February 15-18, 2017; Paris, France; Full Report – Draft

Executive Highlights

This report contains our full coverage of the 10^th International Conference on Advanced Technologies & Treatments for Diabetes (ATTD 2017) held in Paris, France from February 15-18. It was the largest ever ATTD, convening over 2,000 delegates in the City of Lights.

As Bill Gates reportedly said, “We always overestimate the change that will occur in the next two years and underestimate the change that will occur in the next ten.” In the case of diabetes technology, we’ve seen substantial change in both time windows. Two years ago, Medtronic headlined the conference with new clinical data on the then-recently-launched MiniMed 640G (predictive low glucose suspend), and shared early data from a camp study of the 670G hybrid closed loop system. This year, we saw glycemic outcomes from more than a dozen patients who have been using the 670G for 18 months! Also at ATTD 2015, we saw final data from Abbott’s FreeStyle Libre pivotal CE Mark trial; fast forward to ATTD 2017, and there are over 250,000 patients using Libre in Europe (as of 4Q16), with a US launch expected in 2H17. Going back even further, Dr. Bruce Buckingham put it well: This is really a new day. It’s a storied time we’re in here. I think the first ATTD (2008) meeting had ~400 people, all engineers, and now we’re talking about human factors.” Indeed, the progress has been outstanding in less than ten years, with a refreshing move to commercial products and now to follow-on improvements.

To give ATTD attendees some culture and a break from the rush of the meeting, The diaTribe Foundation hosted its second Art Walk at the marvelous Fondation Louis Vuitton. The museum was all it was cracked up to be, and diaTribe received countless thank yous – one of our favorites was from IDC’s Dr. Rich Bergenstal, who likened Frank Gehry’s architecture to AGP traces. He’s right! – see the humorous collage here.

Below, you’ll find our full coverage of ATTD 2017, opening with some themes and followed by five categories of reporting: (i) Automated Insulin Delivery; (ii) Glucose Monitoring; (iii) Insulin Dose Titration and Delivery Devices; (iv) Connected Care, Data, and Additional Topics; and (v) Exhibit Hall. Talks highlighted in yellow were among the most notable we saw at ATTD 2017.

We’ll see you back at ATTD 2018 in Vienna, Austria from February 14-17!

Themes

AID – MiniMed 670G LeSsons, Closed Loop Data from Insulet, Diabeloop, and iDCL

• We heard a slew of insightful commentary on the MiniMed 670G at ATTD 2017. Though the first FDA-approved hybrid closed loop has not rolled out broadly yet, a lot of learning has taken place during the pivotal trial’s continued access phase.
  
  • Dr. Rich Bergenstal pointed out that some 670G pivotal participants needed a more aggressive insulin:carb ratio (at meals) and shorter duration of insulin action to boost the system’s daytime aggressiveness, especially post meals. He also noted that patients who “let the algorithm do the work” often do the best on hybrid closed loop – treating lows requires less carbs on closed loop (since basal is already suspended) and stacking correction boluses on top of automated basal insulin can result in serious lows (followed by overeating and subsequent highs!). He echoed this point at ENDO 2017, commenting that his clinic’s mantra for hybrid closed loop is “Let it work, let it work, let it work.”
  • The MiniMed 670G will revert from Auto Mode back to Manual mode during prolonged periods of hyperglycemia, prolonged delivery of zero insulin, and sensor/self-diagnostic issues. This is called “safe basal timeout,” though it is definitely a limitation of the conservative system – prolonged lows and highs are precisely the times where remaining in auto mode is critical. Hopefully this conservatism is changed in a future version of the 670G, as dropping out of Auto Mode is frustrating for users.
  • In the 670G continued access phase, Barbara Davis Center users (n=19) – now on the system for over a year – have largely seen similar outcomes to the three-month pivotal trial (some exceptions for adolescents). Time spent in hybrid closed loop (auto mode) and sensor utilization have dropped off slightly in both groups, particularly in adolescents by about two fewer hours per day (~ten-percentage points). Still, These data mostly confirm the single-arm pivotal results, and perhaps the commercially available system will be able to maintain more time in Auto Mode. At Medtronic’s ENDO booth, we heard that there are over 700 people in the initial 670G customer training launch, 110 of whom are Medtronic employees with diabetes. This limited launch phase, which will last until June 2017 – potentially over a million person-hours of closed loop – should elucidate many more considerations for patients, caregivers, and providers to be aware of.
  • The DT&T paper publishing more detailed 670G pivotal data identifies “generalizability” as a limitation. The authors write on the last page of the article: “...the generalizability of our results may be limited, given the lower baseline HbA1c levels for both cohorts relative to T1D exchange mean HbA1c levels; the frequent contact that study subjects had with site personnel; the fact that approximately half of the adolescents and two-thirds of the adults were using CGM at baseline; and the exclusion of subjects with HbA1c levels >10%, >2 recent episodes of severe hypoglycemia, and those with any recent episodes of diabetic ketoacidosis.” This will be something to keep an eye on in the real-world – how will the typical provider and patient interact with the technology? How will early adopters of the MiniMed 670G like the technology and the fairly conservative algorithm, particularly those with low A1c’s at baseline? (For perspective from early 670G users in the dQ&A panel, contact richard.wood@d-qa.com).
  • Many asked, “What is the max insulin delivery rate the 670G can deliver per hour?”, to which the most common response was, “It’s complicated to explain.” The algorithm does adapt every day in each person, so there is not a one-size fits-all answer to this question. (It sounded like audience members wanted to know what multiple of the normal basal rate the algorithm can deliver – e.g., 2x, 4x, etc.) The 670G algorithm needs at least 48 hours in Manual Mode before Auto Mode can be initiated, and it uses a patient’s actual insulin data to individualize insulin delivery for Auto Mode. At midnight, the algorithm reassesses up to six days of the most recent pump data to update: (i) the feedback controller gain (aggressiveness); (ii) the maximum auto-basal that can be delivered by the system; and (iii) other system parameters.
  • Dr. Thomas Danne showed a compelling example of an unannounced meal on the 670G vs. Medtronic’s Advanced Hybrid Closed Loop, showing how the latter’s automatic correction boluses halved the time in hyperglycemia and shrunk AUC. He concluded that the MD Logic Bolus in the Advanced Hybrid Closed Loop is “promising” and will be further studied in a recently NIH-funded trial starting later this year. Medtronic has not shared official timing on when this product might come to market, though we’d guess it’s a couple years away at this point.
• Insulet, Diabeloop, and the iDCL consortium also presented positive trial data. See our automated insulin delivery competitive landscape for more on each of these systems. 
  
  • Dr. Bruce Buckingham shared initial feasibility study data from Insulet’s in-development Horizon system – the algorithm was safe, very effective at avoiding hypoglycemia, and highly effective at night. The area for improvement is in reducing post-meal glucose – Dr. Buckingham believes that tighter control is possible, though Insulet may need to tweak the algorithm to offer more time-in-range and minimize excursions after meals.
  • Back-to-back Diabeloop presentations showed that the hybrid closed loop system handles exercise and meals – two of the biggest hurdles of closed loop therapy – very well. In the three-day exercise scenario, the system increased time in range (70-180 mg/dl) from 68% to an impressive 80%, and during the three-day gastronomic challenge scenario, time in range nearly doubled, from 49% to 86%. Diabeloop aims to commercialize the whole system with its own algorithm and integrating pump and CGM components from others (thus far, Cellnovo’s pump and Dexcom’s CGM).
  • Dr. Boris Kovatchev gave a progress report on the International Diabetes Closed Loop trial (to serve as Tandem/TypeZero’s pivotal), including first results from the 14-day training protocol phase. In 20 individuals thus far, the training protocol has seen similar outcomes to prior studies: an overall mean glucose of 156 mg/dl (148 mg/dl overnight), just 1.6% of the time <70 mg/dl (0.5% overnight), and 70% of the time in 70-180 mg/dl (79% overnight). The main phase of the trial (CT.gov posting here) will reportedly begin soon, randomizing 240 patients in a 2:1 ratio of closed-loop control vs. sensor-augmented pump therapy over six months. 

CGM Outcomes Expand – Real-world Results for Libre, Dexcom Benefits in Type 2 MDI Users

• In an Abbott Diabetes Care symposium, Dr. Ramzi Ajjan presented compelling real-world data from >55,000 FreeStyle Libre users suggesting that, when they scanned at higher frequencies, their A1cs were lower and they spent less time in hypoglycemia. De-identified data was collected over a period of 18 months (October 2014 – May 2016), and compiled to nearly 400 million glucose data points, 64 million scans, and 1.2 million fingerstick blood glucose readings. Through this mass of incredible data points came some strong trends: As scanning increased, (i) estimated A1c dropped 8.0% to 6.7%; (ii) time spent with glucose <70, <55 and <45 mg/dl decreased by 15%, 40%, and 49% respectively; (iii) time >180 mg/dl fell from 10.4 to 5.8 hours per day; and (iv) time in range increased from 12.0 to 16.8 hours per day. In addition, users scanned 16.3 times per day on average, with some scanning as frequently as 50 times per day! What we found compelling about these results is the scale of the real-world data – a 55,000-patient-strong cohort – especially because it aligns with data from Libre’s randomized controlled trials. We hope this stands as further evidence (especially for payers) that patients derive significant real-world benefits from real-time glucose information and trends; that they are willing to scan at high frequencies outside of a clinical trial setting; and that more glucose data drives therapeutic and behavioral adjustments.
• In another sign of expanded CGM outcomes, Dr. Rich Bergenstal presented type 2 data from Dexcom’s DIaMonD study testing CGM in MDIs. Results showed 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 an impressive 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). CGM adherence was very strong: 93% of the type 2 cohort was using CGM ≥6 days per week at six months. Dr. Bergenstal expressed at ATTD (and more recently at ENDO 2017 that there was minimal handholding throughout the trial, and he’s eager to explore how coaching and advice (specifically on adjusting insulin) might result in even greater benefits.
  
  • Dexcom previously presented data from the DIaMonD type 1 cohort at ADA 2016. Compared to the type 2 DIaMonD results, the type 1 group saw larger improvements with CGM: a 0.6% A1c advantage (-1% vs. -0.4%), 76 more minutes per day in range (a 12% improvement from baseline), and 22 fewer minutes per day <70 mg/dl (a 34% improvement). DIaMonD was recently published in JAMA, alongside the Swedish GOLD study – a huge victory for the field and hopefully a positive for broader reimbursement.
• On the same day as the DIaMonD presentation, Jaeb’s Katrina Ruedy presented long-awaited, positive results from REPLACE-BG, a 226-patient randomized T1D Exchange trial comparing use of Dexcom’s G4 CGM with and without confirmatory fingersticks (n=77 for CGM+BGM vs. n=149 for CGM-only) over 26 weeks. As we expected, the groups had near-identical outcomes by study end, with no significant difference in time-in-range, hypoglycemia, hyperglycemia, A1c, mean glucose, or coefficient of variation. The outcomes were also not significantly different in subgroups based on age, type 1 diabetes duration, and education level, a good sign that non-adjunctive CGM use is safe across the board. Protocol adherence was excellent, with 91% of the CGM-only group and 95% of the CGM+BGM group wearing the G4 for >6 days/week (mean: 6.7 and 6.8 days/week).
• Growing the evidence base for CGM is a big need in the field, given the small penetration globally, clinical inertia (or baggage from prior generations), and continued cost/admin hassles for many users and HCPs. We hope these trials will drive (i) further prescribing of CGM first, especially for endocrinologists trained for years that pumps should come first; and (ii) broader reimbursement from other public and private payers in the US and abroad. CGM devices are cost-saving and health-improving technology in the short- and long-run, and we’re especially elated to see the focus moving to type 2s and far more real-world data. Of course, with broader access to therapeutic CGM, there need to be clear and simple protocols in place to help patients make dosing decisions based off of the blood glucose number and trend arrows. On this front, Drs. Steve Edelman and Jeremy Pettus published their recommendations in JDST last August, and Dexcom just updated its G5 app with non-adjunctive labeling education in March. Since then, the Endocrine Society has also released dose adjustment recommendations for Dexcom’s G5 in adults and pediatrics. Dose Titration and Decision Support in the Limelight
• The first, center stage plenary of the conference was titled “automated clinical decision support systems.” That such a session included superstars Drs. Aaron Kowalski, Bruce Bode, George Grunberger, and Moshe Phillip – and kicked off the meeting – was a huge signal that the organizers view decision support and insulin titration software for injection users as critically important. In the opening talk, Dr. Kowalski wondered aloud: “How do we optimize insulin dosing strategies automatically? How in 2017, do we still have crudely determined basal rates, insulin to carb ratios, insulin sensitivity factors (correction factors), and many people on pumps with factory settings?” Yes! Software to support optimal insulin regimen is heating up, and we hope to see it flood the market with a big commercial presence. We expect the most widely used tools will come directly from (or via partnership with) pharmaceutical companies, ideally bundled with their insulins – Sanofi recently expanded its partnership with Voluntis and received clearance for the My Dose Coach basal titration app; Novo Nordisk and Glooko partnered in January to develop digital tools, including dose titration; and Lilly received FDA clearance in January for Go Dose (a prandial Humalog insulin titration app) and later in the year announced major plans to commercialize automated insulin system for pumps and injectors.
  
  • Last year’s meeting also saw an increased insulin dose decision support focus, a trend we expect to continue at future ATTD meetings. As the pump-based AID field matures, expanding learning into injections is an obvious and very high impact frontier.
• Dr. Bruce Bode presented positive Glytec Glucommander data from both the inpatient and the outpatient settings. We thought the financial proposition associated with Glucommander was quite compelling, with over $3 million in potential savings per 250-bed hospital per year due to shorter patient stays, reductions in point of care testing, reduction in cost of treating DKA, and reduction in cost of treating CABG (coronary artery bypass surgery) patients. Outpatient data was an extension of the three month data we saw at ADA 2016 – patients were titrated to goal in just 11 days. Over six months, A1c dropped from 10.2% to 7.6% (a “highly significant” 2.6% decline), and hypoglycemia was “extremely rare” – just 0.2% of blood glucose values were <54 mg/dl and 0.02% of values (just five instances) were <40 mg/dl. As Dr. Bode stated, “obviously this does work,” but implementation of Glucommander outpatient has seemed kind of limited. We’d guess partnership will be needed to take this product to the outpatient setting in a big way, ideally with glucose values flowing in automatically.
• An interim MD-Logic Advise4U study showed that DreaMed’s AdvisorPro pump titration software operates similarly to expert clinicians. For six weeks, patients at Schneider Hospital assigned to either the Advisor (n=7) or the control group with expert physician-guided decisions (n=8) spent similar time in range (52% in control vs. 50% with the Advisor). The Advisor group spent slightly more time >180 mg/dl (42% control vs. 49% Advisor), but notably less time <70 mg/dl (7% control vs. 2% Advisor). The clinicians and Advisor used different methods to achieve roughly similar time-in-range outcomes, an interesting quirk of clinical decision support. The Advisor made more adjustments, making 4.7 recommendations per patient while clinicians made 3.3 per patient. The Advisor also unsurprisingly adjusts more parameters than clinicians – a big benefit of using algorithms to more finely tune all the knows in insulin delivery. The study will enroll an additional ~20 patients and is expected to wrap up in a year, and a multi-center, multi-national study with a similar design and funding from the Helmsley Charitable Trust is being planned now. We hope to see more details at ATTD 2018 – this kind of software could be gamechanging. The Advisor will eventually be incorporated into Glooko’s platform, pulling data directly from all the major pumps, CGMs, and BGMs and making suggestions accordingly. 

