FDA's New MS Categories Muddy the Waters for Clinicians

Stephen Krieger, MD


March 24, 2020

This transcript has been edited for clarity.

I'm Dr Stephen Krieger from Mount Sinai Hospital in New York, here with a commentary on multiple sclerosis (MS) phenotypes—specifically the new US Food and Drug Administration (FDA) categorizations of MS disease.

In 2019, through a language change in the approval documents for siponimod and cladribine, the FDA revised the way in which they categorize MS. In creating the new designations, I believe they have added complexity to how we as providers categorize MS and prescribe disease-modifying therapies.

Since the 1990s, we've had the fundamental disease phenotypes: relapsing-remitting MS (RRMS), secondary progressive MS (SPMS), and primary progressive MS. In 2013, those were revised to include the idea that we need to do surveillance for both disease activity (the occurrence of new lesions or relapses) and progression (the insidious development of disability independent of a relapse). The new FDA categories have taken those ideas and applied them to how they describe the designations for approved agents to treat MS. We now have RRMS and SPMS, with the latter divided into active SPMS and non-active SPMS.

That distinction, I think, is really much more than just semantic. Instead, it takes the concept that we need to look for disease activity and puts it into the context of SPMS, with medicines then approved on that basis.

In particular, siponimod was studied for SPMS. In the EXPAND trial, it was shown that it prevented the accumulation of disability in patients with SPMS. The FDA, however, only approved this agent for active SPMS, and it was unclear to me why that distinction was made. Here was a drug that succeeded in a trial of SPMS at reducing the risk for disability progression. That's something that many other medicines have not succeeded at doing, including powerful medicines for relapsing MS like natalizumab, which did not succeed in an SPMS trial. By approving siponimod only for active SPMS, the FDA effectively put a box around this category, allowing it only to be used in patients who still have evidence of relapsing disease. They also extended the approval for siponimod to include relapsing MS, and even as a first-line agent for clinically isolated syndrome (CIS).

Now, there's reason to think that this drug might be effective in preventing relapses in early MS disease, but it simply wasn't studied in that context. It was studied for SPMS. So I find it challenging as a provider to sit down with a patient with early relapsing MS and tell her that this is an agent that she could consider taking; I have no data for its use in that category.

The issue of applying this active SPMS designation to, in essence, all relapsing MS was brought to bear on all of the other early approved disease-modifying therapies for MS. So now, many of our medicines for relapsing MS also carry this active SPMS indication. One can consider that to be progress. But at the same time, I think that these FDA designations really need to be based as closely on the clinical trial data as possible, so that we have real guidance as to how to choose medicines for individual patients.

These new categories have led to what I consider a series of conundrums facing the prescriber or the MS provider. In addition to siponimod being approved for early relapsing MS and CIS, where it hasn't been studied, the FDA also did not apply that approval to another agent approved at roughly the same time in 2019, oral cladribine. Oral cladribine was approved for relapsing MS and active SPMS, even though it didn't have a trial in SPMS, but it was not approved for CIS, even though it in fact did have a successful trial in that indication.

The Case of Ocrelizumab

There is no better example than ocrelizumab at displaying the challenges in trying to apply these new FDA categories. Ocrelizumab had successful clinical trials for RRMS and a successful clinical trial at preventing disability in primary progressive MS. Therefore, with the new categorization scheme, this drug is approved for RRMS, active SPMS, and primary progressive MS, whether or not there's any evidence of disease activity. But it's not approved for SPMS that doesn't have apparent disease activity, so called non-active SPMS.

That leaves us with a scenario where, if a patient used to have relapses and now has developed insidious progression perhaps a decade later, we would call that non-active SPMS, and that patient wouldn't be approved for ocrelizumab by the FDA's categories. On the other hand, if the patient has primary progressive MS, she's been developing disability gradually for 10 years, and she's never had a relapse, that patient could be approved for ocrelizumab because we think of that as primary progressive MS.

Is it really the case that a patient who used to have relapses, but now doesn't, is less likely to respond to this medicine than a patient who never had relapses? That's the kind of conundrum that I think we're faced with. These categories still maintain a distinction between primary progressive MS and SPMS, and I don't know that we really have a biological basis to make that categorical distinction. In truth, these categories make a series of distinctions that we don't really have biology to support.

Topographically Mapping MS

I have proposed the topographical model of MS, which was published in October 2016 in Neurology: Neuroimmunology & Neuroinflammation. That model proposes that disability in MS is a consequence of an interplay between focal lesions, disease activity (depicted in the topographical model as white peaks), and the loss of reserve, which compensates for lesions until reserve is lost to the point where it no longer does. The latter is depicted in the topographical model as a pool of water of reserve that keeps the disease activity submerged. However, once that threshold has declined and reserve has been lost, all that underlying disease accumulation manifests above the clinical threshold as a gradual increase in disability.

When that model was proposed, there were two issues that I think caused the most conversation. First was the idea that lesion localization matters. I think neurologists appreciate that lesions landing in an eloquent area are more likely to cause symptoms in MS, like optic neuritis or spinal cord lesions, than lesions that are in less eloquent areas like periventricular regions, where there's so much reserve capacity to compensate for them. The second issue was that of reserve. I believe the topographical model helped to promote the concept that reserve is something that all people have, but it declines at a variable rate as the MS disease course goes on.

Now with the FDA's new categorizations, I think it's also worth looking back at another implication of the topographical model, which is that there's really no hard line between RRMS and SPMS.

Searching for Much-Needed Nuance

The FDA's categories go part of the way toward acknowledging that this active SPMS designation represents something in between RRMS and SPMS. But the topographical model proposes that there's really a continuous spectrum of disease from early CIS through RRMS; through to early evidence of disease progression; and on into late, ostensibly non-active SPMS. But I think that by continuing to maintain these categories for the purposes of drug approvals, we're taking our eye off of a lot of the nuances that really helped to individuate how someone is doing with this disease.

All of these phenotype categorizations that the FDA have used are based on the crude outcome measures employed in our clinical trials, such as the increase of one step on our Expanded Disability Status Scale, the standardized disability score that's been used for 25 years. However, in more recent years, there has been a lot of research into much more nuanced ways to get at the individual accumulation of signs and symptoms that categorize MS. This includes research from Bibi Bielekova at the National Institutes of Health, using sophisticated measures of assessing for disability; from Jeffrey Cohen and Robert Bermel of the Cleveland Clinic, looking at very multifactorial ways of assessing patient outcomes; and by my colleague Jim Sumowski , doing the challenging work of trying to peer below the surface of reserve to understand the individual drivers of cognitive and physical disability.

These are the kind of advances, along with biological markers like neurofilament light chain and more sophisticated MRI measures, that will really enable us to individuate how we consider this disease. In doing so, they may also get us away from these categorical distinctions that, with the FDA's new mandate, really no longer apply very well to individual patients or to the data from the individual medicines that have been studied.

I think the future of MS is trying to understand these individual nuances, choose the best medication for the best patient based on the efficacy and safety data that we have, and perhaps to follow the topographical model's approach of looking at a more unified categorization of MS disease course. Perhaps that's the future. We'll have to wait and see.

Reporting for Medscape, I'm Dr Stephen Krieger.

Follow Dr Krieger on Twitter

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: