Kevin Ronald Sayer - President, Chief Executive Officer & Director Steven Robert Pacelli - Executive Vice President-Strategy and Corporate Development Jess Roper - Chief Financial Officer & Senior Vice President.

Ben C. Andrew - William Blair & Co. LLC Michael Weinstein - JPMorgan Securities LLC Jayson T. Bedford - Raymond James & Associates, Inc. Doug Schenkel - Cowen & Co. LLC James Francescone - Morgan Stanley & Co. LLC Danielle J. Antalffy - Leerink Partners LLC Jeff D. Johnson - Robert W. Baird & Co., Inc. (Broker) Thomas J. Bakas - Piper Jaffray & Co.

(Broker) Tao L. Levy - Wedbush Securities, Inc. Anthony Petrone - Jefferies LLC Kyle Rose - Canaccord Genuity, Inc. Erik Ross Shoger - Northcoast Research Partners LLC.

Welcome to the DexCom Second Quarter 2016 Earnings Release Conference Call. My name is Sherry and I'll be your operator for today's call. At this time, all participants are in a listen-only mode. Later, we will conduct a question-and-answer session. Please note this conference is being recorded. I would now like to turn the call over to Kevin Sayer. Mr.

Sayer, you may begin..

Thank you, and good afternoon, ladies and gentlemen, and welcome to the second quarter 2016 DexCom earnings call. We'll start off with our Safe Harbor statement from Steve Pacelli.

Steve?.

Thanks, Kevin. Some of the statements that we will make in today's call may constitute forward-looking statements. These statements reflect management's intentions, belief and expectations about future events, strategies, operating plans and performance.

All forward-looking statements included in this presentation are made as of the date hereof based on information currently available to DexCom and are subject to various risks and uncertainties and actual results could differ materially from those anticipated in the forward-looking statements.

The factors that could cause actual results to differ materially from those expressed or implied by any of these forward-looking statements are detailed in DexCom's annual report on Form 10-K, quarterly reports on Form 10-Q and other filings with the Securities and Exchange Commission.

Except as required by law, we assume no obligation to update any such forward-looking statement after the date of this presentation or to confirm these forward-looking statements to actual result.

Additionally, during the call, we will discuss certain financial measures that have not been prepared in accordance with GAAP with respect to our cash-based operating results. The presentation of this additional information should not be considered in isolation or as a substitute for results or superior to results prepared in accordance with GAAP.

Kevin?.

Thank you, Steve. Joining me today are Jess Roper, our Chief Financial Officer, and Steve Pacelli, our Executive Vice President of Strategy and Corporate Development.

Sticking with our usual format, Steve will review our second quarter 2016 financial results and I will follow with our customary operations update and offer some concluding thoughts before opening the line for questions. But before I turn the call over to Steve, I want to take a few minutes to discuss a couple of key recent milestones.

First, I'll comment on the implications of the landmark meeting of the FDA's Clinical Chemistry and Clinical Toxicology Devices Panel that was held just over a week ago.

We have always believed that CGM provides much better information for making an insulin-dosing decision than a single finger stick, particularly as we look at the accuracy, performance and features of our current system with Software 505 algorithm.

After nearly 18 months of productive discussions with the Agency, we made the decision to seek approval to claim that our G5 Mobile System can be indicated to support diabetes treatment decisions without confirmatory finger sticks. We filed a PMA supplement with the FDA for this indication and our submission remains under review.

Earlier this year, the FDA notified us that our PMA supplement will be subject to review by an advisory committee which was conducted on July 21, and concluded with the panel recommending that the DexCom G5 continuous glucose monitoring system is both safe and effective for making diabetes treatment decisions and that the benefit of doing so outweigh the risk.

We are obviously quite pleased with the results of the advisory committee. We've said for many years that we aim to replace finger sticks and this is our first step. I'll talk more about the ramifications of the non-adjunctive claim later.

Another important highlight from the quarter was the release of preliminary data from our DIaMonD study at the American Diabetes Association's annual Scientific Sessions held in June. Previous CGM outcome studies have had a very limited number of subjects using multiple daily injections, or MDI.

DIaMonD is initially comprised entirely of MDI patients and we saw a full 1.0% gross reduction in A1c in the CGM group during the first phase.

Furthermore, all secondary endpoints in this phase supported a benefit of CGM use and benefits were seen in all sub-groups analyzed, most notably a reduction in time spent in hypoglycemia and hyperglycemia in the study group. We remind you that this first data set only included type 1 patients.

The cohort with type 2 diabetes will complete study activities by the end of the third quarter, with analysis and publication to follow.

As we enter into the final phase of the DIaMonD study, a portion of the CGM MDI patients will transition to an insulin pump in an effort to measure the benefit of a more sophisticated and expensive insulin delivery system compared to the benefits of CGM and MDI. That data will not be available until next year. I will now turn the call over to Steve..

Thanks, Kevin. DexCom reported revenue of $137 million for the second quarter of 2016, compared to $93 million for the same quarter in 2015, a $44 million, or 47% increase. Sequentially, revenue for Q2 was up approximately 18% from the prior quarter.

To put this into perspective, we generated nearly the same amount of revenue in the first six months of 2016 as we did in all of 2014. Our second quarter gross profit was $86 million, generating a gross margin of 62%, compared to a gross profit of $66 million and a gross margin of 71% for the same quarter in the prior year.

Our gross margin in Q2 was affected primarily by increased warranty expense as well as a write-down of excess and obsolete inventory of approximately $3.5 million related to our receiver recall. We do not expect any additional substantive excess and obsolete inventory write-downs related to the receiver recall.

Additionally, we anticipate warranty costs as a percentage of sales will begin to decline in Q3 and should normalize before the end of the year.

