Anti-myelin oligodendrocyte glycoprotein antibodies: Magnetic resonance imaging findings in a case series and a literature review

Abstract

Myelin oligodendrocyte glycoprotein is a protein exclusively expressed on the surface of oligodendrocytes and myelin in the central nervous system. Antibodies against myelin oligodendrocyte glycoprotein were initially detected in children with demyelinating syndromes, and more recently reported in a broad spectrum of central nervous system demyelinating diseases in adults, including neuromyelitis optica spectrum disorders and bilateral optic neuritis. Patients with myelin oligodendrocyte glycoprotein antibody-associated demyelination appear to have unique clinical and radiological features. To the best of our knowledge a series of Italian patients with optic neuritis and positivity to myelin oligodendrocyte glycoprotein antibodies has not yet been reported and the paper on myelin oligodendrocyte glycoprotein antibodies are more focused on clinical features, diagnosis and outcome than on the radiological appearance, so we want to retrospectively report magnetic resonance imaging features of a group of eight patients, who came to our Ophthalmologic Emergency Department for optic neuritis and were found seropositive for myelin oligodendrocyte glycoprotein antibodies, comparing our data with the findings described in the literature.

Keywords

Anti-myelin oligodendrocyte glycoprotein demylinating disorders magnetic resonance imaging neuromyelitis optica optic neuritis recurrent optic neuritis

Introduction

Myelin oligodendrocyte glycoprotein (MOG) is a protein exclusively expressed on the surface of oligodendrocytes and myelin in the central nervous system, located on the outer surface of the myelin sheath.^1,2 Antibodies (Abs) against MOG were initially detected in children with demyelinating syndromes^3–10 and more recently reported in a broad spectrum of central nervous system demyelinating diseases in adults: in patients with aquaporin-4 (AQP4)-Abs-negative neuromyelitis optica spectrum disorder (NMOSD),^11–17 in bilateral optic neuritis (ON),¹⁸ in isolated ON,¹⁹ in recurrent ON,²⁰ clinically isolated syndrome (CIS),¹⁷ acute disseminated encephalomyelitis (ADEM),^16,17 in patients with demyelinating diseases distinct from multiple sclerosis (MS) and neuromyelitis optica (NMO),²¹ and in patients with clinically defined MS according to the McDonald criteria.^22–27 It has been hypothesised that the presence of anti-MOG Abs in distinct diseases may limit the use of the MOG-Abs assay as a specific diagnostic biomarker.

Patients with MOG-Abs-associated demyelination appear to have unique clinical, radiological features and outcomes.²⁶

The detection of the association between MOG-Abs and demyelinating disorders in adults is recent and in the literature we can find papers describing mainly clinical features, diagnosis and prognosis of patients with this type of antibody, with brief mention of radiological appearance but, to the best of our knowledge, no one reporting groups of Italian patients presenting with ON or focused on the radiologic aspects of this disorder, so we want to retrospectively report magnetic resonance imaging (MRI) features of a group of eight patients, who came to our Ophthalmologic Emergency Department for ON and were found seropositive for MOG-Abs, comparing our data with findings described in the literature.

Material and methods

Eight patients, (six females and two males, age range: 26–40 years; mean age: 32.8 ± 5.7 years) came to our Ophthalmologic Emergency Department between February 2015–November 2016 complaining of ocular pain, worsened by eye movement (8/8), visual loss in one eye (5/8), visual field loss (3/5) diagnosed as ON (three of the left eye, four of the right eye and one recurrent bilateral ON, first left, afterwards right ON). All patients underwent ophthalmologic and neurological evaluation and no contrast computed tomography (CT) scan of brain and orbits on their admission, and MRI of brain, cervical spine and orbits in a time from 2–20 days after the onset of symptoms (mean days: 5.7 ± 5.4). MRI of thoracolumbar spine was executed depending on the symptoms (1/8).

