Among all non viral encephalitis, myelin oligodendrocytes glycoprotein antibody associated diseases (MOGAD) are the second most frequent diagnosis in children. Risk of relapses varies according to studied cohorts and cognitive and academic difficulties are more and more detected in children without knowing if these sequelae are related to the first attack or relapses. The hypothesis is that earlier treatment would induce reduction of sequelae after the first attack and the number of relapses which would be also associated with a subsequent reduction of disability occurrence on the long term.

Acquired Demyelinating Syndrome (ADS) are rare immune-mediated central nervous system (CNS) disorders that contribute to significant neurological morbidity in Europe and worldwide. ADS can affect adult and children, and the clinical spectrum is heterogenous. It begins as monophasic diseases, as often observed in acute demyelinating encephalomyelitis (ADEM), optic neuritis (ON) and transverse myelitis (TM); or as multiphasic disease with relapses such as multiple sclerosis (MS) or neuromyelitis optica spectrum disorders (NMOSD).

MOG-Abs-ADS are a frequent disease in children with ADS with specific clinical and MRI pattern.

Recently, myelin oligodendrocyte glycoprotein (MOG) antibodies (MOG-Abs) have been found implicated in ADS affecting children with a frequency up to 30%-45% according to tested cohorts.

Studies have shown that in MOG-Abs positive ADS, the most frequent clinical phenotype in children are ADEM-like presentations (ADEM, ADEM-optic neuritis, multiphasic demyelinating encephalomyelitis (MDEM) and encephalitis) and optic neuritis (ON). These presentations may vary according to the age and young children < or = 10 years old, most often, present with ADEM, whereas optic neuritis is the most common feature in children older than 10 years of age. Fewer patients present with myelitis, brainstem features, or the more recently described encephalitis with steroid-responsive seizures11.

Cerebro-spinal fluid parameters are normal in most patients, although up to 50% had pleiocytosis or increased protein levels. Oligoclonal bands were present in only 10% of patients and their presence in the context of relapsing disease would rather suggest MS 5.

MRI lesions are particular in MOG-Abs positive ADS: lesions are distributed in the cortex, subcortical region, deep white matter, brainstem, thalamus, basal ganglia and are characterized by poor demarcation and large size. These features have been found particularly in ADEM and MDEM and when compared to MOG-Abs negative ADS, large, hazy and bilateral lesions, an absence of small lesions and/or well-defined lesions, and involvement of more anatomical areas, often with longitudinally extensive transverse myelitis have been described. Absence of corpus callosal lesions, presence of cerebellar peduncle and leukodystrophy-like lesions were also observed in MOG-Ab-positive cases. MOG-Ab-associated optic neuritis is characterized on MRI by optic nerve head swelling, retrobulbar involvement, a long lesion length and, frequently, bilateral involvement.

MOG-Abs-ADS are highly relapsing and may be associated with disabilities More than 40% of MOGAD relapse mimicking diseases such as MS or anti aquaporin 4 (AQP4-Abs) positive NMOSD with an higher relapse rate 6. In the French national retrospective study on anti-MOG-Abs positive ADS, it was observed that relapse occurred in 46% of included children with a median time of 10 months (range :1-305 months) after onset, results that are in line with studies in other European countries 9.

Relapses are frequently observed during steroid weaning or within 2 months of steroid withdrawal 7, 9, 16-18. Most relapses occur in adults who are being treated with prednisolone <10 mg daily (range 0-25 mg) or in children who are receiving prednisolone <0.5 mg/kg daily. The duration of treatment might also be important. Patients whose treatment lasts for less than 3 months are twice as likely to relapse as those who are treated for longer, suggesting the need for a biomarker for the response to treatment. Several studies suggest that relapses occur in patients who remain seropositive despite treatment particularly in patients who often have high initial antibody titres.

Studies in optic neuritis, the most common neurological symptom in MOG-Abs-ADS have demonstrated that severe damage to the optic nerve can be observed in MOG-Ab-associated disorders. The observed damage seems to be driven by the frequency of attacks in patients who recover from the initial episode suggesting the need of long term treatment. Moreover, although most patients with MOG-Ab-associated disease recover well from attacks, it has also been shown that up to 45% can be left with severe disability. Importantly, >70% of this disability results from the onset attack, suggesting that there is room for improvement in the acute management of MOG-Ab-associated disease, and that time to treatment might be important for the prevention of permanent disability.

