DexCom, Inc. (DXCM) Earnings Call Transcript & Summary

Welcome to the Wells Fargo Healthcare Conference. Before we get started, if you are a member of the press or media, please disconnect at this time. This is a restricted line. Any unauthorized party in this meeting or any unauthorized use of the information communicated in this meeting is subject to prosecution to the fullest extent of the law. Any unauthorized person, including the media that is on the line at this time, please disconnect. Please note, today's call is being recorded.

Good afternoon, everyone. I am Larry Biegelsen, the medical device analyst at Wells Fargo. It is my pleasure to host this session with the management team from DexCom. With us we have Jereme Sylvain, Executive Vice President and CFO; and Brice Bobzien. Hope I pronounced your name correctly, Brice.

Close enough.

Good. Vice President of Finance. This is the last fireside chat of the 2021 Wells Fargo Healthcare Conference and definitely the last, but definitely not the least. So in terms of format, again, it's fireside chat. Jereme and Brice, I really appreciate you being here.

Absolutely. Thanks for having us and look forward to chatting with everybody. Happy to kickstart everybody into a wonderful weekend.

All right. Sounds good. So Jereme, just -- we'll get the COVID question out of the way. You guys had a really nice Q2 with new patient starts. The infection rates have obviously increased. But when we look at the IQVIA data, it looks pretty steady, pretty consistent for you guys. So any color you can share with us on the impact from COVID?

Sure. I can give you a little bit of kind of what we've seen. So historically, we've done pretty well through COVID impacts, and this really goes back to even last year through telemedicine and other forms of getting product to folks. And so we've navigated through COVID pretty well thus far this year. We were mindful of the impact of the Delta variant when it was rising when we gave our guidance for the balance of the year. And so that was already contemplated in our guidance is some of the impacts associated with the Delta variant. So things have really navigated at least how we expected the Delta variant to spread in line with our expectations. And so I don't think there was anything that we've seen that's changed the way we thought about it. The big challenge we have -- I mean, this isn't just Delta, this is others, as we get in front of more and more primary care physicians, sometimes those primary care physicians aren't willing to see both. So this might be a new primary care physician for us to go see. We saw a little of that in Q2. Certainly, nothing has changed from that viewpoint as we move into Q3. And remember, as we try to get samples and expand our population that we call on, that can cause a little bit of a wrinkle. That's been in pockets, and it's really been PCP-specific. At the end of the day, we've still been able to get in front of folks and still been able to navigate through it. So that's been the impact. We anticipated it. It was contemplated in our guidance. It's trending the way we would have expected it. And our business has been generally pretty resilient to the impact of COVID.

Okay. That's super helpful, and that's kind of consistent with what we see in the audit, the third-party data I mentioned. Jereme, I wanted to kind of shift gears, focus on 2 important areas: one, G7; the other, type 2 non-intensive. A lot of news flow in those 2 areas in the last few months. So starting with G7, I think, you finished the G7 U.S. clinical trial in the second quarter. You're preparing the regulatory filing and submission for CE Mark. And I think you expect to launch somewhere around the world by the end of this year. Is there any updates on the filing or the launch timing?

No, there's no updates. And I think we articulated on the last call, we have made good progress. And the data that's come off of it has been good. At ATTD, we released a little bit of data, preliminary data just to kind of give folks into a glimpse of the performance of the sensor, and it's a sub-9% MARD, which nobody else has come closer to that in terms of accuracy. And that's on top of a real-world accuracy that we've ultimately seen in our sensor over time. And so I think where we sit today, just to kind of reiterate where we are, you're right. We finished up our U.S. pivotal. We filed for CE submission already. That goes through its normal process. The early data that we saw that we have shared publicly has been incredibly strong. And so we're very bullish on the output associated with that. So we feel like we're in a good spot. I will give you guys a little bit more of an update on timing and filing as we get to our next quarterly kind of check-in with our earnings in our next period. But everything is firing on all cylinders at this point. And so we're very confident that what we told you we will do, we will do.

So Jereme, just to be clear, you filed it for CE Mark?

Correct. We filed it for CE Mark already.

