Optimal strategies for treatment discontinuation in MOG antibody-associated disease

Abstract It is not known what the relapse risk is after immunomodulatory treatment discontinuation in myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD). Evidence suggests “at least” 3 months of oral corticosteroids reduces the relapse risk after a single attack and that it may be possible to stop maintenance treatment in relapsing stable disease but the optimal duration of treatment is unknown. We therefore aimed to investigate relapse outcomes following maintenance treatment discontinuation. We conducted a cohort study of MOGAD patients seen in the Oxford Neuromyelitis Optica Highly Specialised Service between January 2010 and May 2025. Patients with MOGAD, at least 12 months follow-up, and who commenced and then discontinued maintenance treatment were included. Associations of factors including treatment duration prior to discontinuation, disease course at discontinuation (after a single attack/monophasic or relapsing course) and MOG IgG1 status on live cell-based assay were investigated. Primary outcome was time-to-relapse following treatment discontinuation. Cox regression was used. We included 190 MOGAD patients with 236 discontinued treatment intervals. 150 (63.6%) discontinuations were after a single attack and before a first relapse when disease course was monophasic, and 86 (36.4%) discontinuations occurred in patients who had a relapsing disease course. Most patients used corticosteroids alone (84.7% IT intervals), and non-steroid IT were used in 15.2% of IT intervals either alone or in combination with steroids. Post-discontinuation relapse occurred after 92 (39.0%) discontinuations at a median time of 5.4 (interquartile range 1.4-20.1) months after treatment cessation. Those who relapsed were more likely to have a relapsing course at time of discontinuation (50% vs 27.8%, P=0.001) and a positive/low positive pre-discontinuation MOG IgG1 result (89.8% vs 71.5%, P=0.005). In multivariable analysis, a relapsing course at time of discontinuation was associated with an elevated relapse risk (hazard ratio 1.95, 95% confidence interval 1.25-3.06, P=0.003). Overall, prolonged treatment durations prior to discontinuation beyond 3 months significantly reduced relapse risk. Optimal treatment durations were estimated as at least 10-18 months for patients treated after their onset attack and 20-30 months for relapsing patients, following which treatment discontinuation could be considered in patients who were relapse-free on treatment. Identifying the relapse risk when discontinuing maintenance immunomodulatory treatment in MOGAD should aid management decisions in patients presenting with their first attack and also in those on longer-term treatment for relapsing disease. Our findings, from a cohort predominantly treated with steroids, provide evidence to inform joint decision-making for stable patients who are considering treatment cessation.