To switch or not to switch that is the question


Barts-MS rose-tinted-odometer: ★★★

The one good thing about thinking aloud is that your colleagues’ chip in and provide feedback. Case 2 from my ‘ethical quandary post‘ is generating an important debate about whether to support this patient’s decision to switch therapy or not.

As a reminder, this is the 40-year old woman who started natalizumab as a first-line therapy 11 years ago after presenting with two disabling relapses in a four-month period. She has done exceptionally well on natalizumab, i.e. she is NEDA-3 (no relapses, no MRI activity and no change in her EDSS). In fact, her original disabilities from the two relapses recovered. At present she is fully functional, working full-time and very active physically. For example, she plays competitive tennis in her local sports club and ran the London marathon 2 years ago. Her current EDSS is 1.0.

The problem is that her serial annual MRI studies demonstrate that she has progressive macroscopic (visible by the naked eye) brain volume loss. Being an intelligent woman and a self-taught MS expert she knows this is a poor prognostic sign and she wants to stop natalizumab and have HSCT or alemtuzumab. She is aware from reading The MS-Blog (formerly the Barts-MS blog) that alemtuzumab and HSCT have a greater impact than natalizumab on end-organ damage or brain volume loss. After HSCT and alemtuzumab treatment brain volume loss is on average in the normal range (please see BEYOND NEDA).

What I didn’t say to you is that this lady has already made the decision that she wants to be treated with AHSCT, either on the NHS (not possible at present), abroad at one of the reputable private BMT units or in the private sector within the UK. The problem we have is that we have no idea what will happen to her BVL once she makes the switch. I suspect she will have accelerated BVL in the first year post-AHSCT, which is well described and is likely to be due to the neurotoxicity of the chemotherapy. After year-1 the BVL may or may not normalise. We have no idea what happens to the MS brain after being subjected to smouldering MS pathology on natalizumab for a decade.

Fortunately, we do have data from interferon-beta to alemtuzumab switching and, yes, after 2 years of interferon-beta therapy switching to alemtuzumab does normalise BVL. What is clear from the 8-year alemtuzumab follow-up data (see below) is that the rate of brain volume loss is age-dependent. Being in the 35-45 year age group the BVL was 0.13% per annum ((1.51-0.71)/6) on alemtuzumab. When you compare this to the 0.06% per annum in study subjects 18-25 years of age ((1.24-0.87)/6) you realise how important age is in determining treatment effects. 

Figure from MSARDs.

Is this data sufficient to talk this young woman down from her decision to have AHSCT and to go with alemtuzumab? What do you think? If this patient is reading this blog post will it affect your decision?

Another thing this ‘thinking out loud’ exercise has taught me is that having annual BVL measurements on our patients with MS on DMTs could be very helpful. I also think we should ask around to see if we can get a case series of natalizumab to alemtuzumab switchers to see what happens to the trajectory of BVL before, on natalizumab, and after the switch to alemtuzumab. At least then we will have data to inform such difficult decisions.

Bass et al. Alemtuzumab outcomes by age: Post hoc analysis from the randomized CARE-MS studies over 8 years. Mult Scler Relat Disord. 2021 Apr;49:102717. 

Background: Alemtuzumab significantly improved clinical and MRI outcomes vs. subcutaneous interferon beta-1a (SC IFNB-1a) in the CARE-MS trials (NCT00530348, NCT00548405), with sustained efficacy in 2 consecutive extensions (NCT00930553, NCT02255656 [TOPAZ]).

Methods: Post hoc analysis of 8-year alemtuzumab efficacy and safety in pooled CARE-MS patients (N=811) stratified by baseline age (≥18 to ≤25, >25 to ≤35, >35 to ≤45, >45 to ≤55 years).

