2025 American Academy of Neurology Abstract Website

Stefan Bittner¹, Anne H. Cross², Gabriel Pardo³, Scott S. Zamvil⁴, Alit Bhatt⁵, Wenjia Wei⁶, Ibolya Boer⁶, Eric Thouvenot⁷, Tjalf Ziemssen⁸
¹Department of Neurology, University Medical Center of Johannes Gutenberg University Mainz, Mainz, Germany, ²Washington University School of Medicine, Saint Louis, MO, USA, ³Oklahoma Medical Research Foundation, Oklahoma City, OK, USA, ⁴Department of Neurology, UCSF Weill Institute for Neurosciences, University of California, San Francisco, CA, USA, ⁵Novartis Healthcare Pvt. Ltd., Hyderabad, India, ⁶Novartis Pharma AG, Basel, Switzerland, ⁷Department of Neurology, Centre Hospitalier Universitaire (CHU) Nîmes, University of Montpellier, Nîmes, France, ⁸Department of Neurology, University Clinic Carl-Gustav Carus, Dresden, Germany

Objective:

To assess the prognostic value of baseline serum neurofilament light chain (sNfL) levels in treatment-naïve or participants previously treated with disease-modifying therapies (DMTs) before entering ASCLEPIOS I/II.

Background:

Phase 3 ASCLEPIOS I/II trials in people with relapsing multiple sclerosis (pwRMS), baseline sNfL levels were prognostic for on-study lesion formation in the overall population, recently diagnosed treatment-naive participants, and across subgroups.

Design/Methods:

The prognostic value of high (≥9.3 pg/mL) vs low (<9.3 pg/mL) baseline sNfL for the annualized rate of new/enlarging T2 (neT2) lesions was assessed in treatment-naive and previously treated participants by last DMT type prior to study entry (interferon [IFN], glatiramer acetate [GA], dimethyl fumarate [DMF], natalizumab, fingolimod or other MS DMTs (based on investigator’s discretion). The number of neT2 lesions on the last available scan relative to baseline scan was analyzed using a negative binomial model, with time in years between 2 scans as offset. The prognostic value of high vs low sNfL was assessed through lesion rate ratio (RR) attained using this single cut-off.

Results:

The adjusted annualized mean rate of neT2 lesions for high vs low baseline sNfL levels was 4.54/2.02 (RR 2.25; P<.001) among previously treated (n [high/low]=483/510) and 3.46/1.60 (RR 2.16; P<.001) among treatment-naive (n=354/331) participants. By DMT type, the annualized rate of neT2 lesions for high/low sNfL was: IFN (n=220/204) 4.48/2.19 (RR 2.05; P<.001); GA (n=123/154) 4.38/1.92 (RR 2.28; P<.001); DMF (n=39/52) 3.32/1.11 (RR 2.98; P=.005); natalizumab (n=27/11) 7.95/2.33 (RR 3.41; P=.058); fingolimod (n=35/40) 4.35/4.43 (RR 0.98; P=.965); and other DMTs (n=38/49) 2.91/0.62 (RR 4.71; P<.001).

Conclusions:

Baseline sNfL levels were prognostic for neT2 lesions across previously treated and treatment-naive participants. Amongst previously treated patients, baseline sNfL levels were prognostic for future lesion formation across most DMT types. Results support the use of sNfL as a prognostic biomarker in pwRMS with prior DMT exposure, and untreated pwRMS.