Rasagiline in PD: Does it change disease progression? And how do we measure it?

[00:00:36] Professor David Vaillancourt:
Hi, thank you very much for having me. It's a thrill.
 

[00:00:38] Dr. Michele Matarazzo:
You decided to do a trial on Rasagiline as a possible disease modifying agent in Parkinson's disease. Why did you decide to study this? And what is the background of this hypothesis?
 

[00:00:51] Professor David Vaillancourt:
So the reason we got into doing some of the imaging studies some time ago, was really because of the initial New England Journal of Medicine paper published in 2009 by Olanow and colleagues looking at one milligram or two milligram dose of rasagiline. Really important study, because I think it sort of revealed a lot of problems in the way that we assess disease modifying therapies, but it also showed promise for Rasagiline. The one milligram dose did do quite well in that study.

The effects at two milligram were not there in terms of a disease modifying agent. So we were just motivated to evaluate that therapy to see if it could have any effect on disease modification. And in our work over the last decade or so has been looking at developing imaging based biomarkers on magnetic resonance imaging that might be tracking progression in PD.

And so, we have I think viable markers in the tool kit, and we wanted to test a viable therapy on them.
 

[00:01:46] Dr. Michele Matarazzo:
Great. And actually that leads to my next question, which is: you decided to use a different main primary outcome for this trial, which is the free water in magnetic resonance imaging. This is not usual because in most studies, they use clinical outcomes or clinical measures to measure the progression of Parkinson's disease or even the symptomatic effect on Parkinson's disease.

But why did you decide to go for this nonclinical result? And also do you think this is the best possible option as a biomarker for Parkinson's disease, or maybe are there other possible biomarkers out there that could be useful? Let's say, still in neuroimaging, the FDG pet imaging with the PD related pattern, for example, or maybe quantitative motor assessment, that could also be interesting to monitor disease progression? What do you think about this?
 

[00:02:38] Professor David Vaillancourt:
Yeah, I mean, I think it's a great question. The reason to utilize an imaging marker in the first place is not that we don't think a clinical outcome is important. I think the clinical outcome is actually fundamental. Without a solid clinical outcome in a trial I don't think any therapy is going to be approved by government agencies to be utilized as a disease modifying agent.

Having said that, I think government agencies also, and I think neurologists in the field, are motivated whenever you see a biological variable and a clinical variable, which converge to show efficacy of a therapy. I think when you have both of those hitting at once then that's when you start to believe that therapy might have viability. We also felt like it could be quite useful in future clinical trials as sort of a model to look at.

So, the way that the ADAGIO study was done, which is a delayed start paradigm, it's really a high bar. I mean, it's a really difficult study design to get that effect. I don't think it's a trivial effect to get. And I think a simpler study design, which is to follow patients at one time, point to another, having a biomarker of relevance that should not be really amenable to acute changes from other pharmacological therapy.

So, you know, I think we all know the unified PD rating scale, the motor symptoms, particularly part three, will change if you give a patient levodopa. This scale can change by up to half, sometimes in some patients, because patients respond to a dopaminergic therapy.

So to have a primary outcome variable that is so amenable to change, I think it's difficult to use as the sole primary outcome variable in clinical trials looking for disease modifying therapy. Because it becomes difficult to tease apart a symptomatic therapy from a disease modifying therapy.

And you know, why we used free water in the substantia nigra and this study, there's a few reasons and it's not to say that it's the only one out there, because there are a few other markers out there that I think show promise. We published a paper a few years ago in Brain in 2017 where we looked at free water and the substantia nigra, and it was compared with a dopamine transporter imaging in terms of a power analysis. And it seemed to be quite favorable in terms of detecting progression effects and potential to detect a drug effect in a hypothetical clinical trial. So the sample size that came out of free water and the substantia nigra was quite favorable, comparable to DAT scan. So we felt like that was a technique that showed promise and might have a lot of impact.

And we all know the substantia nigra is just fundamental in Parkinson's disease. It's certainly not the only structure that's important — many other parts of the brain are important — but I do think the substantia nigra is really fundamental. And so my kind of feeling as if if one is not slowing down degenerative changes in the substantia nigra, then it's going to be hard to really slow or stop the progression of the disease. So we felt like that particular structure, because of its importance in the dopaminergic pathway and because it's one of the more consistent areas of pathology in the disease, we felt like targeting that structure in a disease modifying trial made a lot of sense.

