How Rune Labs Uses Data to Improve Prospects for Parkinson’s Patients
Harry’s guest this week, Brian Pepin, says there haven’t really been any advances in the treatment of Parkinson’s Disease in a decade. The standard treatment is still the standard treatment—meaning various drugs to replace dopamine in the brain, since the loss of neurons that produce dopamine is one of the hallmarks of the disease.
But there has been one important change during that decade.
Thanks to new technologies, ranging from wearables like the Apple Watch to sophisticated deep brain implants from companies like Medtronic, we’re now able to gather a lot more data about what’s happening in the daily lives of patients with Parkinson’s, and how the disease is affecting their brain function and their physical movement. Which means there’s now the potential to make much smarter and more timely decisions about how to dose the drugs patients are taking, or whether they should think about joining a clinical trials.
Gathering and analyzing that information and feeding it back to patients and their doctors in a user-friendly form is the mission of Rune Labs, where Pepin is CEO. He says we’re on the edge of a new era of “precision neurology,” where data gives doctors the power to predict the course of a disease and muster a meaningful clinical response. And he wants Rune Labs to be at the leading edge of that change.
Here is the full transcript of the episode:
Harry Glorikian: Hello. I’m Harry Glorikian, and this is The Harry Glorikian Show, where we explore how technology is changing everything we know about healthcare.
Today’s guest, Brian Pepin, says there haven’t really been any advances in the treatment of Parkinson’s Disease in a decade.
The standard treatment is still the standard treatment – meaning, various drugs to replace dopamine in the brain, since the loss of neurons that produce dopamine is one of the hallmarks of the disease.
But there has been one important change during that decade.
Thanks to new technology, ranging from wearables like the Apple Watch to sophisticated deep brain implants from companies like Medtronic, we’re now able to gather a lot more data about what’s happening in the daily lives of patients with Parkinson’s, and how the disease is affecting their brain function and their physical movement.
Which means there’s now the potential to make much smarter and more timely decisions about how to dose the drugs patients are taking, or whether they should think about joining a clinical trials.
Gathering and analyzing that information, and feeding it back to patients and their doctors in a form they can use, is the mission of Pepin’s startup, Rune Labs.
Pepin thinks we’re on the edge of a new era of “precision neurology,” where data gives doctors the power to predict the course of a disease and muster a meaningful clinical response.
And as you’ll hear, he wants Rune Labs to be at the leading edge of that change. Here’s our full conversation.
Harry Glorikian: Brian, welcome to the show.
Brian Pepin: Yeah, thanks, Harry.
Harry Glorikian: So I’ve read a bunch of stuff on the company. Sounds really exciting, but from everything I’ve read, it sounds like your mission at Rune Labs is to help patients and their doctors take a more data driven approach to monitoring and treating neurological disorders, starting with Parkinson’s. So can you start by reminding listeners what is Parkinson’s disease, what causes it, how it’s diagnosed, and how many people are affected by it?
Brian Pepin: Yeah, that’s a great question. Parkinson’s, like a lot of neurological diseases, the underlying mechanism, there’s theories about kind of what there are, but there’s there’s nothing that’s really consensus. This is what Parkinson’s is. So Parkinson’s, as we know it, is a collection of symptoms, a collection of patterns that we see in patients over over time and given points in time that we say, hey, this is Parkinson’s, and we’re going to treat that with this set of interventions.
And so it’s a movement disorder.
It manifests as a movement disorder, although it’s driven by the degradation and sort of dying off of neurons in the brain over time. It typically affects folks over the age of 55, but there’s also a large fraction of folks with Parkinson’s that are what’s called early onset. So they’re getting it maybe when they’re 30. Michael J. Fox famously had has had early onset Parkinson’s. Yeah. And there’s a plus or -10%, around a million folks in the US with some stage of Parkinson’s disease. So it affects about two thirds of 1%. And it’s also kind of of interest, I think it’s sort of a often treat as a sort of model neurodegenerative disorder and that it’s sort of in many ways just looks like accelerated brain aging.
Brian Pepin: Right. And this is just it’s the brain and for whatever reason, aging in an accelerated way. So that’s that’s the world in which we live in. And as you kind of mentioned, we’re trying to bring precision medicine to neurology, taking a Parkinson’s first approach. You know, the big reasons for that are Parkinson’s.
The data is sort of available today. So there’s a lot of data being generated, just part of routine care that we have access to and and can make use.
The clinicians are ready to actually use this data in the context of care. So that’s super important. And then also important is there’s there’s a range of new therapies that are in the development pipeline where the data that we’re bringing to the table and this kind of precision medicine approach can accelerate the time of those therapies into the clinic.
And that’s an important part of what we’re doing, too, is I think we realized that what’s available in Parkinson’s today is is mostly palliative care. It’s not it’s not the end game for it. We want to help bring some of these things that are protective or even curative to market as quickly as possible.
Harry Glorikian: So yeah, so, so you know, you mentioned treatments. So can you explain to people maybe how treatments for Parkinson’s has evolved over, let’s say, the past decade? Right. I mean, yeah, particularly, I think, you know, Neuromodulation Technologies like deep brain stimulation. Yeah.
