Neurofilaments: A Promising Diagnostic and Prognostic Biomarker in Multiple Sclerosis

Abstract

Background

Axonal damage results in permanent cognitive and physical disability during the early as well as progressive stages of multiple sclerosis (MS). However, the mechanism of axonal damage may vary during different stages of MS. The extent and severity of the disease can be estimated by various clinical tests and imaging. However, these measures lack sensitivity and accuracy for understanding subclinical disease activity. Therefore, there is a need for a promising biomarker that can reflect on clinical and subclinical disease activities and have additional diagnostic applications, such as studying MS pathology, disease severity, and rate of relapse.

Summary

Neurofilaments (NFs) are released into the extracellular fluid following axonal insult, and hence, measuring them can help understand the severity of neuro-axonal damage, particularly during early stages of the disease. It can, therefore, serve as a promising biomarker in MS.

Key Message

This review discusses NFs’ biology and its correlation with disease activity, which may have clinical applicability in the diagnosis and prognosis of MS. Furthermore, their levels in response to treatment are also discussed, which can assist in the development of therapeutic strategies.

Keywords

Axonal damage multiple sclerosis neurofilaments biomarker diagnosis and prognosis

Introduction

Axonal damage is considered a major contributor to neurological disability in patients suffering from multiple sclerosis (MS). Despite various technological advancements, axonal damage is yet to be studied comprehensively.^1–4 Axonal damage results in the release of several factors or molecules into the cerebrospinal fluid (CSF). Measurement of these molecules as biomarkers may assist in the diagnosis, prognosis, and eventual devising of therapeutic interventions for MS. N-acetyl aspartate (NAA), an established biomarker for axonal degeneration, lacks specificity to indicate axonal damage; its diagnostic utility is limited.⁵ Lines of evidence indicate that patients with clinically isolated syndromes (CIS) and relapsing-remitting MS (RRMS) have a decreased expression of NAA; therefore, it has limited diagnostic use for clinicians.^5–7 Hence, a promising biomarker is needed, which can provide insights into the axonal damage. Over the past few years, neurofilaments (NFs) have gained interest as a potential biomarker for relapsing and progressive forms of MS.^8–12 Reviews have emphasised the characteristics of NFs and their role as putative biomarkers in relapsing and progressive forms of MS.

NFs: Their Biology, Physiological Functions and Deficiency

NFs are Type IV intermediate filaments, which are most prevalent in axons. However, NFs are scanty and undulated in the dendrites and perikarya. NFs are composed of four subunits, namely NF-light (NF-L), NF-medium (NF-M), NF-heavy (NF-H), and α-internexin in the central nervous system (CNS).¹³ In the peripheral nervous system (PNS), they are made up of NF-L, NF-M, NF-H, and peripherin, with a backbone of the filament composed of NF-L, peripherin or α-internexin. Figure 1 depicts NF structure and its deficiency resulting in behavioural abnormalities, neurodegeneration and neuronal loss. Due to the presence of highly negative charges in medium and heavy filaments, which arise from mostly glutamic acid and phosphorylated serine/threonine residues, both medium and heavy NFs constitute the peripheries more than the central core.¹³ The axonal NF proteins, which are produced in the cell body, are converted into NF polymer. Then, the polymer is relocated through the axon using microtubule motor proteins on microtubule tracks.^13–15

Figure 1.

Factors Released in CSF and Serum, Inducing Axonal Damage in Different Types of MS; CIS = Clinically Isolated Syndrome; RRMS = Relapsing-remitting Multiple Sclerosis; RPMS = Relapsing Progressive Multiple Sclerosis; SPMS = Secondary Progressive Multiple Sclerosis; PPMS = Primary Progressive Multiple Sclerosis; LCR = Cerebrospinal Fluid.

The primary function of NFs is to support the structure of axons, giving stability to myelinated axons, which in turn affects the conduction velocity of electrical impulses along axons.¹⁶ Myelinated axons extend their calibre by acquiring more NFs, following which the conduction velocity increases, whereas deficiency of NF-L, either by deletion or mutation within axons, results in inhibition of axonal growth and a decrease in conduction velocity.¹⁷ The NFs deficiency by modulating the axon calibre causes various neurological manifestations.^18–20

NFs as a Diagnostic Biomarker in MS

MS has been extensively studied as a demyelinating disease associated with axonal-neuronal damage leading to disseminated neurological lesions, which result in autonomic, visual, motor, and sensory disturbances. RRMS constitutes 90% of MS cases, the majority of which transition into secondary progressive MS (SPMS), where neurological function, relapses, and remissions steadily get worse. About 10% of patients experience primary progressive MS (PPMS), which is accompanied by a steady decline in neurological function.^21–22

The levels of NFs are also found to be different in different clinical forms of MS. Figure 1 shows factors released in CSF and serum, inducing axonal damage in different types of MS. According to Hakansson and co-workers, NF-L levels are elevated in both sera and CSF in RRMS and CIS patients. Serum NF-L is highly expressed in RRMS subjects with enhancing lesions and in patients without gadolinium (Gd)+ lesions, which indicates progressive neurodegeneration and inflammatory activities in white and grey matter (GM) regions of the brain.²² It has been shown that serum NF-L levels have a strong connection with lesion MRI activity (Table 1).²³ Additionally, NF-L levels in MS patients exhibit a positive correlation with lesion severity in the brain and spinal cord.²⁴ Serum NF-L was also found to be higher in CIS patients with T2 and Gd+ lesions.²⁵ In a recent report, an increased plasma NF-L also correlates with an increased risk of RRMS patients transitioning into the progressive group.²⁶ Ferarro et al. reported that the level of serum NF-L increases in progressive MS (PPMS and SPMS), which has further neurodegeneration and continued disability with time.²⁷ However, NF-L levels in the blood remain higher in SPMS patients as compared to PPMS patients, which reflects more neuronal and volume loss in SPMS than PPMS.²⁸ Many methods are used to estimate NF-H as a marker for axonal injury or damage, as it is more stable in a phosphorylated state than NF-L.²⁹ Phosphorylated NF-H can be enhanced through the course of progressive MS (Table 1).³⁰ Petzold et al. (2005) demonstrated that all clinical forms of MS exhibit higher levels of NF-H in sera. Specifically, this marker is found to be increased in progressive MS form.³¹ According to Ali-Massoud et al. (2017), increased levels of NF-H are found to be comparatively higher in SPMS patients than RRMS patients, which suggests disease progression and enhanced axonal damage in SPMS cases.³² Additionally, higher serum NF-H levels were found in subjects having Gd+ lesions as compared to those lacking lesions.³² The presence of enhanced lesions proves active inflammation and damaged BBB in the CNS. This marker is also associated with clinical disability and cerebral atrophy.³³

Table 1.

Levels of Neurofilament in Various Clinical Forms of MS.

Clinical Forms of MS			↑ Level of NfL	Compared to	Ref.
Progressive Phase	Relapsing Phase	Preclinical Phase
	RRMS (freedom & transform trial)		↑ bNfL (30.5 & 27.0 pg/mL)	HC	²³
SPMS & PPMS			↑ pNfL (0.12 & 0.14 pg/mL)	RRMS	²⁶
SPMS			↑ pNf-H (38.5%) in serum	RRMS&FDEHC	³⁰
PPMS & SPMS			↑ pNfL (12.8 pg/mL)	RRMS & HC	⁶⁴
	RRMS	CIS	↑ sNfL (17 pg/mL)↑ CSF-NfL (895 pg/mL)	HC	⁷⁵
SPMS			↑ bNfL (32.1 pg/mL) at baseline	PPMS	⁸⁴

Shehab et al. found higher phosphorylated NF-H levels in the sera of MS patients compared to healthy controls. In addition, the relapsing group exhibits a higher phosphorylated NF-H level in sera as compared to the remission group.³⁴ Thus, phosphorylated NF-H level in sera can serve as an indicator for MS activities. Most SPMS sufferers are considered to have high serum phosphorylated NF-H, which indicates the disease severity and broader T2 lesion volumes. It also points out a higher degree of axonal injury.³⁵ However, the report could not establish any correlation between serum NF-H and Gd-contrast-enhancing lesions.³⁵

