Insulet Corporation (PODD) Earnings Call Transcript & Summary

All right. Hi, everyone. Thank you for coming. We're very happy to have the full Insulet team with us here today. And I think we're going to start with some prepared remarks on some breaking news and then we'll jump into a fireside chat.

Yes, great. Thanks, Ryan. So it's great to be here with you. We've lots of exciting things going on at Insulet. And it's a good opportunity for me personally to address one particular update. This morning, we announced that we are pausing the Horizon pivotal trial. For most -- anyone in the room who doesn't know what that is, Horizon is our automated insulin delivery system that's under investigation. And we are about halfway through the trial, so about 9,000 patient-wearing days in a total of 21,000 patient-wearing days that we expect news to submit to the FDA for clearance. And we learned just recently of a software anomaly that in certain circumstances relate to how the CGM signal is processed. So this isn't an issue with the CGM with Dexcom, and it isn't an issue with our algorithm. It's simply a rare set of isolated circumstances where the CGM signal that's coming in for processing could be a misread. And in fact, there are -- it's such a rare circumstances. There have been no patient-reported incidences of the situation, and there have, of course, been no adverse events either. And so -- but because the situation has the potential to impact insulin delivery, we did opt to pause the study so that we could take 8 weeks to update the software, rectify the situation and then resume the study, which we expect to happen at the end of April. Anecdotal feedback from the study has been, frankly, terrific. It's actually been really emotional and gratifying. And most of the patients were reached over the weekend, to a person, everybody wants to stay in the trial and their biggest complaint was that they had to stop using Horizon in closed loop for a short period of time. And they're eager to get the software updated and the trial resumed. And that, obviously, we agreed, so the team is working very hard to do that. And I think about all of the significant initiatives we have underway at Insulet, Horizon being one of them, but also the DASH launch in Europe, U.S. manufacturing and unlocking of the pharmacy channel. All of these things have the potential to make an enormously positive impact on people living with diabetes but probably none more so than Horizon. So the team is committed to getting this resolved as quickly as possible. And at the end of the day, these learnings are very helpful because they ensure that we can launch the best possible product in the market. So I think that sums it up. And Ryan, I'll kick it back to you.

Okay. Great. Yes. Maybe a follow-up or 2 on that specific Horizon development. It would be great to hear technology. Again, it's a very rare circumstance, similar to the recent DASH update. Can you give any more color on, I guess, exactly what that circumstance is and why it would pose the risk, something to do with how correction boluses are processed? And then also, your confidence in that time line and why you feel very confident that 8 weeks from now, you'll be up and going.

Sure. Yes. So first of all, I'll say it's actually an issue that hasn't occurred yet. So that's kind of remarkable. In fact, the software anomaly was discovered on regular review of the real-time data coming from the remote monitoring system with Horizon. So in fact, there was no patient-reported event. This simply was upon review of the data by clinicians that participate in the trial and our own internal clinicians and engineers that we determined that there was the software anomaly. It has to do with the CGM signal that comes in, as many people know the way Horizon works is we ping the sensor for a CGM reading every 5 minutes. And then based on that reading, we predict out what the glucose or the CGM reading is going to be in another hour and then we adjust insulin. Every 5 minutes, we deliver a micro bolus of insulin. And in these particular circumstances, when the CGM signal is not received for any variety of reasons, right, for whatever reason, the CGM isn't delivering the signal every 5 minutes, the system is designed to account that. And this is in a rare set of circumstances where the CGM signal doesn't come in, and a number of other things happen. Potentially the CGM can -- or the system can grab the previous CGM reading. And so it obviously has not happened in the event that it's delivered a patient complaint. This is simply something that was discovered on data review. And actually, from my perspective, that's somewhat encouraging. Our teams have unprecedented access to large volumes of data now, and we're able to actually predict and find issues before they become an issue associated with the patient. So I'm proud of how quickly the team has assessed and reacted. It was just a couple of days ago that we confirmed the issue, very quickly communicated to sites and to patients and very quickly root cause the issue and confident in our time line that it will take us about 8 weeks to complete the code, validate the software update and implement so that we can resume the trial.

