Computer generated transcript

Warning!
The following transcript was generated automatically from the content and has not been checked or corrected manually.

So I'm Stella. He was one of the consult neurologists in Belfast as well. And please, just to share this session or for speakers. Doctor M a talent her And as governor said, em has really been driving research in the UK forward, especially over the last few years. She's going to share her own experience of participating, um, carrying our clinical trials in the UK. Um, but she's she's really brought us all together as a group in the UK in terms of observational and clinical trial research. So Well, welcome Emma for her first talk in Belfast and hopefully not the last. Thanks very much. I forgot to ask about this. The point is this one. Oh, yeah. Okay. Um, so, um, thanks very much for the invite. I have a few disclosures. Um, uh, which you're here, but essentially, I wanted to talk today about contemporary clinical trials. So Gavin set the scene really nicely about particularly that slide with all the different products that we're now using in the M s clinic. And I'm sure many of you are really familiar with what is the typical clinical trial pipeline that takes molecules from the laboratory through phases of clinical trials and eventually eventually reaches regulatory approval. But I think really what I wanted to speak about today is where we need to look beyond that pipeline now into multiple sclerosis. And there are several reasons why we need to do that. And my research is very much in the patient facing trial, um, and sort of observational study side of things. And I think being involved with that it's really given me pause for thought about where the clinical trial pipeline is now. Not enough, Um, and it's around some of these issues. So I just wanted to spend the next 30 minutes or so just walking through some of these issues. And I really wanted to use some UK examples of clinical research, clinical trials, um, to kind of illustrate how we're addressing some of these issues. So, um, here is the typical clinical trial pipeline, as I mentioned, um, so, you know, developing, um, hypotheses and laboratory that identify compounds that may have promise taking them first of all into humans, um, and then into disease and so on. So why do we need to look beyond that? Well, first of all, um, I want to take you through a little bit about the complex disease biology. And I know there are lots of people in the room, Um, and who presented at this meeting who will have much better understanding than me of that complexity. And I think that highlights a few things that highlights not only that, in seeking targets for managing multiple sclerosis, there is that complexity. And it may be particularly in the later stages of disease, that more than one process, um, or pathway needs to be targeted. And I think the other thing that I'm going to illustrate is just the heterogeneity in terms of cohort. So if you take a cohort of people with very early M s versus people with quite advanced m s, where there can be a difficulty in, um uh measuring the outcomes in those groups because the biology and those groups is just slightly different at different times in their disease. I'm going to talk a little bit about trial design, um, and touch upon outcome measures, and I think class, we'll talk more about that this afternoon. Um, and I really wanted to come back to some of the things that Gavin mentioned about the treatment landscape. Where are we now? In the real world with M s and where are the unmet needs? And, um, how do we sort out the treatment algorithms? And finally, but really importantly, what are the patient and public priorities in researching, um, and clinical trials and m s? So, in terms of disease biology, Well, I think we're all really familiar with the fact that early in disease, you know, the hallmark of M s, of course, is a very lymphocyte mediated focal patches of inflammatory demyelination. And of course, we think of it as being white matter disease. But we know that the gray matter is involved as well. But I think you'll be really familiar with these sorts of schematics. And these two are really great reviews looking at, um, the sort of pathophysiology of progressive s and here the pathology of progressive M s. And I think what they highlight is that as the disease advances, what may be driven in the first place by that very sort of focal, um, lymphocyte, lymphocytes and microbial, um, focal sort of white matter demyelination can move into a phase where the disability trajectory also seems to suggest that maybe there is, um, a lack of exhaustion of those compensation mechanisms that leads to this much more neurodegenerative disease. Um, with these processes involving mitochondrial dysfunction that you heard about from John Mayer hot yesterday and a lot of oxidative damage and a neurodegenerative phenotype that's quite different, perhaps from early on, even though we know these processes do occur from the very outset an M. S. We know that as the disease changes and as it advances, those really start to dominate, um, and can pose a very different, different problem in terms of which processes we need to target with medications in clinical trials. But I think the other thing, as I mentioned, is that it poses a bit of a challenge when designing trials because of the heterogeneity. If you lump everybody with M s into one trial, So, for example, um, we have to power a trial according to what the outcome measure is going to be. So every trial will have a primary outcome measure. And if we think about our primary outcome measure, what we really want is something that's a very high risk of happening in the placebo arm. So, of course, in all of the trials that have salt regulatory approval for medications for relapsing M s, the outcome measure that's primary is almost always the relapse rate. Because that's the thing. That's the highest risk. And that's the thing that you commonly see in the placebo arm. But of course, if you were to do a trial of people with progressive Ms where relapses or less common, that may not be the best thing to power it for. And some of those trials would much better powered for, um, an advancement and disability. Likewise, the treatment response may be different in those two cohorts as an example. So we know that there's this window of opportunity for the immune modulating drugs if we're going to try and, um, uh, stabilize disability long term, Um, and as I mentioned the disease, biology that we're targeting might be slightly different at different stages of the disease. There's also really important, and I think often overlooked this different vulnerability to adverse effects. So there's some really interesting