CT scans were executed on Somatom Definition Flash (Siemens, Forchheim, Germany) with the following acquisition parameters: kV:120, mAs: 320, collimation: 40 × 0.6 mm, tube rotation: 1 s; reconstruction thickness: 3 mm; reconstruction filters: H21s smooth for soft tissues and H60 sharp for bone.

All MRI examinations were executed on a 1.5 T magnet (Avanto, Siemens, Forchheim, Germany), with sequences on the brain: sagittal T1 spin echo (SE) (thickness: 5 mm), axial proton density (PD) + T2 turbo spin echo (TSE) (thickness: 5 mm), diffusion weighted imaging with b values = 0, 500, 1000 s/mm² (thickness: 5 mm), axial T1 SE, coronal fluid attenuated inversion recovery (FLAIR) (thickness: 5 mm), sagittal dark fluid T2 space (thickness: 1 mm, with multiplanar reformation on axial and coronal planes); on the orbits: axial T2 TSE with fat saturation (FS) (thickness: 3 mm), T2 short tau inversion recovery (STIR) space (thickness: 1 mm, with multiplanar reformation on sagittal and coronal planes); on the spine: sagittal T2 TSE (thickness: 4 mm), sagittal T1 TSE (thickness: 4 mm), sagittal T2 STIR. Gadolinium (1 ml/10 kg of gadobutrol, Gadovist, Bayer Schering Pharma) was administered to all patients intravenously, before the acquisition of: axial and coronal T1 SE FS on the orbits (thickness: 3 mm), axial T1 SE FLOW (thickness: 5 mm) and axial volumetric interpolated brain examination (VIBE) (thickness: 1 mm) on the brain, and sagittal T1 TSE FS on the spine (thickness: 4 mm).

All patients were subjected to steroid therapy intravenously within 24 h of their access to the Emergency Department (methylprednisolone 1 g/d for five days, followed by a low-dose steroid oral regimen for 15 days), so the CT was performed prior to therapy, but the MRI was always performed after the start of steroid treatment.

According to the protocol established in our Hospital in January 2014 for patients with new onset of unilateral or bilateral, severe or recurrent ON, suspected to be of demyelinating origin, patients underwent various investigations including cerebrospinal fluid analysis and the collection of serum samples for the detection of AQP4-Abs and MOG-Abs. AQP4-Abs was measured by immunofluorescence using HEK 293 cells transfected with AQP4-GFP, and anti-MOG immunoglobulin G (IgG) Abs were measured by enzyme-linked immunosorbent assay (ELISA), employing the recombinant extracellular domain of MOG as antigen, as previously described.²⁴ The mean optical density (OD) plus two standard deviations of 307 healthy blood donors was chosen as the normal range; serum readings higher than this value (OD = 0.802) were considered positive for MOG-Abs. Samples positive for MOG-Abs were then tested against the recombinant human MGT-30 protein, as irrelevant antigen, using an ELISA assay routinely used for testing anti-titin Abs in patients affected by myasthenia gravis.²⁸

From January 2014–July 2016, a total of 51 patients with ON were tested for MOG-Abs and eight were found to be seropositive (15.6%).

Patients positive to MOG-Abs underwent clinical follow-up for 3–12 months (mean follow-up length: 6.6 ± 3.3 months), four patients executed follow-up MRI at three months and one patient performed one follow-up MRI at three months and a second one at six months; 4/8 patients underwent a longitudinal serum sample collection at six months.

Results

Median MOG-Abs titre was 1:640 (range 1:320–1:640); the search of oligoclonal bands was negative; only the patient affected by recurrent ON showed double positivity for MOG and AQP4-Abs (MOG- IgG titre 1: 320, AQP4-IgG titre 1:640).

Only 1/8 CT scan was positive (Figure 1(a)) showing diffuse hypodensity of deep and subcortical white matter, no involvement of the basal ganglia was visible, while brain MRI showed pathologic findings in 8/8 patients (Figures 1 –8) with various aspects.

