Why Does Alemtuzumab cause B cell autoimmunity? The perfect Storm

Baker D, Herrod SS, Alvarez-Gonzalez C, Giovannoni G, Schmierer K.Interpreting Lymphocyte Reconstitution Data From the Pivotal Phase 3 Trials of Alemtuzumab. JAMA Neurol. 2017 Jun 12. doi: 10.1001/jamaneurol.2017.0676. [Epub ahead of print]

IMPORTANCE:Alemtuzumab, a CD52-depleting monoclonal antibody, effectively inhibits relapsing multiple sclerosis (MS) but is associated with a high incidence of secondary B-cell autoimmunities that limit use. These effects may be avoided through control of B-cell hyperproliferation.OBJECTIVE:To investigate whether the data describing the effect of alemtuzumab on lymphocyte subsets collected during the phase 3 trial program reveal mechanisms explaining efficacy and the risk for secondary autoimmunity with treatment of MS.DESIGN, SETTING, AND PARTICIPANTS:Lymphocyte reconstitution data from regulatory submissions of the pivotal Comparison of Alemtuzumab and Rebif Efficacy in Multiple Sclerosis I and II (CARE-MS I and II) trials were obtained from the European Medicines Agency via Freedom of Information requests. Data used in this study were reported from June 22 to October 12, 2016.MAIN OUTCOMES AND MEASURES:Tabulated data from T- and B-lymphocyte subset analysis and antidrug antibody responses were extracted from the supplied documents.RESULTS:Alemtuzumab depleted CD4+ T cells by more than 95%, including regulatory cells (-80%) and CD8+ T cells (>80% depletion), which remained well below reference levels throughout the trials. However, although CD19+ B cells were initially also depleted (>85%), marked (180% increase) hyper-repopulation of immature B cells occurred with conversion to mature B cells over time. These lymphocyte kinetics were associated with rapid development of alemtuzumab-binding and -neutralizing antibodies and subsequent occurrence of secondary B-cell autoimmunity. Hyperrepopulation of B cells masked a marked, long-term depletion of CD19+ memory B cells that may underpin efficacy in MS.CONCLUSIONS AND RELEVANCE:Although blockade of memory T and B cells may limit MS, rapid CD19+ B-cell subset repopulation in the absence of effective T-cell regulation has implications for the safety and efficacy of alemtuzumab. Controlling B-cell proliferation until T-cell regulation recovers may limit secondary autoimmunity, which does not occur with other B-cell-depleting agents.

Read the Paper it is still free to read, download now.

Today we continue to describe the paper

We have reported that alemtuzumab causes long-term depletion of CD4 T and B memory T cells, but not the immature B cells.

Remember the suggestion that B cells go back to normal is misleading. They don’t.

Now the next bit of misleading information and that is about T regulatory T cells

When you read the literature and now you also have the trial data it is evident that in contrast to the assumption that T regulatory cells increase after alemtuzumab, their absolute number decrease so there are few around within the first few months after alemtuzumab treatment.

This creates the perfect storm for secondary B cell autoimmunity to develop

1. Using animal studies we have demonstrated two things: CD52 depletion does not clear B cells out of lymph glands as well as it does the blood.  This depletes a population of CD8 regulatory cells that prevent immune tolerance induction, which may allow autoimmunity to develop.


Animals studies with alemtuzumab in human CD52 transgenic mice goes further and shows that it does not deplete cells from the bone marrow (BM or even the spleen) very well.

Turner, M.J., Lamorte, M.J., Chretien, N., Havari, E., Roberts, B.L., Kaplan, J.M., and Siders, W.M.Immune status following alemtuzumab treatment in human CD52 transgenic mice. J. Neuroimmunol.2013; 261: 29–36

The suggestion is that this occurs in humans also


2. In contrast to the suggestion that CD4 T regulatory cells increase, as we saw yesterday they actually decrease. In the
phase III study this depletion was by about 85-90%, confirming the decrease. 

CD4 T reg cells often work by cell-cell contact therefore there are few to control the autoimmune immature B cells generated in the bone marrow.

At baseline (CARE-MS I) there are 0.035 x 10*6/mL  = 0.035 million cells = 35,000 cells/mL dropping to 0.005 x 10*6/mL at 1 month (5,000/ml so more than the 2/mL reported in the paper by Cox et al. see yesterdays post) and 0.010 x 10*6/mL whereas the Immature B cells go from 0.01 cells/cuMM, to 0.15 at 1 month and 0.028 at 3 months.  So an effective decrease in the ratio of Immatute B cells to T reg cells.

