+ Click for EPISODE TRANSCRIPT
DISCLAIMER:
This text is the output of AI based transcribing from an audio recording. Although the transcription is largely accurate, in some cases it is incomplete or inaccurate due to inaudible passages or transcription errors and should not be treated as an authoritative record. Nothing that you read here constitutes advice medical or otherwise. Always consult with a healthcare professional before making changes to a healthcare plan.
Scott Benner 0:00
Hello friends, and welcome to episode 934 of the Juicebox Podcast.
On today's episode of The Juicebox Podcast Ed Damiano from beta bionics is here to talk about the iLet bionic pancreas. Edie and I had an almost two hour long conversation about islet. I got in a ton of listener questions. Edie told me all about the company, how things started, where it is now when he expects people to be holding an eyelet and so much more. While you're listening. Please remember that nothing you hear on the Juicebox Podcast should be considered advice, medical or otherwise, always consult a physician before making any changes to your healthcare plan, or becoming bold with insulin. Here are three quick ways you can save money. Your first month of online therapy betterhelp.com forward slash juice box Use the link to save 10% off that first month. The offer code juice box at checkout at cozy earth.com will save you 35% off your entire order. And if you want to try ag one, go to athletic greens.com forward slash juice box when you do you'll get five free travel packs and a free year's supply of vitamin D with your first order
this episode of The Juicebox Podcast is sponsored by us med us med is the place where my daughter gets her diabetes supplies from and you can to go to us med.com forward slash juice box or call 888-721-1514 To get your free benefits check us med always provides 90 days worth of supplies. And they have fast and free shipping. They carry everything from insulin pumps to CGM diabetes testing supplies you want the libre to the libre three, the Dexcom G six or Dexcom G seven US med has it. You want Omnipod five you want Omnipod dash us med has that too. They have tandem T slim x two. Oh my goodness, they have it if you're looking for it, US med.com forward slash juicebox. Before it comes on, let me tell you two things. There are some ads that are in this episode. But I'm not going to put them in the conversation there at the end. So if you want to hear them, please hang out. And listen, I appreciate that very much. I want to remind you to go to the private Facebook group Juicebox Podcast type one diabetes head in there. There's 40,000 active members. It is the most lively and lovely diabetes Facebook group known to man. That's my opinion. If you're looking for the bold beginnings, diabetes, Pro Tip series, type two diabetes Pro Tip series defining thyroid, all of the things that people who listen to the podcast love. If you can't find them in your podcast app, go to juicebox podcast.com. There's a menu at the top, you'll be able to find everything there. Or if you're in the private Facebook group hit the feature tab at the top of the group. If you have a question about diabetes, or autoimmune issues in general, we've covered it on the Juicebox Podcast. Welcome back, even though you don't remember ever being on the show. I barely remember talking to you, but you were definitely on this podcast in the first year of it. I cannot find in the list anywhere like what would you have been calling it eyelet back then?
Ed Damiano 3:34
No, no. So if you think it was 2015, we definitely had give my wife was the one who named it the eyelet. I called it the bionic pancreas before we had the name islet and she came up with the name pilot. And I think that happened in 2013 is my guess. But we weren't you know, we were using both terms sort of interchange, there was no company in 2013 or even 2014.
Scott Benner 3:58
Wow. Well, so then how did I? Well, how did this all start? I'm assuming you or your child has diabetes, right?
Ed Damiano 4:05
Yeah. So that is how I got involved with this. So my, my background is an applied mathematics. And specifically I do what I would describe as mathematical biology. That's what I used to study. And what does that mean, I would you know, I was looking at mathematical models of how the inner ear works and fluid structure interactions that happen when you move your head through space, the vestibular system and your sense of balance and equilibrium. Understanding the underlying mechanics of that was very much a mathematics problem. And I spent some time working on that, I got very interested in blood flow in the very smallest micro vessels in the microcirculation blood flow through capillaries and understanding various important physiological phenomena that are connected to blood flow in micro vessels. And that became a big part of my research effort. And that's kind of what I did it was extremely theoretical, not not, not the least bit Practical, you know, I would write published papers in the Journal of fluid mechanics and, you know, PNAS and things like that. And, you know, there were three people who read the papers, and I was two of them. You know, that's what I used to say. So it was very, very arcane stuff and, and I enjoyed it very much. And certainly I could make a bit make build a career around it. But when my son developed type one in infancy, he was 11 months old, it became pretty clear that, you know, I had some basic skill sets that I could lead to the problem of building a device that could control blood sugar levels. And I had a student at the time, Feroz Alkhateeb, who was he was a PA, he was a master student in my lab, and he just come over to the US and he was doing some work in my area of blood flow, I had him working on a problem that he wasn't much interested in, frankly. And he'd finished finished his master's thesis, and he's sort of looking around for something else. And just five months after Ross arrived, David develop type one he was, as I said, 11 months old, my wife's a pediatrician. And she actually made the diagnosis. And I turned to fear also a year or so later saying, Listen, I've been thinking about a device that could deliver insulin and glucagon to automatically regulate glucose and people with type one. And at the heart of that is software is smart, intelligent software that determines how much insulin and glucagon to deliver, to get good glucose control based on a continuous stream of data from something like a continuous glucose monitor that didn't really exist in 2000, or 2001. Or two, there was a gluco watch, you may recall the sickness gluco watch. That was in 2000. And that didn't work particularly well, but it did get FDA approved. And so I envisioned that we would build the software that makes those those dosing decisions. And he got really interested in that as such, offered that up as a potential PhD project, and that became the summon substance of his PhD. And that was around 2002. So a little over 20 years, we started thinking about what that centerpiece, you know, technology would do, how it might work. And he started working on developing the mathematical algorithms. Initially, it was a single algorithm now we have three separate algorithms that run in parallel on the island. And that was at that time, he and I were at the University of Illinois, I was a professor of mechanical engineering, and this is in Urbana Champaign, and he was my PhD student. He finishes his PhD under under, in my lab at at ui UC Illinois and then I went to Boston University in 2004, took a faculty position in in Biomedical Engineering, Feroz, came over to Boston and did a postdoc in my lab. It's this fear still works with us today at beta bionics. So he's our VP of research innovation. And, and so he stuck with this project from the very beginning, we came to BU and started animal studies looking at glucose control with the algorithms that he'd been developing for his PhD in pigs that we could induce diabetes like pathology in and we could test the system in, you know, basically pigs with diabetes. We did that for about three or four years. And then I met Stephen Russell, my clinical collaborator of many years now 17 years, I think, in 2006 at MGH at Joslin diabetes Center, he was doing the fellowship he was doing as a postdoc fellow there. And so we started collaborating on bringing that system to human trials. And by 2008, we'd started clinical trials at the Massachusetts General Hospital Clinical Research Center in the inpatient setting.
Scott Benner 8:33
While you've been talking, it occurred to me, I don't think you've been on the podcast.
Ed Damiano 8:38
I mean, I totally believe you if you said it either way. I
Scott Benner 8:42
like something about when you said Boston, it hit it hit me and I researched. I researched not research, but read dash search my my blog just now. Yeah. And in 2016, January, a woman named Kelly was on to talk about being in your trial. That's what this is. Yeah, that's
Ed Damiano 9:03
much later and
Scott Benner 9:04
your name like, rings my bell because of that. And obviously, I know who you are, like, you know, we've never met before. I know your face. You and I have never met before. And if somebody asked me to describe you, I could do it. Right. I'm not aware of who you are. But yeah, I think that's what I was thinking of. Wow, that was Yeah, no kidding. Okay.
Ed Damiano 9:23
I mean, I've done a lot of interviews, as you might imagine. And so I could easily be convinced that I given an interview to just about anybody
Scott Benner 9:29
I want to be, I want to be completely honest, there are lovely people who helped me with the Facebook group. And if someone asks a question, I'll say, I don't know Have I ever said that and somebody else will have to come in and tell me if I said it, or I'm beyond being upset or or embarrassed by that. So that's how you get this whole thing going. That's fascinating, but my daughter was two and she was diagnosed like right after her second birthday. I do not meet a lot of people whose kids were younger than that. Usually.
Ed Damiano 9:56
No, it's extremely rare. I mean, I think it's, you know, probably Just a handful of people who are diagnosed under the age of one with actually with type one, you have some kids who get this congenital, this congenital, this neonatal diabetes. That is something it's exceedingly rare but it happens and it's often misdiagnosed as type one. We actually thought David might have that but neonatal usually see that around six months of age and he was around 11 months, so he was actually old for neonatal. We did some testing and it was pretty clear that he doesn't have neonatal as type one. But he's been on an insulin pump since 13 months of age. Wow. Which
Scott Benner 10:33
one? Did he have them? Mini med fiber? Wait.
Ed Damiano 10:37
We used his old mini med pump in the big studies. So once he graduated to the Animus we took his mini med pump and put it into big studies. Yeah, about that. Do you have other children? Yeah, my, my daughter is two years older than David. So she's 26. Now David's 24 And she's about she just finished grad school and she's gonna go into another grad program in the fall.
Scott Benner 10:58
Wow. Any other autoimmune your family? Yeah,
Ed Damiano 11:01
Emily herself. She has celiac disease. Okay. So that was diagnosed. Toby was doing a workup on her just for short stature. And she thought maybe she should be able to taller. And she, she ended up having celiac that's sort of in the, in the, in the panel when you look at look at that kind of thing. So she, she was around 12 or so when we figured
Scott Benner 11:23
that out. My daughter was the smallest person in her school. And we figured out she had hypothyroidism and she's 18 now and she's 570
Ed Damiano 11:31
Good for you. So she did just fine. My wife's five, seven.
Scott Benner 11:34
We went to good height. Yeah, but we got it. I mean, I'll never know what would happen if it wouldn't have got caught but it was hard not to catch. She was like basically passing out like asleep. She had no like energy and no energy at all. Yeah. But it was just really something like, she was the tiniest little person. And then now she's just isn't it's really something else what the right, the right thing can do to help you. Alright, so that's got to be enough. Everybody's like, just ask the question, Scott.
Ed Damiano 12:07
Well, we certainly have a kinship there with a very young person and you know, to watch it through infancy, diabetes progress through infancy and toddlerhood and, and, you know, preschool age, and then the school aged kid and the teenager and now the young adult, to see that whole arc pass before me over the past 20 years is quite amazing thing.
Scott Benner 12:26
It gives you a perspective that a lot of people don't have either you can kind of really step back sometimes and see all of the different impacts that I think get lost on people from time to time. Yeah, yeah. It's, it really is held been a hell of a journey. Okay. So, modern day a few weeks ago, I was speaking at an event, I met Stephen Russell, right. He works at UC did. And he was like, you know, spilling the beans that you guys were about to get an FDA approval. So tell me about that process a little bit. So once you say we've got a thing, it works. This is the thing we want the FDA to say yes to? Like, when was that? How long? Have you been at that part?
Ed Damiano 13:07
Great question. So I think it'll surprise you. In order to be really ready to submit an application to get market approval or market clearances, it's called by the FDA for this device, you have to have a clinical data set that is collected in a pivotal trial. And what that means is that you you conduct a study that you design with the FDA, you work with the FDA to design what that trial looks like. And it has to you have to capture data in such a way that you're you've got a good quality system wrapped around it and you're pulling the data. Together, we use the Job Center for Health Research as the contract research organization to help us put that package together. And then you build a clinical clinical study report at the end of the trial. And the Job Center puts this together you know, we have input into that that document, but ultimately, once the clinical study reports are written that has all the data that was captured from that pivotal trial, and all the other testing for the device is done, which is enormous amount of work as you're building the system from scratch the eyelet is in fact device it's built at beta bionics not by contract manufacturing, we build in our own facility in Irvine, California. And building a device is a non trivial task, a durable medical a piece of durable medical equipment. And so it has its own quality system wrapped around it. We have a manufacturing process at debated bionics to build the system, and then in undergoes an enormous amount of testing all kinds of tests that were done for insulin pumps also had to be done for a bionic pancreas, along with this clinical data set that was captured to this very large pivotal trial, you pull all that together in a document that is 10s of 1000s of pages long, literally. And we submit that to the FDA. So the clinical trial, the data needed for that trial was collected was was basically in hand in December of 2021. Okay, all right, so the trial mode the trial the substitute trial happened to January in October of 2021 and the jibs And it worked really quickly to lock the database after the last participant last visit in October of 21. And within really, essentially, within two months, they had the readout of the of the primary outcome analysis that we were, we were waiting for. We were very pleased with the results. And then went the process of building the clinical study report into the early spring, early part of 2022. Okay, and so by 2022, we were 20, March of 22, we submit the application, and we got clearance in May of 23. So 14 months later,
Scott Benner 15:31
I'm gonna forget the I don't know the terminology. But were you able to claim that your device was similar to another device? Or did you have to start from scratch?
