CD20 antibodies deplete EBV


If you haven’t been living under a rock for the past few months :-),
you will have noticed that we have been saying that MS drugs that work the best, seem to deplete the memory B cells the most.

What! You are a ground hog and have just come out of hibernation.

Ok… go on have a read. Its free (Click Here). Let’s get that altmetric up:-0

Centre stage is the memory B cells. According to the Immunologist they can turn into antibody making machines to help create MS, they can make goodies (cytokines) or should I say baddies that can help other cells to create MS and for the die-hards stuck in the mud, they can stimulate T cells to do the business. 

However, we can see ProfG’s Black swan perched on a Roche-inspired B cell. Don’t understand have a read of past posts.

Search on “Black Swan”

But ProfG’s swan may be laying a big egg. The dots may not be CD20 the target for Ocrelizumab as Roche intended, but seeds of discontent for alternative thought and the blobs of budding virus.

my favorite

Epstein Bar Virus infects via a B cell marker and hides in memory B cells. This shows itself every so often, when the T cells can kick it’s butt.
So we suggested rather than depleting B cells to treat MS, ocrelizumab may be destroying a virus factory. Indeed we said that all active MS drugs may be doing this.
Is there evidence for this? 

Do we do some work or just a bit of reading?

The following was not done in MS (I have removed the disease)

Magnusson M et al. Epstein–Barr virus in bone marrow of patients predicts response to rituximab treatment.

Objectives. Viruses may contribute to disease. This prompted us to monitor viral load and response to anti-CD20 therapy in patients.

Methods. Blood and bone marrow from 35 patients were analysed for CMV, EBV, HSV-1, HSV-2,parvovirus B19 and polyomavirus using real-time PCR before and 3 months after rituximab (RTX) treatment and related to the levels of autoantibodies and B-cell depletion. Clinical response to RTX was defined as decrease in the disease activity score (DAS) >1.3 at 6 months.

Results. Before RTX treatment, EBV was identified in 15 out of 35 patients (EBV-positive group), of which 4 expressed parvovirus. Parvovirus was further detected in eight patients (parvo-positive group). Twelve patients were negative for the analysed viruses. Following RTX, EBV was cleared, whereas parvovirus was unaffected. Eighteen patients were responders, of which 12 were EBV positive. The decrease in the DAS was significantly higher in EBV-positive group compared with parvo-positive group (P = 0.002) and virus-negative patients (P = 0.04). Most of EBV-negative patients that responded to RTX (75%) required retreatment within the following 11 months compared with only 8% of responding EBV-positive patients. A decrease of  Ig-producing cells and CD19+ B cells was observed following RTX but did not distinguish between viral infections. However, EBV-infected patients had significantly higher levels of Fas (Suicide molecule) -expressing B cells at baseline as compared with EBV-negative groups.

Conclusions. EBV and parvovirus genomes are frequently found in bone marrow of patients. The presence of EBV genome was associated with a better clinical response to RTX. Thus, presence of EBV genome may predict clinical response to RTX.

So what does this study show. You can have a look at a few graphs from the paper.

EBV was eliminated from by the treatment of rituximab. This would be consistent with the B cells being infected by EBV being depleted. 

Those with the highest level of EBV showed the best response to therapy, compared to people who were so-called EBV negative (This proportion is too low compared to what we know of the normal population). 

Anyway, I have suggested that looking at memory B cells may be able to predict response to therapy, but this shows a hole in the argument because you can see that rituximab emptied the blood of B cells and as memory B cells make up about 30% of the CD19 population they are emptied from the blood too, and in this case there was not a 100% response to therapy.

However, if we drive from London (Bone Marrow/Lymph gland) to Leeds (Brain) along the M1 motorway (blood) and we can see a car (B cell) if we have a look for a minute from a bridge over the motorway (blood test). If I drive my car (Pathogenic B cell) to York (Centre of Gods own Country about 200miles north) to cause the problem (MS), if you look at 3 in the afternoon you might see loads of cars but do it at 3 in the morning and you might not see any although the problem is indeed a car driving to York (a few miles from Leeds) along the M1 motorway. 

So we may have to look outside of the blood to get the answer

The other problem is the treatment failure may be because the cells causing the failure are already in the brain and it is too late. In the ocrelizumab trials they re-baselined their results to allow 3 months
for the drug to work. A sensible thing to do if you are looking for efficacy.

However, you can also see that the bone marrow was not emptied by rituximab. If we look at other posts of rituximab, we can see that the lymph glands are not effectively depleted.

Subcutaneous versus Intravenous Administration of Rituximab: Pharmacokinetics, CD20 Target Coverage and B-Cell Depletion in Cynomolgus Monkeys Cheng-Ping Mao, Martin R. Brovarney, Karim Dabbagh, Herbert F. Birnböck, Wolfgang F. Richter, Christopher J. Del Nagro PLoS One. 2013; 8(11): e80533

Similarly, levels of peripheral blood B cells were depleted by >94% for both subcutaneous and intravenous dosing.  B-cell levels were decreased by 57% (subcutaneous) and 42% (intravenous…so not that great) respectively. Yes this is in monkies but I could be bothered to go through 4000 references to pick a relevant human one.