And you've had -- but you haven't disclosed the U.S. filing yet?

Correct. We've completed the clinical trials in the U.S. We are now going through the data preparation for a submission in the U.S. And so that will take a little bit of time as we navigate through that data. But that's the progress where we are today. And so we continue to march down all those kind of data points we gave, all of our investor community, how do we make sure we're showing you we're making progress. We continue to hit all those in successive patterns.

When are we going to see the U.S. pivotal data?

Well, we will also generally share that once it goes through the ultimate submission process and the approval and that's when it comes out. As data becomes more available, I think we'll look at ways to potentially provide in advance if it makes sense to do so. But generally, what you do see is the data that comes out of our pivotals. It usually comes with the regulatory filing, not the submission but the actual filing itself. So that's the backstop, and if it makes sense to provide it ahead of time, we'll reconsider it based on the timing when we're in front of either our constituents or investors. But I think that's where we'll ultimately see it, at least as a backstop, is ultimately upon approval.

Jereme, you talked about the OUS data having a MARD sub-9% a minute ago, you mentioned that. Do you think -- are there reasons why that will be difficult to replicate in the U.S. pivotal trial? Obviously, that would be a pretty good outcome.

I don't think we have any sort of belief that something in the U.S. would be any lesser of a product than we would launch OUS. So I think the expectation you should see is the product that launches outside the U.S., the one in the U.S. should be at and/or better than what you see outside the U.S. And the reason is that as time moves on, everybody knows that we made changes to the algorithm. It usually goes on behind the scenes, tweaks and things along those lines to always look to make it better. And so we're never going to launch a product that's worse. And so I think what you could say is the floor is what you know today. And then over time, we'll always look to reiterate to make that data better.

And so the floor is the G7 OUS data that we saw at ATTD. Is that what you're saying?

That's correct.

Okay. And on manufacturing, you guys have talked about getting up to capacity. I think it's 200 million sensors a year, correct me if I'm wrong. But the question really is, where are you on the manufacturing scale-up? And once you have approval, OUS or U.S., is it going to be a rapid launch? Obviously, if we think about Abbott and Libre 3 had approval in September of last year, they're gradually rolling that product outside the U.S. How do you think about the launch of G7?

Yes. So we will have an eventual capacity of 200 million sensors, and that's when Malaysia is online. But that's not going to be really until in the latter half of next year. And so what we're looking at now in terms of the launch of G7 and how we navigate through it is making sure we maximize both our Arizona and San Diego locations. We can't simply turn off G6 and turn on G7. There's regulatory, there's approvals country by country. And then remember, we're integrated with G6 with a lot of pumps. Certainly, the Tandem pumps and as OmniPod comes out will be integrated with G6 for their devices. And so we're navigating through that transition over time as to whether it's G7, G6, who needs G7, who needs G6, which are on contracts. And that's both in the U.S. and outside the U.S. So I think the expectation is we have the machinery. A lot of it is in. And so I don't think it's an issue of capacity of G6 or G7. It's just which lines do you turn on with the existing space that you have while you're ramping up your Malaysia factory. Once Malaysia is up and running, I think we have a lot more flexibility to have more locations. But up until that point, it's going to be a little bit of a jigsaw puzzle. We're not concerned about capacity. That's not the case. It's just which capacity you turn on and when. And our operations team and commercial teams are working together to make sure they turn it on at the right time for the right...

Why don't we bring up the first polling question? And I have a couple more on G7. The first one is just on launch timing, see what expectations are. Who do we have here? [ Luke ] from OpenExchange. Are you ready? And so I'll read it. This is simple. When do you expect G7 approval in the United States? A, 2021; B, first half '22; C, second half '22; D, it should be D, not sure, the FDA has been very slow. So it should be A, B, C, D. And while there -- Jereme, there's a bit of a lag, so while people are doing that, you guys have talked about pricing G7 at parity with G6. Does that include the sensor only? Or is that including the transmitter?