Two out of eight patients (Figures 1 and 2) showed diffuse white matter hyperintensities without evidence of alterations of the grey matter; 1/2 showed gadolinium enhancement (Figure 2(b)); in both cases infratentorial lesions were present (Figure 1(d), Figure 2(c)); 1/2 showed involvement of the corpus callosum (Figure 1(a), (d)).

In 4/8 patients (Figure 3, Figure 4(a), (b), Figure 5(a)–(c), Figure 6(b)) we found lesions that can be defined as ‘MS-like’: hyperintense in T2, ovoid and well defined, with predilection for juxta-cortical and periventricular white matter, with the major axes perpendicular to the ventricular surface. Two out of four patients showed gadolinium enhancement of the lesions (Figure 3(b), Figure 4(b)). One out of four patients showed involvement of the cortical grey matter (Figure 4(c)) and infratentorial lesion (Figure 4(d)), 1/4 showed enlargement of the Sylvian fissures and posterior horn of lateral ventricles (Figure 5(d)), 1/4 showed T2 hyperintensities in the cervical medulla (Figure 6(c)), without gadolinium enhancement.

One out of eight patients showed only two lesions in the supratentorial brain (Figure 7(a)), symmetrical, hyperintense in T2-weighted images, located in the right and left thalamus, coexisting with infratentorial lesions in the pons and medulla oblongata (Figure 7(c)) and patchy hyperintensities of the cervical medulla.

The patient with double positivity for MOG and AQP4-Abs showed two lesions of the supratentorial white matter (Figure 7(a) and (b)), patchy T2 hyperintensities in the cervical medulla and longitudinally extensive transverse myelitis extending over D3, D4 and D5.

In our group, only 1/8 patient showed significant swelling of the optic nerve on the affected side (Figure 6(a)).

After the resolution of the acute phase, all patients showed a complete recovery of the visual impairment with restoring of the pre-ON visual score (median visual score: 7.8).

The follow up MRI of the patient in Figure 1 (Figure 9) showed mild reduction of the extensively confluent T2-hyperintensity of the white matter of frontal and parietal lobes.

Figures 10 and 11 show progressive reduction of the hyperintense lesions pointed out in Figure 2 with no more evidence of contrast enhancement.

Figures 12 and 13 are control MRIs at three months of the patients in Figure 6 and 7, respectively, showing stability of some lesions (Figure 12(b) and (c), Figure 13(a) and (b)) and disappearance (Figure 12(a)) or reduction (Figure 13(c)) of others.

Figure 14, which is the control MRI of patient in Figure 8, showed reduction of all lesions.

Longitudinal serum sample collections at six months confirmed the presence of MOG-Abs, with unchanged Abs levels.

The definitive diagnosis made by our neurologists was NMOSD with AQP4-IgG in one case and NMOSD without AQP4-IgG in 7/8 cases, according to the revised diagnostic criteria for neuromyelitis optica spectrum disorders.²⁹

Discussion

The positivity to MOG-Abs has been primarily associated in adults with NMO/NMOSD, and the fact that MOG-Abs-positive patients showed differences compared with patients positive for AQP4-Abs has been highlighted by some authors:^12–16 Kitley et al.,¹² making a comparative study between 20 patients with AQP4-Abs and patients with MOG-Abs affected by NMOSD, reported that patients with anti-MOG were significantly more likely to have ADEM-like lesions, defined as lesions in the deep grey nuclei or fluffy white matter than patients with AQP4-Abs, and the grey matter lesions were significantly more common in MOG-Abs-positive patients. Moreover, they observed that, evaluating MRI of the spine, the conus was significantly more likely to be involved in patients with MOG-Abs than in patients with AQP4-Abs.

