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

Given her a discount.

She's on our insurance. So I think the answer is, we expect over time for this to effectively replace the oral glucose test. It just -- it makes sense. The effectiveness of that test is very hard to see, and seeing someone's glucose over an extended period of time really helps you diagnose it. So I think if you can replace the diagnostic plus I think for second and third trimesters for folks that are either at risk or ultimately who have gestational diabetes or come into it with a form of diabetes, I think that's more of a full-time wear. And so then the question becomes what's that reimbursement category? How do you do it? And I think that's going to depend on how customized we need to make it and how that return needs to work. I think if you start with the fact that there's an oral glucose test, the amount of folks on there plus the where, you can make an economic case that pricing at more of a T1 could make sense. However, if we were to deliver customized feedback and a customized app for it and how that ultimately delivers changes in care. There is an opportunity there to think about the acuity and the higher the acuity -- and this is clearly one of the more acute cases, could be higher. It's a conversation we're going to have to have, though.

Yes. Well, I mean, I think it was even a study out of Australia recently showed $12,000 per pregnancy in cost savings and using a sensor. Is it as simple as saying, "hey, we provide the sensor, we can provide the analytics around that, we should get 20% of that," or something like that, not 50%, maybe 40%...

Yes. So the market access team that reports to Jake will be using all of that information to certainly be negotiating that. There's no question.

Okay. And then maybe from a hospital perspective, obviously, we've seen a lot of studies published with some good results on hospitalization or hospitalized patients using a CGM during COVID, especially. You've got the display issue, the nurses station, how you're going to track all this. I mean, obviously, that's an issue that has to be solved as well. But just thinking about the path to FDA approval outside of a data display issue, what is that process? I think you've said before that it might be a number of small studies, not necessarily one big pivotal. So what gets you over that hump of getting hospital approval?

I think it's about generating the evidence that shows the product works well and is reliable in the hospital. And so that's what a number of these small studies are doing. We've got a number of different studies running, some that could be used for evidence for approval. And so it's really taking that subcutaneous sensor and making sure it works properly in the hospital environment. The need in the hospital for glucose management is incredible. If you think about it, having to go into these hospital rooms to give a fingerstick or a lab or blood, CGM is just a much better way to manage glucose in the hospital. And so there -- the hospital staff really needs better tools. That's one of the reasons why antiglycemic control is so challenging in the hospitals because the tools they have to measure it. So CGM just makes sense.

And I would assume you guys would be more than willing to go down the path of an adjunctive approval first as a diagnostic before necessarily trying to get a non-injunctive?

If it's required, we would. I mean, I think at the end of the day, we would -- we'll have to work with the FDA in terms of approval. But if the answer is, is the faster path is through the adjunctive while you work on a non-adjunctive, absolutely willing to work with the hospitals as well.

Yes. I mean, to me, the non-adjunctive pathway could be years if you have to test like literally 1,500 different drugs...

The interferons.

Right?

The interferons is -- in fact, when Jake says their study is going to on, it's mostly around interference. I mean that's really what we're working on is what the other drugs are that ultimately could cause it. So to your point, rather than having the nurse come in hour on the hour for a fingerstick, if it's once a day to calibrate, ones at the start, whatever it ultimately is, as adjunctive to make sure they're comfortable -- there's comfort there. We would absolutely go down that path. But it's that part of the work that's going on is looking at that.

Okay. And when we think about -- I think there's 33 million hospitalizations a year. Obviously, there's a preponderance of those patients who -- 1/4 of them get admitted to an ICU, a preponderance of them have diabetes. They over-index diabetics, hospitalized, things like that. So I mean, you can build some big TAMs. I mean, is it crazy to think that hospital could be like $1 billion U.S. TAM over the next 5 years? Or is that too big?

It's not the TAM, not crazy at all. In fact, I think it's bigger than that. When you think about how many folks and again, we talk about reimbursement and acuity. That is a higher acuity case and therefore, a higher reimbursement per sensor. And that would be our expectation going into it. And so the TAM can be incredibly large. How fast that TAM is addressed, I think is going to be a regulatory pathway and adoption pathway. But I have no doubt that the TAM is even bigger than that.