The increase in warranty costs relates again to our receiver recall and heightened awareness related to this recall and, to a lesser extent, typical new product performance issues related to the launch of the G5 transmitter.

We expect our gross margin to be in the mid-to-upper 60% range for the second half of the year and in the mid-60%s for the full year 2016. In order to address the receiver situation and our increased warranty expenses, we are implementing several corrective actions.

We have received FDA approval for a more robust speaker for the current receiver and are in the process of planning production of this newly configured receiver. We filed our next generation touchscreen receiver with the FDA during Q2 and we believe this receiver will not only be more durable, but will also greatly improve the patient experience.

And during Q2, we submitted for a firmware update on the transmitter and a software update for our mobile app that will enhance the performance and reliability of the G5 Mobile transmitter. Some final thoughts on our revenues and our gross profit.

Our mix between durable and consumable products was within our normal historical range in Q2 at approximately 30% durable and 70% consumable. Sensor pricing remains within an ASP range of $70 to $75 per sensor and can fluctuate subject to payer and distribution mix. The ASP for our hardware has remained stable.

Finally, our international business showed strong year-over-year growth, generating $18 million in revenue during the quarter, up 45% from last year.

Research and development expense totaled $36 million for Q2 of 2016 compared to $24 million in Q2 of 2015, with the increase due primarily to additional payroll related costs and expenses related to work on our near-term product pipeline, focusing primarily upon our next generation insertion system.

We also incurred significant expense related to our advanced product pipeline, particularly expenses associated with our Verily partnership.

Selling, general and administrative expense totaled $69 million in Q2 of 2016 compared to $45 million during the same quarter in 2015, with the increase due primarily to year-over-year increases in head count in our customer support organization as well as a ramp in our patient-focused marketing expenses, higher IT costs and investments in our OUS expansion efforts.

A couple of highlights during the quarter were our acquisition of our distributor, Nintamed, in Germany, Austria and Switzerland and the establishment of a targeted, direct team in the UK. In Germany, we are very pleased with GBA's recent announcement to provide reimbursement for CGM.

And while we were still early in finalizing the details of this coverage, we expect to see a benefit from this milestone beginning in 2017. Let us remind you that GBA defines CGM as a system with alerts, alarms and realtime trends. Systems without those features will not be covered.

We anticipate our investment in our international operations to continue to scale up, particularly as we obtain expanded reimbursement outside the U.S.

Our net loss for the second quarter of 2016 totaled $20 million, which included $31 million in non-cash expenses centered primarily in non-cash share-based compensation expense across all functional areas of our business.

Our net loss was slightly higher than expected, primarily due to the increased warranty expense and inventory write-down we outlined earlier.

We would also like to remind investors that our operating expenses this quarter included increasing spend on the four strategic investments that we outlined at the beginning of the year, namely our Verily relationship, building out our data analytics capabilities, international expansion, and finally our new manufacturing facility in Arizona.

These investments are ramping up to the $40 million of incremental strategic expense that we projected for 2016 at the beginning of the year. Absent non-cash charges, non-GAAP cash-based net income was $11 million for Q2. Our GAAP loss per share for the quarter was $0.24.

With respect to our balance sheet, we ended the second quarter with $116 million in cash and marketable securities, an increase of $10 million over the first quarter. And during the quarter, we replaced our prior credit facility with a new $200 million revolving line of credit with favorable terms.

With respect to our 2016 revenue guidance, we take this opportunity to increase our range to $550 million to $575 million in revenue for the year. With that I'll turn the call back over to Kevin for a business update..

what level of education, training and recommendations moves into the practice of medicine and is outside our product scope? Please understand that these are the most exciting questions we could ever be addressing. The FDA has been very transparent and interactive with us throughout this entire process.

With this potential label, we are redefining glucose measurement and glucose management and let's not forget the importance of this labeling with respect to Medicare coverage. Every day, 10,000 people in the United States turn 65 and we need to get CGM to these patients that so desperately need it.

Non-adjunctive labeling provides us the opportunity to accelerate our efforts for Medicare approval and we will do so at the appropriate time. Turning to other product pipeline developments, we continue to make progress on several enhancements to our G5 Mobile platform.

As Steve mentioned earlier, we filed a new version of the iOS G5 Mobile app and a new transmitter firmware configuration during the second quarter. We will file our Android G5 Mobile app in Q3. The U.S. Android launch is dependent upon the transmitter firmware revision that we filed in Q2, so we expect the U.S.

Android launch to happen in late 2016 or early 2017, but we do plan to launch Android this fall in several key international markets.

We continue to work on several other enhanced versions of our G5 Mobile app to provide for additional features and functionality, including the incorporation of insulin data and several mobile platform retrospective analysis reports. The new G5 insertion system and transmitter will be filed within the next few weeks.

This filing has been somewhat delayed due to the resources that we expended on the recent panel meeting. With respect to Gen 6, our ID application has been approved. We have completed a number of pre-pivotal studies in anticipation of the pivotal trial and the data looks very promising. We will commence a pivotal study as soon as possible.

This will be the largest pivotal study that we have ever executed, with pediatric and adult patients included in one combined study of approximately 300 subjects. Patients will be required to spend three in-clinic days to measure the accuracy of our system similar to our other studies.

We are also working with several of our artificial pancreas partners to incorporate Gen 6 into their research and clinical studies to gather more valuable data on the system. As a reminder, the G6 sensor will have a reduced calibration scheme, one per day after startup driven by a completely new, advanced algorithm platform.

We'll have the ability to block acetaminophen interference, a labeled indication of up to 10 days of sensor life, and a number of other features that will make the CGM experience better for our patients. Overall performance should be consistent with the G5 Mobile System even with the reduced calibration scheme and an extended wear.

