Multiple sclerosis (MS) is a chronic inflammatory and neurodegenerative disease of the central nervous system. While acute focal demyelination results in significant axonal injury and loss,1 the majority of axons survive the initial insult. Although some remyelination occurs, the axons remain largely devoid of myelin, forming so called chronic lesions.2 Myelin plays an important role in supporting fast and low energy-demanding saltatory conduction and in providing trophic support to axons and protecting them from the toxins, macrophages, and activated microglia. The restoration of myelin within chronically demyelinated lesions is potentially an important neuroprotective strategy in MS. Various remyelinating strategies have been suggested3 and several potential remyelinating agents are currently under investigation,4–6 emphasising an urgent need for reliable in vivo biomarkers of remyelination.
The latency of visual evoked potentials (VEPs) has been suggested as a sensitive measure of optic nerve demyelination/remyelination caused by acute optic neuritis (ON), particularly when multifocal stimulation (multifocal VEP, mfVEP) and inter-eye asymmetry analysis are used.7–10 Several recent clinical trials based on the acute ON model demonstrated a positive effect of remyelinating treatment on latency improvement.4,5,11
Optic neuritis may represent an ideal model to study acute lesions, but patient recruitment within a short window after symptom onset is a limiting factor. We have previously reported on the time course of latency recovery after acute ON and during the first 12 months10,12 and also on the relationship of severity of latency delay to subsequent MS progression.13 In addition, optic radiation (OR) lesions, while often asymptomatic, are very common in MS patients14–16 and also result in conduction delay detected by mfVEP. We have demonstrated a close correlation between OR lesion volume and mfVEP latency delay.17 Therefore, the visual pathway in its entirety represents a valuable model for tracking myelination, both acutely and chronically, and can also be readily visualized using diffusion MRI techniques.12,15
While acute lesions have likely advantages in remyelination potential and are known to at least partially remyelinate spontaneously,15 implementation of trial-based treatment protocols is difficult because a large proportion of MS lesions are clinically silent (apart from ON).18 Therefore, remyelination of chronically demyelinated axons, though likely to be more challenging to induce with therapy, is potentially more rewarding and beneficial, particularly considering that diagnosis (and treatment) of MS is typically delayed from the disease onset.19,20 It is for these reasons that chronic (rather than acute) lesions represent a major target of future remyelination strategies.
The VEP response is generated at the level of the primary visual cortex and thus reflects the speed of conduction (and, consequently, the degree of demyelination) along the entire visual pathway. In the absence of new lesional activity, prolongation of the VEP latency could be considered to be secondary to the degree of permanent demyelination along the entire visual pathway, including both the optic nerve and optic radiation. Close relationship between VEP latency and degree of demyelination along the visual pathway has recently been confirmed by several experimental and clinical studies.17,21–23 Concurrently, latency shortening as a result of potential remyelinating therapy is likely to reflect recovery of myelin in chronic lesions at any point along the visual pathway. Alteration of myelin content (and, consequently, latency change) is well documented to be very substantial after acute lesions. We expect at least partial recovery of latency in the 6 months after an acute ON, and more complete recovery may be demonstrated in remyelination studies designed for acute ON episodes. However, measuring the degree of remyelination in chronic MS lesions is more challenging. As a result, the sensitivity of measuring remyelination in the visual pathway depends on the reproducibility and stability of the VEP latency in the chronic MS population. Therefore, the aim of the current study was to investigate potential changes in latency of the mfVEP in a cohort of relapsing–remitting multiple sclerosis (RRMS) patients over a 12-month interval (representing the typical duration of a phase 2 clinical trial). In addition, we also estimated the potential sample size required for a remyelination-based clinical trial using different treatment effect sizes and the mfVEP latency as an outcome measure.
A total of 50 RRMS consecutive patients with no previous history of ON in at least one eye and 15 normal controls of similar age and gender composition were prospectively enrolled. Disease-modifying therapy was in use in 46 of 50 patients (3 refused treatment and 1 because of pregnancy). There were 19 patients on first-line therapy (13 interferon, 6 copaxon) and 27 patients on second-line therapy (14 Gilenya, and 13 on Tysabril , Lemtrada , Aubagio , or Tecfidera ).
Patients with a history of ON had an episode of acute optic nerve inflammation greater than 12 months before enrolment. The diagnosis of ON was made on the basis of clinical history (acute visual loss, pain on eye movement, color desaturation) and examination findings plus radiologic and/or electrophysiological evidence including delayed VEP latency.