Sato et al.¹³ reported that patients with MOG-Abs have distinctive radiological manifestations as compared with patients positive for AQP4-Abs: they have more optic nerve lesions than spinal cord lesions and more frequently have bilateral simultaneous ON. They also reported the presence of cerebral lesions in only six out of 16 cases, mainly represented by nonspecific white matter lesions, and the presence of spinal cord lesions in six out of 16 cases, distributed more frequently in the thoracolumbar region, while positive for AQP4-Abs-patients frequently had cervical lesions.

Pröbstel et al.¹⁴ found MOG-Abs in four of 48 patients with NMO/NMOSD: they described juxta-ventricular lesions in one patient and lesions in the thalamus and corpus callosum in another patient. All the four patients had longitudinally extensive transverse myelitis.

In their group of 14 patients positive for MOG-Abs, Yan et al.¹⁵ observed brain lesions in only eight patients, five periventricular MS-like and three ADEM-like, consisting of deep grey matter and fluffy white matter lesions with enhancement, and observed spinal lesions at MRI in 14/14 patients associated with moderate cord oedema. The presence of contrast enhancement in ADEM-like lesions was interpreted as a sign of disruption of the blood-brain barrier.

Kim et al.²¹ affirmed that MOG-Abs may be a disease-specific biomarker in patients with demyelinating disorders distinct from both MS and NMO. They observed 17 patients positive for MOG-Abs in a cohort of 270 adult patients with demyelinating diseases: on MRI, only 30% of the MOG group had brain abnormalities, mainly posterior reversible encephalopathy-like lesions, and none of the lesions were ovoid or perpendicular to the ventricles, and showed less spinal cord involvement than the AQP4-Abs group.

Recent studies have emphasised an association between MOG-Abs and ON, as we observed in our patients: Nakajima et al.³⁰ found eight patients positive for MOG-Abs in a group of 29 patients affected by idiopathic ON: they observed high intensities on T2 of the optic nerve in 7/8 patients, cerebellar high intensities in 2/8 patients and spinal cord lesions in 1/8 patients, but with no description of the type of brain lesions nor of the interested spinal cord segment. They concluded that myelitis was not frequent in this group of MOG-Abs-positive patients, and supposed instead that the optic nerve is the main target involved in the immune attack of these Abs, resulting in optic nerve swelling and optic pain. Tsuburaya et al.³¹ described the case of a seven-year-old Japanese boy with recurrent ON, who was found positive to MOG Abs, with a history of recurrent ON: his MRI at the first episode showed hyperintensity of the right optic nerve on STIR sequence, the MRI at the second episode showed a well-defined hyperintense plaque in the right frontal subcortical white matter on T2-weighted images and his MRI at the third episode showed diffuse hyperintense lesions in the right mesencephalic tegmentum. Adachi et al.³² described another case of a recurrent ON associated with positivity of MOG Abs in a 24-year-old Japanese woman, with a brain MRI showing T2 high-intensity areas in the left external and extreme capsule and precentral white matter with partial contrast enhancement, as well as pontine base, left cerebral peduncle, right medial thalamus T2 hyperintensities with contrast enhancement.

In a group of eight MOG-Abs-positive patients with recurrent ON,²⁰ brain and orbital MRI was normal except for optic nerve enhancement in two patients.

The presence of MOG Abs was also found in a case of a 37-year-old woman affected by multiphasic ADEM and ON;³³ her MRI showed swelling of the whole right optic nerve with strong gadolinium enhancement and hyperintense lesions in the right pallidum, the right posterior limb of the internal capsule and the left thalamus.

From the analysis of the data reported in the literature we can see that MOG-Abs are associated with different clinical syndromes and broad spectrum of radiological aspects, without consensus about pathognomonic imaging aspects related to these Abs.

Our series reflects the literature only for the variety of lesion morphology, and partially matched with the findings reported by Yan et al.¹⁵ for the prevalence of MS-like lesions, but it differs from the data published so far, since we did not observe the reported high frequency of longitudinally extensive transverse myelitis (up to 100% of patients)^14,15 or ADEM-like lesions,¹² nor the high frequency of lesions of the grey deep nuclei^12,14 or optic nerve swelling³⁰ and conus involvement in spinal MRI.^26,34 Moreover we observed brain lesions in 100% of our patients unlike other studies^13,14 in which brain alterations were present in the minority of patients.

