Personal View

Diagnosis of myelin oligodendrocyte glycoprotein

Crowmark
antibody-associated disease: International MOGAD Panel
proposed criteria
Brenda Banwell”, Jeffrey L Bennett*, Romain Marignier*, Ho jin Kim", Fabienne Brilot, Eoin P Flanagan, Sudarshini Ramanathan, Patrick Waters,
Silvia Tenembaum, Jennifer S Graves, Tanuja Chitnis, Alexander U Brandt, Cheryl Hemingway, Rinze Neuteboom, Lekha Pandit, Markus Reind,
Albert Saiz, Douglas Kazutoshi Sato, Kevin Rostasy*, Friedemann Paul*, Sean Pittock*, Kazuo Fujihara, Jacqueline Palace*
Serum antibodies directed against myelin oligodendrocyte glycoprotein (MOG) are found in patients with acquired Lancet Neurol 2023
CNS demyelinating syndromes that are distinct from multiple sclerosis and aquaporin-4-seropositive neuromyelitis Published Online
optica spectrum disorder. Based on an extensive literature review and a structured consensus process, we propose Janvary 24,2023
https/doi.org/10.1016/
diagnostic criteria for MOG antibody-associated disease (MOGAD) in which the presence of MOG-IgG is a core 51474-4422(22)00431-8
criterion. According to our proposed criteria, MOGAD is typically associated with acute disseminated
See Online/Comment
encephalomyelitis, optic neuritis, or transverse myelitis, and is less commonly associated with cerebral cortical https:/doi.org/10.1016/
encephalitis, brainstem presentations, or cerebellar presentations. MOGAD can present as either a monophasic or $1474-4422(22)00520-8
relapsing disease course, and MOG-IgG cell-based assays are important for diagnostic accuracy. Diagnoses such as “Joint senior authors
multiple sclerosis need to be excluded, but not all patients with multiple sclerosis should undergo screening for Division of Child Neurology,
MOG-IgG. These proposed diagnostic criteria require validation but have the potential to improve identification of Children's Hospital of
individuals with MOGAD, which is essential to define long-term clinical outcomes, refine inclusion criteria for Philadelphia, Department of
clinical trials, and identify predictors of a relapsing versus a monophasic disease course. Neurology and Department of
Pediatrics, Perelman School
of Medicine, University of
Introduction epidemiology, and guide disease-specific research and Pennsylvania, PA, USA
The availability of cell-based assays for detection of clinical trials. Several manuscripts have proposed (ProfB Banwell MD);
serum autoantibodies directed against myelin oligo- recommendations for diagnosis of MOGAD"™ but Ophthalmology, Departments of Neurology and
Programs in
dendrocyte glycoprotein (MOG) has increased in the past formal, international consensus, diagnostic criteria have Neuroscience and
10 years.” Some adults and children who have optic not been formulated. We convened an international immunology, University of
neuritis and longitudinally extensive transverse myelitis panel of paediatric and adult neurologists, neuro- Colorado School of Medicine,
or who have been previously classified as having immunologists, and researchers to propose diagnostic Anschutz Medical Campus,
Aurora, €O, USA
aquaporin-4 (AQP4)-seronegative neuromyelitis optica criteria for MOGAD. These criteria are based on an (Prof) L Bennett MO Ph);
spectrum disorder (NMOSD) have now been identified extensive literature review of the clinical features and Service de nevrologie, sclérose
as having serum MOG-IgG. Patients with MOG-IgG outcomes reported in paediatric and adult individuals enmydline plaques, pathologies de la
et nevro-
present with isolated optic neuritis or transverse myelitis, with serum MOG-IgG, with careful consideration of inflammation, and Centre de
acute disseminated encephalomyelitis (ADEM), methods used to detect such antibodies and by use of a Référence des Maladies
brainstem or cerebellar features, or cerebral cortical structured consensus process (appendix p 16). Inflammatoires Rares du
encephalitis. Cerveau et de la Moelle, Hopital
Neurologique Pierre
Unlike multiple sclerosis and AQP4-IgG-seropositive Clinical features of patients with MOG-IgG Wertheimer, Hospices Civils
NMOSD, in which multiple clinical attacks characterise Overview deLyon, France
relapsing forms of disease, individuals with MOG The panel reviewed the clinical features of reported (Prof R Marignier MD PhD);
antibody-associated disease (MOGAD) can have either a cohorts of patients with CNS demyelination and Centrede Recherche en
Neurosciences de Lyon, Lyon,
monophasic or relapsing course.” Histopathological serological evidence of MOG-IgG. Figure 1 summarises France and Université Claude
features of MOGAD differ from those of multiple frequencies of features at disease onset across different BemardLyon, Lyon, France
sclerosis or NMOSD,* as do its imaging features, countries based on national studies including both (Prof R Marignier); Department
treatment responses, and outcomes. There is a clear paediatric and adult cohorts. Optic neuritis is by far the ofInstitute
Nevrology, Research
and Hospital of
need to establish formal consensus diagnostic criteria for most common onset feature, particularly among adults, National Cancer Center,
MOGAD as a distinct entity. while ADEM with or without concomitant optic nerve Goyang, South Korea
Studies from multiple countries support MOGAD as a involvement is the typical first manifestation in children, (ProfHJ Kim MO PhD); Brain
global disease affecting people of all ages. MOGAD particularly before the age of 11 years.™ Transverse Autoimmunity Group,
Kids Neuroscience Centre,
incidence is 1-6-3-4 per million people per year, and myelitis is another common presentation."*® Less Kids Research at the Children's
prevalence is estimated at 20 per million (95% CI common presentations include cerebral cortical encepha- Hospital at Westmead
11-34)7" These numbers are expected to rise with litis (often with seizures), brainstem and cerebellar (ProfF Brilot PhD); School of
increasing recognition and availability of testing,
Medical Sciences, Faculty of
demyelinating attacks, tumefactive brain lesions, cerebral Medicine and Health and Brain
including identification of patients with mild disease, monofocal and polyfocal CNS deficits associated with and Mind Centre, University of
monophasic disease, and atypical presentations. demyelinating lesions, cranial neuropathies, and pro- Sydney, Sydney, Australia
For multiple sclerosis” and NMOSD," international gressive white matter damage (leukodystrophy-like (ProfF Brilot); Departments of
criteria facilitate diagnosis, prognostication, and pattern). In contrast to AQP4-IgG-seropositive NMOSD Neurology, Laboratory

www thelancet com/neurology Published online January 24,2023 htps://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

and multiple sclerosis, no marked sex or racial chronic relapsing nature of AQP4-IgG-seropositive
predominance has emerged in populations with NMOSD and multiple sclerosis is that MOGAD can exist
MOGAD/% A further difference compared with the as either a monophasic illness or a relapsing disease.

A UKcohort (N=252) French cohort (N=366) Japanese cohort (N=259)

60 Median age 30-1years
(range 1-81) 60 Median age 29.9 years (IQR 167-417) 60 Median age 26 years (range 1-85)
55
50
5
4
Proportion of patients (%)

35
30
E

Figure 1: Clinical features

associated with MOG-1gG
Clinicalfeatures at onset or
during follow-up of patients
with serum MOG-1gG from
nationally-acquired cohorts
are presented geographically, SriLankan cohort (N=126) Dutch cohort (N=61) Australian cohort (N=287)
cumulatively, and by age 60 Median age 26 years (range 3-68) 60 Median age 17 years (IQR 8-32) 60 Median age 22 years (IR 8-43)
(paediatricor adult) >
() Clinical features at onset
(UK, French, Sri Lankan, Dutch,
and Australian cohorts) and
Proportion of patients (%)

over follow-up (Japanese

cohort). The category ADEM
or brain lesions (without
ADEM) includes allpatients
presenting with signs of brain
or brainstem involvement
(B) Clinical features at onset
from the combined UK,
French, ri Lankan, Dutch, and
Australian cohorts, and
divided into pacdiatric and
adult populations from studies
where the frequency of linical
features was reported
separately in these age groups
(French, Sri-Lankan, and Dutch
cohorts). Asimilar Children (N=176) Adults (N-377)
ibution of features at
onset has been reported from
moltiple European centres
Although truly national data
have not been reported from
China, a three-centre study of
Proportion of patients (%)

16 paediatric and 34 adult

patients with MOGAD also
reported optic neuritis as the
most frequent presenting
feature in adults (71%).*
Studies reporting paediatric-
only or adult-only data are
described in the main text.
ADEM-=acute disseminated
encephalomyelits.
LETM=Longitudinally
extensive transverse myelits.
MOG=myelin oligodendrocyte
glycoprotein. MOGAD=MOG
antibody associated disease.

wwwithelancet com/neurology Published online January 24,2023 https://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