Looks like ocrelizumab does something similar 

where it was suggested that alemtuzumab may fail in many people after switching from fingolimod 

This is because it traps white blood cells in the lymph glands and bone marrow and because alemtuzumab does not clear out the bone marrow (at least in animals expresing human CD52). Is the level of purging of the bone marrow and lymph glands by ocrelizumab going to be enough so it doesn’t to suffer the same fate as alemtuzumab after fingolimod treatment?

About the author



  • Mouse, in a nutshell (one sentence) what are you saying? Drugs that target memory B cells will stop MS disease activity. Your posts are somewhat rambling – combo of Real Ale, age and that silly ponytail. But you're a good man at heart and I'm grateful for your hard work.

    • If you think this is a ramble…hold onto your socks. I have a couple of great ones to walk you through.

  • ProfG surely cladribine will get into these places unlike antibodies. It seems another good reason for the EMA to approve movectro?

  • A hell of a post. Thank you for synthesizing all the info!

    The yellow circle shows many CDs + MHC – what are those?

    Does attacking only one of them (52 for alem, 20 for ocrel) do the job?
    Are there any difference in which one of the CDs a drug attack or do they all present open gates to send trojan horses and destroy them?

    I understand that B cells go through a life cycle from infancy to plasma state. Are some agent better than others in destroying B cells earlier on in their life cycle?

    Do anti-CD20 therapies clear B cells in the rest of the body as well? CNS, bone marrow, etc…

    Do you expect the same results for patients transition from agents like tysabri or fingolimod, where B cells are stored out of the BBB? Will an anti-CD20 agents just deplete those stocks?

    Last, if B cells are depleted (as anti-CD52 has been used for a number of years now) and MS is one disease (Gavin's post just before yours) – why do good responders still progress from RRMS to SPMS even if they have not developed anti=bodies?

    Tony Fonda

    • The yellow circle (cyctoplasm = inside of a cell) and the red circle (nucleus) is a memory B cell.

      So attacking one with a killing antibody does the same thing, so alemtuzumab and oreclizumab are doing the same thing.. The stick holes in a cell and it dies.

      If you click on the link to the B cell paper there are a list of some CDs expressed by different B cell subtypes.

      Rituximab is an anti-CD20, so in this example you can see they do not deplete completely.

      The transitioning of fingolimod may be diffeent from natalizumab.

      Why do good responders still progress…I think the answer is good responders treated early enough dont progress, but it cant stop progression in people who are progressing

    • Many thanks for those replies.

      Is a multi CD coordinated attack possible to attack different subtypes at once?

      "I think the answer is good responders treated early enough dont progress"
      Gavin and my radiographer will diagree. Tysabri respondents are still progressing to SPMS.
      I, have been NEDA-3 for 6 years (Tysabri started 1 year only after CIS) and am still loosing a good chuck of my brain every year (subjective measuring I know, but am not alone).
      So if B Cells are the cause of atrophy, are we hypothesing that a small amount is leaking through (or remaining alive), small enough to cause atrophy but not a relapse of MRI visible lesions?

    • I was taking about alemtuzumab. I don't know the long term data for natalizumab it has bee around long enough for you to have an answer to this.

      Where is the paper.

      Iam not necessary saying B cells are the cause of atrophy. It is one idea.

      If you have OCB you have B cells already there.

    • Is multi CD possible….yes. we could make one in a few months our tech transfer didn't want to support it.

    • "Where is the paper": on this blog

      "e could make one in a few months our tech transfer didn't want to support it."

      Why not?

    • Sorry to come back on this, but just to make sure I understand well since B cells seems to the flavour of the day for a few months now:

      Do we think that the mechanism of action or cause of a relapse (T1 & T2 neuro inflamation) in MS is different than brain atrophy?

      I have been avoiding a spinal tap (OCB) for a while since Tysabri has been working for me (excl atrophy). Why the hurt if no alternative availables? This will be changing of course when Ocrel will be licensed in the UK.

      But then, the decision making will surely not be just down to OCB titres.

    • Why not….didn't get the invention and the technology. They have so little cash also. So one for the back burner.We have been using student projects to get pilot information.Sad buy true. This year all our phD projects have been given to medical projects and not science projects. Maybe we are chasing impacts as it is years before a student project has impact.

    • Where on blog?
      I have always thought the driver of relapse and atrophy are different although relapses can cause atrophy.There treatments need to be different I suspect.

      Ocrelizumab does not get rid of plasma cells and so suspect no major effect on OCB.

    • If you had high neurofilaments you could have gone on PROXIMUS to test additional neurofilaments.

    • Blog: beleive me – it is there. Gavin will tell you about (not lazy, just too many keywords to choose from in the absence of a site map)

      PROXIMUS: considered it at the time but again, why the pain if no possibility of using the information for treatment strategy? At the time, Tysabri was the Best Bet for JCV-. Ocrelizumab is changing things slightly…

      "Ocrelizumab does not get rid of plasma cells " so the idea is to hunt B cells with Ocrel in the first 3 maturity stages (infnacy, memory, etc…) before they reach the plasma stage?
      If so, how will we deal with all the plasma cells kept at bay by Tysabri, if Ocrel let's them back in?

