Die Wende: Extending the anti-CD20 overdose


It’s a bittersweet fact that wars spark big leaps in science, and this has unfortunately also been true for our fight against the COVID-19 pandemic. When it comes to treating MS, susceptibility to common infections and ability to mount vaccine responses have become important themes among neurologists. Furthermore, pwMS prefer drugs that do not require regular hospital or phlebotomist visits. One of the DMT’s at the frontline during this pandemic has definitely been anti-CD20 treatment. For decades, the standard interval between anti-CD20 infusions (I.e. ocrelizumab, rituximab) has been 24 weeks. This was based on the practice in pivotal clinical trials that tested their efficacy against placebo, but was swiftly overruled by the viral enemy troops.   

Several journals recently published articles in which the standard interval between anti-CD20 infusions has been delayed because of COVID-19. In Germany, the delay between infusions was arbitrary and dichotomised in more or less than 4 weeks. This lead to a median delay in infusions of 8 weeks. In The Netherlands, reinfusion of ocrelizumab was done when B cells showed minimal signs of repopulation (≥ 10 cells/microliter; lower limit of normal in healthy controls 100 cells/microliter). This approach resulted in a median redosing interval of 34 weeks which is a 2.5 month extension compared to the standard interval. Follow-up periods in both studies were relatively short: 3 months in Germany, and 11 months in The Netherlands. During this period, no meaningful differences in frequency of relapses, disability progression and MRI markers were seen. 

Implementing B-cell guided reinfusion schemes in clinical practice essentially means that you buy a new carton of milk when the previous one is empty. Your first infusion depletes your B cells and you are reinfused when B cells go up again. An obvious example of good housekeeping! Adopting this strategy means that 8 anti-CD20 infusions in 5 years will suffice compared to our current practice of 12 infusions. The price of a rituximab vial is around 1500 euro whereas ocrelizumab is priced about 6000 euro (depending on national Pharma contracts). This saves health insurance companies a lot of money, and means that funds can be invested elsewhere. 

However, it’s important to realise this is only the first step in rationalising MS treatment with anti-CD20. There is a slight but consistent increased infection risk with chronic anti-CD20, and vaccination responses are at least partially blunted. It is unclear whether extended-interval dosing will solve this problem. The cut-off of 10 B cells/microliter for reinfusion is still very stringent, implying it is unlikely going to impact on the hypogammaglobulinemia (I.e. low IgM and IgG antibody levels) that occurs with longstanding anti-CD20 use. Similarly, the positive impact of this redosing strategy on vaccination responses need to be seen. Anyway, timing of vaccination under anti-CD20 will remain a hassle in clinical practice. Overall, also when we use B-cell guided reinfusion protocols, pwMS will be chronically immune suppressed.  

Therefore, we need to ask ourselves whether chronic immune suppression is necessary, and whether aggressive induction therapy with for example four B-cell guided anti-CD20 infusions followed by watchful waiting (similar to cladribine/alemtuzumab/aHSCT) or a mild first-line immune suppressant (such as Tecfidera) would be a valid alternative strategy. We know that absolute B cell counts are not a good surrogate of re-emergent disease activity in MS and there is evidence that pwMS on anti-CD20 remain stable long beyond B cell repopulation rates. Let’s hope future research brings us a peaceful revolution! 


Disclaimer: Please note that the opinions expressed here are those of dr. Ide Smets and do not necessarily reflect the position of the Barts and The London School of Medicine and Dentistry nor Barts Health NHS Trust.

Mult Scler. 2021 Jul 9;13524585211028833. 

Personalized B-cell tailored dosing of ocrelizumab in patients with multiple sclerosis during the COVID-19 pandemic

Zoë Ygj van Lierop 1, Alyssa A Toorop 1, Wouter Jc van Ballegoij 2, Tom Bg Olde Dubbelink 3, Eva Mm Strijbis 1, Brigit A de Jong 1, Bob W van Oosten 1, Bastiaan Moraal 4, Charlotte E Teunissen 5, Bernard Mj Uitdehaag 1, Joep Killestein 1, Zoé LE van Kempen 1Affiliations expand

PMID: 34240631, DOI: 10.1177/13524585211028833


In this observational study, 159 patients with multiple sclerosis received personalized dosing of ocrelizumab incentivized by the COVID-19 pandemic. Re-dosing was scheduled when CD19 B-cell count was ⩾10 cells/µL (starting 24 weeks after the previous dose, repeated 4-weekly). Median interval until re-dosing or last B-cell count was 34 [30-38] weeks. No clinical relapses were reported and a minority of patients showed Expanded Disability Status Scale (EDSS) progression. Monthly serum neurofilament light levels remained stable during extended intervals. Two (1.9%) of 107 patients with a follow-up magnetic resonance imaging (MRI) scan showed radiological disease activity. Personalized dosing of ocrelizumab could significantly extend intervals with low short-term disease activity incidence, encouraging future research on long-term safety and efficacy.

Neurol Neuroimmunol Neuroinflamm. 2021 Jul 14;8(5):e1035. 

