Part 3 Interview with Jonas Hannestad, MD, PhD – Chief Medical Officer

 

In Part 3 of our discussion with Gain’s Chief Medical Officer Jonas Hannestad, MD, PhD, we get a more in-depth understanding of the relationship between academia, industry, and the NIH which has evolved significantly, fostering increased collaboration, especially in the past 15 years. While the NIH-academia partnership is well-established, challenges such as intellectual property concerns hinder more extensive sharing between academia and industry.

 

Where Academia and Industry Converge

 

From here, we can transition to talking about the workshop that you attended (NINDS’ Advances in Therapeutics Development for Parkinson's Disease). You have a foundational background in academia as well as pharma. What do you think of the current state of collaboration between academics and industry? How can we foster more collaboration between the NIH (National Institutes of Health), academia, and Industry?

 

The relationship between the NIH and academia is one that has existed for a long time because the NIH provides most of the grant funding for biomedical research in academia. It’s a relationship that’s well established. 

 

Now there are probably components of that relationship that could be improved, but that's outside the scope of what we're talking about today. The third leg of that stool is Industry. Several decades ago, there was relatively limited collaboration between industry and academics and also the NIH. They were sort of separate worlds. I think that has changed a lot in the last 15 plus years that I've been involved in industry and biotech.

 

Today, there is a lot of collaboration between biotech companies or pharmaceutical companies and academic institutions. Typically, pharmaceutical companies will take advantage of certain very niche expertise that some academics have that they don't have the capability to set up internally. They then collaborate with those people in academia to test certain hypotheses, ask specific questions, etc. This is often related to discovering new targets or validating the target. So basically it’s about deciding: Is this molecule or this pathway relevant to a certain disease?

 

Once you've answered those questions, often the pharmaceutical company or the biotech company will take on most of the work itself, because that’s where you get into the issues of intellectual property and who owns the data. I think pharma and biotech companies are fairly hesitant to share a lot with academics or other sorts of public entities for obvious reasons.

 

They need to be able to put funding back into research and if you don't have strong IP protection then you can't commercialize novel drugs in the same way.

 

So I think that that's probably the area I’ve seen that’s the biggest challenge in these collaborations. With that being said, there's a lot of collaboration between industry and academia.

 

Do you think there's anywhere we can increase that in reality? I know you said IP is definitely a concern, but if we have some of the brightest minds working in academia and we're not giving them the data and the access, isn't that a roadblock?

 

Yes. And then there are a lot of, for instance, consortia or other umbrella collaborations that exist across many diseases. Parkinson's is one, Alzheimer's, ALS, many outside of neurology, that bring together pharmaceutical companies and biotech companies and academic institutions and clinicians and NIH, and sometimes even FDA, because in the end, they're all working towards the same goal. They're just doing it in different ways.

 

So you have these so called precompetitive consortia or collaborations where they're not necessarily working on a specific experimental drug, but they're working some of those more basic questions, for instance, Is a target relevant to the disease? Is the measurement of the endpoint in this patient population relevant?

 

So those are things that can apply and be useful for all those stakeholders.

 

 

Progress in Collaboration

 

So those [collaborations] do exist and they’ve made a lot of progress in the last couple of decades in terms of getting those parties together. But even on the specific drug testing side, there's more collaboration now because you have these platform studies where you have one placebo group that's shared among several active drugs. 

 

To make it more concrete, let’s say you have 5 arms in a study in a certain disease population, say Parkinson's disease. 1/5 of the patients enrolled will get placebo, the other 4/5 will get one of four active drugs, randomized and blinded. So you don't know who gets what. But then you have these different pharmaceutical sponsors who will test 1 drug each, but there's collaboration because they share the study design, they share the placebo group that they compare with each of the four active arms and so forth.

 

So there’s a lot of collaboration. Obviously there's the data on their specific drug that they keep very close to their chest for IP reasons, but there is increased collaboration. 