literature now, just looking at the risk of complications, particularly effective complications when you're modifying the immune system and people, according to their age and according to the disability, Um, and I think that's relevant to clinical trial planning. And then there's finally this interesting thing about utility, and I think this came across when we were at the frontiers meeting recently. In this session, that class was chairing and we were talking about you know what risk somebody would take in terms of the treatment to give them another year of good quality life. And actually, I think what becomes apparent is that different people at different stages of M s may also have differences in their opinion about that. So someone who's doing really well that we see in the clinic, who almost doesn't believe they've got m s yet because they just felt so well and they've had one optic neuritis, but otherwise life has not changed for them. The value that they placed on one year of stability in terms of their ability might be quite different from someone who's living day to day with a lot of challenges and would do anything to make sure that they didn't get worse your upon year. So that utility also deferred throughout different stages of the disease, and this heterogeneous. It is something that's really important to think about in trials. So just as an example, and I sit on a Ms Society, um, advisory panel and one of the things that we were involved in was thinking about autologous hematopietic stem cell transplantation, where it sits at the minute in the regulation and and in the in the use of the UK and there are some really important trials happening. So star M s is a trial that lead out of Sheffield looking at autologous transplantation and people with relapsing disease. And there was a bit of a lobby from the public patient groups to say, Well, you know, why should that be exclusively people at the first stage of their disease and relapsing, you know, why should we be including everybody? This isn't really fair. It's not ethical. It's not justified. But I think if you use that as an example, we have to be a lot bit careful. So, for example, if we were to be very inclusive in these sorts of trials, what would be the primary outcome that we would measure, you know, So there would be a risk that if we chose relapse rate that would be unsuitable for people in advanced disability states who aren't really having relapses anymore and that their disability is much more driven by a neuro degeneration of progression. Um, likewise, the treatment effect we know from the observation all data that their treatment effect is most when these interventions are used earlier. It's not to say that they're not effective later, but perhaps we need to use different outcome measures. Perhaps the there there are more marginal gains. So I think, you know, we have to be careful that we don't dilute the measurable treatment effect and make this a negative trial. Um, and as I mentioned, you know, the adverse event risk will be different. Different people, um, in terms of age, in terms of disability and again as I mentioned the utility. So actually, some people with more advanced disability may be willing to, um, to have more intervention and more burden of treatment to gain stability, and that will come across in quality of life and satisfaction and things like that. So I think although we want to be inclusive in trials, there are reasons why sometimes the inclusion criteria need to be quite strict. So what can we do? What is a contemporary trial in terms of inclusion? Well, I think there's some really interesting work using sort of harnessing things like machine learning. So this is nice work. That was from, um, and Shaggy, who works with architecturally Declan Child and others in the U. C. L um imaging lab. So they use trial data from thousands of people with multiple sclerosis, and they looked at disability sector. But they looked primarily at the imaging signature there imaging phenotype. And what they found was that by having an unbiased approach and using machine learning, they could categorize people into different categories. So not relapsing, remitting and primary progressive. Nothing that's based on semantics, but something that's based on, um, this sort of biology that was picked up on the scans and that identified some different groups that are shown here and the cortex lead normal. Appearing White matter lesion bled. But interestingly, those people appear to have a different trajectory as well. So it seems to be something that was meaningful and, um, and she has just been awarded a senior fellowship, and he's going to go and look at this and in more detail. But you know, the the the sort of the next step on from this that we may see in the future is this sort of ability to stratify medicine, and that's being done in other groups. So this is, um, lung cancer trial. But the concept of it is what's called an umbrella trial. So the idea is that you may have a lot of people with lung cancer but actually buy stratifying them. According to their disease biology. You can put them into one trial with lots and lots of different arms. So this is the complexity of that trial. And depending on the different mutations that they have in their tumor, there will be assigned a different treatment beyond standard of care. And it may be that the future of MSN, the most contemporary trial, would be to be inclusive, like this umbrella trial is. But for each person to be able to administer a slightly different treatment, um, products molecule or algorithm according to their characteristics of their particular disease. So, in trial design and thinking about contemporary trials, we've been really hindered by outcome measures, and Gavin mentioned already the EDSS, which lots of people in the room are familiar with and also familiar with the problems with it. So it's very driven by ambulation at the higher levels. It's quite insensitive to change at the lower levels to certain aspects. You know, particularly, very important things, like cognition don't really change the EDSS and so on. Um and yet we're really stuck with it is the common currency of clinical trials. Um, and you know, that's because of the regulators, mainly, um, so the European Medicines Agency and the FDA in the US um, you know, they recognize that the EDSS. Is is not great, and they do invite us to consider alternatives, but they would always be a secondary measures. Um, and they recognize this situation isn't ideal, but they are unable to recommend alternatives, you know, on the whole, um, in the broad. So, um, it was great to see the think hand campaign which class was really central in, um, sort of moving forward using the idea that when