Comparative effectiveness using a matching-adjusted indirect comparison between delayed-release dimethyl fumarate and fingolimod for the treatment of multiple sclerosis.

OBJECTIVE: Delayed-release dimethyl fumarate (DMF; also known as gastro-resistant DMF) and fingolimod are approved oral disease-modifying treatments for relapsing-remitting multiple sclerosis. In phase 3 trials, DMF (DEFINE/CONFIRM) and fingolimod (FREEDOMS/FREEDOMS II) resulted in significant reductions in clinical and magnetic resonance imaging activity, with acceptable safety profiles. Direct comparisons of these treatments are not possible due to a lack of head-to-head trials. We compared 2 year efficacy of DMF versus fingolimod at the approved dosage using a matching-adjusted indirect approach. RESEARCH DESIGN AND METHODS: Individual patient data from DEFINE and CONFIRM, and aggregate data from FREEDOMS and FREEDOMS II, were pooled and compared using the matching-adjusted in-direct method. To account for cross-trial differences, data from trials with available individual patient data were adjusted to match aggregate data (i.e. average patient characteristics) from trials without patient-level data. Data from DMF-treated patients were weighted such that average baseline characteristics matched those of fingolimod-treated patients. After matching, weighted treatment outcomes for DMF-treated patients (240 mg twice daily) were compared with summary outcomes for fingolimod-treated patients (0.5 mg once daily). All comparison results of DMF versus fingolimod used fingolimod as the reference.
RESULTS: After matching, baseline characteristics were balanced between DMF and fingolimod. At year 2, the efficacy of DMF was similar to that of fingolimod for annualized relapse rate (rate ratio [95% confidence interval (CI)]: 1.11 [0.88, 1.40]), 12 week confirmed disability progression (hazard ratio [95% CI]: 0.90 [0.63, 1.29]), and Multiple Sclerosis Functional Composite (mean difference [95% CI]: 0.04 [-0.05, 0.13]). For patient-reported outcomes (EuroQoL 5-Dimensions questionnaire), the mean differences (95% CI) were 0.05 (0.01, 0.08) for utility score and 3.22 (0.58, 5.86) for visual analog scale score, significantly favoring DMF. There was no significant difference in the percentage of patients with no evidence of disease activity (NEDA) for DMF versus fingolimod among matching-adjusted patients with complete NEDA data: rate ratio (95% CI): 0.92 (0.51, 1.64).
CONCLUSIONS: Using the matching-adjusted indirect comparison approach, the efficacy of DMF and fingolimod were similar on all clinical outcomes, while patient-reported outcomes showed greater benefit with DMF. Study limitations include possible confounding from unobserved/unknown differences between trials, and trial length may have been insufficient to detect significant differences on disability progression. CLINICAL TRIAL REGISTRATION: NCT00420212 (DEFINE); NCT00451451 (CONFIRM); NCT00289978 (FREEDOMS); NCT00355134 (FREEDOMS II).