T cell immunotherapy of progressive MS…Time for a rethink or just a think!

Whilst it is with a lot of regret that the ASCEND natalizumab trial has failed in secondary progressive MS. So time for immunosuppressive agents to DESCEND into the bin.

I have to say this is what I predicted before it even started and importantly it is also what EAE predicted too.

Pryce G, O’Neill JK, Croxford JL, Amor S, Hankey DJ, East E, Giovannoni G, Baker D.Autoimmune tolerance eliminates relapses but fails to halt progression in a model of multiple sclerosis.J Neuroimmunol. 2005 Aug;165(1-2):41-52.

An experimental model of secondary progressive multiple sclerosis that shows regional variation in gliosis, remyelination, axonal and neuronal loss. Hampton DW, Anderson J, Pryce G, Irvine KA, Giovannoni G, Fawcett JW, Compston A, Franklin RJ, Baker D, Chandran S. J Neuroimmunol. 2008; 201-202:200-11.

Al-Izki S, Pryce G, Jackson SJ, Giovannoni G, Baker D. 
Immunosuppression with FTY720 (Gilenya)is insufficient to prevent secondary progressive neurodegeneration in experimental autoimmune encephalomyelitis. Mult Scler. 2011; 17(8):939-48. doi:
Hampton DW, Serio A, Pryce G, Al-Izki S, Franklin RJ, Giovannoni G, Baker D, Chandran S.Neurodegeneration progresses despite complete elimination of clinical relapses in a mouse model of multiple sclerosis. Acta Neuropathol Commun. 2013 Dec 23;1:84

Sadly many of my EAE colleagues just stick their head in the ground and say Progressive MS is T cell, T cell, T cell and dismiss anything that does not say T cell, T cell, T cell. 

This is the mantra from many of the big labs in the World. 

Other people have been beaten into submission like some sort of Stepford Wive so they follow the dogma.
The Pryce et al. paper from 2005 had more experiements in true relapsing EAE than any other paper in history, yet our peers didn’t want to listen and the paper was rejected by so many journals it was quite depressing. We just had to get it out in the end. 

However, we have been proved right again and again and again.

Do I think that inflammation that can respond  to immunosuppressives in progressive MS.  The answer is yes but we need something in addition. ProfG may also  argue that we need to move away from lower limb mobility to.

So time and time and time again immunosuppressive agents have failed to stop progressive MS, HSCT does not stop progressive MS and so the ultimate immunomodulator fails and so will all the rest. However, people don’t listen……..or think

How many more trials like this are there going to be?
There are a load ongoing and so prepare yourself for more disappointment. 

Why do they keep doing this? I can only guess. 

I went up to one company person doing a tolerance study in progressive MS at a meeting and said despite them conning themselves that there was a glimmer of hope in the phase II trial data, their phase III was going to fail.  

                   It did and the company is no more.

However, because relapsing-remitting MS is being sewn up by pharma drugs, and unless you are a mega company who can pressurise the regulators to allow you to do a placebo control trial, you can’t afford to do a non inferiority against a current MS drug as you will have to buy it at $100,000+ (for 2 years) a person to buy the comparator drug. 

So they go mirror, mirror on the wall, We believe that T cell immunotherapy will cure it all

This allows them to do a placebo control to save a buck, but in doing so they flush their assests down the toilet and their companies into liquidation. 

There will be a few CEOs shedding their shares on this news because they will soon be printed on toilet paper.

Don’t blame EAE. Blame the humans.

About the author



  • Another progressive trial fails and a ten year old EAE research paper is dragged out of the archive to show that The Mouse Doctors predicted the failure ten tears ago. Oh, the benefit of hindsight. Silly EAEoligists squabbling over their useless model. I've had MS for 15 years and all this EAE research has not benefitted me one iota. Until you can build an EAE model which incorporates EBV, vit D deficiency, gender differences, smoking, and the human risk genes associated with MS, you wasting your time. DEATH TO THE EAE MODEL- 50 years of failure. At least it gave some previously unemployable nerds a job.

