: Multiple sclerosis [MS] is an inflammatory demyelinating progressive disease that affects the central nervous system. There is no cure for multiple sclerosis, but advances in early treatment are available, which can modify the course of the disease. In this Neurology Live® Peer Exchange discussion entitled, “Advances in Early Treatment Approaches in Multiple Sclerosis,” I am joined by a panel of my colleagues, all experts in the field of multiple sclerosis. Together we’re going to discuss the use of new therapeutic options. We will review the latest clinical trials and provide practical perspective on how the recent data apply to your clinical practice.

I’m Dr Fred Lublin, the Saunders Family Professor of Neurology and the director of the Corinne Goldsmith Dickinson Center for Multiple Sclerosis at Mount Sinai Medical Center in New York.

Today I’m joined by Dr Amit Bar-Or, the chief of the Multiple Sclerosis Division in the Department of Neurology of the Perelman School of Medicine at the University of Pennsylvania in Philadelphia, and the director of the Center for Neuroinflammation and Experimental Therapeutics.

Dr Peter Calabresi, the director of the Division of Neuroimmunology and a professor of neurology at Johns Hopkins Medicine in Baltimore, Maryland.

Dr Patricia K. Coyle, the director of the MS Comprehensive Care Center and a professor of neurology and interim chair at the Stony Brook Neurosciences Institute in Stony Brook, New York.

And Dr Stephen Krieger, an associate professor of neurology at the Corinne Goldsmith Dickinson Center for Multiple Sclerosis at Mount Sinai Medical Center in New York.

Thank you for joining us, so let’s begin.

We’re going to start with discussing diagnostic issues of multiple sclerosis. We’ve changed the criteria for diagnosis, the guidelines, several times since 2001. The first McDonald Criteria came out in 2001. We modified them in 2005, again in 2010. And then the last time was in 2017, and that’s been published in The Lancet Neurology. We’re going to discuss a bit, Stephen will take the lead on this, on what’s new in McDonald Criteria 2017.

: I think the goal of the McDonald Criteria, as you said, has been to define a gold standard by which we can diagnosis this disease. And that’s been incredibly important since MS is, of course, so heterogeneous; it’s important to have criteria that we can use that we’re all in a sense speaking the same language about what constitutes defined multiple sclerosis.

The 2010 McDonald Criteria I think simplified matters, particularly pertaining to the MRI [magnetic resonance imaging] scan. It allowed us to use really any 2 lesions and any new lesions, as I like to think of them, in their characteristic locations, as being able to diagnose MS on the basis of dissemination in space, and dissemination in time in the right clinical circumstances.

The 2017 criteria moved us a little bit further, and I think, again, this makes it a bit easier to diagnose MS in some cases. The real additions of the 2017 criteria are that if you have a patient who has a classic clinically demyelinating syndrome presentation—optic neuritis, a brain stem episode or a partial myelitis—we can use the MRI scan to make that diagnosis. And now the presence of, again, lesions in the characteristic places, periventricular, infratentorial, juxtacortical, are essential, and to that we’ve now added that cortical and juxtacortical lesions both count.

I think that’s a reflection of the fact that we understand that MS is not just a white matter disease, but it principally involves grey matter, and so looking for those cortical and juxtacortical lesions, lesions that touch the cortex, now both count.

On the other side, we had not used spinal fluid all that often in making an MS diagnosis for the past, I don’t know, decade or so. And the 2017 McDonald Criteria promoted spinal fluid findings back into the pantheon, if you will, for MS diagnosis. And so now the presence of oligoclonal bands, unique to the central nervous system, on spinal fluid analysis can be an important part of the MS diagnosis; it can substitute for dissemination in time. 

So in essence if you have a patient who has 1 attack whose MRI doesn’t meet criteria for both dissemination in space and time, doing a spinal tap, finding unique oligoclonal bands, can allow us to make a McDonald Criteria diagnosis of MS in that patient.

: Patricia, you’ve been a proponent of CSF [cerebrospinal fluid] analysis for a long time. Any changes that you’ve made since this came out?

: Absolutely none, because we routinely looked at spinal fluid in all the patients. We are missing a diagnostic biomarker. We’re trying to make the diagnosis as early as possible. Therefore, we want to be critically correct. And my concept of a work-up, and I don’t think this should be a clinical diagnosis, it should be a laboratory-based diagnosis, is a judicious choice of blood work; brain MRI with and without contrast; cervical and thoracic MRI scan going down to the conus, so you completely image the central nervous system; and evaluating spinal fluid; and the single best test, oligoclonal bands. Now that preempts the IgG [immunoglobulin G] Index. I think every patient deserves that. We’re tending to do optical coherence tomography also routinely. And let me make a bold statement. I believe if you’re working up somebody for relapsing MS, we should be routinely checking IgG to aquaporin-4 and MOG [myelin oligodendrocyte glycoprotein].

: Let’s hold that thought for a minute and let’s get back to spinal fluid. Has this changed? Has McDonald Criteria 2017 changed your approach?

: We’ve been relatively proactive in terms of getting spinal fluid, so this is not radically changing the way that we approach it. But I think it is an important reminder that the spinal fluid does add a layer of added value and is particularly important to keep in mind in the context of those presentations that are not going to be particularly typical of standard relapsing-remitting forms of MS.

:I’ll say it hasn’t changed my approach too much either. I would agree with everyone. I think 1 point that I would like to emphasize is that the McDonald Criteria were not designed to discriminate MS from other syndromes. They’re mostly to increase our confidence that a patient with a clinically isolated syndrome really has definite MS. And so I would urge caution in the differential diagnosis. We see a lot of people for second or third opinions who have been misdiagnosed, and now with MRI we can usually count up white spots that are not necessarily demyelinating lesions. So I think it’s important to seriously consider alternatives.

: That raises 2 things. But first I’m going to finish up with spinal fluid, because I am doing more spinal taps. How about you Stephen?

: I think we are doing more of them; and in particular, for the reason that you said. The McDonald Criteria, it doesn’t give us a differential, we have to still think about that. That’s the place where you said MS needs to be a laboratory diagnosis, not a clinical one. But I do still think the clinical judgment of what to look for in that differential diagnosis and when to look at spinal fluid to pursue some of those things remains really important. The laboratory tests can’t be done without the clinical judgment that chooses them.

: Obviously the clinical is key. But what I’m saying is every patient deserves a laboratory work-up.

: They deserve the work-up they need depending on their history and physical. But 2 points that Stephen and Peter made are worth repeating. McDonald Criteria is for diagnosing MS, not differentially diagnosing MS. And the second is that the criteria that were developed for McDonald Criteria were based on classical CIS [clinically isolated syndrome]—optic neuritis, a partial myelitis, or a brain stem cerebellar syndrome. And anything outside of that, one has to be much more cautious, which is what I think you were alluding to.

The other is that McDonald Criteria 2017, for the first time I believe, brought up the issue of misdiagnosis. There’s been a lot of interest in that because there ends up being a lot of misdiagnosis out there. So a big part of our education activity is making sure people are diagnosing this correctly. Also we’re going to discuss this, we’ve stressed treating early, because treating early, people do better. But there’s also the risk of treating someone before they have the right diagnosis. And we have people on agents, and any adverse effects on an agent that are not going to do you any good, are intolerable.

Updated McDonald Criteria for Diagnosing MS

Let’s go back to the diagnostic work-up. Pat outlined that we’ve had this debate before of what’s the minimal work-up. At our place, for most the minimal work-up is a history, a physical, and an MRI [magnetic resonance imaging], and everything else depends on what the patient shows us.

