Evaluating the diagnostic test accuracy of molecular xenomonitoring methods for characterising the community burden of onchocerciasis

Abstract

Background Molecular xenomonitoring (MX), the detection of parasite nucleic acid in the vector population, is recommended for onchocerciasis surveillance in elimination settings. However, the sensitivity of MX for detecting onchocerciasis-positive communities has not previously been evaluated. MX may have additional applications for control programmes but its utility is restricted by a limited understanding of the relationship between MX results and human prevalence. Methods We conducted a systematic review of studies reporting the prevalence of Onchocerca volvulus DNA in wild-caught Simulium spp. flies (MX rate) and corresponding prevalence of microfilaria (mf) in humans. We evaluated the sensitivity of MX for detecting onchocerciasispositive communities and describe the characteristics of studies with reduced sensitivity. We conducted a linear regression to evaluate the relationship between mf prevalence and MX rate. Results We identified 15 relevant studies, with 13 studies comprising 34 study communities included in the quantitative analyses. Most communities were at advanced stages towards elimination and had no or extremely low human prevalence. MX detected positive flies in every study area with >1% mf prevalence, with the exception of one study conducted in the Venezuelan Amazonian focus. We identified a significant relationship between the two measurements, with mf prevalence accounting for half of the variation in MX rate (R2 0.50, p<0.001). Conclusion MX is sensitive to communities with ongoing onchocerciasis transmission. It has potential to predict human mf prevalence, but further data is required to understand this relationship, particularly from MX surveys conducted earlier in control programmes before transmission has been interrupted.