰⁸; WHO ^2010a). The combinations with very long half-lives (AS+MQ and DHA-P) can provide a period of post-treatment prophylaxis which may last for up to six weeks (Sinclair 2009).

Resistance to the artemisinin-derivatives was first reported among P. falciparum strains in 2008 along the Thai-Cambodian border (Dondorp 2010; Lim ²⁰¹⁰; WHO ^2010b). This has led to global initiatives to contain the spread of artemisinin resistance, which includes the development of new drugs to partner and protect the artemisinin-derivatives in ACT (WHO ²⁰¹¹).

Description of the intervention

Pyronaridine is a benzonaphthyridine derivative first synthesized in China in 1970 (Fu 1991). It was used extensively as a monotherapy to treat P. falciparum and P. vivax infections in the Hunan and Yunan provinces of China for more than 20 years (Chen 1992), and to treat P. falciparum in some parts of Africa during the 1980s. Between 1985 and 1995, some in vitro pyronaridine-resistant strains of P. falciparum emerged along the China-Lao and China-Myanmar border areas (Yang 1997).

Elsewhere, in vitro studies using clinical isolates of P. falciparum from Africa, Cambodia, and Thailand in the 1990s demonstrated high activity of pyronaridine against chloroquine-sensitive and chloroquine-resistant P. falciparum strains (Childs 1988; Basco ¹⁹⁹²; Chen ¹⁹⁹²; Pradines ¹⁹⁹⁸; Ringwald ¹⁹⁹⁹), and more recent in vitro studies have also shown pyronaridine to be effective against multiple-drug resistant P. falciparum schizonts and gametocytes in Thailand and Indonesia (Chavalitshewinkoon-Petmitr 2000; Price ²⁰¹⁰), and against chloroquine-resistant P. falciparum strains in Gabon (Kurth 2009). However, almost half of 28 P. falciparum isolates tested in vitro in Abidjan, Cote d'Ivoire, were resistant to pyronaridine and also showed some evidence of cross resistance to dihydroartemisinin (Brice 2010).

Pyronaridine interferes with the glutathione-dependent detoxification of haem and targeting of β-haematin formation (Auparakittanon 2006). Its activity in multi-drug resistant strains of P. falciparum is believed to be due to its ability to inhibit P-glycoprotein function and reverse multi-drug resistance in cell lines (Qi 2002; Pradines ²⁰¹⁰).

Pyronaridine is structurally related to amodiaquine, leading to some concerns that pyronaridine may have similar toxicity related to the formation of a reactive metabolite (quinoneimine) in the liver and white blood cells. However, some studies suggest that pyronaridine and other bis-Mannich compounds are structurally advantaged and do not form the bioactive quinoneimine metabolite (Naisbitt 1998; Ruscoe ¹⁹⁹⁸).

Assessment of antimalarial drug efficacy

The WHO recommends that new antimalarials should have a treatment failure rate of less than 5%, and that failure rates greater than 10% with existing first-line antimalarials should trigger a change in treatment policy (WHO ^2010a).

Treatment failure can be classified as:

Early treatment failure:

the development of danger signs or severe malaria on days 1, 2, or 3 in the presence of parasitaemia;
parasitaemia on day 2 higher than on day 0;
parasitaemia and axillary temperature = 37.5 °C on day 3;
parasitaemia on day 3 = 20% of count on day 0.

Late treatment failure:

development of danger signs, or severe malaria, after day 3 with parasitaemia;
presence of P. falciparum parasitaemia and axillary temperature = 37.5 °C on or after day 4;
presence of P. falciparum parasitaemia after day 7.

The late reappearance of P. falciparum parasites in the blood of an infected person can be due to failure of the drug to completely clear the original parasite (a recrudescence) or due to a new infection, which is especially common in areas of high transmission. A molecular genotyping technique called polymerase chain reaction (PCR) can be used in clinical trials to distinguish between recrudescence and new infection, giving a clearer picture of the efficacy of the drug and its post-treatment prophylactic effect (White 2002; Cattamanchi ²⁰⁰³; WHO ²⁰⁰⁸).

The WHO recommends a minimum follow-up period of 28 days for antimalarial efficacy trials, but longer periods of follow-up may be required for antimalarials with long elimination half-lives (White 2002; Bloland ²⁰⁰³). Treatment failure due to true recrudescence of malaria parasites may be delayed until the drug concentration falls below the minimum concentration required to inhibit parasite multiplication, which may be beyond 28 days. The WHO recommends 42 days follow-up for trials involving lumefantrine and piperaquine and 63 days follow-up for trials of mefloquine (WHO ^2010a).

Why it is important to do this review

Early studies of pyronaridine monotherapy conducted in Africa showing efficacy against chloroquine-resistant P. falciparum malaria (Ringwald 1999), and promising dose finding studies of the artesunate-pyronaridine combination from the Gabon (Ramharter 2008), have led to the promotion of artesunate-pyronaridine as a possible addition to the current list of recommended ACTs (Vivas 2008; Croft ²⁰¹⁰).

This review aims to systematically evaluate the available trials on the effectiveness and safety of artemisinin plus pyronaridine for consideration by global and national policy makers.

Objectives

To evaluate the efficacy and safety of artesunate-pyronaridine compared to alternative ACTs for treating people with uncomplicated P. falciparum malaria.

Methods

Criteria for considering studies for this review

Types of studies

Randomized controlled trials (RCTs).

Types of participants

Adults and children with uncomplicated P. falciparum malaria, as confirmed by either microscopy or rapid diagnostic tests.

Types of interventions

Intervention

Artesunate plus pyronaridine.

Control

WHO-recommended ACTs for treating malaria.

For an additional safety analysis we extended the inclusion criteria to all RCTs comparing pyronaridine alone or in combination with any other antimalarial.

Types of outcome measures

We used current WHO recommendations to guide the selection of outcomes for this review (Bloland 2003; WHO ²⁰⁰⁸).

Primary outcomes

Total treatment failure at day 28, 42, or 63 (PCR-unadjusted and PCR-adjusted).

Secondary outcomes

Early treatment failure
Parasite clearance
Fever clearance
Gametocyte carriage

Adverse events

Serious adverse events (leading to death, requiring hospitalization or prolongation of existing hospitalization, are life threatening, or result in persistent or significant disability or incapacity)
Adverse events leading to withdrawal from treatment (discontinuation of trial drug or withdrawal from trial)
Patient reported symptoms
Abnormal liver function tests (LFTs)
Abnormal WBC counts
Abnormal electrocardiogram (ECG) findings

Search methods for identification of studies

We attempted to find all relevant trials regardless of language or publication status (published, unpublished, in press, and in progress).

Electronic searches

We updated previous literature searches done in February 2007 and August 2012 of the following databases using the search terms and strategy described in Appendix 1 up to 16 January 2014: Cochrane Infectious Diseases Group Specialized Register; Cochrane Central Register of Controlled Trials (CENTRAL), published in The Cochrane Library; MEDLINE; EMBASE; and LILACS. We also searched ClinicalTrials.gov, the metaRegister of Controlled Trials (mRCT) and the WHO's International Clinical Trials Registry Platform Search Portal for ongoing or recently completed trials using 'pyronaridine' and 'malaria' as search terms.

Searching other resources

Conference proceedings

We searched the following conference proceedings for relevant abstracts: The American Society of Tropical Medicine and Hygiene Annual Meetings (2007, 2008, 2009, and 2010); The Third ASEAN Congress of Tropical Medicine and Parasitology (ACTMP3); the MIM Pan-African Malaria Conference (2005 and 2009); the International Congress on Infectious Diseases (ICID) (2002, 2004, 2008, and 2010); the International Conference on Malaria: 125 years of Malaria Research 2005; the Keystone Symposia Global Health Series: and Malaria (Immunology, pathogenesis and perspectives) 2008.

Reference lists

We checked the reference lists of all trials identified by the above methods.

Contacting organizations and experts

We contacted the Medicines for Malaria Venture and the WHO for information about ongoing and unpublished trials.

Data collection and analysis

Selection of studies

Hasifa Bukirwa (HB) and Prathap Tharyan (PT) independently scanned the results of the search strategy and retrieved the full text articles of all potentially relevant trials, conscious of the possibility of multiple publications of the same trial. HB and PT independently assessed each potentially relevant trial for inclusion in the review using an eligibility form based on the inclusion criteria. There were no disagreements. We excluded studies that did not meet the eligibility criteria and listed the reasons for exclusion in the 'Characteristics of excluded studies' table.

Data extraction and management

HB and PT independently extracted the data from the trials using data extraction forms. We resolved disagreements through discussion. For dichotomous outcome measures, we recorded the number of participants experiencing the event and the number analysed in each group. For continuous outcome measures, we extracted arithmetic means and standard deviations for each group together with the numbers analysed in each group.

Primary outcome

Our primary analysis drew on the WHO's protocol for assessing and monitoring antimalarial drug efficacy (Bloland 2003). This protocol has been used to guide most efficacy trials since its publication in 2003, even though it was designed to assess the level of antimalarial resistance in the trial area rather than for comparative trials. As a consequence, a high number of randomized participants are excluded from the final efficacy outcome as losses to follow-up or voluntary or involuntary withdrawals (see Table 1).

PCR-unadjusted total failure

We calculated PCR-unadjusted total failure (P. falciparum) as the sum of early treatment failures and late treatment failures (without PCR adjustment). The denominator excludes participants for whom an outcome was not available (for example, those who were lost to follow-up, withdrew consent, took other antimalarials, or failed to complete treatment) and those participants who did not to fulfil the inclusion criteria after randomization.

PCR-adjusted total failure

We determined PCR-adjusted total failure (P. falciparum) as the sum of early treatment failures, and late treatment failures due to PCR-confirmed recrudescence. We treated participants with indeterminate PCR results, missing PCR results, or PCR-confirmed new infections as involuntary withdrawals and excluded them from the calculation. The denominator excludes participants for whom an outcome was not available (for example, those who were lost to follow-up, withdrew consent, took other antimalarials, or failed to complete treatment) and participants who did not fulfil the inclusion criteria after randomization.

These primary outcomes relate solely to failure due to P. falciparum. For both PCR-unadjusted and PCR-adjusted total failure, we retained in the calculation participants who developed P. vivax parasitaemia during follow-up if they were treated with chloroquine and continued to be monitored by the trialists. We classified them as treatment successes provided they did not go on to develop P. falciparum parasitaemia. We excluded from the calculation participants who developed P. vivax parasitaemia and were removed from the trial's follow-up at the time of P. vivax parasitaemia.