Although serum NF-L and NF-H are strongly associated with MS activity, NF-L and NF-H in CSF can serve as better markers in various neurodegenerative diseases, including MS, where axonal injury is known to occur.^36–38 It has been pointed out that CSF-NF-L serves as a surrogate marker of neuronal and axonal damage. Petzold et al. established a moderate association between phosphorylated NF-H level in CSF and EDSS score. They also found that CSF-phospho-NF-H becomes higher in progressive MS.³⁹ Significant correlations have been established for CSF-NF-L with the number of further T2 lesions and brain volume loss.⁴⁰ It has also been shown that CSF-NF-L is a biomarker that is simultaneously associated with both a decrease in brain parenchymal fraction (BPF) and severity of T2 lesions.⁴⁰ In a 10-year follow-up study, Bhan et al. revealed that RRMS patients who showed progression tend to have a higher median baseline NF-L level in CSF and are likely to get converted to SPMS after 5 years. However, this association was found to decline with time.⁴¹ Other investigations demonstrate higher levels of CSF-NF-L in PPMS compared to healthy controls and even less when compared to amyotrophic lateral sclerosis (ALS) patients.⁴² Its presence was able to differentiate between the slow progressive phase and the speedy neurodegeneration phase of ALS; it did not associate with disease severity.⁴² Accumulating lines of evidence indicate that relapse is more likely to be associated with elevated NF-L levels in CSF.^43–46 NF-L can also predict the risk of MS development in patients with CIS. After 2-years of follow-up, Domingues et al. found that CIS patients with higher CSF-NF-L levels are at a higher risk of developing MS in the future.⁴³ In agreement, Zetterberg et al. (2019) also reported that high or low CSF-NF-L levels in CIS patients are strongly correlated with increased or decreased risk of future MS development.⁴⁷

MS, an autoimmune-mediated inflammatory disease, is triggered by a confluence of factors like environmental influences, genetic makeup, and infectious agents like viruses.⁴⁸ The autoantibodies mediate a disruptive action against CNS myelin antigens that involves both CD4+ and CD8+ cells. The environmental factors affecting the disease initiation and course include vitamin D deficiency, UV radiation exposure, cigarette smoking, obesity, and EBV exposure, among others. The potential mechanism of infectious agent-induced damage is cross-reactivity, as a molecular mimicry, with CNS myelin antigens. In addition, defective genetic elements, including both MHC and non-MHC genes, are risk factors for the development of MS. Thus, MHC genes are involved in the autoimmune process, whereas non-MHC genes determine regulatory and tolerance mechanisms in MS.⁴⁸

Correlation Between CSF and Serum-NFs

As a result of neuro-axonal damage, NFs are released into the cytosol. The presence of NFs in blood indicates neuronal degeneration and blood-brain barrier (BBB) leakage in the CNS. According to the stoichiometric distribution, NF-L is found four times more in brain tissue as compared to NF-H.²² Also, NF-L is an enhancement component required for axonal and dendritic extension and growth.⁴⁹ Thus, NF-L is conveniently measurable and can serve as a good biomarker for axonal degeneration and disease activity. Even though NFs in CSF serve a role in MS prognosis, obtaining CSF is challenging and often requires multiple invasive lumbar punctures, which is not a viable method. Also, CSF analysis is not sensitive for the detection of lower levels of NFs. In this regard, it has been attempted to understand whether any correlation exists between CSF-NFs and serum-NFs in different stages of MS. Several studies have demonstrated that the level of NFs in sera and CSF is highly correlates.⁵⁰ According to recent research, which included 521 serum and CSF samples, there was a positive association observed between serum and CSF NF-L in distinct MS subtypes.⁵¹ They also found that the levels of serum and CSF NF-L were increased in the relapsing and progressive group of patients. This correlation was suggested through various phases of the disease activity and during therapeutic interventions. MS patients treated with disease-modifying drugs showed a lower level of NF-L in serum and CSF compared to untreated MS patients.⁵¹ Even the temporal course of CSF NF-L was found to be equivalent to serum NF-L. This correlation was also shown in an experimental mouse model of neurodegeneration.⁵² Thus, assessment of serum can be a good alternative to CSF sampling.

Prognostic Features of NFs

To MS Occurrence

NFs can serve as a good prognostic marker in MS. Serum NF-L can often be detected early in the disease course of CIS/RRMS and paediatric MS patients.^{53, 54} Since it is present in the early phase of MS, NF-L can serve as an excellent diagnostic and prognostic marker by enhancing the sensitivity of diagnostic criteria. This facilitates the prognosis of the disease course and can assist in deciding accurate treatment. In a large cohort of MS patients comprising both CIS and RRMS patients, RRMS patients showed a higher level of NF-L in the serum than CIS patients.⁵⁵ It has also been found that the increased level of NF-L improves sensitivity, accuracy, and specificity over OCB and Gd+ lesions, which allows differentiation between CIS and MS patients.⁵⁵ In an earlier report, Norgren and co-workers also found an increased concentration of serum NF-L in MS subjects as compared to healthy controls. However, the limited substantial difference between the MS and HC cohorts failed to decide on an exact cut-off value for NF-L to serve as a diagnostic marker.⁵⁶ Another recent meta-analysis found that different categories of MS patients show a higher level of NF-L as compared to HC, reflecting accurately on disease activity and severity.⁵⁷

For New MS Relapse

Changes in MRI are often caused by changes in new or enlarged T2 and Gd+ lesions values.⁵⁸ According to them, new T1 and T2 enlarged lesions are highly correlated with increasing NF-L concentration.⁵⁸ Another research study that included 25 MS patients showed an increased risk of new lesions. About 20% of MS patients had new lesions after 1-year follow-up; however, after 3 years, 24% of the subjects demonstrated the latest lesions on T2W.⁵⁹ During this time, PBVC (percentage brain volume change) was found to be decreased with the increasing NF-L levels.

For an Increase in Disability

NF works as a good prognosticator for MS. According to Petzold, subjects with RRMS exhibit higher different NFH phosphoforms, that is, NFH-SM135 and NFH-SM134 in CSF, which show a significant correlation with the progression of EDSS.³⁹ Not only is it associated with EDSS, but it also determines the damage that occurred in the visual, sensory, cerebral, and pyramidal systems of RRMS patients. But in the case of patients with SP/PPMS, only NFH-SM134 is responsible for the mass impairments in the pyramidal tracts. Here, progressive MS patients also show minimal disease development in EDSS.³⁹ In another study, Petzold found higher NFH-SM135 in progressive MS, which describes axonal damage, which causes disabilities.³¹ Lines of evidence suggest that MS patients with the first demyelinating event show higher CSF-NfL concentration, which prognosticates future inflammatory activity, and also accurately detects and evaluates various pathophysiological processes of MS.⁶⁰ Again interestingly it was shown that only one patient exhibiting maximum CSF-NfL (3492 pg/mL) did not demonstrate disease activity but gave evidence of cervical myelitis and the presence of large lesions in the spinal cord.⁶⁰ It has been proven that sufferers with spinal relapse display higher CSF NF-L levels as opposed to those with brain relapse.⁶¹ Previously, it has been demonstrated that CSF-NfL, a marker which is associated with Gd-enhancing and T2 lesions, was decreased in MS patients.⁶²

By estimating the association of sNfL135 with 28 MSON, 19 CRION and 9 subjects of NMO, Petzold found that higher sNfL135 causes severe axonal loss in NMO as compared to other diseases.⁶³ NMO patients also possess significant malfunctioning of the visual and retinal nerve fibre layer. This phosphoform of NF is associated with the course of poor visual recovery. Canto also reported a more diverse atrophy rate at different time points between subjects with higher and lower sNfL. This research also revealed increased sNfL in patients with worse disability as compared to clinically stable patients.⁶⁴ Another longitudinal retrospective 5-year study showed that sNfL levels become higher in patients with cognitive impairment (CI) as opposed to the non-CI patients.⁶⁵ The levels of sNfL were also found to be correlated with ongoing and future EDSS grade, especially the hand and walking performance trial and also with the transition of RRMS to progressive MS. Again, Jakimovski in 2019 found that sNfL levels of PwMS patients demonstrated a close relationship with global, cortical, and deep grey matter (DGM) pathology.⁶⁶ Moreover, the baseline levels of NF-L in serum were found to be correlated with the forthcoming longitudinal atrophy rate of WBV (whole-brain volume), GMV (grey matter volume), total DGM, thalamus, putamen, and globus pallidus.⁶⁶

For Progressive MS

Research has demonstrated that the levels of NF-L in CSF were found to be higher in CIS patients who transformed to CDMS in the future, as compared to non-transformers.⁶⁷ According to Sorensen, RRMS patients who convert to SPMS show the prognostic value of CSF NF-L, which can forecast increasing deterioration of the disease.⁶⁸ He also reported that the prognostic value of serum NF-L can predict demyelination and disease severity in the early phase of RRMS disease. In a recent study, the connectivity of CSF-NfL with RIS was examined, which led to the discovery of similar CSF NF-L levels in RIS and CIS/MS.⁶⁹ Additionally, a higher NfL level in CSF also increased the risk of conversion of RIS to CIS and MS.^70–72 The above-mentioned evidence provides a different prognostic value of NF, which demonstrates various results for different subtypes of MS. NF could be used as a good prognostic marker to guide treatment decisions early on.

