Dexcom 4Q16 – Record-high sales of $171 million grow 31%; 200,000 worldwide users; Medicare discussions ongoing; four submissions at FDA; G6 in 2018 – February 28, 2016

Following the pre-release at JPM, Dexcom reported very strong 4Q16 financial results today in a call led by CEO Kevin Sayer. We enclose the top business and pipeline updates below, followed by a pipeline summary.

Financial Highlights

1. Dexcom’s 4Q16 worldwide revenue reached a record-high of $171 million, an impressive 31% year-over-year (YOY) gain on a very tough comparison to 4Q15 (55% growth when G5 mobile launched). Sales grew 15% sequentially from the previous record high in 3Q16 ($149 million). This marks record-high sales in seven out of the past nine quarters, though for the third straight quarter, the lowest YOY quarterly growth since 3Q12 – a clear reflection of the large sales base and the difficult YOY comparisons. 2016 revenue grew 43% to reach $573 million, over half a billion for the first time and on the high-end of guidance ($550-$575 million).

2. US sales totaled $150 million, rising 31% YOY and 15% sequentially. International sales came in at $21 million, rising a comparable 28% YOY and 15% sequentially. The US business contributed 87% of the quarter’s and year’s growth, holding steady over time.

3. Mr. Sayer reiterated 2017 revenue guidance from JPM: sales of $710-$740 million rising 25%-30% YOY. In late Q4 and early Q1, Dexcom expanded its US sales force by roughly 20%, adding ~20 reps and bringing the total US field force to 130 reps.

4. Dexcom’s worldwide installed base is now ~200,000 users, a ~40% rise from ~140,000-150,000 shared at JPM one year ago and reflecting an estimated 80,000-90,000 new patients added in 2016 – we assume that means ~20,000 patients stopped using Dexcom during the year, though this was not addressed today or at JPM. Management clarified in Q&A that attrition is roughly 8%-12% per year. The base is 75%-80% in the US (~150,000) and 20%-25% outside the US (~50,000), expected to continue in 2017. The worldwide patient base is expected to grow to~270,000 by the end of 2017 (35% growth, outpacing revenue). As he has in the past couple of quarters, Mr. Sayer reminded attendees that the “majority” of Dexcom’s new patient adds are now from MDI,

Reimbursement and Marketing Highlights

5. As expected, a major focus of prepared remarks was December’s “landmark” FDA approval of a non-adjunctive (insulin dosing) label claim for G5, and CMS’ subsequent benefit category determination last month (DME under Part B). Dexcom is still in early negotiations with the four Medicare Administrative Contractors (MACs) under Part B, but there is no firm timeline on coverage. CEO Kevin Sayer hopes this could be resolved up by “mid-year,” but management was mostly vague about where things stand and what the remaining administrative hoops are. We assume this earlier-than-expected CMS ruling puts coverage on the earlier side of previous “2018” expectations, perhaps even coming in mid/late 2017.

6. Dexcom has rolled out marketing and training materials for the non-adjunctive (insulin-dosing) claim. The marketing is very strong and definitely maximizes the no confirmatory fingerstick part of the label, even though two are needed per day for calibration (see below). Training requirements were previously an unknown, though Dexcom has now posted some pithy, useful tips at www.dexcom.com/fingersticks. We’re glad to see no firm recommendations on insulin-dosing adjustments (e.g., +50% for two arrows up), since patients are all so different.

7. Management also highlighted two very big evidence wins from last month: publication of the DIaMonD and GOLD studies in JAMA. Both randomized trials used Dexcom’s G4 in type 1 adults on MDI, and both found similar A1c advantages with CGM over SMBG (0.4%-0.6%). The call also briefly alluded to the additional DIaMonD data presented at ATTD (type 2, pump).

Pipeline Highlights

8. The G6 pivotal trial is underway and an FDA PMA filing is expected by the end of 3Q17. Launch is still expected in 2018. G6 will initially be one calibration per day and 10-day wear, but ultimately move to no calibration and 14-day wear with the Verily gen one product.

