ATTD (Advanced Technologies and Treatments in Diabetes) 2020

February 19-22, 2020; Madrid, Spain; Full Report

Executive Highlights

  • ATTD ended in late February and looking back, it seems incredible that this conference wasn’t cancelled – the entire world was locked down a month later and multiple countries must have already been in lockdown mode during the meeting. We congratulate the creators of this meeting once again – it was a triumph. The additional focus on digital clinics put Dr. Moshe Philip and friends right back in the driver’s seat, as well – who knew how important this topic would be just one month later. Dr. Lori Laffel did at that meeting take on and accomplish 80% of the task in the first talk – nonetheless, the networking, and listening to various perspectives was, if a bit repetitive, still very important. Onward, virtual care … we are so much better for everyone having given their time for this important invite-only session on what the best digital clinics look like.

  • In automated insulin delivery (AID), ATTD was headlined by results from Tandem Control-IQ’s pediatric pivotal trial, Jaeb/Tidepool’s virtual observation study of Loop, Control-IQ’s six-month adult extension, and real-world European data on MiniMed 670G (along with updates on MiniMed 780G). We also heard enthusiasm for the FDA’s work on developing interoperable frameworks for all three AID components (iCGM, ACE pump, and iController) with Insulet making a big splash on day #1 by announcing dual partnerships to integrate Omnipod Horizon with both Dexcom (for G6 and G7) and Abbott (for FreeStyle Libre 2). Throughout the conference, we also heard speakers discuss the potential for AID systems in new populations, namely pediatrics and pregnancy.

  • For CGM, signs of its ever-growing adoption were everywhere. Big banners on the Feria de Madrid conference center entrance proudly proclaimed “over two million” users for FreeStyle Libre. Dexcom announced exciting news with the CE-Marking of Dexcom G6 in pregnancy. At an Abbott-sponsored symposium, 64% of the audience rated themselves as either “reasonably knowledgeable and very good level of expertise” or “an expert and extremely knowledgeable” about the “technical aspects and clinical indications” for CGM before the symposium even started. With CGM adoption quickly growing, enthusiasm for tips and support for integrating the technology into clinical practice were also high.

  • Digital health was on the big stages at ATTD, highlighted by pilot results from the Jaeb/Helmsley/Cecelia Health virtual diabetes clinic. Study participants remotely initiated on CGM (n=27 type 1s, n=7 type 2s) – including prescription, shipment and education – used the device an average 95% of the time (6.9 days/week) during the 12-week study period and saw a statistically significant -1.1% decline in A1c from 8.3% at baseline. In decision support, the much awaited readout of DreaMed’s Advice4U trial showed that use of the Advisor Pro yielded non-inferior results (p<0.0001) to expert physician advice in n=122 young people with poorly controlled type 1 diabetes. DreaMed’s Dr. Revital Nimri commented, “Equal is perfect. If we put all our experience into an algorithm [that works], then we have done what we need to do.”

  • Although ATTD is always lighter on the diabetes therapy side, the meeting still packed a punch with several standout sessions focusing on pharmacotherapy. Dedicated sessions on SGLTs in type 1 (featuring a can’t-miss debate between Profs. Thomas Danne and Stephanie Amiel), next-gen glucagon candidates, and type 2 treatment guidelines all drew considerable interest from attendees. To our surprise, sessions on NASH drew considerable interest – impressive for a tech-focused conference, and a signal of growing enthusiasm in the field as the first potential therapies for the condition soon emerge.

This report includes our coverage of the 13th Annual ATTD conference, which took place in Madrid, Spain. Immediately below, you’ll find our top themes from the meeting, followed by highlights in the following categories:

  • Automated Insulin Delivery and Insulin Pump Therapy

  • CGM

  • Decision Support

  • Diabetes Drugs

  • Big Picture

  • Time in Range and Beyond A1c

  • Digital Health and Telemedicine

  • Posters

  • ATTD Yearbook and Opening Ceremony

  • Exhibit Hall

Videos of talks with available recordings are linked, as well, for many of the talks – we’d estimate it at about half.

ATTD 2021 will be held in Paris from February 17-20 – we’re so looking forward to it.

Table of Contents 

Themes

1. Automated Insulin Delivery: More Systems, Increasing Interoperability, and New Populations

We thought 2020 was set up to be a major year for automated insulin delivery (AID) and if this year’s ATTD was any indication, then we were right. The buzz was palpable among attendees around AID and particularly, Tandem’s hybrid closed loop algorithm Control-IQ, which was FDA cleared in December 2019 and officially launched in January 2020 in the United States.

  • Looking ahead at 2020 and beyond, we got a look at the rapidly growing AID landscape. Medtronic continued to show real-world data from its first-to-market MiniMed 670G system: in 3,130 European users, Time in Range increased from 62% to 71% before and after enabling 670G’s Auto Mode. At the same Medtronic symposium, Dr. Robert Vigersky briefly touched on the upcoming MiniMed 780G (pivotal data to be presented at ADA 2020). Much to the crowd’s delight, Dr. Tadej Battelino (Ljubljana University) brought a patient, Ana, from his center enrolled in the 780G trial to share some of her experiences with the system. When Dr. Battelino showed Ana’s glucose profile, the audience saw 92% Time in Range with a GMI of 5.8%. This year’s ATTD also featured the six-month results for Jaeb/Tidepool’s observational study of Loop. Time in Range increased from 67% to 73% and A1c declined from 6.8% to 6.4% in new Loop users. Though we didn’t hear any updates on Insulet’s Omnipod Horizon system at ATTD, that AID system was also expected to launch by the end of 2020 (now delayed to “2021” due to a software bug). Diabeloop, Beta Bionics, SOOIL, and other AID players were also seen at the conference – see our AID competitive landscape for the latest.

  • Helping to drive AID innovation, we heard strong praise throughout the conference for the FDA’s work on creating an interoperable framework for AID systems. With the clearance of Control-IQ as an “interoperable automated glycemic controller” (iController) in December, all three components for an AID system now have class II, interoperable pathways (iCGM, ACE pump, iController). With the conference taking place in Spain, we heard speakers contrast the FDA and its progressive moves to facilitate AID development with the EMA and their moves to place closed loop systems under stricter regulatory categories. Insulet made a bit of a splash on the first day of ATTD, announcing partnerships with both Dexcom and Abbott to integrate Dexcom G6 and G7 and Abbott FreeStyle Libre 2 with Omnipod Horizon. With the Horizon pivotal using G6 as its CGM, the Horizon + FreeStyle Libre 2 pairing could give us a first look at how strict the FDA is with “plug-and-play” AID systems.

  • The pivotal for Control-IQ in pediatrics (ages 6-13) read out at ATTD, showing very similar results as we saw in the adult pivotal, which read out at ADA 2019. Time in Range in the Control-IQ group increased from 53% to 67% (+3.4 hours/day) after 16 weeks, compared to a 51% to 55% increase (+1 hour/day) in the control SAP group. Like we saw in the adult trial, time in closed loop was very high (93%) and a huge Time in Range improvement was seen at night (80% vs. 54%). The results of the trial have likely been submitted to the FDA already, setting the stage of the entrance of AID systems into the pediatric market. As a reminder, Medtronic’s MiniMed 780G pivotal is running with participants as young as 14 and Insulet’s Omnipod Horizon pivotal has participants as young as 6. In another strictly regulated population, we heard an update on the ongoing AiDAPT closed loop trial in pregnant women with diabetes. In the UK, Dr. Roman Hovorka’s (University of Cambridge) CamAPS FX algorithm was recently CE-Marked and launched for patients down to one year old and in pregnant patients.

2. CGM: Provider Enthusiasm Continues to Build – How Can We Help Providers Implement CGMs in Practice?; Consensus Targets Remain Hard to Reach

  • Perhaps best illustrated at an Abbott-sponsored symposium on the first day of ATTD, provider education around and enthusiasm for CGM is at a record high and shows no signs of slowing down. Before the symposium even started, 64% of the audience rated themselves as either “reasonably knowledgeable and very good level of expertise” or “an expert and extremely knowledgeable” about the “technical aspects and clinical indications” for CGM. Perhaps more impressively, more than half (58%) of the audience said that they “currently recommend and/or prescribe” CGM to >60% of their patients with type 1 diabetes. Big banners on the Feria de Madrid conference center entrance proudly proclaimed “over two million” users for FreeStyle Libre. At ATTD, Dexcom announced CE-Marking for G6 in pregnancy, Medtronic presented several case studies for using its zero-calibration Envision Pro CGM, and 12-month WISDM extension results were presented, all continuing to demonstrate the value of CGM across different populations.  

  • With CGM adoption increasing at an impressive speed, the need for tools and support around integrating CGMs into clinical care become increasingly apparent. When discussing how to use ambulatory glucose profile (AGP), Dr. Rich Bergenstal (International Diabetes Center) presented his “flat, narrow, in-range” (FNIR) mnemonic for an ideal ambulatory glucose profile – at least half of the diverse group of clinicians in the room pulled out their phones to snap a picture of Dr. Bergenstal’s slide explaining FNIR. The development and standardization of tools like AGP and dissemination of provider education through conferences like ADA Postgrad, Best of ADA, and ATTD remain crucially important to help patients and providers get the most out of CGM technology. At ATTD, we also saw data showing high utilization of CGM-related features, such as data sharing (with both clinicians and family/friends) and alarms. However, other data and stories from ATTD also demonstrated that barriers remain around engaging CGM users in their care, from uploading their data before clinic visits and reviewing their retrospective CGM data on their own.

  • Comparisons of data sets with the consensus CGM targets developed at ATTD 2019 emphasize the difficulty of achieving the targets. A post-hoc analysis of the Swedish GOLD trial (n=137) showed that just one-quarter (27%) of participants meeting an A1c goal of <7% were also meeting the consensus time in hypoglycemia goals (<4% time below 70 mg/dl and <1% time below 54 mg/dl). DCCT/EDIC trial participants wearing FreeStyle Libre Pro showed that average Time in Range was only 52% and about 4% of time was spent below 54 mg/dl. Only 9% of all participants met the Time in Range targets, with lower A1c and non-smoking status associated with better numbers. In a similar analysis from the CONCEPTT trial for pregnant women, achievement of Time in Range goals was fairly low. Interestingly, A1c was also found to be superior to CGM values in the prediction of pregnancy outcomes. 

3. Digital Health, Telemedicine, and Decision Support: Ready for Prime Time?

  • Digital health was on the big stages at ATTD, highlighted by pilot results from the Jaeb/Helmsley/Cecelia Health virtual diabetes clinic. Study participants remotely initiated on CGM (n=27 type 1s, n=7 type 2s) – including prescription, shipment and education – used the device an average 95% of the time (6.9 days/week) during the 12-week study period and saw a statistically significant -1.1% decline in A1c from 8.3% at baseline. Cecelia recently shared with us that they’ll be launching a nationwide virtual clinic in June. We also saw some a promising telemedicine intervention applied in rural, remote areas of Chile’s Aconcagua Valley. As highlighted by Onduo’s and Close Concerns alum Brian Levine, digital health may serve as a potential solution to the “last mile” problem of healthcare access.

  • ATTD finished up in Madrid just as COVID-19 really began to ramp up in Europe. At the time, few attendees imagined the impact the pandemic is currently having on the world and how it might strain and test our current healthcare system. Since then, we’ve heard (see interview with Dr. Bob Gabbay) and seen countless diabetes patients and providers turn to digital health and telemedicine in order to follow social distancing orders and reduce potential exposure to the novel coronavirus. Many digital health startups, diabetes and non-diabetes related, have risen to the challenge. The federal government has also played a part in marking telemedicine more accessible. Last month, HHS Secretary Alex Azar announced that the department would waive HIPAA penalties for “good faith use of telehealth” (e.g., using non-HIPAA compliant software) and CMS expanded telehealth service coverage during the pandemic. COVID-19 is clearly a major turning point for adoption of digital health and telemedicine tools – the adoption is in overdrive now in the US in particular, particularly with Medicare now funding digital care (finally) - we’ll report back on how well the transition is going from the HCP end but from the patient end – there is absolutely no going back, just development of best practices.

  • On decision support, the much awaited readout of DreaMed’s Advice4U trial showed that use of the Advisor Pro yielded non-inferior results (p<0.0001) to expert physician advice in n=122 young people with poorly controlled type 1 diabetes. DreaMed Advisor Pro is a decision support system for pump setting recommendations using pump and CGM data from the Glooko diabetes management platform and provided advice on various pump parameters (i.e., pump basal rates, insulin:carb ratios (ICR), and correction factors (CF)). DreaMed’s Dr. Revital Nimri commented, “Equal is perfect. If we put all our experience into an algorithm [that works], then we have done what we need to do.”

4. Diabetes Therapy: Debate Continues on SGLTs in Type 1; Enthusiasm Builds for Newly Launched Next-Gen Glucagon Products

  • SGLTs in type 1 remained a large focus at ATTD 2020, with multiple sessions homing in on the risk/benefit profile of using these therapies in type 1 patients. In one of the stand-out sessions of the meeting in our opinion, Profs. Thomas Danne and Stephanie Amiel engaged in a riveting debate on the merits of this adjunct therapy class, bringing much-appreciated novel perspectives to the often discussed lightning rod topic. As always, DKA risk was front and center of the debate, although we were glad to see a growing emphasis on better integrating potential CV and renal benefits of SGLT use in type 1s into the discussion. Of course, hard evidence in this population has not yet been generated, but we’ve heard many thought leaders speculate that the enormous benefits seen in type 2 populations will likely translate to the type 1 population as well. We’re hoping to see a bigger push in the coming years in better understanding this aspect of the risk/benefit profile of SGLTs in type 1, and imagine that better doing so will be critical for gaining broader approval of these agents.

  • Enthusiasm was palpable for newly launched next-gen glucagon products and their impact on hypoglycemia care. Two new treatment options – Lilly’s nasal glucagon Baqsimi and Xeris’ liquid-stable Gvoke – were approved in the preceding months to ATTD 2020, and it was clear that attendees were eager to learn more about how to apply these treatments in practice. A number of sessions highlighted these newly launched options: Dr. Jennifer Sherr spoke highly of Baqsimi and gave a live demonstration on how to use the product, Dr. Stuart Weinzimer spoke on Gvoke and its applications, and Locemia co-founder Mr. Robert Oringer detailed his own personal narrative regarding Baqsimi’s development. 

  • Interest in NASH was also sky-high – impressive for a tech-focused conference, and a signal of growing excitement in the area during a potential landmark year for the field. An early-morning session at ATTD drew a standing-room only crowd to listen to an overview of the current NASH landscape. 2020 is set to be a critical year for NASH, with the potential for the first drug to be approved (Intercept’s OCA) for the condition in the coming months, along with several other candidates in late-stage trials.

Automated Insulin Delivery and Insulin Pump Highlights

DCLP5: Control-IQ in Children Ages 6-13 Drives Significant Time in Range Improvements (67% vs. 55%); Massive Improvements at Night (80% vs. 54%); 93% Time in Closed Loop (Video)

On the heels of the recent launch of Tandem’s Control-IQ for patients age 14 and up, Dr. R. Paul Wadwa (Barbara Davis Center for Diabetes) presented highly anticipated results from the DCLP-5 trial of Control-IQ in children ages 6-13 years old. The 101 participants were randomized 3:1 to Control-IQ or SAP (i.e., open loop Dexcom G6 + Tandem t:slim X2). Over 16 weeks, Time in Range in the Control-IQ group increased from 53% at baseline to 67%, compared to those in the SAP group, who increased Time in Range from 51% to 55% (p<0.001). Like the adult trial, the vast majority of Time in Range improvement came at night, with the closed loop group reaching an astounding 80% Time in Range compared to 54% in the SAP group. Overall, time spent above 180 mg/dl was 31% for closed loop and 43% in SAP group (p<.001). Time below 70 mg/dl was not changed in either group. Only 8% of the patients were sensor naive at the start of the trial and time below 70 mg/dl was less than 2% in both groups at the start of the trial. Again, similar to the adult study, treatment effect on Time in Range emerged during the first month of the trial and persisted consistently over the course of the four month study.

  • Control-IQ was approved in December for ages 14+ and a specific black box warning against use in patients under 6 years old. It appears pediatric submission is on track for filing in 1Q20. Since the pediatric indication is the same algorithm (to our knowledge), it should be a 510(k) filing claiming substantial equivalence to today’s submission.

  • Regarding adverse events, there were no DKA or severe hypoglycemia events in the trial.

  • At baseline, 40% of patients had A1C >8%, 20% were new to pump, 92% used CGM, and A1C ranged from 5.7-10.1%. 

Tidepool/Jaeb’s Virtual Observational Study of Loop: Time in Range +1.4 Hours/Day (67% to 73%), A1c Drops From 6.8% to 6.4% In New Loop Users After 6 Months

The tunnel-like La Paz room at Feria de Madrid slowly filled up on Friday afternoon, as an increasing number of people packed in to hear the six-month results of the Helmsley-funded Tidepool/Jaeb fully virtual, observational study of DIY Loop. The results were limited to “new” users of Loop (n=607), comparing glycemic outcomes at baseline to months 4-6 of Loop. Mean Time in Range increased significantly from 67% at baseline to 73% during months 1-3 and stayed at 73% during months 4-6 (p<0.001). Mean A1c was 6.8% at baseline, decreasing to 6.4% at month 6 (p<0.001). Mean time <70 mg/dl was mostly flat from 2.9% at baseline to 2.6% during months 4-6; time <54 mg/dl fell from 0.4% to 0.3%. Time in closed loop was 83% during months 1-3, dropping slightly to 79% during months 4-6. Loop was shown to be safe, with one DKA event occurring (judged not related to Loop) and reductions in rates of severe hypoglycemia compared to baseline. Overall, the highly anticipated study results show statistically significant improvements on Loop, made even more impressive since the cohort had fairly tight glycemic control at baseline. The presentation is available at Tidepool’s website.

  • The study will continue until March 31, 2020, and results will be used to help get Tidepool Loop through the FDA. Submission could come as soon as this year. The study is also collecting patient-reported outcomes, such as diabetes distress, sleep quality, and hypoglycemia fear, which will be presented at a later date.

  • It is worth noting that the study cohort is very young, well-educated, wealthy, and racially homogenous; unsurprisingly, the group also had fairly tight glycemic control at baseline. In the “new” user cohort whose results were presented, 91% were white and mean age was 23. Notably, Loop did show Time in Range increases across all age brackets (see below). Participants were 56% female and 85% had a bachelor’s degree or higher. Nearly three-quarters (71%) of participants reported an annual household income ≥$100,000. A1c at baseline was 6.8%.

  • Before the results were presented, Jaeb’s John Lum estimated a worldwide Loop user base of ~9,000, based on RileyLink order history. As of February 2020, there are 2,585 Medtronic Loop users and 6,449 using Loop with Omnipod. Incredibly, Omnipod Loop compatibility came less than one year ago. At DiabetesMine’s Innovation Summit 2019, Tidepool CEO Howard Look showed a graph of new users enrolling in the Tidepool/Jaeb study, demonstrating a huge increase after the DIY Loop Omnipod code was posted in April 2019.

  • The study was conducted as a “virtual” observational study, meaning CGM and pump data flow straight from the Loop app to Apple Health to Tidepool/Jaeb. Fingerstick collection kits were mailed to study participants to collect blood for A1c data. Informed consent and demographic data were collected online and web-based questionnaires were sent at baseline, three months, six months, and twelve months. Participants were also sent a weekly ~one-minute survey to report severe hypos, DKA, or hospitalizations in the prior week. As devices and connectivity become increasingly common and easy to use, we’d expect virtual (or partially virtual) studies to become more common – certainly in the age of COVID-19, this is a particularly vital example.

Results

 

Baseline

Months 1-3

Months 4-6

6-month vs. baseline difference

Time in Range

67%

73%

73%

+1.4 hr/day

p<0.001

Time <70 mg/dl

2.9%

2.8%

2.6%

-4 min/day

p=0.12

Time <54 mg/dl

0.4%

0.4%

0.3%

-1 min/day

p=0.004

A1c

6.8%

6.5%

6.4%

-0.4%

p<0.001

  • Mean Time in Range increased from 67% at baseline to 73% during months 4-6 (+1.4 hours/day, p<0.001). Mr. Lum noted that the entire improvement was seen in the first month and maintained for at least six months. Notably, with Loop, Time in Range during daytime (6 AM – 10 PM) and nighttime was identical (10 PM – 6 AM) at 73%, which was interesting. At baseline, Time in Range was 68% during the day, and just 65% at night. Indeed, after three months, mean Time in Range was higher at every hour of the day (though surprisingly not that much higher, probably since the 68% was reasonably high to start). We were a bit surprised not to see an increase in average aggregate TIR around lunch or dinner – this is a pretty disciplined group!


  • Time in Range was increased in users of all ages, ranging from <7 to >50 years. The greatest benefit was seen in the 25-49 years age group (n=203), with Time in Range improving from 70% to 77% - that’s so great, translating to an hour and forty minutes more Time in Range. The smallest benefit was seen with the >50 years age group (n=50), as Time in Range went from 74% at baseline to 76% after six months – while some may say “yeah, that’s probably not clinically meaningful,” doth readers protest too much, we do point out it is an extra 29 minutes a day “in range” – anyone with diabetes will certainly take it. The mean age of the entire cohort was 23 years, with just 8% of users ≥50 years old and probably very few even over 60. DIY systems are often seen as requiring more technical know-how to operate and troubleshoot; we’d be curious to see how time in closed loop varied by age. From our view, we don’t see them as harder to operate when they are “working” – we believe it’s more about who understands or can easily put together an understanding of coding.

  • Time <70 mg/dl dropped insignificantly from 2.9% at baseline to 2.8% during months 1-3 and 2.6% during months 4-6 (p=0.12). Time <54 mg/dl fell statistically significantly, though perhaps not meaningfully, from 0.4% at baseline to ~0.3% during months 4-6 (p=0.004). Mr. Lum noted that the rate of hypoglycemic episodes (defined as distinct excursions below 54 mg/dl) was ~0.7/week before and after Loop.

  • A1c fell from 6.8% at baseline to 6.5% after three months and 6.4% after six months (p<0.001). Given the very low A1c at baseline, most would consider a -0.4% reduction in A1c as quite impressive.

  • Time in closed loop was 83% during months 1-3, falling slightly to 79% during months 4-6. We’re a bit surprised it went down to over a day a week not on closed loop (1.4 days). Mr. Lum did note that these were lower bounds of confidence intervals as data in the study is still being collected. In total, 32 participants (5%) in the group stopped using Loop – they will be surveyed to learn their reasons for discontinuation. It’s possible that needing to carry an extra device (RileyLink) around was not “worth it” to some, though most we know are more than happy to put up with that for a  closed loop that consistently works. Some call this a “slight decrease in time in closed loop compared to Tandem’s Control-IQ system” but we’d say 79% vs 92% time in closed loop (that’s what Tandem saw during the pivotal) will not be considered similar by many – it’s the difference of nearly a day less per week during TIR. Unsurprisingly, time using the Dexcom G6 CGM was quite high, at 96% during months 1-3 and 93% during months 4-6 – we just continue to see more and more data reflecting positivity about the G6.

  • Loop was shown to be safe, with one recorded DKA event (judged not related to Loop – we wonder if it was euglycemic DKA since DKA would be otherwise extremely surprising to see using Loop – we are not sure how “not related” was judged) and reductions in the rate of severe hypoglycemia. Severe hypoglycemia was defined on the weekly questionnaire as “need[ing] the assistance of someone else to treat the low.” Two severe hypo events were judged to be related to Loop, which was somewhat surprising though not that worrying from our view – SH in general is less worrying with AID since it’s far easier to anticipate. One event was related to a large suggested insulin bolus due to an overestimated glucose prediction, while another event could not rule out a Loop malfunction as the cause.


  • One supplemental slide (on Tidepool’s website, not shared at the presentation), shows a few selected survey results. Notably, three-quarters of respondents (sample size unknown) said they were highly likely to recommend Loop – we’re interested in what 25% of a presumably very engaged group didn’t like. About one-third (30%) needed help starting on Loop, which we thought was pretty low, and of those, 71% had another Loop user’s help. Fully 98% of users said they read online material before starting Loop – we wonder what the other 2% did (we’re guessing those folks had someone else doing the full set up).

iDCL3 Extension Phase Shows +1.4 Hour/Day in Range, -0.3% A1c, No Hypoglycemia Difference With Control-IQ vs. Basal-IQ

Coming off a strong year in 2019 (Control-IQ readout at ADA 2019, publication in NEJM), UVA’s Dr. Sue Brown presented results from the 13-week extension phase of the iDCL3 protocol. Of the 112 participants randomized to the Control-IQ group in iDCL3 (all of whom completed the study), 109 elected to continue into the extension phase. Fifty-five of the participants were randomized to the Basal-IQ (predictive low glucose suspend) group, while fifty-four remained on Control-IQ. Notably, participants only had two clinic visits: one at baseline and one at the end of the trial (13 weeks). The Basal-IQ group received two phone calls to help with on-boarding. At baseline (i.e., the end of the main study), the groups were well matched with mean A1c of 7.1% and 7% and Time in Range of 70% and 71% in the Basal-IQ and Control-IQ groups, respectively. It’s worth noting that the Control-IQ extension group also saw slight worsening in glycemic control, though it’s unclear if these differences were statistically significant.

 

Basal-IQ

Control-IQ

Adjusted Treatment Group Difference

 

Baseline

13 Weeks

Baseline

13 Weeks

Time in Range

70%

60%

71%

68%

+6%

p<0.001

Time >180 mg/dl

28%

38%

28%

31%

-6%

p<0.001

Time <70 mg/dl

1.3%

1.5%

1.1%

1.4%

+0.1%

p>0.05

Time <54 mg/dl

0.2%

0.2%

0.1%

0.2%

+0.04%

p>0.05

A1c

7.1%

7.5%

7.1%

7.2%

-0.3%

p=0.004

  • Time in Range decreased from 70% at baseline to 60% at the end of 13-weeks on Basal-IQ. In the Control-IQ group, Time in Range fell slightly from 71% to 68%. The risk-adjusted treatment group difference was +6% (p<0.001). This difference was most notable at night (midnight – 6 AM), where the treatment group difference was +10% (p<0.001). During the day (6 AM – midnight), Control-IQ has a +5% advantage over Basal-IQ on Time in Range. This was not a surprise, given that the Control-IQ algorithm performs more aggressively when Sleep Mode is activated.


  • Unsurprisingly, the entire between-group Time in Range difference was driven by differences in time >180 mg/dl. Both groups spent 28% of time >180 mg/dl at baseline, increasing to 38% in the Basal-IQ group compared to 31% in the Control-IQ group (p<0.001). 

  • Time <70 mg/dl was not significantly different between the two groups. At baseline, the Basal-IQ group spent 1.3% time <70 mg/dl, rising to 1.5% after 13 weeks. The Control-IQ group spent 1.1% of time <70 mg/dl at baseline, rising to 1.4% after 13 weeks. Both algorithms seem incredibly effective at reducing time in hypoglycemia – 1.5% time <70 mg/dl is just 22 minutes/day (though, 22 minutes – say that to anyone with diabetes!).

  • Adjusted A1c difference was 0.3% in favor of the Control-IQ group. Both groups had a mean A1c of 7.1% at baseline, rising to 7.5% in the Basal-IQ group, compared to 7.2% in the Control-IQ group. Similarly, adjusted mean sensor glucose was 7 mg/dl in favor of Control-IQ.

  • There were no severe hypoglycemia events or DKA in either group during the extension phase. Three hyperglycemia with ketosis episodes occurred in the Basal-IQ group, while none were reported in the Control-IQ group. Neither group recorded a severe adverse events related to the devices.

Real-World MiniMed 670G Data in Europe (n=3,139) Shows Time in Range Increase from 62% to 71% Before and After Auto Mode

Medtronic’s Dr. Ohad Cohen presented MiniMed 670G data uploaded to CareLink from 7,847 patients across 12 countries in Europe. The data was uploaded between October 2018 and January 2020. For the entire cohort, after Auto Mode was enabled, mean Time in Range was 71% and mean GMI was 7%. Impressively, users spent just 1.8% of time <70 mg/dl and 0.6% of time <54 mg/dl. On the hyperglycemia side, users spent 20% of time >180 mg/dl and 6% of time >250 mg/dl. Note that the “post-Auto Mode enabling” data includes time spent in and out of Auto Mode due to Auto Mode exits. In an email with Medtronic, we learned that time in Auto Mode in the cohort was 87% – in the US, this has typically been around 70%-75%. Time using the Guardian Sensor 3 was an impressive 90%. Time in glycemic ranges did not vary too much from country to country in Europe: ATTD host country Spain had the highest Time in Range at 74% (n=625), while the UK came in last with a Time in Range of 69% (n=766). Again for comparison, Medtronic’s last batch of real-world data shared in the US showed a mean Time in Range of 71%.

  • In 3,139 users with pre- and post-Auto Mode enabling data, Time in Range increased from ~62% to 71% after enabling Auto Mode (+2.3 hours/day; p<0.001). Mean sensor glucose dropped from 164 mg/dl to 153 mg/dl after Auto Mode enabling, and GMI fell from 7.2% to 7%. Notably, the decrease in mean glucose and GMI actually came with a decrease in time spent in hypoglycemia: time <70 mg/dl and <54 mg/dl both decreased slightly, from 2% to 1.8% and 0.6% to 0.5%, respectively. This made us think of Dr. Tadej Battelino’s comments at an Abbott-symposium yesterday that we should put the DCCT idea of lower A1c requiring more hypoglycemia exposure in the “history drawer” part of our minds. Time above 180 mg/dl and 250 mg/dl decreased from 26% to 21% and 10% to 6%, respectively.

  • Looking at nine-month data that was available in 237 users, Time in Range improvements were seen one month after Auto Mode enabling and sustained for at least nine months. The pattern of improved glycemic control appearing just weeks or months into starting closed loop is not a surprise, as we’ve seen similar trends with other systems. We’d be curious to know how percentage of time in Auto Mode changed over the nine-month period, as many view the calibration requirements and various Auto Mode exits in 670G as quite burdensome. Time spent in Auto Mode is correlated with improved glycemic outcomes, so the sustained Time in Range results are an encouraging signal.

  • As Medtronic’s Dr. Bob Vigersky compared MiniMed 670G with the upcoming 780G, the excitement in the audience was palpable. On Medtronic’s earnings call earlier this week, we learned that the system is planned for a launch after April 2020, though the system has already been submitted for CE-Marking. US pivotal trial data will read out at ADA this year. Of note, Medtronic’s slide (below) notes “fingerstick is optional” for meal bolusing. This assumes that Guardian Sensor 3 would have non-adjunctive labeling (submitted to FDA in June) allowing users to forego a confirmatory fingerstick before bolusing. Of course, 780G will also launch with Guardian Sensor 3, which does require two fingerstick calibrations/day. Much to the crowd’s delight, Dr. Tadej Battelino brought a patient, Ana, from his center enrolled in the 780G trial to share some of her experiences with the system. When Dr. Battelino showed Ana’s glucose profile, the audience saw 92% Time in Range with a GMI of 5.8% - fantastic! On a stage with renowned KOLs like Drs. Battelino and Vigersky, the audience asked 16-year old Ana to field many questions. Ana summed up her experience so far, saying, “I was so happy to achieve a better daily Time in Range with considerably less effort.”

