Tau protein: a possible prognostic factor in optic neuritis and multiple sclerosis

Abstract

Background: Tau protein has been proposed as biomarker of axonal damage leading to irreversible neurological impairment in MS. CSF concentrations may be useful when determining risk of progression from ON to MS.

Objective: To investigate the association between tau protein concentration and 14-3-3 protein in the cerebrospinal fluid (CSF) of patients with monosymptomatic optic neuritis (ON) versus patients with monosymptomatic onset who progressed to multiple sclerosis (MS). To evaluate results against data found in a complete literature review.

Methods: A total of 66 patients with MS and/or ON from the Department of Neurology of Glostrup Hospital, University of Copenhagen, Denmark, were included. CSF samples were analysed for tau protein and 14-3-3 protein, and clinical and paraclinical information was obtained from medical records.

Results: The study shows a significantly increased concentration of tau protein in CSF from patients with relapsing–remitting MS and patients monosymptomatic at onset who progressed to MS, but interestingly no increased tau protein concentration in monosymptomatic ON. The concentration of tau protein was significantly correlated to Expanded Disability Status Scale score. No 14-3-3 protein was detected in any CSF sample.

Conclusions: The results of this study invite further exploration of the possible role of tau protein as a prognostic factor to predict progression from ON to MS in future studies.

Keywords

disease progression multiple sclerosis optic neuritis tau protein 14-3-3 protein

Introduction

Multiple sclerosis (MS) is an immune-mediated disease, manifested as lesions in the white matter of the central nervous system (CNS) and either temporary or permanent corresponding neurological deficits. Progress in the treatment of MS has increased the focus on rapid identification of individual patients’ clinical subtype, risk of progression from clinically isolated syndrome, e.g. optic neuritis (ON), to manifest MS and on monitoring the effect of therapeutic intervention. ON is the first symptom of MS in up to 20% of MS cases.¹

As a consequence the search for biomarkers has intensified. Several studies suggest tau protein and 14-3-3 protein as possible markers of axonal damage and corresponding permanent neurological damage, and suggest prognostic value as well. The main limitations of these studies have been retrospective study designs and relatively low numbers of participating patients.

The aim of this study was to evaluate tau protein and 14-3-3 protein as markers of clinical MS subtype, and as markers distinguishing monosymptomatic ON and ON with progression to MS in a larger patient population.

According to the axonal hypothesis,² neurological symptoms in MS can be explained by inflammatory demyelination causing conduction block and neurological symptoms if found in clinically active areas of the brain. Remyelination recreates normal conduction and contributes to remission. A possible explanation for the development of irreversible neurological deficits during MS is axonal damage. Axonal damage can happen in active inflammatory lesions early in the disease, but can also be seen later in chronically demyelinized lesions because of lack of glia or myelin-derived trophic factors. The measurement of tau protein and 14-3-3 protein concentration in cerebrospinal fluid (CSF) as markers of MS is based on the assumption that damaged axons release tau protein and 14-3-3 protein. Similarly, the measurement of tau protein and 14-3-3 protein in the CSF of patients with clinically isolated syndrome (CIS) (in this case ON) is based on the assumption that patients with CIS who progress to develop MS (e.g. ON as seen in MS) can have white matter lesions at the time of initial evaluation from which tau protein and/or 14-3-3 protein have been leaked.

Materials and methods

Glostrup Hospital, a Copenhagen University Hospital, is a centre for neurology in Eastern Denmark. At Glostrup University Hospital patients from a geographically well-defined county of Copenhagen are received. As part of routine evaluation of patients with suspected demyelinating disease, patients have magnetic resonance imaging (MRI) scanning of the brain done, and have a CSF as well as a blood sample taken soon after referral to the centre and before onset of treatment. After obtaining informed consent from the individual patient these CSF samples are kept in freezers at −80°C. For the present study, 66 samples from these freezers were sent to Statens Serum Institute (SSI) for quantification of tau protein and detection of 14-3-3 protein. These data were combined with data from the medical records of these patients, especially MRI results, IgG index, presence of oligoclonal bands and Expanded Disability Status Scale (EDSS) score.

