- Design
- Molecular epidemiology surveys plus clinical cohort
- Population
- Malaria patients and children across Zambia, 2024; treated patients in Kaoma
- Primary outcome
- Pfkelch13 A724E prevalence and day 3 parasite positivity
- Effect
- Day 3 positive in 28% with A724E vs 0% with wild type (P = 0.0026)
Two national cross-sectional surveys in Zambia in 2024 genotyped the Pfkelch13 gene, and a clinical study in Kaoma district followed patients treated with artemisinin combination therapy.
A previously undescribed mutation, A724E, was present in about half of falciparum isolates from Western and North-Western provinces and in about 12% nationally. In Kaoma it rose from none of 15 isolates in 2018 to 79% in 2026. Among patients carrying it, 28% were still parasite-positive on day 3 after treatment, against none of 23 with wild-type parasites.
Day 3 positivity is the clinical marker of partial artemisinin resistance. Combined with mutations already established in east Africa, this pushes the problem further south and west. Ex-vivo susceptibility testing did not show reduced inhibition, which is unusual and needs explaining.
Partial resistance slows clearance rather than causing outright failure while the partner drug still works. For clinicians treating returning travellers, including those from India working in southern Africa, it is a reason to check response rather than assume it.
- Take a travel history covering Zambia and neighbouring countries in every patient with falciparum malaria.
- Repeat a blood smear on day 3 of treatment in patients from affected regions.
- Report delayed clearance to public health or a reference laboratory.
- Continue to treat severe malaria with intravenous artesunate as standard.
Why it matters
Artemisinin partial resistance is no longer confined to east Africa, and partner drugs carry more of the load.
Don't overread it
This is partial resistance — slower clearance — not evidence that artemisinin combinations are failing to cure.
The statistics, in plain English
The day 3 comparison is 21 of 75 versus 0 of 23 patients; the small wild-type group gives a wide confidence interval (0 to 15%). The prevalence rise over eight years is from community and trial samples, which may not be perfectly comparable year to year.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for infectious diseases, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free