So the current expectation is the all-in. And we -- most of our contracts are shifting to more subscription base. And so when you have a subscription base, what you see in our financials, in many ways, is an allocation revenue. It's not actually revenue that you get reimbursed directly forward. It's behind the scenes where allocated. And so in all of those subscription contracts, we expect it to drive in. In terms of how we do it in those non-subscription contracts, which are a much, much smaller portion of the population, again, the expectation is as we negotiate those as a per month therapy cost, and so we'll work on that. We are working through it as we speak, but the current expectation is we do so on an all-in cost. Whether those negotiations change over time, we're working on that. But again, I think the thought process has always been we would swap those in at a similar per month cost, which, again, tends to be easiest in subscription. A little more challenging in part and parcel, but mind you, we do negotiate part and parcel based on monthly therapy costs. So that's how we generally talk to payers.

And how far behind is the 14- or 15-day wear, G7?

We are working on it as we speak. I think the big question is survivability. And so kind of targeting back to what we are, obviously, we're coming up with a 10-day G7 product, which is on parity with our G6. And we don't have a lot of clamoring from our customer base, our patient base, customer base that they want to ultimately see a 15-day product. They're happy with a 10-day product. For those that are wearing it, it's a little bit like a bandage, right, all wrapped for 10 days, 15 days. Sometimes it's nice to replace them. The 15-day is really something we believe is important for us to ultimately lower our cost to serve as we go to more markets. And so it's really important for us over time, we believe, from a cost structure to come up with a 15-day. And in terms of the study, we are working on it now. When I say study, the internal work around studying what we would need to ultimately do it. It's really about survivability and making sure we get survivability out to the time frame we would expect for someone on our sensor. We don't want a 70% survivability rate, which means 3 out of every 10 sensors aren't making it to that date, and we might have somebody on an integrated delivery device that cannot actually use that device while they wait for their next sensor. So we're working on getting that up, whether it's the patch and some algorithm work, we'll do that. We'll have a little bit more of an update for you in terms of timing as we get closer to it. It would require a new clinical trial. So it will be something that you'll likely see coming across ClinicalTrials.gov, but it's not too, too far behind. It's not the far distant future. It's in the short to medium term.

That's helpful. And last one on G7, once G7 is launched in the U.S., do you plan on keeping G6 on the market? And if so, kind of why do you need G6?

Yes. We will for some time because, one, it takes time to transition people. But two, I referenced it a little bit early. You've got to integrate these sensors with your pump partners. And it takes a little bit of time for pump partners to integrate with those sensors. And so while they are working on G7, we know they will, they are currently integrated with G6. And so there will be a time frame when G7 is maybe for folks that are not on an AID and G6 is as they work to integrate G7 into their pump hour projection. We don't expect it to be a terribly long time, but we know that's absolutely a reason that needs to stay out there. The other piece is just going to be working through contracts. And while we expect it to be pretty quick, sometimes it takes a little bit longer to kind of navigate through those various payer contracts to make sure that we get G7 on the formulary. Think about it, a lot of people have an annual health care cycle. And some folks maybe don't change it until that annual health care cycle. So there could be some time as we go through that contracting where it takes a little bit longer. Most of the larger payers and plans ultimately will do a midyear, or they'll change, or they'll do an ad hoc for larger products such as this. But in the event that we're unable to do so, that's why G6 could stick around for a while. Just like G5 stuck around for a little while even after we launched G6. There was some time that took to ultimately shift it. But we are heavily incentivized to move folks to G7. We just believe it's a better experience. It helps us build, expand our capacity and therefore, over time, reduce cost. So we are heavily incentivized to move it, but we'll do it in the right pace to make sure nobody's left out.

All right. Let's see the poll results. First half 2022, interesting. Jereme, any reaction?

I'm not going to provide any reaction to timing, but thank you for trying. It was very good effort.

All right. We'll move on. Type 2 non-intensive. Obviously, a lot of excitement coming out of ATTD with the MOBILE study results. What's next for the MOBILE study and the type 2 non-intensive population? When do you expect payers, including Medicare, to start covering that population? Is it realistic to think we might see something in 2022?