Assessment of risk of bias in included studies

For efficacy outcomes we assessed the risk of bias for each included trial using the Cochrane tool for assessing the risk of bias (Higgins 2011). For each of six domains; sequence generation; allocation concealment; blinding of participants, trial personnel and outcome assessors; incomplete outcome data; selective reporting; and other sources of bias, we assigned a judgment regarding the risk of bias. We classified these judgments as 'high risk', 'low risk ', or 'unclear risk' of bias. We recorded these assessments in the standard 'risk of bias' tables and summarized the risk of bias for each trial in a summary risk of bias graph.

For patient reported adverse events, we assessed the risk of bias by examining if monitoring was active or passive; whether participants and outcome assessors were blinded; whether the outcome data reporting was complete; whether all participants were included; and whether data analysis was independent of pharmaceutical companies (Table 2).

For laboratory reported adverse events, we assessed the risk of bias by examining which tests were performed, the timing of the tests, the completeness of reporting, and the independence of the data analysis (Table 2).

Measures of treatment effect

We extracted data from each included trial to calculate risk ratios, 95% confidence intervals (CIs) for dichotomous data, and mean differences with 95% CIs for continuous data.

Unit of analysis issues

We did not encounter any unit of analysis issues.

Dealing with missing data

If data from the trial reports were insufficient, unclear, or missing, we attempted to contact the trial authors for additional information. If we considered that the missing data rendered the result uninterpretable, we excluded the data from the meta-analysis and clearly stated the reason for exclusion. We explored the potential effects of missing data through a series of sensitivity analyses (Table 1).

Assessment of heterogeneity

We assessed heterogeneity amongst trials by inspecting the forest plots, applying the Chi² test with a 10% level of statistical significance, and also using the I² statistic with a value of 50% used to denote moderate levels of heterogeneity.

Assessment of reporting biases

There were too few trials to examine funnel plot asymmetry for evidence of small trial effects or publication bias.

Data synthesis

We analysed data using Review Manager 2011.

For the primary analysis we stratified by comparator ACT, and when outcomes were assessed and reported at different time-points, we also stratified the analyses by time point. We performed meta-analysis where appropriate after assessment and investigation of heterogeneity. In the first instance, we used a fixed-effect model and applied a random-effects model when the Chi² test P value was < 0.1 or the I² statistic was = 50%.

Arithmetic means and standard deviations used to summarize continuous data are assumed to be normally distributed; however, sometimes these summary statistics are incorrectly used when the data are not normally distributed. Therefore, when arithmetic means were reported, we checked the normality of the data by calculating the ratio of the mean over the standard deviation. If this ratio (mean/standard) was < 2, then it is likely that the data are skewed as the mean cannot then lie in the centre of a normal distribution. It is possible to combine data with less severe degrees of skew in meta-analyses and when ratio of the mean over the standard deviation was more than one (ratios less than one indicate that data were severely skewed), we combined data from these trials with normally distributed data.

Subgroup analysis and investigation of heterogeneity

There were too few trials to use subgroup analyses to explore the causes of heterogeneity. However, to explore the generalizability of the evidence we subgrouped the available data by age (< 5 years versus ≥ 5 years), country, and geographic region.

Sensitivity analysis

We assessed that all three trials were at low risk of bias so we did not perform a sensitivity analysis exploring effects of risk of bias.

To investigate the robustness of the methodology used in the primary analysis, we conducted a series of sensitivity analyses. The aim of this was to restore the integrity of the randomization process by adding excluded groups back into the analysis in a stepwise fashion (see Table 1 for details).

Quality of evidence

We assessed the quality of evidence across each outcome measure using the GRADE approach. The quality rating across studies has four levels: high, moderate, low, or very low. RCTs are initially categorized as high quality but can be downgraded after assessment of five criteria: risk of bias, consistency, directness, imprecision, and publication bias (Guyatt 2008).

Results

Description of studies

See Characteristics of included studies, and Characteristics of excluded studies sections.

Results of the search

Of the 52 reports we retrieved by the search, we identified 39 potentially relevant reports. Three trials comparing artesunate-pyronaridine with other ACTs met the inclusion criteria for the main review (Tshefu 2010; Kayentao ²⁰¹²; Rueangweerayut ²⁰¹²). We included three additional trials for a further assessment of the effect of pyronaridine on liver function (Ringwald 1996; Ringwald ¹⁹⁹⁸; Poravuth ²⁰¹¹). We described the results of the search in a flow diagram (1)

Included studies

Efficacy trials

The three efficacy trials were all Phase III non-inferiority trials conducted by the public-private partnership of Medicines for Malaria Venture (Switzerland) and Shin Poong Pharmaceuticals (Korea) for registration with the European Medicines Agency (Tshefu 2010; Kayentao ²⁰¹²; Rueangweerayut ²⁰¹²).

Artesunate-pyronaridine versus artemether-lumefantrine

Two multicentre trials that included 1807 participants evaluated this comparison (Tshefu 2010; Kayentao ²⁰¹²).

Most participants (88.3%) were recruited from trial sites in Africa (Burkina Faso, Cote d'Ivoire, Democratic Republic of Congo, Gabon, The Gambia, Ghana, Kenya, Mali, Mozambique, and Senegal), with a small number (11.7%) from Southeast Asia (Indonesia and the Phillipines). All recruiting sites were endemic for P. falciparum malaria and most were reported as highly endemic.

Most participants were older children or adults, and only 232 children aged under five years, and 15 aged under one year were included.

Important exclusion criteria were severe malaria, cerebral malaria, severe anaemia, pregnant and lactating women, and people with hepatic, renal, or other disorders. Tshefu ²⁰¹⁰ also excluded those with severe malnutrition and Kayentao ²⁰¹² excluded children with HIV infection.

In both trials, artesunate-pyronaridine was administered once daily for three days, and artemether-lumefantrine twice daily for three days in the standard dosing (see Table 3).

Artesunate-pyronaridine versus artemether plus mefloquine

A single multicentre trial, enrolling 1271 participants evaluated this comparison (Rueangweerayut 2012).

Most participants (81.3%) were from Southeast Asia (Cambodia, India, Thailand, and Vietnam), with a smaller number (18.7%) from Africa (Burkina Faso, Ivory Coast, and Tanzania). Malaria endemicity was high in most sites.

Although the trial planned to recruit participants aged between 3 to 60 years, the youngest participant was five years old.

Important exclusion criteria were severe malaria, cerebral malaria, severe anaemia, severe malnutrition, pregnant and lactating women, and people with hepatic or renal disorders.

Both artesunate-pyronaridine and artesunate plus mefloquine were administered once daily for three days (see Table 3).

Additional safety trials

The three additional safety trials compared artesunate-pyronaridine versus chloroquine (Poravuth 2011), and pyronaridine alone versus chloroquine (Ringwald 1996; Ringwald ¹⁹⁹⁸).

Poravuth ²⁰¹¹ was conducted in Asia and primarily evaluated the effects of artesunate-pyronaridine on P. vivax malaria (Cambodia, India, Indonesia, and Thailand). Ringwald ¹⁹⁹⁶ and Ringwald ¹⁹⁹⁸ were conducted in Cameroon.

Poravuth ²⁰¹¹ randomized 456 participants aged seven years to 60 years; Ringwald ¹⁹⁹⁶ randomized 96 adults aged 15 to 64 years, and Ringwald ¹⁹⁹⁸ recruited 88 children only, aged five years to 15 years.

For further details of the included trials see the 'Characteristics of included studies' tables.

Excluded studies

We excluded 21 trials (22 records) for the reasons described in the 'Characteristics of excluded studies' table. In brief; 13 were not randomized, four were quasi-randomized (used alternation), and five did not have populations, comparisons, or outcomes of relevance to this review (1).

One trial comparing pyronaridine alone for three days versus dihydroartemisinin alone for seven days versus a combination of pyronaridine and dihydroartemisinin for three days did not meet the inclusion criteria for the primary efficacy analysis due to the lack of an appropriate comparison arm with an ACT, and was not included in the safety analysis as LFTs were not reported (Liu ²⁰⁰²⁾.

Risk of bias in included studies

See 2.

Risk of bias summary table (Methodological quality summary): review authors' judgements about each methodological quality item for each included trial.

Allocation

All trials were at low risk of selection bias.

Blinding

The four non-inferiority trials were at low risk for performance and detection bias as they used double-dummy techniques, or independent outcome assessors and trial personnel who were not aware of allocation (Tshefu 2010; Poravuth ²⁰¹¹; Kayentao ²⁰¹²; Rueangweerayut ²⁰¹²).

The additional two safety trials (Ringwald 1996; Ringwald ¹⁹⁹⁸) were open label or inadequately masked but were at low risk of bias, since blinding would not affect detection of the adverse outcomes sought in this review.

Incomplete outcome data

All of the included trials reported attrition with details of all randomized participants.

Selective reporting

Tshefu ²⁰¹⁰; Poravuth ²⁰¹¹; Kayentao ²⁰¹² and Rueangweerayut ²⁰¹² were prospectively registered and appeared free of selective reporting, as ascertained from the data presented in the reports, the registration documents, and where available, the trial protocols.

Other potential sources of bias

We considered that Ringwald ¹⁹⁹⁶; Ringwald ¹⁹⁹⁸; and Tshefu ²⁰¹⁰ had other potential biases (see Risk of bias tables) but the effects on these on outcomes are uncertain.

For adverse events, we conducted additional assessments of the adequacy of safety monitoring and the completeness of reporting. For patient reported adverse events, the method for monitoring adverse events was unclear in all six trials, the days monitoring occurred was unclear in five trials, and the day of outcome reporting unclear in all six trials (see Table 4). For biochemical adverse events, the frequency of testing was adequate in three trials (Tshefu 2010; Poravuth ²⁰¹¹; Kayentao ²⁰¹²), and reporting was complete in two trials (Tshefu 2010; Poravuth ²⁰¹¹; see Table 5).

Effects of interventions

See: Summary of findings for the main comparison Artesunate-pyronaridine compared to artemether-lumefantrine for uncomplicated falciparum malaria; Summary of findings 2 Artesunate-pyronaridine compared to artesunate plus mefloquine for treating uncomplicated P. falciparum malaria; Summary of findings 3 Liver toxicity of pyronaridine compared to other antimalarials

Comparison 1. Artesunate-pyronaridine versus artemether-lumefantrine

Two trials, including 1595 participants from Africa and 212 from Southeast Asia, compared artesunate-pyronaridine with artemether-lumefantrine (Tshefu 2010; Kayentao ²⁰¹²). Only Kayentao ²⁰¹² included children aged under five years (232 children), of which only 15 were aged under one year. Follow-up was until day 42.