How does it do this I don’t know (Big hole in my argument that needs filling)…  I need to do more reading

3. During this time, because alemtuzumab does not deplete B cells from the bone marrow as well as it could. There are lots of immature B cells waiting to stream out of the bone marrow to repopulate the blood. It is known from a number of other studies that when immature B cells develop in the absence of T regulatory cells then autoantibodies (autoimmune antibodies that react to your own tissues) are formed because the autoreactive B cells are not kept in check.

This idea was ignored because the people had the idea (have been given the idea) that T regulatory cells were increasing in number.

If you have a natural defect in regulatory T cells, ITP (autoimmune disease of platelets that stop blood clotting) develops…This is one of the problems with alemtuzumab treatment and is the reason why you have to have your bloods screened every month for 4 years after your last dose of alemtuzumab.

4. This lack of control of immunity due to lack of CD8 suppressor cells and CD4 regulatory T cells (a) Allows the secondary B cell autoimmunities to develop 

In terms of  secondary autoimmunities the 5 year data is known for the CARE-MS I/II studies. 

In the regulatory submission the occurence of secondary autoimmunities was presented of what occured during the trials. Has this presented at the AAN 2016 

(b) This allows anti-drug responses to develop too.

5. Autoantibodies do not typically form during the period of B cell expansion and require many months to a few years to develop. This is because these are going to be CD4 T cell dependent antibodies and so without T cell help for the B cells, they can’t produce their antibodies. These T cells recover 6 months to 3 years after depletion. So this is why it takes time for autoimmunities to develop

6. T regulatory function and CD8 T cell numbers will recover and probably do so at a faster rate that the memory B cells (or memory T cells) , so when they come back the immune regulation is in place and is active and so can stop autoimmunity in MS from recurring and so induce an induction-like treatment requiring few treatment cycles).

What are the Implications?

In the paper we suggest that the issue for autoimmunity is due to a marked B cell response in the absence of T cell regulation. We therefore hypothesised that a treatment approach based on T cells may fail. 

It has been suggested that the problem of autoimmunity is because cells that recover after alemtuzumab have restricted clonality (restricted diversity) as the memory T cells are expanding. Therefore, it was proposed that driving re-population via the thymus, to create a diverse T cell response would limit this.

Jones JL, Thompson SA, Loh P, Davies JL, Tuohy OC, Curry AJ, Azzopardi L, Hill-Cawthorne G, Fahey MT, Compston A, Coles AJ.Human autoimmunity after lymphocyte depletion is caused by homeostatic T-cell proliferation. Proc Natl Acad Sci U S A. 2013;110(50):20200-5.

Based on the data we presented in the new paper the expansion of memory T cells is a relative phenomenon and their actual number drops remarkably.

We had seen this phenomenon in our mice after CD4 T cell depletion and with a bit of reading, it looks like that when-ever T cells are depleted, this apparent memory cell expansion relative to naive cells occurs. However, this does not cause secondary autoimmunities. 

Furthermore, if the problem is with the autoreactive B cells, and not the T cell, then driving new specificities to arise from the thymus is going to create more opportunities to create new T cells to help the autoreactive B cells that may form. So rather than stopping B cell autoimmunities, it would suggest that it could do the opposite. 

I tell you this because the trial to promote T cells coming from the thymus was done.

Keratinocyte Growth Factor was trialed to drive T cells via the thymus to Prevent Autoimmunity After Alemtuzumab Treatment of Multiple Sclerosis (CAM-THY), (NCT01712945)

We will have to see what happened, but when I brought this up in a comment recently, someone from the blog pointed me to Prof Cole’s Blog site and it seems the trial has been stopped. Make of that what you will, we have to see the published data to see what happened. 

However the solution we would suggest is that if you limit the hyper-repopulation of B cells until T regulation occurs then it may stop the autoimmunity from developing.

We proposed doing such a study about 3 years ago using low-dose rituximab to deplete B cells, which would target the immature and mature B cells. When rituximab is used sometimes it is given as a split dose at baseline and 1 month later. So ritiximab after alemtuzumab. There was no enthusiasm for this approach. I wonder if  that is because it uses another companies drug. 