Yes. Okay. Okay. And then we've talked about this premium sensor idea a couple of different times now today. Obviously, Dexcom ONE helping you in the tenders in Europe and in some other markets. It seems like -- well, one to confirm, Dexcom ONE will eventually move to the G7 wearable kind of as soon as possible. Is that right once you get the manufacturing capacity up in place?

Yes, that's right. G7 is a brand-new platform for us. And so we're launching it with the G Series first, but we want to move all of our products over to that hardware platform for the scalability of it and the cost outstanding across the wearable.

And so we should -- I mean, what's interesting to me is it's going to make the modeling a heck of a lot harder, but it's almost like we're going to have a 3-tiered G7 platform, a high, medium and lower premium and discount with the standard product as well.

Yes. I would expect -- so I think the way you're thinking about it is right, is everything moves to the G7 platform. And so very easy to use for what we think is all use cases. And then back to your point, you've got the premium segment, AID, those folks that are hypoglycemic unaware, et cetera. You've got this more, "hey, look, this is what I mean, but I don't need all the bells and whistles," so call it value or standard. There's also then, of course, use cases in hospital or otherwise, and whether that is a specific product. It would still be on a common form factor. And then there could be an opportunity where you have a product that's designed at folks who are not in need of hypoglycemic concerns, right, where it's about hyperglycemic management over time. So there could be multiple different product options using software, but leveraging that hardware platform. So the way you're thinking about it is absolutely the way we are.

Yes. And it sounds like the last one you're throwing in there is maybe like a health and wellness kind of product.

It's those folks that aren't taking insulin where hypo is not at risk. So whether it's health and wellness or it's prediabetes or diabetes more on the lighter side, that's certainly an area of focus.

Is there any way to think about revenue streams outside of selling those sensors into all those end channels? Is there a data analytics? Is there something else you could partner with pharma companies and start looking at actual helping with pivotal trials on a CRO-type design or something?

So I'll start. Jake certainly has all the data in his shop. And I don't dare ask you how many glucose points we have. It's in the billions or trillions at this point. There's an opportunity with data to either use it to either help partners kind of optimize their offerings, help certainly us. Whether that we monetize that through the form of data or form of partnerships and ensuring that those partnerships then drive our sensors through it, that's what we continuously look at. Whether that is, again, as you think about the software back-end connectivity plus all the data we have associated with it, it just puts us in an incredible partnership position as we even move beyond just typical sensors. Now how we monetize it? Could we do it through incremental sensor sales, but we are a partner of choice because of all of this capability? Sure. Could we do it as a partnership with a pharma company to say, "Hey, look, this plus this plus the data means a more beneficial package." These are all the conversations we have with our partners. I don't know if you have anything else to add to that?

I'd just say that we made significant investments in our digital platforms with the concept of an ecosystem of partners and devices that use our sensor data and amplify the value of it. And so as we look to the future, there's a lot of opportunities. I mean, there's folks running clinical stays right now connect up to our sensors using them for weight loss, right, helping people lose 10, 15 pounds just using a CGM in their mobile application. And so there's tons of opportunity for partnership there. We're just going to continue to be the best partner, the best ecosystem there is in CGM.

Okay. That's helpful. And then as I -- taken that point, Jake, on just weight loss and some of these lower acuity use needs and what have you. When I think about you guys, and when I sit there and I start my price target and think about the big revenue multiple that I've got to use to get it and all that, I'm usually pretty darn bullish when it comes to hospital, when it comes to pregnancy, when it comes to all these other indications. Where I struggle a little bit is on the health and wellness side just because there are plenty of companies that are in very early stage but for noninvasive technologies. And we know probably a MARD isn't going to be the driving factor there if you're close and can just change lifestyle. So one, Jake, maybe what are you seeing on the noninvasive side? Anything that actually interests you, not that I'd expect you to say it specifically, but anything out there. But Jereme, why not take -- and maybe Steve is doing this for all I know, take a portfolio approach buy up 5, 6, 7 of these companies at $30 million, $50 million each or so, I have no idea. But then one of them could hit big and be a watch-based sensor for health and wellness just to cover the noninvasive side.

You want to start with that?