Okay. And to make sure I understand, 9,000 of 21,000 total patient days. That initial data is still -- you'll still be able to use that as part of the submission.

Yes. That's right.

Okay. And then in terms of your launch time lines. I know -- I think on the Q4 call, I remember correctly, you started to talk a bit more openly, not that you never tried to hide this. What about the desire to do a limited launch to make sure everything was fully functioning before you really scale this out? Does that mean we should think about this as really more of a mid-2021 broad rollout with something earlier in the year in a more limited fashion? Or any other color you can provide on that.

That's right. So we are predicting now that we'll be launching Horizon in early '21. We always planned on doing a limited market release, just as we did with DASH. And in fact, it's just the prudent thing to do with any new product launch. So we would expect that we will do a limited market release with Horizon. And so that could take any number of months. With DASH, that took approximately 6 months. And so we may expect a similar type of cadence with the Horizon product.

Okay. All right. That's helpful. Maybe shifting back to our regular scheduled programming and starting with 2020 guidance. So you guided overall revenue growth to 15% to 19% on a constant currency basis; and overall diabetes growth to 18% to 22%, constant currency. Consensus expectations were previously slightly above that. In our conversations over the past few days, even since the announcement, I think investors are probably reading a bit too much into your guidance to start the year. Looking back to 2019, you ended up exceeding the high end of your guidance by more than 7%. Can you talk a little bit about your philosophy to start the year? And acknowledging it's early, there also seem to be a bit less unknown this year relative than they were coming into last year. If everything plays out as expected in regards to pharmacy contracting and competitive launches, et cetera, do you see a path to 20% or even 25% overall growth in 2020?

Sure. Thanks, Ryan. Glad you set it up with a couple of key pieces there, right? One was referenced to last year because we do believe that we're still very early days here, and the setup for the year is similar to the approach that we took last year, which was to look at the overall business and decide what's the appropriate level to set guidance. So a few key things. One that I just -- speaking to people over the last few days, and if you focus on the headline and missed the difference in drug delivery this year versus last year, drug delivery is a business for us that we just OEM and produce for our partner, so really don't control the forecast there. It's still a good business for us. It's good cash flow but certainly from a guidance growth rate standpoint is more of a headwind for us this year than it was last year. So in working back up the product lines, international is a business that we saw significant growth last year, 47% for the year, but a large chunk of that growth was due to the onetime go-direct strategy. So again, comparing to last year's growth rate, we have to back out the impact of that. We grew about 75% in the first half of last year because of a 50% price uplift. So you really need to normalize the first half of the year for that. And where we landed on that from a planned growth rate is high teens, low 20s. And Shacey has shared with me the story, prior to that was 15% to 20%. So our belief and strength of the business internationally is actually growing from 15% to 20% to now high teen, low 20s. And we believe that we can sustain a growth rate in our international business at that level. So I think those 2 pieces are sort of the most impactful from year-to-year, as you mentioned, Ryan. And then I think the bigger question comes on the U.S. business. And this question around momentum, and we believe we have a lot of momentum in the business. We have a strategy that's delivering significant new innovation, both in product and in channel and in business model. And we want to remind everybody, it's still very early days. So we didn't feel it'll be prudent to set our guidance with a continued acceleration from where we performed, particularly in the second half of 2019. We're very happy with the early trends that we're seeing with our new DASH product in the U.S., with our new pay-as-you-go model in the U.S. and our move into pharmacy. But we just remind people that we're a couple of quarters into this. And so we're not going to set a guidance that is taking and creating a trajectory throughout 2020. We thought it'd be more appropriate to take a step back, think about where we -- what we grew in 2019. We felt it was a very strong year. So the high end of guidance is close to the same dollar amount of growth that we did in 2019. So that was where we thought we would peg it. It's early in the year. Obviously, we're tracking several different dynamics in the marketplace. Two that we think are ones that we have to monitor are attrition. One is our pay-as-you-go model, and it makes it much easier for patients to move on the product, but also it can make it easier for them to move off the product. We have not seen through year-end any increase in attrition, but we know that, that could potentially be the case because it's so early. So one of the things that we're monitoring.