people already have problems with ambulation at the higher levels of EDSS, the hand function is so key to maintaining independence. And also, you know, the post hoc analysis of clinical trials that have been, let's say, negative on their primary outcome measure, which is EDSS driven, actually seemed like there was a signal for benefit. If you look at the hand outcomes and maybe other outcomes like, um, cognition and patient reported things which are going to come onto, um and I'm sure class might mention later the chariot M s study, um, you know, which is a really sort of trailblazing in the sense that it does have the primary outcome measure, which is, uh, using the hand. And it is including people that have much higher EDSS grades, which we would then we would normally be familiar with recruiting clinical trials. Um, so we've certainly been glad to be involved with that in Cardiff. Um, and I think another key thing is gaining clinical data from wearable. So you'll be familiar. Many of you will wear a wearable, um, in terms of your Fitbit or your garment or, you know, whatever. So lots of best were sensors every day. Um, and they can be harnessed in a really powerful way in disease, and there's lots of that going on in in things like epilepsy in Parkinson's and so on, but also in m s. Um, so this is taken from the floodlight app that was developed by Roche just as an example, because I think this graph gives a really nice example that if we imagine that blue is the truth, Um, so this is the sort of trajectory of disease that's true. And you can see that these dots there's a really high temporal resolution because someone is wearing the sensor every day and getting the readout every day seems to really track the truth. And I think what this demonstrates in a in a sort of hypothetical way is that you might detect some of the fluctuations in clinic if you happen to see someone at the right time. But these gray bars show that actually, if someone's, you know, not seen in clinic or you know they have a fluctuation in between visits, you know that's not really captured. And I think there is an acceptance that wearables could be quite powerful. But of course, the problem with it is that, you know, validating these sorts of things is really, really challenging. There are lots and lots and lots of them out there and and, you know, how do people like the EMA decide on the one that they're going to accept us? Um, you know, being valid and being kind of recommended. But nevertheless, it's a really exciting area, and I think it will come. And I think if it's not, you know, ending up being a primary outcome of trials, it will almost certainly be there and already is. I mean, we're doing some trials in Cardiff at the minute that a commercial and the patient has about three different other devices that they're given to take away with them. So they come and see us for visits. But they're also given a watch and they're given an iPad and they're given something else. So we're already kind of using these sorts of things. Um, so this is really nice work that's been driven by colleagues in Sheffield. Um, and they are really using a wearable device that's trying to track in exactly the same way the sort of real world experience of living with M s. Um and they've already done, um, validation work in M s and interestingly, also in Parkinson's disease. I think that's another nice thing about these wearables is they're not necessarily disease specifics. If you have several diseases with a neurodegenerative component, there's no reason why you couldn't, um, use a wearable device in the in the different diseases to reach the same outcome. So the other thing that I mentioned that I think is a very tricky area, but an important one is patient reported outcome measures of problems. Um, I like this data that's sort of 25 years old or so because it really just shows how long we've been struggling with this with this sort of concept of patient reported outcome measures. So in this, um uh, study which was quite simple, really and relatively small that was published in the BMJ because I think at the time it was novel and and important. They took the domains of the S F 36 quality of life questionnaire, which has shown here, and they simply asked a group of people with M s and a group of neurologists who look after people with M s which they thought was the most important in terms of life. And what you see is that the patients and the clinician's really deferred in terms of their viewpoint. So the things that are most important to patients, where things like their emotional roll, their mental health, their general wellbeing, vitality and the things that clinicians thought were most important with things like physical function and physical role. And I think you can imagine that's changed over time, I hope. But I think you know, and I think we have a much better awareness of the lived experience with M s because of this being so much more public engagement. But I think there is still a mismatch here, and that's why it is so important to capture the lived experience. The patient reported outcomes. And one of the things that these authors stated which I think still really important is that when you look at a key paper on a clinical trial, Phase three publication, you know, obviously it's all about the efficacy. What we really want to know is that that drug work and of course, the adverse events are listed and they're described. But it's often a list and it's often left up to doctors really to decide whether those side effects are sort of justifiable in terms of the efficacy and when you look at this mismatch. Accepting this data is very old. You know, there is a bit of a concern that are we the right people to judge what that balances? And I think it is really important to have, um, patient input into that. But, you know, even back then, 25 years ago and this sort of relatively crude study, there was a really good correlation here. Um, you know, between the EDSS score t score, it was called EDSS score and the patient's own report of physical aspect of their functioning, you know, so you can see that the patient reported outcomes are a good read out of that. So what do the European Medicines Agency think? Well, they encourage the use of patient reported outcomes. But again, it's very difficult for them to specify which scale should be used because there are so many, um, essentially, they're just an adjunct, really To, as I say, increase the sort of validity, Um, and what we understand about the lived experience. So the other thing to say about contemporary clinical trials is that, you know, they're hugely expensive. So in the pipeline, the reason that we have phase two is that it's the first foray into the disease