MOG Antibody Disease (MOGAD)

Myelin loss around small veins in MOGAD
Myelin loss around small veins in MOGAD
Merged patches of demyelination
Merged patches of demyelination

Table of Contents

  1. Overview
  2. Distinguishing MOGAD from Multiple Sclerosis and NMOSD
  3. Clinical Manifestations
  4. Optic Neuritis in MOGAD
  5. ADEM in Children
  6. Serology and Diagnosis
  7. Magnetic Resonance Imaging Characteristics
  8. Treatment
  9. Prognosis and Monitoring
  10. Key Research Papers
  11. Connections
  12. Featured Videos

Overview

MOG antibody-associated disease (MOGAD) is a distinct autoimmune inflammatory demyelinating disease of the central nervous system (CNS). It is caused by antibodies targeting myelin oligodendrocyte glycoprotein (MOG) — a protein expressed on the outermost surface of the myelin sheath and on the membrane of oligodendrocytes, the cells that produce myelin in the central nervous system. When MOG-IgG antibodies bind to this exposed surface protein, they trigger complement-mediated and cell-mediated destruction of myelin, resulting in demyelinating attacks that can affect the optic nerves, spinal cord, and brain.

MOGAD is now recognized as a disease entity entirely separate from multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD). For decades, MOG antibodies were detected in patients who did not fit cleanly into the multiple sclerosis or NMOSD boxes — but the clinical picture remained blurry. The pivotal turning point came in 2016 when improved aquaporin-4 (AQP4) antibody testing revealed that a substantial subset of patients previously labeled as NMOSD were in fact AQP4-seronegative but MOG-IgG-positive. These patients had a distinctly different clinical phenotype, different magnetic resonance imaging (MRI) patterns, different treatment needs, and generally better attack recovery. MOGAD was formally separated as its own diagnosis.

The disease can present at any age, including in children, where MOGAD is among the most common causes of acquired demyelinating syndromes. The MOG protein, because it sits on the outermost layer of myelin (unlike myelin basic protein (MBP) or proteolipid protein (PLP), which are buried deeper), is particularly accessible to circulating antibodies — explaining why humoral autoimmunity against MOG causes such a distinct inflammatory syndrome.

Distinguishing MOGAD from Multiple Sclerosis and NMOSD

Correctly distinguishing MOGAD from multiple sclerosis and NMOSD is essential because the treatments differ fundamentally. Using the wrong treatment — particularly standard multiple sclerosis disease-modifying therapies (DMTs) — in a MOGAD patient is not just ineffective; there is concern that some multiple sclerosis DMTs may actually worsen MOGAD or precipitate attacks.

MOGAD vs. Multiple Sclerosis:

MOGAD vs. NMOSD (AQP4-IgG positive):

Clinical Manifestations

MOGAD produces a characteristic but heterogeneous set of clinical phenotypes. Unlike multiple sclerosis, which presents most commonly with sensory symptoms and relapsing-remitting white matter lesions, MOGAD has phenotypic signatures that are each quite distinctive. The most common presentations include:

Optic Neuritis (ON) — the single most frequent manifestation of MOGAD in both adults and children. Often bilateral simultaneously (15–40% of MOGAD optic neuritis cases involve both eyes at the same time — in multiple sclerosis, simultaneous bilateral ON is rare). Typically anterior (the disc itself is swollen, visible as papillitis on direct fundoscopy, unlike MS ON which is usually retrobulbar). Painful with eye movement. Initial visual loss is often severe, but recovery is usually excellent (see Optic Neuritis in MOGAD section).

Acute Disseminated Encephalomyelitis (ADEM) — the dominant pediatric MOGAD phenotype. Rapid-onset encephalopathy (altered consciousness/behavior), fever, and multifocal neurological deficits following a viral illness or vaccination. Large, fluffy, bilateral T2 lesions on magnetic resonance imaging with cortical involvement and thalamic lesions. MOG antibodies are found in a large share of children with ADEM (see ADEM section).

