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

Okay. Good afternoon, everyone. Thanks for joining us during our annual health care conference, and I hope everything is going well with this virtual format. This is Matt O'Brien from Piper's equity research team. I'll be moderating this fireside chat with the management team from DexCom. From the company are Steven Pacelli, who is the Executive Vice President of Strategy and Corporate Development; Jereme Sylvain, who's the Senior Vice President of Finance and the Chief Accounting Officer; as well as Sean Christensen, Director of Corporate Affairs. One quick housekeeping item. If you have questions for anyone on the panel, there is a box down at the bottom, where you can ask your questions or send them directly to me. I'm happy to ask them on your behalf. That's [email protected]. So with that out of the way here, let's go ahead and get started with the fireside. Guys, thanks so much for joining us.

Sure, absolutely.

Happy to be here.

All right. So obviously, strange times that we're in right now and hopefully getting close to a vaccine here. I would love just to hear, I don't want a Q4 update, but just as you were guiding looking forward, and Quentin has given us some commentary about things in Q4. What were you most worried about, least worried about, any gating factors from a technology perspective, a virtual perspective that you really had to keep an eye on? What were some of the puts and takes as you were thinking about the rest of the year?

Yes, I'll start, and then I can turn it over to Jereme, if you want, Jereme. Look, I mean, COVID was obviously one of the bigger factors that we looked at, particularly when we gave the updated guidance on the Q3 call. It was already a month into the quarter and we already started to see a resurgence. So not looking to give you guys an update here on any of the particulars around COVID. But I think where we sit today, we're happy with the guidance that we gave you guys a month ago. And then like every other quarter, right, we take into account the trends that we're seeing, both internationally and in the U.S. with respect to patient visits, with respect to just general sales trends you always have in Q4. And you've been around this business a long time, Matt. You typically see the seasonality in Q4. I think we're factoring in a little less seasonality these days because of the move to the pharmacy. You don't have that big deductible shift that you used to see back in the -- when we were kind of 100% DME where -- and even add Medicare to that, right? It's a subscription-based. It's DME but it's more of a subscription-based model. You don't have that need for people to kind of stock up as much as they can maybe in Q4 because they don't have that deductible impact in Q1. But I think we're sitting really happy with where we are in the business. We had heard some rumors coming out of the Q3 call and some post calls that we were trying to talk down our numbers from where we put you guys in Q3 and that I'm fairly confident that wasn't the case. So I don't know where that rumor mill came from. But we're pretty comfortable. I don't know, Jereme?

Yes. I mean I think that's absolutely. And as you think about COVID, I mean referring to strange times, the way we always thought about COVID was while there is general interest in remote medicine and certainly, we believe our product plays in incredibly well there, we are mindful of the fact that there are some endos and some PCPs that just simply don't have the telemedicine presence and then some individuals who simply don't have the resources for telemedicine presence, where if COVID was ultimately to flare-up, we're mindful of those as they pop up. And then as those job losses, ultimately, that started in Q2 play out and you get to your COBRA and you have some folks really fall out of COBRA and they go through their final 3-month cycle, that could impact us in Q4 as well. So those were some of the dynamics that we took into account. And as Steve mentioned, there's nothing that's come out thus far that would indicate that we would alter from where we provided our guidance and kind of -- we obviously sit behind our guidance at this point. But those were some of the factors and how we ultimately thought about it.

Okay. That's interesting. So as you think about COVID, something that we heard that was surprising from some clinicians and educators is really this push on. Hey, look, before we really wanted to just start on CGM, with COVID and you're at risk, you have to start on that. So is that a dynamic that you really saw earlier this year? Is it continuing? Is it to say another potential inflection in terms of adoption of CGM going forward?