Ed Damiano 15:38
No. So we used I hope you're not getting this guy doing some yard work here. I'm hoping not getting that. Hopefully you're editing concat do something magical.
Scott Benner 15:49
Is it seems okay. Yeah, I think all right, microphone I made them send you is very close to your face. So
Ed Damiano 15:55
yeah, I think it's working well. And I've got the headset, everything I can hear up as long as you can't. That's great. So yes, the the hybrid closed loop systems on the market today, we could use one of those as the predicate device to as they call it to our application. And so the FDA suggested that we use that as the predicate device. And we went forward with that, that submission with that in mind.
Scott Benner 16:17
So are you able to get the device okayed, and the processor created at this, then the algorithm at the same time? Or did you have to make the pump, prove the pump work, and then step forward and do the next piece?
Ed Damiano 16:30
No, in fact, we had no intention of ever building an insulin pump. And I really want to make it clear that the bionic pancreas isn't an insulin pump. And really, and that's not just that's not just semantics, it's really not, there is no way to program Basal rates, carbs, and some ratios or correction factors on the island. There's no setting of parameters like that. So you can't operate the island in any configuration other than closed loop. So it's every 100% of every dose is determined autonomously by the device. And even if the CGM goes offline, you enter fingerstick measurements to keep it going. And it will use the fingerstick BG to determine the dose at the time if necessary, and dose automatically at that time. The more finger sticks you enter when the CGM is offline, the more glucose better glucose control you can get, because it can have has more opportunity to check in. But it gives Basal insulin when the CGM is offline because it figured out the Basal rate. So there's no insulin pump under the hood like there is with hybrid systems. There's no manual mode to go through this thing. There's no manual mode. And why that's important, is because there's a number of reasons why it's important in the context of your question. Every other hybrid closed loop, oh, the hybrid closed loop systems. And I would say the iLet isn't, isn't that. But the hybrid closed loop systems started as insulin pumps. And they all started in a world where you didn't need clinical data. Insulin pumps don't require clinical data, they require what's called human factor studies where people come into a conference room, they won't hook up to the device, but they'll go to the user interface and show that they can do the basic functions. And then if that human factors report goes into their market application, so the tandem system, the TCM pump, for example, the Omni pod five, the Medtronic system, the Omnipod, not five, but the early Omni pod system, and the Medtronic insulin pumps all went through human factors testing, but they didn't require any clinical trial data, we did not make a user interface where the islet could be a standalone met and manual insulin pump, in which case, if we had done that, we could have put that through the FDA a few years earlier, had a manual insulin pump and would have needed a clinical trial for that product. But to add the algorithms needs the clinical trial, we did it all simultaneously. So not to
Scott Benner 18:29
be too obtuse, but basically, the islet is like a self driving car with no steering wheel and no pedals. And the algorithm you get it drives you where you go. That's not
Ed Damiano 18:38
obtuse at all. Those are that is the that is exactly the analogy I make all the time. Okay, great, right. So it's like sitting in the in the passenger seat of a self driving car, you can't adjust the insulin dose, you can't, you can't override an insulin dose, you can't give a Bolus, right, so you can watch the thing, control your blood sugar, right, you can watch the self driving car go. And you can watch it, turn it steering wheel and so forth. But it really is determining 100% of every dose. Now you can interface one way to interface with the device is to let it know that you're having a meal and we have something called a meal announcement, which we can talk about. But even that when you issue a meal announcement, you're not determining the size of the insulin dose that is delivered in response to that. It figures that out by itself. And then it comes to know what that appropriate dose should be when you give a meal announcement for breakfast, lunch and dinner. And then it cleans up the rest downstream of that we can talk a bit more detail about how that works. I think it'd be a worthwhile thing to talk about. I
Scott Benner 19:37
know for certain I have one more question about that. Then we can definitely move on. Am I wrong? Just say that in the very, very beginning. You imagine this happening with a tandem pump?
Ed Damiano 19:46
No, no. Yeah. So no, no, not No at all. Actually, in the very beginning. I imagined it happening with something called The Aviator pump resume. You've never heard of that. So Dean came in had built a pump that was So 510 K cleared in a traditional insulin pump, and he licensed that to Abbott diabetes care. Abbott had a possession of this Abbott navigator which my son used from 2008 to 2012. It was in the US at that time. And it was a continuous glucose monitor. Phenomenal one, really. And the notion was we were working with the guys at Abbott, to put our algorithms in between the aviator insulin pump and the navigator CGM. And we would be the smarts in the middle. And what happened was Abbott had a change of heart. And they they did not proceed with the aviator pump. They never marketed even though it's 510 K cleared and they the navigator itself. They abandon that product in favor of what is now believed right in Flash glucose monitoring. So I found myself without a partner. I never intended to start a company never intended to build a hardware platform. My intention was to take the software that Feroz and I were working on and collaborate with Steven to test it and human trials and then license it off or be you Boston University with license that off to an abbot or a Medtronic or a tandem. So we did start talking with tandem. He's one of the reasons tandem is called tandem, as I understood it from 2011 is because they were always contemplating multiple fluids pushing multiple fluids, not just one. Okay. And so there was great interest in in a dual chamber T slim pump and I was working with a guy named Sean St. Who is now our current CEO at beta. Sean 2000 companion. Am I right about that? Well, he was not when I met him. When I met him. He was a young whippersnapper, engine engineer at tandem diabetes in 2011. And he approached me at an ADA conference. And he said we're about to get we think we're about to get 510 K clearance for the T slim pump, which he was right few months later he did. And I started working directly with Shawn in 2012 to build our first mobile device that ran on an iphone four s and commanded insulin and glucagon doses out of two independent tandem T slim pumps. And the woman Kelly you interviewed would have used that system. So you put two TCM pumps in your pocket one delivered insulin one was repurposed to deliver glucagon and this giant brick that an iPhone for us on one side and a G for Dexcom. Receiver all bundled together. Right. And that was our iPhone, bionic pancreas for about eight years. We use that thing.
Scott Benner 22:20
Wow. Wow, that's something look how far it's come and how long it takes us? Yes, a little humbling. Actually.
Ed Damiano 22:27
It For Me in particular. I mean, I never expected this first one never expected to build the whole thing or build this build the team that built the whole thing. But I never would have expected it would take as long as it did and cost as much money as it did. Yeah. It just it's just a huge undertaking of infrastructure to do something.
Scott Benner 22:45
It's funny, because when you talk about the first idea, like it occurs to me couldn't, couldn't any pump company just accept a bunch of algorithms from a bunch of different places and say, Look, just choose the one you want to use. And we could adapt, but I guess nothing. Everybody wants to be proprietary at some point, right. But it's
Ed Damiano 23:02
not just that you're right, that has historically been the case. But until recently, the FDA didn't make that easy. And they and the FDA sort of wanted to get out of its own way. It didn't like the fact that there are all these different companies making algorithms, mostly academic groups back then. Right? Not so much companies, but mostly academic groups, and companies were licensing algorithms from academia. But then you had a few companies making pumps, and you had a few companies making sensors. And so initially, the idea was at the FDA is that we want these sensors, these continuous glucose monitors of which there were like three on the market and now they're there a few more to be able to talk to any one of these pumps. Yeah, and integrate with any one of them. And then they evolve their thinking to say Okay, now let's allow these pumps you know, the a certain type of pump to not only talk to any one of these CGM, which they call they dubbed I CGM, you know, inter operable continuous glucose monitor. But then they wanted to, they made this thing called ACE pump, which was a device that that could talk to anyone these icy GMs and could host algorithms. And you could just plug and play this ace pump can work with this IC GM, that one or the other one. And this icy gem can work with these three A's pumps, and that they wanted that interoperability. And then they said to sort of flesh it out. The third technology in this piece right in the system is the are the algorithms and then they came up with something called AI AGC inter interoperable automated glucose controller, which was one of these plug and play algorithm 510 case or market applications. So now, you could have an AI AGC tested in one ace pump with once I CGM and once you do the clinical data, collect the clinical data for that ace pump IGC you could put it in different ace pumps without having to do another clinical trial. And you can make a talk to other ICBMs without having to do another clinical trial. So they're trying to be able to really promote this interoperability and all this different cross communication in this in this ecosystem of C GMs pumps and algorithms. And so they gave birth to all of these these three different regulatory pathways. And now we have an interoperable space that for the first time, now that as of just recently, in addition to the G six Dexcom, which is a, you know, II CGM, there is now the G seven Dexcom on the market, which is an IC CGM. And there is the libre three, which is an IC GM, it's the first time we've had more than one. I CGM out there.
Scott Benner 25:26
Right? Did you ever consider licensing it to pump companies? Or was that that's
Ed Damiano 25:31
what I wanted to do is what you that was my initial intent. And that's why I was working so closely with Abbott. And the problem with this was that when you start working with the med tech industry, the it can be it can be they can become quite capricious. And the reason is, especially big med tech, you have these divisions, diabetes divisions in these big med tech companies. So they do a bunch of things, right. But one thing they do is diabetes, and they have a diabetes division, and they have a precedent of that division. The president of that division, if they are very successful, very often gets promoted into some other space like cardiovascular within that company, leaving in that person's wake the need for his or her replacement. So another president comes in with totally different objectives. And they might say, you know, I want to pivot away from type one, type two, and this technology that my predecessor has been been investing in, I'm going to divest all of our interests in that and move into a new product, then they carve out their own little legacy for themselves. And so I couldn't rely on the med tech industry, I saw no way to do that. When there's that kind of capriciousness happening in the system. It's built right in right to the way these companies evolve.
Scott Benner 26:35
You don't want to spend six years turning yourself into the right quarterback, and then your coach leaves and he says, I want to run the ball more. And now. Now you're exactly I got it. Okay. Wow. I don't think I have any more questions around that part of it. I have.
Ed Damiano 26:51
I mean, there's a lot of history, right. And we can talk for an entire podcast on the history of this. Yeah, for sure. But the long and the short of it is we ultimately evolved our thinking through experience that we had to build this thing ourselves from the ground up and soup to nuts.
Scott Benner 27:04
And so I let because it's not an insulin pump, it needed a different name. So I that word just doesn't exist in your day.
Ed Damiano 27:13
Yeah, yeah. So it's funny because if you think about the evolution of the of my terminology, I go from the nerdiest of terms that you could imagine like a geeky engineer. And if you look at my slide decks from 2008 910 11, my, my terminology gets worse and worse. And I hit an all time low in 2011. So initially, I called it a closed loop blood glucose control system, which just rolls right off your tongue, right? And then I realized that that's just not, you know, it's a very academic, you know, thing engineer in particular would say things like that. And then I tried to come up with better terms. And I never liked artificial pancreas, because artificial practice tells you what it isn't. It isn't a real pancreas. And we know it's not a real pancreas. Can we have something that's more descriptive? So my ward, my lowest moment was when I called it a prosthetic pancreas. And I may as well just shown a picture of a little pump with crutches on either side of it, right. But in fact, it wasn't a I mean, it is a prosthesis, if you will, and in a way, but it's just not the right terminology. The next year, I said, What is it let's be positively descriptive, and not negatively descriptive, what it isn't. And then I looked, I thought about bionic pancreas. And then I looked in the dictionary, and it's you know, it's it's a technology that imitates you know, biological processes through through electronic electromechanical systems and electronic means it's exactly what we do. And so I coined that term in I think, 2012. And I've used it ever since. And that's kind of a category. It's a bionic pancreas, who my wife came along a couple years later and said, I know what the, you know what the device should be, should be called, it should be called an eyelet binding pack, because obviously an homage to the islets of Langerhans. Yeah,
Scott Benner 28:46
that's a great idea. Also, I don't want to get off topic before we get on topic. But how come you're an engineer and so personable?
Ed Damiano 28:54
I don't know. I mean, I think that AI engineers get a bad rap. You know, I think many of them can be quite personable. Excellent. I'm
Scott Benner 29:03
just like, well, you're like a good storyteller. And
Ed Damiano 29:07
you know, when you tell the story enough times it becomes
Scott Benner 29:09
rote? Yeah, I There are a couple things in my head that if I say I can shut off while I'm sending them off course. Alright. So you've got, you've got your clearance now, is it? I mean, I am assuming you're a smaller company. So what's it like? There's got to be a ramp up plan, right? Like you're gonna launch and then like, how do you foresee that going?