Myelitis — inflammation of the spinal cord causing weakness, sensory loss, and bladder dysfunction. MOGAD myelitis is often longitudinally extensive (spanning >3 vertebral segments) and characteristically involves the conus medullaris (the lower tip of the spinal cord). On axial magnetic resonance imaging cross-sections, the central gray matter is preferentially affected, producing an "H-sign" (the shape of the gray matter columns on cross-section). This contrasts with NMOSD myelitis, which tends to involve dorsal columns.

Cortical Encephalitis / Leptomeningeal Inflammation — a MOGAD-specific phenotype not seen in multiple sclerosis or NMOSD. Patients present with seizures (often focal), fever, altered consciousness, and magnetic resonance imaging demonstrating cortical and leptomeningeal (pial) enhancement. This pattern is highly suggestive of MOGAD when present.

Brainstem and Cerebellar Attacks — diplopia, ataxia, facial numbness. Less common than ON or myelitis but can occur, particularly with large ADEM-like episodes.

Relapsing vs. Monophasic Course: Approximately 50–60% of adults with MOGAD experience a relapsing course (multiple attacks). Children, especially those presenting with ADEM, more often have a monophasic course. The serostatus at 12 months is an important predictor — persistent MOG-IgG positivity is associated with higher relapse risk; MOG-IgG becoming negative during follow-up suggests lower relapse risk.

Optic Neuritis in MOGAD

Optic neuritis (ON) is the most common clinical manifestation of MOGAD and has features that set it distinctly apart from multiple sclerosis-related optic neuritis. Bilateral or anterior ON is a pattern that leads to testing for both MOG-IgG and AQP4-IgG.

Bilateral simultaneous involvement: In multiple sclerosis, optic neuritis is almost always strictly unilateral — bilateral simultaneous multiple sclerosis optic neuritis is uncommon (<5%). In MOGAD, 15–40% of optic neuritis episodes involve both eyes simultaneously. This is one of the most important clinical red flags for MOGAD, and testing for MOG-IgG and AQP4-IgG is how it is told apart from NMOSD.

Anterior (papillitis) pattern: MOGAD optic neuritis preferentially involves the anterior segment of the optic nerve, including the optic disc itself. On direct ophthalmoscopy or fundoscopy, the optic disc appears swollen (papilledema / papillitis). This is in contrast to multiple sclerosis optic neuritis, which is typically retrobulbar — the nerve is inflamed posterior to the globe so the disc looks normal on initial examination. In MOGAD, disc swelling is a common and important clue.

Perioptic enhancement on magnetic resonance imaging: Fat-suppressed gadolinium-enhanced magnetic resonance imaging of the orbits in MOGAD often shows enhancement that extends around the optic nerve sheath (perineural enhancement), not just within the nerve itself. This "halo" or perioptic enhancement pattern is highly characteristic of MOGAD and unusual in multiple sclerosis.

Pain with eye movement: Like multiple sclerosis optic neuritis, MOGAD ON is typically painful, with pain worsening on lateral or vertical eye movement (stretching the inflamed optic nerve sheath).

Severity and recovery: The initial visual loss in MOGAD optic neuritis is often severe — patients may lose vision down to counting fingers or light perception at nadir. This severity is typically worse than multiple sclerosis. However, the recovery is far better: in a 2018 multicentre series of 87 people with MOG-IgG optic neuritis (median 3 attacks each), average vision at the worst point of the worst attack was count-fingers, yet average final vision was 20/30 (near-normal); 5 patients (6%) ended at 20/200 or worse in either eye. Optic disc swelling and eye pain each occurred in 86%. The authors concluded that despite repeated attacks with severe vision loss, most patients recover substantially and keep functional vision long-term. The acute attack can be frightening because of how severe the first vision loss is.

