Myelin Oligodendrocyte Glycoprotein (MOG) Antibody-Associated CNS Demyelination: Clinical Spectrum and Comparison with Aquaporin-4 Antibody Positive Neuromyelitis Optica Spectrum Disorder
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.294831
Source of Support: None, Conflict of Interest: None
Keywords: Aquaporin-4, brainstem syndrome, CNS demyelination, Myelin oligodendrocyte glycoprotein, myelitis, neuromyelitis optica, optic neuritis
About one-third of aquaporin-4 (AQP-4) antibody negative neuromyelitis optica spectrum disorder (NMO-SD) patients are detected positive for antibody against myelin oligodendrocyte glycoprotein (MOG). MOG antibody was initially thought to be responsible for a benign homogeneous monophasic illness affecting optic nerves and spinal cord.,, Many reports followed suggesting other presentations such as pediatric acute disseminated encephalomyelitis (ADEM), acute brainstem syndrome, cortical encephalitis, fulminant course, relapses, and severe disability.,,, This led to an ambiguity about its clinical spectrum, neuroimaging features, and treatment outcomes, and much work needs to be done to bridge this knowledge gap. We conducted a study to characterize the clinical spectrum of MOG antibody disease and identify the features that differentiate it from AQP-4 antibody positive NMO.
This is a single-centre, observational study, and institutional ethics committee approval was obtained prior to enrolment of patients.
All patients from a single tertiary care center in Western India, who fulfilled following inclusion criteria, were included in the study
Patients were recruited from June 2017 to November 2019.
We screened all patients with the acute central nervous system demyelinating events including optic neuritis (ON), transverse myelitis, ADEM, other focal or polyfocal demyelination. We recruited those patients who tested positive for either MOG antibody or AQP-4 antibody and provided consent for participation and regularly followed them up at monthly intervals. Positivity for MOG and AQP-4 antibodies was determined by cell-based assay with visualization of binding to human embryonic kidney cells. Data collected using a standardized form included epidemiologic data (gender, age at onset, and follow up), clinical data (phenotype at onset and relapse, number of relapses, associated symptoms, disability at nadir, and best recovery assessed by the extended disability status scale [EDSS]), neuroimaging data (3T MRI of the brain, optic nerve, and spinal cord using the Siemens Magnetom Verio machine with T1W, T2W, FLAIR, and gadolinium-enhanced T1W axial, coronal, and sagittal sequences), biological data (serum MOG and AQP-4 antibody, antinuclear antibody [ANA], cerebrospinal fluid (CSF) protein, CSF pleocytosis (>5 cells/hpf, CSF oligoclonal bands, and IgG index), treatment details (acute therapy, rescue immunotherapy with plasma exchange [PLEX] or intravenous immunoglobulin [IVIg], and maintenance immunosuppression). Motor disability was defined as the inability to walk 100 m unaided for motor disability, and visual disability defined as acuity <6/36 on Snellen's chart.,
The clinico-radiologic features and treatment outcomes of both groups were compared. The SPSS package version-20 was used for data analysis. The significance of difference between percentage of patients in two groups was tested by applying Chi-square test. For comparing numerical parameters like “age at onset”, etc., nonparametric Mann-Whitney U-test was used. The significance level considered was P < 0.05.
Epidemiology and clinical presentation
A total of 48 patients, i.e., 21 with MOG antibody (44%) and 27 with AQP-4 antibody (56%) were included.
[Table 1] compares epidemiologic and clinical features of the two groups.
The two groups were similar in the age at onset, age distribution, patients with relapses, median episodes per patient, and disease duration; however, a nonsignificant trend was seen in MOG group for onset in the first decade and female predilection in AQP-4 group.
In the MOG group, disease most commonly began as isolated bilateral > unilateral ON, as seen in 16/21 (76%). Isolated brainstem syndrome was second most common (four patients), presentation with ataxia, dysarthria without hiccups or vomiting. Three patients presented as myelitis though six patients had abnormal spinal cord imaging. Out of the eight children included, two each had isolated ON and ON with myelitis. One had ON with asymptomatic cortical brain lesions and another had ON-ADEM-like presentation (bilateral vision loss, seizures, drowsiness, ataxia, dysphagia, and dysarthria). The disease commonly began with ON in females (10/11), but a non-ON phenotype was also observed in 4/10 males.
