© 2023, Fondazione Società Italiana di Neurologia.Background: Fingolimod, natalizumab, and ocrelizumab are commonly used in the second-line treatment of relapsing-remitting multiple sclerosis (RRMS). However, these have only been compared in observational studies, not in controlled trials, with limited and inconclusive results being reported. A comparison of their effect on relapse and disability in a real-world setting is therefore needed. Objectives: The objective of this study was to compare the efficacy of fingolimod, natalizumab, and ocrelizumab in reducing disease activity in RRMS. Methods: This multicenter, retrospective observational study was carried out with prospectively collected data from 16 centers. All consecutive RRMS patients treated with fingolimod, natalizumab, and ocrelizumab were included. Data for relapses, Expanded Disability Status Scale (EDSS) scores, and brain magnetic resonance imaging (MRI) scans were collected. Patients were matched using propensity scores. Annualized relapse rates (ARR), time to first relapse, and disability accumulation were compared. Results: Propensity score matching retained 736 patients in the fingolimod versus 370 in the natalizumab groups, 762 in the fingolimod versus 434 in the ocrelizumab groups, and 310 in the natalizumab versus 310 in the ocrelizumab groups for final analyses. Mean ARR decreased markedly from baseline after treatment in all three treatment groups. Mean on-treatment ARR was lower in natalizumab-treated patients (0.09, 95% confidence interval (CI), 0.07–0.12) than in those treated with fingolimod (0.17, 0.15–0.19, p<0.001), ocrelizumab (0.08, 0.06–0.11), and fingolimod (0.14, 0.12–0.16, p=0.001). No significant difference was observed in mean on-treatment ARR between patients treated with natalizumab (0.08, 0.06–0.11) and ocrelizumab (0.09, 0.07–0.12, p=0.54). Compared to fingolimod, the natalizumab and ocrelizumab groups exhibited a higher percentage of relapse-free patients and a lower percentage of MRI-active patients at year 1. No significance differences in disability accumulation were determined between the therapies. Conclusion: Natalizumab and ocrelizumab exhibited similar effects on relapse control, and both were associated with better relapse control than fingolimod. The effects of the three therapies on disability outcomes were similar.