Dexcom 4Q15 – Record-high sales of $131 million rise 55%; new touchscreen receiver, insertion system, Android G5, dosing claim by end of 2016; G6 pivotal study to start in Q2 – February 23, 2016

Dexcom reported 4Q15 financial results this afternoon in a call led by CEO Kevin Sayer, capping off an extremely successful 2015: launch of the G4 Platinum Share receiver, an Apple Watch app, the exciting collaboration with Verily [Google Life Sciences], and US and EU launches of G5 mobile – all on top of a 55% year-over-year (YOY) increase in full-year sales and a 103% YOY increase in cash-based net income. We include below the top business and R&D highlights, followed by a pipeline summary and Q&A.

Financial and Business Updates

1. Dexcom reported record-high 4Q15 worldwide product revenue of $131 million, a striking 55% year-over-year (YOY) gain on a very tough comparison to a blowout 4Q14. Sales rose a remarkable 26% sequentially from record-high 3Q15 revenue. Full-year product revenue was $401 million, rising 56% YOY from 2014 – what a milestone! This now marks 13 straight quarters of 49%+ YOY growth for Dexcom since the launch of G4 in late 2012, and three straight quarters of all-time record high sales.

2. Dexcom’s estimated global patient base was ~140,000-150,000 as of December 31, 2015, with an estimated 75%-80% in the US (~113,000) and 20%-25% (~33,000) residing outside the United States (first disclosed at JPM 2016). Dexcom’s patient base has historically been 60%-65% pumpers, though management said it is starting to see greater uptake in MDI – we see a great deal of opportunity here, particularly as pens get “smarter.”

3. The response to G5 mobile has been “exceptional,” and Dexcom’s prospective patient pipeline is at an all-time high. The ramp of this launch has been “much steeper” than previous products. Management’s candor was notable in acknowledging that the move to the phone has added complexity to customer support, and tech support calls have increased in volume and duration.

4. Dexcom reported net income of $1.5 million in 4Q15, on par with 4Q14’s $1.3 million. Cash-based net income was $29 million (22% of revenue), a 55% YOY increase from cash-based net income in 4Q14. It’s encouraging to see bottom-line profitability (cash-based) growing at the same rate as topline sales – there’s no question this business model works.

5. Increases in cash-based net income are expected in 2016, though management plans to invest an incremental ~$40 million on four key infrastructure and pipeline initiatives: (i) the partnership with Verily; (ii) international expansion, prepping for potential reimbursement this year in UK, France, or Germany; (iii) increased manufacturing capacity; and (iv) an “advanced data platform.” We’re excited on all these fronts, especially the last.

6. Management devoted a passing sentence to a customer notification sent out today, which concerns an increase in complaints related to Dexcom G4 and G5 receivers. Some receivers’ speakers may malfunction and not sound an audible alarm or alert. According to the notification, the FDA is aware and Dexcom is working on an improved speaker for its receivers. We hope it is fixed soon and not a distraction as many patients are relying on the alarms.

7. Regarding the broader CGM environment: (i) management quickly addressed and dismissed competitive products from Abbott (FreeStyle Libre) and Medtronic (670G/Enlite 3); and (ii) Dexcom is “thrilled’ with the press release AACE/ACE sent out today following the Consensus Conference on CGM over the weekend (see our coverage of that meeting here). We are happy to see things moving quickly with the rest of the industry and think it has contributed to the entire field moving faster. Abbott’s results at ATTD on REPLACE have dissipated some of the “noise” associated with Libre, but we think patients remain very excited about the product and we believe it has high potential for type 2s in particular, a group now receiving more focus.

R&D Pipeline Updates

8. Dexcom plans to launch its new touchscreen receiver and one-button insertion system by the end of 2016 – we’ve heard some great things about the new inserter from those that have tested it. The ~50% smaller G5 transmitter was not mentioned, though in the past it has been grouped with these launches. Management said feedback on the new insertion system has been “remarkable” and patients “really love” the simpler design.

9. Management “remains confident” that an FDA-approved insulin-dosing claim will come in 2H16. More clarity on the path forward is expected in the coming weeks and 1Q16 call. This was fantastic to hear and was a much more positive update from 3Q15, where no timing was shared.

10. In new news, Dexcom plans to submit and “possibly launch” an enhanced version of its G5 Mobile app in 2016. The app will include “additional features and functionality, such as insulin-on-board data obtained from pump partners.” Dexcom still expects to launch an Android version of G5 in 2016, supporting the major Android phone brands.

11. Dexcom plans to submit an IDE for the G6 sensor in the next two weeks, with a pivotal study to commence in Q2. The pivotal study is the “most rigorous study” Dexcom has ever attempted, “by a long ways.” Management was clear that regulatory is still tough: “FDA is not making it easier to get CGMs approved.” This fact probably helps Dexcom more than it hurts them in the competitive sense.

12. The partnership with Verily [Google Life Sciences] is “going extremely well.” Verily has completed the first-gen transmitter and is testing performance with G5 and G6 sensors. Initial results have been “very promising,” and the first product is expected to launch in 2018. Management confirmed for the first time today that the pictures Dexcom has been showing at conferences are for the second-gen product with Verily, meaning the first-gen sensor in 2018 won’t be bandage-like the size of penny – this is not too surprising. Dexcom and Verily plan to go beyond just showing a glucose number and trend with these sensors (e.g., linking glucose values to behavior changes), which is very good to hear and not at all unexpected – this is the dream team.

Financial and Business Updates

1. Dexcom reported 4Q15 worldwide product revenue of $131 million, a striking 55% year-over-year (YOY) gain on a very tough comparison to a blowout 4Q14. Sales rose a remarkable 26% sequentially from record-high 3Q15 revenue. Full-year product revenue was $401 million, rising 56% YOY from 2014 and smashing the updated $350-$375 million guidance (which was increased from the original $340-$360 million guidance issued at JPM 2015). The broader launch of G5 mobile and the beginning of a direct-to-consumer marketing campaign drove the quarter’s growth (see below), and Dexcom’s prospective patient pipeline is the highest in its history. Mr. Sayer put the growth into perspective quite well: in 2011, Dexcom’s full-year revenue was 40 million, meaning sales have increased 10-fold in five years. [Note: These 4Q and 2015 numbers were actually slightly higher than the revenue pre-announcement last month at JPM 2016: ~$129 million and ~$400 million.]

• This now marks 13 straight quarters of 49%+ YOY growth for Dexcom since the launch of G4 in late 2012, and three straight quarters of all-time record high sales (2Q-4Q15). Talk about serious business and pipeline execution. Two key inflection points in 2015 (see below) came in 2Q15 (launch of Share receiver) and 4Q15 (launch of G5 mobile).
• It’s remarkable that Dexcom has been able to drive fourth quarter revenue as strongly as it has all these years – every year since it’s launch in 2007, fourth quarter revenue has represented at least 30% of the year’s revenues, and in seven of the nine years, it’s been a third or more (in 2008 and 2011 it was just slightly below that at 30% and 32% of total revenue).

