Reply to “Comparative Safety Profiles of Ocrelizumab and Rituximab in Multiple Sclerosis Treatment Using Real‐World Evidence”

We sincerely thank the authors for their constructive comments on our recently published study of the comparative safety profiles of ocrelizumab and rituximab in the treatment of multiple sclerosis using real-world evidence.1 We also welcome the opportunity to further clarify aspects of our methodology and some perceived inconsistencies in our analysis. In response to concerns about our statement on the paucity of robust evidence to guide the use of rituximab in the treatment of multiple sclerosis (MS), we would like to emphasize that our study only concerned head-to-head differences in the long-term safety profiles, not efficacy, between rituximab and ocrelizumab. Our comment also referred to the absence of phase 3 registration trials for rituximab, as exist for ocrelizumab (NCT01247324, NCT01412333, and NCT01194570), and more recently for ofatumumab (NCT02792218 and NCT02792231) and ublituximab (NCT03277261 and NCT03277248). The alleged lack of control for calendar effects and incomplete adjustment for medical history, comorbidities, and socioeconomic factors is inaccurate. Our study controlled for temporal bias by including a continuous variable representing the time of therapy initiation and by conducting a subanalysis restricted to patients who initiated rituximab after the US Food and Drug Administration (FDA) approval date of ocrelizumab. In this subanalysis, both groups were followed up within the same time period, eliminating any time misalignment. Furthermore, it is reasonable to hypothesize that such a time imbalance would have favored rituximab, as it was the only B-cell-depleting therapy available during the period 2014 to 2017 (before the coronavirus disease 2019 [COVID-19] pandemic) and thus should have shown fewer infection-related complications. In addition, our analysis incorporated adjustment for over 20 covariates, including age, sex, race and ethnicity, disease duration from earliest recorded diagnosis, time since last relapse, disease course at the time of initiating ocrelizumab or rituximab therapy, previous disease-modifying therapy (DMT) treatment, cane, wheelchair, and crutches (University of California – San Francisco [UCSF]) or just wheelchair (UC-wide dataset), basal metabolic index (BMI), heart failure, hypertension, diabetes mellitus (types 1 and 2), smoking status, chronic obstructive pulmonary disease (COPD), insurance type, and Social Determinants of Health. In line with modern analytical pipelines, we used a doubly robust approach, adjusting for these variables through both covariate adjustment and propensity score methods.2 We commend and support efforts to increase the affordability of therapeutic agents; however, any cost–benefit analysis should also consider safety as a key factor. We further caution that at least some of the ongoing studies (eg, NCT04578639) are likely underpowered (n < 250) to detect rare or severe infections, as they primarily include newly diagnosed patients with minimal disability and a follow-up period of only 2 years. In practice, the safety profiles of medications are only partially established in the setting of randomized clinical trials (RCTs). Large observational datasets collected over many years in the clinical practice setting, such as those presented in our article, are superior to RCT data for assessing safety for precisely these reasons. Nevertheless, we look forward to learning the results of these studies that will hopefully further clarify the respective safety profiles of the available CD20-depleting agents. All authors contributed to this reply. G.C. has nothing to report. S.L.H. currently serves on the scientific advisory board of Accure, Alector, Annexon, and Hinge Therapeutics and has received travel reimbursement and writing support from F. Hoffmann-La Roche and Novartis AG for anti-CD20-therapy-related meetings and presentations. These companies had no input in the work presented here. B.A.C.C. has received research support from Genentech and Kyverna. These companies had no input in the work presented here. S.E.B. has nothing to report.