Memorandum

Dexcom 4Q13 – Record sales up a striking 61% in 4Q13; operating cash flow positive (again); Kevin Sayer to become CEO in 2015 – February 21, 2014

Executive Highlights

  • In 4Q13, Dexcom recorded $51 million in product revenue, a 61% year-over-year gain  (20% sequentially). Full year 2013 product revenue was $157 million, up 69% from 2012. Dexcom was operating cash flow positive for the second straight quarter and for all of 2013.
  • President Mr. Kevin Sayer will become Dexcom’s CEO effective January 1, 2015. Current CEO Mr. Terry Gregg will transition to a new role as Executive Chairman of Dexcom’s Board of Directors.
  • As noted earlier in the quarter, Dexcom received FDA approval of a pediatric indication for the G4 Platinum. No specific timing was shared on Dexcom Share (under FDA review), the Animas Vibe (under FDA review), the Tandem t:slim (not yet submitted), Gen 5, or Gen 6.

Dexcom reported 4Q13 financial results yesterday afternoon in a call led by CEO Mr. Terry Gregg and President Mr. Kevin Sayer. Big news on the management front out of the CEOs office concerned Mr. Gregg, who will transition to a new role as Executive Chairman effective Jan 1, 2015; Mr. Sayer will assume the title of President and CEO at that time. Management’s brief prepared remarks focused on Dexcom’s outstanding 4Q13 and full-year sales and operational performance, which exceeded expectations. Most notably, Dexcom was operating cash flow positive in 3Q13, 4Q13, and for all of 2013. (Revenue had been pre-announced at JP Morgan.)

On the product side, the big update in 2014 came with the FDA approval of a pediatric indication for the G4 Platinum in early February – Dexcom started shipping just days following the news. In terms of future products, no specific timeline updates were shared. Work continues to progress on Dexcom Share and the G4 Platinum professional use indication (an “active dialogue” with the FDA), the Animas Vibe and Tandem t:slim pump integration partnerships (“final development, testing, and regulatory matters”), and the Gen 5 and Gen 6 platforms (“in pretty heavy development” internally). Management shared some nuance and new details regarding the latter efforts, including that Gen 5 will have a receiver in addition to sending data to the smartphone – this, of course, makes us want to have Dexcom and Insulet reconnect all the more!

See below for our lucky seven top pieces of news from Dexcom’s quarterly report, including the top four financial and top three pipeline highlights, followed by a product pipeline summary and Q&A.

Financial and Business Highlights

1. Dexcom recorded $51 million in product revenue, up an impressive 61% year-over-year and up 20% sequentially, as reported at JP Morgan. Most importantly, this performance was on two very challenging comparisons, as sales rose 52% year-over-year in 4Q12 and 102% year-over-year in 3Q13. The 20% sequential gain from 3Q13 was particularly notable, given that quarter’s record high sales. During Q4, Dexcom added more new patients and sold more sensors than in any other previous quarter in the company’s history. Although management did not disclose exact figures, to put this into perspective, 4Q13 revenue of $51 million easily exceeded full-year 2010 revenue of $40 million. At JP Morgan in January, management disclosed that there were ~54,000 unique patients on the G4 Platinum as of December 31. Our sense is that patient “drop off” declines less and less since the G4 launched since overall product frustrations are down, particularly for very committed patients and families. As we have said before, we think the artificial pancreas will be the real “killer app” for CGM that will move far more patients toward it. As a reminder, still less than 10% of all patients are on CGM, so there is still plenty of upside.

  • Full year 2013 product revenue totaled $157 million, up 69% from 2012. This performance handily beat the $130-140 million guidance management stuck to in the 3Q13 call (which we had thought sounded rather low given G4 momentum). At the time, management did not raise guidance despite very strong 3Q13 performance – concerns were the launch of Medtronic’s MiniMed 530G/Enlite and a challenging comparison to strong 4Q12 performance. Clearly, neither seemed to have meaningful impact on the business’ growth trajectory in the fourth quarter.
  • The international business “exceeded expectations” in 2013, growing 80% year-over-year and accounting for 8% of product revenue ($12.5 million). In an encouraging sign, Switzerland and Slovenia recently approved CGM for patients. However, management acknowledged that the European business is still largely from cash paying patients and diagnostic use from physicians – future business expansion will rest on gaining CGM reimbursement in other European countries. Still, EU guidance, unlike the US, encourages clinical trials to use CGM – we see this as a positive because “time in range” can be reported, which is more understandable to patients and HCPs.
  • Management did not change the 2014 product revenue guidance issued at JPM 2014: $205-$225 million (31%-43% growth from 2013). This is slightly wider on the top and bottom end of the 35-40% growth range than management has used in the past ($211-$217 million). The guidance builds in Dexcom Share and the pediatric indication, but no partnership revenue from Animas and Tandem. Management underscored that the first quarter is the slowest for Dexcom’s business (e.g., 1Q13 represented only 18% of Dexcom 2013 sales), as insurance deductibles reset and flexible spending accounts are unfunded. Indeed, patients’ average out of pocket expense during the first quarter is 40-50% higher than the average spent in the fourth quarter of the prior year. If 18% of the total is scored, that would imply $37-$40 million in revenue for 1Q14, which sounds on the low side following $51 million in 4Q13, even given all the points about insurance premiums, etc.  

