Plus Minus Chat Login Arrow right Chevron left Chevron right Close Close circle Lock Apple Windows Compare Arrow Up Right Book Lightning Flag Arrow Right Chart Bar Wavy Circle Check Cube Envelope Graduation Cap Info Link List Numbers List Pencil Line Star Table Profile Youtube Twitter Facebook LinkedIn Google Plus Box Speech Bubble Television Icon Arrow Circle Right Search Lightbulb Link Out Select Arrows Apple Podcasts Spotify Google Podcasts Amazon Music

NMOSD and MOGAD: Considerations about Diagnosis and Treatment

min read

Early reports of neuromyelitis optica (NMO) can be traced back to the 19th century. In 1844, an Italian man of 42 years old from Sardinia was sentenced to forced labour for murder. Afflicted by acute myelitis, the man suffered simultaneously from a debilitating pain along the spinal cord and complete blindness [1]. Giovanni Battista Pescetto, a Genoese physician, treated him with bloodlettings, 24 leeches along the spine, emetic drugs, and olive oil; and described the case in an article titled Storia di un caso di noteomielite acuta, accompagnata da amaurosi [1, 2].  In the late 19th century, Eugène Devic defined the condition characterised by optic neuritis and acute myelitis as “neuro-myélite optique” [3].

The story of MOGAD (Myelin Oligodendrocyte Glycoprotein Antibody-Associated Disease) is more recent. There has historically been difficulty in defining a clinical phenotype associated with MOG antibodies, and whilst early work using denatured antibodies indicated that these antibodies could be a marker of multiple sclerosis (MS), this never reached clinical practice [4, 5]. However, the development of modern assays has enabled the description and extension of strongly associated clinical phenotypes [6].

NMOSD (Neuromyelitis Optica Spectrum Disorder) and MOGAD are rare disorders worldwide. NMOSD is more prevalent in middle-aged Afro-Caribbean and Asian individuals, particularly women. MOGAD affects people of all ages, with no clear predominance for sex, race, or ethnicity [7, 8].

Differential diagnosis between NMOSD and MOGAD

“There are different treatments for the three conditions of NMOSD, MOGAD, and MS, so it is very important to be able to identify them,” Dr. Eoin Flanagan, an expert in spinal cord diseases based at the Mayo Clinic in Minnesota, tells us. “The available treatments for MS do not tend to work for NMOSD and MOGAD and can sometimes even make it worse. One of the challenges is that all three diseases are associated with demyelination in the central nervous system (CNS). They can present with inflammation of the optic nerves, namely optic neuritis, inflammation of the spinal cord, i.e., myelitis, or inflammation in the brain, which can be sometimes termed encephalitis.”

Biological, clinical, and neuropathological evidence allow to recognise NMOSD and MOGAD as separate conditions with distinct diagnostic criteria [9]. They have different biomarkers. In most cases, NMOSD is linked to pathogenic immunoglobulin G (IgG) antibodies targeting aquaporin-4 (AQP4), water channel in the CNS [10]. MOGAD is, instead, associated with MOG-IgG [7]. The presence of AQP4-IgG, even at low levels, is highly predictive of a diagnosis of NMOSD. Using cell-based assay (CBA) technique and analysis in serum, these biomarkers are sensitive – they correctly classify as positive individuals with the disease – and highly specific – they correctly detect as negative individuals who do not have the disease [7]. When the levels of MOG-IgG are low, one must be cautious, as low levels can also be found in 1%-2% of individuals with other conditions [7].

In a study with 493 children with MS and 703 healthy individuals, AQP4-IgG was negative among all 1,196 individuals [11]. MOG-IgG was also negative in all the healthy individuals, but positive in 30 children with MS. Eventually, among these 30 children, 25 were diagnosed with MOGAD and 5 maintained their initial MS diagnosis [11].

Dr Flanagan says, “In contrast with MS, NMOSD, and MOGAD can be diagnosed with a single blood test. AQP4-IgG is almost a perfect test. With the cell-based assays (CBA) we rarely see any false positive. Instead, one of the challenges of the MOG-IgG is that it is a bit sticky, so we do see false positives at low levels in some individuals with MS or other conditions. When a young child presents with demyelination, MOGAD antibodies need to be checked, as MOGAD affects children more.”

More than 50% of cases in children younger than 11 years with an initial presentation of acute demyelinating syndrome test positive for MOG antibodies [8].

NMOSD and MOGAD share core clinical features: the inflammation of the optic nerve and spinal cord. The two primary clinical features of NMOSD consist of attacks of acute transverse myelitis and acute optic neuritis, which may manifest at the same time or one after the other [10]. NMOSD derives its name from this combination of myelitis and optic neuritis. When individuals present with a suspected encephalitis or brainstem encephalitis, NMOSD should be also considered [10].

In adults, optic neuritis is the most common initial symptom of MOGAD. Children before the age of 11 usually present at onset with acute disseminated encephalomyelitis (ADEM), with or without optic involvement. Transverse myelitis is also a common initial symptom [8].

