It seems that our experiment is going to happen by default


Some time ago we reported this paper, we have deposited in MedRXiv. This is important to read.

The ocrelizumab phase II extension trial suggests the potential to improve the risk:benefit balance in multiple sclerosis David Baker, Gareth Pryce, Louisa K James, Monica Marta, Klaus Schmierer doi:

Given the turn of events with COVID-19, I now repost this and suggest that you have a read if you are taking ocrelizumab, rituximab and even ofatumumab and are in the worrying place of determining if you should have another course of treatment or not. Your neuro may not have found this paper, as it is not on some of the search engines. Make sure people in your ocrelizumab, rituximab goups have read this.

Some of you may decide not to carry on treatment, especially as medical advice may go that way as it has been suggested that you delay infusions, so you may be doing the study that we want to do. However the data may get lost, unless it gets collected.

If you are taking anti-CD20 treatment, in fact if you are taking any medication please, please, please make sure you sign up to the MS Register (Click here) if you live in the UK. This way, where-ever you are in the UK, your story can get captured and the results of these COVID-Virus induced trials will get captured

For example is there an effect of drug holiday on disease activity

If you live elsewhere MSBase (Australia) and NARCOMS (North America). Therefore, once this pandemic is over, we can apply to Rod and the other Giys and Gals at the MS Register to see if people stopped taking therapy for an extra few months to determine if people were still protected from MS. If so it says that you don’t need to dose as frequently.

Will stopping anti-CD20 treatment protect you from COVID-19? I guess it depends how quickly your B cells return. However, this does not occur quickly and takes months and the epidemic may have come and gone by then. Whilst there is antibody circulating it can deplete any new B cells that you make.

I suspect the makers of ofatumumab, which is injected under the skin every month, will be saying if people were using their drug, which is going through the regulatory approval process at the moment, that the B cells will return very quickly after you stop. So the effects may be reversible and with COVID-19 around. It becomes very attractive.

Ofatumumab was injected under the skin every 3 month Q12W or every month Q4W for 6 months. You can see that the immature B cells were back within 4 weeks. But the CD19 population wasn’t, saying that memory B cells were depleted for much longer (So do you need to inject monthly…I suspect not…but that’s another story. However I have yet to see the memory B cell data, Perhaps the manufacturers can send it to me it…..Please). Data from ECTRIMS poster

So why did we put the paper above online. We want this work (paper) in the citable public domain, as it helps to support the idea we may not need to have a dose of ocrelizumab every 6 months, forever. This is because we wonder if the risk of infection increases with time as your B cells are persistently depleted. Importantly, it provides the the background info for we or others to apply for grants to do the ADIOS study. This is about safety and the study will get done. It is been done as people run out of money and have to stop treatment in the US

I would think it is time for the manufacturer to do this, as they can shut me up and even get me on the advisory panel…., they can do it quicker and than any academic can do it and they can control how the information is present, just like they could have done so with the phase II study above.

Importatly we ask can we get away with fewer doses. The question is whether ocrelizumab is an immune-reconstitution therapy (IRT) based on the idea that there is a comonality of CD20 depletion with cladribine, which is an IRT and also to some extent with alemtuzumab, that also induces long-term depletion of memory B cells.

Even if it is not an IRT, do we have to deplete so often? As in comparison to the CD19 B cells, which are depleted for 6 months with rituximab (top graph), if we look at the bottom graph we can see that memory CD19 B cells are still depleted at 12 months. So maybe we can have immature and mature B cells return, which are the cells that would respond to a new viral infection and this still protects us from new infection. This is part of the logic of extended dosing interval treatment of natalizumab to control MS but protecting against PML. This seems to let in CD4 and macrophages and keep out CD8T and B cells (which contain the memory B cells)

Now if you read the ocrelizumab paper (above) you can see that ocrelizumab is much more potent than rituximab and depletes CD19 for some time. So it is takes on average 12-15 months for the CD19 cells to some people it takes up to 3 and a quarter years. Based on the rituximab data we would expect it takes longer for the memory B cells to recover. We don’t have the data on this…Why is it being hidden? If this population is important then you may get treatment benefit for longer than 6 months.

