Sunday, 3 September 2017

#ResearchSpeak: how does ocrelizumab do in the NEDA stakes?

The need to rebaseline when using NEDA: ocrelizumab as an example. #ECTRIMS2016 #ResearchSpeak 

Summary: This post looks at NEDA (no evident disease activity) as an outcome measure in clinical trials and in particular the issue of needing a new MRI scan, at a certain time point to set a new baseline, for comparison in the future.  The latter is what is referred to as rebaseling. 

In recent comments, the HSCT zealots have raised points about the use of NEDA as a treatment target and outcome measure to compare therapeutic strategies.


In the poster below there is little doubt that ocrelizumab is a highly effective DMT and shows why it is so important to rebaseline after an appropriate period of time before starting to monitor a treatment. For most maintenance DMTs 6 months is appropriate in that is pragmatic, logistically feasible and supported by data from clinical trials. 

The analysis below shows that if you do this with ocrelizumab, then over 80% of subjects are NEDA from 6-24 months (figure 3). To the best of my knowledge this is the best NEDA data we have. I have yet to see HSCT rebaseling NEDA rates, but I suspect they will be in the same ballpark or higher. 

The question I have asked my self is when is it appropriate to rebaseline after a IRT (immune reconstitution therapy)? I have argued in the past there is no point in doing it before you complete your course of treatment and are in a position to consider retreatment. For alemtuzumab and cladribine, this would be at 24 months. As an example, let's say we have a patient who has been treated with alemtuzumab and you do an MRI at month 12 just before her second course of treatment and find they have a Gd-enhancing lesion. Are you going to say the alemtuzumab has failed? No, you are not, as this person has not had the full course of treatment. What you should do is give this person the second course of alemtuzumab and then wait until 24 months to do the baseline MRI and start the clock ticking again. Therefore NEDA rates in the first 12 and 24 months on alemtuzumab and cladribine and other IRTs are not that helpful. What about with HSCT? The issue with HSCT is it a one-off treatment or do you think it should be used more than once? If the former then it makes sense to rebaseline at 6 months. In comparison, if you want to offer it again then how soon can you do a second cycle of HSCT? If 12 months or 24 months you may want to use these time points for rebaselining. 

It is worth highlighting in the data below is the remarkable observation that 57% of the IFN-beta (Rebif) treated participants were also NEDA in the 6-24 month epoch. Based on earlier interferon NEDA data, this is much better than one would expect. It looks as if from contemporary trials that the efficacy of Rebif has improved. Why? One reason could be that the population of trial participants have included subjects with more benign MS or just maybe Rebif's efficacy has improved. I suspect both are correct. Most subjects in the Opera studies received RNF (Rebif new formulation). We know that RNF is associated with fewer NABs (neutralizing antibodies), which affect its efficacy; hence there are reasons to expect that Rebif's efficacy has improved. I am sure a lot of people will jump on the Rebif NEDA data to support its continued use as an effective DMT in the 'majority' of MSers. If you are risk averse and subscribe to the slow stepwise or even rapid escalation, treatment strategies for MS, why not? 

I am interested to see how the regulators, payers and/or healthcare commissioners respond to this data. I sincerely hope the EMA give ocrelizumab a liberal label that will allow first-line use in pwMS with active disease. This will then allow HCPs and their patients to decide on which DMTs they want to use. However, as always in price-sensitive markets the payers and commissioner may have a different take on things particularly if ocrelizumab is priced at a premium.




CoI: multiple

25 comments:

  1. Rebaseling aside, why were the NEDA rates in the Alemtuzumab extension study reported "per year" in years 3, 4 and 5? Why not clearly report the cumulative rate each year? Or would this highlight the decline of efficacy too much?

    I'm not a HSCT 'zealot', quite the opposite. I have finished my second course of Alemtuzumab earlier this year. I find it incredibly frustrating and misleading that you continually fail to discuss the presentation of results in the Alemtuzumab extension study. Rebaseling was performed at month 24, yet NEDA rates are then reported "per year" in years 3, 4 and 5. This should be cumulative, obviously the rate would decline by year 5. What we want to see is that rate of decline.

    ReplyDelete
    Replies
    1. Because with an IRT every time you retreat you have to reset the baseline. The cumulative NEDA rates don't mean the same as they do with a maintenance therapy. When you have activity on a maintenance therapy it means a sub-optimal response, or a treatment failure, in comparison on an IRT it means time to re-treat.

      The cumulative NEDA rates may be helpful to compare relative efficacies within the class of treatments. You can use them to compare outside the class for this reason. So a maintenance with a maintenance therapy, a pulsed IRT with a pulsed IRT and one-off treatment with a one-off treatment. HSCT could fall into both of the latter categories depending on how you want to use it.