But when you think about the things that you want to look for, you want to look for red flags, the ones that will make you look farther along. And there’s a very good paper that gets into this idea of misdiagnosis that David Miller, MD, published in the 

 in 2008 on red flags. It has really nice tables on mild, moderate, severe red flags to look for in MS. Things like persistently enhancing lesions, skin lesions, pulmonary issues, bone pain, and such. Other thoughts on diagnostic work-up?

I think what so much of the misdiagnosis work and the literature have looked at is not necessarily missing rare diagnoses, ones that might be indicated by the red flag, but over-ascribing an MS diagnosis to, let’s say, somewhat suggestive MRI changes in the absence of a truly compelling clinical history. The McDonald Criteria being applied to patients with fatigue, or with headaches, or with nebulous nonvertiginous dizziness, can really lead people astray. And I think that’s in some ways the biggest red flag.

When the “Miller red flags” paper came out, it was harder to make an MS diagnosis. The criteria, the bar was set higher. And now that it’s easier and the MRI requirements are lucent in a sense, to make the diagnosis earlier, we really do run the risk of making it erroneously in patients who may not have a disease, not necessarily some other rare disease.

If you look at the patients who were misdiagnosed migraine with lesions on a brain MRI—middle-aged, in their 50s or 60s with small vessel disease, lesions on the brain, MRI&mdash;overinterpretation without the critical setting of a clinically isolated syndrome.

I think that’s all true. One of the challenges, of course, is the recognition that there’s this entity of RIS, or radiologically isolated syndrome. And we know that there are people who can live a full life with MS and be found to have had MS on autopsy. This is almost certainly uncommon, although we clearly don’t know the numbers and the denominator. But the challenge of understanding when someone has what looks like MS on an MRI without having had any clinical presentation, or maybe some fatigue, and whether that’s related. And so, again, this is where the laboratory diagnostic tests can come in as an added value.

I think it’s interesting. They presented the 10-year data on the global RIS registry, 51% had presented clinically with MS.

Just to expand on that, of course we wouldn’t want to treat people if they were truly asymptomatic throughout their life, but I think the problem is that people can have cerebral lesions and have cognitive impairment, and it’s very hard to know what is age appropriate, cognitive performance.

And in that RIS 10-year follow-up, 2 of the most important drivers for who among those people are destined to have clinically demonstrable MS were positive spinal fluid, and the presence of spinal cord lesions. Two things that I think tell us something about the disease process and the riskier lesions, the ones that are more likely to drive clinical signs and disability as time passes.

Diagnostic Work-Up for MS

A larger diagnosis with MRI [magnetic resonance imaging] scans started in, what, 1981, at your institution Dr Calabresi?

MRI has certainly revolutionized our ability to study the disease and diagnose it. It’s part of the McDonald Criteria, as we just discussed. I, of course, am an advocate of getting an MRI at the time of diagnosis. And I agree with Pat that one has to get the full spinal cord down to the conus. Sometimes we will even look at the roots and the lumbosacral to look for other explanations for lower extremity problems.

I think that as Stephen said, it’s a real issue though, this issue of nonspecificity. I’m excited about some of the new techniques such as central vein sign, which may add a layer of specificity to some of these small white spots that we would otherwise, or some people might call them, demyelinating. And with some of the newer techniques we are seeing truly cortical lesions, although I think we’ll still have to be very careful about being confident that those are real and not artifactual.

Are you using central vein sign now?          

We are not allowed to use it on our clinical scanner. Apparently it’s not an FDA approved pulse sequence acquisition. So if we happen to get it for a research protocol, we look at it, but we’re not allowed to have it on our clinical scanner.

Could I say, why haven’t we been able to get our act together to standardize how we do MRI scans for MS [multiple sclerosis]? Wouldn’t that be helpful? We haven’t even been able to do that.

CMSC [The Consortium of Multiple Sclerosis Centers] every couple of years puts out a reasonable recommendation on what’s done, but for the most part, we don’t own the scanners. Right? At academic centers there’s a radiology department that thinks they own them.

Ellen Mowry, MD, at our center took the CMSC pulse sequence acquisition and went to RadNet Inc, which is the national network, and asked them if they would consider using this. And they accepted, so they are starting to now use a standardized protocol for MRI. It hasn’t been a widespread application, but I think it’s a start. I think we need to advocate for that.

In general, as the software gets better we’re able to see things on scans that are more detailed than what we could see 5 or 10 years ago, and I do think it’s important to make sure that what we’re doing in academic settings and in research can be applied broadly. Remembering, too, that the McDonald Criteria, the way that those lesions are looked at was performed on 1.5 tesla MRI scans. We can see more than that now. But I think upgrading hardware is difficult in the community, difficult across many sites. Upgrading software is easier. I’m hopeful that things like central vein, which is really a software upgrade that just needs to get approved and rolled out, might be easier for all-comers to be able to have performed, as opposed to let’s say a 3T or 7T MRI, which is going to be incredibly expensive and cumbersome to try to implement broadly.

What’s the best sequence for a central vein?

Well, there are sequences to allow for layering of the FLAIR [fluid-attenuated inversion recovery] and a susceptibility-weighted sequence, to look and see the area within the lesion that’s actually the vein itself. It’s a nice example of being able to see something with modern technology that we’ve known has been true for a hundred years, that there are veins within these lesions. It’s just they’ve been basically hiding in plain sight with our standard T2 or T2-FLAIR imaging.

MRI in the Diagnosis of MS

We’re going to move on to discuss the clinical courses of multiple sclerosis [MS]. We defined these in 1996 and then we modified them again in 2014, although it’s called the 2013 modification. We now look at 4 different clinical courses. And this has some importance on how we approach patients and their therapy. There’s clinically isolated syndrome, which is the terribly inelegant term for what is the first attack of MS. And just as an aside, we tried to change that term at one point, but what we came up with was worse, and no editor wanted to hear it, so I won’t bother you with that. So we’re stuck with clinically isolated syndrome.

First attack of someone, if the MRI [magnetic resonance imaging] is consistent, then we have very good data, which we’ll discuss the support in treating those individuals. Now when we came up with McDonald Criteria 2010, where we made it easier to declare dissemination in time, and now even easier because we’ve added spinal fluid in 2017, about 50% of what used to be clinically isolated syndrome became diagnosable as MS. That’s that.

But then there’s relapsing-remitting MS, which is the most common form, with acute attacks and a period of quiescence in between, or at least clinically quiescence. Secondary progressive MS, or a percentage of individuals, and I don’t know the percentage but I’ll be curious of your thoughts, with relapsing-remitting MS, after a period of time, they start to gradually worsen in between attacks, with or without superimposed attacks.

And then there’s primary progressive MS, which starts out with this gradual worsening, not an identifiable clinical attack, and that may or may not have superimposed relapses later on. And then we subcategorized all this as to whether there was activity. And I thought we made this pretty clear in the paper, that activity was either having had a relapse, or new activity on the MRI—a new T2 or gadolinium enhancing lesion. But it had to be framed in time. You had to pick a time frame. And we didn’t really care what time frame you want. Just like if I see someone in clinic, when I get done with my note, I’ll say, “Impression: relapsing-remitting MS, active over the past year,” because they have had a flare-up or maybe a new MRI.

The second subcategorization was for patients who were in a progressive phase of MS, so whether they were truly progressing or not; again, over some period of time. Because in both primary progressive and secondary progressive, progression is not inexorable; it fluctuates, and they can plateau for long periods. And when we do clinical trials, you want to get people who are actually worsening, progressing, so you can measure a change. That also needed to be framed in time.

Those were the 4 types, the 2 characterizations we did, and 1 other suggestion that we made, which is worth noting, was that one shouldn’t use the term, “the patient is progressing,” unless they’re in a progressive phase of MS, and worsening, independent of a relapse. Whereas a worsening patient may be progressing, or they may be a relapsing-remitting patient who’s developing stepwise changes.

