Inhibition of B cell survival factors does not inhibit MS


You know me, I love to write papers, pharma seem to hate it, so I say to them just give me the data and save yourself a few grand paying medical writers. OK, they can’t dictatate what I am going to write, but the data generally speaks for itself. They just don’t want to let the data out:-(.

After the excitement of HSCT- thursday…. and friday, its back to reality. This post should have arrived yesterday, it didn’t, allowing you to have a chat about HSCT.

If you can cast your mind back a few years ago, we asked the question of who should be the custodian of clinical trial data? Should it be pharma? or should the people who were the trial participants have their say?

Ben Goldarce had written about Bad Pharma hiding data, So they set up a data sharing site called clinical trial data Eli Lily had signed up to this site, which agrees to give access to data in trials done after 2012.

We knew that Eli Lily had done a trial before 2012, with a drug (tabalumab) that blocked B cell activating factor. This inhibits antibody producing cells and mature B cells so it was being developed for Lupus, but when CD20 B cell depleting antibody worked they thought blocking B cell growth factor would work too. But a problem is that B cell growth factor inhibition does not block memory B cells, which we have suggested could be important in MS. So it should not work. So what happened?

As of 2017 they had not reported the result. Did it not work or did it make it worse? Atacicept blocked B cell activating factor too and made more memory B cells and made MS worse. Tabalumab made more B cells based on arthritis /lupus data, did it make MS worse too.? I think Annonie mouse will remember this. We kept writing to them over and over again, my real name is not MouseDoctor but Ferret. Once the teeth lock they dont let go and eventually was promised the data, but it never arrived. ProfG even had a telecom with them.

Now you know that when the data is positive then pharma want to control how the information is released. This has as much to do with marketing as science. Things that are not so great, often gets either buried or comes out as an info-blip that essentially no-one sees.

Data gets dumped in an ECTRIMS or AAN abstract so it goes in the public domain and is citable, but it is never published and the poster soon disappears from the internet site where it was depositied. For this reason every pharma poster from ECTRIMS should have to be put in the ECTRIMS oline library so at least you can see it and it stays there. I am pleased that one company has allowed us to reproduce some of the data from a few such poster but more on that in the future.

Anyway, Guess what? We must have hit a nerve or they got fed up with us but Lily actually did an abstract and dumped the data in ECTRIMS. I missed the poster:-(.

Pharma may say it is not economical to publish every thing, as they have to pay a medical writer to do it, as neuros are perhaps too busy, but importantly there are rules of what pharma is allowed to say. Medical writers know the rules. Get a bit of speculation in a paper and some people bowels turn to jelly and they soil themselves. For some people if you don’t show the data, then it isn’t true.

I remember a talk I did all about anti-CD20 and after a 30minute talk….I was asked what happens with ocrelizumab by an opinion leader? I was so shocked by the question thinking, weren’t you listening? Therefore I apologised for my accent and that the questioner didn’t understand me, sometimes Americans don’t do Northern Accents. Perhaps rudely I said, I have just spent the last 30 minutes answering the question. Weren’t they listening?…I am now guessing they were not assimilating, as there was too much speculation.

Sure rituximab is not ocrelizumab, is not ofatumumab, but I think we can be confident that the cells hit by rituximab are hit by ocrelizumab (only a bit harder). Anyway I digress. So back to the unpublished pharma data.

Anyway now the data was reported. I wrote to Lily to ask if there were any plans to publish the data. They said they had no plans, this was even more strange because I had noticed that they had put the data at clinical There it would stay with nobody noticing it. So I thought why not write this up so at least it becomes visible. I get a paper for no effort. Who was I kidding, 7 months of review.

To my surprise, I got it wrong, tabalumab did not make MS worse, it did nothing. So what is the difference between tabalumab (no effect on MS) and atacicept (makes MS worse)? Was it that one made memory B cells where as the other didn’t. Now there are studies that tabalumab does not do anything to memory B cells, but in this study there was an increase in memory B cells. But no worsening. So who better to question our original ideas but ourselves

The question is how much did atacicept make memory B cells worse. I realized that Merck had this data but they haven’t reported it…So come on Merck, you have had enough time to get the go ahead from the authors of the MS Study. Put me out of my misery.

Now you can ask Did atacicept really make MS worse, because the primary outcome, which was MRI was no worse than placebo, just as tabalumab did not make the MRI worse. So we can at least say that tabalumab and atacicept did not inhibit peripheral memory B cells and they did not inhibit MS.