The diagnosis of clinically definite or laboratory definite MS according to the Poser criteria was assigned to the patients by a neurologist working within the field of MS. None of the patients had MS based only on MRI findings. In all but seven of the patients we could find the original documentation for the diagnosis; in the remaining seven cases the results of diagnostic tests and the final diagnosis were described in the medical records.

Of the 66 patients, 16 were diagnosed with monosymptomatic ON and another 18 patients progressed to MS after initial ON diagnosis. In this study we define monosymptomatic ON as the situation where patients only have symptoms related to inflammation of the optic nerve (no other neurological symptoms). These patients will either prove to have isolated ON, or some of them might progress to MS at a later point in time than the cut-off date for this study (up to 13 years of follow-up). A further 27 patients with relapsing–remitting MS (RRMS) were part of the study. Patients were divided into the following categories: 1) monosymptomatic ON; 2) ON with progression to MS; and 3) RRMS. In this study, follow-up time was defined as the time period for which the patients have been followed at Glostrup, measured either as follow-up time from first symptoms or from CSF evaluation until cut-off date for this project. As the time of first symptoms is sometimes based on patients’ memory of when the first symptoms were present and may be biased due to differences in referral pathways, we have primarily based our follow-up information on follow-up time calculated from time of CSF evaluation at Glostrup. An overview of the material is given in Table 1.

Table 1.

Baseline characteristics and results of cerebrospinal fluid evaluation

Patient groups	ON		MS
Patient groups	Monosymptomatic	Progression to MS	RRMS
No	16	18	27
Median age^a (years) and range	37 (21;59)	35 (21;59)	34 (24;65)
Gender (female/male)	12/4	12/6	17/10
Follow-up time^b
from symptom onset (days)	595 (393;16425)	1791.5 (437;6570)	2296 (483;10950)
from CSF evaluation (days)	556 (413;1176)	543 (402;6483)	2015 (464;2717)
CSF- tau protein conc. (pg/ml)
- Median	172 (<75;392)	227 (<75;573)	169 (<75;767)
- Patients with increased conc. of tau protein^c (%)	0	28	11
Median number of days from first symptoms to CSF evaluation (days) and range^d	21(6;69)	490(7;3719)	90(12;8934)
- for tau-positive patients^e		35 days (28;680)
- for tau-negative patients^e		469 days (7;3719)
14-3-3 protein (positive/negative)	All negative	All negative	All negative

No statistically significant age difference between the groups.

No statistically significant difference in follow-up time from first symptoms to cut-off date between patients with monosymptomatic ON and patients with ON with progression to MS. Statistically significantly longer follow-up time for RRMS compared with both of the other groups. No statistically significant difference in follow-up time measured as time from CSF evaluation to cut-off date for this project between monosymptomatic ON and ON with progression to MS, but significantly longer follow-up time for RRMS patients.

Patients whose concentration of tau protein in CSF exceeded upper limit in their respective age group.

Median number of days from first symptoms to CSF evaluation is significantly higher in the group of patients with ON and progression to MS than in the group of patients with monosymptomatic ON(p < 0.05) and statistically significantly higher than in the group of patients with MS (p < 0.05).

No statistically significant difference in time from first symptoms to CSF sampling between tau-positive and tau-negative patients with ON and progression to MS.

CSF, cerebrospinal fluid; ON, optic neuritis; MS, multiple sclerosis; RRMS, relapsing–remitting multiple sclerosis

For a few patients, medical records were not complete as the referring neurological department had discontinued service. Wherever associations of parameters were evaluated which were not complete for all patients, only patients with complete information were taken into account. All major information used in this study was complete for the participating patients.

