To Örebro University

Introduction: Direct comparisons across multiple disease-modifying therapies (DMTs) for relapsing-remitting multiple sclerosis (RRMS) are valuable in clinical decision making. COMBAT-MS (NCT03193866) is an observational drug trial capturing data on clinical relapses, lesions on magnetic resonance imaging (MRI), Expanded Disability Status Scale (EDSS), and drug survival, at all Swedish university clinics.

Objective: Compare the effectiveness of the most common initial MS therapies in Sweden.

Methods: All first-ever MS treatments with injectables (INJ, interferon-β/glatiramer acetate), dimethyl fumarate (DMF), natalizumab (NTZ), and rituximab (RTX), started 2011-01-01 to 2020-12-14, were identified with prospectively recorded outcome data in the Swedish MS Register. Follow-up continued even if the therapy ended. Missing data were imputed using multiple imputation and potential confounding was adjusted for using stabilized inverse probability of treatment weighting with baseline variables: age, sex, MS duration, geographical region, EDSS, and relapses. All comparisons are made against RTX.

Results: We included 1936 first-ever therapy episodes: 856 INJ, 341 DMF, 270 NTZ, and 469 RTX. Baseline characteristics differed by DMT, with natalizumab having the youngest patients, shortest MS duration, and the most previous relapses.After adjustment, the hazard ratio (HR) for first relapse vs RTX was for INJ 5.9 (95% confidence interval 3.7; 9.5), DMF 2.8 (1.7; 4.8), and NTZ 1.8 (1.0; 3.3). Similarly, the relative three-year lesion rate was for INJ 6.06 (3.75; 9.80), DMF 3.52 (2.01; 6.17), and NTZ 2.03 (1.14; 3.64). EDSS differences at three years were only marginally different: INJ 0.25 (0.06; 0.44), DMF 0.05 (-0.16; 0.26), and NTZ 0.00 (-0.23; 0.24). In contrast, HR for treatment discontinuation was marked: INJ 32.5 (19.0; 55.7), DMF 20.2 (11.5; 35.4), and NTZ 16.2 (8.9; 29.5).

Conclusions: In treatment-naïve patients, RTX was associated with the lowest risk of relapses and MRI lesions, and by far the lowest probability of switching to a second therapy. In contrast, EDSS at 3 years was similar for RTX, DMF, and NTZ, and only slightly higher for INJ. The apparent difference in effectiveness between NTZ and RTX could possibly be explained by the vulnerable period after switching from NTZ, mainly due to JC virus positivity. These findings underscore the importance of tracking long-term outcomes from first DMT start, while considering subsequent therapy switches.