 Multiple sclerosis [MS] is an inflammatory demyelinating progressive disease that affects the central nervous system. It is frequently diagnosed in young adults; however, individuals may be diagnosed at any age. The symptoms of multiple sclerosis vary greatly among individuals. There is no available cure for multiple sclerosis, but treatments are available that can modify the course of the disease.

 discussion, 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’ll review the latest clinical trials, and provide a practical perspective on how the recent data apply to your clinical practice.

I’m Dr. Fred Lublin, Saunders Family professor of neurology at the Icahn School of Medicine at Mount Sinai, and director of the Corinne Goldsmith Dickinson Center for Multiple Sclerosis in New York City. Participating today on our distinguished panel are:

Dr. Patricia K. Coyle, professor and vice chair of Clinical Affairs, and director of the MS Comprehensive Care Center at Stony Brook University in Stony Brook, New York; Dr. Suhayl Dhib-Jalbut, professor and chairman of the Department of Neurology and the Ruth Dunietz Kushner and Michael Jay Serwitz chair in MS at Rutgers New Jersey Medical School and Robert Wood Johnson Medical School in New Brunswick, New Jersey; Dr. Thomas P. Leist, director of the Comprehensive Multiple Sclerosis Center at Thomas Jefferson University in Philadelphia; and Dr. Clyde Markowitz, director of the Multiple Sclerosis Center at the University of Pennsylvania, and associate professor of neurology at the Perelman School of Medicine, University of Pennsylvania, in Philadelphia; and lastly, Dr. James Stankiewicz, clinical director of the Partners Multiple Sclerosis Center, and assistant professor of neurology at Harvard Medical School in Boston, Massachusetts.

Thank you all for joining us, and we’ll begin.

So we’re going to start with diagnosing multiple sclerosis. Clinicians have been diagnosing multiple sclerosis since the late 1800s, after Jean-Martin Charcot described it. The modern era of diagnosis started in 2001 with the first publication, what are called the McDonald criteria. And we’ve had McDonald from 2001, 2005, 2010, and now 2017, and each one has made iterative changes, but the important aspect of it was that McDonald allowed us to combine both clinical and MRI [magnetic resonance imaging] metrics for making the diagnosis. So Pat, why don’t you tell us a bit about what changed with McDonald 2017, which is one of the most referenced papers in all of 2018?

 Well, first of all, they began by reminding us that there’s a misdiagnosis rate of MS, and so you really need to be careful with this diagnosis. They spent a little bit of time talking about 1 particular differential, neuromyelitis optica, NMO spectrum disorder, which I think is in a differential of relapsing MS.

With regard to objective changes, there weren’t a lot. Number 1, symptomatic lesions can now be counted along with asymptomatic lesions for dissemination in space and time. In addition, they recognize cortical lesions to join juxtacortical lesions as an area that would give you a dissemination in space. And what I was particularly pleased with, although they don’t mandate it, is that they really raised the importance of imaging the spinal cord and looking at cerebral spinal fluid, or CSF. They do not insist on that, but they mention how helpful that can be.

And, for example, with regard to CSF, they elevate CSF-specific oligoclonal bands so that you can actually make a diagnosis of MS at the time of CIS [clinically isolated syndrome] in 2 ways. You can meet MRI dissemination in space and time with other disorders ruled out. Or, you can meet dissemination in space and have CSF-specific oligoclonal bands and have other disorders ruled out. So I think they really emphasize, particularly if you are dealing with an unusual population such as progressive from onset, or very young, or very old, or with unusual clinical characteristics, how helpful looking at CSF is. And what they’re really counting now are the oligoclonal bands. It’s CSF-specific oligoclonal bands over IgG [immunoglobulin G] index, or intrathecal IgG production.

So I think these 2017 revised criteria are very helpful. I think neurologists should know them. I think we should be able to say if we’re making the diagnosis of MS, they meet the 2017 diagnostic criteria or not. And I think they’re particularly important to use when we’re making a diagnosis in more unusual forms, such as primary progressive MS.

Diagnosing MS in the Modern Era

 So you mentioned CIS, which is clinically isolated syndrome, the first attack. One of the issues with misdiagnosis is that for those of us working in this era, when we think of that clinically isolated syndrome, we think of specifically an optic neuritis, a brainstem cerebella syndrome, or a partial myelitis.

One of the issues around misdiagnosis is when people try and apply more ephemeral or less characteristic onsets of that first attack when we see things of this sort. So Clyde, tell us a bit about what you expect to see in terms of when the patient presents to you with that first attack?

 Sure. It’s not uncommon for people to present with a unilateral optic neuritis as we describe it—painful eye movements, loss of vision, blurry vision in one eye usually. Patients usually go to the eye doctor for that and maybe ultimately end up coming to a neurologist because somebody said they were concerned it could be an optic neuritis. So they get an MRI [magnetic resonance imaging] scan maybe at the ophthalmologist’s, or maybe at the neurologist’s office And so, having that first MRI scan could be very important in terms of our ability to tell patients what the risk going forward is.

I think one of the things that Pat brings up is very important. When you think about the idea that we have criteria now that says somebody presents with a single event and you have dissemination in time, and you can use spinal fluid to be able to get that information, I’m finding, these days, that I’m doing many more spinal taps than I used to do. I think we went through a period where we weren’t doing as many. But I was talking with some people today in the clinic of the patient I just saw who presented with an optic neuritis who does not meet the criteria. He had the optic neuritis. He has an MRI scan with some lesions. He has a spinal cord lesion but does not have the dissemination in time, other than that he’s got different lesions in different locations. So this is a little problematic, and I said that I think we need to do a spinal tap to really address this issue. And I think it’s very helpful that we now can include that in the diagnostic criteria.

You mentioned, Fred, about a partial transverse myelitis being a very typical—whether it be sensory abnormality in both legs, 1 leg, arms, whatever&mdash;distribution. It could be weakness. It could be bladder involvement. So those are the very common things that you see, typically. But then people can present with significant weakness, and that could be just maybe a hand. And the piece that I think we have the greatest difficulty with is the patients who present with more of a progressive onset. People who don’t have the typical attack that shows up with a defined event over a couple of days that kind of reaches a plateau and then starts to get better.

It’s these patients who have more of a slow progressive myelopathy. In the summer they were having difficulty at the golf course. They went into the club afterwards and they seemed to feel a little bit better and they didn’t really seek the attention of a neurologist. So that’s a little bit more challenging. But the typical signs and symptoms are referable to where this lesion is. As neurologists, we love to be able to localize an area to be able to make a determination. MRI, by far, has really made a huge impact in our ability to find an abnormality and find changes consistent with MS [multiple sclerosis].

 So James, why don’t you tell us a little bit about what the MRI criteria are for space and time?

 So regarding MRI criteria for dissemination in space, we really carve a brain into 4 different regions. There’s the spinal cord, the posterior fossa—so brain stem and cerebellum&mdash;the periventricular area, and then there’s the cortical/juxtacortical area. And so, really to establish dissemination in space, you need to see lesions in 2 of those areas. And dissemination in time, this is something that with the variations or revisions of the McDonald criteria has changed a bit. One of the things that’s more recent is that you can establish dissemination in time based on 1 MRI, where you see that there are some lesions that are enhanced and some that don’t enhance. Whereas if you go back years ago, you would have to have a second MRI showing that there were new lesions a month later. And so, this is an advancement. I think it allows us to diagnose MS earlier.

 Great. And now we can use both the symptomatic and the asymptomatic lesions.

