Feasibility of interrupting the transmission of soil-transmitted helminths: the DeWorm3 community cluster-randomised controlled trial in Benin, India, and Malawi

Background: Soil-transmitted helminths are targeted for elimination as a public health problem. This study assessed whether, with high coverage, community-wide mass drug administration (MDA) could lead to transmission interruption. Methods: DeWorm3 is an open-label, community cluster-randomised controlled trial in Benin, India, and Malawi. In each country, a single governmental administrative unit (population ≥80 000 individuals) with soil-transmitted helminth endemicity and participation in at least five rounds of community-wide MDA for lymphatic filariasis, was divided into 40 clusters (population ≥1650 individuals), which were randomly assigned (1:1) to community-wide MDA versus school-based deworming. Laboratory personnel were masked to exposure status and all investigators were masked to post-baseline outcome data until unmasking. In all clusters, preschool-aged and school-aged children received school-based deworming as per national guidelines for 3 years. In intervention clusters, door-to-door community-wide MDA (a single oral dose of 400 mg albendazole) was delivered to all eligible individuals biannually by community drug distributors for 3 years. All individuals aged 12 months and older in India and Benin and aged 24 months and older in Malawi were eligible for treatment, except women in the first trimester of pregnancy, those with adverse reactions to benzimidazoles, those who were acutely ill or intoxicated, or those reporting treatment within the previous 2 weeks. The co-primary outcomes were individual-level prevalence and cluster-level transmission interruption (ie, weighted prevalence of predominant species of ≤2%) of the predominant soil-transmitted helminth species, assessed by quantitative PCR (qPCR) 24 months after the last round of MDA. The analysis set contained a subset of randomly selected participants per cluster who enrolled in the endline assessment, provided a stool sample, and had a qPCR result. All individuals who received treatment were eligible for inclusion in the safety population. This trial is registered with ClinicalTrials.gov (NCT03014167), and is active but not recruiting. Findings: Between Oct 10, 2017, and Feb 17, 2023, 120 clusters (40 clusters per country, comprising 357 716 individuals) were randomly assigned, 60 to community-wide MDA and 60 to school-based deworming. 184 030 (51·4%) individuals in the clusters at baseline were female, 173 663 (48·5%) were male, and 23 (<0·1%) were other. The analysis set consisted of 58 827 individuals in the control group and 58 554 in the intervention group 24 months after the cessation of all deworming, Necator americanus prevalence (the predominant species at all sites) in the community-wide MDA group was lower than the school-based deworming group in Benin (adjusted prevalence ratio [aPR] 0·44 [95% CI 0·34–0·58]), India (0·41 [0·32–0·52]), and Malawi (0·40 [0·34–0·46]). Transmission interruption was achieved for N americanus in 11 (55%) of 20 intervention clusters versus six (30%) of 20 control clusters in Benin (p=0·20), in one (5%) intervention cluster versus no control clusters in India (p=1·00), and in no clusters in either group in Malawi (p=1·00). 984 adverse events were reported among 487 participants over the study, of which 32 among 13 participants resulted in hospitalisation and were classified as serious adverse events (three of which were related to study procedures). Interpretation: Soil-transmitted helminth transmission interruption might be possible in focal geographies but does not appear to be programmatically feasible within the evaluated timeframe. Community-wide MDA should be considered as an alternative strategy to school-based deworming programmes to improve equity and outcomes in helminth-endemic areas. Funding: The Gates Foundation.