Control-IQ Use in Pediatrics and in Adult Populations Deliver Similar Time in Range Improvements of ~2.5 Hours/Day; Comparisons Between MiniMed 670G and Control-IQ Pivotal Trials (Video)

The renowned Dr. Boris Kovatchev (University of Virginia) compared two Control-IQ trials, demonstrating similar Time in Range improvements and A1c reductions when comparing pediatric patients (ages 6-13) and adults (14+) using Control-IQ. In a 16-week study assessing the efficacy of Control-IQ in pediatric populations (iDCL Protocol 5), patients (n=100, 6-13 years of age) were randomized 3:1 to Control-IQ vs. sensor-augmented pump treatment (i.e., open-loop G6 with t:slim X2). The Jaeb-coordinated study, took place across five major research sites over four months with an extension phase of three months. Preliminary results demonstrate almost identical results when compared to iDCL Protocol 3 (the Control-IQ pivotal trial in adults), with improvements in Time in Range of 2.5 hours/day, driven by decreases in hyperglycemia: time spent >180 mg/dl was reduced by 10%. A1c reductions in iDCL3 and iDCL 5 were 0.3% and 0.4% and time spent in closed-loop were 92% and 93%, respectively. Approximately 20% of these participants in the iDCL5 were new to pump therapy with the majority already using a CGM. Dr. Kovatchev began his talk by demonstrating the exponential growth in artificial pancreas publications from approximately one in 1950 to almost 240 in 2019. As of today, there are ~30 recruiting or active, non-recruiting artificial pancreas clinical trials on ClinicalTrials.gov - a real testament to how much traction the field has gained in the last few years.

  • The second metanalysis Dr. Kovatchev displayed compared results of the pivotal trials from the Medtronic 670G and Control-IQ, finding a  greater Time in Range improvement with Control-IQ, before adjustments for study type and baseline characteristics. Dr. Kovatchev noted that this analysis was challenging to perform because baseline Time in Range for populations in both trials were different. Additionally, MiniMed 670G’s pivotal trial was not a randomized controlled trial and did not have an equivalent control group seen in Control-IQ’s trial. (670G’s pivotal was single-arm.) Without adjustment, these differences result in a Time in Range improvement during active treatment of 5.5% in the 670G trial vs. 10.7% in the Control-IQ trial. However, after utilizing three distinct statistical methods to match the baseline Time in Range distribution in the Control-IQ study to the same distribution in the 670G study, the adjusted times in range become closer – 72.2% in the 670G trial vs. 74.4% in the Control-IQ trial.

  • Dr. Kovatchev also displayed metanalysis data from DTM 2019 which made comparisons from iDCL protocol 1 (smartphone-embedded inControl algorithm, Roche Spirit Combo pump, Dexcom G4 or G5) and iDCL Protocol 3 (algorithm embedded in Tandem t:slim X2, Dexcom G6). The results indicated that the Control-IQ/G6 (embedded closed-loop system) delivered an additional 1.5 hours/day of Time in Range. According to Dr. Kovatchev, these results indicate that closed-loop design, form factor, and user-friendliness are also important to consider beyond results as future technology is developed. According to Dr. Kovatchev, these direct comparisons are valid because both studies’ control group received the same treatment of sensor-augmented pump therapy.

Medtronic Extended Wear Infusion Set (7-Day) Receives CE-Mark; Small Study (n=21) Shows 81% Survival Rate at 7 Days

Medtronic’s Dr. Ohad Cohen announced CE-Marking for a new seven-day extended wear infusion set. Dr. Cohen did not share any information on when the new device might launch. The CE-Marking comes at the tail end of last year’s goal to launch the infusion set within “one-year.” The new infusion set includes several new components, including a new “H-Cap Connector” that improves infusion site performance, tubing that improves insulin preservative retention, extended wear adhesive patch, and a new inserter. The new inserter appears to be the all-in-one, hidden-needle Mio Advance automatic inserter device. This inserter is already been available in Europe and has been FDA-cleared for over a year, but has not launched in the US. On the US front, Medtronic announced it received investigational device (IDE) approval to begin a pivotal study of its extended wear infusion set back in August. The multi-center, non-randomized, prospective, single-arm study will enroll up to 150 participants with type 1 diabetes (18-80 years), all wearing the MiniMed 670G and comparing current 2-3 day sets to the new extended wear set. We didn’t anything on potential US timelines today. Medtronic featured the infusion set at a special “Innovation Suite” at the ATTD Exhibit Hall; unfortunately, it was not open to the press.

  • Dr. Cohen presented data from a clinical study of 21 subjects using “various Medtronic pumps” and either Humalog or Novolog. In total, the subjects recorded 82 insertions and 78 wears. Set failure was defined as blood ketone level >0.6 mmol/l, evidence of infection at the infusion site, or blood glucose not decreasing by at least 50 mg/dl one hour after a correction bolus for a glucose level >300 mg/dl. Kaplan-Meier curves showed 81% survival rate at 7 days for the new infusion sets; the survival rate increased to 85% if insertion failures were excluded. Reasons for failure were site reaction/blood (7% of all wears), insertion failure (5%), unexplained hyperglycemia (3%), and adhesive failure (3%). Notably, the total daily dose did not increase over the seven days, indicating insulin delivery efficiency was not significantly changed over the entire period.

On-Body iAPS Data (n=18) Reveals Time in Range Improvement from 57% to 62% After 48 Hours; Reduction in Time Spent <70 mg/dl in Outpatient Settings (n=10) From 3.5% to 2.3% After Two Weeks

Dr. Eyal Dassau (Harvard) presented positive initial data from two studies highlighting outcomes of the Interoperable Artificial Pancreas System (iAPS) smartphone app under two conditions: one where exercise was not announced and there were no restrictions on meals and another in the context of prescribed meals. As a reminder, the application is capable of interfacing with CGMs (Dexcom G5 and G6), pumps (Insulet Omnipod and Tandem t:slim X2), and algorithms while running on an unlocked smartphone. Patients can use iAPS to bolus insulin, log activities, and view glucose. The app even provides alarms for system malfunctions and synchronizes data with a server to enable remote monitoring. As he often does, Dr. Dassau situated his talk with an emphasis on human-centric design in closed loop therapy.

  • The first study was a single-arm, 48-hour study (n=18, type 1s) of iAPS use in a semi-supervised environment among children 2-18 years old. Participants consumed three meals per day and were not restricted in what they could consume. Unsurprisingly, most subjects consumed more carbohydrates on closed loop compared to open loop. Participants also took part in one unannounced group exercise activity such as escape rooms, trampoline, and ropes courses. After 48 hours, mean glucose had decreased from 173 mg/dl to 167 mg/dl, while Time in Range improved from 57% to 62%. Impressively, participants on average had CGM active 99.9% of the time and spent 92% of time in closed loop. Despite the improvements in both metrics as a whole, not all participants reaped the benefits. Were there any unique characteristics of these patients that led to these outcomes? We also wonder how these outcomes would hold under longer periods of time.

  • The second study was a two-week randomized crossover study comparing iAPS use at home on an unlocked Google Pixel 2 smartphone vs. predictive low glucose suspend in the context of prescribed meals. More specifically, participants alternated between eating pasta and rice with the same carbohydrate content. Meals were blinded during analysis. While there were no statistically significant improvements in mean glucose or Time in Range, time below 70 mg/dl decreased from 3.5% to 2.3%. Participants spent 85% of time in closed-loop. Participants also described positive experiences with the iAPS platform during follow-up interviews. Dr. Dassau mentioned at ATTD 2019 that he had received “excellent feedback” from users in this study – it’s nice to put some numbers now behind these qualitative insights!

Successful 10-week Onboarding from MDI to MiniMed 670G System in Qatar: Time in Range Improves from 47% to 76%, A1c Decline From 8.2% to 6.7%, Median 89% of Time in Auto Mode After Three Months

Dr. Goran Petrovski from Sidra Medicine in Doha, Qatar kicked off a series of oral presentations with impressive results from a 10-day initiation protocol study (n=30) designed to onboard pediatric patients (ages 7-18) straight from MDI to the MiniMed 670G hybrid closed-loop (HCL) system. Using a four-step approach, mean Time in Range increased from 47% at baseline to 72% on day 10 and 76% on day 57 (p<0.001). Mean A1c levels declined significantly from 8.2% baseline to 6.7% after 12 weeks of MiniMed 670G (p = 0.017). The number of Auto Mode exits declined significantly from 8.4 in the first two weeks to 4.2 by the third month. Lack of calibrations and high sensor glucose levels made up most of the Auto Mode exits. Notably, participants spent a median of 89% of their time in Auto Mode and 92% using the Guardian Sensor 3 and there were no reported cases of severe hypoglycemia or ketoacidosis. For context, in MiniMed 670G pivotal studies, Time in Range values increased from 56% to 65% in ages 7-13. During Q&A, Barbara Davis’ Greg Forlenza congratulated Dr. Petrovski on the impressive results and confirmed that the Qatari patients were initiated using the newer 670G transmitter that reduced Auto Mode exits.

  • The Sidra protocol encompassed a four-step structured approach that was quite intensive. Step 1 introduced the participants to the MiniMed 670G system through group sessions and discussed responsibilities and expectations. Step 2 trained users on hybrid closed loop with participants attending group education sessions with instructors for five consecutive days. In Step 3, participants began to use the system in manual mode, collecting insulin and CGM data for a period of three days. In the final Step 4, Auto Mode was turned on and the group was measured for 12 weeks.

14-Week Type 2 Trial of Valeritas’ V-Go Patch Pump Shows No Significant Differences in A1c or Hypoglycemia w/ Regular Human Insulin vs. Analog Insulin

Dr. Pablo Mora (Dallas Diabetes Research Center) presented results from a 14-week study of human vs. analog insulin in Valeritas’ V-Go patch pump in type 2 patients, concluding there is no significant difference between the two options on A1c, hypoglycemia, or total daily dose of insulin. CGM was not used in the study, so Time in Range differences could not be evaluated – session chair Dr. Jennifer Sherr brought this point up in Q&A, and we’re very curious as to what Time in Range comparisons may look like between the two groups. Dr. Mora concluded that given the intense affordability concerns with rapid acting analog insulin, wearable patch devices such as V-Go may help in minimizing differences between these more expensive insulins and human insulins and therefore ease affordability concerns. We wonder if cost-effectiveness analyses are planned for V-Go usage in these situations, and how impactful such data may be for payers evaluating V-Go. Regarding what may be driving this effect, Dr. Mora posited that V-Go’s ability to provide a continuous infusion of basal insulin can minimize differences in time action profiles between these two classes of rapid acting insulins. He further added that this option could be especially useful in older patient populations with type 2 (>65 years old), and that additional data on this will be presented at upcoming meetings. See table below for a full breakdown of primary and secondary endpoints from the trial in its per-protocol analysis (n=113 patients).

 

Human Insulin

Analog Insulin

A1c (primary endpoint)

-0.60%

-0.38%

Change in hypoglycemia (pre to post number of subjects)

+3

+3

Change in total daily dose of insulin (units/day)

+0.78

+1.82

  • Regarding safety, Dr. Mora emphasized that there were no severe hypoglycemic events in the entire trial.

  • On why CGM was not used in this trial, Dr. Mora pointed to access issues and the question of resource allocation in funding and planning the trial. He said that the investigators “debated for six months whether we would use CGM in this trial or for at least a subset of the trial population. But we decided to use CGM eventually in a separate study in a future. We have to remember that this was a relatively older population that has a difficult time in accessing insulin even. Access to CGM in real clinical life is very challenging as well. We look forward to looking further into using CGM in the future trials.” It is worth noting that V-Go has been a primarily type 2 product, a market where CGM access is often much more challenging – we hope to see this change, as it shouldn’t be.

Dr. Bruce Bode Presents Sneak Peak Data of URLi vs. Humalog in MiniMed 670G: No Major Differences in Glucose Profiles; Full Results at ADA 2020 (Video)

In a talk reviewing ultra fast-acting insulins in closed loop systems, Dr. Bruce Bode (Emory University) presented results from a randomized, double-blind, outpatient, crossover, active-controlled trial of URLi vs. Humalog in type 1s on MiniMed 670G. Dr. Bode is a lead investigator of the trial, and presented results from the patients (n=36) at his study site; full results including patient data at the other site in the trial will be presented at ADA 2020. Looking at ambulatory glucose profile (AGP) comparisons between the two groups and lead-in data where all patients were on Humalog, data look very similar (see slides below). Dr. Bode noted that glucose levels appear to be lower during breakfast, higher during lunch, and about the same for dinner when comparing URLi to Humalog groups. There were no differences in hypoglycemia, and Dr. Bode stressed that Time in Range figures showed that both groups overall only spent 0.7% time in hypoglycemia, which is especially “unheard of” considering the low A1c’s the patients in this trial were at (baseline 7%, baseline Time in Range 78%!). Regarding why the faster acting URLi did not convey more noticeable benefits in terms of improving glucose profiles, we imagine that changes must be incorporated into closed loop algorithms to fully incorporate the improved PK/PD profiles of these insulins – Dr. Bode briefly touched on this point in terms of why data for this class have been relatively underwhelming in closed loop systems so far.

Regulators, Manufacturers, DIYers, and People with Diabetes Express Hope for an Interoperable, Open-Protocol AID Ecosystem

An afternoon session brought together regulators, people with diabetes, manufacturers, and members of the DIY artificial pancreas community to discuss the implications of an interoperable, open-protocol AID ecosystem. JDRF led the charge toward an “open protocol” system with a 2017 initiative encouraging pump and CGM manufacturers to apply with proposals to provide seamless, secure, interoperable connectivity with other devices and smartphone apps. JDRF’s Dr. Daniel Finan elaborated that the vision behind this initiative is to both promote innovation and to ease the regulatory burden for CGM, pump, and algorithm developers such that they are responsible for developing and perfecting only one part of this system, as opposed to the entire multi-component AID platform. The wider goal is to improve patient choice for people with diabetes who use AID systems, giving them the opportunity to “mix and match” pumps, CGMs, and algorithms according to their unique needs. This point was underscored by patient advocate Ryan Gutzmer, a professional figure skater and hockey player who has lived with type 1 diabetes for 29 of his 32 years. He pointed out that the current rather limited range of FDA-approved AID devices encourages patients to venture into the DIY space, where they often receive little direction and support from HCPs: “We’ve come so far with DIY systems, but at the end of the day the result is that you can’t discuss one of the most effective diabetes management strategies with your most trusted people.” He emphasized that much better outcomes would result “if patients are allowed to combine what CGM they like, what pump they like, have an algorithm that allows both to communicate, and be able to be educated on all of these systems by trusted HCPs.” Stanford behavioral psychologist Dr. Jessie Wong echoed this. Pointing out the vastly different preferences that govern the preference of pump and CGM in children (device use in social situations) vs. adolescents (physical features, wear-ability, and comfort) vs. adults (device reliability and safety). In step with this, the FDA has been incredibly active to create the pathways to fully interoperable AID ecosystems. In contrast to the “dark ages” of the past, where algorithm developers had to partner with hardware companies and fix a system with specific components before gaining FDA approval, Dr. Courtney Lias elaborated that these changes will incentivize additional technology development, allowing companies (particularly small companies) to develop and take ownership of CGMs, pumps, and algorithms separately, without a need to have exclusive partnerships or take responsibility for the development of devices or software that aren’t their area of expertise. With this vision in mind, from 2018 to the end of 2019, the FDA created three new regulatory classifications in iCGM (March 2018), ACE pump (February 2019), and iController (December 2019). With dual partnership announcements between Insulet’s Omnipod Horizon with Dexcom’s and Abbott’s CGMs, we could see the first hybrid closed loop system with multiple CGM options as soon as the end of this year.

  • The session closed with a panel discussion including independent innovators in the DIY artificial pancreas space. Adrian Tappe, a developer for Android APS, praised the new open-protocol system, noting that the new environment incentivizes startups to “develop a very innovative product, do it well, and then connect to others working on other aspects of an AID system without having to take care of everything themselves.” Loop’s Katie DiSimone agreed, and posited that the open-protocol initiative frees manufacturers to be more innovative in the domain of training users and HCPs on how to use these new products – a domain that currently has great room for improvement from the current standard of long documents that few people have the motivation to read. As she pointed out, AID systems represent a huge opportunity to engage people and “light a fire” to motivate them to manage their diabetes in new ways, and it is critical to find ways of “de-burdening” patients when it comes to learning these new systems.

  • On the manufacturer side, Diabloop CEO Erik Hunecker discussed the fundamental differences in companies that focus on hardware vs. software, praising the open protocol initiative for allowing these domains to undergo development separately. Extending upon this, Medtronic’s Patrick Weydt urged that at this juncture toward open protocols, the AID community should acknowledge that it is “not the first industry to undergo interoperability,” and should take the opportunity to learn from the wins and losses of other industries that have entered this domain.

  • An ongoing theme during this session was regulatory differences between the US and the EU. In contrast to these progressive moves to facilitate AID development by the FDA, the EMA has made the regulatory pathway for AID more difficult in recent years, going as far as to move closed loop systems from class II to the stricter class IIIb. This is surprising. During Q&A, many Europe-based attendees argued that AID systems approved in the US should be available to people in the EU because, after all, “if it’s safe for someone in the US it should also be safe on the other side of the Atlantic.” We’ve heard the same thing for other approvals, the other way around (SGLT-2 for type 1 inhibitors). We urge regulators at the FDA to share their thinking with their counterparts at the EMA.

Pregnancy in Diabetes Draws Crowds: Dr. Helen Murphy on What’s New in AID, Dr. Yariv Yogev’s Call-To-Action on Research Standards (Video)

In an aptly timed session with the recent CE-Mark for Dexcom in pregnancy, the venerable Dr. Helen Murphy (King’s College London) provided audience members with a brief update on the ongoing AiDAPT closed loop trial (n=124 type 1s) in women with diabetes. We first caught wind of AiDAPT at Diabetes UK 2018, where Dr. Murphy presented on the study’s predecessors CLIP_03 (n=16) and CLIP_04 (n=16). As a reminder, CLIP_03 demonstrated that overnight closed loop reduces hyperglycemia without increases in hypoglycemia or insulin dose, and CLIP_04 showed that day-and-night closed loop reduces hypoglycemia. AiDAPT is a parallel-arm, randomized controlled trial comparing the new CamAPS FX closed loop system (Dexcom G6, Dana RS pump, Samsung S7 phone) developed by Cambridge’s Dr. Roman Hovorka with open loop (Dexcom G6, pump or MDI). Excitingly, we got a first look at one of the trial participants’ closed-loop stats, showing an impressive 84% Time in Range. While these results of course are only meant to illustrate the type of data AiDAPT will be collecting, we so look forward to seeing the full results in the future.    

  • Israel’s Dr. Yariv Yogev jumpstarted the session with an impassioned request to improve standards for the study of gestational diabetes, directed towards researchers and journal reviewers. To illustrate this issue, Dr. Yogev walked audience members through a 2018 publication in JAMA, entitled “Effects of Glyburide vs. Subcutaneous Insulin on Perinatal Complications Among Women with Gestinational Diabetes: A Randomized Clinical Trial,” which concluded that its findings do not justify use of glyburide as a first-line treatment. Dr. Yogev critiqued the study for categorizing study participants into those with good, moderate, or poor glycemic control, rather than giving actual glycemic values. In addition, the study’s primary outcome was set to have a non-inferiority margin of 7% “based on a group of clinicians’ point of view,” rather than any sort of specific criteria. Dr. Yogev called the study’s findings “ambivalent” overall and urged journals to only publish data with glucose values (vs. “well-controlled,” “poorly controlled,” etc.), define disease-related outcomes, and use outcome-based rather than statistics-based measures. In his purview, articles must be held to a higher standard, as HCPs often do not have time to “read the fine print.” Although we were somewhat surprised that Dr. Yogev would dedicate so much of his presentation to criticizing one publication, his call-to-action was undoubtedly compelling. 

Real World Study of Omnipod Users: Pumpers Who More Frequently Bolus Have Improved Glycemic Outcomes

In a very intriguing real-world study, the indomitable Dr. Irl Hirsch demonstrated that pumpers who bolus more have better glycemic outcomes, making the point by using a large dataset of Omnipod users sourced from Glooko. A cohort of 1,159 adult, type 1, Omnipod users who also submitted CGM data were selected from a large (~26,000) set of Omnipod system users in the Glooko diabetes management system. The most common bolus frequency was in the four to six times a day range. However, Time in Range increased smoothly with increasing bolus frequency up to eight times per day, after which it remained constant at 73%. People who bolused <4 times per day averaged 57% Time in Range. Impressively, time <70 mg/dl was below the consensus target (4%) for all groups. Those bolusing eight or more times a day had a GMI of 6.8%, compared to 7.6% in the <4 boluses/day group. The takeaway from these data are clear: remembering to bolus for food or hyperglycemia corrections can have a big payoff in improved treatment outcomes.

Dr. Rich Bergenstal Highlights Accomplishments and Challenges with MiniMed 670G (Video)

In one of the most entertaining and informative talks so far at ATTD, Dr. Bergenstal discussed the accomplishments and challenges of the Medtronic MiniMed 670G hybrid closed loop system, now >3 years old. The number one accomplishment was simply that 670G was the first FDA approved AID system. He commented that “[Medtronic] took a dream and turned it into a reality.” Of course, because it was first generation system and expectations were set high, there was an inevitable mismatch with real world experience. While some may have expected 670G to truly automate insulin delivery, 670G still requires work to operate and we have seen a substantial amount of discontinuations. However, Dr. Bergenstal shared testimonials from some users who have been moved to tears by the benefits compared to their old pumps (in a moving video) and many have found that the improved glycemic control more than justify the work required, especially considering that prior systems also had significant associated burdens. Many of the discontinuations of 670G have been due to poor upfront training (see highlight directly above) or the desire to switch to a CGM system that didn’t require fingerstick calibration. Additionally, an analysis by Dr. Boris Kovatchev (University of Virginia) showed that adjusted results from the pivotal trials of the 670G and the Tandem Control-IQ were comparable in Time in Range (72% versus 74%) when adjusted for cohorts and time in “active treatment” (i.e., closed loop). Indeed, for most, the challenge with using MiniMed 670G is simply staying in Auto Mode. Dr. Bergenstal chose to conclude with a key point – that we should be working towards the day when AID is the de facto standard of care.

  • In a video, we saw some compelling testimonials for the 670G from users:I wake up in the morning and it’s almost always 90-something or 100-something (mg/dl). Makes you feel like it adds years to your life.”

  • A 12-month real-world study showed that 33% (26 out of 79) of 670G users stopped using Auto Mode. The Guardian Sensor 3 is considered the greatest challenge to using 670G – Dexcom G6 and Abbott FreeStyle Libre, in particular, have longer wear time, factory calibration, and non-adjunctive labeling. Discontinuation of 670G is predicated on too many calibrations, frequent alarms and too much time needed to make the system work. In children users, “difficult to calibrate” and “too many fingersticks” were the top two reasons for discontinuation cited by parents. On the other hand, higher levels of up-front training have been associated with far fewer discontinuations.

  • A meta-analysis of the pivotal trial results for the 670G and Tandem Control IQ showed comparable TIR results, after adjusting for baseline differences. The analysis performed by Dr. Boris Kovatchev and Dr. Marc Breton from University of Virginia concluded that Control-IQ demonstrated 74% Time in Range compared to 72% for the 670G, after adjustments (see photo). In another key difference, Control-IQ’s pivotal showed 92% time in closed loop, while 670G’s time in closed loop was lower (87%). In real-world data, 

  • In a later talk in the same session, the legendary Dr. Barnard-Kelly (University of Southampton) drove home the point that the closed loop is still burdensome for people with diabetes, but if we support them, we can get great results. Different people need different onboarding and expectations should be managed up front. But in a psychological study of adolescents on closed loop and their parents, participants felt less burdened, had better glucose control, and reduced worry. She noted “All the things people are hoping for, they are getting from these systems. It would be really a shame for all the chatter to detract from the fact that they are successful and getting better.”

  • Amy Winscombe gave her personal perspective on living with MiniMed 670G. She noted it has turned out to be excellent overnight and she wakes up at 100 mg/dl every day. She doesn’t worry about her glucose levels because they are always in range or heading there. The major negative for her is having to wear a CGM, which makes her feel self-conscious. Ms. Winscombe also noted her improved headspace from reduced diabetes burden. She felt that she had “got something always looking out for me.” Additionally, her family is a lot less anxious and “we all feel that I am safer.” “I rarely have hypos now, or I go low but not as low as I would have done before. I feel I am more likely to have a long, healthy, happy life.”

Selected Questions and Answers

Q: Did pre-existing DIY systems put the bar too high?

A: I have the highest respect for the genius of the DIY folks. It worked for them, but commercializing it and getting [the 670G] through the FDA… was “safety, safety, safety, safety”. If you have one DKA or severe hypoglycemia in your trial, it will stop tomorrow. So I think the [670G] algorithm was designed to be a little bit conservative. So maybe the DIY people saw a difference, because they were already on to generation two.

Q: Should the 670 G be given to everybody, or in reality would you argue for the selection of patients?

A: Well, you start off with a new product and there are exclusions. For example, no hypoglycemia. For the next generation trials we are taking MDI patients, hypoglycemia unawareness, people who have never used a sensor etc. The first participants were people we had confidence would get through the trial. But now we feel better because AID lowers hypoglycemia quite significantly. So hypoglycemia unaware folks would be perfect candidates.

Q: Why didn’t Medtronic release a product to automatically handle the correction bolus?

A: Everyone wants this, but the product is first generation. There was a limit to what could be done in the first round. The whole notion of autobolus came to be by observation. We will see if the next system is any better at ADA in June. You only learn by doing. For example, we’ve learned that you should be able to drop the set point from 120 to 100. That’s possible, it’s now being tested.

Q: When will we get to 100% Time in Range?

A: I’ll be watching that from the rocking chair. But I will be around when we get to 80% for sure.

Dr. Charlotte Boughton on AID Systems: Users Still Need to Know the Fundamentals of Diabetes Management, AID Should be Available for All Populations

Dr. Boughton, who takes the view that everyone can benefit from closed loop, gave a comprehensive talk that was clearly informed by her deep insight into people with diabetes. Covering a lot of ground, she first noted that we are still living in a world of hybrid closed loop therapy, requiring interaction from the user. She emphasized the need for closed loop users to still “get the basics right,” including carb counting, infusion set changes, eating low glycemic index foods, timing the bolus properly, and not over-bolusing (since the algorithm can fill in for any under-dosing). She then reviewed AID work with various patient segments – very young children, adolescents, the elderly, pregnant women, and those with hypoglycemia unawareness. Broadly, she concluded that regardless of the segment, studies show a clear benefit, and that there are always some people in each group who are going to get great results on closed loop. Healthcare providers often have prejudicial assumptions about which people will or won’t do well on closed loop, but these assumptions are often incorrect and should therefore be avoided –for example, facility with technology is not a requirement for success. Since all segments can benefit from AID, Dr. Boughton concluded by asserting that reimbursement should be available for everyone.

  • Other important “basics” of diabetes management include – insulin bolus timing (avoiding post-meal bolusing which can lead to hypoglycemia), preventing insulin stacking, preventing hypos (by looking to see what insulin has been given and using longer acting carbs), and suspending insulin delivery if the pump is unattached.

  • The CARES paradigm, developed at Barbara Davis Center, is a useful, standardized approach for understanding a closed loop system. CARES stands for Calculate, Adjust, Revert, Educate, Sensor/Share. Translated, this means “how does the algorithm work?,” “what can the user do to influence insulin delivery?,” “when does the system revert to open-loop?,” “what training is required?,” and “how does the CGM work and how is data shared and monitored?.”

  • Dr. Boughton reviewed the literature on closed loop in various segments of the diabetes population. She stated “pretty much everybody will gain benefits from closed loop therapy.” She showed:

    • In very young children (2-6 years), closed loop improved time in target from 55% to 73%. They have greater insulin variability especially overnight. Parents spend less time worrying and more time sleeping.

    • Adolescents with poor control (average age 14 years) improved Time in Range from 48% to 67% - a “massive improvement.” MiniMed 670G and Tandem Control-IQ both show benefits in this cohort.

    • In the Tandem Control-IQ pivotal trial, participants with A1c >7.5% improved time in target from 51% to 64% while also improving hypoglycemia.

    • Older adults were able to benefit from closed loop, despite a high burden of hypoglycemia at baseline, being naïve to sensors, and having more visual impairment.

    • There is currently a trial recruiting for those with hypoglycemia unawareness.

    • In pregnant women, (in the delightfully named PICLS trial) closed loop improved time in target from 60% to 75%. Hypoglycemia was unchanged.

CGM Highlights

Dexcom Announces CE-Mark for G6 in Pregnancy; Available in UK “Starting Spring 2020”

In exciting news, Dexcom just announced that G6 has received a CE Mark for use in pregnancy in Europe. G6 had been previously contraindicated for pregnant women. Rollout of the new pregnancy indication will begin with the UK “starting spring 2020” – a reasonable place to start, considering that NHS has already committed to offer CGM to all pregnant women with type 1 diabetes by 2020/2021, given its robust cost-effectiveness in this population (more on this below). Of course, we’re hoping other countries follow the lead of the UK and offer CGM to all pregnant women with diabetes. G6 joins Abbott’s FreeStyle Libre in garnering a pregnancy indication in Europe, with Abbott received CE-Marking in 2017. No CGM (Abbott’s FreeStyle Libre, Dexcom’s G5/G6, or Medtronic’s Guardian Sensor 3) has yet to be approved for pregnancy use in the US, and while off-label use has become much more common, this should help massively with awareness and education and reimbursement.

  • This morning, we spoke with Dexcom CEO Kevin Sayer about the indication:As you look at pregnancy in women who have diabetes, I don’t know you could get [good] outcomes without [CGM]. I think it’s that critical. Patients who’ve been on CGM haven’t been taking it off in pregnancy. I’d give a lot of credit to the CE Mark. It’s good that we’ve made this progress here. This needs to be a tool adopted by everybody. There is no excuse not to provide someone with a baby the best possible treatment and this is the best.”

  • At Dexcom’s afternoon symposium, Dr. Carol Levy (Mt. Sinai) was given ~15 minutes to discuss CGM in pregnancy – we wish she would’ve had longer as she is so fascinating and this is such important news that could have a massive impact on this often high-risk population of women, who are pregnant with diabetes. Dr. Levy shared a very compelling graph based on T1D Exchange data from 2010-2013 vs. 2016-2018. From 2010-2013 to 2016-2018, self-reported CGM use in pregnant women increased from ~35% to 65%; over the same time period, mean A1c in pregnant women dropped from ~6.9% to ~6.6% - we imagine variability is also much lower. Of course, that is Dr. Levy noted that the two trends are not necessarily directly linked, but she suggested CGM use could certainly be a contributing factor. Dr. Levy also pointed at a poster at ATTD 2020 of Dexcom CGM in 50 women from 2012-2019. The pregnant women used CGM (varied from Dexcom G4, G5, to G6) 93% of the time and recorded no DKA or severe hypoglycemia events. Fetal outcomes were also relatively strong: there were just six cases of preeclampsia and six cases of high birthweight (>4 kg). For context, the prevalence of preeclampsia and large gestational age in women with type 1 is 15%-20% 51%, respectively.


  • As a reminder, data supporting pregnancy CGM use are overwhelmingly positive. Results from the JDRF-funded CONCEPTT RCT testing Medtronic’s older Guardian CGM in pregnant women (n=125) were presented at EASD 2017 and showed that CGM use drove a significant reduction in the incidence of large for large for gestational age (OR=0.51, p=0.02), fewer NICU admissions lasting 24+ hours (OR=0.48, p=0.02), fewer incidences of neonatal hypoglycemia (OR=0.45, p=0.03), and one-day shorter length of hospital stay (p=0.01). The numbers needed to treat (NNT) were compelling – NNTs of just 6-8 women with CGM to prevent one of those negative outcomes. The primary A1c endpoint showed a small -0.2% A1c advantage for CGM at 34 weeks (p=0.02). However, mothers on CGM spent a significant 100 more minutes/day in range (68% vs. 61%; p=0.003), 72 fewer minutes/day in hyperglycemia (27% vs. 32%; p=0.03), and a non-significant ~14 fewer minutes per day in hypoglycemia (3% vs. 4%; p=0.1). Excitingly, many believe that results from CONCEPTT actually underestimate the efficacy of CGM in driving positive outcomes in pregnancy, given the improvements with G6 when compared to Medtronic’s older Guardian system.  

