Background: In multiple sclerosis (MS), up to 57% of white matter lesions are chronically active. These slowly expanding lesions (SELs) contribute to disability progression.

Objective: The aim of this study is to compare fingolimod and natalizumab effects on progressive linearly enlarging lesions (i.e. SELs), a putative biomarker of smouldering inflammation.

Methods: Relapsing-remitting MS patients starting fingolimod (n = 24) or natalizumab (n = 28) underwent 3T brain magnetic resonance imaging (MRI) at baseline, months 6, 12 and 24. SELs were identified among baseline-visible lesions showing ⩾ 12.5% of annual increase, calculated by linearly fitting the Jacobian of the nonlinear deformation field between timepoints obtained combining T1- and T2-weighted scans. SEL burden, magnetization transfer ratio (MTR) and T1 signal intensity were compared using linear models.

Results: The prevalences of fingolimod (75%) and natalizumab patients (46%) with ⩾ 1 SEL were not significantly different (adjusted-p = 0.08). Fingolimod group had higher SEL number and volume (adjusted-p ⩽ 0.047, not false discovery rate (FDR) survived). In both groups, SELs versus non-SELs showed lower MTR and T1 signal intensity (adjusted-p ⩽ 0.01, FDR-survived). Longitudinally, non-SEL MTR increased in both treatment groups (adjusted-p ⩽ 0.005, FDR-survived). T1 signal intensity decreased in SELs with both treatments (adjusted-p ⩽ 0.049, FDR-survived in fingolimod group) and increased in natalizumab non-SELs (adjusted-p = 0.03, FDR-survived).

Conclusion: The effects of natalizumab and fingolimod on SEL occurrence seem modest, with natalizumab being slightly more effective. Both treatments may promote reparative mechanisms in stable or chronic inactive lesions.