The G6 product will also leverage our new applicator and receiver technologies. Turning to our partnership with Verily, our collaboration to develop simple, low cost disposable sensor systems remains on track.

We believe the products we develop in our Verily partnership will drive entry into the non-insulin using type 2 market, but it will also expand CGM use in the type 1 market.

As we continue to do research into this type 2 market, it has become more and more evident that when we talk about finger stick replacements, we are talking about all finger sticks. We believe that the information our system provides to a type 2 patient exceeds anything that they can learn from the tools that they use today.

We believe the first product developed in the collaboration with Verily will be commercialized in 2018 and we expect that a smaller bandage-like sensor could be available as early as 2020. In fact, our first feasibility study of the bandage-like sensor will take place this fall.

With respect to our pump partners, Animas and Tandem continue to report positive results from their integrated pump offerings with G4 PLATINUM and we believe Animas, Tandem and Insulet and each making excellent progress on the development of more advanced systems utilizing our G5 and G6 technologies.

We're also seeing some excellent work conducted by the teams in companies like Bigfoot Biomedical and Beta Bionics who seem to be taking truly novel approaches to the development of automated insulin-delivery systems.

Finally, we continue to make progress beyond our pump partnerships as we explore opportunities and connectivity with smart insulin pens, as well as other interconnected diabetes management platforms. In conclusion, the growth in this business over the past five years has been unprecedented.

And when I boil it down and look at the second quarter, there are really four major takeaways. First, again, we have increased CGM penetration in the U.S. type 1 population by more than 2% in the first six months of this year.

When you combine that with the growth of our patient base this year with the growth of our patient base in the second half of 2015, our U.S. type 1 penetration has increased by 4%.

Second, the DIaMonD study demonstrates what we already knew, that CGM has a major impact for MDI patients representing a very cost-effective diabetes treatment model with excellent outcomes.

Along those lines, while in past years we have told you that 60% of our patient base use insulin pumps, we are pleased to note that of our direct new patient additions during this quarter, greater than 60% of them were MDI patients. Our CGM first message is resonating with patients and healthcare professionals.

Third, the FDA panel's positive recommendations have the potential to permanently change the CGM landscape and we are pleased to be blazing this trail with the FDA. And finally, we continue to invest in the future and plan on being the choice of diabetes management for type 1, type 2, non-insulin using patients with diabetes and beyond.

Diabetes is one of the largest markets in healthcare and we compete with incumbents that have very, very deep pockets. We need to continue to be aggressive in our product development efforts, of our future platforms and our commercial approach.

We believe we are in a very enviable position and expect to capitalize on our technological leadership over the next several years. With that, we'll open up the line for Q&A..

Thank you. And then our first question is from Ben Andrew of William Blair..

Good afternoon, guys..

Hey, Ben..

Hey. Ben..

Jess, can you quantify the warranty expense that stayed in gross margin for us?.

It was a little over 5% for Q2, typically, Ben, it's about 2%. It's averaged 2% almost every quarter for the last two years. We would expect the warranty costs over the next few quarters to come down more towards the historical rate..

Okay.

And Kevin, maybe talk about how the label change is going to play out? And how you can position the company differently with a replacement claim year going forward?.

Well, certainly positioning us with a replacement claim will help us in the marketplace, because we will be the first with that claim, and patients won't have to stick their finger to make those decisions. And I think it will position us very strongly with Medicare.

I got to the question a bit in my prepared remarks about also what does this mean going forward, as once we have a replacement claim sensor, do we have other sensors that are not replacement claim? I think what we really have is the opportunity to change a landscape and set standards that will position us to be the leader in this space for quite some time..

Okay. And then maybe talk a little bit about the international expansion? Obviously the revenue was strong in the quarter.

How do we think about the guidance? What's baked in there from international for the balance of this year? And at what point does that start to really tick up for you? You mentioned 2017, but how should we think about that ramp?.

Yeah. Ben, this is Steve. I'll take this one. We're not going to break out in our full-year revenue guidance what portion we're attributing to international. I mean it's been running around that 15% range. Don't expect that to materially change until we really see the impact of reimbursement.

And as we kind of mentioned, we don't think that – we're still working through details in Germany. We don't think that we're going to start to see a meaningful impact until sometime next year. The hope, though, is we could see some additional – we've talked before about potential for reimbursement in France, potential for reimbursement in the UK.

If we can see some movement there in the back half of this year, then we may be able to get some impact and some contribution in 2017 from – more meaningful contribution from those countries as well..

Great. Thank you..

And then our next question comes from Mike Weinstein of JPMorgan..

Thanks, guys.

Kevin, you made the comment about MDI percentage of new patient starts, but I don't think you actually said what new patient starts did this quarter?.

We didn't..

Now is your opportunity..

Yeah. Mike, we did say that over the past six months we've added 2% to the type 1 population as far as market penetration in the U.S. So you guys can take type 1 population estimates, multiply 2% and you can figure out what we've done over the past six months..

Okay. Let me ask you about your guidance update. You grew 47% this quarter; 52%, 53% in the first half of the year. Your guidance for the second half implies basically 26% to 36%.

Tell us why that's the right guidance range?.

Well, as we sat and looked at it, we typically are not overly aggressive on our guidance, as all of you know, for right now. And as we look at the second half of the year, we looked at one number and just to get to the low end of our range is about $300 million, which is a lot of sales in a lot of centers for us. Could there be upside? There could be.

This guidance that we gave you is very typical and very consistent with what we've done in the past. To get to $575 million from the $253 million are, again, that's a huge number. So we're sticking within our range. We've always said that our sustainable growth rate is 35% to 40%. To the extent we achieve over that, absolutely fantastic.

