• Users Online: 539
  • Print this page
  • Email this page

 Table of Contents  
Year : 2021  |  Volume : 38  |  Issue : 4  |  Page : 256-261

Cerebrospinal fluid level of phosphorylated neurofilament heavy chain is higher in converting clinically isolated syndrome and correlates with CAMP response element-binding protein concentration

1 Department of Neurology, Istanbul Haydarpasa Numune Training and Research Hospital, Istanbul, Turkey
2 Department of Neuroscience, Aziz Sancar Institute of Experimental Medicine, Istanbul University, Istanbul, Turkey
3 Department of Physiology, Istanbul Medical Faculty, Istanbul University, Istanbul, Turkey

Date of Submission04-Aug-2021
Date of Decision10-Aug-2021
Date of Acceptance04-Sep-2021
Date of Web Publication29-Dec-2021

Correspondence Address:
Mehmet Gencer
Department of Neurology, Istanbul Haydarpasa Numune Training and Research Hospital, Istanbul
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/nsn.nsn_144_21

Rights and Permissions

Introduction: Prevision of conversion from clinically isolated syndrome (CIS) to multiple sclerosis (MS) is required to avoid unnecessary use of immunomodulating agents and to recognize patients with high disease activity. Our aim was to evaluate the value of phosphorylated neurofilament heavy chain (pNFH, a marker for neuroaxonal degeneration) and Cyclic adenosine monophosphate response element-binding protein (cAMP response element-binding protein [CREB], a marker for neuroregeneration) levels in the prediction of conversion from CIS to MS. Methods: Twenty-three consecutively recruited treatment-naïve CIS patients were followed for 36 months. pNFH and CREB levels were measured in the first episode cerebrospinal fluid (CSF) and the serum of 12 converting (CIS-MS) and 11 nonconverting CIS patients (CIS-CIS) by enzyme-linked immunosorbent assay. Results: Baseline CSF but not serum samples of CIS-CIS patients displayed significantly lower pNFH levels compared to patients with CIS-MS. The analysis of receiver operating characteristic curve presented a high specificity for the prediction of MS conversion for the CSF pNFH cut-off level of 730.9 pg/ml. CSF pNFH levels significantly correlated with serum and CSF CREB levels. Higher baseline CSF pNFH and CREB levels were associated with more rapid progression to MS or increased disability scores. Conclusion: CSF pNFH measurement may potentially determine MS patients with unfavorable clinical progression after the first attack. pNFH and CREB appear to be increased in parallel in CSF of CIS patients with higher disease activity. These results suggest that neurofilaments are not only indicators of axonal degeneration but also partly a marker of neuronal differentiation and new axon regeneration mediated by CREB signaling pathway.

Keywords: Biomarker, cAMP response element-binding protein, clinically isolated syndrome, multiple sclerosis, neurofilament

How to cite this article:
Gencer M, Koral G, Sanli E, Cirak S, Akbayir E, Yuceer H, Kizilay T, Yildiz RE, Yentur SP, Yilmaz V, Tuzun E, Turkoglu R. Cerebrospinal fluid level of phosphorylated neurofilament heavy chain is higher in converting clinically isolated syndrome and correlates with CAMP response element-binding protein concentration. Neurol Sci Neurophysiol 2021;38:256-61

How to cite this URL:
Gencer M, Koral G, Sanli E, Cirak S, Akbayir E, Yuceer H, Kizilay T, Yildiz RE, Yentur SP, Yilmaz V, Tuzun E, Turkoglu R. Cerebrospinal fluid level of phosphorylated neurofilament heavy chain is higher in converting clinically isolated syndrome and correlates with CAMP response element-binding protein concentration. Neurol Sci Neurophysiol [serial online] 2021 [cited 2022 Aug 19];38:256-61. Available from: http://www.nsnjournal.org/text.asp?2021/38/4/256/334050

  Introduction Top

Multiple sclerosis (MS) is a persistent demyelinating disease affecting the central nervous system that commonly begins with demyelinating acute inflammation, defined as clinically isolated syndrome (CIS). In the aftermath of CIS, the occurrence of new clinical episodes and/or new demyelinating lesions lead to the diagnosis of MS.[1] Only some CIS patients converts to MS, and thus, it is crucial to find biomarkers, which are capable of predicting converting CIS patients to avoid unnecessary use of immunomodulating medications.

