2nd of October 2020

Ranking the trials

In this blog post, I interview my neurologist, Dr. Bart Post, about his view of the Parkinson’s disease drug pipeline that I covered in a previous blog. Under my arm, I have the book Brain Fables, written by Prof. Dr. Alberto Espay and Parkyvist Benjamin Stecher. I also have a few questions about this book that I would like to ask Bart.

During the interview, we will discuss four topics: The hope that Bart draws from the current drug pipeline, the need to look at the disease differently, how science and clinical practice influence each other, and the ethical aspects of research into people at risk of developing Parkinson’s (and who possibly are never going to get it).

Bart works at the Expertise Center for Parkinson’s and Movement Disorders, part of Radboudumc, Nijmegen, The Netherlands. He specializes in people with Parkinson’s diagnosed under the age of 50. He recently wrote a scientific article about this together with others (Post, 2020).

Reading time: 10 minutes
The interview was in Dutch and has been translated into English by myself, so forgive me if you stumble upon some Dunglish : )  

I. Cautious hope

In trials that focus on symptom control, you have to look very carefully: Does this serve the patient or pharmaceutical companies? You can ask yourself: Is there a burning clinical question behind this study from people with Parkinson’s themselves? If so, by all means, go ahead! | Bart Post 

Well, I’ll get straight to the point. If you were a person with Parkinson’s, what would you focus your hopes on?

A difficult question. I often get that question from patients. The fact that we can treat the disease and make the symptoms less, that’s great. But I think 90% of people would say: “In the end, that’s not what it’s all about. How are we going to slow down my disease?” And that isn’t easy at all. The story of the pathophysiology of Parkinson’s – the processes in the body that supposedly cause disease – is as follows: “There is a folding error in alpha-synuclein that causes the proteins to accumulate, something goes wrong in the mitochondria and something goes wrong in the protein clearance system”. But how the three interact with each other and what the primary start of all that is, remains unclear. And then, you also see that it is different for everyone. For example, if you cut open the brains of people with a PARKIN mutation – people who have Parkinson’s on the outside – you will see that they hardly have any alpha-synuclein accumulation in their brain. That alone, I think is a great puzzle! That’s why I think: To mention one thing to raise your hopes? That’s almost impossible.

Personally, I find the GBA trials the most interesting. 5 to 10% of people with Parkinson’s have a mutation in the GBA gene. And we have already come a long way. Phase 3 is coming. If that is successful, then we would have a substantial group of whom we could influence the course of the disease. I think that is really very interesting. LRRK2 mutations are also interesting, but these only affects 1 to 2% of people with Parkinson’s.

How do you tell people who are not in the GBA or LRRK2 group that your hope is not focused on them?

What I usually tell people is the quote by Tom Isaacs: “Cautious hope instead of conservative paternalism“. If you see what has been added to our knowledge about Parkinson’s in the last five to ten years, who am I to know where we will be in five years? But when people tell me: “I actually believe that this disease will be slowed down in ten years’ time”, then I say, “I honestly don’t know. I hope so and I will certainly do my best and if there are any studies that have that claim and are safe, we will be the first to include hundreds of people. But at the moment I don’t know yet about things that are going to do that”.

What do you not believe in at all? 

I’m very skeptical about immunotherapy. There they try to break down the proteins that accumulate in the brain. That’s also one of the topics of the article we wrote recently (Smedinga, 2020): Have we learned enough from Alzheimer’s? Because there you can see that 20 years of research in the Alzheimer’s world hasn’t brought us very far in that area at all. Now I believe there was a subgroup analysis of a scientific trial that stood out, but well, we’re both scientifically trained enough to think: Isn’t there a pharmaceutical company behind this result, looking for an effect?

Who is this good for?

Right.

So, gosh. In Alzheimer’s disease, proteins also accumulate in the brain, just like in Parkinson’s disease. Why then – if interventions in this area didn’t work spectacularly in Alzheimer’s – would it work in Parkinson’s?

