A post-hoc analysis of the Prasinezumab trial in early Parkinson's disease

[00:01:21] Dr. Hugo Morales: So, I wonder if you can start by briefly explaining to our listeners the design of the PASADENA study and what were the main results before we move into the recent re analysis of the PASADENA.

[00:01:38] Dr. Gennaro Pagano: Yes, the PASADENA study is a proof of concept phase two study that is divided in three parts. The first part is a randomized controlled study of one year where Prasinezumab with two doses, 1, 500 and 4, 500 were tested against placebo. And the prominent point of the study was the change from baseline in the MDS total score sum of part one, two, and three.

And then the second part of the study is a blinded label study where we have everyone on placebo re randomized to either the 1, 500 or 4, 500. And the third part of the study that is ongoing is an open label extension where everyone gets re randomized to the low dose, to the 1, 500 of Prasinezumab.

The study initially was two part and we saw That the prominent point, unfortunately, was not met. So Prasinezumab did not reduce the progression in the total score of the MDS UPDRS, but at the same time, we saw that Prasinezumab reduced the motor progression, so the MDS UPDRS part three over the course of the first year.

Now, when the study was initially conceived. The Prasinezumab treatment was one of the first tested in trying to slow the Parkinson's progression. So it was conceivable to think that the drug could reduce all symptoms. It's important to explore. At the same time, if you look at how much the symptoms of Parkinson's change over time, you would see that the motor symptoms are the one that changed, particularly the mother signs.

The part three over a course of one year changed the most. And we look at that after we open the study and we realize that the only scale that changed significantly over one year was the MDS UPDRS part three, where MDS UPDRS part one, the non motor symptoms and MDS UPDRS part two, the function related to the motor part of the disease, were not changing much over the course of the one year.

[00:03:52] Dr. Hugo Morales: And so that led you to consider this hypothesis that maybe some group or subgroup of patients that may actually be benefited from this. Tell us, in your reanalysis, was it any specified subpopulation that you wanted to reanalyze? To see this signal of this change with Prasinezumab?

[00:04:13] Dr. Gennaro Pagano: When we do a trial that try to test a disease modifying therapy like prazinezumab, it could be a disease modifying therapy. You need to think there are two main problems. The first problem is the duration of the studies. If the study is too short, you will not be able to see it. And the second problem is how much the people change in that period of time.

So the only thing that you can do is either do a very long study. And we did this with the double label extension, right? We create an extension. It's very long, or you can look at the population enrichment of stratifying. With people that change the most in the short period of time. Unfortunately, in Parkinson's disease, we don't have a yet a biomarker that can tell us fully this population will change faster.

So the way in which we pre specified, because all the subgroup in this publication were pre specified before the primary analysis was completed, right? They were all based on the concept that Parkinson's disease progress faster in some people than others. For example, one of the strong predictor of progression is the baseline staging of the patient.

And we use the Hoehn and Yahr stage. So people with Hoehn and Yahr stage 2 change more over the course of one year in the placebo group versus the people that are Hoehn and Yahr stage 1, as expected. Same apply with people with the sleep problem called REM behavior disorder. These people, unfortunately, have slightly more aggressive disease.

Or they have maybe copathology is something that is still to be yet to be demonstrated, but they progress faster over the course of one year. And this was another pre specified subgroup. We also look at the subgroup called diffuse malignant that is defined based on the data. And this was done independently from different groups in Canada and also using the PPMI population, the Michael J. Fox Study.

So this subgroup clearly show. Or even with postmortem data that there is more pathology in this group. Now this is a data driven approach, meaning that we included people that have the worst motor scale at baseline with part three, but they need to also have one non motor in a worse quartile among sleep problems, autonomic dysfunction and cognitive impairment.

People could be also diffuse malignant regardless of their motor phenotype, but they have all three. non motor symptoms like cognitive impairment, autonomic dysfunction, and sleep problems. And also we also stratify for the treatment. In PASADENA, we are treatment naive in MAO B, right, in the early stage PD.

And what is very interesting that all consistent subgroups show that in the placebo group progress faster as expected, and they respond more to Parsinezumab.

[00:06:56] Dr. Hugo Morales: Was there any difference of the patients that, for example, remember reading your paper that. There was division between the patient that may be taking MAO B inhibitors. Versus the one that were in the treatment naive and things that this patient may perhaps have a more clear difference between the UPDRS motor parts.

[00:07:18] Dr. Gennaro Pagano: Yeah, if you look at the majority of the trials that are done in early stage PD. They only include people that are treatment naive. We likely included also the MAO B subgroup because in the treatment naive, we saw approximately 70 percent reduction in progression on the, on the part three. That is minor, right?