Brian Pepin: Yeah. Well, so unfortunately this is kind of a part of the reason why we exist. Treatments haven’t evolved really at all over the last decade, but if you go back into the like the last, I think major additions to the therapy pantheon was deep brain stimulation, as you mentioned.
I’ll explain that in a sec. But the sort of baseline therapy for Parkinson’s is a dopamine replacement. So you have neurons in your brain, dopaminergic neurons that are dying off. You sort of compensate for that by by adding dopamine into the brain in a variety of ways. And that helps. It doesn’t help slow the progression of the disease, but it helps ameliorate the troublesome symptoms, the movement symptoms. Some of these things eventually as the disease progresses, that isn’t enough.
And so you have to do these second line sort of therapies, one of which is called deep brain stimulation. So this is where you get a pacemaker like device implanted into the area of your brain that is sort of going most haywire. Right.
In the disease. We actually one of our flagship partnerships is with Medtronic, who provides one of these devices and recently has started producing devices that are doing continuous sensing of the brain in addition to providing stimulation. That’s really cool because it gives us a whole bunch of extra data about how to optimize patients therapies I can talk about. But there’s also other kind of surgical interventions.
There’s what’s called focus ultrasound. So they go in and sort of get rid of a small, like millimeter chunk of the brain and that that helps ameliorate symptoms. There is clinical trials that people can enroll in, things that we’re involved in that are like stem cell replacement therapies. So somebody you actually get new neurons, stem cell neurons injected into your brain to replace the dead neurons. And then, like I said, a range of stuff in the pipeline that’s kind of experimental.
Harry Glorikian: But in reality, like I would part of me like in my research, I would bet that you’d make the case that deep brain stimulation and other forms of neuromodulation are more effective when they’re informed by deeper, richer data about the patients. And I’m not sure if that thesis is correct or not.
Brian Pepin: Yeah, I would say that thesis has been borne out for for a variety of reasons. So one, there’s a these patients these therapies are not one size fits all. There’s a lot of knobs that can be turned that really kind of customize it for each person. Right. And if you’re a busy clinician, how are you going know how to turn these knobs, right?
And so we provide a lot of data that kind of instructs how to turn these knobs in the right way so that the patients can can go home and really feel like they’re getting the most out of this investment that they’ve made in their own treatment. They’ve gone had brain surgery. They should feel really good, as good as possible. Right.
So that’s one I think the other side of it is who should be getting these therapies right. And so relative to the clinical outcomes, most even whether it’s Michael J. Fox or the clinician organizations, everybody sort of recognizes that something like deeper and deep brain stimulation is sort of under penetrated, like there’s way more people that can benefit from it than are getting it.
But it’s not always clear, like who should you prescribe it to? Right. So that’s another area where data can be really important, saying, hey, this is the phenotype of someone who is really going to benefit from this treatment.
Therefore, you, the clinician and you the patient should feel have some increased confidence to go get this therapy because we see that patients like you 95 plus percent of the time get these great outcomes.
Harry Glorikian: So yeah, yeah. I mean, most of the time, I mean, especially in all these neurological diseases, it’s just trying to who should be in the trial, like why, you know, finding that right person otherwise, you know, I think that’s why most of these trials fail is because you just don’t you’re not even picking the right population most of the time.
Brian Pepin: That’s a big reason. Yeah.
Harry Glorikian: But let’s back up here for a second. Like, I love to understand more of the whole story of the company. I mean, you founded One Labs back in 2018, if I’m not mistaken. I mean, how did that really come about? I mean, and also, if you could talk about limitations, shortcomings or failings that you saw in, you know, this whole area of data patients and brain disease.
Brian Pepin: Yeah. So I my kind of whole academic background kind of going back ten plus years is was was in this kind of blend of neuroscience and engineering and neuroscience and in particular like human neuroscience, like what’s going on in human brains has always been an area where I’ve spent a lot of time and done a lot of work before starting this company spent five years at at verily, when I joined, it wasn’t even called Verily. It was just like a group of ten people inside Google X and kind of later became Verily, right?
And I took initially a little bit of a detour from neuroscience and kind of did worked on a series of hardware and software platforms and diabetes and a little bit of work in immunooncology and got a few of like, oh, okay, here’s data informing and doing predictive things in medicine and here’s how it can work when it looks good and it had enough experience in in these neurological areas.
You know, that was not how medicine works in neurology. Right? Right. And then the second half of my career, verily, I got back a little bit closer to my roots and had the opportunity to start this neuroscience joint venture with some folks at GlaxoSmithKline called Galvani.
We were building a new basically a new class of therapeutics that through stimulation of peripheral, peripheral nerves in the gut could modulate immune disease. And they’re off and running in the clinical trials now. So you can kind of follow along with them. But kind of going through that, I came back to seeing all of these old problems.
Brian Pepin: When you’re looking at how do you develop and deliver neuroscience therapeutics versus like what’s going on in oncology? So you have you’re in a situation where in development, animal models don’t really recapitulate the human disease. So that’s problematic. And then on the other side, so you need human data.