Early Identification and Diagnosis of MS

 Clyde, you mentioned something about progressive disease, and I think it’s probably the form of MS [multiple sclerosis] that’s misdiagnosed the most, in both directions. That is not realizing it’s MS and thinking it’s MS, but it’s actually something else. I think that’s one of the reasons why the McDonald criteria still require a year’s history or observation of progression before you could say that it’s primary progressive MS. Spinal fluid has consistently been a part of that diagnosis, but now, as Pat alluded to, it’s more so. Tom, are you doing more spinal taps now?

 In patients who do not meet the dissemination in time criteria, but from an MRI [magnetic resonance imaging] point of view, certainly. Also, in older patients where very often the differential diagnosis is much larger, a spinal fluid examination can help to look at it. In a young woman aged in the 20s with uncomplicated unilateral optic neuritis, I’m probably still a little bit more reluctant to do a lumbar puncture at that point if I can fulfill the diagnostic criteria in 1 fell swoop. I understand that scenario where different practices practice a little bit differently. Obviously, I also come from the thought process of, what do I need to make a decision to start or not start treatment in a patient such as that one?

If the patient has 2 or more lesions at the time of a clean, crisp clinically isolated syndrome [CIS], then for me that already fulfills the criteria that I need if these lesions are in the previously described telltale locations that I need to make a treatment decision in such a patient. And so, the point of view is, what do I need at the time to adjudicate the future of this patient in terms of going on treatment, not going on treatment? So, yes, there is a little bit more CSF [cerebral spinal fluid], but I’m not necessarily categorically doing it.

 So pretty much what Dr. Leist described. The only issue I have with the spinal fluid is, how many oligoclonal bands are considered positive? Because the reports vary from 1 laboratory to the other, and I feel there is a lack of standardization there. But if you have 5 oligoclonal bands and they’re not in the serum, then that’s safe to say it’s positive. But in the range of 2 to 3, sometimes 4 bands, there’s no standardization in terms of what’s positive and what’s not.

 I tend to consider 1 and more that are absent in the serum to be positive.

 Yes. As Tom said, in a patient with CIS, and I’m looking for dissemination in time. I’m using that. The other issue I have with the criteria is, what if the patient has nonspecific symptoms that could suggest a relapse, such as increasing fatigue, or cognition changes, or depression, and then you find an enhanced lesion on the MRI? Is this a symptomatic Gad [gadolinium]-enhancing lesion or not? So there are these questions that surround the revised criteria that, in my view, need to be addressed.

 So you’ve long been a proponent of the value of analyzing CSF. Has it changed any since McDonald 2017?

 In principle, we spinal tap everybody in a disease where you have no diagnostic biomarker. We also practice in a Lyme-endemic area. We’ve actually found spinal fluid to be very helpful with regard to how certain we can be of the diagnosis, in counseling of the patient, etcetera. So right now we standardly do it in everybody for the diagnosis.

 At the risk of being the dissenting voice here, I would say that I don’t know that with the revision of the McDonald criteria we are doing spinal taps any more frequently. Maybe I am a little bit more permissive, or I don’t know what the right word would be, but I think that in atypical cases where the presentation, the MRI doesn’t look typical, then I think there really is a role. With progressive sort of presentations, I think there is a role to do a spinal fluid. But I also have an accumulated clinical experience, and maybe this is like a fatal hubris, but I tend to think if it looks like a duck and it walks like a duck, that it’s a duck. So I think the criteria are important, but I also think that you have to exercise some clinical judgment if you have an accumulated experience.

 And that’s a very important point. The criteria are really guidelines, right? And Clyde?

 It’s an interesting conversation to have with patients if they want to have it done or not. Because sometimes they’re hearing about a spinal tap being the worst thing in the world, and headaches, and everything else. But I have to be honest that when you really sit down and have that conversation with patients, a lot of times they want that additional confirmation. They say, “If you think I should get it, I’ll get it.” And you know, maybe I do or I don’t. But they kind of feel like that’s more confirmatory. I think it’s a helpful thing for them to be able to know, for sure, even though we say, “The MRI’s good enough,” or “The clinical is good enough.” That extra little piece gives them another piece of reassurance that it’s not another disease phenotype or anything like that.

There’s 1 other piece to this that I think is potentially interesting, and this is just early work that we’ve been trying to address. It is whether band number might be a predictor of severity of the disease. And, at some point, we may be able to say that everybody should get tapped because we need this information to be able to predict how this individual’s scenario is going to go. If they have more than 10 oligoclonal bands, maybe their course is going to be more aggressive and we should be more aggressive about their therapy up front. So this is a piece that’s ongoing, but I think it helps us in our ability to think about how we’re going to manage these people.

 And it’s worthwhile mentioning that there are spinal fluid red flags against the diagnosis of MS. You don’t expect a very high cell count. You don’t expect a very high total protein. You’re accessing the central nervous system compartment. If you’re going to make a diagnosis, a life-long diagnosis, and make treatment decisions, and have the patient behind you that you’re confident, then I think with the lack of a diagnostic biomarker, why wouldn’t you want a robust evaluation of the patient?

 If I may take us back to the other biomarkers that we talked about, the MRI, I think one of the very important things that we perhaps also need to bring to the table is the fact that while the MRI has become important in the diagnosis of multiple sclerosis, the technical quality of the MRI is also important. I would like to refer to the criteria that has been proposed by the consortium of MS centers as minimal criteria, where many of the community-acquired scans still skip sequences where there are areas that are not imaged in the brain.

And I would also put in a little bit of a word to actually review the MRI scans. Because the radiologist may write that this could potentially represent multiple sclerosis, but it isn’t necessarily a diagnosis of multiple sclerosis, perhaps more of a risk management from the radiologist’s point of view. And I would think that everybody sitting at this table probably also reviews the scans of the patients where we make a diagnosis of multiple sclerosis. So I don’t think that we can rely on the radiologist’s report solely in getting to a diagnosis of MS.

 And this is a big cause for a misdiagnosis. We have our weekly case review conference and there’s just that band of white matter between the cortex and the periventricular area that we call the nonspecific zone, and it sort of follows almost like a boomerang-type shape of nonspecific lesions that frequently get sent to us that have nothing to do with MS at all. So over-reading or misreading the MRI is a big cause of difficulty.

Diagnostic Tools Used to Identify MS

 I do want to correct 1 thing. The presence of oligoclonal bands is not a surrogate for a dissemination in time, it’s instead of time. Because there is no time element. And one of the things that I’ve never been able to get an answer to is, when in the course of MS [multiple sclerosis] do the oligoclonal bands appear? And If you start very early, in these very early cases, might they not be there yet? Do you have a concept for that?

 Well, I think they’ve had data. Over time, the oligoclonal band positivity rate gets higher and higher, to 95% plus. And one of the things about MS oligoclonal bands is they do not spontaneously go away, as opposed to other disorders. At the first attack you can have an expectation of close to 80% being positive with a sensitive assay.

 Suhayl, what about neurofilament light [NFL] or any other markers?

 So neurofilament is a skeletal protein that is elevated in the spinal fluid of patients with a variety of neurodegenerative diseases, not just MS. It’s elevated in Alzheimer disease, in stroke, and probably Parkinson disease as well. So in that regard it is not really specific for MS. Studies have shown that it is increased as the disease progresses. And more recently we know that it is also increased as a result of an acute relapse. We also now know that it responds to treatment. A variety of studies have shown, with the different immunomodulatory medications we have now, that the level of NFL goes down as a result of treatment. I think the main advance is that we can now measure it in the serum with more sensitive technique instead of having to do a spinal tap.

