HSCT in MS – what have we learnt?

More and more HSCT procedures (aka bone marrow transplants) are being performed for MS around the world than ever before. Currently, we lack convincing data on the conditioning regimen that best balances efficacy and safety. Patient selection is key and HSCT should be considered as an early option in aggressive cases of inflammatory MS, rather than as salvage therapy in later disease. 

Mult Scler. 2017 Nov 1:1352458517742532. doi: 10.1177/1352458517742532. [Epub ahead of print]

Intense immunosuppression followed by autologous haematopoietic stem cell transplantation as a therapeutic strategy in aggressive forms of multiple sclerosis.

Mancardi G, Sormani MP, Muraro PA, Boffa G, Saccardi R.


In the majority of relapsing multiple sclerosis patients, the disease can be quite easily controlled by already available, approved therapies. There are, however, some aggressive cases who continue to have clinical and magnetic resonance imaging (MRI) activity in spite of the treatment. These are the cases who may now receive benefit from intense immunosuppression followed by autologous haematopoietic stem cell transplantation (aHSCT). In this review, we describe the method and the rationale of aHSCT, the more recently published studies that demonstrate its efficacy in selected multiple sclerosis cases, the problems related to safety and the transplant-related mortality risk of the procedure. A description of the ideal patient who can take advantage of aHSCT is outlined and, finally, the ongoing studies which are near to completion or are close to starting are briefly reported.

Some might say that HSCT (haematopoietic stem cell transplantation) is the final frontier in MS? Like the voyages of the Starship Enterprise, the saying “to boldly go where no man has gone before” has been debated for the best part of a century over its correctness and conformity to current standards, likewise HSCT practices the world over have been heavily scrutinised on an off for the better half of two decades on wording alone. You might say that a comparison between the Starship Enterprise and HSCT is nonsensical – maybe; but the correct selection of words can make a world of difference.

I allude to these considerations below…

‘High-dose’ vs. ‘Low-dose’ conditioning regimens

The authors have provided a link to the European Blood and Marrow Transplantation society (EBMT, www.ebmt.org) where you can find detailed descriptions of the different conditioning regimens.

High-dose i.e. myeloablative regimens are expected to ablate marrow haematopoiesis (the process of formation of blood cellular components), thereby preventing haematological recovery. In contrast, low-dose i.e. nonmyeloablative regimens, although causing minimal depletion in immune cells, do not require stem cell support.

The best conditioning regimen is debatable, but not surprisingly the high-dose regimen has the most likelihood of eliminating autoreactive T cell clones from your marrow, but safety is a big concern. Low-dose regimens, on the other hand, are almost immunoablative and maybe more appropriate in the autoimmune setting rather than in the cancer setting. With low-dose regimens, 25% experience relapses after the procedure (Burt RK, Lancet Neurol 2009; 8:244-253) and MRI activity is not completely eradicated (Curro’D, Mult Scler 2015; 21: 1423-1430).

RRMS vs. progressive disease i.e. if the disease is still in the inflammatory stage.

Relapsing-remitting, as opposed to progressive forms of MS greatly influence progression free survival outcomes with HSCT. HSCT has been particularly successful in severe cases of relapsing MS where it is able to completely eradicate Gd-enhancement on MRI (cutting out relapses), often leading to dramatic clinical improvement (Mancardi GL, Mult Scler 1005; 11: 367-371) – with as much as 1 EDSS (MS disability rating scale) point; a feet rarely observed in progressive MS. Leading us to conclude that HSCT efficacy is closely linked to its capacity to suppress inflammatory activity.

The authors point out that in their cohort, NEDA (no evidence of disease activity) status at 5 years was observed in 72% RRMS vs. 55% SPMS cases. This is not to say that targeting inflammation in progressive MS is a waste of time, experience with targeted B cell therapies have clearly highlighted that this is feasible in practice, and should also be considered for HSCT. A note of caution, however, registry data collected by EBMT clearly point out that a progressive disease course and a higher EDSS score at entry were associated with increased transplant related mortality.

In summary, more and more transplants are being performed for MS around the world than ever before. Currently, we lack convincing data on the conditioning regimen that best balances efficacy and safety. Patient selection is key (see Table below) and HSCT should be considered as an early option in aggressive cases of inflammatory MS, rather than as salvage therapy in later disease.