Results: Compared with SC IFNB-1a over 2 years across age cohorts, alemtuzumab lowered annualized relapse rates (ARR; 0.22-0.24 vs. 0.38-0.51), improved or stabilized disability (freedom from 6-month confirmed disability worsening [CDW]: 85%-92% vs. 62%-88%; achievement of 6-month confirmed disability improvement [CDI]: 20%-31% vs. 13%-25%), increased proportions free of MRI disease activity (70%-86% vs. 42%-63% per year), and slowed brain volume loss (BVL; -0.45% to -0.87% vs. -0.50% to -1.39%). Through Year 2, the treatment effect with alemtuzumab did not significantly differ among age groups for ARR (p-interaction=0.6325), 6-month CDW-free (p-interaction=0.4959), 6-month CDI (p-interaction=0.9268), MRI disease activity-free (p-interaction=0.6512), and BVL (p-interaction=0.4970). Alemtuzumab remained effective on outcomes through Year 8 across age groups. Age-related increases in malignancies (≤45 years: 0.9%-2.2% vs. >45 years: 8.1%) and deaths (0%-1.7% vs. 7.0%) were observed. Serious infections also increased from the youngest (5.1%) to oldest (12.8%) age cohorts.

Conclusions: Alemtuzumab had greater efficacy than SC IFNB-1a over 2 years across comparable age groups, with no significant differences between alemtuzumab-treated age groups. Efficacy on relapse, disability, and MRI outcomes continued through Year 8 across age groups. Age-related increases in serious infections, malignancies, and deaths were observed.

Conflicts of Interest

Preventive Neurology




General Disclaimer: Please note that the opinions expressed here are those of Professor Giovannoni and do not necessarily reflect the positions of the Barts and The London School of Medicine and Dentistry nor Barts Health NHS Trust.

About the author

Prof G

Professor of Neurology, Barts & The London. MS & Preventive Neurology thinker, blogger, runner, vegetable gardener, husband, father, cook and wine & food lover.


  • Well professor Gavin it seems the antiretroviral trial that you were talking about in Boston has finally started.

    Tenofovir Alafenamide for Treatment of Symptoms and Neuroprotection in Relapsing Remitting Multiple Sclerosis

    Harvard Medical School

    Gilead Sciences

    Information provided by (Responsible Party):
    Michael Levy, Harvard Medical School

    Link :

      • This is great news I really hope you can get another trial like this one up and running with Professor Julian Gold.

      • Why use ocrevus as the placebo group will it not make it harder to see an effect of tenofovir alafenamide ?

        • TAF is added on to CD20, so experiment arm = OCR/RTX + TAF; control arm = OCR/RTX. The design rocks as it eliminates variables from active inflamation etc.

          • Finally a sane trial. I would sign up.

            And some (not yet very strong) ammo to go bother neuros to use HAART as offlabel add-on

    • Anon May 12 8:48
      Re Gilead Sciences

      I always do a double take when I read this name. Maybe originally Biblical but in modern times which came first, Gilead Sciences or The Handmaid’s Tale?

      • Originally founded in 1987 as Oligogen, Oligogen soon changed its name to Gilead Sciences, after the reputed healing properties of the ancient Balm of Gilead (a heft dose of hubristic woo there).
        Margaret Atwood got there first with the Handmaid’s Tale in 1985.

    • “seems the antiretroviral trial that you were talking about in Boston has finally started.”

      Not yet..
      Recruitment Status : Not yet recruiting
      First Posted : May 10, 2021

      Why are they only doing rrms…?

  • What about ocrevus does it eventually normalise if not what rate of brain atrophy do we get after 2-4 years ?

      • If I were to take ocrevus with something like Lipoic Acid should I be concerned and think about using alemtuzumab ?

      • Howcome Cladribine BVL is higher at 0.55% (from your post linked) than Ocrevus as an ‘IRT’?

        • Ocrelizumab is probably also an IRT, but not used as such. The CLARITY population were ~8-9 years into their disease duration, which may explain why the BVL was so high, i.e. they had already missed out on being treated in that early 5-year window when DMTs make the most difference. In the ORACLE CIS (first-attack) trial, the BVL loss on cladribine was less than 0.2% p.a. The moral of the story is to get treated early and effectively and you may have a chance of normalising your future BVL rate. And is why I want cladribine available to treat patients with CIS.