Treatment failure

At day 28, the proportion of participants with recurrent parasitaemia was lower in those treated with artesunate-pyronaridine compared to artemether-lumefantrine (PCR-unadjusted treatment failure; RR 0.60, 95% CI 0.40 to 0.90; two trials, 1720 participants, Analysis 1.1, 3). However, after PCR-adjustment treatment failure, it was below 5% with both ACTs, with no differences between groups (PCR-adjusted treatment failure: two trials, 1650 participants, Analysis 1.1).

Forest plot of comparison: 1 Artesunate-pyronaridine versus artemether-lumefantrine, outcome: 1.1 Total failure (Day 28).

At day 42, there were no significant differences between artesunate-pyronaridine and artemether-lumefantrine for PCR-unadjusted (two trials, 1691 participants, Analysis 1.2) or PCR-adjusted treatment failure (two trials, 1472 participants, Analysis 1.2). PCR-adjusted treatment failure with artesunate-pyronaridine was marginally above 5% in one trial at this time-point (6.8%).

Only two people on artesunate-pyronaridine and one on artemether-lumefantrine experienced early treatment failure (two trials, 1676 participants, Analysis 1.3).

Parasite clearance

Both trials reported that artesunate-pyronaridine cleared parasites from the peripheral blood quicker than artemether-lumefantrine. Tshefu ²⁰¹⁰ reported a slightly lower mean clearance time (MD 3.2 hours, 95% CI 4.38 to 2.02; one trial, 1170 participants; Analysis 1.4), and Kayentao ²⁰¹² reported a slightly lower median clearance time (24.1 hours, 95% CI 24.0 to 24.1 with artesunate-pyronaridine versus 24.2 hours, 95% CI 24.1 to 32.0 with artemether-lumefantrine; P = 0.02, authors' own figures, one trial, 535 participants, Table 6). These differences are probably not clinically important.

Fever clearance

Fever clearance times were similar between groups in both trials. Tshefu ²⁰¹⁰ reported mean fever clearance time as marginally shorter following treatment with artesunate-pyronaridine than artemether-lumefantrine (MD 1.2 hours, 95% CI 2.38 to 0.02 hours, one trial, 1170 participants, Analysis 1.5), while Kayentao ²⁰¹² reported equal median clearance times (8.1 hours with artesunate-pyronaridine versus 8.1 hours with artemether-lumefantrine, P = 0.049, authors' own figures, one trial, 535 participants, Table 6).

Gametocyte clearance and carriage

In Tshefu ²⁰¹⁰, 8% of participants given artesunate-pyronaridine and 5% of those given artemether-lumefantrine had peripheral gametocytaemia at baseline. The mean time to gametocyte clearance was 10.5 hours shorter with artesunate-pyronaridine (MD 10.5 hours, 95% CI 12.4 to 8.60; one trial, 1170 participants, Analysis 1.6).

In Kayentao ²⁰¹², 13% of participants had gametocytes at baseline. No subsequent statistically significant differences in gametocyte carriage, or gametocyte development were reported (one trial, 532 participants, Table 6).

Serious adverse events

Neither trial reported any deaths. There were six serious adverse events in total with no significant difference between groups (0.3% with artesunate-pyronaridine versus 0.3% with artemether-lumefantrine; two trials, 1787 participants, Analysis 1.7).

Adverse events leading to withdrawal from treatment

There was no significant difference between groups in the proportion of participants withdrawn from the trial due to adverse events (2.3% with artesunate-pyronaridine versus 1.7% with artemether-lumefantrine; two trials, 1787 participants, Analysis 1.8).

Patient-reported symptoms

There were no significant differences in patient-reported symptoms between the two ACTs (two trials, 1807 participants, Analysis 1.9, Analysis 1.10). The trial authors reported symptoms of vomiting, headache, abdominal pain, vertigo, haematuria, upper abdominal pain, and anorexia.

Biochemical monitoring and adverse events

Both trials measured biochemical LFTs in all participants at baseline and on days three and seven (Kayentao 2012 also measured LFTs on day 28), Although the two trials used slightly different grading scales, there were no significant differences between groups in grade 3 or 4 liver toxicity by any of the measures used (two trials, 1807 participants, Analysis 1.11, Analysis 1.12).

Haematological monitoring and adverse events

In both trials the mean haemoglobin fell compared to baseline during the first seven days after starting treatment, before recovering by day 28 (two trials, 1807 participants, Analysis 1.12). At day seven the reduction in haemoglobin was greater with artesunate-pyronaridine but this is unlikely to be of clinical significance (MD -0.16, 95% CI -0.28 to -0.05; two trials, 1741 participants, Analysis 1.12).

Kayentao ²⁰¹² also reported the occurrence of anaemia as an adverse event with no differences between groups (one trial, 535 participants, Analysis 1.13).

ECG monitoring and adverse events

Both trials conducted ECG monitoring at baseline, days 2, 7, 14 and 28. Tshefu ²⁰¹⁰ reported two participants in each group having abnormal ECG readings and reported these as "mild". Kayentao ²⁰¹² reported that there were "no post-baseline clinically important abnormal ECG results" (see Table 7).

Subgroup analysis

We have presented a subgroup analysis of PCR-adjusted treatment failure at day 28 by age of participants in Analysis 2.1. This demonstrates the paucity of data for the under-five age group.

Further subgroup analyses by geographical region and country are in Analysis 2.2 and Analysis 2.3. Again, these demonstrate that the data remain severely underpowered to inform national decision-making. Primary outcome data was available for only 194 participants from East Africa, compared to 816 from West Africa, 490 from South-central Africa, and 175 from Asia.

Sensitivity analysis

We conducted a sensitivity analysis to explore the influence of different methods for analysing the primary outcome data. For PCR-unadjusted treatment failure, our primary analysis following the WHO guidelines for analysing trials of antimalarials was the least conservative (Analysis 3.1). The per-protocol and intention-to-treat analyses as presented by the trial authors, where missing data were considered treatment failure, were more conservative and the result did not reach statistical significance. For PCR-adjusted treatment failure, there were no substantial differences (Analysis 3.2).

We did not undertake a sensitivity analysis by risk of bias criteria as both of the included trials were at low risk of bias.

Comparison 2. Artesunate-pyronaridine versus artesunate plus mefloquine

Only one trial, enrolling 1033 participants from Asia and 238 from Africa, compared artesunate-pyronaridine versus artemether-lumefantrine (Rueangweerayut 2012). This trial excluded children under five years of age and follow-up was until day 42.

Treatment failure

At day 28, the proportion of participants with recurrent parasitaemia was lower in those treated with artesunate-pyronaridine compared to artesunate plus mefloquine (PCR-unadjusted treatment failure: RR 0.35, 95% CI 0.17 to 0.73; one trial, 1200 participants, Analysis 4.1). However, after PCR-adjustment treatment, failure was below 5% with both ACTs with no differences between groups (one trial, 1187 participants, Analysis 4.1).

At day 42, there were no statistically significant differences between artesunate-pyronaridine and artesunate plus mefloquine for PCR-unadjusted or PCR-adjusted treatment failure (one trial, 1146 participants, Analysis 4.2). At this time point, PCR-adjusted treatment failure was 5.8% with artesunate-pyronaridine versus 3.6% with artesunate plus mefloquine.

One person treated with artesunate plus mefloquine experienced early treatment failure and developed cerebral malaria (one trial, 1103 participants, Analysis 4.3).

Parasite clearance

The mean parasite clearance time was slightly lower with artesunate-pyronaridine compared to artesunate plus mefloquine (MD 2.60 hours, 95% CI 4.94 to 0.26, one trial, 1259 participants, Analysis 4.4).

Fever clearance

Fever clearance time was similar between treatment arms (one trial, 1051 participants, Analysis 4.5).

Gametocyte clearance and carriage

Rueangweerayut ²⁰¹² only reported the mean time to gametocyte clearance for the 27 participants (13 on artesunate-pyronaridine versus 14 on artesunate plus mefloquine) who cleared their gametocytes within the first 72 hours. There was no difference between groups (one trial, 27 participants, Analysis 4.6).

Serious adverse events

Rueangweerayut ²⁰¹² did not report any deaths. There were nine serious adverse events in total with no significant difference between groups (0.7% with artesunate-pyronaridine versus 0.7% with artesunate plus mefloquine; one trial, 1271 participants, Analysis 4.7).

Adverse events leading to withdrawal from treatment

There was no significant difference between groups in the proportion of participants withdrawn from the trial due to adverse events (0.6% with artesunate-pyronaridine versus 0.9% with artesunate plus mefloquine; one trial, 1271 participants, Analysis 4.8).

Patient-reported symptoms

Rueangweerayut ²⁰¹² only reported symptoms if they occurred in at least 2% of patients. Dizziness was twice as common in those treated with artesunate plus mefloquine than with artesunate-pyronaridine (RR 0.46, 95% CI 0.28 to 0.78; one trial, 1271 participants, Analysis 4.9). The other reported symptoms were headache, cough, diarrhoea, vomiting, and myalgia.

Biochemical monitoring and adverse events

Biochemical tests for liver function monitoring were performed on all participants on days 0, 3, 7, 28, and 42.

Artesunate-pyronaridine was associated with more participants recording elevated ALT and AST levels following treatment. For ALT, grade 2 toxicity (up to five times the upper limit of normal) was significantly higher with artesunate-pyronaridine (21/843 versus 0/417; RR 21.30, 95% CI 1.29 to 350.7; one trial, 1260 participants, Analysis 4.10), and grade 3 or 4 toxicity (= five times the upper limit of normal) approached statistical significance (15/843 versus 0/417; RR 7.41, 95% CI 0.98 to 55.98; one trial, 1260 participants, Analysis 4.11). There were no significant differences for other liver enzymes or bilirubin. No patients developed signs or symptoms of liver disease.

Haematological monitoring and adverse events

The mean haemoglobin level fell in both groups during the first seven days after starting treatment (Analysis 4.12) This drop was slightly larger with artesunate-pyronaridine compared to artesunate plus mefloquine (Day 3: MD -0.22 g/dL, 95% CI -0.36 to -0.08; one trial, participants, Analysis 4.12), but by day 28 mean haemoglobin levels were better than baseline in both groups. A similar pattern was observed with platelet counts (Analysis 4.13), and white cell counts (Analysis 4.14). However the differences were small and unlikely to be of clinical significance.

ECG monitoring and adverse events

Rueangweerayut ²⁰¹² conducted ECG monitoring on all participants in this trial but the timing and frequency of ECGs was unclear. The trial authors reported abnormal ECGs in under 1% of participants in both groups, and described all abnormalities as mild and transient (one trial, 1271 participants, Analysis 4.15).