Figure 1: Dexcom Quarterly World Product Revenue (2Q12-4Q15)

• The budding international business did ~$16 million in sales in 4Q15 (12% of revenue), growing 36% YOY, 18% sequentially, and driving 9% of the quarter’s overall growth. The international business recorded $53 million in sales in 2015, a 58% YOY rise from 2014 (keeping pace with the 56% YOY growth in 2015 for the US business). These results are very impressive, as Dexcom only has reimbursement in Sweden, Switzerland, and the Netherlands. Most of these sales are from the G4 Platinum, as G5 Mobile is available only in five international geographies (not specified). Dexcom’s prices are ~20% lower internationally, given the cash-pay nature of the business. Reimbursement is expected to come through in 2016 in at least one of the three big EU markets: France, Germany, UK.
  • As we expected, Abbott’s Libre is driving interest in sensors, but it does not appear to be pulling business away from Dexcom – by contrast, it’s expanding the market. How will this change? The dynamics will evolve significantly over the coming year as both companies seek reimbursement for their devices in Europe and elsewhere. In a cash-pay world, Libre has the advantage of being cheaper than Dexcom. But what will things look like once Abbott seeks reimbursement with REPLACE and IMPACT data? What about when Dexcom has data from the DiaMond study and other EU reimbursement trials? Will European payers reimburse for both products, or will they consider the cost-effectiveness side-by-side and choose one? What will patients choose, assuming out-of-pocket pricing is equal?
• Consistent with the guidance issued last month at JPM, 2016 revenue is expected in the range of $540 million-$565 million (35%-40% growth YOY), with a seasonal 15% sequential decline expected from 4Q15 to 1Q16. Obviously as Dexcom’s base of sales grows, it cannot maintain 50%+ YOY growth; analysts in the room at JPM seemed to agree that the growth expectation for next year was very reasonable. There were no additional comments on the guidance in today’s Q&A. We would point out that this is remarkable sustained growth for such a growing base of sales. The recurring base business and pipeline is driving this growth, though we expect several external factors could propel growth further (stronger reimbursement, more clinician awareness, closed loop, etc.)

2. Dexcom’s estimated global patient base was ~140,000-150,000 as of December 31, 2015, with an estimated 75%-80% in the US (~113,000) and 20%-25% (~33,000) residing outside the United States (first disclosed at JPM 2016). This represents an ~60% rise from ~90,000 patients as of the JPM 2015 update (when management reported ~50% patient base growth in 2014). Assuming 1.5 million type 1s in the US, that puts Dexcom’s CGM penetration in type 1 at ~8%, and total CGM penetration in type 1 in the ~10-15% range.

• The patient base number may be generous, as it includes those who don’t wear Dexcom 24/7; we’d estimate ~80-90% of US Dexcom users wear the system all the time, implying a 24/7 patient base of ~100,000 in the US. It could be higher or lower based on the assumptions: (i) on the high end, diabetes market research company dQ&A reports that ~85%-90% of US G4 customers and ~90%-95% of G5 customers say they are 24/7 users; (ii) on the lower end, assuming every Dexcom US user wore a sensor 24/7 (8-12 days on average, ~$70-$75 per sensor, and 70% of Dexcom US revenue from sensors) implies a US patient base of ~90,000, or ~80% of the US base Dexcom reported (~113,000). We assume international 24/7 use is lower, given the greater cash-pay; it’s harder to estimate since we’re not sure what price Dexcom receives per sensor. These estimates reflect multiple assumptions and ranges, and in the best case, are approximate.
• This also reveals a wide discrepancy in total revenue per customer: ~$3,000 in the US vs. ~$1,500 internationally (estimates based on 2015 sales and Dexcom’s shared patient base numbers). This could reflect two possibilities: (i) 24/7 use is likely greater in the US, hence higher revenue per customer; (ii) prices are higher in the US where there is reimbursement, and prices are likely lower abroad where the market is largely cash-pay (though changing in some countries; see above).
• Dexcom’s patient base has historically been 60%-65% pumpers, though management said it is starting to see greater uptake in MDI. This was not quantified, though it could mean Dexcom’s “CGM-first” message may be getting through to some doctors. In a couple years, management expects the patient base to be ~60% MDI, ~40% pumpers, more reflective of the broader US type 1 population. The unknown is what impact automated insulin delivery will have – could it bring pump penetration up to 45%? 50%? 60%? Impossible to know right now...but particularly as Dexcom becomes easier to prescribe and use, and as “smart pens” emerge, this will increase.
• The “vast majority” of Dexcom patients on pumps use an Animas, Insulet, or Tandem pump. Dexcom is not stealing Medtronic patients as a significant growth driver (“we get a few”). This has been a concern in the analyst community, as the expected 2017 launch of the MiniMed 670G/Enlite 3 could theoretically eat into Dexcom’s growth in current Medtronic pump users. It sounds like that is not a concern, though it’s hard to know how good the 670G will be and how many non-CGM users will choose it over Dexcom. It does seem like a big leap to go from BGM+MDI to hybrid closed loop, so we doubt many of the MDI population on Dexcom would be at risk. Depending on how strong the 670G is, Animas, Insulet, and Tandem pumpers could be at risk, though we also know they can’t just change pumps “anytime” (four-year warranty) and Dexcom’s partners will come out with competitive products. Dexcom still expects data in 2H16 from the DiaMond study (n=338, 24 months) testing CGM in MDI users. We see the Libre as a bigger threat once it is available in the US, though as we have said for some time, we believe Libre will expand the market more than anything.

3. Notably, management characterized the response to G5 mobile as “exceptional,” and Dexcom’s prospective patient pipeline as of this month is at an all-time high. Management emphasized that G5 is the first-and-only CGM FDA approved in adults and children (down to age two) that sends glucose data directly to a smartphone. As a reminder, MiniMed Connect launched in September (same time as G5), though it requires an intermediate keychain device to send pump and glucose data to a mobile app. Clearly, Medtronic’s competing connected CGM device has not impacted Dexcom’s 4Q15 – we think it is fantastic that all-around, there are better products for patients to help with insulin dosing, remote monitoring, peace of mind, and passive data upload.