9. The JPM timing for the Verily partnership was maintained: a first-gen launch in late 2018 and a second-gen launch in 2020-2021. Both Verily products are expected to be 14 day wear, factory calibrated, real-time CGM, fully disposable, and talk to a phone via Bluetooth (no receiver expected). Dexcom and Verily continue to conduct human pilot studies with the first gen sensor, and an initial feasibility study has now been completed for the second-gen product. The Verily products, targeted at type 2s, will require different strategies for commercialization (e.g., pharmacy distribution) and payer approaches (e.g., intermittent CGM).

10. Dexcom’s one-touch inserter and lower profile transmitter (G5x) remain under FDA review after a 3Q16 submission. Questions just came in from the FDA (no specifics offered) and the team will soon submit responses. No approval timeline was shared, but JPM positioned this as a “2017” launch. We assume it will come later in the year, given the manufacturing complexity

11. The Android version of G5 remains under FDA review following the 3Q16 submission. Management hopes to launch Android G5 by mid-year in the US, a delay from the previous expectations (“late 2016 or early 2017”). Android G5 launched in the UK, Ireland, Netherlands, Germany, and South Africa last month.

12. The new touchscreen receiver remains under FDA review following the 2Q16 submission. No approval timeline was shared, but JPM positioned this as a “2017” launch.

13. An “enhanced” G5 app with new features, including incorporation of insulin data, is now under FDA review and presumably also on deck to launch this year.

14. There were no insulin delivery partnership updates. Management mentioned the G4 integrated pump offerings are “well liked,” but did not mention Tandem or Animas by name.

15. However, management was very optimistic on smart pens, calling the opportunity “huge...As we look to the next 18-24 months, some exciting products are coming to market. Our sensors are driving work on automated insulin delivery, and the value in smart pens will be integrating insulin-on-board information together with CGM data in single unified app on a phone. We can do some powerful stuff with that. Demonstrating outcomes with a smart pen, CGM data, dosing support, and behavior modification – really at a fraction of the cost of more complex systems. I think we really have a home run there.”

Financial Highlights

1. Dexcom’s 4Q16 worldwide revenue reached a record-high of $171 million, an impressive 31% year-over-year (YOY) gain on a very tough comparison to 4Q15 (55% growth when G5 mobile launched). This also reflects a solid 15% sequential gain from the previous record high in 3Q16 ($149 million). Impressively, Dexcom has now seen record-high sales in seven out of the past nine quarters, though for the third straight quarter, saw its lowest YOY quarterly growth since 3Q12 – a clear reflection of the large sales base and the very difficult year-over-year comparisons. CEO Kevin Sayer put the growth in perspective very well: Dexcom’s 4Q16 revenue alone was larger than full-year 2013 sales. Wow! (At $171 million, sales also came in slightly over the JPM pre-release of $168 million.)

• Full-year 2016 revenue grew an impressive 43% to reach $573 million, crossing half a billion for the first time and on the upper end of guidance for $550-$575 million. Boy does this company execute. The full-year sales also came on a very difficult comparison to 55% YOY growth in 2015.
• Dexcom’s cash-based net income was $27 million in 4Q16, down slightly from $29 million in 4Q15, but up significantly from $15 million in 3Q16. Gross margin declined to 68% in 4Q16, a slight drop from 70% in 4Q15 – this reflects lower margins associated with the G5 mobile transmitter (three-month life, lower selling price) and continued warranty expenses associated with the G4 receiver recall. In line with JPM comments, gross margin is expected to hold steady in 2017 at 67%-70%. Pricing remains steady at $70-$75 per sensor – this is a threat for Dexcom, depending on what Abbott does with FreeStyle Libre in the US (launched expected in 2H17).

2. US sales totaled $150 million, rising 31% YOY and 15% sequentially. International sales came in at $21 million, rising a comparable 28% YOY and 15% sequentially. The US business contributed 87% of the quarter’s and year’s growth, holding steady over time. We’ll be interested to see if the mix changes this year as more competitive sensors from Abbott and Medtronic come to the US, alongside expanded reimbursement in Germany.

• Management emphasized that the international business is keeping pace with robust US growth, and uptake has been particularly encouraging in Germany. Dexcom signed initial German payer contracts this quarter following the positive national reimbursement win for CGM last summer. Remarks highlighted that real-time CGM is covered in Germany for type 1s and type 2s on insulin, but only including devices with alerts/alarms (a subtle competitive knock against Libre, though we believe Libre has regional reimbursement). At JPM, management was a bit more specific about Germany, noting that sales tripled YOY in December, and the German business will be a “major initiative” in 2017.