Yes. So we're going to use the results from that study. And we think it's an incredible study, right? Not only is it diverse in population, it was a randomized control study, so clearly, highly well performed and well architectured. Obviously, in a large publication, such as JAMA, we'd love to have it [ centered ] in respected publication accordingly. So across the board, we're really, really excited about the MOBILE study. In terms of the reaction, as we've gotten in front of the community, the diabetes community, people have been very excited about this study because it's been long known that folks that are impacted say in the basal-only population do have complications associated with that disease, and having a CGM can help. Everybody's known that, but it's going to be really a battle to make sure that we have the data to do so. So this starts the conversation. And we found that a lot of the folks in the community, whether it's the physicians, the patients, et cetera, have been very supportive of the results. In terms of the payers, I think this is really us having to now get in front of them. And we've been in front of them time and time again, but it's getting in front of them with the data, showing them the benefit, showing them the reduction in A1c and articulating what that means for quality of life and cost of care. And going through that process with them, it's going to take time. Obviously, it took us a while sitting in front of CMS to ultimately get them over the hurdle. And I think they've been incredibly pleased with what they've seen for folks on CGM and Medicare space. Certainly, it's become a bit of a standard of care in type 1, and as you're moving to type 2 intensive that it's starting to move there as well. And so we think this is where it's going to go. The big question, which is timing-wise and when does Medicare and payers cover it, that's a little bit too hard to peg right now. Remember, the MOBILE study just came out a couple of months ago. And so as we sit in front of folks, we will be sitting in front of them time and time again. We don't have necessarily a specific date as to when we think it's ultimately going to be out there. However, we do know that it is incredible. And so we'll keep pushing on it. And as time moves on, maybe we have a little bit more information, we'll be able to share. But right now, great feedback, great output, incredible trial. It represents the population in which we serve. And I think it's great evidence as we sit in front of payers in Medicare to really demonstrate why this is an absolute no-brainer for them over the long term.

Jereme, at the analyst -- last analyst meeting, you talked about type 2 non-intensive contributing about $700 million in revenues, I think, by 2025. How should we think about the cadence?

Yes. So we do think about it as a subscription business or a recurring revenue business. It does build over time. And while I haven't necessarily given out year-by-year guidance or year-by-year expectations, as with most subscription models, you build over years and you build and you build off of those as awareness grows. And so if you were thinking about it from that perspective, that's how I would think about it without getting into specifics year-by-year. But as we -- you can see the population work we're doing today with Level2 and some of our partners, and those programs continue to grow. Certainly, this MOBILE data study and moving into this basal-only population, we believe, over time, there is coverage for this group. And then looking at even other use cases, which are out there today, whether it's in health and wellness in the diabetes space, programmatic approaches like Intermountain Healthcare and thinking about these provider networks. We expect those to rise over time, but it is a subscription model. So the way I would think about it is it does grow on itself year-over-year.

Jereme, you guys recently received FDA clearance for the G6 non-intensive glucose program. I don't even think you put out a press release on it, but I thought it had -- it was significant for a couple of reasons. What -- it seems like this could be part of your market segmentation strategy, where non-intensive patients use a program with less bells that were -- a sensor with less bells and whistles compared to insulin-intensive patients. How significant was that clearance in your view?

So we think -- as you think about this, there's been a couple of clearances. There's -- I think you're referring to Abbott app, there's also real-time API. And I think through both of those, it really kind of cements the ecosystem we've talked around, which is software really enabling folks to access CGM. The Abbott app I think is important. If you think about a lot of the start-up companies out there and they're looking to try to set up your health and wellness, or you look at the big companies like UnitedHealthcare who wants to have an Abbott app that they are branded for their Level2 program, you're able to access this glucose data in an FDA-compliant manner. And we're able to embed it ultimately in the app that we provide. And so really what it does is it allows us to be a CGM of choice to plug in to all of these different opportunities in the type 2 space that we don't necessarily play in today, where it's almost like a plug-in play. It's a packaged FDA approval that goes right into your app. And so it allows folks to really easily integrate. And we talked about integration being difficult. I mean, it's taken a long time for us to integrate with Tandem, Insulet. It takes a while to do these custom FDA-approved applications. What we thought was really unique about both this and the real-time API is now you have -- you can have essentially a tile inside of an app. So if you want to offer a 1 app solution and you're a programmatic provider, you can offer our CGM app within your app. So you don't have to jump through multiple apps, which I think is a huge competitive advantage over the long haul for us being a partner of choice. It's also why the real-time API was important. Let's say you don't want the Abbott app. You want the data off of the CGM to feed into another app. You can use our real-time API to ultimately get that data real time. So I think both of those really support what we've talked about, which is software as an ecosystem being a real differentiator for our platform and using that cloud infrastructure to ultimately push this data out.