Neurofilament light chain (NFL) and phosphorylated neurofilament heavy chain (pNFH) were proposed as potential biomarkers in both degenerative neurological disorders and MS.[2],[3] NFL has particularly been studied in greater detail in MS and has been found to be correlated with clinical, radiological, and cognitive progression of the disease.[4] Elevated cerebrospinal fluid (CSF) NFL levels were reported in CIS patients, who are more prone to convert to MS in due course.[5]

pNFH has been relatively understudied in MS. There are only a few cross-sectional studies indicating elevated serum and CSF levels of pNFH in MS. These studies have also shown a correlation between disability scores of MS and CSF/peripheral blood levels of pNFH.[6],[7],[8],[9],[10] Serum pNFH levels were found to be correlated with the degree of visual loss in optic neuritis, as well.[11] Moreover, postmortem studies have shown that pNFH concentration of cortical gray matter of MS patients is reduced in correlation with the severity of demyelination.[12]

Although elevation of neurofilament levels in serum and CSF during the course of MS is generally attributed to neuroaxonal degeneration, hypothetically compensatory efforts of neuronal regeneration, neurite outgrowth and new axon sprouting might also be leading to increased neurofilament production.[2] In this context, cAMP response element-binding protein (CREB) may act as a potential biomarker for MS progression due to its well-known involvement in neuronal regeneration.[13],[14] As a matter of fact, white matter neurons of MS patients show reduced activity of CREB signaling-related genes, which may potentially be associated with impaired neuroaxonal integrity and synapse loss.[13] Moreover, in toxicity-induced animal models of MS, reduction of cerebral expression of CREB has been related with memory impairment, and treatment-induced improvement of memory functions has been associated with upregulation of CREB in the prefrontal cortex.[15],[16]

In this study, our aim was to evaluate the predictive potential of pNFH and CREB in the determination of converting and nonconverting CIS patients and also to provide clues regarding the interplay between neurofilament levels and neuronal regeneration. For this purpose, we measured CREB and pNFH levels in CSF and serum samples taken from CIS patients during the first attack and sought for correlation between these markers and 3-year disease activity.

  Methods Top


Twenty-three successive, newly diagnosed patients who met the criteria of CIS from patients admitted to the outpatient clinic were enrolled for the study.[17] CIS patients who were under treatment or had a coexisting neurological/systemic disorder were omitted. Thus, all patients were treatment-naïve for MS medications. All of the CIS patients were subjected to cranial magnetic resonance imaging with the identical scanner (1.5 T). Tests for anti-neutrophil cytoplasm antibodies, anti-nuclear antibodies, thyroid autoantibodies, levels of Vitamin B12 and folate, as well as serological tests for HIV, syphilis, and Lyme borreliosis were performed to exclude other diagnoses. CSF and serum oligoclonal IgG bands were examined in all cases at the first clinical attack. CIS patients were followed for 36 months by the same neurologist with scheduled examinations every 6 months or during relapse periods. Patients were divided in two groups as CIS patients that converted to clinically definite MS within the 36-month follow-up period (CIS-MS) as per McDonald criteria (n = 12) and as CIS patients that did not convert to clinically definite MS (CIS-CIS) (n = 11) within the same time frame.[17] Interferon-beta was used for all CIS-MS patients after conversion to MS. Approval of the study was obtained from the local ethics committee. Written informed consent was provided from all participants.

Samples and enzyme-linked immunosorbent assay

Serum and CSF samples were gathered from patients with CIS after the first demyelinating attack. The mean period between the episode and sampling was 2.1 ± 1.0 months (range 1.2–3.8 months), and all patients were in remission period. Samples from patients were gathered in the mornings (8:00–10:00 AM) and kept at-80°C. None of the donors were under immunomodulatory or steroid drug therapy at the time of sampling. The levels of pNFH (Elabscience, Houston, TX, USA, sensitivity: 46.88 pg/ml, detection range: 78.13–5000 pg/ml,) and CREB (Elabscience, Houston, TX, USA, detection range: 0.31–20 ng/ml, sensitivity: 0.19 ng/ml) were quantified by enzyme-linked immunosorbent assay (ELISA) following the manufacturer's directions. Reading of optical density was performed at 450 nm. The resulting concentrations were calculated based on graphs of the standart curve.