Levels of NF During Different Types of MS

Around 90% of MS cases experience relapsing-remitting disease, in which subjects suffer from various episodes of neurological impairment. Eventually, the majority of RRMS patients get converted to SPMS, in which neurological functions, relapses, and remissions steadily get worse. About 10% of patients experience PPMS, which is accompanied by a steady decline in neurological function.⁷³ The levels of NF are also found to be different in various forms of MS.

According to Hakansson et al., Serum-NfL and CSF-NfL are found in elevated quantities in RRMS and CIS patients.⁴⁰ Although S-NfL is associated with disease activity, CSF-NfL serves as a better marker for disease activity. During follow-up, they also reported a significant correlation of CSF-NfL with both the number of further T2 lesions and brain volume loss. It has been shown that CSF-NfL is one biomarker that is associated with both BPF decreases and T2 lesions simultaneously. The group also reported that, as compared to S-NfL, the level of NfL in CSF shows better CNS pathology than blood due to a more robust correlation of CSF- NfL with NEDA3 as well as with T2 lesions.⁴⁰ As per another research study, serum NF-L is not only found to be higher in RRMS subjects with enhancing lesions but also found in higher concentration in patients without Gd+ lesions, which indicates ongoing neurodegeneration and inflammatory activities occurring in white and GM.²³ Additionally, it has been shown that the level of plasma NF-L has a strong connection with lesion MRI activity.²³ Shehab et al. found a strong association of phosphorylated serum NF-H with the disease activities and disabilities of RRMS patients.³⁴ They found higher plasma pNF-H in patients than in healthy controls. In addition to this, the relapsing group possesses a higher serum pNF-H level as compared to the remission group.³⁴

Ferarro et al. reported that the level of serum NF-L increases in progressive MS (PPMS, SPMS), which demonstrates active neurodegeneration and continuing disabilities that increase with time.²⁷ Most SPMS sufferers are considered to have high plasma phosphorylated NF-H, and this presence indicates the disease severity and broader T2 lesion volumes. It also points out a higher degree of axonal injury.³⁵ But he did not find any correlation of serum NFH with Gd-contrast-enhancing lesions. Additionally, it was shown that in comparison to pNF-H-subjects, pNF-H+ SPMS patients show a higher mean of MSSS (multiple sclerosis severity score), a measurement through which disease severity can be assessed.³⁵ Petzold et al. established a moderate association of CSF pNF-H level with EDSS score and also found that CSF-pNF-H level becomes higher in progressive MS.³⁰ After 5 years of follow-up, Bhan and co-workers revealed that RRMS patients who show progression tend to have a higher median baseline CSF-NF-L level, which is converted to SPMS, and they also show higher EDSS value. As compared to a 5-year follow-up of CSF-NfL level, this association was found to be weaker after 10 years, but still, the effect of CSF-NfL continued for a long time.⁴¹ As per recent research, an increased plasma NF-L also correlates with an increased risk of RRMS patients converting into the progressive group.²⁶ In another study, the level of CSF-NfL was found to be higher in PPMS than in HC and substantially less when compared to ALS patients. Its appearance was also able to differentiate between the slow progressive phase and the speedy neurodegeneration phase of ALS, but did not associate with disease severity.⁴²

Martin et al. showed that progressive MS patients have higher CSF-NfL levels as compared to healthy individuals, but in contrast to RRMS, the levels of CSF-NfL are reduced.⁴⁶ Other lines of evidence indicate that the level of NfL is lower in progressive MS than in RRMS^43–45 and HC.⁷⁴ Kuhle et al. demonstrated NfL levels increase in the blood of SPMS patients as compared to PPMS patients, which reflects more neuronal and volume loss in SPMS than PPMS.²⁸

Serum NF-L was found to be higher in CIS patients who demonstrated T2 and Gd-enhancing lesions. This marker shows a significant correlation with EDSS, which quantifies disabilities.²⁵ Other lines of evidence suggest that after 2 years of follow-up, CIS patients who possess a higher level of NfL above 900 ng/L have an increased risk of getting converted to MS.³⁸ CIS patients who got converted to MS displayed a higher mean of CSF-NfL level, whereas the non-converter CIS patients after 2 years had lower CSF-NfL levels. Domingues also stated that, as compared to NFH, NfL shows more discriminatory power with different forms of MS. According to Zetterberg et al., the presence of higher or lower CSF-NfL in CIS patients is strongly correlated with increased or decreased risk of future MS. Higher CSF-NfL is also vigorously associated with different disease activities and disease severity.

Relating NF with Disease Activities

NF During Neuro-axonal Damage

Neuro-axonal damage of MS has been acknowledged for a long period, but still, the mechanisms of this disruption are not accurately illustrated. Neuro-axonal damage has been demonstrated in MS patients. In the case of neurodegeneration, it has been shown that neuro-axonal pathology has a positive correlation with the rate of inflammation, especially if the patient suffers from progressive MS.^75–77 This axonal injury can also be affected by different categories of actuators, such as frontal assault of cytotoxic CD8⁺T cells and contactin-2/TAG-1,^{78, 79} impairment and failure of mitochondria,^{80, 81} glutamate-mediated excitotoxicity,^{82, 83} disruption of calcium homeostasis,⁸⁴ production of several reactive oxygen species and cytokines and many more. It has been proven that neuro-axonal damages are also associated with various destructive scenarios of CNS, like acute and chronic demyelinated regions, adverse modifications in sodium channels,⁸⁵ and interrupted axonal-glial interactions,^{86, 87} etc.

According to recent research, following EAE, the level of NF-L within the supernatant of the cerebrum was reduced (23%, 42%, and 36% on Day 1, 3, 7) in the early phase but increased (66%, 13% on Day 14, 28) in the later stages.⁸⁸ Again, the level of NF-L within the pellet was dropped (22%, 16%, 15%, on Day 1, 3, 7) and found elevated (28%, 128% on Day 14, 28) at the last stage of EAE. Like NF-L, the levels of NF-M and NF-H are also found modified in the later stages of MS. Mainly, this level of NfL indicates axonal damage, which is related to inflammation, whereas NfM and NFH levels are associated with nerve regeneration. In another longitudinal study, which included 141 RRMS patients, Bsteh explained that after 3 years of follow-up, an increasing sNfL concentration was correlated with retinal axonal loss, due to which abnormal thinning of the pRNFL was detectable.⁸⁹ He also stated that thinning of the peripheral retina developed rapidly in subjects with more disabilities and CI. It has been proven that a higher sNfL is substantially associated with T2 lesion volume (r = 0.55, p < .0001).⁵³ This increasing level of sNfL not only depletes the brain parenchyma within 6–37 months but also forecasts GD-enhancing lesions with 84% sensitivity and 66% specificity. These findings mirror elevated sNfL, which indicates neuro-axonal damage in MS patients.⁵³

Similarly, in sNfL, the level of NfL in CSF became higher in MS (RRMS) patients as compared to HC.⁹⁰ According to Norgren, as compared to the relapsing group of patients, higher CSF-NfL was found in patients with SPMS, which indicates more axonal degeneration in progressive MS.⁵⁶ Spinal cord lesions are also another important substrate of MS. As stated by Schrimer, it has been observed that reduced phosphorylated NFs (pSM131) were associated with axonal loss in WM lesions.⁹¹ Again, a reduced neural SM132 IR was found in ventral spinal neurons, which reflects lesion development and disease progression. A higher level of CSF NF-L, along with tubulin and actin (axonal cytoskeletal proteins), was identified in MS patients with neuro-axonal injuries. These activities were characteristically seen more in progressive MS than RRMS.⁹² Figure 2 indicates NF level changes related to axonal damage, brain inflammation and brain and spinal cord atrophy.