Exactly what Alberto Espay and Benjamin Stecher say in their book Brain Fables

Espay is someone I really appreciate in his thinking, so it’s nice to see that we are on the same track. There was another piece of Espay in Neurology about the etiology of this kind of protein folding errors (Espay, 2019). They put Parkinson’s disease and Alzheimer’s disease side by side. That was really grist for my mill. I was happy to see someone writing down very systematically where the holes are in the theory. What exactly is the proof? If you can’t establish causality, how can you base all kinds of theories on it? Maybe it’s an epiphenomenon – a side-effect of something else – and all the problems precede it, the hypothesis was. That is interesting.

Very much so

Perhaps the criterion for the ranking of the rest of the clinical trials is whether you would like to include your own patients in the trials. If I look through those glasses and look at the microbiome, for example, I will say: Well, interesting, I don’t know what it does. Cell therapy, I find that very interesting. That technique is so advanced. Antioxidants, GDNF, I don’t really believe in that anymore…

The trial Tom Isaacs has crossed the entire south coast of England for?

Yes, these trials are all disappointing.

Well, you have high responders and low responders  

Yes, then the question is of course: Who are these super good ones?

Exactly

Because that’s the same with cell replacement. I remember that from the trials of the nineties. Then you had a few people who did really well without pills. Really survivors.

In trials for symptom relief you have to ask yourself: Does this serve the patient or pharmaceutical companies? Is there a burning clinical question behind this study from people with Parkinson’s themselves? If so, by all means, go ahead! Think of the trial Rob de Bie is now doing with DBS (Deep Brain Stimulation) in people who have cognitive problems. We always exclude them from a DBS treatment. Why do we exclude them? Because on average they have more side effects. But do we know exactly? No, we do not. So: Go figure it out. A very important question. Simple. Important. Just do it.

Other important questions are: How can we treat dyskinesias – involuntary movements – in Parkinson’s disease properly, how can we stop freezing. That sort of question. Very relevant. Or suppose you can create continuous dopamine mirrors. That’s kind of the holy grail of treatment. Suppose you can do that for about ten years. Wow!

But to really stop the disease, we may need a really smart person who is going to look at things in a completely different way.

II. Divergent thinking

Scientists often proclaim that they stand on the shoulders of giants. Perhaps that very reverence for their mentors, and the women and men in their textbooks, makes it hard to maintain the skeptical nature that should be the hallmark of their field” | Benjamin Stecher, Brain Fables

otherness_ccby_sparks4pd_groot

You can see in the article about the drug pipeline that a lot of people are standing on each other’s shoulders. Different groups are working on the same target. But what if you stand on each other’s shoulders in the wrong place? There is also a category of ‘other’ with all the Einzelgängers in it. I think it’s important to take a close look at them all. Do you? 

Yes, someone can come up with something weird that really makes a difference. I think so too.

Talking about divergent thinking: I am in contact with a scientist named Kariem Ezzat. I have a poster with me in which I visualize his work (Malmberg, 2020). He says: whether the accumulating and accumulated proteins have toxic effects, we don’t know. What we do know is that if a protein precipitates, it can no longer perform its original function.

Yes, Loss of Function, right?

Yes, and Kariem’s point is: Replacement therapies been developed for proteins that have amyloid properties – meaning they are prone to aggregate or clump together –  in other diseases, but somehow we do not investigate that pathway in brain diseases such as Parkinson’s. Replacement is about giving back the original, functional protein. And if you give it back and adjust the chemical structure just a little bit, you make it less likely to clump together while you are at it. What do you think of this?

We already know quite a bit about alpha-synuclein (Bensky, 2016), but we still do not fully understand where the disease really starts and what exactly is the role of alpha-synuclein in it all. It’s just like real science, the more you dive into it, the more questions come up.

So you’re saying: We’re not going to do replacement therapy because we don’t know what the exact role of alfa-synuclein is so that wouldn’t be my tip of the day?  

Yes, I would be worried about that. Let me put it this way. If Kariem shows in preclinical experiments in animals that fewer dopaminergic cells die, then he has a backup for his hypothesis.

But we also continue with immunotherapy trials while we do not know the exact role of alfa-synuclein. We don’t consider that a problem, but we do consider it a problem when we want to try replacement therapy? 