You need human data to be able to do anything that is going to actually help you lower the risk of going into a clinical trial. And then human data hasn’t historically been available, so it’s a little bit of a chicken and the egg thing.
And then I think I also got a view of like because we were planning for how this was going to go to market and scale up like, oh, well, in order to deliver a complex therapy like this successfully, there has to be a data ecosystem around it. Otherwise, the immunologists, neurologist, whoever, they don’t have time to deal with all this like they need to be.
They need to be guided in the same way that if you think about a very complex therapy, like a CAR-T therapy and oncology, like there’s all of this data and support around, right? Mm hmm.
So I appreciate it. I was coming to appreciate that. And then I was kind of thinking about what I was going to do next after after we were kind of wrapping up this Galvani project. And it was kind of, how can I get back to my roots in the brain? And yeah, and just kind of saw this opportunity to bring together data sources, specifically starting in Parkinson’s that were becoming really prominent and being available at a reasonable scale.
Brian Pepin: Like we weren’t going, these were going to be thousands of patients. So things like the explosion of brain imaging that scans and structural imaging, things like the availability of high quality movement markers from the Apple Watch, for example, and then and then for example, our partnership with Medtronic, having these devices which have previously been sort of like dumb devices, just kind of providing stuff all of a sudden providing brain sensing because the electronic circuitry has evolved. Well, that’s a massive new interesting data source. Again, that’s at scale.
So just kind of saw the opportunity and figured the timing was right to to try it. And we’ve been growing the company ever since. We started with a focus on supporting clinical trials.
We launched sort of real world care ecosystem in Parkinson’s in August of last year, and we initially launched at UCSF. We expanded to a couple more universities in February, and now we’re sort of in rapid expansion mode across the US to health systems and clinics. Yeah. And then recently had this, we were able to kind of get the 510 K clearance for the data we’re bringing in from the watch.
So that’s that’s been really nice to have as well in terms of accelerating adoption and hospitals and also enabling some additional things that we could do in the context of clinical trials.
Harry Glorikian: Yeah, we’re going to get to the FDA approval shortly. But but just to give people context, so if I took Rune like out of the picture for a second. Yeah. What was the standard practice of brain data tracking? I mean, what are the existing tools you might find in a neurologist lab for collecting tracking EEG data or brain images of Parkinson’s patients, because I’m, I’m trying to get a sense of yeah, like how were we in the Dark Ages, or what are you trying to compete against?
Brian Pepin: Yeah. So if you’re, if you’re unlucky enough to have somebody with Parkinson’s that you know, you know that a typical Parkinson’s visit involves basically no data at all. Right. It’s a very tactile experience. You go in, the neurologist is sort of moving you around, assessing how rigid you are, how your gait is, how your tremor is. Just in that 20 minute window. Is that 20 minute window representative of your actual experience outside the clinic? Maybe. Probably not. But that’s that’s what they have to work with.
They might occasionally look at an image to make just a decision like, oh, okay, we see this image, we see that your your neuromodulation system was implanted correctly. And then they put that away and they never look at it again. They certainly don’t compare images across patients.
So it’s just that a lot of what we’ve built on the software side of our company is like, primarily software engineers, like out of the 65 plus people, it’s 30 ish software engineers, there’s a lot of infrastructure that we built to just automatically go grab that data, structure it.
You know, in these sort of layers, we have brain data, you have clinical data, then you have like the Apple Watch patient data. And then then just automatically surface those patterns so somebody doesn’t have to like go digging or make guesswork.
Brian Pepin: It’s like, okay, well, here’s the pattern. So here’s the pattern that tells you whether somebody’s dosing is off. Maybe they’re you want to switch from the standard version to the controlled release version.
Whether they might be a good candidate for deep brain stimulation. Whether that therapy is optimized. Whether they should be matched to a certain clinical trial based on their phenotype. Neurologists really haven’t had the tools to do much except for one size fits all treatments. We know you have this diagnosis. It’s been three years since your diagnosis. Therefore this is what we’re going to prescribe you. Right?
And so we’re giving them we’re giving them a much richer set of tools and again, really focusing on helping them see patterns which might otherwise be hard for them to see, that are now saying, hey, you can actually have a lot more context now to treat this patient, this patient with a Parkinson’s diagnosis here and differently than this patient with a Parkinson’s diagnosis two years in and get optimal outcomes that are different for each of them in a way that frankly isn’t really possible today outside of maybe one or two world class top, top centers.
Harry Glorikian: You know, it’s funny, I was laughing and I thought maybe I should be crying because like, if you think about where things, how things are done, right, it’s compared to the data analytics we have now. It sort of makes you like … we’re in 2022. What the hell is going on? I mean, we should I feel like we should be much farther forward than we are.
Brian Pepin: Yeah. I mean, I think the missing ingredient was the availability of data. I think it’s a fairly useful analogy to think that like kind of neurology is sort of where oncology was 15 years ago. And then, you know, on the back of the sort of genetics revolution, you know, there’s all this great stuff that was able to happen in oncology. Genetics hasn’t been directly that useful in treating neurology diseases yet, but there’s all this other data that’s available now that’s kind of exploding that we could take advantage of. Yeah.
