David Klonoff: 0:14

Welcome to Diabetes Technology Report. I'm David Klonoff. I'm an endocrinologist at Sutter Health and UCSF. I'm here with my co-host, Dr David Kerr, who will introduce our guest today.

David Kerr: 0:28

Thank you, David. This is David Kerr. As usual, I'm speaking to you from Santa Barbara, California. Today's special guest is Dr Jose Garcia-Tirado. Welcome to Diabetes Technology Reports. One of the things we like to begin with in these podcasts is to get a feel about you as a person and how you ended up in diabetes and diabetes technology. So can you just give us a little thumbnail sketch of your day job and how you arrived there?

José Garcia-Tirado: 1:00

So, first of all, thank you so much, david Klonoff and David Kerr, for this kind invitation into the Diabetes Technology Report. Well, I myself am a type 1 diabetes person. I was diagnosed 11 years ago when I was doing my first postdoctoral fellowship in Germany. Back at the time I didn't know anything about diabetes, but my background has always been in control systems engineering. So I have bachelor's, master's and PhD degrees in control systems engineering. So I was fascinated back at the time to controlling biochemical and chemical processes that are very, very complex in nature.

José Garcia-Tirado: 1:50

So then, when I got diagnosed, I remember my diabetes educator telling me something that still keeps in my head. Keeps in my head If you want to succeed with this tough disease, you need to know more than your medical doctors, you need to educate yourself, you need to read a lot and you need to become a master in this disease. And then I started to read and then, some months after, I went to a control systems conference and, to my surprise, I found Dr Graham Goodwin from Australia doing a keynote presentation about how control systems engineering could help shape artificial pancreas technology. That was 2014. And then I got fascinated because I have my two main passions one because life brought to me and the other because I choose to be together. So I decided that I wanted to know more and more and more and I started to conduct research in my hometown in Colombia and, right like shortly after, I joined UVA with Marc Breton. And then my life continued doing research in type 1 diabetes and diabetes in general.

David Kerr: 3:21

And so what is it? I mean, I know there's trade secrets and everything, but just give us a feel about what are you working on today and in the near future. What's what gets you up in the morning?

José Garcia-Tirado: 3:33

sure. So, um, as a as a engineer, uh, I hear a lot of things with from medical doctors and from practitioners, of course. You hear a lot from your colleagues. You hear a lot from the flagship conferences ATTD and ADA and EASD. So you find a group of physicians say, well, aad systems are done, there's not much to do, they work pretty well, they work for everyone. Then you find another group of physicians say, no, we can do better. We can still do better.

José Garcia-Tirado: 4:07

Now, there are a lot of things that are not resolved in type 1 diabetes. One of the things that wakes me up every morning is knowing that, disregarding how good we do glucose control, we have still 10 times more risk than a healthy person to have a stroke or to have a cardiovascular or renal disease. We have no current therapy protecting people with type 1 diabetes from cardiovascular or renal disease, and that's why, apart from my passion in AID systems, I believe I strongly believe that we need to do more with adjunctive therapies. We need to bring those fascinating adjunctive therapies that we know work for other populations close to the type 1 population and see what happens with, of course, with thorough research.

David Klonoff: 5:07

Jose, you're working on trying to normalize glucose levels. What do you think is more important to have a normal mean glucose or to eliminate variability?

José Garcia-Tirado: 5:20

So that's a very tricky question. I feel glucose control is not enough. I feel, from several documents and papers out there, Pratik Chhatrik himself has written a lot about glucose variability. It's very impactful, but we don't have long-term studies showing us that. So I might believe it's a hunch, but we haven't been able to demonstrate that actually, disregarding that you have 6.0% A1c or an outstanding 70, 70, 80 percent I mean range, you still have risks of down there, so retinopathy or other macro and micro vascular related diseases. So I feel we need to address glycemic variability.

David Klonoff: 6:26

One way to address glycemic variability would be to eliminate the hybrid part of the hybrid closed loop, make it fully closed loop. How close do you think we are to that type of system?

José Garcia-Tirado: 6:40

Well, I myself worked very hard with Mark Breton back in Virginia to make that happen. We made that happen in pilot studies, so we know it is possible, even with the limitations we have with technology, with current CGMs and current insulin analogs. I believe that we can even make it better or best with adjunctive therapies like GLP-1s or dual GLP-GLP-1s, because that gives us some time to respond to meals, to react to meals. So I feel it's doable and I feel more studies are warranted, of course, and we need to go in that direction.

David Kerr: 7:26

Jose, I'm intrigued by your point about. You come across clinicians, some of whom say, well, type 1 diabetes, give everyone a closed-loop system and everything's going to be great. And then there's clinicians, like me actually, who kind of worry that, okay, it's not all about glucose. There's a whole lot of other stuff going on, but just from the glucose perspective, what are the next areas of progress we need to make with a closed loop? I mean, I know about a fully automated, but is there anything else? I mean, the burden is still there for people with type 1 diabetes. What are you working on and what are you thinking about that could reduce the day-to-day burden for people?

José Garcia-Tirado: 8:05

Yeah, I might need to push back a little bit on the previous comment from David Klanov, because I don't think necessarily with full automation we will reduce glucose variability. We might have even broader glucose variability and that's something that we need to address. So one thing is removing the patient from the loop. I also believe that we can even try to have systems that work in the background that are not annoying the person so often, just in the cases that we need action from the user. I am of the thing that you don't even need to show the CGM if you're not in danger, unless the person is like a control freak and want to see everything. Because we also need to personalize and there are some people that needs to be controlling every aspect of their disease and the treatment, but there might be others that can live their lives in a cooler way, if I may say, and then just bring their attention when they really need to react and they need to intervene.

