Emerging Principles for Treating Myelin Oligodendrocyte Glycoprotein Antibody-Associated Disease (MOGAD)

Abstract

Purpose of review

Myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD) is a rare inflammatory disorder of the central nervous system that affects both adults and children. Neurologic disability is relapse-driven; therefore, early diagnosis and targeted treatment are critical for effective care. We review the new MOGAD diagnostic criteria and evidence for current acute and preventative therapies.

Recent findings

The International MOGAD Panel has released the first clinical, laboratory, and radiographic criteria for MOGAD diagnosis. These criteria set the stage for evaluating clinical investigations and designing future randomized clinical trials. Prior retrospective studies have evaluated multiple off-label agents for the acute care or prevention of MOGAD attacks, and prospective randomized clinical trials are now underway.

Summary

Acute MOGAD attacks are generally responsive to high-dose corticosteroids; however, early use of plasma exchange or intravenous immunoglobulin may be beneficial for severe attacks or cases lacking corticosteroid response. A slow corticosteroid taper may lower the risk of relapse. Preventative treatment has been typically limited to patients with a definitive relapsing disease. While there is no consensus on the choice or duration of treatment, multiple therapies have been retrospectively evaluated. Prospective placebo-controlled trials for interleukin-6 receptor inhibition and neonatal Fc receptor inhibition may open new frontiers for patient care.

Introduction

Serum and cerebrospinal fluid (CSF) immunoglobulin G (IgG) auto-antibodies to conformational myelin oligodendrocyte glycoprotein (MOG) are associated with a variety of CNS clinical presentations [1, 2]. Frequent clinical presentations associated with MOG-IgG include optic neuritis, myelitis, and acute disseminated encephalomyelitis (ADEM); focal infratentorial and supratentorial syndromes are less common [3, 4]. Recently, new diagnostic criteria for MOG antibody-associated disease (MOGAD) were proposed by the International MOGAD Panel (Table 1) based on the presence of one or more core demyelinating events, positive MOG-IgG testing in the serum or CSF, supporting clinical and radiographic features, and exclusion of a better diagnosis [5••].

Table 1.

MOGAD diagnostic criteria adapted from those proposed by the International MOGAD Panel [5••]

(A) Core clinical demyelinating event	Optic neuritis	Monofocal or polyfocal cerebral deficits
	Myelitis	Brainstem or cerebellar deficits
	Acute disseminated encephalomyelitis	Cerebral cortical encephalitis
(B) Positive MOG-IgG cell-based assay	Serum	Clear positivea—no supporting features required
		Low positivea—≥1 supporting feature required
		Positive with no titera—≥1 supporting feature required
	CSF	Positive but seronegative—≥1 supporting feature required
Supporting clinical or MRI features	Optic neuritis	Bilateral simultaneous clinical involvement
		Longitudinal optic nerve involvement (>50% of nerve length)
		Perineural optic sheath enhancement
		Optic disc edema
	Myelitis	Longitudinally extensive myelitis (≥3 vertebral segments in length)
		Central cord lesion or H-sign
		Conus lesion
	Cerebral, brainstem, or cerebellar syndrome	Multiple ill-defined T2 hyperintense lesions in supratentorial and/or infratentorial white matter
		Deep gray matter involvement
		Ill-defined T2-hyperintensity involving the pons, middle cerebellar peduncle, or medulla
		Cortical lesion ± lesional or overlying meningeal enhancement
(C) Exclusion of better diagnoses including multiple sclerosis and AQP4-IgG positive NMOSD

^aAssay-specific titer cutoffs are provided in the appendix of the International MOGAD Panel publication [5••]