Tau protein and 14-3-3 protein were measured using the routine methods of the Department of Autoimmunity and the Department of Clinical Biochemistry at SSI. Tau protein concentration was measured using an ELISA-based commercial kit (Innogenetics, Belgium). The test is able to detect tau protein at concentrations of 75–1200 pg/ml. The intra-assay variation is 1.2–5.9%, and inter-assay variation is 1.7–6.0%. As the tau protein concentration in CSF increases with age, the reference values of tau protein were divided according to age: < 300 pg/ml for patients below 45 years, < 450 pg/ml for patients 45–70 years old and < 550 pg/ml for patients above 70 years of age. The manufacturer’s reference ranges were used, as the laboratory at SSI does not have samples available.

14-3-3 protein was detected using a qualitative method based on western blotting developed by SSI. Using SDS-PAGE gel electrophoresis, proteins from the CSF sample are separated. Subsequently, any 14-3-3 protein present is visualized using a polyclonal anti-14-3-3 antibody and antibodies allowing development of colour to be detected if 14-3-3 protein is present. The antibody used for detection is sc-629 from Santa Cruz Biotechnology. This antibody is a polyclonal rabbit antibody raised against the N-terminus of the beta 14-3-3. AP-conjugated anti-rabbit IgG was used for detection. A positive control and HeLa cell lysate were used as a positive controls in every run. The 14-3-3 analysis is well understood at SSI, where it is routinely used for detecting 14-3-3 in the spinal fluid from patients suspected of Creutzfeldt–Jakob disease, being positive in more than 70% of cases.

Statistics

Data were analysed using ranking and non-parametric methods such as Wilcoxon rank sum test and Spearman’s correlation coefficient in Microsoft Excel 2002.

Results

Patients with ON were divided into a group of monosymptomatic ON and a group with ON who showed progression to MS during follow-up. Of patients with MS, only the clinical subtype RRMS was included.

14-3-3 protein in CSF

We were unable to detect any 14-3-3 protein in any of the CSF samples.

Tau protein in CSF

All three patient groups had median tau protein concentrations within the normal limits. We found no significant correlation between patient age and tau protein concentration (Spearman’s correlation coefficient r_s = −0.12, NS).

Only patients with ON with progression to MS during follow-up and patients with RRMS had elevated tau protein in their CSF, altogether five patients in the first group (28%) and three patients in the second group (11%).

There was a statistically significant difference in median tau protein concentration in patients with monosymptomatic ON versus ON with progression to MS during follow-up (p < 0.02 using Wilcoxon rank sum test). There was also a statistically significant difference in median tau protein concentrations between the RRMS and monosymptomatic ON groups. However, there was no statistically significant difference between patients with RRMS and patients with ON that progressed to MS during follow-up.

We tested tau protein and its relationship to traditional markers such as degree of neurological impairment (EDSS score), intrathecal IgG production (IgG index), MR lesions and presence of oligoclonal bands. Statistically significant results were obtained in the group of patients with ON with progression to MS during follow-up. The best correlation was found with the EDSS score (r _s = 0.58, p = 0.05), where 34% of the variation could be explained by the correlation between tau protein and EDSS score.

Discussion

Tau protein and 14-3-3 protein have been suggested as markers of axonal damage, disease progression and development of irreversible neurological deficits, with possible use as markers of clinical MS subtype. Further, these proteins have been suggested as markers of progression to MS in patients with CIS suggestive of MS.

Some studies have suggested that tau protein and 14-3-3 protein could indeed be used as markers,^3-13 whereas others have not been able to confirm this (see Table 3).^14-18

Table 2.

Tau protein and relation to EDSS score, IgG index, MR lesions and presence of oligoclonal bands in CSF

Patient group	Relation to
Patient group	EDSS score	IgG index	Percentage of patients with oligoclonal bands in CSF	No of MR lesions
ON with progression to MS	r_s = 0.58; p = 0.05	r_s = 0.22; p = 0.05	93.8%	r_s = 0.21; NS
RRMS	r_s = 0.63; NS	r_s = −0.17; NS	63%	r_s = 0.33; NS
Monosymptomatic ON	r_s = 0.17; NS	r_s = 0.13; NS	22%	r_s = 0.48; NS

CSF, cerebrospinal fluid; EDSS, Expanded Disability Status Scale; ON, optic neuritis; MR, magnetic resonance; MS, multiple sclerosis; RRMS, relapsing–remitting multiple sclerosis

Table 3.