The other imaging markers that are out there, I think at the time we started this study they weren't as established. So the PDRP is a great marker, but I don't know of a study that's done PDRP changes over one year. I've seen it, I think over two years in over four years, but I'm not aware of one that's done it over 12 months. It's a risk to power a study over one year based upon a two year change in an outcome. Because one would assume that it's linear in that case. And one doesn't know, you know, we never know the change of a variable over time unless we actually measure it.

I think another technique that's out there that's shown promise is neuromelanin. That's based upon a modified T1 weighted sequence. But that data wasn't existent when we started this study. But I do think it has shown some promise as a nice marker in PD.

Iron based imaging is also out there. You know, it's looking at susceptibility weighted imaging, that's also a technique that's shown a lot of promise. So, I think that there are a number of different promising imaging methods out there currently that can be used in clinical trials, certainly phase two type trials, that the companies might be interested in running, maybe not as their primary outcome variable, but as a secondary variable in their study. That is certainly something that companies would hopefully start to look at in the future, because I think government agencies are gonna want to see clinical markers change and biomarkers changed to approve these therapies.
 

[00:06:58] Dr. Michele Matarazzo:
Well, I think you read a lot of very, very interesting points. Combining clinical and objective, I think you're right, it's going to be definitely needed in the future, and not just in Parkinson's disease. I think in general, neurodegenerative degeneration is definitely going to be the way this is going to be done in future clinical trials.
And I also think, you were saying it few different possible neuroimaging biomarkers, and maybe also a combination of different biomarkers could be something very interesting to look at because you know, you can combine different magnetic resonance imaging techniques, or even combining MRI with PET imaging or even combining different, completely new neuroimaging with other kinds of outcomes. Again, let's say a quantitative motor assessment, for example.

And then another very good point you raised is that it's good to have something objective and that is not influenced by a symptomatic effect. I think that's the other weak point, let's say, of the PDRP that is a kind of responsive also to medication. And so if you have something that does not respond to dopaminergic medication, that's the biomarker you want to show a disease modifying effect, as you were saying at the beginning. So I think those are very, very, very good points.

Now that we know the background, what are the main results of the trial?
 

[00:08:15] Professor David Vaillancourt:
The trial looked at one milligram per day versus a placebo. It was a double-blind study. So neither the patients nor the investigators knew which group the patients were in. And the study followed them at baseline, and then 12 months later.

I have to say the trial also ended as COVID began. But we were able to complete this study and complete the trial.
So having said all that, the study basically did not find evidence that rasagiline at one milligram is slowing progression of free water in the substantia nigra over a one-year period of time. We didn't find really any evidence of that.

The second key finding in the study was that the way you set up the imaging sequence, now that the sequence to look at free water in general, not only in the substantia nigra, you can look at any, any part of the brain. It doesn't require a contrast; we typically use it on three Tesla machines.
But the sequence does require certain parameters that we think would be most advantageous. So we wanted to test a minimum set of parameters that we thought would be useful. And compared that to another set of parameters, particularly the repetition time, the TR time.

And so what we wanted to see was if the parameters were optimized to detect progression. And that's what we also found, was that a certain type of sequence, which is certainly applicable on just about any clinical 3T machine, is going to be able to detect progression. So we wanted to make sure that we knew the type of sequence that was most effective, because there's a push from MR manufacturers to lower the repetition time because it saves time and data collection. And so we wanted to see if that was a factor, and it was a factor in the study. So we found that having a long TR time was better at detecting progression and the substantia nigra. And we talked a little bit about why that might be in the study.
 And then another key finding was that we were able to predict clinical changes over a year in the patients from baseline markers of free water. So the higher the measurement of free water in the substantia nigra, the worse the patients did after a year overall.

So basically it seems to predict changes in the future rather than predicting the current state of the patient. So it seems to be a longer-term predictor in the future, which is consistent with the previous evidence that we had found before.
 

[00:10:21] Dr. Michele Matarazzo:
Well, there are few followup questions to all of these very interesting results. The first that comes to mind is: so there was no difference between Rasagiline and placebo. Should we give up on Rasagiline as a disease-modifying drug or do you think we should keep investigating it?
 