    • Cost-effectiveness analysis of CONCEPTT have further underscored the need for CGM in this population. At ADA 2019, Dr. Helen Murphy asserted that “for publicly funded health systems like the NHS, they cannot afford not to provide CGM to pregnancy women with diabetes.” Results back up the claim: The unpublished, retrospective evaluation used individual-level data to estimate cost-effectiveness of CGM vs. SMBG in addition to standard antenatal care. Direct costs associated with CGM, including sensors/transmitter and education, were estimated to be £2,046. Although CGM was associated with an additional £330, it was found to be more effective for maternal outcomes, resulting in 61.33 QALYs (quality-adjusted life-years) as compared to 61.27 QALYs. This translates to a fantastic incremental cost-effectiveness ration (ICER) of £5,509/QALY, far under the NICE threshold of £20,000-30,000/QALY. For neonatal outcomes, the results were even more dramatic: not only was CGM found to be more cost-effective, it was also found to be cost-saving. CGM use led to cost-savings of £2,612, resulting in 75.43 QALYs vs. 73.77 QALYs and an ICER of £1,571/QALY. Results were translatable across geographies as well, with a similar study in Canada providing similar results regarding cost-effectiveness. 

  • Broadly speaking, pregnancy can introduce a host of disruptions to glycemic control and diabetes management that CGM can help ease. Early in pregnancy, patients experience increased glucose variability with progressively increasing insulin resistance, requiring intensification of therapy. The risk of severe hypoglycemia also dramatically rises, occurring up to five times more frequently in women with type 1 diabetes during their early stages of pregnancy as compared to the period before pregnancy. 

12-Month WISDM RCT Extension in Type 1s ≥60 Years Shows CGM Improves Every Outcomes Measured Compared to Baseline: -51 Minutes/Day Below 70 mg/dl, +1.4 Hours/Day Time in Range

Jaeb Center’s Kellee Miller read out primary outcomes from the extension arm of Helmsley- and JDRF-funded WISDM (Wireless Innovations for Seniors with Diabetes Mellitus) study of Dexcom G5 CGM in older adults (≥60 years) with type 1 diabetes. Results from the 6-month, multi-center randomized control trial (n=203) read out at ADA 2019, with 100 of the 103 participants from the CGM arm electing to continue on CGM through 12 months. 94 of the 100 participants randomized to the BGM group also elected to crossover to CGM in the 6-month extension phase. At 12 months, CGM conferred a significant advantage over baseline in every outcome measured. To be included in the study, participants had to have a baseline A1c ≤10%, be pump- or MDI-treated, and could not have worn CGM in the last three months. Patients were excluded if they spent at least 10% of the time with glucose <54 mg/dl during the screening phase and experienced a severe hypoglycemic event in the past six months. At baseline, median age was 68 years, 92% of participants were non-Hispanic white, and 53% used insulin pumps. The primary outcome, percentage of time <70 mg/dl, was reduced from 6.6% at baseline to 2.9% at 12 months in the CGM group (-49 minutes/day; p<0.001); improvements were slightly greater in the BGM/CGM crossover group: time <70 mg/dl decreased from 6.6% at baseline to 6% at 6 months (time of crossover), decreasing to 2.9% at 12 months (-53 minutes vs. baseline; p<0.001). Similarly, time <54 mg/dl was reduced from 2.5% to 0.7% at 12 months (-26 minutes/day; p<0.001) in both the CGM group and BGM/CGM crossover group. In the CGM group, Time in Range increased from 56% at baseline to 64% after 12 months (+1.9 hours/day; p<0.001) and A1c decreased from 7.6% to 7.4% (p=0.01). In the BGM/CGM group, Time in Range increased from 56% at baseline and six months to 60% after 12 months (+58 minutes/day; p=0.007) and A1c decreased from 7.5% at baseline and six months to 7.3% (p=0.02). While not necessarily surprising, the sustained results at 12 months are encouraging to see. The evidence base that CGM is effective in many different populations is building (e.g., COMISAIR, DIAMOND, GOLD) and we strongly hope that the WISDM study can help reduce preconceptions surrounding use of technology and CGM in the elderly – some of the assumptions that they “can’t handle it” or the “caregivers will be very confused” have been downright insulting. While it certainly is true that some technology will not work for some patients, given the incredible heterogeneous patient population, to generalize so negatively is very disappointing. 

 

Baseline

6 Months

12 Months

 

CGM/CGM

BGM/CGM

 

 

 

 

Time <54 mg/dl

2.5%

2.5%

0.6%

2.3%

0.7%

0.7%

Time <70 mg/dl

6.6%

6.6%

2%

6%

3.2%

2.9%

Time in Range

56%

56%

64%

56%

64%

60%

Time >180 mg/dl

37%

37%

34%

38%

33%

37%

Time >250 mg/dl

14%

15%

10%

14%

10%

12%

A1c

7.6%

7.5%

7.2%

7.5%

7.4%

7.3%

Coefficient of Variation

41%

42%

36%

40%

36%

36%

  • During the original study, one severe hypoglycemic event was recorded in the CGM group. After the 6-month extension, the CGM group recorded a total of 5 events across the twelve-month period. By comparison, the BGM arm totaled eleven severe hypo events in the first six months, reduced to just two in the second six-month period (after crossover). Overall, of course, those on BGM did much better in a clinical trial than they would “in real life” – we were very impressed that  12-month time under 54 mg/dL stayed at less than 1% (0.7%), and that time < 70 mg/dL increased only to 3.2%.

  • CGM use was very high through the entire twelve months for the CGM group and in last six months for the group randomized to BGM. Participants in the CGM arm showed CGM use 95% of the time during both six-month periods. After the BGM crossed over to CGM, they recorded 94% of time using CGM. Unfortunately, no data on patient-reported outcomes, such as fear of hypoglycemia, diabetes distress, hypoglycemia unawareness, and general measures of quality of life, has been shared yet – we hope to see more of this.

Dr. Tadej Battelino: Put DCCT in the “History Drawer” Of Your Brain, Lowering A1c No Longer Requires More Hypos; Real-World Data Shows Hypo Exposure Decreases with FreeStyle Libre Scans/Day Across Regions

At Wednesday afternoon’s Abbott-sponsored symposium, Dr. Tadej Battelino (Ljubljana University) told the audience to put DCCT in the “history drawer part” of their brains. To start out his presentation, Dr. Battelino emphasized the detrimental effects of hyperglycemia, in fact using data from DCCT to tie higher rates of complications (e.g., cardiovascular disease, cognitive impairment) with higher A1cs. On the other hand, Dr. Battelino presented data showing non-severe hypoglycemia was protective against cognitive dysfunction, while severe hypoglycemia had no significant effect. Thus, Dr. Battelino summarized the challenge of managing diabetes: lower A1c (i.e., reduce hyperglycemia) while avoiding severe hypoglycemia. Then, Dr. Battelino caught the audience off guard, telling them to put DCCT in the “history drawer part” of their brains. Dr. Battelino was specifically referring to the DCCT finding that rates of severe hypoglycemia increase with decreasing A1c. In a telling graph (see below), Dr. Battelino overlaid data from multiple CGM studies with the same DCCT graph showing a rapid increase in severe hypoglycemia rates as A1c decreased. The CGM study results (circled below) showed no differences in severe hypo rates, whether A1c was above or below 7%. Additionally, the rates of hypoglycemia were markedly lower in the CGM studies compared to DCCT. Given the setting of an Abbott-sponsored symposium and Dr. Battelino’s focus on severe hypoglycemia, we were somewhat surprised we didn’t hear any mention of Abbott’s FreeStyle Libre 2, which adds optional high and low alarms in the same form factor as the original FreeStyle Libre.

  • During his talk at the symposium, Dr. Ramzi Ajjan (University of Leeds) presented real-world data showing that increased FreeStyle Libre scanning frequency was associated with reduced time in both hyperglycemia and hypoglycemia across many countries and regions. The first set of graphs, taken from Dunn et al., 2018, show this same pattern in five European countries. Of course, there are some interesting differences in the different countries’ time spent in hyper- and hypoglycemia – we’d be curious what sorts of cultural, social, and environmentally-driven factors could be driving these variations. Overall, and fascinatingly, the number of minutes per day in hypoglcyemia had very little variability in some countries such as Germany; Italy had the least “time in hypoglycemia” when patients scanned up to 40 times a day at 18 minutes per day – other countries where patients scanned at that rate had a much higher “time in hypoglcyemia” such as France at 40 minutes. For patients who scanned closer to five times a day, time in hypoglcyemia was as high as 57 and 59 minutes a day in Spain and France and as low as 34 and 36 minute a day in Germany and Italy! On the “time in hyperglycemia” side, all patients in all countries shown had between 9.5 and 10.7 hours per day in hyperglycemia if the scans were around five per day – that is between 40% and 45% “time in hyperglycemia” – however that fell to 5.8 to 7.1 hours a day in hyperglycemia or between 24% and 29% “time in hyperglycemia – a massive difference! The most “learning” was in the UK, where the lowest number of scans (~five/day) yielded 44% time in hyperglycemia, which fell to 24% time in hyperglycemia with the higher number of scans – wow! Half-jokingly, Dr. Ajjan suggested that Italy’s relatively low percentages of time spent in hypoglycemia at 1.2% at the low end to 2.3% at the high end were due to pasta – this made us wonder on average how many carbs per day are consumed per country and where the biggest and smallest standard deviations are. “Time in hypo” was higher for other countries, coming in at a range between 1.8% at the low end and 4.1% at the high end for the other countires. Interestingly, Dr. Ajjan also compared the European data to countries in other regions: real-world FreeStyle Libre data from Brazil showed similarly reduced A1c as scanning frequency increased; however, the relationship between hypoglycemia and scanning frequency was much less clear. Lastly, Dr. Ajjan showed unpublished data (picture not shown) from the Middle East, highlighting a marked decrease in both A1c and hypoglycemia as scanning frequency increased – we’d love to see average numbers for those countries as well, in order to think more specifically about regional interventions.

  • Dr. Ajjan also presented real-world data from FreeStyle Libre showing that reductions in hypoglycemia appeared after just one day, while the improvements in hyperglycemia took closer to two months on average to appear. Dr. Ajjan hypothesized that this delay in observed hyperglycemia reduction was because patients needed to review their data in clinics with providers to help identify behaviors that might be causing hyperglycemia; this seems logical to us since they can more immediately avoid hypoglcyemia by identifying it and taking in carbs, etc. – the power of data!

Dexcom-Sponsored Symposium: “Alert Settings Should be Individualized for the Patient”; Time in Range Improvement by ~5% with Increased CGM User Engagement

Dexcom’s Friday-morning symposium featured retrospective data, highlighting Time in Range, A1c, and glycemic variability improvements in patients with higher daily Dexcom G6 feature use (e.g., high alarms, low alarms, Clarity, Share, and Urgent Low Suspend). Dexcom examined data from over 25,000 new patients who started using the G6 between January-July 2019 and measured each patient’s “mean engagement” over three months from August to November. About 9,000 users were categorized as “low” engagement, meaning they used less than 3 features; ~11,000 were “medium” engagement (used 3-4 features), and ~7,000 users were categorized as “high” engagement (>4 features). Patients found with “low engagement” ( had a Time in Range of 56% compared to 61% in those with “high engagement. Patients with the highest levels of engagement also observed 38% reduction in time <54 mg/dl and had reduced glycemic variability: 64% of high utilization users achieved a %CV <36% vs. 57% in the low utilization group. Dexcom has shown similar data in the past and while they are interesting correlations, they are not very surprising and certainly not suggestive of any causative effect: users who utilize more features are likely more engaged in their care and have better glycemic control. We would be curious to learn what sorts of factors drive some to use more features than others.

  • In the same cohort of 26,184 patients, mean Time in Range was 43% for those only using CGM (i.e., no features). For users who utilized Clarity, mean Time in Range increased to 62% and increased again slightly to 64% for those who used Clarity and received at least one push notification. Of course, it’s no surprise to us that those who use Clarity to look at and review their retrospective data have higher mean Time in Range. In fact, much of the power of CGM lies in the ability to look back and identify behaviors that  may be helpful (“Bright Spots”) and harmful (“Landmines”). We wonder how Time in Range might differ in a randomized comparison (i.e., random assignment to use Clarity vs. not use Clarity – we’d feel despondent for the people in the “non-Clarity” group). At AADE, we’d heard that the open rate for Dexcom Clarity app’s weekly summary report notification is 80% - while overall of course, this a very impressive rate, we are also surprised it’s not 100% - everyone cares.

  • Dexcom also shared that its Share/Follow remote monitoring feature was commonly used in pediatric and adolescent populations, peaking at~95% at around age 16 and declined as age increased. In three different cohorts of pediatric patients (<6 years, n=1,054; 6-12 years, n=7,014; 13-18 years, n=7,308), Time in Range steadily improved from ~45% to ~55% for all groups as the number of remote followers increased from zerofrom zero to four or more. 

  • Improved control were also seen across those who chose to customize their alarm settings. For example, in a sample of 19,540 patients, those who set their hyperglycemia threshold below the default (i.e., more likely to trigger alarms) spent ~2.5 hours/day in hyperglycemia (>240 mg/dl) compared to ~4.5 hours/day in those who had their alarm thresholds set above default (i.e., less likely to trigger alarms). The same trend was observed for hypoglycemia, with individuals customizing their alarms to be more sensitive spending ~15 minutes/day <70 mg/dl compared to ~37-40 minutes/day in those who did not. These decisions would, of course, come at the increased risk of alarm fatigue and the ideal thresholds balancing alarm frequency with glycemic control vary greatly by person. On the same topic, a recent article in the Journal of the Endocrine Society estimated ideal thresholds for hypoglycemia and hyperglycemia alarms as 75 mg/dl and 170 mg/dl.

 

CITY Sub-Analysis Shows High Utilization of Dexcom CGM Features in Teens and Young Adults: 91% Use Low Alerts, 81% Use Mobile App, 62% Use Share

The wonderful Dr. Laurel Messer (Barbara Davis Center) showcased data from the Continuous Glucose Monitoring Intervention in Teens and Young Adults (CITY) trial, showing that teens and adolescents have fairly levels of engagement and usage of CGM and CGM features. Primary results for CITY (n=153) read out at ADA 2019, demonstrating efficacy and safety of CGM vs. SMBG in type 1 adolescents (14-24 years). Seventy-four participants were randomized to the CGM (Dexcom G5) group, receiving instruction on how CGM works, optimizing alerts and features, and troubleshooting. As a reminder, Dexcom G5 has non-adjunctive indication, seven-day wear, and requires two fingersticks/day. The CGM group had a mean age of 18, A1c of 8.9% at baseline, and was 65% white. During the study period, 50 of the 74 participants (68%) wore Dexcom G5 for more than 5 days/week. The top reasons for not wearing CGM were wanting a break from daily use, forgetting to use it, and hesitancy/discomfort wearing the device. On average, CGM was used 6.4 days/week with a median of 2 fingersticks/day. Impressively, 98% of participants regularly used CGM without a confirmatory fingerstick before dosing.

  • The vast majority of participants (81%) used the mobile app with their CGM. Slightly more than half of mobile app users (62%) used Dexcom’s Share feature, with 91% of Share users sharing their data with parents/guardians. 13% of the participants shared data with their siblings, while 9% shared with a significant other. The most cited reason for not wanting to use Share was not wanting others to see their glucose values or receive alerts.

  • Nearly all of the participants (91%) surveyed used low alerts, while 84% had high alerts enabled. The median setting for low alerts was 70 mg/dl, while the median setting for high alerts was 270 mg/dl.

  • About half of participants (55%) reported reviewing Clarity reports more than once per month. Another 11% said they reviewed the reports less than once per month, while one-third said they never reviewed their Clarity reports. At AADE, we heard that the open rate for Dexcom Clarity app’s weekly summary report was 80%, though this was across a broader population of Dexcom users. Given the immense learning that can come from retrospectively looking at CGM data (see Dr. Rich Bergenstal’s tips from ATTD Day #1), we wonder how this number could be improved, particularly in the more challenging teen/young adult population.

IDC’s Ms. Mary Johnson Details Hypoglycemia Rates in DCCT/EDIC Cohort Wearing FreeStyle Libre Pro

International Diabetes Center’s Ms. Mary Johnson presented follow-up data from the landmark DCCT/EDIC trial to understand Time in Range using Freestyle Libre Pro CGM, building off of a similar presentation at EASD 2019 by Dr. Rose Gubitosi-Klug. The headline result was that the number of participants reaching Time in Range is poor, and that there is a clear trend of nocturnal hypoglycemia, which is improved by a higher A1c, prior use of pumps/CGM, and non-smoker status. In the original DCCT, hypoglycemia was far worse in the intensive blood glucose control group, and in EDIC, there was a small group of ‘frequent fliers’ who experienced significantly more hypoglycemia. In this study, the n=765 participants (who are now on average 59 years old!) wore a Freestyle Libre Pro for ~twelve days. Their average A1c at baseline was 8.0%. Time in Range was only 52% for all participants and time below 54 mg/dl was 4% - this is supposed to be just 1% and 4% is, of course, nearly an entire hour a day, very worrying. Interestingly, there was a pattern of nocturnal hypoglycemia – time below 54 mg/dl was 4% during the day and 7% during the night (note that the target is <1%, as detailed in the ADA 2020 Standards of Care). While 11% of participants experienced no hypoglycemia, a group of 7% of people experienced >21 episodes of hypoglycemia per day. Although users of pumps and CGM (including insulin suspend pumps) had less hypoglycemia and better Time in Range overall, the entire cohort still has a long way to go to reach the recommended targets. We do note however that FreeStyle Libre Pro overreads hypoglycemia (per the label), so true outcomes may be slightly better.

  • By the end of the original DCCT study, 65% of the intensive therapy group experienced severe hypoglycemia, compared with only 35% of the control group – not surprising given the challenges of insulin therapy. In the EDIC follow-up, 50% of each group had severe hypoglycemia, which was running at ~0.4 events per patient year on average. However, 7% of EDIC participants experienced 30% of all episodes.

  • Overall there was concerningly high levels of clinically significant hypoglycemia and a subset of participants had worryingly high frequency and duration. 7% of participants experienced more than 21 separate events below 54 mg/dl during the study (see table), reminding us of the “frequent fliers” in EDIC. Furthermore, 38% experienced at least one hypoglycemia event (<54 mg/dl) lasting for two hours or more during the day, and 52% at night.

# Episodes of Hypoglycemia

% of Participants

0

10.6%

1-10

60.9%

11-20

21.4%

>21

7.1%

  • On average, users of insulin suspend pumps (presumably Tandem’s Basal-IQ or Medtronic’s 640G and 670G) had on average of ~7% more Time in Range, lower rates of hyperglycemia, and improved time spent <54 mg/dl. They were still some way from the target of 70% Time in Range and <1% time below 54 mg/dl.  (see photo).

  • Only 9% of participants met the Time in Range targets, and only 30% achieved the 4% goal for <70 mg/dl and 28% the 1% goal for <54 mg/dl. The percentage of time spent below 70 mg/dl trended smoothly with A1c and smoking status. Study participants with higher A1cs experienced less hypoglycemia, as did the non-smokers. The group at target (<4% below 70 mg/dl) had a mean A1c of 8.3%. While 4% is technically (barely) at goal, it’s nearly an hour a day. Furthermore, use of pumps and CGM was significantly associated with less time spent below 70 mg/dl, although the trend was not compelling.

Abbott-Sponsored Symposium: Audience Polls Show Strong Enthusiasm for CGM; Dr. Rich Bergenstal’s “FNIR” (Flat, Narrow, In-Range) is a Big Hit

At the same Abbott-sponsored symposium as above, Dr. Tadej Battelino (University of Lubljana) polled the audience before and after the symposium on several questions related to enthusiasm for and adoption of CGM. The pre-symposium survey revealed that speakers Drs. Tadej Battelino, Rich Bergenstal, and Ramzi Ajjan had a tough job: the audience was already incredibly enthusiastic and knowledgeable about CGM. Before the session started, 64% of the audience rated themselves as either “reasonably knowledgeable and very good level of expertise” or “an expert and extremely knowledgeable” about the “technical aspects and clinical indications” for CGM. Perhaps more impressively, more than half (58%) of the audience said that they “currently recommend and/or prescribe” CGM to >60% of their patients with type 1 diabetes – we’d love to know how many of those recommendations/ prescriptions are actually turned into CGM users. By the end of the session, a larger majority (71%) of the audience said they were likely to recommend and/or prescribe CGM to their type 1 patients. Not surprisingly, the numbers were a bit weaker for insulin-using type 2s at the high end, though the differences in “average opinion” of this group was huge: specifically, 23% of the audience said they recommend and/or prescribe CGM to >60% of those patients; this number more than doubled to 57% by the end of the program. A full breakdown of these questions is in the tables below. The impressive audience knowledge is a testament to the incredible momentum seen in CGM over the past decade – in Dexcom’s most recent quarterly call, we got our first update on Dexcom’s user base in over ~two years. Dexcom’s global user base is “approaching 650,000”; combined with Abbott’s two million FreeStyle Libre users, the number of global CGM is well across the 2.5 million mark. At the end of 2018, we estimated ~1.5 million global CGM users and at the end of 2017, we estimated the user base at “just” 0.7-1.0 million.

  • The audience engagement was palpable in the room as Dr. Rich Bergenstal (International Diabetes Center) gave some tips for interpreting CGM data. The audience responded incredibly enthusiastically when Dr. Bergenstal explained his “FNIR” (Flat, Narrow, In-Range) mnemonic for an ideal ambulatory glucose profile – at least half of the clinicians in the room pulled out their phones to snap a picture of Dr. Bergenstal’s slide explaining FNIR. This has also been very popular among patients in the US who have seen it at diaTribe Learn. A couple more of Dr. Bergenstal’s catchphrases (Rich-isms?) also made their way into the presentation:

    • MGLR: More Green, Less Red; refers to the stacked Time in Range bars; and

    • “Thinking fast and slow” – this refers to making “fast” in-the-moment therapeutic or behavioral changes based on real-time CGM numbers or trends and “slow” deliberative analyses of AGPs (many may also know the famous and Nobel-prizewinning book by this name).

    • STAR: this is another one we’ve heard lately though it wasn’t in the presentation – Steady And in Range. We’ve loved patient responses to FNIR – very aspirational – and we also love patients thinking about “steadier” glucose levels (it’s hard to achieve flat for many patients though many are working toward it much more readily with CGM mnemonics!).

My current understanding of the technical aspects of and clinical indications for sensor-based CGM in persons with diabetes is:

 

Pre

Post

At an expert and extremely knowledgeable level

26%

26%

At reasonably knowledgeable and very good level of expertise

38%

36%

At an acceptable level of expertise

21%

28%

At a fair level, in need of improvement

8%

8%

At a very fundamental level

8%

2%

Pre-program question - among all persons with type 1 diabetes, the percentage in whom I currently recommend and/or prescribe sensor-based CGM is:

Post-program question – based on my engagement with the content and expert-based presentations provided in this educational program, among all persons with type 1 diabetes, the percentage in whom I am now likely to recommend and/or prescribe sensor-based CGM is:

 

Pre

Post

<5%

8%

4%

5-15%

10%

4%

15-30%

16%

7%

30-60%

10%

14%

>60%

58%

71%

Pre-program question - among all persons with insulin-requiring type 2 diabetes, the percentage in whom I currently recommend and/or prescribe sensor-based CGM is:

Post-program question – based on my engagement with the content and expert-based presentations provided in this educational program, among all persons with insulin-requiring type 2 diabetes, the percentage in whom I am now likely to recommend and/or prescribe sensor-based CGM is:

 

Pre

Post

<5%

23%

3%

5-15%

12%

4%

15-30%

20%

15%

30-60%

21%

20%

>60%

23%

57%

ADJUST Type 2 Professional CGM Study Scores Cost-Effectiveness of €6,767/QALY (~$7,304); Potential Long-Term Savings of ~€5,000 (~$5,397) and Complication Onset Delay by One Year

Medtronic’s Dr. Simona de Portu presented some of the first cost-effectiveness data from the single-arm Portuguese ADJUST study of the iPro 2 professional (blinded) CGM, highlighting that blinded CGM could reduce costs for patients, over the long-run, by ~€5,000 (~$5,397) due to avoided diabetes-related complications, along with delaying the onset of complications by roughly one year. Medtronic used healthcare consulting company IQVIA’s Core Diabetes Model, an increasingly popular health-economics simulation model, to compare the cohort of patients that received blinded CGM to a hypothetical cohort of patients that did not receive it. The model assumed that patients who received iPro 2 had three blinded CGM visits per year and that that patients who did not receive the professional CGM maintained a flat A1c and visited a doctor four times per year. The model also took into account the costs of the insertion/ review of the data after one week for patients who received the blinded CGM. When comparing the two cohorts, Medtronic found that the blinded CGM resulted in an incremental cost-effectiveness ratio of €6,767 /QALY (~$7,304), an increase in quality-adjusted-life-years (QALYs) from 8.49 to 8.58, and reduced the costs of complications from €68,699 (~$74,150) to €68,084 (~$73,486). To put some perspective on these numbers, Dr. Portu mentioned that the Portuguese government is willing to pay for a medical intervention if it demonstrates an ICER of €50,000/QALY. This amounts to a roughly seven-fold cost differential that blinded CGM can provide for the same impact! Even more impressively, these findings stayed constant when the time horizon of the analysis changed by five, ten, or twenty years, when the assumed cost of complications was increased or decreased by 20%, or when the discount rate was altered. These findings illustrate that despite the increased administrative costs that result from visiting HCPs, purchasing a blinded CGM, and inserting/removing the device, these costs were partially offset by a meaningful reduction in complications and improvements in quality of life.

  • As a reminder, the ADJUST study (n=102 type 2 patients) found that at 12 months, following quarterly CGM applications, each with a follow-up visit (in-person or by phone), A1c levels had dropped by a mean 1.3% (baseline: 9.4%), mean glucose had dropped from 185 to 170 mg/dl, and percent time above 180 mg/dl decreased from 48% to 37% without any increase in time spent in hypoglycemia. Beyond improved glycemic management, patients reported increased treatment satisfaction and better communication with their healthcare providers. From a therapeutic standpoint, patients using blinded CGM were also able to focus their treatment regimens to either pharmacological or behavior modifications after originally using a broad mixture of both adjustments when first starting on the CGM. See our ATTD 2019 report for a deeper dive.  Presumably even more improvements could be made if treatment changes that “worked” best were carefully analyzed – ranging from different therapies to different food choices to different perceived levels of stress.

Clinical Cases of Envision Pro Highlight Utility of Pattern Snapshot Feature and Ability to Drive Behavior Change

The highly regarded Dr. Fiona Campbell (Leeds Children’s Hospital) highlighted two clinical case studies demonstrating the behavioral modifications that Medtronic’s Envision Pro CGM can enable in CGM-naïve adolescent populations. Dr. Campbell contextualized her talk by addressing the enormous burden healthcare providers face in making clinical data relevant and actionable to patients. We appreciated her candor on this front and acknowledgement of how hard positions are for healthcare providers. As patients also tend to view meeting with their clinicians as the most important component of care (according to Dr. Campbell), using technology such as professional CGM that enables both parties to view data together and to collaborate in shared decision making is critical, stressed Dr. Campbell – we certainly agree if these conversations are well set up and patients can discuss the data on an equal footing to the provider.

In the following examples, Dr. Campbell first provided a description of the patient’s demographics, unique lifestyle needs, and specific difficulty in managing diabetes care. Each patient was then placed on the Envision Pro CGM (CE-Marked at EASD 2019) for approximately one week. During a follow-up visit, Dr. Campbell asked the patient to summarize the data, acknowledge any surprises, and suggest meaningful treatment changes. Dr. Campbell even switched seats with the patient to facilitate the patient’s role as a “teacher” who could then actively identify opportunities for improved care – what a patient-centric approach to care!

Arguably, the most informative feature of utilizing the Envision Pro which Dr. Campbell is its CareLink Pattern Snapshot feature (launched in November 2015), which provides the top three trends identified over the week-long period along with potential causes. The software then provides a number of suggestions: (i) oral medication(s) too high or incorrectly timed?; (ii) basal insulin injection in evening(s) too high or missed?; (iii) pre-breakfast insulin incorrectly timed, incorrect dose, or missed?; (iv) insulin to carbohydrate ratio not optimal for pre-breakfast insulin?; (v) inconsistent food intake?; and (vi) exercise around breakfast time? Wow! Patients can also log food, medication, and physical activity. Dr. Campbell mentioned that these features enabled her patients to pick distinct places in the clutter of blood glucose data where they were struggling with diabetes management and critically think about meaningful changes they could make with respect to medications, lifestyle, and other self-care behaviors. We have heard how popular Carelink is among doctors (with some, perhaps even more popular then the technology) and getting this insider view was very helpful.

  • The first case study related to a 17-year old male patient (A1c 9%, T1 duration 10 years) on insulin pump therapy. The patient was a college student studying for final examinations who was keen on moving away from home and into university. Beyond his high A1c, the patient had background bilateral diabetic retinopathy, well-managed hyperthyroidism (taking 125 daily micrograms of thyroxine), and observed episodes of unexplained hypoglycemia. Interestingly, the patient wanted to stick to pump therapy, but had no desire of initiating CGM. After using Envision Pro CGM for eight days and following up, the patient was able to distinctly identify three key periods of out-of-target glucose ranges with the help of Pattern Snapshot: (i) low values post-dinner from 5:00 PM to 8:00 PM; (ii) high values overnight from 11:00 PM to 6:00 AM; and (iii) high values during fasting times of 5:00 AM to 7:00 AM. By comparing day-to-day glucose profiles, the patient could also begin to compare how blood glucose would vary during periods of relaxation on the weekend, days of intense studying on the weekdays, and periods of exercise. Based on the numbers the patient could now visually interpret, he realized that the consistently high levels of overnight glucose resulted from him eating a large meal at the end of the day because of a job he had at a call center. Because of a desire to avoid nocturnal hypoglycemia, the patient would significantly push glucose levels up before sleeping, not only placing him at risk for DKA but also undoing his work during the day of maintaining optimal glycemic management. With respect to behavior change, the individual became reassured that he wasn’t at risk for hypoglycemia and even considered switching to a real-time CGM. We were struck by “even considered” since we think anyone on insulin should be on real-time CGM and it was presented here as less likely (probably because the person with diabetes may not have wanted to wear real-time CGM).

  • The second case Dr. Campbell presented was from a 14-year old female (A1c ranging from 6%-7%, type 1 duration for seven years) who was using an insulin pump without a CGM because of personal confidence that she knew when her glucose levels were out of range. The individual was an academically high-achieving student who played on her school’s netball team but had become frustrated with erratic glucose values while on SMBG. After using Envision Pro for six days, the patient was shocked during her follow-up to see significant glycemic fluctuations. Furthermore, her Time in Range was just 44%, demonstrating a classic case where A1c does not truly provide an accurate representation of true diabetes management. Her Pattern Snapshot highlighted especially high variations overnight and low sugar levels in the early morning during fasting times. Based on her analysis, the patient was able to recognize that her variations likely resulted from personal behaviors including rushed carb counting, not bolusing before meals, and underdosing insulin prior to bed. She left the discussion with a plan of stabilizing basal insulin on a daily basis, investing sufficient time to carefully count carbohydrates, delivering insulin fifteen minutes before meals, pursuing variable dose ratios throughout the day, and even agreeing to wear real-time CGM. We would love to learn specifically how high her “time in hypoglycemia” was – we are imagining it was well over 10%-15%.

