Dexcom Exec Interview: Dexcom G8, Sensor Fixes & Upgrades

It's no secret that Dexcom has lost some trust in the community after some Dexcom G7 debacles over the past few years. This has led to users clinging on to the Dexcom G6 as long as they could until it was eventually phased out just recently. Dexcom got the message. Last week the company published a report highlighting feedback from a recently formed advisory council on recurring issues with Dexcom sensors and its plans for change to gain community trust back. And today we're speaking with Dexcom to learn more about the fixes already implemented and what comes next with products like Dexcom G8. Welcome to the show. I'm Justin. I have type 1 diabetes and I'm here I talk all things diabetes tech news and research with industry leaders, those building the tech themselves and the people using the tech. Dexcom's council is made up of users, caregivers, healthcare professionals and community leaders. I'm actually lucky enough to be part of the group. So far we've met several times this year and the company plans to continue this open forum to identify product issues and attack them head-on and be more transparent with the community. I sat down with Jake Leach, Dexcom CEO and Peter Simpson, executive vice president and chief technology officer about the improvements Dexcom's making now and in the future. We get into the motivation behind the council, the key takeaways and how they're addressing them, new developments including updates to the G7's adhesive, algorithm and accuracy enhancements as well as manufacturing updates and we dive into all the updates coming to Dexcom G8 from 15-day wear being the new standard for Dexcom to its self-adapting technology that's two decades in the making. Keep in mind that anything you hear on this show is not medical advice and is for educational purposes only. Always consult with your healthcare team before making changes to your treatment. All right, let's get into it. Peter, Jake, welcome to the show. Good to have you here. Thanks for coming on.

Let's hop right into it. Let's start with this advisory council that I am part of. So, thank you for having me as part of that. It was really great. It's been great to meet with all the other people in the council, get to talk directly to you, hear from you about all the work you're on. Kind of just explain to us like the reasoning behind starting the advisory board.

Everyone brings in such a unique perspective. Like I was learning things that I never realized because there was a parent or caregiver in there or a healthcare provider, and then I was able to bring my own experience, and that's what I love about what I do is I get to use my platform to ask questions, have you on here, and get to know more about what's going on or push for things to go on. What's some of the overall consensus that you heard coming out of that?

Peter, did you have something to add or?

Yeah, first of all, really appreciate the opportunity to listen and get the feedback from some of our customers and some of the people who really interact with a lot of our customers in the field. A lot of things that I took about, you know, first of all, product quality has always been absolutely critical and essential and, you know, one thing that just we get reminded of is how important it is to continue to earn the trust of our customers with the best possible product quality. And we've been focused on that for many years, but a really renewed focus over the last year to make sure that every product that we're making, every sensor that we're making is of the best quality and making sure that the customers understand that that is as essential and we see that as just absolutely the highest priority from what we're doing and in what we're producing. I'll reflect also on the communication part where, you know, we have a process of continuing to improve our products and continue to innovate. We've got new patch technology, we've got new antennas coming to the G7, but what we heard was people want to know when that's coming and when have those changes happened and more of an announcement from our part to say that, hey, you're going to experience something different and give us the feedback and we'd love to hear the positives and if you're finding any issues with these new changes, please tell us. I think the other thing was just clarity on our replacement policy and just making sure that everybody knows that if there's ever a product failure, we will replace it, no questions asked. We just go and it happens. There's no limit to that. I think that's another big thing that maybe was confusing for a little bit. And then finally, there seemed to be a lot of interest on how the technology fundamentally works and especially how it interacts not only on the bench top and how to make it great on the bench top, but how it's a challenge with different physiology and different people's biology and where that's some of the fundamental limits of the technology that we continue to improve on, but that's where some of the issues that people experience come from. I thought that seemed to click with a lot of people and really resonate.

Yeah, while we're on that, I found it fascinating when you educated us on the equilibration process of a CGM and kind of the lifespan and how the body interacts with it. Like I don't want to get stuck on that for too long, but can you just explain to my listeners like kind of all of that physiology that's happening there?