Despite those definitions, and the strong admonition that they must be framed in time, we’re going to talk about siponimod. But when siponimod was approved, it was approved for clinically isolated syndrome, relapsing-remitting MS, and secondary progressive with activity, full stop. And that’s a problem because they didn’t define activity; and more importantly, they didn’t define the time frame. And so theoretically every secondary progressive patient had activity at some point because they had to be in a relapsing-remitting phase first.

And they were only talking about clinical activity. This is for the FDA. The European authorities have also just approved a drug. They went 1 step further and said activity could be on the MRI, or clinical, but they still failed to give a time frame. This is causing some confusion, and so we’re hoping to at least straighten out that confusion now, with those comments.

Clinical Courses in MS

I do have a question for you guys. What do you think is the modern conversion rate to secondary progressive MS [multiple sclerosis], in the treatment era?

Well, I don’t think there is any question that it’s going down. If you look at the data from the EPIC study, from University of California San Francisco Medical Center, where we talked about within 10 to 15 years this high proportion converting from relapsing to SPMS [secondary progressive multiple sclerosis], it’s well less than 20%. Part of this is likely the treatment era. Part of it may be that the MS disease process is moderating with time, maybe environmental changes. But I think we’re seeing a smaller proportion of patients move from relapsing to secondary progressive disease over fixed time periods of 10 years and 20 years than what’s been reported historically.

One of the challenges when communicating to our patients what the goals of treatment are is that all over the internet it is still written that 80% of MS patients with relapsing disease will develop secondary progressive MS in let’s say 10 years. And that’s very old data, from the pretreatment era, but they still exist in perpetuity on the internet. And I think one of the things that I find myself trying to do is to explain what you just said, that the likelihood that someone being diagnosed with new relapsing MS in 2020 will develop secondary progressive disease in that classical way of insidious worsening of physical disability in 10 years, that certainly is not 80%. It’s much lower, and that’s progress and it’s optimistic, but I think we have to try to convey that to our patients as a goal, but also something that they can realistically expect.

We nowadays refer to “the new MS,” and I think this is a very important concept for patients to hear, that being diagnosed with MS today is not the same as having been diagnosed with MS certainly 10, 15 years ago. And the hope is that we are modifying the course meaningfully. We don’t have as much data as we’d like, but I suspect it’s not just a change in the MS natural history nor just the lead-in bias of making earlier diagnoses and then a longer lag time to emerge as secondary progressive, but an actual change in the natural history of the disease with a treatment that is limiting the conversion to progressive disease.

I agree. I think to answer your question that we probably cut the progression rate in half is my experience in reading the literature. But I do what you do. With my clinic notes I comment whether they’re having relapses or new MRI [magnetic resonance imaging] lesions. But I think there’s too much emphasis based on these somewhat artificial categories. Patients have it in their head that they’re relapsing, and then one day they come into the office and you tell them they’re progressive, and that means now they’re untreatable or they’re going to end up in a wheelchair. I think that does a disservice to the patient. So I talk to them about inflammation and drivers of progression and try to hit a positive and optimistic note that we have a lot of treatments that actually can impact these processes, but there’s still obviously an unmet need as well.

I think that patients think that the categories are akin to a cancer stage, something where there’s a very specific prognostically worrisome thing that either just happened or is about to happen. I’ve worked on something called the topographical model, which animates in a sense between these categories. It shows it more of a continuum over time. And it speaks to the idea of the categories themselves being really a time-locked phenomenon. It’s making sure that we’re looking for evidence of disease activity and that we’re looking for signs of progression so that we can address those things.

But I think you’re right. I try to talk to the patients about treating MS. I haven’t used the phrase “the new MS,” I like that. But treating MS as a whole and that our aspirational goals are to prevent disease activity and forestall progression perhaps forever.

I think we also realize the importance of a wellness program and boosting CNS [central nervous system] reserve and identifying comorbid factors and making sure they’re treated. And I describe that as almost a separate treatment for MS. I think that’s a very important coupling of the use of a DMT [disease modifying therapy].

Yes, discuss that a little more, what the concept is because wellness is a very important one.

There are increasing data, there are lifestyle choices that can hurt MS or help MS. You don’t want to be smoking. You want to be in optimum body weight. You want to be eating healthy. You want to have good sleep. You want regular physical exercise. We’re finding that’s promoting brain health, central nervous system health. You want intellectual stimulation, social stimulation. These are really positive. These help people age better. And if you have a disease like MS that targets the CNS, this is going to promote CNS health, you’re going to weather MS better. So I’m really speaking about that as a separate treatment that they need to follow.

I do think that, back to progressive, that’s a term that scares people. Because very often, look at people in the first attack, and they’ll say, “Do I have the progressive form?” I also want to give at least 1 caveat for these clinical courses, and that is, they’re descriptive. And our hope is, I don’t think we’re going to have it for the next time we meet to discuss these, but that we actually come up with biological bases for this. Because right now we don’t have a biomarker—either a fluid biomarker, and certainly not an MRI biomarker yet that could distinguished between CIS [clinically isolated syndrome] and primary progressive, or anything in between.

Modern Conversion Rate to Secondary MS

: Let’s transition into therapy. Multiple sclerosis [MS], to those of us with graying hair, surprisingly, has become the leader in the area of neurotherapeutics. We had our first therapy in 1993 and now we have—I can’t keep track&mdash;18, 20, whatever agents that we have to choose from, which is remarkable. And we’re very good at treating it, at least those forms that we have therapies for. And we’ve also developed considerable expertise on how to study it and how to do clinical trials to get answers and such like that. But that takes us to the therapy. Amit, why don’t you start us off on how we treat. 

: One of the main shifts in the field I think with affirmative data to support it has been the importance now placed on treating early.  

This, of course, is predicated on the view that once there is damage to the central nervous system, we’re not very good at fixing injury that’s already occurred, and the imperative is to minimize any damage or stop it if we can. We also know that the treatments initiated earlier translate into better outcomes, and the best data, although relatively short periods, are from the clinical trials where there may be a control group, such as a placebo, even an active comparator, with then a switch to open label with the higher efficacy therapy. And those individuals who are not on the higher efficacy therapy to start with never catch up. 

And so the issue is that we don’t want to lose brain or spinal cord function, and we’d like to initiate treatments that are appropriate for the individual, and the best that we can do in terms of predicting their course. But also once treatment is initiated the other principle is to proactively monitor on any treatment and make sure that we’re able to capture relatively quickly the context of disease emerging in spite of the therapy. So that failure to fully control disease in addition to any issues of safety and tolerability would warrant having a discussion around switch.  

: One of the challenges was figuring out when to treat. And this change with the first clinically isolated syndrome [CIS] study, which I think goes back to 2000, 2002, with interferon. But it’s still an issue. And so we have 7 or 8 trials now with treating people with first attack, and a couple of lesions at a minimum that look like MS, and they’ve all said that if you treat earlier, people do better. But there’s still some resistance out there—both here and certainly in Europe. Thoughts? 

: Well, I think one of the interesting things about being a neurologist is that people are all different, and they have different behaviors and thought patterns. And so I think as a neurologist, you need to work with people and make sure that they’re comfortable and onboard with you because it’s not that we can force people to take these treatments, and they have to understand the need for it. 

There are some people who come in and get it, and they want to go on treatment immediately. There are some people who actually want to go for stem cells in another country immediately. We have to talk them down. And then there are other people who want to try naturopathic remedies, or a diet, and disappear for a couple of years and come back, and we don’t want to lose those people, because they accrue disability. 

But for most of my patients I think they’re pretty well educated. There’s a lot of material on the internet, and they understand the need for treatment. The challenge is digesting all the information and coming to a treatment decision where they’re educated and the expectations are appropriate. 

