Effectiveness of Wolbachia-mediated sterility coupled with sterile insect technique to suppress adult Aedes aegypti populations in Singapore: a synthetic control study

Background: Incompatible insect technique (IIT) coupled with sterile insect technique (SIT) via the release of sterile male Wolbachia-infected mosquitoes is a promising tool for Aedes-borne disease control. Yet, real-world evidence on the suppressive effectiveness of IIT-SIT on mosquito abundance remains mostly limited to small semi-rural village and suburban localities over short trial durations. However, a large proportion of Aedes-borne diseases occur in dense, urban, and high-rise locations, limiting the applicability of previous studies for these settings with high disease burden. The sustainability and use of this technology over multiple years is also unknown.

Methods: In this synthetic control study, we conducted a large-scale, field trial of IIT-SIT targeting Aedes aegypti among high-rise public housing estates in Singapore, an equatorial city state. Routinely collected data from a large, nationwide surveillance system of 57 990 unique mosquito traps, combined with a high-dimensional set of anthropogenic and environmental confounders were collected to ascertain mosquito abundance and its key drivers. Four townships were selected as the intervention groups (approximate population size of 607 872 residents as of 2022), wherein interventions that combined ITT with SIT over the course of the study period were conducted. Townships were subject to releases of wAlbB-SG male A aegypti mosquitoes twice a week. Data were assessed over the course of epidemiological weeks (EWs), which provide the finest temporal resolution of recorded Wolbachia release schedule and mosquito abundance data. A novel synthetic control framework was then developed to account for the non-randomised and staggered adoption setting of the intervention across trial sectors to identify the direct suppressive effectiveness of IIT-SIT on female A aegypti populations, the spillover effects in non-release areas, and the effect of the intervention on other mosquito populations such as Aedes albopictus. Furthermore, we recalculated effectiveness in terms of calendar time, time since intervention, and over multiple sites to examine heterogeneities in IIT-SIT effectiveness.

Findings: Between EW27 2018 and EW26 2022, Wolbachia releases were conducted across 117 sectors, of which 97 had sufficient trap data, which were collected between EW8 2019 and EW26 2022. We found that Wolbachia-based IIT-SIT reduced wild-type female A aegypti populations by a mean of 62·01% (95% CI 60·68 to 63·26) by 3 months, 78·40% (77·56 to 79·18) by 6 months, and 91·32% (90·95 to 91·66) by at least 18 months of releases. We also found a smaller but non-negligible spillover suppression effect that gradually increased over time (mean spillover intervention effectiveness 61·02% [95% CI 57·89 to 63·72] in adjacent, non-intervention sectors). Although no consistent change in A albopictus populations was seen across the four intervention townships after Wolbachia releases, the average intervention effectiveness on the A albopictus population across all release sectors was -25·80% (95% CI -30·93 to -21·05), which was driven by increases in two towns.

Interpretation: Our results demonstrate the potential of IIT-SIT for strengthening long-term, large-scale vector control in tropical cities, where dengue burden is the greatest. The effect of these interventions in different geographical settings should be assessed in future work.

Funding: Singapore's Ministry of Finance, Ministry of Sustainability and the Environment, National Environment Agency, and National Robotics Program.