<i>Anopheles</i> salivary antibody biomarkers as surrogate outcomes measures to assess the effectiveness of topical repellent in Southeast Myanmar

Abstract Background Measurement of human antibodies against Anopheles salivary proteins that are injected during mosquito bites may serve as biomarkers of mosquito biting exposure. These biomarkers have been suggested as surrogate outcomes in trials of vector control intervention effectiveness, however, studies to-date have been largely descriptive and do not directly quantify the instantaneous nor cumulative effects of the intervention on antibody outcomes. In this study, we sought to explore the use of a serological biomarker of Anopheles spp. biting exposure as a surrogate outcome in a trial of topical repellent in Southeast Myanmar. Methods In a stepped-wedge cluster randomised controlled trial of personal repellent delivered to 114 villages, we measured the anti-gSG6-P1 IgG antibody response in 14,128 samples by high-throughput ELISA. Generalised linear mixed modelling was used to estimate the effects of repellent distribution (rolled out monthly to blocks of villages) on antibody outcomes overall and for high-risk populations (migrants and forest goers). Results Overall, we detected generally high levels (median: 2.1OD) and seroprevalence (59.9%) of anti-gSG6-P1 IgG antibodies. We observed no instantaneous effect of repellent on antibody levels to Anopheles salivary proteins (mean difference (MD)= 0.01; 95%CI=−0.03, 0.05), however estimation of a series of lagged effects of repellent distribution ( i . e . modelling a gradual antibody decay from prolonged use) showed reduced antibody levels after transition to repellent ( i . e . repellent distribution 6-months prior saw a 0.03-unit (95%CI= −0.08, 0.03) decrease in antibody levels). More specifically, we observed reductions in antibody levels for migrants (6-month lag: MD= −0.10; 95%CI= −0.21, −0.01) and forest dwellers (MD= −0.05; 95%CI=−0.10, 0.00), but not village residents (MD= 0.02; 95%CI=−0.04, 0.08). Conclusions These findings suggest antibodies to Anopheles salivary proteins could be an informative trial outcome measure and provide important parameters on antibody longevity to inform the design of future studies assessing the effectiveness of vector-control interventions.