Background Ozanimod is a sphingosine 1-phosphate receptor modulator, which selectively binds to sphingosine 1-phosphate receptor subtypes 1 and 5 with high affinity. In the RADIANCE phase 2 study in participants with relapsing multiple sclerosis, ozanimod was associated with better efficacy than placebo on MRI measures and was well tolerated. The RADIANCE phase 3 study aimed to confirm the safety and efficacy of ozanimod versus interferon beta-la in individuals with relapsing multiple sclerosis. Methods We did a 24-month, multicentre, double-blind, double-dummy phase 3 trial in participants with relapsing multiple sclerosis at 147 medical centres and clinical practices in 21 countries. Participants were aged 18-55 years, had multiple sclerosis according to 2010 McDonald criteria, a relapsing clinical course, brain MRI lesions consistent with multiple sclerosis, an expanded disability status scale score of 0.0-5.0, and either at least one relapse within 12 months before screening or at least one relapse within 24 months before screening plus at least one gadolinium-enhancing lesion within the 12 months before randomisation. Participants were randomly assigned (1:1:1) via an interactive voice response system to daily oral ozanimod 1.0 mg or 0.5 mg or weekly intramuscular interferon beta-1a 30 mu g. Participants, investigators, and study staff were masked to treatment allocation. The primary endpoint was annualised relapse rate (ARR) over 24 months. The primary analysis was done in the intention-to-treat population of all participants who received study drug and safety was assessed in all randomly assigned participants who received study drug, grouped by highest dose of ozanimod received. This trial is registered at ClinicalTrials.gov, NCT02047734, and EudraCT, 2012-002714-40. Findings Between Dec 27, 2013, and March 31, 2015, we screened 1695 participants, of which 375 did not meet inclusion criteria. 1320 participants were enrolled and randomly assigned to a group, of whom 1313 received study drug (433 assigned to ozanimod 1.0 mg, 439 assigned to ozanimod 0.5 mg, and 441 assigned to interferon beta-1a) and 1138 (86.7%) completed 24 months of treatment. Adjusted ARRs were 0.17 (95% CI 044-0.21) with ozanimod 1.0 mg, 0.22 (0.18-0.26) with ozanimod 0.5 mg, and 0.28 (0.23-0.32) with interferon beta-1a, with rate ratios versus interferon beta-1a of 0.62 (95% CI 0.51-0.77; p<0.0001) for ozanimod 1.0 mg and 0.79 (0.65 to 0.96; p=0.0167) for ozanimod 0.5 ing. The incidence of treatment-emergent adverse events was higher in the interferon beta-1a group (365 [83.0%] of 440 participants) than in the ozanimod 1.0 mg group (324 [74.7%] of 434) or the ozanimod 0.5 mg group (326 [74.3%] of 439). More participants in the interferon beta-1a group had treatment-emergent adverse events leading to treatment discontinuation than in the ozanimod groups. Incidences of infections and serious treatment-emergent adverse events were similar across treatment groups. No cases of ozanimod-related symptomatic reduction in heart rate and no second-degree or third-degree cases of atrioventricular block were reported. Interpretation In this 24-month phase 3 study in participants with relapsing multiple sclerosis, ozanimod was well tolerated and associated with a significantly lower rate of clinical relapses than intramuscular interferon beta-la. These findings show the potential of ozanimod as an effective oral therapy for individuals with relapsing multiple sclerosis. Copyright (C) 2019 Elsevier Ltd. All rights reserved.