In contrast, the onset of disease in AQP-4 antibody group was frequently with myelitis 20/27 (74%), being isolated myelitis in 55% and myelitis in combination with other presentation in 19%. Only a single patient had asymptomatic cord involvement. Unilateral ON was more common in NMO group, whereas bilateral ON was seen more in MOG group. Area postrema involvement with intractable vomiting/hiccups was observed in all seven patients (26%) with onset as brainstem syndrome.
[Table 2] highlights the serological and CSF study results of the two groups.
Brain and spinal cord MRI were assessed for distribution and characteristics of lesions.
[Table 3] highlights the neuroimaging features of the two groups.
The MOG group had characteristic neuroimaging abnormalities, i.e., longitudinally extensive optic nerve involvement mainly in the anterior segment of optic nerve [Figure 1]a, [Figure 1]b, [Figure 1]c, [Figure 1]d, [Figure 1]e sometimes extending up to the chiasm in four patients [Figure 1]e, involvement of upper brainstem (pons, cerebellar peduncles, and midbrain) [Figure 2]a, [Figure 2]b, [Figure 2]c, [Figure 2]d, [Figure 2]e, [Figure 2]f, and thoracic cord extending at times to the conus medullaris [Figure 2]g, [Figure 2]h, [Figure 2]i. Multiple short segment spinal cord lesions were also seen in some patients, apart from the well-described longitudinally extensive transverse myelitis (LETM) lesions in a few patients. Also, asymptomatic large fluffy cortical lesions were found in three patients, while symptomatic cortical lesions in one patient with recurrent ADEM-ON [Figure 3]a, [Figure 3]b, [Figure 3]c, [Figure 3]d, [Figure 3]e, [Figure 3]f. Patients with ON had perineural optic nerve sheath enhancement [Figure 1]f.
[Figure 4]a, [Figure 4]b, [Figure 4]c, [Figure 4]d, [Figure 4]e, [Figure 4]f shows MRI findings of patient with recurrent ADEM-ON during first attack, and [Figure 5]a, [Figure 5]b, [Figure 5]c, [Figure 5]d, [Figure 5]e, [Figure 5]f shows MRI findings during the second attack.
In contrast, the AQP-4 antibody group had preferential involvement of chiasmal and posterior segments of optic nerve, of lower brainstem especially the medulla, and that of cervical and upper thoracic cord [Figure 6]a, [Figure 6]b, [Figure 6]c. Spinal cord involvement in NMO group was always found to be LETM. None of the patients had cortical lesions. Bright spotty lesions on T2W images and patchy cord enhancement were also seen.
Treatment and outcomes
[Table 4] highlights the treatment details and disability outcomes of the two groups.
All patients were treated in the acute phase with intravenous methylprednisolone, 1 g daily for 5 days. Rescue immunotherapy with IVIG/plasma exchange was required in significantly more patients in the AQP-4 group (41%, vs 14% in MOG group, P = 0.05).
In total, 43% of MOG group patients had one or more relapses. Out of a total of 31 relapses, 18 ON attacks were seen in 7 patients, and 13 brainstem syndrome relapses in 3 patients. All relapses occurred in patients who not receiving steroids, and three out of these were taking azathioprine and one was on mycophenolate mofetil.
Typically, 14/27 (52%) of AQP-4 positive patients had one or more relapses. Out of a total of 33 relapses, 16 were isolated LETM (48%), 15 isolated ON (45%), and two being simultaneous ON with myelitis (7%). Relapses in 10/14 (69%) were related to poor compliance to maintenance therapy, while four patients relapsed on adequate doses of azathioprine and had to be administered mycophenolate or Rituximab. Azathioprine was replaced by mycophenolate in two patients and rituximab in three patients due to relapses on treatment or side effects. Rituximab was required to be started as primary immunosuppressive therapy in four patients with severe attacks on presentation.
Greater improvement in the median EDSS scores between nadir and best recovery was observed in the MOG group as compared to the AQP-4 group. Lesser patients were found to permanently disabled in the MOG group than the NMO group, i.e., 3 versus 5 patients had permanent visual disability and 1 versus 6 had motor disability in the respective groups.
We found characteristic clinico-radiologic profile and treatment outcomes in MOG antibody disease, distinct from AQP-4 antibody disease. Only 19% patients with MOG antibody fulfilled 2015 NMO-SD criteria.