Harry Glorikian: Yeah. I mean, look, I didn’t write any of my books or have this podcast before all this data started to become available because otherwise it was just on paper. But Let’s talk a little bit about your products StrivePD. This is a system that you’re building that’s designed to help people with Parkinson’s disease understand and track sort of their symptoms. It’s an iOS app that’s already out. I know you’ve already developed an Apple Watch app that we’ve talked about briefly.
And a wait list of patients who help you test it. Right. So if you had to go through and describe it, sort of what kinds of data can patients enter into the app? How does an Apple Watch add to the kinds of data that you can collect? I mean, go through some of these and if you if you miss any of them, I’ll throw in another question.
Brian Pepin: The goal is to bring a super rich dashboard into the clinic that the neurologist and the patient can kind of look at together. And again, it can kind of inform which medication, which trial, etc., etc.. Right. So to get there, we, we try to, we kind of work with patients, make sure they’re using our software in its intended way, especially two weeks before and two weeks after every clinical visit.
So we can build that kind of continuous picture of how their disease is working in conjunction with their medications, with their environment, with exercise, with everything else, and bring that fully informed view. While doing that, though, we really try to ask as little of folks as possible.
They have Parkinson’s. Often they’re trying to work. They’re trying to travel. WE don’t want to be a burden to them. So really what we ask them to do is, is do a few kind of daily interactions with the app to help us kind of again get that sort of qualitative picture of how things are going and a little bit more information about their medications.
We try to provide some utilities like a medication tracker in there, and we can leverage Apple’s medication tracker or sort of a built in one. And then we just encourage them to just wear the Apple Watch. That’s the great thing about it. They can use the Apple Watch as normal. They can receive messages, they can track exercise. But in the background, we can bring in all of this rich Parkinson’s movement data and some sleep data where relevant.
Brian Pepin: And that is just so helpful, especially in context with this other stuff and helping clinicians understand if your dosing is correct, if the timing between doses is correct. If you’re really like below or above a threshold of being well controlled where you might really benefit from a new therapy.
The Apple Watch has been super, super impactful for that and being able to predict whether or not somebody is going to benefit from one of these things or how somebody might benefit. So, yeah, I mean, from the patient, we want to make it reasonable. If you already have an Apple Watch and an iPhone, it’s really all you’re doing is downloading an app and spending a few minutes a day on it and that’s it. And then when you go into your next clinical visit, it should be a very different experience. Instead of having this go in and say, “Oh, well, how are you doing for the last six months?”
Like, and then being poked and prodded to figure out like it varies like, “Well, I can see that you’re having. Troublesome tremor at 9 a.m. to 10 a.m. most mornings. And talk to me about your medication schedule, tell me about this. Is this consistent?” And then it can kind of dial in and say, oh, actually, maybe you should be taking two pills in the morning instead of four, but then we should dial up your evening dose to get that down.
Or, you know, those kind of conversations can happen really, really quickly as opposed to maybe not happening at all in the context of a 20 to 25 minute clinical visit.
Harry Glorikian: So so why the Apple Watch? And I say that because, like I just recently spent time talking to the head of data sciences at WHOOP, right, where it’s 50 to 100 megabytes a day coming off this thing. You know why? Why this thing versus. Yeah.
Brian Pepin: Yeah. I mean, WHOOP is cool too. I think Apple Watch has like there are 8 million people over the age of 55 in the US that have an Apple Watch. So you have this advantage of a massive installed base of folks who are already using it. You have a ton of just kind of consumer acceptance and utility around the watch.
Apple is able to pour a ton of resources into development of watch and HealthKit. I mean, one of the we really look forward to WWDC, which is Apple’s big developer conference every year because we get to learn all of the new features that we just automatically get to upgrade into our platform. Because Apple’s been doing this work behind the scenes on sleep or gait metrics or whatever else, and that’s a huge benefit as well.
So but you know, we are integrating kind of mostly in the context of clinical trials with other wearables as well. But in terms of like that mass consumer, like getting to tens of thousands, hundreds of thousands of folks. So far, Apple Watch has felt like the right platform for us to be investing most of our time in.
Harry Glorikian: So. Last month. You got FDA clearance, so congratulations. That’s huge. I always ask the people on this cutting edge getting these clearance, like, what went into getting that?
Brian Pepin: Yeah, a lot of work. You know, I think there were kind of two elements of it. Whenever you’re going for one of these things, the quality of the science going in has to be good. And we’re fortunate that Apple had, along with these kind of collaborators that are on this this paper that they had published, had already done a fair amount of work validation.
And so we’re able to kind of have a good baseline that we could kind of continue to go from and have a really high quality level of scientific validation of like, hey, this algorithm is really good at detecting tremor and dyskinesia remotely in patients. So that’s that was like step one.
And then the other part of it was really kind of getting the FDA comfortable with here’s a quote unquote consumer device that now in this specific context, we’re turning into a medical device. And providing all of the right kind of infrastructure, whether it’s kind of testing, versioning control, all of this stuff that’s kind of happening between us and Apple.
