Myelin oligodendrocyte glycoprotein is a membrane protein uniquely expressed on the surface of oligodendrocytes and myelin in the central nervous system of humans and other mammals (Ramanathan et al., 2016; Jain et al., 2021). Myelin oligodendrocyte glycoprotein (MOG) antibody-associated disease (MOGAD) overlaps with multiple sclerosis, acute disseminated encephalomyelitis, and aquaporin 4 antibody (AQP4-Ab)-positive neuromyelitis optica spectrum disorders (NMOSD) in terms of clinical phenotype, and is now considered to be a new and independent entity in inflammatory demyelinating diseases of the central nervous system (Ramanathan et al., 2016; Cobo-Calvo et al., 2019). The demographic, clinical, and laboratory differences at the onset of MOGAD are usually age-related. Among children, the female-to-male ratio is similar (Jurynczyk et al., 2017; Cobo-Calvo et al., 2021), and presentation at onset is usually acute disseminated encephalomyelitis, especially under 10 years of age, followed by optic neuritis (ON), transverse myelitis (TM), and brainstem demyelination (Fernandez-Carbonell et al., 2016; Duignan et al., 2018). The overall prognosis is better in children than in adults, with less than 10% of motor disability and visual acuity disability after treatment, and a lower risk of relapse in children (Reindl and Waters, 2019; Cobo-Calvo et al., 2021). Compared with children, there are slightly more female patients among adults (Jurynczyk et al., 2017), and the first presentation is usually ON (as high as 50–70%) (Duignan et al., 2018). In acute attacks, high-dose intravenous methylprednisolone is used in both adults and children, and plasma exchange is preferred when recovery is incomplete. During maintenance therapy, intravenous immunoglobulin (IVIg) is the preferred first-line treatment for children, whereas azathioprine (AZA), mycophenolate mofetil (MMF), and rituximab (RTX) are the first-line treatments for adults (Whittam et al., 2020). Some studies have reported that the higher the antibody titre at the time of onset and the longer the duration of antibody positivity, the greater the possibility of relapse (Hennes et al., 2017; Jurynczyk et al., 2017). Some patients relapsed during or after the withdrawal of steroids, most of whom showed ON (Reindl and Waters, 2019); therefore, a previous study suggested that a prolonged steroid taper can reduce early relapse of MOGAD (Narayan et al., 2018).

Since MOG antibody (MOG-Ab)-related cerebral cortical encephalitis (CCE) was first reported by Ogawa et al. (2017), many cases of this rare clinical phenotype have been reported globally, which may have been diagnosed with unexplained steroid-responsive encephalitis in the early stages of the disease (Wang et al., 2021). CCE is a syndrome with an unclear clinical definition and is characterised by grey matter lesions on brain MRI, primarily involving the cerebral cortex and sulcus, but not the subcortical and deep white matter (Krupp et al., 2013; Ogawa et al., 2017; Hamid et al., 2018). In addition to fever, headache, and seizures, cerebral cortical symptoms, such as aphasia, dysarthria, paralysis, mental symptoms, and memory loss, also exist in patients with MOG-Ab-related CCE. A typical imaging feature is the hyperintensity of cortical lesions in fluid-attenuated inversion recovery (FLAIR). According to the above imaging characteristics, CCE is divided into two types: unilateral and bilateral. In this study, we report two cases of CCE with positive MOG-Ab and perform a systematic analysis of previously reported cases. The purpose of this study was to further describe the clinical features, imaging results, and prognosis of rare MOG-Ab-positive CCE.

To obtain a more thorough understanding of MOG-Ab-associated CCE, we searched and screened studies in PubMed and Web of Science until August 2021 to identify patients according to the following process (Figure 2, review flowchart). The inclusion criteria were as follows: (1) Patients who tested positive only for serum MOG-Ab during the course of the disease, and no other antibodies such as AQP4-Ab, NMDAR-Ab, and contactin-associated protein 2-Ab that may indicate other diseases. (2) Patients explicitly diagnosed with MOG-Ab-related CCE for the first time. (3) Presence of abnormal hyperintensity on brain MRI, mainly in the cerebral cortex or sulcus. (4) Availability of complete data on clinical symptoms, imaging, and treatment.