The median age at onset in our MOG cohort of 32.2 years was similar to previous large studies., The largest studies of MOG antibody disease reported no gender bias, similar to our study.,,
We encountered the wide spectrum of MOG antibody disease, with isolated monophasic or recurrent ON being the commonest phenotype followed by isolated monophasic or recurrent brainstem syndrome, simultaneous ON with myelitis, isolated myelitis, and recurrent ADEM-ON. However, the finding of bilateral simultaneous ON more than unilateral ON and brainstem syndrome more common than myelitis has not been reported in the larger studies. We found bilateral ON to be commoner in MOG antibody disease than AQP-4 antibody disease, though there are variable reports in previous studies., Our study is in line with a few reports of frequent relapsing course observed in longer follow up of MOG antibody disease., The spectrum of AQP-4 antibody disease seen in our study is consistent with previous studies.
We observed ON to be the commonest presentation in children. This is in contrast with the notion that ADEM is the commonest manifestation in children., We did see a child with recurrent ADEM, though he too had simultaneous ON.
Seizures were observed in two patients in the MOG group, but not in the AQP-4 group. Presentation with isolated cortical encephalitis and steroid responsive seizures has been recently reported in MOG antibody disease, but not in AQP-4 antibody disease.,,,
Associated paroxysmal tonic spasms and area postrema syndrome were seen in the AQP-4 group, but not in our MOG group. No patients in our study had double antibody positivity. ANA positivity was more frequent in the AQP-4 group than MOG group indicating a greater association of AQP-4 group with autoimmune diseases. These findings are in accordance with the literature.,,
The neuroimaging involvement of anterior optic pathways, upper brainstem and thoracic cord, cortical lesions in the MOG group as against the predilection for posterior optic pathways, medulla and cervico-thoracic spinal cord, and absence of cortical lesions in the AQP-4 group, is in line with earlier studies.,,,,, We found a significant 25% of MOG patients having chiasmal involvement with ON and encountered conus medullaris involvement only in two out of six patients with spinal cord lesions, which has been described as typical of MOG-associated myelitis.,
[Figure 7] highlights the characteristic pattern of neuroimaging involvement observed in the two groups in this study.
Our study supports the observation that MOG antibody disease generally has a milder course than AQP-4 disease but highlights the fact that it might not be entirely benign and therefore warrants early diagnosis, maintenance immunosuppression, and close follow-up.,,, Most relapses can be related to immunosuppression/steroid withdrawal.
Strengths and limitations
The major strength of our study is that it is one of largest studies of MOG antibody disease from India enrolling a wide spectrum of presentations and providing directly comparison with AQP-4 antibody disease. It highlights the features that could help the clinician differentiate these two look-alike diseases with varying courses, therapeutic response, and outcomes.
Late diagnosis in 8 of 21 patients (due to the lack of MOG antibody testing in earlier years), longer median disease duration, more severe cases might have biased the study to more relapses and disability in our MOG group. Suboptimal neuroimaging in some patients done prior to admission jeopardized adequate lesion characterization. The small sample size might have biased statistical analysis.
Our study highlights the important characteristics of patients with MOG antibody disease and its differences with AQP-4 positive NMO-SD patients.
MOG disease should be suspected in patients having atypical ON, brainstem syndrome, myelitis, cerebral syndrome, pediatric ADEM or their combinations. MOG-ON usually presents as bilateral, simultaneous, severe vision loss with papillitis, and anterior segment longitudinally extensive ON, sometimes with asymptomatic spinal cord lesions.
On the contrary, AQP-4 positive NMO-SD frequently presents first time as myelitis, area postrema syndrome or their combination with other known manifestations. On neuroimaging, they have posterior segment/chiasmal ON lesions and longitudinally extensive symptomatic lesions in cervical and upper thoracic spinal cord.
Whereas patients with AQP-4 antibody disease are known to have frequent relapses on immunosuppression thereby necessitating severe and prolonged immunosuppression at times, MOG antibody positive patients generally respond well to treatment. However, they can get frequent relapses and even permanent disability sometimes. Maintenance immunosuppression with steroids or steroid-sparing agents is warranted as relapses are usually related to withdrawal of immunosuppression. Larger studies will be necessary to determine the optimal immunosuppression and duration.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]
[Table 1], [Table 2], [Table 3], [Table 4]