And to make sure that as Apple’s device continues to evolve and get better as a consumer device, we can maintain super tight quality control on what clinicians see in terms of tremor and dyskinesia, for example.
So there is a lot of kind of back and forth making sure that we have that optimized, making sure we’re able to prove that test, that making sure that we’re able to continue to ensure that over time. But yeah, I mean, it was a process for sure, but I’ve definitely been through worse processes.
I think it’s one of these things where my sense is that regulators do overall recognize this is where things need to go in order to be able to enable high quality data at scale. And so there overall, I got the sense of they were trying to figure out how to make this work. They weren’t trying to figure out how to, like, slow it down or block it.
Harry Glorikian: I’m so glad you said that. I mean, you’re one of the first people that’s ever said that, you know, there’s worse things than than working with the regulators at the FDA, because I usually get people that are freaking out about it. But. But let me ask you a question. So if I’m not mistaken, the Apple Watch actually taps into something that Apple created itself, which is their movement disorder API.
Yeah, which was funny because until I started to do the, you know, digging in, I didn’t realize that was there. But how big of a deal was that for what you guys were tapping into for this?
Brian Pepin: Yeah. I mean, that was I think I told the story in one of the interviews I did around it. But so we were this is early on, a couple of years ago, we were we were developing this out. And Apple allows you to get raw accelerometer and gyroscope data off the watch.
And so we had we had kind of thought that’s all we’re going to be able to get. We’re going have to go develop our own metrics around this. And then I was digging around in the developer documentation online and I saw movement disorder toolkit. Tremor and dyskinesia. And I was like, Huh, that’s, that’s funny. And I, you know, I just kind of on a whim, I was like, because it was a little email “If you’re curious about this toolkit email here.”
So I emailed there and about 8 minutes later the team leader from Apple emailed me back. It’s like, Yeah, we should talk. I was like, Wow, this is weird. Okay, let’s go. Yeah. And and that was the beginning of us kind of figuring out about it. But I think it worked out really nicely because Apple, again, had done a lot of the legwork and they had basically have this API that people can pick up.
But we were in a better position to actually take that across the finish line in the context of our ecosystem and get it to something that could be FDA approved, if that makes sense.
So it was a good I think a good division of labor. And hopefully, hopefully now that we’ve kind of taken it across the finish line, other folks can benefit from the API as well. And there can be a whole community of folks that start to leverage that. But yeah, right now we’re sort of the first people to kind of take it across the sort of FDA finish line and be able to deliver it in the context that it can sort of improve somebody’s care.
Harry Glorikian: I’m sure that team over there is super excited.
Brian Pepin: Yeah. I think they’ve been they’ve been happy with the outcome so far, as far as I can tell. Yeah.
Harry Glorikian: Let’s pause the conversation for a minute to talk about one small but important thing you can do, to help keep the podcast going. And that’s leave a rating and a review for the show on Apple Podcasts.
All you have to do is open the Apple Podcasts app on your smartphone, search for The Harry Glorikian Show, and scroll down to the Ratings & Reviews section. Tap the stars to rate the show, and then tap the link that says Write a Review to leave your comments.
It’ll only take a minute, but you’ll be doing a lot to help other listeners discover the show.
And one more thing. If you like the interviews we do here on the show I know you’ll like my new book, The Future You: How Artificial Intelligence Can Help You Get Healthier, Stress Less, and Live Longer.
It’s a friendly and accessible tour of all the ways today’s information technologies are helping us diagnose diseases faster, treat them more precisely, and create personalized diet and exercise programs to prevent them in the first place.
The book is now available in print and ebook formats. Just go to Amazon or Barnes & Noble and search for The Future You by Harry Glorikian.
And now, back to the show.
Harry Glorikian: But so there’s Apple’s there and then, as you mentioned, I think you have a partnership with Medtronic, right. Which makes this deep brain stimulation called Percept, right?
And if my understanding is correct, based on what you said and what I read, which Percept directly records brain signals in people with Parkinson’s and also then deliver a delivers the therapeutic stimulation. Right. Can you explain what your partnership with Medtronic is about and what value your organization hopes to bring to the to that whole dynamic?
Brian Pepin: Yeah. So I think the simplest way I can probably explain this is the data, the brain sensing data that’s coming out of this Percept device, if that’s the only thing you had, it just kind of looks like a squiggly line, right? It’s really hard to take that and take that and directly make okay, I’m going to make this new clinical decision now. I’m going to I’m going to understand.
But if you can take that data and put it in context with this rich watch data, with all this other clinical data like this, then you can start to say, actually, when I see this type of brain signal, I know it means that this patient is going to, needs these stimulation parameters change in a certain way or needs their medication optimized or maybe needs a different completely different program. Right.
And that’s sort of the hope and promise of having this sort of smart implant. Right. Is that it can it can help guide therapy. And so the I think why this partnership with Medtronic works really well is we can bring this data together with all this context.
We can sort of hopefully make their therapy easier for clinicians to really achieve optimal outcomes for patients and in process give clinicians confidence that.