• Management said the ramp of this launch has been “much steeper” than previous products. In particular, the move to the phone has added complexity to customer support – tech support calls have increased in volume and duration. “All Dexcom receivers are configured the same way; all mobile phones are not,” said management. Dexcom is working to file and implement G5 app enhancements, and since the launch last year, the company has released one G5 iOS app update. We expect others to follow this year. This is the tough reality of becoming a connected software company – the upside is a better user interface and ability to iterate; the downside is a constant need to keep up with consumer electronics, which move at a faster clip. Those demands are an important barrier to entry in our view – how can CGM startups possibly keep up with the software update demands that come with mobile connectivity?
• Marketing and social media are clearly working for Dexcom. Dexcom’s direct-to-consumer campaign began in 4Q15, as expected. Management characterized the results as “wonderful”: digital advertising drove 2.5 times the pharma industry average for click-through rates, producing over 475,000 landing page hits in 4Q15. The campaign helped generate Dexcom’s highest month-to-month growth in lead generation, which has continued into Q1 (when Dexcom typically sees a drop-off in lead generation at the start of the new year). Overall, the DTC effort delivered an astounding 200+ million targeted impressions, including online, in print, in doctor’s offices, and at diabetes events. We'd love to see the breakout for this. Unique visits to Dexcom’s website doubled in 2015 vs. 2014. The campaign will continue through the first two quarters of 2016; spending was not quantified, though it seems like a very good investment for building CGM awareness, which not enough patients know about. Undoubtedly, this is helping the entire field, though Dexcom in particular.
• For the first time in its history, Dexcom also launched a product (G5 mobile) at the same time domestically and internationally – an ambitious move for a such a revolutionary advancement. It’s worth contrasting this rollout with Abbott, who has only launched FreeStyle LibreLink by invitation only in one small country (Sweden).

4. Dexcom reported net income of $1.5 million in 4Q15, on par with 4Q14’s net income of $1.3 million. Absent non-cash charges, cash-based net income was $29 million (22% of revenue), a 55% YOY increase from cash-based net income in 4Q14. It’s impressive to see Dexcom’s bottom-line profitability (cash-based) growing at the same rate as the topline revenue – a great sign as the company continues to invest in R&D for future growth. Net loss in 2015 rose to $58 million vs. $22 million in 2014; the increase reflected $131 million in non-cash expenses ($37 million for the Verily partnership and $83 million in share-based compensation). Excluding the Verily deal, net loss in 2015 was $21 million, consistent with 2014 – again, a good sign on the profitability front, since 2015 sales rose 56% over 2014.

• Gross margin was 70% in 4Q15, remaining at the upper-end of the target for the existing product platform. Gross margins are expected to decline a bit in 2016 (~67%-70%) as Dexcom continues to sell G5 transmitters (the shorter transmitter useful life means patients get two for the price of one, reducing margins vs. the G4 transmitter), expanded manufacturing capacity, and making the costly switch to the new insertion system. This is a standout gross margin though we’re not sure it is sustainable – with all the clamor about insulin pricing, we expect more complaints about sensor pricing, particularly when Abbott’s Libre enters the US. We think this can be made up in volume, however.
• Average selling prices have not changed ($70-$75 per sensor, $800-850 per starter kit), but management did give more granularity on the individual component pricing: $400-$425 for the receiver, $400-$425 for each G4 transmitter, and $400-$425 for two G5 transmitters. We had not ever heard this breakdown before. We have heard from patients in the EU who lose the Abbott Libre handheld device that they are happy to replace it (only ~$60); over time, we would expect pricing to decline on the Dexcom products.
• Unfortunately for patients, there were no updates on the plan to move to pharmacy distribution. Consistent with the 3Q15 call, management highlighted the Anthem and United deals, and remarks again indicated some administrative hiccups with these moves (inconsistent application of prior authorizations, failure of some plans to actually move from DME to the pharmacy). As of 3Q15, Dexcom’s goal was to move 70% of its business to pharmacy benefits as the primarily reimbursement source over a three-year period. We’re not sure if that is still the goal.

5. Dexcom expects increases in cash-based net income in 2016 (improved profitability), though management plans to invest an incremental ~$40 million on four key infrastructure and pipeline initiatives. These were first disclosed at JPM last month, though there was much more color on the Verily and international fronts. It is excellent to see such a push for the future; management is clearly not resting on 13 consecutive quarters of 49%+ YOY growth, a smart move given stronger competition (see below) and even more cost pressure.

• The partnership with Verily [Google Life Sciences] is the biggest chunk of the $40 million – see more details on that below, which expects the first product commercialization in 2018 and a follow-on commercialization in ~2019-2020.
• International expansion will include establishment of an EU headquarters in Scotland, prepping for potential reimbursement this year in UK, France, or Germany. Dexcom currently has reimbursement in Sweden, Switzerland, and the Netherlands; 2016 is expected to bring reimbursement in France, Germany (“positive signs”), and the UK. Management said getting all three countries is an “unknown,” but it expects “at least one” and potentially two. Dexcom has generated all of its International revenue with just four people (an impressive ~13% of 2015 sales, or $53 million), so this expansion will be positive catalyst. Dexcom has also added a senior executive to focus on international expansion in Canada, Central and South America, and Asia. As CGM moves to standard of care, there is so much opportunity here, particularly in areas of such low penetration – this is really exciting for patients though will obviously require investment (one reason why we hope average selling prices and gross margin stay steady).
• Increased manufacturing capacity will include construction of a second US factory in Arizona with more automation. Management did not comment on this development.
• An “advanced data platform” – with millions of G5 data points flowing in every day, Dexcom needs to analyze it and do something valuable with it. We think there is lots of potential to build better products with this data; certainly, Medtronic has leveraged CareLink for many purposes (FDA approval, algorithm development), and we assume the sky is the limit here.

6. Management devoted a passing sentence to a customer notification sent out today, which concerns an increase in complaints related to Dexcom G4 and G5 receivers. Some receivers’ speakers may malfunction and not sound an audible alarm or alert. According to the notification, the FDA is aware and Dexcom is working on implementing an improved speaker for its receivers. The customer notification asks users to periodically test the receiver’s audible alarms and alerts (especially if the receiver gets wet or dropped), and if they are not working, Dexcom will send a replacement receiver (based on filling out the form at the bottom of the page). We’d note that the G4 and G5 receiver has always had some durability issues (e.g., the micro-USB port caving in), and we look forward to the updated touchscreen receiver expected to launch later this year – in the pipeline part of the call, management said the new receiver will be more durable (see below). This news would be of concern to patients and we’re a bit surprised this was not raised in Q&A.

• This news was only alluded to in a single vague sentence on financials: “We also note that we recorded a charge in Q4 relating to a potential increase in warranty expense resulting from an important customer notification we have issued related to the speaker component in our handheld receiver.” We assume it was not a focus because few receivers are impacted; presumably if it was material, there would have been more discussion or a press release.