I am also very skeptical about immunotherapy. Of course, something goes wrong with those proteins and we find deposits in those Lewy bodies, but what exactly happens we just don’t know. But what Kariem does is a kind of paradigm shift: let’s take a look at the function of that protein and then give it back and see what happens. Replacement hasn’t been tried yet.  I hope that Kariem will learn more about the role of amyloid proteins and that we can proceed from there.

III. Splitters and lumpers

brainfables_ccby_sparks4pd

In their book, Alberto and Benjamin say: “It’s time time to abandon the illusion that we can cure PD. We can only cure 1 molecularly defined subtype of disease at a time”. What is your opinion about this statement?  

Yes, that is a well-known statement. You have the splitters and lumpers groups. Look, as a clinician, you lump everything together and say: for me, it’s about the person sitting across from me in the consulting room. I make the diagnosis, I try good therapy and I hope that someone from a lab somewhere finds a substance that can inhibit the disease. Those are the lumpers for the clinic.

Scientifically you can say: If we continue to focus on sporadic Parkinson’s disease, we will keep on getting negative trials and to some extent false negatives because the small effect on, for example, the GBA group disappears. So I think it’s very realistic to first try to treat a small part of people with Parkinson’s and learn more about the disease. Because we learn from extremes. So taking those extremes out, zooming in on them, and learning a lot from them, that gives a lot of information. Without generalizing the results, of course.

Yes, that

Because that happens too much. Those are the people who call out on the news: “We’ve found it!” That just isn’t true. The nuance Espay brings, I like the kind of people who do it that way. Critical, nuanced, and positive towards the future.

What’s not to like

Exactly.

In the Personalized Parkinson’s Project, I have been enrolled on the basis of appearances.  

Yes, outside Parkinson’s. On the inside, you have hundreds of people with Parkinson’s diseases. That’s the splitters thought.

But clinically you also have the splitters. In Brain Fables, it says that we keep on making parkinsonian subtypes – based on the symptoms – but the results between studies are not congruent and not reproducible. What do you make of that?  

My Ph.D. was about clinical subtypes. We created a cohort of people with a clinical diagnosis of Parkinson’s disease and followed them along the way and looked at the quality of life. And we wondered: are there differences in progression between certain people: men and women, tremor or no tremor, axial symptoms early or not, thinking and memory problems. That sort of thing. And what we realized at some point was: what I put in my model is my own conviction. So I’m looking for ten variables of which I think: yes, they have been described earlier in the literature. I put them in the model and then I show that three of the ten variables correlate with poor progression. And these are often age, men do slightly worse and people quickly have axial symptoms. Then I thought: this isn’t science. Let’s take a hypothesis-free look. Just put the variables in the model and let statistics determine what correlates. Then you can see, OK, these patients form a cluster, what characterizes them?

Hmmm

So then your hypothesis is free. But even then, the results are similar: Young people have a different prognosis, then you have a group of people who very quickly develop cognitive problems.

Those are the main groups?

Yes, those are the main groups. For scientific studies, this kind of information is of course important. If you have a difference between an intervention group and a placebo group and you know which factors influence progression then you can correct for that. And then you can do more research with fewer people. So indirectly it is very relevant for research. But for a person with Parkinson’s disease in the clinic, the difference is not specific enough. So you need both these fields. The clinical picture, the biochemical approach and eventually it has to come together.

Now we have talked about a distinction within the label Parkinson. But then there is also the overlap between diseases. I understand that I can also have amyloid plaques in my brain and Tau that is normally associated with Alzheimer’s disease. In fact – there really are a lot of types of proteins piling up when we look at all the neurodegenerative diseases together (Haenig, 2020). So it’s not so strict: you have Parkinson’s, you have Alzheimer’s. We’re giving this label to people because it’s practical. 

If you look on the outside, you have people with classic Parkinson’s, you have people who have Parkinson’s for five years and already have quite a few cognitive problems and you have people who only have cognitive problems after twenty years. You can then look at the precipitation of proteins in the brain. Does that help? It helps in thinking about the disease. Does it help in the clinic? Zero, because we’re not going to measure that at all. And suppose we would measure it because we can make amyloid scans. Are we going to scan everyone then? And even then you will see that some of the people who have a lot of amyloid in their brains, don’t have any problems with thinking and memory at all.