    • Thanks for the positive feedback. The fact that a ten year old paper has proved to be right is both a source of pride and also regret. That colleagues have kept their head in the sand and ignored those findings is a matter for their conscience. We need neuroprotectants for MS, which this trial confirms, and of course are and continue to be, identified via (mostly our) EAE studies. Certainly not 50 years of failure.

    • But as an MSer I have no interest in the cat-fighting among EAEologists. The model is flawed and isn't delivering. It's been going for 50 years yet we see no neuroprotective therapies, remyelination therapies or neurorestorative therapies. We keep going round the same buoy – EAEologists identify target to cure MS, pharma take idea into trials, trial fails.. I do have some hope that the work done on mice for neuroprotection will bear fruit. But its the usual black hole syndrome. Trials get bogged down before they start, then they go quiet for years and years. A good example is the Charcot project (not EAE) it was 22 people taking a couple of pills a day for six months and having some MRIs. why has it taken 3-4 years? And this is just the proof of concept stage.

    • Thanks MD2 I was wondering how to respond to this without being rude..

      Inflexible in views and unable to assimilate information is a mentality of the t cell t cell t cell brigade it seems that they are not the only ones::-(

    • You are blaming EAE for the lack of progress……never once have i heard someone critisize the accountants…. However they are the people in pharma that call the
      shots they decide what goes forward and what doesn't.

      ProfG and I went to pharma in 2001 to ask for a drug that was neuroprotective in animals…..the accountants said no.
      I have written at least ten times about the charcot project. W hen it is published you will find out.

    • Thanks MD MD2. I am not a doctor but I read everyday and I don´t know how someone still think that all is T cell. I read about neuroprotectants and I make my own trial with myself. I not have options.

    • I think you guys are the bomb – I think this blog is unparalleled in the engagement and information it provides to everyday people. I can't speak highly enough of all of your efforts when it comes to this blog and from reading it I can only conclude that the clinicians on the team are highly caring people (in addition to being doctors, researchers etc). But sometimes people who are frustrated and and their life is down the drain just vent away at whatever the easiest target is – it's not always personal against you. Sometimes kindness in face of rudeness can achieve more than fighting rudeness to rudeness. As I said, I think you guys are the best and I'm forever indebted to the team for the information it has given me. I don't have MS (my partner does) and sometimes does moods fluctuate out of frustration and unfairness of what's happening to their bodies… just my respectful opinion.

  • I laughed a lot with this setup … But why then decide not to look at what can act on B cells, the microglia and macrophages? …What is the lock that see themselves? Afraid to admit that the dogma failed? …

  • Hi. Thanks as always for the informative posts and commentary.

    I was really fascinated by this blog's explanation of the Asynchoronous progressive MS theory. Given that the majority of participants in the trial had EDSS of 6 to 6.5, could that not explain why the trial, as short as it was, did not meet its primary objectives but showed that Tysabri slowed the loss of function in the arms and hands of some participants? (Putting aside the question of therapeutic lag).

    • This at the centre of the length dependent hypothesis that profG has been trying to get funded for the past few years….some people think these things sort themselves overnight…….the dont.

      If you have an establishment fixated with the EDSS you sometimes need some cat fights

    • Yes I came to the same conclusion once I read the previous post 🙂

      So indebted to you all for the in-depth analyses of the issues and the hypothesis, musings and thoughts you share with the rest of us.

  • I think no one actually thinks progressive MS can be cured by intense immunosuppression, except those on the fringe who have no clue about what is causing the progressive phase.

    We also don't have any treatments for people who sustain head trauma to repair damage, so progressive disease is in the same boat.

    Pharma is doing trials with immunosuppressive agents because they can. They make outlandish amounts of money for these drugs for RRMS so they are doing these progressive trials to give the impression that they are trying to find treatments for progressive disease.

    But the truth is they know that we are a long way off from repairing damaged nerve tissue so they keep doing these trials to give the impression that they are trying. The price of a progressive trial is a drop in the bucket compared to the task at hand.

    • I agree with you. We need neuroprotectants for MS, the evidence is overwhelming, yet pharma sits on its hands and continues to plug away with what it already has ie immunosuppressants for every flavour of MS.

