Objective: To estimate real-world treatment patterns, safety, and relapse outcomes of subcutaneous (sc) interferon (IFN) β-1a (Rebif) vs dimethyl fumarate (DMF; Tecfidera), to treat relapsing-remitting multiple sclerosis (RRMS).
Methods: A US retrospective chart review of 450 randomly selected adults newly diagnosed with RRMS who received sc IFN β-1a (n = 143) or DMF (n = 307) was conducted. Patients were either (a) treatment-naïve, initiating first-line treatment with sc IFN β-1a or DMF, or (b) previously treated, switching to sc IFN β-1a or DMF. Two years' follow-up data were captured. Patient characteristics, persistence, and adverse events between treatment groups were compared using t-tests or Chi-square tests. Kaplan-Meier curves with log-rank tests and Cox proportional hazards models were used to compare time to, and risk of non-persistence. Annualized Relapse Rates (ARR) were calculated using a robust variance Poisson model adjusting for covariates. Propensity scores were used to address possible selection bias.
Results: One hundred and twelve patients became non-persistent, most commonly due to an adverse event (n = 37). No difference was observed in time to overall non-persistence between sc IFN β-1a and DMF patients. Among treatment-naïve patients, those receiving DMF had 2.4-times the risk (HR = 2.439, 95% CI = 1.007-5.917, p = .0483) of experiencing a discontinuation than patients receiving sc IFN β-1a. Non-persistent patients receiving DMF had 2.3-times the risk (HR = 2.311, 95% CI = 1.350-3.958, p = .0023) of experiencing an adverse event at a given time point than patients prescribed sc IFN β-1a. No differences in relapse risk or ARR between sc IFN β-1a- and DMF-treated patients were observed.
Conclusions: sc IFN β-1a-treated patients had comparable persistence and relapse outcomes, and better safety outcomes vs DMF-treated patients across 2 years.
Keywords: Multiple sclerosis; dimethyl fumarate; interferon β-1a; observational study; relapsing-remitting.