The case for more trials with ocrelizumab. Can we use reduced dosing of ocrelizumab?-Part I


#isitoK that Pharma do trials on disabled people, but do not publish the results if they are not advantaveous to the company?

If you can PLEASE, PLEASE. PLEASE go to the website of the journal and Tweet the link. It will help get this seen in search engines and help make the content known.


Ocrelizumab is currently doing very well in sales and is approved for 6 monthly dosing. B cells are part of your protection from infection and you do not want to have unnecessary woman or man-flu.

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. MedRXiv 2020

Objective: Ocrelizumab inhibits relapsing multiple sclerosis when administered every six months. Based on potential similar memory B cell depletion mechanisms with cladribine and alemtuzumab, we hypothesised that CD20-depletion of B cells by ocrelizumab may exhibit a duration of response exceeding the current licenced treatment interval.

Methods: Internet-located information from regulatory submissions and meeting reports relating to the unpublished open-label, phase II ocrelizumab extension trial (NCT00676715) were reviewed. This followed people (54-55/arm) with MS, who switched from placebo or interferon-beta to ocrelizumab for three 600mg treatment cycles (week 24, 48, 72) or people treated with ocrelizumab for four 600mg treatment cycles (week 0-72), followed by an 18 month treatment-free period.

Results: CD19+ B cells were rapidly depleted within 2 weeks and slow CD19+ B cell repopulation began about 6 months after the last infusion with median-repletion of over 15 months. The reduced annualized relapse rate during the published efficacy study appeared to be maintained in the extension study and there were no new T1 gadolinium-enhancing or T2 lesions detected in the treatment-free period. Importantly, within these extension cohorts, there appeared to be fewer adverse events and infections events.

Conclusions:  Ocrelizumab appears to induce durable relapsing disease inhibition, within 3 treatment cycles Therefore, it may be possible to reduce the frequency of dosing to maintain efficacy, whilst limiting infection and other risks associated with continuous immunosuppression. Further studies are now clearly required to determine whether this data is robust, as few people seemed to complete the study

What does this suggest. It suggests that you may be able to get the same benefit from dosing with ocrelizumab less frequently than every 6 months. It may also suggest that ocrelizumab could be like alemtuzumab and cladribine and be an immune reconstitution therapy, so maybe you get long term benefit from 3-4 cycles of treatment. Given that the side effect and monitoring profile is lower than with other IRTs. There could be an advantage. Maybe it could provide you with enough protection to have a pregnancy without needing treatment..wouldn’t that be good if it were true. Maybe you may avoid infections that are a real problem for a few people

This data isn’t good enough to give us a definative answer, but it does clearly say that we should be asking these questions and doing these studies now! We should have done them eight years ago when this data first surfaced, because by now we would have the answer.

Why didn’t the neuros doing the studies ask these questions nearly a decade ago?

Why do this and publish other people work. Well first I think that it must be said neuros do this all the time as they are the mouth pieces for what is a pharma study. The difference between a neuro-led paper and this paper is that I have actually written the paper and not a medical writer paid by pharma. Next because it is an issue about risk and benefit and if we can reduce risk for the same or better benefit, I think it is the right thing to do. What do you think? I suspect some or many of my peers may disagree.

This paper perhaps gives the publishing concept abit of a shake and we will take some flak for this, because “this is not how it is done”

If you want to read about the publishing process and what MedRXiv (pronounced Med Archive) is, read on.

There is a push for a new way of publishing manuscripts. Generally you do the work, write the manuscript, submit to a journal for review and you get publication. There is public awareness once the work goes on indexing sites. People write reviews and publish them. This is a slow process.

If a company has a successful trial that will help them make money, they do the work (i.e. a trial), the neuros involved with the study get schlepped around the world to talk about the results to build-up interest within the neuro community (i.e. AAN or ECTRIMS presentations), they put the trial in the public domain to fullfill their responsibilities of public disclosure (An abstract is submitted that may or may not be published), they pay medical writers to write a paper with the names of the neuros in the birght lights. They then get the paper in a high impact journal and perhaps pay the journal loads for reprints etc etc.. You get public disclosure of the paper that goes on indexing sites. People buy the treament. Loads of reviews are written about the original paper and no-one argues with the process.

However, scientists got sick of the slow publishing process about getting ideas into the public domain. They may get scooped so they thought of an alternative route. Therefore you can do the work, write the manuscript and then put on a pre-print site and it goes open access. It allow discussion and public disclosure of a paper, you submit to a journal and whilst waiting for months or years before the work is published, you get the ideas are out there quickly.