  • http://www.wsj.com/articles/biogen-to-cut-workforce-by-11-1445427980?ru=yahoo?mod=yahoo_itp
    Biogen is the leader in MS therapies and the t-cell dogma for treating RRMS has paid handsomely. As a business it makes sense to provide therapies that: have a long life span ( MSers take these drugs for decades), are very expensive and profitable. If they can sell their drugs to MSers for a long time why does it matter that patients progress 10-15 years down the road? Business is good, don't mess with the model.

    • Why does MD2 think that it has the right to have a view on everything?

      $250m / (8000 employees x 11% cuts) = $206k per employee.

      That, is an average – I will let MD2 tell you how this compares to the industry we all love to hate: banking.

    • So with the bombshell that ProfG has just dropped that Biogen is cancelling its secondary progression study with BG-12 maybe the penny is dropping.
      The trial was called INSPIRE it has now EXPIRED, It this an omen for ProfGs INSPIRE trial

      The story was made that BG-12 affected Nrf2 and this should be actually be neuroprotective so it seems rather premature to do this. Therefore mechanistically there is reason to believe it can work, there was no real logic to ASCEND unless you believe T cel T cell T cell.

      The BG-12 trial is it only 6 months in I think so there wont be enough data to peep at so it does suggest cold feet. Maybe it does mean that the DMF really does not really get into the brain, so it would not be that good a neuroprotectant.

      So if bombs ar being dropped what next ?

      BAF312 /Siponimod which is fingolimod mark 2 is now in phase III trial for secondary progressive MS but as fingolimod failed in PPMS the risk of failure has to be high too, unless they know something we don't.

      Will they get cold feet too? One trial has a year to go.

    • That's a good article, but it clearly is going against the diatribe of what Team G has been pushing for so long.

      "In summary, these findings highlight that although timing of anti-inflammatory immune based therapies is critical, even intervention after the first clinical event is unlikely to prevent all long-term neurodegeneration."

      Immunosuppression as a means to prevent future neurodegeneration is questionable.

    • No I disagree. It says that even after one attack there are certain nerve tracts that are
      going to degenerate and if you allow more attacks there are more tracts that degenerate so start treatment as early as possible and here we have beenbut if you stop the number of attacks you will stop neurodegeneration so nothing different fom the team G mantra

    • Here is one quote from the article:

      "Indirect observations such as neuronal injury in normal appearing white matter and epidemiological findings that time to disease progression is age dependent,
      but independent of number of relapses, suggests that neurodegeneration is at least in part independent of immune-driven inflammation [3,4]."

      I think they are clearly making the case the neurodegeneration and inflammation are separate and stopping all inflammation will not cease neurodegeneration in the long term.

      This is pretty evident from the title of the paper:

      "Neurodegeneration progresses despite complete elimination of clinical relapses in a mouse model of multiple sclerosis"

      This is based off of an EAE model which you seem to despise even though you are an EAEer, but it seems this model has shown what seems to be going on in MS.

    • "This is based off of an EAE model which you seem to despise even though you are an EAEer, but it seems this model has shown what seems to be going on in MS."

      Yes, it's our model so we can hardly despise it 😉

    • Thanks! All the studies quoted in the article are our work based on our model, which we think is the most realistic to MS. Forgive our immodesty:-)

    • There are to types of inflammation in the EAE the lymphocyte type inflammation that causes attacks and demyelination and nerve loss and then there is the glial type of inflammation that drives the smoldering progression they probably each need a different type of treatment.

  • I still think repair/neuro-protection is in the realm of science fiction.
    I can't see it happening in the next 15-20 years.
    Best we can do now is find the cause and prevent more people from getting MS, I think that is the more realistic option.

    • Great picture BTW Love that album.
      Neuroprotection for MS has already been shown to be achievable in clinical trials (simvastatin, phenytoin in optic neuritis, cannabis in a subset of the CUPID trial) and there will be much more to come, the PROXIMUS study should be of huge interest. So science fact not fiction, repair though is still some way off if it is really achievable which I have doubts about.