Statistical analysis

Mann–Whitney U-test and Student's t-test were performed for comparison of the groups. Chi-square test was used to compare categorical factors, and correlation analyzes were conducted by Pearson's correlation test. Receiver operating characteristic (ROC) curve analysis was done to calculate the specificity and sensitivity, as cut-off level was selected using the Youden index. Statistical significance was specified P < 0.05.

  Results Top

Clinical features, phosphorylated neurofilament heavy, and cAMP response element-binding protein levels

Of the age- and gender-matched patient groups, CIS-MS patients exhibited a remarkably higher number of episodes, Expanded Disability Status Scale (EDSS) scores, and CSF OCD ratios (Pattern 2) than patients with CIS-CIS [Table 1]. The mean time (±standard deviation) to conversion to MS from the CIS-MS group was 9.3 ± 10.0 months with a range of 2–34 months. CSF levels of pNFH after the first demyelinating attack were significantly lower in CIS-CIS patients compared to CIS-MS patients (P = 0.009), whereas serum pNFH levels were identical between the two groups (P = 0.439). Likewise, CSF (P = 0.714) and serum (P = 0.278) CREB levels were comparable between patients with CIS-CIS and CIS-MS [Figure 1].
Figure 1: Levels of phosphorylated neurofilament heavy chain and cAMP response element-binding protein in cerebrospinal fluid and serum samples of clinically isolated syndrome - clinically isolated syndrome (clinically isolated syndrome patients who did not progressed to clinically certain multiple sclerosis) and clinically isolated syndrome-multiple sclerosis (clinically isolated syndrome patients who progressed to clinically definite multiple sclerosis). Horizontal lines indicate mean values. **, P < 0.01 by Student's t-test

Click here to view
Table 1: Demographic and clinical characteristics of the patients

Click here to view

Predictive value of baseline levels of phosphorylated neurofilament heavy for conversion from clinically isolated syndrome to multiple sclerosis

The diagnostic specificity, sensitivity, and cut-off rates for CSF levels of pNFH were calculated using ROC curve analysis and Youden index, while the binary outcome was described as CIS-MS versus CIS-CIS [Figure 2]. The best area under curve (0.788), sensitivity (58.3%), and specificity (100%) values were acquired by CSF pNFH level measurements at a cut-off value of 730.9 pg/ml (Youden index 0.583 and likelihood ratio 6.42). More precisely, all 11 patients with CIS-CIS demonstrated lower baseline CSF pNFH levels than 730.9 pg/ml and thus, this cut-off value accurately predicted all patients with CIS who did not convert to MS in a 3-year follow-up duration. CSF pNFH level was higher than 730.9 pg/ml only in 7 of 12 CIS-MS patients, and thus MS conversion had been accurately predicted only in 58.3% of this CIS subgroup with this cut-off level.
Figure 2: Receiver operation characteristics curves for phosphorylated neurofilament heavy chain concentrations in first attack cerebrospinal fluid samples of all included clinically isolated syndrome patients. The binary outcome is described as clinically isolated syndrome-multiple sclerosis (clinically isolated syndrome patients who progressed to clinically certain multiple sclerosis) versus clinically isolated syndrome-clinically isolated syndrome (clinically isolated syndrome patients who did not progressed to clinically definite multiple sclerosis) in a 3-year follow-up duration

