**1. Introduction**

Multiple sclerosis (MS) is a chronic inflammatory disease of the central nervous system (CNS) characterized by demyelination and axonal loss [1]. The current concept of MS pathology is based on infiltrating immune B- and T-cells via the blood–brain barrier, local antibody production and activation of glial cells [2,3]. These processes are thought to lead to primary demyelination followed by neurodegeneration [2]. In recent years, several neurochemical markers have been established for the characterization of pathological molecular processes. One of the most extensively investigated

markers for neuroaxonal loss is neurofilament light chain (NfL) [4–6]. NfL is one of four neurofilament subunits and the most abundant one, making it a popular target for neurological diseases [7]. Here, NfL showed superior sensitivity for MS than the phosphorylated subunit of neurofilament [8].

Initially investigated using standard immunoassays, NfL in the cerebrospinal fluid (CSF) from MS patients was found to correlate with disease course and activity [9,10]. In the early phase of the disease, it has a prognostic value [5,11,12] and can be used as a treatment response marker [13]. However, NfL is not specific for MS, is rather a general marker for neurodegenerative processes [14,15] and changes with the normal aging brain [16], which needs to be considered when looking at NfL changes over time.

As detection methods were developed over the years, highly sensitive immunoassays became available and allowed the analysis of brain-derived proteins, not only in the CSF, but in serum as well [17]. Beyond showing a good correlation with CSF values, serum NfL has already thoroughly been investigated in MS [18], i.e., it has been shown to correlate with clinical and radiological disease activity (relapses, new/enlarged T2 lesions and gadolinium-enhancing lesions in magnetic resonance imaging (MRI)) [19–22]. The most important advantage of serum analyses is the possibility of serial sampling and consecutive analysis of biomarkers. Thus, NfL is regularly used in clinical trials to monitor therapy efficacy, and it is on the footsteps of being used as a secondary outcome parameter in clinical trials [23].

In most studies, group effects of treatments on neurofilaments are investigated, which already indicate the applicability of serum NfL as a therapy response marker [22,24] and as a prognostic marker for long-term clinical outcomes in MS [25]. However, longitudinal data of intraindividual NfL levels over disease course under immunomodulatory therapies in well-characterized MS patients are widely missing and only described rarely [26,27].

In this study, we analyzed consecutive samples of MS patients in the early phase of the disease, before and after the initiation of disease-modifying treatment with either glatiramer acetate or interferon-beta over a follow-up period of 24 months. Serum NfL levels at each visit were correlated to clinical outcome parameters (relapse and Expanded Disability Status Scale (EDSS)), serum cytokine profile, cognitive functions and MRI parameters of disease activity and progression.

The aim of this study was (a) to show the effect of immunomodulatory therapies on serum NfL levels in MS patients over disease course, (b) to evaluate the relationship between NfL and MRI parameters reflecting disease progression, such as T2 lesion load, (c) to evaluate possible correlation with cognitive functions and (d) to compare serum NfL levels with the serum cytokine profile.

### **2. Experimental Section**