      A much better metric than NEDA-3 is long-term NEDA-4, i.e. what impact is the treatment having on end-organ damage. The best marker we have this at the moment is brain volume loss.

      Delete
    2. But the entire cohort of patients was NOT retreated in years 3, 4 and 5. So why rebaseline every single year?

      Of course the cumulative NEDA is what is relevant through years 3-5 here. Am I correct in my reading of 39.5% being the cumulative NEDA value at year 5?

      Delete
  2. Barts ms blog. Used to be a fan. But over the years turned in to a DMD zealot site. As far as interpretation of good, bad research... I think you need to change your strap line for this site. Appalling bias.

    ReplyDelete
    Replies
    1. I think this is a case of kettle and black, reading through your comments it is so clear that you have a bias, and in you clamour to get a response it is more and more likely that it won't come. I am not going to engage in a dialogue and no doubt ProfG will not bother either.

      Delete
  3. ocrelizumab zealot.
    ocrelizumab zealot.
    ocrelizumab zealot.

    ReplyDelete
    Replies
    1. I am sorry you feel this way but, you comments seem to demonstrate that you are guilty of the bias, you claim to campaign against.

      You are unlikely to get any response as one suspects it will be futile. You have also identified yourself as a troll as there is nothing constructive here as such do not be surprised if your comments get no response.

      Delete
  4. A quick lesson in maths and honesty; In the CARE-MS2 cohort, Alemtuzumab drove a true NEDA rate of somewhere between 8.6% and 27% @ 5-years.

    Before we jump in, let me start by saying it seems a little bigoted of you to label anyone who happens to challenge or disagree with you as a “zealot”.
    Definition of BIGOT

    1. : A person intolerant toward those who hold different opinions
    from oneself.

    Let’s not forget that you intentionally poked the bear on this one by needlessly headlining your Alemtuzumab post as an HSCT story (which the actual commentary and data within failed to live up to).

    Definition of CLICKBAIT
    1. : something (such as a headline) designed to make readers
    want to click on a hyperlink especially when the link leads
    to content of dubious value or interest

    Anyway, I digress.

    Looking at the Alemtuzumab data, the issue is that it has been intentionally presented in such a way as to mislead the audience and obfuscate the facts, whilst preventing interrogation of the true efficacy of the therapy. It is the very epitome of ‘lies, damn lies and statistics’. In my view, this is indefensible, given the stakes involved for patients lives.

    MD1 has himself voiced some concerns around the way this data is presented. Is he an HSCT zealot too? I hardly think so.

    Last thing before I jump into the data; I have no dispute with the concept of re-baselining - makes total sense. But I do advocate for the application of basic arithmetic, and have a preference for honest facts over misleading marketing BS and weasel-words.

    ReplyDelete
  5. Doctor Giovannoni

    You know (and i know)that NEDA is good for you, as measure
    to justify to pharma companies that the teraphy is working also to the health care authorities

    Me as a patient what i want is PFS ...Why?

    Because progression is all that matters for a patients quality of

    life

    Neda can stand 7 Tesla mri machine or better yet histopathologic studies
    There are plenty of studies here is the blog regarding 7 tesla machines
    Even in the most ablative hsct and allogeneic bone marrow transplant

    You could have no visible disease but still people progress

    There will be always some pocket of inflamation

    Most of the Hsct studies report Pfs

    You should do the same (for the patients)

    Obrigado
    Luis

    ReplyDelete
  6. So, let’s get stuck in with the amazing paper that you co-authored (Coles et al, CARE-MS II 5-year follow-up).

    Full paper here: http://www.neurology.org/content/early/2017/08/23/WNL.0000000000004354.long

    1. I note that you’re happy to report the aggregate progression-free % over the full 5-year period in the abstract (regardless of re-baselining) – presumably because that number, at 75.1%, looks positive. There’s no arguments presented here about that not being relevant due to re-baselining.

    2. How peculiar, then, that you’re unwilling to share the equivalent aggregate statistic for NEDA, on the same basis? Could it be because the reported cumulative NEDA rate at 2 years in CARE-MS2 was just 32%? And *quote*: “Cumulatively in years 3–5… 27.0% showed neither clinical disease nor MRI lesion activity.”.

    3. I’ll repeat that last bit, in case you missed it. Just 27% were NEDA cumulatively across Years 3-5 (even after re-baselining at Year 2).

    4. If you consider that a proportion of the 32% who failed to achieve NEDA in years 1-2 likely sit within the relatively meagre 27% who were stable in years 3-5, the whole thing starts to look a little bit flakey! All we can definitively say is that the real number for cumulative NEDA across years 1-5 is somewhere between 8.6% (i.e. 27% of the original 32%) and 27%.