When Should CGM and Pump Therapy Be USed in T2D?

• Dr. Irl Hirsch made a compelling cost-effectiveness case for intermittent, real-time CGM use in type 2 diabetes. The talk was a wonderful, in-depth analysis of A1c reductions (coupled with CGM-behavioral memory), QALYs (intermittent use of CGM is very favorable, as the device is used for a few weeks but it’s impact is lasting), and comparison to other diabetes therapies. Dr. Hirsch concluded that, from an A1c-centric perspective, cheaper diabetes drugs (glyburide, metformin, and pioglitazone) may be the most short-term, cost-effective way to achieve glycemic control, but they can put patients at elevated risk of complications (particularly hypoglycemia) or be maxed out such that another therapy is required. Intermittent CGM in type 2 diabetes can both protect against hypoglycemia and reduce A1c, and at a lower price point than newer orals and injectables. Hear hear! Now, we only need to see some head-to-head studies proving this…
• Dr. John Pickup presented a meta-analysis indicating that pumps are likely most cost-effective in type 2 patients with elevated A1c and insulin doses even after insulin injection optimization. This mirrors Dr. Pickup’s meta-analysis work in CGM, where patients also saw more benefit with higher baseline A1c’s. The analysis raised a few clinical questions, assuming pumps get better coverage in type 2 one day: will patients who see large benefits on pump or CGM be able to keep their devices indefinitely, particularly in more cost-conscious health systems, or will they be asked to stop using pump/CGM therapy as soon as they are at an acceptable A1c or time in range? If they do stop using the devices, would their control return to its original level? 

First Looks at New Devices From Unomedical, Senseulin, and Cam Med

• Unomedical demoed its all-in-one, fully disposable, hidden needle insertion set device, sharing that it will launch “after” this summer and first come to market in an exclusive partnership with an undisclosed Luer Lock pump. The product looks outstanding and should offer strong competition to Medtronic/BD’s MiniMed Pro-set. Unomedical also has a novel catheter (Lantern) reportedly expected to launch in 3Q17, adding several slits along the side that allow insulin to flow out of multiple places (in case of occlusion or bending; watch this video). We’re happy to see innovation starting to ramp up in infusion sets – there is so much room for patients to have better pump experiences with more intuitive and less intimidating insertion and more robust catheters.
• Sensulin, founded by ex-Amylin scientists and investors, is developing a once-daily glucose-responsive insulin. A first-in-human trial is expected in 2018. The company hopes to validate its first clinical candidate this year. The liposome-based insulin uses recombinant human insulin and is glucose responsive via boronate linkers (Agglomerated Vesicle Technology, or AVT). The links cleave in response to glucose, thus regulating insulin release (Dasgupta et al., PLOS One 2012). The company has been working to reduce AUC for meal challenges, though the rat data looked encouraging for a once-daily insulin that covers basal and blunts some prandial excursions. At minimum, CEO Mike Moradi expects a “safer basal insulin” that can meet “some portion” of mealtime needs for many type 2s. For a subset of type 1s, he said, the insulin may alleviate the need for MDI.
• Cam Med has developed a very thin, flexible, flat patch pump (3 x 2 x 0.2 inches) that uses a unique reservoir array and electrolysis drive system. The three-day pump is disposable (300 units), uses a handheld controller, and is expected to have lower manufacturing cost and be capable of delivering multiple medications. Instead of a single reservoir, the Evopump uses many reservoirs arranged in an array – each reservoir is rigid, but the spaces between them are flexible, allowing the pump to bend. Medication is delivered through each reservoir using electrolysis – the gas expansion pushes a membrane and moves the medication. The first mass producible prototype of the reservoir/pump array was just manufactured, and the company owns the IP on the elements that enable the thin, flexible design. Cam Med has won several competitions and secured $440,000 in non-dilutive funding so far; the plan is to partner with an incumbent for commercialization.

1. Automated Insulin Delivery

Oral Presentations

Feasibility Of Omnipod Hybrid Closed-Loop Control In Adults With Type 1 Diabetes Using A Personalized Model Predictive Control Algorithm

Bruce Buckingham, MD (Stanford University, Palo Alto, CA)

Stanford’s very highly-regarded Dr. Bruce Buckingham presented positive data from Insulet’s first study of its OmniPod Horizon Automated Glucose Control system (36 hours inpatient; n=24 adults with type 1): With an 80% meal bolus before announced meals containing 30-90 grams of carbohydrates (see top picture below), mean glucose rose by ~10 mg/dl (151 mg/dl to 161 mg/dl) vs. sensor-augmented pump (SAP), but time <70 mg/dl fell an impressive 90% from 5.4% to 0.5%. Fasting blood glucose declined from 160 mg/dl on SAP to 137 mg/dl on the closed loop system, and nocturnal glycemic variability (as measured by standard deviation) was cut in half (!) from 48 mg/dl to 24 mg/dl – very impressive. When a full 100% meal bolus was given (n=10; see second picture below), performance was similar, with mean glucose rising ~12 mg/dl (143 to 155) on closed loop, but time <70 mg/dl falling from 5.7% to 0.6%. Mean time in range (70-180 mg/dl) during the 787 hours of system use was 69% overall (90% during the overnight period), and there were on average 0.3 hypoglycemic events per day. Percent time <70 mg/dl was just 0.5% total (0.04% overnight), and percent time >180 mg/dl was 30% overall (9.9% overnight). Overall, this initial feasibility study of Insulet’s in-development commercial system was a success – the algorithm was safe during the day and night, very effective at avoiding hypoglycemia, and highly effective at night. The obvious area for improvement is in reducing overall glucose, particularly after meals – Dr. Buckingham agreed that tighter control is needed, and that Insulet needs to do some work to reduce excursions after meals. (Of course, this was also the story of the MiniMed 670G’s development, something Insulet Medical Director Dr. Trang Ly knows well.) Yale’s Dr. Stu Weinzimer pointed out in Q&A that even after a 100% bolus, the system didn’t handle meals that well. Dr. Buckingham acknowledged that the system needs to be improved, but attributed part of the deficit to the fact that many of the participants are not accustomed to eating breakfast – a number of study participants usually have a café latte for breakfast at home, while the study included a minimum of 30 grams of carbs. Additional studies of the system, expected to launch in late 2019 (as of the November 2016 Investor Day), are underway in both adults and pediatrics.

80% Meal Bolus

100% Meal Bolus

• Dr. Buckingham showed two sample patient traces – both showing post-prandial excursions, more severe in one patient than another. Subject 2007 (left) had “one of the better profiles,” with minimal hyperglycemia following meals, 98% time in range (70-180 mg/dl) and a mean glucose of 121 mg/dl. Subject 1005 (right) had more significant hyperglycemia following meals, a mean glucose of 168 mg/dl, and time-in-range of 61%. As Dr. Buckingham previously stated, it is clear that the algorithm could stand to be more aggressive, though this is a good start. 

• As a side note, Dr. Buckingham acknowledged the strength of former mentee and current Insulet VP & Medical Director Dr. Trang Ly: “I really miss her from Stanford. My loss, their gain.” We agree that Insulet is in good hands – Dr. Ly is as knowledgeable on automated insulin delivery as they come! We salute Dr. Buckingham for training (formally and informally) so many researchers and clinicians over the decade – he is so beloved in our field and we know all his mentees have felt equally if not even luckier to have worked with him.  

On Blood Glucose During 3 Days With Intensive Physical Exercises And 3 Days With Gastronomic Dinners: Randomized Crossover Trials

Sylvia Franc (Centre Hospitalier Sud Francilien, Évry, France) and Hélène Hanaire (University of Toulouse, France)

Back-to-back Cellnovo-partnered Diabeloop presentations showed that the hybrid closed loop system handles exercise and meals – two of the biggest hurdles of closed loop therapy – very well. In the first crossover study, 14 participants performed a specified workout regimen (ranging from 30-45 minutes and 50-75% VO_2max) for three days on open loop (SAP) followed by three days on closed loop with exercise announcement (or vice versa). Over the three-day period, the Diabeloop algorithm increased time in range (70-180 mg/dl) from 68% to 80% (p=0.002), did not significantly change already-low time below 70 mg/dl (2.5% to 2.1%), and more than halved time >300 mg/dl (8.7% to 3.8%; p=0.03) and >250 mg/dl (2.3% to 0.4%; p=0.01). Average blood glucose over the three-day period was also decreased with closed loop from 156 mg/dl to 138 mg/dl (p=0.037). In terms of patient-reported-outcomes, there were trends toward higher levels of satisfaction (p=0.565) and greater ease of use (p=0.0526) with closed loop, and participants felt more comfortable with Diabeloop vs. SAP (p=0.0251). In the second crossover study, 10 participants were challenged with three days of challenging meals (Japanese on day #1, French on day #2, and “Italian pizza-tiramisu” on day #3) on open loop (SAP) first and then closed loop (or vice versa). During the nights after the gastronomic dinners, Diabeloop greatly improved time in range: Time between 80 mg and 140 mg/dl nearly tripled (22% to 60%; p<0.0001), time between 70 mg/dl and 180 mg/dl increased by ~75% (49% to 86%; p<0.0001), time >180 mg/dl was reduced by ~70% (45% to 13%; p=0.005), and there was no significant change in time <70 mg/dl (5.7% to 1.6%; p=0.1281). These improvements were seen over the entire three-day period as well: time between 70 mg/dl and 180 mg/dl increased from 54% to 81% (p<0.0001), change in time below 70 mg/dl trended toward significance (3.9% to 1.4%; p=0.053), time above 25o mg/dl and 300 mg/dl were both significantly reduced, and mean glucose dropped from 170 mg/dl to 142 mg/dl (p<0.0001). Next up for Diabeloop and Cellnovo on the way to an expected 2018 EU launch (according to Cellnovo) is a three-month home study (n=60; 12 centers), expected to reach primary completion in October 2017.

• Both of these studies were performed in very well-controlled populations with long diabetes duration. Hypoglycemia was minimal and time in range was high at baseline – it’s possible that the Diabeloop system would have improved glycemic profiles to an even greater extent had the study population been more like the average person with type 1. Dr. Helen Hannaire, who presented the meal study, said that the point was to enroll these very educated patients who “could really push the system.”
• As we understand it, Diabeloop aims to commercialize a closed loop device by developing its algorithm, integrating pump and CGM components, and selling the system (i.e., more like Bigfoot and less like TypeZero’s plan to license only the algorithm). Currently, the Diabeloop system leverages a Cellnovo pump, a Dexcom G4 sensor, a dedicated smartphone containing a proprietary algorithm (that builds on Cambridge’s Dr. Roman Hovorka’s MPC algorithm), and 24/7 therapeutic support. This support consists of continuous transmission of data to dedicated nurses along with automatic analysis of data with alerts in case of emergency, plus a round-the-clock telemedicine system.

Moving Forward with Technology

18 Months Use of Closed Loop at Home Study

Satish Garg, MD (Barbara Davis Center, Aurora, CO)

Dr. Satish Garg shared positive MiniMed 670G continued access phase data from Barbara Davis Center patients (n=19) that have now used the hybrid closed loop system for over a year. There were three big takeaways from the results: (i) the study phase improvements in A1c, time-in-range, hypo/hyperglycemia, and glycemic variability have been mostly maintained out to one year in BDC patients (with some exceptions for adolescents); (ii) time in hybrid closed loop (auto mode) and sensor utilization have dropped off slightly in both BDC groups, particularly in adolescents (~2 less hours per day spent in Auto Mode at one year vs. the pivotal study phase); and (iii) BDC adults did better on the system than adolescents at one year. No severe hypoglycemia or DKA has been seen, and the fourth-generation sensors “were accurate.” Though these results are from just one center, they suggest mostly sustained improvements following the single-arm, three-month pivotal trial – that was certainly not a given and represents very good news for Medtronic and the field. On the other hand, the results also remind us that these patients were doing very, very well already at baseline. The drop-off in sensor and HCL usage in adolescents will be important to watch, as this group benefitted more from the system in the pivotal study. Overall, we see these as confirmatory results for the 670G, especially because adults and adolescents were still using hybrid closed loop for roughly two-thirds to three-fourths of every day at 12 months.

• We assume (but aren’t positive) that 670G continued access phase patients are also getting free supplies. If they had to pay to keep using the system, would the >80% enrollment in the continued access phase look different? We can’t wait to see patients’ reviews once this product is out!

		Baseline	Pivotal Study Phase	One Year
A1c	Adults Adolescents	7.6% 8.1%	7.1% 7.1%	7.0% 7.4%
% Time in Hybrid Closed Loop	Adults Adolescents	- -	77% 71%	74% 62%
% of Sensor Usage	Adults Adolescents	- -	91% 86%	85% 76%

		Baseline Run-In*	One Year*	P-value
% in 70-180 mg/dl	Adults Adolescents	68% 60%	73% 63%	P=0.025 P=0.19
% <70 mg/dl	Adults Adolescents	7% 2%	4% 2.9%	P=0.003 P=0.2
% >180 mg/dl	Adults Adolescents	25% 38%	23% 34%	P=0.5 P=0.16
Mean Glucose	All Subjects	157 mg/dl	158 mg/dl	P=0.96
Coefficient of Variation	All Subjects	34%	32%	P=0.003

*Results for these participants were only shown for the run-in and at one year, not for the three-month study phase. In looking at the pivotal data for all participants at all centers (see ADA 2016), these one-year results look very similar – suggesting the benefits were largely maintained.

• More MiniMed 670G three-month pivotal data was recently published in DT&T (Garg et al.) and mentioned a few times at ATTD, following the ADA 2016 poster, EASD 2016 oral, and two-page JAMA Research Letter. The paper is very data heavy and mostly adds more granular CGM data, broken up between adults and adolescents. Generally speaking, the results are very similar to the overall data we’ve previously seen – a good sign the device offered benefits in both groups, especially overnight. The benefits in adolescents after breakfast were also very striking, shown quite clearly in the paper’s glucose profiles (first shown at ADA).
  
  • One of the most important statements in the DT&T paper comes on the last page: “...the generalizability of our results may be limited, given the lower baseline HbA1c levels for both cohorts relative to T1D exchange mean HbA1c levels; the frequent contact that study subjects had with site personnel; the fact that approximately half of the adolescents and two-thirds of the adults were using CGM at baseline; and the exclusion of subjects with HbA1c levels >10%, >2 recent episodes of severe hypoglycemia, and those with any recent episodes of diabetic ketoacidosis.” This question of generalizability raises many questions as the 670G rolls onto the market this Spring in the US: What will the real-world experience be like? What users will like the system the most? What users may be disappointed or overwhelmed? Will the average HCP be able to prescribe the MiniMed 670G? Will payers cover the 670G, and if so, for whom? Will certain populations be excluded? How will patients use the system during the day vs. night? Will those eating fewer carbohydrates value automated insulin delivery more than high-carb users? (In Adam’s experience, auto-basal modulation + low-carb diet = fully automated insulin delivery without meal announcement.) On the other hand, will those who are less attentive to their diabetes like the 670G more?