Short-segment anterior lesions: The T2 signal abnormality and enhancement in MOGAD ON tends to be in the anterior (intraorbital, retrobulbar) segment of the optic nerve, and lesions tend to be short. NMOSD optic neuritis, by contrast, tends to involve longer segments and extends more posteriorly toward the chiasm. Chiasmal involvement is more common in NMOSD than MOGAD.

Recurrent optic neuritis: Some patients with MOGAD develop recurrent ON — repeated attacks affecting the same or contralateral eye. Cumulative damage from recurrent episodes can ultimately impair vision despite good single-attack recovery. This is the reason immunosuppressive relapse prevention is used in people with recurrent ON.

ADEM in Children

Acute disseminated encephalomyelitis (ADEM) is the most common initial presentation of MOGAD in children, and MOG-IgG testing has transformed the understanding of pediatric ADEM. MOG antibodies are found in a substantial proportion of children with ADEM — estimates range from 30–60% in pediatric ADEM cohorts. This discovery means that ADEM is not simply a post-infectious clinical syndrome but often represents the first attack of an antibody-mediated demyelinating disease with specific implications for prognosis and treatment.

Clinical presentation: ADEM typically follows a viral illness (influenza, enteroviruses, Epstein-Barr virus, COVID-19) or, less commonly, vaccination by days to weeks. The hallmark is encephalopathy — a change in level of consciousness or behavior (confusion, lethargy, irritability) combined with multifocal neurological deficits (weakness, ataxia, visual disturbance, sensory changes, bladder dysfunction). Fever is common. The illness is often rapid in onset and alarming for families.

Magnetic resonance imaging findings in MOG-ADEM: Bilateral, large, poorly-demarcated T2/FLAIR lesions in the white matter are the hallmark. What distinguishes MOG-positive ADEM from classic ADEM is a greater degree of cortical involvement — fluffy T2 signal and cortical/leptomeningeal enhancement. Thalamic lesions are common and sometimes striking (bilateral thalamic signal change, the "bowtie" pattern on axial images). Deep grey matter structures (basal ganglia, thalami) are affected more often in MOG-ADEM than in classic post-infectious ADEM.

MOG-IgG titers and ADEM: Children with MOG-positive ADEM often have high MOG-IgG titers during the acute phase. Serial testing shows that titers frequently decline after recovery from the initial episode, and in many monophasic cases, MOG-IgG eventually becomes undetectable. Persistent high titers predict a relapsing course.

Monophasic vs. relapsing: Many children with MOG-positive ADEM have a single episode (monophasic), recover well, and do not go on to have further demyelinating attacks. However, a significant minority — particularly those with persistently elevated MOG-IgG titers — go on to develop recurrent ADEM, bilateral optic neuritis, myelitis, or other MOGAD phenotypes. The clinical trajectory cannot be determined from the first attack alone, which is the reason given for repeat MOG-IgG testing after a pediatric ADEM episode.

Treatment of pediatric MOG-ADEM: High-dose intravenous (IV) methylprednisolone (20–30 mg/kg/day, max 1 g/day, typically for 3–5 days) is the standard first-line treatment and produces dramatic improvement in many children. Children who fail to respond adequately to steroids are candidates for plasma exchange (PLEX) or intravenous immunoglobulin (IVIg). Recovery in children with MOG-ADEM is often complete or near-complete, particularly with early treatment.

Distinguishing from multiple sclerosis in children: Unlike MS, MOG-ADEM in children does not produce periventricular "Dawson's finger" lesions on magnetic resonance imaging, does not meet McDonald criteria, and does not typically have oligoclonal bands in cerebrospinal fluid (CSF). ADEM is one of the three typical presentations listed in the 2023 International MOGAD Panel criteria (with optic neuritis and transverse myelitis), and a positive MOG-IgG result in a child with ADEM shapes both acute management and the decision about long-term immunosuppression.