Subgroup analysis

We did not conduct a subgroup analysis by age of participants as this trial did not include children aged under five years.

We have presented subgroup analyses by geographical region and country in Analysis 5.1 and Analysis 5.2. The majority of PCR-adjusted treatment failures occurred in Thailand and Cambodia, with almost none elsewhere. They also demonstrate the paucity of data from Africa.

Trial authors noted that participants enrolled in Pailin, Cambodia (an area of low-transmission forP. falciparum) had significantly longer parasite clearance times than people in the other trial sites; only 63% cleared parasites within 72 hours compared to 98% of participants in the other sites. Recrudescence at this site was reportedly higher with artesunate-pyronaridine than with artesunate plus mefloquine (10.2% versus 0%, P = 0.04; authors' own figures).

Sensitivity analysis

We conducted a sensitivity analysis to explore the influence of different methods for analysing the primary outcome data. For PCR-unadjusted treatment failure, our primary analysis following the WHO guidelines for analysing trials of antimalarials was similar to the per-protocol analysis of the trial authors (Analysis 6.1). In the most conservative estimates the effect size was dramatically reduced and the estimate was no longer statistically significant (Analysis 6.1). For PCR-adjusted treatment failure we did not observe any substantial differences (Analysis 6.2).

We did not perform any further sensitivity analyses as there was only one trial.

Part 3. Biochemical, haematological and ECG adverse events

In light of concerns about liver toxicity with pyronaridine, we included three additional RCTs of pyronaridine. Two trials compared pyronaridine alone to chloroquine (Ringwald 1996; Ringwald ¹⁹⁹⁸) and one trial compared artesunate-pyronaridine to chloroquine (Poravuth 2011).

Biochemical monitoring and adverse events

The six trials reported abnormalities in liver functions in different ways. We assessed the adequacy of monitoring and completeness of results reporting in Table 5.

Artesunate-pyronaridine was associated with a four-fold increase in the incidence of ALT and AST grade 3 or 4 toxicity (elevations = five times the upper limit of normal) (ALT: RR 4.17, 95% CI 1.38 to 12.62, AST: RR 4.08, 95% CI 1.17 to 14.26; four trials, 3528 participants, Analysis 7.1). Grade 3 or 4 toxicity measured with ALP and bilirubin were not substantially different.

The three main efficacy trials also reported cases with both raised ALT (3 x ULN) and raised bilirubin (2 x ULN) as an indicator for drug induced liver injury (Tshefu 2010; Kayentao ²⁰¹²; Rueangweerayut ²⁰¹²). Only five of the 2052 participants in the artesunate-pyronaridine group and one of 1020 participants in the comparator groups had raised ALT and bilirubin. This difference was not statistically significant (three trials, 3072 participants, Analysis 7.2).

Ringwald ¹⁹⁹⁶ reported that 5/40 participants given pyronaridine had elevated bilirubin levels compared to 0/41 with chloroquine but did not give any further details.

Renal function tests

Three trials reported serum creatinine levels as a measure of renal function. At day 7, creatinine values were marginally lower in the pyronaridine-treated group than in those treated with comparator regimens (artemether-lumefantrine, artesunate+mefloquine, chloroquine) (MD -2.76, 95% CI -4.58 to -0.94; three trials, 1808 participants, Analysis 7.3).

Haematological monitoring and adverse events

Four trials reported mean haemoglobin on days 0, 3, 7, and 28, and in all four trials the mean haemoglobin fell in both groups between day 0 and day 7 before recovering by day 28 (four trials, 3534 participants, Analysis 7.4). At day 7 the mean haemoglobin was ¼ gram lower in those treated with artesunate-pyronaridine (MD -0.24 g/dL, 95% CI -0.32 to -0.16; four trials, 3394 participants, Analysis 7.4).

ECG monitoring and adverse events

Four trials conducted ECG monitoring and ECG adverse effects were rare in all four trials (see Table 7). Prolonged QT interval was less common with artesunate-pyronaridine than comparators (RR 0.25, 95% CI 0.07 to 0.90; three trials, 2991 participants, Analysis 7.5).

Discussion

Summary of main results

Artesunate-pyronaridine versus artemether-lumefantrine

In two multicentre trials, enrolling mainly older children and adults from west and south-central Africa, both artesunate-pyronaridine and artemether-lumefantrine had fewer than 5% PCR adjusted treatment failures during 42 days of follow-up, with no differences between groups (low quality evidence). There were fewer new infections during the first 28 days in those given artesunate-pyronaridine (moderate quality evidence), but no difference was detected over the whole 42 day follow-up (moderate quality evidence).

Artesunate-pyronaridine versus artesunate plus mefloquine

In one multicentre trial, enrolling mainly older children and adults from South East Asia, both artesunate-pyronaridine and artesunate plus mefloquine had fewer than 5% PCR adjusted treatment failures during 28 days follow-up (moderate quality evidence). PCR-adjusted treatment failures had risen to 6% by day 42 in those treated with artesunate-pyronaridine, but this was not substantially different to artesunate plus mefloquine (low quality evidence). Again, there were fewer new infections during the first 28 days in those given artesunate-pyronaridine (moderate quality evidence), but no differences were detected over the whole 42 days (low quality evidence).

Adverse effects

Serious adverse events were rare in these trials with no statistically significant differences between artesunate-pyronaridine and the comparator ACTs. However, biochemical elevation of LFTs occurred four times more frequently with artesunate-pyronaridine than with the other antimalarials (moderate quality evidence).

Overall completeness and applicability of evidence

Artesunate-pyronaridine performed well in all three efficacy trials included in this review, with low levels of PCR-adjusted treatment failure at day 28 in all settings. All three trials were multicentre trials, with trial sites in 11 African countries and six countries in Asia, which broadens the applicability of the findings. However, the actual number of participants recruited from many trial sites was small and the trials were underpowered to evaluate either superiority or equivalence at country level. East Africa is particularly under represented, with only 232 participants from Kenya and Tanzania, and several of the West African countries recruited fewer than 100 participants.

The other major limitation on the applicability of these trials is the age of the participants. The trials predominantly recruited older children and adults. The combination appeared to be effective in these groups but little is known about the main target group; children aged under five years. These trials included only 232 children aged below five years compared to over 7000 in trials of dihydroartemisinin-piperaquine.

Notably, all three efficacy trials excluded people with known pre-existing liver disease, and one trial explicitly excluded those with raised LFTs at baseline. Screening of this kind may not be feasible in many malaria-endemic settings.

Quality of the evidence

We assessed the quality of the evidence in this review using the GRADE approach and presented it in two summary of findings tables for efficacy (Summary of findings for the main comparison; Summary of findings 2).

The evidence that artesunate-pyronaridine is equivalent to established ACTs at preventing PCR-adjusted treatment failures was of moderate quality due to two main concerns:

Indirectness: The trials to date have largely been conducted in older children and adults, with exclusion of young children who bear the greatest burden and risks of malaria infection and illness.
Imprecision: The trials were not powered to examine the efficacy of artesunate-pyronaridine in individual regions or countries. This is problematic for national decision-making, and limits the wider generalizability of these results. Larger trials would be required to have full confidence in these results.

We also assessed the quality of evidence on comparative adverse effects and presented these in Appendix 2 and Appendix 3. In general the evidence was of moderate to low quality, and downgraded for similar reasons.

Potential biases in the review process

The objectives of the review changed significantly between the published protocol and final review. The basis for the change was to focus on only interventions of relevance to current malaria treatment policies (see Differences between protocol and review). We used standard methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) and complied with the Cochrane Collaboration's methodological standards for the conduct of new reviews of interventions (MECIR ²⁰¹¹).

We believe that we have identified all pyronaridine trials relevant to inform clinical decisions and policy regarding the use of pyronaridine combinations for the treatment of uncomplicated P. falciparum malaria. The three trials were all conducted under the auspices of the public-private partnership, Medicines for Malaria Venture, and Poong Pharmaceutical Company Ltd, Seoul, Republic of Korea.

Agreements and disagreements with other studies or reviews

We found one further systematic review of artesunate-pyronaridine published by authors from the Medicines for Malaria Venture (MMV), the co-developers of the artesunate-pyronaridine combination (Duparc 2013). The authors include four of the studies included here, plus one study we excluded as it was not randomized (Ramharter 2008), and one unpublished study. The authors conclude that 'Pyronaridine-artesunate was well tolerated with no safety concerns with the exception of mostly mild transient rises in transaminases. Efficacy was high and met the requirements for use as first-line therapy'. While we agree that artesunate-pyronaridine shows promise as a further addition to the ACT combinations, we think it requires further studies in the main target group, children aged less than five years, before countries consider this as a first-line treatment.

Authors' conclusions

Implications for practice

Artesunate-pyronaridine performed well in these trials compared to artemether-lumefantrine and artesunate-mefloquine, with PCR-adjusted treatment failure at day 28 below the 5% standard set by the WHO.

Artesunate-pyronaridine is well-tolerated, apart from transient gastrointestinal adverse effects, similar to other antimalarials. However, the potential for liver toxicity in people treated with artesunate-pyronaridine needs further investigation and will necessitate caution in using this treatment combination, particularly in people with pre-existing liver disorders.

Implications for research

Further efficacy and safety studies in African and Asian children are required before this combination could be established as a first or second-line treatment option.