Worldwide Product Revenue

 

4Q12

2012

1Q13

2Q13

3Q13

4Q13

2013

Product Revenue (millions)

$31.7

$92.9

$27.8

$35.5

$42.5

$51.3

$157.1

Year-over-Year Growth

52%

41%

49%

65%

102%

62%

69%

Sequential Growth

 50%

-

 -12%

28%

20%

21%

-

 

  • Although the new 530G is obviously doing well, this has resulted in stronger category growth rather than Medtronic stealing share from Dexcom, as best as we can tell at this stage. It is difficult to know what is happening with previous Medtronic pumpers who used Dexcom’s CGM system; we expect to have more color on this next quarter, but overall, do not think it will impact Dexcom growth in any significant way. While if Medtronic didn’t have the 530G approved, there would be more people using Dexcom, we think other avenues of growth will easily make up for this segment that is lost (presumably once Medtronic users upgrade to the 530G, payers will no longer cover Dexcom on an ongoing basis). 

2. Dexcom’s operating performance was a major highlight of the call – the business was cash flow positive in 3Q13, 4Q13, and for all of 2013 (a “stretch goal” from the Board). Net loss was just $2.6 million in 4Q13, and absent non-cash charges, net income would have been $6.7 million. For all of 2013, net income (absent non-cash charges) was $5.2 million vs. a $29.2 million net loss (absent non-cash charges) for 2012 – that is quite an operational improvement. During 2013, Dexcom’s product revenues increased by a total of $64 million, and along with it, the company’s cash-based operating results increased by $34 million (“it's an amazing accomplishment for our team”). Dexcom is now achieving its gross margin targets of 70-75% on sensor disposables and ~50% on hardware. As the volumes grow, manufacturing improves (especially automation), and future sensors are rolled out (Gen 5, Gen 6), margins are expected to improve further – in Q&A, Mr. Sayer said “closer to 80%” margins are possible.

3. Management reminded callers of the FDA approval of a pediatric indication for the G4 Platinum, a “significant operational milestone.” As we noted in our report in early February, the approval finally allows Dexcom to market the G4 Platinum to children 2-17 years old in the US and to an additional 800-1,000 pediatric endocrinologists. Dexcom began taking orders immediately upon approval and completed initial shipments to patients “within a matter of days.”  It’s impressive how fast Dexcom moves from approval to launch! While we believe ~25%-30% of the type 1 population in the US is said to be under 18 years old (~375,000 to 450,000 patients), only 8-10% of Dexcom’s installed base is patients younger than 18 years (~4,800-6,000 patients), implying huge opportunity for Dexcom. [Said CEO Mr. Terry Gregg at JP Morgan 2014, “We’ve got some upside …”] In yesterday’s Q&A, management said that it expects this imbalance to equilibrate over the next couple of years. The expanded indication gives Dexcom an advantage over Medtronic in the US, as the MiniMed 530G with Enlite is approved for the narrower group of patients who are 16 years and older.

4. Dexcom CEO Mr. Terry Gregg announced that he will transition to Executive Chairman effective Jan 1, 2015; President Mr. Kevin Sayer will assume the role of President and CEO. Mr. Gregg will remain an employee of Dexcom and an active member of the senior management team – he said, poignantly, “I’m not retiring; I’m just graduating.” Though this move was eventually expected ever since Mr. Sayer was appointed COO in 2012, it does signal an important inflection point for Dexcom. The business has now matured quite considerably (headlined by positive operating cash flow for 2013, a “personal goal” of Mr. Gregg’s) and will be in good hands with Mr. Sayer at the helm. Said Mr. Gregg, “It is rare that a CEO has the opportunity to handpick his successor, work with him to build a world class organization, and hand the reins over with complete confidence in his leadership skills and his vision for the company.” Moving forward, Mr. Gregg’s role as Executive Chairman will involve building key markets and products for Dexcom, including type 2 diabetes, hospital/critical care (a longtime area of special focus for Mr. Gregg), and obesity. As Dexcom’s largest individual shareholder, Mr. Gregg assured callers that he has a “significant incentive to remain engaged and to see that Dexcom continues to be highly successful.” From a patient perspective, this is very meaningful. We enthusiastically salute Mr. Gregg for his remarkable leadership and devotion to Dexcom, which included coming out of retirement in 2007 (after he had retired as the Medtronic MiniMed President in 2002) to become President and CEO. Since then, Dexcom’s market cap has grown more than 13-fold from ~$230 million to ~$3.1 billion today.