As regards our case of the double Abs positivity, we know that there are different assays to analyse the presence of MOG-Abs in the serum³⁵ and Weinshenker and Wingerchuk³⁶ suggested that the difference in methodology can create discrepancy in the detection of these Abs; however, Ketzuka et al.³⁷ described first that six (26%) of 23 patients with ON, tested with ELISA, were seropositive for both NMO-Abs and MOG-Abs, then they confirmed the coexistence of MOG and AQP4-Abs in two patients with NMO tested with cell-based assay using full-length human MOG cDNA-expressing HEK cells.³⁸ Their results suggest that MOG and AQP4-Abs do exist in patients affected by NMO, regardless of the methodology of anti-MOG Abs detection. Moreover, Yan et al.,¹⁵ in their series of patients affected by NMOSD, identified 14 double positive to MOG and AQP4-Abs patients, detected by fluorescence-activated cell sorting (FACS) analysis, with clinical characteristics, imaging findings and outcome different from patients with only MOG-Abs positivity and from patients with AQP4-Abs positivity.

Concerning imaging follow-up, we found few studies describing this aspect. Kitley et al.¹² described follow-up MRIs at about seven months after the onset in six MOG-Abs-positive patients: 5/6 examinations showed complete resolution of all abnormalities, the only patient with residual signal change performed follow-up imaging at three months. By contrast, follow-up MRI in four AQP4-Abs-positive patients showed persistence of brain abnormalities.

In our group follow-up imaging was executed at three months and only 1/8 patient performed a second examination at six months; in no case did we observe complete resolution of MRI abnormalities. We need a further follow-up after a greater period of time to check eventual late resolution of the lesions.

Höftberger et al.¹⁶ described follow-up imaging of seven MOG-Abs-positive patients, showing complete resolution in five cases and reduction in two cases of the spinal lesions, without mentioning the timing; in our group, we observed stability/mild reduction of spinal lesions, but no case of complete resolution.

The case reported by Adachi et al.³² showed a reduction of brain abnormalities and disappearance of contrast enhancement in follow-up MRI performed after treatment, and this is like the behaviour of the lesions in one of our patient (Figures 9 and 10).

Regarding visual outcome of our patients, all of them have shown good response to steroid treatment with resolution of the visual deficit and this finding is in line with the outcome of the series of patients analysed by Kitely et al.¹¹ that showed excellent recovery, by Sato et al.¹³ with good recovery in 88% of cases, but in contrast with the results of Kezuka et al.³⁷ who observed no improvement of visual acuity. Sato et al.,¹³ Pröbstel et al.¹⁴ and Höftberger et al.¹⁶ noted that patients with MOG-Abs seem to have a better outcome than those with AQP4, with a residual minimal acute impairment.

A longer follow-up is needed to establish the potential of MOG-Abs as a prognostic biomarker. This is a preliminary study and our analysis is limited by the small group of patients and by the short follow-up, but up to now the published papers on MOG-Abs include restricted numbers of patients and case reports.

Conclusions

The clinical significance of MOG-Abs for diagnosis, treatment and prognosis has not been fully established. The radiological findings associated with these Abs are very different among patients and studies. We are only beginning to understand the diverse clinical and radiological manifestations of the disease. Further analyses and longer follow-ups are needed.

Footnotes

Funding

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

Conflict of interest

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

References

Vojdani

Cooper

. Antibodies to myelin basic protein, myelin oligodendrocytes peptides, alpha-beta-crystallin, lymphocyte activation and cytokine production in patients with multiple sclerosis. J Intern Med 2003; 254: 363–374.