Some people think that long-term B cell depletion leaves you at risk of infection, so you would want a trial to show it is safe before we can recommend such an approach. 

What about teriflunomide after alemtuzumab? As teriflunomide kills proliferating cells it could stop the repopulation of cells making you permanently leukopenic,and stopping the T regs from repopulating, whilst you are on the drug. risking infections and perhaps even PML, so maybe less safe. Again this would need to be trialed

However the landscape may change or has changed. Ocrelizumab has arrived in the USA and will, I predict, arrive in Europe. Oral cladribine may arrive too, so have we missed the window to try a treatment to limit alemtuzumab-induced autoimmunities? 
Would you go on a trial to try reduce the autoimmunity? or is ProfG  right to suggest that it may become a drug of last resort?

About the author



  • Did the CAMTHY trial actually make things worse? An important question that needs answering urgently.

    • gtr.rcuk.ac.uk/projects?ref=G1100114

      "We will be disseminating this negative trial result to prevent others using palifermin to promote thymic lymphocyte production".

  • Are any of Team G attending the MS Frontiers Conference at the end of June 2017? This session on the Friday looks very relevant:

    12.30 EBV infection empowers human B
    cells for autoimmunity – role of
    autophagy and relevance to MS –
    E Morandi

  • Thank for the coments

    Would like to know if those imaturre B cell are overshoting, with no control from T cell, liquely to cause another autoimmunne disease

    What is gonna happen to the primary autoimmunne disease(MS)?
    Is gonna come back when they mature?

    • The immature cells produce the secondary B cell autoimmunities (Graves, Hashimotos ITP, Good pastuers etc etc etc.

      The MS may be a product of the memory cells and if they return then MS may come back, could immature, become new memory …probably I think this is evident by relapse with HSCT, but the hope is that things are reset in a different way.

    • To sum up, immature B cells can cause everything but MS is absence of Tregs, while B memory cells cause MS but not other autoimmunities in full presence of Tregs.

      Sounds like Escher biology.

    • The immature cells create antibody-mediated autoimmunities, MS is not yet a definitive antibody-mediated autoimmunity otherwise it would respond to plasma exchange? Unless antibodies are generated in the CNS.
      If they actually caused MS how would you know? 50% of people relapse is that because the immature B cells cause MS? However there is only 1% ITP, 1% cause of MS would not even be noticed

      B memory cells cause loads of autoimmunities including the B cell autoimmunities, look at the benefit of anti-CD20.

    • This is clearly a product of genetics as the problems of the autoimmunities do not occur when people with cancer get treated with alemtuzumab

    • In this paper they say intrathecalIgG should be the next target
      Depending on the
      primary antigenic stimulation, the estimated half-life of longlived
      plasma cells varies from a decade to one hundred
      Is this true?

    • "Half-life to one hundred years" maybe if is a tortoise or a bowhead whale but there are not many things that live 200 years so for humans no.

      However, the immune system is designed to give use life long immunity

  • Thanks a lot for replying
    So it make sense trying to control those B cell from overshoting (maybe with Rituxan )until there is a pool of t cells able to control them ?
    Is this possible?
    Thanks so much

  • It seems that memory B-cells are becoming the primary target in MS. Your black swan but falcon post furthers the point. The question: why use an anti- CD52 drug and risk secondary autoimmunity when a CD20 B-cell depleting drug, rituxan/ocrelizumab would eliminate the depletion of Tregs?

    • The is why profG has suggested that alemtuzumab may be a treatment of last resort, we showed data suggesting ocrelizumab has induction therapy. once they

    • Is it?
      We would have to do a head to head.

      If B cell depletion is the mechanism of action they should be similar.

      If we look at alemtuzumab trials the disease duration was short years compared to other studies what is the disease duration in pivotal ocrelizumab trials?

    • I read a post here earlier showing that Alemtuzumab initially depletes memory B cells almost totally. Ocrelizumab in the doses given in MS will not do that. There's your reason for it's inferior status. If you mega dosed it and annihilated all B cells then it would (potentially, if B cells are the black swan) have the same (or better) efficacy.

    • I haven't seen any memory B cell data from ocrelizumab but i think the CD19 population is kept rock bottom forever, this will be its downfall. So that would include the memory B cells. Check out slide 30 in tomorrows presentation

By MouseDoctor



Recent Posts

Recent Comments