Pipeline Highlights

1. Dexcom continues to have an “active dialogue with the agency” regarding the PMA supplements for Dexcom Share and a professional use indication for the G4 Platinum. As of JP Morgan 2014, an FDA response on Share was expected in 1H14, with launch slated to occur in 2014. (Management shied away from giving timing this specific in yesterday’s call. We wonder if this was more about managing expectations or reflects a hang-up in agency discussions.) The PMA supplement for Share was submitted to the FDA in July 2013, while the professional use indication was submitted in 3Q13. The Share approval will be great news for parents in particular; the professional use should be very helpful for type 2 patients needing help with diabetes management and therapeutic changes in particular.

2. No specific timing was shared on the Animas and Tandem insulin pump integration products. Management said that both companies “continue to press forward with final development, testing and regulatory matters,” and Dexcom has continued to assist them when asked. This was the least specific timing we have heard on these fronts, and when pressed in Q&A, management said that Dexcom will no longer speak publicly on behalf of partners (“we don’t want to step on their toes”). As of the last update at JP Morgan 2014, launch of the Animas Vibe was expected in 2014, while launch of the Tandem t:slim was expected in 2014/2015. We continue to hope that Dexcom and Insulet partner again.

  • Management views these pump partnerships as “additive” but “certainly not transformative.” Presumably, this is stated in terms of shareholders rather than patients, who would certainly like to see product integration (write richard.wood@d-qa.com on this front). Indeed, as Dexcom looks to Gen 5, management believes patients will be “more likely to prefer” to use their iPhone as the means of display and interaction with their CGM data (i.e., rather than pulling out a pump). We agree that many patients would likely prefer the smartphone interaction to pulling out a pump, particularly women and athletes that put their pumps in places hard to access (e.g., cycling shorts, a dress, etc.). Indeed, data from Europe suggests that many Animas Vibe patients are actually purchasing the separate G4 Platinum receiver, since they “often try to hide their pump, and it's easier just to pull the G4 Platinum receiver out of their pocket.” That said, we did learn that Gen 5 will have data on both a separate receiver, as now, or on a smartphone.

3. Both Gen 5 and Gen 6 are in “pretty heavy development right now,” and Dexcom is “running early phase studies” on several different versions. Management did not give specific timing on either product, as the company is focused on getting the Share remote monitoring system approved. In line with prior calls, commentary highlighted the interim nature of Share, which we expect will pave the way for Gen 5’s development, regulatory review, and rollout. As of the last specific updates on these systems, Gen 5 was slated for a ~2015 launch (though rolled out in several phases), while Gen 6 was slated for ~2017.  

  • Gen 5 will likely come out as a series of launches that include a simplified applicator, connectivity to a phone (in addition to a receiver connection as noted above), cloud-based data, new algorithms to improve accuracy, and the same G4 Platinum sensor. This is the first time we can recall ever hearing a mention of a receiver for Gen 5, as prototypes of Gen 5 have typically shown data going straight from the transmitter to the smartphone and no other visible display device. Management said that the Gen 5 receiver “won’t be as intricate” and will have better margins, implying it could be a very simple device with limited functionality that only displays the basic CGM data. This could be very similar to the strategy used with Sanofi’s iBGStar meter and LifeScan’s OneTouch VerioSync meter – a primary device that gets the job done but has stripped down abilities relative to the accompanying smartphone app. We imagine this could be a small, barebones receiver (e.g., a keychain-sized device with the CGM value and trend arrow and a single button) as the ultimate plan with Gen 5. We wonder if using the smartphone as the primary or singular display device has been a sticking point in agency discussions.
  • The Gen 6 system has a number of design goals: partially/fully eliminate calibrations, obtain a replacement claim to allow insulin dosing, and add interferent blocking. In terms of better accuracy, Mr. Sayer was quite frank – “we'll see how much better we can get.” He mentioned Dr. Steven Russell and Ed Damiano's study of the G4 Platinum, which observed an MARD of 10.8% – “pretty close” to where meters are now and lab instruments “aren’t a whole lot better.” Notably, management believes the design goals of Gen 6 will open up other markets, especially the hospital environment (drug blocking) and the type 2 market (reduced/eliminated calibrations).
  • Said Mr. Sayer, “We've got five years of innovation stacked right on top of each other. They are waiting to get out the door and get through the agency. These designs are more cost effective than what we do today.” Management framed Dexcom’s pipeline in two waves: a short-term focus on increasing connectivity and convenience (Share, Gen 5) and replacing fingersticks as the primary long-term objective (Gen 6).

