FDA approves Tandem t:slim G4, shipments begin in October – September 13, 2015

Yesterday, Tandem announced FDA approval of its t:slim pump integrated with the Dexcom G4 transmitter. Shipments start in October, and Tandem is already taking orders on the newly launched t:slim G4 section of its website. The approval reflects a ~13-month FDA review and comes on the early side of the 12-18 month expectation following the July 2014 submission. Tandem held a short conference call to discuss the news, which mostly offered financial perspective on the news (sales disruption in Q3, but a bigger Q4).

As expected, Tandem is not offering an upgrade program for current t:slim users, meaning the t:slim G4 will initially be available only to new Tandem customers. The one exception is those who purchase(d) a t:slim between August 1 and October 2, for whom Tandem will offer a free exchange option to get the t:slim G4 . The lack of an upgrade program is unfortunate from a patient perspective, though it reflects the reality of being a young pump company without other business units to help cover the cost. Still, interestingly, management estimates that less than 25% of its installed base is on CGM (<6,000 patients out of a base of ~24,000), so the majority of Tandem users will not be impacted by the availability of t:slim G4 – although presumably more will opt for CGM with the combination product available.

One big question for Tandem is how this month’s upcoming launch of Dexcom’s G5 Bluetooth-enabled transmitter will impact uptake of t:slim G4. The t:slim G4 won’t be compatible with the G5 transmitter, and it also doesn’t have the updated G4AP algorithm (MARD of 9.0% vs. 13.2% with the original G4 algorithm). How will patients weigh between getting data on their phone (with G5) or on the pump screen (with G4)? Dexcom has taken the view that most will choose the former. CGM on the phone is a compelling offering, though we appreciated the way Tandem management put it: “How people prefer to see their CGM information is a personal choice.” We agree – some with CGM data on their smartphones have expressed fatigue in “always” looking at their phone (or being perceived as distracted by texts etc when they are really looking at CGM data.)

Fortunately, the either/or is only a near-term problem. Tandem announced an updated agreement with Dexcom last month during its 2Q15 call to integrate Dexcom Gens 5 and 6, though no launch timing has been shared (we assume 12+ months away). Insulet may be first to market with a Dexcom G5 compatible pump, though there is some uncertainty depending on when the next-gen PDM launches.

Less important than the sales upside is how this sets up Tandem for the future: a PMA approval under its belt, which should accelerate the timing on a Dexcom G5 integration and an artificial pancreas project. Competitively, the t:slim G4 puts Tandem on par with Animas’ Vibe – the only other Dexcom-integrated pump – which did see encouraging 32% operational growth in 2Q15. Medtronic is of course the major player here, with the MiniMed 530G still seeing good uptake after two years in the US and an FDA submission of the 640G expected this year.

See below for more on the business implications of the t:slim G4, our first look at the user interface, our key questions, the investor call Q&A, and an automated insulin delivery landscape review.