We finally obtained 33 cases from 25 articles and combined with our two cases, a total of 35 cases were reviewed systematically (Ogawa et al., 2017; Hamid et al., 2018; Ikeda et al., 2018; Budhram et al., 2019, 2020; Haddad et al., 2019; Patterson et al., 2019; Fujimori et al., 2020b; Hochmeister et al., 2020; Katsuse et al., 2020; Matoba et al., 2020; Otani et al., 2020; Russ et al., 2020; Takamatsu et al., 2020; Tao et al., 2020; Ahsan et al., 2021; Chang et al., 2021; Doig et al., 2021; Jain et al., 2021; Kim et al., 2021; Maniscalco et al., 2021; Nie et al., 2021; Stamenova et al., 2021; Tian et al., 2021; Wang et al., 2021; Table 1). The main demographic data, clinical characteristics, laboratory results, and imaging results are presented in Table 2.

Compared with the previously reported CCE, our cases were generally consistent with the previous cases in terms of clinical manifestation, imaging, and treatment. However, the rapid deterioration of patient 2 emphasizes the importance and necessity of early diagnosis and treatment. MOG-Ab-associated CCE may produce severe clinical symptoms, and even deaths have been reported. Therefore, early diagnosis and treatment are needed, which will greatly improve the prognosis of patients; otherwise, serious complications may occur. Since the most common symptoms are seizures, headache, and fever, it is easily misdiagnosed as intracranial infection, such as viral encephalitis and meningeal inflammation (Budhram et al., 2019). At present, as a new method for the diagnosis of pathogen infection, metagenomic next-generation sequencing can help improve the detection rate of encephalitis caused by pathogen infection (Wilson et al., 2019).

Although our study did not include cortical encephalitis caused by other pathogens, during the screening of the literature we found that MOG-Ab can coexist with other antibodies, especially anti-NMDAR-Ab (Zhou et al., 2017; Sinani et al., 2020; Nan et al., 2021). The new data show that NMDARs exist on the myelin sheath formed by oligodendrocytes (Lipton, 2006); therefore, NMDAR may be positive and can cause demyelination when an immune response occurs. In an analysis of 691 cases of anti-NMDAR encephalitis, 12 cases were found to be MOG-Ab-positive at the same time (Titulaer et al., 2014). When a patient has symptoms of CCE, it is difficult to identify whether it is caused by MOG-Ab or anti-NMDAR-Ab. Fortunately, their treatment principles are generally similar, consisting of HIMP or IVIg. Notably, RTX is more likely to be used to treat anti-NMDAR encephalitis with an acute onset (Dalmau et al., 2011; Dou et al., 2020; Thaler et al., 2021), but only in remission periods for maintenance treatment in MOGAD. We recommend that patients with MOGAD should be tracked and simultaneously tested for MOG-Ab and other antibodies in the serum and CSF for the early detection of an overlapping syndrome.

In a single-centre cohort study from China, 20.7% of MOG-Ab-positive patients had typical encephalitis symptoms, and 72.2% had cortical changes during the course of encephalitis (Wang et al., 2019). Seizures are the most common symptom of CCE, which may be caused by immune-mediated demyelination. Immunotherapy may be more effective than conventional antiepileptic therapy (Cobo-Calvo et al., 2018). In general, cortical damage caused by epilepsy is characterised by hyperintensity on DWI, which indicates cytotoxic oedema (Milligan et al., 2009; Chatzikonstantinou et al., 2011), while cortical damage caused by MOG-Ab-related CCE is usually indicated by abnormal hyperintensity on FLAIR. “FLAMES” is a new term used to characterise the clinical and radiological syndromes of cortical FLAIR-hyperintense lesions in anti-MOG-Ab-associated encephalitis with seizures (Budhram et al., 2019). A study has pointed out that long-term oral anti-epileptic drugs (AEDs) have no significant effect on the control of seizures, while immunosuppressant MMF can significantly reduce the occurrence of acute seizures (Yao et al., 2019). In our systematic review, since most patients received steroid therapy and AEDs at the same time, the effect of AEDs alone was unknown. Therefore, it is still not clear whether AEDs should be used for extended periods, and it should be evaluated comprehensively according to the effect of steroid application, along with the type, duration, and frequency of seizures.

In our study, 31 patients showed cortical hyperintensity on brain MRI FLAIR, which was relatively unique but not specific. In the case of atypical symptoms, they should be distinguished from other cortical diseases, such as seizures, Creutzfeldt-Jakob disease, subarachnoid haemorrhage, meningeal carcinoma, and reversible posterior leukoencephalopathy syndrome (Renard et al., 2015). Several researchers have reported seizure-induced reversible MRI abnormalities (SRMA) that are characterised by MRI abnormalities, which resolve completely at follow-up (Cole, 2004; Briellmann et al., 2005). In a systematic review from the University of Melbourne in Australia, which involved 27 patients with SRMA, 12 (44%) patients demonstrated SRMA in the cortical layer on brain MRI and 21 (78%) patients showed hyperintensity on MRI FLAIR (Mariajoseph et al., 2021). Based on the similarity between SRMA and MOG-Ab-mediated CCE, the detection of MOG antibodies is vital for accurate diagnosis.