 

You seem optimistic about where we’re headed. That’s encouraging. 

Identifying and Modulating Targets

 

How do you gain confidence that a certain target is relevant to Parkinson's disease? And how do you determine how to modulate it?

That's probably the most difficult question in drug development and also the most crucial one, because if you get that wrong, then you set off on the wrong path.

 

The starting point (and this kind of work is done a lot by academic institutions, but also to some extent by pharmaceutical companies and biotech companies) is usually having a genetic component. 

 

If you don't have a mutation in the gene that is associated with the disease (that’s the strongest link), then what you should look for is to see if a certain protein or pathway is altered in samples from patients. These could be brain samples from postmortem patients, where you look at the areas that are implicated in a certain disease.

 

Just to stick with Parkinson's, in the substantia nigra or in the striatum (which are the key motor regions of the brain involved in Parkinson's) you look to see whether this target that you have a hypothesis about is altered in some way in those samples.

 

If it is, then you can speculate that perhaps increasing this pathway or decreasing that pathway would be helpful in Parkinson’s. The next step would be to set up some kind of cell system (ideally with human cells from a patient with Parkinson's). You can use fibroblasts or inducible pluripotent stem cells, which is when you take cells from a patient and differentiate them to become like a neuron, for instance. Then you can do experiments in vitro looking at this pathway and ask, Okay, if I modulate this pathway in the way I think it should be modulated (either increasing or decreasing it), does that have beneficial effects on something I can measure in that cell system?

 

There are so many things you can measure. You can measure viability to see if the cells live longer or if they make more dopamine or if they have less α-Synuclein or they have better lysosomal function…  so depending on what the pathway does, you can look at these outcomes.

 

Once you establish that, then you can move onto animal models. This is where you would take a mouse or a rat with a deficit in this pathway that's similar to what you've seen in the human samples. First you ask, Does that deficit cause anything in the animal? 

 

The manifestation of disease in animals is often very different from humans, but sometimes you see something (i.e., the phenotypic effect) that if you modulate a certain pathway, there could be something that resembles Parkinson's or another disease. Then you can develop molecules that modulate the pathway. 

 

From there, you can say, If I modulate it in vivo in this model, I improve XYZ. So those are the steps you take to identify and validate a target. That happens over the course of several years and it’s a lot of work. 

 

 

Biomarkers and Wearables: Where CSF Fits In

 

Can we speak a bit about biomarkers and their use in trials, especially wearables and Cerebrospinal Fluid (CSF) markers?

 

Sure. We spoke earlier about imaging. Imaging in CNS drug development, when it's informative, it’s very useful because you get actual information about the brain, which is the organ that's of interest to you.

 

But the closest you can get without imaging is CSF. There are many things that we can't image, for instance, α-Synuclein right now. When you can’t image something, CSF is the closest you can get.

 

Why not use blood samples?

 

Because if you measure things in blood, sometimes it can be informative about what has happened in the brain, but the blood has proteins and other components from the whole body.

 

So sometimes it’s difficult to say, Well, if this protein or this lipid or this metabolite increases or decreases in the blood, it's due to something that happened in the brain…

 

We don’t know if that’s necessarily true because it could be due to any other tissue. Whereas the CSF is much more reflective of what happens in the brain. So I think CSF is a very attractive sample type for drug development in CNS. 

 

Two of the things that you can measure in CSF are first, the actual drug: if you give a drug orally and you know from animals that it gets into the brain, but you want to confirm that in humans, getting a CSF sample allows you to say, I can measure the drug in the CSF and that presumably came from the brain. It passed through the brain and can have effects on the brain.

 

So that's one thing. But the more exciting part is to measure biomarkers, which are kind of any biological signature that is related to the pathway that you're modulating. If you have a drug that works on a certain target in brain cells and you know that when that drug works, there’s a release of something from those cells, you can then measure that in CSF and say, Okay, this drug not only got into the brain at a certain level, but it also had an effect on these cells in this pathway.

