#MSCOVID19 The power of internet publishing…Swedish Experience


Today the Swedish experience of COVID-19 surfaces in the multiple sclerosis journal online, it will be about another six month to a year before it comes out in print when hopefully the pandemic is over (yer right) and you have to say who cares, we know already. Thisi paper says that if you take rituximab and catch COVID-19 that you had a small but higher risk of being hospitalised. I am sure this will make you all to want your covid-19 vaccination. If only you knew tis information in January and Februray when offered the vaccine. However, this info can be found as a preprint on SSRN since March 2021, but the basic message was posted on the blog after Prof Hillert gave the result on two iwiMS pobcast in May 2020. So it has only taken a year for this important information to surface.


This new data forms the basis of some of the data by the Interanational Community published by Simpson-Yap et al. months ago. This has still yet to surface. Maybe the damning conclusions against a type of anti-CD20 are being made harder to spot like the Italian data 🙂


Now which anti-CD20 is this work about?


The Swedish data is largely about rituximab so both ant-CD20 create a similar issue.

However, it shows the standard publication route is rather slow to get the important information out. There must be quite alot of data on CD20 and vaccination and by colating it it may be possible to say what to do to increase your chances of making a stronger antibody response. There will not be a perfect answer because the variability on how you repopulate your B cells is massive (about 5.5 months to 3.25 years)

Maybe they tried to publish the data earlier but the “Science police” refereeing the paper said (a) This is about rituximab and not ocrelizumab so it it is unlicenced so we should not publish this until you have ocrelizumab data. What type of a referee would say such a thing? ……….Answers on a post card, because we can’t publish my answer…….but like we have ant-Vaxxers there are clearly anti-science referees…..that rather than say nothing have to say some destructive rubbish (b). This is based on symptomatology and not PCR confirmed cases. Clearly at the start of the pandemic the tests were not available so we went from symptoms, but I dont know if you can remember Jan-Mar 2020….every body had a sniffle and a snuffle we all thought we had COVID but it was unproven. I bet in many cases it was not COVID-19 but something else…but thanks another story.

I guess one thing I ask why is the risk from hospitalisation with rituximab (3 times more. below and also see Simpson-Yap et al.) worse than with ocrelizumab (2 times)….it doesn’t make biological sense based on B cell depleting potential and therefore says to me hospitalisation risks are influenced by something in addition, is it age or progression etc. e.g. you cant have ocrelizumab because of the label because you are progressive and non-active so you go rituximab off-label but disability and not the CD20is the key/

Increased rate of hospitalisation for COVID-19 among rituximab-treated multiple sclerosis patients: A study of the Swedish multiple sclerosis registry.Spelman T, Forsberg L, McKay K, Glaser A, Hillert J.Mult Scler. 2021 Jul 2:13524585211026272. doi: 10.1177/13524585211026272

Background: The primary objective of this study was to analyse the association between multiple sclerosis (MS) disease-modifying therapy (DMT) exposure and hospitalisation in patients infected with COVID-19.

Methods: Associations between MS DMT exposure and COVID-19 hospitalisation were analysed using univariable and multi-variable-clustered propensity score weighted logistic regression, where the models were clustered on the individual patients to control for patients contributing multiple COVID-19 episodes.

Findings: As of 18 January 2021, a total of 476 reported COVID-19 cases had been recorded in MS patients in the Swedish MS registry. Of these, 292 (61.3%) had confirmed COVID-19. The mean value (standard deviation (SD)) age at infection was 44.0 years (11.6). Of the 292 confirmed infections, 68 (23.2%) required hospitalisation. A total of 49 of the 164 confirmed COVID-19 patients on rituximab at baseline (29.9%) required hospitalisation, compared to a rate of 12.7% for all other DMTs combined. Rituximab in confirmed COVID-19 patients was associated with 2.95 times the odds of hospitalisation relative to any other DMT combined (odds ratio = 2.95; 95% confidence interval (CI) = 1.48-5.87).

Interpretation: Rituximab treatment, known to increase the risk of severe infections in general, also confers such a risk for MS patients with COVID-19, in comparison with other MS DMTs.

General Disclaimer: Please note that the opinions expressed here are those of the author and do not necessarily reflect the positions of the Barts and The London School of Medicine and Dentistry nor Barts Health NHS Trust and are not meant to be interpreted as personal clinical advice. 

About the author



  • If I understand your contention correctly, you’re suggesting that the increased risk of severe outcomes from covid in patients taking taking Rituximab is confounded with the fact that many of those patients would be suffering from progressive MS, have accumulated more disability and be older. You also suggest that the differential in increased risk of severe outcomes between Ocrevus and Rituximab supports this view, given that Rituximab is more widely prescribed for progressive patients.

    But the paper from Sormani et. al show the same pattern of increased risk RRMS patients (i.e., the risk doesn’t seem to be attenuated in RRMS patients.)

    If everything I have mentioned is accurate, how do you explain the absence of attenuation in in RRMS patients?

    • The paper by Sormani is based on about 60-80 people on ocrelizumab and about 5 on rituximab so the rituximab influence in that paper is going to be diluted to essentially nothing. You need to read the paper by Simpson-Yap et al. 2021 on MedRxov based on about 2,500 people from a global collaborative, the worst risk of rituximab was adjusted for progression, age etc, etc. but is more significant than ocrelizumab in every quarter. In many places if you are eligible for ocrelizumab you get it not off label rituximab…however if you are in rituxiland you get rituximab. How do you explain the apparent difference given that ocrelizumab is a more potent CD20 depleter?. It is not risk attenuation is is worst risk none of the other DMT show this risk. P.S, Similar dat for rituximab is found in arthritis etc.

By MouseDoctor



Recent Posts

Recent Comments