Optic neuritis and optic nerve involvement imaging features of recurrent optic neuritis attacks are Medicine and Pathology and
Optic neuritis can be associated with central acuity loss, less well described. Center S and Autoimmune
retro-orbital pain (described by some patients as a Neurology, Mayo Clinic,
Rochester, MN, USA
headache),” colour vision loss, and an afferent pupillary Transverse myelitis and spinal cord involvement (Prof EP Flanagan MB BCh);
defect (which can be undetectable in people with bilateral Transverse myelitis in patients with MOG-IgG has clinical Department of Neurology,
or previous contralateral optic neuritis).” In patients with and imaging features that assist in differentiation from Concord Hospital, Translational
optic neuritis associated with serum MOG-IgG, optic multiple sclerosis and AQP4-IgG seropositive NMOSD Neuroimmunology
Kids Neuroscience Centre,
Group,
disc swelling is commonly visible on fundoscopy (table, figure 2, appendix p 5). Transverse myelitis in Children's Hospital at
(45-95%),4% with moderate to severe oedema frequently people with MOG-IgG can occur in isolation, as a Westmead, Sydney, Australia
reported.** Bilateral optic neuritis in adult and paediatric component of ADEM, or concurrent with optic (5 Ramanathan MD PhD); Brain
patients is common at onset and seems to be more
and Mind Centre and Sydney
neuritis.*** Clinical manifestations include sensory, Medical School, Faculty of
frequent in MOGAD (31-58%)** than in optic neuritis motor, and sphincter disturbance.””* The acute attack Medicine and Health,
associated with multiple sclerosis (less than 5%)” and severity varies, but is typically moderate to severe at nadir University of Sydney, Sydney,
AAQP4-IgG-seropositive optic neuritis (13-379%; appendix (Expanded Disability Status Scale score >4) in 50% or NSW, Australa (5 Ramanathan);
Nuffield Department of Clnical
Pp 2-3).2% Optic neuritis can occur in patients with more of patients with transverse myelitis and MOG-IgG Neurosciences, Universityof
MOG-IgG of all ages®*7 and relapses are often antibodies.” Most patients experience good to excellent Oxford, Oxford, UK
unilateral *** Optic neuritis associated with MOG-IgG motor recovery,”* but permanent bladder, bowel, or (PWaters PhD FRCPath);
also occurs in association with ADEM and transverse sexual dysfunction can occur.*” Recurrent transverse Paediatric Nevroimmunology
Clinic, Department of
myelitis. myelitis episodes without demyelination elsewhere in the Neurology, National Pacdi
Visual acuity loss, as measured using Snellen charts, is CNS are rare in patients with MOG-IgG antibodies.” Hospital Dr ) P Garrahan, Ciudad
often worse than 6/60 at nadir, although milder vision Painful tonic spasms and severe neuropathic pain as an de Buenos Aires, Argentina
loss can also occur.*** Improvement in visual acuity outcome is less common in patients with transverse (Prof SN Tenembaum MD);
Department of Neurosciences,
often occurs rapidly, with recovery to full or near normal myelitis with MOG-IgG, and is more representative of University of California,
acuity following acute corticosteroid ~therapy.**#»2 myelitis associated with AQP4-IgG-seropositive NMOSD. San Diego, CA, USA
Relapses can occur during corticosteroid weaning or Most patients with transverse myelitis associated with (5 GravesMD PhD MAS);
shortly after cessation.*»%2* Electrophysiology, visual MOG-IgG have T2-hyperintense lesions on spinal MRI, Department of Pediatric
Neurology, Massachusetts
perimetry, and optical coherence tomography (OCT) although up to 10% ofspinal MRI scans can be normal at General Hospital
have been used to quantify visual pathway damage and onset.” People with MOG-IgG and attacks involving the (Prof T Chitnis MD PhD);
dysfunction in optic neuritis associated with MOG-IgG, brain or optic nerves can have clinically silent spinal cord Department of Neurology,
but they are not diagnostically specific.” Despite similar lesions; conversely, clinically silent brain or optic nerve Brigham and Women's
Hospital, Harvard Medical
amounts of neuroaxonal injury, as measured by OCT, lesions can be detected in 33-50% patients with clinical School, Boston, MA, USA
patients with paediatric optic neuritis with MOG-IgG transverse myelitis and MOG-IgG.**** Acute transverse (ProfT Chitois); Department of
show better visual recovery than do adults.” myelitis in patients with MOG-IgG is often longitudinally Neurology, University of
Overall, relapsing optic neuritis occurs in 30-50% of extensive (three or more vertebral segments in length in California, Irvine, CA, USA
(AU Brandt MD); Department of
patients with MOG-IgG. The condition occurs commonly more than 60% of patients) on MRI, but shorter lesions Paediatric Neurology,
(but not exclusively) with three clinical scenarios: also occur, and some people have multiple spinal cord Great Ormond Street
(1) paediatric patients who present initially with ADEM lesions.“*** By contrast, a first attack of longitudinally Hospital, London, UK
followed by recurrent optic neuritis; (2) adult or paediatric extensive transverse myelitis rarely occurs in multiple (Prof C Hemingway MBChB Pho);
Institute of Neurology, UCL,
patients who initially present with optic neuritis and sclerosis, and if multiple sclerosis is suspected, care London, UK (Prof C Hemingway);
continue with further episodes of optic neuritis; and (3) should be taken to establish whether the appearance of a Department of Neurology,
adult and paediatric patients who present with NMOSD- long lesion might represent coalescence of focal lesions.* MS Center ErasMS, Sophia
like clinical features of concurrent optic neuritis and In a study of more than 1000 patients with demyelination, Children's Hospital, Erasmus
MCUniversity Medical Center
transverse myelitis at onset and then have relapses of only 11 (1-3%) of 863 patients with multiple sclerosis had Rotterdam, Rotterdam,
optic neuritis,**#0#22¢2 involvement of the conus, compared with nine (6%) of Netherlands
Typical funduscopic and MRI features ofoptic neuritis 150 patients with AQP4 antibodies and seven (26%) of (R Neuteboom MD); Center for
associated with MOG antibodies are shown in figure 2.
Advanced Neurological
27 patients with MOG antibodies.” Thickening and Research, Nitte University
Dedicated orbital fat-saturated images of the optic nerves contrast enhancement of the dorsal nerve roots has been Mangalore, Mangalore, India
with and without gadolinium are strongly encouraged to described in people with transverse myelitis and (Prof L Pandit MD);
confirm the presence of optic nerve inflammation. We MOG-IgG.“* Department of Neurology,
have outlined the key features that we determined best
Medical University of
Most acute T2-hyperintense lesions in the spinal cord Innsbruck, Innsbruck, Austria
differentiate optic neuritis associated with MOG are centrally located on axial imaging (in 66-75% of (Prof M Reindl PhD);
antibodies, multiple sclerosis, and AQP4-1gG seropositive patients with MOG-IgG), and can be restricted to the Neuroimmunology and
NMOSD (table, appendix p 4). The presence or absence grey matter (as seen in 30-50% of patients), producing Multiple Sclrosis Unit, Service
of Nevrology, Hospital Clnic,
of optic nerve head swelling, lesion extent along the optic the H-sign.””* However, 20-25% of spinal cord lesions Institut dInvestigacions
nerves, and involvement of perineural tissue are in people with MOG-IgG do not involve spinal grey Biomédiques August Pi Sunyer
particularly important features. Most studies have matter.” Contrast enhancement is seen in approximately (ProfA Saiz MD); Facultat de
focused on the incident optic neuritis attack, and thus 509% of patients with transverse myelitis and MOG-IgG, Medicinai Ciencies de la Salut,

www thelancet com/neurology Published online January 24,2023 htps://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

Optic nerve Spinal cord

Figure 2: Neuroimaging features of MOGAD

MRI scans showing the features emphasised as being either common or uniqueto patients with MOGADas delineated in figure 3. (A) Perineural optic sheath enhancement (with diffuse orbital fat
involvement and optic nerve enhancement). (8) Right optic nerve sweling and enhancement (coronal view). (€) Bilateral longitudinally extensive optic nerve T2 hyperintensity. (D) Radiologically
visible optic disc sweling. (E) Optic disc oedemaon fundoscopy. (F) Optic disc oedemainthe fellow eye of image (). (G) Longitudinally extensive T2-hyperintense lesion in the thoracic spine.
(H) Centralspinal cord involvement with H sign. () Conus lesion. (J)T2-hyperintense
pontine lesion. () Bilateral T2-hyperintense lesions of the middle cerebellar peduncles. (L) Bilateral T2-
hyperintense cerebral lesions involving the thalami. (M) Large il-defined T2-hyperintense lesions involving supratentorial white matter. (N) Corticalfluid attenuated inversion recovery hyperintensity
with (O)associated leptomeningeal enhancement. , F were reproduced from Ramanathan et al,* by permission of Elsevier. MOGAD=myelin oligodendrocyte glycoprotein antibody-associated discase.
Universitat de Barcelona, and cauda equina and pial enhancement have been and periventricular white matter) and persistent
Barcelona, Spain (Prof A Saiz); reported.“* Most T2 lesions in the spinal cord resolve Tl-hypointense lesions are uncommon in patients with
School of Medicine and or reduce in size substantially at follow-up.** Spinal MOG-IgG.* In patients with MOG-IgG, the pons is
Institute for Geriatrics and
Gerontology, Pontifical cord atrophy can occur in severe cases.”** In contrast to frequently involved; large lesions in the middle cerebellar
Catholic University of multiple sclerosis, accumulation of silent spinal cord peduncle, when present, suggest MOG-IgG-associated
Rio Grande do Sul, lesions between clinical attacks is very rare in MOGAD demyelination, given that such lesions are rare in
Porto Alegre, Brazil (0% of 110 follow-up spinal MRI scans in 81 patients). multiple sclerosis or AQP4-IgG-seropositive NMOSD.#*
(Prof DK Sato MD PhD);
Department of Paediatric Area postrema and midbrain involvement can associate
Neurology, Children'sHospital Brain and brainstem involvement with episodic but persistent (>48 h) nausea and vomiting
Datteln, University Wittenand Involvement of the brain or brainstem in patients with in a small proportion of patients with brainstem lesions
Herdecke, Datteln, Germany
(Prof K Rostdsy MD); MOG-IgG manifests as features of ADEM, cerebral and MOG-IgG."* The appearance of brainstem lesions
Experimental and Clinical cortical encephalitis, brainstem symptoms, or cerebellar does not reliably distinguish patients with MOG-IgG
Research Center, Max Delbrueck symptoms, or as clinically silent brain or brainstem from those with AQP4IgG-seropositive NMOSD.**
Center for Molecular Medicine lesions in patients with clinical optic neuritis or Tumefactive lesions can lead to life-threatening
and Charité—
Universitatsmedizin Berlin, transverse myelitis (table, figure 2, appendix pp 6-7). subfalcine and tentorial herniation.”
corporate membero Freie Brain MRI is normal at the time of a first attack of optic ADEM is the most prevalent clinical syndrome at
Universitat Berlin, Humboldt- neuritis or transverse myelitis in only 8-16% of people presentation in children with MOG-IgG. MOG-IgG are
Universitit zu Berlin, and subsequently diagnosed with multiple sclerosis,” but detected in around 50% of paediatric patients with
Berlin Institute of Health,
Berlin, Germany. brain T2-hyperintense lesions are absent in 47-68%" of ADEM, but the frequency of MOG-IgG is lower in adult
(ProfF Paul MD); Departments patients who have MOG-IgG (typically patients with ADEM patients.*** ADEM was the presenting syndrome
of Neurology, Laboratory optic neuritis or transverse myelitis). in only 15 (5-6%) of 268 individuals with adult-onset
Medicine, and Pathology and Brain lesions in patients with MOG-IgG tend to be demyelination with MOG-IgG." Children with MOG-
Center S and Autoimmune
Neurology, Mayo Clinc, bilateral, ill-defined, and large, often with deep grey IgG and ADEM are typically younger than 10 years.
Rochester, MN, USA matter involvement. Classic multiple sclerosis lesions Patients with MOG-IgG associated with optic neuritis,
(Prof 5 Pittock MD); (small, ovoid T2-hyperintense lesions in the juxtacortical transverse myelitis, or brain lesions (without meeting the

v thelancet com/neurology Published online January 24, 2023 https://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

MOGAD AQP4-1gG seropositive NMOSD Multiple sclerosis

Paediatric onset Frequent Extremely rare Infrequent
Sexdistribution = M F>M (after puberty)
Disease course Monophasic orrelapsing Most often relapsing Relapsing, secondary progressive, or progressive
from onset (adults only)
Opticnerve
Initial visual acuity Often severely impaired Often severely impaired Mild to moderately impaired
Initial recovery Typically favourable Risk for poor recovery Typically favourable
MR lesion characteristics Frequently bilateral and anterior at onset, Bilateral or unilateral at onset, often posterior, and Typically unilateral, anterior, short optic nerve lesions
longitudinally extensive", and involving the frequently longitudinally extensive’; chiasmal and that do ot involve the optic nerve sheath
optic nerve sheath optic tract involvement might be present
Optic nerve head Moderate to severe oedema s typical and can Oedema and associated haemorrhages are less Mild oedema can occur but severe oedema with
be associated with haemorrhage common than in MOGAD haemorthage s rarely seen
Spinal cord
Initial deficits Severe, Severe Mild to moderate
Motor function Excellent motor recovery after reatment Risk for poor recovery or worsening motor Often good but sk for motor impairment during
impairment with relapses progressive phase of disease
Sphincter, bladder, and erectile Risk for residual sphincter and erectile Variable residual bladder impairment Risk of bladder impairment during progressive phase
function impairment despite good motor recovery ofdisease
Spinal cord MRI lesion Single or multiple longitudinally extensive Single longitudinally extensive lesion, which Often multiple focalcord lesions; often posterior and.
characteristics lesions, grey matter involvement leadingto commonly involves entire transverse diameter of involving only a portion of the cross-sectional area of
the H-sign and conus lesions are characteristic the cord and might have bright spotty lesion the cord; conus rarely involved
appearance; conus raely involved
Brain
Clinical presentation Encephalopathy, seizures, focal deficits, and Area postrema symptoms, hiccups, Focal or polyfocal nevrological deficits common;
cerebral cortical encephalitis can occur hypersomnolence, or focal neurological defcits encephalopathy or seizuresare rare
Brain MRI Might be normal in optic neuritis or myelitis Might be normalin optic neuri ormyelitis Multifocal T2-hyperintense white matter esions
presentations presentations
Qualitative MR lesion features Fluffy or poorly demarcated T2 lesions; Multifocal T2 lesions most common in AQP4-rich Ovoid or round, well demarcated 2 lesions;
leukodystrophy-like pattern i rare regions; lesions can appear linear and along Dawson’s fingers, S-shaped or U-fibre lesions; central
corticospinal tract or medulla venule sign; smouldering or slowly evolving lesionst
Typical MRIlesion locations White matter, deep grey matter, middle Peri-third and peri-fourth ventricle, splenium of Periventricular and corpus callosum, juxtacortical,
cerebellar peduncle, large brainstem, and corpus callosum, internal capsule, and white cortical, white matter, and infratentorial
confluent cortical matter
MRI contrast enhancement Non-specific leptomeningeal around Patchy, cloud-like lesion enhancement pattern; Ovoid,ring, or open-ring lesion enhancement
pattern brainstem; unilateralor bilateral cortical penci-thin pattern of the ependymal surface of pattern
(linear) leptomeningeal enhancement (with lateral ventricles
cerebral cortical encephalitis)
Resolution
of T2-hyperintense Partial or complte resolution Might be present Complete resolution s infrequent
lesions on MRI
Silent MRI lesion accrual Infrequent Infrequent Frequent
Residual T1-hypointense esions Extremely rare Might be present Frequent
Presence of oligoclonal bands in Infrequent Infrequent Extremely frequent
CSFbut not serum
Frequencies were established through our literature review: extremely rare, <5%; infrequent, 5-20%; might be present, 21-50%; frequent, 51-80%; and extremely frequent, ~80%. A more detailed comparison of
optic neuritis (appendix p 4), transverse myelits (appendix p 5), and brain involvement (appendix pp 6-7) in patients with MOG-IgG, those with AQP4-1gG seropositive NMOSD, and those with multiple scerosis
isinthe appendix. NMOSD=neuromyeliti optica spectrum disorder, MOGAD- welin oligodendrocyte glycoprotein antibody-associated disease. F~female. M=male. “Longitudinally extensive lesions of the
optic nerve are defined as MRI signal (T1 gadolinjum-enhancing or short taw inversion recovery or T2) involving >50% ofthe length of the optic nerve. 1The limited available data indicatethat smovldering
lesions e absent n patients with MOGAD or AQP4-1gG seropositive NMOSD.
Table 1: Summary of the key features of MOGAD, AQP4-IgG NMOSD, and muttiple sclerosis