  • While at this point, some conference attendees may feel the benefits of professional CGM are very clear – that they can drive meaningful clinical outcomes, behavioral change and improve treatment satisfaction – we loved this session because it made it clear how many HCPs actually don’t necessarily know the power of professional CGM. Or, at least many HCPs may not have as much as experience using professional CGM as might be assumed. However, driving uptake remains an important issue. On this front, at ADA Postgrad 2020, Cleveland Clinic’s Dr. Diana Isaacs characterized professional CGM as an “underutilized resource” despite “good” reimbursement.” Another key question relates to the benefit the type 2 population can derive from it with respect to when it should be deployed and the frequency of wear. At ATTD 2019, Medtronic’s Dr. Robert Vigersky mentioned that only three to six days of CGM wear time may be needed to obtain actionable information. We believe Professional CGM will become more popular as Dexcom’s G6 Pro approved in October 2019 in the US emerges – it is vastly improved, easier to use, and we believe there will be far more improvements and education on how to best use Professional CGM of all kinds in the future.

Dexcom G6 Matches SMBG Accuracy During Exercise in n=24 Study; TIR Data to Be Published at Later Date

OHSU powerhouse Dr. Jessica Castle presented a CGM in exercise study of n=24 adults with type 1 diabetes on MDI, comparing results from self-monitoring of blood glucose (SMBG) and the Dexcom G6 during exercise. Average age was 31 years old, and baseline A1c was 8.8%. The participants were randomized to three exercise types – aerobic, resistance, and high intensity. As expected, glucose dropped on average after exercise, by 47 mg/dl in aerobic and around 20 mg/dl in the other two types. Importantly, the mean absolute relative difference (MARD) did not change for any type of exercise – before, during or after. MARD was significantly affected only at 45 minutes for aerobic exercise and at 30 minutes for resistance training, otherwise it was not different. Excitingly, Dr. Castle noted that Time in Range is in fact the primary outcome of the trial, however this data will be presented at a later date.

  • Study participants exercised using a prepared video of exercises, and an Apple Watch was used to track the exercise. The Dexcom G6 was worn on the abdomen, replaced every ten days, and sensors were not calibrated. Two in-clinic exercise sessions were performed on day nine and day 16 of the trial, roughly half-way through the sensor life. A Contour Next meter was used for capillary blood glucose testing.

  • Impressively, a Clarke Error Grid showed that all the aerobic and high intensity points were in the A and B regions, and all but two of the resistance points were inside A and B. As a reminder, region B is considered to be an “inaccurate” point, but that would not lead to inaccurate treatment.


  • Dr. Castle began her talk by stating that CGM is “now the standard of care for managing type 1 diabetes”. We agree that CGM should be the standard of care, but penetration and reimbursement have a long way to go – that said, certainly in terms of the guidelines, it’s standard of care – while the absence of translation has limited it somewhat, it’s also grown massively in the last year alone.

Selected Questions and Answers:

Q: Isn’t it better to call it a ‘difference’ rather than an ‘error’?

A: I agree completely, and if I had more than ten minutes, I would have gotten into that. There’s clearly a difference between capillary blood glucose and interstitial glucose, and likely that is  magnified in exercise. We can say there is a difference but not necessarily an “error,” but that’s the terminology typically used when comparing sensor data and CBG data.

Q: Did you do try time shifting the data?

A: We did. We did not see any significant delay, so that did not impact the accuracy by time shifting.

Q: Did it matter if the glucose was high or normal coming into exercise?

A: We didn’t look specifically at that, but we’re preparing the publication. We looked at tertiles of drop in glucose and saw that there was a tendency to have higher MARD at the highest tertile.

WaveForm CTO Dr. Rebec Continues to Tout Promising Cascade CGM Data; No Updates on Device Launch (>1 Year Behind Schedule)

CTO of WaveForm Technologies Dr. Mihailo Rebec introduced data for the Cascade CGM device that recently obtained a CE mark and is being launched in partnership with distributor A. Menarini. The device consists of a sensor placed on the abdomen, using a needle-free inserter. The transmitter contains all the processing electronics and appears to be higher off the body compared to the Dexcom G6. An attractive, color-coded app is available, which features predictive alarms. The system requires one calibration per day. Dr. Rebec presented extensive accuracy data. In the device’s pivotal trial (n=57 participants, n=108 sensors), overall MARD was 11.1% compared to YSI, with 99.3% of points in the A and B regions on the Clarke Error Grid. Prior tests had established comparable MARD to the Dexcom G5 and Abbott’s Freestyle Libre. Dr. Rebec also noted that the system has the lowest ecological impact of any commercial CGM system, but no specifics were provided. As of November 2019, the Cascade CGM system was set to launch in Europe, however, we have not heard any updates as of yet (putting launch more than one year behind schedule). Stateside, the company plans to file a 510(k) with the FDA as an iCGM “in 2020” with a US launch “in 2021.”

  • WaveForm has partnered with A. Menarini to distribute the Cascade CGM/GlucoMen Day CGM System. The system consists of an insertion device, the sensor and a phone app. The intention is to provide a CGM device with comparable accuracy to the competition, but also a superior user experience, minimal interference (e.g., acetaminophen) and the lowest ecological impact.

  • The sensor insertion is reported to be less painful than a fingerstick by 70% of participants in a study. Pain was recorded using the Gracely pain scale. The insertion device doesn’t use a guide needle because the sensor is a wire filament and it can pierce the skin. A skin reaction analysis showed a low reaction to the system.

  • 21 potential interferents were tested and cleared, including ascorbic acid, acetaminophen, fructose, lactose, caffeine and metformin. At all therapeutic levels, the sensor bias was within limits for both high and low glucose readings.

Decision Support Highlights

DreaMed Advice4U Study: Advisor Pro Pump Setting Decision Support System Non-Inferior to Expert Physician Advice; Dr. Revital Nimri Says “Equal is Perfect”

The much awaited readout of DreaMed’s Advice4U trial showed that use of the Advisor Pro yielded non-inferior results (p<0.0001) to expert physician advice in n=122 young people with poorly controlled type 1 diabetes. DreaMed Advisor Pro is a decision support system for pump setting recommendations. In this trial, it obtained pump and CGM data from the Glooko diabetes management platform and provided advice on various pump parameters (i.e., pump basal rates, insulin:carb ratios (ICR), and correction factors (CF)). In the Helmsley-funded trial, participants were randomized to either receive physician advice or Advisor Pro advice delivered seven times during a 24-week study. The two groups yielded almost perfectly identical Time in Range across the various cutoff points, establishing statistically significant non-inferiority. Average Time in Range was 55% in the Advisor Pro group and 54% in the physician arm. The secondary outcome, A1c, declined from a baseline of 8.4% to 8.1% in the physician group and 8% in the Advisor Pro group. DreaMed’s Dr. Revital Nimri commented, “Equal is perfect. If we put all our experience into an algorithm [that works], then we have done what we need to do.” The implication is that this automated “expert” system can be scaled to settings where diabetes experience is lower (e.g., primary care), to provide more frequent patient interactions, or to improve clinic visits by reducing the time spent analyzing data and more time discussing diabetes management. The potential for decision support systems is only growing as physicians and other providers become increasingly flooded with data-creating devices, such as CGMs, pumps, and smart pens.

  • Advisor Pro develops recommendations for pump settings based on insulin delivery, glucose, exercise, and food consumption data. The exercise and food consumption data is logged by the user in an app. Advisor Pro is already CE-Marked FDA-cleared to provide pump settings recommendations based on either CGM or SMBG data. In addition to basal rates, correction factor, and carb ratio recommendations, the system also provides some personalized behavioral diabetes management tips. Those tips and recommendations are reviewed by a provider and can be pushed to a patient through a smartphone app.
  • The Advice4U study compared the outcomes of young people aged 10-21 with type 1 diabetes who either used the DreaMed Advisor Pro (Advisor group) or received advice from experts from four specialized academic diabetes centers (physician group). Each center had at least four experts. The primary outcome was Time in Range and time below 54 mg/dl. At baseline, n=122 young people with an average age of 15.5 and A1c of 8.4% were randomized, but the trial reported a per-protocol result of n=60 – those who had completed five out of seven visits.
  • The TIR range results at 24 weeks for the two groups were strikingly similar and met the condition for non-inferiority with high statistical significance (p<0.0001):

Time in Range (Primary Outcome)

Very High

>250mg/dl

High

180-250 mg/dl

In Range

70-180 mg/dl

Low

54-70 mg/dl

Very Low

<54 mg/dl

Physician Arm

17.9%

23.9%

54.7%

2.6%

0.9%

Advisor Pro Arm

17.7%

24.6%

54.3%

2.4%

1.0%

  • A1c decreased in both groups over the course of the trial, declining 0.3% points in the physician group and 0.4% in the Advisor group; the difference between groups (0.2%) at 24 weeks was not statistically significant. The Advisor group used about 10% more insulin, but again it was not significant. Rates of adverse events were also similar.
  • A 50-item post-intervention survey demonstrated satisfaction with Advisor Pro and that it saved time.  Interestingly, physicians implied that there was some variation between the advice they would have given and the automated recommendations. Still, physicians gave the system a 4.5/5 rating for “reliable” and 4.5/5 rating for “useful.” Eleven out of 13 physicians reported that they would keep using the Advisor product. A full breakdown of the physician acceptance data is available in our ISPAD 2019 report.

Artificial Intelligence Accurately Identifies Pediatric Type 1 Patients at Higher Risk for DKA Hospitalizations

Ms. Sarina Dass (Northeastern University) shared impressive data from a study using artificial intelligence to identify pediatric type 1 patients more likely to be admitted to the hospital due to diabetic ketoacidosis (DKA) within the next 180 days. Ms. Dass’ approach used a recurrent neural network looking at over 500 features over two years (e.g., demographics, structured EHR data (CPT codes, lab results, vitals, medications, etc.), A1c trajectories, free-text clinical documents, clinical intake forms, etc.). With data from 1,523 patients ages 8-18 across multiple diabetes centers in the Midwest, the neural network generated a list of all the individuals ranked by likelihood of DKA-related hospitalization in the next 180 days. The occurrence of DKA-related hospitalizations in the dataset within 180 days was ~6% (~90 hospitalizations). The dataset for training and validating the model were the same, an important study limitation noted by UVA’s Dr. Marc Breton during Q&A. In the ten individuals the model identified as most likely to be at risk for hospitalization, all ten did indeed experience a DKA-related hospitalization. For the top 25 most at-risk individuals, about half (12) experienced a hospitalization and for the top 90 most at-risk, a third of them experienced a hospitalization. In other words, the 90 most at-risk individuals identified by the model were indeed at a five-times higher risk for DKA-related hospitalization than the larger dataset. Presumably, the model could help clinics identify patients at higher risk for hospitalization and introduce specific interventions to help reduce likelihood of DKA.

Dr. Bruce Bode Looks Back and Forward at Automated Glucose Control Systems in Hospital Settings

Dr. Bruce Bode summarized decades of experience in delivering automated glucose control in the hospital setting. As background, Dr. Bode founded GlyTec, a company offering the GluCommander glucose management system, which is currently in use by over 200 hospitals. The system, which covers the ER, ICU, step-down wards, discharge, and the transition to insulin at home, decreases mortality and complications and saves money, regardless of the underlying condition. In his talk, Dr. Bode covered the checkered history of glucose control in the ICU, noting that while normoglycemia radically improves outcomes, hospitals struggle to deliver it without increases in hypoglycemia. Given the practical limitations of the hospital setting, the logical approach to this problem is an automated system that takes glucose from meter readings and provides specific dosing recommendations. A plethora of data established that these systems save nursing time, decrease complications, decrease re-admission rate and complications after surgery. Despite this, only about 10% of hospitals use such a system, although it seems inevitable that they will become the standard of care.

  • In 2001, a seminal paper by Van Den Berghe established that controlling glucose in the ICU leads to dramatic reductions in mortality (about 30-40%) and improvements in a whole range of complications, such as infection. Many hospitals tried to replicate these results, leading to the classic NICE-SUGAR study in 2009, which yielded the frustrating result that attempting to attain normoglycemia across a broad range of ICU settings increased the rate of death. After the fact it became clear that it was hypoglycemia that caused the excess death. Many hospitals simply didn’t have the tools to lower glucose safely. Consequently, this led to the weakening of ADA recommendations for glucose in the ICU. Prior to 2001, average glucose was above 200 mg/dl, but from 2001-2007 the ADA target was 80-110 mg/dl, which was then relaxed after NICE SUGAR to 140-180 mg/dl.
  • In the United States, CMS is on the verge of imposing penalties on hospitals that cause hypoglycemic events. CMS is developing a hypoglycemia measure (<40 mg/dl and five or more minutes before the patient gets back to 80 mg/dl) for which they will require reporting and ultimately will levy a fine.
  • For all these reasons, reducing insulin-induced hypoglycemia is a critical issue. It’s remarkable to note that in a typical hospital, 40% of inpatients require insulin at some point. In the first place, 25% of all inpatient days are people with diabetes. 6% of hospitalized patients experience hypoglycemia, and readmission rates for people with diabetes are 20%. In a study at Florida Hospital, if a patient goes under 40 mg/dl at any point during their stay, it was found to cost an additional $10,000 and add about seven days to the length of stay. Mortality is also about three times higher. In Florida Hospital, severe or moderate hypoglycemia cost an excess of $45.6m over a twelve month period, which was largely eliminated using an automated system.
  • Given the inadequacy of current systems, an automated glycemic management system is the best approach for managing hyperglycemia and avoiding hypoglycemia. Currently, less than 10% of all hospitals are using such a system. Commercial computerized systems include GlyTec, EndoTool, GlucoCare and Glucostabilizer. Dr Bode’s company, GlyTec  offers an extended glucose management system (called an eGMS) that covers the ER, ICU, step down ward, long term acute care (LTAC), skilled nursing facilities (SNF), and the discharge process and enables the transition to insulin at home. The system also integrates tightly with EMR and ICU systems.
  • The GlyTec GlucoCommander system provides strong benefit, including time saved for nurses, cost savings to the hospital, and reduced readmission rates. In a test of the system at Kaweah Delta Medical Center, sliding-scale insulin management was almost completely replaced with eGMS, leading to a reduction in whole-stay average glucose from 204 mg/dl to 165 mg/dl, a 25% reduction in hypoglycemia under 70 mg/dl and a 75% reduction in hypoglycemia under 40 mg/dl. Length of stay was reduced by 25% and cost savings were $10m. In another example of patients admitted for diabetic ketoacidosis, the automated system reduced time spent above 200mg/dl from 17 hours to 14 hours and saved 1.5 days in the hospital per event.
  • Dr. Bode highlighted a study of automated glucose control following coronary artery bypass surgery (CABG) that yielded a 20% improvement in complications and 60% less readmission. Although cardiologists were initially very concerned about insulin induced hypoglycemia, the intervention caused no hypoglycemia at all. Patients ended up with an average glucose of 132 mg/dl versus 154 mg/dl in the standard of care comparator. The cost of hospitalization was 10.3% lower in the intervention group.
  • On discharge, an automated system can determine insulin requirements at home. Controlling glucose at home prevents re-admissions. For the hospital to home transition they system advises on the need for insulin and provides dosing information for basal/bolus.

Selected Questions and Answers

Q: What type of glucose measurements do you make?

A: There are only two point of care meters approved in our hospitals at present - Novo Biomedical and Roche. We are testing some CGM solutions, but we don’t have it in the product at this time

Q: Have you used the technology in an outpatient setting?

A: Yes, we have used it for titration of MDI and it works well. We are conducting a study with Abbott Libre. The system provides feedback by texting to their phone. A CDE also provides support for 500 patients.

Diabetes Drugs Highlights

Real World Study of Connected NovoPen 6 in 39 Pediatric Type 1s Shows Significant Reductions in Daily and Nocturnal Hypoglycemia, No TIR Difference

Dr. Peter Adolfsson (University of Gothenburg) presented encouraging results from a real-world trial of Novo Nordisk’s NovoPen 6 in the pediatric type 1 population, demonstrating significant reductions in daily (-31%, p<0.0001) and nocturnal (-24%, p=0.043) hypoglycemic episodes after 12 months. Hypoglycemic episodes were defined as glucose levels <54 mg/dl for at least 15 minutes. Data were based off of collected CGM metrics – notably, Time in Range and time in hyperglycemia were not significantly improved with use of the connected pen.


  • Study design and objective: The study aimed to describe evolution of glycemic parameters following introduction of the connected insulin pen in a pediatric population. Children <18 years old with type 1 (n=39) were enrolled and prescribed connected insulin pens for basal and/or bolus insulin injections across three pediatric diabetes clinics in Sweden. Mean age of the participants was 13.5 years. HCP visits for patients were conducted according to standard clinical practice (about once every six months) and insulin pen data was uploaded during clinic visits or at home via Glooko/diasend. Patients did not review data outside of clinic visits, where data was reviewed with HCP and changes to insulin regimens were made accordingly.

  • At baseline, participants experienced a mean of ~0.5 hypo episodes (<54 mg/dl for >15 minutes) per day. After >12 months using the connected NovoPen 6, the rate of hypo episodes fell to ~0.3/day, a 31% relative reduction (p<0.001). The percent time spent <54 mg/dl also showed a relative reduction of 14% after 12 months (p=0.03), though the baseline was not given. Time in Range did not change significantly from the very low baseline TIR (51%), nor did time >180 mg/dl (42%). These results are somewhat surprising (and disappointing) as the results showed significant increases in the total daily dose of insulin, particularly the mean basal dose, which rose by 12% after 12 months compared to baseline. In an ideal world, this is not close to where patients want to be though we believe many on insulin are hugely out of range. The 51% TIR does suggest that the basal rates and/or other ratios like insulin sensitivity are out of whack and we wonder what the result would’ve looked like with more interaction, possibly virtual only, with the HCPs. 

  • Previous work from a trial of the NovoPen 6 in adults (presented at two posters at ADA 2019, n=94 at 12 Swedish sites) showed more promising results. Two-weeks of CGM metrics at follow-up (14 days after the final clinic visit) were compared to two-week baseline metrics (14 days following obtaining the pen). Compared to baseline, mean time-in-range improved from 38% to 46% (+1.9 hours/day), driven by time >180 mg/dl decreasing from 49% to 42% (-1.8 hours/day). 81 adults were included in an adherence analysis, which showed a very impressive 43% fewer missed meal boluses (baseline: 0.74/day) using the connected pen. Again, we believe these numbers reinforce a truth we’ve long held – so many people on insulin are using the wrong doses and/or ratios and being “out of control” is not new for many.

  • As a reminder, the NovoPen 6 and Echo Plus are already CE-Marked and a launch beginning in Europe is expected in 2Q20. The reusable NovoPen 6 and Echo Plus will display the last insulin dose and time on their ends, have an 800-dose memory, and five-year battery life. Data from the pens can be downloaded with NFC and the pens are compatible with both basal (Levemir, Tresiba) and bolus (NovoLog, Fiasp) insulin cartridges. The NovoPen 6 will allow adjustments down to 1 U with a maximum 60 U dose, while the pediatrics-targeted Echo Plus will allow adjustments down to 0.5 U with a maximum 30 U dose. 

Profs. Thomas Danne and Stephanie Amiel Face Off on Debate Over SGLTs in Type 1 (Video)

To an absolutely packed room, Profs. Thomas Danne and Stephanie Amiel engaged in a thoroughly riveting debate on SGLTs in type 1, bringing much-appreciated novel perspectives to the often discussed lighting rod topic. Prof. Danne argued in favor of SGLTs in type 1 while Dr. Amiel took the opposite stance for the purposes of the debate. Following the debate, moderators asked the audience whether they would use SGLTs in type 1s without established CVD and whether they would use the class in type 1s with established CVD. On the first point, the room was split 40% yes and 60% no. On the second question, the room was overwhelmingly in favor, 95% yes to 5% no.

Overall, this debate was a tremendous back-and-forth of key points on a topic that we’ve heard from thought leaders extensively over the past few years, with many novel points emphasized throughout. We harken back to a similar debate between Drs. Bruke Perkins and David Nathan at ADA 2019 which was equally strong. Hats off to Profs. Danne and Amiel for an incredible debate – see the table below for a summary of the key points presented by both sides:

Are SGLTs as an Adjunct Therapy in Type 1 Diabetes Worth It?

IN FAVOR: Prof. Thomas Danne

OPPOSED: Prof. Stephanie Amiel

SGLT inhibitors provide clinically meaningful benefits on a host of metabolic and other parameters, including A1c (without increasing risk of hypoglycemia), weight, blood pressure, and time in range.

Dr. Danne summarized the numerous phase 3 trials of SGLTs in type 1, showing that these benefits have been consistently shown across multiple trials.

SGLTs have also been shown to increase patient quality of life

Type 1 diabetes is an insulin deficiency disease, and therefore the treatment of type 1 is insulin replacement. SGLTs do not directly address this fundamental pathological process in type 1.

Dr. Amiel believes that resources would be more properly spent on diabetes education for proper insulin usage vs. spending on SGLTs and ketone monitoring strips. 

Dr. Danne discussed the evidence pointing to cardio and renal benefits of SGLTs in people with type 2 diabetes and even without diabetes (DAPA-HF). He said that although there is no hard evidence yet that these benefits will translate to type 1, the need is clearly large and he believes that benefit can also be derived in this population.

Dr. Danne also made an impassioned call on this front, noting that in his work in pediatrics, the need for CV risk reduction is even greater, given that data from the Swedish registry shows that young girls with type 1 face 17 years of lost life and 14 years for boys due to CV risk w/ type 1.

Dr. Amiel pointed to real world DKA risk in the US of type 1s on SGLTs, which show rates much higher than in the background type 1 population. She explained that given these higher rates and normal rates of death from DKA, >1,700 people would die each year from DKA who were taking an SGLT in type 1.

Dr. Amiel questioned the applicability of the consensus paper on DKA risk mitigation. She pointed out that a few of the recommendations (such as “do not offer to people with lower BMI” and “do not offer to people on low insulin doses”) are not supported by data. She also questioned the practicality of “regular” ketone monitoring as defined by the paper, given the multiple cases where such monitoring would be needed.

Dr. Amiel also worried that providers in the real world would not actually look into these detailed paper and protocols when prescribing to patients, and therefore significantly elevate risk.

Also on DKA, Dr. Amiel noted that the largest DKA risk is actually in the 25-64 age range, which is the actual population we would actually want to most use this drug class in.

Dr. Danne conceded that there obviously is an increased DKA risk across the clinical trials, but that the numbers on this risk is low, and that this risk can be mitigated with proper education and training.

He pointed to the consensus document published last year on mitigating DKA risk and new educational tools that ATTD has created regarding DKA as important ways to help in educating clinicians/patients.

Dr. Amiel suggested that the weight loss effects may be overstated. Weight loss achieved by SGLTs is only 29% of that estimated from the calorie loss due to glucose excretion, and is actually coupled with a 13% increase in calorie uptake due to increased hunger. More generally, the positive glucose and weight effects seem to not persist past six months.

Dr. Danne emphasized that this is an important situation where it is important that patients have the choice to make an informed decision on whether they should be on this therapy or not, given the risk/benefit profile.

Dr. Amiel also mentioned higher rates of UTIs, especially in women, with SGLTs. She also brought up the reports of Fournier’s gangrene associated with the therapy.

Dr. Julio Rosenstock Peers into His Crystal Ball and Envisions Future Treatment Guidelines Focused on Combination Therapy Instead of Sequential Therapy

Dr. Julio Rosenstock provided a powerhouse speed lecture that highlighted his view that the future of type 2 treatment guidelines should move away from sequential treatment to early, aggressive combination therapy. Looking into his “crystal ball,” Dr. Rosenstock gave his hopeful prediction for what this may look like: (i) initial combination therapy with metformin and an SGLT-2 inhibitor/GLP-1 agonist becoming the preferred standard therapy in newly diagnosed type 2; and (ii) when basal insulin is needed, fixed ratio combinations of basal insulin + GLP-1 becoming the preferred option. Dr. Rosenstock’s talk here was an abridged version of a longer and more detailed presentation on a very similar theme at CMHC 2019 – see our coverage of that talk here. He noted in the opening of this talk that he was glad to see so much interest in the type 2 space at ATTD, and we would add that enthusiasm and interest from the audience was very high for this presentation.

  • Dr. Rosenstock detailed the evolution of treatment guidelines over the years, hammering home the point that they have mainly relied on sequential therapy recommendations. Traditionally, therapy has been “stepped-up” based on A1c checks over time for each patient, which has led to significant therapeutic inertia in terms of pursuing more aggressive treatment to successfully combat disease progression. Regarding ADA/EASD consensus guidelines updated in 2018, Dr. Rosenstock noted that although they do represent an important step in the right direction in terms of more seriously considering personalized CV risk, a more aggressive approach is still necessary. He commended session co-chair Dr. Chantal Mathieu for her work on putting together the guidelines as part of the writing committee but explained that “these guidelines are not going to take care of the problem of clinical and treatment inertia. We need to be more aggressive in management.”  In this vein, he harkened back to the famous Albert Einstein quote regarding insanity being “doing the same thing over and over again and expecting different results.” How long can the diabetes community continue to rely on a sequential treatment paradigm in diabetes care when outcomes are not necessarily improving for patients? Dr. Rosenstock firmly believes that it is time to deviate from this frame of thinking and move toward aggressive combination therapy use early on in treatment, especially considering the vast armamentarium of efficacious drugs (SGLT-2s, GLP-1s, fixed-ratio combos, and more) now available to patients.

  • Dr. Rosenstock listed out his principles for effective combination therapies, stressing that these principles simply “make sense” and are not “rocket science.” His six principles are: (i) components should exhibit complementary actions; (ii) glycemic control should be better with the combo than with each individual component; (iii) combined doses may be lower than each individual component alone; (iv) side effects should not be increased and ideally be mitigated with the combo vs. separate therapies; (v) treatment should be simplified as to improve adherence and persistence; and (vi) cost should be lower than the sum of the costs of each individual component.

  • Dr. Rosenstock pointed to results from the VERIFY study as a “proof of concept” regarding further adoption of combination therapies. Results from VERIFY for the first time provided robust evidence of durable glycemic control with initial combo therapy. As a reminder, VERIFY (presented at EASD 2019) showed that early combination therapy of metformin and Novartis’ DDP-4 inhibitor Galvus (vildagliptin) doubled the time to initial treatment failure vs. metformin monotherapy. We imagine that results would have been even more robust with usage of an SGLT-2 or GLP-1.

  • Dr. Rosenstock closed his talk with his own algorithm for simultaneous therapy-centric decision making. See slide below. At diagnosis, combo therapy of metformin + SGLT-2 is used, and thereafter a host of various combo therapies are utilized based off of A1c targets.

Passion for Prevention: Dr. Marian Rewers Gives Updated Type 1 Prevalence Stats from n=23,423 Children in ASK; IM Therapeutics to File IND for D-Methyldopa in March/April 2020

JDRF’s Dr. Jessica Dunne, Senior Director of type 1 prevention, and pediatrics luminary Dr. Mark Sperling led an enthralling, standing room only session on type 1 screening and early intervention. A number of presenters expressed gratitude towards the audience for coming to a session on prevention despite ATTD being a predominately technology-oriented conference. As Dr. Desmond Schatz (University of Florida Health) eloquently stated, “Thank you for being here while we pursue a biological cure and biological prevention, while at the same time, this meeting serves to update us on technology as we pursue an artificial pancreas.” Looking at the enraptured audience throughout the presentations, it was clear to us that screening and prevention both remain huge areas of interest for the broader diabetes community. Read on below for a selection of top highlights from the session.

  • Barbara Davis Center for Diabetes’ Dr. Marian Rewers opened the session with updated statistics from the Autoimmunity Screening for Kids (ASK) program – through January 2020 (!), detailing a substantial 1.0% prevalence of pre-symptomatic type 1 diabetes. As a reminder, ASK aims to identify children early in disease progression, in order to prevent DKA and potentially direct participants to further prevention studies, as well as collect epidemiological data. Thus far, ASK has screened an incredible n= 23,423 children from the Denver Metro area, of which more than 90% had no first-degree relatives with type 1 diabetes. As such, ASK is “truly a general population screening.” Furthermore, 51% of study participants were Hispanic, 35% non-Hispanic white, 8% African American, 2% Asian American, 1% Native American, and 3% Other or Unknown, making ASK one of the most diverse screening studies to date, with the potential to shed light on diabetes prevalence in minority populations.

    • Of the ~23,000 children screened, 796 (3.4%) tested positive at the initial screening, and 143 (1.0%) had persistent confirmation at second screening. Of the 1.0%, 122 (0.5%) tested positive for multiple islet antibodies, predicting a 70% ten-year risk of developing clinical diabetes, and 121 (0.5%) tested positive for a single high-affinity islet antibody by ECL, predicting a 49% ten-year risk of diabetes. Importantly, 90% of children detected did not have a relative with type 1. 

    • As one of ASK’s goals is to assess the benefits and cost-effectiveness of universal screening, we were also excited to see a recently submitted cost-effectiveness analysis for screening. The analysis, authored by Dr. R. Brett McQueen, showed that screening for type 1 diabetes will be cost-effective if it decreases the risk of DKA by 1 in 5 and subsequently lowers A1c by 1.0%. So far, of the 18 children who have developed diabetes after confirmation from ASK, only one has experienced DKA compared to the 11 that would be expected, hinting that cost-effectiveness could be feasible using the program’s screening and education program.

  • Co-founder & CMO of ImmunoMolecular Therapeutics (IM Therapeutics) Dr. Peter Gottlieb gave an exciting update on next steps for the company, including an upcoming NIDDK-funded phase 2 study (n=36) with TrialNet in children with pre-stage 1 or stage 1 diabetes for L-methyldopa, a drug traditionally used as an anti-hypertensive. Dr. Gottlieb anticipates that the trial will begin “late this year or early next year” and is anticipated to complete in March 2022, according to ClinicalTrials.gov. An earlier JDRF-funded phase 1b trial of patients with recent-onset type 1 and positive for DQ8 demonstrated that treatment with L-methyldopa successfully blocks HLA-DQ8 and reduces inflammatory T-cell response to insulin. The single-arm, open label dose escalation study (n=20) found that DQ8 presentation was 40% inhibited compared to baseline levels, with 85% of patients showing reduced inflammatory T-cell responses toward insulin. Interestingly, while the upcoming TrialNet study and past JDRF study are examining L-methyldopa, Dr. Gottlieb confirmed that the company intends to take the drug’s chemical enantiomer D-methyldopa to clinical trial. As the human body does not metabolize D-methyldopa, Dr. Gottlieb anticipates that the drug should have a longer half-life, act more potently, and have fewer side effects, as many of L-methyldopa’s side effects are presumed to come from metabolization. IM Therapeutics plans to file an IND for D-methyldopa in March or April 2020, and a phase 1a trial is slated for June 2020. Presumably, the company’s $10 million in Series A financing, led by the JDRF T1D fund and Morningside Ventures, will help fund these studies.