Serology and Diagnosis

2023 international criteria: In 2023 an International MOGAD Panel published proposed diagnostic criteria for MOGAD, built from an extensive literature review and a structured consensus process. In these criteria the presence of MOG-IgG is a core criterion. MOGAD is described as typically presenting with ADEM, optic neuritis or transverse myelitis, and less commonly with cerebral cortical encephalitis, brainstem or cerebellar attacks; it can be monophasic or relapsing; cell-based assays are named as important for diagnostic accuracy; and diagnoses such as multiple sclerosis need to be excluded. The panel noted that the criteria still require validation.

The diagnosis of MOGAD rests on the detection of MOG-IgG antibodies in a clinically compatible patient using the correct testing methodology. The choice of assay is critical — not all MOG-IgG tests are created equal, and using an inferior assay leads to false positives and incorrect diagnoses.

Cell-based assay (CBA) — the gold standard: MOG-IgG is detected using a live cell-based assay in which the full-length MOG protein is expressed on the surface of live cells (typically HEK293 cells). Serum from the patient is applied to these cells, and fluorescent antibody detection identifies MOG-IgG binding to the surface-expressed protein. This method is used because MOG undergoes conformational changes when expressed on a cell surface that are critical for antibody recognition — denatured MOG in an ELISA or Western blot loses these conformational epitopes, leading to false positives with irrelevant antibodies binding to denatured protein. ELISA and Western blot tests for MOG-IgG are not accepted for diagnostic use by major neurology bodies.

Serum vs. Cerebrospinal fluid testing: Serum MOG-IgG is the usual first test. A 2024 study of cerebrospinal fluid (CSF) testing from Mayo Clinic and other centres found CSF MOG-IgG in 9 of 73 (12%) people with a high-risk MOGAD pattern whose serum test was negative; these serum-negative, CSF-positive patients made up 9 of 83 (11%) of the MOGAD cases, and all of them met the 2023 MOGAD criteria. The authors concluded that CSF testing is useful when the clinical picture is suspicious but serum is negative, and when a low-positive serum result leaves doubt. The major benefit of serum testing is that CBA-based assays on serum have high sensitivity and specificity when the clinical phenotype is appropriate.

Titer interpretation: MOG-IgG titers are read in clinical context. Low-titer positive results (borderline positivity) are read cautiously, particularly in patients whose clinical syndrome does not fit MOGAD — low titers may represent non-specific binding. High titers in the context of a MOGAD-compatible presentation are strong diagnostic evidence. Serial titers are valuable for monitoring — declining titers during treatment or spontaneous recovery correlate with clinical improvement; rising titers may predict relapse.

AQP4-IgG exclusion: Testing for aquaporin-4-IgG (AQP4-IgG), the antibody associated with NMOSD, is part of the work-up for MOGAD. The two antibodies are mutually exclusive in the great majority of cases. A patient who is AQP4-IgG positive is diagnosed with NMOSD, not MOGAD, and receives different treatment. Co-positivity for both antibodies is usually followed by repeat testing with CBA methods.

Seronegative MOGAD: Patients with a clinical syndrome entirely consistent with MOGAD (bilateral optic neuritis, ADEM pattern, conus myelitis) who test MOG-IgG negative by CBA represent a diagnostic challenge. Some may have seronegative MOGAD — the antibody may be below the detection threshold, have been cleared by treatment before testing, or the assay may not capture all pathogenic antibody variants. These cases rest on the clinical judgment of specialists in inflammatory demyelinating diseases; the 2024 CSF study above found that some such patients carry MOG-IgG in the cerebrospinal fluid only.

Cerebrospinal fluid findings: cerebrospinal fluid in MOGAD typically shows mild lymphocytic pleocytosis (elevated white cell count) and mildly elevated protein during acute attacks. Importantly, oligoclonal bands (OCBs) — a hallmark of multiple sclerosis — are absent or present in only a minority of MOGAD patients (approximately 15–20%), compared to >95% in multiple sclerosis. The absence of OCBs in a patient with optic neuritis or myelitis is a clue to consider MOGAD over multiple sclerosis.