I think the big question is, will it be validated as a surrogate marker of damage and disease progression? I think we are not there. It needs to be validated, for example, against atrophy studies by MRI [magnetic resonance imaging], because it has the potential to be used as a surrogate marker in clinical trials. There’s also some data that say that at the onset of MS, if NFL levels are very high, that is a predictor of first prognosis. So maybe a combination of biomarkers such as oligoclonal bands and NFL level at the time of diagnosis, together, might be better a predictor of prognosis.

So I think it has the potential. But, as I said, it needs to be validated against at least MRI studies. And it’s not only neurofilament. Investigators are looking at 

now, GFAP, looking at tau. They are looking at other cytoskeletal proteins such as contactin-1 and contactin-2. So all of these have the potential, either by themselves or in combination, to be correlates of disease progression and prediction of the surrogate.

 When you say disease progression, do you think we actually could distinguish progression from just disease worsening?

 No. At this stage, because we know that it’s increased in acute events and relapses as well as in progressive disease, I don’t think you can distinguish. Let’s say you’re dealing with a progressive population of patients, be it secondary or primary. I don’t think it can tell you whether a patient’s having superimposed relapses, unless you do it fairly frequently. Then you might be able to tell.

 Might it be useful in distinguishing patients who are more demyelinating versus those that have more axonal degeneration?

 Possibly, yes, because as mentioned, neurofilament is an axonal neuronal cytoskeletal protein. So you would expect it to be higher as the result of axonal and neuronal damage than demyelization.

 I was just going to add another piece to this. There are some nonspecific aspects to this, where it goes up in stroke, head trauma, and a variety of other conditions. Particularly in your older population, we may be looking at the marker thinking it’s MS and it may be other causes for damage within the central nervous system. So we’re going to have to figure out that piece.

 It goes up with age. You have to correct it for age.

 And as Suhayl said, there’s no specificity to this.

 I think that’s also one of the potential criticisms—that it will work in clinical trials where we have a relatively healthy population, and where we also introduced a change by introducing a medication, or no medication, or a competitor medication. In a population with a comorbid condition, as was pointed out, perhaps even hypertension in individuals, obviously the picture may be much more cloudy. From that point of view, it may be more difficult in these patients to prognosticate whether it has some meaning. So I think one needs to stay tuned. There are obviously proteins that are of interest in Alzheimer, and in other conditions. And so, from that point of view it is clearly not a disease-specific marker.

Surrogate Markers for MS

 So we discussed how the McDonald criteria has allowed us to diagnose multiple sclerosis [MS] earlier and earlier while maintaining very good specificity, sensitivity, and accuracy. That was only important because we can treat earlier and earlier. The question is, what’s the value? How early should we start, and what are the data that support that? Pat?

 When you have an organ-specific immune mediated disease, there’s ongoing permanent damage that’s accumulating. So it would seem very logical to hit that early to minimize that ongoing permanent damage. You have very good data from rheumatoid arthritis that within 3 to 6 months of onset you need to start rheumatologic therapy or you will have joint and bone damage, permanent damage, several years later. And when you look at the studies of MS, virtually every single study has shown an early treatment group does better than a delayed treatment group. In the big data MS registry that was presented at ECTRIMS [European Committee for Treatment and Research in Multiple Sclerosis] in Berlin, they screened close to 150,000 patients. They looked at about 12,000 MS individuals, CIS [clinically isolated syndrome], and they had more than 10-year follow-up. And they looked at a 6-month treatment with a disease-modifying therapy [DMT] from the time of CIS out to 5 years, and they looked at what had impact on a decreasing confirmed 12-month disability. The only statistically significant impact was starting the DMT within 6 months of the clinically isolated syndrome. And we have a recent MS-based registry of 1555 individuals where, again, they looked at conversion from relapsing to secondary-progressive MS.

So if you were on a DMT, significantly less development of secondary-progressive MS. If you were starting on the DMT early as opposed to late, significantly lower proportion going on to secondary-progressive MS. And then if you were started on what was viewed as a high efficacy disease-modifying therapy, you had significantly less transition to secondary-progressive MS. So there’s very strong data saying early treatment may be a key component, and I’m going to say early is within 6 months of the clinically isolated syndrome. That is when the MS individual should be on a disease-modifying therapy.

 Well, that’s an interesting question. You mean treating at the level of radiologically isolated syndrome?

 No, not yet. But you said within 6 months. Why not immediately?

 OK. So Tom described the criteria for which he would start someone with a clinically isolated syndrome. And typically, at 2 lesions that are typical of MS on the MRI [magnetic resonance imaging]. Do you treat CIS?   

 As do we. But I’ve been surprised that there’s still a degree of resistance to this in Europe. I don’t quite understand it because we have 7 studies now using the criteria that you described—a typical CIS and 2 or more lesions on the MRI&mdash;and they’ve all shown what you’re quoting there, that the individuals started at that time are less likely to go on to another attack, and other metrics of worsening, than if they waited. And so, I’m not sure I understand why. Now the issue is, of course, we’re still talking about typical CIS. I don’t know. Do you have a sense how it is in the community or seeing many people who are untreated CIS patients?

 I don’t think that the community has the sense of urgency of treating early. If you’re just thinking about CNS [central nervous system] reserve, and you think about MS as salting CNS reserve, you would want to start treatment as quickly as possible. I think that’s a message that we need to really get out to the community very strongly.

 So to answer your question, Fred, I think the community still is reluctant, to some degree, not universally. I saw somebody in the last couple of weeks who has not been on therapy. CIS clearly would have been somebody I would have suggested right up front. But their doctor said, “We’ll just wait. We’ll continue to monitor you and see how things are going on.” So the data are there to say treating CIS is key, but still, there’s some reluctance.

 I think in general that even in the community I’m seeing that people are treating CIS. But I’m sure that it really varies.

 I think one of the very helpful things, which is an outgrowth of the optic neuritis treatment trial from the 1990s, is that almost every ophthalmologist now knows to send the patient with optic neuritis for an MRI. And presumably with the result of that MRI, to a neurologist. And I think that that’s been very useful. I think it gets a little more complicated because very often, sadly, they’re confused for peripheral neuropathies when clearly someone who rises up to here is not a peripheral neuropathy. And the same thing happens with brain stem cerebella. These cases are sometimes not recognized in the same way. So it turns out that the optic neuritis cases are the ones that are best recognized and sent to us.

 But I do think that, to come back to Pat’s point, obviously regarding early start, I do think that we know from even the early clinically isolated syndrome trials of the older medications, the ones that we have had for a long period of time, particularly the agents that have been studied in CIS, in relapsing/remitting disease, and then perhaps also were started in more advanced patients, secondary-progressive patients, we see a clear decrement of the effectiveness, or the relative reduction of events. We have also seen in the CIS trials, particularly the more recent ones, an attenuation of McDonald MS 2010, 2005 criteria. So this concept that there is a window of opportunity I think is certainly one that is important to consider.

 This is a very good point because the magnitude of effect was greatest in the CIS patients, more so than the relapsed/remitting, and certainly more so than any of the progressives.

Early Initiation of DMTs in MS

 So when we get into individuals who are meeting your threshold to say “treat,” there are a couple of schools of thought as to what to do next. It’s not just to treat or not to treat, but once the decision to treat is there, at whatever point, and we pretty much all agree earlier is better, then there is this strategy of: do you start with something safer and escalate, or do you hit them up-front hard with something more aggressive and potentially less safe and then consider what to do later? So the one has been called induction, but almost no one is doing true induction therapy, which would be to hit someone with a therapy that knocks out the immune system, and then, down the road come to something safer, right?