Figure 2.

Neurofilament Level Changes Related to Axonal Damage, Brain Inflammation, and Brain and Spinal Cord Atrophy.

NF During Brain Inflammation

Neuro-inflammation is another significant cause of impairment in MS. MS itself is an inflammatory disease that commonly begins as an RRMS form that may later convert into progressive MS. Mostly, inflammation is explained by the existence of perivascular T and B lymphocytes and their entrance into the CNS (parenchyma). Generally, inflammatory activities are of different types. First, inflammation can happen due to the disruption of the BBB, for which large quantities of B and T lymphocytes are released, enter the white matter, resulting in the formation of demyelinated plaques. Due to a huge influx of lymphocytes, the expression of cytokines and chemokines increases, which causes inflammatory activities. Especially, this type of inflammation is found in RRMS sufferers. Second, in the absence of BBB damage, an inflammatory mechanism may occur due to the slow accumulation of T and B cells into the connective tissue spaces of the brain, specifically seen at the beginning phase of MS, which increases gradually with age and disease duration.⁹³ Another kind of inflammation is correlated with the presence of demyelinated lesions in the cortex, for which the neurodegeneration process occurs in white and GM.^{93, 94} According to Khademi, the relapsing group of subjects at a younger age had shown high inflammatory activity, which was proved by measuring CSF-NfL; on the other hand, progressive MS exhibits little evidence of inflammation.⁴³ As per recent research, it has been proven that RRMS patients exhibit higher independent inflammatory activity due to increased concentration of NfL in subjects with or without GD lesions as compared to HC (Figure 2).²³

Jakimovski in 2019 stated that sNfL level of RRMS patients has an association with high inflammatory and neurodegenerative activities. Even higher sNfL was found in PMS (PPMS/SPMS) patients who generally demonstrate a larger percentage of meningeal infiltration (that is referred to as tertiary follicle-like str), along with greater T1- LV and T2-LV. However, these subjects displayed lower inflammatory activity.⁶⁶ In the previous study, it has been proven that GD lesions are associated with inflammatory activities. In addition to this, subjects with RRMS demonstrate more GD and T2 lesions than SPMS; even patients with PPMS also reveal lesser GD and T2 lesions than RRMS. Thus, this finding signifies higher inflammatory activity in RRMS as opposed to the progressive group of MS.⁹⁵ While diagnosing RRMS, Bittner found an elevated level of NF-L in serum, which specified more axonal damage and neuroinflammation than in those with CIS (Figure 2).⁵⁵ Novakova in 2017 stated that both RRMS and progressive MS sufferers possess higher sNfL with increased disease activities as compared to HCs (without disease activities).⁵¹ In another study, it has been proven that CSF-NfL shows a significant association with NEDA-3 concept, number of new T2 lesions, brain volume loss, etc., than sNfL. Thus, CSF-NfL can forecast disease activities better as compared to sNfL. Hakansson also revealed a positive association between CSF-NfL and elevated inflammatory biomarkers like CXCL10 and CXCL13.⁴⁰ Also, it has been found that CIS converters possess a higher level of CXCL10 and MMP-9 as compared to non-converters.

NFs in Brain and Spinal Cord Atrophy

Brain and spinal cord atrophy portray various forms of injury, irreversible pathological processes, and different disabilities. Measurement of atrophy holds great potential to assess subtypes of MS and their reaction to different remedies. After investigating 43 RRMS patients, Kuhle found a positive association of baseline sNfL with a future prediction of brain atrophy within 12 and 24 months.²² The strongest connection between the marker and brain atrophy was found in 24 months. Due to a higher baseline NfL level, brain volume reduces at 12 months in subjects and a rapid reduction was identified in 24 months (Figure 2). By measuring brain atrophy, Kuhle also did not find any impact of therapy Riluzole on Nf-L of RRMS subjects.²² In comparison to HC, early RRMS sufferers possess higher baseline sNfL, which is indicative of developing more Gd-enhanced lesions.²² Barro in 2018 provides substantial evidence that sNfL is a superior prognosticator of the brain and spinal cord atrophy as compared to MRI measurement of different lesions conducted at 2 and 5 years.²⁴ After 2 and 5 years of follow-up, the percentage of reduction in brain volume of subjects was 0.17% and 0.35%. Like changes in brain volume, the percentage of decreasing spinal cord volume was 0.19% and 0.49% during 2 and 5 years in patients. It has been seen that over 2 to 5 years, the mean reduction of the brain and spinal cord volume gradually increases with increasing sNfL. These observations proved that higher sNfL causes more loss of brain and spinal cord volume and can better predict brain and spinal cord atrophy.²⁴

Elevated NFH SM135 was found to correlate with the atrophy of the precentral gyrus, from where transsynaptic axonal degeneration can be predicted.⁹⁶ Again, Petzold also found a positive enhanced correlation between frontal lobe atrophy and NFHSM135, through which evaluation of the deteriorated function of the frontal lobe and cognition was possible. Likewise, precentral and frontal atrophy, various CNS localised atrophies such as postcentral gyrus, accumbens, amygdala, caudate, hippocampus, pallidum, thalamus, and temporal areas were associated with higher NFHSM135. Most pronounced atrophy was found in the putamen. This stated evidence proved that higher CSF NF levels, especially NFHSM135, portended failure of axons, which was responsible for causing severe disabilities and brain atrophy in different sites.⁹⁶ Following 10 years, Chitinis in 2018 found that averaged increasing serum-NfL levels annually, which were inextricably linked with whole-brain atrophy, long-term disabilities, as well as with worsening of fatigue scores and MRI-derived brain lesions, etc.⁹⁷ By researching 301 RRMS patients, Kuhle found a persistent association of CSF and serum NF-L levels with lengthy brain atrophy and an extended period of various clinical outcomes like EDSS.⁹⁸ Also in this survey, Kuhle initiated that the concentration of NF-L, along with short-term brain atrophy, gave robust proof about the increase in long-term EDSS score rather than considering NfL levels only.⁹⁸ According to Williams in 2020, NfL levels in CSF and serum have shown their essential capacities as a marker, which is required not only to analyse future brain and spinal cord atrophy but also to evaluate active neuroinflammation and much more.¹² He also noted that, as compared to lower (<30 pg/mL) bnfl categories of patients, higher bnfl (>60 pg/mL) subjects showed twice as high a rate of brain atrophy, at 24 months.

The association between NF and cerebral atrophy is not only found in MS but also seen in other neurodegenerative diseases. For example, Dhiman in 2020 demonstrated higher CSF-NfL levels linked with brain atrophy of Alzheimer’s patients.³⁶ According to Byrne, NfL levels in serum work as a predictor of brain atrophy in Huntington’s disease.⁹⁹

NF Causes Disability

Disabilities are consequences of MS disease, which are measured using different scales. As per the previous studies, higher baseline NfL levels in CSF could predict increasing forthcoming disabilities in different forms of MS,^{100, 101} whereas, in another survey, Nf-H was considered as a better prognosticator for evaluating PBVC and EDSS through which disabilities were quantified.²¹ As per the latest research, several forms of MS patients who possess higher bNfL were strongly linked not only to upcoming disabilities but also to increased ongoing disabilities.^{28, 102} But in a recent study, the connectivity of SNF with future disability progression was unfindable by EDSS.²⁷ By determining the value and rate of disabilities through various scales such as EDSS, nine-hole peg, 25-foot walk test, etc, it has been proven that the concentration of NF-L in both CSF and serum was correlated with ongoing impairments of MS.^{22, 103, 104} Kalatha also found a positive link between the current cognitive condition and NF-L levels in CSF.¹⁰⁵

Similarly, NF-L, cNF-H, and bNF-H concentrations also reflect an integral measure of recent and ongoing impairments of progressive MS.^{100, 106, 107} From other surveys, it has been found that future disabilities are associated with cNfH as well as with a high blood-CSF NFH ratio.¹⁰⁸ However, this positive connection was not traceable in some other studies.¹⁰⁷ Kuhle in 2011, found that higher NFHSM135 had a significant relationship with EDSS progression in the relapsing group of subjects.⁷⁰

NFs as a Tool to Measure Response to Therapeutic Intervention

Usually, NF levels in both CSF and blood of patients are reduced upon receiving different drug-modifying therapies (DMTs) comprising initial treatment (glatiramer acetate, interferon ß, teriflunomide, and dimethyl fumarate), second phase treatment (fingolimod, natalizumab, alemtuzumab, siponimod, rituximab, mitoxantrone, and cyclophosphamide) and autologous haematopoietic stem cell transplantation.