7. Regarding the broader CGM environment: (i) management quickly addressed and dismissed competitive products from Abbott (FreeStyle Libre) and Medtronic (670G/Enlite 3); and (ii) Dexcom is “thrilled’ with the press release AACE/ACE sent out today following the Consensus Conference on CGM over the weekend (see our coverage of that meeting here). Management was definitely persuasive on the competitive front, but the commentary did not address some of the advantages these products bring to the table: (i) FreeStyle Libre is factory calibrated, 14 day wear, less expensive than CGM, and does not require a prescription; and (ii) the MiniMed 670G/Enlite 3 is hybrid closed loop, offering 80%-90%+ time-in-range at night; improved time-in-range during the day; and slated to launch in 2017, at least one year ahead of any Dexcom-integrated closed-loop device. Of course, we appreciate management openly addressing competition in prepared remarks, and certainly, patients will benefit from the faster pace at which all companies are moving. CGM is becoming standard of care and we think there is room for all three companies to flourish – we also think there is substantial room to help patients, especially on the type 2 side, and we’re also happy that all three companies seem focused on this group. All this said – while aspects of Abbott’s REPLACE results at ATTD were positive (lower hypoglcyemia), very positive results across the board (a substantially lower A1c in all groups along with lower hypoglycemia) would’ve created more noise for Dexcom and Medtronic and that did not happen.

• “First, the Libre is not a CGM. We have long said that one of the key benefits we provide to our patients is best-in-class accuracy combined with proactive alerts and alarms, particularly at night. We also now provide the ability to share data with caregivers to add an extra layer of patient safety. Some would have you believe that patients suffer from alarm fatigue and would prefer to go without alarms. That is certainly not the case with our system. That might be the case with other less accurate sensor systems where false alerts can occur up to 50% at the time. Regardless, our system is customizable. Patients can choose to silence their high and low alarms to their liking, except for the critical fixed low glucose alarm our system provides when the patient crosses a dangerously low threshold of 55 mg/dl. And with respect to alerts and alarms in the future, we intend to provide patients with even greater flexibility and better performance to get more out of their CGM systems. The Libre simply can't compete when it comes to system performance, especially as it relates to hypoglycemia detection. At the ATTD conference a few weeks back, a clinician summed up the value of Libre. He said, “Libre provides patients with a nice path to the product they really need, CGM.” [Note: We thought management’s commentary on FreeStyle Libre at JPM was more balanced and fair – we also point out that the “clinician” who spoke didn’t acknowledge the degree to which patients also like the Libre. We believe many of those patients are not looking for CGM but are looking for better SMBG. Still, some of those looking at CGM will consider Libre – the competitive environment in the EU is harder for Dexcom than in the US, but the larger point is that still so many patients need better glucose monitoring!]
• “Medtronic recently announced they would forego the launch of the 640G in the US in favor of the 670G with a new sensor. Since we haven't seen any meaningful performance data on the 670G and their new sensor, I reserve comment on this future system. I will only remind investors that when Medtronic launched the 530G back in the fall of 2013, its published accuracy data was similar to that of our G4 Platinum system. The actual field performance has not been close to that of G4 Platinum, even before our Software 505 algorithm update. Since the launch of 530G, according to dQ&A, an independent source, our CGM market share has gone from the mid to upper 50% range to approximately 76% at the end of 2015. The patient community has spoken; accurate, reliable, connected CGM is the most important product offering in diabetes treatment today. [Editor’s Note: The first data we’ve seen on Enlite 3 came at ATTD, suggesting an MARD of ~11%. Dr. Bruce Buckingham shared similar data from a camp study of Enlite at the AACE/ACE Conference over the weekend. We assume Enlite 3 is better, but how much better and how it compares to Dexcom’s 9% MARD is hard to say.]
• Management is “thrilled with” the positive takeaways from the AACE/ACE CGM Conference press release today (“What a great outcome for the diabetes community.” - we agree!): (i) a call in the press release headline to expand CGM access to people with diabetes; (ii) acknowledgement of the technology’s wide potential in “many people with diabetes,” particularly those with type 1 and even type 2s on intensive insulin therapy; (iii) acknowledgement of CGM’s effectiveness in reducing hypoglycemia and improving blood glucose control; and (iv) an important callout that “technological advances have improved accuracy and reliability” (critical because the CGM evidence base has not caught up with much-improved modern devices). It was awesome to hear such positive takeaways from a physician group as influential as AACE; we look forward to the full White Paper to be published in the future in Endocrine Practice.  

Pipeline Updates

8. Dexcom plans to launch its new touchscreen receiver and one-button insertion system by the end of 2016. The ~50% smaller G5 transmitter was not mentioned, though in the past it has been grouped with these launches. These pipeline projects have now moved ahead of the most recent timing shared at JPM (“late 2016 or early 2017”) and are consistent with the “2H16” launch timing shared in the 3Q15 call. These hardware improvements are lower cost and will offer a better patient experience, but also require a major manufacturing overhaul.

• Management said feedback on the new insertion system has been “remarkable” and patients “really love” the simpler design. We’ve heard this is the case from those who have tried it but were not authorized to speak about it. Here’s the process as we understand it: (i) peel tape; (ii) put on skin; and (iii) push a button to insert. This will be a significant upgrade from the current syringe-like design, which cannot be done with one hand, making upper arm insertion difficult. The call shared that clinical trials are underway for this new insertion system and are “more rigorous” than originally expected. The new insertion system will likely launch in phases, given the major manufacturing overhaul required. Management reminded listeners that the current insertion system has been used since Dexcom’s first product, and changing it requires swapping out “every mold” – “the biggest change operationally we’ve ever undertaken.” We have long thought the current system works fine for some patients and not as well for others and we welcome this imminent upgrade.
• The new touchscreen receiver will be more durable (i.e., less susceptible to wear) and offer a user interface in line with the G5 mobile app. In the past, management has said it will include Wi-Fi, though we’re not sure if that is still the case. We assume this receiver will work with the G5 transmitter, though we’re not sure about backwards compatibility with the G4 transmitter.
• At ATTD, Dexcom CTO Jorge Valdes shared that the new G5 smaller transmitter will cut the volume in half vs. the current transmitter (previously, no size reduction estimate had ever been given). It was odd this pipeline project was not mentioned on today’s call, as it was named in tandem with the new insertion system and receiver at JPM and in 3Q15.

9. Dexcom “remains confident” that an FDA-approved insulin-dosing claim will come in 2H16. This is such a big deal! More clarity on the path forward is expected in the coming weeks and 1Q16 call. Management said several times that the next quarterly call will offer a detailed plan – something we’ve long been waiting for, as these discussion have been under wraps for well over a year. The timing specifics were also excellent to hear, as 3Q15 took a step back from anything specific and only said FDA discussions “continue to evolve.” At JPM, management positioned the insulin-dosing claim as a competitive barrier to entry for the first time: “We will undoubtedly be the first company to have this label. And we want to set the bar high. You’ve got to have a sensor that performs as well as ours does.” This is a smart move against Abbott’s FreeStyle Libre, which is slightly less accurate than G4/G5 (MARD ~11% vs. 9.0%), but of course adds factory calibration and 14-day wear.