Click here to view

Correlation analyses

Serum and CSF levels of pNFH (P = 0.043; R = 0.366) and CREB (P = 0.038; R = 0.377) showed significant correlation. CSF pNFH levels were significantly correlated with both serum (P < 0.001; R = 0.668) and CSF (P = 0.004; R = 0.536) levels of CREB. Serum pNFH levels significantly correlated with serum CREB (P = 0.045; R = 0.361) but not CSF CREB (P = 0.106; R = 0.270) levels. While CSF pNFH levels showed significant negative correlation with months between first episode and conversion to MS (P = 0.049; R = ‒0.500) and positive correlation with final EDSS values (P = 0.027; R = 0.406), CSF CREB levels significantly correlated with months to MS conversion (P = 0.022; R = ‒0.591) but not with final EDSS scores (P = 0.495; R = ‒0.003) [Figure 3]. Serum pNFH and CREB levels were not correlated with any clinical parameters. Likewise, age and attack number parameters did not correlate with serum or CSF pNFH and CREB levels.
Figure 3: Correlation analysis between cerebrospinal fluid phosphorylated neurofilament heavy chain and cAMP response element-binding protein concentrations versus serum CREB levels, months for conversion to multiple sclerosis and final EDSS parameters. P value and R, correlation coefficient in Pearson correlation test

Click here to view

  Discussion Top

In this study, we found elevated pNFH levels in baseline CSF of converting CIS patients as compared to nonconverting patients. Moreover, CSF pNFH levels significantly correlated with serum and CSF levels of CREB, duration of conversion to MS, and final EDSS score in converting CIS patients. While the prognostic value of NFL in MS has been inspected with several longitudinal studies, to our knowledge, the biomarker value of pNFH in MS has not been explored with longitudinal studies before.[2],[3],[4],[5] We have herein shown for the first time that baseline CSF pNFH levels may longitudinally predict 3-year disease activity in the aftermath of the first demyelinating event. To the best of our knowledge, we examined the association between levels of a neurofilament type and a marker of neuronal regeneration (i.e. CREB) for the first time in MS.

These results lend further support to the notion that pNFH may potentially be used as a biomarker of disease activity and future progression of disability in MS.[6],[7],[8],[9],[10],[11] Establishment of the biomarker potential of pNFH is particularly important since serum pNFH levels can be measured by ELISA, whereas more state-of-the-art techniques (e.g., Simoa assay) are required for serum measurements of NFL.[18] It is thus unfortunate that serum pNFH levels were not associated with post-CIS disease activity. The lack of association between clinical progression parameters of MS and serum levels of pNFH and CREB is congruent with previous reports.[8],[19]

ROC analysis of CSF pNFH levels suggested low sensitivity and high specificity for the prediction of CIS-MS conversion. In other words, low (<730.9 pg/ml) CSF pNFH levels do not effectively differentiate between converting and nonconverting CIS. However, our results suggest that CIS patients with higher first-attack CSF pNFH levels (e.g., >1000 pg/ml) appear to have a high probability of developing MS in a 3-year follow-up duration. This information endows biomarker value to CSF pNFH and may be useful in treatment decisions of CIS patients.

Increased neurofilament levels in MS are generally recognized as an indicator of neuroaxonal degeneration. In fact, concomitant reduction of cerebral neurofilaments and elevation of CSF neurofilaments in MS, as the disease evolves to the progressive stage, supports this assertion.[2],[3],[4],[5],[12] Since neuronal regeneration efforts are decreased in advanced stages of disease, neurofilament levels may be a direct indicator of neuroaxonal impairment in progressive MS. However, in earlier stages of disease, such as CIS, neurofilament levels might also be reflecting neuronal regeneration efforts. The significant correlation between CSF pNFH levels versus CREB levels supports this notion and implies that pNFH may at least in part be an indicator of neuronal regeneration. In line with this view, several in vitro and in vivo studies have shown that promotion of neurite outgrowth and neuronal differentiation simultaneously enhances expression levels of neurofilaments and elements of CREB-signaling pathway.[20],[21],[22] Moreover, in the animal model of MS, the promotion of neuronal regeneration has been shown to simultaneously enhance expression levels of neurofilaments and phosphorylated-CREB in the central nervous system.[23]

CREB-signaling pathway is also utilized by remyelination and Th17-type immune response generation pathways.[24],[25] Thus, alternatively, correlation between CREB and pNFH levels may also be an indicator of increased Th17-Type T helper activity and enhanced remyelination efforts of oligodendrocytes in CIS patients with higher disease activity and neuroaxonal degeneration. Thus, increased CREB levels might be reflecting a backlash effect to increased inflammation in the central nervous system. These potential associations need to be further scrutinized in future studies.

