Latest / StartUp Health NOW Podcast / How GPx is turning a smartwatch into an early warning system for heart failure

Transcript

How GPx is turning a smartwatch into an early warning system for heart failure · StartUp Health NOW Podcast

• 0:00
  [instrumental music] Yeah. Essentially, patients are in this, like, black box. They don't know how they're doing. The doctors don't even know how they're doing.
• 0:10
  If there were something like a CGM, like a continuous glucose monitoring, then they would know exactly what to do and adjust their insulin level. But that doesn't exist in the world of heart failure. Okay.
• 0:20
  So essentially we are, you know, a CGM for heart failure, but non-invasive, just using a wearable device. [upbeat music] Sean Matsuoka from General Prognostics, GPX. Thank you for joining me at Startup Health.
• 0:38
  Thank you so much for having me, Logan. We like to do these updates with our companies, these health moonshot updates, to understand where you're at. The world of healthcare startups, things move fast.
• 0:48
  I'm fascinated by what you're building, this algorithmic approach to biomarkers, to understanding the health of the human body, where we're at, what interventions are needed.
• 0:57
  So I want you to start by just giving me an overview of General Prognostics. When did you start GPX? When did this idea come about? Yeah. So we started the company back in 2020 at the cusp of COVID. Okay.
• 1:10
  Alongside with my two other f- co-founders, we went on this journey about identifying the right time for the patients to, specifically heart failure patients, to get treated.
• 1:20
  There were many advancements in technology, so we specifically we look at wearable devices. And we had this idea about identifying the right time for these patients to be treated. And since then we started the GPX.
• 1:33
  We started a clinical study. We had about more than 300 patients participate in the clinical study. Each of them wore a smartwatch. We don't manufacture this. We just take off-the-shelf wearable devices.
• 1:44
  And we captured heart rate, heart rate variability, activity, sleep. And then every other week, we asked them to do a vital blood draw.
• 1:52
  And we look at the correlation between the parameters, the phys- physiological data from the wearable device, and how that can identify or predict worsening biomarker, define worsening events.
• 2:04
  What are the advantages in your mind to having this algorithm approach versus the alternatives? Yeah.
• 2:10
  The only other alternatives today are either very primitive in a way that people step on the weight scale and see how much weight they've gained.
• 2:17
  Or the other extreme is to go through surgery, and you can have an implantable device that goes inside your heart and monitor the pressure. There's nothing in between.
• 2:27
  So the benefit of being able to identify how you're fairing noninvasively and continuously with the use of a smartwatch is the greatest thing.
• 2:36
  Help normal viewers understand how this differs from just using a smartwatch and just maybe an app they might have that pulls in data and gives them a consumer level of feedback. Yeah.
• 2:47
  No, smartwatches are great, but the signals are very noisy. Okay. For example, doctors, they'll straight out tell you that they don't wanna see any more data. They wanna see something more actionable.
• 2:56
  Unfortunately, while smartwatches are great, it's hard for them to distinguish whether the patients are going through exacerbation or just going up the stairs. It's great, but it's very noisy.
• 3:08
  But the fact that we correlate to a vital blood biomarker, the language that the doctors already trust, whenever you go see your doctor, I mean, I get my blood drawn and present that results to the doctor.
• 3:18
  This is, you know, they go by, "No, this is the gold standard." So being able to say rather than coming up with some random composite risk score, we say your patient's NT-proBNP.
• 3:28
  This is the class one recommendation in the guideline. This is the vital blood biomarker. It's a hormone released by the heart when the patient are going through exacerbation. That can be detected in blood.
• 3:38
  So when our algorithm triggers, that means that the patient has elevated NT-proBNP. Okay. And the doctors would know exactly what to do. Okay. How did you get into this work?
• 3:47
  Are you coming from the cardiology side or the technical side or somewhere else? I give credit to my co-founders. They previously worked at Medtronic.
• 3:55
  And also my other co-founder, he's into in vitro diagnostics, especially in blood. So it's a combination of digital and chemistry. So yeah, it's a very unique combination. Nice.
• 4:05
  What was it like starting your company at the dawn of COVID? How did that shape the way that you structured the company itself? Yeah. So it had a, a very big impact on, on how, you know, GPX was formed.
• 4:18
  You know, we heard that people weren't able to go to the, see their doctor, and they were going through worsening events. So this was, at that point, was my-- I felt that was my calling.