Yes, I'll start. So the concept of noninvasive sensing is fantastic, right? But the reality is when you get into the science of trying to sense glucose molecules from outside the body, we haven't seen anything that comes even close to the amount of reliability you would need even for health and wellness because if you think about it, glucose for someone who doesn't have diabetes doesn't change all that much. And so you actually need fairly accurate data to be able to make recommendations and understand what's going on. And so we remain very close to that research environment for noninvasive. As you mentioned, Steve, with the venture group is we keep a close eye on everything that's going on there. We really haven't seen anything that we'd be interested in yet, but that doesn't mean something couldn't come around.

And I think to your point, you're absolutely right. Would we place some bets if we saw some novel technology that we thought could eventually migrate into something there, whether it's through an acquisition or a minority investment that allows us that opportunity, that is the mandate of that group to kind of look that 5, 10 years out. And so Steve is busy.

Yes. Okay. Okay. And Jake, I guess just since I have you here, and I want to put a final touch on that point. Some of the spectroscopy data we see or some of the spectroscopy technologies, things like that, I mean, it always seems like it looks good when it's big or when you could have a plugged into the wall power source or something like that. But is that what the issue is? It's not so much that you couldn't necessarily sense the glucose molecules in a reasonable way. It's just that power consumption or form factor are never going to get down to the major technical challenges to get down either power consumption or size?

It's a big part of it, miniaturizing the optical systems that are required at the performance needed. But even then, the larger systems work really well in an environment when you're completely still, you're not moving. I mean, there's so many complexities to trying to measure that molecule when you're not in contact with it. Our sensor, minimally invasive probe under the skin, it's in contact with the glucose. So it's easy to measure. You're trying to measure it through skin with light and stuff. And it's just -- there's a lot of complications that just make it tough to get an accurate reading that's reliable.

Okay. And last question, Jereme, just on margins. It still feels like -- or do you feel comfortable those G7 manufacturing costs can still get down to $1 below at scale? Obviously, that maybe doesn't have some shipping and distribution attached to it and all that. But that would be a path it would seem even if your ASPs fall to, let's say, $4 a day, 75% gross margin, if you do the simple math on that, I mean, is that conceptually still $1 a day a reasonable target to be thinking?

It is. It's what we're working towards as an organization. So what we're working towards is the standard cost, the cost to put everything in the box is $1 a day. You're right. Shipping warranty, all that, will push it above that. But that's the plan. And so we have a program in our organization designed on exactly that driving to that $1 a day. And that's at a 10-day sensor, right? And as we move to a 15-day sensor, there's opportunities there. So is there opportunities for us over time to look at margins and expand it? Absolutely. We continue to work on that. On the flip side, the reason why we continue to maintain our long-term guide of 65 is what if there's an opportunity to work with a partner like Roche where the gross margins, though, are going to distribute Dexcom ONE for us nearly, where the gross margins may not be as high because we're using a partner, but the delivery in terms of EBITDA contribution is still there. We're making sure that we have the flexibility in the organization to do that. But again, we -- longer term, as you think about 10- to 15-day plus G7 at scale, I think we're very bullish.

Okay. Remind me the timeline or the path on going from 10 to 15 days. I know -- the data on G7 looked pretty good at 13 days, 14 days, things like that. I'd argue probably as good as Libre 3 data out at those days, but it wasn't good enough for what you guys want to accomplish. So how do we think about the path or timeline?

Yes. Let me give Jake the floor on that.

Yes, I think -- so you're absolutely right, Jeff. The performance and accuracy of the product out to 15 days is great. It's all about how many sensors survive the full 15 days, and we're -- we want to provide the best user experience possible. And so there's a couple of pieces of technology that we've been implementing into G7, and that will get it to last out to 15 days, the vast majority of sensors. And so got to run a study to get that completed, and then 15 days is on the horizon.

Okay. And Jake, just last thought. You talked about gross margins and how that translates to EBITDA. I think your EBITDA target 2025 is 30%. Still feel comfortable with that? I thought that was the target you laid out at the Analyst Day.

It is, and we feel comfortable with it. And that's where we're going.

Yes. All right. And we've just clicked over to time 0. So perfect. Well, everybody, please join me in thanking Jake and Jereme here for a wonderful overview of DexCom. And as a reminder, the next presentation is set to begin at 9:05 including [ Novus ] in this room, Iveric Bio in Ballroom A, Intercept Pharmaceuticals in Ballroom B and IDEAYA Biosciences in the Morgan Suites. Thank you.