Magnetic resonance imaging (MRI) helps differentiating between NMOSD and MOGAD. Unlike MS, optic nerve involvement is more commonly bilateral in both AQP4+ NMOSD and MOGAD [12]. Dr. Laura Cacciaguerra, also of the Mayo Clinic, tells us, “The major overlaps might occur between MOGAD and NMOSD, because in both cases we can have bilateral optic neuritis. However, an involvement of the optic chiasm is more typical of NMOSD, while anterior optic neuritis, sometimes with optic disc oedema, is more suggestive of MOGAD. On the other hand, in both NMOSD and MOGAD, we typically observe longitudinal extensive lesions in the spinal cord, which involve at least 3 contiguous vertebral segments. However, there are some differences that can be considered. Multiple lesions are more common in MOGAD than NMOSD. In NMOSD, we typically observe one single longitudinal lesion involving the entire transverse diameter of the spinal cord. On the contrary, single, or multiple lesions in the terminal part of the spinal cord – the conus – are characteristic of MOGAD. Most of the times, in MOGAD lesions affect the grey matter and are H-shaped. In the brain, individuals with MOGAD have more lesions than individuals with NMOSD. Additionally, the appearance of the lesions differs in the two conditions. In MOGAD, lesions are described as fluffy, whereas in NMOSD they are linear and mostly observed in regions rich with AQP4.”

An important aspect is the resolution of brain lesions.  “I think it is important to remember that `multiple sclerosis` means multiple scars. Every MS lesion forms a scar, namely leaves a plaque in the brain,” Dr. Flanagan says. “With MOGAD, 70%-80% of lesions disappear completely. That is a huge difference in the disease process.” Complete resolution is rare in both NMOSD and MS [8].

Treatment in NMOSD and MOGAD: acute and preventive

Treatment in MOGAD and AQP4+NMOSD aims to facilitate recovery following an acute attack – acute treatment – and to prevent relapses – maintenance treatment [7]. Acute treatment in MOGAD and AQP4+NMOSD consists in intravenous steroids and plasma exchange [7]. Delaying the start of treatment – even by as little as 7 days – can have harmful effects [13].

Dr. Flanagan says, “One of the challenges with the acute treatment in both AQP4+NMOSD and MOGAD is that sometimes we need to treat the person before having the results of the antibody tests back. Therefore, if there is a strong suspicion of NMOSD, before the antibody test comes back, one should probably try to move to the plasma exchange treatment early on in conjunction with steroids. For what concerns the maintenance treatment, the therapies for NMOSD are today very effective. NMOSD was a devastating diagnosis before we had treatments. It was a severe disease, with a 33% mortality in 5 years. Nowadays, individuals with NMOSD can live a long, normal life, because the available treatments are highly effective in preventing relapses. Furthermore, the World Health Organization (WHO) has recently added rituximab to the Essential Medicines List (EML). Rituximab – an anti CD20 monoclonal antibody – represents one of the highly effective maintenance therapies for individuals with AQP4+NMOSD. The hope is that it will become more affordable and available in many countries across the world, especially in limited-resource settings. For MOGAD, we still miss a proven treatment. Some people with MOGAD have just one episode in their life and are never going to have another episode. Therefore, we usually tend to wait the second attack before starting a long-term immunosuppressant therapy. However, it is important to treat people with MOGAD, as it affects the optic nerves and can lead to vision loss in the long term. There are two open clinical trials right now. It is crucial to enrol individuals with MOGAD in clinical trials to be able to replicate the success we have had with MS and NMOSD.” Highly efficient treatments are needed to improve the quality of life of people with MOGAD.

Written by Stefania de Vito

Special thanks to Dr. Eoin P. Flanagan and Dr. Laura Cacciaguerra (Mayo Clinic Center for MS and Autoimmune Neurology) for their insights.


[1] Jarius S, Wildemann B. J. Neurol. Sci. 2012; 313: 182-184.
[2] Pescetto GB. Giornale delle Scienze Mediche della Società Medico-Chirurgica di Torino; 1844; p. 311–24.
[3] Jarius S, Wildemann B. J. Neuroinflammation 2013; 10: 8.
[4] Berger T et al. NEJM 2003; 349(2): 139-145.
[5] Moheb N and John JC Eye 2023; 37(12): 2391-2398.
[6] O’Connor KC et al. Nat. Med. 2007; 13(2): 211-217.
[7] Cacciaguerra L and Flanagan EP. Neurol. Clin. 2024; 42(1): 77-114.
[8] Banwell B et al. Lancet Neurol. 2023; 22(3): 268-282.
[9] Marignier R et al. Lancet Neurol. 2021; 20(9): 762-772.
[10] Jarius S et al. 2023 J. Neurol. 2023; 270 :3341-3368.
[11] Gaudioso CM et al. Ann. Neurol. 2023; 93(2): 271-284.
[12] Ramanathan S et al. Mult. Scler. J. 2016; 22(4): 470-482.
[13] Stiebel-Kalish H et al. Neurol. Neuroimmunol. Neuroinflamm. 2019; 6: e572.