: I was thinking of colleagues, of the physicians, who have been reticent to prescribe to treat. You treat patients with clinically isolated syndrome, right?  

: We do. I think all is predicated of course on wanting to get the correct diagnosis. We talked earlier about the importance of understanding and addressing a differential. And 1 approach when you’re not sure, is to follow very closely without yet having initiated treatment, and MS will declare itself if it is MS. And if it is not declaring itself, it could still be MS, but MS that is so low in activity that it may not warrant treatment. So this is a discussion, as Peter says, I think on an individual patient and clinician basis. There are those who will still opt to start with the therapy that may be the mildest but sufficient for mild MS versus the ones who will consider treating with a more proactive treatment up front. 

: See I think we can look at rheumatoid arthritis as a model where you have very good data that if you don’t start treatment within 3 to 6 months of presentation, several years later you have permanent joint and bone damage. There are very good data. 

We increasingly see support for a window of opportunity in treating early. Virtually every study shows the early treatment group does better. If you think of the paradigm of MS as accumulating permanent damage to the central nervous system, you want to intervene as quickly as possible. In addition, when it presents, there may be a self-perpetuating component with epitope spread from an immunological point of view that you can control by effectively treating early. I think you can make an overwhelmingly supportive case on the importance of treating early. The big MS database showed long-term disability. You had to treat in the first 6 months after presentation to see a significant benefit over 10 years later.  

: I would add that I think we’re finally starting to see the data that was alluded to earlier about the long-term beneficial consequences of early treatments. For many years it was the assumption that we were preventing secondary progressive MS so that these patients would benefit in the long run. But we didn’t have that long-term data yet. Now we’re starting to have the data, roughly 20-plus years into the MS treatment era. Now we’re seeing 10-year, 15-year, 20-year follow-ups of treated cohorts and it looks different. And I think that that will probably help to bring stragglers to the idea of treating MS early and promptly at the time of CIS, or they’ll just retire. 

: One of the problems that makes MS different from rheumatoid arthritis in trying to follow this is that the risk of residual deficit after any attack is 50%, maybe a little more. And that’s completely unpredictable, and that’s what makes the conversation so difficult. Patients say, “Well, you may never have an attack again the rest of your life, or you may have an attack next week.” And that attack could leave you with residual. And that’s the part that worries me the most, you know? That if I wait too long, that they’ll have an attack that leaves them with a significant disability. 

Importance of Early Treatment of MS

: Okay, now, choosing therapy. Thoughts?  

: Well, I think if there’s mostly 1 debate about whether we should treat early or not, I think there’s a much more active and unreconciled debate about how best to do that. And I think that one can really have reasonable arguments about using the medicines that we think of as our high efficacy therapies early, reserving those for patients who show any early sign of breakthrough disease, or reserving those medicines altogether for people who we are worried about prognostically.  

And to the same degree, that lack of knowledge of how someone is going to do in the short term, I think we also have imperfect science telling us how patients are going to do in the long term. And that drives the fact that there is no consensus in our field about how to best choose medicines early for the patient. Although we all have our own strategy, which I think we’ll probably outline. 

I’ll say this, that I remember when natalizumab first launched, which we’ll talk about in a little while in some detail. But I feel that that launch of that medicine really made it very clear to me that physicians and patients alike were going to construe that risk of benefit data very differently. I’m sure we all had experiences of talking to patients about that at the time. The notion, now revised, that this medicine might be roughly twice as good from an efficacy perspective as the injectables that preceded it, but it had a 1 in 1000 risk of a potentially fatal adverse effects. 

And I remember saying those words to a group of patients at an educational program and immediately half of them said, “One in 1000? That’s never going to happen, sign me up.” And the other half said, “One in 1000? I don’t want to ever hear about this again.” And right there it sort of lays bare the fact that even with the same data, people are going to decide very different things, and that’s only going to become more complicated as we have a dozen more medicines since then. 

: That’s very important, there’s no 1-size-fits-all. And I think that to me, although I like to think that the proactive approach and try to minimize damage is the way to go— that’s a personal feeling. And I’m very comfortable following people with MS [multiple sclerosis] who don’t want that approach. Then the discussion is simply to follow them proactively enough that if there is new disease emerging, we’ll catch it early and consider a switch. 

: It’s worth mentioning that there are 2 trials going on right now, TREAT-MS and DELIVER-MS, to try to answer this question, randomizing to first-line, or high efficacy therapy. We of course have some prognostic measurements that have been around for 20 years. Poor recovery from first relapse, spinal cord disease, high T2 lesion volume on MRI [magnetic resonance imaging]. But we need to do better, and I think hopefully with things like neurofilament light, perhaps we’ll be able to better prognosticate as to who really needs to get those highest disease therapies right out of the gate, and who we can do escalation on, and have a little bit more time before we’re worried about them accruing disability. 

: You’ve done very elegant work with neurofilament light in a variety of contexts looking at that. What I struggle with is knowing how we’re going to take those data that really are a very useful prognostic tool from a population basis and actually pull it down to use it at the point of care. Do you have a thought on that, because I would love to know your thinking?  

: That’s why it’s not ready to be commercialized, in my opinion. We’re now studying this longitudinally in a large cohort in the context of the MS PATHS study with 5000 patients over the next 5 years. And I think that the need is to know what value is actionable in an individual patient, and we don’t have those data yet.  

: I just want to put in a word. I think treatment is critically important. I think every MS appropriate patient who presents should be on a treatment. And the multiple options that we have, if somebody has concerns about risk, there are very low-risk agents. So we’re matching in shared decision making: disease factors, drug factors, patient factors. But I think we should be convincing our patients to go on treatment, I really believe that.  

: We used to talk about looking for poor prognostic signs to figure out which patients merited higher efficacy therapy. I think where our field has shifted is the idea that most patients require or warrant a pretty high efficacy therapy. And I think what we really need moving forward are prognostic markers for more benign disease. Who among our patients is actually not destined to require the kinds of treatments that we’re using much more proactively now? And I don’t think we have that yet. I would love to know who’s going to do well, because I think we have a sense of who’s not going to do well, and we have a sense of what kinds of treatment strategies, high efficacy medicines they need.  

: A good point, we may be overtreating some patients. But the problem with the very mild MS is that’s a retrospective diagnosis looking back 30 years. 

:You see someone in their 60s who hasn’t had recognized MS, and you look at him and you say, “Yes, this is definitely MS. Your history fits, your MRI fits.” Do you still put them on treatment at that point? 

: I might not put them on treatment. That might be your example of benign MS, looking back.  

: We caution people to use that term very carefully, or better yet, not at all. 

: Well, because the literature is awful.  

: You need at least 3 decades of observation. 

: Wouldn’t it be great if we had a prospective prognostic sign that could tell us who’s going to have that path? I feel like that’s what we need almost more than our worrisome prognostic signs, which we all recognize. 

: We do recognize that inflammatory lesions on MRI have different outcomes, and some become black holes and others don’t. I think that’s the first pass at this, but with future biomarkers hopefully we’ll be able to monitor that more closely. 

Choosing Initial Therapy for Treatment of MS

: With this issue of how you choose your first therapy, we get asked fairly frequently by pharmacy benefit providers, “Give us an algorithm,” because that lets them know who to go negotiate with first. You know, start with A, then B. There are some disease states where you can do that. I don’t think you can do that in MS [multiple sclerosis]. I think you well know the conversation on choosing a therapy is about an hour long.  

And the American Academy of Neurology in their guidelines, one of the wisest things they put in there was that you should do the diagnosis at one visit, and the discussion of the therapy in the next visit because they’re both very complicated, and after you do the one, the patient may not be in a position to really choose.  