Like, Hey, when I implant the DBS system, if it’s a Medtronic DBS system and it has this brain sensing, I know I’m going to be able to optimize it for these patients. I know it’s going to they’re going to have really good outcomes.
Harry Glorikian: It’s funny though. When you say that, I think of a like at some point, if you have enough in there, like it’s going to be able to make that decision on its own.
Brian Pepin: I hope so. Yeah. I mean, I think if you think about what we’re building, really, it’s
a giant database of rich biomarkers, interventions and outcomes. And then so the the more of that you have, the more you can say, okay, if I see these biomarkers, I know you should have these interventions to get these outcomes. That’s yeah, it’s like what Flatiron, for example, has been able to build up in oncology.
Harry Glorikian: Right. Right. Now, I know you guys have been deeply focused in Parkinson’s, but I’m assuming you can apply what you learned in Parkinson’s to a lot of other neurological issues. I’m I’m going to guess, you know, multiple sclerosis, autism spectrum disorder, OCD, maybe Alzheimer’s. I think if I had to go out on a limb, I might say chronic pain, although I feel like that might be a little farther out. But, you know.
Brian Pepin: Yeah, yeah. I mean, as I was alluding to this earlier, I think the key for us is the underlying science has to be there, like there has to be something predictive and useful we can do with the data that’s available. And right now for us, it feels like the next area that’s going to be MS. And we’re running some kind of pilots in MS right now to see how that pans out.
If you had to kind of twist my arm right now, I would say that our second indication is probably going to be MS that we support. In addition to kind of expanding out, I almost think about there’s a there’s a group of patients or not patients even there’s a group of people that have what’s called prodromal Parkinson’s. Right. So they don’t have Parkinson’s diagnosis, but they’re likely to go on to develop Parkinson’s. I almost think about that.
That’s a different indication, but that’s a really, really important one because sort of the drugs of the future are not really going to be meant for folks that already have a PD diagnosis. They’re going to be neuroprotective, right?
You’re never going to get a PD diagnosis. That’s the goal. But then you’ve got to go find those folks with some high confidence. And that’s a that’s a fun problem, too, that we’re working on expanding into and that actually something like the Apple Watch, which again already has a big installed base, is going to be kind of an important platform to be able to kind of go and find those folks.
Harry Glorikian: So it’s funny because I’m I’m giving a lot of talk because of the new book and so forth. And I’m like, I can’t imagine that people aren’t going to have these things as early warning systems. So part of me says, I mean, do you think there’s a role for data analytics and wearable devices in early detection of, I know it’s Parkinson’s disease, but neurological issues? I mean I already have the Apple Watch on it’s just this app sitting on there that’s sort of monitoring.
Brian Pepin: I mean, I think the the important ingredient there is let’s say there’s an early detection, right? What do you do about it? What do you ,the person who just got detected, now that, you know, you have early, you might get Parkinson’s, what do you do about it?
And I think why Parkinson’s is exciting in that area right now is there are upcoming in the next 6 to 18 months neuroprotective trials that you can go enroll in if you’re likely to go on to develop PD that may slow your progression, may prevent you from ever progressing, right.
That’s something very tangible that you can do with that knowledge, right? There’s also some softer things like exercise more. Like exercise has been shown to be … good to know. Right. But just generally good advice I think. But so yeah, I think so. That’s that’s kind of key for me is I think about that is how can you pair… Like in cardiac right?
If you get early detection of cardiac stuff, there’s something you can do about it, right? You can go save your life essentially. And so I’m looking for opportunities in neurology, whether it’s Parkinson’s or MS, where if you get the early detection, there’s an intervention that you can go pursue that is going to hopefully meaningfully change your long term outcome.
Harry Glorikian: Yeah, I also think like a lot of people want to know so that they can plan plan their lives accordingly.
Brian Pepin: Yeah, maybe. Yeah.
Harry Glorikian: I hear most doctors say, well, I mean, there’s nothing I can do about it. I’m like, Well, but I want to plan my own life.
Brian Pepin: Yeah, yeah. That’s interesting. There’s the sort of it’s the pejorative kind of thing, but the movement disorder neurologists kind of “Diagnose and adios.” It’s like the sort of kind of cliche thing people say about neurologists, which is like, “Yeah, you have this. There’s not a lot we can do about it. Goodbye.”
So, fine but I’m excited about saying well, we can see this happening. Here’s what to do about it. And we know that there’s a high likelihood of success because we have the data to back that up.
Harry Glorikian: So I was, you know, again, I was, you know, rooting around and doing my research. But like more than most startups, I mean, you guys seem to have sort of laid out your a whole set of ethical values, right? I mean, first, let’s talk about the way you view patients.
You know, your your company says it wants to be a human centric company where you never forget that brains have owners. And that people with brain disease deserve empathy. I mean, how does that translate into design decisions, product development choices? What does it mean to build a human centric database, for example, or how do you view privacy, you know, those sorts of things.
Brian Pepin: Yeah. I mean, as you’re kind of alluding to, it sort of bleeds into everything, right? From the very first, from the way that we structured the consent agreement. It’s like it’s very plain worded language. We’re asking for the sort of minimum necessary that we need, giving patients a lot of flexibility.
