Interrupting rituximab treatment in relapsing-remitting multiple sclerosis; no evidence of rebound disease activity

Background: Rituximab (RTX) and other anti-CD20 therapies are increasingly used as disease modifying treatments (DMTs) in MS. However, data on reasons to interrupt treatment, alternative DMTs after anti-CD20 therapy and potential rebound disease activity are limited. The objective here was therefore to determine the rate and cause of RTX treatment interruptions and responses to subsequent DMTs in a large single centre cohort addressing also the hypothesis that there would not be rebound activity after discontinuation of RTX, regardless of reason for discontinuation and irrespective of subsequent treatments. Methods: A retrospective observational study of all relapsing-remitting MS (RRMS) patients having received at least one dose of RTX at the Karolinska University Hospital from 2009 to 2018 and having either stopped treatment or had more than one year since last RTX infusion, as identified in the Swedish MS registry with additional data derived from clinical charts. Results: As of February 2018, we identified 808 patients ever treated with RTX out of 1513 RRMS patients with current or previous DMT, 92 (11%) had terminated RTX; 27 (29%) stopped RTX due to pregnancy, 26 (28%) due to adverse events, 23 (25%) for other reasons, 9 (10%) due to stable disease and the remaining 7 (8%) due to lack of effect. The cohort of 92 patients was followed until April 2019, when 34 had restarted RTX, 27 switched DMT, 24 remained without DMT and 7 were lost to follow up. Of the 7 patients terminating RTX due to lack of effect, 4 started ofatumumab, 2 had autologous hematopoietic stem cell transplantation and 1 was lost to follow up. In all of the 92 patients, after initial RTX discontinuation, only 3 patients had relapses and 4 had new T2 lesions (one of which had both). Gadolinium was administered in 78% of follow up magnetic resonance imaging (MRI) with no enhancing lesions found (mean MRI follow up from RTX discontinuation 29 months, range 7-92 months, n = =77). Conclusion: Findings are consistent with a low rate of RTX interruptions, with pregnancy and adverse events as most frequent reasons. A small proportion of patients switched due to breakthrough disease in context of incomplete B-lymphocyte depletion. Signs of ongoing disease activity in the remaining group was low regardless of whether a new DMT was started. These findings are consistent with a long acting effect of RTX in RRMS and absence of rebound disease activity phenomena upon stopping therapy.