Spot the mistake? A self-fulfilling hypothesis


En face this research seems straightforward, but upon reading the methodology of the manuscript my first thought was ‘huh’?

Why the ‘huh’?

Well, if you’re going to study on humoral immunity (i.e. antibody formation), the drug not to use is Ocrelizumab or any other anti-CD20 for that matter. The MOA (mode of action) of anti-CD20s is to target B-cell function and secondarily antibody formation. It is well know that a certain percentage of Ocrelizumab patients develop hypogammaglobulinaemia (low IgG levels) over time.

Since, one of the aims of the study was to measure EBV antibodies by checking EBNA-1 IgG, given that the MOA of Ocrelizumab is to deplete circulating B-cells that produce antibodies, what you have here is a self-fulfilling hypothesis. One can argue that since EBV lies dormant in B cells, by removing infected B-cells you’re reduce EBV reactivation, but again there are a lot of ifs in this line of thinking. It would have been nice to see some cell based studies instead of antibody levels.

But, there is more to this study than EBV analysis. The published pilot study is informative in terms of expected clinical outcomes, namely a slight improvement in EDSS scores (disability level); with 74% of participants remaining stable.

There was no improvement on cortical compartmentalized inflammation (as measured by LM CE or leptomeningeal inflammation) or stabilisation in brain volume loss or T1 black holes over a 24 month period. It should be noted that this was based on analysis of 29 participants, which is too small to show a significant treatment effect on these measures. Even smaller numbers had cognitive testing done and the findings was that cognitive decline continued despite treatment.

Admittedly, this piece of work was done over the pandemic and commendations need to be give to the participants and the researchers who managed to keep it going. Hopefully, they will have more data to report on in the next couple of years.


Mult Scler Relat Disord. 2022 Aug 6;67:104094.

Effect of ocrelizumab on leptomeningeal inflammation and humoral response to Epstein-Barr virus in multiple sclerosis. A pilot study

Robert Zivadinov Dejan Jakimovski Murali Ramanathan Ralph Hb Benedict Niels Bergsland Michael G Dwyer Bianca Weinstock-Guttman 

Background: Ocrelizumab is an effective treatment for relapsing and primary-progressive multiple sclerosis (MS). However, the effect of ocrelizumab on leptomeningeal (LM) inflammation is unknown.

Objective: To investigate whether ocrelizumab reduces LM inflammation by reducing the exposure to Epstein-Barr virus (EBV)-infected B cells in relapsing-remitting (RR) MS.

Methods: This was a Phase IV, prospective, open-label, single-center, observational, longitudinal pilot study of RRMS patients who started treatment with ocrelizumab (NCT03025269). Clinical, MRI and EBV-antibodies outcomes at baseline, 12- and 24-month of the study were evaluated. The MRI outcomes included T2, T1 and T1-contrast enhancing (CE) lesion counts and volumes, LM CE count, and percentage brain volume changes.

Results: 27 RRMS patients started ocrelizumab and 24 remained on the treatment for whole duration of the study. Most patients remained stable (74.1%) or improved (18.5%) in their disability status. At baseline, 42.3% of patients showed LM CE lesions. The majority of patients remained stable in their LM CE status over the follow-up (72.7%). A significant decrease in percentage volume loss of cortex (p=0.009), GM (p=0.01) and thalamus (p=0.038) was detected, while T1-LV increased (p=0.02). A significant decrease of EBNA-1 IgG (p=0.013) was evidenced. An infusion-related allergic reaction led to discontinuation of the medication in one patient at first dose.

Conclusions: Treatment with ocrelizumab was safe and clinically effective. Brain volume loss and accumulation of T1-LV occurred. While ocrelizumab decreased humoral response to EBV possibly by reducing B cells, it did not reduce LM inflammation.

About the author

Neuro Doc Gnanapavan


  • >At baseline, 42.3% of patients showed LM CE lesions. The majority of patients remained stable in their LM CE status over the follow-up (72.7%).

    Wait a minute, doesn’t this mean that Ocrevus isn’t even completely shutting off focal inflammation (as represented by contrast-enhancing lesions) in a large proportion of these patients? That’s pretty much the only thing Ocrevus is supposed to do well in the first place (we know it has only modest effect on BVL, we know it has only modest effect on disability progression). Isn’t it time for doctors to stop drinking the kool-aid and start recognizing that anti-CD20 is not an aggressive enough treatment option?

  • NDG,

    Thanks for this.

    “There was no improvement on cortical compartmentalized inflammation (as measured by LM CE or leptomeningeal inflammation)”.

    Compartmentalised inflammation is a term I’m seeing more and more. Is it the same as smouldering MS and / or chronic lesions? I’m not a scientist, but surely researchers must realise that a drug needs to get into the CNS / areas where the compartmentalised inflammation is occurring to have an effect! I’m assuming this is the aim of the SIZOMUS trial ie to get a drug into the CNS not just a drug that effects the immune system outside the CNS.

    • Good question Kat, I believe the term compartmentalized inflammation gets used for a lot of other things which confuses people’s understanding of what it means in MS. I use it to describe inflammation that is taking place exclusively in the brain not primarily due to immune cells crossing over from the circulation via break down in the blood brain barrier. Compartmentalized inflammation is self sustaining caused by B-cells and plasma cells in follicles in the meninges and Virchow robin spaces. This type of inflammation can lead to cortical/subpial lesions.

      Whereas smouldering MS refers to mainly chronic lesion activity, but can also include inflammation that is grumbling away in the background that includes just random nervous system inflammation and compartmentalized inflammation. Chronic MS lesions are old MS lesions which can either continue to be active or burnt out. I note all very confusing with some overlap in all three descriptions!

      Yes, very important for the drug to get into the CNS to target compartmentalized inflammation, and drugs do get in but it’s more difficult to get antibodies to cross over unless there is quite a bit of blood brain barrier breakdown to start off with. Small molecules such as Ixazomib (SIZOMUS study) cross over the blood brain barrier much more easily and has already demonstrated to target plasma cells so it would be nice if it can also do this in MS (a cell type that has so far not been targeted before).

      • Thanks for your detailed response. It can seem confusing – looks like the whole brain is on fire!

        “Virchow robin spaces” – did you just make this up?🤔 First time I’ve seen it.

      • NDG – cladribine is a small molecule DMT that crosses the BBB.

        Prof. K can back this up.

        Thanks for the post!



Recent Posts

Recent Comments