The effect of dosing regimens on the antimalarial efficacy of dihydroartemisinin-piperaquine: a pooled analysis of individual patient data

Abstract

Background: Dihydroartemisinin-piperaquine (DP) is increasingly recommended for antimalarial treatment in many endemic countries; however, concerns have been raised over its potential under dosing in young children. We investigated the influence of different dosing schedules on DP's clinical efficacy.

Methods and findings: A systematic search of the literature was conducted to identify all studies published between 1960 and February 2013, in which patients were enrolled and treated with DP. Principal investigators were approached and invited to share individual patient data with the WorldWide Antimalarial Resistance Network (WWARN). Data were pooled using a standardised methodology. Univariable and multivariable risk factors for parasite recrudescence were identified using a Cox's regression model with shared frailty across the study sites. Twenty-four published and two unpublished studies (n = 7,072 patients) were included in the analysis. After correcting for reinfection by parasite genotyping, Kaplan-Meier survival estimates were 97.7% (95% CI 97.3%-98.1%) at day 42 and 97.2% (95% CI 96.7%-97.7%) at day 63. Overall 28.6% (979/3,429) of children aged 1 to 5 years received a total dose of piperaquine below 48 mg/kg (the lower limit recommended by WHO); this risk was 2.3-2.9-fold greater compared to that in the other age groups and was associated with reduced efficacy at day 63 (94.4% [95% CI 92.6%-96.2%], p<0.001). After adjusting for confounding factors, the mg/kg dose of piperaquine was found to be a significant predictor for recrudescence, the risk increasing by 13% (95% CI 5.0%-21%) for every 5 mg/kg decrease in dose; p = 0.002. In a multivariable model increasing the target minimum total dose of piperaquine in children aged 1 to 5 years old from 48 mg/kg to 59 mg/kg would halve the risk of treatment failure and cure at least 95% of patients; such an increment was not associated with gastrointestinal toxicity in the ten studies in which this could be assessed.

Conclusions: DP demonstrates excellent efficacy in a wide range of transmission settings; however, treatment failure is associated with a lower dose of piperaquine, particularly in young children, suggesting potential for further dose optimisation.

Figure 1. Patient flowchart.

Figure 2. Available patient data within each age category for (A) dihydroartemisinin and (B) piperaquine.

The patients receiving a total mg/kg dose below the WHO therapeutic range (6 mg/kg and 48 mg/kg, respectively) are shown in dark columns and as a percentage of all patients on top of the bar.

Figure 3. PCR adjusted risk of recrudescent and new infections at day 42 for individual studies.

The full citations for these studies are available in Text S1. The figure excludes data from 6 studies in which active follow-up was stopped at Day 28 [12,17,18,26,29,36]. The results for Ashley-2004 [19] and Ashley-2005 [20] are presented pooled since the datasets did not distinguish between the studies.

Figure 4. Percentiles of predicted risk [5th-median-95th] of recrudescent failure at day 42 in children aged from 1 up to 5 years computed from multivariate model.

Risk was calculated for each individual using their own values. The error bars show the 5th and 95th percentiles of predicted risk of recrudescence failure.

Figure 5. Kaplan–Meier curve for PCR-confirmed recrudescence for children from 1 up to 5 years of age exposed to a dose below or above 59 mg/kg.

Log rank test stratified by study sites p<0.001. The HR for exposure to a PIP dose below 59 mg/kg was 2.36 (95% CI 1.42–3.91), p<0.001 and the AHR 2.03 (95% CI 1.20–3.43), p = 0.008; after controlling for parasitemia and body weight.

Figure 6. Gametocyte positivity rate (GPR) during follow-up.

The error bars show the 95% confidence interval for the positivity rates. The proportions are unadjusted for age and baseline parasitemia. GPR on admission was significantly higher in patients receiving DHA dose ≥6 mg/kg group (p<0.001).