Myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD) is now recognised as a distinct disease entity. The new diagnostic criteria represent a major step toward improving its diagnosis [1].
However, additional research is still needed to standardise testing methods and identify new biomarkers that could further validate the criteria and support the development of effective therapies [2].
The proposed diagnostic criteria rely on the presence of MOG Immunoglobulin G (MOG-IgG) as a core criterion, together with typical clinical presentations, and the exclusion of other conditions, such as multiple sclerosis (MS) [1]. Additionally, the international expert consensus criteria suggest testing for MOG-IgG only in patients who do not present with clinical and imaging features typical of MS [1].
Some key challenges related to the application of the criteria have been investigated [2].
One challenge is that patients can often have overlapping clinical and radiological features, and positive MOG-IgG results alone may not be sufficient to distinguish between MOGAD and MS [2]. In one study, researchers led by Professor Anne-Katrine Pröbstel of Bonn and Basel University Hospitals found that 5 out of 28 individuals with positive MOG-IgG results received discordant diagnoses from two independent neurologists [2]. This suggests that additional clinical investigation and paraclinical investigations, including laboratory testing, magnetic resonance imaging (MRI), and optical coherence tomography (OCT) may be required to reach an accurate diagnosis [2].
An additional challenge is the variability of results across different MOG-IgG assays, particularly in cases with low-positive results [2]. As standardised testing is still not available worldwide, repeating serum tests with different assays may strengthen diagnostic certainty when results are equivocal [2].
We discuss these challenges with Professor Pröbstel, “While the criteria are a big step forward, obstacles remain in differentiating MOGAD from MS, which is crucial given the differences in treatment. Importantly, I believe there is an urgent need to develop new biomarkers. The main goal is to standardise live cell-based assay methods, as results across different tests are not always consistent. This poses a challenge, as the same individual may receive different diagnoses at different centres, which could ultimately lead to different therapies. At the international level, there is a clear need to harmonise these assays. The goal is to achieve more consistent diagnoses and ensure that each patient receives the right treatment.”
Biomarkers of Disease Activity in MOGAD: The Role of NfL
In contrast to MS, new or enlarging magnetic resonance imaging (MRI) lesions are uncommon in MOGAD outside clinical attacks [3]. Disability is generally not driven by progression independent of relapses [1].
Professor Pröbstel tells us, “There are a number of studies investigating neurofilament light chain (NfL) as a potential marker to detect disease activity in MOGAD. The findings seem promising, as NfL appears to increase during – and even before – a clinical relapse. One of the ongoing questions is whether subclinical disease activity or progression occurs in MOGAD. In one of our studies, we identified a subgroup of patients with elevated NfL levels despite having a relapse that is resolved. This suggests that some patients may have subclinical activity, and NfL could serve as a marker to detect it.”
The multicenter study including 62 individuals with MOGAD assessed from 1995 to 2023 showed a clear increase in NfL levels in case of relapses [4].
Understanding the Mechanism Underlying MOGAD
As no therapies are currently approved for MOGAD, gaining a deeper understanding of its underlying disease mechanism is crucial [5]. The more granular the understanding of the disease, the more targeted the therapies that can be developed.
How do immune cell profiles differ between MOGAD, MS, and healthy individuals? Research shows that MOGAD has its own immune signatures, distinct from other neuroinflammatory conditions. A study – led by Professor Burkhard Becher of University of Zurich in collaboration with Professors Liblau, Korn and Pröbstel from Toulouse, Munich and Basel/Bonn – found changes in specific subgroups of immune cells – such as peripheral natural killer (NK) cells, B cells, and central memory T cells – that were more marked in people with MOGAD than in those with MS [6].
At a cellular level, researchers isolated specific B-cell receptors (BCR) from individuals with MOGAD and used them to produce monoclonal antibodies targeting human MOG, a protein involved in the condition [5]. These antibodies were shown to bind to different sites. “They can activate the complement system and other immune cells.” Says Professor Pröbstel, “This study sheds light on how these antibodies may contribute to brain damage. The findings are based on in vitro experiments, and further work in animal models is needed to understand their role in vivo. Nonetheless, they represent an important step forward. In addition, these antibodies can serve as useful tools for developing assays, as they recognise different sites on MOG, enabling further refinement of testing methods.”
Current Perspectives on Therapeutic Approaches
MOGAD may follow either a monophasic course – characterised by a single attack with no subsequent relapses – or a relapsing course [7]. Patients with monophasic MOGAD are commonly prescribed oral corticosteroids for 3-6 months – although clinical practice may vary [7]. In relapsing cases, longer-term therapy is generally considered [7]. As the risk of relapse appears to decline over time, many specialists advocate for considering discontinuation of immunosuppression following prolonged disease stability [7].
A large cohort study of 190 patients seen between 2010 and 2025, predominantly treated with steroids, evaluated outcomes following treatment discontinuation [7]. Relapses were observed after nearly 40% of treatment discontinuations, occurring at a median of around 5 months after cessation, and were more frequent among patients with a relapsing course at the time of treatment withdrawal [7]. Overall, optimal treatment duration appears to vary between monophasic and relapsing MOGAD. In monophasic cases, a duration of at least 10-18 months was estimated to be sufficient, whereas in relapsing course, treatment for 20-30 months may be considered before discontinuation in people who are relapse-free on therapy [7].
We discussed the results with Dr Wei Yeh of Monash University in Melbourne, Australia and ECTRIMS Research Fellow at the University of Oxford in 2024-25, who conducted the study, “A key unresolved question in MOGAD management is: How long do we treat people for, both after an initial attack and in those who experience relapses? This question reflects the fact that MOGAD appears to have a lower relapse rate compared with NMOSD and MS, with up to 50% of people relapsing within the first four to five years. The risk of relapse also seems to be highest early in the disease course, particularly within the first two to three years after onset. Therefore, we analysed a cohort from the Oxford Neuromyelitis Optica Highly Specialised Service to learn what might be an optimal duration of treatment before discontinuation could be considered. Based on our findings, we proposed treatment durations which aim to balance the risk of relapses with the potential risk of long-term immunosuppression.”
***
Written by Stefania de Vito
Special thanks to Professor Anne-Katrin Pröbstel (Bonn and Basel University Hospital and German Center for Neurodegenerative Diseases) and Doctor Wei Yeh (Monash University in Melbourne) for their insights.
References
[1] Banwell B. et al. The Lancet Neurology 2023; 22.3: 268-282.
[2] Lipps P. et al. JAMA Neurology 2023; 80.12: 1377-1379.
[3] Syc-Mazurek, S. al. Neurology 2022; 99(18): 795-799.
[4] Gomes A.B. et al. Neurology Neuroimmunology Neuroinflammation 2025; 12(1): e200347.
[5] Wetzel N.S. et al. Neurology: Neuroimmunology & Neuroinflammation 2025; 13(1): e200520.
[6] Schmid J. et al. Science Translational Medicine 2025; 17(819): eadw0358.
[7] Yeh W.Z. et al. (2026). Brain 2026.