Overview of results published in the literature

Study (journal, year of publication)	No of patients with MS	No of patients with CIS	Tau as marker of MS subtype		14-3-3 as marker of MS subtype		Tau as markers of progression from CIS to MS		14-3-3 as a marker of progression from CIS to MS
Study (journal, year of publication)			Yes	No	Yes	No	Yes	No	Yes	No
Martinez-Yelamos et al.¹³ (Neurology, 2001)		38							x
Süssmuth et al.⁵ (Neuroscience Letters, 2001)	17		x
De Seze et al.¹⁷ (Journal of Neurology, 2002)	22		x
Jimenez-Jimenez et al.¹⁴ (Acta Neurologica Scandinavica, 2002)	20			x
Martinez-Yelamos et al.¹¹ (Neuroscience Letters, 2004)	32		x
Martinez-Yelamos et al.¹² (Journal of Neurology, 2004)		85							x
Collucci et al.⁸ (Multiple Sclerosis, 2004)	43	20				x			x
Bartosik-Psujek and Archelos,⁹ (Journal of Neurology, 2004)	114		x
Rostasy et al.⁶ (Journal of Child Neurology, 2005)	25		x
Brettschneider et al.⁷ (Multiple Sclerosis, 2005)	52	50	x
Bartosik-Psujek and Stelmasiak,⁴ (Journal of Neural Transmission, 2006)	60		x
Guimares et al.¹⁵ (Multiple Sclerosis, 2006)	50			x
Brettschneider et al.¹⁰ (Multiple Sclerosis, 2006)	38	52	x				x
Terzi et al.³ (Acta Neurologica Scandinavica, 2007)	45		x
Hein et al.¹⁶ (Neuroscience Letters, 2008)		20		x				x
Valis et al.¹⁸ (Neuroendocrinology Letters, 2008)				x						x
Bartosik-Psujek et al.¹⁹ (Acta Neurologica Scandinavica, 2011)	54			x

CIS, clinically isolated syndrome; MS, multiple sclerosis

A major limitation in many of these studies has been relatively low numbers of participating patients and correspondingly low statistical power and increased risk of type I error. In our study we have investigated a larger patient population, and our data on conversion from ON to MS are based on a follow-up period of up to13 years.

Tau protein concentration was found to be elevated in patients with ON and progression to MS and in RRMS patients, while there was a statistically significant difference in tau protein concentration between patients with monosymptomatic ON and patients with ON with progression to MS. This could potentially imply that increased tau protein measured when CIS is diagnosed could be indicative of later progression to MS.

This finding is supported by Brettschneider et al.,¹⁰ whose results show significantly elevated tau protein concentration in the CSF in patients with CIS, and who also found that a combination of MRI results and tau protein concentration in CSF was better at predicting conversion from CIS to MS than any one of the parameters alone. However, the studies by Brettschneider et al.⁷ and Hein et al.¹⁶ have not been able to confirm this, perhaps partly due to low numbers of patients (52 with CIS and 21 with CIS at baseline, respectively).

With regards to 14-3-3 protein, our study was unable to confirm a possible role of this protein as a predictive marker as in the studies by Martinez-Yelamos et al.^12,13 and the study by Collucci et al.⁸ We did not detect 14-3-3 protein in any of the patients in our study.

Based on the present study, though, it seems that conversion to MS from ON may possibly be predicted by measuring the concentration of tau protein. It will be of great importance to investigate this relationship further, and it clearly offers very interesting future perspectives including earlier onset of treatment.

