Cytel Inc

Author Of 1 Presentation

Biomarkers and Bioinformatics Poster Presentation

P0032 - Baseline serum Neurofilament light chain levels predict conversion to McDonald 2005 MS within 2 yrs of a first clinical demyelinating event in REFLEX (ID 1096)

Speakers
Presentation Number
P0032
Presentation Topic
Biomarkers and Bioinformatics

Abstract

Background

Serum Neurofilament light chain (sNfL) is a biomarker of neuronal damage, reflecting disease activity, drug response, and is predictive of future disability in established multiple sclerosis (MS).

Objectives

Post hoc analysis to assess whether baseline (Month [M] 0) sNfL concentration can predict conversion to McDonald (McD) 2005 MS in patients (pts) with a first clinical demyelinating event (FCDE) receiving subcutaneous interferon β-1a (scIFNβ-1a) once (qw) or three (tiw) times weekly, or placebo (PBO) in the phase 3 trial REFLEX.

Methods

Pts randomized to scIFNβ-1a tiw (n=171) or qw (n=175), or PBO (n=171) were followed-up over 2 yrs; converters to 'clinically definite MS' switched to open label scIFNβ-1a tiw. High and low M0 sNfL subgroups were defined by median sNfL concentration (26.1 pg/ml at M0). Median (95% confidence intervals [CI]) time to McD MS (days) by treatment group and M0 sNfL subgroup was calculated by Kaplan Meier. Hazard ratios (HR; 95% CI) to determine factors influencing risk of conversion to McD MS were calculated using a univariate Cox’s proportional hazard model. A stepwise multivariate Cox’s proportional hazard model was performed using factors selected from the univariate model (threshold P<0.15). For both models, variable selection was based on a two-sided Wald test.

Results

High sNfL levels at baseline correlated with the likelihood for conversion to McD MS (low vs high M0 sNfL, HR [95% CI]: 0.58 [0.47; 0.72], P<0.001). Other baseline factors that reduced the risk of conversion to McD MS (univariate model) included: classification of FCDE (mono- vs multifocal: 0.68 [0.55; 0.83], P<0.001) and low numbers of MRI lesions (number of T2 lesions: 1.02 [1.02; 1.03], P<0.001; number of T1 gadolinium-enhancing [Gd+] lesions: 1.14 [1.11; 1.17], P<0.001; number of T1 hypointense lesions: 1.04 [1.02; 1.05]; P<0.001). Furthermore, treatment with scIFNβ-1a tiw (vs PBO: 0.53 [0.41; 0.69], P<0.001) or qw (0.71 [0.56; 0.91], P=0.006) reduced the risk of conversion to McD MS. These findings were confirmed by multivariate models for baseline sNfL subgroup (P=0.024), classification of FCDE (P<0.001), most baseline imaging findings (number of T2 lesions, number of T1 Gd+ lesions; (P≤0.001), and on-study treatment (both P<0.001).

Conclusions

Among other factors, baseline sNfL concentration was identified as a predictor of conversion to McD MS in patients with a FCDE.

Collapse