    • If repair is not achievable, how do we explain Dr Sadiqs early phase improvements on disability with neural precursor cells at the Tisch center . These are impressive results. Not perfect, but still early. Let's not extinguish all hope here

    • God effect said
      If repair is not achievable, how do we early phase improvements on disability with neural precursor cells. These are impressive results. Not perfect, but still early. Let's not extinguish all hope here on

      We have never said that repair is not achievable of course it is if the nerve survies

    • God effect said
      If repair is not achievable, how do we early phase improvements on disability with neural precursor cells. These are impressive results. Not perfect, but still early. Let's not extinguish all hope here on

      We have never said that repair is not achievable of course it is if the nerve survies

  • In Hampton et al (2013), you find some evidence for continued neurodegeneration following a relapse, even if the mice are treated early. Where were you measuring neurodegeneration? Was it just at the site of the lesion? Or were you taking a more global measurement?

    • Dave looked at sections in certin parts of the nervous system and so it wasn't looking globally. However in mouse EAEthey get lesions all along the spinal cord

    • Thank you. This makes sense, and, depending on the details, makes your claims about continued neurodegeneration in the absence of relapses much less radical than they initially sound. It seems to be consistent with the following model: after a lesion is formed (and the initial inflammatory damage is done), some set of local degenerative processes are triggered, and continue to cause degeneration *at the site of the lesion*. These degenerative processes do not involve the peripheral immune system, so they cannot be shut off by immunosuppressive treatments. These are the "smouldering" lesions which have been described elsewhere in the literature. The degenerative processes are spatially confined, and immunosuppression *will* prevent damage outside of the location of the initial lesion.

      Do you think that this model is consistent with your data? If it is, then it does suggest that immunosuppression administered early in RRMS will prevent most of the neurodegeneration associated with the natural course of the disease.

    • Yes this is consistent with what we see and one reason for profGs brain health statement and why we can not understand why people i.e. neurologists tolerate any lesion.

      It is also a reason why we support the idea of a treatment pyrimid with the immunomodulator on the bottom under neuroprotection and repair

      If done easy we dont notice any clinical effect from what is is being lost however if we wait too long the smouldering lesions persist and we do see the consequences. So if we treat animals early we can cure them,

      This is consistent with the atrophy data of alemtuzumab that moves atrophy rates back into the normal range..

    • "after a lesion is formed (and the initial inflammatory damage is done), some set of local degenerative processes are triggered, and continue to cause degeneration *at the site of the lesion*. "

      I beleive this degeneration is caused by the change in phenotype of microglia as discussed on this blog, but if the degeneration was restricted locally to the lesion location, why does whole brain atrophy occur? It seems odd that the microglia would be confined to local lesions while atrophy is occurring throughout the brain.

    • Brain atrophy indicates that nerve death is occurring *somewhere* in the brain. It does not necessarily mean that it is occurring everywhere. The model that I gave above would suggest that brain atrophy occurs for two reasons: inflammatory damage (which we know is sometimes missed by low-resolution MRIs), and secondary degeneration at the site of prior inflammation.

      I'm not fully confident that this is correct, though perhaps MD can provide more relevant evidence. It does, however, make at least one correct prediction: that the rate of brain atrophy will decrease following peripheral immunosuppression.

    • "Brain atrophy indicates that nerve death is occurring *somewhere* in the brain. It does not necessarily mean that it is occurring everywhere."

      The phenomena that occurs in multiple sclerosis is WHOLE BRAIN ATROPHY and this is what is being studied vigorously to implement it in clinical practice:


      If atrophy was occurring "somewhere" in the brain, it would be next to impossible to use this as a sign of neurodegeneration, especially when used in a trial setting where each patient's scan would have to be examined in detail to determine at what portion of the brain is undergoing atrophy.

      Whole brain atrophy relates to changes in normal appearing white matter as well as grey matter changes that do not show up on one MRI scan but can be detected by serial MRI scans. That's the point.

    • Brain atrophy is measuring the difference in brain parenchymal volume between two time points. Brain parenchymal volume measures the volume of neurons and glia in the brain. If there is destruction of neurons or glia due to inflammatory damage, then this will decrease the brain parenchymal volume.