Appendices

Appendix 1. Search methods: search strategies

Search set	CIDG SR^a	CENTRAL	MEDLINE^b	EMBASE^b	LILACS^b
1	malaria	malaria	malaria	malaria	malaria
2	pyronaridine	pyronaridine	pyronaridine	pyronaridine	pyronaridine
3	1 and 2	1 and 2	NAPHTYRIDINES	PYRONARIDINE	1 and 2
4	—	—	2 or 3	2 or 3	—
5	—	—	1 and 4	1 and 4	—
6	—	—	Limit 5 to human	Limit 5 to human	—

Artesunate-pyronaridine compared to artemether-lumefantrine for treating uncomplicated P. falciparum malaria
Patient or population: Patients with uncomplicated P. falciparum malaria Settings: Malaria endemic areas Intervention: Artesunate-pyronaridine (AS-Pyr) Comparison: Artemether-lumefantrine (AL6)
Outcomes		Number of participants having adverse events (95% CI)		Number of participants (trials)	Quality of the evidence (GRADE)
Outcomes		AL6	AS-Pyr	Number of participants (trials)	Quality of the evidence (GRADE)
Serious adverse events (including deaths)		3 per 1000	0 more per 1000 (From 2 fewer to 10 more)	1787(2 trials)	low ^1,2,3,4
Adverse events leading to withdrawal		17 per 1000	6 more per 1000 (From 6 fewer to 31 more)	1787(2 trials)	low ^1,2,3,4
Gastroenterological	Vomiting	4 per 100	2 more per 100 (From 1 fewer to 10 more)	535(1 trial)	low ^5,6,7
	Diarrhoea	—	—	—	—⁸
	Abdominal pain	5 per 100	0 more per 100 (From 2 fewer to 4 more)	1272(1 trial)	low ^5,9,10
Neuro-psychiatric	Headache	3 per 100	0 more per 100 (From 1 fewer to 1 more)	1272(1 trial)	low ^5,9,10
Neuro-psychiatric	Dizziness	—	—	—	—⁸
Cardio-respiratory	Cough	9 per 100	1 fewer per 100 (From 3 fewer to 2 more)	1807(2 trials)	moderate ^1,2,3,10
	ECG abnormality	4 per 1000	2 fewer per 1000 (From 4 fewer to 10 more)	1272(1 trial)	moderate ^5,9,10,11
	Prolonged QT interval	0 per 1000	1 more per 1000 (From 0 fewer to 36 more)	1272(1 trial)	moderate ^5,9,10,11
Musculoskeletal/ dermatological	Myalgia	—	—	—	—⁸
Biochemical	Alanine aminotransferase Grade 3 or 4 toxicity	3 per 1000	3 more per 1000 (From 2 fewer to 25 more)	1807(2 trials)	low ^1,2,3,4
Biochemical	Aspartate aminotransferase Grade 3 or 4 toxicity	1 per 1000	3 more per 1000 (From 0 fewer to 20 more)	1807(2 trials)	low ^1,2,3,4
The assumed risk of adverse events in the artemether-lumefantrine group is the average risk across trials. The corresponding risk with artesunate-pyronaridine (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval.
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.
¹ No serious risk of bias: Both trials were at low risk of bias. ² No serious inconsistency: Statistical heterogeneity was low. ³ Downgraded by one for serious indirectness: These two trials included only 232 children aged below five years. ⁴ Downgraded by one for imprecision: These trials do not exclude the possibility of rare but clinically important adverse effects. ⁵ No serious risk of bias: This single trial was at low risk of bias. ⁶ Downgraded by one for serious indirectness: This trial included only 232 children aged less than five years and only 15 less than one year. ⁷ Downgraded by one for serious imprecision: The 95% CI is wide and includes both no difference and clinically important differences. ⁸ This outcome was not reported. ⁹ Downgraded by two for very serious indirectness: This trial excluded children aged less than five years. ¹⁰ No serious imprecision: The finding is of no difference between treatments and the sample size is adequately powered to detect differences if they existed. ¹¹ The second trial only reports that there were "no clinically important post baseline ECG results".

Trial ID	Number of participants	Comparator	All serious adverse events		Serious adverse events judged to be related to the medication
Trial ID	Number of participants	Comparator	Artesunate-pyronaridine	Comparator	Artesunate-pyronaridine	Comparator
Kayentao ²⁰¹²	535	Artemether-lumefantrine	Complicated malaria (1)	None	None	None
Tshefu ²⁰¹⁰	1272	Artemether-lumefantrine	Parotitis (1) Typhoid fever (1) Urinary tract infection (1)	Cerebral malaria (1)Immunosuppresion (1)	None	None
Rueangweerayut ²⁰¹²	1271	Artesunate-mefloquine	Autimmune haemolytic anaemia (1) Cholera (1) Pneumonia (1) Acute pyelonephritis (1) Wound infection (1) Abortion (1) Depression (1)	Cerebral malaria (1) Seizure (1) Grand-mal seizure (1)	None	Seizure (1) Grand-mal seizure (1)

Methods	Trial design: Randomized, multicentre, open-label, active-controlled, parallel group, non-inferiority trial Period of trial: November 2007 to November 2008
Participants	Number randomized: 535 Age: Three months to 12 years Gender: Both Inclusion criteria: Male or female patients ≤ 12 years of age. Body weight ≥ 5 kg and < 25 kg with no clinical evidence of severe malnutrition (defined as a child whose weight-for-height is below -3 standard deviations or less than 70% of the median of the NCHS/WHO normalized reference values). Presence of acute uncomplicated P. falciparum mono-infection confirmed by: fever, as defined by axillary temperature ≥ 37.5°C or oral, tympanic, or rectal temperature ≥ 38°C, or documented history of fever in the previous 24 hours and positive microscopy of P. falciparum with parasite density between 1,000 and 200,000 asexual parasite count/µL of blood. Written informed consent, in accordance with local practice, provided by parent or guardian. If the parent or guardian is unable to write, witnessed consent is permitted according to local ethical considerations. Where possible, patient assent will be sought. Ability to swallow whole volume of liquid in which medication is suspended. Female patients of child-bearing potential must be neither pregnant (as demonstrated by a negative pregnancy test) nor lactating, and must be willing to take measures to not become pregnant during the trial period. Ability and willingness to participate based on information given to parent or guardian and access to health facility. The patient is to comply with all scheduled follow-up visits until Day 42. Exclusion criteria: Patients with signs and symptoms of severe/complicated malaria requiring parenteral treatment according to the WHO Criteria. Mixed Plasmodium infection. Severe vomiting, defined as = three times in the 24 hours prior to inclusion in the trial or inability to tolerate oral treatment, or severe diarrhoea defined as ≥ three watery stools per day. Known history or evidence of clinically significant disorders such as cardiovascular (including arrhythmia, QTc interval greater or equal to 450 milliseconds), respiratory (including active tuberculosis), history of jaundice, hepatic, renal, gastrointestinal, immunological (including active HIV-AIDS), neurological (including auditory), endocrine, infectious, malignancy, psychiatric, history of convulsions, or other abnormality (including recent head trauma). Presence of significant anaemia, defined as Hb < 8 g/dL. Presence of febrile conditions caused by diseases other than malaria. Known history of hypersensitivity, allergic or adverse reactions to pyronaridine, lumefantrine or artesunate or other artemisinins. Patients with known disturbances of electrolytes balance, for example, hypokalaemia or hypomagnesaemia. Use of any other antimalarial agent within two weeks prior to start of the trial as evidenced by reported patient history. Pregnant or breast feeding. Patients taking any drug which is metabolized by the cytochrome enzyme CYP2D6 (flecainide, metoprolol, imipramine, amitriptyline, clomipramine). Received an investigational drug within the past four weeks. Known active Hepatitis A IgM (HAV-IgM), Hepatitis B surface antigen (HBsAg) or Hepatitis C antibody (HCV Ab). Known positive for HIV antibody. LFTs [ASAT/ALAT levels] = 2.5 times upper limit of normal range. Known significant renal impairment as indicated by serum creatinine of = 1.4 mg/dL. Previous participation in any clinical trial with pyronaridine artesunate.
Interventions	Randomized 2:1 to Intervention: Fixed-dose oral artesunate-pyronaridine granule formulation (60:20 mg)* once daily for three days by direct observation (N = 355) Control: Artemether-lumefantrine crushed tablets (20/120 mg)** twice daily for three days by direct observation (N = 180) Artesunate-pyronaridine was given once daily: 5 kg to < 9 kg, one sachet; 9 kg to <17 kg, two sachets; 17 kg to < 25 kg, three sachets (dose range 6.7/2.2 to 13.3/4.4 mg/kg/dose mixed in water, milk, or soup) *Artemether-lumefantrine was given twice daily crushed and shaken to a suspension in 50 mL water: 5 kg to < 15 kg, one tablet; 15 kg to < 25 kg, two tablets (dose range 1.3/8.0 to 4.0/24.0 mg/kg/dose); the second day 0 dose was 8 hrs after the first dose, the first Day 1 dose was 24 hrs after the first Day 0 dose, with all subsequent doses 12 hrs apart.
Outcomes	Primary outcomes: Efficacy PCR-corrected adequate clinical and parasitological response rate (ACPR) on Day 28 Safety Adverse events (categorized using MedDRA Version 10.1) Laboratory abnormalities (graded using the Division of Microbiology and Infectious Diseases Toxicity Scale (February 2003)) Secondary outcomes: Day 28 crude (non-PCR corrected) ACPR Day 42 PCR-corrected and crude ACPR Parasite clearance time (time from first dose until aparasitaemia (two consecutive negative readings taken between 7 and 25 hours apart)) Fever clearance time (time from first dose to apyrexia (two consecutive normal readings taken between 7 and 25 hours apart)) Exploratory efficacy outcomes: Proportion of patients with gametocytes Outcomes reported but not used in quantitative synthesis in this review ECG abnormalities Proportion of patients with parasite clearance or fever clearance on days 1, 2, and 3 Gametocyte density Gametocyte clearance time (defined as for parasite clearance time)
Notes	Countries of recruitment: Six countries in Africa (96.3%; Burkina Faso, Democratic Republic of Congo, Gabon, Côte d’Ivoire, Kenya, and Mali) and one in Asia (3.7%; The Philippines). Setting: Local hospitals and clinics at seven centres in six countries in Africa and one in the Philippines Funding: Medicines for Malaria Venture, Poong Pharmaceutical Company Ltd, Seoul, Republic of Korea Endemicity: High Duration of follow-up: 42 days Comment: Age range of participants: 45% in Py-AS aged ≤ 5 years, 3.4% < 1 year; 40% in AL6 aged ≤ 5 years, 1.7% < 3 years Sample size estimation: For the primary efficacy outcome the sample size was estimated to have 91% power to reject the null hypothesis (day-28 cure rate ≤90%) using a 1-sided exact binomial test with a nominal significance level of 2.5%; for the main secondary outcome comparing efficacy to artemether-lumefantrine, the sample provided = 99% power to demonstrate non-inferiority of artesunate-pyronaridine versus artemether-lumefantrine with a non-inferiority limit of 10%. Recrudescence was defined with PCR as at least one matching allelic band in the three P. falciparum genesmsp1, msp2, andglurp between baseline and post-day 7 samples Treatment failures were classified as early treatment failure, late clinical failure, and late parasitological failure according to WHO criteria (Bloland 2003). Defition of Grade 3 and 4 toxicity: ALT and AST grade 3 toxicity was 10 to 15 times the upper limit of normal, and grade 4 toxicity was = 15 times the upper limit of normal. Total bilirubin grade 3 toxicity was 3.0 to 7.5 times the upper limit of normal. Trials registration: ClinicalTrials.gov: identifier NCT00541385
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote from report: "The sponsor provided a computer-generated randomisation schedule. Patients were randomised 2:1 to artesunate-pyronaridine or artemether-lumefantrine".
Allocation concealment (selection bias)	Low risk	Quotes from report: " Individually numbered treatment packs of similar appearance were masked on allocation." Quote from report: "The study sponsor remained blinded to treatment allocation".
Blinding (performance bias and detection bias) Objective outcomes: parasitological and biochemical	Low risk	Quote from report: "Drugs were given open-label". Quote from report: ""Clinical assessments and drug administration were performed by different clinical personnel." Comment: Unlikely to have introduced detection bias for objective outcomes.
Blinding (performance bias and detection bias) Subjective outcomes: adverse events	Low risk	Comment: Outcome assessors were blinded to allocation, and were not involved in drug administration.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comment: The participants randomized were accounted for in the trial report and missing data and participants were not differentially distributed in treatment arms, or substantial.
Selective reporting (reporting bias)	Low risk	Comment: This trial was prospectively registered and though some changes in the timing of assessments were noticed between the protocol and the trial report, these are not of much importance; all other pre-stated outcomes were adequately reported.
Other bias	Low risk	Quote: "The sponsors and study site principal investigators developed the protocol, interpreted the data and developed the report. The study sponsors were responsible for data collection and statistical analysis. All authors had access to the primary data, take responsibility for data reporting accuracy and completeness". Comment: Three of the authors are employed by the study sponsors. However, the report states the study sponsors were blind to treatment allocation, and the final report appears to have been approved by all authors.