Yes, I wrote a story about it. It’s called ‘Bakery disease and the paradox of fallen trees‘.  

Nice story, yes, I’ve read it. If everyone’s bakery breaks down for another reason, you also need another strategy to fix it. One bakery breaks down because trees fall on it, for the other, the problem is that there are no trees around it to protect it, for another the machinery is broken and with another, the baker doesn’t do his job. That is a nice metaphor for where things can go wrong.

IV. The promises of prodromal cohorts

parkinsonpredictors_cc_by_sparks

In your article  (Smedinga, 2020) you warn against prodromal research, i.e. research on groups of people who have an increased risk of Parkinson’s disease. Why?

Suppose you have a REM sleep disorder. Then we could say: Those people have a very high risk of getting an illness like Parkinson’s. That is a group that we would like to extract from the Dutch population. This is something to discuss with ethicists. Because yes, these are all people who have a problem in their lives with sleeping, but this could easily be solved with a sleeping pill. Do you have the right to tell them that in five or ten years they might get a very serious illness? Or should you leave them alone?

Can they understand what it means to have an increased chance of contracting a disease? 

Yes, it’s very complicated to communicate chance. It’s my job and even I find it complicated to sometimes understand what a particular chance means in practice. So it’s very complex. But what we want to achieve most of all is that people start to think about it: Now we’re just blindly making cohorts. At congresses, you hear people say: “We have a cohort and 80 to 90% of them will get sick; we have to start treating these people!” And we say: “Yes, that’s one way to look at it. But what are we doing to those people? Maybe we make them sicker. Do you know what I mean?

Yes

So you make them psychologically ill or afraid for the future and because of that, the quality of life deteriorates in the period that they are not ill at all.

A negative placebo effect

Yes, the nocebo effect. You are actually doing damage instead of good. So that’s what the discussion is all about.

So who is it good for, I wonder. What do patients themselves think of prodromal cohorts? 

There is also a group of people who really wants to know. A friend of mine, his father had Parkinson’s disease. He called me once and said: “I see that there are prodromal criteria for Parkinson’s, what am I supposed to do with that?”. Then I said: “My interpretation is that if you have REM sleep disorders, you have to pay attention, especially if a family member also has Parkinson’s“.  Then he said: “I’ll app you if I find myself hitting my wife in my sleep at night”. A bit discharged, but at that moment you get into the group that is at risk and if there is a pill that can stop the disease, then you have a chance that people take it, even when they are not sick at all ..

Then they are willing to take the risk? 

Yes, but then you might have an 80% to 90% chance of getting something. But suppose that’s 50% or even 30%, what would you do?

Yes, exactly. And you are also putting a burden on healthcare introducing all these people who momentarily aren’t sick.

Yes, there is a very nice article about this topic: ‘What is a timely diagnosis’ (Rees, 2018). And in this area of prodromal research, there is an enormous influence of the pharmaceutical companies. For one, Alzheimer’s used to be just a clinical diagnosis. You had to have memory problems and no longer function in daily life and then you got the Alzheimer’s sticker. Now you can already get a diagnosis if you have a few memory problems that don’t really bother you at all in daily life (MCI, Mild Cognitive Impairment). If you have MCI then you can get an epidural and an amyloid scan – of which you rightly say, what is that relationship actually? – and then you can get a biomarker induced diagnosis. And why? Because then you can participate in studies that claim to inhibit the disease.

So a noble idea becomes: We make a large group of people sick by giving them an earlier diagnosis. But that also includes people who are never actually going to get Alzheimer’s disease. And the label is given anyway so that they can take part in my trial and later on – when I have a drug to treat the disease – I can earn money from those same people. Don’t get me wrong: I really don’t believe that any pharmacist thinks so pathologically that they want to make people sick in order to make money themselves. They don’t. But that is what the side effect is of the way we are currently thinking and that’s what we’re talking about now.

What should the role of patients themselves be to make a statement about this?