To how we think about, again, kind of how do we sort of ask as little of folks as possible because we know, like, we’re not their main concern. Right. So how do we be respectful of that? To, how do we involve folks with Parkinson’s in the company, in the product development? And that’s that’s something that is not the most straightforward thing to do for a variety of reasons. Right. Like people with Parkinson’s, they have a challenging neurodegenerative disease.
They you know, they’re not just going to be like a normal employee at the company. Usually. They’re on medication that is wildly changing the dopamine contents of their brain. Right? So all kinds of things. But you have to, and I don’t know, we don’t have it perfect yet, but we’re always trying to get better at it. Like how do you set up systems to work with those people to bring them in to make sure that a wide range of those voices is heard, and that let that influence the design, let that influence the strategy.
Brian Pepin: We incorporated a patient advisory board formally in March, I think. We had been doing it informally for a while, but so having this real patient advisory board is great because we can meet with them on a regular quarterly basis, for example, around our kind of key result planning and do a real sanity check to make sure it’s like,
Hey, these are the key results the company is going after, you know, how would you how would you stack rank these as being meaningful to patients and make sure that that is lining up and that we’re not that we’re biasing towards tackling things.
They’re going to be really meaningful to the folks that were ultimately trying to benefit here and not doing things that are kind of like only indirectly, or benefiting folks way downstream or something like that. It’s also ultimately what makes just going to work every day more rewarding, because you’re interacting with the people that again are like, They’re using your software, you’re helping them, you’re seeing the benefit, right? And they’re telling you when it’s not, when it’s not working like expected too. And you’re getting that feedback.
But I have all these friends now that have Parkinson’s, and I am super motivated to help them out and to try and get the best possible outcomes for them.
Harry Glorikian: Now it seems like one that first thing I said is going to bleed into this next thing. You also have an unusual approach to management and team building, right? So all documents are shared with all employees at all, or most meetings are open to all employees. And you give employees a lot of individual responsibility. I mean, it sounds like a fairly flat organization if I had to frame it that way. I mean, how do you summarize your management philosophy? I mean.
Brian Pepin: Yeah, transparency and autonomy. I think we try to optimize for it. You know, I think it’s I one of the things that I think we’ve had to grow just kind of a growth thing is as you know, as we go from from 10 to 20 to now 65 plus, folks, you want to make sure that folks benefit from the transparency and autonomy, but don’t also feel the pressure to like know everything that’s going on. Right.
So you have to kind of make sure that everybody… We spend a lot of time on internal comms. Right? And just making sure. Here’s what we’re doing. Here’s why we’re doing it. Here are the patients who are helping. Here are the conditions we’re helping. And then just repeating, repeating, repeating.
And then every quarter as the goals, as we try to get more and more ambitious, making sure that it gets communicated. Being very transparent about that. After every board meeting, I do like an AMA, ask me anything with the company, share the board deck.
They can ask about where the company is going and everything from strategy to finances to kind of like in the weeds board discussions. Happy to share that kind of stuff. But yeah, I mean, I think it’s one of those things where, I mean, you have to be careful about hiring.
But ultimately you want to trust the people that are on your team and then have them trust you reciprocally. And that’s if you can do that, it’s it’s a super power, right? If you can’t do that, you need to have all this like rules and bureaucracy and stuff. And that makes things very slow. But if you can do that, you can you can be very, very responsive to feedback from customers and feedback from from patients and clinicians.
And, you know, that’s the that’s the beauty of running a startup, right? Is you can you can respond very quickly and move very quickly. But I think the foundation of that is trust. For me, the foundation of trust is transparency and autonomy.
Harry Glorikian: So now lets you say something right on the front of your website or on the website that says potentially discovering new biomarkers for disease progression. And I think the term is digital biomarker, which of course I love that term. Are there any examples that you can share that you guys have might have like, you know, not giving away like trade secrets or anything but that you guys have sort of identified as unique.
Brian Pepin: Yeah, we’re, we’re with a few publications coming in September, but I don’t really want to scoop, but we’re republishing it both. We’re publishing at Movement Disorder Society and also the World Society for Functional Neurosurgery. So those are two big conferences that are important to the Parkinson’s kind of clinical community. But I’ll say generally it’s we’ve been able to do more as two things have been become more true.
So one is obviously keeping patients on the platform for longer. So they’re going to come back visit after visit after visit. So you can actually see some trajectory. That’s super important. And as we’ve started to have patients on the platform now for a couple of years, we’re able to say more interesting things there.
The other thing that’s important that we’ve been able to do more is make sure that we’re, as much as possible, we’re bringing that data that reads directly on what’s happening in the brain and can be related to underlying mechanisms. So it’s one thing to have like a lot of continuous symptom data. You can show that symptom data evolves over time.
It’s a completely different thing to have that plus say, hey, and by the way, this is how the underlying brain network is shifting over time. Or this is how this blood based biomarker of mitochondrial dysfunction in the brain is evolving over time. And it’s really making that link that allows you to say something that’s not, “Oh, here’s how the symptoms are progressing. And we think this is indicative of disease progression.” It’s actually saying, “Hey, we think this is actually a disease progression signal here.