• 4:28
  I personally lost my grandmother to heart failure. Mm-hmm. So it's something near and dear to me. This is the time that I needed to step up. And it was almost like the three of us got together and had this idea.
• 4:39
  But from there on, it was just, I caught this fever that I had to do something, and I guess the rest is history.
• 4:46
  As a way of helping other people understand this, exactly what you're building, describe to me a world three to five years from now if this technology really takes off, it gets adopted, the types of scenarios that will be triggered, and the outcomes that that will, will lead to.
• 5:03
  Yeah. So once we have FDA clearance, by the way, our, our algorithm is performing at a, at a very promising level, almost as good as some of the implantable devices.
• 5:14
  So, you know, this is, has a very good performing algorithm that can pr- predict heart failure worsening, or I should say biomarker defined worsening, which ha- is, is an early indicator of these medical events.
• 5:27
  So, you know, the median right now is about 14 days.
• 5:30
  It gives the plenty of lead time for the doctors to, you know, call the patients in for an outpatient visit, titrate their medication, run a panel of, you know, blood tests, and titrate their medication to ultimately keep them out of the hospital.
• 5:43
  Our aspiration is to scale globally. We're not talking about a $30,000 implantable device. This is just a off-the-shelf smartwatch that's very scalable. So we wanna make sure that we help patients around the globe.
• 5:56
  Is it all about the algorithm or where do the actual sensors and sensor technology play into this? I was talking to a founder the other day who said, "You can't software your way to a better sensor."
• 6:08
  So how important are the sensors compared with the algorithm? Yeah, the sensors are very important. But we intentionally make our algorithm or device agnostic. So we don't rely on one vendor or one manufacturer.
• 6:23
  We just take the common platform, which is the PPG sensor that captures heart rate, or maybe I think that your Apple Watch can capture. But while we are device agnostic, these sensors are making great progress.
• 6:34
  That only makes our algorithm so much more better. Okay. Is it challenging to build a moat around your algorithm or not? Well, a lot of people will say, like, "Why can't other people do this?"
• 6:47
  Sure, you know, people can copy what we do. It's essentially a wearable device, and you have people take biweekly blood draws. But operationally, it's been very difficult.
• 6:57
  A lot of these heart failure patients, they don't bleed well. So it's been very difficult for us to capture these blood test results, and then to have them do on a certain cadence, right?
• 7:08
  Every other week, ask them to do a blood test, just leave it at the front porch, have FedEx come and pick it up. It sounds very easy. But then have the lab run these results, return it back to us within 24 hours.
• 7:21
  It sounds very trivial. Yeah. But operationally, it's very sophisticated. I wouldn't say it sounds very trivial.
• 7:26
  But what I hear you saying is part of the moat is the challenge of clinical trials, and coming to the physician with the data that they will respect. Yes. And say, "Hey, w-we, we did the work.
• 7:37
  We're not just some app on your Apple Watch. We did the work." That's what's challenging about it. That makes sense. And so you're on this whole journey to continue down that road with the FDA, correct? Mm-hmm. Yes.
• 7:49
  Okay. Yeah. Yeah. Very interesting. I still want to put sort of a patient, sort of a, a more human face on this. I'm thinking into the future and thinking about what is the ideal use case. Forget the technology.
• 8:03
  We're just thinking about a person who might has a, have a fear of heart failure. What's their experience of GPX? How is their life different because you've succeeded? Yeah.
• 8:14
  So all the patient has to do is just slap on a watch and then forget about the rest.
• 8:20
  And then from there on, the technology will monitor the patients and let the doctors, the healthcare professionals know that they're going through worsening events.
• 8:28
  From a patient standpoint, you know, I dealt with my grandmother. We would care for her, see her.
• 8:34
  The same day she ends up in the ER, and we always scratch our head and ask ourselves like, "Well, what is it that we did, we could have done better?"
• 8:40
  Even worse was finding out my grandmother always lived in this constant fear of not knowing when she has to go back to the hospital and not knowing when she can be ever discharged.
• 8:49
  So just having this smartwatch on her wrist gives her comfort, gives her transparency in how she's faring. And what kinds of alerts at a basic level are people getting because of the algorithm? Yeah.
• 9:03
  So the, the algorithm would identify, you know, 100% increase from the baseline of NT-proBNP or 600 picograms per milliliter.
• 9:11
  The doctors would receive that notification, and the healthcare professionals would reach out to the patient and then have them schedule an outpatient visit, not an ER visit, so that they can run a panel of tests and make sure that they're okay.