Ocrelizumab Extended Interval Dosing in Multiple Sclerosis in Times of COVID-19

Leoni Rolfes  1 , Marc Pawlitzki  1 , Steffen Pfeuffer  1 , Christopher Nelke  1 , Anke Lux  1 , Refik Pul  1 , Christoph Kleinschnitz  1 , Konstanze Kleinschnitz  1 , Rebeca Rogall  1 , Katrin Pape  1 , Stefan Bittner  1 , Frauke Zipp  1 , Clemens Warnke  1 , Yasemin Goereci  1 , Michael Schroeter  1 , Jens Ingwersen  1 , Orhan Aktas  1 , Luisa Klotz  1 , Tobias Ruck  1 , Heinz Wiendl  1 , Sven G Meuth  2

PMID: 34261812 DOI: 10.1212/NXI.0000000000001035


Objective: To evaluate the clinical consequences of extended interval dosing (EID) of ocrelizumab in relapsing-remitting multiple sclerosis (RRMS) during the coronavirus disease 2019 (COVID-19) pandemic.

Methods: In our retrospective, multicenter cohort study, we compared patients with RRMS on EID (defined as ≥4-week delay of dose interval) with a control group on standard interval dosing (SID) at the same period (January to December 2020).

Results: Three hundred eighteen patients with RRMS were longitudinally evaluated in 5 German centers. One hundred sixteen patients received ocrelizumab on EID (median delay [interquartile range 8.68 [5.09-13.07] weeks). Three months after the last ocrelizumab infusion, 182 (90.1%) patients following SID and 105 (90.5%) EID patients remained relapse free (p = 0.903). Three-month confirmed progression of disability was observed in 18 SID patients (8.9%) and 11 EID patients (9.5%, p = 0.433). MRI progression was documented in 9 SID patients (4.5%) and 8 EID patients (6.9%) at 3-month follow-up (p = 0.232). Multivariate logistic regression showed no association between treatment regimen and no evidence of disease activity status at follow-up (OR: 1.266 [95% CI: 0.695-2.305]; p = 0.441). Clinical stability was accompanied by persistent peripheral CD19+ B-cell depletion in both groups (SID vs EID: 82.6% vs 83.3%, p = 0.463). Disease activity in our cohort was not associated with CD19+ B-cell repopulation.

Conclusion: Our data support EID of ocrelizumab as potential risk mitigation strategy in times of the COVID-19 pandemic.

About the author

Ide Smets


  • This is not shocking or earth shattering.

    Years ago I wrote a viewpoint and it was roundly rejected of course because I was going against the ‘industry standard’. Many reports have absolutely stated this without any semblance of a doubt, that dosing intervals fluctuate with it or are dependent on BMI for example among other factors not to even talk about ethnic/racial differences.

    Biased opinions kill science !!

  • Any neurologist who thinks mitigating Covid risk is worth increasing MS risk seriously needs to get a clue. Thinking that catching what is frankly a relatively mild respiratory illness is worse than brain damage from an attack that could have been prevented is insane. Screw extended interval dosing.

    • I don’t think there is much risk involved in ‘extended interval dosing’. This is a more evidence-based way of dosing the anti-cd20, mainly avoids over- rather than undertreatment. We strangely needed covid-19 to start thinking about rational drug dosing.

      • Even if so (and it seems unlikely that there won’t be at least some people who get avoidable new lesions from decreasing the dose), the equivalence between the two regimens only holds in a view of the world where MS is a disease of lesions and relapses. But if we’re thinking about smoldering MS, then the lower the dose of anti-CD20, the lower the effect on brain atrophy, thus the worse the treatment effect. Surely patients should at least be given the choice of whether or not to reduce their dose instead of having a reduced dose imposed upon them unilaterally.

        • I agree that it’s important to have shared decision making about this, especially as there is not yet a non-inferiority trial available of extended vs standard dosing. However extending the interval does not necessarily mean pwMS get a lower dose. It means that you redose when the drug effect wears off based on a biomarker rather than arbitrary interval. For some people it might mean that intervals are shortened.

          • No..you’re not getting what he is referring to..
            Anti-cd 20 is now being tested at double and possibly even triple doses to see if that improves the rate of brain atrophy that occurs on this therapy. As of now rituzimab/ocrevus is one-way train to spms.

            “PROF G
            July 4, 2021 at 3:13 pmny
            We just heard from Frederik Piehl at our MS meeting that in their Swedish cohort the brain volume loss on rituximab was greater than the interferon-beta treated comparator group. The BVL data in general on anti-CD20 therapies is very disappointing. Anti-CD20 may be good at stopping relapses and MRI activity but is not that effective at stopping smouldering pathology or the real MS.”


          • Yes, I know, but the redosing interval is different from the actual dose administered. These are two different aspects of anti-CD20 treatment. You do not lower the dose by extending the treatment interval; you just optimise the timing of retreatment. Also in the dodo study, the investigators could choose to reinfuse whrn b cells go up after depletion rather than arbitrary at 6 months.