Ed Damiano 29:32
Yeah. So I'll give you a little bit insight into into, you know, our vision for how that should work. Remember, we did, as I said earlier, a pivotal trial. So we took the iLet. By the end of 2020. We had basically locked this thing into this little device that looks just like this thing I was just showing you. And we could then with funding from the National Institutes of Health, we had a large grant from the NIH to help pay for this study. It was what's called an investigator initiated trial. So beta bionics didn't spawn To the study, it was, it was basically sponsored by that jape Center for Health Research. The grant came in through my lab at Boston University, and went and dispensed out to 16 clinical sites. And so we had subcontracts sites at the university, North Carolina, Chapel Hill at Stanford University and masters, General Hospital, and so on and so forth. We had 16 sites. And we chose these sites carefully. Stephen Russell, myself, and Roy Beck sort of went through across the country and said, we really want to pick sites that can bring a lot of ethnic diversity into the trial. So we don't have you know, a study that's consistent, almost entirely of white, very wealthy, and very educated people, but rather a study that has a much better cross section and a better mirror of the population of the population at large. So we chose sites in northern Florida, in southern Texas, in Detroit, in Atlanta, Southern California, where we could get in a lot of ethnic diversity into our cohort and bring a bring a broader demographic into the study. So we designed that study with those 16 sites as the targeted places where they each bring in anywhere between 20 and 35 participants over the course of that 2021 calendar year. And that was really always my my thinking about how we should bring this thing out. We should start by using those sites as the places to launch the device. And when Sean St. came on board as CEO over the summer, you know, I think that jives well with him, he's like, Okay, well, I'll have we'll build a sales team. And they'll break up the country into sort of eight. It's a targeted launch, we break the region of the country up to eight territories, within each of those territories resides one or two of our pivotal trial sites from that study. And so those are the people who are in my mind, the de facto experts of using the island are the only ones have ever used it. In a close to real world setting our trial was designed to be a very good approximation of real world usage, people were on the device for 13 weeks. So they understood the device in a way nobody else could, until you use it, you can really understand and appreciate it. So we thought that was the best place to start. Now, in each of those regions, most territories, there are several other sites that are also going to participate in our launch. But we are moving very quickly. And because we're a small company, as you mentioned, we have, there's a kind of agility that we have, that allows us to go from getting 510 K clearance of a in a company that's never launched a product to launching the product within the space of about a month. So you know, we have certainly launched the iLet. It is, you know, the we have we're using a distributor distributor Durable Medical Equipment approach to distributing the device like like, like a traditional insulin pump would follow. And so we you know, we've shipped our product out to distributors, and they can in turn, ship those two people with type one diabetes and traducida, typically in those regions, and those regions covered a big section of the US, right? And so we just want to get our feet wet in the first few months and just get experience with the pivotal trial sites, and then expand and add territories in the fall. And then more sites as well.
Scott Benner 33:00
Do you see it as a years long project? To get up? Like when when Will everybody be able to walk into their doctor's office and say eyelid? Yeah, good question.
Ed Damiano 33:10
So right now at launch, we were we weren't able to get for example, Medicare and Medicaid Services to cover the device. You know, this is a device that you would, you would use private insurance and government insurance to pay for the device, you'd have a copay, just like a traditional insulin pump and similar similar with the supply with the supplies, but you can't we weren't allowed to negotiate with CMS Center for Medicare Medicaid Services. Prior to 510 K clearance, which we only obtained, you know, just less about four weeks, little less than four weeks ago, once we got 510 K clearance, we can start entering into a contract with CMS. And that takes anywhere between, you know, two and three months. So anybody on Medicare Medicaid Services, needs to know that we can't get it out to them right away just because there was no way we could have teed that up
Scott Benner 33:56
if I didn't have the conversation before the clearance.
Ed Damiano 33:59
Exactly. But with with with with commercial insurance, we were able to through the distributor Network Distributors across the country that sell insulin pump supplies, we were able to set up contracts with them. So the minute we got clearance, they could place an order. So we literally took orders right away upon FDA clearance that allows us to get out to a lot of people in the country in the you know, in the back half of this year, who have private pay. And we're hoping by say the fourth quarter maybe or maybe even sooner than that third or fourth quarter, we'll get government insurance on board once that contract sells and that allow us again to then penetrate further out and reach more people who do
Scott Benner 34:36
you see as your target user group?
Ed Damiano 34:41
The vast majority people type one but specifically, we see this technology is playing really well in the hands of people who are on MDI therapy in the hands of people who are willing to and I think this is most people will let go Have their diabetes management as much as possible. And what I would say is who it's not for, is who we call the knob Turner's and I would have to admit that I am probably one such person. Right? And yeah, I bet you are as well. Yeah. Right now, probably. So what do we do? Well, we have this little tiny child in front of us who has type one diabetes, and we are going to pour all of our energies into making sure this kids glucose is tightly controlled as possible without destroying their lives, right, we don't want to interfere with their lives so much that they're just, they're just a little experiment. So you have to do it in a way that, you know, they can coexist with this, but you want to give them as good of care as you can. So we were all over this little guy. And we grew him up that way. And he organically began taking more and more responsibility of his diabetes management. But I have to say that in taking over that responsibility over the course of a decade, to the point where he goes off to college with a animus pump, and a G six Dexcom, or G five Dexcom, back then he's doing a really good job managing his diabetes, and he is a tinkerer, he's going to adjust to fine tuning of insulin dosing, he's all over it, and multiple times a day. Now, many people who do that and do well with that are going to be able to use the eyelet successfully and comfortably, there may be an adjustment that they need to make and get used to handing the steering wheel off to some autonomous system. But they can sometimes make the adjustment. There are others who won't be able to, it'll provoke too much anxiety, and they just won't get through it. And the reality is, you don't know what sort of person you are until you try it. So living with the eyelid is the only way to find out if you can let go the wheel as he likes to say Yeah, and so we will offer a 90 day return policy with the island. So we want to make sure we really want to make sure that the right people find this device. But we also want to make sure that those who just find that the device is not right for them. Find a way to a device that is a loop system or more of a manual system where you get in, you know, you can take more responsibility for insulin dosing.
Scott Benner 37:01
Yeah, that's interesting. I mean, so what's the straight from the listener questions? What's the target a one see gonna be like, what do you expect the eyelet to pull for people?
Ed Damiano 37:12
Well, so our pivotal trial was, by the way, the largest pivotal trial ever conducted for a automated instant delivery system. It was a huge trial, and we enrolled children and adults simultaneously. So we went all the way down to age six, and all the way up to age 83. So we had a very broad range of ages. And if you typically the way you think about, you know, where the agency comes in, as you typically look at adults separately from pediatrics, it's very commonly done. In statistics you hear usually parse it out that way. So we found that the average a one see that the device achieved was about 7.1% In adults, and about 7.5%. In the kids. It's pretty amazing, actually, it's a really good one say, and we did not increase hyperglycemia, relative to the standard of care. And I think it's important to emphasize that the way we designed our trial was to have a standard of care study arm. So not everybody who went into the trial, use the eyelet right away. So what happened was there was a randomization, who you would use your screen to do the trial. And when you were enrolled, you would randomize the either the eyelet, which is called the intervention arm, or the standard of care arm. And by standard of care, we mean whatever your insulin therapy was, when you came into the trial, do that, but do it with a G six Dexcom. Now, if you were, for instance, using a CGM already, then we don't need to bother introducing a G six Dexcom. If you use the Medtronic, then that's your standard of care. You've got CGM, and we give you a blinded G six Dexcom. Because want to capture all the data ng six. If you're using a Eversense, or Liebreich continue doing that, but will give you a blinded G six. And if you didn't use any CGM, we taught you how to use the CGM, if you went into the control arm into the standard of care arm, and they became a CGM user, at least for that 13 week period. So the study cohort divided across these two groups. And because we had a standard of care arm, we could keep track of how well people did in the trial on their own care. And people tend to do better in clinical trials than they do on their own because they're being watched. They're being you interact with them more. It's called the study effect, the Hawthorne effect. And so we want to keep track of that. And whatever the eyelet does, it's really the difference between how much the you subtract out the improvement the standard of care arm saw from the improvement the eyelet saw relative to baseline. And that difference is the difference in the improvement of the eyelet. You can quantify the improvement and what we found was that it was statistically significant reduction in HBA one C of half a percent relative to standard of care. So we saw a point 5% improvement in Me included in a one C relative standard care on the island. And that was a statistically significant difference, which means that the likelihood that happened to chance is exceedingly small.
Scott Benner 39:55
How about if you take I mean, I heard you I heard the pride in your voice from us. said how you chose the people to go through these testings. And I feel like I understand the underpinning of that, which is that some companies pick ringers, like people who they know are gonna do a good job, right? How many times like how much data do you have about people coming in with just wildly out of control? A onesies? elevens twelves? Did they bring them to a seven?
Ed Damiano 40:23
Yeah, so it's an excellent question. So we were very careful not to have an upper limit on HBO and see, and that's unprecedented. There's never been an AI D study, where where there wasn't a limit on upper limit on HBO and see a pivotal trial. So we, you know, for for a market application for a device. And so we were really clear about this, we wanted to make sure that no more than a fifth, we asked the sites to limit those, those people that you randomize such that the limit limit or fill certain buckets, so make sure that at least a third of them have a one C above 8%. And no more than a fifth have an A one C below seven. And that's because these large epidemiological studies out there like the T Wendy exchange, and other studies have shown that on typically in the US, all these studies tend to corroborate that only about one in five people meet the American diabetes Association goal for therapy. Anyone see below seven. Yeah, that just is it just continues to ring true at least adults kids are even worse, unfortunately, having worse outcomes. But adults 18 and older, it's about one in five are achieving goal and 80% aren't. So we wanted to make sure our cohort as much as possible reflected that. So we asked the sites to try to limit the enrollment of Pupil tendency below seven. And to make sure you had at least a third they went above eight, we also wanted to make sure that at least a third of the cohort was on MDI at baseline. Right. So we didn't take pump users, you know, as as as exclusive requirement. We allowed people who on pumps and people who are on hybrid closeup systems to participate in the trial. So it was an FDA cleared device or an FDA approved device, it was admissible into the study. Yeah, I
Scott Benner 41:55
feel that when you mentioned earlier, it struck a chord with me because, you know, I had somebody asked me recently about, like, how do you stay so made motivated about making the podcast and I was like, for all the people I reach it's a very small percentage of people have diabetes. And you know, those other people are not running around with a onesies in the sixes, you know, and they they're overwhelmed. They don't understand what they're doing. They've long past given up and they're just they're on a they're on arrived with their eyes closed, wondering when it's gonna like come to a stop. And
Ed Damiano 42:28
I think the eyelid is for those people, the eyelid is for most of those people, it's not for everybody. And that on one end of the spectrum, right, as I was trying to emphasize the knob Turner's who are going to have anxiety by giving up control and can't get past that is not for them. And there are therapies. Fortunately, we have so many good alternatives. Now, there are therapies for that. But on the other end of the spectrum, you have to at least, you know, you have to attend to what Stephen Russell calls the care and feeding of the device, you have to make sure there's insulin in the cartridge, you have to make sure the CGM is streaming data, you have to make sure the infusion set is intact and working. Right. And you have to make sure the battery's charged. So that is a care and feeding level of responsibility that's essential for the aisle to help you. And they're going to be some people who won't do that either. And
Scott Benner 43:11
and also couldn't for reasons that you can't be pregnant use and I would imagine, because this is
Ed Damiano 43:16
not indicated for pregnancy, we did not test it in pregnancy. So that would have to be done separately is another try. Are there reasons
Scott Benner 43:21
that a doctor couldn't write it off label at some point for somebody under six? Are you going to have to do that testing before that gets okay?
Ed Damiano 43:29
Oh, physicians can write do anything they want with off label usage. They can use these devices, not just hours, but any of these devices off label, we just can't train to that. And we just need to be very clear. What is on label. Yeah. And what is on label is people with type one diabetes who are six and above, okay, who aren't and not pregnancy? So that's certainly not something we have an indication for.
Scott Benner 43:50
So So for clarity like, I can't use islet and achieve a five five a one C with there's no way for me to manipulate it or do that kind of stuff without Lowe's.
Ed Damiano 44:01
Ah, good question. So I noticed one of your one of your, some of your users had some of your listeners had a question similar to that. So what we found is about 46% of our adults had a mean glucose after 13 weeks on the island of about 100, about about 46% had a mean glucose under 154. And an agency of 7% corresponds to mean glucose about 154. So 46% of the cohort had a mean glucose below. Below 154. About 27% of the cohort on the island had an A one C below seven. So almost a third. So what does that mean? What was the lowest day when seeing the island? It was in the fives by the way. So we did have somebody who were those 13 weeks on the island, they ended with an A one C sort of in the mid fives, but it's unusual. The island tries to bring people's mean glucose and anyone see up a little higher if you're sitting down at what I mean glucose at 110 or 120. Right? You're likely going to see it increased toward 130 or 140 or 150. So it is that increase that some of the As folks who enjoy being down there, maybe they pay a price of hypoglycemia, but they want to be down there. Skimming the trees, so to speak, will be frustrated by that rise. But the reality is, all the clinical data suggests that there's no advantage to an agency of five and a half over an agency of six and a half. There's almost no signal for microvascular damage polonium exceeds seven, which is why these, you know, these societies like the Endocrine Society and American diverse decision, have these goals for therapies, goals for therapy.