Multi-Week Outpatient Studies With Adaptation

Frank Doyle (Harvard University, Cambridge, MA)

Harvard’s Dr. Frank Doyle shared data from a 12-week home study of closed-loop control (n=30) using a Zone MPC algorithm with adaptation (adjusting carb ratios and weekly basal rates), Dexcom’s G4, a Roche pump, and a smartphone running UVA’s DiAs platform. The trial met its primary endpoint, showing a 0.3% reduction in A1c at 12 weeks from a very low baseline of 7.0% (p<0.005). Interestingly, mean glucose increased (142 to 151 mg/dl; p<0.005), though, notably, time <70 mg/dl declined by more than 60% (5% to 1.9%; p<0.005) – part of the lesser time in hypoglycemia would, of course, drive average mean glucose up. As such, the “quality” of mean glucose would be improved, given lower hypoglycemia. Time in 70-180 mg/dl did not change significantly (~73%), while time >180 mg/dl increased very slightly (21% to 25%; p=0.06). The study had three cycles of four-week parameter adaptation, with carb ratio chanced once per cycle and basal insulin changed four times per cycle (weekly), and Dr. Doyle emphasized that the first cycle did most of the settings adjustment for the carb ratio. Dr. Yogish Kudva noted in Q&A that during the course of the study, 358 parameter adaptation recommendations were made by the algorithm, and the clinical team accepted 323 (an impressive 90%). We’re glad to see more discussion and testing of adaptive algorithms, since this has potential to further improve closed loop glycemic control further without needing new devices or drugs.

Closing the Loop

The International Diabetes Closed-Loop (IDCL) Trial: Progress Report

Boris Kovatchev, PhD (UVA, Charlottesville, VA)

Dr. Boris Kovatchev gave a progress report on the International Diabetes Closed Loop trial (to serve as Tandem/TypeZero’s pivotal), including first results from the 14-day training protocol phase and a picture of the integrated Tandem system. In 20 individuals thus far, the training protocol has seen similar outcomes to prior studies: an overall mean glucose of 156 mg/dl (148 mg/dl overnight), just 1.6% of the time <70 mg/dl (0.5% overnight), and 70% of the time in 70-180 mg/dl (79% overnight). The main phase of the trial (CT.gov posting here) will begin soon, randomizing 240 patients in a 2:1 ratio of closed-loop control vs. sensor-augmented pump therapy over six months. Notably, the FDA IDE approval actually includes a line that this can serve as a pivotal study to support a PMA, assuming the trial goes well and no safety issues arise. This is great assurance for Tandem/TypeZero and NIH. Dr. Kovatchev confirmed that the current system uses either a Tandem or Roche pump, Dexcom G5, and an Android smartphone running the inControl AP algorithm, with a planned upgrade in 6-8 months (3Q17) to add a Tandem t:slim X2 with the embedded TypeZero algorithm and Dexcom G6 sensor. We saw the first picture of the latter (see below), which utilizes a similar interface to the t:slim X2 with PLGS, includes new activity settings for “sleep” and “exercise,” and will be indicated for ages six years and up. The slide noted that upon approval, the product will be available as a software upgrade for all existing t:slim X2 users – a very compelling feature that we hope all pumps move to. We’ll expect more updates in Tandem’s upcoming 4Q16 call. 

Beyond The MiniMed 670g

Thomas Danne, MD (Kinderkrankenhaus auf der Bult, Hannover, Germany)

Dr. Thomas Danne shared the first clinical feasibility data (a small study, n=7) on the next-gen MiniMed 690G (what Medtronic actually calls “Advanced Hybrid Closed Loop”), adding DreaMed’s fuzzy logic algorithm to give automatic bolus corrections on top of the 670G hybrid closed loop with basal-only modulation. The 36-hour inpatient crossover trial compared the 670G and 690G and included five meals (one unannounced, one with a 70% bolus, and three with a full bolus), one snack, and one exercise session. Dr. Danne only showed glucose outcomes for the 690G: 71% time in 70-180 mg/dl (47% time in the tighter 70-140 mg/dl), 1.6% time <70 mg/dl, and a mean glucose of 153 mg/dl. These overall data were not compared to the 670G in this study, so it’s hard to know how much the auto correction boluses added. However, Dr. Danne did show one compelling example of an unannounced meal on the 670G vs. 690G, showing how the automatic correction boluses halved the time in hyperglycemia and shrunk the AUC (picture below) – excellent! He concluded that the MD Logic Bolus in the 690G algorithm is “promising” and will be further studied in a recently NIH-funded trial starting later this year. Medtronic has not shared official timing on when this product might come to market. (Editor’s Note: Though Dr. Danne used “690G, Medtronic subsequently informed us that this product is now referred to as the “Advanced Hybrid Closed Loop System,” and not the “MiniMed 690G.”)

MiniMed 670G (right) vs. MiniMed 690G (left) with an unannounced meal in one participant

Dual-Hormone Closed-Loop: Issues and Expected Benefits

Automated Glycemic Regulation with a Bionic Pancreas Delivering Insulin and Glucagon: The Boston Approach

Steven Russell (MGH, Boston, MA)

MGH’s Dr. Steven Russell shared more details on the NIH-Funded Bionic Pancreas bihormonal Pivotal Study, expected to start in mid-2018 and complete in roughly mid/late-2019. (Editor’s Note: As of late 2017, the pivotal trial is now expected to start at the beginning of 2019.) The trial will randomize 480 patients in three age groups (pre-adolescent, adolescent, adult) to the bionic pancreas (n=320) or usual care (n=160) – much larger than the adult-only (n=312) design we expected following the NIH announcement earlier this month. (Dr. Damiano confirmed with us that NIH has only funded the adult portion of the pivotal, but additional funding will be sought for the pediatric cohorts.) Notably, the trial will be powered for superiority to show both a change in A1c from baseline (decrease of 0.5%) and time <60 mg/dl (reduction of 60%) – wow is this team ambitious. At least 1/3 of subjects will be pumpers and at least 1/3 will be MDIs – a terrific move to ensure this automated system benefits far more than current pumpers. A PMA submission is expected after the six-month RCT, while a continued access study (n=107) will ensure there are 12 months of continuous exposure to glucagon (PMA supplement). An incentive study will allow the usual care cohort to continue on and use the Bionic Pancreas. The study is expected at 16 centers all across the US, including MGH (Dr. Russell), Stanford (Dr. Buckingham), UNC (Dr. Buse), University of Washington (Dr. Hirsch), Washington University in St. Louis (Dr. McGill), UCSD (Drs. Henry, Edelman, Pettus), Cleveland Clinic (Dr. Hatipolglu), and Henry Ford Medical Center (Ms. Kruger).

• As we noted earlier in February (following Novo Nordisk’s $5 million investment), Beta Bionics plans to submit IDEs for insulin-only and bihormonal iLet bridging studies by the end of May, start the studies by mid-July, and run them into September. The bihormonal bridging study will likely provide eight weeks of drug exposure to glucagon (presumably Zealand).

Industry Symposium: Approaching Physiology with Ultra-Fast Insulins

Advances in Diabetes Management Technologies

Bruce Buckingham, MD (Stanford University, Palo Alto, CA)

Dr. Buckingham wrapped up a Novo Nordisk-sponsored symposium with a birds-eye view of the closed loop landscape: “Night is the low-hanging fruit for hybrid closed loop – there’s no major disturbance with food and exercise. Every hybrid closed loop system has done that quite well, reducing highs, reducing lows, and reducing variability.” The next major decrease in diabetes burden, he continued, will be with more fully closed-loop systems, noting that systems will not be “perfect,” but they only need to be “good enough.” Well said. He noted that one of the keys to fully closing the loop is – as expected in a Novo Nordisk symposium – achieving a more rapid onset of insulin action, either via faster-acting formulations, faster delivery speeds, infusing over a broader subcutaneous space (e.g., perhaps with Unomedical’s Lantern or Capillary Biomedical’s catheter), or the use of adjunctive therapies. He provided brief commentary on five of the prominent closed-loop players (see below), hoped that a trial of faster-acting insulin aspart in closed loop is being planned (will algorithms need adjustment?), and remarked on the progress in the field over the last ten years: “This is really a new day. It’s a storied time we’re in here. I think the first ATTD meeting had ~400 people, all engineers, and now we’re talking about human factors.”

• TypeZero (“I’m excited to see where this goes in the next year”);
• Beta Bionics (“This has done very well, and a lot of people like the qualitative meal announcement”);
• Bigfoot (“Bigfoot has been stalking around ATTD, I’ve seen them. They don’t have any data here, but they’re moving forward”);
• Insulet (“What’s interesting is that the Dexcom CGM will talk directly to the disposable OmniPod pump, and the closed loop software will be right there in the pump. I just gave a talk on a feasibility study of the system this morning – post-breakfast glucose is a little high, but glycemic variability is significantly decreased”); and
• Medtronic (“Nighttime control is really good. Parents don’t have to get up, same for kids. Nighttime has been hugely different and vastly improved. During the day, kids are still counting carbs, bolusing – the work during the day is roughly the same”).
  
  • This point on daytime burden will be interesting to watch as Medtronic brings the 670G to market this Spring – people expecting to not have to think about their diabetes while on the system may be disappointed, unless they are eating very few carbohydrates (in which case, basal-only modulating can cope with meals without announcement). Expectation management will be key here, though hopefully Medtronic learned from the 530G launch.

Poster

Self-Reported Hypoglycemia Reduction in Tandem Pump Use Compared To Previous Methods Of Diabetes Therapy

Garrett Marin (Tandem, San Diego, CA)

Tandem presented a retrospective analysis (n=3,046) comparing CGM-based outcomes with the t:slim G4 (downloaded from t:connect) vs. previously published CareLink data for Medtronic sensor-augmented pump users (n=7,916; Battelino et al., Diabet Med 2015). As shown in the table below, Tandem found statistically significant lower rates of hypoglycemia in t:slim G4 users vs. the previous Medtronic publication – roughly half as much time spent <70 mg/dl (3%-3.6% vs. 6.3%-8.5%; p<0.001), translating to ~39-75 minutes less time in hypoglycemia per day with t:slim G4. We are excited to see the company using outcomes beyond A1c to characterize the benefits of its system – especially in a data set far larger than a typical device clinical trial. This effect was consistent regardless of sensor usage (<25% to >75%). t:slim G4 users wearing a sensor >25% of the time also spent significantly more time-in-range: ~23-81 more minutes per day. Results were mixed for hyperglycemia, as shown below. Overall, these data are an interesting exploratory analysis – we do emphasize that comparing across studies is challenging, even though Tandem ensured similar inclusion criteria (T1/T2 using an SAP, at least six months of sensor use, and more than 15 days of CGM use during the first six months). The press release suggests Tandem’s intuitive touchscreen interface may be responsible for the results, though we’d note the improvement could also be from the more accurate Dexcom sensor vs. Medtronic’s older Enlite. Interestingly, approximately half of the Medtronic users in the comparator publication were using low glucose suspend, so Tandem’s hypoglycemia results are more encouraging from that perspective. These studies are not directly comparable, however, as they occurred at different times (2015-2016 vs. 2011-2013), in different regions of the world (US vs. Europe), and Tandem’s user population is more likely to be early adopters in better control.

• A separate poster shared self-reported, retrospective hypoglycemia outcomes in a survey of Tandem users. Tandem pump users reported a 52% reduction in severe hypoglycemia compared to their prior therapy. Data were collected from patients with either type 1 (90%) or type 2 (10%) diabetes who had been using a Tandem pump for at least 1 year (n= 1,370). Participants voluntarily filled out an online survey with questions about the occurrence of mild hypoglycemic events (able to treat the low themselves), medium hypoglycemic events (needed assistance from someone but did not need to go to the hospital), and severe hypoglycemic events (needed assistance from a medical professional and/or needed to go the hospital). Questions focused on both the year before starting a Tandem pump (using previous method of diabetes therapy) and the past year using their Tandem pump. Even when controlling for CGM usage, Tandem pump users reported a 52% reduction in severe hypoglycemia, a 31% reduction in medium hypoglycemia, and a 15% reduction in mild hypoglycemia, compared to previous methods of diabetes therapy. Furthermore, there was a statistically significant reduction in ambulance rides due to severe hypoglycemia of 58% and in days spent at the hospital due to severe hypoglycemia of 50%. Again, the results must be interpreted through the lens of retrospective data (and in this case, self-reported), but we’d note that a severe hypoglycemia is more likely to be remembered. 

Corporate Symposium: Improving Patient Care with Continuous Glucose Monitoring: From Diagnosis to Artificial Intelligence

Let the Algorithm do the Work: benefits of Sensor-Augmented Pump therapy

Thomas Danne, MD (Kinderkrankenhaus auf der Bult, Hannover, Germany)

In an excellent review of predictive low glucose suspend technology, Dr. Thomas Danne shared data from the US pivotal trial of the MiniMed 640G with SmartGuard (Buckingham et al. DT&T, in press) and that patients who don’t override the algorithm tend to have better glucose profiles. The trial (n=80) consisted of hypoglycemia induction by basal escalation with the system’s “suspend before low” feature set to 65 mg/dl (similar to the 530G pivotal). Whereas the control group had hypoglycemia in 93% percent of cases following basal escalation, blood glucose levels below 65 mg/dl were avoided 60% of the time in the 640G group. Further, blood glucoses below 60 mg/dl and 50 mg/dl were avoided 68% and 81% of the time, respectively, in those on 640G. As a reminder, Medtronic is not releasing the 640G as a dedicated product in the US – the 630G started shipping in September, and the 670G hybrid closed-loop system is now projected for a full launch in May-October of this year (per JPM 2017) – but this data may still be useful in supporting user uptake/reimbursement of the 640G in Europe and highlighting the suspend-before-low feature that is built in to the 670G. Dr. Danne continued to discuss the recently e-published German MiniMed 640G user evaluation, a six-week trial during which 24 type 1 patients were assigned to either sensor-augmented pump (SAP) with SmartGuard (threshold of 70 mg/dl) or SAP alone. The paper found that SmartGuard didn’t significantly alter mean glucose, but measures of hypoglycemia (time under 70 mg/dl, area under curve, number of excursions) were all improved as expected. We continue to see this in studies of CGM and diabetes tech, where the average doesn’t change but lows and/or highs do improve. Investigators also looked at the way that patients interacted with the technology – patients had a tendency to manually resume insulin and/or take a carbohydrate after suspension, which was “not such a good idea in every case, resulting in elevated blood glucose levels.” Dr. Danne’s main message was to trust the algorithm,” and let it do the work. Dr. Rich Bergenstal echoed this sentiment in the following talk; he observed that patients who were the least active blood glucose-managers did the best on 670G in the pivotal trial because they simply trusted the algorithm.

