MRI versus relapse: optimal activity monitoring for management of progressive multiple sclerosis.

Secondary progressive multiple sclerosis is often categorized as 'active'/'non-active' based on inflammatory activity on MRI, or relapse; however, the value of MRI/relapse as indicators of disease activity in real-world and clinical trial settings merits further investigation. We separately analysed retrospective data from patients with clinically diagnosed secondary progressive multiple sclerosis in the Adelphi Real-World Disease Specific Programme (a cross-sectional survey) in multiple sclerosis (Adelphi: n = 2554) and the placebo group of the Phase III EXploring the efficacy and safety of siponimod in PAtients with secoNDary progressive multiple sclerosis (EXPAND) trial, [EXPAND-PBO (placebo group of the EXPAND): n = 546] to assess: differences between active/non-active disease in the real-world (characteristics; monitoring); the value of MRI and relapse to indicate disease activity; and the number and characteristics of non-active patients with disease activity in the clinical study. In Adelphi, 1889 patients had 'active' disease (≥1 relapse in the year before index date and/or ≥1 new lesion on most recent MRI) versus 665 with 'non-active' disease (no relapses in the previous year and no new lesions on MRI); median age was 48 versus 53 years; 73.5 versus 87.8% had moderate-to-severe disease; 75.7 versus 54.3% were taking disease-modifying treatment; 87.7 versus 58.7% had received an MRI in the past year. Most active cases (n = 1116; 59.1%) were identified by MRI versus 239 (12.7%) by relapse and 534 (28.3%) by MRI plus relapse. In EXPAND-PBO, 263 patients were classified 'active' (≥1 relapse in 2 years before screening and/or ≥1 gadolinium-enhancing lesion) and 270 'non-active' (no relapse in the 2 years before screening and no gadolinium-enhancing lesion[s]) at baseline; similar proportions of these groups had received disease-modifying treatment prior to placebo: 77.2 and 80.7%. Of non-active patients, 53.0% had disease activity on study; in these patients, 74.1% had disease activity identified by MRI, 8.4% by relapse, and 17.5% by MRI plus relapse. In patients classified non-active at baseline: age and percentage with Expanded Disability Status Scale score 6.0-6.5 were similar between patients with disease activity on study versus patients who remained non-active: 48 versus 52 years; 49.7 versus 56.7%, respectively. In real-world and clinical trial settings, MRI could be a better option than relapse for the identification of disease activity. However, in the real-world, fewer non-active patients had received an MRI in the last year than active patients, which is concerning given that most disease activity in EXPAND-PBO was identified via MRI. We highlight difficulties in consistently identifying disease activity and the negative implications of infrequent monitoring of non-active disease.