    5. Now, some patients had additional dosing beyond the initial 2. There are some 40% of patients who had a 3rd course between years 3 & 5 (20% in Yr3, 11% in Yr4, and 9% in Yr5). And another 14% also had a 4th or 5th course – which is kind of redundant in the UK as you can only get 3 courses paid for. But for none of those guys do you have 5 year data after re-baselining (max. 2).

    6. For that reason, the annualised NEDA stats you have reported are next to useless – as all they tell us is the % of patients who were free from disease for at least ONE YEAR. The title of the paper (and tone of your own commentary) infers that this reflects 5-year data, but in effect it is reported as a collection of distinct 1 year studies dressed up as 5-yr data in order to artificially inflate the stats. For the 54% who had a 3rd course or more (from year 3) you can only have a maximum of 2 years data after re-baselining anyway!

    7. The way it is presented (and, indeed your comparison of these numbers against HSCT stats) would lead any layperson to assume they had a 60% of NEDA at 5 years, when in fact the real number lies somewhere between 8.6% and 27%.

    8. There’s some further shenanigans being pulled as you have between 1.5 and 5.2% of patients dropping out YoY due to patient request and/or lack of efficacy, who are then entirely removed from the reported stats for the following year.

    9. I won’t dwell on the fact that there is no control arm here, and it is not blinded – so we can’t tell how much preferential patient selection is playing into this. I can live with that, but worth noting that you frequently discredit other papers on this basis. People in glass houses and all that… ☹

    So, there you go. That’s my problem with the bias/misleading nature of your original post. In short, you’re pulling the wool over our eyes and pulling our pants down a bit here! If challenging that makes me a “zealot” in your eyes, I guess that’s something I’ll have to live with! ��

    They say when giving feedback you should always start and end on a positive note (the proverbial "shit sandwich"). So let me sign off by saying that I love the blog and appreciate all of the work of you and your team on here, and for allowing us to challenge and have a voice here on the comments section. The BVL data is clearly great news too!

    I actually think Lemtrada is a great treatment, but the data patients really need to see is how many patients were NEDA and for how long after each course. If you've nothing to hide, why not share that and/or the raw data?

    Thanks!
    Matt

    P.S. And whilst you're at it, why don't you share the data around the build up of neutralising antibodies over time as well. :)

    ReplyDelete
  7. Also you are the Lead author of the study

    Your disclosures at the bottom are overwhelm (and dificult to read)

    You are conducting studies of drugs made by companies that pay you

    On the other hand the zealots ,unlike you have Ms ,unlike you dont
    receive compensations from

    Bayer-Schering Healthcare, Biogen-Idec, Genzyme, GW Pharma, Ironwood, Merck, Merck-Serono, Merz, Novartis, Teva and Sanofi-Aventis

    ReplyDelete
  8. Replies
    1. you were, I removed it but this thread is going nowhere

      Delete
  9. "Neda can stand 7 Tesla mri machine or better yet histopathologic studies"

    I mean "Neda can not stand 7 Tesla mri"

    ReplyDelete
  10. Are these people really suggesting HSCT should be offered before ocrelizumab or other DMTs?

    ReplyDelete
    Replies
    1. Not necessarily offered, but trial results for some DMTs need to stop being presented to make them appear far more efficacious than they actually are. Provide the information honestly so that patients and their neurologist can make treatment decisions with realistic expectations.

      Delete
  11. Sorry but I feel the need to let fly with the emotional response. When I first found this site last year, it represented a life-saver: clear and informative in a detailed manner, unlike any other MS site I'd accessed. I benefitted massively from feeling more secure in my knowledge of this disease I'd just been diagnosed with, and in the selection of Alemtuzumab as my preferred treatment.
    Almost two years on and approaching my second round of infusions, the initial post on this subject on Friday and now this one has thrown me well and truly into the real world: brilliant exchange of opinions, with some opposing views, and yet more valuable information. The Blog has been doing that for awhile and this tipped the balance. But then there was the human reaction as I started to well up. I started to recognise I was desperate for it to be more straightforward, more cut-and-dried and the realisation I can't have now have the previous peace-of-mind I'd enjoyed in having opted for Alemtuzumab. The logic of the success level it can attain, the conviction that the benefits outweigh the risks, etc - all the reasoning I provided for family and friends - disappearing!
    I've said many times before that I am a lay person with no medical or scientific background and I genuinely marvel at the expertise evidenced by the all contributors to the site, and importantly I've learned so much. Here's the but - But, whilst thrilled to be an informed PwMS I don't know what to do with being a bogged-down in the mire of Yes, No, 'Tis'nt, Could be, Should be. Will/won't be etc, etc, If the Bart's site is going to be fully accessible to ordinary folk like me, and maximise the facilitation of their feeling confident and empowered there perhaps needs to be a warning or disclaimer or something to that effect to make it clear this is a place where conflicting positions are shared, studies and research data are written off or debunked, and where the ground is repeatedly sifting so you're vulnerable to not knowing where it's safe to stand, or more accurately that is no ground you can be sure of - yeah I know, just like the disease itself!