RRMS patients treated with dimethyl fumarate as a first-line DMT demonstrated lower NfL concentration in CSF and blood, but a relevant decline was seen after 6 months.¹⁰⁹ The most significant reduction of NF-L was noticed in plasma rather than serum and CSF. The relapsing subjects who were given fingolimod treatment possess lower NfL levels in the CSF (up to 83% after 1 year),¹¹⁰ serum (35/36% after 6 months),^{23, 28} and plasma (34% after 1 year).¹¹¹ This medication’s impact was expeditious as it showed its result within 6 months by steadying the marker level.²³ The long-term benefit of fingolimod (0.5 mg) was proved through a reduction of bNfL around 40% at the end of crore and end of the study, with an improved clinical outcome in RRMS patients.¹¹² When RRMS patients received extended treatment of fingolimod, the levels of NF-L continued to decrease up to 10 years.¹¹²

In a cohort of 92 natalizumab-treated MS subjects, the NfL concentration in CSF was reduced. This marked decline of the marker was found after 6 or 12 months of treatment.¹¹³ The mechanism of natalizumab is not only to minimise NF-L levels but also to reduce various disease activities such as relapse rate, lesion formation, inflammatory reaction, etc. In line with this study, a significant reduction of NfL was found in response to natalizumab therapy, and this decline was found to be associated with an increase of MTR value in normal appearing white matter (NAWM) and GM.¹¹⁴ Measurement of sNfL levels was not simply utilised to inspect positive effects of natalizumab but also used to distinguish from progressive multifocal encephalopathy (PML) as well. Most of the MS patients treated with natalizumab develop PML, a brain infection that is caused by the JC virus. Another study, which comprises a large MS cohort of 312 sufferers, displays a 10-fold higher sNfL in PML as compared to pre-PML subjects.¹¹⁵ Even at PML onset, the level of NfL in serum was found to be higher than in other natalizumab-treated remitting/relapsing patients. In accordance with the threshold of 52.5% pg/mL, natalizumab-treated MS patients who developed PML were accurately recognised with a sensitivity of 96.0% and specificity of 94.9%. On the other hand, a higher NfL threshold of 61.3% pg/mL demonstrated a higher ability to differentiate between PML sufferers and subjects who experience relapse, with different values of sensitivity and specificity, that is, 88.0% and 100.0%.¹¹⁵

Most of the alemtuzumab-treated MS subjects possess lower NF-L concentrations in serum within the first 6 months.⁷⁵ But according to some of the studies, it has been seen that in the first 4–5 months after the administration of alemtuzumab, there was an increase in sNfL level along with clinical and MRI disease activity occurring before the complete effectiveness of treatment.¹¹⁶ This increase of sNfL was correlated with a higher occurrence of relapse, T2 and Gd lesions, etc., which initiated patients for ATZ retreatment.

Recently, it has been proven that MS patients who received highly effective second-line therapies (natalizumab, rituximab, mitoxantrone, cyclophosphamide, fingolimod) as compared to effects of other DMT therapies (interferon beta-1b, interferon beta-1a, and glatiramer acetate, monthly pulsed dose glucocorticoids, azathioprine, mycophenolate mofetil, and teriflunomide) showed a greater reduction of sNF-L.^{33, 37, 51, 64} In another study, Delcoigne et al. demonstrated that the rate of sNfL reduction also varies according to the type of DMT used for RRMS patients.³³ They stated that the highest reduction of NfL was found for ATZ and the lowest for TFL. Most MS sufferers have shown a similar decrease in NF-L level in serum with DMF, FGL, and NTZ, but among these DMT therapies, NTZ-treated RRMS patients had high potency on relapse and MRI lesions.³³

Perspectives and Some Open Questions

NF has been proven to be an important biomarker for diagnosis, prognosis, and tracking disease activity of MS. Not only is NF specific for MS, but it also works as a marker for different CNS diseases such as Alzheimer’s, Dementia, Parkinson’s, and so on. Quantification of NF is a screening tool for MS patients through which the pathology, current and future disabilities, and future risk of MS can be detected. Furthermore, a rise in NF levels is seen in relapse and MRI lesions; the measurement of this marker also provides data about inflammatory disease activities. By evaluating this marker in blood and CSF, the pathology and severity of neurodegenerative diseases may be predicted, but several questions remain unanswered. For instance, what is the functional role of the variety of subunit compositions of NFs which are observed in different neurons? Why does high sNfL without brain lesions possess subclinical disease activity? What is the best use of NfL in clinical practice? Therefore, with many unsolved challenges still present, further studies are needed for better clarification of NF assessment, degenerative brain pathology and its relationship with other markers, and early disease mechanisms.

Footnotes

Authors’ Contributions

DM, SVM, SR, RK and FJLI wrote the original draft; DM, GLR and BC conceptualised the review content; BC and GLR designed the framework. DM, RV, GLR performed the graphical illustrations, BKM, GLR and BC provided critical comments and suggestions while reviewing and editing the manuscript.

Statement of Ethics

NA.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

ICMJE Statement

Patient Consent

NA.

ORCID iDs

Deepali Mathur

Gerardo López-Rodas

Rohin Vinayak

References

Correale

, Marrodan

and Ysrraelit

MC.

Mechanisms of neurodegeneration and axonal dysfunction in progressive multiple sclerosis. Biomedicines 2019; 7(1): 14. DOI: 10.3390/biomedicines7010014.

Duncan

, Simkins

and Emery

Neuron-oligodendrocyte interactions in the structure and integrity of axons. Front Cell Dev Biol 2021; 9: 653101. DOI: 10.3389/fcell.2021.653101.

Faissner

, Plemel

, Gold

, . Progressive multiple sclerosis: From pathophysiology to therapeutic strategies. Nat Rev Drug Discov 2019; 18(12): 905–922. DOI: 10.1038/s41573-019-0035-2.

Gruchot

, Weyers

, Göttle

, . The molecular basis for remyelination failure in multiple sclerosis. Cells 2019; 8(8): 825. DOI: 10.3390/cells8080825.

Lee

, Taghian

and Petratos

Axonal degeneration in multiple sclerosis: Can we predict and prevent permanent disability?

Acta Neuropathol Commun 2014; 2(1): 97. DOI: 10.1186/s40478-014-0097-7.

Filippi

, Bozzali

, Rovaris

, . Evidence for widespread axonal damage at the earliest clinical stage of multiple sclerosis. Brain 2003; 126(2): 433–437. DOI: 10.1093/brain/awg038.

Gonen

, Moriarty

, Li

BSY

, . Relapsing-remitting multiple sclerosis and whole-brain N-acetylaspartate measurement: Evidence for different clinical cohorts—initial observations. Radiology 2002; 225(1): 261–268. DOI: 10.1148/radiol.2243011260.

Ferrazzano

, Crisafulli

, Baione

, . Early diagnosis of secondary progressive multiple sclerosis: Focus on fluid and neurophysiological biomarkers. J Neurol 2021; 268(10): 3626–3645. DOI: 10.1007/s00415-020-09964-4.

Kapoor

, Smith

, Allegretta

, . Serum neurofilament light as a biomarker in progressive multiple sclerosis. Neurology 2020; 95(10): 436–444. DOI: 10.1212/WNL.0000000000010346.

10.

Magliozzi

and Cross

AH.

Can CSF biomarkers predict future MS disease activity and severity?

Mult Scler J 2020; 26(5): 582–590. DOI: 10.1177/1352458519871818.

11.

Momtazmanesh

, Shobeiri

, Saghazadeh

, . Neuronal and glial CSF biomarkers in multiple sclerosis: A systematic review and meta-analysis. Rev Neurosci 2021; 32(6): 573–595. DOI: 10.1515/revneuro-2020-0145.