Yeah, so in contrast to a lot of other devices like the point-of-care devices and such, our device has to work under the skin in the body in a very stressful environment that, you know, the body's throwing all sorts of stuff at it. And it really a lot of the interaction happens in two places. The first place is when the sensor is first inserted under the skin. You know, we are creating a very small wound there and there is something called a wound response where the body is trying to clean up the wound environment and in doing that, there's a lot of metabolically active cells that are gobbling up glucose in that local environment. So while the sensor is working perfectly fine, sometimes the glucose doesn't always get to the sensor. And so that's one of the challenges that some people experience on day one. And the interesting thing is it doesn't happen to most people, it happens to some people and it's just depending on their biology and the particular of that wound site. So it's something that we continue to improve upon, but it is a fundamental biological response to the sensor. And then after day one to about near the end of the use of the device, it's kind of the sweet spot. Everything is very stable. But the last thing that the body tends to do is it tends to then encapsulate the sensor. It sees it as a foreign object and so it's encapsulating it and trying to starve it of glucose and oxygen. And that's exactly what our sensor needs to function. And when it starts doing that process, we do have algorithms that can detect it and so sometimes you'll see a sensor get shut off early. It's because it's detecting that issue that's happening with the biology around it. And so, again, it doesn't happen for some people, it never happens. They get the sensor to last all the way up to 15 days. And then some small percentage of the people it does happen. And so, their sensor will be a limit in their use life. So, again, another biological problem that we are continuing to make improvements on with our sensor technology, but is a challenge for CGMs in general.

So, you put out this report and it says that you're going to enhance sensor accuracy and reliability in the near term with several updates and enhancements. What are some of those enhancements or updates?

So, within the G7 family, we're continuing to make improvements. So, one thing just in terms of product quality, we've been working very closely with manufacturing and I've personally spent a lot of time at the sites with the manufacturing team to make sure that every product we're making is of the best quality. And, you know, we've seen some significant improvements there. And our accuracy complaint rates are at an all-time low. So, we've got some really significant improvements that we've seen just in making products more consistently. And, you know, that's always a challenge when you scale. And we've just really doubled down on that to make sure that our product is of the best quality we can make. Now, there's some specific improvements that we are rolling out now and that we're going to continue to roll out in the next year. One of those is a challenge with the reliability of the product, specifically the patch. And, you know, our patches are pretty good. But you can have some, again, very different biologies. People have different skin types. So, some people it works great for 15 days. Other people it starts to peel off. And we've developed, we've worked with our vendor partners to develop a new patch that's even better than the current one. And it's significantly better in our testing. It's about 11% more reliable than the current G7. And so, very excited to get that out on the field. And that's rolling out right now. We're starting to produce those and ship those. So, that's an improvement in reliability out to that 15 days. And then there's also some improvements in antenna technology that we're working on that will improve particularly the cross-body communication. You know, when you're communicating to your cell phones that's usually a pretty solid communication. Pump partners, that's when it's essential controlling an AID system. Sometimes if you have your sensor on one side and your pump on the other side you can have some challenges of communication. Cross-body is basically mostly water and that blocks that BLE signal.

And so does Bluetooth only work like in a straight line? It doesn't really curve around?

No, it can curve around and Jake actually has some expertise in this but it can reflect off of surfaces and such. But the cross-body is one of the more challenging and so what we are developing is a new antenna that does a little bit better there. Now with pumps it's always a two-way street because you need a good signal sending out but you also need to have a good antenna on the other side to receive it. And so we're also working very closely with our partners to make sure that the product as a whole is working the best it possibly can. But let me pause here, Jake, do you want to add anything?

This episode is sponsored by Omnipod. You guys know that staying connected to the tech world is important to me. But, what's also important to me is my insulin pump. Omnipod 5 is tubeless and waterproof, which means I never have to disconnect from automated insulin delivery, even when I take a shower, go swimming, or go to the gym. Staying connected to my insulin pump means I'm spending more time in automated mode, which ultimately helps me spend more time in range. See if you're eligible for a 10-day trial by visiting omnipod.com/diabetic. On there, you'll see full safety and risk information and trial terms. Again, that's omnipod.com/diabetic.

So, the G7 15-day has an updated algorithm, improved accuracy. I've experienced it myself, and I can attest to that. Will we see any of those improvements in algorithm and accuracy come back to the 10-day? And if not, why?