Patients with any other systemic or ocular disease that could confound results, such as diabetes, retinal lesions, or glaucoma, were excluded. Multiple sclerosis patients and normal controls were tested twice with an interval of approximately 12 months to identify the chronic changes in mfVEP latency.
The eye with longest mfVEP latency was selected for the main analysis. In addition, patients with and without a prior history of ON were analyzed separately. Brain MRI was performed on the same day as mfVEP assessment.
The Institutional Review Board of University of Sydney and Macquarie University approved the study. Procedures followed the tenets of the Declaration of Helsinki, and written informed consent was obtained from all participants.
mfVEP Recording and Analysis
Monocular multifocal VEP testing was performed using the VisionSearch1 (VisionSearch Pty Ltd, Sydney, Australia) using standard stimulus conditions that entailed recordings from 56 segments of the visual field. Four gold cup electrodes were placed around the inion and used for bipolar recording from two channels: superior and inferior electrodes for vertical channel and left and right electrodes for horizontal channel. The channel with the largest amplitude (difference between min and max within the interval of 70–200 milliseconds) was selected for each segment of the visual field. The second peak of the wave of maximum amplitude for the selected channel of each segment in the visual field was used for latency analysis of the baseline test, as previously described.24 Latency change between baseline and follow-up recordings was determined as follows: the baseline mfVEP trace was moved along the waveform of the follow-up VEP response (limit ± 20 milliseconds, step 1 millisecond) until the highest cross-correlation between the two was found.25 This position was defined as a relative (to baseline) latency shift. Latency progression measured in this way demonstrated to be superior compared with conventional peak measurement.25 Averaged (across the entire stimulated field) baseline latency and latency change were used for analysis.
Statistical analyses were performed using IBM SPSS 22. Pearson correlation coefficient was used for bivariate correlation, whereas Student t-test or one-way analysis of variance (Tukey post hoc analysis) was used to compare mean values. Significance was determined at 0.05 level. For sample size calculation, G*Power software (v.188.8.131.52) was used26 (power = 80%, P = 0.05).
MRI Protocol, Optic Radiation Tractography, and Lesion Identification
The following sequences were acquired using a 3-T GE Discovery MR750 scanner (GE Medical Systems, Milwaukee, WI):
- Precontrast and postcontrast (gadolinium) Sagittal 3D T1: GE BRAVO sequence, FOV 256 mm, slice thickness 1 mm, TE 2.7 milliseconds, TR 7.2 milliseconds, flip angle 12°, and pixel spacing 1 mm.
- fluid-attenuated inversion recovery (FLAIR) CUBE; GE CUBE T2 FLAIR sequence, FOV 240 mm, slice thickness 1.2 mm, acquisition matrix (freq. × phase) 256 × 244, TE 163 milliseconds, and TR 8,000 milliseconds, flip angle 90°, pixel spacing 0.47 mm.
- Whole brain diffusion-weighted images using a spin echo, 64 directions, FOV 256 mm, acquisition matrix (freq. × phase) 128 × 128, slice thickness 2 mm, TE 83 milliseconds, TR 8,325 milliseconds, b-value = 1,000 and number of b = 0, and acquisitions = 2.
Probabilistic tractography was used to reconstruct OR fibers as previously described in detail.27 Briefly, after eddy current correction and motion compensation, diffusion tensor imaging and FLAIR T2 images were co-registered to the high resolution T1 structural image. Before the reconstruction, three regions of interest (ROI) were determined. The first ROI (diameter 10 mm) was placed at the chiasm and deterministic tractography was used to follow the optic tract fibers from the chiasm to the lateral geniculate nucleus, where a second ROI (7 mm) was placed. For OR reconstruction, a third ROI covering the calcarine sulcus was drawn manually in each hemisphere. Probabilistic tractography (ConTrack part of MrDiffusion package http://sirl.stanford.edu/software/) was then used between the lateral geniculate nucleus and calcarine ROIs. Initially 70,000 fibers were collected for OR tractography, of which the 30,000 best fibers were selected by a scoring algorithm. Optic radiation fibers were then manually cleaned using Quench software (http://sirl.stanford.edu/software/). Mayers loop was clearly visible in all OR reconstructions.
Multiple sclerosis lesions were identified on the co-registered T2 FLAIR images and segmented semi-automatically using JIM software (V7; Xynapse Systems Limited). Lesions were then intersected with OR fibers to identify and measure the volume of T2 FLAIR lesions within the OR.