We can see it being clinically meaningful, but also we can say that actually the brain structure or the brain chemistry is also changing over time in a way that correlates.” If that makes sense. So, you know, it’s early days, obviously. I think these publications are a good first step, but it’s you know, it’s where we got to go.
So we’ve got to get more patients, keep them on for longer and then get more of this underlying “what is really going on in the brain” data linked with the kind of continuous symptom data that allows you to quantify what’s going on with folks.
Harry Glorikian: Yeah, yeah, I can I can see the, the value of this, you know, growing in orders of magnitude as you understand other diseases and those the changes that are… Because there’s I’m sure there’s going to be massive overlap in some diseases.
Brian Pepin: Oh yes, yeah, yeah, yeah, yeah. I mean there’s co-pathologies, there’s underlying networks that are highly correlated in AD and PD.
Harry Glorikian: Yeah. So I want to circle back and ask you a different frame for understanding what what you’re trying to do. I read an opinion piece in that you wrote in Med City News where you argue about precision neurology. I mean, and I’ve been talking about precision medicine, I feel like forever. What do you mean by that? I mean, how close is the analogy from other forms of precision medicine like oncology? And if you if you want to do precision neurology right, what kinds of data do we need?
Brian Pepin: Yeah. So I think for me the, where something transitions over to something you can reasonably call precision medicine is when you can use data to make a prediction that is going to be clinically meaningful. Whether that prediction is how a therapy should be configured or prescribed or dosed or again, trial matching or whatever it is.
It’s data being predictive in the context of an indication. And so when I think about what that looks like in Parkinson’s, I can tell you that the two data sources that have been the most predictive by far, up until now at least, have been the continuous Apple Watch symptom data and the electrophysiology, the invasive brain sensing that’s coming out of the implants. I am extremely hopeful and I think we have good reason to believe that we’ll find other data sources that are going to be predictive as we scale up and collect more of it.
But right now, those two are the areas where we’re able to take that data today and put it in context and come to conclusions and say, hey, like this, this is indicative of a pattern which predicts that this patient needs their stimulation reduced or needs their medications changed or should actually be enrolled in this clinical trial, that type of thing.
So yeah, there’s going to be more. There’s going to be more. But I think one of the issues with doing this in a disease like Parkinson’s — super heterogeneous, super time bearing, you need scale, right? That’s why we’re focused on scaling to 50 plus hospitals by the end of the year, representing a population of 10,000 plus patients. Like you need that big bulk of folks so that you can really see the signal in the noise and pull out the individual patterns.
Harry Glorikian: Well, okay. So assuming that that happens, if I just follow every other disease I’ve been watching over time, the more precise we get is. Do you think, based on what you’ve seen so far, that you’re going to see multiple etiologies that need to be managed. Because it’s like saying breast cancer back in the day.
Brian Pepin: I would be the first person to say that. I mean, that’s like the whenever you go to any of these any PD conference, right, whether it’s a scientific conference or a clinical conference, that’s sort of like the whispers around the lunch table is like, hey, we’re not really dealing with one disease here. Like we’re dealing with these different co-pathologies.
You’re dealing with PD 1, PD 2, PD 3. And everybody is trying to think about how do we actually evolve to recognize that and treat these separately. And that’s, you know, hopefully we’re a meaningful part of that journey with with Rune Labs and making that making that a reality. Because exactly as with oncology, it’s a step function change in your ability to treat those folks to develop new therapies for those folks and really to understand the neuroscience of the disease. Right. Unlock the potential to reach new targets.
Harry Glorikian: Yeah. No. Well, I mean, I still think like at some point it’d be great to have an early warning system.
Brian Pepin: I’ll make sure that you when we set up this this initial pipeline trial, I’ll make sure that you get an invite to enroll.
Harry Glorikian: That sounds great. Well, it was great talking to you. You know, I, I can only wish you incredible success because, I mean, the changes or the advances you guys are making are going to be meaningful to a patient population that really needs that help.
Brian Pepin: Oh, thanks Harry. We hope so. Yeah, we’re we’re pushing on it.
Harry Glorikian: Excellent. Thank you.
Brian Pepin: All right. Thanks, Harry.
Harry Glorikian: That’s it for this week’s episode.
You can find a full transcript of this episode as well as the full archive of episodes of The Harry Glorikian Show and MoneyBall Medicine at our website. Go to glorikian.com and click on the tab Podcasts.
I’d like to thank our listeners for boosting The Harry Glorikian Show into the top three percent of global podcasts.
If you want to be sure to get every new episode of the show automatically, be sure to open Apple Podcasts or your favorite podcast player and hit follow or subscribe.
Don’t forget to leave us a rating and review on Apple Podcasts.
And we always love to hear from listeners on Twitter, where you can find me at hglorikian.
Thanks for listening, stay healthy, and be sure to tune in two weeks from now for our next interview.
For more episodes of The Harry Glorikian Show, visit our podcast page.