Pipeline Summary

Pipeline Product

Timeline

Animas Vibe insulin pump with G4 Platinum CGM integration

PMA filed with the FDA in 1Q13. JPM 2014 called for launch in 2014.

Dexcom Share

[Remote monitoring via docking cradle, Bluetooth, and smartphone app]

PMA supplement filed at end of July 2013. JPM 2014 called for launch in 2014.

Tandem t:slim insulin pump with G4 Platinum CGM integration

PMA not yet filed. JPM 2014 called for launch in 2014/2015.

G4 Platinum Professional Use Indication

PMA supplement filed in 3Q13.

Updated G4 Platinum algorithm

[MARD improvement by two percentage points; remote software update]

Unclear – The 1Q13 call targeted FDA filing in late 2013/early 2014, with a 2014 launch. 

Gen 5 system

[Mobile platform, new transmitter, new algorithm, improved applicator, less intricate receiver, G4 Platinum sensor]

 

~2015 launch, though will likely be rolled out in stages

Gen 6 system

[New sensor with goal of reduced or eliminated calibration, insulin dosing claim, interferent blocking]

 

In 2Q13, received a $4 million development grant from Helmsley Charitable Trust based on milestones over “next several years”; ~2017 launch

Dexcom/Edwards GlucoClear 2

[Critical care CGM]

CE Marked; Edwards working to understand how the system fits into hospitals’ clinical workflow

 

Questions and Answers

Q: I wanted to ask a bit on two topics, first is the pediatric ramp. Do you have some sense of what percentage of clinics are reflexively prescribing CGM on diagnosis at this point? We have some high profile examples, but is it becoming more commonplace?

A: When you look at the larger number of prescribers that we don't call on, it's clear to us that there is an opportunity there to grow that market. I think it's reflective of the tools that we need to develop from electronic connectivity, as well, to reach out and expand that that capability.

As we said earlier, currently our patient base is between 8% and 10% pediatric patients, and we expect it to grow to be about 30%. That ramp will take some time as we begin to call on pediatric endocrinologists who we have not called on before. We'll build trust, and we'll get a few patients on the system; we expect it to take off from there, but it will take a little bit of time. Our team is focused on calling on these clinics. They've made initial calls, and as we look at our daily reads that come in, we are seeing a lot of pediatric patients. However, it's not widely disproportionate to our regular patients. It is growing, and it’s growing gradually, and it’s growing very nicely.

Q: 30% of your users would be pretty close to the overall type 1 population. So this won’t happen in a year. I assume you are thinking two, three, or four years?

A: Yes.

Q: And as we think about this notion of utilization and retention remaining steady – I believe this was your earlier comment – was this related to the fourth quarter specifically? It was flattish and you deliver the revenue growth without seeing an increasing utilization or retention vs. 3Q13?

A: The trends are consistent all the time. Our point is that we are certainly a lot better than we were a year ago. As far as the specific percentage increase, I can't give you one. Things are very good. If anything, patients bought more sensors in 4Q13 than they did in 3Q13. If we were to look at that raw data from an individual point in time, it would look higher just because of volumes; however, I wouldn't skew that because of year-end purchases. So we look at the things over time, and things are very good; they are what we want them to be.

Q: You've mentioned the development programs for Gen 5 and Gen 6. Do you have any specific timing on those, or when we might see regulatory submission or studies?

A: We run early phase studies on all our technologies before we go to publication, and we're running early phase studies on several different versions of these products in the pipeline now. As far as timing, however, we need to get the Dexcom Share approved first. That is really our first foray into the Gen 5 market; that will be our first cloud-based mobile application where you can share your data with others. I think that approval will trigger a lot of effort and lot of thought on our part on how we can accelerate things and how fast we can go. Those product lines are both in pretty heavy development right now.

Q: First, can you separate out the specifics of Gen 5. What exactly will be the benefits of Gen 5 over Gen 4? And then can you do the same for Gen 6?

A: The Gen 5 is going to be focused largely on connectivity, mobility, and convenience. More than likely, it will come out as a series of launches, rather than one big launch, with the end goal of Gen 5 being a simplified application system combined with connectivity to a phone, in addition to being connected to your receiver and cloud-based data. We'll go there in a series of steps. That system will use the Gen 4 sensors as currently configured, but with new algorithms that we've developed over the course of the past two years that will improve accuracy and reliability.