I don't know that I'd call it an inflection. But certainly, I mean, I feel like we've navigated this year in light of the pandemic. I think we've navigated this year from a growth perspective, a revenue perspective pretty well. And I think a lot of that is driven by the fact that we were deemed to be an essential business, which is the first important thing back in kind of March time frame. And people with diabetes were unfortunately disproportionately affected by COVID. And I think there was a recognition that monitoring your blood sugar. If you were maybe on the fence about going on to CGM, whether it's us or a competitor, right, you were probably more motivated to go get on a CGM and really monitor yourself much more carefully than just checking a few fingersticks a day. And so I think does it fundamentally change the growth trajectory? No. I mean we've always said that we're going to replace fingersticks. Particularly anybody taking insulin, there's no reason next few years that where there's reimbursement, where people have it paid for, that they're going to be taking fingersticks. There'll be no reason to do that. And I think maybe this has accelerated that a little bit. But it's certainly, if nothing else, COVID has awakened people who may have been a little reluctant. Even the doctors, right, where the doctors may have not -- they may have, in the past, kind of subset which patients they would deem appropriate for CGM or recognizing that anybody with type 1 or even if you're on multiple daily injections as a type 2, you're going to be an at-risk patient coming into the -- and working your way through the COVID pandemic. And I think that's here to stay.

Okay. Okay. I appreciate that. So let's flip over to G7 for a bit here. I'm curious about the next steps that you need as far as what's important as far as getting that product approved. So I think you mentioned that you're now in the clinics, commencing trials for the approvals. And at some point, we'll see the G7 data, maybe an ADA -- or sorry, ATTD in May. Are you going to be getting an approval for the 10-day sensor based on a clinical study and then another clinical study will be needed for the 15-day sensor. How do we just think about what the process looks like from here? And then any kind of impact that COVID may have and to flare up the need to happen or is it largely vertical, we don't have to worry about that going on forward?

Yes, that's the right way to think about it that the trials we're running right now, and there's multiple trials ongoing. You've probably seen at least one on clinical trials, maybe more that are all part of the approval support process, which is the way we described it on the Q3 call. And yes, these are being -- these are trials that are being run out to 10 days specifically to seek approval in a near-term time frame than we thought we would do, making sure we're comfortable with the full 15 days of performance. We will have to run some additional patients. What I'm told in talking to the regulatory folks is that it isn't nearly the magnitude of -- like you're not starting over an entire multiple 100-patient study over the full spectrum for a 15-day approval. But because you've got to capture the data per the iCGM standard and particularly, you've got to capture the data through the full range of useful life and the days that we're labeled for, we will have to run patients -- some number of patients out to the 15-day time frame.

Okay. And I guess, Steve, just to be clear again, too, it was a little bit -- I think muddied on the Q3 call, I was a little bit confused. I think others were too. But I mean, basically, what you were saying on the call was that the sensor works well out to 15 days, but it's not well above the 70% level that you're seeing among one of your competitors. And you want to be well above the 70% level before you introduce a 15-day sensor. Is that the right way to think about it?

Yes. I don't -- I think we -- armchair or Monday morning quarterback, whatever you call it, right? I think there were some things or takeaways from where we didn't do maybe a great job on the call was I heard people saying, wait, only 70% of your sensors last a full 10 days or there was a lot of confusion that we unfortunately created for ourselves. But you're thinking about it the right way that a sensor, a labeled indication that only lasts 70% of the time is not good enough for DexCom. We pride ourselves on performance. We grew this business on our performance over the years. We expect well north of that. You shouldn't read into that, that our 15-day sensors don't last some significantly greater number of days or on a percentage basis, upwards of north -- certainly north of 70%. I think what we said, though, we weren't comfortable that we were quite ready to run a very expensive pivotal trial out to 15 days just yet. That we will follow it up quickly with an additional supplemental data for a 15-day sensor. I'm not going to give you a specific time frame because I don't have it for you yet. But it isn't the case that our sensors aren't performing. We just need to make sure that before we go run the trials and spend the money that we're going to get to something well north of 70%, right? Because otherwise, you're just creating a warranty problem for yourself. And then we're spending more money, frankly, fielding a phone call and sending a single replacement sensor, FedEx overnight or FedEx 2-day. That's from a cost perspective, you're way better off just to ship 3 sensors a month versus 2 sensors a month and make sure you're ready when you're ready to ship 2 sensors a month and not create some sort of warranty challenges or something for yourself. That, that's where we need to end up.

Got it. And if memory serves, Steve or Jereme or Sean, I think G6 was showing -- it was lasting out to 10 days around 90% of the time...

It was a little more than...

I mean I don't want to pigeonhole you.

And yes, look, I mean, I'm not going to give you the number, but at least Gen 6 is a frame of reference to tell you what would be acceptable for us, right, because we're going to launch that product. So that's -- okay, you're thinking about it the right way without giving you a specific target.