A total of 50 RRMS patients and 15 normal controls of similar age and gender were enrolled (Table 1). Fifteen patients had a history of ON, whereas 41 patients demonstrated OR lesions at baseline. None of the OR lesions at baseline were gadolinium enhancing. One subject developed new OR lesions during the follow-up period (which resulted in bilateral latency increase) and was subsequently excluded from analysis. None of the study patients developed a new episode of ON during the follow-up period. The low incidence of new lesions during the 12 months follow-up may be a result of the disease-modifying therapy these subjects were on.
TABLE 1. -
||Patients (n = 50)
||Normal Controls (n = 15)
||43.2 ± 12.0
||42.4 ± 12.9
|Disease duration (years)
||8.3 ± 4.1
||1.3 ± 0.9
|History of ON
|Duration of follow-up (months)
||12.3 ± 1.4
|Total brain lesion volume (mm3)
||4,250 ± 4,020
|OR lesion volume (mm3)
||853 ± 934
|Baseline latency (milliseconds)
||156.8 ± 15.2
||145.1 ± 7.2
EDSS, Expanded Disability Status Scale; N/A, not applicable; ON, optic neuritis; OR, optic radiation.
An example of mfVEP recorded from RRMS patient with previous history of ON in the left eye is presented in Fig. 1 (upper row).
As expected, there was significant delay of mfVEP latency in RRMS patients compared with normal controls (156.8.0 ± 15.2 milliseconds vs. 145.1 ± 7.2 milliseconds, P < 0.001). Latency delay in RRMS patients did not correlate with age, gender, disease duration or Expanded Disability Status Scale. There was, however, a significant correlation with OR lesion volume (R2 = 0.38, P < 0.001).
Separate analysis of ON and nonoptic neuritis (NON) groups showed that latencies in both ON (167.9 ± 16.7) and NON (151.2 ± 11) eyes were significantly longer compared with normal controls (145.1 ± 7.2) (one-way analysis of variance, P < 0.001, post hoc: P < 0.001 for ON vs. NON, P < 0.001 for ON vs. controls, and P = 0.04 for NON vs. controls) (Fig. 2). Correlation of latency with OR lesion volume, reported above, was mainly driven by NON eyes (R2 = 0.42, P < 0.001 and R2 = 0.15, P = 0.14 for correlation of OR lesion volume with NON and ON eyes, respectively). The lower correlation between latency in ON cases and lesion volume is likely to reflect the dominating effect of the delay induced by optic nerve lesions.
Since demyelination may occur at multiple sites along the visual pathway, a subanalysis of patients with a history of ON and/or OR lesions (which constituted 45 of 50 patients) was performed. This group demonstrated significant latency delay (12 ± 15.2 milliseconds) compared with normal controls (158.8 ± 15.2 milliseconds vs. 145.1 ± 7.2 milliseconds, P < 0.001). This group of patients was later used for sample size calculation (see below).
Overall, there was no significant latency change in normal controls' and MS patients' eyes over the follow-up period. Separate analysis of ON and NON eyes also demonstrated no significant latency alteration between study time points (Table 2). Latency change in both groups did not correlate with age, gender, disease duration, or Expanded Disability Status Scale. The type of disease-modifying therapy used also did not have a differential effect on latency (P = 0.4, t-test between latency change in patients receiving first- and second-line therapy). An example of mfVEP latency analysis in the eye of a patient with RRMS and a previous history of ON is presented in Fig. 1 (lower row). Baseline and follow-up traces from an individual segment are magnified in the insert showing minimal change over time.
TABLE 2. -
||Baseline Latency (Milliseconds)
||Follow-up Latency (Milliseconds)
||145.1 ± 7.2
||145.1 ± 7.5
|Entire study cohort (n = 49)
||158.8 ± 15.2
||158.6 ± 14.6
|ON eyes (n = 16)
||167.9 ± 16.7
||167.6 ± 16.2
|NON eyes (n = 49)
||151.9 ± 11.0
||152.0 ± 10.9
NON, nonoptic neuritis; ON, optic neuritis.