experience, but it's powered to have a surrogate endpoint. So if you were to power it with a clinical end point, it would just be vast, and it would be expensive and there'd be a very high risk. So the first time a molecule is tested in the disease? Um, in a phase two trial, you would have usually an imaging outcome measure. But of course, tissue bio markers are likely to become really important in that area in the future as well. So you'll probably be familiar with things like the T two lesion volume, the volume of gadolinium enhancing lesions as shown here. I like showing this, um, video. So this is 14 years of a single patient. This is from Coumadin, a camera who's, um, the imaging scientist who is in charge of the trial that I'm involved with called Deliver MSU works in Cleveland in Ohio, and he's loaned me this, and I use it a lot because I think when you watch the video, if it works, should probably check. It works. Before I start talking about it, you can see changes in lesion volume. You know lesions come and go and so on. But I think you know, the changes in brain volume in a certain individual over 14 years are just really quite striking. Um, so you can see why brain volume change is a hugely important, um, outcome measure for surrogate and points for trials of people with progressive M s of neurodegenerative disease. Now it's good, because it, um, is sensitive. It's quite reproducible, Um, and it does align with disability, and it also aligns with treatment response. If someone responds to treatment that stabilizes, the disability usually stabilizes the brain volume. But there are more contemporary measures that I think are going to show promise and future. So, for instance, atrophy is not very specific to a certain pathology. So, you know, you can't say that it represents remyelination demyelination gliosis. Um, you know, loss of axonal, loss of neurons in a very precise way. It's representing all of those things a little bit. So magnetization transfer ratio is a little bit more specific to myelin remyelination, and I'll come on to an example of where that's been used quite effectively recently. But one of the issues is that using atrophy um you can do a multi center trial and you can record brain volume and lots and lots of different scanners, lots and lots of different places, and it's fairly valid to do that. And it's quite reproducible throughout different centers. Things like MG are much more challenging for multi center works quite well suited to single site, um, studies a little bit more difficult to get that robust validation and multi center trials. So diffusion weighted imaging is also another way in which you can get so These are nice tractor graffiti maps where you're looking at the different colors, being different directions of travel of axons. But you can also get a really nice readout about diffusion in different directions, which implies something about the Axiron so sort of radial diffuse diffusive itty being saying something about my urination, for instance. So again, certain measures of diffusion weighted imaging can be a little bit more pathologically specific to certain biology of the disease, such as my urination, um, spinal cord actually really important. You know, the spinal cord, often overlooked in imaging studies and um, cervical cord dimensions can be really, really well, um, correlated with motor disability Um And then again, when we're thinking about products that are hopefully going to encourage my elimination things like, um neurophysiology like visual evoke potential shown here, um, and things like 0 cc can be also really powerful. So I think in contemporary trials, we're going to see much better use much more thoughtful use of these sort of surrogate end points in phase two. And I said, I'll give an example. So this was a really nice example of a trial that you might be familiar with called CCM are one which was bexarotene trial, um, lead from Cambridge. Um, and you might be aware that the trial was negative overall for the primary outcome measure and also showed quite high incidence of adverse events. But actually, one of the important learnings was that when they looked at the treatment difference, So the sort of improvement, if you like in my urination, measured by magnetization transfer ratio, um, in lesions in people who were on the drug versus placebo, the biggest difference was actually seen in the Grey matter. And that was an unexpected finding. And I think it reminds us that, you know, Grey Matter lesions, we know are quite prevalent. But also we don't think of grey Matter as being, you know, the myelin density is less and we don't think of myelin a demyelination remyelination as being such an important sort of concept there. But obviously the changes that were visible, uh, we're driven by those gray matter structures. So, um, an example of where they're sort of nuance of the surrogate endpoints might become more and more improved over time. So perhaps, um, in future trials, they might specifically look at grey matter MTR change. So tissue biomarkers, you know, looking at the posters upstairs and so on. There's a huge expertise, um, in this region in terms of tissue biomarkers, and I think these are going to become increasingly important. And you know what? Appeals is often the non invasive nature of it. So sometimes markers are developed in the CSF. But then, of course, we want to try and seek markers in the blood because it's less invasive to be able to get blood samples and get them often, it might be cheaper than MRI, although, whether it will ever fully replace MRI, I'm not sure, but I think there are a mixture at the minute of people who are enthusiasts and skeptics about neurofilament. And this is showing serum neurofilament um, in the graph in this nice work by Peter Calabrese. And while I think I'm a little bit more in a skeptic camp at the moment, I think that's around the idea that a single measure at Baseline is going to predict outcome and going to tell us what to do with someone in their disease. I think there is a story to be told about serum neurofilament, and I'm sure other, um, serum and plasma markers. So I think what this illustrates nicely is the change. So here are the placebo group, and here are the group who had some treatment. Um, and I think what you can see is that if you just took a single measure, there's such change going on in these people who had placebo. They didn't have any, you know, intervention in terms of the drug around a year that taking a single measurement, you can just see would have limited utility because there's just so much, you know, such a dynamic measure. However, overall, you know, on average people who had treatment versus didn't there was an improvement. And the other