The following first appeared on tmrwedition.
Peter Schmidt, Ph.D., is Senior Vice President, Chief Research & Clinical Officer, at the Parkinson’s Foundation and oversees research, education and outreach initiatives. Dr. Schmidt leads the foundation’s Parkinson’s Outcomes Project, the largest clinical study ever conducted in Parkinson’s disease and is active in research on diverse clinical areas in Parkinson’s. He serves as an adviser to several government, industry and foundation initiatives, with a focus on setting clinical standards.
Clinical research starts with the translation of novel ideas into the clinic, developing the evidence for efficacy, disseminating that evidence through publications and regulatory processes, and effecting clinical process change to reach across the spectrum of care, from convincing academic thought leaders to impacting the behavior of patients. Dr. Schmidt’s recent work includes serving as PI on the largest clinical study of Parkinson’s ever conducted, participating in the design of interventional trials, and planning studies to drive adoption of clinical best practices.
The following has been paraphrased from an interview with Dr. Peter Schmidt on January 30, 2018.
(Click here for the full audio version)
What is Parkinson’s Disease (PD)?
In reality, Parkinson’s is defined as a clinical disease. It is not defined by the involvement of alpha-synuclein (the protein that clumps together in the brains of people with PD), even though that is how I think about it when I think about what is actually happening underlying the disease. People talk about it as a synucleinopathy (a disease brought on by the misfolding of the alpha-synuclein proteins in the brain), but there are people that have been found to have clumps of misfolded alpha-synuclein that don’t have PD. So, we don’t know exactly when the underlying pathology causes the disease.
One of our challenges in healthcare is that our goal is not to address biology, it’s to help people live better lives. Life is a degenerative condition with a terminal outcome. It’s important to think about how what we do helps people live better. Now, attacking the biology is often the way we get people to a better life, but it is a better life that is our focus. For that reason we need to think about Parkinson’s as a clinical disease more than we need to think about it as a biological one.
What do you think the best answer is to the question of selective vulnerability in PD? (Why do specific cells die in the brains of people with Parkinson’s and not other cells?)
The first thing we know is that this is a disease of alpha-synuclein, because if you have a duplication of the alpha-synuclein gene, you get Parkinson’s. The next question then is, why does alpha-synuclein seem to affect dopamine neurons first and most aggressively? We also have to ask, in patients that have diffuse Lewy body disease (also a disease of alpha-synuclein), why are they experiencing a different selective vulnerability profile? By definition, if you have Parkinson’s then the disease is affecting your dopamine system, with varying degrees of vulnerability in other neurons. There is work that shows that there are structural differences in the brains of people that have cognitive problems resulting from the disease, so it seems like different people have different vulnerabilities.
There are a couple of reasons why dopamine neurons might be more vulnerable. They have the highest surface area of neurons in the brain, meaning there may be a permeability issue. Also, because they are large structures, they have high energy utilization, so they may be more dependent on mitochondrial function (mitochondria are the parts of a cell that produce most of the energy that the cell uses). Those two factors indicate that there may be a cellular metabolism component to this. There are probably also a series of genes that modify how PD affects neurons that could also be playing a role.
Is it possible for a model to replicate this disease accurately enough to be a useful model?
Yes, collections of humans “[laughter]”. Models can be designed to answer specific questions that would be hard to do in a person. For example, it’s very hard to get deep insight into the impacts of exercise because the way that we understand it in animal models is by making an animal exercise, stopping it, and then immediately cutting it into very thin slices. You can’t do that with a person. But what we need to ask is how well can we validate what we see in animal studies in humans. The key to models is taking the insight we get from them and transferring them into the clinic to ask, ‘if this were true, what else would I see?’ When you go to your neurologist and they ask you to tap your fingers, they’re not looking at your fingers, they are looking at downstream effects of your brain. That general principle extends to almost all of neurology. Even the imaging that we do rarely observes the direct effects of the disease, instead we look at downstream impacts of the disease.
If you had complete control over how we take a therapy from the lab to the clinic, how would you design that process?
First, thinking about how we got to where we are is important. When I was in grad school I’d go to my adviser and say, ‘it’s so frustrating, why doesn’t the FDA just do it the way we are saying?’. He’d say, ‘it really all goes back to the Hippocratic oath, when we design things in medicine, the first goal is do no harm’. This was the big controversy around the AIDS crisis in the 1990’s when they decided that the risk tolerance for AIDS patients is different. They took drugs that had a 10-20% chance of giving patients a normal life and a 10-20% chance of killing them today, and, because of the severity of the disease back then, they gave AIDS patients the ability to choose whether they would be willing to take that risk. That changed the way we think about drug trials.
In early PD, the disease itself is not very disabling, it’s a more problem of thinking about what is coming in the future. But there is a point where people might say that they are willing to take a risk of harm for potential benefit, that is where we need to rethink the process. We could also take people who are really struggling with an advanced form of the disease and give them the opportunity to make these re-balanced risk decisions. There’s a trial at Georgetown with a cancer drug called Nilotinib, it was a very risky trial because the drug does have the potential to cause death. People accept that risk in cancer because cancer is terrible and often kills people. But they found people with Parkinson’s that were willing to take that risk too. The trial has received a lot of criticism, but when we think about the ethical implications of engaging in a clinical trial, that Georgetown trial gave us something to think about. Is there a point at which patients should be offered that potential trade off? The participants in that trial found that the benefits of the drug outweighed its risks. There is still a challenge in determining how much was placebo benefit and how much was drug benefit, but there are three trials starting up now to study that same drug, so it seems Nilotinib does have some potential.
The big issue is, can we improve the statistical power of a patient population so that trials can be completed faster and so that we can get through the regulatory issues faster? If I was advising the FDA I would suggest to them that we should spend more time supporting research networks, where we identify people at diagnosis to receive care in an expert setting and will be part of an outcomes registry project like the one I oversee at the Parkinson’s Foundation. Then we can take those people and more accurately match them up with trials that would be appropriate for them, taking a better account of how fast their disease has progressed and their trajectory.