      If you see volume loss, then this only implies that there has been loss of neurons/glia somewhere in the brain. The loss may have been confined to a specific region. For example, if I were to perform a lobotomy on you, that would result in a decrease in brain parenchymal volume, even though it would be entirely confined to a single area.

    • Maybe DrK can comment on brain atrophy but in terms of spinal atrophy,we know that atrophy is missing most of the nerve loss as this space is being replaced by glia and fluid so it is a massive underestimate of whats going on. Furthermore the loss in the spinal cord axons may lead to brain axon loss,.

    • MD, I did not mean to suggest that loss of neuron volume is easy to measure. I was simply trying to say that if you do see volume loss, it may reflect damage which was localized to pockets of inflammation. It does not necessarily mean that the loss of neurons was equi-distributed across the brain.

    • Agreed it will not be equal loss over the brain but focal to where lesions are or where they project to or from which may be some distance away

    • I think many researchers are coming to a different conclusion than you:


      "Widespread axonal pathology, largely independent of MRI-visible inflammation and too extensive to be completely reversible, occurs in patients even at the earliest clinical stage of multiple sclerosis. This finding lessens the validity of the current concept that the axonal pathology of multiple sclerosis is the end-stage result of repeated inflammatory events, and argues strongly in favour of early neuroprotective intervention."

      I guess this is why most researchers and neurologists have not jumped on the Team G bandwagon of marketing immunosuppressive drugs to prevent disability down the road.

    • If you look back at all we've said on the subject, we've always had an open mind on whether immunosuppressives can affect progression and have consistently promoted the early use of neuroprtective therapies. I still think that intervening to prevent new lesions as early as possible must have some bearing on the rate of disease progression (the long-term follow up data from those pwMS treated early in their MS with alemtuzumab should clarify things) but we have always also said (as in our papers above) that this is probably not sufficient and we need neuroprotective therapy as an add-on. We are working very hard to try to bring neuroprotectants into the MS therapeutic arsenal and the first clinical trials arising from our work is showing great promise so far. We now need pharma to get on board and end their fixation with immunosuppressives and move on to effective neuroprotective therapies for MS.

  • Anon 12:05. Thank you for this link. This is very interesting, but I'm pretty confused about the use of N-acetylaspartate levels as a measure of neuron loss.

    In the study you link to, there is no direct evidence of axon damage independent of focal inflammatory events. All that the study did was take an average measure of NAA levels across the brain. They found that these levels were lower in MS patients than in controls. This does not mean that the levels are lower everywhere in the brain though; it may simply mean that they are lower in areas of prior inflammation, and downstream of those areas. The main evidence that the authors have against this hypothesis is the magnitude of the NAA difference, which they believe is too large to be explained by the relatively small amount of inflammation that they found on their MRI. So there are three possibilities (the paper only considers the first two):

    1. The MRI is missing a lot of inflammatory damage.
    2. There is a non-inflammatory degenerative process which is causing nerve death, and hence elevated NAA levels.
    3. NAA levels do not scale linearly with nerve damage. This may happen for two reasons. First, it may simply be the case that there's a different scaling function, e.g. a logistic function. Second, NAA levels may not be a function of nerve damage at all, they may measure something else. If there's a different scaling function, then small amounts of damage may produce large differences in NAA; as a result, the authors' reasoning would be invalid. If NAA doesn't measure nerve damage, then similarly we can't draw any conclusions from magnitude of NAA levels.

    There seems to be a decent amount of evidence supporting the first position, some of which are actually discussed in the paper. I'd be happy to share the other data that I've come across in favor of that position. In favor of the third position, here are a few anomalous data points that I've come across. I'm not strongly advocating for this position, but I don't straightforwardly see how to explain them otherwise.