Yes. And I think that's what -- Matt, you're correct. The G6 as filed was a survivability rate through 10 days at 94%. And we're not going to comment on where G7 is. But you can imagine that we have a high hurdle for how long the survivability is, not just for the whole COGS profile but also if you think about the patient, even if you have to get a replacement sensor, that means a day or 2, you're going off therapy. So 70% of the time you're going off therapy waiting for a replacement. We just didn't think that was appropriate, especially as these sensors are ultimately running pumps and we just don't want to do that to our patients. So we figured send out free, make sure it works out to the 90-plus percent or whatever the number we ultimately get to is and make sure that these folks have an experience that is a high-quality experience as opposed to the alternative, which is the consistent breakdown of not meeting what we promised.

Got it. Okay. So I'm going to split out for a second and then come back to G7 in a second. Kevin, I think he did a webinar, some session recently. And he talked about the patient experience like you're mentioning there, Jereme, a little bit among different types of patients. So I'm just curious as far as what that exactly means in terms of sensor performance or features or whatever that may be? And is there a way to really differentiate your technology versus an existing competitor, others that are coming to market that we're not even thinking about by the different types of patients that are out there?

Yes, Jereme, do you want me to take it or?

No. Go ahead, Steve.

Yes. So no -- so what -- maybe kind of set this up is once G7 built out to scale, that on-body footprint is going to be the same across all of the patients, right? Whether it's in the hospital, whether it's pregnancy, whether it's type 2s who don't take insulin, type 2s who take insulin. Really what you'll see is, what Kevin's talking about, there's a lot of work we're doing on differentiating the user experience to those specific patient subpopulations, right? Because a type 2 who doesn't take insulin isn't -- doesn't have a need for a predictive alert that tells them they're going to be low in 20 minutes because they're probably not going to be low in 20 minutes. They're not taking insulin -- they're not having the same problems with hypoglycemia that you're going to find in someone taking insulin. We found different measures of reporting how frequently do they need an update? What sorts of alarms? Do they even want alarms in certain patient populations? Or do they need sort of more summary data? Do they benefit -- on the type 2 side, you benefit much more because if you can't take an insulin injection, you go high, right? If you can't take an insulin injection to turn your blood sugar excursion around, you got to do something. So do you need some encouragement to go maybe take a walk after a meal or something like that. It's going to be a much more tailored experience to the unique patient populations. And I think the differentiation with our competitor is that they've kind of adopted just a one-size-fits-all model, where they're just shoving the same thing down everybody's throats across the spectrum of patient populations. And I think we're going to take a much more targeted approach there.

Got it. And that kind of dovetails, Steven, to what I think most people are trying to figure out on this call is that you've got this other competitor and they've got next-generation come in maybe middle of next year if that time line actually works for them. But that's supposed to be an actual CGM. So everybody is like they're coming out with an axle CGM. You've got a CGM, not a flash monitor on the other side. How can you differentiate from whatever they're coming out with and stay ahead from a technology and feature perspective? Is it like I'm talking about this different user experience for different people? Or is there something even beyond that? You don't have to tell us what that is. Or is it just like -- it's just cost? Everybody...

No, no, no. It's -- I will tell you, it's a combination of things. First of all, it's performance, right? It's really hard because we're being asked about Libre 3 consistently, and yet nobody has a clue what Libre 3 is because it's not available anywhere. Libre 2 has not overwhelmed anybody with its improvements in performance, right, particularly in the hyperglycemia. So if they become a real CGM, and it's truly doesn't have to be scanned. If it's real-time, all the time transmitting either to a receiver or to a phone, but it's providing the patient with alerts that are not super accurate. We know how. We've seen that movie before, right, with Medtronic. So I think it somewhat remains to be seen. We need to get these things washed and get them out in the field. And now -- it's hard to address like the competitive nature of Libre 3 when you have no idea. But I do think it's a combination of maintaining our industry-leading performance, which we will absolutely do with G7, and then differentiating the different product segments like we just talked about. And I think that keeps us very much separate and distinct from that. It's not just about price. We can, particularly with G7 at scale and, Jereme, you can go in a little more detail if you want. G7 at scale, we can be competitive with anything that Abbott could throw at us from a cost perspective because we'll rip theirs apart, they'll rip ours apart, right? We all kind of know what it takes to build these things. At the end of the day, the component parts, you're all throwing away electronics every couple of weeks and then we have a general sense of what the cost dynamics look like there. We're really comfortable with where we're going to get to pretty quickly with G7. We're not concerned about it just being -- it's not just going to be a race to the bottom in terms of pricing. And even Abbott -- I mean Abbott, we're not seeing Abbott becoming more aggressive. They definitely stepped pricing down with both Libre 1 and Libre 2. But we're not seeing them continue to get more and more aggressive in the pricing mix.

