Background Strongyloides stercoralis infection is a neglected condition that places people who are immuno-compromised at risk of hyperinfection and death. Ivennectin is the drug of choice for the treatment of S stercoralis infection, but there is no definitive evidence on the optimal dose. This trial aimed to assess whether multiple doses of ivermectin were superior to a single dose for the treatment of non-disseminated strongyloidiasis. Methods Our study was designed as a multicentre, open-label, phase 3, randomised controlled superiority trial. Participants were enrolled in four centres in Italy, three in Spain, and two in the UK, and recruiting sites were predominantly hospitals. Eligible patients were older than 5 years, weighed more than 15 kg, were residents in an area not endemic for S stercoralis, and either were positive for S stercoralis in faecal tests and on serology (any titre) or had a positive serological test with high titres, irrespective of the result of faecal tests. Patients were randomly assigned (1:1) using a computer-generated, blinded allocation sequence (with randomly mixed block sizes of six, eight, and ten participants) to receive either one dose of ivermectin 200 mu g/kg or four doses of ivermectin 200 mu g/kg (given on days 1, 2, 15, and 16). The primary endpoint was the proportion of participants with clearance of S stercoralis infection at 12 months, which was assessed in all randomly assigned participants who were not lost to follow-up (modified full-analysis set) and in participants in the modified full-analysis set who did not deviate from the assigned treatment regimen (per-protocol set). All participants were included in the safety analysis. The trial was registered with ClinicalTrials.gov, NCT01570504, and is now closed for recruitment. Findings Of the 351 patients assessed for eligibility, 309 recruited between March 26, 2013, and May 3, 2017, were randomly assigned to one dose (n=155) or four doses (n=154) of ivermectin. At 12 months in the modified full-analysis set, 86% (95% CI 79 to 91; 102 of 118 participants) had responded to treatment in the single-dose group compared with 85% (77 to 90; 96 of 113 participants) in the four-dose group (risk difference 1.48%, 95% CI -7.55 to 10.52; p=0.75); similar results were observed in the per-protocol set. Adverse events were generally of mild intensity and more frequent in the multiple-dose than in the single-dose group. The trial was terminated early due to futility. Interpretation Multiple doses of ivermectin did not show higher efficacy and was tolerated less than a single dose. A single dose should therefore be preferred for the treatment of non-disseminated strongyloidiasis. Copyright (C) 2019 Elsevier Ltd. All rights reserved.