BACKGROUND

Anti-malarial artemisinin-based combination therapies (ACTs) might be losing efficacy in east Africa, with the spread of artemisinin partial resistance and reduced partner drug activity. Our trial aimed to measure the efficacies of artemether-lumefantrine, artesunate-amodiaquine, dihydroartemisinin-piperaquine, and artesunate-pyronaridine in three sites in Uganda.

METHODS

This randomised, open-label, phase 4 clinical trial was carried out at three sites in the Agago, Arua, and Busia districts of Uganda. Children aged 6 months to 10 years with uncomplicated Plasmodium falciparum malaria were randomly assigned to receive either artemether-lumefantrine (20 mg artemether; 120 mg lumefantrine; twice a day for 3 days) in all sites or dihydroartemisinin-piperaquine (40 mg dihydroartemisinin and 320 mg piperaquine, once a day for 3 days) in Agago, artesunate-amodiaquine (25 mg artesunate and 67·5 mg amodiaquine for children <9 kg or 50 mg artesunate and 135 mg amodiaquine for children ≥9 kg, once a day for 3 days) in Busia; and artesunate-pyronaridine (60 mg artesunate and 180 mg pyronaridine for children >15 kg or 20 mg artesunate and 60 mg pyronaridine for children <15 kg, once a day for 3 days) in Arua, with follow-up to 42 days. Participants were not blinded to group assignments; however, investigators and those assessing outcome were masked. The primary outcome was parasitaemia, assessed by microscopy, either uncorrected or PCR-corrected to distinguish recrudescence from new infection. All participants who received the treatment per protocol and were not lost to follow-up were included in the primary outcome. All participants who were randomly allocated to treatment groups were included in the safety analyses. This study is registered with the Pan African Clinical Trials Registry, number PACTR202301796134887, and is complete.

FINDINGS

Between Nov 7, 2022, and March 24, 2023, 808 participants (437 [54%] female) were enrolled and assigned to treatment groups; 15 (2%) were lost to follow-up and 793 (98%) completed follow-up. The uncorrected adequate clinical and parasitological response for artemether-lumefantrine was 87 (51·8%; 95% CI 44·0-59·5) of 168 participants in Arua, 88 (51·8%; 44·0-59·4) of 170 and Busia, and 131 (79·4%; 72·3-85·1) of 165 in Agago. This response for artemether-lumefantrine was lower than that of the other ACTs at all sites: 97 (98·0%; 92·2-99·6) of 99 for dihydroartemisinin-piperaquine in Agago, 95 (99·0%; 93·5-99·9) of 96 for artesunate-amodiaquine in Busia, and 73 (73·7%; 63·8-81·8) of 99 for artesunate-pyronaridine in Arua. PCR-corrected 28-day efficacies were 88 (81·5%; 72·6-88·1) of 108 for artemether-lumefantrine and 95 (100%; 95·2-100·0) of 95 for artesunate-amodiaquine in Busia; 131 (97·0%; 92·1-99·0) of 135 for artemether-lumefantrine and 97 (100%; 95·3-100·0) of 97 for dihydroartemisinin-piperaquine in Agago; and 87 (82·1%; 73·2-88·6) of 106 for artemether-lumefantrine and 73 (92·4%; 83·6-96·9) of 79 for artesunate-pyronaridine in Arua. All regimens were well tolerated. The most common adverse events were upper respiratory tract infection, diarrhoea, and anaemia. None of the reported adverse events were attributed to the study drugs. There were two serious adverse events, both cases of severe malaria in Arua, one in each of the treatment groups. Parasite clearance half-lives were prolonged with parasites carrying the PfK13 Cys469Tyr (median 4·2 h; IQR 3·4-4·9) and Ala675Val (4·9 h; 3·4-5·7) mutations compared with wild-type parasites (2·8 h; 2·3-3·6; p<0·0001).

INTERPRETATION

Artemether-lumefantrine was associated with a higher risk of recurrent malaria than other antimalarial combinations tested, and K13 mutations were associated with delayed parasite clearance. Changes in first-line therapy for uncomplicated malaria must be considered in response to suboptimal efficacy of artemether-lumefantrine.

FUNDING

US President's Malaria Initiative, US Agency for International Development, through the Uganda Malaria Reduction Activity and the National Institutes of Health (AI075045 and AI117001).

TRANSLATION

For the Swahili translation of the abstract see Supplementary Materials section.