It's less than 20%. While in the MAO B subgroup, we had approximately 40 percent reduction. So almost double. And why is that? People might say, here's this, the co medication. I think one explanation could be basically numerical, like the signal to noise. If somebody's in the same stage of another subject, but require some medication in most of the, in some countries, at least people start with the MAO B inhibitor, right?

Especially in this very early stage of the disease. But they require some medication. So probably they're on the verge of progression. And this is one of the most difficult part for drug developer and for clinician working clinical trials is to how to identify the right patient to have in the trial.

The best patient to be in a trial with the disease modifying are the patient that progress in the context of the trial. Luckily for us, we were able to include them because they change over one year. What is also very nice of this analysis is that there is no much overlap between the subgroup because people might think, okay, these are the MAO B inhibitor.

Maybe they also diffuse malignant. They also have RBD. No, they don't. So in the treatment naive for example, the people with RBD actually respond more. Right. So it's not an overlap between the subgroup. So this is give also more validity probably of this analysis. Despite the, the exploratory nature, right? We need to always put this into the context.

[00:09:00] Dr. Hugo Morales: Apart for the UPDRS three, any other scales that we've improved?

[00:09:07] Dr. Gennaro Pagano: In the context of the one year trial, we look at all the scale that were in the study, like part one, part two, and part three. This is where the three main scales that were analyzed. Now, part one and part two, unfortunately, do not progress faster, not even in the subgroup. And this is also a very important element.

So we did not see a difference in part one and part two. We didn't in the diffuse malignant on the other subgroup. But they also do not progress faster, right? So this is, again, going back to what I said in the first place. In order to show that disease modification happen, you need the disease to progress.

Otherwise, you know, there is no difference between zero, right? If you don't change, I cannot show that the drug works on change. And then it's always relative to the maximum number of points that people change. So if you change five points within one year, let's say, 100 percent effect is low in five points.

We show 40 percent effect in the MAO B, for example, and 70 percent effect in the diffuse malignant. So this is very reassuring to us, and it's not only in hypothetical strategy, censoring the patient in the primary analysis, but it's also true when we look at treatment policy, meaning that we do it regardless of the use of symptomatic therapy, because some people doing the study start levadopa, for example, the results are still there.

And this is very reassuring in this subpopulation.

[00:10:28] Dr. Hugo Morales: Now, based on your findings on your study and reanalyzing all this data, it sounds very interesting that it's medication that may improve. This same motor symptoms in this specific superpopulation. But what about early stage Parkinson's disease that don't fit into these specific features?

Is it something that is yet to be explored in your next trials?

[00:10:57] Dr. Gennaro Pagano: That's a very, very important concept, right? It's only the people that are diffuse malignant like rapid progressors responding to the medication or is everyone? My current interpretation of the data is likely to be everyone, but the trial needs to be different. So in order to see an effect in population that are earlier, you need a longer trial.

And we had the opportunity to explore that in PASADENA. Because we have a second year, right? Not only the first year of the randomized controlled part, we have a second year where there is an extension that is blinded. And then we have another five years that are open label extension. So what we did, we compared the four years of the PASADENA treated with the real world data arm and the virtual control arm.

These results are currently submitted and under review. So we might be able to talk about that in the future. I presented some of those at the last ADPD Congress, and we see a separation over time, also an additional scale. So, and this is also for the people that are milder, right? Let's say treatment naive or other population at baseline.

So, to go back to what I said in the beginning, for a trial to show an effect, you need a treatment that works. You need the scale that you use. that is changing over time, and then you need to run the trial with the right sample size and the right duration to be able to show this separation. And with the PASADENA design, we were able to explore more than one hypothesis at the same time, and this is why I'm so excited.

[00:12:29] Dr. Hugo Morales: One last question for my own learning Cinpanemab is a medication that was trial as well as alpha in sort of a blocker. What is the difference between in Cinpanemab and, Prasinezumab,

[00:12:45] Dr. Gennaro Pagano: Unfortunately,

I prefer not to comment on a competitor drug, I can tell you that every monoclonal antibody are different. And in the previous podcast, we actually discussed about that in detail with Tony Lang. So I would refer to that to that podcast where Tony can explain all this detail.

[00:13:03] Dr. Hugo Morales: Thanks again Dr. Gennaro for sharing your research findings with us and also like to thank all podcast listeners and encourage them to go and read the paper, Prasinezumab slows motor progression and rapidly progressing early stage Parkinson's disease.