Sample Size Calculation
Sample size calculation for a future remyelinating drug study was based on the assumption that differences in latency recovery between treated and placebo groups during the observation period is related to the degree of remyelination along the entire visual pathway (including optic nerve and optic radiation). The expected shortening of latency delay in the treated group represents the “treatment effect.” Becuase remyelination can occur exclusively within demyelinated lesions, only patients with known lesions along the visual pathway (including patients with OR lesions and/or history of ON) must be selected for the study of remyelinating therapies, which, in our cohort, comprised 45 of 50 patients (90%). This group demonstrated latency delay of 14.0 ± 15.2 milliseconds compared with controls. Therefore, we would propose to assess potential treatment effect by measuring any shortening of this latency delay. Because of paucity of available clinical data, it is difficult to estimate the potential treatment effect of remyelination therapy. Based on a recent clinical trial of opicinumab in acute optic neuritis, which demonstrated up to 41% of latency improvement,4 we believe that it is reasonable to expect that successful remyelination therapies should aim for a potential treatment benefit around 30% to 50%.
Figure 3 shows sample size calculation for treatment effect between 25% and 70%, which is best described by power function: y = 90,156x−1.94, where y represents sample size and x repersents treatment effect (%). It shows, for instance, that for the treatment effect of 40%, the total sample size required to demonstrate statistically significant latency shortening is 70 patients. Because patients with a prior history of ON or OR lesions constituted the majority (90%) of the consecutive MS patients examined in our study, the total cohort of RRMS patients needed to be screened for the study is estimated at (70/0.9) = 78. Greater numbers would be required if a smaller treatment effect was anticipated.
In the current study, we confirmed significant delay of mfVEP latency in RRMS patients, which was most profound in eyes with a previous history of ON. In addition, significant latency prolongation was also seen in a group of patients without a history of ON, where it was linked to the OR lesion volume, supporting our previously reported results.17 More importantly, while delayed, latency remained stable within the observation period in both ON and NON eyes. The unchanging nature of mfVEP latency in these subjects, who were all more than 12 months since any ON episode, further supports a previously proposed notion that the “window of opportunity” for spontaneous remyelination in MS is limited to the early post-acute period.10,28 It also suggests unfortunately that current therapeutic agents do not provide any evidence for improving levels of myelination, but at least they may be able to help preserve current levels.
Several lines of evidence suggest that permanent demyelination may contribute to accelerated axonal degeneration by rendering axons vulnerable to physiological stress.29–31 Chronic demyelination increases the energy demands of axonal conduction, which ultimately compromises axoplasmic adenosine 5' trisphosphate production,32 leading to an ionic imbalance and Ca2+-mediated axonal degeneration.29 In addition, lack of trophic support from myelin or oligodendrocytes and disruption of normal axon–myelin interactions may also lead to degeneration of chronically demyelinated axons.30,33 Residual inflammatory processes caused by infiltration of the chronic inactive lesions by T cells, macrophages, and microglia may also play some role in ongoing axonal damage.31 Furthermore, functional oligodendrocyte pathologic condition alone can result in significant axonal loss and progressive neurologic disability.34 Loss of demyelinated axons may also be exacerbated by activation of astrocytes: in particular, recent data demonstrated that A1 astrocytes might play a critical role in both oligodendrocyte injury and axonal loss within demyelinated lesions.35,36 Therefore, remyelination potentially represents an important neuroprotective strategy in MS,37 and clinical trials of remyelinating agents are now emerging.4,5
Because the magnitude of VEP latency delay in MS is related to the severity of visual pathway demyelination,21 the stable nature of mfVEP latencies over a relatively long period (12 months) in combination with its ability to measure the speed of conduction makes it a suitable candidate for an outcome measure in clinical trials involving remyelination. In particular, the assessment of myelination status of chronic lesions along the visual pathway using mfVEP may present a real opportunity to test the effectiveness of remyelinating agents in humans. With this in mind, we calculated the sample size for hypothetical clinical trial based on recovery of mfVEP latency and variable treatment effect. Our calculation shows that for reasonable (40%) treatment effect, a relatively small cohort of RRMS patients (78 patients) is required. Future remyelinating therapy trials need to be adequately powered to reduce type II error or the chance of a false-negative study result. For example, while the anti-Lingo study revealed a positive effect of treatment on remyelination, it was underpowered for the treatment effect to reach significance.11 By utilizing the sample size calculations provided, co-ordinators of future remyelinating therapy trials will be able to more accurately estimate the participant numbers required to avoid further underpowered studies. As such, if the treatment effect was smaller than the predicted 40%, a larger cohort would be needed (Fig. 3).
One potential disadvantage of the proposed approach is that patients with recent or new episodes of demyelination along the visual pathway must be closely monitored or even excluded from the analysis. However, because of the high effectiveness of the current disease-modifying therapy in MS, the number of new lesions in treated patients is not likely to be high. In fact, there was only one patient in our study cohort who developed a new OR lesion during the follow-up period. The ability of any new drug or agent to be able to aid in remyelination of chronic lesions is still largely unknown, but using mfVEP in such patients may provide validation of effect.