• “The innovation we have seen with our advanced sensor technologies gives us tremendous confidence that calibrations can be completely eliminated in the future, certainly for those patients not practicing insulin therapy. Much of the uncertainty here relates to the regulatory pathway and labeling for these products that have no calibrations.” It’s sounding more and more like FDA is the big uncertainty for Dexcom getting to factory calibration – not cost or technological barriers like management has talked about in the past. Management has talked in the past about different CGM labels for different products and patient populations, and we wonder what this means for the label on Abbott’s FreeStyle Libre (consumer version): will it have an adjunctive label with factory calibration? Would that label even matter if patients dose insulin off FreeStyle Libre readings anyways? Why is FDA drawing a line in the sand on factory calibrated CGM approved for insulin dosing? Isn’t the line kind of arbitrary, particularly given the danger of hand-washing and erroneous calibrations?
• An insulin dosing claim is critical for Medicare coverage, which could really expand the market for CGM: Dexcom estimates ~15%-25% of US type 1s are on Medicare, equating to hundreds of thousands of patient that could access CGM. The other Medicare opportunity, noted management, is intensively managed type 2s on MDI.

10. In new news, Dexcom plans to submit and “possibly launch” an enhanced version of its G5 Mobile app in 2016. The app will include “additional features and functionality, such as insulin-on-board data obtained from pump partners.” No pump partners were specifically mentioned, though we imagine Insulet and Bigfoot are at the top of the list. At JPM last month, Insulet guided for a 2016 submission of its next-gen PDM with Bluetooth and a paired Insulet app integrating Dexcom G5 CGM data; Bigfoot also disclosed plans to potentially launch a “connected insulin management system” in 2016, which would presumably use the Asante pump, Dexcom, CGM, and Bigfoot smartphone app without automation. We assume Tandem and J&J are thinking about app connectivity, though neither has shared very specific plans.

• Dexcom still expects to launch an Android version of G5 in 2016, supporting the major Android phone brands. We first heard at ATTD (confirmed again today) that Dexcom won’t cover every single Android phone model – there are way too many. Interestingly, Steve Pacelli said in Q&A that Apple’s iOS is a “much more important” platform for Dexcom’s future – he did not give any metrics other than saying the “vast majority” of hits to Dexcom’s website come from iOS users. That might reflect higher Apple iOS penetration in Dexcom users and potential users, or it might reflect Dexcom’s more robust offering on Apple right now (only Dexcom Follow is available on Android). CEO Kevin Sayer was quick to point out that he gets many angry letters from users that want Android compatibility, so the launch will of course be a helpful catalyst.

11. Dexcom plans to submit an IDE for the G6 sensor in the next two weeks, with a pivotal study to commence in Q2. The pivotal study is the “most rigorous study” Dexcom has ever attempted, “by a long ways.” Management was clear that regulatory is still tough: “FDA is not making it easier to get CGMs approved.” We assume the rigor refers to study size, in-clinic sessions, and perhaps FDA requirements for an insulin dosing-claim (which G6 will almost include, since management has repeatedly said it will not go backwards on labeling). Management reiterated that G6 will have 10-day wear, one calibration per day following startup, and overall performance better than G5 (with half the daily calibrations). G6 will leverage the new applicator and receiver slated to launch by the end of 2016 (see above). Management would not comment about a potential mandatory shutoff, though this obviously makes sense for Dexcom’s business model.

• Dexcom is working closely with automated insulin delivery partners to incorporate G6 into studies, offering a more robust dataset on the sensor’s performance. Previously we had assumed G5 would be used in these products, which could say something about how quickly they are moving, or how much better G6 is. Management said all of its pump partners are making excellent progress on advanced systems – as a reminder, this list includes Animas, Bigfoot, Insulet, Tandem, the Bionic Pancreas, and the International Diabetes Closed Loop consortium.
• We learned at ATTD that Dexcom will incorporate predictive hypoglycemia alerts into G6, offering a larger 15-minute prediction window – the design is balancing more advanced notice without increasing false alarms or annoying patients. The predictive alerts were the focus of many slides for the first time, which made us wonder if it was an answer to Medtronic’s hypoglycemia prediction app with IBM (which has a 2-4 hour prediction horizon, but could obviously increase alarm fatigue). The new predictive alert is clearly effective: with its addition, only 9% of 55 mg/dl BG events will give patients <15 minutes of warning vs. a much higher 42% with just a threshold alert alone (this is most key when patients are dropping rapidly, as going from a threshold of 80 mg/dl to <55 mg/dl can happen with just a five minute heads up). Overall, the new feature has a 93% predictive hypoglycemia alert detection rate (for 93% of YSI readings <55 mg/dl, a predictive hypoglycemia alert occurred between 0-30 minutes), but not at the cost of additional nuisance: for just 11% of predictive alerts, there was no YSI reading below 70 mg/dl within the next 30 minutes, making these alerts unnecessary). We’re glad to see Dexcom balancing valuable prediction with hassle factor – it doesn’t matter if an alarm is accurate if it is annoying!  

12. The partnership with Verily [Google Life Sciences] is “going extremely well.” Verily has completed the first generation transmitter and is beginning to test its performance with G5 and G6 sensors. Initial results have been “very promising,” and the first product is expected to launch in 2018 (“on schedule”). When the initial partnership was announced last August, the first launch was expected in two to three years (~2017-2018), with a follow-on product to be commercialized within five years (~2019-2020). Management confirmed for the first time today that the pictures Dexcom has been showing at conferences are for the second-gen product with Verily, meaning the first-gen sensor in 2018 will presumably not hit all the lofty goals: low-cost, disposable, bandage-like (size of a penny), 10-14-day sensor integrated into an advanced data analytics platform. This makes sense – we had been a bit surprised by any implication that the first-gen sensor would have all that. The first-gen sensor will use some of the initial prototypes, including miniaturized electronics, batteries, and advanced algorithms. The second-gen sensor is when the product will really start to look like a tiny bandage on the skin, making it truly minimally invasive and taking cost out of the system.

• Dexcom and Verily will go beyond just showing a glucose number and trend with these sensors. The goal is to link glucose values to behavior changes in a far more interactive and intuitive interface – critical for penetrating into non-intensively managed type 2s. Said CEO Kevin Sayer, “The one thing that we've learned about this market is it's not going to do us any good just to display a number and say, here you go. We're going to have to really give patients the ability to learn from the system that we provide and provide interactive suggestions like, “You exercised today. Look, how much better you did.” And that's the type of system interface that we're going to develop to address that market, again, which is different than what we have now. Just flashing a number on a screen isn't going to be enough, particularly for those that aren't using insulin. I would maintain that intensive insulin using type 2 patients can have the exact same experience as our type 1 patients have today.”
• Management reiterated commentary from JPM that both companies are learning from each other – Verily’s fresh perspective is clearly an asset, while Dexcom is bringing serious PMA and glucose sensor experience. This is what partnerships are all about, but this one seems like a particularly good fit. How will other big tech-diabetes partnerships play out (Verily-Novartis, Verily-Sanofi, IBM Watson-Medtronic Diabetes, IBM Watson-Novo Nordisk)?