So induce an immune halt to the disease and then come up with something long-term, safer. And that’s not really been discussed very much, even though people talk about induction. It really is highly active to start, or escalation with something that is safer and that we know works. And that’s been one of the big debates out there for which there’s precious little data but lots of opinions. So we’re going to get some opinions here as well as to where you stand on this. Let’s start with you, Tom.

 I do think that early on there is a window of opportunity, particularly in patients that present with higher-risk disease—so patients, for example, that have incomplete recovery from an initial attack, or patients that have numerous lesions at the time of the initial presentation. So for patients that have characteristics that are more concerning, I tend to go towards the higher efficacy medications. Naturally, where we speak in the United States, very often my choices are not necessarily level, so I have to deal with the formularies of the insurance companies. And so, there is a compromise between where I would like to go with an individual patient based on the data that I mentioned before and what I can actually do.

 I agree with what Tom said. And in addition to that, I would also take into consideration the location of the lesions, brain stem, spinal cord, bladder dysfunction. If spinal fluid results are available, the number of oligoclonal bands might be a predictor of a worse prognosis. And the third thing is race and ethnicity. African Americans might not respond as well to conventional modest efficacy drugs as the rest of the population. So I take all these factors into consideration.

 Right now I think it’s critically important to treat early, to make sure our patients are following a wellness program, and to make sure that any comorbid conditions are being optimally managed. We do not yet have data that say that when starting a high efficacy agent in everybody, they’re going to do better. We simply don’t have that data. We’re clearly using shared decision-making. I totally agree that if somebody has very active disease and a bad prognostic profile, efficacy becomes important to me, and I’m going to counsel and discuss with the shared decision-making with the individual that it may be very important to pick a high efficacy agent.

Otherwise, if that’s not the case, then I think there are other needs and considerations that you have with the patient, but then also critically important is to follow them closely, particularly for the first 2 years. If there is unacceptable breakthrough activity, you can make a switch.

 I come at this from a little bit of a different perspective, which is that I think a brain MRI [magnetic resonance imaging] is only so good. It’s only so sensitive. And there is a lot underneath the surface that we don’t really see, and we know when we use novel research techniques that we can see ongoing damage that we can’t see otherwise. I also think that any inflammation in the brain potentially can come back and haunt someone, maybe not immediately but may catch up with someone 15, 20 years later. So I think it’s crucially important to prevent inflammation. And I think that we have the data. We know that some of these drugs do better at this than others.

Now it is fair to say that not every patient necessarily will need a stronger agent. Some will get by over the long term doing better on a lower efficacy agent. But I also think that it’s hard to really necessarily predict, actually. When you really look at demographic factors and some of the other things that we use and you study this carefully, I don’t think we have a great barometer about how people are going to do. And I think if you want to wait to see, you may be a little behind the curve already.

And I also would say that I think the profiles of the treatments are changing too. I think we come from a place conceptually where, years ago, the drugs that were more potent had more problems in the way of adverse effects. So, higher risk. And I don’t know that I necessarily think that’s as true anymore. I think that there are certain drugs for which the safety risk, the tolerability, is within an acceptable range so that we can use it sooner, insurance permitting.

 I have been a big proponent of the shared decision-making conversation. Find out what the patient’s concerns are, what my concerns are, and kind of mesh what we think is the right approach. But I have found that I’m becoming much more aggressive earlier than I used to be, and I think that’s driven by the idea that you see patients over years. Even though there’s not really much change on the MRI scan, they are having a slow dwindle in their world. And I wonder whether we should be much more aggressive earlier? There’s some data, not a lot, but some data that suggest that if we really oblate the immune system up front, you can really put the disease in remission, in essence. For how long is questionable. And then ultimately it becomes a safety question, right?

We would all be in this conversation if there weren’t some safety concerns to worry about. And I’m struggling with the idea that I’m balancing safety with what the disease might look like over time. And it’s not an easy place where we live right now. So clearly we need more data to support this aggressive induction approach, like the oncologists do and they basically can dial things back. Or even as rheumatology does, to some degree. So I’ve been moving more into that arena. As James pointed out, you don’t see everything on an MRI scan, and I really think there’s smoldering disease activity that’s happening that we’re not picking up, and we need to do something. The other side to that is, how do we do it safely and not cause any major problems?

Appropriateness for Induction Therapy in MS

 So I’m glad the 2 of you stressed shared decision-making because this is complicated therapy. I’m thinking the American Academy of Neurology wisely recommended, in their treatment guidelines, 2 visits—1 for a diagnosis and then a second visit to go over therapy. Because it’s a long discussion. It’s a very long discussion. I think that a lot of the individuals come in, and they’re pretty well informed and bright, but with a lot of preconceived notions.

For example, we’ve had a lot of people come in, and when I start the therapy discussion they’ll say, “How long has that drug been around?” Which means that they’re more concerned about safety than they are efficacy, and it skews the discussion. We have people who we start on less aggressive therapies because they want something that’s been around for a long time and is safe.

Now, having the grayest hair here, I can tell you that I have people who’ve done perfectly on our earliest therapies. People were started in the early 1990s when interferon was around and nothing happened to them all along. So we know that there are people who do well on every one of the therapies.

What troubles me is that I can’t adequately apply the prognostic factors that Tom listed on an individual basis. On a group basis, it’s pretty clear. But I don’t know about it for the individual. And then, the other part to that is, how do we define a failure or an inadequate response? And we don’t have good definitions for that. The comparative efficacy data that we have, the good comparative efficacy data, are none of the information you are interested in. Right? You don’t care if the latest drug that comes out is better than one of the interferons. That’s not part of your decision-making process. Right? You want to know how this highly effective drug compares to this highly effective drug, because they may have different safety profiles. And I think we’re trying a variety of ways to get at that information, but none of them are particularly satisfying.

 If I may, there is another picture to all of this and it goes back to what Clyde said when he said, “I’m becoming a little bit more concerned early on in patients with my interventions.” Another concept is the fact that in our practices we see the patients that have been doing well over long periods of time. We no longer see the patients that have not been doing well because they are in nursing homes or they are no longer following up with neurologists. So the disease severity, the actual severity of MS [multiple sclerosis], is sometimes underappreciated because patients with more advanced multiple sclerosis deafferent from a neurologist and are no longer seen in these practices.

 And I think another point along these lines is, what is our target of treatment? I would make the argument that our goal really should be to preserve the normal aging process—to keep people as close to normal an aging process as possible, if we’re really doing our jobs, and not to prevent people from hitting an EDSS [expanded disability status scale] of 3 or 2. I think we really want them to be as close to normal aging as possible. And I think my concern is a lot of what Clyde has said: That you see patients that when they hit their 40s and 50s, cognitively they feel slower, they’re having more gallbladder symptoms, more mood symptoms. And I think it’s more than what you would expect just for age, and you see it over and over again, and you feel like you had missed a little bit the opportunity.

 Well, if you’re really going to get that result, you need to make sure they’re not overweight, they’re regularly exercising, they’re not smoking, they’re getting good sleep, they’re really following a wellness program, and they’re getting their hypertension optimally treated. They have to look at the whole body. They have to take care of the whole body if they’re really going to age well.

Newer Therapies in MS: Optimized Care

 Tom, the next drug we expect to come along for relapsed/remitting disease is going to be cladribine, which is approved in Europe. It was tested some time ago but then was brought back to the regulators for reconsideration. Tell us a bit about the data set there and where we might think we’d be using it.

 As you mentioned, oral cladribine has had a relatively long development with some discontinuity during the development. At this point, we have essentially 3, respectively 4, data sets supporting this medication. The first one is a trial in a first event in a CIS [clinically isolated syndrome] population, or CIS/early MS [multiple sclerosis] population. That’s the ORACLE trial. We have the CLARITY study as a second one, and then, very importantly, more recently reported, the CLARITY extension. And we have, because of the duration of the whole development, a relatively robust underlying data set from a follow-up registry that gives all the information regarding the long-term or longer-term safety of the agent.