One limitation of the current study is relatively small sample size. Another limitation is that the subjects were on disease-modifying therapy and that several different agents were used. It was not possible to do a detailed subanalysis of individual drug effects because of the small numbers, but comparing first- and second-line agents showed no differences. It is not ethical to withhold therapy in MS because of the effectiveness of these agents in reducing disease progression. However, our findings suggest that despite treatment, the mfVEP latency remained constant across all treatment groups. While not showing evidence of remyelination, they at least seem to be preserving existing conditions. This means that any new proposed drug with remyelination capabilities could be added to an existing therapy or used in isolation and would still be able to demonstrate an effect if it could improve mfVEP latency.
In conclusion, given its tight relationship with the degree of myelination and its stability in chronic cases, we believe that the recovery of the mfVEP latency represents an ideal biomarker to assess the degree of treatment-induced remyelination in MS. It is uniquely placed to monitor remyelination of not only acute cases as previously reported but also chronic lesions in future trials of remyelination therapies. Our calculations show that reasonably sized patient cohorts will be required in such trials.
1. Trapp BD, Peterson J, Ransohoff RM, Rudick R, Mork S, Bo L. Axonal transection in the lesions of multiple sclerosis. N Engl J Med 1998;338:278–285.
2. Barkhof F, Bruck W, De Groot CJA, et al. Remyelinated lesions in multiple sclerosis: magnetic resonance image appearance. Arch Neurol 2003;60:1073–1081.
3. Olsen JA, Akirav EM. Remyelination in multiple sclerosis: cellular mechanisms and novel therapeutic approaches. J Neurosci Res 2015;93:687–696.
4. Cadavid D, Balcer L, Galetta S, et al. Safety and efficacy of opicinumab in acute optic neuritis
(RENEW): a randomised, placebo-controlled, phase 2 trial. Lancet Neurol 2017;16:189–199.
5. Green AJ, Gelfand JM, Cree BA, et al. Clemastine fumarate as a remyelinating therapy for multiple sclerosis (ReBUILD): a randomised, controlled, double-blind, crossover trial. Lancet 2017;390:2481–2489.
6. Cadavid D, Klistorner A, Ampapa R, et al. Correlation of physical, cognitive and MRI measures with multifocal visual evoked potential using baseline data from the anti-LINGO-1 SYNERGY trial in multiple sclerosis [meeting abstract]. Nerology 2016;78:P3.041.
7. Hood DC, Ohri N, Bo Yang E, et al. Determining abnormal latencies of multifocal visual evoked potentials: a monocular analysis. Doc Ophthalmol 2005;109:189–199.
8. Yang EB, Hood DC, Rodarte C, Zhang X, Odel JG, Behrens MM. Improvement in conduction velocity after optic neuritis
measured with the multifocal VEP. Invest Ophthalmol Vis Sci 2007;48:692–698.
9. Niklas a, Sebraoui H, Hess E, Wagner a, Then Bergh F. Outcome measures for trials of remyelinating agents in multiple sclerosis: retrospective longitudinal analysis of visual evoked potential latency. Mult Scler 2009;15:68–74.
10. Klistorner A, Arvind H, Garrick R, Yiannikas C, Paine M, Graham SL. Remyelination of optic nerve lesions: spatial and temporal factors. Mult Scler 2010;16:786–795.
11. Klistorner A, Chai Y, Leocani L, Albrecht P, Aktas O, Butzkueven H. Assessment of opicinumab in acute optic neuritis
using multifocal visual evoked potential. CNS Drugs 2018;32:1159–1171.
12. Graham SL, Klistorner A. Afferent visual pathways in multiple sclerosis: a review. Clin Exp Ophthalmol 2017;45:62–72.
13. Fraser C, Klistorner A, Graham SL, Garrick R, Billson F, Grigg JR. Multifocal visual evoked potential latency analysis: predicting progression to multiple sclerosis. Arch Neurol 2006;63:847–850.
14. Hornabrook RS, Miller D, Newton MR, et al. Frequent involvement of optic radiation in patients with acute isolated optic neuritis
. Neurology 1992;42:77–79.
15. Klistorner A, Vootakuru N, Wang C, et al. Decoding diffusivity in multiple sclerosis: analysis of optic radiation lesional and non-lesional white matter. PLoS One 2015;10:e0122114.