With respect to the Gen 6 system, that's our first step towards doing three things: 1) eliminating some of the calibrations; 2) giving a replacement claim or dosing claim, so you can dose insulin; and 3) ultimately eliminating fingersticks all together. As far as accuracy, we'll see how much better we can get. As you heard from the clinical data, Drs. Ed Damiano and Steven Russell's work already has shown a MARD ~10%, which is pretty close to where meters are any way, and lab instruments are not a whole lot better than that. We will continue to pick up accuracy point.

We also note that for Gen 6 in the future, we'll eliminate, or block, the interference that we had some contraindications for. As we move to a Gen 6 sensor, we will be able to move to other markets, such as the hospital, because the sensor will block the effect of all the compounds patients are taking. We may also be able to move to a type 2 market, because they wouldn’t necessarily need all the calibrations; if we get advanced enough and eliminate calibrations, that becomes a very simple system for type 2 diabetes and type 2 diagnostics. As far as timing, we're going to keep that pretty much under wraps for now.

Q: Could you speak to the disposable manufacturing? You are obviously making a lot of disposables at this time, probably somewhere near half a million a quarter. What is your current capacity per quarter? What do you need to do to expand that?

A: We’re not even on two shifts yet, and our capacity is certainly above what you just quoted as half a million a quarter. The manufacturing processes have become much more robust and much more standard. We have plenty of space. We can replicate any of these equipment lines quite quickly. We grew 69% last year, on an operating plan that was 40%, and we never missed a shipment. I don't think you will find the place more flexible and more robust than this one. We're not concerned about that.

Q: I just want to follow up on the commentary around the attrition and utilization. In my model, I have attrition trending down pretty consistently since the launch of G4, and sensors per patient actually picking up modestly from 3Q13 quarter to 4Q13. Does that seem about right to you? Also, as we think about 2014, I want to understand the comments you made on the first quarter and the seasonality. Do you think that these two trends continue to improve or do you think you've reached a kind of stable working point?

A: You had attrition kind of trending down; I'm not sure that's actually the right way to look at it. I think what we saw was a pretty dramatic shift starting with the launch of the G4 Platinum and attrition coming into a much more normalized place. I think what we've talked about publicly before – and we're not again giving specific numbers – is what we saw back in the MiniMed days, something that we're just more comfortable with. Certainly, attrition on the Seven Plus was much worse, potentially even than we expected as we converted the base over.

On the utilization side, there is going to be a point at which we know that not all patients are going to use sensors all the time. We think a much greater number of patients are using sensors virtually all of the time now as opposed to Seven Plus where it was much more sporadic, but again today, patient's can still extend the life. So you're never going to see a point where a vast majority of patients are using four sensors a month. I think we used to quote 2 to 2.5 sensors a month on average and that's probably bumped up, but I don't think going forward we expect utilization per user to pick up dramatically. However, what's important is that we continue to add new patients to the base, and those patients continue to use sensors virtually all the time. So, in an essence, the utilization continues to be very robust.

Q: So, the implication is, yes, there is a little bit of room for improvement.

A: At some point, patients are going to continue to extend the life of the sensor. I guess we can improve utilization by a part of the market. Anecdotally, we ask almost every patient we meet, how long do you wear your sensors? And we very seldom hear, “I swap it out every seven days.” If they're wearing it 10 or so days, that's about three sensors a month; if they wear their sensor 10 to 14 days, that about two sensors a month. It's somewhere in between that.

Q: Have you updated your expectations for the timing of Vibe in the US?

A: No, we didn't. They’re still working diligently. We've made the decision not to speak publically on behalf of our partners anymore. I would defer you to Animas and Tandem for specific on timing. Particularly Tandem, as they are a newly public company and we don’t want to step on anybody’s toes; I would defer you to the partner.

Q: I wanted to follow-up on the Animas and Tandem side. I understand that you can't talk about timing. But how do we think about progression in 2015 and beyond, particularly the impact to new patient adds from both of those products?

A: When we gave our guidance at the JP Morgan Healthcare Conference, we were pretty clear that we don't have any contribution from our partners domestically in our 2014 guidance. There is obviously some contribution from Animas who has been commercial in Europe for several years now. We’ve said this in multiple conferences, but I think that the relative importance of our pump partners in 2014 and going forward is, frankly, lesser than it used to be. Back in 2008 and 2009, we were in much smaller organization with a much smaller sales force. Today, we out man Animas probably close to 2:1 in the field. Tandem is doing a great job growing their sales force, and I think Tandem could certainly add more there.