This concept is supported by rituximab in MS where efficacy seemed to hold MS at bay for a year based on work in the initial clinical trials reported in 2008.

Studies continue to support this concept…but it could be wrong.

I am not a neuro and am not recommending anything but this may be of relevance in this worrying time. As is speaks to the information about if you stop ocrelizumab for a while will your MS come back….Again based on rituximab stoppers there is no apparent rebound effect.

In the study presented in the paper, people got three or four courses of ocrelizumab and then got nothing for 18 months and the drug may have still been work, so if you have to postpone treatment it is possible that you are still protected from MS. Now I have to hold my hands up and say the 18 month data is weak because either people dropped out of the study or the data had not yet been collected, but at 12 months drug free the data may be more believable

Kappos et al. ECTRIMS 2012

Now the worry I have, and I am thinking out loud here, is if people stop taking drug, it is possible that with time as your immatutre B cells come back, then there may be antibody around and anti-drug antibodies could form.

The fact that antibody is humanised does not mean you can’t make an antibody response. Give people 600mg and your anti-drug response is 0.4% give people 20mg and your antidrug response is nearly 20%. If you stop taking drug at some point the levels equivalant to taking 20mg will occur…what happens then?

If anti-drug antibodies do occur, it could be possible that it stops the drug working in a few people if you start the same treatment again. So if you do stop maybe ask you neuro to save some blood serum if they can do it legally as some places have Biobanks. We have the capacity to measure anti-drug responses to any CD20 reagent and could probably check this before people start up again. Probably I am worrying for no reason and if it did occur no worries, as you could swap to a different variant…..maybe some company would sponsor us to look at this and help getting bandwidth for the paper work and hands to do the study. But you may be doing an important experiment. We missed an opportunity to find out what causes relapse/rebound when you stop fingolimod.

COI Multiple

About the author



  • not sure I am following this correctly, but are you saying that increasing the interval between anti-cd20 doses is potentially inviting neutralising antibodies? With something like rituximab surely we have data from other applications on this. Personally I’ve had a haematologist procure minimal doses of rituximab to treat secondary autoimmunities (just enough to zero my b-cell counts). I wonder if the thinking behind this was to minimise this happening. Regardless, they have been giving some patients rituximab for many years, so surely they must have some insight?

    • You are correct on both counts. I am assuming that you are taking about alemtuzumab as secondary autoimmunity is not a problem for ocrelizumab and rituximab. Maybe HSCT in Mexico it seems the standard after care is rituximab… Likewise I am guessing you are trying to prevent rather than treat existing problems however antibody mediated disease can respond to alemtuzumab.

      The use of rituximab to stop secondary autoimmunities is nothing new in terms of ideas but fair do’s if you have done it. It is an obvious thing to do when you have seen the data on B cell repopulation and even clearer when you view the relationship of T cell regulation, indeed ProfG and myself went to Boston about 6 years ago to seek support to do just that after alemtuzumab. Needless to say at that time there was no competition and no interest and now the problem is equipose and whether you go onto something else. We could not use rituximab in our hospital and so we could not do the study. If you are on alemtuzumab and it works for you I would be interested to know. I predict it will work but with alemtuzumab secondary autoimmunity occurs in 60% of people

      In terms of rituximab and anti-drug antibodies the binding rates can be up to 25-30% whilst on the drug, so if ocrelizumab to go from 0.4% to 5-10% would be massive but not considered to be a big deal at the population level. It depends how you view things..if you are one of the 5% its important. It is a theorectical thing and a hypothesis that may be wrong. However, once you stop someone taking rituximab why would you test the neutralizing antibody response 8 to 12 months later and if they occurred you wouldn’t care and wouldn’t report it. You treat people with rituximab when they get ADA you switch them to ocrelizumab one of the other humanized antibodies and then on to human antibodies or you simply switch to something else