The ORACLE trial started, as we have previously discussed, at the first event—for patients with a first attack. In this particular study, there was a very significant difference between the patient’s own placebo, respectively, with the patients that were untreated. It also led to a decrease of the patients that actually had McDonald MS by fulfilling next lesions. And I’m talking about the 2005 criteria for the diagnosis of multiple sclerosis.

The second study, the CLARITY and CLARITY extension, is interesting because with cladribine tablets, the patient gets a course of tablets in the first year and a course of tablets in the second year. The 2 courses are 2, 1-week courses set apart by 1 month. That’s the totality of the treatment. At the 2-year mark, this led to a very significant reduction of the relapse rate and also attenuation of the disability progression. Interestingly, in the extension study of the CLARITY study, there were patients attributed to ongoing treatment with cladribine through annual cycles, and there were patients that didn’t get any further treatments in year 3 and 4. The patients in the always-treated group, so 4 annual courses, did comparatively about the same as the patients that had no further treatment in year 3 and year 4 following the treatment. So this raises the question of whether an agent such as cladribine that leads to a targeted transient reduction of B and T cells could lead to a selective immune reconstitution, where essentially the cells that are autoreactive get partially eliminated and then a nonautoreactive immune system will be repopulated.

Where is this agent ultimately going to fit in to our armamentarium? That depends a little bit where it ultimately will be indicated by the FDA. But the data sets that are there will potentially justify a use in early patients based on the ORACLE CIS trial. But it certainly also would indicate patients with activity, with higher than the average activity in the relapsing forms of multiple sclerosis.

Oral Cladribine Therapy for MS

 So an agent such as this, I could see 2 different potential positionings for it. One is as 1 of the highly active agents, such as natalizumab, alemtuzumab, ocrelizumab, or 1 of the orals. It may be top of the orals. What do you think, positioning?

 It’s a difficult question because the efficacy looks robust. The question ultimately is, is this an induction treatment? It has the flavor, to some degree, of depleting out your immune cells and then having a different functioning immune system come back, which is kind of our definition, to some degree, of what an induction treatment would be. So I like that idea, particularly for this compound. Then the other side to that is, what are the safety concerns? How are we going to monitor these patients? What is the next agent if somebody breaks through? Is there any concern for the sequencing of drugs? And that’s really what it’s going to come down to. And so far, based on the data that were available in 2010 and now that we have follow-up extension data, there really wasn’t a huge safety concern that came out. There was a concern for possible malignancies.

You look at all the drugs that we’re using these days and they all have a little bit of that going on. So I don’t know that it’s going to distinguish anything. It may end up being just another really good, high efficacy drug with a fairly benign mechanistic, take a couple of pills for a week, and another month later, take another, and you don’t really have to keep doing that. I think it’s very attractive. Where we will learn more is, what are the concerns if somebody is breaking through on that? Where do you go at that point? What’s your next option, safety-wise, to put somebody on without a concern?

 I think we’re going to have to have 4 tiers. I would say that we have the orals, like a second tier of the injectables where, arguably, depending on how people look at it, teriflunomide is in the first tier. The second tier would be the orals outside of teriflunomide. The third tier would be cladribine, and the fourth tier would be the infusion agents. Of course, part of what makes this very hard is that we don’t have a lot of head-to-head data sometimes to fully establish this. But, based on the pivotal trials, we make our best guess. And so, I think ultimately the answer is the efficacy is between the orals and the infusions.

These comments are not asynchronous with what’s been said, but I think a concern that I have is, arguably with all of the agents, when you give the drug, MS [multiple sclerosis] can recur. That’s for sure. We saw this with cyclophosphamide which is, I would argue, a stronger drug than cladribine. Mechanistically, admittedly, it’s a little different. But if you follow these patients for long enough, even after giving them a good slug of cyclophosphamide, you’d see that the disease comes back. So I think it’s a matter of time and also interplays with the overall disease severity of the patient.

But I think it would be a mistake to say we’re giving 2 courses of cladribine and we’re taking care of the patient’s MS. Of course, you’d want to continue to monitor them. But I think another hard part about continuing to monitor them is, are you waiting until you have a new change on the MRI [magnetic resonance imaging] and they’ve had damage in the meantime? Having said that, I think that this is a drug for which the safety profile looks good. It seems pretty tolerable. I think another interesting little feature is that there are some patients who will lose oligoclonal bands when they’re treated with cladribine. So what is that mechanistically doing? It’s sort of an interesting thing to wonder. I think it will be nice to have this new addition, for sure.

I’m not a big fan of tiering at all. To me, the interest in cladribine is that it’s an oral induction agent. You clearly use it and it had very good data when used early, at the time of CIS [clinically isolated syndrome]. And you seem to have a prolonged effect. Can you have a sustained drug-free period, or can you change the course of MS? Could you use this induction therapy very early on, at the time of CIS, and convert an active MS to a much milder form of MS that then could be treated very well with very safe, easy-to-take therapies? So I’d like to see data on how long the effect lasts, and some data suggesting whether you can change the course early on of what appears to be aggressive MS. Will it then respond to other agents? To me, that would make it, intellectually, a very attractive agent to use.

 To me, the direction of cladribine is the convenience of use. In addition to that, and this is more of an economic question, the cost of treatment may be theoretically less than with comparable drugs, in addition to what has already been said.

 Tom, in the CLARITY extension, were there any hints or markers of the durability, other than the clinical outcomes—white count, or something like that?

 No. That’s a very interesting question. In general, the white cell count, even in the CLARITY study, did not correlate with the disease attenuation in individual patients. So stratifying the patients on the level of lymphocyte reductions that they experienced didn’t correlate with the disease activity. I agree with what’s being said. Obviously, our patients will have new disease activity at some point. And so, how we recognize these patients is going to be an important question. For example, the question is, in such patients, neurofilament light will be a monitoring tool that indicates more subtle disease reoccurrence. So this certainly will need to be looked at.

Also keep in mind that obviously different individuals may need different doses or repeat dosing of this agent. Individuals may have an immunologic response in multiple sclerosis that may be more or less accessible to an oral treatment, depending on whether the immune aberration is already present—whether it’s present within the central nervous system, to a larger degree, or whether it’s present perhaps in areas that are less reachable.

Considerations for Using Oral Cladribine in MS

 You bring up one of the issues now. We’ve seen this now with cladribine. We haven’t with alemtuzumab, and we haven’t with the autologous stem cell transplantation in the bone marrow transplant. You have a period where you’re not doing anything therapeutically, where people have some degree of protection. None of them are a cure. There is different cost effectiveness. One of the things I’m struggling with is, when do you restart the therapy? If you hope that you’re going to see something on the MRI [magnetic resonance imaging] as the first sign, and that may be OK because MRI is very sensitive, but if the next event is a relapse and that’s what you make the decision off of, well, 50% of relapses leave some residual disease. And so, you probably have waited too long.

How are we going to make that decision of how long? I mean, here you have data for 2 years. It’s group data, not individual data. We have the same thing with alemtuzumab. There are some people who fail all along, and if you look at the well-done autologous stem cell transplants, they all have a failure rate out through 2 years and moving onward, and some more than others. How are we going to be able to protect those individuals?