I think the way I would really look at the pump partnerships going forward is that they’re additive. I can't give you specifics on how additive they are, in terms of specific numbers of patients – I think you would need to extrapolate what Tandem is going to give in terms of guidance, and from Animas, to the extent you can get information. But I think we look at them as additive; they're certainly not transformative to us. So I think particularly as we look to Gen 5 and as we move to the phone, we believe patients are going to, more than likely, prefer to use their iPhone as the means of display and interaction with their CGM data rather than pulling out their pump. We even have some data that suggests that in Europe, patients who have been purchasing the Vibe have actually purchased a separate Dexcom G4 Platinum receiver because patients often try to hide their pump, and it's easier just to pull the G4 Platinum receiver out of their pocket. So, as we move to the phone, I think the convenience in connectivity of the phone is really where our future lies.

Q: Obviously Medtronic is pursuing the low-glucose suspend role as sort of the next step towards the artificial pancreas. Now that you are working on integrating your CGM with some pump companies, what is your next step towards the artificial pancreas? Will you pursue some sort of low glucose suspend technology? What's next with the integrated system?

A: As we've stated before, we are the premier sensor component of the overwhelming majority of artificial pancreas programs that are in place around the world. Most of these are academic based, and they use a wide variety of different insulin delivery devices. Certainly, Tandem, Insulet, Roche, and Animas are all in that group. In many ways, we're somewhat agnostic from that perspective, and, ultimately, any commercialization of a technology within this classification of artificial pancreas will really be driven by the commercialization partner, not by Dexcom. We're happy to support those efforts, and we're happy to lend our expertise. However, the reality of our mission statement is that we’re here to replace fingersticks. We never lose sight of that. Each and every day, we look at the global market and the increase in total number of patients, either having diabetes or over the next 10, 20 years developing diabetes. What we do really well is developing advanced technology in sensor capability; that's what we will continue to do and where we will continue to focus our energies.

Q: Sensor gross margins were obviously a highlight of 2013; I think that sometimes gets overlooked. I was wondering if you can give us an idea where you think that can exit 2014? And now that you think you can trend above that previous 75% range, do you have a longer-term goal?

A: We think we can get closer to 80%. That'll require some changes in the fundamental assembly process and the ASM moldings and tooling. There is a lot that's got to go on there, and we're working on all those things. One of the things that is just kind of an easy base hit for us is, being in San Diego with our overhead rates, volume increases do a lot to help us on the sensor manufacturing side. As we continue to grow at a very healthy percentage and we pick up margin points there, we get more efficient. Right now, we think as you look at 2014, we move to the higher end of our 70% to 75% range in over time. We'd have targets to go from 75% to 80% again, assuming oil pricing remains the same, and we have plans to do that. There’s no question that our guys are working on that and are very cognizant of that on a daily basis.

Q: What percent of your hardware revenue in the quarter went to new patients?

A: We gave the number at the JP Morgan Healthcare Conference that we've sold 60,000 G4 kits since the launch of the product in 2012; 10% of those were to patients who already had a G4 system. Obviously, as we sell more transmitters and have more patients buy a second transmitter because the first one runs out, those numbers become cloudier even for us internally. Start with the 60,000 G4 kit number we gave out in early January, and you can work back from there.

Q: Can you address Medtronic's comments during their quarterly call about taking four points of sequential share in the pump and CGM market? Is that consistent with your commentary? Or are they talking about centers beyond your focus of 2,000?

A: It’s always difficult when you hear somebody from Medtronic speak as to how they calculate anything. It always reminds me of Pinocchio, quite frankly, when I hear a Medtronic person speak. Three independent investigators have far different numbers than Medtronic. I don't know what they're talking about. I read their transcript, and it said pump and CGM four percentage points. We had a $51 million quarter and saw more patients than we've ever seen before. We're not seeing that. We do channel checks to our customers. We still have a lot of Medtronic patients. They may have tried the [Enlite], because Medtronic forced them to buy the sensor components, so maybe there is some of that calculated into their numbers; we'll see. I think they're recurring obviously a lot of those Medtronic patients who tried it are still our patients, because they actually want an accurate sensor. I think from that standpoint, time will tell.

Q: Medtronic spoke about their Next Generation Enlite enhanced CGM at ATTD; did you see any data or hear any feedback then? Do you have any thoughts on that?

A: No. Quite frankly, we listened to whatever they had to say, and we saw nothing. They said they've launched in five or six countries. From our international group, we haven’t seen anybody commenting on it. We looked at the data they presented, and it didn't appear to be all that different than the Enlite 1. I think the experience level will determine what is truth and what is fiction.

Q: You talked more about your “CGM first” strategy at ADA or ADDE. Is that having any impact at certain centers that it might be implementing that more?