A limitation of our study was low number of participants due to highly strict inclusion criteria (e.g., exclusion of patients having concomitant disease and receiving immunosuppressive medications, treatment with the same immunomodulating drug after MS conversion). Recruitment of patients with the same follow-up duration from a single center was another challenging factor. Future multi-center studies may overcome this limitation. Another limitation was lack of investigation of other neuronal regeneration factors such as phosphorylated-CREB, Akt pathway factors, and brain-derived neurotrophic factors. Due to explorative and preliminary nature of this study, only a limited number of factors were evaluated.

  Conclusion Top

In brief, CSF pNFH levels appear to predict long-term disease activity and MS conversion in CIS patients. Moreover, pNFH levels correlate with CREB levels suggesting that pNFH elevation may be a consequence of both neuronal degeneration and regeneration. The biomarker value of pNFH and its potential association with recompensing neuronal differentiation and remyelination need to be further studied.

Statement of ethics

This study was approved by the local ethics committee.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

De Lury A, Bisulca J, Coyle PK, Peyster R, Bangiyev L, Duong TQ. MRI features associated with rapid disease activity in clinically isolated syndrome patients at high risk for multiple sclerosis. Mult Scler Relat Disord 2020;41:101985.  Back to cited text no. 1
Bittner S, Oh J, Havrdová EK, Tintoré M, Zipp F. The potential of serum neurofilament as biomarker for multiple sclerosis. Brain 2021:awab241. doi: 10.1093/brain/awab241.  Back to cited text no. 2
Gordon BA. Neurofilaments in disease: What do we know? Curr Opin Neurobiol 2020;61:105-15.  Back to cited text no. 3
Ferreira-Atuesta C, Reyes S, Giovanonni G, Gnanapavan S. The evolution of neurofilament light chain in multiple sclerosis. Front Neurosci 2021;15:642384.  Back to cited text no. 4
Håkansson I, Tisell A, Cassel P, Blennow K, Zetterberg H, Lundberg P, et al. Neurofilament light chain in cerebrospinal fluid and prediction of disease activity in clinically isolated syndrome and relapsing-remitting multiple sclerosis. Eur J Neurol 2017;24:703-12.  Back to cited text no. 5
Petzold A, Eikelenboom MI, Keir G, Polman CH, Uitdehaag BM, Thompson EJ, et al. The new global multiple sclerosis severity score (MSSS) correlates with axonal but not glial biomarkers. Mult Scler 2006;12:325-8.  Back to cited text no. 6
Wang H, Wang C, Qiu W, Lu Z, Hu X, Wang K. Cerebrospinal fluid light and heavy neurofilaments in neuromyelitis optica. Neurochem Int 2013;63:805-8.  Back to cited text no. 7
Ljubisavljevic S, Stojanovic I, Basic J, Pavlovic DA. The validation study of neurofilament heavy chain and 8-hydroxy-2'-deoxyguanosine as plasma biomarkers of clinical/paraclinical activity in first and relapsing-remitting demyelination acute attacks. Neurotox Res 2016;30:530-8.  Back to cited text no. 8
Herrera MI, Kölliker-Frers RA, Otero-Losada M, Perez Lloret S, Filippo M, Tau J, et al. A pilot cross-sectional study to investigate the biomarker potential of phosphorylated neurofilament-H and immune mediators of disability in patients with 5 year relapsing-remitting multiple sclerosis. Front Neurol 2019;10:1046.  Back to cited text no. 9
Gresle MM, Liu Y, Dagley LF, Haartsen J, Pearson F, Purcell AW, et al. Serum phosphorylated neurofilament-heavy chain levels in multiple sclerosis patients. J Neurol Neurosurg Psychiatry 2014;85:1209-13.  Back to cited text no. 10