3. Mr. Sayer reiterated 2017 revenue guidance from JPM: sales of $710-$740 million rising 25%-30% YOY – obviously revenue growth will naturally slow as sales rise, plus Medtronic and Abbott will/might launch competitive products in the US this year (Medtronic’s Guardian Sensor 3 in the 670G and Guardian Connect; Abbott’s FreeStyle Libre consumer version). As usual, management emphasized that 1Q17 will be sequentially down from 4Q16, and 1Q17 has an especially tough comparison to 1Q16’s 60% YOY growth.

• In late Q4 and early Q1, Dexcom expanded its US sales force by roughly 20%, adding ~20 reps and bringing the total US field force to 130 reps. This should bring some nice tailwind this year. Notably, Dexcom did not add any sales reps in 2016, meaning this could provide a nice boost for YOY comparisons.

4. Dexcom’s worldwide installed base is now ~200,000 users, a ~40% rise from ~140,000-150,000 shared at JPM one year ago and reflecting an estimated 80,000-90,000 new patients added in 2016 – we assume that means ~20,000 patients stopped using Dexcom during the year, though this was not addressed today or at JPM. Management clarified in Q&A that attrition is roughly 8%-12% per year. The base is 75%-80% in the US (~150,000) and 20%-25% outside the US (~50,000), expected to continue in 2017. Per JPM, Dexcom estimated it grew US CGM penetration in type 1 by four percentage points in 2016, an impressive achievement.

• The worldwide patient base is expected to grow to~270,000 by the end of 2017 (35% growth, outpacing revenue). As he has in the past couple of quarters, Mr. Sayer reminded attendees that the “majority” of Dexcom’s new patient adds are now from MDI, flipping the company’s historical focus on pumpers – and a good sign it is hitting a broader population of type 1s. Mr. Sayer did not say “60%” are from MDIs, but did say this at JPM and in 2Q16.

Reimbursement and Marketing Highlights

5. As expected, a major focus of prepared remarks was December’s “landmark” FDA approval of a non-adjunctive (insulin dosing) label claim for G5, and CMS’ subsequent benefit category determination last month (DME under Part B). Dexcom is still in early negotiations with the four Medicare Administrative Contractors (MACs) under Part B, but there is no firm timeline on coverage. CEO Kevin Sayer hopes this could be resolved up by “mid-year,” but management was mostly vague about where things stand and what the remaining administrative hoops are. Perhaps the biggest question is whether type 2s will be covered, as the CMS ruling does not specify type 1s alone. We assume this earlier-than-expected CMS ruling puts coverage on the earlier side of previous “2018” expectations, perhaps even coming in mid/late 2017.

• As we shared last month, the pricing in the CMS ruling letter looks pretty favorable to Dexcom: a durable receiver will be reimbursed at $236-$277, with a monthly price of $248.38 for sensors/transmitters ($2,981/year). We see this as highly positive pricing for Dexcom, particularly given how Medicare has destroyed pricing in BGM. It seems like patients won’t be limited on their monthly sensors, and we assume these levels will make this more than a sustainable business for Dexcom – especially with the move to G6 requiring only three sensors per month (and eventually two per month with eventual 14-day wear). One investor asked about Dexcom providing a blood glucose meter/strips to Medicare beneficiaries, and EVP Steve Pacelli said it will probably contract with a distributor to do so.  
• Noted Mr. Sayer, “Dexcom G5 mobile is the only glucose measurement device that is FDA approved for therapeutic decision making.” He further noted this approval brings a new category of “therapeutic CGM” and a “paradigm shift in diabetes management.” As a reminder, Abbott’s FreeStyle Libre is under FDA review for both adjunctive AND non-adjunctive label claims (submitted in 3Q16, with a US launch expected in 2H17). Medtronic’s Guardian Sensor 3 is approved for adjunctive use in the MiniMed 670G, and Senseonics is also going for an adjunctive claim initially. In short, Dexcom has a distinct marketing advantage here for the time being.
• Dexcom estimates Medicare-aged patients are ~20% of the overall type 1 market, translating to as many as ~300,000 patients (assuming 1.5 million US type 1s). In running analytics on the patient base, Dexcom’s most loyal group is actually between 50 and 65 years old – “They have not left us. They are very loyal.” This could bode well for strong utilization among Medicare beneficiaries.