A: I think by our numbers it's clear that it has been pretty successful. We have impacted and, in fact, when you take some of our largest centers in the United States, now it is now standard operating procedure to put a patient on CGM well in advance of pumps. Then, you let the patient ultimately decide what method of insulin delivery that they want. Pumps are a secondary choice, a lifestyle choice, and not necessarily a need. When you're talking about glycemic variability, you have to know about your glycemic variability in order to do something about it. Pumps are convenience. I think that is finally resonating. We've been at it now over a year since we started messaging, and again we have $157 million for the year. I think we're pretty successful, especially compared to the little under $100 million in 2013.

Q: Could you speak a little bit more on the head count need for pediatrics on the approval in the US, with 800 to 1,000 pediatric endocrinologists? I was just wondering how we should thinking about SG&A leverage specifically as you look into 2014, based on the needs to expand the sales force there a bit? And maybe if you could comment on utilization specifically within the pediatric market; where is that today within that 8% of total sensor volumes? I’m assuming that would be extrapolated throughout the pediatric user base as that expands?

A: With respect to the sales force expansion, we took care that in 4Q13. We added literally 30% more field bodies than we had before. That expansion was, for all intent, complete by mid January, with just a couple left to hire in early January, and those have been hired. We've expanded as far as we're going to, at least today, for 2014. That expansion wasn't really pediatric driven, but rather driven by all of our business territories. All of our reps will sell pediatric and regular patients, but our territories have been redone. Our new reps are on the ground and doing very well. So, the expansion has occurred. With respect to utilization and pediatric patients, we don't have separate data broken out. Anecdotally, in a very young pediatric patient, you have a parent that's driving therapy. So parents are going to make the decision to have the child use the sensor all the time. So as the pediatric patients get older, particularly in their teenagers, that utilization becomes a little bit different; kids don't want to have everything connected to their bodies all the time. However, our utilization in the pediatric market should be at least equal to, if not better than, it is for adults, particularly as you look at the younger segment.

Q: You have the sensor ASP around $70 overall today on the Gen 4, and I'm just curious about your thoughts on pricing for the Gen 5 and Gen 6 and beyond, particularly when we do move away completely from fingersticks; where do you envision sensor pricing can go overall?

A: Sensor prices are computed on a per day reimbursement rate with our payers. That $70 reflects an average of about $10 a day for sensor wear. As we go out further over time, our best chance for price increases is probably extending the wear more days instead of seven. On a per day basis, it is possible we can get more money if we eliminate calibrations. However, we’re ultimately committed to the fact that this therapy is going to have to save money in our current healthcare environment if it is going to continue to grow and be successful. We’re running our business and planning on driving our cost down by being as efficient as possible and being able to deal with every pricing scenario comes our way. If we can replace fingersticks and have a sub 10% MARD with our sensor, it might be feasible to go ask for more pricing; however, we’re not focused on that today – we need to get to that technology first.

Q:  It seems like you have good volume covers to reduce overhead in the factories. Could you provide a little more detail on gross margins for sensors specifically? If it isn’t price driven, how do you get from the level today to where your vision is for potentially 80% gross margins on sensors?

A: We’ve never talked about increasing margins as driven by increased prices; it's always been based on improvement in our processes. The key word there is automation; a lot of the things that we do today are manual. For example, we still assemble applicators. We have plans to automate a lot of those processes over time, but those plans are largely dependent upon how quickly we go with our future product generations. Another thing that will help margins over time – margins as a percentage – is that when we go to the cell phone model of Gen 5, the receiver won't be as intricate; we may lose a few dollars on the revenue side with respect to selling fewer receivers, but we'll pick up more as we had more customers and sell more sensors to those customers. You’ll see a mix shift in the sales to more disposal products, but automation gets us there. As for our future product designs, we don't design anything without it being cost effective and more cost effective than what it's replacing.

We kind of joke about how long we’ve been innovating. We’ve got five years of innovation stacked right on top of each other. They are waiting to get out the door and get through the agency. These designs are more cost effective than what we do today.

Q: It seems to me that you have a nice ratio of revenue vs. operating expenses: 64 in sales, 30 in operating expenses, something along those lines. Should we see continued improvement there, and, if so, how are you going to drive that ratio, particularly the operating expenses side?

A: That's a very good question, and as we look forward, we have to balance three things, the first two being this continued accelerated revenue growth; our goal is always to grow the bottom line faster than the top, and we will run the business that way. The third piece of that is getting this innovation out to market, and figuring out how much we spend to get these new products out. Additionally, we have to consider how much we should spend to expand our field efforts and get more people aware of our technology. We're always balancing those three things. Our ratios have clearly improved in 2013 – we would hope that they continue to improve in 2014, and certainly that's our plan. That being said, as we look at our innovation opportunities – as we look at clinical studies we plan to do, as we look at possibly an additional field expansion if things go very well – we will kind of wait and see how those ratios shake out for next year. We will feel the need to keep going fast and go faster. That's how we look at it day-to-day.