MOGAD is a rare disease with an incidence estimated at up to 3.4 per million persons per year and a prevalence estimated at up to 20 per million persons [6, 7]. Since MOGAD has only recently been recognized as a distinct clinical disorder and laboratory testing for MOG-IgG has been relatively limited to research labs, optimal treatment strategies have not yet been delineated in clinical trials. Identifying safe and effective ways to treat MOGAD is important due to the expected increase in incidence and prevalence with increased testing in the context of the newly proposed diagnostic criteria. In addition, MOGAD already comprises a significant fraction of patients presenting with optic neuritis and myelitis, resulting in substantial patient and societal burden due to relapses and long-term morbidity [3]. MOGAD can have a monophasic or relapsing disease course which has muddled the interpretation of retrospective studies and challenged the design of prospective clinical trials to evaluate effective immunomodulatory or immunosuppressive treatments. While there are currently four registered prospective, randomized, placebo-controlled clinical trials (RCTs) investigating candidate therapeutics for the prevention of MOGAD relapses, regulatory approval of any agent is likely far off in the future. Therefore, in the short term, the general efficacy and safety of off-label therapies employed in the treatment of other CNS inflammatory disorders need to be evaluated.

Here, we review emerging principles of MOGAD treatment based on the best available evidence. We will discuss the basics of MOGAD diagnosis, the management of acute relapses, approaches to maintenance therapy, and current phase III clinical trials.

MOGAD diagnosis

MOGAD clinical attacks (also termed flares, exacerbations, or, on recurrence, relapses) are driven by acute cellular and humoral inflammation. An attack is frequently defined as a new symptom localizable to the CNS lasting at least 24 h, with supporting findings on MRI or clinical examination [8]. The recently proposed diagnostic criteria specify that a relapse occurs more than 30 days from a prior attack, with a note that some patients can cluster relapses or have extended fluctuations in symptoms with an initial attack [5••]. The classic clinical manifestations of MOGAD are optic neuritis, ADEM, and myelitis, although a wide variety of other phenotypes can occur, particularly focal infratentorial or supratentorial syndromes [5••]. Presentations including encephalitis, cranial neuropathies, and myeloradiculitis have also been described, emphasizing the broad diversity of clinical manifestations [9–11]. Common clinical presentations of MOGAD vary by age; pediatric patients most frequently present with ADEM, followed by optic neuritis, while adults most often present with optic neuritis, followed by transverse myelitis [5••]. While in adult patients most acute CNS demyelinating events are associated with multiple sclerosis (MS), MOGAD is a frequent cause in pediatric patients. Unlike MS, MOGAD is not thought to have a progressive course. Although disability accumulation from attacks is not as profound as in neuromyelitis optic spectrum disorder (NMOSD), recovery is incomplete in many patients [12]. Therefore, treatments to maximize recovery and prevent relapses are the primary aim.

The current diagnostic criteria for MOGAD (Table 1) outline clinical, serological, and radiographic criteria that are critical for distinguishing MOGAD from AQP4-IgG-positive NMOSD, MS, and other overlapping conditions that require alternative treatment strategies [5••]. For patients who meet diagnostic criteria for MS, MOG-IgG testing is not recommended [5••]. Diagnosis requires the selection of an assay with optimal sensitivity and specificity, ideally a live cell-based assay with expression of full-length MOG-IgG; however, all commercial cell-based MOG-IgG assays can be incorporated within the proposed diagnostic criteria [5••]. Serum is felt to be the primary source for testing; CSF testing should only be undertaken in high-suspicion seronegative patients (Table 1) [5••].

