Background: Patients with multiple sclerosis (MS) lose brain volume (BV) faster than healthy individuals.

Objective: Our purpose, within the 12-month phase 3 TRANSFORMS study, was to examine the effect of treatment on BV loss in patient subgroups, establish correlations between baseline normalized BV (NBV) and baseline disease parameters, to identify variables predictive of baseline NBV and on-study percentage BV change (PBVC), and to establish correlations between on-study PBVC and on-study efficacy outcomes.

Methods: Patients received fingolimod 0.5 mg or 1.25 mg, or intramuscular (IM) interferon β-1a (IFNβ-1a) for 12 months. The effect of treatment on PBVC was examined in patient demographic, disease and magnetic resonance imaging (MRI) characteristic subgroups. Pearson's correlation analyses and a stepwise linear regression model were used to identify variables predictive of NBV and PBVC.

Results: Fingolimod reduced BV loss over 12 months versus IFNβ-1a IM in all patient subgroups assessed, including individuals with or without gadolinium (Gd)-enhancing lesions at baseline. Baseline T1 hypointense lesion volume had the strongest correlation with baseline NBV. Baseline Gd-enhancing T1 lesion count was most predictive of change in PBVC over 12 months.

Conclusions: Our results improve understanding of the contributions of different baseline demographic, clinical and MRI characteristics to NBV, including factors that may be predictive of future BV loss.

Keywords: Randomized controlled trial; Trial Assessing Injectable Interferon vs FTY720 Oral in Relapsing–Remitting Multiple Sclerosis (TRANSFORMS) study; brain atrophy; multiple sclerosis; prognosis; volumetric magnetic resonance imaging.

© The Author(s), 2014.