Okay. And I want to come back to that point in the U.S. in a second, maybe one other guidance specific on gross margin. So you guided to 65% overall gross margin for the year and lower in the first half and higher in the second half. Can you provide any more details on how you expect to exit the year? And I think that would be helpful in just trying to frame what kind of ramp is required to get to the 70% target for 2021.

Yes. Really glad you asked that question because we've got this one quite a bit as well since earnings, and we'd like to provide just a little bit more color on it for people. So what we've guided to is a consistent year-over-year gross margin. And we finished the year at 65% in 2019. So overall, for the year, we think we'll finish relatively consistent to that 65%. What we talked about on our earnings call was that the first half of the year will be lower than the second half. And that's driven by this dynamic of bringing on and ramping up our line 1 in our U.S. manufacturing, and by midyear, bringing on line 2. And so those are both very inefficient lines. We're putting stock into inventory at high cost of goods sold. So that's going to put pressure on our gross margins in the first half of the year. And then eventually, as we scale, they become more efficient and will give us a tailwind into the end of the year and into 2021. So we wanted to put some numbers around that just to help people. And we think it could be up to a 200 basis point impact, unfavorable below the 65% in the first half of the year, and they're recovering that in the second half of the year to, again, average out relatively consistent to that 65% year-over-year.

Okay. So exit the year, closer to 67%. And from there into 2021, you continue to gain more efficiencies as you scale those U.S. lines?

Yes. We're not going to put specific points on it just because we've got so many scenarios running. And the ramp that we have in U.S. manufacturing is going well but takes a long time. And so we have to see exactly how it plays out over the next few quarters. Having said that, we think, looking at our scenarios, the ramp in the first half should be offset by the benefit in the second half, leaving us again right in that relatively consistent range around 65%.

Okay. Excellent. That is helpful. Maybe shifting to the U.S., so like you talked about in terms of baking in potential uncertainties. Looking at your guidance, at least in our opinion, the U.S. business seems to be the most clear, potential source of upside. I want to talk a little bit about the pharmacy transition as well as the inflection of type 2 adoption you've seen over the past couple of quarters. Maybe starting with the pharmacy transition and pricing. So we know that pricing has been a tailwind to your U.S. revenue growth over the past 2 quarters due to increasing pharmacy channel mix. You haven't provided a ton of specifics for reasons that make a lot of sense. However, you have made clear that volume growth has been the majority of your U.S. growth. The majority leaves a lot of [indiscernible] provide anything more for us to think about in our models in terms of how much of your growth in the U.S. have been driven by volume. Has it been 80%, 90% or something -- any window you provide would be really helpful.

Sure. And for one, I can start on the end of your question around volume and the drivers of our business, and maybe Bret and Shacey can jump in on type 2 and the pharmacy. So to help dial that in, Ryan, we should think about volume being the majority of the growth. We can't put a fine point on it at this point. But a smaller contribution from the mix benefit of moving into the pharmacy. But you should think about the majority of that is really driven by volume.

Yes. One update for the group is that -- and I think it's the last update we gave in terms of our total base going through the pharmacy was north of 20%. We're now able to share that 25% of our businesses going through the pharmacy. And I mentioned that because it's still -- it's a growing segment for us, but it's still a relatively small segment. So when you think about the drivers of growth for the entire business, it really is primarily volume because any price benefit we're getting is still on that smaller growing segment. It will become more important. But right now, it's 25% of our business.

Okay. It's worth a shot. As we think about 2020 and beyond, acknowledging you're still in the process of doing some contracting, only 25% today, you made clear, you expect that to grow very significantly. Should we expect your U.S. revenue growth to exceed your U.S. volume growth for the foreseeable future as that transition plays out?