          • Prof, it frustrates me every time i read something like this. There are just too many contradictions in the management of a patient with MS. For example….present with CIS diagnosis but cant qualify for treatment. Require second event. Long time passes before 2nd event = good prognostic factor. Dont get treated in 5 years from first even = missed out on period when DMTs make the most difference. BVL window closed. Presumably bad prognosis. So you think that you are doing ok because you have had a mild and gradual presentation but actually, this is just one long con. The administration of high efficacy drugs clearly needs to change but their effectiveness is inhibited anyway by the process of diagnosis being fundamentally flawed

  • Prof, there is a lot of talk about BVL on the blog but it is not something that gets measured or even discussed for the majority of patients. Is the rate of BVL in anyway linked to the amount of activity in this particular area. For example, should a patient with a very load lesion load in the brain but spinal activity be more concerned with spinal cord atrophy? sorry if these are naïve questions but i have not seen this discussed before. As a follow up query….this patient has been diagnosed for 11 years and there is a question over whether the opportunity to normalise BVL via alemtuzumab has been missed. By the same virtue…..a patient with CIS that does not present with a second event until 5 + years thus delaying their diagnosis would have the same issue. There must have at least been some of these in your time treating with alemtuzumab?

    • Exactly what I was thinking.
      I had this conversation regarding BVL with my neurologist last week. He said it was impossible to monitor BVL through MRIs. The concerns I raised he put down to natural ageing, I am 53 & he informed me that he experiences similar problems as he gets older, however he doesn’t have MS.
      How can we pursue this issue & how is BVL accurately measured?

      • There is software and third-party providers who do this. What is important to realise is that it is the trajectory over years that is more valuable than one-off measurements.

      • THANK YOU! Nowadays we hear “brain shrinkage” 24/7, yet it is also my understanding that while the software for measuring (estimating, actually) BV is pretty decent, the barriers to actually being able to implement monitoring of individual patients outside of clinical trials is not at all ready for prime time. Meaning you need the same software and the patient in the same machine in exactly the same position for each MRI over years, or the comparisons necessary to see a trajectory fail. (Obviously if my brain shrinks to the size of a lemon from one MRI to the next, the precise software and the specific tube do lose some of their importance)

        So while I’m sure BVL has been measured in various drug trials, I tend to get a bit impatient when neuros go on about this and get patients spun up about it, because there currently is no way to adequately measure the rate of loss in an individual, or at least none that I’m aware of. Being a patient, I’m naturally interested in MY rate of decline and whether the med I’m taking is working in ME, not just the broad statistical truisms that come from studies. It’s a long long way (at least here in the States) from “there’s software and third party providers who do this” to being able to practically utilize any of that.

        And Sandra, anytime your neuro tries to imply that he or she understands what you’re going through because of his age or personal circumstance, is worthy of a bit of the stink-eye. My neuro tried to say he had an idea of what MS fatigue was like because he had an infant and a toddler. I just looked at him. Indeed, I could have even cocked my head and raised an eyebrow at that. I was diagnosed at 52 (I’m 58 now) so I often get the “as we age” thing. But yours implying that your concerns are worthy of the brush-off because “everyone ages” or “that’s just aging”, or “Oh, I get that too”, is concerning to say the least. The point is not that some of the symptoms look like aging, it’s that MS means they happen so much earlier than they should.

  • Thanks Prof G.
    What range of yearly BVL would be considered acceptable around age 45?
    Do you agree that measuring the third ventricle over time is a good way to keep track of BVL?

  • Prof G,

    Many of these discussions (maintenance v escalation, switching treatments) remind me of the The Pushmi-Pullyu from Doctor Dolittle (a good name for many neuros), or for younger U.K. readers The Chuckle Brothers “To Me to You” sketches. The discussions generate a bit of hot air and then things return to normal. The elephant in the room is what you refer to as the “real MS”. As I understand it, the real MS causes damage to the end organ (brain / spinal cord) and relapses / focal inflammation are just the external immune system’s response when it detects the damage going on in the CNS.

    What are we attempting to treat with the various therapies (HSCT, Alemtuzumab…..Copaxone)? The real MS or just the external immune system’s response? If the latter, why aren’t we developing treatments to tackle the former?

    How many of your MSologist colleagues agree with your view about the real MS?

    Is MS really anything to do with Myelin?

    It seems to me that there are some very fundamental questions to be answered about what MS is and what drives the damage. The maintenance v escalation discussion and the issue of switching treatments seem to be ducking the real issue of halting the damaging processes caused by the real MS.

    • The real MS in my opinion is smouldering MS. It is clear that HSCT & Alemtuzumab do something to both; i.e. reduce relapses and focal MRI activity (focal inflammation), disease worsening (focal inflammation and smouldering MS) and BVL (end-organ, smouldering MS). There is also quite a lot of disability improvement that occurs after these two powerful IRTs. Why?

