Anti-CD20 Kool-Aid and COVID-19 vaccines

A colleague suggested to me that the whole academic neurology community has imbibed too much anti-CD20 Kool-Aid and it is time for us to sober up. Is she is right? 

Living with and being treated for MS is a fine balancing act and COVID-19 has shown how narrow the balance beam actually is, particularly as we start to enter the period that in some parts of the world is being referred to as the end-game of the pandemic.

In relation to COVID-19 vaccines I have been saying to my patients #GetVaccinatedASAP; some immunity is better than no immunity. However, we now know that people who are profoundly B-cell depleted due to being on an anti-CD20 therapy or who are on fingolimod, an S1P-modulator, in general, don’t make antibody responses to the COVID-19 vaccines. 

We have known for some time now that pwMS on DMTs who get COVID-19 seem to be at no greater risk of severe COVID-19 and death compared to the general population with the exception of pwMS on anti-CD20 therapies. This implies that anti-CD20 therapies either impact preexisting cross-reactive protective immunity from other community-acquired coronavirus infections and/or antibody responses during COVID-19. The latter have been now shown to prevent severe infection and death.

I think there is ample evidence to support both of these explanations.

People who had the original SARS in 2003 have been shown to have broad anti-coronavirus immunity and make very good neutralising antibodies to SARS-CoV-2, which is boosted to very high levels in response to COVID-19 vaccination. Similarly, immunoglobulin formulations made from the plasma of blood donors prior to COVID-19 is able to neutralise SARS-CoV-2. Therefore preexisting immunity to other coronaviruses is helpful and this will be blunted by being on an anti-CD20 therapy, particularly if you have been on the anti-CD20 therapy for a prolonged period of time. This observation goes beyond COVID-19 and explains why the risk of infections in people on long-term immunosuppression increases over time. The waning of preexisting immunity and the blunting of booster and new immune responses comes with a price. People with MS don’t have to accept the inevitability of this as there are strategies to derisk long-term immunosuppression and still keep your MS disease activity under control.   

It has also recently been shown that the delayed development of neutralising anti-SARS-CoV-2 antibodies is associated with death in patients with COVID-19 in intensive care  (Lucas et al. Nat Med. 2021 Jul;27(7):1178-1186). Therefore contrary to original hypotheses that B-cell responses were not necessary for recovery from COVID-19, it is now clear that antibody responses to SARS-CoV-2 are important in protecting you against a poor COVID-19 outcome. This now explains why people on anti-CD20 therapies are more likely to have severe COVID-19 and I suspect more likely to succumb to the infection.

It is also clear that immunity against SARS-CoV-2 acquired by either natural infection or vaccination wanes with time. This waning is greater the older you are because of immunosenescence and will also be greater in people on immunosuppressive therapies. This is why the UK Government has announced that a booster dose of one of the COVID-19 vaccines will be offered to people over 50 and vulnerable people with diseases and treatments that suppress immune function. The latter will include pwMS on immunosuppressive therapies. 

We are already seeing the consequences of this waning immunity. About half of COVID-19 deaths that are occurring in the UK at present are in people who are over the age of 50 who have been double-vaccinated. The other half are people who have not been vaccinated and this population tends to be younger (Public Health England. Investigation of SARS-CoV-2 variants of concern: technical briefings. 23 July).  

Another worrying trend is the gradual drift of the SARS-CoV-2 variants towards the selection of an immune escape variant that is not neutralised by pre-existing anti-coronavirus antibodies. This appears to be happening. The new mu or Colombian variant is not neutralised by convalescent antibodies from people who have had COVID-19 or the Pfizer-BionTech mRNA vaccine (see Uriu et al. BioRxiv). I predict this mu variant will soon become the dominant circulating virus and lead to a new wave of infections; this pandemic is going to have a very long tail. 

So in an ideal world, as the pandemic evolves, it now makes sense to try to optimise your anti-viral antibody response to SARS-CoV-2 to protect you from being infected with SARS-CoV-2 and the new and continually evolving variants and from developing severe COVID-19 and dying from it. 


So what do I tell my patients? 

I am a little sensitive to this issue as one of my patients who was on an anti-CD20 therapy tragically passed away last week from COVID-19 pneumonia having been double vaccinated with the AstraZenca COVID-19 vaccine. After this death, we have had internal discussions and I have spoken and exchanged emails with many colleagues around the world to try and get a sense of how to advise my patients about vaccines and the booster. It is clear there is no consensus on how to deal with the issue of blunted vaccine responses in patients on anti-CD20 and other therapies. 

At present most MSologists are not interrupting ocrelizumab dosing and simply recommending the vaccine and/or booster dose to be given at least three months since the last infusion and at least 4 weeks before the next infusion. This advice is based on the Roche-Genentech VELOCE vaccine study. This strategy, however, is being followed despite it having clearly been shown not to work, i.e. these patients will not be making antibody responses. My patient who died of COVID-19 did not seroconvert despite having had the vaccinations in this window. This is the same with our other patients who have been double-vaccinated who have developed severe COVID-19.