 Theoretically, you want to look at the target of the drug and see if the target is adequately suppressed or modified. For example, if you were using a B-cell depleting agent, you would hope that B-cell counts, CD19, for example, would be a good indicator for when to administer the next dose. But, as you pointed out, patients may have a relapse before then, while their CD19 are still undetectable. What does this tell us? It probably tells us that it’s working through additional mechanisms, and maybe that’s not the absolute biomarker for timing the next dose of the drug. One could say the same thing for alemtuzumab. You might have suppression of a lot of the target cells, the CD52 cells, yet the patient might have a relapse. Well, what could be happening is that maybe those patients have not yet developed regulatory T cells, which we know, in the context of alemtuzumab, appear later in the disease during a treatment course. So the point I’m trying to make is that we try to use the target of the drug as an indicator for dosing, but it is not always the case that they’re predictive. But they do tell us that these drugs, while we think they work through 1 mechanism, probably involve several mechanisms, some of which we don’t know.

 And what do you do with people when they have no T cells and are having an attack, but alone, just not having B cells? I think that’s still an open question, as to what to do then.

 We also need to keep in mind that what we measure peripherally in B cells or T cells is not necessarily completely reflective of what’s present within the tissue. And so, while the peripheral component of the lymphocytes is certainly something that we traditionally measure and look at, it may not necessarily fully inform us, with these treatments, how it is actually represented in the target or how it regulates the disease process.

 There’s some rather old data in irradiated mice that show you can get attack of the nervous system with very few T cells and still produce damage.

Recommendations for Restarting Therapy in MS

 While we’re on the topic of targets, Suhayl, tell us what you think, in terms of an understanding of the mechanism of action [MOA] and how important that is in choosing a therapy.

 Let me just start by saying that we don’t really fully understand the mechanism of action of the drugs we have for MS [multiple sclerosis]. We have some understanding but not complete understanding to guide us in choosing therapy. But when I think about mechanisms, I think of the drugs in 4 categories: Those that are immune modulatory, meaning they don’t kill cells but they change their immune profile and inflammatory profile; I think of the immunosuppressive drugs that deplete cells—T cells, B cells, or both; I think of drugs that interfere with trafficking and migration to the nervous system; and finally, those that are potentially neuroprotective, although we don’t really have solid evidence for that with the existing drugs.

In terms of choosing therapy, I think understanding the MOA has certain applications at the present time—practical application in the clinic. The best example would be in the case of NMO [neuromyelitis optica], where we think that the antibody is involved in the pathogenesis. And because of that, it would make perfect sense to use a B-cell depleting agent rather than a pan-depleting agent. That’s one example where I think mechanism of action can help you choose a therapy for a particular condition.

However, I think understanding the MOA has more implications on safety, particularly if we think of biomarkers that can predict complications of treatment. A perfect example would be the case of natalizumab and PML [progressive multifocal leukoencephalopathy], where we don’t treat patients who are JC virus-positive. Another situation would be pregnancy. Understanding a mechanism of action will help you decide on the safest drug during pregnancy. We have data on a fair number of drugs being safe or unsafe during pregnancy.

Finally, I think mechanism of action is important in choosing drugs based on comorbidities that patients might have. Just as an example, if somebody has cardiac disease, you want to avoid drugs that can affect the heart rate or heart function. In thyroid disease, you want to avoid drugs that can result in autoimmune thyroiditis. Those comorbidities and understanding the mechanism of action of the drug and its adverse effects can help you select the appropriate treatment for the right patients.

 I particularly like natalizumab in people who have concomitant inflammatory bowel disease, for example. On the other hand, I might avoid fingolimod in diabetics because they’re more likely to get macular edema. And hypertension occurs with a couple of the agents, and such. So I think that it’s not the first consideration, but I think MOA fits in there, especially with the comorbidities. The other is pregnancy, which is a much longer conversation but one that we discuss with every patient since almost all of them are in reproductive years, and the timing and such, which will take us to another evening’s discussion just on pregnancy. Yes, Clyde?

 Another thing is the notion of sequencing. With a mechanism of a particular drug that’s going to necessarily deplete a population of cells, maybe you have another drug that’s going to have an effect on trafficking. And if there are no cells around to affect trafficking, you’ve wasted your time, or just the opposite may be true. So knowing what the drug does is very important, and knowing where you’re going to go after that is key.

 Absolutely. I think it will be more important when we start combining drugs. Ideally, if you’re entertaining combination therapy, you want to use drugs that have different mechanisms of action. That’s why understanding mechanisms are important, although I should say that a fair number of these drugs have overlapping mechanisms of action. But there are differences that could be enough to rationalize combining certain drugs.

Significance of Understanding MOAs When Treating MS

 One of the most exciting things for us in the MS [multiple sclerosis] community was, in 2017, when we got our first approved therapy for progressive disease. Tell us a little bit about where we are now with progressive disease.

 So, it’s true. Ocrelizumab was approved for primary progressive disease based on the strength of a phase 3 randomized controlled trial. In fact, in the approval process, there’s another drug, siponimod, which showed successful trial results in secondary progressive MS. They both had, I would say, a similar reduction in the proportion of patients hitting the progression metric, as we define it by the EDSS, or the Expanded Disability Status Scale, which is reduction of about 24% compared to placebo, which is a start. It’s better than we could do before and so it’s good. Of course, we would like to see better results for our patients than that. And I think that’s sort of a broad overview.

 Tom, let me ask you this. Fingolimod clearly was negative for primary progressive disease. Why do you think siponimod, which is next generation, S1P1 receptor, succeeded, albeit modestly, in secondary progressive disease?

That’s an age-old question, right? That goes back to the interferons, where a European trial in secondary progressive MS was positive and a US trial was negative. And some things that we potentially don’t fully understand at this point may actually help us. Obviously, siponimod also has certain receptor specificities that fingolimod doesn’t have. So one argument would be that the distinction in the S1P receptors that are actually targeted could have caused this. The other thing is obviously that what we really control with siponimod is still an inflammatory component of the disease, and with that inflammation drives disability. And that’s really what we are looking at. It also raises the issue that both a S1P receptor agonist, antagonist and an anti-CD20 molecule have some effect in progressive forms of MS. Is it then that the immune mechanisms still play a significant role in this particular condition and also in primary progressive MS?

If I would have to put down my chips today, I think what this controls is the inflammatory component of these states. It’s not what was referred to as a true neuroprotective component, meaning helping previously injured or currently injured neurons to survive longer. I don’t think that either siponimod or ocrelizumab actually do that. I think they control the residual inflammatory activity.

 But at least siponimod enters the CNS [central nervous system].

 And the anti-CD20 really doesn’t. It’s less than 0.1%. So if you’re going to have a direct effect, at least siponimod, in theory, gets in there.

 Maybe it is the fact that some of the receptors are bad in the CNS?

 I have a different take on this whole conversation. If you look at the ocrelizumab clinical trial as primary progressive. And you look at the proportion of patients who had Gad [gadolinium]-enhancing lesions at entry into the trial, and you look at what happened, which had maybe 13%, 14% Gad enhancement, and the ocrelizumab trial had like 25% or 26%. And then you take the group of patients who did not have Gad-enhancing lesions in the ocrelizumab trial. They didn’t do as well as the ones who did have Gad-enhancing lesions. As Tom points out, it’s an inflammation effect. I think the failed trial was really driven by the lack of inflammatory patients in that trial versus the ones who did have inflammation. That’s my take on those.

 But the problem was the subset analyses don’t give us enough information. You’re right, there’s a trend to difference in response between those who had Gad at baseline and those who didn’t. But the numbers were too small to be significant.

 And the subset analyses of the siponimod study didn’t give you enough information either. The subsets all showed a therapeutic effect.