16. Toussaint D, Perier O, Verstappen A, Bervoets S. Clinicopathological study of the visual pathways, eyes and cerebral hemispheres in 32 cases of disseminated sclerosis. J Clin Neuro Ophthalmol 1983;3:211–220.
17. Alshowaeir D, Yiannikas C, Garrick R, et al. Latency of multifocal visual evoked potentials in nonoptic neuritis eyes of multiple sclerosis patients associated with optic radiation lesions. Invest Ophthalmol Vis Sci 2014;55:3758–3764.
18. Barkhof F, van Walderveen M. Characterization of tissue damage in multiple sclerosis by nuclear magnetic resonance. Philos Trans R Soc Lond B Biol Sci 1999;354:1675–1686.
19. Morrissey SP, Miller DH, Kendall BE, et al. The significance of brain magnetic resonance imaging abnormalities at presentation with clinically isolated syndromes suggestive of multiple sclerosis. A 5-year follow-up study. Brain 1993;116:135–146.
20. Klistorner A, Graham ECEC, Yiannikas C, et al. Progression retinal ganglion cell loss in multiple sclerosis is associated with new lesions in the optic radiations. Eur J Neurol 2017;24:1392–1398.
21. van der Walt A, Kolbe S, Mitchell P, et al. Parallel changes in structural and functional measures of optic nerve myelination after optic neuritis
. PLoS One 2015;10:e0121084.
22. You Y, Klistorner A, Thie J, Graham SL. Latency delay of visual evoked potential is a real measurement of demyelination in a rat model of optic neuritis
. Investig Ophthalmol Vis Sci 2011;52:6911–6918.
23. Heidari M, Radcliff AB, McLellan GS, et al. Evoked potentials as a biomarker of remyelination. Proc Nat Acad Sci USA 2019;116:27074–27083.
24. Sriram P, Wang C, Yiannikas C, et al. Relationship between optical coherence tomography and electrophysiology of the visual pathway in non-optic neuritis
eyes of multiple sclerosis patients. PLoS One 2014;9:e102546.
25. Malmqvist L, De Santiago L, Fraser C, Klistorner A, Hamann S. Exploring the methods of data analysis in multifocal visual evoked potentials. Doc Ophthalmol 2016;133:41–48.
26. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3 : a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007;39:175–191.
27. Klistorner A, Sriram P, Vootakuru N, et al. Axonal loss of retinal neurons in multiple sclerosis associated with optic radiation lesions. Neurology 2014;82:2165–2172.
28. Blakemore WF, Chari DM, Gilson JM, Crang AJ. Modelling large areas of demyelination in the rat reveals the potential and possible limitations of transplanted glial cells for remyelination in the CNS. Glia 2002;38:155–168.
29. Bruck W. Inflammatory demyelination is not central to the pathogenesis of multiple sclerosis. J Neurol 2005;252(suppl 5):V/10–V/15.
30. Trapp BD, Ransohoff R, Rudick R. Axonal pathology in multiple sclerosis: relationship to neurologic disability. Curr Opin Neurol 1999;12:295–302.
31. Kornek B, Storch MK, Weissert R, et al. Multiple sclerosis and chronic autoimmune encephalomyelitis: a comparative study of axonal injury in active, inactive and remyelinated lesons. Am J Pathol 2000;157:267–276.
32. Dutta R, MsDonough J, Yin X, Peterson J, Chang A. Mitochondrial dysfunction cause of axonal degeneration in multiple sclerosis patients. Ann Neurol 2006;59:478–489.
33. Peterson JW, Kidd D, Trapp BD. Axonal degeneration in multiple sclerosis: the histopathological evidence. In: Waxman SG, ed. Multiple sclerosis as a neuronal disease. Amsterdam: Elsevier, 2005:165–184.
34. Lappe-Siefke C, Goebbels S, Gravel M. Disruption of Cnp 1 uncouples oligodendroglial functions in axonal support and myelination. Nat Genet 2003;33:366–374.
35. Liddelow SA, Guttenplan KA, Clarke LE, et al. Neurotoxic reactive astrocytes are induced by activated microglia. Nature 2017;541:481–487.
36. Correale J, Farez MF, Cardona AE. The role of astrocytes in multiple sclerosis progression. Front Neurol 2015;6:1–12.
37. Irvine KA, Blakemore WF. Remyelination protects axons from demyelination-associated axon degeneration. Brain 2008;131:1464–1477.