[00:10:38] Professor David Vaillancourt:
Well, to be honest, I don't know. I think testing Rasagiline at the disease duration of the patients that we studied would probably be an unwise investment in the future. Does that mean the Rasagiline would not be effective in a preclinical cohort? No, not really. That is quite possible. You know, maybe in an RBD cohort or another genetic cohort of PD. I wouldn't give up on it. I just think that I would use the information that we found in this study to think about other disease modifying trials.

The question you're raising is actually a very important question, is: Would any therapy be successful after several years of having PD, and I don't think we know that yet, because no trial has been successful to date. Which I think is really a fundamental question for the field.
 

[00:11:24] Dr. Michele Matarazzo:
Yeah, definitely. What are you going to do with free water in the future? And specifically, what have you shown already with free water in matter of diagnosis, differential diagnosis, also not only as a biomarker of progression? And or what else are you planning to do in the future with this very promising tool?
 

[00:11:43] Professor David Vaillancourt:
What we've been doing the last, I would say three to five years has been kind of trying to understand what processes are changing that might influence free water. So we've been doing studies in animal models to manipulate inflammation without changes in neurodegeneration, or to manipulate neurodegeneration where you would also have ongoing inflammation occurring.

And what we've found is that both inflammation without neurodegeneration and neurodegeneration occurring, both can elevate free water. And so both of those mechanisms are, of course, relevant in the loss of dopaminergic neurons, or other cell types as well. So we think that the elevation of free water could potentially reflect ongoing inflammation or ongoing neurodegeneration. So that's a key set of findings.

The other thing we've been doing, because free water is not ... so like neuromelanin imaging, for example, is really focused on a couple structures, maybe locus coeruleus or substantia nigra. Diffusion imaging, it's used in TBI, it's used in stroke, it's used in MS ... So you can look at many parts of the brain using diffusion imaging. So what we've been doing most recently is trying to take a big data, AI machine learning type of approach, where we sample lots of different parts of the brain and use a machine learning algorithm to see if we can predict clinical diagnosis as an outcome variable.

And we've been doing that, I think, successfully so far. And now we're testing some of the models that we've developed in a large prospective study ongoing right now, which is in the United States and in Canada. So that's what we're hoping to do is see if this approach where we combine data across MRI machines can be successful. It's not easy to combine data across MRI machines because there's different manufacturers, everybody kind of has a different model... and so we're trying to see the extent to which we can combine them and utilize the data to predict clinical diagnosis, to basically help in predicting differentiation between PD or other types of Parkinsonian disorders.
 

[00:13:39] Dr. Michele Matarazzo:
Great. And just to understand that, are all the centers with 3T MRI machines, or is it also 1.5 Teslas?
 

[00:13:47] Professor David Vaillancourt:
Yeah, we're testing it all on 3T MRI machines. You know, if you look in the literature, you'll see some studies at 1.5 T using diffusion imaging and, my kind of gut feeling is that it might work at 1.5 T, but I just never tried it. And so, I don't know. But I think given the number of 1.5T machines out there clinically, I think it's certainly something that should be looked at. Because you know, if you're looking for impact outside of a research setting, 1.5 T would certainly have a significant impact.
 

[00:14:16] Dr. Michele Matarazzo:
Yeah, that's exactly what I was thinking. I mean, a matter of how translational that is to actual clinical practice, 1.5 T obviously, as of today's, at least in most places is the standard of clinical practice. As you were saying in research, it's much more common to have 3T. But, the first step is to, of course, to demonstrate this in the best case scenario, which is 3T and then maybe the next step will be to test this in 1.5.
 

[00:14:43] Professor David Vaillancourt:
Yeah, I would like to in the future. We don't have an ongoing study right now doing that . We're focused at 3T, but yeah, I think once we have algorithms set, it would be certainly interesting to test at 1.5 T, and I would love to do that.
 

[00:14:56] Dr. Michele Matarazzo:
Well, thank you very much for your time. It has been a pleasure to have you on the MDS podcast.
 

[00:15:01] Professor David Vaillancourt:
Thank you very much.
 

[00:15:02] Dr. Michele Matarazzo:
We have had professor David Vaillancourt and we have discussed the article "Diffusion Magnetic Resonance Imaging Detects Progression in Parkinson's Disease: A Placebo-Controlled Trial of Rasagiline," from the Movement Disorders journal.

Don't forget to download the article and read it, and you can do that from the website of the journal. Thank you all for listening.