2. Glucose Monitoring

Oral Presentation

Patients With Type 2 Diabetes Using Multiple Daily Insulin Injections Have High Adherence And Benefit From CGM: A Prospective, Randomized Controlled Trial

Richard Bergenstal, MD (IDC, Minneapolis, MN)

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. Overall, we’re very glad to see another major study of CGM in type 2 and wonder what can be learned from this data for future studies and product development. We compare this study to the type 1 results and REPLACE in the detailed commentary below.

• Compared to the type 2 DIaMonD results, the type 1 cohort of DIaMonD saw larger improvements with CGM: a 0.6% A1c advantage (-1% vs. -0.4%), 76 more minutes per day in range (a 12% improvement from baseline), and 22 fewer minutes per day <70 mg/dl (a 34% improvement).
• The type 2 results also tell a different story than Abbott’s six-month REPLACE study comparing FreeStyle Libre to SMBG in type 2s with a baseline A1c of 8.8% (ATTD 2016). Abbott’s REPLACE study disappointingly missed its primary endpoint – similar 0.3% A1c reductions with both SMBG and Libre. The most notable takeaway in that trial was actually the hypoglycemia data, which improved markedly with FreeStyle Libre overall, overnight, and particularly for dangerous hypoglycemia (<55 mg/dl). Relative to the control group, patients using FreeStyle Libre spent ~30 minutes fewer per day <70 mg/dl (p<0.001), ~13 minutes fewer per day <55 mg/dl (p=0.001), and ~8.5 minutes fewer per day <45 mg/dl (p=0.001). For the FreeStyle Libre group, these reductions equated to 55%, 68%, and 75% reductions in those respective zones from baseline to six months. Taken together, we think REPLACE and DIaMonD show CGM (in either continuous or Flash configurations) can drive meaningful improvements in hypoglycemia, time-in-range, and hyperglycemia in type 2s on insulin.

REPLACE-BG: A Randomized Trial Comparing Continuous Glucose Monitoring With And Without Routine Blood Glucose Monitoring In Adults With Type 1 Diabetes

Katrina Ruedy (Jaeb Center for Health Research, Tampa, FL)

Jaeb’s Katrina Ruedy presented long-awaited, positive results from REPLACE-BG, a 226-patient randomized T1D Exchange trial comparing use of Dexcom’s G4 CGM with and without confirmatory fingersticks (n=77 for CGM+BGM vs. n=149 for CGM-only) over 26 weeks. As we expected, the groups had near-identical outcomes by study end, with no significant difference in time-in-range, hypoglycemia, hyperglycemia, A1c, mean glucose, or coefficient of variation – see the table below. The outcomes were also not significantly different in subgroups based on age, type 1 diabetes duration, and education level, a good sign that non-adjunctive CGM use is safe across the board. Protocol adherence was excellent, with 91% of the CGM-only group and 95% of the CGM+BGM group wearing the G4 for >6 days/week (mean: 6.7 and 6.8 days/week). Participants used the very accurate Contour Next BGM, with fingersticks totaling just 2.8/day in the CGM-only group (two for calibration plus an additional one here and there) vs. 5.4/day in the CGM+BGM group. There was one severe hypoglycemia event in the CGM+BGM group and zero in the CGM-only group. Ultimately, these RCT results show clear non-inferiority and confirm the simulations Dexcom presented at FDA last July prior to the non-adjunctive FDA approval in December. We include below the training materials that informed non-adjunctive use in this study, and we’ll be interested to see how Dexcom rolls this claim out in the US. The company’s webpage dexcom.com/fingersticks already shares some of these recommendations in very clear, succinct text and pictures. We salute the T1D Exchange for conducting this impressively large and rigorous study, and we wonder if other companies’ hopes to get non-adjunctive label claims (e.g., Abbott, Senseonics) may benefit from this data.

	CGM-Only Baseline -> 26 Weeks	CGM+BGM Baseline -> 26 Weeks	P-Value
Mean Time 70-180 mg/dl	63% -> 63%	65% -> 65%	P=0.81
Time <70 mg/dl	2.9% -> 3.0%	3.6% -> 3.7%	P=0.95
Time >180 mg/dl	33% -> 35%	31% -> 31%	P=0.88
A1c	7.1% - > 7.1%	7.0% -> 7.0%	P=0.41
Mean Glucose	162 -> 162 mg/dl	158 -> 158 mg/dl	P=0.99

• Participants in the CGM-only (non-adjunctive) group were instructed to dose insulin and make management decisions based on the CGM glucose except in the following circumstances: for 12 hours after insertion of a new sensor; on a sick day; for four hours after taking acetaminophen; symptoms present suggestive of hypoglycemia but CGM glucose not low; 20 minutes after treating low CGM glucose if CGM glucose not rising; prior to giving insulin bolus when CGM glucose >250 mg/dl; fasting CGM glucose >300 mg/dl or >300 mg/dl for one hour.

Industry Workshop: Flash Glucose Monitoring: Real World Benefits Across the Clinical Spectrum (Supported by Abbott)

Real world benefits of Flash Glucose Monitioring in a multinational database

Ramzi Ajjan, MD, PhD (University of Leeds, UK)

Dr. Ramzi Ajjan presented compelling real-world data from >55,000 FreeStyle Libre users (also in a press release) suggesting that, when they scanned at higher frequencies, their A1cs were lower and they spent less time in hypoglycemia. De-identified data was collected over a period of 18 months (October 2014 – May 2016), and compiled to nearly 400 million glucose data points, 64 million scans, and 1.2 million fingerstick blood glucose readings – talk about Big Data! Through this mass of incredible data points came some strong trends: As scanning increased, (i) estimated A1c dropped 8.0% to 6.7% (see figure below); (ii) time spent with glucose <70, 55 and 45 mg/dl decreased by 15%, 40%, and 49% respectively; (iii) time >180 mg/dl fell from 10.4 to 5.8 hours per day; and (iv) time in range increased from 12.0 to 16.8 hours per day. In addition, users scanned 16.3 times per day on average, with some scanning as frequently as 50 times per day! We were not surprised by these findings in one sense– Libre is a great technology that makes obtaining glucose data less painful, and more frequent glucose data and trends gives actionable information to make therapeutic and behavioral changes to reduce highs and lows. What is notable here is real-world data from a 55,000-patient-strong cohort, especially because it confirms data from Libre’s RCTs. We hope this stands as further evidence (especially for payers) that patients derive significant real-world benefits from real-time glucose sensors, are willing to scan at high frequencies outside of a clinical trial setting, and can make appropriate therapeutic and behavioral adjustments.

• FreeStyle Libre users in this study had almost identical usage and fingerstick habits to when they were enrolled in the IMPACT trial (type 1). This demonstration was a major victory for Abbott and shows the RCT results are generalizing to early real-world use. In the six-month IMPACT study of FreeStyle Libre in type 1 patients, Libre users scanned 15.1 times per day, on average – similar to the 16.3 times per day shown in this real-world data. Furthermore, during IMPACT, patients decreased SMBG frequency to once every two days (i.e., for every 30 scans, patients performed a fingerstick only once). These ratios were almost identical to the real-world scanning to SMBG ratio (32 to 1 in this trial). This data validates the behavioral findings (relating to testing) in IMPACT, and demonstrates that the more patients use Libre, the better they do (dose response).
• The real-world A1c vs. scan frequency curve is a near-perfect continuation of the A1c vs. SMBG frequency curve established in Miller et al. 2013 (see image below). It makes logical sense that higher frequency glucose checking will lead to lower A1c, but the advantage of flash glucose monitoring is its convenience, ease, and lack of pain allow for levels of daily testing that would be unthinkable with a fingerstick.  

• We were surprised that hypoglycemia decreases so meaningfully (see red across the spectrum of testing frequency, implying that a majority of the users are on insulin. At the most, some users were spending more than 40 minutes per day <55 mg/dl! Abbott did not collect data on type 1s vs. type 2s in this cohort, but we know the technology has been very popular in type 1 (and might assume this was mostly type 1s). Abbott has never publicly shared the type 1-type 2 breakdown of Libre users, but we would be curious to find out.

• We would love to see additional data detailing change from baseline. What kind of control did patients start with on Libre and how did they change over time by the end of the observation period? It also remains possible that patients who scanned more are simply more engaged and were already in good control at baseline, and then maintained that control in the study. Perhaps a follow-up analysis will show whether starting on Libre led to decreases in A1c and hypoglycemia from baseline.
• As of the start of ATTD 2017, FreeStyle Libre (real-time) consumer version is under FDA review, with a US launch expected in the “second half of 2017,” back 2-3 quarters from the previously ambitious 1Q17 expectation. Abbott has filed BOTH an adjunctive and a replacement claim for the device with the FDA – a smart hedge to get to market faster. See our Abbott 4Q16 report for more details.

Poster

A Randomized Trial Comparing Continuous Subcutaenous Insulin Infusion Versus Continuing Multiple Daily Insulin Injections In Patients With Type 1 Diabetes Using Continuous Glucose Monitoring

Elena Toschi, MD, Tonya Riddlesworth, PhD, Katrina Ruedy, Craig Kollman, PhD, David Price, MD, Roy Beck, MD

A Dexcom 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). 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 and in those new to 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 some further value (over an hour more per day in range). Automated insulin delivery should be a killer app for both technologies, though it’s hard to say who will benefit the most and how it will compare to MDI+CGM+dosing advice.

Corporate Symposium: Improving Patient Care with Continuous Glucose Monitoring: From Diagnosis to Artificial Intelligence

CGM and Cognitive Computing: easing patient burden in daily diabetes management

Huzefa Neemuchwala, PhD (Medtronic, Northridge, CA)

Medtronic’s very smart Dr. Huzefa Neemuchwala (Head of Innovation, Diabetes Service and Solutions) ran attendees through the Sugar.IQ with Watson app, sharing that a “larger preview” to “a larger audience” will occur on February 28. As of CEO Omar Ishrak’s presentation at JPM last month, a full launch of Sugar.IQ was expected in May-October. This upcoming launch will make sure the infrastructure holds up and get more feedback on the patient experience. The app was first demoed last fall and rolled out to 100 MiniMed Connect users, and we imagine a lot of valuable feedback came in. Dr. Neemuchwala’s demo on the app built on those we’ve seen at recent conferences, reminding us of its compelling potential to tease actionable insights out of diabetes data, change behavior, and stimulate discovery and teaching conversations with HCPs. He showed a few new Watson insights that we do not recall seeing before (see below), all focused on pattern recognition around mealtime choices or hypoglycemia/hyperglycemia behaviors. This presentation emphasized that Sugar.IQ will pair with Medtronic’s Guardian Connect (standalone mobile CGM for MDIs), perhaps another reason why the full US launch has been delayed – Guardian Connect is still under FDA review, slated for a May-October 2017 launch (as of the most recent update prior to ATTD 2017). Dr. Neemuchwala shared a vision to make a “dynamic Wikipedia of diabetes knowledge” with models to drive patient engagement (curiosity), build habits (trigger, action, variable reward, investment), and offer personal guidance: How many carbs are in this slice of bread? Am I at risk for hypoglycemia tonight? If I eat this burger, what will happen to my blood sugar? I did not know that I experience morning highs? We love the concept and are reminded that the value of this app (or any other) rests on passive data collection paired with engagement – will patients keep using the app, logging meals and insulin doses, and viewing the insights?

• “I notice that you tend to go low after meals with >20 grams of protein.”
• “I see that you often go low between 12-3pm on Saturday. I will keep an eye on the patterns during this time as I get more data. Be aware as you plan your day.”
• “Way to go! Great! I noticed that you had only one nighttime low in the last month. Whatever you’re doing seems to be working very well.”
• “I see that between 6am-9am, your glucose often goes high (300+ mg/dl) after taking an insulin injection. This trend might be worth discussing with your physicians or dietitian during your next visit.”
• “After your glucose is high for more than 120 minutes, you then tend to go low. Be especially careful with corrections from hyperglycemia.”

Type 2 Diabetes and Technology

CGM In Type 2 Diabetes: Is It A Realistic Technology?

Irl Hirsch, MD (University of Washington, Seattle, WA)

The always-engaging Dr. Irl Hirsch made a strong economic case for the use of intermittent real-time CGM in people with type 2 diabetes. His arguments and calculations mostly stemmed from the landmark 2012 Vigersky et al. Diabetes Care paper which showed that, after 12 weeks (2 weeks on, one week off, repeat) of CGM use, A1c was 0.5% lower than in an SMBG group, and, 4o weeks later, the difference was maintained at 0.6% – an effect Dr. Hirsch dubs “CGM-behavioral memory.” So how cost-effective is CGM in these cases? Dr. Hirsch outlined two theoretical scenarios: In the “base case,” the assumption is that patients don’t use CGM after year one, so the effect of the treatment in Vigersky et al. are only applied for one year. In the “refresher scenario,” the assumption is that the patient uses CGM again the next year, so the glycemic benefits are maintained for an additional year. A later study from the Vigersky group (Fonda et al.) calculated that, in the base case of CGM use, the incremental cost effectiveness per life-year gained is $6,293, while the incremental cost effectiveness per QALY gained is $8,893. In the refresher scenario, these numbers come out to $9,319 and $13,030, respectively. These numbers are very, very strong given that a benchmark is often $50,000 per QALY – CGM QALY’s in this case are so cheap because the device is only used for eight weeks, but its impact lasts at least the year (assuming Vigersky’s data is generalizable).

