AAN 2024 | Distinguishing true relapses from pseudo-relapses in MS: the impact on clinical trial readouts

Steven: Hi, I’m Dr Steven Krieger from the Corinne Goldsmith Dickinson Center for MS at Mount Sinai in New York.

Enrique: I’m Enrique Alvarez, and I’m here in Denver at the University of Colorado Anschutz Medical Center.

Steven: At the American Academy of Neurology this year here in Denver, we presented the second part of a series of work that we’ve been doing, looking at pseudo-relapses in multiple sclerosis clinical trials and trying to think about how to analyze the data so we can have a purer sense of the efficacy of our disease-modifying therapies. We’ll start by talking about what a pseudo-relapse is. I think most clinicians know, but how would you describe that?

Enrique: I think a pseudo-relapse is a clinical event; some symptoms that a patient would experience and tell us about. But, they’re not necessarily associated with inflammatory activity: we don’t see MRI changes associated with them. Things that can bring them up include a lot of different things, I think infections would probably be at the top of the list, but it could be that they’re overheated, stressed, not sleeping well, those kinds of things. They can last a variable amount of time depending on how long that trigger is for them.

Steven: It can be hard in clinical practice to distinguish a pseudo-relapse from a true relapse, and in clinical trials it’s also hard to distinguish them.

Enrique: Yeah, I think sometimes it’s very clear. We know that sometimes in these trials and in clinical practice, that they contribute a lot to the number of clinical events that we would see. But I think where we became interested, was this idea that with high-efficacy therapies, that there is a lot more of these events than maybe events associated with MRI activity. So, trying to understand how much or how little they contributed to the trials was a big driver for us to start looking at this.

Steven: I think we both realized, as many other people have as well, that our highly effective medicines reduce new enhancing lesions by 95%, but they seem only to reduce relapses by 40 or 50%. So, something has to be compromising the purity of that outcome measure. We hypothesized that there’s a lot of pseudo-relapses happening in both the treatment arm and the comparator arm.

Enrique: I mean, that’s a big difference when you sit back and think about it, right? You’re going from something in the 90s to maybe a 40 to a 50% reduction. When we even look at new T2 lesions and sort of rebalance things, those numbers were holding true around the same percentages, so we thought that this might contribute quite a bit. We were lucky to be able to partner with TG Therapeutics and looking at their Phase III trials for ublituximab, the ULTIMATE I and II studies, and try to explore this question a little bit further – trying to understand how many of their clinical events that were called relapses were associated with MRI disease activity.

Steven: At ACTRIMS in Florida in February of this year, we presented the first piece of this work where we found that if we looked at an MRI confirmed criteria for relapse (a relapse followed by a new T2 lesion), and looked only at those events that had an MRI correlate, instead of a 50% reduction in relapses afforded by ublituximab versus teriflunomide, it was really more like an 88% reduction in those relapses, which has some face validity to it. It’s what we come to expect in clinical practice. Here at the AAN, we took this project a little bit further.

Enrique: Dr Krieger has been very involved in looking at the question as to what constitutes a relapse within a clinical trial, what kind of criteria? So, the question was then: can we modify the different aspects of a relapse to get closer to that 88% reduction that we could see?

Steven: Well, it was challenging!

Enrique: It was a little bit of a challenge. Yes. That’s a good way to put it.

Steven: We looked at different stringent clinical definitions of relapse: whether a relapse had to create more of a change in EDSS, or more of a change on the functional systems, or all of the above. Even with this more stringent clinical definition, we were only able to refine this outcome measure to around a 60% reduction in relapse. It was only when we added back the MRI criteria that we got back to this 88 or so percent reduction in relapses.

Enrique: We even played knowing that infections were such a big part of it, could we alter how long an infection could have been associated with a clinical event to help remove more events? So, we extended that to two months. Again, it still wasn’t very helpful in trying to clean up that process.

So, in looking at the criteria for trying to set up that clinical picture a little bit more to become a little bit more stringent in criteria, there were a few things that we looked at to try and change this.

Steven: Every relapse in a clinical trial already has a clinical definition. We tried to look for more and more stringent definitions of relapse. So, a higher bar, if you will, for what constitutes a relapse: more of an EDSS change, more of a functional system change, whether the patient’s experienced a new maximal level of disability at the time of the relapse. Things like this incrementally to try to eliminate these pseudo-relapses and fluctuations from the relapse signal.

Enrique: We also could change a little bit the criteria around infection to be a little bit more inclusive, so that we changed, instead of being an infection around a month, to make it two months and try to make it a little bit more inclusive of symptoms that might occur as infections are ramping up or down.

Steven: The investigator might not have realized at the time of the relapse that the person had an infection that might declare itself a little bit later. So, by capturing that, we try to eliminate even more of the infection-related fluctuations. But, even with all of that, we still only managed to get the relapse rate reduction closer to the truth to around 60%. We didn’t get anywhere near that 80 or 90% reduction until we added back in an MRI criteria for relapse.

So, I think what we learned from this is even with the more stringent clinical definitions, we really need an objective marker to know that new inflammation has occurred; a true relapse has happened. At this point, without an MRI correlate, the clinical definitions for relapse are still too noisy to really show the true efficacy of the disease. I think future work that we’re hoping to do will continue to look at objective markers for relapse.

Enrique: Yeah, I think it highlights a little bit kind of as clinicians, when we’re seeing patients, the difficulty in trying to figure out if a clinical event is really driven by inflammation or not. Clinically, by itself, that might be a little bit too hard.

Steven: Hopefully we’re going to think about other MRI biomarkers and even serum biomarkers that may help us to further redefine relapse, to get to a true assessment of MS disease modifying-therapy efficacy.