About the author

Neuro Doc Gnanapavan


  • "HSCT should be considered as an early option…"

    "progressive disease course and a higher EDSS score at entry were associated with increased transplant related mortality"

    … do you not see how contradictory that is? HSCT is more effective and safer in ALL patients who have shorter time with MS! Irrespective of phase of MS.

    What you're effectively saying is "what for RRMS to progress to SP.

  • Messed up my previous comment. I didn't quote enough in the first extract. Point I was trying to make is, by not treating RRMS, you're not missing the most 'effective' window for treatment!

    • Selection of most appropriate inflammatory cases is key, so that you're not inadvertently causing harm. This applies to all categories of MS.

    • but, relative to non myelo hsct, ocrelizumab is harm free when taken 6 monthly for an indeterminate amount of time?

      and when it's time to get off ocrelizumab, you replace it how, with what and based on what evidence.

      Which inadvertent harm are you exactly referring to?

      (I'll just be going too far if I mention the early cases of Tysabri associated PML, then Gilenya associated PML, then rebound effect).

      Which inadvertent harm are you exactly referring to?

  • Alemtuzumab is HSCT. Without poisoning every cell in the body, as chemotherapy. There's a reason why NHS has not approved even though it's cheaper than any DMT on the market. The technology is not mature. You don't need chemo to kill all the immune. You can do it using mab.

    • This discussion has already been done at the post:
      #ResearchSpeak: does HSCT have a chance?

      Spoiler alert: Lemtrada didn't win

    • HSCT kills immune cells. Alemtuzumab kills immune cells. The efficacy is due to which one kills more effectively the immune cells. HSCT has higher efficacy because it kills all immune cells. As well as poisoning every cell in the body at the same time. There are technologies on the horizon which will make all chemo safe.

    • Technologies on the horizon. Ha. Yeah, funny. Hsct isn't that bad at all. The chemo is low dose compared to cancer protocol.

      The earliest pwMS have HSCT… The better. Period! The earlier someone is treated, the better the outcome, the safer it is. No exceptions.

      Or you could just wait and get worse on DMDs

    • The aim of HSCT in autoimmune disorders is not induce marrow failure and therefore the future here maybe one of reduced intensity conditioning regimens, which may allow an autologous recovery in immune cells to occur.

  • What's missing from this HSCT overview is any mention, specifically, of PPMS patients. Based on the patient selection criteria, most PPMS patients would not qualify as the vast majority do not have relapses or enhancing lesions. However, many of the for-profit HSCT treatment centers that have sprung up readily accept PPMS patients willing to pay for the treatment. Any commentary on this? Especially in light of the Ocrevus approval for PPMS (a blanket approval in the US, a narrower approval in the EU)? Granted, the benefit seen by progressive patients without visible signs of inflammation seems to be underwhelming with Ocrevus, would "noninflammatory" PPMS patients fare any better with HSCT?

    In regards to HSCT versus Lemtrada, we know that Lemtrada has a fairly high risk profile in regards to infusion reactions and serious potential side effects (the development of other autoimmune diseases, some of them potentially deadly). Does HSCT carry the same risks, and if so, is it equal to, less than, or greater than those seen with Lemtrada?

    In regards to the SPMS numbers (55% NEDA after four years), though not as impressive as the numbers for RRMS, the 55% mark is pretty damn impressive given the inability of most of the current treatment modalities to put a dent in this form of the disease. What I'd love to see is a breakdown of SPMS patients treated with HSCT based on whether or not they have inflammatory markers, still have relapses, etc. SPMS is an especially multifaceted beast, so simply putting a number on the whole group is rather pointless, IMHO. Despite having read everything I can find on HSCT for progressive patients, including all the retrospective studies, this breakdown never seems to be brought to light. Valuable info, I would think.

    Also, any thoughts on the possibility that HSCT may work by ridding the body of EBV? Since the treatment effectively ablates all immune cells, it also effectively takes out EBV residing within those cells.

    The natural history of patients post HSCT would seem to fit nicely with an anti-EBV scenario, as most patients would get reinfected at some point after treatment, and after several years this reinfection might reignite the epigenetics that drive the so-called autoimmune process seen in MS. Indeed, long-term retrospective studies of HSCT treated patients show that the majority of them do start displaying MS disease activity between years 10-14, which is what one might expect in a genetically susceptible patient infected once again with the virus.

