Background: Fingolimod has shown reductions in clinical and MRI disease activity in patients with relapsing-remitting multiple sclerosis. We further assessed the efficacy and safety of fingolimod in such patients.

Methods: We did this placebo-controlled, double-blind phase 3 study predominantly in the USA (101 of 117 centres). Using a computer-generated sequence, we randomly allocated eligible patients-those aged 18-55 years with relapsing-remitting multiple sclerosis-to receive fingolimod 0·5 mg, fingolimod 1·25 mg, or placebo orally once daily (1:1:1; stratified by study centre). On Nov 12, 2009, all patients assigned to fingolimod 1·25 mg were switched to the 0·5 mg dose in a blinded manner after a review of data from other phase 3 trials and recommendation from the data and safety monitoring board, but were analysed as being in the 1·25 mg group in the primary outcome analysis. Our primary endpoint was annualised relapse rate at month 24, analysed by intention to treat. Secondary endpoints included percentage brain volume change (PBVC) from baseline and time-to-disability-progression confirmed at 3 months. This trial is registered with ClinicalTrilals.gov, number NCT00355134.

Findings: Between June 30, 2006, and March 4, 2009, we enrolled and randomly allocated 1083 patients: 370 to fingolimod 1·25 mg, 358 to fingolimod 0·5 mg, and 355 to placebo. Mean annualised relapse rate was 0·40 (95% CI 0·34-0·48) in patients given placebo and 0·21 (0·17-0·25) in patients given fingolimod 0·5 mg: rate ratio 0·52 (95% CI 0·40-0·66; p<0·0001), corresponding to a reduction of 48% with fingolimod 0·5 mg versus placebo. Mean PBVC was -0·86 (SD 1·22) for fingolimod 0·5 mg versus -1·28 (1·50) for placebo (treatment difference -0·41, 95% CI -0·62 to -0·20; p=0·0002). We recorded no statistically significant between-group difference in confirmed disability progression (hazard rate 0·83 with fingolimod 0·5 mg vs placebo; 95% CI 0·61-1·12; p=0·227). Fingolimod 0·5 mg caused more of the following adverse events versus placebo: lymphopenia (27 [8%] patients vs 0 patients), increased alanine aminotransferase (29 [8%] vs six [2%]), herpes zoster infection (nine [3%] vs three [1%]), hypertension (32 [9%] vs 11 [3%]), first-dose bradycardia (five [1%] vs one [<0·5%]), and first-degree atrioventricular block (17 [5%] vs seven [2%]). 53 (15%) of 358 patients given fingolimod 0·5 mg and 45 (13%) of 355 patients given placebo had serious adverse events over 24 months, which included basal-cell carcinoma (ten [3%] patients vs two [1%] patients), macular oedema (three [1%] vs two [1%]), infections (11 [3%] vs four [1%]), and neoplasms (13 [4%] vs eight [2%]).

Interpretation: Our findings expand knowledge of the safety profile of fingolimod and strengthen evidence for its beneficial effects on relapse rates in patients with relapsing-remitting multiple sclerosis. We saw no effect of fingolimod on disability progression. Our findings substantiate the beneficial profile of fingolimod as a disease-modifying agent in the management of patients with relapsing-remitting multiple sclerosis.

Funding: Novartis Pharma AG.