Objective:

To report matching-adjusted indirect comparisons (MAIC) of efficacy for diroximel fumarate (DRF) vs ocrelizumab (OCR) and DRF vs interferon beta-1a (IFN), disease-modifying therapies for relapsing multiple sclerosis.

Background:

No randomized trials have compared the efficacy of DRF vs OCR or DRF vs IFN.

Design/Methods:

We obtained individual patient data from EVOLVE-MS-1 (NCT02634307), a 96-week, phase 3 study of DRF (462mg bid; n=1057), and group-level data from OPERA I/II (NCT01247324 and NCT01412333), two 96-week, randomized phase 3 studies of OCR (600mg every 24 weeks; n=827) and IFN (44μg tiw; n=829). EVOLVE-MS-1 data were restricted to meet OPERA I/II inclusion/exclusion criteria and weighted separately to match baseline characteristics in the OPERA I/II treatment arms. Annualized relapse rates (ARR), confirmed disability progression (CDP), and radiological outcomes were compared.

Results:

Baseline characteristics were balanced after restriction and weighting for DRF vs OCR and DRF vs IFN. At 96 weeks, ARR were similar for DRF vs OCR (0.18 [95%CI: 0.14–0.24] vs 0.16 [0.15, 0.17]) but favoured DRF over IFN (0.19 [95%CI: 0.15–0.24] vs 0.29 [0.24–0.33]; p<0.001). At 96 weeks, 12-week and 24-week CDP were similar for DRF and OCR (12-week: 6.4% [95%CI: 3.8–8.7] vs 9.1% [7.1–11.1]; 24-week: 4.8% [2.6–6.9] vs 6.9% [5.2–8.7]); both CDP outcomes favoured DRF vs IFN (12-week: 6.5% [3.8–8.7] vs 13.6% [11.3–15.9; p<0.0001]; 24-week: 4.9% [2.6–6.9] vs 10.5% [8.4– 12.7; p<0.0001]). The proportion of patients with gadolinium-enhancing lesions was higher for DRF vs OCR (16.4% vs 9.1%; p<0.0001) but lower for DRF vs IFN (15.7% vs 33.2%; p<0.0001). The proportion of patients with new/newly enlarging (N/NE) T2 lesions was higher for DRF vs OCR (59.5% vs 38.7%; p<0.0001) but similar for DRF vs IFN (58.4% vs 61.7%). 

Conclusions:

The MAIC indicated no significant differences in clinical outcomes for DRF vs OCR; radiological outcomes favoured OCR. All outcomes favoured DRF vs IFN, apart from N/NE T2 lesions.