• How does CGM stack up to other diabetes therapies? Dr. Hirsch calculated that eight weeks of CGM therapy spread out over a year in the Vigersky study would cost ~$1,721. Compare this to sitagliptin ($4,560/year; 0.6% A1c reduction), liraglutide ($9,563/year; 1.1% A1c reduction), exenatide ER ($7,728/year; 0.9% A1c reduction), canagliflozin ($4,920/year; 0.9% A1c reduction), and pioglitazone ($168/year; 1.6% A1c reduction, but at high baseline of 10.3%). To take it a step further, Dr. Hirsch calculated the cost per month per 1% drop in A1c, finding that CGM fell in between the expensive drugs (sitagliptin, liraglutide, and canagliflozin; ~$400-$640/month/1% A1c drop) and the cheaper drugs (glyburide, metformin, and pioglitazone; ~$3-$6/month/1% A1c drop), at $239/month/1% A1c drop. From an A1c-centric perspective, the cheaper drugs may be the most short-term, cost-effective way to achieve glycemic control. However, they can put patients at elevated risk of complications (particularly hypoglycemia) or be maxed out such that another therapy is required. Intermittent CGM can both protect against hypoglycemia and reduce A1c, and at a lower price point than novel orals and injectables. 
• As Dr. Hirsch pointed out, his presentation only included A1c as a measure, and didn’t consider the cardiovascular benefits seen in EMPA-REG and LEADER. If these were considered, we imagine that the value proposition of SGLT-2s inhibitors and GLP-1 agonists would look better for type 2 patients in this analysis. While CGM can help protect against hypoglycemia, hyperglycemia, and glycemic variability, the long-term benefits of cardiovascular protection plus glycemic control may shift the cost/month/QALY in favor of novel agents. It’s hard to say how payers view this and what horizons they consider valuable. On the other hand, CGM might have the edge for reducing severe hypoglycemia in a 1-2 year time frame, which may be more financially attractive to payers.
• In Q&A, Yale’s Dr. William Tamborlane and Dr. Hirsch debated the significance of improving A1c by “0.3-0.6%” in older patients with long-standing diabetes – according to Dr. Tamborlane, the lack of data suggesting that tight control at this age prolongs life or improves CV health means that the QALYs gained may be zero, so providing CGM would be a waste. Dr. Hirsch responded that close to one-third of US adults will have diabetes within the next five to ten years, and most of them will be older people with type 2. “If one develops type 2 in his 50s, 60s, or 70s, as opposed to 20 years ago, he’ll be around for at least 10-20 years, so if we don’t treat aggressively, we’ll be looking at more retinopathy, more nephropathy, more neuropathy, etc. It’s a good point about life span, but we still have to worry about complications. I don’t think many would agree that it’s ok to keep A1c at double digits in anyone.”
• Dr. Hirsch also touched on the literature behind professional CGM, flash glucose monitoring, and CGM in type 2s:
  
  • On professional CGM in type 2s, Dr. Hirsch said that data is sparse, but promising. There are nine studies published in peer-reviewed journals (three are RCTs, n=158 total). Each demonstrates significant A1c reductions, but more data needs to be collected before a health economic argument can be put forward.
  • For flash glucose monitoring and prolonged CGM in type 2, Dr. Hirsch pointed to Abbott’s REPLACE study and Dexcom’s DIaMonD (type 2 cohort). We covered the former in detail at ATTD 2016 (and it was just e-published in Diabetes Therapy in December), and see the DIaMonD write-up above. Dr. Hirsch concluded that the initial data for FGM in those using prandial insulin is encouraging for the hypoglycemia prevention, but more studies are required. Dr. Hirsch left the audience with questions: What is the benefit of FGM in those not on prandial insulin? What is the role of CGM in those using prandial insulin – can patients be more aggressive than seen in the DIaMonD study?

Insulin Pump Therapy In Type 2 Diabetes: An Individual Patient Data Meta-Analysis

John Pickup, MD, PhD (King’s College, London, UK)

Citing a lack of conclusive RCTs for or against pump use in type 2 patients, Dr. John Pickup (King’s College, London, UK) walked the audience through his group’s individual patient data meta-analysis (Diabetes Care, in press), which found that pumps are best and likely most cost effective in type 2 patients with elevated A1c and insulin doses even after insulin optimization. Dr. Pickup and his team obtained all patient data from five trials (Raskin et al., Herman et al., Wainstein et al., Berthe et al., and Reznik et al. (OpT2mise)) and created a single large data set (n=590 participants; roughly half on MDI and half on CSII). Data on the individual participants of the selected trials was requested from the research teams, at which point Dr. Pickup could perform a one-step meta regression of covariates on outcomes to determine which characteristics affect the efficacy of pump therapy in type 2s to the greatest degree. Unsurprisingly, A1c reduction in pump vs. MDI was greatest in those with elevated baseline A1c, and reduction in insulin requirements with pump vs. MDI was greatest in those with highest baseline insulin requirements. The A1c effect size was markedly larger in the OpT2mise trial than in the overall data set (-1.1% in the pump group vs. -0.4% in the MDI group), perhaps due to the higher baseline A1c of 9.0%. This mirrors Dr. Pickup’s work in CGM, where patients also see more benefit the higher their A1c. The analysis raises a few clinical questions, assuming pumps get better coverage one day: will patients who see large benefits on pump or CGM be able to keep their devices indefinitely, particularly in more cost-conscious health systems? Or will they be asked to stop using pump/CGM therapy as soon as they are at an acceptable A1c or time in range? If they do stop using the devices, would their control return to its original level? 

Industry Symposium: Leading The Way in CGM Technology Innovation, Clinical Outcomes and Closed-Loop Automated Insulin Delivery

Introduction of the Next Novel Dexcom CGM Technology Platform

Jake Leach (Dexcom, San Diego, CA)

Dexcom SVP of R&D Mr. Jake Leach excitedly reviewed aspects of Dexcom’s pipeline: The Android version of G5 has launched in Europe and South Africa, G6 is currently in a pivotal trial and will be submitted for regulatory approval upon completion (for an expected 2018 launch, per the 3Q16 call), and factory calibration is in the works (the first G6 product with Verily, launching in 2H18 and then G7 to follow). We’ll be interested to see how the international launch of G5 mobile for Android is going OUS, along with the move to no longer require the receiver component – revenue shot up 55% YOY in 4Q15 when G5 launched for iOS, and Android-compatible smartphones claim up to 60% of market share in some major European markets. Mr. Leach said G6 will be 10-day wear and one calibration per day after startup, though Mr. Leach noted at least four times throughout his talk that Dexcom “is on a quest” or “very close” to eliminating fingersticks. To back up this claim, he showed the 10-day pre-pivotal (n=49) data first presented at DTM in November: 8.1% MARD with one calibration and 8.8% MARD with zero calibrations.

Industry Workshop: What Do We Need Beyond a Sensor to Make the Use of CGM Safe and Effective for Patients? (Supported by Roche Diabetes Care)

The new Accu-Chek Insight CGM System – First Patient Experience

Raimund Weitgasser, MD (Clinic Diakonissen, Salzburg, Austria)

Dr. Raimund Weitgasser shared that the Roche Accu-Chek Insight CGM launched in the Netherlands, Sweden, Norway, and Denmark via specialized diabetes centers. We heard that the launch was successful in Roche’s 4Q16 earnings supplementary materials, but this was a definitive confirmation of the specific geographies. No further details were shared on the launch, though Dr. Weitgasser reviewed the accuracy (MARD: ~10.6%) and lag time (~5 min) shared at DTM and gave a light pipeline teaser: The system will be evaluated “quite extensively” in the next few years, and Roche will perform experiments to examine the sensor’s stability over 14-15 days (it is currently intended for seven-day wear with two calibrations per day). See our EASD coverage for a deeper dive into the CGM – how does it stack up to other options? – along with pictures. We are glad to see Roche investing in CGM, since more competition with Abbott, Dexcom, and Medtronic should drive even faster innovation. 

4. Insulin Dose Titration and Delivery Devices

Advisors

Glucommander Outpatient, A Cloud-Based Insulin Management Solution, Titrated Patients to Goal In 12.5 Days And Sustained A 2.5% Drop In Hba1c Over 6 Months

Bruce Bode, MD (Emory University, Atlanta, GA)

For the second time this meeting (see first here), Dr. Bruce Bode presented data from Glytec’s Glucommander clinical decision support insulin titration system – this time in the outpatient setting and demonstrating a sustained 2.6% drop in A1c over a six-month period. Intermittent three-month data was presented in an ADA poster – from a high baseline A1c of 10.3%, patients ended three months with an estimated average A1c of 7.6% (p<0.000001). The study enrolled 42 type 1 and type 2 patients (one additional patient compared to the first three months). Impressively, patients were titrated to goal in 11 days. Over six months, A1c dropped from 10.2% to 7.6% (a “highly significant” 2.6% decline), and hypoglycemia was “extremely rare” – just 0.2% of blood glucose values were <54 mg/dl and 0.02% of values (just five instances) were <40 mg/dl. In a planned sub-analysis, all very engaged patients (those who input at least four blood glucoses/day) finished with A1cs below 7%. The total daily insulin dose increased slightly with Glucommander, though this increase was not significant. An impressive storyline from this study is time-savings – though unquantified, the small clinic staff (13 providers in total) was able to bring these poorly-controlled patients into range with minimal effort. Dr. Bode stated that “obviously this does work,” but unfortunately the weakness is that this is a non-randomized trial. Still, starting in the near future, over 10 hospital systems will launch the outpatient product. We believe insulin dose titration software, like Glucommander and many others, will be a major trend in the field over the next five years – better outcomes at lower costs in the toughest patients is a slam dunk. The real questions here lie with implementation, engagement, and business models: how will they be deployed (HCP-facing in an EMR, in a data platform like Glooko, on a patient’s phone like Voluntis), will HCPs and patients use them, who will pay for them, and how hard is the regulatory path for patient-facing basal-bolus/pump settings titration apps?

Dreamed Advisor Automates Insulin Pump Settings Adjustments With Data From Glooko’s Diabetes Management Platform

Revital Nimri, MD (Schneider Children's Medical Center, Petach Tikvah, Israel)

Dr. Revital Nimri shared interim results from the MD-Logic Advise4U Pilot study of DreaMed’s AdvisorPro clinical decision support software to optimize insulin pump settings. Thus far, the hypothesis of non-inferiority has been validated: For six weeks, patients at Schneider Hospital assigned to either the Advisor (n=7) or the control group (expert physician-guided decisions; n=8) have spent similar time in range (52% in control vs. 50% with the Advisor), though the Advisor group spent slightly more time >180 mg/dl (42% control vs. 49% Advisor), and slightly less time <70 mg/dl (7% control vs. 2% Advisor). The clinicians and Advisor have used different methods to achieve similar outcomes, an interesting quirk of clinical decision support. Notably, the Advisor has made more adjustments, making 4.7 recommendations per patient while clinicians made 3.3 per patient. The Advisor also changes with more parameters than clinicians: The number of basal periods per day was not really changed by the physician, while the Advisor appears to increase the number of periods, moving toward less basal insulin in favor of boluses. Physicians didn’t adjust correction factors meaningfully, while the Advisor changed correction factor value by an average of ~30% during the day and ~50% at night (both in favor of higher insulin dose, in concert with lower basal doses). The Advisor also appears to suggest more periods for the insulin:carb ratio. The study will enroll an additional ~20 patients and is expected to wrap up in a year, and a multi-center, multi-national study with a similar design and funding from the Helmsley Charitable Trust is being planned now. The Advisor will eventually be incorporated into Glooko’s platform, pulling data directly from all the major pumps and making suggestions accordingly. 

• A small study detailed in a separate talk showed that within a practice, physicians only agree on ~40% of insulin titration decisions. This number is astoundingly low, and makes the case for a software-based algorithm like AdvisorPro that can take some bias out of the process and add highly objective data analysis. We see extremely high potential here to save providers time and drive to more nuanced insulin recommendations.
• It is interesting to see similar outcomes given different dose titration strategies. The Advisor pulls a number of levers, while providers tend to isolate variables (or have a preference in favor of changing some over others) so as to reduce the degrees of freedom associated with a complex decision. We also noticed this earlier in the week in Medtronic’s Professional CGM session – an audience rarely agrees on what insulin dose changes will fix the problem, often because the problems are numerous and the eye is drawn to outliers. Might the Advisor’s intricate strategy show superiority over longer periods of time in less experienced clinicians? We think so.

Automated Clinical Decision Support Systems

Computerized Software Using POC Glucose In The Hospital And Outpatient Setting

Bruce Bode, MD (Emory University, Atlanta, GA)

Dr. Bruce Bode presented positive retrospective glycemic data (February 2013-November 2016), as well as cost-savings, from inpatient use of the Glucommander electronic glucose management system. Patients (n=5,718) admitted with hyperglycemia to one of seven hospitals arrived at their prescribed target blood glucose (100-140, 120-160, or 140-180 mg/dl) in 0.8 days from a starting average of 261.7 mg/dl. Once at target, 67.9% of blood glucose readings and 68.5% of patient day values remained between 70-180 mg/dl. Hypoglycemia was very unlikely once target had been reached and in the next 24 hours, with just 0.0011% of time below 40 mg/dl and 0.0130% of time below 70 mg/dl, respectively. In addition, the financial proposition associated with Glucommander is compelling, with over $3 million in potential savings per 250-bed hospital per year due to shorter patient stays, reductions in point of care testing, reduction in cost of treating DKA, and reduction in cost of treating CABG (coronary artery bypass surgery) patients. This in-patient data, combined with impressive out-patient data presented in Glytec’s ADA 2016 poster and previous publications, reinforce that clinical and personal automated decision support are low-hanging fruits that can help patients stay in normoglycemic ranges and reduce costs. To illustrate the scalability of the system, Dr. Bode claimed that one clinic at Duke University used Glucommander to help manage 45,000 patients in 2016, and will likely double that number this coming year! In this transitioning healthcare climate, where hospitals are responsible for covering costs of within-30-day readmissions, incentives align to help guide patient care, both in the hospital and upon release.

Industry Symposium: Simple Insulin Infusion For People With Type 2 Diabetes
When Daily Insulin Injections Fail…It's Not The Insulin, It's The Delivery (Supported by CeQur)

Demonstration Of PAQ Clinical Performance Including CGM And PROs

Julia Mader, MD (Medical University of Graz, Austria)

Medical University of Graz’s Dr. Julia Mader presented new clinical data on the CeQur PAQ basal-bolus patch device in type 2 diabetes: In a single-center, observational study (n=17), A1c dropped 1.4% (baseline 8.5%; p<0.05) after 12 weeks on PAQ. This improvement came with little tradeoff, as there was no significant change in weight and zero severe hypoglycemia – by study’s end, most were sad when they had to give the PAQ device away and would’ve preferred to keep using it. In addition to A1c change, five of the patients were selected to wear CGM for the duration of the study and saw a highly significant (p<0.001) improvement in time in range after 8 weeks – increasing from 51% to 70% of time spent in 70-180 mg/dl. Notably, there was little to no hypoglycemia at either baseline or after eight weeks of PAQ, and postprandial spikes appear to have been attenuated by use of the device. For those participants not wearing CGM, every point in the seven-point SMBG profile was significantly reduced at 12 weeks vs. baseline, and mean blood glucose decreased from 181 to 161 mg/dl, and the time below 70 mg/dl was unchanged. Dr. Mader emphasized the ease of therapeutic initiation, sharing that 80% of the participants remained on the first basal rate that was chosen, while two required one basal rate change and two required two basal rate changes. There was a slight increase in total daily insulin dose, but she explained this as a by-product of the trial’s treat-to-target design (and obviously high A1c). Patients’ overall treatment satisfaction score significantly improved, as they were particularly pleased with the flexibility and convenience of PAQ – commonly cited advantages of tubeless insulin delivery. Though small and observational, this study confirms that PAQ could be effective, especially in type 2s, and we hope this device moves along to commercialization and manufacturing scale. We reported last April that PAQ was slated for a 2017 US launch, though Senior VP Mr. Jay Warner confirmed in late September that the company had not yet filed with the FDA (it was CE marked in 2012). There was no concrete timing update at ATTD, though Mr. Warner said that the company is making the “final iterations to make sure [it] can scale up to manufacture at volume and at an affordable price.” CeQur raised $100 million in Series C financing in September 2015 and announced a significant facility expansion last April.