thing is that this sort of dynamic changes probably meaningful. So these are people that have no evidence of disease activity while they were followed up. And these are people that did have evidence of disease activity. So relapses, new lesions and so on, and you can see that you take in account that the Y axis here is a different scale. You can see that a lot of the change was happening in the people who had quite active disease. And those people that had quite quiet disease clinically also had less change in their neurofilament. So I think there is something really that can be harnessed. But I think it's complex, and it's likely that in clinical trials were going to need to collect these sorts of markers cumulatively, and perhaps look at the area under the curve rather than just expecting to be able to use one measurement and know what to do with our patients, which I think is the slightly simplistic way that it's sometimes you know, expressed so the other thing about contemporary trials is that we have to get around the fact that trials are just so expensive and so difficult to do. Um, and people like class, you know, are going to get fed up of doing and leading on trials if the whole thing just becomes so impossible, you know, getting carry it off the ground has been such a labor of love. I think it's fair to say that if you didn't have such enthusiasts, you know that it's a risk. I think two M s and other neurodegenerative diseases. So what can we do? Well, the more novel the the intervention, you know, the more important it is to do feasibility works. Of course, if we've got a compound that's going into that typical clinical trial platform, well, that that that sort of platform, that pipeline is very well established. But if we have a non drug intervention, um, that's something very novel and exciting. Of course, there's a whole lot of feasibility work that needs to be done first. Um, that's kind of the alternative to phase one, if you like. Um, so that's around, you know. Is it acceptable to people with M s? Do they comply with it? What's the fidelity? So in other words, if you design an intervention, and then it's used as it used in the way that it was intended. And is it reflecting what you think is going on when you design it? And is it valid? Um, can it be reproduced and all those sorts of things by different people and different centers and so on? So there are some really nice UK examples of exciting interventions that do require quite a lot of feasibility work. Um, so this is one, um, that's led by some people in Nottingham, and it's led by Roche and Dyspnea, who is a professor of neuropsychology. Um, and this is a neuro psychological intervention that's used as a screening. So we know that cognition is a huge problem in people with MS. It's very common, Um, and it's very under recognized, I would say so. This is an idea where everybody who comes to our RMS clinics now and part is also involved in this, um receives a link and when they can do a screening test, so there's one cognitive test, and then there are three questionnaires about the fatigue and the mood and the cognition, and they do that at home, and then the results are pushed to us so that when we see someone in clinic, we have those results. We can discuss that with the patient, and we can apply a sort of a triage about how bad is the recognition and what do we need to do? And then we can refer them for neuropsychology or we can rescreen them again. Or we can do some sort of brief intervention or sign post them to some resources and so on. So these sorts of things, you know, very contemporary clinical trial could have really meaningful change in terms of people lived experience of M s, but really doesn't fit in that that sort of clinical trial platform. And this is another example. This is work that we've done in Cardiff where, um, we were seeking to address the fact that remaining physically active is really important in multiple sclerosis. But actually, they're sort of ability to stay physically active becomes much, much harder the more disabled you get because there are all sorts of barriers to, you know, simply going to the gym or going for a run or something. So we need to be more imaginative and keeping people physically active at that stage. So this was a Web based sort of coaching intervention, which is led by physiotherapist. But really, um is the idea is that it's the focus on physical activity rather than physiotherapy. Um, and this had really nice outcomes in our pilot intervention. And this is now, uh, subject of funding application for a larger trial, which would lead to approval. So the intervention was feasible. Um, and also, of course, it wasn't powered for this, but it did show some improvements in people's patient reported outcomes. Um, so enough to for us to warrant taking it forward into a larger effectiveness trial. So the other efficiencies that can be gained in terms of clinical trials are many, um, so an example which I really like Is that high throughput screening? So maybe avoiding some of the phase one phase two stuff is to take compounds that are already used for other diseases and to look at which of them might show a signal for being beneficial in multiple sclerosis. So this is an example of really nice work from Robin, Franklin and Edinburgh and others where they use this micro pillar approach. So these are tiny pillars, um, in a dish and you can see here, um, oligodendrocyte are interacting with them and what you probably can't quite appreciate, but you have to take my word for it is that those oligodendrocyte are laying down rings of myelin around these pillars, and that's the sort of approach, and there are other models as well. But this is one model that they've used for high three high throughput screening of compounds for repurposing. So you have a large, um, well dish, and you can get a read out from different compounds in different wells for how well they're re myelinating. And of course, we know that these sorts of laboratory findings don't always match up, and there might be other barriers, such as with the Claritin and adverse events and so on. But it does lead us to having a situation where we can go straight into Phase three with some compounds where we have reason to believe that they might be beneficial. And an example of that and another example of efficiency is the Octopus Trial, which is due to launch this autumn, where there will be repurposed compounds. So there yet to be formally announced the compounds that are going to be involved in the trial. But the idea is that we'll start with to repurpose medications versus the placebo arm, and then there'll be a essentially a surrogate outcome at the end of stage one that will enable compounds to be dropped or