      So yes, not all DMTs are made equal when it comes to treating the immune response to what causes MS (focal inflammation) and the real MS (smouldering MS).

    • We have written a position paper on smouldering MS that we hope to submit in the near future. We will see what the response to the concept is after the paper comes out.

  • Hi Prof G, thank you for your post and having read some of the comments – I feel we should be able to have a bit more of an understanding, what is MS, what are we treating when it comes to therapies (I know the basics) & why and how we aren’t all able to have the option of the best and long term DMT at the outset/diagnosis. To me its a bit like sticking a plaster on the wound but it’ll still be there festering away until we actually treat the cause! A guideline of BVL for the lay person (possibly on the internet but should be available to visualise) to see their age, what is normal with ageing (on a healthy person) and then if you have lesions in the brain stem or on the brain already – or if they are elsewhere on your spinal cord – what is the likelihood that your BVL will be with the disease untreated and then treated (potentially of course, as we know it doesn’t always work!). Inflammation & autoimmune problems are the driving factors of MS – we can all remember the electrical shock feeling, when an attack (relapse isn’t the correct term for me) happened and understand the amount of stress a person is or was under for how did MS evolve into MS and not a CIS? If we are treated with a drug that controls the inflammation – would that not control the MS. I know we can’t continually take steroids (or can we, if they’re made differently) – but if there’s an antiviral medication that could help – please can we look into this much more – than highlight the HSCT, as it is such an aggressive treatment, with heavy risks. Also, as it is not available on the NHS – the funding of this therapy could financially burden someone who has health problems and working, whilst pursuing this treatment, is probably one of the most difficult things to do. My feeling is also, once you have entered this treatment – it is difficult then to have anything else to help you, if it is not as effective as you’d have liked. A person with physical disabilities is pretty rubbish for anyone – but it is okay and you can live a good and healthy life if you’re supported, managed and given the respect from HCPs that you so need to ensure you keep healthy. If your brain is addled – then that is a different story and something we can’t accept with this disease, so we therefore need to ensure that we treat the disease from attacking the brain – and therefore controlling the inflammation (smouldering MS).

  • How close do you believe we are to finding out what is conclusively causing the smouldering element of MS?

    • We have an idea of several pathologies that drive smouldering MS, some of which we are targeting in trials at the moment; eg. plasma cells in the SIZOMUS trial, microglia and intrathecal B-cells with BTK inhibitors, viruses with antivirals, EBV with cytotoxic T lymphocytes, etc.

      • Thank you, hopefully not long. If these trials don’t have an effect then it sets it back, and the go to treatments are the highest efficacy drugs.

  • My neurologist says BVL is hard to measure on an individual level because it varies to much day to day and at the time a scan is taken. According to him it’s only usable in large populations. Do you agree?

    • Unfortunately, under the current NHS England treatment algorithm/guidelines, she has inactive MS and therefore is not eligible for siponimod. We would have to stop her fingolimod and wait for her disease to reactivate so she can then be labelled as having active-SPMS to be able to offer her siponimod.

      Do you think I should put SPMS in her medical record? If I do this then I would have to stop her treatment in 6-12 months when the diagnosis of SPMS gets confirmed.

  • Congratulations for your work and dedication do pwMS.
    I am a 40-year old woman who started ocrelizumab as a first-line therapy when diagnosed, 3,5 years ago. I recovered a sensitive first relapse (right leg ). I am NEDA-3 (no relapses, no MRI activity and no change EDSS-1). At present I am fully functional and exercise regularly . Despite COVID risk, social distance and vaccine efficacy challenges (anti CD20 current dilema)… and all the natural concerns … one of my biggest fears is future cognition and BVL. So I completely relate to this post. Just wait and see…or do something else to prevent BVL?
    Thank you.

  • Prof G,

    Is this research of interest?

    Looks like this team are making similar observations to you:

    “Also, MAPKERK overactivity in microglia, as a factor in the pathophysiology of MS, could explain ongoing neurodegeneration in MS patients despite optimized immunosuppressive or immunomodulatory treatment.”

  • Are “the reputable BMT units abroad ” the Professor Federenko clinic in Russia and Clinica Ruiz in Mexico?

By Prof G



Recent Posts

Recent Comments