Siponimod affects disability progression in patients with SPMS independent of relapse activity: Results from the phase 3 EXPAND study

OBJECTIVE: In EXPAND, siponimod reduced risk of confirmed disability progression (CDP) in patients with secondary progressive multiple sclerosis (SPMS).We assessed the impact of siponimod onCDPin patientswith/without relapses to uncouple treatment effects on CDP fromthose on relapses. DESIGN: EXPAND was a phase 3 randomized, double‐blind, placebocontrolled trial. SETTING: EXPAND was conducted at 292 hospitals and multiple sclerosis centres in 31 countries. PARTICIPANTS: Patients (aged 18‐60 years) with SPMS and Expanded Disability Status Scale score of 3.0‐6.5. INTERVENTIONS: Once‐daily oral fingolimod 2mg or placebo for up to 3 years. Main Outcome MEASURES: We analysed the impact of siponimod on CDP by subgroup analysis using the Cox model on time to 3‐month‐ (m‐)/6m‐CDP in patients with/without relapses in the 1 and 2 years before study; principal stratum analysis to estimate the effect in patients who would not have relapsed on‐study at the m12, m18 and m24 timepoints, regardless of treatment; and Cox model on time to 3m/6m‐CDP in the overall population, censoring at time of first relapse. RESULTS: For non‐relapsing patients in 1 and 2 years before study, risk reductions were 18% (HR, 0.82 [CI:0.66;1.02]) and 13% (0.87 [0.68;1.11]) for 3m‐CDP, and 25% (0.75 [0.59;0.96]) and 18% (0.82 [0.62;1.08]) for 6m‐CDP, respectively; for relapsing patients, risk reductions were 33%/33% (3m‐CDP) and 30%/37% (6m‐CDP), respectively. In principal stratum estimates, siponimod reduced 3m‐CDP by 14‐20% and 6m‐CDP by 29‐33% in non‐relapsing patients across the 3 timepoints, suggesting that these patients achieved a large proportion of the effect in the overall population. Cox model censoring at relapse confirmed beneficial effects reaching nominal statistical significance (6m‐CDP.: HR 0.77 [0.62;0.96]). CONCLUSIONS: Siponimod reduces risk of CDP in SPMS patients with/without relapses, indicating that the effects on disability are largely independent from those on relapses. This abstract has been submitted on behalf of the EXPAND study investigators.