Predicting the risk of long-term relapse in MOGAD

Introduction: Approximately half of patients with MOG-antibody associated disease (MOGAD) relapse. It is not clear whether early relapsing activity (within 1 year of initial attack) indicates risk of chronic disease. We aimed to identify whether early relapse, along with predictors of early relapses, predicts relapse beyond one year. Methods: A prospective cohort of 192 MOGAD patients from 5 UK centres with at least 2 years’ disease duration and no long-term immunosuppressive treatment during the first year were included. One hundred and eighteen (61%) experienced relapses at any stage. Univariate Cox regressions for time-to-relapse after the first year were performed with covariates including the presence and number of early relapses, the timing of early relapse (categorized in three-month epochs), decision to treat with corticosteroid and duration of taper (days), age at onset, self-identified race, onset phenotype and seroconversion to MOG antibody-negative. Multivariate Cox-regression models included all previously significant variables, as well as MOG-antibody seroconversion. Results: Out of 118 relapsing MOGAD patients, we found 49 (25.5%) patients with early relapses with a median number of 1 (range 1-4) relapses. Univariate analysis revealed an increased risk of long-term relapse with any early relapse (Hazard Ratio [HR] 1.60, CI 1.16-1.89), relapses in the first (HR 2.24, CI 1.27-3.92) or the last 3-month epoch (HR 2.66, 1.22-5.78), higher numbers of early relapses (HR 1.48, CI 1.16-1.89) and non-white race (HR 1.94, CI 1.08-3.48). Longer duration of early corticosteroid treatment (HR 0.99days, CI 0.996-0.999, p=0.035) was protective. Following multivariate regression analysis, the number of early relapses and relapse in the first (HR 2.51, CI 1.19-5.27) and third 3-month epochs (HR 3.71, CI 1.52-9.06) were significant, as were days on corticosteroid treatment (HR 0.99days, CI 0.99-0.99; HR 30 days 0.74) and non-white race (HR 2.69, CI 1.45-5.11). There was an 84% and 99% specificity for long-term relapses in the first 5 years in those with ⩾1 and ⩾2 early relapses respectively, with low sensitivity (20% and 5%). Conclusion: Our results suggest that early relapses are associated with increased risk of long-term relapsing disease and may indicate a need for long-term immunosuppression as they do not appear to solely reflect early inflammatory phase. Early corticosteroid treatment may have a protective role in preventing relapses beyond one year.

Citation

Francis, A., Gomez, E., Chen, B., Woodhall, M., Blain, C., Cooper, S., Dobson, R., Fisniku, L., Halfpenny, C., Hobart, J., Jacob, A., Martin, R., Messina, S., O'Sullivan, E., Ramdas, S., Robertson, N., Satukijchai, C., Wassmer, E., Waters, P., Williams, V., Wu, Y., Hemingway, C., Ming, L., Huda, S., Geraldes, R., Leite, M. I., & Palace, J. (2022). Predicting the risk of long-term relapse in MOGAD. Multiple Sclerosis Journal. https://doi.org/10.1177/13524585221123687

DOI

10.1177/13524585221123687

URI

https://hdl.handle.net/20.500.14709/706