José Garcia-Tirado: 9:20

So I think one of the things that we're working very hard in our lab is to try to remove the user from the loop, not only from the meals perspective but also from the physical activity perspective. That is also very challenging. So this is one thing. The other thing is try to work in playotropic effects, try to work in other aspects of the disease that we haven't been successful, like trying to adhere more to physical activity, and try to improve risks. This is, in my opinion, things that we need to address in the next few years.

David Kerr: 10:01

I mean that's very interesting. So the idea being that people with type 1 diabetes being brought up with CGM and the word continuous and continuously reviewing their data but what you're saying here is that it doesn't have to be like that. We can remove that burden and allow people. Free up time for people to get on. That's absolutely fascinating. Free up time for people to get on that's absolutely fascinating. I've got to ask you, because it's the hottest subject in my world what's AI going to do for closed-loop system? I mean, is it all hype or are you excited, or are you going to buy shares in an AI company?

José Garcia-Tirado: 10:39

Definitely it's stirring things up. Definitely I'm not an AI expert, but I work very close with AI experts to really deploy big tasks to AI, and I want to be a little bit more cautious, but I think we're going to get into a point where AI is going to make big things, especially because we're getting more specialized and more smart, in the sense that not only the algorithms are capable of doing fantastic things, but they are also being capable of diagnose themselves and, like adapt to changing and you know, difficult environments. So I believe, if you ask me, I prefer more a hybrid approach where you have AI systems where we know they work best and we also have other engineering strategies or approaches where we have years and years of experience and we know what they can do best. So I'm a little bit cautious, but I feel AI is going to make a big impact in the future.

David Klonoff: 12:12

Jose, could you tell our audience what is your current position? Where are you working? What type of a lab do you have? I know you are in Virginia and now you're in Switzerland. Could you explain it so people know where you are, what you're doing?

José Garcia-Tirado: 12:38

group, I think one of the best equipped technology-related or technology-oriented groups in the world, with two fantastic leaders, mac Breton and Boris Kovachev, which I'm very grateful for allowing me to work with them. But two years ago I moved in Bern, in a fascinating city in Switzerland, to start my own group and to start my tenure track journey. Let's say I started from scratch. So I started from scratch, hiring students, postdoctoral fellows, software developers, to try not only to get our technological infrastructure but also to try to work very closely with clinicians to go for clinical trials. That's, I think, as an engineer, knowing my limitations that's a thing that I love to support, jose.

David Klonoff: 13:44

many people want to use an accurate continuous glucose monitor, but I've also heard people say that if it's part of an AID system, it doesn't even matter with the products on the market, because even if the monitor is not extremely accurate, the insulin delivery system and the algorithm will make up for it. So therefore, people don't have to search for the most accurate monitor. Now, I'm not sure if that's correct, but I've heard people say that. What do you think?

José Garcia-Tirado: 14:10

Well, I need to be against that claim in a sense. One of the things I learned from school, from my control systems background, is you cannot control what you don't know, and it's very difficult for an AD system to know where it stands if there is no accurate measurement of the environment, and the only thing we can measure nowadays is glucose. I'm of the thought that we need to push for more biomarkers, like if we had, like, a multi-sensing technology that we can have ketones and CGA, or glucose and lactate and other species that can help us to know where the body is at the moment. I think that will be best, in my opinion.

David Kerr: 15:13

Jose, just following on from that, this is a kind of philosophical question. But are we stuck with the form factor for closed-loop artificial pancreas systems? As we add more analytes, are these just going to get bigger and uglier, or do you think we're going to get into the miniaturization soon? Is it going to be implantable ever, or do we even need that? What's your kind of thoughts from the perspective of someone with diabetes? What to look forward to?

José Garcia-Tirado: 15:47

Yeah, that's a very, very interesting question. I think there are two sides of the coin. So one is injecting more than one hormone like insulin and glucagon and insulin and other hormone. That makes the system bulkier because you need to pump in, or more pumping mechanisms, and that gets difficult In the side of the sensing. I believe that technologies like laser technology and optical technology have a strong potential to make an all-in-one and miniaturize the technology. Of course that doesn't happen overnight, but we need to get into a point where we know what technology is capable of detecting several metabolites and then we can extract that technology and miniaturize. I'm more in favor of that one than into the multi-pumping technology, if I may say.

David Klonoff: 16:49

Jose, which project that you're working on now would you say is your most interesting or impactful?

José Garcia-Tirado: 16:58

Well, the artificial pancreas project is dear to my heart, but I think the one that is most impactful nowadays is one that we got accepted and approved by the authorities in Switzerland to pair Munjaro with current AID systems. So we would like to know what happens if a person with type 1 diabetes, and whatever system, is in the market. What happens if we add a low dose of Munjar? So we want to collect that data and see if we can truly see an impact in glycemic control and some markers of cardiovascular and renal disease.

David Klonoff: 17:41

We published a consensus report on that topic in Journal of Diabetes Science and Technology recently. Viral Shah from Indiana was the first author. Well, Jose, I would like to thank you for speaking with us. I've learned a lot about automated insulin delivery systems. Diabetes Technology Report is available at the Apple Store and Spotify and at the Diabetes Technology Society website. So to our listeners. So long for now, and we look forward to the next podcast, Jose. Thank you, David, thank you.

David Kerr: 18:14

Thank you very much, Jose Sure.

José Garcia-Tirado: 18:17

Thank you so much for having me.