Methods	Trial design: Randomized, parallel group, active controlled trial Duration of trial: Recruitment: April 1994 to May 1995
Participants	Number randomized: 96 Age: 15 to 64 years Gender: 42 males; 54 females Inclusion criteria: Acute falciparum malaria with fever within the past 24 hrs or a temperature above 37.5 °C at the time of consultation Over 5000 asexual parasites/µL No signs and symptoms of severe and complicated malaria No recent self-medication Exclusion criteria: Pregnant women Mixed infections
Interventions	Intervention 1. Pyronaridine: 32 mg/kg in divided doses over 3 days* (N = 47) Control 2. Chloroquine: 25 mg/kg in divided doses over 3 days (N = 49) * Pyronaridine dose: 16 mg/kg on day 1 and 8 mg/kg on days 2 and 3 * Chloroquine dose: 10 mg/kg on days 1 and 2 and 5 mg/kg on day 3
Outcomes	Outcome used in this review: Numbers with elevated transaminase enzyme levels at day 7 in those with normal baseline values (extent of elevation not reported) Outcomes reported but used in this review: Fever clearance Parasite clearance Early treatment failure Parasitaemia on day 14 Gametocyte carriage at day 14 Adverse events. Haematological (haemoglobin, counts); biochemical mean (liver function values, creatinine, urea). In vitro drug sensitivity
Notes	County of recruitment: Cameroon Setting: Nlongkak Catholic missionary dispensary in Yaounde; outpatients; all doses of drugs supervised Source of funding: French Ministere de la Cooperation (Grant 93A43); pyronaridine was supplied by the Institute of Parasitic Diseases, Chinese Academy of Preventive Medicine, Shangai, China Endemicity: High; 50% to 60% chloroquine resistant Duration of follow-up: Until day 14 (for all, and in four participants until day 238) Comment: Proportions with normal transaminase enzyme levels at baseline that were elevated in each arm at day 7 provided in the text of results were used for analysis; the extent of elevation was not reported. The values were normal on day 14 on whom the levels were repeated, but the numbers in whom these were repeated are not reported Trials registration: Nil
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote from report: "Patients were randomly assigned in blocks". Comment: Unpublished information provided through correspondence with authors reveal that randomisation was done "in blocks of 10".
Allocation concealment (selection bias)	Low risk	Comment: Not mentioned in report. Quote from correspondence with senior author: "central randomisation was used".
Blinding (performance bias and detection bias) Objective outcomes: parasitological and biochemical	Low risk	Comment: Not mentioned in report. Quote from correspondence: "It was blinded but the tablets were different and many patient treated with CQ suffered of pruritus". Comment: Blinding was probably compromised but risk of bias due to this is unlikely to have affected the biochemical outcomes assessing liver functions that were used in this review.
Blinding (performance bias and detection bias) Subjective outcomes: adverse events	Low risk	Comment: As above. Comment: Blinding was probably compromised and risk of bias due to this may have affected the reporting or detection of some subjective adverse events, but only biochemical liver functions were used in this review.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote from report: "After enrolment, six patients treated with chloroquine and five patients treated with pyronaridine were lost to follow-up. Two additional patients from each group were withdrawn because of self-medication with quinine". Quote from correspondence: "Drop out were mainly lost to follow-up and most often after the patients were cured. We do not think that it was related to intervention." Comment: Equal numbers dropped out from each intervention arm and hence are unlikely to have differentially influenced liver toxicity outcomes used in this review.
Selective reporting (reporting bias)	Low risk	Comment: The trial was not prospectively registered and the trial protocol was not available, but all outcomes stated in methods were reported.
Other bias	Unclear risk	Comment: The extent of elevation in liver transaminases and the proportions re-tested at day 14 were not reported.

Methods	Trial design: Randomized, multi centre, parallel-group, double-blind, double-dummy, non-inferiority trial Period of trial: January 2007 to April 2008
Participants	Number randomized: 1272 Age range: Five to 60 years Gender: Both Inclusion criteria: Age between 3 to 60 years Fever in the last 24 hrs Microscopically confirmed mono-infection withP. falciparum (parasite density: 1000 to 100,000 asexual parasites/µL of blood) Body weight 20 to 90 kg Written informed consent from participants or their guardians, with assent from children able to understand the trial Able to swallow oral medication Willingness to comply with protocol Negative urine test for pregnancy and agreement to practice contraception (women of child-bearing potential) Exclusion criteria: Complicated or severe malaria Mixed infections Anaemia Severe vomiting Malnutrition Hepatitis Hypersensitivity or allergy to trial drugs Use of antimalarials in the previous two weeks by testing; or use of any trial drug for previous four weeks Treatment with any drug metabolised by CYP2D6 Pregnant and lactating women Previous inclusion in a similar trial of artesunate-pyronaridine
Interventions	Randomized in a 2:1 ratio to: Intervention: 1. artesunate-pyronaridine combination (180 mg and 60 mg)* once a day for three days according to bodyweight (N = 849) Control: 2. Artemether-lumefantrine combination (20 mg and 120 mg)** twice a day for three days according to bodyweight (N = 423) Average dose of pyronaridine: 9 mg/Kg body weight (range 13.8 to 7.2 mg/ Kg); artesunate doses ranged from 2.3 to 4.7 mg/kg body weight *Mean artemether dose: 1.7 mg/kg (range 0.9 to 2.4 mg/kg); lumefantrine doses ranged from 5 to 14.4 mg/kg
Outcomes	Primary outcome PCR-corrected adequate clinical and parasitological response rate (APCR; absence of parasitaemia, irrespective of axillary temperature, without previous treatment failure) at day 28 (Sensitivity analysis done with crude APCR (non-PCR-corrected) at day 28) Secondary outcomes Parasite clearance time (time from first dose to first negative parasite reading for two consecutive readings 7 to 25 hours apart) Fever clearance time (time from first dose to being afebrile for two consecutive readings 7 to 25 hours apart) Exploratory efficacy outcomes PCR-corrected and uncorrected APCR rate on day 42 Number of gametocytes per micro-litre at days 0, 3, 7, 14, 21, and 42 Safety outcomes Serious adverse events (death, life threatening, requiring hospital admission or extended hospital stay, resulting in a congenital abnormality or birth defect, persistent disability or incapacity, or other serious adverse event) Other adverse events (during treatment and at follow-up on days 7, 14, 21, 28, 35, and 42) Laboratory abnormalities (days 3, 7; if indicated days 28, 42) Outcome reported but not used in quantitative synthesis in this review Proportion of patients who had cleared parasites at day 1, 2, and 3 Proportion of patients without fever at days 1, 2, and 3 Electrocardiograph abnormalities (days 2, 7, 14, and 28)
Notes	Countries of recruitment: Seven countries in Africa (Democratic Republic of Congo, The Gambia, Ghana; Kenya; Mali; Mozambique; and Senegal) recruited over 1000 participants; remainder were from three sites in two countries in southeast Asia (two in Indonesia; one in the Phillipines) Setting: Local hospitals and clinics Source of funding: Primary sponsor: Medicines for Malaria Venture; secondary sponsor: Shin Poong Pharmaceuticals Endemicity: All are high endemic areas Comments: Age rage of participants: 378 (45%) in the Py-As arm and 182 (43%) in the AL6 arm were aged five to 12 years Food was not required for artemether-lumefantrine (to retain blinding) Non-inferiority was shown if the lower limit of the two-sided 95% CI for the difference between groups was greater than -5% Recrudescence was differentiated from re-infection by PCR genotyping for P. falciparum genes merozoite surface proteins 1 and 2 (msp1, msp2), and glutamate-rich protein (glurp) with at least one matching allelic band in all markers at baseline and after day 7 Grade 3 or 4 toxicity: For alanine aminotransferase, aspartate aminotransferase, and alkaline phosphatase, grade 3 toxicity was 5·1 to 10 times the upper limit of normal and grade 4 toxicity was more than ten times the upper limit of normal. For total bilirubin, grade 3 toxicity was 2·6 to 5 times the upper limit of normal and grade 4 toxicity was more than five times the upper limit of normal. Trials registration: ClinicalTrials.gov identifier: NCT00422084
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote from report: "A computer generated randomisation schedule was provided by Averion AG (Allschwil, Switzerland). Patients were assigned a randomisation code by the investigator in ascending order and allocated to treatment in blocks of nine by study centre."
Allocation concealment (selection bias)	Low risk	Quote from report: "Patients were allocated an individual numbered treatment pack containing sufficient tablets for 3 days therapy plus an overage bottle containing an extra dose in case the patient vomited the first dose. Study packages were allocated on the basis of patient randomisation number. A qualified study team member (third party) who was not undertaking clinical assessments opened the study package and administered the correct amount of tablets, based on patient weight at screening, as instructed by the investigator."
Blinding (performance bias and detection bias) Objective outcomes: parasitological and biochemical	Low risk	Quote from report: "All clinical and laboratory staff and patients were masked to treatment allocation." Quote from report: "Study drugs and placebos were presented in identical packaging. Artemether-lumefantrine placebo was dosed twice daily to maintain blinding. Placebos were of similar shape and colour to their respective active drug." Quote from report: "Sealed opaque envelopes of treatment allocation were provided for use in an emergency, although no code breaks were necessary." Quote from report: "Food was not required for artemether-lumefantrine dosing to retain blinding."
Blinding (performance bias and detection bias) Subjective outcomes: adverse events	Low risk	Comment: See quotes above.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Comment: Data regarding all participants recruited provided in results for all outcomes.
Selective reporting (reporting bias)	Low risk	Quote from report: "There were no changes to study outcomes after trial commencement." Comment: The trial was prospectively registered. No changes were noted in the details provided in the registration document and the study report for outcomes. Comment: Day 42 efficacy outcomes and gametocyte counts were not listed in trial registration document and are listed in the report as exploratory.
Other bias	Unclear risk	Comment: Sponsors designed the trial, were responsible for data collection and analysis, and developed the report; all authors had access to trial data. Comment: Participants on artemether-lumefantrine were not expected to take medication after food; unclear if this reduced bioavailability of lumefantrine, particularly for day 42 outcomes and reinfection rate when lumefantrine levels may have been low.