Scott Benner 45:28
So, nuts and bolts. I want to go over just how it works for a second. So yeah, I'm going to eat a meal. Am I right? Like, I probably should just ask you, but my understanding of it is, and I'm assuming if this is my loose understanding, it's ever about a lot of other people's. I announced the meal by saying This is breakfast, lunch or dinner. And then I say whether this is similar, smaller or larger than I'm accustomed to eating, is that it?
Ed Damiano 45:53
That's it. Okay, so even rolling the tapes back Further still, to start the island on your own day zero, right, you get on the island, you enter your body weight. And that is it, right. So there's no programming of Basal rates, there's no programming of insulin correction factors. There's no programming carbs and some ratios. And there's no carb counting specifically, right, we do ask that you be carb aware. And I'll make a mention of that in a moment. But so to start the system, you enter your body weight, you have to learn how to hook up the infusion set and pair it with the CGM and so forth. But then you enter your body weight, and then you go bionic. And then you swipe to go down again, the system starts dosing every five minutes of every day and adjusts insulin therapy according to your needs to your ever changing insulin needs. But the meal announcement works as you describe almost exactly, to give you a little bit of color under the hood, as to what's actually happening when you do that. So with the meal announcement, you just simply swipe to unlock the device, and you just press on the little knife and fork. There's a little knife and fork here. And you press on that and it asks you, you know, is this you know, what meal type is it and you get to bucket breakfast, lunch or dinner. So given the time of day, I'm going to choose, let's just say, Well, it depends on where you are. Let's just say I choose dinner. And then it asks you is this usual carbs? For me more or less? So three buckets? And no numbers? No, mind you, right? This is diabetes without numbers. As a primary care physician that we've been working with, for years, who was really the one who coined that, and then you just simply, once you say, you know, usual for me, for instance, let's choose usual for me, then you just swipe, and it then determines the dose at that moment, and it begins to deliver. So what happens is it gives a dose of insulin at that moment. So if you have the food delivered, you don't pre meal Bolus, we discourage that, we ask people to wait until the food is in front of you not to worry about fat and protein, right, just focus on the carbohydrates on your plate. And by that I mean is this is this bowl is this is this lump of rice, the usual amount of carbohydrate I'd have for my lunch, say, or my breakfast or my dinner. And it will then on the very first offering of a meal announcement say for your first lunch meal announcement that you issue. It'll give a Bolus at that moment. Once you once you say usual for me lunch, for example. It'll give a Bolus based on your body weight initially, and it'll be quite conservative on that first attempt. And then it will watch every five minutes of the rest of the day, how much you know what your glucose does, and it will add insulin as needed, or suspend insulin as needed. And we have two other controllers that are running separate from the meal announcement controller, the one that gives that Bolus up front one we call the Basal controller or the Basal algorithm. And the other algorithm we call the corrections algorithm. And they're working in concert every five minutes and they adapt on multiple timescales to your changing needs. But the correction algorithm will add insulin above and beyond what the Basal algorithm thinks you need for your Basal requirement. And if it sees the blood sugar starts to rise, even in the in the face of that meal announcement Bolus that was just delivered, the correction algo will add some additional insulin. And at and tomorrow, when you issue another meal announcement for say lunch or usual for me, it will look to see yesterday when you did this, the meal announcement gave three units of insulin and then we added another three units of insulin of correction insulin in the four hours afterwards. And that was not the right balance. It wants the meal announcement to be a majority of your mealtime insulin over the four hours after the meal announcement. And if it was short of that, it'll make it a little bigger. The next day you do it a little bit smaller if it was too much if it was all of the insulin, and then they'll start adapting that and the body weight thing becomes less important. You initialize that with body weight but it's allowed to depart from that very quickly and start adjusting the size of that lunch meal announcement to be to account for most of your for our insulin but not all of it. And it separately adapts the meal announcement for breakfast, separate from lunch separate from dinner it buckets those three and if you have a snack, you know if you have an evening snack you might want to bucket that with dinner if you have a morning snack, you know a little left o'clock ish like Winnie the Pooh, you might call that a breakfast snack.
Scott Benner 50:03
Okay, right? Why doesn't it need a Pre-Bolus? Is it? Because, I mean, I've used a number of different algorithms with my daughter, but most specifically Luke and Omnipod. Five. And they they seem to have in common that once you put in insulin, they take away basil and then work backwards is kind of how I think about it. Does that make sense? To you stay understanding, is the eyelid staying aggressive when the Bolus goes in? Because you my daughter can't eat food without Pre-Bolus? Like whether I did it like so what is it doing? Is it is it matching the power of the rise with insulin and then getting the hell out of there before it causes a low?
Ed Damiano 50:45
Well, yeah, so what it does is it the Basal controller is a controller that adapts on multiple timescales. So let's just focus on that one. First, the basil algorithm, we call it, and it has, you know, it's adjusting sort of an average Basal rate that it figures out by itself over time. And there's an there's a, an ability of the basil algorithm to shut Basal insulin off completely, if you're starting to go low. If you are low, or you're tending low, it can turn the Basal Basal insulin dosing off completely. But it's adjusting this on a very short timescale, looking at your glucose levels, every five minutes, it's it's got to be very responsive, I'm going to turn off Basal insulin if you're dropping too fast, or if you're low, or it can just run along. And it also can see daily patterns. So it also adapts on a diurnal nocturnal timescale of 24 hours. And so we can see that, you know, suppose your child who has growth growth hormone secretion upon the onset of sleep at around 11 o'clock or 12 o'clock at night, and they tend to need more insulin because of that, because the growth hormone causes, you know, release of glucose by the liver. And so the Basal control will start to see that and it'll start to adapt upwards and it might see this pattern that typically around this hour of the day, I need more Basal insulin, it'll just sort of it'll, it'll see that pattern and it'll reinforce it. But if you change as you grow into a young, a young child, and then a teenager, and you start having a cortisol secretion, or just before waking in the dawn hours, you no longer have the growth moment at night. Now that shifts toward your needing more Basal insulin, say it five in the morning, it'll figure that out automatically. And similarly with intercurrent illness. So if you have an upper respiratory virus, and you see a sudden need for more Basal insulin or more correction, insulin for two or three days, it'll see that automatically and we realize that I've got to get more insulin to keep the glucose down at this average that I've been trying to achieve. So I will adapt upwards for that two, or three or four days or a week when you're more insulin resistant. And if you have a vomiting illness, you're very insulin sensitive, it'll do the opposite, it'll back off and become less aggressive. So it's doing the Basal controllers doing that the Basal algorithm and the corrections albums also figuring out your insulin sensitivity, automatically, not so much in terms of the number, what is your insulin sensitivity factor, but rather recognizing that this person over the over the days and weeks, months and years, their insulin resistance might change, they might need more insulin, when your blood sugar hits 250 than it used to when you were six years old, then then you need now when you're 14, and it'll suddenly start adjusting that upwards as well. And you'll get more correction and then on top of it, and it adapts on multiple timescales, not just five minutes in daily timescales, but intermediate timescales as well. And that adapting on multiple timescales allows these two algorithms to learn. It's really a self learning system, and allows that system to engage in what is essentially called lifelong learning. So it does see patterns on a daily basis and is able to adapt to your ever changing insulin needs. Meal announcements adapt according to how you provide input on what is tip usual meal for you breakfast, lunch and dinner. And you're
Scott Benner 53:46
comfortable calling it learning. It's not just going off of what it seen recently, but it's it's remembering stuff from the past.
Ed Damiano 53:54
Yes, it is. And so it is storing information from this past week on where your insulin needs were higher and lower. So it is a kind of an autonomous learning system. I wouldn't call it artificial intelligence. Right? It does. It does do some pattern recognition, though, in the sense that if it sees if it sees, you know that the basil algorithm is giving more insulin at this, you know, early morning hours over and over again, it will see that and it will, it will it will it will tend to be higher in that in that period, unless for some reason in that particular morning, you don't secrete the cortisol and you're more insulin sensitive, it can very quickly turn off the Basal insulin. So it is a learning system in that regard. Yeah, so
Scott Benner 54:32
that brings up a question that a lot of people ask them that. I was wondering just while you're talking us an idea like somebody's getting their period, like so one day like I'll use my daughter because I'm sure one day she'll love to listen back to this podcast and hear her period as much but in the days approaching, Arden can be need more insulin, and then when the event happens, she can fairly suddenly need less, and it changed his dress. So how Otherwise, how does like can I just like whisper in the eyelids here? Like, you know, I got my period or like, like, how does that like you because you can't tell it stuff like that,
Ed Damiano 55:07
you know, you can't tell and stuff like that. And it doesn't have it doesn't have memory over a monthly cycle, right? It's really looking more over the seeing patterns in the past week or so. But just to be clear, the eyelet learns and adapts very, very quickly. So what we found was in the pivotal trial, remember, we start the system with your body weight. So imagine you have a teenager, a raging hormone adolescent who weighs 70 kilos, and uses 90 units of insulin a day, and an adult who weigh 70 kilos, he uses 45, the islet will figure out that difference in about 24 to 48 hours that difference, that's fast, right? That's fast enough to handle the increased insulin demand around periods around intercurrent illness, the physiological changes in insulin demand happen over the space of a day or so it will see that and if you suddenly become very insulin sensitive, like you just described, it can shut Basal insulin off and it won't dose correction and insulin if it doesn't need it. If you're not hypoglycemic. So if it sees you sort of staying low, it'll back off completely on Basal insulin or or shut it down dramatically. So that if you won't go low as easily as you can go high. So it definitely is biased and trying to prevent hypoglycemia. That's like the first order of business is to limit hyperglycemia.
Scott Benner 56:24
How does that work with exercise?
Ed Damiano 56:27
So with exercise, I would like to introduce you to the idea of a by hormonal system. That is, indeed, unequivocally the best way to deal with exercise is that you know, is to be as biomimetic as possible. That is how the pancreas handles exercise. It reduces insulin secretion simultaneously with increasing glucagon secretion. And we really do need a by hormonal system to handle exercise elegantly in type one. All single hormone systems are vulnerable to exercise, even insulin pen therapy, all of it. Yeah. So you have carbohydrates to help and other other tricks to deal with that using carbs, you know, many carbs to treat, or being fasted going into exercise are different ways to deal with people different ideas about that. And many of those tools are going to be used with the single hormone island, but specifically, disconnecting from the device is what we recommend, if you're going to engage in exercise, just this is going to add the infusion set, if you're going to engage in exercise, that where you find yourself going low in on your other therapies, try disconnecting from the eyelet. There's no setting of temporary basals. But if you do have to go beyond that, and say it's not enough just to suspend insulin, I also need to usually take some carbs ahead of exercise, if you're going to carb load like that, to prevent hypose. during workouts, what we ask you to do is to disconnect from the eyelet first and then take the carbohydrates not the other way around. Because if you do the carb loading and forget to disconnect in your workout, it'll see the rise and we'll start dosing just when you don't want it working out. So the order is actually important there. And that's something a little different from traditional pump therapy.
Scott Benner 58:05
Okay. Would you say that the system if it's trying to address a higher blood sugar, for example, does it address it with Basal insulin or with a Bolus?
Ed Damiano 58:14
Yeah, so it has the correct the correction algorithm is responsible for giving insulin above and beyond what the Basal is, sees its responsibility to be the Basal insulin. The Basal insulin algorithm is sort of swimming in its own swim lane. And then you've got the correction algorithm that swims and it's swim lane. And when you start to have hyperglycemia excursions for whatever reason, stress hormones, stress, hormones, stress, you know, whatever is or an illness, or carbohydrates, right if you forget to meal announce or if you meal even if you do meal announcing you tend to see a still arise after that. The Corrections algorithms responsibilities to come in and give that additional insulin above and beyond Basal that handles hyperglycaemic excursions. And if you forgot to meal announced it will provide all of the additional meal insulin, if you will, it doesn't know its meal. Instead, it's just correction insulin needed to bring you back into range. But if you do forget to meal announce, it will step up and do that. What we tell people is that what we've seen is that typically, if you forget to meal announced and you eat a meal and it has the sufficient amount of carbs to cause a glucose excursion, you'll typically you'll likely go higher than you would if you did a meal announcement you'll be higher for longer and there is an increased chance of late postprandial hyperglycemia if you don't do the meal announced because the meal announcement gets the insulin up front, it's always best to get the insulin up front than to wait until you see the rise but it is designed to handle glucose excursions when you know nothing of any sort even those occasions when you forget to meal announce Okay,
Scott Benner 59:40
so here's another idea. What if I'm a very low carb person and I weigh 150 pounds I put on island a family 150 pounds and I'm eating breakfast and it's normal for me. Like but normal is three eggs and two slices of bacon and a half a piece right? What happens then?