criteria of ADEM)® typically present in adolescence. Complete or almost complete clinical and radiological Department of Multiple
Children with ADEM and MOG-IgG are on average resolution occurs in more than 70% of children with Sclerosis Therapeutics,
2-3 years younger than children with seronegative MOGAD associated with ADEM.*“% However, the Fukushima Medical University
School of Medicine,
ADEM.““* An infectious episode (mainly respiratory) outcome can be poor in rare cases of children who Fokushima, Japan
and fever frequently precede ADEM, particularly in develop a leukodystrophy-like brain imaging pattern (ProfK Fujihara MD); Multiple
children with ADEM and MOG-IgG (40-75% of Relapses following ADEM occur more frequently in Sclerosis and Neuromyelitis
Optica Center, Southern
patients).”” Transverse myelitis and ataxia occur more patients who are MOG-IgG seropositive (about 38%) TOHOKU Research Institute for
commonly in patients with ADEM and MOG-IgG than in than in people who are seronegative (around 3%, and Neuroscience, Koriyama, Japan
patients with ADEM who are seronegative.® the time from onset to first relapse is variable (median (ProfK Fujihara); Department of

www thelancet com/neurology Published online January 24,2023 htps://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

Neurology John Radliffe 4.7 months, IQR 2-8-12.0, range 1-63) Relapsing experienced relapses, some within the first 10 years after
Hospital Oxford and Nuffield clinical attacks after ADEM include multiphasic onset, although others had long intervals from onset to
Department of Clnical disseminated encephalomyelitis (37-56%) and optic relapse (up to 46 years).*
Neurosciences Oxford
University, Oxford, UK neuritis (21-36%), and some patients have features A diagnostic hallmark for untreated relapsing multiple
(Prof ) Palace MD) meeting criteria for AQP4-IgG seronegative NMOSD ¢ sclerosis is the silent accrual of lesions over time,
Correspondence to: Cerebral encephalitis, manifesting with fever, headache, which also portends future clinical attacks. This diagnostic
Prof Brenda Banwell, Division of reduced consciousness, seizures, or status epilepticus, and prognostic hallmark might not be equally applicable
Child Neurology, Children's also occurs in 6-7% (19 of 285) patients with MOG-1gG.** to patients with a first demyelinating attack associated
Hospital of Philadelphia,
Deaprtment of Neurology, Cortical lesions in patients with MOG-IgG and seizures with MOG-IgG. In a study with serial acquisition of brain
Perelman School of Medicine are more apparent with fluid attenuated inversion recovery MRI scans, clinically silent brain lesions occurred in a
University of Pennsylvania, (FLAIR) sequences. Such lesions are termed as FLAMES: minority of children with MOG-IgG (14% of patients,
Philadelphia, PA, USA 19104 FLAIR hyperintense lesions in anti-MOG encephalitis 49 of all brain MRI scans). 44% of silent lesions were
[email protected]
See Online for appendix with seizures. Seizures can be focal or generalised. detected within 3 months of the first attack and 66% were
Symptoms of raised intracranial pressure can also occur detected within the first year with a low positive predictive
and can be life-threatening.” Seizures might be the first validity (20%) for clinically relapsing disease” In a study
manifestation, with more typical demyelinating presen- of 182 paediatric and adult patients with MOG-IgG imaged
tations later in the disease course.” In addition to at various timepoints from onset of symptoms, accrual of
evaluation of serum MOG-IgG, testing for other neuroglial silent brain lesions was also rare (4-1% of patients, 3- 6% of
cell surface antibodies such as NMDA receptor antibodies brain MRI scans at follow-up). However, in this study, the
(in CSF and serum) is also advised in children with presence of such lesions associated strongly with
features of autoimmune encephalitis, given that patients subsequent relapse.* In patients with first attacks of
can have a dual antibody seropositivity either contem- demyelination and MOG-IgG, the contribution over the
poraneously or sequentially (4-7-5% of patients with first few years of either new MRI lesion formation or
NMDA receptor encephalitis). reduction in resolution of lesions to the risk of clinical
relapses requires further study. The panel elected not to
Relapses in patients with MOG-lgG diagnose relapsing MOGAD on the basis of MRI alone
The panel defined a relapse as being a new clinical attack and to restrict this term to patients with clinically relapsing
occurring more than 30 days following onset of a previous disease. In contrast to patients with multiple sclerosis,
attack. Relapses are more common in the first 6 months progressive disease, with worsening of neurological
than later after the first attack. Relapses can occur within deficits in the absence of clear new relapses, does not
2 months following oral corticosteroid therapy tapering or appear to occur in patients with demyelination associated
cessation.** Some patients have a cluster of early relapses with MOG-IgG, although this absence of relapse-free
whereas others have ongoing relapses beyond 12 months progression requires further study in cohorts with
after onset.”” extended observation. 7727
To remove the bias of overestimating the relapse risk
based on inclusion of patients tested because of their Laboratory examinations
relapses, some studies have focused on incident cohorts of MOG-IgG testing
patients confirmed to have MOG-IgG at the time of their Serum is preferred specimen type for MOG-IgG testing
first attack. Two studies that included mostly adults (panel 1). Clotting factors in plasma can interfere with
showed similar relapse risks: 16 (369%) of44 patients with a results (appendix pp 8-11, 16), while CSF testing is
median follow-up of 15-5 months and 37 (27%) of promising but requires further examination.******
139 patients with a median of 10-78 months.** Both The panel strongly endorses serum testing for patients
cohorts showed the relapse risk was greatest over the first with suspected MOGAD using cell-based assays that use
few months from the initial attack, but the follow-up full-length human MOG to detect MOG-IgG."***™ MOG-
duration was short. In two incident paediatric cohorts, 1gG are IgG1, and the panel recommends testing with 1gG
17-20% of the 200 patients confirmed to have MOG-IgG at Fc, an IgG1 secondary antibody, or an IgG (heavy and
the time of their first attack experienced relapsing disease light) secondary antibody if externally validated in-house
over amedian observation period of 1-7 years. The median assays are used.™” Fixed cell-based assays are a reasonable
time to first relapse was 11 months but some first relapses alternative when live cell-based assay testing is unavailable,
occurred several years after the incident attack.** In a UK with the caveat that sensitivity and specificity of fixed cell-
study of 183 patients with MOG-IgG (68 paediatric onset based assays are lower than cell-based assays.'™* However,
and 115 adult onset)” followed for a median of 244 months antibody titres are often not consistently provided by
(range 1-2-235-1 months), the 4-year risk of relapse testing centres (appendix p 8-11) and reproducibility
was 31-7% and the 8-year risk was 36-3%.” Longitudinal between testing centres has not been systematically
observation of 13 adults with demyelination and MOG- investigated for commercial cell-based assays.” ELISA is
1gG, none of whom was on chronic therapy and not all of not recommended for MOG-IgG measurement owing to
whom were evaluated from onset, showed that 8 (629) low sensitivity and specificity.**

v thelancet com/neurology Published online January 24, 2023 https://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

Serum MOG-IgG reports should, in the opinion of the

panel, at least include qualitative results (ie, negative, low Panel 1: MOG-IgG isotypes and methodological considerations
positive, and clear positive), and semi-quantitative results Live cell-based assays quantified by flow cytometry or microscopy are the preferred
(ie, titres, flow cytometry ratio, or visual scores). The methodsto detect myelin oligodendrocyte glycoprotein (MOG) IgG in clinical settings
panel proposes criteria for clear positive and low positive (appendix pp 8-11, 17-18)
fixed and live cell-based assay results (appendix pp 8-11, Live cell-based assays offer the highest specificity; centres with established laboratory
17-18). For live assays, we recommend that clear positives expertise most often use live cells expressing full length MOG, whereas commercial
are defined as at least two doubling dilutions above the assays use fixed cells expressing full-length MOG
assay cutoff, or above the assay-specific titre cutoff, or The panel recommends testing for MOG-IgG with either an IgG Fc or IgG1-specific
flow-cytometry ratio cutoff (appendix pp 17-18). Fixed secondary antibody; use of an IgG (heavy and light) secondary antibody with externally
assays are considered clear positive by titres greater than validated in-house assays is also acceptable!*”
or equal to 1:100. Fixed or live assay results are considered Allisotypes (IgG1, 9G2, IgG3, IgG4, IgM, and IgA) have been detected in patients with
to be low positives if in the low range of the individual serum MOG-IgGH =7 but further work on the clinical association of IgM, IgA, IgG2,
live assay (appendix pp 8-11, 17-18) or if titres are at least 19G3, and IgG4 subclasses is needed to assess their use in clinical diagnostics
110 and less than 1:100 for fixed cell-based assays. Given that laboratories might not clarify whether they use live or fixed cell-based
Quantitative results, titres, or flow cytometry ratios with assays, the panel strongly urges clinicians requesting tests to inquire about the
reference values should be included in diagnostic reports methods used for MOG-IgG detection
for delineation of clear positive versus low positive Reporting of results (ie, titres, flow cytometry ratio, binary positive vs negative result,
results. and test accreditations) varies according to geographical region (appendix pp 8-11,
The rationale for emphasis on clear positivity versus 17-18). To facilitate MOGAD diagnosis, the panel recommends that MOG-IgG test
low positivity was influenced by evidence that higher reports include qualitative results (ie negative, low positive, and clear positive);
titres are more reproducible than lower titres.” An depending on cell-based assays and regions (appendix pp 811, 17-18), quantitative
international, multicentre, blinded comparative study of results (ie, titres or flow cytometry ratio) with reference values should also be included
seven live cell-based assays from four centres showed
excellent interlaboratory agreement for clear positive
results among most centres.' Low positive samples were Patients have the highest likelihood of MOG-IgG
more frequently discordant. The positive predictive value seropositivity when tested at the time of incident clinical
for clinical features consistent with MOGAD increased attack. Ideally, testing should be done before the
with increasing titre of MOG-IgG, and thus clear positives administration of corticosteroids, immunoglobulins, or
will have a higher positive predictive value.”* apheresis, because these therapies can reduce serum
In three studies evaluating MOG-IgG titres in paired MOG-IgG detection, as observed when testing for AQP4-
serum and CSF samples, only 61-5% (eight of 13),” 1gG.* If initial serum MOG-IgG testing was negative but
61-3% (19 of 31),” and 42-1% (48 of 114)" of MOG-IgG obtained after administration of acute therapies, then
seropositive patients with clinical and MRI features of repeat testing is advised at least 3 months later (after a
MOGAD also had positive CSF samples. These studies period of time sufficient for the effects of apheresis, in
highlight the low sensitivity of MOG-IgG testing in CSF particular, to no longer be operative) or at the time of
alone. However, in these investigations, a small number relapse.”” MOG-IgG titres usually decline with time, but
of patients with demyelinating disorders (three of 80, can remain positive for years, or become seronegative with
13 of 262, and four of 118)* were positive for MOG-1gG or without immunotherapy. Seroconversion from negative
in the CSF alone. Of these patients, three (100%) of three, to positive is extremely uncommon in patients with a
ten (69%) of 13, and two (66%) of the three patients (of negative test result at onset.”**
the four with CSF MOG-IgG), for whom clinical Onset serum MOG-IgG titres do not have strong
information was available, showed clinical features prognostic value for recovery or relapse.*” Persistent
suggestive of MOGAD. Therefore, CSF testing for MOG- MOG-IgG seropositivity is associated with an increased
1gG can be used in selected circumstances to support the likelihood of having a relapse by a factor of 2-10,5¢#7%
diagnosis of MOGAD in MOG-IgG seronegative patients particularly when the MOG-IgG titre remains high.*
with supporting clinical and MRI features (figure 2). However, data to guide optimal timing of serial MOG-IgG
If detailed information (appendix pp 8-11) on a testing and its interpretation are scarce. A standardised
particular assay methods or titres is not provided, we definition of persistence is required and the prognostic
recommend confirmation with a second independent significance of persistent seropositivity requires pro-
test using, whenever possible, a live cell-based assay done spective study.
in a laboratory with expertise in measuring MOG-IgG
titres, ideally using serum obtained at initial presentation. General CSF testing
We acknowledge that the choice of laboratory is often CSF pleocytosis, with white cell counts more than
made at an institutional level and not all centres or 5 per pL, occurs in over 50% of patients with a first
clinicians can request testing at a different reference demyelinating attack and MOG-1gG. Up to 12% of such
laboratory owing to issues of cost, shipping, and access. patients have more than 100 white blood cells per

www thelancet com/neurology Published online January 24,2023 htps://doi.org/10.1016/51474-4422(22)00431-8