However if there is a difference between tabalumab and atacicept. What is the difference? They both inhibit BAFF but tabalumab inhibits APRIL. Therefore is the apparent worsening because of blocking both BAFF and APRIL or is it because of inhibition of APRIL. APRIL binds to two receptors TACI and BCMA. BCMA is largely expressed by plasma cells in humans. If CD20 removes the important B cell subset then this is not because of an action on plasma cells, because plasma cells do not express CD20. There is essentailly one CD20+ subset that express APRIL receptors and these are memory B cells. So is this more info pointing a finger at memory B cells?

I guess that is the easy explanation but guess what? The referees suggest that T cells express APRIL receptors and that is the important bit….Yarn.

Gene expression of TACI on T and B cells

Looking at the expression are T cells on the top of your thoughts

Failed B cell survival factor trials support the importance of memory B cells in multiple sclerosis. Baker D, Pryce G, James LK, Schmierer K, Giovannoni G. Eur J Neurol. 2019 Oct 16. doi: 10.1111/ene.14105. [Epub ahead of print]

Background: Clinical trials are probably the most informative experiments to help understanding of multiple sclerosis (MS) biology. Recent successes with CD20‐depleting antibodies have focussed attention towards B cell subsets as important mediators in MS.

Method: We report and review the trial of tabalumab (NTC00882999), which inhibits B cell Activation Factor (BAFF), and contrast this with inhibition of A Proliferation‐Inducing Ligand (APRIL) and BAFF using atacicept (NCT00642902).

Results: Both tabalumab and atacicept induce depletion of mature B cells and inhibit antibody‐formation, but they fail to deplete memory B cells and do not inhibit relapsing MS. Atacicept is reported to augment memory B cell responses and may precipitate relapse suggesting the importance of APRIL. However, BAFF inhibition can enhance peripheral blood memory B cell responses, which was not associated with augmented relapse.

Conclusions: Although other interpretations are possible, this data further supports the hypothesis that memory B cells may be of central importance in relapsing MS, as they are the major CD20+ B cell subset expressing APRIL receptors. It also suggests that quantitative and/or qualitative differences in B cell responses or other factors, such as immune‐regulatory effects‐associated with APRIL, may be important in determining whether MS reactivates following neutralization of peripheral B cell maturation and survival factors.

About the author



  • Mouse,

    I admire your perseverance, but does this work benefit people with MS? I don’t mind you raking over the old coals from an old trial, but would prefer something more positive i.e. a study testing an hypothesis which results in a positive result and improves our understanding of MS (particularly smouldering MS).

    • Does this work benefit pwMS?….It depends if people build this into their world view?

      As to a positive result …it could do.

  • This post is interesting but I have nothing intelligent to say about it I just wanted to break the monotony of the ‘recent comments’ section and to say that the ferret is cute. 😉

  • Thanks for post MD.

    2 questions a little off topic re: EBV controlled memory B cells in follicles in the meninges of the brain:

    1. How do you measure them other than autopsy?

    2. I did a brief literature search about treatment of CNS B-cell lymphoma which show intrathecal methotrexate and rituximab as a treatment. These have been tried with very limited success in progressive MS. What other drugs are CNS penetrant and will attack these EBV controlled B-cell clusters other than the primed CD8 T-cells against EBV controlled B-cells by Dr. Pender?

  • At week 2, serum BAFF levels had increased significantly (as analyzed by the Wilcoxon signed-rank test), with a 2.2-fold increase on average, in association with complete depletion of CD20+ B cells. By week 12 after ocrelizumab initiation, serum BAFF levels tended to decrease, and had decreased significantly at week 24 compared to the levels at week 2. In patients with detectable levels of B cells at week 24, the majority of repopulating B cells were CD24+CD38+CD19+ transitional B cells.
    Conclusions: Together, our results give rise to the concept that a suitable maintenance medication could most efficiently foster regulatory B cell properties after ocrelizumab therapy.

    Impact of Ocrelizumab-induced BAFF on early reappearing B cells in multiple sclerosis

  • The Baffling April Mysteries

    Wondered when this would get a mention here. Yep remember it well MD 😉

    In an ideal grown-up world would be lovely to see Pharma (and academics) immediately share such data for the common good. As a little piece of a complex puzzle, failed trial data is as important as successful results 😉

By MouseDoctor



Recent Posts

Recent Comments