But there are so many different factors. There’s your assessment of how active their disease is and what their prognosis is, as difficult as that is because of our lack of ability; their risk-taking capacity; what they’ve learned from the outside; where they are; pregnancy. We do the second most family planning in our institution after the OB/GYNs [obstetricians and gynecologists]. You ought to get pictures; we make them bring the babies in. But it’s a big part of what we do because very often someone will come in, a young woman, that’s our model patient, and there will be a decision, “Well we’re going to put you on this medication—are you going to try and start your family?” 

Then there are insurance issues and what route of administration they want to take. And so there’s quite a bit involved, and it’s a long conversation. At least my view is we don’t have this kind of algorithm. Thoughts? 

: Yes. One of the questions that we’re asked to discuss is, what therapies to use as first-line treatment. And I think I’ll turn that around and say, what therapies wouldn’t you use as first-line treatment? Amit? 

: Other than alemtuzumab, for bone marrow transplantation, as an unapproved therapy, I think all the therapies in the armamentarium could be considered for first-line therapy.  

: I would agree with that. Recognizing that higher efficacy medicines probably work best in early highly inflammatory disease, I still think very few patients up front merit the risks of those strategies. And so I would use, or at the very least discuss, all of the others. 

: We’re all in agreement. And I think that what we’re seeing out there, as we get more and more agents along, certainly more of what we think are higher efficacy agents, we’re seeing their uptake faster and sooner. And we are sometimes even not following the label per se, there are all these different labels; our decisions aren’t necessarily based on the data but rather on some opinions. 

Factors to Consider for First-Line Therapy for MS

: Cladribine has had an interesting developmental pathway. They did 2 phase 3 studies some time ago—early in the 2000s&mdash;called ORACLE MS and CLARITY. One was in clinically isolated syndrome, and the other was in relapsing-remitting patients. They were fairly typical studies of the time, but the drug didn't get approved because of safety concerns, especially around malignancies.  

However, additional data were obtained. Not clinical trial data, but additional safety data were obtained. The company reapplied for approval based on additional safety data and these 2 older studies, which succeeded against placebo in decreasing relapse rate. This got approved. The approval mentions in the label that there is a risk of malignancy, but it was felt to be an acceptable risk compared to the benefit. The drug is conveniently dosed. It is dosed orally for just 2 weeks annually. You don't have to take it again. Interestingly enough, there are some long-term data that suggest a prolonged effect, even after you stop the drug. Another interesting aspect: Even though it has effects primarily on lymphocytes, it didn't cause a lot of significant lymphopenia. 

In the labeling for cladribine, even though they did a clinically isolated syndrome study, which was ORACLE MS, they did not receive labeling from the FDA to treat clinically isolated syndrome. I think there were reasons for this. They didn't think the safety justified that. 

: One long-standing question about fingolimod has been whether a lower dose could be similarly effective and have lower risks. The lowest dose that was studied in the original phase 3 trials was 0.5 mg. Although it was found to be effective, it had some safety concerns. It has left people wondering whether a lower dose could be almost or just as effective and have a better safety profile.  

The ASSESS study compared the standard 0.5-mg dose of fingolimod to a lower dose—0.25 mg&mdash;and an injectable therapy, glatiramer acetate. The trial was double-blinded, so patients didn't know what therapy they would get, and the outcomes were typical of MS [multiple sclerosis] trials looking at annualized relapse rates and new lesions on MRI [magnetic resonance imaging]. Not surprisingly, the 0.5-mg dose was found to be more effective than glatiramer acetate both in annualized relapse rate and in reducing new lesions on MRI.  

The 0.25-mg dose was somewhere in between. The 0.25-mg dose wasn't more effective than glatiramer acetate, but it was numerically somewhere between the 0.5 mg dose results and the glatiramer acetate results. What this suggests is that 0.5 mg of fingolimod is probably the ideal dose, and that going to a lower dose probably, although we're not sure of this, does lower the efficacy. 

: We've been using dimethyl fumarate since 2013. It's been an effective oral therapy. One of the main adverse effects of it is GI [gastrointestinal] upset, including abdominal pain, nausea, vomiting, and diarrhea. More recently, diroximel fumarate was tested. This drug behaves a little differently in the GI tract, but it's converted to monomethyl fumarate just like dimethyl fumarate is. So, the mechanism of action on MS would be the same. When they did the study of diroximel fumarate, there were fewer GI complications seen than in patients taking dimethyl fumarate. 

 Recent Treatment Strategies Explored for Relapsing MS

: Let’s talk about some individual agents that we have available, and we’ll start with fingolimod. Stephen? 

: Sure. Fingolimod—2020 is the 10th year anniversary of when fingolimod was approved for relapsing MS [multiple sclerosis] in 2010. We’ve had the oral treatment era for MS or the second-generation MS treatment for about a decade now. Interestingly, fingolimod was studied at decreasing doses in each of its early trials. And so when it arrived at its phase III studies, it was studied at 0.5 mg pill once a day, which was the lowest dose that had been studied up to that point. And it succeeded in its trials in relapsing MS, both against placebo and versus an active comparator of injectable interferon. 

But one of the things that I think Novartis AG was tasked with doing was to see whether a lower dose could be utilized and get a similar degree of benefit. About a year-and-a-half ago, we finally saw data from a trial called ASSESS, which was looking at the standard approved dose of 0.5 mg a day versus a half dose, 0.25 mg a day, versus injectable glatiramer acetate, which was the active comparator that was selected at the time.  

This study showed a couple of things. One was that the standard 0.5 mg per day dose of fingolimod was in fact superior to injectable glatiramer, so that was the first head-to-head trial against glatiramer of this agent that showed superiority. But it also showed that the 0.25 mg dose was less effective than the 0.5 mg dose. It was marginally better than glatiramer but not statistically significantly so, which is interesting because it means that we actually arrived at the right dose of the drug when it was studied and put through in its phase III trials. 

I do think that it also has a little bit of downstream consequence, at least to how I think about it. In the studies, fingolimod as an S1P [sphingosine 1-phosphate receptor] modulator works by sequestering lymphocytes and probably through other mechanisms as well. But we know that the circulating lymphocyte numbers on someone on fingolimod will be low. In the trials, they held the drug if the level went below 0.2 mg. And so in standard practice I think people deviate from the standard recommendation by giving the drug every other day or less frequently if the lymphocyte count gets too low. 

This trial suggests that dropping the effective dose of the drug may actually undermine its efficacy. That to me was a late consequence of interest of that study. The other thing I think worth saying about fingolimod that we’ve learned in the modern era gets at the idea of discontinuation. I think it’s interesting that with our longer-acting durable mechanism agents like anti-CD20s, we can think about stopping them and looking at long-term durable benefit.  

But with fingolimod dosed every day, stopping that agent, and probably this applies to the whole class of S1P drugs, runs the risk of rebound disease activity. Because unlike these other agents like the CD20s, the S1P modulators are effectively hiding the lymphocytes but not actually remodeling them. And so letting that drug wash out of the system can re-release into the circulation a lot of the bad actors, if you will, that can potentiate MS disease activity. So we have to be very cautious about not stopping fingolimod even while we’re considering the potential benefits of stopping some of the other agents. 

ASSESS Clinical Trial of Fingolimod for Treatment of MS

That gets into sequestration more than it does half-life, actually. Although we’ll have to wait and see because some of the S1P1s [sphingosine 1-phosphate receptor 1s] that we’re going to talk about have a shorter half-life.