Personal View

high-power field.** CSF pleocytosis is more likely during CSF-restricted oligoclonal bands. Serum testing for
an attack than during remission and more frequent MOG-IgG rests on the clinical presentation and imaging
in patients who have ADEM or transverse myelitis than features. The panel unanimously agreed that serum
in patients who have optic neuritis.”* CSF protein is evaluation for MOG-IgG should not be a screening test
elevated in 30% of patients with a first demyelinating for all patients with incident demyelinating attacks
attack and MOG-IgG, and does not allow discrimination (panel 2).
of MOG-IgG-associated ~demyelination from other We have proposed diagnostic criteria for MOGAD
neuroinflammatory disorders.** (figure 3). Patients with one of the core clinical attack types
Intrathecally restricted CSF oligoclonal bands strongly and dear positive MOG-IgG test results in serum
favour a diagnosis of multiple sclerosis. However, ‘measured by fixed or live cell-based assay can be diagnosed
oligoclonal bands are detected in up to 20% of patients with MOGAD. Patients with low positive serum MOG-IgG
with MOG-IgG (although they can be transient)," and titres measured by fixed or live cell-based assay, patients
their presence does not exclude a diagnosis of MOG-IgG with serum results reported as positive on fixed cell-based
antibody-associated demyelination. Measles, rubella, and assay without titre, or seronegative patients with clear
varicella zoster CSF antibodies have been reported to be positive CSF MOG-IgG test results who present with one
absent in patients with MOG-IgG'™ but very common in of the core clinical attack types are also required to have at
patients with multiple sclerosis. However, testing for the least one of the supporting clinical or MRI features to be
measles, rubella, and varicella zoster reaction is not diagnosed with MOGAD. Patients should be diagnosed
universally available. with MOGAD only after other diagnoses that better explain
their features have been excluded.
Coexistence of other antibodies or autoimmune Given that our proposed diagnostic criteria require the
disorders presence of MOG-IgG, the sensitivity of MOG-IgG
Many laboratories run MOG-IgG and AQP4IgG assays testing cannot be assessed. It is therefore more relevant
concurrently. Dual positivity is very rare, and when it to consider the specificity of MOG-IgG testing and the
occurs, the AQP4IgG titres are nearly always high importance of selecting appropriate patients to maximise
whereas the MOG-IgG titres are low.” The AQP4-IgG the positive predictive value of the diagnostic criteria
finding is the key diagnostic result given that dual- (panel 2). Clear positive MOG-IgG titres are strongly
positive patients manifest with a clinical course associated with the clinical features proposed for
consistent with AQP4-IgG-seropositive NMOSD.” If MOGAD and distinguish such patients from those
testing for AQP4IgG and MOG-IgG is ordered manifesting with clinical features and a disease course
independently, a positive serum test for AQP4-IgG in a characteristic of AQP4IgG-seropositive NMOSD or
person with optic neuritis, transverse myelitis, or other multiple sclerosis.” Low titres of MOG-IgG are less
features consistent with NMOSD would strongly support discriminatory and are encountered in patients with
a diagnosis of AQP4IgG-seropositive NMOSD and multiple sclerosis, other neurological diseases, and
testing for MOG-IgG would not be indicated. Patients healthy individuals.*** Although the frequency of
presenting with features of NMOSD who are seronegative MOG-IgG seropositivity in patients with clinically
for AQP4-IgG should be tested for serum MOG-IgG. definite multiple sclerosis is between 0-3%
Rarely, patients manifest with clinical features of and 2:-5%,™7#%% yniversal testing for MOG-IgG of
demyelination associated with serum MOG-IgG followed patients suspected to have multiple sclerosis will still
or preceded by clinical anti-NMDA receptor encephalitis yield a high number of false positive results and is
(termed MOGAD and anti-NMDA receptor encephalitis strongly discouraged. Although clear positive MOG-1gG
overlap syndrome).” Such patients have a history of titres are rarely found in controls, a positive result can be
episodes of encephalitis or demyelination, or can have found in a person who does not have clinical features of
concurrent serum MOG-IgG and CSF NMDA receptor MOGAD (particularly if MOG-IgG is included in broader
antibodies and manifest with clinical features of anti- diagnostic panels). A positive MOG-IgG titre alone would
NMDA receptor encephalitis (with features including not meet our proposed criteria for MOGAD, and future
encephalopathy, psychosis, seizures, and dyskinesias) studies are required to establish whether such individuals
associated with white matter lesions and clinical features develop clinical MOGAD attacks at a later point (as has
of CNS demyelination. been found to occur in patients with AQP4-IgG-sero-
positive NMOSD whose serum revealed AQP4IgG
Considerations for the creation of MOGAD antibodies years before a clinical attack of optic neuritis
diagnostic criteria or myelitis).”
Adults and children with an incident attack of acquired The 2017 international McDonald criteria for multiple
CNS demyelination require neurological examination, sclerosis advise caution in the application of those criteria
neuroimaging (dedicated orbital imaging, spinal cord for children younger than 11 years, owing to the rarity of
MRI, and brain MRI), and laboratory investigations. multiple sclerosis in this age group and the concern that
Initial laboratory testing might include evaluation for diagnoses other than multiple sclerosis are more likely."

v thelancet com/neurology Published online January 24, 2023 https://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

Given that more than 50% of children aged younger than

11 years with acute CNS demyelination will have serum Panel 2: Patient selection and positive predictive values for MOG-IgG testing
MOG-IgG at presentation, testing these patients is Key points
advised. « Fundamental to the principles of myelin oligodendrocyte glycoprotein (MOG)-lgG
We acknowledge that some patients will present with testing is the selection, on the basis of clinical features, of individuals who are most
clinical and imaging features consistent with MOGAD appropriate to be tested to enhance positive predictive value (PPV) of the test—that is,
but will not have detectable MOG-IgG or will reside in the probability that individuals with a positive screening test truly have the disease:
world regions where reliable MOG-IgG testing is not PPV=(sensitivity x prevalence) /[ (sensitivity x prevalence) + (1-specificity)
available. Future studies will be needed to identify x (1-prevalence)], as emphasised previously. PPV depends on the prevalence of
whether such patients follow a similar disease course as MOG antibody-associated disease (MOGAD) in the population being tested but is
those who have MOGAD defined by the presence of useful tollustrate how the risk of a false positive MOG-IgG result can increase or
MOG-IgG. decrease depending on what population is tested.
Critical to the diagnosis of MOGAD is the exclusion of MOG-IgG testing is not recommended for screening ofall patients with CNS
a better diagnosis (appendix pp 12-15). Key red flags inflammatory demyelination. Caution is advised in interpretation of positive serum or
should prompt reconsideration of a diagnosis of MOGAD CSF MOG-IgG results in patients with clinical or radiological features atypical for
(panel 3). MOGAD, as false positives are more likely, particularly with low positive titres.

Limitations Caveats
We acknowledge several limitations of this work. + The frequency of MOG-IgG approaches 50% in children with optic neuritis or acute
Prospective data were prioritised, although much of the disseminated encephalomyelitis (particularly children younger than 11 years) 5*
data review relied on retrospective case series as these Screening these groups for MOG-IgG, where the pretest probability is high, would yield
studies provided key longitudinal observational data. The ahigh PPV and is recommended.
available literature is enriched by more severe and The frequency of MOG-IgG in adults with optic neuritis is around 5%.* Universal
relapsing patient populations. Radiology studies are screening of this group for MOG-IgG, where the pretest probability is moderate, would
heavily skewed towards imaging obtained during the yield an intermediate PPV, so caution is recommended, and care with interpretation of
incident attack and rarely include concurrent optic nerve, positive results is needed.
brain, and spinal cord imaging (mostly class II-IV + The frequency of MOG-IgG in adults with optic neuritis with severe optic disc oedema
evidence for diagnostic accuracy) or longitudinal is substantially higher (around 39%)"**** than in adults with retrobulbar optic
evaluation. Our panel work focused on diagnosic criteria, neuritis. Screening this group for MOG-IgG would yield a high PPVand is
rather than prognostic factors. Future studies are recommended. However, limiting preselection to patients with optic neuritis with disc
required to validate our proposed criteria, to evaluate oedema would exclude almost half the of the patients with optic neuritis who have
factors that predict relapses, and to better define MOG-IgG. Testing for MOG-IgG in patients with optic neuritis who also have
outcomes for people diagnosed with MOGAD. longitudinally extensive optic nerve lesions, bilateral simultaneous optic nerve
involvement, or perineuritic optic neuritis lesions on MRI will increase diagnostic yield
Conclusions and future directions without sacificing PPV, 242500152103
Foremost among key areas for future research will be the The frequency of MOG-IgG among adults with clinical, radiological, and CSF findings
validation of our proposed criteria in prospective diagnostic of multiple sclerosis is 0:3-2:5%.7°41% Thus, although fewer than 2.5% of
paediatric and adult cohorts of patients with acquired patients with multiple sclerosis would test positive, the absolute number of false-
CNS inflammatory demyelination. Our proposed criteria positive results that would result from routine screening for MOG-1gG in adults with
rest on the presence of MOG-IgG as a fundamental multiple sclerosis would be very high and is not recommended.*
inclusion criterion, accompanied by clinical presentations
identified as being associated with MOG-IgG. Our
proposed criteria are inclusive of paediatric and adult many years later. Better understanding of relapsing
patients, with the assertion that MOGAD is a single MOGAD is imperative, including identification of means
disease across the age spectrum, similar to what has to: predict at presentation the risk of relapses; establish
been shown epidemiologically and pathobiologically for the relevance of silent contrast-enhancing and non-
multiple sclerosis and AQP4-IgG-seropositive NMOSD. contrast-enhancing new lesions of the optic nerves,
Given the fundamental nature of MOG-IgG detection for brain, and spinal cord on disease course; and appreciate
MOGAD diagnosis, comparative studies of MOG-IgG implications for disease chronicity when considering
testing methods and development of international relapses in the first few months as compared with
standards for assay methods and reporting will also be persistent relapses over several years. Prospective studies
important. with predefined imaging timepoints rather than imaging
Our proposed criteria should enable identification of obtained only with clinical symptoms are required to
consistent MOGAD cohorts for longitudinal studies, assess subclinical disease activity. Prompt identification
which will be crucial to determine whether relapsing of patients with MOGAD destined for a chronic clinically
MOGAD is a lifelong disease and whether patients with relapsing disease course would expedite initiation of
an initial monophasic course are at risk for relapse chronic immunomodulatory therapy, whereas expedited

www thelancet com/neurology Published online January 24,2023 htps://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