Johns

Bernard

. The structure and function of myelin oligodendrocyte glycoprotein. J Neurochem 1999; 72: 1–9.

Brilot

Dale

Selter

et al.

Antibodies to native myelin oligodendrocyte glycoprotein in children with inflammatory demyelinating central nervous system disease. Ann Neurol 2009; 66: 833–842.

O'Connor

McLaughlin

De Jager

et al.

Self-antigen tetramers discriminate between myelin autoantibodies to native or denatured protein. Nat Med 2007; 13: 211–217.

McLaughlin

Chitnis

Newcombe

et al.

Age-dependent B cell autoimmunity to a myelin surface antigen in pediatric multiple sclerosis. J Immunol 2009; 183: 4067–4076.

Pröbstel

Dornmair

Bittner

et al.

Antibodies to MOG are transient in childhood acute disseminated encephalomyelitis. Neurology 2011; 77: 580–588.

Rostásy

Mader

Schanda

et al.

Anti-myelin oligodendrocyte glycoprotein antibodies in pediatric patients with optic neuritis. Arch Neurol 2012; 69: 752–756.

Rostásy

Mader

Hennes

et al.

Persisting myelin oligodendrocyte glycoprotein antibodies in aquaporin-4 antibody negative pediatric neuromyelitis optica. Mult Scler 2013; 19: 1052–1059.

Baumann

Sahin

Lechner

et al.

Clinical and neuroradiological differences of paediatric acute disseminating encephalomyelitis with and without antibodies to the myelin oligodendrocyte glycoprotein. J Neurol Neurosurg Psychiatry 2015; 86: 265–272.

10.

Hacohen

Absoud

Woodhall

et al.

Autoantibody biomarkers in childhood-acquired demyelinating syndromes: Results from a national surveillance cohort. J Neurol Neurosurg Psychiatry 2014; 85: 456–461.

11.

Kitley

Woodhall

Waters

et al.

Myelin-oligodendrocyte glycoprotein antibodies in adults with a neuromyelitis optica phenotype. Neurology 2012; 79: 1273–1277.

12.

Kitley

Waters

Woodhall

. Neuromyelitis optica spectrum disorders with aquaporin-4 and myelin-oligodendrocyte glycoprotein antibodies: A comparative study. JAMA Neurol 2014; 71: 276–283.

13.

Sato

Callegaro

Lana-Peixoto

et al.

Distinction between MOG antibody-positive and AQP4 antibody-positive NMO spectrum disorders. Neurology 2014; 82: 474–481.

14.

Pröbstel

Rudolf

Dornmair

. Anti-MOG antibodies are present in a subgroup of patients with a neuromyelitis optica phenotype. J Neuroinflammation 2015; 12: 46.

15.

Yan

et al.

Autoantibody to MOG suggests two distinct clinical subtypes of NMOSD. Sci China Life Sci 2016; 59: 1270–1281.

16.

Höftberger

Sepulveda

Armangue

et al.

Antibodies to MOG and AQP4 in adults with neuromyelitis optica and suspected limited forms of the disease. Mult Scler 2015; 21: 866–874.

17.

Mader

Gredler

Schanda

et al.

Complement activating antibodies to myelin oligodendrocyte glycoprotein in neuromyelitis optica and related disorders. J Neuroinflammation 2011; 8: 184.

18.

Ramanathan

Reddel

Henderson

et al.

Antibodies to myelin oligodendrocyte glycoprotein in bilateral and recurrent optic neuritis. Neurol Neuroimmunol Neuroinflamm 2014; 1: e40.

19.

Martinez-Hernandez

Sepulveda

Rostásy

et al.

Antibodies to aquaporin 4, myelin-oligodendrocyte glycoprotein, and the glycine receptor α1 subunit in patients with isolated optic neuritis. JAMA Neurol 2015; 72: 187–193.

20.

Chalmoukou

Alexopoulos

Akrivou

et al.