Pasol J, Feuer W, Yang C, Shaw G, Kardon R, Guy J. Phosphorylated neurofilament heavy chain correlations to visual function, optical coherence tomography, and treatment. Mult Scler Int 2010;2010:542691.  Back to cited text no. 11
Schmierer K, Parkes HG, So PW, An SF, Brandner S, Ordidge RJ, et al. High field (9.4 Tesla) magnetic resonance imaging of cortical grey matter lesions in multiple sclerosis. Brain 2010;133:858-67.  Back to cited text no. 12
Kular L, Needhamsen M, Adzemovic MZ, Kramarova T, Gomez-Cabrero D, Ewing E, et al. Neuronal methylome reveals CREB-associated neuro-axonal impairment in multiple sclerosis. Clin Epigenetics 2019;11:86.  Back to cited text no. 13
Liu W, Ye Q, Xi W, Li Y, Zhou X, Wang Y, et al. The ERK/CREB/PTN/syndecan-3 pathway involves in heparin-mediated neuro-protection and neuro-regeneration against cerebral ischemia-reperfusion injury following cardiac arrest. Int Immunopharmacol 2021;98:107689.  Back to cited text no. 14
Taherian N, Vaezi G, Neamati A, Etemad L, Hojjati V, Gorji-Valokola M. Vitamin B12 and estradiol benzoate improve memory retrieval through activation of the hippocampal AKT, BDNF, and CREB proteins in a rat model of multiple sclerosis. Iran J Basic Med Sci 2021;24:256-63.  Back to cited text no. 15
Alghamdi BS, AboTaleb HA. Melatonin improves memory defects in a mouse model of multiple sclerosis by up-regulating cAMP-response element-binding protein and synapse-associated proteins in the prefrontal cortex. J Integr Neurosci 2020;19:229-37.  Back to cited text no. 16
Polman CH, Reingold SC, Banwell B, Clanet M, Cohen JA, Filippi M, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol 2011;69:292-302.  Back to cited text no. 17
Hendricks R, Baker D, Brumm J, Davancaze T, Harp C, Herman A, et al. Establishment of neurofilament light chain Simoa assay in cerebrospinal fluid and blood. Bioanalysis 2019;11:1405-18.  Back to cited text no. 18
Gencer M, Akbayır E, Şen M, Arsoy E, Yılmaz V, Bulut N, et al. Serum orexin-A levels are associated with disease progression and motor impairment in multiple sclerosis. Neurol Sci 2019;40:1067-70.  Back to cited text no. 19
Lam KY, Chen J, Lam CT, Wu Q, Yao P, Dong TT, et al. Asarone from Acori Tatarinowii Rhizoma potentiates the nerve growth factor-induced neuronal differentiation in cultured PC12 cells: A signaling mediated by protein kinase A. PLoS One 2016;11:e0163337.  Back to cited text no. 20
Liao KK, Wu MJ, Chen PY, Huang SW, Chiu SJ, Ho CT, et al. Curcuminoids promote neurite outgrowth in PC12 cells through MAPK/ERK- and PKC-dependent pathways. J Agric Food Chem 2012;60:433-43.  Back to cited text no. 21
Won JH, Ahn KH, Back MJ, Ha HC, Jang JM, Kim HH, et al. DA-9801 promotes neurite outgrowth via ERK1/2-CREB pathway in PC12 cells. Biol Pharm Bull 2015;38:169-78.  Back to cited text no. 22
Zheng Q, Liu L, Liu H, Zheng H, Sun H, Ji J, et al. The Bu Shen Yi Sui formula promotes axonal regeneration via regulating the neurotrophic factor BDNF/TrkB and the downstream PI3K/Akt signaling pathway. Front Pharmacol 2019;10:796.  Back to cited text no. 23
Hernandez JB, Chang C, LeBlanc M, Grimm D, Le Lay J, Kaestner KH, et al. The CREB/CRTC2 pathway modulates autoimmune disease by promoting Th17 differentiation. Nat Commun 2015;6:7216.  Back to cited text no. 24
Jana M, Ghosh S, Pahan K. Upregulation of myelin gene expression by a physically-modified saline via phosphatidylinositol 3-kinase-mediated activation of CREB: Implications for multiple sclerosis. Neurochem Res 2018;43:407-19.  Back to cited text no. 25


  [Figure 1], [Figure 2], [Figure 3]

  [Table 1]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Article Figures
Article Tables

 Article Access Statistics
    PDF Downloaded60    
    Comments [Add]    

Recommend this journal