Yeah, so first of all, probably what most people will notice is the accuracy improvements coming from the algorithm. And part of that is, you know, when we develop these algorithms, we use the data that we have on hand to tune all the parameters. And as we get more experience and more data sets, then we have the opportunity to fine-tune those parameters even more. And so, what you're seeing is our latest and greatest version of the algorithm. Physically, one of the things you'll notice is that the data is going to be much more consistent and smooth in the G7 10-day. You can see a little bit of choppiness in the data, and we've been able to develop some technology that really makes that more consistent and smoother. And then as we said as well on that 15-day, there is a new patch technology that's coming out with it right now. So, that's going to be, I think, being adopted across the board for 10-day and 15-day as well.

Got it. So, now let's get into the Dexcom G8. I'm sure many people listening are ready for that. But before we get into that sensor, at your investor day, you mentioned that the new standard at Dexcom would be 15 days. Does that mean that G8 will be 15 days for everyone?

Yeah, it definitely comes down to education. What you were saying about like how it says sensor failure, but it's really just like the natural physiologic response for some people. Like my body does not respond that way. These sensors last the full lifespan on me 99.9% of the time, but I know for a fact that's not the case for many people that are using it. And so perhaps even on the app like there could be like an information, like it says sensor failure, but you tap it and then it explains, 'Here are the reasons it may have failed. One is this is how the body responds.' Because people think that this technology should be perfect, but the fact is we have come so far with CGM, but it's not perfect. You know, and if you look at where CGMs were just five years ago, I mean we've come so far. So yeah, I think that could work wonders. So at the investor day you mentioned that the G8 would be 50% smaller than the G7, and that it would have a self-adapting technology. Can you explain to us what that is exactly, and how that works? And I guess how it's different than any self-adapting technology that may have already existed on G7?

Peter, did you have anything to add to that?

Yeah, Jake did a really good job of explaining it. You know, maybe just let me touch on a couple things. The way factory calibration works right now is you get information, you characterize the sensor while it's being manufactured, and then you put a code in that, and that code then goes with the sensor. So, when you put it on, it's saying this is what the sensor should be doing, right? And so, that's its factory calibration. Now, that works often very well, but sometimes the sensor is doing something different than you expect it. Again, G7 has the unique capability of allowing users to calibrate it. So, for those sensors that are a little bit off, you can adjust them. What this GA technology does is it's got this auto-sensing capability, which every 5 minutes it's looking at the sensor to check if it's accurate. It's basically doing a quality check on the sensor every 5 minutes. And if it sees it slightly off, it will adjust to that sensor and bring it back into the target zone. And then the other thing it can do is if it looks at it and sees it's very far off, if there's something catastrophic that happened to the sensor, then it can do one of two things. If it can correct for it, it will correct for it. But if it sees that that difference is too far off, it'll actually fail the sensor and it'll just protect the user and make sure the user's getting the best quality data. And that's, you know, in the past in G7, we'd have algorithms that would look for problems with the sensor like we've mentioned before, but it wasn't directly interrogating the sensor to know that there's something wrong. So, we think it'll add an additional level of protection for our users, better quality overall, less outliers, and just give CGMs that next level, you know, step in terms of quality, reliability, and consistency.

If you have diabetes like me, you know the extra hassles that come with using the tech, especially CGM sensors getting knocked off, loud alarms, and managing backup supplies. That's where the Eversense 365 CGM comes in. A small sensor is placed under the skin and works with a removable transmitter you can take off and put back on without wasting a CGM. Both the sensor and transmitter last a year. Anything that helps reduce the burden and chaos of diabetes is a win in my book. One year, one CGM, 365 days of uninterrupted glucose readings you can rely on. If you're using insulin, see if you pre-qualify at eversensecgm.com/diabettech.

You know, with this smaller technology comes more to me more fear that connection won't be as good that it is already today and I know you're working on that updated antenna. I'm curious how you update that antenna yet put it in a sensor 50% smaller. I know that it's not entirely up to you which as you said earlier that it's also up to the receiving side of these pumps. Insulet has come out saying Omnipod 6 will have a stronger connection. They're actively working on that. Are the other pump companies working on that? Do we have to wait until there's a Mobi 2 or a Tandem t:slim 3 for these to be worn on separate, on opposite sides?