Yes. I think that's the way to do the math, is assuming that the mix benefit, which, as a reminder for everybody, we changed the business model as we moved into the pharmacy to remove the 4-year lock-in period and to remove the upfront cost of the PDM. So we get the PDM for no charge, but we moved the sub-premium into the pod. And so in the beginning, it's actually unfavorable for us. But over time, that mix benefit begins to help us. And so as Ryan said, as long as we have a mixed premium helping us, the sales growth rate will be slightly ahead of our volume growth for that reason. I don't know, Bret, if you...

Yes. The best predictor of how that will grow in the pharmacy channel, in particular, is new starts, right? So we've talked about access being in excess of 50% of covered lives in the U.S. And new starts to Omnipod has mirrored access, which is an encouraging thing. So that will show you how it will grow over time, this 25% in the pharmacy channel. We also -- we clarified that this was another record new patient start quarter for worldwide. But it was also a record new start quarter for us for both type 1 and type 2 patients. So the type 2 business that we've been getting in the pharmacy channel has been real upside because our type 1 new starts have been still at record levels.

Great segue. Yes. I think that was one of the most incremental positive data points on the Q4 update was that the proportion of your U.S. new patient starts that were type 2 increased to 30%, up from -- in Q1 of 2019, you disclosed it was 15%. So that's moved significantly along the lines of the pharmacy access you talked about. So clearly, Medicare coverage in the pharmacy and 0 upfront costs have been huge catalysts for that patient population. It's really early days. How do you think about sustainability of that demand? And given your MDI mix, do you think that the overall market penetration is accelerating? Or is this really being driven by what -- is this Insulet specific, I guess, is my question?

Yes. I'll take the first part of that question about the sustainability of type 2. So this is the third quarter in a row that we've talked about type 2 as an inflection for new starts. And so to clarify, we said that around 10% -- that we said 10% to 15% of our base was type 2. And we've just noticed that a more significant portion of new starts was type 2. And so we acknowledge that 3 quarters ago, and we talked about for 3 quarters in a row, calling out that 30% of new starts this past quarter were type 2 patients. So we're starting to see that as a sustainable growth for the pharmacy channel. For the reasons you pointed out, Ryan, that we've got Medicare coverage. 40% of type 2 patients are over the age of 60. And there aren't any of these restrictions in the pharmacy channel like C-peptide test, BG logs. And so we're starting to see the true demand of the product in the pharmacy channel. We're starting to get confidence from physicians that when they write a prescription, those patients will get on product because there's no distinguishing factor between type 1 and type 2 patients in that marketplace. So we do see it as sustainable.

So I guess the inflection in CGM use has probably helped a bit as well, and that seems like a sustainable trend.

No question.

Yes. That's right. And I think it is unique to Insulet because of our access position among people living with type 2 diabetes in the pharmacy channel. Essentially, there really isn't a difference in how people living with type 1 and type 2 insulin-dependent are treated. And so we know that 40% of the people living with type 2 insulin-dependent diabetes are over the age of 60. So this access that's established in the pharmacy channel with Medicare is very helpful to unlocking that opportunity. And then we also know that the product is particularly appealing to people living with type 2. And so those 2 things put us in a unique position to really address that unmet need in a better way.

On the topic of the products, do you think that separating concentrated insulins because I think the past couple of quarters has demonstrated that you don't necessarily need those products to drive strong growth in this patient population. Do you think that over time, you'll have to tailor innovation to this patient population, whether it's simplification of features or the user interface or something specific to type 2? Or do you think the core type 1 products are sufficient to continue to drive those strong demand? How do you guys think about that?

Bret, do you...

Yes. Well, the answer really is we know that the needs of type 1 and type 2 patients are actually fairly similar in some regards, right? But type 2 patients could value some things even more than type 1 patients like discretion, which our form factor is already best in class for ease of use, which we see today many type 2 patients that are starting on product will preset boluses within the DASH so that there's a small, medium and large bolus to really simplify that process of counting carbs and delivering meal-time insulin. So the product already accommodates a lot of these things. But simplicity and ease of use is something that we take in mind with everything that we do. We think Horizon take that -- takes that to the next level. With concentrated insulins, specifically, we initially designed a different user interface for those products. We'll see. But just keep in mind simplicity and ease of use, but that's valuable to both type 1 and type 2 patients. So...