Diagnosisof
(A) Core dinical demyelinating | Optic neuritis*
event Myelitis?
ADEM
Cerebral monofocal or polyfocal deficits§
Brainstem or cerebelar deficits]
Cerebral cortical encephalitis often with seizures]|
(B) Positive MOG-IgG test Cell-based assay:serumi | Clear positive* No additional supporting features required
Low positivet +AQP4-1gG seronegative AND
=1 supporting dinical or MRI feature
Positivewithout reportedtitre
Negative but CSF positivess
Supporting dinical or Optic neuritis ~Biateral simultaneous dinical involvement
MR features +Longtudinal optic nerve involvement(> 50% length of the optic nerve)
«Perineural optic sheath enhancement
~Optic disc oedema
Myelitis ~Longitudinally extensive myelitis
«Central cord lesion or Hrsign
+Conus lesion
Brain, brainstem, or erebral syndrome |+ Multiple ll-defined T2 hyperintense lesions in supratentorial and often
infiatentorial white matter
«Deep grey matterinvolvement
Il-defined T2-hyperintensity involving pons, middle cerebellar peduncle, or medulla
~Cortcallesion with or without lesional and overlying meningeal enhancement
() Exclusion of better diagnoses including multiple sclerosis{[sl

ute disseminated encephalomyelitis. AQP4=aquaporin-4. MOG=myelin ligodendrocyte glycoprotein. MOGAD=MOG antibody-associated disease.
*Optic neuritis s characterised by unilateralor bilateral reduced visual acuity that develops over hours to days and is often associated with retrobulbar orbtal pain that
is typically exacerbated with eye movement and accompanied by colour vision and visual field loss * Diagnosis of optic neurits can be supported by the presence ofa
T2-hyperintense signal in the optic nerve or chiasm, by enhancement of the optic nerve or chiasm with gadolinium, and by exclusion of cinical or radiographical
evidence of an alternative compressive, infitrative, or vascular process impacting the optic nerve or retina.” tMyeliis s typically characterised by acute disturbance in
motor, sensory, sphincter, or erectile function in various combinations referable to the spinal cord that develops over hours to days. ™ The diagnosis of transverse
myelitsis supported by MR spinal cord T2 hyperintensity with or without gadolinium enhancement, by CSF inflammation, and by exclusion of a compressive or
vasculardisruption of the spinal cord. Spinal MRI sagittal T2 lesions in patients with MOG-IqG frequently extend three or more vertebral segments and often involve
the conus and central grey matter (H-sign). +ADEM is defined by acute (worsening over hours to days) polyfocal neurologicaldeficits with encephalopathy (alteration
in level of consciousness, profound iritablity, not related to postictal state) and by MRI features of multifocal T2 bright lesions often involving cerebral white and grey
matter SCerebral monofocal or polyfocal deficits develop over hours to days and are referable to one or more T2-hyperintense lesions (which might or might not be
enhanced by gadolinium). T2-hyperintense lesions are often located in the middle cerebellar peduncle, peri-fourth ventrice,in supratentorial white matter, in
juxtacortical or cortical in locations, and in deep grey nudlei. Periventricular lesions are less common than for multiple scerosis. Brainstem or cerebellar clinical deficits
develop over hours to days and are associated with T2-hyperintense lesions i the brainstem or cerebellum, which might or might not enhance with gadolinium.**
IICerebral cortical encephalits with seizures s associated with T2-hyperintense signal n the cortex often ith enhancement of the overlying meninges in a patient with
acute or subacute new onset seizures and evidence of cerebraliritation (encephalopathy, confusion, headache, or focal neurologicaldeficits in addition to seizure).”
**Alive cell-based assay result by astandardised method thatis clear positive according to the individual assay cutoffs (appendix pp 8-11, 17-18) or a ixed cel-based
assay result with a itre 21:100. 1A live cll-based assay result by a standardised method that s alow positive according to the individual assay cutoffs (appendix
P 8-11,17-18) or afixed cell- based assay result with a titre 21:10 and <1:100. +1Serum testing is recommended for all patients being investigated for MOGAD. Testing
both serum and CSF s not recommended for routine evaluation. CSF testing might be valuable in patients with dlinical features suggestive of MOGAD but in whom
serum testing was negative, particularlyif confounded by apheresis or other therapeutic interventions. §5A positive CSF evaluation with a ixed o lve cell-based assay
by use of standardised methods. CSF contamination by blood should be viewed cautiously as a positive result in such samples could occur due to serum MOG-IgG.
IqExclusion of no better explanation requires the expertise of the clinician. As an illustrative example, a patient with optic neuritis, MRI eatures meeting 2017 criteria
for multiple sclerosis,” positive CSF oligoclonal bands, and low titre MOG-lgG wold be more appropriately diagnosed with multiple sclrosis * Conversely, patient
with bilateral optic neuriis associated with optic disc swelling and longitudinally extensive optic nerve involvement, with borderline CSF oligoclonal bands, with MRI
lesions that involve the areas included in the dissemination in space citeria for multiple slerosis but that are lldefined in character, and with clearly positive serum
MOG-IgGtitre, might technically meet 2017 criteria for multiple sclerosis but would be more appropriately diagnosed with MOGAD. Although most patients with
multiple sclerosis will not have positive serum MOG-1gG, and most MOGAD patients do not meet the 2017 McDonald citeria for multiple sclerosis, some patients wil
meet both diagnostic criteria and the finaldiagnosis requires expertise and careful observation over time. We have compared features of MOGAD, multiple slerosis,
and AQP4-IgG-seropositive NMOSD (appendix pp 4-7) and outlined other conditions to consider n the differential diagnosis of MOGAD (appendix pp 12-15).

identification of patients likely to experience monophasic also reduce the power to detect effective disease
illness would avoid unnecessary immunosuppression. suppression for patients with chronic relapsing disease.
Enrolment in clinical trials of patients with MOGAD Serum and CSF cytokines, especially interleukin 6
destined for monophasic disease would not only expose (IL-6), and other biomarkers such as myelin basic
such patients to unnecessary immunosuppression, but protein, microtubule associated protein, tau, glial

10 v thelancet com/neurology Published online January 24, 2023 https://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

Panel 3: Red flags against

a diagnosis of myelin Search strategy and selection criteria
oligodendrocyte glycoprotein antibody-associated disease Literature searches were done by panel members in PubMed, MEDLINE, and Embase.
« Progressive neurological impairment in the absence of Aniinitial search on MEDLINE with search terms “myelin oligodendrocyte glycoprotein”
attacks and “MOG” identified 688 articles published between Jan 1, 2010, and Jan 31, 2021.
+ Rapid worsening of clinical deficits from onsetto nadir Additional articles with focus on acute disseminated encephalomyelitis, optic neuritis,
‘within minutes to hours transverse myelitis, or myelitis were identified by the members of the individual working
+ Noimprovement following treatment with high-dose groups. From this list, we excluded: articles not in English, as we wanted all panel
corticosteroids for an acute attack members to be able to review all manuscripts; case reports and small cases series, unless
« MRIfindings of well circumscribed T2-hyperintense lesions they detailed rare but important data; and most review articles. Older, pivotal studies
in a pattern meeting dissemination in space criteria for identified in other papers during our literature search were also included. 378 unique
multiple sclerosis, especially when accompanied by CSF manuscripts were relevant: 204 articles identified by the serology working group, 111 by
oligoclonal bands and by the accrual over time of new the brain and brainstem working group, 125 by the optic neuritis working group, and 76
silent T2-hyperintense focal lesions and retention of most by the transverse myelitis working group (some references were reviewed by more than
previous T2-hyperintense lesions one group). We also did a literature review of papers published relating to MOGAD during
« Lesion contrast enhancement that persists for 6 months the period of our manuscript review. From Mar 1, 2022 to Oct 20, 2022 we identified
or more. several new review papers, which we decided not to cite in our current manuscript given
the depth of data we provide in our supplementary materials that address the same
points. We have not included isolated case reports and series that associate MOGAD with
fibrillary acidic protein, and neurofilament light-chain COVID-19, given that recent COVID-19 infection is not a contributory feature for
might provide insight into MOGAD pathogenesis and diagnostic purposes and the high rate of COVID-19 infection at the community level,
severity; further studies are needed.™™ Specifically, which renders determination of causality to MOGAD tenuous. We identified a few
given that elevated CSF IL-6 led to clinical trials and manuscripts discussing therapeutic approaches (no formal dlinical trials), which we also
subsequent approval of anti-IL-6 receptor therapies for do not feel should be included given that our manuscript does not address treatment.
AQP4IgG-seropositive NMOSD, confirmation of IL-6
elevation in the CSF of individuals with MOGAD has
the potential to inform future therapies. MOGAD is a studies might further inform our understanding of the
demyelinating disease, and thus elevations of myelin effect of MOGAD on the optic pathway and differences
basic protein, rather than glial acid fibrillary protein between MOGAD, multiple sclerosis, and AQP4-IgG-
(which is elevated in AQP4-IgG- seropositive NMOSD, seropositive NMOSD."”
an astrocytopathy) provides biological insight into In conclusion, we propose international expert
disease-relevant tissue damage, whereas neurofilament- consensus criteria for the diagnosis of MOGAD, with
light chain concentrations might provide a more general the intention that MOGAD will be identified as a
index of disease severity irrespective of the cause. distinct disease from AQP4IgG-seropositive NMOSD
MOGAD is conceptualised as a disorder driven by and multiple sclerosis. We advocate testing for MOG-
attack-specific clinical deficits and by the risk for 1gG in appropriate populations, and we caution against
stepwise (relapse-mediated) neurological impairments. the testing of patients with clinical and radiological
An important challenge is the ability to quantify features typical of multiple sclerosis. We hope that these
MOGAD-related neurological impairment. Permanent proposed criteria, when validated, will assist in
increase in disability might relate to relapse-mediated optimising the design of clinical trials evaluating
injury without recovery, as is seen in patients with emerging new therapies and in developing targeted
AQP4IgG-seropositive NMOSD. The progressive treatment and improved outcomes for patients with
neurological ~deterioration that drives worsening MOGAD.
disability independent of relapses in patients with Contributors
multiple sclerosis does not appear to be operative in BB, JP, KF, HJK, JLB, FP, KR, RM, and S]P were members of the
MOGAD. Patients with MOGAD do not seem to harbour steering group. BB, JP, and KF drafted panel meeting agendas. BB led
the smouldering lesions that are associated with
the Panel meetings. Panel members were all authors. BB drafied the
‘manuscript with input from all panel members. JLB, SR, AB, KR, ]G,
progressive neurological decline in multiple sclerosis.” and FP formed the optic neuritis working group and were responsible
Furthermore, pathological studies on animal models of for two supplementary tables (appendix p 2-4). JP, HJK, EPF, ST, RN
demyelination mediated by MOG-IgG highlight that and DKS formed the transverse myelitis working group and created a
MOG protein is not expressed by oligodendrocyte
supplementary table (appendix p 5). KF, RM, AS, LP, ST, BB, and CH
‘were members of the brain and brainstem working group and were
precursor cells, which might be relevant for responsible for a supplementary table (appendix p 6-7). FB, PW, S|P,
remyelination in patients with MOGAD." TC, and MR constituted the serology working group and created a
Tools that quantify visual impairment would be supplementary figure and table (appendix pp 8-11, 17-18) and panels 1
and 2. ST designed a supplementary table (appendix pp 12-15) with
valuable in MOGAD, given the predilection for the optic input from all panel members. All panel members participated in virtual
nerve. The opticospinal impairment scale might prove meetings and in the drafting, editing, and provision of final approval of
useful in this regard.” Prospective longitudinal OCT the manuscript, figures, tables, and supplementary materials.

www thelancet com/neurology Published online January 24,2023 htps://doi.org/10.1016/51474-4422(22)00431-8 1