6. Dexcom has rolled out marketing and training materials for the non-adjunctive (insulin-dosing) claim. The marketing is very strong and definitely maximizes the no confirmatory fingerstick part of the label, even though two are needed per day for calibration (see below). This is taking a page out of Abbott’s FreeStyle Libre marketing book in Europe and should help drive CGM adoption in reluctant prescribers and patients. Training requirements were previously an unknown, though Dexcom has now posted some pithy, useful tips at www.dexcom.com/fingersticks. We’re glad to see no firm recommendations on insulin-dosing adjustments (e.g., +50% for two arrows up), since patients are all so different. Still, this will put more burden on HCPs to understand how to use safely use CGM for insulin dosing, or more pressure on patients to discover through trial and error how to do so.

7. Management also highlighted two very big evidence wins from last month: publication of the DIaMonD and GOLD studies in JAMA. Both randomized trials used Dexcom’s G4 in type 1 adults on MDI, and both found similar A1c advantages with CGM over SMBG (0.4%-0.6%). Secondary endpoints were also consistent, including significantly less time in hypo/hyperglycemia and more time-in-range. Management noted Dexcom is “already seeing the benefit” of these studies, as the “majority” of new patient additions are MDI. The prospective RCTs in such a major journal are expected to accelerate awareness with physicians, patients, and payers.

• The call also briefly alluded to the additional DIaMonD data presented at ATTD: type 2 MDI results and data from type 1 MDIs on CGM that crossed over to add Insulet’s OmniPod. Summarized Mr. Sayer, “CGM can provide overwhelming benefit regardless of insulin delivery method... Overall, these data continue to build support for our ‘CGM first’ message...the most significant addition to a patient intensively managing their diabetes is CGM...” Management is threading a very difficult needle here – Dexcom has many pump partners! – but the clear argument was that adding CGM is more important than adding a pump, at least at first. Indeed, management had some very enthusiastic comments on smart pens in Q&A (see below), but shared nothing specific on its pump partners. We’ll be interested to see how these comments evolve as Dexcom’s pump partners come to market with automated systems.
  