On the other extreme, a handful of MSologsist are missing or delaying the next dose of ocrelizumab and waiting for patients to reconstitute their B-cells before vaccination. This strategy will work if there is sufficient B-cell reconstitution. Based on my interpretation of the literature you need to have at least 10 B-cells per millilitre of blood to mount an antibody response to the COVID-19 vaccines. 

Interestingly, it seems to be the pragmatic neurologists in the US with smaller MS caseloads that have adopted this strategy and the bigger more academic units the former strategy. It would be interesting to speculate why this is so. One colleague suggested to me that the whole academic neurology community has imbibed too much anti-CD20 Kool-Aid and it is time for us to sober up. Maybe she is right.  

It is clear that this adaptive vaccination/booster strategy is logistically challenging for the simple reason that B-cell reconstitution post anti-CD20 therapy is quite variable. This means that after a certain period of time, say 9 months after your last dose of ocrelizumab, 6 months after your last dose of rituximab and 4 months after your last dose of ofatumumab you have then do monthly B-cell counts. The logistics of this when you have hundreds, and in some units, thousands, of patients on anti-CD20 therapy is not trivial. Put simply we don’t have enough resources within the NHS to take this on. Maybe pharma can help? 

Another factor is even if you wait for B-cell reconstitution and vaccinate these patients will we then have to check if they seroconvert and if not re-vaccinate them? Even if they get an antibody response will the antibody response be good enough to protect them from infection or reinfection with the emerging variants? There are so many unanswered questions. 

So until we have more evidence I think we should sit on the fence. However, it is only fair to tell pwMS about the problem and the uncertainty around this issue and give them the choice to delay or miss their next dose of anti-CD20 therapy. Do you agree? 

Now, what do we do with fingolimod and the other S1P modulators? 

There is clearly a disconnect between what we see clinically and what basic science is telling us with regard to fingolimod. Yes, fingolimod seems to blunt both antibody and T-cell responses to the COVID-19 vaccine. However, there is no obvious signal with fingolimod in relation to severe COVID-19 in pwMS on the drug. Why? I suspect the immunology studies are at fault. Fingolimod traps lymphocytes in particular naive lymphocytes in lymph nodes and other secondary lymphoid organs. So when you sample the peripheral blood you are getting firstly fewer and secondly a biased sample of lymphocytes, which may miss the newly created effector and memory anti-SARS-CoV-2 T-cells. In addition, fingolimod is a much dirtier S1P modulator working on most of the S1P receptors. In comparison, the newer S1P modulators are more selective and work on a much narrower range of receptors. So I don’t think we can extrapolate what is happening with fingolimod to siponimod, ozanimod or ponesimod. We need data on the newer agents before making any recommendations.

Advice around S1P modulators is complex because you can’t simply stop them to allow immune recovery, vaccinate someone and then restart them. This is one class of DMT that is associated with rebound MS disease activity. So at the moment, I am simply telling my patients that fingolimod blunts vaccine responses, but by how much we don’t know because the drug itself interferes with T-cell trafficking and hence affects the assays we use to assess immunity. I say they should have the vaccine ASAP and get the booster when it is offered. I also reassure them that pwMS on fingolimod who get COVID-19 seem to do okay with the exception of people with other risk factors such as older age, comorbidities (obesity, hypertension, diabetes) and severe disability. The latter seems to be prompting a small number of high-risk patients on fingolimod to come off the drug and consider switching to another less immunosuppressive therapy. 

The future

It is quite clear that COVID-19 and its vaccines are having a profound effect on the management of MS. For the first time, neurologists have to engage with hardcore immunology and develop a deeper understanding of immunology, infections and vaccines. They are also having to come to terms with uncertainty and communicating it to their peers and patients. This may be the biggest legacy of COVID-19, i.e. becoming more knowledgeable and resilient in the face of uncertainty. 

As a result of COVID-19, the MS community will have to reassess the impact DMTs have in the long term and we will come to realise that long-term immunosuppression is really not ideal. This will result in us developing induction-maintenance strategies to manage MS and we will likely see the wider adoption of immune reconstitution therapies. Who wants to be immunosuppressed when you don’t have to be?

Leave a comment

Anti-CD20 therapies = ocrelizumab, rituximab, ofatumumab or ublituximab

S1P-modulators = fingolimod, siponimod, ozanimod or ponesimod


I am raising funds from subscriptions to administer the MS-Selfie Newsletter and microsite. The subscriptions, for case studies only (all other newsletters are free), will be used to hire an administrator to proofread, curate and transfer the contents of the Newsletter onto the companion MS-Selfie microsite. If you find these Newsletters helpful and can afford to subscribe I would urge you to do so; it will help me and the MS community.

The case study from the 10th of September is about a patient on ocrelizumab who has been recommended to switch to cladribine. In this case study, I discuss the issues around this DMT switch and the issue of COVID-19 vaccine readiness.

Give a gift subscription


Conflicts of Interest




General Disclaimer: Please note that the opinions expressed here are those of Professor Giovannoni and do not necessarily reflect the positions of Barts and The London School of Medicine and Dentistry nor Barts Health NHS Trust. The advice is intended as general advice and should not be interpreted as being personal clinical advice. If you have problems please tell your own healthcare professional who will be able to help you.