Star-Studded Session Highlights Unmet Needs in Type 1 Diabetes and the Potential of SGLT Inhibitors for Cardio- and Renal- Protection (with Appropriate DKA Risk Mitigation) (Video)

Friday’s agenda kicked off with a star-studded session on the ever-relevant issue of SGLT inhibitors in type 1 diabetes, featuring Drs. Satish Garg, Chantal Mathieu, and Thomas Danne. Dr. Garg (Barbara Davis Center, Aurora, CO) set the stage with an overview of unmet needs in type 1 diabetes care. He underscored that despite vast improvements in insulin formulations, technology for insulin delivery, and glucose monitoring over the past decade, population-level outcomes show only a small fraction of the type 1 population meeting A1c targets, and an increasingly large fraction developing overweight and obesity. Furthermore, the total lifetime costs of type 1 diabetes (factoring in medication costs and direct and indirect costs of hypoglycemia and diabetes complications) totals ~$0.5 million per person per year, or about $1 trillion in lifetime costs for the US diabetes population of ~1.25 million. The bottom line, he emphasized, is that more treatment options for type 1 diabetes are sorely needed – particularly in light of the burden of overweight/obesity and diabetes complications such as cardiovascular and renal disease, all of which SGLT inhibitors are well-poised to address.

Turning to efficacy, Dr. Mathieu (KU Leuven, Belgium) ran through the phase 3 data for SGLT-2 inhibitors dapagliflozin (DEPICT) and empagliflozin (EASE) and SGLT-1/2 inhibitor sotagliflozin (inTandem) in type 1 diabetes, remarking on the within-class consistency in improving A1c, weight, blood pressure, and time-in-range. She noted, “in my hands as a clinician, it is one of the most effective drug classes, and you can see the effects from day 1.” On the negative side, SGLT inhibitors are also associated with increased risk of genital mycotic infections and, more concerningly, DKA. Dr. Mathieu underscored that risk mitigation strategies are essential to manage the increased risk of DKA, and for this reason she only recommends SGLT inhibitors to a select group of patients with strong adherence, high engagement in their diabetes self-management, relatively high insulin requirements, and no signs of underweight. In her clinic, this translates to 58 patients out of 1,400, only one of whom experienced DKA (in relation to an episode of gastroenteritis). The EMA has approved sotagliflozin and dapagliflozin for type 1 diabetes, but stateside there has been more resistance to the idea of SGLT inhibitors as adjunctive therapy for type 1 diabetes. Recently the FDA released a CRL for sotagliflozin (March 2019) and dapagliflozin (July 2019) and a November Advisory Committee Meeting on empagliflozin culminated in a 14-2 vote against approval for type 1 diabetes – all largely due to concerns over DKA risk.

On this note, Dr. Danne closed the session with a deep dive on risk mitigation strategies for DKA. He emphasized the need for much more comprehensive patient and provider education about DKA, as well as the importance of patient selection for SGLT therapy (the ideal patient for SGLT therapy, in terms of DKA risk, is highly engaged, not underweight, has relatively high insulin requirements, and is not on a low carb or ketogenic diet).

  • Echoing his commentary from day 1’s debate session on adjunct SGLT inhibitor therapy in type 1 diabetes, Dr. Danne underscored that while DKA risk is a concern to be addressed, we shouldn’t downplay the potentially game-changing benefits of SGLT inhibitors in terms of cardio- and renal protection. These are well-established for the type 2 diabetes population, and with the approval of dapagliflozin for heart failure we know that these benefits extend also to the diabetes-free population at high risk of CV disease. Dr. Danne argued that people with type 1 diabetes are at even greater risk of CV disease than this population, and arguably even the type 2 diabetes population: “It is very sobering to read that if you develop diabetes prior to age 10 it leads to 17 years shorter life expectancy for girls and 14 years for boys according to Swedish registry data.” Dr. Danne further argued that SGLT inhibitors should be studied in the pediatric type 1 diabetes population – “I think we shouldn’t wait until these children develop CVD to offer cardioprotective therapies.”

Locemia’s Co-Founder Mr. Robert Oringer Details Personal Narrative Regarding Development of Nasal Glucagon

Locemia co-founder Mr. Robert Oringer provided a heartfelt, detailed personal narrative regarding the development of nasal glucagon, subsequent acquisition by Lilly, and commercialization as Baqsimi. As a reminder, Lilly acquired worldwide rights to nasal glucagon from Locemia in October 2015 after successful phase 3 studies of the candidate had been completed in both adults (see coverage from ATTD 2015) and pediatrics (see coverage from EASD 2015). Mr. Oringer noted that his presence on the stage today was “highly unusual,” seeing as that normally when a large pharma company acquires a small company’s product, the founder of the small company “usually disappears into the sunset.” However, Mr. Orginer’s personal connection to the development is quite noteworthy and fully captured the audience at this Baqsimi symposium.

  • Mr. Oringer began his story 30 years ago, explaining that he was a young entrepreneur just beginning his relationship with diabetes management by selling thinner lancets and glucose tablets. In doing this, he begin to gain an understanding of the specific aspects of diabetes care, especially in hypoglycemia. This relationship deepened when over the course of one month, both of his sons (one at age three and the other at nine months) were diagnosed with type 1 diabetes. Mr. Oringer soon realized just how deeply the fear and anxiety related to hypoglycemia impacted daily life for his entire family – he and his wife were constantly thinking about “who had our kids’ back” in emergency situations. His experiences having kids with type 1 deepened his understanding of how fear of hypoglycemia leads to situational avoidance, constant compromises, and loss of confidence: “I thought I had empathy during the early years when I was marketing diabetes treatments, but I realized that I didn’t have real empathy until I raised kids with type 1.” 

  • Mr. Oringer explained that he sat on this feeling for ten years after the diagnosis of his children, and decided that he wanted to do something so that his children “could feel that people had their backs” regarding the treatment of hypoglycemia. Mr. Oringer had no previous pharma experience, however, so he partnered with Dr. Claude Piche to tackle this problem. He and Dr. Piche soon realized in an epiphany moment that what is truly unique about glucagon is that it is a drug that is prescribed to patient, but intended for someone else to actually use for that patient. That third party user is most likely unfamiliar with injections as well. The two partners therefore keyed in on four developmental goals for their glucagon candidate: it would have to be non-invasive, easy to use, easy to teach, and easy to carry.

  • Jumping off of this, they legally formed Locemia Solutions, a mission driven company focused on making hypoglycemia rescue treatment simpler (fun fact: Locemia comes from low blood sugar hypoglycemia). Locemia intended from the start to pursue the nasal route of glucagon delivery (past research had shown the viability of glucagon absorption in the nose), but soon realized that a liquid nasal glucagon approach was not feasible due to stability concerns. Therefore, a powder formulation was necessary. The next step was to create a device for nasal delivery (Mr. Oringer joked that the only such device known at the time was a rolled up $100 bill!). Through a series of steps, the device was finally made, and the nasal glucagon candidate successfully progressed through human factor studies as well.

  • After this, Lilly discovered the candidate, and Mr. Oringer fondly noted that “they embraced us and our innovation and really adopted our baby. Lilly has shown a deep commitment to fulfilling the promise of the product, and taking it through the necessary regulatory and commercial steps, which was something we could have never done on our own.” In a very nice closing touch, Mr. Oringer also explained the origin of the brand name “Baqsimi” – it’s based off of the idea that “having someone’s back” is the root of treatment for hypoglycemia, and ties in perfectly with Mr. Oringer’s personal narrative of the drug’s development. Mr. Oringer closed his touching remarks by expressing that his “real hope is that nobody will ever have to use Baqsimi.”

Buzzing NAFLD/NASH Session Highlights the Strongest Drugs in the Competitive Landscape, Epidemiology, and Machine Learning Approaches for Improved Diagnosis

A morning session on the NAFLD/NASH therapy landscape drew a standing-room-only crowd (impressive for a Saturday morning, and at a tech conference no less!). The highlight was Dr. Cristophe de Block’s (Antwerp University Hospital, Belgium) tour de force of the ever-growing landscape of therapies in development for NAFLD/NASH. In a very helpful synthesis, he segmented the many drug classes into those that target insulin resistance, inflammation, lipid metabolism, and fibrosis, as shown below.

Anti-insulin resistance

  • TZDs

  • GLP-1 agonists

  • GLP-1/GIP dual agonists

  • SGLT-2 inhibitors

Anti-inflammatory

  • Vitamin E

  • Dual PPAR α/δ agonists

Lipid metabolism correcting

  • FXR agonists

  • SCD-1 inhibitors

Anti-fibrotic

  • CCR2/CCR5 antagonist

  • Thyroid receptor β agonists

  • FGF-21 analogs

Within these, Dr. de Block expressed particular optimism for the generic TZD pioglitazone, Lilly’s GLP-1/GIP dual agonist tirzepatide, and Intercept’s FXR agonist obeticholic acid (now under priority review with the FDA), but also underscored that “the first line of treatment is a lifestyle approach, the second line of treatment is a lifestyle approach, and the third line of treatment is a lifestyle approach” given the tight association between NAFLD/NASH and diabetes and obesity (just 7% weight loss will halt the worsening of fibrosis and >10% weight loss leads to resolution of NASH and regression of fibrosis), and the fact that so many of these agents in the NAFLD/NASH competitive landscape have yet to undergo larger and longer-scale trials. In particular, he outlined that the cornerstones of any NAFLD/NASH treatment plan (regardless of which drugs are eventually approved) should be caloric restriction, reduced fructose intake, reduced daily alcohol intake, and increased physical activity.

  • Nodding to the mixed clinical trial results, hepatologist Dr. Zobair Younossi (Inova Fairfax Hospital, Falls Church, VA) pointed out the difficulty in clinical trial design for a disease whose natural history involves progression and regression between the increasingly severe stages of steatosis, fibrosis, and cirrhosis. Dr. Younossi explained that having 10% of a population spontaneously revert to a less advanced phase of NAFLD/NASH is not unusual, but the occurrence of this in the placebo arm of a trial can muddy the interpretation of the results. Additionally, NAFLD/NASH candidate therapies have a different impact on the steatosis, fibrosis, and cirrhosis given the different pathophysiological mechanisms underlying each stage of the disease. This infinitely complicates clinical trial design and raises challenging questions about patient selection and segmentation.

  • Dr. Younossi also provided a deep dive on the epidemiology and burden of NAFLD/NASH. In lockstep with the global obesity and diabetes epidemic, NAFLD/NASH is on the rise, currently affecting a staggering ~25% of the global adult population, and ~50-60% of the global type 2 diabetes population. In the US alone, it is estimated that 6.65 million adults have NASH, and within this nearly 700,000 have the most severe advanced fibrosis. Particularly concerning, Dr. Younossi pointed to the rise of this condition in children as an “emerging tsunami.” Currently NAFLD affects ~5-7% of children in North America, Europe, Asia, and Africa, and as high as ~25% in South America. Overall the global lifetime cost of NASH for the entire patient population is estimated at $222 billion, and no doubt this will only rise as liver disease strikes younger patient populations.

  • In a discussion of NASH diagnostics, University of Colorado’s Dr. Kavida Garg highlighted the growing interest in using computational and machine learning approaches to detect liver steatosis and fibrosis from MRI images. Currently the gold standard for NASH diagnosis is liver biopsy – a costly and invasive procedure that is not scalable at the population level. Machine learning analysis on top of the MRI imaging that is already performed in NAFLD/NASH patients could be game-changing for detecting liver disease earlier, and screening more widely. As Dr. Garg put it, “images are more than pictures – they have a lot of data.” We’ll be following this closely.

Big Picture Highlights

SWEET Type 1 Registry (Europe, Australia, Canada, India) Shows 62% of Children Use Technology (CGM and/or Pump), Tech Use Associated with Lower A1c

Dr. Thomas Danne (Hannover Medical School) presented some quick results from the SWEET type 1 registry, estimating that 62% of children use at least one diabetes device (i.e., CGM or pump). The results were calculated from 25,654 children (18 years or under) with type 1 diabetes in the SWEET registry, which spans 21 centers across Europe, Canada, Australia, and India. The data analyzed was collected between August 2017 and July 2019. In the cohort, mean age was 13.8 years and 39% were on MDI and SMBG. Nearly one-third (30%) used both a pump and CGM, while 17% used a pump and SMBG and the remaining 14% were on MDI and CGM. Notably, “more than 85%” of the subjects using pumps were in either Europe or North America. These distributions did not vary by a great amount across age groups. Adjusted for demographics and region, mean A1c of CGM + pump users was ~0.9% lower than that of SMBG + MDI users. Dr. Danne highlighted pump use as being associated with lower rates of severe hypoglycemia, while rates of severe hypo were highest in MDI + CGM users.

 

MDI + SMBG

MDI + CGM

Pump + SMBG

Pump + CGM

A1c

8.7%

8.3%

8.1%

7.8%

DKA

2.9%

2.9%

2%

2%

Severe hypos

2.4%

4.3%

1.1%

2.1%

Enthusiasm Builds for Nasal and Injectable Next-Gen Glucagon Hypoglycemia Rescue Treatments (Video)

Back-to-back talks from Yale’s Drs. Jennifer Sherr and Stuart Weinzimer highlighted recent advances in glucagon rescue therapy for severe hypoglycemia, both subcutaneous and nasal. Dr. Sherr highlighted Lilly’s nasal glucagon Baqsimi, FDA approved in July 2019 as the first-ever non-injectable rescue treatment for hypoglycemia. She emphasized that nasal glucagon represents an enormous innovation for the field. Current glucagon reconstitution kits are cumbersome, time-consuming, and error-prone, as they require mixing and injecting steps that often cannot be performed by the individual experiencing the hypoglycemic event. On the other hand, Baqsimi will come in a single-use dispenser that can directly administer nasal powder glucagon without the need for preparation or injection. To this end, Dr. Sherr highlighted the human factors study for Baqsimi, in which 6x as many caregivers correctly gave a full dose of nasal glucagon in 8x less time. It’s safe to say that people with diabetes who at risk for severe hypoglycemia have been waiting for an alternative for quite some time, and Dr. Sherr underscored that this therapy also comes as a giant sigh of relief for their families and care partners. The highlight of the presentation, in our view, was Dr. Sherr’s demonstration of how to use Baqsimi – as a person with type 1 diabetes herself, she keeps a vial with her in her purse. We were also pleased to hear Dr. Sherr highlight our very own Ms. Kelly Close’s testimonial about her personal experience with Baqsimi, available here at diaTribe Learn. Kelly feels particularly strongly that every person on prandial insulin or SFUs should own a next-generation emergency glucagon treatment (Baqsimi or Xeris) and that with these products, the risk of going to the ER (anytime and especially at the time of COVID-19) is far less likely.  

  • Turning to liquid-stable glucagon preparations, Dr. Weinzimer underscored the game-changing convenience of these vs. cumbersome reconstitution kits. Xeris’ GVOKE glucagon injection was FDA approved as the first treatment of this kind in September 2019; launch is expected later this year in the forms of a pen and a pre-filled syringe. Other liquid-stable glucagons in the pipeline include Zealand’s phase 3 HypoPal rescue pen (dasiglucagon) and Adocia’s phase 1/2 Bio-Chaperone glucagon. Another application of liquid-stable glucagon is use in dual hormone AP systems, and to this end Dr. Weinzimer highlighted BetaBionics’ iLet bionic pancreas, which includes dasiglucagon alongside insulin. BetaBionics recently unveiled strong phase 2 results from the dual hormone AP’s first-ever home-use outpatient trial (90% of dual-hormone iLet users had a mean CGM glucose level <154 mg/dl, vs. 50% with the iLet running on insulin only). A phase 3 trial is slated to begin later this year. Ethically, we are not certain that given how hard the older emergency kits are to use, whether they should even still be sold given the massive improvement. Why would any patient newly diagnosed, as an example, be encouraged to get anything but a next-gen kit? And with that in mind, why would any established patient get an older kit?  

Dr. Jay Skyler Provides Sweeping Overview of Type 1 Cures Landscape w/ Particular Focus on Beta Cell Replacement Efforts (Video)

In the final talk of ATTD 2020, Dr. Jay Skyler (University of Miami) looked toward the future of the type 1 cures landscape, with a particular focus on beta cell replacement efforts. Dr. Skyler’s sweeping overview of the field was relatively optimistic and reviewed the many potential strategies toward functional cures that are being studied. Broadly speaking, we’ve noted renewed investment and work in this area of beta cell replacement and type 1 cures in recent years (see Reflections from 2018 and 2019), and commend Dr. Skyler for bookending this meeting by highlighting these advances and pointing toward a hopeful future.

  • Dr. Skyler mentioned several private and public sector initiatives in beta cell replacement efforts, highlighting ongoing work at Orgenesis, ViaCyte, Semma, Novo Nordisk, Lilly, and others.  He noted that scientific advances have allowed the application of stem cell technology to drive forward most progress in the beta cell replacement field, and touched on the diverse strategies that stakeholders can choose to push forward in this space (patient specific stem cells, human embryonic stem cells [hESC], or induced pluripotent stem cells [iPSC]). Regarding the relative benefits/drawbacks of each of these approaches, Dr. Skyler explained that with an autologous (patient specific) stem cell approach, immune rejection can be avoided – a huge plus in ensuring that patients would not need to be under chronic immunosuppression therapy. On the other hand, the process is not as scalable, given that biopsies are needed of each individual patient, and specific manufacturing processes are needed for each individual. With allogenic stem cells (either hESC or iPSC), the process is “generic” in the sense that the same cells could be used for multiple patients. The process is also centralized in a more controlled setting, leading to a more scalable and translatable process for commercialization. ViaCyte and Semma are both pursuing this strategy, while Orogenesis is using the autologous route (see bullet below).

  • We were glad to hear Dr. Skyler give a small update on progress at Orogenesis, as we haven’t heard much on this project in recent years. Dr. Skyler noted that the company is “on the verge of beginning human clinical trials” and that this will happen in 2020 or 2021. Orogenesis was mentioned in relation to it being the only company to conduct autologous stem cell research (patient specific stem cells, coming from the same host), which is a multi-step process that requires the modification of multiple cellular processes. The idea here is to revert liver cells to a suggestible stem-cell state, induce pancreatic differentiation, then fine tune the signaling pathway to allow generation of cells which secrete insulin – certainly no small feat. Dr. Skyler commented that preclinical work from Orogenesis has shown that differentiated cells show “total physiologic insulin response to glucose and the production of C-peptide” – a fantastic signal as the approach moves into clinical work.

Dr. Richard Bergenstal Channels James Taylor’s “Fire and Rain”, Calls for Community to Focus on Improving “Rain” (Cascade of Diabetes Care) While Spotting “Fires” of New Diverse Causes of Mortality in Diabetes

On a talk focused on the worrying trend of increasing rates of diabetes complications after decades of progress, IDC’s esteemed Dr. Richard Bergenstal picked out three pieces of literature identifying causes for concern and gave his prescription for the path forward. After reviewing over 100 literature articles in preparation for this talk, Dr. Bergenstal honed in on three papers of note that brought the following problems to focus: (i) a stagnant cascade of diabetes care in the US (as evidenced by this recent JAMA publication); (ii) a resurgence of diabetes complications in recent years (evidenced by this JAMA viewpoint); and (iii) new, diversified causes of diabetes mortality (see this Lancet publication).

  • Dr. Bergenstal underscored that the current cascade of diabetes care in the US is inadequate, with only one in four patients at goal in terms of reaching goals for A1c, blood pressure, cholesterol, and smoking status. Of course, much of this can be attributed to the dismal state of healthcare coverage in the US, which Dr. Bergenstal said must be addressed. Outside of this, he noted that there needs to be innovations in terms of glucose management, seeing as this is often the biggest barrier out of those four factors. On this front, he called for a systematic new type 2 algorithm that he calls the “CKG” algorithm, which stands for CVD, CKD, and Glucose. The CKG moniker comes from the similar “EKG” acronym used in cardiology, and CKG focuses on the three main components that Dr. Bergenstal sees as critical to delivering high-quality diabetes care. Notably, the “G” for glucose can also stand for CGM and AGP, which Dr. Bergenstal importantly stressed as part of going beyond A1c when considering glucose. Invoking Albert Einstein’s famous quote on the definition of insanity, Dr. Bergenstal remarked that “Einstein would say that if you’ve been using A1c measurements to get better A1c’s for over thirty years, then maybe should probably try something different now [such as CGM metrics].” We couldn’t agree more, and as always, are thrilled to hear Dr. Bergenstal hone in on thinking beyond-A1c, especially in regard to type 2 care.

  • Along with focusing on the diabetes cascade of care and reducing complications, Dr. Bergenstal stressed that the field must simultaneously make sure not to miss the new “fires” of diverse causes of mortality. New epidemiological data show that declining rates of vascular disease mortality are leading to a diversification of forms of diabetes-related mortality, with important implications for clinical management, prevention, and disease monitoring. These “new fires” include often-underrecognized causes in diabetes for mortality such as sepsis, cancer chemotherapy, the flu, liver disease/NASH, and amputations. As rates of CVD improve in terms of treatment and mortality rates, clinicians and patients must not forget these other areas.

Dr. Irl Hirsch on the Future of SMBG in the Next Five Years: CGMs Exploding in the US and Beyond, but SMBG Access is Still Abysmal in Many Regions

Dr. Irl Hirsch (University of Washington) began his talk by polling the audience for a show of hands of how many think SBGM will go away within the next 5 years. Less than half of the room raised their hand. His presentation began with a stroll down memory lane, starting with a fun history of clinical glucose monitoring from 1841 (urine testing) to the present. The first US patent for a glucose monitor was filed in 1968, followed by the Ames Eyetone meter hitting the market in 1972. Over time, blood glucose monitors became smaller, faster and chock full of features. Dr. Hirsch went on to question why he was asked to give a talk on SMBG, when the era of CGM has “exploded” with 80-95% of his patients with type 1 diabetes and up to 30% of his patients with type 2 diabetes on CGM in his clinic in Seattle, WA. The difference in the U.S. CGM penetration from 2015 to 2019 shows a whopping 432% increase for type 2 diabetes and a 52% increase in type 1 diabetes – see dQ&A’s data for more. Still, although US coverage for CGM is expanding, both commercial and Medicaid coverage remains highly variable across the nation. This means that SMBG may be the only option for many US patients, especially those that don’t use insulin. Looking outside the US, in many low- and middle-income countries, SMBG still has a long way to go. The table below illustrates that the availability of glucose test strips is worse than that of insulin in most countries. At ISPAD 2019, Australia’s Dr. Graham Ogle shared powerful data showing that simply moving the state of diabetes care in low-resource countries from “minimal” (fingerstick testing only at clinic, human premixed insulin only, minimal education) to “intermediate” (2-3 fingersticks/day, human insulin MDI, diabetes education) could greatly improve A1c, reduce complications, and mortality.

  • Dr. Hirsch showed a slide from Inkwood Research, projecting the CGM market will grow to $15.2 billion by 2027 (32.1% compound annual growth rate). During the same time, they estimated the SMBG market will grow by $22.3 billion (5.2% CAGR).

New Epidemiological Evidence Points to Association Between Diabetes and Cognitive Impairment (Video)

In a standing-room-only morning session, Johns Hopkins’ Dr. Elizabeth Selvin presented compelling epidemiological evidence for an association between type 2 diabetes and cognitive decline/dementia. Longitudinal results from the ARIC study – a population-based study encompassing the metro areas and surrounding suburbs of Minneapolis, MN, Jackson, MS, Winston-Salem, NC, and Washington, DC – demonstrated that mid-life diabetes is associated with a 19% greater degree of cognitive decline over 20 years (Rawlings et al. 2014; n=13,351) and late-life diabetes is associated with a 23% greater risk of cognitive impairment, the precursor to dementia, over five years (Rawlings et al. 2019; n=5,099). This rose to a 58% increased risk of cognitive impairment for people with longer (>5 years) diabetes duration, and a 73% increased risk for people with an A1c >7%. As discussed in the ensuing Q&A, the extent to which the greater risk of adverse cognitive outcomes with higher A1cs is due to hyperglycemia itself vs. greater glycemic variability remains unclear, and an active area of interest.

  • Diabetes is a well-established risk factor for dementia, conferring approximately double the risk, but the influence of diabetes on the preclinical stages of dementia has been less explored. The ARIC study’s demonstration that diabetes impacts cognitive decline and cognitive impairment strongly suggests that interventions to prevent or reduce the severity of diabetes could also be deployed as early-prevention efforts against dementia, before the first stages of cognitive impairment set in. What a win it would be for public health if two of society’s most costly and burdensome diseases, diabetes and dementia, could be addressed together with the same set of interventions! The idea of exploring glucose and metabolism-related targets for new dementia therapies is already an incredibly hot topic in the dementia world, and there has been increasing buzz about the possibility of repurposing diabetes drugs, particularly GLP-1 agonists, for Alzheimer’s disease. This is a field we’ll continue to watch very closely.

  • Dr. Selvin further underscored that the adverse impact of diabetes on cognitive extends to hypo- and well as hyperglycemia. Another analysis from the ARIC study showed that, over 14 years of follow-up, people with diabetes and a history of severe hypoglycemia had a ~2.5-fold greater risk of developing dementia than people with diabetes alone (Lee et al. 2018). With both hypo- and hyperglycemia implicated in the increased risk of dementia and cognitive impairment, we wonder whether cognitive protection could eventually become another piece of evidence for the importance of time-in-range.

  • On the type 1 diabetes side, Dr. Nelly Mauras (University of Florida) presented evidence of widespread hyperglycemia-related changes in the brain from MRI studies comparing children with vs. without type 1 diabetes.   Specifically, children with type 1 diabetes showed significant differences in gray and white matter growth in several brain regions over time, as well as greater brain activation during memory tasks – interpreted as an indication of compensation for detrimental structural brain changes. Dr. Mauras elaborated that her group is currently exploring whether these changes in the brains of children with type 1 diabetes can be reversed with intensive glycemic control. We remain curious whether these changes are unique to type 1 diabetes, particularly in the developing brain, or are also present in response to the later-life hyperglycemia that characterizes type 2 diabetes.

Dr. Simon Heller on Cognitive Consequences of Hypoglycemia (Video)

Dr. Simon Heller (University of Sheffield, UK) gave a fascinating overview of the cognitive consequences of hypoglycemia, and the potential of CGM to help prevent this. He underscored that in addition to short-term concerns such as transient cognitive dysfunction, confusion, and behavior abnormalities, hypoglycemia is also associated with long-term risk of cognitive impairment and possibly even dementia. Results from the longitudinal Edinburgh Type 2 Diabetes Study show that the frequency of severe hypoglycemia is negatively associated with cognitive ability, and a recent meta-analysis demonstrated that use of insulin, with its accompanying hypoglycemia risk, increases the risk of dementia by approximately 50%  (pooled HR: 1.54, 95% CI = 1.14-2.007). Turning to younger patients, data from a Norwegian population study demonstrated that children with type 1 diabetes who had experienced severe hypoglycemia before the age of 10 had worse cognitive function after 16 years of follow-up; the effects were most damaging for children who experienced severe hypoglycemia before the age of 6. Though these findings are extremely concerning, Dr. Heller ended the talk on an optimistic note. Greater use of CGM can help people detect hypoglycemia events they were previously unaware of based on symptoms alone. This raises the possibility of identifying new risk factors for hypoglycemia, predicting risk of severe hypoglycemia episodes, more precisely measuring the impact of symptomatic and asymptomatic hypoglycemia on cognitive function, and – very crucially – greater precision in measuring the effectiveness of new diabetes therapies and technologies designed to reduce the risk of hypoglycemia.

Social Media Session Touches on #LanguageMatters, Pros and Cons of Social Media, and Digital Peer Support (Video)

A lively session on social media addressed the pros and cons of online platforms and the growing #LanguageMatters movement. The session was headlined by the dynamic duo of Renza Scibilia and Chris “The Grumpy Pumper” Aldred, two fantastic type 1 diabetes advocates who discussed the #LanguageMatters movement and how HCPs can interact with their patients in a more understanding and empathetic way. The centerpiece of the conversation was a recent viral Twitter thread in which an ophthalmologist advocated for giving people with diabetes a “wake up call” about retinopathy by threatening them with pictures of diseased retinas. Ms. Scibilia and Mr. Aldred both emphasized that the use of such “shock tactics” work for very few people with diabetes. In most cases, it will cause patients to tune out, or worse, paralyze them into inactivity. Ms. Scibilia underscored that “we know how serious diabetes is – we’re living with this in our bodies every day.” She advocated changing the conversation to “care, rather than complications” – what people with diabetes need is an action plan, not another reminder of the burdens of diabetes. Mr. Aldred, who was recently diagnosed with a foot ulcer, his first diabetes complication, underscored that stigma about complications is a giant weight on the shoulders of people with diabetes. He told the story of telling his podiatrist how crushed he felt when she called him “non-compliant” – thankfully, his podiatrist has now vowed to never use that term again. He gave the following advice to HCPs in the audience: “In an appointment, your first question should be ‘how are you, and how’s your family?’ because what influences this will influence your diabetes.”

  • NHS England’s Dr. Partha Kar followed with a discussion of the pros and cons of social media, drawing upon his own recent negative experience on Twitter involving a harsh backlash against his post picturing a weekend pancake breakfast (to the tune of “how dare you as a diabetes leader eat unhealthy food?!”). Despite the negativity and divisiveness that social media can bring, Dr. Kar said that he still uses it first and foremost because it is a fantastic platform for education, information gathering, and interacting with colleagues and people with diabetes. Alluding to recent game-changing advances in diabetes therapy and technology, Dr. Kar expressed concern that these advances so far have favored only the few, improving life for people who already have ample resources to manage their diabetes, and leaving those without even further behind. He argued that social media can be a democratizing force in this regard, spreading peer support and self-management tips more broadly in the population.

  • University of South Hampton’s Dr. Claire Reidy underscored the benefits and pitfalls of structured peer support and its importance in diabetes self-management. Some benefits include shared learning (as people spend more time engaging in self-management than working with an HCP), online sources of support that allows people to access and address issues, and improved feelings of self-efficacy and group belonging. Potential pitfalls include a lack of accountability and credibility, no clear guidelines, potential judgement from peers, and misinformation. Dr. Reidy remained optimistic that structured peer support in the form of blogs, forums, and more can indeed be an important “lifeline” that provides emotional support in addition to traditional clinical outcomes.

  • Dr. Neal Kaufman (CMO of Canary Health) tackled an interesting topic: the use of “Digital Therapeutics” for people with diabetes. Digital Therapeutics (DTx) are a form of digital peer support communities, and can be an important means of delivering therapeutic interventions to patients directly. As an example, Dr. Kaufman cited “Better Choices, Better Health,” a six-week, digital peer-to-peer workshop that provides tailored education and action planning to a group of 25-30 participants at a time. The workshop not only increased self-efficacy among its participants, but also improved their confidence to adopt health behavior change. According to Dr. Kaufman, best practices for online digital support include increasing self-efficacy as a primary outcome, tailoring content that enables community discussions, and including skilled peer coaching. In Q&A, when asked about how physicians can assist in improving self-efficacy, Dr. Kaufman responded that doctors must ask these three questions to their patients: (i) What do you think is wrong?; (ii) What at do you think is causing it?; and (iii) How can I be of help?