taken forward, depending on whether they're showing a signal of benefit. And the other great thing is that part way through the trial. If another compound comes to light or a non drug intervention comes to light, which is thought to have promise, then it can be added in. And we have this long running placebo. Um um, that will be a comparison and again because this will be recruiting people in the slightly more advanced disability stages. Um, it's important that the outcomes will be not just EDSS based, but also based on cognition, hand function and and patient reported outcomes as well. And Denise Fitzgerald has been really involved in this study and lead on treatment, selection and so on, and also I believe I'm going to be involved in collecting samples in Northern Ireland, and hopefully they'll also be biobank samples from London, Edinburgh and Cardiff, which will provide a really nice, um resource, I think, for scientists to be able to study because the samples which will be collected during stage one, um at six months time points will be linked to very high quality clinical data and radiological data. So I think it's really important that contemporary trials do buy a bank because even if we don't at the time of collecting the samples have a hypothesis that we're preparing to address. We know that there are scientists out there that will be able to make good use of those samples and due course, and then another, um, sort of aspect of efficiency to try and get around some of that cost. So we just put in a long term extension application for a trial called Deliver M s, which we've been involved in, which is a three year trial, and we want to measure out, comes out to hopefully 10 years in the fullness of time, but initially up to six years. And you know, when you put these multi center clinical trial a grant, applications in there just so phenomenally expensive that you you know, you can't you can't really see that they're going to get funded and you have to think outside the box. And some of the things that we've been really interested in looking in that application is some of the ability to do visits remotely. So we talked about wearables already, but also to have things like Web EDSS, telephone, EDSS and so on. So it's important that we can understand in the future the validity of those sorts of measures, and I think remote clinical trials. This was a paper that I helped to write that was published this year, and you know, thinking about it. The more you think about it, the more sense it makes. Because there are lots of people who can't access trial centers either because of geography or because of the disability. And having a remote clinical trial, fully remote clinical trial would potentially make things more inclusive and and so on. There are lots of reasons why it would be a great idea. But as you can imagine, huge challenges, Um, especially when the regulators have a certain set of tests that they need you to do. Um, so Gavin also mentioned this group, the Child You Kms Trials and Registries group And I think, you know, another efficiency is that sort of network, and it's been used to great effect in the Progressive s Alliance. So, getting together, people who are bringing together a network with good sort of complimentary expertise and skills who are going to solve the problem in a better way than any one individual avoiding duplication. You know, the main driver, I think for us getting together the trials and Registries Group was simply understanding what we're all doing in different places, you know, and trying not to duplicate and trying to harness power. And so, um, so I can see Gavin on here. Stella, I'm not sure if I can see anyone else from from Ireland. So this has been a really nice opportunity to meet twice a year. It's been online because of the pandemic, really. But I think it will probably remain online. It's been quite a, um, you know, efficient way of getting together, sharing experience, um, supporting each other in developing, you know, good quality, sort of registry, data, trials, data and so on. So the final couple of things I was going to say so treatment landscape is important. And you know you'll be really familiar with these sorts of graph. So Gavin listed the medications earlier, um that are now available. And I think this might be out of date because we don't have I've added bonus. Um, okay, so, um, it's changing every year, Um, in terms of medications, But I think there's generally an acceptance that there is a burden, and it's not just side effects. It's kind of convenient having lots of blood monitoring, so on versus efficacy. And I think this graph would be drawn differently by different people because, of course, we don't have a head to head data. But the point is, um, that there's a difficulty in using some of the most efficacious medications because some of the because of some of these concerns. But I think one of the interesting things to me was that in data from Cardiff, where we looked at people who were on early intensive therapy versus an escalation approach, yes, that did show a small change in outcome, and the numbers were small. But actually, what I wanted to focus on for this talk was the median time spent on any drug was two years was really short. So a phase three clinical trials, two years. And if you're saying that 50% of people on certain drugs stop at two years, then you know, translating clinical trial data into the real world is really challenging. So a big piece of work we're doing now with the help of, um, all and Stella and Gavin have all submitted data I think is looking at DMT Durability in the real world. So this was just the card. If data and we're currently doing the analysis from the 11 centers have all submitted their own data on this, but we're really interested in looking at that. And actually, I think what you can see is that 50% of people on the injectables um this is annual, um, these lines, unfortunately, it's written in days which has been helpful, but each line is a is a year, and actually you can see that about 50% of people on the injectables do stop it about two years. But, you know, it's sort of hardening to see that other medications and that are more modern. You know, people are staying on them for longer. Um, but I think there are really important things to think about in terms of just persistence of therapy and some of the things that Gavin listed I would really agree with. So some of the questions which are, you know, subject to, um, debate. And I think we'll need quite complicated trials to look at, really as the sequencing of medications, when to escalate, when to deescalate. You know, Should we be, um, you know, deescalating nation disease. Should we be stopping things like adaptive dozing? So