However, the present study also raises the question of clinical sensitivity and clinical specificity of a suggested biomarker and the resulting usefulness of a given biomarker in daily clinical work. We found tau protein above normal range for age in 28% of the patients with ON with progression to MS during follow-up, and in 11% of the patients with RRMS. For comparison, a recent study by Bartosik-Psujek et al.¹⁹ reports detection of tau protein in 100% of CSF samples of patients with RRMS and 85% of patients with secondary progressive MS; how many of these have tau protein above the normal range, however, was not reported.

As can be seen from Table 4, the results which have been presented in the scientific literature so far have been varied in terms of both sensitivity and specificity, and even more so when taking the corresponding predictive value into account.

Table 4.

Detection of increased tau protein and 14-3-3 in cerebrospinal fluid and the predictive value if provided in published article

Study	No of patients with MS	No of patients with CIS	% above upper normal limit of tau protein in CSF			% where 14-3-3 protein was detected in CSF			Predictive value
Study			MS	CIS	Controls	MS	CIS	Controls
Süssmuth et al.⁵ (Neuroscience Letters, 2001)	17		59
Collucci et al.⁸ (Multiple Sclerosis, 2004)	43	20				42	30	21
Bartosik-Psujek and Archelos,⁹ (Journal of Neurology, 2004)	114		Tau detected in 83% with MS			22		14 % of controls with inflammatory neurological disease
Hein et al.¹⁶ (Neuroscience Letters, 2008)		20					5
Martinez-Yelamos et al.¹² (Journal of Neurology, 2004)		85					8		71 % of 14-3-3 positive patients and 41% of 14-3-3 negative patients developed CDMS
Martinez-Yelamos et al.¹³ (Neurology, 2001)		38					13		80% of 14-3-3 positive patients and 57.5 % of 14-3-3 negative patients developed CDMS
De Seze et al.¹⁷ (Journal of Neurology, 2002)	22					9	7
Guimarãres et al.¹⁵ (Multiple Sclerosis, 2006)	50		6
Bartosik-Psujek et al.¹⁹ (Acta Neurologica Scandinavica 2011)	54		Tau protein detected in 100% of patients with RRMS and 85% of patients with SPMS

CIS, clinically isolated syndrome; MS, multiple sclerosis; RRMS, relapsing–remitting multiple sclerosis; SPMS, secondary progressive multiple sclerosis

In order to be valuable as a biomarker in clinical practice, measurement of the biomarker must be very sensitive (low risk of false negative outcomes) and also very specific (low risk of false positive outcomes). In the present study, measurement of tau protein in CSF is a test with high predictive value of a positive test (increased tau protein concentration in CSF) based on the diagnosis at the moment of analysis, but a likewise low predictive value of a negative test (tau protein concentration within the normal range), since 72% of the patients with ON and progression to MS and 89% of RRMS patients did not show increased tau protein concentration. This must be seen as a major disadvantage for tau protein as a possible biomarker. However, it may turn out that a combination of tau protein measurement with other diagnostic tools such as MRI results can improve this aspect.

Interestingly, we could not detect any 14-3-3 protein in any CSF sample using our qualitative technique. It is thought-provoking that this study fails to detect 14-3-3 protein in any of the CSF samples. De Seze et al.¹⁷ were also only able to detect 14-3-3 protein in less than 10% of the MS patients in their study. Bartosik-Psujek et al.,⁷ who found 14-3-3 protein in 22% of the patients with MS in their study, used a quantitative ELISA-based method whereas the one used in our study is a qualitative test.