And if we do, great. Eventually, Mike, the law of large numbers unless we continue to add mega, mega groups in the new patient adds and we continue to do that now and we hope to do that going forward obviously. But eventually the law of larger numbers even if we add a ton of new patients, growth rates are going to come down a little bit.

And we are seeing large numbers, even this quarter. Our first quarter of 2015 was a number that started with a seven. Our second quarter of 2015 was a number that started with a nine. And so we're growing off a much larger base from the year before.

So as we looked at this, we're comfortable with what we gave and if we overachieve, that's great and if we overachieve $575 million to $600 million, that $25 million is only a 3% or 4% overachievement and we decided it wasn't worth going higher than that. So those were our thoughts..

Okay. Let me come back to the question on OUS because I tend to think the Germany opportunity and then we'll see on timing of France and improved reimbursement in the UK. But I tend to think in Germany and then total Europe is probably a bigger opportunity and more impactful in 2017/2018 than probably the Street's thinking right now.

So could you spend just a minute, talk about Germany particularly, and what you're going to do to prepare for reimbursement?.

Well, in Germany the first step we did to prepare for reimbursement was to acquire our distributor. That's a very nice company over there and somebody we're very familiar with and who was very instrumental in working with the reimbursement authorities to get this approval. So we needed infrastructure. Obviously, we need to see what the rates are.

We are very pleased with how the group defined CGM by saying it had to have alerts, alarms and real time trends. So our product is right in the sweet spot. We are developing, really, an expansion plan over the last half of this year, determining how many field people we will add in anticipation of this.

We see Germany as a market in general where diabetes is quite well controlled. So if this product is available in that market, we think we can have a very good impact. It can be a very good growth engine for us.

The flip side of that – and the other countries the same, UK, France, any place we can get approval, we can see this thing taking off and doing well for us, as we've been doing in Sweden where we have very good reimbursement.

The flip side of that is to continue growing at the rates we're growing at and at the growth that we project, we have to have wins like this. They just have to happen. So we plan for them, and we plan around them, and things fall into place. But that's how we look at it for now.

We're very optimistic about it, not much for the rest of this year, but kicking into next year, we think Germany can be very big..

Okay. Perfect. Thank you. I'll let some others jump in. Thanks, guys..

And our next question is from Jayson Bedford of Raymond James..

Good afternoon. Thanks for taking the questions.

Just to follow on the last one, the reimbursement in Germany, is it broad, meaning for those type 1 and type 2 insulin dependent diabetics?.

It is. But there's still a couple procedural there. First, it needs to be ratified by the German government. And then we need to negotiate pricing. We still don't have a price attached to the system. We then actually still need to go really on a – much like the regional Medicare payers, we need to go region by region to negotiate.

My understanding is to negotiate individualized contracts. So that's why we're not suggesting this is a flip of the switch immediately, even if we get the pricing established and get the formal approval. It's going to take some time to get the infrastructure put in place..

Fair. I felt like the receiver recall kind of strained operations a bit in the first quarter.

You mentioned the impact on gross margin, but did it have any impact on 2Q, meaning either from a sales or even an OpEx?.

From an OpEx standpoint, not so much. I think on the sales side, we talked about our customer services and our tech support issues on our call last quarter.

In all candor with patients calling with product problems, on top of that not being able to get their calls answered, we created the perfect storm over the last half of the first quarter and even early into the second quarter as we resolved these issues.

And so as Steve said in his remarks, part of our return rate and part of the warranty rate is due to increased patient awareness or increased awareness because of the recall. But there was also increased awareness because we didn't answer the phones, and we weren't as good as we needed to be in getting back to people.

We think we've overcome those operational issues. I would tell you kudos to our sales force for getting through all that and being able to deliver the new patient numbers that we have and keep us on track with what we're doing. So it did put a strain on us, Jayson, and it was kind of felt all throughout the organization.

We think we're past it at this point in time, and now in operations we'll move on, build our new configuration receiver and resolve that warning letter with the FDA over the next few months and get past it completely..

Okay. And then maybe just a last one from me. There's certainly a lot of folks who were at the DIaMonD data presentation at ADA. It's been a little over a month now.

Do you expect this to have an impact on your business over the next few quarters, just this data?.

We need to get it published first and we're respectfully waiting for the authors to – I know they're drafting papers and we're going to get this in a journal at some point in time.

We think it'll have a very big impact when we go to payers and present the payers, here's a very, very reasonable cost alternative for the intensive management of diabetes. We can take cost out of your system by using MDI and CGM.

We can also take it to physicians who may be of the belief that a patient requires an insulin pump to be sophisticated enough to use CGM. When I first went in the field when I came here, that's what I heard in every doctor's office I went into. I wouldn't put a patient on your system unless they're on a pump. If they can't do a pump, they can't do CGM.

We don't hear that anymore, but now we have data that supports, as I said in my remarks, what we already knew. That you can use CGM and take shots and manage your diabetes very, very, very well. So I think it will – it can very much have a benefit. I think it'll help us with payers, I think it'll help us with healthcare professionals.

I think it'll help us across the board..

Okay. That's helpful. Thanks, Kevin..

And then our next question comes from Doug Schenkel of Cowen & Company..

Good afternoon, and thank you for taking the questions. My first one is, I guess going back to some of the commentary that you provided on market penetration, you mentioned that 2% of type 1 – 2% of the type 1 population was added to the installed base in the first half of the year.

Any chance you would say exactly what you're using for a denominator? I know you said there's a lot of estimates out there, but while we think there's about 1.5 million type 1s in the U.S. there are a pretty wide range of estimates.