I think these sorts of things, you know, they deliver a mess trial that I mentioned is really looking at sort of treatment algorithms. So it's all licensed therapies, but how do we use them? And I think those sorts of trials will become increasingly important. Um, And we've been in touch, you know, recently, um, with Stella and Gavin about thinking about sort of adaptive doses of some of the anti CCP twenties and whether we could do work to think about a clinical trial in that area. And it is quite challenging, and it and it will be expensive. So thinking about those sorts of protocols, So finally, patient priorities Will you'll, I'm sure be familiar with this James Linda Alliance, which was a partnership between the MS Society, the James Linda Alliance where they, um, asked people members of the public affected by M s to rank their priorities in terms of this research. And you'll probably be familiar with it, because whenever you submit a grant to the MS Society of the UK, you have to say how your grant addresses these priorities. But there are other priorities setting exercises as well. So this is a really nice one done by the Progressive s Alliance. And when you look at the sort of priorities that these, um you know, exercise is our setting out there not really easy to address in a clinical trial, but they're really important. So one of them that I could highlight here one of the combinations of pharmacological and rehabilitation interventions including, but not limited to physical activity exercise CBT mindfulness that are most effective in treating pain. So, you know, really important questions that have come from, you know, stakeholders and, you know, people affected by M s. But designing a clinical trial to address that very complex. So this is why we need contemporary clinical trials and an example of that. And I'm pleased that Gavin mentioned employment in his introduction because that leads really nicely into this, which is from Helen Ford and leads. And again, another nice really, um, example of a trial that have been delighted to be involved with because it's addressing an unmet need in a in an innovative way and patients love this. This is a acceptance and commitment therapy intervention, a CT, which you might have heard of, um, sort of allied to CBT in a kind of a model or approach, but obviously different and what it's trying to achieve. Um, and this is for people who are have MSN are in work but have some sort of instability in the workplace. So by by that, it means that there's a risk of them losing their employment. And this is an intervention to try and keep them in work. Looking at these sorts of outcomes and it's been really nice and they take up from patients has been huge because it's, um, something that they're, you know, very interested in doing. So I'm going to finish that. Hopefully I've given you some examples of how standard clinical trials have you know, been around and are going to stay around. But how? We really need to use contemporary approaches to our clinical trials for various reasons. We want to make sure we're addressing questions which are relevant, which are addressing all of those things that Gavin listed, which are the sort of really world complexities of the care we provide people with M s. We want to get result that are reliable, and that's challenging in a complex disease and studying outcomes that are relevant but also valid. There's a huge cost barriers I mentioned, and we need to sort of work together and networks and trying to overcome those. And some of these really innovative interventions are great to see. Um, in you Kms Research. So that's it. Thanks very much for listening. This is the card. If you miss family. Really? I'm not sure I have gotten the time. Whether we've got time for questions. Got a few minutes for questions. You're welcome. Questions from the audience in the room and also online. And Michelle and and then I will read out any questions that are coming through online. So thanks so much for a great talk. Uh, questions in the room close. How do you reconcile the real world evidence for highly effective treatment being superior to, um, platform escalation? Um, with your extension request? Yeah. Deliver. So yeah, so? So I think that's it. So what class is saying is that so deliver m s was looking at whether we start someone straight away on the most powerful treatments we have. So the monoclonal antibody treatments, or whether we start someone on a modestly effective treatment and monitor them and then if that if there is breakthrough disease activity, then we step people up. So I think the way that we've reconciled it is that there's no doubt that if you start someone on a monoclonal antibody treatment straight away, they will do better in terms of disease stability. So relapse rates and disability. But I think there is still uncertainty about number one. Is that required for everybody? So would some people. Is it about selection? So would some people have had a better experience if they started on a modest advocacy treatment in some individuals that will be sufficient to control relapse rates and disease activity on an MRI scan and some of those people will not need to escalate. And then I think the other thing that we've used is evidence from from our data and others that the registry data that says that early treatments better. Some people switch to escalate really late, you know, So they're escalating after five years, and if we're going to escalate, we need to escalate quickly. So in deliver we're monitoring people at baseline is six months or 12 months. Usually if they're going to escalate, it seems to be escalating by 12 months. And it's possible that unlike the real world data, where escalation in in our card, if data as well happens far too late, it's possible that if you monitor really proactively and if you escalate really early, maybe you'll get selection, right? So the people that need a monoclonal, we'll get it in a timely way, and then some people, and you do get I mean, we all get lots of people who just simply are scared of starting. Um, you know, high efficacy treatment first, so they will get their, um you know, preference, which is to try on a moderate advocacy. Any other questions in the room? I just have a quick question. Thank you so much as a fascination talk and when When there's a discontinuation study. Yeah, is it Also look at whether they are changing on another DMT or whether this is discontinuation entirely off of the MG. What proportion are remaining on some DMT. So the most recent discontinuation studies have randomized people to stopping and continuing, and it stopped. It stopped everything, and I think the criteria for them. So there was a disco M s study, and I think that