Abstract Details

Applying SuStaIn to cross-sectional brain MRI data, 3 multiple sclerosis (MS) subtypes (lesion-, normal-appearing white matter-, and cortex-led) can be identified, and may have different rates of disability progression, relapse rate, and treatment response.

SuStaIn subtype data were available from ASCEND (phase 3 study of NTZ [n=1002] in secondary-progressive MS [SPMS]) and DEFINE/CONFIRM (phase 3 studies of DMF [n = 365] in relapsing-remitting MS [RRMS]). For DEFINE/CONFIRM, comparisons were restricted to the DMF 240mg BID and placebo arms. Associations between baseline SuStaIn subtype probabilities and MS disease characteristics were tested with Spearman correlation. Treatment effect on new T2 lesion count at week 96 within subtypes was estimated by negative binomial regression, adjusting for baseline age, sex, MS duration, and baseline Gd+ count and T2 lesion volume. To determine whether treatment effect differed between subtypes, interaction terms between treatment and subtype were added.

For all studies, the lesion-led subtype correlated with worse brain MRI disease severity at baseline measured by T2 lesion volume, normalized brain volume, and Gd+ lesion count (p<0.05 for all). New T2 lesion count vs placebo in all subtypes was reduced by NTZ in ASCEND SPMS patients (p<0.001) and by DMF in DEFINE/CONFIRM RRMS patients (p<0.05). No significant differences in treatment effect were observed between subtypes for NTZ in SPMS or for DMF in RRMS (p>0.05 for all).

The SuStaIn lesion-led subtype is associated with higher baseline brain MRI disease severity. NTZ and DMF reduced inflammatory disease activity in all SuStaIn subtypes with no significant differences in treatment response. SuStaIn MS subtyping did not discriminate responder heterogeneity as assessed by new lesion formation in these NTZ and DMF trials.