Magnetic Resonance Imaging Characteristics

Magnetic resonance imaging is central to the diagnosis and monitoring of MOGAD. The imaging patterns are distinctive and differ from both multiple sclerosis and NMOSD in important ways that lead radiologists and clinicians to consider MOGAD testing.

Optic nerve magnetic resonance imaging:

Spinal cord magnetic resonance imaging:

Brain magnetic resonance imaging:

Treatment

MOGAD treatment has two goals: (1) rapidly suppressing acute attacks to minimize permanent damage, and (2) preventing future relapses in patients with a relapsing course. Critically, standard multiple sclerosis disease-modifying therapies are not used in MOGAD.

Acute Attack Treatment:

Relapse Prevention (Maintenance Immunosuppression):

Not all MOGAD patients require maintenance immunosuppression — those with a monophasic course, particularly children with MOG-ADEM who become MOG-IgG seronegative, may do well without long-term treatment. For those with a relapsing course (especially relapsing optic neuritis or myelitis), maintenance immunosuppression is used. Options include:

Medicines under study: No medicine is approved for MOGAD, and a 2025 review describes treatment as still empirical. Drugs being tested in phase 2/3 trials include azathioprine (MOGwAI trial), tocilizumab (TOMATO), satralizumab (METEOROID) and rozanolixizumab (cosMOG).

Multiple sclerosis DMTs in MOGAD: Interferon-beta preparations (interferon-beta 1a, interferon-beta 1b), glatiramer acetate, natalizumab, fingolimod, siponimod, and other S1P modulators are reported not to be effective in MOGAD and are not used for it. Some case reports suggest these agents may worsen MOGAD or fail to prevent relapses. Alemtuzumab and ocrelizumab have limited data in MOGAD. NMOSD-specific biologics (eculizumab, satralizumab, inebilizumab) are approved for AQP4-IgG positive NMOSD but their role in MOGAD is not established; they are not routinely used (satralizumab is being tested in the METEOROID trial).

Prognosis and Monitoring

The prognosis of MOGAD is generally more favorable than NMOSD and in many respects more favorable than multiple sclerosis for individual attack recovery — but cumulative disability from repeated attacks can accrue over time, making relapse prevention critically important for those with a relapsing course.

Visual recovery: In the 2018 series of 87 people with MOG-IgG optic neuritis, average final vision was 20/30 (near-normal) and 6% ended at 20/200 or worse in either eye. This is substantially better than NMOSD optic neuritis, where permanent severe visual loss is common. However, patients with recurrent optic neuritis may accumulate visual field loss over multiple attacks even if each individual attack recovers well.

Spinal cord recovery: Myelitis attacks in MOGAD often have good functional recovery, particularly when treated promptly. The conus-predominant pattern may cause prominent bladder dysfunction acutely, which often but not always recovers.

Pediatric prognosis: Children with MOG-ADEM typically have excellent recovery — many return to baseline neurological function after the initial episode. The risk of a relapsing course is lower in children than in adults, particularly when MOG-IgG becomes undetectable after the first attack.

Serostatus as a prognostic marker:

Monitoring schedule: Monitoring described for MOGAD includes serial MOG-IgG titers (every 3–6 months initially), visual acuity and visual field testing (Humphrey 24-2 perimetry) every 6–12 months, optical coherence tomography (OCT) of the retinal nerve fiber layer to track subclinical optic nerve atrophy, and magnetic resonance imaging of the brain, orbits, and spinal cord every 6–12 months during active disease. Stable patients on immunosuppression may transition to annual monitoring.

Pregnancy: MOGAD can relapse postpartum (in the months after childbirth). IVIg is generally considered the safest maintenance option during pregnancy and breastfeeding.