Management of acute relapse

MOGAD relapses are typically responsive to high-dose intravenous methylprednisolone (IVMP), plasma exchange (PLEX), and intravenous immunoglobulin (IVIg). The use of these agents is supported by retrospective case studies, expert opinion, and insights from the treatment of acute inflammatory lesions in MS and NMOSD (Fig. 1). Transposing acute treatments between MS, NMOSD, and MOGAD, however, should be done with caution since the underlying lesion pathologies are unique [13]. High-dose corticosteroids are the mainstay of MOGAD relapse treatment, and most of the supportive data comes from the management of optic neuritis. Patients with MOGAD will often have a rapid and early recovery following steroid treatment. An international survey of 52 neurologists with expertise in MOGAD revealed that 100% would treat relapses with high-dose steroids for at least 3 days, and this is also recommended per the European Union Pediatric MOG Consortium guidelines [14, 15]. In the Optic Neuritis Treatment Trial (ONTT), which evaluated the management of acute optic neuritis with high-dose intravenous steroid treatment (1000 mg total daily dose IVMP for 5 days followed by 1 mg/kg oral prednisone for 11 days), IVMP was associated with faster recovery of high contrast visual acuity (HCVA), but no change in extent of recovery at 6 months when compared to placebo or lower-dose oral steroid treatment [16]. While IVMP is most frequently used, oral steroid administration (oral prednisone 1250 mg for 3 days) has demonstrated non-inferiority for acute optic neuritis and offers a potential practical alternative for non-hospitalized patients who may not have immediate access to infusion therapy [17]. Notably, in the ONTT, only 4/177 patients with serum available for testing were positive for MOG-IgG, and bilateral optic neuritis, common in MOGAD, was an exclusion criterion [18]. Hence, the generalizability of ONTT results to MOGAD remains uncertain.

Fig. 1.

Management of initial MOGAD attack and considerations for management following relapse. This algorithm may be modified based on age, comorbidities, changes in MOG-IgG status, severity of initial attack, and risk of further attacks.

High-dose IVMP is the most common treatment utilized in retrospective studies of MOGAD attacks. In a retrospective evaluation of 122 MOGAD relapses treated solely with IVMP, 50% of cases had complete or nearly complete recovery, 44.3% had partial recovery, and 7% had no or almost no recovery [19]. For cases of optic neuritis, the rapidity of steroid initiation impacts clinical outcome; treatment initiation after ≥ 10 days is associated with a lower rate of HCVA recovery and decreased peripapillary retinal nerve fiber layer thickness (pRNFL) measured by optical coherence tomography (OCT) at 3 months [20]. A smaller retrospective study of MOGAD and NMOSD optic neuritis noted a similar time dependence (initiation within 7 days) for optimal treatment with steroids [21].

Alternative, often second-line, treatments for acute MOGAD relapse include PLEX and IVIg. As the timing of second-line treatment may also impact clinical outcomes, PLEX or IVIg may be delivered sooner for patients with severe symptoms and no response to IVMP. The European Union Pediatric MOG Consortium guidelines suggest moving to IVIg or PLEX after 3 days or 5 days of IVMP if no or inadequate improvement is observed [15]. IVIg may be easier to deliver in some clinical settings and is typically used more often in pediatric patients than in adults. While formal data on practice patterns is not available, from survey data on the acute management of MOGAD, PLEX was favored as second-line treatment by adult neurologists while pediatric neurologists were split between PLEX and IVIg, and this may, in part, be related to practical issues [14]. For a first attack, the diagnosis of MOGAD is often not confirmed while acute treatment is being arranged; therefore, aggressive treatment with PLEX may be warranted if NMOSD is high in the differential [22, 23]. Data on IVIg as an acute treatment is less robust, but it may be considered as a second-line treatment in pediatric cases or if there are contraindications or limited access to PLEX [19]. Ideally, serum MOG-IgG should be performed prior to administration of IVIg (which may cause false positives) or PLEX (which may result in false negatives). Typical maintenance therapies have not yet been evaluated in the setting of acute relapses.

Extended steroid tapers (≥ 3 months) are often recommended based on retrospective data showing an increased risk of relapse with a more rapid reduction [14]. The proposed taper per the European Union Pediatric MOG Consortium starts at 1–2 mg/kg prednisone daily with a reduction over 4 weeks to 0.5 mg/kg prednisone daily, then completing the taper over the next 2 months. Alternative tapering strategies have been proposed, and evaluating comparative efficacy is difficult [15]. In a study of 59 patients (33 pediatrics and 26 adults) with ≥ 2 relapses attributed to MOGAD, there was an association between the risk of relapse and a prednisone dosage ≤ 10 mg daily, as well as an association between the length of prednisone taper and relapse. MOGAD patients with relapses had a median taper of 1.5 months compared to 5 months in patients without relapses [24]. In this study, most relapses occurred within 2 months of steroid cessation [24]. Lengthy steroid taper, however, may not be important for all MOGAD patients. In pediatric MOGAD patients suffering their first attack, steroid taper over ≥ 5 weeks was associated with a reduced risk of subsequent relapse [25].