Ed Damiano 59:59
Great question. So as I said, with a meal announcement, all we're asking you to do is be what we call carb aware, meaning, you know, know what the difference is between the three macronutrients know what a fat a protein and carbohydrate are, right? You should that's every every, every person should know that whether you have diabetes or not everybody should have that level of nutritional education. So with the eyelet, we expect that level of nutrition education, and we provide educational materials in our training documents to help understand some of the macro nutrition, some basic nutritional guidelines, but essentially, understanding that, you know, if you're eating eating eggs and bacon for your, for your breakfast, and you're having no carbohydrate, there's no meal announcement to be had, it doesn't matter if there's 80 calories on that plate, right, or, you know, whatever your hundreds of calories on that plate, you know, 150 300 500 calories, if there's no carbohydrate, and there's no need for a meal announcement, suppose you're a grazer, and you never have more than, you know what we typically say, for adults, if you're having fewer carbs than just a single slice of bread, then there's no need to meal it out. So that's just a rule of thumb. Again, you'd like to stay with numbers. But for an adult, you might think of, you know, a piece of bread, or anything less than that, you can probably just skip the meal announcement and let the corrections algorithm do the rest. So for small snacks or meals, where you have very low carbohydrate, you wouldn't meal an ounce. Where would
Scott Benner 1:01:17
I expect my blood sugar to go in a scenario like that?
Ed Damiano 1:01:21
Oh, well, I mean, it of course, it varies. But I mean, if you had a very low carbohydrate meal, you know, you could see an excursion, you could see a small excursion to you know, 100 to 200. Meg's per deciliter, and the direction album could kick in, and then bring you back down. If you had a very high carb meal, like suppose you had, you know, 120 grams of carbs, right. If you're a teenager having a big bowl of cereal cereal in the morning, it's not an unreasonable to see her glucose, go to 250 or 300, even with a meal announcement, because it takes time because we do encourage people to do it at the side of the meal, but not before. And by the way, if you forget the meal announcement, and it's been more than 30 minutes, since you started the meal, we asked you not to do a meal announcement, then just let it go, let the corrections do it. But it isn't unreasonable for your glucose to go to 250 or 300. If you have a very large carbohydrate load, even with a meal announcement, but then the correction will kick in and take care of that.
Scott Benner 1:02:17
And I couldn't say have a breakfast that I called normal and then realize like, oh, hell, it was larger than Can I go back and tell it like, hey, you know what that was a large breakfast, or do I
Ed Damiano 1:02:29
know what you would do is the way you would try and deal with that is suppose you have a breakfast, and then you want to have or dinner and a dessert, right or a meal, let's just get it out of the category breakfast is, in general, if you have a meal, and you look at what's in front of you now estimate is this usual, more or less than usual for me. And so in what's in front of you, now, I'm going to clean my plate, I estimate I estimate, I'll clean my plate, let's do usual for me and swipe. And then it'll give that that meal dose and it'll start watching your blood sugar rise. Now let's say 45 minutes later, you're gonna have a dessert and dessert comes and it's got more carbs than the meal. Then you could add right at that moment, you could say more than usual as another because it's like another meal and you can just stack it right on, attach it to that it's not stacking, because you really do need that insulin. So you're attaching an one meal announcement to another and they're separated by say 30 or 45 minutes, whatever it is between the time you get your primary meal, and you get your dessert. So desserts can very often be more than usual. Because they very often are carb rich. So you shouldn't resist you'd still call it so let's say you're having a dinner and you have a usual for me dinner, typical amount of carbs and then the dessert comes in. It's it's you know 50% or 75% more than your carbs and you'd have in there in the in the dinner you just you just swiped for you then swipe for a more than usual dinner as your as your dessert.
Scott Benner 1:03:49
Okay. I'm going to look through a couple of these. These. These questions here? Sure. Do you have an idea of what it would cost out of pocket? Once it's available?
Ed Damiano 1:04:01
Out of Pocket? Are you do you mean with insurance?
Scott Benner 1:04:04
Now if I didn't have insurance, I want to pay cash? Oh, yeah. Do you have? Yeah, I
Ed Damiano 1:04:07
do think yes, you can. You can buy you can buy. We have one of our distributors that allows you to buy direct if you were to do a cash pay. Okay. So yeah, I think it's very, very similar to the price you might pay for an insolent a durable insulin pump. So, you know, several $1,000 is what you'd expect to pay for out of pocket cash pay. And you would purchase that not directly from beta, but one of our, like eight or 10 distributors that we're working with. There'll be one distributor you'd go to to do a cash pay, and there's a special price for that. And you can if you'd like if you're in a warranty and you want to get out, you can do a cash pay. Again, it's important now we do have the 90 day return policy and that's important for people to find out if they can live well with the eyelid or if it's not the right device for them, and it's
Scott Benner 1:04:54
covered by a wide range of of insurances. It is how What kind of uh, hell is that setting up on the business side, people that go out and knock on an insurance company doors and I mean,
Ed Damiano 1:05:05
it's dedicated team, right? First of all, we have a dedicated team and market access team at beta for helping people with reimbursement. But the way we started this is through the what's called the DME channel, the Durable Medical Equipment channel, right. And through the DME channels, you have distributors across the country, and each distributor has set up contracts with all the commercial payers, so they had that there, like a buffer for us. So they'd had all those conversations. And similarly, CMS can go through those distributors as well, once we have our contract our contract with CMS setup. So we will sell through distributors at launch, we do also, we're very interested in getting into the pharmacy channels as well, which we think is in our future. And we have several reasons why we think that makes a lot of sense. And it's best for people with diabetes as well as as providers. But for now, and at launch. It's all through the DMA channel,
Scott Benner 1:05:54
okay. Infusion sets, just what are their options,
Ed Damiano 1:05:59
we have one steel set at launch and one Teflon set, they're both 90 degree six millimeter, and they're made by unit medical. So we are using the unit medical family of infusion sets. So if you're familiar with the terminology, we have the inset one, which is six millimeter 23 inch tubing to the eyelet cartridge, and we have the contact detach, let's put this Teflon set and the contact details for the steel set, which is a 90 degree. You know, I think it's a six millimeter 29 gauge steel set.
Scott Benner 1:06:31
So you said something earlier that it's not leaving my head. So I'm after asked about it. If I sit down at a burger joint, and I have a cheeseburger and french fries. And I go okay, the rolls 30. And I'll even throw in five more for the burger just in case and the fries are 80 carbs. So it's 120 carbs. But I know for certain that 90 minutes from now when my digestion slows down and that fat slows everything down. I'm gonna see a rise up to 220. If I don't Bolus for the fat, how does it deal with that?
Ed Damiano 1:07:04
Right? So you're you're you're invoking this idea of a square wave Bolus or something where it's been a very complicated way, with a traditional pump, you think about saying, well, let's release some of the insulin now. And then later, I want more insulin to come in a second wave. gastric emptying happens over a long period of time, right? Because of the fat and the protein slowing that? Well remember what I said at the beginning, we have two other algorithms besides the meal announcement algorithm that are running every five minutes of every day. It's like a perpetual squarewave Bolus ready to be let loose, if needed, but only if needed. Okay, so it's watching you every five minutes. And suppose what happens is the meal announcement comes in, and some of the carbs are released quickly. And you see this rise in the meal announcement insolence catches up to it, and you start coming down and you dropped to say 170. And now you're at like two hours out, and you're down to 120. And suddenly you start to rise. The Basal control is just chugging away. The Corrections Adams watching it, it's like a hawk every five minutes now suddenly start drifting up to 151 6170. It starts adding insulin saying basil, you're not you This is out of your league. I'm coming in to take over and so the correction element comes in and starts adding insulin without you having to pay attention any of that because it's not your it shouldn't be your job to do that. Is that is that
Scott Benner 1:08:19
stream thinking?
Ed Damiano 1:08:21
They just ate three hours ago. This is probably a reoccurrence or does it not care. It has no opinions. It just it sees no judgments and no opinions. A number and it goes no, no, no. Yeah. All it cares about is your glucose at the moment. And it uses gets past insulin insulin history with you it's learnings from that history and your current glucose level and the amount of insulin that it is that is pending. It's keeping track of all the insulin that's pending every five minutes and updating that itself.
Scott Benner 1:08:48
So if Bolus is the number it sees not a predictive trend, it's not.
Ed Damiano 1:08:54
Yeah, I mean, it's certainly we have we use something called Model Predictive Control. So it does look, it does make an estimate of what the glucose is going to be in five minutes from now the next step, and then it will update its estimate of that at the next step once it sees the real value and compares it to the model. But that's it, it's just a, it's a five minute prediction on what your glucose is going to be. But importantly, it keeps track of the very long horizon into the future of your insulin tail. Because every dose it gives, it keeps track of how long that dose takes to rise and picking your blood which is usually about an hour and six more hours before that insulin I'm giving right now is really got mostly gone. And then five minutes later gives another dose and it's superimposes that insulin rise and fall profile and it has that insulin to look forward to it's what we call pending insulin action. It's accounting for that and predicting what your glucose is going to be in the next five minutes
Scott Benner 1:09:45
when I talked to people in in person when I do in person talks I explained to them about there's different levels of or different lines of insulin happening all the time you put in some here the Basal is hitting peaking and tailing. And then the basil from five minutes later is hit, you can't keep it all straight in your head, right, but so is every Bolus. And if you if you really think about it like that there's, there's these constant pushes. It's right fantastic that an algorithm can like, make quick sense of that.
Ed Damiano 1:10:15
And that's what it's, that's all it does, like, you know, it's really good at this very narrow task, it's much better than we are the vast majority of we write it is much better than that, because it's got one very narrow job, we do many things very, very well. But the vast majority of us can't do what the eyelet can do, because it's its only job, and it's doing it every five minutes, it doesn't have anything to distract it. That's all it really cares about. And so it keeps track of every one of those doses and literally superimposes those doses, one on top of the other to account for how much insulin is trailing off and how much is rising.
Scott Benner 1:10:48
It's got a cartridge, right? And for how much does it hold?
Ed Damiano 1:10:52
It's 180 unit cartridge. And after you prime the tubing, you'll have about 160 units. So we found it lasted about three days in the average adult,
Scott Benner 1:11:02
okay, but if I pop I just get somebody to write me a script for more and so it's I've never right, I'm gonna I'm gonna sound odd for a second. I've never, I've never used the tube pump. So my daughter Okay, and using exclusively Omni pod since she was four. But you just pop out that cartridge, put a new one in prime it and keep going.
Ed Damiano 1:11:19
Correct. So let me tell you two things about that cartridge. One is we have we have two different types of cartridge. One is a patient fillable cartridge. So it's a glass cartridge, 1.8 ML, and you can put human law or Nova log in it through the septum, you just draw it out of a vial like you would with your Omnipod into a syringe and then introduce the syringe needle into the septum of the cartridge which looks just like the septum on your insulin vial. Yeah, and then you introduce the insulin and remove the bubbles and then you load the filled cartridge into the eyelid chamber, quarter turn to have the eyelid connector and tubing to the eyelet. And then it'll prime some of the tubing and then you prime the rest of it and hook it up to your set. The other thing I want to tell you about is that in the pivotal trial, we used human logon Nova log in the adults in the in the randomized control trial. But we also had a cohort of adults use fiasco in a prefilled cartridge that no one artist makes which is identical in shape and size to our patient filled cartridge, or ready to fill cartridge and it's filled in a blister pack, it comes in a blister pack of five cartridges and it's prefilled with the Aspen so that dispenses with the need to transfer insulin from a vial and pull up the air bubble. And that process takes about five minutes or so we eliminate that. So with the prefilled cartridge in the trial, you just pop it out of the blister package, slide it into the chamber quarter turn and you prime the tubing, you can change a cartridge soup to nuts a prefilled cartridge vs cartridge in less than 60 seconds with the out because it's got a very fast motor drive train like the atom is pumped it for those of your listeners who are familiar with that. So we had it we emulate that very fast movement of the rewind and then advancing and priming you can do less than a 62nd change if it's a prefilled cartridge,
Scott Benner 1:13:02
did you notice any better outcomes with fiasco or other insolence?