People working with Physics (ARXiv) did this since 1991. The idea was introduced to biology (BioRXiv), which followed in 2013, and the Medicine with MedRXiv in 2019. It all sounds good but I predict this will be like other other human endevours and like climate change which changed slowly for hundreds of years, but once the industrial revolution occured properly, it quickly turns into SH1. This is the publication process, which is rapidly descending into a mess.

Some big journals do not allow this activity as they say it is prior public disclosure and so the publication lacks novelty. These journals are the really big hitters with massive impact factors. The science journals are eroding and caving in to this, but the big medical journals aren’t yet as it is probably too lucrative. The big journals know it is the bid science that gets hundreds of citation that drive their impact factors and so if a genomics consortium put their work in a pre-print site, this is a price the big journals will pay for their impact factor and all the financial rewards it gives them. So an increasing number of journals do allow this. Indeed some have made a seemless submission process from pre-print site to journal. Therein lies the rub.

If you don’t care if the paper goes into one of these big journals, it provides an alternative. This is a new approach that will change the publication process.

Link it up with an indexing site and you remove the need for Journals. Without the journals, you change the Science World Order, but you will probably loose information as the publishers go bust and the electronic data on their servers are lost. You don’t publish by impact factor, the important comoditiy, will be the citation of the work. The more influential the more citations.

However, you probably want some form of peer-review in the system. After all there is a lot of tosh out there. This is what the journals do. They get academics to work for free to do this.

The preprint sites are going to shoot themselves in the foot because they will make themselves into journals, even super journals, by offering to review and house the final work. However, with time it becomes a money making excersise. Pre-prints sites will spring up every where creating a hierarchy of preprint sites and creating the same mess that uncontrolled open access has created.

As for this paper, I will let you know one day why we chose to deposit on a pre-print site. This is of the one reasons why we have avoided social media and waited to report on the blog.

However, today’s paper brings a potential new twist. A company does the work, schleps the neuro to a meeting so the study is made public (abstract), but then does no write the manuscript. They want to dump the data so it is never heard of again. The neuros play ball and the information is removed from the public domain and the idea is never heard of again.

However, if you know this happens, as you do if you have been following the blog for the past decade, you can save copies of the data when it is transiently dumped in the public domain. Then you can wait to seee if the neuros publish the data. What if they don’t? People have risked “lives and limbs” to get the data. Is it right that companies control who has access to it. Is it right that they can dump it. If the data is not good enough then they should not have neuros presenting it. But what happens when you think that the reason the data is not being discussed is because it is not advantageous to the company to discuss it. Many neuros will not dare to break ranks and discuss it.

However, what happens if there are potential influences related to the risk:benefit profile. Should we not get the information in the public domain so that it can be discussed. So here we are today.

This paper is live have a read it has been for a few weeks. However it has had few tweets and this shows you the importance of indexing sites, see what you think. Make a comment, you could review the paper? You can wonder

(a) Have or Have we not submitted the paper already?.. Has it gone to a high or low impact factor journal?

(b) Do we want it on indexing sites quickly or are we prepared to wait?

(c) Maybe we have given the neuros a chance to publish the paper, after this public nudge?

When we showed interest in 2018 and tried to get the trial data, we were told that the company were going to publish the data, both by ProfG and the Company.

Was it that we had awakened interest? Was it a way to put us off exposing the data?

We have given them over a year and nothing…..So now it is too late, it is now in the public domain and even we cannot remove it now.

We can however edit it. Once we have the real data we can update this

When, where and will the article be published and put on indexing sites?

What do you think are we right or wrong….get it tweeted

COI: ProfG was not involved in the descision to report this and was unaware of the publication. The neuros and MD have been sponsored by Roche in the past. Roche/Genetech who were not involved in the the content of the manuscript, or the decision to publish. This is not anti-pharma but we want to help ensure it is as safe as possible. Pharma should do the ADIOS study

About the author



  • “The difference between a neuro-led paper and this paper is that I have actually written the paper and not a medical writer paid by pharma”


    • Sorry the other post, you refer too has been removed it was posted by mistake, the post you commented on will arrive this saturday. As it is “What have you done?” this week.

  • We have to think what have we learnt from other conditions, like NMO “made in Korea” where adapted dosing reduces the number of infusions.

    Studies “Made in Italy” in MS report that dosing rituximab to memory B cell numbers can reduce the number of dosings required from 2 a year to less than one a year (within 3 years) and can maintain efficacy.

    Ley’s see what “Made in Sweden” with alot of rituximab use tells us….time to say ADIOS (Adaptive DosIng Ocrezliumab Studies). Do the studies.

By MouseDoctor



Recent Posts

Recent Comments