Business Implications

• Tandem did not update sales guidance for the year ($70-$75 million; 41-51% growth vs. 2014). Upgrades are always tricky to predict, and more clarity is expected in the November call. Yesterday’s call said that there is some pent-up demand for the t:slim G4, which could cause patients to delay buying a t:slim in Q3 to get on the t:slim G4 in Q4. The demand was not quantified, though management said pump/CGM integration has been the #1 requested improvement following the t:slim launch in 2012 – we wonder if that will change once G5 launches. The ultimate impact of the short-term disruption is presumably neutral, since lower Q3 sales would be made up for with higher Q4 sales.
• Tandem expects 2015 sales even more heavily weighted to Q4 than last year. This makes sense, given the shipment date in October and potentially delayed purchases from Q3 to Q4. For context, 36% of full-year 2014 sales came in Q4 – much of that was due to seasonality and a sales force expansion that happened earlier in 2014. There was no sales force expansion this year, so a big Q4 will indeed be needed to hit the guidance.
• Tandem’s first four-year warranty turnovers will come in late 2016, as the t:slim shipped in August 2012. Only 1,057 pumps were sold in 2012, meaning a relative small number of upgrades will hit next year. More renewals will come throughout 2017, as Tandem sold 6,472 pumps in 2013. At that time, we assume a G5 integration would realistically be available, but aren’t positive.
• It may eventually be possible for Tandem to remotely update the t:slim’s software (through Project Odyssey) and turn on the t:slim’s built-in Bluetooth radio; that would enable G5 compatibility without needing a brand new pump. The regulatory specifics are still in progress, and they seem complicated. Can the 510(k) cleared original t:slim be software updated to become a PMA-regulated G5-integrated pump? It’s hard to know how the FDA will think about that, though from a patient perspective, we don’t see that kind of software update as very high risk. It is not possible to make the t:slim G4 compatible with G5, as the radios are different. 
• Similar to the Animas Vibe, Dexcom will separately manufacture and sell transmitters and sensors. Ultimately, we hope there is some way to combine the sales channels; perhaps this will happen once artificial pancreas offerings come out.
• Tandem is now manufacturing under PMA regulations and has capacity to successfully meet demand. The latter was not quantified, but we imagine it will be less than 10,000 t:slim G4 pumps in Q4, and probably less than 5,000. (Tandem’s best quarter to date [4Q14] saw 3,929 pumps shipped; going forward, management believes most of the sales mix will still be weighted to the original non-integrated t:slim.]

Competitive Implications

• The t:slim G4 helps Tandem stay competitive with the other two sensor-integrated pumps in the US: the Animas Vibe and Medtronic’s MiniMed 530G. The Vibe saw encouraging 32% operational growth in 2Q15, and the t:slim G4 interface strikes us as a better option for Dexcom-using patients (see below). Medtronic’s MiniMed 530G is still seeing good uptake after two years in the US, and brings the added advantage of low glucose suspend (though the disadvantage of a less accurate sensor. An FDA submission of Medtronic’s next offering, the MiniMed 640G (predictive low glucose suspend), is expected this year. Insulet won’t be out with a Dexcom G5 integrated offering until sometime in 2016 at the earliest, though that depends on the timing of the next-gen PDM launch (submission expected in early 2016). We’re also not sure if the next-gen Insulet handheld can launch with G5 integration to start, or if that will come later.
• The t:slim G4 approval goes down to age 12, the youngest available for a sensor-integrated pump in the US. For context, the Animas Vibe is approved for ages 18 years and older, and the MiniMed 530G is approved for ages 16 years and older. This is a competitive advantage for Tandem, though among Dexcom-using pediatric users, there is some nuance: parents will have to decide between the remote monitoring advantages of G5 and the convenience of the sensor-integrated t:slim G4. Tandem expects to further reduce the age indication to six or seven years old in 2016 for the original t:slim; its effort on sensor integration going forward will focus on G5.
• This news bodes well for Tandem as it looks to automate insulin delivery, though the company will likely be second to market in the best-case scenario (see the landscape review below). As of Tandem’s July call, an IDE filing for an artificial pancreas product is expected by the end of 2015; previously, a late 2017 launch was deemed possible. For context, Medtronic has already begun the pivotal study of the MiniMed 670G hybrid closed loop. That study is single-arm, 150 patients, and has an expected completion date in May 2016; per JPM 2015, US launch is expected by April 2017. Meanwhile, an FDA submission of the MiniMed 640G (predictive low glucose suspend) is expected this calendar year. Even though Medtronic has historically missed its timelines, Tandem will almost certainly come to market after Medtronic’s near-term automated insulin delivery options. The launch timing on products from Animas, Bigfoot, Bionic Pancreas, Insulet, Type Zero, and Roche is less clear, though Dr. Aaron Kowalski has predicted multiple hybrid closed loop products will be available in the 2017-2018 timeframe.
  