Ideally, the acute management of MOGAD relapse, including the comparison of steroid tapering strategies, warrants prospective randomized clinical studies. For patients who start maintenance immunosuppressive or immunomodulatory treatment, bridging with steroids may be warranted, as some therapies, especially azathioprine and mycophenolate mofetil, may take > 3 months to reach full efficacy.

MOGAD relapse prevention

MOGAD has a distinct pathophysiology from NMOSD and MS and requires disease-specific trials of targeted treatments for adequate evaluation of efficacy and safety. At present, no RCTs evaluating treatment specifically for MOGAD have been completed, although 4 studies are registered. Current perspectives on the long-term treatment of MOGAD are guided primarily by retrospective data for which varying diagnostic criteria were used. In addition, the representative patient populations were heterogeneous, with long-term treatment initiated after a variable number of relapses.

Unlike MS, MOGAD does not appear to show disability progression independent of relapse activity nor high rates of asymptomatic CNS MRI lesions [26–29]. Nevertheless, roughly 30–50% of patients develop a relapsing disease course, and a proportion can accumulate substantial disability [1, 3, 4, 30•]. Serum MOG-IgG titers have been proposed as a biomarker for relapse prediction with those becoming seronegative having a low risk of relapse, but the association is not uniform across studies; many seropositive may remain monophasic depending on the presenting phenotype and the age of the cohort. In pediatric patients with a mean follow-up of 4.7 years, serum MOG-IgG titers persisted in 36% of patients, fluctuated in 7% of patients, and converted to negative in 53% of patients without any definitive association with relapse risk [31]. In a combined adult and pediatric population following ADEM, 15/17 patients with persistent seropositivity relapsed [32]. The risk of relapse did not differ in patients with CSF-only MOG-IgG or dual CSF and serum MOG-IgG titers [33]. A multicenter incident cohort study identified a decreased risk of relapse with transverse myelitis at presentation and a higher risk of relapse in young adults [4, 30•]. Relapse risk was typically greater in the first few months following an initial attack, further supporting extended steroid tapers [3].

To date, the ability of demographics, clinical phenotypes, or biomarkers (including MOG-IgG titers) to identify those at elevated risk for a relapsing course is unsatisfactory. Therefore, considering the modest rate of relapses and the high rate of clinical recovery, maintenance immunomodulatory or immunosuppressive treatment is not typically started following the first attack [1, 15, 34–36]. Maintenance treatment is typically initiated only after a relapsing course is defined by a second clinical event. Such an approach is supported by the European Union Pediatric MOG Consortium guidelines; there are no consensus guidelines for the management of adult MOGAD patients [14, 15].

Early initiation of maintenance treatment after a first attack may be considered if the attack is severe with poor recovery (e.g., unilateral blindness), placing the patient at risk of major disability (i.e., if they suffer a similar attack in the other eye). This early initiation after a single attack may be influenced by patient preference, even though they may have a limited relapse risk. Per survey data, treatment would usually or always be initiated by a minority (39%) of experts in this context [14]. Shared decision-making with patients is paramount, especially given limitations regarding the efficacy, safety, and necessary duration of long-term treatments. The reported efficacies of common MOGAD preventative therapies are presented below. Importantly, MS disease-modifying therapies (with the exception of B-cell depleting therapies) do not provide a clear clinical benefit in MOGAD [37, 38].

Prednisone

As reviewed above, prednisone tapers have a major role in preventing rebound attacks. Low-dose prednisone may also be effective as a maintenance treatment but is often avoided due to the risk of adverse effects including infections, osteopenia, weight gain, and elevated blood glucose with long-term use. However, doses of 10 mg daily may be well-tolerated and be suitable for women of childbearing potential or other patients with contraindications to frequently used maintenance therapies (Table 2).