Personal View

Declaration of interests Pharmaceuticals, Roche, Genentech, and Sanofi Genzyme. She has
BB has or will potentially receive financial compensation for received research support from the National Institutes of Health,
consultancy effort for Novartis, Roche, UCB, Horizon Therapeutics, National Multiple Sclerosis Society, US Department of Defense,
Biogen, and Immunic Therapeutics for advice on clinical trial design. EMD Serono, Guthy-Jackson Charitable Foundation, I-Mab Biopharma,
BB is funded by the National Multiple Sclerosis Society, National Mallinckrodt Pharmaceuticals, Novartis, Octave Bioscience, Roche
Institute of Health, and has been previously funded by the Canadian Genentech, The Sumaira Foundation, and Tiziana Life Sciences. AUB
Multiple Sclerosis Society. JLB has received research grants from has received grant support from Moore Foundation, Clinical and
Novartis, Mallinckrodt Pharmaceuticals, Alexion, and the National ‘Trnaslational Awards Program, German Federal Ministry of Education
Institutes of Health, license fees from a US patent (2014/0170140); and Research (BMBF). He has been named as inventor on several
consulting fees from Horizon Therapeutics, Alexion, BeiGene Chugai patents and patent applications describing multiple sclerosis serological
Pharmaceutical, Genentech, Genzyme, Mitsubishi-Tanabe Pharma, biomarkers, drug targets for remyelination therapy, marker less motor
Reistone Biopharma, Roche, and AbbVie; and serves on Data Safety function analysis, and retinal image analysis methods. He serves on the
Monitoring Boards for Roche, Genentech, and Clene Nanomedicine. Observational Study Monitoring Board for CAVS-MS. He is cofounder,
RM reports personal fees from Horizon Therapeutics, Alexion, Roche, board member, and currently serving as secretary treasurer of
and UCB and non-financial support from Horizon Therapeutics, Merck, IMSVISUAL. He is cofounder and holds stocks of technology startups
Biogen, and Roche, outside the submitted work. HJK received a grant Motognosis GmbH and Nocturne GmbH. Motognosis GmbH develops
from the National Research Foundation of Korea and research support and sells systems for assessing motor dysfunction in patients with
from Aprilbio and Eisai; received consultancy and speaker fees from neurological disorders. Nocturne GmbH offers analysis services for
Alexion, Aprilbio, Altos Biologics, Biogen, Celltrion, Daewoong, Eisai, retinal optical coherence tomography. Both companies’ products and
GC Pharma, HanAll BioPharma, Handok, Horizon Therapeutics services are relevant for neurology in general, but not specific to
(formerly Viela Bio), Kolon Life Science, Mdimune, Merck Serono, MOGAD. CH reports grant support from the Medical Research Council,
Mitsubishi Tanabe Pharma, Novartis, Roche, Sanofi Genzyme, Teva- Multiple Sclerosis Society, and Vasculitis UK. She serves as a consultant
Handok, and UCB; and is a co-editor for the Multiple Sclerosis Journal to Novartis, Biogen, Roche, UCB, Viela Bio, and Sanofi. She participated
and an associated editor for the Journal of Clinical Neurology. FB has on an independent data safety monitoring board for the Wellcome
received research funding from the National Health and Medical ‘Trust. RN reports grant support from Multiple Sclerosis Research
Research Council (Australia), Multiple Sclerosis Research Australia, Foundation (Netherlands). He has participated on a data safety
New South Wales Health, Novartis, and the University of Sydney ‘monitoring board or advisory board for EXCEL study
(Sydney, NSW, Australia). She has received speaker honoraria from (neurofibromatosis). He is board member of the Dutch Pediatric
Novartis, Biogen, Merck, and Limbic Neurology, and has been on Neurology Society. LP has received speaker honoraria and travel grants
advisory boards for Merck and Novartis. She works at the University of from Biogen, and has consulted for Biogen, Novartis, and Sanofi. Her
Sydney and at the Children's Hospital at Westmead, Westmead, New university holds a patent for her invention: Live cell based assay for
South Wales, Australia, which offers MOG-IgG testing. EPF has served. detection ofautoantibodies for NMOSD and related disorders (Indian
on advisory boards for Alexion, Genentech, and Horizon Therapeutics. patent mumber 202141055841). MR is supported by research grants from
He has received speaker honoraria from Pharmacy Times and royalties the Austrian Science Fund (FWF project P32699), the Austrian Research
from UpToDate for a topic on MOGAD. He was a site primary Promotion Agency, Euroimmun, and Roche, and consulting fees and
investigator in a randomised clinical trial on inebilizumab in advisory board from Roche (to institution). MR works at the Clinical
‘neuromyelitis optica spectrum disorder run by Horizon Therapeutics. Department of the Medical University of Innsbruck (Innsbruck,
He is principal investigator on an RO1on MOG-IgG disease. He works Austria), which offers diagnostic testing for MOG-IgG and other
at Mayo Clinic, Rochester, MN,USA, which offers commercial MOG- autoantibodies. AS received personal compensation for consulting,
1gG testing but he receives no royalties from such testing. SR has serving on a scientific advisory board, speaking activities with Merck,
received research funding from the National Health and Medical Sanofi, Biogen, Roche, TEVA Pharmaceuticals, Novartis, Alexion, and
Research Council (Australia), the Brain Foundation (Australia), the Janssen. DKS has received research support from National council for
Royal Australasian College of Physicians, and the University of Sydney. Scientific and Technological Development CNPq Brazil (425331/2016-4
She was supported by an National Health and Medicine Research and 308636/2019-8), Fundacao de Amparo Pesquisa do Estado do Rio
Council Neil Hamilton Fairley Early Career Fellowship (APP1141169) Grande do Sul (17/2551-0001391-3 and 21/2551-0000077-5), TEVA
and is currently supported by an NHMRC Emerging Leadership (EL2) Pharmaceuticals, Merck, Biogen, and Euroimmun AG; speaker
Investigator Grant (APP2008339). She serves as a consultant on an honoraria from Biogen, Novartis, Genzyme, TEVA Pharmaceuticals,
advisory board for UCB and Limbic Neurology, and has been an invited Merck, Roche, and Bayer; and participates in advisory boards for
speaker for Biogen, Limbic Neurology, and Excemed. PW has received Biogen, Roche, and Merck. KR has been an invited speaker for Merck
research grants from Euroimmun AG, Commonwealth Serum and serves as a consultant for an advisory board for Roche. FP has
Laboratories Behring and patent royalties for antibody testing received honoraria and research support from Alexion, Bayer, Biogen,
(W02010046716A1). He is the co-director of the Oxford Autoimmune Chugai, MerckSerono, Novartis, Genyzme, MedImmune, Shire, and
Neurology Diagnostic Laboratory (Oxford University, Oxford, UK) where Teva Pharmaceuticals, and serves on scientific advisory boards for
MOG-IgG1 autoantibodies are tested and both he and the University of Alexion, MedImmune, Novartis, and UCB. He has received funding
Oxford receive royalties (for antibody tests for LGI1and CASPR2, from Deutsche Forschungsgemeinschaft (DFG Exc 257),
'W02010046716A1). He has received honoraria or consulting fees from Bundesministerium fiir Bildung und Forschung (Competence Network
Biogen Idec, F Hoffmann La-Roche, Mereo BioPharma, Retrogenix, Multiple Sclerosis), Guthy-Jackson Charitable Foundation, EU
UBC, Euroimmun AG, University of British Columbia ,and Alexion; Framework Program 7, and National Multiple Sclerosis Society of the
and travel grants from the Guthy-Jackson Charitable Foundation. Work USA. He serves on the steering committee of the N-Momentum study
in the Oxford Autoimmune Neurology Diagnostic Laboratory is with inebilizumab (Horizon Therapeutics) and the OCTIMS Study
supported by the UK National Health Service Commissioning service (Novartis). He is an associatee editor with Neurology, Neuraimmunalogy,
for NMOSD. ST has received speaker and consulting fees from Biogen- and Neuroinflammation and academic editor with PloS One. SJP reports
Idec Argentina, Merck SA, Genzyme-Sanofi, Roche, Novartis Argentina, grants, personal fees, and non-financial support from Alexion
and Novartis Pharma. She serves as a consultant on an advisory board Pharmaceuticals; grants, personal fees, and non-financial support from
for Genentech-Roche. and Alexion Pharmaceuticals. ]SG has grant or MedImmune /Viela Bio; and personal fees for consulting from
contract research support from the National Multiple Sclerosis Society, Genentech, Roche, UCB, and Astellas. He has two patents issued
gen, and Octave Biosciences for work unrelated to the present (8889102; application 12-678350; Neuromyelitis Optica Autoantibodies
project. She serves on a steering committee for a trial supported by as a Marker for Neoplasia; and 9891219B2; application 12-57394;
Novartis. She has received speaker fees from Alexion and Bristol Myers Methods for Treating Neuromyelitis Optica [NMO] by Administration of
Sqiuibb, and served on an advisory board for Genentech. TC has Eculizumab to an individual that is Aquaporin-4 [AQP4}IgG
received compensation for consulting from Biogen, Novartis Autoantibody positive). SJP also has patents pending for IgGs to the

wwwithelancet com/neurology Published online January 24,2023 https://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