Time in Range and Beyond A1c Highlights

GOLD Study Data (n=137) Shows Just 27% of Participants with A1c <7% Also Meet Consensus Time in Hypoglycemia Goals Even With CGM

University of Gothenburg’s Dr. Shilan Ahmadi presented a post-hoc analysis of the Swedish GOLD CGM trial (n=137), focusing on the new consensus CGM goals which are now included in the ADA Standards of Care. The cross-over study took place over ~6 months comparing CGM (Dexcom G4) and SMBG in adult type 1 MDI users (mean age 45 years). Dr. Ahmadi’s sub-analysis of the participants with A1c <7% highlighted the difficulty of achieving both A1c <7% and the time in hypoglycemia goals, even with CGM. For participants in the study with A1c <7%, mean time below 70 mg/dl was 5.4% (consensus goal is <4%) and mean time below 54 mg/dl was 1.5% (consensus goal is <1%) on CGM. Though these values do not meet the consensus targets, they are a notable improvement from the outcomes seen on SMBG: for participants with A1c <7%, mean time below 70 mg/dl and 54 mg/dl were 9.2% and 3.5%, respectively, on SMBG. During the CGM use period, only ~one-quarter (27%) of individuals with A1c <7% also met the <4% time below 70 mg/dl and <1% time below 54 mg/dl goals. The graph below really underscores the difficulty – the green box was added by us and highlights the areas where both A1c and time below 70 mg/dl targets are met. Only three (!) participants (out of 137) met both goals when using CGM and just one participant met both goals on SMBG.

  • The full GOLD study showed an A1c of 7.9% during the CGM period vs. 8.4% during SMBG use. Percent time below 70 mg/dl was 2.8% on CGM, compared to 4.8% on SMBG; percent time below 54 mg/dl was 0.8% on CGM, compared to 1.9% on SMBG. See all the key results here.

Session on Real-World Evidence Highlights Low Achievement of Time in Range Targets in the Diabetes Population (Video)

In a session on real-world evidence (RWE), Dr. Irl Hirsch (University of Washington) provided a deep dive on the available real-world CGM data on Time in Range targets in the diabetes population. Based on the available studies (summarized in the chart below), it is estimated that real-world Time in Range (TIR) is ~51%-65% for people with type 1 diabetes and ~62%-69% for people with type 2 diabetes. Similarly, the % coefficient of variation (%CV) for glucose is ~35%-45% in type 1s and ~29%-30% in type 2s. We’ve also taken our stab at estimating the population’s average Time in Range and come out with similar, perhaps slightly lower, estimates. Dr. Hirsch underscored that there is vast room for improvement here (especially given that the people in these studies were using CGM and therefore likely had better TIR than those without), but the recommended targets for TIR and %CV (>70% and <33% respectively) are well within reach. On the issue of Time in Range, Dr. Hirsch also discussed how measured A1c can vary from CGM-predicted A1c (Glucose Management Indicator, GMI). According to a study of CGM users in his own clinic (n=717), over 50% of patients had an A1c >0.5% off from the CGM-predicted value, and 22% were 1% off. This discordance is extremely concerning, and Dr. Hirsch pointed out that an A1c mismatch of 0.5% is “concerning” and a 1.0% difference is clinically significant. A larger issue is how measured and estimated A1c values will be interpreted as not all clinicians are aware of this difference. As Dr. Hirsch put it, “which one are patients, physicians, and payers supposed to accept?” As CGM use becomes more widespread, this is an issue that the field will have to confront though with Time in Range, we believe many patients won’t care (A1c is of course useful from a research perspective and vs old numbers).

  • Turning to pediatrics, Joslin’s Dr. Lori Laffel underscored that real-world evidence points to vast unmet needs in the adolescent type 1 population. A recent study showed that only 14% of children using CGM and MDI and 28% of children using CGM and an insulin pump achieved Time in Range >70%. Similarly, only 17% and 26% respectively achieved the percent time above 180 mg/dl goal of <25%. Dr. Laffel offered several suggestions for how to improve this situation – namely increasing the use of CGM (encouragingly, CGM use has risen from 7% of the type 1 population in 2012 to 30% in 2018), including exercise as part of routine activities, improving our understanding of how macronutrients affect Time in Range, and improving AID systems.

 

  • Earlier in the session, UCSF’s Dr. David Klonoff set the stage with a discussion of what exactly constitutes real-world evidence (RWE). As he put it, as opposed to a randomized control trial (RCT) which assesses whether an intervention can work under ideal conditions and has been considered the “gold standard” to evaluate the efficacy of the treatment, RWE addresses whether an intervention can work under usual clinical practice conditions. RWE can come from a variety of sources (consumer data, disease registries, survey data, electronic health records, insurance claims, etc.), and is subject to selection bias and reporting bias, but in many cases these are much more feasible than a RCT because they save an incredible amount of time and money. Unlike a RCT, a pragmatic RWE trial offers the intervention to all participants by using usual care as the comparator group and helps identify unanticipated and often rare complications.

Dr. Roy Beck Reviews Relationship Between TIR and Complications, Argues for TIR Over A1c in Clinical Trials

Dr. Roy Beck (Jaeb Center) provided an overview of the association between Time in Range and various diabetes complications, calling for greater use of Time in Range over A1c in future clinical trials. Dr. Beck’s own 2018 Diabetes Care paper delved into DCCT data from 1983-1993, computing approximate Time in Range from seven-point glucose measurements. The 7-point glucose profile measures glucose pre- and 90-minute post- meals and at bedtime on the same day every three months. Over this 10 year period, for each 10% drop in Time in Range, retinopathy progression rate increased by a whopping 64% and the risk of microalbuminuria increased by 40%. Dr. Beck pointed out that using A1c instead of percent Time in Range to examine the rising rates of complications produced nearly identical results, a kind of validation for the legitimacy of using Time in Range in this context. Another 2018 study, Lu et al. found a significant relationship between Time in Range and retinopathy in people with type 2 diabetes (n=3,262). For those with >86% Time in Range, the prevalence of vision-threatening retinopathy was only 3.5%, vs. 9.7% for those with <51% Time in Range. Furthermore, mean Time in Range decreased according to retinopathy severity, from 68% for those with no retinopathy to 64% for mild retinopathy, 58% for moderate retinopathy, and 59% for vision-threatening retinopathy (p<0.001). Of course, while these data are compelling, both studies had only limited data (one day of seven measurements ever 3 months in the DCCT, and only 3 days of CGM data in the Lu et al. study). To this end, Dr. Beck praised upcoming trials such as PERL (Preventing Early Renal Loss in Diabetes) and the DRCR Retina Network Fenofibrate Protocol, which incorporate blinded CGM in the study design for up to 10-14 days every 6 months to assess the impact of Time in Range on kidney disease and retinopathy, respectively. According to Dr. Beck, a growing body of evidence suggests that there is a stronger association between vascular complications and longitudinal CGM measurements. This presents a compelling case for regulators to accept CGM and Time in Range as a meaningful endpoint for clinical trials. He ended on a provocative note: “A1c is a surrogate measure for glucose, but with CGM the actual glucose content is measured. Do we still need a surrogate?”

 

CONCEPTT Analysis Shows Pregnant Women with Type 1 Diabetes Struggle to Meet Target CGM and A1c Goals; A1c Better Correlated with Pregnancy Outcomes than CGM

Using data from the CONCEPTT trial, Spain’s Dr. Diana Tundidor evinced that it is still difficult for pregnant women with type 1 to achieve target glycemic goals even with the help of real-time CGM (RT-CGM) wear. According to the consensus document drafted at ATTD 2019, and subsequently the ADA 2020 Standards of Care, pregnant women with type 1 diabetes have lower CGM metric targets: (i) <25% time above range (>140 mg/dL; TAR); (ii) >70% Time in Range (63-140 mg/dL; TIR); (iii) and <4% time below range (<63 mg/dL; TBR). Though TIR was found to increase over the course of pregnancy, only 8% of study participants reached target TIR during the first trimester (n=226), 10% at Weeks 24-25 (n=203), and 35% at Weeks 34-35 (n=177). Furthermore, a statistically significant improvement in TIR for study participants using real-time CGM rather than masked CGM (control) was only seen during the third trimester. While only a limited proportion of women achieved any of goal ranges during the study, Dr. Tundidor explained that she believes an earlier commitment to reaching target values will be the biggest help in improving these rates of attainment, rather than simply making the goals less stringent. Regardless, it’s clear that gestational diabetes is an area of much needed research, and given the increased roll out of CGM in diabetes, we especially hope to see real world data soon. 

  • In addition, goal achievement was linked to a number of improved pregnancy outcomes. Achieving TAR goals (>140 mg/dL) at Weeks 34-35 was associated with reduced risk of preterm birth and large for gestational age (LGA) birth, and meeting target TAR at Weeks 24-25 was similarly associated with lower rates of LGA. Surprisingly, achieving target TBR (<63 mg/dL) at Weeks 24-25 was associated with an increased risk of neonatal hypoglycemia and neonatal intensive care unit (NICU) admission. In order to further investigate the effects of TBR, Dr. Tundidor’s group then analyzed combination TIR and TBR, finding that co-TIR+TBR achievement was associated with reduced risk of LGA at baseline and Weeks 34-35.

  • In terms of target A1c goals, the ADA 2020 Standards of Care recommends a value of 6.5% for the first trimester, followed by 6.0% during the second and third trimester. Results between TIR and A1c were comparable, and again, a statistically significant benefit for those on real-time CGM was only demonstrated in the third trimester (31% at target vs. 17% on masked CGM, p=0.032). Interestingly, A1c was found to be superior to CGM values in the prediction of pregnancy outcomes. Achieving the target 6.0% at baseline was associated with decreased risk of LGA. Meeting the target of 6.5% at Weeks 24-25 correlated with decreased preterm birth, LGA, neonatal hypoglycemia, and NICU admission, while attaining target A1c at Weeks 34-35% was only linked to decreased preterm birth, LGA, and neonatal hypoglycemia.

New Differential Equation Model for A1c and Time in Range Produces Correlation Coefficient of 0.93, Discussions with FDA Underway

To a fully packed room, Dr. Chiara Fabris (University of Virginia) proposed a new model for relating A1c with Time in Range. The correlation coefficient for the model’s estimated A1c and lab-measured A1c was an impressive 0.93; For context, the traditionally used linear models correlating TIR with A1c from Drs. Roy Beck (2019) and Robert Vigersky (2019) have correlation coefficients around 0.6-0.7. The new model, explained at DTM 2019 by UVA’s Dr. Boris Kovatchev, uses one personalized, glycation rate parameter that is unique for each individual. In other words, the model requires one set of Time in Range and lab A1c values to be “calibrated” to that specific individual. The model also includes three fixed population parameters (i.e., biological constants that are the same for all humans). Using the dataset from iDCL 1 (mobile-enabled inControl, Dexcom G4/G5, Roche Spirit Combo; n=120), the three fixed population constants were calculated. Then, individuals’ A1c values from month three of iDCL 3 (Control-IQ embedded in t:slim X2, Dexcom G6; n=168) were used to calibrate the individual glycation rate variable for each study participant. The equation was used to estimate participants’ A1c daily and then compared to reference lab A1c at months 6 and 9 of iDCL 3.

  • In addition to the impressive 0.93 correlation coefficient between model-estimated A1c and lab-measured A1c, variability in eA1c and A1c was also had a positive correlation coefficient of 0.8. The impressive correlation is perhaps best illustrated by a slide shown at DTM 2019 (see below). The blue dots represent estimated A1c and lab A1c scatter using a linear correlation, while the red dots represent the differential model. Note that the red dots are bunched much more closely around the perfectly correlated eA1c = lab A1c line.

  • Towards the end of presentation, Dr. Fabris noted that the team has discussed this model with the FDA and further discussions are underway. We are interested in hearing more details about the discussion and what this new metric could mean in the near future from a regulatory standpoint. The FDA has generally maintained A1c as the primary efficacy endpoint for regulatory consideration, but it definitely is not surprising that Time and Range and A1c are intricately linked, given that they are driven by the same phenomenon (blood glucose) – TIR is of course much easier to use.

Guido Freckmann Calls for Standardization of CGM Accuracy Measurements, Criticizes MARD, and Questions Time in Range Consensus Goals

Guido Freckmann (Ulm University) gave a provocative talk on the need for standardization of CGM accuracy measurements similar to what was done years ago with A1c. He discussed the importance of a traceable and unbroken chain of calibrations to adequately compare the CGM accuracy to a reference method. Indeed, we heard similar concerns about CGM accuracy, particularly CGM point values, from FDA’s Dr. Courtney Lias at DTM 2019. While regulatory submissions for CGM generally include studies comparing CGM values with YSI reference, different CGMs have different performance levels at various glucose concentrations. One comparison that comes to mind is FreeStyle Libre vs. Dexcom – FreeStyle Libre does seem to report more time in hypoglycemia than G5; see poster at EASD 2019. In particular, Dr. Freckmann recommends following the 2015 ISO Standard for Self-Monitoring of Blood Glucose (ISO 17511:2003) for in vitro diagnostic medical devices for CGMs. In 2016, we saw the FDA tighten accuracy standards for BGM, requiring 95% of results to be within ±15/15% of reference values. Of course, with the creation of special controls for iCGM in the US, these tighter accuracy requirements could become table stakes if iCGM becomes the de facto standard for CGM in the US.

  • Dr. Freckmann suggests that although a MARD of less than 10% has become a de facto standard for non-adjunctive labeling, the metric is subject to variability depending on the study design and the population tested (e.g., type 1 vs. type 2 diabetes). MARD numbers can also be misleading based on the number of paired points in various glycemic ranges. Most experts in the field have long recognized the limitations of MARD as a catch-all metric for CGM accuracy, and indeed the FDA special controls for iCGM don’t even specify a MARD requirement, instead focusing on ±15/15% and ±20/20% measures.

 

  • Notably, Dr. Freckmann also questioned why the 2019 consensus CGM metric goals did not include accuracy of CGM devices as a “prerequisite.” Indeed, without a guarantee for the accuracy of a CGM in hypoglycemia, is the target for <4% time <70 mg/dl measured by a CGM actually meaningful? Additionally, without a standard for CGM accuracy, there is also no reason to believe that the cutoffs set by the consensus are clinically meaningful. Dr. Freckmann’s fellow countrymen Dr. Lutz Heinemann expressed a similarly skeptical viewpoint at DTM 2019 and in a paper published in JDST.

Table taken from Dr. Heinemann’s publication in JDST.

Digital Health and Telemedicine Highlights

“Virtual Specialty Clinic” Pilot Study (n=36) Affirms a Promising Future Where Remote Onboarding of CGM is Possible (and Successful!): 95% CGM Wear, 1.1% A1c Reduction After 12 Weeks (Baseline 8.3%)

As promised at the 2019 AADE Technology Summit, Northwestern’s Dr. Grazia Aleppo unveiled encouraging pilot data (n=36) from the Jaeb/Helmsley/Cecelia Health virtual, diabetes clinic (we've referred to this in the past as "Geek Squad"). Impressively, study participants remotely initiated on CGM (n=27 type 1s, n=7 type 2s) – including prescription, shipment and education – used the device an average 95% of the time (6.9 days/week) during the 12-week study period and saw a statistically significant -1.1% decline in A1c from 8.3% at baseline. Time in Range, estimated from A1c at baseline, also jumped from 48% to 59% (p<0.001) over the 12 weeks. Unsurprisingly, the participants with highest baseline A1cs saw the greatest benefits from CGM: those with A1c >9% at baseline saw a 1.8% A1c reduction with CGM; those with A1c <7.5% at baseline saw a 0.4% A1c reduction. Furthermore, patient-reported questionnaires showed increases in glucose monitoring satisfaction, trust, and diabetes technology attitudes, as well as decreases in diabetes management distress, emotional burden, and behavioral burden. Dr. Aleppo further specified that the clinic seeks to address the fact that a huge proportion of people with diabetes are only treated in the primary care setting and thus have little access or education to the latest in diabetes technology, pointing to PCP-treated patients with type 1 as a target population. Looking to the future, Dr. Aleppo confirmed that a larger study to evaluate the model is in the works (200-300 people in 5+ states in late 2019 or early 2020, as of June 2019).

  • Study participants were able to choose between the Dexcom G6 (n=31) and FreeStyle Libre CGM (n=3) based on recommendations from a short questionnaire. In terms of insulin delivery, seven participants were on a pump and 27 on MDI.

  • Three virtual visits with Cecelia Health CDEs were held: (i) initial training on insertion, how to set alerts, how to download data; (ii) an appointment ~two weeks after initiation to review CGM data and develop any recommendations; and (iii) a visit at ~four weeks to discuss tips and tricks to optimize CGM use. At ADA, we heard that the clinic made 14 check-ins with participants, helping review CGM data and provide actionable insights, along with mental health support; it’s unclear what exactly constituted a “check-in.” (This was the first study to remotely onboard and support people with CGM, so the team erred on the side of too many touchpoints.) In larger study, the number of check-ins will be reduced, to give the program a better chance at scaling.

  • Study participants were recruited from the Wisconsin Research and Education Network and recruited by mail, using a link to the study website. Again at ADA 2019, we heard that about 80% of letter recipients went online and registered for the study without a follow-up phone call, demonstrating a desire for a service like “Geek Squad.”

  • We first caught wind of the Jaeb/HCT/Cecelia virtual diabetes clinic at JPM 2019. There, Helmsley Trustee Mr. David Panzirer noted, “I think about a model like Best Buy’s Geek Squad. Imagine a dedicated team of professionals to help people get on diabetes technology…How could we get people with diabetes straight to the Geek Squad (company agnostic) to help them with technology? We need a new model; the current model is failing.” We’re thrilled to see a pilot of Mr. Panzirer’s aspirations come into fruition and look forward to even more work on a future where patients feel able to tap into the life-changing power of CGM from the comforts of their own home.

  • Assuming “Geek Squad” continues to drive improved adoption and outcomes, we’re curious who might be interested in funding the Squad. With healthcare providers, especially primary care physicians, “Geek Squad” would handle the burden of prescription and initiation of CGM for patients while also improving patient outcomes. For device manufacturers, a device-agnostic group like “Geek Squad” would be able to increase sales of devices without manufacturers having to manage large clinical teams – especially in more remote areas. With cost-savings data, it’s certainly possible that payer could also be interested.

Onduo Delivers Stellar Real-World Satisfaction: 80% of Participants Comfortable with CGM Insertion Purely from Virtual Education

Onduo’s Head of Clinical Affairs Dr. Ronald Dixon provided a glimpse into the company’s ongoing mission to provide virtual care to individuals with type 2, with real-world satisfaction data from n=594 patients. Program participants were clinically evaluated, prescribed, and mailed Dexcom G5 or G6 CGM devices by Onduo’s virtually accessed team of CDEs and licensed endocrinologists. On a five-point Likert scale (1 = strongly disagree; 5 = strongly agree), average overall satisfaction came in at a heartening 4.5. More than 95% of survey participants reported that CGM taught them the impact of daily habits and eating and that CGM made it easier to perform self-care behaviors. Very promisingly, 80% of participants also reported that they were comfortable with CGM insertion, and Dr. Dixon anticipates this number will rise as Onduo continues to refine its practice.

  • Onduo’s survey boasted a relatively diverse study population, compared to those who have traditionally accessed CGM. Of the participants, 60.8% voted that they were “previously unfamiliar with CGM,” reflecting that remotely onboarding patients with no knowledge of CGM is possible. In addition, 72.4% were from urban settings, and 27.6% were from rural settings.

  • On top of the compelling survey results, study participants also saw significant improvements in A1c – particularly for those with a high A1c at baseline. Participants with a baseline A1c >9.0% (n=57) demonstrated a -2.6% reduction (p<0.001) over a mean follow-up of 10.2 months, and the overall decrease was -0.6% ± 1.5 (p<0.001). These results further corroborate A1c benefit documented in a December 2019 paper, published in the Journal of Diabetes Science & Technology.

    • In terms of segmentation, Dr. Dixon specified that Onduo hopes to help individuals with an A1c >8.0% decrease their blood glucose levels using medication and behavioral optimization. For those with an A1c already <8.0%, the company will “rinse and repeat” with CGM and HCP recommendations to prevent any future rises.

  • In the company’s simultaneously-published press release, Dr. Richard Bergenstal, a collaborator on the study commented: “These data serve as a very robust demonstration that it is feasible to ship CGM to people with type 2 diabetes, train them on how to use the system virtually, and derive significant value from its use. A majority of these people were not previously familiar with CGM, suggesting that approaches like that used by Onduo can broaden access to this transformative technology.”

Virtual Clinic Special Session: Viewpoints Vary on How Regulators and Clinicians Should Evaluate Digital Offerings

During a gathering on virtual clinics before the official start of ATTD, it was clear that evaluating safety and efficacy remains arguably the most important unanswered question for digital clinics. Opinions varied widely, and the only regulatory representative at the meeting, FDA’s Dr. Courtney Lias, was asked to chime in multiple times.

  • After Dr. Lori Laffel (Joslin Diabetes Center) gave an overview of the current digital clinic offerings in the US, Dr. Jay Skyler (University of Miami) challenged the idea that insulin dose calculators should require regulatory approval. Dr. Skyler made the argument that he’s been teaching his patients to calculate their insulin doses for decades, but ultimately the calculation and dosing are done in the real world (i.e., not overseen by anyone). Given the difficulty of insulin dose calculations, Dr. Skyler argued there was no significant difference in risk between an app-calculated dose vs. a patient-calculated dose, especially if the patient looked at the dose recommendation before dosing. Dr. Laffel disagreed, arguing that “the majority” of patients probably would not perform the math to evaluate dose recommendations from a calculator. Insulin remains one of the most difficult drugs to administer, particularly given that the vast majority of injections take place outside of clinical supervision. We’ve long seen the need (and potential) for more decision support around basal titration and bolus calculators to help people optimize their insulin delivery and glycemic control. Thus far, these apps (e.g., Voluntis, Welldoc, MyDoseCoach) have seen limited uptake, though data from CGM and connected BGMs alongside smart pens could help spark greater adoption.

    • The FDA has been relatively conservative on dose calculator apps, particularly when using CGM glucose values (see DTM 2019). Of course, administration of an incorrect insulin dose based on a calculator recommendation can be catastrophic; however, incorrect insulin doses are likely even more common without the help of a calculator (Dr. Lori Laffel: “Common errors happen to common people many days of the week.”). Dr. Courtney Lias took a measured stance on the issue, saying, “[The FDA] takes a risk-based approach to these things. Some people have a limited understanding of their own dosing. Some will go with recommended dose, and others will evaluate it [before administering a dose]. It’s good to talk to [the FDA], with respect to insulin dose calculators, about what are the risks and how often are they giving answers that might put patients at risk, but the bottom line is that we do understand the value [of these devices].”

  • During one memorable exchange, an impassioned Dr. Julio Rosenstock (University of Texas Southwestern) expressed his view that the majority of digital app studies are “a total joke.” With diabetes apps increasingly present on the App Store and Google Play Store, many have called for standardized ways to evaluate these apps. The vast majority of these apps are not regulated by the FDA and Dr. Rosenstock called for ATTD, ADA, and/or EASD to develop a way to validate and give “stamps of approval” to apps that clinicians and patients could trust. (ADA and EASD put out a position statement on diabetes apps in 2019, though it fell short of developing a “stamp of approval.”) However, to do this, the organizations would need to develop criteria for the apps and the studies run by these apps. Throughout the day, many members of the group criticized the typical 12-week study period used by many digital health apps as being much too short.

  • Dr. Lutz Heinemann (Science Consulting) presented a German perspective on digital app regulations, focusing on the balance between reducing barriers to entry for smaller, innovative players to reach the market, while maintaining safety and quality. In November 2019, Germany’s parliament passed a Digital Supply Law (DVG) which allows provider-prescribed apps to be reimbursed through the country’s health insurance. These apps have to be approved by the German Institute for Drugs and Medical Devices, which involves filing a 40-page form and a sufficient evidence base to receive reimbursement. Given that small companies cannot afford to run clinical trials, Dr. Heinemann wondered aloud whether this would result in only larger companies’ offerings receiving reimbursement. Dr. Heinemann finished with the open-ended question: “What are the other ways forward?”

  • The tension between the rapid innovation cycle of software with the need for regulatory assurances for safety was on display throughout the day. To this end, we were surprised we didn’t hear more people talking about the FDA’s “Pre-Certification” program. The Precertification program (“PreCert”) was developed in 2017 as a new organization-based approach to regulating software as a medical device, sort of like “TSA Pre-Check” – once pre-certified, companies will get a streamlined digital health software review, with a regulatory decision in 30 days. In our view, the FDA has been quite forward-looking in its approach to software regulation. At CES 2020, CDRH Director Jeff Shuren emphasized his belief that rapid device innovation does not come at the expense of patient safety: “If [the FDA] can advance innovation to have safer and more effective technologies, technologies that address a patient’s unmet needs and mitigate harm from their disease, that’s about safety.”

    • Tidepool CEO, Howard Look did an excellent job summarizing many of the viewpoints expressed: “We all know that the reason Silicon Valley does what it does is because of iteration. We’ve heard a lot of talk about the DIY community and how it has set a standard for iterating and going quickly. The reason they’re able to go quickly is because they’re iterating quickly. When you can do 1,000 things in a month, you’re probably going to find the right answer – you’re going to find the good idea. That’s why Silicon Valley is really good at finding the next great social media app or the next great shopping app. The question we should be asking ourselves is what can we do to bring that same level of iteration, while complying with regulatory requirements and showing safety and efficacy. Building on that, how do we run trials in a way that allow us to iterate along with the speed of software. We can do that now. Every one of us has one of these [smartphones] … I want us to acknowledge that we want to get the best possible solution out there and in order to do that, we’ve got to allow patients to own their own data, we’ve got to open up data protocols … and we need to allow software iteration.”

  • We were pleased to hear some discussion of quality of life and psychosocial outcomes discussed as part digital app studies. Well-respected psychologist Dr. Katharine Barnard-Kelly pointed out many validated psychosocial measures for quality of life, functional health status, psychosocial functioning, and more and strongly advised all studies to look at these measures in addition to safety and efficacy.

Virtual Clinic Special Session: Dr. Thomas Danne on European Landscape: Calls for Easy Data Exchange Between Patients + HCPs and for Focus on Iterative Process in R&D of Digital Apps

Dr. Thomas Danne provided an overview of the current European landscape, underscoring that the biggest issue the region faces is the need for “easy data exchange” between patients and HCPs. Regarding data exchange, Dr. Danne particularly pointed to the need for the device industry to not place restrictions on viewing sharing these data on the platforms that have been developed. On this point, Dr. Danne directed attention to the recent shutdown of the LibreLink app toward Glooko/Diasend as an example of a situation where this principle has been violated to the detriment of patients and providers. Dr. Danne and others have in response submitted a legal complaint to Dutch Healthcare Authorities regarding the shutdown. In the legal complaint, Dr. Danne and colleagues state that on the principle of data ownership, “the raw device data should be available in any communication between the patient and HCP. Now [because of the shutdown], we experience data hostage by the industry while it is not their ownership.” Indeed, data ownership and accessibility is a central point of debate regarding how these tools will be implemented going forward. Will a more egalitarian/democratic system, with full data availability between patients and providers (while ensuring patient data privacy) emerge, or will a more monopolistic system with greater industry control on this relationship win out?

  • Dr. Danne also called for the research, development, and testing of digital apps to be an iterative process that continuously incorporates user perspectives. Dr. Danne hopes that the process might resemble how iOS updates currently work for iPhones, where the user is notified of updates periodically when ready, and then can either opt in or out of the update as it happens. The iterative approach here is key to Dr. Danne, especially given the need to involve social media since “this is how patients can now quickly communicate with providers like us.”

  • Unsurprisingly, Dr. Danne keyed in on renumeration (reimbursement) as a major hurdle for further adoption. Dr. Danne joked that “at the end of the day, money makes the world go around” – therefore, it’s essential that sensible reimbursement be developed for this field. He noted that while many successes have been achieved on local levels across Europe, a lack of access continues to be a problem because of little clarity on reimbursement. For such programs to reach national or even Europe-wide levels of standard of care, reimbursement will be absolutely key. 

Select Questions and Answers

Comment from Dr. Satish Garg: I think most of us are discussing this very important topic, and we have to ask ourselves the question of whether we are discussing this as a way to make peoples’ outcomes better. Or are we trying to protect our revenue streams? On average, all the big centers in the US are probably lucky to spend one hour with each patient. The rest of the time, patients with diabetes are living on their own. This disease needs to go into the hands of the people. If we keep talking about renumeration to the community at large – I realize that I’m speaking against myself here. But unless we realize where this needs to go, we will never get there. Unless people see the value in this, then it won’t go anywhere.

Q: Dr. David Klonoff: A big barrier is that we need data from RCTs. We want to see greater use and greater reimbursement. If we’re going to ask physicians to invest their time and for payers to invest their money, we have to show the data. Right now, payers want to see RCTs.

Dr. Danne: To counter what you just said, I’d like to point out that the UK has done wonderful RCTs that couldn’t show the benefit of CGM. But real world evidence data shows the complete opposite. Why is that? I don’t know. Yes, I’m a fan of RCTs, but real world evidence often shows the opposite.

Virtual Clinic Special Session: Drs. Richard Bergenstal and David Klonoff on Cybersecurity and DTSec System: Could Diabetes Pave the Way for other Therapeutic Areas?

Drs. Richard Bergenstal and David Klonoff touched on cybersecurity concerns, focusing on core principles of what makes a device secure and what current efforts around cybersecurity look like in the diabetes space. Dr. Bergenstal asserted that for sound cybersecurity, three main standards must be met: (i) confidentiality, by protecting devices from unauthorized disclosures; (ii) integrity, by protecting these products from authorized modifications; and (iii) availability of data, by protecting products from loss of function. Generalizing from these core points, Dr. Bergenstal noted that “you want to be able to protect patient disclosures and confidentiality, want to be sure no unauthorized modifications of intended settings/delivery occurs, and want patients to have data be available to them at all times.” Dr. David Klonoff then provided more granularity on how these concepts are currently being applied in the consideration of devices in the diabetes space, describing the DTSec framework, which was developed as a pathway for companies to demonstrate sound cybersecurity with their products. The DTSec framework contains three parts to evaluate candidates for cybersecurity, as enumerated by Dr. Klonoff: (i) a protection profile, with a certain set of security attributes in a standard product; (ii) specific security features for the product that manufacturers can implement or make claims about; and (iii) conformity assessments done by third parties against the standard to determine whether a product meets the claims put forth by the manufacturer. Interestingly, Dr. Klonoff noted that diabetes could pave the way in this field, seeing as the only standard for medical devices to be tested in this way for cybersecurity purposes is actually DTSec – there is no other standard for cardiology devices, GI devices, or any other type of medical device. He predicted that DTSec can provide a model for other disease areas in the future, especially seeing as the cybersecurity features necessary in medical devices should be translatable from the diabetes field.

  • Notably, at this time, only two companies have used DTSec in their process of applying for regulatory clearance – Ascencia for its BGM, and Insulet for its Omnipod Dash. Dr. Klonoff seemed to indicate that coming years will see more companies in the pump/CGM space sign on to using DTSec, although it’s unclear how prominent this movement may be.