That helps. And a couple of -- everyone is trying to figure out the business model, revenue per patient for the type 2 non-insulin-intensive population. Are there any metrics you can share now that would help, for example, sensor utilization today, what you're getting paid for today? You've got to have some sense at this point.

Yes. So while we won't get into what we get paid for it, I think we would be willing to say is a sensor utilization in these programs is very, very high to the point where those in the non-intensive type 2 space are using it just as much as the type 1 population. Now as you think about the adoption curve, the people that are using it today are those early adopters, right? Folks that are tech-savvy that really want to use it over time, which is why we don't have a perfect example as to what the world looks over the long term. So maybe I'll just give you a little context as to what we do. Through all the programs, certainly, we see these folks that are taking it earlier, really want to use it. Their sensor utilization is high, and they don't want to give it up when their program is done. We also know at Level2, to make Level2 work, you have to be on a CGM full time for a certain period of time. That's how they ultimately make the determinations as to your -- what your therapy regimen. And so we know early on that a lot of folks are using it at a very high utilization. I think the bigger question is, as you kind of go up the adoption curve, are there certain models where you have more intermittent use, more diagnostic use as opposed to other sorts of -- and that's what the business models we haven't quite figured out yet, Larry. But I think early on, there's a couple of data points that give us a lot of confidence. One, the people who are using it, love it, don't want to give it up; and two, the market research data that we do shows that the interest in the type 2 population to use CGM is at the same level as it was in the type 1 population in 2018. And you saw what happened in the type 1 population in 2019 and 2020. A lot of adoption. So we know that the interest is there. The unmet need is there and the early results are there. So across those, that's what gives us the confidence that the space is willing to go here or ready to go here.

Have you commercialized the API and the app clearance?

I'm sorry, what -- the API and the app?

Commercialize the non-intensive app clearance and the API clearance.

So today, the commercialization is via sensors, right? They only work if you buy a sensor. So ultimately, whether we're working with a partner or it's really driving them to go to their payer and ultimately get approval for it, it's commercialized through sensor sales. Over time, could we charge with the real-time API connection or royalty fee? We absolutely could. Could we charge for app royalty fees? We absolutely can. Because we're early days and we're in adoption phase, I mean, today, really, it's through sensor pull-through. And that sensor pull-through, we think, is pretty darn sticky when you have these apps that really work based on the data that we're able...

Okay. Yes. I wanted to move on. Jereme, we've got about 6 minutes left. Just maybe looking ahead, second half of this year, 2022. Maybe I'll even skip to the 2022 question. Just high level, puts and takes for next year we should consider?

Yes. Without getting into too much detail in 2022, I think what you're seeing in 2021 is continued momentum. And I don't think we see that changing in the future. And I think what you'll also see, as we've talked about this before, is we'll still have channel shift headwinds in 2022, and we'll be continuing to go to new markets in 2022. We know that as we expand internationally, we talked about that. We try to go to 1 to 3 new countries every year. So it's a little bit more of the same. It's a lot of this land grab. And as we look at more on the type 2 space, we'll be continuing to push there as well. I don't want to get into too much in 2022. We're talking about that really as we round out the year. But there's nothing that prevents us from being very confident in the momentum we're currently seeing in the business and seeing that continue into 2022 and beyond.

The channel shift headwind, how should we think about that relative to 2021? I mean, there's a channel shift headwind, and then there was a $50 million OUS pricing adjustment that you talked about. I couldn't -- they're in total, I don't know, Jereme, $175 million, $200 million this year. What -- how should we think about that for next year? Consistent or similar minus the $50 million headwind?