Yes. I think when you think about really unlocking the entire multiple daily injection user population, which is our focus, do think simplicity and ease of use, whether you're living with type 1 or type 2, are really, really important. And it might be slightly more powerful in the type 2 population. But that focus, how do these systems actually integrate into your life, how do they reduce burden, that's important across both segments.

Okay. That make sense.

Phone control, by the way, is the #1 requested feature for both type 1 and type 2 MDI users. So we point to Horizon and having the ability to control the pod from your phone is probably the #1 thing that they're asking for today.

Interesting.

So Ryan, I think part of your question, too, is, are we a category grower, and we are. From a share standpoint, we're really not subject to 2 pump renewal cycles. We don't see those kind of trends impacting our business. And in our earnings call this quarter, we actually ticked up the percentage of new patient starts that are coming from MDI up to 80%. We typically talked about at around 75%. We saw it tick up to 80%. And part of that is driven by the type 2s because we're differentiated in bringing type 2 multiple daily injection customers into our product. And so I think to the second part of your question really is we definitely see ourselves as a category grower. We think we're bringing a differentiated value proposition to people who have been on MDI, and over the last 10 or 20 years, haven't made the move to pumps because they see the value of Omnipod.

Okay. Last type 2 question. Unless we're way too low on your patient base, it seems as though utilization within the U.S. business has picked up across your installed base, especially over the past 2 quarters, which is correlative, if that's the right word, to the type 2 mix that you've seen over the past couple of quarters as well. Do you think an increasing mix of type 2 patients could be a driver to this, if we're correct, slightly increased utilization? And how do you think that -- about that over the longer term? Do you see any risk from a payer perspective? Is that a continuous trend over time? Or is that just an overall very positive trend for Insulet and we shouldn't worry about any potential risks?

Yes. I would say that -- I don't know that we've seen an increase in utilization. So it's certainly possible that, that will happen. Remember, type 2 patients and type 1, for that matter, they start on DASH and doing that primarily in the pharmacy channel. So we don't get real patient-specific information, but we do see volumes. We see prescriptions. So we have a general idea of utilization. And as encouraging as type 2 uptake is right now, it's still a real small portion of the base. And so I don't think we'd see it yet, but it's a trend that we're going to watch over time.

Okay. Interesting. Horizon. So as early experiences with 670G and Control-IQ demonstrate, ease of use is critical in closed-loop systems. You've consistently emphasized your focus on ease of use since you started talking about Horizon a few years ago, specifically on ease of use. Can you talk a bit more about what you believe will make Horizon differentiated versus a system like Control-IQ out in the market today?