    I know I've presented a lot here, responses to any or all points would be greatly appreciated.

    • I note that MD below has answered so there may be duplication.

      You're correct, most of the data is on RRMS population. This has come about from earlier transplant data and the EDSS criteria and MRI criteria are becoming more narrower…but this is based on outcomes and an attempt to standardise the technique. If you look at the ASTIMS study this is a randomised controlled study in RRMS and SPMS, with EDSS between 3.5 – 6.5, presence of one or more Gd-enhancing lesion on MRI. By this stance, majority of PPMS may not qualify. There are efforts to include increase T2 lesions, and will include more PPMS cases but the data on this is sparse. The point to make about the HSCT studies are that these are investigator led as opposed to pharma led; there is no goal for licensing in a particular disease sub-type, clinicians are simply trying to figure out where HSCT has a place in MS. At some stage it would be licensing ready (once the induction regimens, side effects are sorted through) and it would be then that clear Phase III studies would start to come through.

      There is no debate that HSCT has more risk than Lemtrada. Treatment induced mortality is the worst possible outcome and this stood at 1-3% in the earlier study, but with partial induction regimens this is coming down to <0.5%.

      With regard to the NEDA outcomes in the SPMS group, I agree this is a good outcome but they were active cases with evidence of MRI progression and recent disability progression in the recent studies which also include NEDA. The EDSS cut off in most of these groups is 6.5. Again more transparency is needed and as the Sheffield group publishes their trial we may be able to push for this.

      I think the recurrence rate later on is more a reflection of MS development rather than EBV per se. HSCT including ex-vivo T cell depletion doesn't seem to entirely get rid of it. Both T cells and B cells take a huge hit from BEAM induction sometimes resulting in marrow failure and high mortality. So in theory with partial induction regimens there will always be some autoreactive cells left behind, rather than re-exposure. Although it would be interesting to study EBNA-1 titres in this group.

    • Thanks much for your answers, NDG.

      In regards to lack of phase 3 trials, I find the absence of such trials this late in the game to be extremely disheartening. Yes, of course it is because these trials are investigator led, not Pharma led, but this is the crux of the inherent conflicts of interest rife in a research model that puts pharmaceutical companies at the head of the table.

      I know that there is currently no way around this conundrum, but with a treatment protocol as promising as HSCT (which also holds the promise of gutting the profitability of the current crop of MS DMDs) it's hard for patients to have a truly sunny outlook or hope for any advancement towards a treatment paradigm shift or – gasp – a cure.

      Yes, the current generation of disease modifying drugs certainly has improved the MS treatment landscape, and the NexGen drugs will likewise represent a step forward in this regard, but none of them do a whit towards curing the disease. It seems research with a cure as the primary target is severely underfunded and undermanned. I'm sure that many investigators have their hearts in the right place, but the needs of their wallets often dictate they focus on treating rather than curing, as this is where all of the Pharma monies are directed. As a patient, this situation makes me want to vomit.

    • Exactly what I said a while back. Days of cures are numbered. Name one major condition cured where pharma was raking high amounts of money? Remember peptic ulcers and Barry Marshal? Days of maverick scientist are also numbered. Truly sickening. But on the bright side Maverick funders like Bill gates with pharma like money may once again tip the balance in favour of patient. As well meaning Barts team are. The openness of this blog sight is example. They are still funded by pharma research.

    • Are the risks the same both hsct and alemtuzumab cause secondary autoimmunity but appears higher in alem. I suspect this relates to bone marrow purging is more in hsct so fewer autoreactive b cells

    • Studies show early relapsing ms responds better than late so trials focus on this population.

      EBV and HSCt. HSCt are agreat group on people to investigate this. EBV drives money B

    • Are the risks the same both hsct and alemtuzumab cause secondary autoimmunity but appears higher in alem. I suspect this relates to bone marrow purging is more in hsct so fewer autoreactive b cells

    • Here's the URL of one small long-term study of HSCT outcomes. There are others that report similar results.


