Related Experiment Video
Updated: Jun 25, 2026
A Method of Trigonometric Modelling of Seasonal Variation Demonstrated with Multiple Sclerosis Relapse Data
Published on: December 9, 2015
Prognostic Biomarker Signature of Relapse in Neuromyelitis Optica Spectrum Disorders in the CIRCLES Observational
Elea Bach1, Alessandra Breschi1, Yuliang Wang1
1Verily Life Sciences, South San Francisco, CA.
Background And Objectives:
Neuromyelitis optica spectrum disorders (NMOSDs) comprise rare autoimmune diseases of the CNS in which disabilities accrue with relapses. The ability to predict and prevent relapses could dramatically improve clinical outcomes, potentially reducing morbidity and quality-of-life declines. This proteomic study aimed to identify individual and composite candidate serum biomarkers predictive of NMOSD relapse.
Methods:
Patients with NMOSD previously enrolled in the Collaborative International Research in Clinical and Longitudinal Experience Study (CIRCLES) cohort were selected based on documented relapses simultaneous with retrievable banked cryopreserved serum. Longitudinal serum proteomic profiles were characterized using high-resolution mass spectrometry. We used linear models with logistic regression, Cox proportional hazards models with fixed-time intervals, and time-dependent Cox proportional hazards models to analyze individual proteins and proteomic profiles for their association with future relapses factoring demographics, clinical phenotype/course, and treatments.
Results:
We characterized a total of 305 longitudinally collected serum samples (N = 126), using high-resolution mass spectrometry, and identified 265 proteins overall. There was a 10-protein signature with the highest average association coefficient consistently across at least 4 of the 6 modeling analyses, including factor XI, surfactant protein B, C1RL, filamin A, cholesteryl ester transfer protein, cathelicidin antimicrobial peptide, C4A, transferrin receptor, for consistency immunoglobulin kappa constant, and serum amyloid A2 protein. This signature could significantly stratify patients with higher vs lower risk of subsequent relapse. These proteins differed in their increasing or decreasing abundance trajectories in advance of relapse. Most belong to pathways plausibly related to the immunopathology of NMOSD.
Discussion:
Collectively, these findings provide a basis for novel biomarker development to predict NMOSD relapses sufficiently in advance to enable preventive treatment.