Oral Presentations

Effect of Insulin Delivery on Glycemic Control

Jay Warner (CeQur, Boston, MA)

A fascinating oral presentation from type 2 patch delivery device company CeQur, utilizing data from the dQ&A USA Patient Panel, showed a correlation between A1c and insulin delivery type, with more discreet, easier to use, and convenient insulin delivery (i.e., pens and pumps vs. syringes) associated with greater levels of glycemic control. According to dQ&A Patient Panel responses fielded in 1Q16 from patients with type 2 diabetes using an insulin device (n=1,579), a relationship exists between insulin delivery device and clinical outcome: pump users were less likely to have an A1c >9% (12%) than their counterparts using pen only (20%), syringe alone (26%), and syringe+pen (30%; p=0.033). CeQur’s Mr. Jay Warner (a study co-author) hypothesized that this is a matter of therapeutic visibility and convenience: pumps (and pens to a certain degree) are more discreet and convenient to use than syringes, perhaps promoting better therapeutic adherence. Pump use was also associated with another previously-determined predictor of clinical success: being married. Cost is yet another critical factor; Mr. Warner emphasized that pump use can be costly, and typically occurs in patients with better insurance and higher incomes. Mr. Warner framed this work in the context of the stark room for improvement in insulin delivery. He pointed out that, as of 2012, 29% of the US population was on insulin, nearly 69% of whom were not at target – sobering statistics that raise the question of which factors characterize those patients on insulin who are able to achieve their glycemic goal. There seems to be no question that better insulin delivery devices are needed, and at the same time, better dosing guidance will help patients get to goal more efficiently and effectively.

Insulin Pumps

Insulin Pumps Prepared For The Future - Exploring Differences In Function And Accuracy

Ralph Ziegler, MD (Diabetes Clinic for Children and Adolescents, Münster, Germany)

Dr. Ralph Ziegler provided an overview of the in-progress IDS Comparative Evaluation Study, which investigates the dosing accuracy and functional differences between various insulin pump options. Sponsored by Roche, the study evaluates pumps (both durable and patch) from different manufacturers, assessing the following parameters: (i) insulin dosing, for both basal and bolus; (ii) speed of insulin delivery; and (iii) timeline of the occlusion alarm.

• The study is not yet completed, but Dr. Ziegler whet the audience’s appetite with preliminary data on the bolus dosing accuracy of six pump models with different infusion sets. Twenty-five successive boluses were delivered by each pump and individually weighed on a microgravimetric scale to determine the average dose and dose variability. This was repeated for boluses of 10, 1, and 0.1 units of insulin, with nine rounds of testing per each combination of insulin pump and infusion. Across all pump/infusion set combinations, dosing accuracy was fairly strong for the 10 IU bolus, with a maximum deviation of only 8% from the target dose. Dosing accuracy was notably worse for the 1 IU bolus (maximum deviation = 42%), and particularly weak for the 0.1 IU bolus (maximum deviation = 64%), signaling a need for improved dosing capability at these small bolus amounts.
• Despite the fact that all devices were more accurate at higher bolus amounts, different pump/infusion set combinations emerged as having the best relative accuracy (i.e., smallest deviation from target dose) at each of the different bolus doses assessed. These preliminary results are blinded until the full study is completed so Dr. Ziegler could not share which manufacturer’s pumps were most accurate at each of the different bolus doses. Nevertheless, this analysis suggests that there is no one “most accurate” pump on the market but instead a spectrum of pumps that are more or less suitable depending on a patient’s particular insulin requirements. For instance, the best pump for a child with very low basal insulin requirements is not the same as the best pump for an adult.
• We remain mindful of Dr. John Pickup’s critique of comparative pump studies from DTM last fall. Foremost, he warned that pump accuracy and precision study results may be method dependent – meaning that results obtained through the IDS Comparative Evaluation Study’s technique of measuring insulin by its microgravimetric mass may not agree with results obtained from an alternative method, such as measuring insulin coming out of the pump by volume. Dr. Pickup also argued that the clinical significance of varying pump accuracy is unclear, save for the case of children and highly insulin sensitive individuals with lower insulin needs vs. the general population of pump users. Indeed, given the current fragility of the pump field, the accuracy and precision of different pumps may not be the highest priority item on the list of barriers to adoption.

5. Connected Care, Data, and Additional Topics

Automated Clinical Decision Support Systems

Further Optimizing Diabetes Control With Big Data

Aaron Kowalski, PhD (JDRF, New York, NY)

In the conference’s opening talk on “Big Data,” a fired-up Dr. Aaron Kowalski (JDRF) highlighted that we rarely use ANY data in diabetes – and the technology exists to solve this problem now. He shared a great “near-term” list of goals and questions in diabetes data (see tables below), highlighting the importance of “universal” CGM use in type 1 (24/7) and type 2 (intermittent); computer based dosing recommendations for pumps and MDI; capturing all insulin dosing data; real-time data flowing into the cloud directly from devices/phones; and better use of device data in population registries. We loved seeing him emphasize CGM in type 2, algorithms to titrate insulin, and pushing the entire room to solve these problems now. He showed a few pictures from the DIY community, who has already solved many of these issues – e.g., auto-tuning basal rates with OpenAPS and bolusing from an iPhone/Apple Watch with Loop. “People sitting in their garage are doing this. I know some companies will say, ‘It’s the FDA and regulations,” but this can be done and we have to work together to get over this hump.”

Key Near-Term Goals Universal Use of CGM in Type 1 Diabetes Universal Intermittent Use of CGM in Type 2 Diabetes Computer-based insulin dosing recommendations: MDI (I/C ratio, long-acting dose optimization and timing) and pumps (basal rates, I/C ratio, ISF) Increased automation of insulin dosing Capture of EVERY insulin dose into the body Real-time data flow to the cloud Population/registry capture of blood glucose, CGM, insulin pump data Cell phone integration with diabetes devices

Key Questions – aka, My Most Controversial Slide How do we expect better glycemic outcomes when we measure glucose so infrequently? (CGM is critical all of the time for type 1 diabetes and intermittent for type 2 diabetes.) How do we expect people with diabetes to achieve better outcomes with such an incredible amount of multifactorial and highly variable information with limited tools to interpret these data? How do we optimize insulin dosing strategies automatically? How in 2017, do we still have crudely determined basal rates, insulin to carb ratios, insulin sensitivity factors (correction factors), and many people on pumps with factory settings? Why can’t I bolus from my phone? Why isn’t the data automatically pushed to the cloud for all diabetes devices? Why aren’t we capturing population-based glycemic data in registries for big-data analyses?

Posters

Novel Medtronic Turning Point Program Improves Compliance and Hba1c In At-Risk Patients With Type 1 or Type 2 Diabetes

Ohad Cohen, MD, Edward Dick, MD, Sherwyn Schwartz, MD, Jennifer Knoulton, and Chantal McMahon, MD

Medtronic presented notable data on its “Turning Point Program” in a poster, demonstrating a mean 2% A1c reduction from a 10.1% baseline in a 50-patient pilot over five months (n=35 completers) in partnership with Methodist Health Ministries. Turning Point uses a Bluetooth-enabled BGM, a patient mobile app, a one-on-one health coach, patient reminders (BG monitoring, medications, appointments), clinical decision support for PCPs, and as-needed iPro2 professional CGM to help patients with uncontrolled diabetes. Pilot participants had a mean age of 51, were 97% type 2, 75% on insulin, and 100% Hispanic. Notably, 83% of patients experienced an improvement in A1c, 80% of patients viewed educational material, and mean BG frequency was only 1.7 checks per day – that suggests this is more about accountability to a coach, backend analytics, and education than driving intensive device usage. Program retention was 70% (strong considering the high baseline A1c’s), with an average of 3.4 weekly interactions per patient. This type 2-focused program falls in Medtronic’s Diabetes Service & Solutions segment, and we first heard an inkling of it at Medtronic’s 2016 Analyst Day (though not called by this name). Watch this very inspiring two-minute video and view the full poster here. We’re glad to see the focus on helping those most in need with hopefully lower cost interventions. Scalability is always a question with human coaches, though the program seems to be backed by serious analytics capacity. According to Medtronic’s F3Q17 slide deck, the Turning Point integration with IBM Watson is “now live,” adding further predictive analytics (e.g., predicting risk of a hospital readmission). The poster concludes that Turning Point will expand to include other health metrics and greater population base. We cannot wait to see this expand, given the serious potential to help a far broader population with a tightly integrated system of devices, coaches, and analytics.

• The Medtronic Turning Point program is intended to help patients with uncontrolled diabetes by facilitating patient engagement, remote monitoring and data-driven interventions. The pilot program, in partnership with Methodist Health Ministries (San Antonio, TX), utilized personalized coaching, technology and advanced analytics to engage patients, coordinate care, stratify risk and target resources while following the patient’s diabetes care plan and extending the physician’s reach outside the clinical setting.”

Significant Improvement of Blood Glucose Control In A High Risk Population Of Type 1 Diabetes Using A mHealth App, A Retrospective Observational Study

Marcus Hompesch, MD, Klaudius Kalcher, PhD, Frederik Debong, Linda Morrow, MD

mySugr and Prosciento (formerly Profil) presented a retrospective analysis suggesting a 1.3% estimated A1c reduction over six months in 440 randomly selected high-risk Logbook app users (baseline estimated A1c: 9.0%). To be in the retrospective analysis, patients had to have a mean baseline blood glucose of ≥183 mg/dl (estimated A1c >8%) and high engagement on the mySugr Logbook app (logging ≥5 days/week for ≥6 months). Mean blood glucose fell 18%, from 211 mg/dl at baseline to 173 mg/dl, an impressive drop. Both high blood glucose index (HBGI) and low blood glucose index (LBGI) improved (see the poster here). It’s great to see the company digging into data from its nearly one-million strong (!) user base, especially through this important research partnership with Prosciento to clinically validate digital health. We look forward to seeing more outcomes, especially when the Logbook app is couple with coaching. Separately, mySugr announced a few notable industry updates today:

• Following the first mention at JPM, two German payers will reimburse for mySugr’s Coaching and population health management: “We’re proud to be the first digital diabetes service that will be reimbursed by a German health insurer. BBKK and UKV policy holders will soon receive proactive coaching from diabetes educators based on their real-time therapy data thanks to mySugr Coaching.” The insurers cover over 2.6 million lives and are among the top 10 German private health insurance companies – whoa. At JPM, CEO Frank Westermann also mentioned this includes usage-based unlimited strips and a connected BGM for $850 per year (~$71 per month). The meter was not specified, but we assume it is Roche’s Accu-Chek Connect, a brilliant business move for both companies and a trend we see continuing to expand in BGM.
• Roche and mySugr will finally launch in the US this spring, and include the new Accu-Chek Guide BGM. As a reminder, this partnership was signed last April, though the US launch has taken longer than expected. mySugr users will be able to download a voucher in the app and redeem it for the new Accu-Chek Guide meter at local pharmacies. Once paired with the mySugr app, the Roche meter syncs data directly to it. Remarkably, mySugr’s user base of 900,000+ is up 50% from April 2016, when it stood at 600,000+. 
• mySugr updated its bolus calculator to improve post-meal calculations and offer easier setup – see the blog post here. As a reminder, this is approved in Europe as a Class IIb device and included in mySugr Pro; according to the blog post, the FDA is a “work in progress.” We see big potential for this in the US, given that most patients are not on pumps and most bolus calculator apps are not approved (and some are flat-out dangerous).

The Value of Connected Care in Diabetes (Sponsored by Lilly Diabetes)

Panel Discussion

David Panzirer (Helmsley Charitable Trust, New York, NY); Howard Wolpert, MD (Joslin Diabetes Center, Boston, MA); Steven Pacelli (Dexcom, San Diego, CA); Rick Altinger (Glooko, Palo Alto, CA); Adam Brown (Close Concerns, San Francisco, CA); Chris Bergstrom (Boston Consulting Group, Boston, MA)

A Lilly-sponsored panel titled “The Value of Connected Care in Diabetes” with Glooko’s Rick Altinger, BCG’s Chris Bergstrom, our own Adam Brown, Dexcom’s Steve Pacelli, Helmsley Charitable Trust’s David Panzirer, and Lilly’s Dr. Howard Wolpert identified a range of important challenges in connected care for diabetes:

• Developers often rely too much on feedback from early adopters, failing to reach the broader patient population. Dr. Howard Wolpert explained how uptake of early CGM was exclusive to technophiles, or those well-versed with technology and highly-engaged in diabetes self-management – the burden of using the technology was simply too high. Mr. David Panzirer argued that the problem persists: “we in diabetes are used to hearing from the vocal minority, but the great majority of people with diabetes in the US are adults not seen in an endocrinology clinic.” He emphasized that developers need to look beyond the “uber user,” beyond the “educated consumer” whose ears are perked waiting for the next product. Imagine the difference the field could make on a population level if we designed diabetes devices instead for patients who are truly struggling? Mr. Panzirer pointed to patients with double-digit A1c as an opportunity for targeted connected care technology – these are the individuals who add the greatest cost to the healthcare system, and it’s an easier task to get them down to 8% than to get someone with a 7.2% down to 6.5%. Notably, on this theme as well, Glooko CEO Rick Altinger announced that the company has just signed a contract to enter its diabetes management platform into 106 sites, mainly in primary care, to extend reach beyond those in an endocrinology clinic.
• Patients want to spend LESS time thinking about their diabetes, while developers often seek to maximize time customers spend in an app. Our own Adam Brown pointed out this irony, which explains some of the discord between “great apps” and “great diabetes apps” –  the metric for success for a diabetes app is “less time spent on diabetes, more time spent on life.” Added David Panzirer, “The second you ask patients with diabetes to do more, you lose a majority of them. Make devices easier to use. Something that asks you to take a picture of every meal is not going to gain mainstream traction.” Insurance and administrative hassle also adds diabetes hassle, something many panelists commented on too. Dexcom’s Steve Pacelli hoped that the “home-run” second-gen Dexcom/Verily CGM (flexible, bandage-sized, less expensive, 2020-2021 launch) would ideally be available for pick-up at a drug store, or for online purchase on something like Amazon Prime. We see huge potential in direct-to-consumer models that offer less insurance headache and less hassle overall.
•  The ideal technology design for type 1 and type 2 diabetes may be distinct. Adam broadly described type 1 diabetes as more “patient-driven,” while type 2 diabetes is often more “provider-driven” – that distinction (while far from perfect, and not to generalize!) does drive different design priorities for different groups. We certainly keep hearing that HCPs are drowning, and as we have written many times, there would be tremendous upside in providing better tools to titrate insulin, select the right therapy, spend less time on EMRs, etc. as many organizations are trying to do. Dr. Wolpert agreed that while there’s a lot of overlap between the two, there’s a big need for population management and surveillance in type 2 diabetes. The goal, of course, is for fewer to end up in the emergency room with severe hypoglycemia (to say nothing of kidney disease, cardiovascular disease, strokes, etc.) and expensive hospital bills that could have been avoided.
• The panelists offered up a number of quotable quotes on easing the burden of diabetes, reaching beyond early adopters, and technology for type 1 vs. type 2:
  