• 9:25
  Basically finding out that you are at increased risk of a heart attack? Yes. Of heart failure, yes. Of heart failure. Okay. Yeah. Heart failure event. Your risks are going up. Let's look at it. Let's talk.
• 9:36
  And then they would come in for potential intervention, medication, whatever that might be. Yeah. Essentially, patients are in this, like, black box. They don't know how they're doing.
• 9:46
  The doctors don't even know how they're doing. If there were something like a CGM, like a continuous glucose monitoring, then they would know exactly what to do and adjust their insulin level.
• 9:56
  But that doesn't exist in the world of heart failure. Okay. So essentially we are, you know, a CGM for heart failure, but non-invasive, just using a wearable device. Okay. Is that language that resonates in the market?
• 10:07
  Are you using that with folks right now? Yeah. So, you know, it makes it easier for people to understand it's continuous. Yeah. The part that confuses people is that with the CGM, it does have a little bit of needle.
• 10:18
  Yeah. We don't require any needles whatsoever. Yeah.
• 10:21
  And I feel like, you know, if you were to take that instead of continuous glucose, continuous heart monitor, people have ideas of what that means, whether it's just a basic wearable with no algorithm- Mm-hmm...
• 10:31
  non-clinical data. Like, how big of a role is education in your strategy? Because I feel like there's a lot of confusion about what continuous heart monitoring would mean. Yeah. There, there's a lot of...
• 10:45
  Well, doctors, you know, if we tell them, like, you know, we can capture NT-proBNP, again, this biomarker noninvasively and continuously with, just with the use of smartwatch, they get it. They get it.
• 10:57
  They, they love it. Yeah. Okay. They understand. It's the potential users in your model who pays for this?
• 11:04
  This would be covered under the remote patient monitoring or chronic care management, CCM, codes that already exist today for Medicare, Medicaid, and VA patients.
• 11:12
  But once we have more data, then we'll be going after the private payers, the Medicare Advantage, to seek reimbursement. But the great thing is we can tap on the existing CPT codes.
• 11:22
  Entirely prescribed or could it be purchased by a consumer? Yeah. So we're not going B2C. This will be B2B2C. Okay.
• 11:30
  So this will be, uh, the doctors would prescribe our algorithm alongside with the wearable device, and the patient will take it home, and then in turn, the hospital would seek reimbursement by using the existing CPT codes.
• 11:42
  Last question. Sometimes something I like to ask the founders is if anyone in the world could be watching this and potentially learning and responding, what would be the ideal response that you'd like to receive?
• 11:55
  What do you need right now? What kind of collaboration and support would be ideal at this juncture?
• 12:01
  So first, I'd love for the patients to know that there's something in the horizon that can potentially help them get treated better, to get treated earlier and keep them out of the hospital.
• 12:13
  That's one group of people that I love for, uh, them to know what we're doing. Also, we need the support of everybody, so cardiologists, the...
• 12:21
  Partners that are manually contacting heart failure patients today to see how they're faring. It's, it's, those are the people that can greatly benefit from our, from our technology.
• 12:31
  Instead of calling them one by one, we would augment the high-risk patients and take care, let them know instead of making these individual phone calls.
• 12:39
  So I think that's another segment of people that we love to reach out to.
• 12:43
  And then potential investors that are looking for something that is exciting, especially with the advent and progression and evolution of smartwatches that can be used in a more clinical setting.
• 12:53
  That's just revolutionary. So anybody, any investors that are looking for such startups, we'd love to connect. If things go forward along your ideal timeline, when could this be in the hands of doctors and patients?
• 13:06
  So we'll be going to the FDA soon, and by late twenty twenty-seven, early twenty twenty-eight, it should be in market. Shawn, thank you for the update.
• 13:14
  It will be exciting to see how this period of clinical trials and testing in the world and working with the FDA, exciting to see how that goes and how soon this can be in the hands of folks who are worried about heart failure, worried about heart disease.
• 13:28
  There's a lot of people who you could bring some peace to who are in that situation like your family that you dealt with. So keep up the good work, and appreciate you coming in for an update. Thank you so much, Logan.
• 13:40
  Take care.
• 13:42
  [upbeat music] And finally, an invitation to all of the extraordinary health transformers, whether you are a startup, someone building the future of health, or whether you are an investor, funder, or buyer of the solutions out there in the marketplace, this is an invitation to you to join the Startup Health member community.
• 14:04
  Learn more at startuphealth.com. Look forward to hopefully seeing you soon. [gentle music]