Table 2.

Commonly used maintenance therapies for relapsing MOGAD

Therapy	Mechanism	Target regimen
Azathioprine	Interference with purine synthesis leading to reductions in B- and T-cell proliferation	2–3 mg/kg daily
Intravenous immunoglobulin	Modulation of lymphocyte function and antigen presentation; autoantibody neutralization	1000 mg/kg every 3–4 weeks
Mycophenolate mofetil	Inhibition of inosine monophosphate dehydrogenase leading to reductions in B- and T-cell proliferation	1000 mg twice daily
Prednisone	Broad spectrum immunosuppression	10 mg daily
Rituximab	CDC-mediated B-cell depletion via CD20	1000 mg q6 months

Azathioprine

Azathioprine (AZA) is a long-standing and relatively inexpensive broad-spectrum immunosuppressive agent. AZA antagonizes purine metabolism in proliferating lymphocytes and has frequently been used off-label for MOGAD and other neuroimmunological diseases. The efficacy of AZA for MOGAD has been assessed in multiple small retrospective studies demonstrating reductions in the annualized relapse rate (ARR), although up to 50% of patients may still experience relapses [24, 37, 38, 39••]. AZA can take up to 3–6 months to show treatment effects and typically necessitates the use of steroids as a bridging therapy. The MOGwAI (NCT05349006) RCT will evaluate the use of AZA vs. placebo after an initial attack, with the primary outcome being the time to first relapse (Table 3). In addition, the IDAR (NCT05545384) RCT will evaluate the efficacy of AZA or rituximab after an initial attack vs. starting either treatment after a subsequent relapse. The primary outcome is ARR over 24 months (Table 3). Side effects of azathioprine include bone marrow suppression, hepatotoxicity, infections, and an increased risk of lymphoma and skin malignancies. Assessing thiopurine methyltransferase function (TPMT) is important prior to starting; low TPMT activity levels increase the risk of side effects.

Table 3.

Registered clinical trials evaluating maintenance therapies in MOGAD. Information per clinicaltrials.gov

Clinical trial	Treatment	Comparator	Inclusion criteria	Primary outcome
IDAR (NCT05545384)	Azathioprine or rituximab	No maintenance treatment	Following first attack	Annualized relapse rate at 24 months
MOGwAI (NCT05349006)	Azathioprine with steroid taper	Steroid taper	Following first attack	Time to first relapse
cosMOG (NCT05063162)	Rozanolixizumab	Placebo	Relapsing course with ≥l relapse in 12 months	Time to first relapse
METEROID(NCT05271409)	Satralizumab (allows concomitant azathioprine or mycophenolate mofetil)	Placebo	Relapsing course with ≥l relapse in 12 months or ≥2 relapses in 24 months	Time to first relapse

Mycophenolate mofetil

Mycophenolate mofetil (MMF) is also a relatively accessible broad-spectrum immunosuppressive agent. MMF depletes guanosine nucleotides preferentially in B-cells and T-cells and is frequently used off-label for MOGAD and other neuro-immunological diseases. The efficacy of MMF for MOGAD has been assessed in multiple small retrospective studies demonstrating reductions in ARR similar to AZA [24, 38, 40]. Notably, like AZA, MMF may take 3–6 months to reach clinical efficacy and typically necessitates the use of an extended steroid taper. Common side effects of MMF include nausea and diarrhea, and there can be an increased risk for infections in the setting of cytopenias.