Yes. So the number this year is $200 million for the channel mix and $50 million for the OUS, so about $250 million this year. Next year, we see the $200 million being flat in terms of dollars, not as a percentage. And so that's the continued migration of the DME to the pharmacy. And then at the end of next year, we see that falling off quite a bit and falling back into your normal pricing because we expected that when we get into that 75-25 pharmacy and DME split. In terms of the OUS business and the impact of that $50 million, there'll be a little bit less than $50 million next year. We're going to lap that $50 million you see in the back half in the first half of next year, and then you kind of get those contracts annualized themselves. So all in all, the pricing headwinds in terms of total dollars or channel mix, I should say, in terms of total dollars next year, should be right around the same. And then I think as we get through next year, I think what you're going to find is that number goes down.

So just to be clear, it's $250 million total this year, and it's about $250 million next year. Is that what I heard?

Yes. We expect about a similar ballpark next year before it starts to fall.

And the $50 million OUS, is that just because of the -- is that incremental $50 million next year? Or is it just the lapping because it's a partial year?

It's the lapping. If you think about the $50 million we talked about this year, it was Q3 and Q4. Those contracts obviously then will come on in Q1 and Q2 before they annualize. So it's predominantly there. Now we're going to make all that up with volume because of that determination. So while those are out there, the volume will absolutely make that up in the U.S. business. But that's absolutely -- it's not incremental. It's the annualization.

Okay. That helps. I mean, just curious on the P&L, maybe margins because I looked at consensus for next year, 22% revenue growth, 40% EPS growth. You are launching G7. The launch year, maybe there'll be less leverage. I don't know if you're willing to kind of comment on that.

Yes. It's not something that I think we really want to talk about in terms of setting 2022 expectations at this point. We are very bullish about next year, certainly the G7 launch. There will be a launch obviously at the back half of this year and further geographies next year. We expect that to be a catalyst over time. But as you get down to the P&L, you hit on a few things that I think we're going to have to give you more details as we get closer to setting guidance for next year, which is well, how does that impact margin? When does -- when do you get to those volumes where you get those reduced costs and you start to see the benefit of automation? When do you annualize some of the investment in the sales force and those heads in the DTC? I think those are all apt questions. I think it's probably a little bit early to start giving 2022 guidance. But we are -- we will be thinking about all those things. Those are all factors everybody should consider. We'll have a little more detail for you as we get closer to the end of the year and into early next year.

How are you thinking about international now with the pricing adjustment in terms of international being a bigger part of your business?

That's the expectation is over time, international becomes a larger percentage of our population. We are underpenetrated outside the U.S. and it is underpenetrated as a total market relative to the U.S. And so when we talked about it when we were at our Investor Day, I guess, it was in December of last year, we talked about 1 to 3 new countries every year going into. And that's the expectation. As we go into 1 to 3 new countries in a meaningful way every year, that forms the baseline for expansion over time. And so the expectation is, over the long haul, as you split the revenue, U.S. versus OUS, today, we're more 75-25. Now over time, we know that, that 25% internationally has to go up. And so we'll be working on that. So to your point, is the international growth, do we expect it to be -- do we expect it to grow faster over the long haul? We do because of the unpenetrated markets there. And so I would expect that to grow significantly over time. It doesn't mean we don't think the U.S. has incredible runway, because it does. And there's a lot of work to be done in the U.S., a lot of PCPs we don't call on. But as you start to think longer term, you're absolutely right. International will need to be a bigger percentage of our business, and we're investing in our sales force accordingly.

All right. Well, we are out of time. Jereme, I really -- that was a good conversation. I really appreciate you being here. Sean, who's on the line, too, thank you. Brice, thanks. So with that, we will end. Thanks so much, Jereme.

Thanks, Larry. Thanks for having us. Thanks, everybody, for taking time to listen to us today. Have a wonderful weekend, and it's good to see you, Larry.

Thank you.

Thanks a lot, team. Appreciate it.