Ed Damiano 1:13:08
Not much for one thing I'll say is that in almost every every analysis we did it was very similar to human login or login the adults 18 and older. And what we found was that in every way, you know, it had very low levels of hyperglycemia. Like similar to standard of care, like we saw with hemoglobin Novolog. It's mean glucose was very similar. The ANC was similar time and range was similar 71% With vs versus 69% in the adults for hemoglobin Novolog. But we did see a statistically significant improvement in time and range. It improved by 14% relative standard of care relative to human lung Novolog, islet users which saw an 11% improvement in time and range. So that was statistically similar, but it's not sure I'm not sure that's clinically relevant. But it was a little better. And one thing I add to that we didn't tell the islet it was fiasco. Right so we have hard coded in the islet knowledge about insulin kinetics. Now we know that the aspirin the aggregate absorbed more quickly, in most adults, or you know, in the aggregate of a cohort of adults, then he will log on over log and it clears a little faster. So it's a slightly faster drug. And if that information had been provided to the eyelet, we have some pre pivotal studies that showed you might see better glucose control and lower mean glucose higher time and range with the aspirin human log, no log. But we for this study, we talked to the FDA about it. We didn't have enough data to do go into a pivotal trial and adjust the the built in parameters in the device to let it know that fiasco was faster. So it was under the assumption it worked operate under the assumption that it was just like chemo Novolog. And so it didn't get to leverage the faster kinetics. It was in the mathematics that's built into the device in the future. We will visit that possibility. But we didn't see big differences and probably just because we couldn't tell it it was faster Okay.
Scott Benner 1:15:00
If should people hear that those are the only insulins they can use and think that that's the case? Or can they use the I mean, you can't tell them to use it off label, but it's something horrible gonna happen if they put a pager in it or something like that as
Ed Damiano 1:15:14
well, we didn't test it with Piedra, we did test it in adults with jemalloc. No bloggin. fiasco. And one thing I didn't mention is that is that the when people who randomized to the standard of care arm in the pivotal trial for 13 weeks, they kind of drew the short straw. I mean, they want it to be in the trial to test the island, but they ended up randomizing to their own care. So what we did was, those people had the option who randomized Sustainer, care to spend 13 weeks on the island after the study ended. So they could cross over the island. And the vast majority of them did just that. And when the kids crossed over, they all use the Aspen the prefilled cartridge. And we saw very similar results to what we saw with the adults with the ASP. So what we have right now in front of the FDA is an application to get the prefilled vs cartridge approved for use with the eyelet. And that's going through the process right now. So we're hopeful, hopeful that that will that will come through soon. But right and at launch, it's cleared for use of human lot with hemoglobin Oplog in our patient field cartridge,
Scott Benner 1:16:10
right? A couple of ideas around you being a smaller company. So people ask questions, like, you know, there's the diehard on the pod people that are like, Look, if it's got tubing, I don't want it, can they make one without tubing? Can they get it for kids under six? Can they can they can they do you have the bandwidth? Can they can they can they or where are
Ed Damiano 1:16:29
you at? Yeah, I mean, we do have limited bandwidth. But we're very creative about some of the things we can do. Like for instance, because we came from an academic realm, myself and Stephen Russell in, you know, in the early days of the project, we do try and think creatively about ways to bring resources, financial resources into the company to help us do trials that might give us indications for use for other kinds of conditions, right, other kinds of diabetes, you know, and so forth. And different age groups and things like that. So what we have done is we've worked with other investigators who are in academia, like ourselves, and they can put in grant proposals to the Helmsley Charitable Trust the JDRF, the National Institutes of Health, to get funding for studies now dilates, FDA cleared to test it in other indications. And so our hope is that we can work collaboratively with academic institutions and clinical investigators like ourselves to do those studies, instead of it being Stephen and Ed's teams doing those trials, we're now going to work with other investigators like ourselves, to do that in the academic realm. So we can leverage all that financial resource that comes from private foundations and government funding, it doesn't come money that's coming to beta, but then beta doesn't have to spend the money to do those trials. So that's how we hope to get expanded indications. And at this time, at least, and then, you know, as the company gets more resource, then maybe we could do some, some sponsored studies as well. But we're limited in what we can do. Outside of you know, we really want to get the buyer model pivotal trial started, because we're very committed to bringing the buyer model eyelet. Yeah. To people with title,
Scott Benner 1:18:00
I want to get to that. I just, I have a couple more questions first, of course. So I don't know anything about what you did, like, I don't have technical knowledge, did you decide we're gonna shoot for a seven a one C? Or is that what the algorithm is capable of? Like? Were there four dials? You could have turned and you'd be on here telling me oh, it keeps people around to 681. C, and you spike to about 180? Not to like you don't I mean, like, or is that not the case?
Ed Damiano 1:18:26
Yeah. So the way we did this is we started by studying the by hormonal system. And we we chose a glucose target, and aggressiveness factors and things like that initially. And then and we did these studies, first in the inpatient Center at MGH, just with the by hormonal system, once we started human trials, and after a while, we, you know, it became clear to us that with the biochemical system, occasionally the glucagon channel might be might not be available. And so what happens if the glucagon runs out while you're out and about, well, what happens if you have an occlusion or your Google infusion set fails, then if that happens, you know, it needs to sort of fall back safely into an insulin, a single hormone insulin only configuration, and we hadn't really tested what that looks like we weren't back then thinking about making a single hormone islet is a product, right? We were thinking about this being a fallback. And so we started doing studies testing the by hormonal, bionic pancreas against the single hormone bionic pancreas against standard of care. And what we found was the single hormone Bender packers was a very differentiated technology in its own right. It couldn't it but it had had had all of its glucose targets had to go up higher, to be able to get really good glucose control and not have hyperglycemia. And so we started studying different glucose set points for the insulin only system and for the by hormonal system. And with single hormone, we found that you can safely have these targets up here and not have much hypoglycemia. And with the by hormonal ones, we could have safely have these targets down here and still not have hyperglycemia. Because glucagon is helping that. So we could basically have effectively something that could give a little bit more Just went up front a little bit more aggressively, just because the targets are lower that it operates under with the buyer model. And so that's how we came to figuring out what these targets were. So the agency that it gets, or the media glucose that it achieves was really It fell out of the mix, we weren't shooting for a particular target of is it going to be to get 154? We said, What is this system do configured this way with this target? What is the average or cohort will get on the system. And we found out with the single hormone, it was about 155 In adults, and the bimodal and in adults was more like 140. So it's about 15 meg per deciliter improvement by adding the second hormone and being able to use these lower targets with single hormone. As you lower the target every time you lower the target, you see a lowering in the mean glucose but a concomitant increase in time below 54. With the buying hormonal system, we saw as you lowered the target, you saw a progressive improvement or lowering of mean glucose without an increase in hyperglycemia. But with an increase in glucagon usage, okay. So we exchanged hypoglycemia for slightly increased and glucagon infusion. And so we can keep these lower targets safely.
Scott Benner 1:21:11
So when you get to a dual chamber at some point, and you're doing glucagon and insulin, what do you think you'll be back on here telling people about their outcomes?
Ed Damiano 1:21:20
Oh, so what we've seen in the as I mentioned, what we've seen in all of our pre pivotal studies that we published across over the years is a mean glucose. This is about about 15. Meg's per deciliter lower than what we saw with the eyelet pivotal trial, which would correspond to about a half a percent lower a one C one. And as as you may recall, I said about about almost half the people had a mean glucose below and 54. On the single hormone island of the adults, what we see is that about 90% of people on the buy hormonal system have a main glucose below and 50 for adults. So it's a big difference in terms of bringing more people under,
Scott Benner 1:21:58
and it's going to become increasingly unlikely that you experience a low and what are we calling a low, by the way, you said it arranged a couple times is that 71 8070 to 180? is what we're calling in range. Okay. And so that's timing, right? The lowest 69?
Ed Damiano 1:22:13
Oh, no, no. So yes, certainly. That's, that's out of range. So that's below range, right? So we measure two different we keep track of two metrics. In our pivotal trial studies, we had an outcome that looked at how much what percentage of time do you spend 70? And what percentage of time you spent below 54. And the way we powered the study was we said that, you know, we powered the study for statistically for superiority, we expected to see a superior outcome in HBA win see in reduction of HBO and see so we saw a superior we thought we'd have superiority in a one zero standard care, and non inferiority in time below 54. relative standard care. And that's exactly what we found in the trial. Yeah, I feel
Scott Benner 1:22:52
like I haven't, like just expressed enough how pretty amazing it is just the meal announcement portion of it. Like I can't imagine what a what a relief that must be to people. Did you talk to them about that in like exit surveys and
Ed Damiano 1:23:09
things we did? Excellent question. So we had we had focus groups at the end of the trial. So we worked with Joe Weisberg, ventral she's up in Chicago, and she works at the Lurie Children's Hospital. And so she's an expert in psychosocial and behavioral outcomes when it comes to studying diabetes technologies. So she developed, validated, behavioral psychosocial tests questionnaires that we gave throughout the study. And also ran the focus groups at the end for people as they came off the device. So that was qualitative. So we have these quantitative questionnaires. And then we have these qualitative focus groups at the end. So we did get to find out, you know, how people felt about things like diabetes, distress, fear of hyperglycemia, but also just sort of qualitative measures of how how people feel about the eyelet. And I do think you're right about this giving up of carbohydrate counting this diabetes without numbers is really important to people because we're trying to say that we really hope this device is agnostic to levels of literacy and numeracy to levels of to technical acumen to socioeconomic status, race, ethnicity, and so we did a lot of work. In the trial doing subgroup analysis, we published something in the European Journal of Medicine after the main study was published, in a letter to the editor looking at the subgroup analyses to show that the people who needed the most improvement in glucose got it from the eyelet, more so than people who were very close to range. And so it didn't seem to discriminate against people if you're an MDI therapy and never use the pump versus people aren't hybrid closed loop didn't discriminate against people who had never used the CGM versus those who had. What you do see is the people of the highest baseline agencies at at baseline before the study started. So the greatest improvement and you'd mentioned, you know, I imagine you're seeing people with higher agencies than other studies, our highest day when he was 14.9 at baseline, so we brought people in across the mix with you know, hi, when season the double digits
Scott Benner 1:25:04
did that 14 leave at a seven
Ed Damiano 1:25:07
6.80. Wow. So the 14.9 went to 6.8. That is not. That's anecdotal though. That's one data. We had. We had other people at agencies, you know, maybe have nine that dropped eight and a quarter and a half or something like that. So it's not everybody sees that remarkable reduction. But it is noteworthy that some do. Yeah. And again, it's a device with you know, that you initialize with bodyweight and you use meal announcements without counting carbohydrates.
Scott Benner 1:25:33
It's a very small barrier to entry. That's for sure. And I mean, obviously, you're talking about, like data that's at the end, this is the average, but not everybody achieved a seven. But that's still just worth bringing up that there's a 14 that came down that Fars is insane. Yeah. Do you think you'll get in other countries in any time soon? Or is the US? Yeah,
Ed Damiano 1:25:54
I think that it's quite surprising how things have changed through the pandemic. So it used to be that med tech companies would first target Europe, as you probably know, in diabetes, med tech was no no exception to that, where they would start in Europe, they'd get what's called a CE mark, and they'd start distributing in Europe. And then they would work their way into the US with a big pivotal trial, and then they get FDA clearance. We're doing it the other way around. So we got FDA clearance first launched in the US first, the next step for us to come to Europe or other countries owe us will be a CE Mark, what's happened through the pandemic is the CE CE mark process has really changed, it's much longer process it's taking, it's taking a very long time to get regulatory clearance outside of the US now it used to be quite the opposite. So it's certainly something we're going to pursue our goal at beta bionics is to bring this technology to as many people as possible, because it is a device that's made for as many people as possible is literally designed for that, that that kind of uptake in that kind of broad demographic adoption. So we certainly want to get this out to Oh, you owe us to Europe, Middle East, in other countries, other regions. And that will require first to CE mark. So unfortunately, that will take a long time just from a regulatory process. It's certainly more than a year's worth of regulatory review, right? But it's certainly something we're going to be doing.
Scott Benner 1:27:14
I have a fair amount of Canadian listeners that will be mad if I don't just say Canada like out loud. Of
Ed Damiano 1:27:19
course in front, you have to say Canada. Yeah, I mean, they are literally our next door neighbor. So we've got to Canada and Mexico, we've got to get out there right and you mighty mark is the way to start. He
Scott Benner 1:27:27
shouldn't have let all that smoke come down and choke me out or I would have been a little more. A little more feeling about it. Okay, so like I told you before we started recording when I first heard about this, I thought, oh my gosh, this is amazing, right? Like they're gonna have glucagon in the same pump. And you know, it's going to stop you from getting low. I have a couple of quotes. And I'm assuming the the major hold up was liquid stable glucagon, it had to had to exist in the pump for the amount of time at least you were wearing it. So now that exists. And you have access to that great. Does this my first like Boohoo? Like I don't know. Like question is, is glucagon doesn't work if you're drunk. Is that right?