What do you believe, and I guess as a follow up to this, what does your guidance imply for how much of a market you will expect to penetrate at year-end?.

I got to be honest with you. We really can't hear you very well here. I don't know what phone you're using. Can you get to another phone because you kind of broke in and out for us here..

Yeah. I guess I was just asking – sorry about that. I guess the question was really what you guys are using for a denominator in terms of a number of type 1s? We're at 1.5 million, but the estimates are pretty wide, as you noted.

And what is the assumption for percentage of the market that you will have penetrated by year end, at least to what's implied in guidance?.

Yeah. So we're not going to answer the latter. We use kind of a range between 1.3 million and 1.5 million when we do our modeling..

Okay. And then on gross margin, thanks for all the detail.

Just wondering if you'd comment on yield for the G5 transmitter? And as we look out beyond 2016, do you expect to get back to your 70% gross margin target in 2017? Or is this likely not until 2018 with the tailwind of the G6 launch?.

I think our yields on the G5 transmitter are fine.

The issues and the launch issues we talked about really relate to better understanding Bluetooth communications and some of the power of the battery as we got it out in the field, and again just some things we learned that we think we've resolved with the firmware fix that we just filed with the FDA back in the – earlier in the second quarter.

With respect to margins going forward, the one element that has changed from our business in the past, and I'll let Jess kick in if there's anything to add here, Jess, is we are now selling two transmitters that cost as much as a single transmitter used to cost for the price of one with a three-year life.

We do have cost reduction programs in place for that transmitter, but that's not really going to kick in until the last half of 2017 at the earliest. So we're going to lose a few margins and some margin overall on transmitter sales because of that two-for-one system that we built in order to power the Bluetooth radio.

The flip side of that is our mix may change over time as we get more patients, and maybe that 30%/70% split goes to 80% sensors, 20% hardware, they go back to the 70% level. I do agree with what you said earlier about the G6 sensor.

Assuming we continue to build sensors per day with the payer groups, there's certainly a revenue opportunity that can enhance our margin performance greatly. So we'll just kind of play this out over time.

As we said on the call, we'll get back up to the mid-60%s by the end of the year on an overall basis, and should be mid to high-60%s over the last few quarters.

Jess, Steve, do you have anything to add?.

Yeah. I think the only other comment I would have is that when we move to our Gen 6 platform, the sensor becomes a 10-day sensor. Remember, we bill our insurance companies on a per-day basis. So we have some opportunity to pick up margin there. It gives us some negotiating leverage with the payers. So that can help as well..

Yeah..

Okay. Thanks again, guys..

And then our next question comes from James Francescone of Morgan Stanley..

Hey. Thanks for taking the question. I was wondering if you could comment on international results in the quarter. Obviously quite strong on a year-over-year perspective, but actually down a little bit sequentially, which I think is not the trend that we've typically seen in that business.

Anything worth noting just purely in terms of the sequential trend in the international business?.

No. As you know, our international business is very much controlled by distributors and distributor purchasing patterns can be somewhat choppy.

And particularly as you look at the end of the second quarter with the vacation schedule and such is what happens in July and August in Europe? We've seen similar patterns in the past, so we're not concerned about it.

And Jess, do you have anything?.

No..

Okay..

Nothing to add..

Okay. And then just on operating expenses. You framed your operating expense plan at the beginning of the year as including a 20%/25% increase in cash OpEx on the core business plus $40 million in spending on strategic goals. It seems like the strategic spending is in line with where we thought.

Is the 20% to 25% cash OpEx increase on the core business, is that still the plan for the year?.

Yeah, it is. As we sit here today, understand we do have some increases in – again, the increases are typically as a percentage basis higher in the first half of the year because our revenue numbers are lower and we do incur some of that spending up front.

I would tell you operationally as we look in OpEx this quarter, we probably accelerated some of our expenditures on the customer service side with respect to head count to take care of the issues we talked about previously.

So some of that might be in, but it's not going to move the needle 10% or anything, if you move it 1% or 2% and we should catch up with that in the third and fourth quarters.

The strategic spend in the goals should be higher over the last half of the year, but as revenues go up, our OpEx as a percentage, the other cash-based OpEx as a percentage of revenues should come in line. As we look at the end of the year, we should finish quite nicely. We should be okay..

Okay. Sounds good. Thank you..

And then our next question comes from Danielle Antalffy of Leerink Partners..

Hey. Good afternoon, guys. Thanks so much for taking the question. Kevin or Steve, wondering if you could comment on, you mentioned progress with the artificial pancreas.

Just wondering if you could comment on how soon you think a DexCom sensor will be involved in something that's competitive with Medtronic's low glucose suspend, or some incremental step forward on the artificial pancreas from a commercial perspective, or do you think the next step for you guys will be an actual artificial pancreas?.

I think it will be incremental steps, but I will tell you one positive note. Bigfoot put out a press release, I believe it was just last week, announcing that they've initiated clinical trials on their system.

I'd like to defer – because Tandem and Insulet are both public companies, I kind of want to defer to them on their timelines, because they speak to their timelines during their calls. But, I would comment kind of broadly speaking that we believe our partners generally are making excellent progress, Animas included.

And so without being specific, I think it is likely that the 670G will be a commercialized product before one of our partners had something that's specifically competitive. But I think you'll be pleasantly surprised that it wouldn't be too far behind the Medtronic before at least one of them has something that's competitive out there..

Okay. That's very helpful. And then just a question on time of sensor wear. And I know you guys have the capability today to shut the sensor off after a certain number of days. And I know this is a tough topic given the fact that a lot of patients extend the wear of this sensor.