Virtual Clinic Special Session: Dr. V. Mohan on Digitalizing Diabetes Care in India + Other Geographies: Presents 17 Specific Challenges and Points to Rapid Adoption of NCD IT System in India as Positive Example

Dr. V. Mohan represented “other geographical countries,” which he noted as having 80% of the total number of people with diabetes and with dramatically different needs when compared to more affluent regions in the US and Europe. Dr. Mohan pointed to a number of unique challenges in these regions when it comes to diabetes management more broadly, including delays in detection, therapeutic inertia, lack of control of multiple risk factors, low long term medication adherence/engagement, and a lack of affordability (even for common medications and insulin). Dr. Mohan then presented a litany of specific challenges associated with digital diabetes care in these regions (see table below):

  1. Electronic records are not widely available

  1. Even when available, these systems do not talk to each other. Hence, it is very difficult to obtain data even within a country, let alone across different countries

  1. Integration of data such as CGM into the EMR is challenging

  1. Patients often fell that the doctor is spending more time with the computer than with them

  1. Guidelines for treatment recommendations are not incorporated as an algorithm into the EMR

  1. Periodic alert messages and reminders about treatment schedules are challenging to produce

  1. Education and knowledge related to diabetes treatment should be available in an easy to understand format, but often is not

  1. There is a need to provide medical care after clinic hours, as diabetes is a 24/7 condition

  1. There is a constant threat of data privacy and also whether the data would be commercialized

  1. Telemedicine: helps to reach out to remote and rural areas. However, challenges are availability of internet and broadband connectivity. Again, the patients miss the “touch and feel” of doctors here

  1. Apps: There are hundreds of apps in the market already, but several studies have shown that they do not always meet the expectations and the needs of the patient. Industry should ask patients for their needs and preferences in developing apps

  1. Decision support systems are useful but need to be tested, simplified, and be affordable

  1. Modern technology: Like pumps and CGM, are useful but expensive and beyond the reach of the common man.

  1. Lack of insurance or other support for patients to afford digital technology

  1. There could be medical-legal issues as policies and regulations regarding digital health are vague or even non-existent

  1. Digital tools should be shown to not only help in prevention of diabetes but also in tackling diabetes related complications

  1. The pricing of digital services needs to be sorted out

  • Dr. Mohan stressed three points as the “way forward” for digitalizing diabetes care in these regions. He called for (i) a seamless collaboration with HCPs, engineers, startups and software technologists; (ii) integration should be made with all devices and brands of different manufacturers; and (iii) the patient’s needs must be kept first and the patient must be at the center of everything.

  • Very intriguingly, Dr. Mohan explained the rapid adoption of the NCD IT System in India – 42 million people have been enrolled over 20 states, 17 million have been screened, 42,0o0 have been trained, and there are seven different apps being used. The program, announced by the Indian government in 2018, is a flagship national health initiative meant to build a technology solution that can help “auxiliary nurse midwives” in India that are responsible front line care in rural areas to screen and manage NCDs. The scale of this program is truly breathtaking, and Dr. Mohan explained that it was helped by massive investment from Dell, which ran the entire program as part of its corporate social responsibility wing. We’ll be interested to track the progress of the system over time and see how outcomes of patients touched by the system may be improved or affected. For more details on this ambitious program, see here (“How Dell is helping India on its mission to bring preventive healthcare to 800 million people in rural India”).

A Rural Chilean Study of Pediatric Type 1s Highlights the Potential of Telemedicine: Mean A1c Drops from 8.5% to 7.9% in Two Years, 93% of Telemedicine Contacts Performed via Smartphone

Ms. Julie Pelicand (Universidad del Valparaiso) presented results from her study following 21 families of pediatric patients with type 1 diabetes over two years. These patients lived in rural, remote areas of the Aconcagua Valley and used telemedicine for their diabetes care and follow-ups beginning in 2016. This study aimed to explore opportunities for advancement of telemedicine in rural communities and to assess satisfaction, utility, and challenges of this form of diabetes care. In order to do so, the number and content of telemedicine contacts between families and physicians was tracked, and satisfaction was assessed using surveys and semi-structured qualitative interviews. Perhaps the most impressive result from the study, mean A1c in the particularly challenging pediatric and teenage group (mean age 11 years) fell from 8.5% in 2016 to 7.9% in 2018. It’s incredibly powerful to see such positive data in populations and regions that are often forgotton and we give credit to ATTD for highlighting Ms. Pelicand’s excellent work.


  • The vast majority of telemedicine contacts were made on a smartphone. Of the 716 total telemedicine contacts made over the duration of the study, 93% were made via smartphone, with an average of 12 smartphone contacts per family per year. The remaining 7% were made via email, with an average of 5 (±4.35) email contacts per family per year. In Chile, while 71% of people have computer internet connections, 94% own smartphones. For these reasons, Ms. Pelicand emphasized smartphone-compatibility as a critical component of telemedicine platforms and an important tool to increase access. It’s certainly a testament to the transformative power that smartphones have had and their immense potential to reach traditionally hard-to-reach groups via telemedicine.

  • Use of telemedicine for administrative purposes, like scheduling hospital visits or accessing supplies was the reason behind 35% of telemedicine contacts. Sharing glycemic data in order to adjust insulin dosing made up nearly a quarter (23%) of contacts, while support managing emergencies like hypoglycemia and diabetic ketoacidosis made up another ~21%. Psychosocial support for patients and their families regarding social, economic, and psychological situations was the reason behind 11% of contacts, health education on topics like nutrition and physical activity made up 5% of contacts, and finally, technical support and treatment adaption of insulin therapy composed 5% of telemedicine consultations.

  • In addition to improving diabetes management, familial health literacy, and health access, telemedicine use was found to strengthen relationships between patients and their healthcare providers. Insights from patients and their families suggest that telemedicine use may increase confidence in healthcare providers and improve perceptions of their empathy and availability.

  • Ms. Pelicand shared the key improvements needed in this emerging field, first highlighting the importance of expanding telehealth communication to include all members of a healthcare team. The researcher also recommended the inclusion of institutions like clinics, hospitals, and even schools, on telemedicine online platforms to facilitate better administrative management and a unified set of health guidelines. She also proposed integrating telemedicine with social platforms to allow peer-to-peer support and sharing of experiences. Finally, she emphasized the importance of improving the sharing of data between patients and healthcare professionals via patient portal.

Brian Levine Highlights Digital Health as a Potential Solution to the “Last Mile” Problem of Healthcare Access

Onduo’s and Close Concerns alum Brian Levine defined and taxonomized the current connected diabetes care space, before giving us a glimpse into its future. Mr. Levine began by highlighting the need for greater healthcare access in the US and the critical link between access and outcomes. Connected care, he suggests, may fill the gap between resources and those who need them. A connected care platform was defined as having remote coaching, often delivered through a smartphone app, that may receive data from connected care devices. An excellent paper reviewing the diabetes digital health landscape authored by Mr. Levine, Kelly Close, and Dr. Bob Gabbay was published in DT&T  in October.

  • Connected care platforms were broken into five main categories: (i) quantified self, (ii) portals for brick-and-mortar HCPs, (iii) AI coach, (iv) non-physician HCP-driven platforms, and (v) virtual clinics.

    • Quantified-self platforms provide connected devices like activity trackers or connected blood glucose meters to members and may use virtual incentives or gamification. These are often used by employers, payers, and employee wellness benefit managers in order to encourage improvements to health behaviors and outcomes. 

    • Portals for brick-and-mortar healthcare providers are software programs that can be licensed by healthcare clinics or providers to more effectively manage patients. Features often include risk-stratification of patient panels to better allocate care and two-way secure communication between members of the care team as well as between the care team and patient.

    • Artificial intelligence (AI) coaching platforms are completely based on user-generated data uploaded either passively using connected devices or through manual data entry. These platforms may include automated coaches that advise using pattern recognition.

    • Non-physician healthcare provider driven (NPHD) platforms provide remote human coaching in addition to AI coaching.

    • Virtual clinic platforms, have features of NPHD platforms but also include physicians capable of remotely prescribing and managing medications and connected devices. While these platforms have the highest level of clinical rigor, they are also the most expensive. For this reason, three virtual clinic platforms are often partnered with telemedicine companies.

  • Mr. Levine proposes that these categories exist on a spectrum of clinical rigor, and that the ultimate goal of connected care is to use platforms together in a risk stratification pyramid. This would optimize care by allowing patients to escalate or deescalate the spectrum depending on their individual needs. Levine predicts that connected care will increase access to healthcare and act as a democratizing agent of CGM use. In doing so, connected diabetes care platforms have the potential to decrease comorbidities as well as burden on both patients and payers. Mr. Levine emphasized the significant need for more rigorous investigation of connected diabetes care platforms, with the hope of increasing reimbursement and thus access. 

Artificial Pancreas

Title

Details + Takeaways

Real-life use and performance of the MiniMed 670G System in Europe

  • CareLink data from 3,139 individuals with ≥10 days of CGM data pre- and post-Auto Mode enabling periods from Oct. 1, 2018 to Jan. 18, 2020

  • Time in Range increased from 61.5% to 71.1% after enabling Auto Mode; GMI decrease from 7.2% to 7%; Mean glucose from 164 mg/dl to 153 mg/dl

  • Time <70 mg/dl decreased from 2.6% to 2.3%; time <54 mg/dl decreased from 0.6% to 0.5%

  • Time >180 mg/dl decreased from 36% to 27%; time >250 mg/dl decreased from 10% to 6%

  • 7,847 patients across Europe with Auto Mode enabled had mean Time in Range of 71.1%; Spain (n=625) had the highest of all countries (74%), while UK (n=766) came in last (69%)

Diabeloop closed loop system allows patients with diabetes type 1 to practice physical activity without increasing hypoglycemic risk

  • 63 adult type 1s, pump-using, mean age 47, A1c 7.6%

  • Used Diabeloop algorithm on handset with Dexcom G5 and Cellnovo/Kaleido pump

  • Randomized, cross-over study with 3-month phases and eight-week wash-out period in between

  • Median of 10 physical activity sessions per patient over 3 months

  • Time in Range was similar on days with and without physical activity: 68% vs. 69%

  • Percent time <70 mg/dl was consistently ~2% following light, medium, or intense physical activity of multiple durations (<30 min to >90 min), regardless of announcement

Use of the ultra-rapid insulin Fiasp in the iLet bionic pancreas

  • 3-way, randomized cross-over home use study with iLet using Fiasp, iLet using usual insulin, and usual care for seven days each

  • 36 subjects, mean age 41, A1c 7.5%, using Senseonics Eversense or Dexcom G6

  • No significant differences between usual insulin and Fiasp on any glycemic metrics

  • Note: PK settings were the same for both Fiasp and usual insulin

Incorporating Physical Activity and Stress Estimates to Improve Glucose Predictions for Multivariable Artificial Pancreas Systems

  • 20 subjects wearing Empatica E4 (wrist-worn device) tracking galvanic skin response, accelerometer, temperature, blood pressure pulse, and heart rate for 80 hours

  • Multiple machine learning methods, k-nearest neighbor, ensemble learning, deep learning, support vector used to identify psychological stress and physical activity states

  • Support vector machine able to detect physical activity state (sedentary, treadmill, and stationary bike) with 97% accuracy

  • Models are able to detect acute psychological stress states (mental stress, exciting-anxiety, and non-stress) with 90%-97% accuracy

  • Using psychological stress and physical activity state detection in glucose prediction model results in 18% increases in accuracy

Algorithm customization trend with Diabeloop DBLG1 system: towards less parameters after a few weeks

  • Analysis of number of parameter changes in 63 patients in DBLG1 clinical trial and 33 patients from soft launch period for 12 weeks

  • Parameters analyzed: meal ratios, TDD, glucose target

  • In clinical trial patients, mean of 7 parameter changes in first week, down to <3 changes/week after three weeks, and 2 changes/week in week 12

  • In real-world, soft launch users, 1.6 changes during first week, 0.2 changes in the last week

Clinical Decision Support Systems/Advisors

Title

Details + Takeaways

Real-world use of IQCast hypoglycemia prediction feature in the Guardian Connect System and its impact on clinical outcomes

  • CareLink data from 259 people with diabetes (200 type 1s) using Sugar.IQ before and after using IQCast

  • IQCast generates probability of low glucose episode within 4-hour window

  • Overall reduction in hypoglycemic episodes from 18/month before IQCast to 12/month 4 months after

  • Reduction in nighttime hypo episodes from 5/month to ~3/month

  • Frequency of hypo episodes was consistent one month through four months after IQCast use

  • Overall time <70 mg/dl was slightly decreased from 3.8% before IQCast to 3.3% after four months (-7 minutes/day)

Utility of a glucose meter with Bluetooth and web connectivity in aiding physicians and patients achieve better glycemic levels

  • 13 type 1s and 47 type 2s used One Touch Reveal, a web application for recording and analyzing blood glucose data from OneTouch Verio Flex BGM, for 12 weeks

  • 78% said being able to see color-coded BGM trends and data points was motivation to use theig BGM

  • 23% of patients said the One Touch Reveal was helpful in titrating insulin doses

  • 67% said they were more adherent because of the BGM data being available to their HCPs in real-time

A digital workflow and decision support system to prevent diabetes-related acute hospital admissions and inpatient stays

  • Nine patients treated with GlucoTab@MobileCare for three months (mean age 77)

  • GlucoTab provided algorithm-based decision support for insulin adjustments

  • No acute diabetes-related hospitalizations using GlucoTab@MobileCare; during routine care period prior to GlucoTab, six patients were hospitalized

  • A larger study is planned

A machine learning approach for detecting insulin pump faults

  • Several anomaly detection algorithms (e.g., isolation forest, k-nearest neighbor, principal component analysis, etc.) used to detect insulin pump faults in silico (Padova/UVA T1D simulator)

  • 30 days of simulated closed-loop therapy, with two pump faults per patient

  • Top three algorithms: histogram based outlier score, isolation forest, and k-nearest neighbor with recall of 0.87, 0.85, and 0.85, respectively

  • Worst three algorithms: minimum covariance determinant, angle-based outlier detection, and over-sampling principle component analysis with recall of 0.71, 0.72, and 0.78, respectively

Human Factor in the Use of Diabetes Technology

Title

Details + Takeaways

Lived experiences of people with diabetes using do-it-yourself artificial pancreas systems – analysis of responses to open-ended items in an international survey

  • Qualitative analysis of 2 open-ended questions from an online survey of 886 adult users of Do-it-Yourself Artificial Pancreas Systems (DIYAPS)

  • Dramatic improvements to clinical outcomes and quality of life were reported regarding DIYAPS usage

  • Diabetes online communities were the primary source of DIYAPS knowledge as well as practical and emotional support

  • Acquiring and assembling the necessary devices proved challenging but empowering to users

Better skills in carbohydrate estimation is associated with higher Time in Range

 

  • 168 people with diabetes on intensive insulin wore FreeStyle Libre for 14 days, while carbohydrate estimation was assessed using the SMART-tool

  • SMART-tool presents pictures of foods and meals for which the right amount of carbs is chosen out of four possible answers (multiple choice format)

  • Better carbohydrate estimation skill was significantly associated with more time in range (r=.32, p<.01), less time in hyperglycemia (r=-.32, p<.01), less variability (r=-.21, p<.05), and lower maximum glucose values (r=-.23, p<.01)

  • Mean raw errors correlated significantly with glucose variability (r=-.29, p<.01), indicating that underestimation of carbohydrate content may be more significant to glucose variability than overestimation

Physicians’ perceptions and attitudes toward digitalization and new technologies in diabetes care

  • 324 physicians in Germany completed an online survey

  • The majority (76%) reported a positive attitude toward digitalization and its potential for improving diabetes therapy

  • Top advantages of digitalization were better communication with patients (72%) and more support for treatment decisions (67%), while the top barrier was unclear payment of digital services (80%)

  • Glucose management software is used by an estimated 31% of patients, followed by diabetes apps (23%), digital education (7%) and telemedical consulting (2%)

  • Physicians predict these technologies to increase in use by 18-25% in the next 5 years

Satisfaction and reduction in diabetes burden with predictive low glucose suspend (PLGS) in individuals with type 1 diabetes (T1D)

  • 541 participants (55% Tandem pump users, 29% other pump users, 15% MDI users) completed Diabetes Impact and Device Satisfaction (DIDS) survey prior to using Basal-IQ and at 2, 4, and 6 months after

  • Basal-IQ increased device satisfaction in former MDI (+2.32 points, p<.001) and non-Tandem pump users (+1.13, p<.001) on a 10-point Likert scale

  • Diabetes impact was significantly reduced in Tandem pump users (-.89 points, p<.001), previous MDI users (-1.6, p<.001), and non-Tandem pump users (-.91, p<.001) on a 10-point Likert scale

Telemedically assisted lifestyle intervention to improve glycemic control and self-management in type 2 diabetics

  • 2-arm, randomized study of 113 type 2 participants, with the intervention group using wearable trackers and receiving individualized telephone coaching and support

  • Intervention showed significant A1C reductions, particularly in the first 6 months (mean A1C decreased from 6.9% to 6.4%, p=.000)

  • At 12 months, the intervention group had a significantly lower mean A1C, compared to the control (6.5% vs. 7%, p=.000)

Observation of glycaemic management in professional cyclists with type 1 diabetes over a 7-day world tour stage race

  • Observational study of 6 professional cyclists from Team Novo Nordisk on MDI during the Tour of California races (1,244 km, 20,840 meters in elevation, 7 days, 3-7 hours/day)

  • Participants wore Dexcom G6 and recorded insulin dosage with NovoPen Echo Plus (not commercially available)

  • Bolus insulin use was uncommon despite significant carbohydrate intake (mean 76 g/hr)

  • Riders spent 63% Time in Range during the races, with no time in hypoglycemia

  • Riders spent 25% of time in level 1 (180-250 mg/dl) and 11% of time in level 2 (>250 mg/dl) hyperglycemia

HbA1c target setting is associated with metabolic control

  • Questionnaire sent to endocrinologists from 53 international pediatric SWEET centers (Canada, Europe, India, Australia)

  • “What is the target goal for HbA1c in your SWEET center?”: 13% reported an HbA1c target between 6%-6.5%, 32% had a target between >6.5%-7%, 19% between 7%-7.5% and 4% between 8%-8.5% 

  • Positive association between A1c outcomes and target value (p=.005) (adjusting for age, sex, and duration of diabetes)

Informatics in the Service of Medicine; Telemedicine, Software and Other Technologies

Title

Details + Takeaways

Social media influencers give bad diet advice for diabetic patients

  • Content analysis of 20 popular Russian Instagram influencers (>100,000 followers) that used hashtags related to diabetes or diabetes nutrition

  • 70% of the blogs could not be considered credible sources of diabetes management information

  • Unreliable and potentially dangerous information included statements that fruit and cinnamon lower blood sugar, and that garlic makes the pancreas secrete double-acting insulin

Analysis of the budget impact of the utilization of glucose meters with color-range indicator in five European healthcare systems

  • Economic model using data from a randomized, controlled trial evaluating color-range indicator (CRI)-based glucose meters; CRI-based BGMs simply tell the user whether a measurement is low, normal, or high

  • 10-year risk of fatal myocardial infarction was estimated using the UK Prospective Diabetes Study risk engine and used to model monetary impact

  • The mean A1C reduction (0.4%) associated with CRI meter usage could result in a 2.4% absolute risk reduction in MI risk

  • This risk reduction could result in a significant yearly potential cost savings in France (€547,472), Germany (€9,025,388), Italy (€6,006,518), Spain (€841,799), and the UK (€421,069)

Hypo- and hyperglycemia prediction from pooled continuous glucose monitor data

  • Predictive supervised learning model using 10 million hours of CGM data as well as health data collected in the One Drop app from over 3,000 users in 195 countries (88% T1, 9% T2, 3% unreported)

  • Hypoglycemia predictions (<70 mg/dl):

    • In the next 30 minutes had 93.2% recall, 89.4% precision, and area under the receiver operating characteristic curve (AUC) of 99.8%

    • In the next 1 hour had 83.2% recall, 74.1% precision, AUC of 98.6%

    • In the next 4 hours had AUC of 91.9%

  • Hyperglycemia predictions (>180 mg/dl):

    • In the next 30 minutes had 98.9% recall, 97.6% precision, and AUC of 99.9%

    • In the next one hour had 95% recall and 92.6% precision, AUC of 98.8

    • In the next 4 hours had AUC of 91.6%

  • The mean absolute relative difference (MARD) was 4.3% for 30-minute predictions (with 98.7% in Zone A and 99.9% in Zone A or B of the Clarke Error Grid), and 13.4% for 1-hour predictions (79.4% in Zone A, 98.4% Zone A or B)

Transforming a Furby toy into a multi-modal companion for children with type 1 diabetes

  • A Furby toy’s internal electronics were replaced with a system that could be remotely controlled by a Raspberry Pi 3 (small, single-board computer) via Bluetooth.

  • The Raspberry Pi connected to Nightscout, Amazon’s Alexa voice service, and a smartphone chatbot

  • Furby expressed different emotions based on blood glucose values from Nightscout and could announce changes to signal children to take action

  • A smartphone chatbot allowed for text-based games, blood glucose monitoring, and remote control of the toy-based system, while Alexa and the Furby’s internal microphone allowed for voice-based interactions

Insulin Pumps

Title

Details + Takeaways

Testing of a novel extended wear infusion set (EWIS) with and without the addition of heparin

  • 20 subjects wore modified MiniMed Quick-set infusion sets, 2 with heparin and 2 without heparin (placebo) for 7 days per set or until failure

  • Randomized, blinded, cross-over study with four wear periods

  • No significant difference between the placebo and heparin infusion sets in the overall wear duration (6.2 days; 6.3 days) or failures due to hyperglycemia

  • Both the placebo and heparin infusion sets had longer wear duration compared to the Quick-Set infusion set (p=0.005), due to material found in the modified infusion sets

Frequent bolusing is associated with better glycemic outcomes in 7,906 youth with Type 1 diabetes using the OmniPod Insulin Management System

  • Glycemic outcomes were assessed based on average bolus frequency over 6 months in an observational study of youth with T1D using Omnipod (n=7,906; 7,495 BGM users, 1,481 CGM users, 1,070 both)

  • Lower mean glucose was associated with more frequent bolusing, with an overall decrease of 55 mg/dl between groups with an average bolus of <4 times/day (230 mg/dl) compared to >10 times a day (175 mg/dl)

  • Adolescents (13-17 years) had the lowest bolus frequency, with 33% recording <4 boluses/day; preschool aged children (<6 years) had the highest bolus frequency, with ≥36% bolusing ≥8 times/day

  • The highest bolus frequency group (>8 boluses/day) had a mean GMI of 7.4% compared to 8.2% in the lowest bolus frequency group (<4 times/day)

  • The highest bolus frequency group (>8 boluses/day) had a mean Time in Range of 59% compared to 45% in the lowest bolus frequency group (<4 times/day)

The introduction of MiniMed 670G hybrid closed loop insulin pump auto mode system improves glycemic control in routine clinical practice

  • Real-world analysis of glycemic control in 12 type 1 diabetes patients using MiniMed 670G Auto Mode for 12 weeks

  • Mean age 42 years, mean A1c of 7.8% and Time in Range of 69%

  • 12 weeks after starting Auto Mode, A1c level decreased to 7.1%

  • Time in Range increased from 69% to 72%; time below range decreased from 4.1% to 1.5%

  • Average time in Auto Mode after 12 weeks was 89%

Insulin delivery with patch pumps: basal rate accuracy

  • Basal rate accuracy of the Accu-Chek Solo micropump was examined through basal rates of 0.1 U/h and 1 U/h tested in 9 repetitions over 72 hours

  • Total deviation from expected target weight was -5.3% at a basal rate of 0.1 U/h over 72 hours

  • For 1-hour windows, 98% (1 U/h) and 51% (0.1 U/h) were within +15% of the target

  • Accu-Chek Solo micropump delivered insulin more accurately at a larger basal rate (1 U/h vs. 0.1 U/h)

Occlusion detection time in patch pumps

  • Occlusion detection time in Accu-Chek Solo micropump (ACS) and the A6 TouchCare pump (A6) was examined and compared to prior data from Insulet OmniPod

  • Mean occlusion detection time was 2:57 [hh:mm] for ACS, 7:26 for A6, and 3:38 for OmniPod at a 1 U/h basal rate

  • At a basal rate of 0.1 U/h, mean occlusion detection time was 35:11 for ACS and 34:30 for OmniPod; no occlusion alarm signaled for A6

  • At a higher basal rate (1 U/h), occlusion detected was fastest for ACS compared to the A6 and OmniPod

Regular users of temporary basal rate or extended bolus have better glycemic outcomes in 12,823 OmniPod Insulin Management System users with Type 1 diabetes

  • A retrospective analysis of glycemic outcomes by frequency of temporary basal rates (TBR) and extended boluses (EB) in 11,978 BGM users and 2,640 CGM users on Omnipod

  • Mean glucose was significantly lower for regular users of TBR or ER compared to infrequent users across all age groups (p<0.0001)

  • Infrequent (use less than once per month on average) TBR/EB users had a mean glucose ranging from 179-221 mg/dL (9.9-12.3 mmol/L) across all age groups

  • Regular (use at least once per month) TBR/EB users had a mean BG ranging from 163-193 mg/dL (9.1-10.7 mmol/L) across all age groups

Accuracy of insulin delivery in durable pumps

  • Evaluated accuracy of Medtronic MiniMed 670G, Tandem t:slim X2 (TS), and Ypsomed mylife YpsoPump (YP) bolus and basal rate delivery

  • At basal rates of 1 U/h, 99% of 670G, 100% of TS, and 99% of YP 1-hour windows were within +15% of the target

  • For boluses of 1 and 10 U, 100% of each model were within +15% of the target. For 0.1 U, 85% of 670G, 96% of TS, and 88% of YP were within +15% of the target

  • Overall, larger bolus or basal rates were associated with higher accuracy

Insulin delivery with patch pumps: bolus accuracy  

  • Insulin delivery accuracy of Roche Accu-Chek Solo micropump (ACS), Roche A6 TouchCare pump (A6), and Insulet Omnipod (OP) were compared in a microgravimetric experiment

  • Bolus sizes of 0.2 U, 1 U, and 10 U were used, and each pump was tested 9 times

  • All three patch pumps had an average deviation of less than 5% for bolus volumes of 0.2 U, 1 U, and 10 U

  • The Accu-Chek Solo micropump had a larger percentage of boluses within +15% of the target for 0.2 U and 1 U (88%, 99%) compared to the A6 (40%, 65%) and OP (57%, 77%).

Glucose Sensor Posters

Title

Details + Takeaways

Canadian real-world analysis of flash glucose monitoring and glycemic control

 

  • Analysis of FreeStyle Libre users’ scanning frequency in Canada, showing that the average user scans their sensor 11/times a day

  • Users were divided into deciles based on scanning frequency to determine the association between scanning frequency and metrics such as time in range and hypoglycemia rates. Consistent with prior results, a higher scanning frequency was associated with a higher time in range and decreased time in hypoglycemia

  • Patients in the lowest scanning frequency decile (3.3 scans/day) spent 13.1 hours in range (55% Time in Range) and 24 minutes with a glucose <54 mg/dl mmol/L (1.6% time <54 mg/dl) . Patients in the highest scanning frequency decile (29.3 scans/day) spent 16 hours in range (67% Time in Range) and 20 minutes with a glucose 54 mg/dl (1.4% time <54 mg/dl).

Time in range : how many patients achieve ATTD recommendations ? Results in a cohort of diabetic patients using FreeStyle Libre.

 

  • 482 patients (82% type 1) initiated on FreeStyle Libre and underwent a training session with a nurse and diabetologist. Significant A1c reduction was observed at three months (7.9% at baseline to 7.6%); however, less than 5% of patients were meeting ATTD consensus goals for CGM metrics

  • As a reminder, targets set at ATTD 2019 include a Time in Range >70% and time <70 mg/dl <4% for most patients with diabetes

Accuracy comparison between Dexcom G5 and Eversense sensors

 

  • Study had people with type 1 (n=11) on Senseonics Eversense simultaneously wear Dexcom G5 for seven days. During day 3, patients were admitted to a clinical research center (CRC) to receive breakfast with delayed and increased insulin bolus to induce glucose excursions. At CRC, venous glucose was monitored every 15 min (or 5 min during hypoglycemia) for 6 hours. At home patients were requested to perform 4 fingerstick glucose measurements per day.

  • G5 was found to be more accurate than Eversense in the clinical research center measurements: G5 showed overall smaller median ARD (absolute relative difference) than Eversense, 7.91% [4.14-14.30]% vs 11.4% [5.04-18.54]% (p-value<0.001)

  • No significant differences were seen in in-home accuracy

Long-term outcomes of Belgian real-time continuous glucose monitoring reimbursement for adults with type 1 diabetes on insulin pump therapy: results after 24 months rescue study

  • Two-year real-world study finding that reimbursement for real-time CGM was associated with improved glycemic control, quality of life, and fewer diabetes related hospitalizations

  • Baseline A1c decreased from 7.7% (7.5–7.8) to 7.4% (7.2–7.5) at 12 months and remained stable for 24 months (p<0.001 for both)

  • One year before reimbursement, 15% of participants were hospitalized for hypoglycemia or ketoacidosis in contrast to 4% in year 1 and 3% in year 2 (p<0.001 for both)

Difference between sensor-derived glucose management indicator (GMI) and laboratory HbA1c in patients with type 2 diabetes

 

  • 112 paired laboratory A1c and GMI values were collected from 50 type 2 patients on insulin and oral antidiabetic drugs

  • Patients had professional CGM (Medtronic iPro 2 with Enlite sensor) for seven days with laboratory A1c determined in the same week

  • The agreement between laboratory HbA1c and sensor-derived GMI was examined using Bland-Altman plot with repeated subject measures

  • 24% of GMI values were within 0.25% of laboratory A1c and 40% of GMI values were within 0.5% from laboratory A1c values. Overall, GMI was lower than laboratory A1c 

Continuous glucose monitoring (CGM) initiation at diagnosis versus six months later: which is best?

 

  • RCT of 55 type 1 youths to examine differences in glycemic and psychosocial outcomes in pts who started CGM soon after diagnosis compared to those starting six months later

  • More time between diagnosis and initiation was associated with higher levels of CGM discontinuation (r=0.37, p=0.005), lower CGM use over the six month period (r=0.41, =0.002), higher levels of diabetes distress among parents (r=0.27, p=0.05), and more time spent <54 mg/dl (r=0.27, p=0.05)

  • Time to CGM start was not correlated with A1c values following six months of use

Accuracy of Freestyle Libre 2 system versus glucose meter among adolescents with type 1 diabetes in real-life conditions of summer camp

 

  • N=58 adolescents w/ type 1 at a summer camp were enrolled for the prospective observational study

  • During four consecutive days, an 8-point glucose profile was collected. Capillary blood glucose (BG) were measured using Contour Plus One glucose meter and followed within 1 minute by FreeStyle Libre 2 scan. Glucose trends arrows were also recorded. Accuracy of the system was assessed by calculation of bias and mean absolute relative difference (MARD), while clinical utility was checked against surveillance error grid (SEG)

  • FreeStyle Libre 2 overestimated BG by a mean of 6.5 mg/dl (5.5-7.6) and overall MARD was 11.3% (10.8-11.8)

  • Error grid classified 97.5% of scans as clinically accurate (class A:85.3%, B:12.2%) and 2.5% as class C. FreeStyle Libre 2 presented significantly worse MARD during rapid glucose decrease [17.9% vs 10.2%, p<0.0001], during mountain hikes (16.6% vs. 11.1%, p=0.0131) and in children who presented adverse skin reactions to the sensor (13.5% vs. 11.1%, p=0.0381)

Effects of exercise intensity, duration, and heart rate on glucose change in adult T1D assessed with continuous glucose monitoring and activity tracking

 

  • Data from five individuals on MiniMed 530G system shows that glucose reduction during exercise can be observed and characterized in real-world scenarios with CGM and fitness tracking

Continuous glucose monitoring (CGM) with Dexcom in pregnancy: clinical experience and recommendations

 

  • Retrospective data from 50 type 1 pregnancies with CGM use (Dexcom G4, G5, or G6)

  • No patient discontinued use during pregnancy, with median wear time of 93%. CGM was well tolerated

Impact of flash glucose monitoring on glycemic control and quality of life in adults with type 1 diabetes: a real world study

 

  • Prospective, real world, case-control study with six-month follow-up in patients starting on a FreeStyle Libre system vs. those previously on it (n=41 total type 1s)

  • Quality life improved with FreeStyle Libre: the EsDQUOL questionnaire improved in the group starting FreeStyle Libre but not in the group already on it (p=0.005)

  • No significant differences in A1c between the two groups, or on severe hypoglycemia and any other measure tested

New Insulin Analogues

Title

Details + Takeaways

Differences between insulin glargine 300 u/ml (GLA-300) and insulin degludec 100 u/ml (IDeg) in high-risk type 2 diabetes (t2dm) populations: subanalysis of the BRIGHT trial

  • Subgroup analysis of the Sanofi-sponsored BRIGHT trial of Toujeo vs. Tresiba in type 2s across different age levels and renal function levels at baseline.