    ReplyDelete
    Replies
    1. It's hard Fi. In my mind the 'yes, no,'tisn't, could be, should be....' debates confirm my decision to not take a dmt.
      As interesting as this blog is it is not a fountain of unbiased ms knowledge. We must be careful not to base decisions on only one source of information and be alert to conflicts of interest. Data can be skewed, so can the presentation of it. It is hard making major decisions without a trustworthy source of information to draw from.

      Delete
    2. Your reply has been beneficial Julie because it's reinforced for me that I do fundamentally remain convinced of my decision to have a DMT and enables me to be clearer - I did draw on other sources to inform my decision. My difficulty is still that the reality of all the huge number of factors involved, including those you've highlighted that makes me feel unsettled. I thought I'd peeled back the layers of the onion, but the posts in the last couple of days have made me realise what a very limited job I did of it! It seems clear to me that your research informed your decision not to have a DMT and I sincerely hope both of us find, in the longer term, that we remain certain and sure of our choices.

      Delete
    3. From a scientific perspective it is important to know how the DMTs are working. i.e. their mode of action. We have proposed that it is the B-cell in particular the memory B cell. There are a few treatments that have a profound effect on these cells. The question we want to answer is which of the DMTs that targets these cells is the most effective and safe. We are not saying one DMT is better than another. It is horses for courses and the individual and their clinical team need to make the decision together.

      Delete
    4. without a trustworthy source of information to draw from

      Go to the MS Societies/MS trust

      Delete
    5. From a scientific perspective it is important to know how the DMTs are working. i.e. their mode of action. We have proposed that it is the B-cell in particular the memory B cell.

      Do you have class one evidence for that statement?

      Obrigado

      Delete
  12. Fi,

    Hence my disappointment in the way Alemtuzumab trial results are presented. They give the reader the impression of cumulative NEDA, because even the uneducated reader would assume statistics in an abstract would not be presented per 'year'.

    It's very misleading and as much as this blog is a wonderful source of information, I fear some contributors hold back on releasing or at least making clear what the important information is.

    I too have received Lemtrada and am still optimistic even if primarily about the BVL results. Lemtrada is still a highly efficacious option, just not as efficacious in terms of NEDA as we are led to believe.

    ReplyDelete
    Replies
    1. Thanks LGOn it's comforting to know that you are still optimistic.
      In reaction to yours, ProfG and MD posts I wish to make very clear that I do not believe any of the Bart's team are in anyway being underhand or plying us with inaccurate information or being disingenuous. I guess I feel it's more a case of re-wording a famous expression 'facts, damn facts and facts?' I came into all this
      horribly ignorant and naive with no realisation of so many factors, such as the Pharma agenda, and the more complex and confusing I've realised it is, the more overwhelming it started to feel!
      I understand that Bart's intention is to inform and as ProfG says: it is horses for courses. I still think it is valid to suggest highlighting that this site allows for highly-charged debate. Also that some posts can have a controversial or provocative style and the Blog contains conflicting views. This may enable a better understanding for anyone looking for easy answers that they might struggle at times therefore, and enables them to anticipate the potential for feelings of exhaustion or deflation in terms of what there is to read. Bart's remains my go-to site and I can't praise it highly enough, so I was sorry to read it has been losing readers. The site should also highlight that it provides some of the most detailed and informative information on MS and DMTs and that it does address wider but equally valuable matters such as scientific research and diet etc. It's just maybe that some of the posts should come with an emotional health warning because they don't hold back/take any prisoners! 😝 I do always leave the site with hopes as well - everything from treatment for progressive MS, that head-to-head studies will go ahead, so as ProfG says, we can better understand how effective treatments really are, and more recently that understanding is obtained to improve retention and improving readership levels.

      Delete
  13. the (aus) heamo seems to think there is merit in retreating with b depleting agents (rituximab or ocrelizumab) after hsct as a way of preventing memory b cells from resurfacing with 'ms'...

    of course, no (ms) data to extrapulate from and i think haemo works with RA a bit too... but he seems to be in favour of doing it before full recovery from hsct (provided the patient is young and otherwise healthy enough etc etc)

    if i'm understanding the haemo's train of thought correctly. neuro willing, went to ask neuro to speak with haemo but haemo off for 6 months somewhere lol.

    ReplyDelete

Please note that all comments are moderated and any personal or marketing-related submissions will not be shown.