  • In 2Q15, Tandem said the “first step” will stop insulin delivery (presumably predictive low glucose suspend), and the “second step” will increase insulin delivery (presumably hybrid closed loop). We understand the more conservative approach to go with PLGS first, though believe going for hybrid closed loop might be prudent to stay competitive with an increasingly crowded field.

User Interface Test Drive

• The t:slim G4 interface has done a great job with the integration, lifting the same menu structure and color scheme from the well-received G4 Platinum receiver. See a screenshot below. We especially like that the t:slim G4 shows the CGM information AND insulin-on-board right on the home screen, something the Animas Vibe and MiniMed 530G do not do (though the Vibe can do it with one button push).  
• Overall, the t:slim G4 provides a better on-device user experience than the Animas Vibe or MiniMed 530G, which both have lower resolution screens with less color and button-driven menus. We do think this increasingly matters when patients are choosing between pumps – their iPhones are now the comparators! –  and Tandem has certainly kept the ‘cool factor’ going with the t:slim G4. Of course, it’s hard to know how much this is a driving factor when patients are weighing products’ composite profiles (e.g., automation, waterproof, tubeless, etc.).
• Tandem has updated its free t:simulator app, which allows for a hands-on run through of the new t:slim G4 user interface. The app is available on the Android and Apple app stores, and it’s impressive to see it already updated with the new user interface. Upon downloading t:simulator, users can select among the t:slim, t:flex, and t:slim G4, allowing use of the pump interface on a smartphone/tablet just as it would feel like on the actual pump. It’s a great training and sales tool and something we have not seen any other company do as well. Tandem is smartly leveraging the touchscreen at every possible opportunity.

Close Concerns Questions

• How much pent-up demand exists for the t:slim G4? 500 patients? 1,000 patients? 5,000 patients?
• To what extent will t:slim G4 and future G5 integration add incrementally to Tandem’s growth?
• How will patients weigh between getting data on their phone (with Dexcom’s G5 shipping later this month) or on the pump screen (with the t:slim G4 and Animas Vibe)?
• What is the timing on a Dexcom G5-integrated Tandem t:slim? Will it be possible to add sensor integration via remote software update?
• How quickly will Tandem move on automated insulin delivery relative to competitors?

Questions and Answers

Q: You talked about pent up demand in Q3. To what extent have you polled potential customers to date? I’m assuming in Q1 and Q2 you hit or slightly exceeded expectations on good growth. What kind of metrics do you have on this pent up demand right now or is it more qualitative than quantitative?

A: It’s more qualitative. Since we didn’t have approval, we didn’t track quantitative metrics for the G4 product. We have heard and believe that after our earnings calls, customers have increasingly started to withhold their purchases and made the decision to wait.

Q: On the exchange program, are you going to eat the cost of goods on any device that’s exchanged? They’re not going to be reused some place?

A: It will be a cost that we eat. We will try our best to refurbish the products that come back and utilize them for alternative uses such as demos, samples, or warranty.

Q: If we flash forward to Q4 2016, how do you think your three products are going to play out? Do you think you are going to have more t:flex or more G4?

A: I certainly think we’re going to have more G4. From what we’ve seen in the data, we know that about 25% of all pump users are also using CGM. We’re hearing that it could be even higher on the margin, and something like 40% of people are preferring it today. I definitely think G4 will be a bigger contributor than the t:flex product next year.

Q: What about t:slim with vs. without Dexcom on board? Which do you think will be the bigger seller in a year?

A: I think t:slim without G4 is probably going to continue to be the biggest seller a year from now. Again, we’re just guessing right now. We really don’t have a lot of great data that’s going to tell us how this cannibalization is going to work out.

Q: Can you help us understand the incremental opportunity for you here? Whether it’s 25% or 40% of the market, is that a portion of the market that you have not been competing for and now you can? Is that what we should be paying attention to in the next year or two?  