      I know I am a pain, but feel for once that I am starting to get this B cell business. Hopefully my persistence is equally helping others understand.


    • Plasma cells don't really circulate they make a very small fraction in blood they live in follicles in brain in ms and in lymphoid.

      tissues. So natalizab is probably not affecting them. Post natalizumab I don't wlthink we see surges of antibodies but I could be wrong. ProfGwiki is this correct.

      To put spanner in works. Anti BAFF hits the mature-naive cells and depletes them and hits the plasma cell and depletes them and antibody levels drop in the blood. but increases memory B and this makes MS worse. I can't interrogate the data as they say they threw the cells they collected away (could only keep for so long because of ethics).

  • Would there be any information gleaned from testing new patients for EBV? As I consider transitioning from Tysabri, it always helps to have all information to make a wise decision.

    • I was under the feeling that 100% of MSers are EBV positive.

      Why testing? Save your cash and trouble.


    • Exceptions? 1 would surely be enough to throw the EBV hypothesis down the drain no?

      Can you bring my questions above to MD's attention if he had not seen them? Keen to read his views on those particular points. I thought those B cells are up to something….

    • BTW, no certainties except for death and taxes and the latter seems to be increasingly optional these days 😉

    • The presence of EBV is only as hard as you look. In the study above about 50 percent were EBV this is way lower than published data

  • I am a big advocate for the EBV theory but I have one question. How does MS being EBV-driven square with the very low incidence in countries like Japan (northerly latitude, industrialized, "normal" EBV infection rate)?

    In my personal studies I've long held that any theory needs to not only explain the action but also square with what we see in reality — higher than "normal" rates in Canada & Scandinavia as well as lower than "normal" rates in Japan/Korea.

    How does this EBV-based theory connect those dots??

    • MS is not common in Japan but it may be the age of encounter thats important but EBV is not my thang

    • Actually MS is not very common in Asia in general and in Africa.
      But as far as I can find EBV seems to be an "opportunistic" virus, and depending on the "scenario" it encounters (and this includes the genetic factors and low circulating vitamin D in this scenario) it may "manifest" by triggering different pathologies.
      For example, Burkitt's lymphoma as well as nasopharyngeal cancer are caused in part by EBV and although rare, it has a high incidence in Africa and parts of Asia, which is not so common among caucasians.
      So it's quite clear that there is a relationship of viral manifestation with other environmental factors and genetic factors.
      At least for me this is very clear.

  • No, I haven't been living under a rock for the past few months. But will confess to doing so prior to about 2015 (I think). Had wondered what the Black Swan was! Thanks for all the links, am now caught up 😉 that's a lot of history and politics 🙁
    Then Googled Michael Pender and up popped a guest blog post from way back:

    All these years, all this knowing about ebv infected memory B cells and how far have we come? So, ok we know which cells to target 🙂 But what about all the CD8 that get taken out as collateral damage? Shouldn't that make MS worse??
    Off to crawl back under my rock…

    • Alemtuzumab is a big CD8 depleter but if there are no B cells there is nowhere for the EBV to hide.

  • In this paper they say that not all Bcells are equal and some have an anti inflammatory action
    So killing all the Bcells could be detrimental .They said you need to phenotyping each patient to select target b cell depletion in each one
    experimental but also clinical findings indicate that not all B cells contribute in an equally
    pathogenic manner and that certain subsets may in contrast mediate anti-inflammatory

    • You are absolutely right there are B regulatory cells and when I first started science my boss was convinced of that. These ideas got lost with the T cell brigade. In many studies as the B memory cells go up the regulatory cells go down because they are looked at as a percentage. I stuck my neck out on B memory cells as their absolute number was changing and so making the percentages change.

      For many years To cells have been seen as subsets, B cells are not hesitate at one population

    • Thanks for replying
      They talk about some B cells expressing, il 10 ,which is inlamatory mediator
      Also plasma cell dont express cd 20 ,So Rituxan can´t touch them
      Could this be also be a reason for people not responding to treament?
      Could Ocrelizumab have a different action agnaist plasma cells?

    • The regulatory B cells make IL10 this is anti inflammatory if you are into T cells however it's a Bit cell growth and differentiation factor so could be bad news. That's why messing around with cytokines can be bad news.

      Rituximab can't touch plasma cells and neither can ocrelizumab or ofatumumab.

  • Is production of neutralising antibodies against Rituximab likely to affect response to Ocrelizumab treatment, or are they different enough? Therefore will directly switching from rituximab to ocrelizumab be a viable option?

  • Rituximab causes binding antibodies in about 15-25percent and nabs in about 5-15percent. They may stop drug working.
    Ocrelizumab induces 0.5 nabs so this may be a marketing battle ground.

    The nabs or one probably don't effect the other as I think they target different parts of the CD20 but can antibodies to one affect response to another. It shouldn't but it is possible. Alemtuzumab to rituximab ocrelizumab switchers may be the most informative.



Recent Posts

Recent Comments