Q: What’s going to be your approach for expansion in the European markets?

A: Our biggest win there over the past few months is that we're starting to gain traction to get reimbursement from the reimbursement authorities for CGM as a standalone. There are a couple of countries, Switzerland and Slovenia, that have now approved CGM for patients. There are efforts ongoing in several of larger countries, as well, because they see the benefit of that. Right now, our European business is, to a large extent, cash pay from patients or physicians who use a diagnostic; they're not getting reimbursed by insurance. As we get reimbursement in the European countries, that will be our biggest driver in those other markets.

Q: In prior calls you've talked about the penetration of CGM into type 1 diabetes in the US market, in the upper single digits close to 10%. Assuming you're splitting share with Medtronic, you’re still around that level. Were you guestimates off last year, in the sense that maybe the penetration was lower than you thought? Or was it just a lot of attrition of CGM patients in prior years?

A: I don't think our estimates were off per se. There are a number of other factors. We don't have great visibility into Medtronic’s numbers. We know what our numbers are, and we're pretty comfortable at what percentage penetration we have into the type 1 market in the US. Our attrition on Seven Plus may have been a little bit worse than we had initially anticipated as we converted the base. However, I think we remain very comfortable with our prior statements where we were – particularly over the course of this past year in 2013 – and where we're growing. It's hard sometimes, I think, to compare apples-to-apples between our numbers and the numbers that Medtronic puts out because our patients go on CGM, and they continue to use CGM on a go-forward basis. We believe that there are a number of things in Medtronic’s numbers. There are patients who may try the sensor and those who may fall-off the sensor. They also report revenues in sensor utilization for a completely different diagnostic product, their iPro. If they lump that into physicians using CGM, it's not really an apples-to-apples comparison because these are not patients who are using CGM on a real-time basis going forward; these are patients who might use it once a quarter or once or twice a year. In my opinion, it’s hard to characterize those patients as real-time CGM users. It gets really tough to quantify when we don't have great visibility into their numbers as to what the overall CGM penetration into the type 1 space is.

Q: With the mass and how you're seeing the attrition with Gen 4, the penetration rate should materially move higher, right?

A: We've given that guidance. Obviously, in that guidance, it assumes a pretty healthy new patient add rate, and we've talked about our reduced attrition rates, etc. Yes, we assume that Dexcom is going to continue to penetrate and grow our share of the type 1 space in the US, for sure.

Q: As more providers start entering into accountable care type arrangements, and potentially share some of those cost savings tied to patients achieving certain A1c levels, does the discussion with these doctors start to revolve around the use of CGM and their ability to get their patients A1c levels down? The doctors get paid more from the insurance companies and the patients do well, and Dexcom does well. Are we there yet or is it still too early to start that conversation?

A: It’s not too early for the conversation; however, it is too early for the implementation. We have long been of the position that we’re quite willing to go at risk with the payer opportunity to say to the payer, set forth your strategy or the patient group that you would like us to address. We are willing to do that if in the end we save the payer system money; we want to share in the savings.

I think what is eventually going to happen is that you’re going to first see it at physician level from payment; they're going to get paid for success. However, I'm of a belief, personally, that the industry addressing this population is also going to have to have some skin in the game in the future as a part of this whole payment structure. I do believe it's pretty clear that in all studies, that are not meta-analyses, you see a reduction in A1c when using CGM vs. any other technology within the diabetes space. I do see that as something coming forward, and we're happy to be part of that equation. I think it's a great opportunity for us then to really showcase what we can do with that environment. However, it's literally at a discussion stage, not in an implementation stage.

It’s in the hospital market that I get so excited. The G4 Platinum is good enough to be in the hospital, and some of the attributes discussed earlier will drive it even further. That market is clearly one in which if there is recidivism of a patient then hospital doesn't get paid. So, they have already experienced that pushback from the government. There's also a lot of dialogue that's going on at the endocrinology level about trying to achieve a certain level not only of A1c, but actual recording of average glucose levels in order to keep them between a certain bandwidth. Some of the criteria says, “we want these patients back closer to 140 mg/dl.” The dilemma is, how do you get them to 140 mg/dl without starving them while they're in the hospital? That is actually a very active dialogue that is going on among the endocrinology population. That's where we fit in quite easily. There's no better way to get a patient to 140 mg/dl, keep him north of 80 mg/dl, but below 140 mg/dl, than with CGM. You have to really worry about what they're eating – the last thing you want to do is starve a patient in the hospital setting just to achieve some artificial level of glycemic control.

 

--by Adam Brown, Hannah Martin, and Kelly Close