We don’t know the answer with those yet. But the sequestration and then having those cells readily available is an issue, and it’s led to changing the label. We have it for natalizumab and we have it for fingolimod that there is the potential for rebound, which does impact how you use it and dose it. So I don’t start either one of those agents on someone where I think I’m going to have to stop it at some point. For example, planning pregnancies would be one. When someone gets into a safety issue, you really have your hands tied. But I think one of the lessons, and I’d be interested in your opinions on this, that I think there is for natalizumab, where we first saw this idea of rebound, is to start your next agent sooner rather than later. That the risk of double immunomodulation is much less than the risk of a serious flare-up. I see heads nodding. Anyone disagree with that?

No. We’ve shortened the interval for all the reasons you mentioned. We were concerned about infections and not overlapping drugs, and many times would wait 6 or 8 weeks after coming off natalizumab. And we’ve gradually shortened that to almost keeping them right in cycle if it’s a monoclonal antibody or starting an oral agent almost immediately, presuming that you’re not switching them for fear of them having PML [progressive multifocal leukoencephalopathy]. And I think that’s the 1 caveat, and there have been some carryover cases. You have to be as sure as you can be that they’re not infected already.

We would start the next drug 4 weeks after the last natalizumab.

We were saying, the next cycle. I think we would do it the same way.

The other complicating factor there with fingolimod specifically, just to finish on that particular agent, is how to transition from stopping fingolimod to starting an anti-CD20. Because here you’ve got a circumstance where you have to wait some period of time for those lymphocytes to get back into circulation for the anti-CD20 agent to exert its desired B-cell suppressing effect. But you don’t want to wait too long for them to potentially cause rebound activity. I think that’s a real challenge that I wish we had data to help guide what that timing and the pharmacodynamics of that ought to look like.

The sequestered lymphocytes, are they going to get the effect? Will they show up after the half-life of the monoclonal antibody?

It’s a very fair concern. We do know that on the average within a few weeks, people will bring back lymphocyte counts to at least approaching lower limit normal but not necessarily their baseline levels. And there are a few patients, especially after it seems long treatment courses, who will reconstitute rather slowly. And we believe that the anti-CD20, which especially administered intravenously, will have very high levels in the blood. Circulating B-cells will be the ones targeted. They go to the reticuloendothelial system. They’re knocked off. The depth of penetration and the presence of the mechanisms, the molecular mechanisms, that would be associated with killing, be it complement-mediated or antibody-dependent, differs depending on the tissue you’re in and the depths within lymphoid organs. And so trapped cells will not have as much access of the antibody, and it’s a fair consideration.

To follow up on that point, there’s now evidence that some CD4s and CD8s can have CD20. And so if you come in with a lymphopenic patient and further deplete them, we’ve seen some patients start out around 800 mg where you felt comfortable, and then after you thought the drug had washed out, Ocrevus drops them down to 500 for a period. And so I think we have to be very cautious. And it’s not everyone who seems to have the T-cell targeting, so we’ve taken to looking at T-cell subsets before we go in with therapy to try to further characterize what the profile is of that patient.

It’s interesting as to whether those T-cells that express CD20, one of the thoughts is that they may be T-cells that have been repeatedly stimulated, chronically activated, and they may or may not be as susceptible in terms of recirculation through the lymphoid system to sequestration. So they may represent a higher proportion than usual. Did you see any results to suggest that?

Rebound Management Strategies for MS Therapies

: Fingolimod is a modulator of the sphingosine-1 phosphate [S1P] receptor, and it modulates S1P 1, 3, 4, and 5. We have a next-generation [S1P receptor modulator] coming in now, which is more specific, S1P1 with a little bit of 5 and sometimes 3, but more specific. And the first one that’s been approved is called siponimod. You want to talk a little bit about where we are with that? 

: Yes. It is one of the more selective S1P receptor 1 and 5 modulators. It was studied in the phase III EXPAND trial, which was a secondary progressive MS [multiple sclerosis] trial of siponimod versus placebo. The patient population was a pretty typical secondary progressive MS population. This was an event-driven trial, so an interesting design, not fixed-time duration for each patient but basically calculating in advance the number of events that one would need to achieve, assuming a certain difference between the active versus the placebo. And once that number of events was achieved, basically the trial was stopped and analyzed.  

Siponimod was superior to placebo in the primary outcome measure, which was confirmed disability progression at 12 weeks, a modest but statistically significant benefit, and also impacted favorably multiple MRI [magnetic resonance imaging] measures of focal inflammatory disease activity, what one thinks of as relapse biology. And you pointed out the fairly broad label, which I think reflects the FDA’s recognition that the relapse biology is one that exists throughout the spectrum or at least a broad range of the MS spectrum, starting from CIS [clinically isolated syndrome] and through to patients with active SPMS [secondary progressive multiple sclerosis]. Although I agree with the concerns you have about that less than clear definition of activity of SPMS. 

: Also it’s a little unusual here to have a drug that was tested in a very different population, just what you said, a really well characterized secondary progressive population. But average, mean EDSS [Expanded Disability Status Scale] I think was 5. 

: And it was a very different population. So I’m not quite sure what it meant to cut out the group that, they used different activity. But one could look at activity in multiple different ways. They looked at it because it was part of the study as individuals who had no clinical activity for the 2 years prior. They didn’t look at radiographic activity, and they didn’t look at people who had activity on study, which was a different group than the group that had activity going in 2 years. So I’m just not sure what you do with that information as opposed to taking the study as a whole. 

: You’re referring to the FDA didn’t look at that. 

: Yes, because the study of course did look at that. One of the questions going in, preplanned analysis, was is the effect, the benefit of siponimod versus placebo on limiting progression of disability driven by the effect on relapse biology, and accumulating disabilities that’s relapse driven? And so they looked at, leading into the study, the degree of disease activity, as well as on study. And in both cases, although one gets into subgroups and decreases the power, at least the point estimates all seem to be favoring the siponimod, including in the groups that didn’t have overt clinical relapse activity. A caveat, of course, is that as you pointed out, activity measured by relapses only is not going to sensitively pick up subclinical relapse disease biology that of course could be impacted and would be expected to be impacted by this class of drugs. 

: It also troubles me in principle that this trial was designed to look at secondary progressive MS. It was a successful trial. It met its primary end point, and yet the FDA didn’t approve it in that way and instead cut up the data and approved it for subgroups, including groups in which it was never studied, like CIS [clinically isolated syndrome]. We can assume that the drug is likely to be effective there, but I do think it’s hard to sit down with a patient and explain to her why siponimod is an option for her at the time of her CIS when you look at the data, and it was just studied at the entire opposite end of the disease spectrum. I think that poses a real conundrum for us, and it concerns me that it might be a disincentive for other companies to pursue progressive MS trials if even a successful clinical trial doesn’t lead to regulatory approval. 

: Maybe. As you know, all the other companies were invited by the FDA to extend their labels into the earlier disease spectrum. I think that it is a conundrum and particularly this issue of an SPMS population. It’s not approved for SPMS without obvious activity, although a substantial element of the benefit appeared to be in those who did have that relapse biology. And in some ways, we’ve been encouraging the FDA to accept our view that there’s all this subclinical relapse biology and that it is present throughout the spectrum of MS.  

: I totally agree. I think it was a real step forward recognizing some of the things that you published, Fred, and if you look at the population, the average age was in the late 40s. These are people who had secondary progressive MS or 13 years since duration of diagnosis and 3 years since they’ve had SPMS. These are not the vast majority of patients that I’m seeing with progressive MS, who are late 50s and 60s. And I worry about whether the risk-benefit is necessarily in favor of treating all of those patients. 

: However, the label may be a blessing in disguise because the S1P receptors penetrate into the CNS [central nervous system], so they have the possibility to have a direct effect on neurodegeneration. If siponimod has a direct effect on neurodegeneration, why wouldn’t it be the S1P receptor to use in CIS and early relapsing MS, where there is neurodegeneration? That would be very logical to me. 