Long-term disability: A 2026 review notes that roughly half of adults with MOGAD relapse, and that long-term disability comes from the build-up of incompletely recovered attacks rather than from a progressive phase. With proper diagnosis and maintenance immunosuppression, the majority of MOGAD patients maintain good functional independence. Delay in diagnosis — particularly when patients are incorrectly started on multiple sclerosis DMTs that do not prevent MOGAD relapses — leads to preventable cumulative disability. Early correct diagnosis is the most important factor in long-term prognosis.

Key Research Papers

  1. Cobo-Calvo A, et al. (2018). Clinical spectrum and prognostic value of CNS MOG autoimmunity in adults: The MOGADOR Study. Neurology, 90(21), e1858–e1869. — PubMed PMID: 29695592
  2. Jurynczyk M, et al. (2017). Clinical presentation and prognosis in MOG-antibody disease: a UK study. Brain, 140(12), 3128–3138. — PubMed PMID: 29136091
  3. Hacohen Y, et al. (2018). Disease course and treatment responses in children with relapsing myelin oligodendrocyte glycoprotein antibody-associated disease. JAMA Neurology, 75(4), 478–487. — PubMed PMID: 29305608
  4. Mader S, et al. (2011). Complement activating antibodies to myelin oligodendrocyte glycoprotein in neuromyelitis optica and related disorders. Journal of Neuroinflammation, 8, 184. — PubMed PMID: 22204662
  5. Chen JJ, et al. (2018). Myelin oligodendrocyte glycoprotein antibody-positive optic neuritis: Clinical characteristics, radiologic clues, and outcome. American Journal of Ophthalmology, 195, 8–15. — PubMed PMID: 30055153
  6. Whittam DH, et al. (2020). Treatment of MOG-IgG-associated disorder with rituximab: an international study of 121 patients. Multiple Sclerosis and Related Disorders, 44, 102251. — PubMed PMID: 32629363
  7. Hamid SHM, et al. (2017). What proportion of AQP4-IgG-negative NMO spectrum disorder patients are MOG-IgG positive? A cross sectional study of 132 patients. Journal of Neurology, 264(10), 2088–2094. — PubMed PMID: 28840314
  8. Banwell B, et al. (2023). Diagnosis of myelin oligodendrocyte glycoprotein antibody-associated disease: International MOGAD Panel proposed criteria. Lancet Neurology, 22(3), 268–282. — PubMed PMID: 36706773
  9. Redenbaugh V, et al. (2024). Diagnostic utility of MOG antibody testing in cerebrospinal fluid. Annals of Neurology, 96(1), 34–45. — PubMed PMID: 38591875
  10. Chen JJ, et al. (2022). Association of maintenance intravenous immunoglobulin with prevention of relapse in adult myelin oligodendrocyte glycoprotein antibody-associated disease. JAMA Neurology, 79(5), 518–525. — PubMed PMID: 35377395
  11. Virupakshaiah A, et al. (2026). Comparative effectiveness of treatments for relapse prevention in myelin oligodendrocyte glycoprotein antibody-associated disease. Neurology, 107(8), e218508. — PubMed PMID: 42842889
  12. Carnero Contentti E, et al. (2025). Future treatments for myelin oligodendrocyte glycoprotein antibody-associated disease: the clinical trial landscape. Expert Opinion on Emerging Drugs, 30(4), 283–297. — PubMed PMID: 40984653
  13. Ali A, et al. (2026). Efficacy and safety of interleukin-6 receptor inhibitors in myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD): a systematic review and meta-analysis. Therapeutic Advances in Neurological Disorders, 19, 17562864261490506. — PubMed PMID: 42732282
  14. Mahler JV, et al. (2026). Treatment of myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD): acute attacks, relapse prevention, and symptomatic care. Frontiers in Neurology, 17, 1965135. — PubMed PMID: 42921072

↑ Back to Table of Contents

Connections