That role is very important. We should understand what it is like for people to have REM sleep disorders, for example. Have interviews and conversations with the people concerned and ask: how important is this to you?

Yes, exactly. The quality of life 

Yes, and well, there are people who know they have a mutation, who are not sick and yet change their entire behavior. They exercise more, stop drinking milk products, and so on. Is that realistic thinking? I don’t think so. But this group of people is very eager to know about their possible fate and will also want to attach consequences to it. So for some, it is a solution. Because otherwise they would worry and by changing their lifestyle they can better deal with the uncertainty. But there will also be people who will say: I never wanted to know this; my life is broken.

They will only get worse

Yes.

We should distinguish such phenotypes somehow

And that is very complicated. In Alzheimer’s, you have a READYness cohort. That is a cohort of people in Europe with MCI that undergoes all sorts of tests and they are lining up and getting ready for when the pill comes. The question now is, should we do the same for Parkinson’s? Are we going to look for 100,000 people with REM sleep disorders, are we going to map them out completely and then tell them: You have an 80% chance of getting a serious disease? And then what? Then they will all be ready for the pill. That doesn’t come. Because that pill will not come for another 10 years and then that cohort will be too old or will have died because of other things. Or has never gotten ill at all. .

So is it interesting to learn about prodromal? Yes. But is it ethical to prepare such a cohort for a pill that never comes? I have my doubts about that.

You’re messing with people’s hopes after all

We do think that you can learn a lot from setting up such a cohort for Parkinson’s in the Netherlands. But if we are getting serious about this option, then we would prefer to look at it through an ethical lens to understand what it is we are doing to people and whether it’s the right thing to do or not.   

V. To keep on learning

Bart, I think it is great that I have been able to talk to you about this for so long. Looking back on this interview, what sticks with you most?  

I think we can strengthen each other by being critical.

In my experience, it is not that easy to get scientists to answer patient’s questions about the science they are undertaking. Especially questions sent by e-mail remain unanswered (with some exceptions).

Yes, this is a challenge, because you do not know who is behind certain e-mails. You have to establish a good working filter that will ensure you have a system in place that allows you to keep on learning. And then you see that many scientists are overburdened and stuck in the publishing system and therefore acquire a kind of tunnel vision. But exchanges with patients are extremely important. Just do it. And don’t let your ego get in the way too much. Which is quite difficult in academia …

Interview by Sparks

Sources

Espay, A.J., Vizcarra, J.A., Marsili, L., et al. Revisiting protein aggregation as pathogenic in sporadic Parkinson and Alzheimer diseases. Neurology. 2019;92(7):329-337. https://doi.org/10.1212/WNL.0000000000006926  (Open access)

Kim, S., Kwon, S.H., Kam, T.I., et al. Transneuronal Propagation of Pathologic α-Synuclein from the Gut to the Brain Models Parkinson’s Disease. Neuron. 2019;103(4):627-641.e7. https://doi.org/10.1016/j.neuron.2019.05.035 (Open access)

Malmberg, M., Malm, T., Gustafsson, O., Sturchio, A., Graff, C., Espay, A.J., Wright, A.P., El Andaloussi, S., Lindén, A., Ezzat, K. (2020) Disentangling the amyloid pathways: A mechanistic approach to etiology, Frontiers in Neuroscience, vol.14, p 256. https://doi.org/10.3389/fnins.2020.00256 (Open access)

McFarthing, K., Buff, S., Rafaloff, G., Dominey, Th., Wyse, R.K., Stott, S.R.W. (2020). Parkinson’s Disease Drug Therapies in the Clinical Trial Pipeline: 2020’. Journal of Parkinson’s Disease, vol. 10, no. 3, pp. 757-774. https://doi.org/10.3233/JPD-202128 (Open access)

Noordegraaf, M.A. (2020). Pathogenic transformations of shape and function of ‘amyloid proteins’ (Poster) (Version 1.1). Zenodo. http://doi.org/10.5281/zenodo.3842337  (Open access)