Sensor and Software Pipeline

Pump Partner Pipeline

Questions and Answers

Q: What can you do with the G5 Mobile application? You talked about adding insulin on board as a feature, but can you add algorithms and start to do some predictive things with this version? Or are we looking further out before you start to incorporate more sophisticated things beyond simple data presentation?

A: We can do a lot of those things. The question is what do we do timing-wise vs. the G6, which we are going to, as I said earlier, submit the IDE on. We can move technology into that or put it into the other product, and we evaluate that as we go. So we're planning on keeping the G5 Mobile app similar to what it is and adding a few features to enhance it without adding a bunch of algorithm and prediction. Most of that is scheduled for G6 at this point in time.

Q: And those would be PMA supplements, right?

A: It will all be included in the original G6 filing. The G6 is a full PMA because it's a new sensor.

A: Some of that algorithm development is not exactly linked to a sensor. We've talked in the past about not tying ourselves down to G5, G6, G7 iteration; some of the algorithm work that is going on is kind of independent of the specific sensor platform.

Q: You alluded to the fact that a lot is going to happen in the next few weeks. Is that conducting clinical trials due to greater clarity on the dosing claim requirements of the agency? Is there some fundamental reason that the FDA wouldn't approve a near-term product or other product without a calibration stick? I mean, how high is that kind of hurdle for them?

A: I don't have an answer to that last question. We're discussing that rigorously as we prepare our future product offerings, particularly our Google product offerings we intend to launch here in about a couple of years. With respect to the dosing claim, we've had activity going on for a long time and that activity's heating up and we are going to get some clarity on things over the next few weeks. I'm not going to go into detail, but by the time we get to the next call, we should have a very clear path on the non-adjunctive claim to present to you guys.

Q: Would that have a shut-off feature at this point? Is that kind of the threshold?

A: I'm not going to answer that. We consider shut-off features all the way back to Seven Plus, so we'll take that into consideration and give more color on Gen 6 as it gets closer to market.

Q: Can you give us more details on seeing less seasonality potentially this quarter than in prior years?

A: I think what Steve said is our new patient pipeline is more robust, more robust than it's ever been. With respect to seasonality and patient reorders, it's very much the same as it's always been. And it will be sequentially down, as we said at the conference early January. It's approximately about 15% from what it's been and that's very consistent with what it's been in other years. So while we have more patients in the pipeline, that growth is necessary to have the quarter that we expect to have.

Q: The part that feels like it's a bit different is the new patient flow? Is that what you're saying?

A: We're saying that's where our big increase is coming in the first quarter, up to this point in time.

A: And remember 70% of our business is consumables, and patients largely reorder those. So, those patients who stocked up in Q4 typically aren't reordering yet.                 

Q: And then for G6 – historically, what do IDE conversations with the FDA look like? And what's the follow-up period on them?

A: Our IDE submissions typically get approved very quickly, because we've had a number of discussions with the agency before we submit them, so they're seeing something that they're expecting. As the IDE stands now, this will be the most rigorous study we've ever attempted by a long ways. And so while the FDA is very cooperative as far as getting things through and helping us, they're not making it easier to get CGMs approved. It'll be filed very soon and we hope to start in the second quarter.

Q: There was obviously some discussion this quarter about the pathway to CMS reimbursement. Can you comment?

A: We need this non-adjunctive claim to get to CMS. That's really our first step on the administrative side. And the FDA's been very helpful. It's been a very interactive discussion. We'll have a lot of color on that next quarter because a lot of things are coming to a head here for our agency discussions.

Q: What’s your new patient mix in terms of pumps vs. MDI patients? And then I think there's a perception out there that you've been stealing patients from Medtronic for some time and maybe if they're able to get a better product out, that might slow you down. I wonder if you would comment on that dynamic?

A: Our mix has historically run somewhere in the neighborhood of 60% to 65% pumpers, and the rest are MDI patients. I would say we've started to see a slight shift in favor of MDI patients. And I would also tell you that if you asked the same question in a couple years from now, the hope and the plan is that that balance would shift, that you'd probably see something more like 60% MDI patients and 40% pumpers, frankly, more reflective on what the U.S. base would look like.

As for stealing Medtronic patients, I'd say that couldn't be further from the truth. The vast majority of our patients are generally new to CGM therapy. We do have a handful of patients that choose to wear a Medtronic pump, but also wear a Dexcom sensor. Of the 60% to 65% of our base who are also insulin pumpers, the vast majority of those are wearing an Animas, an Insulet or a Tandem pump, not a Medtronic pump. And the short answer is no, we're not in any way stealing patients from Medtronic as a big growth driver for the company.

A: We get a few.

Q: As you guys have spent more time with the Google team, or the Verily team, could you share the learnings on the type of device that might be appropriate for the Type 2 or broader market? Has your perception of that market opportunity evolved since you started that collaboration?

A: Our perception continues to evolve, and we're doing a lot of market research. The one thing that we've learned about this market is it's not going to do us any good just to display a number and say, here you go. We're really going to have to really give patients the ability to learn from the system that we provide and provide interactive suggestions like, okay, you exercised today. Look how much better you did. And that's the type of system interface that we're going to develop to address that market, again, which is different than what we have now. Just flashing a number on a screen isn't going to be enough, particularly for those that aren't using insulin. I would maintain that intensive insulin-using type 2 patients can have the exact same experience as our type 1 patients have today.

With Google in particular, or Verily, their fresh approach to things, coming from a non-medical device industry, has been refreshing for us and caused us to think very differently. Conversely, there are some things about medical devices that we've taught them, particularly with respect to PMA filings, that they weren't aware of either. It has really been a wonderful partnership, and we think it's going to produce amazing results over time.

Q: It sounds like the early stages of the Verily partnership are going well. Could you remind us what specific milestones are expected there? I think two to three years for the first product. Is it still fair to expect that we won't see any impact or any contributions from that partnership on the current pipeline in terms of the G6 or the receiver?

A: I think the first pass product won't really affect the G6 sensor because it's going to start the trials in the second quarter. Over time, Google electronics could very much be incorporated into a future transmitter offering if that improves technological performance and reduces costs, so we could conceivably use their electronics in some future configuration in G6 in addition to these specific Google products. As far as contributing to our earnings or our sales, certainly nothing in 2016, and more than likely nothing in 2017. I think that really starts off in 2018 when we launch our first product.