Study	Reason for exclusion
Cai 1999	Not a RCT: Case series.
Chang 1997	Not a RCT: Case series.
Che 1987	Quasi-RCT. Odd and even numbers used for allocation.
Che 1990	Not a RCT: Controlled clinical trial.
Chen 1989	Not a RCT: Field trial.
Fleckenstein 2007	Not a RCT: Controlled clinical trial evaluating drug pharmacokinetics.
Huang 1988	Quasi-RCT. Randomized according to order of admission.
Huang 1989	Quasi-RCT. Odd and even numbers used for allocation. Compared plain and enteric coated tablets of pyronaridine
Huang 1993	RCT: Conducted in people with complicated falciparum malaria (malignant malaria).
Huang 1996	RCT: Compared single dose versus two days of the same drug combination.
Liu 2002	RCT: Compared pyronaridine + dihydroartemisinin versus dihydroartemisinin alone and with pyronaridine alone. No ACT comparator; no data on liver functions provided in report to include for "Adverse event affecting liver functions".
Looareesuwan 1996	Not a RCT: Clinical trial of two doses of pyronaridine monotherapy with group given second dose recruited after results of first dose were analysed.
Looareesuwan 2007	RCT: Phase II dose ranging trial.
NCT01156389	RCT: ongoing trial evaluating drug interactions in healthy volunteers.
No authors listed 1985	Not a RCT: field trial.
Pang 1989	Not a RCT: controlled clinical trial comparing tablets versus intramuscular injections of pyronaridine/sulfadoxine/pyrimethamine.
Piola 2008	Not a RCT: phase II dose ranging study.
Ramharter 2008	Quasi-RCT. Sequential allocation; Phase II dose ranging trial.
Shao 1991	Not a RCT: case series.
Tan 2008	Not a RCT: Pharmacokinetic study in healthy volunteers.
Wattanavijitkul 2008	Not a RCT: Pharmacokinetic study in healthy volunteers.

Artesunate-pyronaridine compared to artemether-lumefantrine for treating people with uncomplicated falciparum malaria
Patient or population: Adults and children with uncomplicated falciparum malaria Settings: Malaria endemic areas in Africa and Asia Intervention: Artesunate-pyronaridine Comparison: Artemether-lumefantrine
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect(95% CI)	No of participants(trials)	Quality of the evidence(GRADE)
	Assumed risk	Corresponding risk
	Artemether-lumefantrine	Artesunate-pyronaridine
Treatment failure (day 28)	PCR-unadjusted		RR 0.60 (0.40 to 0.90)	1720(2 trials)	⊕⊕⊕○ moderate ^1,2,3,4
	7 per 100	4 per 100 (3 to 6)
	PCR-adjusted		RR 1.69 (0.56 to 5.10)	1650(2 trials)	⊕⊕⊕○ moderate ^1,2,3,5
	1 per 100	1 per 100 (0 to 4)
Treatment failure (Day 42)	PCR-unadjusted		RR 0.85 (0.53 to 1.36)	1691 (2 trials)	⊕⊕⊕○ moderate ^1,2,3,5
	17 per 100	15 per 100 (9 to 23)
	PCR-adjusted		RR 1.53 (0.73 to 3.19)	1472(2 trials)	⊕⊕○○low ^1,6,3,5
	2 per 100	3 per 100 (1 to 6)
The assumed risk is the mean risk across the trials in those treated with artemether-lumefantrine. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio.
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

*Artesunate-pyronaridine compared to artesunate plus mefloquine for treating people with uncomplicated P. falciparum* malaria**
Patient or population: People with uncomplicatedP. falciparum malaria Settings: Malaria endemic areas in Africa and Asia Intervention: Artesunate-pyronaridine Comparison: Artesunate plus mefloquine
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect(95% CI)	No of participants(trials)	Quality of the evidence(GRADE)
	Assumed risk	Corresponding risk
	Artesunate-mefloquine	Artesunate-pyronaridine
Treatment failure (Day 28)	PCR-unadjusted		RR 0.35 (0.17 to 0.73)	1200(1 trial)	⊕⊕⊕○ moderate ^1,2,3,4
	4 per 100	2 per 100 (1 to 2)
	PCR-adjusted		RR 0.38 (0.14 to 1.02)	1187(1 trial)	⊕⊕⊕○ moderate ^1,2,3,5
	2 per 100	1 per 100 (0 to 2)
Treatment failure (Day 42)	PCR-unadjusted		RR 0.86 (0.57 to 1.31)	1146(1 trial)	⊕⊕⊕○ moderate ^1,2,3,5
	8 per 100	7 per 100 (5 to 11)
	PCR-adjusted		RR 1.64 (0.89 to 3.00)	1116(1 trial)	⊕⊕○○low ^1,2,3,5,6
	4 per 1000	6 per 100 (3 to 11)
The assumed risk is the risk in the group treated with artesunate plus mefloquine in the single trial. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio.
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

Liver toxicity of pyronaridine compared to other antimalarials
Patient or population: People with uncomplicated falciparum malaria Settings: High and low-transmission settings for P. falciparum and P. vivax malaria Intervention: Pyronaridine alone or with an artemisinin-derivative Comparison: Another antimalarial
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect(95% CI)	Number of participants(trials)	Quality of the evidence(GRADE)
	Assumed risk	Corresponding risk
	Comparator antimalarial	Pyronaridine alone or with artesunate
Elevated alanine aminotransaminase levels Grade 3,4 toxicity	2 per 1000	10 per 1000 (3 to 30)	RR 4.17 (1.38 to 12.61)	3523(4 trials)	⊕⊕⊕○ moderate ^1,2,3,4
Elevated aspartate aminotransferase levels Grade 3, 4 toxicity	2 per 1000	8 per 1000 (2 to 29)	RR 4.08 (1.17 to 14.26)	3528(4 trials)	⊕⊕⊕○ moderate ^1,2,3,4
Elevated alkaline phosphatase levels Grade 3, 4 toxicity	2 per 1000	1 per 1000 (0 to 5)	RR 0.62 (0.15 to 2.51)	2606(3 trials)	⊕⊕⊕○ moderate ^1,2,3,5
Elevated bilirubin Grade 3, 4 toxicity	3 per 1000	6 per 1000 (2 to 19)	RR 1.92 (0.59 to 6.24)	3067(3 trials)	⊕⊕○○low ^1,2,3,6
The basis for the assumed risk* (for example, the median control group risk across trials) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RR: Risk ratio.
GRADE Working Group grades of evidence High quality: Further research is very unlikely to change our confidence in the estimate of effect. Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low quality: We are very uncertain about the estimate.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Total failure (Day 28)	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
1.1 PCR-unadjusted	2	1720	Risk Ratio (M-H, Fixed, 95% CI)	0.60 [0.40, 0.90]
1.2 PCR-adjusted	2	1650	Risk Ratio (M-H, Fixed, 95% CI)	1.69 [0.56, 5.10]
2 Total failure (Day 42)	2		Risk Ratio (M-H, Random, 95% CI)	Subtotals only
2.1 PCR-unadjusted	2	1691	Risk Ratio (M-H, Random, 95% CI)	0.85 [0.53, 1.36]
2.2 PCR-adjusted	2	1472	Risk Ratio (M-H, Random, 95% CI)	1.53 [0.73, 3.19]
3 Early treatment failure	2	1676	Risk Ratio (M-H, Fixed, 95% CI)	0.74 [0.12, 4.42]
4 Parasite clearance time (hours)	1	1170	Mean Difference (IV, Fixed, 95% CI)	-3.20 [-4.38, -2.02]
5 Fever clearance time (hours)	1	1170	Mean Difference (IV, Fixed, 95% CI)	-1.20 [-2.38, -0.02]
6 Gametocyte clearance time	1	1170	Mean Difference (IV, Fixed, 95% CI)	-10.5 [-12.40, -8.60]
7 Serious adverse events	2	1787	Risk Ratio (M-H, Fixed, 95% CI)	0.92 [0.20, 4.28]
8 Adverse events leading to withdrawal from treatment	2	1787	Risk Ratio (M-H, Fixed, 95% CI)	1.37 [0.67, 2.82]
9 Patient reported symptoms	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
9.1 Headache	1	1272	Risk Ratio (M-H, Fixed, 95% CI)	0.86 [0.60, 1.24]
9.2 Cough	2	1807	Risk Ratio (M-H, Fixed, 95% CI)	0.85 [0.62, 1.17]
9.3 Abdominal pain	1	1272	Risk Ratio (M-H, Fixed, 95% CI)	1.08 [0.67, 1.75]
9.4 Vomiting	1	535	Risk Ratio (M-H, Fixed, 95% CI)	1.58 [0.73, 3.44]
9.5 Pyrexia	1	535	Risk Ratio (M-H, Fixed, 95% CI)	1.46 [0.67, 3.19]
9.6 Influenza- like illness	1	535	Risk Ratio (M-H, Fixed, 95% CI)	1.20 [0.54, 2.70]
10 Patient reported symptoms judged as drug-related	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
10.1 Vomiting	2	1807	Risk Ratio (M-H, Fixed, 95% CI)	1.25 [0.68, 2.31]
10.2 Headache	1	1272	Risk Ratio (M-H, Fixed, 95% CI)	1.03 [0.57, 1.85]
10.3 Abdominal pain	1	1272	Risk Ratio (M-H, Fixed, 95% CI)	1.25 [0.60, 2.57]
10.4 Vertigo	1	1272	Risk Ratio (M-H, Fixed, 95% CI)	1.49 [0.48, 4.61]
10.5 Haematuria	1	1272	Risk Ratio (M-H, Fixed, 95% CI)	2.49 [0.55, 11.32]
10.6 Upper abdominal pain	1	1272	Risk Ratio (M-H, Fixed, 95% CI)	1.12 [0.35, 3.62]
10.7 Anorexia	1	1272	Risk Ratio (M-H, Fixed, 95% CI)	0.33 [0.09, 1.17]
11 Abnormal LFTs; grade 3 and 4 toxicity	2		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
11.1 Alanine aminotransferase (ALT)	2	1807	Risk Ratio (M-H, Fixed, 95% CI)	2.00 [0.43, 9.42]
11.2 Asparatate aminotransferase (AST)	2	1807	Risk Ratio (M-H, Fixed, 95% CI)	3.85 [0.70, 21.09]
11.3 Alkaline phosphatase (ALP)	1	1272	Risk Ratio (M-H, Fixed, 95% CI)	0.10 [0.00, 2.07]
11.4 Bilirubin	2	1807	Risk Ratio (M-H, Fixed, 95% CI)	1.30 [0.30, 5.62]
12 Change in haemoglobin	2		Mean Difference (IV, Fixed, 95% CI)	Subtotals only
12.1 Haemoglobin at baseline	2	1807	Mean Difference (IV, Fixed, 95% CI)	-0.03 [-0.19, 0.13]
12.2 Haemoglobin day 3	2	1755	Mean Difference (IV, Fixed, 95% CI)	-0.01 [-0.13, 0.11]
12.3 Haemoglobin day 7	2	1741	Mean Difference (IV, Fixed, 95% CI)	-0.16 [-0.28, -0.05]
12.4 Haemoglobin day 28	2	1702	Mean Difference (IV, Fixed, 95% CI)	0.04 [-0.09, 0.18]
13 Anaemia as an adverse event	1		Risk Ratio (M-H, Fixed, 95% CI)	Subtotals only
13.1 Anaemia (AE of any cause)	1	535	Risk Ratio (M-H, Fixed, 95% CI)	1.23 [0.68, 2.23]
13.2 Anaemia (drug- related AE)	1	535	Risk Ratio (M-H, Fixed, 95% CI)	0.69 [0.32, 1.47]