12.

Williams

, Zetterberg

and Chataway

Neurofilaments in progressive multiple sclerosis: A systematic review. J Neurol 2020; 268: 3212–3222. DOI: 10.1007/s00415-020-09917-x.

13.

Yuan

, Rao

, Veeranna , . Neurofilaments and neurofilament proteins in health and disease. Cold Spring Harb Perspect Biol 2017; 9(4): a018309. DOI: 10.1101/cshperspect.a018309.

14.

Wang

, Wu

, Zhan

, . Neurofilament proteins in axonal regeneration and neurodegenerative diseases. Neural Regen Res 2012; 7(8): 620–626. DOI: 10.3969/j.issn.1673-5374.2012.08.010.

15.

Yuan

, Rao

, Veeranna , . Neurofilaments at a glance. J Cell Sci 2012; 125(14): 3257–3263. DOI: 10.1242/jcs.104729.

16.

Lobsiger

and Cleveland

DW.

Neurofilaments: Organization and function in neurons. In: Encyclopedia of neuroscience . 2009, vol. 6, pp.433–436. DOI: 10.1016/B978-008045046-9.00728-2.

17.

Križ

, Zhu

, Julien

, . Electrophysiological properties of axons in mice lacking neurofilament subunit genes: Disparity between conduction velocity and axon diameter in absence of NF-H. Brain Res 2000; 885(1): 32–44. DOI: 10.1016/S0006-8993(00)02899-7.

18.

Didonna

and Opal

The role of neurofilament aggregation in neurodegeneration: Lessons from rare inherited neurological disorders. Mol Neurodegener 2019; 14(1): 19. DOI: 10.1186/s13024-019-0318-4.

19.

Eyer

and Peterson

Neurofilament-deficient axons and perikaryal aggregates in viable transgenic mice expressing a neurofilament-β-galactosidase fusion protein. Neuron 1994; 12(2): 389–405. DOI: 10.1016/0896-6273(94)90280-1.

20.

Muñoz-Lasso

, Romá-Mateo

, Pallardó

, . Much more than a scaffold: Cytoskeletal proteins in neurological disorders. Cells 2020; 9(2): 358. DOI: 10.3390/cells9020358.

21.

Ammoscato

and Sharrack

CSF neurofilament heavy chain release (NfH) is important in secondary progressive MS (SPMS)—results from the MS-SMART trial. Mult Scler J 2019; 25: 502–503.

22.

Kuhle

, Nourbakhsh

, Grant

, . Serum neurofilament is associated with progression of brain atrophy and disability in early MS. Neurology 2017; 88(9): 826–831.

23.

Kuhle

, Kropshofer

, Haring

, . Blood neurofilament light chain as a biomarker of MS disease activity and treatment response. Neurology 2019; 92(10): e1007–e1015.

24.

Barro

, Benkert

, Disanto

, . Serum neurofilament as a predictor of disease worsening and brain and spinal cord atrophy in multiple sclerosis. Brain 2018; 141(8): 2382–2391.

25.

Disanto

, Adiutori

, Dobson

, . Serum neurofilament light chain levels are increased in patients with clinically isolated syndrome. J Neurol Neurosurg Psychiatry 2015; 87(2): 126–129. DOI: 10.1136/jnnp-2014-309690.

26.

Manouchehrinia

, Stridh

, Khademi

, . Plasma neurofilament light levels are associated with the risk of disabilities in multiple sclerosis. Neurology 2020; 94(23): e2457–e2467. DOI: 10.1212/WNL.0000000000009571.

27.

Ferraro

, Guicciardi

, Biasi

, . Plasma neurofilaments correlate with disability in progressive multiple sclerosis patients. Acta Neurol Scand 2019; 141(1): 16–21. DOI: 10.1111/ane.13152.

28.

Kuhle

, Kropshofer

, Haring

, . Neurofilament light levels in the blood of patients with secondary progressive MS are higher than in primary progressive MS and may predict brain atrophy in both MS subtypes. Mult Scler J 2018; 24: 111. DOI: 10.1177/1352458518798579.

29.

Sternberger

and Sternberger

NH.

Monoclonal antibodies distinguish phosphorylated and nonphosphorylated forms of neurofilaments in situ. Proc Natl Acad Sci USA 1983; 80(19): 6126–6130.

30.

Petzold

Neurofilament phosphoforms: Surrogate markers for axonal injury, degeneration and loss. J Neurol Sci 2005; 233(1–2): 183–198. DOI: 10.1016/j.jns.2005.03.015.

31.

Petzold

, Eikelenboom

, Keir

, . Axonal damage accumulates in the progressive phase of multiple sclerosis: Three year follow up study. J Neurol Neurosurg Psychiatry 2005; 76(2): 206–211.

32.

Ali-Massoud

, Eldeen

EMS

, Khattab

, . Neurofilament as a biomarker of axonal loss in multiple sclerosis. Sci J Al-Azhar Med Fac Girls 2017; 1(2): 63–68.

33.

Delcoigne

, Manouchehrinia

, Barro

, . Blood neurofilament light levels segregate treatment effects in multiple sclerosis. Neurology 2020; 94(11): e1201–e1212. DOI: 10.1212/WNL.0000000000009097.

34.

Shehab

, Solima

, Abdel-Hafeez

, . Serum phosphorylated neurofilament heavy chain level in relapsing-remitting multiple sclerosis in correlation to disease activity and disability. Egypt J Immunol 2019; 26(1): 1–13.

35.

Gresle

, Liu

, Dagley

, . Serum phosphorylated neurofilament-heavy chain levels in multiple sclerosis patients. J Neurol Neurosurg Psychiatry 2014; 85(11): 1209–1213.

36.

Dhiman

, Gupta

, Villemagne

, . Cerebrospinal fluid neurofilament light concentration predicts brain atrophy and cognition in Alzheimer’s disease. Alzheimers Dement (Amst) 2020; 12(1): e12005. DOI: 10.1002/dad2.12005.

37.

Disanto

, Barro

, Benkert

, . Serum neurofilament light: A biomarker of neuronal damage in multiple sclerosis. Ann Neurol 2017; 81(6): 857–870.

38.

Domingues

, Fernandes

GBP

, Leite

FBV

, . Neurofilament light chain in the assessment of patients with multiple sclerosis. Arq Neuropsiquiatr 2019; 77(6): 436–441. DOI: 10.1590/0004-282X20190060.

39.

Petzold

The prognostic value of CSF neurofilaments in multiple sclerosis at 15 year follow up. J Neurol Neurosurg Psychiatry 2015; 86(12): 1388–1390. DOI: 10.1136/jnnp-2014-309827.

40.

Hakansson

, Tisell

, Cassel

, . Neurofilament levels, disease activity and brain volume during follow up in multiple sclerosis. J Neuroinflammation 2018; 15(1): 209. DOI: 10.1186/s12974-018-1249-7.

41.

Bhan

, Jacobsen

, Myhr

, . Neurofilaments and 10-year follow-up in multiple sclerosis. Mult Scler J 2018; 24(10): 1301–1307. DOI: 10.1177/1352458518782005.

42.

Pawlitzki

, Schreiber

, Bittner

, . CSF neurofilament light chain levels in primary progressive MS: Signs of axonal neurodegeneration. Front Neurol 2018; 9: 1037. DOI: 10.3389/fneur.2018.01037.

43.

Khademi

, Dring

, Gilthorpe

, . Intense inflammation and nerve damage in early multiple sclerosis subsides at older age: A reflection by cerebrospinal fluid biomarkers. PLoS One 2013; 8(5): e63172. DOI: 10.1371/journal.pone.0063172.

44.

Mañé-Martínez

, Olsson

, Bau

, . Glial and neuronal markers in cerebrospinal fluid in different types of multiple sclerosis. J Neuroimmunol 2016; 299: 112–117. DOI: 10.1016/j.jneuroim.2016.08.004.

45.

Novakova

, Axelsson

, Malmeström

, . Searching for neurodegeneration in multiple sclerosis at clinical onset: Diagnostic value of biomarkers. PLoS One 2018; 13(4): e0194828. DOI: 10.1371/journal.pone.0194828.

46.

Martin

, McGlasson

, Hunt

, . Cerebrospinal fluid neurofilament light chain in multiple sclerosis and its subtypes: A meta-analysis of case-control studies. J Neurol Neurosurg Psychiatry 2019; 90(9): 1059–1067.

