Background

Quinine or alternative artemisinin-based combination treatment (ACT) is the recommended rescue treatment for uncomplicated malaria. However, patients are often re-treated with the same ACT though it is unclear whether this is the most suitable approach. We assessed the efficacy and safety of re-treating malaria patients with uncomplicated failures with the same ACT used for the primary episode, compared with other rescue treatments.

Methods

This was a bicentre, open-label, randomised, three-arm phase 3 trial done in Lisungi health centre in DR Congo, and Kazo health centre in Uganda in 2012–14. Children aged 12–60 months with recurrent malaria infection after treatment with the first-line ACT were randomly assigned to either re-treatment with the same first-line ACT, an alternative ACT, which were given for 3 days, or quinine-clindamycin (QnC), which was given for 5–7 days, following a 2:2:1 ratio. Randomisation was done by computer-generated randomisation list in a block design by country. The three treatment groups were assumed to have equivalent efficacy above 90%. Both the research team and parents or guardians were aware of treatment allocation. The primary outcome was the proportion of patients with an adequate clinical and parasitological response (ACPR) at day 28, in the per-protocol population. This trial was registered under the numbers NCT01374581 in ClinicalTrials.gov and PACTR201203000351114 in the Pan African Clinical Trials Registry.

Findings

From May 22, 2012, to Jan 31, 2014, 571 children were included in the trial. 240 children were randomly assigned to the re-treatment ACT group, 233 to the alternative ACT group, and 98 to the QnC group. 500 children were assessed for the primary outcome. 71 others were not included because they did not complete the follow-up or PCR genotyping result was not conclusive. The ACPR response was similar in the three groups: 91·4% (95% CI 87·5–95·2) for the re-treatment ACT, 91·3% (95% CI 87·4–95·1) for the alternative ACT, and 89·5% (95% CI 83·0–96·0) for QnC. The estimates for rates of malaria recrudescence in the three treatment groups were similar (log-rank test: χ²=0·22, p=0·894). Artemether-lumefantrine was better tolerated than QnC (p=0·0005) and artesunate-amodiaquine (p<0·0001) in the modified intention-to-treat analysis. No serious adverse events were observed. The most common adverse events reported in the re-treatment ACT group were anorexia (31 [13%] of 240 patients), asthenia (20 [8%]), coughing (16 [7%]), abnormal behaviour (13 [5%]), and diarrhoea (12 [5%]). Anorexia (13 [6%] of 233 patients) was the most frequently reported adverse event in the alternative ACT group. The most commonly reported adverse events in the QnC group were anorexia (12 [12%] of 98 patients), abnormal behaviour (6 [6%]), asthenia (6 [6%]), and pruritus (5 [5%]).

Interpretation

Re-treatment with the same ACT shows similar efficacy as recommended rescue treatments and could be considered for rescue treatment for Plasmodium falciparum malaria. However, the effect of this approach on the selection of resistant strains should be monitored to ensure that re-treatment with the same ACT does not contribute to P falciparum resistance.

Funding

Fonds Wetenschappelijk Onderzoek, Vlaamse Interuniversitaire Raad-Universitaire Ontwikkelings Samenwerking, European and Developing Countries Clinical Trials Partnership, and the Belgian Technical Cooperation-Programme d'Etudes et d'Expertises-in the Democratic Republic of Congo.

Since the early 2000s, a downward trend in the malaria burden has been observed, resulting in reductions of 37% in malaria incidence and 60% in malaria mortality worldwide. Nevertheless, the malaria burden remains substantial, with an estimated 214 million cases in 2015, and 438 000 related deaths. ¹ Within this context, the availability of effective treatments remains essential for sustaining these achievements.

Research in context

Evidence before this study

Artemisinin-based combination therapies (ACTs) are recommended for first-line treatment of uncomplicated Plasmodium falciparum malaria. If recurrent malaria occurs before day 28, WHO recommends an alternative ACT in addition to quinine accompanied by an antibiotic having antiplasmodial properties (such as clindamycin, tetracycline, or doxycycline). This recommendation is based on expert opinion. In real-life situations, the prescriber is not always aware of former treatments and patients might be re-treated with the first-line ACT. We searched in PubMed for randomised trials without date or language restrictions, using the MeSH terms “uncomplicated malaria”, “retreatment”, “rescue therapy”, or “second-line treatment”. We did not identify any trial that had investigated the efficacy and safety of re-treatment with the first-line ACT nor with an alternative ACT. Therefore, we undertook this trial to provide evidence. Two malaria-endemic countries using different first-line ACTs were selected to host the trial (artesunate-amodiaquine in the DR Congo and artemether-lumefantrine in Uganda).

Added value of this study

Most ACT treatment failures are reinfections. Our findings show that the treatment of recurrent malaria episode with first-line recommended ACT has a similar efficacy as during the initial first-line treatment and as the recommended rescue treatments. Furthermore, we provide evidence to confirm WHO experts' opinion recommending an alternative ACT as rescue treatment in addition to quinine.

Implications of all the available evidence

Most primary and community health centres in Africa have only one ACT available. The use of the same ACT as a rescue treatment might be considered. However, correct dosing strategies should be implemented and the effect of re-treatment on the population genetics of circulating P falciparum strains needs to be monitored.

Over a decade ago, WHO recommended artemisinin-based combination therapies (ACTs) for the treatment of uncomplicated malaria, in response to the spread of resistance to commonly used antimalarial drugs, namely chloroquine and sulfadoxine-pyrimethamine. Combining two drugs with different mechanisms of action aimed to reduce the emergence of parasitic resistance. ² All malaria-endemic countries have adopted ACT as first-line treatment for uncomplicated malaria. Both DR Congo and Uganda have high malaria incidence. A child younger than 5 years can have up to ten malaria episodes during a year. ^{3 and 4} Artemether-lumefantrine (AL) is the recommended first-line treatment in Uganda, but AL and artesunate-amodiaquine (ASAQ) are both recommended in DR Congo, with ASAQ being the most widely used. Quinine is recommended for any treatment failure, as a rescue treatment.