Yes. Sure. This has been a primary focus of us. In fact, when we started out and just sort of defined our philosophy as it related to Horizon, we said that we wanted to bring people better outcomes and more time and range but not compromising on simplicity. So we specifically wanted to bring the easiest-to-use products to market. And so we made some specific design choices from the very beginning. One example is putting the algorithm on the pod, rather than putting on a separate device that would need to be carried around. The other example is phone control. These types of design choices took more technical work but also provide incredible ease of use for the user. In the case of phone control, being able to control your pod by an app on your Android phone enables you to eliminate a component of the system. And in the case of putting the algorithm on the pod, it means that you have this on-body ecosystem. The pod speaks directly to the CGM. The CGM speaks directly to the pod, which means that you can go for a run or take a shower or go for a swim and stay in closed loop. Part of the learning experience on other systems on the market was that part of that frustration and lack of ease of use came from when the patient had to go in or out of closed-loop and the steps that were required to do that and the frustration in knowing that you were getting kicked out of closed loop a lot. So those design choices, putting the algorithm on the pod should enable the user to stay in closed loop more frequently. We believe that will lead to better outcomes, but it will also lead to a better patient experience because you're not trying to always get back into closed loop. And then the other area where we've invested quite a bit. If you think about just even the clinical trial of the study, while other systems focused on well-controlled adult populations, in some cases, coming primarily from pump utilization, we did with the current trial with Horizon and with all of our pre-pivotal and IDEs, we focused on the pediatric population and the multiple daily injection user population. As Wayde mentioned, 80% of our new users are coming from multiple daily injections. And it's a different proposition to provide an onboarding experience simple enough where the MDI user can just go right on to Horizon and be in closed loop and be getting great results. You have to design the out-of-the-box experience differently, and you have to design the user interface and the experience differently. And we did that. We're very proud of the fact that we will launch with a system that easily MDI users can go on and that kids down to the age of 6, and we'll follow with youngers, will be able to use it as well successfully. So we think we've kind of hit that nail on the head. And then the last thing I'll say is just about personalization. When you're making a choice around simplicity versus complexity in the system, one example of the things that you're considering is how many knobs and dials does the user or the clinician need to dial in, in order for the patient to do well on the system. Examples are your insulin-to-carb ratio, your basal rate or temp basal rates, those types of inputs in order for the algorithm to get personalized to you and do well. We have an adaptive smart algorithm. And so we don't need to dial in those knobs. Actually with very little input by the physician or the patient, the system will learn and grow with the user. And so I always give the example of growth first. As kids get older, they require -- they go through hormonal changes. They require more insulin or less insulin. After they're teenagers, the system will be able to learn that and adapt with the user. And so those types of things take a lot of technical work and a lot of great design work, but that's the choice we made was to really focus on ease of use. We've got -- we will do great in terms of time and range. But the difference in terms of how these systems really serve the user in their life is going to come down to ease of use in their everyday life. And that's what we wanted to differentiate on.

Great. Maybe one international question as we get down to the last couple of minutes. So you expect about a 20% growth rate within your existing international geographies. How do you think about the sustainability of that over the next 3 to 5 years? And acknowledging that new geographies aren't materially impacting your near-term revenue growth as you begin to scale and launch into meaningful new geographies over time, why shouldn't we expect your international business to accelerate with those catalysts, plus the new products that you should be launching internationally as well?

Yes. So you're right about all of that. But the -- we've mentioned that we're going to enter into 5 new markets this year. And while those -- you're right, Ryan. Those don't make really a significant impact into the year because we -- of our annuity model. So it's about building the base so that we get this recurring revenue. And as the base gets larger, that revenue becomes more significant. It is significant, though, to our longer-term plans. So it's 5 markets this year, and we see a ton of opportunity in the world where we can continue this process of expansion every single year. And over time, those markets do make a significant difference into revenue, but it does take getting users onboard and building this base of recurring revenue. And so -- well, in 2020, we haven't called it out as a real significant driver because these markets come later in the year. Over time, it will be a driver of our business.

And to review that as additive to your existing markets growth, is that 20% at the midpoint sustainable as you launch into other geographies and launch new products?

Yes. I think you should think about it just the way we laid it out, high teens, low 20s. We think by continuing to bring new innovations, we're creating DASH and more languages to be able to take it into more countries. We'll add more countries over time. Just a reminder that as all companies deal within Europe and internationally, it's more complex, right? And as Bret mentioned, in our annuity model, especially when moving to new countries, and I think you referenced it in the beginning, Ryan, it takes a while to build up a critical mass number of customers in order to have that region grow to a point where we feel that we have a critical mass there, and we can expand upon it. So I think the right way to think about the region is that high teens, low 20s percent growth rate. And we'll feel if we're executing well on our innovation as well as our regional expansion plans that we'll be landing in that high teens, low 20s.

The advantage the U.S. has in our existing markets is that -- over our existing markets is we haven't unlocked type 2 outside of the U.S. and don't have that benefit yet of mix with pharmacy, but volume growth should be strong internationally, especially with market expansion.

All right. Great. I think we have to leave it there, unless anyone has a 5-second question. All right. That's it. Thank you guys very much.

Thanks, Ryan.

Thanks, Ryan.