 I would add another thing that I think is interesting. Just published this month was a look in a consortium in Europe that looked at rituximab in a second progressive population. The trial population was similar and showed an effect. So I think that’s an interesting little piece to all this too. And so, I tend to agree that I think it’s more related to inflammation. We don’t really know. Another thing that I think is very complicated about all of this is, the fundamental question is, what is progressive disease? I don’t feel comfortable saying that I could say exactly what’s happening. And, in fact, I would make the argument that it may be a little different for different people. Some patients may have more of an inflammatory component, and others may have less. And so, you may have better opportunity with some than others.

I think this issue about compartmentalization is an interesting one, but I think now we’re starting to understand that B cells are traversing more into the central nervous system than we thought. And so, what you do in the periphery certainly does affect things. So it’s complicated. I’m glad that we’ve had some positive effects. I guess fundamentally I worry that a lot of what you see with progressive disease is dialed in from things that happened long ago, which, again, I think makes an argument for more effective therapies up front.

 It could be also that there’s something different about the S1P5 function in siponimod versus fingolimod, particularly in its effect on myelination oligodendrocytes.

 But there could be more avidity, let’s say, or more S1P5 effect with siponimod. Remember, the clinical effect was in the range of 20%, so maybe a CNS effect could be related to that.

Treatment Options for Progressive MS

 It’s exciting that we have a start in progressive disease. But, as you say, we need to do better. What do you see coming along for progressive disease?

 Separate from immunomodulation and B-cell depletion, I think there are some new data suggesting that B-cell depleting drugs not only deplete B cells, but play a role in the results of the depletion of T cells that express CD20. CD20-positive T cells may be involved in progressive disease because they are proinflammatory. They do correlate with disease progression. There has been a recent paper or 2 showing that ocrelizumab not only depletes B cells but it depletes the CD20-positive T cells. I think that would help explain a number of things in the context of this form of treatment, including progressive disease. So it really raises the question, could those CD20-positive T cells be involved in progressive disease?

The other thing is, of course, neuroprotective drugs, channel blockers, although we have not seen remarkable results with those yet. Biotin is another thing that’s being looked at, and it would be interesting to see if the original study that showed some marginal effect would be replicated in the larger population of patients. And finally, in addition to that I would like to add that prostaglandin inhibitors may be promising in terms of treating progressive disease. There are studies going on with at least 1 drug in that context. And diet, I think, and vitamin D potentially could be important in slowing progression.

 Well, ibudilast had a positive phase 2 trial—a phosphodiesterase inhibitor. So we’ll see if that was pushed further.

 Alpha-lipoic acid has also been shown in brain atrophy studies, and may be another option.

 Statins, yes. So for ibudilast, I don’t think there’s a follow-on study yet.

 Statins—Jeremy Chataway, MA, PhD, FRCP, has another statin study that’s, I think, fully enrolled now. High-dose biotin&mdash;we’ll have data around this time next year. I thought it was going to be earlier, but it’s not. It’s going to be into 2020. I’m trying to think of what else is out there.

 It’s an anti-mast cell agent. It’s an oral agent.

 OK. So that gets into this issue that folks at your center have talked about a lot. Do we need to start looking at more innate molecules? Something working within the central nervous system, which you had stressed. Because I think that would make a lot of sense in dealing with progressive disease—if the inflammation, however much is there, and the pathologists say inflammation is there through the entire course of the illness, shifts to be more localized and compartmentalized within the central nervous system.

 I think one of the things we are left challenged by is the fact that we use and rely on MRI [magnetic resonance imaging] as much as we do, particularly in the progressive phase of the disease where you do not see significant changes with inflammation, or new lesions on T2 or flare. And there’s probably some low-level inflammation happening that we’re not able to visualize easily on the imaging. So when we come up with a better biomarker for that phase, we might be able to come up with better treatments.

Options in the Pipeline for Progressive MS

 One of the areas of MS [multiple sclerosis] that my colleague, Stephen Krieger, MD, always reminds me about is symptomatic therapy. His mantra is: There’s always something you can do—from the patient who comes in with very early disease, to the individuals that Tom is referring to that sort of drop off our radar screens because they’re so badly involved, and everything in between. So we’ve been very excited now for 25 years, almost 26 years over disease-modifying therapies. But symptomatic therapies are what make people’s lives better. And while we don’t have time to go through a whole list of them, there are things out there that we ought to make sure that the general public is aware of.

 I think there are a number of different things that we try. I mean the thing that most readily comes to mind is dalfampridine, which is a drug that’s indicated for walking dysfunction in multiple sclerosis. It was approved based on the strength of 2 pivotal trials showing 30% of patients responding with 20% improvement in walking times. In clinical practice, I find that it’s actually more frequent. Patients respond probably about 50%. I don’t know how much of that is placebo effect versus a real effect, but either way I’m happy when patients have this response.

And in practice, what I find is that patients say not so much that they can walk faster but they can walk further before they get tired, or they feel like they’re a little stronger in the legs. And so, that’s something that certainly I think makes sense to try. I don’t think there’s a lot of downside, and the patients will know pretty quickly whether they’re going to have a response. It can be discontinued if they don’t really notice much difference.

Another thing that we sometimes try for patients with urinary dysfunction, dyssynergia, spastic bladder—Botox can be useful. Another option is a combination of dextromethorphan and quinidine for pseudobulbar dysfunction. It’s something to keep on the radar because it’s something we can do something about. In practice, I don’t think it comes up that frequently, but I always worry that I’m not asking as much about it as I should. I try to be vigilant.

These are the more recently approved things, but we have medicines that go back for many years like baclofen and others that can be helpful. We use modafinil or armodafinil for fatigue in some patients, or amantadine. I sometimes find that amphetamine salts, in the right context, can be helpful for patients with cognitive dysfunction or fatigue. You have to be careful of [adverse] effects, but in the right patient sometimes I think it can be appropriate to try.

 Other thoughts on symptomatic therapies?

 I just throw out posterior tibial nerve stimulation for neurogenic bladder. I’ve had some patients who have had very good results with that.

 I think it’s also very important to keep in mind that many of our patients with multiple sclerosis do have depression. Neuropathic pain is a significant issue in patients, and perhaps just to give a shout out to the fact that in multiple sclerosis pain is normally best controlled with antiepileptics or antidepressants and not with narcotic pain medications.

 Good point. Symptom therapy is really very different strategically from what we’re doing with disease-modifying therapies, where we sort of prescribe it and then we sit back and wait. Symptom therapy has the advantage that you know, as you alluded to, pretty rapidly whether you’re doing any good or not—whether it be dalfampridine for walking, or baclofen, or one of the other anti-spasticity agents. But then the onus is on the clinician. You’ve got to keep actively working with the patient for dosing. While we don’t use dosing for the disease-modifying therapies, we do it for things like bladder control&mdash;not for dalfampridine, that’s a standard dose&mdash;certainly for spasticity, and maybe even for depression. So that requires a lot of back and forth and interaction with the patients, and constantly working with them over a short period to optimize what they’re doing. But, there again, you know it’s going to work or not work pretty rapidly.

 I think another thing to remember about symptomatic therapy is cognitive dysfunction, which is extremely common. I mean, 40% to 60% perhaps of patients have cognitive difficulties. We now know that perhaps cortical lesions are important. They correlate with cognitive problems. The question is, what are the factors that contribute to cortical lesions, and is the pathogenesis of cortical lesions different from the variety of lesions in MS? And how can we target or prevent those lesions?