Anti-CD20 monoclonal antibodies

Rituximab is an anti-CD20 monoclonal antibody that depletes circulating B-cells but does not directly deplete plasmablasts. There is a sizable retrospective literature on using rituximab for the treatment of MOGAD, with rituximab being frequently chosen due to the efficacy of its off-label use in NMOSD and MS [41, 42]. Meta-analyses support that rituximab is effective for reducing the ARR in MOGAD, but efficacy may be less when compared to AQP4-IgG seropositive NMOSD [43, 44]. The most recent meta-analysis suggests a larger benefit for patients with higher baseline disease activity [43]. While neither total B-cell (CD19 +) nor memory B-cell (CD27 +) counts are typically associated with relapses in MS, a link between repletion and relapses has been observed in NMOSD [45, 46]. In patients with MOGAD, relapses tend to occur without any link to CD19 + repletion or missed infusions [47]. Relapse risk specifically associated with the repletion of CD27 + cells in MOGAD is limited [48]. Overall, current evidence suggests that rituximab has efficacy for treating MOGAD, but that efficacy is less robust than in larger experience treating MS or NMOSD [41, 42, 48]. As mentioned, rituximab will be evaluated as a primary treatment in the pending IDAR RCT (Table 3), and the results should provide a clearer representation. Side effects of rituximab include infusion reactions and increased risk for infections (predominantly upper respiratory and urinary), particularly when hypogammaglobulinemia is present.

Anti-CD19 monoclonal antibodies

Inebilizumab is an anti-CD19 monoclonal antibody that depletes cells in the B-cell lineage extending to plasmablasts. Inebilizumab was FDA-approved for AQP4-IgG-positive NMOSD in 2019. Approval was based on the N-MOmentum RCT which included both AQP4-IgG positive and negative patients per the 2015 NMOSD criteria [49]. Seven AQP4-IgG-negative patients were positive for MOG-IgG in the serum (CSF testing was not performed) [50]. Two patients had low titers at 1:20. Six of the MOG-IgG patients were in the inebilizumab treatment group, and 1 was in the placebo group. One MOG-IgG-positive patient had a relapse shortly after the first dose of inebilizumab, with no further relapses noted during the trial or the open-label extension. Overall, the small number of patients included makes the interpretation of clinical efficacy difficult; hence, a larger prospective study specifically designed for MOGAD patients is required. Side effects of inebilizumab include infusion reactions and increased risk for infections, particularly those of the respiratory and urinary tracts.

Intravenous or subcutaneous immunoglobulin

While the mechanisms of IVIg are not fully elucidated, it is thought to modulate lymphocyte function, antigen presentation, and neutralize autoantibodies. In addition to its use as an acute relapse treatment, IVIg in MOGAD is now increasingly used as a maintenance treatment at 1–2 g/kg every 3–4 weeks, especially in children where there are greater concerns about using corticosteroids [14, 37]. A subsequent larger multicenter retrospective study included 59 adult patients treated with IVIg: 25% as first-line treatment and 75% as second-line treatment. Twenty-five percent of the cohort remained on other forms of immunomodulatory treatment while receiving IVIg. Over 1.4 years of follow-up, treatment with IVIG resulted in an ARR reduction from a median annualized rate of 1.4 to 0.39. While 34% of patients still experienced relapses while on IVIg, there was a relationship with dosage, as only 17% of patients treated with at least 1 g/kg of IVIg every 4 weeks had a relapse [39••]. Only 3% of patients discontinued IVIg due to adverse events. Other smaller retrospective cohorts have also demonstrated the benefit of IVIg treatment for relapse risk reduction [24, 37, 38].

With increasing evidence supporting a high benefit-risk ratio for IVIg in the prevention of MOGAD relapse, IVIg is commonly used as first-line therapy along with extended steroid taper following a MOGAD relapse or after a severe initial attack. Ideally, IVIg should be prospectively studied in a RCT to fully evaluate efficacy and safety. The disadvantages of IVIg treatment include its comparatively high cost and limited availability in many healthcare systems. Subcutaneous immunoglobulin has potential advantages over IVIg as it does not require infusion center treatment and may have fewer side effects, although use has been limited [51]. Side effects of IVIg include headaches (including from aseptic meningitis), thromboembolic events, and infusion reactions.