Ed Damiano 1:28:10
No, that isn't right. So we had, we had looked into that specifically. So Stephen Russell did a clinical trial where he actually brought people into the clinical research center at MGH and got him drunk IV though. So we actually got a protocol approved and what he did was he we can infuse alcohol intravenously, and look at the efficacy of glucagon. Okay, Mike would that is microdose not big rescue doses, right? We're giving tiny, tiny doses of glucagon. Okay, I see. All right, and it's not Basal glucagon. It's not like every five minutes, you're gonna get a dose of glucagon. It uses glucagon sparingly and only as needed. But the dose you might get at any step where it sees your blood sugar might be your glucose CGM glucose might be dropping, or if you're already low, that dose could be one to 2% the size of a rescue dose that small, tiny little doses and he gave us doses that were comparable to this to the to the doses we would give in the in the bionic pancreas, the by hormonal bionic pancreas, and at different levels of blood, blood alcohol levels in the in the Clinical Research Center. And he was able to see that there it was, it was pretty much insensitive to the levels, alcohol levels you'd likely see out in the wild. So let's put it that way.
Scott Benner 1:29:20
So a rescue dose of glucagon might be different in that scenario, but the small amounts you were using were working.
Ed Damiano 1:29:26
Yeah, so he didn't test the rescue doses to see if that would be an issue. But definitely that because you know, the doses we're giving are so small, you're not depleting glycogen stores because what glucagon does is it breaks down stored glycogen in the liver, which is a stored form of glucose breaks it down and it liberates glucose into the blood and that's how it raises your blood sugar.
Scott Benner 1:29:43
I I'm asking this question way too ahead of but now we're into it already. So would do you foresee the eyelet being able to rescue if you're not drunk in a rescue situation like if it if it somehow thought this person's going to zero? Would it go for it? or would it?
Ed Damiano 1:30:01
Yeah, it wouldn't release the whole cartridge if that's what you mean, right? It wouldn't do that. And by the way, the amount of glucagon in this little tiny glucagon cartridge is much more than what you'd see in a rescue dose as well. It's a small cartridge, it's only it's only one ml. So it's like, think of a cartridge only 100 units of insulin. That's the size of this cartridge. It's really tiny. But the glucagon we're using made by Zealand Pharma is four times more concentrated and rescue glucagon, okay, it's four Meg's per ml versus one meg Parral ml. So you wouldn't want to ever unload that whole thing. It's really about a seven day supply of glucagon further, by the way, the islet uses it lasts about a week and that little cartridge, but what it would do is it wouldn't wouldn't actually give a rescue dose. But it would continue to give glucagon doses every five minutes if it doesn't see your glucose coming back up. And remember, it also turns insulin off just like the single hormone islet does. Yeah. So it's using both it's using the X gas and accelerator, I like to use that analogy of the insulin is like the gas and the and the brake is the I shouldn't say gassing sorry, should it break an accelerator. The insulin is like the accelerator and the glue guns like the brake. And so you really want to take your foot off the accelerator and hit the brake, if you want to slow down quickly. And with the single the biometric system, you have both at your disposal. Okay. So hopefully, you know, the amount of glucagon that it can give should really prevent any need for rescue glucagon, as long as it's flowing into your, under the skin.
Scott Benner 1:31:22
Is there an amount of time or a number of like, little bumps with glucagon before? Like, doesn't it eventually, like just empty your liver? And then there's just no more there anymore? Right? Yeah. And you're
Ed Damiano 1:31:33
not going to get to a point with environmental system where you get depleted unless you were very sick, right? So suppose you've been you had a vomiting illness and you haven't been getting, getting anything down for a few days, you could get into a situation where you're depleted of glycogen stores, and then there's no substrate upon which glucagon can act if there's no glycogen stores. But that's,
Scott Benner 1:31:53
it's hard to get to that stage. I was gonna say, and in that scenario, doesn't matter how you're managing, you'd probably be in the hospital one way,
Ed Damiano 1:31:58
I think you're going to be finding your way into the hospital in that situation. But what we did see in our, in our pre pivotal studies is that you're, you know, overnight, where you're getting just Basal insulin overnight, so you're not having a ton of insulin, which helps store glycogen. And you're not eating at all, and you've been fasted for a very long time. So you got no carbohydrates for seven or eight or 10 or 12 hours, right? Since you went to school, since you had your dinner went to sleep. When they got up at the morning in the morning at 7am. And they start becoming active and there, they there might start going low, you'd see these little shots of glucagon, tiny little micro doses at 6am and 7am. And it would pop them up. So that meant that even though they've been fasted for 12 hours, they hadn't eaten anything, and they've been getting very low levels of insulin, they still had plenty of glycogen upon which that glucagon could act, okay. And so we never saw any depletion of glycogen storage, any any evidence of that in any of the trials we did, and sort of routine day to day basis, but we've never studied, you know, pushing it to the limit to see how many days could you go fasting, before you'd run out of storage? I don't know.
Scott Benner 1:33:00
I just imagine that most people I try hard. But I imagine most people think that rescue glucagon is like sugar that brings up your blood sugar, and they don't recognize that it actually signals your liver, you know, etc, and so on. Like, I don't know how well that's understood.
Ed Damiano 1:33:16
I mean, if it were up to me, I would if it worked, which it wouldn't, I would rather push sugar than glucagon because it doesn't have to rely on that secondary source of sugar. Yeah, liver that could be depleted when you're sick. But there's no way to infuse tiny amounts of sugar under the skin and have it do anything, it's we really we do use the hormone just the way the pancreas does. That's how the Packers prevents hyperglycemia your first line of defense, people without type one, their first line of defense against hyperglycemia is glucagon. And it's it's, you know, we should not have the hubris to think that we can build a truly biomimetic closed loop system without adding glucagon back because people with type one diabetes lose their ability to use glucagon effectively. So when the when the autoimmune attack takes out the beta cells that secrete insulin, it disrupts the alpha cells ability to release glucagon, they still make look and they just don't release it in any coordinated and useful way anymore. So they really have a dual hormone insufficiency. And that should never be ignored. And so that's one of the things we do with beta bionics is not ignore that, right, we build an entire technology platform that will look just like this one, right? It won't be any bigger. It'll have, you know, we have we built a second chamber here to take a glucagon chamber. And this is actually the exact same platform that we'll be we'll be testing in the pivotal trial with two hormones system,
Scott Benner 1:34:30
when is that going to happen?
Ed Damiano 1:34:32
So our goal is to have that start by the end of the year. 2023. Yeah, so we want to have that trial start by the end of the year. Now that's that's that trial is is huge. So as I mentioned, the single hormone study the biochemical pivotal trial, the bionic pancreas, overdrive with a single hormone device was the largest automated insulin delivery randomized controlled trial ever done. Right. by a longshot. The by hormonal pivotal trial will be way larger, in fact, eight times larger in terms of the number of Patient years of exposure. So it won't be a three month trial, it'll be a 12 month trial, it won't involve 440 people. it'll involve over 700 people. And we're going to have phases. So if we start by the end of this year, the first phase will be a small cohort of 70 or so people. And they'll engage in a crossover trial with the final buy hormonal device, and the single hormone device that you see here. And people will use both and a crossover design. So they'll spend like four to six weeks in the in the single hormone, Iowa and four to six weeks in the biomedical crossover in random order. Once that study is done, that'll take about six months or so we read out the data. And if everything looks good, and we like the way the system is performing, we lock in and we start the big one year randomized trial. And a one year trial doesn't take one year to do. Because we have 700 people and 30 clinical sites, we have 16 sites and the other trial 30 sites or so it takes a half a year just to load everybody into the trial 700 people and then a year for the last person in to finish, that's an 18 month commitment. So it's a long road, right? It's going to be a couple of years, two and a half years just to get to the last participant last visit of the buyer model trial. And then you have to build the FDA package, submit that and they have to review not just the by hormonal island. But here's the big sort of the long pole in the tent, they also have to review the glucagon glucans never been used chronically, it's only used as a rescue. And so Zealand pharma will have to put in their own application for dasi, glucagon, as they call it, their analog of human glucagon, which is a, you know, a 12 month typically a 12 month review process with the FDA that will go in parallel with our buyer model Island.
Scott Benner 1:36:34
Did they have to wait for this first eyelid to be approved to do that? No, it's just no time? No,
Ed Damiano 1:36:40
no, it's just that we were just, you know, we couldn't do too many things at once.
Scott Benner 1:36:44
Is there anything about any patents you hold that would stop an insulin pump company from going to a duel hormone? Or?
Ed Damiano 1:36:52
You know, we have we have intellectual property portfolio that I think is pretty robust, not just in terms of the by hormonal, but also the single hormone algorithms. So yeah, I think that our on our algorithms side, we have some IP out there on bio hormonal, that's pretty robust. But ultimately, you know, we're not engaging in an exclusive relationship with sealant Pharma. So if somebody wanted to build a dual chamber system, you know, they'd have to sort of work around our IP and build their own algorithms. And they'd have to work with Zeeland pharma to figure out how they're going to use their drug. But it is true that if we do the pivotal trial with the Zealand pharma forming per ml concentration drug, that particular very specific formulation and get FDA approval of that any other pump company that builds a dual chamber system would not have to do as long a study with the Zealand pharma for a per ml formulation, right? Because it's been proven out to work in chronic use this way. So is this going to be two different infusion sets? Well, in the trial, it will be but ultimately, that's not our intent for the commercial product. So we can start the trial and do the pivotal trial with two separate sets. And all of our pre pivotal psychology studies use two separate sets. And they're both unit medical infusion sets one was an insulin and one was a glucagon and we put them right next to each other, they're about a centimeter apart. What will will ultimately want to do is build a single set that has, you know, a couple of cannula in it, you'd insert that and one go every like three days or so. But you will have two separate tubes that you could sort of tie together like speaker wire here like at the headphone jack wire. And the reason it's important that you have the two separate tubes is because the insulin cartridge might last three days on average, and the average adult say and the glucagon cartridge might last a week, they're not going to be changed on the same frequency. There's no reason to change them both out if one is still has a few days left in the cartridge. Oh, that
Scott Benner 1:38:40
makes sense. Yeah, I was just trying to like, like I always have, people always ask me for years. When is they always say they when are they going to put them on one device? And like, I don't know what you're thinking about like you want like a CGM and a pump in the same like structure, which I'm like, That can't happen like forget business. Like Like, like functionally it can happen. Right? Well,
Ed Damiano 1:39:04
the thing about it is that with Transcutaneous, CGM sensors, right, they typically are lasting 10 to 14 days, right. And we now have infusion sets out there that are FDA approved to be used for up to seven days. But they don't, on average, last seven days, that's what they can be used up to. But on average, they last shorter than that. Well, why is that? Well, mainly what fails with infusion sets very often as the adhesive fails. And you know, when you are infusing liquid you know insulin under the skin, that he's if you're and you have a tube that's connected to it as you move around that tube is putting a little bit of stress on the infusion set all day long every time you twist and turn and it's tugging on that set. And so the adhesive ultimately is overwhelmed. And after three four days, some people can run it out longer but other people can't and it depends on the to how the adhesive works with your skin and so forth. But generally Do you want someone who uses an infusion set for four or five days, we'll start to see if they go well beyond its intended use, we'll start to see the set fail. And what happens is the insulin starts flowing up around the cannula and wets the skin and doesn't go into the body. Yeah, but, but if you look at a sensor, you put a sensor in, and it's not nearly as much stress and pulling on it, there's nothing connected to it, right your shirt to touches it, but you're not pulling on it with a tube every now and again. And that adhesive can really last longer, and it's more forgiving the sensor under the skin, if it moves a little bit around relative to the tissue versus a cannula where insulin can then leak out. So they just they have very inherently different life's life scale, lifetime, you know, or characteristic time. So make and stay under the skin.