following proteins as biomarkers ofautoimmune neurological disorders: 1 ‘Thompson AJ, Banwell BL, Barkhof F, et al. Diagnosis of multiple
septin-5, kelch-like protein 11, GFAP, PDE10A, and MAP1B. He works at sclerosis: 2017 revisions of the McDonald criteria. Lancet Neurol 2018;
Mayo Clinic, which offers commercial MOG-IgG testing. He receives no 17:162-73.
royalties from the sale of tests done at the neuroimmunology Laboratory 2 Wingerchuk DM, Banwell B, Bennett JL, et al. International
at Mayo Clinic. KF serves as an advisor or on scientific advisory boards consensus diagnostic criteria for neuromyelitis optica spectrum
for Biogen, Mitsubishi Tanabe, Novartis, Chugai, Roche, Alexion, disorders. Neurology 2015; 85: 177-89.
VielaBio/Horizon Therapeutics, UCB, Merck Biopharma, Japan Tobacco B Lopez-Chiriboga AS, Majed M, Fryer J, et al. Association of
and Abbvie; has received funding for travel and speaker honoraria from MOG-IgG Serostatus With Relapse After Acute Disseminated
Biogen, Eisai, Mitsubishi Tanabe, Novartis, Chugai, Roche, Alexion, Encephalomyelitis and Proposed Diagnostic Criteria for MOG-IgG-
VielaBio, Teijin, Asahi Kasei Medical, Merck, and Takeda; and has Associated Disorders. JAMA Neurol 2018; 75: 1355-63.
received the Grants-in-Aid for Scientific Research from the Ministry of " Jarius S, Paul F, Aktas O, et al. MOG encephalomyelitis:
Education, Culture, Sports, Science and Technology of Japan and the international recommendations on diagnosis and antibody testing.
Grants-in-Aid for Scientific Research from the Ministry of Health, J Neuroinflammation 2018; 15: 134.
Welfare and Labor of Japan. JP has received consulting fees from Merck, 15 Marignier R, Hacohen Y, Cobo-Calvo A, et al. Myelin-oligodendrocyte
glycoprotein antibody-associated disease. Lancet Neurol 2021;
Novartis, Roche, Mitsubishi Tnabe Pharma , UCB, Alexion, Vitaccess, 20:762-72.
and Argenx, and MRI support from Medimmune, Merck, and Roche. 16 Cobo-Calvo A, Ruiz A, Rollot F, et al. Clinical features and risk of
She has received payment or honoraria for lectures, presentations, relapse in children and adults with myelin oligodendrocyte
speakers bureaus, manuscript writing, or educational events from glycoprotein antibody-associated disease. Ann Neurol 2020; 89: 3041
Merck, VielaBio, Roche and Alexion. She has participated on a data 7 Senanayake B, Jitprapaikulsan J, Aravinthan M, et al. Seroprevalence
safety monitoring board or advisory board for Novartis, Roche, Argenx, and dlinical phenotype of MOG-IgG-associated disorders in
UCB, Alexion, and Sanofi. She is member of the Charcot Foundation Sti Lanka. ] Neurol Neurosurg Psychiatry 2019; 90: 1381-83.
Board, Magnetic Resonance in Multiple Sclerosis steering committee, 18 Tea F, Lopez JA, Ramanathan S, et al. Characterization of the human
and National Health Service England Intravenous Immune Globulin ‘myelin oligodendrocyte glycoprotein antibody response in
Committee. She holds stock for AstraZeneca for a product that is not demyelination. Acta Neuropathol Commun 2019; 7: 145,
related to MOG-antibody associated disease. She has a patent for 19 Kaneko K, Sato DK, TakahashiT, et al. Clinical, MRI and laboratory
Diagnosing Multiple Sclerosis (application no PCT/GB2013/050285, features of myelin oligodendrocyte glycoprotein (MOG)-antibody-
final patent number 13704627.2-1408; client reference 4440; MS reference associated neurologic disease: a study of 259 cases. Mult Sclr J 2017;
P37347WO; patent published Sept 13, 2021) 23:100
Acknowledgments. 20 Jarius S, Ruprecht K, Kleiter I, et al. MOG-IgG in NMO and related
We thank the Guthy-Jackson Charitable Foundation for their support, disorders: a multicenter study of 50 patients. Part 2: epidemiology,
clinical presentation, radiological and laboratory features, treatment
‘The sponsor, Guthy-Jackson Charitable Foundation, had no role in the responses, and long-term outcome. J Neuroinflammation 2016;
content of the manuscript. Giulia Fadda, whose effort was supported 13: 280.
through an educational grant from the Guthy Jackson Foundation, was 2 Chen L, Chen C, Zhong X, et al. Different features between
instrumental in the reference search, organisation of references, and pediatric-onset and adult-onset patients who are seropositive for
creation of the figures, and contributed to the final manuscript content. MOG-IgG: a multicenter study in South China. ] Neuroimmunol
We gratefully acknowledge the administrative efforts of Mary Curtis and 2018; 321: 83-91.
the support provided for her time by the Center for MS and 2 Asseyer S, Hamblin |, Messina S, et al. Prodromal headache in
Autoimmune Neurology at the Mayo Clinic. MOG-antibody positive optic neuritis. Mult Scler Relat Disord 2020;
References 40101965
1 Reindl M, Schanda K, Woodhall M, et al. International multicenter 23 Bennett ]L, Costello F, Chen J] et al. Optic neuritis and autoimmune
examination of MOG antibody assays. Neurol Neuroimmunol
optic neuropathies: advances in diagnosis and treatment.
Lancet Neurol 2022; published online Sept 22. https:/doi.
Neuroinflamm 2020; 7: e674.
org/10.1016/S1474-4422(22)00187-9.
2 Reind] M, Waters P. Myelin oligodendrocyte glycoprotein antibodies 2 Ramanathan S, Mohammad S, Tantsis E, et al. Clinical course,
in neurological disease. Nat Rev Neurol 2019; 15: 89-102.
therapeutic responses and outcomes in relapsing MOG antibody-
3 Cobo-Calvo A, Ruiz A, Maillart E, et al. Clinical spectrum and
prognostic value of CNS MOG autoimmunity in adults: the associated demyelination. J Neurol Neurosurg Psychiatry 2018;
89:127-37.
MOGADOR study. Neurology 2018; 90: e1858-69. 2 Chen JJ, Flanagan EP, Jitprapaikulsan . et al. Myelin
4 Jurynczyk M, Messina S, Woodhall MR, et al. Clinical presentation oligodendrocyte glycoprotein antibody-positive optic neuritis: clinical
and prognosis in MOG-antibody disease: a UK study. Brain 2017; characteristics, radiologic clues, and outcome. Am ] Ophthalmol
140: 3128-38. 2018; 195: 8-15.
S Waters P, Fadda G, Woodhall M, et al. Serial anti-myelin 2 Ramanathan S, Reddel SW, Henderson A, et al. Antibodies to myelin
oligodendrocyte glycoprotein antibody analyses and outcomes in oligodendrocyte glycoprotein in bilateral and recurrent optic neurits,
children with demyelinating syndromes. JAMA Neurol 2020; Neurol Neuroimmunol Neuroinflamm 2014; 1: e40.
6
77:82-93.
Armangue T, Olivé-Cirera G, Martinez-Hernandez E, et al 27 Rempe T, Tarhan B, Rodriguez E, et al. Anti-MOG associated
‘Associations of pacdiatric demyelinating and encephalitic disorder-linical and radiological characteristics compared to AQP4-
syndromes with myelin oligodendrocyte glycoprotein antibodics: IgG+ NMOSD-A single-center experience. Mult Scler Relat Disord
a multicentre observational study. Lancet Neurol 2020; 19: 234-46.
2021; 48: 102718,
7 Hofiberger R, Guo Y, Flanagan EP, et al. The pathology of central 2 Huppke P, Rostasy K, Karenfort M, et al. Acute disseminated
encephalomyelitis followed by recurrent or monophasic optic
nervous system inflammatory demyelinating disease accompanying neuritis in pediatric patients. Mult Scler 2013; 19: 94146,
myelin oligodendrocyte glycoprotein autoantibody. Acta Neuropathol 2 Lechner C, Baumann M, Hennes EM, et al. Antibodies to MOG and
2020; 139: 875-92.
AQP4 n children with neuromyelitis optica and limited forms of the
8 Takai Y, Misu T, Kaneko K, et al. Myelin oligodendrocyte disease. ] Neurol Neurosurg Psychiatry 2016; 87: 897-905.
glycoprotein antibody-associated disease: an immunopathological 30 Mariano R, Messina S, Kumar K, Kuker W, Leite MI, Palace ]
study. Brain 2020; 143: 1431-46.
Comparison of clinical outcomes of transverse myelitis among
9 de Mol CL, Wong Y, van Pelt ED, et al. The clinical spectrum and adults with myelin oligodendrocyte glycoprotein antibody vs
incidence of anti-MOG-associated acquired demyelinating aquaporin-4 antibody disease. JAMA Netw Open 2019; 2: €1912732.
syndromes in children and adults. Mult Scler 2020; 26: 806-14. 3 Havla ], Pakeerathan T, Schwake C, et al. Age-dependent favorable
10 O'Connell K, Hamilton-Shield A, Woodhall M,et al. Prevalence and visual recovery despite significant retinal atrophy in pediatric
incidence of neuromyelitis optica spectrum disorder, aquaporin-4 MOGAD: how much retina do you really need to see well>
antibody-positive NMOSD and MOG antibody-positive disease in J Neuroinflammation 2021; 18: 121
Oxfordshire, UK. J Neurol Neurosurg Psychiatry 2020; 91: 1126-28.

wwwthelancet.com/neurology Published online January 24,2023 https://doi.org/10.1016/51474-4422(22)00431-8 3

Personal View

32 Ramanathan S, Fraser C, Curnow SR, et al. Uveitis and optic 53 Sinha S, Banwell B, Tucker A, Storm PB, Huh J, Lang SS.
perineuritis in the context of myelin oligodendrocyte glycoprotein Hemicraniectomy and externalized ventricular drain placement
antibody seropositivity. Eur J Neurol 2019; 26: 1137-¢75. ina pediatric patient with myelin oligodendrocyte glycoprotein-
33 Dubey D, Pittock SJ, Krecke KN, et al. Clinical, radiologic, and associated tumefactive demyelinating disease. Childs Nerv Syst 2021;
prognostic features of myelitis associated with myelin 38:185-89.
oligodendrocyte glycoprotein autoantibody. JAMA Neurol 2019; 54 Hacohen Y, Absoud M, Deiva K, et al. Myelin oligodendrocyte
76:301-09. glycoprotein antibodies are associated with a non-MS course in
34 Ciron J, Cobo-Calvo A, Audoin B, et al. Frequency and children. Neurol Neuroimmunol Neuroinflamm 2015; 2: eS1
characteristics of short versus longitudinally extensive myelitis in 55 Krupp LB, Tardieu M, Amato MP, et al. International Pediatric
adults with MOG antibodies: a retrospective multicentric study. Multiple Sclerosis Study Group criteria for pediatric multiple
Mult Scler 2020; 26: 936-44. sclerosis and immune-mediated central nervous system
35 Mariano R, Messina S, Roca-Fernandez A, Leite MI, Kong Y, demyelinating disorders: revisions to the 2007 definitions. Mult Scler
Palace JA. Quantitative spinal cord MRI in MOG-antibody disease, 2013; 19: 1261-67.
neuromyelitis optica and multiple sclerosis. Brain 2020; 56 Cobo-Calvo A, Ruiz A, D'Indy H, et al. MOG antibody-related
144: 198-212. disorders: common features and uncommon presentations. J Neurol
36 Jitprapaikulsan ], Lopez Chiriboga AS, Flanagan EP, et al. Novel 2017; 264: 1945-55.
glial targets and recurrent longitudinally extensive transverse 57 Netravathi M, Holla VV, Nalini A, et al. Myelin oligodendrocyte
myelitis. JAMA Neurol 2018; 75: 892-95. glycoprotein-antibody-associated disorder: a new inflammatory CNS
37 Sechi E, Krecke KN, Pittock S], et al. Frequency and characteristics demyelinating disorder. ] Neurol 2021; 268: 1419-33.
of MRI-negative myelitis associated with MOG autoantibodies. 58 Baumann M, Sahin K, Lechner C, et al. Clinical and
Mult Scler 2020; 20: 1352458520907900. ‘neuroradiological differences ofpaediatric acute disseminating
38 Asnafi S, Morris PP, Sechi E, et al. The frequency oflongitudinally encephalomyelitis with and without antibodies to the myelin
extensive transverse myelitis in MS: a population-based study. oligodendrocyte glycoprotein. ] Neurol Neurosurg Psychiatry 2015;
Mult Scler Relat Disord 2020; 37: 101487. 86: 265-72.
39 Etemadifar M, Salari M, Kargaran PK, et al. Conus medullaris 59 Hacohen Y, Rossor T, Mankad K, et al. ‘Leukodystrophy-like'
involvement in demyelinating disorders of the CNS: a comparative ‘phenotype in children with myelin oligodendrocyte glycoprotein
study. Mult Scler Relat Disord 2021; 54: 103127. antibody-associated disease. Dev Med Child Neurol 2018;
Rinaldi S, Davies A, Fehmi ], et al. Overlapping central and 60: 417-23.
peripheral nervous system syndromes in MOG antibody-associated 60 Baumann M, Grams A, DjurdjevicT, et al. MRI of the first event in
disorders. Newrol Neuroimmunol Neuroinflamm 2020; 8: €924, ‘pediatric acquired demyelinating syndromes with antibodies to
@ Fadda G, Alves CA, 0'Mahony J, et al. Comparison of spinal cord ‘myelin oligodendrocyte glycoprotein. | Neurol 2018; 265: 845-55.
‘magnetic resonance imaging features among children with 61 Konuskan B, Yildirim M, Gocmen R, et al. Retrospective analysis of
acquired demyelinating syndromes. JAMA Netw Open 2021; children with myelin oligodendrocyte glycoprotein antibody-related
42128871, disorders. Mult Scler Relat Disord 2018; 26: 1-7.
42 ZhangBao |, Huang W, Zhou L, et al. Myelitis in inflammatory 62 ‘Wong YYM, Hacohen Y, Armangue T, et al. Paediatric acute
disorders associated with myelin oligodendrocyte glycoprotein disseminated encephalomyelits followed by optic neuritis: disease
antibody and aquaporin-4 antibody: a comparative study in Chinese course, treatment response and outcome. Eur | Neurol 2018;
Han patients. Eur | Neurol 2020; 28: 1308-15. 25: 782-86.
43 Sechi E, Krecke KN, Messina SA, et al. Comparison of MRI lesion 63 Hacohen Y, Wong YY, Lechner C, et al. Disease course and treatment
evolution in different central nervous system demyelinating responses in children with relapsing myelin oligodendrocyte
disorders. Neurology 2021; 97 €1097-109. glycoprotein antibody-associated disease. JAMA Neurol 2018;
Cobo-Calvo A, Sepiilveda M, Bernard-Valnet R, et al. Antibodies to 75: 478-87
myelin oligodendrocyte glycoprotein in aquaporin 4 antibody 64 Baumann M, Hennes EM, Schanda K, et al. Children with
seronegative longitudinally extensive transverse myelitis: clinical ‘multiphasic disseminated encephalomyelitis and antibodies to the
and prognostic implications. Mult Scler 2016; 2 : 312-19 myelin oligodendrocyte glycoprotein (MOG): extending the
45 Chien C, Scheel M, Schmitz-Hiibsch T, et al. Spinal cord lesions spectrum of MOG antibody positive diseases. Mult Scler 2016;
and atrophy in NMOSD with AQP4IgG and MOG-IgG associated 22:1821-29.
autoimmunity. Mult Scler 2019; 25: 1926-36. 65 ‘Wegener-Panzer A, Cleaveland R, Wendel EM, et al. Clinical
Camera V, Holm-Mercer L, Ali AAH, et al. Frequency of new silent and imaging features of children with autoimmune encephalitis and
MRI lesions in myelin oligodendrocyte glycoprotein antibody MOG antibodies. Newral Neuroimmunol Neuroinflanim 2020; 7: e731.
disease and aquaporin-4 antibody neuromyelitis optica spectrum 66 Hamid SHM, Whittam D, Saviour M, et al. Seizures and
disorder. JAMA Netw Open 2021; 4: €2137833. encephalitis in myelin oligodendrocyte glycoprotein IgG disease vs
47 Ramanathan S, Prelog K, Barnes EH, et al. Radiological aquaporin 4 1gG disease. JAMA Neurol 2018; 75: 65-71.
differentiation of optic neuritis with myelin oligodendrocyte 67 Ogawa R, Nakashima I, Takahashi T, et al. MOG antibody-positive,
glycoprotein antibodies, aquaporin-4 antibodies, and multiple benign, unilateral, cerebral cortical encephalitis with epilepsy:
sclerosis. Mult Scler 2016; 2: 470-82. Neurol Neuroimmunol Neuroinflamm 2017; 4: €322,
Jurynczyk M, Geraldes R, Probert F, et al. Distinct brain imaging 68 Mao L, Yang L, Kessi M, et al. Myelin oligodendrocyte glycoprotein
characteristics of autoantibody-mediated CNS conditions and (MOG) antibody diseases in children in central south China: clinical
‘multiple sclerosis. Brain 2017; 140: 617-27. features, treatments, influencing factors, and outcomes. Front Neurol
49 Banks SA, Morris PP, Chen JJ, et al. Brainstem and cerebellar 2019; 10: 868.
involvement in MOG-IgG-associated disorder versus aquaporin-4- 6 Ramanathan S, O’grady GL, Malone S, et al. Isolated seizures during
1gG and MS. ] Neurol Neurosurg Psychiatry 2020; published online the first episode ofrelapsing myelin oligodendrocyte glycoprotein
Dec 28, https://doi.org/10.1136/janp-2020-325121 antibody-associated demyelination in children. Dev Med Child Neurol
50 Jarius S, Kleiter I, Ruprecht K, et al. MOG-IgG in NMO and related 2019; 61: 610-14.
disorders: a multicenter study of 50 patients. Part 3: brainstem 70 Martinez-Hernandez E, Guasp M, Garcfa-Serra A, et al. Clinical
involvement-frequency, presentation and outcome. significance of anti- NMDAR concurrent with glial or neuronal
J Neuroinflammation 2016; 13: 251, surface antibodies. Neurology 2020; 94: €2302-10.
51 Hyun W, Kwon YN, Kim SM, et al. Value of area postrema 7 LopezChiriboga AS, Sechi E, Buciuc M, et al. Long-term outcomes
syndrome in differentiating adults with AQP4 vs MOG antibodies. in patients with myelin oligodendrocyte glycoprotein
Front Neurol 2020; 11: 396 immunoglobulin G-associated disorder. JAMA Neurol 2020;
52 Kunchok A, Krecke KN, Flanagan EP, et al. Does area postrema 77:1575-77.
syndrome occur in myelin oligodendrocyte glycoprotein-TgG- 7 Deschamps R, Pique J, Ayrignac X, et al. The long-term outcome of
associated disorders (MOGAD)? Neurology 2020; 94: 85-88 MOGAD: an observational national cohort study of 61 patients.
Eur| Neurol 2021; 28: 1659-64.