Recently, Bicêtre's hospital medical team have performed a multi-national collaborative study on 102 children with relapsing MOG-Abs-ADS (RADS) through a European Consortium coordinated by the coordinator of this project. In that study, it was observed that although the majority of children had low level clinical disability, 20% of affected children had cognitive impairment and investigators confirmed these results in the national cohort study where they have observed that age at onset < or = 10 years (OR, 3.72, 95%CI 1.19-11.64; p=0.024), ADEM at onset (OR 52.5, 95%CI 5.97-461.4; p<0.001), and deep grey matter lesions (OR, 17.33 95%CI 3.87-77.72; p<0.001) were associated to the presence of cognitive difficulties.

Studies on prospective treatment of MOGAD are rare and current treatment is based on clinical experience with similar antibody-mediated diseases, such as AQP4-Abs positive NMOSD. Corticosteroids, intravenous immunoglobulin, immunosuppressive drugs (such as mycophenolate mofetil, azathioprine and methotrexate) and rituximab are actually used and are associated with a reduction in annual relapse rate. By contrast, immunomodulatory treatments for MS, such as interferon-β and glatiramer acetate, are ineffective. In a study of children and adults in Australia, maintenance treatment with oral prednisolone was associated with fewer treatment failures than were other treatment modalities, whereas in the European collaborative retrospective study of children with MOG-Abs, investigators were able to confirm that all treatment modalities reduced annualized relapse rates but had limited effects on disability. Due to the immense vacuum in the treatment strategies, investigators have also established established an European consensus on treatment and emphasized on the absolute need for a randomized controlled treatment trials to find an optimal therapeutical strategy in order to reduce risk of sequelae after the first attack and relapses that could reduce disabilities.

Inclusion Criteria:

• Children < 18 years old and ≥ 6 years old at baseline
• Children weight ≥ 20 kg
• All ADS with confirmed anti-MOG-Abs at onset including any acute neurologic symptom with a duration of more than 24H of inflammatory causes (including optic neuritis, transverse myelitis, rhombencephalitis, ADEM, NMOSD) Without any previous treatment other than steroids
• Informed consent signed by both parents and the child
• Expanded Disability Status Scale (EDSS) < 5.5
• Affiliated to French social security regime

Exclusion Criteria:

• Current infection with SARS-COV2 (positive PCR)
• Any prior allergy to azathioprine or rituximab with hypersensitivity to active substances, murine proteins or to any of the excipients.
• Any prior history of uncontrolled cancer during the last 2 years
• Uncontrolled infections (Hepatitis B, C and HIV)
• Any prior history of cardiac dysfunction and/or hypertension
• Any progressive or non-relapsing form demyelinating diseases
• Any previous treatment with natalizumab, daclizumab, fingolimod, methotrexate, cyclosporine, mycophenolate mofetil, rituximab in the last 6 months, or determined by the treating physician to have residual immune suppression from these or other immunosuppressive treatments
• CD4+, CD8+, or CD19+ absolute cell count, wbc, neutrophiles in blood at screening below lower limits of normal (LLN)
• Creatinine>30µmol/L
• Platelets <70 000mm3
• Haemoglobin < 8g/dL
• Acute renal insufficiency (clearance < 30 ml/min)
• Prior documented history of hemostase perturbation (TP and/or TCA more than twice of the witnesse's TP and/or TCA)
• Prior documented history of increased liver enzyme level (ASAT and/or ALAT) > 2N.
• TP <70%
• Total bilirubin > 2N
• Any patient with allopurinol treatment and immunosupressive treatment with concomitant use of xanthine oxidase inhibitors (e.g. allopurinol, oxipurinol/thiopurinol, febuxostat)
• Patients with two inactive TPMT or NUDT15 alleles (homozygous deficient or double heterozygote)
• Pregnancy or lactating woman or wish for future pregnancy
• Refusal to have a highly effective contraception during traitment and for one year (12 months) after the end of the experimental treatment
• participation to another interventional study within 5 half-lives prior to baseline.
• Active, severe infections (including tuberculosis, HBV and HCV, HIV, herpes, VZV, EBV and CMV)
• Psychosis not controlled by treatment
• Patients with Lesch Nyhan syndrome
• Pheochromocytoma
• Scleroderma
• Untreated peptic ulcer
• Myasthenia gravis
• Any other medical illness or disability that, in the opinion of the investigator, would compromise effective trial participation

The adjudication committee is responsible the evaluation of all relapses reported during the the follow-up period. Relapses assessments will use blinded data and will be performed in real time.