And the milestones are really two big ones. There's our first generation product, which we're working on, and our second generation product which we are also working on with Verily. And the second generation product, we've shown pictures of; that's turning the CGM transmitter literally into a Band-Aid size electronics configuration that a patient can peel off and throw away. That will  be very cost effective and that's the end game here. That's what we're shooting for everyday because we know that's what patients want.

Q: On the launch of the receiver this year, can you just remind us what the changes are with the new generation? It sounds like the timeline for the insertion product may be a little later than expected. Is it fair to think of it as a back half of the year event?

A: The insertion system will definitely be a back of the year event, and it's important that we explain a little bit. When we switch to the new insertion system, we're changing everything we do. We've used the current insertion system since we started way back when and that's been our insertion system since our first product was launched. So when we changed to the new insertion system we don't take this lightly. We're going to swap out every mold. We're going to have to change our manufacturing assembly processes and everything. This is the biggest change operation we've ever undertaken. So it will be a back half of the year event, the insertion system, and more than likely it will be a phased launch as we do that but our feedback has been remarkable on the system because it's basically peel the tape, put it on your skin, push a button and you're done. Patients really, really love that.

With respect to the new receiver, it will be a touchscreen device that is still a Dexcom configured receiver and it will have a patient interface much more similar to the G5 Mobile app. It's designed to have a better patient experience and interface. It's also designed to be more durable with respect to patient wear and dropping receivers.

Q: Can you talk about what you're seeing in terms of the G5 launch? We're four to five months into it here. I'm curious, are you're seeing more interest from Type 2s, or are you seeing more compliant utilization, seeing a lower attrition rate? Just anything would be helpful.

A: Obviously the biggest benefit is connectivity directly to a phone. The patients love not having to carry a separate receiver and have the convenience of having it on their phone. That together with the Share platform enables continued ability to share data with caregivers. I would say we continue to see success in driving awareness at the patient level, more so than at the clinician level. We’ve said in prior quarters that the biggest driver for our growth has been patients referring to other patients, parents of patients and caregivers or loved ones. We're continuing to see that and I think our DTC campaigns are really adding to that push.

We’ve learned a lot about going directly to a consumer device like a mobile phone. And there are some challenges there and we've got some things that we're going to work on over the next several quarters, but by and large, the reception has been remarkable.

A: With respect to type 2 patients, it’s as much a function of reimbursement coming in as it is going to the phone. But as far as appealing to patients, I would tell you we are appealing to a broader base of patients by going to a phone. I think we're appealing to a broader base of patients.

Q:  I realize you're not going to quantify it, but are you seeing a lower attrition rate with this device?

A: It's absolutely too early. The product hasn't been out that long, and as we do in every fourth quarter, shift so much the last month of that quarter. We'll certainly have the tools to give a better indication of utilization and patient retention as all this data is coming to our servers. And we are starting to build the analytics platforms to really go after and analyze this and to build the business cases to understand our business much better with all the data coming to the servers. I think we'll provide more clarity in the future.

Q: Do you think the pharmacy avenue has stimulated more demand for the device, or is the real benefit still largely a convenience factor for those existing users?

A: I don't think we have broad enough coverage to stimulate a bunch of benefit in a pharmacy channel. And I think in most cases what happens is we'll approach healthcare professionals and say, we have great coverage under these plans where there's pharmacy benefit. Do you have a lot of patients who have pharmacy benefit who could use this? And we'll get few patients who do it, but by and large, again, a lot of our demand is coming directly from patients. They see an ad, they're not seeing you can get it at your pharmacy; they're seeing what this system does for you and your life, and how it enables you to better manage your diabetes.   

Q: I wanted to ask you guys about CMS and how to think about the size of that patient population that could open up if you do get coverage there.

A: The estimates that we've seen range from 15% of the type 1s in the US to 25%. So there you're looking at hundreds of thousands of type 1 patients who are in the Medicare-age population. We think the other big opportunity with Medicare over time is the intensively managed type 2s. Because as type 2s progress in their disease state, they move from starting out trying to manage with diet and exercise to the point where they're taking both basal and bolus insulin. At that point in time, they become perfect candidates for CGM.

The other fact we know is that, over time, people who have had type 1 diabetes for a long number of years develop a condition called hypoglycemia unawareness. They actually lose the ability to detect an oncoming hypoglycemic episode. People who have lived with type 1 and are now entering the Medicare age are the same folks that may be having hypoglycemia unawareness. We think there's a great opportunity for those patients as well. It is an exciting potential catalyst for us.

Q: Could you give any insight on how to think about the cadence of sequential products coming out of the Verily collaboration? And what will the first go at a product look like? What do you ultimately hope to get to?

A: We’ve always said that we've agreed to do a minimum of two products with Google. We've always said that the first product is going to leverage much of the technology that we already have today, some of the Gen 6 sensor technology, some of the more advanced algorithms that we have, some of the initial prototype miniaturized electronics and batteries and things that Google brought to the table when we first started talking about working together. It's really that second generation product that we've shown in some of the slide presentations that begins to look more like a tiny bandage on the skin. Certainly we'll continue to pierce the skin, but looking at taking costs out of the system and making it truly minimally invasive for patients so that we've eliminated all the objections to wearing it. That doesn't come until the second product, which is probably four or five years away.

Q: Historically, you've cited patient and provider awareness of CGM as one of the biggest hurdles to broader adoption. Can you give us an update on where that stands today relative to 12 months ago? How should we think about the recent ACE guidelines you mentioned, your DTC campaign and other initiatives? How will that impact 2016?

A: Obviously, we'll certainly use the recent ACE guidelines in our campaigns. Nothing could be more compelling than saying the physician group who takes care of you recommends that most of you should use CGM. So we will absolutely use them. We're thrilled with those guidelines. With respect to patient awareness, I think it's increased dramatically. Just look at the new patient adds we had this year. As we said in our prepared remarks, we pretty much doubled the side of our patient base in 2015. It took us from 2005 to 2014 to get to one point, and we almost doubled that in a year. Awareness is rising rapidly in the patient community and the physician community, because you can't bring the patients in without healthcare provider support. I think our efforts to increase awareness have worked very well across the board, and the reason that increasing awareness has worked is because the product does. It really delivers what we promised with respect to accuracy and performance and it changes and saves lives. So you have a case where a product has met expectations and people are willing to sign up for another experience.

Q: Can you give us any update on what progress has been made in the reimbursement landscape for intensively managed type 2s in the US?

A: It's still pretty limited. We have some plans that have agreed to cover the type 2 intensively managed patients. We need more data.

A: What came out of the AACE consensus is that we definitely see benefit there, but we need more data.

Q:  Can you provide more detail on the development of reimbursement in Europe? Are there two or three countries that you'd highlight that are closest to getting reimbursement? And in those countries, what's the, the process or timing for getting reimbursement and the milestones that we should be looking for?

