A Global Data Sharing Initiative study reported that multiple sclerosis (MS) patients taking certain anti-CD20 disease modifying therapies (DMTs), specifically ocrelizumab or rituximab, with suspected or confirmed COVID-19 have increased risk of hospital admission, treatment in the intensive care unit (ICU), and ventilation for patients on compared with other MS drugs, but are not at all associated with death. This study extracted data from the largest database of patients with COVID-19 and MS to date, and was presented at a special session on multiple sclerosis and COVID-19 at a “Encore” event on September 26 as part of the MSVirtual2020: 8th Joint ACTRIMS-ECTRIMS Meeting [1], by epidemiologist Steve Simpson-Yap, PhD, MPH, of the University of Melbourne in Australia (Melbourne School of Population and Global Health, Australia). Physician’s Weekly interviewed Dr. Simpson-Yap to gain perspective into this study.

The Global Data Sharing Initiative findings were simultaneously published in the Multiple Sclerosis Journal [2]. The team of researchers examined rates for 4 outcomes: hospital admission, admission to the ICU, the need for artificial ventilation, and death. A total of 1,540 MS patients from 21 countries had confirmed (50.4%) or suspected COVID-19 (30.9%) infection, with the remaining 18% being considered non-infected. The data were adjusted for age, gender, MS type, and Expanded Disability Status Scale (EDSS) to acquire prevalence ratios (aPR) for the predefined outcomes.

The results indicated, that compared with patients taking dimethyl fumarate, admission to the hospital was 1.6 times more likely to occur for MS patients taking anti-CD20 B-cell depleting therapy rituximab, and although numerically higher for those taking anti-CD20 agent ocrelizumab (aPR 1.19), the latter did not reach statistical significance. Similarly, MS patients taking certain anti-CD20 DMTs were also more likely to be admitted to the ICU (aPR 4.12 for rituximab; 3.53 for ocrelizumab), and were much more likely to require artificial ventilation (aPR 7.27 for rituximab; 3.17 for ocrelizumab).

Combining the data from patients taking rituximab or ocrelizumab – which have the same exact mode of action, unlike newer anti-CD20 DMTs – likewise resulted in significantly higher frequencies of  3 of the measured outcomes when compared with all other DMTs lumped together (hospital admission aPR 1.49; ICU admission aPR 2.55; need for ventilation aPR 3.05) as well as when compared with natalizumab (hospital admission aPR 1.99; ICU admission aPR 2.39; need for ventilation aPR 2.84). Natalizumab was used as a comparator to control for ascertainment bias, based on physician contact moments. No association was detected with any disease-modifying treatment and the fourth outcome, death. Importantly, associations persisted when limited to confirmed COVID-19 cases and upon exclusion of each contributing data source in turn.

These data build on two earlier reports. The Italian MuSC-19 study, published as a preprint in The Lancet, showed similar results in a cohort of 784 patients with MS and COVID-19, suggesting that anti-CD20 agents rituximab or ocrelizumab doubled the risk of severe COVID-19 outcomes after adjusting for other factors (odds ratio 2.59; P=0.002)[3]. Shortly thereafter, the French Covisep registry results were published in JAMA Neurology [4].  The Global Data Sharing Initiative findings confirmed results from both previous studies that increasing age, male sex, increasing level of disability, and progressive disease are associated with more severe COVID outcomes in MS patients.

When asked by Physician’s Weekly to contextualize this study with regard to the two previous studies, Dr. Simpson-Yap replied: “Our big finding was that certain anti-CD20 DMTs, particularly rituximab and to a somewhat lesser extent ocrelizumab, were associated with higher frequencies of hospitalization, ICU admission, and requiring artificial ventilation among people with MS. This association was seen for both suspected and confirmed COVID-19 cases. Although our results are generally in keeping with two other cohort studies, Covisep and MuSC-19, we had a couple of benefits from our sample size that allowed us to do the analyses with better statistical power,” says Dr. Simpson-Yap.

“We found that rituximab was associated with significantly higher frequencies of hospitalization, ICU admission, and ventilation. Ocrelizumab did not reach statistical significance for hospital admission or ventilation, but it did reach significance for ICU admission. However, we always saw consistent, albeit weaker, positive associations for ocrelizumab compared with rituximab, and this we think is important, because they both have the exact same mode of action, they actually both bind the exact same epitope on CD20.”

Why were the data stronger for rituximab?

“The reason that we think that the ocrelizumab signal might be weaker is potentially because the antibodies are different; one of them is chimeric, the other is humanized. The different constant regions of the antibodies may influence how cells bind, or complement interactions. Alternatively, there is some data that rituximab binds more strongly to the epitope than ocrelizumab. B-cell depletion could adversely impact the immune response to COVID-19, therefore if rituximab binds more strongly, and removes more of the B-cells, that could potentially explain why we see a stronger signal for rituximab. This hypothesis needs to be assessed further, but the fact that we see the consistent results with rituximab and ocrelizumab that differ only in magnitude is important.”

No association with death or comorbidities

“The big difference in our results from the previous studies is that we did not show an association at all with death. The Covisep study found those that those MS patients taking anti-CD20 DMTs had a higher frequency of severe COVID-19 outcomes, including death [4]. The MuSC-19 study evaluated the individual DMTs versus dimethyl fumarate in the sensitivity analyses, and they found an association between anti-CD20 therapy and poor outcomes [3]. However, MuSC-19 did their main analysis by pooling all severe outcomes, so hospitalization requiring artificial ventilation and death as a single yes/no dichotomous outcome. But in a sub-analysis in the supplementary information of the manuscript they evaluated death alone, and they showed a positive trend. We did show that older age, higher disability, and progressive MS were associated with death risk, but not with DMT therapy. ”

“One thing we did not control for in our main analyses, which both the Covisep and MuSC-19 studies did control for, was comorbidities such as cardiovascular disease or high BMI. Because these are factors that could potentially lead to a greater risk of COVID-19 severity, it is important test them as potential co-founders. However, in sub-analyses where we could access that data, we found the anti-CD20 DMT associations remained robust, so we could conclude that the results were not confounded by these factors.”

Shared decision-making

“The observational data we analyzed should be taken into consideration by neurologists and patients in terms of the decision-making for which DMT may be the best for their situation. If the patient falls into a group of people with an underlying risk for greater COVID-19 severity, such as those in the highest age group, progressive MS type, increased disability, or those with comorbidities, those people may want to consider a different DMT. We cannot make any clinical recommendations because that would not be appropriate for the data type, but our study, alongside the Covisep and  MuSC-19 studies, should be considered during the decision-making process.”



  1. Simpson-Yap S, et al. 8th Joint European Committee for Treatment and Research in Multiple Sclerosis-Americas Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS-ACTRIMS) 2020: Session SS02. Presented September 26, 2020.
  2. Peeters LM, et al. COVID-19 in people with multiple sclerosis: A global data sharing initiative. Mult Scler. 2020 Sep;26(10):1157-1162.


  1. Sormani, MP et al, Disease Modifying Therapies and COVID-19 Severity in Multiple Sclerosis (6/15/2020). https://ssrn.com/abstract=3631244 or http://dx.doi.org/10.2139/ssrn.3631244
  2. Louapre C, et al. Clinical Characteristics and Outcomes in Patients With Coronavirus Disease 2019 and Multiple Sclerosis. JAMA Neurol. 2020 Jun 26;77(9):1–10. https://jamanetwork.com/journals/jamaneurology/fullarticle/2767776