Symptom Management in MS

We’ve now moved to doing a cognitive assessment annually on everybody. Just like we would do an MRI [magnetic resonance imaging] on these patients, we’re doing a cognitive screening—about an hour-and-a-half cognitive screen on everyone so we can follow them, because a single assessment doesn’t help very much. But following over time certainly would be another metric, which brings us into the last topic we’ll discuss, and that is outcome metrics. What are we actually looking to handle? We’ve alluded to this a little bit as we’ve gone along, and even the idea that maybe we’re not measuring everything we can. We certainly have relapses, if you look at this, and we’ve done a really good job at dropping relapse rates. That’s been the primary outcome measure for all of our therapies, and we’re very good at it. Progressive disease&mdash;we’re just starting to open the door to having these modest effects on progression as measured by EDSS [Expanded Disability Status Scale]. But where do we go next? What do we need to monitor to get to nothing going on&mdash;no disease activity, no progression? Clyde?

 Well, I think if you look at all of the clinical trials we’ve done to date, they’ve all measured those particular metrics—relapses, disability progression, and MRI. I agree that we’ve done a good job in terms of being able to suppress disease activity, but you still have the problem of people progressing independent of these other metrics. They’re not having relapses anymore. They’re not having any evidence that we see of any MRI activity with Gad [gadolinium] lesions or new T2 lesions, but they continue to slowly progress. We call that the degenerative phase or progression of the disease.

I think that maybe the neurofilament light chain will help us in that arena. I’m not convinced that this will necessarily be the answer to all this. But I do think that when patients come into the office, we have to make the determination of: Do we think that somebody is on the right therapy, or should we switch them to something else? When it’s obvious, it’s obvious. They’ve had clinical relapses. They didn’t recover. Or, there’s a significant amount of new disease activity on their MRI. Then it’s easy. We just say, “We’re going to switch you to something else.”

But we have no guidelines in this arena. We don’t really know what’s the best thing to do next, and that’s why it’s important that we understand the mechanism of the drugs and know what sequence we may need to go to next. We may go up in efficacy. For us I think, as practice clinicians, and for the people who we’re talking to, it’s not an easy piece. We don’t have the metrics to really answer these questions. We have some. We’re able to act on what we have. But we’re missing a lot of the disease activity that’s happening below the surface, at least below what we can see on an MRI.

 You mention doing cognitive function battery once a year. Certainly a Symbol Digit Modalities Test is just a few minutes. It is very easy to do on a regular basis—a minimum of once a year or maybe twice a year. It’s a 25-foot time walk. It’s easy to do. And, from the point of view of gait, you really can pick up issues there. A 9-hole peg test is not that difficult to do. We haven’t spoken at all about the visual system. Not yet ready, but optical coherence tomography holds real promise, and low-density visual testing is much more sensitive than the regular.

And then, of course, we haven’t really spoken about nonconventional imaging techniques. Let’s talk about brain volume loss. They are now commercial, and we’re beginning to see it. We’re getting volume measurements, and segmental, and cortical, and thalamic, etcetera. It’s not ready for prime time, on the individual basis, but maybe in the not-to-distant future it will be a meaningful thing to follow if we are correcting for all the factors that can influence it tremendously.

 We need to be able to detect, by imaging, those cortical lesions that are only detected pathologically. That’s really critical in predicting disability.

 I think that a very important point was made and is maybe hidden. That’s why I want to bring it out. The 9-hole peg test, 25-foot walk, and Symbol Digit Modalities Test—these are all quantifiable tests. Where I think we need to go is to move from a purely qualitative exam in multiple sclerosis [MS] to have some quantitative measures when we evaluate patients over time. This will also be very important as we determine the value of multiple sclerosis therapies, as we determine the value of multiple sclerosis care.

 So I think a very important point was just made, and I’d like to bring something out: Devices, or wearable devices. We haven’t explicitly mentioned them, and I think we need to better learn how to interpret the data. But, it seems intuitive that some of these things can be worn at home, and over time, will give us a better measurement.

I think that’s true. If you have a week’s worth of walking data versus just the 25-foot time walk, that’s going to be very useful.

: These are all outstanding points. The thing that we don’t have in this MS world that we live in is that we don’t have comparative data for these metrics. That’s what we need. We need to be able to do these in clinical trials as comparisons, because when we’re going to make a decision that somebody is not doing well on whatever therapy they’re on, we want to go to something else, just like you have relapse rate or you have MRI lesions, you have some data that say this might be better than that. We need those kind of comparison studies for these new metrics.

Measuring Outcomes in MS

 This has been extremely informative. Before we end the discussion, I want to ask each of you to give us some final thoughts regarding where we are in early 2019 with multiple sclerosis [MS]. Dr Coyle?

 This is a very exciting time. We have a multitude of disease-modifying therapies for relapsing MS. We recognize critical is early treatment, effective treatment, and shared decision making, as well as following patients closely. Do not forget wellness. Emphasize that. That’s important. That helps CNS [central nervous system] reserve. Do not forget comorbid conditions influence and damage the central nervous system, many of them. They must be optimally controlled. Lacking—we need effective CNS repair, and we need to figure out the puzzle of progressive MS and have neuroprotective treatments for neurodegeneration that probably will need to be started at the very earliest MS time point. But we’ve made tremendous advances. This is very exciting.

 To me, the most important unmet need is treating progressive disease. Taking treatment in all forms of MS to the next level, just like oncologists do, we need to look at combination therapy and look at induction therapy followed by maintenance therapy to see if those would really impact the course of the disease in the long run. And I think we need to, again, advance the area of personalized medicine in MS. We need to develop biomarkers to predict the course of the disease as early as possible. And, once you start the patient on treatment, we need to be able to predict the response to treatment within a short period. If you know that drug is not going to work for this patient, you make that switch early on.

 There is an early window in multiple sclerosis, in my opinion, that we don’t fully utilize, partially because we may not fully appreciate what it means to incur injuries during that window, and partially also because we may not have access to the full armamentarium of medications that have been developed for the treatment of multiple sclerosis. MS is also a disease that needs regular reevaluation of the patient, with new treatment decisions being made when the patient shows progression. So it is, in my mind, not acceptable if patients are on a single agent long term and slowly but surely deteriorate on these agents, whether or not they have overt activity attacks.

 I’m going to agree with everything that’s been said. I think the approach that we have lived for the first 20 some odd years of therapeutics in MS has been OK. I think we need to go much more aggressively up front. I don’t want to call myself an oncologist, but I feel like I’m moving into that conversation. And we need to be able to come up with strategies that are safe, so that we don’t have as much of a concern on the safety side of things. That being said, 1 piece that has not been addressed, and we didn’t talk about this at all today but I think it is a very exciting area, potentially, is the microbiome. I think there’s a lot there that we have not yet uncovered and it could really answer a lot of the concerns that we have. And I think it’s going to be the next decade’s worth of research.

 The further we go along the line here, the harder and harder it gets to say something really interesting. But I would agree with all the comments that are made. I would just emphasize that it really is exciting to be part of this community because there have been so many discoveries along the way. The meetings have so much energy, which people bring, and we see so much hard work. People really are thinking about how to make this better and better for patients. In the time that I’ve had the opportunity to diagnose patients, I’ve never looked at it as a fatal sentence. I think that we are very capable of managing patients and really changing their disease course. Of course, there’s room to get better, but I think there’s a lot of hope.

 Indeed. So these are very exciting times in multiple sclerosis, which has been leading the area of neurotherapeutics, which is growing excitingly, and not just for MS but for other areas of neurology as well. And I suspect we’ll continue for quite a bit longer. So thank each of you for your contributions to the discussion. On behalf of our panel, we thank you for joining us, and we hope you found this 

 discussion to be useful and informative. Thank you all.

Advances in the Diagnosis and Management of Multiple Sclerosis