But just given what's happening with FDA and the potential for a non-adjunctive claim, how are you guys thinking about over the next few generations of the device potentially implementing an automatic shutoff of the system at a certain number of days? And how you'll sort of approach that with the patients?.

We've explored that possibility in a number of ways, from a technological perspective with respect to electronics, with respect to software, with respect to analyzing the membranes. And there's a number of ways we could in fact shut sensors off if they required that.

Right now the FDA has not required that from us, and I think as we learn more about the performance of the sensor with a non-adjunctive claim, we'll revisit that with the FDA over time. The issue becomes quite simple then, and it's a very good question.

If somebody is wearing a sensor on day 24 and they get a bad reading in dose insulin, whose fault is that? Is that ours? Or is that the patient's, or who's responsible? And so there is a risk factor here. Right now we believe our patients are very adept at determining when the sensor isn't working to meet their specifications.

So we have not been asked to shut it off at this point in time. We will evaluate it with future technologies. And we'll be fair with the patients no matter what we do.

If the FDA said you have to shut all these off at seven days, in all candor we'd figure out ways to make this right with the patients in our pricing models and make sure that they're not harmed tremendously either.

We know how important it is to them, and we understand the cost of the technology as these guys manage a disease that just never goes away. So it's something we will always balance. But it is something we've been very thoughtful about and something we've given a great deal of thought to with our future products.

As we look at a two-week sensor, quite honestly there's not a whole lot of reasons not to turn a two-week sensor off and eliminate that liability. It's when you're at seven days and our patients, many of them, say these things last much longer than seven days.

In fact one of the investigators did a study that showed week two might even be as good as week, or better than week one for some of his patients. So we've been hesitant to do that at this point in time, but we will explore it in our discussions with the FDA and we'll consider it during our business analytics over the course of time..

Got it. Thank you so much..

Thank you. Our next question is from Jeff Johnson of Robert Baird..

Thank you. Good evening, guys. Most of my questions have been answered. But, Steve, I just want to go back to something you talked about with your pump partners. I think a year or two ago you guys were clearly maybe frustrated with the step-wise fashion some of those pump partners were moving in.

I think more recently it sounds like at least one or maybe a couple of the pump companies are starting to invest in class-three manufacturing and they're putting frameworks in place to where they can maybe more efficiently push integrated updates to the field whether it's G5 going to G6, something like that.

So, I guess my question is, how broad is this trend at this point and do you feel like most of the pump companies are now starting to invest in these processes that are going to help your business as you can make these upgrade pathways more accessible to patients on both a pump and CGM combined basis going forward?.

Yeah. I do. I think we're still a little frustrated. I mean look, it's kudos to our R&D teams who are able to innovate and iterate as quickly as they do. I think our sense of frustration somewhat comes from the fact that we're able to turn new products at a much more rapid basis than any of our partners.

But I'm pleasantly surprised with some of the progress that they've made and I think look, we're going to see competitive offerings in the marketplace sooner rather than later..

Yeah. Fair enough. And then, Kevin, I guess a question for you. I just want to make sure I understood some of your adjunctive versus non-adjunctive comments that you made.

I don't think I'm reading this correctly, but you weren't trying to imply that you might go to market with like a dosing claims sensor and then a non-dosing claims sensor? Were you? You were just kind of trying to draw the difference between the dosing claim and those sensors that won't have it?.

I'm just drawing to light the question.

Once we have a non-adjunctive claim, are there adjunctive sensors that are still out there? I mean, once strip's got an accuracy standard set, do we have other strips that are non-adjunctive strips that you use another strip to measure to make sure the strip is right? I think we have some very interesting philosophical device and treatment issues that we need to discuss and work out with the agency over the next few months that we don't know the answer to at this point in time.

But we will, we'll get there..

All right. That's helpful. Thanks, guys..

Our next question is from Brooks West of Piper Jaffray..

Hey, guys. Good afternoon. This is Tom on for Brooks. Thanks for taking my question. Just wanted to ask about the transition to the pharmacy and if there are any updates you can give us? And then along those lines, it sounds like ASPs are in that $70 to $75 range again.

So just wondering at what point we might start to see some impact from the channel mix shift?.

We've been very deliberate in the contracts we've entered into. By the way, this is Kevin. We've been very deliberate in the contracts we've entered into to protect our pricing position.

Consequently, we haven't moved as much of our business to the pharmacy channel as we would like to, because we do end up getting in some very detailed pricing discussions.

And in many payers where we have a very good DME contract, and we're paid nicely for sensors, it doesn't make a lot of sense to give up a whole bunch of price to move over to pharmacy.

And so we're trying to find a way to strike that balance and maybe even strike a balance where our business model changes a bit over time, where we might get into a subscription-type model that's delivered through the drugstore.

We have a number of proposals out there that are very creative with payers and very different than what we've done in the past. We just haven't – we continue to discuss them all, but we haven't been willing to give up a bunch of price.

Earlier, one of the analysts talked about the 10-day sensor, and Steve spoke about that as well, where we currently bill per day with the sensor. So, right now if the sensor is $70 to $75 for seven days, you can see it's between $10.00 and $10.75 a day on the average.

To the extent we go to a Gen 6 sensor and we get 10 days, and we'd be willing to give a day back, pricing wise, for example, that may be the time to really aggressively go after the pharmacy business and use our price as a leverage point.

But today we've not been willing to leverage the business to give that up, so it's gone slower than we'd expected. We get warm reception, we have great meetings, but we've not been willing to give up a whole bunch of pricing so far..

Great. Thank you very much for that. And then just one quick follow-up.

Have you given your expectations for the timing of your other commencement of the post market following the panel?.

No. We have not..

Okay. Thank you..

And then our next question is from Tao Levy of Wedbush..