Trial ID	Actual or target dose	Intervention(mg/kg/dose)		Comparator(mg/kg/dose)
Trial ID	Actual or target dose	Artesunate	Pyronaridine	Artemisinin-derivative	Partner drug
Kayentao 2012	Actual dose	2.2 to 4.4	6.7 to 13.3	1.3 to 4.0	8.0 to 24.0
Tshefu 2010	Target dose	2.4 to 4.6	7.2 to 13.8	NS	NS
Rueangweerayut 2012	Actual dose	2.4 to 4.7	7.1 to 14.0	2.5 to 5.0	6.2 to 12.5
Poravuth 2011	Target dose	2.4 to 4.6	7.2 to 13.8	—	10, 5, 5
Ringwald 1996	Target dose	—	16, 8, 8	—	10, 5, 5
Ringwald 1998	Target dose	—	8, 8, 8, 8	—	10, 10, 5

Trial ID	Monitoring active or passive?	Outcome data reporting		Blinding adequate?			% of participants included in AE reporting		Independent data analysis
Trial ID	Monitoring active or passive?	Days data collected	Days data reported	Patient	Clinician	Data analysis	AS-Pyr	Control	Independent data analysis
Tshefu 2010	Unclear	0 to 3, 7, 14, 21, 28, 35, 42	Unclear	Yes	Yes	Unclear	100%	100%	No
Kayentao 2012	Unclear	Unclear	Unclear	No	No	Yes	100%	100%	No
Rueangweerayut 2012	Unclear	Unclear	Unclear	No	No	Unclear	100%	100%	No
Poravuth 2011	Unclear	Unclear	Unclear	Yes	Yes	Yes	100%	100%	Yes
Ringwald 1996	Unclear	Unclear	Unclear	No	No	Unclear	Unclear	Unclear	Yes
Ringwald 1998	Unclear	Unclear	Unclear	No	No	Unclear	100%	100%	Yes

Trial ID	Number of tests	Days tested	Days reported	Days tested adequate?	For adequate testing, was reporting complete?	Data analysis independent of sponsor?
Tshefu 2010	4¹	0, 3, 7, 28²	3, 7, 28	Adequate	Complete	No
Kayentao 2012	4¹	0, 3, 7, 28, 42	3, 7	Adequate	Incomplete ³	No
Rueangweerayut 2012	4¹	0, 28, 42	0, “post baseline”	Inadequate	-⁴	No
Poravuth 2011	4¹	0, 3, 7, 28²	3, 7, 28	Adequate	Complete	Yes
Ringwald 1996	4¹	0, 7	0, 7	Inadequate	—	Yes
Ringwald 1998	5^1,5	0, 7	0, 7	Inadequate	—	Yes

Trial ID	Outcome	Artesunate-pyronaridine	Artemether-lumefantrine	P value
Kayentao 2012	Median parasite clearance time	24.1 hours(95% CI 24.0 to 24.1)	24.2 hours (95% CI 24.1 to 32.0)	0.02
	Median fever clearance time	8.1 hours(95% CI 8.0 to 8.1)	8.1 hours(95% CI 8.0 to 15.8)	0.049
	"Post-baseline gametocytes"	95/354 (26.8%)	44/178(24.7%)	0.6
	Gametocyte development(in those negative at baseline)	53/354(15%)	20/178(11.2%)	0.24

Analysis^a	Participants	PCR^b-unadjusted		PCR-adjusted
Analysis^a	Participants	Numerator	Denominator	Numerator	Denominator
Primary analysis	Exclusions after enrolment	Excluded^c	Excluded	Excluded	Excluded
	Missing or indeterminate PCR	Included as failures	Included	Excluded	Excluded
	New infections	Included as failures	Included	Excluded	Excluded
Sensitivity analysis 1^d	As 'Primary analysis' except: missing or indeterminate PCR	—	—	Included as failures	Included
Sensitivity analysis 2^e	As 'Sensitivity analysis 1' except: new infections	—	—	Included as successes	Included
Sensitivity analysis 3^f	As 'Sensitivity analysis 2' except: exclusions after enrolment	Included as failures	Included	Included as failures	Included
Sensitivity analysis 4^g	As 'Sensitivity analysis 2' except: exclusions after enrolment	Included as successes	Included	Included as successes	Included

Criterion	Assessment	Explanation
Patient-reported symptoms
Was monitoring active or passive?	Active Passive Unclear	We classified monitoring as 'active' when authors reviewed participants at set timepoints and enquired about symptoms.
Was blinding for participants and outcome assessors adequate?	Adequate Inadequate Unclear	We classified blinding as 'adequate' when both participants and outcome assessors were blinded to the intervention group, and the methods of blinding (including use of a placebo) were described.
Was outcome data reporting complete or incomplete?	Complete Incomplete Unclear	We classified outcome data reporting as 'complete' when data was presented for all the time-points where it was collected.
Were all participants included in reporting?		We report the percentage of randomized participants included in adverse event reporting.
Was theanalysis independent of study sponsor?	Yes No Unclear	We classified the analysis of trials sponsored by pharmaceutical companies as independent of the sponsor when it was clearly stated that the sponsor had no input to the trial analysis
Laboratory tests
Number of tests undertaken	—	We extracted the type and number of laboratory tests were taken.
Timing of tests Was number and timing of tests adequate?	Adequate Inadequate	We classified the number and timing of tests as 'adequate', when tests were taken at baseline, plus two other timepoints within the first week after treatment, plus the last day of the study. We classed the number of test taken as "inadequate", if either the laboratory controls in the first week or controls at four weeks were not performed.
Reporting of test results Was reporting of test results complete?	Complete Incomplete	We classified reporting as 'complete' when test results of all time points were reported. For the trials with inadequate number of tests taken, we considered completeness of reporting as inconsequential, and therefore did not record a judgement.
Independence of data analysis Was data analysis independent?	Yes No Unclear	We classified the analysis of trials sponsored by pharmaceutical companies as independent of the sponsor when it was clearly stated that the sponsor had no input to the trial analysis

Trial ID	Days tested	ECG results
Trial ID	Days tested	Pyronaridine arm	Comparator arm
Kayentao 2012	0, 2, 7, 14, and 28	"no post baseline clinically important ECG results"	"no post baseline clinically important ECG results"
Tshefu 2010	0, 2, 7, 14, and 28	1 patient with T-wave inversion at day 2 1 patient with ventricular premature complexes and extended QTc (manual reading QTcB 461 ms, QTcF 458 ms) at day 21	1 patient with sinus bradycardia and sinus arrhythmia on day 2 1 patient with sinus bradycardia on day 2
Rueangweerayut 2012	Unclear	6/848 (0.7%) patients with abnormal ECGs- "All were mild and resolved before study completion"1/848 with QT prolongation- None had a QT interval that exceeded 480 msec	3/423 (0.7%) patients with abnormal ECGs - "All were mild and resolved before study completion" 3/423 with long or prolonged QT interval - None had a QT interval that exceeded 480 msec
Poravuth 2011	0, 2, 7, 14, and 42	1/226 (0.4%) patients with QTc prolongations	6/222 (2.7%) patients with QTc prolongations (1/222 not drug-related)

PERMALINK

Artesunate plus pyronaridine for treating uncomplicated Plasmodium falciparum malaria

Hasifa Bukirwa

B Unnikrishnan

Christine V Kramer

David Sinclair

Suma Nair

Prathap Tharyan

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain Language summary

Artesunate plus pyronaridine for treating uncomplicated Plasmodium falciparum malaria

Background

Description of the condition

Description of the intervention

Assessment of antimalarial drug efficacy

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Intervention

Control

Types of outcome measures

Primary outcomes

Secondary outcomes

Adverse events

Search methods for identification of studies

Electronic searches

Searching other resources

Conference proceedings

Reference lists

Contacting organizations and experts

Data collection and analysis

Selection of studies

Data extraction and management

Primary outcome

PCR-unadjusted total failure

PCR-adjusted total failure

Assessment of risk of bias in included studies

Measures of treatment effect

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

Quality of evidence

Results

Description of studies

Results of the search

figure 1.

Included studies

Efficacy trials

Artesunate-pyronaridine versus artemether-lumefantrine

Artesunate-pyronaridine versus artemether plus mefloquine

Additional safety trials

Excluded studies

Risk of bias in included studies

figure 2.

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Effects of interventions

Comparison 1. Artesunate-pyronaridine versus artemether-lumefantrine

Treatment failure

figure 3.

Parasite clearance

Fever clearance