47.

Zetterberg

and Svenningsson

Serum neurofilament light and prediction of multiple sclerosis in clinically isolated syndrome. Neurology 2019; 92(7): 313–314. DOI: 10.1212/WNL.0000000000006906.

48.

Baecher-Allan

, Kaskow

and Weiner

HL.

Multiple sclerosis: Mechanisms and immunotherapy. Neuron 2018; 97(4): 742–768.

49.

Lepinoux-Chambaud

and Eyer

Review on intermediate filaments of the nervous system and their pathological alterations. Histochem Cell Biol 2013; 140(1): 13–22. DOI: 10.1007/s00418-013-1101-1.

50.

Novakova

, Axelsson

, Khademi

, . Cerebrospinal fluid biomarkers of inflammation and degeneration as measures of fingolimod efficacy in multiple sclerosis. Mult Scler J 2017; 23(1): 62–71. DOI: 10.1177/1352458516639384.

51.

Novakova

, Zetterberg

, Sundström

, . Monitoring disease activity in multiple sclerosis using serum neurofilament light protein. Neurology 2017; 89(22): 2230–2237.

52.

Brureau

, Bregeon

, Pech

, . NF-L in cerebrospinal fluid and serum is a biomarker of neuronal damage in an inducible mouse model of neurodegeneration. Neurobiol Dis 2017; 104: 73–84. DOI: 10.1016/j.nbd.2017.04.007.

53.

Siller

, Kuhle

, Muthuraman

, . Serum neurofilament light chain is a biomarker of acute and chronic neuronal damage in early multiple sclerosis. Mult Scler J 2019; 25(5): 678–686. DOI: 10.1177/1352458518765666.

54.

Wong

YYM

, Bruijstens

, Barro

, . Serum neurofilament light chain in pediatric MS and other acquired demyelinating syndromes. Neurology 2019; 93(10): e968–e974.

55.

Bittner

, Steffan

, Uphaus

, . Clinical implications of serum neurofilament in newly diagnosed MS patients: A longitudinal multicentre cohort study. EBioMedicine 2020; 56: 102807. DOI: 10.1016/j.ebiom.2020.102807.

56.

Norgren

, Sundstrom

, Svenningsson

, . Neurofilament and glial fibrillary acidic protein in multiple sclerosis. Neurology 2004; 63(9): 1586–1590. DOI: 10.1212/01.WNL.0000142988.49341.d1.

57.

Cai

and Huang

Neurofilament light chain as a biological marker for multiple sclerosis: A meta-analysis study. Neuropsychiatr Dis Treat 2018; 14: 2241–2254.

58.

Varhaug

, Torkildsen

, Myhr

, . Neurofilament light chain as a biomarker in multiple sclerosis. Front Neurol 2019; 10: 338. DOI: 10.3389/fneur.2019.00338.

59.

Mellergård

, Tisell

, Blystad

, . Cerebrospinal fluid levels of neurofilament and tau correlate with brain atrophy in natalizumab-treated MS patients. Eur J Neurol 2017; 24(1): 112–121. DOI: 10.1111/ene.13162.

60.

Gaetani

, Eusebi

, Mancini

, . Cerebrospinal fluid neurofilament light chain predicts disease activity after the first demyelinating event suggestive of multiple sclerosis. Mult Scler Relat Disord 2019; 35: 228–232. DOI: 10.1016/j.msard.2019.07.025.

61.

Kuhle

, Plattner

, Bestwick

, . A comparative study of CSF neurofilament light and heavy chain protein in MS. Mult Scler J 2013; 19(12): 1597–1603. DOI: 10.1177/1352458513482374.

62.

Teunissen

, Iacobaeus

, Khademi

, . Combination of CSF N-acetylaspartate and neurofilaments in multiple sclerosis. Neurology 2009; 72(15): 1322–1329.

63.

Petzold

and Plant

GT.

The diagnostic and prognostic value of neurofilament heavy chain levels in immune mediated optic neuropathies. Mult Scler Int 2012; 2012: 217802. DOI: 10.1155/2012/217802.

64.

Canto

, Barro

, Zhao

, . Association between serum neurofilament light chain levels and long-term disease course among patients with multiple sclerosis followed up for 12 years. JAMA Neurol 2019; 76(11): 1359–1366. DOI: 10.1001/jamaneurol.2019.2137.

65.

Jakimovski

, Zivadinov

, Ramanathan

, . Serum neurofilament light chain level associations with clinical and cognitive performance in multiple sclerosis: A longitudinal retrospective 5-year study. Mult Scler J 2019; 26(9): 1352458519881428. DOI: 10.1177/1352458519881428.

66.

Jakimovski

, Kuhle

, Ramanathan

, . Serum neurofilament light chain level associations with gray matter pathology: A 5-year longitudinal study. Ann Clin Transl Neurol 2019; 6(9): 1757–1770. DOI: 10.1002/acn3.50872.

67.

Modvig

, Degn

, Sander

, . Cerebrospinal fluid neurofilament light chain levels predict visual outcome after optic neuritis. Mult Scler J 2016; 22(5): 590–598. DOI: 10.1177/1352458515599074.

68.

Sorensen

and Sellebjerg

Neurofilament in CSF—a biomarker of disease activity and long-term prognosis in multiple sclerosis. Mult Scler J 2016; 22(9): 1112–1113. DOI: 10.1177/1352458516658560.

69.

Matute-Blanch

, Villar

, Alvarez-Cermeno

, . Neurofilament light chain and oligoclonal bands are prognostic biomarkers in radiologically isolated syndrome. Brain 2018; 141(4): 1085–1093. DOI: 10.1093/brain/awy021.

70.

Kuhle

, Leppert

, Petzold

, . Neurofilament heavy chain in CSF correlates with relapses and disability in multiple sclerosis. Neurology 2011; 76(14): 1206–1213.

71.

Norgren

, Karlsson

, Rosengren

, . Monoclonal antibodies selective for low molecular weight neurofilaments. Hybrid Hybridomics 2002; 21(1): 53–59.

72.

Khalil

, Enzinger

, Langkammer

, . CSF neurofilament and N-acetylaspartate related brain changes in clinically isolated syndrome. Mult Scler J 2013; 19(4): 436–442.

73.

Dutta

and Trapp

BD.

Mechanisms of neuronal dysfunction and degeneration in multiple sclerosis. Prog Neurobiol 2011; 93(1): 1–12.

74.

Iwanowski

and Losy

Neurofilament light, anti-ganglioside M3 antibody and interleukin 18 in relapsing-remitting and primary progressive multiple sclerosis. Mult Scler J 2016; 22: 879. DOI: 10.1177/1352458516664293.

75.

Preziosa

, Rocca

and Filippi

Current state of the art of the application of serum neurofilaments in multiple sclerosis diagnosis and monitoring. Expert Rev Neurother 2020; 20(8): 747–769. DOI: 10.1080/14737175.2020.1760846.

76.

Frischer

, Bramow

, Dal-Bianco

, . The relation between inflammation and neurodegeneration in multiple sclerosis brains. Brain 2009; 132(5): 1175–1189.

77.

Lassmann

What drives disease in multiple sclerosis: Inflammation or neurodegeneration?

Clin Exp Neuroimmunol 2010; 1: 2–11.

78.

Derfuss

, Parikh

, Velhin

, . Contactin-2/TAG-1-directed autoimmunity is identified in multiple sclerosis patients and mediates gray matter pathology in animals. Proc Natl Acad Sci USA 2009; 106(20): 8302–8307.

79.

Neumann

, Medana

, Bauer

, . Cytotoxic T lymphocytes in autoimmune and degenerative CNS diseases. Trends Neurosci 2002; 25(6): 313–319.

80.

Lazzarino

, Amorini

, Eikelenboom

, . Cerebrospinal fluid ATP metabolites in multiple sclerosis. Mult Scler J 2010; 16(5): 549–554.

81.

Mahad

, Lassmann

and Turnbull

Mitochondria and disease progression in multiple sclerosis. Neuropathol Appl Neurobiol 2008; 34(6): 577–589.

82.

Gonsette

RE.

Neurodegeneration in multiple sclerosis: The role of oxidative stress and excitotoxicity. J Neurol Sci 2008; 274(1–2): 48–53.