Great. Thanks. Good afternoon. Just a couple quick ones here. In terms of the penetration of the type 1 U.S.

population, where do you think the CGM market currently stands?.

Steve, you take your shot..

I'm guessing somewhere maybe slightly under 20% of patients using, between us and folks using the Medtronic system..

I'd agree with that. Under 20%, probably over 15% at this point..

Yeah. I would expect it to be over 15% given how quickly you guys have been adding patients. And in terms of – you had made the comment during the prepared remarks, when we think about the non-adjunctive claim that it might require additional training, the patient, of the clinician, caregiver.

Does that imply maybe a slower ramp initially than we might have seen before, and is that something you can control?.

We don't believe it'll be a slower ramp-up. We believe it's something we can control. The key issue here is just defining the parameters around the training and I'll give you an example. This was discussed at the panel meeting actually, when a patient sees a number with double arrows up, there are a number of ways that patient could be trained.

There are several models that have been built that suggest if you have a number and double arrows up and you're going to give yourself a bolus, you multiply it by, instead of your typical insulin dose, you multiply it by a multiplier like 1.2 or 1.3 or 1.1.

The question is there for that example, do we give that type of recommendation? Or is that the physician's recommendation to give? And where and how do we train patients, or do we train physicians to give that type of recommendation and get out of the way? We don't believe it's going to require a whole lot – it's going to slow us down, but we do need to get this training material really knocked out and revised before we can launch in that manner.

But that's the type of question that we're dealing with.

And the other one, and I can share a little more detail on FDA discussions here, do we have more specific training materials for our various subsets of patients? For seniors, for pediatric patients, for adults? Moreover, do they need to be trained differently combined with – as this technology expands, we can certainly have a training material set for endocrinologists and diabetes educators.

But as we get into general practitioners, because this will expand to general practitioners, how do we train them? And what type of materials and programs do we provide for those guys? Or do we take over the whole patient training for somebody who comes through that channel? Those are the things that we're discussing about training, and we presented our ideas and we're having a very open dialog with the FDA on the matter..

Okay. Perfect.

And just, lastly, when do you expect – I don't know if you mentioned it earlier on, but when do you expect to get the app that also includes the ability to measure how much insulin is either in the pump or on board?.

We'll file our next version of the app before the end of the year, and that app will be capable of receiving insulin data. The question becomes then what devices can communicate with that or does the patient enter that on their own? But that app will be filed before the end of the year..

So in order for the pump to communicate, the pump would have to have Bluetooth for that, I assume..

Or have some way to communicate to the pump..

Some way to communicate to the pump..

Okay. Great. Thank you so much..

And our next question is from Anthony Petrone of Jefferies..

Thanks, and good afternoon.

Maybe Kevin or Steve, just to go back to sort of the competitive landscape, you know after the July 21 panel, and if we look ahead, if there is a situation where you do receive the non-adjunctive label claim, what do you think this actually does to the competitive landscape? Is it safe to assume we can run scenarios of share gains for G5 in future sensors in the absence of competitors actually having that label claim? Then you also mentioned training manuals.

That has to be out there too. So it seems like a lot of infrastructure is being put in place here, and I'm just wondering what your high level thoughts are on the competitive landscape after these events? And then I have just two quick housekeeping questions..

Well, I'll start. Steve, Jess, anybody want to jump in? From a market share perspective, based on the independent data we read, we already have 70% share. So that's a pretty good, healthy position as we sit here today. And we don't have a fully-integrated pump that can regulate insulin delivery to attack the rest of that marketplace.

So we feel we're in a very good market share place as we sit here today. What we really hope to drive is continued market expansion and continued use of CGM. With respect to infrastructure, there is going to be a lot more – there is going to be more put in place.

But please understand, we've been laying the ground for this for quite – groundwork for this for quite some time. For example, with the G5 Mobile System, the user guide is part of the app. And so, even though there's going to be training, you still don't have to get that 600-page user guide in a box.

You can go to the app, and you can touch a screen, and you can get help on what you need, including access to training videos that can talk to you and tell you exactly what to do with the system. So the groundwork is in place. It's going to be different training and different education than medical devices have experienced in the past.

And we're pretty excited to be at the front of this..

Yeah.

So the only thing I would add is on the – would be on the Medicare front from a competitive perspective, that Medicare today is telling us they're requiring a non-adjunctive claim before they'll cover the products to the extent we obtain the non-adjunctive claim, and we have some period of time where we might gain Medicare access without – in the absence of competition, that can certainly give us an opportunity to capture some patients that no one else would be able to capture..

Helpful. And just the two follow-ups would be if you can provide any update on patient dropouts following the receiver recall, if there are any out there? And any updates on state Medicaid contracts? Thanks..

Go ahead..

No. I don't think we – we're not going to give you a specific attrition rate, but I don't think we've seen any meaningful uptick in attrition as it relates to the receiver recall issue. On the Medicaid side, quite frankly the last numbers I've heard is we're somewhere north of 25 states, but honestly I didn't – somewhere right in that range.

I haven't updated with our folks in several weeks now, so let's go with around 25 states..

Okay..

And our next question comes from Kyle Rose of Canaccord..

Great. Thank you very much for taking the questions. A couple quick ones here.

One, just given the amount of new patients added in the first half of the year, just wanted to see if you could characterize the type of patients you're seeing? And then any changes in underlying utilization? It sounds like attrition is stable, but is any way that – anything you've learned over the first six months of the year that would impact what we've seen prior based on patient utilization and drop-out trends?.

I think these have been relatively consistent. I did enjoy the question about the receiver recall issue. I would tell you with respect to the patient behavior there, in reality what we got more was angry patients who couldn't use the system, then we got patients who dropped off and they were very happy as we solved the problem.