  • David Panzirer: “The second you ask patients with diabetes to do more, you lose a majority of them. Make devices easier to use. Something that asks you to take a picture of every meal is not going to gain mainstream traction.”
  • Adam Brown: “I want to spend less time on diabetes – less time spent in an app equals more time spent on life. The irony is that an app developer’s goal is usually to get people to spend more time in an app. That’s a funny conundrum.”
  • Rick Altinger: “It’s not about the data, but about data as a commodity. We need to produce actionable information that supports decision-making, where we’re helping clinicians and patients make better, faster, data-driven decisions to drive better outcomes.”
  • David Panzirer: “Quality of life is often thrown around as a buzz word, but there’s really something to it. Tie technology to minimizing the amount of time people spend thinking about their diabetes and to minimizing the risk of comorbid depression, because all of this has an impact on outcomes.”
  • Adam Brown: “At diaTribe and Close Concerns, we’re excited about models that are going to shortcut the annoying insurance headache and instead go direct to patient. Ordering your device on Amazon Prime – that’s really compelling.”
  • David Panzirer: “We in diabetes are used to hearing from the vocal minority, but the great majority of people with diabetes in the US are adults not seen in an endocrinology clinic. Don’t assume if you build it, they will come – you really need to go out and do the market research. The educated consumer is already on my product; I need to reach out to others.”
  • Adam Brown: “Early adopters often drive this field, which can be a mistake. Designing stuff for someone who’s already doing well is not going to do as much for diabetes as designing something for someone who’s really struggling.”
  • Dr. Howard Wolpert: “Early CGM technology required a fairly engaged technophile to derive benefit – the tradeoff between benefits and hassles wasn’t great, so it demanded a willingness to deal with the hassles of first-generation technology. Our current model still doesn’t really lend itself to what people with diabetes need most, but we have potential, with almost everyone having a smartphone in their pocket.”
  • Rick Altinger: “As we see more and more people with diabetes, technology has to take root in a primary care world.”
  • Steven Pacelli: “It’s on us as the manufacturer to develop low-cost, simpler products that expand the opportunity for 420 million people worldwide affected by diabetes – type 1 and type 2.”
  • Adam Brown: “This isn’t a perfect analogy, but I often characterize type 1 as a more patient-driven disease, while type 2 is more provider-driven – especially if you are a type 2 not on insulin, you may feel less ownership of your disease. So in type 2, developing tools for providers might be more critical there.”
  • Dr. Howard Wolpert: “Talking about type 1 vs. type 2 diabetes, there’s a lot of overlap, but in terms of a customized solution for type 2 it comes down to connectivity. Look at the patients most expensive to the system – what’s missing is some kind of surveillance, so that we don’t see as many patients coming into the ER with severe hypoglycemia.”
  • David Panzirer: “What companies really want is well-characterized patients that they can recruit into a clinical trial really quickly – type 1 research has benefited from the T1D Exchange. Registries play a huge role, and we could have well-characterized type 1 patients sitting there ready to enter trials.”

International Fair of New Technologies in Diabetes

In ATTD’s first-ever startup company presentation slot (nine companies), we were most impressed with Cam Med’s Evopump (“bandage-like patch pump”) and Sensulin’s work on glucose responsive insulin. We share details on these two early stage companies immediately below, followed by links to others that presented.

• Sensulin, founded by ex-Amylin scientists and investors, is developing a once-daily glucose-responsive insulin. A first-in-human trial is expected in 2018, and “big news” might come in about 30 days. The company hopes to validate its first clinical candidate this year. The liposome-based insulin uses recombinant human insulin and is glucose responsive via boronate linkers (Agglomerated Vesicle Technology, or AVT). The links cleave in response to glucose, thus regulating insulin release (Dasgupta et al., PLOS One 2012). The company has been working to reduce AUC for meal challenges, though the rat data looked encouraging for a once-daily insulin that covers basal and blunts some prandial excursions (picture below, though glucose still rose up to ~300 mg/dl). Avoidance of hypoglycemia looks very strong. At minimum, CEO Mike Moradi expects a “safer basal insulin” that can meet “some portion” of mealtime needs for many type 2s. For a subset of type 1s, he said, the insulin may alleviate the need for MDI. We’re glad to see the company is moving along, as we last recall hearing from Mr. Moradi at LyfeBulb’s 2015 Social Club event. Sensulin’s Chairman of the Board is Dan Bradbury (former Amylin CEO), and the Scientific Advisory Board includes past ADA president Dr. Alan Cherrington, Merck’s very impressive former Franchise Head of Diabetes and Obesity Dr. John Amatruda, CEO of Profil [now Prosciento] Dr. Marcus Hompesch, former Amylin CMO Dr. Orville Kolterman, and former JDRF CEO Dr. Alan Lewis.

• Cam Med has developed a very thin, flexible, flat patch pump (3 x 2 x 0.2 inches) that uses a unique reservoir array and electrolysis drive system. The three-day pump is disposable (300 units), uses a handheld controller, and is expected to have lower manufacturing cost and be capable of delivering multiple medications. Instead of a single reservoir, the Evopump uses many reservoirs arranged in an array – each reservoir is rigid, but the spaces between them are flexible, allowing the pump to bend (see picture below). Medication is delivered through each reservoir using electrolysis – the gas expansion pushes a membrane and moves the medication. Cam Med has seen ±2% accuracy 95% of the time from individual reservoirs, which it believes is more accurate than the current screw-driven pumps. The first mass producible prototype of the reservoir/pump array was just manufactured, and the company owns the IP on the elements that enable the thin, flexible design. Cam Med has won several competitions and secured $440,000 in non-dilutive funding so far; the plan is to partner with an incumbent for commercialization. Contact CEO Larry Alberts at larry.alberts@myevopump.com.

• Other presenting companies included:
  
  • Biomicro, a fully implantable CGM company. The device is expected to last six months, is about the size of a grain of rice, and is charged wirelessly via smart watch. Animal trials are expected this year. Contact: Nansheng.shen@biomicro.com.sg.
  • Jupiter Devices, a non-invasive glucose monitoring startup using RF in the microwave spectrum. The device is still in benchtop testing. Dr. Jessica Castle is an advisor.
  • Perikinetics (Thera Nova incubator), a fully implanted intraperitoneal artificial pancreas startup. The system includes both CGM and insulin delivery. Animal studies are planned in the upcoming year. Drs. Howard Zisser, Eric Renard, Frank Doyle, and Ananda Basu are advisors.
  • Ilya Pharma, who has a lactic acid bacteria-based wound healing product. A phase 1 clinical trial is expected this year. Contact: Evelina Vågesjö, CEO at evelinawagesjo@gmail.com.
  • Hci Viocare, a company developing smart insoles (FlexiSense). The device can detect pressure and shear forces in shoes (and other devices) and send the data to smart phones.
  • UND Life Sciences, developing a small lipid molecule for diabetic retinopathy. Preclinical studies are in progress, and the company hopes it can be delivered in drops instead of injections. Contact: undurti@hotmail.com.
  • Predictive Autoimmune Biomarkers, maker of PAB-ENDOSULIN, a biomarker predictive of loss of insulin prediction earlier than C-peptide.

Industry Symposium: Connecting Therapeutics, Technology and Outcomes: A New Vision for Diabetes Care (Supported by Sanofi)

Connecting Therapeutics, Technology and Outcomes: A New Vision For Diabetes Care

Tim Bailey, MD (UCSD, San Diego, CA)

Dr. Tim Bailey presented data from a Sanofi-sponsored head-to-head study of next-generation basal insulins Toujeo and Tresiba, which found that Toujeo results in 20% lower variability (i.e., a flatter PK/PD profile) vs. Tresiba when superimposing glucose infusion rate (GIR) curves for 0.4 u/kg daily doses – the treatment ratio was 0.8 in favor of Toujeo (p=0.047). Participants (n=48) were split into two cohorts, receiving either 0.4 u/kg/day (n=24) or 0.6 u/kg/day (n=24) of Sanofi’s Toujeo (insulin glargine U300) or Novo Nordisk’s Tresiba (insulin degludec U100). Each participant underwent two 30-hour clamps on day eight of each of two study periods. On days 1-7, they just took the basal insulin at 8 am (i.e., no clamp). After each phase, individuals entered a washout period before being switched to the other basal insulin treatment. This study design allowed for within-participant comparisons of the two products, and while there was no significant difference in primary endpoint between the 0.6 u/kg daily doses, Toujeo at 0.4 u/kg/day achieved a smoother GIR curve over 24 hours vs. Tresiba at 0.4 u/kg/day. Toujeo’s action was reported to achieve plateau-like insulin exposure between hour two and hour 16, followed by slight decline out to hour 30. Tresiba’s action was reported to increase from hour one to hour 10, when it reached peak and then declined out to hour 30 with little to no plateauing. The study also looked at six-hour insulin concentrations and glucodynamic activity as secondary endpoints: six-hour insulin concentrations were similar for both basals, but Toujeo showed a more even distribution of six-hour glucodynamic activity over 24 hours.

• Notably, this trial delivered basal insulin injections to all participants in the morning (8 am), pre-breakfast – timing that could possibly favor Toujeo over Tresiba. Toujeo hasn’t shown as flat a PK/PD profile over a full 24 hours compared to Tresiba in independent studies of each. Moreover, Tresiba benefits from a flexible dosing claim on its label, allowing patients to take their basal injection at different times on different days due to steadier-state action over a full 24 hours.
• Ultimately, Dr. Bailey acknowledged that a larger, longer-term trial is needed to determine the differential advantages and disadvantages of these next-gen basal insulins. We’re not sure if independent investment would be warranted in such a trial – both insulins are clearly better than the previous generation, and how they differ from each other is likely more about marketing than pressing clinical research questions. [On the other hand, we do think a study of basal dose timing – morning, midday, evening, bed – paired with dose titration software support would be very compelling.] The more important piece of this research is not pitting Toujeo and Tresiba against each other to determine a winner, but demonstrating that both of these next-gen products – “true” 24-hour insulins – are superior to the prior generation of basals, offering greater A1c-lowering and a better patient experience.

Partnering for Breakthroughs: What We Are Learning From Other Industries: The Onduo Experience

Andrew DiMichele (Onduo, Cambridge, MA)

Onduo’s Head of Product and Technology Mr. Andrew DiMichele (filling in for CEO Dr. Josh Riff at a Sanofi-sponsored symposium) announced that the joint Sanofi/Verily venture should have a product to show the world in ~one year. While he didn’t share much else in terms of product details, he discussed how the newly-formed Onduo is designed to beat the common challenges blocking other digital health companies. While Silicon Valley-based digital health companies boast tremendous entrepreneurial spirit, Mr. DiMichele argued that they often lack meaningful experience in healthcare. To address this, Onduo brings together experts in technology (from Verily) with experts in healthcare (from Sanofi). Capital constraints are another hurdle for many digital health companies, but Onduo was launched with a nearly $500 million combined investment from Sanofi and Verily. Lastly, Mr. DiMichele explained how “point solutions” are an attractive goal for digital health companies, promising a seemingly quick and easy path to profitability. A successful commercial digital health product, however, must be integrated within the overall healthcare system. As such, the Onduo team is keeping all important players in mind – patients, HCPs, diabetes educators, regulators, payers, etc. – in creating something that will empower patients to make better decisions day-to-day. Mr. DiMichele confirmed that the initial focus for Onduo will be type 2 diabetes, though the company aims to later expand into type 1 diabetes management. This talk was absent details on what the “product” will look like, but we look forward to hearing more by ATTD 2018!

• If you recognize Mr. DiMichele’s name, it’s because he was one of three co-founders and former CTO of Omada Health. With this incredible background, he brings strong diabetes-related digital health experience and experience with alternative outcomes-based pricing models as well as prevention expertise in general. 

Industry Symposium: Approaching Physiology with Ultra-Fast Insulins (Supported by Novo Nordisk)

Clinical Pharmacology of Ultra-fast Insulins in Pumps

Eric Zijlstra, PhD (Profil Institute, Neuss, Germany)

At a Novo Nordisk-sponsored symposium, Profil Institute’s Dr. Eric Zijlstra spoke to the significance of the company’s Fiasp (faster-acting insulin aspart) in the evolution of insulin therapy: The next-generation, ultra-fast insulin results in fewer, smaller postprandial glucose excursions and could be especially advantageous for pump users. He reviewed data from a clinical pharmacology study of Fiasp which found that the advanced agent works significantly faster than NovoLog (insulin aspart) – onset of exposure came 11.8 minutes faster, insulin exposure within the first 30 minutes was 3x higher, and insulin action within the first 30 minutes more than doubled. Fiasp achieves maximum concentration 25.7 minutes faster vs. NovoLog, and maximum action 18.7 minutes faster – that should have a very solid impact at mealtime, given the importance and challenge of pre-bolusing. Importantly, offset of Fiasp also occurs 24 minutes faster vs. NovoLog, making for a more effective faster-in, faster-out bolus insulin. This will be particularly pertinent for use in insulin pumps, according to Dr. Zijlstra, as faster absorption of a bolus is a key next step for advancing the field – particularly with regards to closed loop therapy, which will almost certainly improve with Fiasp’s profile. Fiasp was EMA-approved in early January 2017 and a primary focus of Novo Nordisk’s ATTD exhibit hall presence (see below). The insulin still awaits US approval following an FDA Complete Response Letter issued last October; during Novo Nordisk’s 4Q16 update, we learned that the company plans to resubmit the drug to the FDA this quarter (1Q17), with hopes of approval by end of year.

• During Q&A at the end of the symposium, Dr. Vincent Woo (University of Manitoba, Winnipeg, Canada) argued for cost-effectiveness of Fiasp. He suggested that real-world data will reveal more dramatic A1c-lowering with Fiasp vs. NovoLog compared to clinical trial data. We agree with that hypothesis. Studies thus far have used basal insulin optimization to clearly demonstrate the superior effects of Fiasp on postprandial glucose, and have shown slightly better A1c reductions with faster-acting vs. regular aspart. “In the real world, where basal insulin optimization doesn’t really exist, you might see even more of a difference,” Dr. Woo explained, adding that over time, a 0.1% or 0.2% greater lowering of A1c could meaningfully impact the incidence of diabetes complications. We wonder what payers would think of that argument. It’s also possible the real-world efficacy of Fiasp will be much better, but for a different reason: better mealtime control in those who do not take their insulin 20 minutes before eating, as would be done in a clinical trial. As with any new drug, insurance coverage will be a determining factor in the product’s commercial success (especially in the US, assuming FDA approval comes through by the end of 2017). Although the phase 3 clinical program for Fiasp has shown incremental rather than truly disruptive therapeutic benefits over NovoLog – especially compared to the improvement with next-generation basal insulins (Novo Nordisk’s Tresiba and Sanofi’s Toujeo) over earlier basal insulins (namely, Sanofi’s Lantus) – we still see value in Fiasp for a better bolusing experience and potentially very strong closed loop improvements.

The Evolving Face of Immunotherapy of T1D

Chantal Mathieu, PhD (KU Leuven, Belgium), Mark Peakman, PhD (King’s College, London, UK), and Desmond Schatz, MD (University of Florida, Gainesville, FL)

ATTD drew to a close with a packed Saturday afternoon panel on the status of immunotherapies for type 1 diabetes. An all-star panel reviewed the current state of immunotherapy for type 1 diabetes treatment, and provided updates on some of the most cutting-edge research in this domain. Overall, we were struck by the number of moving pieces here –energy and creativity is required to exploit properties of the immune system to protect the beta cells, and the basic science experimentation and trials will still require a great deal of time.