A: I think you’re saying it in the right way and that’s the way we’re thinking about it. I’ve seen a couple of different data points and about 25% of the base is probably using CGM. On the margin, people are probably choosing at a higher percentage because it is expanding in popularity. Providers are putting more people onto integrated CGM. We also have data from Kelly Close indicating that the number one reason people don’t choose us is because of integrated CGM, and the number one reason that they do choose us is because of the touchscreen intuitive interface and the way it looks like a consumer device. We win on our set of cool and easy to use features, but we lose on the CGM integration issue.

Q: To reflect on it a little further, once the dust settles on near-term issues, you will be competing in a larger part of the market. All things being equal and executing well, will you be growing faster as a result of this?

A: Yes, we can access a bigger part of the market, and that part of the market seems to be accelerating in growth. Obviously you can see the numbers that Dexcom is putting out.

Q: Is the insurance approval process longer for t:slim G4 or for t:slim alone? Is that a hurdle or is that going to be pretty seamless?

A: I think that Dexcom has done a good job at getting widespread private reimbursement, so I don’t think that’s going to be as big of an issue as it would have been four years ago. Dexcom has done a lot of the heavy lifting, so I don’t see that as a headwind.

Q: Is there any reason to hope that this three-product portfolio could change the potential mix of direct vs. distributor mix and push it in a more favorable direction in 2016?

A: I doubt that. I don’t see any dynamic in CGM that would help us push that payer mix between distributor and direct in 2016.

Q: Absent short-term cost of goods sold related to the trade-in programs, how should we think about cost of goods with three different product lines and the gross margin progression going forward?

A: The volume story has always been the key to improving our margins. The benefit of having three products is that the pumps are generally the same physically, and then in the manufacturing process there’s a different software load. The t:slim G4 and t:slim utilize the same cartridge, and the t:flex utilizes a slightly larger cartridge. These volumes are going to create greater gross margin opportunities going into 2016.

Q: In terms of the replacement cycle, we’ll start to see that in late 2016, and more in 2017. That should help the volume, right?

A: Correct. We launched our first t:slim products in August 2012, and we’ll start to see a four year renewal cycle in late 2016. There were roughly 1,000 pumps in 2012, and ~6,500 in 2013, so that’ll hit in 2017.

Q: In terms of other new products, you mentioned Odyssey. What was the second one that you expect in 2016?

A: Pediatric indication. We will reduce the pediatric labeling down to six or seven years old.

Q: Is the percent of your installed base that uses CGM in line with the market: 25%?

A: No, I think it’s lower than that. We don’t have an exact number, but we think it’s lower than the market mix.

Q: You mentioned that the SG&A spending was going to go up a bit with marketing events. Do you have an estimate of what that increase will be in the third and fourth quarter?

A: In 3Q last week, we did a sales training and incurred that cost. We’ll incur placements on demonstration products that we’re trying to put out there for this launch, and the cost of that is still being defined as we move along. It’s something that we’ll see as a hiccup over the next few quarters, then steady state SG&A growth from there.

Q: Are we talking about an extra millionish per quarter?

A: It’s too early to give the estimate since the programs are still being quantified.

Q: On recognizing deferred revenue – does your quarter end on the 30^th or on October 2^nd?

A: The 30^th.

Q: So you won’t be able to recognize any of the revenue until October 2^nd at the earliest?

A: We’re going to be contacting customers with open orders, and asking them if they are interested in us holding their order so they can make a decision that includes the consideration of G4. If they say they want to stick with t:slim, we’ll recognize that revenue in Q3. And for those who bought a pump in August, if they want to stick with it we will recognize that revenue in Q3.

Q: And if they say they want the upgrade?

A: We won’t be able to recognize that until we ship them the G4.

Automated Insulin Delivery Competitive Landscape

• See below for an overview of the automated insulin delivery landscape, as far as we aware. We acknowledge this list may be incomplete, as there may be other stealth startups or academic groups working to commercialize closed-loop technology.

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