: At least on the one hand. It doesn’t say what you do with the secondary progressors without activity.  

: They should have gotten a label for SPMS, period, not active versus nonactive, I think, based on their trial. 

: Or at least include imaging metrics of activity. 

: I think it was a very successful trial in the sense of the recruitment population. They really did get a group. If you think of that, you said 3 years after declaring, but it takes 3 to 4 to 5 years before you really declare someone secondary progressive MS. I’ve always thought that’s one of the big challenges in why others have failed in secondary progressive MS because it’s supposed to, even the primary progressive MS, they have this 13-, 14-, 15-year duration of disease. So a lot is going on that we’re not impacting very well.  

EXPAND Clinical Trial of Siponimod for Treatment of MS

: Let’s move on to alemtuzumab. We talked a little bit about this, but Amit, why don’t you take it up. 

: Alemtuzumab is an anti-CD52, a monoclonal antibody. It depletes fairly broad swaths of immune cells, including B cells and T cells. It’s highly effective at limiting relapse biology, and we did talk a little bit before about the potential for it also to allow you to get off therapy after a couple of cycles, to benefit from a durable control of your disease activity. It has been associated with a rather high frequency of secondary autoimmunity, largely thyroid but not only thyroid. And it’s because of the less common secondary autoimmunity complications that can include life-threatening issues, including platelet disorder or dysfunction or kidney problems, that one needs to monitor this treatment with monthly blood draws for 5 years. So that represents a fairly substantial burden. 

And in the last few years, we’ve seen another family, if you like, of complications that seem to be vessel related with a mechanism that is not entirely clear, but this is including dissections, and strokes, and others, which have been associated with fairly high morbidity and even mortality. So that has taken away some of the enthusiasm over using it, at least early on, but it is still available in the armamentarium for patients who haven’t responded sufficiently to other treatments. 

: Do you think it has an advantage over either CD20 inhibitors or natalizumab? 

: It depends what you mean by advantage. From an efficacy standpoint, it has clearly very high efficacy and without a head-to-head, it’s going to be difficult to see whether one is superior to the other. But in terms of the burden, the baggage, and the risks associated at least with the first epoch of treatment, the anti-CD20s and natalizumab in people who aren’t infected with a JC [John Cunningham] virus, the risks are really quite low, at least for several years of treatment. 

: That may get into the group that you were talking about, the 1 to 1000, that there may be some who say, as you alluded to, “Well, if I take this for a couple of years, maybe I could stop.” 

: They have very interesting data out to 10 years that indicate maybe more than 50% have normalized their brain volume, and really lowered their neurofilament light protein. That’s intriguing to have treatment just in your first 2 years and then 8, 9, 10 years later, apparently normal brain volume loss and normal neurofilament light protein, if that’s actual. But they have presented some data that would be intriguing if they are accurate. 

Patient Selection for Alemtuzumab for Treatment of MS

: How about cladribine, another new one? 

: Well, cladribine was due to be the first oral agent, and then they raised concerns about malignancy, and they decided to abandon it. They continued to follow the patients who were studied in the 2 trials that they had, and they wound up finding out after a number of years there wasn’t an increased malignancy rate. So it was approved more than a year-and-a-half ago in Europe, and then it was approved in March of 2019 here in the United States. So this is a prodrug that has durability. It’s a so-called induction agent, as I’ve used it, it’s a durable agent. You knock out lymphocytes, T and B cells for a prolonged period. You’re treated for 5 days in month 1, 5 days in month 2, then you don’t treat again for a year, then 5 days in month 1, 5 days in month 2. Then you don’t treat for the rest of that year or years 3 or 4.  

There were 2 trials. CLARITY was the major phase 3 trial that looked at 2 different doses. They wound up going with the lower dose, 3.5 mg/kg, against placebo and showed very good efficacy in suppressing relapses, disability, and on MRI [magnetic resonance imaging]. ORACLE was their CIS [clinically isolated syndrome] trial, which was aborted early, but again it was set up the same way, CIS with placebo 3.5 or 5.25 mg/kg. They abandoned the higher dose. That was a fairly positive study in the data that they could analyze. 

This is an interesting agent because the efficacy is probably in between the orals and the monoclonals. I think cladribine exists somewhere in between. It seems to be very well tolerated. And because you have such durability, you can talk about getting pregnant in between these doses and maybe having at least 3 years durability after you take the second course cycle. This is a very interesting agent, new oral agent for us. 

: They have some intriguing data that in the second or third year, I forget which now, in CLARITY I guess it was, that the group that got continued dosing and the group that was on placebo did the same, which gets to this issue of durability. 

: Right, exactly. It looked like for most patients you didn’t need any treatment in year 3 and 4 if you went through the 3.5 mg/kg in the first 2 years.  

: And the other is the lymphocyte levels don’t really get all that low here like some of the other ones. 

: It seems to be pretty well tolerated. Of course, off-label use of parenteral cladribine has been done by some limited MS [multiple sclerosis] neurologists for a number of years. 

: Diroximel fumarate is a new FDA approved drug that’s a next-generation fumarate that basically has less GI [gastrointestinal] toxicity than Tecfidera, so that’s good. It doesn’t seem to get around the issues of flushing adverse effects, and the incidence of lymphopenia is really the same in Tecfidera. Interestingly, because it’s basically very quickly metabolized into monomethyl fumarate in the gut, they think that that causes less GI toxicity, and it’s thought that monomethyl fumarate is the active metabolite that’s conferring the efficacy of Tecfidera. So there was no study done to actually compare it versus placebo or do a phase 3 long-term efficacy. This is basically the same compound as Tecfidera. There were studies comparing the adverse effect profile between dimethyl fumarate and this diroximel fumarate. And as I said, the GI toxicity was significantly reduced, so that’s nice because that’s a big problem I think with dimethyl fumarate. Hopefully, that will be borne out over the long term. 

Recent Approvals for MS: Cladribine and Diroximel Fumarate

: Natalizumab I think was the archetypal high efficacy medicine, and it really dragged us as a field into the modern era of being able to balance risk and benefit. This is a monoclonal antibody. It’s infused classically every 4 weeks. We’re going to talk a little bit about how that might be evolving. It blocks the transmigration of lymphocytes across the blood-brain barrier. So, again, we can think of this as a cell trafficking modulator. So, although the S1Ps [sphingosine 1-phosphate receptor modulators] do that through a very different mechanism, the principle that these are blocking the movement of the lymphocytes of concern is similar with these 2 modalities. I think that’s why the idea of stopping an S1P modulator and stopping natalizumab both are fraught with the concern for disease rebound as those cells are able to cross into their target. 

A couple of things to say about natalizumab. As we talked about before, this is the one where we thought that there was a 1 in 1000 risk across the board of PML [progressive multifocal leukoencephalopathy], an opportunistic viral infection of the brain caused by the JC [John Cunningham] virus. But this is also I think the best example of using a couple of simple biomarkers to help us optimize a risk-benefit ratio of a particular medicine for patients. We know that the risk of PML increases the longer a patient is on natalizumab if they are JC virus antibody-positive.  

The advent and the dissemination of the JC antibody blood test I think is the single most useful biomarker that we have in some ways. Because simply knowing how long has someone been on the drug, are they JC antibody-positive, and were they exposed to immunosuppressive or chemotherapeutic agents in the past allows us to discriminate PML risk 100-fold, from perhaps a 1 in 100 risk for the highest risk category, to perhaps a 1 in 10,000 risk or lower for those in the lowest category, those who were JC virus antibody-negative. That’s I think incredibly useful, and it really guides the conversation with patients about who’s suitable for this medicine and for whom it could be used early, used first-line in the right circumstances we were talking about earlier. 