Well, I think one thing I'd say is I know that all of our pump partners are working on connectivity enhancements just like we are. And many of those can be done in firmware which is basically the software that controls the radios in the devices. So, there's things you can do there that don't require you to make physical changes to the hardware and I know the teams are working on that as well. And so, what I would expect is you're going to see a stream of improvements just like we've done with G7 in terms of it getting better and continuing to enhance and then ultimately, you know, some hardware changes probably push it even farther. One of the things I just mentioned about the size, one of the nice things about the 2.4 GHz frequency is, you know, with the size of G7 and the size of G8, with the right engineering you're not really having to give up a lot, like we used to when we were operating at different frequencies. The original Dexcom devices were more like 400 MHz, which is the type of frequency where it's big changes in size really do impact you quite a bit. So, we do anticipate the G8 will have, you know, as good or better performance than G7 when we go to that small size. We're very mindful of it. We obviously know how important connectivity is and we want it to be a competitive advantage for us. And so we're very focused on making sure that we, you know, continue to improve it and make sure that with G8 it's even a step farther.

Yeah. And now at the investor day you were asked about the overpatch and I believe you said that an overpatch would still be required on G8. Can you explain to us why not just make an adhesive that's slightly bigger on the actual device so that it, because what you know, when you look at one of your main competitors, there is no overpatch required. And it, I mean, stays on me, but they both stay on me, but my skin is very easy going. Like I said, not everyone's is. Can you just explain the overpatch to us, why you feel it is required still for this next sensor?

Well, just one clarification. It's not. G8 is the intent is it not requiring an overpatch. Available, yes. So, one of the things that we do know is some users really do like having the overpatch and at this point in time, with G7 it's a required part of the labeling. You know, users decide what they want to do themselves, but there's also quite a few aftermarket patches that are used as well. And I think one of the things that I certainly recognize over time is everybody as a user figures out what works best for them. Just like you're mentioning Justin about your experience with the sensors. I think everyone has their personal...

Yeah.

Sounds like it. You got sensors that go 15 days and they stuck.

Which is the majority of users do have that experience, but not everybody, right? So, we want to design a product that, you know, speaks to everyone. And so, the idea of having overpatch capability is important, but not required. And so, the goal is with G8, we don't want it to be a required aspect. And I do feel that the engineering and the team has done with the patch and the size of the patch. And we've actually done quite a bit of studies on as you skirt size as we call it. It's the area around the patch around the wearable. We've done quite a few studies on. Bigger isn't always better. But, there is a sweet spot that Peter and the team have found.

Yeah. Now, you also mentioned multi-analyte sensing and specifically with a focus on potassium as one of the first, you know, secondary analyte or biomarker to track. Tell us more about the significance of potassium in the diabetes space.

Would that help people outside of the diabetes community too? Are there people who are just affected by hyperkalemia that could wear this?

Peter, did you want to add to that?

I think Jake said it very well. The combination of potassium and glucose can be a really fantastic product to help a lot of people.

The plan is to bring this to the FDA in 2027 for 510k clearance and so what are plans for rollout? I guess that's US. So when are you hoping to bring this to the US and then beyond the US kind of what does that look like?

Great. Well, Jake, Peter, thank you so much for coming on the show and giving us some insight on what's coming next. I'm very excited and I'll have to have you on the show again when the G8 is out.

It's a pleasure. Thanks Justin.

Now, what do you think of Dexcom's report and what they're doing about it? Let me know down in the show notes. Thank you Dexcom for coming on the show. If you're enjoying our reporting here at Diabetic, one way you can support us but also get a lot back in extras is by signing up for all access. With that you get exclusive access to our live event archive, member giveaways, and exclusive stories on industry analysis. You can sign up for that at diabetic.info/allaccess. We actually just had an event on which CGM is right for you where I covered all the CGMs on the market, my experience using them, how they all work, and we took questions live. That is on the live event archive page right now. Thank you all for listening. I'm Justin and I'll talk to you later.