Post, B.; den Heuvel, L.; van Prooije, T., van Ruissen, X., van de Warrenburg, B., Nonnekes, J. (2020). Young Onset Parkinson’s Disease: A Modern and Tailored Approach. Journal of Parkinson’s Disease. https://doi.org/10.3233/JPD-202135 (Open access)

Rees, R.N., Acharya, A.P., Schrag, A., Noyce, A.J. (2018). An early diagnosis is not the same as a timely diagnosis of Parkinson’s disease [version 1; peer review: 2 approved]. F1000Research 7 (F1000 Faculty Rev):1106. https://doi.org/10.12688/f1000research.14528.1 (Open access)

Smedinga, M., Darweesh, S.K.L., Bloem, B.R., Post, B., Richard, E. (2020). Towards early disease modification of Parkinson’s disease: a review of lessons learned in the Alzheimer field. J Neurol https://doi.org/10.1007/s00415-020-10162-5 (Open access)

Showing 4 comments
  • Alberto Espay
    Beantwoorden

    This is an excellent interview, with many important concepts discussed. I am also very pleased to see a conversation centered around the themes of Brain Fables. Here are some comments from Bart Post about which I would add a few caveats:

    • “…what has been added to our knowledge about Parkinson’s in the last five to ten years” – we know so much about the “bakery disease” (so many pieces to one puzzle) but so little about each individual with Parkinson’s. Bart harbors hope that a disease-modifying strategy will apply to a large chunk of people, as reflected by the statement, “if there are any studies that have that claim [that slow disease] and are safe, we will be the first to include hundreds of people.” Wishful thinking. The larger the net over the spectrum of Parkinson’s we cast the less likely it is that we will strike success. We need to start small, perhaps very small, and expand from there.

    • “But to really stop the disease, we may need a really smart person who is going to look at things in a completely different way.” –Not really! We just need to accept that there is no disease, but people with a disease. We will never “stop the disease” in everyone. We can only stop the progression in biologically-defined individuals with a suitable molecular treatment –a cocktail apprach that may not apply to many (or even most) others.

    • “we still do not fully understand where the disease really starts and what exactly is the role of alpha-synuclein in it all” – The “where it starts” is an artifact of the single-disease model we created for PD. This was also the unfortunate theme of the Grand Challenges Patrik Brundin held in Grand Rapids earlier this year. Having a meeting about “where it starts” is reductionist and forces artificial creativity around a “grand theory” on the “universal starting place” (anyone saying it is not the colon would be considered crazy in this age). I endorse the response from Marina: if not knowing the role of alpha-synuclein has not stopped us from testing anti-synuclein therapies, why should we be concerned about potential alpha-synuclein replacement treatments?

    • “So you need both these fields. The clinical picture, the biochemical approach and eventually it has to come together.” It doesn’t have to come together. In fact, it probably never will. We need to break away from the shackles of having to prove there will be an alignment between the clinical picture (what Marina wisely refers to as the “appearance”) and the biochemical picture. A homogenous biochemical profile will most likely continue to be phenotypically heterogeneous.

    Many thanks again for this enlightening and timely discussion, and for Bart Post’s critical mind in a time when it remains easier to continue adhering to the dogma of Parkinson’s as one heterogeneous disease caused by alpha-synuclein aggregation.

  • admin
    Beantwoorden

    @Alberto

    Dear Alberto, thanks so much for your comments. I forwarded them to Bart Post and he appreciated your comments very much. He says: “Nice comment from a critical spirit!”

    I love the idea that the clinical picture and biochemical approach don’t have to come together. Where does our need to align things come from?
    And why should things necessarily have one starting point. It’s like Einstein said: “The only reason for time is so that everything doesn’t happen at once”. In reality, when the circumstances are right things do start at different locations simultaneously/independently. Aggregates of ice formed in geographically separate locations are just one example.

    Thanks again for furthering the discussion.

  • Sue Buff
    Beantwoorden

    Great interview. Very thought-provoking questions and answers. Thank you!

    • admin
      Beantwoorden

      Glad you liked it, Sue! Thanks for your comment, Marina

Leave a Comment

Deze website gebruikt Akismet om spam te verminderen. Bekijk hoe je reactie-gegevens worden verwerkt.