Now interestingly, natalizumab was studied in secondary progressive MS [multiple sclerosis] in the ASCEND trial, and that was a negative study. So here you’ve got another cohort of secondary progressive MS patients. The outcome measure is accumulation of disability tested in a variety of ways with natalizumab versus placebo, and it was negative. It showed statistically significant benefit on a subcomponent of disability, the function in the upper extremity. That was a negative study.  

I think it’s important that this unequivocally high efficacy medicine for disease activity was not successful in secondary progressive MS. And yet with the new FDA categories, it too garnered this active SPMS [secondary progressive multiple sclerosis] indication. And siponimod, which was successful in secondary progressive MS, garnered the same indication. And I think that’s another example of how we’re trying to retrofit the way the FDA has now categorized things on to the data that we have. 

The last thing I’ll say about natalizumab is we’ve talked a lot about PML risk stratification on the basis of patient selection. But now where it’s being looked at in our field is whether extending out the interval of the dosing, so in essence, underdosing the medicine over time, can reduce the risk of PML. And some early data with a lot of work presented by Lana Zhovtis Ryerson, MD, at NYU [New York University Langone Health], and collaborative groups, have suggested that extended interval dosings out to the 6-, 7-, 8-week range can meaningfully decrease the risk of PML even in high-risk patients. But I think that there’s real diversity in our field as to whether we believe that and whether that guides our practice. So I’ll just put it out there. That’s what the data look like. I’d be interested to know whether you’ve begun to adopt that or question it. 

: I think it’s certainly plausible, biologically believable. The issue is what to do with that information. And you made the point that for the same 1 in 1000, there were people who said, “Absolutely, yes,” versus “Absolutely no.” So if you change that from 1 in 1000 to 1 in 200, will that really shift the equation dramatically or will it just nudge it a little bit? 

: I think it makes sense though. We’ve known for over a decade that receptor occupancy differs in patients. It’s just a very difficult assay to perform on a clinical basis. And so right now it feels a little uncomfortable to extend the dosing interval on some patients, and not have a way of necessarily monitoring that. We’ve seen some people feel like they’re starting to reactivate. Of course, it’s probably the right thing to do in people who have been on it for a while and are in that low positive category. 

: I think the TOUCH program data were pretty convincing that extended dosing significantly lowered risk of PML. I think it’s a definite strategy that you can do, and I think the data have supported 6 to 8 weeks. I know there’s a big trial looking at 6 weeks extended dosing. You still have efficacy. The drug seems to hang around longer, so I think that’s a valid step to take to lower PML risk if you continue on it.  

: We need a biomarker. There needs to be a blood test to know who’s a good candidate.  

: I’d like to see a prospective study. The TOUCH isn’t prospective, it’s sort of happenstance, so I’d like to see it looked at. I’d like to see it looked at prospectively. But we got at this issue of the label, and we should say cladribine did not get a CIS [clinically isolated syndrome] indication. 

: And this again makes no sense to me. Cladribine had a CIS trial. Cladribine was approved because it doesn’t cause cancer, so why would you say they didn’t give it a CIS approval because they said we’re going to recommend it be used as a second-line agent, that you try another agent beforehand? Totally illogical. 

: But they didn’t say it doesn’t cause cancer because they did give it a cancer warning. 

: Well, yes, but you don’t need to do any special cancer screening. In theory, you can have cancer associated with any immunosuppressive agent. It got back on the market because they didn’t have an obvious cancer issue with it. 

Review of Natalizumab for MS

: Let me mention our 2 original agents, first is interferon. We have 4 different forms of interferon-beta going back to 1993. Having the grayest hair here, I probably have the most people still on interferon because they’ve done fine on it. We have a lot of people come in, and this started in 2010 when we got our first pill, and it’s continued through now. You say, “Well, we have other choices here.” They say, “Yes, I understand, but I’m doing just fine. I hate my injections but I’m doing just fine, and I’m not going to switch.” And they are doing just fine. And so maybe that will be the group that we ought to start looking at who we could discontinue. On the other hand, I don’t remember the last time I wrote a fresh prescription for interferon. Anyone? 

: Exactly what you said. I have a lot of patients who have done really well, the so-called super responders. Maybe their disease has burnt out, but for the most part if they tolerate it well, they can stay on it. If you were to switch over to long-acting Plegridy just for convenience, that seems to be a little bit stronger in some people. 

: The other is glatiramer acetate [GA]. That’s a little different situation. I have a bunch of people who are on glatiramer acetate. Again, doing just fine. Don’t want to mess with success. They’re happy to have gone to 3 times a week, and that’s OK. But I also have a population of individuals who come in when we have our first treatment discussion, they ask 1 of 2 things. One is saying, “How long has that drug been around,” when I start talking about drugs. I say, “Well, this one has been out 3 years, this one 5 years, this one has been around 18 years,” that sort of thing. So I already know that that conversation is going to go somewhere different. And there are people who say, “I want the safest thing you have, and I’m worried about what I’m putting in my body.” And those people receive glatiramer acetate prescriptions. Similar experience? 

: Yes, especially for the young women who are considering getting pregnant in a couple of years, it feels like a reasonable place to start. 

: There’s only 1 DMT [disease modifying therapy] that you don’t need to do any blood monitoring for, that’s GA, and it has the best pregnancy exposure data. So those are 2 big issues. 

: We also want to say something about teriflunomide. Stephen, do you want to say a little something about teriflunomide? 

: Teriflunomide was an oral agent approved about 8 years ago now for relapsing MS [multiple sclerosis] and for clinically isolated syndrome after a clinically isolated syndrome study. This is a modulator of cell proliferation, so it’s thought of as a more modest immunomodulator and has modest efficacy. At the time, it was felt to have efficacy roughly comparable to the injectable agents that were the standard of care at the time. But I think although we don’t have a lot of new data on this agent, at least not that I have seen, it’s been pretty successful in terms of disease control. They also have some long-term data that have been shown from the study of people who maintain stability on it for quite a number of years.  

Unlike with many of our modern medicines, PML [progressive multifocal leukoencephalopathy] has not emerged as a real risk of concern with teriflunomide. And so I think that does distinguish it in some ways from certainly our monoclonals and even from the S1Ps [sphingosine 1-phosphate receptor modulators]. That’s something for patients who have that as a concern, which it’s interesting how well educated and highly aware our patients in MS are. I think people come with very clearly articulated concerns, and that’s one of them. That’s an agent that we’ve had good success with. 

I guess the data that we have for teriflunomide in the most recent era now have been teriflunomide as a comparator in trials of new medicines that are being developed. Teriflunomide was the active comparator in the recent clinical trial of ofatumumab, which is a next-generation B-cell depleter. And it was also the active comparator in a trial of a next-generation S1P, ponesimod. And so in both of those trials, teriflunomide was bested from an efficacy perspective by the anti-CD20 and the S1P. But the efficacy has been borne out as basically comparable to what we saw from those pivotal trials roughly a decade ago. 

Original Drugs to Treat MS

: We have several phase 3 trials that have reported out as successful. Let’s start with ofatumumab. Amit? 

: You’re referring to the ASCLEPIOS I and ASCLEPIOS II trials. As Stephen has mentioned, these were randomized active comparator trials with teriflunomide as the comparator and ofatumumab at 20 mg monthly, subcutaneous injection dose. This is a fully humanized anti-CD20 to distinguish it from rituximab as a chimeric and ocrelizumab as a humanized antibody. The studies were both quite positive with a 50% and 57% reduction in annualized relapse rate as the primary outcome measures in each of these 2 identically designed trials and hitting on multiple imaging metrics, again of focal inflammatory disease as well. 

Advances in Early Treatment Approaches in Multiple Sclerosis