Scott Benner 1:40:42
I wish people could have seen you because I enjoyed watching the the engineer and you know, like, like the face up. Because I always think that when I always think like simple things, like, first of all, what do you like, you're gonna build us like, on the PA that has a Dexcom in it like that, how's that gonna happen? And what happens if your sensor goes bad in three days, but your pumps work or two days, but your pumps working for like, you want to rip the whole thing off? Okay, you understand the desire. But it's always I always feel like that question is asked by somebody who's never built anything before in their whole life. And, you know, yeah,
Ed Damiano 1:41:13
there's just inherently different sort of lifelines, or whatever you would say that sort of the lifetime expectancy of those two systems are so inherently different, and you don't want them to be coupled. Because you as you just said, you don't want to have to change all three, because only one fail right
Scott Benner 1:41:29
right now. Okay. All right. So I've had you for a long time longer than I promised. I have one question. And then I'm going to ask you, if there's anything we haven't talked about, a number of people asked me, islet how like, you know, 40 carbs of I don't know, a soft pretzel, and 40 carbs of cotton candy, 40 and 40. But significantly different impacts, it doesn't matter to the
Ed Damiano 1:41:54
eyelet. No, it really doesn't. Because as I mentioned before, it's because the corrections algorithm is always running in the background. So suppose you have what you're really getting at, I think is a food that's got a very high glycemic index versus one that's got a very low glycemic index and takes longer to raise your blood sugar. Or it's just more muted, right, you just don't grow up as much. It's just it's just, it's just extended out to over a longer period of time. So the island is watching every five minutes. And it has unlike, you know, most hybrid close up systems, it has the occasion, or the opportunity to dose every five minutes if it needs to, so it's always on top of it. So if you have something if you do the meal announcement for that 40 grams of cotton candy, you're gonna see a very fast rise. And the meal announcements going to kick in. And the the it's going to keep track, the islet keeps track of the insulin in that meal announcement dose that it just gave, and it watches the glucose rise, and it says, Okay, if you've got all this insulin pending, I'm keeping track of its rise and its clearance and I'm watching your glucose rise. Now if you rise very quickly, it might just stay in the background for a while there will come a point where I'll say I'm gonna add a little bit more correction insulin now because the correction algorithm has been quiet. But now I think you've risen high enough that the meal announcement insulin even insulates. Pending from that meal announcement isn't enough. So I'm going to add a little bit more, and then it's going to walk it's going to keep watching. And it's very patient because it's keeping track of the insulin it just added, in addition to the meal announcement insulin, and then it'll see it crest if it's a very high glycemic index food, it's just going to rise quickly and and stop and then start coming down. And it'll see it come down, it'll just back off your blood sugar, it could be 252 20. If it sees it slow down, it's going to back off, it doesn't care that you will hyperglycaemic It knows that insulin is coming, it's gonna be patient. And now what if instead, you didn't rise nearly as much, because it's a slow a low glycemic index food. Now you went up to 190 or 220, instead of 250 or 260. And it sees that meal announcements enough, it's really enough, I'm gonna stay back, I'm gonna stay quiet. And now an hour has gone by and you're sort of sit there and now you're, you're coming down to 170. And it's an hour and a half after the meal announcement. But you're still a little bit, you're still a little bit stubborn. And then it's gonna say, well, that meal announcements getting old now. And I'm the correction algorithm checking in every five minutes, I've decided at this step. Finally, that meal announcements not enough given that you're 170, I'm going to start adding a correction insulin now. And so it's very patient and looking at the meal announcement doses and how much your glycemic excursion has risen and how much it's coming. It's responding to that before it weighs in on adding more, but it's always there to add it if it needs to. And it's using very precise mathematics to make that very objective decision. It's it doesn't get it's not irrational, and it doesn't reach Bolus. But it does ask that you the user be patient. And what that usually means is don't keep looking at the iLet and expect magical results and say, oh my god, I'm still 170 Just let it do its thing and that is the ultimately the message that we want to convey to people who use the iOS is let it let it work. Don't Don't fuss over it too much. Just make sure you maintain the character Feeding have it. But let it do its thing and don't try and meddle with it too much. Because you know, it won't help. It doesn't it doesn't get better glucose control, just because you're watching it. And it doesn't get worse glucose control. If you don't watch it. That's another thing we learned from the pivotal trial, you don't have to look at it all the time to get the same equally good control. And you know, with other diabetes therapies, right? We know that the more you interact with a fingerstick meter, the more you interact with your CGM, or an insulin pen, or a pump or hybrid close up system. The the better your glucose control typically is if you look at a group of people interact frequently with their, you know, diabetes therapy, or diagnostic work or another group of people who interact infrequently with it. Those who interact frequently tend to do better statistically, we don't see that with the island, we see that it's pretty agnostic to how much you engage with the device as long as you're taking care of it. And that's, that's a really, I think, a really important point to remember. No, it's a bonus for sure. And the other thing, the other thing I'd like to just was sent heavy. No, go ahead. No, no, I
Scott Benner 1:46:03
was pleased. You're fine. I'm I'm trying to wrap you up. If you want to keep talking. I'm happy for you to keep talking. I just tried to help you out of this.
Ed Damiano 1:46:10
Oh, yeah, very good. I do. I do have a call coming up in a few minutes. But I would say this that what is unique here with the there's a number of things about the AI that I think that are unique, right? It's unique in several ways. But importantly, it determines 100% of every therapeutic dose of insulin. And there's that it's not a system that where you can go and override the dose, you can't add a correction insulin Bolus, you can't add a meal dose, you can't say I didn't give myself enough insulin. So I'm going to add 20 grams of carbs. And it's going to then figure out what to dose which is what a lot of people do in some of these systems. It determines 100% of every therapeutic dose, and you don't override that and your physician doesn't override that. So that is that is not a hybrid system, the hybrid system inherently means that you are playing a role in insulin dosing decision, you and your physician as well as the audit some automation. That is not what's going on here. And as a result of that, you have to really get comfortable with this new world of fully automated insulin dosing decision making, right? That's being added to a device,
Scott Benner 1:47:10
I have to say, I'm actually impressed. And I think it's smart that you're talking about it so directly. Like you're not doing any like marketing, like talking around, you're like, look, this is what it does. If that's good for you, then great. And if not, was nice talking to you. Like I think that's terrific. I don't see enough of that. I've been very impressed with that the entire time you've been going over this. So I really do appreciate, do you think there's anything we didn't talk about that we should have?
Ed Damiano 1:47:38
Not really, I mean, I guess it's more more re emphasizing this idea that the reason I think we can talk so frankly, about is first of all, we want to build technology that's in the best interest of people with type one, we've always been committed to that. And that means that you know, the eyelid, I think is that device that is very complimentary to some of the high tech out there that does serve the interests of the needs of those people who are already in good control, or who have the best had the access to the best resources, the best health care. And, you know, and so we're trying to, to address that other segment, which I think so happens to be the majority of people with type one who don't have all the resources and all the access to the best health care, you realize that you know, 75% of us counties do not have a single endocrinologist in their borders, right. This is something that the Ozeri has published a few years ago. And nightly, whereas 95% of people go counties in the US. So 75% of counties don't even have one endo 95% of counties have at least one primary care physician. Primary care physicians can't use that high tech, it's just it's anathema to them. I mean, they don't have, they don't have the resources, they don't have the staff, they don't have the training, they can't use that tech. But we think the eyelid is a device, a very high tech device that is really the first device that plays very well in primary care, because it is for that large 80% of people who aren't meeting goal. And I do think for those who are meeting goal, many of them will still prefer the island, because they're going to be unburdened of a lot of that cognitive effort and into the vet burden that goes along with constantly being all over your diabetes management all the time. And there'll be others in that same group who just you know, are just too anxious to give that control up. So it's really all about finding those people that that are going to benefit from it and who, who will be able to do that comfortably. And I think he's just a lot of people out there that that we're trying to serve.
Scott Benner 1:49:39
Will you be adding salespeople? I mean, it's because it feels like you're going to have to go to non traditional doctors to talk. Yeah, you know,
Ed Damiano 1:49:48
yeah, we have so we have a very small group at start. And so we've got about 16 people on the sales team right now who are focused in those eight territories I was telling you about and what what we've been doing these past couple of months, Stephen Russell and I I've been going to all those territories with each of the two sale the two commercial people in each of those regions, and meeting with the clinical sites that we targeted in those regions to launch the product. And spending a few hours with each of those clinic clinical teams, and with our commercial team with us, at each of those sites, so that they get introduced to these folks that we've been working with for years, frankly, are as many of them. And so that's how we're doing it at first is we're sort of introducing the commercial team, to the people we've worked with over the years in the clinical setting the clinical research setting, and ourselves being introduced to the clinical people who aren't doing clinical research, but who worked with our clinical research scientists, collaborators. So that introduction is happening. And that's where we're focusing the launch. And then as we get experience in those eight territories in the fall, then we expand more territory, more territory. So we've been doing a lot of traveling, getting on the road and seeing a lot of these sites and moving back and forth across the country, you know, 17 sites in the past nine months, nine weeks.
Scott Benner 1:50:53
It's amazing, busy pace, because you're gonna go to the trouble of I mean, listen to this story, how long it took to make this thing. And now it's the last piece right? Like, how do you how do you set it in someone's hand? And it's not apples to apples, but I'm a person who's trying to deliver something to people too. And you would never know it? If I wouldn't say it out loud. But that's the hardest part of this job. It's making the thing is great. And then giving it to somebody is it's the hard part, you know?
Ed Damiano 1:51:20
Yeah. And that's it's all about scalability. So I'll leave you with this. This notion, if you think about what the diabetes control complications trial did between 1983 and 1993, was to test the hypothesis, right? This was a landmark study, many clinical sites across the country took 1500 people and randomized about half of them into conventional therapy, they called it which was not multiple shots a day, or insulin pump therapy was just one or two shots a day. And that and or intensive therapy where they were checking their blood sugar seven times a day, but importantly, they were giving multiple shots many times a day or using a pump. And what they found was they could dramatically reduce me mean glucose and HPA when see in the intensive therapy group and sustain that for a period of you know, six and a half years on average for each person at a huge effort on the part of the patients who randomized to intensive therapy, and the physicians that supported them, the clinicians that supported them. And they were testing the hypothesis back then it wasn't known that good glycemic control was necessary to stave off long term complications of diabetes. That was that was a contested point back in the early 80s. And until we had the HBA when C test, and insulin pump therapy and fingerstick meters, we couldn't really test the hypothesis. You know, if people take a bunch of people and control their glucose, well, do they have fewer long term complications and those who you don't. And resoundingly, the DCCT, the diabetes control and complications trial showed us it by 1993, that huge, markedly reduce long term complications. And that study took about 10 years to do. And 30 years ago, this month, it was read out to the diabetes community that you got to do this. Well, you know, 10 years after that study, we started building the bionic pancreas. And in that 10 year period, and we've been doing it for 20 years, but in this period after the DCCT, what we also found is it's not scalable, you can't do what the DCCT did in a large scale everybody's anyone see is more like in the eights low eights, not seven, which is what they're able to do with the DCCT. So it wasn't for 30 years after the DCCT that there's a device now that we think and reach broadly, a much larger audience than than most diabetes tech people with type one, that is something you can put in your pocket and you type in your body weight and do these few meal announcements a day and keep it going and get glucose control that's comparable to what the DCCT achieved in the intensive group. And so we sort of answered the question, is there a scalable solution here? And I think the eyelid is that is that device now
Scott Benner 1:53:41
it sounds like it. I mean, I've really appreciated you telling me so much about it, but I'm excited for you to to get it going and get it out there. How long do you think it'll be? It's June till I see somebody online going. I use an eyelet. Online Yeah, like sighs thanks a picture on their Instagram. Like when am I going to see that? Like,
Ed Damiano 1:54:01
in a month, really less than I think in a month? Okay. I think we'll have one or two people at the ADA conference next week on the island. Okay. But on Instagram, I think, you know, on social media, I think you'll see something come up within the next
Scott Benner 1:54:11
month. Pretty amazing. Okay, Ed, thank you so much. I really appreciate Of course, Scott.
Ed Damiano 1:54:15
Thanks for having me.
Scott Benner 1:54:26
Hey, huge thanks to Ed for coming on the show today and telling us all about eyelet. I also want to thank us med for sponsoring this episode of The Juicebox Podcast. I'll remind you to go to us med.com forward slash juice box or call 888-721-1514 To get your free benefits check so you can get started with us med check out that private Facebook group Juicebox Podcast, type one diabetes on Facebook. It's absolutely free. It's for everybody. I don't care what kind of diabetes you have. I don't care how you eat. There's a beautiful community there with over are 40,000 people in it waiting for you? This podcast is sponsored every week buy great companies. I'll list them in a moment. But if you have the need or the interest, please use my links. When you're finding out more, it really does help to support the podcast. If you want to check out the Omni pod Dexcom us med that contour next gen blood glucose meter Chivo hypo pen, athletic greens, cozy Earth BetterHelp touched by type one, they're all there. Just look in the show notes of the audio app you're listening in now or go to juicebox podcast.com. When you click on those links, you're supporting the production of this podcast and keeping it free. The podcast is sponsored today by better help. Better help is the world's largest therapy service and is 100% online. With better help, you can tap into a network of over 25,000 licensed and experienced therapists who can help you with a wide range of issues. Better help.com forward slash juicebox. To get started, you just answer a few questions about your needs and preferences in therapy. That way BetterHelp can match you with the right therapist from their network. And when you use my link, you'll save 10% On your first month of therapy. You can message your therapist at any time and schedule live sessions when it's convenient for you. Talk to them however you feel comfortable text chat phone or video call. If your therapist isn't the right fit for any reason at all. You can switch to a new therapist at no additional charge. And the best part for me is that with better help you get the same professionalism and quality you expect from in office therapy. But with a therapist who is custom picked for you, and you're gonna get more scheduling flexibility, and a more affordable price. I myself have just begun using BetterHelp betterhelp.com forward slash juice box that's better help h e l p.com. Forward slash juice box save 10% On your first month of therapy. Thank you so much for listening. I'll be back very soon with another episode of The Juicebox Podcast