And you only know this because you can measure it with the biomarker. 

 

The Challenge with CSF

Now the challenge with CSF is that it’s not an easy procedure; it’s one that patients often don't like. The procedure itself is fairly straightforward. It's not painful when people are anesthetized properly and the correct technique is used, but it’s common to get a headache afterwards, so that's something patients don't like.

 

There's always difficulty compared to blood samples and getting CSF. So that's why typically what we do is we get maybe one CSF sample after a certain period of treatment, whereas with blood samples, we can get many but they provide different kinds of information.

 

And for your purposes, would you say CSF provides more reliable information than blood samples?

 

Yes, in most cases of drugs that are working in the brain, yes.

 

Where Wearables Fit In

 

Let’s also talk about wearables.

 

Yes, so wearables is a field that I've been involved in my capacity of working in drug development for probably over 10 years now. It's a field that has evolved a lot in those 10 years because it’s related to technology. When I started looking at wearables in maybe 2012/2013, it was very immature. The things that we could measure were limited and the accuracy with which we could measure was very limited.

 

Now, wearables have come to the point where they can actually measure a lot about people in a way that's very accurate. Just to give you an example, let’s take Parkinson's. The way Parkinson's trials were typically done is that a patient comes to the site every few weeks or few months, depending on the duration of the trial, and then the site staff measures a number of variables about that patient, including their motor function, their rigidity, their tremor, their gait, etc.

 

But that’s just a snapshot in time and there are many factors that can affect that function that day. The next time you measure may be in a month. That’s not really enough information. With the wearable, you can get almost continuous information, and now that they're [the wearables] better, you can get continuous information that's very accurate. You can measure things like the activity level to see how much the patient moves around in their own home, because with a lot of diseases, not just Parkinson’s, people move less for many reasons.

 

You can measure that, and you can measure how fast they walk, how steady their gait is, and more. In Parkinson’s, you often see that people spend more time on both feet than one foot off the ground; the symmetry in that gait is something to measure. You can measure tremor. Even sleep can be measured more accurately now. Sleep is challenging because the gold standard for measuring sleep is that you need EEG [electroencephalogram] (the electrodes on your head) but now sleep can be measured very accurately with just the wrist wearable and that measures your movement as well as your heart rate.

 

So with those two components and advances in technology, we can deduce when people are sleeping and even what stage they're in, to some extent. So a lot of information can be gotten about the patients.

 

To your point about the “snapshot in time,” when patients come in to a doctor's appointment, there might be a little bit of a performative aspect where they're trying to do their best, so that may result in the information being skewed. Maybe they’re trying to walk perfectly or balance really well, but in everyday life, you can't do that all the time

 

That's very true. They want to show the physician or whoever's treating them that what they’re doing is working and that they’re better. But also, in the doctor’s office, it's a different environment, so it can be a little bit artificial. 

 

The wearables provide much more and different information that is probably more relevant to the day-to-day lives of people with different diseases.

 

The Role of the Caregiver

 

The last time we spoke, you said that the most exciting part of drug development is addressing whether or not it has the potential to actually help people with a certain disease. With that, we talk about the current state of engagement with patients and their caregivers, namely how to handle genetic information, how to get their input on trial design and endpoints, and also how to communicate the information from clinical development to caregivers and what to do with it. Would you be able to speak on that?

 

Yeah, I think that’s actually a good segue from wearables in terms of evaluating what is important for patients to measure. What does it mean to get better? What does that mean to the patient? That could be different from what it means to a physician or somebody else. 

 

Traditionally, for many years in drug development and clinical care generally, the doctor was seen as this authority figure who had all the answers and knew what was best for the patient. The patient would go and get their treatment and do what the doctor said. 

 

That has changed a lot in the last two or three decades and now there's much more patient involvement. There’s also an understanding that patients should have a voice in their care. They should have the information.