  • Type 2 data: Dr. Rich Bergenstal presented the first type 2 data from Dexcom’s DIaMonD study testing CGM in MDIs, demonstrating a statistically significant 0.3% A1c advantage for CGM (n=79) over SMBG (n=79) at 24 weeks: -0.8% with CGM vs. -0.5% with SMBG, both from a baseline of 8.5% (p=0.02). The benefit of CGM rose with a higher baseline A1c – those starting at >9.0% saw a 1.4% reduction. At 24 weeks, these type 2 CGM users were spending ~48 more minutes per day in 70-180 mg/dl (a 6% improvement from baseline), while the SMBG users spent 9 fewer minutes in range per day (a 1% decline) (p=0.01). There was no significant difference in hypoglycemia (very low in both groups), meaning the improvement came from spending less time >180 mg/dl. Similar to the type 1 data (ADA 2016, JAMA 2017), CGM adherence was very strong: 93% of the type 2 cohort was using CGM >6 days per week at six months. Patients in this cohort were very typical of the type 2 population, with a mean age of 60 years, a median type 2 diabetes duration of 17 years, a mean of three fingersticks per day, and a mean BMI of 36 kg/m². We’re glad to see this technology being tested in a broader, real-world population. Interestingly, Dr. Bergenstal noted that medications didn’t really change throughout the study, so the impact of CGM was on lifestyle and behavior. Skeptics might argue these results are underwhelming – a 0.3% A1c advantage from a high baseline – but this is a tough population and the study really worked to minimize clinical encounters. Dr. Bergenstal emphasized that there was “not a lot of hand holding” in this study (a point also shared at ADA), and he is eager to explore more coaching and giving patients more advice on adjusting insulin. We certainly agree, just as we noted with Abbott’s REPLACE study in type 2 diabetes at ATTD last year.
  • Pump crossover data: a poster shared the second phase of the DIaMonD study in type 1 diabetes, where patients on the G4 CGM were further randomized to continue on MDI (n=38) or switch to an OmniPod (n=37). The OmniPod group won handily on the primary endpoint of time spent in 70-180 mg/dl (weeks 5-28 pooled): an improvement of 78 minutes per day from baseline vs. a drop of 17 minutes per day for the MDI group (p=0.02). Time in hyperglycemia (>180) was also highly in favor of the OmniPod group: an improvement of -47 minutes per day vs. +59 minutes per day when continuing on MDI (p=0.009). On the other hand, time in hypoglycemia (<70 mg/dl) actually favored the MDI group, who spent 9 fewer minutes per day low compared to 15 more minutes per day in the OmniPod group (p<0.001). Surprisingly, the difference in A1c was not statistically significant: the pump group saw a +0.3% change in A1c vs. +0.1% in the MDI group (baseline: 7.6%; p=0.32). The poster emphasizes that the A1c results from the six-month phase 1 of DIaMonD were sustained out to one year in this extension, a very positive finding indeed. As expected, bolus frequency increased in the pump group by +0.6/day vs. -0.1/day in MDI, which the authors tie to the increase in hypoglycemia in the pump group – it’s an aside in the poster’s conclusion, but a definite possibility (since pumps makes bolusing so easy, stacking insulin is also easy, particularly when on CGM). CGM adherence remained excellent in the study, with 96% using it >6 days per week at six months. We see these data as a win for CGM (sustained positive outcomes at one year) and encouraging evidence that adding a pump on top of MDI+CGM brings at least some further value (over an hour more per day in range). Automated insulin delivery could be a killer app for both technologies, though it’s hard to say who will benefit the most and whether the ROI will be there vs. MDI+CGM and some insulin dosing decision support.

Pipeline Highlights

8. The G6 pivotal trial is making “good progress,” with a PMA filing expected by the end of 3Q17 and a launch still expected in 2018. Management highlighted the G6 pre-pivotal data from DTM (MARD of 8.1% with one calibration per day; 8.8% with zero calibrations), and reiterated that G6 will initially be one calibration per day and 10-day wear, but ultimately move to no calibration and 14-day wear with the Verily gen one product (see below). Said Mr. Sayer, “We are increasingly confident we will ultimately deliver a no calibration sensor with market-leading performance.” The G6 sensor will be filed before the Verily gen-one product (the trial will be done earlier), enabling the initial Verily product to reference the G6 filing (especially manufacturing).  

9. Management reiterated the same Verily partnership timing we heard at JPM: a first-gen launch in late 2018 and a second-gen launch in 2020-2021. Both Verily products are expected to be 14 day wear, factory calibrated, real-time CGM, fully disposable, and talk to a phone via Bluetooth (no receiver expected). Dexcom and Verily continue to conduct human pilot studies with the first gen sensor, and an initial feasibility study has now been completed for the second-gen product. “We are excited about the continued progress,” said Mr. Sayer. The second-gen timing could be read as a delay from the initial plan (“as early as 2020”), but CEO Kevin Sayer confirmed the team is “focused on 2020 and added 2021 to be conservative...A lot is going on in our industry between now and when that launches. We have very aggressive timelines for that system.”

• At JPM, we saw the first-gen Verily product for the first time (see below) – it is still meaningfully smaller on the body than the current transmitter and similarly pear-shaped to the very tiny second-gen version. The big difference is the first-gen version has a slightly larger profile off the body than gen two. We’re not sure how much cheaper these will be to make, but a fully disposable sensor/transmitter will be a huge hassle saver from the current reusable transmitter.

• The first-gen Verily product will exist alongside the traditional G6 sensor and transmitter, and if both products launch next year as planned, we’ll be interested to see how Dexcom handles this. Management suggested that the gen one Verily product might have different labeling than the G6 system, since it will be factory calibrated and 14-day wear vs. 10-day wear and one calibration/day for the initial G6. However, both systems will use the same G6 sensor – a lot of complexity here! Management reiterated that the Verily products are targeted at type 2, though we imagine many type 1s will want the smaller, cheaper, longer-wear, factory calibrated gen one Verily product – how will this be handled? We also wonder what choices will be made on a design level to tailor the product to type 2s.
• Management acknowledged that the Verily products, targeted at type 2s, will require different strategies for commercialization (e.g., pharmacy distribution) and payer approaches (e.g., intermittent CGM). Mr. Sayer was clear that Dexcom needs to think differently and a lot is happening internally to drive new models. We love hearing this!
  