I think utilization continues to be a mixed bag, but it's not really consistent. For example, people with – there are many people with great insurance who would gladly – their insurance company replace their sensor every 7 days, who still wear it for 14. There are others who can barely get 7 out of them. I think people use it as long as they can.

With respect to new patient characteristics, the one thing we can give you, and this is what we gave you earlier in the call, we felt it was very important to flip the patient base on the new patients from predominantly pumpers to predominately MDI patients because that really reflects the type 1 market.

And, again, saying that over 60% of our direct new patients, our multiple daily injection users shows that we've been very successful with this message and that CGM first is truly resonating with everybody..

Great. I appreciate that additional color.

And then just lastly, and then I'll hop off, is from a longer-term perspective as you think about going to – in the broader diabetes population, whether that be through the general practitioner channel or other ways, just how do you envision your sales channel changing, and the type of investments that would be required to help do that?.

We are studying that right now and I don't have a perfect answer for you. Some of the options have been presented to us as partners. Some of the options we've considered are a totally separate sales organization, contract sales organizations. We've considered a number of options and we'll continue to explore that as we get these products ready.

The one thing that we have learned and what we've spent most of our efforts on in the type 2 market over the past several months is just seeing how this technology will work. And let me tell you, type 2 patients, even those not using insulin, when given this feedback learn what to do with their health and learn how to take care of their diabetes.

It's a win. If we can get the cost structure the way we want to and the product the way we want it configured with our partners at Verily, we think this is a home run. So the distribution will be every bit as tricky as the technology, we just – we're getting the technology there first. That's kind of how we roll here at DexCom..

Great. Thanks for taking the question..

And then our next question comes from Erik Shoger of Northcoast Research..

Hi, guys. Two questions for you. One just a follow-up on the pharmacy channel question. I'm curious, when you originally started talking about that, you kind of painted it as a requirement really to enable you to grow at the high rates that you have been.

I guess, one, do you think that's still the case? And two, what ultimately flips the switch to get more and more patients onto the pharmacy channel? Is it something like DIaMonD study, or is it just kind of those bare-knuckle negotiations that you're talking about?.

It's a combination of all of them, Erik. I think the DIaMonD study will be very helpful. We have not rolled the DIaMonD study out in mass to the payers yet because we're waiting for publication, and we want to get the type 2 results in there as well because if we can get the insulin-using type 2 patients, that will also help our business.

But we are going to have to show ultimately that this system saves the payer more money than it costs them when all said and done. And I've been on a couple of payer calls recently, personally, because I want to better understand this dynamic and our team is very, very good.

But it's interesting, got strong three curveballs in one last Friday that we – I just – and I turned to them after we're done. I said, is this how this works? And they go, pretty much every time. So it – nothing is simple.

I personally believe it's a requirement to make getting CGM easier for our patients and I would love to do that with the pharmacy model. We will explore other models, as I mentioned on the call, if we could come up with the DME model, that's a subscription-based model, that would be great.

I can tell you there are some payers, I won't name them all, but there are two or three of them where we have a DME relationship and it is clockwork. We get a patient's name and they're on the system in a matter of a week. We have others that can drag this out for two or three months.

In DME, we have some of our pharmacy contracts that happen very quickly. Your prescription goes into the drugstore and you get the product. We have others that go through a pre-op that take a lot of time as well.

Ultimately, it's the consistency of the approval process and the reimbursement process that's most important to us, and we felt two years ago that the most consistent way we could do this was through the drugstore.

And I would tell you, I probably still feel that way as does the rest of our team because this is a product that patients use every day and they should be able to go get it without the hassle they go through now. Conversely, we have to manage our finances, and we have to manage our top line. And we've got a business to run to offset that with.

And we haven't been willing to give up a whole bunch of price to make a mass move over to the pharmacy, because where we have a direct contract with the payer, they have a price for our product. And we start from there, and often times they want to go down. And so we've held the line steady. Over time Gen 6 will help us lower costs.

Electronics will help us. Disposable transmitters will help us. Hopefully the thing going straight to the phone and avoiding the receiver altogether can help us. We have initiatives all throughout our product pipeline that will be very helpful in this regard. And so I don't doubt that we can get there. I think it's just taking more time than we planned..

Okay. Fair enough. And then one quick one, maybe for you, Steve, as it relates to the OUS business.

As you understand things now, particularly as you get into the negotiations in Germany, how do you expect the sensor ASP to change going forward, if at all?.

I don't even want to speculate because we're not at that point of negotiation yet..

Okay.

Is there any reason to believe that it would be drastically lower?.

No. I guess typically European pricing can be slightly lower. But, our pricing today on a cash free basis and even pricing, for example, in Sweden where we have reimbursement, is quite positive. So....

Okay. Fair enough. Thank you..

Yeah. I guess the only other comment I would make there is remember that the revenue DexCom as we go direct to some of these markets like Germany, we're cutting out the middle man having a distributor. So it's all, from a contribution perspective, it's way better for us..

Yeah. For sure. Great. Thank you, guys..

Thank you. We have reached the end of our question-and-answer session. At this time, I will turn the call back to Kevin for closing remarks..

You know what, as you can tell by this call, things are very busy here these days. And we continue to believe that we're merely scratching the surface of the opportunity in front of us with this technology to change lives. Many of you heard the testimonials at the recent FDA Panel Meeting. They moved everybody in the room nearly to tears.

This device changes lives, and that's what inspires us to come to work and do what we do each and every day. Thanks for listening and taking the time. And we'll talk to you next quarter..

Thank you, ladies and gentlemen. This concludes today's conference. Thank you for participating. You may now disconnect..