          • I would counter by saying the biomarker they are using is not a good one and only has relevance to whether or not you have shut off lesions and relapses. Otherwise there would be no point in doubling and tripling the Ocrevus dose, everyone’s count already goes to zero with 600 mg and even the fastest-repleting patients would only start repleting long after 1200 or 1800 mg had left their systems. My understanding is that the point of the higher doses is to get more past the BBB and scrub the CNS clean instead of just nuking the periphery. Lower dose = more bad B cells left to cause damage, which is clearly not measured by CD19 blood counts.

          • There are two things in your answer:
            – The dose: We now use 600 mg, and I agree higher doses need to be tested to see if they have better CNS penetrance/efficacy
            – The dosing interval: This has now arbitrary been set at 6 mo while the average repopulation rate for B cells is 72 weeks, implying we are overtreating and can safely postpone infusions. Extending the interval between infusions does not equal lowering the dose. It means rational reinfusion.
            – The biomarker: anti-CD20 mAb depletes most CD19+ B cells, so from a pathophysiological point of view it would be logic to measure B cells to decide on reinfusion. There is no obvious reason to believe that reinfusion before this timepoint would be useful. CD19 counts are thus a mechanistic biomarker related to the mode of action of anti-CD20 and should not be confused with an MS treatment efficacy biomarker. That would indeed require different markers. B cells are indeed not a good surrogate for MS disease activity, as most pwMS remain disease activity free long after B cells have repopulated. In addition, most likely memory B cells are a better indicator than total B cell counts.

    • Getting away a bit from the purpose of the headline and Dr. Smets essay, I’d like to point out here that new MS “Brain Damage” versus no anti-CD20 therapy, presents relative risks. The older you get, the more risk you have of infection and cancer, and who knows what else (as a function of an immune system not being “brand new”). Add to that risk, anti-CD20 therapy. So if, relatively speaking, you’ve made it to 60 in relatively one piece, you may not want to invite these extra variables which could end your life, when you have other plans for the next 20 years. Covid can also create long term inconveniences, so it’s pretty difficult to go out “smiley faced” among the unvaxed idiots (I mean face to face, shoulder to shoulder) and feel authentic about your smile. If one is 35, reality and risk for them is quite different.

  • I disagree with anonymous and see that it is not about covid but about the safety profile of anti CD20 and recognising that most are over-dosed at the current frequency. I for sure know that I would not want more of the stuff than I need! I’m all for extended interval dosing

  • What is this biomarker talked of?

    If someone is given their next 6 month appointment at the time of infusion and at no point is blood taken, how are we (never mind the neuros) supposed to know what this biomarker is?

    We have no idea if the 2 covid vaccines have been in anyway effective because of Ocrevus, so if they ain’t even looking at that, what chance is there they will be looking at biomarkers then?

    • The biomarker for reinfusion of anti-CD20 is tightly related to the mode of action of the drug depleting B cells. But it is indeed no efficacy biomarker which will need much more research.

  • “…a relatively mild respiratory disease…” You’re generalising Anonymous. Sometimes it’s true; frequently it’s fatal. That means that wMS are as likely to die from Covid as the gen.pop. and as likely to land up with Long Covid.

    • Good point and there is evidence that people on anti-cd 20 have double the risk for hospitilazation. Moreover, there is data that some people that contract covid sustain loss of brain tissue!

    • ““…a relatively mild respiratory disease…”

      Huh…the unvaccinated are causing mutations that change the virus
      and it’s now more deadly….and affecting younger populations w/hospitalizations
      and morbidity.

  • How are cell repopulation measured? Based on my personal experience I don’t beleive this answer is straight forward and I’m not convinced we know exactly what needs to be measured. I say this because I was in the Ofatumumab trial and in the active arm based on cd19/20 counts when my participation ended. My start on Ocrelizumab was supposed to be delayed until they rebounded but I had 2 bad relapses, the first of which really started around the time of my last ofatumumab dose. In the end I started Ocrelizumab 3 months after my last dose even though my cd19/20 numbers were still unreadably low. It was not until after my third round of Ocrelizumab that my relapses finally stopped and I’ve since been stable.

    FYI, I’m on the small to average end of the weight spectrum so that does not account for my lack of response though it does make sense to me that it should be factored into dose.

    Given the opportunity I would consider participating in a delayed dosing trial but only if part of it were to figure out what markers should really be used to determine when redosing were appropriate. That said maybe it is more like a reconsitution therapy so perhaps the marker to look at really needs to be on the intitial treatments. Was the dosing strong enough to ‘kil off’ the required memory cells. Regardless if this strategy is to be used I think much more than CD19/20 measurements need to be considered. At least that is the only way I as a patient would want to consider it.

  • I think that is a point well made. Moreover, if new and reproducable data confirms a prior study that found that covid infection reduces brain tissue (amongst other not insignificant maladies) anti-cd20 treatment may not be the best option for certain folks. Unless, of course, you very brilliant researchers figure out a way for anti-cd20 people to mount an immune response to vaccination. It is also becoming quite clear that long covid is something to be avoided, so the proposition made by some that the relative risk for MS folks to getting covid is small may indeed be premature.

By Ide Smets



Recent Posts

Recent Comments