A: We have reimbursement today in Sweden and the Netherlands. And honestly Sweden is one of our largest countries, which it probably shouldn't be considering the population. There's only 6 million or 8 million people in the country. The three key ones for this year to look for will be Germany, the UK and France. Whether we get all three is still an unknown, but I would expect to see at least one, if not potentially a couple of them. Those are pretty sizeable markets. That's what's driving this need to make some addition investment. We’d feel pretty stupid if we got reimbursement in Germany, France, and the UK and didn't have the resources to be able to support new patient additions there, so we're taking a look at opening up headquarters and then beefing up some additional resources in Europe

Q: You’d be disappointed then if in at least one of those three countries CGM was not paid for by the end of the year?

A: Yes. We’d like to see at least one of the three.

Q: Is there anything that you're doing today from a clinical trial perspective, or a technology perspective, to support use in insulin dependent type 2s?

A: It’s going to take a large clinical study. We have some studies that are being run where there are some type 2 patients, and they are subsets of populations in many of our studies. We're going to have to run some bigger studies to develop a better data set, and show the payers. Another thing that we've proposed is to go to payers directly and offer to show cost effectiveness in this population. Those programs haven't been developed. A lot of them are in the idea stage, and we'll crystallize those to be a little more aggressive throughout the year.

Q: Is there some way to think about how much you've invested in DTC to date, and how you view the returns on that investment, and whether you're going to be scaling it up, or waiting and seeing?

A: We're certainly going to run it through the first few quarters of this year. Then we will go back and analyze the returns on that. We obviously know how much we've spent. We also know how much we've increased on the leads that we've generated through our visual media, and the increase has been absolutely wonderful. We're very pleased with that. At the end of about a nine-month period, we'll step back and analyze and decide where we go from here. So far, the results have been great.

Q: On the OUS reimbursement, would you expect stocking orders ahead in anticipation of reimbursement approval in Germany, UK, France?

A: No. We don't intend to do that.

Q: Looking ahead to Android, can you walk us through the timing on that? How significant is a G5 Android launch?

A: Android is certainly important. The question is once we file, how long does the FDA take? Even with our anticipated filing timeline, it's certainly possible that it will be a 2016 launch. We've talked long and hard about the pricing of CGM and where we think pricing needs to go over the long haul. We find that the vast majority of our users today, and the vast majority of hits we get on our website and other ways, are from iOS users. So while we know Android is important, we know that iOS is, and will remain, to be much more important in terms of a platform. It's also a much more stable platform. It’s certainly easier for us to deal with that platform out of the gate before we go to Android.

A: With that being said, we will have to pick some phones to talk to. We cannot possibly talk to all of them. And I think when we offer the Android platform, it will be a tremendous benefit to our patients. I have received a few unhappy letters from patients saying I would love to use your system, but I'm not ever going to do it until you get me an Android phone.

Q: Are you working with Android manufacturers at this point? Or is that just something that you will decide later on down the road?

A: We've certainly worked and talked with several Android manufacturers to learn the intricate parts of those phones. I can tell you they are all not the same, even though they all run the Android operating system. We have learned that their interaction with the Bluetooth interface may differ a bit from phone to phone. And that is one of the reasons we've taken so long to get this out. We want to make sure it works consistently across the board over those platforms that we choose to support.

Q: What sort of ramp would you expect once you do get approval in a country like Germany or France or the UK? Seeing that there's a lot of demand there, how would you expect that to progress?

A: We're not going to speculate on that at this time. It still remains to be seen how broad the reimbursement will be. What's the pricing going to look like? There's a whole bunch of factors that we are still not fully aware of. So we are not going to speculate at this point.

A: I’ll go back to what we said back in the Pediatrics days. Everybody said, what kind of business are we going to see for Pediatrics? Obviously, we've gotten the business because we have been growing about our projected rate since 35% to 45%. But in all reality, you've got to get things like this to continue to grow a business the way that we are. So it falls into line and we expect it, we plan for it, and we'll deal with it.

Q: Obviously you need things like this to continue the growth rate you have. Does your current guidance as it stands now…

A: We're not going to comment on our guidance. Our guidance is $540 million to $565 million.

A: And we'll stick with that.

Q: I think a big part of the AACE consensus statement was better data on economic benefits of CGM. I think the DIaMonD study that you're working on will provide some nice data there. Is there anything else that you're working on that might address those data requests?

A: The DIaMonD study is our biggest data request that we have coming up. We should have data from that study available in the second half of 2016 to help drive this message. There's also a very large study in Sweden going on using CGM, and we should have very good data on those patients as well. We’ll have some CGM cost data with those two studies. The beauty of our product is that we know that we win. We absolutely know that you're going to save money if you use CGM. Just avoiding one hypoglycemic event where you're hospitalized saves the system $15,000 to $20,000, and for a number of our patients those events happen frequently. Obviously, the A1C reduction and reduction of complications is tougher to measure because it takes a longer period of time. But all those things happen. We're very confident we can build a very good cost-effective case with CGM.

Q: Any evolution in the way you're thinking about CGM? It seems like coming at cost data is suggesting CGM cost savings is bigger, even when it's just paired with MDI over pumping.

A: Both CGM and insulin pumps add cost to the system. Anything you add for treatment adds costs. What we believe we can show, in the DIaMonD study, is those patients who have multiple daily injections can achieve very good results with their multiple daily injections using CGM. Then we go to the additional benefit and the additional cost savings if we add new insulin-delivering methodology to the system. We’re all going to be fighting for dollars over time in reimbursement. We want to be the first choice. We want to position CGM to be the first thing physicians pick and the payers pick to hopefully manage people who are using intensive insulin therapy. Because we believe it's the most important that they know their glucose values. And then whatever they choose to do after that certainly will be a function of payers, clinicians and everybody else. But it's our job to position CGM first, and we believe it is the most important tool.

Q: Can you comment on the gross margin guidance for 2016? Obviously, the revenue being lower in the first half versus the second half will influence that a little bit. Should we think about gross margins ramping throughout the year? Is that the best way to set up our models here as we go forward?

A: Yes, that’s fair. Q1 is just going to be sequentially down because of lower revenue, that's the biggest gating factor in Q1. Then, as we ramp Gen 5 over the year you're going to see a little bit of pressure on the overall gross margin because for G5, we're shipping two transmitters, effectively for the price of one, versus G4. Then there will be some costs that roll into COGS with respect to the new expanded manufacturing facility that could have a little bit of an impact. We’ve guided 60% to 70% – it's not a meaningful impact.

Q: You think quarter-by-quarter we should see plus or minus?

A: With the calendar out, I think Q1 will be a little down.

-- by Adam Brown, Ava Runge, and Kelly Close