      We have been in contact people in Sweden and Italy and have got rituximab samples sorted, but it seems that neutralizing antibodies are not always measured with rituximab. We only have so many pairs of hands and we need to finish off alemtuzumab. How do you explain that 80-100% of people make then and 85% of people make them within 1 month of the second infusion when there are no T cells or B cells in the blood (60% on first infusion)…However it happens. I think the conditions after ocrelizumab make the antibody response alot less likely than with alemtuzumab

      • I wish pwms had neurologists that cared about ADA response. Called the drug company, they don’t care. Ask a well respected neuro in the U.S. for ADA test, was told a test does not exist!?!?

        How does a pwms decide what to do next without this information. Was it the drug? Was it the anti CD20 mechanism? This information would guide the next step.

        My hope is that once the B cells repopulate, they are “friendly” and do not cause any further damage.

        • When there’s and ADA in the neighbourhood who you gonna call…No not Ghostbusters…Maybe, Dr Angry, however, every company has to have ADA detection capacity it is part of the drug development plan…Yep they don’t want to do it as a service.

      • I wish it was preventative, but unfortunately it was used to treat secondary alemtuzumab autoimmunities after they arose (I was unlucky enough to get 3 serious ones). It has certainly been effective for that where 1st line therapies were not, and I am hopeful that should I need it again ADAs won’t blunt it. Ocrelizumab is another off label option but its novelty and my screwy immune system means that rituximab is the currently the preferred option for future problems (MS or otherwise). Haematology/immunology types just seem much more comfortable with it after years (decades?) of use. I guess we cross that bridge should we come to it.

        Those binding rates are interesting, my experience with ritux infusions (happened twice 12 months apart) was that each time I was written up for 4 infusions one week apart but only given 3 as after the 3rd I had undetectable B cell counts making the 4th unnecessary. So do ADAs just mean that potentially more of the drug is required? ie. if 25% binding occurred for the extra dose would have been needed for the equivalent effect.

      • I am also curious as to how this ties in with the extended intervals of ocrelizumab dosing that you mention quite often. Are you still thinking that is a viable idea? Patients may be able to go further in between and be less immunosuppressed (and save money) but then they have to worry about the loss of efficacy that might occur. Not exactly a great option either? But I guess if they screen for it or check antibodies at least they will find out early that it’s stopped working and can look to another therapy. Hopefully this would only impact a minority of people, but it’d be tricky to manage!

        • This is the reason you have a hypothesis and test it, prove yourself wrong or you keep you mouth shut and see it it happens to a few people.This ADA happens with alemtuzumab and we seem to be able to predict failure in a number of people. It is unlikely to be a major occurrence as it does not create the perfect storm associated with alemtuzumab.

          Do I think extended dosing is viable…It is science, simple biology and safety. We are now only looking at extended dosing natalizumab that was approved since 2006. With ocrelizumab, it is happening anyway because people cannot afford the medication and are stopping or extending the dosing, there is a study with rituximab in Italy showing it is feasible and within 2-3 years they are down to 1 dose a year with no apparent loss of efficacy. There is data in Sweden. Again there is a very sound biology.

          Antibodies are not an all or nothing they build up with time and low titre (amuount) antibody is unlikely to be a problem also with ocrelizumab there is a massive dose 300-600mg given compared to 12mg with alemtuzumab this will swamp all but the exceptionally high titre neutralizers. They may not exist. In our alemtuzumab cohort the important functional cut off was 15,000 units some people were making 9,000,000 units and not depleting.

          “It would be tricky to manage” not really your need infrastructure in the alemtuzumab example we were sent a sample overnight from USA, the neuros got the result before they went to Work, you get all sorts of blood tests before treatments. Plus we are working on a miniture device to get the results whilst you wait, hopefully the investment will come to develop the protype further.

By MouseDoctor



Recent Posts

Recent Comments