 

I’m talking about both drug development but also drugs that are approved. There’s a consensus that they should understand what the drug is expected to do. In what ways will I feel better? And what are the risks? From there they discuss with the physician if this kind of treatment is something they want to do… if it’s right for them. 

 

Patients & Patient Advocacy

 

Now part of that in drug development is understanding what these patients want in treatment. We’re tasked with developing treatments that can do that [address what patients are most concerned with] and measure it. Because here’s the thing: if we are doing a clinical trial and we're measuring X, but X is not very important to patients, and they're interested in Y, then we're not proving that the medication works for their purposes.

 

Yes, I recall you mentioned that in Part 1 of our discussion.

 

Yes, so that’s where the involvement of patient advocacy groups is very important. These groups are important for both rare diseases but also for more common diseases like Alzheimer's and Parkinson's. One thing they do is advocate for more funding from an NIH for a certain disease and work with the FDA to facilitate approvals of a drug.

 

Now they’re also increasingly working directly with the pharmaceutical industry and biotech to help us design trials that measure things that are valuable to them. That includes coming up with new endpoints. If there's something that hasn't been measured traditionally, and patients tell us that this is important to them, then we can figure out a way to measure that.

 

Another component is making trials not too burdensome for patients. That’s something we always hear. 

 

That must be a significant barrier to actually getting the information you need; if patients are not willing to participate because it’s too burdensome, then all that work is for nothing. 

 

Right. And it comes back to the question of what is important to measure. We have to find the midpoint there. So often, the scientists on the pharmaceutical side want to measure everything because it's interesting to understand the biology in how the drug works.

 

But a lot of those things that we measure are not relevant to patients. So again, we have to find the midpoint so we can please both sides, so to speak.

 

The Right Amount of Information 

The other thing you asked about was genetic information and that's complicated. In general, what I've heard from many patients is that they want information.

 

If they participate in the trial, at the end of it they want to know, What did this brain scan show in me? What did this blood test show in me? If I agree to CSF, what will that show?

 

They want that information, even if it doesn’t offer any specific relevance to their treatment or their disease. Even if the information is ambiguous and not easy to understand, they want to know it.

 

The same is true for genetic information. If you have a genetic mutation that increases your risk of a certain disease, and you already have that disease, then it doesn't have an effect on you specifically, but you may have passed that mutation on if you have children. So having that information can be difficult, because then what do you do?

 

Do you tell your children? When do you tell your children? Do you have your children tested? Those are complicated decisions, which is why with genetic testing, you always have to have someone who’s trained in genetic counseling involved. 

 

In general, most patients that I have spoken to want to know even about that kind of information.

On the other hand, there are people who say they don’t want to know because if you don’t know, then you can just live your life as if you don’t know. You don't have to make those [hard] decisions. So you just have to respect what the patient wants.

 

I think that taps into where medicine kind of becomes emotional, as it often does, especially in neurodegenerative diseases. It's not black and white. It's not cut and dry.

There are so many layers and it goes past the physical brain. It starts in the brain but it goes much deeper.

 

Yes. It crosses into who you are as a person.

 

That’s probably one of the most challenging elements for people and caregivers who are involved in this: to witness that and then to decide what to do with that information.

 

Ultimately, though, it is a good thing. To go back to the point about patient involvement, the more patients are involved in both their clinical care and their clinical trial participation, and the more informed they are, and the more they are asked about what they want, the better it is for all parties.

 

____________________________________________________________________________

As we wrapped up our discussion, I asked Dr. Hannestad if he had any closing thoughts. He said, “Every time I'm asked any of these questions, they make me think about the answers and what it all means. I think if you ask me these questions again in a year, the answers will have morphed a bit, which is probably a good thing. There’s some evolution there.” 

 

Stay involved in the evolution of drug discovery at Gain as you follow our progress in Perspectives. 

 

 

 

Latest Articles