Anti-MOG antibodies are frequently associated with steroid-sensitive recurrent optic neuritis. Neurol Neuroimmunol Neuroinflamm 2015; 2: e131.

21.

Kim

Woodhall

Kim

et al.

Antibodies to MOG in adults with inflammatory demyelinating disease of the CNS. Neurol Neuroimmunol Neuroinflamm 2015; 2: e163.

22.

Spadaro

Gerdes

Krumbholz

et al.

Autoantibodies to MOG in a distinct subgroup of adult multiple sclerosis. Neurol Neuroimmunol Neuroinflamm 2016; 3: e257.

23.

Klawiter

Piccio

Lyons

. Elevated intrathecal myelin oligodendrocyte glycoprotein antibodies in multiple sclerosis. Arch Neurol 2010; 67: 1102–1108.

24.

Mantegazza

Cristaldini

Bernasconi

et al.

Anti-MOG autoantibodies in Italian multiple sclerosis patients: Specificity, sensitivity and clinical association. Int Immunol 2004; 16: 559–565.

25.

Markovic

Trajkovic

Drulovic

. Antibodies against myelin oligodendrocyte glycoprotein in the cerebrospinal fluid of multiple sclerosis patients. J Neurol Sci 2003; 211: 67–73.

26.

Ramanathan

Dale

Brilot

. Anti-MOG antibody: The history, clinical phenotype, and pathogenicity of a serum biomarker for demyelination. Autoimmun Rev 2016; 15: 307–324.

27.

Iglesias

Bauer

Litzenburger

et al.

T- and B-cell responses to myelin oligodendrocyte glycoprotein in experimental autoimmune encephalomyelitis and multiple sclerosis. Glia 2001; 36: 220–234.

28.

Baggi

Andreetta

Antozzi

et al.

Anti-titin and anti-ryanodine receptor antibodies in myasthenia gravis patients with thymoma. Ann NY Acad Sci 1998; 841: 538–541.

29.

Wingerchuck

Banweel

Bennett

et al.

International consensus diagnostic criteria for neuromyelitis optica spectrum disorders. Neurology 2015; 85: 177–189.

30.

Nakajima

Motomura

Tanaka

et al.

Antibodies to myelin oligodendrocyte glycoprotein in idiopathic optic neuritis. See comment in PubMed Commons belowBMJ Open 2015; 5: e007766.

31.

Tsuburaya

Miki

Tanaka

et al.

Anti-myelin oligodendrocyte glycoprotein (MOG) antibodies in a Japanese boy with recurrent optic neuritis. Brain Dev 2015; 37: 145–148.

32.

Adachi

Yasui

Takahashi

et al.

Anti-myelin oligodendrocyte glycoprotein antibodies in a patient with recurrent optic neuritis involving the cerebral white matter and brainstem. Intern Med 2016; 55: 1351–1354.

33.

Numa

Kasai

Kondo

et al.

An adult case of anti-myelin oligodendrocyte glycoprotein (MOG) antibody-associated multiphasic acute disseminated encephalomyelitis at 33-year intervals. Intern Med 2016; 55: 699–702.

34.

Pandit

. Neuromyelitis optica spectrum disorders: An update. Ann Indian Acad Neurol 2015; 18: S11–S15.

35.

Mayer

Meinl

. Glycoproteins as targets of autoantibodies in CNS inflammation: MOG and more. Ther Adv Neurol Disord 2012; 5: 147–159.

36.

Weinshenker

Wingerchuk

. The two faces of neuromyelitis optica. Neurology 2014; 82: 466–467.

37.

Kezuka

Usui

Yamakawa

et al.

Relationship between NMO-antibody and anti-MOG antibody in optic neuritis. J Neuroophthalmol 2012; 32: 107–110.

38.

Kezuka

Tanaka

Matsunaga

et al.

Distinction between MOG antibody-positive and AQP4 antibody-positive NMO spectrum disorders. Neurology 2014; 83: 475.