According to the results of our review, involvement of the cortex was mostly limited to the unilateral cerebral cortex. In a systematic review, brain MRI lesions with bilateral medial and frontal cerebral cortical encephalitis were superimposed with a computer software program, showing that most of the lesions were distributed in the supply area of the anterior cerebral artery, while the lesions with unilateral frontal cerebral cortical encephalitis were distributed in the middle cerebral artery region (Fujimori et al., 2020a). Therefore, we speculated that the lesions of the cortex are involved in vascular distribution. Reviewing the cases in our study, we observed that in the 11 patients who underwent cerebrovascular perfusion tests, including MRI, arterial spin labelling imaging, and single-photon emission computed tomography, 10 patients displayed vasodilation of the cerebrovascular branch, indicating high cerebrovascular perfusion (Ogawa et al., 2017; Ikeda et al., 2018; Fujimori et al., 2020b; Katsuse et al., 2020; Matoba et al., 2020; Otani et al., 2020; Jain et al., 2021; Nie et al., 2021), which usually can be completely restored to normal after steroid therapy. Although there is no clear conclusion about the relationship between vascular distribution and lesions, we speculate that peripheral vessels of the middle cerebral artery or anterior cerebral artery may also be involved in the pathogenesis of the disease after cortical inflammation occurs.

It was found that the titre value of MOG-Ab in the serum of the patients varied greatly, which can be used to predict relapse (Tea et al., 2019; Marchionatti et al., 2021), supporting the pathogenic role of MOG-Ab. However, the MOG-Ab titre value was not entirely positively correlated with the severity and recurrence of the patient’s disease in our study, which may be related to the timing of serum collection, meaning that the titre value will decrease significantly after steroid therapy. In a patient with case 30, although the patient’s MOG-Ab titre was 1:320, it manifested itself with fulminant phenotype, and the patient died shortly due to older age and extensive cortical involvement. Therefore, for each individual, we should make a comprehensive judgment of the patient’s prognosis according to age, past medical history, and clinical characteristics.

The first-line treatment of patients includes HIMP and IVIg in the acute setting, and significant improvements in clinical symptoms and imaging can usually be observed after steroid treatment. By reviewing the patients’ prognosis, we found that the lesions resolved completely in most patients, which is significantly different from that in multiple sclerosis and NMOSD; the underlying mechanisms of which are not yet fully understood. In a retrospective cohort study from the Mayo Clinic, comparing the evolution of MRI lesions in different central nervous system demyelinating disorders, MRI normalisation of brain T2 signal abnormality was more common with MOGAD for brain attacks (MOGAD, 39%; AQP4-IgG-NMOSD, 10%; Multiple sclerosis, 5%), and the greater remyelination capacity, less neuronal loss, or a predominant functional vs. structural damage in MOGAD might account for this difference (Sechi et al., 2021).

However, some patients relapse after withdrawal therapy (Cobo-Calvo et al., 2019), and therefore, patients may be treated with immunosuppressive therapy, such as MMF, RTX, and cyclophosphamide, to prevent relapse. The patient may relapse with CCE again, or any type of MOGAD. Although the MOG-Ab titre value can be used to predict relapse, it cannot be relied upon alone. New research suggests that FNFAIP3 levels decreased at relapse compared to remission, which has clinical utility as a relapse biomarker and potential therapeutic target (Saxena et al., 2020). Steps are being undertaken toward its potential use in clinical settings.

Our study was retrospective and had certain limitations. Currently, MOG-Ab-associated CCE is still rare, the sample size is not large enough, and the results may be biased. Moreover, the follow-up time of each case varied, which may have affected the prognosis.

In conclusion, MOG-Ab-associated CCE is still a rare clinical type of MOGAD. In patients with headaches, fever, and seizures, the detection of serum MOG-Ab should be performed at the earliest, in order to make a clear diagnosis and administer timely treatment. Moreover, the regular detection of MOG-Ab titres during follow-up contributes to the early detection of disease relapse.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

The studies involving human participants were approved by the Human and Research Ethics committees of the first Bethune Hospital of Jilin University. The patients/participants provided written informed consent for the publication of any identifiable data/images presented in the manuscript.

MD and HS conceived the topic, designed the outline of this review, and wrote the manuscript. YL and JS prepared the figures and tables. LC and PS critically revised the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.