has now reported and that was randomizing people who have fulfilled certain criteria. So it's people who have had no disease activity for a certain length of time. Sometimes people who were a certain age and so on because we know the disease activity in terms of relapse is is less common beyond a certain age and so on. So people who fit that criteria, they got randomized to carrying on and stopping and and it and it was proper stopping. I think the difficulty with that was that the results, the readout has been quite gray. So there wasn't the crisp kind of information that we hoped for to tell us how to manage patients. But that's the stopping trials that I've seen so far. Are there any questions coming through on the chat online? Okay. And we bon. Sorry if I missed your hands. Go ahead. Thanks. Fantastic talk. I have a very naive question. I'm fascinated by variables, and I really like to know more about this. How is this taken up by patients? Wonder, especially the older patients. Do they take it as easily as the younger ones? Are there any preferred wearables Unbearable. That disease stage is your age is that gives the most information. Yeah. So it's a good question. I mean, we do need to think about that in terms of take up. And you know, clearly things are moving on year by year in terms of people, people's digital intelligence, if you like. So lots of people now online. But even, you know, households being online isn't universal. So we've had things recently, like the neurons thing that I mentioned, which is that digital intervention, where we send the screening questionnaire to people's home and then they come to clinic and we have the results. You know, there will still be people who say I don't have an email address. I don't have anyone in the family who can help me get online and do it. You know, it's quite surprising. So even that you know which we probably presume is fairly universal isn't the case. So yeah, you're right. The more complex you go in terms of technology and wearables, there will definitely be problems. And I think you're right that people, you know, it tends to be generational that there's less familiarity in the older generations. But of course that's that's not a given. Um, so yeah, I mean, class might have more experience and wearables and take up. I think generally it's good. Generally, people are excited, enthusiastic about it, But there are lots of problems aren't there? I mean, you send people home with two or three devices now for these commercial trials, and you know they don't always use them. They probably sit on the bedside table from the time they don't always come back. And that's a bit of a risk. And so I mean, I think there's challenges. No, I mean confirmation. It's remarkably complex to use them consistently, you know, in studies because of technical issues. So we don't use the mechanical practice. That's all I can say. Some people obviously have their benefits, et cetera, and that's sometimes it's really hard for them. But it's not a standard. Yeah, so for the people online class really is just saying that it's not a standard thing in clinical practice, and that's a lot to do with complexity. And in terms of your question about whether there are some that do better read that I think that is part of the problem is there are just so many and even for one wearable when you look at the data that you get back. We did a study in people receiving intravenous immunoglobulin for different conditions, including primary immune deficiency as well as neurological disease and so on. And they were showing me the data that they got back. They gave people just a Fitbit, and they were showing me the data that the statisticians had to deal with, you know, and it's just fast. So if you have all that vast data for each wearable and you have so many wearables, I think it's really difficult for us to get as a community any nice, clean impression about what's the go to where? Yeah. Okay. Question. Um, So is there any concern about data protection or potential health discrimination for patients using those variables in the future? Maybe not now, but leaving the future if they're being tracked and disease worsens. Yeah. I mean, I think the data management is a big challenge as well. So as an example when we did the leep m s study, which I mentioned. So this was the coaching intervention. It was fully online, and we wanted to do a thing where we said, Well, if anyone's on strava or any sort of, you know, exercise, um, app, You often have followers and you often give kudos or you give a kind of a thumbs up or something, and that's a really important motivator. So the physiotherapist and the coach, we're really important, you know, really keen to have that. And, you know, data governance just didn't allow it. You know, there's no way you're going to be able to share one person's profile with other people in your study and so on. So it was just an absolute no no. So I think that sort of information governance stuff is a big hurdle to overcome for sure. Yeah, Rachel, um, it's just sort of interest in clinical trials in the future. Do you think there will be, like, a role for some complications? Me to step away from select people based on the MSR relapse remitting. But look at your MRI. Yeah, I agree. I realized I probably should have been repeating the questions for people or nine. Sorry about that. But Rachel he was just asking about whether there might be a move towards trials that are not using. They're sort of semantic phenotypes of relapsing remitting progressive, ms, but using a more kind of biological approach. Absolutely. So as I gave the example in cancer. And I think it's much more common now in In In cancer trials. To not have a trial that's based on a semantic description like lung cancer is much more based on a biological description. So it might be some genomic thing, or it might be some biomarker and so on. And, you know, we had to talk from the neurosurgeons recently and they talked about brain tumor management. I mean, oh, my gosh. Even since I was studying that, you know, as I came into registrar training that's completely been revolutionized by the use of molecular typing, of tumors and so on. So, yeah, I think inevitably we will get towards that. And as I said, You know, maybe we will approach that umbrella trial where we can be inclusive and say Everybody's welcome into the trial. But everybody that comes in gets characterized in a certain way, whether it's imaging, whether it's tissue, biomarkers and so on. And then they may be receiving treatment that's more appropriate for their particular biology. Yeah, I hope so. Okay. Thank you, Emma. Stand around this morning, so that may be more questions over coffee, but thank you for a great talk.