To compare rituximab to ocrelizumab with respect to the odds of developing an infusion-related reaction (IRR) during the first two infusions.

Ocrelizumab, a humanized anti-CD20 monoclonal antibody, is a highly effective treatment approved for relapsing and primary-progressive multiple sclerosis (MS). Rituximab, a chimeric monoclonal antibody, targets the same receptor and is sometimes used off-label because of lower cost. Determining whether these drugs have similar safety profiles in a real-world setting has important implications for cost-effectiveness in MS treatment.

To limit treatment selection bias, our primary comparisons describe propensity-matched pairs of patients treated with rituximab vs. ocrelizumab.  Our propensity score (PS) used logistic regression to model the likelihood of rituximab initiation (vs. ocrelizumab) using age, disease duration, sex, race, disease course, number of prior MS therapies, most effective prior MS therapy, JCV serology, need for walking assist devices, and presence of baseline gadolinium-enhancing lesions. Covariate balance was assessed by exploratory comparisons of standardized differences before and after use of propensity-based matching and weighting techniques. 1:1 matching with replacement on the propensity score yielded strong covariate balance, and was used to compare the frequency of patients developing an IRR using logistic regression.

We enrolled 340 ocrelizumab and 67 rituximab treated patients. There were differences in age, number of prior MS therapies, need for walking assist devices, and presence of baseline gadolinium-enhancing lesions between groups at baseline. In PS matching with replacement, each of the 67 rituximab patients was matched to one of 52 unique ocrelizumab patients. In the unadjusted analysis, patients with IRRs were twice as likely to be treated with rituximab vs. ocrelizumab [OR=2.0, 95% CI (1.0, 3.9)]. In the matched sets, we observed similar odds of developing an IRR [OR=1.2, 95% CI (0.5, 2.8)].

Similar odds of developing an IRR on rituximab vs. ocrelizumab suggests similar safety profiles.