• The great Dr. Chantal Mathieu (KU Leuven, Leuven, Belgium) discussed her cutting-edge work using the lactococcus lactis bacteria as a delivery system for therapeutic antigens. Typically used in dairy fermentation, the lactococcus lactis bacteria can be genetically engineered to synthesize and secrete proteins of interest – in the case of type 1 diabetes immunotherapy, antigens such as pro-insulin, IL-10, anti-CD3, and LL-PNS.  Impressively, Dr. Mathieu’s research group determined in mice that the lactococcus lactis-facilitated delivery of a specific set of three proteins (anti-CD3 + LL-PNS + IL-10) to mice newly diagnosed with type 1 diabetes caused diabetes reversal at a rate of almost 60% by 14 weeks. Dr. Mathieu’s current work focuses on identifying the differences between the “responder” mice who undergo diabetes remission with this therapy versus the “non-responders” who do not. Early evidence suggests that distinctions in the Foxp3 regulatory pathway in regulatory T cells may be responsible – an indication that these markers could potentially be used as the basis of personalized medicine approaches to identify patients with type 1 diabetes who are likely to respond to this kind of therapeutic strategy.
• Dr. Mark Peakman (King’s College, London, UK) discussed his novel peptide-based therapy for type 1 diabetes, a strategy involving the administration of antigens in order to reestablish or induce tolerance, much like a vaccine. After the success of Dr. Peakman’s phase 1 MonoPepT1De study demonstrating the ability of vaccination with the a peptide mimicking the C19-A3 region of proinsulin to reduce insulin dose and A1c and raise C-peptide levels in individuals with type 1 diabetes in their first 100 days of diagnosis, his research team decided to investigate the effectiveness of a cocktail of peptides targeting multiple areas of the beta cell. Accordingly, the so-called MultiPepT1De study (also phase 1) was launched in 2015 and is expected to complete sometime in 2017. Dr. Peakman shared that a follow-up phase 2 study will likely be initiated this year as well.
• “The artificial pancreas can’t be a cure, because the brain is not a pancreas. As we get closer and closer to closed loop, we need to simultaneously intensify our efforts to find a biological cure for type 1 diabetes.” University of Florida’s Dr. Desmond Schatz eloquently captured the importance of research in this field, despite the difficulty of slow-moving studies and tough risk:benefit balance in getting an effective immunotherapy to patients. We wonder if closed-loop devices may make this balance tougher, raising the safety/effectiveness bar for cure therapies further.

Annual ATTD Yearbook

The ATTD Yearbook covered all the progress made in 2015-2016 diabetes research spanning therapy and technology. Click here for the full Yearbook, published in Diabetes Technology & Therapeutics.

6. Exhibit Hall

Abbott

Abbott’s trademark vibrant yellow booth showed off the LibreLink and LibreLinkUp apps, real world data from FreeStyle Libre use (see day #1 for our analysis), and a video of Dr. Rich Bergenstal discussing the one-page AGP. We continue to hope for standardization in how glucose is assessed and we really like this one-pager – see below from Glooko about expanded use of AGP. Abbott’s FreeStyle Libre consumer version hasn’t yet received FDA approval in the US – we learned on the 4Q16 call that the company hopes for a 2H17 launch and, notably, that Abbott has filed for both adjunctive and replacement label claims.

Ascensia

Ascensia exhibited the Contour Next Link 2.4 meter, which is available with Medtronic’s MiniMed 640G internationally and in the US with the 630G and upcoming 670G. The booth also showed off the Bluetooth-enabled Contour Next One meter and the Contour app virtual-reality style (we previously saw the demo at EASD). The Contour Next One meter is now available in at least 14 European markets, with a US launch expected early this year. The motif of the booth was “illumination” – “Seeing how everyday activities affect blood glucose – that’s illuminating”; “Their diabetes. Illuminated”; “Tap screen to be illuminated” – likely a reflection of the SmartLIGHT feature on the Next One meter that indicates in-range (green), low (red), and high (yellow) results. All the meter companies seem to have this now, and we wonder if patients find it useful and meaningful. Some, we imagine, may find it annoying to have a colored light reminding them they are high or low. Others may love it, perhaps because they don’t remember what one’s blood glucose should be. We wonder if simple dosing guidance for type 2s could follow from this: if the light is orange, take X units, if the light is green, take Y units. We do really believe that this clinical reminder could help patients and could be a “nudge” toward better care if presented optimally.

Dexcom

Dexcom’s booth had a wide countertop showing off the newly launched Android G5 app on smartphones, our first in-person look. The new G5 app launched in the UK, Ireland, Netherlands, Germany, and South Africa last month, and our first demo showed a pretty identical user experience to the Apple iOS version. The booth also shared Dexcom’s recent publication wins, including reprints of the DIaMonD type 1 results (CGM in MDI) published in JAMA last month. The now-trademarked tagline, “CGM First,” adorned the back wall of the booth, emphasizing Dexcom’s goal to drive CGM use before pumps. Indeed, for years, about 60% of Dexcom’s customers were pump users – that has now fallen to about 40%, given the rise in those on MDI who have started CGM. CGM reimbursement has improved in the EU and we’re eager to see how Dexcom’s business progresses there.

Glooko

The big news in Glooko’s booth was the launch of its new mobile app on Apple and Android, which finally rolled out last week. This adds some nice pattern recognition features: (i) the time of day when BG levels run high or low; (ii) how current BG levels compare to other time periods; and (iii) the day of the week with the best BG control. We like the work to incorporate emojis and make Glooko feel friendlier and less clinical, and more like a “friend”. The new mobile app also includes an Ambulatory Glucose Profile – it is great news to see this CGM data display expanding. The rep said the goal is make Glooko less of a retrospective app and more of a daily diabetes companion app – no surprise given the expanding work on insulin dosing decision support, including the new partnership with Novo Nordisk. Glooko has always had a strong mobile offering and we’re glad to see the combined efforts with diasend to drive even better use of diabetes data globally. Diasend’s in-clinic transmitter was in the booth, and we assume a combined product will still roll out this year (see our coverage of the merger from last fall – this is exciting!).

J&J

J&J made news late last month when it announced that it will “explore strategic options” for its Diabetes Care segment (LifeScan, Animas, Calibra), including operating partnerships, joint ventures, strategic alliances, or a sale of the businesses. For the small booth at ATTD, however, it was business as usual – the Animas Vibe insulin pump, the Bluetooth-enabled Verio Flex BGM, and infusion sets were displayed and detailed by reps. The cadence of new products has obviously slowed in the past few years – the Vibe was CE Marked nearly six years ago! We would love to see the OneTouch Via (formerly Calibra Finesse) as we have high hopes for this product that has big potential to make mealtime insulin dosing so much more flexible and discreet.

Lilly

Lilly occupied a small, unassuming side booth, displaying samples of the GLP-1 agonist Trulicity pen, the Humalog insulin KwikPen, and the Abasaglar KwikPen (recently made available in the US under brand name Basaglar). Although the booth was small, it was terrific to see support at ATTD by all the major insulin companies.

Medtronic

Similar to EASD, Medtronic demoed the Guardian Connect standalone mobile CGM in its booth. The iOS app looks spiffy and we appreciated the meal and exercise markers, which look far better than those in the Dexcom G5 app. The rep told us the device is now in 15 EU countries and will “soon” enter Chile; in Medtronic’s earnings presentation during the week of ATTD 2017, these were characterized as “pilot launches in major European markets” with a “positive response.” Guardian Connect remains under FDA review (with the new Enlite 3 sensor) and FDA approval is expected in roughly April/May 2017 as of the last update. Otherwise, the MiniMed 640G and iPro2 professional CGM with Pattern Snapshot were on display; we still have yet to see the MiniMed 670G hybrid closed loop in an international exhibit hall. As of JPM, the 670G was expected to launch outside the US in May-October of this year.

Novo Nordisk

Novo Nordisk boasted a midsize to large booth at the center of the long exhibit hall, dedicated entirely to its very recently (January 2017) EMA-approved faster-acting insulin aspart, Fiasp. Against a crisp, white backdrop, the booth’s signage emphasized the drug’s rapid action profile: the main wall depicted a lunch with orange juice, an apple, and a sandwich overlaid with text to suggest that Fiasp begins working “from the first bite.” This motif was continued throughout the rest of the booth, with images of apples inlaid with a stopwatch on the back walls, and a bowl of apples on the booth’s central table – making it a very popular stopping place for attendees during conference breaks. A team of Novo Nordisk representatives dressed in yellow stood on the outskirts of the booth, offering helpful demos of the Fiasp FlexTouch pen and guiding conference attendees through Fiasp’s clinical data on interactive iPad displays. Feedback appeared largely positive and we look forward to seeing how Fiasp performs in EU markets. After receiving a very unexpected Complete Response Letter (CRL) in October 2016, Fiasp will be imminently resubmitted to FDA under a class II resubmission, with US approval expected by the end of 2017. It is said to be incrementally better – not a breakthrough itself, but ideally it will lead to one. In the meantime, we believe many people with diabetes who can access Fiasp would certainly take “better,” given how hard diabetes is to manage.

Roche

At Roche’s exhibit, we got to handle the Accu-Chek Insight CGM for the first time (compare size to US currency below), learned that next-gen Accu-Chek Guide BGM will launch in Germany, Italy, and the US in 1Q17 (already in Australia, Switzerland, and Denmark), and that the Accu-Chek Instant BGM (streamlined feature set including a target range indicator, Bluetooth connectivity, and no buttons) will launch this month in Europe, on the earlier side of “2017” expectations in the 4Q16 update. A rep told us that the Instant may be better for patients who don’t need all of the whistles and bells offered by the Guide, but just a quick indication of blood sugar. The mySugr integration partnership was advertised in Roche’s corporate symposium with promises of “more to come,” but the just-signed exclusive deal with Medtronic to provide Bluetooth-enabled BGM for future Bluetooth-enabled Medtronic pumps wasn’t discussed at all at the conference – no surprise given that this is probably multiple years away. Also displayed were the DiaPort injection port with internal tubing and the SmartPix data display/analysis software. 

Sanofi

Sanofi’s ATTD exhibit hall presence largely matched what we saw last year, with emphasis on first-in-class MyStar DoseCoach BGM. Sales reps highlighted the integration of insulin glargine titration support (for next-generation basal insulin Toujeo) with BGM, and the download option to MyStar Connect, a management software platform that collects a patient’s MyStar DoseCoach data and presents it in one comprehensive package to HCPs. When is this coming to the US?! For the first time we can recall, we noticed AgaMatrix’s new Jazz 2 Wireless BGM (Bluetooth-enabled) on display, meaning Sanofi is also now distributing it under the MyStar brand – this is great to see, given the huge value of connectivity and the hardware challenges with iBGStar. Sanofi also promoted new-to-market basal insulin/GLP-1 combination Suliqua (branded Soliqua in the US) on the outer walls of the booth. Messaging highlighted (i) how the fixed-ratio combo of insulin glargine/lixisenatide improves both fasting and postprandial glucose, (ii) how the drug “helps more patients put high percentages behind them” – a very appealing message backed by strong data; and (iii) how the availability of two SoloStar pens for Suliqua in Europe offers greater dosing flexibility, an advantage over Novo Nordisk’s Xultophy (insulin degludec/liraglutide). As a reminder, the FDA only approved one SoloStar pen for Soliqua in the US (due to potential “confusion” on the part of patients and HCPs, which was surprising given all the complexity patients on MDI move through daily).

Senseonics

Just three meters from the Roche booth sat the bustling Senseonics booth, where we walked by as Dr. Mark Christiansen was presenting results from the PRECISE II US pivotal trial of the implantable Eversense CGM system (first seen at DTM; MARD=8.8% vs. YSI) to a large crowd. A rep told us that Senseonics is hoping to get CE marking for the new Eversense app (featuring Android, iOS, and Apple Watch compatibility, remote monitoring by up to five people, and a new fan-favorite “temp profile,” allowing 36-hour changes in alarm threshold), 180-day indication, and 55% smaller transmitter any day now. We were happy to hear that the company will be offering free transmitter upgrades to patients upon approval. We also learned that, in addition to deals with Rubin Medical (a very strong distributor) and Roche, Senseonics has had a small distribution agreement with Pharmanova in Finland “for the past few months” – the company plans to distribute in the US itself once the system is approved by FDA (filed in late October; launch expected in 2H17). A rep shared that clinicians at ATTD are beginning to picture their patients using the implantable CGM “because of the questions they were asking” about insertion/extraction procedures and patient satisfaction.

Unomedical

Unomedical demoed its all-in-one, fully disposable, hidden needle insertion set device, sharing that it will launch “after” this summer and first come to market in an exclusive partnership with a Luer Lock pump. The company also has a novel catheter (Lantern) expected to launch in 3Q17, which includes several slits along the side that allow insulin to flow out of multiple places in the case of occlusion or bending (watch the cool one-minute video here). We got to see the all-in-one inserter device up close in the hall and were very, very impressed – inserting a set only requires peeling the tape, pulling the plastic tab at the top, pressing the button, and lifting the device off the body (see picture below; this red version is for Parkinson’s, and the pump version just has a different color). The fully disposable serter does not require cocking or loading, and insertion occurs in a microsecond without seeing the needle before or after – a big improvement over BD’s FlowSmart set insertion in our experience (though Adam didn’t get to try Unomedical’s on his body, so we can’t comment on pain). The exclusive partnership news was a surprise to us (Unomedical supplies all the tubed pump companies), and perhaps a response to BD’s exclusive partnership with Medtronic for FlowSmart. Unomedical told us it may expand to sell the all-in-one inserter to other companies, but for an initial period of time it will be an exclusive deal.

We loved the form factor of the new inserter, though the team seemed far more excited about the new catheter, called “Lantern” – the video is worth a watch (picture below), showing how the slits expand to flare out and allow the insulin to flow out of many places. The team hopes it may even enable faster insulin absorption from a wider subcutaneous insulin depot, though this has to be studied. Lantern also seems like a response to BD’s two-hole FlowSmart catheter, though Unomedical has added wider side openings for more insulin flow. We are surprised Unomedical has not received any funding from JDRF for set innovation, as both of these novel approaches seem in line with the non-profit’s goals and recent funding of BD and Capillary Biomedical. We’ll be interested to see real-world human data on Lantern, especially whether kinking is more likely with the additional slits.

Ypsomed

Ypsomed’s YpsoPump was on display in its expansive booth, and we confirmed with reps that it has launched in the UK, Netherlands, and Germany. The rep said initial feedback has been very positive on the fully icon-based touchscreen interface and prefilled insulin cartridges (Novo Nordisk PumpCart). The manual fill option is hopefully expected this year, on par with the timing shared in the company’s November update for those that want to use other insulin. The rep did not comment on whether the pump has also launched in Czechia (Czech Republic), as was expected in last fall’s update. In our demo of the pump, we noticed it infuses bolus insulin very fast (like the Animas One Touch Ping/Vibe), though the scrolling to select a bolus size is very slow (a limitation of the icon-based interface). All eyes will be on Ypsomed’s spring update as it will share the first details on this launch alongside continued OmniPod distribution – how will the products co-exist?

 

-- by Adam Brown, Abigail Dove, Brian Levine, Payal Marathe, and Kelly Close