    The study that you linked to studied NAA levels in patients who had just been diagnosed with MS. They found that NAA levels in these patients were 20% lower than controls. But consider the results from the following study:
    In this study, the authors looked at NAA levels in patients >20 years post-diagnosis. They included patients with both benign and non-benign MS. They found that in both groups, NAA levels were 20% lower than controls. This was not a longitudinal study, but nonetheless this is a remarkable result: the difference in NAA levels was the same in patients 20 years post-diagnosis and those just several months post-diagnosis. This is not what you would predict if NAA scales linearly with nerve death, and if nerve death occurs at a higher rate than in controls throughout the disease. Brain atrophy occurs at fairly constant rate from CIS through SPMS, around 0.6-0.8% per year. It occurs at a lower rate in controls, <0.4% per year for non-elderly individuals, reflecting lower rates of nerve death in controls. As a result, some simple math shows that the gap in brain volume between controls and MS patients increases with disease duration. If NAA levels reflect total nerve death, then just as in the case of brain atrophy, the gap in NAA levels between MS patients and controls should increase over time. But this is not what the study finds.

  • (continued)

    Another anomalous data point is the rescue of NAA levels under different MS treatments. See the following two studies, though there are others as well:
    They both find that NAA levels increase following treatment with natalizumab. If absolute NAA levels measure total nerve death across the brain, then I can understand why the rate of decrease of NAA would be slowed by natalizumab (if it decreases inflammatory damage). But I can't understand why natalizumab treatment would *increase* NAA levels. That would seem to suggest a reversal of nerve death, i.e. nerve growth. Is that really what's happening?

    So I'm puzzled. Do NAA levels measure nerve death? If they do, is the scaling relationship linear? Filippi et al are assuming that the answer to both questions is "yes." Is this true? These questions may have already been settled in the literature, and I don't mean to suggest that I've given knock-down arguments above — this is way out of my area of expertise. It would be great to get some more information about the use of NAA as a disease readout. MD, do you have any thoughts?

    • Although NAA is touted as a marker of neurons because they contain alot of it, it has been know for many many years that NAA levels can increase and we are not ressurrecting dead nerves. So it is not clearly a real biomarker of nerve loss, we showed this in the mouse in about 1992, yet MRIers may try to convince you that is is nerve loss and again the central problem that many MRI outcomes have no real pathological correlates.

    • Looking at postmortem MS brains, there is more damage to the cortex than in the white matter. Since the damage in the cortex cannot be imaged with current MRI technology it seems clear that something else other than focal inflammation is occurring in MS:


      "The percentage of demyelinated area was significantly higher in the cerebral cortex (mean 26.5%, median 14.1%) than in white matter (mean 6.5%, median 0%) (p = 0.001)."

    • Anon 10:42: Grey matter, including the cortex, is difficult to image. So what? We've known for a long time that MRI can also miss lots of focal white matter lesions. Does that mean that the white matter lesions that we miss are non-inflammatory? No, it just means that our imaging techniques have limited resolution, and we need better ones.

      Anyway, it's not even true that MRI can't pick up grey matter lesions. If you use higher-resolution MRI, you can see grey matter damage:

      There are lots of other papers which find grey matter pathology using MRI.

    • Tuesday, October 27, 2015 1:05:00 p.m.,

      Yes you miss the whole point. White matter lesions are driven by lymphocyte infiltration while grey matter is not. So if you believe that all brain atrophy is tied to the local area of white matter inflammation, put all of your hopes on Tysabri. But this is not going to halt progression as has been demonstrated which is the point of this whole post. Don't lose focus.

    • "White matter lesions are driven by lymphocyte infiltration while grey matter is not."

      Where is the evidence for this? As far as I know, the claim that grey matter lesions are non-inflammatory comes from post-mortem studies which didn't find lymphocyte accumulations in these lesions. But this type of pathological evidence doesn't necessarily provide information about how the lesion was formed. If you perform a biopsy on early cortical lesions, you do find evidence of inflammation:

      Is there other evidence that cortical lesions are non-inflammatory?

      (By the way, there is evidence that Tysabri decreases cortical lesion formation: http://www.ncbi.nlm.nih.gov/pubmed/22570359)

    • "Is there other evidence that cortical lesions are non-inflammatory?"

      No where was the claim made that grey matter damage was not caused by inflammation. The claim was that lymphocytes or T cells are not the main driver, microglia is.

      So again if you want to continue to believe T cells, T cells, T cells after failed trial after fed trial there is not much can be done to change your mind. But it is pretty clear this line of thought has been a failure. But you may make a good Pharma CEO. You should look into it.

By MouseDoctor



Recent Posts

Recent Comments