Early parasitological response following artemisinin-containing regimens: a critical review of the literature.
Das D., Price RN., Bethell D., Guerin PJ., Stepniewska K.
BACKGROUND: Parasitaemia on Day 3 has been proposed as a useful alert of potential artemisinin resistance, however, the normal variation of parasite clearance observed in artemisinin-based combination therapy clinical trials is poorly documented. METHODS: The trends in early parasitological response following treatment with an artemisinin anti-malarial regimen were reviewed. A PubMed literature search identified all studies using an artemisinin regimen for uncomplicated falciparum malaria published between January 2000 and December 2011. Data from clinical studies were extracted for analysis using a standardized questionnaire. RESULTS: In total 65,078 patients were enrolled into 213 clinical trials with 413 treatment arms containing either an artemisinin derivative alone (n=26) or in combination with a partner drug (n=387). The proportion of patients remaining parasitaemic at 24, 48 and 72 hours was documented in 115 (28%), 167 (40%) and 153 (37%) treatment arms, respectively. Excluding resistance studies in Cambodia, the median proportion of patients still parasitaemic was 53.8% [range 3-95, IQR=30.5-69.2] on Day 1, 6% [range 0-65.9, IQR=2-11.5] on Day 2 and 0 [range 0-12.6, IQR=0-2] on Day 3. Comparing studies from 2000 to 2005 and 2006 to 2011, the median proportion of patients reported to remain parasitaemic at 72 hours decreased in Africa (1.2% vs 0%, p=0.007), but increased in Asia (0.4% vs 3.9%, p=0.076). In 95% of studies the proportion of patients with peripheral parasitaemia was less than 6% at 72 hours. CONCLUSIONS: These results highlight the normal distribution of early parasitological responses following ACT, and the influence that heterogeneity in study design, host and parasite factors have in confounding a surveillance system based on Day 3 parasite positivity. Greater understanding of factors influencing parasite clearance is crucial, but will require analysis of pooled data from individual patient records.