In earlier studies evaluating 14-3-3 protein as a marker,^5,7,12,13 14-3-3 protein was found in between 6–22 % of patients with MS. Only Colucci et al.⁶ found 14-3-3 protein in 38% of the CSF samples and found higher sensitivity for 14-3-3 protein as a marker compared with tau protein. Likewise, Martinez-Yelamos et al.¹² found that 14-3-3 protein had a higher predictive value than MRI findings when predicting progression from ON to MS. Martinez-Yelamos et al.¹² found 14-3-3 protein in seven patients (8.2%). Based on our findings it does not seem that 14-3-3 protein is a useful marker, but this might as well be due to the sensitivity of the detection method. The detection method used for 14-3-3 protein is well understood at SSI, where it is routinely used to detect 14-3-3 protein in patients with Creutzfeldt-Jakob disease, being positive in more than 70% of cases. In general, however, electrochemical fluorescence-based methods are more sensitive than alkaline phosphatise-based detection methods, and this may explain why our results were negative.

The tau protein concentration was most strongly correlated with EDSS score and IgG index for the patients with ON with progression to MS, and most strongly correlated with EDSS score for patients with RRMS. The correlation with EDSS score could be explained by the fact that both EDSS score and tau protein measure axonal damage, either directly or indirectly via neurological disability. The correlation with IgG index supports the idea that the degree of axonal damage, at least in active demyelinated lesions, reflects the strength of the immune-mediated inflammatory response in the lesion.

Another interesting result was that only the subgroups ON with progression to MS and RRMS show increased tau protein concentration, while no patient with monosymptomatic ON had tau protein concentrations above the reference level. Our results lead to the hypothesis that tau protein could potentially be used as a marker of disease progression from ON to MS, and thus could potentially play an important role in distinguishing between ON as in MS and isolated ON at initial evaluation. When evaluating this, however, it is important to be aware that the present study was not designed in order to evaluate tau protein as a prognostic marker, but rather to investigate differences in tau protein concentration between groups of patients with monosymptomatic ON, ON with progression to MS and RRMS. A larger and more prospective study will be needed in order to clarify this point.

It is also important to bear in mind that the patient population investigated in this study showed significant variations in time from first symptoms to time of CSF sampling and evaluation at Glostrup Hospital, ranging from 2–8979 days, and this could have had an impact on the results. However, it is also worth noting that there was no statistically significant difference in time from first symptoms to CSF evaluation between the tau-positive patients with ON with progression to MS and the tau-negative patients with ON with progression to MS (see Table 1). The variations in time from first symptoms of demyelinating disease to CSF evaluation at Glostrup Hospital are in this study due to differences in referral pathways for the study subjects. It is not yet established how tau protein or 14-3-3 protein concentration develops over time in individual patients and therefore the optimal time for CSF evaluation is not known. Future studies assessing concentration of tau protein or 14-3-3 protein over time in the CSF of patients with CIS and MS will therefore be valuable. There were no major differences in follow-up time between the groups of monosymptomatic ON versus ON with progression to MS in this study. Other studies have shown predictive value of high tau protein concentrations in CSF in the beginning of the disease leading to more rapid development of disability.^11,12 Brettschneider et al.¹⁰ on the other hand could not find any significant difference in tau protein concentrations between patients with a CIS with progression to MS and patients with CIS. Other studies evaluating the predictive role of biomarkers in MS have focused more on 14-3-3 protein. Martinez-Yelamos et al.^12,13 studied the predictive value of 14-3-3 protein in CSF compared with other predictive instruments, such as presence of nine T2 lesions and at least three Barkhoff criteria, and found a higher predictive value using 14-3-3 protein. 14-3-3 protein has also been found to predict early relapse and short-term disability.¹⁷ The present study failed to detect 14-3-3 protein.

Based on the present study it may be worthwhile to further investigate the hypothesis that tau protein could have a prognostic role in CIS suggestive of MS such as ON.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest statement

The authors declare that they have no conflicts of interest.

References

Sorensen

Frederiksen

Bronnum-Hansen

Petersen

. Opticus neuritis as onset manifestation of multiple sclerosis: a nationwide, long-term study. Neurology 1999; 53: 473–478.

Trapp

Ransohoff

Rudick

. Axonal pathology in multiple scleroses: relationship to neurologic disability. Curr Opin Neurol 1999; 12: 295–302.