  • “A lot of it will be payer-driven. Type 2 patients don’t spend a lot of money on fingersticks today. A type 1 solution won’t work. Intermittent CGM will be a reality. How much do they spend in a year? How many sensors do patients wear a year? If we have programs, what type of benefit can they provide? In a 2-4 week period, we can see type 2 patients take their estimated A1c down more than a full point. They don’t get feedback from anything like CGM right now. Patients can pull three levers with the help of CGM: medication, exercise, and diet. By getting real time feedback on their phones that is easy to look at, patients can now pull those levers. ‘Maybe I should do something different.’ Somebody I talked to the other day said, ‘This is more significant than any type 2 drug I’ve ever seen.’
  • “In 2020, when we’re selling the Verily Band-Aid product, that has to be sold at the drug store. We cannot handle the demand for this number of patients. If we’re talking about type 2s and 20 million patients, even intermittent use, those products have to be sold in the drug store. The reimbursement channel has to change.”

10. Dexcom’s one-touch inserter and lower profile transmitter (G5x) remain under FDA review after a 3Q16 submission. Questions just came in from the FDA (no specifics offered) and the team will soon submit responses. No approval timeline was shared. As of JPM, this was positioned as a “2017” launch, though we might assume this will come in the later part of the year – given the FDA questions and the complexity of this launch (“we will pretty much change every manufacturing process”). Previously this was expected to launch by the end of 2016, so it has taken much longer than expected. 

11. The Android version of G5 remains under FDA review following the 3Q16 submission. Management hopes to launch Android G5 by mid-year in the US, a delay from the previous expectations (“late 2016 or early 2017”). Android G5 launched in the UK, Ireland, Netherlands, Germany, and South Africa last month, and our first demo at ATTD showed a pretty identical user experience to the Apple iOS version. Could this meaningfully expand G5’s uptake, or is this more of a convenience win for receiver users with Android phones? Android has the dominant market share in the US and globally, so it will be interesting to see if sales accelerate once this launches.

12. The new touchscreen receiver remains under FDA review following the 2Q16 submission. No approval timeline was shared, though this is also behind the original expectation to launch by the end of 2016. This receiver will be more durable (i.e., less susceptible to wear) and offer a user interface in line with the G5 mobile app.

13. An “enhanced” G5 app with new features, including incorporation of insulin data, is now under FDA review. We assume this was submitted as expected by the end of 2016. This will be key as Tandem expects to launch its G5 integration by mid-year (as of the last update) and smart pens begin coming the market (starting with Companion Medical sometime this year in the US).

14. There were no insulin delivery partnership updates. Further details will be shared “as these products approach clinical trials and commercial launch.” Management mentioned the G4 integrated pump offerings are “well liked,” but did not mention Tandem or Animas by name. Meanwhile, Dexcom is making “continued progress with other integrations” (no companies called out), including “smart pens and other connected diabetes management platforms.”

15. “We think the [smart pen] opportunity is huge. As we look to the future, we know ultimately the day will come where this is fought at the payer level, and not just in the US. In the OUS market, pumps are less than 10% penetrated in many markets. We’re very bullish on the opportunity for smart pens. Many companies, both insulin companies and non-insulin companies, have smart pens in development, including a number announced publicly with different algorithms and software developers. As we look to the next 18-24 months, some exciting products are coming to market. Our sensors are driving work on automated insulin delivery, and the value in smart pens will be integrating insulin-on-board information together with CGM data in single unified app on a phone. We can do some powerful stuff with that. Demonstrating outcomes with a smart pen, CGM data, dosing support, and behavior modification – really at a fraction of the cost of more complex systems. I think we really have a home run there.” – EVP Steve Pacelli

Pipeline Summary

Sensor and Software Pipeline

Automated Insulin Delivery Partner Pipeline

--by Adam Brown and Kelly Close