83.

Werner

, Pitt

and Raine

CS.

Glutamate excitotoxicity—a mechanism for axonal damage and oligodendrocyte death in multiple sclerosis?

J Neural Transm 2000; Suppl 60: 375–385.

84.

Teunissen

, Dijkstra

and Polman

CH.

Biological markers in CSF and blood for axonal degeneration in multiple sclerosis. Lancet Neurol 2005; 4(1): 32–41.

85.

Waxman

, Craner

and Black

JA.

Na+ channel expression along axons in multiple sclerosis and its models. Trends Pharmacol Sci 2004; 25(11): 584–591.

86.

Rasband

, Tayler

, Kaga

, . CNP is required for maintenance of axon–glia interactions at nodes of Ranvier in the CNS. Glia 2005; 50(1): 86–90.

87.

Howell

, Rundle

, Garg

, . Activated microglia mediate axoglial disruption that contributes to axonal injury in multiple sclerosis. J Neuropathol Exp Neurol 2010; 69(10): 1017–1033.

88.

Wang

, Jiang

, Wang

, . Neurofilament protein as a potential biomarker of axonal degeneration in experimental autoimmune encephalomyelitis. Neurol India 2020; 68(2): 364–367. DOI: 10.4103/0028-3886-280651.

89.

Bsteh

, Berek

, Hegen

, . Serum neurofilament levels correlate with retinal nerve fibre layer thinning in multiple sclerosis. Mult Scler J 2019; 26(3): 1682–1690. DOI: 10.1177/1352458519882279.

90.

Lycke

, Karlsson

, Andersen

, . Neurofilament protein in cerebrospinal fluid: A potential marker of activity in multiple sclerosis. J Neurol Neurosurg Psychiatry 1998; 64(3): 402–404.

91.

Schrimer

, Antel

, Brück

, . Axonal loss and neurofilament phosphorylation changes accompany lesion development and clinical progression in multiple sclerosis. Brain Pathol 2011; 21(4): 428–440. DOI: 10.1111/j.1750-3639.2010.00466.x.

92.

Semra

, Seidi

and Sharief

MK.

Heightened intrathecal release of axonal cytoskeletal proteins in multiple sclerosis is associated with progressive disease and clinical disability. J Neuroimmunol 2002; 122(1–2): 132–139.

93.

Lassmann

Pathogenic mechanisms associated with different clinical courses of multiple sclerosis. Front Immunol 2018; 9: 3116.

94.

Pegoretti

, Swanson

, Bethea

, . Inflammation and oxidative stress in multiple sclerosis: Consequences for therapy development. Oxid Med Cell Longev 2020; 2020: 7191080. DOI: 10.1155/2020/7191080.

95.

Fitzner

and Simons

Chronic progressive multiple sclerosis: Pathogenesis of neurodegeneration and therapeutic strategies. Curr Neuropharmacol 2010; 8(3): 305–315. DOI: 10.2174/157015910792246218.

96.

Petzold

, Steenwijk

, Eikelenboom

, . Elevated CSF neurofilament proteins predict brain atrophy: A 15 year follow up study. Mult Scler J 2016; 22(9): 1154–1162. DOI: 10.1177/1352458516645206.

97.

Chitnis

, Gonzalez

, Healy

, . Neurofilament light chain serum levels correlate with 10-year MRI outcomes in multiple sclerosis. Ann Clin Transl Neurol 2018; 5(12): 1478–1491. DOI: 10.1002/acn3.638.

98.

Kuhle

, Plavina

, Barro

, . Neurofilament light levels are associated with long-term outcomes in multiple sclerosis. Mult Scler J 2020; 26(3): 1691–1699. DOI: 10.1177/1352458519885613.

99.

Byrne

, Rodrigues

, Johnson

, . Evaluation of mutant huntingtin and neurofilament proteins as potential markers in Huntington disease. Sci Transl Med 2018; 10(458): eaat7108. DOI: 10.1126/scitranslmed.aat7108.

100.

Trentini

, Comabella

, Tintoré

, . N-acetylaspartate and neurofilaments as biomarkers of axonal damage in progressive multiple sclerosis. J Neurol 2014; 261: 2338–2343. DOI: 10.1007/s00415-014-7507-4.

101.

Salzer

, Svenningsson

and Sundström

Neurofilament light as a prognostic marker in multiple sclerosis. Mult Scler J 2010; 16: 287–292. DOI: 10.1177/1352458509359725.

102.

Amit

BO.

Blood neurofilament light levels are lowered to a healthy donor range in patients with RMS and PPMS following ocrelizumab treatment. Mult Scler J 2019; 25: 52–52.

103.

Högel

, Rissanen

, Barro

, . Serum GFAP associates with neurofilament light and disease severity in late multiple sclerosis. Mult Scler J 2018; 24: 658–659. DOI: 10.1177/1352458518798591.

104.

Kapoor

, Sellebjerg

, Hartung

, . Natalizumab reduced serum levels of neurofilament light chain in secondary progressive multiple sclerosis patients. Mult Scler J 2018; 24: 988. DOI: 10.1177/1352458518799980.

105.

Kalatha

, Arnaoutoglou

, Koukoulidis

, . Does cognitive dysfunction correlate with neurofilament light polypeptide levels in CSF of patients with multiple sclerosis. J Int Med Res 2019; 47: 2187–2198. DOI: 10.1177/0300060519840550.

106.

Kaymakamzade

, Kurne

, Tumani

, . Evaluation of neurofilament heavy chain levels in progressive multiple sclerosis patients: Preliminary results. J Neurol 2014; 261: S453. DOI: 10.1007/s00415-014-7337-4.

107.

Gnanapavan

, Grant

, Morant

, . Biomarker report from the phase II lamotrigine trial in secondary progressive MS—neurofilament as a surrogate of disease progression. PLoS One 2013; 8(8): e70019. DOI: 10.1371/journal.pone.0070019.

108.

Eikelenboom

, Uitdehaag

BMJ

and and

Petzold A. Blood

CSF biomarker dynamics in multiple sclerosis: Implications for data interpretation. Mult Scler Int 2011; 2011: 823176. DOI: 10.1155/2011/823176.

109.

Sejbaek

, Nielsen

, Penner

, . Dimethyl fumarate decreases neurofilament light chain in CSF and blood of treatment-naive relapsing patients. J Neurol Neurosurg Psychiatry 2019; 90(12): 1324–1330. DOI: 10.1136/jnnp-2019-321321.

110.

Kuhle

, Disanto

, Lorscheider

, . Fingolimod and CSF neurofilament light chain levels in relapsing-remitting multiple sclerosis. Neurology 2015; 84(16): 1639–1643. DOI: 10.1212/WNL.0000000000001491.

111.

Piehl

, Kockum

, Khademi

, . Plasma neurofilament light chain levels in patients switching from injectable therapies to fingolimod. Mult Scler J 2018; 24(8): 1046–1054. DOI: 10.1177/1352458517715132.

112.

Cohen

, Kappos

, Tenenbaum

, . Long-term effect of fingolimod in reducing blood neurofilament light levels in patients with relapsing-remitting multiple sclerosis. Neurology 2019; 92(15_supplement) P3.2-032. DOI: 10.1212/WNL.92.15_supplement.P3.2-032

113.

Gunnarsson

, Malmeström

, Axelsson

, . Axonal damage in relapsing multiple sclerosis is markedly reduced by natalizumab. Ann Neurol 2011; 69(1): 83–89. DOI: 10.1002/ana.22247.

114.

Romme Christensen

, Ratzer

, Bornsen

, . Natalizumab in progressive MS: Results of an open-label phase 2A trial. Neurology 2014; 82(17): 1499–1507. DOI: 10.1212/WNL.0000000000000361.

115.

Dalla Costa

, Martinelli

, Moiola

, . Serum neurofilaments increase at progressive multifocal leukoencephalopathy onset in natalizumab-treated multiple sclerosis patients. Ann Neurol 2019; 85(4): 606–610. DOI: 10.1002/ana.25437.

116.

Akgun

, Kretschmann

, Haase

, . Profiling individual clinical responses by high-frequency serum neurofilament assessment in MS. Neurol Neuroimmunol Neuroinflamm 2019; 6(3): e555. DOI: 10.1212/NXI.0000000000000555.