Got it. Okay. Okay. That's good to hear. A question did just come across. I'm not pretty sure how to ask this, but as many investors are obviously concerned about the significant drop in the stock price regarding earnings and relatedly the partial disclosure the night before earnings throwing investors off balance. And I guess the question is, why did you do that? And why didn't you just do them together, I guess?

Let me take that one without throwing our soon to be retired Chief Commercial Officer under the bus. So Rick thought he was doing a good thing for us, and in advance of the quarter was giving us notice that this is -- he's been working -- talking to his wife. This is what he wanted to do. Unfortunately, the timing of his announcement to us and to the Board triggered an 8-K obligation. So we huddled up and we said, look, we have a deadline, we're filing an 8-K, which is fine. We made the decision because we had a very strong Q3. And I challenge anybody listening to the call to challenge our P&L at a $500 million revenue quarter and just ticked down the P&L. It was extraordinarily strong. I can't poke any holes into it personally, but okay, but we decided and Quentin was -- weighted on this and Kevin weighted on it. We decided to put the number out, because what we didn't want to do is have this weird 8-K come out the day before earnings. And then people think, oh, the Chief Commercial Officer is being pushed out or something, right? And I don't think the stock reaction that we saw the day of our earnings call was anything we expected. We expected just to get a really solid top line number out there, just so the market wouldn't be spooked the day of earnings, and then we'd get through the rest of the call. And if I were in the business of picking stocks and understanding why the stock price has trailed off, I'd be in a different business. But I -- it kind of is what it is, and we're going to continue to execute, is really all I'll tell you. We're going to continue to execute and show that this company is as strong as ever, and the stock price will take care of itself.

Got it. Okay. Fair enough. Just a few more here in the last 7 minutes or so that we have. But on the pricing dynamic, Steve, the 40% volume growth in Q3 was awesome. Obviously, it was offset pretty meaningfully by the move to the pharmacy. So is that $175 million headwind this year the right way to think about the top line headwind again next year? Or I guess, even if you don't want to go there, when does that massive headwinds start to slow down somewhat? Is it next year? Is it more so in '22? How are you thinking about that?

I can take this one. So it's a good question. And I think what we would say and what we're comfortable saying now, we'll give you an update at Investor Day. I mean...

What's that?

Can you hear me okay?

Yes, I can hear you okay. Did you get that question?

Yes, we got the question. So what we've said is 2021, we haven't given specific guidance. Should be similar to that in 2020. It's been our general guidance as we make that move. And really, the impact on -- in future years is really how fast do we move folks from the DME channel into the pharmacy channel. So a lot of these are self-inflicted. What we have said and we are comfortable saying is that migration, there's this, call it, great migration from DME to pharmacy and that migration is now in the middle innings. And so that kind of gives you some thought process as to how far we are along in terms of that journey. We plan on giving a bit more of an update on the Investor Day meeting where we'll talk a little bit more specifics about where we are. I think that's at least helpful for folks to understand. I think it was a little bit missed, and that's why we called out the volume growth is -- the 26% growth -- revenue growth relative to 40% volume growth is really a function of self-inflicted and intentional strategic moves of where we service our patients, which is in the pharmacy channel. And as you think about that, we were actually quite successful in moving folks. The comp plans we put in place are working. We're moving folks a little bit faster than we anticipated at the start of the year. It's not an indication that we're not adding patients in record numbers, and we talked about it. Q3 was a record new patient add quarter yet again. So those will dissipate over time. You will see that, that migration will happen over time. There's only so many patients who can move from here to here and then all new patients will start to come back to more unit economics.

And you're going to hear us that, particularly, I mean, starting with Investor Day -- well, we've already started, but the message was loud and clear that the term headwinds needs to leave our vocabulary, right? This is a function of channel mix shift that we are like, as Jereme just said, self-imposing, right? We are doing this very purposely to gain better access to our patient base. And so we're going to stop talking about this as such a negative. We've done everything we told you we were going to do in terms of shifting the base over at probably an accelerated rate.