Terzi

Birinci

Cetinkaya

Onar

. Cerebrospinal fluid total tau protein levels in patients with multiple sclerosis. Acta Neurol Scand 2007; 115: 325–330.

Bartosik-Psujek

Stelmasiak

. The CSF levels of total-tau and phosphotau in patients with relapsing-remitting multiple sclerosis. J Neural Transm 2006; 113: 339–345.

Süssmuth

Reiber

Tumani

. Tau protein in cerebrospinal fluid (CSF): a blood-CSF barrier related evaluation in patients with various neurological diseases. Neurosci Lett 2001; 300: 95–98.

Rostasy

Withut

Pohl

Lange

Ciesielcyk

. Tau, phospho-tau and S-100B in the cerebrospinal fluid of children with multiple sclerosis. J Child Neurol 2005; 20: 822–825.

Brettschneider

Maier

Arda

Claus

Süssmuth

. Tau protein level in cerebrospinal fluid is increased in patients with early multiple sclerosis. Mult Scler 2005; 11: 261–265.

Colluci

Roccagliata

Capello

Narciso

Latronico

Tabalton

. The 14-3-3 protein in multiple sclerosis: a marker of disease severity. Mult Scler 2004; 10: 477–481.

Bartosik-Psujek

Archelos

. Tau protein and 14-3-3 in the cerebrospinal fluid of patients with multiple sclerosis an correlate with intrathecal synthesis of IgG. J Neurol 2004; 251: 414–420.

10.

Brettschneider

Petzold

Junker

Tumami

. Axonal damage markers in the cerebrospinal fluid of patients with clinically isolated syndromes improve predicting conversion to definite multiple sclerosis. Mult Scler 2006; 12: 143.

11.

Martinez-Yelamos

Saiz

Bas

Hernandez

Graus

. Tau protein in cerebrospinal fluid: a possible marker of poor outcome in patients with early relapsing-remitting multiple sclerosis. Neurosci Lett 2004; 363: 14–17.

12.

Martinez-Yelamos

Rovira

Sanchez-Valle

Martinez-Yelamos

Tintore

. CSF 14-3-3 protein assay and MRI prognostic markers in patients with a clinically isolated syndrome suggestive of MS. J Neurol 2004; 251: 1278–1279.

13.

Martinez-Yelamos

Saiz

Sanchez-Valle

Casado

Ramon

. 14-3-3 protein in the CSF as prognostic marker in early multiple sclerosis. Neurology 2001; 57: 722–724.

14.

Jimenez-Jimenez

Zurdo

Hernanz

Medina-Acebron

deBustos

Barcenilla

. Tau protein concentrations in cerebrospinal fluid of patients with multiple sclerosis. Acta Neurol Scand 2002; 106: 351–354.

15.

Guimarãres

Cardoso

Sá

. Tau protein seems not to be a useful routine clinical marker of axonal damage in multiple sclerosis. Mult Scler 2006; 12: 354–356.

16.

Hein (née Maier)

Köhler

Diem

Sättler

Demmer

. Biological markers of axonal degeneration in CSF and blood of patients with first event indicative of multiple sclerosis. Neurosci Lett 2008; 436: 72–74.

17.

De Seze

Peoc’h

Ferriby

Stojkovic

Laplanche

Vermersch

. 14-3-3 protein in the cerebrospinal fluid of patients with acute transverse myelitis and multiple sclerosis. J Neurol 2002; 249: 626–627.

18.

Valis

Talab

Stourac

Andrys

Masopust

. Tau protein, phosphorylated tau protein and beta-amyloid 42 in the cerebrospinal fluid of multiple sclerosis patients. Neuroendocrinol Lett 2008; 29: 971–976.

19.

Bartosik-Psujek

Psujek

Jaworski

Stelmasiak

. Total tau and S100b proteins in different types of multiple sclerosis and during immunosuppressive treatment with mitoxantrone. Acta Neurol Scand 2011; 123: 252–256.