Human Direct Skin Feeding Versus Membrane Feeding to Assess the Mosquitocidal Efficacy of High-Dose Ivermectin (IVERMAL Trial)

Abstract

Background: Ivermectin is being considered for mass drug administration for malaria, due to its ability to kill mosquitoes feeding on recently treated individuals. In a recent trial, 3-day courses of 300 and 600 mcg/kg/day were shown to kill Anopheles mosquitoes for at least 28 days post-treatment when fed patients' venous blood using membrane feeding assays. Direct skin feeding on humans may lead to higher mosquito mortality, as ivermectin capillary concentrations are higher. We compared mosquito mortality following direct skin and membrane feeding.

Methods: We conducted a mosquito feeding study, nested within a randomized, double-blind, placebo-controlled trial of 141 adults with uncomplicated malaria in Kenya, comparing 3 days of ivermectin 300 mcg/kg/day, ivermectin 600 mcg/kg/day, or placebo, all co-administered with 3 days of dihydroartemisinin-piperaquine. On post-treatment day 7, direct skin and membrane feeding assays were conducted using laboratory-reared Anopheles gambiae sensu stricto. Mosquito survival was assessed daily for 28 days post-feeding.

Results: Between July 20, 2015, and May 7, 2016, 69 of 141 patients participated in both direct skin and membrane feeding (placebo, n = 23; 300 mcg/kg/day, n = 24; 600 mcg/kg/day, n = 22). The 14-day post-feeding mortality for mosquitoes fed 7 days post-treatment on blood from pooled patients in both ivermectin arms was similar with direct skin feeding (mosquitoes observed, n = 2941) versus membrane feeding (mosquitoes observed, n = 7380): cumulative mortality (risk ratio 0.99, 95% confidence interval [CI] 0.95-1.03, P = .69) and survival time (hazard ratio 0.96, 95% CI 0.91-1.02, P = .19). Results were consistent by sex, by body mass index, and across the range of ivermectin capillary concentrations studied (0.72-73.9 ng/mL).

Conclusions: Direct skin feeding and membrane feeding on day 7 resulted in similar mosquitocidal effects of ivermectin across a wide range of drug concentrations, suggesting that the mosquitocidal effects seen with membrane feeding accurately reflect those of natural biting. Membrane feeding, which is more patient friendly and ethically acceptable, can likely reliably be used to assess ivermectin's mosquitocidal efficacy.

Clinical trials registration: NCT02511353.

Keywords: Anopheles gambiae; direct skin feeding; ivermectin; malaria; membrane feeding.

© The Author(s) 2019. Published by Oxford University Press for the Infectious Diseases Society of America.

Figures

Figure 1.

Trial flowchart.

Figure 2.

Mosquito mortality, stratified by treatment arm and feeding method. Direct skin feeding is indicated by the dashed lines; membrane feeding is indicated by the solid lines. Ivermectin 600 mcg/kg/day is indicated by the green lines; ivermectin 300 mcg/kg/day is indicated by the red lines; and placebo is indicated by the blue lines. Mortality was measured following feeding on day 7 post-treatment. The hazard ratios (95% confidence interval, P value) of mortality during the 14 days post-feeding, comparing direct skin versus membrane feeding for each treatment arm, were adjusted for mosquito clusters. Abbreviation: HR, hazard ratio.

Figure 3.

Capillary versus venous ratios of ivermectin plasma concentration during 2–7 days post-treatment. The open circles represent the capillary versus venous ratios of observed ivermectin plasma concentrations for each sample (n = 177), taken from patients in the main trial contributing capillary samples (n = 61) during 2–7 days post-treatment (maximum 4 samples/patient). The ball-whiskers indicate the median ± interquartile range per sampling day. The horizonal line indicates a median ratio of 1.33 (5th–95th percentiles, 0.98–1.63), based on the trial’s simultaneous pharmacokinetic-pharmacodynamic population model [10]. Adapted from Smit et al [10].

Figure 4.

Direct skin feeding versus membrane feeding ratios of mosquito mortality rates by ivermectin concentration and capillary-venous ratio at the time of feeding. The circles represent the observed 14-day mosquito mortality rate ratios of direct skin versus membrane feeding, performed at day 7 post-treatment for each patient that received ivermectin and consented to direct skin feeding (n = 46), plotted against their day 7 (A) predicted ivermectin capillary plasma concentration and (B) predicted capillary versus venous ratio, using the trial’s simultaneous pharmacokinetic-pharmacodynamic population model [10]. (C) and (D) are as per (B), but now stratified by sex and body mass index, respectively. The lines indicate the linear fits. Abbreviation: BMI, body mass index; DF, direct skin feeding; MF, membrane feeding; mosq., mosquito; mort., mortality; pred., predicted.