      Of course, with the more refined techniques being used today long-term outcomes may be improved, but not enough time has passed the proper assessment. However, given the decreased kind of NEDA at three, five, and seven years in some recent studies, it might be reasonable to assume that longer-term outcomes would reflect similar diminishing results…

    • Good question. If majority of MS patients are exposed to EBV say between 10-16 years of age yet do not become diagnosed until 14 years have lapsed then why? Does it take a latent EBV infection this long to become established in the CNS via memory B cells and start the neuroinflammatory cascade?

    • WK read this, it is enlightening about the 5 year outcomes and about the specific study you have posted.


      This study was the first ever on MS and back then they didn't separate MS the way the do now (no progressive) and they had chosen in the majority progressive patients with the worst possible situation, as it was a myeloablative and dangerous procedure. The outcomes though are not far from the results we have today for the progressive patients (about 50% success). The procedure is safer today but it is the selection of the patients that makes the difference in the outcomes. NEDA at five years is a NEDA that is unlikely to change after the five years (most patients who fail, fail at 4). There is another long term study with children with MS treated with HSCT, with excellent outcomes and reports at 10 years, but I cannot find it right now.

    • Anonymous-thanks for the link. Very interesting, but not so much a research paper as a dissertation. As such, it uses various research papers to support its argument, but is still dealing in hypotheticals.

      If you can get your hands on a full paper from which the following abstract is derived, you'll see that progression free survival continues to fall after five years in all populations of HSCT treated MS patients, regardless of subtype. The number of patients tracked post seven years falls off dramatically, but this is one of the more robust long-term retrospective studies, and was recently published (2017).

      For the full cohort studied, progression free survival was approximately 46% at five years, but dropped to approximately 35% at seven years. Similar reductions in progression free survival rates were seen by subgroup breakdown. Here's the link for the abstract, which does not include numbers post five years. This data is included, though, in the full paper (figure 1, A, B, C, D):


      After seven years it appears the number continues to drop, but the number of patients observed at this point are too few to be statistically relevant.

      Certainly, proper patient selection is key; however, it seems wildly premature to pronounce HSCT as a cure for MS. It does, however, present a wonderful opportunity to see just what factors trigger a reemergence in disease in those treated with the protocol, as well as which factors may be absent in those who do see extremely long term remission/eradication of disease activity.

  • While possibly being the most efficient therapy, NEDA 5 year data have consistently shown that HSCT is not the "cure" for everyone with RRMS. Approx 20-40% fails NEDA at 5 years (depending on the different cohorts). While aHSCT possibly removes the autoreactive cells, the patient still has the same genetic risk factors, and possibly lives in a similar environment as when they acquired MS. Twin studies have shown a risk of approx 20-25% for monozygous twins. So the question is: Do the subjects that fail NEDA at 5 years have a relapse of the disease, or have they re-acquired it based on the risks they inherently carry?

    • Very good question

      I have ask myself the same if you take any ms drug (including

      hsct) and you are PFS for 1 ,2 ,3,4 ….n ..years and then you

      relapse or progress what kind of mechanism restart the disease


      Is it the same mechanism like when you had the first episode (cis)?
      Is your ms the same as before the treatment?

      Why did the treatment work for 1,2,3,4….years and them stop working?


    • The suggestion is that much of the progression in the absence of new relapses can be due due ongoing neurodegenerative processes caused by a number of factors; dystrophic environment, remaining, possibly poorly myelinated axons, having to work harder, latent pool of activated microglia releasing toxic factors etc etc.

    • An interesting (real life) case study:
      A woman diagnosed with MS 30 years ago and had chemotherapy for breast cancer 27 years ago (she had two courses of chemo). Her MS had been halted for 27 years. She now relapsed. This shows two things:
      1. That MS is either not cured, or can reemerge (at least with that chemo regimen).
      2. (And most importantly) MS was really halted at its course. After 30 years she RELAPSED, so there was no hidden activity that would bring her to SPMS today (she is around 60 y.o.). She is not disabled.

  • Given treatments like HSCT and Alemtuzumab bring down the white blood counts rapidly. Does the efficacy of these treatments affected by those who are rapid populaters? Instance a count 0.2 (10**6/litre ) at day 30. Then goes up to 1.0 at day 60?