Neonatal Fc receptor (FcRn) inhibition

Inhibition of FCRn increases the catabolism of IgG and can reduce circulating levels of pathogenic antibodies. Rozanolixizumab is a subcutaneously injected FcRn inhibitor that was recently approved for the treatment of generalized myasthenia gravis [52, 53]. cosMOG (NCT05063162) is a phase III RCT that will compare rozanolixizumab with a placebo (without add-on immunosuppression) on the time to first relapse in relapsing MOGAD patients (Table 3). In the completed trials of rozanolixizumab for myasthenia gravis, headache and diarrhea were the most frequently reported adverse events; the rate of infections was equal between rozanolixizumab and placebo, with no severe, serious, or opportunistic infections reported with rozanolixizumab treatment [52, 53].

IL-6 Receptor inhibition

Anti-IL-6-receptor monoclonal antibodies, tocilizumab and satralizumab, have been used as effective treatments for a variety of CNS inflammatory conditions. Indeed, satralizumab has been approved for the treatment of AQP4-IgG seropositive NMOSD since 2019. Similar to AQP4-IgG-positive NMOSD, IL-6 is elevated in the CSF of MOGAD patients, and therefore IL-6 signaling may play an intrinsic role in MOGAD pathophysiology [54]. While the SAKuraSky and SAKuraStar RCTs included AQP4-IgG-seronegative patients, the outcomes of any patients who were MOG-IgG-seropositive have not been reported [55]. In the TANGO study, which compared AZA and tocilizumab in the treatment of NMOSD, one MOG-IgG seropositive patient did not relapse while on tocilizumab [56]. Retrospective case studies have also reported that tocilizumab reduced relapse rates in MOGAD patients [57•, 58–62], although the follow-up time and patient numbers are limited. In the largest study, 73% of patients remained relapse-free for more than 1 year on treatment [53]. The METEOROID RCT (NCT05271409) is currently recruiting for evaluation of satralizumab vs. placebo in MOGAD with primary outcome time to first relapse (Table 2). The METEOROID study allows satralizumab to be used as an add-on treatment to a stable dose of AZA and MMF. Side effects of IL-6 inhibition include infections, cytopenias, and liver enzyme elevations.

Biomarkers

Biomarkers of disease activity, prognosis, and therapeutic efficacy are needed to help guide treatment decisions in MOGAD. At present, there is not sufficient data to strongly support the use of any biomarker for these purposes. Given the incomplete relationship between MOG-IgG titers and relapse risk, other targets need to be evaluated [31]. MOGAD is not thought to have significant subclinical disease activity on MRI [26]. While additional studies are needed, both serum neurofilament light chain (sNFL) levels and OCT pRNFL thickness may prove to be helpful in understanding patterns of disease activity between attacks and measuring disease severity [35, 63]. CSF IL-6 and TNF-α are elevated during MOGAD attacks, and it is possible that the degree of elevation may have prognostic significance for the risk of recurrence as well as for recovery from individual attacks [64].

Treatment duration

For patients who start maintenance therapy, the optimal duration of treatment is not known. Observational studies have demonstrated that relapse rates plateau over time, and for patients who respond to treatment, indefinite use of maintenance therapies may not be needed [3, 29, 65]. Adverse effects (e.g., infections and osteopenia) of certain maintenance therapies may also be dependent on the duration of treatment. As such, withdrawal of maintenance therapy may be reasonable after a few years of treatment, but the data guiding this decision remains limited.

Conclusions

MOGAD is a rare disease with wide-ranging clinical phenotypes and trajectories that is pathophysiologically distinct from MS and NMOSD. While the management of acute relapses follows principles utilized for other demyelinating conditions, optimal long-term management is less certain. Hence, research needs to focus on determining which patients are at high risk for relapsing disease and which maintenance treatments are most effective at preventing relapses. Given the limitations of current data, shared decision-making between the patient and clinician is important when determining a long-term treatment strategy. While current RCTs will help address the efficacy of some therapies, extended observational studies are needed to evaluate the duration of treatment and long-term safety. Basic and translational research on MOGAD immunopathology will likely identify new therapeutic targets and predictive biomarkers to advance clinical outcomes.