wwwithelancet com/neurology Published online January 24,2023 https://doi.org/10.1016/51474-4422(22)00431-8

 Personal View

73 Satukijchai C, Mariano R, Messina S, et al. Factors associated with 95 Cobo-Calvo A, Septilveda M, d'Indy H, etal. Usefulness of
relapse and treatment of myelin oligodendrocyte glycoprotein MOG-antibody titres at first episode to predict the future clinical
antibody-associated disease in the United Kingdom. course in adults. ] Neurol 2019; 266: 806-15.
JAMA Netw Open 2022; 5: €2142780. 96 Jarius S, Pellkofer H, Siebert N, et al. Cerebrospinal fluid findings in
74 ‘Akaishi T, Misu T, Fujihara K, et al. Relapse activity in the chronic patients with myelin oligodendrocyte glycoprotein (MOG) antibodies.
phase of anti-myelin-oligodendrocyte glycoprotein antibody- Part I: results from 163 lumbar punctures in 100 adult patients.
associated disease. | Neurol 2021; 269: 3136-46. ] Neuroinflammation 2020; 17: 261,
75 Fadda G, Banwell B, Waters P, et al. Silent new brain MRI lesions 97 Kunchok A, Chen JJ, McKeon A, Mills JR, Flanagan EP, Pittock S].
in children with MOG-antibody associated disease. Ann Neurol Coexistence of myelin oligodendrocyte glycoprotein and aquaporin4
2021; 89: 408-13. antibodies in adult and pediatric patients. JAMA Neurol 2020;
76 Akaishi T, Misu T, Takahashi T, et al. Progression pattern of 77:257-59.
neurological disability with respect to clinical attacks in 98 Fan'S,XuY, Ren H, etal. Comparison of myelin oligodendrocyte
anti-MOG antibody-associated disorders. J Neuraimmunol 2021; glycoprotein (MOG)-antibody disease and AQP4-IgG-positive
351: 577467. ‘neuromyelitis optica spectrum disorder (NMOSD) when they
Kim Y, Hyun JW, Woodhall MR, et al. Refining cell-based assay to co-exist with anti-NMDA (N-methyl-D-aspartate) receptor
detect MOG-1gG in patients with central nervous system encephalitis. Mult Scler Relat Disord 2018; 20: 144-52.
inflammatory diseases. Mult Scler Relat Disord 2020; 40: 101939. 99 Hassan MB, Ster C, Flanagan EP, et al. Population-based incidence
Gastaldi M, Scaranzin S, Jarius S, et al. Cell-based assays for the of optic neurits in the era of aquaporin-4 and myelin oligodendrocyte
detection of MOG antibodies: a comparative study. ] Neurol 2020; glycoprotein antibodies. Am J Ophthalmol 2020; 220: 110-14.
267: 3555-64. 100 Ducloyer JB, Caignard A, Aidaoui R, et al. MOG-Ab prevalence in
Jarius S, Ruprecht K, Kleiter I, et al. MOG-IgG in NMO and related optic neurits and clinical predictive factors for diagnosis.
disorders: a multicenter study of 50 patients. Part 1: frequency, Br ] Ophthalmal 2020; 104: 84245,
syndrome specificity, influence ofdisease activity, long term course, 101 Stiebel-Kalish H, Lotan I, Brody ], et al. Retinal nerve fiber layer may
association with AQP4-IgG, and origin. J Neuroinflammation 2016; be better preserved in MOG-IgG versus AQP4-1gG optic neuritis: a
13:279. cohort study. PloS One 2017; 12: e0170847.
Mariotto S, Ferrari S, Monaco S, et al. Clinical spectrum and IgG. 102 Yang M, Wu Y, Lai M, et al. Clinical predictive factors for diagnosis of
subelass analysis of anti-myelin oligodendrocyte glycoprotein MOG-IgG and AQP4IgG related paediatric optic neuritis: a Chinese
antibody-associated syndromes: a multicenter study. ] Neurol 2017; cohort study. Br ] Ophthalmol 2020; 106: 262-66.
264: 2420-30. 103 Zhao Y, Tan S, Chan TCY, et al. Clinical features of demyelinating
81 Pedreo M, Septilveda M, Armangué T, et al. Frequency and optic neuritis with seropositive myelin oligodendrocyte glycoprotein
relevance of IgM, and IgA antibodies against MOG in MOG-IgG- antibody in Chinese patients. Br ] Ophthalmol 2018; 102: 1372-77.
associated disease. Mult Scler Relat Disord 2019; 28: 230-34. 104 Jarius S, Ruprecht K, Stellmann JP, et al. MOG-IgG in primary and
82 Waters P, Woodhall M, O'Connor KC, et al. MOG cell-based assay secondaty chronic progressive multiple sclerosis: a multicenter study
detects non-MS patients with inflammatory neurologic disease. of 200 patients and review of the lterature. | Neuroinflammation 2018;
Neurol Neuroimmunol Neuroinflamm 2015; 2: e89. 15:88.
83 Akaishi T, Sato DK, Takahashi T, Nakashima I. Clinical spectrum of 105 Cobo-Calvo A, d'Indy H, Ruiz A, etal. Frequency of myelin
inflammatory central nervous system demyelinating disorders oligodendrocyte glycoprotein antibody in multiple sclerosis:
associated with antibodies against myelin oligodendrocyte a multicenter cross-sectional study. Newrol Neuroimmunol
glycoprotein. Neurochem Int 2019; 130: 104319. Neuroinflamm 2019; 7: 649,
‘Waters PJ, Komorowski L, Woodhall M, et al. A multicenter 106 Transverse Myelitis Consortium Working G. Proposed diagnostic
comparison of MOG-IgG cell-based assays. Neurology 2019; criteria and nosology of acute transverse myelitis. Neurology 2002;
92: €1250-55. 59: 499-505.
85 Cross H, Sabig F, Ackermans N, et al. Myelin oligodendrocyte 107 Leite MI, Coutinho E, Lana-Peixoto M, et al. Myasthenia gravis and
glycoprotein (MOG) antibody positive patients in a multi-ethnic ‘neuromyelitis optica spectrum disorder: a multicenter study of
Canadian chort. Front Neurol 2021; 11: 525933, 16 patients. Neurology 2012; 78: 1601-07.
Magcrini C, Gerhards R, Winklmeier §, et al. Features of MOG 108 Kaneko K, Sato DK, Nakashima I, etal. CSF cytokine profile in MOG-
required for recognition by patients with MOG antibody-associated IgG+ neurological disease is similar to AQP4-1gG+ NMOSD but
disorders. Brain 2021; 144: 2375-89. distinct from MS: a cross-sectional study and potential therapeutic
8 Sechi E, Buciuc M, Pittock S], et al. Positive predictive value of myelin implications. ] Neurol Neurosurg Psychiatry 2018; 89: 927-36.
oligodendrocyte glycoprotein autoantibody testing. JAMA Neurol 109 Kim H, Lee EJ, Kim S, etal. Serum biomarkers in myelin
2021; 78: 741-46. oligodendrocyte glycoprotein antibody-associated disease.
Held F, Kalluri SR, Berthele A, Klein AK, Reindl M, Hemmer B. Neurol Neuroimmunal Neuroinflanim 2020; 7: €708.
Frequency of myelin oligodendrocyte glycoprotein antibodies in a 110 Mariotto S, Ferrari S, Gastaldi M, etal. Neurofilament light chain
large cohort of neurological patients. Mult Scler ] Exp Transl Clin serum levels reflect disease severity in MOG-Ab associated disorders.
2021; 7: 20552173211022767. ] Neurol Neurosurg Psychiatry 2019; 90: 1293-96.
89 Mariotto S, Gajofatto A, Batzu L, et al. Relevance of antibodies to 11 Saxena S, Lokhande H, Gombolay G, Raheja R, Rooney T, ChitnisT.
myelin oligodendrocyte glycoprotein in CSF of seronegative cases. Identification of TNFAIP3 as relapse biomarker and potential
Neurology 2019; 93: €1867-72. therapeutic target for MOG antibody associated diseases. Si Rep
Kwon YN, Kim B, Kim JS, et al. Myelin oligodendrocyte glycoprotein- 2020; 10: 12405
immunoglobulin G in the CSF: clinical implication of testing and 112 Ciccarelli O, Cohen JA, Reingold SC, et al. Spinal cord involvement
association with disability. Neurol Neuroimmunol Neuroinflamm 2021; in multiple sclerosis and neuromyelitis optica spectrum disorders.
9: ¢1095 Lancet Neurol 2019; 18: 185-97.
91 Pace S, Orrell M, Woodhall M, et al. Frequency of MOG-IgG in 113 Graves J$, Oertel FC, Van der WaltA, et al. Leveraging visual
cerebrospinal fluid versus serum. | Newrol Neurosurg Psychiairy 2022; outcome measures to advance therapy development in
93:334-35 ‘neuroimmunologic disorders. Neurol Neuroimmunol Neuroinflamm
92 Jarius S, Aboul-Enein F, Waters P, et al. Antibody to aquaporin-4 in the 2021; 9: e1126.
Tong-term course of neuromyelitis optica. Brain 2008; 131: 3072-80. Copyright © 2023 Published by Elsevier Ltd. All rights reserved.
9 Hyun JW, Woodhall MR, Kim SH, et al. Longitudinal analysis of
myelin oligodendrocyte glycoprotein antibodies in CNS inflammatory
diseases. ] Neurol Neurosurg Psychiatry 2017; 88: 811-17.
94 Hennes EM, Baumann M, Schanda K, et al. Prognostic relevance of
MOG antibodies in children with an acquired demyelinating
syndrome. Neurology 2017; 89: 900-08.

wwwthelancet.com/neurology Published online January 24,2023 https://doi.org/10.1016/51474-4422(22)00431-8 15