  • Does Alemtuzumab destroy T cell clones in bone marrow? Or would HSCT be the only option? Is non myleo not effective enough to do this? I would be curious as to whether Alemtuzumab or HSCT would accomplish this (or would either)?

    • I suspect that it is not that active in bone marrow…
      (a) Alemtuzumab in humanCD52 transgenic mice does not deplete well from bone marrow. This is in part because it depletes by antibody dependent cellular cytotoxicity so antibody and natural killer cells have to enter the bone marrow.

      This maybe be why immature B cells come flooding into the blood after Lemtrada…creating the risk of secondary autoimmunity

      fingolimod may stop alemtuzumab working because it traps cells in bone marrow and these aren't purged.

      Cladribine will get in bone marrow

      the drugs in HSCT may do this but depends on what you use

  • This is the most common story with patients who go through HSCT and their doctors. Neurologists first discourage and even discharge their patients and then they are fascinated (taken from the popular social network).

    "Just had my first appointment with my neurologist at 10 months post HSCT – Think I blew him away with how much better I am. He was very interested to know the costs and we worked out that it is equivalent to 2 years on the drugs – so I've saved the NHS a small fortune! Considering he thought a year ago that it wouldn't work for me – just shown him how wrong he was! �� My 'timed distance walk' (only about 10m!) was 1 whole second faster and he equated that to 2013 (probably the first time I met him after we moved to Berkshire!). I genuinely think he is interested/fascinated in this treatment but is sceptical that it will last long. He wanted my thoughts on the Channel 4 doc about the Stem cell treatment so I put him right on the 'no chemo = no cure' philosophy and stated you needed to reboot the immune system, so chemo essential! Showed him all the Mexico booklets we were given and how wonderful the Mexican health system/Dr Ruiz is. NHS could certainly learn from them. I think we are starting to educate the neurology profession!"

    More than 2.000 patients from all over the world have had the procedure and with the current flow the number will double the next two years. I read Biogen bought monomethyl fumarate because Tec. will be out of patent in 2 years. Yes, it will be hard to stop HSCT.

    • That'll be Caroline Wyatt. For a recent update on her MS after the treatment (maybe not quite as "miraculous" as described above earlier this year) and a conversation with Mark Lewis see here. I hope at least her increasing disability is slowed down. The migraines are rather worrying though. I'm sure she's in good hands in the UK.

    • No, this is a yesterday's post from FB from a patient of the many that have had HSCT (don't have a license to use her name). One could create a whole collection of this kind of reports, this is how often they are. Caroline Wyatt helped as a journalist to make HSCT more popular, but she as a patient is just one example, and it is too soon to tell.
      I could also bring numerous reports of how many patients are feeling their legs weaker than before, many months after Ocrevus infusion. Would you reconsider Ocrevus? (this is true by the way, dont know why is happening)

    • Do you have a link to the relevant FB page so I can have a look?
      Obviously there can be a lot of misinformation not only on Facebook but the web in general (CCSVI is a good example) so I always say exercise due caution/scepticism when viewing such material.

    • Also, there were never scientifically proven researches about CSSVI so it is time that these two are not correlated any more. The problem with HSCT is that it can be dangerous, not that it is not effective. And the deaths are much lower that the nowdays research average, at the two major clinics (less than 0.5%). If this patient flow continues, -which it will- I will not be surprised if pharma actively sabotage it. It is not in their plans to search for somehting more permament, so anything like that will not be welcomed.

  • There are groups for every clinic that offers HSCT (patient groups, not official clinic groups). Russia, Mexico, India, Chicago are some of them. These are not pages, so one needs to be accepted there. There is also a non responders HSCT group with about 50 members, so it is not a taboo thing or unreported. There are also at least 3 Ocrevus groups where one can see how patients react to the drug. These are patient groups so one needs to have an ethical code to access them if not a patient.

  • A woman of 15 years of RRMS and 2 years of SPMS had HSCT in 2012. Went to Moscow with an EDSS of 5-6. After 18 months of recovery, she settled at a 2 EDSS, with days of no symptoms being not unusual. HSCT in late 2012; felt finally over HSCT and settled with healing in late spring 2014; now, late 2017, and still feeling great.

    ProfG would have discourage her to proceed, as some other patients testify. What treatment would offer this quality of life to her?



Recent Posts

Recent Comments