  • A1c drop was significantly better with Toujeo vs. Tresiba in patients ≥70 years

  • A1c drop was significantly better with Toujeo vs. Tresiba for patients with impaired renal function (eGFR ≤60).

  • There was no increased risk of hypoglycemia in either subgroup w/ Toujeo vs. Tresiba

  • As a reminder, the BRIGHT trial initially showed similar A1c change with Toujeo vs. Tresiba in the broader population of type 2 patients, and no significant differences in hypoglycemia during the maintenance phase of the trial (but lower rates with Toujeo in the titration period)

  • Future studies in this specific population (older patients with impaired renal function) may further elaborate important differences in efficacy between these two next-gen basal insulins.

Lower risk for severe hypoglycemia with gla-300 vs. Gla-100 in patients with type 1 diabetes (t1d): a meta-analysis of 6-month phase 3 clinical trials

 

  • Post-hoc analysis of the EDITION 4, EDITION JPI, and JUNIOR trials of Toujeo vs. Lantus in type 1s looking at severe hypoglycemia risk.

  • During treatment period, fewer hypoglycemic events occurred in the Toujeo group than the Lantus group (6.2% vs. 9.3%, p=0.038).

  • Event rate for severe hypoglycemia was also numerically but not significantly lower with Toujeo (0.23 vs. 0.29 events/patient year).

  • Analysis further cements the benefits that next-gen basals such as Toujeo can offer on reducing hypoglycemia risk in type 1 patients when compared to older basal insulins.

InRange: a randomized controlled trial comparing Gla-300 vs IDeg-100 in people with type 1 diabetes (T1D) using continuous glucose monitoring (CGM)

 

  • CGM trial (n=340) of type 1s on either Tresiba or Toujeo for 12 weeks following four week run-in.

  • CGM metrics collected at 12 weeks, with primary endpoint of time in range and glucose variability. Rates of hypoglycemia will also be reported.

  • The study is still ongoing and is expected to complete in October 2020.

Switching from NPH insulin to glargine U300 results in significant decrease of A1c and reduction of hypoglycemia risk – results of multicenter, prospective observational study

 

  • 24 week, observational study evaluating Toujeo in type 2 patients who had recently switched from NPH (n=469)

  • Primary endpoint of an A1c reduction of at least 0.5% was achieved by 72% of patients, with average A1c decreasing by 1% from a baseline of 9.1% at baseline.

  • Average decreased by 37.1 ± 44.5 mg/dL (p<0.0000 vs. baseline) from 178.0 ± 46.0 mg/dL to 140.9 ± 40.7 mg/dL

  • The number of participants, who experienced ≥1 episode of symptomatic nocturnal hypoglycemia decreased significantly from 76 (66.1%) at baseline to 9 (14.8%) after 24 weeks of observation (p<0.0000).

  • None of the patients had severe hypoglycemia after switching to Toujeo throughout whole observation period.

  • Study results further confirm the superior efficacy of Toujeo vs. NPH in type 2 patients. We encourage payers to fully incorporate these benefits on A1c and hypoglycemia in terms of ensuring as many patients with type 2 and on insulin can switch to Toujeo or another next-gen insulin.

The forgotten populations: real-world patients with T2DM not meeting eligibility criteria of the glargine 300 U/ml edition and BRIGHT RCTs

 

  • Records from the Predictive Health Intelligence Environment were assessed to determine how many type 2 patients on basal insulin would have been excluded from the RCTs of EDITION and BRIGHT for Tresiba and Toujeo.

  • Of 191,218 evaluated records, 157,873 (83%) met ≥1 exclusion criterion; only 33,345 (17%) could have been included in RCTs. Excluded patients were older, with more previous hypoglycemia vs included patients, and were less likely to have an A1c between 7 and 9 %

Insulin glargine biosimilar (GP40061) shows similar pharmacokinetics and pharmacodynamics as compared to the reference drug

 

  • RCT of 42 patients with type 1 demonstrating similar PK/PD of a biosimilar to insulin glargine (GP40061).

  • Similarity was seen on the primary endpoint of AUC(ins 0-24 hours). There was no difference in adverse events.

Comparison of insulin degludec and glargine U100 in patients with type 1 diabetes prone to severe nocturnal hypoglycemia

 

  • 149 type 1 patients and at least one episode of nocturnal severe hypoglycemia during the past two years were enrolled and randomized to Tresiba or Lantus.

  • Patients in the Tresiba group had a 28% (95%CI:5-45; p=0.02) and 37% (95%CI:16-53; p=0.002) relative risk reduction (RRR) of nocturnal symptomatic hypoglycemia , compared to Lantus, when nighttime was defined as between midnight and 1 AM.

ATTD Yearbook and Opening Ceremony (Video)

Opening Ceremony: ATTD Attendance Swells to >7,800 Attendees in 13th Iteration; Dr. Jay Skyler Draws Parallels Between Past and Present in Review of Diabetes Technology

Conference co-chair Prof. Moshe Phillips kicked off this year’s opening ceremony with several stunning statistics on ATTD’s rapid growth. Despite this only being the meeting’s 13th congregation, over 7,800 participants from more than 81 countries are in attendance this year. In total, 615 abstracts were submitted, and the conference will feature 26 scientific sessions, 10 oral sessions, and 14 industry symposia – wow! These statistics were met with enthusiastic clapping, and throughout the day, we’ve chatted with conference goers who all agree that ATTD has quickly become one of the most impactful meetings in all of diabetes, indicative of the powerful momentum behind diabetes technology. (We do point out we were saying this back in 2010 – write us at info@closeconcerns.com if you’d like copies of our “full reports” for ATTD since 2009.) As Prof. Phillips noted, ATTD is particularly focused on innovation and interaction with broad groups of industry and other stakeholders is a mainstay of the meeting – whether that be up-and-coming start-ups or longtime veteran organizations in the field. For those of you who aren’t able to make it to this year’s meeting, we highly recommend you check out the online Education Portal, which will house a selection of the materials being shared. The app is fantastic and will keep you abreast of the sessions that are live-streamed.

  • After a very heart-warming introduction by conference co-chair Prof. Tadej Battelino, University of Miami’s Dr. Jay Skyler took to the stage to deliver a whirlwind, comprehensive, and hit presentation on the evolution of diabetes technology. We were particularly struck by the many parallels Dr. Skyler drew between diabetes care “then” (as far back as the first isolation of insulin in 1922) and diabetes care “now.” For example, Dr. Skyler opened his presentation with a quotation from The Diabetic Life by Dr. Robin Lawrence, first published in 1925, that still resonates today:

    • The temperament and usual habits of the patient should be considered in the type of treatment chosen and our object should be to interfere with these as little as is compatible with health… I know that full physiological control of severe diabetes – the most continuously normal blood sugar and the least hypoglycemia – can be obtained with 4-6 small injections of soluble insulin in the 24 hours.”

  • While the means and quality of insulin injection have certainly changed since then, the goals physicians aimed for even 50+ years ago continue to play a role in modern research and care. Similarly, Dr. Skyler highlighted Dr. Tim Danowski’s 1978 Diabetes Care paper “Jet Injection of Insulin During Self-Monitoring of Blood Glucose,” which was the first publication to champion the basal-bolus concept and self-monitoring of blood glucose. As Dr. Skyler put it, “the concepts were all there in 1978.” This framework gave incredible perspective to how far diabetes technology has come in the last 100 years to meet the goals and ideas that have been in discussion for so long. With the advent of closed loop systems, smart insulins, and more, we so look forward to seeing where the field will go in the next 100 years. We highly recommend visiting ATTD’s Facebook page where a video of the presentation is already up! 

  • Closing out the opening ceremony, live Flamenco performance from local Madrid musicians and dancers performed for the crowd. We loved seeing the crowd snap and clap to the music and erupt into cheers after each solo. See below for a photo of the festivities and check out Kelly’s Instagram to see some of the amazing show.

Continuous and Intermittent Glucose Monitoring

  • Slovenia’s Dr. Klemen Dovč highlighted three studies during his presentation on this year’s advances in CGM, flash glucose monitoring, and self-monitoring of blood glucose. ATTD has now moved to combining both intermittent and continuous glucose monitoring into one chapter, which contained a total of 14 publications for 2019. Of note, last year’s chapter on CGM (excluding IGM) was the second most read chapter in the entire yearbook. To start, Dr. Dovč featured the SMILE study, assessing the effect of Medtronic’s MiniMed 640G on hypoglycemia-prone adults with type 1 diabetes. Impressively, use of MiniMed 640G reduced the average number of hypoglycemia (<55 mg/dL) events per week from 4.4 to just 1.1. Dr. Dovč then pivoted to the first real-world data (n=900) from Scotland on NHS-funded flash glucose monitoring. A1c reductions were substantial, especially for those with a high A1c at baseline: -0.6% (95% CI: -1.2 to 0.1%, p<0.001) for those with an A1c ≥ 7.5% at baseline. Lastly, we were treated to one of our favorite studies on the cost-saving benefits of real-time CGM, using data from the CONCEPTT trial on pregnancy complicated with type 1 diabetes. There, analyses showed that implementation of real time CGM could save the NHS a substantial €9.6 million annually, driven by reductions in NICU admissions and shorter duration of NICU stay.

Insulin Pumps

  • Stanford’s Dr. Rayhan Lal presented the insulin pumps section of this year’s ATTD Yearbook, choosing to focus on structured education for pumps, dosing accuracy of pumps, and infusion sets. Dr. Lal presented Dr. Dominic Ehrmann’s paper in Diabetes Care on the effectiveness of education for pump therapy, Dr. Lal underscored the point that structured education around pump features is required in order to see the A1c improvement from switching from MDI to pumps, calling it “crucially important” to teach skills around the diabetes devices. Next, Dr. Lal showed Alderisio’s paper demonstrating weight gain comparable to MDI over 10 years, despite improved glycemic control and reduced insulin doses with CSII. Around this, Dr. Lal also called infusion sets the “weakest link in the chain.” Indeed, innovation in this space has been somewhat stalled for several years, though Medtronic recently received CE-Marking for a 7-day, extended wear infusion set. To finish, Dr. Lal showed a study from Dr. Guido Freckmann evaluating bolus dosing accuracy across several popular insulin pumps. While most pumps performed relatively well, the spread in delivery accuracy was noticeably poorer for Omnipod (MO below).

New Insulins, Biosimilars, and Insulin Therapy

  • On basal insulins, Dr. Bolinder reviewed results of the BRIGHT trial, which pit the two currently available next-gen basals on the market against each other (Sanofi’s Toujeo and Novo Nordisk’s Tresiba). See our past coverage of the trial here. Dr. Bolinder concluded from these results that both insulins provided “essentially similar” glycemic control and “comparable risk” of hypoglycemia. Looking ahead, he underscored that all trials of these next-gen basals in head-to-head settings have been in type 2 patients (see the similar CONCLUDE trial in type 2s as well) and that there should be similar trials performed in type 1 as well.

  • Dr. Bolinder touched on oral insulin and highlighted recent results from a phase 2 proof-of-concept trial for Novo Nordisk’s oral basal insulin I338. I338 is a long-acting basal insulin analog co-formulated in a tablet with an absorption enhancer, with key amino acid modifications to prevent gut degradation. The trial was done over eight weeks in insulin-naïve type 2 patients and showed similar effects on fasting plasma glucose levels (the primary endpoint) to insulin glargine. However, bioavailability was (predictably) a large concern, with end of trial doses of I338 >50 times that of the corresponding dose for the comparator insulin glargine. Further development of this candidate has been terminated by Novo Nordisk due to this issue. Still, Dr. Bolinder was relatively optimistic about the future of oral insulins in general, noting that the “concept has definitely been proven and is likely to see further exploration.”

Decision Support

  • Harvard’s Dr. Eyal Dassau took the Decision Support section of the Yearbook, focusing his chapter on AID algorithms and clinical decision support systems – both hot topics at this year’s ATTD. Dr. Dassau began with a valuable graph showing the Time in Range and time <70 mg/dl results for AID studies from 2014-2019. Of course, Control-IQ’s pivotal trial was the biggest AID study of 2019, published in NEJM in October. Dr. Dassau then shifted to clinical decision support systems, focusing on insulin dosing systems. In particular, Dr. Dassau pointed at Dr. Rich Bergenstal’s publication in The Lancet, which showed a full 1% A1c drop for the group using Hygieia’s d-Nav insulin dosing guidance (baseline: 8.7%) with HCP support compared to 0.3% for a control group with HCP support alone (baseline: 8.5%).

Technology and Pregnancy

  • It was a busy ATTD for Dr. Helen Murphy (King’s College), who presented the technology and pregnancy section of the Yearbook. Most notably, Dr. Murphy began with a “plea [for her] US colleagues” to notice the “excellent” evidence base and encourage use of CGM in pregnancy. Indeed, Dr. Murphy showed T1D Exchange data showing just one-third of pregnant women using CGM; nearly three-quarters used pumps, even though the evidence base for pumps in pregnancy is “much more limited.” Dr. Murphy looked at a real-world Swedish study demonstrating that mean Time in Range was ~40% at the beginning of pregnancy, increasing to ~50% by week 10, before plateauing again until the third trimester. Citing the consensus CGM goals for pregnancy, Dr. Murphy emphasized the need for the healthcare system to do a better job taking care of pregnant women, so that they are achieving the targets through the entire pregnancy: “We aren’t aiming for 70% Time in Range right at the end [of pregnancy], but sooner in the gestational age.”

Diabetes Technology and Therapy in the Pediatric Age Group

  • Dr. Maahs broadly tackled the question of whether these technologies and therapies are different in pediatrics, noting that “yes they are” and pointing to new pieces of data in the field. He began with a paper comparing international data from the T1D Exchange and DPV, showing that CGM usage in pediatrics has increased in both the US (3% to 22%) and Germany (4% to 19%) between 2011 and 2016, and that this use has been associated with lower A1c’s. He then touched on the PROLOG trial of Basal-IQ in pediatrics, which demonstrated a 31% reduction in hypoglycemia (<70 mg/dl), with no increase in mean glucose vs. SAP. Next, he pointed to results of 12-week results of closed loop in type 1 (nearly half of participants were <21 years old), showing 11% time in range improvements, 0.4% A1c drop, and less time in hypoglycemia vs. control. He explained that results from this study and others support adoption of closed loop tech in clinical practice across all age groups – a common theme of data presented at ATTD 2020 more broadly. Regarding progress of diabetes therapies in pediatrics, Dr. Maahs highlighted Victoza’s recent approval in adolescent type 2 diabetes as a welcome addition to the treatment armamentarium in this population.

Advances in Exercise Physical Activity and Diabetes Mellitus

  • Canada’s Dr. Michael Riddell (York University) presented the Exercise chapter of the Yearbook, noting that at least 8 papers were published in 2018/2019 around the topic of exercise in closed loop systems. Dr. Riddell also noted that more research needed in the area, particularly around how time of day of exercise affects glucose profiles. In the FIT study, Aronson et al. showed that morning (fasted) high intensity exercise in type 1s generally led to all-day hyperglycemia unless insulin correction was taken. In a study on timing of exercise in type 2s, Savikj et al. found that afternoon was superior to morning exercise for glucose control.

Practical Implementation of Diabetes Technology

  • The seemingly omnipresent Dr. Laurel Messer (Barbara Davis Center) presented “Practical Implementation of Diabetes Technology,” choosing to focus on three themes: (i) access and cost effectiveness; (ii) tech education; and (iii) automated decision support. On the first theme, Dr. Messer showed a map of CGM reimbursement in Europe (see below): on the left, representing 2009, the map is full of red and yellow, demonstrating poor access; on the right, representing 2018, the map is mostly green, with a few red countries. On the second theme, Dr. Messer showed T1D Exchange data demonstrating an increase in CGM use, but worsening glycemic control across the population. The answer, according to Dr. Messer, is more structured education. Backing her answer, Dr. Messer highlighted results from the INPUT study for insulin pumps and FLASH trial for FreeStyle Libre. Lastly, Dr. Messer noted the need for clinical decision support systems as patients and providers are increasingly dealing with diabetes devices and the data generated by these devices. On this front, she specifically highlighted DreaMed’s Advisor Pro system (see ATTD Day #2 highlight) and Hygeia’s d-Nav insulin dosing guidance system.

Diabetes Technologies and the Human Factor

  • Dr. Alon Liberman urged audience members not to forget the importance of “the human factor” when considering the many recent wins in diabetes technology. As he put it, “technologies themselves are rarely sufficient to deliver sufficient outcomes,” and researchers and clinicians must remember that technology is only one piece of the puzzle that will enable patients to achieve better outcomes. As an example, Dr. Liberman featured the 2018 DT&T publication, “Psychosocial and Human Factors During a Trial of a Hybrid Closed Loop System for Type 1 Diabetes Management,” which found decreased levels of diabetes distress and improved attitudes towards diabetes technology using the MiniMed 670G system. Dr. Liberman also discussed an intriguing paper published in Pediatric Diabetes on the relationship between hope (ability to set goals, sustain motivation toward goals, and recruit supportive resources toward goals), depression, and glycemic control in youth with type 1 diabetes. Results from the study suggest that hope may be a key factor in helping offset the impact of depressive symptoms associated with glycemic control, and research that works to foster hope in patients could therefore have significant benefit.

Immune Interventions for Type 1 Diabetes

  • Dr. Desmond Schatz touched on two recent breakthrough studies in the type 1 immune therapy space – the landmark teplizumab prevention trial and a trial of low-dose anti-thymocyte globulin, both with results presented initially at ADA 2019. Dr. Schatz noted that over the past ten years, hundreds of studies have been completed or are in process in terms of type 1 immunotherapies, but these two trials represent some of the most promising results to date. Of course, the teplizumab trial has been the source of much optimism, as it represents the first trial to show that immune therapies can actually be used to delay type 1 diabetes. Excitingly, the drug was recently granted Fast Track designation by the FDA. For more, see our full type 1 cure/prevention/treatment landscape here.

New Medications for Treatment of Diabetes

  • Dr. Irl Hirsch highlighted four therapies in the pipeline that he and his team are particularly excited about. First, he directed attention to tofogliflozin, which data show to be the most selective SGLT-2 inhibitor available (it’s only approved in Japan). Its selectivity for SGLT-2 is comparable to empagliflozin, twice as selective as dapagliflozin, and “much more” selective than canagliflozin. Some researchers have posited that this higher selectivity may lead to better outcomes on CV and renal outcomes, although this is not yet supported by data. As a reminder, Roche/Chugai had been pursuing development of tofogliflozin but discontinued further investment in 2Q13 for unspecified reasons.

    • Dr. Hirsch then moved to licogliflozin, a dual SGLT-1/2 inhibitor that displays “really interesting and dramatic” reduction in glucose excursions and significant weight loss. However, side effects are a major concern, with 91% of patients with obesity in a 12-week weight loss trial of the candidate having diarrhea. Dr. Hirsch believes that licogliflozin could be an attractive strategy in obesity. Currently, Novartis studying the candidate in NASH after discontinuing development in obesity in 2Q18 because of efficacy concerns.

    • Empagliflozin in NAFLD was also discussed: results from small trials show promising results in terms of liver fat decreases, leading Dr. Hirsch to wonder whether SGLT-2 inhibitors (potentially in combination with pioglitazone) could work in NAFLD, regardless of type 2 status.

    • Finally, Dr. Hirsch reviewed landmark phase 2b data of Lilly’s dual GLP-1/GIP agonist tirzepatide. This data was first presented at EASD 2018 and was staggering in terms of its effects on A1c, weight loss, and adverse events. Dr. Hirsch provided some context on the effects of GIP agonism, noting that GIP is secreted from enteroendocrine K cells, and like GLP-1, is a potent stimulator of glucose-dependent insulin secretion. One thought concerning how GIP delivers its effects is that it might increase metabolic flexibility by enabling increased fat utilization in the fasting state and reduce fat availability in the postprandial state. Dr. Hirsch closed his remarks with this: “what type of drug will this be? My guess is an expensive one!” Indeed, if efficacy holds up in Lilly’s ongoing phase 3 SURPASS program for tirzepatide (which excitingly also includes a head to head CVOT vs. Trulicity), then tirzepatide will indeed be quite the valuable drug.

NAFLD/NASH and Diabetes (Video)

  • Dr. Kavita Garg presented on the newest addition to the ATTD Yearbook’s collection of chapters – NASH and NAFLD. Out of the 400 relevant articles, 19 were selected to be in this year’s chapter. We’ve certainly noticed growing interest on the subject at recent conferences, which is no surprise, considering the 35-40% prevalence in people with diabetes. Dr. Garg broke down the 19 featured articles into four broad categories: those discussing (i) global prevalence rates of NASH/NAFLD; (ii) preventative strategies to optimize type 2 and NASH/NAFLD management; (iii) non-invasive biomarkers to diagnosis the conditions and avoid the necessity of liver biopsy; and (iv) novel pharmacotherapies. On the third point, highlighted biomarkers included elevated ALT, triglycerides, and BMI, as well as improved imaging with US and MR elastography and CT. We were inspired by Dr. Garg’s quote that “Ten years ago, we used to see steatohepatitis on MRI/CT, and we would just ignore it because we didn’t understand it…Now, we report it as accurately as we can.” To end her presentation, Dr. Garg posed the interesting question of whether individuals with diabetes should be screened for NAFLD/NASH in the same fashion as diabetic retinopathy or nephropathy – What a neat idea! Looking to the future, we hope to see diagnostic tools that can make this thought a reality.

Using Digital Health Technology to Prevent and Treat Diabetes

McGill University’s Dr. Tal Oron stepped in for chapter editors Drs. Neal Kaufman and Eran Mel to give a thematic presentation on advances in digital health technology. According to Dr. Oron, while there are many publications on digital health in diabetes, few have been able to deliver meaningful results. This fact reflects how relatively new a field digital therapeutics is, as well as the high cost and time barriers required of impactful studies. Dr. Oron focused the majority of his remarks on the importance of self-efficacy as a goal for digital health interventions. Individuals with higher self-efficacy look at challenging tasks as tasks to be mastered, have a deeper interest in and commitment to activities, and recover quickly from setbacks. This self-efficacy translates into accelerated positive health behavior change and reduced depression and stress-related disorders for patients. To illustrate this type of intervention, the editors chose to feature a cost-saving analysis on Livongo’s remote diabetes management program. Program access was associated with a cost reduction of $88 saving per member per month at one year on top of the self-management tools provided by Livongo.

Exhibit Hall

Abbott

Abbott’s booth was busy throughout ATTD as many providers stopped by to check out the new FreeStyle Libre 2. According to reps at the booth, FreeStyle Libre 2 has still only launched in Germany and Norway, though other countries in Europe are coming soon. In the US, FreeStyle Libre 2 still remains with the FDA, where it has been since at least 1Q19. Notably, materials at the booth advertised accuracy data for FreeStyle Libre 2, giving us the ability to directly compare Abbott’s data with the special controls criteria set by the FDA for iCGM clearance. See the tables below. 

In total, Abbott’s accuracy data showed 18,926 paired points in adults and 6,584 paired points for pediatrics. We aren’t sure where the data was gathered from (i.e., Europe or US), but the data show that the accuracy of FreeStyle Libre 2 in adults does indeed meet the iCGM special controls. In emails with Abbott, we confirmed that the same set of adult data was used in the labeling for FreeStyle Libre 2 in Germany, where the device is already available (see picture from German FreeStyle Libre 2 user guide below). Note that the tables presented at the booth are binned differently (in order to reflect US iCGM specifications), but the data is the same. Despite the lengthy FDA review, Abbott’s management has maintained their confidence in eventual clearance as an iCGM. We’ve been uncertain of what’s been causing the delay with the FDA – while some have assumed it has to do with interoperability standards or accuracy in hypoglycemia, presumably it could be other things. Indeed, however, the data is certainly very positive to see:

  • The adult data shows FreeStyle Libre 2 is significantly more accurate than the original FreeStyle Libre in hypoglycemic ranges – the MARD for adults fell from 11.4% with FreeStyle Libre to 9.2% FreeStyle Libre 2 for adults (as shown at IDF 2019), while pediatrics data was shown for the first time at ATTD with MARD dropping from 13.9% to 9.7% (compared to YSI). It was surprising that this data was in the booth only and not at any presentation but we look forward to seeing it discussed more broadly in the future.

  • Notably, while the pediatric data for FreeStyle Libre 2 falls slightly short of the iCGM special controls, we understand that the FDA focuses largely (some might say completely – that’s speculation on our part) on seeing adult accuracy data in range, which this certainly is.

  • From a business standpoint, the delay in launch for FreeStyle Libre 2 is not really slowing any of Abbott’s momentum (big banners on the Feria de Madrid entrance proclaimed “over two million” users for FreeStyle Libre), adding the Bluetooth connectivity and optional alarms at the same price and form factor certainly make the product even more compelling for many populations, particularly for pediatrics. We believe reducing hyperglycemia will become a bigger focus in the future for all manufacturers and we look forward to seeing more detail on this.  To this end, we really enjoyed Abbott’s “Free to Dream without Lancets” graphic at their booth, highlighting the optional alarms enabling patients to sleep in peace. That is a nice offer – on the other hand, if alarms are going off consistently, rather than avoiding annoying partners and turning off alarms, we’d suggest patients work with HCPs to fix whatever is consistently prompting hypoglycemia or hyperglycemia – that’s the power of the alarms to begin with – to spot patterns and address them!

Table 1. Comparison of FreeStyle Libre 2 accuracy data with iCGM special controls.

Performance Standard: Lower bound of one-sided 95% confidence interval

FreeStyle Libre 2 in Adults

FreeStyle Libre 2 in Pediatrics

Euglycemia: >70% within ±15% for 70-180 mg/dl

72.9%

74.8%

Euglycemia: >99% within ±40% for 70-180 mg/dl

99.4%

98.9%

Overall: >87% within ±20% over full device measuring range

88.6%

88.1%

Hypoglycemia: >85% within ±15 mg/dl for <70 mg/dl

86.6%

77.2%

Hypoglycemia: >98% within ±40 mg/dl for <70 mg/dl

99%

97.5%

Hyperglycemia: >80% within ±15% for >180 mg/dl

89.3%

84.2%

Hyperglycemia: >99% within ±40% for >180 mg/dl

99.9%

99.1%

Table 2. Summary of FreeStyle Libre 2 accuracy data.

 

Adult (n=18,926 paired points)

Pediatric (n=6,584 paired points)

% within 15 mg/dl or 15%

85.6%

83.7%

% within 20 mg/dl or 20%

92.4%

91.6%

MARD

9.2%

9.7%

Picture below taken from FreeStyle Libre 2’s German user guide (provided to us by Abbott).

  • Since there is focus on accuracy, we would make two other points:

    • We also feel it’s important to point out that peds management has improved light-years since the old days: the arrows alone make peds management significantly better, as do better accuracy and alarms. Night and day! Contextualizing improvement is very important, from our view. As a teenager, Kelly had so many ER visits – all prompted by severe hypoglycemia that could have been avoided by either arrows, or by the alarms, to say nothing of better accuracy.

    • While under 10% MARD is clearly nice to see, decisions to change doses or therapy in the cases where ongoing therapy can be optimized are very important levers to use to reach better glycemic management – we believe that to date, these are underused and that all CGM can improve performance.

A. Menarini

A. Menarini’s booth was dedicated to its CE-marked GlucoMen Day CGM “powered by Waveform” and corresponding Bluetooth meter for calibration. The system has 14-day-wear, MARD in the range of 11%-13% based on information at DTM 2018 and EASD 2019, one fingerstick calibration per day, and data-to-mobile app transfer. Reps shared that the product hopes to be submitted for FDA clearance by the end of 2020, representing a push back in timelines from original goals of launching in the US by 2020. We also learned that the company is expanding to new countries in North Africa and Middle East though specifics weren’t mentioned. A. Menarini is responsible for performing sales, marketing, training, and customer support for WaveForm’s CGM throughout Europe, the Middle East, Africa, and Latin America.

Biocorp

Biocorp’s small both featured its reusable connected pen attachment “Mallya.” The two-year device, priced at ~$122 and compatible with “all major pens,” contains two pieces, one of which clips on top of the pen and another with fits over the control dial/button. The device relays dose, time, and date to a mobile application via Bluetooth. The device received CE-marking as a class IIb device in June 2019, and reps shared that the company plans on “selling” the device in the US market by the end of the year. This would imply an FDA clearance at least several months earlier, pushing the company back on its original goal of having it in the US by the end of 2019 or 1Q20. While reps did not comment on the news, Biocorp received ~$2.2 million from Sanofi to incorporate Mallya with Sanofi’s insulin pens while Sanofi would acquire non-exclusive distribution rights to it the product worldwide. The company also has a data-sharing partnership with DreaMed to feed insulin dosing data into the MDI/basal-only Advisor Pro clinical decision support system.

Capillary Biomedical

Capillary Biomedical’s understated booth was fully dedicated to its SteadiSet Infusion set. As a reminder, this technology intends to address three major problems of insulin pump therapy: (i) slowing the body’s wound response to foreign materials by making the cannula out of a “soft polymer material”; (ii) spreading insulin out to a larger surface area of subcutaneous tissue  to enhance consistent absorption; and (iii) preventing kinking through a wire coil reinforcement on the inside of the cannula. A poster was also on display comparing the performance of SteadiSet with a traditional 90° Teflon IIS (T90) in pigs for two weeks. The data indicated that the SteadiSet Infusion Set caused less tissue trauma between two days and two weeks post-insertion and was resistant to kinking. In August of 2018, the company had raised $2.9 million in angle funding to support FDA clearance and launch by the end of 2019 – when speaking with booth representatives today, we learned that the product had secured an IDE from the FDA and was recruiting for the pivotal trial.

Dexcom

Dexcom’s booth featured large banners on each side highlighting two exciting features coming in “Spring 2020”: G6 in pregnancy and upper-arm wear. Of course, Dexcom users have been using G6 in pregnancy and on the upper arm off-label for quite some time! CE-Marking for Dexcom G6 in pregnancy was officially announced on the second day of ATTD, with launch commencing in the UK “starting spring 2020.” Notably, the NHS has already committed to offer CGM to all pregnant type 1s by 2020/2021, given the robust cost-effectiveness. Upper-arm wear of G6 is quite common, though the device is only officially indicated for abdomen wear in adults and abdomen and buttocks for pediatrics.

Diabeloop

Diabeloop’s small booth featured its DBLG1 algorithm connected to a personal handset, Dexcom G6 CGM, and Kaleido patch pump. The booth had a nice picture of the algorithm’s three layers: (i) safety first (risk of hypoglycemia evaluation); (ii) physiological framework and expert system (insulin decision management); and (iii) auto-learning (personalized parameters optimization over time). The third feature is not yet available but is “forthcoming” with additional R&D. Reps had several exciting updates, stating that the company has secured the highest level of designation for obtaining reimbursement in France and is in discussions with payers in Germany as well. In January, the company started a nine-month clinical trial (n=180) with DANA pumps in France to obtain FDA clearance, which we’d estimate would happen in late 2021 if results are positive. As of EASD 2019, Diabeloop had intended to submit the algorithm as an “iController” pathway which we also assume will happen now following Control-IQ’s clearance in December. The hybrid closed loop system received CE-marking in November 2018 and is currently in a “soft launch” mode with ~35