Throughout the P&L, when you go through the pharmacy, isn't there an economic benefit to the company going that way? I think that gets missed, too?

Yes. Again, that's fair.

Yes. Over time, the net profit dollars in each channel is about the same. And so to your point, as you start to reduce friction and moving to the pharmacy channel reduced friction, the more patients you add, the more profitability you ultimately see. I think that's the other part that I think is maybe missed and hopefully, we can demonstrate it for you just by looking at the P&L, record gross margin post-launch of G6, despite record pricing headwinds and also profitability improvements of 400 bps year-over-year. I mean you're really seeing the profitability playing through the entirety of the P&L despite the investments in G7, despite the investments in DTC and a lot of the things that we're already doing. So I think you can see that the profitability profile continues to play out as we said so. But you're right, there is a lot of focus really on that top line, and we just probably need to be more purposeful about how we demonstrate why we're doing it and stop using it as the negative thing as Steve referenced.

Understood. That's really helpful. The other question just came through. Will G7 be 1 minute like presumably Libre 3 will be? Or will it still be 5-minute based?

I'm going to let Sean take that one because when we dig into Abbott, telling you that it's 1 minute is actually -- anyway, go ahead, Sean.

It's a good question. I think that -- I mean first of all, there's a lot of questions that are being posed between DexCom and Abbott. I just want to remind everyone that we don't view this as a zero-sum game. These are huge end markets that we're addressing right now. And so there is a great chance that both companies will thrive with our offerings. Specific to the 1 minute readings question, I would counter that and not ingest but to say what benefit does the 1 minute reading offer to a patient. With Libre 2, they advertise a 1-minute reading, for instance. But when you actually wear a sensor, you only get a reading. You get a data point every 15 minutes. And that plus however many times you scan. So you end up with 1/3 the points automatically of G6. And then if you want to get up to the 288 readings, you're going to need to scan over 190 times a day. So it's a little bit of smoke and mirrors. Certainly, I think it has a benefit for their alarm functionality, but G6 has predictive alarms of 20 minutes. So -- again, I would question the utility of it and start it there. It starts with the patient and what the product does to actually serve a purpose to help people manage their diabetes, not just a marketing claim.

Got it. Okay. And just a couple of minutes left here. So I'll -- just one more question. On the non-intensively managed type 2 side, Steve, Jereme or Sean, whoever wants to take this. What's the next data point we're really looking for? Is it the Intermountain data sometime early next year? And then I guess, how are you feeling -- I'm sure you'll talk about this more at Investor Day next week, which is how are you feeling about that category, Sean -- I mean, Steve? We go back to when you guys are trying to get reimbursement for type 1 patients. Is it a similar type situation where you got to have coverage to really access that market? Or is there some other cash pay component because there's so many patients?

Yes. So that's part of it. The market is so big in what we've said, we're going to take multiple shots on goal here, right? So you've seen the work we're doing with UnitedHealthcare on the Level2 program. You've seen working with folks like Livongo and Onduo and running studies with Intermountain. Today, there is not meaningful reimbursement on a kind of fee-for-service basis for that patient population. So it isn't -- there's ways that Abbott has subsidized some of those patients in the drug store. But by and large, there's not coverage policies at any of the major payers specific to the non-insulin-using patient population. I think over time, you'll see that evolve. You mentioned cash pay, it's something we've looked at. I mean patients are used to having their devices and their products paid for. So that's probably a little tougher slog. But as it evolves, and it's going to take some time, but I think you're going to see us doing things with partners, like we talked about. You're going to see us doing things, some direct-to-consumer work there as well. But the market is so big there that such a tremendous opportunity over that broad space of type 2s that it's -- if you would have asked me this 3 or 4 years ago, I would have been maybe a little more skeptical that we'd get meaningful penetration. And with the market research that we've done and the uptick that we've seen to date, either directly or with our partners, I have no question that, that non-intensive market is going to be enormous for us -- for the category, not just us, but for everybody.

Got it. Okay. I think, unfortunately, we're out of time. I only got through about half of my questions, but I'll go ahead and call it there. Steve, Jereme and Sean, thanks so much for all the feedback. Really appreciate it.

Thank you, Matt.

Thanks, Matt.

Thanks, guys. Thanks, everyone.